Insight into the origin of the limited activity and stability of p-Cu2O films in photoelectrochemical proton reduction

(1)

Insight

into

the

origin

of

the

limited

activity

and

stability

of

p-Cu

2

O

ﬁlms

in

photoelectrochemical

proton

reduction

A.Wouter

Maijenburg

a,b,

*

,

Michel

G.C.

Zoontjes

a

,

Guido

Mul

a,

*

a

PhotoCatalyticSynthesisGroup,MESA+InstituteforNanotechnology,UniversityofTwente,P.O.Box217,7500AEEnschede,TheNetherlands b

PresentAddress:ZIKSiLi-nano,Martin-Luther-UniversityHalle-Wittenberg,Karl-Freiherr-von-Fritsch-Straße3,06120Halle(Saale),Germany

ARTICLE INFO Articlehistory:

Received15February2017

Receivedinrevisedform16May2017 Accepted17May2017

Availableonline19May2017 Keywords: p-Cu2Ophotocathode electronscavenger photoelectrochemistry gaschromatography ABSTRACT

Theoriginofinstabilityofp-Cu2OﬁlmsdepositedonaplatinizedSisubstratewhenusedasphotocathode

inphotoelectrochemicalwatersplitting,wasstudiedintheabsenceorpresenceofaprotectivelayerof

RuO2.Whenappliedat+0.3Vvs.RHEandatpH7,p-Cu2Oﬁlmswerefoundtoshowaslightlymorestable

performanceascomparedtophotoelectrochemicalmeasurementsreportedintheliteratureat0Vvs.

RHEandunderacidicconditions.Inaddition,thestabilityandthephotocurrentinducedbytheCu2O

filmsweresignificantlyimprovedwhenH2O2wasaddedtotheelectrolyte,whichisexplainedbyefficient

scavengingofelectrons,yieldingoxygenandwaterasconﬁrmedbygaschromatography(GC).Also,other

electronacceptorsimprovedthephotocatalyticperformanceofthep-Cu2Oﬁlms,demonstratingthatthe

transferofphoto-excitedelectronstoprotonsadsorbedonthesurfaceistheratedeterminingstepin

p-Cu2Obasedphoto-electrochemicalwatersplitting.WeconﬁrmedthatdepositionofRuO2improvesthe

stabilityoftheﬁlms,buttotheexpenseofadecreaseinphotocurrentdensity.Theresultsprovidedinthis

studyrationalizetheattachmentofaneffectiveH2evolutioncatalystasameanstosigniﬁcantlyimprove

thestabilityofp-Cu2Oelectrodes.

(http://creativecommons.org/licenses/by/4.0/).

1.Introduction

Oneofthetechniquesthatisemergingasapromisingmethod for storage of renewable energy, is photocatalytic or photo-electrochemicalsplittingofwater,yieldinghydrogen(andoxygen) gas.Unfortunately,theefficiency ofmanycellconfigurationsin solar-to-hydrogenconversionislow,becauseeitherthebandgap oftheappliedsemiconductorsistoolargeandonlyphotonsofthe UVregionofthesolarspectrumcanbeefficientlycaptured(i.e. TiO2, ZnO and SrTiO3), or the conduction band minimum is

positionedunfavourably,sothatwater(proton)reductionisnot thermodynamicallyfeasible(forexamplethecaseforFe2O3and

WO3) [1,2]. In order to achieve overall water splitting using

semiconductorswithasmallbandgap,atwo-photonsystembased on a Z-scheme can be employed, which requires two semi-conductors (a H2- and an O2-evolving photocatalyst) that are

electricallyconnected viaan electronmediator [1,3,4]. Alterna-tively,bothsemiconductorscanbeconnectedthroughanOhmic

contact, constituting a so-called p-ntype photoelectrochemical diode[5,6].Recently,Zoontjesintroducedaninnovativedesignofa watersplittingdevice,containingaPtdividerseparatinghydrogen and oxygen, which is simultaneously used as Ohmic contact betweenthephotocathodeandanode[7].

Todate,manyn-typeoxidesemiconductorshavebeenreported to bepromising photo-anodes stimulatingwater oxidation (i.e. WO3[8]andBiVO4[9]),buttheefﬁciencyand

photoelectrochem-ical stability of p-type oxide semiconductors suitable for H2

formationislimited.Despitedevelopmentofnovelp-type(oxide) photocathodematerialssuchasCuFeO2[10]andCu2ZnSnS4[11],

p-Cu2Oisstilloneofthemostpromisingp-typeoxidematerialsfor

photocatalyticandphotoelectrochemicalwatersplitting.p-Cu2O

has many favourable properties: it is abundant, cheap and environmentallybenign,anditsbandgapof1.9-2.2eVissuitable for solarenergy conversion in the visible light region[12–14]. Furthermore, the conduction band minimum lies signiﬁcantly negative of the equilibrium potential of water reduction [14]. Unfortunately, the stability of p-Cu2O is low and requires

improvement.Cu2Ohasalimitedphotoelectrochemicalstability

inaqueoussolutions,sincetheredoxpotentialsforthereduction and oxidation of Cu2OtoCu and CuO,respectively, arelocated

withinitsbandgap[14].Inpreviousreports,differentstrategies * Correspondingauthor.

E-mailaddresses:wouter.maijenburg@chemie.uni-halle.de(A.W.Maijenburg), g.mul@utwente.nl(G.Mul).

http://dx.doi.org/10.1016/j.electacta.2017.05.114

Electrochimica

Acta

(2)

wereemployedfortheenhancementofthephotoelectrochemical stabilityofCu2O,includingatomiclayerdepositionofalayerof

ZnO:AlandTiO2withorwithoutanadditionalRuO2layer[14,15],

protection with a layer of carbon [16], improvement of the crystallinity and reduction of defects by annealing [17], and chemicaletchingofthesurfacebyMV2+_reduction_[18]_._The_ﬁrst

threetechniqueshavetheadvantagethatthephotocurrentwas simultaneouslyenhancedwiththestability(seeTable1),buttheir disadvantageisthatthephotocurrentstilldecreasessigniﬁcantly withintheﬁrst20minutes;onlytheprotectionwithanadditional RuO2 layer seemstobecapableof improvingthe

photoelectro-chemicalstability of Cu2Oas photocathode,showing94% ofits

initialactivityafter8hofillumination[15].Inanotherstudyfrom Sowersetal.,itisproposedthatthecrystalorientationandsurface termination of Cu2O play a signiﬁcant role in determining its

photoelectrochemicalstability,andtheyshowthatCu+_-terminated

(111)surfacesaremorestablethanthemorecommonlyobtained O2-terminated(100)surfaces[19].

