Lecture Notes in Chemistry 99

Solar Light Harvesting with Nanocrystalline Semiconductors

Oleksandr Stroyuk

Lecture Notes in Chemistry

Volume 99

Series editors

Barry Carpenter, Cardiff, UKPaola Ceroni, Bologna, ItalyBarbara Kirchner, Bonn, GermanyKatharina Landfester, Mainz, GermanyJerzy Leszczynski, Jackson, USATien-Yau Luh, Taipei, TaiwanEva Perlt, Bonn, GermanyNicolas C. Polfer, Gainesville, USAReiner Salzer, Dresden, Germany

The Lecture Notes in Chemistry

The series Lecture Notes in Chemistry (LNC) reports new developments inchemistry and molecular science-quickly and informally, but with a high qualityand the explicit aim to summarize and communicate current knowledge for teachingand training purposes. Books published in this series are conceived as bridgingmaterial between advanced graduate textbooks and the forefront of research. Theywill serve the following purposes:

• provide an accessible introduction to the field to postgraduate students andnonspecialist researchers from related areas,

• provide a source of advanced teaching material for specialized seminars, coursesand schools, and

• be readily accessible in print and online.

The series covers all established fields of chemistry such as analytical chemistry,organic chemistry, inorganic chemistry, physical chemistry including electrochem-istry, theoretical and computational chemistry, industrial chemistry, and catalysis. Itis also a particularly suitable forum for volumes addressing the interfaces ofchemistry with other disciplines, such as biology, medicine, physics, engineering,materials science including polymer and nanoscience, or earth and environmentalscience.

Both authored and edited volumes will be considered for publication. Editedvolumes should however consist of a very limited number of contributions only.Proceedings will not be considered for LNC.

The year 2010 marks the relaunch of LNC.

More information about this series at http://www.springer.com/series/632

Oleksandr Stroyuk

Solar Light Harvestingwith NanocrystallineSemiconductors

123

Oleksandr StroyukLaboratory of Organic Photovoltaics andElectrochemistry

L.V. Pysarzhevsky Institute of PhysicalChemistry

KievUkraine

ISSN 0342-4901 ISSN 2192-6603 (electronic)Lecture Notes in ChemistryISBN 978-3-319-68878-7 ISBN 978-3-319-68879-4 (eBook)https://doi.org/10.1007/978-3-319-68879-4

Library of Congress Control Number: 2017955261

© Springer International Publishing AG 2018This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or partof the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmissionor information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilarmethodology now known or hereafter developed.The use of general descriptive names, registered names, trademarks, service marks, etc. in thispublication does not imply, even in the absence of a specific statement, that such names are exempt fromthe relevant protective laws and regulations and therefore free for general use.The publisher, the authors and the editors are safe to assume that the advice and information in thisbook are believed to be true and accurate at the date of publication. Neither the publisher nor theauthors or the editors give a warranty, express or implied, with respect to the material contained herein orfor any errors or omissions that may have been made. The publisher remains neutral with regard tojurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer NatureThe registered company is Springer International Publishing AGThe registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

The solar light conversion and storage is currently one of the most blossominginterdisciplinary fields of science converging the physical chemistry, physics ofsolid state, optics, photochemistry, electrochemistry, catalysis, and many otherresearch directions. The present textbook is intended to give a perspective on thecurrent state of photochemical systems for the solar light harvesting based onnanocrystalline semiconductor materials and assemblies. The book chapters providean account on various aspects of such systems, including the solar water splittingand evolution of molecular hydrogen, the photosynthetic processes of CO2 and N2

reduction, and the photoelectrochemical solar cells based on nanoparticulatesemiconductor materials. A special focus is made on a “nano” aspect of semi-conductor photocatalysis—the role of nanocrystals and size effects in the solarenergy conversion, the design of semiconductor nanostructures with tailored pho-tochemical properties, and the perspectives of nanophotocatalysis and photovoltaicsystems based on semiconductor quantum dots.

The introduction provides a brief account on various concepts of the solar lightharvesting using the bulk and nanocrystalline semiconductors as well as a shorthistorical account on the development of various photochemical and photovoltaiclight conversion technologies.

The first chapter is an introduction to the photochemistry of semiconductornanoparticles (NPs). It highlights basic principles of the selection of semiconductormaterials for the applications in the solar light harvesting and requirements to theoptical and electrophysical properties of photoactive semiconductor NPs. Thechapter is focused on special features of the nanocrystalline semiconductors, inparticular, on the quantum size effects and a unique capability of semiconductorNPs for the photoinduced charging. We discuss the most prominent size effects inthe photochemistry of semiconductor NPs such as a dramatic enhancement of thephotocatalytic/photoelectrochemical activity of nanocrystalline semiconductors ascompared to their bulk counterparts, a crucial role of the surface charge traps in thephotochemical processes, the effects of NP shape and porosity, the charging-induced changes in the NP photoreactivity, etc.

v

The second and third chapters provide a review of the current state of the art inthe semiconductor-based light-harvesting systems for the water splitting and thereduction of carbon dioxide and dinitrogen. The semiconductor-catalyzed photo-chemical water splitting for the hydrogen production as a green and sustainable fuelis discussed in detail. A review of the photocatalytic systems for the photosyntheticreduction of water, CO2, and N2 encompasses the systems based on thedye-sensitized oxide nanocrystalline semiconductors, binary semiconductorheterostructures, a survey of the visible-light-sensitive metal-chalcogenidenanophotocatalysts, and new and emerging nanostructured photocatalysts andcocatalysts of these photosynthetic processes.

The fourth chapter introduces the reader to the semiconductor-based photo-electrochemical solar cells designed for the conversion of solar light into electriccurrent. As the topic of dye-sensitized liquid-junction semiconductor solar cells hasrecently been broadly covered elsewhere, the discussion is limited mostly to thesemiconductor nanoparticle-sensitized solar cells with liquid electrolytes, where thelight conversion occurs as a result of a cyclic series of photochemical/photocatalyticprocesses and secondary “dark” redox reactions.

The fifth chapter provides a concise account on typical synthetic approachesused for the preparation of various semiconductor nanomaterials—the colloidalNPs, nanocrystalline powders, thin films, binary and more complexnanoheterostructures, and nanocomposites of semiconductors with other functionalcomponents, such as metal NPs, carbonaceous compounds, etc.

The final sixth chapter has a methodological character and acquaints the readerswith the experimental methods using light as a probe of the structure, electro-physical, photophysical, and photochemical properties of nanocrystalline semi-conductors and related heterostructures. The chapter discusses the methods ofabsorption and photoluminescence spectroscopy, flash photolysis, and other spec-troscopic techniques that can be used to gain insights into the photochemicalbehavior of semiconductor NPs.

I would like to thank all persons who helped me in writing this book, particu-larly, to my wife, Dr. Alexandra Raevskaya, a trusted friend and colleague, for hersteady support and discussions on the book subject. Also, I appreciate deeply theexperience and skills acquired by my coauthoring with senior peers from L.V.Pysarzhevsky Institute of Physical Chemistry, National Academy of Sciences ofUkraine—Prof. Anatoliy Kryukov, Prof. Stepan Kuchmiy, and Prof. VitaliyPokhodenko. Recently, we have published a comprehensive book on semicon-ductor nanophotocatalysis (Publishing House “Akademperiodika”, Kyiv, Ukraine,2014) and the coauthoring of this book has been a source of invaluable experienceand constant inspiration for me. As this book was written during my stay inTechnical University of Dresden as a Marie-Skłodowska Curie Fellow, the supportof European Union’s Horizon 2020 Research and Innovation Program (GrantNo. 701254) is deeply appreciated.

vi Preface

I hope that the present book will be useful both for a novice reader who starts ajourney into the exciting world of the solar light-harvesting science and for anadvanced reader who is already familiar with the field and seeks an informativereview on principal topics of the solar light conversion, such as the solar cells andsemiconductor-based artificial photosynthesis.

Kiev, Dresden Dr. Oleksandr Stroyuk2016–2017

Preface vii

Contents

1 Basic Concepts of the Photochemistry of SemiconductorNanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Light Absorption by Bulk and Nanocrystalline

Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61.2 Influence of Surface States on the Photochemical

Properties of Semiconductor NPs . . . . . . . . . . . . . . . . . . . . . . . . 141.3 Influence of Size Dependences of CB and VB Levels . . . . . . . . . 191.4 Photoinduced Charging of Semiconductor NPs . . . . . . . . . . . . . . 23References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

2 Semiconductor-Based Photocatalytic Systemsfor the Solar-Light-Driven Water Splitting and HydrogenEvolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392.1 Photocatalytic Systems Based on the Wide-Band-Gap

Semiconductors and Sensitizers . . . . . . . . . . . . . . . . . . . . . . . . . 412.2 Photocatalytic Systems Based on the Binary and More Complex

Semiconductor Heterostructures . . . . . . . . . . . . . . . . . . . . . . . . . 482.3 Photocatalytic Systems Based on the Metal-Doped

Wide-Band-Gap Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . 552.4 Photocatalytic Systems Based on the Nonmetal-Doped

Wide-Band-Gap Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . 582.5 Photocatalytic Systems Based on the Metal-Sulfide

Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 612.6 Emerging Semiconductor Photocatalysts for the Solar Hydrogen

Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712.7 New-Generation Co-Catalysts for the Photocatalytic Hydrogen

Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 852.8 Stoichiometric Water Splitting Under the Illumination

with the Visible Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

ix

3 Semiconductor-Based Photocatalytic Systems for the ReductiveConversion of CO2 and N2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1273.1 Photocatalytic Reduction of Carbon Dioxide . . . . . . . . . . . . . . . 1283.2 Photocatalytic Fixation of Dinitrogen . . . . . . . . . . . . . . . . . . . . . 146References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

4 Semiconductor-Based Liquid-Junction PhotoelectrochemicalSolar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1614.1 Principles and Designs of Semiconductor NP-Sensitized

Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1614.2 Basic Photoelectrochemical Characteristics of SSSCs . . . . . . . . . 1664.3 Nanocrystalline Photoanodes Produced by the Ex Situ

Deposition of Sensitizer NPs . . . . . . . . . . . . . . . . . . . . . . . . . . . 1704.4 Nanocrystalline Photoanodes Produced by the In Situ

Deposition of Sensitizer NPs . . . . . . . . . . . . . . . . . . . . . . . . . . . 1824.5 Making Progress in SSSCs—Toward More Efficient

and Less Toxic Photoelectrodes . . . . . . . . . . . . . . . . . . . . . . . . . 2034.6 Nanocrystalline Semiconductor Counter-Electrodes

for SSSCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

5 Synthesis of Nanocrystalline Photo-Active Semiconductors . . . . . . . . 2415.1 Colloidal Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2425.2 Nanocrystalline Powdered Semiconductors . . . . . . . . . . . . . . . . . 2515.3 Nanocrystalline Films of Photo-Active Semiconductors . . . . . . . . 2605.4 Mesoporous Photo-Active Semiconductor Nanomaterials . . . . . . 2645.5 Spatially Organized Nanocrystalline Photo-Active

Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2685.6 Nanocrystalline Photo-Active Semiconductors on Carriers . . . . . . 2715.7 Doped Semiconductor Nano-Photocatalysts . . . . . . . . . . . . . . . . 2735.8 Bi- (Multi-) Component Photo-Active Semiconductor

Nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2755.9 Photo-Active Semiconductor/Metal Nanostructures . . . . . . . . . . . 281References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284

6 Probing with Light—Optical Methods in Studiesof Nanocrystalline Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . 3196.1 A Brief Characterization of the Spectral Studies

of Nano-Semiconductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3196.2 Studies of Nano-Photocatalysts by the Electron Absorption

Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3216.3 Luminescence Spectroscopy as a Tool for the Studies

of Nanocrystalline Semiconductors . . . . . . . . . . . . . . . . . . . . . . 3296.4 Studies of Nanocrystalline Semiconductors

by the Time-Resolved Photolysis Techniques . . . . . . . . . . . . . . . 339

x Contents

6.5 Studies of Nanocrystalline Semiconductors Using RamanScattering Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350

6.6 Studies of Colloidal Semiconductor-Based Systems UsingDynamic Light Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373

Contents xi

Abbreviations and Symbols

AIS Silver indium sulfide, Ag-In-SCB Conduction bandCBD Chemical bath depositionCE Counter electrodeCIS Copper indium sulfide, Cu-In-SCNP Carbon nanoparticleCVD Chemical vapor depositionCys CysteinCZTS Copper zinc tin sulfide, Cu2ZnSnS4DDT DodecanethiolDEL Double electric layerDLS Dynamic light scatteringDMSO Dimethyl sulfoxideDSSC Dye-sensitized solar cellEDTA Ethylenediaminetetraacetic acidEDX Energy-dispersive X-ray spectroscopyEIS Electrochemical impedance spectroscopyEMA Effective mass approximationEPR Electron paramagnetic resonanceFTO Fluorine-doped tin oxide (transparent conductive glass)FWHM Full width on half-maximumGCN Graphitic carbon nitride (g-C3N4)GO Graphene oxideGSH GlutathioneHDA Hexadecyl amineHOMO Highest occupied molecular orbitalHS Hollow sphereHTT Hydrothermal treatmentICL Inorganic complex ligandsIE Ion exchange

xiii

IPCE Incident photon to current efficiencyIR InfraredITO Indium tin oxide (transparent conductive glass)LO Longitudinal optical phononLUMO Lowest unoccupied molecular orbitalMAA Mercaptoacetic acidMOF Metal–organic frameworkMPA Mercaptopropionic acidNB Non-stationary bleachingNHE Normal hydrogen electrodeNP NanoparticleNR NanorodNT NanotubeNS NanosheetNW NanowireOA Oleic acidODE OctadeceneOLA OleylamineOTE Optically transparent electrodePCE Power conversion efficiencyPEC PhotoelectrochemicalPEG Polyethylene glycolPEI PolyethyleneiminePL PhotoluminescencePVA Polyvinyl alcoholPVP PolyvinylpyrrolidoneQSE Quantum size effectQY Quantum yieldRGO Reduced graphene oxideSEM Scanning electron microscopySILAR Successive ionic layer adsorption and reactionSPP Sodium polyphosphateSPR Surface plasmon resonanceSO Surface optical phononSSSC Semiconductor-sensitized solar cellSTEM Scanning transmission electron microscopyTBT Titanium tetrabutoxideTEA TriethanolamineTEM Transmission electron microscopyTGA Thioglycolic acidTMD Transition metal dichalcogenideTOP TrioctylphosphineTOPO Trioctylphosphine oxideTTIP Titanium tetraisopropoxideUV Ultraviolet

xiv Abbreviations and Symbols

VB Valence bandVis Visible lightWZ WurtziteXRD X-ray diffractionZAIS Zinc-doped silver indium sulfide, Zn-Ag-In-SZB Zinc blendeaB Bohr radius of excitonEg BandgapEF Fermi energyECB Conduction band potentialEVB Valence band potentiale-CB Conduction band electronh+VB Valence band holee–tr/h

+tr Trapped electron/trapped hole

m*e Effective mass of CB electronm*h Effective mass of VB holehv Light quantum energykbe Wavelength of fundamental absorption band edge of a semiconductorJsc Short-circuit current densityVoc Open-circuit voltageRCT Charge transfer resistanceη Total light conversion efficiencyD Donor of electronA Acceptor of electronFF Fill factor of a voltage–current characteristic

Abbreviations and Symbols xv

Introduction

…nature is not in a hurry and mankind is.

Giacomo Ciamician 1912.

A solar light-harvesting system can be defined in the broadest terms possible as acombination of light-sensitive moieties (molecules, metal complexes, supramolec-ular complexes, nanodimensional inorganic or organic particles, biomolecules, andtheir assemblies) with various electron mediators, cocatalysis, etc. that serves toabsorb the incoming solar light and converts the luminous energy into the form ofelectrical or chemical energy available for the future utilization or for the immediatechemical/electrochemical reactions.

According to this definition, the artificial light-harvesting systems (that is, thesystems devised by humans inspired by the natural light-harvesting systems) can betentatively divided into three major classes: (i) the systems for the light-to-currentconversion or the so-called solar cells, (ii) the systems for the artificial photosyn-thesis and accumulation/storage of the luminous energy in the form of stablechemical products, and (iii) the systems for immediate utilization of the solar energyas a driving force for various, mostly destructive, photochemical/photocatalytictranformations of chemical species. The latter systems are broadly used in theenvironmental photocatalysis, where the solar light energy is applied to decomposevarious persistent organic and inorganic contaminants and the harmful biota both inthe gas phase and waters.

In the present book, we will focus on the former two types of systems thatproduce photocurrent or/and stable energy-rich products and, therefore, can be usedfor the conversion of the light energy in the form suitable for a long-term storageand a broad distribution to potential consumers.

The efficient harvesting of solar light was an ever-inspiring dream of many greatscientists, from philosophers to middle-age alchemists to academic chemists andphysicist. The idea of mimicking the natural photosynthesis to accumulate theenormous energy flux supplied annually by Sun was discussed already by theancient philosophers, but, probably, the first to put it into correct words and to

xvii

present it to a broad audience was Italian chemist Giacomo Ciamician, oftenhonored as the Father of Photochemistry [1−4]. In his famous speech published inScience (1912) [5], he formulated the idea of photocatalysis to be applied for thesolar light harvesting: “… the solar energy that reaches a small tropical country …is equal annualy to the energy produced by the entire amount of coal mined in theworld!”, “… By using suitable catalyzers, it should be possible to transform themixture of water and carbon dioxide into oxygen and methane, or to cause otherendo-energetic processes” [1, 2]. Ciamician recognized the main problem of thesolar light harvesting that still restricts a broad implementation of thephotocatalytic/photoelectrochemical technologies, that is, a relatively low intensityof the solar light flux reaching the Earth surface as compared to the “concentrated”energy that can be produced from traditional fossil fuels.

The Ciamician’s speech on the future of solar light conversion technologies is sovivid and precise that a large portion of its deserved to be cited unchanged anduninterrupted: “Where vegetation is rich, photochemistry may be left to the plantsand by rational cultivation, as I have already explained, solar radiation may beused for industrial purposes. In the desert regions, unadapted to any kind ofcultivation, photochemistry will artificially put their solar energy to practical uses.On the arid lands there will spring up industrial colonies without smoke andwithout smokestacks; forests of glass tubes will extend over the plants and glassbuildings will rise everywhere; inside of these will take place the photochemicalprocesses that hitherto have been the guarded secret of the plants, but that willhave been mastered by human industry which will know how to make them beareven more abundant fruit than nature, for nature is not in a hurry and mankind is.And if in a distant future the supply of coal becomes completely exhausted, civi-lization will not be checked by that, for life and civilization will continue as long asthe sun shines! If our black and nervous civilization, based on coal, shall befollowed by a quieter civilization based on the utilization of solar energy, that willnot be harmful to progress and to human happiness.”

The progress of the mankind confirmed decisively the correctness of Ciamician’sprophesies. Indeed, the steady development by using sustainable raw sources andregenerative energy sources, like the solar energy, wind energy, geothermal energy,etc. nowadays not only determines the competitiveness of a country’s economy butcan be the sole way to deal with a grave challenge of the global climatic changescaused by CO2 and heat emissions.

It is expected that the energy demands of the humankind will increase to 50 TWin 2050 requiring the alternative non-carbon energy sources because the energyproduction in such a scale will invariably cause dramatic disruptions in the globalclimate balance [6]. At the same time, more energy is sent by the Sun to the Earth’ssurface in one hour that the humankind consumes in a year [3] and this energy ispotentially available to every country and every person and waits to be harvestedand used.

The solar light harvesting now comes by two parallel and sometimes intertwinedways of progress (Fig. 1). One is the realization of the Ciamician’s dream ofartificial photosynthesis, that is, the endothermic conversion of extensively

xviii Introduction

abundant compounds—water, N2, and CO2 into the chemical products that can beprocessed later releasing the accumulated energy or, alternatively, used as valuablechemical raw materials for more complex syntheses.

Such photosynthetic processes require a series of concerted reactions involvingmany electrons, protons, and other components and, therefore, they can be realizedwith acceptable efficiency only in the presence of catalysts and photocatalysts.

The second way is the direct solar light energy conversion into the electric powerthat occurs in the so-called solar cells. The main component of the solar cell is alight absorber which is light-excited and supplies nonequilibrium electrons andelectron vacancies—holes into an electric circuit, thus resulting in the photocurrentgeneration.

Numerous studies carried out around the globe in the last three–four decadesshowed that both types of the light-harvesting systems can be successfully realizedby using photosensitive semiconductor materials. Indeed, some semiconductorsshow characteristics ideal for the solar light harvesting. The semiconductors haveintense and continuous absorption bands that cover entire UV, visible, and,sometimes, near IR (NIR) ranges of the solar spectrum. This feature arises from thepossibility of the electron transition from a continuous filled valence band (VB) intoa continuous vacant conduction band (CB) under the excitation with the light of anyenergy higher than the distance between CB and VB, that is, the bandgap Eg.Typically, the semiconductors are crystalline and robust and reveal uncomparablyhigher photochemical stability than the light-harvesting molecular species and

Fig. 1 A scheme illustrating the most important semiconductor-based systems for the solar lightharvesting

Introduction xix

metal coordination compounds. Finally, some of the semiconductor materials usedto convert/store the solar light are extremely abundant in the nature, such as siliconor iron oxide.

In comparison to the semiconductors, the molecular and metal-complex lightharvesters typically have much narrower absorption bands and are prone to variousphotochemical transformations resulting in the deterioration of theirlight-absorption capability. Also, typically quite a complex chemistry is needed forthe preparation of efficient light-harvesting molecules, complexes, and assembliesthat can compete with the natural photosynthesizing species and semiconductormaterials.

From the historical perspective, the semiconductor-based photocatalytic, pho-tovoltaic, and photoelectrochemical (PEC) systems stem from the same roots. Forexample, the photocatalytic water splitting can be realized in a combinedphotochemical/photoelectrochemical regime in a water-splitting solar cell [7].

Today, probably nobody knows for sure the exact priority of the first photo-catalytic experiments using semiconductors, but it is generally agreed upon that areal breakthrough in the area was inspired by the works of A. Fujishima and K.Honda on the PEC splitting of water on a biased titanium dioxide electrode. Indeed,TiO2 was an ideal material for the semiconductor-based light-harvesting systemsdue to availability, low cost, stability, and the millennia-long story of utilization as apigment [8]. However, titania is only sensitive to the UV and a tiny (around 5%)portion of the visible light. This utter limitation on the background of so manyunique positive properties inspired and continues to inspire numerous concepts andmethods for the sensitization of TiO2 to the visible light by coupling it with thehighly absorbing species or by altering its band structure imparting the TiO2

crystals with the capability to absorb the visible light.In recent two–three decades, the photochemistry of semiconductor materials

experiences a real explosive growth associated with the miniatuarization of thesemiconductor crystals to the nanometer dimensions. A transition from the micro-to nanoscale opens huge perspectives and potential of tailoring/designing theproperties of semiconductor light harvesters via variations in the crystal size, shape,nanoparticle (NP) association mode, NP surface chemistry, doping, etc.

Evolution of photocatalytic synthetic processes with the participation ofsemiconductors. The present book focuses on the photocatalytic processes resultingin the accumulation of the light energy in the form of endothermic chemical sub-stances such as the molecular hydrogen as a main product of the water reductionand various products of multi-electron reduction of atmosphere-abundant CO2 andN2. The details of working principle and examples of corresponding photocatalyticsystems are discussed in Chap. 2 (the water reduction) and Chap. 3 (the reduction ofCO2 and N2). Here, we provide only a very general description of basic principlesof the semiconductor-based photocatalytic systems introducing the reader into thisfield of photochemistry.

The photocatalytic process starts when a semiconductor crystal absorbs a lightquantum with the energy equal to or (typically) higher than the width of the for-bidden band (or the bandgap Eg, process 1 in Fig. 2). The photogenerated charge

xx Introduction

carriers can then recombine either via a radiative pathway 2 emitting a photolu-minescence quantum or via a non-radiative pathway 3 providing a vibrationalenergy to the crystal lattice and adsorbed species.

Some portion of the charge carriers migrates in the crystal and can reach thesurface (processes 4) where the carriers get “trapped” by various lattice defects(vacancies, adatoms, undercoordinated atoms, etc.) as well as by the adsorbedspecies. The photogenerated VB hole typically has a high oxidation potential andgets filled with an electron from various donor species present in the system, forexample, from water molecules or OH– ions, the process resulting in the evolutionof molecular oxygen as a final product. The photogenerated CB electrons “seek” foraccepting species present in the system, such as protons or H3O

+ ions reducingthem to the molecular hydrogen (or CO2– to CO, CH2O, formate, CH4, and otherproducts). After the withdrawal of both CB electron and VB hole to the acceptingand donating species, respectively, the semiconductor crystal regains its originalstate thus finishing the photocatalytic cycle. Provided that fresh donors andacceptors are constantly supplied to the surface of semiconductor crystal, it canfunction as a “pump” transferring electrons from the donors to the acceptors at theexpense of the solar light energy.

Pioneer reports on the semiconductor-mediated photocatalytic processesappeared as early as in 1920–1930s dealing mostly with the photobleaching of dyesin the presence of titania crystals [7−9]. In the late 1960s, the studies of the waterphotoelectrolysts on titania electrodes were started by Fujishima and Honda, whoreported in 1969–1972 on the photoelectrochemical water splitting on asingle-crystal rutile electrode illuminated with the UV light [8, 10, 11]. In theproposed “artificial photosynthesis” scheme, the processes of water reduction andoxidation were separated in space—the oxidation of H2O to O2 took place on theilluminated and externally biased titania electrode with the participation of VBholes, while the reduction of H2O to H2 occurred in a separate vessel on theplatinum foil that accepted electrons from the titania CB (Fig. 3a, b).

Fig. 2 A general layout of a semiconductor-based photocatalytic system

Introduction xxi

The miniatuarization of semiconductor crystals to several microns and the directdeposition of Pt NPs on the titania surface allowed to compose thesuspension-based photocatalytic systems where both water reduction and oxidationtook place on the suspended microparticle aggregates. Such systems can functionwithout an external bias as the photogenerated electrons can migrate through the netof contacting crystals, thus avoiding the electron–hole recombination, contrary tothe single-crystal titania electrodes (Fig. 3c). The next step in the developmentof the water-splitting systems was the introduction of “fuels”—the so-called“sacrificial” electron donors that provided electrons for the water reduction, beingoxidized to CO2 and other products.

When the process is performed in the presence of air, a water-splitting systemconverts into a system for the oxidation of sacrificial donors, as the electrons aretransferred to O2 and the net result of the photocatalytic process is the total oxi-dation of the introduced organic species to carbon dioxide and other inorganiccompounds (nitrates, sulfates, phosphates, etc.)—the so-called photocatalytic“mineralization” of organic compounds. Such systems started to live an indepen-dent life and constituted a special direction of the semiconductor photocatalysisdealing with the photocatalytic destruction of harmful organic compounds, thewater decontamination, the air purification, the mitigation of harmful microorgan-isms, etc. [4, 8−10]. The development of such environment-oriented photocatalyticsystems took place in 1980–1990s simultaneously with the advancement in thesemiconductor-based light-energy-accumulating systems.

A future progress of the environmental semiconductor photocatalysis can bereadily grasped by “reviewing” the current reviews on this subject [4, 8−10, 12,13]. This progress encompasses the systems for drinking water and air purification,the development of self-cleaning and photoactive building materials, variousanti-bacterial coatings, etc. The environmental photocatalysis blossomed in 1990–2000s resulting in the first real commercial implementations of the photocatalytictechnologies. As noted in [4], starting from 2000, a steady flux of more than 1300

Fig. 3 A layout (a) and a schematic energy diagram (b) of the pioneer PEC cell for the watersplitting proposed by Fujishima and Honda; (c) illustration of a photocatalytic water-splittingsystem based on a suspended semiconductor photocatalyst. Reprinted with permissions from [10].Copyright (2012) American Chemical Society

xxii Introduction

international patents on the photocatalytic technologies was observed, contrary to afew tens per year before 1990.

An ever higher attention is currently paid to the semiconductor-mediated pho-tocatalytic “synthetic” reactions allowing to form C–C, C–N, C–S, and other bondsand to realize the photo-driven syntheses of valuable industrial chemical products[3, 4]. The implementation of semiconductor photocatalysis in the organic synthesisis typically impeded by two obstacles. The first one is a low selectivity of thephotocatalytic reactions on the semiconductor materials. Indeed, the VB holes inthe most extensively studied semiconductor photocatalyst—TiO2—exist in the formof highly reactive HO• radicals that can oxidize indiscriminately almost everyorganic compound introduced into the system. Recently, various approaches weredeveloped to selective oxidative and reductive processes on the semiconductorphotocatalysts, including the doping, surface state engineering, introduction ofmetal-complex cocatalysts and enzymes, etc. [4]. The second problem consists in atypically low light intensity of the conventional light sources. To enhance thephotochemical transformations using such low-intensity light fluxes, a specialattention is paid to the design of new reactor types including the fluidized bedreactors and the continuous-flow systems [4]. Simultaneously, the technologies ofsolar light concentration are under the development resulting in the first promisingresults [4].

The issues of the photochemical production of non-carbonaceous fuels and theutilization of renewable energy sources advanced greatly starting from 2000s, whenthe perils of the over-abundant CO2 emission as a result of enormous consumptionof the fossil fuels, were truly realized. The realization and apprehension of clearevidence of the hazardous global climatic changes resulted in a shift of the energypolicy of most developed economies and the eve of 2010s was marked by astrongly renewed interest in all the spectrum of light-harvesting technologies,including the semiconductor-based solar cells and the photocatalytic systems pro-ducing hydrogen from water or mimicking the natural photosynthesis by convertingCO2 and N2 by using the solar light energy.

The need in efficient photochemical light conversion systems was stronglysupported by the development of chemistry and photochemistry of nanocrystallinesemiconductors [11]. The studies of nanoparticulate semiconductors and variousrelated composites showed an unprecedented and virtually unlimited variabilityof the properties and functional characteristics that can be achieved by varying thesize, shape, and composition of semiconductor nanocrystals, as well as a way of thespatial organization of NP-based systems on the nano-, micro-, and macroscalelevels [11].

The so-called “Holy Grail” of the photocatalytic technologies of water splittingis an apparent quantum efficiency of 30% at 600 nm [11]. Currently, theseveral-percent efficiency at wavelengths as long as 500 nm was achieved with thetotal light conversion efficiency below 0.1%. Therefore, new-generation semicon-ductor photocatalysts with a bandgap of around 2 eV and lower are needed to createphotocatalytic water-splitting systems feasible for the practical implementation.

Introduction xxiii

Evolution of semiconductor-based solar cells. The attempts of the light energyconversion into the electric power stem from the photoelectric effect discovered byHenri Bequerel in 1839 [14, 15]. It should be noted that Bequerel observed theillumination-induced electric current between two electrodes immersed in a liquidelectrolyte. Later, the photoeffect was also observed for solid electrolytes and afocus of the solar cell studies was turned to solid photovoltaic materials for manydecades to come, till the dye-sensitized liquid-junction solar cells appeared on theresearch scene and started their fast progress.

The first to appear was the silicon-based solar cells. Silicon has a bandgap of 1.1eV which is close to the optimal value of *1.3 eV (peak of the solar irradiationspectrum) necessary for the achievement of the highest photoconversion efficiency.It is a photochemically stable (at least in the crystalline state), low-toxic, andearth-abundant material [6, 16]. The first efficient solar cell on the crystalline siliconwas produced by Bell laboratory in 1954 and showed an efficiency of 6% [14, 16].Today, such cells dominate the solar light converter market along with the amor-phous Si-based sells [14]. Together, such cells occupy around 80% of the globalsolar cell market [16, 17].

The Si-based solar cells, typically named the first-generation semiconductorsolar cells (Fig. 4), though being unrivalled in terms of the ratio of conversionefficiency versus production cost, have a number of shortcomings. First, the siliconsolar cell technology requires a huge amount of very pure silicon. Typically, thephotovoltaic technologies used rejected materials from the semiconductor industryand the necessary amount of raw materials can be maintained only if both industriesare developed with the same rate, which is doubtful in view of a recent drasticgrowth of interest to the solar energy harvesting. The solar cells based on amor-phous silicon, which is much less expensive, emerged in 1960–1970s with a firstcommercial cell available in 1981 [14].

Another fundamental shortcoming of the silicon is the indirect character of theinterband electron transitions (see Chap. 1) resulting in a comparatively low linearabsorption coefficient. At least a 100-lm-thick silicon layer is required for thecomplete solar light absorption on the Earth surface, thus putting limitations on theminimal thickness (and, therefore, the cost) of the solar cells.

The above shortcomings stimulated a search for new conceptions/materials, forexample, the application of new crystalline forms of Si with a reduced thickness,like Si nanoribbons, amorphous silicon, cadmium chalcogenides, ternary copper–indium–chalcogenides, and more complex quaternary compositions [14]. It wasfound already in mid-1980s that some of the materials, in particular, Cu2S and

Fig. 4 Evolution of semiconductor-based solar cells

xxiv Introduction

CdTe (studied since 1960s), CuInSe2, CuGaSe2, CuInS2 and their alloys Cu(In, Ga)(S, Se)2 (CIGS/CIGSe, studied from 1070s), and kesterite Cu2ZnSnS4 (studied from1990s), sputtered as thin films can be used as the solar cell light absorbers [14, 17].Most of such compounds are the direct-bandgap semiconductors with highabsorption coefficients and, therefore, much thinner (1–2 lm) absorber layers arerequired for the efficient solar light harvesting [6]. Additionally, the alloying ofseveral components can be used for a precise variation of the absorber bandgap.

The thin-film cells can be formed on a variety of substrates (flexible or rigid,conductive, or insulating) with a broad spectrum of techniques, including plasmavacuum deposition, chemical vapor deposition, electrodeposition, sputteringmethods, etc. [18].

A typical thin-film cell is designed as a p–n junction and includes a 1-lm-thicksputtered molybdenum contact on soda-lime glass, 1–2-lm p-type-conductingmetal-chalcogenide absorber layer and a thin layer (*50 nm) n-type-conductingsemiconductor material (for example, CdS), a transparent layer of undoped ZnOthat prevents the shunting, and, finally, a conductive layer of Al-doped zinc oxidewith a thickness of around 120 nm (Fig. 5a).

The thin-film cells were christened as the second-generation semiconductor cells(Fig. 4) and showed efficiencies of 20% and higher [19]. However, the cell pro-duction puts very rigorous requirements to the purity of sources used for thethin-film sputtering, and therefore, the second-generation cells are very expensiveand found applications mostly in the aerospace industry [14, 18], where the cell costis not so critical.

Recently, the thin-film solar cell technology gained a new impetus by discov-ering the possibility of using NP or molecular precursor “inks” for the preparationof metal-chalcogenide thin-film absorbers. The ternary and quaternary compoundscan be prepared by using the well-established methods of the colloidal chemistryand concentrated to the form of inks. The inks can be deposited on any desirablesubstrate by the conventional inkjet printing and annealed in a non-oxidativeatmosphere resulting in the film solidification and formation of good-qualityabsorber layers [6]. In a similar way, other components of the solar cell (metalcontact, n-type component, barrier layers, etc.) can be prepared as the NP inks so

Fig. 5 A typical layout of (a) thin-film CIGS-based solar cell; (b) organic/inorganicperovskite-based solar cell. (c) an energy level diagram for the CH3NH3PbI3-based cell pre-sented in (b). Reprinted with permissions from [6] (a) and [20]. Copyright (2010, 2015) AmericanChemical Society

Introduction xxv

that the entire solar cell can be produced by using the inkjet printing technologiesmaking such solar cells competitive to the conventional silicon-based devices [6].

A breakthrough in the efficiency of thin-film bulk-heterojunction solar cells isexpected by using special light harvesters—semiconductor NPs capable of themulti-exciton generation. When some of the narrow-bandgap semiconductor NPs,such as PbX (X = S, Se, Te), CdTe, InP, Si, etc. [21], are illuminated with the lightenergy exceeding considerably the NP bandgap, one absorbed light quantum can beaccommodated by generating several electron–hole pairs. This effect is possible dueto a special feature of semiconductor NPs—the so-called “phonon bottleneck”. Theterm is used to refer to the slow thermalization of the primary electron–hole pairfavoring to the channelling of lattice energy to other routes, in particular, into thegeneration of additional electron–hole couples till the energy excess (hv − Eg) iscompletely accommodated in the excited NP [21]. The phenomenon ofmulti-phonon generation gives a hope to surpass the fundamental Shockley–Queisser limit of 31−33% light conversion efficiency achievable for asingle-junction solar cell [21].

In recent years, the thin-film solar cells based on the organo-inorganic per-ovskites have emerged coming the way to around 20% efficiency in mere 5 yearsfrom the publication of the first reports [20, 22, 23]. Such perovskites combine highabsorption coefficients, a variable bandgap that can be tuned across the entirevisible and NIR ranges, an unprecedently large electron mobility and charge dif-fusion coefficient, the tolerance to point defects and grain boundaries, and relativesimplicity of the cell formation [22]. The most popular materials are CH3NH3PbX3

(X = Cl, Br, I). Typically, a perovskite layer is sandwiched between an electrontransport layer (an inorganic wide-bandgap semiconductor like TiO2, SnO2, andZnO) and a transparent hole transporting layer, for example, a derivative of spir-obifluorene (spiro-OMeTAD, Fig. 5b, c) [20]. However, for a successful imple-mentation of such solar cells, a number of quite critical issues should be addressed,including the efficient recycling of Pb-containing cells after their utilization, as wellas a low chemical and photochemical stability of the lead-based organo-inorganicperovskites.

The idea of devising solar cell with two electrodes—a light-sensitivephotoanode/photocathode and a catalytically active counter electrode connectedby a liquid electrolyte—stems directly from the Bequerels experiments on thephotoelectric effect [15, 24]. As discussed above, this idea was realized by A.Fujishima and K. Honda in their PEC cell for the water splitting on a rutile singlecrystal. However, TiO2 can absorb only a fraction of the solar irradiation and,therefore, the spectral sensitivity range of TiO2 crystals should be extended tolonger wavelengths either by modifying the band structure of the crystal or byintroducing visible-light-absorbing species on the crystal surface. The idea of thesensitization of wide-bandgap semiconductors, like TiO2, with external molecularabsorbers is also one of the oldest conceptions of the solid-state photochemistry,introduced by Vogel in 1883 for silver halide emulsions used in the photographicprocess [15, 24]. A combination of a liquid electrolyte (“liquid-junction”)-basedPEC cell with the dye sensitization approach gave rise to the dye-sensitized solar

xxvi Introduction

cells (DSSCs) belonging to the third-generation solar cells (Fig. 4), according to thegenerally accepted classification.

The sensitization effect was also observed long ago for the narrow-bandgapsemiconductors deposited to/formed on the surface of the wide-bandgap semi-conductors, for example, for Ag2S produced by an ion exchange on the surface ofAgBr (AgI) [25]. It is, therefore, logically inevitable that the conception of DSSCswas later extended to the semiconductor-sensitized solar cells (SSSCs) with thesame liquid-junction architecture as in the DSSCs [26−29].

The DSSC research was strongly stimulated by the fuel crisis of 1973 and over athousand papers on DSSCs emerged in a few years from the start of the studiesattesting to an explosive growth of the DSSC field [15, 24, 27, 30]. A regenerativeDSSC is designed to return to its original state after a cycle of the photoinduced andsecondary (“dark”) chemical reactions on the electrodes with a net result of gen-erating the electric power from the light energy (Fig. 6).

The solar light is absorbed in the DSSCs by a molecular sensitizer—a dye or ametal complex. Ref. 30 provides an extensive review of the basic classes of organicdyes and metal coordination compounds that were tested as sensitizers of theliquid-junction DSSCs. It was found that the highest efficiency is observed for theDSSCs with monolayer-adsorbed sensitizer molecules because competitive pro-cesses of the intermolecular interactions (charge transfer, formation of excimers,etc.) result in a loss of the light-harvesting efficiency at higher coverages [30].

To achieve a high-light absorption with a single dye molecule layer on thewide-bandgap semiconductor surface, it was suggested to use mesoporous elec-trodes with a high specific surface area. As a result, the typical DSSCs comprise amesoporous dye-sensitized TiO2 (or ZnO or SnO2) photoanode that supplies thephotogenerated electrons into the electric circuit and regenerates the original statevia the oxidation of a redox shuttle in the electrolyte [15, 24, 27, 30].

Typically, the mesoporous layer has a thickness of *10 lm and consists ofloosely aggregated 10−30-nm particles resulting in 50−60% porosity and completepermeability with the liquid electrolyte. The mesoporous layer is formed on theoptically transparent electrodes (OTEs), such as indium tin oxide (ITO) or

Fig. 6 (a) Scheme of a dye-sensitized liquid-junction solar cell. (b) A 900-cm2 glass-basedsandwich DSSC module. The device consists of six serial-connected so-called meander-typecurrent-collecting parts; (c) A building-integrated DSSC demonstrator from Dyesol. Reprintedwith permissions from [30]. Copyright (2010) American Chemical Society

Introduction xxvii

fluorine-doped tin oxide (FTO). The enormous experimental work on the selectionand modification of the sensitizers to achieve the highest light conversion efficiencypushed to the summit a class of Ru–bipyridyl complexes as the most efficient andstable light harvesters [15, 24, 27, 30]. Also, a large variety of counter electrodes,electrolyte compositions, and redox shuttles were probed, and the highest lightconversion efficiencies were achieved for the Ru–bpy complex sensitizers, Ptcounter electrodes, and iodide/iodate redox couples in mixed water/polar organicsolvents [30].

A dye absorbs the solar light and injects an electron into CB of thewide-bandgap support or, alternatively, directly into the subbandgap states of themetal oxide NP originating from the surface defects. The one-electron-oxidized dyeregenerates its original state by accepting an electron from a reducing componentof the redox shuttle, for example, iodide ions (Fig. 6a) [30]. The oxidized form ofredox shuttle, I�3 ions, diffuses through the electrolyte layer (*50 lm) to the Ptcounter electrode, where it is reduced to I− by the electrons coming from thephotoanode through the electric circuit.

If no appropriate redox couple is present in a DSSC, the photogenerated holescan oxidize water to O2, while the electrons transferred to the counter electrode canreduce water to hydrogen and the DSSCs perform as a photochemical solar cell forthe water splitting as discussed above. The light conversion efficiency of such cellscan be boosted by introducing a hole scavenger that acts as a consumable fuelenhancing the H2 generation and suppressing the oxidation of water [15, 24].

As discussed in Ref. [32], the DSSC concept is a good example of a system,where the performance of the overall device is better than that of the separatecomponents. Indeed, the mesoporous titania cannot absorb efficiently the solar lightand also does not conduct electric current. The conventional Ru–bpy sensitizersdegrade very quickly when illuminated in solutions without any oxide support andredox shuttles. However, a combination of all the components into a united systemresults in a solar cell that can generate the electric current densities of up to 20mA/cm2 and exhibits stable performance for more than 15 years in the outdoor solarillumination [30]. The DSSC technology is the one that has come the way from theearly laboratory concepts to the small pilot cells and, finally, to the large-scalecommercial realization (see some examples in Fig. 6b, c).

The toughest challenge for the DSSCs still to be met is to surpass a threshold of15% efficiency [30]. The thermodynamics of the DSSC design allows to achievethis value; however, quite spectacular efforts applied in the field of DSSC in tworecent decades resulted in only *11% efficiency for the best-performing cells.A small ratio of applied efforts to the achieved efficiency increments observed in therecent years stimulated the studies of other liquid-junction cell designs, in partic-ular, the above-mentioned SSSCs with the semiconductor sensitizer introduced inthe form of NPs. The SSSCs with liquid electrolyte are the main subject of Chap. 4.

The working principle of the SSSCs is essentially based on a combination of thelight-driven photocatalytic processes on a photoanode (photocathode) and theelectrocatalytic processes on a counter electrode (Fig. 7), very similar to the

xxviii Introduction

above-discussed DSSCs. Here also, the light is absorbed by the NPs of anarrow-bandgap semiconductor, for example, CuInS2 (the photosensitizer NPs),resulting in the electron transfer to the wide-bandgap porous semiconductor metaloxide layer (TiO2 or ZnO). The VB hole of the sensitizer NPs is then filled at theexpense of the oxidation of sulfide ions—one of the components of the redoxcouple present in the liquid electrolyte and having a very high adsorption affinity tothe surface of sensitizer NPs.

The elemental sulfur produced as a result of the S2– photooxidation gets boundby the polysulfide species and diffuses to the CuxS counter electrode where it isreduced by the electrons arriving from the photoanode and with this, the PEC cycleis finished leaving the cell in exactly the same state as before the light adsorption.

The SSSCs started with a modest few-percents efficiency of the light harvestingbut showed an accelerated growth and achieved in 2015−2016, a promising effi-ciency higher than 11%. The potential of such cells is still to be realized to a fullextent. There exists a general optimism toward such semiconductor NP-sensitizedliquid-junction solar cells in the research community, which was vividly expressedby P. Kamat in his essay “Quantum Dot Solar Cells. The Next Big Thing inPhotovoltaics” [31]. The NP-sensitized SSSC field is one of the principal focusesof the present book and comes under a detailed discussion in Chap. 4.

It is instructive to conclude the introduction to the semiconductor-based solarcells with some numerical data on the current top efficiencies. As reported in 2015in a regularly updated solar cell efficiency table [19], the highest light conversionsachieved are 25.6% and 11.4% for polycrystalline and amorphous silicon,respectively, 28.8%—for the gallium arsenide thin-film cells, 21%—for thethin-film CdTe-based cells, and 11.9%—for the DSCCs [19]. The highest reportedefficiency for the NP-sensitized solar cells is currently around 12% [32]. Forcomparison, the top efficiency of a solar cell based exclusively on the organicsemiconductors and the charge transport layers is around 11%, while the highest

Fig. 7 A layout of the photoelectrochemical cycle in the liquid-junction SSSC based onlight-harvesting CuInS2 NPs and polysulfide redox couple

Introduction xxix

light conversion efficiency of 37.9% was reported for the multi-junction thin-filmsemiconductor cells [19].

References

1. Venturi M, Balzani V, Gandolfi MT (2005) Fuels from solar energy. A dream of GiacomoCiamician, the father of photochemistry. Proceedings of the Solar World Congress, Orlando,Florida, 2005.

2. Albini A, Dichiarante V (2008) The ‘belle ´epoque’ of photochemistry. Photochem PhotobiolSci 8:248–254. doi:10.1039/b806756b

3. Protti S, Fagnoni M (2009) The sunny side of chemistry: green synthesis by solar light.Photochem Photobiol Sci 8:1499–1516. doi:10.1039/b909128a

4. Ravelli D, Dondi D, Fagnonia M, Albini A (2008) Photocatalysis. A multi-faceted concept forgreen chemistry. Chem Soc Rev 38:1999–2011. doi:10.1039/b714786b

5. Ciamician, G., The Photochemistry of the Future, Science, 1912, 36, 385.6. Habas SE, Platt HAS, van Hest MF, Ginley DS (2010) Low-Cost inorganic solar cells: From

ink to printed device. Chem Rev 110:6571–6594. doi:10.1021/cr100191d7. Walter MG, Warren EL, McKone JR, Boettcher SW, Mi Q, Santori EA, Lewis NS (2010) Solar

water splitting cells. Chem Rev 110:6446–6473. doi:10.1021/cr10023268. Hashimoto K, Irie H, Fujishima A (2007) TiO2 Photocatalysis: A Historical Overview and

Future Prospects. AAPPS Bulletin 17:12–28.9. Serpone N, Emeline AV, Horikoshi S, Kuznetsov VN, Ryabchuk VK (2012) On the genesis of

heterogeneous photocatalysis: a brief historical perspective in the period 1910 to themid-1980s. Photochem Photobiol Sci 11:1121–1150. doi:10.1039/c2pp25026h

10. Teoh WY, Scott JA, Amal R (2012) Progress in heterogeneous photocatalysis: From classicalradical chemistry to engineering nanomaterials and solar reactors. J Phys Chem Lett 3:629−639. doi:10.1021/jz3000646

11. Maeda K (2011) Photocatalytic water splitting using semiconductor particles: History andrecent developments. J Photochem Photobiol C 12:237–268. doi:10.1016/j.jphotochemrev.2011.07.001

12. Chen H, Nanayakkara CE, Grassian VH (2012) Titanium dioxide photocatalysis inatmospheric chemistry. Chem Rev 112:5919−5948. doi:10.1021/cr3002092

13. Ljubas D (2005) Solar photocatalysis—a possible step in drinking water treatment. Energy30:1699–1710. doi:10.1016/j.energy.2004.11.010

14. Goetzberger A, Luther J, Willeke G (2002) Solar cells: past, present, future. Sol Energy MaterSol Cells 74:1–11.

15. M Grätzel (2001) Photoelectrochemical cells. Nature 414:338–344. doi:10.1038/3510460716. Green MA (2009) The path to 25% silicon solar cell efficiency: History of silicon cell

evolution. Prog Photovolt: Res Appl 17:183–189. doi:10.1002/pip.89217. Unold T, Schock HW (2011) Nonconventional (Non-Silicon-Based) photovoltaic materials.

Annu Rev Mater Res 41:297–321. doi:10.1146/annurev-matsci-062910-10043718. Chopra KL, Paulson PD, Dutta V (2004) Thin-Film solar cells: An overview. Prog Photovolt:

Res Appl 12:69–92. doi:10.1002/pip.54119. Green MA, Emery K, Hishikawa Y, Warta W, Dunlop ED (2015) Solar cell efficiency tables

(Version 45). Prog Photovolt: Res Appl 23:1–9. doi:10.1002/pip.257320. Ke W, Fang G, Liu Q, Xiong L, Qin P, Tao H, Wang J, Lei H, Li B, Wan J, Yang G, Yan Y

(2015) Low-Temperature Solution-Processed Tin Oxide as an Alternative ElectronTransporting Layer for Efficient Perovskite Solar Cells. J Am Chem Soc 137:6730−6733.doi:10.1021/jacs.5b01994

xxx Introduction

21. Nozik AJ, Beard MC, Luther JM, Law M, Ellingson RJ, Johnson JC (2010) SemiconductorQuantum Dots and Quantum Dot Arrays and Applications of Multiple Exciton Generation toThird-Generation Photovoltaic Solar Cells. Chem Rev 110:6873–6890. doi:10.1021/cr900289f

22. Song TB, Chen Q, Zhou H, Jiang C, Wang HH, Yang Y, Liu Y, Youab J, Yang Y (2015)Perovskite solar cells: film formation and properties. J Mater Chem A 3:9032–9050. doi:10.1039/c4ta05246c

23 Grätzel M (2014) The light and shade of perovskite solar cells. Nature Mater 13:838–842.doi:10.1038/nmat4065

24. M Grätzel (2003) Dye-sensitized solar cells. J Photochem Photobiol C 4:145–153. doi:10.1016/S1389-5567(03)00026-1

25. James TH (1977) The Theory of The Photographic Process. Macmillan, New York.26. Rühle S, Shalom M, Zaban A (2010) Quantum dot sensitized solar cells. ChemPhysChem

11:2290–2304. doi:10.1002/cphc.20100006927. Hodes G (2008) Comparison of dye- and semiconductor-sensitized porous nanocrystalline

liquid junction solar cells. J Phys Chem C 112:17778–17787. doi:10.1021/jp803310s28. Kamat PV (2008) Quantum dot solar cells. Semiconductor nanocrystals as light harvesters.

J Phys Chem C 112:18737–18753. doi:10.1021/jp806791s29. Kamat PV, Tvrdy K, Baker DR, Radich JG (2010) Beyond photovoltaics: semiconductor

nanoarchitectures for liquid-junction solar cells. Chem Rev 110:6664–6688. doi:10.1021/cr100243p

30. Hagfeldt A, Boschloo G, Sun L, Kloo L, Pettersson H (2010) Dye-Sensitized Solar Cells.Chem Rev 110:6595–6663. doi:10.1021/cr900356p

31. Kamat PV (2013) Quantum dot solar cells. The next big thing in photovoltaics. J Phys ChemLett 4:908–918. doi:10.1021/jz400052e

32. Sharma D, Jha R, Kumar S (2016) Quantum dot sensitized solar cells: recent advances andfuture perspectives in photoanode. Sol Energy Mater Sol Cells 155:294–322. doi:10.1016/j.solmat.2016.05.062

Introduction xxxi

Chapter 1Basic Concepts of the Photochemistryof Semiconductor Nanoparticles

A strong interest to photochemical (photocatalytic, photoelectrochemical, andphotoelectro-catalytic) processes with the participation of semiconductors wasobserved starting from 1970ths and resulted in the rise of semiconductor photo-chemistry as an independent discipline with exciting perspectives of applications inthe chemical industrial synthesis, solar energy conversion, environmental protec-tion, etc. [1–6]. The semiconductor photochemistry combined conceptions andknowledge of “classic” molecular photochemistry, catalysis, solid state physics,spectroscopy and other disciplines.

In recent years, a renaissance of the semiconductor photochemistry is observedassociated with successful developments in the physics and chemistry ofnanocrystalline semiconductors, in particular the semiconductor nanoparticles(NPs) displaying size dependences of optical and electrophysical characteristicsdeemed before to be fundamental and invariable for the corresponding “bulk”semiconductor materials [7–12].

The so-called “quantum size effects” (QSEs) are of a special importance for thesemiconductor photochemistry, the term QSEs referring to all possible sizedependences of fundamental electro-physical properties of semiconductors with acrystal size smaller than a certain “critical” value. The QSEs originate from thespatial confinement of the photogenerated electrons and holes (or a boundelectron-hole pair—exciton) in the volume of NPs typically smaller than the excitondiameter (or doubled exciton Bohr radius aB) for a given semiconductor material [8,10, 13–15]. Among the typical QSEs are size dependences of the bandgap Eg,spectral parameters and intensity of absorption and photoluminescence (PL) bands,oscillator strengths of optical excitonic transitions, exciton binding energy, as wellas a gradual transformation of continuous energy bands [conduction band (CB) andvalence band (VB)] into a spectrum of discrete electron states as the NP size isdecreased. The latter effect makes possible another important size effect—thephotoinduced charging of semiconductor NPs, that can affect strongly their pho-tochemical behavior.

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_1

1

The critical size for the QSEs to be observable can be associated with de Brogliewavelength of electron or exciton diameter 2aB [7, 8, 10, 16]. Two cases of spatialexciton confinement can be distinguished—a weak confinement, when the NP sized is close to 2aB, and a strong confinement at d < 2aB. The aB depends on thechemical nature of a semiconductor and can vary in a relatively broad range(Table 1.1). As a result, the range of NP sizes, where QSEs can be observed, is broad.

The dynamics of photochemical and photo-electrochemical (PEC) processeswith the participation of semiconductors depends largely on their electro-physicalparameters, in particular, the bandgap and energies (electrochemical potentials) ofconduction and valence bands—ECB and EVB, respectively [1, 9]. A photochemicalprocess starts after the absorption of a light quantum hv, which is possible under acondition hv � Eg. The absorption results in an electron transition from VB to CB,leaving a hole in VB (Fig. 1.1). Typically, the electron-hole pair has an excess ofvibrational energy (hv–Eg) which is accommodated by the crystal lattice till theelectron and hole reach the CB bottom and VB top corresponding to the ECB andEVB levels (“thermalization” process).

Table 1.1 Bohr excitonradius aB for somesemiconductors [7, 10]

Semiconductor aB, nm

CdS 2.4

ZnS 1.5

CdSe 3.9

TiO2 0.8–1.9

ZnO 4.8

CuCl 0.7

PbI2 1.9

PbS 18

PbSe 46

Fig. 1.1 Photoinducedcharge transfers between asemiconductor, acceptor A,and donor D

2 1 Basic Concepts of the Photochemistry of Semiconductor …

Afterwards two key processes occur—the migration of charge carriers to thecrystal surface and their interfacial transfer to other components of the photo-chemical system. The dynamics of interfacial charge transfer, in turn, depends onrelative positions of the electron and hole potentials and the redox potentials of anelectron acceptor A and an electron donor (a hole acceptor) D. The charge transferis possible when ECB is more negative than the redox-potential of the acceptorE0A/A•−) and EVB is more positive than the redox potential of the donor E0(D/D•+)[9] (Fig. 1.1), that is,

ECB\E0 A=A��ð Þ ð1:1Þ

and

EVB [E0 D=D�þð Þ ð1:2Þ

In “classic” photochemical systems based on bulk semiconductors, where ECB

and EVB are the characteristic constants for a given semiconductor (when measuredin the bulk but varying near the crystal surface as a result of double electric layervariations), only those acceptors and donors, which comply with the conditions(1.1) and (1.2), respectively, can participate in the photochemical transformationson the semiconductor crystal surface. A free energy of the interfacial electrontransfer can be expressed as [14, 17].

DG0 ¼ e ECB�E0 A=A��ð Þ� � ð1:3Þ

A similar expression can be written also for the hole transfer. Thus, the design ofphotochemical systems based on bulk semiconductors with size-invariable Eg, ECB,and EVB is limited to the selection of appropriate electron acceptors and donors.Some influence can have also the adsorption of potential-determining ions. Forexample, the CB potential of metal oxide semiconductors, such as TiO2 and ZnOcan be tuned by pH variations as ECB(pH) = ECB

(pH 0) − 0.059pH.In the case when both conditions (1.1) and (1.2) are satisfied, a photoexcited

semiconductor donates an electron to A and accepts an electron from D and thusregenerates its original (prior to the light absorption) state. Such events can,therefore, happen many times in a cyclic manner and the photochemical processoccurs in a photocatalytic regime. In the case of neutral A and D species, theelectron transfers generate an anion-radical A•− and a cation-radical D•+. If A andD are ionic species they decrease and increase the oxidation state, respectively.

The electron transfers to A and from D should occur at a comparable rate,otherwise, the semiconductor typically undergoes reductive or oxidation photo-corrosion, depending on the carrier type that gets accumulated in the crystal. Thecharacter of secondary (“dark”) processes depends on the nature and reactivity ofthe intermediary A•− and D•+ species generated in the primary charge transferevents. The dark stages include the formation of neutral free radicals or stable finalproducts, their interaction and reactions with original A and D species, etc.

1 Basic Concepts of the Photochemistry of Semiconductor … 3

Apart from the selection of A and D species, the design of a photochemicalsystem based on bulk semiconductors includes also the selection of a semicon-ductor that absorbs in an appropriate spectral “window”. For example, the solarlight harvesting applications require the absorption spectrum of the semiconductorto be maximally overlapping with the solar irradiation spectrum near the Earthsurface, that is, the materials with Eg range of 1.5–2.5 eV (see Table 1.2) [1]. Inother applications, such as photopolymerization and photolithography, the lightsensitivity range of semiconductors may be limited by the UV spectral domain(Eg = 3–4 eV, Table 1.2).

Another criterion for the selection of photo-active semiconductors for photo-catalytic applications, is the chemical/photochemical stability of the semiconductor.For example, ZnO dissolves both in acidic and alkaline media and can corrodeinteracting with the products of photo-catalytic reactions. Also, it easily transformsinto ZnS upon a contact with sulfide anions and thus additional precautions shouldbe taken when using ZnO-based materials in liquid-junction solar cells withsulfide/polysulfide electrolytes. Cadmium sulfide is one of the “universal” photo-catalysts and sensitizer materials for solar cells. However, it is prone to oxidativephotocorrosion resulting in dissolution (CdS + 2O2 = CdSO4) when exposed toillumination in tha absence of strong electron donors, such as Na2S or Na2SO3.

Finally, when large-scale applications are anticipated for thesemiconductor-based photo-chemical systems, the factors of a low cost and avail-ability as well as a low toxicity can be weighted and considered as well. The latter

Table 1.2 Band gap Eg andthe absorption band edposition (kbe) of some bulksemiconductors [7, 10, 26]

Semiconductor Eg, eV (approx. kbe, nm)

CdS (cubic) 2.4 (520)

ZnS (cubic) 3.6 (350)

ZnS (hexagonal) 3.8 (330)

PbS 0.41 (3030)

In2S3 2.0 (620)

Bi2S3 1.3 (1000)

MoS2 1.23 (1010)

MnS 3.0 (415)

CuInS2 1.55 (800)

CdSe 1.74 (715)

AgCl 3.3–3.5 (355–380)

TiO2 (anatase) 3.2 (390)

TiO2 (rutile) 3.0 (415)

ZnO 3.2 (390)

Fe2O3 2.0–2.2 (565–620)

SnO2 3.5 (355)

CeO2 3.4 (365)

BiVO4 2.4 (520)

Bi2MoO6 2.6 (480)

4 1 Basic Concepts of the Photochemistry of Semiconductor …

factor largely undermines possible applications of CdS- and PbS-based systems, aswell as recently emerging Pb-based organo-inorganic perovskite materials for solarcells [18, 19]. The CuInS2-based materials have almost perfect bandgap for thesolar light harvesting, but they contain relatively rare indium and therefore, alter-natives are constantly probed using Earth crust abundant materials for solar cellabsorbers, such as kesterite Cu2ZnSnS4 [20–22], that combines availability, lowtoxicity and a bandgap matching to the solar spectrum.

As a result of the above-discussed criteria, that sometimes become contradictory,the selection of appropriate semiconductor photocatalysts or solar cell absorbermaterials is often a complex task that is still to be fulfilled in a satisfactory manner.A typical example is the photo-catalytic systems for the stoichiometric watersplitting, where there is no “universal” semiconductor photocatalyst that combinessimultaneously a high photoactivity, appropriate band gap and CB/VB positionsand a high chemical stability (see a detailed discussion of such systems in Chap. 2).

Going down to nanometer dimensions of semiconductor crystals opened newpossibilities for the design of semiconductor-based light conversion systemsassociated both with the size/volume ratio effects and with the changes of funda-mental photophysical/electrophysical properties of semiconductors.

The transition from microcrystalline to nanocrystalline state of semiconductorsresults in a number of quantitative changes, including an increase in the totalsurface area and the surface-to-volume ratio and an increase in the population ofvarious defects—dangling bonds, dislocations, undercoordinated atoms, vacancies,adventitious doping, etc. These changes, that can be designated as geometrical/morphological ones, affect invariably the physical and chemical properties ofsemiconductor particles. In particular, an increase of surface atoms from around 1%for a 10-nm particle to *50% for 1–2-nm NP, can influence strongly the ther-modynamic properties of the NP as a whole, such as melting and phase transitiontemperatures, heat capacitance, solubility, etc.

Formation of various surface defects impacts the adsorption capacity and,therefore, the catalytic and photocatalytic processes with the participation ofsemiconductor NPs that obligatorily include intermediary steps of the reactantadsorption and product desorption. Besides, the NP surface defects can participatedirectly in the interfacial charge transfers and act as recombination sites thusinfluencing the photochemical and PL properties of semiconductor NPs. Thenano-dispersed semiconductors can sometimes crystallize in the phases unstable forthe bulk counterparts (phase size effect) and the relative stability of various latticepolymorphs and phase transition temperatures can differ drastically for bulk andnanocrystalline semiconductors. The size effects can also affect magnetic propertiesof semiconductor NPs, electric conductivity, diffraction of X-rays, Raman scatter-ing and other properties.

As the crystal size comes into the range of *10 nm and smaller, theabove-mentioned morphological effects are joined by the quantum size effects—[7,14, 23–25] altering the fundamental electron structure of the semiconductor. First,an expansion of the bandgap is observed, resulting in a corresponding increase ofthe photogenerated charge carrier potentials. As the NP size decreases, the

1 Basic Concepts of the Photochemistry of Semiconductor … 5

continuous energy bands transform into sets of quantized discrete occupied andvacant levels.

This phenomenon can be readily visualized by imagining the reverse process offormation of semiconductor NPs from molecular species, for which the electrontransition energy should be much higher than in the bulk solids. Figure 1.2 illus-trates schematically a change of the excitation energy for a typical semiconductorphotocatalyst—cadmium sulfide during the transition from molecular [Cd(SR)4]

2−

ion (SR—aliphatic thiol) to larger polynuclear [Cd10S4(SR)16]4− and

[Cd32S14(SR)36] clusters, then to *5-nm CdS NPs and, finally, to the bulk CdS. Asthe number of structural CdS units increases, the number of binding andnon-binding orbitals increases as well. As a result, the CdS NPs with discrete(quantized) energy levels appear in an intermediary size range, then the distancebetween the quantized levels decreases gradually and, finally the adjacent levelsmelt into continuous energy bands, typical for larger CdS NPs and bulk cadmiumsulfide crystals.

The photochemical reactions with the participation of both bulk and nanocrys-talline semiconductors start from the light absorption and formation of the elec-tronically excited state. The above discussion shows that the character of lightabsorption can be strongly affected by size effects. The following section introducesthe reader to the basics of light absorption phenomena in semiconductors.

1.1 Light Absorption by Bulk and NanocrystallineSemiconductors

The electromagnetic irradiation excites both atomic and electronic subsystems of asemiconductor crystal. The extinction of a light flux as a result of absorption in thesemiconductor crystal can be described by the Lambert-Beer equation I = I0e

−al,

Fig. 1.2 Size-dependent variation of the electron excitation energy for CdS-based molecular andnanocrystalline species. The energies of electron transitions for molecular ions and clusters areprovided in [156], the bandgap of nanocrystalline CdS taken from [64]

6 1 Basic Concepts of the Photochemistry of Semiconductor …

where I0 and I is an incident light intensity and the light intensity at a distancel from the crystal surface, respectively, a is a linear absorption coefficient showingthe probability of light absorption per a length unit (cm−1).

In the case of optically transparent colloidal semiconductor NPs the absorptioncoefficient can be expressed as a = ecl, where e is the molar absorption coefficient,c is the molar semiconductor concentration (in moles per L), l is the optical pass inthe cuvette.

Principal types of electron transitions in semiconductor crystals are summarizedin Fig. 1.3a.

Fundamental absorption. Transition type 1 corresponds to the fundamentalabsorption in the semiconductor crystal and results in the generation of free chargecarriers—a conduction band electron (e�CB) and a valence band hole (hþ

VB). Thistransition can occur if the light quantum energy is larger than or equal to thebandgap of the semiconductor, hv � Eg.

The fundamental absorption can originate from the electron transition of twotypes—direct transitions and indirect transitions. The absorption intensity isdetermined by the transition probability which can be assessed by the selectionrules [7]. The direct (or vertical) interband transitions occur in the semiconductorshaving the lowest points of the potential curves E(k) on VB and CB, where E andk are the energy and the quasi-impetus of the electron, one above the other(Fig. 1.3b, transition 1). For the realization of the direct electron transitions, thelight quantum energy should be equal to or higher than the direct bandgap.

The indirect electron transitions can occur in semiconductors having displaced(in the k space) minimums of the potential curves of the ground and excited states.The electron comes from the VBmaximum to the CBminimum (Fig. 1.3b, curve 1//).The indirect electron transitions require an additional energy supplied from the

Fig. 1.3 Photoinduced electron transitions in a semiconductor. In a 1—fundamental absorption;2, 3—absorption on free charge carriers; 4–7—absorption on defects/impurities; 8—intra-bandgapabsorption; 9—exciton absorption; 10—absorption as a result of exciton dissociation. In b direct(1) and indirect (1/, 1//) electron transitions induced by the fundamental light absorption

1.1 Light Absorption by Bulk and Nanocrystalline Semiconductors 7

vibrational energy pool of the lattice, that is by the vibrational quanta of lattice—phonons with typical energies of a single phonon varying from 20 to 70 meV [7, 8,26]. Thus, the direct electron transitions involve two particles (electron and hole)and at hv = Eg result only in the electron excitation of the semiconductor. Theindirect transitions involve three particles (electron, hole, and phonon) and requireboth light and vibrational energy. As a process involving three particles is muchless probable than a process with the participation of only two particles, the proba-bility of indirect electron transitions and the intensity of corresponding absorptionbands is much lower than the corresponding parameters of a direct transition. Indirectelectron transitions can be realized also in the direct-bandgap semiconductors(Fig. 1.3b, transition 1/) as a result of absorption of light quanta with the energy muchhigher than Eg.

Absorption on free charge carriers. The photoexcitation of electrons and holesdelocalized in the corresponding energy bands results in the absorption of freecarriers and transitions within the range of electron states available in the corre-sponding bands (Fig. 1.3a, transitions 2 and 3). The free-carrier absorption bandsare continuous and reside in the IR range of the spectrum.

Absorption on defects/impurities. The light absorption can occur as a result oflocalized electron transitions from defect- or impurity atom-related levels into theconduction band (Fig. 1.3a, transitions 4 and 5) or from the valence band—on thelocalized intra-bandgap levels (transitions 6 and 7). The defect/impurity absorptioncan be observed as a “tail” below the absorption band edge (at hv = Eg). As thedensity of impurity/defect-related states is much lower than the density of states inthe allowed bands, the absorption coefficients of the impurity/defect-related bandsare typically by several orders of magnitude lower than corresponding coefficientsof the fundamental absorption. If the concentration of donor and acceptor defects inthe semiconductor is relatively high, the so-called donor-acceptor couples can formthat can absorb light resulting in the donor-acceptor electron transitions (Fig. 1.3a,transition 8).

Exciton absorption. The photoexcited electron can come free into the CB, or,alternatively, remain bound with the hole by the Coulomb interactions forming ahydrogen-like e−…h+ quasi-particle or exciton (Fig. 1.3a, transition 9) that canmigrate through the crystal. The exciton has an own discrete set of levels situatedbelow the CB bottom. The exciton can dissociate either as a result of lightabsorption or under the influence of the thermal lattice energy. Upon the excitondissociation the electron becomes free and delocalized in CB (Fig. 1.3a, transition10). The exciton radius can be estimated using Eq. (1.4) which is similar to theBohr equation for the hydrogen atom:

aB ¼ �h2ee2

1m�

eþ 1

m�h

� �; ð1:4Þ

8 1 Basic Concepts of the Photochemistry of Semiconductor …

where ћ is the reduced Planck constant (h/2p); e is the dielectric constant of thesemiconductor; e is the electron charge; m�

e , m�h are the effective masses of CB

electron and VB hole.The effective masses do not correspond to any measure of the inertia of the

charge carriers and reflect the influence of the periodic potential of the semicon-ductor lattice on the movement of charge carriers. The effective masses of electronand hole depend on the semiconductor composition (Table 1.3) and typically arepresented as a portion of the electron rest mass me.

The doubled aB can be regarded as a “borderline” NP diameter for the QSEs tobe observable. As mentioned earlier, two regimes of the spatial exciton confinementcan be distinguished—weak confinement at 2aB < d � k, where k is the averageexcitation wavelength) and strong confinement at d < 2aB. In the former regime,the exciton experiences spatial confinement but the movement of charge carriersstill results in a displacement of the mass center of the exciton (i.e. the NP size islarger than the exciton diameter). In the strong confinement regime, when the NPsize becomes smaller than the exciton diameter, the charge carriers move at a steadyexciton mass center resulting in a strong dependence of the electron properties ofNPs on their size.

The spatial exciton confinement in semiconductor NPs results in an increase ofthe exciton energy and bandgap (Eg

nano) as compared to the bulk material (Egbulk).

The size-dependent energy gain (DE = Egnano − Eg

bulk) can be estimated by using aneffective mass approximation (EMA) based on the assumptions of parabolic bandedges and size-independ effective masses [7, 8, 16, 25, 27, 28]:

DE ¼ p2�h2

2R2

1m�

eþ 1

m�h

� �� 1:786e2

eR� 0:248R�

y ð1:5Þ

The first term in Eq. (1.5) depends on R2 (R is the NP radius) and corresponds tothe exciton energy increase due to the spatial confinement in a “potential box”, thatis in the NP volume. The second term describes the energy of Coulombic inter-actions between electrons and holes and increases in a reverse proportion to the NPsize. R�

y is a Rydberg energy accounting to the correlation of the electron and hole

Table 1.3 Effective electron(m�

e /me) and hole (m�h/me)

masses for somesemiconductors [7, 26]

Semiconductor m�e m�

h

CdS 0.2 0.8

ZnS 0.27 0.58

PbS 0.1 0.1

Ag2S 4.55 7.8

CdSe 0.13 0.44

ZnSe 0.17 0.06

PbSe 0.05 0.05

CdTe 0.11–0.14 0.35–0.8

TiO2 *30 *3

ZnO 0.27 0.50

1.1 Light Absorption by Bulk and Nanocrystalline Semiconductors 9

movement. The two latter terms in Eq. (1.5) counter-weight the size-dependent Eg

increment, however, in the case of medium and strong confinement the first termdominates and the latter two terms are typically neglected.

The results of experimental and theoretical studies of size dependences of DE forvarious semiconductors, in particular, for CdS [16, 29–34], CdSe [33, 35–37],CdTe [33, 38–40], ZnO [41–44], PbS [45, 46], PbSe [46–48], etc. showed thatEMA and Eq. (1.5) describe adequately only the case of weak exciton confinementin semiconductor NPs.

The semiconductor NPs in the strong confinement regime experience a gradualtransformation of bulk-like continuous band structure into sets of molecular-likediscrete energy levels. Such NPs are similar to large molecular clusters andtherefore are often referred to as semiconductor nanoclusters [13, 15]. For theseNPs the basic EMA assumption of the parabolic bands is not valid anymore andEMA fails to predict adequately the size-dependence of Eg. For a correct descrip-tion of the size dependences of electronic parameters of semiconductor nanoclustersother models are applied instead of the infinitely deep potential well model, inparticular, the model of a finite-depth potential well or various semi-empiricalquantum chemical calculations [32, 44, 48–53]. Therefore, the EMA model shouldbe applied with a caution and when the discrepancy between predicted andexperimental values becomes too high it is better to use empirical calibration curvesplotted on the basis of numerous measurements by electron microscopy, X-raydiffraction, and other techniques. Examples of such calibration curves are discussedin Chap. 6.

An increase of the bandgap due to the QSEs can be observed in the absorptionspectra of semiconductor NPs as a “blue” (hypsochromic) shift of the absorptionband edge (kbe), which is the larger the smaller NP size is. Figure 1.4 exemplifies

Fig. 1.4 Normalized absorption spectra of colloidal CdS NPs (a) and CdSe NPs (b). The NP sizeis a 8 nm (curve 1) and 1.8 nm (curve 2); b 2.5 nm (curve 1) and 1.8 nm (curve 2)

10 1 Basic Concepts of the Photochemistry of Semiconductor …

this phenomenon for two photo-active semiconductors—CdS [54–56] and CdSe[57], showing strong size-dependent blue kbe shifts, especially for ultra-small NPs(d < 2 nm).

The blue shift of the absorption band edge is observed for all semiconductor NPsexperiencing the spatial exciton confinement, its magnitude depending on the NPsize d and the electron and hole effective masses. The highest bandgap incrementsDE are typical for semiconductor NPs with a small m�

e (m�e � 1) and therefore,

relatively large aB (aB � 2–5 nm). For example, in the case of PbSe (m�e = 0.05me

[58], aB = 4–6 nm [59]) DE reaches a record value of 2.8 eV (relative to the bulkmaterial) as the NP size in decreased to 2–3 nm [58]. A considerable, up to 2 eV,increment of the bandgap can be achieved by reducing the size of PbS NPsfrom *20 to 2 nm [60]. A moderate Eg increment (DE = 0.1–0.2 eV) is typical forthe semiconductors with small exciton radii (see Table 1.1)—CuCl (aB = 0.7 nm[7]), PbCl2 (aB = 1.9 nm [7]), TiO2 (aB = 0.8–1.9 nm [61]), ZnS (aB = 1.5 nm[62]), etc.

The spatial confinement of electron/hole movement in semiconductor NPsresults also in an increase of the energy of Coulomb interaction between electronand hole in the exciton, that is, to an increase of the exciton binding energy Eex ininverse proportion to the NP radius R [7, 8, 63]:

Eex ¼ e2 � 3eRð Þ�1 ð1:6Þ

The exciton binding energy in bulk semiconductors does not exceed severalmeV and the exciton can easily dissociate under the influence of lattice vibrations:ðe�. . .hþ Þ ! e�CB þ hþ

VB. By this reason, the absorption bands corresponding to theenergy states of the exciton can be observed in bulk crystals only at very lowtemperatures. According to Eq. (1.6), a decrease of the NP size is accompanied byan increase of the exciton binding energy and Eex can become larger than thevibrational lattice energy (kT 25 meV at 300 К) already in the weak excitonconfinement regime, the NPs revealing excitonic absorption peaks even at roomtemperature (Fig. 1.4).

The intensity of light absorption by semiconductor NPs is also affected by thespatial exciton confinement. As Fig. 1.4 shows for CdS and CdSe NPs, a decreaseof the NP size is accompanied by the “concentration” of absorbance within theexcitonic peak which becomes more and more narrow as the NP size is reduced.This effect originates from increased overlapping of the wave functions of electronand hole and a corresponding increase in the exciton generation probability. Inspectroscopic terms, this effect results in an increase of the oscillator strength of thefirst excitonic transition that can be calculated using Eq. (1.7) [7, 8]:

f ¼ 2m�e

�h2DE lj j2 UðRÞj j2; ð1:7Þ

where DE is the transition energy, |l|2 is the dipole transition moment, |U(R)|2 is theelectron and hole wave function overlapping factor proportional to (aB/R)

3.

1.1 Light Absorption by Bulk and Nanocrystalline Semiconductors 11

Equation (1.7) anticipates an increase of the oscillator strength of the excitonictransition proportionally to a decrease of the NP volume (R3). This model foundexperimental evidence, in particular, in the optical properties of size-selected CdSNPs [30, 31, 33, 36, 56, 64]. Figure 1.5 shows that a decrease of the CdS NP sizefrom around 5 to 2 nm results in an increase of the oscillator strength of theexcitonic transition by more than an order of magnitude. Similar results werereported for CdSe NPs [33, 63, 65–67] and CdTe NPs [33, 39].

As the bandgap of semiconductor NPs increases with a size decrease the posi-tions of CB and VB levels change as well shifting to more negative and to morepositive values, respectively. The shifts—DECB and DEVB can be estimated for agiven NP diameter d by using Eqs. (1.8) and (1.9) [9, 16].

DECB ¼ h2

2m�ed

2 ð1:8Þ

DEVB ¼ h2

2m�hd

2 ð1:9Þ

It is obvious that for the quantum-sized semiconductor NPs the free energy ofelectron/hole transfer (Eq. 1.3) depends on the NP size. Therefore, the NP sizebecomes an additional “fitting” parameter allowing for varying the energy levelalignment in the photochemical system without actual changes in its chemicalcomposition.

Up to date, a considerable massive of experimental evidence was accumulatedon the size dependences of the photochemical activity of many semiconductor NPs[9–12, 14]. The reported results showed that the QSEs can not only accelerate thephotochemical/photocatalytic processes with the participation of semiconductorNPs but can even result in an expansion of a range of semiconductor materials thatcan be used as photocatalysts as well as of substrates that can be involved inphotochemical transformations for a given semiconductor.

Despite the fact that an increase in the photochemical activity is observed almostroutinely when nanocrystalline semiconductors are used instead of bulk

Fig. 1.5 Relative oscillatorstrength of the first excitonictransition (f/fex) of CdS NPson the NP radius R and R−3

(insert); fex is the oscillatorstrength of the first excitonictransition in bulk CdScrystals, fex = 0.0256 [157]

12 1 Basic Concepts of the Photochemistry of Semiconductor …

counterparts, the interpretation of this phenomenon is often fragmentary or biased.In particular, the acceleration of photochemical reactions is typically associatedonly with the size-dependence of CB/VB energies. At that, other important factorsthat can affect both thermodynamics and kinetics of the interfacial charge carriertransfers are discarded or not taken into account. A large volume of data on the sizedependences of photophysical and primary photochemical processes that is accu-mulated to date using spectral and kinetic methods, in particular, by the PL spec-troscopy and the flash photolysis, is used quite rarely in the interpretations of thesize dependences of photocatalytic and/or photo-electrochemical activity ofnanocrystalline semiconductor materials. In this view, the present book is inten-tionally focused on the potential of spectroscopic methods for the studies ofnanocrystalline semiconductors in an attempt to fill the gap between photophysicsand photochemistry of semiconductor NPs (see Chap. 6 for details).

The studies of semiconductor NPs by means of the stationary and time-resolvedPL measurements showed that the photogenerated free charge carriers get “trapped”quite rapidly by various defects of the NP lattice. As the NPs are characterized by ahigh surface-to-volume ratio and largely disordered surface with a lot of under-coordinated surface atoms, the most part of the lattice defects belong to the NPsurface. The defect-related electron states are located in the bandgap and have,therefore, a localized character. Depending on the distance (in terms of energy)between the defect-related states and CB/VB edges one distinguishes between“shallow” and “deep” charge traps. The trapping of charge carriers results in adecrease of their energy depending on the trap “depth” and the carriers do not movefreely anymore and become localized. Further electron-hole recombination pro-cesses can occur either at the encounter between freely moving electron/hole and atrapped hole/electron or via the electron tunneling between the trapped electronsand holes.

As the charge carriers get trapped their chemical potential decreases as comparedto original ECB/EVB level and, therefore, the trapping can affect strongly furtherchemical reactions with the participation of the photogenerated electrons and holes.Typically, the trapping of carriers in semiconductor NPs occurs very fast, in afemtosecond/picosecond time range and, as a result, almost exclusively trappedcarriers participate in the chemical reactions while having electrochemicalpotentials different from those that one can expect from Eqs. (1.8) and (1.9).However, despite the obvious influence of the charge trapping on the photo-chemical activity of semiconductor NPs only scattered reports on the analysis ofpossible consequences of this phenomenon on the photochemical properties of NPscan be found.

The phenomenon of photoinduced charging of semiconductor NPs is anotherexample of a QSE that draws undeservedly low attention with respect to thephotochemical NP behavior, despite quite elaborate studies of this effect by spectraland kinetic methods. As the NP size decreases the continuous energy bandstransform into sets of discrete levels. If the rate of interfacial transfer of electrons ismuch lower than the rate of hole transfer the discrete levels in CB get filled withexcessive electrons and thus the photogeneration of new electron-hole pairs requires

1.1 Light Absorption by Bulk and Nanocrystalline Semiconductors 13

a higher energy, as compared to Eg. Simultaneously with the charging-inducedincrease in the optical bandgap, the surface traps become filled as well thusresulting in inhibition of the electron-hole recombination and in a considerableincrease of the potential of the surface double charged layer. All these factors arefavorable for the interfacial electron transfer and produce a unique photo-chemicalbehavior of charged NPs differing strongly from that of “regular” non-charged NPs.A vivid example of such behavior is the participation of the photocharged semi-conductor NPs in the processes prohibited thermodynamically for regular NPs(discussed below).

1.2 Influence of Surface States on the PhotochemicalProperties of Semiconductor NPs

The surface states of semiconductor NPs influence both radiative electron-holerecombination processes and the interfacial electron transfers. Both types of pro-cesses are competing and, therefore, by observing the evolution of PL properties ofsemiconductor NPs we can make some conclusions on the photoinduced chargetransfer dynamics and the photochemical reactions in general. The discussion ofthese interrelations between PL and charge transfer requires a concise characteri-zation of the PL phenomena prior the examination of the role of surface states in thephotochemistry of semiconductor NPs.

The photoexcited semiconductor NPs relax to the ground state by severalcompeting routes, in particular by non-radiative electron-hole recombination (1.10),charge carrier trapping by NP defects (1.11, 1.12), direct radiative recombinationbetween free charge carriers (1.13), and interfacial transfer of free carriers toacceptor and donor substrates.

e�CB þ hþVB ! n phð Þ ph�phononð Þ ð1:10Þ

e�CB ! e�tr e�tr�trapped electron� � ð1:11Þ

hþVB ! hþ

tr hþtr �trapped hole

� � ð1:12Þ

e�CB þ hþVB ! hvlum hvlum�PL quantumð Þ ð1:13Þ

The photoexcitation of semiconductor NPs as small as *2aB results in thegeneration of charge carriers both in the NP volume and on the NP surface [68, 69].The charge carriers produced in the NP volume can then migrate to the defect-richsurface and get captured by the surface traps. The migration process is typicallyvery fast, for example around 10 fs for 3–5-nm CdS NPs [68]. The charge carrierscan be trapped (localized) by the lattice defects, undercoordinated surface atoms,admixtures and adsorbed substrates. The charge trapping results in the generation ofactive species, typically of a radical (ion-radical) nature that can participate in the

14 1 Basic Concepts of the Photochemistry of Semiconductor …

following redox reactions on the NP surface. For example, the electron trapping onthe TiO2 NP surface produces Ti3+ ions that can be detected by the electronparamagnetic resonance (EPR) and optically, by a characteristic absorption bandwith a peak at 600–900 nm (see Chap. 6) [70–80]. In oxygen-containing systems,the electron can be trapped on the NP surface as a O2

•− anion-radical [72, 81] whichis one of the main actors in numerous photocatalytic processes of the oxidation oforganic compounds. The trapped hole can exist on the TiO2 NP surface in the formof a OH• radical [81] or a O•− radical [70–72] that can also be detected by EPR [70,72, 75–77] and by characteristic absorption bands [71, 79, 80]. The hole trapping inmetal-sulfide NPs produces S•− anion-radicals with characteristic absorption fea-tures [82, 83].

A special feature of semiconductor NPs differing them from bulk counterparts isa very short time of the charge carrier migration to the NP surface (sm), as comparedto the characteristic electron-hole recombination time (sr). For example, sm can beas small as 10 ps for a 10-nm TiO2 NPs, while sr is by several orders of magnitudelarger—around 100 ns [14, 84]. Owing to such drastic difference in the charac-teristic times, the primary separation and trapping of the photogenerated electronsand holes in NPs occur very efficiently. As the NP size is increased from *10 nmto *1 lm, the migration time (which depends on R2) grows to around 100 ns [14,84] and the micro-crystal thus loses the favorable conditions for the charge sepa-ration existing in the case of a nanocrystal. By this reason, an external electric fieldshould be applied to the micro-crystalline semiconductors for the charge migrationto be competitive to the recombination.

The semiconductor NPs also reveal a different structure of thesemiconductor/electrolyte interface as compared to the corresponding bulk mate-rials. A typical length of a depleted layer in the semiconductor microcrystals is byan order of magnitude larger than the linear size of quantum-sized semiconductorNPs [14] (Fig. 1.6). A distribution of the potential from the center of a semicon-ductor crystal to the distance r from the center, Du(r), can be described byEq. (1.14) [14]

DuðrÞ ¼ kT6e

r � ðr0 �WÞLD

� 2

1þ 2ðr0 �WÞr

� ; ð1:14Þ

where r0 is the crystal radius, W is a band bending area length (Fig. 1.6); k isBoltzmann constant, T is temperature, LD is a Debye length depending on thecharge density ND, LD = (e0ekT/2e

2ND)1/2 (e, e0 is the dielectric constant of semi-

conductor and electrolyte, respectively).For low-doped semiconductors with e * 10, LD is on the order of 102 nm [85].

Therefore, for a 1–10-nm particle r0 � LD and Eq. (1.14) can be simplified toEq. (1.15), indicating that Du is the same both on the surface and in the volume ofsemiconductor NPs, and no appreciable band bending takes place, that can impedethe interfacial charge transfer (Fig. 1.6).

1.2 Influence of Surface States on the Photochemical … 15

Du ¼ kT6e

r0LD

� 2

ð1:15Þ

The lack of the band bending on the semiconductor/electrolyte interface resultsin a dramatic acceleration of the charge transfer from semiconductor NPs, ascompared to the bulk crystals of the same composition. For example, hole migrationto the surface of colloidal TiO2 NPs and the interfacial transfer to SCN− ions occurwithin 50 fs after the photoexcitation [86]. The photoinduced electron transfer fromCdSe and CdS NPs to adsorbed methylviologen (4,4/-dimethylbipyridyl cation,MV2+) requires 70 fs and 200–300 fs, respectively [87, 88].

The charge transfer competes with the radiative electron-hole recombination thatcan occur via two different mechanisms [69, 89]. The first route of PL generation isthrough direct interband recombination of the photogenerated carriers. As thisrecombination occurs very often between the exciton-bound charge carriers, this PLtype is typically referred to as the excitonic PL [process (1.13)]. The excitonic PLband maximum position is close to the absorption band edge and approximatelycorresponds to Eg (Fig. 1.7a). For the direct-bandgap semiconductor NPs, theexcitonic PL decays in a nanosecond time range and reveals a quantum yield of10−3–100 depending on the NP composition and synthesis mode.

The second-type PL originates from the recombination of free charge carrierswith counterparts trapped by the NP lattice defects—either between a free hole anda trapped electron (1.16) or between a free electron and a trapped hole (1.17)[69, 90].

Fig. 1.6 Potential variation on the semiconductor/electrolyte interface for semiconductormicrocrystals (a) and nanocrystals (b)

16 1 Basic Concepts of the Photochemistry of Semiconductor …

e�tr þ hþVB ! hv=lum ð1:16Þ

e�CB þ hþtr ! hv==lum ð1:17Þ

The energy states corresponding to the traps reside in the bandgap—the electrontraps De are located lower than the CB bottom, while the hole traps Dh typically areabove the VB top. As a result, the bands of defect-related (or donor-acceptor, DA)PL are typically shifted to lower wavelengths as compared to kbe (Fig. 1.7a). As thetrap states can differ by “depth” and the local surrounding, the trap state energyspectrum is typically quite broad and mirrored by a large spectral width of thedefect-related PL. The DA luminescence emission at room temperature is a phe-nomenon typical for semiconductor NPs with a high and disordered surface area,while for the bulk semiconductors the DA PL can typically be observed only incryogenic conditions [7].

Mechanisms and dynamics of the defect-related PL depend strongly on thesynthesis conditions, size and surface chemistry of semiconductor NPs. Therecombination is supposed [90–93] to occur at an encounter of two opposite chargecarriers, of which one is localized in a deep trap (De/h > kT), while other migratingin CB/VB or trapped by “shallow” traps (De/h � kT) that can be ionized under theinfluence of lattice vibrations. A shift between the maximum of excitonic PL band(or bandgap) and defect-related PL band characterizes a depth of the traps relativeto the corresponding band edges. If both charge carriers are trapped the PL origi-nates, most probably, from the electron tunneling between the electron and hole trapstates. Some examples of analytical extraction of trap energies from PL spectra ofsemiconductor NPs are discussed in Chap. 6.

Fig. 1.7 a Normalized absorption (curve 1) and PL (curve 2) spectra of colloidal 2.5-nm CdSeNPs [158–160]. b Absorption (curve 1) and PL (curves 2–4) spectra of colloidal 1.8-nm CdS NPs[54]. The PL spectra were registered in 5 ns (2), 20 ns (3), and 50 ns (4) after the laser pulse

1.2 Influence of Surface States on the Photochemical … 17

In real semiconductor NPs, both electron and hole have quite a broad spectrumof traps differing both by the depth (energy) and the distance r between electron andhole traps. The pairs of opposite charges that are closer to each other experience astronger Coulomb interaction than the pairs separated by a larger distance. Thedistance distribution affects the emitted PL spectrum according to Eq. 1.18 [8, 90]:

Elum ¼ Eg� Dh�Deð Þþ e2=er ð1:18Þ

The probability of the radiative recombination is inversely proportional to r and,therefore, the average distance between trapped electrons and holes graduallyincreases in the course of radiative recombination. As a result, the third member ofEq. (1.15) and the energy of emitted PL quanta gradually decrease. This effect canbe observed as a red shift of the PL band maximum in the course of PL decay(Fig. 1.7b).

The surface traps can sometimes affect the dynamics and even the mechanism ofphotochemical and photocatalytic reactions in a rather decisive manner. Forexample, CdSe NPs can act as a photocatalyst of the one-electron reduction ofMV2+ to cation-radical as well as of the reduction of MV+• to a neutral form MV0

[94]. At the same time, CdxZn1−xS NPs with roughly the same CB potential (atx = 0.25) can photocatalyze only the first of these processes and reveal no activityin the photochemical generation of MV0. A detailed analysis of PL properties ofboth CdSe and CdxZn1−xS NP presented in Chap. 6 allowed to conclude that thedifference in the photochemical behavior originates from a different depth of theelectron traps in both semiconductors. Thus, the electron trapping in CdSe NPsdecreases the electron energy only slightly and does not impede it from the inter-facial transfer to MV+•, while in the case of CdxZn1−xS NPs the electrons “fall” toodeeply into the traps losing considerably in the chemical potential and the capabilityof MV+• reduction [94].

Colloidal ZnS NPs cannot reduce CO2 to CO2−• despite the fact the CB potential

is negative enough for this process to occur [95]. This fact is interpreted as a resultof a deep trapping of the photogenerated electron resulting in a loss of energy ofaround 1 eV. As additional HS– ions are introduced into the system, they fill thesurface vacancies and block the trapping, and the ZnS NPs gain the ability to reduceCO2 at the expense of the oxidation of H2PO�

2 as a sacrificial donor [95]. Thisexample demonstrates vividly the possibility of influencing the dynamics ofphotochemical/photocatalytic processes by a proper modification of the semicon-ductor NP surface.

During the photocatalytic reduction of CdII and ZnII on the surface of CdS andZnS NPs, respectively, the ions adsorb selectively on the NP surface and createadditional electron traps capable of participation in the radiative recombination [96,97]. As a result, an increase in the CdII/ZnII concentration results in the deteriorationof photocatalytic activity of CdS (ZnS) NPs [96, 97]. Doping of CdS and CdxZn1−xS NPs by small amounts of BiIII or CuII introduces additional deep electron trapsmediating and increasing the electron transfer from NPs to molecular oxygen [98].

18 1 Basic Concepts of the Photochemistry of Semiconductor …

These examples show that an intentional modification of the surface of semicon-ductor NPs via adsorption or implantation of ionic species can be used to influencethe dynamics of the photochemical processes on the NP surface.

1.3 Influence of Size Dependences of CB and VB Levels

A basic condition of photochemical/photocatalytic processes with the participationof semiconductor NPs is a correspondence between the CB and VB levels and theredox potentials of the electron acceptors and donors adsorbed on the NP surface.

The VB potentials of typical photochemically active bulk semiconductors, suchas CdS (1.6 V versus normal hydrogen electrode (NHE) [99]), ZnS (1.8 V vs. NHE[95]), TiO2, ZnO, SnO2, WO3 (EVB > 2.5 V vs. NHE [1, 14]), and Fe2O3 (1.6 Vvs. NHE [100]) are relatively high. As a result, an increase of the EVB level inducedby QSEs, as a rule, does not affect strongly the photochemical activity of thesesemiconductors in oxidative reactions with the participation of VB holes. At thesame time, the CB potentials of many photoactive semiconductors are located onlyslightly above the NHE [1, 14] and even small variations in ECB can affect quitespectacularly their photochemical activity. Also, for most photo-active semicon-ductors m�

e < m�h, and, according to Eqs. (1.8) and (1.9), we can expect that a

size-dependent variation of the CB level will be much larger than the correspondingchange in the EVB potential.

Two very important consequences of the QSEs can be envisaged for semicon-ductor NPs as a result of the size-dependence of ECB and EVB levels, in particular,(i) demonstration of photochemical properties by NPs of a semiconductor that isabsolutely passive in the bulk form and (ii) enhancement of thephotochemical/photocatalytic activity of semiconductors with a decrease of thecrystal size. It should be noted, however, that the increase of CB and VB potentialper se does not guarantee realization of these two effects, because the photo-chemical activity of semiconductor NPs depends not only on the charge carrierenergy but also on the dynamics of primary photophysical/photochemical pro-cesses, recombination rate, and many other factors.

Photochemical activity of nanocrystalline semiconductors passive in the formof bulk materials. The photochemical activity and photocatalytic properties arerevealed by a comparatively narrow number of inorganic semiconductors [1] andmost reported semiconductor photocatalysts have a relatively wide bandgap (Eg >2.5 eV). At the same time, the range of reactions potentially possible for thenarrow-bandgap semiconductors (Eg < 2.5 eV) is limited by substrates with theredox-potentials intermediary between ECB and EVB levels of semiconductorphotocatalysts. Therefore, at a low Eg the primary photoinduced charge transfers arelimited and invariably characterized by a low free Gibbs energy (Eq. 1.3).

The bandgap expansion with a decrease of the size of semiconductor NPs canovercome these limitations. In particular, the QSEs can result in an expansion of the

1.2 Influence of Surface States on the Photochemical … 19

range of semiconductor photocatalysts due to the introduction of new materials, thathave no photochemical activity in the bulk form. For example, as the size of PbSeNPs is reduced from d > 100 nm to 5–10 nm, this semiconductor becomes sus-ceptible to the reductive photocorrosion with the formation of Pb0 and can alsoparticipate in the photocatalytic redox-processes thermodynamically forbidden forbulk PbSe, such as the MV2+ and water reduction [58]. Also, as opposite to the bulkmaterials, CdSe NPs smaller than 5 nm can act as a photocatalyst of the water andCO2 reduction [58].

Molybdenum disulfide is photocatalytically passive in the oxidation of phenoland its derivatives when introduced in the form of either bulk crystals and 8–10-nmparticles [101, 102]. At the same time, the formation of the oxidation products wasdetected chromatographically in the presence of 4–5-nm MoS2 NPs (Fig. 1.8a). Therise of photocatalytic activity of small MoS2 NPs was assigned [102] to asize-dependent increase of the VB potential because the generation of very activeOH• radicals is only possible for 4–5-nm MoS2 NPs (Fig. 1.8b).

This system can also be used for the illustration of another special feature of thephotochemistry of semiconductor quantum-sized NPs. A size-dependent increase ofthe photogenerated charge carrier energies, though being extremely positive for thephotoinduced charge transfers, is invariably accompanied by a blue shift of theabsorption NP threshold. Figure 1.8c shows that molybdenum disulfide loosesstrongly the ability for the visible light absorption as the NP size is reduced from 8–10 nm to 4–5 nm. This limitation has a general character—one should weight gainsin charge energies and losses in the visible light harvesting when designing aphoto-catalytic/photoelectrochemical system based on the quantum-sized semi-conductor NPs.

The photocatalytic reduction of benzophenone in acetonitrile can occur only inthe presence of CdS NPs smaller than 4 nm [103]. Cadmium sulfide reveals also a

Fig. 1.8 a Kinetic curves of the photocatalytic oxidation of phenol in the presence of MoS2 NPsand nanocrystalline TiO2 Evonik P25; b band positions for various semiconductors with respect tothe water oxidation redox-potential (at pH 7); c absorption spectra of colloidal MoS2 NPs ofdifferent sizes. Reprinted with permissions from [102]. Copyright (1999) American ChemicalSociety

20 1 Basic Concepts of the Photochemistry of Semiconductor …

photocatalytic activity in the reductive decomposition of sodium selenosulfate [104]and reduction of Ni(II) with sodium sulfide [105] when present in the form of6–8-nm particles, while bulk CdS is inert in these processes.

PbS NPs smaller than 3 nm reveal photocatalytic properties in the MV2+

reduction [106]. A similar effect is observed for CdSe and CdTe as the NP size isreduced to 3–5 nm [94, 107].

ZnS nanocrystals can be used as a photocatalyst of the CO2 reduction while bulkzinc sulfide reveals no activity in this reaction [95, 108]. Analysis of the band edgepositions of differently sized ZnS crystals [95] showed that the photoactivity of ZnSNPs originates from a size-dependent increase of the CB potential that becomesmore negative than the redox potential of CO2/CO2

•− reduction (−1.9 V vs. NHE).A size decrease of Si NPs from 3–4 to 1–2 nm renders them active in the

photocatalytic reduction of some organic dyes and CO2 [109]. As opposite to themicrocrystalline zinc oxide, ZnO NPs can initiate the methylmethacrylate pho-topolymerization [110]. Similarly, MnO2 NPs revealed photocatalytic properties inthe oxidative coupling of b-naphtol, untypical for the bulk material [111].

Acceleration of photocatalytic processes as a result of size-dependent increaseof the CB and VB energies. A phenomenon of the acceleration of photocatalyticprocesses with a decrease of the semiconductor crystal size is broadly observed. Inmany cases, this effect stems not only from an increase of the specific surface areaand the generation of surface defects participating in the interfacial charge transfersbut also from changes in the band edge energies due to the QSEs. As discussedearlier, an increase in absolute ECB and EVB values results in a correspondingincrease in the free energy of electron transfers, which affects the rate of photo-catalytic processes. For example, CdS NPs [112] and ZnS NPs [113] were found tobe much more efficient photocatalysts of the Rhodamine B degradation as com-pared to the corresponding bulk materials. A reduction of CdS and ZnS NP sizefrom 5 to 2 nm and from *3 to 1.6 nm, respectively, results in a *5-foldacceleration of the photocatalytic dehalogenation of polyhalogenated aromatics[114]. Zeolite-hosted In2S3 NPs revealed a much higher photocatalytic activity inthe hydrogen evolution from aqueous solutions, as compared to the bulk indiumsulfide (Fig. 1.9a) as well as a high stability and reusability due to NP-host inter-actions [115]. Similarly, CdS nanocrystals formed in or anchored to the zeolites andmesoporous hosts revealed an enhanced photocatalytic activity in the hydrogenevolution from aqueous sulfide/sulfite solutions as compared to bulk CdS [116–119]. A decrease of the CdS NP size from 5 to 3.8 nm, though being comparativelysmall, results in almost 40-fold acceleration of the photocatalytic reduction of somenitro-aromatic compounds (Fig. 1.9b) [120].

The specific rate of photoinduced buthylmethacrylate polymerization in thepresence of CdS NPs was found to depend on NP size increasing by 60–70% as theNP size is reduced from *8 to 3.8 nm [121]. This effect is a result of the directmonomer photoreduction by CB electrons which is possible only in the case of thesmallest 3.8-nm CdS NPs. For larger NPs, the photopolymerization can be initiatedonly by radicals generated via the oxidation of 2-propanol (solvent) with VB holes.

1.3 Influence of Size Dependences of CB and VB Levels 21

The photocatalytic methylviologen reduction rate was also found to dependconsiderably on the size of colloidal CdS NPs [103, 122, 123] and In2S3 NPs [123].A reduction of CdS NP size from 5 to 3 nm results in a *0.2 eV increment of theCB potential providing 4–5-fold acceleration of the photoinduced electron transferfrom CdS NPs to MV2+. The dependence of the photoinduced electron transfer rateconstant on ECB is linear when presented in the coordinates of Tafel equation,which is a typical relation between the rate of an electrochemical reaction and anover-voltage of the electrode charge transfer. In the current case, the over-voltageDE is provided by a difference between the size-dependent CB potential of CdSNPs and the MV2+/MV•+ pair redox-potential. Therefore, the Tafel equation can bewritten as lg(k/kbulk) = −aDE = −a(ECB(R) − E0(MV2+/MV•+)), where kbulk is therate constant for bulk CdS, a is a constant.

A study of the photocatalytic MV2+ reduction on the surface of a broad series ofsemiconductor NPs differing both by the composition and the size [94] showed thatsimilar Tafel-like dependences between the methylviologen reduction rate (orquantum yield of MV+• radical) and ECB of colloidal semiconductor NPs are typicalallowing to predict the efficiency of this process for any given NPs with the knownECB (Fig. 1.10a).

A Tafel-like dependence was also observed between the rate of the photocat-alytic nitrate reduction to NH3 and ECB of size-selected CdS NPs (Fig. 1.10b)[124]. The NPs studied in [124] belong to a very narrow size range of 2.0–2.2 nm,corresponding to the regime of strong spatial confinement. As a result of strongQSEs, a NP size reduction by mere 0.2 nm supplies an appreciable 0.25 V incre-ment of the CB potential and a rate increase by a factor of 5–6 [124].

A direct relationship between the electrochemical characteristics of the photo-generated charge carriers and NP size was experimentally proven for CdS NPs[125, 126] and Bi2S3 NPs [127]. In particular, as the CdS NP size is reduced from

Fig. 1.9 a Kinetic curves of the hydrogen evolution over platinized zeolite/In2S3 NPs (■),non-platinized zeolite/In2S3 NPs (▲), and platinized (♦) and non-platinized (▼) bulk In2S3 undervisible light illumination (k > 430 nm); b a ratio of rate constants of the photocatalytic andnon-catalytic reduction of nitrotoluene as a function of CdS NP size. Reprinted with permissionsfrom [115] (a) and [120] (b). Copyright (2006, 2008) Elsevier

22 1 Basic Concepts of the Photochemistry of Semiconductor …

4.5 to 3.9 nm the gap between anodic (oxidative) and cathodic (reductive) currentpeaks increases from 2.63 to 3.39 eV in line with a corresponding broadening of theoptical bandgap from 3.06 to 3.23 eV [125]. A similar correlation between asize-dependent increment of the bandgap and a distance between the anodic andcathodic current peaks was established also for colloidal CdTe NPs [128].

1.4 Photoinduced Charging of Semiconductor NPs

Under very intense illumination several electron-hole couples (excitons) can begenerated simultaneously in each semiconductor NPs. Interactions of two excitonsgive rise to various non-linear optical phenomena, their amplitude dependingnon-linearly on the light flux intensity. Typically, the non-linear effects dependquite strongly on the NP size [7, 8, 38, 63, 129–137].

Some of the non-linear effects can also be observed in the case of relativelylow-intensity excitation, for example, for the AM1.5 light flux. In particular, PbXNPs (X = S, Se) display a pronounced tendency to multi-exciton generation [138–141]. This phenomenon is observed when PbX NPs are excited at hv * nEg

(n = 3–10). As the thermalization of hot carriers is slow for small PbX NPs, theexcitation energy can be accommodated by the generation of several (at least two)electron-hole pairs and, thus the light harvesting efficiency of PbX NP-based cantheoretically be higher than 100–200% [140].

Another non-linear optical phenomenon can be observed when the rates ofinterfacial transfers of electrons and holes are strongly different resulting in the

Fig. 1.10 a A relationship between the quantum efficiency of the photocatalytic methylviologenreduction Ф(MV•+) and ECB of 5.0-nm ZnS NPs (point No. 1), 3.0-nm CdTe NPs (2), 5.5–5.6-nmCdSe NPs (3), Cd0.25Zn0.75S NPs (4), Cd0.50Zn0.50S NPs (5), Cd0.63Zn0.33S NPs (6), and CdS NPswith the size d = 6.5–6.6 nm (7), Cd0.75Zn0.25S (8), CdS NPs with d = 10–11 nm (9), and 4.8-nmZnO NPs (10). The solid line represents a linear fit of the presented data; b Tafel plot of the naturallog of the measured current density versus the reaction over-voltage for amine-capped CdS NPs.Reprinted with permissions from [94] (a) and [124] (b). Copyright (2010 a and 1997 b) Elsevier(a) and American Chemical Society (b)

1.3 Influence of Size Dependences of CB and VB Levels 23

population of semiconductor NPs with an excessive charge. Typically for the metalchalcogenide NPs, the VB hole trapping and subsequent reactions both withchalcogenide (oxide) lattice and adsorbed substrates are quite fast, while electrontransfer can be obstructed by many factors, such as a relatively low CB potential,resistance of NPs to the reductive corrosion, non-availability of suitable adsorbedacceptors, etc. As a result, the semiconductor NPs are typically populated withexcessive electrons that fill the available states near the CB bottom.

The excessive electrons are relatively long-lived and the NPs can be excitedmany times while being in the charged state. At that, the transition of each newelectron from VB to CB requires a higher energy, because a portion of the loweststates near the CB edge gets occupied by the excessive electrons (the so-called“electrons-spectators”) and this portion increases with an increase of the excessivecharge density. Therefore, the excitation of a charged NP requires a higher energy,than for the “normal” NP, Eg + DEB, where DEB—is an excess necessary to pushan electron to the nearest available free electron state in CB. Obviously, DEB

depends on the excessive charge density and, thus, on the light intensity andNP size (volume). This phenomenon is often referred to as Burstein-Moss effect[142–144], while DEB is called a Burstein shift [143, 144].

The Burstein-Moss effect was observed for the first time in strongly doped InSb[7]. Typically, bulk semiconductor crystals are moderately doped and have quite ahigh density of states near the CB edge. Therefore, an excessive charge density highenough to induce an appreciable optical shift cannot be achieved for bulk semi-conductors even at intense photoexcitation. The situation changes dramatically forthe semiconductor NPs, which have a tiny volume and partially quantized CB as aresult of the QSEs. When a strong light pumping is applied to excite thequantum-sized NPs, an excessive charge with a density of the order of ne *1026 m−3 can be created comparable to the free electron gas density of typicalmetals, *1028 m−3 [69, 143, 144].

A partial filling of CB states with excessive electrons can be observed as a blueshift of the fundamental absorption band edge (Fig. 1.11a) or as a negative“bleaching” band in differential absorption spectra of charged semiconductor NPs.The photoinduced blue shift of kbe as well as the intensity and spectral width of thenon-stationary bleaching (NB) bands of semiconductor NPs depend on the exces-sive charge density ne. When ne becomes comparable with the free charge densityin metals the Burstein-Moss effect can be described by Eq. (1.19) that relates theoptical shift DEB = EF − ECB with ne [142, 144].

DEB ¼ ð1þ m�e

m�hÞ h2

2m�e

3ne8p

� 32

�4kT

" #ð1:19Þ

In view of the obvious dependence of ne on the NP volume, the Burstein-Mosseffect in semiconductor NPs has a pronounced size-dependent character. For thesemiconductor NPs residing in the strong confinement regime, that is, at R < aB, ahigh DEB can be reached already at a comparatively moderate photoexcitation

24 1 Basic Concepts of the Photochemistry of Semiconductor …

power, even under the stationary illumination [56, 144]. Apparently, at an equal nethe shift DEB will increase with a NP size decrease. As kbe of quantum-sized NPsshifts to lower values with a size decrease, the NP band maximum reveals asize-dependent blue shift as well (Fig. 1.11b). Finally, as the rate of interfacialtransfer of excessive electrons depends also on the NP size, the NB relaxation ratealso reveals a size-dependence, the smaller NPs discharging faster (Fig. 1.11c) [56,64, 130, 143, 145, 146].

The relaxation of NB bands in the differential spectra (and the return of kbe to theoriginal position in conventional absorption spectra) corresponds to the interfacialtransfer of excessive charge to other components of the system—the solvent, dis-solved oxygen or other electron acceptors. For example, kinetic curves of the NPdecay presented in Fig. 1.11c reflect gradual consumption of the excessive elec-trons in the charge transfer to oxygen molecules: e− + O2 ! O2

•−. Oppositely, theNB amplitude increases, when additional electron donors are introduced into thesystem. For example, the introduction of Na2SO3, Na2S, N2H4 or (CH3)2CHOH, ainto aqueous CdS and CdxZn1−xS colloids results in 3–4-fold increase of the NBband intensity as a result of the efficient VB hole capture [56].

The excessive charge density depends on the light intensity and the NP size andcomposition. In the case of deaerated aqueous CdS colloids the charge accumula-tion results in the cathodic (reductive) photocorrosion and a partial transformationof CdS into metallic Cd [97, 99, 125] and, therefore, the Burstein-Moss effect forCdS NPs can be observed only under pulse photoexcitation in air- (oxygen-) sat-urated colloidal solutions, where the excessive charge is withdrawn by O2 after eachlight pulse, preserving the NP stability [56, 147]. In the case of colloidal ZnO NPslarger than 5 nm resistant to the reductive photocorrosion, the charge accumulation

Fig. 1.11 a Absorption spectra of colloidal ZnO NPs in ethanol prior to (curve 1) and after theillumination (curve 2), kexc = 310–370 nm, curve 3 is a difference between curves 1 and 2 [148,151, 161]; b NB bands of colloidal CdSe NPs with an average size of 2.7 nm (curve 1), 3.0 nm(2), and 3.2 nm (3) [104, 162]; c normalized kinetic curves of the NB decay in the NP bandmaximum (k = 345 nm) for colloidal ZnO NPs with an average size of 3.7 nm (curve 1) and4.4 nm (2) [161, 163]

1.4 Photoinduced Charging of Semiconductor NPs 25

results in a blue kbe shift even under the stationary photoexcitation (Fig. 1.11a) andpersists for many hours [148–150]. Using Eq. (1.19) one can show [150–152] thatthe case of DEB = 0.2 eV presented in Fig. 1.11a corresponds to the accumulationof 3–4 excessive electrons by each ZnO NP. The air admission into the illuminatedsolution results in the instant backward shift of kbe to the original position due tothe transfer of excessive electrons to oxygen. The same mechanism of NB relax-ation is valid for the case of pulse photoexcitation of air-saturated ZnO colloids(Fig. 1.11b). The charge transfer to oxygen occurs much faster for 3.7-nm ZnO NPsas compared with larger 4.4-nm NPs indicating a higher photochemical activity ofsmaller ZnO NPs.

The Burstein-Moss effect can result in a considerable enhancement of thephotocatalytic activity of semiconductor NPs [11, 12, 143–145]. In this view,studies of the characteristics and decay dynamics of NB band can provide uniqueinformation on the photoinduced charge transfer kinetics on the NP/electrolyteinterface. Some examples of the application of pulse photolysis and NB phe-nomenon for probing of the photochemical behavior of semiconductor NPs arediscussed in details in Chap. 6.

The accumulation of an excessive charge alters significantly the photophysicaland electrophysical properties of nanocrystalline semiconductors as well as thedynamics of interfacial charge transfer. In particular, it results in a cathodicpolarization of NPs, that is, in a change of the potential of the double electric layer(DEL) on the NP surface, especially of its dense Helmholzian component [143].A charging-induced increase of the optical bandgap can be used to calculate anexcessive charge density ne and an average number of excessive electrons per NP,Ne, from Eq. (1.19). Then, a charging-altered DEL potential E* can be estimatedusing the reported typical values of the DEL capacity C of colloidal semiconductorNPs (around 0.06–0.10 F/m2 [153]) as [143]

E� ¼ ECB þNe=C ð1:20Þ

The influence of the Burstein-Moss effect on the interfacial charge transfer canbe illustrated by photochemical processes occurring in mixed aqueous colloidalsolutions containing *10-nm CdS NPs and size-selected 3.6–6.6-nm CdTe NPs[154]. The pulse photoexcitation of such colloids results in the electron transferfrom CdS NPs to CdTe NPs evidenced by quenching of the NB band of CdS NPs.By using the above-discussed methodology, an average number of transferredelectrons per CdTe NP DNe can be estimated from the reduction in the NB bandintensity. As shown in Table 1.4, DNe increases with an increase of the CdTe NPsize. An energy level scheme for the CdS–CdTe system (Fig. 1.12a) shows that thephotoinduced electron transfer from the stationary CB level of CdS NPs to the CBlevel of CdTe NPs of any size meets a thermodynamic barrier. Therefore, it wassupposed [154] that electrons come to CdTe NPs not from the stationary CdS CBlevel, but from a non-equilibrium higher-energy state E* generated as a result of thephotoinduced charging of CdS NPs (Fig. 1.12a). Indeed, estimations performedusing Eq. (1.20) showed that the accumulation of an excessive charge on the CdS

26 1 Basic Concepts of the Photochemistry of Semiconductor …

NP/electrolyte interface under pulsed photoexcitation can result in an increment ofthe DEL potential as large as *0.7 eV, making possible the electron transfer fromcharged CdS NPs to CdTe NPs of any size studied (Fig. 1.12a).

The dynamics of photoinduced electron transfer from the charged CdS NPs toCdTe NPs is governed by the over-voltage, E* − ECB(CdTe) depending on the sizeof CdTe NPs. Table 1.4 and Fig. 1.12 show that a decrease of the CdTe NP sizefrom 6.6 to 3.0 nm results in a shift of the CB potential from around −0.9 V (versusNHE) to around −1.3 V (versus NHE), thus decreasing the electron transferover-voltage and reducing the efficiency of this process.

Table 1.4 Bandgap Eg and CB potential ECB of size-selected CdTe NPs, the electron transferover-voltage E* − ECB and the average number of transferred electrons per CdTe NP DNe [154]

dCdTe, нм Eg, eV ECB, V (NHE) E* − ECB, V DNe

3.0 2.24 −1.3 0.3 0.5

3.2 2.12 −1.2 0.4 0.6

5.0 1.82 −1.0 0.6 1.6

6.6 1.74 −0.9 0.7 4.5

Fig. 1.12 a Energy levelscheme for a colloidal systemcontaining CdS NPs andsize-selected CdTe NPs;b Rate of the photoinducedreductive corrosion of ZnONPs as a function of the NPEg and size [148, 151]

1.4 Photoinduced Charging of Semiconductor NPs 27

In some cases, the above-discussed transition of a semiconductor from aphotochemically-passive to a photochemically-active state can originate from asimultaneous contribution of the size-dependence of ECB energy and the photoin-duced NP charging. For example, colloidal ZnO NPs in ethanol are resistant to thereductive photocorrosion if the NP size is larger than *5 nm. However, the sta-tionary illumination of ZnO colloids with the smaller NPs results in the formationof metallic Zn [148, 151]: ZnO + 2e�tr + H2O ! Zn0 + 2OH−. The photoreductionstarts only after the development of a photoinduced Burstein shift of 0.15–0.18 eV.In the size range of 3.7–4.4 nm the photocorrosion rate is directly proportional tothe of ZnO NP bandgap (Fig. 1.12b).

As the photocorrosion does not demand any reactants to diffuse to the ZnO NPsurface, the sole reason for the increased photoactivity of ZnO NPs smaller than4.8 nm can be a size-dependent increase of the energy of photogenerated chargecarriers [148, 151]. Indeed, the CB potential of ZnO NPs shifts from −0.60 V(versus NHE) to −0.74 V (versus NHE) as the NP size is reduced from 4.8 to3.7 nm. At the same time, the photocorrosion occurs only after the photoinducedcharging of ZnO NPs, that is, under a cathodic polarization. According toEq. (1.19), a Burstein shift of DEB = 0.15 eV corresponds to an additional shift ofthe CB potential of around 0.20 B [148]. Therefore, the total size- and polarizationshifts can increase the CB level in ZnO NPs to E* = −0.80 V for 4.8-nm particlesand to −0.94 V for the smallest 3.7-nm particles (Table 1.5).

It was reported that the reductive dissolution of 5–6-nm ZnO NPs starts at thepotentials more negative than Ecorr = −0.8 V (versus NHE) [155]. As shown inTable 1.5, the reductive photocorrosion is impossible for 4.8-nm ZnO NPs, even inthe state of the photoinduced polarization (|E*| < |Ecorr|). For smaller ZnO NPs,however, the condition |E*| > |Ecorr| is valid and the NPs become unstable whenilluminated by UV light in deaerated solutions.

The above-discussed examples show that the capability of semiconductor NPsfor the photoinduced charging can have a number of far-reaching consequences fortheir photochemical behavior. First, an increase of the over-voltage of the interfacialcharge transfer results in an acceleration of the photochemical reactions.Additionally, the filling of the surface traps with excessive electrons blocks theradiative and non-radiative recombination channels and adds to an increased pho-toactivity of the charged NPs. Second, the accumulation of an excessive negativecharge by semiconductor NPs should favor to multi-electron processes allowing toavoid one-electron reduction steps that sometimes can require very high redox

Table 1.5 Somecharacteristics of size-selectedZnO NPs

Eg,eV

ECB, V(NHE)

d,nm

E*, V(NHE)

E* − Ecorr,V

3.43 −0.58 4.8 −0.78 0.02

3.48 −0.61 4.4 −0.81 −0.01

3.50 −0.63 4.1 −0.83 −0.03

3.57 −0.68 3.9 −0.88 −0.08

3.63 −0.72 3.7 −0.92 −0.12

28 1 Basic Concepts of the Photochemistry of Semiconductor …

potentials. This factor is of a special importance for the photocatalyticmulti-electron CO2 and N2 reduction discussed in Chap. 3. Finally, the photoin-duced charging of semiconductor NPs creates a new high-energy excited state E*

that can participate in processes impossible for “conventional” uncharged NPs, inparticular, with the participation of substrates with the redox potential more neg-ative than the CB level of uncharged semiconductor NPs.

Concluding the brief and basic discussion of the photochemical behavior ofsemiconductor NPs, we can outline a number of special features differingnanocrystalline semiconductors from their bulk counterparts. These special featuresarise from the phenomena of the spatial exciton confinement, the participation ofsurface states in photochemical reactions and the photoinduced charging of semi-conductor NPs.

An extremely high surface-to-volume ratio of semiconductor NPs with a size ofa few nanometers creates numerous surface states (corresponding to surface defects,vacancies, dangling bonds, etc.) that can actively participate in the photophysicalprocesses, in particular, by trapping the photogenerated charge carriers. At that, thenature, energy, and density of the surface states can sometimes dictate the possi-bility and rate of photochemical/photocatalytic processes with the participation ofsemiconductor NPs.

The quantum size effects in semiconductor NPs, that is, size-dependent variationof basic electrophysical parameters, such as the bandgap, CB and VB levels, etc.,can result in a dramatic enhancement of the photochemical processes as thesemiconductor crystal size is reduced to a few nanometers. Also, semiconductorspassive in the form of microcrystals, can become photochemically active whenintroduced as nanocrystals as a result of altered energies of the photogeneratedcharge carriers.

A fundamental difference between the photochemical properties of semicon-ductor NPs and bulk counterparts is, therefore, in the occurrence of photochemicalprocesses, when there is no mutual correspondence between the redox potentials ofreactants and CB/VB levels of the semiconductor crystal. This feasibility originatesnot only from the size-dependent increase of the energy of charge carriers but alsofrom the phenomenon of photoinduced charging of semiconductor NPs resulting ina radical change of the thermodynamics and kinetics of the interfacial chargetransfers.

References

1. Grätzel M (ed) (1983) Energy resources through photochemistry and catalysis. AcademicPress, Inc., New York

2. Fox MA, Dulay MT (1993) Heterogeneous photocatalysis. Chem Rev 93:341–357. doi:10.1021/cr00017a016

3. Linsebigler AL, Lu G, Yates JT Jr (1995) Photocatalysis on TiO2 surfaces: principles,mechanisms, and selected results. Chem Rev 95:735–758. doi:10.1021/cr00035a013

1.4 Photoinduced Charging of Semiconductor NPs 29

4. Shiraishi Y, Hirai T (2008) Selective organic transformations on titanium dioxide-basedphotocatalysts. J Photochem Photobiol C 9:157–170. doi:10.1016/j.jphotochemrev.2008.05.001

5. Imahori H, Umeyama T (2009) Donor-acceptor architecture on semiconducting electrodesfor solar energy conversion. J Phys Chem C 113:9029–9039. doi:10.1021/jp9007448

6. Liu G, Wang L, Yang HG et al (2010) Titania-based photocatalysts—crystal growth, dopingand hetero-structuring. J Mater Chem 20:831–843. doi:10.1039/B909930A

7. Gaponenko SV (1996) Optical properties of semiconductor nanocrystals. University Press,Cambridge

8. Wang Y, Herron N (1991) Nanometer-Sized semiconductor clusters: materials synthesis,quantum size effects, and photophysical properties. J Phys Chem 95:525–532. doi:10.1021/j100155a009

9. Kryukov AI, Kuchmiy SY, Pokhodenko VD (2000) Energetics of electron processes insemiconductor photocatalytic systems. Theor Experim Chem 36:63–81. doi:10.1007/BF02529022

10. Rogach A (ed) (2008) Semiconductor nanocrystal quantum dots: synthesis, assembly,spectroscopy and applications. Springer-Verlag GmbH, Vienna

11. Stroyuk OL, Kryukov AI, Kuchmiy SY, Pokhodenko VD (2005) Quantum size effects inphotonics of semiconductor nanoparticles. Theor Experim Chem 41:67–91. doi:10.1007/s11237-005-0025-9

12. Stroyuk OL, Kryukov AI, Kuchmiy SY, Pokhodenko VD (2005) Quantum size effects insemiconductor photocatalysis. Theor Experim Chem 41:207–228. doi:10.1007/s11237-005-0042-8

13. Henglein A (1989) Small-particle research: physicochemical properties of extremely smallcolloidal metal and semiconductor particles. Chem Rev 89:1861–1873. doi:10.1021/cr00098a010

14. Hagfeldt A, Grätzel M (1995) Light-induced redox reactions in nanocrystalline systems.Chem Rev 95:49–68. doi:10.1021/cr00033a003

15. Eychmüller A (2000) Structure and photophysics of semiconductor nanocrystals. J PhysChem B 104:6514–6528. doi:10.1021/jp9943676

16. Brus L (1986) Electronic wave functions in semiconductor clusters: experiment and theory.J Phys Chem 90:2555–2560. doi:10.1021/j100403a003

17. Adams DM, Kamat PV, Marcus RA et al (2003) Charge transfer on the nanoscale: currentstatus. J Phys Chem B 107:6668–6697. doi:10.1021/jp0268462

18. Zhao Y, Zhu K (2016) Organic–inorganic hybrid lead halide perovskites for optoelectronicand electronic applications. Chem Soc Rev 45:655–689. doi:10.1039/c4cs00458b

19. Grätzel M (2014) The light and shade of perovskite solar cells. Nat Mater 13:838–842.doi:10.1038/nmat4065

20. Sharma D, Jha R, Kumar S (2016) Quantum dot sensitized solar cells: recent advances andfuture perspectives in photoanode. Sol Energy Mater Sol Cells 155:294–322. doi:10.1016/j.solmat.2016.05.062

21. Aldakov D, Lefrançois A, Reiss P (2013) Ternary and quaternary metal chalcogenidenanocrystals: synthesis, properties and applications. J Mater Chem C 1:3756–3776. doi:10.1039/C3TC30273C

22. Thomas SR, Chen CW, Date M et al (2016) Recent developments in the synthesis ofnanostructured chalcopyrite materials and their applications: a review. RSC Adv 6:60643–60656. doi:10.1039/C6RA05502H

23. Beydoun D, Amal R, Low G, McEvoy S (1999) Role of nanoparticles in photocatalysis.J Nanopart Res 1:439–458. doi:10.1023/A:1010044830871

24. El-Sayed MA (2004) Small is different: shape-, size-, and composition-dependent propertiesof some colloidal semiconductor nanocrystals. Acc Chem Res 37:326–333. doi:10.1021/ar020204f

25. Alivisatos AP (1996) Perspectives on the physical chemistry of semiconductor nanocrystals.J Phys Chem 100:13226–13239. doi:10.1021/jp9535506

30 1 Basic Concepts of the Photochemistry of Semiconductor …

26. Landolt-Börnstein (1982) Numerical data and functional relationships in science andtechnology, Group III, vol 17b—Semiconductors, Sect. 3.10.1. Springer, Berlin

27. Khairutdinov RF (1998) Chemistry of semiconductor nanoparticles. Russ Chem Rev67:109–122

28. Alivisatos AP (1996) Semiconductor clusters, nanocrystals, and quantum dots. Science271:933–937

29. Shankar SS, Chatterjee S, Sastry M (2003) Synthesis of CdS nanoparticles within thermallyevaporated aerosol OT thin films. Phys Chem Commun 6:36–39. doi:10.1039/B303919F

30. Murakoshi K, Hosokawa H, Saitoh M et al (1998) Preparation of size-controlled hexagonalCdS nanocrystallites and the characteristics of their surface structures. J Chem Soc FaradayTrans 94:579–586. doi:10.1039/A707192B

31. Vossmeyer T, Katsikas L, Giersig M et al (1994) CdS nanoclusters: synthesis, character-ization, size dependent oscillator strength, temperature shift of the excitonic transitionenergy, and reversible absorbance shift. J Phys Chem 98:7665–7673. doi:10.1021/j100082a044

32. Nosaka Y, Shigeno H, Ikeuchi T (1995) Formation of polynuclear cadmium-thiolatecomplexes and CdS clusters in aqueous solution studied by means of stopped-flow andNMR spectroscopies. J Phys Chem 99:8317–8322. doi:10.1021/j100020a066

33. Yu WW, Qu L, Guo W, Peng X (2003) Experimental determination of the extinctioncoefficient of CdTe, CdSe, and CdS nanocrystals. Chem Mater 15:2854–2860. doi:10.1021/cm034081k

34. Choi KM, Shea KJ (1994) Amorphous polysilsesquioxanes as a confinement matrix forquantum-sized particle growth. Size analysis and quantum size effect of CdS particles grownin porous polysilsesquioxanes. J Phys Chem 98:3207–3214. doi:10.1021/j100063a026

35. Murray CB, Norris DJ, Bawendi MG (1993) Synthesis and characterization of nearlymonodisperse CdE (E = sulfur, selenium, tellurium) semiconductor nanocrystallites. J AmChem Soc 115:8706–8715. doi:10.1021/ja00072a025

36. Soloviev VN, Eichhöfer A, Fenske D, Banin U (2000) Molecular limit of a bulksemiconductor: size dependence of the “band gap” in CdSe cluster molecules. J Am ChemSoc 122:2673–2674. doi:10.1021/ja9940367

37. Rivera-Marquez A, Rubin-Falfan M, Lozada-Moralez R et al (2001) Quantum confinementand crystalline structure of CdSe nanocrystalline films. Phys Stat Sol (a) 188:1059–1064

38. Brysch A, Bour G, Neuendorf R, Kreibig U (1999) Nonlinear optical spectroscopy ofembedded semiconductor clusters. Appl Phys B 68:447–451. doi:10.1007/s003400050647

39. Rajh T, Micic OI, Nozik AJ (1993) Synthesis and characterization of surface-modifiedcolloidal cadmium telluride quantum dots. J Phys Chem 97:11999–12003. doi:10.1021/j100148a026

40. Sharma SN, Kohli S, Rastogi AC (2005) Quantum confinement effects of CdTe nanocrystalssequestered in TiO2 matrix: effect of oxygen incorporation. Physica E 25:554–561. doi:10.1016/j.physe.2004.08.110

41. Wong EM, Hörtz PG, Liang CJ et al (2001) Influence of organic capping ligands on thegrowth kinetics of ZnO nanoparticles. Langmuir 17:8362–8367. doi:10.1021/la010944h

42. Viswanatha R, Sapra S, Satpati B et al (2004) Understanding the quantum size effects inZnO nanocrystals. J Mater Chem 14:661–668. doi:10.1039/B310404D

43. Pesika NS, Stebe KJ, Searson PC (2003) Relationship between absorbance spectra andparticle size distributions for quantum-sized nanocrystals. J Phys Chem B 107:10412–10415.doi:10.1021/jp0303218

44. Green M, Wakefield G, Dobson PJ (2003) A simple metalorganic route to organicallypassivated mercury telluride nanocrystals. J Mater Chem 13:1076–1078. doi:10.1039/B210563B

45. Miyoshi H, Yamachika M, Yoneyama H, Mori H (1990) Photochemical properties of PbSmicrocrystallites prepared in nafion. J Chem Soc Faraday Trans 86:815–818. doi:10.1039/FT9908600815

References 31

46. Kang I, Wise W (1997) Electronic structure and optical properties of PbS and PbSe quantumdots. J Opt Soc Am B 14:1632–1646. doi:10.1364/JOSAB.14.001632

47. Sashchiuk A, Langof L, Chaim R, Lifshitz E (2002) Synthesis and characterization of PbSeand PbSe/PbS core–shell colloidal nanocrystals. J Cryst Growth 240:431–438. doi:10.1016/S0022-0248(02)01156-9

48. Allan G, Delerue C (2004) Confinement effects in PbSe quantum wells and nanocrystals.Phys Rev B 70:245321

49. Joswig JO, Springborg M, Seifert G (2000) Structural and electronic properties of cadmiumsulfide clusters. J Phys Chem B 104:2617–2622. doi:10.1021/jp993512c

50. Gurin VS (1996) Electronic structure of CdS cores in Cd thiolate complexes and clusters.J Phys Chem 100:869–872. doi:10.1021/jp9527620

51. Li J, Wang LW (2003) High energy excitations in CdSe quantum rods. Nano Lett 3:101–105. doi:10.1021/nl020231z

52. Hu J, Wang L, Li L et al (2002) Semiempirical pseudopotential calculation of electronicstates of CdSe quantum rods. J Phys Chem B 106:2447–2452. doi:10.1021/jp013204q

53. Woggon U, Wind O, Gindele F et al (1996) Optical transitions in CdSe quantum dots: fromdiscrete levels to broad gain spectra. J Lumin 70:269–280. doi:10.1016/0022-2313(96)00060-9

54. Rayevska OE, Grodzyuk GY, Dzhagan VM et al (2010) Synthesis and characterization ofwhite-emitting CdS quantum dots stabilized with polyethyleneimine. J Phys Chem C114:22478–22486. doi:10.1021/jp108561u

55. Raevskaya AE, Grodzyuk GY, Stroyul OL et al (2010) Preparation and spectral properties ofhigh-efficiency luminescent polyethyleneimine-stabilized CdS quantum dots. Theor ExpChem 46:233–238. doi:10.1007/s11237-010-9145-y

56. Raevskaya AE, Stroyuk OL, Kuchmiy SY (2003) Optical characteristics of colloidalnanoparticles of CdS stabilized with sodium polyphosphate and their behavior during pulsephotoexcitation. Theor Exp Chem 39:153–160. doi:10.1023/A:1024933023783

57. Raevskaya AE, Grodzyuk GY, Korzhak AV et al (2010) Preparation and optical propertiesof polyethyleneimine-stabilized colloidal CdSe and CdSxSe1−x quantum dots. TheorExperim Chem 46:397–402. doi:10.1007/s11237-011-9173-2

58. Nedeljkovic JM, Nenadovic MT, Micic OI, Nozik AJ (1986) Enhanced photoredoxchemistry in quantized semiconductor colloids. J Phys Chem 90:12–13. doi:10.1021/j100273a005

59. Steckel JS, Coe-Sullivan S, Bulovic V, Bawendi MG (2003) 1.3 lm to 1.55 lm tunableelectroluminescence from PbSe quantum dots embedded within an organic device. AdvMater 15:1862–1866. doi:10.1002/adma.200305449

60. Nenadovic MT, Rajh T, Mi OL (1985) Size quantization in small semiconductor particles.J Phys Chem 89:397–399. doi:10.1021/j100249a004

61. Kavan L, Stoto T, Grätzel M et al (1993) Quantum size effects in nanocrystallinesemiconducting titania layers prepared by anodic oxidative hydrolysis of titaniumtrichloride. J Phys Chem 97:9493–9498. doi:10.1021/j100139a038

62. van Dijken A, Janssen AH, Smitsmans MH et al (1998) Size-selective photoetching ofnanocrystalline semiconductor particles. Chem Mater 10:3513–3522. doi:10.1021/cm980715p

63. Klimov VI (2000) Optical nonlinearities and ultrafast carrier dynamics in semiconductornanocrystals. J Phys Chem B 104:6112–6123. doi:10.1021/jp9944132

64. Raevskaya AE, Stroyuk OL, Kuchmii SY (2004) Photocatalytic oxidation of hydrosulfideions by molecular oxygen over cadmium sulfide nanoparticles. J Nanopart Res 6:149–158.doi:10.1023/B:NANO.0000034719.30620.d3

65. Schmelz O, Mews A, Basché T et al (2001) Supramolecular complexes from CdSenanocrystals and organic fluorophors. Langmuir 17:2861–2865. doi:10.1021/la0016367

66. Leatherdale CA, Woo WK, Mikulec FV, Bawendi MG (2002) On the absorption crosssection of CdSe nanocrystal quantum dots. J Phys Chem B 106:7619–7622. doi:10.1021/jp025698c

32 1 Basic Concepts of the Photochemistry of Semiconductor …

67. Striolo A, Ward J, Prausnitz JM et al (2002) Molecular weight, osmotic second virialcoefficient, and extinction coefficient of colloidal CdSe nanocrystals. J Phys Chem B106:5500–5505. doi:10.1021/jp020170t

68. Zhang JZ, O’Neil RH, Roberti TW (1994) Femtosecond studies of photoinduced electrondynamics at the liquid-solid interface of aqueous CdS colloids. J Phys Chem 98:3859–3864.doi:10.1021/j100065a049

69. Zhang JZ (2000) Interfacial charge carriers dynamics of colloidal semiconductor nanopar-ticles. J Phys Chem B 104:7239–7253. doi:10.1021/jp000594s

70. Cropek D, Kemme PA, Makarova OV et al (2008) Selective photocatalytic decomposition ofnitrobenzene using surface modified TiO2 nanoparticles. J Phys Chem C 112:8311–8318.doi:10.1021/jp712137x

71. Yoon M, Chang JA, Kim Y et al (2001) Heteropoly acid-incorporated TiO2 colloids as novelphotocatalytic systems resembling the photosynthetic reaction center. J Phys Chem B105:2539–2545. doi:10.1021/jp003736r

72. Scotti R, Bellobono IR, Canevali C et al (2008) Sol–gel pure and mixed-phase titaniumdioxide for photocatalytic purposes: relations between phase composition, catalytic activity,and charge-trapped sites. Chem Mater 20:4051–4061. doi:10.1021/cm800465n

73. Fitzmaurice DJ, Frei H (1991) Transient near-infrared spectroscopy of visible lightsensitized oxidation of I− at colloidal TiO2. Langmuir 7:1129–1137. doi:10.1021/la00054a019

74. Boschloo G, Fitzmaurice D (1999) Spectroelectrochemical investigation of surface states innanostructured TiO2 electrodes. J Phys Chem B 103:2228–2231. doi:10.1021/jp984414e

75. Dimitrijevic NM, Saponjic ZV, Rabatic BM et al (2007) Effect of size and shape ofnanocrystalline TiO2 on photogenerated charges. An EPR study. J Phys Chem C 111:14597–14601. doi:10.1021/jp0756395

76. Ke SC, Wang TC, Wong MS, Gopal NO (2006) Low temperature kinetics and energetics ofthe electron and hole traps in irradiated TiO2 nanoparticles as revealed by EPR spectroscopy.J Phys Chem B 110:11628–11634. doi:10.1021/jp0612578

77. Berger T, Sterrer M, Diwald O et al (2005) Light-induced charge separation in anatase TiO2

particles. J Phys Chem B 109:6061–6068. doi:10.1021/jp040429378. Rabani J, Yamashita K, Ushida K et al (1998) Fundamental reactions in illuminated titanium

dioxide nanocrystallite levels studied by pulsed laser. J Phys Chem B 102:1689–1695.doi:10.1021/jp973411j

79. Yoshihara T, Katoh R, Furube A et al (2004) Identification of reactive species inphotoexcited nanocrystalline TiO2 films by wide-wavelength-range (400–2500 nm) transientabsorption spectroscopy. J Phys Chem B 108:3817–3823. doi:10.1021/jp031305d

80. Tang J, Durrant JR, Klug DR (2008) Mechanism of photocatalytic water splitting in TiO2.Reaction of water with photoholes, importance of charge carrier dynamics, and evidence forfour-hole chemistry. J Am Chem Soc 130:13885–13891. doi:10.1021/ja8034637

81. Fu H, Zhang L, Zhang S et al (2006) Electron spin resonance spin-trapping detection ofradical intermediates in N-doped TiO2-assisted photodegradation of 4-chlorophenol. J PhysChem B 110:3061–2065. doi:10.1021/jp055279q

82. Kamat PV, Dimitrijević NM, Fessenden RW (1988) Photoelectrochemistry in particulatesystems. 7. Electron-transfer reactions of indium sulfide semiconductor colloids. J PhysChem 92:2324–2329. doi:10.1021/j100286a029

83. Kamat PV, Ebbesen TW, Dimitrijević NM, Nozik AJ (1989) Primary photo-chemical eventsin CdS semiconductor colloids as probed by picosecond laser flash photolysis. Chem PhysLett 157:384–389. doi:10.1016/0009-2614(89)87267-7

84. Jortner K, Rao CNR (2002) Nanostructured advanced materials. Perspectives and directions.Pure Appl Chem 74:1491–1506

85. Nozik AJ, Memming R (1996) Physical chemistry of semiconductor–liquid interfaces.J Phys Chem 100:13061–13078. doi:10.1021/jp953720e

References 33

86. Yang X, Tamai N (2001) How fast is interfacial hole transfer? In situ monitoring of carrierdynamics in anatase TiO2 nanoparticles by femtosecond laser spectroscopy. Phys ChemChem Phys 3:3393–3398. doi:10.1039/B101721G

87. Matylitsky VV, Dworak L, Breus VV et al (2009) Separation in multiexcited CdSe quantumdots mediated by adsorbed electron acceptors. J Am Chem Soc 131:2424–2425. doi:10.1021/ja808084y

88. Logunov S, Green T, Marguet S, El-Sayed MA (1998) Interfacial carriers dynamics of CdSnanoparticles. J Phys Chem A 102:5652–5658. doi:10.1021/jp980387g

89. Rogach A (ed) (2008) Semiconductor nanocrystal quantum dots: synthesis, assembly,spectroscopy and applications. Springer, Vienna

90. Chestnoy N, Harris TD, Hull R, Brus LE (1986) Luminescence and photophysics ofcadmium sulfide semiconductor clusters: the nature of the emitting electronic state. J PhysChem 90:3393–3399. doi:10.1021/j100406a018

91. Ramsden JJ, Grätzel M (1984) Photoluminescence of small cadmium sulphide particles.J Chem Soc Faraday Trans 1(80):919–933. doi:10.1039/F19848000919

92. Kamat PV, Patrick B (1992) Photophysics and photochemistry of quantized ZnO colloids.J Phys Chem 96:6829–6834. doi:10.1021/j100195a055

93. Ramsden JJ, Webber SE, Grätzel M (1985) Luminescence of colloidal cadmium sulfideparticles in acetonitrile and acetonitrile/water mixtures. J Phys Chem 89:2740–2743. doi:10.1021/j100259a007

94. Stroyuk O, Raevskaya AE, Kozytskiy AV, Kychmiy SY (2010) Electron energy factors inphotocatalytic methylviologen reduction in the presence of semiconductor nanocrystals.J Photochem Photobiol A 210:209–214. doi:10.1016/j.jphotochem.2009.11.019

95. Kanemoto M, Shiragami T, Pac C, Yanagida S (1992) Semiconductor photocatalysis.Effective photoreduction of carbon dioxide catalyzed by ZnS quantum crystallites with lowdensity of surface defects. J Phys Chem 96:3521–3526. doi:10.1021/j100187a062

96. Stroyuk OL, Raevskaya AE, Korzhak AV, Kuchmiy SY (2007) ZnS nanoparticles: spectralproperties and photocatalytic activity in metals reduction reactions. J Nanopart Res 9:1027–1039. doi:10.1007/s11051-006-9183-5

97. Stroyuk OL, Shvalagin VV, Raevskaya AE et al (2003) Photocatalysis of the reduction ofCd2+ ions by CdS nanoparticles in isopropyl alcohol. Theor Experim Chem 39:341–346.doi:10.1023/B:THEC.0000013985.94005.c3

98. Dzhagan VM, Stroyuk OL, Raevskaya AE et al (2010) Spectroscopic and photochemicalstudy of Ag+-, Cu2+-, Hg2+-, and Bi3+-doped CdxZn1−xS nanoparticles. J Colloid InterfaceSci 345:515–523. doi:10.1016/j.jcis.2010.02.001

99. Henglein A (1997) Nanoclusters of semiconductors and metals: colloidal nanoparticles ofsemiconductors and metals: electronic structure and processes. Ber Bunsenges Phys Chem101:1562–1572. doi:10.1002/bbpc.19971011103

100. Faust BC, Hoffmann MR, Bahnemann DW (1989) Photocatalytic oxidation of sulfur dioxidein aqueous suspensions of a-Fe2O3. J Phys Chem 93:6371–6381. doi:10.1021/j100354a021

101. Wilcoxon JP (2000) Catalytic photooxidation of pentachlorophenol using semiconductornanoclusters. J Phys Chem B 104:7334–7343. doi:10.1021/jp0012653

102. Thurston TR, Wilcoxon JP (1999) Photooxidation of organic chemicals catalyzed bynanoscale MoS2. J Phys Chem B 103:11–17. doi:10.1021/jp982337h

103. Yanagida S, Ogata T, Shindo A et al (1995) Semiconductor photocatalysis: size control ofsurface-capped CdS nanocrystallites and the quantum size effect in their photocatalysis. BullChem Soc Jpn 68:752–758. doi:10.1246/bcsj.68.752

104. Raevskaya AE, Stroyuk OL, Kuchmiy SY (2006) Preparation of colloidal CdSe and CdS/CdSe nanoparticles from sodium selenosulfate in aqueous polymers solutions. J ColloidInterface Sci 302:133–141. doi:10.1016/j.jcis.2006.06.018

105. Stroyuk OL, Korzhak AV, Raevskaya AE, Kuchmiy SY (2004) Photocatalysis of the releaseof molecular hydrogen from aqueous solutions of sodium sulfite by composite CdS/Ninanoparticles. Theor Experim Chem 40:1–6. doi:10.1023/B:THEC.0000020758.50103.8f

34 1 Basic Concepts of the Photochemistry of Semiconductor …

106. Torimoto T, Sakata T, Mori H, Yoneyama H (1994) Effect of surface charge of4-aminothiophenol-modified PbS microcrystal photocatalysts on photoinduced chargetransfer. J Phys Chem 98:3036–3043. doi:10.1021/j100062a048

107. Kovalenko MV, Bodnarchuk MV, Stroyuk OL, Kuchmiy SY (2004) Synthesis, optical, andphotocatalytic properties of quantum-sized particles of CdTe. Theor Experim Chem 40:220–225. doi:10.1023/B:THEC.0000041806.60632.04

108. Kanemoto M, Hosokawa H, Wada Y et al (1996) Semiconductor photocatalysis. Part 20.Role of surface in the photoreduction of carbon dioxide catalyzed by colloidal ZnSnanocrystallites in organic solvent. J Chem Soc Faraday Trans 92:2401–2411. doi:10.1039/FT9969202401

109. Kang Z, Tsang CHA, Wong NB et al (2007) Silicon quantum dots: a general photocatalystfor reduction, decomposition, and selective oxidation reactions. J Am Chem Soc 129:12090–12091. doi:10.1021/ja075184x

110. Hoffman AJ, Yee H, Mills G, Hofmann MR (1992) Photoinitiated polymerization of methylmethacrylate using Q-sized ZnO colloids. J Phys Chem 96:5540–5546. doi:10.1021/j100192a066

111. Jana S, Pande S, Sinha AK, Pal T (2008) Synthesis of superparamagnetic b-MnO2

organosol: a photocatalyst for the oxidative phenol coupling reaction. Inorg Chem 47:5558–5560. doi:10.1021/ic800499q

112. Guo Y, Zhang H, Wang Y et al (2005) Controlled growth and photocatalytic properties ofCdS nanocrystals implanted in layered metal hydroxide matrixes. J Phys Chem B105:21602–21607. doi:10.1021/jp054400q

113. Zhao Q, Xie Y, Zhang Z, Bai X (2007) Size-selective synthesis of zinc sulfide hierarchicalstructures and their photocatalytic activity. Cryst Growth Des 7:153–158. doi:10.1021/cg060521j

114. Yin H, Wada Y, Kitamura T, Yanagida S (2001) Size-selective synthesis of zinc sulfidehierarchical structures and their photocatalytic activity. Environ Sci Technol 35:227–231.doi:10.1021/cg060521j

115. Shen S, Guo L (2006) Structural, textural and photocatalytic properties of quantum-sizedIn2S3-sensitized Ti-MCM-41 prepared by ion-exchange and sulfidation methods. J Sol StateChem 179:2629–2635. doi:10.1016/j.jssc.2006.05.010

116. Sathish M, Viswanathan B, Viswanath RP (2006) Alternate synthetic strategy for thepreparation of CdS nanoparticles and its exploitation for water splitting. Inter J Hydrogen En31:891–898. doi:10.1016/j.ijhydene.2005.08.002

117. Sathish M, Viswanath RP (2007) Photocatalytic generation of hydrogen over mesoporousCdS nanoparticle: effect of particle size, noble metal and support. Catal Today 129:421–427.doi:10.1016/j.cattod.2006.12.008

118. Shen S, Guo L (2008) Growth of quantum-confined CdS nanoparticles inside Ti-MCM-41 asa visible light photocatalyst. Mater Res Bull 43:437–446. doi:10.1016/j.materresbull.2007.02.034

119. Bao N, Shen L, Takata T et al (2007) Facile Cd-thiourea complex thermolysis synthesis ofphase-controlled CdS nanocrystals for photocatalytic hydrogen production under visiblelight. J Phys Chem C 111:17527–17534. doi:10.1021/jp076566s

120. Datta A, Priyam A, Bhattacharyya SN et al (2008) Temperature tunability of size in CdSnanoparticles and size dependent photocatalytic degradation of nitroaromatics. J ColloidInterface Sci 322:128–135. doi:10.1016/j.jcis.2008.02.052

121. Stroyuk OL, Granchak VM, Korzhak AV, Kuchmii SY (2004) Photoinitiation ofbuthylmethacrylate polymerization by colloidal semiconductor nanoparticles. J PhotochemPhotobiol A 162:339–351. doi:10.1016/S1010-6030(03)00371-X

122. Matsumoto H, Uchida H, Matsunaga T et al (1994) Photoinduced reduction of viologens onsize-separated CdS nanocrystals. J Phys Chem 98:11549–11556. doi:10.1021/j100095a041

123. Nosaka Y, Ohta N, Miyama H (1990) Photochemical kinetics of ultrasmall semiconductorparticles in solution: effect of size on the quantum yield of electron transfer. J Phys Chem94:3752–3755. doi:10.1021/j100372a073

References 35

124. Korgel BA, Monbouquette HG (1997) Quantum confinement effects enable photocatalyzednitrate reduction at neutral pH using CdS nanocrystals. J Phys Chem B 101:5010–5017.doi:10.1021/jp971016n

125. Haram SK, Quinn BM, Bard AJ (2001) Electrochemistry of CdS nanoparticles: a correlationbetween optical and electrochemical band gaps. J Am Chem Soc 123:8860–8861. doi:10.1021/ja0158206

126. Weib H, Fernandez A, Kisch H (2001) Elektronische Halbleiter-träger-wechselwirkung –

ein neuer Effekt in der Halbleiterphotokatalyse. Angew Chem 113:3942–3945127. Riley DJ, Waggett JP, Wijayantha KGU (2004) Colloidal bismuth sulfide nanoparticles: a

photoelectrochemical study of the relationship between bandgap and particle size. J MaterChem 14:704–708. doi:10.1039/B311517H

128. Haram SK, Kshirsagar A, Gujarathi YD et al (2011) Quantum confinement in CdTequantum dots: investigation through cyclic voltammetry supported by density functionaltheory (DFT). J Phys Chem C 115:6243–6249. doi:10.1021/jp111463f

129. Burda C, El-Sayed MA (2000) High-density femtosecond transient absorption spectroscopyof semiconductor nanoparticles. A tool to investigate surface quality. Pure Appl Chem72:165

130. Haase M, Weller H, Henglein A (1988) Photochemistry of colloidal semiconductors. 26.Photoelectron emission from cadmium sulfide particles and related chemical effects. J PhysChem 92:4706–4712. doi:10.1021/j100327a030

131. Jacobsohn M, Banin U (2000) Dependence of second harmonic generation in CdSenanocrystal quantum dots. J Phys Chem B 104:1–5. doi:10.1021/jp9925076

132. Wuister SF, de Mello Donega C, Meijerink A (2004) Influence of thiol capping on theexciton luminescence and decay kinetics of CdTe and CdSe quantum dots. J Phys Chem B108:17393–17397. doi:10.1021/jp047078c

133. Tan GL, Yang Q, Hömmerlich U et al (2004) Linear and non-linear optical properties ofcapped CdTe nanocrystals prepared by mechanical alloying. Opt Mater 27:579–584. doi:10.1016/j.optmat.2004.06.005

134. Wu F, Lewis JW, Kliger DS, Zhang JZ (2003) Unusual excitation intensity dependence offluorescence of CdTe nanoparticles. J Chem Phys 118:12–16. doi:10.1063/1.1533733

135. Zheng MJ, Zhang LD, Zhang JG (2001) Size dependence of non-linear optical properties ofSiO2 thin films containing InP nanocrystals. Appl Phys A 73:183. doi:10.1007/s003390000615

136. Yang C, Yu P, Wang J et al (2000) Synthesis and characterization of poly(vinylpyrrolidone)-modified zinc oxide nanoparticles. Chem Mater 12:2268–2274. doi:10.1021/cm9907817

137. Ganeev RA, Baba M, Morita M et al (2004) Thermally induced and Kerr-induced opticalnonlinearities of a pseudoisocyanine solution at 532 nm. J Opt A 6:1076

138. Shabaev A, Efros AL, Nozik AJ (2006) Multiexciton generation by a single photon innanocrystals. Nano Lett 6:2856–2863. doi:10.1021/nl062059v

139. Ellingson RJ, Beard MC, Johnson JC et al (2006) Highly efficient multiple excitongeneration in colloidal PbSe and PbS quantum dots. Nano Lett 5:865–871. doi:10.1021/nl0502672

140. Nozik AJ (2008) Multiple exciton generation in semiconductor quantum dots. Chem PhysLett 457:3–11. doi:10.1016/j.cplett.2008.03.094

141. Midgett AG, Luther JM, Stewart JT et al (2013) Size and composition dependent multipleexciton generation efficiency in PbS, PbSe, and PbSxSe1−x alloyed quantum dots. Nano Lett13:3078–3085. doi:10.1021/nl4009748

142. Kamat PV, Dimitrijević NM, Nozik AJ (1989) Dynamic Burstein-Moss shift in semicon-ductor colloids. J Phys Chem 93:2873–2875. doi:10.1021/j100345a003

143. Savinov EN, Nagorky VE, Parmon VN (1994) Influence of excessive charge of colloidalcadmium sulfide particles on the rate of interfacial charge transfer. Khim Fiz 13:56–65

144. Liu C, Bard AJ (1989) Effect of excess charge on band energetics (optical absorption edgeand carrier redox potentials) in small semiconductor particles. J Phys Chem 93:3232–3237.doi:10.1021/j100345a069

36 1 Basic Concepts of the Photochemistry of Semiconductor …

145. Albery WJ, Brown GT, Darwent JR, Saeivar-Iranizad E (1985) Time-resolved photoredoxreactions of colloidal CdS. J Chem Soc Faraday Trans 1(81):1999–2007. doi:10.1039/F19858101999

146. Bavykin DV, Savinov EN, Parmon VN (2000) Studies on the kinetics of interfacial electrontransfer sensitized by colloidal CdS. J Photochem Photobiol A 130:57–61. doi:10.1016/S1010-6030(99)00196-3

147. Kamat PV, Shanghavi B (1997) Interparticle electron transfer in metal/semiconductorcomposites. Picosecond dynamics of CdS-capped gold nanoclusters. J Phys Chem B101:7675–7679. doi:10.1021/jp9709464

148. Shvalagin VV, Stroyuk OL, Kuchmiy SY (2004) Role of quantum size effects in thecathodic photocorrosion of ZnO nanoparticles in ethanol. Theor Experim Chem 40:378–382.doi:10.1007/s11237-005-0003-2

149. Cao G (2004) Nanostructures & nanomaterials: synthesis, properties & applications.Imperial college press, London

150. Stroyuk OL, Shvalagin VV, Kuchmiy SY (2004) Photochemical synthesis, spectral-opticaland electrophysical properties of composite nanoparticles of ZnO/Ag. Theor Exp Chem40:98–104. doi:10.1023/B:THEC.0000028904.52818.e7

151. Stroyuk OL, Shvalagin VV, Kuchmii SY (2005) Photochemical synthesis and opticalproperties of binary and ternary metal-semiconductor composites based on zinc oxide.J Photochem Photobiol A 173:185–194. doi:10.1016/j.jphotochem.2005.02.002

152. Stroyuk OL, Shvalagin VV, Kuchmiy SY (2007) Photochemical synthesis of ZnO/Agnanocomposites. J Nanopart Res 9:427–440. doi:10.1007/s11051-006-9086-5

153. Liver N, Nitzan A (1992) Redox properties of small semiconductor particles. J Phys Chem96:3366–3373. doi:10.1021/j100187a035

154. Bodnarchuk MI, Kovalenko MV, Stroyuk OL, Kuchmiy SY (2004) Photoinduced electrontransfer between CdS and CdTe in colloidal solutions. Theor Exp Chem 40:287–292. doi:10.1023/B:THEC.0000049074.31907.81

155. Hoyer P, Weller H (1995) Potential-dependent electron injection in nanoporous colloidalZnO films. J Phys Chem 99:14096–14100. doi:10.1021/j100038a048

156. Frenzel J, Joswig JO, Seifert G (2007) Optical excitations in cadmium sulfide nanoparticles.J Phys Chem C 111:10761–10770. doi:10.1021/jp071125u

157. Leistikow MD, Johansen J, Kettelarij AJ et al (2009) Size-dependent oscillator strength andquantum efficiency of CdSe quantum dots controlled via the local density of states. PhysRev B 79:045301

158. Dzhagan VM, Valakh MY, Raevskaya AE et al (2007) Resonant Raman scattering study ofCdSe nanocrystals passivated with CdS and ZnS. Nanotechnology 18:285701

159. Raevskaya AE, Stroyuk AL, Kuchmiy SY et al (2007) Optical study of CdS- andZnS-passivated CdSe nanocrystals in gelatin films. J Phys Condens Matter 19:386237

160. Dzhagan VM, Valakh MY, Raevskaya AE et al (2008) Size effects on Raman spectra ofsmall CdSe nanoparticles in polymer films. Nanotechnology 19:305707

161. Stroyuk OL, Dzhagan VM, Shvalagin VV et al (2010) Size-dependent optical properties ofcolloidal ZnO nanoparticles charged by photoexcitation. J Phys Chem C 114:220–225.doi:10.1021/jp908879h

162. Raevskaya AE, Stroyuk OL, Kuchmiy SY et al (2006) Growth and spectroscopiccharacterization of CdSe nanoparticles synthesized from CdCl2 and Na2SeSO3 in aqueousgelatine solutions. Colloids Surf A 290:304–309. doi:10.1016/j.colsurfa.2006.05.038

163. Stroyuk OL, Shvalagin VV, Kotenko IE et al (2010) Photochemical reduction of sulfur inethanol in the presence of ZnO nanoparticles. Theor Exp Chem 46:218–224. doi:10.1007/s11237-010-9143-0

References 37

Chapter 2Semiconductor-Based PhotocatalyticSystems for the Solar-Light-Driven WaterSplitting and Hydrogen Evolution

The research and development of new technologies for the conversion and storageof inexhaustible solar light energy were boosted several decades ago by the 1970thfuel crisis and a strategic need for sustainable power sources that can serve asalternatives to the fossil fuels. The basic idea was to accumulate the solar lightenergy as the electricity as well as to store it in the form of highly endothermic andeco-friendly fuels, in particular, molecular hydrogen produced by the photochem-ical splitting of water.

