Influence of the metal center of metalloprotoporphyrins on the electrocatalytic CO2 reduction to formic acid

ContentslistsavailableatScienceDirect

Catalysis Today

jo u rn al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / c a t t o d

Influence of the metal center of metalloprotoporphyrins on the electrocatalytic CO 2 reduction to formic acid

Yuvraj Y. Birdja

, Jing Shen

, Marc T.M. Koper

aLeidenInstituteofChemistry,LeidenUniversity,POBox9502,2300RALeiden,TheNetherlands

bChemistryandChemicalEngineeringDepartment,HunanInstituteofEngineering,Xiangtan,China

a r t i c l e i n f o

Articlehistory:

Received22August2016

Receivedinrevisedform23January2017 Accepted27February2017

Availableonline23March2017

Keywords:

Carbondioxidereduction Hydrogenevolution Formicacidformation Immobilizedmolecularcatalysts Metalloporphyrins

a b s t r a c t

Electrocatalyticconversionofcarbondioxidehasgainedmuchinterestforthesynthesisofvalue-added chemicalsandsolarfuels.Importantissuessuchashighoverpotentialsandcompetitionofhydrogen evolutionstillneedtobeovercomefordeeperinsightintothereactionmechanisminordertosteerthe selectivitytowardsspecificproducts.Hereinwereportonseveralmetalloprotoporphyrinsimmobilized onapyrolyticgraphiteelectrodefortheselectivereductionofcarbondioxidetoformicacid.Noformic acidisdetectedonCr-,Mn-,Co-andFe-protoporphyrinsinperchloricacidofpH3,whileNi-,Pd-,Cu- andGa-protoporphyrinsshowonlyalittleformicacid.Rh,InandSnmetalcentersproducesignificant amountsofformicacid.However,thefaradaicefficiencyvariesfrom1%to70%dependingonthemetal center,thepHoftheelectrolyteandtheappliedpotential.Thedifferentiationofthefaradaicefficiency forformicacidonthesemetalloprotoporphyrinsisstronglyrelatedtotheactivityoftheporphyrinfor thehydrogenevolutionreaction.CO2reductiononRh-protoporphyrinisshowntobecoupledstronglyto thehydrogenevolutionreaction,whilstonSn-andIn-protoporphyrinsuchstrongcouplingbetweenthe tworeactionsisabsent.TheactivityforthehydrogenevolutionincreasesintheorderIn<Sn<Rhmetal centers,leadingtofaradaicefficiencyforformicacidincreasingintheorderRh<Sn<Inmetalcenters.

In-protoporphyrinisthemoststableandshowsahighfaradaicefficiencyofca.70%,atapHof9.6and apotentialof−1.9VvsRHE.Experimentsinbicarbonateelectrolytewereperformedinanattemptto qualitativelystudytheroleofbicarbonateinformicacidformation.

©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).

1. Introduction

Inthepastfewdecadestheconsequencesofanthropogeniccar- bondioxideaccumulationintheatmospherehavebeenaddressed repeatedly.Itisnowadaysgenerallyacceptedthattheincreasing carbondioxidelevels intheatmosphere poseseriousproblems ifnoactionistaken[1].TheemissionofCO₂hasincreasedsince theindustrialrevolution,leadingtoanincreasedamountofCO2in theatmosphere[2].AccumulationofatmosphericCO₂leadstothe greenhouseeffectwhichcontributestoglobalwarmingandclimate change.Anotherimportantsustainabilityissueisthedepletionof fossilfuelsourceswhichiscausedbyanincreasingworldpopu- lationandachanginglifestyle,resultinginanincreasingenergy demand.Mankindisstillstronglydependentonfossilfuelsforits energyconsumption.Adrawbackoftheuseoffossilfuelsisthe

∗ Correspondingauthor.

E-mailaddress:m.koper@chem.leidenuniv.nl(M.T.M.Koper).

productionofCO2aftercombustionwhichisoftensimplyreleased in the atmosphere. In the past couple of years much research hasbeendonetomitigateCO₂accumulation[3–6]andsearchfor renewableenergysources[7–9].ApromisingwaytoutilizeCO2

isbyelectrochemicalCO2 conversion.Comparedtoothermeth- odstheadvantageofelectrochemicalconversion,whenoperational onanindustrialscale,istwo-fold:itreducestheCO₂emissionsin theatmosphereononehand,anditproducesrenewablefuelsand commoditychemicalsontheother.Moreover,additionaladvan- tagesarethefactthattheprocesscanbecarriedoutatambient conditions,watercanbeusedashydrogensourceand,iftheelec- tricityusedisproducedfromrenewablesources,onecancontribute toacompletelysustainablecarboncycle.However,therearestill significanthurdleswhich shouldbeovercomeorcircumvented, suchasthecompetitionofthehydrogenevolutionreaction,high overpotentialsforCO₂reduction,poorsolubilityofCO₂inaqueous mediaandthepoorselectivityforspecificfuels.Forimplementa- tioninthecurrentinfrastructure,liquidfuelsaremoreconvenient thangaseousfuels.Thereforemuchworkhasbeenfocusedonthe http://dx.doi.org/10.1016/j.cattod.2017.02.046

0920-5861/©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).

selectiveproductionofmethanol,ethanolandformicacid.These fuelscanbeemployeddirectlyinfuelcellstoproduceelectricity (e.g.DirectMethanolFuelCellandDirectFormicAcidFuelCell).

ElectroreductionofCO2toformicacidislesscomplexandthere- foreeasiertooptimizecomparedtoalcohols,asonlytwoelectrons needtobetransferred.

Molecularcatalystshave been gaining more attentionlately as they are relatively inexpensive and more abundantly avail- ablecomparedto(noble)metalcatalysts.Theyusuallyshowhigh activitiesandgood selectivitiesforvarious reactionswhich can betunedbymodifyingthecatalystwithadditionalligands,using electron-donating or electron-withdrawing groups [10–13]. In homogeneouselectrocatalysis,thesecatalystsareoftendissolved innon-aqueous solvents,astheyare poorlysoluble inaqueous electrolytes.Furthermore,largeamountsofcatalystareneededto dissolveintheelectrolyte.CO2canbindtoametalcenterwhich resultsinitsactivation[14]bya weakenedC–Ointeractiondue totransferofelectrondensity.Aftercoordinationtoametalcen- ter,reactionscantakeplacewhichwereinitiallynotpossiblefor freeCO2.Manydifferentmolecularcatalystshavebeenreported forCO₂ reductionsuchasporphyrins,phthalocyanines,cyclams, phosphinesandpolypyridines[15,16].

