ContentslistsavailableatSciVerseScienceDirect
Synthetic
Metals
j o u r n al hom ep a ge :w w w . e l s e v i e r . c o m / l o c a t e / s y n m e t
Enhanced
carrier
extraction
efficiency
in
organic
solar
cells
with
reduced
space-charge
limited
transport
by
engineered
vertical
compositional
profile
Ming-Kun
Lee
a,
Tsung-Han
Kuo
b,
Kuo-Wen
Kong
a,
Yun-Ru
Hong
b,
Jen-Chun
Wang
c,
Sheng-Fu
Horng
a,∗,
Hsin-Fei
Meng
daDepartmentofElectricalEngineering,NationalTsingHuaUniversity,Hsinchu300,Taiwan,ROC bInstituteofPhotonicsTechnologies,NationalTsingHuaUniversity,Hsinchu300,Taiwan,ROC
cDepartmentofMaterialsScienceandEngineering,NationalTsingHuaUniversity,Hsinchu300,Taiwan,ROC dInstituteofPhysics,NationalChiaoTungUniversity,Hsinchu300,Taiwan,ROC
a
r
t
i
c
l
e
i
n
f
o
Articlehistory: Received1May2012
Receivedinrevisedform25August2012 Accepted12September2012
Available online 13 October 2012 Keywords:
Polymersolarcells Space-chargelimitedeffect Fillfactor
Layerthickness Lightbias Quantumefficiency
a
b
s
t
r
a
c
t
Amulti-layeredpolymersolarcell(PSC)withagradedcompositionalprofiletoreducespace-charge limited(SCL)transportwasdesignedandanalyzedindetails.Wefoundthatbyincreasingtheeffective carrierlifetimefortheslowerphoto-carriers,SCLeffectsthatdegradethefillfactorandthepower con-versionefficiencycanbealleviatedinPSCswithathickactivelayer.ThereductionofSCLeffectswas evidencedbyadiminisheddependenceoftheincidentphoton-to-electronconversionefficiency(IPCE) onlightbias,whichalsoindicatestheimportanceoflightbiasincharacterizingtheoperatingIPCEof PSCswiththickactivelayerswhichoperateinSCLregion.Basedonourresults,devicearchitectureto enhanceFFwasproposed.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
Polymer solar cells (PSCs) have attracted much attention
aspotential renewable energy sources because of their
advan-tagesof light weight, low fabrication cost, flexibility,and easy
manufacturing[1,2].ConventionalPSCsgenerallycomprisea
bulk-heterojunction(BHJ)withinterpenetratingnetworkofdonor(D)
andacceptor(A)thatgivesmoreinterfaces,atwhichthe
photo-generatedexcitonscanbeefficientlydissociated,yieldingahigh
photo-carriergenerationefficiency.Muchprogresshasbeenmade
and the power conversion efficiency (PCE) of BHJ PSCs based
onpoly(3-hexylthiophene)(P3HT)andfullerenederivative
(6,6)-phenylC61 butyric acid methyl ester (PCBM)was reported to
exceed5%[3–5].
TheconstitutingmaterialsinconventionalBHJdevicesusually
exhibitlowintrinsiccarriermobility.Inaddition,thedisrupting
(discontinuous)natureofdonorandacceptordistributioninBHJ
layerseverelyretardsthecarriertransport. Asaresult,theBHJ
layerswerecharacterizedbylowercharge-carriermobilitiesand
∗ Correspondingauthor.Tel.:+88635715131x42578;fax:+88635752120. E-mailaddresses:[email protected],[email protected] (S.-F.Horng).
