Chapter 5

CHAPTER 5

Particulate Oxynitrides forPhotocatalytic Water SplittingUnder Visible Light

KAZUHIKO MAEDA AND KAZUNURI DOMEN*

The University of Tokyo, Department of Chemical System Engineering,7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan*Email: [email protected]

5.1 Introduction

Catalytic splitting of pure water into hydrogen and oxygen in the presence ofsemiconductor powders with visible light is a promising approach from theviewpoint of converting solar energy into chemical energy. The reaction (eq.(1)) is typical of uphill-type reaction with a large positive change in the Gibbsfree energy (DG0 238 kJmol1).

2H2O ! 2H2O2 1

whose half reactions are described as follows:

2H 2e ! H2 2

2H2O 4h ! O2 4H 3

The research was initially triggered by the potential of TiO2-based photo-electrochemical reactions for the decomposition of water into H2 and O2.

1

Since then, a range of semiconductor materials have been examined aspowder photocatalysts in view of the emphasis placed on large-scale

RSC Energy and Environment Series No. 9

Photoelectrochemical Water Splitting: Materials, Processes and Architectures

Edited by Hans-Joachim Lewerenz and Laurence Peter

r The Royal Society of Chemistry 2013

Published by the Royal Society of Chemistry, www.rsc.org

109

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09

hydrogen production.26 For efficient solar energy conversion, a photocatalyticmaterial that can work under visible light (l4400 nm) the main componentof sunlight needs to be devised.

Figure 5.1 shows a schematic illustration of the basic principle of overallwater splitting on a semiconductor particle. Under irradiation with an energyequivalent to or greater than the band gap of the semiconductor photocatalyst,electrons in the valence band are excited into the conduction band, leavingholes in the valence band. These photogenerated electrons and holes causereduction and oxidation reactions, respectively. To achieve overall watersplitting, the potential equivalent to the bottom of the conduction band must bemore negative than the reduction potential of H1 to H2 (0V vs. NHE at pH 0),while the top of the corresponding potential for valence band must be morepositive than the oxidation potential of H2O to O2 (1.23V vs.NHE). Therefore,the minimum photon energy thermodynamically required to drive the reactionis 1.23 eV, which corresponds to a wavelength of ca. 1000 nm, in the near in-frared region. Accordingly, it would appear possible to utilize the entire spec-tral range of visible light (400olo800 nm). However, there are activationbarriers in the charge transfer processes between the photocatalyst and watermolecules, necessitating a photon energy greater than the band gap of thephotocatalyst to drive the overall water-splitting reaction at a reasonable rate.In addition, the back reaction; that is, water formation from H2 and O2, mustbe strictly inhibited, and the photocatalysts themselves must be stable in thereaction. Although there is a large number of materials that possess suitableband-gaps, materials that function as a photocatalyst for overall water splittingare limited due to other factors, as will be mentioned later.

Since 1980, many metal oxides have been examined as photocatalysts forwater splitting. They include transition metal oxides containing metal ions ofTi41, Zr41, Nb51, Ta51, or W61 with d0 electronic configuration and typicalmetal oxides having metal ions of Ga31, In31, Ge41, Sn41, or Sb51 with d10

Figure 5.1 Basic principle of overall water splitting on a semiconductor particle.Reprinted with permission from Reference 3. Copyright 2007 AmericanChemical Society.

110 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

electronic configuration.4,5 More than 100 metal oxides have been reported toexhibit water splitting activity, and some have achieved high quantum effi-ciencies of several tens of percent. However, metal oxide photocatalysts inprinciple work only under UV light (lo400 nm), as explained later. Althoughsome non-oxide materials such as CdS and CdSe have been examined, par-ticularly for visible-light catalysis, successful photocatalytic systems have yet tobe achieved, primarily due to the lack of oxygen productivity and inherentinstability of the materials.7,8

The two major obstacles to the development of powder photocatalysts arethe discovery of new stable photocatalytic materials and the construction of asuitable visible light-driven photocatalytic system. The application of metaloxides for photocatalysis in the visible-light region is complicated by the deepvalence band positions (O2p orbitals) of these materials, resulting in a band gapthat is too large to harvest visible light.9 In general, the conduction band of ametal oxide photocatalyst is formed by empty d orbitals of a transition metal ors,p orbitals of a typical metal, which lie above the water reduction potential ofH2O (0V vs. NHE at pH 0). On the other hand, the potential of the valenceband, which consists of O2p orbitals (ca. 3 V), is considerably more positivethan the water oxidation potential (1.23V). Therefore, the band gaps of metaloxide photocatalysts inevitably become wider than 3 eV, if they meet thethermodynamic requirement for water splitting.

Since the N2p orbital has a higher potential energy than the O2p orbital, itwould be interesting to use a metal nitride or metal oxynitride as a photo-catalyst. Figure 5.2 shows the schematic band structures of the metal oxideNaTaO3 and (oxy)nitride BaTaO2N, both of which have the same perovskitestructure.3 The top of the valence band (i.e. the highest occupied molecularorbital, HOMO) of the metal oxide consists of O2p orbitals. When N atoms are

Figure 5.2 Schematic band structures of a metal oxide (NaTaO3) and metal (oxy)ni-tride (BaTaO2N).Reprinted with permission from Reference 3. Copyright 2007 AmericanChemical Society.

Particulate Oxynitrides for Photocatalytic Water Splitting Under Visible Light 111

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

partially or fully substituted for O atoms in a metal oxide, the HOMO of thematerial is expected to shift above that of the corresponding metal oxidewithout affecting the level of the bottom of the conduction band (i.e. the lowestunoccupied molecular orbital, LUMO). DFT band structure calculations forBaTaO2N indicated that the HOMO consists of hybridized N2p and O2p or-bitals, whereas the LUMO is mainly composed of empty Ta5d orbitals. TheHOMO for the (oxy)nitride is at a higher potential energy than that of thecorresponding oxide due to the contribution of N2p orbitals, making the bandgap small enough to respond to visible light (o3 eV). Similar results were ob-tained in calculations for other (oxy)nitrides. In this chapter, (oxy)nitrides arepresented as non-oxide type materials for photocatalytic water splitting undervisible light, focusing on recent research progress made by the authors group.