Direct photochemical water splitting to gaseous hydrogen and oxygen can occuronly under the illumination with highly energetic quanta at the wavelength k shorterthan 240 nm [1]. However, such irradiation is completely absorbed by the atmo-sphere and does not reach the Earth surface. To overcome this obstacle, the watersplitting is realized in the presence of photocatalysts—the substances capable ofabsorbing longer-wavelength light quanta (k > 300 nm) and inducing chemicaltransformations of water molecules.

Inorganic semiconductors are probably the most broadly studied photocatalystsof water splitting. The semiconductor photocatalysts combine a high photosensi-tivity with a photochemical activity, stability, availability and relative simplicity ofpractical implementation. It should be noted that the photocatalytic andelectro-photocatalytic (photoelectrochemical) processes with the participation ofsemiconductor nanomaterials are very similar by the nature and start with the sameprimary act of light quantum absorption resulting in the generation of anelectron-hole couple. Differences between photocatalytic and photoelectrochemical/photoelectrocatalytic processes arise mainly on the secondary steps of the chargecarrier migration to the reaction participants. By this reason, both types of processescan be regarded as photocatalytic ones occuring in “usual” and electrochemicalregimes and discussed together.

Molecular hydrogen can be produced in photocatalytic systems of two types:(a) water splitting systems where stoichiometric amounts of H2 and O2 are pro-duced simultaneously, and (b) systems with a so-called “sacrificial” donor which isconsumed irreversibly supplying electrons for the water reduction.

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_2

39

Stoichiometric (total) water splitting is accompanied by the energy accumulationand a free energy increment DG = 238 kJ/mole [2, 3]. Such process requires asemiconductor photocatalyst with a valence band (VB) potential more positive thanthe water oxidation potential (1.23 V vs. normal hydrogen electrode (NHE) atpH 0) and a conduction band (CB) potential more negative than the NHE potential(E = 0.0 V at pH 0). Therefore, a minimal light quantum energy required for thesemiconductor-driven water splitting is 1.23 eV. Invariable losses accompanyinginterfacial charge transfers as well as over-voltages of the H2 and O2 formationincrease this minimal energy to 1.7–1.9 eV [2, 3]. Therefore, the photocatalyticconversion of solar light energy should be the most favorable from the energeticviewpoints for semiconductors with a band gap (Eg) around 1.7–1.9 eV and acorresponding fundamental absorption band edge at kbe = 650–730 nm.

The wider-band-gap semiconductors with kbe < 400 nm can also be used for thewater splitting. However, due to a relatively small fraction of the UV light in thesolar flux at the Earth surface, the conversion efficiency in such systems is typicallynot higher than 1–2%. Therefore, successful application of wide-band-gap semi-conductors for the water splitting can be achieved only by expansion of their lightsensitivity range to the visible domain of the spectrum. This effect can be achievedeither by doping with metal/non-metal additives during the semiconductor synthesisor by various post-synthesis modifications.

It should be noted that the semiconductors-based systems for the total watersplitting have not yet showed reasonably high conversion efficiency as a result of afast recombination of the oppositely charge photogenerated charge carriers as wellas of primary intermediates—hydrogen atoms and hydroxyl radicals. A muchhigher conversion efficiency was achieved in the photo-catalytic systems withsacrificial donors. The range of sacrificial donors is very broad including inorganicsulfur compounds (H2S and alkali metal sulfides, sulfites, thiosulfates, thionates,etc.), hydrazine and aliphatic amines (triethylamine, triethanolamine (TEA), etc.),aliphatic alcohols (methanol, ethanol, 2-propanol), carboxylic acids (formic acid,ethylenediaminetetraacetic (EDTA) acid, etc.), carbohydrates and other organicsubstances, in particular those abundant in the broadly available and sustainablesource—the fermented bio-mass.

In the donor-based systems the photocatalytic process includes following typicalstages: (i) excitation of a semiconductor photocatalyst by a light quantum with aproper (typically above-band-gap) energy, (ii) the interfacial transfer of a CBelectron to an adsorbed water molecule followed by its reduction(e− + H2O ! H• + OH•), (iii) filling of a VB hole with an electron from a sacri-ficial donor (h+ + D ! D+•). This cycle requires the CB potential of a semicon-ductor photocatalyst to be more negative than the water reduction potential in givenconditions and the VB potential—to be more positive than the oxidation potentialof a sacrificial donor (or water molecules). Figure 2.1 provides a graphic review ofband edge positions for a series of semiconductor materials relative to the standardpotentials of water reduction and oxidation. The figure shows separately thesemiconductors suitable (a) and unsuitable (b) for the evolution of the solarhydrogen from water.

40 2 Semiconductor-Based Photocatalytic Systems …

Typically, the semiconductor-based photocatalytic systems for the hydrogenproduction include a co-catalyst, that has no inherent photochemical activity but iscapable of increasing dramatically the efficiency of semiconductor photocatalysts.Metal particles (Pt, Pd, Rh) deposited either on the semiconductor surface or on thesurface of an inert carrier are typical co-catalysts for the semiconductor-basedphotocatalytic systems. The co-catalyst accepts and accumulates the charge carriersphotogenerated in the semiconductor crystals inhibiting their recombination as wellas contributes to a lowering of the water reduction overvoltage.

In recent years the studies of new light energy conversion systems based onsemiconductor photocatalysts and photoelectrodes have bloomed in leadingresearch centers [5–34]. The research focused also on the photosyntheticmicroorganisms and other photoactive bio-systems capable of the molecularhydrogen evolution [35–37]. The present chapter obviously cannot encompass thewhole variety of papers reporting on the photochemical water splitting. It aimsmainly to highlight typical and most important directions of the recent research aswell as to give the reader a notion of the current state of the area and its futuredevelopment.

2.1 Photocatalytic Systems Based on the Wide-Band-GapSemiconductors and Sensitizers

The wide-band-gap semiconductors, mostly metal oxides, belong to a large groupof light-sensitive materials broadly studied as photocatalysts of the water reduction.The spectral sensitivity range of such materials can be expanded to longer wave-lengths by combining them with dyes-sensitizers that absorb strongly UV and nearIR light.

Fig. 2.1 CB and VB energylevels for somesemiconductingphotocatalysts with respect toNHE (ENHE) and vacuum(Evac). Reprinted withpermission from Ref. [4].Copyright (2015) The RoyalSociety of Chemistry

2 Semiconductor-Based Photocatalytic Systems … 41

Upon absorption of the visible and near IR light a sensitizer gets excited from theground singlet state S0 into the first (or a higher) singlet excited state S1 (Sn). The S1

state can either return to S0 via emitting fluorescence or via the radiationless internalconversion. It can also convert into the first triplet excited state T1 or inject anelectron into the conduction band (CB) of a semiconductor. After that, the waterreduction occurs either on the semiconductor surface or (most often) on the surfaceof a metal co-catalysts (Fig. 2.2). The role of spectral sensitizers is typically playedby organic dyes or metal complexes (Fig. 2.3). The basic operation principles andthe state-of-the-art of the photocatalytic H2 evolution with the dye-sensitizedsemiconductors are comprehensively outlined in a recent review [38].

The most studied sensitized systems are based on titanium(IV) dioxide. Forexample, the hydrogen evolution under the illumination with the visible light(Vis-illumination) was observed in the presence of TiO2/Pt heterostructures mod-ified by eosin [39, 40], derivatives of phenothiazine [41, 42], triphenylamine [43]and perylene [44], by various complexes of PtIV [45], ZnII [46] and NiII [47], copperphthalocyanine and ruthenium bipyridyl complexes [39]. Eosins adsorbed on thesurface of Na2Ti2O4(OH)2 nanotubes (NTs) or MCM-41 zeolite modified by TiO2

nanoparticles (NPs) in the presence of the photodeposited Pt NPs act as spectralsensitizers of the hydrogen evolution from aqueous TEA solutions [48, 49].A sensitization effect was also observed in a similar system based on eosin Y andN-doped TiO2 NPs [50].

Hydrogen generation from water/acetonitrile/КI occurs at the expense of I−

oxidation under the Vis-illumination of the platinized titania and layered K4Nb6O17

sensitized by adsorbed coumarin and merocyanine dyes [51]. In the latter case, aneffect of Pt NP localization on the photocatalyst activity was observed. Thehydrogen formation rate over the K4Nb6O17/Pt composites with Pt NPs formedinside the interlayer space was found to be much higher than in similar systemswhere the metal NPs were distributed evenly between the inner and outer surface ofthe semiconductor or deposited only onto the outer semiconductor surface. Theeffect is caused by a side reaction of I�3 complex with the CB electrons.

The eosin Y acts as a “universal” sensitizer for a series of layered wide-band-gapmagnesium, calcium and strontium titanates [52]. The highest photocatalyticactivity in the hydrogen evolution from aqueous diethanolamine solutions was

Fig. 2.2 Scheme of a photocatalytic system for the hydrogen evolution based on a TiO2/Ptheterostructure and a sensitizer (S). S0, S1, S+•—sensitizer in the ground state, excited state andoxidized state, respectively, D—sacrificial donor

42 2 Semiconductor-Based Photocatalytic Systems …

observed for SrTiO3 modified by 0.5 wt.% Pt. Co3O4 NPs sensitized by eosin Yshowed a high activity in the water reduction under the Vis-illumination in theabsence of any additional co-catalysts [53].

Adsorption of 1,1/-dinaphtyl-2,2/-diol on the surface of TiO2 NPs results in theformation of a charge-transfer complex with an intense absorption band centered at550–600 nm. The photoexcitation of the complex into a charge-transfer absorptionband leads to the hydrogen evolution from aqueous TEA solutions with a quantumyield (QY) of 0.02% [54]. The photocatalytic hydrogen evolution from aqueousglycerol solutions was observed for TiO2/Pt nanoheterostructures sensitized byinorganic tungsten-containing heteropolyacids [55, 56].

Molecular and metal complex dyes were successfully used to sensitize not onlymetal oxide photocatalysts but also semiconductors of other types, such as cad-mium sulfide [57] and graphitic carbon nitride (g-C3N4, GCN) [58]. TheVis-illumination of aqueous GCN suspensions in the presence of eosin Y, TEA, andPt NPs resulted in the hydrogen evolution with a QY of around 19% [58]. In similarphotocatalytic systems, g-C3N4 was sensitized by erythrosin [59, 60] and copperphtalocyanine [61]. GCN sensitized by ZnII phthalocyanines revealed a compara-tively high quantum yield of H2 evolution reaching 3.05% and a spectral sensitivityof up to 750 nm [62].

Starting from 1980th, various RuIII/II complexes with bipyridyl ligands werebroadly studied as sensitizers of the hydrogen production and the studies in thisdirection are still advancing. For example, a photocatalytic system for the hydrogen

Fig. 2.3 Structure of some molecular sensitizers used in the semiconductor-based photocatalyticsystems for hydrogen evolution

2.1 Photocatalytic Systems Based on the Wide-Band-Gap … 43

production comprising Ru2+ tris-bipyridyl complexes, TiO2 NPs and hydrogenaseas a co-catalyst was reported [63]. The hydrogen evolution under theVis-illumination of aqueous solutions of sacrificial donors (methanol [64, 65] orTEA [66]) was observed in the presence of mesoporous TiO2 modified by Pt NPsand mono- and bidentate Ru2+ bipyridyl complexes.

A strong electrostatic interaction between Ru(bpy)32+ cation and the negatively

charged surface of K4Nb6O17 nanoscrolls produced by the exfoliation of the bulkpotassium niobate results in efficient electron phototransfer from the excited sen-sitizer to the semiconductor CB. The rate of photocatalytic hydrogen evolution fromaqueous EDTA solutions is by an order of magnitude higher in the case ofK4Nb6O17 nanoscrolls than for the bulk semiconductor [67]. The H2 evolution QYfrom EDTA solutions in the presence of H4Nb6O17 and HCa2Nb3O10 nanoscrollsmodified by platinum NPs and Ru(bpy)3

2+ and Ru(bpy)2(4,4/-(PO3H2)2bpy)

2+

complexes reached 20–25% [68].New sensitizers of titanium dioxide—binuclear RuIII complexes with separate

fragments connected by an azobenzene “bridge”were reported in [69]. As opposite to“classical” sensitizers of such type that typically adsorb strongly on the semiconductorsurface, the bonding between the sensitizer and the photocatalyst is weak in this case.The weak coupling allows for the photooxidized sensitizer to desorb from the semi-conductor surface inhibiting a reverse electron transfer and accelerating the photo-catalytic hydrogen evolution from aqueous solutions of methanol or TEA.

A recent extensive review of the sensitized H2 evolution in the semiconductor-based systems [38] outlined principal challenges that still need to be met in thisarea. Most dyes have relatively narrow absorption bands, typically in the Vis rangeand an expansion of the light-harvesting range into the near IR is a vital challengeto be addressed. Some strategies aimed at resolving this problem includeco-sensitization of semiconductor nanomaterials with combinations of dyes havingcomplementary absorption spectra; fabrication of heterostructures with dyes,narrow-band-gap semiconductors, and conductive polymers; search for ligandscapable of bonding to the semiconductor surface and forming intense ligand-to-metal charge transfer absorption bands, etc.

The second challenge lies in a typically low stability of the molecular sensitizers.The organic dyes suffer from the photodegradation as a result of alternative reac-tions involving the singlet and triplet excited dyes, while the metal complexes areprone to photoinduced ligand exchange and photosolvation reactions resulting inthe deterioration of their light-harvesting ability. Attempts of abating this probleminclude a proper modification of the semiconductor surface to mitigate secondaryreactions as well as a rational design of the dye structure to reduce the possibility ofthe excited state relaxation pathways competing with the charge injection.

In recent years, a new research direction formed focusing on thevisible-light-induced photocatalytic activity of heterostructures of wide-bandgapsemiconductors with noble metal NPs, the latter exhibiting a surface plasmonresonance in the visible spectral range. This effect was christened as “plasmonicphotocatalysis” [16, 70, 71] and was first accepted sceptically, but a number ofreports on various photocatalytic transformations and photoelectrochemical

44 2 Semiconductor-Based Photocatalytic Systems …

processes that can be performed by illuminating the semiconductor/metal NPs withthe visible light was growing steadily, showing good perspectives of this phe-nomenon for the solar light harvesting [16, 34, 70–73].

The NPs of noble metals—gold, and silver reveal intense absorption bands in thevisible spectral range as a result of electron gas oscillations in a surface layer of themetal NPs that is referred to as surface plasmon resonance (SPR). The SPR effectcan be observed only for NPs (roughly smaller than 100 nm) and not for thecorresponding bulk metals. The spectral parameters of SPR absorption band dependon the metal type, NP size and shape, dielectric parameters of the dispersivemedium (solvent), nature of species adsorbed on the NP surface, on the proximityof neighboring metal NPs and many other factors [70–72]. For sphericalnon-aggregated silver and gold NPs the SPR maxima can be found around 390–400and 530–550 nm, respectively.

The SPR absorption of gold NPs, though being quite intense and fitting to thesolar spectrum, does not result in an interband electron transition and generation ofadditional free charge carriers, as it happens at the above-bandgap photoexcitationof semiconductors. Therefore, the Au NPs cannot act similarly to conventionalmolecular spectral sensitizers that inject an electron into the wide-bandgap semi-conductor after the photoexcitation. The fact fed the skepticism concerning thereality of the “plasmon photocatalysis” phenomenon when it was only emerging inthe field of solar light harvesting. Meanwhile, more and more reports on the pho-tocatalytic transformations occurring under excitation into the SPR band of variousgold/semiconductor heterostructures were steadily accumulated, some reports pro-viding photoaction spectra (dependences of the QY of a photoreaction on theexcitation wavelength) coinciding with the absorption spectra of Au NPs [74–79].In attempts to interpret these processes, several alternative mechanisms were pro-posed including the heat transfer from Au NPs to the semiconductor resulting in theinterband electron transition, ionization of the surface states of semiconductor NPsunder the influence of the electromagnetic field of SPR-excited Au NPs, and others.However, a number of recently reported scrupulous and sophisticated studiesshowed that Au NPs excited into the SPR band can indeed inject “hot” electronsinto the CB of wide-bandgap semiconductors, such as titania, in the cases when theFermi level of photoexcited metal NPs shifts higher than the Schottky barrier on thesemiconductor-metal interface (Fig. 2.4) [13, 34, 70–73].

For the plasmonic NPs smaller than 20 nm the hot electrons exhibit a broadspectrum of energies falling within the range from EF,M to EF,M + hv, while largerparticles exhibit much smaller hot electron energies close to EF,M and therefore forthe larger metal NPs the probability of the hot electron injection is much lower.

The electrons with an energy lower than the Schottky barrier relax through theelectron-electron and electron-phonon interactions. After the hot electron injection,a metal NP recompenses via a hole transfer to a water molecule (resulting in the O2

evolution) or to another sacrificial donor, similarly as it happens with the pho-toexcited molecules of dye sensitizers or the photoexcited semiconductor NPs. Thehot electron injection probability depends also on the distance to the semiconductorsurface that should be covered by a hot electron before the internal relaxation

2.1 Photocatalytic Systems Based on the Wide-Band-Gap … 45

occurs, as well as on the density of states on the semiconductor surface that canaccommodate a hot electron [71]. In this chapter, the effect of plasmonic lightabsorption in the semiconductor-based photocatalytic for the hydrogen evolutionwill be discussed only concisely. A series of recent reviews covers the issue ofplasmonic photocatalysis much more extensively and can serve as a perfect guidefor further development of this area [13, 16, 34, 70–73, 80].

The most popular plasmonic photocatalyst for hydrogen production is probably aTiO2/Au combination. The SPR-enhanced H2 evolution under illumination with thevisible light (typically with k > 420–450 nm) was observed in the presence ofnanocrystalline TiO2/Au heterostructures [74, 81–84], N-doped TiO2 decoratedwith Au NPs [78], mesoporous TiO2/Au composites [75] and aerogels [77], porousflat TiO2/Au electrodes [85], TiO2/Au photonic crystals [86]. Mixed Au/Pt NPsdeposited onto the surface of TiO2 nanosheets can play a double role, the goldproviding SPR for the visible light harvesting, while Pt acting as a co-catalyst ofhydrogen evolution [87]. The photoaction spectrum of TiO2/Au composite as aphotocatalyst of H2 evolution was found to be very similar to the absorptionspectrum (Fig. 2.5) indicating unambiguously on the participation of SPR-excitedgold NPs in the photochemical transformations.

Direct participation of Au NPs in the photocatalytic reaction was clearlydemonstrated for a mesoporous TiO2/Au heterostructure evolving hydrogen fromaqueous solutions of ascorbic acid when excited into narrow spectral windows of500 ± 20 and 550 ± 20 nm [75]. No H2 was detected in such conditions for thepure titania. It is notable that the excitation into the 500 ± 20 nm window results ina higher rate of hydrogen evolution because the energy of hot electrons depends on

Fig. 2.4 Plasmonic energy conversion: electrons from occupied energy levels are excited abovethe Fermi energy. Hot electrons with energies high enough to overcome the Schottky barrieruSB = uM − vS are injected into the conduction band Ec of the neighboring semiconductor, whereuM is the work function of the metal and vS is the electron affinity of the semiconductor. DOS isthe density of states, EF,M and EF,S–Fermi level of the metal and metal/semiconductorheterojunction, Ev—valence band of semiconductor. Reprinted with permissions from Ref. [72].Copyright (2014) Macmillan Publishers Limited

46 2 Semiconductor-Based Photocatalytic Systems …

the excitation energy and the probability of injection is higher for theshorter-wavelength light.

The effect of SPR-induced enhancement of the photocatalytic/photoelectrochemical H2 evolution is of general nature and can be observed forother photoactive semiconductors, such as nanocrystalline CdS [88] and Ta2O5/Ta3N5 [89], ZnO nanorods (NRs) [76, 90], La2Ti2O7 nanosheets [77]. The CdS/Auheterostructures exhibited not only an enhanced activity in the photocatalytic waterreduction but also a much higher photostability in aqueous Na2S/Na2SO3 solutionsas compared to the individual CdS [88].

A spectacular plasmon enhancement of the photocatalytic/photoelectrochemicalH2 evolution was also observed for branched ZnO nanowires (NWs) decorated withgold NPs [76]. The deposition of Au NPs onto a highly developed surface ofbranched ZnO NWs resulted in a much broader spectral response extending to700–750 nm. The incident-photon-to-current-efficiency (IPCE) spectra (analogs ofphotoaction spectra) of ZnO and ZnO/Au NWs excited by UV light (Fig. 2.6,panel 1) are roughly the same revealing no appreciable spectral differences andcorresponding to the direct interband electron excitation of the semiconductorphotocatalyst. However, the ZnO/Au heterostructures, as opposite to bare ZnONWs, revealed a spectral response in the visible range with the band shape mim-icking closely the SPR band shape of gold NPs (Fig. 2.6, panel 2).

Recently, the family of “plasmonic” photocatalysts was joined by GCN/Aunanoheterostructures. Graphitic carbon nitride absorbs only a limited portion of thevisible light up to 460–470 nm and can be sensitized to longer-wavelength irra-diation by the deposition of Au NPs [91, 92].

Similarly to gold, Ag NPs exhibit an intense SPR band in the visible spectralrange and can induce the effect of spectral sensitization when excited into the SPRband, however, in this case the sensitization effect is not so obvious, as for gold,because the SPR band maximum of Ag NPs is closer or even overlapped with theabsorption spectra of the most photoactive semiconductors. The effect of

Fig. 2.5 Absorption andphotoaction spectra of TiO2

Evonik P25 and a P25/Auheterostructure. Reprintedwith permissions from Ref.[74]. Copyright (2016)American Chemical Society

2.1 Photocatalytic Systems Based on the Wide-Band-Gap … 47

plasmon-enhanced H2 evolution was reported for N-doped TiO2/Ag heterostruc-tures [78], ZnO/Ag [93], GCN/Ag [94]. The ZnO NRs decorated with triangular Agnanoprisms revealed a higher plasmon-activated photoactivity as compared withsimilar heterostructures based on regular spherical Ag NPs as a result of a strongelectromagnetic field generated on the prism edges [94].

2.2 Photocatalytic Systems Based on the Binary and MoreComplex Semiconductor Heterostructures

Absorption of the visible light by a narrow-band-gap component of binary semi-conductor composites also results in the electron injection to the CB of awide-band-gap component, where, with the participation of a co-catalyst, hydrogenformation occurs. The photogenerated hole remains separated from the electron andreacts with a donor. Such spatial separation of the charge carriers is a reason fortypically high rates of the photocatalytic hydrogen evolution over binaryhetero-structures composed of narrow-band-gap metal sulfides and wide-band-gapmetal oxides [95–100]. Figure 2.7 shows a scheme of charge transfers in a

Fig. 2.6 IPCE spectra of P-ZnO, B-ZnO, Au/P-ZnO, and Au/B-ZnO NW photoanodes collected at1.23 V versus NHE in a wavelength window of 300–420 nm (panel 1) and 420–850 nm (panel 2).Reprinted and adapted with permissions from Ref. [76]. Copyright (2014) American ChemicalSociety

48 2 Semiconductor-Based Photocatalytic Systems …

photocatalytic system based on a very popular TiO2/CdS composite. In the furtherdiscussion we will define binary and more complex heterostructures by listing theircomponents one after other separated by a “/” symbol. Typically we will put to theleft of the slash a “basic” component of the heterostructure, for example, awide-bandgap semiconductor (TiO2) onto which another component, such as anarrow-bandgap sensitizer (CdS) is deposited or attached.

One of the most broadly studied semiconductor sensitizers for the hydrogenevolution is cadmium sulfide as well as related solid solutions, such as cadmiumzinc sulfide. For example, Vis-sensitive photocatalysts of the hydrogen evolutionfrom aqueous solutions of 2-propanol or Na2S–Na2SO3 were formed by thedeposition of CdS NPs on the surface of nanocrystalline titania [101, 102]. Thephotoactivity of the heterostructures increases remarkably with a decrease of theCdS NP size as a result of a size-dependent increase of the CB energy of CdS NPs[103, 104]. The photocatalytic activity of such systems can be further boosted bymodification with fullerenes acting as photoelectron acceptors [105].

Ternary TiO2/CdS/Pt heterostructures can be used for the photocatalytic H2

evolution directly from the sea water after addition of sacrificial donors (Na2S andNa2SO3) [106]. An important factor governing the photocatalytic properties ofternary TiO2/CdS/Pt composites in the water reduction is a “correct” spatial orga-nization of components [107–109]. A photocatalyst produced by the Pt NP pho-todeposition on the surface of preliminarily formed binary TiO2/CdSheterostructure showed by an order of magnitude lower photoactivity than similarcomposites prepared by the CdS NP deposition onto pre-formed TiO2/Ptheterostructure [107]. The same photocatalytic behavior is typical for a broad rangeof ternary TiO2/CdS/M composites, where M = Au, Ag, Pd, Pt [110].

TiO2/CdS/Pt heterostructures produced by the impregnation of TiO2/CdS com-posites with chloroplatinic acid followed by the thermal Pt(IV) reduction exhibiteda higher photoactivity in the H2 evolution than similar composites produced via thephotocatalytic Pt(IV) reduction [108]. In this case, the difference in photoactivityalso owes to the fact that the thermally deposited Pt NPs are attached mostly to the

Fig. 2.7 Scheme of spatialseparation of thephotogenerated chargecarriers in a CdS/TiO2

heterostructure and the H2

formation under theVis-illumination

2.2 Photocatalytic Systems Based on the Binary … 49

TiO2 surface, where the water reduction takes place, while the photo-depositedmetal NPs are distributed randomly between the CdS and TiO2 NPs.

Ternary WO3/CdS/Au heterostructures built on the basis of inverted WO3 opalsare more active photocatalysts of the water splitting than their analogs producedfrom randomly structured tungsten oxide. An advanced photoactivity of theopal-based photocatalysts stems from a more efficient light absorption due to themultiple scattering and refraction of light in the regular pores of the opals [111].

A shape anisotropy of zinc oxide NRs [112] and nanobelts [113] favors to thespatial charge carrier separation in ZnO/CdS heterostructures reflecting in a highphotocatalytic activity in the H2 evolution from water/methanol mixtures.

The ion exchange capability of a Ti(IV)-modified MCM-41 zeolite was used toform 2.5-nm CdS NPs in the zeolite pores [114]. After the Pt NP photodepositionsuch heterostructure exhibits a high photocatalytic activity in the hydrogen evo-lution from aqueous sodium sulfite solutions exceeding strongly that of bulk cad-mium sulfide. The photocatalytic H2 evolution from aqueous TEA solutions wasalso observed in the presence of CdS NPs immobilized on MCM-41 with a fractionof Si atoms replaced with Zr and Ti [115].

The heterostructures of CdS NPs [116–118] and Cd0.5Zn0.5S NPs [119] withTiO2 NTs are efficient Vis-sensitive photocatalysts of the hydrogen evolution fromaqueous Na2S/Na2SO3 solutions. The CdS NP deposition the surface of TiO2

nanoplates [120] and meso-porous microspheres [121] with prevailingly exposed{001} facets yields efficient photocatalysts of the water reduction by lactic acid[120]. The photoactivity of such heterostructures exceeds that of similar compositesproduced from conventional titania crystals because the {001} lattice face of titaniaexhibits a relatively higher efficiency of the interfacial electron transfer [120, 121].

Spatial separation of the photogenerated charge carriers between the hosttitanosilicate matrices ETS-4 and ETS-10 comprising ultra-thin (–O–Ti–O–Ti–O–)x“quantum wires” and CdS NPs deposited into the host pores results in a high pho-toactivity of such heterostructures in the H2 evolution from aqueous Na2S/Na2SO3

solutions [122]. Similar approaches were used to introduce CdS NPs into the inter-layer galleries of layered titanates [123–127], niobates [128–130] and tantalates [129,131, 132], as well as layered mixed ZnII and CrIII hydroxides [133]. In such com-posites, thewater reduction toH2 occurs on co-catalyst NPs (Pt, Ni orRuO2) depositedon the outer photocatalyst surface, while the oxidation of a sacrificial donor (Na2S orNa2SO3) involves CdS NPs attached to the inner surface of the layered host material.Due to the spatial separation of the charge carriers the photocatalytic activity of thecomposites exceeds strongly that of individual cadmium sulfide or a mechanicalmixture of CdS and a layered metallate [123, 124, 128].

To achieve favorable conditions for the formation of CdS NP-basedheterostructures and to promote photocatalytic processes with their participation,a preliminary treatment of layered host materials is often performed aimed at anexpansion of the interlayer galleries. For example, the intercalation of propylamineand [Pt(NH3)4]Cl2 complex into the interlayer space of HNbWO6 expands con-siderably the inner voids between the layers favoring to the secondary intercalationwith CdII and ZnII [134]. The annealing and sulfurization of such material resulted

50 2 Semiconductor-Based Photocatalytic Systems …

in a HNbWO6/Cd0.8Zn0.2S/Pt heterostructure exhibiting a Vis-light-driven photo-catalytic activity in the hydrogen evolution from aqueous solutions of sodiumsulfite.

A treatment of co-deposited CdS and TiO2 NPs with titanium(IV) chloridefollowed by he annealing [135] assures the formation of TiO2/CdS heterostructureswith a good mechanical and electronic contact between the CdS and TiO2 NPsfavoring to the charge transfers between the components. The highest photocurrentand photocatalytic activity in the hydrogen generation were observed at 80 wt.%titania content [135].

Directed migration of the photogenerated charge carriers—from a layer ofcadmium selenide to TiO2 NTs through an intermediary CdS layer in ternary TiO2/CdS/CdSe heterostructures contributes to their high photoelectrochemical activityin the hydrogen evolution from aqueous solutions of Na2S/Na2SO3 or ethyleneglycol with QY reaching *9.5% [136]. A similar effect was observed fornanoheterostructures formed by CdS “nanoflowers” grown on the surface of TiO2

NT arrays (Fig. 2.8a) [137].A very efficient charge transport from the visible-light-sensitive CdSe NPs to the

thin (*5 nm thick) titania NSs results in a strong non-additive enhancement of thephotocatalytic hydrogen evolution from aqueous Na2S/Na2SO3 solutions [138].Coupling of the TiO2 NSs to CdS NPs via a molecular bridge—bifunctionalmercaptopropionic acid (MPA) anion allows to double the H2 evolution efficiencyas compared to the bare CdSe NPs, while direct (without linkers) deposition of thesensitizer NPs onto the TiO2 NSs increases the efficiency by another *100%(Fig. 2.8b). An electron paramagnetic resonance (EPR) study showed that Ti4+ ionscan be converted into Ti3+ by the photogenerated CB electrons and act as chargetransfer mediators to the CdSe NPs. Due to the fact, the annealing of TiO2 NSs that

Fig. 2.8 a A scheme of the photoelectrochemical H2 evolution with “TiO2 nanotube/CdSnanoflower” heterostructures; b The rate of photocatalytic hydrogen evolution in the presence ofTiO2 nanosheets and NS/CdSe heterostructures. Reprinted with permissions from Ref. [137](a) and [138] (b). Copyright (2015, 2016) Elsevier (a) and American Chemical Society (b)

2.2 Photocatalytic Systems Based on the Binary … 51

caused their aggregation and a loss of the surface area had a detrimental effect onthe photocatalytic activity of both bare TiO2 NSs and TiO2 NS/CdSe nanocom-posites [138].

The sol-gel deposition of 10–20-nm titania NPs on the surface of microcrys-talline cadmium sulfide followed by the photodeposition of Pt NPs results in aternary CdS/TiO2/Pt composite that reveals photocatalytic properties in thehydrogen evolution from aqueous Na2S/Na2SO3 solutions [139]. Other platinumgroup metals can also act as co-catalysts of this process forming the followingactivity sequence: Pt > Rh > Pd > Ru. Isotopic studies in a similar system, whereH2S was used as a sacrificial electron donor, showed that H2 is evolved at theexpense of the decomposition of both H2O and H2S [140].

CdS/TiO2 heterostructures based on cadmium sulfide NWs [141] exhibited amuch higher photocatalytic activity in the H2 evolution from aqueous Na2S/Na2SO3

solutions than non-modified CdS NWs. Spatial separation of the photogeneratedcharge carriers between the heterostructure components results in the separation ofoxidative and reductive steps of the process—the water reduction to H2 occurs onthe TiO2 NPs, while the sacrificial donors are oxidized on the surface of CdS NWs.

Despite the fact that sacrificial donors, especially sodium sulfide and sulfite canefficiently quench the oxidative photocorrosion of cadmium sulfide, some inevitablerelease of inherently toxic CdII ions can be expected for the CdS-based photocatalysts.This hazard stimulates a constant search for other less toxic narrow-bandgap sensi-tizers capable of competing with cadmium sulfide in the hydrogen evolutionefficiency.

A particular attention in this search is paid to ternary and quaternarymetal-chalcogenide NPs, such as indium-based chalcopyrite CuInS2 and AgInS2(AgIn5S8) NPs and quaternary kesterite Cu2ZnSnS4 NPs. These compounds haverelatively narrow bandgaps of around 1.4–1.8 eV and reveal strong absorptionbands covering the entire visible spectral range thus making such NPs ideal lightharvesters for the photocatalytic hydrogen evolution systems.

The CuInS2/TiO2 [142, 143] and TiO2/AgIn5S8 [144] heterostructures revealed aphotocatalytic activity under the photoexcitation over almost the whole visiblespectral range. The sensitization of Ag NP-decorated ZnO NW arrays with CuInS2NPs results in *100-fold enhancement of the photoelectrochemical hydrogen pro-duction efficiency under the Vis-illumination as compared to the original NWs [145].

The quaternary kesterite NPs were successfully employed as a light harvester forthe photoelectrochemical hydrogen production over a ZnO/CdS/Cu2ZnSnS4heterostructure based on ZnO NWs [146]. The mutual positions of the CB and VBlevels of the components are ideally suitable for a cascade transfer of the photo-generated electrons from the outer kesterite layer to the CdS buffer layer to the ZnONW layer (Fig. 2.9a). After the cascade the electrons are collected into the electriccircuit and transferred to a Pt counter electrode, where the H2 evolution occurs,while the CdS/Cu2ZnSnS4 (CZTS) light-harvesting layer is regenerated via theoxidation of a sacrificial donor (Na2S/Na2SO3) [146]. The photocurrent (and cor-respondingly, H2 on the counter electrode) is generated under the illumination in the

52 2 Semiconductor-Based Photocatalytic Systems …

entire visible range (400–700 nm) with the light-to-current conversion efficiencyreaching *45% (Fig. 2.9b).

Quaternary NPs of other types, such as Cu–Ga–In–S NPs [147], are also cur-rently probed as spectral sensitizers with the aim of combining a high absorptivityin the visible spectral range and a “suitable” band positions for the efficient chargetransfer to TiO2.

Among the binary non-toxic semiconductor sensitizers, a special attention isfocused on bismuth and antimony chalcogenides that combine a high sensitivity tothe visible light, a relative stability and band positions favorable for the chargeinjection into TiO2, ZnO, and other wide-bandgap semiconductor materials.Thermal hydrolysis of thiourea in the presence of Bi(NO3)3 and nanocrystallineTiO2 yields TiO2/Bi2S3 heterostructures manifesting a photocatalytic activity in theVis-light-driven H2 production from aqueous Na2S/Na2SO3 solutions [148]. Thephotoactivity of the composite was found to be much higher than that of bismuthsulfide alone and maximal—at the equimolar content of the components [148]. TheTiO2/Bi2S3 composites produced by a solvothermal method from 10 to 15-nmtitania NPs exhibited photocatalytic properties in the hydrogen evolution fromwater/methanol mixtures [149].

Spatial separation of negative and positive charge carriers in thenanoheterostructures of titania and copper(I,II) oxides as well as the capability ofcopper oxides of accumulating electrons and decreasing the water reduction over-voltage allowed to carry out the photocatalytic H2 evolution under the illuminationwith the visible light [138, 150–160]. A p/n heterojunction also forms on theinterface between TiO2 and copper phosphide Cu3P NPs enabling efficient sepa-ration of the photogenerated charge carriers and the water reduction with anapparent QY (measured at a certain wavelength) of 4.6%, which is by an order ofmagnitude higher than for sole titania NPs [161].

Fig. 2.9 a A scheme of charge transfers in ZnO NW/CdS/Cu2ZnSnS4 (CZTS) system; b IPCEspectra of ZnO NW-based heterostructures with CdS and CZTS NPs. Reprinted with permissionsfrom Ref. [146]. Copyright (2015) The Royal Society of Chemistry

2.2 Photocatalytic Systems Based on the Binary … 53

The photocatalytic Vis-light-driven formation of hydrogen was observed also inthe presence of In2O3/In2S3 [162], CuO/ZnO [163], In1−xGaxN/ZnO [164], CuFeO2/SnO2 [165], RuO2/TiO2 [166], and CuAlO2/TiO2 [167] nanoheterostructures.

Along with the development of photocatalytic systems based on traditionalsemiconductors, a search is also performed for new photosensitive semiconductingmaterials combining the visible light sensitivity with a capacity to act as spectralsensitizers for wide-band-gap semiconductors. At that, a special attention is paid tocarbon materials—fullerenes, carbon NTs, etc. For example, a composite of mul-tiwall carbon NTs with titania modified by Ni NPs exhibited a photocatalyticactivity in the water reduction when excited by the visible light [113, 168]. It wasassumed that the photoexcitation of carbon NTs results in the electron injection intothe TiO2 CB followed by the electron transfer to the Ni NPs where the final act ofthe water reduction occurs. The oxidized NTs are then regenerated at the expense ofmethanol oxidation.

Recently, good perspectives were shown for the sensitization of wide-bandgapsemiconductor materials with carbonaceous nanostructured species, such as carbonNPs and nanodispersed carbon nitride. The carbon NPs can be produced bythermal/electrochemical decomposition of a variety of organic precursors andcontain a partially aromatic carbon core and an outer shell abundant with variousfunctional groups [169, 170]. They absorb light in broad and intense bandsextending throughout the visible range and can strongly bind to the most of thephotoactive wide-bandgap semiconductors typically used for the photocatalyticprocesses. A comprehensive account of recent successes and challenges associatedwith the utilization of carbon NPs in the photocatalysis can be found in [169]. Forexample, the nanocrystalline titania can be sensitized to the visible light by thecarbon NPs [171, 172] produced via hydrothermal treatment of vitamin C [171] orby the electrochemical destruction of graphite [172, 173]. Such heterostructuresexhibited almost by an order of magnitude higher photocatalytic activity in the H2

evolution from water/methanol mixtures under the illumination with the “white”light (k > 400 nm) as compared to the bare titania.

Graphitic carbon nitride is often called a “rising star” of the semiconductorphotocatalysis as it combines a unique set of properties including chemical stability,sensitivity to the visible light, “appropriate” positions of CB and VB energiesallowing both for the water reduction and oxidation to occur simultaneously. Thismaterial will be discussed in details later in the section devoted to new photoactivematerials. Here, we only mention the role of GCN as a component of composite H2

evolution photocatalysts. It was found that spatial separation of the photogeneratedcharg carriers imparts TiO2/GCN heterostructures with a photocatalytic activity inthe Vis-light-driven hydrogen evolution from water with no sacrificial donors [174–176] as well as from aqueous solutions of methanol [177] or TEA [178].

The exfoliation of GCN into a-few-layer or even single-layer CN sheetsincreases strongly its activity as a hydrogen evolution photocatalyst, both in theindividual state and when incorporated into complex nanoheterostructures. It isreported that the photocatalytic activity of composites of titania NRs with the GCN

54 2 Semiconductor-Based Photocatalytic Systems …

nanosheets produced by an ultrasound treatment of the bulk GCN is by far higherthan the photoactivity of a mixture of TiO2 NRs and the unexfoliated GCN [179].

As GCN has a bulk bandgap of 2.7 eV it can also be a subject to the spectralsensitization. Such effect was achieved for CdS/GCN [180] and ZnIn2S4/GCN[181] composites, as well as for CdSe NP-decorated hollow GCN spheres [182].

The GCN NSs can be used as a “mat” to accommodate wide-bandgap semi-conductor NPs. For example, GCN/TiO2 heterostructures produced by thesolvothermal deposition of titania NPs onto GCN NSs demonstrated the rates ofphotocatalytic hydrogen evolution by *10 and *20 times higher than thoseobserved in the presence of sole TiO2 and the bulk GCN [183]. A similar effect wasalso achieved for *20-nm InVO4 nanocrystals grown on the GCN sheets [184].

2.3 Photocatalytic Systems Based on the Metal-DopedWide-Band-Gap Semiconductors

Doping of the wide-band-gap semiconductors with metal ions introduces newoccupied local states in the band gap that can be excited by the visible light andsupply electrons to CB (Fig. 2.10). The CB electrons participate then in the waterreduction while the holes localized on the dopant states get filled by electrons froma sacrificial donor or water [137, 185–189].

Visible-light-sensitive photocatalysts of the hydrogen evolution from aqueoussolutions of sodium sulfite were prepared by doping ZnS with PbII [190], NiII [191]and CuII [192]. Such photocatalysts can function without additional co-catalysts.The visible-light absorption by these compounds originates from the photoinducedelectron transition from the local dopant states in the CB of zinc sulfide. The pho-tocatalytic activity of ZnS:PbII is maximal at 1.4 wt.% lead content and decreasesconsiderably at a higher dopant concentration (more than 2%) as a result of theformation of a separate PbS phase. Additional co-doping of the photocatalyst withhalogen anions results in a 3-fold increase of the photoactivity, the effect originatingfrom a relaxation of the lattice strain and a decrease of the number of non-radiativerecombination sites [192]. At an optimal dopant concentration of 4.3 wt.% the

Fig. 2.10 A photocatalyticsystem for the hydrogenproduction based on NiII-doped titania

2.2 Photocatalytic Systems Based on the Binary … 55

ZnS:CuII-based photocatalytic system exhibits an apparent hydrogen production QYof 3.7% [192]. A maximal increment of the photocatalytic H2 evolution rate afterdoping of zinc sulfide with Cu(II)—by a factor of 11 is observed after the intro-duction of *5 mol% copper [193]. At the same time, for ZnS:NiII the peakphotocatalytic activity in the water reduction was achieved already at 0.1 mol% Nicontent [191].

Doping of titania with BiIII imparts this semiconductor with a photocatalyticactivity in the H2 evolution from water/ethanol mixture under the Vis-illumination[194]. In a similar way, doping of SrTiO3 with Cr [195, 196] and Rh ions [197,198] yields Vis-light-sensitive photocatalysts of the hydrogen evolution fromaqueous methanol [195, 196] and pure water splitting [197, 198].

The substitution of Ti4+ with Cr3+ or Fe3+ in titania crystals requires a com-pensation of the excessive negative lattice charge and induces self-oxidation of CrIII

to CrVI and the release of molecular oxygen [199–202]. The recombination ofcharge carriers at anion vacancies forming after the O2 subtraction decreases thephotocatalytic activity. To balance the charge and to increase the stability andactivity of Cr-doped titania an equimolar amount of Ta5+ or Nb5+ should addi-tionally be introduced into the lattice. An increase of the photocatalytic activity of adoped semiconductor as a result of the charge compensation was also observedfor the co-doping of TiO2 and SrTiO3 with combinations of Ni2+/Ta5+ [199] andCr3+/Sb5+ [200].

A strong doping-induced photoactivity enhancement of a semiconductor host inthe water reduction was observed after the introduction of Bi3+ into NaTaO3 [203],Zn2+ into SrTiO3 and BaTiO3 [204], Ag

+ into BiVO4 [205], cations of Y, La, Ce orYb into NaTaO3 [206], and Zn2+ into Ga2O3 [207].

The photocatalytic water reduction to H2 under Vis illumination was reported forZnS and SrTiO3 doped with La3+ [208], Ni2+-doped InTaO4 and InNbO4 [209].After the deposition of a co-catalyst (Pt, RuO2, NiOx) the latter two systemsdemonstrated an apparent QY of up to 0.66% (at k = 400 nm). Besides doping withNi2+, InTaO4 can be turned into a Vis-sensitive photocatalyst of the H2 evolution byintroducing Mn, Fe, Co, and Cu cations [210]. Doping with chromium turns aBa2In2O5/In2O3 heterostructure into a “universal” photocatalyst capable of thehydrogen evolution from water and water/methanol mixtures in the presence of Ptor Ni as well as of the O2 evolution from aqueous AgNO3 (electron acceptor)solutions [211].

A broad range of dopants—CrVI, FeIII, CoII, NiII, RuII, and PdII was used toconvert nanocrystalline Bi2O3 (Eg = 2.8 eV) into a Vis-sensitive photocatalyst ofthe water reduction [212]. Doping with palladium(II) resulted in the best charac-teristics, the fact apparently originating from in situ Pd(II) photoreduction Pd0

which can act as a co-catalyst.Almost in each system based on doped semiconductors there exists an optimal

dopant concentration range where the maximal photoactivity is observed, while ahigher dopant amount deteriorates the semiconductor activity in the water reduc-tion. For TiO2 doped with Ni2+ [213, 214] or Bi3+ [194], the maximal rate of thephotocatalytic hydrogen evolution from water/alcohol mixtures was observed at a

56 2 Semiconductor-Based Photocatalytic Systems …

1% mol. dopant content [213]. This effect is typically associated with a hindrance ofthe free migration of photogenerated charge carriers in the bulk of highly-dopedsemiconductor crystals because of an abundance of the local dopant states acting ascharge traps.

In some cases doping results in a fusion of the dopant states and a “top” of thehost VB. The effect narrows the band gap (and increases the Vis-light sensitivity ofa semiconductor) without the emergence of additional local states in the forbiddenband. For example, doping of indium titanate with a mixture of nickel and chro-mium cations results in a fusion of Ni3d, Cr3d, Ti3d/In5sp, and O2p orbitalsyielding a Vis-light-sensitive In12NiCr2Ti10O42 photocatalyst (Eg = 2.14 eV) of theH2 evolution from water/methanol mixtures [215, 216], more efficient than“mono-doped” In6NiTi6O22 (Eg = 2.48 eV) and In3CrTi2O10 (Eg = 2.00 eV) [215].

Sometimes a variation in the metal dopant nature allows switching the semi-conductor activity between the water reduction and the water oxidation. Forexample, doping of SrTiO3 with Mn2+ or Ru3+ impart this semiconductor with aphotocatalytic activity for the Vis-light-driven oxygen evolution from aqueousAgNO3 solutions [217]. At the same time, doping of strontium titanate with Ru, Rh,or Ir (1 wt.%) cations and the deposition of 0.1 wt.% Pt converts this wide-band-gapsemiconductor into a Vis-light-sensitive photocatalyst of the water reduction bymethanol demonstrating an apparent H2 evolution QY of 5.2% at 420 nm [217].

The radio-frequency magnetron sputtering technique is typically used [218–224]to produce titania films that exhibit a visible light sensitivity originating from astoichiometry deviation, that is a gradient of the O/Ti atomic ratio from the surfaceto the bulk of the films. A post-synthesis hydrothermal treatment of the filmsenhances considerably their photocatalytic activity in the water reduction as a resultof an increase of the film crystallinity and the specific surface area [222]. Thismethod was also applied to produce Ti foils decorated with titania nano-columnsoriented normally to the film surface [225] with an O/Ti ratio changing from 2.00on the column top to 1.93 at the site of the column contact with the substrate. Afterthe modification of an opposite side of a Ti/TiO2 foil with Pt NPs, it was used as aVis-light-sensitive photocatalyst of the total water splitting in a combined reactorwith membrane-separated compartments for the water reduction to H2 and the wateroxidation to O2 (Fig. 2.11a) [218, 225]. Such design allows avoiding the recom-bination between primary products of the reduction (H atoms) and the oxidation(OH radicals), which is one of the main factors limiting the H2 evolution efficiency.

Recently, a so-called “black” titania emerged as a new visible-light-sensitivephotocatalyst of the water reduction [29]. The “black” TiO2 is typically producedby treating titania with hydrogen or aluminium resulting in a massive reduction ofTi4+ to Ti3+, the latter imparting the material with a characteristic blackish-gray toblack color (Fig. 2.11b, insert).

According to [226], reduction with Al yields much deeper reduced TiO2−x

samples with the absorbance extending to longer wavelengths as compared to thehydrogen-processed titania (HP-TiO2), the absorption band encompassing the entirevisible and near IR ranges (Fig. 2.11b). A strong light-harvesting capability of the“black” TiO2 results in much higher photocurrents/H2 evolution rates in the

2.3 Photocatalytic Systems Based on the Metal-Doped … 57

photoelectrochemical/photocatalytic systems as compared to those with conven-tional nanocrystalline TiO2 powders [226], NTs [227] or mesoporous TiO2 [228].

2.4 Photocatalytic Systems Based on the Nonmetal-DopedWide-Band-Gap Semiconductors

A partial oxygen substitution in a metal oxide semiconductor lattice by othernon-metals—nitrogen, carbon, sulfur, etc., was found to be one of the most versatilemethods of tailoring the band gap of semiconductor photocatalysts. The p-orbitalsof a dopant typically have a higher energy than the p-orbitals of oxygen, so thedopant introduction results in a narrowing of the band gap without appreciableshifts of the CB edge (Fig. 2.12). The effect is explored by a so-called “banddesign” concept, that is, tailoring of the band gap and the VB position of semi-conductor photocatalysts by non-metal dopings [2, 137, 185, 187–189].

The introduction of nitrogen into the lattice of titanium dioxide NPs achieved byTiO2 synthesis in the presence of ammonia [50], results in a retardation of the NPsgrowth during the calcination, a decrease of the average NP size from 20 to 14 nm,and a shift of the light sensitivity threshold of titania to longer wavelengths. Also,the doping generates oxygen vacancies on the TiO2 NP surface that promoteadsorption of a sensitizer—eosin [50]. Such sensitized TiO2:N NPs showed a3-times higher photocatalytic activity in the hydrogen evolution from aqueous TEAsolutions as compared to undoped TiO2 NPs.

Fig. 2.11 a A combined photocatalytic reactor for simultaneous evolution of H2 and O2 fromwater in the presence of a “Pt/Ti foil/Ti NTs” heterostructure; b absorption spectra of original TiO2

and products of titania reduction with hydrogen (HP-TiO2) and aluminium (TiO2−x). Yellowbackground—solar AM1.5 irradiation spectrum. Insert: photographs of conventional nanocrys-talline titania (Evonik P25) and black TiO2−x produced by the reduction with Al. Reprinted withpermissions from Refs. [222] (a) and [226] (b). Copyright (2008, 2015) Elsevier (a) and AmericanChemical Society (b)

58 2 Semiconductor-Based Photocatalytic Systems …

N-doped TiO2 produced by the titania calcination with urea [229, 230] absorbsthe visible light with k < 600 nm and exhibits a photocatalytic activity in theVis-light-driven H2 production from aqueous solutions of Na2SO3 [229] andmethanol [230]. Of two forms of surface nitrogen—the chemisorbed N and thenitrogen substituting O atoms in the oxide lattice, it is the latter that imparts titaniawith a Vis-sensitivity and the enhanced photocatalytic activity. Among N-dopedtitania materials a higher photocatalytic activity in the Vis-light-induced watersplitting is typically observed for the mesoporous TiO2 [231–233]. The photocat-alytic activity of TiO2:N in the water reduction can be further enhanced by com-bining it with Pt NPs [234] or other electron acceptors such as graphene derivatives[235, 236].

Annealing of tantalum oxide in a stream of ammonia and water vapors (to preventthe formation of tantalum nitride) yields tantalum oxynitride TaON absorbing visiblelight in a range of k < 500 nm (Fig. 2.13a) [236–239]. The material demonstrates ahigh photocatalytic activity in the water oxidation to O2 (with an apparent QY of 30%at 420–500 nm excitation), but possesses a negligible photoactivity in the waterreduction, even in the presence of Pt (QY of 0.2% at 420–500 nm). Oppositely to pureoxide semiconductors, forwhich platinum is typically the best co-catalyst of hydrogenevolution, for the N-doped semiconductors a much higher activity is observed for aruthenium co-catalyst (QY of 0.8% and 2.1% in the presence ofmethanol and ethanol,respectively). The photocatalytic deposition of Ru NPs produces 2–4-nm particlesexhibiting a higher catalytic activity than 20–50-nm Ru NPs formed by the conven-tional impregnation/annealing [237]. Complete substitution of O with nitrogen yieldstantalum nitride Ta3N5 with Eg = 2.1 eV (Fig. 2.13a) that is also an active photo-catalyst of the water splitting [238].

The annealing in ammonia stream was used to produce zirconium oxynitrideZr2ON2 from ZrO2 [240]. The fusion of N2p-orbitals and O2p-orbitals in the VB ofzirconium oxynitride results in the bandgap shrinking to 2.6 eV. After the pho-todeposition of 5 wt.% Pt, Zr2ON2 crystals exhibited a photocatalytic activity in theH2 evolution from aqueous solutions of methanol as well as the O2 evolution fromsilver nitrate solutions under the Vis-illumination. Solar-light-sensitive photocata-lysts of the water reduction/oxidation were prepared by a partial nitridation of ZrO2/Ta2O5 composite [241]. In a similar way, layered LaTaON2 and Y2Ta2O5N2

Fig. 2.12 A scheme of aphotocatalytic system for thehydrogen production based onN-doped titania

2.4 Photocatalytic Systems Based on the Nonmetal-Doped … 59

perovskites were produced [242] exhibiting a photocatalytic activity in thehydrogen evolution from water/ethanol mixtures in the presence of Pt and Ru NPs.

The introduction of a nitrogen dopant into the Sr2Nb2O7 perovskite yields aseries of Sr2Nb2O7−xNx compounds that preserve the layered structure and, due to acontribution of the N2p-orbitals into the VB, exhibit a photocatalytic activity in thehydrogen evolution from water/methanol mixtures under the Vis-illumination[243].

A nitrogen-doped solid solution of gallium and zinc oxides (Ga1−xZnx)(N1−xOx)with x = 0.18 was used as a photocatalyst of the water reduction that, in a com-bination with a mixed co-catalyst Rh2–yCryO3, exhibited a QY of 6% at 420–440 nm [244]. The co-catalyst was produced in situ via the photocatalytic reductionof KCrO4 over a (Ga1−xZnx)(N1−xOx)/Rh composite that, in turn, was synthesizedby the photocatalytic deposition of Rh NPs [245]. A layer of chromium oxide onthe metal surface prevents reverse reactions between H2 and O2 allowing the(Ga1–xZnx)(N1−xOx)/Rh2−yCryO3 heterostructure to function as a photocatalyst ofthe total water splitting. Active co-catalysts for this system were also produced bythe semiconductor impregnation with a mixture of rhodium salts and rutheniumcarbonyl Ru3(CO)12 followed by annealing [246–248]. The nitridation of a mixtureof germanium and zinc oxides yields a compound (Zn1+xGe)(N2Ox) that exhibits aphotocatalytic activity in the water reduction under illumination with the visiblelight [249].

N-doping of indium oxide narrows its band gap from 3.5 to 2.0 eV and impartsthe semiconductor with a sensitivity to the visible light [250]. The photocurrent(proportional to the water reduction rate) generated under the Vis-illumination bythe In2O3:N electrode is by a factor of 2 higher than that for the undoped indiumoxide, and by a factor of almost 50—than the photocurrent generated by a TiO2:Nreference photoelectrode [250].

The photoelectrochemical hydrogen generation was realized in a system, wherea TiO2 NT array incorporated with Pt NPs acted as a cathode, while a photoanode

Fig. 2.13 Diffuse reflectance spectra of Ta2O5, TaON, and Ta3N5; b a layout of thephotoelectrocatalytic system for water splitting based on C-doped and Pt-decorated TiO2 NTs.Reprinted with permissions from Refs. [238] (a) and [251] (b). Copyright (2003, 2007) Elsevier(a) and American Chemical Society (b)

60 2 Semiconductor-Based Photocatalytic Systems …

was formed by the carbon-doped TiO2 NTs (Fig. 2.13b) [251]. The photoanodewas produced by the sonoelectrochemical anodization of titanium foil in a mixtureof NH4F with ethylene glycol followed by the annealing in the H2 atmosphere. Thecell demonstrated a photocurrent QY of 8.5% [251].

By the calcination of titanate NTs at 600 °C in a CO stream, 8–42 mol% carboncan be introduced without the formation of a separate titanium carbide phase [252].The fusion of O2p- and C2p-orbitals in the VB results in the bandgap shrinking to2.2 eV and a corresponding expansion of the spectral sensitivity range. By com-bining the carbon-doped TiO2 NTs (a photoanode) with Pt (a cathode) theVis-light-driven water splitting to H2 and O2 was achieved [252, 253].

An alternative approach to the C-doped TiO2 consists in the burning of Ti foilsin the carbon-enriched flame [254, 255]. The carbon doping results in a bandgapreduction from 3.20 to 2.65 eV as well as in the formation of a filled sub-band1.6 eV above the VB top. This material was tested as a photoanode for the pho-toelectrochemical water splitting and showed a QY of 13% under the illuminationwith “white” light in aqueous 5.0 M NaOH solution. The C-doping increaseselectrode surface porosity favoring additionally to the photoelectrochemical reac-tion [254, 255].

Sulfur-doped TiO2 nanocrystals produced by a mechanochemical treatment of amixture of titania with S8 were used for the photoelectrochemical water splitting underthe Vis-illumination [256]. The photocatalytic water reduction or oxidation (dependingon the type of co-catalyst—Pt or IrO2) under the Vis-illumination (420–480 nm) withthe participation of indium-lanthanum oxysulfides was observed in [257].

2.5 Photocatalytic Systems Based on the Metal-SulfideSemiconductors

Among the “solar” hydrogen production systems based on narrow-band-gapsemiconductors a leading role is evidently played by metal-sulfide photocatalysts,mainly CdS, that is, however, photochemically unstable and liable to the photo-corrosion. By this reason, a further development of the photocatalytic systems forthe hydrogen production based on metal-sulfide semiconductors requires newmethods of the photocorrosion mitigation. Also, a search is performed for newmetal-sulfide materials that do not contain cadmium, lead, and other acutely toxicmetals. Among the challenges in this field is also a search for new co-catalysts thatdo not contain precious platinum group/noble metals. New ways of photoactivityenhancement of the metal-sulfide semiconductors are constantly probed via acareful design of composite materials and harnessing of the quantum size effectsinherent in the nanocrystalline metal-chalcogenide semiconductors.

The anti-corrosion stability can be achieved via a combination of metal-sulfideNPs with various photochemically passive carriers. For example, CdS NPs formedin the pores of zeolites were found to be photochemically stable when used as a

2.4 Photocatalytic Systems Based on the Nonmetal-Doped … 61

photocatalyst of the hydrogen evolution from water/alcohol mixtures [128, 258–263]. The photoactivity of such heterostructures depends considerably on the carrierstructure and increases from zeolite L to SBA-15 to zeolite Y [258].

A combination of a high photocatalytic activity and photostability was observedfor CdS NPs stabilized by the colloidal silica [264] and organic polymers [265,266], as well as for CdS NPs deposited on the surface of carbon nanofibers [267],aluminium oxide [268], silica gels [269–271], and glasses [272, 273]. Theglass-incorporated CdS NPs revealed a high photostability and can be used asvisible-light-sensitive photocatalysts for the hydrogen evolution from aqueous H2Ssolutions with a QY of 17–18% [272].

Interaction between cadmium(II) salts and polyvinylidene sulfide yields5–30-nm CdS NPs regularly dispersed over the polymer surface [274]. Suchcomposite exhibits photochemical stability and a high (up to 20%) QY of thehydrogen evolution from aqueous H2S solutions.

The interest to nanocrystalline cadmium sulfide is greatly stimulated by a strongdependence of the electron properties of CdS NPs on the crystal size (d) at d < 5–6 nm. A variation of the CdS NP size in this range is accompanied by pronouncedchanges of the optical and photochemical NP properties allowing for a tuning of thespectral sensitivity range and efficiency of the NP-based photocatalytic systems.This feature is excellently exemplified by the photocatalytic systems for thehydrogen production based on size-selected CdS NPs decorated with Pt NPco-catalyst [275]. The H2 evolution QY decreases from around 17 to *11%, as theCdS NP size increases from 2.8 to 4.6 nm. The dependence was interpreted in termsof a size-dependent driving force of photoinduced charge transfer from CdS tovacant states of Pt NPs (Fig. 2.14a). However, the photoactivity increase iscounter-weighted by a “blue” shift of the absorption band edge of CdS NPs as theirsize is reduced (Fig. 2.14b), resulting in a partial loss of the solar light harvestingability. To compensate for this detrimental effect, a double-chamber photocatalyticreactor was proposed [275], where shorter-wavelength light is absorbed selectivelyby smaller and more active 2.8-nm CdS NPs, while a longer-wavelength portion ofthe light passes through the first chamber and is absorbed in the second chamber bylarger CdS NPs. In this way, a 50% increment of the H2 evolution efficiency can beachieved for the double-chamber systems as compared with a single photocatalyticreactor with a mixture of smaller and larger CdS NPs [275].

Similarly to cadmium sulfide, a reduction in the size of MoS2 NPs from 15–25to *5 nm was found to result in almost doubled photocatalytic activity in the H2

evolution, both types of NPs being much active than bulk molybdenum sulfide [276].Alternatively to CdS, bismuth, indium, iron, ruthenium, zirconium and silver

sulfides, as well as related solid solutions can be used as water reduction photo-catalysts. For example, RuS2 NPs immobilized on thiol-modified polystyrene beadsexhibited a photocatalytic activity in the hydrogen evolution from water/2-propanol[277].

The earth abundant pyrite FeS2 was shown to be a suitable candidate for theVis-light photo-electrochemical H2 generation [278].

62 2 Semiconductor-Based Photocatalytic Systems …

A solvothermal synthesis of tin sulfide starting from Sn(II) salts yields a mixedSnS2/SnS nanostructure with the intertwined tin(IV) and tin(II) sulfide layers.A lattice mismatch between the layers generates inherent sulfur vacancies resultingin a Eg narrowing by around 0.3 eV (to *2 eV) as compared with pure SnS2 [279].Such composite showed an excellent activity in the H2 evolution under illuminationwith a “blue” LED source (400–500 nm).

Layered zinc indium sulfide and its composites with metal NPs exhibiteda photocatalytic activity in the Vis-light-driven H2 evolution from aqueousNa2S/Na2SO3 solutions [280–292]. The photoactivity of ZnIn2S4 was found toincrease proportionally to the post-synthesis hydrothermal treatment duration as wellas to the concentration of cetyltrimethyl ammonium bromide acting as a template.The dependence was assumed to originate from a deformation of the ZnIn2S4 crystallattice resulting in a dipole moment in the semiconductor interlayer space that favorsto the photogenerated charge carriers separation. The copper(II) doping of ZnIn2S4expands its spectral sensitivity to around 800 nm, the maximal rate of photocatalytichydrogen production registered at 0.5 wt.% copper content [293].

Good perspectives as Vis-sensitive photocatalysts of the water reduction couldbe envisaged for a number of ternary/multinary metal sulfides with a narrow bandgap that suit perfectly as visible light harvesters and can potentially induce the waterphotodecomposition—CaIn2S4 with a band gap of 1.76 eV [294, 295], AgGaS2(Eg = 2.48 eV) [296], CuGaS2 [297], CuIn1−xGaxS [298], (CuGa)1−xZn2xS2 [299,300], Zn1−2x(CuGa)xGa2S4 [301], Cu3SnS4 (1.38 eV) [302], and Cu2ZnSnS4(Eg = 1.75 eV) [303–307]. Some of these materials were studied as nanocrystallinematerials, while for others the effects of nano-scaling are still to be explored.

A photocatalytic activity in the hydrogen evolution from aqueous Na2S/Na2SO3

solutions under the Vis-illumination was observed for mesoporous agglomerates of

Fig. 2.14 Band diagram (a) and absorption spectra (b) of 2.8–4.6-nm CdS NPs; c a scheme of adual photocatalytic reactor for the H2 evolution. Reprinted with permissions from Ref. [275].Copyright (2015) The Royal Society of Chemistry

2.5 Photocatalytic Systems Based on the Metal-Sulfide … 63

CuInS2 NPs (Eg = 1.53 eV) modified by Pt NPs [308], for CuIn5S8 [309] andCuIn0.7Ga0.3S2 films [310], CuInS2/NaInS2 nanoheterostructures [311], (CuIn)xZn2(1−x)S2 (x = 0.01–0.50) microspheres [312–317], non-stoichiometric Cu–In–Zn–SNPs attached to reduced graphene oxide (RGO) sheets [318], as well as for thenanocrystalline (CuAg)xIn2xZn2(1−2x)S2 solid solutions [319–322]. The Vis-lightsensitivity of these materials originates from a contribution of Cu3d- andS3p-orbitals into the VB and In5s5p- and Zn4s4p-orbitals—into the CB of mixedsulfide semiconductor.

Nanoporous ZnS–In2S3–Ag2S solid solutions demonstrated a photocatalyticactivity in the water reduction without additional co-catalysts [323]. Similarly,ternary sulfide AgInZn7S9 (Eg = 2.3 eV) is capable of the photocatalytic hydrogenevolution from water with no co-catalysts and electron donors present in the system[324]. When used together with Pt NPs and a sacrificial donor (Na2S/Na2SO3), thisquaternary photocatalyst showed an H2 production QY of *15%. CdIn2S4 NTsrevealed the photocatalytic properties in the water reduction with no additionalsacrificial donors with a QY of up to 17% [325]. A high photocatalytic activity inthis process was also observed for ZnIn2S4/CdIn2S4 [326, 327], In2S3/ZnIn2S4[328], and CdS/ZnIn2S4/RGO heterostructures [329]. A mixed sulfide AgIn5S8(Eg = 1.77 eV) modified by Pt NPs was used as a Vis-sensitive photocatalyst of thehydrogen evolution from aqueous sulfide/sulfite solutions with a QY of 5.3% [330].

Various layered metal-sulfide nanomaterials, both in the form of nanometergrains and especially as a few-layer (single-layer) NSs are of a great potential forthe solar H2 production. For example, layered NaInS2 (Eg = 2.3 eV) was found tobe an efficient photocatalyst of the water reduction with a QY of 6% [331], whilebulk indium sulfide remains inactive in this reaction. The 2.5-nm In2S3 NPs pro-duced by ion exchange/sulfidation in the pores of titania-containing Ti-MCM-41zeolite exhibited pronounced photocatalytic properties in the water reduction [332].The activity stems from an efficient photoinduced electron transfer from In2S3 NPsto the zeolite host and inhibition of the subsequent electron-hole recombination.

First-principles calculations showed perspectives of single-layer and a-few-layerzirconium sulfide as a visible-light-sensitive photocatalyst (Eg = 1.9–2.0 eV) of thewater splitting [333], though these predictions still require an experimentalverification.

A comprehensive review of the photochemical water splitting on layered tran-sition metal dichalcogenides (TMDs) can be found in [334]. The exfoliation ofsome of TMDs into single-layer sheets can result in a dramatic increase of theoptical bandgap thus providing an additional driving force for the water splittingprocesses as illustrated in Fig. 2.15a for MoS2. A band diagram, shown inFig. 2.15b for a series of reported single-layer TMDs, provides a notion of possiblecandidates for the utilization in the photocatalytic/photoelectrochemical systems forthe water splitting.

Ultra-thin sheets of MnSb2S4 with Eg = 1.9 eV and a thickness of 0.76 nmproduced by the spontaneous thermal exfoliation of hydrazine-intercalated bulkmaterials revealed promising properties as a photocatalyst of the water reductionwith a peak QY of 0.14% [335].

64 2 Semiconductor-Based Photocatalytic Systems …

The films of layered Bi2S3 with Eg = 1.28 eV produced by the electrodepositioncan be used as an efficient and stable photocatalyst of the Vis-light-driven hydrogenevolution from aqueous Na2S solutions [336]. In a similar photocatalytic systembased on a Bi2S3/zeolite Y composite the QY of H2 evolution reached 0.12% [337].

Mixed sulfides CdxZn1−xS are typically capable of the photocatalytic waterreduction without additional co-catalysts [338–341]. For these semiconductors, as arule, a dome-shaped relationship between the photocatalyst composition (theparameter x) and the rate of hydrogen evolution is observed. Such dependence isquite non-trivial because both CB and VB potentials of CdxZn1−xS solid solutionsincrease as the Cd is gradually substituted with Zn (Fig. 2.16) and, therefore, amonotonous dependence of the photocatalytic activity of cadmium-zinc sulfidecrystals on their composition should be expected.

The exact position of the maximum on this relationship still remains contro-versial, most probably, due to differences in synthesis methods of CdxZn1−xS solidsolutions that can affect considerably their photochemical behavior. For example, itwas found [338] that a maximal apparent QY of hydrogen evolution from aqueoussulfide/sulfite solutions, 10.2% at 420 nm, corresponds to x = 0.8. According to[339], the highest H2 production QY with the participation of CdxZn1–xS crystalscan be observed in a range of x = 0.25–0.30. Studies of the photocatalytic activityof cadmium-zinc sulfides precipitated on paper revealed two distinct photoactivitymaxima corresponding to x = 0.5 and 0.2 [342]. The peak activity of CdxZn1−xSmicrospheres produced by hydrothermal synthesis [343] was found at x = 0.1.

A detailed transient flash photolysis study [344] showed that a dependencebetween the capability of CdxZn1−xS NPs to accumulate an excessive negativecharge (photoinduced polarization of NPs) and the NP composition also has adome-shaped character. The maximum position on this dependence corresponds toa maximum position on the dependence between the composition of CdxZn1−xSNPs and the QY of hydrogen evolution (Fig. 2.17). Therefore, a direct relationship

Fig. 2.15 a Energy diagram for bulk and single-layer MoS2; b band edge positions of single-layersulfide TMDs relative to the vacuum level, including the redox potentials for the H+/H2 and O2/H2O couples at pH = 0. Reprinted with permissions from Ref. [334]. Copyright (2015) The RoyalSociety of Chemistry

2.5 Photocatalytic Systems Based on the Metal-Sulfide … 65

between the photocatalytic activity of CdxZn1−xS NPs and their electric capacitancecan be concluded from these dependences.

The photocatalytic water reduction by electron donors was reported for morecomplex semiconductors based on cadmium-zinc sulfide, such as Cd0.1SnxZn0.9−2xS solid solutions [345]. The compound with x = 0.01 exhibited a 1.5-fold higherphotocatalytic activity than undoped Cd0.1Zn0.9S. Doping of Cd0.5Zn0.5S with BiIII

increased considerably the QY of photocatalytic H2 evolution from aqueousNa2S/Na2SO3, solutions that reached *10% at 0.1 mol% dopant [346].

Along with the stability issue, various strategies are probed to enhance thephotoactivity of metal-sulfide narrow-bandgap semiconductors suitable for the solarwater splitting. At that, the most fruitful approaches include using (i) loosely

Fig. 2.16 CB and VB potentials of CdxZn1−xS solid solutions with various Cd/Zn ratios.Reprinted with permissions from Ref. [8]. Copyright (2010) Elsevier

Fig. 2.17 a Dependence between the quantum yield Ф(H2) of the photocatalytic H2 evolutionfrom aqueous Na2SO3 solution and the Cd molar fraction x in CdxZn1−xS NPs. b Effect of thephotoetching time of nanocrystalline Pt/CdS on the photocatalytic H2 evolution rate. Reprinted andadapted with permission from Ref. [348]. Copyright (2008) Elsevier

66 2 Semiconductor-Based Photocatalytic Systems …

aggregated nanocrystalline and mesoporous metal sulfides; (ii) biphase nanocrys-talline metal-sulfides and materials with a graded (gradient) composition;(iii) chemical/photochemical treatment of a nanocrystalline metal-sulfide aimed atthe elimination of the surface defects/ligands; (iv) metal-sulfide NPs with an ani-sotropic shape; (v) composites with the water oxidation and reduction processesseparated in space.

This list and selected examples given below provide a mere illustration of avariety of the possible ways of influencing/enhancing the photoactivity ofmetal-sulfide semiconductors. The attractiveness of the metal-sulfide photocatalyticsystems for the solar H2 production can be also enhanced by using broadly avail-able raw materials and contaminants as sacrificial donors.

An ultrasound treatment of reaction mixtures during deposition of cadmiumsulfide on the surface of aluminum and magnesium oxides favors to the formationof mesoporous CdS with an average pore diameter of 5.5 nm and a particle size of4–6 nm [347]. Such materials exhibited a high photocatalytic activity in thehydrogen evolution from Na2S/Na2SO3 solutions in the presence of Pt group metalswith the catalytic activity of metals increasing from Rh to Pd to Pt.

The photocatalytic activity of nanocrystalline CdS can be boosted by a photo-chemical treatment in aqueous air-saturated solutions of formic acid [348]. Thetreatment decreases the NP size as a result of the oxidative photocorrosion.Simultaneously, the cleavage of a surface NP layer eliminates the surface defectsparticipating in the electron-hole recombination and thus the overall photocatalyticactivity of CdS NPs is increased by more than 2 times (Fig. 2.17b) [348].

A ligand shell on the surface of colloidal cadmium sulfide NPs is necessary toensure the individual character of each NP and prevent their aggregation. However,the ligands even as small as MPA can present an obstacle for the photoinducedelectron transfer to water molecules and inhibit the photocatalytic hydrogen evo-lution. As shown in [349], the CdS NPs “stripped” from the surface ligands andstabilized only by a surface charge reveal a two orders of magnitude higher pho-toactivity in the H2 evolution from aqueous Na2SO3 solutions as compared to theMPA-capped NPs with the same NP size. The colloidal CdS NPs stabilized elec-trostatically by an outer shell of sulfide ions revealed a 8–9 times higher efficiencyof the photocatalytic H2 evolution from water/hydrazine mixtures than similar NPscapped with MPA [350].

Typically, different crystalline modifications of a semiconductor differ in theband energies and can be combined to produce heterostructures, where efficientseparation of the photogenerated charge carriers becomes possible. The mostwell-known example of such a heterostructure is the commercially availablenanocrystalline titania Evonik P25, consisting of 70–80% anatase and 20–30%rutile, that is extensively used as a benchmark photocatalyst for comparing thephotoactivity of different (and not only oxide) semiconductor materials. The anataseand rutile have slightly offset CB potentials enabling a one-way migration of thephotogenerated electrons from anatase to rutile.

Similar heterostructures can be composed of different-phase metal sulfidesemiconductors. For example, cubic CdS NPs (Eg = 2.6 eV) can be deposited on

2.5 Photocatalytic Systems Based on the Metal-Sulfide … 67

the surface of hexagonal microcrystalline cadmium sulfide with Eg = 2.3 eV [351].The charge separation occurring in this system due to a difference in the CBpositions imparts the composite with a high photocatalytic activity in the hydrogenevolution from aqueous sodium sulfite solutions. Similarly to the above-discussedCdS/TiO2/Pt system [107], here also the co-catalyst localization plays an importantrole. The highest activity was observed for a composite produced by the photo-catalytic deposition of Pt NPs on the microcrystalline CdS prior to the formation ofa layer of cubic CdS NPs.

A dramatic acceleration (by around 500 times) of the photocatalytic H2 pro-duction was observed after the deposition of a thin (*2.5 nm) shell of cubic CdSon the surface of hexagonal CdS NRs [352]. The concentric core/shell CdS NRsalso exhibited unrivaled photostability and even the possibility of hydrogen evo-lution in aerobic conditions, indicating a very efficient spatial separation of H2 andO2 generation events on different locations of the photocatalyst surface.

Mixed-phase elongated CdxZn1−xS NRs with thin hexagonal wurtzite layers“sandwiched” between thicker cubic zinc blend domains were reported to be anefficient hydrogen evolution photocatalyst [341]. The CB and VB edge offsets andan internal electric field existing in such heterostructures result in a directed flow ofthe photogenerated CB electrons from wurtzite (WZ) to zinc blend (ZB) and the VBholes—in the opposite direction (Fig. 2.18a). In this way, the water reduction andsulfite oxidation occur at different sites and the electron-hole recombination isefficiently suppressed without additional metal co-catalysts [341]. In the case ofZnIn2S4, the CB potential of a cubic modification (−1.5 V vs. NHE) is morenegative than that of a hexagonal phase (–1.1 V vs. NHE) and thus, a combinationof the ZB and WZ NPs results in a flow of the photogenerated electrons from thecubic to the hexagonal zinc indium sulfide phase [291]. As a result, the ZB/WZZnIn2S4 composite is a highly superior photocatalyst of the water reduction ascompared to individual ZB and WZ phases (Fig. 2.18b).

A number of photochemically active metal chalcogenides can form solid solu-tions with almost ideally mixed components. As discussed above, cadmium andzinc sulfides form solid-solution compounds with the composition varying frompure CdS to pure ZnS via intermediate CdxZn1−xS phases with intermixed metalcations and a joint S2− sublattice. As CdII and ZnII have a different reactivity tosulfide-ions and the sulfides of cadmium and zinc have a different solubility, it ispossible, by properly adjusting the synthesis conditions, to form CdxZn1−xS crystalswith a gradient of cadmium or zinc concentration, for example, the crystals enrichedwith CdII on the surface.

The cadmium and zinc sulfide possess quite different bandgaps (2.4 and 3.8 eVfor bulk cubic CdS and ZnS, respectively) and differ also in the CB and VBenergies, the CB level changing from −0.8 V versus NHE for CdS to −1.8 V versusNHE for ZnS [8]. In a graded CdxZn1−xS crystal with a CdII-enriched outer layerthe CB potential on the crystal surface is, therefore, lower (closer to that of pureCdS) than in the bulk of the crystal, where it grows and shifts closer to the CBpotential of pure ZnS. The CB level gradient directs the photogenerated electrons tothe surface and prevents their recombination with the photogenerated VB holes.

68 2 Semiconductor-Based Photocatalytic Systems …

This charge separation principle was realized for a CdxZn1−xS/SiO2

heterostructure produced by the sulfurization of a graded CdxZn1−xO/SiO2 com-posite [271]. The surface area of CdxZn1−xS crystals can be cleaved layer by layervia the bombardment with heavy Ar+ ions thus allowing to reveal with the XPS agraded structure of such crystals, the Cd to Zn molar ratio decreasing from *1 to 2on the surface to *1–3 in the bulk of the crystals (Fig. 2.19a).

The graded crystal structure favors to the directed flow of photogeneratedCB electrons along a ECB downfall, that is, toward the crystal surface, where theycan react with water evolving gaseous hydrogen. Due to this effect, the gradedCdxZn1−xS/SiO2 heterostructure showed a high photocatalytic activity in the water

Fig. 2.18 a Upper panel: HRTEM image of a typical CdxZn1−xS NR. The blue squares and greenarrows index the segments of WZ and ZB structures. Lower panel: migration of charge carriers inCdxZn1−xS ZB/WZ nanojunctions; b kinetic curves of the hydrogen evolution in the presence ofsole ZB and WZ ZnIn2S4 as well as a ZB/WZ nanojunction. Insert: an HRTEM image of theZB/WZ nanojunction. Reprinted and adapted with permissions from [341] (a) and [291] (b).Copyright (2016) Americal Chemical Society (a) and Elsevier (b)

Fig. 2.19 a Illustration of a graded structure of CdxZn1−xS crystals; b photocatalytic H2 evolutionactivity of ZnS/SiO2, CdS/SiO2, CdxZn1−xS/SiO2 and CdxZn1−xS with different structures underthe visible light irradiation. Reprinted and adapted with permissions from Ref. [271]. Copyright(2016) The Royal Society of Chemistry

2.5 Photocatalytic Systems Based on the Metal-Sulfide … 69

reduction leaving behind both the individual sulfides and cubic/hexagonal CdxZn1−xS NPs with a regular non-graded structure (Fig. 2.19b). The electron flowarriving on the crystal surface reduces undercoordinated surface cadmium speciesto Cd0 that acts as a metal co-catalyst of the photoprocess and therefore, noadditional co-catalyst is necessary [271].

The liability of metal-sulfide semiconductors to the photocorrosion can be usedto transform the surface layer of the sulfide crystals into a metal oxide or a mixedoxide/sulfide composition with a tunable band structure. For example, the photol-ysis of ZnS microcrystals under 254-nm UV light at the ambient air pressure andmoisture converts the surface layer of the microcrystals into a graded ZnSxO1−x

interface [353]. A combination of the surface etching with Auger spectroscopicprobing revealed that the surface layer of such photochemically treated ZnSmicrocrystals is composed of a zinc oxysulfide solid solution NPs with the oxygencontent decreasing from the surface to the crystal bulk (Fig. 2.20a). The gradedregion extends to 80–100 nm matching the depth of light penetration into the zincsulfide microcrystals.

As the CB level of pure ZnO (*−0.5 V vs. NHE at pH 7) is lower than the CBlevel of ZnS, the CB electrons photogenerated within the graded ZnSxO1−x layerare directed toward the surface, where they can participate in the water reduction.Similarly to the case of CdxZn1−xS, this effect results in a drastic enhancement ofthe hydrogen evolution, that proceeds also without additional co-catalysts due to apartial reduction of surface Zn species to Zn0 [353]. The photocatalytic activity ofthe photoproduced ZnSxO1−x layer depends on the duration of photolysis(Fig. 2.20b), most probably, due to a balance between the thickness of oxidized

Fig. 2.20 a Distribution of Zn, S, C, and O atoms in the photolyzed single ZnS crystal derivedfrom the Auger spectroscopic data. The etching time of 10 min roughly corresponds to a 100-nmetching depth. The signal of adventitious carbon stems from surface contaminations. b The rate ofthe photocatalytic H2 evolution (RH) from water/ethanol mixture under UV illumination(k > 320 nm) in the presence of microcrystalline ZnS, ZnS/ZnSxO1−x heterostructures producedby the photolysis and a mechanical mixture of ZnS and ZnO. Reprinted with permissions fromRef. [353]. Copyright (2016) Elsevier

70 2 Semiconductor-Based Photocatalytic Systems …

layer and the light penetration depth, and supersedes by an order of magnitude theactivity of a mechanical mixture of ZnS and ZnO of the same composition.

A photocatalytic activity increase can be achieved by using semiconductorcrystals with an anisotropic (non-spheric) shape, where the charge migration rate isdifferent depending on the crystal axis. For example, CdS NWs with a length of3–4 lm and a diameter of 50 nm act as a more efficient photocatalyst of H2 evolutionfrom aqueous Na2S/Na2SO3 solution than isotropic nanocrystalline CdS [354].

The commercial attractiveness of the photocatalytic systems for the hydrogenproduction based on the narrow-band-gap semiconductors can be enhanced by a cutin the costs of sacrificial donors. For example, a CdS/LaMnO3 heterostructure canbe used as a Vis-sensitive photocatalyst for the hydrogen production either fromconventional sulfide/sulfide donors [355] or from the broadly available biomass[356]. The rate of hydrogen evolution from the biomass under the poly–chromaticillumination is comparable to that observed in the sulfide/sulfite/containing systemsdue to the presence of perfect donors—methanol and formic acid in the partiallyfermented biomass.

A special interest is evoked by the issue of the utilization of various industrialwastes as sacrificial donors for thewater reduction. This approach can be illustrated bythe photocatalytic hydrogen evolution from hydrogen sulfide solutions in ethanola-mine and other aliphatic amines in the presence of CdS/Pt nanoheterostructures [357].Such solutions are abundantly produced as wastes of the carbon and natural gasindustries as well as in the technologies of crude oil desulfurization. The aminesreadily dissolve hydrogen sulfide and favor its ionization and proton release that canbe reduced to the molecular hydrogen. At the same time, a high solubility of poly-sulfide anions as products of the S2− oxidation promotes desorption of S2�x from thephotocatalyst surface and prohibits reverse reactions between polysulfide and thephotogenerated CB electrons.

The sulfide-capped colloidal CdS NPs were found to be an excellent photocat-alyst of H2 evolution by using aqueous hydrazine as a sacrificial electron donor[350]. The hydrazine contains 12 wt.% of hydrogen and is considered as apromising liquid hydrogen carrier with N2 as a sole product of the oxidativedecomposition.

2.6 Emerging Semiconductor Photocatalysts for the SolarHydrogen Production

A constant search for new photocatalysts of the water splitting is currently under way.Each new compound emergingwithin the focus of attention of possible photocatalyticapplications is typically tested as a hydrogen evolution photocatalyst. The emergingphotocatalysts are of exceedingly broad scope making almost impossible their rig-orous classification, the most trend-making of them being new narrow-bandgapsemiconductors with a lattice formed by metal ions and oxygen (oxides, metallates,

2.5 Photocatalytic Systems Based on the Metal-Sulfide … 71

etc.), metal-chalcogenide semiconductors, carbonaceous species like graphitic carbonnitride and carbon nanoparticles, and metal-organic frameworks.

Some of the new semiconductor photocatalysts can be referred to as “exotic”because they have only recently entered the attention spotlight of the photocatalysiscommunity and only a little is known about the potential and perspectives of suchmaterials. The following subsection provides an account of the photocatalytichydrogen evolution systems based on the semiconductor materials that can becurrently characterized as “emerging” photocatalysts.

Photocatalysts with a chalcogenide lattice. The examples of hydrogen-evolvingphotocatalysts with a chalcogenide-based lattice are confined mostly to theabove-discussed sulfide semiconductors because metal selenide and telluridesemiconductors have typically too low bandgaps to induce the waterreduction/oxidation reactions and reveal an unacceptably low photostability even inthe presence of sacrificial electron donors.

The photocatalytic hydrogen evolution from sulfide/sulfite solutions under theVis-illumination of CdSe nanobelts [358] was a rare example of the photochemicalactivity of cadmium selenide and, probably, the first evidence of its capability toinduce the water reduction. Later, size dependences of the photocatalytic activity ofCdSe NPs in the water reduction were reported [359, 360]. The bandgap of CdSeNPs is broadened considerably as the NP size decreases from 3.1 to *1.8 nm dueto the QSEs resulting in a steep growth of the photocatalytic activity with a NP sizereduction (Fig. 2.21a). No activity is typically observed for the CdSe NPs largerthan 3–4 nm.

As the size of CdSe NPs is reduced, a strong “blue” shift of the NP absorptionband edge is observed (Fig. 2.21a, insert) resulting in a drastic loss of visible lightharvesting capability. Nevertheless, smaller CdSe NPs still reveal a high photo-catalytic activity (when the H2 evolution rate is normalized to the light absorbance)as a result of a larger driving force of the CB electron transfer to water molecules[361]. This example again illustrates that a reasonable trade-off between the lightharvesting capacity and the CB energy should be maintained for the quantum-sizedsemiconductor NPs to attain the maximal QY of the solar hydrogen production[359, 360]. Also, the control of surface chemistry of the nanocrystalline CdSe playsa crucial role in the photocatalytic water reduction. For example, passivation of thesurface defects of CdSe NRs tipped with Pt NPs with an atomically-thick CdS layerincreases the H2 production rate by a factor of 6–7 [362].

Nanocrystalline CdSe was used as a light-harvesting material in a photoanodedesigned for the water oxidation to O2 with the hydrogen evolution occuring on acounter electrode [363]. The cadmium selenide was protected against the corrosion/photocorrosion by a sputtered layer of metallic cobalt that was partially convertedinto a water oxidation co-catalyst—CoPO4 by the oxidative etching in a phosphatebuffer [363].

The reported assortment of other (Cd-free) metal selenide photocatalysts ofwater reduction is apparently limited to the nanocrystalline c-In2Se3 (a Pt NPco-catalyst, TEA as a sacrificial donor, Eg * 1.6 eV) [364].

72 2 Semiconductor-Based Photocatalytic Systems …

Photocatalysts with oxygen-containing lattice. A broad range of variousvisible-light-sensitive oxides, both stoichiometric and mixed-valence, as well asnumerous metallates (ferrites, aluminates, cuprates, stannates, borates, etc.) can beused as photocatalysts of the solar H2 production in the bulk and nanocrystallineforms. Among these compounds a special attention was paid to layered perovskites(tantalates, niobates, titanates, ferrites, etc.) used as bulk materials, NPs, NRs,nanofibers, hollow spheres, etc. and exhibiting a lucrative combination of a highstability and activity in the hydrogen evolution.

The CuGa2O4 and CuGa2−xFexO4 spinels were recently introduced asVis-sensitive photocatalysts of the H2 evolution from aqueous H2S solutions in thepresence of a NiO/RuO2 co-catalyst [365]. A photocatalytic activity in the hydrogenevolution from aqueous sulfide/sulfite solutions was observed for CuLaO2

(Eg = 2.33 eV) modified with the photodeposited Pt NPs [366], as well as forCuLaO2.62 [367] and CuAlO2 [368].

Ultra-thin (*3 nm) tin niobate NSs produced by the hydrothermal exfoliation ofthicker 2D SnNb2O6 particles showed a remarkable photocatalytic activity in thevisible-light-driven water reduction, which is 4 and 14 times higher than the activityof the 50-nm thick particles and bulk niobate, respectively [369].

Nanocrystals of CuFe2O4 and ZnFe2O4 [370, 371] were successfully tested asphotocatalysts of the water reduction under the illumination with the visible light.The photocatalytic properties in similar systems were reported for MnO2 [372],Ga2O3 [373], NiO [374], Sn3O4 [375], nanocrystalline LaFeO3 [376, 377] andLaMgxTa1−xO1+3xN2−3x (x � 1/3) perovskites [378], nanoparticles of SrSnO3

[379] and MnCo2O4 [380], copper borate [381], a family of M2BiSbO7 (M = Ga,Fe, Gd) [382], as well as for ZnAg3SbO4 [383]. Also, a number of newUV-light-sensitive semiconductor photocatalysts for the water reduction wasreported in recent years, such as titanium phosphate [384], gallium borate [385],Zn2GeO4 [386, 387], LaCO3OH [388], Sm2GaTaO7 [389], that can potentially besensitized by molecular dyes and narrow-band-gap semiconductor NPs.

Xe lamp spectrum2.0 nm2.8 nm4.0 nm

Fig. 2.21 a Theabsorption-corrected rate ofthe photocatalytic H2

evolution with theparticipation of size-selectedCdSe NPs. Insert: absorptionspectra of 2.0–4.0 nm CdSeNPs overlapped with a typicalspectrum of a Xe lamp (usedfor the simulation of the solarspectrum). Reprinted andadapted with permissionsfrom Ref. [359]. Copyright(2012) The Royal Society ofChemistry

2.6 Emerging Semiconductor Photocatalysts … 73

A new family of In-based photocatalysts of the water reduction under theVis-illumination [370] includes InVO4 (Eg = 1.9 eV), InNbO4 (Eg = 2.5 eV), andInTaO4 (Eg = 2.6 eV). Differences in the band gap among these compoundsoriginate from a contribution of the V3d-, Nb4d- and Ta5d-orbitals in the respectiveCB positions. Mesoporous InVO4 produced by a template synthesis exhibited ahigher photocatalytic activity in the hydrogen evolution than the non-porousnanocrystalline indium vanadate [390].

A recently reported series of homologous compounds ZnxIn2O3+x (x = 4, 5, or 7)showed a unique combination of a strong light harvesting capability with a highmobility of the photogenerated charge carriers [391]. A layered structure of suchcompounds comprised of light-absorbing Zn(In)O4(5) layers alternated with InO6

layers provides abundant charge collection sites and transport channels (Fig. 2.22a).An enhanced sensitivity of these compounds to the visible light as compared toindividual ZnO and In2O3 (Fig. 2.22b) originates from a hybridization between theO2p and In4d orbitals in the Zn(In)O4(5) layers resulting in an upward shift of the VBedge and a reduced bandgap (2.57–2.67 eV depending on x) [391]. Under thefull-spectrum illumination the mixed compounds revealed an almost 5-fold higherphotocatalytic activity in the water reduction than the parent zinc and indium(III)oxides (Fig. 2.22c).

Photocathodic production of solar hydrogen. In conventional photoelectro-chemical systems for the solar hydrogen production a visible-light-sensitivesemiconductor/heterostructure attached to a conductive substrate acts as a pho-toanode. The light-excited photoanode oxidizes water or a sacrificial donor presentin the electrolyte and injects a photogenerated electron into the conductivesubstrate-collector, that transfers it to an outer electrical circuit. Afterward, theelectrons come to a cathode, where the water reduction to molecular hydrogen takesplace. In such a system, the reduction and oxidation half-reactions are separated inspace and the H2 evolution efficiency is determined by the catalytic properties of thecathode and the photoactivity of the anode responsible for the donor oxidation.

In recent years, alternative photoelectrochemical systems emergedwhere the targetreaction—the water reduction to H2 takes place as a direct consequence of the lightabsorption by a photocathode, while the photogenerated VB holes are transferred intothe electric circuit to a counter anode, where the water oxidation and O2 evolutionoccur. An account of the current progress in this field can be found in [392].

The assortment of semiconductor materials suitable for the photocathodeapplications is rather limited, because they should comply with a set of rigorousrequirements, in particular, the sensitivity to the visible light (that is, have a rela-tively narrow enough bandgap) and a high conduction band potential, that is, ECB

negative enough to induce the water reduction (Fig. 2.23) [392]. Also, a photo-cathode should be coupled with an appropriate oxygen-evolving electrocatalyst toensure the cyclic performance of the photoelectrochemical system and the efficientregeneration of the photocathode.

As compared to the photoanodes comprised typically of n-type semiconductorsand prone to the photocorrosion, the photocathodes usually use p-type conducting

74 2 Semiconductor-Based Photocatalytic Systems …

semiconductor materials that appear much more stable toward the photocorrosionprovided that the photogenerated electrons are efficiently transferred to watermolecules.

Copper oxides are very promising visible-light-sensitive materials for the pho-tocathodes of the hydrogen-evolving photoelectrochemical systems [393]. Toachieve efficient light harvesting with copper oxides it is suggested to use Cu2O(CuO) NW arrays rather than conventional planar semiconductor electrodes [393,394]. A mixed Cu2O/CuO heterostructure can be easily sensitized with a layer ofcopper sulfide via the ion exchange reaction and additionally decorated with aPt NP co-catalyst exhibiting quite spectacular 3.6% efficiency of the solar lightconversion [395]. The reduced graphene oxide (RGO) was found to be an efficient

Fig. 2.22 a Schematic representation of a Zn4In2O7 photocatalyst with different functional parts;b photographs of mixed zinc indates and pure zinc and indium oxides; c the average photocatalytichydrogen production rate under the full range irradiation (k � 250 nm) and the visible lightirradiation (k � 420 nm). Reprinted with permissions from Ref. [391]. Copyright (2016)Americal Chemical Society

Fig. 2.23 Band edgepositions of several typicalphotocathodic semiconductormaterials. Reprinted withpermissions from Ref. [392].Copyright (2015) The RoyalSociety of Chemistry

2.6 Emerging Semiconductor Photocatalysts … 75

co-catalyst for 1D Cu2O photocathodes and a promising candidate to replace theconventional noble metal co-catalysts [394].

Another promising and stable photocathode material is nickel oxide. As NiO hasa weak absorbance in the visible spectral range it can be coupled with other robustand strongly-absorbing semiconductors, for example, graphitic carbon nitride. Dueto a band edge offset in the NiO/GCN heterostructure, the nickel oxide layer acceptsthe photogenerated VB holes from GCN while the CB electrons of GCN reducewater to H2 [396]. A similar cascade hole transfer from GCN to a p-type semi-conductor is observed in a GCN/CoSe2 composite attached to the top of a Simicrowire array photocathode [397].

As Fig. 2.23 shows, silicon can also be utilized as a photocathode material forthe solar hydrogen production. The Si photocathodes are typically designed as NWarrays [397–400] that can be additionally decorated with other semiconductors[397, 399] and metal co-catalysts (Pt) [399], and covered with a titania layer toprotect the photocathode from the photocorrosion [399]. The activity of silicon canbe boosted by making porous photocathodes with a highly developed surface areafrom Si NPs [401]. Amorphous silicon coupled to a triple Ni–Mo–Zn alloy as ahydrogen evolution catalyst and a Co-containing water oxidation catalyst forms aphotoelectrochemical cell for the hydrogen evolution with a QY of 4.7% under the1 Sun illumination [402].

The visible-light-induced H2 evolution was observed on a photocathode formedby the CuInS2 nanodisks grown epitaxially on cubic Au NPs (Fig. 2.24a) [403]coupled to a Pt counter-electrode. The water reduction process was assumed toinvolve both the photogenerated CB electrons of CuInS2 and the “hot” electronsinjected into the semiconductor from the plasmon-excited Au NP seeds. A searchinto other possible ternary/quaternary narrow-bandgap metal-chalcogenide materi-als for the H2-evolving photocathodes is still performed resulting in ever morecomplex and highly tunable compositions, like Cu–In–Ga–Se–S [404] and Zn–Cu–In–Ga–Se [405].

A combination of a photocathode and a photoactive anode allows constructing awater splitting system functioning without any externally applied bias. For exam-ple, a titania oxygen-evolving photoanode can be combined with a hydrogen-evolving photocathode comprised of the visible-light-sensitive composite of Znphthalocyanine and fullerene C60 decorated with Pt NPs (Fig. 2.24b) [406].

A photoelectrochemical system for the H2 evolution with no external bias wasassembled from a CuGaS2/RGO photocathode and a BiVO4/CoOx photoanode[297]. The RGO layer was deposited by the direct photocatalytic reduction ofgraphene oxide on the surface of CuGaS2 photocathode enhancing dramatically itsphotoresponse to the visible light.

New “exotic” inorganic photocatalysts. Some new photocatalysts for waterreduction were recently reported that can be referred to as “exotic” because of theirquite rare (or otherwise unreported) photoactivity in the redox-processes. Many ofsuch new compounds were only tested as bulk (microcrystalline) powders to date,but are nevertheless discussed here because a strong enhancement of the

76 2 Semiconductor-Based Photocatalytic Systems …

photocatalytic activity can be anticipated for such compounds in the nanocrystallinestate thus posing a challenge for further studies.

For example, a recently discovered fibrous modification of the red phosphoruswas reported to have photocatalytic properties in the water reduction [407]. Thephotoactivity of such materials can be enhanced considerably by decreasing thecrystal dimensions via the growth restriction on the silica fibers or by an ultrasoundtreatment of the bulk material [407]. The red P was reported to have the CBpotential around −0.25 V versus NHE at pH 0 which is sufficient for the waterreduction, while the VB holes (EVB > 1.5 eV) can oxidize either water or a broadrange of sacrificial electron donors [408].

A community of rare photocatalysts of hydrogen evolution was recently joinedby silicon carbide [409–414] capable of the water reduction even in the absence ofsacrificial donors, as well as by gallium nitride with a CB potential by 0.5 V morenegative than the water reduction potential [415]. Such difference appeared to besufficient to overcome the hydrogen evolution overvoltage and produce H2 fromaqueous solutions of methanol and Na2S/Na2SO3 under the Vis-illuminationwithout additional co-catalysts.

Iron silicide b-FeSi2 revealed a photocatalytic activity in the hydrogen evolutionfrom aqueous dithionite solutions even under the illumination with the near-IR light[416]. A Vis-sensitive photocatalyst of the H2 evolution from aqueous solutions offormic acid was prepared via the deposition of Pt NPs on the surface of poly-crystalline Si [417]. The photocatalytic water reduction was also observed formacroporous silicon [418], mesoporous Si coupled to a noble-metal-freenon-stoichiometric cobalt phosphate co-catalyst [419], and Si nanowires loadedwith an iron phosphite co-catalyst [420].

Layered siloxene NSs (Fig. 2.25a) produced by a topotactical transformation ofcalcium silicide in water revealed photocatalytic properties in the water reductionwithout additional co-catalysts and sacrificial electron donors [421]. The siloxene is

Fig. 2.24 Schemes of the photoelectrochemical water splitting using Au/CuInS2 (a) and ZnII

phthalocyanine (ZnPc)/fullerene C60/Pt (b) heterostructured photocathodes. Reprinted withpermissions from Refs. [403] (a) and [406]. Copyright (2016) American Chemical Society(a) and The Royal Society of Chemistry (b)

2.6 Emerging Semiconductor Photocatalysts … 77

a direct bandgap semiconductor with a strong visible light harvesting capability anda suitable bandgap of 2.5 eV. The position of both CB and VB bands of thesiloxene are respectively negative and positive enough for both water reduction andoxidation to occur on the photocatalyst surface under the photoexcitation with thevisible light (Fig. 2.25b) [421].

Iron silicide was reported to be able to evolve hydrogen from aqueous solutions ofsodium dithionite even when illuminated with the NIR light [416]. It is a uniquesemiconductor material because it combines a narrow band gap of 0.8 eV with a CBposition of around −0.65 V (vs. NHE) which is unprecedently high for such anarrow-bandgap material. Titanium disilicide is a narrow-bandgap semiconductorthat can harvest light over the entire visible spectral range and serve as a stablephotocatalyst for the hydrogen production without additional sacrificial donors [422].

Carbon NPs. Carbon NPs (CNPs) have recently shown a high promise forapplications concerning with the light emission and absorption, including thephotocatalytic solar light harvesting [169, 170]. The CNPs can be synthesized quiteeasily by the thermal decomposition of a single or mixed organic precursor at 200–300 °C. The structure of CNPs is still a subject of discussions, as numerous XPSstudies showed them to contain simultaneously aliphatic sp3-hybridized and aro-matic sp2-hybridized carbon, sometimes also the amine- and pyridine-like nitrogen.The surface of CNPs is typically decorated with hydroxyl and carboxyl groups. TheCNPs are often called carbon quantum dots (QDs), however, the usage of this termseems not to be justified as no information on the possibility and character of QSEsis available for CNPs.

The CNPs can emit bright and broadband PL in the visible spectral range with abroad gamut—from blue to red, the fact determining the perspectives of the CNPsfor luminescent bio-labeling. Similarly to the exact internal structure of the CNPs,the PL origins and mechanisms are also a subject of vivid discussions. A specialclass of CNPs is constituted by graphene quantum dots, that is, small pieces ofgraphene functionalized with oxygen-containing groups and N dopants that reveal astrong dependence of the spectral properties on their lateral size [169, 170].

The synthesis of CNPs typically requires very simple and available precursorsmaking them an excellent candidate for the mass-scale photocatalytic applications.

Fig. 2.25 A schematic molecular structure (a) and band structure (b) of siloxene. Reprinted andadapted with permissions from Ref. [421]. Copyright (2016) The Royal Society of Chemistry

78 2 Semiconductor-Based Photocatalytic Systems …

For example, the carboxylate-terminated 2–3-nm CNPs can be produced via thepyrolysis of sodium salt of ethylenediaminetetraacetic acid (EDTA) at 350 °C. TheCNPs can be then coupled to Pt NPs and the CNP/Pt heterostructure and used as anefficient visible-light-sensitive photocatalyst of the H2 evolution from aqueoussolutions of dihydronicotinamide adenine dinucleotide (NaDH) which is a conve-nient electron/proton sources frequently used in a biochemical research practice[423].

The CNPs produced by a hydrothermal treatment of flower pollens was suc-cessfully used to sensitize GCN NSs, the GCN/CNPs assembly evolving hydrogenunder the illumination with the visible light from methanol/water mixtures with aPt NP co-catalyst [424].

Pure CNPs produced from multi-wall carbon NTs were found to possess thecapability of reducing water to hydrogen in water/methanol mixtures withoutadditional co-catalysts [425]. In the presence of Pt NPs, the CNPs revealed asuperior photoactivity as compared with the nanocrystalline TiO2 Evonik P25.

The electrochemical etching of graphite electrodes is another convenient methodof the CNPs production yielding stable suspensions of 4–5-nm CNPs. Such NPs arecrystalline with a lattice regularity of 0.321 nm typical for the parental graphiteindicating the CNPs to be small fragments of graphite stabilized by an outer shell offunctional groups (mostly COOH as revealed by the XPS) [426]. The CNPs can becoupled with layered MoS2 into a material with a pronounced photoelectrochemicalactivity in the water reduction process [427].

The scalable and benign carbonization of vegetable raw materials (such asspinach, peas, and others) was reported to produce CNPs capable of strongadsorption on the surface of nanocrystalline titania [428]. The CNPs/TiO2 com-posites were tested as a photocatalyst of the hydrogen evolution from methanol/water mixtures.

Graphitic carbon nitride. The graphitic carbon nitride is probably one of the firstartificially synthesized organic polymers but it emerged only relatively recently as auniversal and visible-light sensitive photocatalyst with brilliant perspectives in thedomains of the solar energy harvesting and the environmental photocatalysis [429–433].

Similarly to graphite, GCN is formed by planar infinite single layers that have anaromatic character and are stacked by the van-der-Waals forces (pp-interactions)with an interplanar distance of around 0.34 nm. The term “carbon nitride” does notreflect exactly the structure of single layers as they are composed of heptazine(tris-s-triazine) heterocycles bounded through tertial amine N atoms into an infinitenetwork with an intra-network periodicity of around 0.68 nm.

Two alternative structures of the single layer carbon nitride (SLCN) are pro-posed, one describing its as a regular net-like polyheptazine (Fig. 2.26a), the otherpostulating that a SLCN is composed on infinite 2D polyheptazine ribbons boundtogether by numerous hydrogen bonds (Fig. 2.26b) [429–435]. The exact structureof SLCN still remains a subject of discussion, however, the second structure seemsto be more realistic. It can be seen that the term “carbon nitride” is only a con-venient approximation to describe the stacked multilayer polyheptazine structure

2.6 Emerging Semiconductor Photocatalysts … 79

because the composition of the bulk undoped material is close C3N4, however, thisterm does not reflect the exact chemical nature of this material, though used broadlyby the historical reasons and convenience.

GCN is a direct bandgap semiconductor with the VB and CB formed respec-tively by filled and vacant C2p and N2p orbitals. The bandgap of bulk GCN isaround 2.7 eV and can vary by 0.1–0.2 eV as a result of differences in the syn-thesis, possible adventitious doping, and structural defects and, therefore, GCNabsorbs the UV and a portion of the visible light in a range of k < 460 nm. Also,the GCN has a very “suitable” position of both CB and VB levels that arerespectively negative and positive enough to allow the photogenerated chargecarriers to participate in a variety of redox-processes, including the reduction andoxidation of water [429, 431–434]. Reported data on the exact position of CB andVB levels of GCN reveal some scatter, most probably due to the above-mentioneddifferences in the synthesis nuances and structural imperfection and, most often, canbe found at around −1.0 and +1.7 V versus NHE [431–433].

The GCN has comparatively high thermal, chemical and photochemical stabilityas well as a low toxicity that distinguish this material from other inorganic semi-conductors with similar Eg, ECB and EVB parameters. The bulk GCN can be pro-duced in copious amounts from a variety of affordable precursors, such as melamine(1, 3, 5-triaminotriazine), dicyandiamide, and urea by the thermal treatment at 400–600 °C. By introducing various heteroatomic additives GCN can be doped with P,S,B etc. on the stage of the material formation [436]. Alternatively, the GCN can beannealed or treated with aggressive oxidizing/reducing agents after the synthesis tovary the C/N ratio and thus to modify its spectral and photochemical properties[429, 431–434, 436]. Various supramolecular assemblies of triazine derivativeswith other carbocyclic and heterocycles compounds can be used for the synthesis ofGCN with tailored electron properties, for example with an increased hydropho-bicity or a more extended aromatic system [436–439]. By performing the thermalcondensation of precursors in the presence of templates the GCN can be producedas porous solids. Finally, similarly to graphite, GCN can be exfoliated to form

Fig. 2.26 Alternative models of the single layer carbon nitride structure. Reprinted withpermissions from Ref. [435]. Copyright (2014) American Chemical Society

80 2 Semiconductor-Based Photocatalytic Systems …

a-few-layer and single-layer moieties or, alternatively, subjected to a partialcleavage to produce GCN NPs.

A comprehensive account on the photocatalytic systems for hydrogen evolutionand other processes based on GCN can be found in reviews [430–432, 436, 440].

The compact GCN has a relatively low specific surface area of 5–15 m2/g thuslimiting the efficiency of the charge carrier transfer from the photoexcited GCN toother components of the photocatalytic system. To increase the GCN surface areavarious “hard” (for example silica NPs [182, 441, 442]) and “soft” (organicpolymers [443], sucrose [444]) templates are used resulting in the mesoporousGCN samples active in the photocatalytic H2 evolution.

Using of uniform SiO2 micro-beads as a hard template allows producing hollowGCN spheres with a wall thickness of around 50 nm and an inner void of severalhundred nm in diameter [182, 442] (Fig. 2.27a). The wall can then be modified witha co-catalyst, like layered MoS2 [442] or Pt NPs [182] as well as with a sensitizer,for example, CdS NPs [182], forming a complex photocatalyst for the solarhydrogen evolution. The “hollow sphere” architecture is favorable for the photo-catalytic process as the co-catalyst is typically localized on the outer walls and thusthe water reduction/water oxidation half-reactions can be spatially separated. Also,the hollow spheres (HSs) are well known for the ability of a more efficient lightharvesting as a result of multiple light scattering in the inner voids of the spheres.

Alternatively, the GCN can be grafted to the developed surface of a photo-chemically inert carrier, such as mesoporous [445] and macroporous silica [446] ora zeolite [447, 448].

As in the case of inorganic semiconductor photocatalysts of the water reduction,GCN typically requires a co-catalyst to efficiently evolve hydrogen. For example,metal NP (Pt [444, 449–451], Au [91, 92], Cu [452], Ni [453]) and alloys (Au–Pt[448], Ni-Pt [454]) co-catalysts can be attached to the GCN surface or formedin situ by the chemical/photochemical deposition. Similarly to titania, GCN can becoupled with Au NPs to form plasmonic photocatalysts for the H2 production [91,92]. GCN can interact with MoS2 NSs [442, 455, 456], RGO [457], and multiwall

Fig. 2.27 TEM image (a) and EDX element mapping (b) of a GCN hollow sphere decorated withMoS2 nanosheets. c TEM image of a CdS-decorated GCN hollow sphere. Reprinted withpermissions from Refs. [442] (a) and [182] (b). Copyright (2015–2016) Elsevier

2.6 Emerging Semiconductor Photocatalysts … 81

carbon NTs [458–460] via the pp interactions forming visible-light-sensitive lay-ered “dyads” for the hydrogen production. The assortment of sacrificial donors thatcan be applied in the GCN-based systems for H2 evolution is also quite broad [436]including traditional TEA [449–451, 453], carbon acids [448], hydrazine [454], andNa2S/Na2SO3 [180].

Similarly to other wide-bandgap semiconductors that can harvest only a part of thevisible spectrum, GCN can be sensitized to longer wavelengths by organic dyes andnarrower-bandgap semiconductor NPs. The interactions between organic dyes andGCN are especially strong due to the possibility of a multiple bonding between a dyeand GCN via functional (carboxyl) groups present of the edges and outer planes of theGCN particles and via the pp-stacking between the extensive aromatic systems of thedye and outer sheets of the multilayer GCN particles. In this way, visible/NIR-light-sensitive photocatalysts for the H2 evolution were produced by couplingGCN with eosin Y [461, 462], erythrosin B [463], thiazole orange [464], and Znphthalocyanines (an apparent QY of around 3% was observed at 730 nm) [62].

As GCN has a band structure similar to that of cadmium sulfide it can beintroduced into various binary heterostructures with inorganic semiconductors,where offsets between the CB and VB edges of the components favor to an efficientspatial separation of the photogenerated charge carriers [436]. For example, thephotocatalytic activity of TiO2 NRs increases by an order of magnitude after thedecoration with GCN NSs [179].

A large variety of oxide/metallate semiconductor photocatalysts can be producedby a solvothermal treatment in supercritical conditions. Introduction of GCN intothe reaction mixtures offers a surface for the nucleation/deposition of oxide semi-conductors and typically results in a decreased crystal size as compared to theconventional solvothermal synthesis. In this way, the heterostructures of GCN withC,-N-doped TiO2 [183], CuFe2O4 [449], CdS [180, 465], ZnIn2S4 [181] wereproduced with an enhanced photocatalytic activity in the H2 evolution as comparedto both individual GCN and the inorganic component.

A prolonged thermal treatment of GCN at 550–650 °C can also generatenumerous lattice defects (“pinholes”) as a result of the residual ammonia elimina-tion and splitting of the intra-layer bonds. The distortion of single layers results alsoin the expansion of the GCN and a partial exfoliation to NSs [466]. The formationof point defects, new surface edges and reduction in the GCN particles thicknesstypically result in an increase of the photocatalytic activity of this material in thewater reduction [466].

The ultrasound-assisted exfoliation of bulk GCN in 2-propanol produces 2-nmthick NSs revealing an enhanced photocatalytic activity in the solar hydrogenproduction as compared to the compact g-C3N4 [467]. The exfoliation of GCN canbe facilitated by the preliminary intercalation with sulfuric acid [468] in this caseleading to the predominant (around 60 mass%) formation of SLCN. The SLCNshowed a 3-times higher photoactivity than the non-treated GCN.

The GCN nanoribbons around 2 lm in length, *200 nm in width and 3 nmthick were produced by “chemical scissors”, that is by the treatment with a mixtureof concentrated HNO3/H2SO4 [469]. The molar C/N ratio decreases to *0.63

82 2 Semiconductor-Based Photocatalytic Systems …

versus 0.76 for the stoichiometric C3N4 indicating the formation of carbonvacancies that can act as charge traps retarding the electron-hole recombination. Asa result, the nanoribboned GCN showed a 20 times higher photocatalytic activity inthe H2 evolution as compared to the starting bulk material [469].

GCN can be effectively disintegrated by adding water to the bulk GCN mixedwith the concentrated H2SO4 due to a strong exothermic effect (Fig. 2.28a) [450].Following ultrasound-assisted exfoliation produces 2–3-nm thick NSs exhibiting amuch higher photocatalytic activity in the H2 evolution from aqueous TEA solu-tions (Fig. 2.28b) [450].

Atomically thin carbon nitride nanomeshes can be produced by the solvothermalexfoliation of mesoporous GCN intercalated with 2-propanol [451]. Along with ahighly developed surface area, the holey SLCN revealed a higher bandgap of 2.75 eV(as compared to 2.59 eV for the starting material) and a CB level by *0.5 eV morenegative than that of mesoporous GCN. Due to these favorable factors, the SLCNnanomeshes showed an apparent QY of the H2 production of 5.1% at k > 420 nm,which is the highest reported for the exfoliated carbon nitrides [451].

Alternatively to the conventional polyheptazine-based GCN, formed by poly-heptazine networks, a special attention is currently brought to polytriazine networksthat can be produced by versatile synthetic approaches and additionally doped toobtain visible-light-harvesting photocatalysts of the hydrogen production [470–474].

Metal-organic frameworks (MOFs). The metal-organic frameworks can also berated as an “emerging star” of the semiconductor photocatalysis with a number ofreports on various redox-processes catalyzed by the photoexcited MOFs increasingdrastically in recent years [475, 476]. MOFs are formed by metal-organic complexunits linked by “bridge” bi- or tri-functional ligands into the 2D/3D continuousnetworks. Typical bridge ligands for the construction of the 2D and 3D frameworksare aromatic bi- and tri-carboxylic acids. MOFs combine a high light sensitivitywith an almost unlimited versatility of possible building blocks and substituents thatcan alter/modify the MOF structure in a desirable manner. High-intensityligand-to-metal (metal-to-ligand) electron transitions impart them with a strong

Fig. 2.28 a Schematic illustration of the GCN disintegration and exfoliation with H2O/H2SO4;b kinetic curves of the H2 evolution over the bulk and nanosheet GCN. Reprinted and adapted withpermissions from Ref. [450]. Copyright (2015) The Royal Society of Chemistry

2.6 Emerging Semiconductor Photocatalysts … 83

light harvesting capability, while the 2D/3D networks are favorable for the directedcharge transport and form a system of regular pores that can be accessed by thewater and sacrificial donors. The current state-of-the-art in the photocatalyticapplications of MOFs is highlighted by comprehensive reviews [475, 476].

Similarly to the photocatalysis with inorganic semiconductors, no definitemodels were proposed allowing to predict a photocatalytic activity for a given MOFor a class of MOFs and therefore the quest for new photoactive MOFs is performedmostly empirically. As with photoactive inorganic semiconductors, there is a “club”of selected MOFs exhibiting a high photocatalytic activity and by this reasonappearing most frequently in the researchers’ spotlight.

One of such MOFs is UiO-66 formed by ZrO6 octahedra linked with p-dibenzoicacid and its derivatives (Fig. 2.29a) [477–480]. The UiO-66-type MOFs combineregular 3D porous structure, a high stability and sensitivity to the visible light. Theintroduction of functionalities into the bridge molecules opens a way of affecting thepore characteristics.

For example, 3 nm Pt NPs can be introduced into the inner voids ofUiO-66-NH2 MOF acting as a co-catalyst of the photocatalytic water reduction[480]. Such MOF/Pt heterostructure reveals a much higher photoactivity as com-pared to an analog with Pt NPs attached to the outer surface of the MOF grains.A spectral response of the UiO-66 MOF can be extended by the sensitization withdyes [483], CdS NPs [481, 482] and CdxZn1−xS NPs [484]. After modification withRGO, a CdS/UiO-66 assembly becomes a much more active photocatalyst of the H2

production from aqueous Na2S/Na2SO3 solutions as compared with the “classical”TiO2/CdS heterostructure (Fig. 2.29a).

Similar 3D structures can be assembled using oxo-zirconium units with atetracarboxylate-derived zinc porphyrine ZnTCPP (Fig. 2.30) [482]. The MOFcontains a relatively large inner channel where a hydrogenase-biomimeticmetal-organic co-catalyst can be placed. The assembly shows a photocatalyticactivity in the solar H2 production from the aqueous ascorbic acid solutions [482].

Fig. 2.29 a Schematic structure of UiO-66 MOF; b rate of the H2 evolution using variousphotocatalysts. Reprinted with permissions from Refs. [483] (a) and [479] (b). Copyright (2014–2015) The Royal Society of Chemistry

84 2 Semiconductor-Based Photocatalytic Systems …

A key step in the photoprocess is a charge transfer from the singlet excited state ofZn porphyrinate bridges to the incorporated co-catalyst, where the subsequentproton reduction occurs.

A popular family of photoactive MOFs includes a number of MIL-100 com-plexes formed by various central ions and tricarboxylic acids (like 1,3,5-benzenetricarbioxylic acid) [484, 485].

A Fe3+-based MIL-100 has a broad spectral response in the visible range and canbe used as a photocatalyst of the H2 evolution from water/CH3OH mixturesenhanced by the additional deposition of Pt NPs [484]. A La3+-based 3D MOF is awide-bandgap compound with Eg = 3.7 eV that can be sensitized to the visible lightby CdSe NPs [485]. The MOFs with a high sensitivity to the visible light andphotocatalytic activity in the H2 evolution can be produced by using derivatives ofazo-dyes [486] and rhodamines [487] as bridge ligands.

Another face of the application of MOFs in the photocatalysis is in using them asprecursors for the preparation of highly dispersed visible-light-sensitive H2-evol-ving materials, for example, C,H-doped iron oxides [488].

2.7 New-Generation Co-Catalysts for the PhotocatalyticHydrogen Production

The most reported semiconductor-based photocatalytic systems for the hydrogenproduction contain obligatorily a co-catalyst, typically, a noble metal (Pt, Pd, Rh)that is characterized by a much lower water reduction over-voltage as comparedwith the semiconductor photocatalysts. The co-catalyst accepts the photogeneratedCB electrons from the photocatalyst and then participates in the water/protonsreduction, atomic hydrogen accumulation and recombination to the molecularhydrogen. In the photoelectrocatalytic systems where the photoexcitation of a

Fig. 2.30 a Structural building units of Zr6O8(CO2)8(H2O)8, and b structural building unit ofZnTCPP. c Model MOF structure. d Structure of the co-catalyst. e Model structure of the MOF/co-catalyst assembly. Color scheme: Zr, green; Zn, dark gray; C, light gray; O, red; N, blue; Fe,light green; S, yellow. Reprinted with permissions from Ref. [484]. Copyright (2014) The RoyalSociety of Chemistry

2.6 Emerging Semiconductor Photocatalysts … 85

semiconductor anode results in the oxidation of water with simultaneous hydrogenevolution on a counter electrode the photoanode is typically coupled with appro-priate oxygen evolution catalysts [489–491].

Recent studies showed that the “conventional” noble metal co-catalysts can besuccessfully substituted with much less expensive and more available compounds,in particular, iron group metal NPs [492], molybdenum sulfide [426, 455, 456, 493–504], oxide [505], and nitride [506], nickel oxide [507, 508] and hydroxide [408,509], nickel sulfide [510, 511], nitride [512] and phosphide [513], tungsten carbide[514], as well as with cobalt oxide [515, 516], phosphate [176], and phosphide[517]. For example, in the CdS-based systems, the introduction of mere 0.2 wt.%MoS2 results in a rather drastic 36-fold increase of the rate of photocatalytichydrogen evolution from aqueous lactic acid solutions [493]. The catalytic activityof MoS2 in the water reduction is typically associated with an efficient chargecarriers separation between CdS and MoS2, as well as with a well-known capabilityof molybdenum disulfide to the hydrogen activation.

The catalytic activity of MoS2 can be boosted by the exfoliation of bulk layeredmaterial into single or a-few-layer NSs [502, 518]. As the layers of bulkMoS2 are kepttogether by relatively weak van-der-Waals forces the exfoliation can be achieved by aconventional ultrasound treatment. The sonication of bulk MoS2 in the presence ofCdS NRs results in a composite photocatalyst of the H2 evolution from aqueous lacticacid solutions that is far more active than bare CdS NRs or a mechanical NR mixturewith unexfoliated molybdenum disulfide (Fig. 2.31a) [502].

Apart from the most stable semiconducting 2H-phase of MoS2, it can form anallotropic 1T modification that is characterized by the metallic conductance. Theexfoliation of 1T-MoS2 was found to produce an even more efficient co-catalyst ofthe hydrogen evolution for the CdS NR photocatalyst, than conventional exfoliated2H-MoS2 [503] (Fig. 2.31b).

A similar catalytic activity in the H2 evolution in the presence of nanoparticulateCdS was observed for ultrathin composite cobalt selenide/reduced graphene oxideNSs that revealed a half-metallic character [519].

The in situ photodeposition of Ni, Co, or Cu NPs on the surface of Cd0.4Zn0.6Sresulted in a 5-fold acceleration of photocatalytic hydrogen evolution from aqueousNa2S/Na2SO3 solutions [492]. The modification of nanocrystalline Cd0.2Zn0.8Swith 3 wt.% CuS increases strongly the photocatalytic water reduction QY upto *37% at k = 420 nm [520]. The catalytic properties of copper sulfide were alsoobserved in a photocatalytic hydrogen production system based on a visible-light-sensitive CuO/Al2O3 composite [521]. Nickel sulfide can act as a co-catalyst of thephotocatalytic water reduction on the nanocrystalline CdS [522, 523], Cd0.5Zn0.5S[524], and a ZnS1−x−0.5yOx(OH)y/ZnO heterostructure [525]. Cobalt sulfiderevealed a catalytic activity in the photocatalytic water reduction on GCN [526].

Nickel phosphide Ni2P NPs were proven to act as a “universal” co-catalyst of thehydrogen evolution for a broad range of semiconductor photocatalysts includingTiO2, CdS and GCN (Fig. 2.31c) and different sacrificial agents, such as lactic acid,TEA, and methanol [527].

86 2 Semiconductor-Based Photocatalytic Systems …

Various biological molecules, for example, hydrogenase [63, 528] as well as ironcomplexes mimicking an active center of the hydrogenase [482, 529], have alsogood perspectives as co-catalysts in the photocatalytic systems for the H2 genera-tion. A recent account on the hybrid artificial photosynthesis systems based on thesemiconductor light harvesters and biomimetic metal-complex co-catalysts can befound in [530].

The reduced graphene oxide is often used as a hydrogen evolution co-catalyst,however, it can play other unique roles, such, for example, as a conductive 2D“mat” for assembling of various components of a photocatalytic system [531]. GOis typically produced by the ultrasound-assisted exfoliation of layered graphiteoxide which, in turn, can be obtained by treating graphite with strong oxidizingagents such as KMnO4/H2SO4 or KClO3/HNO3. The single (or a-few-layer) GOderived by the exfoliation can then be reduced by a variety of agents, like NaBH4

and hydrazine, or via a photochemical/electrochemical/microwave treatment [531].

Fig. 2.31 a, b Rate of the photocatalytic hydrogen production in the presence of CdS NRs, amechanical mixture of bulk 2H-MoS2 (BM) and CdS NRs and a composite of CdS NRs withultrathin 2H-MoS2 NSs (UM) and 1T-MoS2 NSs produced by the sonication. c Rate of thephotocatalytic H2 production from aqueous solutions of sacrificial donors with the participation ofnanocrystalline TiO2, CdS, GCN and their composites with Ni2P NPs. Reprinted with permissionfrom Refs. [502] (a), [503] (b), and [527] (c). Copyright (2016) The Royal Society of Chemistry

2.7 New-Generation Co-Catalysts for the Photocatalytic … 87

RGO bears residual functional groups such as –COH and –COOH that caninteract with semiconductor NPs and metals bringing then together on the surface ofRGO NSs. At the same time, the RGO sheets typically possess a good conductivityclose to that of pristine graphite. Thus RGO enables a good electric contact amongthe components of a photocatalytic system that are assembled but spatially sepa-rated on the 2D RGO mat. The intermediatory role of RGO NSs ensuring theelectric contact between spatially separated semiconductor and metal NPs wasobserved for TiO2/RGO/Ag [532] and TiSi2/RGO/RuO2 [422]. Thiolated RGO NSscan strongly interact both with CdS NPs and dendritic Pt nanocrystals assemblingthem into a photocatalytic system for the hydrogen production from aqueous lacticacid solutions [533].

RGO can also substitute noble metal co-catalysts revealing in some cases anappealing catalytic activity for the hydrogen generation. For example, the rate ofhydrogen evolution in binary TiO2 NS/RGO NS heterostructure is more than 40times higher than for the individual titania [534]. A catalytic effect of RGO was alsoobserved in the photocatalytic H2 evolution systems based on nanocrystalline CdS[500, 535, 536], Zn-doped CIS NPs [500], ZnIn2S4 [292], Cu2O [537], CaIn2O4

[538], BiPO4 [539], Bi2WO6 [540], K4Nb6O17 [541], and GCN [457].

2.8 Stoichiometric Water Splitting Under the Illuminationwith the Visible Light

One of the conditions for the successful total water splitting consists in the spatialseparation of the water reduction and water oxidation processes necessary to pre-vent reverse reactions between hydrogen and oxygen [2]. This challenge is typicallyaddressed by using two separate electrode cells connected with a membrane. In thefirst cell the water reduction proceeds at the expense of a mediator oxidation, thenthe oxidized mediator diffuses through the membrane into the second cell where itparticipates in the water oxidation to O2 (Fig. 2.32).

Fig. 2.32 Scheme of aphotocatalytic system for totalwater splitting based ontantalum oxynitride, tungstenoxide and I�

�IO�

3 mediatorcouple

88 2 Semiconductor-Based Photocatalytic Systems …

Such coupled systems are often compared to the Z-scheme of the natural pho-tosynthesis [2, 137, 187–189, 542, 543]. The scheme is very convenient as it allowsto design separately the cathode and anode cells and then combine them by anappropriate redox-pair, for example, Fe3+/Fe2+ or I�

�IO�

3 . The photocatalytichydrogen evolution can occur on the surface of SrTiO3/Pt doped with Cr3+/Ta5+ atthe expense of the iodide ions oxidation to IO�

3 . The oxygen is evolved in acomplementary cell with the regeneration of I− on the surface of a WO3/Ptheterostructure. Such system showed an apparent QY of around 0.1% at 420 nm[544, 545]. In the same manner, a hydrogen evolution cell based on avisible-light-sensitive GaInP2/Pt composite was coupled with an oxygen evolutioncell based on AgCl sensitized with silver bromide [546].

A number of cathode/anode cells connected by the iodide/iodate mediator wasproposed, for example TiO2(anatase)/Pt (H2 evolution)—TiO2(rutile)/Pt (O2 evo-lution) [547], TaON/Pt (H2 evolution)—WO3/Pt (O2 evolution) [548], ATaO2N/Pt,A = Ca, Sr, Ba (H2)—WO3/Pt (O2) [549], TaON/Pt (H2)—TaON-RuO2 (O2) [550],TaON/ZrO2/Pt,Ru (H2)—WO3/Pt (O2) [241], sensitized layered H4Nb6O17/Pt (H2)—WO3/Pt (O2) [551], BiVO4 (H2)—Rh-doped SrTiO3/Ru (O2) [198, 552],BaZrO3/BaTaO2N (H2)—WO3/TiOx (O2) [553]. In the most part of such systems,the selective water oxidation takes place as a result of efficient IO�

3 adsorption onthe semiconductor surface despite the presence of a large excess of I− [547].

In a cell, where the photocatalytic water reduction occurs on the surface ofrhodium-doped SrTiO3/Pt and the water oxidation proceeds with the participationof BiVO4, Fe

3+/Fe2+ redox couple is used as a mediator [554, 555]. The systemperforms under the Vis-illumination (k < 500 nm) with an apparent QY of 0.3% at440 nm. The Fe3+/Fe2+ couple was also employed as an electron carrier in a totalwater splitting system based on Rh-doped SrTiO3 and WO3 [556].

Alternatively, the Z-scheme can be realized by combining an H2-evolving and anO2-evolving cell by a conducting bridge. For example, the RhCrOx-loaded LaMg1/3Ta2/3O2N crystals, acting as an H2 evolution photocatalyst can be combined withMo-doped BiVO4 crystals as an oxygen-evolving photocatalyst on a shared goldsubstrate (Fig. 2.33a) [557]. Both components are capable of absorbing the visiblelight. Such composite exhibits a 5-times higher photocatalytic activity in the watersplitting than a combination of corresponding suspensions. The Au substrate acts asa transport layer enabling filling of the tantalate holes with the electrons photo-generated in bismuth vanadate [557].

A similar role of an electron mediator in a Z-scheme photocatalyst can be playedby the photoreduced GO (Fig. 2.33b) [558]. As opposite to the RGO, produced by aconventional reduction with hydrazine, the photoreduced RGO showed a muchmore expressed hydrophilic character binding strongly both to an H2-evolvingphotocatalyst (Rh-doped SrTiO3 decorated with Ru NPs) and to an O2-evolvingphotocatalyst (BiVO4).

By using spatially organized semiconductor materials—NRs, NTs, layeredsubstances, etc. a spatial separation of the water reduction and water oxidation sitescan be achieved within a single photocatalyst, and in such a way a short-circuited

2.8 Stoichiometric Water Splitting Under the Illumination … 89

photo-electrochemical cell can be designed. For example, in a system based on acomposite of titania NTs with Pt NPs [559, 560] the water reduction and oxidationhalf-reactions occur on different components—on Pt NPs and TiO2 NTs, respec-tively, allowing to reach a water splitting QY of 0.6% [559]. In the case of aNi/NiOx/In0.9Ni0.1TaO4 heterostructure, the co-catalyst (Ni/NiOx) particles act as ahydrogen evolution cathode, while the surface of the Ni-doped indium tantalate—asan oxygen evolution anode [561]. A QY of the total water splitting in such systemreached 0.66%. A separation of the cathodic and anodic water cleavage processeswas also realized in the case of a composite photo-catalyst produced by the inter-calation of Fe2O3 NPs into the interlayer voids of HTiNb(Ta)O5 [562].

The hydrogen evolution and oxygen evolution processes become naturallyseparated in the case of two-sided nanocrystalline titania films produced by themagnetron sputtering with one side of the film decorated by Pt NPs. The photo-catalytic hydrogen evolution from aqueous H2SO4 solution occurs on the TiO2/Ptside of the film under the Vis-illumination, while on another side the water oxi-dation in the presence of NaOH takes place [221].

By oxidizing titanium foil in the presence of water vapors and NaF thin films ofF-doped titania were prepared, exhibiting a photoelectrochemical activity in thetotal Vis-light-driven water splitting [563]. The photoelectrochemical water split-ting under the Vis-illumination was also observed in the case of nanocrystallineTiO2 films etched in HF solution [564] and was attributed to the formation ofVis-light-absorbing titanium oxyfluoride species on the film surface.

Along with the above-described complex systems where the spatial separation ofcathodic and anodic processes is organized intentionally, alternative semiconductorcompounds capable of simultaneous water oxidation and reduction under theVis-excitation are explored continuously. One of the first “universal” photocatalystsof the kindwas a solid solutions of gallium nitride and zinc oxide (Ga1−xZnx)(N1−xOx)withEg varying in a range of 2.58–2.76 eV. Gallium-zinc oxy-nitride exhibited a high

Fig. 2.33 a Schematic band diagram of the (RhCrOx/LaMg1/3Ta2/3O2N)/Au/BiVO4:Mo photo-catalyst; b scheme of the water splitting in a photocatalytic system based on Rh:SrTiO3/Ru andBiVO4 coupled via RGO sheets. Reprinted with permissions from Refs. [502] (a) and [558] (b).Copyright (2011, 2016) American Chemical Society

90 2 Semiconductor-Based Photocatalytic Systems …

photochemical stability and, in the presence of RuO2, acted as a photocatalyst of thewater splitting to H2 and O2 under the Vis-illumination [565]. A solid solution ofbismuth and yttrium tungstates, BiYWO6 (Eg = 2.71 eV) was also used as a totalwater splitting photocatalyst that showed an apparent QY of 0.17% at 420 nm in thepresence of RuO2 and Pt/Cr2O3 co-catalysts [566]. Porous films of BiVO4 producedby the thermal decomposition of vanadium oxyacetyl acetonate [567] and BiVO4/Cu2O heterostructures [568] showed a photocatalytic activity in the water splittingunder external bias [567]. The photoelectrochemical water splitting was realized onthe surface of nanocrystalline hematite a-Fe2O3 films synthesized by the thermaldecomposition of ferrocene or iron pentacarbonyl [569, 570].

A visible-light-sensitive photoelectrochemical cell for the total water splittingwas tested [571], where a RuII bipyridyl complex served simultaneously as an“antennae” and as a “bridge” connecting to the titania NPs via phosphate groupsand simultaneously—to the IrO2 NPs via COOH groups. The sensitizer photoex-citation results in the electron transfer through the following chain (Fig. 2.34):water molecules (water oxidation to O2) ! IrO2 NPs ! sensitizer ! TiO2

NPs ! Pt cathode ! water molecules (water reduction to H2).Three new types of total water-splitting photocatalysts functioning under the UV

and a portion of visible light were proposed in [572–574]: BiMNbO7 (M = AlIII,GaIII, InIII), InMO4 (M = NbV, TaV) and BiMO4 (M = NbV, TaV) with Eg in arange of 2.4–2.7 eV. The photoactivity of the compounds increases considerably inthe presence of NiO or Pt co-catalysts. The stoichiometric water splitting under theVis-illumination was also observed in the presence of CaTaO2N perovskite [575],complex solid solutions In–Ni–Ta–O–N [576] and Bi–Y–V–O [577], as well asgallium borate Ga4B2O9 [385].

A feasibility of the photocatalytic decomposition of water confined in an innervolume of the single-wall carbon NTs was shown in [578]. Illumination of the NTsresults in the evolution of a gaseous mixture with 80% hydrogen fraction.