Comparingthedifferentelectrolytesandpotentialsusedinthe literaturetothePourbaixdiagramofCu[20],surprisinglymany photoelectrochemical measurements reported in the literature wereperformedatconditionsinwhichCu2Othermodynamically

corrodes in the absence of illumination. From the Pourbaix diagram it can beobserved that shifting the applied potential from0Vvs.RHEto0.3Vvs.RHEandbyusinganelectrolytewith neutralpH,Cu2Oisalreadymorestable,andtheuseofaprotective

coatingisnotprincipallynecessary[20].Therefore,weusedthese conditions(pH7and0.3Vvs.RHE)inthepresentstudyforthe evaluationof thephotocatalyticactivityof differentCu2O ﬁlms

depositedonaplatinizedSisubstrate,servingasamodelsubstrate forthemetallicdividerproposedbyZoontjesetal.[7]. Further-more, we studied the effect of different hole and electron scavengers on both the photocatalytic activityand the photo-electrochemicalstabilityofCu2Oﬁlmsintheabsenceorpresence

ofaRuO2layer,inordertoevaluatethelimitingfactorsthatresult

in the poor performance of Cu2O. Although it is known that

electronacceptors,suchasH2O2,improvethestabilityofCu2O,we

providemoredetail in thechemistryand Faradaic efﬁciency of theseCu2O ﬁlmsby combining PECmeasurements with

ultra-sensitivereal-timeGasChromatographymeasurements(allowing H2andO2detectionintheppbrange).

2.Experimentaldetails

Allchemicalsusedwerepurchased fromcommercialsources andusedwithoutfurtherpuriﬁcation.Coppersulphate pentahy-drate(CuSO4

5H2O,p.a.quality)andsodiumsulphate(Na2SO4,p.a.

quality) were purchased from Boom Chemie;lactic acid(extra pure), sodium hydroxide (NaOH, purity 98.5%), sulfuric acid (H2SO4, 96% in water) and dipotassium phosphate trihydrate

(K2HPO4

3H2O,purity>99%)werepurchasedfromAcrosOrganics;

chloroplatinic acid hexahydrate (H2PtCl6

6H2O, powder and

chunks), potassium perruthenate (KRuO4), potassium sulphate

(K2SO4,purity99%),hydrogenperoxide(H2O2,30wt%inH2O),

sodium persulphate (Na2SO8, purity98%), sodium sulphite

(Na2SO3, purity 98-100%) and methanol were purchased from

Sigma-Aldrich. Double distilled water with a resistivity of 18.2M

V

cmwasusedinallexperiments.

Cu2OlayersweremadebyelectrodepositiononaplatinizedSi

wafer as the working electrode. The platinized Si wafer was obtainedbydepositionofalayerof13nmTi,and subsequently 100nmPtbysputtering,usingacommerciallyavailable<100>Si waferandahome-buildsputteringsetup,byapplyinga6.6mbar argonatmosphereanda200Wdirectcurrent.Asimilarplatinized Si wafer was used as counter electrode during the deposition procedure,andAg/AgClin3MKCl(MetrohmAutolab)wasusedas referenceelectrode.TheelectrodeswereconnectedtoanAutolab PGSTAT128Npotentiostat.p-Cu2Owasdepositedat0.4Vvs.Ag/

AgCl fromanaqueouselectrolyte containing0.02MCuSO4 and

0.4Mlacticacid.Beforedeposition,thesolutionwasadjustedtopH 11usingNaOHand heatedinawater bathtoa temperatureof 60C.

RuO2layersweredepositedbyphoto-enhanced

electrodeposi-tioninadifferentthree-electrodecellconnectedtoaVersastat4 potentiostat (Princeton). Here, the working electrode was the previouslydepositedCu2Oﬁlm,thecounterelectrodeconsistedof

aPtmeshconnectedtoaPtwire,andAg/AgClin3MKCl(BASi)was usedasthereferenceelectrode.Thereactorconsistedofa25mL opticalglasscuvette(HellmaAnalytics),whichﬁlteredmostofthe emittedUV light.In this cell,theRuO2 layerwas depositedfor

15min at 33.7

m

A/cm2 _from _an _aqueous _solution _containing

1.3mMKRuO4underconstantilluminationwithanAM1.5Gsolar

simulator(model10500,ABETTechnologies)[15].

Photoelectrochemical(PEC)measurementswereperformedin an electrolyte consisting of 0.1MK2SO4 with or without an

additional0.1Mofanelectron-orholescavenger(H2O2,S2O82,

methanol,SO32,PO43).Asareference,theelectrolytedescribed

byParacchinoetal.(1.0MNa2SO4with0.1MK2HPO4atpH4.9)

was used [14]. Linear Sweep Voltammetry (LSV) curves were measuredaspartofacyclicvoltammogram(CV),measuredinthe positivedirectionfrom0.5Vvs.Ag/AgClto0.5Vvs.Ag/AgClat 0.05V/s,ofwhichthesecondhalfoftheﬁrstCVwastakenasthe LSVwhenthecurvewasmeasuredinthenegativedirectionfrom 0.5V vs. Ag/AgCl to 0.5V vs. Ag/AgCl. For prolonged PEC measurements,a 300WXelampwithAM1.5G ﬁlter(Newport Corporation)withanautomaticshutterwasusedincombination withthesameelectrochemicalcellandpotentiostat.Thepotential was converted to the RHE reference electrode by the Nernst equation:

Eðvs:RHEÞ¼Eðvs:Ag=AgClÞþEAg=AgClðrefÞþ0:0591V

pH;(1)

E_Ag=AgClðrefÞ¼0:1976Vvs:RHEat 25C:(2)

Next to the PEC measurements, the photoelectrochemical stabilityoftheCu2OﬁlmsbeforeandafterPECmeasurementswas

evaluatedbyXRDandSEM.Foranalysisofthecrystalstructureof Cu2O and thepresenceof Cu and/orCuOin theCu2Olayers,a

BrukerD2powderdiffractometer(equippedwithaCuK

a

source) was used. A Nova 600-nanolab HRSEM instrument (FEI Instru-ments)wasusedforimagingoftheCu2Osurface.

GasChromatography(GC) measurementswereperformedin combinationwiththePECmeasurementsinordertoinvestigate Table1

OverviewofstabilityenhancementstrategiesandobtainedphotocurrentsofCu2O.

Stabilityenhancedby Electrolyteused Potentialused

(Vvs.RHE)

Photocurrent obtained(mA/cm2₎

Ref.

