Effect of polyvinyl alcohol additive on the pore structure and morphology

ContentslistsavailableatScienceDirect

MaterialsScienceandEngineeringC

journalhomepage:www.elsevier.com/locate/msec

Effectofpolyvinylalcoholadditiveontheporestructureandmorphologyofthefreeze-casthydroxyapatiteceramics

KaiHuiZuo,Yu-PingZeng⁎,DongliangJiang

ShanghaiInstituteofCeramics,ChineseAcademyofSciences,Shanghai200050,China

articleinfoabstract

Poroushydroxyapatite(HAP)ceramicswithdifferentmorphologieswerefabricatedbythefreezecastingmethod.ThemorphologiesofHAPceramicsweremodifiedbyadjustingtheconcentrationofpolyvinylalcohol(PVA)additiveintheHAPslurries.HAPceramicswithoutPVAadditivewerecomposedofnon-interconnectedmacroscopiclamellarporesandporousceramicwalls.WithPVAadditive,theHAPceramicsweremadeupofsmalllamellarporesorthree-dimensionalreticulateporesandporousceramicwalls.PVAadditivehadnoeffectonthephasecompositionofHAPceramics.TheopenporosityandporeconnectivitywereimprovedbecauseoftheadditionofPVA.

©2009ElsevierB.V.Allrightsreserved.

Articlehistory:

Received28July2008

Receivedinrevisedform14October2009Accepted6November2009

Availableonline13November2009Keywords:

PorousceramicsFreezecastingPorosityMorphologyHydroxyapatite

1.Introduction

Themechanicalandbiologicalpropertiesofporousbio-ceramics,suchashydroxyapatite(HAP)andcalciumphosphate(TCP)aregreatlyaffectedbytheporesize,poremorphologyandporosity,etc.Therefore,itisworthwhilefinetuningtheporestructuresofporousbio-ceramics.Porousceramicshavebeenfabricatedbymanymethods,suchaspolymericspongeimpregnation,pore-former,extrusion,sol–gel,freezecastingprocess,andsoon[1–6].Amongthem,freezecastingisanovelandsimpletechnique,becauseofitsadvantagessuchaswidelycontrollableporosity,relativelygoodmechanicalstrength,andenvironmentalfriendliness,etc[7,8].

Nowadays,mostresearchesinthefreeze-castporousceramicsarefocusedonthepurewater-basedceramicslurriesintheabsenceofanyorganicadditives,andtheobtainedceramicshavenon-interconnectedlamellarpores[7–9].Recently,Devilleetal.andArakiandHalloranobtainedinterconnectedporeswithdifferentmorphologiesbycontrollingthecoolingratesandthesolidloadingsofslurries[10,11].Themodificationofporemorphologyenlargestheapplicationfields,suchasseparationfilters,catalystsupports,sensorandbiomaterials,etc.Itiswellknownthatceramicslurriesusedintapecasting,slipcastingandgelcastingusuallycontainorganicadditives,suchasPVAasbinder,polyethyleneglycol(PEG)asplasticizer[12–14].Theseorganicadditiveshavegreateffectsonthemicrostructuresandpropertiesofceramics.Thus,itcanbeconjec-

turedthattheorganicadditivesshouldalsoaffectthemicrostructuresandpropertiesofthefreeze-castceramics.

Inthepresentstudy,theporousHAPceramicswerefabricatedbythefreezecastingmethod.Withtheaimtocontroltheporousmorphologywithoutchangingthesolidloadingofslurryandthecoolingrate,anorganicadditivePVAwasaddedintotheHAPslurry.TheeffectofPVAonthemorphologyofporousHAPceramicswasdiscussed.Inaddition,thephasecompositionandopenporositywerealsoinvestigated.

2.Materialsandmethods2.1.Preparationofspecimens

Hydroxylapatitepowder(d50=0.4µm)waspreparedbywetmechanochemicalmethodwithCa(OH)2(Analytical-grade,ShanghaiChemicalReagentCorp.,China),(NH4)2HPO4(Analytical-grade,ShanghaiChemicalReagentCorp.,China)anddeionizedwater.AqueousHAPslurrywithinitialsolidloadingof50.0wt.%waspreparedbymixingpowderwithadispersantagent(ammoniumpolyacrylate,BkGiulini,Germany)indeionizedwater.TheslurrywasballmilledwithZrO2ballsfor24h.TheinitialHAPslurrywasnamed50Hslurry.ThenthepreparedPVA(ShanghaiChemicalReagentCorp.,China)watersolutionwasaddedintotheslurryandthemixedslurrywasfurtherstirredbymagneticforcefor0.5hbeforefreezing.WhentheratioofpurePVAtoHAPpowderwas2.0wt.%,thefinalsolidloadingoftheslurrywaschangedinto46.1wt.%.Themixedslurrywasnamed50H2Pslurry.Table1liststhepropertiesofslurriesusedinthework.