Immobilizationofmolecularcatalystsshouldinprinciplecom- binethebestofbothworlds:molecularcatalysisonheterogeneous surfaces[17,18].Asystemwithhighefficiencyandselectivitycan be createdwhereby the catalyst structure can be modified by additionofspecificligandse.g.tochangethecatalystelectronic propertieswhichisveryusefulformechanisticstudiesorcontrol- lingtheselectivity.Furthermore,onlyasmallamountofthecatalyst isrequiredandevencatalystsinsolubleincertainsolventscanstill beusedwhenimmobilized onasurface.Deactivation processes oftenencounteredinhomogeneoussystems,suchasdimerization andaggregationofthecatalyst,canbecircumvented.Inthelit- eraturemanydifferent systemsbased onimmobilizedcatalysts havebeendescribed:covalentattachmentofthecatalystbyusing e.g.aryldiazoniumsalts,4-aminopyridine[19–21],non-covalent attachmentofthecatalyst(drop-casting)[22,23]anddispersionof thecatalystinpolymerfilms[24–27].

In this work we will focus on one type of molecularcata- lysts,namelymetalloprotoporphyrins(MPPs)whichareasubgroup of the more general metalloporphyrins. Metalloporphyrins are complexesthatconsistofametalcenterwithinaheterocyclicmac- rocyclecomposedoffourpyrrolegroupsattachedtoeachother withmethinebridges.Protoporphyrinshavetwovinyl,twopro- pionicacidandfourmethylgroupsattachedtotheporphyrinring asshowninFig.1.Metalloprotoporphyrinsareanimportantpre- cursorforessentialmoleculesinbiologysuchastheheme-group inourredbloodcellsandchlorophyllsinplants(protoporphyrins withrespectivelyanFe²⁺andMg²⁺metalcenters).Furthermore ithasbeenshownthat (immobilized)metallo(proto)porphyrins aregoodcatalystsfore.g.theoxygenreductionreaction,hydrogen evolutionreaction,carbondioxidereductionandnitratereduction [28,10,16,29].SavéantandRobertandcoworkershaveconducted extensiveresearchonmolecularcatalystsforCO₂ reductionand howtoinfluencetheselectivity[30,31].Theyrecentlyshowedthat COorHCOOHisproducedbychangingthemetalcenterofthecom- plex[32].Theselectivityissueisimportantfroma fundamental pointofviewandthereforealsothesubjectoftheoreticalstudies [33,34],inwhichithasbeenshownthatadifferentbindingmode ofCO₂tothemetalcenterleadstoeitherCOorHCOOH.

Inthispaper,theselectivitytowardsformicacidisinvestigated experimentallywheretheroleofthemetalcenterisscrutinized bystudyingthepH effect,theconcomitanthydrogenevolution and thenature of the electroactivespecies (i.e., CO₂ or HCO⁻₃) on different metalloprotoporphyrins immobilized on pyrolytic graphite.

Fig.1. Chemicalstructuremetalloprotoporphyrins.

2. Experimental

The electrochemical experiments were performedin a one- compartment three-electrode cell at room temperature and ambientpressure.Allglasswarewasfirstcleanedbyboilingina 1:1mixtureofconcentratedsulfuricandnitricacid.Beforeeach experimenttheglasswarewasboiledthreetimesinultrapurewater (MilliporeMilliQgradientA10system,18.2Mcm).Thelong-term electrolysisexperimentswerecarriedoutinatwo-compartment three-electrode cell(H-type cell), where theworking electrode (WE)andcounterelectrode(CE)compartmentswereseparatedby anafionmembrane(Nafion115).

TheWEwasapyrolyticgraphite(PG)discwithadiameterof 5mmor 10mm usedina hanging meniscusconfiguration.The largePGelectrodes wereusedfor HPLC measurementstogen- erate largeramounts ofproducts.The reportedcurrent density isnormalized bythegeometricsurface areaoftheWE.A plat- inumgauzeandareversiblehydrogenelectrode(RHE)inthesame electrolytewereusedasCEandreferenceelectrode(RE),respec- tively.Unlessmentionedotherwise, thepotentials inthispaper arereferredtothisreferenceelectrode.Priortoeachexperiment, theWEwaspolishedwithsandpaper(firstP600andthenP1000) andultrasonicatedforapproximately2–3mininwater.Blankcyclic voltammogramswererecordedatascanrateof500mVs⁻¹until astablevoltammogramwasobtained(typicallyaround50cycles) toensureacleanPGsurface.Afterimmobilizationofporphyrinsa blankcyclicvoltammogramwasrecordedagaininordertoqualita- tivelyverifytheimmobilizationoftheporphyrin(asshowninFig.

S.1intheSI).Metalloprotoporphyrins(MPPs)wereimmobilizedon PGbydropcastingfroma0.01MboratesolutionofpH10inwhich theporphyrinwasdissolvedtoaconcentrationof0.5mM[35].

Electrolyte solutions were prepared with ultrapure water.

Pyrolyticgraphiteworkingelectrodeswerecutin-housefroma highpurity pyrolytic graphite plate(PY001009, GraphiteStore, USA).Highpuritychemicalswereused:perchloricacidandphos- phoricacid(MerckSuprapur),potassium phosphatemono- and dibasic (SigmaAldrich, TraceSelect), potassium phosphate trib- asic(SigmaAldrich,Reagentgrade),sodiumperchlorate (Sigma Aldrich,ACSreagent)andpotassiumbicarbonate(SigmaAldrich, tracemetalsbasis).AlltheporphyrinswerepurchasedfromFron- tierScientificandusedwithoutfurtherpurification.Beforeeach experimentthecellwasdeaeratedwithargon(Linde,Argon6.0 Scientific).ForCO2reduction,thecellwassaturatedwithCO2by

Fig.2. BulkpHoftheusedelectrolytesbeforeandafterCO2saturation.

purgingthecellwithCO2(Linde,Carbondioxide4.5)foratleast 20min.Hydrogen6.0Scientific(Linde)wasusedfortheRHEref- erence electrode. A -AutolabType III(MetrohmAutolab B.V.) wasusedfor voltammetricexperiments withsamplecollection andanIviumStatorCompactStat(IviumTechnologies)wasused toperformchronoamperometryandelectrochemicalimpedance spectroscopy.Experimentswerecarriedoutinunbufferedperchlo- ricacidofpH1and3andinavarietyofdifferentphosphatebuffer solutionsinthepHrangeof3–12(seeTableS.1intheSuppor- tingInformation).AlthoughtheactualpHwillbelowerwhenthe electrolyteissaturatedwithCO2,thedifferentelectrolyteswillbe referredtobytheirpHasmeasuredpriortoCO₂saturation.Fig.2 showstherelationbetweenthepHbeforeandafterCO₂saturation indicatingthatthepHafterCO2 saturationisapproximatelythe same(≈6.6)foralmostallneutralandalkalinephosphatebuffers.

