1

Supportinginformation

IodideMediatedReductiveDecompositionofDiazoniumSalts:

TowardsMildandEfficientCovalentFunctionalizationofSurface-

SupportedGraphene

YuanzhiXia,aCristinaMartin,b,cJohannesSeibel,aSamuelEyley,cWimThielemans,cMarkvanderAuweraer,a

KunalMalia*andStevenDeFeytera*

aDepartmentofChemistry,DivisionofMolecularImagingandPhotonics,KULeuven,Celestijnenlaan200F,

B-3001Leuven,Belgium,bDepartamentodeQuímicaFísica,FacultaddeFarmacia,UniversidaddeCastilla-La

Mancha,02071Albacete,Spain,cDepartmentofChemicalEngineering,SustainableMaterialsLab,KULeuven,

campusKulakKortrijk,EtienneSabbelaan53,8500Kortrijk,Belgium

Correspondingauthor:kunal.mali@kuleuven.be,steven.defeyter@kuleuven.be

Contents:

1. Experimentalmethods.

2. Aschematicshowingthefunctionalizationprocess(FigureS1).

3. Comparisonoftheefficiencyofcovalentgraftingachievedinthisworkwithsomeliteraturereports

togetherwithcomparisonoftheexperimentalconditions(TableS1).

4. RamanspectraoffunctionalizedSLGondifferentsubstrates(FigureS2).

5. XPSsurveyspectraofpristineandfunctionalizedSLG/SiO2(FigureS3).

6. Dataderivedfromcarbon1shighresolutionXPSspectrumoftheSLG/SiO2samples(TableS2).

7. Highresolutionnitrogen1sXPSspectrumofNBDfunctionalizedSLG/SiO2(FigureS4).

8. Datafornitrogen1sXPSspectrumofNBDfunctionalizedSLG/SiO2(TableS3).

9. AFMimagesshowingtheedgeoffunctionalizedSLG-Siandthecorrespondingtopographicalline-

profilesreflectingthelayerthicknessoffunctionalizationlayer(FigureS5).

10. AdditionalAFMandSTMdataforthedependenceofreactiontimeoncovalentgrafting(FigureS6).

11. RamanspectrumofthepristineHOPGsurfacepriortofunctionalization(FigureS7).

12. RepresentativeAFMdatashowinghowtheaveragelayerthicknesswasobtained(FigureS8).

13. Comparisonoflayerthicknessmeasuredacrossunfunctionalizedandscratchedareas(FigureS9).

14. AFMdataforthefilmsobtainedusinglongreactiontimes(30min)(FigureS10).

15. STMandRamandataforthespontaneousreactionofNBDwiththeHOPGsurface(FigureS11).

16. AdditionalAFMdatafortheconcentrationdependenceofcovalentgrafting(FigureS12).

Electronic Supplementary Material (ESI) for Nanoscale.This journal is © The Royal Society of Chemistry 2020

2

17. VariationinthesurfacecoverageofthecovalentlygraftedfilmasafunctionofNBDconcentration

obtainedfromAFMdata(FigureS13).

18. AdditionalAFMdataforthe[KI]/[NBD]solutionmoleratiooncovalentgrafting(FigureS14).

19. Characterizationofthereactionmixture(NBD+KI)usingUV-Visspectroscopy(FigureS15).

20. ThetimeevolutionoftheUV-visabsorbanceofthereactionmixtureatdifferent[KI]/[NBD]ratiosin

aqueoussolution(FigureS16).

21. Time-dependentevolutionoftheabsorbanceofNBDinaqueoussolution(FigureS17).

22. UV-VisabsorptionspectrumofaqueousNBDasafunctionoftime(monitoredoveraperiodof1.5

hours)(FigureS18).

-------------------------------------------------------------------------------------------------------------------------------------

Experimentalsection:

Materials.4-Nitrobenzenediazonium(NBD)tetrafluoroborate(97%)andpotassiumiodide(KI)(99.9

%)werepurchasedfromSigma-Aldrich.Allchemicalswereusedwithoutfurtherpurification.Highpurity

water(Milli-Q,Millipore,18.2MΩ·cm)wasusedforall theexperiments.Thehighlyorderedpyrolytic

graphite (HOPG, gradeZYB,AdvancedCeramics Inc., Cleveland,USA) substrateswere freshy cleaved

beforeuse.CVD-growngraphenesamples(1×1cm2)transferredtoSi++/SiO2(300nm),SLG/Auand

SLG/CuwereobtainedfromGrapheneaandwereusedasreceived.

