This is an author-deposited version published in:
http://oatao.univ-toulouse.fr/
Eprints ID: 8732
To link to this article: DOI:
10.1016/j.electacta.2011.11.029
URL:
http://dx.doi.org/10.1016/j.electacta.2011.11.029
To cite this version:
Ponrouch, Alexandre and Taberna, Pierre-Louis and
Simon, Patrice and Palacín, M. Rosa On the origin of the extra capacity at
low potential in materials for Li batteries reacting through conversion
reaction. (2012) Electrochimica Acta, vol. 61 . pp. 13-18. ISSN
0013-4686
!
O
pen
A
rchive
T
oulouse
A
rchive
O
uverte (
OATAO
)
OATAO is an open access repository that collects the work of Toulouse researchers and
makes it freely available over the web where possible.
Any correspondence concerning this service should be sent to the repository
administrator:
staff-oatao@listes-diff.inp-toulouse.fr
!
On
the
origin
of
the
extra
capacity
at
low
potential
in
materials
for
Li
batteries
reacting
through
conversion
reaction
Alexandre
Ponrouch
a,b,
Pierre-Louis
Taberna
b,c,
Patrice
Simon
b,c,
M.
Rosa
Palacín
a,b,∗aInstitutdeCiènciadeMaterialsdeBarcelona(ICMAB-CSIC),CampusdelaUAB08193Bellaterra,Catalonia,Spain bALISTORE-ERIEuropeanResearchInstitute
cUniversitéPaulSabatier,CIRIMAT,UMRCNRS5085,Toulouse31062,France
Keywords: Energystorage Conversionreactions Lithiumbatteries Interfacialstorage
a
b
s
t
r
a
c
t
Thepossibilityofinterfacialstorageatlowpotentialforelectrodematerialsreactingthroughconversion reactionswasevaluated.Theamountofchargethatcouldbestoredthroughtheproposedinterfacial mechanismwasestimatedforarangeofdifferentmaterialsandfoundtobemuchlowerthanthose observedexperimentally.Moreover,theslopeofthepotentialdecayandtheinfluenceofthecurrentin theextentofstoredcapacityforexperimentscarriedoutincompositeelectrodescontainingCo3O4are
notconsistentwithacapacitive-likemechanism.Insummary,noevidenceforcapacitivestoragecouldbe found,ourresultsbeinginagreementwiththeprocesstakingplaceatlowpotentialbeingsolelyrelated toelectrolytedecomposition.
1. Introduction
Thedevelopmentoflithiumbasedbatterieswithhighenergy density is one of the expected breakthroughs in the current quest for energy storage systems with enhanced performance
[1,2].Progressinthestudyofinsertioncompoundsenabledthe developmentandcommercialisationoflithiumionbatteriesbut a paradigmatic shift in energy density will only be achieved with the use of electrodes operating through alternative reac-tionmechanisms.Whilematerialselectrochemicallyformingalloys with lithium [3] are starting to become a commercial reality, thoseoperatingthroughconversionreactions[4]exhibit promis-ingexpectative.Themaindrawbackforthosematerialsisthatthe largeavailableelectrochemicalcapacityisachievedattheexpense ofmajorstructuralchangesintheelectrodethataredifficultto “buffer”.Theelectrochemicalformationofalloyswithlithiumhas beenstudiedforlongand a numberofsuccessfulstrategies [5]
havebeendevelopedtoovercometheseintrinsicshortcomings.On thecontrary,thestudyofconversionreactionmaterialsismuch morerecentandtherearestillsomekeyissuesthatdeserve under-standingatfundamentallevelsuchasvoltagehysteresis orlow coulombicefficiencyonthefirstcycle,beforecommercial imple-mentationcanberealisticallydiscussed.
∗ Correspondingauthorat:InstitutdeCiènciadeMaterialsdeBarcelona (ICMAB-CSIC),CampusdelaUAB08193Bellaterra,Catalonia,Spain.
E-mailaddress:rosa.palacin@icmab.es(M.R.Palacín).