Currently, studies are underway aimed at the search and development of newsemiconductor materials capable of acting as photocatalysts of the stoichiometricwater splitting, for example, GCN [579, 580], GCN/TiO2 nanoheterostructures[176], composites of the CNPs with BiVO4 [581], etc.

Concluding the discussion of the large massif of experimental data on thephotocatalytic systems for hydrogen production based on nanocrystalline semi-conductor materials, we outline very generally the principal directions, where thehighest efforts are currently applied and where a future progress could be expected.

New principles of functioning of the photocatalytic systems can open rich andunexpected directions of progress. For example, utilization of the quantum sizeeffects allowed to create much more efficient systems for the solar water splittingbased on conventional semiconductor materials, to design highly efficientnanoheterostructures based on the same materials but in different phase composi-tions and grain size, and to “invoke life” into some semiconductor materials that arepassive in the bulk form but reveal pronounced photocatalytic properties in the

2.8 Stoichiometric Water Splitting Under the Illumination … 91

water reduction/splitting when introduced in a nanocrystalline form. Also, thephenomenon of plasmonic photocatalysis is a vivid example of the new principle oflight energy harvesting making a great impact on the development ofsemiconductor-based photocatalytic light-harvesting systems.

New photocatalysts should be continuously searched for, in particular, amongthe available and Earth-abundant materials. The example of graphitic carbonnitride, that was known since the middle of 19th century but discovered as anexcellent photocatalyst only at the end of 20th century, shows that new solutions forthe challenges of the solar light harvesting can be just before our eyes and wait to bediscovered and realized. Very high expectations are currently associated withinexpensive and abundant materials based on carbon NPs, that can be producedfrom a variety of available natural sources, as well as with ternary and morecomplex metal chalcogenides based on broadly available copper, tin, zinc and otherelements, that can act as excellent harvesters of the visible and near IR solarirradiation.

New co-catalysts are continuously discovered and such materials can enhancedramatically the performance of conventional semiconductor-based photocatalyticsystems as well as to reduce a need for expensive noble and platinum-group metals.

Finally, new sacrificial donors derived from sustainable sources, such as bio-mass, when coupled to the above-discussed benefits of new photocatalysts andco-catalyst can make the photocatalytic water splitting a really competitive andlucrative process and assist to its broad implementation in our everyday life. Thisroad should be paved by simultaneous development of the theoretical backgroundsof the solar-light-induced water splitting and predictive modelling of the mostoptimized designs and constructions of the photochemical reactors and watersplitting solar cells as well as their operational regimes [582].

Fig. 2.34 Scheme of aphotoelectrochemical systemfor the stoichiometric watersplitting based on aheterostructure of TiIV andIrIV oxides and avisible-light-sensitive RuII

complex (S). Details of thesystem—in [571]

92 2 Semiconductor-Based Photocatalytic Systems …

References

1. Calvert JG, Pitts, JN Jr (1966) Photochemistry. Wiley, New-York, London, Sydney2. Maeda K, Domen K (2007) New non-oxide photocatalysts designed for overall water

splitting under visible light. J Phys Chem C 111:7851–7861. doi:10.1021/jp070911w3. van de Krol R, Liang Y, Schoonman J (2008) Solar hydrogen production with

nanostructured metal oxides. J Mater Chem 18:2311–2320. doi:10.1039/B718969A4. Babu VJ, Vempati S, Uyar T, Ramakrishna S (2015) Review of one-dimensional and

two-dimensional nanostructured materials for hydrogen generation. Phys Chem Chem Phys17:2960–2986. doi:10.1039/c4cp04245j

5. Walter MG, Warren EL, McKone JR et al (2010) Solar water splitting cells. Chem Rev110:6446–6473. doi:10.1021/cr1002326

6. Kitano M, Hara M (2010) Heterogeneous photocatalytic cleavage of water. J Mater Chem20:627–641. doi:10.1039/B910180B

7. Maeda K, Domen K (2010) Photocatalytic water splitting: recent progress and futurechallenges. J Phys Chem Lett 1:2655–2661. doi:10.1021/jz1007966

8. Jing D, Guo L, Zhao L et al (2010) Efficient solar hydrogen production by photocatalyticwater splitting: from fundamental study to pilot demonstration. Inter J Hydrogen En35:7087–7097. doi:10.1016/j.ijhydene.2010.01.030

9. Maeda K (2011) Photocatalytic water splitting using semiconductor particles: history andrecent developments. J Photochem Photobiol, C 12:237–268. doi:10.1016/j.jphotochemrev.2011.07.001

10. Abe R (2010) Recent progress on photocatalytic and photoelectrochemical water splittingunder visible light irradiation. J Photochem Photobiol C 11:179–209. doi:10.1016/j.jphotochemrev.2011.02.003

11. Moriya Y, Takata T, Domen K (2013) Recent progress in the development of (oxy)nitridephotocatalysts for water splitting under visible-light irradiation. Coord Chem Rev 257:1957–1969. doi:10.1016/j.ccr.2013.01.021

12. Du P, Eisenberg R (2012) Catalysts made of earth-abundant elements (Co, Ni, Fe) for watersplitting: recent progress and future challenges. En Environ Sci 5:6012–6021. doi:10.1039/C2EE03250C

13. Warren SC, Thimsen E (2012) Plasmonic solar water splitting. En Environ Sci 5:5133–5146.doi:10.1039/C1EE02875H

14. Valdes A, Brillet J, Grätzel M et al (2012) Solar hydrogen production with semiconductormetal oxides: new directions in experiment and theory. Phys Chem Chem Phys 14:49–70.doi:10.1039/C1CP23212F

15. Zhou H, Qu Y, Zeid T, Duan X (2012) Towards highly efficient photocatalysts usingsemiconductor nanoarchitectures. En Environ Sci 5:6732–6743. doi:10.1039/C2EE03447F

16. Wang P, Huang B, Dai Y, Whangbo MH (2012) Plasmonic photocatalysts: harvestingvisible light with noble metal nanoparticles. Phys Chem Chem Phys 14:9813–9825. doi:10.1039/C2CP40823F

17. Horiuchi Y, Toyao T, Takeuchi M et al (2013) Recent advances in visible-light-responsivephotocatalysts for hydrogen production and solar energy conversion—from semiconductingTiO2 to MOF/PCP photocatalysts. Phys Chem Chem Phys 15:13243–13253. doi:10.1039/C3CP51427G

18. Prevot MS, Sivula K (2013) Photoelectrochemical tandem cells for solar water splitting.J Phys Chem C 117:17879–17893. doi:10.1021/jp405291g

19. Zhang K, Guo L (2013) Metal sulphide semiconductors for photocatalytic hydrogenproduction. Catal Sci Technol 3:1672–1690. doi:10.1039/C3CY00018D

20. Osterloh FE (2013) Inorganic nanostructures for photoelectrochemical and photocatalyticwater splitting. Chem Soc Rev 42:2294–2320. doi:10.1039/C2CS35266D

References 93

21. Fresno F, Portela R, Suarez S, Coronado JM (2014) Photocatalytic materials: recentachievements and near future trends. J Mater Chem A 2:2863–2884. doi:10.1039/C3TA13793G

22. Zhang P, Zhang J, Gong J (2014) Tantalum-based semiconductors for solar water splitting.Chem Soc Rev 43:4395–4422. doi:10.1039/C3CS60438A

23. Ran J, Zhang J, Yu J et al (2014) Earth-abundant cocatalysts for semiconductor-basedphotocatalytic water splitting. Chem Soc Rev 43:7787–7812. doi:10.1039/C3CS60425J

24. Huang ZF, Pan L, Zou JJ et al (2014) Nanostructured bismuth vanadate-based materials forsolar-energy-driven water oxidation: a review on recent progress. Nanoscale 6:14044–14063. doi:10.1039/C4NR05245E

25. Ida S, Ishihara T (2014) Recent progress in two-dimensional oxide photocatalysts for watersplitting. J Phys Chem Lett 5:2533–2542. doi:10.1021/jz5010957

26. Hisatomi T, Kubota J, Domen K (2014) Recent advances in semiconductors forphotocatalytic and photoelectrochemical water splitting. Chem Soc Rev 43:7520–7535.doi:10.1039/C3CS60378D

27. Vaneski A, Schneider J, Susha AS, Rogach AL (2014) Colloidal hybrid heterostructuresbased on II–VI semiconductor nanocrystals for photocatalytic hydrogen generation.J Photochem Photobiol C 19:52–61. doi:10.1016/j.jphotochemrev.2013.12.001

28. Zong X, Wang L (2014) Ion-exchangeable semiconductor materials for visible light-inducedphotocatalysis. J Photochem Photobiol C 18:32–49. doi:10.1016/j.jphotochemrev.2013.10.001

29. Chen X, Liu L, Huang F (2015) Black titanium dioxide (TiO2) nanomaterials. Chem SocRev 44:1861–1885. doi:10.1039/C4CS00330F

30. Moniz SJA, Shevlin SA, Martin DJ et al (2015) Visible-light driven heterojunctionphotocatalysts for water splitting—a critical review. En Environ Sci 8:731–759. doi:10.1039/C4EE03271C

31. Wang M, Han K, Zhang S, Sun L (2015) Integration of organometallic complexes withsemiconductors and other nanomaterials for photocatalytic H2 production. Coord Chem Rev287:1–14. doi:10.1016/j.ccr.2014.12.005

32. Gholipour MR, Dinh CT, Beland F, Do TO (2015) Nanocomposite heterojunctions assunlight-driven photocatalysts for hydrogen production from water splitting. Nanoscale7:8187–8208. doi:10.1039/C4NR07224C

33. Tachibana Y, Vayssieres L, Durrant JR (2012) Artificial photosynthesis for solarwater-splitting. Nat Photonics 6:511–518. doi:10.1038/nphoton.2012.175

34. Zhou N, Lopez-Puente V, Wang Q et al (2015) Plasmon-enhanced light harvesting:applications in enhanced photocatalysis, photodynamic therapy and photovoltaics. RSC Adv5:29076–29097. doi:10.1039/C5RA01819F

35. Ghirardi ML, Dubini A, Yu J, Maness PC (2009) Photobiological hydrogen-producingsystems. Chem Soc Rev 38:52–61. doi:10.1039/B718939G

36. Szacilowski K, Macyk W, Drzewecka-Matuszek A et al (2005) Bioinorganic photochem-istry: frontiers and mechanisms. Chem Rev 105:2647–2694. doi:10.1021/cr030707e

37. Allakhverdiev SI, Kreslavski VD, Thavasi V et al (2009) Hydrogen photoproduction by useof photosynthetic organisms and biomimetic systems. Photochem Photobiol Sci 8:148–156.doi:10.1039/B814932A

38. Zhang X, Peng T, Song S (2016) Recent advances in dye-sensitized semiconductor systemsfor photocatalytic hydrogen production. J Mater Chem A 4:2365–2402. doi:10.1039/c5ta08939e

39. Nada AA, Hamed HA, Barakat MH et al (2008) Enhancement of photocatalytic hydrogenproduction rate using photosensitized TiO2/RuO2-MV2+. Inter J Hydrogen En 33:3264–3269. doi:10.1016/j.ijhydene.2008.04.027

40. Jin Z, Zhang X, Lu G, Li S (2006) Improved quantum yield for photocatalytic hydrogengeneration under visible light irradiation over eosin sensitized TiO2–Investigation of differentnoble metal loading. J Mol Catal A 259:275–280. doi:10.1016/j.molcata.2006.06.035

94 2 Semiconductor-Based Photocatalytic Systems …

41. Tiwari A, Mondal I, Pal U (2015) Visible light induced hydrogen production overthiophenothiazine-based dye sensitized TiO2 photocatalyst in neutral water. RSC Adv5:31415–31421. doi:10.1039/C5RA03039K

42. Lee J, Kwak J, Ko KC et al (2012) Phenothiazine-based organic dyes with two anchoringgroups on TiO2 for highly efficient visible light-induced water splitting. Chem Commun48:11431–11433. doi:10.1039/C2CC36501D

43. Tiwari A, Pal U (2015) Effect of donor-donor-p-acceptor architecture of triphenylamine-basedorganic sensitizers over TiO2 photocatalysts for visible-light-driven hydrogen production.Inter J Hydrogen En 40:9069–9079. doi:10.1016/j.ijhydene.2015.05.101

44. Chen S, Li Y, Wang C (2015) Enhancement of visible-light-driven photocatalytic H2

evolution from water over g-C3N4 through combination with perylene diimide aggregates.RSC Adv 5:15880–15885. doi:10.1016/j.apcata.2015.03.026

45. Zhang J, Du P, Schneider J et al (2007) Photogeneration of hydrogen from water using anintegrated system based on TiO2 and platinum(II) diimine dithiolate sensitizers. J Am ChemSoc 129:7726–7727. doi:10.1021/ja071789h

46. Astuti Y, Palomares E, Haque SA, Durrant JR (2005) Triplet state photosensitization ofnanocrystalline metal oxide electrodes by zinc-substituted cytochrome c: application tohydrogen evolution. J Am Chem Soc 127:15120–15126. doi:10.1021/ja0533444

47. Bala S, Mondal I, Goswami A et al (2014) Synthesis, crystal structure and optical propertiesof a naphthylbisimide-Ni complex: a framework on TiO2 for visible light H2 production.Dalton Trans 43:15704–15707. doi:10.1039/C4DT02006E

48. Li Q, Jin Z, Peng Z et al (2007) High-efficient photocatalytic hydrogen evolution on EosinY-sensitized Ti − MCM41 zeolite under visible-light irradiation. J Phys Chem C 111:8237–8241. doi:10.1021/jp068703b

49. Li Q, Lu G (2007) Visible-light driven photocatalytic hydrogen generation on EosinY-sensitized Pt-loaded nanotube Na2Ti2O4(OH)2. J Mol Catal A 266:75–79. doi:10.1016/j.molcata.2006.10.047

50. Li Y, Xie C, Peng S et al (2008) Eosin Y-sensitized nitrogen-doped TiO2 for efficient visiblelight photocatalytic hydrogen evolution. J Mol Catal A 282:117–123. doi:10.1016/j.molcata.2007.12.005

51. Abe R, Sayama K, Arakawa H (2004) Dye-sensitized photocatalysts for efficient hydrogenproduction from aqueous I− solution under visible light irradiation. J Photochem Photobiol A166:115–122. doi:10.1016/j.jphotochem.2004.04.031

52. Puangpetch T, Sommakettarin P, Chavadej S, Sreethawong T (2010) Hydrogen productionfrom water splitting over Eosin Y-sensitized mesoporous-assembled perovskite titanatenanocrystal photocatalysts under visible light irradiation. Inter J Hydrogen En 35:12428–12442. doi:10.1016/j.ijhydene.2010.08.138

53. Zhang N, Shi J, Niu F et al (2015) A cocatalyst-free Eosin Y-sensitized p-type of Co3O4

quantum dot for highly efficient and stable visible-light-driven water reduction and hydrogenproduction. Phys Chem Chem Phys 17:21397–21400. doi:10.1039/c5cp02983j

54. Ikeda S, Abe C, Torimoto T, Ohtani B (2003) Photochemical hydrogen evolution fromaqueous triethanolamine solutions sensitized by binaphthol-modified titanium(IV) oxideunder visible-light irradiation. J Photochem Photobiol A 160:61–67. doi:10.1016/S1010-6030(03)00222-3

55. Fu N, Lu G (2009) Hydrogen evolution over heteropoly blue-sensitized Pt/TiO2 undervisible light irradiation. Catal Lett 127:319–322. doi:10.1007/s10562-008-9681-4

56. Fu N, Lu G (2009) Photo-catalytic H2 evolution over a series of Keggin-structure heteropolyblue sensitized Pt/TiO2 under visible light irradiation. Appl Surf Sci 255:4378–4383. doi:10.1016/j.apsusc.2008.11.056

57. Kuchmiy SY, Korzhak AV, Guba NF et al (1995) Sensitization of cadmium sulfide bycyanine dyes in the photocatalytic production of hydrogen. Theoret Experim Chem 31:370–374. doi:10.1007/BF00531245

References 95

58. Xu J, Li Y, Peng S et al (2013) Eosin Y-sensitized graphitic carbon nitride fabricated byheating urea for visible light photocatalytic hydrogen evolution: the effect of the pyrolysistemperature of urea. Phys Chem Chem Phys 15:7657–7665. doi:10.1039/C3CP44687E

59. Wang Y, Hong J, Zhang W, Xu R (2013) Carbon nitride nanosheets for photocatalytichydrogen evolution: remarkably enhanced activity by dye sensitization. Catal Sci Technol3:1703–1711. doi:10.1039/C3CY20836B

60. Xu J, Li Y, Peng S (2015) Photocatalytic hydrogen evolution over erythrosin B-sensitizedgraphitic carbon nitride with in situ grown molybdenum sulfide cocatalyst. Inter J HydrogenEn 40:353–362. doi:10.1016/j.ijhydene.2014.10.150

61. Takanabe K, Kamata K, Wang X et al (2010) Photocatalytic hydrogen evolution ondye-sensitized mesoporous carbon nitride photocatalyst with magnesium phthalocyanine.Phys Chem Chem Phys 12:13020–13025. doi:10.1039/C0CP00611D

62. Song S, Guo Y, Peng T et al (2016) Effects of the symmetry and carboxyl anchoring groupof zinc phthalocyanine derivatives on g-C3N4 for photosensitized H2 production. RSC Adv6:77366–77374. doi:10.1039/c6ra15890k

63. Reisner E, Fontecilla-Camps JC, Armstrong FA (2009) Catalytic electrochemistry of a[NiFeSe]-hydrogenase on TiO2 and demonstration of its suitability for visible-light drivenH2 production. Chem Commun 550–552. doi:10.1039/B817371K

64. Peng T, Ke D, Cai P et al (2008) Influence of different ruthenium(II) bipyridyl complex onthe photocatalytic H2 evolution over TiO2 nanoparticles with mesostructures. J PowerSources 180:498–505. doi:10.1016/j.jpowsour.2008.02.002

65. Lakshminarasimhan N, Bae E, Choi W (2007) Enhanced photocatalytic production of H2 onmesoporous TiO2 prepared by template-free method: role of interparticle charge transfer.J Phys Chem C 111:15244–15250. doi:10.1021/jp0752724

66. Li J, E Y, Lian L, Ma W (2013) Visible light induced dye-sensitized photocatalytichydrogen production over platinized TiO2 derived from decomposition of platinum complexprecursor. Inter J Hydrogen En 38:10746–10753. doi:10.1016/j.ijhydene.2013.02.121

67. Maeda K, Eguchi M, Youngblood WJ, Mallouk TE (2008) Niobium oxide nanoscrolls asbuilding blocks for dye-sensitized hydrogen production from water under visible lightirradiation. Chem Mater 20:6770–6778. doi:10.1021/cm801807b

68. Maeda K, Eguchi M, Lee SH et al (2009) Photocatalytic hydrogen evolution fromhexaniobate nanoscrolls and calcium niobate nanosheets sensitized by ruthenium(II)bipyridyl complexes. J Phys Chem C 113:7962–7969. doi:10.1021/jp900842e

69. Peng T, Dai K, Yi H et al (2008) Photosensitization of different ruthenium(II) complex dyeson TiO2 for photocatalytic H2 evolution under visible-light. Chem Phys Lett 460:216–219.doi:10.1016/j.cplett.2008.06.001

70. Ueno K, Misawa H (2013) Surface plasmon-enhanced photochemical reactions.J Photochem Photobiol C 15:31–52. doi:10.1016/j.jphotochemrev.2013.04.001

71. Valenti M, Jonsson MP, Biskos G et al (2016) Plasmonic nanoparticle-semiconductorcomposites for efficient solar water splitting. J Mater Chem A 4:17891–17912. doi:10.1039/c6ta06405a

72. Clavero C (2014) Plasmon-induced hot-electron generation at nanoparticle/metal-oxideinterfaces for photovoltaic and photocatalytic devices. Nat Photonics 8:95–103. doi:10.1038/NPHOTON.2013.238

73. Nguyen BH, Nguyen VH (2015) Recent advances in research on plasmonic enhancement ofphotocatalysis. Adv Nat Sci Nanosci Nanotechnol 6:043001. doi:10.1088/2043-6262/6/4/043001

74. Mendez-Medrano MG, Kowalska E, Lehoux A et al (2016) Surface modification of TiO2

with Au nanoclusters for efficient water treatment and hydrogen generation under visiblelight. J Phys Chem C 120:25010–25022. doi:10.1021/acs.jpcc.6b06854

75. Fang J, Cao SW, Wang Z et al (2012) Mesoporous plasmonic Au–TiO2 nanocomposites forefficient visible-light-driven photocatalytic water reduction. Inter J Hydrogen En 37:17853–17861. doi:10.1016/j.ijhydene.2012.09.023

96 2 Semiconductor-Based Photocatalytic Systems …

76. Zhang X, Liu Y, Kang Z (2014) 3D branched ZnO nanowire arrays decorated withplasmonic Au nanoparticles for high-performance photoelectrochemical water splitting.ACS Appl Mater Interfaces 6:4480–4489. doi:10.1021/am500234v

77. DeSario PA, Pietron JJ, DeVantier DE et al (2013) Plasmonic enhancement of visible-lightwater splitting with Au–TiO2 composite aerogels. Nanoscale 5:8073–8083. doi:10.1039/c3nr01429k

78. Ingram DB, Linic S (2011) Water splitting on composite plasmonic-metal/semiconductorphotoelectrodes: evidence for selective plasmon-induced formation of charge carriers nearthe semiconductor surface. J Am Chem Soc 133:5202–5205. doi:10.1021/ja200086g

79. Meng F, Cushing SK, Li J et al (2015) Enhancement of solar hydrogen generation bysynergistic interaction of La2Ti2O7 photocatalyst with plasmonic gold nanoparticles andreduced graphene oxide nanosheets. ACS Catal 5:1949–1955. doi:10.1021/cs5016194

80. Hou W, Cronin SB (2012) A review of surface plasmon resonance-enhanced photocatalysis.Adv Funct Mater 23:1612–1619. doi:10.1002/adfm.201202148

81. Wu B, Liu D, Mubeen S et al (2016) Anisotropic growth of TiO2 onto gold nanorods forplasmon-enhanced hydrogen production from water reduction. J Am Chem Soc 138:1114–1117. doi:10.1021/jacs.5b11341

82. Chen JJ, Wu JCS, Wu PC, Tsai DP (2011) Plasmonic photocatalyst for H2 evolution inphotocatalytic water splitting. J Phys Chem C 115:210–216. doi:10.1021/jp1074048

83. Mubeen S, Hernandez-Sosa G, Moses D et al (2011) Plasmonic photosensitization of a wideband gap semiconductor: converting plasmons to charge carriers. Nano Lett 11:5548–5552.doi:10.1021/nl203457v

84. Priebe JB, Radnik J, Lennox AJJ et al (2015) Solar hydrogen production by plasmonicAu − TiO2 catalysts: impact of synthesis protocol and TiO2 phase on charge transferefficiency and H2 evolution rates. ACS Catal 5:2137–2148. doi:10.1021/cs5018375

85. Zhan Z, An J, Zhang H et al (2014) Three-dimensional plasmonic photoanodes based onAu-Embedded TiO2 structures for enhanced visible-light water splitting. ACS App MaterInterfaces 6:1139–1144. doi:10.1021/am404738a

86. Zhang Z, Zhang L, Hedhili MN et al (2013) Plasmonic gold nanocrystals coupled withphotonic crystal seamlessly on TiO2 nanotube photoelectrodes for efficient visible lightphotoelectrochemical water splitting. Nano Lett 13:14–20. doi:10.1021/nl3029202

87. Chen C, Kuai L, Chen Y et al (2015) Au/Pt co-loaded ultrathin TiO2 nanosheets forphotocatalyzed H2 evolution by the synergistic effect of plasmonic enhancement andco-catalysis. RSC Adv 5:98254–98259. doi:10.1039/c5ra17732d

88. Yu G, Wang X, Cao J et al (2016) Plasmonic Au nanoparticles embedding enhances theactivity and stability of CdS for photocatalytic hydrogen evolution. Chem Commun52:2394–2397. doi:10.1039/c5cc10066f

89. Luo Y, Liu X, Tang X et al (2014) Gold nanoparticles embedded in Ta2O5/Ta3N5 as activevisible-light plasmonic photocatalysts for solar hydrogen evolution. J Mater Chem A2:14927–14939. doi:10.1039/c4ta02991g

90. Wu M, Chen WJ, Shen YH et al (2014) In situ growth of matchlike ZnO/Au plasmonicheterostructure for enhanced photoelectrochemical water splitting. ACS Appl NaterInterfaces 6:15052–15060. doi:10.1021/am503044f

91. Zeng J, Song T, Lv M et al (2016) Plasmonic photocatalyst Au/g-C3N4/NiFe2O4

nanocomposites for enhanced visible-light-driven photocatalytic hydrogen evolution. RSCAdv 6:54964–54975. doi:10.1039/c6ra08356k

92. Patnaik S, Martha S, Madras G, Parida K (2016) The effect of sulfate pre-treatment toimprove the deposition of Au-nanoparticles in a gold-modified sulfated g-C3N4 plasmonicphotocatalyst towards visible light induced water reduction reaction. Phys Chem Chem Phys18:28502–28514. doi:10.1039/c6cp04262g

93. Zhang X, Zhao J, Wang S et al (2014) Shape-dependent localized surface plasmon enhancedphotocatalytic effect of ZnO nanorods decorated with Ag. Inter J Hydrogen En 39:8238–8245. doi:10.1016/j.ijhydene.2014.03.153

References 97

94. Yi J, She X, Song Y et al (2016) A silver on 2D white-C3N4 support photocatalyst formechanistic insights: synergetic utilization of plasmonic effect for solar hydrogen evolution.RSC Adv 6:112420–112428. doi:10.1039/c6ra23964a

95. Kryukov AI, Kuchmiy SY, Korzhak AV et al (1993) Synergism in an organizedphotocatalytic system based on cadmium sulfide and titanium complexes. Theoret ExpChem 29:312–316. doi:10.1007/BF00532100

96. Kryukov AI, Kuchmiy SY, Kulik SV et al (1994) New photocatalysts: dispersedsemiconductor materials with microheterojunctions. Theoret Exp Chem 30:182–186.doi:10.1007/BF00531180

97. Kryukov AI, Kuchmiy SY, Pokhodenko VD (1994) Molecular design in photo-catalysis:physicochemical principles for designing high-efficiency photocatalytic oxidation-reductionsystems. Theoret Exp Chem 30:141–157. doi:10.1007/BF00534653

98. Kryukov A, Kuchmiy SY, Pokhodenko VD (1997) Nanostructural composite photocatalystsbased on polycrystalline cadmium sulfide. Theoret Exp Chem 33:306–321. doi:10.1007/BF02522707

99. Kryukov AI, Kuchmiy SY, Pokhodenko VD (2000) Energetics of electron processes insemiconductor photocatalytic systems. Theoret ExpChem36:69–89. doi:10.1007/BF02529022

100. Fedoseev BI, Savinov EI, Parmon VN (1987) Photocatalytic hydrogen evolution from Na2Ssolution in the presence of highly disperse cadmium and copper sulfides. Kinetics Catal28:1111–1115

101. Ogisu K, Takanabe K, Lu D et al (2009) CdS nanoparticles exhibiting quantum size effect bydispersion on TiO2: photocatalytic H2 evolution and photoelectrochemical measurements.Bull Chem Soc Jpn 82:528–535. doi:10.1246/bcsj.82.528

102. Hirai T, Suzuki K, Komasawa I (2001) Preparation and photocatalytic properties ofcomposite cds nanoparticles-titanium dioxide particles. J Colloid Interface Sci 244:262–265.doi:10.1006/jcis.2001.7982

103. Stroyuk OL, Kryukov AI, Kuchmiy SY, Pokhodenko VD (2005) Quantum size effects in thephotonics of semiconductor nanoparticles. Theoret Exp Chem 41:67–91. doi:10.1007/s11237-005-0025-9

104. Stroyuk OL, Kryukov AI, Kuchmiy SY, Pokhodenko VD (2005) Quantum size effects insemiconductor photocatalysis. Theoret ExpChem41:207–228. doi:10.1007/s11237-005-0042-8

105. Lian Z, Xu P, Wang W et al (2015) C60-decorated CdS/TiO2 mesoporous architectures withenhanced photostability and photocatalytic activity for H2 evolution. ACS Appl MaterInterfaces 7:4533–4540. doi:10.1021/am5088665

106. Ji SM, Jun H, Jang JS et al (2007) Photocatalytic hydrogen production from naturalseawater. J Photochem Photobiol A 189:141–144. doi:10.1016/j.jphotochem.2007.01.011

107. Park H, Choi W, Hoffmann MR (2008) Effects of the preparation method of the ternary CdS/TiO2/Pt hybrid photocatalysts on visible light-induced hydrogen production. J Mater Chem18:2379–2385. doi:10.1039/B718759A

108. Jang JS, Choi SH, Kim HG, Lee JS (2008) Location and state of Pt in platinized CdS/TiO2

photocatalysts for hydrogen production from water under visible light. J Phys Chem C112:17200–17205. doi:10.1021/jp804699c

109. Cui X, Jiang G, Zhu M et al (2013) TiO2/CdS composite hollow spheres with controlledsynthesis of platinum on the internal wall for the efficient hydrogen evolution. Inter JHydrogen En 38:9065–9073. doi:10.1016/j.ijhydene.2013.05.062

110. Zhou H, Pan J, Ding L et al (2014) Biomass-derived hierarchical porous CdS/M/TiO2

(M = Au, Ag, pt, pd) ternary heterojunctions for photocatalytic hydrogen evolution Inter JHydrogen En 39:16293–16301. doi:10.1016/j.ijhydene. 2014.08.032

111. Cui X, Wang Y, Jiang G et al (2014) A photonic crystal-based CdS–Au–WO3

heterostructure for efficient visible-light photocatalytic hydrogen and oxygen evolution.RSC Adv 4:15689–15694. doi:10.1039/C4RA01415D

112. Vaishnav JK, Arbuj SS, Rane SB, Amalnerkar DP (2014) One dimensional CdS/ZnOnanocomposites: an efficient photocatalyst for hydrogen generation. RSC Adv 4:47637–47642. doi:10.1039/C4RA08561B

98 2 Semiconductor-Based Photocatalytic Systems …

113. Yang G, Yan W, Zhang Q et al (2013) One-dimensional CdS/ZnO core/shell nanofibers viasingle-spinneret electrospinning: tunable morphology and efficient photocatalytic hydrogenproduction. Nanoscale 5:12432–12439. doi:10.1039/C3NR03462C

114. Shen S, Guo L (2008) Growth of quantum-confined CdS nanoparticles inside Ti-MCM-41 asa visible light photocatalyst. Mater Res Bull 43:437–446. doi:10.1016/j.materresbull.2007.02.034

115. Liu Z, Shen S, Guo L (2012) Study on photocatalytic performance for hydrogen evolutionover CdS/M-MCM-41 (M = Zr, Ti) composite photocatalysts under visible light illumina-tion. Inter J Hydrogen En 37:816–821. doi:10.1016/j.ijhydene.2011.04.052

116. Zhang YJ, Yan W, Wu YP, Wang ZH (2008) Synthesis of TiO2 nanotubes coupled withCdS nanoparticles and production of hydrogen by photocatalytic water decomposition.Mater Lett 62(2008):3846–3848. doi:10.1016/j.matlet.2008.04.084

117. Xing C, Jing D, Liu M, Guo L (2009) Photocatalytic hydrogen production overNa2Ti2O4(OH)2 nanotube sensitized by CdS nanoparticles. Mater Res Bull 44:442–445.doi:10.1016/j.materresbull.2008.04.016

118. Kim JC, Choi J, Lee YB et al (2006) Enhanced photocatalytic activity in composites of TiO2

nanotubes and CdS nanoparticles. Chem Commun 48(2006):5024–5026. doi:10.1039/B612572G

119. Chen Y, Guo L (2012) Highly efficient visible-light-driven photocatalytic hydrogenproduction from water using Cd0.5Zn0.5S/TNTs (titanate nanotubes) nanocomposites withoutnoble metals. J Mater Chem 22:7507–7514. doi:10.1039/C2JM16797B

120. Qi L, Yu J, Jaroniec M (2011) Preparation and enhanced visible-light photocatalytic H2-production activity of CdS-sensitized Pt/TiO2 nanosheets with exposed (001) facets. PhysChem Chem Phys 13:8915–8923. doi:10.1039/C1CP20079H

121. Gao B, Yuan X, Lu P et al (2015) Enhanced visible-light-driven photocatalytic H2-production activity of CdS-loaded TiO2 microspheres with exposed (001) facets. J PhysChem Sol 87:171–176. doi:10.1016/j.jpcs.2015.08.018

122. Guan G, Kida T, Kusakabe K et al (2005) Photocatalytic activity of CdS nanoparticlesincorporated in titanium silicate molecular sieves of ETS-4 and ETS-10. Appl Catal A295:71–78. doi:10.1016/j.apcata.2005.08.010

123. Shangguan W, Yoshida A (2001) Synthesis and photocatalytic properties ofCdS-intercalated metal oxides. Sol En Mater Sol Cells 69:189–194. doi:10.1016/S0927-0248(01)00020-4

124. Shangguan W, Yoshida A (2002) Photocatalytic hydrogen evolution from water onnanocomposites incorporating cadmium sulfide into the interlayer. J Phys Chem B106:12227–12230. doi:10.1021/jp0212500

125. Dinh CT, Pham NH, Kleitz F, Do TO (2013) Design of water-soluble CdS–titanate–nickelnanocomposites for photocatalytic hydrogen production under sunlight. J Mater Chem A1:13308–13313. doi:10.1039/C3TA12914D

126. Xiao J, Peng T, Ke D et al (2007) Synthesis, characterization of CdS/rectoritenanocomposites and its photocatalytic activity. Phys Chem Miner 34:275–285. doi:10.1007/s00269-007-0146-x

127. Parayil SK, Baltrusaitis J, Wu CM, Koodali RT (2013) Synthesis and characterization ofligand stabilized CdS-Trititanate composite materials for visible light-induced photocatalyticwater splitting. Inter J Hydrogen En 38:2656–2669. doi:10.1016/j.ijhydene.2012.12.042

128. Ryu SY, Choi J, Balcerski W et al (2007) Photocatalytic production of H2 on nanocompositecatalysts. Ind Eng Chem Res 46:7476–7488. doi:10.1021/ie0703033

129. Shangguan W (2007) Hydrogen evolution from water splitting on nanocompositephotocatalysts. Sci Technol Adv Mater 8:76–81. doi:10.1016/j.stam.2006.09.007

130. Choi J, Ryu Y, Balcerski W et al (2008) Photocatalytic production of hydrogen on Ni/NiO/KNbO3/CdS nanocomposites using visible light. J Mater Chem 18:2371–2378. doi:10.1039/B718535A

References 99

131. Chen W, Gao H, Yuan J et al (2013) Structure characteristics of CdS/H1.9K0.3La0.5Bi0.1Ta2O7 and photocatalytic activity for hydrogen evolution under visiblelight. Inter J Hydrogen En 38:10754–10760. doi:10.1016/j.ijhydene.2013.02.067

132. Yu J, Lei SL, Chen TC et al (2014) A new CdS/Bi1−xInxTaO4 heterostructured photocatalystcontaining solid solutions for H2 evolution from water splitting. Inter J Hydrogen En39:13105–13113. doi:10.1016/j.ijhydene.2014.06.148

133. Zhang G, Lin B, Yang W et al (2015) Highly efficient photocatalytic hydrogen generation byincorporating CdS into ZnCr-layered double hydroxide interlayer. RSC Adv 5:5823–5829.doi:10.1039/C4RA11757C

134. Wu J, Lin J, Yin S, Sato T (2001) Synthesis and photocatalytic properties of layeredHNbWO6/(Pt, Cd0.8Zn0.2S) nanocomposites. J Mater Chem 11:3343–3347. doi:10.1039/B103838A

135. So WW, Kim KJ, Moon SJ (2004) Photo-production of hydrogen over the CdS–TiO2

nanocomposite particulate films treated with TiCl4. Inter J Hydrogen En 29:229–234. doi:10.1016/S0360-3199(03)00211-8

136. Wang H, Zhu W, Chong B, Qin K (2014) Improvement of photocatalytic hydrogengeneration from CdSe/CdS/TiO2 nanotube-array coaxial heterogeneous structure. Inter JHydrogen En 39:90–99. doi:10.1016/j.ijhydene.2013.10.048

137. Kudo A, Miseki Y (2009) Heterogeneous photocatalyst materials for water splitting. ChemSoc Rev 38:253–278. doi:10.1039/B800489G

138. Moon GD, Joo JB, Lee I, Yin Y (2014) Decoration of size-tunable CuO nanodots on TiO2

nanocrystals for noble metal-free photocatalytic H2 production. Nanoscale 6:12002–12008.doi:10.1039/C4NR03521F

139. Jang JS, Ji SM, Bae SW et al (2007) Optimization of CdS/TiO2 nano-bulk compositephotocatalysts for hydrogen production from Na2S/Na2SO3 aqueous electrolyte solutionunder visible light (k � 420 nm). J Photochem Photobiol, A 188:112–119. doi:10.1016/j.jphotochem.2006.11.027

140. Jang JS, Kim HG, Borse PH, Lee JS (2007) Simultaneous hydrogen production anddecomposition of H2S dissolved in alkaline water over CdS–TiO2 composite photocatalystsunder visible light irradiation. Inter J Hydrogen En 32:4786–4791. doi:10.1016/j.ijhydene.2007.06.026

141. Jang JS, Kim HG, Joshi UA et al (2008) Fabrication of CdS nanowires decorated with TiO2

nanoparticles for photocatalytic hydrogen production under visible light irradiation. Inter JHydrogen En 33:5975–5980. doi:10.1016/j.ijhydene.2008.07.105

142. Li C, Xi Z, Fang W et al (2015) Enhanced photocatalytic hydrogen evolution activity ofCuInS2 loaded TiO2 under solar light irradiation. J Sol State Chem 226:94–100. doi:10.1016/j.jssc.2015.02.011

143. Guo K, Liu Z, Han J et al (2014) Hierarchical TiO2–CuInS2 core–shell nanoarrays forphotoelectrochemical water splitting. Phys Chem Chem Phys 16:16204–16213. doi:10.1039/C4CP01971G

144. Li K, Xu J, Zhang X et al (2013) Low-temperature preparation of AgIn5S8/TiO2

heterojunction nanocomposite with efficient visible-light-driven hydrogen production.Inter J Hydrogen En 38:15965–15975. doi:10.1016/j.ijhydene.2013.09.147

145. Cheng Z, Zhan X, Wang F et al (2015) Construction of CuInS2/Ag sensitized ZnO nanowirearrays for efficient hydrogen generation. RSC Adv 5:81723–81727. doi:10.1039/c5ra14188e

146. Choi Y, Baek M, Zhang Z et al (2015) A two-storey structured photoanode of a 3DCu2ZnSnS4/CdS/ZnO@steel composite nanostructure for efficient photoelectrochemicalhydrogen generation. Nanoscale 7:15291–15299. doi:10.1039/c5nr04107d

147. Kandiel TA, Takanabe K (2016) Solvent-induced deposition of Cu–Ga–In–S nanocrystalsonto atitanium dioxide surface for visible-light-driven photocatalytic hydrogen production.Appl Catal B 184:264–269. doi:10.1016/j.apcatb.2015.11.036

148. Brahimi R, Bessekhouad Y, Bouguelia A, Trari M (2007) Visible light induced hydrogenevolution over the heterosystem Bi2S3/TiO2. Catal Today 122:62–65. doi:10.1016/j.cattod.2007.01.030

100 2 Semiconductor-Based Photocatalytic Systems …

149. Kim J, Kang M (2012) High photocatalytic hydrogen production over the band gap-tunedurchin-like Bi2S3-loaded TiO2 composites system. Inter J Hydrogen En 37:8249–8256.doi:10.1016/j.ijhydene.2012.02.057

150. Senevirathna N, Pitigaka P, Tennakone K (2005) Water photoreduction with Cu2O quantumdots on TiO2 nanoparticles. J Photochem Photobiol A 171:257–259. doi:10.1016/j.jphotochem.2004.10.018

151. Zhang S, Zhang S, Peng F et al (2011) Electrodeposition of polyhedral Cu2O on TiO2

nanotube arrays for enhancing visible light photocatalytic performance. ElectrochemCommun 13:861–864. doi:10.1016/j.elecom.2011.05.022

152. Kumar DP, Reddy NL, Kumari MM et al (2015) Cu2O-sensitized TiO2 nanorods withnanocavities for highly efficient photocatalytic hydrogen production under solar irradiation.Sol En Mater Sol Cells 136:157–166. doi:10.1016/j.solmat.2015.01.009

153. Zhang S, Peng B, Yang S et al (2013) The influence of the electrodeposition potential on themorphology of Cu2O/TiO2 nanotube arrays and their visible-light-driven photocatalyticactivity for hydrogen evolution. Inter J Hydrogen En 38:13866–13871. doi:10.1016/j.ijhydene.2013.08.081

154. Li L, Xu L, Shi W, Guan J (2013) Facile preparation and size-dependent photocatalyticactivity of Cu2O nanocrystals modified titania for hydrogen evolution. Inter J Hydrogen En38:816–822. doi:10.1016/j.ijhydene.2012.10.064

155. Sharma D, Upadhyay S, Satsangi VR et al (2014) Improved photoelectrochemical watersplitting performance of Cu2O/SrTiO3 heterojunction photoelectrode. J Phys Chem C118:25320–25329. doi:10.1021/jp507039n

156. Jung M, Scott J, Ng YH et al (2014) CuOx dispersion and reducibility on TiO2 and its impacton photocatalytic hydrogen evolution. Inter J Hydrogen En 39:12499–12506. doi:10.1016/j.ijhydene.2014.06.020

157. Bandara J, Udawatta C, Rajapakse C (2005) Highly stable CuO incorporated TiO2 catalystfor photocatalytic hydrogen production from H2O. Photochem Photobiol Sci 4:857–861.doi:10.1039/B507816D

158. Yu Z, Meng J, Li Y et al (2013) Efficient photocatalytic hydrogen production from waterover a CuO and carbon fiber comodified TiO2 nanocomposite photocatalyst. Inter JHydrogen En 38:6649–16655. doi:10.1016/j.ijhydene.2013.07.056

159. Xu S, Du AJ, Liu J et al (2011) Highly efficient CuO incorporated TiO2 nanotubephotocatalyst for hydrogen production from water. Inter J Hydrogen En 36:6560–6568.doi:10.1016/j.ijhydene.2011.02.103

160. Kumar SP, Shankar MV, Kumari MM et al (2013) Nano-size effects on CuO/TiO2 catalystsfor highly efficient H2 production under solar light irradiation. Chem Commun 49:9443–9445. doi:10.1039/C3CC44742A

161. Yue X, Yi S, Wang R et al (2016) A novel and highly efficient earth-abundant Cu3P withTiO2 “P–N” heterojunction nanophotocatalyst for hydrogen evolution from water.Nanoscale 8:17516–17523. doi:10.1039/c6nr06620h

162. Yang X, Xu J, Wong T et al (2013) Synthesis of In2O3–In2S3 core–shell nanorods withinverted type-I structure for photocatalytic H2 generation. Phys Chem Chem Phys 15:12688–12693. doi:10.1039/C3CP51722E

163. Liu Z, Bai H, Xu S, Sun DD (2011) Hierarchical CuO/ZnO “corn-like” architecture forphotocatalytic hydrogen generation. Inter J Hydrogen En 36:13473–13480. doi:10.1016/j.ijhydene.2011.07.137

164. Rajaambal S, Mapa M, Gopinath CS (2014) In1−xGaxN@ZnO: a rationally designed andquantum dot integrated material for water splitting and solar harvesting applications. DaltonTrans 43:12546–12554. doi:10.1039/C4DT01268B

165. Derbal A, Omeiri S, Bouguelia A, Trari M (2008) Characterization of new heterosystemCuFeO2/SnO2 application to visible-light induced hydrogen evolution. Inter J Hydrogen En33:4274–4282. doi:10.1016/j.ijhydene.2008.05.067

References 101

166. Nguyen-Phan TD, Luo S, Vovchok D et al (2016) Visible light-driven H2 production overhighly dispersed ruthenia on rutile TiO2 nanorods. ACS Catal 6:407–417. doi:10.1021/acscatal.5b02318

167. Brahimi R, Bessekhouad Y, Bouguelia A, Trari M (2007) CuAlO2/TiO2 heterojunctionapplied to visible light H2 production. J Photochem Photobiol A 186:242–247. doi:10.1016/j.jphotochem.2006.08.013

168. Ou Y, Lin J, Fang S, Liao D (2006) MWNT–TiO2: Ni composite catalyst: a new class ofcatalyst for photocatalytic H2 evolution from water under visible light illumination. ChemPhys Lett 429:199–203. doi:10.1016/j.cplett.2006.08.024

169. Yu H, Shi R, Zhao Y et al (2016) Smart utilization of carbon dots in semiconductorphotocatalysis. Adv Mater 28:9454–9477. doi:10.1002/adma.201602581

170. Cao S, Yu J (2016) Carbon-based H2-production photocatalytic materials. J PhotochemPhotobiol C 27:72–99. doi:10.1016/j.jphotochemrev.2016.04.002

171. Wang J, Gao M, Ho GW (2014) Bidentate-complex-derived TiO2/carbon dot photocatalysts:in situ synthesis, versatile heterostructures, and enhanced H2 evolution. J Mater Chem A 2(2014):5703–5709. doi:10.1039/C3TA15114J

172. Yu H, Zhao Y, Zhou C et al (2014) Carbon quantum dots/TiO2 composites for efficientphotocatalytic hydrogen evolution. J Mater Chem A 2:3344–3351. doi:10.1039/C3TA14108J

173. Tang Y, Hao R, Fu Y et al (2016) Carbon quantum dot/mixed crystal TiO2 composites via ahydrogenation process: an efficient photocatalyst for the hydrogen evolution reaction. RSCAdv 6:96803–96808. doi:10.1039/c6ra17597j

174. Wang J, Huang J, Xie H, Qu A (2014) Synthesis of g-C3N4/TiO2 with enhancedphotocatalytic activity for H2 evolution by a simple method. Inter J Hydrogen En 39:6354–6363. doi:10.1016/j.ijhydene.2014.02.020

175. Zang Y, Li L, Xu Y et al (2014) Hybridization of brookite TiO2 with g-C3N4: avisible-light-driven photocatalyst for As3+ oxidation, MO degradation and water splitting forhydrogen evolution. J Mater Chem A 2:15774–15780. doi:10.1039/C4TA02082K

176. Li Y, Wang R, Li H et al (2015) Efficient and stable photoelectrochemical seawater splittingwith TiO2@g-C3N4 nanorod arrays decorated by Co-Pi. J Phys Chem C 119:20283–20292.doi:10.1021/acs.jpcc.5b05427

177. Pany S, Parida KM (2015) A facile in situ approach to fabricate N, S-TiO2/g-C3N4

nanocomposite with excellent activity for visible light induced water splitting for hydrogenevolution. Phys Chem Chem Phys 17:8070–8077. doi:10.1039/C4CP05582A

178. Zhong X, Jin M, Dong H et al (2014) TiO2 nanobelts with a uniform coating of g-C3N4 as ahighly effective heterostructure for enhanced photocatalytic activities. J Sol State Chem220:54–59. doi:10.1016/j.jssc.2014.08.016

179. Jiang Y, Guo S, Hao R et al (2016) A hybridized heterojunction structure between TiO2

nanorods and exfoliated graphitic carbon nitride sheets for hydrogen evolution under visiblelight. CrystEngComm 18:6875–6880. doi:10.1039/c6ce01442a

180. Cheng F, Yin H, Xiang Q (2017) Low-temperature solid-state preparation of ternary CdS/g-C3N4/CuS nanocomposites for enhanced visible-light photocatalytic H2-productionactivity. Appl Surf Sci 391:432–439. doi:10.1016/j.apsusc.2016.06.169

181. Liu H, Jin Z, Xu Z et al (2015) Fabrication of ZnIn2S4–g-C3N4 sheet-on-sheetnanocomposites for efficient visible-light photocatalytic H2-evolution and degradation oforganic pollutants. RSC Adv 5:97951–97961. doi:10.1039/c5ra17028a

182. Zheng D, Zhang G, Wang X (2015) Integrating CdS quantum dots on hollow graphiticcarbon nitride nanospheres for hydrogen evolution photocatalysis. Appl Catal B 179:479–488. doi:10.1016/j.apcatb.2015.05.060

183. Chen W, Liu TY, Huang T et al (2015) A novel yet simple strategy to fabricate visible lightresponsive C, N-TiO2/g-C3N4 heterostructures with significantly enhanced photocatalytichydrogen generation. RSC Adv 5:101214–101220. doi:10.1039/c5ra18302b

102 2 Semiconductor-Based Photocatalytic Systems …

184. Hu B, Cai F, Chen T et al (2015) Hydrothermal synthesis g-C3N4/Nano-InVO4

nanocomposites and enhanced photocatalytic activity for hydrogen production under visiblelight irradiation. ACS Appl Mater Interfaces 7:18247–18256. doi:10.1021/acsami.5b05715

185. Kudo A (2007) Recent progress in the development of visible light-driven powderedphotocatalysts for water splitting. Inter J Hydrogen En 32:2673–2678. doi:10.1016/j.ijhydene.2006.09.010

186. Ni M, Leung M, Leung D, Sumathy K (2007) A review and recent developments inphotocatalytic water-splitting using TiO2 for hydrogen production. Renew Sustain En Rev11:401–425. doi:10.1016/j.rser.2005.01.009

187. Matsuoka M, Kitano M, Takeuchi M et al (2007) Photocatalysis for new energy production:recent advances in photocatalytic water splitting reactions for hydrogen production. CatalToday 122:51–61. doi:10.1016/j.cattod.2007.01.042

188. Lee JS (2005) Photocatalytic water splitting under visible light with particulate semicon-ductor catalysts. Catal Surv Asia 9:217–227. doi:10.1007/s10563-005-9157-0

189. Kudo A (2003) Photocatalyst materials for water splitting. Catal Surv Asia 7:31–38. doi:10.1023/A:1023480507710

190. Tsuji I, Kudo A (2003) H2 evolution from aqueous sulfite solutions under visible-lightirradiation over Pb and halogen-codoped ZnS photocatalysts. J Photochem Photobiol A156:249–252. doi:10.1016/S1010-6030(02)00433-1

191. Kudo A, Sekizawa M (2000) Photocatalytic H2 evolution under visible light irradiation onNi-doped ZnS photocatalyst. Chem Commun 1371–1372. doi:10.1039/B003297M

192. Kudo A, Sekizawa M (1999) Photocatalytic H2 evolution under visible light irradiation onZn1−xCuxS solid solution. Catal Lett 58:241–243. doi:10.1023/A:1019067025917

193. Mei Z, Ouyang S, Zhang Y, Kako T (2013) Ultrafine Zn1−xCuxS (0 � x � 0.066)nanocrystallites for photocatalytic H2 evolution under visible light irradiation. RSC Adv3:10654–10657. doi:10.1039/C3RA41076E

194. Wu Y, Lu G, Li S (2009) The doping effect of Bi on TiO2 for photocatalytic hydrogengeneration and photodecolorization of rhodamine B. J Phys Chem C 113:9950–9955. doi:10.1021/jp9009433

195. Ishii T, Kato H, Kudo A (2004) H2 evolution from an aqueous methanol solution on SrTiO3

photocatalysts codoped with chromium and tantalum ions under visible light irradiation.J Photochem Photobiol A 163:181–186. doi:10.1016/S1010-6030(03)00442-8

196. Wang D, Ye J, Kako T, Kimura T (2006) Photophysical and photocatalytic properties ofSrTiO3 doped with Cr cations on different sites. J Phys Chem B 110:15824–15830. doi:10.1021/jp062487p

197. Iwashina K, Kudo A (2011) Rh-doped SrTiO3 photocatalyst electrode showing cathodicphotocurrent for water splitting under visible-light irradiation. J Am Chem Soc 133:13272–13275

198. Kato H, Sasaki Y, Shirakura N, Kudo A (2013) Synthesis of highly active rhodium-dopedSrTiO3 powders in Z-scheme systems for visible-light-driven photocatalytic overall watersplitting. J Mater Chem A 1:12327–12333. doi:10.1039/C3TA12803B

199. Niishiro R, Kato H, Kudo A (2005) Nickel and either tantalum or niobium-codoped TiO2

and SrTiO3 photocatalysts with visible-light response for H2 or O2 evolution from aqueoussolutions. Phys Chem Chem Phys 7:2241–2245. doi:10.1039/B502147B

200. Kato H, Kudo A (2002) Visible-light-response and photocatalytic activities of TiO2 andSrTiO3 photocatalysts codoped with antimony and chromium. J Phys Chem B 106:5029–5034. doi:10.1021/jp0255482

201. Hwang DW, Kim HG, Lee JS et al (2005) Photocatalytic hydrogen production from waterover M-doped La2Ti2O7 (M = Cr, Fe) under visible light irradiation (k > 420 nm). J PhysChem B 109:2093–2102. doi:10.1021/jp0493226

202. Hwang DW, Kim HG, Jang JS et al (2004) Photocatalytic decomposition of water–methanolsolution over metal-doped layered perovskites under visible light irradiation. Catal Today93–95:845–850. doi:10.1016/j.cattod.2004.06.084

References 103

203. Kanhere P, Zheng J, Chen Z (2012) Visible light driven photocatalytic hydrogen evolutionand photophysical properties of Bi3+ doped NaTaO3. Inter J Hydrogen En 37:4889–4896.doi:10.1016/j.ijhydene.2011.12.056

204. Zou JP, Zhang LZ, Luo SL et al (2012) Preparation and photocatalytic activities of two newZn-doped SrTiO3 and BaTiO3 photocatalysts for hydrogen production from water withoutcocatalysts loading. Inter J Hydrogen En 37:17068–17077. doi:10.1016/j.ijhydene.2012.08.133

205. Xue Y, Wang X (2015) The effects of Ag doping on crystalline structure and photocatalyticproperties of BiVO4. Inter J Hydrogen En 40:5878–5888. doi:10.1016/j.ijhydene.2015.03.028

206. Jana P, Montero CM, Pizzarro P et al (2014) Photocatalytic hydrogen production in thewater/methanol system using Pt/RE:NaTaO3 (RE = Y, La, Ce, Yb) catalysts. Inter JHydrogen En 39:5283–5290. doi:10.1016/j.ijhydene.2013.12.182

207. Shimura K, Yoshida H (2012) Effect of doped zinc species on the photocatalytic activity ofgallium oxide for hydrogen production. Phys Chem Chem Phys 14:2678–2684. doi:10.1039/C2CP23220K

208. Kudo A, Niishiro R, Iwase A, Kato H (2007) Effects of doping of metal cations onmorphology, activity, and visible light response of photocatalysts. Chem Phys 339:104–110.doi:10.1016/j.chemphys.2007.07.024

209. Zou Z, Arakawa H (2003) Direct water splitting into H2 and O2 under visible light irradiationwith a new series of mixed oxide semiconductor photocatalysts. J Photochem Photobiol A158:145–162. doi:10.1016/S1010-6030(03)00029-7

210. Zou Z, Ye J, Sayama K, Arakawa H (2002) Photocatalytic hydrogen and oxygen formationunder visible light irradiation with M-doped InTaO4 (M = M n, Fe Co, Ni and Cu)photocatalysts. J Photochem Photobiol A 148:65–69. doi:10.1016/S1010-6030(02)00068-0

211. Wang D, Zou Z, Ye J (2005) Photocatalytic water splitting with the Cr-doped Ba2In2O5/In2O3 composite oxide semiconductors. Chem Mater 17:3255–3261. doi:10.1021/cm0477117

212. Gurunathan K (2004) Photocatalytic hydrogen production using transition metal ions-dopedc-Bi2O3 semiconductor particles. Inter J Hydrogen En 29:933–940. doi:10.1016/j.ijhydene.2003.04.001

213. Jing D, Zhang Y, Guo L (2005) Study on the synthesis of Ni doped mesoporous TiO2 and itsphotocatalytic activity for hydrogen evolution in aqueous methanol solution. Chem PhysLett 415:74–78. doi:10.1016/j.cplett.2005.08.080

214. Liu Q, Ding D, Ning C, Wang X (2015) Black Ni-doped TiO2 photoanodes forhigh-efficiency photoelectrochemical water-splitting. Inter J Hydrogen En 40:2107–2114.doi:10.1016/j.ijhydene.2014.12.064

215. Wang D, Ye J, Kitazawa H, Kimura T (2007) Photophysical and photocatalytic properties ofthree isostructural oxide semiconductors In6NiTi6O22, In3CrTi2O10, and In12NiCr2Ti10O42

with different 3d transition metals. J Phys Chem C 111:12848–12854. doi:10.1021/jp0678599

216. Wang D, Zou Z, Ye J (2005) Photocatalytic H2 evolution over a new visible-light-drivenphotocatalyst In12NiCr2Ti10O42. Chem Phys Lett 411:285–290. doi:10.1016/j.cplett.2005.05.124

217. Konta R, Ishii T, Kato H, Kudo A (2004) Photocatalytic activities of noble metal ion dopedSrTiO3 under visible light irradiation. J Phys Chem B 108:8992–8995. doi:10.1021/jp049556p

218. Kitano M, Takeuchi M, Matsuoka M et al (2007) Photocatalytic water splitting usingPt-loaded visible light-responsive TiO2 thin film photocatalysts. Catal Today 120:133–138.doi:10.1016/j.cattod.2006.07.043

219. Fukumoto S, Kitano M, Takeuchi M et al (2009) Photocatalytic hydrogen production fromaqueous solutions of alcohol as model compounds of biomass using visible light-responsiveTiO2 thin films. Catal Lett 127:39–43. doi:10.1007/s10562-008-9769-x

104 2 Semiconductor-Based Photocatalytic Systems …

220. Kitano M, Tsujimaru K, Anpo M (2008) Hydrogen production using highly active titaniumoxide-based photocatalysts. Topics Catal 49:4–17. doi:10.1007/s11244-008-9059-2

221. Selli E, Chiarello GL, Quartarone E et al (2007) A photocatalytic water splitting device forseparate hydrogen and oxygen evolution. Chem Commun 5022–5024. doi:10.1039/B711747G

222. Matsuoka M, Kitano M, Fukumoto S et al (2008) The effect of the hydrothermal treatmentwith aqueous NaOH solution on the photocatalytic and photoelectrochemical properties ofvisible light-responsive TiO2 thin films. Catal Today 132:159–164. doi:10.1016/j.cattod.2007.12.032

223. Kitano M, Takeuchi M, Matsuoka M et al (2005) Preparation of visible light-responsiveTiO2 thin film photocatalysts by an RF magnetron sputtering deposition method and theirphotocatalytic reactivity. Chem Lett 34:616–617. doi:10.1246/cl.2005.616

224. Dholam R, Patel N, Adami M, Miotello A (2008) Physically and chemically synthesizedTiO2 composite thin films for hydrogen production by photocatalytic water splitting. Inter JHydrogen En 33:6896–6903. doi:10.1016/j.ijhydene.2008.08.061

225. Matsuoka M, Kitano M, Takeuchi M et al (2005) Photocatalytic water splitting on visiblelight-responsive TiO2 thin films prepared by a RF magnetron sputtering deposition method.Topics Catal 35:305–310. doi:10.1007/s11244-005-3838-9

226. Wang Z, Yang C, Lin T et al (2013) Visible-light photocatalytic, solar thermal andphotoelectrochemical properties of aluminium-reduced black titania. En Environ Sci6:3007–3014. doi:10.1039/c3ee41817k

227. Cui H, Zhao W, Yang C et al (2014) Black TiO2 nanotube arrays for high-efficiencyphotoelectrochemical water-splitting. J Mater Chem A 2:8612–8616. doi:10.1039/c4ta00176a

228. Zhang K, Zhou W, Zhang X et al (2016) Large-scale synthesis of stable mesoporous blackTiO2 nanosheets for efficient solar-driven photocatalytic hydrogen evolution via anearthabundant low-cost biotemplate. RSC Adv 6:50506–50512. doi:10.1039/c6ra06751d

229. Yuan J, Chen M, Shi J, Shanguang W (2006) Preparations and photocatalytic hydrogenevolution of N-doped TiO2 from urea and titanium tetrachloride. Inter J Hydrogen En31:1326–1331. doi:10.1016/j.ijhydene.2005.11.016

230. Sreethawong T, Laehsalee S, Chavadej S (2009) Use of Pt/N-doped mesoporous-assemblednanocrystalline TiO2 for photocatalytic H2 production under visible light irradiation. CatalCommun 10:538–543. doi:10.1016/j.catcom.2008.10.029

231. Sreethawong T, Laehsalee S, Chavadej S (2008) Comparative investigation of mesoporous-and non-mesoporous-assembled TiO2 nanocrystals for photocatalytic H2 production overN-doped TiO2 under visible light irradiation. Inter J Hydrogen En 33:5947–5957. doi:10.1016/j.ijhydene.2008.08.007

232. Liu SH, Syu HR (2013) High visible-light photocatalytic hydrogen evolution of C,N-codoped mesoporous TiO2 nanoparticles prepared via an ionic-liquid-template approach.Inter J Hydrogen En 38:13856–13865. doi:10.1016/j.ijhydene.2013.08.094

233. Kim H, Monllor-Satoca D, Kim W, Choi W (2015) N-doped TiO2 nanotubes coated with athin TaOxNy layer for photoelectrochemical water splitting: dual bulk and surfacemodification of photoanodes. En Environ Sci 8:247–257. doi:10.1039/C4EE02169J

234. Lin WC, Yang WD, Huang IL et al (2009) Hydrogen production from methanol/waterphotocatalytic decomposition using Pt/TiO2−xNx catalyst. En Fuels 23:2192–2196

235. Pei F, Liu Y, Xu S et al (2013) Nanocomposite of graphene oxide with nitrogen-doped TiO2

exhibiting enhanced photocatalytic efficiency for hydrogen evolution. Inter J Hydrogen En38:2670–2677. doi:10.1016/j.ijhydene.2012.12.045

236. Pei F, Xu S, Zuo W et al (2014) Effective improvement of photocatalytic hydrogen evolutionvia a facile in-situ solvothermal N-doping strategy in N-TiO2/N-graphene nanocomposite.Inter J Hydrogen En 39:6845–6852. doi:10.1016/j.ijhydene.2014.02.173

237. Hara M, Nunoshige J, Takata T et al (2003) Unusual enhancement of H2 evolution by Ru onTaON photocatalyst under visible light irradiation. Chem Commun 3000–3001. doi:10.1039/B309935K

References 105

238. Hara M, Hitoki G, Takata T et al (2003) TaON and Ta3N5 as new visible light drivenphotocatalysts. Catal Today 78:555–560. doi:10.1016/S0920-5861(02)00354-1

239. Suzuki TM, Nakamura T, Saeki S et al (2012) Visible light-sensitive mesoporous N-dopedTa2O5 spheres: synthesis and photocatalytic activity for hydrogen evolution and CO2

reduction. J Mater Chem 22:24584–24590. doi:10.1039/C2JM33980C240. Mishima T, Matsuda M, Miyake M (2007) Visible-light photocatalytic properties and

electronic structure of Zr-based oxynitride, Zr2ON2, derived from nitridation of ZrO2. ApplCatal A 324:77–82. doi:10.1016/j.apcata.2007.03.017

241. Maeda K, Terashima H, Kase K et al (2008) Surface modification of TaON with monoclinicZrO2 to produce a composite photocatalyst with enhanced hydrogen evolution activity undervisible light. Bull Chem Soc Jpn 81:927–937. doi:10.1246/bcsj.81.927

242. Liu M, You W, Lei Z et al (2004) Water reduction and oxidation on Pt–Ru/Y2Ta2O5N2

catalyst under visible light irradiation. Chem Commun 2192–2193. doi:10.1039/B407892F243. Ji SM, Borse PH, Kim HG et al (2005) Photocatalytic hydrogen production from water–

methanol mixtures using N-doped Sr2Nb2O7 under visible light irradiation: effects of catalyststructure. Phys Chem Chem Phys 7:1315–1321. doi:10.1039/B417052K

244. Maeda K, Teramura K, Domen K (2008) Effect of post-calcination on photocatalytic activityof (Ga1−xZnx)(N1−xOx) solid solution for overall water splitting under visible light. J Catal254:198–204. doi:10.1016/j.jcat.2007.12.009

245. Maeda K, Teramura K, Lu D et al (2007) Roles of Rh/Cr2O3 (core/shell) nanoparticlesphotodeposited on visible-light-responsive (Ga1−xZnx)(N1−xOx) solid solutions in photocat-alytic overall water splitting. J Phys Chem C 111:7554–7560. doi:10.1021/jp071056j

246. Maeda K, Teramura K, Lu D et al (2006) Characterization of Rh–Cr mixed-oxidenanoparticles dispersed on (Ga1−xZnx)(N1−xOx) as a cocatalyst for visible-light-drivenoverall water splitting. J Phys Chem B 110:13753–13758. doi:10.1021/jp061829o

247. Teramura K, Maeda K, Saito T et al (2005) Characterization of ruthenium oxide nanoclusteras a cocatalyst with (Ga1−xZnx)(N1−xOx) for photocatalytic overall water splitting. J PhysChem B 109:21915–21921. doi:10.1021/jp054313y

248. Maeda K, Teramura K, Takata T et al (2005) Overall water splitting on (Ga1−xZnx)(N1−xOx)solid solution photocatalyst: relationship between physical properties and photocatalyticactivity. J Phys Chem B 109:20504–20510. doi:10.1021/jp053499y

249. Lee Y, Terashima H, Shimodaira Y et al (2007) Zinc germanium oxynitride as aphotocatalyst for overall water splitting under visible light. J Phys Chem C 111:1042–1048.doi:10.1021/jp0656532

250. Reyes-Gil KR, Reyes-Garcia EA, Raftery D (2007) Nitrogen-doped In2O3 thin filmelectrodes for photocatalytic water splitting. J Phys Chem C 111:14579–14588. doi:10.1021/jp072831y

251. Mohapatra SK, Misra M, Mahajan VK, Raja KS (2007) Design of a highly efficientphotoelectrolytic cell for hydrogen generation by water splitting: application of TiO2−xCx

nanotubes as a photoanode and Pt/TiO2 nanotubes as a cathode. J Phys Chem C 111:8677–8685. doi:10.1021/jp071906v

252. Park JH, Kim S, Bard AJ (2006) Novel carbon-doped TiO2 nanotube arrays with high aspectratios for efficient solar water splitting. Nano Lett 6:24–28. doi:10.1021/nl051807y

253. Liu Z, Pesic B, Raja KS et al (2009) Hydrogen generation under sunlight by self orderedTiO2 nanotube arrays. Inter J Hydrogen En 34:3250–3257. doi:10.1016/j.ijhydene.2009.02.044

254. Shaban YA, Khan SU (2008) Visible light active carbon modified n-TiO2 for efficienthydrogen production by photoelectrochemical splitting of water. Inter J Hydrogen En33:1118–1126. doi:10.1016/j.ijhydene.2007.11.026

255. Shaban YA, Khan SU (2009) Carbon modified (CM)-n-TiO2 thin films for efficient watersplitting to H2 and O2 under xenon lamp light and natural sunlight illuminations. J Sol StateElectrochem 13:1025–1036. doi:10.1007/s10008-009-0823-4

106 2 Semiconductor-Based Photocatalytic Systems …

256. Randeniya LK, Murphy AB, Plumb IC (2008) A study of S-doped TiO2 for photoelec-trochemical hydrogen generation from water. J Mater Sci 43:1389–1399. doi:10.1007/s10853-007-2309-z

257. Ogisu K, Ishikawa A, Teramura K et al (2007) Lanthanum-indium oxysulfide as a visiblelight driven photocatalyst for water splitting. Chem Lett 36:854–855. doi:10.1246/cl.2007.854

258. Ryu SY, Balcerski W, Lee TK, Hoffmann MR (2007) Photocatalytic production of hydrogenfrom water with visible light using hybrid catalysts of CdS attached to microporous andmesoporous silicas. J Phys Chem C 111:18195–18203. doi:10.1021/jp074860e

259. Hirai T, Nanba M, Komasawa I (2003) Dithiol-mediated incorporation of CdS nanoparticlesfrom reverse micellar system into Zn-doped SBA-15 mesoporous silica and their photocat-alytic properties. J Colloid Interface Sci 268:394–399. doi:10.1016/j.jcis.2003.09.011

260. Khatamian M, Oskoui MS, Haghighi M (2014) Photocatalytic hydrogen generation overCdS–metalosilicate composites under visible light irradiation. New J Chem 38:1684–1693.doi:10.1039/C3NJ01348K

261. Peng R, Zhao D, Baltrusaitis J et al (2012) Visible light driven photocatalytic evolution ofhydrogen from water over CdS encapsulated MCM-48 materials. RSC Adv 2:5754–5767.doi:10.1039/C2RA20714A

262. Peng R, Wu CM, Baltrusaitis J et al (2013) Ultra-stable CdS incorporated Ti-MCM-48mesoporous materials for efficient photocatalytic decomposition of water under visible lightillumination. Chem Commun 49:3221–3223. doi:10.1039/C3CC41362D

263. Peng R, Lin C, Baltrusaitis J et al (2014) Insight into band positions and inter-particleelectron transfer dynamics between CdS nanoclusters and spatially isolated TiO2 dispersedin cubic MCM-48 mesoporous materials: a highly efficient system for photocatalytichydrogen evolution under visible light illumination. Phys Chem Chem Phys 16:2048–2061.doi:10.1039/C3CP52801D

264. Henglein A, Gutierrez M (1983) Photochemistry of colloidal metal sulfides. 5. Fluorescenceand chemical reactions of ZnS and ZnS/CdS co-colloids. Ber Bunsenges Phys Chem87:852–858. doi:10.1002/bbpc.19830871005

265. Deshpande A, Shah P, Gholap RS, Gupta NM (2009) Interfacial and physico-chemicalproperties of polymer-supported CdS-ZnS nanocomposites and their role in the visible-lightmediated photocatalytic splitting of water. J Colloid Interface Sci 333:263–268. doi:10.1016/j.jcis.2009.01.037

266. Lunawat PS, Senapati S, Kumar R, Gupta NM (2007) Visible light-induced splitting ofwater using CdS nanocrystallites immobilized over water-repellant polymeric surface. Inter JHydrogen En 32:2784–2790. doi:10.1016/j.ijhydene.2007.04.001

267. Kim YK, Kim M, Hwang SH et al (2015) CdS-loaded flexible carbon nanofiber mats as aplatform for solar hydrogen production. Inter J Hydrogen En 40:136–145. doi:10.1016/j.ijhydene.2014.11.011

268. Hirai T, Bando Y, Komasawa I (2002) Immobilization of CdS nanoparticles formed inreverse micelles onto alumina particles and their photocatalytic properties. J Phys Chem B106:8967–8970. doi:10.1021/jp020386v

269. Hirai T, Nanba M, Komasawa I (2002) Dithiol-mediated immobilization of CdSnanoparticles from reverse micellar system onto Zn-doped silica particles and their highphotocatalytic activity. J Colloid Interface Sci 252:89–92. doi:10.1006/jcis.2002.8430

270. Yu G, Geng L, Wu S et al (2015) Highly-efficient cocatalyst-free H2-evolution oversilica-supported CdS nanoparticle photocatalysts under visible light. Chem Commun51:10676–10679. doi:10.1039/C5CC02249E

271. Yu G, Zhang W, Sun Y et al (2016) A highly active cocatalyst-free semiconductorphotocatalyst for visible-light-driven hydrogen evolution: synergistic effect of surfacedefects and spatial bandgap engineering. J Mater Chem A 4:13803–13808. doi:10.1039/c6ta03803d

References 107

272. Bb Kale, Baeg JO, Apte SK et al (2007) Confinement of nano CdS in designated glass: anovel functionality of quantum dot–glass nanosystems in solar hydrogen production. J MaterChem 17:4297–4303. doi:10.1039/B708269J

273. Apte SK, Garaje SN, Valant M, Kale BB (2012) Eco-friendly solar light driven hydrogenproduction from copious waste H2S and organic dye degradation by stable and efficientorthorhombic CdS quantum dots–GeO2 glass photocatalyst. Green Chem 14:1455–1462.doi:10.1039/C2GC16416G

274. Kanade KG, Baeg JO, Mulik UP et al (2006) Nano-CdS by polymer-inorganic solid-statereaction: visible light pristine photocatalyst for hydrogen generation. Mater Res Bull41:2219–2225. doi:10.1016/j.materresbull.2006.04.031

275. Li W, O’Dowd G, Whittles TJ et al (2015) Colloidal dual-band gap cell for photocatalytichydrogen generation. Nanoscale 7:16606–16610. doi:10.1039/c5nr04950d

276. Baldovi HG, Latorre-Sanchez M, Esteve-Adell I et al (2016) Generation of MoS2 quantumdots by laser ablation of MoS2 particles in suspension and their photocatalytic activity for H2

generation. J Nanopart Res 18:240–248. doi:10.1007/s11051-016-3540-9277. Hirai T, Nomura Y, Komasawa I (2003) Immobilization of RuS2 nanoparticles prepared in

reverse micellar system onto thiol-modified polystyrene particles and their photocatalyticproperties. J Nanoparticle Res 5:61–67. doi:10.1023/A:1024422226598

278. Barawi M, Ferrer IJ, Flores E et al (2016) Hydrogen photoassisted generation by visible lightand an earth abundant photocatalyst: pyrite (FeS2). J Phys Chem C 120:9547–9552. doi:10.1021/acs.jpcc.5b11482

279. Li G, Su R, Rao J et al (2016) Band gap narrowing of SnS2 superstructures with improvedhydrogen production. J Mater Chem A 4:209–216. doi:10.1039/c5ta07283b

280. Shen S, Zhao L, Guo L (2008) Cetyltrimethylammoniumbromide (CTAB)-assistedhydrothermal synthesis of ZnIn2S4 as an efficient visible-light-driven photocatalyst forhydrogen production. Inter J Hydrogen En 33:4501–4510. doi:10.1016/j.ijhydene.2008.05.043

281. Lei Z, You W, Liu M et al (2003) Photocatalytic water reduction under visible light on anovel ZnIn2S4 catalyst synthesized by hydrothermal method. Chem Commun 2142–2143.doi:10.1039/B306813G

282. Guijun M, Hongjian Y, Xu Z et al (2008) Photocatalytic splitting of H2S to producehydrogen by gas-solid phase reaction. Chin J Catal 29:313–315. doi:10.1016/S1872-2067(08)60029-7

283. Shen S, Zhao L, Guo L (2009) Crystallite, optical and photocatalytic properties ofvisible-light-driven ZnIn2S4 photocatalysts synthesized via a surfactant-assisted hydrother-mal method. Mater Res Bull 44:100–105. doi:10.1016/j.materresbull.2008.03.027

284. Shen J, Zai J, Yuan Y, Qian X (2012) 3D hierarchical ZnIn2S4: the preparation andphotocatalytic properties on water splitting. Inter J Hydrogen En 37:16986–16993. doi:10.1016/j.ijhydene.2012.08.038

285. Fan WJ, Zhou ZF, Xu WB et al (2010) Preparation of ZnIn2S4/fluoropolymer fibercomposites and its photocatalytic H2 evolution from splitting of water using Xe lampirradiation. Inter J Hydrogen En 35:6525–6530. doi:10.1016/j.ijhydene.2010.04.036

286. Li Y, Zhang K, Peng S et al (2012) Photocatalytic hydrogen generation in the presence ofethanolamines over Pt/ZnIn2S4 under visible light irradiation. J Mol Catal A 363–364:354–361. doi:10.1016/j.molcata.2012.07.011

287. Chaudhari NS, Bhirud AP, Sonawane RS et al (2011) Ecofriendly hydrogen production fromabundant hydrogen sulfide using solar light-driven hierarchical nanostructured ZnIn2S4photocatalyst. Green Chem 13:2500–2506. doi:10.1039/C1GC15515F

288. Li F, Luo J, Chen G et al (2014) Hydrothermal synthesis of zinc indium sulfide microsphereswith Ag+ doping for enhanced H2 production by photocatalytic water splitting under visiblelight. Catal Sci Technol 4:1144–1150. doi:10.1039/C3CY00952A

108 2 Semiconductor-Based Photocatalytic Systems …

289. Shang L, Zhou C, Bian T et al (2013) Facile synthesis of hierarchical ZnIn2S4submicrospheres composed of ultrathin mesoporous nanosheets as a highly efficientvisible-light-driven photocatalyst for H2 production. J Mater Chem A 1:4552–4558. doi:10.1039/C3TA01685D

290. Xu Z, Li Y, Peng S et al (2012) NaCl-assisted low temperature synthesis of layered Zn-In-Sphotocatalyst with high visible-light activity for hydrogen evolution. RSC Adv 2:3458–3466. doi:10.1039/C2RA01159J

291. Chen Y, He J, Li J et al (2016) Hydrilla derived ZnIn2S4 photocatalyst with hexagonal-cubicphase junctions: A bio-inspired approach for H2 evolution. Catal Commun 87:1–5. doi:10.1016/j.catcom.2016.08.031

292. Tian F, Zhu R, Zhong J et al (2016) An efficient preparation method of RGO/ZnIn2S4 forphotocatalytic hydrogen generation under visible light. Inter J Hydrogen En 41:20156–20171. doi:10.1016/j.ijhydene.2016.08.063

293. Shen S, Zhao L, Zhou Z, Guo L (2008) Enhanced photocatalytic hydrogen evolution overCu-doped ZnIn2S4 under visible light irradiation. J Phys Chem C 112:16148–16155. doi:10.1021/jp804525q

294. Ding J, Sun S, Yan W et al (2013) Photocatalytic H2 evolution on a novel CaIn2S4photocatalyst under visible light irradiation. Inter J Hydrogen En 38:13153–13158. doi:10.1016/j.ijhydene.2013.07.109

295. Ding J, Li X, Chen L et al (2016) Au–Pt alloy nanoparticles site-selectively deposited onCaIn2S4 nanosteps as efficient photocatalysts for hydrogen production. J Mater Chem A4:12630–12637. doi:10.1039/c6ta04468a

296. Zhou Q, Kang SZ, Li X et al (2015) AgGaS2 nanoplates loaded with CuS: an efficient visiblephotocatalyst for rapid H2 evolution. Inter J Hydrogen En 40:4119–4128. doi:10.1016/j.ijhydene.2015.01.143

297. Iwase A, Ng YH, Amal R, Kudo A (2015) Solar hydrogen evolution using a CuGaS2photocathode improved by incorporating reduced graphene oxide. J Mater Chem A 3:8566–8570. doi:10.1039/C5TA01237F

298. Yu X, An X, Shavel A et al (2014) The effect of the Ga content on the photocatalytichydrogen evolution of CuIn1−xGaxS2 nanocrystals. J Mater Chem A 2:12317–12322. doi:10.1039/C4TA01315H

299. Kato T, Hakari Y, Ikeda S et al (2015) Utilization of metal sulfide material of (CuGa)1−xZn2xS2 solid solution with visible light response in photocatalytic and photoelectrochem-ical solar water splitting systems. J Phys Chem Lett 6:1042–1047. doi:10.1021/acs.jpclett.5b00137

300. Kandiel TA, Huttona GA, Reisner E (2016) Visible light driven hydrogen evolution with anoble metal free CuGa2In3S8 nanoparticle system in water. Catal Sci Technol 6:6536–6541.doi:10.1039/c6cy01103a

301. Quintans CS, Kato H, Kobayashi M et al (2015) Improvement of hydrogen evolution undervisible light over Zn1−2x(CuGa)xGa2S4 photocatalysts by synthesis utilizing a polymerizablecomplex method. J Mater Chem A 3:14239–14244. doi:10.1039/C5TA02114F

302. Chen F, Zai J, Xu M, Qian X (2013) 3D-hierarchical Cu3SnS4 flowerlike micro-spheres:controlled synthesis, formation mechanism and photocatalytic activity for H2 evolution fromwater. J Mater Chem A 1:4316–4323. doi:10.1039/C3TA01491F

303. Kush P, Deori K, Kumar A, Deka S (2015) Efficient hydrogen/oxygen evolution andphotocatalytic dye degradation and reduction of aqueous Cr(VI) by surfactant freehydrophilic Cu2ZnSnS4 nanoparticles. J Mater Chem A 3:8098–8106. doi:10.1039/C4TA06551D

304. Yu X, Shavel A, An X et al (2014) Cu2ZnSnS4-Pt and Cu2ZnSnS4-Au heterostructurednanoparticles for photocatalytic water splitting and pollutant degradation. J Am Chem Soc136:9236–9239. doi:10.1021/ja502076b

305. Gonce MK, Dogru M, Aslan E et al (2015) Photocatalytic hydrogen evolution based onCu2ZnSnS4, Cu2ZnSnSe4 and Cu2ZnSnSe4−xSx nanofibers. RSC Adv 5:94025–94028.doi:10.1039/c5ra18877f

References 109

306. Yu X, An X, Genç A et al (2015) Cu2ZnSnS4–PtM (M = Co, Ni) nanoheterostructures forphotocatalytic hydrogen evolution. J Phys Chem C 119:21882–21888. doi:10.1021/acs.jpcc.5b06199

307. Zhang ZX, Chong RF, Meng YN et al (2015) High temperature recrystallization of kersteriteCu2ZnSnS4 towards enhanced photocatalytic H2 evolution. Inter J Hydrogen En 40:13456–13462. doi:10.1016/j.ijhydene.2015.08.032

308. Zheng L, Xu Y, Song Y et al (2009) Nearly monodisperse CuInS2 hierarchicalmicro-architectures for photocatalytic H2 evolution under visible light. Inorg Chem48:4003–4009. doi:10.1021/ic802399f

309. Gannouni M, Assaker IB, Chtourou R (2015) Photoelectrochemical cell based on n-CuIn5S8film as photoanodes for photocatalytic water splitting. Inter J Hydrogen En 40:7252–7259.doi:10.1016/j.ijhydene.2015.04.057

310. Guan Z, Luo W, Feng J et al (2015) Selective etching of metastable phase induced anefficient CuIn0.7Ga0.3S2 nano-photocathode for solar water splitting. J Mater Chem A3:7840–7848. doi:10.1039/C5TA01259G

311. Hu P, Ngaw CK, Tay YY et al (2015) A “uniform” heterogeneous photocatalyst: integratedp–n type CuInS2/NaInS2 nanosheets by partial ion exchange reaction for efficient H2

evolution. Chem Commun 51:9381–9384. doi:10.1039/C5CC02237A312. Zhang X, Du Y, Zhou Z, Guo L (2010) A simplified method for synthesis of

band-structure-controlled (CuIn)xZn2(1−x)S2 solid solution photocatalysts with high activityof photocatalytic H2 evolution under visible-light irradiation. Inter J Hydrogen En 35:3313–3321. doi:10.1016/j.ijhydene.2010.01.111

313. Zhang X, Yang M, Zhao J, Guo L (2013) Photocatalytic hydrogen evolution withsimultaneous degradation of organics over (CuIn)0.2Zn1.6S2 solid solution. Inter J HydrogenEn 38:15985–15991. doi:10.1016/j.ijhydene.2013.10.014

314. Huang Y, Chen J, Zou W et al (2015) Enhanced photocatalytic hydrogen evolutionefficiency using hollow microspheres of (CuIn)xZn2(1−x)S2 solid solutions. Dalton Trans44:10991–10996. doi:10.1039/C5DT01269D

315. Lin y, Zhang F, Pan D (2012) A facile route to (ZnS)x(CuInS2)1−x hierarchical microsphereswith excellent water-splitting ability. J Mater Chem 22:22619–22623. doi:10.1039/C2JM35166H

316. Tang X, Tay Q, Chen Z et al (2013) Cu–In–Zn–S nanoporous spheres for highly efficientvisible-light-driven photocatalytic hydrogen evolution. New J Chem 37:1878–1882. doi:10.1039/C3NJ00266G

317. Xu M, Zai J, Yuan Y, Qian X (2012) Band gap-tunable (CuIn)xZn2(1−x)S2 solid solutions:preparation and efficient photocatalytic hydrogen production from water under visible lightwithout noble metals. J Mater Chem 22:23929–23934. doi:10.1039/C2JM35375J

318. Tang X, Tay Q, Chen Z et al (2013) CuInZnS-decorated graphene nanosheets for highlyefficient visible-light-driven photocatalytic hydrogen production. J Mater Chem A 1:6359–6365. doi:10.1039/C3TA01602A

319. Zhang G, Zhang W, Wang P et al (2013) Stability of an H2-producing photo-catalyst (Ru/(CuAg)0.15In0.3Zn1.4S2) in aqueous solution under visible light irradiation. Inter J HydrogenEn 38:1286–1296. doi:10.1016/j.ijhydene.2012.11.033

320. Zhang G, Zhang W, Minakata D et al (2013) The pH effects on H2 evolution kinetics forvisible light water splitting over the Ru/(CuAg)0.15In0.3Zn1.4S2 photocatalyst. Inter JHydrogen En 38:11727–11736. doi:10.1016/j.ijhydene.2013.06.140

321. Tsuji I, Kato H, Kudo A (2006) Photocatalytic hydrogen evolution onZnS − CuInS2 − AgInS2 solid solution photocatalysts with wide visible light absorptionbands. Chem Mater 18:1969–1975. doi:10.1021/cm0527017

322. Tsuji I, Kato H, Kobayashi H, Kudo A (2005) Photocatalytic H2 evolution undervisible-light irradiation over band-structure-controlled (CuIn)xZn2(1−x)S2 solid solutions.J Phys Chem B 109:7323–7329. doi:10.1021/jp044722e

110 2 Semiconductor-Based Photocatalytic Systems …

323. Li Y, Chen G, Zhou C, Sun J (2009) A simple template-free synthesis of nanoporous ZnS–In2S3–Ag2S solid solutions for highly efficient photocatalytic H2 evolution under visiblelight. Chem Commun 2020–2022. doi:10.1039/B819300B

324. Kudo A, Tsuji I, Kato H (2002) AgInZn7S9 solid solution photocatalyst for H2 evolutionfrom aqueous solutions under visible light irradiation. Chem Commun 1958–1959. doi:10.1039/B204259B

325. Kale BB, Baeg JO, Lee SM et al (2006) CdIn2S4 nanotubes and “marigold” nanostructures: avisible-light photocatalyst. Adv Func Mater 16:1349–1354. doi:10.1002/adfm.200500525

326. Yu Y, Chen G, Wang G, Lv Z (2013) Visible-light-driven ZnIn2S4/CdIn2S4 compositephotocatalyst with enhanced performance for photocatalytic H2 evolution. Inter J HydrogenEn 38:1278–1285. doi:10.1016/j.ijhydene.2012.11.020

327. Chen X, Li L, Zhang W et al (2016) Fabricate globular flower-like CuS/CdIn2S4/ZnIn2S4with high visible light response via microwave-assisted one − step method and itsmultipathway photoelectron migration properties for hydrogen evolution and pollutantdegradation. ACS Sustainable Chem Eng 4:6680–6688. doi:10.1021/acssuschemeng.6b01543

328. Mei Z, Ouyang S, Tang DM et al (2013) An ion-exchange route for the synthesis ofhierarchical In2S3/ZnIn2S4 bulk composite and its photocatalytic activity under visible-lightirradiation. Dalton Trans 42:2687–2690. doi:10.1039/C2DT32271D

329. Hou J, Yang C, Cheng H et al (2013) Ternary 3D architectures of CdS QDs/graphene/ZnIn2S4 heterostructures for efficient photocatalytic H2 production. Phys Chem Chem Phys15:15660–15668. doi:10.1039/C3CP51857D

330. Chen D, Ye J (2007) Photocatalytic H2 evolution under visible light irradiation on AgIn5S8photocatalyst. J Phys Chem Sol 68:2317–2320. doi:10.1016/j.jpcs.2007.07.059

331. Kudo A, Nagane A, Tsuji I, Kato H (2002) H2 evolution from aqueous potassium sulfitesolutions under visible light irradiation over a novel sulfide photocatalyst NaInS2 with alayered structure. Chem Lett 31:882–883. doi:10.1246/cl.2002.882

332. Shen S, Guo L (2006) Structural, textural and photocatalytic properties of quantum-sizedIn2S3-sensitized Ti-MCM-41 prepared by ion-exchange and sulfidation methods. J Sol StateChem 179:2629–2635. doi:10.1016/j.jssc.2006.05.010

333. Li S, Wang C, Qiu H (2015) Single- and few-layer ZrS2 as efficient photocatalysts forhydrogen production under visible light. Inter J Hydrogen En 40:15503–15509. doi:10.1016/j.ijhydene.2015.08.110

334. Wang F, Shifa TA, Zhan X et al (2015) Recent advances in transition-metal dichalcogenidebased nanomaterials for water splitting. Nanoscale 7:19764–19788. doi:10.1039/c5nr06718a

335. Lin Z, Ning S, Yang Z et al (2016) Large-scale preparation of heterometallic chalcogenideMnSb2S4 monolayer nanosheets with a high visible-light photocatalytic activity for H2

evolution. Chem Commun 52:13381–13384. doi:10.1039/c6cc07127a336. Bessekhouad Y, Mohammedi M, Trari M (2002) Hydrogen photoproduction from hydrogen

sulfide on Bi2S3 catalyst. Sol En Mater Sol Cells 73:339–350. doi:10.1016/S0927-0248(01)00218-5

337. Abdi A, Denoyelle A, Commenges-Bernole M, Trari M (2013) Photocatalytic hydrogenevolution on new mesoporous material Bi2S3/Y-zeolite. Inter J Hydrogen En 38:2070–2078.doi:10.1016/j.ijhydene.2012.11.085

338. Zhang K, Jing D, Xing C, Guo L (2007) Significantly improved photocatalytic hydrogenproduction activity over Cd1−xZnxS photocatalysts prepared by a novel thermal sulfurationmethod. Inter J Hydrogen En 32:4685–4691. doi:10.1016/j.ijhydene.2007.08.022

339. del Valle F, Ishikawa A, Domen K et al (2009) Influence of Zn concentration in the activityof Cd1−xZnxS solid solutions for water splitting under visible light. Catal Today 143:51–56.doi:10.1016/j.cattod.2008.09.024

340. Chan CC, Chang CC, Hsu CH et al (2014) Efficient and stable photocatalytic hydrogenproduction from water splitting over ZnxCd1−xS solid solutions under visible lightirradiation. Inter J Hydrogen En 39:1630–1639. doi:10.1016/j.ijhydene.2013.11.059

References 111

341. Du H, Liang K, Yuan CZ et al (2016) Bare Cd1−xZnxS ZB/WZ heterophase nano-junctionsfor visible light photocatalytic hydrogen production with high efficiency. ACS Appl MaterInterfaces 8:24550–24558. doi:10.1021/acsami.6b06182

342. Roy AM, De GC (2003) Immobilisation of CdS, ZnS and mixed ZnS–CdS on filter paper:effect of hydrogen production from alkaline Na2S/Na2S2O3 solution. J PhotochemPhotobiol A 157:87–92. doi:10.1016/S1010-6030(02)00430-6

343. Zhang X, Jing D, Liu M, Guo L (2008) Efficient photocatalytic H2 production under visiblelight irradiation over Ni doped Cd1−xZnxS microsphere photocatalysts. Catal Commun9:1720–1724. doi:10.1016/j.catcom.2008.01.032

344. Stroyuk AL, Raevskaya AE, Korzhak AV et al (2009) Photocatalytic production ofhydrogen in systems based on CdxZn1−xS/Ni

0 nanostructures. Theoret Exp Chem 45:12–22345. Kimi M, Yuliati L, Shamsuddin M (2011) Photocatalytic hydrogen production under visible

light over Cd0.1SnxZn0.9−2xS solid solution photocatalysts. Inter J Hydrogen En 36:9453–9461. doi:10.1016/j.ijhydene.2011.05.044

346. Peng S, An R, Li Y et al (2012) Remarkable enhancement of photocatalytic hydrogenevolution over Cd0.5Zn0.5S by bismuth-doping. Inter J Hydrogen En 37:1366–1374. doi:10.1016/j.ijhydene.2011.09.140

347. Sathish M, Viswanath RP (2007) Photocatalytic generation of hydrogen over mesoporousCdS nanoparticle: effect of particle size, noble metal and support. Catal Today 129:421–427.doi:10.1016/j.cattod.2006.12.008

348. Li Y, Du Y, Peng S et al (2008) Enhancement of photocatalytic activity of cadmium sulfidefor hydrogen evolution by photoetching. Inter J Hydrogen En 33:2007–2013. doi:10.1016/j.ijhydene.2008.02.023

349. Chang CM, Orchard KL, Martindale CM, Reisner E (2016) Ligand removal from CdSquantum dots for enhanced photocatalytic H2 generation in pH neutral water. J MaterChem A 4:2856–2862. doi:10.1039/c5ta04136h

350. Jana MK, Gupta U, Rao CNR (2016) Hydrazine as a hydrogen carrier in the photocatalyticgeneration of H2 using CdS quantum dots. Dalton Trans 45:15137–15141. doi:10.1039/c6dt02505f

351. Silva LA, Ryu SY, Choi J et al (2008) Photocatalytic hydrogen production with visible lightover Pt-interlinked hybrid composites of cubic-phase and hexagonal-phase CdS. J PhysChem C 112:12069–12073. doi:10.1021/jp8037279

352. Li K, Han M, Chen R et al (2016) Hexagonal@Cubic CdS Core@Shell nanorodphotocatalyst for highly active production of H2 with unprecedented stability. Adv Mater28:8906–8911. doi:10.1002/adma.201601047

353. Baran MP, Korsunskaya NE, Stara TR et al (2016) Graded ZnS/ZnSxO1−x heterostructuresproduced by oxidative photolysis of zinc sulfide: structure, optical properties andphotocatalytic evolution of molecular hydrogen. J Photochem Photobiol A 329:213–220.doi:10.1016/j.jphotochem.2016.07.003

354. Jang JS, Joshi UA, Lee JS (2007) Solvothermal synthesis of CdS nanowires forphotocatalytic hydrogen and electricity production. J Phys Chem C 111:13280–13287.doi:10.1021/jp072683b

355. Kida T, Guan G, Minami Y et al (2003) Photocatalytic hydrogen production from water overa LaMnO3/CdS nanocomposite prepared by the reverse micelle method. J Mater Chem13:1186–1191. doi:10.1039/B211812B

356. Kida T, Guan G, Yamada N et al (2004) Hydrogen production from sewage sludgesolubilized in hot-compressed water using photocatalyst under light irradiation. Inter JHydrogen En 29:269–274. doi:10.1016/j.ijhydene.2003.08.007

357. Ma G, Yan H, Shi J et al (2008) Direct splitting of H2S into H2 and S on CdS-basedphotocatalyst under visible light irradiation. J Catal 260:134–140. doi:10.1016/j.jcat.2008.09.017

358. Frame FA, Carroll EC, Larsen DS et al (2008) First demonstration of CdSe as aphotocatalyst for hydrogen evolution from water under UV and visible light. Chem Commun2206–2208. doi:10.1039/B718796C

112 2 Semiconductor-Based Photocatalytic Systems …

359. Holmes MA, Townsend TK, Osterloh FE (2012) Quantum confinement controlledphotocatalytic water splitting by suspended CdSe nanocrystals. Chem Commun 48:371–373. doi:10.1039/C1CC16082F

360. Grigioni I, Bernareggi M, Sinibaldi G et al (2016) Size-dependent performance of CdSequantum dots in the photocatalytic evolution of hydrogen under visible light irradiation.Appl Catal A 518:176–180. doi:10.1016/j.apcata.2015.09.021

361. Zhao J, Holmes MA, Osterloh FE (2013) Quantum confinement controls photocatalysis: afree energy analysis for photocatalytic proton reduction at CdSe nanocrystals. ACS Nano7:4316–4325. doi:10.1021/nn400826h

362. Rasamani KD, Li Z, Sun Y (2016) Significant enhancement of photocatalytic water splittingenabled by elimination of surface traps in Pt-tipped CdSe nanorods Nanoscale 8:18621–18625. doi:10.1039/c6nr06902a

363. Costi R, Young ER, Bulović V, Nocera DG (2013) Stabilized CdSe-CoPi compositephotoanode for light-assisted water oxidation by transformation of a CdSe/Cobalt metal thinfilm. ACS Appl Mater Interfaces 5:2364–2367. doi:10.1021/am400364u

364. Yang S, Xu CY, Yang L et al (2016) Solution-phase synthesis of g-In2Se3 nanoparticles forhighly efficient photocatalytic hydrogen generation under simulated sunlight irradiation.RSC Adv 6:106671–106675. doi:10.1039/c6ra21784b

365. Gurunathan K, Baeg JO, Lee SM et al (2008) Visible light active pristine and Fe3+ dopedCuGa2O4 spinel photocatalysts for solar hydrogen production. Inter J Hydrogen En33:2646–2652. doi:10.1016/j.ijhydene.2008.03.018

366. Saadi S, Bouguelia A, Derbal A, Trari M (2007) Hydrogen photoproduction over newcatalyst CuLaO2. J Photochem Photobiol A 187:97–104. doi:10.1016/j.jphotochem.2006.09.017

367. Koriche N, Bouguelia A, Trari M (2006) Photocatalytic hydrogen production over newoxide CuLaO2.62. Inter J Hydrogen En 31:1196–1203. doi:10.1016/j.ijhydene.2005.08.015

368. Koriche N, Bouguelia A, Aider A, Trari M (2005) Photocatalytic hydrogen evolution overdelafossite CuAlO2. Inter J Hydrogen En 30:693–699. doi:10.1016/j.ijhydene.2004.06.011

369. Zhou C, Zhao Y, Shang L et al (2016) Facile synthesis of ultrathin SnNb2O6 nanosheetstowards improved visible-light photocatalytic H2-production activity. Chem Commun52:8239–8242. doi:10.1039/c6cc03739a

370. Ye J, Zou Z, Arakawa H et al (2002) Correlation of crystal and electronic structures withphotophysical properties of water splitting photocatalysts InMO4 (M = V5+, Nb5+, Ta5+).J Photochem Photobiol A 148:79–83. doi:10.1016/S1010-6030(02)00074-6

371. Zhu H, Fang M, Hyang Z et al (2016) Novel carbon-incorporated porous ZnFe2O4

nanospheres for enhanced photocatalytic hydrogen generation under visible light irradiation.RSC Adv 6:56069–56076. doi:10.1039/c6ra05098k

372. Pai YH, Tsai CT, Fang SY (2013) Enhanced photocatalytic hydrogen generation with PtNanoparticles on multi-phase polycrystalline microporous MnO2 photocatalyst. J PowerSources 223:107–113. doi:10.1016/j.jpowsour.2012.09.024

373. Jin S, Wang X, Ju M et al (2015) Effect of phase junction structure on the photocatalyticperformance in overall water splitting: Ga2O3 photocatalyst as an example. J Phys Chem C119:18221–18228. doi:10.1021/acs.jpcc.5b04092

374. Hu C, Chu K, Zhao Y, Teoh WY (2014) Efficient photoelectrochemical water splitting overanodized p-Type NiO porous films. ACS Appl Mater Interfaces 6:18558–18568. doi:10.1021/am507138b

375. Manikandan M, Tanabe T, Li P et al (2014) Photocatalytic water splitting under visible lightby mixed-valence Sn3O4. ACS Appl Mater Interfaces 6:3790–3793. doi:10.1021/am500157u

376. Tijare SN, Joshi MV, Padole PS et al (2012) Photocatalytic hydrogen generation throughwater splitting on nano-crystalline LaFeO3 perovskite. Inter J Hydrogen En 37:10451–10456. doi:10.1016/j.ijhydene.2012.01.120

References 113

377. May KJ, Fenning DP, Ming T et al (2015) Thickness-dependent photoelectrochemical watersplitting on ultrathin LaFeO3 films grown on Nb:SrTiO3. J Phys Chem Lett 6:977–985.doi:10.1021/acs.jpclett.5b00169

378. Pan C, Takata T, Nakabayashi M et al (2015) A complex perovskite-type oxynitride: the firstphotocatalyst for water splitting operable at up to 600 nm. Angew Chem Int Ed 54:1–6.doi:10.1002/anie.201410961

379. Lee CW, Kim DW, Cho IS et al (2012) Simple synthesis and characterization of SrSnO3

nanoparticles with enhanced photocatalytic activity. Inter J Hydrogen En 37:10557–10563.doi:10.1016/j.ijhydene.2012.04.063

380. Shibli SMA, Arun PS, Raj AV (2015) Exploration of octahedrally shaped MnCo2O4 catalystparticles for visible light driven photocatalytic water splitting reaction. RSC Adv 5:19393–19399. doi:10.1039/C4RA12646G

381. Liu J, Wen S, Zou X et al (2013) Visible-light-responsive copper(II) borate photocatalystswith intrinsic midgap states for water splitting. J Mater Chem A 1:1553–1556. doi:10.1039/C2TA00522K

382. Luan J, Guo N, Chen B (2014) Hydrogen production with Ga2BiSbO7, Fe2BiSbO7 andGd2BiSbO7 as photocatalysts under visible light irradiation. Inter J Hydrogen En 39:1228–1236. doi:10.1016/j.ijhydene.2013.11.020

383. Mahapure SA, Palei PK, Nikam LK et al (2013) Novel nanocrystalline zinc silverantimonate (ZnAg3SbO4): an efficient & ecofriendly visible light photocatalyst withenhanced hydrogen generation. J Mater Chem A 1:12835–12840. doi:10.1039/C3TA12883K

384. Guo S, Han S (2014) Constructing a novel hierarchical 3D flower-like nano/micro titaniumphosphate with efficient hydrogen evolution from water splitting. J Power Sources 267:9–13.doi:10.1016/j.jpowsour.2014.05.011

385. Wang G, Jing Y, Ju J et al (2015) Ga4B2O9: an efficient borate photocatalyst for overallwater splitting without cocatalyst. Inorg Chem 54:2945–2949. doi:10.1021/ic5031087

386. Liang J, Xu J, Gu Q et al (2013) A novel Zn2GeO4 superstructure for effective photocatalytichydrogen generation. J Mater Chem A 1:7798–7805. doi:10.1039/C3TA11374D

387. Yan S, Wan L, Li Z, Zou Z (2011) Facile temperature-controlled synthesis of hexagonalZn2GeO4 nanorods with different aspect ratios toward improved photocatalytic activity foroverall water splitting and photoreduction of CO2. Chem Commun 47:5632–5634. doi:10.1039/C1CC10513B

388. Pan B, Xie Q, Wang H et al (2013) Synthesis and photocatalytic hydrogen production of anovel photocatalyst LaCO3OH. J Mater Chem A 1:6629–6634. doi:10.1039/C3TA01553J

389. Ruiz-Gomez MA, Torres-Martinez LM, Figueroa-Torres MZ et al (2013) Hydrogenevolution from pure water over a new advanced photocatalyst Sm2GaTaO7. Inter J HydrogenEn 38:12554–12561. doi:10.1016/j.ijhydene.2012.11.131

390. Ye J, Zou Z, Oshikiri M et al (2002) A novel hydrogen-evolving photocatalyst InVO4 activeunder visible light irradiation. Chem Phys Lett 356:221–226. doi:10.1016/S0009-2614(02)00254-3

391. Lv M, Liu G, Xu X (2016) Homologous compounds ZnnIn2O3+n (n = 4, 5, and 7) containinglaminated functional groups as efficient photocatalysts for hydrogen production. ACS ApplMater Interfaces 8:28700–28708. doi:10.1021/acsami.6b10951

392. Huang Q, Ye Z, Xiao X (2015) Recent progress in photocathodes for hydrogen evolution.J Mater Chem A 3:15824–15837. doi:10.1039/c5ta03594e

393. Luo J, Steier L, Son MK et al (2016) Cu2O nanowire photocathodes for efficient and durablesolar water splitting. Nano Lett 16:1848–1857. doi:10.1021/acs.nanolett.5b04929

394. Dubale AA, Su WN, Tamirat AG et al (2014) The synergetic effect of graphene on Cu2Onanowire arrays as a highly efficient hydrogen evolution photocathode in water splitting.J Mater Chem A 2:18383–18397. doi:10.1039/c4ta03464c

395. Dubale AA, Tamirat AG, Chen H-M (2016) A highly stable CuS and CuS–Pt modifiedCu2O/CuO heterostructure as an efficient photocathode for the hydrogen evolution reaction.J Mater Chem A 4:2205–2216. doi:10.1039/c5ta09464j

114 2 Semiconductor-Based Photocatalytic Systems …

396. Dong Y, Chen Y, Jiang P et al (2016) A novel g-C3N4 based photocathode forphotoelectrochemical hydrogen evolution. RSC Adv 6:7465–7473. doi:10.1039/c5ra23265a

397. Basu M, Zhang ZW, Chen CJ et al (2016) CoSe2 embedded in C3N4: an efficientphotocathode for photoelectrochemical water splitting. ACS Appl Mater Interfaces 8:26690–26696. doi:10.1021/acsami.6b06520

398. Li S, Zhang P, Song X, Gao L (2015) Photoelectrochemical hydrogen production of TiO2

passivated Pt/Si-nanowire composite photocathode. ACS Appl Mater Interfaces 7:18560–18565. doi:10.1021/acsami.5b04936

399. Bao XQ, Cerqueira MF, Alpuimab P, Liu L (2015) Silicon nanowire arrays coupled withcobalt phosphide spheres as low-cost photocathodes for efficient solar hydrogen evolution.Chem Commun 51:10742–10745. doi:10.1039/c5cc02331a

400. Lewis NS (2016) Developing a scalable artificial photosynthesis technology throughnanomaterials by design. Nat Nanotechnol 11:1010–1019. doi:10.1038/nnano.2016.194

401. Chandrasekaran S, McInnes SJP, Macdonald TJ (2015) Porous silicon nanoparticles as ananophotocathode for photoelectrochemical water splitting. RSC Adv 5:85978–85982.doi:10.1039/c5ra12559f

402. Reece SY, Hamel JA, Sung K, Jarvi TD, Esswein AJ, Pijpers JJH, Nocera DG (2011)Wireless solar water splitting using silicon-based semiconductors and earth-abundantcatalysts. Science 34:645–648. doi:10.1126/science.1209816

403. Patra BK, Khilari S, Pradhan D, Pradhan N (2016) Hybrid dot-disk Au-CuInS2nanostructures as active photocathode for efficient evolution of hydrogen from water.Chem Mater 28:4358–4366. doi:10.1021/acs.chemmater.6b01357

404. Chae SY, Park SJ, Han SG et al (2016) Enhanced photocurrents with ZnS passivated Cu(In,Ga)(Se, S)2 photocathodes synthesized using a nonvacuum process for solar water splitting.J Am Chem Soc 138:15673–15681. doi:10.1021/jacs.6b09595

405. Kaneko H, Minegishi T, Nakabayashi M et al (2016) Enhanced hydrogen evolution undersimulated sunlight from neutral electrolytes on (ZnSe)0.85(CuIn0.7Ga0.3Se2)0.15 photocath-odes prepared by a bilayer method. Angew Chem Int Ed 55:15329–15333. doi:10.1002/anie.201609202

406. Abe T, Fukui K, Kawai Y et al (2016) A water splitting system using an organophotocathodeand titanium dioxide photoanode capable of bias-free H2 and O2 evolution. Chem Commun52:7735–7737. doi:10.1039/c6cc01225f

407. Hu Z, Yuan L, Liu Z et al (2016) An elemental phosphorus photocatalyst with a record highhydrogen evolution efficiency. Angew Chem Int Ed 55:9580–9585. doi:10.1002/anie.201603331

408. Dang H, Dong X, Dong Y et al (2014) Enhancing the photocatalytic H2 evolution activity ofred phosphorous by using noble-metal-free Ni(OH)2 under photoexcitation up to 700 nm.RSC Adv 4:44823–44826. doi:10.1039/c4ra06867j

409. Gao Y, Wang Y, Wang Y (2007) Photocatalytic hydrogen evolution from water on SiCunder visible light irradiation. React Kin Catal Lett 91:13–19. doi:10.1007/s11144-007-5064-x

410. Wang M, Chen J, Liao X et al (2014) Highly efficient photocatalytic hydrogen production ofplatinum nanoparticle-decorated SiC nanowires under simulated sunlight irradiation. Inter JHydrogen En 39:14581–14587. doi:10.1016/j.ijhydene.2014.07.068

411. Wang D, Guo Z, Peng Y, Yuan W (2015) Visible light induced photocatalytic overall watersplitting over micro-SiC driven by the Z-scheme system. Catal Commun 61:53–56. doi:10.1016/j.catcom.2014.12.008

412. Wang Y, Guo X, Dong L et al (2013) Enhanced photocatalytic performance of chemicallybonded SiC-graphene composites for visible-light-driven overall water splitting. Inter JHydrogen En 38:12733–12738. doi:10.1016/j.ijhydene.2013.07.062

413. Liu H, She G, Mu L, Shi W (2012) Porous SiC nanowire arrays as stable photocatalyst forwater splitting under UV irradiation. Mater Res Bull 47:917–920. doi:10.1016/j.materresbull.2011.12.046

References 115

414. Wang B, Wang Y, Lei Y et al (2016) Mesoporous silicon carbide nanofibers with in situembedded carbon for co-catalyst free photocatalytic hydrogen production. Nano Res 9:886–898. doi:10.1007/s12274-015-0971-z

415. Kida T, Minami Y, Guan G et al (2006) Photocatalytic activity of gallium nitride forproducing hydrogen from water under light irradiation. J Mater Sci 41:3527–3534. doi:10.1007/s10853-005-5655-8

416. Yoshimizu M, Kobayashi R, Saegusa M et al (2015) Photocatalytic hydrogen evolution overb-iron silicide under infrared-light irradiation. Chem Commun 51:2818–2820. doi:10.1039/C4CC08093A

417. Yoneyama H, Matsumoto N, Tamura H (1986) Photocatalytic decomposition of formic acidon platinized n-type silicon powder in aqueous solution. Bull Chem Soc Jpn 59:3302–3304.doi:10.1246/bcsj.59.3302

418. Jang YJ, Ryu J, Hong D et al (2016) A multi-stacked hyperporous silicon flake for highlyactive solar hydrogen production. Chem Commun 52:10221–10224. doi:10.1039/c6cc04775k

419. Zhang H, Li A, Wang Z et al (2016) Decorating mesoporous silicon with amorphous metal–phosphorous-derived nanocatalysts towards enhanced photoelectrochemical water reduction.J Mater Chem A 4:14960–14967. doi:10.1039/c6ta05725j

420. Lv C, Chen Z, Chen Z et al (2015) Silicon nanowires loaded with iron phosphide foreffective solar-driven hydrogen production. J Mater Chem A 3:17669–17675. doi:10.1039/c5ta03438h

421. Li S, Wang H, Li D et al (2016) Siloxene nanosheets: a metal-free semiconductor for watersplitting. J Mater Chem A 4:15841–15844. doi:10.1039/c6ta07545b

422. Mou Z, Yin S, Zhu M et al (2013) RuO2/TiSi2/graphene composite for enhancedphotocatalytic hydrogen generation under visible light irradiation. Phys Chem Chem Phys15:2793–2799. doi:10.1039/c2cp44270a

423. Wu W, Zhan L, Ohkubo K et al (2015) Photocatalytic H2 evolution from NADH with carbonquantum dots/Pt and 2-phenyl-4-(1-naphthyl)quinolinium ion. J Photochem Photobiol, B152:63–70. doi:10.1016/j.jphotobiol.2014.10.018

424. Liu Q, Chen T, Guo Y et al (2016) Ultrathin g-C3N4 nanosheets coupled with carbonnanodots as 2D/0D composites for efficient photocatalytic H2 evolution. Appl Catal B193:248–258. doi:10.1016/j.apcatb.2016.04.034

425. Yang P, Zhao J, Wang J et al (2015) Pure carbon nanodots for excellent photocatalytichydrogen generation. RSC Adv 5:21332–21335. doi:10.1039/c5ra01924a

426. Ming H, Ma Z, Liu Y et al (2012) Large scale electrochemical synthesis of high qualitycarbon nanodots and their photocatalytic property. Dalton Trans 41:9526–9531. doi:10.1039/C2DT30985H

427. Zhao S, Li C, Wang L et al (2016) Carbon quantum dots modified MoS2 withvisible-light-induced high hydrogen evolution catalytic ability. Carbon 99:599–606.doi:10.1016/j.carbon.2015.12.088

428. Wang J, Ng YH, Lim YF, Ho GW (2014) Vegetable-extracted carbon dots and theirnanocomposites for enhanced photocatalytic H2 production. RSC Adv 4:44117–44123.doi:10.1039/c4ra07290a

429. Thomas A, Fischer A, Goettmann F et al (2008) Graphitic carbon nitride materials: variationof structure and morphology and their use as metal-free catalysts. J Mater Chem 18:4893–4908. doi:10.1039/b800274f

430. Zhou L, Zhang H, Sun H et al (2016) Recent advances in non-metal modification ofgraphitic carbon nitride for photocatalysis: a historic review. Catal Sci Technol 6:002–7023.doi:10.1039/c6cy01195k

431. Wen J, Xie J, Chen X, Li X (2017) A review on g-C3N4-based photocatalysts. Appl Surf Sci391:72–123. doi:10.1016/j.apsusc.2016.07.030

432. Zheng Y, Lin L, Wang B, Wang X (2015) Graphitic carbon nitride polymers towardsustainable photoredox catalysis. Angew Chem Int Ed 54:12868–12884. doi:10.1002/anie.201501788

116 2 Semiconductor-Based Photocatalytic Systems …

433. Wang X, Blechert S, Antonietti M (2012) Polymeric graphitic carbon nitride forheterogeneous photocatalysis. ACS Catal 2:1596–1606. doi:10.1021/cs300240x

434. Zhu J, Xiao P, Li H, Carabineiro SA (2014) Graphitic carbon nitride: synthesis, properties,and applications in catalysis. ACS Appl Mater Interfaces 6:16449–16465. doi:10.1021/am502925j

435. Liu J, Wang H, Antonietti M (2016) Graphitic carbon nitride “reloaded”: emergingapplications beyond (photo)catalysis. Chem Soc Rev 45:2308–2326. doi:10.1039/C5CS00767D

436. Patnaik S, Martha S, Parida KM (2016) An overview of the structural, textural andmorphological modulations of g-C3N4 towards photocatalytic hydrogen production. RSCAdv 6:46929–46951. doi:10.1039/c5ra26702a

437. Zhang G, Wang X (2013) A facile synthesis of covalent carbon nitride photocatalysts byCo-polymerization of urea and phenylurea for hydrogen evolution. J Catal 307:246–253.doi:10.1016/j.jcat.2013.07.026

438. Chuang PK, Wu KH, Yeh TF, Teng H (2016) Extending the p-Conjugation of g-C3N4 byIncorporating Aromatic Carbon for Photocatalytic H2 Evolution from Aqueous Solution.ACS Sustainable Chem Eng 4:5989–5997. doi:10.1021/acssuschemeng.6b01266

439. Fan X, Zhang L, Cheng R et al (2015) Construction of graphitic C3N4-based intramoleculardonor − acceptor conjugated copolymers for photocatalytic hydrogen evolution. ACS Catal5:5008–5015. doi:10.1021/acscatal.5b01155

440. Cao S, Yu J (2014) g-C3N4-based photocatalysts for hydrogen generation. J Phys Chem Lett5:2101–2107. doi:10.1021/jz500546b

441. Wang X, Maeda K, Chen X et al (2009) Polymer semiconductors for Artificialphotosynthesis: hydrogen evolution by mesoporous graphitic carbon nitride with visiblelight. J Am Chem Soc 131:1680–1681. doi:10.1021/ja809307s

442. Zheng D, Zhang G, Hou Y, Wang X (2016) Layering MoS2 on soft hollow g-C3N4

nanostructures for photocatalytic hydrogen evolution. Appl Catal A 521:2–8. doi:10.1016/j.apcata.2015.10.037

443. Liang Q, Li Z, Yu X et al (2015) Macroscopic 3D porous graphitic carbon nitride monolithfor enhanced photocatalytic hydrogen evolution. Adv Mater 27:4634–4639. doi:10.1002/adma.201502057

444. He F, Chen G, Zhou Y et al (2015) The facile synthesis of mesoporous g-C3N4 with highlyenhanced photocatalytic H2 evolution performance. Chem Commun 51:16244–16246.doi:10.1039/c5cc06713h

445. Shen S, Zhao D, Chen J et al (2016) Enhanced photocatalytic hydrogen evolution overgraphitic carbon nitride modified with Ti-activated mesoporous silica. Appl Catal A521:111–117. doi:10.1016/j.apcata.2015.11.004

446. Qiao S, Mitchell RW, Coulson B et al (2016) Pore confinement effects and stabilization ofcarbon nitride oligomers in macroporous silica for photocatalytic hydrogen production.Carbon 106:320–329. doi:10.1016/j.carbon.2016.05.039

447. Chen X, Jun YS, Takanabe K et al (2009) Ordered mesoporous SBA-15 type graphiticcarbon nitride: a semiconductor host structure for photocatalytic hydrogen evolution withvisible light. Chem Mater 21:4093–4095. doi:10.1021/cm902130z

448. Xu L, Jin B, Zhang J et al (2016) Efficient hydrogen generation from formic acid using AgPdnanoparticles immobilized on carbon nitride-functionalized SBA-15. RSC Adv 6:46908–46914. doi:10.1039/c6ra06071d

449. Cheng R, Fan X, Wang M et al (2016) Facile construction of CuFe2O4/g-C3N4 photocatalystfor enhanced visible-light hydrogen evolution. RSC Adv 6:18990–18995. doi:10.1039/c5ra27221a

450. Tong J, Zhang L, Li F et al (2015) Rapid and high-yield production of g-C3N4 nanosheetsvia chemical exfoliation for photocatalytic H2 evolution. RSC Adv 5:88149–88153. doi:10.1039/c5ra16988g

References 117

451. Han Q, Wang B, Gao J et al (2016) Atomically thin mesoporous nanomesh of graphiticC3N4 for high-effciency photocatalytic hydrogen evolution. ACS Nano 10:2745–2751.doi:10.1021/acsnano.5b07831

452. Fan M, Song C, Chen T et al (2016) Visible-light-drived high photocatalytic activities of Cu/g-C3N4 photocatalysts for hydrogen production. RSC Adv 6:34633–34640. doi:10.1039/c5ra27755h

453. Bi L, Meng D, Bu Q et al (2016) Electron acceptor of Ni decorated porous carbon nitrideapplied in photocatalytic hydrogen production. Phys Chem Chem Phys 18:31534–31541.doi:10.1039/c6cp05618k

454. Xu L, Liu N, Hong B et al (2016) Nickel–platinum nanoparticles immobilized on graphiticcarbon nitride as highly efficient catalyst for hydrogen release from hydrous hydrazine. RSCAdv 6:31687–31691. doi:10.1039/c6ra01335j

455. Ge L, Han C, Xiao X, Guo L (2013) Synthesis and characterization of composite visiblelight active photocatalysts MoS2–g-C3N4 with enhanced hydrogen evolution activity. Inter JHydrogen En 38:6960–6969. doi:10.1016/j.ijhydene.2013.04.006

456. Jin X, Fan X, Tian J et al (2016) MoS2 quantum dot decorated g-C3N4 compositephotocatalyst with enhanced hydrogen evolution performance. RSC Adv 6:52611–52619.doi:10.1039/c6ra07060d

457. Sun Q, Wang P, Yu H, Wang X (2016) In situ hydrothermal synthesis and enhancedphotocatalytic H2-evolution performance of suspended rGO/g-C3N4 photocatalysts. J MolCatal A 424:369–376. doi:10.1016/j.molcata.2016.09.015

458. Suryawanshi A, Dhanasekaran P, Mhamane D et al (2012) Ioubling of photocatalytic H2

evolution from g-C3N4 via its nanocomposite formation with multiwall carbon nanotubes:electronic and morphological effects. Inter J Hydrogen En 37:9584–9589. doi:10.1016/j.ijhydene.2012.03.123

459. Ge L, Han C (2012) Synthesis of MWNTs/g-C3N4 composite photocatalysts with efficientvisible light photocatalytic hydrogen evolution activity. Appl Catal B 117–118:268–274.doi:10.1016/j.apcatb.2012.01.021

460. Wang N, Li J, Wu L et al (2016) MnO2 and carbon nanotube co-modified C3N4 compositecatalyst for enhanced water splitting activity under visible light irradiation. Inter J HydrogenEn 41:22743–22750. doi:10.1016/j.ijhydene.2016.10.068

461. Gao LF, Wen T, Xu JY et al (2016) Iron-doped carbon nitride-type polymers ashomogeneous organocatalysts for visible light-driven hydrogen evolution. ACS Appl MaterInterfaces 8:617–624. doi:10.1021/acsami.5b09684

462. Min S, Lu G (2012) Enhanced electron transfer from the excited Eosin Y to mpg-C3N4 forhighly efficient hydrogen evolution under 550 nm irradiation. J Phys Chem C 116:19644–19652. doi:10.1021/jp304022f

463. Zhang H, Li S, Lu R, Yu A (2015) Time-resolved study on xanthene dye-sensitized carbonnitride photocatalytic systems. ACS Appl Mater Interfaces 7:21868–21874. doi:10.1021/acsami.5b06309

464. Mori K, Itoh T, Kakudo H et al (2015) Nickel-supported carbon nitride photo-catalystcombined with organic dye for visible-light-driven hydrogen evolution from water. PhysChem Chem Phys 17:24086–24091. doi:10.1039/c5cp04493f

465. Cao SW, Yuan YP, Fang J et al (2013) In-situ growth of CdS quantum dots on g-C3N4

nanosheets for highly efficient photocatalytic hydrogen generation under visible lightirradiation. Inter J Hydrogen En 38:1258–1266. doi:10.1016/j.ijhydene.2012.10.116

466. Zhang Z, Zhang Y, Lu L et al (2017) Graphitic carbon nitride nanosheet for photo-catalytichydrogen oduction: the impact of morphology and element composition. Appl Surf Sci391:369–375. doi:10.1016/j.apsusc.2016.05.174

467. Yang S, Gong Y, Zhang J et al (2013) Exfoliated graphitic carbon nitride nanosheets asefficient catalysts for hydrogen evolution under visible light. Adv Mater 25:2452–2456.doi:10.1002/adma.201204453

118 2 Semiconductor-Based Photocatalytic Systems …

468. Xu J, Zhang L, Shi R, Zhu Y (2013) Chemical exfoliation of graphitic carbon nitride forefficient heterogeneous photocatalysis. J Mater Chem A 1:14766–14772. doi:10.1039/C3TA13188B

469. Bu X, Bu Y, Yang S et al (2016) Graphitic carbon nitride nanoribbon for enhancedvisible-light photocatalytic H2 production. RSC Adv 6:112210–112214. doi:10.1039/c6ra23218c

470. Schwinghammer K, Mesch MB, Duppel V (2014) Crystalline carbon nitride nanosheets forimproved visible-light hydrogen evolution. J Am Chem Soc 136:1730–1733. doi:10.1021/ja411321s

471. Li L, Fang W, Zhang P et al (2016) Sulfur-doped covalent triazine-based frameworks forenhanced photocatalytic hydrogen evolution from water under visible light. J Mater Chem A4:12402–12406. doi:10.1039/c6ta04711d

472. Lin L, Ou H, Zhang Y, Wang X (2016) Tri-s-triazine-based crystalline graphitic carbonnitrides for highly efficient hydrogen evolution photocatalysis. ACS Catal 6:3921–3931.doi:10.1021/acscatal.6b00922

473. Bhunia MK, Melissen S, Parida MR (2015) Dendritic tip-on polytriazine-based carbonnitride photocatalyst with high hydrogen evolution activity. Chem Mater 27:8237–8247.doi:10.1021/acs.chemmater.5b02974

474. Liu H, Chen D, Wang Z et al (2017) Microwave-assisted molten-salt rapid synthesis ofisotype triazine-/heptazine based g-C3N4 heterojunctions with highly enhanced photocat-alytic hydrogen evolution performance. Appl Catal B 203:300–313. doi:10.1016/j.apcatb.2016.10.014

475. Li Y, Xu H, Ouyang S, Ye J (2016) Metal–organic frameworks for photocatalysis. PhysChem Chem Phys 18:7563–7572. doi:10.1039/c5cp05885f

476. So MC, Wiederrecht GP, Mondloch JE et al (2015) Metal–organic framework materials forlight-harvesting and energy transfer. Chem Commun 51:3501–3510. doi:10.1039/c4cc09596k

477. Xiao JD, Shang Q, Xiong Y et al (2016) Boosting photocatalytic hydrogen production of ametal-organic framework decorated with platinum nanoparticles: the platinum locationmatters. Angew Chem Int Ed 55:9389–9393. doi:10.1002/anie.201603990

478. Zhou JJ, Wang R, Liu XL et al (2015) In situ growth of CdS nanoparticles on UiO-66metal-organic framework octahedrons for enhanced photocatalytic hydrogen productionunder visible light irradiation. Appl Surf Sci 346:278–283. doi:10.1016/j.apsusc.2015.03.210

479. Lin R, Shen L, Ren Z et al (2014) Enhanced photocatalytic hydrogen production activity viadual modification of MOF and reduced graphene oxide on CdS. Chem Commun 50:8533–8535. doi:10.1039/c4cc01776e

480. He J, Wang J, Chen Y et al (2014) A dye-sensitized Pt@UiO-66(Zr) metal–organicframework for visible-light photocatalytic hydrogen production. Chem Commun 50:7063–7066. doi:10.1039/c4cc01086h

481. Su Y, Zhang Z, Liu H, Wang Y (2017) Cd0.2Zn0.8S@UiO-66-NH2 nanocomposites asefficient and stable visible-light-driven photocatalyst for H2 evolution and CO2 reduction.Appl Catal B 200:448–457. doi:10.1016/j.apcatb.2016.07.032

482. Sasan K, Lin Q, Mao CY, Feng P (2014) Incorporation of iron hydrogenase active sites intoa highly stable metal–organic framework for photocatalytic hydrogen generation. ChemCommun 50:10390–10393. doi:10.1039/c4cc03946g

483. Ragon F, Campo B, Yang Q et al (2015) Acid-functionalized UiO-66(Zr) MOFs and theirevolution after intra-framework cross-linking: structural features and sorption properties.J Mater Chem A 3:3294–3309. doi:10.1039/C4TA03992K

484. Wang D, Song Y, Cai J et al (2016) Effective photo-reduction to deposit Pt nanoparticles onMIL-100(Fe) for visible-light-induced hydrogen evolution. New J Chem 40:9170–9175.doi:10.1039/c6nj01989g

References 119

485. Vaddipalli SR, Sanivarapu SR, Vengatesan S (2016) Heterostructured Au NPs/CdS/LaBTCMOFs photoanode for efficient photoelectrochemical water split-ting: stability enhancementvia CdSe QDs to 2D-CdS nanosheets transformation. ACS Appl Mater Interfaces 8:23049–23059. doi:10.1021/acsami.6b06851

486. Sun X, Yu Q, Zhang F et al (2016) A dye-like ligand-based metal—organic framework forefficient photocatalytic hydrogen production from aqueous solution. Catal Sci Technol6:3840–3844. doi:10.1039/c5cy01716e

487. Dong XY, Zhang M, Pei RB et al (2016) A crystalline copper(II) coordination polymer forthe efficient visible-light-driven generation of hydrogen. Angew Chem Int Ed 55:2073–2077. doi:10.1002/anie.201509744

488. Xu JY, Zhai XP, Gao LF et al (2016) In situ preparation of a MOF-derived magneticcarbonaceous catalyst for visible-light-driven hydrogen evolution. RSC Adv 6:2011–2018.doi:10.1039/c5ra23838b

489. Nellist MR, Laskowski FAL, Lin F, Mills TJ, Boettcher SW (2016) Semiconductor-electrocatalyst interfaces: theory, experiment, and applications in photoelectrochemicalwater splitting. Acc Chem Res 49:733–740. doi:10.1021/acs.accounts.6b00001

490. Orlandi M, Dalle Carbonare N, Caramori S, Bignozzi CA, Berardi S, Mazzi A, El Koura Z,Bazzanella N, Patel N, Miotello A (2016) Porous versus compact nanosized Fe(III)-basedwater oxidation catalyst for photoanodes functionalization. ACS Appl Mater Interfaces8:20003–20011. doi:10.1021/acsami.6b05135

491. Ronconi F, Syrgiannis Z, Bonasera A, Prato M, Argazzi R, Caramori S, Cristino V,Bignozzi CA (2015) Modification of nanocrystalline WO3 with a dicationic perylenebisimide: applications to molecular level solar water splitting. J Am Chem Soc 137:4630–4633. doi:10.1021/jacs.5b01519

492. Wang J, Li B, Chen J et al (2012) Enhanced photocatalytic H2-production activity of CdxZn1−xSnanocrystals by surface loading MS (M = Ni Co, Cu) species. Appl Surf Sci 259:118–123.doi:10.1016/j.apsusc.2012.07.003

493. Zong X, Yan H, Wu G et al (2008) Enhancement of photocatalytic H2 evolution on CdS byloading MoS2 as cocatalyst under visible light irradiation. J Am Chem Soc 130:7176–7177.doi:10.1021/ja8007825

494. Ji J, Guo L, Li Q et al (2015) A bifunctional catalyst for hydrogen evolution reaction: theinteractive influences between CdS and MoS2 on photoelectro-chemical activity. Inter JHydrogen En 40:3813–3821. doi:10.1016/j.ijhydene.2015.01.075

495. Jia T, Kolpin A, Ma C et al (2014) A graphene dispersed CdS–MoS2 nanocrystal ensemblefor cooperative photocatalytic hydrogen production from water. Chem Commun 50:1185–1188. doi:10.1039/C3CC47301E

496. Lu Y, Wang D, Yang P et al (2014) Coupling ZnxCd1−xS nanoparticles with graphene-likeMoS2: superior interfacial contact, low overpotential and enhanced photocatalytic activityunder visible-light irradiation. Catal Sci Technol 4:2650–2657. doi:10.1039/C4CY00331D

497. Zhu B, Lin B, Zhou Y et al (2014) Enhanced photocatalytic H2 evolution on ZnS loadedwith graphene and MoS2 nanosheets as cocatalysts. J Mater Chem A 2:3819–3827. doi:10.1039/C3TA14819J

498. Zhao H, Dong Y, Jiang P et al (2015) In situ light-assisted preparation of MoS2 on graphiticC3N4 nanosheets for enhanced photocatalytic H2 production from water. J Mater Chem A3:7375–7381. doi:10.1039/C5TA00402K

499. Laursen AB, Kegnaes S, Dahl S, Chorkendorff I (2012) Molybdenum sulfides—efficient andviable materials for electro- and photoelectrocatalytic hydrogen evolution. En Environ Sci5:5577–5591. doi:10.1039/C2EE02618J

500. Liu M, Li F, Sun Z et al (2014) Noble-metal-free photocatalysts MoS2–graphene/CdS mixednanoparticles/nanorods morphology with high visible light efficiency for H2 evolution.Chem Commun 50:11004–11007. doi:10.1039/C4CC04653F

501. Redman DW, Kim HJ, Stevenson KJ, Rose MJ (2016) Photo-assisted electrodeposition ofMoSx from ionic liquids on organic-functionalized silicon photoelectrodes for H2 generation.J Mater Chem A 4:7027–7035. doi:10.1039/C5TA09684G

120 2 Semiconductor-Based Photocatalytic Systems …

502. Kumar DP, Hong S, Reddy DA, Kim TK (2016) Noble metal-free ultrathin MoS2nanosheet-decorated CdS nanorods as an efficient photocatalyst for spectacular hydrogenevolution under solar light irradiation. J Mater Chem A 4:18551–18558. doi:10.1039/c6ta08628d

503. Du P, Zhu Y, Zhng J et al (2016) Metallic 1T phase MoS2 nanosheets as a highly efficientco-catalyst for the photocatalytic hydrogen evolution of CdS nanorods. RSC Adv 6:74394–74399. doi:10.1039/c6ra10170d

504. Sasikala R, Gaikwad AP, Jayakumar OD et al (2015) Nanohybrid MoS2-PANI-CdSphotocatalyst for hydrogen evolution from water. Colloids Surf A 481:485–492. doi:10.1016/j.colsurfa.2015.06.027

505. Du H, Xie X, Zhu Q et al (2015) Metallic MoO2 cocatalyst significantly enhancesvisible-light photocatalytic hydrogen production over MoO2/Zn0.5Cd0.5S hetero-junction.Nanoscale 7:5752–5759. doi:10.1039/C4NR06949H

506. Ma B, Liu Y, Li J et al (2016) Mo2N: An efficient non-noble metal cocatalyst on CdS forenhanced photocatalytic H2 evolution under visible light irradiation. Inter J Hydrogen En41:22009–22016. doi:10.1016/j.ijhydene.2016.08.133

507. Khan Z, Khannam M, Vinothkumar N et al (2012) Hierarchical 3D NiO–CdS heteroar-chitecture for efficient visible light photocatalytic hydrogen generation. J Mater Chem22:12090–12095. doi:10.1039/C2JM31148H

508. Lin CY, Lai YH, Mersch D, Reisner E (2012) Cu2O|NiOx nanocomposite as an inexpensivephotocathode in photoelectrochemical water splitting. Chem Sci 3:3482–3487. doi:10.1039/C2SC20874A

509. Bi G, Wen J, Li X et al (2016) Efficient visible-light photocatalytic H2 evolution overmetal-free g-C3N4 co-modified with robust acetylene black and Ni(OH)2 as dualco-catalysts. RSC Adv 6:31497–31506. doi:10.1039/c6ra03118h

510. Yuan YP, Cao SW, Yin LS et al (2013) NiS2 Co-catalyst decoration on CdLa2S4nanocrystals for efficient photocatalytic hydrogen generation under visible light irradiation.Inter J Hydrogen En 38:7218–7223. doi:10.1016/j.ijhydene.2013.03.169

511. Parida KM, Biswal N, Das DP, Martha S (2010) Visible light response photocatalytic watersplitting over CdS-pillared zirconium–titanium phosphate (ZTP). Inter J Hydrogen En35:5262–5269. doi:10.1016/j.ijhydene.2010.03.017

512. Sun Z, Chen H, Zhang L et al (2016) Enhanced photocatalytic H2 production on cadmiumsulfide photocatalysts using nickel nitride as a novel cocatalyst. J Mater Chem A 4:13289–13295. doi:10.1039/c6ta04696g

513. Cao S, Chen Y, Wang CJ et al (2014) Highly efficient photocatalytic hydrogen evolution bynickel phosphide nanoparticles from aqueous solution. Chem Commun 50:10427–10429.doi:10.1039/C4CC05026F

514. Yang JS, Ham DJ, Lakshminarasimhan N et al (2008) Role of platinum-like tungsten carbideas cocatalyst of CdS photocatalyst for hydrogen production under visible light irradiation.Appl Catal A 346:149–154. doi:10.1016/j.apcata.2008.05.020

515. Yan Z, Wu H, Han A et al (2014) Noble metal-free cobalt oxide (CoOx) nano-particlesloaded on titanium dioxide/cadmium sulfide composite for enhanced photocatalytichydrogen production from water. Inter J Hydrogen En 39:13353–13360. doi:10.1016/j.ijhydene.2014.04.121

516. Yuan J, Wen J, Gao Q et al (2015) Amorphous Co3O4 modified CdS nanorods withenhanced visible-light photocatalytic H2-production activity. Dalton Trans 44:1680–1689.doi:10.1039/C4DT03197K

517. Cao S, Chen Y, Hou CC et al (2015) Cobalt phosphide as a highly active non-precious metalcocatalyst for photocatalytic hydrogen production under visible light irradiation. J MaterChem A 3:6096–6101. doi:10.1039/C4TA07149B

518. Yuan YJ, Ye ZJ, Lu HW et al (2016) Constructing anatase TiO2 nanosheets with exposed(001) facets/layered MoS2 two-dimensional nanojunctions for enhanced solar hydrogengeneration. ACS Catal 6:532–541. doi:10.1021/acscatal.5b02036

References 121

519. Cao S, Chen Y, Kang L et al (2015) Enhanced photocatalytic H2 evolution by immobilizingCdS nanocrystals on ultrathin Co0.85Se/RGO—PEI nanosheets. J Mater Chem A 3:18711–18717. doi:10.1039/c5ta04910e

520. Zhang L, Jiang T, Li S et al (2013) Enhancement of photocatalytic H2 evolution onZn0.8Cd0.2S loaded with CuS as cocatalyst and its photogenerated charge transfer properties.Dalton Trans 42:12998–13003. doi:10.1039/C3DT51256H

521. Zhang L, Liu YN, Zhou M, Yan J (2013) Improving photocatalytic hydrogen evolution overCuO/Al2O3 by platinum-depositing and CuS-loading. Appl Surf Sci 282:531–537. doi:10.1016/j.apsusc.2013.06.006

522. Wang X, Liu M, Chen Q et al (2013) Synthesis of CdS/CNTs photocatalysts and study ofhydrogen production by photocatalytic water splitting. Inter J Hydrogen En 38:13091–13096. doi:10.1016/j.ijhydene.2013.03.016

523. Zhang J, Qiao SZ, Qi L, Yu J (2013) Fabrication of NiS modified CdS nanorod p–n junctionphotocatalysts with enhanced visible-light photocatalytic H2-production activity. Phys ChemChem Phys 15:12088–12094. doi:10.1039/C3CP50734C

524. Li N, Zhou B, Guo P et al (2013) Fabrication of noble-metal-free Cd0.5Zn0.5S/NiS hybridphotocatalyst for efficient solar hydrogen evolution. Inter J Hydrogen En 38:11268–11277.doi:10.1016/j.ijhydene.2013.06.067

525. Li Y, Lin S, Peng S et al (2013) Modification of ZnS1−x−0.5yOx(OH)y–ZnO photocatalystwith NiS for enhanced visible-light-driven hydrogen generation from seawater.Inter J. Hydrogen En 38:15976–15984. doi:10.1016/j.ijhydene.2013.09.149

526. Zhu Y, Xu Y, Hou Y et al (2014) Cobalt sulfide modified graphitic carbon nitridesemiconductor for solar hydrogen production. Inter J Hydrogen En 39:11873–11879. doi:10.1016/j.ijhydene.2014.06.025

527. Chen Y, Qin Z (2016) General applicability of nanocrystalline Ni2P as a noble-metal-freecocatalyst to boost photocatalytic hydrogen generation. Catal Sci Technol 6:8212–8221.doi:10.1039/c6cy01653g

528. Yasomanee JP, Bandara J (2008) Multi-electron storage of photoenergy using Cu2O–TiO2

thin film photocatalyst. Solar En Mater Solar Cells 92:348–352. doi:10.1016/j.solmat.2007.09.016

529. Jian JX, Ye C, Wang XZ et al (2016) Comparison of H2 photogeneration by [FeFe]-hydrogenase mimics with CdSe QDs and Ru(bpy)3Cl2 in aqueous solution. Energy EnvironSci 9:2083–2089. doi:10.1039/c6ee00629a

530. Wen F, Li C (2013) Hybrid artificial photosynthetic systems comprising semicon-ductors aslight harvesters and biomimetic complexes as molecular cocatalysts. Acc Chem Res 4:2355–2364. doi:10.1021/ar300224u

531. Yeh TF, Cihlar J, Chang CY et al (2013) Roles of graphene oxide in photocatalytic watersplitting. Mater Today 16:78–84. doi:10.1016/j.mattod.2013.03.006

532. Yang Y, Liu E, Dai H et al (2014) Photocatalytic activity of Ag-TiO2-graphene ternarynanocomposites and application in. hydrogen evolution by water splitting. Inter J HydrogenEn 39:7664–7671. doi:10.1016/j.ijhydene.2013.09.109

533. Fang Z, Wang Y, Song J et al (2013) Immobilizing CdS quantum dots and dendritic Ptnanocrystals on thiolated graphene nanosheets toward highly efficient photocatalytic H2

evolution. Nanoscale 5:9830–9838. doi:10.1039/c3nr03043a534. Xiang Q, Yu J, Jaroniec M (2011) Enhanced photocatalytic H2-production activity of

graphene-modified titania nanosheets. Nanoscale 3:3670–3678. doi:10.1039/C1NR10610D535. Hong Y, Shi P, Wang P, Yao W (2015) Improved photocatalytic activity of CdS/reduced

graphene oxide (RGO) for H2 evolution by strengthening the connection between CdS andRGO sheets. Inter J Hydrogen En 40:7045–7051. doi:10.1016/j.ijhydene.2015.04.005

536. Yu X, Du R, Li B et al (2016) Biomolecule-assisted self-assembly of CdS/MoS2/graphenehollow spheres as high-efficiency photocatalysts for hydrogen evolution without noblemetals. Appl Catal B 182:504–512. doi:10.1016/j.apcatb.2015.09.003

122 2 Semiconductor-Based Photocatalytic Systems …

537. Xu H, Li X, Kang S et al (2014) Noble metal-free cuprous oxide/reduced graphene oxide forenhanced photocatalytic hydrogen evolution from water reduction. Inter J Hydrogen En39:11578–11582. doi:10.1016/j.ijhydene.2014.05.156

538. Ding J, Yan W, Sun S et al (2014) Fabrication of graphene/CaIn2O4 composites withenhanced photocatalytic activity from water under visible light irradiation. Inter J HydrogenEn 39:119–126. doi:10.1016/j.ijhydene.2013.10.077

539. Pan B, Wang Y, Liang Y et al (2014) Nanocomposite of BiPO4 and reduced graphene oxideas an efficient photocatalyst for hydrogen evolution. Inter J Hydrogen En 39:13527–13533.doi:10.1016/j.ijhydene.2014.02.031

540. Sun Z, Guo J, Zhu S et al (2014) A high-performance Bi2WO6—graphene photocatalyst forvisible light-induced H2 and O2 generation. Nanoscale 6:2186–2193. doi:10.1039/c3nr05249d

541. Hong Z, Li X, Kang S et al (2014) Enhanced photocatalytic activity and stability of thereduced graphene oxide loaded potassium niobate microspheres for hydrogen productionfrom water reduction. Inter J Hydrogen En 39:12515–12523. doi:10.1016/j.ijhydene.2014.06.075

542. Alexander BD, Kulesza PJ, Rutkowska I et al (2008) Metal oxide photoanodes for solarhydrogen production. J Mater Chem 18:2298–2303. doi:10.1039/B718644D

543. Kudo A, Kato H, Tsuji I (2004) Strategies for the development of visible-light-drivenphotocatalysts for water splitting. Chem Lett 33:1534–1539. doi:10.1246/cl.2004.1534

544. Sayama K, Mukasa K, Abe R et al (2002) A new photocatalytic water splitting system undervisible light irradiation mimicking a Z-scheme mechanism in photosynthesis. J PhotochemPhotobiol A 148:71–77. doi:10.1016/S1010-6030(02)00070-9

545. Sayama K, Mukasa K, Abe R et al (2001) Stoichiometric water splitting into H2 and O2

using a mixture of two different photocatalysts and an IO�3

�I� shuttle redox mediator under

visible light irradiation. Chem. Commun 2416–2417. doi:10.1039/B107673F546. Schürch D, Currao A, Sarkar S et al (2002) The silver chloride photoanode in photoelectro–

chemical water splitting. J Phys Chem B 106:12764–12775. doi:10.1021/jp0265081547. Abe R, Sayama K, Sugihara H (2005) Development of new photocatalytic water splitting

into H2 and O2 using two different semiconductor photocatalysts and a shuttle redoxmediator IO�

3

�I�. J Phys Chem B 109:16052–16061. doi:10.1021/jp052848l

548. Abe R, Takata T, Sugihara H, Domen K (2005) Photocatalytic overall water splitting undervisible light by TaON and WO3 with an IO�

3

�I� shuttle redox mediator. Chem Commun

3829–3831. doi:10.1039/B505646B549. Higashi M, Abe R, Takata T, Domen K (2009) Photocatalytic Overall Water Splitting under

Visible Light Using ATaO2N (A = Ca, Sr, Ba) and WO3 in a IO�3

�I� Shuttle Redox

Mediated System. Chem Mater 21:1543–1549. doi:10.1021/cm803145n550. Higashi M, Abe R, Ishikawa A et al (2008) Z-scheme overall water splitting on

modified-TaON photocatalysts under visible light (k < 500 nm). Chem Lett 37:138–139551. Abe R, Shinmei K, Koumura N et al (2013) Visible-light-induced water splitting based on

two-step photoexcitation between dye-sensitized layered niobate and tungsten oxidephotocatalysts in the presence of a triiodide/iodide shuttle redox mediator. J Am ChemSoc 135:16872–16884. doi:10.1021/ja4048637

552. Jia Q, Iwase A, Kudo A (2014) BiVO4–Ru/SrTiO3: Rh composite Z-scheme photo-catalystfor solar water splitting. Chem Sci 5:1513–1519. doi:10.1039/C3SC52810C

553. Maeda K, Lu D, Domen K (2013) Solar-driven Z-scheme water splitting using modifiedBaZrO3–BaTaO2N solid solutions as photocatalysts. ACS Catal 3:1026–1033. doi:10.1021/cs400156m

554. Kato H, Hori M, Konta R et al (2004) Construction of Z-scheme type heterogeneousphotocatalysis systems for water splitting into H2 and O2 under visible light irradiation.Chem Lett 33:1348–1349. doi:10.1246/cl.2004.1348

References 123

555. Kato H, Sasaki Y, Iwase A, Kudo A (2007) Role of iron ion electron mediator onphotocatalytic overall water splitting under visible light irradiation using Z-scheme systems.Bull Chem Soc Jpn 80:2457–2464. doi:10.1246/bcsj.80.2457

556. Bae SW, Ji SM, Hong SJ et al (2009) Photocatalytic overall water splitting with dual-bedsystem under visible light irradiation. Inter J Hydrogen En 34:3243–3249. doi:10.1016/j.ijhydene.2009.02.022

557. Pan Z, Hisatomi T, Wang Q et al (2016) Photocatalyst sheets composed of particulateLaMg1/3Ta2/3O2N and Mo-Doped BiVO4 for Z-scheme water splitting under visible light.ACS Catal 6:7188–7196. doi:10.1021/acscatal.6b01561

558. Iwase A, Ng YH, Ishiguro Y et al (2011) Reduced graphene oxide as a solid-state electronmediator in Z-scheme photocatalytic water splitting under visible light. J Am Chem Soc133:11054–11057. doi:10.1021/ja203296z

559. Paulose M, Mor GK, Varghese OK et al (2006) Visible light photoelectrochemical andwater-photoelectrolysis properties of titania nanotube arrays. J Photochem Photobiol A178:8–15. doi:10.1016/j.jphotochem.2005.06.013

560. Khan MA, Akhtar MS, Woo SI, Yang OB (2008) Enhanced photoresponse under visiblelight in Pt ionized TiO2 nanotube for the photocatalytic splitting of water. Catal Commun10:1–5. doi:10.1016/j.catcom.2008.01.018

561. Zou Z, Ye J, Arakawa H (2002) Surface characterization of nanoparticles of NiOx/In0.9Ni0.1TaO4: effects on photocatalytic activity. J Phys Chem B 106:13098–13101. doi:10.1021/jp0216225

562. Jang JS, Kim HG, Reddy VR et al (2005) Photocatalytic water splitting over iron oxidenanoparticles intercalated in HTiNb(Ta)O5 layered compounds. J Catal 231:213–222.doi:10.1016/j.jcat.2005.01.026

563. Wu G, Chen A (2008) Direct growth of F-doped TiO2 particulate thin films with highphotocatalytic activity for environmental applications. J Photochem Photobiol A 195:47–53.doi:10.1016/j.jphotochem.2007.09.005

564. Kitano M, Iyatani K, Tsujimaru K et al (2008) The effect of chemical etching by HF solutionon the photocatalytic activity of visible light-responsive TiO2 thin films for solar watersplitting. Topics Catal 49:24–31. doi:10.1007/s11244-008-9064-5

565. Maeda K, Takata T, Hara M et al (2005) GaN:ZnO solid solution as a photocatalyst forvisible-light-driven overall water splitting. J Am Chem Soc 127:8286–8287. doi:10.1021/ja0518777

566. Liu H, Yuan J, Shangguang W, Teraoka Y (2008) Visible-light-responding BiYWO6 solidsolution for stoichiometric photocatalytic water splitting. J Phys Chem C 112:8521–8523.doi:10.1021/jp802537u

567. Sayama K, Nomura A, Arai T, Sugita T et al (2006) Photoelectrochemical decomposition ofwater into H2 and O2 on porous BiVO4 thin-film electrodes under visible light andsignificant effect of Ag ion treatment. J Phys Chem B 110:11352–11360. doi:10.1021/jp057539+

568. Bornoz P, Abdi FF, Tilley SD et al (2014) A bismuth vanadate-cuprous oxide tandem cellfor overall solar water splitting. J Phys Chem C 118:16959–16966. doi:10.1021/jp500441h

569. Saremi-Yarahmadi S, Upul Wijayantha KG, Tahir AA, Vaidhyanathan B (2009)Nanostructured a-Fe2O3 electrodes for solar driven water splitting: effect of doping agentson preparation and performance. J Phys Chem C 113:4768–4778. doi:10.1021/jp808453z

570. Saremi-Yarahmadi S, Tahir AA, Vaidhyanathan B, Upul Wijayantha KG (2009) Fabricationof nanostructured a-Fe2O3 electrodes using ferrocene for solar hydrogen generation. MaterLett 63:523–526. doi:10.1016/j.matlet.2008.11.011

571. Youngblood WJ, Lee SHA, Kobayashi Y et al (2009) Photoassisted overall water splitting ina visible light-absorbing dye-sensitized photoelectrochemical cell. J Am Chem Soc131:926–927. doi:10.1021/ja809108y

572. Zou Z, Ye J, Arakawa H (2003) Photocatalytic water splitting into H2 and/or O2 under UVand visible light irradiation with a semiconductor photocatalyst. Inter J Hydrogen En28:663–669. doi:10.1016/S0360-3199(02)00159-3

124 2 Semiconductor-Based Photocatalytic Systems …

573. Oshikiri M, Boero M, Ye J et al (2002) Electronic structures of promising photocatalystsInMO4 (M = V, Nb, Ta) and BiVO4 for water decomposition in the visible wavelengthregion. J Chem Phys 117:7313–7318

574. Zou Z, Ye J, Arakawa H (2001) Photophysical and photocatalytic properties of InMO4

(M = Nb5+, Ta5+) under visible light irradiation. Mater Res Bull 36:1185–1193. doi:10.1016/S0025-5408(01)00607-9

575. Xu J, Pan C, Takata T, Domen K (2015) Photocatalytic overall water splitting on theperovskite-type transition metal oxynitride CaTaO2N under visible light irradiation. ChemCommun 51:7191–7194. doi:10.1039/C5CC01728A

576. Li Y, Jiang S, Xiao J, Li Y (2014) Photocatalytic overall water splitting under visible lightover an In–Ni–Ta–O–N solid solution without an additional cocatalyst. Inter J Hydrogen En39:731–735. doi:10.1016/j.ijhydene.2013.10.092

577. Liu H, Yuan J, Jiang Z et al (2011) Novel photocatalyst of V-based solid solutions foroverall water splitting. J Mater Chem 21:16535–16543. doi:10.1039/C1JM11809A

578. Guo DZ, Zhang GM, Zhang ZX et al (2006) Visible-light-induced water-splitting inchannels of carbon nanotubes. J Phys Chem B 110:1571–1575. doi:10.1021/jp055929q

579. Maeda K, Wang X, Nishihara Y et al (2009) photocatalytic activities of graphitic carbonnitride powder for water reduction and oxidation under visible light. J Phys Chem C113:4940–4947. doi:10.1021/jp809119m

580. Zhao G, Huang X, Fina F et al (2015) Facile structure design based on C3N4 formediator-free Z-scheme water splitting under visible light. Catal Sci Technol 5:3416–3422.doi:10.1039/C5CY00379B

581. Wu X, Zhao J, Guo S et al (2016) Carbon dot and BiVO4 quantum dot composites foroverall water splitting via a two-electron pathway. Nanoscale 8:17314–17321. doi:10.1039/c6nr05864g

582. Haussener S, Hu S, Xiang C, Weber AZ, Lewis NS (2013) Simulations of the irradiation andtemperature dependence of the efficiency of tandem photoelectrochemical water-splittingsystems. Energy Environ Sci 6:3605–3618. doi:10.1039/C3EE41302K

References 125

Chapter 3Semiconductor-Based PhotocatalyticSystems for the Reductive Conversionof CO2 and N2

Semiconductor-based photocatalytic systems aimed at the reduction of carbondioxide and dinitrogen are continuously studied for more than 30 years [1].A gradual shift from micro- to nanocrystalline semiconductor photocatalysts, whichis, probably, the main trend in modern semiconductor photocatalysis/photoelectrochemistry, allowed to achieve attractively high quantum efficienciesof the CO2 and N2 conversion as well as to apply a potent array of spectral methodsfor the elucidation of mechanistic aspects of these important photoreactions.A decrease of the photocatalyst crystal size to a few nanometers allows not only tointensify the photocatalytic synthetic reactions but also to engineer the surface andband structure of the nano-photocatalysts to direct the reactions toward desirableproducts.