Aswewillseelater,therearedifferencesbetweentheseinitially neutralandalkalineelectrolytesintermsoffaradaicefficiencies and product distribution. Therefore for a better discrimination betweenthedifferentelectrolyteswerefer tothem bytheini- tialbulkpH.ForcorrectmeasurementsversustheRHEscale,the LuggincapillaryandtheRHEcompartmentwerealsofilledwith CO₂ saturated electrolyte. Furthermore,under reduction condi- tions,especially whenH₂ isevolved,thepHneartheelectrode (“localpH”)maydifferfromthepHinthebulk.Thismayaffectthe onsetpotentialforproductformationaswellasproductselectiv- ity.Asquitehighcurrentswereobtained,especiallyattheø10mm PGelectrode,ohmicdropcouldnotbeneglected.Hence,thesolu- tionresistancewasdeterminedbypotentiostaticelectrochemical impedancespectroscopy.Theobtainedvaluewasusedtocorrect thevoltammogramsaftertheelectrochemicaldatawascollected.

Forchronoamperometry(electrolysisexperiments)thepotentio- stat’sIRcompensationfunctionwasusedtocompensateforohmic dropduringmeasurement.DetailscanbefoundintheSupporting InformationSection2.

OnLineElectrochemicalMassSpectrometry(OLEMS)wasuti- lizedforthedetectionofvolatilereactionproducts.Atip,which is placedclose (≈10␮m)totheelectrode surface, continuously collects volatilereaction productsfrom theelectrode interface.

Thetiphasadiameterof0.5mmandconsistsofaporousteflon cylinder(averageporesize10–14␮m)inaKel-Fholderandiscon- nectedtoanEvoLutionmassspectrometer(EuropeanSpectrometry SystemsLtd.)byaPEEKcapillary[36].AQuadrupoleMassSpec- trometerPrismaQMS200(Pfeiffer)isbroughttovacuumwitha TMH-071Pturbomolecularpump(60ls⁻¹,Pfeiffer)andaDuo2.5 rotaryvanepump(2.5m³h⁻¹,Pfeiffer).Priortoexperiments,the tipwascleanedin0.2MK2Cr2O7 in2MH2SO4 andrinsedwith

ultrapurewater.ASEMvoltagebetween1200Vand2400Vwas usedforthedifferentmassfragments.Allmassfragmentsshoweda decayduringmeasurementwhichistheresultofslowequilibration ofthepressureinthesystem.Thiswascorrectedforbysubtracting adoubleexponentialfittodatapointswherenochangeinactivity isobservedfromthewholedataset.Themassfragmentsshownin thispaperareallbackgroundcorrectedinthismanner.

OnlineHighPerformanceLiquidChromatography(onlineHPLC) isthetechniqueemployedtoanalyzenon-volatilereactionprod- ucts.AsimilartiptotheoneforOLEMS,however,withoutaporous tefloncylinder,isplacedneartheelectrodesurface[37].Samples witha volumeof60␮lwerecollectedwitha fractioncollector (FRC-10A,Shimadzu)atarateof60␮lmin⁻¹(LC-20ATpump,Shi- madzu).Sincethescanrateofthepotentialsweepduringsample collectionwas1mVs⁻¹,eachsampleheldtheaveragedconcentra- tionofa60mVpotentialdifference.Thesampleswereanalyzed aftervoltammetrywithHPLC(ProminenceHPLC,Shimadzu).The sampleswereplacedinanauto-sampler(SIL-20A)whichinjects 20␮lofthesampleintothecolumn.AnAminexHPX87-H(Bio- Rad)columnwithaMicro-GuardCationHCartridge(Bio-Rad)in frontwereused.Theeluentwas5mMH2SO4andtheeluentflow rate0.6mlmin⁻¹.Thecolumnand therefractiveindex detector (RID-10A)weremaintainedatatemperatureof35^◦C.

The reported results were all reproduced at least twice andallelectrochemicalmeasurements(CV, OLEMS,onlineHPLC andChronoamperometry)showedqualitativelythesameresults (HCOOH trend,current densities,onset potentials, etc.).Froma quantitativepointofview,theresultssometimescouldbeslightly different,which is ascribedtoa differentPGsurface each time afterpolishingthePGelectrodeascanbeobservedfromtheblank voltammogramsrecordedpriortotheexperiments.Thisisalsothe reasonthatforthestandarderroranalysisintheHPLCresults,only theconcentrationanalysisofliquidphase istakenintoaccount (concentrationsareoftenlowandthebaselinenoisy,whichleads todifferentpeakareasfordifferentexperiments).

3. Resultsanddiscussion

3.1. ActivityofmetalloprotoporphyrinsinperchloricacidpH3

Foraproperattributionoftheeffectofthemetalcenteronthe activityorselectivityfor theelectrocatalyticCO₂ reduction, the influenceofthebaresubstrate(PG)andtheporphyrinmacrocycle shouldfirstbeknown.Therefore,CO₂reductiononpristinePGand onmetal-lessorfreebasePPimmobilizedonPG(denotedasPP- PG)wasinvestigated.WithonlineHPLCduringvoltammetryitwas confirmedthatnoformicacidorotherliquidproductsareproduced atanypotentialonpristinePGnoronPP-PG.On-LineElectrochem- icalMassSpectrometryshowedthattheonlyproductfromCO2

reductiononPGandPP-PGisH2 (seeFig.S.2intheSupporting Informationandalsoreference[38]).

Someoftheinvestigatedprotoporphyrinswithcertainmetal centersproducenoformicacidoramountslowerthanthedetec- tionlimit.TheseprotoporphyrinshaveaCr,Mn,CoorFecenter.

WithOLEMS,nogaseousproductsotherthanH₂areobservedon MnPP,InPP,CrPP,SnPPandGaPP(Fig.S.4).OnFePP,RhPPandCuPP smallamountsofCH₄aredetectedandonNiPPsmallamountsof COandCH₄aredetectedasshowninFig.S.5.CoPPwasrecently showntoproduceCOatpH3withhighfaradaicefficiencybutno formicacidwhichisinagreementwithourresults.Moreover,a loweramountofCOwasformedatpH1togetherwithmethane andasmallamountofformicacid[38].Comparingthecurvesin Fig.3,it canbeseenthat PP-PGhasalowercurrent compared tothepristinePG.TheotherMPPsshowhighercurrents,indicat- ingthatthemetalloprotoporphyrinsareactiveforthehydrogen

Fig.3. LinearsweepvoltammetryinCO2saturated0.001MHClO4+0.099MNaClO4. Scanrate:1mVs⁻¹.

evolutionreaction(HER)orCO2reduction,whilePPpresumably blocksactivesitesforHERandCO₂reductiononPG.Thisconfirms thatthemetalcenterisofimportanceforcatalysis.Thefluctuations intheCVatverynegativepotentialsaretheresultofH2 bubble formationandtheirdetachmentfromtheelectrodesurface.The currentspikesappearskewedinsomeofthelatervoltammograms becauseofthepost-measurementOhmicdropcorrection.

InadditiontotheMPPswhich arenotactiveforformicacid productionfromCO₂,thereisasetofMPPswhichproducetrace amountsofformicacid,asshowninFig.4a.ThesearetheMPPswith Ni,Ga,PdorCumetalcenters.Theresultshavebeenreproducedand thestandarderrorsintheconcentrationareshowninthegraphsas well.OLEMSresultsagaindonotshowsignificantamountsofother CO2reductionproductsbesidesH2.