lowcharacteristicchargecollectionlengths[6].Accordingtothe
model proposed by Goodmanand Rose[7], when thecharged
carriermobilities areunbalanced and theblendlayerthickness
islargerthanthecharacteristic chargecollectionlength,
space-charge limited (SCL) transport will be the dominant transport
mechanismofthephoto-carriersintheBHJ.Thisleadstoa
one-halfpowerphotocurrentvs.effectivevoltage(Jph–Veff)relationship
aroundthe operating point ofthe BHJsolarcells, reducing the
changeinthephotocurrentwithrespecttoappliedvoltage,and
thefillfactor (FF)aswell.Althoughvariousmethodshavebeen
proposedtoenhancetheP3HTordering,andthustoimprovethe
balanceincarriermobilities[3,4,8,9],theresultedvertical
com-positiongradientofPCBMandP3HTintheBHJisunfavorablefor
thechargeextraction[10].Therefore,thereisatradeoffbetween
theFFandblendlayerthickness,whichisrelatedtoabsorption,
forconventionalBHJsolarcells.Althoughtheuseofhigh
mobil-itymaterials[11]andmodifiedcathodestructurewithinterlayer
[12,13]helpreducetheSCLeffects,theuseofconventionaland
well-knownmaterialsisstilladvantageoustoelucidatetheorigin
ofFFdegradation.Inthispaper,weproposedaschemetoreduce
theSCLeffectinBHJsolarcellsbyintroducingaP3HT-richlayer
betweentheactiveblendlayerandthebottomanode.Wefound
thattheSCLeffectscanindeedbereduced,leadingtoanenhanced
FF.Ourresultsthussuggestthepossibilitytoincreasethe
absorp-tionlayerthicknesswhilemaintainingahighFF.
0379-6779/$–seefrontmatter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.synthmet.2012.09.007
Fig.1. LayerstructurefordeviceP3IandIusedinthisstudy.
AccordingtoGoodmanandRose[7],thephotocurrent(Jph)due
todoubleextractionofelectron–holepairsfromdeviceswith
non-injectingcontactsatloweffectivevoltage(Veff)isgivenby
Jph=gel[1b+b]
Veffss l2 (1)whereg,e,l,s,and
sdenotethephoto-carriergenerationrate,theelectroniccharge,thelayerthickness,thelifetimeandthe
mobil-ityofslowcarriers,respectively,andb=ss/ffistheratioofthe
productofcarriermobilityandlifetimeofslowcarrierstothatof
fastcarriers.Sincethecontactsforpracticalorganicphotovoltaic
devicesarenotcompletelynon-injecting,wefollowBlometal.to
reducephenomenallythecarrierinjectionbysubtractingthe
mea-suredilluminatedcurrent(Jlight)fromthemeasureddarkcurrent
(Jdark)[14].Thephotocurrent(Jph)andtheeffectivevoltage(Veff)
inEq.(1)arethusgivenbyJph=Jdark−JlightandVeff=V0−V,
respec-tively,whereV0,thecompensationvoltagein[11],istheapplied
voltage(V)atwhichJdarkequalsJlight.
AthigherVeff,orequivalentlyatVlowerthanopen-circuit
volt-age(Voc),theJ–Vcharacteristicexhibitsatransitiontoaone-half
powerlawoftheeffectivevoltageduetoSCLeffects,andthe
tran-sitionvoltageisgivenby
v
tr= l 2b24(1−b)ss (2)
AtevenhigherVeff,whichcorrespondstoreverseappliedbias,all
photo-carriersareextractedfromtheactivelayer.Asaturated
pho-tocurrentwhichisindependentofvoltagewillresult.Thesaturated
photocurrentisgivenby
Jph=egl (3)
InorderfortheFFtoincrease,thehalf-powerregionneedstobe
reduced,requiringalargerbinEq.(2).Onewaytoincreasethe
bistoincreasetheeffectivecarrierlifetimefortheslow
photo-carriers. Since the photo-carrier recombination in PSCsfollows
Langevinrecombination,anincreasedeffectivecarrierlifetimecan
beachievedbyreducingtheconcentrationoftheothertypeof
car-riers.Wetherefore addbetweentheactiveblendlayerandthe
bottomanodeaP3HT-richlayer,withinwhichtheelectron
concen-tration,andthustherecombinationaswell,isreduced.Theweight
ratioofP3HTandPCBMfortheP3HT-richlayeris3:1andthatofthe
activelayerforthecontroldeviceis1:1.Thestructuresforthe
con-troldevice(I)andthedevicewithinsertionofP3HT-richlayer(P3I)
areshowninFig.1.ThislayerstructurefordeviceP3Iissimilarto
thatinvestigatedbyLiangetal.[15],albeitwiththeP3HTelectron
blockinglayerreplacedbyaP3HT-richlayerinsertedforreduced
carrierrecombination.However,aswillbeanalyzedindetailsin
whatfollows,ourdevicewilloperatedifferentlyduetothethicker
activelayer.