⁎Correspondingauthor.Tel.:+862152415203;fax:+862152413903.E-mailaddress:yuping-zeng@mail.sic.ac.cn(Y.-P.Zeng).0928-4931/$–seefrontmatter©2009ElsevierB.V.Allrightsreserved.doi:10.1016/j.msec.2009.11.003

284

Table1

Compositionofslurriesusedinthework.Slurry70H50H50H2P50H6P50H12P

K.H.Zuoetal./MaterialsScienceandEngineeringC30(2010)283–287

InitialsolidloadingRatioofpurePVAtoFinalsolidloading

(beforeaddingPVA,wt.%)HAPpowder(wt.%)(afteraddingPVA,wt.%)70.050.050.050.050.0

002.06.012.0

70.050.046.140.033.3

Theresultantslurrieswerepouredintosiliconerubbermoldswithinnerdimensionof40×6×8mm.Thewallthicknessofthemoldwas5mm.Allsampleswerefrozeninachamberwithtemperatureof−18°C(illustratedinFig.1a).Aftercompletesolidification,thesamplesweremovedintoalyophilizer(TLG-A,ZhongkeBiologicCo.Ltd,China),followedbydryingat4Pavacuumfor24h.Afterfreezedrying,thegreenbodieswereheatedupto550°Candheldfor2htoremoveorganicadditivesandsinteredat1300°Cfor1hinairatmosphere.Theheatingratesofde-organicandsinteringprocesswere3°C/minand10°C/min,respectively.TheobtainedporousHAPceramicswerenamedas70H,50H,50H2P,50H6Pand50H12Pceramics,respectively.2.2.Evaluationandmeasurements

PhaseanalysisofthesinteredporousHAPceramicswasconductedbyX-raydiffraction(XRD,D/max2550V,Rigaku,CuKα,λ=0.106nm).Morphologywasobservedbyscanningelectronmicroscopy(SEM,JSM-6700F,JEOL,Japan).TheopenporosityandrelativedensityofHAPceramicsweremeasuredbytheArchimedesmethodwithdistilledwaterasliquidmedium.3.Resultsanddiscussion3.1.Phaseanalysis

Fig.2showstheXRDpatternsoftheoriginalHAPpowder,thesintered50Hand50H2Pporousceramics.ThevisiblepeaksinallsamplescanbeindexedtoHAPphase(JCPDS09-0432).Noprocessingresiduephaseisfoundinsinteredsamples.TheresultindicatesthePVAadditivehasnoeffectonthephasecompositionofHAPceramics.3.2.PorosityofHAPceramics

Table2showstherelativedensityandporosityofHAPceramics.TherelativedensityofsinteredHAPceramicsiscalculatedbythemeasureddensity(ratioofweighttototalvolume)andtheoretical

Fig.2.XRDpatternsofHAPpowderandsinteredporousHAPceramics.

density(3.16g/cm3).TheopenporosityoftheHAPceramicswithPVAadditiveisalmostthesameasthetotalporosity.Therefore,itisreasonabletosupposethatmostoftheporesareopen.However,apartofporesinceramicswithoutPVAadditiveisclosed.Table2alsoshowsthattheporositiesofHAPceramicswithPVAadditivearehigherthanthoseofceramicswithoutPVA.Andtheporosityof70Hceramicsislowerthanthatof50Hceramics.TheresultsindicatethatthecontentsofwaterandPVAbothaffecttheporositiesofsinteredceramics.AfteraddingthePVAwatersolutionintotheHAPslurries,theporescausedbysublimatingicecrystalandburningoutofPVAincrease.Thisisthereasonwhytheporosityof50H6P,50H2Pand50H12Parehigherthanthatof50Hceramics.Atthesametime,thesizesoflamellarporesdecreasewithaddingPVA,whichresultsinthesmallerporosityof50H12Pcomparedwiththatof50H2Pceramics.