Conversionreactionisthetermappliedtodefinethe electro-chemicalreactionof abinarytransitionmetalcompound, MaXb (M=transitionmetal,X=O,S, F, P,N,...)withlithium toyield metallicnanoparticlesembeddedinamatrixofLicX.Duetothefull reductionofthetransitionmetaltothemetallicstatetheyresultin remarkablyhighcapacityvalues.Moreover,electrochemical capac-itiesexceedingthetheoreticalvalueare generallyobserved[4]. Thesignatureofsuchadditional reversiblecapacity isasloping curveatlowpotentials(generallybelow0.8Vvs.Li+/Li,seeFig.1A). Whiledecompositionoftheelectrolyteuponreductionwith forma-tionofagel-likepolymericfilmwhichisconsumeduponfurther oxidationhasbeenprovedbydiverseelectrochemicaltechniques
[6–9],analternativeenergystoragemechanismtermed “interfa-cialstorage”hasalsobeenproposedtoexplainsuchextracapacity (Fig.1B).Thelatterissupportedbytheoreticalcalculations[10–12]
andbasedonatwophasecapacitivebehaviourattheM/LicX inter-facesinthereducedelectrodeswhichwouldallowforthestorage ofLi+ionsontheLi
cXsideandelectronsontheMside.Capacitive andfaradaicphenomenabeingintrinsicallydifferent,theaimof thecurrentstudywastoascertainwhetherproofoftheexistence ofsuchcapacitivestoragecouldbeachievedinpresenceof con-comitantformationofthegel-likepolymericfilmthroughfaradaic electrolytedecomposition.
2. Experimental
Compositeelectrodesmimickingindustrialtechnologieswere prepared withCo3O4 as active materialsfrom slurries(65wt.%
Fig.1.(A)Typicalpotentialversuschargecapacityprofilecurve.Usualvaluesfor Q/F(correspondingtob·cinEq.(1))rangebetween2and8molofelectronspermole ofMaXb)[4].(B)Schemesofthetwoproposedredoxmechanismstakingplaceat
lowpotential.
Co3O4,8wt.%of polyvinylidenefluoride binder(PVDF, Arkema) and27wt.%ofSuperPcarbon(CsphereafterfromTimcal)in N-methylpyrrolidone (NMP, Aldrich))mixed for 15hby magnetic stirringwiththreeintermediate10minsonicationsteps[13].These weretapecastedona20mmthickcopperfoil(Goodfellow)with a250mmDoctor-Bladeandfurtherdriedat120◦Cundervacuum. Oncedried,0.8cm2diskelectrodeswerecutandpressedat8tprior totesting.
Electrochemical testing was performed in two electrode Swagelokcellsforcomposite electrodesusingadiskofLimetal foil(Chemetall)ascounterandreferenceelectrodeand1MLiPF6 inEC:DMC1:1(LP30,Merck)electrolyte.Electrochemicalcycling experimentsweremadeingalvanostaticmodewithpotential lim-itation(GCPL)usingaBio-LogicVMP3potentiostatatverydifferent ratesrangingfromC/10to10C(1CbeingoneLi+insertedin1h). Reproducibilitywascheckedbyassemblyoftwincells.
3. Resultsanddiscussion
Capacitiveelectrochemicalphenomena aremuch faster than faradaicprocessesinvolvingaredoxreaction,asillustratedbytheir respectivetimeconstants:secondsfortheformerandminutesor hoursforthelatter[14].Thus,andinordertodiscriminatebetween faradaicand capacitiveprocesses in materials reactingthrough conversionreaction,we decidedto evaluatethe kineticsofthe diverseredoxprocessestakingplaceincompositeCo3O4 contain-ingelectrodes.Theresultsofgalvanostaticcyclingatdifferentrates forbothelectrodesareexhibitedinFig.2.