ThemostinterestingMPPsforthisstudyareSnPP,InPPandRhPP astheseareabletoproducesignificantamountsofformicacidfrom CO2asseeninFig.4b.AswiththeotherMPPstherearenosignif- icantamountsofreactionproductsotherthanH₂ observedwith OLEMS.Alloftheseprotoporphyrinsshowthesametrendinformic acidconcentrationduringLinearSweepVoltammetry,withSnPP andRhPPhavingaslightlylessnegativeonsetpotentialforHCOOH formationthanInPP.InandSnmetalcenterswerealsoidentifiedto beactiveforCO2reductiontoformicacidonphtalocyanineswhich aresimilarmolecularcatalyststoporphyrins[39].Moreinterest- ingisthefactthatRhPPproducessignificantamountsofformic

Fig.5. CO2reductiononPGinRhPPcontainingelectrolytesandimmobilizedRhPP onPG.Electrolytesolution0.001MHClO4+0.099MNaClO4.Scanrate:1mVs⁻¹.

acidfromCO₂.IthasbeenreportedthatInandSnmetalelectrodes mainlyproduceHCOOH,whileRhmetalmostlyformsH2[40].Rh metalonlyshowsactivityforCO₂reductionatelevatedpressures [41].Rhcomplexeshavebeenshowntobeactiveforhydrogena- tionofCO2 toformicacid,albeitthattheyoftendonotoperate inaqueousmediawithoutspecificligands[42–44].Muchresearch hasbeendoneonCO₂hydrogenationtoformicacid,butelectrocat- alyticreductionofCO2toformicacidonRhporphyrinsorsimilarRh molecularcatalystshasnotbeeninvestigatedindepth.Olderwork existsonRhcomplexesdissolvedinnon-aqueousmedia[45,46]for whichformicacidproductionwasalsoobserved.

InFig.5,acomparisonismadebetweenimmobilizedRhPPon PGanddifferentamountsofRhPPintheelectrolytewithapristine PGelectrode.ThecurrentdensitiesandthetrendsinHCOOHcon- centrationaresimilar.However,atacertainpotentialtheformic acidconcentrationreachesamaximumandstartstodecreasefor RhPPinsolution,whiletheRhPP-PGelectrodecontinuestopro- duceformicacid.ThemaximumforRhPP-PGishigherandatmore negativepotentialscomparedtoRhPPinsolution.Anotherobser- vationisthatalargerconcentrationofRhPPinsolutionleadsto aloweramountofformic acidproduced. Thiscanbeexplained byinhibition of activesiteswithmore RhPPmolecules present in solution. These results show thesuperiority of immobilized

Fig.4.Formicacidproductiononmetalloprotoporphyrinsin0.001MHClO4+0.099MNaClO4.Scanrate:1mVs⁻¹.

Fig.6.CO2reductiononformicacidproducingMPPsin(a)0.1MHClO4and(b)phosphatebufferofpH6.8.Scanrate:1mVs⁻¹.

metalloprotoporphyrinswithrespecttometalloprotoporphyrins dissolvedintheaqueouselectrolyte.

3.2. Activityofmetalloprotoporphyrinsinotherelectrolytes

TheformicacidproducingMPPsfromthepreviousparagraph werestudiedfortheirpHdependenceisinvestigatedinorderto obtainmoremechanisticinsight.Theaimistodeducetheoriginof theactivityofRhPPforCO₂reductiontoformicacidandidentify possibledifferencesbetweentheMPPsaswellastoidentifythe optimalconditionsforformicacidformation.

Inthe0.1MHClO₄electrolyte,alloftheseMPPsproducesignif- icantlylessformicacidthaninpH=3HClO₄electrolyteasshown inFig.6a.Interestingly,adifferencebetweenthethreeformicacid producingMPPsisobserved.InpH1InPPandRhPPstillproduce someHCOOHwhileSnPPonlyshowsnegligibleamountsofHCOOH.

ThecurrentgeneratedbytheInPPismuchsmallercomparedto thatoftheotherporphyrins.Thisisascribedtoa loweractivity forhydrogenevolution,whichalsoinfluencesthefaradaicefficien- ciesaswillbediscussedinafollowingsection.Moreover,theonset potentialsofthecurrentprofilesoftheMPPsarequitedifferent, inthatRhPPhastheleastnegativeonsetpotentialandInPPthe mostnegativeonsetpotential.Thisisprobablyassociatedwitha differenceinactivityfortheHERontheseMPPs.InpH6.8(0.1M

phosphatebuffer)comparableamountsofformicacidareformed asinpH3,asillustratedinFig.6b.

Differentphosphatebuffersolutionswereusedforthestudyin electrolyteswithapHrangefrom3till12.Eventhoughthebulk pHwillbelowerwhentheelectrolyteissaturatedwithCO₂,the differentelectrolyteswillbereferredtobytheirpHasmeasured priortoCO₂saturation.Theformicacidproductionduringtheneg- ativepotentialsweeponthethreeHCOOH-producingMPPsinthe differentelectrolytesisshowninFigs.7a–9a.Thecurrentprofilesof RhPPshowaplateaubetween−0.9Vand−1.5V(seeinset)while InPPandSnPPshownosuchplateaucurrent(thisisalsovisible inFigs.4b and6b).OnRhPPtheformic acidconcentrationpro- fileshowsamaximumwhichappearstobeatpotentialswithin thisplateauregion.Moreover,theHCOOHconcentrationprofiles forRhPPareslightlyshiftedtowardspositivepotentialsforhigher pHwhereasthoseforInPPandSnPPdonotshowsuchapotential shift.Thisisbettervisiblewhentheconcentrationprofilesarenor- malizedbytheirmaximumconcentration(asshowninFig.S.6in theSI).InthesameFigs.7b–9bthehydrogenevolutioncurrenton thethreeMPPsisshowninthesameelectrolytes.Inthesevoltam- mogramsnocurrentplateauregionisobservedasforCO₂reduction onRhPP,implyingthatthisplateaufeatureisspecificallyrelatedto CO₂reduction.Moreover,thereisacleardistinctionbetweenthe currentprofilesinthedifferentphosphatebufferelectrolytesforall

Fig.7.CO2reduction(a)andhydrogenevolution(b)onRhPPinelectrolyteswithdifferentpHs.

Fig.8. CO2reduction(a)andhydrogenevolution(b)onInPPinelectrolyteswithdifferentpHs.

Fig.9. CO2reduction(a)andhydrogenevolution(b)onSnPPinelectrolyteswithdifferentpHs.