In recent years, the photochemical conversion of CO2 got under a renewedspotlight focus because of the global climatic changes induced by theover-abundant anthropogenic CO2 emissions. Also, understanding of the photo-catalytic CO2 transformations on the surface of semiconductor NPs can shed lighton the pre-biotic photosynthesis of simplest organic molecules that could happenbillions of years ago [2, 3]. In those ages, the oceans were saturated with H2S, theEarth atmosphere was of reductive character and the solar irradiation was a lotstronger. Such conditions favored to the formation of colloidal metal sulfide NPsand their participation in the CO2 and N2 reduction to compounds with C–C andC–N bonds—acetates, propionates, ethane, ethanol, urea, amines, etc. [4, 5] thatcould be used as a feedstock for the synthesis of more complex organic compounds.

Molecular nitrogen is very inert in ambient conditions and can be fixed typicallyin biochemical processes occurring in the roots of legumes, some bacteria, as wellas during electric discharges in the atmosphere, thus enriching the soils withwater-soluble nitrogen compounds. Alternatively, N2 can be reduced to ammonia athigh pressures and temperatures over a catalyst in the chemical industry. A possiblesubstitution of this process with new energy-saving ways of N2 fixation in mild(ambient) conditions can have enormous economic and social effect and stimulates

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_3

127

the incessant search for new photochemical processes and photocatalysts that couldmake such technology truly sustainable and competitive one [6–8].

3.1 Photocatalytic Reduction of Carbon Dioxide

The photocatalytic reduction of CO2 into hydrocarbons or oxygenates is oftencalled the artificial photosynthesis, because it combines the processes of solar lightaccumulation in the form of the products of photo-transformations—CO, CH4,CnH2n+2, CH3OH, HCHO, HCOOH, etc., with the oxidation of water to oxygen,similarly as it happens in green plants [9–13].

The idea of mimicking the natural photosynthesis aimed at the conversion ofCO2 and at the release of oxygen was constantly in the focus of attention for morethan three decades. Various approaches were probed including biological conver-sion, thermal hydrogenation processes, electrochemical reduction and photocat-alytic reduction of CO2 [9–17]. The latter approach is very attractive because thebreaking of C=O bonds in the very stable CO2 molecule needs a high supply ofenergy that can be easily obtained with the light quanta of UV and visible spectralrange. The reduction of CO2—one the most stable forms of C(IV) requires also thepresence of electron-supplying agents, typically water or other sacrificial andabundant electron donors—hydrogen, H2S, SO2, amines, etc. (Fig. 3.1a).

The photocatalytic reduction of CO2 into CH4 and CH3OH is a highlyendothermic reaction with a free Gibbs energies equal to 702.2 kJ/mol(CO2 + 2H2O = CH3OH + 3/2O2) and 818.3 kJ/mol (CO2 + 2H2O = CH4 + 2O2),respectively [9, 11, 13, 16].

Among the advantages of the photocatalytic CO2 conversion are relatively mildconditions, simultaneous mitigation of climatic changes caused by ever-increasinganthropogenic CO2 emission, the possibility of formation of C–C bonds in the form

Fig. 3.1 Schematic illustration of different steps in the photocatalytic CO2 reduction with H2Oover a heterogeneous photocatalyst. The dotted lines indicate the thermodynamic potentials forwater oxidation and CO2 reduction into CO, CH4, and CH3OH. Reprinted with permissions fromRef. [16]. Copyright (2016) The Royal Society of Chemistry

128 3 Semiconductor-Based Photocatalytic Systems …

of hydrocarbons, ethers and carbonic acids and others. From the other side, theefficiency of the photocatalytic CO2 reduction is restricted by a relatively low lightflux intensity near the Earth surface as well as a low selectivity of the photopro-cesses toward the most desirable products [9–11, 13, 14, 16].

In the light of recent global awareness of perils associatedwith climate changes, theproblems of CO2 conversion and utilization find a new appreciation. According to thedecisions made on the Paris Global Climat Conference in 2012 [18], each of 192participating countrieswill work up its own strategy of diminishing ofCO2 productionand conversion of already emitted carbon dioxide and the efforts aimed at the pho-tocatalytic conversion will undoubtedly have a new impetus [10, 11, 13, 14, 16].

Historically, the first studies of the photocatalytic reduction of CO2 were per-formed on ZnO, GaP, ZnS, then the circle of semiconductor photocatalysts wasextended to TiO2, CdS, SiC, as well as various niobates, tungstates and germanates(Fig. 3.1b) [16, 19].

The efficiency and route of the photocatalytic CO2 reduction depend on a varietyof factors, including the composition and band structure of photocatalyst, surfacechemistry, composition of the reactant mixture, the spectral composition of theexciting light, etc. In this view, multiple approaches are developed simultaneouslyto the design of the photocatalytic CO2-converting systems. In general, the pho-tocatalyst design ideology is similar to that used for the water-splitting system withaccount to the specifics of the chemistry and photochemistry of carbon dioxide. Thedesign includes: (1) band structure engineering, such as doping, using quantum sizeeffects and solid state alloying for the band edge manipulating; (2) combination ofvarious semiconductors with matching CB and VB levels; (3) introduction ofsurface vacancies/defects enabling adsorption and conversion of CO2 molecules;(4) expansion of the spectral sensitivity range of the photocatalysts by sensitizationwith organic dyes, metal complexes and inorganic narrow bandgap NPs; (5) intro-duction of additional co-catalysts/co-adsorbents; (6) morphological design ofsemiconductor photocatalysts on the nano-level (shape/lattice face/phase engi-neering) creating favorable conditions for the separation and directed transport ofthe photogenerated charge carriers; (7) combination of various subsystems into theZ-schemes where the CO2 reduction and the water/hydrogen oxidation are sepa-rated in space [9, 15, 16, 20, 21]. Also, an important role is attributed to the constantsearch of new photocatalysts, in particular, among layered materials such as inor-ganic perovskites and carbonaceous compounds, metal-organic frameworks(MOFs) and other classes of compounds [9–12, 20–22].

The photocatalytic conversion of carbon dioxide with the participation ofnanocrystalline semiconductors is a multi-faced phenomenon that can be analyzedand systematized from various aspects, for example, from the viewpoint of materialscience or from the viewpoint of catalysis. In the former case, a classification ofreported of semiconductor-based systems for CO2 reduction can be based oncomposition and structure of the photoactive semiconductor phase. Such approachhas been used in Chap. 2 for the classification of the photocatalytic water-splittingsystems. It allows grasping the versatility of photocatalytic systems and a scope of

3.1 Photocatalytic Reduction of Carbon Dioxide 129

materials that can be used in the design of the solar-light-driven systems for CO2

fixation. We continue with such kind of classification in the present chapter.Nanocrystalline semiconductors as photocatalysts of CO2 conversion. Zinc

sulfide has relatively high CB potential (around −1.8 V vs. NHE [1]) and thereforethis semiconductor is one of the most viable candidates for the role of a photo-catalyst of the CO2 reduction. The nanocrystalline ZnS-based photocatalysts can beproduced by a variety of methods, including the controlled precipitation [23, 24],ion-exchange and hydrothermal synthesis, the latter producing the mostphoto-active materials [25]. Zinc sulfide NPs attached to the surface of montmo-rillonite were used as a working photosensitive body in reactors of variousgeometry for CO2 conversion to methanol, methane, and CO [26]. This studyreported a distinct dependence of the yield of various reduction products on thereactor geometry.

Cadmium sulfide NPs with a much lower CB potential (ECB = −0.8 V vs. NHE)can also be used for theCO2 reduction provided a suitable cocatalyst is introduced intothe photocatalytic system. In particular, the efficient Vis-light-induced conversion ofCO2 into CO was observed for CdS NP assemblies with carbon monoxide dehy-drogenase [27]. Hexagonal colloidal CdS NPs in N,N′-dimethylformamide(DMF) were found to be a visible-light-sensitive photocatalyst of the CO2 reductionwith carbon monoxide as a main product [28]. The addition of excessive Cd2+ ionswas found to affect positively the efficiency of CO2-to-CO conversion and the max-imal CO yield was observed at a molar ratio of excessive Cd2+ to CdS close to 0.2(Fig. 3.2a). A combined PL and EXAFS (extended X-ray absorption fine structurespectroscopy) showed that Cd2+ adsorbs on CdS NPs surface building up the metalsublattice and creating in this way sulfur vacancies [28]. The sulfur vacancies (□)serve as selective sites for the CO2 reduction facilitating the formation of surfacecomplexes between Cd2+ ions, sulfur vacancies and two CO2 molecules and facili-tating the two-electron reduction of carbon dioxide to CO (Fig. 3.2b).

Fig. 3.2 a Effect of Cd2+ excess on the CO formation in the presence of CdS NPs in DMF;b Scheme of CdS NP-photocatalyzed formation of CO with the participation of sulfur vacanciesintroduced by excessive Cd2+. Reprinted with permissions from Ref. [28]. Copyright (1997)American Chemical Society

130 3 Semiconductor-Based Photocatalytic Systems …

The visible light energy can be harvested and used for the reduction of CO2 byternary CdIn2S4 microspheres produced by a hydrothermal synthesis [29]. Themicrospheres are characterized by a comparatively narrow band gap (1.68 eV)which is not favorable for a deep reduction of CO2 but allows to produce dime-thoxymethane and methyl formate when the reaction is performed in methanol [29].

The productivity of metal sulfide microspheres can be boosted by the modificationwith metal NPs or conjugated polymers enhancing the electron-hole separation. Inparticular, the decoration of Bi2WO6 microspheres with polyaniline, polypyrrole, orpolythiophene allows to increase the rate of the photocatalytic CO2 reduction tomethanol and ethanol [30]. The highest activity increment, by a factor of 2.8 ascompared to bismuth tungstate, was observed for polythiophene. The molar ratio ofCH3OH–C2H5OH produced from carbon dioxide was around 3:1 [30].

Recently, open-framework zeolite-like structures consisting of nanosizedmetal-chalcogenide nanoclusters were tested as visible-light-sensitive photocata-lysts of the CO2 reduction. The mixed zinc-germanium-sulfide-based frameworkswere found to photocatalyze the carbon dioxide reduction with water to methaneand the conversion efficiency is affected by the incorporation of third metal cations(Au3+, Pd2+) into the framework [31].

The photocatalytic CO2 reduction was quite broadly studied in titania-basedsystems. In particular, three nanocrystalline TiO2 polymorphs—anatase, rutile, andbrookite were subjected to comparative studies both in the pristine form and after atreatment with ionized helium flow creating oxygen vacancies on the NP surface[32]. Such treatment resulted in a remarkable (up to 10-times) increase of the rate ofphotocatalytic CO2 reduction to CO and CH4 on anatase and brookite, while rutileretained a low activity in these processes even after the treatment. In situ diffusereflectance infrared Fourier transform spectroscopy revealed that the enhancementeffect originates from a higher efficiency of CO��

2 anion radical formation onoxygen vacancies and Ti3+ sites of the plasma-treated titania NPs [32].

A detailed mechanistic study of photoinduced events on the surface of titaniaNPs in contact with CO2-saturated aqueous solutions [33] revealed multiple in-termediates produced with CB electrons (H atoms and CO��

2 radicals) and with VBholes (OH radicals and CO��

3 radicals). Among the secondary intermediates,CH3O� and CH�

3 radicals were detected by the electron paramagnetic resonance. Inview of the versatility of intermediary species, the principal pathway of the CO2

reduction can be affected by a broad variety of factors. In particular, the TiO2 NPsize can affect the selectivity of the CO2 reduction. As the titania NP size isdecreased the number of undercoordinated Ti(IV) ions on the NP surface increasesmaking the deoxygenation of CO2 to CO more and more favorable [34]. Also, thenature and formation rate of the reduction products can be strongly affected bypreferably exposed crystal faces because different TiO2 facets were found to havedifferent band structure and band edge positions [35]. Thus, local heterojunctionsbetween different facets can form in the nanocrystalline TiO2-based systemsfavoring to various reduction pathways.

3.1 Photocatalytic Reduction of Carbon Dioxide 131

The photoactivity of titania in the carbon dioxide conversion can be increased byintroducing co-catalysts, such as metal NPs, that can collect the photogeneratedelectrons and favor to multi-electron processes. As shown in [36], the rate ofphotocatalytic CO2 reduction with water vapor to methane increases from Ag to Rhto Au to Pd to Pt in line with an increase of the electron-accepting capability of themetal NPs. For a given metal, the CO2 conversion increases with a decrease of themetal NP size and typically shows a dome-shaped dependence on the metal content[36]. The CO2 reduction efficiency is limited by a competing reaction of waterreduction to H2 that can be suppressed by the deposition of an additional MgO layeron the photocatalyst surface.

The attachment of titania to the pores of zeolites and mesoporous silica is atraditional method of increasing the TiO2 NP stability and the photocatalytic CO2

reduction efficiency [19]. A sol-gel transformation of Ti(IV) precursors in the poresof HZSM-5 zeolite [37] or TUD-1 mesoporous silica [38] resulted in highly dis-persed TiO2 NPs, the titania-based composites revealing remarkable photocatalyticactivity in the CO2 reduction.

The photocatalytic deposition of Ag NPs onto the titania surface results in analmost 10-fold increase of the efficiency of CO2 conversion into methanol due tothe photoinduced electron accumulation of the metal NPs [39].

A sol-gel transformation of Ti(IV) precursor around the Au NPs was used toproduce the so-called “yolk-shell” nanostructures (Fig. 3.3a, b) with a sole Au NPencapsulated into a mesoporous hollow titania sphere [40]. Excitation of the surfaceplasmon resonance (SPR) in the Au NPs creates a local electromagnetic field(Fig. 3.3c) affecting the photophysical and charge transfer processes in the TiO2

shell. As a result, the yolk-shell structure exhibited an increased rate of the pho-tocatalytic CO2 conversion to methane as well as in the generation of C2H6, con-trary to the bare TiO2 shells and TiO2 P25 [40].

Titania 12-nmNPs anchored to the reduced graphene oxide (RGO) sheets revealeda superior photocatalytic activity in the CO2 reduction to methane over the bare TiO2

NPs and a similar heterostructure based on the non-exfoliated graphite oxide [41].

Fig. 3.3 a, b TEM images of Au/TiO2 “yolk-shell” nanostructures; c simulated spatialdistribution of the local electromagnetic field enhancement on the x–y plane for the yolk-shellAu/TiO2 heterostructure. Reprinted with permissions from Ref. [40]. Copyright (2015) The RoyalSociety of Chemistry

132 3 Semiconductor-Based Photocatalytic Systems …

A solvothermal treatment of titania nanofibers imparts them with the meso-porosity (Fig. 3.4a) resulting in a spectacular 6- and 25-fold enhancement of thephotocatalytic activity in the CO2 reduction to CH4 as compared with thenon-treated titania nanofibers and commercial nanocrystalline TiO2 Evonik P25,respectively [42].

The single-crystalline anatase nanocubes with exposed {100} and {001} facets(Fig. 3.4c, d) were produced by a combined hydrothermal/calcination synthesis andapplied as a UV-sensitive photocatalyst of the reductive CO2 conversion to methaneand methanol [43]. According to the Mott-Schottky measurements, the TiO2

nanocubes have a more negative CB potential as compared to titania nanowires andTiO2 P25 (Fig. 3.4e) favoring to the CO2 reduction not only to methane but also toCH3OH (Fig. 3.4f).

The sodium niobate and tantalate perovskites were used as UV-sensitive pho-tocatalysts of the CO2 reduction to CO, CH4, and CH3OH [44]. The lattice typeinfluences quite strongly the photoactivity of nanocrystalline sodium niobate, therate of photocatalytic CO2 reduction being 2-fold higher for the cubic NaNbO3 ascompared to the orthorhombic polymorph [45].

Ultra-thin WO3 nanosheets (NSs) produced by the oriented attachment oftungsten oxide NPs (Fig. 3.5a–c) revealed enhanced photocatalytic performance inthe CO2 reduction to methane under the illumination with simulated solar light [46].Since the bulk WO3 is passive in this process (Fig. 3.5e), the photoactivity of WO3

NSs consisting of only six repeating unit cells of monoclinic WO3 was ascribed tothe quantum size effects resulting in an increase of the CB potential above the

Fig. 3.4 a, b SEM images of mesoporous TiO2 nanofibers; c, d TEM (c) and SEM (d) images ofTiO2 nanocubes; e energy diagram for titania nanocubes (TC) and titania nanowires (TW); f ratesof CH4 and CH3OH production with the participation of TC, TW and P25. Reprinted withpermissions from Refs. [42] (a, b) and [43] (c–f). Copyright (2014, 2015) The Royal Society ofChemistry

3.1 Photocatalytic Reduction of Carbon Dioxide 133

standard potential of CO2 reduction to CH4 (Fig. 3.5d) [46]. Similar effects wereobserved for thin (*10 nm) Bi2WO6 NSs [47].

Gallium oxide NPs decorated with ultra-small (*1 nm) Ag NPs were found tobe an efficient photocatalyst for the CO2 reduction to carbon monoxide [48]. Thelayered double zinc-gallium hydroxy-carbonates [Zn3Ga(OH)8]2CO3 � mH2Orevealed a photocatalytic activity in the CO2 conversion to CO and CH3OH underthe UV illumination [49]. The photocatalyst modification with Ag or Au NPsincreased the photoconversion efficiency by around 70–80%.

Ultra-thin ZnGa2O4 NSs assembled into mesoporous microspheres [50] as wellas Zn2GeO4 nanobelts [51] revealed a high photocatalytic activity in the CO2

reduction to CH4 with the simultaneous water oxidation to O2.Mixed CuFeO2/CuO films can be relatively easily prepared by the electrodepo-

sition and applied as a photocatalyst of the CO2 reduction to formate at the expense ofwater oxidation to O2 [52, 53]. The participation of carbon dioxide in the formategeneration was unambiguously confirmed by 13C isotopic studies [52]. The photo-catalysts gradually degrade due to partial copper reduction, but can be easily recov-ered via the oxidative annealing and used continuously for more than a month [52].

The BiOCl nanoplates can be rendered photocatalytically active in the CO2

conversion by generating oxygen vacancies under the UV illumination [54]. The O

Fig. 3.5 a–c TEM images of ultra-thin WO3 NSs; d energy diagrams for NS and bulk WO3;e kinetic curves of the CH4 accumulation for WO3 NSs and commercial bulk tungsten oxide.Reprinted with permissions from Ref. [46]. Copyright (2012) American Chemical Society

134 3 Semiconductor-Based Photocatalytic Systems …

vacancies favored to enhanced CO2 adsorption and capture of the photogeneratedcharge carriers suppressing their recombination.

Silicon NPs (1–4 nm) produced by the ball milling can be used as avisible-light-sensitive photocatalyst of the CO2 reduction with water exhibiting ahigh selectivity towards the formation of HCHO [55]. The photoreduction isobserved only on Si NPs with an open surface, while the surface passivation withalkyl derivatives deteriorates the photocatalytic properties of Si NPs completely,indicating on a crucial role of CO2 adsorption on the active surface sites [55].

Similarly to the water-splitting systems discussed in Chap. 2, the photocatalyticconversion of CO2 is constantly probed on new semiconductor compounds, inparticular on graphitic carbon nitride (GCN) or metal organic frameworks (MOFs).In particular, GCN modified with a Co-bipyridylate and a CoOx co-catalysts wasused as a photocatalyst of the CO2 deoxygenation into CO (Fig. 3.6a) [56]. GCN isabundant with nitrogen vacancies that act as strong binding sites for CO2 mole-cules, while the Co-based species accelerate electron transfers between GCN, CO2,and triethanolamine [56]. The photoaction spectra of CO and H2 (a by-product)generation follow the absorption spectrum of GCN attesting to the photocatalyticcharacter of the photoinduced transformation of CO2 in this system (Fig. 3.6b).

Similarly to the H2-evolving systems, the exfoliation or nano-structuring ofGCN typically results in an enhancement of the photocatalytic CO2 conversionefficiency. For example, the ammonia-assisted thermal exfoliation of GCN yields ananoplate-like material with a plate thickness of around 3 nm [57]. A decrease inthe GCN particle thickness is accompanied by a blue shift of the absorption bandand an increase in the CB potential (Fig. 3.6b) as revealed by the Mott-Schottkymeasurements. The higher reducing potential of the photogenerated CB electrons ofnanostructured GCN is reflected in an increased rate of the photocatalytic methanoland methane generation from carbon dioxide (Fig. 3.6c) [57].

Porphyrin-incorporated Zr-based MOFs were successfully tested as asolar-light-driven photocatalyst of the CO2 reduction with water to HCOOH [58].A hybrid of MOF-253 with Ru(CO)2Cl2 complex displayed a visible-light-drivenphotocatalytic activity in the CO2 reduction to formate anions [59].

Cobalt imidazole zeolitic MOF (Co-ZIF-9) was shown to act as a co-catalyst forthe CO2 adsorption and activation in a combination with a TiO2 photocatalyst of theCO2 reduction to CO and methane, increasing the photoconversion efficiency by afactor of 2 as compared to the bare titania [60].

MOFs can also be used as a “host” for various molecular species—electro-catalysts, photosensitizers, etc., allowing to transfer the CO2 reduction into theheterogeneous regime and increase the stability and turnover numbers of the cat-alysts. For example, a manganese bipyridyl complex, Mn(bpydc)(CO)3Br (bpy-dc = 5,5′-dicarboxylate-2,2′-bipyridine) was incorporated into a highly robust Zr(IV)-based MOF UiO-67 (Fig. 3.7a).

The assembly was then sensitized with a [Ru(dmb)3]2+ (dmb = 4,4′-

dimethyl-2,2′-bipyridine) complex and used as a photocatalyst of the CO2 reductionto formate with 1-benzyl-1,4-dihydronicotinamide as a sacrificial electron donor[61]. The cyclic sequence of photoinduced electron transfers in this system can be

3.1 Photocatalytic Reduction of Carbon Dioxide 135

presented by Fig. 3.7b. As compared to homogeneously soluted Mn complex, theMOF-incorporated assembly retained prolonged stability, partially because the rigidframework prohibited dimerization of a one-electron-reduced Mn complex.

Similar strategy of incorporating a molecular metal-complex photocatalyst into arobust MOF structure to prevent undesirable side reactions and to enhance thephotocatalyst stability was realized for ReI(CO)3(bpydc)Cl, bpydc = 2,2′-bipyridine-5,5′-dicarboxylate incorporated into the Zr-based MOF UiO-67 [61].The combination of a molecular photocatalyst with a MOF host affords anunprecedented flexibility in structure variation. In particular, the number of Recomplexes per unit cell of the MOF can be varied as n = 0, 1, 2, 3, 5, 11, 16, and24. The highest photocatalytic activity in the CO2 reduction to CO was observed forn = 3 [62] (Fig. 3.8a), as a result of a fine balance of the proximity betweenphotoactive centers needed for the cooperatively enhanced photocatalytic activity.The most active Re complex/MOF composite photocatalyst with n = 3 wasdeposited onto the surface of Ag nanocubes (Fig. 3.8b) with a strong SPR on thecube edges. The SPR-induced local electromagnetic field promoted charge

Fig. 3.6 a Scheme of GCN/Co–bpy–CoOx photocatalyst; b diffuse reflectance spectrum of GCN(dashed line) and the rates of CO and H2 generation (bars). Insert in (b): kinetic curves of CO andH2 accumulation; c, d Energy diagram (c) and the rates of photocatalytic CH4 and CH3OHgeneration (d) for bulk and thermally-exfoliated nanostructured (NS) GCN. Reprinted withpermissions from Refs. [56] (a, b) and [57] (c, d). Copyright (2014) American Chemical Society(a, b) and (2017) The Royal Society of Chemistry (c, d)

136 3 Semiconductor-Based Photocatalytic Systems …

separation processes in the Re complex/MOF layer resulting in a 7-fold enhance-ment of the photocatalytic CO2 deoxygenation [62].

Sensitizer-based systems for CO2 photoreduction. Nanocrystalline titania sen-sitized with a well-known Ru(II)-bipyridyl complex (N719) coupled with a Ptcounter electrode was used for the visible-light-driven photoreduction of CO2 roHCOOH, HCHO, and CH3OH in a two-vessel geometry (Fig. 3.9a) [63]. A portionof the TiO2 film was “stained” with the dye and connected by an I�

�I�3 -containing

electrolyte to a Pt counter electrode. This part of the film acted as a DSSC providingelectrons for the rest of the TiO2 film where the CO2 reduction to oxygenates tookplace (Fig. 3.9b). The photogenerated holes were transferred by iodine/iodideredox-shuttle (see Chap. 4 for details on the DSSC design) from TiO2/dye surface tothe Pt counter electrode and then went via the electric circuit to another TiO2 filmplaced into a second vessel and separated from the first one by a Nafion membrane.Water was oxidized on this second TiO2 film to O2 providing electrons for the CO2

reduction in the first vessel thus completing the photoelectrocatalytic cycle [63]. Insuch way, the products of CO2 reduction were shielded from the re-oxidation on the

Fig. 3.7 a Scheme of the synthesis of UiO-67-Mn(bpy)(CO)3Br; b Mechanism of thephotocatalytic reduction of CO2 with sensitized UiO-67-Mn(bpy)(CO)3Br. Reprinted withpermissions from Ref. [61]. Copyright (2015) American Chemical Society

3.1 Photocatalytic Reduction of Carbon Dioxide 137

anode surface. The rate of CO2 conversion in such system can be increased con-siderably by an external bias.

Nanocrystalline titania decorated with a series of Nile-Red-type dyes and Pd NPswas used as a photocathode for the conversion of CO2 into methanol with thesimultaneous water oxidation on a W-doped BiVO4-based photoanode modifiedwith a cobalt phosphate co-catalyst (Fig. 3.9c) [64]. Both electrodes were immersedinto aqueous KHCO3 solution and the formation of CH3OH from carbonate ionsand O2—from water molecules was confirmed by isotopic studies. The photosyn-thetic system was additionally biased with a voltage of *0.6 V from an inde-pendent silicon solar cell [64].

A tandem principle was realized for a nickel(II) oxide photocathode sensitizedwith a supramolecular Ru(II)-Re(I) complex assembly and coupled to a tantalumoxynitride photoanode modified with a cobalt oxide co-catalyst (Fig. 3.9d) [65].The nickel oxide-based photocathode showed a selectivity toward the formation ofCO. The system also used water as an electron donor and a low external biasof *0.3 eV. The p-type CuGaO2 photocathode senstized by a similarsupramolecular assembly of Ru(II)-Re(I) complexes revealed a photoelectrochem-ical activity in the CO2 reduction by water without an additional external bias [66].

Fig. 3.8 a Scheme of “Re complex/MOF/Ag nanocube” composite formation; b TEM of thecomposite particles. Reprinted with permissions from Ref. [62]. Copyright (2017) AmericanChemical Society

138 3 Semiconductor-Based Photocatalytic Systems …

A Ru-bipyridyl sensitizer was also applied to extend the spectral sensitivityrange of nanocrystalline CuCo2O4 [67]. The sensitized spinel acts as a photocatalystof the CO2 deoxygenation to CO.

Mesoporous N-doped Ta2O5 microspheres sensitized with a Ru-bipyridylcomplex showed a photocatalytic activity in the visible-light-driven CO2 conver-sion into HCOOH [68].

Composite TiO2/Zn phthalocyanine NPs produced by a combined microwave/hydrothermal process can be applied as a selective photocatalyst of the CO2

reduction to methanol under simulated solar light illumination [69].The functionalization of TiO2 NPs with various aminosalicylic acids

(ASA) results in the formation of surface charge transfer complexes that extend theabsorption range of titania far into the visible range (Fig. 3.10a, blue curve) [70].The photoinduced charge transfer in such complexes occurs directly from the

Fig. 3.9 a Photocatalytic reduction of CO2 in a two-vessel reactor; b working principle of thetwo-vessel reactor; c, d schematic design of photoelectrochemical systems for the CO2 conversionwith TiO2-based (c) and NiO-based (d) photo-electrodes. Reprinted with permissions from Refs.[63] (a, b), [64] (c), and [65] (d). Copyright (2013) Elsevier (a, b), (2017) The Royal Society ofChemistry (c), and (2016) American Chemical Society (d)

3.1 Photocatalytic Reduction of Carbon Dioxide 139

highest occupied molecular orbital of ASA molecules into the CB of TiO2 NPs andthen—to adsorbed CO2 molecules. As a result, the absorption band edge of aternary TiO2-ASA-CO2 systems shifts even further into the visible range as com-pared to the binary TiO2-ASA charge transfer complex (Fig. 3.10a, red curve).

The light sensitivity range of GCN can be greatly extended by coupling it withCo porphyrin [71] or Co bipyridyl complexes [72]. The sensitized GCN showed a5-fold (Co bipyridylate) and a 13-fold (Co porphyrinate) increments of the rate ofphotocatalytic CO2 reduction to CO as compared to the individual GCN. Also, bydecreasing the lateral size and thickness of GCN particles the photoconversionefficiency can be additionally enhanced to a CO yield of 17 lmol/g/h [71].A dependence of the reaction rate on the photoexcitation wavelength (the pho-toaction spectrum) mimics the GCN/Co-porphyrin absorption spectrum(Fig. 3.10b) showing two distinct sensitivity ranges of GCN (k < 470 nm) andCo-porphyrin (k > 500 nm).

In tandem systems comprising GCN and Ru(II)-bipyridyl complexes, the rateand pathway of the CO2 reduction depend on the bipyridyl substituent X in the4-position [73]. In particular, the tandems based on X = COOH and X = PO3H2 themain product of the CO2 reduction was HCOOH, while for X = CH2PO3H2 thephotoprocess yielded CO and HCOOH with a relatively high selectivity towardcarbon monoxide (40–70%). The difference arises from a photoinduced transfor-mation of the CH2PO3H2-substituted Ru bipyridyl complex into a polymeric spe-cies active specifically toward the CO formation [73]. Polymeric Ru-bipyridylcomplexes were also used as sensitizers/electrocatalysts to increase the rate ofphotocatalytic CO2 reduction to formate over indium phosphide [74].

Doped semiconductor nano-photocatalysts of CO2 conversion. Doping of thewide-bandgap semiconductors, such as titanium dioxide, with metal ions/non-metalatoms is one of the most frequent and productive strategies both of increasing the

Fig. 3.10 a Absorption spectra of TiO2 NPs, binary TiO2-3ASA charge transfer complex, andternary TiO2-2ASA-CO2 system; b Rate of CO production over GCN/Co-porphyrin hybrid (bars)compared with the hybrid absorption spectrum (red line). Reprinted with permissions from Refs.[70] (a) and [71] (b). Copyright (2013) American Chemical Society (a) and (2017) Elsevier (b)

140 3 Semiconductor-Based Photocatalytic Systems …

photoreaction efficiency due to the recombination inhibition and of extending thespectral sensitivity range of the photocatalysts as a result of the participation oflocalized/delocalized dopant levels in the light absorption. In Chap. 2 we havediscussed doping methods for the photocatalytic H2-evolving systems. Thisapproach is also broadly used in the photosynthetic CO2-conversion systems [19].

For example, the incorporation of Pd, Cu, and Mn ions into the TiO2 latticeimparts the semiconductor with the sensitivity in the visible range (Fig. 3.11a) [34,35]. The dopants are present in the titania lattice as –O–M–O– fragments and canactively participate in the trapping of the phogenerated charge carriers—the VBholes in the case of Pd and Cu and the CB electrons in the case of Mn (Fig. 3.11b).As a result, doping increases the efficiency of the photocatalytic CO2 reduction tomethane [75].

In-doped nanocrystalline TiO2 was applied as a photo-active phase for the pho-tocatalytic CO2 conversion in microchannel monolith photoreactors [76]. Thesesystems produce a broad range of reduction products with a product populationdecreasing in the following sequence: CO > CH4 > C2H6 > C2H4 > C3H6. After amulti-parameter optimization of the photoreactor performance, the quantum yields ofCO and methane reached 0.1 and 0.022%, respectively. Doping of the mesoporoustitaniawith Inwas also reported to change the basic CO2 reduction product fromCO toCH4 increasing the light harvesting efficiency by a factor of around 8 [77].

Doping with Ni2+ prohibits the growth of TiO2 nanocrystals and the anatase-to-rutile conversion during the thermal synthesis and results in a partial substitutionof Ti4+ with nickel ions [78]. The dopant extends the absorption edge of titania tolonger wavelength and provides traps for the photogenerated charge carriers thusdecreasing the recombination efficiency. The Ni-doped nanocrystalline titania loa-ded onto the quartz optical fibers can be used in a monolith reactor for the CO2

reduction with water vapors under the UV/Vis illumination [78]. A similar effect onthe growth of titania NPs was observed for Ce doping [79]. After the deposition onmesoporous SBA-15 silica TiO2-Ce was applied as a visible-light-sensitive pho-tocatalyst of the CO2 conversion into CO and CH4. The photocatalytic conversionof CO2 in the presence of TiO2-Cu/SBA-15 composite with a 45 wt.% loading of

Fig. 3.11 a Diffuse reflectance spectra of TiO2 and Cu, Pd, and Mn-doped (1 mol%) titania; b,c mechanistic diagrams of the CO2 photoreduction on TiO2–Cu (b) and TiO2–Mn (c). Reprintedwith permissions from Ref. [75]. Copyright (2017) American Chemical Society

3.1 Photocatalytic Reduction of Carbon Dioxide 141

the photoactive component yields methanol as a principal product [80]. Thenanocrystalline titania doped with Ce(IV) revealed an extended spectral sensitivityrange and the highest photocatalytic activity in the CO2 reduction at 0.28 mol%dopant content [81].

Nanocrystalline ZnS doped with Ni(II) showed a high selectivity toward theformation of methyl formate as a result of the photocatalytic CO2 reduction inmethanol [25]. The highest yields were observed at a *0.3 mol% dopant content.

The melamine pyrolysis in the presence of ammonium molybdate results in theformation of mesoporous Mo-doped GCN [82]. The Mo doping enhanced the GCNabsorptivity in the visible range proportionally to the dopant content (Fig. 3.12a).The Mo-doped GCN revealed an enhanced photocatalytic activity in the CO2

reduction to CO and CH4 compared with the pure g-C3N4 [82].Nitrogen-doped anatase NPs (10–20 nm) with predominantly exposed

(001) faces covered with RGO sheets were tested as a photocatalyst of the CO2

reduction with water to methane [83]. Coupling with RGO resulted in an 11-foldincrease of the CO2 conversion efficiency.

Oxygen-enriched titania NPs were produced by the thermal decomposition of aperoxy-titanium complex [84]. The oxygen doping shifted the absorption thresholdof TiO2 NPs from 390 nm to around 420 nm favoring to the visible-light-drivenphotocatalytic conversion of CO2 into methane.

Titania NT arrays produced by the anodization of a Ti foil followed by thehydrothermally-assised doping with vanadium and nitrogen (Fig. 3.13a) werereported to be an efficient photocatalyst of the CO2 reduction to CH4 [85].

The calcination of a zeolite-like zinc-imidazolate framework on air results in theformation of mesoporous ZnO that retains the zeolitic structure up to 300 °C, whileconverting to wurtzite zinc oxide at higher temperatures [86]. Along with thisconversion, carbon doping of ZnO takes place as well as the deposition of a

Fig. 3.12 Diffuse reflection spectra of a Mo-doped GCN with a different dopant content (valueson figure correspond to the molar% of Mo with respect to melamine) and b carbon-doped ZnOnanostructures (values correspond to the calcination temperature). Reprinted with permissionsfrom Refs. [82] (a) and [86] (b). Copyright (2016) Elsevier (a) and The Royal Society ofChemistry (b)

142 3 Semiconductor-Based Photocatalytic Systems …

carbonaceous layer on the ZnO surface imparting the material with the visible lightsensitivity (Fig. 3.13b). As a result, the C-doped mesoporous ZnO can be used as asolar-light-driven photocatalyst of the CO2 reduction to methanol, the conversionefficiency being around 6-times higher than for the undoped reference ZnO NRs[86]. Similarly, the annealing of titania layers deposited onto a carbon nanospheretemplate yielded hollow TiO2 spheres with a surface carbon layer and carbon-dopedinorganic matrix [84]. The doped titania hollow spheres showed twice as highphotocatalytic activity in the CO2 reduction to CH4 as the reference TiO2 P25.

Binary semiconductor nanoheterostructures for CO2 photoreduction. Thephotocatalytic conversion of CO2 in the presence of methanol over a CuO/TiO2

heterostructure results in the preferential formation of methyl formate [22]. At that,methanol serves as a reactant and as a VB hole scavenger, while CO2 is reduced bythe photogenerated CB electrons [22]. Deposition of Cu2O NPs onto titania wasreported to result in enhanced adsorption of CO2 and simultaneous inhibition ofwater adsorption [87].

Simultaneously, Pt NPs deposited onto TiO2 acts as electron “pools” promotingmulti-electron photoinduced reactions. The summary effect of Cu2O and Pt NPsresults in complete suppression of the water reduction on the titania surface and theselective reduction of CO2 with CH4 as a sole product of the photoreaction [87].Titania and Cu2O can be separated in space and used as a photoanode and acathode, respectively, in a photoelectrochemical system for the CO2 reduction. Inthis way, the Cu2O can be protected against the oxidative photocorrosion with thephotogenerated VB holes [88].

Ordered and hierarchically porous CeO2/TiO2 heterostructures were produced byusing SBA-15 zeolite as a sacrificial template and used as a visible-light-sensitivephotocatalyst of the CO2 reduction with water [89]. The enhancement factor of thephotocatalytic reduction of CO2 to CO over the TiO2/CeO2 nanocompositesdepends on the crystalline structure of TiO2, being the highest for the rutile/CeO2

composites of all the titania polymorphs [90].

Fig. 3.13 a SEM image of V,N-doped titania NT arrays [85]; b, c SEM and TEM images ofcarbon nanosphere template (b) and C-doped titania hollow spheres (c); d STEM image andelement distribution profiles for a TiO2 hollow sphere (b–d). Reprinted with permissions from Ref.[134]. Copyright (2017) The Royal Society of Chemistry

3.1 Photocatalytic Reduction of Carbon Dioxide 143

By using self-ordered polystyrene microspheres as a sacrificial template orderedmacroporous titania can be produced. The decoration of such materials with ananolayer of ceria extends its spectral sensitivity to the visible range [91]. Themacroporous TiO2/CeO2 retains an ordered character of the original polystyrenephotonic crystal favoring to a higher light absorptivity and can be used as asolar-light-driven photocatalyst of the CO2 reduction to CO with water [91].

Titania coupling with GCN allows for the spatial separation of charge carrierssince the photogenerated electrons are collected in the lower-positioned titania CBand the VB holes (photogenerated in both semiconductors) are supplied to thereactants through the VB of carbon nitride. As a result of the charge separation, thetitania/GCN composites displayed an ehnanced photocatalytic activity in the CO2

reduction [92]. A similar enhancement effect in the CO2 reduction to methane wasobserved also for GCN/KNbO3 [93] and GCN/NaNbO3 [94].

Efficient separation of the photogenerated charge carriers can be realized in abinary heterostructure of boron carbide B4C with GCN [95]. The p-type conductingsemiconductor B4C forms with the n-type GCN a heterojunction with a favorableband edge offsets, allowing for the interfacial electron transfer from the photoex-cited B4C to GCN. The VB hole in B4C is filled by oxidizing H2 to atomichydrogen and by transferring the photogenerated CB electrons to GCN (Fig. 3.14a),while CO2 is reduced primarily on the surface of platinum particles deposited ontothe graphitic carbon nitride. The chain electron transfers result in the photocatalyticreduction of CO2 to CO and further hydrogenation of carbon monoxide to methaneand ethane [95].

The sheet-like GCN can be used as a “mat” for the growth of other nanocrys-talline semiconductors, such as indium oxide. The GCN/In2O3 heterostructuresrevealed an enhanced photocatalytic activity in the reduction of water and CO2 dueto spatial separation of the CB electrons and VB holes between the heterostructurecomponents [96].

On the other hand, the GCN sheets can be deposited as a shell around inorganiccore crystals, such as NRs and NWs. The core/shell LaPO4/GCN NWs (Fig. 3.14b,c) displayed an enhanced photocatalytic activity in the CO2 conversion into CO ascompared with individual components [97]. The GCN NPs with a size of around3 nm formed via the urea condensation on the surface of TiO2 brookite nanocubes(Fig. 3.14d, e) acted as a spectral sensitizer allowing to reduce CO2 selectively toCH4 under Vis-light illumination [98].

Vanadium-doped TiO2 sensitized by graphene NSs was used as a photocatalystfor the model endothermic conversion of potential environmental pollutants intophotosynthetic products [99]. On the first step, methylene blue dye as a modelpersistent pollutant was mineralized to CO2 and then carbon dioxide reducedphotocatalytically into CH4, CH3OH, and CH3CH2OH with an apparent quantumefficiency of *5% at 420 nm [99].

By combining Fe2O3 NTs with branched SnO2 NRs with predominantly exposed{110} and {101} faces the photoelectrocatalytic reduction of CO2 to methanol canbe accelerated by more than 7 times [100].

144 3 Semiconductor-Based Photocatalytic Systems …

In metal chalcogenide/titania composites with a proper CB and VB level posi-tions, the CO2 reduction and the donor (for example, water) oxidation occur on thesurface of titania and metal chalcogenide NPs, respectively. Both branches of thephotoprocess can be additionally separated in space to avoid the re-oxidation ofCO2 reduction products and to promote the formation of C–C bonds. For example,by combining two subsystems—the Moorella thermoacetica bacteria decoratedwith CdS NPs and TiO2 NPs loaded with Mn(III) phthalocyanine into a Z-system,the CO2 reduction to acetic acid can be achieved [101]. Both subsystems areconnected by a donor/acceptor cysteine/cystine couple. The cysteine (Cys) getsoxidized on the surface of CdS NPs to cystine (CySS) and then CySS is regeneratedto Cys on the surface of the phthalocyanine-functionalized titania (Fig. 3.15a).

Titania NTs can be sensitized to the visible light by CdS and Bi2S3 NPs [102].The decoration of TiO2 NTs with bismuth sulfide NPs favors to the formation ofmethanol and increases the total CO2 reduction rate by a factor of 2.2 as comparedto the bare titania NTs [102]. Titania NTs sensitized by mixed CdSeTe NPs wereused as a photoelectrocatalyst of the CO2 reduction to methanol in a two-cell reactor[103]. The bandgap of mixed cadmium selenide-telluride NPs was adjusted to1.24 eV (corresponding to kbe * 1000 nm), thus allowing for harvesting the entirevisible and near-IR irradiation.

The efficiency of photocatalytic reduction of CO2 with water vapor over TiO2/CdSe nanoheterostructures was found to depend on the CdSe NP size as a result ofa size-dependence of the cadmium selenide CB position (Fig. 3.15b) [104]. Themain reduction product was methane with CH3OH, CO, and H2 present as sec-ondary admixtures.

Fig. 3.14 a Scheme of photoinduced charge separation in p-B4C/n-GCN heterojunction(potentials are given versus NHE). b, c TEM images of LaPO4/GCN core/shell nanowires. d,e TEM images of TiO2/GCN nanocomposite. Reprinted with permissions from Refs. [95] (a), [97](b, c), and [98] (d, e). Copyright (2016, 2017) Elsevier

3.1 Photocatalytic Reduction of Carbon Dioxide 145

Mixed cadmium zinc sulfide Cd0.2Zn0.8S NPs were deposited onto a UiO-66MOF resulting in a nanocomposite with enhanced spatial separation of the pho-togenerated charge carriers [105]. The heterostructure can be used as a photocat-alyst of the CO2 reduction to methanol at the expense of water oxidation, showingremarkable chemical/photochemical stability.

3.2 Photocatalytic Fixation of Dinitrogen

The photochemical reduction of dinitrogen (N2) is probably second by importance,after the CO2 fixation, photosynthetic process that can be used directly for theaccumulation and storage of the solar energy in a chemical form. Also, the N2

reduction to ammonia can provide the ways to valuable raw materials of chemicalindustry and fertilizers. Nowadays, the most important process of the N2 fixation isa conventional catalytic Haber-Bosch reaction between N2 and H2. This process isthough energy-demanding, requires non-renewable feedstocks for the hydrogengeneration and suffers from the catalyst poisoning. Therefore, alternative ways ofthe N2 fixation are constantly developed including thermal and non-thermalplasma-based processes, biological dinitrogen fixation, metal-complex catalysis andphotocatalytic N2 transformations [106].

The dinitrogen reduction results in energy accumulation of 678 kJ/mol and therealization of this process using the solar light can potentially allow substitutingmodern energy-demanding catalytic technologies with mild photosynthesis-likeprocesses thus contributing to the alleviation of global climate and energy diver-sification problems.

The feasibility of photochemical reduction of N2 to ammonia and traces of N2H4

over desert sands of various origins was first shown in 1983 [107]. The N2 fixationefficiency was found to depend on the content of TiO2 in the sand samples. Thiswork indicated that close to *107 tons of dinitrogen per year can be converted into

Fig. 3.15 a Scheme of a “Moorella thermoacetica—CdS /TiO2–Mn phthalocyanine” tandemsystem for the CO2 conversion into acetic acid; b Energy diagram of TiO2/CdSe heterostructureswith bulk and nanocrystalline cadmium selenide. Reprinted with permissions from Refs. [101](a) and [104] (b). Copyright (2016, 2010) American Chemical Society

146 3 Semiconductor-Based Photocatalytic Systems …

ammonia in the semiarid desert conditions under the solar light illumination of thedesert sands [107].

A special feature of the photocatalytic nitrogen fixation is a strong dependenceof the conversion efficiency on the nature of lattice defects of semiconductorphotocatalysts. In some cases, the presence of defects—anionic and cationicvacancies, is an obligatory condition for a semiconductor to reveal photocatalyticproperties in the N2 reduction [108–110]. For example, the microwave treatment ofGCN results in the formation of numerous pores as well as in the generation ofnitrogen vacancies via the NH3 elimination [108]. The newly generated vacanciesact as perfect N2 adsorption sites because of matching physical sizes of a dinitrogenmolecule and a nitrogen vacancy [110]. Also, the vacancies can trap the photo-generated charge carriers thus allowing to avoid the recombination processes.Finally, the generation of vacancies with corresponding mid-bandgap electronicstates results in an increase of the light absorbance of GCN in the visible spectralrange (Fig. 3.16a) [108, 110].

Typically, there exists an optimal vacancy density producing the highest pho-tocatalytic activity in the N2 reduction. For example, for the microwave-treatedGCN such density can be created at a 25-min treatment providing the best per-formance for the photoinduced ammonia generation (Fig. 3.16b) [108].

Figure 3.16b shows that the dinitrogen reduction kinetics is very similar to thekinetics of the water reduction to H2, the process rate remaining roughly constantfor many hours of illumination. Also, the microwave-treated GCN retains a steadyphotocatalytic activity for a long illumination period 20 h and more (Fig. 3.16b,insert).

Similarly to the heat treatment, the porosity and nitrogen vacancies in GCN canbe introduced by a treatment with concentrated acids, such as HCl [109] and HNO3

[111]. The HCl treatment, in particular, results in a spectacular *13-fold increment

Fig. 3.16 a Diffuse reflectance spectra of GCN produced at 550 °C (CN-550) and the products ofmicrowave treatment with a varied duration (denoted as MCN-x, x—duration); b kinetic curves ofthe ammonia accumulation over different GCN samples. Reprinted with permissions from Ref.[108]. Copyright (2016) Elsevier

3.2 Photocatalytic Fixation of Dinitrogen 147

of photocatalytic GCN activity in the N2 reduction to ammonia as compared withthe bulk material [109].

The thermal condensation of an adduct produced in methanol at the interactionbetween melamine and concentrated nitric acid was found to yield a sponge-likeporous GCN material abundant with nitrogen vacancies. The vacancies impart suchmaterials with advanced photocatalytic properties in the dinitrogen conversion intoammonia [112]. As compared to GCN produced by the conventional melaminecondensation, the porous sponge-like GCN exhibited a *27-times higher photo-catalytic activity.

In a similar way, a series of metal-sulfide NPs and metal-sulfide/GCNnanoheterostructures were prepared, in particular, Cd–Zn–Sn–S [113] andCd–Zn–Sn–S/GCN [114], Cd–Zn–Mo–S/GCN [115], Cd–Ni–Mo–S [116]. Suchcomposites are abundant with the sulfur vacancies in metal chalcogenide NPs thatfavor to the adsorption and activation of N2 molecules.

Aqueous chalcogels comprising ultra-small nanosized [Mo2Fe6S8(SPh)3]3+ and

[Sn2S6]4− sub-units (SPh is a thiophenolate anion) were found to photocatalyze the

reduction of N2 to ammonia under the solar light illumination thus mimickingFe–Mo–S active centers of the N2-fixating micro-organisms [117].

Oxygen vacancy-rich BiOCl NPs revealed photocatalytic properties in the N2

reduction [118]. As the lattice of bismuth oxychloride is strongly anisotropic, theoxygen vacancies on {001} and {010} lattice facets have non-equal energies andcan coordinate N2 molecules in a different way (Fig. 3.17a), thus favoring to dif-ferent pathways of the N2 reduction. The dinitrogen photoreduction on {001} faces,where N2 molecules are coordinated by a sole nitrogen atom, results in the pref-erential formation of ammonia (Fig. 3.17b), while the photoreduction of N2 coor-dinated to the {010} face by both N atoms yields hydrazine as a principalintermediate, followed by the N2H4 conversion into NH3 (Fig. 3.17c) [118].Similarly, BiOBr NSs with preferentially exposed {001} facets revealed a highphotocatalytic activity in the N2 reduction to ammonia due to abundant presence ofoxygen vacancies [119]. The process requires no additional hole scavengers orco-catalysts and proceeds at the ambient humidity, pressure, and temperature.

The 2–5-nm bismuth monoxide NPs were reported to be an efficient photocat-alyst of N2 reduction to ammonia capable of producing up to *1230 mmol NH3

per (g � h) which is by three orders of magnitude faster than for conventionalFe-doped titania photocatalysts [120]. The most probable reason for the highphotoactivity of BiO NPs is the strong adsorption and activation of N2 molecules onBi-rich (and, therefore, rich with oxygen vacancies) NP surface.

The introduction of co-catalysts capable of coordinating N2 molecules andweakening of the triple N–N bond results in a pronounced enhancement of thephotocatalytic dinitrogen reduction. For example, mixed Ru(II) complexes withEDTA and chloride anions can coordinate molecular nitrogen and facilitate N2

reduction by the photoelectrons generated in silver-doped CdS/RuO2/Ptnanoheterostructure [121]. The isotopic 15N studies confirmed that N2 was the solenitrogen source in the photoproduced ammonia.

148 3 Semiconductor-Based Photocatalytic Systems …

A modification of partially exfoliated and protonated GCN with RGO NSsincreases the rate of photocatalytic N2 reduction to ammonia by factors of around42, 8, and 4 as compared to the bulk GCN, exfoliated GCN and a composite of bulkGCN with RGO, respectively [122]. The strong PL quenching in the composites ofexfoliated GCN and RGO attests to an efficient charge transfer from the photo-catalyst (GCN) to the co-catalyst (RGO).

A composite of titania with poly(3-methylthiophene) revealed a photocatalyticactivity in the visible-light-driven conversion of N2 into ammonia and ammoniumsalts when exposed to solar-like “white” light at the ambient humidity and tem-perature [123].

Similarly to the CO2 photoconversion, doping is also a potent tool forinfluencing the photocatalytic properties of semiconductor nanomaterials in thenitrogen fixation. At that, the most spectacular efficiency increments were observedfor doping with iron in its various forms. For example, FeIII-doping of the

Fig. 3.17 a Crystal structure of BiOCl and the corresponding cleaved {001} and {010} facets(left part); the terminal end-on adsorption structure of N2 on {001} surface and side-on bridgingadsorption structure of N2 on {010} surface of BiOCl (right part); b, c kinetic curves of NH3

(b) and N2H4 (c) formation of {001} and {010} facet exposed BiOCl. Reprinted with permissionsfrom Ref. [118]. Copyright (2016) The Royal Society of Chemistry

3.2 Photocatalytic Fixation of Dinitrogen 149

nanocrystalline titania with preferentially exposed {101} facets results in an almost4-fold increase in the photocatalytic N2 reduction rate [124].

Though typically the role of iron dopants is explained in a conventional way bytrapping of the photogenerated charge carriers, a more probable reason for theselective activation with Fe dopants can be in the formation of donor-acceptorbonds between N2 and Fe centers thus resulting in an activation of otherwise inertdinitrogen molecule. In particular, by doping of GCN with iron the photocatalyticammonia generation rate can be increased by a factor of 13 and higher [125]. Ironions were found to be incorporated into the interstitial positions of GCN andstabilized by electron-rich heptazine fragments via donor-acceptor interactions. TheFe centers participate in the N2 chemisorption and the photoinduced electrontransfers from GCN to adsorbed dinitrogen as confirmed by the density functionaltheory calculations [125]. In particular, HOMO of N2 become delocalized whileLUMO is hybridized with the iron-related orbitals as a result of the Fe–N2 inter-actions, thus facilitating the electron transfers from GCN to N2.

Copper(I)—nitrogen vacancy couples in Cu+-doped GCN serve as the N2

adsorption and activation sites promoting photoinduced electron transfers from thesemiconductor to adsorbed dinitrogen molecules converting them to ammonia ions[126].

Doping with iron enhances the photocatalytic N2 reduction over mesoporousTa2O5 by a factor of two with an optimal Fe loading of around 1 wt.% [127].

The iron can be introduced as Fe2O3 deposited onto other wider-bandgapsemiconductors to facilitate the photoinduced charge transfers. In particular, TiO2/Fe2O3 systems displayed a much higher efficiency of the photocatalytic N2

reduction to ammonia and hydrazine as compared to sole titania (Fig. 3.18a) [128].The formation of both products can be imagined as a step-wise dinitrogen reductionand the addition of H atoms produced from the water reduction with CB electronsto a N2 molecule (Fig. 3.18b).

The photocatalytic fixation of dinitrogen was also observed in the presence ofnanocrystalline iron titanate and confirmed unambiguously by the isotopic studies[129]. Iron titanate also exhibited a suppressed activity in the undesirable process ofthe photocatalytic NH3 oxidation as compared with pure titania.

Hydrated iron oxide NPs stabilized in the cavities of Nafion membranes werefound to photocatalyze a nitrogen fixation cycle involving both the N2 reductionand oxidation in aerated aqueous solutions [130]. The photogenerated CB electronsreduce dinitrogen to ammonia, while the VB holes oxidize water to O2 and N2—tonitrite, as shown by the following brutto-equations [130]:

6e�CB + N2 + 6Hþ ! 2NH3; 2H2O + 4hþVB ! O2 + 4Hþ ;

6hþVB + N2 + 4H2O ! 2NO�

2 + 8Hþ ; 2hþVB + N2 + O2 + 2H2O ! 4Hþ + 2NO�

2 :

The photocatalytic N2 reduction is promoted in conditions allowing formulti-electron processes to occur, as the total conversion of a dinitrogen moleculeto two ammonia molecules requires six electrons. To create such conditions various

150 3 Semiconductor-Based Photocatalytic Systems …

approaches are probed including the introduction of electron-collecting co-catalysts,for example, noble metal NPs.

The introduction of a co-catalyst is probably the most straightforward way offacilitating the multi-electron N2 reduction. In particular, the decoration of TiO2/Fe2O3 composites with Pd NPs that can collect the photogenerated electrons resultsin an appreciable increase of the photocatalytic NH3 and N2H4 generation rate(Fig. 3.18a) [128]. Nanostructures of titania with ruthenium, rhodium, palladium,and platinum NPs were reported to have photocatalytic properties in the N2

reduction to ammonia, the catalytic activity of metal NPs decreasing asRu > Rh > Pd > Pt [131]. The activity sequence mirrors the efficiency of primaryseparation of the photogenerated electrons and holes between metal NPs andsemiconductor crystals, respectively.

The silicon NR arrays decorated with Au NPs were shown to reduce atmosphericN2 to ammonia in a photoelectrochemical regime, the ammonia yield increasingunder an elevated N2 pressure [132]. In the presence of sulfite ions as a sacrificialelectron donor, the photoprocess directly yielded ammonia sulfate which is a fer-tilizer of industrial importance.

Exfoliated layered semiconductor materials with a high electron density can alsofavor to multi-electron transfers as exemplified for the ultra-thin MoS2, while bulkmolybdenum disulfide is passive in the nitrogen reduction [133]. Partially exfoli-ated GCN displayed around 5-times higher photocatalytic activity in the N2

reduction to NH3 as compared to the pristine bulk graphitic carbon nitride [122].Concluding the discussion of the photosynthesis-like systems for the photocat-

alytic CO2 and N2 conversion, we define the “hottest” pathways of further progressof this very promising field.

Similarly to the hydrogen photoproduction systems, here a search for newsemiconductor nanomaterials with unexpected properties is of the paramountimportance. For example, a high potential can be noted for the metalorganicframeworks allowing for a precise molecular design and control of the geometry

Fig. 3.18 a Rate of the photocatalytic generation of NH3 and N2H4 from N2 over differentphotocatalysts. Reprinted with permissions from Ref. [128]. Copyright (2017) Elsevier

3.2 Photocatalytic Fixation of Dinitrogen 151

and composition of the active sites, which is even more important for multi-electronCO2 and N2 reduction to desired products, than for the reduction of water to H2.

A spatial design of semiconductor-based nanoheterostructures and nanoassem-blies aimed at the separation of the reduction and oxidation sites and the inhibitionof the re-oxidation of CO2 and N2 reduction products is, therefore, a next importantthing in the progress of such photocatalytic synthetic systems. Such design caninclude studying of various hierarchical structures, like hollow spheres withincorporated co-catalysts and sensitizers, multi-faceted materials with a differentreactivity of faces toward the CO2 and N2 reduction, spatial separation of thereduction and oxidation semi-reactions by transforming a photocatalytic processinto a photoelectrochemical one, etc.

Similarly to the water splitting systems and even to a much higher extent, theefficiency of the photocatalytic reduction of CO2 and N2, as well as the selectivityof these processes, can be influenced by the co-catalysts of multi-electron reactions.The challenge of creating efficient catalysts of concerted 4–8-electron processes isof the multi-disciplinary nature, requiring a convergence of efforts in the photo-chemistry, electrochemistry, and catalysis and promising in the future to make theCO2 and N2 conversion technologies competitive to the presently used catalyticprocesses.

The photocatalytic systems for CO2 and N2 conversion can be enhanced andmodified by doping and/or creating of additional lattice defects—vacancies. Thevacancies can favor to the CO2 and N2 adsorption and, at the same time, vary in theadsorption geometry, thus providing possibilities for the formation of differentproducts and determining not only the efficiency but also the selectivity of thephotocatalytic transformations. Also, as the natural processes of the CO2 and N2

photofixation occur mostly in the living microorganisms, the photocatalytic systemsutilizing both the potential of semiconductor nanomaterials and that ofbio-mimicking approaches (ferments and the analogs of the active ferment centers,bacteria, etc.) can pave the way to very efficient and selective phototransformationsof dinitrogen and carbon dioxide.

References

1. Grätzel M (ed) (1983) Energy resources through photochemistry and catalysis. AcademicPress, Inc, New York

2. Zhang XV, Martin ST, Friend CM et al (2004) Mineral-assisted pathways in prebioticsynthesis: photoelectrochemical reduction of carbon (IV) by manganese sulfide. J Am ChemSoc 126:11247–11253. doi:10.1021/ja0476415

3. Zhang XV, Ellery SP, Friend CM et al (2007) Photodriven reduction and oxidation reactionson colloidal semiconductor particles: implications for prebiotic synthesis. J PhotochemPhotobiol A 185:301–311. doi:10.1016/j.jphotochem.2006.06.025

4. ChittendenGJ, Schwartz AW (1981) Prebiotic photosynthetic reactions. Biosystems 14:15–325. Olson JM, Blankenship RE (2004) Thinking about the evolution of photosynthesis.

Photosynth Res 80:373–386

152 3 Semiconductor-Based Photocatalytic Systems …

6. Shiavello M, Sclafani A (ed) (1985) Photoelectrochemistry, photocatalysis and photoreac-tion. NATO-ASI Series, Reidel, Dordrecht

7. Shiavello M (ed) (1988) Photocatalysis and environment trends and application. Kluwer8. Davies JA, Boucher OL, Edwards JG (1995) The question of artificial photosynthesis of

ammonia on heterogeneous catalysts. In: Neckers OC, Volman DH, von Bünou G(eds) Advances in photochemistry. New York, Wiley

9. Tu W, Zhou Y, Zou Z (2014) Photocatalytic conversion of CO2 into renewable hydrocarbonfuels: state-of-the-art accomplishment, challenges, and prospects. Adv Mater 26:4607–4626.doi:10.1002/adma.201400087

10. Nikokavoura A, Trapalis C (2017) Alternative photocatalysts to TiO2 for the photocatalyticreduction of CO2. Appl Surf Sci 391:149–174. doi:10.1016/j.apsusc.2016.06.172

11. Wen F, Li C (2012) Hybrid artificial photosynthetic systems comprising semiconductors aslight harvesters and biomimetic complexes as molecular cocatalysts. Acc Chem Res46:2355–2364. doi:10.1021/ar300224u

12. Kim W, McClure BA, Edri E, Frei H (2016) Coupling carbon dioxide reduction with wateroxidation in nanoscale photocatalytic assemblies. Chem Soc Rev 45:3221–3243. doi:10.1039/c6cs00062b

13. Roy SC, Varghese OK, Paulose M, Grimes CA (2010) Toward solar fuels: photocatalyticconversion of carbon dioxide to hydrocarbons. ACS Nano 4:1259–1278. doi:10.1021/nn9015423

14. Ozin GA (2015) Throwing new light on the reduction of CO2. Adv Mater 27:1957–1963.doi:10.1002/adma.201500116

15. Qu Y, Duan X (2013) Progress, challenge and perspective of heterogeneous photocatalysts.Chem Soc Rev 42:2568–2580. doi:10.1039/c2cs35355e

16. Chang X, Wang T, Gong J (2016) CO2 photo-reduction: insights into CO2 activation andreaction on surfaces of photocatalysts energy. Environ Sci 9:2177–2196. doi:10.1039/c6ee00383d

17. Chen Y, Lewis NS, Xiang C (2015) Operational constraints and strategies for systems toeffect the sustainable, solar-driven reduction of atmospheric CO2. Energy Environ Sci8:3663–3674. doi:10.1039/c5ee02908b

18. Atwater HA (2016) Photonics and light science wish list from COP21 in Paris. ACSPhotonics 3:155–157. doi:10.1021/acsphotonics.5b00749

19. Izumi Y (2013) Recent advances in the photocatalytic conversion of carbon dioxide to fuelswith water and/or hydrogen using solar energy and beyond. Coord Chem Rev. doi:10.1016/j.ccr.2012.04.018

20. Dhakshinamoorthy A, Navalon S, Corma A, Garcia H (2012) Photocatalytic CO2 reductionby TiO2 and related titanium containing solids. Energy Environ Sci 5:9217–9233. doi:10.1039/c2ee21948d

21. Sun H, Wang S (2014) Research advances in the synthesis of nanocarbon-basedphotocatalysts and their applications for photocatalytic conversion of carbon dioxide tohydrocarbon fuels. Energy Fuels 28:22–36. doi:10.1021/ef401426x

22. Qin S, Xin F, Liu Y et al (2011) Photocatalytic reduction of CO2 in methanol to methylformate over CuO–TiO2 composite catalysts. J Colloid Interface Sci 356:257–261. doi:10.1016/j.jcis.2010.12.034

23. Kanemoto M, Hosokawa H, Wada Y et al (1996) Role of surface in the photoreduction ofcarbon dioxide catalysed by colloidal ZnS nanocrystallites in organic solvent. J Chem SocFaraday Trans 92:2401–2411. doi:10.1039/FT9969202401

24. Zhou R, Guzman MI (2014) CO2 reduction under periodic illumination of ZnS. J PhysChem C 118:11649–11656. doi:10.1021/jp4126039

25. Chen J, Xin F, Qin S, Yin X (2013) Photocatalytically reducing CO2 to methyl formate inmethanol over ZnS and Ni-doped ZnS photocatalysts. Chem Eng J 230:506–512. doi:10.1016/j.cej.2013.06.119

26. Kočí K, Reli M, Kozák O et al (2011) Influence of reactor geometry on the yield of CO2

photocatalytic reduction. Catal Today 176:212–214. doi:10.1016/j.cattod.2010.12.054

References 153

27. Chaudhary YS, Woolerton TW, Allen CS et al (2012) Visible light-driven CO2 reduction byenzyme coupled CdS nanocrystals. Chem Commun 48:58–60. doi:10.1039/c1cc16107e

28. Fujiwara H, Hosokawa H, Murakoshi K et al (1997) Effect of surface structures onphotocatalytic CO2 reduction using quantized CdS nanocrystallites. J Phys Chem B101:8270–8278

29. Jiang W, Yin X, Xin F et al (2014) Preparation of CdIn2S4 microspheres and application forphotocatalytic reduction of carbon dioxide. Appl Surf Sci 288:138–142. doi:10.1016/j.apsusc.2013.09.165

30. Dai W, Xu H, Yu J et al (2015) Photocatalytic reduction of CO2 into methanol and ethanolover conducting polymers modified Bi2WO6 microspheres under visible light. Appl Surf Sci356:173–180. doi:10.1016/j.apsusc.2015.08.059

31. Sasan K, Lin Q, Mao C, Feng P (2016) Open framework metal chalcogenides as efficientphotocatalysts for reduction of CO2 into renewable hydrocarbon fuel. Nanoscale 8:10913–10916. doi:10.1039/c6nr02525k

32. Liu L, Zhao H, Andino JM, Li Y (2012) Photocatalytic CO2 reduction with H2O on TiO2

nanocrystals: comparison of anatase, rutile, and brookite polymorphs and exploration ofsurface chemistry. ACS Catal 2:1817–1828. doi:10.1021/cs300273q

33. Dimitrijevic NM, Vijayan BK, Poluektov OG et al (2011) Role of water and carbonates inphotocatalytic transformation of CO2 to CH4 on titania. J Am Chem Soc 133:3964–3971.doi:10.1021/ja108791u

34. Lee D, Kanai Y (2012) Role of four-fold coordinated titanium and quantum confinement inCO2 reduction at titania surface. J Am Chem Soc 134:20266–20269. doi:10.1021/ja309871m

35. Yu J, Low J, Xiao W et al (2014) Enhanced photocatalytic CO2-reduction activity of anataseTiO2 by coexposed 001 and 101 facets. J Am Chem Soc 136:8839–8842. doi:10.1021/ja5044787

36. Xie S, Wang Y, Zhang Q et al (2014) MgO- and Pt-promoted TiO2 as an efficientphotocatalyst for the preferential reduction of carbon dioxide in the presence of water. ACSCatal 4:3644–3653. doi:10.1021/cs500648p

37. Tong Y, Chen L, Ning S et al (2017) Photocatalytic reduction of CO2 to CO over the Ti–Highly dispersed HZSM-5 zeolite containing Fe. Appl Catal B 203:725–730. doi:10.1016/j.apcatb.2016.10.065

38. Hamdy MS, Amrollahi R, Sinev I et al (2014) Strategies to design efficient silica-supportedphotocatalysts for reduction of CO2. J Am Chem Soc 136:594–597. doi:10.1021/ja410363v

39. Liu E, Kang L, Wu F et al (2014) Photocatalytic reduction of CO2 into methanol over Ag/TiO2 nanocomposites enhanced by surface plasmon resonance. Plasmonics 9:61–70. doi:10.1007/s11468-013-9598-7

40. Tu W, Zhou Y, Li H et al (2015) Au@TiO2 yolk–shell hollow spheres for plasmon-inducedphotocatalytic reduction of CO2 to solar fuel via a local electromagnetic field. Nanoscale7:14232–14236. doi:10.1039/c5nr02943k

41. Tan LL, Ong WJ, Chai SP et al (2013) Reduced graphene oxide-TiO2 nanocomposite as apromising visible-light-active photocatalyst for the conversion of carbon dioxide. NanoscaleRes Lett 8:465. doi:10.1186/1556-276X-8-465

42. Fu J, Cao S, Yu J et al (2014) Enhanced photocatalytic CO2-reduction activity of electrospunmesoporous TiO2 nanofibers by solvothermal treatment. Dalton Trans 43:9158–9165.doi:10.1039/c4dt00181h

43. Xu Q, Yu J, Zhang J et al (2015) Cubic anatase TiO2 nanocrystals with enhancedphotocatalytic CO2 reduction activity. Chem Commun 51:7950–7953. doi:10.1039/c5cc01087j

44. Fresno F, Jana P, Reñones P et al (2017) CO2 reduction over NaNbO3 and NaTaO3

perovskite photocatalysts. Photochem Photobiol Sci 16:17–23. doi:10.1039/c6pp00235h45. Li P, Ouyang S, Xi G et al (2012) The effects of crystal structure and electronic structure on

photocatalytic H2 evolution and CO2 reduction over two phases of perovskite-structuredNaNbO3. J Phys Chem C 116:7621–7628. doi:10.1021/jp210106b

154 3 Semiconductor-Based Photocatalytic Systems …

46. Chen X, Zhou Y, Liu Q et al (2012) Ultrathin, single-crystal WO3 nanosheets bytwo-dimensional oriented attachment toward enhanced photocatalystic reduction of CO2 intohydrocarbon fuels under visible light. ACS Appl Mater Interfaces 4:3372–3377. doi:10.1021/am300661s

47. Zhou Y, Tian Z, Zhao Z et al (2011) High-yield synthesis of ultrathin and uniform Bi2WO6

square nanoplates benefitting from photocatalytic reduction of CO2 into renewablehydrocarbon fuel under visible light. ACS Appl Mater Interfaces 3:3594–3601. doi:10.1021/am2008147

48. Yamamoto M, Yoshida T, Yamamoto N et al (2015) Photocatalytic reduction of CO2 withwater promoted by Ag clusters in Ag/Ga2O3 photocatalysts. J Mater Chem A 3:16810–16816. doi:10.1039/C5TA04815J

49. Kawamura S, Puscasu MC, Yoshida Y et al (2015) Tailoring assemblies of plasmonicsilver/gold and zinc–gallium layered double hydroxides for photocatalytic conversion ofcarbon dioxide using UV-visible light. Appl Catal A 504:238–247. doi:10.1016/j.apcata.2014.12.042

50. Liu Q, Wu D, Zhou Y et al (2014) Single-crystalline, ultrathin ZnGa2O4 nanosheet scaffoldsto promote photocatalytic activity in CO2 reduction into methane. ACS Appl MaterInterfaces 6:2356–2361. doi:10.1021/am404572g

51. Liu Q, Zhou Y, Kou J et al (2010) High-yield synthesis of ultralong and ultrathin Zn2GeO4

nanoribbons toward improved photocatalytic reduction of CO2 into renewable hydrocarbonfuel. J Am Chem Soc 132:14385–14387. doi:10.1021/ja1068596

52. Kang U, Park H (2017) A facile synthesis of CuFeO2 and CuO composite photocatalystfilms for the production of liquid formate from CO2 and water over a month. J MaterChem A 5:2123–2131. doi:10.1039/c6ta09378g

53. Zhang T, Wang X, Huang X et al (2016) Bifunctional catalyst of ametallophthalocyanine-carbon nitride hybrid for chemical fixation of CO2 to cycliccarbonate. RSC Adv 6:2810–2818. doi:10.1039/c5ra21058e

54. Zhang L, Wang W, Jiang D et al (2015) Photoreduction of CO2 on BiOCl nanoplates withthe assistance of photoinduced oxygen vacancies. Nano Res 8:821–831. doi:10.1007/s12274-014-0564-2

55. Peng F, Wang J, Ge G et al (2013) Photochemical reduction of CO2 catalyzed by siliconnanocrystals produced by high energy ball milling. Mater Lett 92:65–67. doi:10.1016/j.matlet.2012.10.059

56. Lin J, Pan Z, Wang X (2014) Photochemical reduction of CO2 by graphitic carbon nitridepolymers. ACS Sustainable Chem Eng 2:353–358. doi:10.1021/sc4004295

57. Xia P, Zhu B, Yu J et al (2017) Ultra-thin nanosheet assemblies of graphitic carbon nitridefor enhanced photocatalytic CO2 reduction. J Mater Chem A 5:3230–3238. doi:10.1039/c6ta08310b

58. Xu HQ, Hu J, Wang D et al (2015) Visible-light photoreduction of CO2 in a metal–organicframework: boosting electron–hole separation via electron trap states. J Am Chem Soc137:13440–13443. doi:10.1021/jacs.5b08773

59. Sun D, Gao Y, Fu J et al (2015) Construction of a supported Ru complex on bifunctionalMOF-253 for photocatalytic CO2 reduction under visible light. Chem Commun 51:2645–2648. doi:10.1039/C4CC09797A

60. Yan S, Ouyang S, Xu H et al (2016) Co-ZIF-9/TiO2 nanostructure for superior CO2

photoreduction activity. J Mater Chem A 4:15126–15133. doi:10.1039/C6TA04620G61. Fei H, Sampson MD, Lee Y et al (2015) Photocatalytic CO2 reduction to formate using a Mn

(I) molecular catalyst in a robust metal-organic framework. Inorg Chem 54:6821–6828.doi:10.1021/acs.inorgchem.5b00752

62. Choi KM, Kim D, Rungtaweevoranit B et al (2017) Plasmon-enhanced photocatalytic CO2

conversion within metal-organic frameworks under visible light. J Am Chem Soc 139:356–362. doi:10.1021/jacs.6b11027

References 155

63. Qin G, Zhang Y, Ke X et al (2013) Photocatalytic reduction of carbon dioxide to formicacid, formaldehyde, and methanol using dye-sensitized TiO2 film. Appl Catal B 129:599–605. doi:10.1016/j.apcatb.2012.10.012

64. Jia Y, Xu Y, Nie R et al (2017) Artificial photosynthesis of methanol from carbon dioxideand water via a Nile red-embedded TiO2 photocathode. J Mater Chem A 5:5495–5501.doi:10.1039/c6ta10231j

65. Sahara G, Kumagai H, Maeda K et al (2016) Photoelectrochemical reduction of CO2

coupled to water oxidation using a photocathode with a Ru(II)–Re(I) complex photocatalystand a CoOx/TaON photoanode. J Am Chem Soc 138:14152–14158. doi:10.1021/jacs.6b09212

66. Kumagai H, Sahara G, Maeda K, Higaghi M, Abe R, Ishitani O (2017) Hybrid photocathodeconsisting of a CuGaO2 p-type semiconductor and a Ru(II)–Re(I) supramolecularphotocatalyst: non-biased visible-light-driven CO2 reduction with water oxidation. ChemSci 8:4242–4249. doi:10.1039/C7SC00940B

67. Jiang M, Gao Y, Wang Z, Ding Z (2016) Photocatalytic CO2 reduction promoted by aCuCo2O4 cocatalyst with homogeneous and heterogeneous light harvesters. Appl Catal B198:180–188. doi:10.1016/j.apcatb.2016.05.055

68. Suzuki TM, Nakamura T, Saeki S et al (2012) Visible light-sensitive mesoporous N-dopedTa2O5 spheres: synthesis and photocatalytic activity for hydrogen evolution and CO2

reduction. J Mater Chem 22:24584–24590. doi:10.1039/c2jm33980c69. Qian W, Wei W, Jianfeng C et al (2012) Novel synthesis of ZnPc/TiO2 composite particles

and carbon dioxide photo-catalytic reduction efficiency study under simulated solar radiationconditions. Colloid Surfaces A 409:118–125. doi:10.1016/j.colsurfa.2012.06.010

70. Finkelstein-Shapiro D, Petrosko SH, Dimitrijevic NM et al (2013) CO2 preactivation inphotoinduced reduction via surface functionalization of TiO2 nanoparticles. J Phys ChemLett 4:475–479. doi:10.1021/jz3020327

71. Zhao G, Pang H, Liu G et al (2017) Co-porphyrin/carbon nitride hybrids for improvedphotocatalytic CO2 reduction under visible light. Appl Catal B 200:141–149. doi:10.1016/j.apcatb.2016.06.074

72. Walsh JJ, Jiang C, Tang J, Cowan AJ (2016) Photochemical CO2 reduction usingstructurally controlled g-C3N4. Phys Chem Chem Phys 18:24825–24829. doi:10.1039/c6cp04525a

73. Kuriki R, Ishitani O, Maeda K (2016) Unique solvent effects on visible-light CO2 reductionover ruthenium(II)-complex/carbon nitride hybrid photocatalysts. ACS Appl MaterInterfaces 8:6011–6018. doi:10.1021/acsami.5b11836

74. Sato S, Arai T, Morikawa T et al (2011) Selective CO2 conversion to formate conjugatedwith H2O oxidation utilizing semiconductor/complex hybrid photocatalysts. J Am Chem Soc133:15240–15243. doi:10.1021/ja204881d

75. Yan Y, Yu Y, Huang S et al (2017) Adjustment and matching of energy band of TiO2-basedphotocatalysts by metal ions (Pd, Cu, Mn) for photoreduction of CO2 into CH4. J PhysChem C 121:1089–1098. doi:10.1021/acs.jpcc.6b07180

76. Tahir M, Amin NS (2013) Photocatalytic CO2 reduction and kinetic study over In/TiO2

nanoparticles supported microchannel monolith photoreactor. Appl Catal A 467:483–496.doi:10.1016/j.apcata.2013.07.056

77. Tahir M, Amin NS (2015) Indium-doped TiO2 nanoparticles for photocatalytic CO2

reduction with H2O vapors to CH4. Appl Catal B 162:98–109. doi:10.1016/j.apcatb.2014.06.037

78. Ola O, Maroto-Valer MM (2014) Role of catalyst carriers in CO2 photoreduction overnanocrystalline nickel loaded TiO2-based photocatalysts. J Catal 309:300–308. doi:10.1016/j.jcat.2013.10.016

79. Zhao C, Liu L, Zhang Q et al (2012) Photocatalytic conversion of CO2 and H2O to fuels bynanostructured Ce–TiO2/SBA-15 composites. Catal Sci Technol 2:2558–2568. doi:10.1039/c2cy20346d

156 3 Semiconductor-Based Photocatalytic Systems …

80. Yang HC, Lin HY, Chien YS et al (2009) Mesoporous TiO2/SBA-15, and Cu/TiO2/SBA-15composite photocatalysts for photoreduction of CO2 to methanol. Catal Lett 131:381–387.doi:10.1007/s10562-009-0076-y

81. Matějová L, Kočíb K, Reli, et al (2014) Preparation, characterization and photocatalyticproperties of cerium doped TiO2: on the effect of Ce loading on the photocatalytic reductionof carbon dioxide. Appl Catal B 152–153:172–183. doi:10.1016/j.apcatb.2014.01.015

82. Wang Y, Xu Y, Wang Y et al (2016) Synthesis of Mo-doped graphitic carbon nitridecatalysts and their photocatalytic activity in the reduction of CO2 with H2O. Catal Commun74:75–79. doi:10.1016/j.catcom.2015.10.029

83. Ong WJ, Tan LL, Chai SP et al (2014) Self-assembly of nitrogen-doped TiO2 with exposed001 facets on a graphene scaffold as photo-active hybrid nanostructures for reduction ofcarbon dioxide to methane. Nano Res 7:1528–1547. doi:10.1007/s12274-014-0514-z

84. Tan LL, Ong WJ, Chai SP, Mohamed AR (2014) Band gap engineered, oxygen-rich TiO2

for visible light induced photocatalytic reduction of CO2. Chem Commun 50:6923–6926.doi:10.1039/C4CC01304B

85. Lu D, Zhang M, Zhang Z et al (2014) Self-organized vanadium and nitrogen co-dopedtitania nanotube arrays with enhanced photocatalytic reduction of CO2 into CH4. NanoscaleRes Lett 9:272. doi:10.1186/1556-276X-9-272

86. Liu S, Wang J, Yu J (2016) ZIF-8 derived bimodal carbon modified ZnO photocatalysts withenhanced photocatalytic CO2 reduction performance. RSC Adv 6:59998–60006. doi:10.1039/C6RA11264A

87. Xiong Z, Lei Z, Kuang CC et al (2017) Selective photocatalytic reduction of CO2 into CH4

over Pt-Cu2O TiO2 nanocrystals: the interaction between Pt and Cu2O cocatalysts. ApplCatal B 202:695–703. doi:10.1016/j.apcatb.2016.10.001

88. Chang X, Wang T, Zhang P et al (2016) Stable aqueous photoelectrochemical CO2 reductionby a Cu2O dark cathode with improved selectivity for carbonaceous products. Angew ChemInt Ed 55:8840–8845. doi:10.1002/anie.201602973

89. Wang Y, Li B, Zhang C et al (2013) Ordered mesoporous CeO2-TiO2 composites: highlyefficient photocatalysts for the reduction of CO2 with H2O under simulated solar irradiation.Appl Catal B 130–131:277–284. doi:10.1016/j.apcatb.2012.11.019

90. Zhao J, Wang Y, Li Y et al (2016) Phase-dependent enhancement for CO2 photocatalyticreduction over CeO2/TiO2 catalysts. Catal Sci Technol 6:7967–7975. doi:10.1039/C6CY01365A

91. Jiao J, Wei Y, Zhao Z et al (2014) Photocatalysts of 3D ordered macroporous TiO2-supported CeO2 nanolayers: design, preparation, and their catalytic performances for thereduction of CO2 with H2O under simulated solar irradiation. Ind Eng Chem Res 53:17345–17354. doi:10.1021/ie503333b

92. Reli M, Huo P, Šihor M et al (2016) Novel TiO2/C3N4 photocatalysts for photocatalyticreduction of CO2 and for photocatalytic decomposition of N2O. J Phys Chem A 120:8564–8573. doi:10.1021/acs.jpca.6b07236

93. Shi H, Zhang C, Zhou C, Chen G (2015) Conversion of CO2 into renewable fuel over Pt–g-C3N4/KNbO3 composite photocatalyst. RSC Adv 5:93615–93622. doi:10.1039/c5ra16870h

94. Shi H, Chen G, Zhang C, Zou Z (2014) Polymeric g-C3N4 coupled with NaNbO3 nanowirestoward enhanced photocatalytic reduction of CO2 into renewable fuel. ACS Catal 4:3637–3643. doi:10.1021/cs500848f

95. Zhang X, Wang L, Du Q et al (2016) Photocatalytic CO2 reduction over B4C/C3N4 withinternal electric field under visible light irradiation. J Colloid Interface Sci 464:89–95.doi:10.1016/j.jcis.2015.11.022

96. Cao SW, Liu XF, Yuan YP et al (2014) Solar-to-fuels conversion over In2O3/g-C3N4 hybridphotocatalysts. Appl Catal B 147:940–946. doi:10.1016/j.apcatb.2013.10.029

97. Li M, Zhang L, Fan X et al (2017) Core-shell LaPO4/g-C3N4 nanowires for highly active andselective CO2 reduction. Appl Catal B 201:629–635. doi:10.1016/j.apcatb.2016.09.004

References 157

98. Li K, Peng B, Jin J et al (2017) Carbon nitride nanodots decorated brookite TiO2 quasinanocubes for enhanced activity and selectivity of visible-light-driven CO2 reduction. ApplCatal B 203:910–916. doi:10.1016/j.apcatb.2016.11.001

99. Zou JP, Wu DD, Luo J et al (2016) A strategy for one-pot conversion of organic pollutantsinto useful hydrocarbons through coupling photodegradation of MB with photoreduction ofCO2. ACS Catal 6:6861–6867. doi:10.1021/acscatal.6b01729

100. Yang Z, Wang H, Song W et al (2017) One dimensional SnO2 NRs/Fe2O3 NTs with dualsynergistic effects for photoelectrocatalytic reduction CO2 into methanol. J Colloid InterfaceSci 486:232–240. doi:10.1016/j.jcis.2016.09.055

101. Sakimoto KK, Zhang SJ, Yang P (2016) Cysteine–cystine photoregeneration for oxygenicphotosynthesis of acetic acid from CO2 by a tandem inorganic–biological hybrid system.Nano Lett 16:5883–5887. doi:10.1021/acs.nanolett.6b02740

102. Li X, Liu H, Luo D et al (2012) Adsorption of CO2 on heterostructure CdS(Bi2S3)/TiO2

nanotube photocatalysts and their photocatalytic activities in the reduction of CO2 tomethanol under visible light irradiation. Chem Eng J 180:151–158. doi:10.1016/j.cej.2011.11.029

103. Wei W, Yang Z, Song W et al (2017) Different CdSeTe structure determined photoelec-trocatalytic reduction performance for carbon dioxide. J Colloid Interface Sci 496:327–333.doi:10.1016/j.jcis.2016.11.054

104. Wang C, Thompson RL, Baltrus J, Matranga C (2010) Visible light photoreduction of CO2

using CdSe/Pt/TiO2 heterostructured catalysts. J Phys Chem Lett 1:48–53. doi:10.1021/jz9000032

105. Su Y, Zhang Z, Liu H, Wang Y (2017) Cd0.2Zn0.8S@UiO-66-NH2 nanocomposites asefficient and stable visible-light-driven photocatalyst for H2 evolution and CO2 reduction.Appl Catal B 200:448–457. doi:10.1016/j.apcatb.2016.07.032

106. Cherkasov N, Ibhadon AO, Fitzpatrick P (2015) A review of the existing and alternativemethods for greener nitrogen fixation. Chem Eng Proc 90:24–33. doi:10.1016/j.cep.2015.02.004

107. Schrautzer GN, Strampach N, Hui LN et al (1983) Nitrogen photoreduction on desert sandsunder sterile conditions. Proc Natl Acad Sci USA 80:3873–3876

108. Ma H, Shi Z, Li Q, Liu N (2016) Large-scale production of graphitic carbon nitride withoutstanding nitrogen photofixation ability via a convenient microwave treatment. Appl SurfSci 379:309–315. doi:10.1016/j.apsusc.2016.04.085

109. Ma H, Shi Z, Li Q, Li S (2016) Preparation of graphitic carbon nitride with large specificsurface area and outstanding N2 photofixation ability via a dissolve-regrowth process. J PhysChem Sol 99:51–58. doi:10.1016/j.jpcs.2016.08.008

110. Dong G, Ho W, Wang C (2015) Selective photocatalytic N2 fixation dependent on g-C3N4

induced by nitrogen vacancies. J Mater Chem A 3:23435–23441. doi:10.1039/c5ta06540b111. Raevskaya AE, Panasiuk YV, Korzhak GV et al (2017) Photocatalytic H2 production from

aqueous solutions of hydrazine and its derivatives in the presence of nitric-acid-activatedgraphitic carbon nitride. Catal Today 284:229–235. doi:10.1016/j.cattod.2016.12.024

112. Wu G, Gao Y, Zheng B (2016) Template-free method for synthesizing sponge-like graphiticcarbon nitride with a large surface area and outstanding nitrogen photofixation abilityinduced by nitrogen vacancies. Ceram Internat 42:6985–6992. doi:10.1016/j.ceramint.2016.01.086

113. Hu S, Chen X, Li Q et al (2016) Effect of sulfur vacancies on the nitrogen photofixationperformance of ternary metal sulfide photocatalysts. Catal Sci Technol 6:5884–5890. doi:10.1039/c6cy00622a

114. Hu S, Li Y, Li F et al (2016) Construction of g-C3N4/Zn0.11Sn0.12Cd0.88S1.12 hybridheterojunction catalyst with outstanding nitrogen photofixation performance induced bysulfur vacancies. ACS Sustainable Chem Eng 4:2269–2278. doi:10.1021/acssuschemeng.5b01742

158 3 Semiconductor-Based Photocatalytic Systems …

115. Zhang Q, Hu S, Fan Z et al (2016) Preparation of g-C3N4/ZnMoCdS hybrid heterojunctioncatalyst with outstanding nitrogen photofixation performance under visible light viahydrothermal post-treatment. Dalton Trans 45:3497–3505. doi:10.1039/c5dt04901f

116. Cao Y, Hu S, Li F et al (2016) Photofixation of atmospheric nitrogen to ammonia with anovel ternary metal sulfide catalyst under visible light. RSC Adv 6:49862–49867. doi:10.1039/c6ra08247e

117. Banerjee A, Yuhas BD, Margulies EA et al (2015) Photochemical nitrogen conversion toammonia in ambient conditions with FeMoS-chalcogels. J Am Chem Soc 137:2030–2034.doi:10.1021/ja512491v

118. Li H, Shang J, Shi J et al (2016) Facet-dependent solar ammonia synthesis of BiOClnanosheets via a proton-assisted electron transfer pathway. Nanoscale 8:1986–1993. doi:10.1039/c5nr07380d

119. Li H, Shang J, Ai Z, Zhang L (2015) Efficient visible light nitrogen fixation with BiOBrnanosheets of oxygen vacancies on the exposed 001 facets. J Am Chem Soc 137:6393–6399.doi:10.1021/jacs.5b03105

120. Sun S, An Q, Wang W et al (2017) Efficient photocatalytic reduction of dinitrogen toammonia on bismuth monoxide quantum dots. J Mater Chem A 5:201–209. doi:10.1039/c6ta09275f

121. Taqui Khan MM, Rao NN (1991) Stepwise reduction of coordinated dinitrogen to ammoniavia diazinido and hydrazido intermediates on a visible light irradiated Pt /CdS � Ag2S/RuO2

particulate system suspended in an aqueous solution of K[Ru(EDTA-H)Cl]2H2O.J Photochem Photobiol A 56:101–111

122. Hu S, Zhang W, Bai J et al (2016) Construction of a 2D/2D g-C3N4/rGO hybrid heterojunc–tion catalyst with outstanding charge separation ability and nitrogen photofixationperformance via a surface protonation process. RSC Adv 6:25695–25702. doi:10.1039/c5ra28123g

123. Ogawa T, Kitamura T, Shibuya T, Hoshino K (2004) Characterization and materialconditions of conducting polymer/titanium oxide hybrid systems used for dinitrogen fixationunder ordinary pressure and temperature. Electrochem Commun 6:55–60. doi:10.1016/j.elecom.2003.10.015

124. Zhao W, Zhang J, Zhu X et al (2014) Enhanced nitrogen photofixation on Fe-doped TiO2

with highly exposed (1 0 1) facets in the presence of ethanol as scavenger. Appl Catal B144:468–477. doi:10.1016/j.apcatb.2013.07.047

125. Hu S, Chen X, Li Q et al (2017) Fe3+ doping promoted N2 photofixation ability ofhoneycombed graphitic carbon nitride: the experimental and density functional theorysimulation analysis. Appl Catal B 201:58–69. doi:10.1016/j.apcatb.2016.08.002

126. Hu S, Qu X, Bai J, Li P, Li Q, Wang F, ong L (2017) Effect of Cu(I)–N active sites on the N2

photofixation ability over flowerlike copper-doped g-C3N4 prepared via a novel moltensalt-assisted microwave process: the experimental and density functional theory simulationanalysis. ACS Sustainable Chem Eng 5:6863–6872. doi:10.1021/acssuschemeng.7b01089

127. Yue C, Trudeau ML, Antonelli D (2005) Mesoporous tantalum oxide photocatalysts forSchrauzer-type conversion of dinitrogen to ammonia. Canadian J Chem 83:308–314. doi:10.1139/v05-018

128. Lashgaria M, Zeinalkhania P (2017) Photocatalytic N2 conversion to ammonia usingefficient nanostructured solar-energy-materials in aqueous media: a novel hydrogenationstrategy and basic understanding of the phenomenon. Appl Catal A 529:91–97. doi:10.1016/j.apcata.2016.10.017

129. Linnik O, Kisch H (2006) On the mechanism of nitrogen photofixation at nanostructurediron titanate films. Photochem Photobiol Sci 5:938–942. doi:10.1039/b608396j

130. Tennakone K, Ileperuma OA, Bandara JMS et al (1991) Simultaneous reductive andoxidative photocatalytic nitrogen fixation in hydrous iron(iii) oxide loaded nafion films inaerated water. J Chem Soc Chem Commun 579–580

131. Ranjit KT, Varadarajan TK, Viswanathan B (1996) Photocatalytic reduction of dinitrogen toammonia over noble-metal-loaded TiO2. J Photochem Photobiol, A 96:181–185

References 159

132. Ali M, Zhou F, Chen K et al (2016) Nanostructured photoelectrochemical solar cell fornitrogen reduction using plasmon-enhanced black silicon. Nat Commun 7:11335. doi:10.1038/ncomms11335

133. Sun S, Li X,WangW et al (2017) Photocatalytic robust solar energy reduction of dinitrogen toammonia on ultrathin MoS2. Appl Catal B 200:323–329. doi:10.1016/j.apcatb.2016.07.025

134. Wang W, Xu D, Cheng B et al (2017) Hybrid carbon@TiO2 hollow spheres with enhancedphotocatalytic CO2 reduction activity. J Mater Chem A 5:5020–5029. doi:10.1039/c6ta11121a

160 3 Semiconductor-Based Photocatalytic Systems …

Chapter 4Semiconductor-Based Liquid-JunctionPhotoelectrochemical Solar Cells

4.1 Principles and Designs of SemiconductorNP-Sensitized Solar Cells

The photoelectrochemical light-harvesting systems constitute an important part ofthe assay of available solar light conversion approaches, along with the photo-voltaic light conversion and endothermal photochemical reactions such as thehydrogen production, CO2 reduction, etc. [1–12]. Today, the realm ofsemiconductor-based solar cells is dominated (up to 85%) by “classic” photovoltaicsystems based on single-crystal and polycrystalline silicon with a light conversionefficiency reaching 14–19 and 8–10%, respectively [2, 4, 12]. At the same time, ahigh price of the single-crystalline Si stimulates a search for alternative technologiesbased on more available materials, such as amorphous silicon, thin-filmCdTe-based heterostructures [2, 12], organic conjugated polymers [2, 13–15],liquid-junction solar cells [2, 4–11], etc.

The photovoltaic light-harvesting systems are contingently categorized into thefirst-generation, second-generation, and third-generation solar cells depending onthe operating principles [4, 8]. The first-generation group encompasses “classic”Si-based cells [4, 12] with a p/n junction responsible for the separation of thephotogenerated charge carriers. Due to fundamental reasons, such as an indirectnature of electron transitions, the fabrication of inexpensive thin-film Si-based solarcells that can be implemented on the broadest scale, is impossible.

This problem is solved partially in the second-generation solar cells based onsemiconductor thin films coupled to an optically transparent electrode (OTE) and acounter electrode. The most vivid example of the second-generation solar cell is an“n-CdS/p-CdTe” system [2, 12]. Today, the second-generation cells occupy around15% of the solar cell market. The high light conversion efficiencies can be achievedin such systems by combining several p/n junctions, however, the multi-layer cellsare expensive and can be rationally used in specific applications, for example, in theaerospace field [2]. Also, the production of thin-film solar cells requires

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_4

161

high-precision technologies and corresponding equipment, for example, thegas-phase or molecular-beam epitaxy, the magnetron sputtering, etc. [2, 7, 12].

The third-generation solar light harvesters include solar cells with nanocrys-talline semiconductor electrodes, in particular, the nanostructured metal oxides,such as TiO2, ZnO, and SnO2 attached to the surface of various OTEs [4, 6]. Themetal oxide semiconductors have relatively large band gaps (Eg > 3 eV) and thusthey are capable of harvesting only a small portion of the solar light. To cover thevisible spectral range the metal oxide electrodes should be sensitized by variouscompounds that can efficiently absorb visible (and near-IR) light and transfer thephotogenerated charge carriers to a wide-bandgap component [7, 8]. Typically themetal oxides can be sensitized by organic dyes and strongly-absorbing metalcomplexes in the so-called dye-sensitized solar cells (DSSCs) or, alternatively, byNPs of a narrow-bandgap semiconductor capable of the visible light absorption—inthe semiconductor-sensitized solar cells (SSSCs).

Both types of cells are composed of a light-harvesting photoanode, a counterelectrode and liquid electrolyte containing a redox couple capable of the electrondonation to the photoanode and recovering its original state on the counter electrodeby accepting an electron. Alternatively, an acceptor can be reduced on a pho-toexcited photocathode and then oxidized on a counter electrode (CE), thus com-pleting the light-harvesting cycle. The cells of both types are typically referred to as“the liquid-junction solar cells”. Primary photoinduced processes in such solar cellsinclude the photoinduced charge separation and the oxidation/reduction of a dis-solved substrate, resembling the photocatalytic semiconductor-driven processes. Asa result, the range of substances used both as the wide-bandgap matrices (scaffolds),the dyes and complexes used as sensitizers and the narrow-bandgap NPs are typicaland similar both for the photocatalytic reactions and for the photoelectrochemical(PEC) third-generation liquid-junction solar cells. The energy criteria used for theselection of appropriate components of a PEC system are also very similar to thoseapplied in the semiconductor-based photocatalytic systems and require an energycorrespondence between the CB level of a narrow-bandgap sensitizer (or a LUMOlevel of the photoexcited dye-sensitizer), the CB level of a wide-bandgap oxidescaffold, and the redox level of the electron “shuttling” couple present in theelectrolyte (Fig. 4.1a).

Alternatively, the nanocrystalline semiconductors can be coupled with othermaterials, such as the conjugated polymers, fullerene derivatives, organicoligo-dyes, organic-inorganic perovskites, etc., forming solid p/n-junctions wherethe photoinduced separation of electron and hole can occur, similarly to thesecond-generation thin-film solar cells. Such cells are referred to as “the bulkheterojunction solar cells” [8]. The highest reported light-conversion efficiency forthe bulk-heterojunction solar cells is around 11% [13]. As we focus in this book onthe photochemical light-harvesting, that is, the processes involving chemicaltransformations of the participants, we will focus predominantly on theliquid-junction SSSCs, where the solar light energy conversion occurs as a result ofconcerted photochemical/chemical transformations of the cell components.

162 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Fig. 4.1 a Energy diagram illustrating the band positions of some semiconductor materialstypically used in the third-generation solar cells relative to the redox levels of some popularelectron-shuttling redox-couples; b A working principle of the dye-sensitized liquid-junction solarcell with a titania-based photoanode. Reprinted with permissions from Ref. [11]. Copyright (2001)Nature Publishing Group

4.1 Principles and Designs of Semiconductor … 163

The DSSC action is based on the photoinduced electron transfer from a pho-toexcited dye-sensitizer into the conduction band (or some mid-bandgapsurface-related states) of the wide-band-gap metal oxide (TiO2, ZnO, SnO2) andthen—to the electric circuit (Fig. 4.1a) [11, 16]. The one-electron oxidizeddye-sensitizer is reduced by a component of the redox couple present in solution.Typically, the iodide/iodate I�

�IO�

3 couple is used in the DSSCs. In this case, thedye recovers its original state by oxidizing iodide to iodate and then IO�

3 ions arereduced to I− again on a counter electrode (typically Pt, Au, Ag, etc.), thus finishingthe PEC cycle. The highest light conversion efficiency, 12%, was achieved for themesoporous titania scaffolds sensitized by RuL(NCS)3 complexes, where L is abipyridyl-based ligand [16].

In SSSCs the solar light is harvested by NPs of a narrow-bandgap semiconductor(the term “narrow” here is relative, it is used rather to distinguish suchvisible-light-sensitive NPs from the wide-bandgap metal oxide materials), forexample, CdS, CdSe, CuInS2, or InP [4–9, 17, 18]. The sensitizer NPs absorbvisible light quanta with the energy higher than the bandgap resulting in an electroncoming from VB to CB (Fig. 4.2a). Then, the CB electron is transferred from thesensitizer to the metal oxide scaffold and then—to the electric circuit and, finally, tothe counter electrode. The transfer is only possible if the CB level of the sensitizerNPs is higher than the CB of the wide-bandgap component (Fig. 4.2b).A photogenerated VB hole of the sensitizer NP is filled with an electron from aredox couple component (in this case from sulfide ions) producing an oxidized formof the shuttle (elemental sulfur). The shuttle is then regenerated on thecounter-electrode accepting an electron and finishing the PEC cycle. The holetransfer from the sensitizer to the scaffold VB is impossible, as the VB level of themetal oxide resides deeper than the sensitizer VB level (Fig. 4.2b). In this way, thephotogenerated electron and hole are reliably separated between the photoanodecomponents [8, 18].

Fig. 4.2 Schemes illustrating a action principle of a liquid-junction solar cell sensitized byvisible-light-sensitive semiconductor NPs and b energy diagram of a cell comprising TiO2, CdS,S2−/S0 electron-shuttling couple and a counter electrode

164 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

The SSSCs have a number of advantages over DSSCs, in particular, (i) a broadvariability of the electron parameters (Eg, CB and VB level positions) as a result ofsize and composition variations of the sensitizer NPs [4–9, 17, 18]; (ii) the possi-bility of multi-exciton generation in some narrow-bandgap NP materials, in par-ticular PbS, PbSe, PbTe, CdSe, InAs, InP, CdTe, and Si [10]; (iii) a more robustelectron contact between the scaffold and the sensitizer NPs as compared tomolecular sensitizers, that are typically adsorbed via the surface bridge OH groups[5].

Efficient spatial separation of the photogenerated electrons and holes is the mostimportant condition to be met by a binary semiconductor heterostructure to be anefficient SSSC photoanode. For this, the heterostructure should have the mutual CBand VB positions similar to those depicted in Fig. 4.2b [4–9]. In real systems, evenif the basic energy condition is met, concurrent recombination processes alwaystook place, resulting in a loss of the light conversion efficiency. The losses areaccounted for by the electron-hole recombination in the sensitizer NPs precedingthe electron transfer, by the charge capturing in the surface traps of sensitizer NPs,by the recombination of the injected electron with components of the electrolyte, byside photocatalytic reactions, etc.

The liquid-junction SSSCs provide also a number of advantages over the bulkheterojunction solar cells, where a donor and an acceptor contact directly. In par-ticular, the liquid-junction SSSCs do not suffer from the “non-ideality” of theheterojunction due to the presence of a liquid electrolyte that envelops the entiresurface of the electrodes and ensures a good electric contact between the pho-toanode and the counter-electrode. Also, the liquid-junction SSSCs can be producedin a relatively simple way without any high-precision equipment orelaborate/unique laboratory techniques, such as the high vacuum, ultra-cleanenvironment, ultra-high-pure semiconductors, etc. At the same time, theliquid-junction SSSCs suffer from a relatively low chemical stability [4].

The most widely used wide-bandgap scaffold is mesoporous/nanocrystallinetitania that is characterized by a chemical stability, a high electron mobility and lowrecombinational losses [3–9, 18, 19]. Another popular metal oxide scaffold is zincoxide, that exhibits a number of quite unique properties including spectacularquantum size effects, the capability of photoinduced charge accumulation, a highphotoactivity and relative simplicity of preparation. However, ZnO is chemicallyunstable as compared to TiO2 and suffers from degradation in acidic/basic solutionsas well as in the presence of sulfide/polysulfide electrolyte. At the same time, thisinstability can be exploited to modify ZnO scaffolds or to convert them in binaryheterostructures, as will be shown in this chapter later. The available syntheses ofnanocrystalline ZnO also provide a virtually unlimited variety of morphologies andgeometries—from single-crystalline nanorods (NRs) to intricate ordered 3Dstructures [19, 20].

The most popular narrow-bandgap sensitizers for the liquid-junction SSSCs aremetal sulfideNPs (CdS, PbS,CuInS2, AgInS2),metal selenideNPs (CdSe, CdSxSe1−x,CdSexTe1−x, PbSe), and binary metal chalcogenide nanocomposites (CdS/PbS,CdS/CdSe, CdS/CuInS2, etc.). Starting from 2009 the feasibility of using

4.1 Principles and Designs of Semiconductor … 165

organic-inorganic Pb-based perovskites CH3NH3PbHal3 (Hal—Br, I) as spectralsensitizers for the liquid-junction solar cells was shown [21]. Today, the perovskitesolar cells form an independent and rapidly developing branch of the photovoltaicsreaching the light conversion efficiencies of around 18% [22–26]. However, theperovskite-based materials also suffer from chemical/photochemical instability andcontain toxic lead making the solar cells recycling a challenge still to be properly met.

In the SSSCs based on the narrow-bandgap metal chalcogenide NPs the highestefficiencies of light conversions were achieved with aqueous sulfide/polysulfideelectrolytes [4–9, 18]. The S2�

�S2�x electron-shuttling couple ensures high pho-

tocurrents and photovoltages in such SSSCs and simultaneously inhibits theundesirable photocorrosion of the sensitizer NPs. Along with S2�

�S2�x , other redox

couples are constantly probed, including I��I�3 , Fe CNð Þ2�6

.Fe CNð Þ3�6 ,

ðCo o - phenð Þ2þ3.ðCo o - phenð Þ3þ3 , etc. [4]. The alternative redox couples are

typically introduced when the sulfide/polysulfide shuttle is impossible to use, forexample, in the case of CH3NH3PbHal3 perovskite or Sb2S3, which dissolves in thepresence of S2− ions [21, 27].

The counter electrodes for the liquid-junction SSSCs are typically selected for aparticular redox couple, because a CE should be catalytically active with respect tothe reduction/oxidation of the electron-shuttling species [28]. The Pt-based counterelectrodes are used for the electrolytes with I�

�I�3 redox-couple, however, in the

polysulfide media platinum is rapidly deactivated as a result of poisoning [4, 18,28]. The highest electrocatalytic activities with respect to the polysulfide elec-trolytes of SSSCs were found for a number of transition metal sulfides, in particular,CoS, CuxS, PbS, and NiS, attached to conductive substrates [28, 29]. Typically,such materials are stable in the presence of S2− and reveal a high electrocatalyticactivity toward oxidation/reduction of the S2�

�S2�x shuttle. A search for new and

more efficient CE materials is constantly performed [30, 31] biasing to morecomplex structures [32, 33], for example, the composites of metal sulfides withgraphene derivatives [34, 35].

4.2 Basic Photoelectrochemical Characteristics of SSSCs

The illumination of a SSSC with the light corresponding to the absorption band ofsensitizer NPs results in the photocurrent generation. The photocurrent density (thecurrent per a surface area) is limited by the radiative recombination in the sensitizerNPs and several types of non-radiative recombination processes involving thesensitizer and metal oxide NPs and the electrolyte components [6, 9]. The photo-generated valence band holes in the sensitizer NPs are filled with electrons from adonating component of the shuttling couple—sulfide anions:

166 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

S2� + 2hþ ! S ð4:1Þ

S + S2�x�1 ! S2�x x = 1. . .7ð Þ ð4:2Þ

The shuttling couple is then regenerated on the surface of a counter electrode:

S2�x + 2e� ! S2�x�1 + S2� ð4:3Þ

The electron migration through the electric circuit that connects a photoanode anda counter electrode results in the current characterized typically by the short-circuitphotocurrent density, Jsc, measured at a zero voltage. The open-circuit photovoltage,Voc, is the second important characteristic of SSSCs measured at J = 0 and corre-sponding to a difference between the Fermi level of the photoanode, EF, and theredox-potential of the shuttling couple in the electrolyte E(Red/Ox) [7, 18]:

Voc ¼ EF�E Red/Oxð Þ: ð4:4Þ

The Fermi level of the photoanode resides between the work function of theconductive OTE and the CB potential of the wide-bandgap oxide scaffold [4, 7, 18].

Typically, the SSSCs are illuminated with a solar light simulator emitting theso-called AM1.5 light flux. The AM1.5 flux is characterized by an intensity of100 mW/cm2 and has a spectral distribution very similar to that of the solar irra-diation near the Earth surface [4–11], but xenon and mercury high-pressure lampsare also used for the SSSC characterizations similarly to the photocatalyticlight-harvesting systems. Figure 4.3a illustrates the solar irradiation spectra justoutside the Earth atmosphere (AM0) and near the Earth surface (AM1.5). The“wells” in the AM1.5 spectrum are associated with the selective absorption of somewavelengths by the atmosphere gases (oxygen, water, CO2) [36]. The figure showsalso a spectrum of a black body heated to 5800 K which is an ideal irradiationspectrum for a solar simulator.

The basic parameter of a SSSC is a total light power conversion efficiency η [7, 8]:

g ¼ ðJsc � Voc � FFÞ=Pin; ð4:5Þ

where FF is the fill factor of the voltage-current characteristics, Pin is the incominglight flux intensity, mW/cm2. All parameters necessary for the calculation of η canbe determined from the voltage-current curve for a given SSSC (Fig. 4.3b) [7, 18].The fill factor can be calculated as

FF = Pactual=Ptheoretical; ð4:6Þ

where Ptheoretical = Jsc � Voc is a theoretically highest possible power for the givenSSSC, while Pactual is the experimentally measured cell power that can be deter-mined as a maximum on the dependence of the cell power on the applied voltage(Fig. 4.3c).

4.2 Basic Photoelectrochemical Characteristics of SSSCs 167

A ratio of the number of photogenerated electrons to the number of absorbedphotons is referred to as a photocurrent quantum yield c orincident-photon-to-current-efficiency, IPCE [7, 18]:

c ¼ 1241� Jsc= k� Pactualð Þ � 100%; ð4:7Þ

where Jsc is presented in A � cm−2, Pactual—in W � cm−2, k—in nanometers. Thevalue of 1241 is a combination of fundamental constants (h � c � 109/e), whereh is the Planck constant (6.62 � 10−34 J � s−1), c is the light velocity in a vacuum(3 � 108 m � s−1), e is the electron charge (1.602 � 10−19 C).

A dependence of IPCE on the excitation wavelength is referred to as the pho-tocurrent quantum yield spectrum (IPCE spectrum or photo-action spectrum).The IPCE spectrum typically coincides with the absorption spectrum of the solarcell photoanode (or the photocathode) [6, 7, 18] and supplies information on thespectral sensitivity range and the efficiency of a photo-electrode. As an example,IPCE spectra of a TiO2/CdSe photoanode are presented in Fig. 4.3d showing that

Fig. 4.3 a Solar spectrum outside the Earth atmosphere (AM0) and near the Earth surface(AM1.5) and a black body irradiation at 5800 K (the ideal spectrum for a solar simulator), duE/dkis a normalized light flux [36]. b, c Characteristics of a SSSC that can be derived from the voltage–photocurrent (b) and power—voltage (c) dependences. (d) IPCE spectra of a TiO2/CdSephotoanode and a ternary TiO2/CdSe/ZnS heterostructure. Reproduced with permissions fromRefs. [36] (a) and [18] (b–d). Copyright (2011, 2012) The Royal Society of Chemistry

168 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

such heterostructure can harvest the solar light up to *700 nm with the efficiencythat increases substantially, by around 50%, when a protective ZnS shell isdeposited onto the surface of CdSe NPs [18].

The ways of the modification of wide-bandgap metal oxide scaffolds withnarrow-bandgap sensitizer NPs are quite versatile and affect considerably the lightconversion efficiency of the final SSSCs. The sensitizer NPs can be producedseparately (ex situ) and adsorbed/deposited from a colloidal solution onto thescaffold surface. Typically the oxide surface is preliminary functionalized by a“bridge” bifunctional molecule that can interact simultaneously with the hydroxy-lated oxide surface and with sensitizer NPs (Fig. 4.4) [7, 18].

In particular, metal oxide surfaces can be modified with mercapto-carboxylicacids HOOC–R–SH (mercaptoacetic, mercaptopropionic, etc.) simply by immers-ing the photoanode into an aqueous acid solution. The carboxylic group is attachedto the oxide surface via hydrogen bonding between –COOH and surface –COHgroups while –SH groups can efficiency interact with the undercoordinated metalions on the surface of metal chalcogenide sensitizer NPs, such as CdTe, CdSe, CdS,PbS, PbSe, etc. Alternatively, the ex situ synthesized sensitizer NPs can bedeposited onto oxide surfaces by the electrophoretic deposition [7, 18, 37].

Chemical bath deposition (CBD) is performed by the immersion of a metaloxide film into a hot solution containing metal and chalcogenide precursors thatform sensitizer NPs during the slow decomposition. Hydrolytically unstable com-pounds are typically used as a chalcogenide source, such as thiourea or thioac-etamide (release of S2−) or sodium selenosulfate (release of Se2−). Slowchalcogenide release allows for the uniform nucleation and controlled growth of thesensitizer NPs.

Successive ionic layer adsorption and reaction (SILAR) is also broadly used forthe preparation of chalcogenide/oxide heterostructures. In this method, a metaloxide film is immersed consecutively into a solution containing metal ions and into

Fig. 4.4 Scheme illustrating the deposition of ex-situ synthesized sensitizer NPs onto the surfaceof oxide scaffold pre-modified with mercaptopropionic acid. Yellow and red circles correspond tothe metal oxide and metal chalcogenide sensitizer NPs, respectively. Reprinted with permissionsfrom Ref. [7]. Copyright (2010) American Chemical Society

4.2 Basic Photoelectrochemical Characteristics of SSSCs 169

a solution containing chalcogenide ions. As a result of multiple repetitions of suchprocedure, the content and size of semiconductor NPs can be increased in a con-trolled manner. Despite the seemingly trivial character of the method, the pho-toanodes produced by SILAR show quite high light conversion efficiencies thusfavoring to the actual domination of this method in the preparation of various SSSCcomponents.

Electrochemical deposition is a convenient and potent method for the formationof chalcogenide/oxide heterostructures based on the electrochemical decompositionof chalcogenide precursors with the release of X2− ions, similarly to the CBDmethod. Metal chalcogenide films produced by the electrodeposition typically showa high adhesion to the oxide surface, while the sensitizer NP size can be tailored byvarying the current density, temperature, and the electrolyte composition.

Chemical vapor deposition (CVD) is based on the gas-phase interaction ofprecursors and nucleation of the sensitizer NPs on a substrate [38]. The method canbe used both for the deposition of metal oxide scaffolds with a precisely controlledmorphology and for the formation of narrow-bandgap sensitizer NPs. The SSSCphotoanodes can also be prepared by spray pyrolysis [39], molecular beam epitaxy[40], and ultrasound-assisted deposition [41].

Recently, photocatalytic deposition was introduced as an emerging method forthe formation of metal chalcogenide (sulfide, selenide) NPs using inherent photo-chemical activity of the most popular TiO2 and ZnO scaffolds [42]. The photode-position showed a broad variability of the sensitizer NP parameters and goodperspectives for the SSSC-related applications.

Finally, good perspectives can be envisaged for various chemical transforma-tions of unstable ZnO scaffolds, for example, ion exchange reactions that canproduce a variety of binary and more complex chalcogenide/oxide heterostructuressuch, for example, as reported in [43] for the preparation of ZnO/ZnxCd1−xSecomposites.

Below in this chapter we will discuss the most popular methods used for theformation of SSSC components—the photoanodes, photocathodes, and counterelectrodes. We should note that the literature on the synthesis of metal-chalcogenideNPs is of enormous volume and the reports discussed below are only a smallfraction of it confined to the examples of using the ex situ produced NPs as spectralsensitizers of SSSCs.

4.3 Nanocrystalline Photoanodes Produced by the Ex SituDeposition of Sensitizer NPs

Deposition of the ex situ synthesized sensitizer NPs. The ex situ deposition isbroadly used for the preparation of photoanodes for the liquid-junction SSSCs [44].Table 4.1 summarizes the PEC parameters of some of the reported SSSCs producedby using the ex situ deposition of sensitizer metal-chalcogenide NPs. In thisapproach, the sensitizer NPs are synthesized separately by using well-known

170 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

synthetic protocols and then deposited onto the surface of oxide scaffold most oftenby using bifunctional molecules-linkers. A typical and very popular linker ismercaptopropionic acid (MPA), HS–CH2CH2–COOH, that can bind strongly to thesurface of titanium (zinc) oxide via the carboxyl group and simultaneously to forma coordination bond with the NP surface cations via the mercapto-group [44].The MPA molecule is short enough to allow electron transfer from the photoexcitedNPs to the wide-bandgap scaffold. Recently, linear aminoalkanoic acids [45] andphosphonoalkanoic acids [46] were introduced as alternatives to the SH-basedmolecular linkers for the attachment of ex situ produced sensitizer NPs.

The attractiveness of the ex situ deposition appears, in the first place, in broadpossibilities of the variation of composition and size of metal-chalcogenide NPs aswell as in the selection of an appropriate molecule-linker. The synthetic approachestypically used to produce NPs are “heating up” and “hot injection” methods [44].The syntheses occur is organic solvents with high boiling temperatures capable ofthe coordination to the surface of growing NPs passivating them against the growthand aggregation. Both methods allow for a precise control over the size and sizedistribution of NPs and, in the case of a shell formation, also over the thickness ofthe shell.

The heating-up method consists in the thermal decomposition of metal andchalcogen precursors (or a single precursor) in high-boiling-point solvents at 180–280 °C [44]. Oleylamine (OLA) is very often used as a reaction medium as it canserve both as a high-boiling-T solvent and as a passivating ligand capable ofcoordination to the NP surface in the form of a monolayer. At the same time, it candissolve sulfur and selenium or other chalcogen precursors, thus acting as a uni-versal reaction medium. A similar role can be played by combinations of oleic acid(OA) and trioctylphosphine (TOP). Another popular composition for the heating upprocedure combines the paraffin as an inert medium, Cd oleate and solutions ofelemental chalcogens (S, Se, Te) in TOP as the NP precursors.

The size of growing NPs is determined by the pyrolysis duration and the NPgrowth can be quenched at any desirable moment by a sharp temperature reduction.In such a way, the size-selected CdSe [47, 48], CdSexTe1−x [49], CuInS2 [50–53]and alloyed ZnSe-AgInSe2 NPs [54] can be prepared (Fig. 4.5a, b). In the case ofmetal sulfide NPs, dodecanthiol (DDT) is often used in various roles—as a solvent,coordinating ligand and sulfur source [50–53]. For example, by varying the dura-tion of heat treatment of CuInS2 (CIS) NPs in DDT from 10 to 90 min the NP sizecan be smoothly increased from 2.9 to 5.3 nm (Fig. 4.5c, d) [51].

In the hot-injection approach, the metal precursors are dissolved in OLA (ormixtures of OLA with octadecene (ODE) or TOP) and kept at an elevated temper-ature [44]. Additional ligands can also be added to the reaction mixture to allow amore precise control of the NP characteristics, such as trioctylphosphine oxide(TOPO) or hexadecylamine (HDA). Then the temperature of the solution is increased(up to 320 °C) to promote the decomposition of a chalcogen precursor dissolved inTOP, which is then rapidly injected thus creating favorable conditions for thehomogeneous NP nucleation. The mixture is then cooled down to 250–270 °C and

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 171

Tab

le4.1

Examples

ofSS

SCsprod

uced

bytheex-situ

sensitizerNPdepo

sitio

n

Photoanode

material

Molecular

linker

Eg(NPs),eV

Counter

electrode

J sc,mA/cm

2Voc,V

FF,%

η,%

Reference

TiO

2/CdT

e/CdS

MAA

1.56

Au

13.60

0.682

413.80

[85]

TiO

2/CdT

e/CdS

eMPA

*1.4

PbS

7.77

0.602

522.42

[66]

ZnO

NW/CdS

eMPA

*2.0

Pt2.10

0.5–

0.6

*30

0.4a

[55]

TiO

2/Au/CdS

e/P3

OT

MAA

*2.0

Pt0.91

0.610

380.66

[56]

TiO

2/CdS

eMAA

*2.0

Cu xS

8.53

0.56

462.21

[73]

TiO

2/CdS

eCys

*2.3

PbS

2.3

0.48

460.83

[77]

TiO

2/CdT

eMPA

2.48

Pt3.61

850

662.02

a[80]

TiO

2/CdS

0.17Se

0.87

MAA

*1.9

Pt8.72

650

39.3

2.23

[86]

TiO

2/CdS

e 0.45Te 0

.55

MPA

1.55

Cu xS

19.35

571

57.5

6.36

[49]

TiO

2/CdS

eSb

S 43−

*2.0

Cu xS

6.17

510

531.67

[57]

TiO

2/CdS

eMPA

2.0

Cu xS

15.93

619

65.8

6.49

[58]

ZnO

NR/CdS

Cys

*2.6

Pt2.42

550

500.67

[78]

TiO

2/CuInS

2S2

−*1.8

Pt4.14

543

49.4

1.11

[50]

TiO

2/CdS

/CuInS

2MAA

*1.6

Carbon

8.21

489

371.47

[82]

TiO

2/CuInS

2:Zn

MPA

*1.4

Cu xS

19.73

580

586.66

[71]

TiO

2/CIS

DDT

*1.6

Cu xS/RGO

10.10

501

472.38

[51]

TiO

2/CIS

(exsitu/in

situ)

MPA

*1.8

Cu xS

7.72

570

421.84

[53]

TiO

2/CIS/CdS

MPA

*1.55

Cu xS

16.9

560

454.2

[69]

TiO

2/CIS

Cys

+MAA

*1.8

Cu xS

12.82

640

544.44

[83]

TiO

2/ZnS

e-AgInS

e 2/CdS

MPA

*1.6

Cu xS

8.8

*500

431.9

[54]

TiO

2/CdS

xSe 1

−x

no–

Cu xS/RGO

11.2

557

513.20

[65]

TiO

2/CdS

eno

*1.83

Pt3.0

524

270.4

[47]

TiO

2/CdS

eNRs

no*1.83

PbS

9.7

564

492.7

[37]

TiO

2/CuInS

2/CdS

no*1.83

Cu xS/RGO

15.65

529

473.91

[52]

Notethetablereportsthehighestη

values

achieved

inthecorrespondingpapers;the

cells

wereillum

inated

with

AM1.5lig

ht(100

mW/cm

2 )ifnotstatedotherw

ise;redoxcoupleisS2

−/S

x2−if

notstatedotherw

ise;in

somecasesascattering

layerw

asappliedon

topof

thephotoanodesto

increase

efficiency

andaZnS

layerw

asdepositedonto

thephotoanode

bySILARto

increase

PEC

efficiency

(see

original

refs.)

I 2/I−redoxcouple

was

used;P3

OTispoly(3-octylthiophene)

The

values

oftotallig

htconversion

efficiency

ηareintentionally

highlig

hted

inbold

characters

172 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

kept at this T for a certain time to let the NPs grow and rapidly cooled to room T toquench the NP growth.

The hot-injection approach allows for a better size and size distribution control,as compared to the heating-up method, as the steps of nuclei formation and sub-sequent growth are separated in time. However, for the synthesis of the NPs of adesirable size a precise control over the reaction duration and temperature isrequired. The method is successfully applied to produce the size-selected NPs ofCdSe [55–64], CdSxSe1−x [65], CdTe [66], PbS [67, 68], CuInS2 [69], AgInS2 [70],Zn-doped CuInS2 [71], and Cu2ZnSnS4 [72]. For example, in this way, by con-trolling the duration of post-injection aging the size-resolved series of 2.3, 2.6, 3.0,and 3.7-nm CdSe NPs can be produced (Fig. 4.6) [60] as well as the size-selectedfractions of 2.9–6.6 nm PbS NPs [67].

After the NP growth quenching a shell of another semiconductor with a widerbandgap can be grown on the NP surface by the second round of injections of theshell material precursors [44]. The repetition of such injections allows to precisely

Fig. 4.5 a, b Absorption spectra of colloidal CdSeTe, CdTe, and CdSe NP (a) and sensitizedTiO2 films (b). Insets: Photographs of colloidal NPs (a) and TiO2-based sensitized photoanodes.c Absorption spectra of the size-selected CuInS2 NP in toluene. d Photographs of TiO2/CuInS2photoanodes with the size-selected sensitizer NPs. Reprinted with permissions from Refs. [49] (a,b) and [51] (c, d). Copyright (2013, 2014) American Chemical Society

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 173

tune the thickness and composition of the protective shell. In this way, a ZnS shellis typically grown on the CdSe NPs [59], and CdSe shell—on CdTe NPs [66].

The metal-chalcogenide NPs produced by both methods are tightly covered withhydrophobic organic ligands (OLA or TOP) and interact weakly with the polarsurface of oxide wide-bandgap materials. To achieve efficient adsorption of the NPson the oxide surface typically a bifunctional bridge-ligand is used as theabove-discussed MPA [44]. The linker is introduced in two ways. The first and amore straightforward way is soaking of the TiO2 scaffold with MPA (or anotherlinker, like MAA [56]) solution followed by a prolonged (60–70 h) incubation ofthe MPA-modified TiO2 electrode in a colloidal NP solution in non-polar solvents(toluene, CHCl3, etc.). In this way, the titania surface can be decorated by CdSe [56,60, 61, 64], CdTe [66], and AgInS2 NPs [70] grown by the hot-injection synthesis.

The amount of adsorbed CdSe NPs and, therefore, the light harvesting capabilityof the TiO2/CdSe photoanode, can be increased considerably (in 5–6 times) by amultiple precipitation/redispersion of HDA/TOPO-stabilized CdSe NPs in toluene.This procedure results in the elimination of residual unbound ligands and partialdesorption of the ligands from the NP surface thus enabling the subsequent inter-action with the MPA-treated TiO2 scaffold [64]. However, removal of the ligand bythe washing procedure also results in the NP aggregation on the titania surface andtherefore some of the adsorbed NPs are not really attached to TiO2 and cannotparticipate efficiently in the charge transfer. Thus, an optimal precipitation/redispersion cycle number exists (2 as reported in [64]) providing a balancebetween the amount of the adsorbed NPs and their aggregation state.

The second way consists in the ex situ attachment of MPA to the NP surface viaa ligand exchange [44]. In this approach, the organic phase containing NPs isbrought into the contact with a polar solution (methanol, dimethylformamide(DMF), water) containing a molecule-linker and then the bi-phase mixture issubjected to vigorous mechanical or ultrasonic shaking. The MPA gradually sub-stitutes the OLA rendering the NP surface polar and transferring the MPA-capped

Fig. 4.6 Absorption spectra of toluene NP colloids (a) and IPCE spectra of TiO2/CdSeheterostructures (c) both containing 3.7, 3.0, 2.6, and 2.3 nm CdSe NPs. b, d Photographs ofcolloidal solutions (b) and TiO2/CdSe films (d) containing the size-selected CdSe NPs. Reprintedwith permissions from Ref. [60]. Copyright (2008) American Chemical Society

174 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

NPs into the polar solvent. Then oxide (TiO2, ZnO) film is immersed into the polarNP-MPA solution and the NPs are efficiently adsorbed. This approach was used todecorate the nanocrystalline TiO2 films with CdSe [58, 63], CdSe/ZnS [59],CdSexTe1−x [49], PbS [67], CIS [69], Zn-doped CIS [71], and alloyed ZnSe–AgInSe2 NPs [54].

The adsorption of mercaptocarboxylate-capped NPs on the surface of TiO2

depends strongly on pH of the NP solution. At neutral pH (around 8) the carboxylgroups of the stabilizers are mostly protonated and the NPs can bind tightly to themesoporous oxide scaffold, thus blocking the surface layer and hindering furtherportions of the NPs from penetration into the bulk of the oxide film [73]. Also, as pHbecomes lower the hydrodynamic size of MPA-capped CdSe NPs is reported toincrease considerably indicating that agglomeration of the NPs takes place, furtherlowering the NP absorption efficiency [58]. At an elevated pH (higher than 10) themercaptocarboxylate ligands are mostly ionized and charged negatively thus expe-riencing electrostatic repulsion from the surface of TiO2 that is also negativelycharged. This repulsion, however, favors to the NP diffusion deeper into themesoporous oxide scaffold and results in better adsorption and higher loadings of theNPs, which can be further increased by elevating the temperature of NP solution[73]. Also, it is reported that the deprotonated thiolate group can form much stronger(by around 40 times) coordination bonds with Cd(II) ions on the surface of CdSeNPs [58] and, therefore, the NPs appear to be much more resistant to the agglom-eration in such conditions. Due to these factors the light conversion efficiency of theSSSCs based on the MPA-terminated CdSe NPs generally increases considerablywith an increase of pH of the solution used to deposit NPs onto the titania surface(Fig. 4.7a).

The same ligand exchange methodology can be applied to the ternarycadmium-free NPs, such as CIS and AgInS2 (AIS) chalcopyrites. The OLA-cappedCIS NPs can be rendered water-soluble by the ligand exchange with MPA or sulfideions [50]. The light conversion efficiency on TiO2/CIS heterostructures dependsstrongly on the size of a capping ligand and expectedly increases when bulky OLAor DDT is substituted with smaller MPA and S2− (Fig. 4.7b, gray bars) [50].

Even more dramatic changes can be observed in the rate constant of the electrontransfer from the capped-CIS NPs to TiO2—kET increases by around an order ofmagnitude after OLA (or DDT) is exchanged to smaller MPA and S2− species(Fig. 4.7b, red bars). The changes in the electron transfer dynamics are also illus-trated by a drastic decrease of the charge transfer resistance between CdSe NPs andTiO2 as OLA or DDT are substituted with MPA and sulfide ions (Fig. 4.7b, bluecircles).

In a similar way, the nanocrystalline ZnO-based photoanodes can be produced.For example, TOPO/HDA ligands on the surface of CdSe NPs can be substituted byMPA in methanol solutions resulting in good adsorption of the ligand-exchangedCdSe NPs on the surface of ZnO nanowires (NWs) [55]. The coverage of ZnONWs with NPs can be increased substantially via a preliminary treatment of theNWs with oxygen plasma [55]. The treatment can influence the ZnO NWs inmultiple ways—it can charge the surface attracting the negatively charged CdSe–

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 175

MPA NPs, eliminate surface impurities that prevent efficient NP adsorption, andproduce dangling (non-compensated) bonds tending to interact with the NPs [55].

Similar approaches were applied to anchor PbS NPs onto the mesoporous ZnO[68] and CdSe NPs—on ZnO nanotubes (NTs) [62]. A series of different moleculeswere tested as linkers for the attachment of PbS NPs to ZnO films [68], in particularoxalic, malonic and thioacetic acids, MAA (which is often referred to as thiogly-colic acid, TGA) and MPA as well as hexanedithiol. The highest photoresponseswere obtained for the ZnO/PbS systems with the ligands having a free –SH groupavailable for binding to the undercoordinated Pb atoms on the PbS NP surface—TGA, MPA, and hexanedithiol (Fig. 4.8a) [68].

Iodide-capped 6–7 nm PbSe NPs (Fig. 4.8b) were attached to the ZnO surfaceby cysteine (HS–CH2–CH(NH2)–COOH, Cys) [74]. Such NPs impart the zincoxide films with the spectral sensitivity to 1800–1900 nm as confirmed by bothabsorption and IPCE spectral measurements (Fig. 4.8c, d). The 6–7 nm PbSe NPsreside in the strong quantum confinement regime resulting in a considerableincrease of the CB potential and making possible the photoinduced electron transferinto CB of the ZnO scaffold, contrary to the PbSe bulk materials (Fig. 4.8e) [74].

The attempts to apply the ligand exchange with MPA to the ternary CIS NPsresulted in strong NP aggregation during the phase transfer into water. To cir-cumvent the aggregation effect a two-step ligand exchange was proposed [53]. Onthe first stage, the original DDT ligands were partially replaced with OA and theresulting NPs dispersed in water with the help of an ultrasound treatment. Oleic acidreplaces partially DDT and enters the ligand shell of CIS NPs interacting with thealkyl chains of neighboring thiol molecules, while –COOH group remains in theouter part of the shell. In alkaline solutions, the carboxyl group of OA is depro-tonated and protects the CIS NPs against aggregation via the electrostatic repulsion

Fig. 4.7 a Dependence of the light conversion efficiency for SSSCs based on the MPA-cappedCdSe NPs on pH of colloidal solution used for the photoanode preparation. Plotted using datapresented in [58]. b Light conversion efficiency η, electron transfer rate constant kET and chargetransfer resistance RCT calculated from the electrochemical impedance spectra of TiO2/CISphotoanodes based on differently capped CIS NPs. Plotted using data reported in [50]

176 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

between the NPs [53]. On the second step, the CIS NPs are brought into the contactwith a large excess of MPA that substitutes both residual DDT and OA, producingthe non-aggregated water-soluble NPs that can easily be attached to TiO2 [53].

Kesterite Cu2ZnSnS4 NPs prepared by the hot-injection method can be renderedwater-soluble by a ligand exchange with graphene oxide reduced by aromatic thiols[72]. The sheets of reduced graphene oxide (RGO) produced by this method aredecorated with C–SH and C=S groups that can coordinate to the NP surface,similarly to MPA and MAA.

Along with the organic bifunctional linkers, other types of small molecules andmetal complexes are probed as potential linkers for the attachment ofmetal-chalcogenide NPs to oxide surfaces. Of particular interest aremetal-chalcogenide inorganic complex ligands (ICL), such as SnS4

3−, SbS43− and

AsS33− [57]. Such ICL can be relatively easily produced by the dissolution of

corresponding metal sulfides in an excess of Na2S, they have a high affinity both tothe NPs and the oxide surface and can efficiency stabilize the NPs because of arelatively high negative charge. Also, the ICL-capped NPs can readilyself-assemble in tightly packed single layers which is very favorable for the for-mation of uniform SSSC photoanodes. Similarly to MPA (MAA), ICLs can beintroduced by a simple ligand exchange and promote the phase transfer of CdSeNPs into stable aqueous solutions [57].

A detailed study of the CdSe-ICL systems with a combination of the UV pho-toelectron spectroscopy and time-resolved photoluminescence (PL) spectroscopyrevealed a correlation between the LUMO position of the complexes relative to ECB

Fig. 4.8 a Scheme of a supposed binding of different ligands to the ZnO surface; b TEM ofas-synthesized OA-capped PbSe NPs; c, d absorption (b) and IPCE (c) spectra of colloidal PbSeNPs (c) and ZnO/PbSe heterostructures (d); e Band energy level diagram depicting the relevantenergy levels of bulk PbSe and 6–7 nm PbSe NPs and ZnO crystal. Reprinted with permissionsfrom Refs. [68] (a) and [74] (b–e). Copyright (2011, 2016) American Chemical Society

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 177

of CdSe NPs, the rate of photoinduced charge transfer from CdSe NPs to the titaniascaffold, and the PEC efficiency of such TiO2/NPs assemblies [57]. In particular,the charge transfer rate constant was found to be the highest (1.4 � 1011 s−1) forthe Sn-ICL, that has the lowest LUMO level relative to the ECB of CdSe NPs, anddecreasing for Sb-ICL (5.1 � 1010 s−1) and As-ICL (3.4 � 1010 s−1) because thebarrier for the electron transfer (a delta between LUMO and ECB) increases from0.21 eV for Sn-ICL to 0.77 eV for Sb-ICL to 1.26 eV for As-ICL (Fig. 4.9a).

However, the PEC light conversion efficiency does not follow this trend. It ismaximal for the Sb-ICL (1.67%), lower—for As-ICL (1.24%) and much lower(0.61%)—for the Sn-ICL. The authors of Ref. [57] hypothesized that the PECperformance of CdSe-ICL-TiO2 assemblies is determined not only by the efficiencyof electron transfer from CdSe NPs but also by the rate of valence band holetransfer to the electrolyte that meets the highest barrier for the Sn-ICL (Fig. 4.9a).

Similarly to the above-discussed polyelectrolyte-assisted multi-layer NPadsorption, the ICL-terminated CdSe NPs can be deposited as multilayers byalternating the deposition of negatively charged NPs with the adsorption of Cd orZn cations [57]. In this way, by using Cd2+ the PEC efficiency of a TiO2/CdSeheterostructure with the Sb-ICL linker can be increased to 1.84% for a four-layerdeposition of the sensitizer NPs (Fig. 4.9b).

Alternatively, the NPs can be attached to the titania surface without linkers. Forthis, the native ligands (OLA, TOPO, etc.) can be partially or even completelyeliminated by multiple washing of the NP precipitate with methanol [61] or CH2Cl2[48]. The direct adsorption can result in even closer interaction between the NPsand titania scaffold. For example, the charge transfer rate constant measured by thetime-resolved PL for CdSe NPs treated with methanol was found to be more than 3times higher (7.2 � 109 s−1 vs. 2.3 � 109 s−1), than for similar TiO2/CdSe com-posites produced using the MPA linker [61]. However, this method also suffersfrom the aggregation of CdSe NPs devoid of their ligand shell and, therefore, thewashing conditions should be chosen very carefully to achieve high PECcharacteristics.

Fig. 4.9 a Energy diagram for CdSe NPs, TiO2 scaffold, and three ILCs. b Jsc and light powerconversion efficiency (PCE) for SSSCs based on 1–4 layer CdSe/Sb-ICL NPs. Reprinted withpermissions from Ref. [57]. Copyright (2015) American Chemical Society

178 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

The preparation of photoanodes is typically finalized by the deposition of aprotective layer of ZnS [48, 49, 58, 66, 69, 71, 73] or a CdS layer [52, 54, 69], mostfrequently, using the SILAR technique. A ZnS (CdS) layer protects the NPs fromthe photochemical and “dark” corrosion and prevents the “leakage” of photogen-erated charge carriers from NPs into the electrolyte.

Direct aqueous synthesis. As shown by the above discussion, themercaptocarboxylate-capped CdX (X = S, Se, Te) NPs can easily adsorb on theoxide surface, being in this way very similar to the conventional Ru-complexsensitizers of the dye-sensitized solar cells, where the dyes bind to the TiO2 surfacevia the –COOH groups [73]. In view of the complexity of the ex situ synthesis andthe post-synthesis ligand exchange, a direct synthesis of themercaptocarboxylate-terminated NPs in water and other polar solvents is greatlypreferable. As the synthesis temperature is restricted by the solvent boiling point(100 °∁ for water) the direct synthesis does not allow such precise size variationsand structural perfection of the NPs as the above-discussed heating-up andhot-injection approaches. However, the direct syntheses provide another, quitepowerful methods of size variation and thus can be strong competitors to other exsitu synthetic protocols in view of their simplicity and a “green” nature.

Typically, CdSe NPs can be synthesized directly in water via the interactionbetween a chalcogen precursor and Cd(II) complexes with ligands-stabilizers. Inthis way, 2.3-nm CdSe NPs stabilized by MAA were produced in aqueous solutions[73] that can be directly adsorbed on the surface of mesoporous TiO2.

Ultra-small colloidal core/shell CdSe/CdS NPs can be produced by a directaqueous synthesis [75] and used as a sensitizer of mesoporous TiO2 (Fig. 4.10) in aSSSC with polysulfide electrolyte and a copper sulfide-based counter electrode[76]. The photoanodes were prepared by simply soaking the titania film with col-loidal solutions of 1.8–2.0-nm CdSe/CdS NPs (Fig. 4.10a, b). The sensitizer NPspenetrate the bulk of mesoporous titania film very uniformly showing identicalatomic cadmium, selenium, and sulfur contents both near the FTO transparentelectrode, in the bulk of TiO2 film and on the film surface (Fig. 4.10c–e). TheCdSe/CdS NPs absorb light in a spectral range of k < 450−460 nm and reveal ahigh chemical and photochemical stability. The total light conversion efficiency in aSSSC with the ultra-small NP-based FTO/TiO2/CdSe/CdS photoanode and an FTO/TiO2/Cu2S CE formed by the sulfidation of photocatalytically deposited Cu NPs isas high as 6.3% [76].

An aqueous synthesis of cadmium selenide NPs using cysteine anions as acapping agent results in the formation of ultra-small CdSe NPs with a well-resolvedabsorption maximum at 422 nm [77] that allows to identify these NPs ad theso-called “magic-size clusters”—stable ultra-small CdSe NPs with a well-definednumber of monomeric CdSe units in each NPs. At 80 °C the synthesis yields“regular” CdSe NPs with a bandgap around 2.32–2.34 eV, corresponding to theaverage size of around 2.5 nm. Both NP types can easily be attached to the titaniasurface upon the TiO2 film immersion into the colloidal CdSe solution [77].Cysteine can be applied also for the aqueous synthesis of CdS NPs with the NP size

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 179

varying from 2.3 to 2.8 nm [78]. The cysteine-capped CdS NPs readily adsorb onthe surface of ZnO NRs producing uniform ZnO/CdSe nanoheterostructures.

The MPA-capped CdTe NPs can be produced directly in aqueous solutions viathe sodium tellurite reduction with NaBH4 in the presence of Cd(II) salts under themicrowave heating and then adsorbed onto the surface of ZnO NRs [79].

