a photo differential scanning calorimetric study Producing critical exponents from gelation for various photoinitiatorconcentrations; Progress in Organic Coatings

(1)

ContentslistsavailableatSciVerseScienceDirect

Progress

in

Organic

Coatings

j o ur na l ho me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / p o r g c o a t

Producing

critical

exponents

from

gelation

for

various

photoinitiator

concentrations;

a

photo

differential

scanning

calorimetric

study

Zekeriya

Do˘gruyol

a

_,

_Nergis

_Arsu

b,∗

_,

_Sevnur

_Keskin

_Do˘gruyol

b

_,

_Önder

_Pekcan

c,∗∗

a_Department_of_Physics,_Yıldız_Technical_University,_Davutpas¸_a_Campus,₃₄₂₂₀_Istanbul,_Turkey b_Department_of_Chemistry,_Yıldız_Technical_University,_Davutpas¸_a_Campus,₃₄₂₂₀_Istanbul,_Turkey c_Faculty_of_Arts_and_Science,_Kadir_Has_University,_Cibali,₃₄₃₂₀_Istanbul,_Turkey

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received27December2010

Receivedinrevisedform15October2011 Accepted12December2011

Available online 24 February 2012 Keywords: Criticalexponent Epoxyacrylate Photo-DSC Photoinitiatorconcentration

a

b

s

t

r

a

c

t

Photoinitiated radical polymerization of an 80wt% epoxy diacrylate (EA) and 20wt% tripropy-leneglycoldiacrylate(TPGDA)mixture withvarious 2-Mercaptothioxanthone(TX-SH)photoinitiator concentrationswasstudiedbyusingphoto-differentialscanningcalorimetric(Photo-DSC)technique. Photopolymerizationreactionswerecarriedoutunderthesameconditionsoftemperatureandlight intensity.Itwasobservedthatallconversioncurvesduringgelationatvariousphotoinitiator concentra-tionpresentnicesigmoidalbehaviorwhichsuggestsapplicationofthepercolationmodel.Thecritical time,wherepolymerizationreachesthemaximumrate(Rpmax)iscalledtheglasstransitionpoint(tg).

Thegelfractionexponents,ˇwereproducedfromtheconversioncurvesaroundtg.Theobservedcritical

exponentswerefoundtobearound0.55,predictingthatthegelsystemobeysthepercolationmodel. Rpmaxandﬁnalconversion(Cs)valueswerefoundtobeincreasedasthephotoinitiatorconcentration

wasincreased.Ontheotherhandtgvaluesdecreasedasphotoinitiatorconcentrationwasincreased,

1. Introduction

Photopolymerizableformulationsconsistofthreemajor com-ponents:apre-polymer,monomerasadiluentandaphotoinitiator. Inadditiontothesecomponents,additives(pigments,co-initiators, etc.)arealsooftenapartof thesesystems.Themostexpensive partoftheformulationsandplayingakeyroleinUV-curable sys-temsisaphotoinitiatorcapableofabsorbinglightoftheappropriate wavelengthandproducingreactivespecies,ions,orradicalspecies, whichareabletoinitiatepolymerizationofthemultifunctional monomersoroligomers[1,2].Photoinitiatedradical polymeriza-tionmaybeinitiatedbybondcleavage(TypeI)andH-abstraction type(TypeII) initiators[3].TypeIIphotoinitiatorsarebasedon compoundswhosetripletexcitedstatesreadilyreactwith hydro-gendonors,therebyproducinginitiatingradicals(Scheme1)[4–6]. Becauseof thebimolecular radicalgenerationprocess,theyare generallyslowerthanTypeIphotoinitiators,whichformradicals unimolecularly.

Typical Type II photoinitiators include benzophenone and derivatives,thioxanthones,benzyl,andquinones,whilealcohols,

∗ Correspondingauthor.Tel.:+902123834186;fax:+902123834134. ∗∗ Correspondingauthor.Tel.:+902125336532;fax:+902125336515.