Covalent functionalization experiments. Appropriate amounts of the diazonium salt and KI were

dissolvedinMilli-Qwatertopreparethesolutionswiththedesiredmolarconcentration.Thereaction

mixturewaspreparedpriortousebymixing100µLdiazoniumsaltsolutionwith100 µLKIsolutionin

aglassvial.Immediatelyaftermixing,100 µLofthemixedsolutionwasdropcastedontofreshlycleaved

HOPG/graphenesurface.Afteragiventime,thereactionwasstoppedbyrinsingwiththeHOPGsurface

withethanolfollowedbywaterandtherinsingwasrepeatedthreetimes.TheHOPG/graphenesubstrates

weredriedunderastreamofargon.

AFMandSTMcharacterization.AFM imagingwasperformedwith aCypherES (AsylumResearch)

systemat32°Cintappingmodeattheair/solidinterface.OMCL-AC160TS-R3probes(springconstant

∼26N/m)witharesonancefrequencyaround100kHzwereused.AllSTMexperimentswereperformed

atroomtemperature(21–23°C)usingaPicoLE(Keysight)machineoperatinginconstant-currentmode

STMtipswerepreparedbymechanicalcuttingofPt/Irwire(80%/20%,diameter0.25mm).Scanning

ProbeImagingProcessor(SPIP6.3.5)softwarefromImageMetrologyApSwasusedforAFMandSTM

imageprocessing.Theexperimentswererepeatedin2–3sessionsusingdifferenttipstocheckforre-

producibilityandtoavoidexperimentalartefacts,ifany.

3

Ramanspectroscopy.Ramanexperimentswereperformedatroomtemperature(21−23°C)usingan

OmegaScopeTM1000(AIST-NT).LaserlightfromaHe–Nelaser(632.8nm)wasreflectedbyadichroic

mirror(Chroma,Z633RDC)andthenfocusedontothesamplesurfacebyusinganobjective(MITUTOYO,

BDPlanApo100×,N.A.0.7).Theopticaldensityatthesamplesurfacewasabout800kWcm−2.Raman

scatteringwascollectedwiththesameobjectiveanddirectedtoaRamanspectrograph(HoribaJY,iHR-

320)equippedwithacooledcharge-coupleddevice(CCD)cameraoperatingat−100°C(Andor,DU920P)

throughthedichroicmirror,apinholeandalongpassfilter(Chroma,HQ645LP).Accumulationtimefor

allspectrawas6s.ForfunctionalizedHOPG/graphenesamples,pointspectrawereobtainedonatleast

9differentlocationsthatareseparatedfromeachotherbyatleastamillimeter.

UV/Vis spectroscopy. The UV-Vis absorption spectra were recorded using a Lambda 950

spectrophotometerwithblockedbeamandblankcorrections.FormonitoringthereactionbetweenNBD

andKI,300µLeachofNBDandKIsolutionsweremixedtogetheratappropriatemoleratiosandthe

mixedsolutionwasquicklytransferredtoathinquartzcell(0.1cm×1cm×4cm).Thecellwasplaced

intotheUV-visspectrometertomonitorthereactionbycontinuallytakinganabsorptionspectrumevery

3minutes.

4

Fig.S1.AschematicshowingthecovalentfunctionalizationprotocolinthecaseofHOPG.Priortodrop

castingontotheHOPGsurface,theaqueousNBDandKIsolutionswithappropriateconcentrationswere

mixed inaglassvial.Themixedsolutionwas immediatelydropcastedon toa freshlycleavedHOPG

surface.Afteragiven time, thereactionwasstoppedbyrinsing thesurfacewithethanol followedby

waterandtherinsingwasrepeatedthreetimes.TheHOPG/graphenesubstratesweredriedusingargon

andthensubjectedtofurthercharacterization.Picturesinthelowerpartofthefigureshowtheevolution

ofnitrogengasbubblesimmediatelyafterdropcastingofthereactionmixture.Reactionconditions:10

mM NBD + 10 mM KI, reaction time = 10 minutes. (b, c) Representative AFM images for the NBD

functionalizedandpristineSLG/SiO2surfaces,respectively.