A variation of the heating duration (7–30 min) allows tunig of the average sizeof CdTe NPs from 4 to 9 nm. Alternatively, CdTe NPs can be formed at theexpense of Te reduction by NaBH4 in boiling aqueous solutions containing cad-mium perchlorate and MPA [80, 81]. Such NPs can be deposited onto the TiO2

surface either by spontaneous adsorption from the solution [80] or by thedrop-casting and evaporation of CdTe colloid on the TiO2 scaffold surface [81]. Asthe MPA-capped CdTe NPs are charged negatively, the multi-layer deposition ofthe NPs onto titania is possible together with a positively charged polyelectrolyte—poly(dimethyl diallyl ammonium chloride) [80]. The procedure includes a cyclicadsorption of the polyelectrolyte and MPA-capped NPs attracted to each other bythe electrostatic forces.

The ternary CIS NPs formed directly in aqueous solutions in the presence ofTGA can be used as “inks” to sensitize porous TiO2 substrates via a simpleimmersion technique [82].

Fig. 4.10 a Cross-sectional SEM image of mesoporous TiO2 film soaked with colloidalCdSe/CdS NP solution (a), the elemental analysis was performed in the points numbered 1–5 andthe results presented for Cd, Se, and S in (e); b TEM/HRTEM images of colloidal CdSe/CdS NPs;c, d Cd (c) and Se (d) atom distribution in the cross section of the TiO2/CdSe/CdS film

180 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Similarly to CdTe, CIS NPs can be produced directly in aqueous solutions ofvarious sulfur-containing ligands (TGA, MPA, glutathione (GSH), Cys) under themicrowave heating [83]. By applying a co-linker (for example, TGA to theCys-capped CIS NPs) the amount of NPs adsorbed on titania can be increasedconsiderably resulting in a drastic (by more than 20 times) increase in the lightconversion efficiency [83]. The additional ligand is supposed to participate in thechemical reduction of S–S fragments that form on the NP surface as a result ofpartial oxidation of the primary ligand and hinder the NP adsorption on the TiO2

surface. A PL quenching study showed that the rate constant of electron transferfrom CIS NPs to TiO2 is the highest for Cys linker, 9.5 � 1010 s−1, decreasing to7.1 � 1010 s−1 for a more bulky GSH. The same tendencies were found also forAIS and CdSxSe1−xS NPs [83].

The MAA-stabilized CIS/ZnS NPs penetrate uniformly the volume of meso-porous TiO2 films revealing a homogeneous composition of the resulting TiO2/CISboth along the cross section of the films (Fig. 4.11a, b) and across the outer filmsurface (Fig. 4.11c). The stability and PEC activity of CIS NPs both increase uponthe deposition of a thin ZnS shell on the CIS NP surface. The TiO2/CIS/ZnScomposites act as visible-light-sensitive photoanodes in the SSSCs with polysulfideelectrolyte and copper sulfide-based counter electrodes with a total conversionefficiency of around 8% [84].

Mixed ex situ/in situ approach. In this approach, the TiO2 films are immersedinto polar solutions where the primary nuclei of sensitizer NPs form. The growthand attachment of the metal-chalcogenide to the oxide surface occur simultaneouslyduring the following heat treatment in the solvothermal conditions resulting in auniform NP distribution over the film volume. The NP size can be tailored byvarying the duration or/and temperature of the heat treatment. For example,

Fig. 4.11 Cross-sectional SEM images of FTO/TiO2/CIS/ZnS photoanode (a, b); results of theenergy-dispersive X-ray spectroscopic determination of elemental composition of the photoanodecross section (b) and the outer surface (c) [84]

4.3 Nanocrystalline Photoanodes Produced by the Ex Situ … 181

a hydrothermal treatment (HTT) of the nanocrystalline TiO2 films in a solution ofprimary CdTe nuclei, produced by the injection of NaHTe into aqueous solution ofcadmium(II) mercaptoacetate, yields CdTe NPs anchored to the titania surface viathe MAA bridge [85]. The average size of CdTe NPs can be varied from around 3to 6 nm by increasing the HTT temperature from 80 to 160 °C. Additionally, a CdSshell was spontaneously deposited onto the CdTe NPs as a result of MAAhydrolysis rendering the NPs stable toward the air oxidation [85].

The nuclei for the preparation of CdSe NPs can be produced by the injection ofNaHSe into alkaline aqueous Cd(II)-MAA solution under an inert atmosphere [73].In a similar way, a mixture of Na2S and NaHSe, prepared via the reduction ofselenium with NaBH4, was used to form the nuclei of MAA-stabilized CdSxSe1−xNPs [86]. A photoanode is then produced by a HTT of a TiO2 film immersed intothe nuclei solution.

Electrophoretic deposition. The metal-chalcogenide NPs can be deposited ontothe surfaces of oxide substrates indiscriminately of the NP surface chemistry by theelectrophoretic deposition technique. In this method, two electrodes—FTO/TiO2

and bare FTO are immersed into the NP solution and a voltage of 60–200 V � cm−1 is applied between the two electrodes, with the FTO/TiO2 filmtypically connected to the positive terminal of the power supply unit [51, 65]. Themagnitude of applied voltage depends on the solvent polarity and increases from60 V/cm for aqueous solutions to 200 V/cm for toluene [37, 47]. The NPs move inthe electrostatic field and deposit as a uniform layer on the polarized titania surface.Most probably, the NPs are stripped from a portion of their protecting ligand layerin the process of the field-stimulated adsorption, coming, therefore in a closecontact with the oxide scaffold.

The electrophoretic deposition was successfully applied to decorate poroustitania scaffolds with CdSe [47], CdSxSe1−x [65], and CIS NPs [51, 52]. Themethod can be also used to form multi-layer structures as shown by thelayer-by-layer deposition of composition-selected CdSxSe1−x NPs with anincreasing bandgap simply by switching between the NP solutions [65].

The electrophoretic deposition was found to be especially attractive to produceuniform sensitizer layers when conventional adsorption of MPA-capped NPs isinefficient for some reasons, for example, in the case of elongated CdSe NRs thatmeet difficulties in penetrating the mesopores of titania scaffolds [87].

4.4 Nanocrystalline Photoanodes Produced by the In SituDeposition of Sensitizer NPs

The in situ formation of sensitizer NPs occurs directly on the surface ofwide-bandgap oxide as a result of chemical reactions taking place in the oxidesurface layer with the participation of surface functional groups or the chargecarriers generated in/injected into the wide-bandgap material. In the former case,

182 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

metal ions are first adsorbed on the oxide surface, then the interaction between theadsorbed metal ions and chalcogen anions takes place resulting in a layer of metalchalcogenide NPs.

The two ideologies of NP deposition—in situ and ex situ are in constant com-petition, as each of them provides unique possibilities of the control of morphologyand properties of the NPs and/or NP-oxide interface, but, at the same time, each hasinherent drawbacks and limitations. In particular, the in situ deposition by SILARor CBD ensures a good contact between the metal oxide scaffold and the NPsensitizer, because the NP formation starts immediately on the surface oxide layerand typically an intermediary thin oxide/chalcogenide layer forms between theoxide and NPs ensuring perfect electron transport from the photoexcited NPs to theoxide scaffold. The NPs deposited by ex situ methods are invariably separated fromthe oxide surface by a ligand shell, which can be thin, for example, in the case ofusing MAA or MPA ligands, but nevertheless affecting negatively the efficiency ofelectron transfer from NPs to the porous oxide layer. However, the ex situ for-mation allows tuning the size, shape, and composition of the NPs bywell-established synthetic protocols with a precision typically unachievable for thein situ methods. As a result, the properties of NP/oxide photoanodes dependstrongly on the method of preparation, in particular, the way of NP deposition andboth in situ and ex situ methods are constantly developing and brought to thecomparison.

Deposition of sensitizer NPs by SILAR. The method is very simple from theexperimental viewpoint but, at the same time, it allows to produce a variety ofmetal-chalcogenide NP-based heterostructures exhibiting quite high efficiencies ofthe light conversion. Typically, the SILAR procedure consists in the immersion of awide-bandgap porous oxide film (TiO2, ZnO) into a solution of a metal precursor(soluble salts) for some time necessary for the adsorption/desorption equilibrium tosettle, then the film is extracted, washed with pure water and immersed into anothersolution containing chalcogenide X2−/HX− ions (X = S, Se, Te) or a chalcogenideprecursor that can readily decompose producing the chalcogenide ions. At that, athin (ideally—a monomolecular) layer of metal-chalcogenide forms on the surfaceof the oxide film. The procedure is them repeated many times each cycle producingan additional layer of the metal chalcogenide. In this way, the metal chalcogenidelayer thickness (and in some cases—the NP size) is determined by a number N ofthe SILAR cycle repetitions that can vary from 2–3 to tens for manual preparationand even to hundreds if a mechanically-controlled setup is used for the filmpreparation. Typically, the SILAR produces smooth layers of sensitizer NPs cov-ering the entire surface of the wide-bandgap material and does not block the poresin the oxide scaffold that can still be freely penetrated by the electrolyte after the NPdeposition. The SILAR can be applied to different wide-bandgap scaffold materials(TiO2, ZnO, In2O3) and morphologies (mesoporous films, NRs, NWs, nanosheets,and nanoplates, etc.), NP sensitizers (CdX, PbX, CuX, ternary/multinary NPs) andperformed from different solvents (water, methanol, ethanol, acetone). Table 4.2demonstrates some examples of the SSSCs where the visible-light-sensitive com-ponent was produced by the SILAR technique.

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 183

As the most frequent case, CdS NPs are deposited by SILAR using aqueous oralcohol solutions of Cd(II) nitrate and Na2S on the mesoporous titania [88–91],TiO2 NTs [92] and nanosheets [93], ZnO NWs [94, 95], nanoplates [91, 96] andNRs [97], producing visible-light-sensitive TiO2/CdS and ZnO/CdS heterostruc-tures. Typically, cadmium sulfide forms a dense layer on the TiO2 (ZnO) surfaceshielding it from the electrolyte. According to the optical data, the CdS layerconsists of separate NPs with the NP size and the layer thickness in generaldepending on the number of SILAR cycles.

At primary steps of the SILAR procedure the thickness and absorbance of theCdS layer depend almost linearly on the SILAR cycle number. Figure 4.12 showsthat the thickness of CdS layer deposited by the SILAR onto the ZnO NWsincreases continuously from 3 to around 12 nm as N grows from 10 to 120 [94].A comparison of TEM and optical absorption data indicates that for a given

Table 4.2 Some examples of SSSCs produced by the SILAR deposition of sensitizer NPs

Photoanode material Eg(NPs),eV

Counterelectrode

Jsc,mA/cm2

Voc,V

FF,%

η,%

Reference

ZnO/CdS (ZnONWs)

2.4 Pt 7.2 n/r n/r 3.53 [94]

TiO2/CdS/CdSe(meso-TiO2)

*1.8 Au 11.9 0.51 53 3.20 [116]

TiO2/CdS (TiO2

NTs)2.4 Pt 6.4 0.46 41 1.19 [92]

TiO2/CdS (TiO2

nanosheets)*2.1 CuxS 9.3 0.59 42 2.29 [93]

TiO2/ZnS/CdS/ZnS *2.4 CuxS 10.3 0.64 57 3.69 [88]

TiO2/CdS/Bi2S3/ZnS *1.5 CuxS 9.3 0.50 54 2.52 [89]

TiO2/CdSe *1.8 CuxS 15.9 0.58 57 5.21 [90]

ZnO/Ag2S (ZnONWs)

1.0 Au 11.4 0.26 38 1.10 [95]

TiO2/Ag2S n/r Pt 7.3 0.33 41 0.98 [105]

TiO2/AgSbS2 1.7 Au 2.4 0.32 n/r 0.34 [101]

TiO2/AgBiS2 1.32 Pt 7.6 0.18 39 0.53 [99]

TiO2/PbS/CdS n/r CuxS 10.9 0.44 46 2.21 [103]

TiO2/PbS/Pb0.2Cd0.8S/CdS

*2.2 CuxS 10.3 0.42 32 1.37 [104]

TiO2/PbS:Hg2+ *1.0 CuxS 30.0 0.40 47 5.58 [108]

TiO2/CuxTe *1.0 Carbon 8.6 0.50 16 0.69 [118]

TiO2/CuInS2/Bi2S3 *1.5 CuxSa n/r n/r n/r 0.44 [100]

TiO2/Cu2Se/CuInS2 1.5 Pt 6.5 0.59 32 1.22 [107]

Note The table reports the highest η values achieved in corresponding papers; the cells wereilluminated with AM1.5 light (100 mW/cm2) if not stated otherwise; redox couple is S2−/Sx

2− if notstated otherwise; in some cases a scattering layer was applied on top of the photoanodes tpoincrease efficiency (see original refs.) n/r—not reportedaelectrolyte contained Na2S and Na2SO3

184 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

ZnO/CdS nanoheterostructure the CdS layer thickness is comparable to the size ofCdS NPs [94] indicating a monolayer coverage of the ZnO NWs with CdS NPs of adifferent size.

The PEC activity of the TiO2/CdS and ZnO/CdS heterostructures produced bythe SILAR also depends on the cycle number N and increases at first, then decreasesrevealing a distinct maximum. For example, Jsc produced by the NW-basedZnO/CdS heterostructures increases till N = 30 and then comes to a saturation valuefor a much higher N (up to 120) [94]. At the same time, for TiO2/CdS compositesproduced from the anodized titania NTs the light conversion efficiency grows up toN = 5 and falls considerably at a higher SILAR cycle number (Fig. 4.13a, curve 1)[92]. The efficiency decrease is associated with the blockage of NT openings thatprohibits the electrolyte penetration and the regeneration of sensitizer NPs.

Fig. 4.12 TEM images of ZnO NWs decorated with a CdS layer by the SILAR with a differentcycle number N. Reproduced with permissions from Ref. [94]. Copyright (2009) The RoyalSociety of Chemistry

Fig. 4.13 a Efficiency of SSSCs based on TiO2 NT/CdS (curve 1) and meso-TiO2/Ag2S (curve 2)heterostructures as a function of the SILAR cycle number N, plotted using data reported in [92](curve 1) and [105] (curve 2); b Photocurrent generation spectra of ZnO nanoplates (curve 1) andZnO/CdS heterostructures produced at N = 5 (curve 2), 10 (3), 20 (4), 60 (5), and 200 (6). Y is thephotocurrent quantum yield [98]

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 185

On the contrary, for the mesoporous TiO2 (ZnO) with no particular spatialarrangement the light conversion efficiency typically saturates at a certainN (varying for the scaffold nature) and almost does not change at a higher numberof the SILAR cycle repetitions [91, 92, 94]. These examples show that the pho-toanode efficiency dependence on the amount of SILAR-deposited NPs can varyquite strongly reflecting differences in the shape and surface chemistry of thewide-bandgap scaffolds.

The CdS NP-based heterostructures produced by the SILAR reveal anotherintriguing property that is often overseen by the researchers. As discussed above,the thickness of cadmium sulfide layer and the size of CdS NPs both increase as thecycle number N is elevated. The size increase is accompanied by a lowering of thebandgap of CdS NPs and a corresponding “red” shift of the band edges both inoptical absorption and photocurrent spectra of TiO2/CdS (ZnO/CdS) heterostruc-tures. The bandgap shrinking indicates a continuous weakening of the quantum sizeeffects in the growing CdS NPs and one can expect that for NPs larger than thedoubled Bohr exciton radius in cadmium sulfide (typically for d > 10 nm) noquantum size effects will be observed and the bandgap of CdS NPs will reach thevalue of Eg = 2.4 eV typical for the bulk cadmium sulfide. However, the TiO2/CdS(ZnO/CdS) heterostructures produced at a relatively high SILAR cycle number,N > 20–30, reveal distinctly lower Eg values, that can be as small as 2.0–2.2 eVcorresponding to the band edge of around 600 nm (Fig. 4.13b) [98].

The phenomenon of the bandgap of SILAR-produced CdS NPs being lower thanthe bulk value appears to be registered quite frequently but not paid due attention[92, 93, 95]. Gary Hodes et al. were the first group to assess this phenomenonsystematically [97], probing several alternative explanations, including formation ofa type II heterojunction between the wide-bandgap scaffold (ZnO in this case) andCdS NPs, possible effects of adsorbates and surface states on the band structure,contribution of Cd-enriched faces of CdS NPs into the band edge positions, and,finally, the participation of sub-bandgap states in the light absorption and thephotocurrent generation.

The analysis of Hodes et al. allowed to conclude definitely [97] that theextension of the edges of absorption and IPCE bands in the spectra of ZnO/CdS(TiO2/CdS) heterostructures originates from a large contribution of sub-bandgapstates introduced by a high structural disorder of the lattice of CdS NPs depositedby the SILAR, unlikely all other deposition methods. This conclusion found sup-port in the studies of CdS-based photoanodes on various scaffolds (ZnO, TiO2,In2O3) by the resonant Raman spectroscopy [91] confirming a high structuraldisorder of the SILAR-produced CdS NPs as indicated by the appearance ofcharacteristic disorder-activated vibrational modes.

Apart from cadmium sulfide, the SILAR technique was successfully applied forthe preparation of photoanodes comprising NPs of ZnS [88, 89], Bi2S3 [89, 99, 100],Sb2S3 [101, 102], PbS [103, 104], Ag2S [95, 99, 101, 105, 106] and others. Recently,this method was extended for the synthesis of ternary metal-chalcogenide NPs, suchas CIS [100, 107]. In this case, each SILAR cycle includes the successive adsorptionof copper(II) and indium(III) followed by their interaction with sulfide ions.

186 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Figure 4.14 shows how the absorbance of some toxic-metal-free sensitizer NPsgrows as the SILAR cycle number is continuously increased.

The morphology of oxide/chalcogenide heterostructures produced by the SILARdepends strongly on the interaction between the oxide surface and metal ions in theforming NPs as well as on the metal chalcogenide layer thickness. In particular, theSILAR-deposited CdS NPs typically tend to form a uniform and dense layer on thesurface of both titanium and zinc oxides. On the contrary, Ag2S [95, 105, 106] andPbS [103] NPs formed by the SILAR are typically larger and distributed randomlyover the oxide surface (Fig. 4.14a, insert) thus leaving a portion of the surfaceaccessible for the electrolyte and for the potential recombination between thereduced/oxidized electrolyte species and the photogenerated holes/electrons,respectively. This recombination is one of possible reasons for a generally lowerlight conversion efficiency of Ag2S-based (Fig. 4.13a, curve 2) and PbS-based(Table 4.2) heterostructures as compared to their CdS-based counterparts.

A larger size of the SILAR-deposited Ag2S and PbS NPs with respect to CdSNPs reflects faster aggregation of the less-soluble silver sulfide particles already onthe stage of the primary nuclei formation. At the same time, the Ag2S-basedphotoanodes also reveal a typical volcano-shaped dependence of the light con-version efficiency on N (Fig. 4.13a, curve 2) indicating on a general character andreasons for such dependence for various sensitizer NPs.

Apart from individual metal-chalcogenide NPs, the SILAR allows to depositmixed metal solid-solution chalcogenide NPs and to dope NPs with another metal[108]. For example, by using Pb(II) and Cd(II) precursors individually and as amixture, a graded multi-layer TiO2/PbS/PbxCd1−xS/CdS photoanodes were formedrevealing a light conversion efficiency of around 1.4% [104] (Table 4.2). Byvarying the composition of a mixture of Cd(II) and Zn(II) salts, a series of ZnO/CdxZn1−xS photoanodes can be produced with a tunable spectral response andpositions of the CB and VB levels [109]. As the reactivity of both metals towards

Fig. 4.14 Absorption (a, c) and transmission (b) spectra of TiO2-based photoanodes produced ata different SILAR cycle number (given on figures) with Ag2S NPs (a), AgBiS2 NPs (b), andCuInS2 NPs (c). Insert in (a) TEM of Ag2S NPs on the TiO2 surface. Reprinted with permissionsfrom Refs. [105] (a), [99] (b), and [100] (c). Copyright (2010 (a), 2013 (b), 2015 (c)) Elsevier (a),American Chemical Society (b), The Royal Socienty of Chemistry (c)

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 187

sulfide anions is typically different, the composition of mixed NPs can differ quitenotably from the solution composition set during the SILAR procedure. In partic-ular, the alloyed layer in the above-mentioned TiO2/PbS/PbxCd1−xS/CdS compositecontains 20 mol% lead while only 5 mol% Pb(II) was present in the precursorsolution [104]. Therefore, the composition of a mixed photoanode should be ver-ified in each specific case as, for example, in the SILAR-produced ZnO/CdxZn1−xSsystem, where a correlation between the real and nominal Zn content was deter-mined independently by optical, Raman and energy-dispersive X-Ray spectro-scopies [109].

The CdxZn1−xS solid solution is a perfect “polygon” for probing dependences ofthe photochemical/PEC activity on the CB and VB energies because both valuescan be easily varied by changing the NP composition at more or less constant sizeand lattice parameters. As mentioned above, the photovoltage in a liquid-junctionSSSC depends on the difference between the redox-potential of the electron shut-ting couple in the electrolyte and the Fermi energy of the photoanode. In thecadmium-zinc-sulfide-based SSSCs the latter can be approximately assumed to belinearly dependent on ECB of CdxZn1−xS NPs. The conduction band potential ofmixed CdxZn1−xS NPs varies from ECB(CdS) = −0.8 V (versus normal hydrogenelectrode, NHE) to ECB(ZnS) = −1.8 V (versus NHE). The details of calculationsof composition-dependent band potentials of CdxZn1−xS NPs can be found inChap. 6.

The open-circuit photovoltage in the SSSCs based on the ITO/ZnO/CdxZn1−xSphotoanodes, Voc(x), grows with an increase in the Zn content (Table 4.3). Thephotovoltage increment with respect to Voc of a CdS-based photoanode,DVoc(x) = Voc(CdS) − Voc(x) generally follows the corresponding increment of theCB potential, DECB(x), showing a considerable deviation between the twoparameters only for the smallest studied Cd content at x = 0.62.

A decrease in the Cd content is also accompanied by an increase in the potentialcorresponding to the maximal power of the CdxZn1−xS-based SSSCs. As the sen-sitizer is changed from CdS to mixed cadmium-zinc-sulfide NPs with x = 0.62 thispotential grows by around 250 mV [109], similarly to the corresponding increase inthe CB potential of the sensitizer NPs (220 mV, Table 4.3), as well as to theopen-circuit photovoltage of the corresponding cells (280 mV, Table 4.3). A close

Table 4.3 Molar Cd fraction x in CdxZn1−xS NPs (determined by the energy-dispersive X-rayspectroscopy), bandgap Eg of CdxZn1−xS determined by the optical absorption spectroscopy), thephotocurrent density at a dark immersion potential, Jph, the open-circuit photovoltage Voc,DECB(x) and DVoc(x) parameters for the solar cells based on CdxZn1−xS NPs [109]

x Eg, eV Jph, mA � cm−2 Voc, mV DECB(x), mV DVoc(x), mV

0.62 2.69 0.43 1.46 220 280

0.74 2.62 0.47 1.37 170 190

0.83 2.53 0.40 1.30 105 120

0.95 2.47 0.29 1.24 56 60

1.09 2.40 0.23 1.18 0 0

188 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

increment of the photovoltage, around 310 mV, was also observed for thesecond-generation thin-film solar cells based on CdxZn1−xS/CdTe heterostructures,where x was varied from 1.00 to 0.65 [110].

The photocurrent density Jph of the CdxZn1−xS-based photoanodes grows as x isdecreased from 1.00 to 0.75–0.80, but then falls at a higher Zn(II) content(Table 4.3, Fig. 4.15a, curve 1) [109]. Such behavior was also observed for theSILAR-produced TiO2/CdxZn1−xS heterostructures at x < 0.75 and explained by astrong blue shift of the absorption band edge, kbe, of the sensitizer NPs, resulting ina partial loss of the solar light harvesting capability [111]. A similar reason isresponsible for a Jph decrease in the case of ZnO/CdxZn1−xS system because thedependence between the absorbance-normalized Jph and the CB potential shows theexpected monotonous ascending behavior (Fig. 4.15a, curve 2). The figure showsthat the absorbance-normalized photocurrent density increases by a factor of almost4 as x is lowered from 1.0 to 0.62 indicating a crucial role of the energy bandpositions of sensitizer NPs for the SSSC performance.

The relationship between the composition-variable CB potential increment ofCdxZn1−xS NPs and the normalized photocurrent density can be described by Tafelequation DE = a + blogi, where a and b are coefficients and DE is an over-voltageof the interfacial electron transfer from cadmium-zinc-sulfide NPs to the ZnOscaffold. Accordingly, the dependence of log(i/Aint) on the energy gap between adonor level (ECB of sensitizer NPs or the related ECB(x) − ECB(CdS) difference)and an acceptor level (ECB of ZnO scaffold) is linear (Fig. 4.15b).

The linear Tafel dependences are typical for systems where no barrier exists forthe interfacial charge transfer and the transfer rate is determined predominantly bythe difference between the donor and acceptor level energies. In thecadmium-zinc-sulfide-based photoanodes a photogenerated electron migrates from

Fig. 4.15 a Photocurrent density Jph (curve 1) and Jph normalized to the total absorbance of ITO/ZnO/CdxZn1−xS films at k > 400 nm (curve 2) as a function of the composition-dependentconduction band potential ECB(x) of CdxZn1−xS NPs; b ECB(x) − ECB(CdS) versus logarithm ofthe normalized photocurrent density

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 189

CdxZn1−xS NPs to ZnO NPs to ITO, thus decreasing its energy. By using reporteddata about the free electron energy versus the vacuum level for CdS, ZnO, and ITO[16, 112–115], the conduction band potentials of CdS and ZnO in 0.01 M Na2Saqueous electrolyte can be estimated as −0.9 V (versus NHE) and −0.6 to −0.5 V(NHE), respectively, while the accepting level of ITO is estimated to be at around−0.1 to 0 V (NHE). Therefore, in the TIO/ZnO/CdS system there exists a favorablethermodynamic band alignment for the cascade electron transfer from CdS to ZnOto ITO and the efficiency of this process is expected to increase as ECB becomesmore negative with an increase in the Zn(II) content, in accordance with theabove-discussed experimental results.

The CdSe NPs can also be deposited by SILAR, typically under an inertatmosphere to avoid the photoanode contamination with elemental selenium. TheSe2− ions come from SeO2 reduced in situ by NaBH4 [90, 116] or directly fromNa2Se [96]. Similarly to CdS, an increase in the SILAR cycle number results bothin the growth of CdSe absorbance and in a decrease of the average Eg of thedeposited CdSe NPs. For example, the bandgap of CdSe NPs deposited on themesoporous TiO2 decreases from *2.5 eV to around 1.8 eV as the SILAR cyclenumber is elevated from 3 to 8–10 (Fig. 4.16a). Using a well-established correlationcurve between Eg and average size of CdSe NPs the size of CdSe NPs can beestimated to be around 2.6 nm at N = 3, increasing to *8 nm for N = 7, and tohigher size values at the further repetition of the SILAR deposition cycles.

To protect CdSe NPs during the PEC experiments, the photoanode was coveredby a ZnS shell, also using simple SILAR technique with a small (1–2) number ofcycles [116]. Such a thin ZnS layer prohibits the photocorrosion of CdX NPs as

Fig. 4.16 a Bandgap and average size of cadmium selenide NPs in TiO2/CdSe heterostructure asa function of the SILAR cycle number N (plotted basing on data reported in [90]); b–e TEMimages of TiO2 crystals before PbSe deposition (b) and after (c), 2 (d), and 3 (e) SILAR cycles ofPbSe deposition. The scale bar is 5 nm. The red figures is a Pb-to-Se ratio as determined by theenergy-dispersive X-ray spectroscopy. Reprinted with permissions from Ref. [117]. Copyright(2012) American Chemical Society

190 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

well as the recombination of the photogenerated charge carriers with the electrolytespecies as discussed in details below in this chapter.

The less-soluble PbSe NPs can be deposited using sodium selenosulfate as a Se2− source that is quite stable on air in the absence of metal ions allowing for a simpleand reliable SILAR procedure to be performed in the ambient conditions [117].Surprisingly, even a single SILAR cycle produced well defined 2–3 nm PbSe NPsrandomly distributed over the surface of mesoporous TiO2. As the cycle number iselevated to 3 the NP size increases to 4–5 nm and the PbSe NPs become enrichedwith Pb (Fig. 4.16b).

Frequently, the SILAR deposition is followed by a heat treatment, either topromote the crystallization of the NP deposit or to induce a chemical transformationof the deposited precursors. For example, CuxTe NPs can be produced on themesoporous TiO2 by a thermal treatment of a copper(II) tellurite layer deposited bythe SILAR [118]. Annealing of a layered ZnO/CuS/Sb2S3 heterostructure results inthe copper(II) reduction with sulfide ions and simultaneous formation of ternaryCuSbS2 NPs [102]. Similarly, the thermal treatment of two separate metal sulfidelayers pre-deposited by the SILAR on the mesoporous titania yields AgSbS2 [101]and AgBiS2 NPs [99].

Electrodeposition of sensitizer NCs. The electrodeposition methods are basedon the electrochemical reduction of chalcogenide precursors resulting in the releaseof X2− anions (S2−, Se2−, Te2−). The chalcogenide then interacts with metal ionsadsorbed on the surface of a wide-bandgap scaffold which serves as a workingelectrode. The method is typically fast and can potentially be applied to prepare abroad variety of metal sulfide, selenide, and telluride NPs. For example, CdS NPscan be electrochemically deposited from hot aqueous electrolytes or water/DMSOmixtures containing Cd(II) nitrate, and thiourea or elemental sulfur. The elec-trodeposition is typically performed under the galvanostatic control and elevatedtemperature (around 90 °C). In this way, CdS NPs were successfully electrode-posited onto the surface of ZnO NRs (Fig. 4.17a–c) [119, 120], ZnO NTs [121],and hierarchical TiO2 microspheres [122].

Cadmium selenide and telluride NPs were electrodeposited in a similar way byusing Na2SeSO3 [122, 123] and K2TeO3 [124] as selenium and tellurium sources,respectively. As a rule, the electrodeposition yields relatively large NPs in the formof dense NP layers with a thickness of several tens of nanometers showing bulk-likebandgaps [119–121, 125].

In cases of oxide scaffolds with a largely anisotropic morphology, like NW orNT arrays, a homogeneous distribution of the electrodeposited metal-chalcogenideNPs can be achieved by applying frequency-controlled electrodeposition regimes.For example, the electrodeposition of CuInS2 NPs onto ZnO NR arrays(Fig. 4.17d–f) by using square current pulses with a frequency of 1 kHz produces arelatively smooth CIS layer distributed evenly from the top of ZnO NR down to theNR contact site with the FTO surface (Fig. 4.17f) [125].

At the same time, with a 1 Hz pulse or under the continuous electrodepositionthe CIS layer deposits predominantly on the ZnO NR tops forming a dense layerthat hinders the electrolyte penetration to the inter-NR space (Fig. 4.17e). The

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 191

reason for such morphology of the CIS layer is the depletion of the inter-NR spacewith the reactants at the continuous/quasi-continuous deposition resulting in thepreferred CIS NP deposition on the outer border of the NR array.

By applying a 1 kHz pulses the rates of the CIS NP growth and the diffusion offresh portions of reactants to the ZnO NR surface can be equilibrated favoring to theformation of a smooth sensitizer NP layer. It should be noted, however, that theoptimal frequency for the electrodeposition is unique for a given scaffold mor-phology and electrolyte composition [125].

The electrodeposition is one of the most frequently used methods for the for-mation of Cu2O NPs both on bare conducting substrates and on the surface of oxidescaffolds. Typically, the electrodeposition is performed from alkaline solutionscontaining Cu(II) complexes with lactic acid anions [126, 127].

Chemical Bath Deposition (CBD) of sensitizer NPs. The CBD is also a rela-tively simple deposition method requiring the immersion of a wide-bandgap sub-strate into a hot bath containing metal and chalcogen precursors. The X2− anions are

Fig. 4.17 SEM (a, b) and TEM (c) images of a ZnO NR tip (a), ZnO NR tip with electrodepositedCdS (b) and a fragment of the ZnO/CdS NR heterostructure (c). d–f SEM images of ZnO NR array(d) and ZnO NR/CuInS2 heterostructures produced by the electrodeposition with a pulse frequencyof 1 Hz (e) and 1 kHz (f). Reprinted with permissions from Refs. [120] (a–c) and [125] (d–f).Copyright (2012) Elsevier (a–c) and (2015) The Royal Society of Chemistry (d–f)

192 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

slowly released as a result of the hydrolytic (solvolytic) decomposition of thechalcogen precursor, while the metal ions are bound by a complexing agent toprevent rapid formation of a deposit and to level off the rates of X2− release andprecipitate formation.

The metal sulfide NP deposits are typically produced using Na2S2O3, thiourea orthioacetamide as sulfur sources that can be slowly decomposed in alkaline media inthe ambient conditions. Similarly to the SILAR, the CBD deposition can be usedfor the preparation of complex ternary metal sulfide-based heterostructures. Inparticular, the CBD from aqueous acidic solutions of copper(II) and bismuth(III)nitrates in the presence of Na2S2O3 can be used to produce meso-TiO2/CuBiS2heterostructures [128]. A solvothermal treatment of a trilayer meso-TiO2/Ag2S/In2S3 heterostructure, where the indium sulfide layer was deposited by the CBDusing thiourea as a sulfur source results in TiO2/AgInS2 composites.

Metal selenides can be deposited by the CBD using sodium selenosulfate as aSe2− source that can be easily prepared by dissolving elemental Se in hot aqueousNa2SO3 solutions. Typically, to deposit uniform layers of CdSe NPs onto thewide-bandgap scaffold a thin layer of “seed” CdS NPs is preliminarily formed bythe SILAR [129–132]. Then the TiO2/CdS (ZnO/CdS) heterostructure is immersedinto an alkaline (pH 11–12) aqueous solution containing Cd(II), nitrilotriacetic acidas a Cd(II) complexing agent and Na2SeSO3 at room or lowered temperatureresulting in the growth of a uniform CdSe NP layer.

The deposited CdSe NP layer thickness can be controlled by varying the CBDduration. As an example, a variation of the CDB time from 5 to 50 h can be used totailor the thickness of a CdSe layer deposited on TiO2/CdS heterostructure from 20to 180 nm [132]. This effect was used [132] to probe the light-to-current conversionefficiency of multi-layer CdSe NP coverings on the surface of compact titania. Thecompact scaffolds were chosen to minimize possible distortions of the multi-layeruniformity caused by the curvature of TiO2 mesopores. It was found that Jsc gen-erated by such multilayer TiO2/CdS/CdSe heterostructures increases as the CdSelayer thickness is increased to around 100 nm as a result of an enhancement of thelight absorbance by the photoanode (Fig. 4.18a, red bars). At higher thicknesses,however, the photocurrent generation efficiency drops most probably due to anincrease in the distance the photogenerated carriers need to pass before they can becollected by the TiO2 layer.

The normalization of Jsc to the CdSe layer absorbance showed that, indeed, forthe film thickness of 20–100 nm, the photocurrent generation efficiency is almostthe same, while lowering noticeably for higher thicknesses of the CdSe layer(Fig. 4.18a, blue bars). This result illustrates a fundamental difference between thedye-sensitized and NP-sensitized PEC solar cells—the maximal efficiency of theDSSCs is typically observed at a monolayer coverage of the TiO2 surface with adye-sensitizer, while in the SSSCs a multilayer absorber can be used and thus thelight harvesting efficiency can be increased dramatically.

Another important conclusion drawn from the results of [132] was that theSSSCs do not necessarily need a mesoporous wide-bandgap scaffold with a highlydeveloped surface area, as is the case for DSSCs. Using relatively compact TiO2

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 193

films composed of around 120 nm titania crystals the authors of [132] obtained amuch higher light conversion efficiency than in the case of the mesoporous titania.The calculations performed in [132] basing on an optimized CdSe NP absorberthickness of 100 nm and the maximal TiO2 layer thickness providing 100% col-lection of all injected charge carriers, 4 lm, showed that the best hypotheticalconfiguration of the titania scaffold for the SSSCs is a periodic array of verticallyaligned hexagonally packed TiO2 NRs with a diameter of *80 nm and a distancebetween the neighbouring NRs of *250 nm (Fig. 4.18b). Such a layer cansimultaneously serve as a light scattering layer further increasing the light har-vesting efficiency.

Photochemical deposition of sensitizer NPs. The most papers on the photo-catalytic deposition of narrow-bandgap NPs onto wider-bandgap photoactivescaffolds are focused on metal-sulfide NPs. The metal sulfide photodeposition istypically achieved via the photocatalytic decomposition of sulfur-containing metalcomplexes or via the photocatalytic reduction of elemental sulfur [42]. The formerapproach can be exemplified by the formation of TiO2/MoS2 and TiO2/WS2heterostructures via the photocatalytic decomposition of (NH4)2MoS4 and(NH4)2WS4 complexes on the surface of nanocrystalline titania (Fig. 4.19a) [133,134]. The photoprocess involves the central ion reduction with the photogeneratedtitania CB electrons followed by the deposition of nanocrystalline MoS2 or WS2.

The photocatalytic reduction of elemental sulfur in ethanol solutions in thepresence of the nanocrystalline ZnO or TiO2 and a corresponding metal salt was usedfor the preparation of colloidal ZnO/CdS [135] and ZnO/ZnS NPs [136], as well ascomposite films of TiO2/PbS [137–139] (Fig. 4.19b), TiO2/CdS [138, 140–142](Fig. 4.19c), TiO2/Ag2S [143, 144] (Fig. 4.19d), and TiO2/CuxS [138]. When aTiO2/Au

0 nanostructure is used as a photocatalyst, a ternary TiO2/Au/CdS

Fig. 4.18 a Photocurrent density JSC and normalized JSC as functions of the CdSe NP layerthickness on the TiO2 film surface; b Illustration of an optimal TiO2 scaffold structure for SSSCs.Adapted (a) and reprinted (b) with permissions from Ref. [132] Copyright (2012) AmericanChemical Society

194 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

heterostructure can be easily produced where cadmium sulfide is deposited as a thin,1–2 nm, layer on the surface of gold nanocrystals [145] (Fig. 4.20a, b).

The size of photodeposited metal-sulfide NPs can be controlled by varying thephotocatalytic reaction conditions, in particular, the illumination intensity andduration, the reactant concentrations, the composition and morphology of thephotocatalyst, etc. [42, 139, 144–146]. For example, by changing the duration ofthe photocatalytic CdS deposition on TiO2/Au heterostructure the size of CdS shellsgrown on the Au cores can be varied in a broad range (Fig. 4.20a) [145]. The sizeof photodeposited NPs can be also tuned by introducing a stabilizer that restricts thegrowth of the photoreaction product [42, 139]. In particular, by decreasing theMAA concentration from 0.04 M to zero the average size of photocatalyticallydeposited PbS NPs can be increased by more than an order of magnitude—fromaround 5 to 70 nm [139]. The introduction of MAA during the photodeposition ofcadmium sulfide on the CdS nuclei (pre-deposited by the SILAR) reduced theaverage size of final CdS NPs from 6 to 4 nm [142].

A detailed study of a TiO2/Au/CdS heterostructure produced by the photocat-alytic deposition [145] showed that cadmium sulfide is deposited as a thin layerpredominantly on the surface of Au NPs resulting in Au/CdS core/shell composites(Fig. 4.20b). This phenomenon was explained by efficient separation of the pho-togenerated electrons and holes between the Au NPs and the nanocrystalline titaniasupport, respectively [145]. At that, the reduction of sulfur (or cadmium) with the

Fig. 4.19 TEM/HRTEM images of the photocatalytically deposited MoS2 NPs (a), PbS NPs (b),CdS NPs (c), and Ag2S NPs (d) on the nanocrystalline titania surface. Reprinted with permissionsfrom Refs. [134] (a), [137] (b), [140] (c), and [143] (d). Copyright (2011 (a, b, d) and 2009 (c))Elsevier (a), The Royal Society of Chemistry (b), and American Chemical Society (c, d)

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 195

photogenerated electrons takes place mostly on the surface of Au NPs resulting inthe formation of a CdS shell.

Similar effects of the charge carrier separation were supposed to account for theformation of spatially-organized ZnO/CdS [135, 146] and TiO2/CdS [138]heterostructures, where CdS is present as CdS NTs [135, 146] or NRs [138](Fig. 4.20c). When cadmium sulfide starts to deposit on the oxide surface, a ZnO–CdS (TiO2–CdS) heterojunction forms where the photogenerated electrons andholes can be spatially separated. As the oppositely charges carriers are attracted toeach other, the photoinduced redox reactions occur predominantly at the oxide-CdSinterface resulting in the growth of new CdS NPs at the same place and geometricalenvironment yielding ordered NTs and NRs. As a result, the morphology of pho-tocatalytically produced TiO2/CdS composites differs drastically from the mor-phology of similar heterostructures synthesized by the conventional CBD procedure(Fig. 4.20c, d). Also, no effects of spatial organization of the photodeposited NPswere observed when the deposited metal-sulfide is photochemically-passive andcannot supply the photogenerated charge carriers to the oxide-sulfide heterojunctionas, for example, in the case of the photodeposited TiO2/CuxS heterostructures(Fig. 4.20e).

The mechanism of metal-sulfide (MS) NPs deposition on the surface ofmetal-oxides (M/Os) via the photocatalytic sulfur reduction in ethanol can bepresented as follows [42]:

Fig. 4.20 a TEM images of TiO2/Au/CdS heterostructure produced at a different duration of thephotocatalytic CdS deposition tp; b HRTEM image of a Au/CdS core/shell NP on titania surface;c–e Atomic force microscopic images of TiO2/CdS (c, d) and TiO2/CuxS (e) composite filmsproduced by the photocatalytic deposition (c, e) and CBD (d). Reprinted with permissions fromRefs. [145] (a, b) and [138] (c–e). Copyright (2006 (a, b), 2009 (c–e)) Nature Publishing Group(a, b) and Elsevier (c–e)

196 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

M=O + hm ! M=O e�CB þ hþVB

� �;

2hþVB + CH3CH2OH ! 2Hþ + CH3CHO,

2e�CB þ Sx x ¼ 1. . .8ð Þ ! Sx�1 + S2�; ð4:8Þ

M2þ þ 2e�CB ! M0; ð4:9Þ

M2þ + S2� ! MS, M0 + S0 ! MS:

The photoprocess can proceed by two routes—via the direct reduction of sulfurto S2− with the photogenerated CB electrons [an “ionic” route (8)] or, alternatively,via the reduction of metal cations to M0 [“atomic” route, (9)], or via both routessimultaneously [42].

The feasibility of the photocatalytic deposition for the formation of SSSCphotoanodes was first shown for the TiO2-based systems [137, 141, 143, 144].Recently, nanocrystalline ZnO/CdS heterostructures were also shown to be quiteefficient photoanodes of the liquid-junction SSSCs [147, 148]. In both cases, thephotoanodes produced by the photocatalytic deposition revealed an increased PECactivity as compared to similar heterostructures formed using the ex situ synthe-sized CdS NPs or by the SILAR, as shown in Table 4.4 on the example oftitania-based SSSCs [141].

Figure 4.21 shows some time-resolved PEC responses from the TiO2/CdS andZnO/CdS heterostructures produced by the SILAR and photocatalytic depositionand having similar composition and optical properties [147, 148]. The illuminationof ITO/TiO2/CdS or ITO/ZnO/CdS photoanodes immersed into aqueous 0.01 MNa2S electrolyte by the “white” light with k > 400 nm results in a rise of photo-voltage and photoinduced current between the photoanode and a Pt counter elec-trode. The photovoltage is roughly the same for the ITO/TiO2/CdS films producedby both methods (Fig. 4.21a) which is expected for the systems with the similarchemical composition.

At the same time, the sensitization of both TiO2 and ZnO via the photocatalyticdeposition of CdS NPs results in much higher photocurrent densities as comparedto the SILAR-produced analogs (Fig. 4.21b, c). In the case of ITO/TiO2/CdS, theabsorption-normalized photocurrent density generated by the photochemicallyproduced anode is 5 times higher than for the SILAR-produced heterostructure(Fig. 4.21b), while in the case of ITO/ZnO/CdS the photodeposited cadmium

Table 4.4 Performance of SSSCs based on TiO2/CdS photoanodes produced by differentmethods [141]

CdS NP deposition method Jsc, mA � cm−2 Voc, V FF η (%)

Photocatalytic deposition 6.5 0.7 0.7 2.5

SILAR 2.7 0.7 0.7 1.2

Adsorption of ex situ synthesized CdS NPs 0.5 0.6 0.6 0.15

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 197

sulfide NPs reveal twice as high efficiency as CdS NPs formed by SILAR(Fig. 4.21c). The photoexcitation of non-sensitized ZnO films does not produce anyappreciable photo-response (Fig. 4.21c, curve 3).

The observations show that the metal oxide—cadmium sulfide heterostructuresproduced by the photocatalytic deposition of CdS NPs are capable of more efficientspatial separation of the photogenerated electrons and holes between the compo-nents as compared to the products of SILAR procedure.

The conclusion is strongly supported by the results of a comparativetime-resolved laser photolysis study of TiO2/CdS composites produced by thephotodeposition and CBD [149] and discussed in details in Chap. 6. The reportshowed that primary separation of the photogenerated charge carriers and formationof intermediates—Ti3+ in the nanocrystalline titania (a trapped electron) and S•− inthe CdS NPs (a trapped hole) occurs by an order of magnitude more efficient for thephotochemically-formed TiO2/CdS as compared to the analog produced by CBD(the atomic force microphotographs of both heterostructures are depicted inFig. 4.20c and d, respectively).

The photocurrent density generated by the illuminated ZnO/CdS heterostructuresincreases in a direct proportion to the amount of sensitizer NPs, which, in turn,depends on the SILAR cycle number N and the photodepositon duration(Fig. 4.22a).

However, after the normalization to the light absorbance, the photoanodes dif-fering in the CdS NP content show more or less the same efficiency of lightconversion (Fig. 22b). As shown earlier in many examples, as well as in [147, 149]for ITO/ZnO/CdS heterostructures, an increase in the SILAR cycle number resultsin a considerable growth of the CdS NP size. The results presented in Fig. 4.22bshow, therefore, that the size factor is of low importance for the ZnO/CdSheterostructures and the overall light conversion efficiency is affected rather by thesensitizer content than by the NP dimensions. The reason for the lack of size

Fig. 4.21 Temporal changes of voltage (a) and current (b, c) induced by the illumination (“hv on”moments)/extinction (“hv off” moments) registered for ITO/TiO2/CdS (a, b) and ITO/ZnO/CdS(c) photoanodes produced by the SILAR (curves 1) and photocatalytic deposition (curves 2).Curve 3 (c) corresponds to the bare ITO/ZnO film

198 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

dependence can be in very favorable conditions for the spatial charge separationthat exists even in the bulk ZnO–CdS and TiO2–CdS heterojunctions and should beeven more advantageous in the case of nanocrystalline semiconductors.

The TiO2/Ag2S nanoheterostructures produced by the photocatalytic silver sul-fide deposition were successfully tested as photoanodes for the hydrogen-evolvingPEC solar cells [143]. The bandgap of Ag2S NPs was found to become narrowerwith an increase in the photodeposition duration indicating an increase in the silversulfide NP size. The highest efficiency of the solar light harvesting was observed forAg2S NPs with Eg = 1.75 eV [143].

The silver sulfide NPs can also be produced by the sulfidation of AgNPs depositedvia the photocatalytic Ag+ reduction. The TiO2/Ag2S films produced by this methodfrom titania NTs showed a light conversion efficiency of 1.23% when applied as aphotoanode in a liquid-junction SSSC with the polysulfide electrolyte [144].

The TiO2/PbS heterostructures can be prepared by the photocatalytic lead sulfidedeposition from ethanol solutions of lead perchlorate and S8 [137]. Such filmsshowed a photocurrent density of 1.71 mA/cm2 (at 0 V vs. Ag/AgCl) when illu-minated by the AM1.5 light in aqueous solutions of Na2S and Na2SO3. Additionalexamples of SSSCs based on the photocatalytically-produced photoanodes arepresented in Table 4.5.

The UV-illumination of titania films immersed into degassed ethanol solutionsof S8 and SbCl3 results in the deposition of amorphous antimony sulfide, theamount of Sb2S3 growing with an increase in the photodeposition duration [150].The annealing in N2 atmosphere yields crystalline TiO2/Sb2S3 heterostructures thatcan be used as visible-light sensitive SSSC components.

Fig. 4.22 Dependence of the photocurrent density (a) and absorbance-normalized photocurrentdensity (b) generated by the ITO/ZnO/CdS photoanodes on the SILAR cycle number (curves 1)and the photodeposition duration (curves 2)

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 199

The amorphous antimony sulfide is deposited in the form of spheroidal particleswith a size of 150–300 nm making Sb2S3 NPs clearly visible in the background ofmuch smaller titania nanocrystals (Fig. 4.23a, b). According to theenergy-dispersive X-ray spectroscopy (EDX) analysis, the atomic Sb-to-S ratio,1:1.3, is close to the expected stoichiometric value. The TiO2/Sb2S3 films retaintheir morphology after the annealing (Fig. 4.23c).

Along with the nanometer-sized deposits, some much larger spherical particlescan be observed on the surface of TiO2/Sb2S3 films at a lower SEM magnification

Table 4.5 Some examples of SSSCs produced by the photodeposition of sensitizer NPs

Photoanode CE I, mW � cm−2 Redox-couple species η (%) Reference

TiO2/PbS Pt 100 Na2S/Na2SO3 0.16 [137]

TiO2/CdS Pt 100 I−/I2 2.51 [141]

TiO2/Ag2S Pt 100 Na2S/Na2SO3 0.29 [143]

TiO2 NR/Ag2S Pt 10047

S2−/Sx2− 0.19

1.27[144]

ZnO NR/CdS/CdSe CuxS 100 S2−/Sx2− 2.03 [241]

Fig. 4.23 SEM images of starting nanocrystalline TiO2 film (a) and TiO2/Sb2S3 (b–d) films withamorphous (b) and crystalline Sb2S3 (b). Image (d) was taken at a lower magnification. e SEMimage of photodeposited metallic Sb; f Photocurrent density Jphoto at 0.1 V versus Ag/AgCl forFTO/TiO2 and FTO/Sb2S3 electrodes. Photodeposition time is indicated on the bars. Reprintedfrom Ref. [150] with permissions. Copyright (2015) Elsevier

200 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

(Fig. 4.23d). The EDX analysis of such spheres showed them to be stronglyenriched with antimony (Sb:S is 13:1 in the bulk of the spheres and 5:1—on theirsurface). Also, the photodeposition in the absence of sulfur produced similar butstrongly aggregated microparticles of metallic antimony (Fig. 4.23e). The presenceof such particles in the photodeposited TiO2/Sb2S3 composites can, therefore, betaken as an indication that the photocatalytic deposition of antimony sulfide pro-ceeds, most probably, via the atomic route of Sb0 formation followed by theantimony sulfidation with S8.

Because of the ready dissolution of antimony sulfide in the polysulfide elec-trolytes, the TiO2/Sb2S3 composites require alternative electron-shuttling couples tobe used or, alternatively, a sacrificial electron donor that is consumed irreversiblysupplying electrons to the photoanode, such as ascorbic acid [150]. The lightconversion efficiency of the composites with crystalline antimony sulfide under theillumination with the “white” light in aqueous solutions of ascorbic acid grows withan increase in the sensitizer content, that is with an increase in the photodepositionduration (Fig. 4.23f). At the same time, the amorphous antimony sulfide depositsrevealed no photoactivity even showing an adverse light-shielding effect on thetitania.

Preparation of SSSC photoanodes by ions exchange. The ion exchange (IE) isa quite straightforward method for the formation of various metal oxide/metalchalcogenide nanoheterostructrures, that is typically applied to chemically unstablezinc oxide. Both Zn2+ ions and O2− anions in the ZnO lattice can be substituted byother metal cations and chalcogenide anions, respectively, producing less solublemetal chalcogenides. For example, the nanocrystalline ZnO films immersed intoaqueous solutions of Se2−-generated species (Na2SeO3 + NaBH4) gain yellowcolor indicating the formation of zinc selenide [151]. The extinction of ZnO/ZnSefilms in the visible spectral range increases with an increase of the ZnO layerthickness (proportional to the electrodeposition duration of original zinc oxide films[151]) as more and more ZnO is converted into zinc selenide (Fig. 4.24a).

The presence of a ZnO-related shoulder in the extinction spectra of ZnO/ZnSeheterostructures indicates a partial character of the ion exchange. An analysis of thespectral curves showed that the bandgap corresponding to the new spectral featurein the visible range is 2.74–2.75 eV, which is typical for bulk hexagonal zincselenide [152, 153]. The partial transformation of ZnO into ZnSe does not induceappreciable change in the morphology of zinc oxide films (Fig. 4.24b, c). An EDXanalysis of the film cross-section (Fig. 4.24d) showed that selenium is distributedevenly in the volume of the films (Fig. 4.24e) mimicking the distribution of zincand oxygen (Fig. 4.24f, g) and indicating that the entire ZnO film is accessible forthe IE reaction.

The IE procedure can be repeated again producing ternary and more complexoxide/chalcogenide nanocomposites. For example, ZnO nanowires (NWs) wereconverted into ZnO/ZnSe heterostructures by a partial IE of oxygen (Fig. 4.25(I)a,b), then Zn2+ ions in the ZnSe layer were partially substituted with Cd2+ thusproducing ternary ZnO/ZnSe/CdSe nanocomposites retaining the wire-like shape ofthe original zinc oxide NWs (Fig. 4.25(I)c–e) [43, 154].

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 201

The morphology of heterostructures and amount of the incorporated cadmiumselenide depend on the temperature of the ZnSe conversion into CdSe [43]. Bysubstituting O2− in ZnO nanocables with sulfide ions ZnO/ZnS composites wereproduced, which then were transformed into ZnO/CdS by exchanging Zn2+ with

Fig. 4.24 a Extinction spectra of ZnO (curve 1) and ZnO/ZnSe (curves 2–5) films produced bythe IE from zinc oxide films electrodeposited during 2 min (curve 2), 4 min (3), 6 min (4), and10 min (5). See deposition conditions in [151]; b–d SEM images of ZnO/ZnSe film, e–g atomdistribution maps for Se (e), Zn (f), and O (g) produced by the EDX analysis [151]. The size ofimages in (d–g) is around 10 � 10 lm

Fig. 4.25 I SEM images of (a) a ZnO NW array, b ZnO/ZnSe NW, c, d Zn0.7Cd0.3Se NWsprepared by reacting a ZnO/ZnSe nanocable with Cd2+ at 50 °C (c), 90 °C (d), and 140 °C (e). IIDiffused reflectance spectra of (a) as-prepared ZnO NR; b CdS/ZnO nanocable, c ZnO/CdS0.61/CdSe0.39 (50 °C) nanocable, and d ZnO/CdS0.24/CdSe0.76 (90 °C) nanocable arrays; III Scheme ofthe fabrication process of ZnO/AgInS2 NR arrays [158]. Reprinted with permissions from Refs.[43] (I), [155] (II), and [158] (III). Copyright (2011–2014) American Chemical Society

202 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Cd2+ (Fig. 4.25(II), curves a, b). Finally, sulfur in CdS was partially substitutedwith selenium resulting in a considerable red shift of the absorption edge of theprospective ZnO/CdSxSe1−x photoanode (Fig. 4.25(II), curves c, d) due to theformation of an alloyed cadmium sulfoselenide layer [155].

The IE transformation of ZnO is often used to form a thin blocking zincchalcogenide layer preventing the charge leakage and the recombination in thephotoanodes prepared by the following deposition of ex situ synthesized sensitizerNPs. In this way, ZnS and ZnSe blocking layers were formed on the ZnO surfaceprior to the deposition of the visible-light-harvesting Cu2ZnSnS4 NPs [156].

The ion exchange methods were probed for the preparation of Cd- and Pb-freephotoanodes of the liquid-junction SSSCs. For example, by substituting Zn2+ inZnO/ZnS heterostructures by Ag+ the ZnO/Ag2S composites can be produced,which can then be subjected to a partial IE with Sb3+ to produce ZnO/AgSbS2heterostructures [157]. In a similar way, by introducing In3+ into ZnO/Ag2Sheterostructure produced by the IE, visible-light-sensitive ZnO/AgInS2 photoan-odes were formed (Fig. 4.25(III)) [158].

4.5 Making Progress in SSSCs—Toward More Efficientand Less Toxic Photoelectrodes

The photoelectrochemical SSSCs are nowadays in a constant progress and steadyefforts are applied to increase their efficiency in the solar light harvesting and theirattractiveness as compared with competing photovoltaic technologies. These effortscan be categorized into several main trends that will be discussed in this subsection.

The photocurrent generation efficiency is limited to a far extent by theelectron-hole recombination in the sensitizer NPs as well as on the interfacesbetween the wide-bandgap scaffold and the electrolyte and between the metal oxideand the sensitizer NPs. To suppress the recombination and minimize losses of thephotogenerated charges various approaches are developed, one of the most simpleand, at the same time, efficient being the formation of additional “protective” or“buffer” semiconductor layers either on the sensitizer NP surface, or between thesensitizer and the oxide scaffold, or between the metal oxide and OTE [s066, s067,s068, s070].

As discussed in details in Chap. 1, the properties of sensitizer NPs can changedramatically in a critical size range of around 1–10 nm as a result of the quantumsize effects. This feature opens possibilities of enhancing the photoresponse ofsensitizer NPs, that is increasing the energy and transfer rate of CB electrons andVB holes through variations of the sensitizer NP size. By combining severalsize-selected sensitizer NPs in a “cascade” structure with an outer NP layer havingthe highest CB potential and an inner, closest to the scaffold, layer having thelowest CB potential, the directed transport of the photogenerated CB electrons canbe organized, from the smaller NPs in the outer NP layer to the larger NPs in theinner layer to the metal oxide scaffold [159–161]. Such a cascade structure allows

4.4 Nanocrystalline Photoanodes Produced by the In Situ … 203

suppressing the recombinative processes as the CB electron gets physically sepa-rated from the VB hole and cannot come in the reverse direction through thepotential barrier. Similar cascade structures can also be organized by using differentsensitizer materials with correspondingly matched CB levels [159–161].

Finally, the attractiveness of SSSCs can be enhanced by utilizing new nontoxicand Earth-abundant sensitizer materials instead of Cd- and Pb-based NPs thatdominate today in the photoelectrochemical SSSCs [159, 160, 162]. The lowtoxicity of cell components is an important issue that can even counterweight thehigh efficiency of such heavy-metal containing materials as CdSe or PbS. As a vividexample, the organo-inorganic Pb-containing perovskites can be mentioned, such asCH3NH3PbHal3 (Hal = Cl, Br, I). The perovskites became a rapidly rising star ofthe photovoltaics making progress in the light harvesting efficiency from severalpercents in 2009 to more than 21% in the recent years [22, 26, 163–171]. However,the organo-inorganic perovskites suffer from chemical and photochemical insta-bility and the problem of possible lead leakage is a grave concern that can impede abroad implementation of the photovoltaic technologies based on such materialsdespite their high efficiency. In this view, constant efforts are applied in thescreening and testing of new narrow-band-gap semiconductor materials, in par-ticular among the more complex ternary and quaternary metal chalcogenides in thehope of finding reasonably efficient, abundant and low-toxic light-harvestingmaterials [159–162, 172, 173].

Finally, the metal oxide scaffold, though not participating directly in the lightharvesting, can strongly influence the rate of secondary charge transfer processesand limit the total light conversion efficiency of a SSSC. Also, the enhanced lightscattering from specially designed metal oxide nano-architectures can influence in apositive way the light harvesting efficiency of the cell as a whole. In this view,constant efforts are applied for the design of new morphologies of metal oxidescaffolds favoring to the accommodation of sensitizer NPs and affecting their lightabsorption.

Suppression of the recombination by barrier layers. Charge losses in thephotoanodes consisting of the mesoporous wide-bandgap scaffold and sensitizerNPs can originate from several recombination processes, in particular, (i) theelectron-hole recombination in the volume of NPs, (ii) the recombination of anelectron injected into the metal oxide scaffold with a hole left in a sensitizer NP, and(iii) “leakage” of electrons migrating along the mesoporous network of TiO2

(ZnO) toward OTE as a result of the interactions with the electrolyte that permeatesthe whole photoanode volume. Each of the above-discussed recombination path-ways can be addressed separately by introducing a special barrier layer that hindersthe recombinative charge losses but at the same time does not impede the direc-tional electron transfer from the sensitizer NPs to the metal oxide to the OTE.

Suppression of the recombination in the sensitizer NPs. The recombination in thesensitizer NPs occurs predominantly via structural defects introducing additionalstates in the NP bandgap and allowing for the thermal or radiative dissipation of thelight excitation energy. As opposite to bulk counterparts, NPs have a much largersurface-to-volume ratio and a lot of structural defects reside on the NP surface.

204 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

These defects (unsaturated bonds, cation/anion vacancies, etc.) can be passivated byappropriate ligands or a shell of other metal chalcogenide semiconductor, typicallywith a larger bandgap, such as ZnS. The passivating layers can be very conve-niently deposited by the SILAR procedure allowing for a quite precise control overthe thickness of the barrier layer. For example, the light conversion efficiency ofTiO2/PbS heterostructures can be enhanced by the deposition of a CdS layer, bothlead sulfide and cadmium sulfide layers deposited by the SILAR [103, 174]. As theSILAR cycle number is increased to 4–5 the photocurrent increases, reaches asaturation value and then decreases because a too thick protective CdS shell exerts alight-shielding effect on the PbS NP sensitizer (Fig. 4.26a) [103].

An additional *60% enhancement of the light conversion efficiency on TiO2/PbS photoanodes can be achieved by the SILAR deposition of a mixed PbxCd1−xSlayer prior to the deposition of the passivating CdS shell [104].

A passivation effect can be achieved by the deposition of a ZnS layer ontoCuInSe2 sensitizer NPs anchored to the titania surface [175]. The zinc sulfide layerincreases the photostability of sensitizer NPs and shifts the NP absorption edge tolonger wavelength allowing for a more efficient harvesting of the solar light. Theeffect is caused by a partial penetration of the wavefunctions of photogeneratedcharge carriers from the 4-nm CuInSe2 NPs into the ZnS layer resulting in aweakening of the exciton confinement effect in the sensitizer NPs. In this case, thereexists an optimal number of the SILAR deposition cycles (Fig. 4.26b), because athicker ZnS shell impedes the electron transfer from the electrolyte species toCuInSe2 NPs. A similar passivation effect of a ZnS shell was observed also forCdS/CdSe NPs [176].

CdTe NPs can be passivated with a CdSe shell that contributes to the lightabsorbance and serves as an electron acceptor resulting in photogenerated electronand hole separation in the CdSe shell and CdTe core, respectively [66].

Fig. 4.26 a Total light conversion efficiency of TiO2/PbS/CdS-based (a), TiO2/CuInSe2-based(b) and TiO2/CuInS2-based (c) SSSCs as a function of the SILAR number of PbS and CdSdeposition (a) and ZnS deposition (b) as well as the nature of protective shell (c). Reprinted/adapted with permissions from Refs. [103] (a), [175] (b), and [107] (c). Copyright (2011–2015)American Chemical Society

4.5 Making Progress in SSSCs—Toward More Efficient … 205

By choosing appropriate shell materials and deposition sequence the lightconversion efficiency of TiO2/CuInS2-based SSSCs was increased from *1% forthe bare sensitizer NPs to *4.5% for a layered CuInS2/CdSe/ZnSe heterostructure(Fig. 4.26c) [107]. A higher efficiency of the ZnSe-based photoanode as comparedto the ZnS-based counterpart originates from the enhanced visible light absorbanceof zinc selenide contributing to the overall solar light harvesting. Similar effects ofincreased light absorbance and recombination suppression account for a higher lightconversion efficiency of TiO2/CdS/CdSe photoanodes covered with a ZnSe pro-tective layer (η = 6.4%) as compared to a similar heterostructure passivated with aZnS layer (η = 4.9%) [177].

Alternatively, the surface states of sensitizer NPs responsible for the recombi-nation losses can be passivated by molecular ot ionic species, thus leaving thesurface of sensitizer NPs fully open for interactions with redox species in theelectrolyte. For example, a passivation effect was observed upon the adsorption ofaliphatic amines on CdS NPs, which increases the light conversion efficiency from1.45 to 2.35% [178].

Adsorption of a layer of 4-tert-butylpyridine on a TiO2/CdS/CdSe photoanodepassivated with a ZnS shell adds around 1% to the cell efficiency [176]. However,the amine adsorption prior to the sensitizer NP deposition results in the deteriorationof the cell performance due to intervention of the amine layer into the chargetransfer between TiO2 and CdS/CdSe NPs. A similar adverse effect on the cellperformance has also the adsorption of a layer of electron-withdrawing molecules,such as 4-cyanopyridine [176].

A layer of mercaptophenol deposited onto TiO2/PbSe heterostructures wasreported [117] to improve the hole tunneling to the electrolyte from lead selenideNPs thus contributing to the light harvesting efficiency. The recombination ofphotogenerated charge carriers in TiO2/Sb2S3 heterostructures can be suppressed bythe deposition of an outer shell of a conjugated polymer—poly-3-hexylthiopheneresulting in an enhancement of the light conversion efficiency from 3.2 to 4.2%[27]. Lead sulfide NPs can be passivated by adsorption of halogenide ions [179,180], most probably due to the formation of a thin surface layer of lead halogenides.

Suppression of the recombination on the interface between the sensitizer NPsand the metal oxide scaffold. Provided the CB levels of the sensitizer NPs and TiO2

(ZnO) are favorable for the electron transfer from NPs to oxide, the sensitizer NPphotoexcitation results in the extremely fast electron transfer from NPs to theneighboring TiO2 (ZnO) layer, leaving a hole in the NP valence band. Potentially,the electron in the titania CB can recombine with the hole in the sensitizer VB,similarly that it occurs in the DSSCs after the electron injection from the pho-toexcited dye-sensitizer to the titania scaffold. To prevent such charge losses ablocking layer is often introduced between the sensitizer NPs and TiO2

(ZnO) constituting a potential barrier for the injected electron on its way back to theparental metal chalcogenide NPs.

Similarly to the above discussed passivation of TiO2/CdS photoanodes with anouter ZnS shell, the deposition of an intermediary zinc sulfide layer between titaniaand CdS or CdxZn1−xS NPs results in an increment of the light conversion

206 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

efficiency from 3.06 to 3.69% as a result of shielding of the titania layer both fromthe charge leakage to the electrolyte and against the reverse electron transfer to thesensitizer NPs [88]. Improved charge collection in the TiO2/ZnS/CdS/ZnSheterostructure with the inner and outer ZnS protective shells can be clearlyexemplified by the electrical impedance spectra (Fig. 4.27a). The larger is theradius of the Nyquist plots for a given photoanode the larger is the electriccapacitance Cl on the photoanode/electrolyte interface, which is a quantitativemeasure of the charge collection efficiency. The Cl values obtained for TiO2/CdS,TiO2/CdS/ZnS, and TiO2/ZnS/CdS/ZnS by a simulation of the impedance spectrawith an equivalent circuit were 1516, 2217, and 2586 lF, respectively, indicatingon the better charge collection for the double passivated photoanode due to lowerelectron-hole recombination [88].

A screening search for potential materials for a barrier layer between TiO2 andCuInS2 (CIS) NPs showed that a number of metal chalcogenides can suppress thereverse electron transfer from TiO2 to CIS. These include cadmium, copper andindium chalcogenides, the most efficient being CuxSe and In2Se3 allowing toincrease the light conversion efficiency by a factor of 3 and higher (Fig. 4.27b)[107]. The electron-hole recombination suppression by buffer layers was provedunambiguously by an increase of the open-circuit voltage. The highest efficiency ofindium selenide barrier layer was explained by a combination of favorable factors,including a proper band alignment (the CB of In2Se3 stays between the CB levels ofTiO2 and CIS NPs), formation of an intermediate CuInSxSe1−x layer allowing for abetter orientation of crystalline planes of CIS NPs with respect to those of titaniaNPs, and, finally, by a layered character of indium selenide that “smears” uniformlyon the titania surface.

Similarly to zinc sulfide, CdS layers can be placed between the titania scaffoldand CIS NPs [82] and on top of binary TiO2/CIS heterostructure [52] resulting inboth cases in an improvement of the cell performance.

Fig. 4.27 a Electrical impedance spectra of SSSCs with different photoelectrodes in the form ofNyquist-plots. b Light conversion efficiency (η) and open-circuit voltage (Voc) for TiO2/CuInS2photoanodes with different buffer layers placed between TiO2 and CuInS2 NPs. Reprinted/adaptedwith permissions from Refs. [88] (a) and [107] (b). Copyright (2016) The Royal Society ofChemistry (a) and (2013) American Chemical Society (b)

4.5 Making Progress in SSSCs—Toward More Efficient … 207

The light conversion efficiency on Ag2S NP-decorated TiO2 NT arrays can beincreased from 0.22 to 0.28% by placing a barrier recombination-blocking ZnOlayer between titania and sensitizer NPs [106]. Alternatively, a TiO2 NP layer canbe inserted between ZnO NR scaffold and CdS/CdSe NPs to suppress the electronleakage from zinc oxide to the electrolyte. This approach results in a spectacularincrease of the light conversion efficiency from 1.54 to 3.14% [130].

A Mg-doped ZnO layer placed between the ZnO scaffold and PbS NPs canefficiently suppress the backward electron transfer from zinc oxide to the metalchalcogenide NPs due to a favorable cascade CB level positions of all threecomponents [181].

Suppression of the charge leakage from the sensitizer NPs and the metal oxidescaffold to the electrolyte. After the injection from photoexcited sensitizer NPs toadjacent mesoporous TiO2 (ZnO) layer, the electron migrates through the networkof contacting metal oxide NPs till it reaches OTE and comes into the electric circuit.At that, a possibility exists for the electron to be captured by the components of theelectrolyte, for example, by water or H3O

+ ions that can reduce the photoanodeperformance considerably. To avoid such losses the TiO2/NPs heterostructures aretypically covered with an additional protective layer of wide-bandgap materials,most often, zinc sulfide that creates a barrier for the electron to reach the electrolyte.In the case of ZnO-based photoanodes such passivation can occur directly in thepolysulfide electrolyte as a result of a partial anion exchange and in situ ZnOtransformation into ZnS. A more general approach consists in the formation of athin ZnS layer by several SILAR deposition cycles. The protective ZnS layer alsoprovides a stronger contact between the sensitizer NPs and TiO2 (ZnO) as well asprotects the light-harvesting metal chalcogenide NPs from corrosive processes thatcan occur during the photoelectrochemical events in the SSSC.

In the case of titania scaffolds, the photoanode can be relatively easily “insu-lated” from the charge leakage by soaking with TiCl4, followed by the hydrolysisand annealing. As a result, a thin and evenly distributed layer of TiO2 NPs isformed on the photoanode surface. The titania layer prevents charge transfer to theelectrolyte from both the sensitizer NPs and the metal oxide scaffold as well asprovides a better contact between the sensitizer NPs and the metal oxide transportlayer. This procedure resulted in more than 150% increment of the light conversionefficiency when applied to TiO2/CdS photoanodes [182].

Cascade designs of SSSCs. The cascade design of photoanodes/photocathodesof SSSCs can be realized in several ways, in particular, (i) by using several differentsemiconductor photo–electrode materials with favorable CB and VB level offsets;(ii) by using alloyed solid solution compounds with a varied or spatially gradientstructure to create a CB (VB) offset from the outer surface of the photoelectrodetoward OTE, and (iii) by combining NPs of the same semiconductor but of differentsize that reveal a strong size-dependence of the CB and VB levels and placing themin the order of decreasing CB (VB) potential from the outer photoelectrode surfacetoward OTE.

Cascades of different semiconductor NPs. The formation of a cascade of two andmore different semiconductor NPs aimed to produce a descending gradient of the

208 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

CB level (or an ascending gradient of the VB level) is, probably, the moststraightforward way of achieving enhanced spatial separation of the photogeneratedcharge carriers and suppressing their recombination in the sensitizer NPs. Suchcascade effects were reported for TiO2/ZnO/CdS [183], TiO2/CdS/CdSe [116, 122],SnO2/TiO2/CdS/CdSe [184], ZnO/CdS/CdSe [185], TiO2/PbS/CdS [103, 104],TiO2/CuInS2/CdS [69, 186], and TiO2/ZnIn2S4/CdS [187]. Typical CB/VB levelalignments in some successful cascade systems are presented in Fig. 4.28.

A ternary cascade system of PbSe, CdS, and carbon NPs (the so-called carbondots) was applied as a sensitizer for the titania scaffolds (Fig. 4.28b) [188] resultingin almost 5% efficiency of the solar light conversion and the spectral sensitivityextending over 1000 nm.

Fig. 4.28 Energy schemes of some successfully realized cascade photoelectrode designs—SnO2/TiO2/CdS/CdSe (a), TiO2/ZnIn2S4/CdS (b), TiO2/PbSe/CdS/C (c), and TiO2/CuInS2/CdS (d).Reprinted with permissions from Refs. [184] (a), [178], (c) [188] (d) [186]. Copyright (2011,2015). The Royal Society of Chemistry (b–d)

4.5 Making Progress in SSSCs—Toward More Efficient … 209

As a rule, the cascade effect results in a non-additive enhancement of the lightconversion efficiency as compared to the sum of efficiencies of the SSSCs based onseparate components. For example, in a TiO2/CdS/CdSe cascade the reported lightconversion efficiency is much higher (3.2% [116], 4.81% [122]) than the sum ofefficiencies in TiO2/CdS-based (0.39% [116], 1% [122]) and TiO2/CdSe-basedSSSCs (2.29% [116], 2.69% [122]).

Cascades of alloyed metal chalcogenide NPs. Some metal chalcogenide semi-conductors, for example, CdS and ZnS, CdS and CdSe, can form solid solutions ofany varied composition and, as a consequence, with varied CB and VB positions.By combining several composition-selected sensitizer NPs a cascade structure canbe arranged for the directed electron migration from the metal-chalcogenide layer tothe metal oxide scaffold and then—to the electric circuit.

By changing the Se-to-S ratio in the mixed CdSxSe1−x NPs one can producemulti-colored, green to orange-yellow to red, light absorbers with a varied bandgap(Fig. 4.29a) [65]. Arranging of the CdSxSe1−x NPs by increasing Eg from the TiO2

surface to the outer photoanode zone results in a cascade structure reaching the lightconversion efficiency of 3%. The quantitative data on relative band edge positionsof titania and composition-selected CdSxSe1−x determined by the UV photoelectronspectroscopy were reported (Fig. 4.29b) [189], allowing for a precise design ofsuch photoanodes. The SSSCs with the composition-selected ZnO/CdxZn1−xSenanocable array photoanodes revealed the light conversion efficiencies of up to4.74% [43].

Some synthetic approaches, for example, the temperature-gradient chemicalvapor deposition [190], or the photochemical transformation [191], were applied toprepare heterostructures with a continuous gradient structure, where the directedcharge flow can be realized as well, for example, ZnO/CdxZn1−xSe [190], ZnO/

Fig. 4.29 a Absorption spectra and photographs (taken under the UV illumination) of colloidalcomposition-selected CdSxSe1−x NPs; b energy diagram of the TiO2/CdSxSe1−x heterojunctionswith different x. Reprinted with permissions from Refs. [65] (a) and [189] (b). Copyright (2012,2013) American Chemical Society (a) and The Royal Society of Chemistry (b)

210 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

ZnOxS1−x [191]. Such heterostructures were [190] or potentially can be [191]applied as the photoanode materials in the SSSCs.

Cascades of the size-selected metal-chalcogenide NPs. In a similar way, dif-ferently-sized NPs of the same semiconductor can be arranged in layers on theoxide scaffold surface in order of increasing Eg enabling the photogenerated elec-trons to migrate from the outer sensitizer NP layer to the metal oxide transport layerand into the circuit. Such a “rainbow-cell” design was successfully realized forTiO2/CdSe photoanodes with the ex situ produced 2.3–3.7 nm NPs [60, 63] as wellas with the in situ deposited CdSe NPs [73]. Differently sized (2.0–4.5 nm) CdTeNPs were also tested in a cascade structure with ZnO NR array scaffolds both asdirect sensitizers and as energy relays for the indirect excitation of an additional dyesensitizer [79].

Lox-toxic alternative sensitizer NPs for SSSCs. A search of newnarrow-bandgap semiconductor materials capable of efficient light harvesting andtheir testing as the light-sensitive components of the photoanodes and photocath-odes is, probably, the current hot spot in the SSSC field. Each and every newcompound that potentially can be used as a light absorber is probed, including new(and sometimes quite exotic) Cd-free and Pb-free metal chalcogenides, ternary andquaternary chalcopyrites and kesterites, emerging materials such as carbon dots andmany others.

Some of new absorber materials were discussed in the sections devoted to theformation of photoanodes using the ex situ and in situ synthesized sensitizer NPs.This subsection collects recent reports on new (promising and potentially promis-ing) sensitizer materials for the SSSCs that were not scrutinized before in thischapter. Some of the SSSCs examples are summarized in Table 4.6.

The most straightforward way to the low-toxic SSSCs is to probe the sensitizerswhich are similar to the dominating cadmium and lead selenides. In this way,alternative binary chalcogenide sensitizers were studied, including Sb2S3 [27, 150],Bi2S3 [89, 100], FeS2 [192], ZnSe [151, 154, 193] and Sb2Se3 [194].

In the SSSC design with new sensitizer materials, not only a correspondencebetween the absorption spectrum of the sensitizer NPs and the solar irradiationspectrum should be taken into account, but also a proper alignment of the CB and/orVB levels of the sensitizer NPs and other photoelectrode components, including themetal oxide scaffold and the passivating layers. For example, in the bismuthsulfide-based systems, the deposition of an intermediary CdS layer between Bi2S3NPs and mesoporous TiO2 film creates a cascade structure (Fig. 4.30a), favoring tothe photoinduced electron transfer from the sensitizer NPs into the circuit, while inthe TiO2/Bi2S3/CdS structure the sensitizer blocks the electron transfer from cad-mium sulfide and thus the efficiency of the light conversion with such photoanode isonly 0.56% as compared to 2.52% for the TiO2/CdS/Bi2S3 photoanode [89].

As discussed in the previous subsections, great expectations in the SSSCdevelopment are associated with the ternary and more complex metal chalcogenideNPs that combine a high absorptivity in the visible and near-IR ranges with anunprecedented flexibility of properties via the variations in the NP composition,size, shape, doping, etc. [173, 195].

4.5 Making Progress in SSSCs—Toward More Efficient … 211

Tab

le4.6

Someexam

ples

ofSS

SCsbasedon

variou

sCd-

andPb

-freeNPsensitizerNPs

Photoano

dematerial

Eg,eV

CE/redox

-cou

ple

J sc,mA

�cm

−2

Voc,V

FF,%

η,%

Reference

ZnO

/ZnS

e/CdS

*2.3

Cu xS/S2

−/0

2.29

440

270.27

[193

]

TiO

2/Sb

2S3

TiO

2/Sb

2S3/P3

HT

1.65

Pt/Co2

+/3+

12.0

530

503.20

[27]

12.2

667

514.20

TiO

2/Ag 2S

1.0

Pt/S

2−/0

10.3

290

330.98

[144

]

ZnO

/ZnS

/FeS

2n/r

Pt/I−/0

0.87

390

360.12

[192

]

TiO

2/BiOI

1.85

Pt/I−/0

1.52

490

510.38

[242

]

TiO

2/CuInS

2/CdS

*1.6

Cu xS/S2

−/0

16.9

560

454.20

[243

]

ZnO

/AgInS

21.8

Pt/I−/0

3.8

540

350.72

[196

]

TiO

2/AgInS

21.8

Au/S2

−/0

4.62

450

390.80

[197

]

TiO

2/AgInS

2/In

2S3

*1.7

Pt/S

2−/0

7.87

320

280.70

[198

]

TiO

2/CuInS

e 2/ZnS

1.22

Cu xS/S2

−/0

26.93

528

578.10

[175

]

TiO

2/CdS

/CuInS

2*1.5-1.6

C/S

2−/0

8.12

489

371.47

[82]

TiO

2/AgS

bS2

1.7

Au/S2

−/0

2.42

320

n/r

0.34

[101

]

TiO

2/CuB

iS2

2.1

Cu xS/S2

−/0

6.87

250

360.62

[128

]

TiO

2/AgB

iS2

1.32

Pt/S

2−/0

7.61

180

390.53

[99]

TiO

2/Cu 2ZnS

nS4

n/r

Pt/I−/0

0.41

560

580.13

[72]

ZnO

/ZnS

e/Cu 2ZnS

nS4

1.5

Cu xS/S2

−/0

10.46

490

432.20

[156

]

TiO

2/CuInS

2:Zn

n/r

Cu xS/S2

−/0

20.65

586

587.04

[71]

TiO

2/CuInS

e 1.4S 0

.6n/r

Cu xS/S2

−/0

10.5

550

603.45

[199

]

TiO

2/CuInS

2/CdS

*1.8

Cu xS/S2

−/0

15.65

529

473.91

[52]

TiO

2/CuInS

2

TiO

2/AgInS

2

Cu xS/S2

−/0

12.82

640

544.20

[83]

9.75

432

652.62

Note1sun(A

M1.5)

ifno

tstated

otherw

ise;

with

outerZnS

layertypically

212 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

For example, the ternary CuInS2 and AgInS2, and correspondingnon-stoichiometric CIS and AIS compounds are used in a constantly broadermanner [52, 53, 69–71, 82, 83, 100, 125, 196, 197], tending to gradually substituteCdS and CdSe in the SSSCs.

In the case of CIS(AIS)-based photoanodes formed by the attachment of ex situsynthesized sensitizer NPs on the surface of metal oxide scaffolds the size of NPscan be finely tuned by adjusting the synthesis conditions and so the size depen-dences of the light conversion efficiency can be conveniently probed. In the case ofCIS/AIS NPs, similarly to other earlier discussed examples, such dependences aretypically volcano-shaped (Fig. 4.31a, blue bars) as a result of a counter-balance ofthe CB level increase and a blue absorption band edge shift both observed with aNP size decrease (Fig. 4.31b), the latter effect resulting in a partial loss of the solarlight absorption [51].

Typically the size-dependence of η mimics the size variation of PL emissionintensity (Fig. 4.31a, red bars) because both the light emission and the photocurrentgeneration compete with the non-radiative electron-hole recombination. Therefore,the PL spectroscopy can be used as a diagnostic tool for assessing/predicting thePEC activity of CIS/AIS-based photoanodes as discussed later in Chap. 6.

The efficiency of AIS-based SSSCs can be boosted by the sensitizer NP passi-vation with indium sulfide protective layers. At that, the highest performance wasobserved for the photoanodes with two In2S3 layers (Fig. 4.30b). The first layer isplaced between AIS NPs and the TiO2 scaffold to mediate the electron transfer fromAIS to titania and simultaneously to impede the reverse electron transfers. Thesecond In2S3 layer is deposited on the photoanode surface followed by the depo-sition of an additional protective ZnS layer—to mediate the hole transfer from AISNPs to sulfide ions in the electrolyte [198].

Fig. 4.30 Schematic energy diagrams of TiO2/CdS/Bi2S3 and TiO2/Bi2S3/CdS heterostructures(a), and TiO2/In2S3/Ag-In-S/In2S3/ZnS photoanode (b). Reprinted with permissions from Refs.[89] (a) and [198] (b). Copyright (2015) Elsevier (a) and The Royal Society of Chemistry (b)

4.5 Making Progress in SSSCs—Toward More Efficient … 213

The light sensitivity range of SSSCs based on the ternary NPs can be furtherextended by using ZnS-protectedCu-In-SeNPswithEg = *1.3 eV [175]. The SSSCssensitized with Ag-In-Se NPs also can harvest light down to *860 nm (1.44 eV,Fig. 4.32a) [54]. By doping Ag-In-Se NPs with Zn a series of (AgIn)xZn2(1−x)Se2 solidsolution NPs can be produced with CB and VB levels shifting continuously to lowervalues as x is increased (Fig. 4.32b). The offset between the CB levels of sensitizer NPsand titania increases with a decrease in x, but simultaneously Eg expands as well,resulting in a loss of the solar light harvesting capability (Fig. 4.32c, curves 1, 2). As aresult, the light conversion efficiency of Zn-doped AISe NP-based SSSCs shows a

Fig. 4.31 a PL QY (blue bars) of size-selected CIS NPs and light conversion efficiency of TiO2/CIS heterostructures based on corresponding NPs (red bars). Adapted (a) and reprinted (b) withpermissions from Ref. [51]. Copyright (2015) American Chemical Society

Fig. 4.32 a, b IPCE spectra (AgIn)xZn2(1−x)Se NP-based photoanodes and energy diagram ofcorresponsding sensitizer NPs; c band gap (curve 1) and ECB (curve 2, versus vacuum level) of(AgIn)xZn2(1−x)Se (ZAIS) NPs as a function of x. Yellow bars indicate the light conversionefficiency η (given in black figures) of corresponding TiO2/ZAISe/CdS photoanodes as well asTiO2/CdS heterostructure. Reprinted (a, b) and adapted (c) with permissions from Ref. [54].Copyright (2014) American Chemical Society

214 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

volcano-shaped dependence on the Zn content with a maximal value of 1.9% observedat x = 0.5 [54] (Fig. 4.32c, bars). By similar reasons, the SSSCswithmixedCuInSxSe1−x NPs show the highest light conversion efficiency at an intermediary chalcogenidecomposition corresponding to x around 0.4 [199].

A progress in the ternary CIS/AIS NPs applications for the SSSCs stimulated asearch for other ternary and more complex formulations that revealed new andpromising metal chalcogenide sensitizers, in particular Cu2SnS3 [200]. Such NPscan easily be electrodeposited on the metal oxide substrates from aqueous elec-trolytes containing Cu and Sn citrate complexes and Na2S2O3 and provide efficientlight harvesting in the range of hv > 2 eV [200].

The ternary Bi- and Sb-based sensitizers crystallize in a variety of compositionrevealing a plethora of band gaps favorable for the solar light harvesting *1 eVfor Cu3SbS4 [201], *1.3 eV for AgBiS2 [99], *1.7 eV for AgSbS2 [101, 157]and Cu12Sb4S13 [201], 2.1 eV for CuBiS2 [128].

The quaternary kesterite Cu2ZnSnS(Se)4 materials that find broad applications inphotovoltaics in the form of microcrystalline thin films [202–205] can be producedas NPs [206] that have a great potential for application as the SSSC sensitizers [72,156].

Design of new architectures of the metal oxide scaffolds for more efficientSSSCs. The principles and approaches of the design of efficient light-scattering andcharge-collecting metal oxide scaffolds for the SSSCs are generally the same asthose devised for the dye-sensitized solar cells [16, 19, 20, 26, 207, 208]. Since thesensitizer NPs usually have higher extinction coefficients than typical dye sensi-tizers the high specific surface area of the scaffold is not so critical for the SSSCs asit is for the DSSCs, however, a high porosity of the metal oxide layer is stillwelcomed for a better contact between the sensitizer NPs and the electrolyte. At thesame time, a high contact area enables also a higher recombination rate and so acertain optimal scaffold porosity is always required to achieve the highest SSSCperformance.

The oxide scaffold accepts the photogenerated electrons from the sensitizer NPsand transfers them further into the circuit and the efficiency of this process dependsdirectly on the lattice perfection and the defect density in the oxide layer. As theelectron travels trough the mesoporous metal oxide scaffold it “visits” *102 oxideNPs and in each NP there exists a probability of the recombination with adsorbedelectrolyte species [15, 16, 19, 26, 29, 208, 209]. In this view, single-crystallineoxide NRs or NWs provide a better conductance of electrons than the polycrys-talline mesoporous frameworks abundant with the interparticle interfaces and defectstates.

The electron mobility is estimated to be hundreds of times higher for ZnO NWsas compared with the conventional mesoporous ZnO films [15, 19, 20, 26, 209]. Asa result, the SSSC designs based on ZnO NWs [19, 20, 43, 55, 94, 95, 124, 154,190, 208, 209], and NRs [79, 119, 120, 125, 130, 155, 156, 196, 210, 211](Fig. 4.33a) were successfully realized. The ZnO NRs and NR arrays can be quiteconveniently formed by the electrodeposition, the NR length, and diameter con-trolled by the electrolysis duration [120]. Alternatively, the NW- and NR-based

4.5 Making Progress in SSSCs—Toward More Efficient … 215

ZnO scaffolds can be produced by a hydrothermal treatment using hexam-ethylenetetramine as a structure-directing agent [55, 125, 130, 154–156, 196, 210](see synthesis details in Chap. 5). Similarly, titania NRs produced by the HTT wereprobed as a transport layer in SSSCs (Fig. 4.33b) [123, 144, 209].

Even in the case of mesoporous materials, a considerable breakthrough wasachieved for the ordered scaffolds, such, for example, as mesoporous titania NTarrays. The lateral movement of charge carriers is strongly confined in such NTsmaking the carriers to move along the main NT axis toward OTE. Also, the NTsreveal a strong light scattering, thus increasing the light absorption probability inthe sensitizer NPs. Both titania and ZnO polycrystalline NTs are typically producedby the electrochemical etching. The ZnO NTs can be synthesized by the elec-troetching of previously electrodeposited ZnO NRs (Fig. 4.33c) [121, 185, 212],while titania NT arrays (Fig. 4.33d) are typically formed by the etching of titaniafoils in various fluorine-based electrolytes [92, 106, 213]. The etching procedurescan also be applied to produce porous ZnO NTs [62, 158, 214].

Fig. 4.33 SEM images of arrays of ZnO NWs (a), TiO2 NWs (b), ZnO NTs (c), and TiO2 NTs(d). Reprinted with permissions from Refs. [119] (a), [123] (b), [212] (c), and [92] (d). Copyright(2011–2013) American Chemical Society (a, c) and The Royal Society of Chemistry (b, d)

216 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

In the context of the idea of a strong light scattering by the metal oxide scaffold,inverse opal oxide structures were introduced both in the DSSCs and SSSCs,allowing to “capture” the incoming light and subject it to the multiple scatteringwithin the absorbing NP layer [207]. Typically, titania photonic crystals are pro-duced by self-assembling of the polystyrene microspheres into opals, soaking theopalescent assembly with titanium(IV) precursors and calcination producing thehollow-sphere inverse replica of the original opal [81]. Rutile hollow spheres can beproduced by the laser-induced melting of titania NPs and applied as a light scat-tering layer in SSSCs [129].

4.6 Nanocrystalline Semiconductor Counter-Electrodesfor SSSCs

A counter electrode is an important functional constituent of the SSSCs as it col-lects electrons photogenerated in the visible-light-sensitive photoanode and cat-alyzes the reduction of a component of the redox couple that was oxidized by thephotogenerated conduction band holes on the photoanode [30, 31]. As theliquid-junction SSSCs operate most often with aqueous (aqueous/methanolic)solutions of sodium sulfide/polysulfide, that is with a S2−/S0 redox couple, the CEmust fulfill several basic requirements. It should be catalytically active with respectto the transformations of the redox couple (in this case, to catalyze reduction ofsulfur to S2−), it must provide sufficiently high surface area to avoid any diffusionresistances on the electrolyte/CE interface and, finally, the CE should be stable inthe polysulfide electrolyte.

At the rise of the studies on the SSSCs it was realized that Pt CE, traditional andthe “best” one for the dye-sensitized liquid-junction solar cells exhibits a lowefficiency and a low stability in the polysulfide electrolyte due poisoning of the Ptsurface with sulfur species. At the same time, some metal sulfides were recognizedas very promising materials for the CE of SSSCs with the polysulfide electrolytes[30, 31], such as CuxS [33, 215–225], CoSx [33, 215, 218, 226–228], NiS [215,229], PbS [35, 219, 229, 230], etc. Also, the nanostructured films of Cu3Se2 grownon FTO by CBD were found to reveal superior catalytic properties in SSSCs ascompared to copper sulfide-based CEs [231]. Such materials revealed excellentcatalytic properties with respect to the S2−/S0 redox couple as well as a long-termstability.

Recently, good perspectives for various porous carbon materials as CEs for thepolysulfide-based SSSCs as well as for metal sulfide composites with the carbonmaterials were also recognized. For a deeper analysis of the current state-of-the-artin this area, the reader is referred to recent excellent reviews on the classificationand special features of metal-sulfide, carbonaceous and composite CE materials forSSSCs [31, 232]. Table 4.7 summarizes some of the reported SSSCs with metalsulfide and carbon CE and provides an overview of the achieved efficiencies of thelight power conversion in such systems. A large part of these studies was performed

4.5 Making Progress in SSSCs—Toward More Efficient … 217

with similar TiO2/CdS/CdSe photoanodes enabling a comparison of various typesof the CE materials. As the first several rows of Table 4.7 show, the Pt CEs appearto be much inferior to the copper and cobalt sulfide films produced on the FTOplates by the SILAR deposition, the sulfidation of metal copper, or theelectrodeposition.

Similarly to the photoanodes, the metal-sulfide-based CEs can easily be pro-duced by the SILAR from aqueous solutions of metal salts and sodium sulfide. Thismethod can be generally applied to form porous copper, cobalt, nickel and leadsulfide films on the nanocrystalline mesoporous ITO [215]. In the cells with a TiO2/CdS/CdSe photoanode, the CuxS and CoS CEs demonstrate a superior activity ascompared to NiS and PbS as well as to the conventional Pt (Fig. 4.34a). Thecatalytic activity of the best CuxS CEs was found to depend strongly on the SILARcycle number N most probably due to the incomplete ITO coverage at small Ns.

The activity of various CE materials is closely related to the dynamics of thecharge transfer on the CE/electrolyte interface. As a measure of the charge transferefficiency the charge transfer resistance RCT can be adopted, which is typicallydetermined by studying the CE with the electrochemical impedance spectroscopy(EIS). A higher RCT indicates a lower charge transfer rate and, correspondingly, thelower catalytic activity in the sulfur reduction reflected in a lower light conversionefficiency of the entire SSSCs. For example, for a series of Pt, CoS and CuSelectrodes RCT measured by EIS is 560, 48, and 6 O/cm2, respectively [218]varying in line with the light conversion efficiency of the corresponding cells with aTiO2/CdxZn1−xS/CdSe/ZnS photoanode (Table 4.7).

The relative activity of various CE materials can be evaluated from the “dark”current-voltage curves, that is the J–V dependences registered with no illuminationapplied in a three-electrode scheme with a metal sulfide film as a working electrode,some auxiliary electrode (Pt) and a reference electrode. The higher is the catalyticactivity of the CE material toward S2−/S0 couple, the steeper is the J–V dependence,that is an increment/decrement of J with varied V.

Figure 4.34b shows such J–V curves for a series of CuxS CE produced by theSILAR on the non-porous ITO glass and mesoporous ITO films with a differentnumber of deposition cycles. As can be seen, the catalytic effect is very small fornonporous substrates as compared to the porous ones and increases with anincreased amount of the CuxS catalyst. For the maximal cycle number N = 12 theJ–V dependence is closer to Y axis as compared with the curve for Pt complyingwith the higher efficiency of the porous CuxS film as a CE.

The surface area of CuxS-based materials can be increased by coupling coppersulfide particles to conductive/semiconductive substrates with a high surface area.For example, the composites of CuxS NPs with RGO exhibited quite high activityas CE with TiO2/CdS/CdSe photoanodes [217].

Similarly to the photoanodes, the metal-sulfide CEs with a high specific surfacearea can be produced by using other nanostructured semiconductors as a platform. Inparticular, ZnO/CuxS heterostructures can be easily prepared by the SILAR depo-sition of copper sulfide onto ZnO NRs (Fig. 4.35a) [220]. The CuxS forms a uniformlayer on the NR surface gaining from the high surface area of the ZnO substrate.

218 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Tab

le4.7

Someexam

ples

ofSS

SCsprod

uced

bywith

variou

scoun

terelectrod

es

Photoano

deCou

nter

electrod

e(CE)

CEform

ationmetho

dJ sc,mA

�cm

−2

Voc,mV

FF,%

η,%

Reference

TiO

2/CdS

/CdS

ePt

Cu xS

CoS

SILAR

SILAR

6.70

9.38

9.52

370

420

370

22 37 47

0.56

1.47

1.41

[215]

TiO

2/CdS

/CdS

eCu xS/Cumesh

sulfidatio

n11

.54

478

593.27

[216]

TiO

2/CdS

/CdS

ePt

Cu xS/RGO

sulfidatio

n11

.318

.446

052

031 46

1.60

4.40

[217]

TiO

2/CdS

-ZnS

/CdS

e/ZnS

Pt

CoS

CuS

e/d

e/d

9.1

11.2

13.9

470

520

550

35 32 35

1.6

1.9

2.7

[218]

TiO

2/CuInS

2/CdS

/ZnS

Pt

CuS

PbS

doctor

blade/precipitatio

n14

.217

.018

.3

430

550

580

37 42 45

2.3

4.0

4.7

[219]

TiO

2/CdS

mesop

orou

scarbon

matrixcarbon

ization

4.31

540

46.7

1.08

[244]

TiO

2/CdS

/CdS

eCoS

2sulfidatio

n14

.44

510

56.5

4.16

[227]

TiO

2/CdS

/CdS

eCuS

/CoS

CBD

17.11

n/r

55.4

4.1

[33]

TiO

2/CdS

/CdS

eCoS

CBD

14.95

454

50.5

3.4

[226]

TiO

2/CdS

/CdS

ecarbon

foam

carbon

ization

6.85

510

501.75

[244]

TiO

2/CdS

/CdS

ePbS

sulfidatio

n9.28

554

593.01

[230]

ZnO

/CdS

e/CdS

carbon

foam

carbon

ization

12.6

685

423.60

[245]

TiO

2/CdS

/CdS

eZnO

/PbS

CBD

13.28

633

56.6

4.76

[229]

ZnO

/CdS

/CdS

eZnO

/CuS

SILAR

14.48

740

354.18

[220]

TiO

2/CdS

/CdS

eCu 3Se

2CBD

13.10

567

63.4

4.71

[231]

TiO

2/CdS

e/ZnS

Nifoam

/Cu xS

sulfidatio

n9.95

581

61.4

3.55

[246]

TiO

2/CdS

e/CdS

Ca-do

pedCu xS/RGO

SILAR/electroph

o-retic

depo

sitio

n16

.26

520

332.73

[34]

TiO

2/CdS

e/CdS

PbS

CBD

12.17

644

594.61

[35]

(con

tinued)

4.6 Nanocrystalline Semiconductor Counter-Electrodes for SSSCs 219

Tab

le4.7

(con

tinued)

Photoano

deCou

nter

electrod

e(CE)

CEform

ationmetho

dJ sc,mA

�cm

−2

Voc,mV

FF,%

η,%

Reference

TiO

2/CdS

e/CdS

Cu 1

.8Splatelets

CBD

19.1

*60

045

5.16

[221]

TiO

2/CdS

WO

3−x

carbon

/WO

3−x

electrod

eposition

7.9

8.86

1004

951

44.5

51.5

3.66

4.60

[247]

TiO

2/CdS

e xTe 1

−x

Cu xS

carbon

/Ti

electrod

eposition

carbon

ization

20.61

20.67

698

803

61.2

68.6

8.79

11.39

[234]

TiO

2/CdS

e xTe 1

−x

carbon

/Cu xS

sulfidatio

n21

.27

655

608.40

[222]

TiO

2/CdS

/CdS

eCuInS

2

carbon

/CuInS

2

exsitu/doctorblade

13.43

14.16

518

512

52 603.63

4.32

[238]

ZnO

/ZnS

e/CdS

eCu 1

.8S

Cu 2Sn

S 3solvotherm

alsynthesis

10.51

11.46

822

810

42.3

43.7

3.65

4.06

[223]

TiO

2/CdS

eCu 2ZnS

nS4

Cu 2ZnS

nSe 4

spraydepo

sitio

n/selenizatio

n10

.53

15.49

520

540

40 522.19

4.35

[240]

TiO

2/CdS

/CdS

eCu 2ZnS

n(S,Se) 4

exsitu/dropcasting

12.71

550

433.01

[237]

NoteAM1.5ifno

tstated

otherw

ise;

ZnS

layertypically

(see

refs.)PS

redo

x-coup

leifno

tstated

otherw

ise

220 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Additionally, the CuxS layer reveals its own typical sheet-like morphology of sep-arate particles forming the layer. The activity of ZnO/CuxS heterostructure as a CEfor a ZnO/CdS/CdSe photoanode-based SSSC depends on the SILAR cycle number[220]. The light conversion efficiency increases with N increasing from 2 to 6 andsupersedes the activity of individual CuxS (Fig. 4.35b). Then η decreases as N iselevated from 6 to 8 following closely the variation of the charge transfer resistanceRCT. The similarity of both trends shows that at N > 6 the copper sulfide layer on the

Fig. 4.34 a Light conversion efficiency in the SSSCs with TiO2/CdS/CdSe photoanodes anddifferent CE; b Current−voltage curves of ITO glass-supported CuS(x), ITO porous film-supportedCuS(x), and Pt CEs (x—SILAR cycle number). Reprinted with permissions from Ref. [215].Copyright (2013) American Chemical Society

Fig. 4.35 a SEM of ZnO (a) and ZnO/CuS−6 NRs (b) cover produced by the SILAR; b Lightconversion efficiency in SSSCs with ZnO/CdS/CdSe photoanodes and CuS-based CEs, and chargetransfer resistance for CuS and ZnO/CuS counter electrodes (c). Reprinted with permissions fromRef. [220]. Copyright (2016) The Royal Society of Chemistry

4.6 Nanocrystalline Semiconductor Counter-Electrodes for SSSCs 221

ZnO surface becomes too thick for the efficient electron transfer to sulfur species inthe electrolyte, while at a smaller N the coverage of CuxS is insufficient for the fasttransfer of all electrons incoming to the counter electrode, while the ZnO alone has alow catalytic activity in the conversion of S0 to S2−.

Alternatively to the SILAR, the metal sulfide catalysts—CoS, NiS, CuS, PbS,can be deposited to the surface of ZnO NRs by CBD [229]. Also in this case theZnO NRs provide a high-surface-area framework for the metal sulfide loading thatis easily accessible by the electrolyte. Due to a high electron mobility in zinc oxideand single-crystalline character of ZnO NRs, the NRs offer an efficient electronpathway from the circuit to the metal sulfide catalyst layer. Unlikely the previousstudies on the SILAR-deposited sulfides, the highest light conversion efficiency(4.76%) was observed for a ZnO/PbS CE-based SSSC.

The relative activity of the ZnO/metal sulfide heterostructures as CEs can bevividly anticipated from a comparison of their “dark” J–V characteristics(Fig. 4.36a). By the angle between the Y axis and corresponding J–V curves theheterostructures form the following row: ZnO/PbS > ZnO/CuS > ZnO/NiS > ZnO/CoS > Pt. Exactly the same decreasing sequence is observed for the basic PECactivity parameters of the solar cells with a TiO2/CdS/CdSe photoanode and theabove-discussed CEs (Fig. 4.36b). The ZnO/PbS electrodes demonstrated excellentstability after multiple (more than 50) cyclic J–V measurements [229] indicating onthe robustness of such CE architecture that is necessary for the applications in theSSSCs.

Similarly to the preparation of photoanodes, the photocatalytic deposition of anactive metal-sulfide phase can be a good alternative to both SILAR and CBDpreparations of CEs. As mentioned before, a copper sulfide film can be photocat-alytically deposited onto mesoporous titania immersed into ethanol solution ofcopper perchlorate and S8 and illuminated with the UV light [138]. The CuxSdeposit exerts a light-shielding effect and slows the photoprocess after the

Fig. 4.36 a Cyclic voltammetry (CV) measurements of electrodes formed from ZnO NRs withmetal sulfides and Pt in a polysulfide solution with a scan rate of 100 mV � s−1; b Lightconversion efficiency, open-circuit voltage and fill factor for SSSCs with different ZnO-based CEs.Reprinted with permissions from Ref. [229]. Copyright (2016) The Royal Society of Chemistry

222 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

deposition of first portions of copper sulfide because it absorbs the incident light butcannot participate in the photo–chemical reactions. To avoid the light shielding andto produce thicker and more robust copper sulfide films it was proposed to performthe photodeposition in two stages, with the photocatalytic formation of copper NPson the surface of TiO2 or ZnO films followed by their transformation into CuxS in areaction with sodium sulfide (or polysulfide) that can take place in situ in the Na2Sxelectrolyte [233]. The photodeposited Cu0 NPs can act as a co-catalyst that col-lected the photogenerated electrons from the metal oxide photocatalyst and accel-erate the two-electron reduction of Cu2+ [239] resulting in a much higher coppercontent as compared to the direct photodeposition of CuxS.

Figure 4.37a shows SEM images of a ZnO/Cu0 composite produced by thephotocatalytic deposition of copper on the surface of ZnO films. Copper isdeposited in the form of NPs with a broad size distribution—from tens to hundreds

Fig. 4.37 SEM images of ZnO/Cu0 (a, b) and ZnO/CuxS (c, d) films produced by the copperphotodeposition [233]

4.6 Nanocrystalline Semiconductor Counter-Electrodes for SSSCs 223

nm attached to ZnO platelets and even growing through the platelets in some places(Fig. 4.37b), Such morphology indicates that primary smaller Cu NPs participate asco-catalysts in the photodeposition resulting in Cu0 deposition exclusively on theprimary metal NPs with no additional nuclei forming in the system.

When the ZnO/Cu0 films are immersed into aqueous sodium polysulfide solu-tions the spontaneous sulfidation takes place and the faceted Cu NPs transform intospherical aggregates of the nanometer-thin copper sulfide platelets (Fig. 4.37c, d).According to EDX, the atomic Cu/S ratio in the film was 1.3–1.4 indicating on anon-stoichiometric character of the copper sulfide coating as well as on a possiblepartial sulfidation of ZnO microplatelets.

The platelet-like morphology is quite typical for the products of the sulfidationof both Cu0 [216] and Cu2O [235]. The secondary aggregation of separate coppersulfide nanoplatelets is driven most probably by a spheroidal shape of startingphotodeposited Cu0 NPs, because in a similar system produced by a ions substi-tution of Zn(II) in ZnO nanoplatelets with Cu(II) the shape of secondary CuxSnanoplatelet aggregates mimicked the shape of starting ZnO microplatetes [236].

The ZnO/CuxS nanostructured films produced by the above-discussed pho-toassisted deposition were used as CEs with a ZnO/CdS photoanode showingaround 25% higher efficiency of the solar light harvesting as compared to similarsystems where the ZnO/CuxS counter electrode was produced by the ion exchange[236].

Typically, the metal-sulfide CEs are non-transparent and, therefore, the cellshould be illuminated through the semi-transparent photoanodes thus requiring thephotoanode to be formed on a transparent conductive substrate. To expand therange of possible conductive electrodes and use, for example, metal-based elec-trodes, such, for example, as anodized Ti foils with titania NT arrays, transparent orsemitransparent CE are required. The solution to this problem can be found inutilizing metal meshes with micro-/nano-layers of catalytically active materialsformed on their surface. For example, by contacting a copper mesh with polysulfidesolution a Cu/CuxS heterostructrure can be produced (Fig. 4.38a–d) [216]. Such Cumesh is light-transparent and conductive and supports a *2–3 lm-thick CuxSlayer that is composed of separate thin copper sulfide sheets thus providing a largecontact area between the CuxS layer and the polysulfide electrolyte.

Lead sulfide was found to reveal “dark” and photochemical catalytic propertiesin the sulfur reduction to S2− [219]. As PbS combines a p-type photoresponse withbroad absorption bands extending to the NIR range it can be used as a photocathodein a SSSCs with a TiO2/CuInS2/CdS photoanode and the polysulfide electrolyte.Under the illumination, such cell outperforms similar cells based on Pt and CuxSCE even though copper sulfide has superior electrocatalytic activity toward the S2−/S0 redox couple. The PbS-based SSSC showed a light conversion efficiency of4.7% that is *15% higher than the CuxS-based analog [219].

The higher efficiency of the PbS-based cell originates from two factors. The firstis in additional photocurrent produced by the photoexcitation of PbS cathode due tothe photostimulated reduction of sulfur in polysulfide anions (Fig. 4.38e). Thesecond factor is an increase in the total Voc of the cell due to the contribution of a

224 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

photovoltage produced by the photoanode (Voc 1 on Fig. 4.38e) to the photovoltagegenerated by the PbS photocathode (Voc 2 on Fig. 4.38e). Naturally, only a smallportion of the light incoming the cell penetrates the photoanode and reaches thephotocathode. However, the combination of “dark” catalytic activity of PbS withthe photoinduced catalytic effect from this residual light allows outperforming thephotochemically inert copper sulfide.

High power conversion efficiencies in the SSSCs with CuxS, CoS and othersulfides stimulated further searches for alternative metal chalcogenide materials. Forexample, cobalt disulfide pyrite CoS2 which is quite abundant in the Earth crust wastested as a CE with the polysulfide electrolyte and a TiO2/CdS/CdSe photoanodedemonstrating a reasonably high energy conversion efficiency of 4.16% [227].

More complex ternary chalcopyrite and quaternary kesterite metal chalcogenideswere also tested as CE materials, simultaneously with probing of their potential as alight-harvesting component of the SSSCs. In particular, both individual and mixedquaternary chalcogenides with a general formula Cu2ZnSn(S1−xSex)4 were found tobe catalytically active towards the S2−/S0 redox-couple, their activity dependingstrongly on the CE composition and morphology [237]. The light conversionefficiency in SSSCs based on the kesterite CE and a TiO2/CdS/CdSe photoanodereaches a peak value of 3.01% at x = 0.5 and then drops as the selenium content isincreased (Fig. 4.39a).

All the tested kesterites were more active than Pt (1.24%). The charge transferresistance RCT shows a similar dependence on the CE composition passing througha minimum for x = 9.5 (Fig. 4.39a), however, it is relatively small both forCu2ZnSnS4 and Cu2ZnSn(S0.5Se0.5)4, while the corresponding SSSCs differ in ηalmost by 100%. Such a difference in the CE activity at a relatively low RCT wasattributed to a much more developed surface area of the sulfoselenide kesterite(Fig. 4.39b) as compared to the individual Cu2ZnSnS4 (panel c). The non-uniformmorphology of Cu2ZnSn(S0.5Se0.5)4 arises most probably from the inhomogeneous

Fig. 4.38 a–d SEM images of semitransparent CuxS/Cu mesh CE; e Schematic energy banddiagram and charge transfer processes in the SSSC with PbS-based photo-active CE. Reprintedwith permissions from Refs. [216] (a–d) and [219] (e). Copyright (2013) Elsevier (a–d) and TheRoyal Society of Chemistry (e)

4.6 Nanocrystalline Semiconductor Counter-Electrodes for SSSCs 225

nucleation due to a difference in the Se and S ionic radii favoring to a highercatalytic activity of the mixed sulfoselenide kesterite.

An abundant source of various CE materials was found by combiningmetal-sulfide NPs with diverse carbonaceous materials [247], such as soot, meso-porous carbon foams with a developed surface area [222, 234, 238, 248], carbonNTs [31, 247], RGO [31, 34, 228]. Individual carbon nanomaterials can also beapplied as cathode materials in the SSSCs with visible-light-sensitive photoanodesand polysulfide or iodine/iodide redox-shuttles, in particular, mesoporous carbons[234, 249], mesocellular carbon foams [245].

Partial sulfidation of flexible Cu/Ni films was shown to yield stretchable counterelectrodes for the CdSe-sensitized SSSCs [246]. Such CEs can be combined withflexible photoanodes based on the titania-modified plastics showing a light con-version efficiency of 3.55% as well as a good chemical and mechanical robustness[246].

Concluding the discussion of various aspects of the semiconductorNP-sensitized photoelectrochemical solar cells we should note that this researcharea seems to be in its very blossom stage, especially if compared with thedye-sensitized solar cells, where a certain saturation is currently observed. Theprogress in SSSCs occurs simultaneously in many directions, it includes (i) constantemergence of new nano-materials for the light-harvesting photoanodes and catalyticcounter electrodes, especially among the Earth-abundant and low-toxic semicon-ductors and carbonaceous materials; (ii) steady development of design conceptionsto orchestrate the photoinduced electron transfers in composite photoelectrodessuch as the cascade design, the bandgap and CB/VB level design, the scaffold

Fig. 4.39 a Light conversion efficiency and charge transfer resistance RCT for TiO2/CdS/CdSe/ZnS photoanodes coupled with Cu2ZnSn(S,Se)4 CEs with a different S to Se ratio;b, c Morphology of Cu2ZnSn(S1−xSex)4 films with x = 0.8 (b) and x = 0 (c). Reprinted withpermissions from Ref. [237]. Copyright (2013) American Chemical Society

226 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

morphology design, etc.; (iii) deep investigations into the factors limiting the lightconversion efficiency, such as various recombination processes and interfacialbarriers. The total light conversion efficiencies, both in absolute values and in theincrements achieved in recent years as compared to the earlier studies, inspire astrong optimism and show a good future for this exciting area of the photochemicalsolar light conversion research.

References

1. Manassen J, Cahen D, Hodes G, Sofer A (1976) Electrochemical, solid state, photochemicaland technological aspects of photoelectrochemical energy converters. Nature 263:97–100.doi:10.1038/263097a0

2. Bhubaneswari P, Iniyanb S, Ranko G (2011) A review of solar photovoltaic technologies.Renew Sus En Rev 15:1625–1636. doi:10.1016/j.rser.2010.11.032

3. Kamat PV (2013) Energy outlook for planet earth. J Phys Chem Lett 4:1727–1729. doi:10.1021/jz400902s

4. Rühle S, Shalom M, Zaban A (2010) Quantum dot sensitized solar cells. ChemPhysChem11:2290–2304. doi:10.1002/cphc.201000069

5. Hodes G (2008) Comparison of dye- and semiconductor-sensitized porous nanocrystallineliquid junction solar cells. J Phys Chem C 112:17778–17787. doi:10.1021/jp803310s

6. Kamat PV (2008) Quantum dot solar cells. Semiconductor nanocrystals as light harvesters.J Phys Chem C 112:18737–18753. doi:10.1021/jp806791s

7. Kamat PV, Tvrdy K, Baker DR, Radich JG (2010) Beyond photovoltaics: semiconductornanoarchitectures for liquid-junction solar cells. Chem Rev 110:6664–6688. doi:10.1021/cr100243p

8. Kamat PV (2007) Meeting the clean energy demand: nanostructure architectures for solarenergy conversion. J Phys Chem C 111:2834–2860. doi:10.1021/jp066952u

9. Kamat PV (2013) Quantum dot solar cells. The next big thing in photovoltaics. J Phys ChemLett 4:908–918. doi:10.1021/jz400052e

10. Nozik AJ (2010) Nanoscience and nanostructures for photovoltaics and solar fuels. NanoLett 10:2735–2741. doi:10.1021/nl102122x

11. Grätzel M (2001) Photoelectrochemical cells. Nature 414:338–344. doi:10.1038/3510460712. Mlinar V (2013) Engineered nanomaterials for solar energy conversion. Nanotechnology

24:04201113. Fan X, Zhang M, Wang X et al (2013) Recent progress in organic–inorganic hybrid solar

cells. J Mater Chem A 1:8694–8709. doi:10.1039/C3TA11200D14. Hodes G, Cahen D (2012) All-solid-state, semiconductor-sensitized nanoporous solar cells.

Acc Chem Res 45:705–713. doi:10.1021/ar200219h15. Huang J, Yinb Z, Zheng Q (2011) Applications of ZnO in organic and hybrid solar cells.

Energy Environ Sci 4:3861–3877. doi:10.1039/C1EE01873F16. Anta JA, Guillen E, Tena-Zaera R (2012) ZnO-based dye-sensitized solar cells. J Phys

Chem C 116:11413–11425. doi:10.1021/jp301002517. Selinsky RS, Ding Q, Faber MS et al (2013) Quantum dot nanoscale heterostructures for

solar energy conversion. Chem Soc Rev 42:2963–2985. doi:10.1039/C2CS35374A18. Yang Z, Chen CY, Roy P, Chang HT (2011) Quantum dot-sensitized solar cells

incorporating nanomaterials. Chem Commun 47:9561–9571. doi:10.1039/C1CC11317H19. Zhang Q, Dandeneau CS, Zhou X, Cao G (2009) ZnO nanostructures for dye-sensitized

solar cells. Adv Mater 21:4087–4108. doi:10.1002/adma.200803827

4.6 Nanocrystalline Semiconductor Counter-Electrodes for SSSCs 227

20. Weintraub B, Zhou Z, Li Y, Deng Y (2010) Solution synthesis of one-dimensional ZnOnanomaterials and their applications. Nanoscale 2:1573–1587. doi:10.1039/C0NR00047G

21. Kojima A, Teshima K, Shirai Y, Miyasaka T (2009) Organometal halide perovskites asvisible-light sensitizers for photovoltaic cells. J Am Chem Soc, 131(17):6050–6051

22. Grätzel M (2014) The light and shade of perovskite solar cells. Nat Mater 13:838–842.doi:10.1038/nmat4065

23. Petrovic M, Chellappan V, Ramakrishna S (2015) Perovskites: solar cells & engineeringapplications—materials and device developments. Sol Energy 122:678–699. doi:10.1016/j.solener.2015.09.041

24. Chen J, Zhou S, Jin S et al (2016) Crystal organometal halide perovskites with promisingoptoelectronic applications. J Mater Chem C 4:11–27. doi:10.1039/c5tc03417e

25. Meng L, You J, Guo TF, Yang Y (2016) Recent advances in the inverted planar structure ofperovskite solar cells. Acc Chem Res 49:155–165. doi:10.1021/acs.accounts.5b00404

26. Zhao Y, Zhu K (2016) Organic–inorganic hybrid lead halide perovskites for optoelectronicand electronic applications. Chem Soc Rev 45:655–689. doi:10.1039/c4cs00458b

27. Heo JH, Im SH, Kim HJ et al (2012) Sb2S3-sensitized photoelectrochemical cells: opencircuit voltage enhancement through the introduction of poly-3-hexylthiophene interlayer.J Phys Chem C 116:20717–20721. doi:10.1021/jp305150s

28. Hodes G, Manassen J (1980) Electrocatalytic electrodes for the polysulfide redox system.J Electrochem Soc 127:544–549. doi:10.1149/1.2129709

29. Chen H, Zhu L, Liu H, Li W (2013) ITO porous film-supported metal sulfide counterelectrodes for high-performance quantum-dot-sensitized solar cells. J Phys Chem 117:3739–3746. doi:10.1021/jp309967w

30. Meng K, Chen G, Ravindranathan Thampi K (2015) Metal chalcogenides as counterelectrode materials in quantum dot sensitized solar cells: a perspective. J Mater Chem A3:23074–23089. doi:10.1039/c5ta05071e

31. Wang S, Tian J (2016) Recent advances in counter electrodes of quantum dot-sensitizedsolar cells. RSC Adv 6:90082–90099. doi:10.1039/c6ra19226b

32. Eskandari M, Ahmadi V, Ghahary R (2015) Copper sulfide/lead sulfide as a highly catalyticcounter electrode for zinc oxide nanorod based quantum dot solar cells. Electrochim Acta151:393–398. doi:10.1016/j.electacta.2014.11.037

33. Yang Z, Chen CY, Liu CW et al (2011) Quantum dot–sensitized solar cells featuring CuS/CoS electrodes provide 4.1% efficiency. Adv Energy Mater 1:259–264. doi:10.1002/aenm.201000029

34. Khalili SS, Dehghani H (2016) Ca-doped CuS/graphene sheet nanocomposite as a highlycatalytic counter electrode for improving quantum dot-sensitized solar cell performance.RSC Adv 6:10880–10886. doi:10.1039/c5ra24053k

35. Gopi C, Bae J, Venkata-Haritha M et al (2015) One-step synthesis of solution processed timedependent highly efficient and stable PbS counter electrodes for quantum dot-sensitized solarcells. RSC Adv 5:107522–107532. doi:10.1039/c5ra22715a

36. Nayak PK, Garcia-Belmonte G, Kahn A et al (2012) Photovoltaic efficiency limits andmaterial disorder. Energy Environ Sci 5:6022–6039. doi:10.1039/C2EE03178G

37. Salant A, Shalom M, Tachan Z et al (2012) Quantum rod-sensitized solar cell: nanocrystalshape effect on the photovoltaic properties. Nano Lett 12:2095–2100. doi:10.1021/nl300356e

38. Abrutis A, Plausinaitiene V, Skapas M et al (2008) Chemical vapor deposition ofchalcogenide materials for phase-change memories. Microelectron Eng 85:2338–2341.doi:10.1016/j.mee.2008.09.014

39. Perednis D, Gauckler LJ (2005) Thin film deposition using spray pyrolysis. J Electroceram14:103–111. doi:10.1007/s10832-005-0870-x

40. Park YM, Andre R, Kasprzak J et al (2007) Molecular beam epitaxy of CdSe epilayers andquantum wells on ZnTe substrate. Appl Surf Sci 253:6946–6950. doi:10.1016/j.apsusc.2007.02.012

228 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

41. Bang JH, Suslick KS (2010) Applications of ultrasound to the synthesis of nanostructuredmaterials. Adv Mater 22:1039–1059. doi:10.1002/adma.200904093

42. Tada H, Fujishimaa M, Kobayashi H (2011) Photodeposition of metal sulfide quantum dotson titanium(IV) dioxide and the applications to solar energy conversion. Chem Soc Rev, 40(7):4232–4243. doi:10.1039/C0CS00211A

43. Xu J, Yang X, Wang H et al (2011) Arrays of ZnO/ZnxCd1−xSe nanocables: band gapengineering and photovoltaic applications. Nano Lett 11:4138–4143. doi:10.1021/nl201934k

44. Kershaw SV, Susha AS, Rogach AL (2013) Narrow bandgap colloidal metal chalcogenidequantum dots: synthetic methods, heterostructures, assemblies, electronic and infraredoptical properties. Chem Soc Rev 42:3033–3087. doi:10.1039/c2cs35331h

45. Rivera-González N, Chauhan S, Watson DF (2016) Aminoalkanoic acids as alternatives tomercaptoalkanoic acids for the linker-assisted attachment of quantum dots to TiO2.Langmuir 32:9206–9215. doi:10.1021/acs.langmuir.6b02704

46. Calzada R, Thompson CM, Westmoreland DE et al (2016) Organic-to-aqueous phasetransfer of cadmium chalcogenide quantum dots using a sulfur-free ligand for enhancedphotoluminescence and oxidative stability. Chem Mater 28:6716–6723. doi:10.1021/acs.chemmater.6b03106

47. Salant A, Shalom M, Hod I et al (2010) Quantum dot sensitized solar cells with improvedefficiency prepared using electrophoretic deposition. ACS Nano 4:5962–5968. doi:10.1021/nn1018208

48. Martinez-Ferrero E, Mora Sero I, Albero J et al (2010) Charge transfer kinetics in CdSequantum dot sensitized solar cells. Phys Chem Chem Phys 12:2819–2821. doi:10.1039/b924970b

49. Pan Z, Zhao K, Wang J et al (2013) Near infrared absorption of CdSexTe1−x alloyedquantum dot sensitized solar cells with more than 6% efficiency and high stability. ACSNano 7:5215–5222. doi:10.1021/nn400947e

50. Penga Z, Liua Y, Wub L et al (2016) Influence of surface states of CuInS2 quantum dots inquantum dots-sensitized photo-electrodes. Appl Surf Sci 388:437–443. doi:10.1016/j.apsusc.2015.12.004

51. Jara DH, Yoon SJ, Stamplecoskie KG, Kamat PV (2014) Size-Dependent photovoltaicperformance of CuInS2 quantum dot-sensitized solar cells. Chem Mater 26:7221–7228.doi:10.1021/cm5040886

52. Santra PK, Nair PV, Thomas KG, Kamat PV (2013) CuInS2–sensitized quantum dot solarcell. Electrophoretic deposition, excited-state dynamics, and photovoltaic performance.J Phys Chem Lett 4:722–729. doi:10.1021/jz400181m

53. Chang CC, Chen JK, Cp Chen et al (2013) Synthesis of eco-friendly CuInS2 quantumdot-sensitized solar cells by a combined ex situ/in situ growth approach. ACS Appl MaterInterfaces 5:11296–11306. doi:10.1021/am403531q

54. Kameyama T, Douke Y, Shibakawa H et al (2014) Widely controllable electronic energystructure of ZnSe − AgInSe2 solid solution nanocrystals for quantum-dot-sensitized solarcells. J Phys Chem C 118:29517–29524. doi:10.1021/jp508769f

55. Leschkies KS, Divakar R, Basu J et al (2007) Photosensitization of ZnO nanowires withCdSe quantum dots for photovoltaic devices. Nano Lett 7:1793–1798. doi:10.1021/nl070430o

56. Zarazua I, De la Rosa E, Lopez-Luke T et al (2011) Photovoltaic conversion enhancement ofCdSe quantum dot-sensitized TiO2 decorated with Au nanoparticles and P3OT. J PhysChem C 115:23209–23220. doi:10.1021/jp207744n

57. Jin H, Choi S, Xing G et al (2015) SnS44−, SbS4

3−, and AsS33− metal chalcogenide surface

ligands: couplings to quantum dots, electron transfers, and all-inorganic multilayeredquantum dot sensitized solar cells. J Am Chem Soc 137:13827–13835. doi:10.1021/jacs.5b05787

References 229

58. Meng X, Du J, Zhang H, Zhong X (2015) Optimizing the deposition of CdSe colloidalquantum dots on TiO2 film electrode via capping ligand induced self-assembly approach.RSC Adv 5:86023–86030. doi:10.1039/c5ra17412k

59. Jung MH, Kang MG (2011) Enhanced photo-conversion efficiency of CdSe–ZnS core–shellquantum dots with Au nanoparticles on TiO2 electrodes. J Mater Chem 21:2694–2700.doi:10.1039/c0jm03431b

60. Kongkanand A, Tvrdy K, Takechi K et al (2008) Quantum dot solar cells. Tuningphotoresponse through size and shape control of CdSe-TiO2 architecture. J Am Chem Soc130:4007–4015. doi:10.1021/ja0782706

61. Pernik DR, Tvrdy K, Radich JG, Kamat PV (2011) Tracking the adsorption and electroninjection rates of CdSe quantum dots on TiO2: linked versus direct attachment. J PhysChem C 115:13511–13519. doi:10.1021/jp203055d

62. Chouhan N, Yeh CL, Hu SF et al (2011) Photocatalytic CdSe QDs-decorated ZnOnanotubes: an effective photoelectrode for splitting water. Chem Commun 47:3493–3495.doi:10.1039/c0cc05548d

63. Chen J, Lei W, Deng WQ (2011) Reduced charge recombination in a co-sensitized quantumdot solar cell with two different sizes of CdSe quantum dot. Nanoscale 3:674–677. doi:10.1039/c0nr00591f

64. King LA, Riley DJ (2012) Importance of QD purification procedure on surface absorbanceof QDs and performance of QD sensitized photoanodes. J Phys Chem C 116:3349–3355.doi:10.1021/jp210290j

65. Santra PK, Kamat PV (2013) Tandem-layered quantum dot solar cells: tuning thephotovoltaic response with luminescent ternary cadmium chalcogenides. J Am Chem Soc135:877–885. doi:10.1021/ja310737m

66. Itzhakov S, Shen H, Buhbut S et al (2013) Type-II quantum-dot-sensitized solar cellspanning the visible and near-infrared spectrum. J Phys Chem C 117:22203–22210. doi:10.1021/jp312190x

67. Hyun BR, Zhong YW, Bartnik AC et al (2008) Electron injection from colloidal PbSquantum dots into titanium dioxide nanoparticles. ACS Nano 2:2206–2212. doi:10.1021/nn800336b

68. Krüger S, Hickey SG, Tscharntke S, Eychmüller A (2011) Study of the attachment of linkermolecules and their effects on the charge carrier transfer at lead sulfide nanoparticlesensitized ZnO substrates. J Phys Chem C 115:13047–13055. doi:10.1021/jp200935x

69. Li TL, Lee YL, Teng H (2012) High-performance quantum dot-sensitized solar cells basedon sensitization with CuInS2 quantum dots/CdS heterostructure. Energy Environ Sci5:5315–5324. doi:10.1039/c1ee02253a

70. Cheng KC, Law WC, Yong KT et al (2011) Synthesis of near-infrared silver-indium-sulfide(AgInS2) quantum dots as heavy-metal free photosensitizer for solar cell applications. ChemPhys Lett 515:254–257. doi:10.1016/j.cplett.2011.09.027

71. Pan Z, Mora-Seró I, Shen Q et al (2014) High-efficiency “Green” quantum dot solar cells.J Am Chem Soc 136:9203–9210. doi:10.1021/ja504310w

72. Xu A, Tao H, Chen S et al (2015) A novel approach to utilize thiol reduced graphene oxideas linker molecule for Cu2ZnSnS4 sensitized solar cell. Inter J Hydrogen Energy 40:15933–15939. doi:10.1016/j.ijhydene.2015.09.068

73. Wang H, Luan C, Xu X et al (2012) In situ versus ex situ assembly of aqueous-basedthioacid capped CdSe nanocrystals within mesoporous TiO2 films for quantum dot sensitizedsolar cells. J Phys Chem C 116:484–489. doi:10.1021/jp209987q

74. King LA, Parkinson BA (2016) Photosensitization of ZnO crystals with iodide-capped PbSequantum dots. J Phys Chem Lett 7:2844–2848. doi:10.1021/acs.jpclett.6b01133

75. Raevskaya AE, Stroyuk OL, Panasiuk YV et al (2016) A new route to very stablewater-soluble ultra-small core/shell CdSe/CdS quantum dots. Nano-Struct Nano-Objects.doi:10.1016/j.nanoso.2015.12.001

230 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

76. Raevskaya AE, Alontseva VV, Kozitskiy AV et al (2016) Photoelectrochemical propertiesof titania nanoheterostructures with low-dimensional cadmium selenide particles. Ther ExpChem 52:152–162

77. Coughlin KM, Nevins JS, Watson DF (2013) Aqueous-phase linker-assisted attachment ofcysteinate(2−)-capped CdSe quantum dots to TiO2 for quantum dot-sensitized solar cells.ACS Appl Mater Interfaces 5:8649–8654. doi:10.1021/am402219e

78. Zhou C, Zhou L, Xu J, Gan Y (2016) Controllable synthesis of CdS quantum dots and theirphotovoltaic application on quantum-dot-sensitized ZnO nanorods. J Solid StateElectrochem 20:533–540. doi:10.1007/s10008-015-3075-5

79. Sarkar S, Makhal A, Lakshman K et al (2012) Dual-sensitization via electron and energyharvesting in CdTe quantum dots decorated ZnO nanorod-based dye-sensitized solar cells.J Phys Chem C 116:14248–14256. doi:10.1021/jp3046593

80. Lan GY, Yang Z, Lin YW et al (2009) A simple strategy for improving the energyconversion of multilayered CdTe quantum dot-sensitized solar cells. J Mater Chem 19:2349–2355. doi:10.1039/b817000b

81. Nehme AS, Haydous F, Halaoui L (2016) Amplification in light energy conversion atQ-CdTe sensitized TiO2 photonic crystal, photoelectrochemical stability in Se2− electrolyte,and size-dependent type II Q-CdTe/CdSe formation. J Phys Chem C 120:4766–4778.doi:10.1021/acs.jpcc.5b11478

82. Hu X, Zhang Q, Huang X et al (2011) Aqueous colloidal CuInS2 for quantum dot sensitizedsolar cells. J Mater Chem 21:15903–15905. doi:10.1039/c1jm12629f

83. Chang JY, Li CH, Chiang YH et al (2016) Toward the facile and ecofriendly fabrication ofquantum dot-sensitized solar cells via thiol coadsorbent assistance. ACS Appl MaterInterfaces 8:18878–18890. doi:10.1021/acsami.6b05411

84. Raevskaya AE, Rosovik OP, Kozytskiy AV et al (2016) Non-Stoichiometric Cu-In-S@ZnSnanoparticles produced in aqueous solutions as light harvesters for liquid-junctionphotoelectrochemical solar cells. RSC Adv 6:100145–100157. doi:10.1039/C6RA18313A

85. Yu XY, Lei BX, Kuang DB, Cy Su (2011) Highly efficient CdTe/CdS quantum dotsensitized solar cells fabricated by a one-step linker assisted chemical bath deposition. ChemSci 2:1396–1400. doi:10.1039/c1sc00144b

86. Song X, Wang M, Deng J et al (2013) One-step preparation and assembly of aqueouscolloidal CdSxSe1−x nanocrystals within mesoporous TiO2 films for quantum dot-sensitizedsolar cells. ACS Appl Mater Interfaces 5:5139–5148. doi:10.1021/am4009924

87. Nakamura R, Makuta S, Tachibana Y (2015) Electron injection dynamics at the SILARdeposited CdS quantum dot/TiO2 interface. J Phys Chem C 119:20357–20362. doi:10.1021/acs.jpcc.5b06900

88. Lee YS, Gopi CV, Venkata-Haritha M, Kim HJ (2016) Recombination control inhigh-performance quantum dot-sensitized solar cells with a novel TiO2/ZnS/CdS/ZnSheterostructure. Dalton Trans 45:12914–12923. doi:10.1039/c6dt02531e

89. Esparza D, Zarazúa I, López-Luke T et al (2015) Photovoltaic properties of Bi2S3 and CdSquantum dot sensitized TiO2 solar cells. Electrochim Acta 180:486–492. doi:10.1016/j.electacta.2015.08.102

90. Hossain MA, Jennings JR, Shen C et al (2012) CdSe-sensitized mesoscopic TiO2 solar cellsexhibiting >5% efficiency: redundancy of CdS buffer layer. J Mater Chem 22:16235–16242.doi:10.1039/c2jm33211f

91. Malashchonak MV, Mazanik AV, Korolik OV et al (2015) Influence of wide band gap oxidesubstrates on the photoelectrochemical properties and structural disorder of CdS nanopar-ticles grown by the successive ionic layer adsorption and reaction (SILAR) method.Beilstein J Nanotechnol 6:2252–2262. doi:10.3762/bjnano.6.231

92. Wang X, Zheng J, Sui X et al (2013) CdS quantum dots sensitized solar cells based onfree-standing and through-hole TiO2 nanotube arrays. Dalton Trans 42:14726–14732.doi:10.1039/C3DT51266E

References 231

93. You T, Jiang L, Han KL, Deng WQ (2013) Improving the performance of quantumdot-sensitized solar cells by using TiO2 nanosheets with exposed highly reactive facets.Nanotechnology 24:245401

94. Tak Y, Hong SJ, Leeb JS, Yong K (2009) Fabrication of ZnO/CdS core/shell nanowirearrays for efficient solar energy conversion. J Mater Chem 19:5945–5951. doi:10.1039/b904993b

95. Hwang I, Yong K (2013) Environmentally benign and efficient Ag2S-ZnO nanowires asphotoanodes for solar cells: comparison with CdS-ZnO nanowires. ChemPhysChem14:364–368. doi:10.1002/cphc.201200876

96. Emin S, Fanetti M, Abdi FF et al (2013) Photoelectrochemical properties of cadmiumchalcogenide-sensitized textured porous zinc oxide plate electrodes. ACS Appl MaterInterfaces 5:1113–1121. doi:10.1021/am3027986

97. Rabinovich E, Hodes G (2013) Effective bandgap lowering of CdS deposited by successiveionic layer adsorption and reaction. J Phys Chem C 117:1611–1620. doi:10.1021/jp3105453

98. Malashchonak M, Streltsov EA, Mazanik A et al (2015) Band-gap and sub-band-gapphotoelectrochemical processes at nanocrystalline CdS grown on ZnO by successive ioniclayer adsorption and reaction method. Thin Solid Films 589:145–152. doi:10.1016/j.tsf.2015.04.057

99. Huang PC, Yang WC, Lee MW (2013) AgBiS2 semiconductor-sensitized solar cells. J PhysChem C 117:18308–18314. doi:10.1021/jp4046337

100. Wan Y, Han M, Yu L et al (2015) Fabrication and photoelectrochemical properties of TiO2/CuInS2/Bi2S3 core/shell/shell nanorods electrodes. RSC Adv 5:78902–78909. doi:10.1039/c5ra14548a

101. Ho YR, Lee MW (2013) AgSbS2 semiconductor-sensitized solar cells. ElectrochemCommun 26:48–51. doi:10.1016/j.elecom.2012.10.003

102. Liu Z, Huang J, Han J et al (2016) CuSbS2: a promising semiconductor photo-absorbermaterial for quantum dot sensitized solar cells. Phys Chem Chem Phys 18:16615–16620.doi:10.1039/c6cp01688j

103. Braga A, Gimenez S, Concina I et al (2011) Panchromatic sensitized solar cells based onmetal sulfide quantum dots grown directly on nanostructured TiO2 electrodes. J Phys ChemLett 2:454–460. doi:10.1021/jz2000112

104. Kim J, Choi H, Nahm C et al (2013) Graded bandgap structure for PbS/CdS/ZnSquantum-dot-sensitized solar cells with a PbxCd1−xS interlayer. Appl Phys Lett 102:183901

105. Tubtimtae A, Wu KL, Tung HY et al (2010) Ag2S quantum dot-sensitized solar cells.Electrochem Commun 12:1158–1160. doi:10.1016/j.elecom.2010.06.006

106. Chen C, Xie Y, Ali G et al (2011) Improved conversion efficiency of Ag2S quantumdot-sensitized solar cells based on TiO2 nanotubes with a ZnO recombination barrier layer.Nanoscale Res Lett 6:462. doi:10.1186/1556-276X-6-462

107. Chang JY, Lin JM, Su LF, Chang CF (2013) Improved performance of CuInS2 quantumdot-sensitized solar cells based on a multilayered architecture. ACS Appl Mater Interfaces5:8740–8752. doi:10.1021/am402547e

108. Lee JW, Son DY, Ahn TK et al (2012) Quantum-dot-sensitized solar cell withunprecedentedly high photocurrent. Sci Rep 3:1050. doi:10.1038/srep01050

109. Kozitskiy AV, Stroyuk OL, Kuchmiy SY et al (2013) Morphology, optical, andphotoelectrochemical properties of electrodeposited nanocrystalline ZnO films sensitizedwith CdxZn1−xS nanoparticles. J Mater Sci 48:7764–7773. doi:10.1007/s10853-013-7598-9

110. Lee JH, Song WC, Yi JS et al (2003) Growth and properties of the Cd1−xZnxS thin films forsolar cell applications. Thin Solid Films 431–432:349–353. doi:10.1016/S0040-6090(03)00526-1

111. Antoniadou M, Kondarides DI, Dionysiou DD, Lianos P (2012) Quantum dot sensitizedtitania applicable as photoanode in photoactivated fuel cells. J Phys Chem C 116:16901–16909. doi:10.1021/jp305098m

112. Henglein A (1984) Catalysis of photochemical reactions by colloidal semiconductors. PureAppl Chem 56:1215–1224. doi:10.1351/pac198456091215

232 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

113. Zeug N, Bücheler J, Kisch H (1985) Catalytic formation of hydrogen and carbon-carbonbonds on illuminated zinc sulfide generated from zinc dithiolenes. J Am Chem Soc107:1459–1465. doi:10.1021/ja00292a001

114. Sadhu S, Patra A (2012) Lattice strain controls the carrier relaxation dynamics in CdxZn1−xSalloy quantum dots. J Phys Chem C 116:15167–15171. doi:10.1021/jp304901w

115. Sun XW, Huang JZ, Wang JX, Xu Z (2008) A ZnO nanorod inorganic/organicheterostructure light-emitting diode emitting at 342 nm. Nano Lett 8:1219–1223. doi:10.1021/nl080340z

116. Zewdu T, Clifford JN, Hernandez JP, Palomares E (2011) Photo-induced charge transferdynamics in efficient TiO2/CdS/CdSe sensitized solar cells. Energy Environ Sci 4:4633–4638. doi:10.1039/c1ee02088a

117. Guijarro N, Lana-Villarreal T, Lutz T et al (2012) Sensitization of TiO2 with PbSe quantumdots by SILAR: how mercaptophenol improves charge separation. J Phys Chem Lett3:3367–3372. doi:10.1021/jz301528a

118. Srathongluan P, Kuhamaneechot R, Sukthao P et al (2016) Photovoltaic performances ofCu2−xTe sensitizer based on undoped and In3+-doped TiO2 photoelectrodes and assembledcounter electrodes. J Colloid Interface Sci 463:222–228. doi:10.1016/j.jcis.2015.10.052

119. Bu Y, Chen Z, Li W, Yu J (2013) High-efficiency photoelectrochemical properties by ahighly crystalline CdS-sensitized ZnO nanorod array. ACS Appl Mater Interfaces 5:5097–5104. doi:10.1021/am400964c

120. Yao CZ, Wei BH, Meng LX et al (2012) Controllable electrochemical synthesis andphotovoltaic performance of ZnO/CdS core–shell nanorod arrays on fluorine-doped tinoxide. J Power Sources 207:222–228. doi:10.1016/j.jpowsour.2012.01.154

121. Qi X, She G, Liu Y et al (2012) Electrochemical synthesis of CdS/ZnO nanotube arrays withexcellent photoelectrochemical properties. Chem Commun 48:242–244. doi:10.1039/c1cc15674h

122. Yu XY, Liao JY, Qiu KQ et al (2011) Dynamic study of highly efficient CdS/CdSe quantumdot-sensitized solar cells fabricated by electrodeposition. ACS Nano 5:9494–9500. doi:10.1021/nn203375g

123. Ai G, Sun W, Gao X et al (2011) Hybrid CdSe/TiO2 nanowire photoelectrodes: fabricationand photoelectric performance. J Mater Chem 21:8749–8755. doi:10.1039/C0JM03867A

124. Wang X, Zhu H, Xu Y et al (2010) Aligned ZnO/CdTe coreshell nanocable arrays on indiumtin oxide: synthesis and photoelectrochemical properties. ACS Nano 4:3302–3308. doi:10.1021/nn1001547

125. Tang Y, Wang P, Yun JH et al (2015) Frequency-regulated pulsed electrodeposition ofCuInS2 on ZnO nanorod arrays as visible light photoanodes. J Mater Chem A 3:15876–15881. doi:10.1039/c5ta03255e

126. Zainun AR, Tomoya S, Noor UM et al (2012) New approach for generating Cu2O/TiO2

composite films for solar cell applications. Mater Lett 66:254–256. doi:10.1016/j.matlet.2011.08.032

127. Wei H, Gong H, Wang Y et al (2011) Three kinds of Cu2O/ZnO heterostructure solar cellsfabricated with electrochemical deposition and their structure-related photovoltaic proper-ties. CrystEngComm 13:6065–6070. doi:10.1039/c1ce05540b

128. Suriyawong N, Aragaw B, Shi JB, Lee MW (2016) Ternary CuBiS2 nanoparticles as asensitizer for quantum dot solar cells. J Colloid Interface Sci 473(2016):60–65. doi:10.1016/j.jcis.2016.03.062

129. Wang H, Miyauchi M, Ishikawa Y et al (2011) Single-crystalline rutile TiO2 hollow spheres:room-temperature synthesis, tailored visible-light-extinction, and effective scattering layerfor quantum dot-sensitized solar cells. J Am Chem Soc 133:19102–19109. doi:10.1021/ja2049463

130. Tian J, Zhang Q, Uchaker E et al (2013) Constructing ZnO nanorod array photoelectrodesfor highly efficient quantum dot sensitized solar cells. J Mater Chem A 1:6770–6775. doi:10.1039/c3ta11056g

References 233

131. Lin KH, Chuang CY, Lee YY et al (2012) Charge transfer in the heterointerfaces ofCdS/CdSe cosensitized TiO2 photoelectrode. J Phys Chem C 116:1550–1555. doi:10.1021/jp209353j

132. Shalom M, Buhbut S, Tirosh S, Zaban A (2012) Design rules for high-efficiencyquantum-dot-sensitized solar cells: a multilayer approach. J Phys Chem Lett 3:2436–2441.doi:10.1021/jz3010078

133. Ho W, Yu JC, Lin J et al (2004) Preparation and photocatalytic behavior of MoS2 and WS2nanocluster sensitized TiO2. Langmuir 20:5865–5869. doi:10.1021/la049838g

134. Kanda S, Akita T, Fujishima M, Tada H (2011) Facile synthesis and catalytic activity ofMoS2/TiO2 by a photodeposition-based technique and its oxidized derivative MoO3/TiO2

with a unique photochromism. J Colloid Interface Sci 354:607–610. doi:10.1016/j.jcis.2010.11.007

135. Shvalagin VV, Stroyuk OL, Kuchmiy SY (2007) Photocatalytic formation of porousCdS/ZnO nanospheres and CdS nanotubes. Theor Exp Chem 43:229–234. doi:10.1007/s11237-007-0026-y

136. Stroyuk OL, Shvalagin VV, Kotenko TE et al (2010) Photochemical reduction of sulfur inethanol in the presence of ZnO nanoparticles. Theor Exp Chem 46:218–224. doi:10.1007/s11237-010-9143-0

137. Ma B, Wang L, Dong H et al (2011) Photocatalysis of PbS quantum dots in a quantumdot-sensitized solar cell: photovoltaic performance and characteristics. Phys Chem ChemPhys 13:2656–2658. doi:10.1039/C0CP02415E

138. Zhukowskiy MA, Stroyuk OL, Shvalagin VV et al (2009) Photocatalytic growth of CdS,PbS, and CuxS nanoparticles on the nanocrystalline TiO2 films. J Photochem Photobiol A:Chem 203:137–144. doi:10.1016/j.jphotochem.2009.01.007

139. Jin-nouchi Y, Akita T, Tada H (2010) Ultrafast photodeposition of size-controlled PbSquantum dots on TiO2. ChemPhysChem 11:2349–2352. doi:10.1002/cphc.201000229

140. Fujii M, Nagasuna K, Fujishima M et al (2009) Photodeposition of CdS quantum dots onTiO2: preparation, characterization, and reaction mechanism. J Phys Chem C 113:16711–16716. doi:10.1021/jp9056626

141. Jin-nouchi Y, Naya SI, Tada H (2010) Quantum-dot-sensitized solar cell using a photoanodeprepared by in situ photodeposition of CdS on nanocrystalline TiO2 films. J Phys Chem C114:16837–16842. doi:10.1021/jp1062226

142. Fujishima M, Nakabayashi Y, Takayama K et al (2016) High coverage formation of CdSquantum dots on TiO2 by the photocatalytic growth of preformed seeds. J Phys Chem C120:17365–17371. doi:10.1021/acs.jpcc.6b04091