ALDofZnO:AlandTiO2 1.0MNa2SO4and0.1MKxH3-xPO4(pH4.9) 0 7.6 [14]

ZnO:Al,TiO2andRuO2 0.5MNa2SO4and0.1MKxH3-xPO4(pH5.0) 0 5 [15]

Clayerprotection 1.0MNa2SO4(pH7) 0 3.95 [16]

Annealing 0.5MNa2SO4(pH7) 0.3 0.143 [17]

(3)

the mechanistic aspects involved in the photoelectrochemical watersplittingoftheCu2O-RuO2ﬁlmsinthedifferentelectrolytes.

Tothisend,aCompactGC(Interscience)equippedwithapulsed dischargedetector (PDD)was used witha 7N heliumpurge of 10mL/min. The combined GC and PEC measurements were performedina home-buildPECcellwithaquartzwindowand a gas inlet and outlet for the GC measurements. Before each measurement,theelectrolytewas purgedwithHeovernight to removeasmuchO2andN2fromthesolutionaspossible.Thenext

morning,theO2deﬁcientelectrolytewasinsertedintothePECcell

andpurgedfor1extrahourbeforestartingtheGCmeasurement. 3.Resultsanddiscussion

Fig.1a shows thesurfaceof anas-depositedCu2Oﬁlmwith

cubic crystals ranging in size from 100nm to 1

m

m. The XRD pattern shownin Fig.1c (blackcurve) demonstrates that Cu2O

crystalsgrow preferentiallyinthe(111) directionwithaminor fractioninthe(220)direction,inducedbythepresenceoflactic acidintheprecursorsolution.Also,asmallCupeakwasobserved, which was apparently deposited along with Cu2O. The Pt

diffractionlineisduetothePtﬁlmpresentonSi,usedassubstrate fortheCu2Oﬁlm.Afterphotoelectrochemical(PEC)measurements

for30mininneutralsolution(0.1MK2SO4,pH7),itwasobserved

thatthesurfaceoftheCu2Ocrystalsroughened(Fig.1b).Asthe

XRDpatternsofthesamplesbeforeandafterPECmeasurements aresimilar(Fig.1c),itisexpectedthatrougheningonlytookplace atthetopsurfaceanddidnotlargelyaffectthecrystalintegrityof Cu2O.Mostimportantly,noadditionalpeaksformetallicCuwere

foundfor the XRD measurement of the p-Cu2O ﬁlm after PEC

measurement. As can be seen from Fig. 1d, the measured photocurrent was quite stable and decreased only by 10% in the ﬁrst few minutes, followed by stable performance in the remainder of the measurement (30min total). Also, the dark current only slightly increased during the ﬁrst few min of measurement, but thenstayed constant. However,it should be notedthatweobservedamuchsmallerphotocurrentdensitythan valuesreportedintheliteratureat0.3V[14].Thistotalcurrent density of0.06mA/cm2 _yields _the _equivalent _of _0.11 _C/cm2

passedthroughthecellafter30min.Assumingallchargeisusedto

reduce Cu2O tometallic copper(2 electrons per Cu2O unit), a

maximumof136nmofCu2Oisreduced(usingthemolarmassof

Cu2Oof143g/mol,andadensityof6g/cm3).Asthisisonlyabout

7%ofthetotalthicknessoftheﬁlm(2.5

m

maccordingtotheSEM cross section, see Fig. 3b), this could explain the observed degradationbeinglimitedtothetopsurfaceofthep-Cu2Oﬁlm

(Fig.1b). Furthermore, we observed largecathodic and anodic spikes in current when the light was turned on and off, respectively; thesespikesindicatesigniﬁcant accumulation and recombination of photoexcitedstates(charges), suggesting that theH2evolution rateis lowcompared totheformation rate of

electron-holepairsbylightabsorption[21].

Even though the measurement conditions used seemed to provide a relatively higher photoelectrochemical stability, the photocurrentdensityof0.06mA/cm2asmeasuredunderthese conditionsisfarfromidealforitsuseasaphotocathode.Toconﬁrm thatthelargetransientinphotocurrentuponswitchingthelighton oroffisduetolimitingcatalyticperformanceofthep-Cu2Oﬁlms

forreductionofwater,weinvestigatedwhetherreductionofH2O2

orS2O82(electronscavengers)wouldenhancethephotocurrent

[22,23].TheresultisshowninFig.2.BoththeuseofH2O2(red

curve)andS2O82(greencurve)resultedinasigniﬁcantincreasein

the photocurrent, as well as an increase in the dark current (Fig. 2a). Especially noteworthy is the observed stability in photocurrent for thesample with H2O2 added;after 27min of

illumination,thecurrentdensitiesasmeasuredunderillumination andinthedarkwereevenslightlyhigherasobservedafter10min ofillumination,whileslightlyincreasingtheirdifference. Further-more,thelargeandfastcathodicandanodicspikesasobservedin

Fig.1dwerenotobservedinaH2O2containingsolution,indicating

electrontransfertoH2O2isquiteefﬁcient.Theoriginoftheslow

transient in current response (to less negative values) after illumination was switched off is not exactly understood, but might be related to passivation of the electrochemically active surfacestatesoftheCu2Ocathode.TheXRDpatternofthissample

didnotshowsigniﬁcantdifferencesbeforeandafterexperiment, andonlyaminorchangeinsamplesurfacestructurewasobserved (redcurveinFig.2b,andFig.2c,respectively).Theenhancementin current densityuponilluminationusing H2O2 asscavengerwas

0.94mA/cm2_.

Fig.1.(a,b)Top-viewSEMimagesofaCu2Ofilm(a)beforeand(b)afterPECmeasurementin0.1MK2SO4(pH7)for30minatanappliedpotentialof0.3Vvs.RHE;(c)XRD patternoftheCu2Ofilmbefore(blackcurve)andafter(bluecurve)photocurrentmeasurement:DiffractionpeakscanbeassignedtoPt(JCPDS#88-2343),Cu2O(JCPDS# 05-0667)andCu(JCPDS#03-1018);(d)PECmeasurementoftheCu2Ofilmin0.1MK2SO4atanappliedpotentialof0.3Vvs.RHE:thelightwasturnedonandoffduringthePEC measurement(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.).