3.3.MicrostructuresofHAPceramicswithoutPVAadditive

TheSEMmicrographsofHAPceramicswithoutPVAadditiveareshowninFig.3.Theresultsindicatethatthe50Hand70Hceramicsarecomposedofceramicwallsandnon-interconnectedmacroscopiclamellarpores,whicharegeneratedbysublimatinglamellaricecrystals.Fig.3banddshowstheSEMmicrographsoftheHAPceramicwalls,whicharecomposedofsmallporesandHAPparticles.TheporesintheceramicwallregionsarecausedbythesmallicecrystalssublimatingandHAPparticlesstacking.

Intheexperiment,sincethecoefficientofthermalconductivityofthesiliconerubbermoldisrelativelow,theicecrystalsgrowinthedirectionofthetemperaturegradientfromthetopofslurrytothebottom

Fig.1.Schematicillustrationsoffreezingprocessandcross-sectionparalleltothegrowingfrontofcolumnarice,(a)freezingprocess,(b)cross-sectionparalleltothegrowingfrontoficecrystals.

K.H.Zuoetal./MaterialsScienceandEngineeringC30(2010)283–287

Table2

RelativedensityandporosityofHAPceramics.CeramicRelative

density(%)70H50H50H2P50H6P50H12P

79.676.439.537.743.2

Total

porosity(%)20.423.660.562.356.8

OpenporosityobtainedbyArchimedesmethod(%)16.219.860.162.156.0

285

(illustratedinFig.1b).Duringthefreezingprocess,HAPparticlesareexpelledfromtheformingicecrystalsandentrappedwithinchannelsamongthelamellaricecrystals.Simultaneously,HAPparticlesrestraintheicegrowth.Becausetheslurrywithhighersolidloadinghashigherviscosityandasmallervolumeofwater,theicegrowingresistanceisstrongerthanthatinslurrywithlowersolidloading.Thereforetheicecrystalsinthe70Hslurryaresmallerthanthoseinthe50Hslurry.Afterthesublimationandsinteringprocesses,theHAPceramicswithlamellarporesandceramicwallscanbeobtained(Fig.3).Andthesizeoflamellarporesinthe70Hceramicsissmallerthanthatof50Hceramics.TheresultssuggestthattheporesizeofHAPceramicscanbeadjustedbychangingthesolidloadingofslurries,whilethemorphologyoftheceramicscanbetunedbyaddingPVAinslurries.3.4.MicrostructuresofHAPceramicswithPVAadditive

Fig.4showstheSEMmicrographsofHAPgreenbodiesafterfreezedrying,wherethePVAadditivehasnotbeenremoved.ThePVAmainlyexistsneartheporesasdiscontinuousphase,whentheconcentrationofPVAinslurriesislow(Fig.4aandb).AstheconcentrationofPVAin

slurriesincreases,PVAcaneasilyformtheinterconnectedmembraneencapsulatingHAPpowders(Fig.4c,d,eandf).

Fig.5showstheSEMmicrographsofsinteredHAPceramics.Theresultsindicatethatthe50H2Pand50H6Pceramicsarebothcomposedofsmallerlamellarpores.Comparedwithporesof50Hceramics,theselamellarporesof50H2Pand50H6Pceramicsarenotmacroscopic.The50H12Pceramiciscomposedofthree-dimensionalreticulatepores.Theobviousdifferenceliesinthatthenon-interconnectedporesintheHAPceramicswithoutPVAadditivegraduallychangeintotheinterconnectedporesintheHAPceramicswithPVAadditive.

Figs.4and5showthatthefreezingprocessofslurrywithPVAadditivecanbedividedintotwostages.Inthefirststage,thetemperatureofslurryishigherthan0°C,thePVAgelsandliquidwaterstillexists.ThegelledPVAwillencapsulatetheHAPpowdersandformthepowderwalls.Inthesecondstage,theslurryfallsintothesubzerotemperature,thewatersolidifiesandpowdercontainedbygelledPVAwillrestraintheicegrowth.Therefore,thecompletefrozenbodyconsistsoftwophases:oneisthepowderwallcontainedbygelledPVAandanotherisicecrystal.Aftersublimationandsintering,theicecrystalsandPVAdisappearandporousceramicsform.