NBDKI

HOPG FunctionalizedHOPG

Rinsing

1. Ethanol2. H2O

T=0sec T=10sec T=2min T=4min T=8min T=10minHOPG Dropletremoved

Afterrinsinganddrying

Functionalized SLG/SiO2 Pristine SLG/SiO2

a)

b) c)

5

TableS1.ComparisonoftheID/IGvaluesobtainedforfunctionalizedofSLG/SiO2inthisworkwiththose

reported in the literaturetogetherwithreactionconditionsused.Seereferencesat thebottomof the

document.

Fig.S2.RamanspectraoffunctionalizedSLGofdifferentunderlyingsubstrates.Reactiontime=10min,

[NBD]=[KI]=10mM.(a)RamanspectraofpristineandfunctionalizedSLG/Cu.(b)Ramanspectraof

pristineandfunctionalizedSLG/Au.

XPSAnalysis

Spectra were recorded on a Kratos Axis Supra X-ray Photoelectron Spectrometer employing a

monochromat-edAlKα(hν=1486.7eV,150W)X-raysource,hybrid(magnetic/electrostatic)opticswith

a slot aperture, hemispherical analyser, multichannel plate and delay line detector (DLD) with an

emissionangleof0°relativetothesamplesurfacenormal.Surveyspectrawereacquiredwithapass

energyof160eVandhigh-resolutionspectrawith20eV.Theresultingspectrawereprocessedusing

CasaXPSsoftware.BindingenergywasreferencedtoAg3d5/2at368.21eVmeasuredonsputtercleaned

silver under the same analysis conditions, on the same day as the samples. Good electrical contact

betweenthegraphenelayerandthespectrometerwasensuredbyuseofsilverpaint“track”fromthetop

ofthesampletothesamplebar.Componentsinhighresolutionspectrawerefittedusingthe“LA(α,m)”

lineshapeforsymmetricpeaksandthe“LF(α,β,w,m)”line-shapeforasymmetricpeakscorresponding

ID/IG Reaction time Temperature Solvent EnviromentThis work 3 10 min RT Water Ambient conditions

Ref. 1 2,8 1 h 30 min and 15 min RT Ethanol Glovebox ( ˂0.1 ppm O2, ˂ 0.1 ppm H2O, Ar)Ref. 2 1,4 16h 30 min 30 ℃ Water Ambient conditionsRef. 3 1,2 3 Days RT 1,2-DME Ar purging/Strict exclusion of air or moistureRef. 4 1,2 20 h 150 ℃ THF Glovebox ( ˂0.1 ppm O2, ˂ 0.1 ppm H2O, Ar)Ref. 5 0,8 20 h RT CH3CN N2-purgedRef. 6 1,1 48 h RT DME Glovebox ( ˂0.1 ppm O2, ˂ 0.1 ppm H2O, Ar)

6

toanumericalconvolutionofLorentzianfunctions(withexponentsαandβforthehighbindingenergy

andlowbindingenergysides)withaGaussian(widthm)andinclusionoftail-damping(w)toprovide

finiteintegrationlimits.DetailsoftheselineshapefunctionsareavailableintheCasaXPSdocumentation

online.7

Empirical relative sensitivity factors supplied by Kratos Analytical (Manchester, UK) were used for

quantification. Use of these relative sensitivity factors does not account for any attenuation due to

overlayersorothersurfacecontaminationandassumesauniformdepthdistributionofelementswithin

theinformationdepthofthesample.Matrixeffectsarealsodiscounted.8,9Quotedstandarddeviations

resultfromMonteCarlosimulationsoftheerrorinpeakmodellinganddiscountallother(potentially

larger)sourcesoferror.

Duringfittingofthecarbon1sspectrathefollowingconstraintsweremadeinordertoprovideastable

andmeaningfulfittotheexperimentaldata:

1. ThelineshapeusedfortheC=Cgraphenepeakwasleftunchangedbetweentheunmodified

grapheneandNBDfunctionalizedgraphenesamples.

2. TheFWHMwasfixedforallsymmetriccomponentpeaksascribedtospeciesassociatedwith

thenitrophenylmoiety,oxidationorC-Cdefectsinthegraphenesubstrate.

3. TherelativeintensityoftheC-NO2andC=C(nitrophenyl)peakswerefixedtocorrespondwith

theratioexpectedfromthenitrophenylmoiety.

4. TheC-C(defect)componentbindingenergywasfixedat285.0eV.

Survey spectra of the samples (Fig. S2) revealed the expected carbon, nitrogen (for theNBDgrafted

sample),oxygenandsilicon.Additionally,alowconcentrationofironwaspresentinboththeunmodified

andNBDgraftedgraphenesamples.Theoriginofthisironspeciesisunknown.TheNBDgraftedgraphene

alsoshowedalowlevelofiodinecontaminationfromthegraftingprocess.