MPPs.Theonsetofhydrogenevolutionseemsstronglyrelatedto thepHofthephosphatebuffers.AthighpH(11.6)thereislessneg- ativeonsetofH₂ evolutionandveryhighcurrents,whilelowpH (4and5.8)showsamorenegativeonsetpotentialforHERandthe currentsarelow.Ingeneral,thecurrentsaremuchhigherforRhPP comparedtoInPPandSnPP,indicatingthebetteractivityofRhPPfor theHER.ComparingthecurrentprofilesforCO₂reductionandHER onallthreeMPPs,theHERissomewhatinhibitedbyCO2reduction asthereductiononsetisdelayedtomorenegativepotentialand thecurrentsaremuchlower.ForRhPPtheonsetoftheHERisclose to/withintheplateauregion.ThemaximumHCOOHproduction alsolieswithinthispotentialrangeandshiftstopositivepoten- tialswithhigherpH.AsthemaximumofHCOOHproductioncan beinterpretedtobetheresultofthecompetitionbetweenhydro- genevolutionandCO2 reduction,itseemsplausibletoassociate thecurrentplateauwiththiscompetition.Thefactthatthecurrent plateauisnotobservedforInPPandSnPPcouldbeduetothedecou- plingorweakcouplingbetweenCO2reductionandHERonthese materials.TheonsetpotentialoftheHERismorenegativeforSnPP andInPPcomparedtoRhPP.ThecompetitionofCO₂reductionand HERisimportantforhighfaradaicefficienciestowardsHCOOHas willbediscussedinalatersection.

To investigate thepH effect more quantitatively, we define the onset potential for HCOOH and H2. The onset potential is

determinedbased ontheHCOOHconcentrationprofileandthe currentprofiles.Theonsetpotentialbasedontheconcentration isdefinedastheaveragepotentialofthepotentialswherethecon- centrationofHCOOHreaches0.01mM,0.03mMand0.05mM.The potentialscorrespondingtothesedifferentconcentrationsleadto asimilartrend.ThisisshowninFig.S.7intheSupportingInfor- mation.Similarly,theonsetpotentialbasedonthecurrentdensity isdefinedastheaverageofthepotentialscorrespondingtodif- ferentcurrentdensitieswithintherangeof0.25–1mAcm⁻².The onsetpotentialsareconverted totheNHEscale,and thetrends forthe3MPPsareshowninFig.10.Thedatawithintheshaded areacorrespondtothephosphatebufferelectrolytes.Notethatthe analysisoftheonsetpotentialsisonlyperformedforabettercom- parisonbetweenthedifferentMPPsandnottoderiveunderlying mechanisticinformation.ThedifferenceinpHbeforeandafterCO2

saturationasmentionedintheexperimentalsection,willnotaffect theseresults,sincealltheMPPsaremeasuredinthesameelec- trolyte.Thebuffercapacityoftheelectrolytemayplayanimportant roleontheselectivityforCO₂reduction[47,48].AsshowninFig.

S.10,ahigherbuffercapacityleadstohighercurrentsandaffectsthe catalyticactivityoftheporphyrintowardsformicacid.Therefore, theelectrolytepHisnottheonlyfactorinfluencingthecatalytic activityoftheimmobilizedporphyrinsTheonsetpotentialsatpH 1are quitedifferent,becausetheinfluenceof protonreduction

Fig.10.Onsetpotentialsbasedon(a)thecurrentforCO2-saturatedsolution,(b)theHCOOHconcentrationprofile,(c)themaximumconcentrationofHCOOHand(d)the currentoftheHER.

isdominanthere.AscanbeseeninFig.10a,theonsetpotential showsalineardependenceonpHforallMPPs.ThepHdependence (slope)isnotthesameforthethreeMPPs.Theonsetpotentialof SnPPshiftswith53±2mVpH⁻¹whileforInPPandRhPPtheyshift with36±5 and35±4mVpH⁻¹.Aslopeof 59mVpH⁻¹ onthe NHEpotentialscalewouldindicatea concertedproton–electron transfermechanism[49].ItisdifficulttosayifSnPPfollowsacon- certedproton–electron transfermechanism solelybased onthe analysisoftheonsetpotentials.However,theresultsindicatethat thereisprobablyadifferenceinmechanismbetweenSnPPonone handand RhPPand InPPontheotherhand.Anotherdistinction betweenSnPPandRhPPorInPPisthefactthatatpH11.6only smallamountsofHCOOHareformedonSnPP,whilereasonable amountsofHCOOHareproducedonInPPandRhPP.ForRhPPsim- ilarpHdependencesareobservedfortheonsetofHCOOHandfor H2 formation(Fig.10aandd),indicatingthattheCO2 reduction toHCOOHonRhPPiscoupledtotheconcomitantHER,asalready suggestedbefore.InPPandSnPPdonotexhibitaclearlinearpH dependenceforHER(seeFig.10d),whichsuggeststhatthetrend observedinFig.10aisrelatedtotheCO₂reductionratherthanto theHER.OnInPPandSnPP,CO₂reductionandHERarenotstrongly coupled,asconcludedbefore.OnRhPPthepotentialsofmaximum HCOOHproduction(Fig.10c)showasimilarslopetothatofthe onsetpotentialsofHCOOH(Fig.10b)confirmingtheearlierobser- vationofapotentialshiftoftheconcentrationprofileswithpH,in contrasttoInPPandSnPPwhereasimilarslopeisonlyobserved

fortheHCOOHmaximaandonsetpotentialofthecurrent(Figs.10a andc).Furthermore,theonsetpotentialsofRhPParealwaysatmore positivepotentialscomparedtothoseofInPPandSnPP.Thisdiffer- enceinonsetpotentialsconfirmsthatRhPPismoreactiveforthe HER.

3.3. Faradaicefficiencies

ForafurthercomparisonoftheactivitybetweentheMPPs,the faradaicefficienciesweredeterminedduring2helectrolysisina two-compartmentcell.Thefaradaicefficienciesforformicacidin HClO₄,pH3forallthethreeMPPsatdifferentpotentialsareshown asfunctionoftimeinFig.11.ForRhPPandSnPPthefaradaiceffi- cienciesalwaysdecaywithtimetovaluesof≈1–2%.Interestingly InPPseemstoreachasteadystatevalueof≈10–15%.Thisindicates thattheimmobilizedporphyrinsonPGarenotverystableordeac- tivateratherquickly,especiallyRhPP-PGandSnPP-PG.Improving theperformanceoftheimmobilizedporphyrinsisbeyondthescope ofthispaper.However,webelievethatthedeactivationisrelated tothecatalystinsteadofdepositionofpoisoningspeciesonthe surfaceorblockageoftheactivesitesbyintermediates.Asshown inFig.S.9theblankvoltammogramoftheimmobilizedporphyrin onPGiscomparedwiththeblankvoltammogramafterCO2reduc- tionorHER.Itcanbeseenthattheporphyrin-specificredoxpeaks havedisappearedandthebackgroundcurrenthaschanged.There- forewe believe thatthe decreasein activityis associated with

Fig.11.Faradaicefficienciesin0.001MHClO4+0.099MNaClO4ondifferentMPPs asfunctionoftime.