2. Experiment
The devices (I and P3I) were prepared as follows. Indium
tin oxide (ITO) coated glass was cleaned by acetone and
iso-propanol toremove residual organicmaterials. UVO treatment
formorehydrophilicsurfacefollowed.Holetransportingmaterial,
PEDOT:PSS(AI4083),wasspunonITOsubstrateswithathickness
of45nm.Thedevicewasheatedonahotplateat200◦Cfor5min
inairforadriedfilm.WethenloadedthePEDOT-coatedITO
sub-stratesintoagloveboxinnitrogenambient.FordeviceP3I,amixed
solutionofP3HTandPCBM(P3HT:PCBM=3:1inweight)intoluene
withaP3HTconcentrationof8.5mg/mlwasspunat3000rpmfor
30stoachieveaP3HT-richlayerof70nmthickness.Toluenerather
thanmorecommonlyemployed1,2-dicholobezene(DCB)wasused
inthisworksinceitismorecompatiblewiththefabricationprocess
ofthebufferlayertechnique[16]whichweusedtocircumventthe
interlayermixingproblem.
The devicewasannealed ona hotplateat 140◦Cfor 10min
to remove the residual solvent and to make it more resistant
againsttheinterlayermixing.Following[16],afterspinningthe
1,2-propyleneglycolasbufferliquid,solutionwhichconsistsofP3HT
andPCBMatweightratioof1:1intoluenewithaP3HT
concentra-tionof17mg/ml,wasspunat1000rpmfor30sontheliquidsurface
immediatelytoformamulti-layereddevice(deviceP3I)withatotal
activelayerthicknessof300nm.Thefilmwasannealedagainona
hotplateat140◦Cfor10min.Forthesingle-layeredcontroldevice
(deviceI),whichconsistedofonlyablendinglayerofP3HTand
PCBMwithweightratioof1:1,solutionwhichconsistsofP3HTand
PCBMatweightratioof1:1intoluenewithaP3HTconcentration
of17mg/ml,wasspunat550rpmfor30s,aconditionwhichwas
optimizedtoachieveanequalactivelayerthickness(300nm)for
bothdevices.Aheattreatmentwasthencarriedoutonahotplate
at140◦Cfor20min,whichequalsthetotalannealingtimeforthe
multi-layereddevice.Subsequently,atopmetalelectrode
consist-ingofcalcium(50nm)andsilver(80nm)wasthermallydeposited
underapressureof2.3×10−6Torr.Theareaofactivedevicewith
thedepositedtopelectrodewas4mm2.
3. Characterization
TheJ–V characteristicsofthedevices weremeasuredwitha
Keithley2400sourcemeasurementunitusinga100mW/cm2AM
1.5Gsolarsimulator(San-eiElectric,XES-301S).Theintensityof
theincidentsolarilluminationwascalibratedbyasilicon
photo-diode(HAMAMATSUS1337-BR).Thelightintensitywasreduced
byneutraldensityfilters(Thorlab)toachievedifferent
illumina-tionlevels.Theincidentphoton-to-electronconversionefficiency
(IPCE)wasconductedusingameasurement system(model
QE-R)builtbyEnliTechnologyCo.,Ltd.Wide-spectrumlightsource
waschoppedanddiffractedintoseparatedmonochromatic
nar-rowbands,andeachofwhichwasprojectedontothedeviceunder
testing.Anotherwide-spectrumlightsourcewasusedasoptional
lightbias.Thephotocurrentgeneratedbyincidentmonochromatic
light wasconverted and amplifiedtoanACvoltagebya
trans-impedanceamplifier, whichwasthen measuredby aDAQ card
(NationalInstrument)atthechopperfrequency.
4. Resultsanddiscussion
Theabsorptionspectraforactivelayerofbothdevices(Iand
P3I) are shown in Fig. 2. Because the thickness of the active
Table1
ThedeviceparametersincludingJsc,Voc,FF,PCE,Vmax,JmaxandV0fordevicesP3IandI.