ResultsalsoindicatethattheconcentrationofPVAhasgreateffectontheporesizesandmorphologiesofHAPceramics.WhentheconcentrationofPVAinslurriesisrelativelow,thegelledPVAdoesnotformcontinuum,asseeninFig.4aandb.Theicecrystalswilleasilygrowintobigcrystals,resultinginbiglamellarporesaftersublimationandsinteringprocesses.WithincreasingtheconcentrationofPVA,PVAformsinterconnectedflakyshapegels.Theicecrystalswouldencounterbiggerresistances,resultinginsmallerlamellarpores(Fig.5candd)thanthatintheceramicswithlowerPVAaddition,suchas50H2Pceramics.WhentheconcentrationofPVAreaches12wt.%,PVAformstheinterconnectedandimmovablenetwork(Fig.4eandf).Theice

Fig.3.SEMmicrographsofHAPceramicswithoutPVAadditiveaftersintering(cross-sectionverticaltotheicefront),(a)50Hceramic,(b)ceramicwallof50Hceramic,(c)70Hceramic,(d)ceramicwallof70Hceramic.

286K.H.Zuoetal./MaterialsScienceandEngineeringC30(2010)283–287

Fig.4.SEMmicrographsofHAPgreenbodiesafterfreezedrying,wherethePVAadditivehasnotbeenremoved(cross-sectionverticaltotheicecrystalfront),(a,b)50H2P,(c,d)50H6P,(e,f)50H12P.

crystalsencounterastrongresistancesothatthelamellaricecrystalscannotform,resultinginthree-dimensionalreticulateporesinceramics(Fig.5eandf).4.Conclusions

TheHAPceramicswithdifferentmorphologieswerefabricatedbythefreezecastingmethod.ThesolidloadingofslurryandtheconcentrationofPVAadditivehavegreateffectsontheprocessofPVAgelationandmannersoficecrystalsgrowth,resultingindifferentmorphologiesandsizesofporesinHAPceramics.ThemorphologiesofHAPceramicswithoutPVAadditivearecomposedofnon-interconnectedmacroscopiclamellarporesandporousceramicwalls.AfteraddingPVAintoHAPslurries,theporemorphologieschangeintosmallerlamellarpores.WhentheconcentrationofPVAis12wt.%,three-dimensionalreticulateporesareobtained.MostoftheporesinceramicswithPVAadditiveareopen.TheopenporosityisimprovedbecauseoftheadditionofPVAandmoreporesmakefor

interconnections.TheHAPceramicswithsmallporesfabricatedwithPVAadditioncanbeusedasmedicinecarrierandbasalbodyofHAPceramicswithlargeporesinordertoimprovethestrength,andsoon.Acknowledgements

TheauthorsthankforthefinancialsupportfromtheNationalNaturalScienceFoundationofChina(projectno.50902140)andthesupportofInnovativeFoundationofShanghaiInstituteofCeramics,ChineseAcademyofScience.Thecommentsofthereviewersareappreciated.References

[1][2][3][4][5][6]

W.Schwartz,H.Somers,A.V.Somers,USpat.(1963)No3090094.X.W.Zhu,D.L.Jiang,S.H.Tan,Mater.Sci.Eng.A323(2002)232.C.Falamaki,M.Naimi,A.Aghaie,J.Eur.Ceram.Soc.24(2004)3195.A.Diaz,S.Hamphshire,J.Eur.Ceram.Soc.24(2004)413.L.J.Zu,S.J.Luo,J.Mater.Pro.Tech.114(2001)1.

MVerdenelli,S.Paarola,F.Chassagneux,J.M.Létoffé,H.Vincent,J.P.Scharff,J.Bouix,J.Eur.Ceram.Soc.23(2003)1207.

K.H.Zuoetal./MaterialsScienceandEngineeringC30(2010)283–287287

Fig.5.SEMmicrographsofHAPceramicswithPVAadditiveaftersintering(cross-sectionverticaltotheicefront),(a,b)50H2P,(c,d)50H6P,(e,f)50H12P.

[7][8][9][10]T.Fukasawa,M.Ando,T.Ohji,S.Kanzaki,J.Am.Ceram.Soc.84(2001)230.S.Deville,E.Saiz,R.K.Nalla,A.P.Tomsia,Science311(2006)515.T.Fukasawa,Z.Y.Deng,M.Ando,J.Am.Ceram.Soc.85(2002)2151.S.Deville,E.Saiz,A.P.Tomsia,Biomaterials27(2006)80.[11][12][13][14]K.Araki,J.W.Halloran,J.Am.Ceram.Soc.88(2005)1108.D.Hotza,P.Greil,Mater.Sci.Eng.A202(1995)206.

J.D.Yates,S.J.Lombardo,Mater.Sci.Eng.A337(2002)297.M.A.Janney,O.O.Omatete,USpat.(1991)No5028362.