deactivationoftheporphyrinstructureordetachmentofthepor- phyrinfromPG.Itis assumedthat thevery negativepotentials ultimatelyaredestructivefortheimmobilizedporphyrins.RhPP showslowerfaradaicefficienciescomparedtoSnPPwhichinturnis lowercomparedtoInPP.Thefaradaicefficienciesdeterminedafter 10mininHClO4pH=3areplottedagainsttheappliedpotentialsin Fig.13a.Thepotentialwherethehighestfaradaicefficienciesare obtainedinpH3isaroundE=−1.3VforRhPP,aroundE=−1.9V forInPP,andapproximatelyE=−1.5VforSnPP.Thistrenddepends

ontheelectrolyte,ascanbeseeninFig.12b,wherethesameis depictedforaphosphatebufferofpH5.8.Whenonlyexperiments atanappliedpotentialofE=−1.5Vareconsidered,theinfluence ofthepHcanberevealedasseen inFig.S.8intheSI. Theini- tialfaradaicefficiencyofthedifferentMPPsasa functionofpH isshowninFig.13.ThefaradaicefficiencyofInPPatpH9.6seems tobeoptimalandreachesavaluecloseto70%.

3.4. Electroactivespecies

ThefactthateveninquitealkalineelectrolytesHCOOHispro- duced,couldindicatethatnot(only)CO₂istheelectroactivespecies butforinstancebicarbonateisinvolvedintheformationofformate.

IntheliteratureonelectrocatalyticCO2reductiontherehasbeen controversyabouttherealelectroactivespeciesduringCO₂reduc- tionatdifferentpHandatdifferentcatalysts.OftendissolvedCO₂ hasbeenidentifiedastheelectroactivespecies[50,51],butthere arealsosystemswherebicarbonatehasbeensuggestedtobethe keyreactantspecificallyfortheformationofformicacid/formate [52–56].

ToprobetheroleofHCO⁻₃,westudieddifferentconcentrations ofKHCO₃aselectrolytewithandwithoutCO₂saturation.InFig.14a itisshownthatahigherconcentrationofKHCO3leadstoalarger amountofHCOOHformedonRhPPwhennoCO2isspargedthrough thesolution.IftheelectrolyteissaturatedwithCO₂anevenlarger amountofHCOOHisproduced.ThisisobservedforInPPandSnPP aswell,asshowninFig.14b.TheHCOOHconcentrationonInPPand SnPPisincreasedbyafactorofmorethan10whenCO₂ispurged throughthesolution.TheseresultsgiveafirstimpressionthatCO₂ shouldbethedominantelectroactivespecies,however,itisstill notconclusiveenoughtoruleoutHCO⁻₃asreactivespecies.OLEMS experimentsinKHCO₃ withandwithoutCO₂bubblingshownin Fig.15,indicatethatCO2doeshaveaninfluenceastheH2evolution andCO₂consumptionaredelayed.EvenwithoutCO₂ saturation, theCO₂masssignaldecreasesduringthenegativepotentialsweep implyingthattheoriginoftheformedHCOOHistheKHCO3elec- trolyteitself.However,duringthenegativepotentialsweepthepH inthevicinityoftheworkingelectrodeincreasesasaresultofthe reactionsshowninEqs.(1)and(2).ThishigherlocalpHcouldlead toalocalconversionofCO2toHCO⁻₃ (Eq.(3)),whichmayinfluence thedirectbicarbonatereductionmechanismtoformate.Inprevi- ousstudies,directbicarbonatereductionhasbeensuggestedon palladium,leadandcopperelectrodes[53–56].Theexperiments performedinthisstudyonlyprovidequalitativeinformationabout theinfluenceofCO₂ andHCO⁻₃.Thesimultaneousmeasurement ofthelocalconcentrationsofformate,CO2andbicarbonatedur- ing voltammetryis crucial in order to shedmore light on this debate.

2H⁺+2e⁻→H₂ (1)

2H₂O+2e⁻→H₂+2OH⁻ (2)

CO2+OH⁻→HCO⁻₃ (3)

4. Conclusions

Inthisstudyweinvestigatedtheinfluenceofthemetalcen- terofmetalloprotoporphyrinsfortheelectrocatalyticreductionof CO₂towardsformicacid.WefoundthatSn,InandRhmetalcen- tersareabletoproducesignificantamountsofHCOOHwhileNi, Ga,PdandCumetalcentersonlyshowtraceamountsofHCOOH.

MetalloprotoporphyrinswithCr,Mn,CoorFecentersdonotpro- ducemeasurableamountsofHCOOH.Moreover,immobilizingthe MPPsonPGshows increasedactivityand stabilitycomparedto homogeneouscatalysiswiththecomplexinsolution.

Fig.12. Faradaicefficienciesdeterminedatt=10min,asfunctionofpotentialin(a)0.001MHClO4+0.099MNaClO4and(b)phosphatebufferofpH5.8.

Fig.13.Faradaicefficienciesdeterminedatt=10min,atE=−1.5VasfunctionofpH.

Thehydrogenevolutionreactionplaysanimportantroleinthe differenceinactivitytowardsHCOOHformationonRhPP,InPPand SnPP.RhPPisthemostactiveforHERandInPPtheleastactive.

Consequently,InPPshowshighfaradaicefficiencytowardsHCOOH

formationfromCO2andRhPPlowfaradaicefficiency.SnPPliesin betweenwithmoderateactivityforHERand faradaicefficiency towardsHCOOH.Allcatalystsdeactivatewithtime,however,the deactivation of InPP seems to be less compared to RhPP and SnPP.

AllthreeHCOOH-producingMPPsshowapHdependencefor CO2reductionaswellasforHER.AtverylowpH,theprotonreduc- tionisdominantwhichresultsinlittleornoHCOOHformed.Atvery alkalinepHs,theHERalsoseemstobedominant,leadingtopoor selectivitytowardsHCOOH.TheoptimalpHisaround9.6forInPP andbetween7and10forSnPP.Theidealsituation(highestfaradaic efficiencies)isdifferentforthethreeMPPsintermsofelectrolyte solutionandappliedpotential.Thebestperformanceobservedin thisstudyisforInPPinpH9.6electrolytewherefaradaicefficiencies of≈70%wereobtained.

This workhighlights some important properties of metallo- protoporphyrinsforelectrocatalyticCO2reductiontoformicacid.

However,thereisnoconsensusyetabouttheelectroactivespecies for the formation of formic acid. Quantitative measurement of HCO⁻₃,CO2andHCOOHconcentrationsduringvoltammetrywould benecessary inthis respect.For a detailed investigationof the mechanism,thesupportoftheoreticalcalculationssuchasinRef.

[34]isalsodesirable.

Fig.14.CO2/HCO⁻₃reductioninKHCO3onthedifferentMPPs.Scanrate:1mVs⁻¹.

