Surface&CoatingsTechnology235(2013)699–705
ContentslistsavailableatScienceDirect
Surface&CoatingsTechnology
journalhomepage:www.elsevier.com/locate/surfcoat
HighpowerdiodelasercladdingofFe–Co–B–Si–C–Nbamorphouscoating:Layeredmicrostructureandproperties
Y.Y.Zhua,Z.G.Lia,?,R.F.Lia,M.Lia,K.Fenga,Y.X.Wua,b,T.Wadac,H.KatocabcShanghaiKeyLaboratoryofMaterialsLaserProcessingandModi?cation,SchoolofMaterialsScienceandEngineering,ShanghaiJiaoTongUniversity,Shanghai200240,ChinaStateKeyLaboratoryofMetalMatrixComposites,SchoolofMaterialsScienceandEngineering,ShanghaiJiaoTongUniversity,Shanghai200240,ChinaInstituteforMaterialsResearch,TohokuUniversity,2-1-1Katahira,Aobaku,Sendai980-8577,Japan
articleinfoabstract
Fe–Co–B–Si–C–Nbamorphouscompositecoatingabout0.6mminthicknesswasfabricatedbyusingaone-steplasercladdingmethod.Microstructuresandphaseswereinvestigatedbyscanningelectronmicroscopy,transmissionelectronmicroscopy,X-raydiffractionandelectronprobemicroanalysis.Microhardnessandwearresistancetestswereconductedtoevaluatethemechanicalpropertiesofthiscoating.Neartheinterfaceofcoating/substrate,theregionshowedalayeredmicrostructure,whichcouldbegenerallycategorizedintothreelayers:layerI(columnardendritesphase),layerII(equiaxeddendritesphase)andlayerIII(amorphous-particlecompositephase).Themainreasonforthislayeredstructurewasduetothedifferenceinchemicalcompositionsofthethreelayers.Withregardtomechanicalproperties,themicrohardnessandwearresistanceoftheFe–Co–B–Si–C–Nbamorphouscompositecoatingalsoexhibitedlayeredcharacteristics.ThemeanvalueofthemicrohardnessforlayerI,layerIIandlayerIIIwas729,680and1245HV,respectively.Thefrictioncoef?cientofthetransitionallayerIIIwas0.28timeslowerthanthatofthesubstrateunderthesameslidingfrictioncondition.
?2013ElsevierB.V.Allrightsreserved.
Articlehistory:
Received20November2012
Acceptedinrevisedform27August2013Availableonline4September2013Keywords:LasercladdingAmorphousCoating
MicrostructureMicrohardnessWearresistance
1.Introduction
Amongthefamiliesofbulkmetallicglasses(BMGs),Fe-basedBMGshaveattractedmuchattentionbecauseoftheirhighyieldandfracturestrength,excellentwearresistanceandrelativelylowcost[1–3].How-ever,theirindustrialapplicationsarenotquitewidespreadowingtothelimitedsize.Theproductionofthistypeofmaterialsintheformofcoatingscanbeaninterestingtechnologicalachievement.Extensiveef-fortshavebeenmadetowidentheapplicationthroughtheintroductionofamorphousalloyscoatings[4–8].Thepreparationofamorphouscoat-ingsbyusinglasercladdingtechniqueisthoughttobeonefeasibleapproach[9–15].