5.2 Oxynitrides Having Early-Transition Metal Cations

Some particulate (oxy)nitrides containing early transition-metal cations arestable and harmless materials, and can be readily obtained by nitriding a cor-responding metal-oxide powder.1026 Note that this type of material differsfrom materials doped with nitrogen. Figure 5.3 shows the UV-visible diffusereflectance spectra (DRS) for certain (oxy)nitrides containing transition-metalcations of Ti41, Nb51 and Ta51 with d0 electronic configuration. It is clear thatthese (oxy)nitrides possess absorption bands at 500750 nm, corresponding toband-gap energies of 1.72.5 eV that are estimated from the onset wavelengthsof the absorption spectra. As expected from DFT calculations, these (oxy)ni-trides have band-edge potentials suitable for overall water splitting, as revealed

Figure 5.3 UV-visible diffuse reflectance spectra for (oxy)nitrides containing Ti41,Nb51, and Ta51.Reprinted with permission from Reference 3. Copyright 2007 AmericanChemical Society.

112 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

by UV photoelectron spectroscopy (UPS) and photoelectrochemical (PEC)analysis.16 For example, it has been revealed that the tops of the valence bandsare shifted to higher potential energies in the order Ta2O5oTaONoTa3N5,whereas the potentials of the bottoms of the conduction bands remain almostunchanged, as shown in Figure 5.4.

As overall water splitting is generally difficult to achieve due to the up-hillnature of the reaction, photocatalytic activities of a number of (oxy)nitrides forwater reduction or oxidation were examined in the presence of methanol orsilver nitrate as a sacrificial reagent. The reactions using sacrificial reagentsare not overall water splitting reactions, but are often carried out as testreactions for overall water splitting.3,4 Table 1 lists the photocatalytic activitiesof some d0-(oxy)nitrides for H2 or O2 evolution in the presence of sacrificialreagents. These (oxy)nitrides exhibited relatively high photocatalytic activityfor water oxidation to form molecular oxygen under visible irradiation(l4420 nm), with some exceptions, such as ATaO2N (ACa, Sr, Ba) andLaTaON2. Among this group, TaON achieved the highest quantum efficiencyof 34% (entry 5, Table 1).12 It should be noted that water oxidation involving a4-electron process can be achieved by (oxy)nitrides with high quantum effi-ciencies, although silver nitrate is used as an electron accepter in such a system.In contrast, nitrogen-doped TiO2 displayed negligible activity for this reactionunder the same reaction conditions (entry 11, Table 1). (Oxy)nitrides alsoproduced H2 from an aqueous methanol solution upon visible irradiation whenloaded with nanoparticulate Pt as a cocatalyst for H2 evolution. In most cases,a low level of N2 evolution accompanied the initial stage of photocatalyticreactions over these catalysts, indicating that the (oxy)nitride is partially

Figure 5.4 Schematic illustration of band structures of Ta2O5, TaON and Ta3N5.Reprinted with permission from Reference 16. Copyright 2003 AmericanChemical Society.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

decomposed by the photogenerated holes (instead of water oxidation) ac-cording to the reaction:

2N3 6h ! N2 4

However, the production of N2 was completely suppressed as the reactionprogresses, and no change in the XRD pattern of the (oxy)nitrides as a result ofthe reaction has been detected. It thus appears that (oxy)nitride photocatalystsare essentially stable in the individual water reduction and oxidation reactions.

5.3 Improvement of Water Reduction Activity ofd0-(Oxy)nitrides

It should be noted that the activities for H2 production are about an order ofmagnitude lower than for O2 evolution, although the photocatalytic perform-ance for H2 evolution is stable. In this context, several new modifications toenhance the H2 evolution rate of (oxy)nitrides have been pursued. Here weintroduce such an attempt to reduce the particle size of Ta3N5. NanoparticulateTa3N5 can be readily prepared from nanoparticulate Ta2O5 precursor preparedthrough a precipitation method.21 Compared to the conventional bulk-typeTa3N5 particles (300500 nm), nanoparticulate Ta3N5 with a smaller particlesize of 3050 nm was shown to exhibit enhanced photocatalytic activity for H2evolution from aqueous solution containing methanol as an electron donorunder visible light. Diffuse reflectance spectroscopy and PEC measurements

Table 5.1 Photocatalytic activities of (oxy)nitrides with d0 electronicconfiguration for H2 or O2 evolution in the presence of sacrificialreagents under visible light (l4420 nm).a

Entry OxynitrideBand gapenergyb/eV

Activity/mmol h1

Ref.H2c O2

d

1 LaTiO2N 2.0 30 41 152 Ca0.25La0.75TiO2.25N0.75 2.0 5.5 60 153 TiNxOyFz 2.3 0.2 43 17, 194 CaNbO2N 1.9 1.5 46 145 TaON 2.5 20 660 126 Ta3N5 2.1 10 420 137 CaTaO2N 2.4 15 0 188 SrTaO2N 2.1 20 0 189 BaTaO2N 1.9 15 0 1810 LaTaON2 2.0 20 0 1811 N-doped TiO2 2.6

e n.d. 1.6 17, 19

aReaction conditions: 0.20.4 g of catalyst, 200mL aqueous solution containing sacrificial reagents,300W xenon lamp light source, Pyrex top irradiation-type reaction vessel with cutoff filter.bEstimated from onset wavelength of diffuse reflectance spectra.cLoaded with nanoparticulate Pt as a cocatalyst, reacted in the presence of methanol (10 vol%),sacrificial reagent.dSacrificial reagent: silver nitrate (0.01M).eEnergy gap estimated from a shoulder peak (400500 nm).

114 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

suggested that the enhancement was due to the promotion of the water re-duction process originating from the lower defect density of the material. It isalso possible to prepare nanosized Ta3N5 particles through a template ap-proach using a mesoporous carbon nitride (mpg-C3N4),

22 which has a potentialto reduce or oxidize water with visible light as well.27,28 This method allows oneto prepare Ta3N5 nanoparticles with a primary particle size that in principlereflects the pore size of mpg-C3N4. Nanoparticles Ta3N5 having the smallestparticle size, which was prepared from mpg-C3N4 with a pore size of 7 nm,showed approximately one order of magnitude higher activity than bulkTa3N5, as shown in Figure 5.5. In addition to the low particle size and highsurface area, the low density of defect sites would also result in an improvementof the activity, since fewer defect sites would be favorable for electron migrationfrom the Ta3N5 bulk to the surface and/or electron transfer from the con-duction band of Ta3N5 to the loaded Pt.