Generallyspeaking,thefirstreduction ofoxidesexhibiting a conversionreactionmechanismwithrespecttolithiumexhibits
Fig.2.(A)PlotofpotentialversuschargecapacityprofileforCo3O4composite
elec-trodesuponthefirstcycleatdifferentCratesand(B)plotofthechargecapacity percentageforeachprocess(i.e.insertiona;conversionbandextracapag)with respecttothetotalcapacityachieveduponthefirstreduction.
threecomponents:(i)anoccasionalcertaindegreeoflithium inser-tioninthestructure(whichwewilldenotea),(ii)theconversion reactionitself(denotedb)andiii)analmostlinearpotentialdecay below1.15VvsLi+/Li(notedg)responsiblefortheextracapacity (seeFig.1).Fig.2Aexhibitsthepotentialversuscapacityprofiles forCo3O4containingelectrodescycledatratesrangingfromC/10 to10C.Thetotalvalueofthecapacityisfoundtodecreasewith increasingrateasexpectedandgenerallyobservedintheliterature forconversionreactionmaterials.Therelativevalueofa, corre-spondingtotheinsertionstepisfoundtodecreasewithincreasing therateinagreementwithpreviousreportssuggestinga correla-tionbetweenthestabilityoftheintercalatedphaseandthecurrent density[15].Interestingly,thecapacityinvolvedinbothbandg processesisreducedwithincreasingcyclingrate,butthisdecrease seemstobeenhancedforthegcomponent(extracapacity).This resultisconfirmedbytheplotoftherelativevalueofthea,band g componentswithrespecttothetotalcapacity(Fig.2B)achieved uponthefirstreduction.Indeed,thiswouldmeanthatthe kinet-icsoftheconversionreaction(ˇ),whichcorrespondstoafaradaic processarefasterthanthatoftheg processoccurringatlow poten-tial.Thisisinfullagreementwiththeobservedcatalyticelectrolyte decompositionattheselowpotentials[16]butdoesnotseemto beconsistentwithinterfacialstoragebeingthemainprocess tak-ingplace.Indeed,much fasterkineticsshouldbeexpectedfrom capacitiveprocesses.
Bearinginmindthetwohypotheticalmechanismsresponsible fortheextracapacityatlowpotential,namely(i)the decomposi-tionoftheelectrolyteand(ii)thechargestorageattheinterface between themetallic nanoparticles and the lithiated matrix, it seemsstraightforwardtoconcludethatinthefirstcasethe spe-cific surfaceof thematerial(before conversionreaction)would playanimportantroleasthechargeassociatedwiththe decom-positionoftheelectrolyteisproportionaltothesurfaceincontact withtheelectrolyte.ThisisinagreementwithresultsfromDelmer etal.[17]forRuO2withanextracapacityof31%ofthetotalfirst dischargefornanoparticles(ca.60nm)and24%forca.10mmsize particlesandalsofromourpreviousworkonCo3O4withparticle sizesofca.35nmand1mmexhibitingextracapacitiesof35%to 28%,respectively[13].
Capacitive behaviour is characterized by a linear potential decay/increaseinthepotential–discharge/chargecurve,theslope ofwhichisequaltothecurrentdividedbythecapacitance[18].The factthattheslopeobserveduponoxidationisgenerallylargerthan uponreductionforconversionreactionmaterials(seeforinstance
Figs.1Aand2A)isalsoinagreementwithelectrolytedegradation beingthemainphenomenonaccountingforextracapacityatlow potentials.
Hypotheticalcapacitive-likestorageattheinterfacebetween MandLicXwouldbedifficulttorecordusingconventional tech-niquessinceMandLicXareboundtothesamecurrentcollector andhenceexhibitthesamepotential.Nonethelesswefound inter-estingtoestimatetherelativeamountofchargethatcouldbestored bysuchamechanism,whichwouldbeproportionaltotheextent oftheM/LicXinterface.Itiscommonlyadmittedintheliterature thatmetallicnanoparticleswithdiameterbetween2and10nm areformedattheendofaconversionreaction[19,8,20]somewhat independently ontheinitialparticlesize.Considering a general conversionprocess,
MaXb+(b·c)Li +
+(b·c)e−
=aM+bLicX (1)
MassesofmetalMandlithiatedmatrix(i.e.LicX)obtainedatthe endoftheconversionprocesscanbeexpressedasfollows: mM=a·MM.m MX i MMX i (2) mLicX=a·MLicX·m MX i MMX i (3) mMandmLicXbeingthemassofmetalMandlithiatedmatrixLi
cX,
MMandMLicXbeingtheirrespectivemolecularweight,andmMX i andMMX
i beingtheinitialmassandmolecularweightofthe tran-sitionmetalcompound.