7

Fig.S3.XPSSurveyspectraofunmodifiedSLG/SiO2(black)andNBDfunctionalizedSLG/SiO2(red).

TableS2:Dataderivedfromcarbon1shighresolutionspectraofthegraphenesamples

Component Bindingenergy/eV FWHM/eV Area/% St.Dev./%

PristineSLG/SiO2

C=Cgraphene 284.1 0.84 90.56 0.73

C-O 286.2 0.92 0.68 0.10

π*←π 290.2 4.66 6.43 0.44

π*←π 294.3 3.77 2.34 0.91

NBDfunctionalizedSLG/SiO2

C=Cgraphene 284.2 0.80 44.92 0.46

C=Cnitrophenyl 284.8 1.30 24.66 0.25

C-C 285.0 1.30 4.06 0.18

C-NO2 285.6 1.30 4.93 0.05

C-O 286.2 1.30 8.41 0.25

C=O/C-NH3+ 287.6 1.30 1.94 0.18

O-C=O 288.7 1.30 1.15 0.22

π*←π 290.5 2.89 3.94 0.23

π*←π 292.9 4.87 6.00 0.65

8

Thenitrogen1shighresolutionspectrum(Fig.S3)oftheNBDfunctionalizedgraphenesampleshows

significant reduction of the nitro group to amines, protonated amines and other nitrogen containing

species. This has previously been linked to photochemical reduction of nitrobenzene by X-ray

irradiation,10however,testsofX-rayinducedreductiononmultiplesampleareasshowthatthiswasnot

asignificantcontributortothereducednitrogenenvironmentsinthisdataset.Thenitrogen1sdatais

summarizedinTableS3.

Fig.S4.Nitrogen1shighresolutionspectrumofNBDfunctionalizedSLG/SiO2.

TableS3.Datafornitrogen1sspectrumofNBDfunctionalizedSLG/SiO2.

Component Bindingenergy/eV FWHM/eV Area/% St.Dev./%

-NH2 399.8 1.43 20.63 1.68

-NH3+ 400.8 1.43 13.21 2.08

-NO 403.0 1.43 4.70 0.89

-NO2 405.7 1.43 61.46 2.16

9

Fig.S5.AFMimagesshowingtheedgeoffunctionalizedSLG-SiO2andthecorrespondingtopographical

line-profilesreflectingthelayerthicknessoffunctionalizationlayer.(a)Schematicillustrationoftheedge

of functionalizedSLG-SiO2. (b)Theedgeof functionalizedSLG-SiO2 characterizedbyAFM.Three line-

profilesperformedbasedontheamplifiededgeoffunctionalizedSLG-SiO2.(c)Thecorrespondingline-

profilesreflectingthelayerthicknessoffunctionalizationlayer.Theheightofthefunctionalizationlayer,

graphene and silica substrate indicate that the bottom of the empty areas (corrals) within the

functionalizationlayerisgraphene.

Fig.S6.AdditionalAFM(a)andSTM(b)dataforthedependenceofreactiontimeoncovalentgraftingof

NBD. [KI] =10mM, [NBD]=10mM. STMdata shows typical bright features that arise fromgrafted

moleculesevenwhenthedropletofthereactionmixtureiswashedawaywithinasecondortwoseconds

fromthesurfacehighlighting theefficiencyof the functionalizationprocess.Tunnelingparameters in

STM:(2s)Iset=0.1nA,Vbias=-0.6V;(10s)Iset=0.07nA,Vbias=-0.5V;(1min)Iset=0.1nA,Vbias=-0.5V;(6

min)Iset=0.05nA,Vbias=-0.82V.

10

Fig.S7.RamanspectrumofthepristineHOPGsurfacepriortofunctionalization.TheabsenceoftheD

bandindicatesthattheintrinsicdefectdensityonunfunctionalizedHOPGsamplesislow.

FigureS8.RepresentativeAFMdatasetshowinghowaverage topographicalheightof thecovalently

graftedfilmswasobtained.Typically,between3to5AFMimagesobtainedondifferentpartsofsame

samplewereconsideredforheightanalysis.Withintheseimages,heightprofilesweremeasuredacross

threedifferentlocationsandthemeasuredvalueswereaveraged.Theaveragethicknessofthefilmfor

thisdatasetis1.5 ±0.1nm.Reactionsconditionsforthisdataset:10mMKI,10mMNBD,reactiontime

=10min.TheAFMdataobtainedonallothersampleswastreatedusingthesameapproach.