21

Reducing the particle size of TaON, which exhibits higher activity thanTa3N5 (see Table 1), has been examined in a similar manner. However, it wasdifficult to suppress the generation of defects during the nitridation process,because TaON is an intermediate phase between Ta2O5 and Ta3N5.

23 As aresult, the as-obtained TaON did not show an appreciable increase in photo-catalytic activity, despite a relatively small particle size (5080 nm). Althoughthe calcination of solid photocatalysts after preparation has been demonstratedto be an effective approach for reducing the density of defects,29,30 such post-calcination appears to be disadvantageous for TaON due to its low thermal

Figure 5.5 Time courses of 0.5wt% Pt-loaded Ta3N5 samples for H2 evolution fromaqueous methanol solution under visible light (l4400 nm), along with aTEM image of Ta3N5 nanoparticles. Catalyst (0.2 g); an aqueousmethanol solution (20 vol%, 400mL); light source, high pressure mercurylamp (450W); inner irradiation-type reaction vessel made of quartz orPyrex with or without an aqueous NaNO2 solution (2M) filter.Reprinted with permission from Reference 22. Copyright 2010 The RoyalSociety of Chemistry.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

stability. In fact, post-calcination of TaON under O2 or N2 results in a decreasein activity.31 The elimination of defects in TaON therefore remains a challenge.

To overcome this, a new method to reduce surface defects on TaON wasapplied using ZrO2 as a protector through a surface modification techni-que.24 Specifically, monoclinic ZrO2 nanoparticles are dispersed on the surfaceof Ta2O5 precursor before nitridation, thereby protecting Ta

51 cations in theTaON surface from being reduced by H2 (derived from disassociation of NH3at high temperatures) during nitridation, as schematically illustrated inFigure 5.6. As has been observed for other (oxy)nitrides such as LaTiO2N

15

and TiNxOyFz,19 the reduction of Ta51 cations in TaON during nitridation

generates reduced tantalum species, which can act as recombination centersbetween photogenerated electrons and holes. On the other hand, when ZrO2 isloaded on the surface of Ta2O5, tantalum cations at the interface betweenTa2O5 (and/or TaON) and the loaded ZrO2 are expected to interact with ZrO2and thereby become more cationic. As a result, the formation of reducedtantalum species on the TaON surface is effectively suppressed by prior loadingwith nanoparticulate ZrO2. As expected, the as-prepared ZrO2-modified TaON(represented as ZrO2/TaON) exhibited enhanced water reduction behaviorcompared to unmodified TaON. This enhancement of activity is attributed tothe reduced density of defects in ZrO2/TaON, which contributes to a lowerprobability of undesirable electron-hole recombination in ZrO2/TaON than in

Figure 5.6 Nitridation of ZrO2/Ta2O5 composite to produce ZrO2/TaON whilesuppressing the production of reduced tantalum species (defect sites)near the surface of the material.Reprinted with permission from Reference 24. Copyright 2008 TheChemical Society of Japan.

116 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

TaON. This material is also applicable to a building block for H2 evolution in atwo-step water splitting system under visible light, as will be presented later.

Improvement of water reduction was also attempted for a mixed metaloxynitride, CaTaO2N.

26 CaTaO2N was successfully prepared by nitriding amixture of layered Ca-Ta oxide (RbCa2Ta3O10 or HCa2Ta3O10) and CaCO3under NH3 flow. The as-prepared CaTaO2N consisted of aggregated nano-particles with a primary particle size of 50100 nm, and had a lower density ofanionic defects. Compared to CaTaO2N prepared from a bulk-type Ca-Taoxide in a conventional manner, CaTaO2N derived from the layered oxidesexhibited an enhanced photocatalytic activity (about 2.5 times) for H2 evo-lution under visible light.

5.4 Ge3N4, a Typical Metal Nitride with d10 Electronic

Configuration

As described above, (oxy)nitride photocatalysts having the ability to reduceand/or oxidize water under visible light had been composed of transition-metalcations of Ti41, Nb51, or Ta51 with d0 electronic configuration. From theviewpoint of the electronic band structure, however, d10-based semiconductingmaterials are advantageous over the d0 configurations as a photocatalyst in thatwhile the top of the valence band consists of O2p orbitals, the bottom of theconduction band is composed of hybridized s,p orbitals of typical metals.5 Thehybridized s,p orbitals possess large dispersion, leading to increased mobility ofphotogenerated electrons in the conduction band and thus high photocatalyticactivity. This has stimulated study of (oxy)nitrides with the d10 electronicconfiguration as photocatalysts for overall water splitting.

(Oxy)nitrides with a d10 electronic configuration are therefore of interest aspotentially efficient photocatalysts for overall water splitting. In evaluatingsuch d10-compounds according to this hypothesis, b-Ge3N4 was examined as awater-splitting photocatalyst.32 This compound can be readily prepared bynitriding GeO2 powder at 11231223K for 15 h. HR-TEM images and anelectron diffraction pattern of the as-prepared b-Ge3N4 indicated it to consist ofprimary well-crystallized rod-like particles with a hexagonal crystal system. Thespecific surface area of the as-prepared b-Ge3N4 was ca. 2.4m

2/g.It was found that b-Ge3N4 loaded with RuO2 nanoparticles functions as a

photocatalyst for overall water splitting. This was the first case involving a non-oxide photocatalyst for the cleavage of pure water. Since this discovery, thephotocatalytic properties and effects of post-treatment to enhance the activityof Ge3N4 have been examined in detail.