Thetotalvolumeofthesephasescanthenbeexpressedas(4) and(5),withubeingthedensityofaphaseu:
VM=a·MM· mMX i (MMX i ·M) (4) VLicX=a·MLicX· mMX i (MMX i ·LicX) (5) Finally,thenanoparticlesofmetalbeingconsideredasspheres withagivenradiusr,wecanexpresstheirtotalsurfaceattheend oftheconversionreactionasfollows:
SM=3·a·MM· m MX i (r·MMX i ·M) (6) Inadditiontothetotalmetallicnanoparticlesurfacearea(Eq.
(6)),thesesimplecalculationsleadtoageneralformulatocalculate thevolumechange(givenin%)inducedbytheconversionreaction
(7): V%
change=100−100×
(VM+VLiX)
ViMX (7)
Fig.3displaysthedataobtainedsolvingEqs.(6) and(7) for mostofthematerialsreportedtodatereactingthroughconversion reaction[4]notformingalloyswithlithiumandconsidering1gof precursormaterialand5nmaveragemetallicparticlediameter.All valuesareplottedwithrespecttothevolumeexpansionandsurface areaofmetallicnanoparticlespergramofMaXb.Weconsidered flu-orides(formationofLiFduringconversion):CuF2,TiF3,VF3,CoF2, FeF3,NiF2andCrF3;oxides(formationofLi2O):Cu2O,MnO,CuO, MoO3,Fe3O4,FeO,CoO,Mn2O3,NiO,Fe2O3,MoO2,Co3O4,RuO2, Cr2O3andMnO2;sulfides(formationofLi2S):Cu2S,MnS,CuS,FeS, WS2,MoS2,NiS,CoS2andFeS2;andalsosomenitrides(formation ofLi3N):CrNandCu3Nandphosphides(formationofLi3P):FePand CuP2.Thelownumberofphosphidesandnitridesresultsfromthe scarcityofthedensityvaluesin[21].
Asexpected,a generaltrend appearsforthevolumechange experiencedduringconversiondependingontheoxidationstateof X(c)whichdeterminesthestoichiometryofthebinaryLicXmatrix. Thehigherc,thelargertheamountofLi+reactingintheconversion processpermoleofMaXbandhencethelargertheexpected vol-umechange(cf.Fig.3).Indeed,thecalculatedvolumechangeswere foundtovaryfrom11%to30%forfluorides(c=1),from65%to165% fordivalentoxidesandsulphides(c=2)and from195%to235% forphosphidesandnitrides(c=3).Additionally,valueofa/blarger than1resultsinlowervolumechange,asexemplifiedbythe val-uesobtainedforCu2O(22%),Cu2S(48%)andCu3N(40%)whichare muchlowerthanvaluescalculatedforotherMaOb,MaSbandMaNb compounds(cf.Fig.3).Whileotherfactorssuchascost,availability, operationpotentialandsoonaredecisiveintermsofestimating thepotentialinterestofelectrodematerials,basingexclusivelyin theoreticalcapacityandvolumeexpansionbothtrivalentfluorides (e.g.TiF3,VF3,FeF3andCrF3)andFe3O4appearasmostinteresting caseexamplestobestudiedindetail.
Thevaluesofthemetallicnanoparticlesurfaceareasforallthe abovementionedcompoundsaredepictedinFig.3andrangefrom aboutca.50to130m2pergramofM
aXb.Interestingly,most com-pounds developmetallicsurface areasaround100m2 pergram ofMaXb.Comingbacktotheexperimentalextracapacityvalues
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0 1 2 3 4 (1 /Q e x tr a ) / m A h -1g (C rate)1/2 (C rate)-1/2
A
B
0 100 200 300 400 500 600 700 0 1 2 3 4 Q e x tr a / m A h g -1 Qouter 1/QtotalFig.4.Plotsof(A)Qextraversus(Crate)−1/2and(B)1/Qextraversus(Crate)−1/2.Red
linesarelinearextrapolationofthecapacityatinfinitelyfastandslowCratesin(A) and(B),respectively.