E-mailaddresses:narsu@yildiz.edu.tr(N.Arsu),pekcan@khas.edu.tr(Ö.Pekcan).

ethers,aminesandthiolsareusedashydrogendonors.AmongType II photoinitiators, thioxanthone (TX) derivatives in conjunction withtertiaryaminesareefﬁcientphotoinitiatorswithabsorption characteristicsthat comparefavorablywithbenzophenones[7]. Wehavereported[8]theuseofathiolderivativeofthioxanthone (TX-SH)asaphotoinitiatorforfreeradicalpolymerization.Amajor advantageofthisinitiatorisrelatedtoitsonecomponentnature.It canserveasbothtripletphotosensitizer[9]andahydrogendonor [10].Thus,thisphotoinitiatordoesnotrequireanadditional co-initiator,i.e.,aseparatemolecularhydrogendonor.Themechanism ofthephotoinitiationisbasedontheintermolecularreactionofthe triplet3_TX-SH*_with_the_thiol_moiety_of_ground_state_TX-SH._The

resultingthiylradicalinitiatesthepolymerization(Scheme2). Therateofphotoinitiationisdirectlyproportionaltothesquare rootofthephotoinitiatorconcentration,anincreaseinthelevel ofphotoinitiator would beexpected toenhance thecurespeed offormulation. Severalreports claimthat thecurerateand the degreeofpolymerizationofaformulationwillatﬁrstincreasewith increasingphotoinitiatorconcentration,andafterpassingthrough amaximumoroptimumconcentration,theeffectwillrapidlyfall off.Thiseffectwilllargelydependupontheabsorbance charac-teristicsofthephotoinitiatorandthatofthecurableformulation [11].

AthigherphotoinitiatorconcentrationstheabsorbanceofUV lightappearstogeneratea higherconcentrationoffreeradicals

(2)

nearthesurfaceofthefilm.Thishigherconcentrationof photoini-tiatorandthereforefreeradicalsatthesurfaceisthoughttoblock sufficientenergyfrompenetratingfilm,preventingphotoactivation belowagiventhickness.Belowthislevelaninsufficientnumber offreeradicalsaregeneratedinordertoinitiateandsustainthe polymerizationprocess,andthereforetherateofpolymerization decreases.

Photoinitiatorswhichhaveahighmolarextinctioncoefficient wouldbeexpectedtogiverisetothisunevendistributionmore readilythanphotoinitiatorswhichhavealowextinctioncoefficient anddonotabsorbUVlightintensity.Thephotoinitiated polymer-izationofacrylatesandmethacrylatesisoneofthemostefficient processesfortherapidproductionofpolymericmaterialswithwell definedproperties.Thesematerialshavefoundwidespreaduseas coatings,imagingmaterials,photoresistsandpolymericmaterials formanyotherapplications.Thephotoinitiatorplaysakeyrolein UV-curablesystemsbygeneratingthereactivespecies,freeradicals orions,whichwillinitiatethepolymerizationofthe multifunc-tionalmonomersandoligomers[3–6,21].

Thebulk freeradicalcross-linking copolymerization(FCC) is dividedintothreestages:lowconversionstage,geleffectstageand glasseffectstage[12–14].Itwasobservedthatmonomer conver-sionﬁrstincreasesveryslowlybutthenitacceleratesbecauseof thegeleffect[15].Whenthereactiontemperatureislowerthanthe glasstransitionpointofthepolymertheglasseffectstageoccursas thelaststageofpolymerization.Theglasstransitiontemperature ofpolymersiscustomarilydeﬁnedasthetemperatureatwhich therelaxationtimeonthemonomerscalereaches100s[16]. Rad-icalchainpolymerizationsareoftencharacterizedbythepresence ofanauto-accelerationinthepolymerizationrateasthereaction proceeds[17].

NorrishandSmith[18] postulatedthat theincreased viscos-itycausedbymonomersbeingconvertedtopolymerresultedin adecreaseinthemobilityofthegrowingchains,makingitmore difﬁcultforthemtodiffusetogetherandterminate.Thetermgel effectwasusedduetothecharacteristicriseinviscosity accompa-nyingthedramaticincreaseinpolymerconversion[19].Burnett and Melville [19], Schulz and Harbort[20] each independently

Scheme 2.Photoinitiated free radical polymerization by using 2-Mercapto-thioxanthone(TX-SH).

averagedvalueforthecriticalexponentˇwasfoundtoobeythe percolationmodel,predictingtheuniversalbehaviorholdsnearthe glasstransitionpoint.Theproducedglasstransitionpointtg,

max-imumconversionrateRpmaxandﬁnalconversionvaluesCswere

foundtobestronglycorrelatedwiththeintroducedphotoinitiator concentration.