33,34 The relationship between thestructural characteristics and the photocatalytic performance of Ge3N4 has alsobeen investigated.35 Unfortunately, the band gap of b-Ge3N4 is about 3.8 eV,which is responsive only to ultraviolet light.34 Figure 5.7 shows the dependenceof the photocatalytic activity of RuO2-loaded b-Ge3N4 for overall watersplitting on the wavelength of incident light, along with a typical DRS ofb-Ge3N4. The rate of H2 and O2 evolution in overall water splitting decreased


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

with increasing the wavelength of the incident light, due to a decrease in thenumber of incident photons from the light source. No H2 and O2 evolution wasobserved when the reaction was carried out at wavelengths longer than 400 nmor in the dark. These results indicate that the sharp absorption band at 350 nm,indicative of the band gap transition from the valence band formed by N2porbitals to the conduction band formed by Ge4s,4p hybridized orbitals, con-tributes to promoting overall water splitting on RuO2-loaded b-Ge3N4. Inaddition, it is clear that the broad absorption in the visible-light region as-signable to impurities and/or defect sites is not available for the reaction. Thissituation forced us to search for a new d10 compound that can function undervisible light.

5.5 GaN-ZnO and ZnGeN2-ZnO Solid Solutions

To devise a new oxynitride with d10 electronic configuration that can de-compose water under visible light, the solid solution of GaN and ZnO,(Ga1xZnx)(N1xOx), was examined.

36,37 The (Ga1xZnx)(N1xOx) solid solu-tion is typically synthesized by nitriding a mixture of Ga2O3 and ZnO. Elem-ental analyses by inductively coupled plasma optical emission spectroscopy(ICP-OES) revealed that the ratios of Ga to N and Zn to O in the as-preparedmaterial are close to unity, and N and O concentrations increase with theGa and Zn concentrations, respectively. The atomic composition is controllableby changing the nitridation conditions.37 X-ray diffraction and neutronpowder diffraction analyses showed that the prepared samples are solid

Figure 5.7 (Left) Dependence of photocatalytic activity of 1 wt% RuO2-loaded b-Ge3N4 for overall water splitting on wavelength of incident light: l4 (A)200, (B) 300, and (C) 400 nm. Catalyst (0.5 g); an aqueous H2SO4 solution(1M, 350400mL); light source, high pressure mercury lamp (450W);inner irradiation-type reaction vessel made of quartz or Pyrex with orwithout an aqueous NaNO2 solution (2M) filter. (Right) UV-visiblediffuse reflectance spectrum of b-Ge3N4.Reprinted with permission from Reference 34. Copyright 2007 AmericanChemical Society.

118 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

solutions of GaN and ZnO with no interstitial sites and no large disorder in thematerial.38

While GaN and ZnO both have band gap energies greater than 3 eV and thusdo not absorb visible light, the (Ga1xZnx)(N1xOx) solid solution has ab-sorption edges in the visible region. Figure 5.8 shows the UV-visible DRS forseveral samples. The absorption edge of (Ga1xZnx)(N1xOx) lies at a wave-length longer than those of GaN and ZnO and shifts to longer wavelengths withincreasing Zn and O concentration (x) in (Ga1xZnx)(N1xOx). The band gapenergies of the solid solutions are estimated to be 2.42.8 eV based on the DRS.It is thus clear that the visible-light-response of (Ga1xZnx)(N1xOx) originatesfrom the presence of ZnO in the crystal. Initially, the origin of the visible-lightabsorption was thought to be band-gap narrowing of GaN due to pdrepulsion between Zn 3d and N 2p electrons in the upper part of the valenceband.36,37 Follow-up studies on the electronic structure of GaN-rich(Ga1xZnx)(N1xOx) using photoluminescence spectroscopy and DFT calcula-tions suggested that this material absorbs visible light via electron transitionsfrom the Zn acceptor level to the conduction band while maintaining the band-gap structure of the host GaN, as illustrated in Figure 5.9.39,40 Because theconcentration of Zn in (Ga1xZnx)(N1xOx) is relatively high, the acceptor level,which is filled with electrons derived from O donor levels (or thermal exci-tation), is likely to behave as an impurity band with a high density of states.Electrons can thus be transferred from the Zn acceptor level to the conductionband by visible-light absorption.

The as-prepared (Ga1xZnx)(N1xOx) exhibited little photocatalyticactivity for water decomposition even under UV irradiation. However, modi-fication by surface deposition of metal oxide nanoparticles as cocatalysts

Figure 5.8 UV-visible diffuse reflectance spectra for various (Ga1xZnx)(N1xOx)solid solutions.Reprinted with permission from Reference 3. Copyright 2007 AmericanChemical Society.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

resulted in clearly observable H2 and O2 evolution. The activity was stronglydependent on the loaded cocatalyst, and a mixed-oxide of Rh and Cr (Rh2yCryO3) is the most effective for (Ga1xZnx)(N1xOx) among the co-catalystmaterials examined.41,42 Figure 5.10 displays a typical time course of water

Figure 5.9 Expected energy level diagram for impurity levels in undoped GaN,Zn-doped GaN, and GaN-rich (Ga1xZnx)(N1xOx) solid solutions. Arrowsdenote photoabsorbtion, photoluminescence, and thermal relaxation.Reprinted with permission from Reference 40. Copyright 2010 AmericanChemical Society.

Figure 5.10 Time courses of overall water splitting on (Ga1xZnx)(N1xOx) loadedwith Rh2yCryO3 (1wt% Rh and 1.5wt% Cr) under (A) UV (l4300 nm)and (B) visible light irradiation (l4400 nm). Reactions were performedin pure water (370mL) with 0.3 g of catalyst powder under illuminationfrom a high-pressure mercury lamp (450W).Reprinted with permission from Reference 42. Copyright 2011 ElsevierB. V.

120 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

splitting over Rh2yCryO3-loaded (Ga1xZnx)(N1xOx). As can be seen, thisphotocatalyst produced both H2 and O2 in the stoichiometric ratio (H2/O2 2)without noticeable degradation. The photocatalytic performance was alsodependent on the pH of the reactant solution.43 The photocatalyst exhibitedstable and high photocatalytic activity in an aqueous solution at pH 4.5 for aslong as three days. The photocatalytic performance at pH 3.0 and pH 6.2 wasmuch lower, attributable to corrosion of the cocatalyst and hydrolysis of thecatalyst. It has also been confirmed by X-ray diffraction, X-ray photoelectronspectroscopy, and X-ray absorption fine structure that the crystal structure ofthe catalyst and the valence state of both the surface and bulk do not changeeven after reaction for three days at optimal pH (4.5). These results offer usefulguidelines for further research using other oxynitrides.