Lasercladdingisafeasibletechniquetoproducemetallicglasssur-face/coatingsincethecoolingratescanachieve103–106K/s,muchhigherthanthecriticalcoolingrates(Rc)ofBMGswithgoodglassformingabilities(GFA)[16].Laserpower,laserscanningspeedandpow-derfeedingratearethemaintechnologicalparametersthatcontrolthecoatingstructures.Therefore,therelationshipsamongtheprocessingparameters,microstructuresandpropertiesofthecoatingshavebeenstudiedoverthelasttwodecades[10,17–19].However,fewstudieshavebeenfocusedonthevariationsofmicrostructuresandpropertydistributionsatdifferentdepthsofthecoatingsanditsmechanisms.Chemicalcompositionandmicrostructurearealwaysinhomogeneous
?Correspondingauthor.Tel.:+862134202837;fax:+862134203024.E-mailaddress:lizg@sjtu.edu.cn(Z.G.Li).0257-8972/$–seefrontmatter?2013ElsevierB.V.Allrightsreserved.http://dx.doi.org/10.1016/j.surfcoat.2013.08.050
inthecladdedcoatingduetotheextremelynon-equilibriumsolidi?-cationconditionduringlasercladding.Therefore,theunderstandingofmicrostructuraldevelopmentundernon-equilibriumconditionsiscriticalinthedesignofnewmaterials[20].Forexample,functionallygradedmaterialsareaclassofmaterialswiththeircompositionandmi-crostructurechangedgraduallyortailoredintentionallyfromonesidetotheother,resultinginacorrespondingvariationintheirproper-ties[21].Itissigni?canttostudythevariationofmicrostructuresofthecoatingsanddiscusstheformationmechanismsduringrapidsolid-i?cation.Lasercladdingsolidi?cationstartsbyepitaxialgrowthonthecubicmetalsubstrate[22].Ifthecladdingmaterialcompositionisthesamewiththesubstrate,thecladlayerswillhaveacrystallographicori-entationsimilartothatofthesubstrate.Bezen?onetal.[23,24]foundthatepitaxylossmayalsooccurwhenpowderwithacompositiondifferentfromthesubstratewasinjectedintothemeltpool,duetothenucleationandgrowthofdifferentphases.Inthiswork,akindofFe–Co–B–Si–C–Nbamorphouscompositecoatingwasfabricatedbyusingaone-steplasercladdingmethod.Afterwards,thedistributionsofmicrostructure,phase,microhardnessandwearresistanceoftheamorphouscompositecoatingwerestudied.2.Experimentaldetails
Fe34Co34B20Si5C3Nb4powderswereproducedbyhighpressureArgasatomization.Thesizeofthepowdersrangedfrom60μmto70μmasshowninFig.1(a).Fig.1(b)showstheXRDpro?leofthepowder
700Y.Y.Zhuetal./Surface&CoatingsTechnology235(2013)699–705
Table1
Parametersoflasercladdingprocess.Power(W)1900
Scanspeed(mm/s)50
Feedingrate(g/min)30
Carriergas(Ar)(l/min)3
Shieldinggas(Ar)(l/min)15
Fig.1.(a)Micrographand(b)XRDpatternofthepowder.
usingCu-Kαradiation.Thepresenceofbroadhalowithdiffusedinten-sityintheXRDpro?lecon?rmedthattheprecursorpowderusedinthepresentstudywasamorphousinnature.Lowcarbonsteelwasselectedasthesubstratewiththesizeof150mm×20mm×8mm.Allsub-stratesweremachinedandpolished,thendegreasedbyacetoneanddriedintheair.A3.5kWhighpowerdiodelaser(ROFIN,DL035Q)
withacoaxialpowderfeedingnozzlewasusedtoperformthelasercladdingprocess.Thewavelengthofthelaserwas808nmand940nm,andthespotwas3.3mm×2mmrectangular-shaped.Thebasematerialsurfacepositionwasinthefocalplane.
Manyparametersneedtoadjustduringlasercladdingprocess.Theheatinputbythelasersourcemustbewell-controlledtoachieveade-sirabledilutionratiobetweencoatingandsubstrate[25–27].Theeffectoflasercladdingfactors(dilutionratioandlaserscanningspeed)onthestructureandamorphousfractionofthecoatingwasstudiedinourpreviousresearch[28].Theresultrevealedthatoptimumparameters(lowerdilutionratioandhigherscanningspeed)cangethigherfractionoftheamorphousphaseofthecoatings.Forthescanningspeedof50mm/s,thelaserpowerwas1700,1800,1900and2000W.Thecross-sectionofeachsamplewasobservedtoobtainitsbondingcharac-teristics.Theresultsshowedthatthedilutionratiowashigherathigherlaserpower,themetallurgicalbondingwasnotgoodatlowerlaserpower.ThemicrostructuresofthecoatingatdifferentdilutionsareshowedinFig.2.Itcanbeseenthatthecoatingwithahigherdilutionratioisnotcomposedofamorphousphase.Optimumparameterswerechosencorrespondingtothebestsamplewithbothlowdilutionratio,higheramorphousfractionandgoodmetallurgicalbondingbetweensubstrateandcoating.ThedetailedcladdingparametersarelistedinTable1.