(4)

Fig.2. (a)PECmeasurementsofp-Cu2Ofilmsin0.1MK2SO4withandwithoutscavengers:(bluecurve)noscavenger,(redcurve)0.1MH2O2and(greencurve)0.1MNa2S2O8; the(magentacurve)wasmeasuredinanaqueoussolutioncontaining1.0MNa2SO4with0.1MK2HPO4atpH4.9:allmeasurementswereperformedatanappliedpotentialof 0.3Vvs.RHEandthelightwasturnedonandoffduringthemeasurements;(b)XRDpatternsoftheCu2Ofilms(blackcurve)beforeandafterphotocurrentmeasurements (samecolorindexasin(a)):DiffractionpeakswereassignedtoPt(JCPDS#88-2343),Cu2O(JCPDS#05-0667)andCu(JCPDS#03-1018);(c-e)top-viewSEMimagesofCu2O filmsafterPECmeasurementin0.1MK2SO4with(c)0.1MH2O2or(d)0.1MNa2S2O8atpH7,andin(e)1.0MNa2SO4with0.1MK2HPO4atpH4.9(Forinterpretationofthe referencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.).

Fig.3.(a)Top-viewand(b)cross-sectionSEMimagesofaCu2OfilmwithaRuO2toplayer;(c)PECmeasurementofCu2O-RuO2filmsin0.1MK2SO4withandwithout scavengers:(bluecurve)noscavenger,(redcurve)with0.1MH2O2and(greencurve)with0.1MNa2S2O8:allmeasurementswereperformedatanappliedpotentialof0.3Vvs. RHEandthelightwasturnedonandoffduringthemeasurementsinordertoobtainabetterinsightintothedifferencebetweenthephotocurrentandthedarkcurrent;(d-f) top-viewSEMimagesofCu2O-RuO2filmsafterPECmeasurementin0.1MK2SO4(d)withoutscavenger,(e)with0.1MH2O2and(f)with0.1MNa2S2O8(Forinterpretationof thereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.).

(5)

Using S2O82 as electron scavenger, the current increase

betweendarkandilluminatedoperationwashighandamounted to 1.34mA/cm2_. _However, _as _can _be _seen _from _Fig. ₂_a, _the

obtained current was not stable during the duration of the measurementasboththecurrentdensityunderilluminationand inthedarkincreasedsigniﬁcantlyovertime.Thisincreasecanbe relatedtoachangeinmorphologyandanincreaseinsurfacearea of the electrode as observed in Fig. 2d, although a minor contributionfrom Pt cross-over fromanode to cathodecannot beexcluded.AstheXRDpattern(greencurveinFig.2b)doesnot show a phase change, this might indicate that Cu2O

photo-electrodesslowlydissolvewhenexposedtoelectrolytescontaining S2O82. In addition, it should be noted that S2O82 reduction

(reactions(3)and(4))inducescurrentdoublingwhenusedasan electronscavenger: S2O28þe!SO 4 þ 1 2O2 ð3Þ SO4þH2O!SO24 þOH þHþaq; ð4Þ

where denotes a radical. Reaction (1) requires photon absorption (one photon per electron), while current doubling occursduetosubsequentreactionoftheformedOHradicalsto OH (accepting a second electron (not requiring an absorbed photon))[24].Therefore,theincreaseinphotocurrentmightnot necessarily scale with an increase in Cu2O induced photon

efﬁciency.

Itshouldbenotedthatthepossibilityofphotocurrentdoubling (duetotheformationofOHradicalsbyH2O2+e-!OH+OH)is

alsosuggestedfortheuseofH2O2[25–28],andthatespeciallyin

thecaseofGaAs,thecathodicreductionofH2O2wasassociated

withchemicaletchingoftheGaAssurface[26,27].Inthisrespect, wewouldliketostressthatinthecaseofp-Cu2Oelectrodes,we

observedtheoppositeeffect,namelyahigher photoelectrochem-icalstability oftheCu2OﬁlmsuponadditionofH2O2.Sincethe

extent of photocurrent doubling is strongly dependent on the semiconductorused,aswellasonitssurfaceorientationandthe appliedlightintensity,thecontributionofcurrent-doublingtothe enhancementinphotocurrentinducedbyH2O2cannotbeeasily

quantiﬁedandrequiresfurtherinvestigation.

Finally,theelectrolytethatwaspreviouslyusedbyParacchino et al. (1.0M Na2SO4 with 0.1MK2HPO4 at pH 4.9) was used

(magentacurveinFig.2a)[14].Asexpectedfromtheirpublication andthePourbaixdiagramofcopper,theobtainedphotocurrentin thissolutionquicklydiminishedastheCu2OlayerreducedtoCu.

Theobserveddifferencebetweenphotocurrentanddarkcurrent wasminimalafter10minofillumination.InstabilityofthisCu2O

ﬁlmwasalsoclearlyvisiblewhenanalysingthesamplesurface, whichturnedintonanoparticleswithadiameterof40nmwhile maintaining the overall cubic structure typical for Cu2O ﬁlms

(Fig.2e).Furthermore,anadditionalandsigniﬁcantCupeakwas foundintheXRDpatternofthis_ﬁlm(Fig.2b,magentacurve),in agreementwiththeblackcolour of thissample after measure-ment.

Additionally,alsotheuseofthewell-knownholescavengers methanolandSO32wasinvestigated,aswellastheuseofPO43at

pH 7, but these electrolytes did not show any difference in behaviourcomparedtotheelectrolytewithoutaddedscavenger (resultsnotshown).

Although the previous results indicated a lower rate of photoreduction of Cu2O under the conditions used here, the

additionalprotectiveeffectofRuO2depositeddirectlyontopofthe

Cu2Oﬁlmwasalsoevaluatedundertheseconditions.TheRuO2

layerconsistsofsmallamorphousnanoparticlesmergedtogether

into acontinuous ﬁlmfollowing theoverallmorphologyof the underlyingCu2Oﬁlm,aswasobservedbySEM(Fig.3aandb)and

thelackofadditionalXRDlines(shownbelowinFig.6g(black curve)).The PECchronoamperogramfor theCu2O-RuO2 sample

measured in an electrolyte without scavenger (blue curve in

Fig.3c)showsaverylowtransientcurrentresponseafterthelight sourcewasturnedonoroff.WhenH2O2orS2O82areaddedtothe

solution,anincreaseindarkcurrentandphotocurrentisapparent, as was also thecase for Cu2Osamples withoutRuO2 (Fig.2a).

Comparing the photocurrent values of the samples with and withoutRuO2,itcanbeobservedthatthephotocurrentdecreased

uponadditionoftheRuO2layer(-0.43mA/cm2vs.0.94mA/cm2

and 0.30mA/cm2 _vs. _1.34_mA/cm2 _for _H

2O2 and S2O82,

respectively),butthestabilityincreasedsigniﬁcantly;the perfor-manceisparticularlystablewhenH2O2ispresentintheelectrolyte

(Fig.3c).AscanbeseenfromFig.3e,thePECmeasurementwith addedH2O2didnotinﬂuencethemorphologyoftheCu2O-RuO2

sampleatall,whilethemorphologyofthesamplecontainingRuO2

alsoappearedstabilizedintheabsenceofascavenger(Fig.3d).On theotherhand,themorphologyofthesampleinthepresenceof S2O82changeddramatically(Fig.3f),whichindicatesthatS2O82

issocorrosive,thataRuO2protectionlayerbecomesineffective.