Ingeneralgelationprocesshasbeenknownasageometrical phasetransitionratherthanathermalphasetransitionandownsa universalclassbyitself.However,asfarasthecriticalphenomenon isconcernedbehaviorofgelationmaybemadeanalogoustosecond orderphasetransition.

Understandingthegelationphenomenonwasﬁrstintroduced byFloryandStockmayer[23,24]onaspeciallatticecalledBethe latticeonwhichtheclosedloopswereignored.Analternativeto theclassical theoryisthe latticepercolationmodel [25] where monomersarethoughttooccupythesitesofaperiodiclatticeand thechemical bondsascorrespondingtotheedgesjoiningthese sitesrandomlywithsomeprobabilityp,whichistheratioofactual numberofbondsthathavebeenformedbetweenthemonomers, tothetotalpossiblenumberofsuchbonds.Thegelpointcanbe identiﬁedwiththepercolationthresholdpc,where,inthe

thermo-dynamiclimit,theincipientinﬁniteclusterstartstoformandthe systemexhibitsviscoelasticrigidity[26,27].

Fromthepointoftheuniversality,thepredictionsofthesetwo theoriesaboutthecriticalexponentsforthegelationarequite dif-ferent.ThecriticalexponentˇforthegelfractionG(thestrength oftheinﬁnitenetworkinpercolationlanguage)nearthegelpoint, canbedeﬁnedasEq.(1),

G˛(p−pc)␤, p→p+c(abovepercolation) (1)

where the Flory–Stockmayer theory (so-called the classical or mean-ﬁeldtheory)gives,ˇ=1independentofthedimensionality, whilethepercolationstudiesbasedoncomputersimulationsgive ˇaround0.43inthree-dimension[25,28].Thesetwouniversality classesforgelationproblemareseparatedbyaGinzburgcriterion [29]thatdependsuponthechainlengthbetweenthebranchpoints aswellastheconcentrationofthenon-reactingsolvent.The vul-canizationoflonglinearpolymerchainsbelongstothemean-ﬁeld class. Criticalpercolation describesthepolymerizationof small multifunctionalmonomers[25,27].

2. Materialsandmethods 2.1. Materials

2-Mercaptothioxanthone[8](TX-SH)wassynthesized accord-ing tothe previously described procedure. Dimethylformamide (DMF, 99+%, Aldrich) was distilled over CaH2 under reduced

pressure.Epoxydiacrylate(EA)andTripropyleneglycoldiacrylate (TPGDA)wereobtainedfromCognisFrance.

(3)

Fig.1.HeatﬂowspectraofphotopolymerizationofEA/TPGDAwithvarious pho-toinitiator(TX-SH)concentrationsirradiatedat25◦_C_by_UV_light_with_an_intensity

of40mW/cm2_.

2.2. Photodifferentialscanningcalorimetry(Photo-DSC)

ThephotoinitiatedpolymerizationofEA/TPGDAinthepresence ofvarying(0.01,0.02,0.05,0.10,0.25,0.50and0.75wt%)TX-SH concentrationwasperformedinaPhoto-DSCsetup(TA-DSCQ100). UVlightwasappliedfromamediumpressuremercurylampwith 40mW/cm2_light _intensity_for₅_min._The_weight_of_the_samples

2±0.1mgwasplacedintoanopenaluminumliquidDSCpan.The measurementswerecarriedoutunderidenticalconditions(i.e.an isothermalmodeatroomtemperature(24.92◦C)andanitrogen ﬂowof50mL/min).