Microstructuresandphaseidenti?cationsatdifferentdepthsofthecoatingwerestudiedbyscanningelectronmicroscopy(SEM,JSM-7600),X-raydiffraction(XRD,D/max2550VL/PC)withCuKαradiationandscanningtransmissionelectronmicroscopy(STEM)withenergydispersiveX-rayspectroscopy(EDS,JEM-2100F).Chemicalcompositionswerestudiedbyelectronprobemicroanalysis(EPMA,JXA-8230)withtheprobesizeof500nm.Thecompositionsat8dif-ferentspotsintheinterestedregionswereaveraged,whichwereusedforsemi-quantitativeconsiderationsonthephasecomposi-tions.Themicrohardnessalongthedepthdirectionofthecompositecoatingwastestedusingaloadof0.5Nandloadingtimeof10s.Drywearbehaviorofthetransitionallayerswasevaluatedwiththero-tatingwearmachine(MMW-1A).Specimenswerecutintocuboidswiththesizeof7mm×7mm×8mmandusedasarotatingpin.TheschematicdiagramofweartestsystemwasshowninFig.3.Toinvestigatethetribologicalpropertiesofdifferentlayers,the
Fig.2.(a)SEMimageofthecoatingwithlowerdilutionratio;(b)SEMimageofthecoatingwithhigherdilutionratio.
Y.Y.Zhuetal./Surface&CoatingsTechnology235(2013)699–705701
specimenswerepolishedatdifferentthicknessestoreachdifferentlayersandthenthelayersweretestedinplanview.Duringthetest,therotatingpinwasputunderaforceloadof200Nagainstasta-tionarydiskmadeofsinteredAl2O3ceramic.Arotationalspeedof60r/min,withadurationof60minwasselectedasanotheroper-atingparameter.Wearmasslosswastestedinthisexperiment.ThewornsurfaceswereexaminedundertheSEMwithEDSinthesec-ondelectronicmodeto?ndoutthewearmechanism.3.Resultsanddiscussion
Fig.4(a)showsthemacrostructureofthecoatingwithasmoothsurface.Fig.4(b)showstheimageofthecoatingincross-section,whichindicatesalowdilutionratiowithathicknessofapproximately0.6mm.Nocracksorvoidsarefoundinthecross-section.Fig.4(c)isthemagni?edregionthatismarkedbyarectangularframeinFig.4(b).ItcanbeseenfromFig.4(c)thatthecoating(neartheinterfaceofcoating/substrate)exhibitsalayeredmicrostructure,whichcouldbegenerallycategorizedintothreelayers:layerI(columnardendrites),layerII(equiaxeddendritesandafewwhiteparticles)andlayerIII(graymatrixandmanywhiteparticles).Thecolumnardendritelayer(~10μm)directlyadjacenttothesteelsubstrateshowsgoodmetal-lurgicalbondingattheinterfaceofthecoating/substrate.Towardsthetopofthecoating,thecolumnardendritesarefollowedbytheequiaxeddendriteswiththewidthofabout8μm.Therestofthe
Fig.3.Theschematicdiagramofweartestsystem.
Fig.4.(a)Macrostructureofthecoating;(b)SEMimageofthecross-section;(c)microstructureofthecoatingneartheinterfaceofcoating/substrate;(d)microstructureofthetopofthecoating;(e)schematicillustrationofthegradedcoating.