A solid solution of ZnGeN2 and ZnO, (Zn11xGe)(N2Ox), which is a similarmaterial to (Ga1xZnx)(N1xOx) in terms of crystal structure and optical ab-sorption, has also been found to be an active and stable photocatalyst foroverall water splitting under visible light.44,45 Both ZnGeN2 and ZnO arewide-gap semiconductors with band gaps larger than 3 eV, although the bandgap of ZnGeN2 is dependent on the crystal structure and composition.

46 The(Zn11xGe)(N2Ox) solid solutions can be prepared in a method similar to(Ga1xZnx)(N1xOx), but with a mixture of ZnO and GeO2 as the startingmaterial. The crystal structure of the synthesized sample was confirmed byboth Rietveld analysis and neutron powder diffraction to be wurtzite withspace group P63mc.

44 In the solid solution between ZnGeN2 and ZnO, theoxygen atoms are replaced with nitrogen sites. The band gap of the material isestimated to be 2.52.7 eV based on the absorption edges, which is smallerthan the band gaps of b-Ge3N4 (ca. 3.8 eV), ZnGeN2 (ca. 3.3 eV), and ZnO(ca. 3.2 eV). Figure 5.11 shows DRS spectra of ZnGeN2, ZnO, and a solidsolution between the two, along with the proposed scheme of the origin of thevisible-light response. The visible-light response of the material originatesfrom the wide valence bands consisting of N2p, O2p, and Zn3d atomicorbitals and pd repulsion between Zn3d and N2p and O2p electrons in theupper part of the valence bands. Therefore, the absorption edge of(Zn11xGe)(N2Ox) tends to be located at longer wavelengths with increasingZnO content.

Neither ZnGeN2 nor ZnO alone exhibited photocatalytic activity for overallwater splitting under UV irradiation. However, (Zn11xGe)(N2Ox) becamephotocatalytically active under visible irradiation when loaded with a propercocatalyst. Overall water splitting on a modified (Zn11xGe)(N2Ox) proceededby band gap photoexcitation from the valence band formed by N2p, O2p, andZn3d atomic orbitals to the conduction band consisting of Ge4s,4p hybridizedatomic orbitals. As demonstrated in the (Ga1xZnx)(N1xOx) section earlier, thephotocatalytic activity of the transition metal-loaded catalysts was improvedmarkedly by coloading Cr oxide.45 The largest improvement in activity wasobtained by loading the base catalysts with both 3 wt % Rh and 0.2 wt % Cr.It was confirmed that catalytic gas evolution of Rh2yCryO3-loaded(Zn11xGe)(N2Ox) is stable for as long as 60 h.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

In addition to cocatalyst-loading, preparation conditions of (Zn11xGe)(N2Ox)and the kind of precursors had a significant impact on the photocatalytic activity.As mentioned earlier, (Zn11xGe)(N2Ox) are typically prepared by reaction ofGeO2 and ZnO under a NH3 flow at 1123 K. With increasing nitridation time, thezinc and oxygen concentrations decreased due to reduction of ZnO and volatil-ization of zinc, and the crystallinity and band gap energy of the product increased.The highest activity for overall water splitting was obtained for the sample afternitridation for 15 h. Structural analyses revealed that the photocatalytic activity of(Zn11xGe)(N2Ox) for overall water splitting depends heavily on the crystallinityand composition of the material. The preparation route had a strong influence onphotocatalytic performance of (Zn11xGe)(N2Ox).

47 Improving the homogeneityof the (Zn11xGe)(N2Ox) powder, reducing the number of superficial defects orincreasing the crystallinity, showed a direct positive impact on the photocatalyticactivity for overall water splitting. On the other hand, photoreduction of water, ahalf reaction of overall water splitting, occurred with low Zn/Ge ratios and wateroxidation requires a high crystallinity. Moreover, overall water splitting wasachieved only if the crystal phase is active enough for photoreduction.

A

(a)

(b)B

C

Figure 5.11 UV-visible DRS traces for (A) ZnO, (B) ZnGeN2, and (C) a solidsolution between the two, (Zn1.44Ge)(N2.08O0.38).Reprinted with permission from References 44 and 45. Copyright 2007American Chemical Society.

122 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

5.6 Two-Step Water Splitting Mimicking NaturalPhotosynthesis in Green Plants

Despite the efforts mentioned above, overall water splitting using these d0-(oxy)nitrides has yet to be achieved. One plausible explanation for this is thatthey still have a high density of defect sites that can act as recombinationcenters for photogenerated electrons and holes. Nevertheless, Z-scheme overallwater splitting under visible light has been achieved using several combinationof (oxy)nitrides in the presence of an IO3

/I shuttle redox mediator.TaON modified with Pt was first applied as a component for H2 evolution in

a two-step water splitting system.48 When it was combined with Pt-loaded WO3as an O2 evolution photocatalyst, overall water splitting was achieved undervisible light in the presence of an IO3

/I shuttle redox. One reason for theeffectiveness of TaON as a H2 evolution component in an IO3

/I-basedZ-scheme system is the good ability of TaON to oxidize I into IO3

. However,this in turn implies that it is difficult to apply TaON into an O2 evolution system;I ions are more susceptible to oxidation than water, thereby hindering wateroxidation catalysis. Actually, neither Pt/TaON nor TaON are active for O2evolution half reaction from aqueous NaIO3 solution. Interestingly, modificationof TaON with nanoparticulate RuO2 resulted in observable O2 evolution.

49

Structural analyses and (photo)electrochemical measurements revealed that theactivity of RuO2/TaON is strongly dependent on the generation of optimallydispersed RuO2 nanoparticles, which simultaneously promote both the reductionof IO3

and oxidation of water.50 Thus, Z-scheme water splitting under visiblelight has been achieved using modified TaON catalysts.

ZrO2/TaON, introduced earlier, exhibits higher performance for H2 evolutionin Z-scheme water splitting than TaON. In particular, combining Pt-loadedZrO2/TaON with Pt/WO3 and an IO3

/I shuttle redox mediator achievedstoichiometric water splitting into H2 and O2 under visible light, yielding anAQY of 6.3% under irradiation by 420.5 nmmonochromatic light under optimalconditions, six times greater than the yield achieved using a TaON analog.51 Tothe best of the authors knowledge, this is the highest reported value to date for anon-sacrificial visible-light-driven water-splitting system.