Thereaction heat liberatedin thepolymerizationis directly proportionaltothenumberofacrylatesreactedinthesystem.By integratingtheareaundertheexothermicpeak,theconversionof theacrylategroups(C)ortheextentofthereactionwasdetermined accordingtoEq.(2):

C= Ht

Htheory

0

(2) whereHtisthereactionheatevolvedattimetandH₀theoryis

thetheoreticalheatforcompleteconversion.Areactionheatforan acrylatedoublebondpolymerizationofHtheory

0 =86kJ/molwas

used[30].Therateofpolymerization(Rp)isdirectlyrelatedtothe heatﬂow(dH/dt)byEq.(3):

Rp=

dC dt

= (dH/dt) Htheory 0 (3)

3. Resultsanddiscussion 3.1. Photo-DSCresults

Photo-DSCmeasurements supply thereactiondatain which themeasuredheatﬂowcanbeconverteddirectlytotheultimate percentageconversionandpolymerizationrateforagiven formu-lation.Atypicalheatﬂowandrateofpolymerizationcurvesversus reactiontimefordifferentinitiatorconcentrationsarepresentedin Figs.1and2,respectively.

Conversioncurves produced fromFig.2 areshown in Fig.3 againstreactiontime.ItisseeninFigs.1–3,that,withincreasing initiatorconcentration,theextentofheatﬂow,therateof polymer-izationandconversionincreased.

Increasedinitiatorconcentrationsprovidehigherefﬁciencyfor initiation,leadingtomoreformedprimaryradicalsinthesolution. Thisenhancesboththepolymerizationrateandﬁnalconversion

Fig.2. RateofpolymerizationspectraofphotopolymerizationofEA/TPGDAwith variousphotoinitiator(TX-SH)concentrationsirradiatedat25◦CbyUVlightwith anintensityof40mW/cm2_.

ratio.Thereductionoftheinductionperiodisalsofavoredbya highconcentrationofphotoinitiator.

3.2. Kineticassessmentofphotopolymerization

Theeffectsofphotoinitiatorconcentrationonthe photopoly-merization kineticswerepresented as shown inFigs. 4–6.It is interestingtonotethattgvaluesdecreaseasthephotoinitiator

con-centrationincrease,followingtheoppositebehaviorofRpmaxand

ﬁnalconversion,Cs.

Onenaturallymayexpectthatthehigherphotoinitiator concen-trationshouldreducethetimerequiredtoachieveglassformation andtheirpercolationduringgelationprocess.Ontheotherhand, asexpected,bothRpmaxandCsincreaseasthephotoinitiator

con-centrationisincreased.

3.3. Criticalexponentsduringgelation

Theaboveexperimentalﬁndingscanbequantiﬁedintermsof gelationtheoriesasfollows:ingelationtheorytheconversion fac-tor,p,alonedeterminesthebehaviorofthegelationprocess,though pmaydependontemperature,concentrationofmonomers,and time.Ifthetemperature,lightintensityandconcentrationarekept constant,thenpwillbedirectlyproportionaltothereactiontime,

Fig.3.ConversionspectraofphotopolymerizationofEA/TPGDAwithvarious pho-toinitiator(TX-SH)concentrationsirradiatedat25◦_C_by_UV_light_with_an_intensity

(4)

Fig.4.Theeffectofphotoinitiator(TX-SH)concentrationontgvalueforEA/TPGDA.

Fig.5.Theeffectofphotoinitiator(TX-SH)concentrationontherateof photopoly-merizationforEA/TPGDA.

t.Thisproportionalityisnotlinearoverthewholerangeofreaction timebutitcanbeassumedthatinthecriticalregion,i.e.aroundthe criticalpoint|p−pc|islinearlyproportionaltothe|t−tc|.

There-fore,belowthecriticalpoint(i.e.,fort<tc)conversionmeasures

theweightaveragedegreeofpolymerization.Abovetc,however

Fig.6.Theeffectofphotoinitiator(TX-SH)concentrationontheﬁnalconversionof photopolymerizationforEA/TPGDA.

Fig.7. Doublelogarithmicplotoftheconversionversustimecurvesabovetgfor

variousphotoinitiator(TX-SH)concentrations.

conversionmeasuressolelythegelfractionG,thefractionofthe monomersthatbelongtothemacroscopicnetwork.