Fig.15. H2andCO2signalsduringCO2/HCO3reductionin0.5MKHCO3.Scanrate:1mVs⁻¹.

Acknowledgement

J.S.acknowledgestheawardofagrantfromtheChineseSchol- arshipCouncil.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,athttp://dx.doi.org/10.1016/j.cattod.2017.02.

046.

References

[1] I.Dincer,Environmentalimpactsofenergy,EnergyPolicy27(1999)845–854.

[2]R.K.Pachauri,L.A.Meyer(Eds.),ClimateChange2014:SynthesisReport.

ContributionofWorkingGroupsI,IIandIIItotheFifthAssessmentReportof theIntergovernmentalPanelonClimateChange,IPCC,Geneva,Switzerland, 2014.

[3]S.Bachu,CO2storageingeologicalmedia:role,means,statusandbarriersto deployment,Prog.EnergyCombust.Sci.34(2008)254–273.

[4]S.I.Plasynski,J.T.Litynski,H.G.McIlvried,R.D.Srivastava,Progressandnew developmentsincarboncaptureandstorage,Crit.Rev.PlantSci.28(2009) 123–138.

[5]M.Aresta,A.Dibenedetto,Thecontributionoftheutilizationoptionto reducingtheCO2atmosphericloading:researchneededtoovercomeexisting barriersforafullexploitationofthepotentialoftheCO2use,Catal.Today98 (2004)455–462.

[6]N.S.Spinner,J.A.Vega,W.E.Mustain,Recentprogressintheelectrochemical conversionandutilizationofCO2,Catal.Sci.Technol.2(2012)19–28.

[7]N.L.Panwar,S.C.Kaushik,S.Kothari,Roleofrenewableenergysourcesin environmentalprotection:areview,Renew.Sustain.EnergyRev.15(2011) 1513–1524.

[8]H.Lund,Renewableenergystrategiesforsustainabledevelopment,Energy32 (2007)912–919.

[9]R.Ba ˜nos,F.Manzano-Agugliaro,F.G.Montoya,A.Alcayde,C.Gil,J.Gómez, Optimizationmethodsappliedtorenewableandsustainableenergy:a review,Renew.Sustain.EnergyRev.15(2011)1753–1766.

[10]A.Lü,Y.Fang,M.Zhu,S.Huang,Z.Ou,K.M.Kadish,Dioxygenreduction catalyzedbysubstitutedirontetraphenylporphyrinsinacidicmedia,J.

Porphyr.Phthalocyanines16(2012)310–315.

[11]C.Costentin,S.Drouet,M.Robert,J.-M.Savant,Alocalprotonsourceenhances CO2electroreductiontoCObyamolecularFecatalyst,Science338(6103) (2012)90–94.

[12]S.-i.Yamazaki,Y.Yamada,S.Takeda,M.Goto,T.Ioroi,Z.Siromaa,K.Yasuda, Effectsofp-substituentsonelectrochemicalCOoxidationbyRh

porphyrin-basedcatalysts,Phys.Chem.Chem.Phys.12(2010)8968–8976.

[13]A.Ghosh,Acomparisonofortho,meta,andparasubstituenteffectsin tetraphenylporphyrins:insightsintothenatureoftheporphyrin-phenyl electronicinteraction,J.Mol.Struct.(Theochem)388(1996)359–363.

[14]M.E.Vol’pin,I.S.Kolomnikov,Reactionsofcarbondioxidewithtransition metalcompounds,PureAppl.Chem.33(4)(1973)567–582.

[15]E.E.Benson,C.P.Kubiak,A.J.Sathrum,J.M.Smieja,Electrocatalyticand homogeneousapproachestoconversionofCO2toliquidfuels,Chem.Soc.

Rev.38(2009)89–99.

[16]J.L.Inglis,B.J.MacLean,M.T.Pryce,J.G.Vos,Electrocatalyticpathwaystowards sustainablefuelproductionfromwaterandCO2,Coord.Chem.Rev.256 (2012)2571–2600.

[17]R.W.Murray,Chemicallymodifiedelectrodes,Acc.Chem.Res.13(1980) 135–141.

[18]W.A.Herrmann,C.W.Kohlpaintner,Water-solubleligands,metalcomplexes, andcatalysts:synergismofhomogeneousandheterogeneouscatalysis, Angew.Chem.Int.Ed.Engl.32(1993)1524–1544.

[19]A.Yes¸ilda˘g,D.Ekinci,Covalentattachmentofpyridine-typemoleculesto glassycarbonsurfacesbyelectrochemicalreductionofinsitugenerated diazoniumsalts.Formationofrutheniumcomplexesonligand-modified surfaces,Electrochim.Acta55(2010)7000–7009.

[20]T.Atoguchi,A.Aramata,A.Kazusaka,M.Enyo,

Cobalt(II)-tetraphenylporphyrin-pyridinecomplexfixedonaglassycarbon electrodeanditsprominentcatalyticactivityforreductionofcarbondioxide, J.Chem.Soc.,Chem.Commun.(1991)156–157.

[21]T.Atoguchi,A.Aramata,A.Kazusaka,M.Enyo,ElectrocatalyticactivityofColl TPP-pyridinecomplexmodifiedcarbonelectrodeforCO2reduction,J.

Electroanal.Chem.318(1991)309–320.

[22]T.Yoshida,K.Kamato,M.Tsukamoto,T.Iida,D.Schlettwein,M.Kaneko,D.

Wohrle,SelectiveelectrocatalysisforCO2reductionintheaqueousphase usingcobaltphthalocyanine/poly-4-vinylpyridinemodifiedelectrodes,J.

Electroanal.Chem.385(1995)209–225.

[23]T.deGroot,ElectrochemistryofImmobilizedHemesandHemeProteins(PhD thesis),EindhovenUniversityofTechnology,2007.

[24]A.Deronzier,J.-C.Moute,Functionalizedpolypyrroles.Newmolecular materialsforelectrocatalysisandrelatedapplications,Acc.Chem.Res.22 (1989)249–255.

[25]J.Roncali,Electrogeneratedfunctionalconjugatedpolymersasadvanced electrodematerials,J.Mater.Chem.9(1999)1875–1893.

[26]A.Bettelheim,B.A.White,S.A.Raybuck,R.W.Murray,Electrochemical polymerizationofamino-,pyrrole-,andhydroxy-substituted tetraphenylporphyrin,Inorg.Chem.26(1987)1009–1017.

[27]A.Giraudeau,D.Schaming,J.Hao,R.Farha,M.Goldmann,L.Ruhlmann,A simplewayfortheelectropolymerizationofporphyrins,J.Electroanal.Chem.

638(2010)70–75.

[28] R.Boulatov,Metalloporphyrincatalystsforoxygenreduction,in:M.T.M.

Koper(Ed.),FuelCellCatalysis.ASurfaceScienceApproach,JohnWiley&

Sons,Inc.,NewYork,2009,pp.637–693,Chapter18.