143. Nagasuna K, Akita T, Fujishima M, Tada H (2011) Photodeposition of Ag2S quantum dotsand application to photoelectrochemical cells for hydrogen production under simulatedsunlight. Langmuir 27:7294–7300. doi:10.1021/la200587s

144. Hu H, Ding J, Zhang S et al (2013) Photodeposition of Ag2S on TiO2 nanorod arrays forquantum dot-sensitized solar cells. Nanoscale Res Lett 8:10. doi:10.1186/1556-276X-8-10

145. Tada H, Mitsui T, Kiyonaga T et al (2006) All-solid-state Z-scheme in CdS-Au-TiO2

three-component nanojunction system. Nat Mater 5:782–786. doi:10.1038/nmat1734146. Stroyuk OL, Shvalagin VV, Raevskaya AE et al (2008) Photochemical formation of

semiconducting nanostructures. Theor Exp Chem 44:205–231. doi:10.1007/s11237-008-9037-6

147. Kozytskiy AV, Stroyuk OL, Kuchmiy SY et al (2013) Effect of the method of preparation ofZnO/CdS and TiO2/CdS film nanoheterostructures on their photoelectrochemical properties.Theor Exp Chem 49:165–171. doi:10.1007/s11237-013-9310-1

148. Kozitskiy AV, Stroyuk OL, Kuchmiy SY (2015) Photoelectrochemical properties of a solarcell based on FTO/ZnO/CdS (photoanode) and FTO/ZnO/CuxS (counter-electrode)heterostructures. Theor Exp Chem 51:203–209. doi:10.1007/s11237-015-9417-7

149. Stroyuk OL, Kuchmiy SY, Zhukovskii MA et al (2009) Effect of the method of productionof TiO2/CdS film nanoheterostructures on the effectiveness of photoinduced chargeseparation. Theor Exp Chem 45:302–307. doi:10.1007/s11237-009-9097-2

234 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

150. Kozytskiy AV, Stroyuk OL, Skoryk MA et al (2015) Photochemical formation andphotoelectrochemical properties of TiO2/Sb2S3 heterostructures. J Photochem Photobiol A:Chem 303:8–16. doi:10.1016/j.jphotochem.2015.02.005

151. Kozytskiy AV, Stroyuk OL, Kuchmiy SY (2013) Inorganic photoelectrochemical solar cellsbased on nanocrystalline ZnO/ZnSe and ZnO/CuSe heterostructures. Catal Today 230:227–233. doi:10.1016/j.cattod.2013.09.043

152. Zhou SX, Wang YG, Han W, Wang N (2008) I−V characteristics of metal−oxide−ZnSenanowire structure. J Phys Chem C 112:18644–18650. doi:10.1021/jp8068046

153. Yang L, Xie R, Liu L et al (2011) Synthesis and characterization of ZnSe nanocrystals byW/O reverse microemulsion method: the effect of cosurfactant. J Phys Chem C 115:19507–19512. doi:10.1021/jp204798y

154. Xu J, Yang X, Yang QD et al (2012) Arrays of CdSe sensitized ZnO/ZnSe nanocables forefficient solar cells with high open-circuit voltage. J Mater Chem 22:13374–13379. doi:10.1039/c2jm31970e

155. Lu Z, Xu J, Xie X et al (2012) CdS/CdSe double-sensitized ZnO nanocable arrayssynthesized by chemical solution method and their photovoltaic applications. J Phys Chem C116:2656–2661. doi:10.1021/jp208254z

156. Akram MA, Javed S, Islam M et al (2016) Arrays of CZTS sensitized ZnO/ZnS and ZnO/ZnSe core/shell nanorods for liquid junction nanowire solar cells. Sol Energy Mater SolCells 146:121–128. doi:10.1016/j.solmat.2015.11.034

157. Han J, Liu Z, Guo K et al (2015) AgSbS2 modified ZnO nanotube arrays forphotoelectrochemical water splitting. Appl Catal B 179:61–68. doi:10.1016/j.apcatb.2015.05.008

158. Han J, Liu Z, Guo K et al (2014) High-efficiency AgInS2-modified ZnO nanotube arrayphotoelectrodes for all-solid-state hybrid solar cells. ACS Appl Mater Interfaces 6:17119–17125. doi:10.1021/am5047813

159. Duan J, Zhang H, Tang Q et al (2015) Recent advances in critical materials for quantumdot-sensitized solar cells: a review. J Mater Chem A 3:17497–17510. doi:10.1039/C5TA03280F

160. Pietryga JM, Park YS, Lim J et al (2016) Spectroscopic and device aspects of nanocrystalquantum dots. Chem Rev 116:10513–10622. doi:10.1021/acs.chemrev.6b00169

161. Sharma D, Jha R, Kumar S (2016) Quantum dot sensitized solar cells: recent advances andfuture perspectives in photoanode. Sol Energy Mater Sol Cells 155:294–322. doi:10.1016/j.solmat.2016.05.062

162. Thomas SR, Chen CW, Date M et al (2016) Recent developments in the synthesis ofnanostructured chalcopyrite materials and their applications: a review. RSC Adv 6:60643–60656. doi:10.1039/C6RA05502H

163. Manser JS, Saidaminov MI, Christians JA et al (2016) Making and breaking of lead halideperovskites. Acc Chem Res 49:330–338. doi:10.1021/acs.accounts.5b00455

164. Wang D, Wright M, Elumalai NK, Uddin A (2016) Stability of perovskite solar cells. SolEnergy Mater Sol Cells 147:255–275. doi:10.1016/j.solmat.2015.12.025

165. Pedesseau L, Sapori D, Traore B et al (2016) Advances and promises of layered halidehybrid perovskite semiconductors. ACS Nano 10:9776–9786. doi:10.1021/acsnano.6b05944

166. Colella S, Mazzeo M, Rizzo A et al (2016) The bright side of perovskites. J Phys Chem Lett7:4322–4334. doi:10.1021/acs.jpclett.6b01799

167. Roose B, Ummadisingu A, Correa-Baena JP, Saliba M, Hagfeldt A, Grätzel M, Steiner U,Abate A (2017) Spontaneous crystal coalescence enables highly efficient perovskite solarcells. Nano Energy 39:24–29. doi:10.1016/j.nanoen.2017.06.037

168. Correa-Baena JP, Tress W, Domanski K, Anaraki EH, Turren-Cruz SH, Roose B, Boix PP,Grätzel M, Saliba M, Abate A (2017) Identifying and suppressing interfacial recombinationto achieve high open-circuit voltage in perovskite solar cells. Energy Environ Sci 10:1207–1212. doi:10.1039/C7EE00421D

References 235

169. Bi D, Yi C, Luo J, Decoppet JD, Zhang F, Zakeeruddin SM, Li X, Hagfeldt A, Grätzel M(2016) Polymer-templated nucleation and crystal growth of perovskite films for solar cellswith efficiency greater than 21%. Nat Energy 1:16142. doi:10.1038/nenergy.2016.142

170. Divitini G, Cacovich S, Matteocci F, Cina L, Di Carlo A, Ducati C (2016) In situobservation of heat-induced degradation of perovskite solar cells. Nat Energy 1:15012.doi:10.1038/nenergy.2015.12

171. Capasso A, Matteocci F, Najafi L, Prato M, Buha J, Cina L, Pellegrini V, Di Carlo A,Bonaccorso F (2016) Few-Layer MoS2 flakes as active buffer layer for stable perovskitesolar cells. Adv Energy Mater 6:1600920. doi:10.1002/aenm.201600920

172. Reiss P, Carriere M, Lincheneau C et al (2016) Synthesis of semiconductor nanocrystals,focusing on nontoxic and earth-abundant materials. Chem Rev 116:10731–10819. doi:10.1021/acs.chemrev.6b00116

173. Aldakov D, Lefrançois A, Reiss P (2013) Ternary and quaternary metal chalcogenidenanocrystals: synthesis, properties and applications. J Mater Chem C 1:3756–3776. doi:10.1039/C3TC30273C

174. Sato K, Ono K, Izuishi T et al (2016) The effect of CdS on the charge separation andrecombination dynamics in PbS/CdS double-layered quantum dot sensitized solar cells.Chem Phys 478:159–163. doi:10.1016/j.chemphys.2016.03.014

175. Kim JY, Yang J, Yu JH et al (2015) Highly Efficient Copper indium selenide quantum dotsolar cells: suppression of carrier recombination by controlled Zns overlayers. ACS Nano9:11286–11295. doi:10.1021/acsnano.5b04917

176. Dong J, Zhu Y, Jia S, Zhu Z (2016) Blocking the back reaction in quantum dot sensitizedsolar cells via surface modification with organic molecules. RSC Adv 6:14224–14228.doi:10.1039/c5ra26168f

177. Huang F, Zhang Q, Xu B et al (2016) A comparison of ZnS and ZnSe passivation layers onCdS/CdSe co-sensitized quantum dot solar cells. J Mater Chem A 4:14773–14780. doi:10.1039/c6ta01590e

178. de la Fuente MS, Sánchez RS, González-Pedro V et al (2013) Effect of organic and inorganicpassivation in quantum-dot-sensitized solar cells. J Phys Chem Lett 4:1519–1525. doi:10.1021/jz400626r

179. Lan X, Voznyy O, García Pelayo, de Arquer F et al (2016) 10.6% certified colloidalquantum dot solar cells via solvent-polarity-engineered halide passivation. Nano Lett16:4630–4634. doi:10.1021/acs.nanolett.6b01957

180. Lan X, Voznyy O, Kiani A et al (2016) Passivation using molecular halides increasesquantum dot solar cell performance. Adv Mater 28:299–304. doi:10.1002/adma.201503657

181. Zhang X, Johansson EMJ (2017) Reduction of charge recombination in PbS colloidalquantum dot solar cells at the quantum dot/ZnO interface by inserting a MgZnO buffer layer.J Mater Chem A 5:303–310. doi:10.1039/c6ta07775g

182. Hwang JY, Lee SA, Lee YH, Seok SI (2010) Improved photovoltaic response ofnanocrystalline CdS-sensitized solar cells through interface control. ACS Appl MaterInterfaces 2:1343–1348. doi:10.1021/am900917n

183. Zhua G, Pana L, Xua T et al (2011) Cascade structure of TiO2/ZnO/CdS film for quantumdot sensitized solar cells. J Alloys Comp 509:7814–7818. doi:10.1016/j.jallcom.2011.05.043

184. Chen Z, Wei C, Li S et al (2016) CdS/CdSe Co-sensitized solar cells based on hierarchicallystructured SnO2/TiO2 hybrid films. CdS/CdSe co-sensitized solar cells based on hierarchi-cally structured SnO2/TiO2 hybrid films. doi:10.1186/s11671-016-1493-7

185. Kuang PY, Su YZ, Xiao K et al (2015) Double-Shelled CdS- and CdSe-cosensitized ZnOporous nanotube arrays for superior photoelectrocatalytic applications. ACS Appl MaterInterfaces 7:16387–16394. doi:10.1021/acsami.5b03527

186. Chen C, Ali G, Yoo SH et al (2011) Improved conversion efficiency of CdS quantumdot-sensitized TiO2 nanotube-arrays using CuInS2 as a co-sensitizer and an energy barrierlayer. J Mater Chem 21:16430–16435. doi:10.1039/C1JM13616J

236 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

187. Mahadik MA, Shinde PS, Cho M, Jang JS (2015) Fabrication of a ternary CdS/ZnIn2S4/TiO2

heterojunction for enhancing photoelectrochemical performance: effect of cascadingelectron–hole transfer. J Mater Chem A 3:23597–23606. doi:10.1039/c5ta07454a

188. Kokal RK, Kumar PN, Deepa M, Srivastavab AK (2015) Lead selenide quantum dots andcarbon dots amplify solar conversion capability of a TiO2/CdS photoanode. J Mater Chem A3:20715–20726. doi:10.1039/c5ta04393j

189. Chen Z, Peng W, Zhang K et al (2012) Surface ion transfer growth of ternary CdS1−xSexquantum dots and their electron transport modulation. Nanoscale 4:7690–7697. doi:10.1039/C2NR31703F

190. Li H, Cheng C, Li X et al (2012) Composition-graded ZnxCd1−xSe@ZnO core − shellnanowire array electrodes for photoelectrochemical hydrogen generation. J Phys Chem C116:3802–3807. doi:10.1021/jp204747w

191. Baran MP, Korsunskaya NE, Stara TR et al (2016) Graded ZnS/ZnSxO1−x heterostructuresproduced by oxidative photolysis of zinc sulfide: structure, optical properties andphotocatalytic evolution of molecular hydrogen. J Photochem Photobiol A: Chem329:213–220. doi:10.1016/j.jphotochem.2016.07.003

192. Wang M, Qin H, Fang Y et al (2015) FeS2-sensitized ZnO/ZnS nanorod arrays for thephotoanodes of quantum-dot-sensitized solar cells. RSC Adv 5:105324–105328. doi:10.1039/c5ra23281c

193. Ning Z, Tian H, Yuan C et al (2010) Solar cells sensitized with type-II ZnSe–CdS core/shellcolloidal quantum dots. Chem Commun 47:1536–1538. doi:10.1039/c0cc03401k

194. Kim J, Yang W, Oh Y et al (2017) Self-oriented Sb2Se3 nanoneedle photocathodes for watersplitting obtained by a simple spincoating method. J Mater Chem A 5:2180–2187. doi:10.1039/c6ta09602f

195. Torimoto T, Kameyama T, Kuwabata S (2014) Photofunctional materials fabricated withchalcopyrite-type semiconductor nanoparticles composed of AgInS2 and its solid solutions.J Phys Chem Lett 5:336–347. doi:10.1021/jz402378x

196. Sasamura T, Okazaki K, Kudo A et al (2012) Photosensitization of ZnO rod electrodes withAgInS2 nanoparticles and ZnS-AgInS2 solid solution nanoparticles for solar cell applica-tions. RSC Adv 2:552–559. doi:10.1039/c1ra00423a

197. Kadlag KP, Patil P, Rao MJ et al (2014) Luminescence and solar cell from ligand-freecolloidal AgInS2 nanocrystals. CrystEngComm 16:3605–3612. doi:10.1039/c3ce42475h

198. Wang Y, Zhang Q, Li Y, Wang H (2015) Preparation of AgInS2 quantum dot/In2S3co-sensitized photoelectrodes by a facile aqueous-phase synthesis route and theirphotovoltaic performance. Nanoscale 7:6185–6192. doi:10.1039/c4nr06458e

199. McDaniel H, Fuke N, Pietryga JM, Klimov VI (2013) Engineered CuInSexS2−x quantumdots for sensitized solar cells. J Phys Chem Lett 4:355–361. doi:10.1021/jz302067r

200. Guo Y, Yin X, Yang Y, Que W (2016) Construction of ZnO/Cu2SnS3 nanorod array filmsfor enhanced photoelectrochemical and photocatalytic activity. RSC Adv 6:104041–104048.doi:10.1039/c6ra22674d

201. van Embden J, Latham K, Duffy NW, Tachibana Y (2013) Near-Infrared AbsorbingCu12Sb4S13 and Cu3SbS4 nanocrystals: synthesis, characterization, and photoelectrochem-istry. J Am Chem Soc 135:11562–11571. doi:10.1021/ja402702x

202. Mitzi DB, Gunawan O, Todorov TK et al (2011) The path towards a high-performancesolution-processed kesterite solar cell. Sol Energy Mater Sol Cells 95:1421–1436. doi:10.1016/j.solmat.2010.11.028

203. Siebentritt S, Schorr S (2012) Kesterites—a challenging material for solar cells. ProgPhotovolt Res Appl 20:512–519. doi:10.1002/pip.2156

204. Siebentritt S (2013) Why are kesterite solar cells not 20% efficient? Thin Solid Films 535:1–4. doi:10.1016/j.tsf.2012.12.089

205. Liu X, Feng Y, Cui H et al (2016) The current status and future prospects of kesterite solarcells: a brief review. Prog Photovolt Res Appl 24:879–898. doi:10.1002/pip.2741

206. Guo Q, Hillhouse HW, Agrawal R (2009) Synthesis of Cu2ZnSnS4 nanocrystal ink and itsuse for solar cells. J Am Chem Soc 131:11672–11673. doi:10.1021/ja904981r

References 237

207. Toyoda T, Shen Q (2012) Quantum-dot-sensitized solar cells: effect of nanostructured TiO2

morphologies on photovoltaic properties. J Phys Chem Lett 3:1885–1893. doi:10.1021/jz3004602

208. Lin L, Zhain T, Bando Y, Golberg D (2012) Recent progress of one-dimensional ZnOnanostructured solar cells. Nano Energy 1:91–106. doi:10.1016/j.nanoen.2011.10.005

209. Tiwari JN, Tiwari RN, Kim KS (2012) Zero-dimensional, one-dimensional,two-dimensional and three-dimensional nanostructured materials for advanced electrochem-ical energy devices. Progr Mater Sci 57:724–803. doi:10.1016/j.pmatsci.2011.08.003

210. Chen J, Li C, Eda G et al (2011) Incorporation of graphene in quantum dot sensitized solarcells based on ZnO nanorods. Chem Commun 47:6084–6086. doi:10.1039/c1cc10162e

211. Zhu Z, Qiu J, Yan K, Yang S (2013) Building high-efficiency CdS/CdSe-sensitized solarcells with a hierarchically branched double-layer architecture. ACS Appl Mater Interfaces5:4000–4005. doi:10.1021/am400235g

212. Sun B, Hao Y, Guo F et al (2012) Fabrication of Poly(3-hexylthiophene)/CdS/ZnO coreshellnanotube array for semiconductor-sensitized solar cell. J Phys Chem C 116:1395–1400.doi:10.1021/jp206067m

213. Mukherjee B, Smith YR, Subramanian V (2012) CdSe nanocrystal assemblies on anodizedTiO2 nanotubes: optical, surface, and photoelectrochemical properties. J Phys Chem C116:15175–15184. doi:10.1021/jp208879f

214. Yang K, Xu C, Huang L et al (2011) Hybrid nanostructure heterojunction solar cells fabrica–ted using vertically aligned ZnO nanotubes grown on reduced graphene oxide.Nanotechnology 22:405401

215. Chen H, Zhu L, Liu H, Li W (2013) ITO porous film-supported metal sulfide counterelectrodes for high-performance quantum-dot-sensitized solar cells. J Phys Chem C117:3739–3746. doi:10.1021/jp309967w

216. Yang YY, Zhang QX, Wang TZ et al (2013) Novel tandem structure employingmesh-structured Cu2S counter electrode for enhanced performance of quantumdot-sensitized solar cells. Electrochim Acta 88:44–50. doi:10.1016/j.electacta.2012.09.094

217. Radich JG, Dwyer R, Kamat PV (2011) Cu2S reduced graphene oxide composite forhigh-efficiency quantum dot solar cells. overcoming the redox limitations of S2−/Sn

2− at thecounter electrode. J Phys Chem Lett 2:2453–2460. doi:10.1021/jz201064k

218. Balis N, Dracopoulos V, Bourikas K, Lianos P (2013) Quantum dot sensitized solar cellsbased on an optimized combination of ZnS, CdS and CdSe with CoS and CuS counterelectrodes. Electrochim Acta 91:246–252. doi:10.1016/j.electacta.2013.01.004

219. Lin CY, Teng CY, Li TL et al (2013) Photoactive p-type PbS as a counter electrode forquantum dot-sensitized solar cells. J Mater Chem A 1:1155–1162. doi:10.1039/c2ta00251e

220. Eskandari M, Ghahary R, Shokric M, Ahmadi V (2016) Zinc oxide/copper sulfide nanorodsas a highly catalytic counter electrode material for quantum dot sensitized solar cells. RSCAdv 6:51894–51899. doi:10.1039/c6ra11034g

221. Savariraj AD, Rajendrakumar G, Selvam S et al (2015) Stacked Cu1.8S nanoplatelets ascounter electrode for quantum dot-sensitized solar cell. RSC Adv 5:100560–100567. doi:10.1039/c5ra20965j

222. Guo W, Du Z, Zhao Q et al (2016) Controlled sulfidation approach for coppersulfide − carbon hybrid as an effective counter electrode in quantum-dot-sensitized solarcells. J Phys Chem C 120:16500–16506. doi:10.1021/acs.jpcc.6b05211

223. Xu J, Yang X, Wong TL, Lee CS (2012) Large-scale synthesis of Cu2SnS3 and Cu1.8Shierarchical microspheres as efficient counter electrode materials for quantum dot sensitizedsolar cells. Nanoscale 4:6537–6542. doi:10.1039/c2nr31724a

224. Zhang H, Yang C, Du Z et al (2017) Graphene hydrogel-based counter electrode for highefficiency quantum dot-sensitized solar cells. J Mater Chem A 5:1614–1622. doi:10.1039/c6ta08443e

225. Venkata-Haritha M, Gopi CV, Lee YS, Kim HJ (2016) Phase transformations of novel CuxSnanostructures as highly efficient counter electrodes for stable and reproducible quantumdot-sensitized solar cells. RSC Adv 6:101185–101197. doi:10.1039/c6ra23763k

238 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

226. Yang Z, Chen CY, Liu CW, Chang HT (2010) Electrocatalytic sulfur electrodes forCdS/CdSe quantum dot-sensitized solar cells. Chem Commun 46:5485–5487. doi:10.1039/c0cc00642d

227. Faber MS, Park K, Cabán-Acevedo M et al (2013) Earth-abundant cobalt pyrite (CoS2) thinfilm on glass as a robust, high-performance counter electrode for quantum dot-sensitizedsolar cells. J Phys Chem Lett 4:1843–1849. doi:10.1021/jz400642e

228. Khalili SS, Dehghani H, Afrooz M (2017) Composite films of metal doped CoS/carbonallotropes: efficient electrocatalyst counter electrodes for high performance quantumdot-sensitized solar cells. J Colloid Interface Sci 493:32–41. doi:10.1016/j.jcis.2017.01.005

229. Gopi C, Venkata-Haritha M, Lee Y, Kim H (2016) ZnO nanorods decorated with metalsulfides as stable and efficient counter-electrode materials for high-efficiency quantumdot-sensitized solar cells. J Mater Chem A 4:8161–8171. doi:10.1039/c6ta02415g

230. Tachan Z, Shalom M, Hod I et al (2011) PbS as a highly catalytic counter electrode forpolysulfide-based quantum dot solar cells. J Phys Chem C 115:6162–6166. doi:10.1021/jp112010m

231. Wang S, Shen T, Bai H et al (2016) Cu3Se2 nanostructure as a counter electrode for highefficiency quantum dot-sensitized solar cells. J Mater Chem C 4:8020–8026. doi:10.1039/c6tc02309f

232. Wu M, Lin X, Wang Y, Ma T (2015) Counter electrode materials combined with redoxcouples in dye- and quantum dot-sensitized solar cells. J Mater Chem A 3:19638–19656.doi:10.1039/c5ta03682h

233. Kozytskiy AV, Stroyuk OL, Skoryk M, Kuchmiy SY (2015) Photoassisted formation ofCuxS-based cathodes for CdS-sensitized solar cells with S2−/Sx

2− electrolyte. PhotochemPhotobiol Sci 14:942–947. doi:10.1039/C4PP00314D

234. Du Z, Pan Z, Fabregat-Santiago F et al (2016) Carbon counter-electrode-basedquantum-dot-sensitized solar cells with certified efficiency exceeding 11%. J Phys ChemLett 7:3103–3111. doi:10.1021/acs.jpclett.6b01356

235. Jayathilaka K, Kapaklis V, Siripala W, Jayanetti J (2012) Sulfidation of electrodepositedmicrocrystalline/nano-crystalline cuprous oxide thin films for solar energy applications.Semicond Sci Technol 27:125019

236. Kozytskiy AV, Stroyuk OL, Breslavskiy AV et al (2013) Nanostructured films of CuxS—counter electrodes for solar cells based on FTO/ZnO/CdS heterostructures and sulfide/poly–sulfide redox couple. Theor Exp Chem 49:213–218. doi:10.1007/s11237-013-9317-7

237. Cao Y, Xiao Y, Jung JY et al (2013) Highly electrocatalytic Cu2ZnSn(S1−xSex)4 counterelectrodes for quantum-dot-sensitized solar cells. ACS Appl Mater Interfaces 5:479–484.doi:10.1021/am302522c

238. Zhang X, Huang X, Yang Y et al (2013) Investigation on new CuInS2/Carbon compositecounter electrodes for CdS/CdSe cosensitized solar cells. ACS Appl Mater Interfaces5:5954–5960. doi:10.1021/am400268j

239. Stroyuk OL, Kuchmiy SY, Kryukov AI, Pokhodenko VD (2010) Semiconductor catalysisand photocatalysis on the nanoscale. Nova Science Publishers, Inc., New York

240. Zeng X, Zhang W, Xie Y et al (2013) Low-cost porous Cu2ZnSnSe4 film remarkablysuperior to noble Pt as counter electrode in quantum dot-sensitized solar cell system. J PowerSources 226:359–362. doi:10.1016/j.jpowsour.2012.11.023

241. Wang X, Liu H, Shen W (2016) Controllable in situ photo-assisted chemical deposition ofCdSe quantum dots on ZnO/CdS nanorod arrays and its photovoltaic application.Nanotechnology 27:085605

242. Zhang Y, Pei Q, Liang J et al (2015) Mesoporous TiO2-based photoanode sensitized by BiOIand investigation of its photovoltaic behavior. Langmuir 31:10279–10284. doi:10.1021/acs.langmuir.5b02248

243. Wang R, Shang Y, Kanjanaboos P et al (2016) Colloidal quantum dot ligand engineering forhigh performance solar cells. Energy Environ Sci 9:1130–1143. doi:10.1039/C5EE03887A

References 239

244. Paul GS, Kim JH, Kim MS et al (2012) Different hierarchical nanostructured carbons ascounter electrodes for CdS quantum dot solar cells. ACS Appl Mater Interfaces 4:375–381.doi:10.1021/am201452s

245. Seol M, Ramasamy E, Lee J, Yong K (2011) Highly efficient and durable quantum dotsensitized ZnO nanowire solar cell using noble-metal-free counter electrode. J Phys Chem C115:22018–22024. doi:10.1021/jp205844r

246. Du Z, Tong J, Guo W et al (2016) Cuprous sulfide on Ni foam as a counter electrode forflexible quantum dot sensitized solar cells. J Mater Chem A 4:11754–11761. doi:10.1039/c6ta04934f

247. Kumar PN, Kolay A, Kumar SK et al (2016) Counter electrode impact on quantum dot solarcell efficiencies. ACS Appl Mater Interfaces 8:27688–27700. doi:10.1021/acsami.6b08921

248. Sudhagar P, Ramasamy E, Cho WH et al (2011) Robust mesocellular carbon foam counterelectrode for quantum-dot sensitized solar cells. Electrochem Commun 13:34–37. doi:10.1016/j.elecom.2010.11.006

249. Jiao S, Du J, Du Z et al (2017) Nitrogen-doped mesoporous carbons as counter electrodes inquantum dot sensitized solar cells with a conversion efficiency exceeding 12%. J Phys ChemLett 8:559–564. doi:10.1021/acs.jpclett.6b02864

240 4 Semiconductor-Based Liquid-Junction Photoelectrochemical …

Chapter 5Synthesis of Nanocrystalline Photo-ActiveSemiconductors

Photochemical activity of a semiconductor substance depends on a variety of dif-ferent properties, including the spectral sensitivity range, band gap Eg, positions(potentials) of the conduction band ECB and valence band EVB, mobility of pho-togenerated charge carriers and density of donor and acceptor states, surfacemorphology, adsorption capability, etc. Only a limited number of the reportedsemiconductors has a “complete” set of characteristics necessary for the photo-catalytic action, mostly from the AIVBVI and AIIBVI groups. The photochemicalactivity was broadly reported for metal oxides (mostly TiO2, ZnO and rarely—WO3, Fe2O3, SnO2, Bi2O3, etc.), metal chalcogenides (most frequently—CdS,CdxZn1−xS, ZnS and rarely—CdSe, CdTe, In2S3, HgS, MoS2, etc.), and salts ofmetal based acids—metallates (for example, Na2Ti2O7, NaTaO3, SrTiO3, etc.). Thephotochemical activity of other semiconductors is reported much scarcely.

The situation, when the multiple selection criteria are met by only a limitedsemiconductor substances led to a dual character of the development in the syn-thetic aspects of the photo-active nanocrystalline semiconductors. The first, rela-tively minor, direction consists in a search and testing of new semiconductorsamong more and more complex and exotic substances with reported or yet unre-ported semiconductor properties. The second, major, direction combines the studiesof nanocrystalline materials with new structures but produced from “usual” (dis-cussed above) photoactive semiconductors. This group includes nanocrystallinepowders and films, mesoporous and layered semiconductors, as well as dopednanodispersed semiconductors and composites of semiconductors with othersemiconductors, metals, conjugated polymers, carbon allotropes, and other sub-stances. Each new photocatalyst/photoelectrocatalyst and a new structure exhibitboth advantages and drawbacks associated with the issues of activity, stability,technological applicability, etc. In this view, a constant exploration of new syntheticways to the photo-active materials in both directions is of a high importance for theprogress of the semiconductor-based light harvesting systems.

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_5

241

5.1 Colloidal Semiconductors

Colloidal semiconductors were studied as photocatalysts/photoelectrocatalysts formore than four decades. They are popular owing to a low light scattering enablingextensive spectral characterization, relative simplicity of the synthesis and stabi-lization, quasi-homogeneity of the reaction systems allowing for detailed kineticstudies as well as a variety of reproducible methods of the semiconductor particlesize variation by means of the “classical” colloidal chemistry. Besides, the opticaltransparency and homogeneity of colloidal semiconductors made them perfectmodels for the studies of photophysical and primary photochemical events allowingto shed light on the mechanisms of many photochemical/photocatalytic/photoelectrochemical transformations.

This section presents an overview of the syntheses and stabilization of colloidalsemiconductors exhibiting photochemical activity—metal oxides, sulfides, sele-nides as well as some other classes. A special accent is made on the size variation ofcolloidal semiconductors, typically achieved by a post-synthesis treatment. All theliterature sources discussed in the present chapter reported some or otherphotochemical/photoelectrochemical/photocatalytic process, both of endothermicnature, such as the water splitting and CO2 reduction and of the oxidative nature, inparticular, decomposition of inorganic and organic compounds, the water oxidation,etc.

Metal chalcogenides. Colloidal metal sulfides are typically produced in reac-tions between soluble metal salts and hydrogen sulfide or its soluble salts as well aswith the substances releasing sulfide/hydrosulfide ions during the hydrolysis, suchas thiourea (SC(NH2)2 + 2H2O + 3OH− ! HS− + 2NH4OH + CO3

2−) or thioac-etamide (CH3CSNH2 + OH− + 2H2O ! HS− + CH3COOH + NH4OH). Theselenide ions are typically generated by treating selenium (2Se + N2H4 +2OH− ! 2HSe− + N2 + 2H2O) or selenite salts (4SeO3

2− + 3BH4− + H2O !

4HSe− + H2BO3− + 4OH−) with strong reducing agents, or via the decomposition

of sodium selenosulfate Na2SeSO3 in alkaline solutions (SeSO32− + OH− !

HSe− + SO42−). Telluride ions can be conveniently produced by the electro-

chemical reduction of metallic tellurium in alkaline solutions (Te + 2e− ! Te2−).Most of the photoactive metal chalcogenides have a low solubility in water,

while the reaction between chalcogenide and metal (or a metal complex) ions istypically very fast. As a result, such reactions yield highly aggregated precipitateswhen performed without additional substances—stabilizers. The stabilizer stops themetal chalcogenide crystal formation on the stage of nanoparticles (NPs) preventingfurther growth due to the strong adsorption on the NP surface [1, 2]. By thestabilization mechanism, the stabilizers can be assorted into several types, includingpotential-depending ions (one of the ions forming the NP lattice), organicsulfur-containing acids and alcohols, organic/inorganic polymers, as well as col-loidal nanoparticulate stabilizers.

242 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

Stabilization of colloidal metal chalcogenides with ionic agents can be achievedby introducing an excess of the metal cations over the amount needed for thebinding of the chalcogenide anions. After the metal chalcogenide formation theexcessive metal ions adsorb on the surface of colloidal NPs and impart them with apositive charge. The charge creates an electrostatic barrier preventing theinter-particle interaction [2, 3]. For example, colloidal CdS particles can be stabi-lized by a Cd2+ excess in N,N-dimethylformamide (DMF) [4]. The stabilization ofCdS NPs by excessive Cd2+ ions is broadly reported for aqueous and methanol/ethanol solutions [5–9]. A similar stabilization effect of a Zn2+ excess was alsoreported for 2–5 nm ZnS particles in DMF, acetonitrile, and methanol [10, 11].

The metal ion-stabilized colloidal semiconductors typically reveal a relativelylow stability that depends strongly on the temperature and solution pH as well as onthe presence of other ion admixtures that can neutralize the surface charge. Muchmore stable colloidal metal chalcogenide NPs can be synthesized in the presence oforganic mercapto-acids, mercapto-alcohols, and other bifunctional organic com-pounds. The mercapto- compounds can form strong covalent bonds with theunder-coordinated metal ions on the NP surface. The stabilizer layer forms a stericor electrostatic (in the case of bifunctional mercapto-acids with ionized carboxylgroup) barrier preventing the agglomeration of colloidal NPs. In particular, thephoto-active CdS NPs were produced via the stabilization with mercaptoacetic acid(MAA) and its salts [12, 13], thiophenol [14], alkyl thiols with C6, C12 and C18

alkyl radicals [15], and mercaptoethanol [16]. Similar approaches were used tostabilize PbS [17], ZnS [18], Bi2S3 [19], In2S3 [16], and CdTe NPs [20].

A pronounced NP stabilization effect was also reported for someamino-compounds capable of forming complexes with the metal ions on the NPsurface. The amine-assisted stabilization was reported for CdS [21, 22], PbS [23],and CdTe NPs [24].

Stabilization with polymers. The metal sulfide NPs can be reliably stabilized inaqueous solutions by inorganic polyanions such as sodium polyphosphate (SPP) thatcan strongly adsorb on the NP surface and create a dense steric/electrostatic barrieragainst the NP aggregation. The SPP composition can be described as (NaPO3)n withn varying in a broad range with a distribution maximum at n = 6. The polyphos-phates combine mild buffer properties, chemical stability, and inertness towardtypical photochemical reactions taking place on the surface of metal-chalcogenideNPs. As a result, SPP was broadly used to produce colloidal NPs of CdS [16, 25–29],CdxZn1−xS [17], Bi2S3 and Sb2S3 [30–33], and Ag2S [34].

The bulky organic polymers such as polyvinyl alcohol (PVA),polyvinylpyrrolidone (PVP), polyethylene glycol (PEG), gelatin, polyacrylamide,peptides, polyethyleneimine (PEI) etc., can also adsorb on the surface of metalchalcogenide NPs providing a thick steric barrier preventing the NPs from con-tacting each other. These polymers can be used for the stabilization of metal-sulfideNPs (CdS [35–37], CdxZn1−xS [17], PbS [16, 38]) in water and in polar solventssuch as acetonitrile, tetrahydrofuran, DMF or methanol. Finally, the metal-sulfidenano-photocatalysts can be stabilized by colloidal particles of inert materials, suchas silica, that interact with the NPs via electrostatic forces and shield the

5.1 Colloidal Semiconductors 243

inter-particle attraction. For example, colloidal CdS NPs were stabilized by thecolloidal Ludox© silica [39].

Size variation of colloidal metal chalcogenide nano-photocatalysts. As the sizeof metal-chalcogenide NPs is reduced, a size dependence of the photochemicalproperties of metal-chalcogenide NPs becomes more and more pronounced. Themost distinct size dependences can be observed in the regime of strong spatialexciton confinement, typically for the sizes of 1–10 nm (the so-calledquantum-sized NPs). Therefore, the size variation of colloidal NPs, as well as thefocusing/defocusing of size distribution, have a paramount importance for thestudies of special features of the quantum-sized NPs and for the progress of the lightharvesting using nanocrystalline metal chalcogenide semiconductors in general.The synthetic approaches that were briefly discussed above provide two basic waysof NP size variation: (i) through variation of the synthesis conditions; (ii) through apost-synthesis treatment of the previously prepared NPs.

The first group of methods is based on the variation of the ratio between theprimary nuclei formation rate and the nuclei growth rate. The nuclei formation rateVn depends on the oversaturation of the solution with respect to the low-solublesubstance [3]: Vn = kn � (C − C0) � C0

−1, where kn is the nucleation rate constant,C and C0 are the concentrations of over-saturated and saturated solutions, respec-tively. For most metal-chalcogenides typically C � C0, and, therefore, the nucleiformation rate is constant and very high. The nucleation lowers the over-saturationand, after reaching a certain critical concentration, the formation of new nucleistops, while the present nuclei continue to grow. The nuclei growth rate Vg can beexpressed [3] as Vg = kg � DS(C − C0) � d−1 � DSC � d−1, where k2 is a nucleigrowth rate constant, D is a diffusion coefficient, S is the NP surface area, d is adiffusion layer length where the concentration changes from C to C0.

Therefore, at Vn = const, any influence that lowers Vg results in a decrease of thesize of final metal-chalcogenide NPs. The typical factors are the viscosity andtemperature of the solution, the concentration of the reactants, the presence of astabilizer as well as the stabilizer type and content.

An increase in the solution viscosity results in a slowing of themetal-chalcogenide monomers diffusion toward the growing NP surface and can beused as an efficient tool for affecting the NP size. For example, the absorption bandedge kbe of ZnS NPs synthesized in aqueous solutions can be found at 330 nmcorresponding to an average size of dav = 7 nm. As the viscosity of the solution isincreased via partial water substitution with glycerol the absorption band shows ablue shift indicating a decrease of the average NP size. The ZnS NPs produced inpure glycerol are characterized by kbe = 297 nm corresponding to dav = 4 nm(Fig. 5.1a) [40]. A similar dependence of the average NP size on the solventviscosity was observed for CdS NPs produced in glycerol, ethylene glycol, ethanol,and water.

The average size of metal-chalcogenide NPs decreases with a lowering of thesolvent temperature. The diffusion coefficient can be expressed as D = kTB−1,where B is a constant depending on the shape of colloidal NPs. As the temperatureis decreased the diffusion coefficient of metal-chalcogenide monomers decreases as

244 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

well, resulting in a slowing of the NP growth. For example, the average size of CdSNPs stabilized by thiophenol can be tuned from 3.8 to 5.0 nm via an elevation ofthe solution temperature from 5 to 25 °C [14]. A temperature increase from 4 to40 °C during the growth of SPP-stabilized CdSe NPs in aqueous solutions results ina pronounced shift of the absorption band edge from 650 to 585 nm (Fig. 5.1b)indicative of dav increase from 4 to 8 nm [41].

The average size of colloidal NPs grows also with an increase in the reactantconcentration. For example, as the CdCl2 and Na2S content is elevated from1 � 10−4 M to 1 � 10−3 M the bandgap of CdS NPs forming in the presence ofSPP decreases from 2.64 to 2.50 eV indicating a dav increase from 6.5–6.6to *10 nm [25]. At the same time, for the metal chalcogenides a ratio between theconcentrations of metal salt and chalcogenide source is typically a much moreimportant factor than the absolute reactant concentrations. By introducing an excessof the metal or chalcogen one can strongly influence the size and optical propertiesof colloidal NPs in a broad range. In particular, the presence of a 100% excess ofsodium sulfide during the synthesis of aqueous CdS NPs results in a *70 nm blueshift of kbe (Fig. 5.1c) attesting to a decrease of the average NP size from *9 to4 nm.

The synthesis performed with a Na2S excess allows also to focus the size dis-tribution of CdS NPs. The width of size distribution can be evaluated from anabsorption maximum width. As Fig. 5.1c shows the size distribution of CdS NPsdecreases from dav ± 40% for the stoichiometric Cd2+:S2− ratio to dav ± 20% forcolloidal solutions produced with a 100% Na2S excess.

A size variation can be achieved by varying the nature and concentration ofstabilizers. A dependence between the stabilizer content and the average size ofmetal-chalcogenide NPs was reported for the CdS NPs capped with thioglycerol[42], CdS and In2S3 NPs protected by mercaptoethanol [16], and MPA-stabilizedBi2S3 NPs [43]. In particular, by varying the metal/thioglycerol ratio the average

Fig. 5.1 Absorption spectra of a colloidal ZnS NPs synthesized in water (curve 1), glycerol(curve 4) and water:glycerol mixtures with a ratio of 1:2 (curve 2) and 2:1 (curve 3); b CdSe NPssynthesized in water at 4 °C (curve 1) and 40 °C (curve 2); c CdS NPs synthesized at [CdCl2]:[Na2S] = 1:2 (curve 1) and 1:1 (curve 2), dashed lines reflect approximations of the first excitonicmaxima with Gaussian curves

5.1 Colloidal Semiconductors 245

size of CdS and ZnS NPs can be tuned in the range of 3.8–7.2 nm [42] and 1.8–3.5 nm [18], respectively.

Of special interest for the NP size variation are organized micro/nano-objects, inparticular, the lipid vesicles and inverted micelles. The inverted “water in oil”microemulsions can be formed by the surfactant-assisted water solubilization innon-polar organic solvents. The most precise NP size variation is reported for theinverted micellar systems based on the non-ionogenic Triton X-100 and anionicsurfactant sodium bis-octadecyl sulfosuccinate (Aerosol OT or AOT, Fig. 5.2a).

The size of water drops solubilized inside the AOT micelles (Fig. 5.2b) can beprecisely tailored by changing the ratio of molar concentrations of water and thesurfactant, w = [H2O]/[AOT]. The average radius r of the water droplets can becalculated as r3 � (r − L)−3 = 1 + V2 � (wV1)

−1, where L is the linear size of anAOT molecule (1.5 Å), V1 is the volume of a water molecule (30 Å3), V2 is thevolume of an AOT molecule (825 Å3) [44]. One can tune the average size of metalchalcogenide NPs by varying the size of water droplets where the interactionbetween metal salts and chalcogenide sources takes place. Figure 5.2c exemplifiesthis approach for CdS NPs synthesized in an inverted “water/AOT/heptane” system,where a nearly linear dependence between dav and w was observed [44, 45]. At thesame time, a very uniform distribution of the solubilized water droplets—“nanoreactors” by their size allows reaching a very narrow size distribution ofsemiconductor NPs formed in such media.

The size-selected 2–8 nm CdS NPs can be extracted from the inverted micellarmedia by using the alkyl thiol-grafted silica, yielding visible-light-driven photo-catalysts of the hydrogen evolution from water/2-propanol solutions [46]. A broadassortment of MoS2 NPs in a size range of 2–15 nm can also be produced in theAOT-based micellar media [47, 48].

An alternative way of tailoring the average size of metal-chalcogenidenano-photocatalysts is a post-synthesis treatment of raw colloidal solutions wherean ensemble of differently sized NPs is present. The most frequent are thesize-selective fractionation and the thermal treatment.

Fig. 5.2 Molecular structure of AOT (a), layout of an inverted “water-in-heptane” micelle (b) anddependence between the size of CdS NPs and w (plotted using the data reported in [44, 45])

246 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

The size-selective fractionation can be achieved owing to a different adsorptioncapability of differently sized NPs (for the case of gel chromatographic separation)or to a different rate of propagation in the gel in an external electric field. Forexample, the CdS NPs can be separated by the electrophoresis of an NP-richpolyacrylamide gel [27, 49]. After the electrophoresis, the gel parts containingdifferently sized NPs can be separated mechanically and the NPs extracted bywater.

The thermal treatment of polydisperse metal chalcogenide colloids results inOstwald ripening, that is in the growth of larger NPs at the expense of the disso-lution of smaller NPs [3]. The driving force of the process is a difference in thesurface tension of the smaller and larger NPs. The smaller NPs with an excess ofsurface energy tend to dissolve creating a concentration gradient in the treatedsolution. The gradient results in a mass transfer from smaller to larger NPs resultingin the growth of larger NPs and complete dissolution of smaller NPs.

The Ostwald ripening that focuses the NP size distribution is far from being thesole result of the thermal treatment. In a 2–5 nm particle a large portion of atomsresides on the NP surface in a partly under-coordinated and/or disordered state.These atoms become natural “traps” for the photogenerated charge carriers resultingin the radiative and non-radiative recombination competing with the photochemicalreactions. The thermal treatment of colloidal solutions is accompanied by areconstruction of the NP surface layer and the elimination (partial or sometimescomplete) of such defects. The NP surface ordering results thus in an increase of thephotochemical activity and aggregative stability of colloidal semiconductors.

A vivid example of the thermal treatment effect is provided by aqueous colloidalCdSe NPs stabilized by SPP. Heating of the solutions at the boiling point (around98 °C) for 2 h results in a large “red” shift of the absorption band edge indicating aconsiderable increase in the NP size (Fig. 5.3a) [41, 50]. The figure also shows thatthe thermal treatment results in a steeper absorption edge indicating a narrower sizedistribution of CdSe NPs in the treated colloids in accordance with theabove-discussed Ostwald ripening mechanism.

In the frames of this mechanism the NP volume, or r3 (r is the NP radius),increases linearly with the treatment duration t [3]. Linear dependences presented inFig. 5.3b indeed show that the Ostwald ripening is a principal NP growth mech-anism for CdSe NPs at higher temperatures [41]. At the same time, cooling of thecolloidal CdSe solutions down to 4 °C allows to “freeze” the existing sizedistribution.

Metal oxides. Colloidal metal oxides are typically produced by the hydrolysis ofmetal precursors with a partial or complete dehydration of an intermediary metalhydroxide. The most broadly studied semiconductor photocatalyst—titaniumdioxide can be synthesized in the form of colloidal NPs via a sol-gel method basedon the hydrolysis of inorganic salts (TiCl4, TiOSO4, TiOCl2, Ti(SO4)2, etc.) ororganic ethers of Ti(IV) followed by the polycondensation of the intermediatehydroxy-compounds [3, 51]:

5.1 Colloidal Semiconductors 247

Ti ORð Þ4 + H2O ! Ti ORð Þ3 OHð Þ + ROH ! � � � ! Ti OHð Þ4R ¼ CH3CH2; CH3ð Þ2CH, CH3 CH2ð Þ3� �

2Ti(OR)xðOH)y ! ðOH)yðOR)xTi-O-Ti(OR)xðOH)y�1 + H2O

The process starts with the formation of a colloidal TiOx sol, which thentransforms into a highly viscous gel. The sol-gel transformation is completed withthe gel drying into a xerogel and the xerogel annealing (or hydrothermal/microwavetreatment).

To produce colloidal TiO2 only two first stages of the sol-gel transformation areneeded. For example, the TiCl4 hydrolysis in water at *0 °C yields colloidalcrystalline 3–5-nm titania NPs [52–60]. In a similar way, 5–10-nm ZrO2 NPs can beproduced from ZrCl4 [55–57]. The hydrolysis of Ti(IV) ethers, such as titaniumtetraisopropoxide (TTIP) in acidic aqueous solutions yields larger, mostly amor-phous titania NPs. The crystallinity can be enhanced by carrying out the TTIPhydrolysis in dry alcohols [49, 61]. The TiO2 NPs can then be extracted by thevacuum evaporation of the solvent and redispersed in water [61, 62].

Similarly to the above-discussed metal chalcogenides, colloidal titania NPs havea partially amorphous structure and a surface abundant with defects [63] that induceinter-particle interaction and agglomeration. The stability of TiO2 colloids can beenhanced by adding amines and some organic polymers, for example, PVA [63–68].Similar stabilization methods can be also applied to colloidal ZrO2 [66, 67] andSnO2 [69].

A synthesis of TiO2 NPs in micro-capsules of polyelectrolytes—polyallylamineand polystyrene sulfonate in the presence of the PVA stabilizer results in theformation of photocatalytic microreactors capable of the photocatalytic productionof urea from CO2 and nitrate ions at the expense of PVA oxidation [70].

Fig. 5.3 a Absorption spectra of SPP-stabilized aqueous colloidal CdSe NPs before (curve 1) andafter (curve 2) thermal treatment at *98 °C for 2 h. (b) “r3—t” dependences for CdSe NPsproduced at 4 °C (curve 1), 20 °C (curve 2), and 40 °C (curve 3) [41]

248 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

Titania NPs were stabilized by EDTA in anhydrous 2-propanol then extractedand redispersed in aqueous solutions [71]. Dodecylsulfonate anion was found to bean efficient stabilizer of colloidal ZrO2 NPs produced from zirconiumtetra-isopropoxide.

The sol-gel method can be realized in a two-stage scheme, when amorphous Ti(OH)4 is first produced via the hydrolysis of TiCl4 or TTIP and then a peptizingagent is added to the precipitate, typically, HNO3 or HCl. A prolonged (severaldays) interaction between the precipitate and the acid at room T results in completedissolution of Ti(OH)4 and the formation of crystalline TiO2 NPs. The size oftitania NPs depends on the acid concentration and the peptization temperature andcan be varied in a range of 3–40 nm [72–85].

A sol-gel synthesis in the inverted micellar solutions were used to produceultra-small titania NPs. For example, TiO2 NPs as small as 0.5 nm were formed inthe inverted “water/AOT/heptane” systems [44]. The TiCl4 hydrolysis in aninverted micellar medium formed by water, cetyl dimethyl benzyl ammoniumchloride, and benzene was applied to synthesize 0.7–0.9 nm TiO2 NPs [86].

Highly crystalline colloidal titania NPs can be produced by a post-synthesishydrothermal treatment (HTT) of the as-prepared colloids. In this method, thecolloidal solutions were kept in the supercritical conditions at 150–250 °C in steelTeflon-lined autoclaves for 12–48 h. By using colloidal TiO2 produced from TTIPin a water/ethanol mixture in the presence of nitric acid as a raw material, anatasenanocrystals were produced by the HTT with the average size varying from 7 to25 nm depending on the TTIP concentration (Fig. 5.4a) and the water/alcohol ratio(Fig. 5.4b) [87].

Zinc hydroxide is resistant to the dehydration in aqueous solutions and, there-fore, ZnO NPs cannot be produced directly in water by the hydrolysis techniques[88, 89]. In view of this, colloidal ZnO NPs are typically prepared in anhydrousaliphatic alcohols (ethanol, 2-propanol) via the interaction between zinc acetate

Fig. 5.4 Size variation of TiO2 NP produced by HTT as a function of TTIP concentration (a) andwater/ethanol volume ratio (b) (plotted using the data reported in [87])

5.1 Colloidal Semiconductors 249

with sodium, potassium or lithium hydroxide [90–95]. Then the solvent is evapo-rated and ZnO NPs can be redispersed in water.

When the Zn acetate hydrolysis in alcohols starts, very small, <1 nm, ZnOnuclei are first formed. As the process continues, the ZnO nuclei grow into 5–6 nmcrystals [88, 89]. The growth can be clearly observed by a continuous red shift ofthe absorption band edge of ZnO colloids. To ensure the stability of forming ZnONPs in alcohols the post-synthesis annealing at 50–60 °C is typically requiredresulting also in an increased crystallinity of the ZnO NPs [90–96]. By extractingportions of the colloid during the annealing several fractions of ZnO NPs can beproduced with an average NP size of 2.9–4.1 nm [90].

The size of ZnO NPs in alcohols can be also varied by changing the concen-tration of a zinc salt and an alkali. An increase of the zinc acetate and NaOHconcentration by an order of magnitude results in a red shift of the absorption bandedge of ZnO NPs from 343 to 356 nm indicating an increase of dav from 3.7 to4.4 nm [97].

Stable colloidal 3–6 nm ZnO NPs can be produced in DMSO in the interactionbetween zinc acetate and tetraalkyl ammonium hydroxides (Fig. 5.5) [98, 99]. Theaverage size of ZnO nanocrystals does not depend on the nature of the tetraalkylammonium cation and can be tailored mainly by varying the duration and temper-ature of the post-synthesis thermal treatment (Fig. 5.5a). A variation of the treatmentduration at a constant T or a T variation at a constant duration allows for the preciseadjustment of the average size of ZnO nanocrystals in the range of 3–6 nm(Fig. 5.5b).

Fig. 5.5 a Absorption spectra of ZnO/SiO2 NPs produced from ZnO NPs subjected to the thermaltreatment at 60 °C for 1 min (curves 1), 5 min (curves 2), 15 min (curves 3), 60 min (curves 4),and 120 min (curves 5). b TEM images of ZnO/SiO2 NPs produced after 1 min aging before thedeposition of a SiO2 shell. The scale bar is 50 nm. Insert in (b): size distribution of ZnO/SiO2 NPs

250 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

The growth of ZnO nanocrystals can be terminated at any time by freezing of thecolloidal solution or by introducing tetraethyl orthosilicate that hydrolyzes resultingin the formation of core–shell ZnO@SiO2 nanocrystals [98, 99]. The core-shellZnO/SiO2 NPs can be incorporated into polymer films, e.g. PVA, and dried topowders where the nanoparticulate character of ZnO cores is preserved [99]. Also,the DMSO-based colloids can be mixed with a variety of solvents, including waterand benzene.

5.2 Nanocrystalline Powdered Semiconductors

Metal chalcogenides. The nanocrystalline metal sulfide powders can be synthesizedthrough the interactions of sulfur-containing precursors (Na2S, thiourea, thioac-etamide) with metal salts, often in the presence of various structure-directing agents,such as polymers, amino-compounds, surfactants. As a rule, the synthesis is com-plemented with a HTT of metal sulfide in the form of a colloidal solution or aprecipitate immersed in the parental solution. The HTT product is then separatedfrom the supernatant, dried and annealed in an inert atmosphere. For example, thehexagonal 5–8 nm CdS crystals can be produced by the HTT of aqueous solutions ofcadmium acetate and thiourea at 150 °C for 24 h [100]. By introducing ethylene-diamine as a structure-directing agent the nanocrystals can be converted into CdSnanowires (NWs) with a diameter of *50 nm and a length of 3–4 lm (Fig. 5.6a)[101]. In a similar way, diethylenetriamine was used to produce ZnS nanorods(NRs) [102]. In the presence of PVP and the same HTT conditions, the processyields spherical 1.5–2.0 lm agglomerates composed of 12-nm ZnS NPs [103].

By introducing a mixture of two metals, ternary cadmium-indium-sulfide andzinc-indium-sulfide nanopowders can be produced [104]. In the case of CdIn2S4,the HTT in aqueous solutions produces spherical NPs, while in methanol nanotubes(NTs) with a diameter of 25 nm are predominantly formed [105]. The nanocrys-talline ZnIn2S4 powders were synthesized by the HTT of aqueous solutions of metalsalts, thioacetamide, and a surfactant—cetyl trimethyl ammonium bromide orsodium dodecyl sulfonate [106, 107]. The reaction yields porous hierarchicalZnIn2S4 microspheres with a size of 100–400 nm composed of *20-nm units(Fig. 5.6b). After the photodeposition of Pt NPs, such materials act as excellentphotocatalysts of the hydrogen evolution from aqueous sulfide/sulfite solutions[105–109] as discussed in details in Chap. 2. Similar photocatalytic properties wereobserved for the nanocrystalline Cu2WS4 (Fig. 5.6c) and CdLa2S4 powders pro-duced by the HTT [110, 111].

Nanocrystalline powdered metal selenides were obtained via the reduction ofselenites and selenates of various metals, or through the hydrothermal decompo-sition of metal selenosulfates. For example, the HTT of aqueous solutions con-taining zinc nitrate, Na2SeSO3, N2H4 and EDTA at 180 °C for 2 h yieldsnanocrystalline ZnSe powders (Fig. 5.6d) [112].

5.1 Colloidal Semiconductors 251

Metal oxides. The photocatalytic properties of nanocrystalline metal oxidesdepend on a broad variety of factors, in particular, on the NP size and crystalstructure, surface area and acidity, porosity, density of surface hydroxyl groups andstructural defects, etc. [113]. The synthetic approaches to nanocrystalline metaloxides should, therefore, provide broad and reliable means of varying theseparameters in a controllable manner.

As discussed above, nanocrystalline titanium dioxide is typically produced bythe hydrolysis of salts and alkoxides of Ti(IV), drying of the forming gel and theHTT/annealing of the xerogel. This general scheme is illustrated in Fig. 5.7 and caninclude a broad variety of additional steps and options. For example, the hydrolysisof TiCl4 in cold diluted aqueous acid solutions followed by the aging at 80 °C for24 h yields the nanocrystalline mixtures of anatase (3–4 nm) and rutile (13–15 nm)[114]. The oxidative TiCl3 hydrolysis in the presence of ammonia and H2O2 andannealing at 300 °C produces the nanocrystalline anatase powders with an averagegrain size of 13 nm [115]. Adsorption of TTIP on amorphous carbon followed bythe hydrolysis with the air moisture and annealing at 400–600 °C can be used to

Fig. 5.6 SEM images of CdS nanowires (a), ZnIn2S4 microspheres (b), Cu2WS4 nanoplates (c),and ZnSe nano-“stars” (d). Reprinted with permissions from Refs. [101] (a) and [112] (d),copyright (2007, 2008) American Chemical Society; [106] (b) and [110] (c), copyright (2008,2010) Elsevier

252 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

form 10–20-nm titania crystals for the photocatalytic hydrogen evolution fromwater/ethanol mixtures [116].

The nanocrystalline titania with a grain size of 15 nm was prepared through theTTIP hydrolysis in the presence of ice acetic and nitric acids followed by the HTTat 220 °C for 12 h [117]. The hydrothermal treatment of aqueous solutions ofTiOSO4, H2TiO(C2O4)2 or TiO(NO3)2 results in the nanocrystalline TiO2 powderswith a NP size of 20–50 nm [118].

Monodisperse titania nanocrystals were synthesized by a multiple repetition ofthe TiCl4 addition to hot (80 °C) water and the neutralization of released acetic acidwith ammonia. The treatment results in the dissolution of smaller NPs allowing tofocus the size distribution of titania NP ensemble. The final size of TiO2 NPs growsgradually from 17 to 22 nm as the number of repetitions of the procedure isincreased from 7 to 30 [119, 120].

The principal factors affecting the phase composition and grain size of nanocrys-talline titania powders formed by the HTT are the temperature and duration of thetreatment [121, 122], as well as the parental solution acidity [123, 124]. Figure 5.8ashows a scheme of the formation of three typical titania phases from the individualTiO6 octahedra [125]. As the HTT temperature is increased from 100 to 300 °C theaverage size of titania nanocrystals grows from 7 to 18 nm (Fig. 5.8b) [121].

Fig. 5.7 Layout of thesol-gel synthesis ofnanocrystalline titaniaphotocatalysts

5.2 Nanocrystalline Powdered Semiconductors 253

The phase composition of titania nanocrystals depends on the grain size. Thesmallest reported TiO2 NPs (5–10 nm) crystallize in the anatase modification, largerNPs reveal a structure of brookite, while even larger crystals can be typically foundas a rutile admixture [124, 125].

The Ti(IV) hydrolysis in strongly acidic solutions results mostly in nanocrys-talline rutile. The rutile NRs can be also produced via HTT of TiCl4 solutionscontaining chloroform [126]. A prolonged (around 3 months) aging of titanate NTsin diluted aqueous solutions of mineral acids (HCl, H2SO4) results in the NTconversion into ultrasmall (around 3 nm) rutile nanocrystals [127]. The HTT ofprotonated titanates, such as H2Ti4O9 � 1.2H2O in acid solutions yields titania inthe form of nanocubes and NRs with a size depending on the HCl concentration[128].

Along with the HTT temperature and duration the structural characteristics ofnanocrystalline TiO2 can also be affected by the composition and pressure ofgaseous phase present in the autoclave. For example, HTT of a titania gel at 80 °Cin deaerated conditions yields aggregated 30-nm titania crystals. However, in thepresence of air with other conditions being equal the synthesis results in TiO2 NRswith a length of up to 200 nm [129].

The microwave treatment of titania gels is an alternative to HTT and allows toenhance the crystallinity of the products and to prepare smaller NPs. As reported in[130], the microwave heating of a TiO2 gel applied instead of the conventional HTTresults in a decrease of the grain size from 30–40 to 15–20 nm.

The growth of TiO2 NPs during the thermal treatment can be retarded by theintroduction of compounds capable of the crystallization simultaneously with titania,for example, SiO2. The nanocrystalline TiO2/SiO2 powders can be formed by thesimultaneous hydrolysis of titanium tetra-butoxide (TBT) and tetraethylorthosilicate

Fig. 5.8 a Formation of three typical titania phases from the individual TiO6 octahedra. Reprintedwith permissions from Ref. [125]. Copyright (2008) Elsevier. b The average size of TiO2

nanocrystals as a function of the HTT temperature (plotted using the data reported in [121])

254 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

or produced by mixing separately prepared TiO2 and SiO2 sols [131, 132]. The silicapresent in such composites inhibits the growth of TiO2 NPs and suppresses thetransformation of anatase into rutile during the annealing [131, 133–136]. As thesilica fraction in TiO2/SiO2 powders is elevated to 0.3 the average size of anatasecrystals decreases from 17 to 4 nm, while the specific surface area grows from 32 to240 m2/g (Fig. 5.9a) [131].

Nanocrystalline TiO2/SiO2 powders can also be produced by the hydrolysis of Ti(IV) alkoxides on the surface of 150–160 nm silica nanobeads. The final size oftitania NPs in such heterostructures depends on the annealing temperature growingfrom 4–5 to 11–12 nm with an increase in the calculation T from 500–600 to1000 °C [137] (Fig. 5.9b).

The growth of TiO2 NPs can also be restricted by performing the Ti(IV)hydrolysis in the presence of surfactants and bulky organic molecules. For example,nanocrystalline titania can be prepared by the hydrolysis of TBT in ethanol in thepresence of dodecylamine at 40–80 °C [138]. The amine is then removed duringthe calcination at 350–500 °C.

The introduction of surfactants allows also to influence the shape of the TiO2

nanocrystals. The hydrolysis of TiCl4 or TBT in the presence of sodium dodecylbenzyl sulfonate followed by the calcination yields elongated ellipsoid crystals witha length of up to 500 nm, while by using sodium dodecyl sulfonate or hydrox-ypropyl cellulose TiO2 nanocubes and NRs, respectively, can be synthesized [139].

Nanocrystalline titania is often produced with the triblock copolymers such asPluronic 123 [140–142]. The polymer composition can be described as (EO)20–(PO)70–(EO)20, where EO and PO are ethylene oxide (–CH2CH2O–) and propyleneoxide (–CH2(CH3)2CHO–) fragments. The method yields rutile NRs with adiameter of 10–20 nm and a length of 100–200 nm with the surface-anchored 5–15-nm anatase NPs [143].

In the case of nanocrystalline titania produced by the oxidative TiCl3 hydrolysisin ethanol in the presence of PEGs the size and phase composition of

Fig. 5.9 Average size of TiO2 nanocrystallites in the TiO2/SiO2 heterostructures (a), curve 1;(b) and the surface are of TiO2/SiO2 composite ((a), curve 2) as functions of the silica fraction(a) and the annealing temperature (b). Plotted using the data reported in [131] (a) and [137] (b)

5.2 Nanocrystalline Powdered Semiconductors 255

nano-dispersed products depend on the molecular mass M of the polymer. In arange of M = 200–4000 g/mole the synthesis yields anatase crystals with a size of35–75 nm (Fig. 5.10a), while at a higher M brookite nanocrystals as large as 70–80 nm can be observed [144].

The hydrolysis of Ti(IV) alkoxides in anhydrous alcohols occurs much slowerthan in water allowing to achieve a high crystallinity of the products without asolvothermal treatment. For example, the sol-gel TTIP transformation in ethanol/2-propanol followed by the gel drying at 70 °C and the xerogel annealing at 200–500 °C resulted in 7–12 nm titania crystals with a preferential (90%) anatase lattice[145].

The size of titania NPs can be controlled when Ti(IV) compounds are subjectedto the hydrolysis in micellar solutions. Such approach was applied to produce15-nm anatase crystals in water droplets solubilized by hexadecyl trimethylammonium bromide in cyclohexanol [146]. The simultaneous hydrolysis of TTIPand ammonium vanadyl in a micellar “water/Triton X-100/heptane” system fol-lowed by the thermal treatment produced 8–13 nm crystals of mixedvanadium-titanium oxide [147].

Highly dispersed titania powders are frequently produced by the pyrolyticmethods. For example, anatase nanocrystals can be synthesized by burning TiCl4 ina hydrogen/oxygen mixture [148]. By varying the TiCl4 feed rate the particle sizecan be tuned in a range of 15–30 nm, while by adjusting the rate of the gas mixture,the oxygen content and the flame temperature in a range of 1400–1700 °C one canvary the anatase fraction from 40 to 80 wt.%.

Nanocrystalline titania is often prepared by using self-igniting mixtures thatinclude a Ti(IV) precursor, oxidant, and a substance-fuel. The first two componentsare often present in a single compound—titanyl nitrate TiO(NO3)2, while the fuelrole is typically delegated to various amino-acids and polyalcohols. Such mixturesignite when heated to 300–350 °C and burn steadily with the formation of highly

Fig. 5.10 a Titania nanocrystal size as a function of the molecular mass (values on the graph) ofPEG present at the oxidative TiCl3 hydrolysis; b ZnO NP size as a function of the HTTtemperature. Plotted using the data reported in [144] (a) and [158] (b)

256 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

dispersed TiO2. A popular combination for the self-igniting mixtures is titanyl nitrate—glycine [149–153] that burns as described by the following brutto-equation: 9TiO(NO3)2 + 10H2NCH2COOH ! 9TiO2# + 14N2 + 20CO2 + 25H2O.

In the recent years, the synthesis of titania nanofibers and nanowires (NWs) hascome into the focus as such materials find broad applications for the preparation ofphotoactive paper, tissues, and reinforced polymers. The TiO2 NWs are typicallyprepared using PVP fibers [154] or layered protonated titanates [155] as templates.The titania NRs and NWs can also be formed by the electrophoretic deposition ofTiO2 NPs into the anodized alumina pores followed by the calcination and finaldissolution of the Al2O3 matrix [156].

Similarly to titania, nanocrystalline ZnO is most often produced using thehydrothermal syntheses. For example, the HTT of zinc acetate solution in ethanol at90 °C for 12 h produces elongated ZnO (40 � 80 nm) crystals [157]. As the HTTtemperature is elevated from 150 to 500 °C the average size of ZnO nanocrystalsgrows from 10 to 90 nm (Fig. 5.10b) [158].

The HTT of aqueous solutions containing zinc acetate and histidine at 150 °Cyields a variety of products depending on the reactants concentration—nanoprisms,hollow microspheres, NR aggregates, etc. [159].

Nanocrystalline ZnO was synthesized by burning a solution containing glycineand Zn(OH)2 at 1500–1800 °C [160], as well as by the thermal decomposition ofzinc hydroxycarbonate [161] and Zn[N(SiCH3)2]2 [162]. The mechanochemicaltreatment of ZnCl2 and Na2CO3 diluted with NaCl yields nanocrystalline ZnO witha grain size of 30–60 nm [163].

The zinc oxide NRs/NWs on substrates are typically produced by a two-stage“seeding” method [164]. On the first stage, a layer of an aqueous solution of zincnitrate and hexamethylenetetramine is deposited onto a rotating plate and annealedat around 200 °C. The ZnO NPs produced by this method serve as nuclei for theformation of ZnO NRs during the HTT of the substrate plate in the same solution at95 °C for 10–30 h. The ZnO NRs on substrates can also be conveniently producedby the thermal evaporation/redeposition (Fig. 5.11a) [165].

Fig. 5.11 Electron microscopic images of ZnO NRs on Zn foil (a), macroporous WO3 (b), BiOImicrospheres (c). Reprinted with permissions from Refs. [165] (a), [166] (b), and [170] (c).Copyright (2008) American Chemical Society (a, c) and The Royal Society of Chemistry (b)

5.2 Nanocrystalline Powdered Semiconductors 257

The synthetic methods described for TiO2 and ZnO can be extended to otheroxide and metallate semiconductors. A set of examples is provided below illus-trating a general character of the methods of synthesis of thermally stablephoto-active metal oxides.

The calcination of closely packed polystyrene microspheres soaked with a(NH4)6H2W12O40 solution was used to synthesize macroporous nanocrystallineWO3 (Fig. 5.11b) [166]. The WO3 NRs with a diameter of *8 nm can be formedby the decomposition of ammonium metatungstate in the pores of SBA-15 zeolite[167]. The WO3 NW arrays formed by a solvothermal synthesis on the conductiveglass can be used as the visible-light-sensitive photoanodes of solar cells [168].

The dehydration of Bi(OH)3 under the ultrasound irradiation in the presence ofPVP yields nanocrystalline Bi2O3 with a grain size of 40–100 nm [169]. The HTTof aqueous solution containing bismuth(III) nitrate, sodium halogenide (NaCl,NaBr, NaI) and ethylene glycol results in platelet-shaped BiOX (X = Cl, Br, I)nanocrystals [170, 171]. The HTT of BiOI nanoplates in aqueous ammonia solu-tions induces in their agglomeration into microspheres (Fig. 5.11c) with a muchhigher photoactivity as compared to original nanoplates [172].

Metal titanates and zirconates are typically produced by the co-hydrolysis of Ti(IV) or Zr(IV) with the corresponding metal salts followed by the HTT of gels orprecipitates. This method was applied for the preparation of nanocrystalline SrTiO3

with a grain size of 8–10 nm [173]. Of the three syntheses of nanocrystallinestrontium titanate—the co-hydrolysis of Ti(IV) and Sr(II) compounds (a final NPsize of 30 nm), a solid-state high-temperature reaction (140 nm), and themechanochemical treatment of a precursor mixture (30 nm), it is the co-hydrolysisthat yielded the most efficient photocatalysts of the H2 evolution fromwater/methanol mixtures [174].

Nanocrystalline potassium niobate K4Nb6O17 was produced by the HTT ofzirconium oxide in highly basic solutions (1.0 M KOH) at 150–200 °C (Fig. 5.12)[175–177]. To increase the photo-activity of the layered material it was intercalatedwith tetrabutyl ammonium hydroxide that expands the interlayer galleries makingthem more accessible for reactants [177].

Metal tungstates are typically produced from tungstic acid or related alkali metalsalts, often introduced as the polynuclear compounds. For example, the interactionbetween Bi2O3 and (NH4)2O � 12WO3 � 5H2O in the presence of diethylenetri-aminepentaacetic acid yields nanocrystalline Bi2WO6 powders [178]. Sometimes,this method is modified by the microwave [179, 180] or solvothermal treatment [180]to produce more active photocatalysts. The HTT of solutions containing Bi(NO3)3and sodium tungstate results in plate-shaped Bi2WO6 nanocrystals [181, 182]. Thenanoplate thickness depends on the HTT temperature and duration (Fig. 5.13a) andincreases from 20 to 100 nm as the HTT is prolonged from 4 to 24 h.

The HTT of a solution of Na2WO6, Zn(NO3)2 and cetyl trimethyl ammoniumbromide in a concentration high enough for the formation of cylindric micelles, at120–140 °C results in rod-shaped Zn2WO4 nanocrystals [183]. At a higher HTT

258 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

temperature the synthesis yields conventional Zn2WO4 nanocrystals showing noshape anisotropy. The size of such nanocrystals, similarly to Bi2WO4, can be tunedby varying the HTT duration (Fig. 5.13b).

Nanocrystalline vanadates can be produced using the whole assortment ofsynthetic approaches discussed above for oxides and other metallates. For example,plate-shaped BiVO4 nanocrystals with a thickness of 10–40 nm can be produced bythe HTT of solutions containing ammonium vanadate, bismuth nitrate, and astructure-directing agent—sodium dodecyl benzyl sulfonate [184]. The microwaveheating of a mixture of cerium nitrate and NH4VO4 in the presence of PEG-600yields nanocrystalline cerium orthovanadate with a particle size of 15–20 nm [185].

Fig. 5.12 TEM images of niobate nanosheets exfoliated by using propylamine(PA) hydrochloride (a) and niobate nanoscrolls (b) produced with tetrabuthyl ammonium(TBA) hydroxide; (c) scheme of K4Nb6O17 niobate nanosheets and nanoscrolls synthesis.Reprinted with permissions from Ref. [177]. Copyright (2008) Elsevier

5.2 Nanocrystalline Powdered Semiconductors 259

5.3 Nanocrystalline Films of Photo-Active Semiconductors

One of the most popular methods of the preparation of porous nanocrystallinesemiconductor films consists in a slow (with a rate of 1–2 mm per min) extractionof a substrate (typically a glass plate, a conductive glass plate—ITO, FTO, a metalfoil of polymer fibers) from a colloidal solution of semiconductor NPs that containssimultaneously pore-forming polymeric additives and other functional components,in particular those increasing the solution viscosity (Fig. 5.14a). The method istypically referred to as the “dip coating” [3].

Fig. 5.13 Thickness of Bi2WO4 nanoplates (a) and size of Zn2WO4 nanocrystals (b) as a functionof the HTT duration (plotted using the data reported in [182] (a) and [183] (b)

Fig. 5.14 a Synthesis of nanocrystalline TiO2 on glass. b Dependence of the thickness andsurface area of the titania film on the dip coating repetition number (plotted using the data reportedin [195])

260 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

The film thickness depends on the number of dip coating repetition, while theparameters of porous structure can be affected by the type and size of pore-formingadditive and the thermal treatment conditions. The drying and calcination ofdip-coated films result in nanoporous coatings formed by loosely aggregatedsemiconductor NPs. For example, titania films were prepared from TTIP or TiCl4 inthe presence of Triton X-100 [186–190], Pluronic 123 [191] and F127 [192],non-ionogenic Tween 20 [193], methyl cellulose [194], terpineol with hydrox-ypropyl cellulose [195], PVA [196], and PEG [197]. In the latter case, a lineardependence between the dip coating repetition number and the film thickness wasobserved—the latter increased from 0.9 to 5.4 lm as the repetition number waselevated from 5 to 30 [195] (Fig. 5.14b).

The dip coating can be performed from a titania sol synthesized in an invertedmicellar solution «water/Triton X-100/hexane» [128], a suspension of TiO2 NPsproduced by the HTT and dispersed in water by the ultrasound treatment [197], andSnO2 sols synthesized via the SnCl4 hydrolysis in 2-propanol [198]. Relatively thick(*10 lm) and optically transparent titania films can be produced from TiO2 solsmixed with highly viscous PEG CarbowaxTM [199]. The films of TiO2 [200–202],WO3 [203] and Bi2MoO6 [204] can be dip-coated onto wafers and optical fibers.

Negatively charged subnanometer titania sheets can be produced via the exfo-liation of alkali metal titanates assisted by the intercalation of tetrabutyl ammoniumcations. In the presence of a cationic surfactant—octadecyl ammonium chloridethese sheets accumulate on the solution/air interface and can be dip-coated on glassproducing ultrathin (0.75 nm) TiO2 � nH2O films with a lateral size of up to 1 lm[205].

Another popular method of the formation of nanocrystalline films from colloidalsemiconductors was christened “spin coating”. In this approach, the colloidalsolution/suspension is deposited onto a rapidly rotating substrate (Fig. 5.15). Undera combination of the centrifugal and centripetal forces, the solution is distributedevenly on the substrate surface [3]. The film thickness depends mainly on therotation speed. In this way, photo-active films are frequently produced from thecommercial nanocrystalline TiO2 Evonik P25 [206–208] or titania sols with addedviscous agents [82, 209–212]. For example, photo-active porous nanocrystallineTiO2 can be synthesized using carbon nanospheres produced by the HTT of glucosesolutions as a pore-forming agent [213] as well as conventional PEGs [214].

Nanocrystalline titania can be formed as films in mild conditions by the liquidphase deposition. In a typical procedure, a glass plate is immersed into a solution,where the hydrolysis of TiF4 [215] or TiF6

2− [216, 217] occurs in the presence ofboric acid that binds fluoride ion in the form of BF4

−. Ordered films of TiO2 onquartz were produced by multiple repetitions of the successive adsorption of neg-atively charged titania NPs and positively charged polydiallyl dimethyl ammoniumcations [218].

Porous films of titania form as a result of the NP deposition in an externalelectric field. For example, the electric field applied to a suspension of TiO2

(Evonik P25) with PVA results in the opening of channels between the polymer

5.3 Nanocrystalline Films of Photo-Active Semiconductors 261

globules favoring to the formation of inner branched pores in the final annealedtitania films [219].

Photoactive nanocrystalline films can be deposited from the gas phase usingexclusively physical processes or their combinations with chemical reactions [3]. Inthe first group of methods, the growth of nanocrystalline films results from thesubstance condensation on the surface of a substrate in the form of NPs, while in thesecond case it occurs via the decomposition (oxidation) of a gaseous precursor andthe product deposition on the surface of a heated substrate.

Of special interest are methods of mild deposition allowing to form photo-activecoatings on thermally unstable substrates. For example, such films can be depositedvia the bombardment with TiO2 clusters accelerated by an electron beam [220]. Theclusters are produced distantly from the substrate at the titanium vapor oxidationwith O2 (Fig. 5.16a). Also, the films can be formed by the gas-phase “cold”deposition. This method was realized for titania aerosol particles caught andtransported by a nitrogen stream at a rate of 300–1200 m/s [221, 222].

Photoactive titania films were produced by the thermal decomposition of TiCl4[206, 207] or titanyl 2,4-penthadionate [223]. The consecutive pyrolysis of SnCl4and titanyl acetylacetonate yields mixed nanocrystalline SnO2/TiO2 with a thick-ness up to *800 nm [224].

Thin-film photoelectrodes for the water splitting were synthesized by the flameaerosol deposition [225]. The pyrolysis of aerosol produced from a solution of zincand indium chlorides with thiourea admixtures was used to form nanocrystalline

Fig. 5.15 Synthesis of nanocrystalline titania films on glass by spin coating

262 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

ZnIn2S4 films that were successfully applied as a photocatalyst of the waterreduction [226].

The atomic layer deposition is often used for films preparation on porous sub-strates [3, 227]. In this method, one of the precursors is injected into the reactionchamber, then the chamber is cleaned with a gas carrier, another precursor isinjected, and the chamber is flooded with the inert gas again (Fig. 5.16b). Thegrowth of a nanocrystalline film occurs due to alternating surface reactions in thestate of adsorption saturation, resulting in uniform and structurally perfect films.

Tungsten oxide [228] and iron oxide [229] films produced by the atomic layerdeposition were used as electrodes in the photoelectrochemical water-splitting cells.The nanocrystalline titania films formed by the magnetron sputtering on glass andmetal substrates can also be applied as photocatalysts of this process [230, 231].A post-synthesis etching of such films in HF solutions results in an enhancement ofthe light conversion efficiency due to an increase of the surface area and the densityof donor centers.

A combined gas-phase sputtering of TiO2 and Pt on different sides of titaniumfoils yielded photoelectrodes for the water splitting, where the oxidation of water toO2 occurs on the nanocrystalline TiO2 film, while the water reduction to H2 takesplace on platinum [232]. A similar method was applied to form nanocrystallinetitania films with a gradient O/Ti ratio increasing from 1.93 near the substratesurface to 2.00—on the outer film surface [233]. The presence of Ti(III) in suchfilms makes them an active photocatalyst of the hydrogen evolution fromwater/methanol mixtures under the Vis-illumination [233].

Fig. 5.16 Formation of TiO2 films on thermally unstable substrates via the bombardment withcharged (TiO2)n

+ clusters [220] (a) and via the “cold” deposition (b) [221, 222]

5.3 Nanocrystalline Films of Photo-Active Semiconductors 263

The laser pulse deposition allows forming nanocrystalline semiconductor filmsat relatively low temperatures. In particular, this method was applied to produceCaFe2O4 films showing photoelectrocatalytic properties in the hydrogen evolution[234].

The electrodeposition is broadly used to produce films on various conductivesubstrates. For example, electrodeposited hematite a-Fe2O3 films doped with Pt[235], Cr or Mo [236] were used for the water photoelectrolysis. NanocrystallineCu2O films can be deposited onto conductive glasses (ITO) from solutions con-taining copper sulfate and lactic acid [237]. The electrodeposition was also appliedto produce ZnIn2S4 films with a grain size of around 20 nm from solutions of zincand indium chlorides and Na2S2O3 [238].

An anodic treatment of mixed Al/Ti film in a solution of phosphoric and oxalicacids results in the formation of porous nanocrystalline Al2O3/TiO2 films ready forthe photocatalytic applications [239]. An electrochemical etching of silicon yieldsnanocrystalline Si films with a particle size of 1–4 nm that were successfully usedfor the photocatalytic CO2 reduction [240]. The electrochemical oxidation of asilver layer in chloride-containing electrolytes resulted in the nanocrystalline AgClfilms [241].

5.4 Mesoporous Photo-Active SemiconductorNanomaterials

The smaller are the semiconductor crystals the higher is their surface energy. Thenanocrystals are thermodynamically non-stable and tend to lower their surfaceenergy via the agglomeration. This process is typically unfavorable for the synthesisof nano-photocatalysts and is inhibited by introducing various stabilizers thatadsorb on the NP surface and lower the surface tension. However, the tendency ofsemiconductor NPs to agglomeration can be used for the preparation of a specialkind of porous photo-active materials. For this aim, special substances are added tothe reaction mixtures that do not inhibit the agglomeration completely but preventthe NPs from the conversion into a dense body during the annealing. As a result, thesynthesis yields framework structures formed by loosely aggregated semiconductornanocrystals with nanometer voids between the separate NPs—the mesoporousmaterials [242]. The pore size can be tailored by performing the synthesis in thepresence of surfactant micelles or polymer globules of a given size that, after beingdestructed during the calcination, leave the voids of a corresponding size. To form amesoporous structure both the size of semiconductor NPs and the pore size shouldbe in a nanometer range [242].

The mesoporous semiconductor materials reveal a set of features very favorablefor the photocatalysis and solar cell applications [243]. They have a high specificsurface area (100–300 m2/g) and a developed system of mesopores that can adsorbreactants in the inner pore volume and accumulate them till as long as the capillary

264 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

condensation occurs. The high pore curvature allows for the strong adsorption andlong retention of reactants in the pores allowing for a more efficient contact with thephotogenerated charge carriers as compared to the non-porous or macroporousmaterials. As the mesoporous materials are formed by the NP building blocks eachcontacting with many neighboring NPs, there exists a unique opportunity for themigration of the photogenerated charge carriers along the NP network decreasingthe probability of recombination losses. Finally, in the case of hollow microsphereswith mesoporous walls an effect of the multiple light scattering and reflection in theinner voids can emerge favoring to a higher light absorption.

As the annealing of mesoporous material precursors is obligatory in almost allcases for the elimination of pore-forming agents, the assortment of reportedmesoporous materials is confined mostly to thermally stable oxide semiconductors[242]. For the production of mesoporous metal chalcogenide materials, alternativemethods are typically proposed, such as the anchoring of the metal sulfide NPs onthe inner walls of zeolites and membranes followed with the host dissolution, theultrasound treatment of colloidal solutions, the photochemical methods [244], etc.

Mesoporous oxide photocatalysts. The mesoporous metal oxides are typicallyproduced by the hydrolysis of corresponding precursors in the presence of variousfunctional additives followed by the HTT, drying and final calcination. A broadvariety of surfactants and polymers were reported as pore-forming additives for thesynthesis of mesoporous TiO2 and related nano-heterostructures (TiO2/SiO2 [245],TiO2/ZrO2 [246], TiO2/VOx [247], TiO2/In2O3 [248], etc.) including Pluronicfamily (Fig. 5.17a) [246, 249–255], agarose (Fig. 5.17b) [256]), Triton X-100[257], Tween surfactants [250], PEGs [250, 258], polyatomic alcohols (glucose,sorbitol [247]), bulky organic molecules such as 4,4′-dioxydiphenylpropane [259],triethanolamine [260]), dodecyl- and tetradecyl amine [261], lauril amine [262],cetyl trimethyl ammonium bromide [245, 263–267], alkyl phosphates (C12–C18)[268], etc.

Fig. 5.17 TEM (a) and SEM (b, c) images of mesoporous titania produced using Pluronic 123(a) and agarose (b), CdxZn1−xS microspheres (c). Reprinted with permissions from Refs. [249] (a),[256] (b), and [282] (c). Copyright (2008) Elsevier (a) and American Chemical Society (b, c)

5.4 Mesoporous Photo-Active Semiconductor Nanomaterials 265

The textural characteristics of mesoporous titania can be varied by changing thenature/size of pore-forming agents, and the conditions of HTT and final calcination.For example, an average size of NPs forming the framework of mesoporous TiO2

produced with PEGs decreases from 17 to 13–14 nm with an increase in theaverage molecular mass of the polymer from 200 to 20,000 g/mole (Fig. 5.18a). Asthe calcination temperature is elevated from 350 to 650 °C the NP size increasesfrom 10 to 35–37 nm (Fig. 5.18b, curve 1), while the pore size grows from 7–8 to15–16 nm (Fig. 5.18b, curve 2) [258].

The mesoporous titania with a particle size of 18–20 nm and a surface area of70 m2/g was produced by the TTIP hydrolysis in the presence of HCl followed bythe gel aging at 80 °C during 12 h and the calcination at 450 °C [269]. The materialwith 2–20 nm pores was synthesized from a mixed TiO2/SiO2 oxide via theselective silica dissolution in HF solutions [270]. The TBT hydrolysis in thepresence of acetylacetone during the HTT results in 100–200-nm TiO2 NRs.Titania NP frameworks form with a NP size of 5–10 nm and a pore size of 7–12 nmcan be observed on the NR surface [271]. A synthesis in water/ethanol solutionsyields porous materials with a bimodal pore size distribution—2–8 nm for theintra-aggregate pores and 40–50 nm for the pores between aggregates of 5–10 nmTiO2 particles [272, 273].

The high-intensity ultrasound treatment during the TTIP hydrolysis acceleratesthe nuclei formation and favors to their condensation and agglomeration intomesoporous 100–200-nm microspheres composed of 10 nm titania crystals sepa-rated by 4 nm pores [274].

A mesoporous material composed of nitrogen-doped TiO2 nanocrystals wasproduced by the calcination of a precipitate forming in a reaction between NH4OHand a complex [TiO(C2O4)2]

2− ion [275]. The size of titania NPs and pores in themesoporous framework depend distinctly on the calcination temperature.

Fig. 5.18 Particle size (a and b, curve 1) and pore diameter (b, curve 2) of mesoporous TiO2

produced in the presence of PEGs of a different molecular weight (a) and at a different calcinationtemperature (b) (plotted using the data reported in [258])

266 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

Similar methods were used to produce other mesoporous oxides for the photo-catalytic and photoelectrochemical applications, for example, Ta2O5 and Nb2O5. Incontrast to titania, the porous structure of mesoporous tantalum and niobium oxidesis not stable at higher T and can collapse during the annealing. To preserve themesoporous framework these oxides are annealed in a two-stage regime—at alower T (350 °C) to eliminate the organic pore-forming agent and at a higher T(500 °C)—to allow for the oxide crystallization [276]. Alternatively, silica NPs canbe introduced into the mesoporous oxides enhancing their stability at elevated T.The SiO2 NPs can then be removed by the chemical etching [277]. Such methodsyield mesoporous materials with an oxide NP size of 8–12 nm and a pore diameterof 4–5 nm that revealed photocatalytic properties in the hydrogen evolution fromwater/alcohol mixtures [276–278].

The protonation of K4Nb6O17 niobate followed by the intercalation with tetra-butyl ammonium cations results in the exfoliation of the layered material intoultra-thin niobate sheets. The sheet deposition on MgO followed by the calcinationand acidic etching results in mesoporous Nb2O5 revealing photocatalytic propertiesin the water splitting [279].

Mesoporous InVO4 synthesized in the presence of cetyl trimethyl ammoniumbromide was applied as a visible-light-sensitive photocatalyst of the H2 evolutionfrom aqueous solutions of oxalic acid [260].

The annealing of Ni(OH)2 NPs deposited from ethanol onto a conducting glasswas used to produce mesoporous nickel oxide photoelectrodes [280].

Mesoporous metal chalcogenide nanophotocatalysts. Metal chalcogenidematerials are far less stable during the thermal treatment and, therefore, require mildmethods to be produced in the mesoporous form, such, for example, as the ultra-sound treatment. This method was successfully used for the preparation of meso-porous CdS with a pore size of 5–6 nm formed by loosely aggregated 4–6 nm NPs[281]. The sonochemical treatment of a hot (80 °C) aqueous solution containingcadmium and zinc acetates, thioacetamide, and sodium dodecyl sulfonate, yieldsmesoporous CdxZn1−xS formed by 3–5 nm particles associated into porous 60–100 nm spheres (Fig. 5.17c) [282].

The HTT of precursor solutions in methanol is a relatively mild route to me-soporous metal sulfide nano-materials, for example, CdIn2S4 nanotubes that exhibitphotocatalytic activity in the water reduction [105].

The thermal decomposition of a complex of zinc sulfide with ethylenediamineZnS(en)0.5 results in porous platelet-like ZnS and ZnO particles with a size of20 � 30 nm active as photocatalysts of the H2 evolution from aqueousNa2S/Na2SO3 solutions [283]. Porous ZnS microaggregates formed by 4–10 nmparticles were synthesized via the HTT of micellar solutions composed of water,cetyl trimethyl ammonium bromide, cyclohexanone and pentanol and containingzinc acetate and urea [284]. The microsphere size increases from 200 nm to 1.5 lmas the ratio of water to surfactant is elevated from 8 to 32.

5.4 Mesoporous Photo-Active Semiconductor Nanomaterials 267

5.5 Spatially Organized Nanocrystalline Photo-ActiveSemiconductors

Mesoporous hollow microspheres. Recently, a considerable interest was observedfor the photochemical properties of hollow metal-oxide mesoporous microspheres[243]. Such hollow spheres (HSs) can be produced using the so-called “sacrificial”pore-forming microsphere-templates (SiO2, organic polymeric globules, carbonmicrospheres, etc.). In a typical synthesis, the TTIP hydrolysis is performed on thesurface of polystyrene latex microparticles (Fig. 5.19a–c) [285]. The latex is formedby the styrene polymerization in a PVP solution in the presence of sodium chlorideand is composed of highly monodisperse spherical *200-nm particles. The cal-cination of hydrolytic products yields monodisperse titania HSs with a size of 250–260 nm [285]. In a similar manner TiO2/SiO2 [286] and ZnO HSs [287] can beproduced (Fig. 5.19d–f).

Fig. 5.19 (a, b, c) Polystyrene microbeads (a), microbeads coated with titania (b) and titania HSs(c); (d, e, f) ZnO HSs produced from the microbeads of a different size—450 nm (d), 700 nm (e),and 1100 nm (f). Reprinted with permissions from Refs. [285] (a–c) and [287] (d–f). Copyright(2008) Elsevier (a–c) and American Chemical Society (d–f)

268 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

The TBT hydrolysis in a sulfonated polystyrene latex with a globule size of200 nm followed by calcination results in TiO2 HSs with an inner void diameter of150 nm and the 15-nm-thick mesoporous walls [288]. A prolonged (around 72 h)HTT of spherical TiO2 NP aggregates converts them into HSs [289]. Similarmethods were used also to convert dense cubic agglomerates of ZnSn(OH)6 NPsinto hollow micro-cubes [282].

TiO2 HSs can be prepared by the HTT of (NH4)2TiF6 [290] or TBT in2-propanol in the presence of glucose [291, 292]. The role of a pore-forming agentis played by carbon microspheres formed on the first step of HTT. Titania HSs witha size of 70 nm and a wall thickness of 4–7 nm were synthesized by burning Ti(IV)acetylacetonate in a stream of oxygen [293]. TiO2 HSs with a size of up to 550 nmand pore size of 4–5 nm were reported to form in the presence of suspended NaClcrystals in ethanol [294].

The TiOSO4 hydrolysis in alcohols containing glycerol and diethyl ether allowsforming HSs without additional pore-forming templates [295]. Depending on thesynthesis conditions, this method can yield dense titania microspheres, HSs, as wellas more complex geometries, for example, HSs enclosing porous titaniamicroparticles in the inner voids.

The iron(III) oxide HSs were produced by the FeCl3 hydrolysis in ethyleneglycol in the presence of sodium dodecyl benzyl sulfonate under the microwaveheat treatment [296, 297]. The HSs are formed from Fe2O3 nanoplates appearing onthe first step of the synthesis.

An alternative method for the preparation of hollow metal oxide spheres is basedon the local crystallization induced by the presence of fluoride anions. The crys-tallization results in the Ostwald ripening of amorphous semiconductor particlesthat propagates from the microsphere center to the outer surface and results in theformation of inner voids. This effect was used for the preparation of Sn-dopedtitania HSs via the HTT of Ti(SO4)2 solutions with SnCl4 and NH4F [298].

Nanotubes. Titania and titanate NTs have good perspectives for the utilization invarious light-harvesting systems [243]. NTs are typically produced via the HTT oftitania powders in strongly alkaline solutions or by the electrochemical etching oftitania foils in fluoride or phosphate electrolytes [299–301].

The HTT of nanocrystalline TiO2 in alkaline media results in the formation oftitanate nanosheets (NSs) with similar characteristics irrespectively of the phasecomposition or size of the original titania crystals. When such NSs are treated withacids (HCl, HNO3 or H2SO4) the alkali metal ions are exchanged with protons andthe protonated form of titanates spontaneously rolls into the NTs (Fig. 5.20).

A broadly used method for the preparation of mesoporous TiO2 NTs consists inthe electrochemical etching (anodization) of metallic titanium [302–316] in solu-tions containing H2O2, NaF/HF, alkali phosphates, ethylene glycol and otheradditives. The etching results in the formation of closely packed arrays of poly-crystalline NTs with a length of 1–6 lm, an inner void diameter of 15–120 nm anda wall thickness of 10–20 nm.

The anodization of titanium foils with a surface treated preliminarily withammonium persulfate results in bilayer nanostructures [316]. An upper 100 nm

5.5 Spatially Organized Nanocrystalline Photo-Active Semiconductors 269

layer is formed by the anatase nanorods with a diameter of around 20 nm, while alower 250 nm layer is composed of the mesoporous NTs with the outer and innerdiameters of 120 and 80 nm, respectively.

Layered nanophotocatalysts. Layered semiconductor materials such as tita-nates, niobates, and tantalates of alkali metals have the inter-layer sub-nanometergalleries that can accommodate substrates of the photochemical and photocatalytictransformations thus resembling porous nanomaterials. Typically, such layeredphotocatalysts are prepared by the sintering of corresponding oxides with cesium,potassium or sodium carbonates. The produced metallates are then converted in aprotonated form by the ion exchange in acidic solutions (Fig. 5.21).

In some cases, the protonated metallates are additionally treated with tetrabuthylammonium salts that can intercalate into the interlayer galleries and expand themor, otherwise, disrupt the layer structure completely resulting in the exfoliation toultra-thin metallate NSs [317–322]. Such materials are broadly studied as photo-catalysts of the water splitting [318, 319, 321].

The ultra-thin metallate NSs can be used as building blocks for the preparationof more complex multi-component nanomaterials. For example, HCa2Nb3O10

perovskite NSs with a thickness of around 1 nm were combined with Pt and IrO2

NPs by a bridge molecule—3-aminopropyl trimethoxysilane producing a photo-catalyst of the total water splitting [322].

The intercalation of TiO2 NPs into the interlayer galleries of a partially exfoli-ated H0.67Ti1.83□0.17O4 � H2O titanate (□ is an anion vacancy) results in a 4–5-fold increase of photocatalytic activity of the composite layered material in thewater splitting [319]. The RhCl3 hydrolysis on the surface of potassium niobatenanoscrolls produced by the exfoliation of layered niobate with tetrabuthylammonium hydroxide results in tubular Rh(OH)3/K4Nb6O17 nanostructures with a

Fig. 5.20 a Synthesis of titanate NTs from TiO2 NPs (a). TEM images of titanate NTs (b). SEMimages of anodized titania NT arrays (cross-sectional view (c) and upper view (d)). Reprinted withpermissions from Ref. [303] (c, d). Copyright (2007) Elsevier

270 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

diameter of 30 nm and a wall thickness of 5 nm. The tubular nanostructures are100 nm long and decorated with sub-nanometer rhodium hydroxide NPs.

The alternating interaction of silica NPs with negatively charged potassiumniobate sheets and polydiallyl dimethyl ammonium cations resulted in porous“core/shell” structures that were used as active photocatalysts of the hydrogenevolution from water/methanol mixtures [323].

5.6 Nanocrystalline Photo-Active Semiconductorson Carriers

Porous nano-photocatalysts can be prepared by anchoring semiconductor NPs onthe surface of zeolites and molecular sieves, in the interlayer galleries of clays, etc.[243, 324]. Alternatively the semiconductor NPs can be formed directly on thesurface of a porous carrier that provides abundant sites for the nucleation andsimultaneously restricts the NP growth.

TiO2 NPs were formed in the pores of natural silicates (mordenite, bentonite,palygorskite, kaolinite, etc. [325–336]) and ceramics [337]. Porous titania-basednanomaterials were synthesized basing on various zeolites, in particular zeolite Y[338–340], b [340], MCM-41 [341–345], SBA-15 [329, 346, 347], ZSM-5

Fig. 5.21 a Preparation and modification of layered metallates (on the example of NaNb3O8); b,c TEM images of exfoliated calcium niobate sheets (b) and Ca2Nb3O10 sheets with photodepositedPt NPs used as a photocatalyst of the water splitting. Reprinted with permissions from Refs. [321](b) and [322] (c). Copyright (2007–2008) American Chemical Society

5.5 Spatially Organized Nanocrystalline Photo-Active Semiconductors 271

[340, 348–350], etc. In a typical synthesis, the zeolite is impregnated with asolution of Ti(IV) alkoxide and annealed.

The zeolite carriers were successfully applied also for the preparation of metalsulfide-based porous nanophotocatalysts. The synthesis proceeds via an ionexchange step when the alkali cations are exchanged with cadmium, indium, zincions. The following sulfidation of such zeolites yields metal-sulfide NPs with a sizesmaller than the pore size of the zeolite host [351, 352] allowing in future for thediffusion of reactants of the photocatalytic transformations. For example, thedeposition of 2.5–2.6 nm CdS [352] and In2S3 NPs [351] in the pores of atitanosilicate Ti-MCM-41 yields efficient photocatalysts for the water reduction.The water splitting can also be photocatalyzed by CdS NPs incorporated into thepores of zeolite Y [353–355], b [353], L [355], SBA-15 [355, 356], and ZSM-5[281, 353], aluminium and magnesium oxides [281], as well as into the interlayergalleries of potassium niobate [354].

Semiconductor NPs are often stabilized in the perfluorinated Nafion membranes.The membrane consists of polytetrafluoroethylene fragments alternated with lateralperfluorovinyl ether fragments with the terminal sulfonic groups. It contains reg-ularly separated 4–5 nm voids that have a hydrophilic character due to the presenceof sulfonic groups [357]. This feature, coupled with mechanical and chemicalstability, and a low light absorption in UV and visible spectral range make theNafion a perfect host for the formation of semiconductor NPs. TheNafion-stabilized NPs are photostable and accessible for typical substrates of thephotocatalytic and photoelectrochemical reactions [357]. For example, the Nafionmembrane impregnation with TTIP followed by the treatment with boiling waterresulted in 4 nm TiO2 NPs that were used as a photocatalyst of the CO2 reduction[358, 359]. CdS NPs can be formed in the Nafion by reacting CdCl2 and thioac-etamide [357]. In contrast to Na2S, thioacetamide can diffuse through the membraneproviding a much more uniform distribution of CdS NPs.

The H2 evolution photocatalysts were produced by the deposition of 2–5 nmCdS NPs onto the surface of polystyrene micro-globules [360]. For the samepurposes CdxZn1–xS NPs were attached to the Whatman paper [361].

Porous photocatalysts were produced by impregnating the mesoporous graphitewith a TiO2 sol [362]. The carrier can be produced by the styrene polymerization inthe presence of mono-disperse (30 nm) colloidal SiO2 NPs followed by the car-bonization in an Ar stream. The deposition of TTIP from the gas phase on thesurface of activated carbons with the following hydrolytic transformation yields 10–50-nm TiO2 crystals dispersed in the host material pores. WO3 NPs deposited onthe amorphous carbon were used as a visible-light-sensitive harvesting componentof the photoelectrochemical systems [363].

By impregnating porous alumina membranes with TiF4 titania NTs were pro-duced with an inner diameter of 5–100 nm, depending on the synthesis duration[364]. Titania dispersions can be also attached to fibers and tissues using silica solsas binding additives [364]. Also, the photo-active semiconductor NPs can adhere tovarious thermally stable materials, such as metals or ceramics [365, 366].

272 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

5.7 Doped Semiconductor Nano-Photocatalysts

As discussed in Chaps. 2 and 3, doping of the wide-bandgap semiconductor withmoieties that can be excited by the visible light is one of the most broadly usedways of spectral sensitization [51, 367–371]. The metal ion doping results in dis-crete states in the band gap that can be involved in the electron transitions.Additionally, the dopant atoms can serve as traps for the photogenenared chargecarriers helping to avoid the electron-hole recombination. Very often the dopantatoms can also inhibit to some extent the growth of metal oxide semiconductor NPsduring the sol-gel synthesis as well as to stabilize some low-temperature phases(anatase in the case of titania) during the annealing. Also, the doping typicallystrengthens the mesoporous frameworks preventing them from collapse at the highT calcination.

Doping with non-metals (N, C, S, P, B, etc.) in the metal oxides results in apartial substitution of the lattice oxygen. At that, the p-orbitals of the dopant layhigher (in terms of the energy) than the p-orbitals of oxygen in the valence band,thus narrowing the band gap and making the semiconductor photo-active under theillumination with the visible light.

Doping with metals. The most frequent method of doping metal oxides withmetal ions consists of the introduction of dopant salts into the reaction mixturesbefore the hydrolysis and sol-gel transformation of the precursors. Alternatively, thedoped metal oxides can be deposited in the pyrolytic processes fromdopant-containing solutions. The metal oxides can also be bombarded with ionbeams and subjected to the thermal or mechanochemical treatment in the presenceof a dopant.

The hydrolysis of Ti(IV) precursors in the presence of dopant salts serves as ageneral method for the synthesis of doped titania powders and films. Themetal-doped TiO2 is often produced from the self-igniting mixtures. For example,Pd-doped [372] and Sm-doped [373] nanocrystalline titania was synthesized byburning a mixture of titanyl nitrate, fuel (glycine [372] or ethylene glycol and citricacid [373]) and a dopant nitrate.

The collisions of accelerated ionized dopant beam with the surface of titaniaresult in the incorporation of dopant ions into the oxide lattice. This method wasapplied to modify TiO2 with vanadium [374, 375] and chromium [376], bothdopant penetrating 70–80 nm deep from the crystal surface into the titania crystalbulk without the formation of separate phases or surface compounds.

Nanodisperse titania is often doped with alkali earth ions (Mg, Ba) to reduce thecrystal size and to enhance the thermal stability of the anatase modification duringthe annealing [377, 378]. This method allows to decrease the titania crystal sizefrom 20 to 5–10 nm and to produce pure anatase nanomaterials.

Metal sulfide nano-photocatalysts can be doped with ions of metals formingmetal sulfides with a lower solubility. Such metals (for example, Bi, Sb, Ag, Cu)can substitute the main ions (Cd, Zn, In) in the near-surface layer of the sulfide

5.7 Doped Semiconductor Nano-Photocatalysts 273

crystals forming point defects, islands or separate NPs depending on the dopantconcentration [379].

Doping with non-metals. Two basic methods are used to synthesizenonmetal-doped oxide semiconductors: (i) by introducing corresponding additivesinto the reaction mixtures where the formation and aging of oxide NPs take place,and (ii) by a post-synthesis treatment of metal oxide NPs in a solution or gas streamcontaining dopant compounds.

The most popular doped metal oxide in the photocatalysis and related solar lightharvesting technologies is nitrogen-doped titania. TiO2:N samples were producedby introducing nitrogen-containing admixtures into the solutions where thehydrolysis of Ti(IV) of the oxidative hydrolysis of Ti(III) takes place [380, 381].The most frequent sources of nitrogen are NH4OH [204, 275, 382–389] andammonium salts [388], hydrazine [388–391], urea [380, 381, 392–397], alky-lamines [398–400], ethylenediamine [401], hexamethylenetetramine [402], bipyr-idyl [403]. The product of sol-gel transformation is typically subjected to the HTTand annealed at 350–550 °C. In some cases, mixtures of various N sources are usedto increase the total nitrogen amount in the final product. For example, by using amixture of NH4OH and N2H4 up to 0.45 wt.% nitrogen can be introduced into thetitania lattice while for the separately taken nitrogen sources this level is onlyaround 0.24 wt.% [389].

The nitrogen dopant can be introduced into the nanocrystalline powders and NTsof TiO2 [404] and TiO2/SiO2 composites [405] by the thermal treatment in theammonia stream [404–412]. The treatment results in the surface nitridation of themetal oxide but does not alter the crystal structure as a whole. The nitrogen dopanttypically shows an absorption “shoulder” at 400–550 nm depending on the Ncontent making titania sensitive to the visible light.

TiO2:N materials can be conveniently produced by the mechanochemicaltreatment of commercial nanocrystalline titania (Evonik P25) with hexam-ethylenetetramine followed by the annealing at 400 °C to decompose the amine[413]. Ammonium salts were also used as a nitrogen source during themechanochemical synthesis of TiO2:N [388].

The post-synthesis nitridation can be achieved by applying a high-voltageelectric discharge to the nanocrystalline titania in the nitrogen atmosphere [414].Photochemically active TiO2:N samples were reported to form as a result of thepartial oxidation of nanocrystalline TiN [388, 415]. The nitrogen doping wasobserved at the magnetron sputtering of nanocrystalline titania [416] and ZnO [417]films.

Similarly to nitrogen, carbon can be introduced during the sol-gel transforma-tions [418–421]. The carbon atoms can either substitute oxygen in the titania lattice[419] or remain in the interstitial positions [421], in both cases making the metaloxide sensitive to the visible light. The ultrasound-assisted electrochemical etchingof titanium foils in solutions of ethylene glycol and NH4F followed by theannealing yields TiO2 NTs doped with carbon [422]. Alternatively, TiO2 NTs canbe annealed in a mixture of argon and acetylene producing TiO2:C [423].

274 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

Nanopowders of TiO2:C were produced from the self-igniting mixtures such as acomposition of TiCl4, citric acid and ethylene glycol [424].

The hydrolytic and pyrolytic methods were applied to introduce sulfur dopants.For example, the sol-gel transformation of TBT in the presence of thiourea yieldsTiO2:S samples containing 1–3 wt.% sulfur and demonstrating photocatalyticproperties in the hydrogen production under the Vis-illumination [425]. The TBThydrolysis in the presence of NH4F can be used to synthesize TiO2:F samples [426,427]. Alternatively, TiO2:F was prepared by the HTT of acidic TiF4 solutions [428].Nanocrystalline TiO2:F films are typically formed by the annealing of titanium foilswith a surface layer of sodium fluoride solution [429] as well as by the HTT oftitanium foils in HF solutions [430, 431].

The hydrolysis of Ti(IV) tetraethoxide in a hot solution of H3PO4 [432] or HPO3

[433] combined with the sol-gel transformation and annealing results in TiO2:Pnano-powders sensitive to the visible light. The sol-gel conversion of TTIP in thepresence of a mixture of boric acid and B(CH3CH2O)3 leads to the samples of TiO2:B with a grain size of 10–15 nm [431, 434–436]. The anodization of titanium foilsin H3BO3 solutions yields boron-doped TiO2 NTs with a spectral response atk < 540 nm [435].

5.8 Bi- (Multi-) Component Photo-Active SemiconductorNanostructures

Binary semiconductor nanostructures with “concerted” positions of the conductionand valence bands allowing for the efficient interfacial charge transfer between thecomponents is one of the most potent ways of achieving enhanced spatial separa-tion of the photogenerated charge carriers and boosting of the quantum efficiency ofthe photocatalytic transformation.

Metal chalcogenide/metal oxide composites. Very popular are “metal oxide—metal chalcogenide” nanostructures that can be produced by forming chalcogenideNPs in the presence of colloidal or nanopowdered metal oxide component. Forexample, TiO2/CdS heterostructures were synthesized by impregnating commercialTiO2 (Evonik P25) with a Cd(II)-thiourea complex solution followed by the thermaltreatment in the nitrogen stream [437]. Similar nanostructures were obtained by thedeposition of MPA-capped CdS NPs on the TiO2 surface [438, 439].

TiO2/CdS and ZrO2/CdS composites can be easily produced by the consecutiveimpregnation of nanocrystalline TiO2 and ZrO2 films with solutions of cadmiumacetate and sodium sulfide [440]. In the same way, PbS, CdS, Ag2S, Sb2S3, andBi2S3 NPs were anchored to the surface of TiO2, SnO2, Nb2O5, Ta2O5 films [441].

TiO2/PbS nanostructures with 3-nm PbS NPs were formed on the surface ofEvonik P25 titania by the CBD from hot alkaline aqueous solutions of lead acetateand thiourea [442]. This method was also applied to produce CdS NPs onthe surface of anodized titania NTs [443], mesoporous TiO2 [444], titania NRs

5.7 Doped Semiconductor Nano-Photocatalysts 275

[127, 445, 446] and NWs [447], as well as on rod-like ZnO nanocrystals [448, 449].Cadmium sulfide NPs can be formed on the titanate NTs by exchanging alkali metalions on the NT surface with Cd2+ followed by the treatment in a hot thioureasolution [450, 451]. By purging H2S into a solution of bismuth iodide and MPA inthe presence of nanocrystalline TiO2 films TiO2/Bi2S3 nanoheterostructures wereproduced [452].

Mesoporous TiO2/CdS composites can be synthesized by the SILAR method[214, 453]. This method can be extended to the preparation of TiO2/CdxZn1−xS[454], TiO2/In2S3 [455], TiO2/PbS [456, 457], and TiO2/Cu2S [458].

Tubular TiO2/CdS nanostructures produced by the ultrasound-assisted elec-trodeposition of cadmium sulfide NPs (Fig. 5.22a, b) revealed a high photoelec-trochemical activity in the water splitting [459]. The photoelectrochemically activeternary ZnO/CdSe/CdS heterostructures can be formed by the successive depositionof CdSe and CdS NPs (Fig. 5.22c, d) [460]. In a similar way, ternaryZnO/ZnS/AgInS2 [461] and ZnO/CdS/PbS [462] film heterostructures were pro-duced as well as nanotubular TiO2/CdS/CdSe composites [463] and SnO2/CdS/CdSenanostructures [464].

The TTIP hydrolysis in the presence of 40-nm CdS crystals followed by thethermal treatment at 450 °C yields CdS/TiO2 heterostructures comprising 9–10-nmanatase NPs (Fig. 5.23) [465]. In a similar manner, TiO2 NPs were anchored to thesurface of CdS NWs [466, 467].

The 5-nm CdS and CdSe NPs produced in tetrahydrofuran in the presence oftributhyl phosphine and 2,3-dimercaptosuccinic acid can be attached covalently to

Fig. 5.22 Array of anodized TiO2 NTs (a) and TiO2/CdS heterostructure (b). c, d TernaryZnO/CdS/CdSe heterostructure. Reprinted with permission from Refs. [459] (a, b) and [460] (c,d). Copyright (2010, 2011) Elsevier

276 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

the titania surface via the mercapto-group. Mercaptocarboxylic acids HS–(CH2)n–COOH (n = 1–3) [468–471], as well as mercaptopropyl-trimethoxysilane [472]were also used as “bridges” to bind CdS and TiO2 NPs.

The photochemically active HNbWO6/(Pt, Cd0.8Zn0.2S) heterostructures weresynthesized by the successive intercalation of an ammonia complex of platinum anda mixture of cadmium and zinc acetate followed by the sulfidation [473]. Using asimilar approach, layered K4Ce2M10O30 (M = Ta, Nb) materials were decoratedwith CdS NPs [474].

The chalcogenide/oxide heterostructures can also be produced by theco-precipitation of both components or by mixing separately prepared NPs. Forexample, the simultaneous precipi–ta–tion of zinc sulfide from ZnSO4 and theTiOSO4 hydrolysis in the presence of thioacetamide was used to prepare porousZnS–TiO2 heterostructures with a specific surface area of up to 200 m2/g com-prising 15–20-nm ZnS NPs and 6–7 nm titania crystals [475]. The binary com-posites can be produced by mixing colloids of CdS and TiO2 [44, 476–479], ZnOand CdS [480], CdS and AgI [479].

An additional treatment of a mixture of cadmium sulfide and titania NPs withtitanium(IV) chloride results in a stronger electric contact between CdS and TiO2

NPs favoring to the spatial separation of the photogenerated charge carriers [481].TiO2/CdS nanostructures can be formed by the electrochemical sulfur reduction

in DMSO on the surface of titania NT arrays in the presence of cadmium salts[482]. In a similar tellurium reduction process TiO2/CdTe NTs were obtained [483].

The nanoheterostructures comprising cadmium selenide NPs are typically syn-thesized in colloidal selenide solutions synthesized in the high-boiling-pointorganic solvents [484] or in the aqueous solutions in a reaction between Na2SeSO3

and cadmium salts in the presence of nitrilotriacetic acid [472, 485–487].

Fig. 5.23 Rod-like CdS/TiO2 heterostructures as a photocatalyst of the hydrogen evolution.Reprinted with permissions from Ref. [466]. Copyright (2008) Elsevier

5.8 Bi- (Multi-) Component Photo-Active Semiconductor Nanostructures 277

Alternatively, selenide ions can be generated by reducing Se (IV)/Se (VI) withN2H4 [488]. Then, colloidal CdSe NPs can be coupled with colloidal TiO2 [489] orwith titania nanopowders [402] and mesoporous materials [485, 486, 490–495],TiO2 NTs and NWs [487, 496]. Similarly, ZnO/CdSe nanowhiskers were synthe-sized [497]. Such heterostructures are broadly used for the preparation of lightharvesting cell electrodes [485–487, 490–497].

Oxide nanocomposites. Binary oxide nanostructures are typically formed by theimpregnation of a major component with salt solutions followed by the annealing,or by the co-hydrolysis, thermal treatment of salt mixtures, pyrolysis, electrode-position, etc.

The photoactivity of nanocrystalline titania was increased by coupling it withZnO [498], Fe(OH)3 [499, 500], Ni(OH)2 [501], Cu(OH)2 [499], and MnOx [502].Such heterostructures can be produced by a sequence of the impregnation of EvonikP25 with metal salt solutions, salt hydrolysis, and annealing. The calcination ofnanopowdered TiO2 soaked with copper(II) salts yields TiO2/CuxO compositeswith x close to 1 after the treatment at 350 °C and increasing to 2 at 450 °C [503].The TiO2/CuO heterostructured photocatalysts of the hydrogen evolution wereprepared in a similar way from titania NTs [504].

ZnO–TiO2 nanostructures can be produced by the simultaneous hydrolysis ofTBT and zinc nitrate in the presence of sodium dodecyl benzyl sulfonate followedby the annealing [505] or, alternatively, by the HTT of solutions of titanium andzinc chlorides, urea [506] or ammonia [507].

By annealing the coprecipitated oxides at 500 °C the SnO2–ZnO and TiO2–WO3

composites were produced [508–510]. Tungsten oxide NPs affect the crystallizationof titania rendering the anatase modification of TiO2 stable up to 800 °C anddecreasing the average titania NP size from 7 nm (with no WO3 present) to 2.5 nmfor the composites containing 4 wt.% WO3.

Binary nanocrystalline TiO2/Cu2O films were produced by a two-step synthesis[511]. In the first step, a suspension TiO2 nanocrystals (Evonik P25) ultrasonicallydispersed in a Triton X-100 solution was deposited onto the conductive ITO glassplate, then a layer of copper(I) oxide NPs was electrodeposited as a second step.

The co-hydrolysis was applied to produce photo-active Fe2O3–TiO2 [512],SnO2–TiO2 [513, 514], and ZnO–TiO2–SiO2 nanostructures [515]. The HTT ofalkaline solutions of tin (IV) chloride and zinc acetate yields mesoporous SnO2–

ZnO composites [s1124].The TTIP hydrolysis in the presence of colloidal SnO2 NPs results in

“core/shell” SnO2/TiO2 composites [516]. In a similar method, Fe2O3/TiO2 with aniron oxide core [517] and TiO2/Fe2O3 with a titania core were synthesized [518,519], as well as InVO4/TiO2 [520], TiO2/ZrO2 [521], and Bi2O3/TiO2

heterostructures [522]. The HTT of potassium titanate results in K0.3Ti4O7.3(OH)1.7NWs with a length of 0.5–1.5 lm and 10–20-nm anatase crystals attached to theNW surface [523]. The anion exchange followed by the hydrolysis of adsorbed Fe(III) was applied to form HTiNb(Ta)O5/Fe2O3 heterostructures [524]. In a similarway, the photoactive TiO2/MnO2 composites were produced from porous MnO2

[525]. TiO2/SnO2 can be formed via the pyrolysis of tetramethyl tin in the presence

278 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

of TiO2 Evonik P25 [526] as well as by the pyrolytic decomposition of mixedSnCl4 and TiO[C5H7O2]2 solutions [527]. The pyrolysis of solutions containingammonia complexes of zinc and iron or (NH4)10W12O41 was used for the prepa-ration of Fe2O3–ZnO and WO3–ZnO heterostructures [528]. The electron beamsputtering was applied to form ordered WO3 NR arrays coated with a titanianano-layer [529].

The titania NT arrays were decorated with hematite NPs by the electrodeposition[530]. A relatively short contact time (1 h) between the nanotubes and Fe(III)solution results in a selective precipitation of a-Fe2O3 NPs on the NT tips.A prolonged contact (24 h) yields titania NTs densely covered with hematite NPsboth on the inner and outer surface (Fig. 5.24) [530].

Semiconductor heterostructures with carbon nanomaterials. The synthesis ofnanostructures containing various carbon forms was partly discussed earlier in thesection dedicated to the NP-on-a-carrier photocatalysts. However, the role of car-bons is not confined only to the inert support of photoactive semiconductor NPs.The carbon nanomaterials are reported to participate actively in the charge sepa-ration processes both in primary photoprocesses, by donating/accepting electrons,and in the secondary steps of the photocatalytic reactions as co-catalysts oradsorbents [531, 532]. The active carbons used in the photocatalytic systems are

Fig. 5.24 Titania NT arrays (a, b) and TiO2/Fe2O3 (c, d) nano–composites produced by theelectrodeposition during 1 h (c) and 24 h (d). Reprinted with permissions from Ref. [530].Copyright (2011) American Chemical Society

5.8 Bi- (Multi-) Component Photo-Active Semiconductor Nanostructures 279

carbon NTs, fullerenes, graphite, graphite oxide and graphene oxide as well asreduced graphene oxide (RGO).

The carbon NTs can be decorated with titania NPs by the Ti(IV) alkoxidehydrolysis [531–541] or by drying the NT suspensions with nanocrystalline TiO2

[542, 543]. In a similar way, the carbon NTs were modified by ZnO NPs [544]. Thenanocomposites of titania with carbon NRs or fullerene were produced by the TBThydrolysis in the presence of NTs or C60 that were partially pre-oxidized with m-chloroperbenzoic acid [545].

The microwave treatment of carbon NTs in solutions of zinc acetate andthioacetamide results in NT/ZnS heterostructures (Fig. 5.25a) [546]. Similar meth-ods were applied to synthesize NT/ZnSe [547] and NT/CdSe composites [548].

The methane decomposition at 600 °C on the surface of Fe(III)-doped TiO2

nanocrystals starts with the formation of amorphous carbon islands which thentransform into the carbon NTs with the edges attached to the titania crystals(Fig. 5.25b) [549].

Recently the composite semiconductor nano-photocatalysts covered with a layerof amorphous or graphitic carbon were intensely studied. The titania nanocom-posites with graphitic carbon can be produced by the thermal annealing of a mixtureof Evonik P25 with polyvinyl alcohol [550]. The carbon covering TiO2

nanocrystals prohibits the transformation of anatase into rutile at the temperature ofup to 800 °C.

The thermal decomposition of glucose adsorbed on the ZnO NR surface underthe microwave illumination in an inert atmosphere results in ZnO/C nanostructureswith the carbon forming a uniform and dense 3–4-nm shell over the core semi-conductor [551]. The HTT of TiO2/C nanocomposite produced by the glucosedecomposition allows enhancing the graphitization of the carbon layer [552].Similar structures based on 10–20-nm ZnO nanocrystals were also produced by theglucose carbonization and graphitization at 800 °C in the N2 stream [553]. Theultrasonic treatment of graphite suspensions with titania yields intercalated

Fig. 5.25 Composites of carbon NTs with ZnS (a) and TiO2 NPs (b). Reprinted with permissionsfrom Refs. [546] (a) and [549]. Copyright (2008,2009) American Chemical Society (a) andElsevier (b)

280 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

graphite/TiO2 composite that can be used as a light-harvesting component of thesolar cells [554].

In the case of composite TiO2/graphite films deposited from the gas phase, theintroduction of graphite was found to decrease the size of TiO2 NPs and to enhancetheir photoactivity [555]. Carbon/WO3 nanostructures produced by the gas-phasedeposition onto the carbon paper exhibited a photoelectrochemical activity underthe Vis-illumination [363].

An explosive growth of interest in the chemistry of graphene and its derivativeswas stimulated by the mechanical exfoliation of graphite reported in 2004 by Geimand Novoselov [556, 557]. The most popular methods of the production of gra-phene or, more exactly, partially reduced graphene oxide are based on the reductionof grapehene oxide (GO), that can be obtained by the exfoliation of graphite oxide,while the latter is synthesized by the oxidation of graphite [558, 559]. One of themain challenges is to prevent the RGO aggregation into graphitic structures duringthe GO reduction. This aim can be achieved by reducing GO in the presence ofsemiconductor NPs that can arrange along the basal plane of GO and RGO particlesinteracting with the functional groups on their surface and making impossible therestacking and restoration of the original graphitic layered structure. In the case ofphoto-active semiconductor NPs, such as CdS, CdSe, TiO2, ZnO, etc. the photo-chemical and catalytic properties of the corresponding composites with RGO aretypically expressed much strongly that those of the individual semiconductor NPs.The fact opens broad possibilities of the synthesis of new nano-catalysts andnano-photocatalysts [560, 561].

The RGO/semiconductor nanocomposites are typically produced by the HTT ofa nano-dispered semiconductor (or its precursors) with a suspension (colloidalsolution) of GO. Alternatively, the RGO-based composites can be synthesized bythe in situ photocatalytic reduction of GO over semiconductor NPs [562].

The nanostructures of C60 fullerene with semiconductor NPs are formed by thehydrolysis of metal alkoxides in the presence of suspended fullerene (TiO2/C60

[563]) or by mixing colloidal semiconductors and fullerenes (CdSe/C60 [564]).

5.9 Photo-Active Semiconductor/Metal Nanostructures

Photocatalytic transformations of many substrates, for example, the reduction ofwater and CO2, CO decomposition, alcohols dehydrogenation, etc., require addi-tional co-catalysts, typically metal NPs that accumulate electrons photogenerated inthe nanocrystalline semiconductor photocatalyst and accelerate secondary “dark”redox processes.

One of the most popular methods of the synthesis of semiconductor/metalnanostructures is based on the semiconductor impregnation with a metal salt followedby the chemical reduction with a suitable agent (NaBH4, hydrogen, hydrazine, etc.).This method was applied to producemesoporous TiO2/Ag [261, 565], TiO2/Au [565–567], TiO2/Cu [568], TiO2/Ni [569] composites as well as plate-like KTiNbO5/Au

5.8 Bi- (Multi-) Component Photo-Active Semiconductor Nanostructures 281

nanostructures (Fig. 5.26a) [570]. The photo-active composites with an “inverted”structure formed by a 10-nm titania layer on 15-nm silver nanocrystals were preparedby the TTIP hydrolysis on the surface of pre-synthesized Ag NPs (Fig. 5.26b) [571].

Nanostructured ITO/Au/TiO2 electrodes produced by the electrodeposition ofgold NPs with the following formation of a titania layer on their surface cangenerate photocurrent when illuminated both by UV and the visible light [572].

TiO2/Pt nanocomposites revealing photoactivity in many reactions can be syn-thesized by the impregnation of commercial TiO2 Evonik P25 with platinum saltsand the reduction with formaldehyde [573] or hydrogen [573–576]. Alternatively,metal NPs can be anchored to the semiconductor surface by the electrophoreticdeposition [577–579].

The reduction of silver ions on the surface of ZnO NRs yields photoactiveZnO/Ag nanostructures (Fig. 5.26c) [580]. The NR-shaped ZnO/Ag compositeswere produced by a two-step synthesis, when the silver NWs are first formed by thesolvothermal synthesis in PVP, and then the ZnO NRs were deposited on the metalNW surface by the Zn(NO3)2 hydrolysis in the presence of hexamethylenetetramine[581].

The visible-light-sensitive Si/Pt electrodes were produced by reducing H2PtCl6with silicon in HF solutions [582]. Nanostructured CdS/Au [583] and TiO2/Aufilms [584] can be formed by the deposition of positively charged metal NPs ontothe nanocrystalline semiconductors charged negatively by adsorbed mercaptoac-etate or citrate anions.

The ultrasound treatment of semiconductor suspensions in the presence of metalsalts is a relatively mild and convenient method for the preparation of varioussemiconductor/metal composites. This method was used to form TiO2/M (M = Pt,Au, Pd) [585] and TiO2/Au–Pd heterostructures [586]. The ultrasonic treatment ofEvonik P25 suspension in a solution of Au(S2O3)2

3− also yields TiO2/Au nanos-tructures with the metal NP size increasing from 1.8 to 3.0 nm as the precursorconcentration is elevated from 1 to 8 wt.% [587].

Fig. 5.26 TEM images of selected semiconductor/metal nanocomposites: the sheet-like KtiNbO5/Au composite (a), core/shell Ag/TiO2 (b) and rod-like ZnO/Ag heterostructure (c). Reprinted withpermissions from Refs. [570] (a), [571] (b), and [580] (c). Copyright (2007, 2011) Elsevier (a) andAmerican Chemical Society (b, c)

282 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

The incubation of a solution containing 4 nm Au NPs and Na2ZnO4 at 90–95 °Cresults in the formation of photoactive Au/ZnO nanostructures with a 2.0–2.5 nmzinc oxide shell [588].

TiO2/Pt heterostructures were prepared by themicrowave treatment of titania NWsin ethylene glycol in the presence of H2PtCl6 [589], as well as by the pyrolysis of amixture of TiO2 Evonik P25 with hexachloroplatinic acid [590]. The solvothermaltreatment of methanol solutions of a H2−2x[Pt(NH3)4]xTi4O9 � 0.25H2O complexproduced from Pt(NH3)4Cl2 and H2Ti4O9 � 0.25H2O yields fibrous TiO2/Ptnanocomposites [591].

When titania NTs are used as a carrier for Pt NPs the charge of metal precursoraffects strongly the size of Pt NPs. For example, the surface protons of titania NTscan be exchanged with the positively charged platinum-ammonia complexes. Thefollowing annealing yields highly dispersed Pt NPs with a size of around 2 nm[592], active as a co-catalyst of the photocatalytic water splitting. At the same time,the thermal treatment of titania NTs impregnated with H2PtCl6 results in muchlarger Pt crystals (20–50 nm) showing a low activity in this photoprocess [592].

The thermal treatment of silver compounds introduced during the synthesis oftitanium dioxide or into the porous TiO2 materials results in TiO2/Agheterostructures [207, 266, 593, 594]. The photoactive ZnO/Ag composites formafter the calcination of a product of the co-precipitation of Ag(I) and Zn(II)hydroxides [595], as well as by the HTT of solutions containing zinc acetate,AgNO3 and hexamethylenetetramine [596]. The thermal treatment was also appliedto synthesize photoactive TiO2/Cu nanostructures [597, 598].

The flame pyrolysis of precursors is a universal method for the synthesis ofsemiconductor/metal nanocomposites. It was used, in particular, to synthesize TiO2/Mnanostructures (M = Ag, Au, Au–Ag, Pt) [599]. The flame pyrolysis of a mixture ofTTIP and gold(III) dimethylacetylacetonate was used to prepare a TiO2/Au photo-catalyst of the water splitting [600].

The pulsed electrodeposition of metal NPs is also a convenient method for thedecoration of titania NT arrays [601] and mesoporous TiO2 films [602] with goldNPs.

Along with the chemical methods for the preparation of semiconductor/metalnanostructures the practice of photocatalysis broadly uses the in situphotochemical/photocatalytic reduction of metal compounds on the surface ofmicro- and nanocrystalline semiconductors. All the above discussed heterostruc-tures can be produced by the photocatalytic reduction and the assortment ofreported composites includes TiO2/Hg [58], TiO2/Cd [603, 604], TiO2/Zn, TiO2/Mn, TiO2/Tl [604], ZnO/Pt [605–607], ZnO/Au [608], ZnO/Ag [158, 605], ZnO/Cu[605, 609, 610], ZnO/Pd [611], etc. The addition of dye sensitizers allowsaccomplishing the photocatalytic deposition of metals under the Vis-illumination.In this way, TiO2/Pt, TiO2/Pd, TiO2/Au [s1375], TiO2/Ag [612–614], and ZnO/Ag[96] heterostructures were synthesized.

The semiconductor/metal heterostructures produced by the photodeposition arebroadly studied in the water splitting processes and other endothermic reactions.For example, the mesoporous TiO2/Ni and TiO2/Cu formed via the photocatalytic

5.9 Photo-Active Semiconductor/Metal Nanostructures 283

reduction of metal salts were then applied as photocatalysts of the hydrogen evo-lution from water/ethanol mixtures [615, 616].

The applicability area of the semiconductor/metal composites synthesized by thephotodeposition is not confined to the photocatalysis. For example, spectralparameters of the surface plasmon resonance bands of silver NPs photodepositedonto the surface of nanocrystalline titania films depend strongly on the dielectricproperties of the solvent. The fact allows applying the TiO2/Ag heterostructures assolvatochromic sensors for the detection of alcohols, urea, glucose and other sub-strates [617–620]. The nanostructures formed by the photodeposition of Ag NPs ontitania nanotubes [621] and the nanocrystalline TiO2 films [617, 618] revealedphotochromic properties—the films become colorless under the Vis-illuminationand restore the color under the UV illumination. Local fluctuations of the magneticfield created by the Ag NPs deposited photocatalytically on the surface ofnanocrystalline titania result in a giant enhancement of the photoluminescence ofTiO2/Ag heterostructures [622].