~ 1.5nm

~ 1.5nm

~ 1.4nm

Image1

~ 1.5nm

~ 1.4nm

~ 1.4nm

Image3

1.5nm

1.5nm

1.6nm

Image2

11

Fig.S9.Comparisonbetweenthelayerthicknessobtainedusingtopographicalheightmeasuredacross

(a, b) unfunctionalized areas with no physisorbed deposits, (c, d) unfunctionalized areas with

physisorbeddepositsand(e,f)anareaobtainedviascratchingawaythefilmusingtheAFMtipincontact

mode(sameareaasthatinc).Reactionconditions:50mMKI,50mMNBD,reactiontime=10min.The

unmodified areas seen in panel (c) are possibly occupied by some physisorbed material. The

topographicalheightofthephysisorbeddepositisaround0.6nmasevidentfromheightprofileprovided

inpanel(d).

~ 2.6nm

Measuredacrossunfunctionalizedarea

~0.6nm

Measuredacrossscratchedarea

~ 2.6nm

~ 2.6nm

Measuredacrossunfunctionalizedarea

12

Fig.S10.ArepresentativeAFMimageshowingthatanincreaseinthereactiontimeto30minutesdoes

not lead to any further increase in the layer thickness than that obtained at 10minutes. The layer

thicknessforthissamplewasfoundtobearound1.5 ±0.1nmwhichiscomparabletothatobtained

usingshorterreaction(10min)timesusingthesameconcentrationofreagents.

Fig.S11.SpontaneousreactionofNBDwiththeHOPGsurface.(a)ArepresentativeSTMimageshowing

absence of grafted film on the HOPG surface. (b). Raman spectrum of the HOPG surface allowed to

spontaneouslyreactwithNBD(10mM)for10min.

1.5 ± 0.1 nm

13

Fig.S12.AdditionalAFMandSTMdataforthedependenceofNBDconcentrationonthecovalentgrafting

of NBD. [NBD]/[KI] = 1:1 and reaction time = 10 min. The STM images are provided for those

concentrationsforwhichAFMdataisprovidedinthemaintext.

Fig. S13. Variation in the surface coverage of the covalently grafted film as a function of NBD

concentrationasmeasuredfromAFMimages.KIwasusedinequimolaramount.Thereactiontimewas

keptconstantat10min.Thesurfacecoverageofthegraftedfilmswasobtainedbymeasuringtheareaof

the emptyunfunctionalized regions. Each value for surface coveragewasobtainedby estimating the

surfacecoverageoftheemptyareasfrom5AFMimages(2µm´2µm)andthencalculatingthesurface

coverageofthefilm.

AFM

STM

14

Fig.S14.AdditionalAFMdataforthe[KI]/[NBD]solutionmoleratiooncovalentgraftingonthesurface

ofHOPG.[NBD]=10mMandreactiontime=10min.

Fig.S15.UV-VisspectraforNBD,KIandapotentialby-productofthereaction1-iodo-4-nitrobenzene

(INB). The absorptionmaxima for triiodate (I3-)11 and 4,4’-dinitro-1,1’-biphenyl (DNBP)12 have been

reportedearlier.

15

Fig.S16.ThetimeevolutionoftheUV-visabsorbanceofthereactionmixtureatdifferent[KI]/[NBD]

ratiosinaqueoussolution.InallcasestheNBDconcentrationisfixedat1mMandKIconcentrationwas

variedfrom0.1mM(a)to7mM(e).

FigureS17.TimeevolutionoftheUV-visabsorbanceofNBDandKI.Thedashedlinerepresentsthebest

fitfor1storderkinetics(A-A0=B[1-exp-kt].Inthisequationkisthereactionrateconstantforthesystem

andBisdefinedasfunctionoftheopticalpathlength(l)andthedifferencebetweenthemolarextinction

coefficient(B=lc0(Ɛl-Ɛf).