Using ZrO2/TaON instead of TaON as a H2 evolution photocatalyst alsoallowed one to apply a modified Ta3N5 as an O2 evolution photocatalyst, ex-tending the available wavelength for O2 evolution up to 600 nm.

52 In this case,not only the use of ZrO2/TaON but also modification of Ta3N5 is very im-portant. Modification of Ta3N5 with nanoparticulate Ir and rutile titania(R-TiO2) achieved functionality as an O2 evolution photocatalyst in a two-stepwater-splitting system with an IO3

/I shuttle redox mediator under visible light(l4420 nm) in combination with a Pt/ZrO2/TaON H2 evolution photocatalyst.The loaded Ir nanoparticles acted as active sites to reduce IO3

to I, while theR-TiO2 modifier suppressed the adsorption of I

on Ta3N5, allowing Ta3N5 toevolve O2 in the two-step water splitting system.

Although the absorption edge of TaON and ZrO2/TaON (H2 evolutionphotocatalysts) is limited to 520 nm in wavelength, replacing these


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

photocatalysts with BaTaO2N extended the wavelength available for H2 evo-lution up to 660 nm.53 The photocatalytic performance of BaTaO2N for H2evolution can be improved by 69 times as a result of forming a solid solutionwith BaZrO3, due to an enhanced driving force for the redox reactions and thereduction of defect density.54

Thus, several combinations of (oxy)nitrides for Z-scheme water splitting inthe presence of an IO3

/I shuttle redox mediator have been achieved, asschematically illustrated in Figure 5.12. The absorption wavelength availablefor the individual H2 and O2 evolution has been increased up to 660 nm and600 nm, respectively. The next challenges remain in the promotion of electrontransfer between two semiconductors and in the suppression of backwardreactions involving shuttle redox mediators.

5.7 Photoelectrochemical Water Splitting Usingd0-Oxynitrides

(Oxy)nitrides are also useful as photoelectrode materials. While the separationof H2 and O2 produced in water splitting reaction is one of the biggest issues tobe solved toward practical application, however, it is in principle possible toproduce H2 and O2 separately in a PEC cell, as shown in Figure 5.13. Besides,in this approach, not only light energy but also electric energy can be put intothe system for operation, resulting in a relatively high incident photon-to-current conversion efficiency (IPCE).

As mentioned earlier, most (oxy)nitrides have band edge potentials suitablefor water splitting under visible light (o3 eV). This suggests that they are, inprinciple, capable of working as photoelectrodes to split water even without an

Figure 5.12 A schematic illustration of a two-step water-splitting system based onoxynitrides in the presence of an iodate/iodide shuttle redox mediator.The absorption edges of photocatalysts are also shown.

124 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

externally applied potential. This is distinctly different from the conventionalmetal-oxide based visible-light-responsive electrodes such as WO3 and Fe2O3,which need an external bias for operation, due to their conduction band po-tential being located at more positive potential than water reductionpotential.55,56

Several (oxy)nitrides such as TaON,31,57 Ta3N5,5860 and LaTiO2N,

61,62 havebeen studied as water-splitting photoelectrodes, and it was found that theyfunction as photoanode materials due to their n-type semiconducting character.Abe et al. have developed a simple method for preparing efficient photoanodesof TaON and Ta3N5.

57,60 In this method, TaON or Ta3N5 particles are firstdeposited on a conductive glass (fluorine-doped tin oxide; FTO) substrate byelectrophoretic deposition, and then a necking treatment is applied to formeffective contacts between the deposited particles. Figure 5.14 displays thestructure of electrodes after various treatments and the corresponding per-formance for PEC water splitting. The as-deposited TaON electrode, even aftercalcination at 673K, gave little photoresponse (Figure 5.14(a)). However,TaCl5 treatment followed by calcination at 723K in air resulted in an appre-ciable increase in the photocurrent (Figure 5.14(b)). This increase in photo-current was attributed to the presence of Ta2O5 bridges that promote electrontransfer between TaON particles. Subsequent heating of the same electrodesample under NH3 significantly increased the photocurrent (Figure 5.14(c)).XPS analysis indicated that most Ta2O5 bridges transformed into TaON,greatly facilitating the interparticle electron transport in the electrode structure.Modification of this electrode with colloidal IrO2, which is known to be aneffective catalyst for water oxidation,63 further enhanced the photocurrent(Figure 5.14(d)). The onset potential of IrO2-loaded post-necked TaON elec-trode was about 0.25V vs. RHE, indicating that the electrode can potentiallysplit water into H2 and O2 even without an applied bias. IrO2 loading was alsofound to improve the stability of TaON electrodes during photoelectrolysis,with nearly stoichiometric H2 and O2 evolution in a two-electrode

Figure 5.13 Photoelectrochemical water splitting systems using n-type semiconductorphotoanode (a), p-type semiconductor photocathode (b), and tandemsystem (c).Reprinted with permission from Reference 6. Copyright 2011 Elsevier B. V.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

configuration in the presence of an applied bias greater than 0.6V vs. Pt wirecathode. A similar result was observed when Ta3N5 was employed as an anodematerial. The IPCE of the optimized TaON and Ta3N5 electrode recorded at

(a) (b)

(c) (d)

Figure 5.14 (Top) Schematic illustration of post-necking process and IrO2 loading onTaON, and (bottom) Currentpotential curves in aqueous 0.1M Na2SO4solution (pH 6) under chopped visible light irradiation (l4400 nm) forTaON electrodes as-prepared (a), treated by TaCl5 (b), heated in NH3(c), and loaded with IrO2.Reprinted with permission fromReference 6. Copyright 2011 Elsevier B. V.

126 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

1.15V vs. RHE was ca. 76% at 400 nm and 31% at 500 nm, the former of whichis the highest IPCE among oxynitride or nitride semiconductor photoelectrodesreported so far.

In terms of available wavelength, photons with wavelengths up to 600 nmare available for PEC water oxidation with Ta3N5

5860 and LaTiO2N.61,62

However, no photoanode (including metal oxides) with a band gap smallerthan 2.0 eV (corresponding to a 600 nm absorption band) that can oxidizewater without application of an external potential (except through the use of atandem cell configuration64) has been reported to date. As the band gap ofa given material is decreased, the driving forces for water oxidation andreduction should inevitably decrease, making water splitting more difficult.