Nowourexperimentalﬁndingswereinterpretedbyconsidering thequasi-staticpropertiesofthegelneartheglasstransitionpoint inthelanguageofpercolation[31].Thedoublelogarithmicplot oftheconversionversus|t− tc|wasanalyzed,whichproducedthe

criticalexponent,ˇfromtheslopeofthestraightlineduringﬁtting thedatainFig.7.

Heretheimportantproblemwastheprecisedeterminationof theglasstransitionpointand thecriticalregion.Inparticular,a smallshiftintc resultsin alargeshiftin thecriticalexponent.

Suchalog–logplotrevealsthatdatashouldbeparticularlyaccurate nearthecriticalpoint.Usuallythecriticalpointcanthenbe deter-minedbyvaryingtcinsuchawayastoobtaingoodscalingbehavior

overthegreatestrangein|t−tc|,iftheexperimentsareperformed

againsttime.Thetimecorrespondingtothemaximumrateof poly-merizationwaschosenasthecriticaltime,tcwhichmaybenamed

astheglasstransitionpoint,tgforthephotoinitiatedgelationunder

consideration.Theplotoflog(conversion)versuslog|t−tc|above

tgforthegelationofEA/TPGDAwiththreedifferentphotoinitiator

concentrationsarepresentedinFig.8.

Theslopesofthestraightlinesproducedgelfractionexponent ˇatthegiveninitiatorconcentrations.The(valuescalculatedfor thegelationofEA/TPGDAinallinitiatorconcentrationsarelisted inTable1togetherwithtg,Rpmaxandﬁnalconversion,Cs.Here

ithastobenotedthattheaveragevalue(=0.56)ofthecalculated (valuesabovetgstronglysuggestthattheglassyregions

perco-lateduringphotoinitiatedgelformationforallthesamplesunder

Table1

Experimentallyobservedparametersmeasuredby “Photo-DSC”and calculated via“percolationtheory”forvariousphotoinitiatorconcentrationsduring diacry-late,EA/TPGDA,photopolymerizationat25◦CbyUVlightwithanintensityof 40mW/cm2_. [TX-SH](wt%) tga(s) 10−3×Rpmaxb(s−1) Csc(%) ˇd 0.01 10.6 7 34 0.52 0.02 6.6 8 41 0.56 0.05 3.7 12 44 0.57 0.10 3.2 18 52 0.57 0.25 2.5 22 59 0.57 0.50 2.4 25 62 0.57 0.75 2.7 27 69 0.55

a_Time_to_reach_glass_transition_point. b_The_maximum_{polymerization}_rate. c _The_ﬁnal_conversion_of_double_bond. d _Beta_value.

(5)

Fig.8. Doublelogarithmicplotoftheconversionversustimecurvesabovetgfor

(a)0.02%,(b)0.10%and(c)0.50%TX-SHconcentration,respectively.Thevaluesofˇ exponentweredeterminedfromtheslopeofthestraightlines.

consideration,bypredictingthattheybelongtothesame univer-salityclass,thatis,theyobeythepercolationtheory.

4. Conclusions

InthisworkitisseenthatPhoto-DSCtechniquecanbeusedto measurethecriticalexponents,(duringgelformationfromEAto TPGDAmixturesforvariousphotoinitiatorconcentrations.Ithas tobeemphasizedthat (values do notvary duringgelation for allsamplespreparedinvariousinitiatorconcentrationsindicating

thatthegelationinthisparticularsystemobeystheuniversalityof percolationmodel.Howeveritwasobservedthattheother gela-tionparameterssuchastg,RpmaxandCspresentedconsiderable

variationsdependingonthephotoinitiatorconcentration. Acknowledgments

TheauthorsacknowledgeYıldızTechnicalUniversityResearch Foundation (20-01-02-02),TUB˙ITAK (TB-1820), and theTurkish StatePlanningOrganization(24-DPT-01-02-01)fortheirﬁnancial support.

References

[1] N.Arsu,I.Reetz,Y.Yagci,K.M.Mishra,in:K.M.Mishra,Y.Yagci(Eds.),Handbook ofVinylPolymers,CRCPress,London,2009,pp.141–204.