702Y.Y.Zhuetal./Surface&CoatingsTechnology235(2013)699–705
Fig.5.XRDpatternsofthecoatingatdifferentdistancesfromtheinterfaceofcoating/substrate.
coatingmicrostructureislayerIIIwhichisgraymatrixwithwhiteparticlesembeddedinitrandomly.Fig.3(d)isthemagni?edregionthatismarkedbytherectangularframeinFig.4(b)atthetopofthecoating.Itcanbeseenthatthecrystallinephasewasembeddedinthegrayphaseatthetopofthecoating.Fig.4(e)showstheschematicillustrationofthecompositecoating.
Fig.5showstheXRDpatternsofthecoatingatdifferentdistances.Atthetopofthecoating,theXRDpatternpresentssomepeakscorre-spondingtothecrystallinephasesFe2Bandbody-centeredcubic(bcc)-Fe.Inthemiddleofthecoating,thepatternshowsabroadhalopeakatadiffractionangleof44°(2theta).FewweakdiffractionpeaksoftheNbCphasecanalsobeseenoverthebroadhalo.Itindicatesthatthemicrostructureinthisregionisamixtureofamorphousandcrystallinephases.Neartheinterfaceofthecoating/substrate,thepeaksofthebcc-FephasecanbeclearlyseeninXRDpattern.Thelatticeparameterofthebcc-Fephaseinthesubstrateis0.2866nmwhichcor-respondstothatofpurebcciron[PDF#06-696].Alongwiththeheightofthecoating,thelatticeparameterofthebcc-Fephasemonotonously
Fig.6.TEMimagesofdifferentlayersofthecoatingandcorrespondingSAEDpatternsandEDSspectrum:(a)layerIII;(b)EDSspectrumofthewhiteparticleinlayerIII;(c)layerII;(d)layerI.
Y.Y.Zhuetal./Surface&CoatingsTechnology235(2013)699–705703
Fig.7.Microhardnessvaluesmeasuredacrossthecoating.
decreases.IthasbeenreportedthatthelatticeparameterdecreaseswithincreasingcontentofCointhebcc-Fephase[29].Thisresultsug-geststhatthecontentofCoinbcc-Fephasedecreaseswithdecreasingheightofthecoating.
TheaveragecompositionsoftheregionsAandBareFe36Co35B16Si6-C3Nb4andFe58Co23B8Si5C4Nb2(at.%),respectively.ThecompositioninregionAisrelativelyclosetothatofthepowder.However,incompari-sontoregionA,regionBisenrichedbyFewhichagreeswiththechangeoflatticeparameterofbcc-FeobservedintheXRDpatterns.Thissug-geststhatthediffusionofFefromsubstratetocoatinglayeroccurredduringlasercladdingtreatment.Duetothesecompositionalchanges,themicrostructureofthecoatingresultedinamorphousandcrystallineheterogeneousstructure.AtlayerIIIwheretheaveragecompositionissimilartothatofpowder,themelthadanoptimumcompositionforamorphousformationthereforeamorphousmatrixwithsmallvolumeofNbCwasformed.However,atlayersIandIIwherethealloycompo-sitiondeviatesfromtheoptimumone,amorphousphasedidnotformedbutthedendriteofbcc-Fephasepredominantlyprecipitatedduringsolidi?cation.
Fig.6showstheTEMimagesobtainedfromdifferentlayersoftheFe–Co–B–Si–C–Nbcompositecoating.Fig.6(a)showstheTEMimageobtainedfromlayerIII.Itcanbeseenthatthegraymatrixisanamor-phousphaseaccordingtotheSAEDpattern.Thewhiteparticleisacrystallinephasewithface-centeredcubic(fcc)structure.TheSTEM–EDSlinescananalysisshowsthatthecontentofelementsNbandCismuchhigherinthewhiteparticlethanintheamorphous
Fig.8.Frictioncoef?cient–timecurvesofthetoplayer,layerIII,layersIandII,andsubstrate.