In such a situation, a niobium-based oxynitride, SrNbO2N, was shown toachieve the functionality as a photoanode to oxidize water even without ap-plying an external bias.65 SrNbO2N is a perovskite-type oxynitride consisting ofrelatively earth-abundant metals that has a band gap of ca. 1.8 eV.66 SrNbO2Npowder also has the ability to photocatalytically reduce or oxidize water intoH2 or O2 in the presence of proper electron donors or acceptors, respectively.When particulate SrNbO2N modified with colloidal IrO2 is employed as aphotoelectrode for visible-light-driven water splitting, it acts as an activephotoanode to oxidize water, even without an externally applied potential.Nearly stoichiometric H2 and O2 evolution was observed during photo-electrolysis in a neutral electrolyte of Na2SO4 (pHE6) when a potential of1.0-1.55V vs. RHE was applied.Figure 5.15 shows the dependence of photocurrent generated from the post-

necked SrNbO2N electrode modified with colloidal IrO2 at 0.95V vs. RHE onthe cutoff wavelength of the incident light, along with the DRS of SrNbO2N.

Figure 5.15 Dependence of photocurrent generated from the IrO2-modifiedSrNbO2N electrode (6 cm

2) at 0.95V vs. RHE in aqueous 0.1M Na2O4solution upon the cutoff wavelength of the incident light. Diffuse reflect-ance spectrum for SrNbO2N is also shown.Reprinted with permission from Reference 65. Copyright 2011 AmericanChemical Society.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

The photocurrent was decreased with cutoff wavelength, reaching almost zeroat 700 nm, which corresponds to the absorption edge of SrNbO2N. This resultindicates that PEC water oxidation occurs through light absorption bySrNbO2N, and that an absorption band at longer wavelengths, assignable toreduced Nb species (e.g. Nb41) which are defects of the material, does notcontribute to the reaction. The main problem of this electrode material is thelow IPCE, which was 0.2% at 400 nm in the presence of a 1.23V (vs.RHE) bias.

5.8 Conclusions

A range of (oxy)nitride materials for photocatalytic overall water splitting hasbeen developed. (Oxy)nitrides have been found to function as stable photo-catalysts for water reduction and oxidation under visible irradiation, and the(Ga1xZnx)(N1xOx) and (Zn11xGe)(N2Ox) solid solutions with d

10 electronicconfiguration have been shown to achieve overall water splitting under visiblelight without noticeable degradation. Some of d0-type (oxy)nitrides are ap-plicable to a two-step water splitting system that can harvest a wide range ofvisible photons. Carefully prepared photoelectrodes based on d0-(oxy)nitridessuch as TaON and Ta3N5 exhibit excellent IPCEs for splitting water undervisible light in the presence of an external bias.

References

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37.2. J. S. Lee, Catal. Surv. Asia, 2005, 9, 217.3. K. Maeda and K. Domen, J. Phys. Chem. C, 2007, 111, 7851.4. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.5. Y. Inoue, Energy Environ. Sci., 2009, 2, 364.6. R. Abe, J. Photochem. Photobiol. C: Photochem. Reviews, 2010, 11, 179.7. R. Williams, J. Chem. Phys., 1960, 32, 1505.8. A. B. Ellis, S. W. Kaiser, J. M. Bolts and M. S. Wrighton, J. Am. Chem.

Soc., 1977, 99, 2839.9. D. E. Scaife, Solar Energy, 1980, 25, 41.

10. F. Tessier and R. Marchand, J. Alloys and Compounds, 1997, 262263, 410.11. M. Jansen and H. P. Letschert, Nature, 2000, 404, 980.12. G. Hitoki, T. Takata, J. N. Kondo, M. Hara, H. Kobayashi and

K. Domen, Chem. Commun., 2002, 1698.13. G. Hitoki, A. Ishikawa, T. Takata, J. N. Kondo, M. Hara and K. Domen,

Chem. Lett., 2002, 31, 736.14. G. Hitoki, T. Takata, J. N. Kondo, M. Hara, H. Kobayashi and

K. Domen, Electrochem., 2002, 70, 463.15. A. Kasahara, K. Nukumizu, G. Hitoki, T. Takata, J. N. Kondo, M. Hara,

H. Kobayashi and K. Domen, J. Phys. Chem. A, 2002, 106, 6750.

128 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

16. W.-A. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara,M. Kawai, Y. Matsumoto and K. Domen, J. Phys. Chem. B, 2003,107, 1798.

17. K. Nukumizu, J. Nunoshige, T. Takata, J. N. Kondo, M. Hara,H. Kobayashi and K. Domen, Chem. Lett., 2003, 32, 196.

18. D. Yamasita, T. Takata, M. Hara, J. N. Kondo and K. Domen, Solid StateIonics, 2004, 172, 591.

19. K. Maeda, Y. Shimodaira, B. Lee, K. Teramura, D. Lu, H. Kobayashi andK. Domen, J. Phys. Chem. C, 2007, 111, 18264.

20. K. Maeda, B. Lee, K. Teramura, D. Lu, H. Kobayashi and K. Domen,J. Phys. Chem. C, 2007, 111, 18264.

21. K. Maeda, N. Nishimura and K. Domen, Appl. Catal. A: Gen., 2009,370, 88.

22. L. Yuliati, J.-H. Yang, X. Wang, K. Maeda, T. Takata, M. Antonietti andK. Domen, J. Mater. Chem., 2010, 20, 4295.

23. K. Maeda, H. Terashima, K. Kase and K. Domen, Appl. Catal. A: Gen.,2009, 357, 206.

24. K. Maeda, H. Terashima, K. Kase, M. Higashi, M. Tabata and K. Domen,Bull. Chem. Soc. Jpn., 2008, 81, 927.

25. S. S. K. Ma, K. Maeda and K. Domen, Catal. Sci. Technol., 2012, 2,818.

26. R. Sasaki, K. Maeda, Y. Kako and K. Domen, Appl. Catal. B: Environ.,2012, 128, 72.

27. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson,K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76.

28. X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou, X. Fuand M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680.