[29]J.Shen,Y.Y.Birdja,M.T.M.Koper,Electrocatalyticnitratereductionbyacobalt protoporphyrinimmobilizedonapyrolyticgraphiteelectrode,Langmuir31 (30)(2015)8495–8501.

[30]C.Costentin,M.Robert,J.-M.Savant,Currentissuesinmolecularcatalysis illustratedbyironporphyrinsascatalystsoftheCO2-to-COelectrochemical conversion,Acc.Chem.Res.48(2015)2996–3006.

[31]A.Maurin,M.Robert,Noncovalentimmobilizationofamoleculariron-based electrocatalystoncarbonelectrodesforselective,efficientCO2-to-CO conversioninwater,J.Am.Chem.Soc.138(2016)2492–2495.

[32]L.Chen,Z.Guo,X.-G.Wei,C.Gallenkamp,J.Bonin,E.Anxolabhre-Mallart, K.-C.Lau,T.-C.Lau,M.Robert,Molecularcatalysisoftheelectrochemicaland photochemicalreductionofCO2withearth-abundantmetalcomplexes.

SelectiveproductionofCOvsHCOOHbyswitchingofthemetalcenter,J.Am.

Chem.Soc.137(34)(2015)10918–10921.

[33]J.Shen,M.J.Kolb,A.J.Göttle,M.T.M.Koper,DFTstudyonthemechanismof theelectrochemicalreductionofCO2catalyzedbycobaltporphyrins,J.Phys.

Chem.C120(29)(2016)15714–15721.

[34]J.S.Yoo,R.Christensen,T.Vegge,J.K.Norskov,F.Studt,Theoreticalinsightinto thetrendsthatguidetheelectrochemicalreductionofcarbondioxideto formicacid,ChemSusChem9(4)(2016)358–363.

[35]M.T.deGroot,M.T.M.Koper,Redoxtransitionsofchromium,manganese, iron,cobaltandnickelprotoporphyrinsinaqueoussolution,Phys.Chem.

Chem.Phys.10(2008)1023–1031.

[36]A.H.Wonders,T.H.M.Housmans,V.Rosca,M.T.M.Koper,On-linemass spectrometrysystemformeasurementsatsingle-crystalelectrodesin hangingmeniscusconfiguration,J.Appl.Electrochem.36(2006)1215–1221.

[37]Y.Kwon,M.T.M.Koper,CombiningvoltammetrywithHPLC:applicationto electro-oxidationofglycerol,Anal.Chem.82(2010)5420–5424.

[38]J.Shen,R.Kortlever,R.Kas,Y.Y.Birdja,O.Diaz-Morales,Y.Kwon,I.

Ledezma-Yanez,K.J.P.Schouten,G.Mul,M.T.M.Koper,Electrocatalytic reductionofcarbondioxidetocarbonmonoxideandmethaneatan immobilizedcobaltprotoporphyrin,Nat.Commun.6(2015)8177.

[39]N.Furuya,K.Matsui,Electroreductionofcarbondioxideongas-diffusion electrodesmodifiedbymetalphthalocyanines,J.Electroanal.Chem.271 (1989)181–191.

[40]M.Azuma,K.Hashimoto,M.Hiramoto,M.Watanabe,T.Sakata,

Electrochemicalreductionofcarbondioxideonvariousmetalelectrodesin low-temperatureaqueousKHCO3media,J.Electrochem.Soc.137(6)(1990) 1772–1778.

[41]Y.Hori,ModernAspectsofElectrochemistry,vol.42,Springer,NewYork, 2008.

[42]F.Gassner,W.Leitner,Hydrogenationofcarbondioxidetoformicacidusing water-solublerhodiumcatalysts,J.Chem.Soc.,Chem.Commun.19(1993) 1465.

[43]J.-C.Tsai,K.M.Nicholas,Rhodium-catalyzedhydrogenationofcarbondioxide toformicacid,J.Am.Chem.Soc.114(1992)5117–5124.

[44]F.Hutschka,A.Dedieu,M.Eichberger,R.Fornika,W.Leitner,Mechanistic aspectsoftherhodium-catalyzedhydrogenationofCO2toformicacids–a theoreticalandkineticstudy,J.Am.Chem.Soc.119(1997)

4432–4443.

[45]S.Slater,J.H.Wagenknecht,ElectrochemicalreductionofCO2catalyzedby Rh(diphos)2Cl,J.Am.Chem.Soc.106(1984)5368–5370.

[46]C.M.Bolinger,N.Story,B.P.Sullivan,T.J.Meyer,Electrocatalyticreductionof carbondioxideby2,2^-bipyridinecomplexesofrhodiumandiridium,Inorg.

Chem.27(1988)4582–4587.

[47]R.Kas,R.Kortlever,H.Yilmaz,M.T.M.Koper,G.Mul,Manipulatingthe hydrocarbonselectivityofcoppernanoparticlesinCO2electroreductionby processconditions,ChemElectroChem2(3)(2015)354–358.

[48]A.S.Varela,M.Kroschel,T.Reier,P.Strasser,ControllingtheselectivityofCO2

electroreductiononcopper:theeffectoftheelectrolyteconcentrationand theimportanceoflocalpH,Catal.Today260(2016)8–13.

[49]M.T.M.Koper,Theoryofmultipleproton–electrontransferreactionsandits implicationsforelectrocatalysis,Chem.Sci.4(2013)2710–2723.

[50]P.vanRysselberghe,G.J.Alkire,J.M.McGee,Polarographicreductionofcarbon dioxide.III.Descriptionandinterpretationofthewaves,J.Am.Chem.Soc.68 (1946)2050–2055.

[51]T.E.Teeter,P.VanRysselberghe,Reductionofcarbondioxideonmercury cathodes,J.Chem.Phys.22(1954)759–760.

[52]M.Spichiger-Ulmann,J.Augustynski,Remarkableenhancementoftherateof cathodicreductionofhydrocarbonateanionsatpalladiuminthepresenceof caesiumcations,Helv.Chim.Acta69(1986)632–634.

[53]M.Spichiger-Ulmann,J.Augustynski,Electrochemicalreductionof bicarbonateionsatabrightpalladiumcathode,J.Chem.Soc.FaradayTrans.

81(1985)713–716.

[54]E.A.Kolyadko,B.I.Podlovchenko,S.Lu,Reductionofcarbondioxideon mercurycathodes,J.Electroanal.Chem.373(1994)185–187.

[55]B.Innocent,D.Pasquier,F.Ropital,F.Hahn,J.-M.Leger,K.B.Kokoh,FTIR spectroscopystudyofthereductionofcarbondioxideonleadelectrodein aqueousmedium,Appl.Catal.B:Environ.94(2010)219–224.

[56]R.Kortlever,K.H.Tan,Y.Kwon,M.T.M.Koper,Electrochemicalcarbondioxide andbicarbonatereductiononcopperinweaklyalkalinemedia,J.SolidState Electrochem.17(2013)1843–1849.