Jsc(mA/cm2) Voc(v) FF(%) PCE(%) Vmax(v) Jmax(mA/cm2) V0(v)
P3I 9.96 0.63 48.8 3.12 0.46 6.78 0.74
I 9.01 0.63 41.6 2.35 0.42 6.59 0.675
Fig.2.TheabsorptionspectrumoftheactivelayerfordeviceP3IandI.
overlapasexpected.Fig.3(a)showsthemeasureddarkcurrent
den-sity(Jdark)andtheilluminatedcurrentdensity(Jlight)asafunctionof
appliedvoltage(V)forbothdevicesundersimulatedAM1.5Glight
illuminationat100mW/cm2.Inaddition,thecompensation
volta-ges(V0)forbothdevicesarelabeled.Thephotovoltaicparameters
forbothdevicesaresummarizedinTable1.FromTable1,the
inser-tionoftheP3HT-richlayerindeviceP3IhasalittleeffectonVoc,as
comparedtodeviceI.ThisisnotsurprisingsinceVocdependseither
ontheelectrodemetalworkfunctionorthedonor/acceptorenergy
levels,whichareallidenticalforbothdevices.However,the
short-circuitcurrentdensityJsc (FF)increasesbyapproximately10.5%
(17.3%)from9.01mA/cm2(41.6%)to9.96mA/cm2(48.8%),yielding
animprovedPCEby32.3%from2.35%to3.12%.Sincethe
absorp-tionremainsunchanged(Fig.2)forbothdevices,theimprovements
shouldarisefromtheenhancedcarrierextractionefficiency.From
Fig.3(a),thedarkcurrentdensityofdeviceP3Iismorethanthatof
deviceIatforwardbias,revealingthatthecarrierinjectionfromthe
electrodeisnotsuppressedfordeviceP3IascomparedtodeviceI.
Inaddition,thedarkcurrentdensityofdeviceP3Ishowsanorderof
magnitudehigherthanthatofdeviceI.Thereasonofthisincreasein
darkcurrentmaybeattributedtothebufferlayerprocessinwhich
high-boilingpoint(188◦C)1,2-propyleneglycolbufferwasusedto
circumventtheinterlayermixingprobleminpreparingdeviceP3I.
Theresidual1,2-propyleneglycolaftertheannealing(at140◦C)
seemstoactaschemicaldopantsandincreasesthereversedark
current.Amorebalancedinjectionwhichenhancesthe
recombina-tionamongtheinjectedcarriersmayalsoplayarole.Thereforethe
improvedcarrierextractionefficiencyfordeviceP3Ishouldnotbe
attributedtosuppressedcarrierinjection,whichcancauseslossin
photocurrentduetotherecombinationbetweenthephoto-carriers
andtheinjectedcarriers.
Forbettercomparison,thedarkcurrentdensity(Jdark)vs.applied
voltage(V)characteristicsforbothdevicesareplottedinsemi-log
scaleinFig.3(b).Itisclearthat,whilethereisonlyslight
differ-encebetweenthecurrentforbothdevicesatforwardbias,thedark
currentofdeviceP3IishigherthanthatofdeviceIbyanorderof
magnitudeatreversebias.Thisisdifferentfromwhatwasobserved
in[15],inwhichthedarkcurrentissuppressedatreversebiasbythe
insertedP3HTelectronblockinglayer.Theabsenceofsuppressed
carrierinjectionfromtheITOanodecanbeunderstoodbynoting
thattheinsertedP3HT-richlayerin ourdevice(P3I)consistsof
PCBM,whichallowsforelectrontransport.OurinsertedP3HT-rich
layer,however,revertstheconventionallyobservedunfavorable
compositionprofile,whichexhibitsahigherconcentrationofPCBM
attheITOside,toafavorableonethatfacilitatescarriercollection.In
addition,aswillbeshowninwhatfollows,theinsertionofa
P3HT-richlayerreducescarrierrecombination.Therefore,increasedJsc
wasobtainedeventhoughtheabsorptionremainedunchanged.