29. K. Sayama, H. Arakawa and K. Domen, Catal. Today, 1996, 28, 175.30. H. Kominami, S. Murakami, Y. Kera and B. Ohtani, Catal. Lett., 1998,

56, 125.31. R. Abe, T. Takata, H. Sugihara and K. Domen, Chem. Lett., 2005,

34, 1162.32. J. Sato, N. Saito, Y. Yamada, K. Maeda, T. Takata, J. N. Kondo,

M. Hara, H. Kobayashi, K. Domen and Y. Inoue, J. Am. Chem. Soc.,2005, 127, 4150.

33. Y. Lee, T. Watanabe, T. Takata, M. Hara, M. Yoshimura and K. Domen,J. Phys. Chem. B, 2006, 110, 17563.

34. K. Maeda, N. Saito, D. Lu, Y. Inoue and K. Domen, J. Phys. Chem. C,2007, 111, 4749.

35. K. Maeda, N. Saito, Y. Inoue and K. Domen, Chem. Mater., 2007,19, 4092.

36. K. Maeda, T. Takata, M. Hara, N. Saito, Y. Inoue, H. Kobayashi andK. Domen, J. Am. Chem. Soc., 2005, 127, 8286.

37. K. Maeda, K. Teramura, T. Takata, M. Hara, N. Saito, K. Toda,Y. Inoue, H. Kobayashi and K. Domen, J. Phys. Chem. B, 2005,109, 20504.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

38. M. Yashima, K. Maeda, K. Teramura, T. Takata and K. Domen, Chem.Phys. Lett., 2005, 416, 225.

39. T. Hirai, K. Maeda, M. Yoshida, J. Kubota, S. Ikeda, M. Matsumura andK. Domen, J. Phys. Chem. C, 2007, 111, 18853.

40. M. Yoshida, T. Hirai, K. Maeda, N. Saito, J. Kubota, H. Kobayashi,Y. Inoue and K. Domen, J. Phys. Chem. C, 2010, 114, 15510.

41. K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue andK. Domen, Nature, 2006, 440, 295.

42. K. Maeda, K. Teramura, N. Saito, Y. Inoue and K. Domen, J. Catal.,2006, 243, 303.

43. K. Maeda, K. Teramura, H. Masuda, T. Takata, N. Saito, Y. Inoue andK. Domen, J. Phys. Chem. B, 2006, 110, 13107.

44. Y. Lee, H. Terashima, Y. Shimodaira, K. Teramura, M. Hara,H. Kobayashi, K. Domen and M. Yashima, J. Phys. Chem. C, 2007,111, 1042.

45. Y. Lee, K. Teramura, M. Hara and K. Domen, Chem. Mater., 2007,19, 2120.

46. T. Misaki, X. Wu, A. Wakahara and A. Yoshida, Proc. Int. WorkshopNitride Semiconductors IPAP Conf. Series 1, 2000, 685.

47. F. Tessier, P. Maillard, Y. Lee, C. Bleugat and K. Domen, J. Phys. Chem.C, 2009, 113, 8526.

48. R. Abe, T. Takata, H. Sugihara and K. Domen, Chem. Commun., 2005,3829.

49. M. Higashi, R. Abe, A. Ishikawa, T. Takata, B. Ohtani and K. Domen,Chem. Lett., 2008, 37, 138.

50. K. Maeda, R. Abe and K. Domen, J. Phys. Chem. C, 2011, 115, 3057.51. K. Maeda, M. Higashi, D. Lu, R. Abe and K. Domen, J. Am. Chem. Soc.,

2010, 132, 5858.52. M. Tabata, K. Maeda, M. Higashi, D. Lu, T. Takata, R. Abe and

K. Domen, Langmuir, 2010, 26, 9161.53. M. Higashi, R. Abe, K. Teramura, T. Takata, B. Ohtani and K. Domen,

Chem. Phys. Lett., 2008, 452, 120.54. T. Matoba, K. Maeda and K. Domen, Chem. Eur. J., 2011, 17, 14731.55. C. Santato, M. S. Ulmann and J. Augustynski, J. Phys. Chem. B, 2001,

105, 936.56. S. D. Tilley, M. Cornuz, K. Sivula and M. Gratzel, Angew. Chem., Int. Ed.,

2010, 49, 6405.57. R. Abe, M. Higashi and K. Domen, J. Am. Chem. Soc., 2010, 132,

11828.58. A. Ishikawa, T. Takata, J. N. Kondo, M. Hara and K. Domen, J. Phys.

Chem. B, 2004, 108, 11049.59. D. Yokoyama, H. Hashiguchi, K. Maeda, T. Minegishi, T. Takata,

R. Abe, J. Kubota and K. Domen, Thin Solid Films, 2011, 519, 2087.60. M. Higashi, K. Domen and R. Abe, Energy & Environ. Sci., 2011, 4, 4138.61. C. Le Paven-Thivet, A. Ishikawa, A. Ziani, L. L. Gendre, M. Yoshida,

J. Kubota, F. Tessier and K. Domen, J. Phys. Chem. C, 2009, 113, 6156.

130 Chapter 5

Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

62. N. Nishimura, B. Raphael, K. Maeda, L. Le Gendre, R. Abe, J. Kubotaand K. Domen, Thin Solid Films, 2010, 518, 5855.

63. A. Harriman, I. J. Pickering, J. M. Thomas and P. A. Christensen,J. Chem. Soc., Faraday Trans. 1, 1988, 84, 2795.

64. O. Khaselev and J. A. Turner, Science, 1998, 280, 425.65. K. Maeda, M. Higashi, B. Siritanaratkul, R. Abe and K. Domen, J. Am.

Chem. Soc., 2011, 133, 12334.66. Y. Kim, P. M. Woodward, K. Z. Baba-Kishi and C. W. Tai, Chem. Mater.,

2004, 16, 1267.


Dow

nloa

ded

by U

nive

rsity

of

Lan

cast

er o

n 17

/01/

2015

21:

41:1

8.

Publ

ishe

d on

02

Oct

ober

201

3 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/9

7818

4973

7739

-001

09View Online

http://dx.doi.org/10.1039/9781849737739-00109

fig1fig1fig2fig2fig3fig3fig4fig4fig5fig5fig6fig6fig7fig7fig8fig8fig9fig9fig10fig10fig11fig11fig12fig12fig13fig13fig14fig14fig15fig15

Chapter 5

Documents

Transcript of Chapter 5