A linear sweepvoltammogram (LSV) in thedark and under illuminationprovidesinformationabouttheonsetpotentialofa photocathodeinaspeciﬁcelectrolyte,whichisameasureofthe requiredoverpotentialforH2evolutionandthereforeameasure

forthesuitabilityofthephotocathodeforwatersplitting.When connecting the photocathode to a photoanode in a photoelec-trochemical diode, a more positive onset potential is highly desirable[29].Fig.4showsaselectionofLSVcurvesmeasuredin lightanddarkconditionsondifferentsamples(Cu2OandCu2

O-RuO2) with or without the addition of H2O2 or S2O82 to the

electrolyte. Fromthis ﬁgure, it can beobserved that the onset potentialforaCu2Oﬁlmwithouttheuseofanelectronscavenger

(0.4Vvs.RHE)isveryclosetothepotentialusedfor chronoam-perometricmeasurements(0.3Vvs.RHE),whichexplainsthelow currentdensities observedforthissample.Boththeadditionof H2O2 and S2O82 provided an onset potentialof 0.8V vs. RHE,

explainingthemuch highercurrent densitiesinthepresenceof thesecompounds.WhenaCu2O-RuO2ﬁlmismeasuredwithout

Fig.4. LSVmeasurementsof:(darkbluecurves)Cu2Ofilmin0.1MK2SO4,(red curves)Cu2Ofilmin0.1MK2SO4and0.1MH2O2, (greencurves)Cu2Ofilmin 0.1MK2SO4and0.1MNa2S2O8,(lightbluecurves)Cu2O-RuO2filmin0.1MK2SO4, and (magentacurves)Cu2O-RuO2film in0.1MK2SO4and 0.1MH2O2;Forall samples,thesolidcurvewasmeasuredunderAM1.5Gillumination,andthedotted curvewasmeasuredinthedark(Forinterpretationofthereferencestocolourin thisfigurelegend,thereaderisreferredtothewebversionofthisarticle.).

(6)

scavenger,theonsetpotentialliesevenmorepositivethan1.1Vvs. RHE,whilethedifferencebetweenthedarkcurrentandthecurrent measuredunderilluminationissmall.WhenmeasuringaCu2

O-RuO2 sample with H2O2 added, the onset potential becomes

slightly more negative (1.0V vs. RHE), while the difference between the dark current and the current measured under illuminationismorepronouncedatpotentialsmorenegativethan 0.6Vvs.RHE.Atmorepositivepotentials,thedifferencebetween thecurrentmeasuredunderillumination andin thedarkstays ratherlowforthesemeasurementconditions.

The photoelectrochemical stability of a Cu2O-RuO2 ﬁlm was

further investigated during a long-term PEC measurement in whichH2O2wasaddedtotheelectrolyte(Fig.5).Fig.5ashowsthat

the photocurrentdecreases from 1.9mA/cm2 _to _1.0_mA/cm2

duringtheﬁrst7hoursofthemeasurement,butincreasedagain afteradditionoffreshH2O2.Therefore,thisdecreasein

photocur-rentcanbeascribedtotheconsumptionofH2O2.Fig.5bshowsthat

thisCu2O-RuO2sample was photoelectrochemicallystable even

aftermore than 11hoursof illumination, which underlinesthe importanceoffastelectronextractionfromthephotocathodefor improvedstability.

In order toanalysethecathodically inducedchemical trans-formations,weperformedPECmeasurementsinacellthatwas connectedtoagaschromatograph(GC)(Fig.6;seeFig.6gfora schematicrepresentationofthecell).InFig.6a,c,e,theresultsfrom thecombinedPECandGCmeasurementsareshown,inwhichthe amountsofH2andO2evolvedduringthemeasurementaredirectly

comparedtothemeasuredcurrentdensity.Fromthemeasurement withoutscavenger(Fig.6a),it canbeseen thatbothH2 andO2

evolvedfromthePECcelluponilluminationoftheCu2O-RuO2ﬁlm

whileapplyingabiasof0.3Vvs.RHE,buttheydidnotevolveina 2:1ratioaswouldbeexpectedfromstoichiometricwatersplitting: much more O2 than H2 evolved. We propose the following

reactionstakingplaceonaCu2O-RuO2cathodeandaPtanodein

0.1MK2SO4(pH7): Oxidation: H2Oþ2hþ!2Hþþ 1 2O2 E ¼1:299VRHE E_¼_0:816VRHE ðonPtCEÞ ð5Þ

Reduction: 2Hþþ2e!H2andE=0VRHEE=0.413VRHE (6)

(on Cu2O-RuO2) Cu2OþH2Oþ2e!2Cuþ2OH E=0.360

VRHEE=0.053VRHE (7)

ThetransientinH2evolution(peakinFig.6a)canbeexplained

bytherapidlyrisingoxygenconcentrationinthecell.Eventhough a continuous purge is applied,we suspect that the reactionof hydrogenwithoxygentoformwaterbecomessignificant(the‘back reaction’,presumablyoccurringoverthePtanode). Atthesame time, the high concentration of oxygen formed can only be explainedbyasacrificialreductionreaction,whichsuggeststhat despitetheRuO2filmasignificantfractionofCu2Oisincontact

withthe0.1MK2SO4electrolyte,reactingaccordingtoreaction(5).

The formationof Cuis indeed observedby XRD(bluecurvein

Fig.6g).ItshouldalsobenotedthattheO2evolutionstillcontinues

after the solar simulator was turned off. This can likely be explained by the formed Cu particles, decreasing the cathode resistance,andallowingelectrochemicalO2formation(anode)and

reductionofCu2O(cathode)tocontinueattheappliedpotential

(0.3V vs.RHE). Assoonasthe potentialwas releasedfromthe sample, the O2 formation rate dropped to zero within a few

minutes,followingthetransientbehaviourofthereactor. ForthesampleforwhichH2O2wasusedaselectronscavenger,

H2 and O2 were formed upon photoelectrochemical reaction

(Fig. 6c). During the measurement, it was observed that gas bubblesaccumulatedonthesamplesurfacebeforebeingreleased, which explains the sharp spikes in the GC signals and the continuouschangeinthemeasuredcurrentdensity.BothH2and

O2evolvedatalargerratethanforthesamplewithoutscavenger,

butagaindidnotevolveina2:1ratio;againmuchmoreO2thanH2

evolved.Contrarytothesamplewithoutscavenger,Cupeakswere absentintheXRDspectraforthissampleafterreaction(redcurve inFig.6g),andalsothestructureofthesamplesurfacestayedmore or less intact (Fig. 6d). This means that in this case the photoreductionofCu2OtoCuisnotthecompetingreactionwith

H2formation,butthereductionofH2O2toH2O(reaction(11)).This

conﬁrmsthatH2O2isindeedusedasanelectronscavengerduring

photoelectrochemical water splitting using a Cu2O-RuO2 ﬁlm.