observedforCo3O4(Fig.2A),areca.600mAhg−1,corresponding to2160Cg−1.Thiscapacitybeingachievedbetween1and0Vvs Li+/Li,thechargestoredattheinterfacewouldbe2160Fg−1.Thus, ifonlyinterfacialstorageattheCo/Li2Ointerfacewas responsi-blefortheextracapacityandtakingintoaccountthat100m2of metalliccobaltnanoparticlespergramofCo3O4areproducedat theendoftheconversionreaction,theinterfacialcapacitancevalue wouldbe2160mFcm−2.Knowingthatstandardmaximumvalue forthedoublelayerspecificcapacitanceatthesurfaceofametalis aboutca.30mFcm−2[18],thisvalueissurprisinglyhigh(70times largerthandoublelayerstoragecapacity).Thus,itseems straight-forwardtoconcludethatifinterfacialstorageattheM/LicXtakes place,itwouldonlyaccountforasmallpercentageofthe exper-imentallyobservedcapacity,therestbeingeitherfullyrelatedto electrolytedecompositionorotherprocessesinvolving acharge transfer.
Itisworthmentioningthatsimilarconsiderationsled,about40 yearsago,totheconclusionthatdoublelayerchargingcontribution wasnegligiblewithrespecttotheverylargecapacitiesobservedfor RuO2,IrO2,etc.cycledinaqueousacidicsolution[22,18].Indeed, evenifacapacitivebehavior(i.e.constantcurrentwithlinearsweep ofthepotential)wasobservedinthosecases,theobtained capaci-tiesareabout10–100timeslargerthancapacitiesachievablewith accumulationofchargeatthedoublelayerinterface.Furthermore, Ardizzoneetal.demonstratedthatthecapacitivebehaviorofRuO2 inacidicsolutionisnotonlygovernedbysurfaceprocesses[23]. Indeed,byextrapolatingthevalueofthechargeatinfinitelyslow andfastsweepratestheywereabletodiscriminatebetweenbulk andsurfacecontributiontothecapacitiesrecordedand demon-stratedasignificantcontributionofthebulkofthematerial.Later on, this demonstrationledto thebetter comprehension ofthe
chargestorageofRuO2involvingfaradaicreactionswithdiffusion ofprotonsthroughthebulkofthematerial[18].
Fig.4Adisplaystheplotsoftheextracapacity(denotedQextra thereafter)versus(Crate)−1/2.Fig.4Bdisplaystheplotsof1/Q
extra versus(Crate)−1/2.SimilarplotswereproposedbyArdizzoneetal.
[23]inordertoevaluatethesurfacecharge(denotedQouter),the totalcharge(calledQtotal)andthechargerelatedtothebulk contri-butionofthecapacity(calledQinner).Thiswasdonebyextrapolating thevalueofthecapacity atinfiniteCrateinFig.4A(Qouter),by extrapolatingthevalueofthecapacityataninfinitelyslowCrate
inFig.4B(1/Qtotal).FinallyQinnercanbeevaluatedaccordingtoEq.
(8):
Qinner=Qtotal−Qouter (8)
FromFig.4AandB,450mAhg−1and710mAhg−1canbe esti-mated for Qouter and Qtotal, respectively. Therefore, Qinner is ca. 260mAhg−1,representingabout37%ofthetotal extracapacity. Thisdemonstratestheinvolvementofakineticallylimitedprocess. Aspreviouslydiscussed,thekineticlimitationinthecaseofRuO2 cycledinacidicsolutionisduetothediffusionofprotonthroughthe bulkoftheelectrode.InthecaseofCo3O4cycledinanonaqueous electrolyteitwould besurprisingthata kineticallylimited pro-cesscanoccurduringaninterfacialstorageasnomasstransport isinvolved.Bycontrast,theformationofapolymericfilmatthe surfaceoftheelectrodebydecompositionoftheelectrolyte(SEI formation)willobviouslybekineticallylimited bythediffusion oftheelectroactiveelectrolytespeciesthatwilldecomposeatthe interface.