200 250 300 350 4000,0

0,5

1,0

1,5

2,0

2,5

KI/NBD5mM / 1mM

t0= 0 mintf= 2.3 hours

Abs

orba

nce

Wavelength / nm200 250 300 350 400

0,00,51,01,52,02,53,03,54,04,55,05,56,06,57,0

KI/NBD7 mM / 1mM

t0= 0 mintf= 1.23 hours

Abs

orba

nce

Wavelength / nm

200 250 300 350 4000,0

0,5

1,0

1,5KI/NBD1mM / 1mM

t0= 0 mintf= 4 hours

Abs

orba

nce

Wavelength / nm200 250 300 350 400

0,0

0,5

1,0

1,5

2,0

2,5

KI/NBD3mM / 1mM

t0= 0 mintf= 3 hours

Abs

orba

nce

Wavelength / nm200 250 300 350 400

0,0

0,5

1,0

KI/NBD0.1 mM / 1 mM

t0= 0 mintf= 4 hours

Abs

orba

nce

Wavelength / nm

NBD

I-

0 4000 8000 12000 160000.00

0.05

0.10

0.15

a)

A 0-A

Time / s

pseudo primer orderwith graphiteA0-A= B[1-exp(-kt)]k=4.23*10-4 s-1

c2= 0.92

0 5000 10000 150000.00

0.02

0.04

0.06

0.08

0.10

b)

pseudo primer orderwithout graphiteA0-A= B[1-exp(-kt)]k=1.54*10-4 s-1

c2= 0.98

A 0-A

Time / s

0 1000 2000 3000 40000.00

0.05

0.10

A 0-A

Time / s

pseudo primer orderwith graphiteA0-A= B[1-exp(-kt)]k=6.1*10-4 s-1

c2= 0.98

c)

16

Fig.S18.AbsorptionspectrumofaqueousNBDasa functionof time(monitoredoveraperiodof1.5

hours).TheabsorbanceofaqueousNBDsolutiondoesnotchangemuchindicatingextremelyslowrate

ofdecompositionintheabsenceofeitherKIorgraphitepowder.

References:

1 T.Wei,M.Kohring,M.Chen,S.Yang,H.Weber,F.Hauke,andA.Hirsch,Angew.Chem.Int.Ed.,

2020,59,1–6.

2 Q.Wang,Z.Jin,K.Kim,A.Hilmer,G.Paulus,C.Shih,M.Ham,J.Sanchez-Yamagishi,K.Watanabe,

T.Taniguchi,J.Kong,P.Jarillo-HerreroandM.Strano,Nat.Chem.,2012,4,723–732.

3 J.Englert,C.Dotzer,G.Yang,M.Schmid,C.Papp,J.Gottfried,H.Steinruck,E.Spiecker,F.Hauke

andA.Hirsch,Nat.Chem.,2011,3,279–286.

4 T.Wei,O.Martin,S.Yang,F.Hauke,andA.Hirsch,Angew.Chem.Int.Ed.2019,58,816–820.

5 Z.Sun,C.Pint,D.Marcano,C.Zhang, J.Yao,G.Ruan,Z.Yan,Y.Zhu,R.Haugeand J.Tour,Nat.

Commun.2011,2,559.

6 G.Abellań,M.Schirowski,K.Edelthalhammer,M.Fickert,K.Werbach,H.Peterlik,F.Hauke,and

A.Hirsch,J.Am.Chem.Soc.2017,139,5175−5182.

7 N. Fairley, Lorentzian Asymmetric Lineshape,

http://www.casaxps.com/help_manual/manual_updates/LA_Lineshape.pdf, (accessed 2

September2017).

8 D.BriggsandJ.T.Grant,Eds.,SurfaceAnalysisbyAugerandX-rayPhotoelectronSpectroscopy,

IMPublications,Manchester,2003.

200 250 300 350 4000,0

0,5

1,0

1,5260

NBD aqueous solutiont0= 0 mintf= 1. 5 hours

Abs

orba

nce

Wavelength / nm

17

9 S.Hofmann,Auger-andX-RayPhotoelectronSpectroscopyinMaterialsScience,SpringerBerlin

Heidelberg,Berlin,Heidelberg,2013,vol.49.

10 K.Roodenko,M.Gensch,J.Rappich,K.Hinrichs,N.EsserandR.Hunger,J.Phys.Chem.B,2007,

111,7541–7549.

11. A.E.BurgessandJ.C.Davidson.J.Chem.Edu.2012,89,814-816.

12. Bio-RadLaboratories, Inc.SpectraBase;SpectraBaseCompoundID=639lhqsWLzpSpectraBase

Spectrum ID=KxKsK2Msvsa http://spectrabase.com/spectrum/KxKsK2Msvsa and Dorr, F.;

Gazis,E.,UVatlasoforganiccompounds,1966,1,D10/6.