SincenoCuisformedinthiscase,noO2wasformedafterthelight

source was turned off while the potential was still applied. Furthermore,Ptfromthesubstrateand/orthecounterelectrode canalso actas acatalyst forhydrogen peroxidedecomposition (reaction(12)),whichcontributestothenonstoichiometricH2and

O2formation.Especiallythemuchmorethan5timesincreasein

observedO2evolutioninthecaseofaddedH2O2ascomparedto

the measurement without added electron scavenger (compare

Fig.6a,cwithFig.3c)andtheobservedO2formationbefore(and

after) illumination, show the signiﬁcance of this catalytic disproportionation reaction.In summary, withtheuse of H2O2

asanelectronscavenger,thefollowingreactionstakeplace: Fig.5. (a)Long-termPECmeasurementofaCu2O-RuO2ﬁlmin0.1MK2SO4and

0.1MH2O2atanappliedpotentialof0.3Vvs.RHEandwithchoppedlight.0.1mL H2O2wasaddedafter7hourstobringitsconcentrationbackto0.1M;(b)top-view SEMimageoftheCu2O-RuO2ﬁlmafterlong-termPECmeasurementin0.1MK2SO4 and0.1MH2O2.

(7)

Fig.6. (a,c,e)CombinedPEC(at0.3Vvs.RHE)andGCmeasurementsofCu2O-RuO2filmsin(a)0.1MK2SO4withoutscavenger,(c)0.1MK2SO4with0.1MH2O2and(e)1M Na2SO4and0.1MK2HPO4adjustedtopH4.9.(b,d,f)Top-viewSEMimagesoftheCu2O-RuO2filmsaftercombinedPECandGCmeasurementsin(b)0.1MK2SO4,(d) 0.1MK2SO4and0.1MH2O2,and(f)1MNa2SO4and0.1MK2HPO4atpH4.9.(g)SchematicrepresentationofthePECcellusedforthecombinedPECandGCmeasurements showingtheidealcaseofstoichiometricwatersplitting;forsimplicity,theAg/AgClreferenceelectrodeusedisnotshown.(h)XRDmeasurementsoftheCu2O-RuO2films (blackcurve)beforeand(blue,red,magentacurves)aftercombinedPECandGCmeasurementsin(bluecurve)0.1MK2SO4,(redcurve)0.1MK2SO4and0.1MH2O2,and (magentacurve)1MNa2SO4and0.1MK2HPO4atpH4.9:DiffractionpeakswereassignedtoPt(JCPDS#88-2343),Cu2O(JCPDS#05-0667)andCu(JCPDS#03-1018)(For interpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.).

(8)

Oxidation: H2Oþ2hþ!2Hþþ12O2 and E=1.229VRHE E=

0.816VRHE (8)

(on Pt CE) H2O2þ2OHþ2hþ!O2þ2H2O, E=0.146 VRHE

E=0.297VRHE (9)

Reduction: 2Hþþ2e!H2andE=0VRHEE=0.413VRHE (10)

(on Cu2O-RuO2) H2O2þ2Hþþ2e!2H2O E=1.776VRHE

E=1.334VRHE (11)

Catalysis:(onPt) 2H2O2!O2þ2H2O. (12)

ForthesamplethatwasmeasuredinanelectrolyteatpH4.9,as waspreviouslyusedbyParacchinoetal.[14],H2formationwasnot

signi_ﬁcantandpredominantlyO2 wasformed(Fig.6e).

Further-more,weobservedthatthissamplelostitselectricalcontactafter approximately1hourofPECmeasurement.It wasattemptedto restartthePECmeasurement,but ascanbeobservedfromthe large switching of the current density around 2:15hours, the contactwiththesamplewasnotregained.Whenthesamplewas taken out of the PEC cell, it was observed that the film was detachedfromtheSi/Pt substrate,and SEM confirmed thatthe sample was reduced during the measurement (Fig. 6f), in agreement with the presence of a significant Cu peak in the XRDspectrumafterthemeasurement(magentacurveinFig.6g). Duringtheperiodthat thepotentialwas appliedtothesample withouthavingpropercontact,itwasobservedthatO2 formed,

andtheO2formationwentbackto0afterthepotentiostatwas

turnedoffwhilethelightwasstillturnedon.Thisdependenceof theO2formationontheavailabilityofanappliedpotentialtothe

sampleandnotontheavailabilityoflightcanbeascribedtothe formation of Cu onthis sample. In summary, with the use of

Na2SO4 and K2HPO4 as theelectrolyte at pH4.9, thefollowing

reactionstakeplaceonaCu2O-RuO2ﬁlm:

Oxidation H2Oþ2hþ!2Hþþ 1 2O2 E ¼1:229VRHE E¼0:816VRHE onptCEÞ ð ð13Þ

Reduction Cu2OþH2Oþ2e!2Cuþ2OH:

E¼0:360VRHE E¼0:053VRHE

ðonCu2ORuO2Þ

ð14Þ In summary, Fig. 7 shows the location of the valence and conduction band of Cu2O including the position of all redox

reactions(reactions (5)-(14)) as discussed above.The reactions indicatedinredandbluearethereactionsthattakeplaceinthe presenceandabsenceofH2O2intheelectrolyte,respectively,as

observedbytheGCmeasurementsshowninFig.6.Ascanbeseen inFig.7,forbothcases(withorwithoutH2O2)thesecoloredredox

reactionpairscorrespondtothereductionreactionwiththemost negative redox potential possible in combination with the oxidationreactionwiththemostpositiveredoxpotentialpossible. Furthermore,wecanobservethattheredoxreactions involving H2O2(inred)arethermodynamicallyabletotakeplace

spontane-ously without the presence of illuminated Cu2O, as was also

observedinFig.6c,althoughthepresenceofilluminatedCu2Odid

increasetheefﬁciencyofthesereactions.Theenergeticallymore favorableredoxreactionswithoutH2O2addedtotheelectrolyte(in

blue)doneedthepresenceofilluminatedCu2Oinordertotake

place,aswasalsoobservedinFig.6candFig.6e.