References

1. Henglein A (1989) Small-particle research: Physicochemical properties of extremely smallcolloidal metal and semiconductor particles. Chem Rev 89:1861–1873. doi:10.1021/cr00098a010

2. Eychmüller A (2000) Structure and photophysics of semiconductor nanocrystals. J PhysChem B 104:6514–6528. doi:10.1021/jp9943676

3. Burda C, Chen X, Narayanan R, El-Sayed MA (2005) Chemistry and properties ofnanocrystals of different shapes. Chem Rev 105:1025–1102. doi:10.1021/cr030063a

4. Fujiwara H, Hosokawa H, Murakoshi K et al (1997) Effect of surface structures onphotocatalytic CO2 reduction using quantized CdS nanocrystallites. J Phys Chem B101:8270–8278. doi:10.1021/jp971621q

5. Stroyuk OL, Granchak VM, Korzhak AV, Sya Kuchmii (2004) Photoinitiation ofbuthylmethacrylate polymerization by colloidal semiconductor nanoparticles. J PhotochemPhotobiol, A 162:339–351. doi:10.1016/S1010-6030(03)00371-X

6. Shiragami T, Ankyu H, Fukami S et al (1992) Semiconductor photocatalysis: visible lightinduced photoreduction of aromatic ketones and electro-deficient alkenes catalyzed byquantised cadmium sulfide. J Chem Soc, Faraday Trans 88:1055–1061. doi:10.1039/FT9928801055

7. Stroyuk OL, Shvalagin VV, Raevskaya AE et al (2003) Photocatalysis of Cd2+ ionsreduction in 2-propanol by CdS nanoparticles. Theoret Exp Chem 39:331–336

8. Hoffman AJ, Mills G, Yee H, Hoffmann MR (1992) Q-sized CdS: synthesis, characteri-zation, and efficiency of photoinitiation of polymerization of several vinylic monomers.J Phys Chem 96:5546–5552. doi:10.1021/j100192a067

9. Shiragami T, Pac C, Yanagida S (1990) Visible-light-induced two-electron-transferphotoreductions on CdS: Effect of morphology. J Phys Chem 94:504–506. doi:10.1021/j100365a002

10. Kanemoto M, Hosokawa H, Wada Y et al (1996) Semiconductor photocatalysis. Part 20.Role of surface in the photoreduction of carbon dioxide catalyzed by colloidal ZnSnanocrystallites in organic solvent. J Chem Soc, Faraday Trans 92:2401–2411. doi:10.1039/FT9969202401

284 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

11. Kanemoto M, Shiragami T, Pac C, Yanagida S (1992) Semiconductor photocatalysis.Effective photoreduction of carbon dioxide catalyzed by ZnS quantum crystallites with lowdensity of surface defects. J Phys Chem 96:3521–3526. doi:10.1021/j100187a062

12. Korgel BA, Monbouquette HG (1997) Quantum confinement effects enable photocatalyzednitrate reduction at neutral pH using CdS nanocrystals. J Phys Chem B 101:5010–5017.doi:10.1021/jp971016n

13. Zhang JZ, O’Neil RH, Roberti TW (1994) Femtosecond studies of photoinduced electrondynamics at the liquid-solid interface of aqueous CdS colloids. J Phys Chem 98:3859–3864.doi:10.1021/j100065a049

14. Datta A, Priyam A, Bhattacharyya SN et al (2008) Temperature tunability of size in CdSnanoparticles and size dependent photocatalytic degradation of nitroaromatics. J ColloidInterface Sci 322:128–135. doi:10.1016/j.jcis.2008.02.052

15. Fisher CH, Henglein A (1989) Photochemistry of colloidal semiconductors. 31. Preparationand photolysis of CdS sols in organic solvents. J Phys Chem 93:5578–5581. doi:10.1021/j100351a049

16. Weib H, Fernandez A, Kisch H (2001) Elektronische Halbleiter-träger-wechselwirkung –

ein neuer Effekt in der Halbleiterphotokatalyse. Angew Chem 113:3942–394517. Nenadović MT, Čomor MI, Vasić V, Mićić OI (1990) Transient bleaching of small PbS

colloids. Influence of surface properties. J Phys Chem 94:6390–6396. doi:10.1021/j100379a044

18. Nanda J, Sapra S, Sarma DD et al (2000) Size-selected zinc sulfide nanocrystallites:synthesis, structure, and optical studies. Chem Mater 12:1018–1024. doi:10.1021/cm990583f

19. Riley DJ, Waggett JP, Wijayantha KGU (2004) Colloidal bismuth sulfide nanoparticles: aphotoelectrochemical study of the relationship between bandgap and particle size. J MaterChem 14:704–708. doi:10.1039/B311517H

20. Haram SK, Kshirsagar A, Gujarathi YD et al (2011) Quantum confinement in CdTequantum dots: investigation through cyclic voltammetry supported by density functionaltheory (DFT). J Phys Chem C 115:6243–6249. doi:10.1021/jp111463f

21. Warrier M, Lo MKF, Monbouquette H, Garcia-Garibay MA (2004) Photocatalytic reductionof aromatic azides to amines using CdS and CdSe nanoparticles. Photochem Photobiol Sci3:859–863. doi:10.1039/B404268A

22. Kumar A, Mital S (2004) Synthesis and photophysics of 6-dimethylaminopurine-cappedQ-CdS nanoparticles—a study of its photocatalytic behavior. Inter J Photoenergy 6:61–68.doi:10.1155/S1110662X04000091

23. Torimoto T, Sakata T, Mori H, Yoneyama H (1994) Effect of surface charge of4-amino-thiophenol-modified PbS microcrystal photocatalysts on photoinduced chargetransfer. J Phys Chem 98:3036–3043. doi:10.1021/j100062a048

24. Nakashima T, Sakashita M, Nonoguchi Y, Kawai T (2007) Sensitized photopolymerizationof an ionic liquid-based monomer by using CdTe nanocrystals. Macromol 40:6540–6544.doi:10.1021/ma0707988

25. Raevskaya AE, Stroyuk OL, Sya Kuchmii (2004) Photocatalytic oxidation of hydrosulfideions by molecular oxygen over cadmium sulfide nanoparticles. J Nanopart Res 6:149–158.doi:10.1023/B:NANO.0000034719.30620.d3

26. Stroyuk OL, Sobran IV, Korzhak AV et al (2008) Photopolymerization of water-solubleacrylic monomers induced by colloidal CdS and CdxZn1−xS nanoparticles. Colloid PolymSci 286:489–498. doi:10.1007/s00396-007-1824-4

27. Matsumoto H, Uchida H, Matsunaga T et al (1994) Photoinduced reduction of viologens onsize-separated CdS nanocrystals. J Phys Chem 98:11549–11556. doi:10.1021/j100095a041

28. Nosaka Y, Miyama H, Terauchi M, Kobayashi T (1988) Photoinduced electron transfer fromcolloidal cadmium sulfide to methylviologen: a picosecond transient absorption study.J Phys Chem 92:255–256. doi:10.1021/j100313a003

References 285

29. Nosaka Y, Fox MA (1986) Effect of light intensity on the quantum yield of photoinducedelectron transfer from colloidal cadmium sulfide to methylviologen. J Phys Chem 90:6521–6522. doi:10.1021/j100282a021

30. Raevskaya AE, Korzhak AV, Stroyuk OL, Kuchmiy SY (2005) Spectral, optical andphotochemical properties of ZnS nanoparticles. Theoret Exp Chem 41:105–109

31. Raevskaya AE, Korzhak AV, Stroyuk OL, Kuchmiy SY (2005) Photocatalytic reduction of(II) with the participation of ZnS nanoparticles. Theoret Exp Chem 41:231–235

32. Raevskaya AE, Korzhak AV, Stroyuk OL, Kuchmiy SY (2005) Photocatalysis of theformation of ZnS/Au heterostructure via reduction of complex gold ions by ZnSnanoparticles. Theoret Exp Chem 41:359–364

33. Stroyuk OL, Raevskaya AE, Korzhak AV, Kuchmiy SY (2007) Zink sulfide nanoparticles:spectral properties and photocatalytic activity in metals reduction reactions. J Nanopart Res9:1027–1039. doi:10.1007/s11051-006-9183-5

34. Kryukov AI, Zinchuk NN, Korzhak AV et al (2004) Optical and catalytic properties of Ag2Snanoparticles. J Mol Catal A 221:209–221. doi:10.1016/j.molcata.2004.07.009

35. Meahcov L, Sandu I (2004) Colloidal CdS fluorescence quenching by MV2+ undercontinuous irradiation. J Fluorescence 14:181–185. doi:10.1023/B:JOFL.0000016289.30498.e9

36. Rossetti R, Brus LE (1986) Picosecond resonance Raman scattering study of methylviologenreduction on the surface of photoexcited colloidal CdS crystallites. J Phys Chem 90:558–560. doi:10.1021/j100276a014

37. Kanade KG, Baeg JO, Mulik UP et al (2006) Nano-CdS by polymer-inorganic solid-statereaction: visible light pristine photocatalyst for hydrogen generation. Mater Res Bull41:2219–2225. doi:10.1016/j.materresbull.2006.04.031

38. Kamat PV, Dimitrijević NM, Fessenden RW (1988) Photoelectrochemistry in particulatesystems. 7. Electron-transfer reactions of indium sulfide semiconductor colloids. J PhysChem 92:2324–2329. doi:10.1021/j100319a044

39. Henglein A (1982) Photochemistry of colloidal cadmium sulfide. 2. Effects of adsorbedmethyl viologen and of colloidal platinum. J Phys Chem 86:2291–2293. doi:10.1021/j100210a010

40. Igumenova TI, Vasil’tsova OV, Parmon VN (1996) Formation and photocatalytic propertiesof Q-sized nanoparticles of various transition metal sulphides in the inner cavities of lecithinvesicles modified with sodium dodecylsulphate. J Photochem Photobiol, A 94:205–213.doi:10.1016/1010-6030(95)04175-3

41. Raevskaya AE, Stroyuk OL, Kuchmiy SY (2006) Preparation of colloidal CdSe andCdS/CdSe nanoparticles from sodium selenosulfate in aqueous polymers solutions. J ColloidInterface Sci 302:133–141. doi:10.1016/j.jcis.2006.06.018

42. Bavykin DV, Savinov EN, Parmon VN (2000) Studies on the kinetics of interfacial electrontransfer sensitized by colloidal CdS. J Photochem Photobiol, A 130:57–61. doi:10.1016/S1010-6030(99)00196-3

43. Variano BF, Hwang DM, Sandroff CJ et al (1987) Quantum effects in anisotropicsemiconductor clusters: colloidal suspensions of Bi2S3 and Sb2S3. J Phys Chem 91:6455–6458. doi:10.1021/j100310a006

44. Sant PA, Kamat PV (2002) Interparticle electron transfer between size-quantized CdS andTiO2 semiconductor nanoclusters. Phys Chem Chem Phys 4:198–203. doi:10.1039/B107544F

45. Yagi I, Mikami K, Ebina K et al (2006) Size-dependent carrier dynamics in CdSnanoparticles by femtosecond visible-pump/IR-probe measurements. J Phys Chem B110:14192–14197. doi:10.1021/jp061073t

46. Hirai T, Nanba M, Komasawa I (2002) Dithiol-mediated immobilization of CdSnanoparticles from reverse micellar system onto Zn-doped silica particles and their highphotocatalytic activity. J Colloid Interface Sci 252:89–92. doi:10.1006/jcis.2002.8430

47. Thurston TR, Wilcoxon JP (1999) Photooxidation of organic chemicals catalyzed bynanoscale MoS. J Phys Chem B 103:11–17. doi:10.1021/jp982337h

286 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

48. Wilcoxon JP, Samara GA (1995) Strong quantum-size effects in a layered semiconductor:MoS2 nanoclusters. Phys Rev B 51:7299–7302. doi:10.1103/PhysRevB.51.7299

49. Matsumoto H, Matsunaga T, Sakata T et al (1995) Size dependent fluorescence quenching ofCdS nanocrystals caused by TiO2 colloids as a potential-variable quencher. Langmuir11:4283–4287. doi:10.1021/la00011a019

50. Raevskaya AE, Stroyuk OL, Kuchmiy SY et al (2006) Growth and spectroscopiccharacterization of CdSe nanoparticles synthesized from CdCl2 and Na2SeSO3 in aqueousgelatine solutions. Colloids Surfaces A 290:304–309. doi:10.1016/j.colsurfa.2006.05.038

51. Chen X, Mao SS (2007) Titanium dioxide nanomaterials: synthesis, properties, modifica-tions, and applications. Chem Rev 107:2891–2959. doi:10.1021/cr0500535

52. Serpone N, Lawless D, Khairutdinov R (1995) Size effects on the photophysical propertiesof colloidal anatase TiO2 particles: size quantization or direct transitions in this indirectsemiconductor? J Phys Chem 99:16646–16654. doi:10.1021/j100045a026

53. Hykrdova L, Jirkovsky J, Mailhot G, Bolte M (2002) Fe (III) photoinduced and Q-TiO2

photocatalysed degradation of naphthalene: comparison of kinetics and proposal ofmechanism. J Photochem Photobiol, A 151:181–193. doi:10.1016/S1010-6030(02)00014-X

54. Ahn WY, Sheeley SA, Rajh T, Cropek DM (2007) Photocatalytic reduction of 4-nitrophenolwith arginine-modified titanium dioxide nanoparticles. Appl Catal B 74:103–110. doi:10.1016/j.apcatb.2007.01.016

55. Pelet S, Grätzel M, Moser JE (2003) Femtosecond dynamics of interfacial and intermolec-ular electron transfer at eosin-sensitized metal oxide nanoparticles. J Phys Chem B107:3215–3224. doi:10.1021/jp027358x

56. Huber R, Moser JE, Grätzel M, Wachtveitl J (2002) Real-time observation of photoinducedadiabatic electron transfer in strongly coupled dye/semiconductor colloidal systems with a6 fs time constant. J Phys Chem B 106:6494–6499. doi:10.1021/jp0155819

57. Huber R, Spörlein S, Moser JE et al (2000) The role of surface states in the ultrafastphotoinduced electron transfer from sensitizing dye molecules to semiconductor colloids.J Phys Chem B 104:8995–9003. doi:10.1021/jp9944381

58. Chen LX, Rajh T, Wang Z, Thurnauer MC (1997) XAFS studies of surface strutures of TiO2

nanoparticles and photocatalytic reduction of metal ions. J Phys Chem B 101:10688–10697.doi:10.1021/jp971930g

59. Dimitrijevic NM, Rajh T, Saponjic ZV et al (2004) Light-induced charge separation andredox chemistry at the surface of TiO2/host-guest hybrid nanoparticles. J Phys Chem B108:9105–9110. doi:10.1021/jp049028d

60. Dimitrijevic NM, Saponjic ZV, Bartels DM et al (2003) Revealing the nature of trappingsites in nanocrystalline titanium dioxide by selective surface modification. J Phys Chem B107:7368–7375. doi:10.1021/jp034064i

61. Bahnemann D, Henglein A, Lilie J, Spanhel L (1984) Flash photolysis observation of theabsorption spectra of trapped positive holes and electrons in colloidal TiO2. J Phys Chem88:709–711. doi:10.1021/j150648a018

62. Ban T, Kondoh S, Ohya T et al (2003) Photocatalysis of a transparent titanate aqueous solprepared from titanium tetraisopropoxide and tetramethylammonium hydroxide.J Photochem Photobiol, A 156:219–225. doi:10.1016/S1010-6030(02)00404-5

63. Serpone N, Sharma DK, Moser J, Grätzel M (1987) Reduction of acceptor relay species byconduction band electrons of colloidal titanium dioxide: light-induced charge separation inpicosecond time domain. Chem Phys Lett 136:47–51. doi:10.1016/0009-2614(87)87296-2

64. Moser J, Grätzel M (1983) Light-induced electron transfer in colloidal semiconductordispersions: single vs. dielectronic reduction of acceptors by conduction-band electrons.J Am Chem Soc 105:6547–6555. doi:10.1021/ja00360a003

65. Matylitsky VV, Lenz MO, Wachtveitl J (2006) Observation of pH-dependentback-electron-transfer dynamics in alizarin/TiO2 adsorbates: importance of trap states.J Phys Chem B 110:8372–8379. doi:10.1021/jp060588h

References 287

66. Walters KA, Gaal DA, Hupp JT (2002) Interfacial charge transfer and colloidalsemiconductor dye-sensitization: mechanism assessment via stark emission spectroscopy.J Phys Chem B 106:5139–5142. doi:10.1021/jp015540c

67. Ramakrishna G, Ghosh HN (2001) Emission from the charge transfer state of xanthenedye-sensitized TiO2 nanoparticles: a new approach to determining back electron transfer rateand verifying the Marcus inverted regime. J Phys Chem B 105:7000–7008. doi:10.1021/jp011291g

68. Kölle U, Moser J, Grätzel M (1985) Dynamics of interfacial charge-transfer reactions insemiconductor dispersions. Reduction of cobaltoceniumdicarboxylate in colloidal TiO2.Inorg Chem 24:2253–2258. doi:10.1021/ic00208a026

69. Müller BR, Majoni S, Meissner D, Memming R (2002) Photocatalytic oxidation of ethanolon micrometer and nanometer-sized semiconductor particles. J Photochem Photobiol, A151:253–265. doi:10.1016/S1010-6030(02)00010-2

70. Shchukin DG, Möhwald H (2005) Photosynthesis inside polyelectrolyte capsules: effect ofconfined media. Langmuir 21:5582–5587. doi:10.1021/la050429+

71. Zang L, Rodgers MAJ (2000) Diffusion-controlled charge transfer from excited Ru(bpy)32+

into nanosized TiO2 colloids with EDTA. J Phys Chem B 104:468–474. doi:10.1021/jp992636m

72. Fessenden RW, Kamat PV (1995) Rate constants for charge injection from excited sensitizerinto SnO2, ZnO, and TiO2 semiconductor nanocrystallites. J Phys Chem 99:12902–12906.doi:10.1021/j100034a032

73. Weng YX, Wang YQ, Asbury JB et al (2000) Back electron transfer from TiO2

nanoparticles to FeIII(CN)63−: origin of non-single-exponential and particle size independent

dynamics. J Phys Chem B 104:93–104. doi:10.1021/jp992522a74. Yang X, Tamai N (2001) How fast is interfacial hole transfer? In situ monitoring of carrier

dynamics in anatase TiO2 nanoparticles by femtosecond laser spectroscopy. Phys ChemChem Phys 3:3393–3398. doi:10.1039/B101721G

75. Fernandez A, Caballero A, Gonzalez-Elipe AR et al (1995) In situ EXAFS study of thephotocatalytic reduction and deposition of gold on colloidal titania. J Phys Chem 99:3303–3309. doi:10.1021/j100010a047

76. Rabani J, Yamashita K, Ushida K et al (1998) Fundamental reactions in illuminated titaniumdioxide nanocrystallite levels studied by pulsed laser. J Phys Chem B 102:1689–1695.doi:10.1021/jp973411j

77. Yoshihara T, Katoh R, Furube A et al (2004) Identification of reactive species inphotoexcited nanocrystalline TiO2 films by wide-wavelength-range (400–2500 nm) transientabsorption spectroscopy. J Phys Chem B 108:3817–3823. doi:10.1021/jp031305d

78. Hao YQ, Wang YF, Weng YX (2008) Particle-size-dependent hydrophilicity of TiO2

nanoparticles characterized by Marcus reorganization energy of interfacial charge recom-bination. J Phys Chem B 112:8995–9000. doi:10.1021/jp802532w

79. Martini I, Hodak JH, Hartland GV (1999) Dynamics of semiconductor-to-dye electrontransfer from anthracene dyes bound to different sized TiO2 particles. J Phys Chem B103:9104–9111. doi:10.1021/jp990302r

80. Ghosh HN (1999) Charge transfer emission in coumarin 343 sensitized TiO2 nanoparticle: adirect measurement of back electron transfer. J Phys Chem B 103:10382–10387. doi:10.1021/jp9918611

81. Kim J, Lee D (2007) Size-controlled interparticle charge transfer between TiO2 andquantized capacitors. J Am Chem Soc 129:7706–7707. doi:10.1021/ja071329o

82. Behar D, Rabani J (2001) Laser photolysis of TiO2 layers in the presence of aqueous iodide.J Phys Chem B 105:6324–6329. doi:10.1021/jp003176w

83. Skinner DE, Colombo DP, Cavaleri JJ, Bowman RM (1995) Femtosecond investigation ofelectron trapping in semiconductor nanoclusters. J Phys Chem 99:7853–7856. doi:10.1021/j100020a003

288 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

84. Grabner G, Quint RM (1991) Pulsed-laser-induced charge-transfer reactions in aqueousTiO2 colloids. A study of the dependence of transient formation on photon fluence.Langmuir 7:1091–1096. doi:10.1021/la00054a012

85. Yonezawa Y, Kometani N, Sakaue T, Yano A (2005) Photoreduction of silver ions in acolloidal titanium dioxide suspension. J Photochem Photobiol, A 171:1–8. doi:10.1016/j.jphotochem.2004.08.020

86. Zhang R, Gao L, Zhang Q (2004) Photodegradation of surfactants on the nanosized TiO2

prepared by hydrolysis of the alkoxide titanium. Chemosphere 54:405–411. doi:10.1016/S0045-6535(03)00588-5

87. Chae SY, Park MK, Le SK et al (2003) Preparation of size-controlled TiO2 nanoparticlesand derivation of optically transparent photocatalytic films. Chem Mater 15:3326–3331.doi:10.1021/cm030171d

88. Bahnemann DW, Kormann C, Hoffmann MR (1987) Preparation and characterization ofquantum size zinc oxide. A detailed spectroscopic study. J Phys Chem 91:3789–3798.doi:10.1021/j100298a015

89. Bahnemann DW (1993) Ultrasmall metal oxide particles: preparation, photophysicalcharacterization, and photocatalytic properties. Isr J Chem 33:115–136. doi:10.1002/ijch.199300017

90. Wang CY, Böttcher C, Bahnemann DW, Dohrmann JK (2003) A comparative study ofnanometer sized Fe (III)-doped TiO2 photocatalysts: synthesis, characterization and activity.J Mater Chem 13:2322–2329. doi:10.1039/B303716A

91. Kamat PV, Patrick B (1992) Photophysics and photochemistry of quantized ZnO colloids.J Phys Chem 96:6829–6834. doi:10.1021/j100195a055

92. Hoffman AJ, Yee H, Mills G, Hofmann MR (1992) Photoinitiated polymerization of methylmethacrylate using Q-sized ZnO colloids. J Phys Chem 96:5540–5546. doi:10.1021/j100192a066

93. Henglein A (1993) Physicochemical properties of small metal particles in solution“microelectrode” reactions, chemisorption, composite metal particles, and theatom-to-metal transition. J Phys Chem 97:5457–5471. doi:10.1021/j100123a004

94. Su S, Lu SX, Xu WG (2008) Photocatalytic degradation of reactive brilliant blue X-BR inaqueous solution using quantum-sized ZnO. Mater Res Bull 43:2172–2178. doi:10.1016/j.materresbull.2007.08.029

95. van Dijken A, Meulenkamp EA, Vanmaekelbergh D, Meijerink A (2000) The kinetics of theradiative and nonradiative processes in nanocrystalline ZnO particles upon photoexcitation.J Phys Chem B 104:1715–1723. doi:10.1021/jp993327z

96. Stroyuk OL, Shvalagin VV, Kuchmiy SY (2007) Photochemical synthesis of ZnO/Agnanocomposites. J Nanopart Res 9:427–440. doi:10.1007/s11051-006-9086-5

97. Stroyuk OL, Dzhagan VM, Shvalagin VV et al (2010) Size-dependent optical properties ofcolloidal ZnO nanoparticles charged by photoexcitation. J Phys Chem C 114:220–225.doi:10.1021/jp908879h

98. Panasiuk YV, Raevskaya OE, Stroyuk OL et al (2014) Colloidal ZnO nanocrystals indimethylsulfoxide: a new synthesis, optical, photo- and electroluminescent properties.Nanotechnology 25:075601. doi:10.1088/0957-4484/25/7/075601

99. Raevskaya AE, Panasiuk YV, Stroyuk OL et al (2014) Spectral and luminescent propertiesof ZnO/SiO2 core/shell nanoparticles with size-selected ZnO cores. RSC Adv 4:63393–63401. doi:10.1039/C4RA07959K

100. Bao N, Shen L, Takata T et al (2007) Facile Cd-thiourea complex thermolysis synthesis ofphase-controlled CdS nanocrystals for photocatalytic hydrogen production under visiblelight. J Phys Chem C 111:17527–17534. doi:10.1021/jp076566s

101. Jang JS, Joshi UA, Lee JS (2007) Solvothermal synthesis of CdS nanowires forphotocatalytic hydrogen and electricity production. J Phys Chem C 111:13280–13287.doi:10.1021/jp072683b

References 289

102. Xiong S, Xi B, Wang C et al (2007) Tunable synthesis of various wurtzite ZnS architecturalstructures and their photocatalytic properties. Adv Func Mater 17:2728–2738. doi:10.1002/adfm.200600891

103. Wang X, Wan F, Han K et al (2008) Large-scale synthesis well-dispersed ZnS microspheresand their photoluminescence, photocatalysis properties. Mater Charact 59:1765–1770.doi:10.1016/j.matchar.2008.04.005

104. Song L, Wei H, Xu H, Zhan J (2010) Ethanol-thermal synthesis of Cd1−xZnxS nanoparticleswith enhanced photodegradation of 4-chlorophenol. Mater Res Bull 45:1396–1400. doi:10.1016/j.materresbull.2010.06.044

105. Kale BB, Baeg JO, Lee SM et al (2006) CdIn2S4 nanotubes and marigold nanostructures: avisible-light photocatalyst. Adv Func Mater 16:1349–1354. doi:10.1002/adfm.200500525

106. Shen S, Zhao L, Guo L (2008) Cetyltrimethylammoniumbromide (CTAB)-assistedhydrothermal synthesis of ZnIn2S4 as an efficient visible-light-driven photocatalyst forhydrogen production. Inter J Hydrogen En 33:4501–4510. doi:10.1016/j.ijhydene.2008.05.043

107. Shen S, Zhao L, Guo L (2009) Crystallite, optical and photocatalytic properties ofvisible-light-driven ZnIn2S4 photocatalysts synthesized via a surfactant-assisted hydrother-mal method. Mater Res Bull 44:100–105. doi:10.1016/j.materresbull.2008.03.027

108. Chai B, Peng T, Zeng P et al (2011) Template-free hydrothermal synthesis of ZnIn2S4floriated microsphere as an efficient photocatalyst for H2 production under visible-lightirradiation. J Phys Chem C 115:6149–6155. doi:10.1021/jp1112729

109. Fan WJ, Zhou ZF, Xu WB et al (2010) Preparation of ZnIn2S4/fluoropolymer fibercomposites and its photocatalytic H2 evolution from splitting of water using Xe lampirradiation. Inter J Hydrogen En 35:6525–6530. doi:10.1016/j.ijhydene.2010.04.036

110. Jing D, Liu M, Chen Q, Guo L (2010) Efficient photocatalytic hydrogen production undervisible light over a novel W-based ternary chalcogenide photocatalyst prepared by ahydrothermal process. Inter J Hydrogen En 35:8521–8527. doi:10.1016/j.ijhydene.2010.04.170

111. Kale BB, Baeg JO, Kong K et al (2011) Self assembled CdLa2S4 hexagon flowers,nanoprisms and nanowires: novel photocatalysts for solar hydrogen production. J MaterChem 21:2624–2631. doi:10.1039/C0JM02890H

112. Cao F, Shi W, Zhao L et al (2008) Hydrothermal hynthesis and high photocatalytic activityof 3D wurtzite ZnSe hierarchical nanostructures. J Phys Chem C 112:17095–17101. doi:10.1021/jp8047345

113. Li G, Ciston S, Saponjic ZV et al (2008) Synthesizing mixed-phase TiO2 nanocompositesusing a hydrothermal method for photooxidation and photoreduction applications. J Catal253:105–110. doi:10.1016/j.jcat.2007.10.014

114. Kanna M, Wongnawa S (2008) Mixed amorphous and nanocrystalline TiO2 powdersprepared by sol–gel method: Characterization and photocatalytic study. Mater Chem Phys110:166–175. doi:10.1016/j.matchemphys.2008.01.037

115. Melghit K, Al-Rabaniah SS, Al-Amri I (2008) Low temperature preparation andcharacterization of nanospherical anatase TiO2 and its photocatalytic activity on Congored degradation under sunlight. Ceram Internat 34:479–483. doi:10.1016/j.ceramint.2006.11.011

116. Tang W, Chen Z, Katoh S (2004) Preparation of a nanocrystalline TiO2 photocatalyst using adry-process with acetylene black. Chem Lett 33:1200–1201. doi:10.1246/cl.2004.1200

117. Tang J, Durrant JR, Klug DR (2008) Mechanism of photocatalytic water splitting in TiO2.Reaction of water with photoholes, importance of charge carrier dynamics, and evidence forfour-hole chemistry. J Am Chem Soc 130:13885–13891. doi:10.1021/ja8034637

118. Kolen′ko YV, Churagulov BR, Kunst M et al (2004) Photocatalytic properties of titaniapowders prepared by hydrothermal method. Appl Catal B 54:51–58. doi:10.1016/j.apcatb.2004.06.006

290 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

119. Neppolian B, Yamashita H, Okada Y et al (2004) Preparation of TiO2 photo-catalysts bymulti-gelation and their photocatalytic reactivity for the degradation of 2-propanol. ChemLett 33:268–269. doi:10.1246/cl.2004.268

120. Neppolian B, Yamashita H, Okada Y et al (2005) Preparation of unique TiO2 nanoparticlephotocatalysts by a multigelation method for control of the physicochemical parameters andreactivity. Catal Lett 105:111–117. doi:10.1007/s10562-005-8013-1

121. Chou TP, Zhang Q, Russo B et al (2007) Titania particle size effect on the overall per–formance of dye-sensitized solar cells. J Phys Chem C 111:6296–6302. doi:10.1021/jp068939f

122. Nakade S, Saito Y, Kubo W et al (2003) Influence of TiO2 nanoparticle size on electrondiffusion and recombination in dye-sensitized TiO2 solar cells. J Phys Chem B 107:8607–8611. doi:10.1021/jp034773w

123. Li Y, Lee NH, Hwang DS et al (2004) Synthesis and characterization of nano titania powderwith high photoactivity for gas-phase photo-oxidation of benzene from TiOCl2 aqueoussolution at low temperatures. Langmuir 20:10838–10844. doi:10.1021/la0489716

124. Testino A, Bellobono IR, Buscaglia V et al (2007) Optimizing the photocatalytic propertiesof hydrothermal TiO2 by the control of phase composition and particle morphology.A systematic approach. J Am Chem Soc 129:3564–2575. doi:10.1021/ja067050+

125. Di Paola A, Cufalo G, Addamo M et al (2008) Photocatalytic activity of nanocrystallineTiO2 (brookite, rutile and brookite-based) powders prepared by thermohydrolysis of TiCl4 inaqueous chloride solutions. Colloids Surfaces A 17:366–367. doi:10.1016/j.colsurfa.2007.11.005

126. Jia H, Xu H, Hu Y et al (2007) TiO2@CdS core–shell nanorods films: fabrication anddramatically enhanced photoelectrochemical properties. Electrochem Commun 9:354–360.doi:10.1016/j.elecom.2006.10.010

127. Bavykin DV, Kulak AN, Shvalagin VV et al (2011) Photocatalytic properties of rutilenanoparticles obtained via low temperature route from titanate nanotubes. J PhotochemPhotobiol, A 218:231–238. doi:10.1016/j.jphotochem.2011.01.006

128. Yu JC, Tang HY, Yu J et al (2002) Bactericidal and photocatalytic activities of TiO2 thinfilms prepared by sol–gel and reverse micelle methods. J Photochem Photobiol, A 153:211–219. doi:10.1016/S1010-6030(02)00275-7

129. Xia XH, Liang Y, Wang Z et al (2008) Synthesis and photocatalytic properties of TiO2

nanostructures. Mater Res Bull 43:2187–2195. doi:10.1016/j.materresbull.2007.08.026130. Jung JM, Wang M, Kim EJ et al (2008) Enhanced photocatalytic activity of Au-buffered

TiO2 thin films prepared by radio frequency magnetron sputtering. Appl Catal B 84:389–392. doi:10.1016/j.apcatb.2008.04.020

131. Hou YD, Wang XC, Wu L et al (2008) N-Doped SiO2/TiO2 mesoporous nanoparticles withenhanced photocatalytic activity under visible-light irradiation. Chemosphere 72:414–421.doi:10.1016/j.chemosphere.2008.02.035

132. Wilhelm P, Stephan D (2007) Photodegradation of rhodamine B in aqueous solution viaSiO2@TiO2 nano-spheres. J Photochem Photobiol, A 185:19–25. doi:10.1016/j.jphotochem.2006.05.003

133. Xu G, Zheng Z, Wu Y, Feng N (2009) Effect of silica on the microstructure andphotocatalytic properties of titania. Ceram Internat 35:1–5. doi:10.1016/j.ceramint.2007.09.008

134. Ismail AA, Ibrahim IA, Ahmed MS et al (2004) Sol–gel synthesis of titania-silicaphotocatalyst for cyanide photodegradation. J Photochem Photobiol, A 163:445–451.doi:10.1016/j.jphotochem.2004.01.017

135. Anderson C, Bard AJ (1995) An improved photocatalyst of TiO2/SiO2 prepared by a sol–gelsynthesis. J Phys Chem 99:9882–9885. doi:10.1021/j100024a033

136. Kominami H, Yukishita K, Kimura T et al (2008) Direct solvothermal formation ofnanocrystalline TiO2 on porous SiO2 adsorbent and photocatalytic removal of nitrogenoxides in air over TiO2–SiO2 composites. Topics Catal 47:155–161. doi:10.1007/s11244-007-9016-5

References 291

137. Lim SH, Phonthammachai N, Pramana SS, White TJ (2008) Simple route to monodispersedsilica-titania core-shell photocatalysts. Langmuir 24:6226–6231. doi:10.1021/la703899j

138. Li K, Ding YY, Guo J, Wang D (2008) Surface electron structures and mechanism ofnonradiative transitions on crystalline TiO2 nanoparticles. Mater Chem Phys 112:1001–1007. doi:10.1016/j.matchemphys.2008.07.010

139. Liao DL, Liao BQ (2007) Shape, size and photocatalytic activity control of TiO2

nanoparticles with surfactants. J Photochem Photobiol, A 187:363–369. doi:10.1016/j.jphotochem.2006.11.003

140. Scotti R, Bellobono IR, Canevali C et al (2008) Sol–gel pure and mixed-phase titaniumdioxide for photocatalytic purposes: Relations between phase composition, catalytic activity,and charge-trapped sites. Chem Mater 20:4051–4061. doi:10.1021/cm800465n

141. Neppolian B, Celik E, Anpo M, Choi H (2008) Ultrasonic-assisted pH swing method for thesynthesis of highly efficient TiO2 nano-size photocatalysts. Catal Lett 125:183–191. doi:10.1007/s10562-008-9563-9

142. Choi H, Stathatos E, Dionysiou DD (2007) Effect of surfactant in a modified sol on thephysicochemical properties and photocatalytic activity of crystalline TiO2 nanoparticles.Topics Catal 44:513–521. doi:10.1007/s11244-006-0099-1

143. Jiu J, Isoda S, Adachi M, Wang F (2007) Preparation of TiO2 nanocrystalline with 3–5 nmand application for dye-sensitized solar cell. J Photochem Photobiol, A 189:314–321.doi:10.1016/j.jphotochem.2007.02.021

144. Stengl V, Bakardjieva S, Murafa N et al (2007) Preparation, characterization andphotocatalytic activity of optically transparent titanium dioxide nanoparticles. Mater ChemPhys 105:38–46. doi:10.1016/j.matchemphys.2007.04.037

145. Carrera R, Vázquez AL, Arce E et al (2007) Photodecomposition of NO by sol–gel TiO2

catalysts under atmospheric conditions: Effect of the method on the textural and morphologicproperties. J Alloys Compounds 434–435:788–791. doi:10.1016/j.jallcom.2006.08.334

146. Zhang S, Yu Q, Chen Z et al (2007) Nano-TiO2 particles with increased photocatalyticactivity prepared by the miniemulsion method. Mater Lett 61:4839–4842. doi:10.1016/j.matlet.2007.03.054

147. Kubacka A, Fuerte A, Martinez-Arias A, Fernandez-Garcia M (2007) Nanosized Ti–Vmixed oxides: Effect of doping level in the photocatalytic degradation of toluene usingsunlight-type excitation. Appl Catal B 74:26–33. doi:10.1016/j.apcatb.2007.01.011

148. Jang HD, Kim SK, Kim SJ (2001) Effect of particle size and phase composition of titaniumdioxide nanoparticles on the photocatalytic properties. J Nanopart Res 3:141–147. doi:10.1023/A:1017948330363

149. Marimuthu A, Madras G (2008) Photocatalytic oxidative degradation of poly(alkylacrylates)with nanoTiO2. Ind Eng Chem Res 47:2182–2190. doi:10.1021/ie0712939

150. Vinu R, Madras G (2008) Photocatalytic degradation of poly(acrylamide-co-acrylic acid).J Phys Chem C 112:8928–8935. doi:10.1021/jp801887t

151. Surolia PK, Lazar MA, Tayade RJ, Jasra RV (2008) Photocatalytic degradation of 3,3′-dimethylbiphenyl-4,4′-diamine (o-tolidine) over nanocrystalline TiO2 synthesized by sol–gel, solution combustion, and hydrothermal methods. Ind Eng Chem Res 47:5847–5855.doi:10.1021/ie800073j

152. Priya MH, Madras G (2006) Photocatalytic degradation of nitrobenzenes with combustionsynthesized nano-TiO2. J Photochem Photobiol, A 178:1–7. doi:10.1016/j.jphotochem.2005.06.012

153. Vinu R, Madras G (2008) Kinetics of simultaneous photocatalytic degradation of phenoliccompounds and reduction of metal ions with nano-TiO2. Environ Sci Technol 42:913–919.doi:10.1021/es8025648

154. Alves AK, Berutti FA, Clemens FJ et al (2009) Photocatalytic activity of titania fibersobtained by electrospinning. Mater Res Bull 44:312–317. doi:10.1016/j.materresbull.2008.06.001

155. Li W, Liu C, Zhou Y et al (2008) Enhanced photocatalytic activity in anatase/TiO2(B)core-shell nanofiber. J Phys Chem C 112:20539–20545. doi:10.1021/jp808183q

292 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

156. Lin Y (2008) Photocatalytic activity of TiO2 nanowire arrays. Mater Lett 62:1246–1248.doi:10.1016/j.matlet.2007.08.021

157. Ren X, Han D, Chen D, Tang F (2007) Large-scale synthesis of hexagonal cone-shaped ZnOnanoparticles with a simple route and their application to photocatalytic degradation. MaterRes Bull 42:807–813. doi:10.1016/j.materresbull.2006.08.030

158. Wang H, Xie C (2008) Effect of annealing temperature on the microstructures andphotocatalytic property of colloidal ZnO nanoparticles. J Phys Chem Solids 69:2440–2444.doi:10.1016/j.jpcs.2008.04.036

159. Ye C, Bando Y, Shen G, Golberg D (2006) Thickness-dependent photocatalyticperformance of ZnO nanoplatelets. J Phys Chem B 110:15146–15151. doi:10.1021/jp061874w

160. Dodd AC, McKinley AJ, Saunders M, Tsuzuki T (2006) Effect of particle size on thephotocatalytic activity of nanoparticulate zinc oxide. J Nanopart Res 8:43–51. doi:10.1007/s11051-005-5131-z

161. Liqiang J, Baifu X, Fulong Y et al (2004) Deactivation and regeneration of ZnO and TiO2

nanoparticles in the gas phase photocatalytic oxidation of n-C7H16 or SO2. Appl Catal A275:49–54. doi:10.1016/j.apcata.2004.07.019

162. Wang G, Chen D, Zhang H et al (2008) Tunable photocurrent spectrum in well-oriented zincoxide nanorod arrays with enhanced photocatalytic activity. J Phys Chem C 112:8850–8855.doi:10.1021/jp800379k

163. Sun T, Qiu J, Liang C (2008) Controllable fabrication and photocatalytic activity of ZnOnanobelt array. J Phys Chem C 112:715–721. doi:10.1021/jp710071f

164. Wan Q, Wang TH, Zhao JC (2005) Enhanced photocatalytic activity of ZnO nanotetrapods.Appl Phys Lett 87:083105. doi:10.1063/1.2034092

165. Wang Y, Li X, Lu G et al (2008) Highly oriented 1-D ZnO nanorod arrays on zinc foil:direct growth from substrate, optical properties and photocatalytic activities. J Phys Chem C112:7332–7336. doi:10.1021/jp7113175

166. Sadakane M, Sasaki K, Kunioku H et al (2008) Preparation of nano-structured crystallinetun–gsten (VI) oxide and enhanced photocatalytic activity for decomposition of organiccompounds under visible light irradiation. Chem Commun 48:6552–6554. doi:10.1039/B815214D

167. Sadakane M, Tamura N, Kanome N et al (2011) Preparation of crystalline tungsten oxidenanorods with enhanced photocatalytic activity under visible light irradiation. Chem Lett40:443–444. doi:10.1246/cl.2011.443

168. Su J, Feng X, Sloppy JD et al (2011) Vertically aligned WO3 nanowire arrays grown directlyon transparent conducting oxide coated glass: synthesis and photoelectrochemical properties.Nano Lett 11:203–208. doi:10.1021/nl1034573

169. Zhang J, Shi F, Lin J et al (2008) Self-assembled 3-D architectures of BiOBr as a visiblelight-driven photocatalyst. Chem Mater 20:2937–2941. doi:10.1021/cm7031898

170. Zhang X, Ai Z, Jia F, Zhang L (2008) Generalized one-pot synthesis, characterization, andphotocatalytic activity of hierarchical BiOX (X = Cl, Br, I) nanoplate microspheres. J PhysChem C 112:747–753. doi:10.1021/jp077471t

171. Xiao X, Zhang WD (2010) Facile synthesis of nanostructured BiOI microspheres with highvisible light-induced photocatalytic activity. J Mater Chem 20:5866–5870. doi:10.1039/C0JM00333F

172. Sun M, Li D, Zhang W et al (2010) Rapid microwave hydrothermal synthesis of GaOOHnano-rods with photocatalytic activity toward aromatic compounds. Nanotechnology21:355601. doi:10.1088/0957-4484/21/35/355601

173. Chang CA, Ray B, Paul DK et al (2008) Photocatalytic reaction of acetaldehyde over SrTiO3

nanoparticles. J Mol Catal A 281:99–106. doi:10.1016/j.molcata.2007.08.014174. Liu Y, Xie L, Li Y et al (2008) Synthesis and high photocatalytic hydrogen production of

SrTiO3 nanoparticles from water splitting under UV irradiation. J Power Sources 193:701–707. doi:10.1016/j.jpowsour.2008.05.057

References 293

175. Zhang G, He F, Zou X et al (2008) Hydrothermal preparation and photocatalytic propertiesof sheet-like nanometer niobate K4Nb6O17. J Phys Chem Solids 69:1471–1474. doi:10.1016/j.jpcs.2007.10.095

176. Zhou C, Chen C, Wang Q (2011) High photocatalytic activity of porous K4Nb6O17

microsphere with large surface area prepared by homogeneous precipitation using urea.J Mol Catal A 339:37–42. doi:10.1016/j.molcata.2011.02.009

177. Sarahan MC, Carrol EC, Allen M et al (2008) K4Nb6O17-derived photocatalysts forhydrogen evolution from water: nanoscrolls versus nanosheets. J Solid State Chem181:1681–1686. doi:10.1016/j.jssc.2008.06.021

178. Zhang S, Zhang C, Man Y, Zhu Y (2006) Visible-light-driven photocatalyst of Bi2WO6

nanoparticles prepared via amorphous complex precursor and photocatalytic properties.J Solid State Chem 179:62–69. doi:10.1016/j.jssc.2005.09.041

179. Wu L, Bi J, Li Z et al (2008) Rapid preparation of Bi2WO6 photocatalyst with nanosheetmorphology via microwave-assisted solvothermal synthesis. Catal Today 131:15–20. doi:10.1016/j.cattod.2007.10.089

180. Bi J, Wu L, Li J et al (2007) Simple solvothermal routes to synthesize nanocrystallineBi2MoO6 photocatalysts with different morphologies. Acta Mater 55:4699–4705. doi:10.1016/j.actamat.2007.04.034

181. Wu J, Duan F, Zheng Y, Xie Y (2007) Synthesis of Bi2WO6 nanoplate-built hierarchicalnest-like structures with visible-light-induced photocatalytic activity. J Phys Chem C111:12866–12871. doi:10.1021/jp073877u

182. Zhang C, Zhu Y (2005) Synthesis of square Bi2WO6 nanoplates as high-activityvisible-light-driven photocatalysts. Chem Mater 17:3537–3545. doi:10.1021/cm0501517

183. Lin J, Lin J, Zhu Y (2007) Controlled synthesis of the ZnWO4 nanostructure and effects onthe photocatalytic performance. Inorg Chem 46:8372–8378. doi:10.1021/ic701036k

184. Zhang L, Chen D, Jiao X (2006) Monoclinic structured BiVO4 nanosheets: hydrothermalpreparation, formation mechanism, and coloristic and photocatalytic properties. J PhysChem B 110:2668–2673. doi:10.1021/jp056367d

185. Mahapatra S, Nayak SK, Madras G, Guru Row TN (2008) Microwave synthesis andphotocatalytic activity of nano lanthanide (Ce, Pr, and Nd) orthovanadates. Ind Eng ChemRes 47:6509–6516. doi:10.1021/ie8003094

186. Bouras P, Lianos P (2005) Photodegradation of dyes in aqueous solutions catalyzed byhighly efficient nanocrystalline titania films. J Appl Electrochem 35:831–836. doi:10.1007/s10800-005-1616-9

187. Bouras P, Stathatos E, Lianos P (2007) Pure versus ion-metal-doped nanocrystalline titaniafor photocatalysis. Appl Catal B 73:51–59. doi:10.1016/j.apcatb.2006.06.007

188. Strataki N, Bekiari V, Kondarides DI, Lianos P (2007) Hydrogen production byphotocatalytic alcohol reforming employing highly efficient nanocrystalline titania films.Appl Catal B 77:184–189. doi:10.1016/j.apcatb.2007.07.015

189. Strataki N, Bekiari V, Stathatos E, Pianos P (2007) Effect of aggregation of dyes adsorbedon nanocrystalline titania films on the efficiency of photodegradation. J PhotochemPhotobiol, A 191:13–18. doi:10.1016/j.jphotochem.2007.03.023

190. Yu JC, Yu J, Ho W, Zhao J (2002) Light-induced super-hydrophilicity and photocatalyticactivity of mesoporous TiO2 thin films. J Photochem Photobiol, A 148:331–339. doi:10.1016/S1010-6030(02)00060-6

191. Liu Y, Wang X, Yang F, Yang X (2008) Excellent antimicrobial properties of mesoporousanatase TiO2 and Ag/TiO2 composite films. Micropor Mesopor Mater 114:431–439. doi:10.1016/j.micromeso.2008.01.032

192. Angelome PC, Andrini L, Calvo ME et al (2007) Mesoporous anatase TiO2 films: use ofTi K XANES for the quantification of the nanocrystalline character and substrate effects inthe photocatalysis behavior. J Phys Chem C 111:10886–10893. doi:10.1021/jp069020z

193. Chen Y, Dionysiou DD (2008) Bimodal mesoporous TiO2–P25 composite thick films withhigh photocatalytic activity and improved structural integrity. Appl Catal B 80:147–155.doi:10.1016/j.apcatb.2007.11.010

294 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

194. Habibi MH, Esfahani MN, Egerton TA (2007) Photochemical charactrization andphotocatalytic properties of a nanostructure composite TiO2 film. Inter J Photoen 2007:1–8

195. Negishi N, Matsuzawa S, Takeuchi K, Pichat P (2007) Transparent micrometer-thick TiO2

films on SiO2-coated glass prepared by repeated dip-coating/calcination: characteristics andphotocatalytic activities for removing acetaldehyde or toluene in air. Chem Mater 19:3808–3814. doi:10.1021/cm070320i

196. Guan H, Zhu L, Zhou H, Tang H (2008) Rapid probing of photocatalytic activity ontitania-based self-cleaning materials using 7-hydroxycoumarin fluorescent probe. AnalChem Acta 608:73–78. doi:10.1016/j.aca.2007.12.009

197. Yang J, Chen C, Ji H et al (2005) Mechanism of TiO2-assisted photocatalytic degradation ofdyes under visible irradiation: Photoelectrocatalytic study by TiO2-film electrodes. J PhysChem B 109:21900–21907. doi:10.1021/jp0540914

198. Ercan A, Bakir U, Karakas G (2006) Photocatalytic microbal inactivation over Pd dopedSnO2 and TiO2 thin films. J Photochem Photobiol, A 184:313–321. doi:10.1016/j.jphotochem.2006.05.001

199. Mills A, Elliott N, Hill G et al (2003) Preparation and characterisation of novel thick sol–geltitania film photocatalysts. Photochem Photobiol Sci 2:591–596. doi:10.1039/B212865A

200. Ao CH, Lee SC, Yu JC (2003) Photocatalyst TiO2 supported on glass fiber for indoor airpurification: effect of NO on the photodegradation of CO and NO2. J Photochem Photobiol,A 156:171–177. doi:10.1016/S1010-6030(03)00009-1

201. Sunada K, Watanabe T, Hashimoto K (2003) Studies on photokilling of bacteria on TiO2

thin film. J Photochem Photobiol, A 156:227–233. doi:10.1016/S1010-6030(02)00434-3202. Xu J, Ao Y, Fu D et al (2008) Photocatalytic activity on TiO2-coated side-glowing optical

fiber reactor under solar light. J Photochem Photobiol, A 199:165–169. doi:10.1016/j.jphotochem.2008.05.019

203. Yang B, Barnes PRF, Zhang Y, Luca V (2007) Tungsten trioxide films with controlledmorphology and strong photocatalytic activity via a simple sol–gel route. Catal Lett118:280–284. doi:10.1007/s10562-007-9188-4

204. Nakamura N, Tanaka T, Nakato Y (2004) Mechanism for visible light responses in anodicphotocurrents at N-doped TiO2 film electrodes. J Phys Chem B 108:10617–10620. doi:10.1021/jp048112q

205. Umemura Y, Shinohara E, Koura A et al (2006) Photocatalytic decomposition of analkylammonium cation in a Langmuir-Blodgett film of a titania nanosheet. Langmuir22:3870–3877. doi:10.1021/la0530743

206. Mills A, Elliott N, Parkin IP et al (2002) Novel TiO2 CVD films for semiconductorphotocatalysis. J Photochem Photobiol, A 151:171–179. doi:10.1016/S1010-6030(02)00190-9

207. Brook LA, Evans P, Foster HA et al (2007) Highly bioactive silver and silver/titaniacomposite films grown by chemical vapour deposition. J Photochem Photobiol, A 187:53–63. doi:10.1016/j.jphotochem.2006.09.014

208. Ohko Y, Nakamura Y, Fukuda A et al (2008) Photocatalytic oxidation of nitrogen dioxidewith TiO2 thin films under continuous UV-light illumination. J Phys Chem C 112:10502–10508. doi:10.1021/jp802959c

209. Yoon KH, Noh JS, Kwon CH, Muhammed M (2006) Photocatalytic behavior of TiO2 thinfilms prepared by sol–gel process. Mater Chem Phys 95:79–83. doi:10.1016/j.matchemphys.2005.06.001

210. Mishra PR, Shukla PK, Srivastava ON (2007) Study of modular PEC solar cells forphotoelectrochemical splitting of water employing nanostructured TiO2 photoelectrodes.Inter J Hydrogen En 32:1680–1685. doi:10.1016/j.ijhydene.2006.10.002

211. Li D, Guo Y, Hu C et al (2004) Preparation, characterization and photocatalytic property ofthe PW11O39

7−/TiO2 composite film towards azo-dye degradation. J Mol Catal A 207:183–193. doi:10.1016/S1381-1169(03)00491-6

References 295

212. Gao R, Stark J, Bahnemann DW, Rabani J (2002) Quantum yields of hydroxyl radicals inilluminated TiO2 nanocrystallite layers. J Photochem Photobiol, A 148:387–391. doi:10.1016/S1010-6030(02)00066-7

213. Ao Y, Xu J, Fu D, Yuan C (2008) Preparation of porous titania thin film and itsphotocatalytic activity. Appl Surf Sci 255:3137–3140. doi:10.1016/j.apsusc.2008.08.102

214. Piris J, Ferguson AJ, Blackburn JL et al (2008) Efficient photoinduced charge injection fromchemical bath deposited CdS mesoporous TiO2 probed with time-resolved microwaveconductivity. J Phys Chem C 112:7742–7749. doi:10.1021/jp800527r

215. Song XM, Wu JM, Yan M (2008) Photocatalytic and photoelectrocatalytic degradation ofaqueous Rhodamine B by low-temperature deposited anatase thin films. Mater Chem Phys112:510–515. doi:10.1016/j.matchemphys.2008.06.009

216. Yu JG, Yu HG, Cheng B et al (2003) The effect of calcination temperature on the surfacemicrostructure and photocatalytic activity of TiO2 thin films prepared by liquid phasedeposition. J Phys Chem B 107:13871–13879. doi:10.1021/jp036158y

217. Gu DE, Yang BC, Hu YD (2007) A novel method for preparing V-doped titanium dioxidethin film photocatalysts with high photocatalytic activity under visible light irradiation. CatalLett 118:254–259. doi:10.1007/s10562-007-9179-5

218. Sagawa T, Kotani M, Nada H et al (2003) Photoinduced reduction of methylviologen withTiO2/polymer films. Chem Lett 32:962–963. doi:10.1246/cl.2003.962

219. Zhao Y, Zhang X, Zhai J et al (2008) Enhanced photocatalytic activity of hierarchicallymicro-/nano-porous TiO2 films. Appl Catal B 83:24–29. doi:10.1016/j.apcatb.2008.01.035

220. Takeuchi M, Yamashita H, Matsuoka M et al (2000) Photocatalytic decomposition of NO ontitanium oxide thin film photocatalysts prepared by an ionized cluster beam technique. CatalLett 66:185–187. doi:10.1023/A:1019095406121

221. Burlacov I, Jirkovsky J, Kavan L et al (2007) Cold gas dynamic spraying (CGDS) of TiO2

(anatase) powders onto poly(sulfone) substrates: microstructural characterisation andphotocata–lytic efficiency. J Photochem Photobiol, A 187:285–292. doi:10.1016/j.jphotochem.2006.10.023

222. Ryu J, Park DS, Hahn BD et al (2008) Photocatalytic TiO2 thin films by aerosol-deposition:from micron-sized particles to nano-grained thin film at room temperature. Appl Catal B83:1–7. doi:10.1111/j.1551-2916.2009.03391.x

223. Ohko Y, Tatsuma T, Fujishima A (2001) Characterization of TiO2 photocatalysis in the gasphase as a photoelectrochemical system: behavior of salt-modified systems. J Phys Chem B105:10016–10021. doi:10.1021/jp011459y

224. Shinde PS, Sadale SB, Patil PS et al (2008) Properties of spray deposited titanium dioxidethin films and their application in photoelectrocatalysis. Sol En Mater Sol Cells 92:283–290.doi:10.1016/j.solmat.2007.09.001

225. Thimsen E, Rastgar N, Biswas P (2008) Nanostructured TiO2 films with controlledmorphology synthesized in a single step process: performance of dye-sensitized solar cellsand photo watersplitting. J Phys Chem C 112:4134–4140. doi:10.1021/jp710422f

226. Li M, Su J, Guo L (2008) Preparation and characterization of ZnIn2S4 thin films depositedby spray pyrolysis for hydrogen production. Inter J Hydrogen En 33:2891–2896. doi:10.1016/j.ijhydene.2008.04.008

227. Kemell M, Pore V, Tupala J et al (2007) Atomic layer deposition of nanostructured TiO2

photocatalyst via template approach. Chem Mater 19:1816–1820. doi:10.1021/cm062576e228. Liu R, Lin Y, Chou LY et al (2011) Splitting by tungsten oxide prepared by atomic layer

deposition and decorated with an oxygen-evolving catalyst. Angew Chem Int Ed 50:499–502. doi:10.1002/ange.201004801

229. Klahr BM, Martinson ABF, Hamann TW (2011) Photoelectrochemical investigation ofultrathin film iron oxide solar cells prepared by atomic layer deposition. Langmuir 27:461–468. doi:10.1021/la103541n

230. Kitano M, Tsujimaru K, Anpo M (2008) Hydrogen production using highly active titaniumoxide-based photocatalysts. Topics Catal 49:4–17. doi:10.1007/s11244-008-9059-2

296 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

231. Dholam R, Patel N, Adami M, Miotello A (2008) Physically and chemically synthesizedTiO2 composite thin films for hydrogen production by photocatalytic water splitting. Inter JHydrogen En 33:6896–6903. doi:10.1016/j.ijhydene.2008.08.061

232. Selli E, Chiarello GL, Quartarone E et al (2007) A photocatalytic water splitting device forseparate hydrogen and oxygen evolution. Chem Commun 5022–5024. doi:10.1039/B711747G

233. Kitano M, Takeuchi M, Matsuoka M et al (2005) Preparation of visible light-responsiveTiO2 thin film photocatalysts by an RF magnetron sputtering deposition method and theirphotocatalytic reactivity. Chem Lett 34:616–617. doi:10.1246/cl.2005.616

234. Cao J, Kako T, Li P et al (2011) Fabrication of p-type CaFe2O4 nanofilms forphotoelectrochemical hydrogen generation. Electrochem Commun 13:275–278. doi:10.1016/j.elecom.2011.01.002

235. Hu YS, Kleiman-Shwarsctein A, Forman AJ et al (2008) Pt-doped a-Fe2O3 thin films activefor photoelectrochemical water splitting. Chem Mater 20:3803–3805. doi:10.1021/cm800144q

236. Kleiman-Shwarsctein A, Hu YS, Forman AJ et al (2008) Electrodeposition of a-Fe2O3

doped with Mo or Cr as photoanodes for photocatalytic water splitting. J Phys Chem C112:15900–15907. doi:10.1021/jp803775j

237. Nian JN, Hu CC, Teng H (2008) Electrodeposited p-type Cu2O for H2 evolution fromphotoelectrolysis of water under visible light illumination. Inter J Hydrogen En 33:2897–2903. doi:10.1016/j.ijhydene.2008.03.052

238. Yu H, Quan X, Zhang Y et al (2008) Electrochemically assisted photocatalytic inactivationof Escherichia coli under visible light using a ZnIn2S4 film electrode. Langmuir 24:7599–7604. doi:10.1021/la800835k

239. Chu SZ, Inoue S, Wada K et al (2003) Highly porous (TiO2-SiO2-TeO2)/Al2O3/TiO2

composite nanostructures on glass with enhanced photocatalysis fabricated by anodizationand sol–gel process. J Phys Chem B 107:6586–6589. doi:10.1021/jp0349684

240. Kang Z, Tsang CHA, Wong NB et al (2007) Silicon quantum dots: a general photocatalystfor reduction, decomposition, and selective oxidation reactions. J Am Chem Soc 129:12090–12091. doi:10.1021/ja075184x

241. Schürch D, Currao A, Sarkar S et al (2002) The silver chloride photoanode inphotoelectrochemical water splitting. J Phys Chem B 106:12764–12775. doi:10.1021/jp0265081

242. Soller-Illia G, Sanchez C, Lebeau B, Patarin J (2002) Chemical strategies to design texturedmaterials: from microporous and mesoporous oxides to nanonetworks and hierarchicalstructures. Chem Rev 102:4093–4138. doi:10.1021/cr0200062

243. Shchukin DG, Sviridov DV (2006) Photocatalytic processes in spatially confined micro- andnanoreactors. J Photochem Photobiol, C 7:23–39. doi:10.1016/j.jphotochemrev.2006.03.002

244. Stroyuk OL, Shvalagin VV, Raevskaya AE et al (2008) Photochemical formation ofsemiconducting nanostructures. Theoret Experim Chem 44:205–231

245. Zhang X, Yang H, Zhang F, Chan KY (3007) Preparation and characterization of Pt–TiO2–

SiO2 mesoporous materials and visible-light photocatalytic performance. Mater Lett61:2231–2234. doi:10.1016/j.matlet.2006.08.053

246. Smirnova N, Gnatyuk Y, Eremenko A et al (2006) Photoelectrochemical characterizationand photocatalytic properties of mesoporous TiO2/ZrO2 films. Inter J Photoen 1–6. doi:10.1155/IJP/2006/85469

247. Mohamed MM, Bayoumy WA, Khairy M, Mousa MA (2008) Structural features andphotocatalytic behavior of titania and titania supported vanadia synthesized by polyolfunctionalized materials. Mircopor Mesopor Mater 109:445–457. doi:10.1016/j.micromeso.2007.05.055

248. Deshpande AS, Shchukin DG, Ustinovich E et al (2005) Titania and mixedtitania/aluminum, gallium, or indium oxide spheres: sol–gel/template synthesis andphotocatalytic properties. Adv Func Mater 15:239–245. doi:10.1002/adfm.200400220

References 297

249. Yuan S, Sheng Q, Zhang J et al (2008) Synthesis of thermally stable mesoporous TiO2 andinvestigation of its photocatalytic activity. Micropor Mesopor Mater 110:501–507. doi:10.1016/j.micromeso.2007.06.039

250. Oh ST, Choi JS, Lee HS et al (2007) H2O-controlled synthesis of TiO2 with nanosizedchannel structure through in situ esterification and its application to photocatalytic oxidation.J Mol Catal A 267:112–119. doi:10.1016/j.molcata.2006.11.028

251. Chen L, Yao B, Cao Y, Fan K (2007) Synthesis of well-ordered mesoporous titania withtunable phase content and high photoactivity. J Phys Chem C 111:11849–11853. doi:10.1021/jp072070z

252. Fan X, Yu T, Wang Y et al (2008) Role of phosphorus in synthesis of phosphatedmesoporous TiO2 photocatalytic materials by EISA method. Appl Surf Sci 254:5191–5198.doi:10.1016/j.apsusc.2008.02.038

253. Huang J, Wang X, Hou Y et al (2008) Synthesis of functionalized mesoporous TiO2

molecular sieves and their application in photocatalysis. Micropor Mesopor Mater 110:543–552. doi:10.1016/j.micromeso.2007.06.055

254. Zhang L, Yu JC (2003) A sonochemical approach to hierarchical porous titania spheres withenhanced photocatalytic activity. Chem Commun 2078–2079. doi:10.1039/B306013F

255. Ismail AA, Bahnemann DW (2011) Mesostructured Pt/TiO2 nanocomposites as highlyactive photocatalysts for the photooxidation of dichloroacetic acid. J Phys Chem C115:5784–5791. doi:10.1021/jp110959b

256. Wang X, Mitchell DRG, Prince K et al (2008) Gold nanoparticle incorporation into poroustitania networks using an agarose gel templating technique for photocatalytic applications.Chem Mater 20:3917–3926. doi:10.1021/cm703509f

257. van der Meulen T, Mattson A, Österlund L (2007) A comparative study of the photocatalyticoxidation of propane on anatase, rutile, and mixed-phase anatase–rutile TiO2 nanoparticles:role of surface intermediates. J Catal 251:131–144. doi:10.1016/j.jcat.2007.07.002

258. An T, Liu J, Li G et al (2008) Structural and photocatalytic degradation characteristics ofhydrothermally treated mesoporous TiO2. Appl Catal A 350:237–243. doi:10.1016/j.apcata.2008.08.022

259. Shen Q, Yang H, Gao J, Yang J (2007) Low-temperature fabrication of porous anatase TiO2

film with tiny slots and its photocatalytic activity. Mater Lett 61:4160–4162. doi:10.1016/j.matlet.2007.01.045

260. Sang L, Dai H, Sun J et al (2010) Fabrication of the hydrogen-evolving photocatalyst withmesoporous structure. Inter J Hydrogen En 35:7098–7103. doi:10.1016/j.ijhydene.2010.01.019

261. Wang HW, Lin HC, Kuo CH et al (2008) Synthesis and photocatalysis of mesoporousanatase TiO2 powders incorporated Ag nanoparticles. J Phys Chem Solids 69:633–636.doi:10.1016/j.jpcs.2007.07.052

262. Sreethawong T, Suzuki Y, Yoshikawa S (2006) Platinum-loaded mesoporous titania bysingle-step sol–gel process with surfactant template: photocatalytic activity for hydrogenevolution. Compt Rend Chim 9:307–314. doi:10.1016/j.crci.2005.05.015

263. Kozlova EA, Vorontsov AV (2007) Influence of mesoporous and platinum-modifiedtitanium dioxide preparation methods on photocatalytic activity in liquid and gas phase.Appl Catal B 77:35–45. doi:10.1016/j.apcatb.2007.07.001

264. Tian G, Fu H, Jing L et al (2008) Preparation and characterization of stable biphase TiO2

photocatalyst with high crystallinity, large surface area, and enhanced photoactivity. J PhysChemC 112:3083–3089. doi:10.1021/jp710283p

265. De Witte K, Ribbens S, Meynen V et al (2008) Photoatalytic study of P25 and mesoporoustitania in aqueous and gaseous environment. Catal Commun 9:1787–1792. doi:10.1016/j.catcom.2008.02.013

266. Ao Y, Xu J, Fu D, Yuan C (2008) Preparation of Ag-doped mesoporous titania and itsenhanced photocatalytic activity under UV light irradiation. J Phys Chem Solids 69:2660–2664. doi:10.1016/j.jpcs.2008.06.100

298 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

267. Song C, Wang D, Xu Y, Hu Z (2011) Preparation of Ag-TiO2 hollow structures withenhanced photocatalytic activity. Mater Lett 65:908–910. doi:10.1016/j.matlet.2010.12.015

268. Dai Q, Shi LY, Luo YG et al (2002) Effect of templates on the structure, stability andphotocatalytic activity mesostructured TiO2. J Photochem Photobiol, A 148:295–301.doi:10.1016/S1010-6030(02)00056-4

269. Lakshminarasimhan N, Bae E, Choi W (2007) Enhanced photocatalytic production of H2 onmesoporous TiO2 prepared by template-free method: role of interparticle charge transfer.J Phys Chem C 111:15244–15250. doi:10.1021/jp0752724

270. Matsunaga T, Yamaoka H, Ohtani S et al (2008) High photocatalytic activity ofpalladium-deposited mesoporous TiO2/SiO2 fibers. Appl Catal A 351:231–238. doi:10.1016/j.apcata.2008.09.020

271. Pavasupree S, Ngamsinlapasathian S, Nakajima M et al (2006) Synthesis, characterization,photocatalytic activity and dye-sensitized solar cell performance of nanorods/nanoparticlesTiO2 with mesoporous structure. J Photochem Photobiol, A 184:163–169. doi:10.1016/j.jphotochem.2006.04.010

272. Wang G (2007) Hydrothermal synthesis and photocatalytic activity of nanocrystalline TiO2

powders in ethanol–water mixed solutions. J Mol Catal A 274:185–191. doi:10.1016/j.molcata.2007.05.009

273. Yu J, Zhang L, Cheng B, Su Y (2007) Hydrothermal preparation and photocatalytic activityof hierarchically sponge-like macro-/mesoporous titania. J Phys Chem C 111:10582–10589.doi:10.1021/jp0707889

274. Yu JC, Zhang L, Yu J (2002) Rapid synthesis of mesoporous TiO2 with high photocatalyticactivity by ultrasound-induced agglomeration. New J Chem 26:416–420. doi:10.1039/B109173E

275. Fang J, Wang F, Qian K et al (2008) Bifunctional N-doped mesoporous TiO2 photocatalysts.J Phys ChemC 112:18150–18156. doi:10.1021/jp805926b

276. Sreethawong T, Ngamsinlapasathian S, Suzuki Y, Yoshikawa S (2005) Nanocrystallinemeso–porous Ta2O5-based photocatalysts prepared by surfactant-assisted templating sol–gelprocess for photocatalytic H2 evolution. J Mol Catal A 235:1–11. doi:10.1016/j.molcata.2005.03.021

277. Noda Y, Lee B, Domen K, Kondo JN (2008) Synthesis of crystallized mesoporous tantalumoxide and its photocatalytic activity for overall water splitting under ultraviolet lightirradiation. Chem Mater 20:5361–5367. doi:10.1021/cm703202n

278. Chen X, Yu T, Fan X et al (2007) Enhanced activity of mesoporous Nb2O5 forphotocatalytic hydrogen production. Appl Surface Sci 253:8500–8506. doi:10.1016/j.apsusc.2007.04.035

279. Abe R, Shinohara K, Tanaka A et al (1997) Preparation of porous niobium oxides bysoft-chemical process and their photocatalytic activity. Chem Mater 9:2179–2184. doi:10.1021/cm970284v

280. Zhu H, Hagfeldt A, Boschloo G (2007) Photoelectrochemistry of mesoporous NiOelectrodes in iodide/triiodide electrolytes. J Phys Chem C 111:17455–17458. doi:10.1021/jp077134k

281. Sathish M, Viswanath RP (2007) Photocatalytic generation of hydrogen over mesoporousCdS nanoparticle: effect of particle size, noble metal and support. Catal Today 129:421–427.doi:10.1016/j.cattod.2006.12.008

282. Wang L, Tang K, Liu Z et al (2011) Single-crystalline ZnSn(OH)6 hollow cubes viaself-templated synthesis at room temperature and their photocatalytic properties. J MaterChem 21:4352–4357. doi:10.1039/C0JM03734F

283. Jang JS, Yu CJ, Choi SH et al (2008) Topotactic synthesis of mesoporous ZnO and ZnSnanoplates and their photocatalytic activity. J Catal 254:144–155. doi:10.1016/j.jcat.2007.12.010

284. Li Y, He X, Cao M (2008) Micro-emulsion-assisted synthesis of ZnS nanospheres and theirphotocatalytic activity. Mater Res Bull 43:3100–3110. doi:10.1016/j.materresbull.2007.11.016

References 299

285. Kartsonakis IA, Liatsi P, Danilidis I et al (2008) Synthesis, characterization and antibacterialaction of hollow titania spheres. J Phys Chem Sol 69:214–221. doi:10.1016/j.jpcs.2007.08.071

286. Song X, Gao L (2007) Fabrication of hollow hybrid microspheres coated with silica/titaniavia sol–gel process and enhanced photocatalytic activities. J Phys Chem C 111:8180–8187.doi:10.1021/jp071142j

287. Deng Z, Chen M, Gu G, Wu L (2008) A facile method to fabricate ZnO hollow spheres andtheir photocatalytic property. J Phys Chem C 112:16–22. doi:10.1021/jp077662w

288. Syoufian A, Nakashima K (2008) Degradation of methylene blue in aqueous dispersion ofhollow titania photocatalyst: study of reaction enhancement by various electron scavengers.J Colloid Interface Sci 317:507–512. doi:10.1016/j.jcis.2007.04.027

289. Cao FL, Wang JG, Lv FJ et al (2011) Photocatalytic oxidation of toluene to benzaldehydeover anatase TiO2 hollow spheres with exposed 001 facets. Catal Commun 12:946–950.doi:10.1016/j.catcom.2011.03.007

290. Yu J, Wang G (2008) Hydrothermal synthesis and photocatalytic activity of mesoporoustitania hollow spheres. J Phys Chem Sol 69:1147–1151. doi:10.1016/j.jpcs.2007.09.024

291. Ao Y, Xu J, Fu D, Yuan C (2008) Photoelectrochemical application of hollow titania film.Electrochem Commun 10:1812–1814. doi:10.1016/j.elecom.2008.09.015

292. Ao Y, Xu J, Fu D, Yuan D (2009) Visible-light responsive C, N-codoped titania hollowspheres for X-3B dye photodegradation. Micropor Mesopor Mater 118:382–386. doi:10.1016/j.micromeso.2008.09.010

293. Lee SC, Lee CW, Lee SC, Lee JS (2008) Optical and photocatalytic properties of TiO2 withhollow nanostructure. Mater Lett 62:564–566. doi:10.1016/j.matlet.2007.09.008

294. Yang L, Zhu L, Liu C et al (2008) Synthesis and photocatalytic property of porous TiO2

microspheres. Mater Res Bull 43:806–810. doi:10.1016/j.materresbull.2007.06.044295. Li H, Bian Z, Zhu J et al (2007) Mesoporous titania spheres with tunable chamber stucture

and enhanced photocatalytic activity. J Am Chem Soc 129:8406–8407. doi:10.1021/ja072191c

296. Cao SW, Zhu YJ (2008) Hierarchically nanostructured a-Fe2O3 hollow spheres: preparation,growth mechanism, photocatalytic property, and application in water treatment. J PhysChem C 112:6253–6257. doi:10.1021/jp8000465

297. Cao SW, Zhu YJ, Cheng GF, Huang YH (2010) Preparation and photocatalytic property ofa-Fe2O3 hollow core/shell hierarchical nanostructures. J Phys Chem Sol 71:1680–1683.doi:10.1016/j.jpcs.2010.09.006

298. Yu J, Liu S, Zhou M (2008) Enhanced photocatalytic activity of hollow anatasemicrospheres by Sn4+ incorporation. J Phys Chem C 112:2050–2057. doi:10.1021/jp0770007

299. Grimes CA (2007) Synthesis and application of hyghly ordered arrays of TiO2 nanotubes.J Mater Chem 17:1451–1457. doi:10.1039/B701168G

300. Bavykin DV, Friedrich JM, Walsh FC (2006) Protonated titanates and TiO2 nanostructuredmaterials: synthesis, properties, and applications. Adv Mater 18:2807–2824. doi:10.1002/adma.200502696

301. Bavykin DV, Walsch FC (2010) Titanate and titania nanotubes: synthesis, properties andapplications. RSC Nanoscience and Nanotechnology, N 12. R.S.C., London

302. Zhuang HF, Lin CJ, Lai YK et al (2007) Some critical structure factors of titanium dioxidenanotube array in its photocatalytic activity. Environ Sci Technol 41:4735–4740. doi:10.1021/es0702723

303. Zhang G, Huang H, Zhang Y et al (2007) Highly ordered nanoporous TiO2 and its photo–catalytic properties. Electrochem Commun 9:2854–2858. doi:10.1016/j.elecom.2007.10.014

304. Zhang Z, Yuan Y, Shi G et al (2007) Photoelectrocatalytic activity of highly ordered TiO2

nanotube arrays electrode for azo dye degradation. Environ Sci Technol 41:6259–6263.doi:10.1021/es070212x

300 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

305. Bae S, Shim E, Yoon J, Joo H (2008) Enzymatic hydrogen production by light-sensitizedanodized tubular TiO2 photoanode. Sol En Mater Sol Cells 92:402–409. doi:10.1016/j.solmat.2007.09.019

306. Wu X, Ling Y, Sun J et al (2008) Photoelectrocatalytic degradation of methylene blue dyeon titania nanotube array film. Chem Lett 37:416–417. doi:10.1246/cl.2008.416

307. Quan X, Ruan X, Zhao H et al (2007) Photoelectrocatalytic degraation of pentachlorophenolin aqueous solution using a TiO2 nanotube film electrode. Environ Pollution 147:409–414.doi:10.1016/j.envpol.2006.05.023

308. Bae S, Shim E, Yoon J, Joo H (2008) Photoanodic and cathodic role of anodized tubulartitania in light-sensitized enzymatic hydrogen production. J Power Sources 185:439–444.doi:10.1016/j.jpowsour.2008.06.094

309. Allam NK, Shankar K, Grimes CA (2008) Photoelectrochemical and water photoelectrolysisproperties of ordered TiO2 nanotubes fabricated by Ti anodization in fluoride-free HClelectrolytes. J Mater Chem 18:2341–2348. doi:10.1039/B718580D

310. Paulose M, Mor GK, Varghese OK et al (2006) Visible light photoelectrochemical andwater-photoelectrolysis properties of titania nanotube arrays. J Photochem Photobiol, A178:8–15. doi:10.1016/j.jphotochem.2005.06.013

311. Zlamal M, Macak JM, Schmuki P, Krysa J (2007) Electrochemically assisted photocatalysison self-organized TiO2 nanotubes. Electrochem Commun 9:2822–2826. doi:10.1016/j.elecom.2007.10.002

312. Mohapatra SK, Raja KS, Mahajan VK, Misra M (2008) Efficient photoelectrolysis of waterusing TiO2 nanotube arrays by minimizing recombination losses with organic additives.J Phys Chem C 112:11007–11012. doi:10.1021/jp7100539

313. Sohn YS, Smith YR, Misra M, Subramanian VR (2008) Electrochemically assistedphotocatalytic degradation of methyl orange using anodized titanium dioxide nanotubes.Appl Catal B 84:372–378. doi:10.1016/j.apcatb.2008.04.021

314. In S, Nielsen MG, Vesborg PCK et al (2011) Photocatalytic methane decomposition oververtically aligned transparent TiO2 nanotube arrays. Chem Commun 47:2613–2615. doi:10.1039/C0CC02570D

315. He X, Cai Y, Zhang H, Liang C (2011) Photocatalytic degradation of organic pollutants withAg decorated free-standing TiO2 nanotube arrays and interface electrochemical response.J Mater Chem 21:475–480. doi:10.1039/C0JM02404J

316. Zhang H, Liu P, Liu X (2010) Fabrication of highly ordered TiO2 nanorod/nanotubeadjacent arrays for photoelectrochemical applications. Langmuir 26:11226–11232. doi:10.1021/la1005314

317. Paek MJ, Kim TW, Hwang SJ (2008) Effects of hydronium intercalation and cationsubstitution on the photocatalytic performance of layered titanium oxide. J Phys Chem Sol69:1444–1446. doi:10.1016/j.jpcs.2007.09.012

318. Ohtani B, Ikeda S, Nakayama H, Nishimoto S (2000) Shape- and size-selectivephotocatalytic reactions by layered titanic acid powder suspended in deaerated aqueousalcohol solutions. Phys Chem Chem Phys 2:5308–5313. doi:10.1039/B006497L

319. Choy JH, Lee HC, Jung H, Hwang SJ (2011) A novel synthetic route to TiO2-pillaredtitanate with enhanced photocatalytic activity. J Mater Chem 11:2232–2234. doi:10.1039/B104551M

320. Sakai N, Ebina Y, Takada K, Sasaki T (2004) Electronic band structure of titaniasemiconductor nanosheets revealed by electrochemical and photoelectrochemical studies.J Am Chem Soc 126:5851–5858. doi:10.1021/ja0394582

321. Compton OC, Carroll EC, Kim JY et al (2007) Calcium niobate semiconductor nanosheetsas catalysts for photochemical hydrogen evolution from water. J Phys Chem C 111:14589–14592. doi:10.1021/jp0751155

322. Compton OC, Mullet CH, Chiang S, Osterloh FE (2008) A building block approach tophotochemical water-splitting catalysts based on layered niobate nanosheets. J Phys Chem C112:6202–6208. doi:10.1021/jp711589z

References 301

323. Zhou H, Sabio EM, Townsend TK et al (2010) Assembly of core-shell structures forphotocatalytic hydrogen evolution from aqueous methanol. Chem Mater 22:3362–3368.doi:10.1021/cm903839t

324. Kuwahara Y, Yamashita H (2011) Efficient photocatalytic degradation of organics diluted inwater and air using TiO2 designed with zeolites and mesoporous silica materials. J MaterChem 21:2407–2416. doi:10.1039/C0JM02741C

325. Yang X, Zhu H, Liu J et al (2008) A mesoporous structure for efficient photocatalysts:anatase nanocrystals attached to leached clay layers. Micropor Mesopor Mater 112:32–44.doi:10.1016/j.micromeso.2007.09.017

326. Nikazar M, Golivand M, Mahanpur K (2007) Clinoptylolite-supported titania as aphotoatalyst of disperse yellow 23 decomposition in water. Kin Catal 48:230–236

327. Miao S, Liu Z, Han B et al (2006) Synthesis and characterization of TiO2–montmorillonitenanocomposites and their application for removal of methylene blue. J Mater Chem 16:579–584. doi:10.1039/B511426H

328. Menesi J, Körösi L, Bazso E et al (2008) Photocatalytic oxidation of organic pollutants ontitania–clay composites. Chemosphere 70:538–542. doi:10.1016/j.chemosphere.2007.06.049

329. Liu S, Yang JH, Choi JH (2006) Microporous SiO2–TiO2 nanosols pillared montmorillonitefor photocatalytic decomposition of methyl orange. J Photochem Photobiol, A 179:75–80.doi:10.1016/j.jphotochem.2005.07.016

330. Zhang G, Ding X, Hu Y et al (2008) Photocatalytic degradation of 4BS dye by N, S-codopedTiO2 pillared montmorillonite photocatalysts under visible-light irradiation. J Phys Chem C112:17994–17997. doi:10.1021/jp803939z

331. Chen J, Liu X, Li G et al (2011) Synthesis and characterization of novel SiO2 and TiO2

co-pil–lared montmorillonite composite for adsorption and photocatalytic degradation ofhydrophobic organic pollutants in water. Catal Today 164:364–369. doi:10.1016/j.cattod.2010.11.014

332. Salama TM, Ali IO, Mohamed MM (2007) Synthesis and characterization of mordenitesencapsulated titania nanoparticles: photocatalytic degradation of meta-chlorophenol. J MolCatal A 273:198–210. doi:10.1016/j.molcata.2007.03.071

333. Sun Z, Chen Y, Ke Q et al (2002) Photocatalytic degradation of cationic azo dye by TiO2/bentonite nanocomposite. J Photochem Photobiol, A 149:169–174. doi:10.1016/S1010-6030(01)00649-9

334. Zhang GK, Ding XM, He FS et al (2008) Low-temperature synthesis and photocatalyticactivity of TiO2 pillared montmorillonite. Langmuir 24:1026–1030. doi:10.1021/la702649v

335. Zhao D, Zhou J, Liu N (2007) Surface characteristics and photoactivity of silver-modifiedpalygorskite clays coated with nanosized titanium dioxide nanoparticles. Mater Character58:249–255. doi:10.1016/j.matchar.2006.04.024

336. Koči K, Matejka V, Kovař P et al (2011) Comparison of the pure TiO2 and kaoli-nite/TiO2

composite as catalyst for CO2 photocatalytic reduction. Catal Today 161:105–109. doi:10.1016/j.cattod.2010.08.026

337. Wang WY, Irawan A, Ku Y (2008) Photocatalytic degradation of Acid Red 4 using atitanium dioxide membrane supported on a porous ceramic tube. Water Res 42:4725–4732.doi:10.1016/j.watres.2008.08.021

338. Dubey N, Labhsetwar NK, Devotta S, Rayalu SS (2007) Hydrogen evolution by watersplitting using novel composite zeolite-based photocatalyst. Catal Today 129:428–434.doi:10.1016/j.cattod.2006.09.041

339. Zhang J, Hu Y, Matsuoka M et al (2001) Relationship between the local structures oftitanium oxide photocatalysts and their reactivities in the decomposition of NO. J PhysChem B 105:8395–8398. doi:10.1021/jp012080e

340. Cosa G, Galletero MS, Fernandez L et al (2002) Tuning the photocatalytic activity oftitanium dioxide by encapsulation inside zeolites exemplified by the cases of thianthrenephotooxygenation and horseradish peroxidase photodeactivation. New J Chem 26:1448–1455. doi:10.1039/B201397E

302 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

341. Zheng S, Gao L, Zhang Q, Guo J (2000) Synthesis, characterization and photocatalyticproperties of titania-modifed mesoporous silicate MCM-41. J Mater Chem 10:723–727.doi:10.1039/A908799K

342. Shankar H, Rajasudha G, Karthikeyan A et al (2008) Synthesis, characterization andphotocatalytic activity of nanotitania loaded W-MCM-41. Nanotechnology 19:315711

343. Bouazza N, Lillo-Rodenas MA, Linares-Solano A (2008) Photocatalytic activity of TiO2-based materials for the oxidation of propene and benzene at low concentration in presence ofhumidity. Appl Catal B 84:691–698. doi:10.1016/j.apcatb.2008.06.002

344. Zheng S, Gao L (2002) Synthesis and characterization of Pt, Au or Pd clusters depositedtitania-modified mesoporous silicate MCM-41. Mater Chem Phys 78:512–517. doi:10.1016/S0254-0584(02)00353-X

345. Reddy EP, Sun B, Smirniotis PG (2004) Transition metal modified TiO2-loaded MCM-41catalysts for visible- and UV-light driven photodegradation of aqueous organic pollutants.J Phys Chem B 108:17198–17205. doi:10.1021/jp047419m

346. De Witte K, Busuioc AM, Meyen V et al (2008) Influence of the synthesis parameters ofTiO2–SBA-15 materials on the adsorption and photodegradation of rhodamine-6G.Micropor Mesopor Mater 110:100–110. doi:10.1016/j.micromeso.2007.09.035

347. Phanikrishna Sharma MV, Durga Kumari V, Subrahmanyam M (2008) TiO2 supported overSBA-15: an efficient photocatalyst for the pesticide degradation using solar light.Chemosphere 73:1562–1569. doi:10.1016/j.chemosphere.2008.07.081

348. Panpa W, Sujaridworakun P, Jinawath S (2008) Photocatalytic activity of TiO2/ZSM-5composites in the presence of SO4

2− ion. Appl Catal B 80:271–276. doi:10.1016/j.apcatb.2007.11.029

349. Neren Ökte A, Yilmaz Ö (2008) Photodecolorization of methyl orange by yttriumincorporated TiO2 supported ZSM-5. Appl Catal B 85:92–102. doi:10.1016/j.apcatb.2008.07.025

350. Kitano M, Matsuoka M, Ueshima M, Anpo M (2007) Recent developments in titaniumoxide-based photocatalysts. Appl Catal A 325:1–14. doi:10.1016/j.apcata.2007.03.013

351. Shen S, Guo L (2006) Structural, textural and photocatalytic properties of quantum-sizedIn2S3-sensitized Ti-MCM-41 prepared by ion-exchange and sulfidation methods. J Sol StateChem 179:2629–2635. doi:10.1016/j.jssc.2006.05.010

352. Shen S, Guo L (2008) Growth of quantum-confined CdS nanoparticles inside Ti-MCM-41 asa visible light photocatalyst. Mater Res Bull 43:437–446. doi:10.1016/j.materresbull.2007.02.034

353. Sathish M, Viswanathan B, Viswanath RP (2006) Alternate synthetic strategy for thepreparation of CdS nanoparticles and its exploitation for water splitting. Inter J Hydrogen En31:891–898. doi:10.1016/j.ijhydene.2005.08.002

354. Ryu SY, Choi J, Balcerski W et al (2007) Photocatalytic production of H2 on nanocompositePhotocatalysts. Ind Eng Chem Res 46:7476–7488. doi:10.1021/ie0703033

355. Ryu SY, Balcerski W, Lee TK, Hoffmann MR (2007) Photocatalytic production of hydrogenfrom water with visible light using hybrid catalysts of CdS attached to microporous andmesoporous silicas. J Phys Chem C 111:18195–18203. doi:10.1021/jp074860e

356. Hirai T, Nanba M, Komasawa I (2003) Dithiol-mediated incorporation of CdS nanoparticlesfrom reverse micellar system into Zn-doped SBA-15 mesoporous silica and theirphotocatalytic properties. J Colloid Interface Sci 268:394–399. doi:10.1016/j.jcis.2003.09.011

357. Wang S, Liu P, Wang X, Fu X (2005) Homogeneously distributed CdS nanoparticles inNafion membranes: preparation, characterization, and photocatalytic properties. Langmuir21:11969–11973. doi:10.1021/la051072c

358. Pathak P, Meziani MJ, Li Y et al (2004) Improving photoreduction of CO2 with homogene–ously dispersed nanoscale TiO2 catalysts. Chem Commun 1234–1235. doi:10.1039/B400326H

359. Pathak P, Meziani MJ, Castillo L, Sun YP (2005) Metal-coated nanoscale TiO2 catalysts forenhanced CO2 photoreduction. Green Chem 7:667–670. doi:10.1039/B507103H

References 303

360. Lunawat PS, Senapati S, Kumar R, Gupta NM (2007) Visible light-induced splitting ofwater using CdS nanocrystallites immobilized overwater-repellent polymeric surface. Inter JHydrogen En 32:2784–2790. doi:10.1016/j.ijhydene.2007.04.001

361. Roy AM, De GC (2003) Immobilisation of CdS, ZnS and mixed ZnS–CdS on filter paper.Effect of hydrogen production from alkaline Na2S/Na2S2O3 solution. J PhotochemPhotobiol, A 157:87–92. doi:10.1016/S1010-6030(02)00430-6

362. Yi X, Liqin D, Lizhen A et al (2008) Photocatalytic degradation of rhodamine B and phenolby TiO2 loaded on mesoporous graphitic carbon. Chin J Catal 29:31–36. doi:10.1016/S1872-2067(08)60014-5

363. Zhang X, Lu X, Shen Y et al (2011) Three-dimentional WO3 nanostructures on carbonpaper: photoelectrochemical property and visible light driven photocatalysis. ChemCommun 47:5804–5806. doi:10.1039/C1CC10389J

364. Zhang X, Du AJ, Lee P et al (2008) Grafted multifunctional titanium dioxide nanotubemembrane: Separation and photodegradation of aquatic pollutant. Appl Catal B 84:262–267.doi:10.1016/j.apcatb.2008.04.009

365. Plesch G, Gorbar M, Vogt UF et al (2009) Reticulated macroporous ceramic foam supportedTiO2 for photocatalytic applications. Mater Lett 63:461–463. doi:10.1016/j.matlet.2008.11.008

366. Skorb EV, Ustinovich EA, Kulak AI, Sviridov DV (2008) Photocatalytic activity of TiO2:In2O3 nanocomposite films towards the degradation of arylmethane and azo dyes.J Photochem Photobiol, A 193:97–102. doi:10.1016/j.jphotochem.2007.06.012

367. Fujishima A, Rao TN, Tryk DA (2000) Titanium dioxide photocatalysis. J PhotochemPhotobiol, C 1:1–21. doi:10.1016/S1389-5567(00)00002-2

368. Linsebigler AL, Lu G, Yates JT (1995) Photocatalysis on TiO2 Surfaces: principles,mechanisms, and selected results. Chem Rev 95:735–758. doi:10.1021/cr00035a013

369. Nowotny MK, Sheppard LR, Bak T, Nowotny J (2008) Defect chemistry of titaniumdioxide. Application of defect engineering in processing of TiO2-based photocatalysts.J Phys Chem C 112:5275–5300. doi:10.1021/jp077275m

370. Anpo M (2004) Preparation, characterization, and reactivities of highly functional titaniumoxide-based photocatalysts able to operate under UV–visible light irradiation: approaches inrealizing high efficiency in the use of visible light. Bull Chem Soc Jpn 77:1427–1442.doi:10.1246/bcsj.77.1427

371. Liu G, Wang L, Yang HG et al (2010) Titania-based photocatalysts—crystal growth, dopingand heterostructuring. J Mater Chem 20:831–843. doi:10.1039/B909930A

372. Vinu R, Madras G (2008) Synthesis and photoactivity of Pd substituted nano-TiO2. J MolCatal A 291:5–11. doi:10.1016/j.molcata.2008.07.005

373. Xiao Q, Si Z, Yu Z, Qiu G (2008) Characterization and photocatalytic activity of Sm3+-doped TiO2 nanocrystalline prepared by low temperature combustion method. J AlloysComp 450:426–431. doi:10.1016/j.jallcom.2006.10.137

374. Zhou J, Takeuchi M, Zhao XS et al (2006) Photocatalytic decomposition of formic acidunder visible light irradiation over V-ion-implanted TiO2 thin film photocatalysts preparedon quartz substrate by ionized cluster beam (ICB) deposition method. Catal Lett 106:67–70.doi:10.1007/s10562-005-9192-5

375. Zhou J, Takeuchi M, Ray AK et al (2007) Enhancement of photocatalytic activity of P25TiO2 by vanadium-ion implantation under visible light irradiation. J Colloid Interface Sci311:497–501. doi:10.1016/j.jcis.2007.03.007

376. Takeuchi M, Yamashita H, Matsuoka M et al (2000) Photocatalytic decomposition of NOunder visible light irradiation on the Cr-ion-implanted TiO2 thin film photocatalyst. CatalLett 67:135–137. doi:10.1023/A:1019065521567

377. Ventkatachalam N, Palanichamy M, Arabindoo B, Murugesan V (2007) Alkaline earth metaldoped nanoporous TiO2 for enhanced photocatalytic mineralisation of bisphenol-A. CatalCommun 8:1088–1093. doi:10.1016/j.catcom.2006.10.025

378. Venkatachalam N, Palanichamy M, Murugesan V (2007) Sol–gel preparation andcharacterization of alkaline earth metal doped nano TiO2: Efficient photocatalytic

304 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

degradation of 4-chlorophenol. J Mol Catal A 273:177–185. doi:10.1016/j.molcata.2007.03.077

379. Kryukov AI, Kuchmiy SY, Pokhodenko VD (1997) Nanostructured composite photocat-alysts based on polycrystalline cadmium sulfide. Theoret Exp Chem 33:306–321

380. Yin S, Aita Y, Komatsu M, Sato T (2006) Visible-light-induced photocatalytic activity ofTiO2−xNy prepared by solvothermal process in urea–alcohol system. J Eur Ceram Soc26:2735–2742. doi:10.1016/j.jeurceramsoc.2005.05.012

381. Gandhe AR, Fernandes JB (2005) A simple method to synthesize N-doped rutile titania withenhanced photocatalytic activity in sunlight. J Sol State Chem 178:2953–2957. doi:10.1016/j.jssc.2005.06.034

382. Xu JH, Dai WL, Li J et al (2008) Simple fabrication of thermally stable apertured N-dopedTiO2 microtubes as a highly efficient photocatalyst under visible light irradiation. CatalCommun 9:146–152. doi:10.1016/j.catcom.2007.05.043

383. Matsumoto T, Iyi N, Kaneko Y et al (2007) High visible-light photocatalytic activity ofnitrogen-doped titania prepared from layered titania/isostearate nanocomposite. Catal Today120:226–232. doi:10.1016/j.cattod.2006.07.047

384. Li Y, Xie C, Peng S et al (2008) Eosin Y-sensitized nitrogen-doped TiO2 for efficient visiblelight photocatalytic hydrogen evolution. J Mol Catal A 282:117–123. doi:10.1016/j.molcata.2007.12.005

385. Gao B, Ma Y, Cao Y et al (2006) Great enhancement of photocatalytic activity ofnitrogen-doped titania by coupling with tungsten oxide. J Phys Chem B 110:14391–14397.doi:10.1021/jp0624606

386. Li H, Li J, Huo Y (2006) Highly active TiO2N photocatalysts prepared by treating TiO2

precursors in NH3/ethanol fluid under supercritical conditions. J Phys Chem B 110:1559–1565. doi:10.1021/jp055830j

387. Rhee CH, Bae SW, Lee JS (2005) Template-free hydrothermal synthesis of high surface areanitrogen-doped titania photocatalyst active under visible light. Chem Lett 34:660–661.doi:10.1246/cl.2005.660

388. Livraghi S, Chierotti MR, Giamello E et al (2008) Nitrogen-doped titanium dioxide active inphotocatalytic reactions with visible light: a multi-technique characterization of differentlyprepared materials. J Phys Chem C 112:17244–17252. doi:10.1021/jp803806s

389. Sun H, Bai Y, Liu H et al (2009) Photocatalytic decomposition of 4-chlorophenol over anefficient N-doped TiO2 under sunlight irradiation. J Photochem Photobiol, A 201:15–22.doi:10.1016/j.jphotochem.2008.08.021

390. Sun H, Bai Y, Jin W, Xu N (2008) Visible-light-driven TiO2 catalysts doped with low-con–centration nitrogen species. Sol En Mater Sol Cells 92:76–83. doi:10.1016/j.solmat.2007.09.003

391. Li D, Huang H, Chen X et al (2007) New synthesis of excellent visible-light TiO2 − xNx

photocatalyst using a very simple method. J Sol State Chem 180:2630–2634. doi:10.1016/j.jssc.2007.07.009

392. Cheng P, Deng C, Gu M, Dai X (2008) Effect of urea on the photoactivity of titania powderprepared by sol–gel method. Mater Chem Phys 107:77–81. doi:10.1016/j.matchemphys.2007.06.051

393. Yuan J, Chen M, Shi J, Shanguang W (2006) Preparations and photocatalytic hydrogenevolution of N-doped TiO2 from urea and titanium tetrachloride. Inter J Hydrogen En31:1326–1331. doi:10.1016/j.ijhydene.2005.11.016

394. Bacsa R, Kiwi J, Ohno T et al (2005) Preparation, testing and characterization of doped TiO2

active in the peroxidation of biomolecules under visible light. J Phys Chem B 109:5994–6003. doi:10.1021/jp044979c

395. Sreethawong T, Laehsalee S, Chavadej S (2008) Comparative investigation of mesoporous-and non-mesoporous-assembled TiO2 nanocrystals for photocatalytic H2 production overN-doped TiO2 under visible light irradiation. Inter J Hydrogen En 33:5947–5957. doi:10.1016/j.ijhydene.2008.08.007

References 305

396. Wang Y, Zhou G, Li T et al (2009) Catalytic activity of meso-porous TiO2 − xNx

photocatalysts for the decomposition of methyl orange under solar simulated light. CatalCommun 10:412–415. doi:10.1016/j.catcom.2008.10.007

397. Chen D, Yang D, Geng J et al (2008) Improving visible-light photocatalytic activity ofN-doped TiO2 nanoparticles via sensitization by Zn porphyrin. Appl Surf Sci 255:2879–2884. doi:10.1016/j.apsusc.2008.08.032

398. Gole JL, Stout JD, Burda C et al (2004) Highly efficient formation of visible light tunableTiO2−xNx photocatalysts and their transformation at the nanoscale. J Phys Chem B108:1230–1240. doi:10.1021/jp030843n

399. Jagadale TC, Takale SP, Sonawane RS et al (2008) N-doped TiO2 nanoparticle based visiblelight photocatalyst by modified peroxide sol–gel method. J Phys Chem C 112:14595–14602.doi:10.1021/jp803567f

400. Xu J, Ao Y, Fu D, Yuan C (2008) A simple route to synthesize highly crystalline N-dopedTiO2 particles under low temperature. J Cryst Growth 310:4319–4324. doi:10.1016/j.jcrysgro.2008.07.045

401. Zhao Y, Qiu X, Burda C (2008) The Effects of sintering on the photocatalytic activity ofN-doped TiO2 nanoparticles. Chem Mater 20:2629–2636. doi:10.1021/cm703043j

402. Lopez-Luke T, Wolcott A, Xu L et al (2008) Nitrogen-doped and CdSequantum-dot-sensitized nanocrystalline TiO2 films for solar energy conversion applications.J Phys Chem C 112:1282–1292. doi:10.1021/jp077345p

403. Sano T, Negishi N, Koike K et al (2004) Preparation of a visible light-responsivephotocatalyst from a complex of Ti4+ with a nitrogen-containing ligand. J Mater Chem14:380–384. doi:10.1039/B311444A

404. Feng C, Wang Y, Jin Z et al (2008) Photoactive centers responsible for visible-lightphotoactivity of N-doped TiO2. New J Chem 32:1038–1047. doi:10.1039/B719498F

405. Ozaki H, Iwamoto S, Inoue M (2007) Marked promotive effect of iron onvisible-light-induced photocatalytic activities of nitrogen- and silicon-codoped titanias.J Phys Chem C 111:17061–17066. doi:10.1021/jp0751211

406. Yang S, Gao L (2008) Photocatalytic activity of nitrogen doped rutile TiO2 nanoparticlesunder visible light irradiation. Mater Res Bull 43:1872–1876. doi:10.1016/j.materresbull.2007.06.058

407. Morikawa T, Ohwaki T, Suzuki K et al (2008) Visible-light-induced photocatalyticoxidation of carboxylic acids and aldehydes over N-doped TiO2 loaded with Fe, Cu or Pt.Appl Catal B 83:56–62. doi:10.1016/j.apcatb.2008.01.034

408. Yin S, Liu B, Zhang P et al (2008) Photocatalytic oxidation of NOx under visible LED lightirradiation over nitrogen-doped titania particles with iron or platinum loading. J PhysChem C 112:12425–12431. doi:10.1021/jp803371s

409. Shouxin L, Xiaoyun C, Xi C (2006) Preparation of N-doped visible-light response nanosizeTiO2 photocatalyst using the acid-catalyzed hydrolysis method. Chin J Catal 27:697–702.doi:10.1016/S1872-2067(06)60037-5

410. Balcerski W, Ryu SY, Hoffmann MR (2007) Visible-light photoactivity of nitrogen-dopedTiO2: photo-oxidation of HCO2H to CO2 and H2O. J Phys Chem C 111:15357–15362.doi:10.1021/jp074989o

411. Abe H, Kimitani T, Naito M (2006) Influence of NH3/Ar plasma irradiation on physical andphotocatalytic properties of TiO2 nanopowder. J Photochem Photobiol, A 183:171–175.doi:10.1016/j.jphotochem.2006.03.013

412. Sano T, Puzenat E, Guillard C et al (2008) Degradation of C2H2 with modified-TiO2

photocatalysts under visible light irradiation. J Mol Catal A 284:127–133. doi:10.1016/j.molcata.2008.01.014

413. Yin S, Yamaki H, Komatsu M et al (2003) Preparation of nitrogen-doped titania with highvisible light induced photocatalytic activity by mechanochemical reaction of titania andhexamethylenetetramine. J Mater Chem 13:2996–3001. doi:10.1039/B309217H

306 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

414. Shen H, Mi L, Xu P et al (2007) Visible-light photocatalysis of nitrogen-doped TiO2

nanoparticulate films prepared by low-energy ion implantation. Appl Surf Sci 253:7024–7028. doi:10.1016/j.apsusc.2007.02.023

415. Cui X, Ma M, Zhang W et al (2008) Nitrogen-doped TiO2 from TiN and its visible lightphotoelectrochemical properties. Electrochem Commun 10:367–371. doi:10.1016/j.elecom.2007.12.037

416. Liu B, Wen L, Zhao X (2008) The structure and photocatalytic studies of N-doped TiO2

films prepared by radio frequency reactive magnetron sputtering. Sol En Mater Sol Cells92:1–10. doi:10.1016/j.solmat.2007.07.009

417. Lin HF, Liao SC, Hung SW (2005) The dc thermal plasma synthesis of ZnO nanoparticlesfor visible-light photocatalyst. J Photochem Photobiol, A 174:82–87. doi:10.1016/j.jphotochem.2005.02.015

418. Wang H, Zhang Z, Chang W (2008) A novel one-step photocatalytic synthesis of benzo[d]oxazol-2(3H)-one with C-doped TiO2 nanoparticle. Chem Lett 37:1156–1157. doi:10.1246/cl.2008.1156

419. Huang Y, Ho W, Lee S et al (2008) Effect of carbon doping on the mesoporous structure ofnanocrystalline titanium dioxide and its solar-light-driven photocatalytic degradation ofNOx. Langmuir 24:3510–3516. doi:10.1021/la703333z

420. Treschev SY, Chou PW, Tseng YH et al (2008) Photoactivities of the visible-light-activatedmixed-phase carbon-containing titanium dioxide: the effect of carbon incorporation. ApplCatal B 79:8–16. doi:10.1016/j.apcatb.2007.09.046

421. Wang X, Meng S, Zhang X et al (2007) Multi-type carbon doping of TiO2 photocatalyst.Chem Phys Lett 444:292–296. doi:10.1016/j.cplett.2007.07.026

422. Mohapatra SK, Misra M, Mahajan VK, Raja KS (2007) Design of a highly efficientphotoelectrolytic cell for hydrogen generation by water splitting: application of TiO2 − xCx

nanotubes as a photoanode and Pt/TiO2 nanotubes as a cathode. J Phys Chem C 111:8677–8685. doi:10.1021/jp071906v

423. Hu X, Zhang T, Jin Z et al (2008) Fabrication of carbon-modified TiO2 nanotube arrays andtheir photocatalytic activity. Mater Lett 62:4579–4581. doi:10.1016/j.matlet.2008.08.051

424. Xiao Q, Zhang J, Xiao C et al (2008) Solar photocatalytic degradation of methylene blue incarbon-doped TiO2 nanoparticles suspension. Sol En 82:706–713. doi:10.1016/j.solener.2008.02.006

425. Randeniya LK, Murphy AB, Plumb IC (2008) A study of S-doped TiO2 for photoelec-trochemical hydrogen generation from water. J Mater Sci 43:1389–1399. doi:10.1007/s10853-007-2309-z

426. Xu J, Ao Y, Fu D, Yuan C (2008) Low-temperature preparation of F-doped TiO2 film and itsphotocatalytic activity under solar light. Appl Surf Sci 254:3033–3038. doi:10.1016/j.apsusc.2007.10.065

427. Xu J, Ao Y, Fu D, Yuan C (2008) Synthesis of fluorine-doped titania-coated activatedcarbon under low temperature with high photocatalytic activity under visible light. J PhysChem Sol 69:2366–2370. doi:10.1016/j.jpcs.2008.03.017

428. Zhou JK, Lv L, Yu J et al (2008) Synthesis of self-organized polycrystalline F-doped TiO2

hollow microspheres and their photocatalytic activity under visible light. J Phys Chem C112:5316–5321. doi:10.1021/jp709615x

429. Wu G, Chen A (2008) Direct growth of F-doped TiO2 particulate thin films with highphotocatalytic activity for environmental applications. J Photochem Photobiol, A 195:47–53.doi:10.1016/j.jphotochem.2007.09.005

430. Wu G, Wang J, Thomas DF, Chen A (2008) Synthesis of F-doped flower-like TiO2

nanostructures with high photoelectrochemical activity. Langmuir 24:3503–3509. doi:10.1021/la703098g

431. Khan R, Kim SW, Kim TJ, Nam CM (2008) Comparative study of the photocatalyticperformance of boron-iron Co-doped and boron-doped TiO2 nanoparticles. Mater ChemPhys 112:167–172. doi:10.1016/j.matchemphys.2008.05.030

References 307

432. Yu HF (2007) Photocatalytic abilities of gel-derived P-doped TiO2. J Phys Chem Sol68:600–607. doi:10.1016/j.jpcs.2007.01.050

433. Lin L, Lin W, Zhu Y et al (2005) Phosphor-doped titania—a novel photocatalyst active invisible light. Chem Lett 34:284–285. doi:10.1246/cl.2005.284

434. Zaleska A, Sobczak JW, Grabowska E, Hupka J (2008) Preparation and photocatalyticactivity of boron-modified TiO2 under UV and visible light. Appl Catal B 78:92–100.doi:10.1016/j.apcatb.2007.09.005

435. Li J, Lu N, Quan X et al (2008) Facile method for fabricating boron-doped TiO2 nanotubearray with enhanced photoelectrocatalytic properties. Ind Eng Chem Res 47:3804–3808.doi:10.1021/ie0712028

436. In S, Orlov A, Berg R et al (2007) Effective visible light-activated B-doped and B,N-codoped TiO2 photocatalysts. J Am Chem Soc 129:13790–13791. doi:10.1021/ja0749237

437. Tristao JC, Magalhaes F, Corio P, Sansiviero MTC (2006) Electronic characterization andphotocatalytic properties of CdS/TiO2 semiconductor composite. J Photochem Photobiol, A181:152–157. doi:10.1016/j.jphotochem.2005.11.018

438. Demeestere K, Dewulf J, Ohno T et al (2005) Visible light mediated photocatalyticdegradation of gaseous trichloroethylene and dimethyl sulfide on modified titanium dioxide.Appl Catal B 61:140–149. doi:10.1016/j.apcatb.2005.04.017

439. Yin H, Wada Y, Kitamura T et al (2001) Enhanced photocatalytic dechlorination of1,2,3,4-tetrachlorobenzene using nanosized CdS/TiO2 hybrid photocatalyst under visiblelight irradiation. Chem Lett 30:334–335. doi:10.1246/cl.2001.334

440. Blackburn JL, Selmarten DC, Nozik AJ (2003) Electron transfer dynamics in quantumdot/titanium dioxide composites formed by in situ chemical bath deposition. J Phys Chem B107:14154–14157. doi:10.1021/jp0366771

441. Vogel R, Hoyer P, Weller H (1994) Quantum-sized PbS, CdS, Ag2S, Sb2S3, and Bi2S3particles as sensitizers for various nanoporous wide-bandgap semiconductors. J Phys Chem98:3183–3188. doi:10.1021/j100063a022

442. Hong JS, Choi DS, Kang MG et al (2001) Photocurrent instability of PbS-sensitized TiO2

electrodes in S2− and SO32− solution. J Photochem Photobiol, A 143:87–92. doi:10.1016/

S1010-6030(01)00455-5443. Sun WT, Yu T, Pan HY et al (2008) CdS quantum dots sensitized TiO2 nanotube-array

photoelectrodes. J Am Chem Soc 130:1124–1125. doi:10.1021/ja0777741444. Shalom M, Dor S, Rühle S et al (2009) Core/CdS quantum dot/shell mesoporous solar cells

with improved stability and efficiency using an amorphous TiO2 coating. J Phys Chem C113:3895–3898. doi:10.1021/jp8108682

445. Das K, De SK (2009) Optical properties of the type-II core-shell TiO2@CdS nanorods forphotovoltaic applications. J Phys Chem C 113:3494–3501. doi:10.1021/jp8083953

446. Chen H, Fu W, Yang H et al (2010) Photosensitization of TiO2 nanorods with CdS quantumdots for photovoltaic devices. Electrochim Acta 56:919–924. doi:10.1016/j.electacta.2010.10.003

447. Xia M, Wang F, Wang Y et al (2010) TiO2 nanowires sensitized with CdS quantum dots andthe surface photovoltage properties. Mater Lett 64:1688–1690. doi:10.1016/j.matlet.2010.05.003

448. Nayak J, Sahu SN, Kasuya J, Nozaki S (2008) CdS–ZnO composite nanorods: synthesis,characterization and application for photocatalytic degradation of 3,4-dihydroxy benzoicacid. Appl Surf Sci 254:7215–7218. doi:10.1016/j.apsusc.2008.05.268

449. Lee W, Min SK, Dhas V et al (2009) Chemical bath deposition of CdS quantum dots onvertically aligned ZnO nanorods for quantum dots-sensitized solar cells. ElectrochemCommun 11:103–106. doi:10.1016/j.elecom.2008.10.042

450. Zhang YJ, Yan W, Wu YP, Wang ZH (2008) Synthesis of TiO2 nanotubes coupled withCdS nanoparticles and production of hydrogen by photocatalytic water decomposition.Mater Lett 62:3846–3848. doi:10.1016/j.matlet.2008.04.084

308 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

451. Xing C, Jing D, Liu M, Guo L (2009) Photocatalytic hydrogen production overNa2Ti2O4(OH)2 nanotube sensitized by CdS nanoparticles. Mater Res Bull 44:442–445.doi:10.1016/j.materresbull.2008.04.016

452. Peter LM, Upul Wijayantha KG, Jason Riley D, Waggett JP (2003) Band-edge tuning inself-assembled layers of Bi2S3 nanoparticles used to photosensitize nanocrystalline TiO2.J Phys Chem B 107:8378–8381. doi:10.1021/jp030334l

453. Smith YR, Subramanian VR (2011) Heterostructural composites of TiO2 mesh-TiO2

nanoparticles photosensitized with CdS: a new flexible photoanode for solar cells. J PhysChem B 115:8376–8385. doi:10.1021/jp110200s

454. Lin ZQ, Lai YK, Hu RG et al (2010) A highly efficient ZnS/CdS@TiO2 photo-electrode forphotogenerated cathodic protection of metals. Electrochim Acta 55:8717–8723. doi:10.1016/j.electacta.2010.08.017

455. Gao C, Li J, Shan Z et al (2010) Preparation ans visible-light photocatalytic activity of In2S3/TiO2 composite. Mater Chem Phys 122:183–187. doi:10.1016/j.matchemphys.2010.02.030

456. Patil RS, Lokhande CD, Mane RS et al (2007) Room temperature PbS nanoparticle growth,incubation in porous TiO2 electrode for photosensitization application. J Non-Cryst Sol353:1645–1649. doi:10.1016/j.jnoncrysol.2007.01.014

457. Lee W, Lee J, Min SK et al (2009) Effect of single-walled carbon nanotube in PbS/TiO2

quantum dots-sensitized solar cells. Mater Sci Eng, B 156:48–51. doi:10.1016/j.mseb.2008.11.014

458. Ratanatawanate C, Bui A, Vu K, Balkus KJ (2011) Low-temperature synthesis of copper(II) sulfide quantum dot decorated TiO2 nanotubes and their photocatalytic properties. J PhysChem C 115:6175–6180. doi:10.1021/jp109716q

459. Liu Y, Zhou H, Zhou B et al (2011) Highly stable CdS-modified short TiO2 nanotube arrayelectrode for efficient visible-light hydrogen generation. Inter J Hydrogen En 36:167–174.doi:10.1016/j.ijhydene.2010.09.089

460. Seoul M, Kim H, Kim W, Yong K (2010) Highly efficient photoelectrochemical hydrogengeneration using a ZnO nanowire array and a CdSe/CdS co-sensitizer. ElectrochemCommun 12:1416–1418. doi:10.1016/j.elecom.2010.07.035

461. Sasamura T, Okazaki K, Tsunoda R et al (2010) Immobilization of ZnS-AgInS2 solidsolition nanoparticles on ZnO rod array electrodes and their photoresponse with visible lightirradiation. Chem Lett 39:619–620. doi:10.1246/cl.2010.619

462. Liu C, Liu Z, Li Y et al (2011) CdS/PbS co-sensitized ZnO nanorods and its photocatalyticproperties. Appl Surf Sci 257:7041–7046. doi:10.1016/j.apsusc.2011.02.133

463. Huang S, Zhang Q, Huang X et al (2010) Fibrous CdS/CdSe quantum dot co-sensitized solarcells based on ordered TiO2 nanotube arrays. Nanotechnology 21:375201

464. Anower Hossain M, Jennings JR, Koh ZY, Wang Q (2011) Carrier generation and collectionin CdS/CdSe-sensitized SnO2 solar cells exhibiting unprecedented photocurrent densities.ACS Nano 5:3172–3181. doi:10.1021/nn200315b

465. Bessekhouad Y, Chaoui N, Trzpit M et al (2006) UV-vis versus visible degradation of AcidRed II in a coupled CdS/TiO2 semiconductors suspension. J Photochem Photobiol, A183:218–224. doi:10.1016/j.jphotochem.2006.03.025

466. Jang JS, Kim HG, Joshi UA et al (2008) Fabrication of CdS nanowires decorated with TiO2

nanoparticles for photocatalytic hydrogen production under visible light irradiation. Inter JHydrogen En 33:5975–5980. doi:10.1016/j.ijhydene.2008.07.105

467. Jang JS, Choi SH, Kim HG, Lee JS (2008) Location and state of Pt in platinized CdS/TiO2

photocatalysts for hydrogen production from water under visible light. J Phys Chem C112:17200–17205. doi:10.1021/jp804699c

468. Lawless D, Kapoor S, Meisel D (1995) Bifunctional capping of CdS nanoparticles andbridging to TiO2. J Phys Chem 99:10329–10335. doi:10.1021/j100025a040

469. Dibbell RS, Watson DF (2009) Distance-dependent electron transfer in tethered assembliesof CdS quantum dots and TiO2 nanoparticles. J Phys Chem C 113:3139–3149. doi:10.1021/jp809269m

References 309

470. Hirai T, Suzuki K, Komasawa I (2001) Preparation and photocatalytic properties ofcomposite CdS nanoparticles—titanium dioxide particles. J Colloid Interface Sci 244:262–265. doi:10.1006/jcis.2001.7982

471. Peter LM, Jason Riley D, Tull EJ, Upul Wijayantha KG (2002) Photosensitization ofnanocrystalline TiO2 by self-assembled layers of CdS quantum dots. Chem Commun 1030–1031. doi:10.1039/B201661C

472. Lee YL, Huang BM, Chien HT (2008) Highly efficient CdSe-sensitized TiO2 photoelectrodefor quantum-dot-sensitized solar cell applications. Chem Mater 20:6903–6905. doi:10.1021/cm802254u

473. Wu J, Lin J, Yin S, Sato T (2001) Synthesis and photocatalytic properties of layeredHNbWO6/(Pt, Cd0.8Zn0.2S) nanocomposites. J Mater Chem 11:3343–3347. doi:10.1039/B103838A

474. Shangguan W (2007) Hydrogen evolution from water splitting on nanocompositephotocatalysts. Sci Technol Adv Mater 8:76–81. doi:10.1016/j.stam.2006.09.007

475. Stengl V, Bakarjieva S, Murafa N et al (2008) Visible-light photocatalytic activity of TiO2/ZnS nanocomposites prepared by homogeneous hydrolysis. Micropor Mesopor Mater110:370–378. doi:10.1016/j.micromeso.2007.06.052

476. Kumar A, Jain AK (2003) Photophysics and photocatalytic properties of Ag+-activatedsandwich Q-CdS–TiO2. J Photochem Photobiol, A 156:207–218. doi:10.1016/S1010-6030(02)00396-9

477. Yu JC, Wu L, Lin J et al (2003) Microemulsion-mediated solvothermal synthesis ofnanosized CdS-sensitized TiO2 crystalline photocatalyst. Chem Commun 1552–1553.doi:10.1039/B302418K

478. Evans JE, Springer KW, Zhang JZ (1994) Femtosecond studies of interparticle electrontransfer in a coupled CdS-TiO2 colloidal system. J Chem Phys 101:6222–6225. doi:10.1063/1.468376

479. Gopidas KR, Bohorquez M, Kamat PV (1990) Photophysical and photochemical aspects ofcoupled semiconductors. Charge-transfer processes in colloidal CdS-TiO2 and CdS-AgIsystems. J Phys Chem 94:6435–6440. doi:10.1021/j100379a051

480. Hotchandani S, Kamat PV (1992) Charge-transfer processes in coupled semiconductorsystems. Photochemistry and photoelectrochemistry of the colloidal CdS-ZnO system.J Phys Chem 96:6834–6839. doi:10.1021/j100195a056

481. So WW, Kim KJ, Moon SJ (2004) Photoproduction of hydrogen over the CdS–TiO2

nano-composite particulate flms treated with TiCl4. Inter J Hydrogen En 29:229–234. doi:10.1016/S0360-3199(03)00211-8

482. Chen S, Paulose M, Ruan C et al (2006) Electrochemically synthesized CdSnanoparticle-modified TiO2 nanotube-array photoelectrodes: Preparation, characterization,and application to photoelectrochemical cells. J Photochem Photobiol, A 177:177–184.doi:10.1016/j.jphotochem.2005.05.023

483. Seabold JA, Shankar K, Wilke RHT et al (2008) Photoelectrochemical properties ofheterojunction CdTe/TiO2 electrodes constructed using highly ordered TiO2 nanotubearrays. Chem Mater 20:5266–5273. doi:10.1021/cm8010666

484. Lu ZX, Zhang ZL, Zhang MX et al (2005) Core/shell quantum-dot-photosensitizednano-TiO2 films: Fabrication and application to the damage of cells and DNA. J PhysChem B 109:22663–22666. doi:10.1021/jp054472h

485. Diguna LJ, Shen Q, Sato A et al (2007) Optical absorption and ultrafast carrier dynamicscharacterization of CdSe quantum dots deposited on different morphologies of nanostruc-tured TiO2 films. Mater Sci Eng, C 27:1514–1520. doi:10.1016/j.msec.2006.06.036

486. Shen Q, Katayama K, Yamaghuchi M et al (2005) Study of ultrafast carrier dynamics ofnanostructured TiO2 films with and without CdSe quantum dot deposition using lens-freeheterodyne detection transient grating technique. Thin Solid Films 486:15–19. doi:10.1016/j.tsf.2004.11.238

310 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

487. Shen Q, Sato T, Hashimoto M et al (2006) Photoacoustic and photoelectrochemicalcharacterization of CdSe-sensitized TiO2 electrodes composed of nanotubes and nanowires.Thin Solid Films 499:299–305. doi:10.1016/j.tsf.2005.07.019

488. Ho W, Yu JC (2006) Sonochemical synthesis and visible light photocatalytic behavior ofCdSe and CdSe/TiO2 nanoparticles. J Mol Catal A 247:268–274. doi:10.1016/j.molcata.2005.11.057

489. Robel I, Kuno M, Kamat PV (2007) Size-dependent electron injection from excited CdSequantum dots into TiO2 nanoparticles. J Am Chem Soc 129:4136–4137. doi:10.1021/ja070099a

490. Shen Y, Bao J, Dai N et al (2009) Speedy photoelectric exchange of CdSe quantumdots/mesoporous titania composite system. Appl Surf Sci 255:3908–3911. doi:10.1016/j.apsusc.2008.10.004

491. Loef R, Houtepen AJ, Talgorn E et al (2009) Study of electronic defects in CdSe quantumdots and their involvement in quantum dot solar cells. Nano Lett 9:856–859. doi:10.1021/nl803738q

492. Im SH, Lee YH, Seok SI et al (2010) Quantum-dot-sensitized solar cells fabricated by thecombined process of the direct attachment of colloidal CdSe quantum dots having a ZnSglue layer and spray pyrolysis deposition. Langmuir 26:18576–18580. doi:10.1021/la1034382

493. Salant A, Shalom M, Hod I et al (2010) Quantum dot sensitized solar cells with improvedefficiency prepared using electrophoretic deposition. ACS Nano 4:5962–5968. doi:10.1021/nn1018208

494. Guijarro S, Lana-Villarreal T, Shen Q et al (2010) Sensitization of titanium dioxidephotoanodes with cadmium selenide quantum dots prepared by SILAR: photoelectrochem-ical and carrier dynamics study. J Phys Chem C 114:21928–21937. doi:10.1021/jp105890x

495. Fuke N, Hoch LB, Koposov AY et al (2010) CdSe quantum-dot-sensitized solar cellwith *100% internal efficiency. ACS Nano 4:6377–6386. doi:10.1021/nn101319x

496. Hossain MF, Biswas S, Zhang ZH, Takahashi T (2011) Bubble-like CdSe nanoclusterssensitized TiO2 nanotube arrays for improvement in solar cell. J Photochem Photobiol, A217:68–75. doi:10.1016/j.jphotochem.2010.09.020

497. Lin CJ, Chen S, Liou YH (2010) Wire-shaped electrode of CdSe-sensitized ZnO nanowirearrays for photoelectrochemical hydrogen generation. Electrochem Commun 12:1513–1516.doi:10.1016/j.elecom.2010.08.021

498. Wang H, Wu Z, Liu Y, Sheng Z (2008) The characterization of ZnO–anatase–rutilethree-component semiconductor and enhanced photocatalytic activity of nitrogen oxides.J Mol Catal 287:176–181. doi:10.1016/j.molcata.2008.03.010

499. Wang N, Li J, Zhu L et al (2008) Highly photocatalytic activity of metallichydroxide/titanium oxide nanoparticles prepared via a modified wet precipitation process.J Photochem Photobiol, A 198:282–287. doi:10.1016/j.jphotochem.2008.03.021

500. Wang N, Zhu LH, Li J, Tang HQ (2007) A novel Fe(OH)3/TiO2 nanoparticles and its highphotocatalytic activity. Chin Chem Lett 18:1261–1264. doi:10.1016/j.cclet.2007.08.020

501. Yu J, Hai Y, Cheng B (2011) Enhanced photocatalytic H2-production activity of TiO2 by Ni(OH)2 cluster modification. J Phys Chem C 115:4953–4958. doi:10.1021/jp111562d

502. Mohamed MM, Othman I, Mohamed RM (2007) Synthesis and characterization of MnOx/TiO2 nanoparticles for photocatalytic oxidation of indigo carmine dye. J PhotochemPhotobiol, A 191:153–161. doi:10.1016/j.jphotochem.2007.04.017

503. Lalitha K, Sadanandam G, Durga V et al (2010) Highly stabilized and finely dispersedCu2O/TiO2: a promising visible sensitive photocatalyst for continuous production ofhydrogen from glycerol:water mixtures. J Phys Chem C 114:22181–22189. doi:10.1021/jp107405u

504. Xu S, Du AJ, Liu J et al (2011) Highly efficient CuO incorporated TiO2 nanotubephotocatalyst for hydrogen production from water. Inter J Hydrogen En 36:6560–6568.doi:10.1016/j.ijhydene.2011.02.103

References 311

505. Liao DL, Badour CA, Liao BQ (2008) Preparation of nanosized TiO2/ZnO compositecatalyst and its photocatalytic activity for degradation of methyl orange. J PhotochemPhotobiol, A 194:11–19. doi:10.1016/j.jphotochem.2007.07.008

506. Chen D, Zhang H, Hu S, Li J (2008) Preparation and enhanced photo-electrochemicalperformance of coupled bicomponent ZnO–TiO2 nanocomposites. J Phys Chem C 112:117–122. doi:10.1021/jp077236a

507. Abdel Aal A, Barakat MA, Mohamed RM (2008) Electrophoreted Zn–TiO2–ZnOnanocomposite coating films for photocatalytic degradation of 2-chlorophenol. Appl SurfSci 254:4577–4583. doi:10.1016/j.apsusc.2008.01.049

508. Song H, Jiang H, Liu X, Meng G (2006) Efficient degradation of organic pollutant with WOx

modified nano TiO2 under visible irradiation. J Photochem Photobiol, A 181:421–428.doi:10.1016/j.jphotochem.2006.01.001

509. Ke D, Liu H, Peng T et al (2008) Preparation and photocatalytic activity of WO3/TiO2

nanocomposite particles. Mater Lett 62:447–450. doi:10.1016/j.matlet.2007.05.060510. Li XZ, Li FB, Yang CL, Ge WK (2001) Photocatalytic activity of Wox–TiO2 under visible

light irradiation. J Photochem Photobiol, A 141:209–217. doi:10.1016/S1010-6030(01)00446-4

511. Yasomanee JP, Bandara J (2008) Multi-electron storage of photoenergy using Cu2O–TiO2

thin film photocatalyst. Sol En Mater Sol Cells 92:348–352. doi:10.1016/j.solmat.2007.09.016

512. Celik E, Yildiz AY, Ak Azem NF et al (2006) Preparation and characterization of Fe2O3–

TiO2 thin films on glass substrate for photocatalytic applications. Mater Sci Eng, B 129:193–199. doi:10.1016/j.mseb.2006.01.013

513. Seftel EM, Popovici E, Mertens M et al (2008) SnIV-containing layered double hydroxidesas precursors for nano-sized ZnO/SnO2 photocatalysts. Appl Catal B 84:699–705. doi:10.1016/j.apcatb.2008.06.006

514. Chen LC, Tsai FR, Fang SH, Ho YC (2009) Properties of sol–gel SnO2/TiO2 electrodes andtheir photoelectrocatalytic activities under UV and visible light illumination. ElectrochimActa 54:1304–1311. doi:10.1016/j.electacta.2008.09.009

515. Ismail AA (2008) Single-step synthesis of a highly active photocatalyst for oxidation oftrichloroethylene. Appl Catal B 85:33–39. doi:10.1016/j.apcatb.2008.06.025

516. Bedja I, Kamat PV (1995) Capped semiconductor colloids. Synthesis and photoelectro-chemical behavior of TiO2-capped SnO2 nanocrystallites. J Phys Chem 99:9182–9188.doi:10.1021/j100022a035

517. Penpolcharoen M, Amal R, Brungs M (2001) Degradation of sucrose and nitrate over titaniacoated nano-hematite photocatalysts. J Nanopart Res 3:289–302. doi:10.1023/A:1017929204380

518. Nishijima K, Ohtani B, Yan X et al (2007) Incident light dependence for photocatalyticdegradation of acetaldehyde and acetic acid on S-doped and N-doped TiO2 photocatalysts.Chem Phys 339:64–72. doi:10.1016/j.chemphys.2007.06.014

519. Nishijima K, Fujisawa Y, Murakami N et al (2008) Development of an S-doped titaniananotube (TNT) site-selectively loaded with iron (III) oxide and its photocatalytic activities.Appl Catal B 84:584–590. doi:10.1016/j.apcatb.2008.05.014

520. Xiao G, Wang X, Li D, Fu X (2008) InVO4-sensitized TiO2 photocatalysts for efficient airpurification with visible light. J Photochem Photobiol, A 193:213–221. doi:10.1016/j.jphotochem.2007.06.027

521. Lee BT, Han JK, Gain AK et al (2006) TEM microstructure characterization of nano TiO2

coated on nano ZrO2 powders and their photocatalytic activity. Mater Lett 60:2101–2104.doi:10.1016/j.matlet.2005.12.102

522. Xu J, Ao Y, Fu D, Yuan C (2008) Synthesis of Bi2O3–TiO2 composite film withhigh-photocatalytic activity under sunlight irradiation. Appl Surf Sci 255:2365–2369.doi:10.1016/j.apsusc.2008.07.095

312 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

523. Tawkaew S, Chareonpanich M, Supothina S (2008) Preparation and photocatalytic study offibrous K0.3Ti4O7.3(OH)1.7—anatase TiO2 nanocomposite photocatalyst. Mater Chem Phys111:232–237. doi:10.1016/j.matchemphys.2008.03.031

524. Jang JS, Kim HG, Reddy VR et al (2005) Photocatalytic water splitting over iron oxidenanoparticles intercalated in HTiNb(Ta)O5 layered compounds. J Catal 231:213–222.doi:10.1016/j.jcat.2005.01.026

525. Jothiramalingam R, Wang MK (2007) Synthesis, characterization and photocatalytic activityof porous manganese oxide doped titania for toluene decomposition. J Hazard Mater147:562–569. doi:10.1016/j.jhazmat.2007.01.069

526. Cho J, Denes FS, Timmons RB (2006) Plasma processing approach to molecular surfacetailoring of nanoparticles: improved photocatalytic activity of TiO2. Chem Mater 18:2989–2996. doi:10.1021/cm060212g

527. Martinez AI, Acosta DR, Cedillo G et al (2005) Effect of SnO2 on the photocatalyticalproperties of TiO2 films. Thin Solid Films 490:118–123. doi:10.1016/j.tsf.2005.04.060

528. Li D, Haneda H (2003) Photocatalysis of sprayed nitrogen-containing Fe2O3–ZnO andWO3–ZnO composite powders in gas-phase acetaldehyde decomposition. J PhotochemPhotobiol, A 160:203–212. doi:10.1016/S1010-6030(03)00212-0

529. Smith W, Zhao YP (2009) Superior photocatalytic performance by vertically aligned core–shell TiO2/WO3 nanorod arrays. Catal Commun 10:1117–1121. doi:10.1016/j.catcom.2009.01.010

530. Jeon TH, Choi W, Park H (2011) Photoelectrochemical and photocatalytic behaviors ofhematite-decorated titania nanotube arrays: energy level mismatch versus surface specificactivity. J Phys Chem C 115:7134–7142. doi:10.1021/jp201215t

531. Woan K, Pyrgiotakis G, Sigmund W (2009) Photocatalytic carbon-nanotube–TiO2

composites. Adv Mater 21:2233–2239. doi:10.1002/adma.200802738532. Eder D (201) Carbon nanotube-inorganic hybrids. Chem Rev 110:1348–1385. doi:10.1021/

cr800433k533. Luo Y, Liu J, Xia X et al (2007) Fabrication and characterization of TiO2/short MWNTs

with enhanced photocatalytic activity. Mater Lett 61:2467–2472. doi:10.1016/j.matlet.2006.09.051

534. Gao B, Peng C, Chen GZ, Li Puma G (2008) Photo-electro-catalysis enhancement on carbonnanotubes/titanium dioxide (CNTs/TiO2) composite prepared by a novel surfactant wrappingsol–gel method. Appl Catal B 85:17–23. doi:10.1016/j.apcatb.2008.06.027

535. Yu H, Quan X, Chen S et al (2008) TiO2–carbon nanotube heterojunction arrays with acontrollable thickness of TiO2 layer and their first application in photocatalysis. J PhotochemPhotobiol, A 200:301–306. doi:10.1016/j.jphotochem.2008.08.007

536. Wang W, Serp P, Kalck P et al (2008) Preparation and characterization of nanostructuredMWCNT-TiO2 composite materials for photocatalytic water treatment applications. MaterRes Bull 43:958–967. doi:10.1016/j.materresbull.2007.04.032

537. Chen LC, Ho YC, Guo WS et al (2009) Enhanced visible light-induced photoelectrocatalyticdegradation of phenol by carbon nanotube-doped TiO2 electrodes. Electrochim Acta54:3884–3891. doi:10.1016/j.electacta.2009.02.001

538. Yen CY, Lin YF, Hung CH et al (2008) The effects of synthesis procedures on themorphology and photocatalytic activity of multi-walled carbon nanotubes/TiO2 nanocom-posites. Nanotechnology 19:045604

539. Xia XH, Jia ZJ, Yu Y et al (2007) Preparation of multi-walled carbon nanotube supportedTiO2 and its photocatalytic activity in the reduction of CO2 with H2O. Carbon 45:717–721.doi:10.1016/j.carbon.2006.11.028

540. Wang H, Wang HL, Jiang WF, Li ZQ (2009) Photocatalytic degradation of2,4-dinitrophenol (DNP) by multi-walled carbon nanotubes (MWCNTs)/TiO2 compositein aqueous solution under solar irradiation. Water Res 43:204–210. doi:10.1016/j.watres.2008.10.003

References 313

541. Wang H, Wang HL, Jiang WF (2009) Solar photocatalytic degradation of2,6-dinitro-p-cresol (DNPC) using multi-walled carbon nanotubes (MWCNTs)–TiO2

composite photocatalysts. Chemosphere 75:1105–1111. doi:10.1016/j.chemosphere.2009.01.014

542. Yao Y, Li G, Ciston S et al (2008) Photoreactive TiO2/carbon nanotube composites:synthesis and reactivity. Environ Sci Technol 42:4952–4957. doi:10.1021/es800191n

543. Tang ZR, Li F, Zhang Y et al (2011) Composites of titanate nanotube and carbon nanotubewith high mineralization ratio for gap-phase degradation of volatile aromatic pollutant.J Phys Chem C 115:7880–7886. doi:10.1021/jp1115838

544. Zhu LP, Liao GH, Huang WY et al (2009) Preparation, characterization and photocatalyticproperties of ZnO-coated multi-walled carbon nanotubes. Mater Sci Eng, B 163:194–198.doi:10.1016/j.mseb.2009.05.021

545. Oh WC, Jung AR, Ko WB (2009) Characterization and relative photonic efficiencies of anew nanocarbon/TiO2 composite photocatalyst designed for organic dye decomposition andbactericidal activity. Mater Sci Eng, C 29:1338–1347. doi:10.1016/j.msec.2008.10.034

546. Wu H, Wang Q, Yao Y et al (2008) Microwave-assisted synthesis and photocatalyticproperties of carbon nanotube/zinc sulfide heterostructures. J Phys Chem C 112:16779–16783. doi:10.1021/jp8069018

547. Wu HQ, Wang Q, Yao YZ et al (2009) Microwave-assisted synthesis and highlyphotocatalytic activity of MWCNT/ZnSe heterostructures. Mater Chem Phys 113:539–543.doi:10.1016/j.matchemphys.2008.08.004

548. Maeda K, Eguchi M, Youngblood WJ, Mallouk TE (2008) Niobium oxide nanoscrolls asbuilding blocks for dye-sensitized hydrogen production from water under visible lightirradiation. Chem Mater 20:6770–6778. doi:10.1021/cm801807b

549. Tsubota T, Ono A, Murakami N, Ohno T (2009) Characterization and photocatalyticperformance of carbon nanotube material-modified TiO2 synthesized by using the hot CVDprocess. Appl Catal B 91:533–538. doi:10.1016/j.apcatb.2009.06.024

550. Kim TW, Lee MJ, Shim WG et al (2008) Adsorption and photocatalytic decomposition oforganic molecules on carbon-coated TiO2. J Mater Sci 43:6486–6494. doi:10.1007/s10853-008-2978-2

551. Guo Y, Wang H, He C et al (2009) Uniform carbon-coated ZnO nanorods:microwave-assisted preparation, cytotoxicity, and photocatalytic activity. Langmuir25:4678–4684. doi:10.1021/la803530h

552. Zhong J, Chen F, Zhang J (2010) Carbon-deposited TiO2: Synthesis, characterization, andvisible photocatalytic performance. J Phys Chem C 114:933–939. doi:10.1021/jp909835m

553. Zhang L, Cheng H, Zong R, Zhu Y (2009) Photocorrosion suppression of ZnO nanoparticlesvia hybridization with graphite-like carbon and enhanced photocatalytic activity. J PhysChem C 113:2368–2374. doi:10.1021/jp807778r

554. Hu X, Huang K, Fang D, Liu S (2011) Enhanced performances of dye-sensitized solar cellsbased on graphite-TiO2 composites. Mater Sci Eng, B 176:431–435. doi:10.1016/j.mseb.2010.12.016

555. Sellappan R, Zhu J, Fredriksson H et al (2011) Preparation and characterization ofTiO2/carbon composite thin films with enhanced photocataloytic activity. J Mol Catal A335:136–144. doi:10.1016/j.molcata.2010.11.025

556. Novoselov KS, Geim AK, Morozov SV et al (2004) Electric field effect in atomically thincarbon films. Science 306:666–669. doi:10.1126/science.1102896

557. Meyer JC, Geim AK, Katsnelson MI et al (2007) The structure of suspended graphenesheets. Nature 446:60–63. doi:10.1038/nature05545

558. Dreyer DR, Park S, Bielawski CW, Ruoff RS (2010) The chemistry of graphene oxide.Chem Soc Rev 39:228–240. doi:10.1039/B917103G

559. Park S, Ruoff RS (2009) Chemical methods for the production of graphenes. NatNanotechnol 4:217–224. doi:10.1038/nnano.2009.58

314 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

560. Kamat PV (2010) Graphene-based nanoarchitectures. Anchoring semiconductor and metalnanoparticles on a two-dimentional carbon support. J Phys Chem Lett 1:520–527. doi:10.1021/jz900265j

561. Zhu H, Wei J, Wang K, Wu D (2009) Applications of carbon materials in photovoltaic solarcells. Sol En Mater Sol Cells 93:1461–1470. doi:10.1016/j.solmat.2009.04.006

562. Stroyuk OL, Andryushina NS, Kuchmiy SY et al (2015) Photochemical processes involvinggraphene oxide. Theoret Exp Chem 51:1–29. doi:10.1007/s11237-015-9393-y

563. Kamat PV, Bedja I, Hotchandani S (1994) Photoinduced charge transfer between carbon andsemiconductor clusters. One-electron reduction of C60 in colloidal TiO2 semiconductorsuspensions. J Phys Chem 98:9137–9142. doi:10.1021/j100088a008

564. Brown P, Kamat PV (2008) Quantum dot solar cells. Electrophoretic deposition of CdSe-C60

composite films and capture of photogenerated electrons with nC60 cluster shell. J Am ChemSoc 130:8890–8891. doi:10.1021/ja802810c

565. Kathiravan KA, Sathish Kumar P, Renganathan R, Anandan S (2009) Photoinduced electrontransfer reactions between meso-tetrakis(4-sulfonatophenyl)porphyrin and colloidalmetal-semiconductor nanoparticles. Colloids Surf A 333:175–181. doi:10.1016/j.colsurfa.2008.09.042

566. Dai W, Zheng X, Yang H et al (2009) The promoted effect of UV irradiation on preferentialoxidation of CO in an H2-rich stream over Au/TiO2. J Power Sources 188:507–514. doi:10.1016/j.jpowsour.2008.12.028

567. Bannat I, Wessels K, Oekermann T et al (2009) Improving the photocatalytic performance ofmesoporous titania films by modification with gold nanostructures. Chem Mater 21:1645–1653. doi:10.1021/cm803455k

568. Abdulla-Al-Mamun M, Kusumoto Y, Ahmmad B, Shariful Islam M (2010) Photocatalyticcancer (HeLa) cell-killing enhanced with Cu–TiO2 nanocomposite. Topics Catal 53:571–577. doi:10.1007/s11244-010-9489-5

569. Yoshinaga M, Yamamoto K, Sato N et al (2009) Remarkably enhanced photocatalyticactivity by nickel nanoparticle deposition on sulfur-doped titanium dioxide thin film. ApplCatal B 87:239–244. doi:10.1016/j.apcatb.2008.08.023

570. Lin HY, Chang YS (2010) Photocatalytic water splitting for hydrogen production onAu/KTiNbO5. Inter J Hydrogen En 35:8463–8471. doi:10.1016/j.ijhydene.2010.06.006

571. Wang W, Zhang J, Chen F et al (2008) Preparation and photocatalytic properties of Fe3+-doped Ag@TiO2 core–shell nanoparticles. J Colloid Interface Sci 323:182–186. doi:10.1016/j.jcis.2008.03.043

572. Sakai N, Fujiwara Y, Takahashi Y, Tatsuma T (2009) Plasmon-resonance-based generationof cathodic photocurrent at electrodeposited gold nanoparticles coated with TiO2 films.ChemPhysChem 10:766–769. doi:10.1002/cphc.200800704

573. Ikuma Y, Bessho H (2007) Effect of Pt concentration on the production of hydrogen by aTiO2 photocatalyst. Inter J Hydrogen En 32:2689–2692. doi:10.1016/j.ijhydene.2006.09.024

574. Miyao T, Suzuki Y, Naito S (2000) Hydrogen formation by the photodecomposition ofwater over Pt/TiO2 suspended in a water in oil emulsion. Catal Lett 66:197–200. doi:10.1023/A:1019032614298

575. Ohtani B, Iwai K, Nishimoto S, Sato S (1997) Role of platinum deposits on titanium(IV) oxide particles: structural and kinetic analyses of photocatalytic reaction in aqueousalcohol and amino acid solutions. J Phys Chem B 101:3349–3359. doi:10.1021/jp962060q

576. Sun B, Vorontsov AV, Smirniotis PG (2003) Role of platinum deposited on TiO2 in phenolphotocatalytic oxidation. Langmuir 19:3151–3156. doi:10.1021/la0264670

577. Subramanian V, Wolf EE, Kamat PV (2001) Semiconductor-metal composite nanostruc-tures: to what extent do metal nanoparticles improve the photocatalytic activity of TiO2

films. J Phys Chem B 105:11439–11446. doi:10.1021/jp011118k578. Dawson A, Kamat PV (2001) Semiconductor-metal nanocomposites. Photoinduced fusion

and photocatalysis of gold-capped TiO2 (TiO2/gold) nanoparticles. J Phys Chem B 105:960–966. doi:10.1021/jp0033263

References 315

579. Subramanian V, Wolf EE, Kamat PV (2003) Influence of metal/metal ion concentration onthe photocatalytic activity of TiO2-Au composite nanoparticles. Langmuir 19:469–474.doi:10.1021/la026478t

580. Zheng Y, Zheng L, Zhan Y et al (2007) Ag/ZnO heterostructure nanocrystals: synthesis,characterization, and photocatalysis. Inorg Chem 46:6980–6986. doi:10.1021/ic700688f

581. Gu C, Cheng C, Huang H et al (2009) Growth and photocatalytic activity of dendrite-likeZnO@Ag heterostructure nanocrystals. Cryst Growth Design 9:3278–3285. doi:10.1021/cg900043k

582. Lombardi I, Marchionna S, Zangari G, Pizzini S (2007) Effect of Pt particle size anddistribution on photoelectrochemical hydrogen evolution by p-Si photocathodes. Langmuir23:12413–12420. doi:10.1021/la7016165

583. Sheeney-Haj-Ichia L, Pogorelova S, Gofer Y, Willner I (2004) Enhanced photoelectro-chemistry in CdS/Au nanoparticle bilayers. Adv Func Mater 14:416–424. doi:10.1002/adfm.200305430

584. Lana-Villarreal T, Gomez R (2005) Interfacial electron transfer at TiO2 nanostructuredelectrodes modified with capped gold nanoparticles: the photoelectrochemistry of wateroxidation. Electrochem Commun 7:1218–1224. doi:10.1016/j.elecom.2005.08.031

585. Mizukoshi Y, Makise Y, Shuto T et al (2007) Immobilization of noble metal nanoparticleson the surface of TiO2 by the sonochemical method: photocatalytic production of hydrogenfrom an aqueous solution of ethanol. Ultrason Sonochem 14:387–392. doi:10.1016/j.ultsonch.2006.08.001

586. Mizukoshi Y, Sato K, Konno TJ, Masahashi N (2010) Dependence of photocatalyticactivities upon the structures of Au/Pd bimetallic nanoparticles immobilized on TiO2

surface. Appl Catal B 94:248–253. doi:10.1016/j.apcatb.2009.11.015587. Tian B, Zhang J, Tong T, Chen F (2008) Preparation of Au/TiO2 catalysts from Au (I)–

thiosulfate complex and study of their photocatalytic activity for the degradation of methylorange. Appl Catal B 79:394–401. doi:10.1016/j.apcatb.2007.11.001

588. Haldar KK, Sen T, Patra A (2008) Au@ZnO core-shell nanoparticles are efficient energyacceptors with organic dye donors. J Phys Chem C 112:11650–11656. doi:10.1021/jp8031308

589. Liu Z, Guo B, Hong L, Jiang H (2005) Physicochemical and photocatalytic characterizationsof TiO2/Pt nanocomposites. J Photochem Photobiol, A 172:81–88. doi:10.1016/j.jphotochem.2004.11.008

590. Yi Z, Wei W, Lee S, Jianhua G (2007) Photocatalytic performance of plasma sprayedPt-modified TiO2 coatings under visible light irradiation. Catal Commun 8:906–912. doi:10.1016/j.catcom.2006.09.023

591. Yin S, Sato T (2005) Photocatalytic activity of platinum loaded fibrous titania prepared bysol–vothermal process. J Photochem Photobiol, A 169:89–94. doi:10.1016/j.jphotochem.2004.05.038

592. Alam Khan M, Shaheer Akhtar M, Woo SI, Yang O (2008) Enhanced photoresponse undervisible light in Pt ionized TiO2 nanotube for the photocatalytic splitting of water. CatalCommun 10:1–5. doi:10.1016/j.catcom.2008.01.018

593. Shan Z, Wu J, Xu F et al (2008) Highly effective silver/semiconductor photocatalyticcomposites prepared by a silver mirror reaction. J Phys Chem C 112:15423–15428. doi:10.1021/jp804482k

594. Moonsiri M, Rangsunvigit P, Chavadej S, Gulari E (2004) Effects of Pt and Ag on thephotocatalytic degradation of 4-chlorophenol and its by-products. Chem Eng J 97:241–248.doi:10.1016/j.cej.2003.05.003

595. Zhou G, Deng J (2007) Preparation and photocatalytic performance of Ag/ZnOnano-composites. Mater Sci Semicon Proc 10:90–96. doi:10.1016/j.mssp.2007.05.003

596. Zhang Y, Mu J (2007) One-pot synthesis, photoluminescence, and photocatalysis of Ag/ZnOcomposites. J Colloid Interface Sci 309:478–484. doi:10.1016/j.jcis.2007.01.011

316 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

597. Araña J, Doña-Rodriguez JM, Herrera Melian JA et al (2005) Role of Pd and Cu ingas-phase alcohols photocatalytic degradation with doped TiO2. J Photochem Photobiol, A174:7–14. doi:10.1016/j.jphotochem.2005.03.003

598. Wu NL, Lee MS (2004) Enhanced TiO2 photocatalysis by Cu in hydrogen production fromaqueous methanol solution. Inter J Hydrogen En 29:1601–1605. doi:10.1016/j.ijhydene.2004.02.013

599. Chiarello GL, Aguirre MH, Selli E (2010) Hydrogen production by photocatalytic steamreforming of methanol on noble metal-modified TiO2. J Catal 273:182–190. doi:10.1016/j.jcat.2010.05.012

600. Chiarello GL, Forni L, Selli E (2009) Photocatalytic hydrogen production by liquid- andgas-phase reforming of CH3OH over flame-made TiO2 and Au/TiO2. Catal Today 144:69–74. doi:10.1016/j.cattod.2009.01.023

601. Lai Y, Zhuang H, Xie K et al (2010) Fabrication of uniform Ag/TiO2 nanotube arraystructures with enhanced photoelectrochemical performance. New J Chem 34:1335–1340.doi:10.1039/B9NJ00780F

602. Pang H, Li Y, Guan L et al (2011) TiO2/Ni nanocomposites: biocompatible and recyclablemagnetic photocatalysts. Catal Commun 12:611–615. doi:10.1016/j.catcom.2010.12.015

603. Ruvarac-Bugarčić IA, Šaponjić ZV, Zec S et al (2005) Photocatalytic reduction of cadmiumon TiO2 nanoparticles modified with amino acids. Chem Phys Lett 407:110–113. doi:10.1016/j.cplett.2005.03.058

604. Somasundaram S, Ming Y, Chenthamarakshan CR et al (2004) Free radical-mediatedhetero–geneous photocatalytic reduction of metal ions in UV-irradiated titanium dioxidesuspensions. J Phys Chem B 108:4784–4788. doi:10.1021/jp036729m

605. Wood A, Giersing M, Mulvaney P (2001) Fermi level equilibration in quantum dot—metalnanojunctions. J Phys Chem B 105:8810–8815. doi:10.1021/jp011576t

606. Siemon U, Bahnemann D, Testa JJ et al (2002) Heterogeneous photocatalytic reactionscomparing TiO2 and Pt/TiO2. J Photochem Photobiol, A 148:247–255. doi:10.1016/S1010-6030(02)00050-3

607. Rabani J, Behar D (1989) Quenching of aqueous colloidal zinc oxide fluorescence byelectron and hole scavengers: effect of a positive polyelectrolyte. J Phys Chem 93:2559–2563. doi:10.1021/j100343a063

608. Subramanian V, Wolf EE, Kamat PV (2003) Green emission to probe photoinducedcharging events in ZnO–Au nanoparticles. Charge distribution and Fermi level equilibration.J Phys Chem B 107:7479–7485. doi:10.1021/jp0275037

609. Vigil E, Gonzales B, Zumeta I et al (2005) Preparation of photoelectrodes with spectral res–ponse in the visible without applied bias based on photochemically deposited copper oxideinside a porous titanium dioxide film. Thin Solid Films 489:50–55. doi:10.1016/j.tsf.2005.04.098

610. Lu L, Hu S, Lee HI et al (2007) Photoinduced growth of Cu nanoparticles on ZnO fromCuCl2 in methanol. J Nanopart Res 9:491–496. doi:10.1007/s11051-006-9087-4

611. Liqiang J, Baiqi W, Baifu X et al (2004) Investigations on the surface modification of ZnOnanoparticle photocatalyst by depositing Pd. J Sol State Chem 177:4221–4227. doi:10.1016/j.jssc.2004.08.016

612. Bae E, Choi W (2003) Highly enhanced photoreductive degradation of perchlorinatedcompounds on dye-sensitized metal/TiO2 under visible light. Environ Sci Technol 37:147–152. doi:10.1021/es025617q

613. Liu Y, Liu C, Zhang Z, Wang C (2001) The surface enhanced Raman scattering effects ofcomposite nanocrystals of Ag–TiO2. Spectrochim Acta A 57:35–39. doi:10.1016/S1386-1425(00)00326-7

614. He J, Yang P, Sato H et al (2004) Effects of Ag-photodeposition on photocurrent of an ITOelectrode modified by a hybrid film of TiO2 nanosheets. J Electroanal Chem 566:227–233.doi:10.1016/j.jelechem.2003.11.031

References 317

615. Korzhak AV, Ermokhina NI, Stroyuk OL et al (2008) Photocatalytic hydrogen evolutionover mesoporous TiO2/Metal nanocomposites. J Photochem Photobiol, A 198:126–134.doi:10.1016/j.jphotochem.2008.02.026

616. Ng YH, Ikeda S, Harada T et al (2008) Photocatalytic route for synthesis of hollow porouscarbon/Pt nanocomposites with controllable density and porosity. Chem Mater 20:1154–1160. doi:10.1021/cm702034w

617. Naoi K, Ohko Y, Tatsuma T (2004) TiO2 films loaded with silver nanoparticles: control ofmulticolor photochromic behavior. J Am Chem Soc 126:3664–3668. doi:10.1021/ja039474z

618. Uddin MJ, Cesano F, Bertarione S et al (2008) Tailoring the activity of Ti-basedphotocatalysts by playing with surface morphology and silver doping. J PhotochemPhotobiol, A 196:165–173. doi:10.1016/j.jphotochem.2007.07.037

619. Tanahashi I, Iwagishi H, Chang G (2008) Localized surface plasmon resonance sensingproperties of photocatalytically prepared Au/TiO2 films. Mater Lett 62:2714–2716. doi:10.1016/j.matlet.2008.01.023

620. Tanahashi I (2007) Photocatalytic preparation of Ag/TiO2 films and their localized surfaceplasmon resonance sensing properties. Bull Chem Soc Jpn 80:2019–2023. doi:10.1246/bcsj.80.2019

621. Paramasivam I, Macak JM, Ghicov A, Schmuki P (2007) Enhanced photochromism of Agloaded self-organized TiO2 nanotube layers. Chem Phys Lett 445:233–237. doi:10.1016/j.cplett.2007.07.107

622. Zolotavin P, Permenova E, Sarkosov O et al (2008) Two-photon luminescence enhancementof silver nanoclusters photodeposited onto mesoporous TiO2 film. Chem Phys Lett 457:342–346. doi:10.1016/j.cplett.2008.04.034

318 5 Synthesis of Nanocrystalline Photo-Active Semiconductors

Chapter 6Probing with Light—Optical Methodsin Studies of NanocrystallineSemiconductors

The studies of photochemical and photocatalytic processes involving the semi-conductor NPs are performed using a broad range of modern physical and chemicalmethods applied to determine the NP size and structure, their spectral, photo-physical and other properties. The most frequently used are electron spectroscopy inthe absorption, transmission and reflection modes, photoluminescence spec-troscopy, electron microscopy (in the scanning and transmission modes), X-raysdiffraction, etc. Nuances of the photochemical properties of semiconductor NPs canbe revealed using the lamp and laser flash photolysis, the electron paramagneticresonance and the Raman spectroscopy. The voltammetry is often used to deter-mine the potentials of conduction and valence bands of semiconductor NPs, theadsorption/desorption methods—for the determination of the specific surface areaand pore size of semiconductor nanophotocatalysts. A detailed description of thesemethods is far beyond the scope of the present book. This chapter is confined to themethods using light to probe the properties of nanocrystalline semiconductors andthus shedding light on the structure and properties of these fascinating objects as aresult of their interaction with the probing irradiation.

6.1 A Brief Characterization of the Spectral Studiesof Nano-Semiconductors

The spectral studies of optical properties of the nanocrystalline semiconductorsprovide the most ample information when applied to optically transparent colloidalsystems and films. The family of spectral methods includes the electron absorptionspectroscopy, stationary and time-resolved luminescence spectroscopy, Ramanspectroscopy, as well as the time-resolved pulse spectroscopy (the flash photolysis)in the femto-microsecond time domains (Fig. 6.1).

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4_6

319

The studies of semiconductor NPs by the electron absorption spectroscopy (in atransmission mode—for colloidal NPs and films, in a reflection mode—fornanocrystalline powders and thick films) allow to determine the type and energy ofinterband electron transitions in the nano-objects and to calculate the bandgap Eg orthe energy of the first exciton transitions E1 for the case of strong confinementeffects. For the quantum-sized semiconductor NPs, Eg depends on the NP dimen-sions and the NP size d can be calculated by using reported empiricalEg(d) dependences, while the size distribution of NPs can be evaluated from thespectral width of the excitonic absorption band.

Basing on the optical bandgap Eg and the reported values of bulk CB and VBpotentials of a given semiconductor the ECB and EVB potentials can be evaluated forthe NPs of a given size. In the case of solid-solution nanophotocatalysts, such asCdxZn1−xS or CdSxSe1−x, the reported bandgap dependences on the composition(x value) can be used to determine the NP composition (for the NPs with weak or noconfinement effects).

The analysis of a longer-wavelength spectral range of the absorption spectra of semi–conductor NPs corresponding to hv < Eg and originating from the absorbance on thelattice defects provides information on the energy and density of local defect-relatedelectron states residing in the forbidden band of semiconductor NPs. Finally, theanalysis of the surface plasmon resonance bands of metal NPs (Au, Ag, Cu) in

Fig. 6.1 Family of spectral methods for the studies of nanocrystalline semiconductors

320 6 Probing with Light—Optical Methods in Studies …

semiconductor/metal nanocomposites allows evaluating the size of metal NPs, themetal content, electron gas density, conductivity and other characteristics.

The photoluminescence (PL) spectroscopy is used to determine the energy ofphotons emitted as a result of the electron-hole recombination and the energy anddensity of local defect-related states acting as the electron/hole trapping sites. In thecase of direct (excitonic) PL, the size and size distribution of semiconductor NPscan be derived from the reported empirical EPL(d) dependences.

The observations of PL decay dynamics, as well as the PL quenching by varioussubstrates, provide ample information on the rate of the radiative recombination andmechanisms of the interaction between the semiconductor NPs and other compo-nents of the photocatalytic and photoelectrochemical systems.

The Raman spectroscopy is used to study the structure of nanocrystallinesemiconductors, especially the “core/shell” composites. The method can also revealthe phase composition of NPs, their size and a degree of the NP lattice disorder.

The time-resolved (flash) photolysis, both in the conventional transmission modeand in the mode of diffuse reflectance, is a powerful tool for probing the primaryphotochemical processes that occur with the participation of semiconductor NPsand nanoheterostructures. The observations of the spectral parameters and decaykinetics of the absorption bands in the non-stationary differential spectra registeredunder the pulsed illumination provide information on the structure and reactivity ofthe short-lived intermediates. Complementary, the intensity and relaxation rate ofthe “negative” non-stationary bleaching bands bear valuable information on thecapability of semiconductor NPs to the accumulation of an excessive charge duringthe light pulse as well as the dynamics of following charge transfers to the acceptingcomponents of the photochemical system.

The above spectral methods can be used both separately and in a complex way.For some systems, combinations of the spectral methods (for example, the PLspectroscopy and flash photolysis) can provide quite unique information on thestructure and properties of nano-semiconductors and the products of photochemicalreactions on their surfaces.

6.2 Studies of Nano-Photocatalysts by the ElectronAbsorption Spectroscopy

The basic features of the interaction between the electromagnetic irradiation andsemiconducting materials, the nature of absorption bands in electron spectra of thesemiconductors, as well as the influence of spatial exciton confinement on theoptical properties of semiconductor NPs were discussed in Chap. 1. Here we focuson the methodology of extracting the information on the type and energy of electrontransitions, on the electrophysical parameters of semiconductor NPs, as well on asthe NP size and size distribution from the electron absorption spectra.

6.1 A Brief Characterization of the Spectral Studies … 321

Determination of basic electrophysical parameters of nanocrystalline semi-conductors. The character of light absorption by semiconductor NPs depends onmany factors, in particular on the light wavelength, th semiconductor composition,the density of “alien” atoms and inherent structural defects, the free charge carrierdensity, etc.

The fundamental light absorption by a semiconductor results in the generation ofa couple of a conduction band electron (e�CB) and a valence band hole (hþ

VB) boundby the Coulomb force. The energy of electron transition from VB to CB depends onthe bandgap Eg (forbidden band between VB and CB). The bandgap value sets therange of spectral sensitivity of the semiconductor which is also a function of the NPsize in the case of quantum-sized semiconductor NPs.

The bandgap can be estimated using the absorption threshold, that is, from thewavelength corresponding to the fundamental absorption band edge (kbe) asEg = 1241/kbe, where Eg is expressed in eV, kbe—in nm. It can be calculated as anintersection point between the abscissae axis and a tangent to the linear section ofthe absorption band edge—the “tangent” method (Fig. 6.2a).

As the longer-wavelength band edge of an ensemble of the quantum-sizedsemiconductor NPs is formed mostly by a fraction of larger NPs, the bandgapderived from the tangent method should be regarded as a lower limit of thebandgaps in the ensemble. A more precise determination of the average band gapcan be realized in the cases where a distinct excitonic absorption maximum ispresent in the absorption spectra corresponding to a transition between quantized(discrete) VB/CB levels (Fig. 6.2b). Typically, the colloidal semiconductor NPs areproduced in the form of rather polydisperse NP ensembles with a 15–20% sizedistribution and the first excitonic maximum becomes smeared for such systemsmaking possible only the tangent-based bandgap determination. Both methods forthe Eg determination are valid only for the direct interband electron transitions. Incases where the type of transition is unknown, it can also be deduced from theabsorption spectrum.

Fig. 6.2 Examples of thedetermination of the bandgapof colloidal ZnO NPs (a) andCdS NPs (b) by the tangentmethod (a) and from theposition of the first absorptionmaximum (b) [90, 124, 125]

322 6 Probing with Light—Optical Methods in Studies …

Determination of the type of electron transition. In general, the fundamentallight absorption by a semiconductor can originate from the direct and indirectinterband electron transitions (see Chap. 1). The absorption bands corresponding tothe indirect transitions are typically featureless and do not have a distinct edge. Inthe case of a direct electron transition, the fundamental absorption band typicallyreveals a distinct edge at hv = Eg (Fig. 6.3a) and a low light absorption at hv < Eg

tha latter originating both from the size distribution of semiconductor NPs and fromthe electron transitions with the participation of mid-bandgap defect-related levels.This section of the spectrum appears linear in the coordinates of Urbach Eq. (1) thatdescribes the dependence of the light absorption coefficient a on the light quantumenergy

ln a ¼ bhvkT

; ð1Þ

where b is a constant characterizing the structure disordering, k is the Boltzmannconstant, T is temperature.

The fundamental absorption spectrum, that is, the dependence of the funda-mental absorption coefficient a on the light quantum energy is described by ageneral Eq. (2):

a ¼ Aðhv� EgÞn

hv; ð2Þ

where A is a constant, while n depends on the transition type and can be equal to ½and 2 for the allowed direct and indirect transitions, respectively, and to 3/2 and 3—for the forbidden direct and indirect transitions.

The n constant can be determined from a tangent of the absorption spectrumlinearized in the coordinates ln(ahv) versus ln(hv − Eg). For example, for colloidalZnS NPs n was found to be 0.50 ± 0.02 (Fig. 6.3a, insert) indicating that the

Fig. 6.3 a Absorptionspectrum of colloidal ZnSsolution. Insert: the spectrumlinearized in the coordinatesln(ahv) versus ln(hv − Eg);b absorption spectrum of ZnSNPs presented as d{ln(ahv)}/d{hv} versus hv. The dashedline indicates a discontinuitypoint [126]

6.2 Studies of Nano-Photocatalysts by the Electron … 323

fundamental absorption band edge of ZnS NPs corresponds to an allowed directinterband electron transition.