4. Conclusions
Inthisstudyweevaluatedthepossibilityofinterfacialstorageat lowpotentialforelectrodematerialsreactingthroughconversion reactionsformingM/LicXcompositesattheendofreduction.While thiswouldbehardlymeasurableexperimentallywithinasingle electrode,theslopeofthepotentialdecayandtheinfluenceofthe currentintheextentofstoredcapacitydonotseemtobeconsistent withacapacitive-likemechanism.Also,simplegeometrical calcula-tionsindicatethatinterfacialstoragewouldonlyaccountforavery smallpercentofthetotalextracapacityvalue.Thesecalculations havebeengeneralizedtoallmaterialsreactingthrough conver-sionreactionswhicharecomparedintermsofvolumeexpansion andspecificsurfaceofmetallicnanoparticlesattheendof reduc-tion.Bulkandsurfaceprocessescontributiontocapacityhasbeen ascertainedusingthemethodreportedin[23].Insummary,our resultsindicatethatinterfacialstorage,ifany,wouldbe negligi-blewithrespecttoelectrolytedecompositiontoaccountforthe extracapacity observedat lowpotentialin conversionreaction materials.
Acknowledgements
WeacknowledgeMinisteriodeCienciaeInnovaciónforgrant MAT2011-24757andaregratefultoALISTORE-ERImembersfor helpfuldiscussions.
References
[1]M.R.Palacín,Chem.Soc.Rev.38(2009)2565. [2]M.Armand,J.M.Tarascon,Nature451(2008)652.
[3]C.M.Park,J.H.Kim,H.Kim,H.J.Sohn,Chem.Soc.Rev.39(2010)3115. [4]J.Cabana,L.Monconduit,D.Larcher,M.R.Palacín,Adv.Mater.22(2010)E170. [5]D.Larcher,S.Beattier,M.Morcrette,K.Edström,J.C.Jumas,J.M.Tarascon,J.
Mater.Chem.17(2007)3759.
[6] G.Gachot,S.Grugeon,M.Armand,S.Pilard,P.Guenot,J.M.Tarascon,S.Laruelle, J.PowerSources178(2008)409.
[8]S.Grugeon,S.Laruelle,R.Herrera-Urbina,L.Dupont,P.Poizot,J.-M.Tarascon, J.Electrochem.Soc.148(2001)A285.
[9]M.Dollé,P.Poizot,L.-J.Dupont,M.Tarascon,Electrochem.Solid-StateLett.5 (2002)A115.
[10]P.Balaya,A.J.Bhattacharyya,J.Jamnik,Y.F.Zhukovskii,E.A.Kotomin,J.Maier, J.PowerSources159(2006)171.
[11]J.Jamnik,J.Maier,Phys.Chem.Chem.Phys.5(2003)5215.
[12] Y.F.Zhukovskii,P.Balaya,E.A.Kotomin,J.Maier,Phys.Rev.Lett.96(2006) 058302.
[13] A.Ponrouch,M.R.Palacín,J.PowerSources196(2011)9682. [14]P.Simon,Y.Gogotsi,Nat.Mater.7(2008)845.
[15] D.Larcher,G.Sudant,J.-B.Leriche,Y.Chabre,J.-M.Tarascon,J.Electrochem.Soc. 149(2002)A234.
[16]S.Laruelle,S.Grugeon,P.Poizot,M.Dollé,L.Dupont,J.-M.Tarascon,J. Elec-trochemSoc.149(2002)A627.
[17]O.Delmer,P.Balaya,L.Kienle,J.Maier,Adv.Mater.20(2008)501.
[18]B.E.Conway,ElectrochemicalSupercapacitors:ScientificFundamentalsand TechnologicalApplications,Kluwer,Norwell,MA,1999.
[19]P.Poizot,S.Laruelle,S.Grugeon,L.Dupont,J.-M.Tarascon,Nature407(2000) 496.
[20] B.Varghese,M.V.Reddy,Z.Yanwu,C.S.Lit,T.C.Hoong,G.V.SubbaRao,B.V.R. Chowdari,A.T.S.Wee,C.T.Lim,C.-H.Sow,Chem.Mater.20(2008)3360. [21] D.R.Lide,HandbookofChemistryandPhysics,84thed.,CRCPressLLC,New
York,2003.
[22] S.Trasatti,G.Buzzanca,J.Electroanal.Chem.29(1971)1.