Theactivitydeterminationandevaluationofstructuralchanges conﬁrmthat the performanceand stability of (RuO2 protected)

Cu2Ocanbesigniﬁcantlyimprovedifefﬁcientelectrontransferis

induced,whichcouldbeachievedbyanefﬁcientprotonreduction catalyst.AsourdatashowthatRuO2isnotsufﬁcientlyeffectivefor

thispurpose[17,30],wealsoperformedpreliminaryexperiments withadditionalPtnanoparticles.Thesepreliminaryexperimentsin ourlaboratoryhaveshownthatdepositionofPtparticlesontopof Cu2OandCu2O-RuO2ﬁlms,withoutmodiﬁcationofthe

composi-tionandstructureofCu2Oisnottrivial,andphotodepositionofPt

particlesresultedinan insigniﬁcant currentdensityincrease as compared to that obtained in the presence of H2O2.

Non-destructivePtdepositionbyphysicalmethodssuchassputtering or e-beam evaporation is highly desired for obtaining more efﬁcientandstableCu2Ophotocathodes.Indeed,Grätzeland

co-workersalreadysolvedthisissuebyapplyingtheirchampion bi-layercoatingofAl-dopedZnOandTiO2viaatomiclayerdeposition

(ALD), afterwhich theywereable tosuccessfullydeposit Ptby either photodeposition or E-beam evaporation [14,15]. On the otherhand,wewouldliketoemphasizehere,thatinthisworkwe triedtoworkaroundthenecessityoftheseALDcoatings,which wouldmaketheﬁnalphotoelectrodesystemmoreeasilyscalable. Provided that Pt nanoparticles or another hydrogen evolution catalyst can be deposited with a non-destructive and scalable depositiontechniqueontopoftheCu2O-RuO2ﬁlms,our

experi-mentsshowthat it shouldindeedbepossibletoobtain photo-electrochemicallymorestablep-Cu2Ophotocathodeswithoutthe

needtogrowpassivationlayersviaALD.

Anothermethodforimprovementoftheefﬁciencyofp-Cu2Oas

aphotocathodeinphotoelectrochemicalwatersplittingwouldbe to use nanostructures like nanowires, nanowire networks or nanocubes [31]. Especially for Cu2O with a diffusion lengthof

minoritychargecarriers(electronsinthecaseofp-Cu2O)ofonly

20–100nm, and an absorption depth near the bandgap of approximately 10

m

m [16,32–34], the use of one-dimensional Fig.7.Energybanddiagramofp-Cu2Oﬁlmsincludingtheredoxpotentialsforall

reactionstakingplaceduringthephotoelectrochemicalexperimentsasdescribed above(Reactions(5)-(14)).ThereactionsinbluearetakingplacewithoutH2O2 addedtotheelectrolytesolution,thereactionsinredtakeplacewhenH2O2isadded asanelectronscavenger,andthereactionsinblackareeithernotconsideredoronly takingplaceatthestartofthemeasurementuntiltheCu2Oreductionreactiontakes over(Forinterpretationofthereferencestocolourinthisﬁgurelegend,thereaderis referredtothewebversionofthisarticle.).

(9)

nanostructures is likely to be highly beneﬁcial to increase photocurrentefﬁciencies[35].

4.Conclusions

Itwasfoundthatthephotoelectrochemicalstabilityofp-Cu2O

ﬁlms used for photoelectrochemical water splitting can be signiﬁcantly enhanced by the use of milder conditions (pH 7 and0.3Vvs.RHE)thantypicallyappliedintheliterature.Although aphotoelectrodepositedRuO2layerfurtherenhancedthestability,

themeasuredphotocurrentslightlydecreased.Signiﬁcant perfor-mance was obtained when H2O2 was added as an electron

scavengertotheelectrolyte:themeasuredphotocurrent(forH2O2

reduction) was increased and also the photoelectrochemical stability was that high, that the Cu2O-RuO2 ﬁlm could be

illuminatedformorethan 11hourswithoutnoticeable photore-duction to Cu. By combining gas chromatography with the performedphotoelectrochemicalmeasurements,itwasconﬁrmed thatH2O2servedasan electronscavengerduring

photoelectro-chemical water splitting using Cu2O-RuO2 ﬁlms: therefore, the

additionofasuitableH2evolutioncatalystishighlyrecommended

forobtainingmoreefﬁcientCu2Ophotocathodes.Unfortunately,Pt

nanoparticlespreparedbyphotodepositionappearedincapableof providinghighhydrogenproductionratesandstability.

Acknowledgements

FinancialsupportfromtheNetherlandsTechnologyFoundation (STW)isgratefullyacknowledged.KaiHan(UniversityofTwente) isgratefullyacknowledgedforhishelpwiththeGCmeasurements. AlexanderMilbrat (University ofTwente) isgratefully acknowl-edgedfor hishelp withtheSEM measurements and foruseful discussionsconcerningthedesignofaPECcellforcombinedPEC andGCmeasurements.

References

[1]A.Kudo,Z-schemephotocatalystsystemsforwatersplittingundervisiblelight irradiation,MRSBull36(2011)32–38.

[2]R.Van DeKrol, Y.Liang,J. Schoonman, Solarhydrogenproductionwith nanostructuredmetaloxides,J.Mater.Chem.18(2008)2311–2320. [3]H.Kato,M.Hori,R.Konta,Y.Shimodaira,A.Kudo,ConstructionofZ-scheme

typeheterogeneousphotocatalysissystemsforwatersplittingintoH2andO2 undervisiblelightirradiation,Chem.Lett.33(2004)1348–1349.

[4]A.Kudo,H.Kato,I.Tsuji,Strategiesforthedevelopmentofvisible-light-driven photocatalystsforwatersplitting,Chem.Lett.33(2004)1534–1539. [5]A.J.Nozik,Photochemicaldiodes,AppliedPhysicsLetters30(1977)567–569. [6]A.J.Nozik,Photoelectrochemistry:Applicationstosolarenergyconversion,

AnnualReviewofPhysicalChemistry29(1978)189–222.

[7]M.G.C.Zoontjes,PhDthesis:Visible-light-inducedwatersplittingonachip, UniversityofTwente,Enschede,2015.

[8]M.G.C.Zoontjes,M.Huijben,J.Baltrusaitis,W.G.VanDerWiel,G.Mul,Selective hydrothermalmethodtocreatepatternedandphotoelectrochemically effectivePt/WO3interfaces,ACSAppl.Mater.Interfaces5(2013)13050–13054.