Thephotocurrentdensities (Jph)vs. effectivevoltage(Veff)in
log–logscaleareplottedinFig.4andthemaximumpowerpoints,
VmaxandImaxinTable1,arealsomarked.WealsoincludeinFig.4
auxiliarylineswithslope1(0.5),whichrepresentbalanced(SCL)
transportforphoto-carriers.AsshowninFig.4theSCLdominated
region,whichfollowstheauxiliarylineofslope0.5,isnarrower
fordeviceP3IthanthatfordeviceI.Sincethemaximumpower
pointsarelocatedintheSCLregion,VmaxandImaxfordeviceP3Iare
largerthanthosefordeviceI(Fig.4).TheimprovementinPCEand
FFfordeviceP3IisthusattributedtoreduceSCLeffectswiththe
insertionoftheP3HT-richlayer.Itisalsonotablethatcurrent
satu-rationathigheffectivevoltage,whichrepresentsthephoto-carrier
generationrateinEq.(3)[7],isobservedforbothdevicesandthe
saturationphotocurrentdensitiesare12.5mA/cm2and12mA/cm2
Fig.3.(a)Theilluminatedcurrentdensity(Jlight)underAM1.5Gat100mW/cm2anddarkcurrentdensity(Jdark)vs.voltage(V)fordeviceP3IandI.Thecompensationvoltage (V0)islabeledfordeviceP3IandI;(b)Jdark–Vcharacteristicsplottedinsemi-logscale.
Fig.4. Thephotocurrentvs.effectivevoltagecharacteristicsinlog–logscale.
fordeviceP3IandI,respectively.Thephoto-carriergenerationrate
exhibitsonly4.17%increasefordeviceP3Iascomparedtodevice
I,andcannotaccountforthe32.8%enhancementinPCE. Thisis
againdifferentfromwhatwasobservedinRef.[15],inwhichthe
improvedPCEresultsmostlyfromincreasedphoto-carrier
genera-tionrate.
TounderstandthedifferencebetweenourresultsandRef.[15],
weestimatethechargecollectionlengthwhichcharacterizesSCL
transport.Typicalreportedvaluesforcarrierlifetime(10s[17])
andholemobility(1.5×10−8m2/Vs[18])areusedandacharge
col-lectionlengtharound300nmat0.6VwasobtainedforP3HT:PCBM
BHJPSCs.While thisestimatedvalueisclosetotheactivelayer
thicknessinourdevices,itissignificantlylargerthanthat(175nm)
inRef.[15].ThedeviceinRef.[15]operatesthereforeinthelinear
regionandtheimprovementinPCEresultsmostlyfromincreased
photo-carrier generation rate. On the other hand, our devices
employmuchthickeractivelayersandtheenhancementinPCE
withinsertedP3HT-richlayerarisesfromreducedSCLeffects.
Itisremarkablethatthetransitionfromthebalancedtransport
regiontotheSCLregionisdeterminedbythetransitionvoltageVtr
giveninEq.(2).AnincreasedvalueofbinEq.(2)leadstoalargerVtr,
extendsthelinearregionandthusincreasesthePCEandFF.From
Fig.4,Vtrincreasesby34.8%from0.155to0.209V.Thisincrease
inVtristentativelyattributedtotheincreaseintheeffective
car-rierlifetimeduetoreducedrecombination.Assumedthatf
f,theproductofcarriermobilityandlifetimeofthefastcarrierswhich
areelectronsinourcase,remainsunchangedandb1,VtrinEq.(2)
dependslinearlyonb.Therefore,aroughestimateof34.8%increase
inbwiththeinsertionoftheP3HT-richlayercanbeobtained.
TofurtherinvestigatetheeffectsoftheinsertedP3HT-richlayer
onSCLtransport,theshort-circuitcurrentdensitiesunderdifferent
illuminationlevelsweremeasuredandtheresultswereplottedin
log–logscaleinFig.5.AsshowninFig.5,deviceIexhibitsapower
dependenceof0.72±0.04,whichisclosetothepredictedvalueof
0.75forSCLphotocurrent[7].Ontheotherhand,anexponentof
0.94±0.06wasobtainedfordeviceP3I,indicatingalleviatedSCL
effects.ThisresultisconsistentwiththeobservationinFig.5,
cor-roboratingthattheinsertionofP3HT-richlayerreducesSCLeffects.
TheIPCEofdeviceIandP3Iwithandwithoutaone-sunDClight
biaswereplottedinFig.6(a)and(b).Sincebothdevicesexhibit
nearlyidenticalabsorption(Fig.2),enhancedquantumefficiency
observedindeviceP3I,ascomparedtodeviceI,indicatesimproved
chargecollectionefficiency.Itisalsointerestingtonotethatthe
Fig.5.Theshort-circuitcurrentdensityvs.incidentlightintensityinlog-logscalefor(a)deviceP3Iand(b)deviceI.