Table2
Statisticresultsonfrictionforce,wearmasslossandfrictioncoef?cientduringwearprocess.Region
AverageWearFrictioncoef?cientfrictionmassforce(N)loss(g)RangeAverageRelativewearToplayer205.770.0070.129–0.15410.144620.39LayerIII
206.470.0030.0974–0.11510.104860.28LayersIandII206.560.0110.2285–0.25360.244120.66Substrate
204.21
0.055
0.3424–0.3857
0.36816
1
matrix(Fig.6(b)).ThisresultagreeswellwiththeXRDpatternwhichindicatedformationinthemiddlesectionofthecoatingandofacompositeconsistingofanamorphousphaseandNbCphase.TEMimagesofFig.6(c)and(d)areselectedfromlayerIIandlayerI,respectively.ThesetworegionsshowbccstructuresasindicatedbytheSAEDpatterns.CombiningSAEDpatternswiththeEPMA,SEMandXRDresults,theequiaxedandcolumnardendritesaresup-posedtobebcc-Fephase.
Accordingtotheaboveresults,itcanbeobservedthattheamor-phouscompositephaseismainlyformedinlayerIIIandsomecrys-tallinephasesformedatlayersIandIIofthecoatingwhenusingaone-steplasercladdingmethod.Atthebottomofthemoltenpool,thesubstrateactsasaheatsinkandtheheatisdissipatedthroughthesubstrateeasily[30].Therefore,astrongthermalgradientGexistsattheinterfaceofthecoating/substrate.Still,itishardtoformanamor-phousphaseatlayersIandIIbecausethechemicalcompositiondevi-atesfromthenominalcompositionbasedontheEPMAresults.
AsshowninlayerII(Fig.4c),equiaxedgrainscannucleateandgrowaheadofthecolumnardendriticwhenthereisaregionofundercooledliquid.Thiscanleadtothetransitionfromacolumnartoanequiaxeddendriticmorphology,referredtoasCET(columnartoequiaxedtransition)[31–33].AccordingtoG?umann'smodel[34,35],duringtheadvanceofthecolumnardendriticfront,soluteisrejectedaheadofthesolidi?cationinterface.Theliquidsandtemper-aturepro?lesleadtoaconstitutionallyundercooledregionaheadofthedendriticfront.Inthisundercooledregion,nucleationmayhappenifthemaximumgrowthundercoolingislargerthantheundercoolingre-quiredfornucleation.Andthecompositionisafactorthathasastrongin?uenceontheshapeandmagnitudeoftheundercooledregion[34,35].Asthesolid/liquidinterfacemoves,thecompositionoftheliquidisclosetothenominalcompositionandtheamorphousphaseasthemainphaseinthecoatingisformedunderthelargeundercoolingcondition.4.Properties
Fig.7showsthemicrohardnessofthecoatingatvariousdepthsfromthecoatingtothesubstrate.ThemeanmicrohardnessvalueoflayerI,layerIIandlayerIIIis729,680and1245HV,respectively.Itcanbein-ferredthatthehighermicrohardnessvaluesinlayerIIIareattributedtotheNbCparticlesembeddedintheamorphousmatrix.Towardsthebottomofthecoating,thevaluesofmicrohardnessofthecoatingde-creased.Thisisprobablyduetothecrystallinephaseneartheinterface.Anabruptdropinmicrohardnessisobservedattheinterface,whichisduetoanarrowmixingzoneofthecladdingandsubstratematerials.
Thefrictioncoef?cient–timecurvesofthecoatingareshowninFig.8.ItisclearthatlayerIIIandthetoplayerexhibitalowerfrictionco-ef?cientthanlayersIandII,whilethesubstratehasthehighestfrictioncoef?cient.Thedetailedstatisticdataonfrictionforce,wearmasslossandfrictioncoef?cientarelistedinTable2.Accordingtothedata,layerIIIexhibitedalowfrictioncoef?cientunderroom-temperaturedryweartestconditions.Fig.9showsthewornsurfacemorphologiesofdifferentlayersatthesameweartestparameters.Thewornsurfacemorphologyofthetoplayerissmoothandexhibitsnoobvioussharpgrooves(seeFig.9(a)).Themagni?cationoftherectanglemarkedin