[9]T.Kim,K.S.Choi,NanoporousBiVO4photoanodeswithdual-layeroxygen

evolutioncatalystsforsolarwatersplitting,Science343(2014)990–994. [10]C.G.Read, Y.Park, K.S. Choi,Electrochemicalsynthesisofp-type CuFeO2

electrodesforuseinaphotoelectrochemicalcell,JournalofPhysicalChemistry Letters3(2012)1872–1876.

[11]L. Rovelli, S.D. Tilley, K. Sivula, Optimization and stabilization of electrodepositedCu2ZnSnS4photocathodesforsolarwaterreduction,ACS

Appl.Mater.Interfaces5(2013)8018–8024.

[12]C.M.McShane,K.S.Choi,Junctionstudiesonelectrochemicallyfabricatedp-n Cu2Ohomojunctionsolarcellsforefﬁciencyenhancement,PhysicalChemistry

ChemicalPhysics14(2012)6112–6118.

[13]L.-k.Tsui,L.Wu,N.Swami,G.Zangari,Photoelectrochemicalperformanceof electrodepositedCu2OonTiO2nanotubes,ECSElectrochemistryLetters1

(2012)D15–D19.

[14]A.Paracchino,V.Laporte,K.Sivula,M.Grätzel,E.Thimsen,Highlyactiveoxide photocathodeforphotoelectrochemicalwaterreduction,NatureMaterials10 (2011)456–461.

[15]S.D.Tilley,M.Schreier,J.Azevedo,M.Steﬁk,M.Graetzel,Rutheniumoxide hydrogenevolutioncatalysisoncompositecuprousoxidewater-splitting photocathodes,Adv.Funct.Mater.24(2014)303–311.

[16]Z.Zhang,R.Dua,L.Zhang,H.Zhu,H.Zhang,P.Wang,Carbon-layer-protected cuprousoxidenanowirearraysforefﬁcientwaterreduction,ACSNano7 (2013)1709–1717.

[17]K.Maeda,N.Saito,L.Daling,Y.Inoue,K.Domen,Photocatalyticpropertiesof RuO2-Loaded(-Ge3N4foroverallwatersplitting,JournalofPhysicalChemistry

C111(2007)4749–4755.

[18]F.Amano,T.Ebina,B.Ohtani,Enhancementofphotocathodicstabilityofp-type copper(I)oxideelectrodesbysurfaceetchingtreatment,ThinSolidFilms550 (2014)340–346.

[19]K.L. Sowers, A. Fillinger, Crystal face dependenceof p-Cu2Ostability as

photocathode,JournaloftheElectrochemicalSociety156(2009)F80–F85. [20]B.Beverskog,I.Puigdomenech,Revisedpourbaixdiagramsforcopperat25to

300(C,JournaloftheElectrochemicalSociety144(1997)3476–3483. [21]L.M.Peter,Dynamicaspectsofsemiconductorphotoelectrochemistry,Chem.

Rev.90(1990)753–769.

[22]A.N.Ökte,M.S.Resat,Y.Inel,Inﬂuenceofhydrogenperoxideandmethanolon thephotocatalyticdegradationof1,3-dihydroxybenzene,Toxicol,Environ. Chem.79(2001)171–178.

[23]H. Dotan,K. Sivula, M.Grätzel, A. Rothschild,S.C. Warren, Probing the photoelectrochemicalpropertiesofhematite(a-Fe2O3)electrodesusing

hydrogenperoxideasaholescavenger,EnergyEnviron.Sci.4(2011)958–964. [24]J. Schneider, D.W. Bahnemann, Undesired role of sacriﬁcial reagents in

photocatalysis,JournalofPhysicalChemistryLetters4(2013)3479–3483. [25]R.Memming,MechanismoftheElectrochemicalReductionofPersulfatesand

HydrogenPeroxide,JournalofTheElectrochemicalSociety116(1969)785– 790.

[26]B.P. Minks, G. Oskam, D. Vanmaekelbergh, J.J. Kelly, Current-doubling, chemicaletchingandthemechanismoftwo-electronreductionreactions atGaAs,JournalofElectroanalyticalChemistryandInterfacial

Electrochemistry273(1989)119–131.

[27]B.P.Minks,D.Vanmaekelbergh,J.J.Kelly,Current-doubling,chemicaletching andthemechanismoftwo-electronreductionreactionsatGaAs,Journalof ElectroanalyticalChemistryandInterfacialElectrochemistry273(1989)133– 145.

[28]J.E.A.M.vandenMeerakker,Thereductionofhydrogenperoxideatsiliconin weakalkalinesolutions,ElectrochimicaActa35(1990)1267–1272. [29]M.G.Walter,E.L.Warren,J.R.McKone,S.W.Boettcher,Q.Mi,E.A.Santori,N.S.

Lewis,Solarwatersplittingcells,Chem.Rev.110(2010)6446–6473. [30]M.Tian,W.Shangguan,J. Yuan,S.Wang,Z.Ouyang,Promotioneffectof

nanosizedPt,RuO2andNiOxloadingonvisiblelight-drivenphotocatalystsK 4Ce2M10O30(M=TaNb)forhydrogenevolutionfromwaterdecomposition,Sci.

Technol.Adv.Mater.8(2007)82–88.

[31]A.W.Maijenburg,A.N.Hattori,M.DeRespinis,C.M.McShane,K.S.Choi,B.Dam, H.Tanaka,J.E.TenElshof,Niandp-Cu2Onanocubeswithasmallsize

distributionbytemplatedelectrodepositionandtheircharacterizationby photocurrentmeasurement,ACSAppl.Mater.Interfaces5(2013)10938– 10945.

[32]P.E. De Jongh, D. Vanmaekelbergh, J.J. Kelly, Photoelectrochemistry of electrodepositedCu2O,JournaloftheElectrochemicalSociety147(2000)

486–489.

[33]C.J.Engel,T.A.Polson,J.R.Spado,J.M.Bell,A.Fillinger,Photoelectrochemistryof porousp-Cu2Oﬁlms,JournaloftheElectrochemicalSociety155(2008)F37–

F42.

[34]S.Sunkara,V.K.Vendra,J.H.Kim,T.Druffel,M.K.Sunkara,Scalablesynthesis andphotoelectrochemicalpropertiesofcopperoxidenanowirearraysand ﬁlms,CatalysisToday199(2013)27–35.

[35]J.Luo,L.Steier,M.-K.Son,M.Schreier,M.T.Mayer,M.Grätzel,Cu2ONanowire

PhotocathodesforEfﬁcientandDurableSolarWaterSplitting,NanoLetters16 (2016)1848–1857.