Fig.7.(a)ExtractedintrinsicJ–VcurvesfordeviceI(opencircle)andP3I(solidsquare);extractedJ–Vcurvesanditsdecompositionintoalinearandashiftedquadratic componentfor(b)deviceP3Iand(c)deviceI.In(b)and(c)averticaldashedlinewasshowntodenotetheshifted(transition)voltage,largerthanwhichthequadraticSCL currenttakeseffect.
Table2
TheextractedparametersincludingRs,RshandJphfordevicesP3IandI. Rs(cm2) Rsh(cm2) Jph(mA/cm2)
P3I 14.6 407 10.4
I 9.26 312 9.44
IPCEfordeviceIshowsmuchstrongerdependenceontheDClight
biasthan deviceP3I.Thiscanbeunderstoodbynotingthatthe
DClightbiascontributesmorebackgroundcarrierswhichleadto
segregatedchargedistributionnearbothsidesinthedevices.This
chargesegregationcanscreentheelectricfieldinthebulkregion.
FordeviceswithdominantSCLtransport,suchasdeviceI,reduced
electricfieldinthebulkregionleadstodecreasedcollection
effi-ciency,andthusdecreasedIPCE.Ontheotherhand,sincetheSCL
effectsindeviceP3Iarealreadymuchreduced,littlechangeinIPCE
wasobservedwithDClightbias.
Althoughthecontactstotheactivelayerarecrucialtothe
car-rierextractionaswellasinjection[13]andaffectinturnthecarrier
concentrationwithintheactivelayer,SCLeffectsareintrinsically
bulkeffectswithwhichthespacechargemodulatesthebulk
elec-tricfield.Itwouldthereforeberevealingtoremovetheeffectsdue
tothecontacts,whichmanifestmostlyasparasiticseriesresistance,
andtostudytheintrinsiccurrent–voltage(I–V)characteristicofthe
devices.Wefollowedaprocedurewhichweproposedpreviously
forextractingtheintrinsicI–Vcharacteristicsandtheseries
resis-tancewithoutassumingtheI–Vfunctionalform[19].Inshort,the
extractedI–VrepresentsthebestfitbulkI–Vrelationship,assuming
constantOhmic(linear)I–Vcharacteristicforthecontacts.The
pro-cedurerequiresI–Vmeasurementswithtwoslightlydifferentlight
illuminationlevels.Besidestheoneswithacalibratedsolar
simula-toratAM1.5G(100mW/cm2),theotherI–Vcurveswereobtained
withamicroscopecoverglassplacedontopoftheOSCs,leading
toalightattenuationofabout8%.Theroot-mean-squareerroris
calculatedfortherangefromVD=0toopen-circuitvoltage(Voc)
inthemeasuredI–Vcharacteristicwithlesslightillumination.The
detailedcalculationprocedurecanbefoundin[19].
Theextractedparametersincludingseriesresistance(Rs),shunt
resistance(Rsh)andphoto-inducedcurrentdensity(Jph)were
sum-marizedinTable2.Apparently,theRshofdeviceP3Iisincreasedas
comparedtothatofdeviceI,indicatingreducedrecombinationof
photo-carrierswiththeinsertionoftheP3HT-richlayer.
Fig. 7(a) shows the extracted intrinsic J–V curves for both
devices.Asitwasshownin[19]thattheintrinsiccurrentdensity
ofOSCsatlowforwardbiascomprisesofalinearandaquadratic
component,whichweretentativelyattributedtohoppingandSCL,
respectively,wesubtractthelinearcomponentfromtheextracted
J–VandfitthesubtractedJ–Vwithashiftedquadratictermaround
thevoltagethatcorrespondstoVmax;theresultswereshownin
Fig.7(b)and(c)fordeviceP3IandI,respectively.FromFig.7(b)and
(c),itwasclearthatSCLeffectsappearatlowerbiasfordeviceIthan
deviceP3I.Itisalsounderstandablethatbothvoltagesarelarger
thanthoseobtainedpreviouslyfromthemeasuredI–Vcurves,since
anegativevoltagedropdevelopsacrossRswiththeextracted
pho-tocurrentforthemeasuredI–V.