[2]C.G.Roffey,PhotopolymerizationofSurfaceCoatings,JohnWileySons,New York,1982,pp.72–100.

[3]S.P.Pappas,UVCuringScienceandTechnology,TechnologyMarketingCorp., NorwalkCT,1978,pp.2–13.

[4]J.P.Fouassier,Photoinitiation,PhotopolymerizationandPhotocuring,Hanser Verlag,Munich,1995,pp.1–7.

[5]K.Dietliker,Chemistry&TechnologyofUV&EBFormulationforCoatings, Inks&Paints,London,1991,pp.76–77.

[6]R.S.Davidson,in:D.Bethel,V.Gold(Eds.),AdvancesinPhysicalChemistry, AcademicPress,1983,pp.1–130.

[7]M.J.Davis,J.Doherty,P.N.Godfrey,P.N.Gren,J.R.A.Yung,M.A.Parrish,Journal oftheOilandColourChemistsAssociation11(1978)256–263.

[8]L.C¸okba˘glan,N.Arsu,Y.Yagci,etal.,Macromolecules36(2003)2649–2653. [9] S. Keskin, S. Jockusch, N.J. Turro, N. Arsu, Macromolecules 41 (2008)

4631–4634.

[10]S.K.Dogruyol,Z.Dogruyol,N.Arsu,JournalofPolymerScienceA:Polymer Chemistry19(2011)4037–4043.

[11] Z.Do˘gruyol,F.Karasu,G.Temel,D.K.Balta,M.Aydın,S.Keskin,Ö.Pekcan,N. Arsu,in:J.P.Fouassier,X.Allonas(Eds.),BasicsandApplicationsof Photopoly-merizationReactions,ResearchSignpost,Trivandrum,2010,pp.161–173. [12]J.Qin,W.Guo,Z.Zhang,Polymer43(2002)1163–1170.

[13] H.K.Mahabadi,K.F.O’Driscoll,JournalofPolymerChemistryEdition15(1977) 283–300.

[14]I.A.Maxwell,G.T.Russell,MacromolecularTheoryandSimulations2(1993) 95–128.

[15] D.Kaya,Ö.Pekcan,PhaseTransition4(2004)359–373.

[16] M.D.Ediger,C.A.Angell,S.R.Nagel,JournalofPhysicsandChemistry100(1996) 13200–13212.

[17]A.M.North,in:A.D.Jenkins,A.Ledwith(Eds.),TheInﬂuenceofChainStructure ontheFreeRadicalTerminationReaction,Wiley-Interscience,1974. [18] R.G.W.Norrish,R.R.Smith,Nature150(1942)336–337.

[19]G.M.Burnett,H.W.Melville,ProceedingsoftheRoyalSocietyofLondon189 (1947)494–507.

[20] G.V.Schultz,G.Harbort,DieMakromolekulareChemie1(1947)106–139. [21]Z.Do˘gruyol,F.Karasu,D.K.Balta,N.Arsu,Ö.Pekcan,PhaseTransition81(2008)

935–947.

[22]Z.Do˘gruyol,N.Arsu,Ö.Pekcan,JournalofMacromolecularSciencePartB: Physics48(2009)745–754.

[23]P.J.Flory,JournaloftheAmericanChemicalSociety63(1941)3083–3090. [24]W.H.Stockmayer,JournalofChemicalPhysics11(1943)45–55.

[25]M.Sahimi,ApplicationofPercolationTheory,Taylor&Francis,London,1993, pp.169–192.

[26]S.Arbabi,M.Sahimi,PhysicalReviewB47(1993)695–702. [27]S.Arbabi,M.Sahimi,PhysicalReviewLetters65(1990)725–728. [28]A.Aharony,PhysicalReviewB22(1980)400–414.

[29]C.P.Lusignan,J.C.Wilson,T.H.Mourey,R.H.Colby,PhysicalReviewE60(1999) 5657.

[30]E.Andrejewska,M.Andrzejewski,JournalofPolymerSciencePartA:Polymer Chemistry36(1998)665–673.