The first derivative of the Eq. (2)—function (3) reveals a discontinuity point athv = Eg (Fig. 6.3b) allowing for a more precise determination of the bandgap:

d lnðahvÞf gdðhvÞ ¼ n

hv� Eg: ð3Þ

For a given n the bandgap Eg can be calculated as a cross point between thelinear section of “(ahv)1/n – hv” dependence and the abscissae (hv) axis.

Example—the determination of electrophysical parameters of solid-solutionCdxZn1−xS NPs. The nanocrystalline materials based on mixed cadmium-zincsulfide are quite broadly used as the photocatalysts of various redox processesincluding the reduction of metal ions and methylviologen, evolution of hydrogenfrom aqueous solutions of sacrificial donors, photopolymerization, etc. The analysisof CdxZn1−xS-based photocatalytic systems and possible mechanisms of the pho-tocatalytic reactions requires information on the electrophysical properties ofCdxZn1−xS NPs of any given composition. The methodology of determination ofthe composition-dependent Eg, ECB, and EVB parameters of cadmium-zinc sulfidesolid-solutions proposed in [1–3] is of a general character and can be applied toother solid solutions of semiconductor materials for which the bulk properties ofseparate components are known.

The SPP-stabilized aqueous colloidal CdxZn1−xS NPs with x varied from 0 to 1 canbe easily produced via the interaction between a mixture of cadmium(II) and zinc(II)chlorides and sodium sulfide [2]. An XRD study of the dried residuals from suchcolloids showed that the reaction yields cubic CdxZn1−xS solid solution NPs with anaverage size of 6 nm, irrespectively of the composition [2]. For such NPs, the quantumsize effects are expressed only weakly and therefore the position of the fundamental

Fig. 6.4 Absorption spectraof colloidal CdxZn1−xS NPsproduced at x = 0 (curve 1),0.25 (curve 2), 0.50 (curve 3),0.75 (curve 4), and 1.00(curve 5) [1–3]

324 6 Probing with Light—Optical Methods in Studies …

absorption band edge is determined mainly by the NP composition, not their size(Fig. 6.4) [1–3].

In the family of bulk (microcrystalline) CdxZn1−xS solid solutions the compo-sition variation from pure cadmium sulfide (x = 1.0) to pure zinc sulfide (x = 0)results in a change of the valence band (VB) potential from 1.6 V versus normalhydrogen electrode (NHE) for CdS [4] to 1.8 V (NHE) for ZnS [4]. This VBvariation is small comparatively to a large conduction band (CB) variation from−0.8 V for CdS up to −1.8 V for ZnS. Therefore, the Ebulk

VB of composition-variedCdxZn1−xS can be approximately presented by a linear combination (4) [1, 2]:

EbulkVB xð Þ ¼ xEbulk

VB CdSð Þ þ 1� xð ÞEbulkVB ZnSð Þ: ð4Þ

The bandgap Ebulkg of bulk CdxZn1−xS crystals can be calculated using the

reported empirical expression (5) [5, 6]:

Ebulkg ¼ 3:6� 1:78xþ 0:61x2: ð5Þ

Basing on Ebulkg and Ebulk

VB calculated for a given x the bulk CB potential can beexpressed as

ECB xð Þ ¼ EVB xð Þ � Eg xð Þ: ð6Þ

Table 6.1 presents the Eg, ECB, and EVB parameters of CdxZn1−xS NPs (with noQSEs observable for such NPs) calculated using Eqs. (4–6).

For the colloidal CdxZn1−xS NPs, for which the QSEs become possible, thebandgap can be calculated by using the above discussed “tangent” method from theposition of kbe. The QSE influence on the positions of ECB and EVB is typicallyaccounted for by using the effective mass approximation and the values of effectiveelectron and hole masses determined as linear combinations of the correspondingparameters of the individual cadmium and zinc sulfide [5]:

Table 6.1 Bandgap andconduction and valence bandpotentials of bulk cubicCdxZn1−xS solid solutionscrystals of a variedcomposition calculated usingEqs. (4–6)

x Eg, eV EVB, V (NHE) ECB, V (NHE)

1.0 2.40 1.60 −0.80

0.9 2.47 1.62 −0.85

0.8 2.54 1.64 −0.90

0.7 2.63 1.66 −0.97

0.6 2.73 1.68 −1.05

0.5 2.85 1.70 −1.15

0.4 2.97 1.72 −1.25

0.3 3.11 1.74 −1.37

0.2 3.26 1.76 −1.50

0.1 3.43 1.78 −1.65

0.0 3.60 1.80 −1.80

6.2 Studies of Nano-Photocatalysts by the Electron … 325

m�e xð Þ ¼ xm�

e CdSð Þ þ 1�xð Þm�e ZnSð Þ; and m�

h xð Þ ¼ xm�h CdSð Þ þ 1�xð Þm�

h ZnSð Þ

where m�e (CdS) = 0.20m0, m�

e(ZnS) = 0.27m0, m�h(CdS) = 0.70m0, m�

h(ZnS) =0.58m0 [7], m0 is the free electron rest mass.

Then, the potentials of allowed bands of CdxZn1−xS NPs of any given compo-sition and size can be estimated using Eqs. (7) and (8), the m�

e(x) and m�h(x) pa-

rameters calculated from the bulk effective masses and the bandgap Eg determinedfrom the absorption spectra of colloidal solutions:

ECB xð Þ ¼ EbulkCB xð Þ � m�

h m�e þm�

h

� ��1ðEg � Ebulkg Þ; ð7Þ

EVB xð Þ ¼ EbulkVB xð Þþm�

e m�e þm�

h

� ��1ðEg � Ebulkg Þ ð8Þ

Determination of the average size d and size distribution of semiconductorNPs using empirical Eg(d) calibration curves. The effective mass approximation(EMA) describes the size scaling of the bandgap of semiconductor NPs with a sized falling into the regime of weak/moderate exciton confinement (see Chap. 1). Forthe NPs with medium-to-strong QSEs, the basic assumptions of EMA becomeinvalid and the calculational results do not correspond to the experimental mea-surements by the direct methods (such as TEM). In this view, for a number ofsemiconductor NPs empirical calibration curves were proposed that combine thedata reported in various papers (in some cases, tens of separate papers, as in the caseof cadmium selenide).

The calibration curves correlate the optical bandgap of size-selected semicon-ductor NPs with the results of direct size measurements using electron microscopyas well as indirect measurements by the light scattering and the wide angle/smallangle X-Ray scattering. Figures 6.5, 6.6, 6.7, 6.8 and 6.9 illustrate some of thereported empirical data on the “bandgap (first excitonic maximum)—NP size”dependences for a series of semiconductor NPs that are typcially used in thenano-photocatalysis and the photoelectrochemical systems, in particular for CdS(Fig. 6.5), ZnO (Fig. 6.6), ZnS (Fig. 6.7), PbS (Fig. 6.8), and CdSe (Fig. 6.9). Thecomputational curves presented in the figures are derived from the “classical” EMAas well as some other reported modeling approaches, while the experimental datawere collected by TEM and the wide angle/small angle X-ray scattering.

The figures show that EMA agrees well with the experimental data only in thecase of CdS NPs. For other presented semiconductor NPs this approximation resultsin a strong size over-estimation already at d = 4–6 nm. The data show that theEMA, though being comparatively simple and universal, should be applied with acaution. In the cases of strong differences between the results of EMA and the directsize measurements, such, for example, as CdSe NPs revealed, alternative and moresophisticated theoretical models or empirical calibration curves built on the

326 6 Probing with Light—Optical Methods in Studies …

experimental results of TEM and small-angle X-ray scattering measurementsshould preferably be used to evaluate the NP size.

The empirical size calibration curves for the determination of the size of CdSeand CdTe NPs based on tens of separate reports can be found in [8]. Such curvesactually accumulate the long history of studies of the NPs produced by differentmethods and, therefore, the determination of the NP size from optical absorption

Fig. 6.5 Calibration Eg(d) data for CdS NPs. Curve 1 is derived from the EMA withm�

e = 0.204m0 and m�h = 0.70m0, curve 2 results from the pseudo-potential calculations [18], curve

3 is produced by a finite-depth potential well model (3.6 eV, m�e = 0.18m0, m�

h = 0.53m0) [18,127], curves 4–7 were reported as results of various modeling in [128] (curve 4), [129, 130] (curve5), [131] (curve 6), and [132] (curve 7). The results of TEM/XRD are presented by the hollowcircles [133], hollow squares [134], hollow triangles [135], filled areas [136] and filled triangles[137]

Fig. 6.6 CalibrationEg − d data for ZnO NPs.Curve 1 corresponds to theEMA with m�

e = 0.27m0,m�

h = 0.50m0 [138–140],curves 2 and 3 are the resultsof modeling presented in[141] (curve 2) and [9] (curve3).The results of TEM aregiven as the hollow squares[138], the results of the smallangle X-Ray scattering arepresented by the hollowcircles [138, 142, 143]

6.2 Studies of Nano-Photocatalysts by the Electron … 327

Fig. 6.7 CalibrationEg − d data for ZnS NPs.Curve 1 corresponds to theEMA with Ebulk

g = 3.6 eV,m�

e = 0.27m0, m�h = 0.58m0

[144, 145], curve 2 wasderived by modeling in [132].The TEM measurements aregiven by the hollow circles[146], squares [147], andtriangles [148]

Fig. 6.8 CalibrationEg − d data for PbS NPs.Curve 1 corresponds to theEMA with Ebulk

g = 0.41 eV,m�

e = 0.08m0, m�h = 0.075m0,

curves 2–4 reflect results ofmodelling reported in [149](curve 2), [150] (curve 3), and[151] (curve 4). TEM resultsare presented by the hollowsquares [151] and circles[152]; the filled trianglespresent results of the smallangle X-ray scattering [152]

Fig. 6.9 CalibrationE1 − d data for CdSe NPs,where E1 is the position of thefirst excitonic maximum inabsorption spectra. Curve 1corresponds to the EMA,curve 2–4 is the result ofmodeling reported in [153,154] (curve 2), [134, 155](curve 3), and [156, 157](curve 4). The TEM data arepresented by the hollowsquares [158], circles [159],and triangles [19], the filledtriangles present the data ofsmall angle X-ray scattering[24]

328 6 Probing with Light—Optical Methods in Studies …

spectra using such well-elaborated calibration curves is expected to be one of themost reliable and universal methods [8].

The graphical calibration dependences can be presented in an analytical form tomake the size determination even more convenient. Such analytical expressionsassociate the size of NPs with one of the parameters that can be derived from theabsorption spectra—the wavelength of band edge/furst excitonic maximum kbe/k1,the bandgap Eg or the first excitonic maximum energy E1. Several expressions ofthe kind are presented below for CdS NPs (9) [9], CdSe NPs (10–12) [8–10], CdTeNPs (13, 14) [8, 9], and ZnO NPs (15) [11].

dCdS ¼ �6:6521� 10�8k3be þ 1:9557� 10�4k2be�9:2352� 10�2kbe þ 13:2 ð9Þ

dCdSe ¼ 1:6122� 10�9k41�2:6575� 10�6k31 þ 1:624210�3k2�0:4277k1 þ 41:57

ð10Þ

dCdSe ¼ 59:60816�0:54736k1 þ 1:8873� 10�3k21�2:85743� 10�6k31 þ 1:62974� 10�9k41

ð11Þ

Eg CdSeð Þ ¼ 1:858þ 0:220d2 þ 0:008dþ 0:373� ��1 ð12Þ

dCdTe ¼ 9:8127� 10�7k31�1:7147� 10�3k21 þ 1:0064k1�194:84 ð13Þ

Eg CdTeð Þ ¼ 1:596þ 0:137d2 þ 0:206� ��1 ð14Þ

Eg ZnOð Þ ¼ 3:41þ 3:87� d�1:83 ð15Þ

Such expressions describe quite correctly the dependences of electron andoptical properties of the semiconductor NPs on their size in a broad range, typicallyfor d = 2–10 nm.

6.3 Luminescence Spectroscopy as a Tool for the Studiesof Nanocrystalline Semiconductors

The photoluminescence (PL) spectroscopy, both in the stationary and dynamicregimes, is one of the richest and most versatile methods for the studies of theexcited states of nanocrystalline semiconductors. Here, the potential of thesemethods is discussed on the example of solid solution cadmium-zinc sulfide NPs[1–3].

The optical, photochemical and photocatalytic properties of CdxZn1−xS NPs canbe tuned in a broad range by varying both the NP size and composition. The studies

6.2 Studies of Nano-Photocatalysts by the Electron … 329

of composition-selected CdxZn1−xS NPs synthesized in identical conditions can bevery useful to gain an insight into the nature and mechanisms of the defect-relatedPL in individual nanocrystalline cadmium and zinc sulfides [1, 3]. Despite thenumerous efforts on both nanosized and bulk crystals of these semiconductors, themechanisms of defect-related PL in CdS and ZnS still remain a subject of vividdiscussions. The present chapter also provides an example of the application of theresults of PL spectroscopy for the interpretation of special kinetic features of thephotocatalytic processes, in particular, the photocatalytic reduction of 4,4/-dime-thylbipyridyl (methylviologen) cation, with the participation of CdxZn1−xS andCdSe NPs.

We note also, that a detailed account of the structure of PL spectra of semi-conductor NPs, PL mechanisms and decay dynamics depending on the NP com-position and size, PL quenching by various substrates, etc. is beyond the scope ofthe present book. A deep insight into this subject can be found in specializedcomprehensive monographs [12, 13].

PL spectra of CdxZn1−xS NPs. The position of fundamental band edge ofCdxZn1−xS NPs synthesized in aqueous solutions depends only slightly on thenature of a stabilizer used and remains unchanged upon the NP introduction into thepolymer films [2, 3]. This allows to compare the results of PL spectroscopic studiesof both colloidal CdxZn1−xS and the polymer films with incorporated cadmium-zincsulfide NPs which are incomparably more stable than the original colloidal solu-tions. The PL spectra of such films reveal broad emission bands (Fig. 6.10a, leftpart) shifted considerably to lower energies (by 0.5–1.2 eV) as compared to thecorresponding absorption bands indicating unambiguously on the defect-relatednature of such emission [14, 15].

A comparatively large characteristic PL decay time, around 100 ns, also atteststo the participation of “deep” (in terms of energy relative to the edges of CB andVB) charge traps in the radiative electron-hole recombination [16, 17]. The trap

Fig. 6.10 a Absorption(curves 1a–5a) and PL spectra(curves 1b–5b) of gelatinfilms with CdxZn1−xS NPs.x = 0 (curves 1), 0.25 (2),0.50 (3), 0.75 (4), and 1 (5),PL was excited byhvexc = 2.81 eV; b PL energyin the maxima of D1 and D2

bands registered under theillumination withhvexc = 2.81 eV (filledcircles) and 3.82 eV (hollowcircles)

330 6 Probing with Light—Optical Methods in Studies …

states reside deep in the bandgap and typically arise from the point defects on theNP lattice and the undercoordinated surface atoms [16, 17].

The PL band shape of CdxZn1−xS can be approximated by a combination of twoGaussian profiles with the maxima at 2.16–2.36 eV (denoted as D1, Table 6.2) and1.85–2.00 eV (D2). The distance between D1 and D2 almost does not depend onx indicating that PL is of the same origin for all the NP compositions. As the molarfraction of zinc is increased in CdxZn1−xS NPs the PL bands show a blue shift,however, the shift magnitude is much lower than the corresponding shift of theabsorption band edge. Simultaneously, the gap between the PL band maximum andthe absorption band edge (the Stokes shift) increases by a factor of 2 and more asthe NP composition is varied from CdS to ZnS.

At least two alternative mechanisms of the defect-related PL in CdS and ZnSNPs [15, 18–21], and CdxZn1−xS NPs [22, 23] are proposed incurring the radiativerecombination either between a deeply trapped electron (e�tr ) with a free or shal-lowly trapped hole h+ (Fig. 6.11, route 1), or between a deeply trapped hole (hþ

tr )and a free or shallowly trapped electron e− (Fig. 6.11, route 2). In the latter case, aconsiderable increase of the PL energy hvem should be expected with an increase inthe molar Zn(II) fraction in CdxZn1−xS NPs because the ECB increases by around1.0 eV with x varied from 1 (CdS) to zero (ZnS). Alternatively, in the case of theradiative recombination with the participation of free or shallowly trapped holes(route 1) hvem is expected to depend only slightly on the NP composition, becauseEVB increases by mere 0.2 eV from CdS to ZnS. Table 6.2 shows that the PLmaxima shifts do not exceed 0.2 eV with a variation of x from 1.0 to 0, indicatingthat the PL results from the radiative recombination with the participation of deeplytrapped electrons e�tr .

Additional arguments in favor of the involvement of e�tr in the PL emission canbe derived from the size-dependence of the PL band position [3]. The effective holemass is much higher than the effective electron mass for both ZnS and CdS and,therefore, the CB level should be affected by the quantum size effects in a muchstronger way than the VB level. In other words, ECB would increase faster than EVB

as the NP size is decreased and we can expect a much stronger size dependence forthe PL route 2 with the participation of free e− than for the route 2 with theparticipation of free holes.

Table 6.2 Bandgap (Eg), PLbands maxima (D1, D2),FWHM, relative intensity (I2/I1), and the characteristic PLdecay time (s) of CdxZn1−xSNPs incorporated into gelatinfilms [3]

x Eg, eV PLmaximum,eV

FWHM, eV I2/I1 s, ns

D1 D2 D1 D2

1.00 2.6 2.16 1.85 0.34 0.51 4.57 115

0.75 2.7 2.18 1.88 0.35 0.52 3.89 130

0.50 2.9 2.22 1.92 0.38 0.55 2.26 140

0.25 3.2 2.26 1.97 0.39 0.59 2.13 155

0.00 3.6 2.36 2.00 0.49 0.64 0.27 –

Note accuracy of s determination is ±10 ns

6.3 Luminescence Spectroscopy as a Tool for the Studies … 331

The positions of D1 and D2 bands were found to be unchanged as the averagesize of CdS NPs was decreased from 5 to 4 nm. Additionally, the size-dependentPL data can be obtained by probing the NP ensemble with a light of differentenergy, that is, by selectively exciting smaller or larger NPs within the NPensemble. Figure 6.10b shows the results of such measurement for two differentexcitation energies hvexc = 2.81 eV (441.7 nm) and 3.82 eV (325.0 nm) forcomposition-selected 6–8 nm CdxZn1−xS NPs. It is clear, that the variation in D1

position does not exceed 50 meV for the case of the whole NP ensemble excitationat 3.82 eV and for the case of the selective excitation of a fraction of larger NPs at2.81 eV. This variation decreases as x is decreased and the exciton confinementbecomes less strong. The position of D2 does not depend on hvexc in the wholecompositional range allowing to assign this PL band to surface defect states becausethe I2/I1 intensity ratio is higher for hvexc = 3.82 eV when smaller NPs are involvedin the photoexcitation and PL emission.

The above-discussed observations show that the PL of CdxZn1−xS NPs is, mostprobably, the result of the radiative recombination between a deeply trappedelectron and a free/shallowly trapped hole [3]. This interpretation is corroborated bythe reported depths of electron trap levels in cadmium and zinc sulfides, that varyfrom 0.7–1.0 eV for ZnS [18, 20, 21, 24] to 0.4–0.7 eV for CdS [15, 21, 24].Figure 6.12 illustrates the primary photophysical processes for individual CdS andZnS NPs that can occur for any mixed CdxZn1−xS NPs in the frames of the pro-posed mechanistic interpretation. The defect-related states corresponding to D1 andD2 are presented as narrow bands with a width equal to the FWHM (full width onhalf-maximum) of the respective PL bands (see Table 6.2). This scheme is valid fora low density of the photogenerated charge when a possible change of the doubleelectric layer on the NP surface due to the electrons accumulation can be neglected.

Table 6.2 shows FWHMs of the D1 and D2 band for the composition-selectedCdxZn1−xS NPs. The bandwidth depends mostly on an energy distribution of thetrap states because a variation of the average NP size from 5 to 8 nm was not

Fig. 6.11 Schematic ofpossible routes ofdefect-related radiativeelectron-hole recombinationin semiconductor NPs

332 6 Probing with Light—Optical Methods in Studies …

accompanied by any appreciable variation in FWHM. The spectral width of D2

band is almost twice as high as that of the D1 band indicating that the latter is, mostprobably, associated with the radiative recombination on the NP surface defectshaving a broader energy distribution as compared to the volume NP defects. Anincrease in FWHM can also be expected for the electron transition D�

2 between thedeep electron and hole traps (Fig. 6.12).

As the molar Cd(II) fraction is increased both total PL intensity and the relativecontribution of the D2 band (I2/I1 ratio in Table 6.2) increase as well. The followinginterpretation of this tendency was suggested in [3]. The spatial exciton confine-ment becomes stronger as x is increased (due to an increase in aB from ZnS to CdS)resulting in an acceleration of the radiative recombination. This tendency is alsofavored by a decrease of the electron trap depth making stronger the overlappingbetween the wavefunctions of trapped electrons and holes. At that, the PL decaytime is reduced (see below) indicating an acceleration of the non-radiativerecombination as well. The latter factor limits the increase of PL intensity. Both theprobability of the non-radiative recombination and the radiative process D2 increase

Fig. 6.12 Possible routes of the defect-related radiative recombination in CdS and ZnS NPs withthe energies of corresponding electron transitions [3]

6.3 Luminescence Spectroscopy as a Tool for the Studies … 333

with x and, correspondingly, with aB, as a result of more probable migration of thephotogenerated charge carriers onto the NP surface [16, 17].

The kinetic PL decay curves of the composition-selected CdxZn1−xS NPs almostcoincide with the PL decay curve for the pure gelatin in the range of 0–20 ns(Fig. 6.13). The longer components of PL decay can be extracted by using a linearcombination of several single-exponential functions (16) [16, 17]:

IðtÞ ¼X

Ai expð�t=siÞ; ð16Þ

where i = 1–4, the amplitude Ai and the lifetime si are the fitting parameters. Theapproximation revealed three “fast” components common for all the studied filmsand related to the polymer (s1 = 1.0 ns, s2 = 5.0 ns, s3 = 10 ns). The presence of afourth component s4 = 120–150 ns (Table 6.2) is typical only for theNP-containing films and is, therefore, related to the radiative recombination insemiconductor NPs. The life time value varies only slightly in a broad range ofregistration wavelengths (550–750 nm) indicating the recombination occurs via theD1 and D2 channels with a comparable rate.

PL spectroscopy in the interpretation of the photocatalytic MV2+ reductionkinetics. The photocatalytic reduction of viologens—4,4/-derivatives of bipyridyloccupies a special place in the semiconductor photocatalysis. The most renownedof viologens—bication of 4,4/-dimethylbipyridyl or methylviologen (MV2+)(Fig. 6.14a) is soluble in water and can strongly adsorb on the surface of varioussemiconductors serving as an excellent electron relay between the semiconductorcrystals and other components of the photochemical system. The one-electronreduction of MV2+ results in the formation of intensely colored cation-radical MV•+

that is ideal for the spectrophotometric detection even at concentrations as low as10−7–10−8 M. The standard redox potential of the MV2+/MV•+ couple is inde-pendent of the solution pH and has a very favorable position (E0 = −0.44 V vs.

Fig. 6.13 Kinetic PL decaycurves ofcomposition-selected CdxZn1−xS NPs in gelatin films, PLexcitation—at 3.82 eV, PLregistration—at 1.81 eV

334 6 Probing with Light—Optical Methods in Studies …

NHE) between ECB of many photo-active semiconductors and typical substrates ofthe photocatalytic reactions (such as water or typical sacrificial electron donors).

The photocatalytic reduction of MV2+ was one of pioneer reactions studied forthe colloidal semiconductor NPs. The methylviologen appeared to be a very con-venient substrate for the studies of quantum size effects in the nano-photocatalystsbecause the quantum yield of MV2+ photoreduction typically can change quitestrongly with small variations of the CB level position induced by the spatialexciton confinement in the semiconductor NPs. The kinetics of MV2+ photore-duction and the character of PL quenching of semiconductor NPs by methylvio-logen ions often bear important information on the structure, lattice imperfectionand surface charge of nano-photocatalysts as well as on the dynamics of interfacialcharge transfers with the participation of semiconductor NPs. In this section, wefocus on special features of the kinetics of MV2+ reduction on the surface ofCdxZn1−xS and CdSe NPs that were interpreted using the PL spectroscopy [25, 26].In a similar way, the PL spectroscopy can be applied to the studies of othernanocrystalline semiconductors and systems.

Figure 6.14b shows the kinetic curves of MV•+ accumulation during the pho-tocatalytic reduction of MV2+ by sodium sulfite (a sacrificial donor) in the presenceof CdS, Cd0.25Zn0.75S, and CdSe NPs. In the case of CdSe NPs, the process rate ismuch higher on an initial stage as compared to CdS and Cd0.25Zn0.75S NPs,however, it decreases gradually as the reaction proceeds and finally MV•+ starts todisappear in some secondary reaction. At the same time, the MV•+ concentrationremains constant in the “dark” conditions and in the absence of CdSe NPs, the factsindicating that the consumption MV•+ is the result of a photocatalytic process,similarly to the main reaction of the MV2+ photoreduction.

The photochemical reduction of MV2+ by sodium sulfite without any photo-catalysts also reveals a “plateau” on the kinetic curves [25] due to the secondaryone-electron reduction of MV•+ to the colorless MV0 as a result of the directphotoexcitation of the dimerized MV•+ [26]:

Fig. 6.14 Structure ofmethylviologen cation (a) andkinetic curves (b) of thephotocatalytic reduction ofMV2+ by sodium sulfite in thepresence of CdS NPs (curve1), Cd0.25Zn0.75S NPs (curve2), and CdSe NPs (curve 3)[25]

6.3 Luminescence Spectroscopy as a Tool for the Studies … 335

2MV�þ , (MV�þ Þ2ðMV�þ Þ2 þ hv ! MV2þ þMV0

The similar conversion of methylviologen cation-radical into the neutral MV0

was reported for the photocatalytic reduction of MV2+ with the participation of Ru(bpy)3

3+ complex [27]. These data allow concluding that MV•+ disparition in thephotocatalytic system based on CdSe NPs also results from the photocatalyticmethylviologen cation-radical reduction to MV0. This conclusion leads to a ques-tion—why MV•+ is not reduced to MV0 in similar photocatalytic systems based onCdS and Cd0.25Zn0.75S NPs?

A diagram presented in Fig. 6.15 shows that the photogenerated CB electrons ofboth CdS and CdSe NPs have a potential high enough to reduce MV2+ to MV•+.However, the second process of MV•+ reduction to MV0 can occur only in the caseof CdSe NPs with ECB = −1.40 V (NHE).

Similar kinetic differences of MV•+ accumulation can be observed for anothercouple of photocatalysts—CdSe and Cd0.25Zn0.75S (Fig. 6.14b, curves 2, 3) thathave close ECB, respectively, −1.40 and −1.45 V. Here the CB potential of bothnano-photocatalysts is high enough to reduce MV•+ to MV0, however, this processoccurs only in the case of cadmium selenide NPs.

To interpret these kinetic differences, a conventional energy scheme of thephotocatalytic system that accounts only for the potentials of free CB electrons andVB holes, should be re-considered and alternative ways of the electron transfer canbe taken into account, in particular, with the participation of deep electron traps onthe NP surface associated with the corresponding sub-bandgap levels (Fig. 6.15)[25]. The exact position of such sub-bandgap states can be derived from the PLspectra of CdSe and Cd0.25Zn0.75S NPs.

Such a scheme describes in a much more comprehensive way possible electrontransfers with the participation of nano-photocatalysts abundant with various sur-face defects. The photogenerated electrons which have escaped the radiationlessrecombination and the interfacial transfer to MV2+ have the option to be trapped bythe surface states. The trapping results in a lowering of the electron energy(chemical potential) by around 0.2 eV in the case of CdSe NPs, but still, the trappedelectron can be transferred to the adsorbed methylviologen cation-radical. Forcadmium-zinc sulfide the trapping-associated loss is much higher reaching up to1.2 eV and making impossible the following electron transfer to MV•+.

The presented example shows that the interpretation and prediction of thephotocatalytic properties of nanocrystalline semiconductors require a combinedanalysis of the thermodynamic characteristics of a semiconductor, in particular, theECB and EVB levels, together with the spectral data on the nature, concentration andenergy of the surface defects, which act as the traps of the photogenerated chargecarriers and contribute greatly to the photochemical behavior of semiconductorNPs.

336 6 Probing with Light—Optical Methods in Studies …

PL spectroscopy in the studies of Cu–In–S NP-sensitized PEC solar cells. Thechalcopyrite copper indium sulfide (CIS) NPs can be prepared with large variationsin the stoichiometry while preserving the chalcopyrite structure with no additionalbinary phases [28–31]. Both stoichiometric CuInS2 and non-stoichiometric Cu–In–S NPs are photochemically stable and reveal broad absorption bands extending toaround 800 nm (for the bulk CuInS2 Eg = 1.5 eV) and quite high absorptioncoefficients exceeding 105 cm−1 [28–30]. A combination of these features makesCIS NPs a good candidate as a visible light harvester both for the solid-statephotovoltaic solar cells and for the liquid-junction photoelectrochemical(PEC) solar cells [28–34].

The CIS and core/shell CIS/ZnS NPs used as spectral sensitizers of the TiO2-based photoanodes in the PEC solar cells showed a remarkable correspondencebetween the PEC activity and PL properties [35]. In particular, at a constant Zn:Curatio and at a variation of the copper and indium content the PL intensity ofcore/shell CIS/ZnS NPs in solutions follows the PEC activity of the TiO2/CIS/ZnSheterostructures produced from such colloids. The data reported in [35] is anexample of the application of the PL spectroscopy as a diagnostic tool for theinvestigations of light-harvesting PEC systems.

Similarly to the broadly studied cadmium chalcogenide NPs, the formation of aZnS shell on the surface of CIS NPs results in a drastic enhancement of the PLintensity as well as in a considerable growth of the PEC activity of NPs coupled to

Fig. 6.15 Scheme of possible electron transitions in the photocatalytic systems comprising CdS,CdSe, or Cd0.25Zn0.75S NPs, MV2+, and Na2SO3. Reprinted with permissions from Ref. [25].Copyright (2010) Elsevier. hvPL is a defect-related PL quantum. The sub-bandgap levels arepresented as the bands with a center and width corresponding to the maximum and FWHM of thePL bands of semiconductor NPs

6.3 Luminescence Spectroscopy as a Tool for the Studies … 337

titanium dioxide. In particular, an increase in the molar Zn:Cu ratio from 0 (no zinc)to 20 resulted in a more than the 30-fold increment in the PL efficiency (Fig. 6.16a,curve 1). The photocurrent density measured for the TiO2/CIS/ZnS photoanodesbased on such NPs was found to increase from around 2.0 mA/cm2 for theuncovered “core” Cu–In–S NPs to 2.5 mA/cm2 for the core/shell CIS/ZnS NPswith a Zn:Cu ratio of 1:1 (Fig. 6.16a, curve 2). As opposite to the PL properties, afurther increase in the Zn content results in the deterioration of PEC activity ofCIS/ZnS NPs and for the thickest ZnS shell and the highest PL efficiency, atZn:Cu = 20:1, the photocurrent density is three times lower than for the originalcore CIS NPs. Apart from the PL enhancement, the doping of Cu–In–S NPs withzinc(II) occurs resulting in a blue shift of the absorption band edge and a con-comitant shift of the PL band maximum. These phenomena are well reported andcaused by the penetration of Zn2+ ions into the Cu–In–S core resulting in the latticereconstruction and band gap widening [30, 34, 36].

The dependencies presented in Fig. 6.16a can be understood by assuming thatthe deposition of a ZnS shell results in the elimination of surface structural defectsacting as the non-radiative recombination sites. At that, the competing processes ofthe radiative electron-hole recombination in CIS/ZnS NPs and the interfacialelectron transfer in the TiO2/CIS/ZnS heterostructures become much more efficientresulting in the PL enhancement and the photocurrent increase. However, as theZnS shell becomes thicker it impedes the interfacial electron transfer and promotesat the same time the electron-hole recombination in the CIS core. Therefore, at ahigher ZnS content, we observe a sharp decrease of the PEC activity of the TiO2/CIS/ZnS heterostructures with a steady growth of the PL intensity of core/shellCIS/ZnS NPs used for the preparation of the photoanodes.

Fig. 6.16 a Photocurrent density (curve 1) obtained for the TiO2/CIS/ZnS photoanodes and PLintensity of CIS/ZnS NPs (curve 2) as a function of the molar Zn:Cu ratio during the ZnS shelldeposition. b Photocurrent density (curves 1, 2) for the TiO2/CIS (curve 1) and TiO2/CISZnS(curve 2) photoanodes and PL intensity of CIS/ZnS NPs (curve 3) as a function of xCu [35]

338 6 Probing with Light—Optical Methods in Studies …

As can be seen from Fig. 6.16a, the PEC measurements showed that the peakphotocurrent density is achieved at a comparatively low Zn(II) content, that is, for arelatively thin ZnS shell on the surface of CIS NPs. At a Zn:Cu ratio of 1:1 (optimalfor the solar cell performance) the absorption spectra of uncovered and core/shellNPs are essentially identical) indicating that the Zn(II) doping effect is small tonegligible and can be ignored in such conditions.

An increase of copper content in the CIS NPs results in a gradual transformationof the structure of the fundamental absorption band of colloidal NPs and a generalgrowth of the absorbance in the UV and visible spectral range, but the absorptionspectra of uncovered CIS and core/shell CIS/ZnS NPs were almost identical foreach given xCu. Both types of NPs, the core CIS and core/shell CIS/ZnS, revealedalso similar dependences of the PEC activity on the composition (Fig. 6.16b).

The photocurrent density generated by the TiO2/CIS/ZnS photoanodes increasesas the xCu is changed from 0.2 to 1.0, then it peaks at a NP composition corre-sponding to Cu:In:S = 1:5:10 and decreases at xCu = 2.5 almost to the same valueas for the NPs with the minimal copper content. For the most active absorber NPswith xCu = 1.0, the increment of the PEC activity introduced by the ZnS shelldeposition reaches around 25% (Fig. 6.16b, curves 1, 2).

The dependence of the integral PL intensity of Cu–In–S/ZnS NPs on the xCufollows closely the relationship between the PEC activity and the copper contentand reaches the maximal value at xCu = 1.0 as well, dropping quite considerably ata higher copper content (Fig. 6.16b, curve 3). The PL intensity is, therefore, areliable indicator allowing to anticipate the PEC activity of colloidal CIS NPsbasing on the spectral PL data.

6.4 Studies of Nanocrystalline Semiconductorsby the Time-Resolved Photolysis Techniques

The flash photolysis allows generating a relatively high amount of photoexcitedmolecules and short-lived intermediates including radicals during a short lightpulse, allowing for their spectral identification and studies of the decay kinetics.This feature is a basic distinction between the flash photolysis and the conventionalstationary methods used to study the photochemical processes. The present sectionaims to exemplify the potential of the lamp and laser flash photolysis on severalsystems with the semiconductor metal sulfide NPs and more complexnanoheterostructures. It shows the versatility of the pulse photolysis in the studieson the accumulation of the excessive charge in the semiconductor NPs, thedynamics of interfacial transfer of the excessive electrons, and the photochemicaltransformations of NPs under the powerful photoexcitation. The metal sulfide NPsare unstable in the charged state and prone to the reductive photocorrosion andtherefore, the photocharging events cannot be studied for such NPs in the regime ofstationary illumination.

6.3 Luminescence Spectroscopy as a Tool for the Studies … 339

Pulse photoexcitation of CdxZn1−xS NPs and discharging dynamics: a case ofrelatively weak lamp pulse photoexcitation. The pulse photoexcitation of aqueouscolloidal SPP-stabilized CdxZn1−xS NPs induces a reversible blue shift of thefundamental absorption band edge that can be conveniently observed as anon-stationary bleaching (NB) band in the differential absorption spectra(Fig. 6.17a), as discussed earlier in Chap. 1. The shift is induced by the accumu-lation of excessive electrons near the CB edge resulting in a complete filling of theavailable states and an increase in the energy necessary for the photogeneration ofnew CB electrons—the so-called dynamic Burstein-Moss effect.

The NB band intensity of cadmium and cadmium-zinc sulfide NPs is propor-tional to the excessive charge density and decreases by a factor of *3 in the first1 ms after the light pulse (Fig. 6.17b) reaching zero in 3–5 ms after the photoex-citation. In aerated aqueous solutions with no other donor/acceptor present therelaxation of the NB band indicates the capture of the excessive electrons byoxygen and water molecules.

Figure 6.17a shows that the NB band intensity of the mixed CdxZn1−xS NPs is5–6 times higher than for CdS NPs indicating that the mixed NPs have a morepronounced capability of accumulating the excessive charge. Such a difference canoriginate from multiple reasons. As the molar Cd(II) fraction is decreased the PLintensity of CdxZn1−xS drops as well resulting in an increase of the number oflong-lived electrons capable of participating in the chemical reactions on themicrosecond time scale. Additionally, the FWHM of D1 and D2 PL bands ofCdxZn1−xS NPs (that is, the spectrum of possible mid-bandgap electron states) aswell as the trap depth increase as the Zn content is elevated (Table 6.2) favoring tothe accumulation and retaining of a higher excessive negative charge under thepulse photoexcitation.

The kinetic NB decay curves can be fitted with linear combinations of twosingle-exponential functions (Fig. 6.17b, solid line), similarly to the

Fig. 6.17 a Non-stationary bleaching spectra of colloidal Cd0.67Zn0.33S NPs (curve 1) and CdSNPs (curve 2); b Kinetic curve of the NB decay registered for Cd0.67Zn0.33S NPs at kreg = 440 nm.The hollow circles are experimental data, while the solid line represents a two-exponential fitting

340 6 Probing with Light—Optical Methods in Studies …

above-discussed PL bands of CdxZn1−xS NPs. The fact may be in indication on atleast two independent routes of the excessive electrons consumption. The charac-teristic lifetime of the first channel is 100–150 ls, while for the second one it ishigher than 650–700 ls. A comparison of the results of the flash photolysis and thePL spectroscopy allowed to conclude [3] that the two processes under discussioninvolve the electrons trapped by D1- and D2-type states, respectively. The fasterprocess with t1 = 100–150 ls can be associated with the consumption of electronstrapped by the surface D2 traps more easily accessible for the acceptors, while theslower process with t2 > 650 ls can be assigned to the acceptor reactions withelectrons trapped by the volume states D1.

Table 6.3 shows the characteristics NB decay lifetimes t1 and t2 at a differentregistration wavelength (kreg) for CdxZn1−xS with x = 0.67 and 0.40. In both cases,t2 increases at longer kreg, while t1 remains more or less constant. As a variation ofthe NB registration wavelength allows to probe selectively the CdxZn1−xS NPfractions of a different size, the observed “t2 − kreg” dependence can be interpretedas a result of the size-dependence of the rate of e�tr migration to the NP surface.

In the presence of electron-donatingNa2SO3, theNB signal intensity ofCdxZn1−xSNPs increases in a linear manner with the donor concentration (Fig. 6.18a, curve 1).At the same time, the concentrational dependence for another donor—Na2S revealed amaximum (Fig. 6.18a, curve 2) caused by the continuous photogeneration of poly-sulfide anions S2�n capable of capturing the excessive electrons ofCdSNPs [37, 38]. Inthe presence of intentionally introduced polysulfide, even at a comparatively smallS2�n concentration (*1 � 10−4 M) a considerable quenching of the NB signal isobserved. The possibility of disulfide reduction (S2�2 + 2e− = 2S2−) is additionallyconfirmed by the comparison of the standard potential of this reaction (−0.52 V vs.NHE) with the conduction band potential of CdS NPs (−0.8 V, NHE).

The systems, where S2�x is used to capture and shuttle the photogenerated chargecarriers, occupy an important place in the solar light harvesting with nanocrystallinesemiconductors. As shown in Chap. 4, the sulfide/polysulfide redox shuttle is one ofthe most efficient and popular in the electrolytes of the semiconductorNPs-sensitized solar cells. Despite the broad studies into the factors affecting thelight conversion efficiency and photocurrent generation mechanisms in such sys-tems, the primary light-induced charge transfer between CdS NPs and S2�x species,as well as the nature and fate of short-lived intermediates of these processes stillrequire a deeper understanding.

As mentioned earlier, the introduction of polysulfide into the colloidal CdS NPsolutions results in the quenching of NB bands, and a new broad transient absorptionband can be observed with a smeared maximum at 580–590 nm (Fig. 6.18b).

Table 6.3 Characteristic TBdecay lifetimes t1 and t2 forCdxZn1−xS NPs at differentkreg [3]

kreg, nm 425 430 435 440

t1, ls x = 0.40 140 130 130 140

x = 0.67 110 110 100 120

t2, ls x = 0.40 910 940 970 1780

x = 0.67 640 720 940 1570

6.4 Studies of Nanocrystalline Semiconductors by the Time-Resolved … 341

It is well reported that a relative content of various S2�x species (with a different x)depends primarily on the ratio of S2− and S0 during the polysulfide formation, as wellas on the solution pH [39–41]. The S2�x distribution diagrams presented in [39–41]show that for [S2−]:[S0] = 1 (the conditions of the experiments under discussionhere) the solutions contain predominantly S2�3 and S2�4 anions in a ratio close tounity.

The analysis of reported literature data [42–46] allowed to assign the positivetransient band peaked at kmax = 580–590 nm to S��3 anion-radical. Other radicalsthat can potentially form in given experimental conditions absorb in a differentspectral range, for example, O��

2 at kmax = 240–245 nm [47], HS•—kmax = 290–330 nm [48], free S•−—kmax = 260 nm [47], S•− bound on the CdS surface anddepending on the NP size—kmax = 450–500 nm [48–50], H2S2

•−—kmax = 380 nm[47], S��2 —kmax = 400 nm [42, 43, 46], SO��

2 —kmax = 365 nm [47],S��4 —kmax = 513 nm [46].

The most probable way of S��3 generation in the presence of CdS NPs is theoxidation of polysulfide species by the photogenerated CdS valence band holes:

hþVB þ S2�3 ! S��3

As the sulfide ions get oxidized during the pulse photolysis to the elementalsulfur, the latter interacts with the excess of S2− producing polysulfide and the S2�x

Fig. 6.18 a Dependences of the NB signal intensity for 6–8 nm CdS NPs on the concentration ofNa2SO3 (curve 1) and Na2S (curve 2), b Transient differential absorption spectra of colloidal CdSsolution containing 1 � 10−3 M Na2S registered consecutively one after the other (curves 1–3).A total number of pulses in the registration range of 410–690 nm (10 pulses per point in average)is 280 (1), 560 (2), and 840 (3). Curve 4 obtained for CdS solution with a sodium polysulfideaddition. [CdS] = 1 � 10−3 M [37, 160]

342 6 Probing with Light—Optical Methods in Studies …

concentration gradually increases during the flash photolysis experiment as evi-denced by a gradual increase of the transient band intensity (Fig. 6.18b).

The decay kinetics of the S2�x pulse photolysis products cannot be describedeither by the first-order or second-order models indicating a complex decaymechanism with several possible reaction of S��3 , for example the recombination(second-order reaction) [45] or a reaction with the molecular oxygen [47]. Asoxygen is present in a concentration at least by 4 orders higher than the intermediateconcentration this reaction is of the pseudo-first kinetic order.

S��3 + O2 ! S03 + O��2

S��3 + S��3 ! S2�6

The characteristic lifetime of S��3 produced in the presence of CdS NPs is 2–3times higher than in homogeneous solutions by the direct photoexcitation ofpolysulfide, most probably due to the S��3 adsorption on the NP surface and theinhibition of both decay processes.

Pulse photoexcitation of CdxZn1−xS NPs and discharging dynamics: a case ofrelatively strong laser pulse photoexcitation. Under the photoexcitation with laserpulses with a photon density at least by an order of magnitude higher than in thecase of lamp flash photolysis, CdxZn1−xS NPs show some special features due tothe probability of the formation of several electron-hole couples in a sole NPswithin the time scope of a laser pulse.

Illumination of colloidal CdxZn1−xS with the laser pulses at k = 355 nm resultsin spectral changes similar to those observed in the case of lamp photolysis—aBurstein blue shift of the absorption band edge and a rise of the NB band in thedifferential absorption spectra indicating the accumulation of an excessive charge.The NB band relaxes during tens-hundreds microseconds (Fig. 6.19a) owing to thereactions between excessive electrons and the solution components. As x is reducedfrom 1.0 to 0.2 the NB band maximum shifts to shorter wavelengths from 460 to375 nm (Fig. 6.19b) mimicking the corresponding shift of the fundamentalabsorption band edge of CdxZn1−xS NPs [51, 52].

When oxygen is bubbled through the colloidal solution the intensity of NB banddecreases indicating on the electron scavenging by oxygen molecules [52]:e� þO2 ! O��

2 (Fig. 6.19a). The oxygen exerts only a partial NB quenching and,therefore, the above reaction cannot be the sole or even the main fate of theexcessive electrons generated by the laser pulse [52]. Taking into account thephotocatalytic properties of CdxZn1−xS in the reduction of water to hydrogen(discussed in Chap. 2), the reduction of water by excessive electrons can beregarded as a principal mechanism of the NB decay in the case of cadmium-zincsulfide NPs [52]: 2e− + 2H2O ! H2 + 2OH−. Therefore, by studying the NBdecay dynamics we can derive important information on the mechanism and lim-itating factors of the photocatalytic water reduction.

6.4 Studies of Nanocrystalline Semiconductors by the Time-Resolved … 343

Another special feature of the pulse-excited CdxZn1–xS NPs is a volcano-shapeddependence between the NP composition (x) and the NB band intensity(Fig. 6.19c). As x is decreased from 1.0 to around 0.6 the NB band intensityincreases, reaches a peak value at x = 0.6–0.7, and decreases at a further increase ofthe Zn(II) fraction in CdxZn1−xS NPs down to zero at x = 0.2.

The interpretation of the dome-shaped dependence presented in Fig. 6.19c wasperformed [52] basing on a model introduced for the oxygen one-electron reductionprocess with the participation of the radiolytically-charged Ag NPs [53]. Accordingto the model, the NP discharging rate (decay of a charge Q) in this reaction dependson the oxygen reduction over-voltage DE and can be expressed as

� dQdt

¼ ke�aFRTDE; ð17Þ

where a is a constant, F is the Faraday number, R is the universal gas constant, T istemperature (К).

The over-voltage can be expressed as DE = –Q/C, where C is the electriccapacitance of semiconductor NPs, Q is a charge per a NP at a moment t. Byintegrating Eq. (17) and a logarithmic transformation, Eq. (18) can be derived [52]:

Q ¼ RTCaF

lnð aRTCk

Þ � ln t� �

: ð18Þ

A dependence between the excessive voltage Q and the optical density of NBband can be expressed as Q = bDD, where b is a constant [54]. Therefore,

DD ¼ A� B ln t; ð19Þ

Fig. 6.19 a Kinetic curves of NB decay registered in the NB band maximum in N2-saturated(curve 1) and O2-saturated (curve 2) colloidal CdS solutions. b, c Positions of the NB bandmaximum (b) and the NB band intensity (c) of colloidal CdxZn1−xS of a different composition [52]

344 6 Probing with Light—Optical Methods in Studies …

where

A ¼ RTCaFb

lnð aRTCk

Þ and B ¼ RTCaFb

: ð20Þ

Figure 6.20a shows that there indeed exists a linear dependence between DD andln(t), which is in accordance with the experimental results. The tangent of suchlinear dependence and, therefore, the NB decay curve shape appear to be deter-mined mostly by the electric capacitance of the semiconductor NPs.

By making reasonable assumptions on the values of a and b [52] the capacitanceof a CdxZn1−xS NP of a given composition can be calculated using Eq. (20). Inparticular, the a = 0.3 can be taken by analogy with the photocatalytic reduction ofMV2+ in the presence of SPP-stabilized CdS NPs reported in [55], while b = 0.3was calculated in [52] from a dependence of the NB band intensity on the laserpulse intensity. The NP capacitance C was found to be 0.014 F � L−1 for CdS NPs,0.043 F � L−1 for CdxZn1−xS NPs with x = 0.8 and 0.068 F � L−1—with x = 0.5.The laser flash photolysis experiments reported in [52] were performed usingCdxZn1−xS NPs with an equal average size of 6.0 ± 0.5 nm regardless of theircomposition. The electric capacitance per NP estimated using the size value wasfound to be 0.2 � 10−18 F (x = 1.0), 0.7 � 10−18 F (x = 0.8), and 1.1 � 10−18 F(x = 0.5), which is close to the lower limit of the reported range of the capacitanceof colloidal 10 nm CdS NPs determined by other methods, (6–60) � 10−18 F [54].

An increase in the electric capacitance of CdxZn1−xS NPs with increasingx originates, most probably, from a disordering of the mixed sulfide lattice causedby the difference in the ionic radii of Cd2+ and Zn2+ and a different rate of the CdSand ZnS formation during the co-precipitation. The lattice defects generated at thesynthesis can act as charge traps prohibiting free migration of the photogenerated

Fig. 6.20 a Kinetic NB decay curves registered in the NB band maxima of CdS NPs (curve 1),Cd0.8Zn0.2S NPs (curve 2), and Cd0.5Zn0.5S (curve 3) presented as DD(t) versus –ln(t) [52]. b,c NB decay curves of CdS NPs (b) and Cd0.8Zn0.2S NPs (c) in the corresponding NB band maxima(480 nm in (a) and 460 nm in (b)) registered without additions (curves 1) and in the presence ofZnSO4 (curves 2) and Na2S (curves 3). Insert in (b) presents kinetic curves as DD(t) vs. –ln(t),solid lines correspond to the linear fits of the experimental data. Insert in (c) presents PL spectra ofCdS NPs (curve 1), Cd0.5Zn0.5S NPs (curve 2), and Cd0.2Zn0.8S NPs (curve 3) [52]

6.4 Studies of Nanocrystalline Semiconductors by the Time-Resolved … 345

charge carriers and their recombination. Another possible reason for the x-depen-dence of the capacitance of CdxZn1−xS NPs can be a deviation from the feedstoichiometry set at the synthesis (a ratio of Cd(II) and Zn(II) concentrations) [2]due to the formation of various hydroxo-complexes of zinc that react much slowlywith sulfide anions. These forms of Zn(II) can adsorb on the NP surface and also actas traps capturing the photogenerated electrons and thus increasing the overall NPcapacitance.

The above assumptions are corroborated by the results of the flash photolysis ofCdS and Cd0.8Zn0.2S NPs in the presence of intentional additions ofover-stoichiometric ZnSO4 and Na2S [52]. Figure 6.20b shows that a Zn2+ additiononly slightly affects the NB band intensity DD0, but results in an increase of thetangent of the DD—ln(t) dependence (insert) indicating an increase of the NPcapacitance (see Eq. 19). At the same time, an addition of Na2S results in a con-siderable reduction of DD0 and the NP capacitance. A similar effect of both addi-tions was observed for the Cd0.8Zn0.2S NPs (Fig. 6.20c), but in this case theintroduction of Zn(II) decreases the NB band intensity and this effect becomes morepronounced with a further decrease of x.

A deviation from the non-stoichiometry of CdxZn1−xS NPs results also in adrastic increase in the PL intensity as x is decreased from 1.0 to 0.2 (Fig. 6.20c,insert), because the Zn(II) species adsorbed on the NP surface can participate bothin the electron accumulation and the radiative electron-hole recombination. Anincrease in the charge accumulation rate will inevitably cause a decrease in the PLefficiency. In the case of polymer-incorporated CdxZn1−xS NPs (see discussionabove) the polymer passivates efficiently the surface under-coordinated Zn(II) andan inverse tendency is observed [3].

The discussed results of the flash photolysis coupled with the PL spectroscopydata allowed to conclude [52] that a volcano-shaped dependence between the NBband intensity and the composition of CdxZn1−xS NPs (Fig. 6.19c) originates fromthe overlap of two tendencies—(i) an increase of the NP capability of accumulatingthe excessive charge with a decrease of x and (ii) a increase of the probability of theradiative electron-hole recombination with a decrease of x.

Laser flash photolysis of TiO2/CdS film nanoheterostructures. As discussed inChaps. 2–4, TiO2/CdS is one of the most broadly studied photo-activenanoheterostructures. The application of the flash photolysis to the studies ofcharge separation in this composite allowed to derive important and quite uniqueinformation on the dynamics of the photogenerated charge carriers in the TiO2/CdSheterojunction with a CdS layer produced by different methods.

The fundamental band edge of nanocrystalline titania deposited on glass [56] isaround 360–370 nm (Fig. 6.21a, curve 1) and, therefore, the laser pulses withk = 355 nm (3.5 eV) can excite interband electron VB–CB transitions.A differential absorption spectrum of the TiO2 films reveal a broad band in therange of 670–710 nm with a peak at k = 680–690 nm (Fig. 6.21a, insert) [57]. Theband can be assigned to a long-lived intermediate as the transient signal shows nosigns of decay during hundreds of ls after the pulse extinction (Fig. 6.21b). Similarbands are typically observed [58, 59] in the case of the CB electron capture by deep

346 6 Probing with Light—Optical Methods in Studies …

traps (Ti4+ ions) resulting in the Ti3+ generation (Ti4þ þ e�CB ! Ti3þ ). The VBholes are also rapidly captured by the hole traps—typically surface hydroxideanions (hþ

VB þOH� ! OH�) [58] or interact with the donor compounds adsorbedon the TiO2 NP surface. A large portion of the photogenerated charge carriersdecays in the recombination processes, that have a predominantly non-radiativecharacter for the nanocrystalline titania [58].

To increase the transient Ti3+-related band intensity a thin transparent layer ofsucrose was applied on top of the TiO2 films that does not interfere with the lightabsorption but supplies additional electrons as a sacrificial donor [60].

When nanocrystalline cadmium sulfide is deposited on the titania surface by thechemical bath deposition (CBD) a new absorption band appears with an edge at510–520 nm (Fig. 6.21a, curve 3) while the optical density of the TiO2/CdS film atthe laser wavelength (3550 nm) increases to 1.75 indicating a complete lightabsorption by cadmium sulfide. Such TiO2/CdS nanohetero–structure showedalmost zero intensity of the transient signal at 670–710 nm (Fig. 6.21b, curve 1).The fact indicates that the efficiency of interfacial electron transfer from the pho-toexcited CdS to TiO2 followed by the electron capture and formation of Ti3+ isvery low for this TiO2/CdS composite, despite the favorable thermodynamic con-ditions (ECB(CdS) = −0.8 V vs. NHE [61], ECB(TiO2) = −0.3 V at pH 7 [61]).

The electron transfer can be hindered by an interfacial barrier between titania andCdS, because the CBD of cadmium sulfide typically yields hexagonal CdS NPs[62], while TiO2 is crystallized in a cubic anatase modification [63]. Also, theCDB-deposited CdS NPs show drastic recombinational losses of the photogener-ated charge carriers as evidenced by a large increase of the transient signal intensity

Fig. 6.21 a Absorption spectra of the nanocrystalline TiO2 film (curve 1) and TiO2/CdSnanoheterostructures (curves 2, 3) produced by the photocatalytic CdS deposition (2) and the CBD(3) [56]. Insert in (a): transient absorption spectrum of TiO2 film excited by the laser pulses withk = 355 nm. b Kinetic curves of the transient signal decay at k = 680 nm registered for theCBD-deposited TiO2/CdS films (curves 1, 2) and the photodeposited TiO2/CdS films (curves 3, 4).Curves 2 and 4 registered after the deposition of a sucrose layer [57]

6.4 Studies of Nanocrystalline Semiconductors by the Time-Resolved … 347

in the presence of sucrose capable of capturing of the photogenerated holes(Fig. 6.21b, curves 1, 2) and interfering with the recombinative processes.

The flash photolysis of similar TiO2/CdS nanoheterostructures produced by thephotocatalytic deposition of cadmium sulfide NPs [56] results in the same tran-sients, however, the signal intensity is much higher in the presence of sucrose inthis case (Fig. 6.21b, curves 2, 4). After a correction on the light absorbance, thedifference in the transient signal intensity between both nanoheterostructuresincreases additionally by a factor of 3. The observations indicate that the efficiencyof the photoinduced interfacial electron transfer from CdS to TiO2 and the for-mation of Ti3+ species is by an order of magnitude higher for the photocatalyticallyproduced TiO2/CdS nanoheterostructure as compared with the analog synthesizedby the conventional CBD [56].

As opposite to the bare titania films, the TiO2/CdS nanocomposites revealed asecond quite intense transient signal in the range of 420–570 nm peaked at 470–500 nm (Fig. 6.22a). This band can be assigned to surface-adsorbed S•− radicalsformed via the photogenerated VB hole capture by the deep traps (lattice S2−

anions).The transient band observed for the CDB-deposited TiO2/CdS nanocomposite

seems to be composed of two spectral components. However, the kinetic decaycurves registered for this band on different wavelengths (460, 490, and 520 nm) arethe same and, therefore, describe the decay of a single short-lived intermediate. The

Fig. 6.22 a Transient differential absorption spectra of the TiO2/CdS nanohetero–structuresproduced by the CBD (curve 1) and the photocatalytic CdS deposition (curve 2); b Normalizedkinetic curves of the transient signal decay for the CBD-produced TiO2/CdS registered at 500 nm(curve 1) and for the photodeposited composite registered at 470 nm (curve 2) [57]

348 6 Probing with Light—Optical Methods in Studies …

band distortion can be caused by an overlap of the positive transient signal with a NBband of CdS NPs bearing excessing negative charge (as discussed in the previoussection). The NB band maximum for the given TiO2/CdS nanoheterostructure isexpected to be at 480–500 nm thus indeed overlapping with the absorption band ofS•− anion-radical. The NB band presence indicates that the photogenerated CdS CBelectrons accumulate on CdS NPs as a result of a low efficiency of the interfacialtransfer to TiO2 NPs. A reconstruction of the S•− absorption band (Fig. 6.22a,dashed line) shows that the band maximum should be observed at 485–490 nm.

The intensity of S•− related band of the photocatalytically-produced TiO2/CdSnanocomposites is twice as high as for the CBD-produced analog, its peakblue-shifting to 465 nm. The first observation is in accordance with an increase inthe Ti3+ signal intensity at 670–710 nm, while the second one illustrates awell-reported blue shift of the adsorbed S•− band maximum with a decrease of theCdS NP size [64, 65].

The decay dynamics of the sulfur anion-radical is also different for theCBD-deposited and the photodeposited TiO2/CdS nanoheterostructures(Fig. 6.22b). The kinetic curves have a complex shape that cannot be fitted with asimple first-order or second-order kinetic model. The complexity attests to severalsimultaneous reactions with the participation of S•−. Also, it can arise from a sizedistribution of CdS NPs.

Figure 6.22b shows that the CBD-deposited CdS NPs show a sharp decrease ofthe transient signal intensity in the first 3–5 ls after the laser pulse followed by aslower signal relaxation till the zero level (at t > 50 ls). At the same time, no fastcomponent can be observed in the decay curves of the photodeposited TiO2/CdSnanocomposites (Fig. 6.22b, curve 2). The decay is generally slower and more thana half of the photogenerated S•− anion-radical survives as long as 50–100 ls after theexciting pulse. The differences in the decay curve shape cannot be explained solelyby possible differences in the rate of radical recombination (S•− + S•− ! S2

2−) orthe interaction with oxygen (S•− + O2 ! S0 + O2

•−).Taking into account the presence of excessive electrons on CdS NPs in the

CBD-deposited TiO2/CdS nanoheterostructure the different decay dynamics can beassigned to the recombination of the excessive electrons and S•− anion-radicals (Cd(II)S•− + e�tr ! CdS) [56]. In the case of photoproduced TiO2/CdS nanocompos-ites, this process is blocked by the efficient electron transfer from CdS NPs to thetitania scaffold. Formally, the latter reaction corresponds to the electron recombi-nation with the deeply trapped hole. The recombination of free charge carriers inCdS NPs is typically over in 1–10 ns after the photoexcitation [64], but the processcan be extended to 200–300 ns if one of the carriers gets captured by the deep trap[3, 64]. It can be safely assumed that the recombination between the deeply trappedelectron and deeply trapped hole will occur by 1–2 orders of magnitude slower, thuscorresponding to the discussed time scale.

6.4 Studies of Nanocrystalline Semiconductors by the Time-Resolved … 349

6.5 Studies of Nanocrystalline Semiconductors UsingRaman Scattering Spectroscopy

The Raman spectroscopy is used much rarer for the studies of semiconductornano-photocatalysts and photo-electrodes as compared to the electron absorptionand PL spectroscopies. However, in many cases, the Raman spectroscopy providesa unique information on the nano-photocatalyst structure, especially in the studiesof mixed compounds and multi-component nanostructures.

For colloidal and thin-film nanocrystalline semiconductors comprising a smallamount of a target phase, the resonant Raman spectroscopy appeared to be the mostproductive, when the samples are excited by the wavelength corresponding to thespectral range of maximal absorbance, for example, into the excitonic band max-imum. For example, the Raman spectrum of CdSe NPs incorporated into the gelatinfilms shows no distinct semiconductor-related features under the illumination withk = 647.1 nm which is not absorbed by the NPs. At the same time, the resonantexcitation at k < 550 nm, that is, into the absorption band of cadmium selenideNPs, allows to register the characteristic CdSe phonon bands. The peak positionsdepend on the excitation wavelength as a result of the selective photoexcitation ofdifferently sized fractions of CdSe NPs in the incorporated NP ensemble [66].

The Raman spectrum of the nanocrystalline sample can be used for the deter-mination of the phase composition because many photochemically active semi-conductors have characteristic vibrational frequencies. For example, the mainphonon mode (longitudinal optical phonon—LO) of CdS, CdSe, and CdTe can beobserved at 305 (Fig. 6.23a), 210, and 170 cm−1 [67]. The characteristic LO fre-quencies of ZnO, ZnSe, and ZnSe are 350, 580, and 250 cm−1 (Fig. 6.23c),respectively [67]. The anatase modification of titanium dioxide has six activevibrational modes, with the most intense in the Raman spectra being at 144 cm−1

Fig. 6.23 Raman spectra of CdS NPs deposited onto ZnO surface (a) [161], nanocrystalline TiO2

film (b, curve 1) and TiO2/Sb2S3 nanoheterostructures with amorphous (curve 2) and crystalline(curve 3) antimony sulfide NPs [162], and ZnSe NPs (c) [163]

350 6 Probing with Light—Optical Methods in Studies …

(Fig. 6.23b, curve 1) allowing for a secure spectral distinction between the anataseand rutile [68].

The presence of LO overtones at a double (2LO) and a triple (3LO) mainfrequency (Fig. 6.23a) is a sign of high crystallinity and high structural order of theNP lattice. In the case of Sb2S3 NPs deposited on titania as a spectral sensitizer, theRaman spectroscopy can provide a definite proofs of amorphous (Fig. 6.23b, curve2) of crystalline (curve 3) character of antimony sulfide, the latter revealing a muchmore resolved picture of possible vibrational modes of the crystalline stibnitelattice.

The Raman spectra of small NPs with a developed surface area often revealadditional spectral “shoulders” shifted to lower frequencies as compared to the mainLO peak (Fig. 6.23c). Such peaks are typically assigned to surface optical(SO) phonons of NPs and a ratio of LO and SO phonon peaks can be used as ameasure of the surface area and/or structural disorder of the semiconductor NPs.

In some cases, the Raman spectroscopy can be used to evaluate the size ofsemiconductor NPs. Similarly to the size determination from the optical absorptionspectra, this procedure is only possible for the semiconductors with reportedempirical correlations (or calculated ones) between the NP size and some of thespectral parameters of Raman spectra, such as the position or FWHM of the LOpeak. For example, the size of CdSe NPs can be estimated from the spectral positionof the LO peak for the size range of d < 6–7 nm (Fig. 6.24a). The dependenceoriginates from the spatial phonon confinement resulting in a gradual decrease ofthe LO peak frequency as the NP size is reduced [69–72]. A correlation between theFWHM of the most intense Raman peak of TiO2 NPs and the average NP size isreported (Fig. 6.24b) allowing the NP size to be estimated in the range of d < 15–20 nm.

Fig. 6.24 Correlations between the size and PL frequency of CdSe NPs (a) and between FWHMof Eg(1) phonon peak and the size of TiO2 NPs (b). Adapted with permissions from [69] (a) and[164]. Copyright (1998, 2005) The American Physical Society

6.5 Studies of Nanocrystalline Semiconductors … 351

The Raman spectroscopy, especially in the combination with the X-raydiffraction and the energy-dispersive X-ray spectroscopies, is a powerful methodfor the determination of the composition of mixed solid-solution semiconductornanomaterials, such as CdxZn1−xS and CdSxSe1−x. The phonon peak positions inthe Raman spectra of these compounds change in a monotonous way with the NPcomposition between the positions of the spectral signals of separate components.Figure 6.25 exemplifies this approach for the CdSxSe1−x solid-solution NPs [73].By using the cadmium sulfoselenide NPs of a known composition, a calibrationcurve can be plotted for the composition-dependent positions of phonon frequencies(Fig. 6.25b) that can be used for the determination of the composition of CdSxSe1−xsamples with unknown x.

The latter approach was implemented for the determination of real compositionof ZnO/CdxZn1−xS photoanodes produced by the SILAR [74]. The SILAR depo-sition from aqueous mixed solutions of Cd(II) and Zn(II) nitrate results in theformation of CdxZn1−xS on the ZnO film surface evidenced by a new absorptionband with the edge shifting to shorter wavelength as the molar cadmium fraction,x0 = [Cd(II)]/([Cd(II)] + [Zn(II)]), was decreased (Fig. 6.26a). In particular, adecrease of x0 from 0.9 to 0.1 results in a blue shift of the absorption band edge ofCdxZn1−xS from 500–505 nm (Eg = 2.46–2.48 eV) to 460–465 nm (Eg = 2.67–2.70 eV). As discussed earlier in this chapter, the size of cadmium-zinc sulfide NPsdoes not depend considerably on their composition and, therefore, the shift of theabsorption band edge of the SILAR-deposited CdxZn1−xS NPs can be assignedexclusively to a variation of the NP composition. The estimations performed usingthe above-described approach and the empirical Eq. (5) showed that the real molarCd(II) fraction in CdxZn1−xS NPs derived from the absorption spectra, xabs, is much

Fig. 6.25 a Raman spectra of colloidal CdSxSe1−x NPs at x = 0 (curve 1), 0.2 (curve 2), 0.5(curve 3), 0.8 (curve 4), and 1.0 (curve 5). Insert: compositional dependences of the LOCdS

(squares) and LOCdSe (circles) peaks of CdSxSe1−x NPs [165]. Solid lines represent similardependences reported in [166]. Reprinted with permission from Ref. [73]. Copyright (2010)Springer. b Dependence of the LO peak position on the composition of CdxZn1−xS NPs. Plottedusing data reported in [67]

352 6 Probing with Light—Optical Methods in Studies …

higher than the nominal Cd(II) fraction x0 and varies from 0.94 to 0.67 as x0 isdecreased from 0.9 to 0.1. Such a strong difference between x0 and xabs originates,most probably, from a different adsorption of Cd2+ and Zn2+ ions on the ZnOsurface, the partial hydrolysis of Zn(II), as well as the lower solubility of CdS.

Figure 6.26b–d shows the Raman spectra of ITO/ZnO/CdxZn1−xS films excitedat kexc = 325 nm (3.82 eV) and 514.5 nm (2.42 eV). The spectra reveal a peak at300–320 cm−1 that was assigned to the LO mode of CdxZn1−xS NPs. Thelow-frequency wing of the peak at 250–270 cm−1 may be ascribed to the lightscattering on the SO phonons as discussed earlier for ZnSe NPs. A largesignal-to-noise ratio in the Raman spectra of ZnO/CdxZn1−xS films prepared atx0 = 1.00, 0.75, and 0.50 registered at kexc = 514.5 nm attests to the resonantcharacter of the scattering as a result of close energies of the exciting quanta and thebandgaps of CdxZn1−xS NPs.

The composition of CdxZn1−xS solid solution can be determined with a goodaccuracy either from the LO phonon frequency mLO using the well-known empirical

Fig. 6.26 a Normalized absorption spectra of the ITO/ZnO/CdxZn1−xS films synthesized atx0 = 1.0 (curve 1), 0.9 (curve 2), 0.6 (curve 3), 0.4 (curve 4), and 0.1 (curve 5). b Raman spectra ofthe ITO/ZnO/CdxZn1−xS films produced at x0 = 0.25 (curves 1), 0.50 (curves 2), 0.75 (curves 3),and 1.0 (curves 4). kexc = 514.5 nm (b) and 325 nm (c, d) [74]

6.5 Studies of Nanocrystalline Semiconductors … 353

equation x = 0.013 � (mLO − 303) [75] or directly from the LO peak position usingthe reported calibration curves [76–78]. The real molar Cd(II) ratios in CdxZn1−xSNP deposited on the ZnO surface determined in this way, xRRS, are presented inFig. 6.27. Despite the fact that xabs and xRRS were estimated for two sets ofITO/ZnO/CdxZn1−xS samples with different x0, they perfectly match and comple-ment each other thus producing a reliable calibration curve allowing to determinethe real composition of SILAR-deposited CdxZn1−xS NPs with x0 varied from 1.0to zero.

Complementary to the optical data, the surface of ITO/ZnO/CdxZn1−xS filmswas studied by EDX allowing to quantify the atomic composition of the films. Itshowed that the atomic Cd:S ratio is very close to 1:1 at x0 = 1 and decreases withdecreasing x0 indicating the formation of mixed cadmium-zinc sulfide. Similarly tothe optical data, the EDX shows a strong deviation of the real composition of thefilms relative to the Cd:Zn ratio set at the SILAR procedure. The values of realcomposition of CdxZn1−xS NPs determined by EDX, xEDX, appeared to be in aperfect accordance with the results of the optical absorption and Raman spectro-scopies (Fig. 6.27) attesting to the high accuracy and reliability of theabove-discussed optical methods.

The Raman spectroscopy of nanoheterostructures comprising two and morecomponents provides ample information on their structure and allows to distinguishbetween the core/shell NPs with a continuous shell and an island-like shell. Also, itscan reveal an effect of the interdiffusion of the components having close parametersof the crystal lattice. For example, the resonant Raman spectra of core/shellCdSe/ZnS NPs show a distinct phonon peak at 300 cm−1 typical for cadmiumsulfide (Fig. 6.28) indicating the interdiffusion of the materials of the shell (zincsulfide) and the core (cadmium sulfide) [79–81].

Fig. 6.27 Calibrationdependence between themolar Cd(II) fraction x0 in amixed Cd(II)-Zn(II) solutionused for the SILARdeposition of CdxZn1−xS andthe real molar Cd(II) fractionin ITO/ZnO/CdxZn1−xSnanocomposites. The valueswere determined using theoptical absorptionspectroscopy (xabs, squares),the resonant Ramanspectroscopy (xRRS, circles),and the energy-dispersiveX-ray spectroscopy (xEDX,diamonds) [74]

354 6 Probing with Light—Optical Methods in Studies …

The Raman spectroscopy was fruitfully used in the studies of binary CdS/CdSenanoheterostructures synthesized by the photocatalytic reduction of Na2SeSO3 onthe surface of cadmium sulfide NPs [82]. The resonant Raman spectra of CdS/CdSeshowed the LO peaks of both components with the CdSe LO peak growing with anincrease in the photodeposited CdSe content (Fig. 6.28b). The main phonon peak of6–7-nm CdS NPs used as a photocatalyst can be found at 302 cm−1 shifting slightlyto lower frequencies as compared to the bulk cadmium selenide (305 cm−1 [67]).The small shift magnitude attests to a weak phonon confinement in such NPs.However, the CdS NPs revealed a much higher FWHM of the LO band (70–80 cm−1) as compared to both the photodeposited CdSe (40 cm−1) and the CdSeNPs prepared in “dark” conditions via the interaction between Na2SeSO3 andCdCl2 (10–20 cm−1) [79, 83–85]. Most probably, the fact is associated with a highdensity of the bulk and surface lattice defects in CdS NPs.

The interdiffusion produces a weak signal at 500 cm−1 which is a combinationof the second-order vibrations LOCdS + LOCdSe [79] (Fig. 6.28b). The stronginterdiffusion is also typical for the core/shell CdSe/CdS NPs produced in anon-catalytic way [79, 83]. As the lattice constants of CdSe and CdS are slightlydifferent, the contact between the two semiconductors results in the diffusion ofsulfur atoms into the bulk of cadmium selenide and the formation of a mixedCdSxSe1−x layer. The process results in a larger surface disordering in CdS NPsmanifesting as an increased SOCdS intensity for the photocatalytically producedCdS/CdSe NPs.

The LO phonon peak of CdSe can be observed as a low-intensity shoulder at185 cm−1 for low CdSe contents but appears as a well-resolved peak at 200 cm−1

for the highest cadmium selenide content (13 mol%). The peak is shifted by around10 cm−1 to lower frequencies as compared to the bulk CdSe (210 cm−1 [67])

Fig. 6.28 Resonant Raman spectra of (a) CdSe NPs (curve 1) and the core/shell CdSe/ZnS NPs(curve 2) incorporated into the gelatin films (kexc = 457.9 nm) [79, 80]; b CdS NPs (curve 1) andthe photocatalytically formed CdS/CdSe nanoheterostructures (curves 2–5) in the gelatin films(excitation at 441.7 nm, CdSe contant is 4 mol% (curve 2), 7 mol% (curve 3), 10 mol% (curve 4),and 13 mol% (curve 6) [82]; c CdS/CdSe nanoheterostructure with 13 mol% CdSe (relative toCdS content) registered at a different excitation wavelength [82]

6.5 Studies of Nanocrystalline Semiconductors … 355

indicating a considerable phonon confinement, that is, a small size of the pho-todeposited CdSe NPs.

The spectral width and position of the LO bands do not reflect directly the sizeand size distribution of CdSe NPs because the spectra were registered under theresonant conditions (kexc = 441.7 nm) when a spectral contribution of the NPs withthe bandgap energy closest to the excitation energy is the highest in the ensemble.At the same time, an increase in the phonon peak intensity with a growing CdSecontent indicates that a fraction of the resonantly excited CdSe NPs becomes larger.As the excitation wavelength is increased the resonance conditions for the selectiveexcitation of CdSe NPs become more and more favorable. As a result, the ratio ofLOCdSe and LOCdS peaks distinctly grows as kexc is increased from 441.7 to514.5 nm (Fig. 6.28c) indicating that the laser energy (2.41 eV) is close to thebandgap of the photodeposited CdSe NPs. The fact can be taken as an indicationthat cadmium selenide is indeed photodeposited as separate NPs, not assub-nanometer 2D islands, for which the resonance energy is expected to be muchhigher. The driving force for the formation of 3D NPs can be supplied by the latticeconstant mismatch of CdS and CdSe resulting in a compressive stress that can berelaxed via the transformation of primary 2D CdSe islands into the 3D NPs simi-larly to the well-reported Stransky-Krastanov transformation of epitaxial AIIIBV

semiconductor nano-islands [86].The ratio of the main phonon mode and its overtone 2LO depends on the

electron-phonon interaction in the semiconductor lattice amounting to I2LO/ILO = 0.3–0.4 similarly to 2–5 nm CdSe NPs produced by a non-catalytic method[87], the fact additionally proving the formation of separate 3D CdSe NPs as aproduct of the photocatalytic deposition. This example of CdS/CdSenanoheterostructures demonstrates quite clearly the capabilities of the Ramanspectroscopy in the studies of composite semiconductor nanoheterostructures thatcan potentially be used in the photocatalytic and photoelectrochemicallight-harvesting systems.

6.6 Studies of Colloidal Semiconductor-Based SystemsUsing Dynamic Light Scattering

The dynamic light scattering (DLS) or the laser photon correlation spectroscopy canbe used as an alternative or a complementary method to the transmission electronmicroscopy (TEM) in the studies of colloidal systems with semiconductor NPs andother components dispersed in a liquid medium. The method is based on thedetection of fluctuations of the elastic (Reighley) light scattering by the colloidalparticles changing their position chaotically in the Brownian movement [88]. Themethod allows determining the diffusion coefficient of NPs (polymer globules, largemolecules, etc.) or a distribution of the diffusion coefficients—for the polydisperse

356 6 Probing with Light—Optical Methods in Studies …

colloidal systems. Finally, the size/size distribution of the colloidal species can becalculated using the well-known Einstein-Stokes equation.

One of the undoubtful advantages of DLS is the possibility to probe “live”colloids, while the preparation of samples for TEM, scanning electron microscopy(SEM) and X-ray diffraction (XRD) requires the destruction of a colloidal systemand the extraction of a dispersed nano-phase that can possibly be accompanied bythe phase transitions and particle size changes. Simultaneously with the determi-nation of the diffusion coefficient, the DLS can provide important information onthe structure of a double electric layer of the NP surface and the surface charge.Modern DLS setups allow determining the size of colloidal NPs down to 1 nm withan accuracy of ±0.1 nm [89].

On the other hand, the application area of the DLS method is confined tocolloidal systems stabilized by adsorbed ions or relatively small ligands. TheEinstein-Stokes equation gives the hydrodynamic size of NPs, that is, the size of acolloidal micelle composed of the NP “core”, a layer of stabilizer molecules and asolvation shell moving as a whole entity in the Brownian movement. As a result,DLS cannot typically be applied to the polymer-stabilized NPs because the poly-mers are present in the form of globules as large as several hundred nm thusmasking the target NPs.

In the case of colloidal systems with no bulky polymers present, the DLSmethod can provide quite precise determination of the size of colloidal semicon-ductor NPs. For example, this method can distinguish colloidal ZnO NPs differingonly slightly by the average size [90]. As the starting reagent concentration isincreased from 2 � 10−3 to 2 � 10−2 M the absorption band edge of resultingcolloidal zinc oxide NPs shifts from *345 to 355 nm (Fig. 6.29a) correspondingto a variation of the average NP size from 3.7 to 4.4 nm. The results of DLSpresented in Fig. 6.29 show the feasibility of the reliable determination of a minutedifference in the ZnO NP size that can be observed in the absorption spectra owingto a strong size dependence of the bandgap. The positions of size distributionmaxima determined by the DLS (Fig. 6.29b, c) and the average NP size derived

Fig. 6.29 a Normalized absorption spectra (a) and size distributions obtained from the DLSmeasurements (b, c) of colloidal ZnO NPs in ethanol synthesized at a starting reactantconcentration of 2 � 10−3 M (a, curve 1; b) and 2 � 10−2 M (a, curve 2; c) [90]

6.6 Studies of Colloidal Semiconductor-Based Systems … 357

from the optical absorption spectra (the values presented in Fig. 6.29b, c) almostcoincide.

The DLS size determination was successfully applied to track the growth ofultra-small core/shell CdSe/CdS NPs stabilized in aqueous solutions by mixed Cd(II) complexes with ammonia and mercaptoacetate ions [91]. The growth occursduring the thermal treatment of a starting solution at 90–95 °C and the NP size canbe varied by adjusting the treatment duration. As the heating proceeds the bandgapof CdSe QDs determined from the electron absorption band edge decreases grad-ually from 2.89 to 2.70 eV after 45 min heating (Table 6.4). These values arestrongly shifted to higher energies as compared with the band gap of bulk CdSe,1.75 eV, due to the strong spatial confinement of the photogenerated charge carriersin very small CdSe NPs.

The average size d of the CdSe NPs was estimated from Eg using thewell-known empirical calibration curve presented in Fig. 6.9 to be as small as1.9 nm (Table 6.4) for the colloid produced by the 2-min heating and growing to2.2 nm after the 45-min heating. A TEM study of the CdSe colloid produced by the2-min thermal treatment (Fig. 6.30a) showed the CdSe NPs to be 1.8–2.0 nm insize revealing a high degree of the size uniformity. The interparticle aggregationwas minimized due to an electrostatic barrier of charged Cd(II)-NH3-mercaptoa-cetate complexes adsorbed on the NP surface. Despite such a small size, the NPsshowed a good crystallinity with a well-resolved interplanar distance of3.5 ± 0.1 Å typical for the cubic CdSe (Fig. 6.30b, c). A scatter of CdSe NP sizearound the average value did not exceed 0.5 nm (Fig. 6.30d).

The DLS spectroscopy confirmed the presence of individual ultra-small CdSeNPs in the colloidal solution (Fig. 6.30e, Table 6.4). The species in the startingsolution are characterized by an average hydrodynamic size of 1.8 nm whichincreases upon the thermal treatment to 2.4 nm for the CdSe NPs produced by the2-min heating and grows up to 3.5 nm for the CdSe NPs formed after the thermaltreatment for 45 min (Fig. 6.30e, curve 6; Table 6.4). Therefore, for the colloidalCdSe solution heated for 2 min the NP size estimated from the spectral curvecorresponds to the TEM data and agrees with the DLS results, the differencebetween d and dDLS indicating the existence of a half-nm-thick surface stabilizerlayer on the NP surface.

A larger discrepancy between d and dDLS for the colloidal solution produced atthe 45-min heating was ascribed to the formation of a shell on the surface of CdSe

Table 6.4 Band gap Eg, sized, and hydrodynamic sizedDLS of CdSe NPs subjectedto thermal treatment duringtime t [91]

t, min Eg, eV d, nm dDLS, nm

0 – – 1.8

2 2.89 1.9 2.1

4 2.83 2.0 2.5

5 2.80 2.1 2.7

15 2.73 2.2 2.8

30 2.72 2.2 3.0

45 2.70 2.2 3.5

358 6 Probing with Light—Optical Methods in Studies …

NPs, that is clearly visible for the DLS spectroscopy but does not contribute sub-stantially to the position of the band edge in the electron absorption spectra. It wasassumed [91] that the discrepancy between the size of CdSe NPs derived from theoptical absorption (and TEM) data and the hydrodynamic NP size that increasesduring the thermal treatment originate from the formation of a protecting CdS layeron the surface of CdSe NPs as a result of the partial hydrolysis of mercaptoacetateanions in strongly alkaline solutions at 90–95 °C. This assumption found con-vincing proofs in the results of the Raman and X-ray photoelectron spectroscopy ofsuch core/shell CdSe/CdS NPs [91]. Thus, the DLS spectroscopy combined withoptical absorption spectroscopy and TEM can be a powerful tool for probing thestructure of semiconductor NPs and nanocomposites even at the low size scale.

The application area of the DLS spectroscopy is not limited to “rigid” inorganicNPs having an invariable size and shape. This method can also be applied to probe“soft” systems, in particular those containing ultra-thin layers of various photoac-tive materials, such as molybdenum or tungsten dichalcogenides. Of special interestare the DLS studies of colloidal graphene oxide (GO) and reduced graphene oxide(RGO) used very often as co-catalysts of various light processes and as componentsof the photoelectrochemical solar cells [92–101].

The GO sheets feature a random alternation of the aromatic graphene areas of thesp2-hybridized carbon and the oxidized regions, where the sp3-hybridized carbonatoms are bound to various oxygen-containing functional groups—the epoxy andhydroxyl groups, carboxyl groups, etc. [92, 93, 102, 103]. As a result, the GOsheets are flexible and can attain a scrolled or crumpled shape [104–106]. Thetransformation of GO particles into a nonplanar conformation is also favored by the

Fig. 6.30 TEM (a), high-resolution TEM (b, c) images and size distribution (d) of CdSe/CdS NPs(2 min thermal treatment). e Hydrodynamic size distribution in starting CdSe nuclei solution(curve 1) and after the thermal treatment for 1 min (curve 2), 3 min (curve 3), 5 min (curve 4),17 min (curve 5), and 45 min (curve 6) [91]

6.6 Studies of Colloidal Semiconductor-Based Systems … 359

formation of hydrogen bonds between the functional groups from isolated frag-ments of the same GO sheet, either directly or via the bridge water molecules [92,93, 103, 105, 107]. In the case of RGO sheets, the intraparticle bonds can alsooriginate from the pp stacking interactions between the isolated aromatic areas ofRGO sheets [105, 107, 108]. Therefore, the properties of GO(RGO)-based colloidalsystems depend strongly on the shape and aggregation of the RGO sheets that canchange with the variations of solvent properties, in particular, pH, ionic strength,and temperature [103, 105–107, 109, 110].

The rare reports on the effects of pH and ionic strength on the shape of colloidalGO (RGO) discuss mostly indirect TEM observations of the sheet aggregatesproduced by the solvent evaporation/extraction [105–107] or the mathematicalmodeling of the shape changes [105]. At the same time, direct observations of theshape evolution of GO (RGO) particles in colloidal solutions can be made by thenondestructive DLS method [106, 110]. Such direct studies of the shape evolutionand aggregation are of special interest for RGO, which is the most frequently used2D material in the light-harvesting systems and solar cells. However, the studies aretypically obstructed by instability of the colloidal particles caused by the presenceof ionic residuals from decomposition of a reducing agent used to convert GO intoRGO [103, 107, 109]. To avoid the introduction of chemical reductants, the pho-tochemical reduction of colloidal GO can be applied, as it does not require anyadditional reagents except for water. Also, the photoreduction does not change pHand ionic strength of the solution and allows to vary smoothly the photoreduction“depth” by adjusting the light intensity and/or the illumination duration (exposure)[111–113].

A DLS study of aqueous GO colloids showed that the average hydrodynamicsize of colloidal particles varies from *150 to * 550 nm with a distributionmaximum at dDLS = 320 nm (Fig. 6.31a, curve 1) [114]. The absence of planarGO/RGO particles larger than a half-micron typically observed in the atomic forcemicroscopic images indicates that the GO (RGO) particles are crumpled as a resultof the pp stacking and the H-bond formations between various functional groups.The GO/RGO particles deposited from colloidal solutions onto hydrophobic sub-strates, such as carbon films and conductive FTO glass, preserve this partially orstrongly crumpled shape which can be visualized by the TEM/SEM measurements(Fig. 6.31b, c).

The photoreduction of GO, even at a starting stage (first 30 min illumination),results in considerable changes of the sheet properties manifesting as a drastic growthof dDLS up to 520 nm (Fig. 6.31a, curve 2). According to the absorption, Raman, andinfrared absorption spectroscopy this time range is also characterized by the mostvivid changes in the structure and bandgap energy of RGO [115, 116], in particular, ina considerable increase of the fraction of aromatic carbon in the RGO sheets. Thesechanges become only deeper at further illumination, however, the hydrodynamic sizeof colloidal RGO starts to change in a reverse direction decreasing to 360 and 300 nmfor 60 and 90-min light exposure, respectively (Fig. 6.31a, curves 3, 4). When thephotoreduction is finished at the 180-min exposure, the RGO particles are

360 6 Probing with Light—Optical Methods in Studies …

characterized by an average hydrodynamic size of 260 nm (Fig. 6.31a, curve 5),which is smaller than dDLS of the starting GO particles.

The hydrodynamic size of planar GO (RGO) particles is an effective valuedepending primarily on the way the sheets are crumpled. The DLS studies ofcolloidal dispersions of the single-layer graphene, MoS2, and WS2 [117] showedthat a dependence between the the lateral size L of these planar particles and theirhydrodynamic size dDLS can be expressed as dDLS = aLb, where a = 5.9 ± 2.2,b = 0.66 ± 0.06. Therefore, the observed photoinduced size evolution of colloidalGO/RGO particles indicates the changes of the sheet crumpling character as a resultof the intra-sheet interactions.

The mechanism of such shape evolution can be illustrated by a scheme presentedin Fig. 6.31d. The crumpling of starting GO is caused by the formation ofintra-sheet hydrogen bonds between the functional groups in different fragments ofthe GO sheets including the “bridge” water molecules (the case I in Fig. 6.31d)[118, 119]. The feasibility of spontaneous folding of GO sheets and the formationof 0.42-nm thick folds between the sheet fragments was confirmed by the molecularmodeling [120].

The GO photoreduction results the primary stage in abrupt changes in the sheetcomposition, in particular, in a partial elimination of epoxide and hydroxy groupsfrom the basal GO plane and the restoration of its aromatic character. At that, mostof the H-bonds interconnecting the folds disappear and the RGO sheet becomesmore unfolded, the fact mirrored by an increase of the hydrodynamic size (the caseII in Fig. 6.31d). As the RGO becomes photoreduced deeper and deeper, the aro-matic character of the basal plane becomes more expressed and new folds start toform between the sheet fragments via the pp stacking of the sp2-hybridized RGOfragments. The formation of new folds is also favored by increasingly hydrophobiccharacter of the photoreduced GO, as the RGO sheets tend to minimize their contact

Fig. 6.31 a Hydrodynamic size distribution of colloidal GO at pH 6 before the photoreduction(curve 1) and after the illumination with the UV light for 30 min (curve 2), 60 min (curve 3),90 min (curve 4), and 180 min (curve 5). b, c SEM (b) and TEM (c) images of a crumpled GOparticle. d Scheme of the photoinduced changes in the shape and structure of colloidal GOparticles [114]

6.6 Studies of Colloidal Semiconductor-Based Systems … 361

with the polar medium. As calculations showed [119], a distance between thefold-forming fragments of the RGO sheets is around 0.39 nm, which is close to theinterlayer distance in the bulk graphite (0.34–0.38 nm). A pronounced tendency ofRGO to the folding and crumpling was broadly reported [121–123]. A decrease inthe interactions between RGO and water and a more compact crumpling of theRGO sheets as a result of the pp stacking results in the fact that the deeplyphotoreduced RGO has a lower dDLS as compared to the starting GO.

The above-discussed model finds support in the results of pH-dependent DLSspectroscopy of the photoreduced GO. As pH of GO/RGO colloids is lowered to 2,the sheet aggregation is observed in all cases. At that, the RGO aggregate sizedepends on the reduction depth and varies from around 400 nm for the original GOto 2.5–3 lm for the most deeply reduced RGO. A pH increase to 11, on thecontrary, results in a decrease in dDLS—down to 100–120 nm for the RGO with thehighest reduction depth (Fig. 6.32a, curves 1, 2). The latter observation shows thetendency of RGO sheets to crumple and to minimize the surface contacting with thepolar medium with an increased ionic strength. The dialysis purification of alkalineRGO colloids removes the alkali and the hydrodynamic size of RGO sheets returnsto a starting value of 280 nm (Fig. 6.32a, curves 2, 3). Then, as NaOH is added forthe second time, the hydrodynamic size of RGO particles decreases again(Fig. 6.32a, curves 3, 4). These observations indicate that the pH-induced shapechanges of colloidal RGO have a dynamic character, and the conformation ofcolloidal RGO sheets can “adapt” to the polarity and ionic strength of the dispersivemedium.

The reversible character of the conformational changes of RGO sheets is alsoevidenced by changes of the hydrodynamic size induced by the interactions of the

Fig. 6.32 Hydrodynamic size distribution of the photoreduced colloidal RGO sheetspre-illuminated for 180 min at pH 6 (curve 1) and treated in two consequences: a after elevatingpH to 10 with NaOH (curve 2), reducing pH to 7 by the dialysis (curve 3) and, again, elevating pHto 10 with NaOH (curve 4); b after elevating pH to 10 with NaOH (curve 2), and adding methyleneblue (curve 3) or sodium salt of pyrene sulfonic acid (curve 4) [114, 116]

362 6 Probing with Light—Optical Methods in Studies …

RGO sheets with aromatic molecules. The latter are capable of the strongadsorption on the RGO basal plane via pp stacking. The absorption can provoke theunfolding of the crumpled RGO particles. For example, introduction of methyleneblue dye or sodium salt of the pyrene sulfonic acid into the colloidal RGO (with thehighest photoreduction depth) solution results in a dDLS increase from 120 to 290–300 nm (Fig. 6.32b).

The methylene blue and pyrene derivatives have a well-known affinity to thearomatic fragments of RGO sheets [93]. An additional electrostatic repulsionbetween the neighboring adsorbed molecules that bear a positive (methylene blue)or negative charge (pyrene sulfonic acid anion) evidently overwhelms theintra-sheet pp stacking interactions between the fragments of crumpled RGOresulting in the sheet unfolding and an increase of the hydrodynamic size.

It should be noted that the results discussed here, along with the data of [110],where the DLS spectroscopy was successfully applied to study reversible interac-tions between the colloidal GO sheets and DNA molecules at variations of thesolution temperature, demonstrate a high potential and a unique character of thismethod in studying subtle effects, such as the conformational changes of singlelayer 2D sheets under the external stimuli directly in colloidal solutions.

Concluding the discussion of experimental methods that exploit the interactionof nanocrystalline semiconductors with light to probe the structure and properties ofsuch nanomaterials we note that this discussion has an introductory character anddoes not pretend on a comprehensive characterization of the whole variety ofoptical and spectroscopic methods applied nowadays for the investigations ofnanocrystalline materials. We aimed to provide a general notion on the possibilitiesof using the light to study semiconductor NPs and nanoheterostructures, high-lighting only some the most useful or unique capabilities of such methods. Also, theoptical methods constitute only a small fraction of the versatility of modern arsenalof techniques used to get an insight into the intimate details of the structure andproperties of nanocrystalline objects.

Recently some disbalance can be observed in the appreciation and application of“classical” optical characterization methods as compared to other modern charac-terization methods, most of them very demanding and sophisticated in the instru-mental sense. The researchers strive to characterize their nanomaterials with thelargest possible array of structural methods, such as the XRD, TEM, SEM, atomicforce microscopy, X-ray photoelectron and UV photoelectron spectroscopy, nuclearmagnetic resonance, etc., trying to collect as much information as possible on thestructure of such light-harvesting materials. From the other hand, the utilization ofsuch a versatile array of techniques is, at least partially, caused by more and morerigid standards and demands to the instrumental level of studies of semiconductornanomaterials put forth by the authoritative scientific journals. However, very oftenthe presented results are analyzed only superficially. At the same time, the opticalmethods that gained deserved esteem from the early stages of the studies ofnanocrystalline semiconductors and nanoheterostructures are applied in a less andless habitual way and retreat undeservedly to the background of modern method-ology. We hope that this chapter will help at improving this disbalance by

6.6 Studies of Colloidal Semiconductor-Based Systems … 363

delivering a proper impression on the capabilities of the optical methods in col-lecting the most versatile information on the electron, photophysical and structuralcharacteristics of the nanocrystalline semiconductor light-harvesting materials.

References

1. Stroyuk OL, Sobran IV, Korzhak AV et al (2008) Photopolymerization of water-solubleacrylic monomers induced by colloidal CdS and CdxZn1−xS nanoparticles. Colloid PolymSci 286:489–498. doi:10.1007/s00396-007-1824-4

2. Raevskaya AE, Stroyuk OL, Kryukov AI et al (2006) Structural and optical characteristics ofCdxZn1−xS nanoparticles stabilized in aqueous solutions of polymers. Theor Exp Chem42:181–185. doi:10.1007/s11237-006-0035-2

3. Stroyuk OL, Dzhagan VM, Kuchmii SY et al (2007) Nanosecond and microsecond decay ofphotogenerated charges in CdxZn1−xS nanoparticles. Theor Exp Chem 43:297–305. doi:10.1007/s11237-007-0037-8

4. Grätzel M (ed) (1983) Energy resources through photochemistry and catalysis. AcademicPress, Inc., New York

5. Korgel BA, Monbouquette HG (2000) Controlled synthesis of mixed core and layered (Zn,Cd)S and (Hg, Cd)S nanocrystals within phosphatidylcholine vesicles. Langmuir 16:3588–3594. doi:10.1021/la990139r

6. Petrov DV, Santos BS, Pereira GA, de Mello Donegá C (2002) Size and band-gapdependences of the first hyperpolarizability of CdxZn1−xS nanocrystals. J Phys Chem B106:5325–5334. doi:10.1021/jp010617i

7. Kasap S, Capper P (eds) (2006) Springer handbook of electronic and photonic materials.Springer Science + Business Media, Inc, Berlin

8. de Mello Donega C, Koole R (2009) Size dependence of the spontaneous emission rate andabsorption cross section of CdSe and CdTe quantum dots. J Phys Chem C 113:6511–6520.doi:10.1021/jp811329r

9. Monticone S, Tufeu R, Kanaev AV (1998) Complex nature of the UV and visiblefluorescence of colloidal ZnO nanoparticles. J Phys Chem B 102:2854–2862. doi:10.1021/jp973425p

10. Jasieniak J, Smith L, Embden J, Mulvaney P (2009) Re-examination of the size-dependentabsorption properties of CdSe quantum dots. J Phys Chem C 113:19468–19474. doi:10.1021/jp906827m

11. Schoenhalz AL, Arantes JT, Fazzio A, Dalpian GM (2010) Surface and quantumconfinement effects in ZnO nanocrystals. J Phys Chem C 114:18293–18297. doi:10.1021/jp103768v

12. Gaponenko SV (1998) Optical properties of semiconductor nanocrystals. University Press,Cambridge

13. Rogach A (ed) (2008) Semiconductor nanocrystal quantum dots: synthesis, assembly,spectroscopy and applications. Springer, Vienna

14. Wang Y, Herron N (1991) Nanometer-sized semiconductor clusters: materials synthesis,quantum size effects, and photophysical properties. J Phys Chem 95:525–532. doi:10.1021/j100155a009

15. Ye C, Fang X, Wang M, Zhang L (2006) Temperature-dependent photoluminescence fromelemental sulfur species on ZnS nanobelts. J Appl Phys 99:063504. doi:10.1063/1.2181311

16. Eychmüller A (2000) Structure and photophysics of semiconductor nanocrystals. J PhysChem B 104:6514–6528. doi:10.1021/jp9943676

17. Burda C, Chen X, Narayanan R, El-Sayed MA (2005) Chemistry and properties ofnanocrystals of different shapes. Chem Rev 105:1025–1102. doi:10.1021/cr030063a

364 6 Probing with Light—Optical Methods in Studies …

18. Nanda K, Kruis F, Fissan H (2001) Energy levels in embedded semiconductor nanoparticlesand nanowires. Nano Lett 1:605–611. doi:10.1021/nl0100318

19. Qu L, Peng X (2002) Control of photoluminescence properties of CdSe nanocrystals ingrowth. J Am Chem Soc 124:2049–2055. doi:10.1021/ja017002j

20. Chen W, Wang Z, Lin Z, Lin L (1997) Absorption and luminescence of the surface states inZnS nanoparticles. J Appl Phys 82:3111–3115. doi:10.1063/1.366152

21. Tang H, Xu G, Weng L et al (2004) Luminescence and photophysical properties of colloidalZnS nanoparticles. Acta Mater 52:1489–1494. doi:10.1016/j.actamat.2003.11.030

22. Cizeron J, Pileni MP (1995) Solid Solution of CdyZn1−yS nanosize particles made in reversemicelles. J Phys Chem 99:17410–17416. doi:10.1021/j100048a016

23. Cizeron J, Pileni MP (1997) Solid solution of CdyZn1−yS nanosized particles: photophysicalproperties. J Phys Chem B 101:8887–8891. doi:10.1021/jp9713571

24. Melamed NT (1957) Sulfur vacancy emission in ZnS phosphors. Phys Rev 107:1727–1728.doi:10.1103/PhysRev.107.1727.2

25. Rayevska AE, Stroyuk OL, Kozytskiy AV, Kuchmiy SY (2010) Electron energy factors inphotocatalytic methylviologen reduction in the presence of semiconductor nanocrystals.J Photochem Photobiol A 210:209–214. doi:10.1016/j.jphotochem.2009.11.019

26. Feilchenfeld H, Chumanov H, Cotton TM (1996) Photoreduction of methylviologenadsorbed on silver. J Phys Chem 100:4937–4943. doi:10.1021/jp952329q

27. Ferrer B, Llabrés i Xamena FX, García H (2007) Hollow organosilica spheres as hosts:Photoinduced electron transfer between Ru(bpy)3

2− and methylviologen. Inorg Chim Acta360:1017–1022. doi:0.1016/j.ica.2006.07.107

28. Zhong H, Bai Z, Zou B (2012) Tuning the luminescence properties of colloidal I-III–VIsemiconductor nanocrystals for optoelectronics and biotechnology applications. J PhysChem Lett 3:3167–3175. doi:10.1021/jz301345x

29. Liu S, Su X (2014) The synthesis and application of I-III–VI type quantum dots. RSC Adv4:43415–43428. doi:10.1039/C4RA05677A

30. Leach ADP, Macdonald JE (2016) Optoelectronic properties of CuInS2 nanocrystals andtheir origin. J Phys Chem Lett 7:572–583. doi:10.1021/acs.jpclett.5b02211

31. Kolny-Olesyak J, Weller H (2013) Synthesis and application of colloidal CuInS2 semicon–ductor nanocrystals. ACS Appl Mater Interfaces 5:12221–12237. doi:10.1021/am404084d

32. Tang Y, Wang P, Yun JH et al (2015) Frequency-regulated pulsed electrodeposition ofCuInS2 on ZnO nanorod arrays as visible light photoanodes. J Mater Chem A 3:15876–15881. doi:10.1039/C5TA03255E

33. Wang S, Yang X, Zhu Y et al (2014) Solar-assisted dual chamber microbial fuel cell with aCuInS2 photocathode. RSC Adv 4:23790–23796. doi:10.1039/C4RA02488E

34. Pan Z, Mora-Sero I, Shen Q et al (2014) High-efficiency “Green” quantum dot solar cells.J Am Chem Soc 136:9203–9210. doi:10.1021/ja504310w

35. Raevskaya AE, Rosovik OP, Kozytskiy AV et al (2016) Non-Stoichiometric Cu-In-S@ZnSnanoparticles produced in aqueous solutions as light harvesters for liquid-junctionphotoelectrochemical solar cells. RSC Adv 6:100145–100157. doi:10.1039/C6RA18313A

36. Zhang B, Wang Y, Yang C et al (2015) The composition effect on the optical properties ofaqueous synthesized Cu–In–S and Zn–Cu–In–S quantum dot nanocrystals. Phys ChemChem Phys 17:25133–25141. doi:10.1039/C5CP03312H

37. Stroyuk OL, Raevskaya AE, Kuchmii SY (2004) Oxidation of polysulfide ions induced byCdS nanoparticles under pulsed photolysis conditions. Theor Exp Chem 40:130–135.doi:10.1023/B:THEC.0000028910.49933.67

38. Raevskaya AE, Stroyuk OL, Kuchmiy SY (2004) Photocatalytic oxidation ofhydrosulfide-ions by molecular oxygen over cadmium sulfide nanoparticles. J NanopartRes 6:149–158. doi:10.1023/B:NANO.0000034719.30620.d3

39. Vorobets VS, Kovach SK, Kolbasov GY (2001) Ionic equilibria in sulfide/polysulfidesolutions with an account for the effect of ionic coupling. Ukr Chem J (in Russian) 67:12–16

40. Giggenbach W (1972) Optical spectra and equilibrium distribution of polysulfide ions inaqueous solution at 20°. Inorg Chem 11:1201–1207. doi:10.1021/ic50112a009

References 365

41. Licht S, Hodes G, Manassen J (1986) Numerical analysis of aqueous polysulfide solutionsand its application to cadmium chalcogenide/polysulfide photoelectrochemical solar cells.Inorg Chem 25:2486–2489. doi:10.1021/ic00235a003

42. Meyer B (1976) Elemental sulfur. Chem Rev 76:367–388. doi:10.1021/cr60301a00343. Chivers T, Drummond I (1972) Characterization of the trisulfur radical anion S3

− in bluesolutions of alkali polysulfides in hexamethylphosphoramide. Inorg Chem 11:2525–2527.doi:10.1021/ic50116a047

44. Gruen DM, McBeth RL, Zielen AJ (1971) Nature of sulfur species in fused salt solutions.J Am Chem Soc 93:6691–6693. doi:10.1021/ja00753a070

45. Martin RP, Doub WH, Roberts JL, Sawyer DT (1973) Electrochemical reduction of sulfur inaprotic solvents. Inorg Chem 12:1921–1925. doi:10.1021/ic50126a047

46. Annenkova VS, Antonik LM, Haliullin AK et al (1983) Formation of sodium sulfide anionradicals in dipolar aprotic solvents. J General Chem (in Russian) 53:2409–2410

47. Mills G, Schmidt KH, Matheson MS, Meisel D (1987) Thermal and photochemical reactionsof sulfhydryl radicals. Implications for colloid photocorrosion. J Phys Chem 91:1590–1596.doi:10.1021/j100290a060

48. Melnikov MY (1994) Photochemistry of organic radicals. Moscow University Publishing,Moscow (in Russian)

49. Vossmeyer T, Katsikas L, Giersig M et al (1994) CdS nanoclusters: synthesis, character-ization, size dependent oscillator strength, temperature shift of the excitonic transitionenergy, and reversible absorbance shift. J Phys Chem 98:7665–7673. doi:10.1021/j100082a044

50. Kovach SK, Vorobets VS, Vasko AT (1992) Potentiometric titration using asulfide-selective electrode. Ukr Chem J 58:491–494

51. Dzhagan VM, Stroyuk OL, Rayevska AE et al (2010) Spectroscopic and photochemicalstudy of Ag+-, Cu2+-, Hg2+-, and Bi3+-doped CdxZn1−xS nanoparticles. J Colloid InterfaceSci 345:515–523. doi:10.1016/j.jcis.2010.02.001

52. Stroyuk OL, Raevskaya AE, Korzhak AV et al (2009) Photocatalytic production ofhydrogen in systems based on CdxZn1−xS/Ni

0 nanostructures. Theor Exp Chem 45:12–22.doi:10.1007/s11237-009-9057-x

53. Henglein A, Lilie J (1981) Storage of electrons in aqueous solution: the rates of chemicalcharging and discharging the colloidal silver microelectrode. J Am Chem Soc 103:1059–1066. doi:10.1021/ja00395a011

54. Savinov EN, Nagornyi VE, Parmon VN (1994) Influence of excessive charge of colloidalcadmium sulfide particles on interfacial electron transfer rate. Chem Phys (in Russian)13:56–65

55. Matsumoto H, Uchida H, Matsunaga T et al (1994) Photoinduced reduction of viologens onsize-separated CdS nanocrystals. J Phys Chem 98:11549–11556. doi:10.1021/j100095a041

56. Zhukowskiy MA, Stroyuk OL, Shvalagin VV et al (2009) Photocatalytic growth of CdS,PbS, and CuxS nanoparticles on the nanocrystalline TiO2 films. J Photochem Photobiol, A203:137–144. doi:10.1016/j.jphotochem.2009.01.007

57. Stroyuk OL, Kuchmii SY, Zhukovskii MA et al (2009) Effect of the method of production ofTiO2/CdS film nanoheterostructures on the effectiveness of photoinduced charge separation.Theor Exp Chem 45:302–307. doi:10.1007/s11237-009-9097-2

58. Serpone N, Lawless D, Khairutdinov R, Pelizzetti E (1995) Subnanosecond relaxationdynamics in TiO2 colloidal sols (particle sizes Rp = 1.0–13.4 nm). Relevance to heteroge-neous photocatalysis. J Phys Chem 99:16655–16661. doi:10.1021/j100045a027

59. Bahnemann D, Henglein A, Lilie J, Spanhel L (1984) Flash photolysis observation of theabsorption spectra of trapped positive holes and electrons in colloidal TiO2. J Phys Chem88:709–711. doi:10.1021/j150648a018

60. Teoh WY, Mädler L, Beydoun D et al (2005) Direct (one-step) synthesis of TiO2 andPt/TiO2 nanoparticles for photocatalytic mineralisation of sucrose. Chem Eng Sci 60:5852–5861. doi:10.1016/j.ces.2005.05.037

366 6 Probing with Light—Optical Methods in Studies …

61. Duonghong D, Ramsden J, Grätzel M (1982) Dynamics of interfacial electron transferprocesses in colloidal semiconductor systems. J Am Chem Soc 104:2977–2985. doi:10.1021/ja00375a006

62. Bessekhouad Y, Chaoui N, Trzpit M et al (2006) UV-vis versus visible degradation of AcidRed II in a coupled CdS/TiO2 semiconductors suspension. J Photochem Photobiol A183:218–224. doi:10.1016/j.jphotochem.2006.03.025

63. Liu Y, Wang X, Yang F, Yang X (2008) Excellent antimicrobial properties of mesoporousanatase TiO2 and Ag/TiO2 composite films. Micropor Mesopor Mater 114:431–439. doi:10.1016/j.micromeso.2008.01.032

64. Rajh T, Micic OI, Lawless D, Serpone N (1992) Semiconductor photophysics. 7.Photoluminescence and picosecond charge carrier dynamics in cadmium sulfide quantumdots confined in a silicate glass. J Phys Chem 96:4633–4641. doi:10.1021/j100190a090

65. Kamat PV, Dimitrijević NM, Fessenden RW (1987) Photoelectrochemistry in particulatesystems. 6. Electron-transfer reactions of small CdS colloids in acetonitrile. J Phys Chem91:396–401. doi:10.1021/j100286a029

66. Dzhagan VM, Valakh MY, Raevskaya AE et al (2008) Size effects on Raman spectra ofsmall CdSe nanoparticles in polymer films. Nanotechnology 19:305707

67. Landolt-Börnstein (1982) Numerical data and functional relationships in science andtechnology. Group III, Vol. 17b—Semiconductors, Sect. 3.10.1. Springer, Berlin

68. Chen X, Mao SS (2007) Titanium dioxide nanomaterials: synthesis, properties, modifica-tions, and applications. Chem Rev 107:2891–2959. doi:10.1021/cr0500535

69. Trallero-Giner C, Debernardi A, Cardona M et al (1997) Optical vibrons in CdSe dots anddispersion relation of the bulk material. Phys Rev B 57:4664. doi:10.1103/PhysRevB.57.4664

70. Zhang JY, Wang XY, Xiao M et al (2002) Lattice contraction in free-standing CdSenanocrystals. Appl Phys Lett 81:2076–2078. doi:10.1063/1.1507613

71. Hwang YN, Shin S, Park HL et al (1996) Effect of lattice contraction on the Raman shifts ofCdSe quantum dots in glass matrices. Phys Rev B 54:15120. doi:10.1103/PhysRevB.54.15120

72. Rolo AG, Vasilevskiy MI (2007) Raman spectroscopy of optical phonons confined in semi–conductor quantum dots and nanocrystals. J Raman Spectr 38:618–633. doi:10.1002/jrs.1746

73. Raevskaya AE, Grodzyuk GY, Korzhak AV et al (2011) Preparation and optical propertiesof polyethyleneimine-stabilized colloidal CdSe and CdSxSe1−x quantum dots. Theor ExpChem 46:416–421. doi:10.1007/s11237-011-9173-2

74. Kozitskiy AV, Stroyuk OL, Kuchmiy SY et al (2013) Morphology, optical, andphotoelectro–chemical properties of electrodeposited nanocrystalline ZnO films sensitizedwith CdxZn1−xS nanoparticles. J Mater Sci 48:7764–7773. doi:10.1007/s10853-013-7598-9

75. Yükselici H, Persans PD, Hayes TM (1995) Optical studies of the growth of Cd1−xZnxSnanocrystals in borosilicate glass. Phys Rev B 52:11763. doi:10.1103/PhysRevB.52.11763

76. Azhniuk YM, Gomonnai AV, Lopushansky VV et al (2007) Morphology, optical, andphotoelectrochemical properties of electrodeposited nanocrystalline ZnO films sensitizedwith CdxZn1−xS nanoparticles. J Phys 92:012044. doi:10.1007/s10853-013-7598-9

77. Arora AK, Rajalakshmi M (2000) Resonance Raman scattering from Cd1−xZnxS nanopar-ticles dispersed in oxide glass. J Appl Phys 88:5653. doi:10.1063/1.1321025

78. Sahoo S, Dhara S, Sivasubramanian V, Kalavathi S, Arora AK (2009) Phonon confinementand substitutional disorder in Cd1−xZnxS nanocrystals. J Raman Spectrosc 40:1050. doi:10.1002/jrs.2232

79. Dzhagan VM, Valakh MY, Raevskaya AE et al (2007) Resonant Raman scattering study ofCdSe nanocrystals passivated with CdS and ZnS. Nanotechnology 18:285701

80. Dzhagan VM, Valakh MY, Raevska AE et al (2009) The influence of shell parameters onphonons in core–shell nanoparticles: a resonant Raman study. Nanotechnology 20:365704

81. Raevskaya AE, Stroyuk OL, Kuchmiy SY et al (2007) Optical study of CdS- andZnS-passivated CdSe nanocrystals in gelatin films. J Phys: Condens Matter 19:386237

References 367

82. Dzhagan VM, Raevskaya AE, Stroyuk OL et al (2009) Resonant Raman spectroscopy ofconfined and surface phonons in CdSe-capped CdS nanoparticles. Phys Stat Sol C 6:2043–2046. doi:10.1002/pssc.200881755

83. Dzhagan VM, Valakh MY, Raevskaya AE et al (2007) Temperature-dependent resonantRaman scattering study of core/shell nanocrystals. J Phys Conf Series 92:012045

84. Dzhagan VM, Valakh MY, Raevskaya AE et al (2008) Characterization of semiconductorcore-shell nanoparticles by resonant Raman scattering and photoluminescence spectroscopy.Appl Surf Sci 255:725–727. doi:10.1016/j.apsusc.2008.07.018

85. Raevskaya AE, Stroyuk AL, Kuchmiy SY (2006) Preparation of colloidal CdSe andCdS/CdSe nanoparticles from sodium selenosulfate in aqueous polymers solutions. J ColloidInterface Sci 302:133–141. doi:10.1016/j.jcis.2006.06.018

86. Krasil’nik ZK, Lytvyn P, Lobanov DN et al (2002) Microscopic and optical investigation ofGe nanoislands on silicon substrates. Nanotechnology 13:81–85

87. Raevskaya AE, Stroyuk OL, Kuchmiy SY et al (2006) Growth and spectroscopiccharacterization of CdSe nanoparticles synthesized from CdCl2 and Na2SeSO3 in aqueousgelatine solutions. Colloids Surfaces A 290:304–309. doi:10.1016/j.colsurfa.2006.05.038

88. Berne BJ, Pecora R (1976) Dynamic light scattering. Wiley, New York89. www.malvern.com/ZetasizerNanoZS90. Stroyuk OL, Dzhagan VM, Shvalagin VV et al (2010) Size-dependent optical properties of

colloidal ZnO nanoparticles charged by photoexcitation. J Phys Chem C 114:220–225.doi:10.1021/jp908879h

91. Raevskaya AE, Stroyuk OL, Panasiuk YV et al (2016) A new route to very stablewater-soluble ultra-small core/shell CdSe/CdS quantum dots. Nano-Objects, Nano-Struct.doi:10.1016/j.nanoso.2015.12.001

92. Zhu Y, Murali S, Cai W et al (2010) Graphene and graphene oxide: synthesis, properties,and applications. Adv Mater 22:3906–3924. doi:10.1002/adma.201001068

93. Dreyer DR, Park S, Bielawski CW, Ruoff RS (2010) The chemistry of graphene oxide.Chem Soc Rev 39:228–240. doi:10.1039/B917103G

94. Bai S, Shen X (2012) Graphene-inorganic composites. RSC Adv 2:64–98. doi:10.1039/C1RA00260K

95. An X, Yu JC (2011) Graphene-based photocatalytic composites. RSC Adv 1:1426–1434.doi:10.1039/C1RA00382H

96. Shen J, Zhu Y, Yang X, Li C (2012) Graphene quantum dots: emergent nanolights forbioimaging, sensors, catalysis and photovoltaic devices. Chem Commun 48:3686–3699.doi:10.1039/C2CC00110A

97. Singh V, Joung D, Zhai L et al (2011) Graphene based materials: past, present and future.Progr Mater Sci 56:1178–1271. doi:10.1016/j.pmatsci.2011.03.003

98. Zhang Y, Tang ZR, Fu X, Xu YJ (2011) Engineering the unique 2D mat of graphene toachieve graphene-TiO2 nanocomposite for photocatalytic transformation: what advantagedoes graphene have over its forebear carbon nanotube? ACS Nano 9:7426–7435. doi:10.1021/nn202519j

99. Xiang Q, Yu J, Jaroniec M (2012) Graphene-based semiconductor photocatalysts. Chem SocRev 41:782–796. doi:10.1039/C1CS15172J

100. Yao J, Sun Y, Yang M, Duan Y (2012) Chemistry, physics and biology of graphene-basednanomaterials: new horizons for sensing, imaging and medicine. J Mater Chem 22:14313–14329. doi:10.1039/C2JM31632C

101. Huang X, Zeng Z, Fan Z, Liu J, Zhang H (2012) Graphene-based electrodes. Adv Mater24:5979–6004. doi:10.1002/adma.201201587

102. Erickson K, Erni R, Lee Z et al (2010) Determination of the local chemical structure ofgraphene oxide and reduced graphene oxide. Adv Mater 22:4467–4472. doi:10.1002/adma.201000732

103. Konkena B, Vasudevan S (2012) Understanding aqueous dispersibility of graphene oxideand reduced graphene oxide through pKa measurements. J Phys Chem Lett 3:867–872.doi:10.1021/jz300236w

368 6 Probing with Light—Optical Methods in Studies …

104. Wen X, Garland CW, Hwa T et al (1992) Crumpled and collapsed conformations of graphiteoxide membranes. Nature 355:426–428. doi:10.1038/355426a0

105. Whitby RLD, Gun’ko VM, Korobeynik A et al (2012) Driving forces of conformationalchanges in single-layer graphene oxide. ACS Nano 6:3967–3973. doi:10.1021/nn3002278

106. Zangmeister CD, Ma X, Zachariah MR (2012) Restructuring of graphene oxide sheets intomonodisperse nanospheres. Chem Mater 24:2554–2557. doi:10.1021/cm301112j

107. Whitby RLD, Korobeynik A, Gun’ko VM et al (2011) pH-driven physicochemicalconformational changes of single-layer graphene oxide. Chem Commun 47:9645–9647.doi:10.1039/C1CC13725E

108. Kim K, Lee Z, Malone BD et al (2011) Multiply folded graphene. Phys Rev B 83:245433.doi:10.1103/PhysRevB.83.245433

109. Park S, An J, Piner RD et al (2008) Aqueous suspension and characterization of chemicallymodified graphene sheets. Chem Mater 20:6592–6594. doi:10.1021/cm801932u

110. Tang L, Wang Y, Liu Y, Li J (2011) DNA-directed self-assembly of graphene oxide withapplications to ultrasensitive oligonucleotide assay. ACS Nano 5:3817–3822. doi:10.1021/nn200147n

111. Stroyuk OL, Andryushina NS, Shcherban ND et al (2012) Photochemical reduction ofgraphene oxide in colloidal solution. Theoret Exp Chem 48:1–11. doi:10.1007/s11237-012-9235-0

112. Shulga YM, Martynenko VM, Muradyan VE et al (2010) Gaseous products of thermo- andphoto-reduction of graphite oxide. Chem Phys Lett 498:287–291. doi:10.1016/j.cplett.2010.08.056

113. Matsumoto Y, Koinuma M, Kim SY et al (2010) Simple photoreduction of graphene oxidenanosheet under mild conditions. ACS Appl Mater Interfaces 2:3461–3466. doi:10.1021/am100900q

114. Andryushina NS, Stroyuk OL, Yanchuk IB, Yefanov AV (2014) A dynamic light scatteringstudy of photochemically reduced colloidal graphene oxide. Colloids Polym Sci 292:539–546. doi:10.1007/s00396-013-3134-3

115. Prezioso S, Perrozzi F, Donarelli M et al (2012) Large area extreme-UV lithography ofgraphene oxide via spatially resolved photoreduction. Langmuir 28:5489–5495. doi:10.1021/la204637a

116. Stroyuk OL, Andryushina NS, Kuchmy SY et al (2015) Photochemical processes involvinggraphene oxide. Theor Exp Chem 51:1–29. doi:10.1007/s11237-015-9393-y

117. Lotya M, Rakovich A, Donegan JF, Coleman JN (2013) Measuring the lateral size ofliquid-exfoliated nanosheets with dynamic light scattering. Nanotechnology 24:265703

118. Dreyer DR, Park S, Bielawski CW, Ruoff RD (2010) The chemistry of graphene oxide.Chem Soc Rev 39:228–240. doi:10.1039/B917103G

119. Shul’ga YM, Vasilets VN, Baskakov SA et al (2012) Photoreduction of graphite oxidenanosheets with vacuum ultraviolet radiation. High En Chem 46:117–121. doi:10.1134/S0018143912020099

120. Feng L, Wu L, Qu X (2013) New horizons for diagnostics and therapeutic applications ofgraphene and graphene oxide. Adv Mater 25:168–186. doi:10.1002/adma.201203229

121. Zhang B, Li L, Wang Z et al (2012) Radiation induced reduction: an effective and cleanroute to synthesize functionalized graphene. J Mater Chem 22:7775–7781. doi:10.1039/C2JM16722K

122. Bittolo BS, Piccinini M, Mariani A et al (2011) Wettability and switching of electricalconductivity in UV irradiated graphene oxide films. Diamond Rel Mater 20:871–874.doi:10.1016/j.diamond.2011.04.013

123. Zhang HH, Liu Q, Feng K et al (2012) Facile photoreduction of graphene oxide by an NAD(P)H model: Hantzsch 1,4-Dihydropyridine. Langmuir 28:8224–8229. doi:10.1021/la301429g

124. Rayevska AE, Grodzyuk GY, Dzhagan VM et al (2010) Synthesis and characterization ofwhite-emitting CdS quantum dots stabilized with polyethyleneimine. J Phys Chem C114:22478–22486. doi:10.1021/jp108561u

References 369

125. Raevskaya AE, Grodzyuk GY, Stroyuk OL et al (2010) Preparation and spectral propertiesof high-efficiency luminescent polyethyleneimine-stabilized CdS quantum dots. Theor ExpChem 46:233–238. doi:10.1007/s11237-010-9145-y

126. Stroyuk OL, Raevskaya AE, Korzhak AV, Kuchmiy SY (2007) Zink sulfide nanoparticles:spectral properties and photocatalytic activity in metals reduction reactions. J Nanopart Res9:1027–1039. doi:10.1007/s11051-006-9183-5

127. Matsumoto H, Sakata T, Mori H, Yoneyama H (1996) Preparation of monodisperse CdSnanocrystals by size selective photocorrosion. J Phys Chem 100:13781–13785. doi:10.1021/jp960834x

128. Li J, Xia JB (2000) Hole levels and exciton states in CdS nanocrystals. Phys Rev B62:12613–12616. doi:10.1103/PhysRevB.62.12613

129. Lippens P, Lannoo M (1989) Calculation of the band gap for small CdS and ZnS crystallites.Phys Rev B 39:10935–10942. doi:10.1103/PhysRevB.39.10935

130. Wang Y, Herron N (1990) Quantum size effects on the exciton energy of CdS clusters. PhysRev B 42:7253–7255. doi:10.1103/PhysRevB.42.7253

131. Yang CC, Jiang Q (2006) Size effect on the bandgap of II–VI semiconductor nanocrystals.Mater Sci Eng B 131:191–194. doi:10.1016/j.mseb.2006.04.016

132. Sapra S, Sarma D (2004) Evolution of the electronic structure with size in II-VIsemiconductor nanocrystals. Phys Rev B 69:125304. doi:10.1103/PhysRevB.69.125304

133. Gorer S, Ganske J, Hemminger J, Penner R (1998) Size-selective and epitaxialelectrochemical/chemical synthesis of sulfur-passivated cadmium sulfide nanocrystals ongraphite. J Am Chem Soc 120:9584–9593. doi:10.1021/ja981676l

134. Baskoutas S, Terzis A (2006) Size-dependent band gap of colloidal quantum dots. J ApplPhys 99:013708. doi:10.1063/1.2158502

135. Murakoshi K, Hosokawa H, Saitoh M et al (1998) Preparation of size-controlled hexagonalCdS nanocrystallites and the characteristics of their surface structures. J Chem Soc, FaradayTrans 94:579–586. doi:10.1039/A707192B

136. Nandakumar P, Vijayan C, Murti Y (2002) Optical absorption and photoluminescencestudies on CdS quantum dots in Nafion. J Appl Phys 91:1509–1514. doi:10.1063/1.1425077

137. Leistikow MD, Johansen J, Kettelarij AJ et al (2009) Size-dependent oscillator strength andquantum efficiency of CdSe quantum dots controlled via the local density of states. PhysRev B 79:045301. doi:10.1103/PhysRevB.79.045301

138. Wood A, Giersig M, Hilgendorff M et al (2003) Size effects in ZnO: the cluster to quantumdot transition. Austr J Chem 56:1051–1057. doi:10.1071/CH03120

139. Wong EM, Hörtz PG, Liang CJ et al (2001) Influence of organic capping ligands on thegrowth kinetics of ZnO nanoparticles. Langmuir 17:8362–8367. doi:10.1021/la010944h

140. Brus LE (1984) Electron-electron and electron-hole interactions in small semiconductorcrystallites: the size dependence of the lowest excited electronic state. J Chem Phys80:4403–4444. doi:10.1063/1.447218

141. Fonoberov VA, Balandin AA (2004) Radiative lifetime of excitons in ZnO nanocrystals:The dead-layer effect. Phys Rev B 70:195410. doi:10.1103/PhysRevB.70.195410

142. Wang YS, Thomas PJ, O’Brien P (2006) Nanocrystalline ZnO with ultraviolet lumines-cence. J Phys Chem B 110:4099–4104. doi:10.1021/jp0566313

143. KwakH, TiagoML,Chelikowsky JR (2008)Quantum confinement and strong coulomb corre–lation in ZnO nanocrystals. Sol State Commun 145:227–230. doi:10.1016/j.ssc.2007.11.004

144. van Dijken A, Janssen AH, Smitsmans MHP et al (1998) Size-selective photoetching ofnano–crystalline semiconductor particles. Chem Mater 10:3513–3522. doi:10.1021/cm980715p

145. Kumbhojkar N, Nikesh VV, Kshirsagar A, Mahamuni S (2000) Photophysical properties ofZnS nanoclusters. J Appl Phys 88:6260–6264. doi:10.1063/1.1321027

146. Souici AH, Keghouche N, Delaire JA et al (2006) Radiolytic synthesis and optical propertiesof ultra-small stabilized ZnS nanoparticles. Chem Phys Lett 422:25–29. doi:10.1016/j.cplett.2006.02.013

370 6 Probing with Light—Optical Methods in Studies …

147. Ghosh PK, Jana S, Nandy S, Chattopadhyay KK (2007) Size-dependent optical anddielectric properties of nanocrystalline ZnS thin films synthesized via rf-magnetronsputtering technique. Mater Res Bull 42:505–514. doi:10.1016/j.materresbull.2006.06.019

148. Nanda J, Sapra S, Sarma DD, Chandrasekharan N, Hodes G (2000) Size-selected zinc sulfidenanocrystallites: synthesis, structure, and optical studies. Chem Mater 12:1018–1024.doi:10.1021/cm990583f

149. Kane RS, Cohen RE, Silbey R (1996) Theoretical study of the electronic structure of PbSnanoclusters. J Phys Chem 100:7928–7932. doi:10.1021/jp952869n

150. Kang I, Wise W (1997) Electronic structure and optical properties of PbS and PbSe quantumdots. J Opt Soc Am B 14:1632–1646. doi:10.1364/JOSAB.14.001632

151. Wang Y, Suna A, Mahler W, Kasowski R (1987) PbS in polymers. From molecules to bulksolids. J Chem Phys 87:7315–7322. doi:10.1063/1.453325

152. Miyoshi H, Yamachika M, Yoneyama H, Mori H (1990) Photochemical properties of PbSMicrocrystallites prepared in Nafion. J Chem Soc, Faraday Trans 86:815–818. doi:10.1039/FT9908600815

153. Sapra S, Shanthi N, SarmaDD (2002) Realistic tight-bindingmodel for the electronic structureof II-VI semiconductors. Phys Rev B 66:205202. doi:10.1103/PhysRevB.66.205202

154. Lippens PE, Lannoo M (1991) Electronic structure of II–VI semiconductor nanocrystals.Mater Sci Eng B 9:485–487

155. Wang LW, Zunger A (1996) Pseudopotential calculations of nanoscale CdSe quantum dots.Phys Rev B 53:9579–9582. doi:10.1103/PhysRevB.53.9579

156. Yu WW, Qu L, Guo W, Peng X (2003) Experimental determination of the extinctioncoefficient of CdTe, CdSe, and CdS nanocrystals. Chem Mater 15:2854–2860. doi:10.1021/cm034081k

157. Rogach AL, Kornowski A, Gao M et al (1999) Synthesis and characterization of a size seriesof extremely small thiol-stabilized CdSe nanocrystals. J Phys Chem B 103:3065–3069.doi:10.1021/jp984833b

158. Shiang JJ, Kadavanich AV, Grubbs RK, Alivisatos AP (1995) Symmetry of annealedwurtzite CdSe nanocrystals: assignment to the C3v point group. J Phys Chem 99:17417–17422. doi:10.1021/j100048a017

159. Murray CB, Norris DJ, Bawendi MG (1993) Synthesis and characterization of nearlymonodisperse CdE (E = sulfur, selenium, tellurium) semiconductor nanocrystallites. J AmChem Soc 115:8706–8715. doi:10.1021/ja00072a025

160. Raevskaya AE, Stroyuk OL, Kuchmii SY (2003) Optical characteristics of colloidalnanoparticles of CdS stabilized with sodium polyphosphate and their behavior during pulsephotoexcitation. Theor Exp Chem 39:158–165. doi:10.1023/A:1024933023783

161. Kozytskiy AV, Stroyuk OL, Kuchmiy SY et al (2014) Photoelectrochemical and Ramancharacterization of nanocrystalline CdS Grown on ZnO by successive ionic layer adsorptionand reaction. Thin Solid Films 562:56–62. doi:10.1016/j.tsf.2014.03.070

162. Kozytskiy AV, Stroyuk OL, Skoryk MA et al (2015) Photochemical formation andphotoelectrochemical properties of TiO2/Sb2S3 heterostructures. J Photochem Photobiol A303–304:8–16. doi:10.1016/j.jphotochem.2015.02.005

163. Kozytskiy AV, Stroyuk OL, Kuchmiy SY (2014) Inorganic photoelectrochemical solar cellsbased on nanocrystalline ZnO/ZnSe and ZnO/CuSe heterostructures. Catal Today 230:227–233. doi:10.1016/j.cattod.2013.09.043

164. Swamy V, Kuznetsov A, Dubrovinsky LS et al (2005) Finite-size and pressure effects on theRaman spectrum of nanocrystalline anatase TiO2. Phys Rev B 71:184302. doi:10.1103/PhysRevB.71.184302

165. Azhniuk YM, Hutych YI, Lopushansky VV et al (2007) Interplay of factors affecting Ramanscattering in cadmium chalcogenide nanocrystals in dielectric media. J Phys: Conf Ser79:012017

166. Bersani D, Loticci PP (1992) Confinement effects on the LO-phonons in CdSexS1−x dopedglasses. Phys stat sol (b) 174:575–582. doi:10.1002/pssb.2221740227

References 371

Index

AAbsorption band edge/threshold, 4, 244, 322Absorption spectrum, 167, 321, 357Aerosol AT, 246Alloyed nanoparticles, 210Alumina membrane, 273Aluminium oxide, 272AM1.5 light flux, 58, 167Anatase, 67, 131, 141, 254, 273, 278, 350Anistropic shape, 71, 147, 191Anodization, 61, 142, 269Antimony sulfide, 199, 350Artificial photosynthesis, 87, 128Atomic force microscopy, 198Atomic layer deposition, 263Average size, 326

BBand alignment, 209Band bending, 15Band design, 58Barrier layer, 207Bequerel, xxivBifunctional molecules, 171, 243Binary heterostructures/nanocomposites, 48,

143, 275Biomass, 71, 92Bio-mimicking, 152Bipyridyl, 91, 135, 137, 140Bismuth oxide, 56, 148, 258Bismuth oxyhalogenide, 147, 258Bismuth sulfide, 53, 65, 145“Black” titania, 57Blocking layer, 203, 206Boron carbide, 144Brookite, 131, 254Bulk heterojunction solar cells, 162Burstein-Moss effect, 23, 340

CCadmium selenide, 18, 21, 51, 72, 145, 178,

190, 328, 335, 350Cadmium sulfide, 6, 43, 130, 145, 184, 196,

275, 347Cadmium sulfo-selenide, 210, 352Cadmium telluride, 26, 329Cadmium zinc sulfide, 18, 65, 146, 188, 324,

327, 330, 340, 352Calibration curve, 326, 351Carbon-doped titania, 61Carbon materials, 54, 217, 226, 279Carbon microspheres, 269Carbon nanoparticles, 54, 78, 209Carbon nanotubes, 54, 79, 82, 280Carbon vacancy, 83Cascade charge transfer, 52, 190, 203Cascade conduction band levels, 208Cascade design, 208, 211Catalytic activity, 218, 226Cathodic polarization, 26, 28Ceramics, 272Cerium oxide, 144Chalcopyrite, 52, 175Charge carrier migration, 15Charge collection efficiency, 206Charge compensation, 56Charge leakage, 203, 206Charge migration, 265Charge separation, 15, 68, 196, 199Charge transfer, 16Charge transfer complex, 43, 139Charge transfer rate constant, 178Charge transfer resistance, 175, 218, 221, 225Charge trapping, 13, 141, 330, 336Chemical bath deposition, 192, 222, 347Chemical vapor deposition, 170Chromium oxide, 60

© Springer International Publishing AG 2018O. Stroyuk, Solar Light Harvesting with Nanocrystalline Semiconductors,Lecture Notes in Chemistry 99, https://doi.org/10.1007/978-3-319-68879-4

373

Ciamician, xvii, xviiiClays, 271Clusters, 6, 10, 131, 262CO2conversion, 127, 273Cobalt phosphate, 72, 77Cobalt sulfide, 218Co-catalyst, 41, 81, 86, 132, 135, 151, 224, 281Colloidal semiconductors, 242, 319, 356Colloidal titania, 248Colloidal zinc oxide, 250Conduction band potential, 2, 188Conjugated polymer, 131, 149, 205Copper antimony sulfide, 191Copper indium selenide, 205Copper indium sulfide, 52, 76, 213, 337Copper oxide, 53, 75, 143, 192Copper phosphide, 53Copper sulfide, 86, 196, 218, 223Copper telluride, 191Copper tin sulfide, 215Core/shell, 144, 195, 251, 337, 354, 358Coulomb interaction, 6, 10, 322coumarine, 42Counter electrode, 217Cysteine, 145, 176

Dde-Broglie wavelength, 2Defect-related photoluminescence, 16, 330Defects, 8Density functional theory, 150Design of photocatalysts, 129Diffusion coefficient, 244Dip coating, 260Dipole moment, 63Direct aqueous synthesis, 179Direct electron transition, 7, 323Discharging rate, 344Dodecanthiol, 171Doping, 55, 58, 140, 149, 273Double charged layer, 14, 26, 357Drop-casting, 180Dye sensitized solar cell, 137, 161, 164, 217Dynamic light scattering, 356

EEffective mass, 9, 326Effective mass approximation, 9, 326Einstein-Stokes equation, 357Electric capacitance, 65, 206, 344Electrochemical deposition, 170Electrochemical etching, 79, 216, 264, 269Electrochemical impedance spectroscopy, 218

Electrochemical reduction, 191, 242, 277Electrodeposition, 191, 215, 264Electron absorption spectroscopy, 321Electron acceptor, 3Electron affinity, 46Electron beam sputtering, 279Electron donor, 3Electron-hole recombination, 14, 203, 330, 337Electron mobility, 215, 222Electron paramagnetic resonance, 15, 51, 131Electron transition, 321, 323Electron trap, 18, 332Electrophoretic deposition, 182, 257Electrophysical parameters, 321Energy diagram, 214, 225, 336Energy-dispersive X-ray spectroscopy, 188,

200, 354Environmental photocatalysis, 79Eosin, 42, 58, 82Erythrosin, 43, 82Evonik P25, 67, 261, 274Excessive negative charge, 65, 340, 349Exciton absorption, 8, 322Exciton binding energy, 11Exciton Bohr radius, 1, 8Excitonic photoluminescence, 16Exfoliation, 44, 54, 64, 73, 82, 86, 135, 261,

267Exposed facets, 50, 131, 133, 142, 147Extinction spectrum, 201

FFermi level, 45, 167Ferrites, 73Fill factor, 167Finite-depth potential well, 327First-principles calculations, 64Flash photolysis, 65, 198Flexible electrode, 226Free Gibbs energy, 3, 128Fujishima and Honda, xx, xxi, xxii, xxviFullerene, 49, 54, 76, 281Fundamental absorption, 7, 322

GGallium nitride, 77Gallium oxide, 60, 134Gold nanoparticles, 195Gradient composition, 67, 69, 210Graphene, 59, 78, 144, 281Graphene oxide, 177, 281, 359Graphite, 273, 281Graphitic carbon, 280

374 Index

Graphitic carbon nitride, 43, 54, 79, 92, 135,142, 147

HHeating up synthesis, 171Heterojunction, 144, 346Hodes G., 186Hollow sphere, 55, 81, 132, 143, 217, 265, 268Hot electron, 45, 76Hot-injection synthesis, 171Hydrazine, 67, 71, 147Hydrodynamic size, 357, 360Hydrolysis, 247, 265, 274Hydrothermal treatment, 215, 249, 257

IImpedance spectrum, 206Incident-photon-to-current efficiency, 47, 168,

214Indirect electron transition, 7, 323Indium oxide, 60, 144Indium phosphide, 140Indium selenide, 72Indium sulfide, 21Industrial wastes, 71Ink, 180Inorganic complex ligands, 177In situ deposition, 183Intercalation, 50, 82, 90, 270, 277Interdiffusion, 354Interfacial charge transfer, 3, 275, 337, 347Intermediate, 131, 198, 337, 342, 346Internal electric field, 68Inverted (inverse) opal, 50, 217Inverted micelles, 246Ion exchange, 170, 201, 224, 272Iron oxide, 144, 150, 269, 279Iron silicide, 78Isotopic studies, 134, 148, 150

KKamat P., xxixKesterite, 5, 52, 77, 215, 225

LLambert-Beer equation, 6Laser photocorrelation spectroscopy, 356Laser pulse deposition, 264Lattice defects, 13, 82, 147, 345, 355Layered material, 50, 64, 151, 258, 270Layered metal chalcogenide, 64Lead selenide, 20, 176, 191

Lead sulfide, 21, 187, 199, 222, 224, 328Life-time, 330, 334, 341Ligand exchange, 174Light absorption, 6Light conversion efficiency, 162, 166, 218Light harvesting cycle, 162Light harvesting system, xvii, xix, xx, 92, 161,

167, 241, 269, 356, 360Light scattering layer, 194, 216Light-shielding effect, 201Light-to-current conversion efficiency, 53Linear absorption coefficient, 7Liquid-junction solar cells, 162Liquid phase deposition, 261Loosely aggregated nanoparticles, 261, 264Luminescence/photoluminescence

spectroscopy, 329, 335

MMagic-size clusters, 179Magnetron sputtering, 57, 90, 263Mechanochemical treatment, 61, 257, 274Mercaptopropionic acid, 51, 171Merocyanine, 42Mesoporous cadmium sulfide, 67, 267Mesoporous framework, 266, 273Mesoporous materials, 264Mesoporous metal chalcogenide, 267Mesoporous silica, 132, 141Mesoporous titania, 44, 59, 141, 165, 266Metal complex dye, 43Metallate, 73, 258, 270Metal mesh, 224Metal-organic framework, 83, 135Metal sulfide photocatalyst, 61, 245Methylviologen, 16, 21, 22, 334Micellar solution, 267Microemulsion, 246Microsphere, 131, 134, 144, 258, 266Microwave treatment, 254, 259Mid-bandgap states, 147Molar absorption coefficient, 7Molecular orbital, 140Molecular photocatalyst, 136Molybdenium disulfide, 20, 62, 86, 151, 194,

359Monolith reactor, 144Mott-Schottky, 133, 135Multi-electron process, 28, 130, 133, 135Multi-exciton generation, 23, 165Multi-layer structures, 182

Index 375

NNafion membrane, 137, 150, 272Nanocrystalline films, 260Nanocrystalline powders, 252Nanorod, 194, 215, 222Nanoscroll, 44, 259Nanosheet, 46, 73, 133, 149, 259, 269Nanotube, 42, 51, 269, 283Nanotube array, 60, 142, 216, 224, 269Nanowire, 47, 52, 175, 201, 215, 257N-doped titania, 59, 142, 266, 274Nickel oxide, 76, 138, 267Nickel sulfide, 86Niobate, 44, 50, 73, 133, 267Nitridation, 59, 274Nitrogen fixation, 127, 146Nitrogen vacancy, 135, 147Noble metal, 86Non-stationary bleaching, 24, 340, 349Nyquist plot, 206

OOleylamine, 171Open-circuit photovoltage, 167, 188Optical fiber, 261Optical phonon, 350Oscillator strength, 11Ostwald ripening, 247, 269Oxidative photocorrosion, 143Oxide/chalcogenide heterostructures, 187, 196,

201, 275, 346Oxygen evolution, 57Oxygen vacancy, 58, 131, 134, 147, 148Oxysulfide, 61, 70

PPaper, 273, 281Paris Climate Conference, 129Passivating ligand, 171Perovskite, 5, 60, 73, 91, 133, 166, 270Phase composition, 253, 350Phase size effect, 5Phonon, 7Phonon confinement, 351, 356Photoaction spectrum, 45, 47, 135, 140, 168Photocatalyst, 39Photocatalytic microreactor, 248Photocatalytic system, xx, xxi, xxii, xxiii, 39,

41–43, 49, 54–56, 58, 59, 61, 62, 62, 67,71, 81, 85, 87, 88, 90–92, 127, 129, 130,152, 162

Photocathode, 74, 138, 224Photochemical/photocatalytic deposition, 49,

52, 59, 81, 170, 194, 222, 283, 348

Photochromic properties, 284Photocorrosion, 28, 67, 70Photocurrent density, 166Photocurrent generation efficiency, 203Photocurrent spectrum, 186Photoelectric effect, xxiv, xxviPhotoelectrocatalytic system, 86Photoelectron spectroscopy, 177Photoetching, 66Photoinduced charge accumulation, 165Photoinduced charging, 13, 23Photoinduced electron transfer, 22, 150Photoinduced polarization, 65Photoluminescence quenching, 181Photoluminescence spectroscopy, 213Photolysis, 70Photonic crystal, 46, 144, 217Photopolymerization, 21Photovoltage, 188Phthalocyanine, 42, 76, 82, 139, 145Plasmonic photocatalyst, 44, 81Platinum group metals, 52p/n Heterojunction, 53Polycondensation, 247Polyelectrolyte, 180Polymer films, 330, 350, 355polystyrene latex/microparticles, 268Pore size distribution, 266Porphyrin, 84, 135, 140Post-synthesis treatment, 246, 274Power conversion efficiency, 167Prebiotic photosynthesis, 127Primary nuclei, 181, 244Protective shell, 169, 174, 179, 208, 359Pulse photoexcitation, 25Pulse photolysis, 26Pyrolysis, 262, 283

QQuantum confinement, 176Quantum-sized nanoparticles, 20, 244, 322Quantum size effects, 2, 5, 133, 203, 325Quaternary metal chalcogenide, 52, 70, 225

RRadiative recombination, 18“Rainbow-cell” design, 211Raman spectroscopy, 186, 350, 353, 359Reactor geometry, 130Redox potential, 19Red phosphorus, 77Reduced graphene oxide, 64, 76, 84, 87, 132,

142, 149, 218, 281, 359Resonant excitation, 350, 356

376 Index

Reverse electron transfer, 207, 213Rhodamine, 85Ruthenium bipyridyl, 42, 44Rutile, 67, 254Rydberg energy, 9

SSacrificial donor, 40, 92Scaffold, 162, 215Schottky barrier, 45Selection rules, 7Self-igniting mixtures, 256, 273Semiconductor-metal nanostructures, 281Semiconductor-sensitized solar cells, 161Sensitization, 84, 273Sensitizer, 42, 137, 139, 163, 283Short-circuit photocurrent density, 167Silica nanoparticles, 267, 271Silicon carbide, 77Silicon nanoparticles, 76, 135, 151Silicon solar cell, 138, 161Siloxene, 77Silver antimony sulfide, 191Silver bismuth sulfide, 191Silver indium sulfide, 52, 213Silver sulfide, 187, 199Single-layer sheets, 64, 361Singlet excited state, 85Single-wall carbon nanotube, 91Size dependence, 72, 213, 331, 341Size distribution, 244, 253, 326, 349, 356Size-selected fractionation, 247Size-selected nanoparticles, 62, 173, 211, 326,

357Size variation, 244Solar cell market, 161Solar light simulator, 167Solar spectrum, 58Sol-gel method, 247, 274Solid solution, 64, 68, 91, 187, 210, 324, 352Solvatochromic sensor, 284Spatial confinement, 9Spatial design, 152Spatial organization, 49, 89, 196Spatial separation, 49, 52, 144, 198, 209, 275Spectral methods, 319Spin coating, 261Stokes shift, 331Structural defects, 204Structural disorder, 186, 351Structure-directing agent, 251Sub-bandgap states, 186

Successive ionic layer adsorption and reaction,169, 183, 218, 275, 352

Sulfidation, 224, 226Sulfur-doped titania, 61Sulfur vacancy, 63, 130, 147Surface defects, 5, 67, 333, 336, 355Surface optical phonon, 351Surface plasmon resonance, 44, 136, 284Surface states, 14, 341Surface-to-volume ratio, 204

TTafel equation, 22, 189Tandem, 138, 140Tantalates, 50Tantalum nitride, 59Tantalum oxide, 59Tantalum oxynitride, 59Template, 63, 81, 144Ternary metal chalcogenide, 52, 76, 186, 193,

225Thioglycolic acid, 176Third-generation solar cells, 161Time-resolved laser photolysis, 198, 339Time-resolved photoluminescence, 177, 334Tin sulfide, 63Titanate, 42, 50, 258, 269Titania nanotubes, 145Titania-silica composites, 255Titanium oxyfluoride, 90Titanosilicate, 272Total water splitting, 57, 270Trapped electron/hole, 198, 332Tungstate, 258Tungsten disulfide, 194Tungsten oxide, 50, 133, 258Type II heterojunction, 186

UUltra-small nanoparticles, 10, 179, 249, 358Ultrasound treatment, 267, 282Urbach Equation, 323

VValence band potential, 2Volcano-shaped dependence, 187, 213, 344Voltage-current curve, 167

WWater oxidation, 59Water splitting, 89Work function, 46Wurtzite, 68

Index 377

XX-ray photoelectron spectroscopy, 359X-Ray scattering, 326

ZZeolite, 50, 62, 81, 132, 143, 271Zinc blende, 68

Zinc indium sulfide, 63Zinc oxide, 60, 142, 165, 175, 201, 249, 327,

352, 357Zinc selenide, 201, 206, 350Zinc sulfide, 18, 130, 208, 323, 328Zirconate, 258Z-scheme, 89, 129, 145

378 Index