5. Conclusion
Insummary,ourresultsdemonstratedthatbytailoringthe
com-positionalprofileofblendlayerwithaninsertedP3HT-richlayer,
theeffectivecarrierlifetimecanbeincreasedandthedetrimental
SCLeffectscanbealleviated.Withourresults,itisexpectedthat
theadditionofinterlayers,preferablywithhighercarriermobility
forbettercarrierextractionandwithlessheterojunctionand/or
largerbarriertoreducethecarrierconcentrationoftheothertype
toreducerecombination,betweentheactivelayerandthe
elec-trodeswillenhancetheFF.Itisthereforepossibletoincreasethe
FFandthePCEfordeviceswithathickactivelayerwithasuitably
designedverticalcompositionprofileorlayerstructure.Ourresults
alsoshowthatDClightbiasisimportanttocharacterizethe
operat-ingcharacteristicsofPSCswhenSCLeffectsdominatephoto-carrier
carriertransportneartheoperatingpoints.
Acknowledgment
ThisworkwassupportedbytheNationalScienceCouncil(NSC)
References
[1] F.C.Krebs,SolarEnergyMaterialsandSolarCells93(2009)394.
[2] F.C.Krebs,M.Jørgensen,K.Norrman,O.Hagemann,J.Alstrup,T.D.Nielsen,J. Fyenbo,K.Larsen,J.Kristensen,SolarEnergyMaterialsandSolarCells93(2009) 422.
[3] G.Li,V.Shrotriya,J.Huang,Y.Yao,T.Moriarty,K.Emery,Y.Yang,Nature Materials4(2005)864.
[4]W.Ma,C.Yang,X.Gong,K.Lee,A.J.Heeger,AdvancedFunctionalMaterials15 (2005)1617.
[5] A.J.Moulé,K.Meerholz,AdvancedMaterials20(2008)240.
[6] J.Xue,B.P.Rand,S.Uchida,S.R.Forrest,JournalofAppliedPhysics98(2005) 124903.
[7] A.M.Goodman,A.Rose,JournalofAppliedPhysics42(1971)2823.
[8] J.Peet,J.Y.Kim,N.E.Coates,W.L.Ma,D.Moses,A.J.Heeger,G.C.Bazan,Nature Materials6(2007)497.
[9]Y.M.Chang,L.Wang,JournalofPhysicalChemistryC112(2008)17716. [10]M. Canpoy-Quiles,T. Ferenczi,T.Agostinelli, P.G.Etchegoin, Y.Kim, T.D.
Anthopoulos,P.N.Stavrinou,D.D.C.Bradley,J.Nelson,NatureMaterials7(2008) 158.
[11]S.Fabiano,Z.Chen,S.Vahedi,A.Facchetti,B.Pignataroc,M.A.Loi,Journalof MaterialsChemistry21(2011)5891.
[12]J.Wagner,M.Gruber,A.Hinderhofer,A.Wilke,B.Bröker,J.Frisch,P.Amsalem, A.Vollmer,A.Opitz,N.Koch,F.Schreiber,W.Brütting,AdvancedFunctional Materials20(2010)4295.
[13]Z.He,C.Zhong,X.Huang,W.Y.Wong,H.Wu,L.Chen,S.Su,Y.Cao,Advanced Materials22(2011)4636.
[14] V.D.Mihailetchi,L.J.A.Koster,J.C.Hummelen,P.W.M.Blom,PhysicalReview Letters93(2004)216601–216611.
[15] C.W. Liang, W.F. Su, L. Wang, Applied Physics Letters 95 (2009) 133303.
[16] S.R.Tseng,S.C.Lin,H.F.M.Meng,H.H.Liao,C.H.Yeh,H.C.Lai,S.F.Horng,C.S. Hsu,AppliedPhysicsLetters88(2006)163501.
[17] G.G.Belmonte,P.P.Boix,J.Bisquert,M.Sessolo,H.J.Bolink,SolarEnergy Mate-rialsandSolarCells94(2010)366.
[18] L.J.A.Koster,V.D.Mihailetchi,P.W.M.Blom,AppliedPhysicsLetters88(2006) 093511.
[19] M.K.Lee,J.C.Wang,S.F.Horng,H.S.Meng,SolarEnergyMaterialsandSolarCells 94(2010)578.