Effect of the crystallinity of diamond coatings on cemented carbide inserts on their cutting performance in milling

Effect

of

the

crystallinity

of

diamond

coatings

on

cemented

carbide

inserts

on

their

cutting

performance

in

milling

G.

Skordaris

a

,

K.-D.

Bouzakis

(1)

a,b,

*,

T.

Kotsanis

a

,

A.

Boumpakis

a

,

F.

Stergioudi

a

,

D.

Christo

_filos

c

_,

_O.

_Lemmer

d

_,

_W.

_Kölker

d

_,

_M.

_Woda

d

a

LaboratoryforMachineToolsandManufacturingEngineering,MechanicalEngineeringDepartment,AristotleUniversityofThessaloniki,Greece

b

Turkish-GermanUniversityinIstanbul,Turkey

c_{ChemicalEngineeringDepartment&LaboratoryofPhysics,FacultyofEngineering,AristotleUniversityofThessaloniki,Greece} d_{CemeConAG,Germany}

1. Introduction

Micro-crystalline diamond coatings deposited on cemented carbidesubstratescanbeeffectivelyappliedinmachiningof non-ferrous materials such as of aluminium alloys, carbon fibre reinforced plastics etc. [1–4]. Due to the superior adhesion characteristics of MCD coatings and to improved tribological propertiesofnano-crystallinediamondones,variousmicro-and nano-crystalline layer coating systems on cemented carbide substrates are manufacturedand used effectively in machining procedures[5–7].Theappliedparametersduringthehotfilament chemicalvapourdepositionprocesssuchasofsubstrate tempera-ture,totalpressure etc.affectsignificantlythefilmgrowth and structureandinthiswaytheirproperties[8].

Thispaperaimsatinvestigating,forthefirsttime,theeffectof theresultingcrystallinityofdiamondcoatingsduetotheapplied depositionparameters ontheirfatiguestrengthatambientand elevated temperature and wear behaviour in milling. In this context,twogroupsofmicro-crystallinecoatingsweredeposited onsame cementedcarbide substrates byvarying the substrate temperature.Ramanspectrawereusedtocheckthecrystallinityof the deposited diamond coatings. For evaluating the fatigue strengthoftheproducedcoatings,inclinedimpacttestsat25C and300C werecarriedout [9,10]. Thepreparedcoatedinserts

wereusedinmillingaluminiumfoamforassessingtheircutting performance. Raman spectroscopy was also conducted on the remainingworncoatingintheimpactcraterandonthetoolrake within the chip contact area after milling. Via the detected crystallinitychanges,itwaspossibletoexplainthedifferentwear evolutionsduringtheimpacttestandmillingwhenusing micro-crystallinecoatingsatvarioussubstratetemperaturesdeposited. 2. Experimentaldetails

The applied cemented carbide inserts specifications are illustrated at the bottom of Fig. 1. These were coated with micro-crystallinediamondcoatingsviathehotfilamentmethod using aCC800/9DiaCEMECONcoatingmachine. Hereupon,two insert’sbatchesweremanufacturedatvarioussubstrate tempera-turesduringthedepositionprocess.Inthe_firstinsertbatchnamed as T1, the substrate temperature was adjusted at 900C. For preparing the secondbatch (T2), this temperaturewas slightly increased.Thefilamenttemperatureamountedtoapproximately 2000Candthetotalpressureto30mbar.Atacarbontohydrogen ratioof1%,andagasflowof2l/min,thecoatinggrowthratewas around 0.5

m

m/h. For thecoating thickness ofabout 5

m

m, the overallprocesstimewasequaltoroughly19h.

Theinclinedimpacttestatvariousloadsandtemperaturesup to300Cwasusedtocheckthefatiguestrengthoftheprepared micro-crystalline diamond coatings. The applied device in the conductedinvestigationswas constructedbytheLaboratoryfor Machine Tools and Manufacturing Engineeringof the Aristotle UniversityofThessalonikiinconjunctionwithCemeConAG[9,10]

(seeFig.1a).Highpressureairatatemperatureequaltothetest CIRPAnnals-ManufacturingTechnology68(2019)65–68

ARTICLE INFO Keywords: Diamondcoating Fatigue Wear ABSTRACT

Micro-crystallinediamond(MCD) coatingsweredepositedoncementedcarbideinsertsatdifferent temperaturesusinghotfilamentchemicalvapordepositiontechnique.Forinvestigatingtheeffectofthe developeddiamondcrystallinityonthefatiguestrengthandwearbehaviourofthepreparedMCDcoated inserts,inclinedimpacttestsandmillinginvestigationswereconductedcorrespondingly.Ramanspectra wererecordedforcapturingthecrystallinephasesafterthefilmdepositionandtheirpotentialchanges aftertheimpactand millingexperimentsinduced bythemechanicalandthermalloads.Thus, the explanationofthecuttingperformanceoftheemployeddiamondcoatedinsertswithvariouscrystalline phaseswasenabled.

©2019PublishedbyElsevierLtdonbehalfofCIRP.

*Correspondingauthorat:LaboratoryforMachineToolsandManufacturing Engineering(LMTME),MechanicalEngineeringDepartment,AristotleUniversityof Thessaloniki,Greece.

E-mailaddress:bouzakis@eng.auth.gr(K.-D.Bouzakis).

ContentslistsavailableatScienceDirect

CIRP

Annals

-

Manufacturing

Technology

j o u r n a lh o m e p a g e :h t t p : / / e e s . e l s e v i e r . c o m / c i r p / d e f a u l t . a s p

https://doi.org/10.1016/j.cirp.2019.04.056

onewasemployedforremovingpotentiallyoccurredballindenter orcoatingdebris.Thetimecourseoftheappliedimpactloadsignal is illustrated in Fig. 1b. The developed impact imprints were evaluatedthrough3Dmeasurementsbytheconfocalmicrscope

m

SURF of NANOFOCUS AG. The Raman spectroscopy on the untreatedaswellastreateddiamondcoatingswerecarriedout usingaLabRAMHRspectrometer.Themillinginvestigationswere conductedemployingathree-axisnumerically-controlledmilling centre using aluminium foam as workpiece material. This workpiece material consists of various hard phases, as related optical microscopy observations using standard metallographic techniquesrevealed(seeFig.1c).Moreover,duetothestructureof theworkpiecematerial,intensedynamicloadsaredevelopedon thecuttingedgeofthecoatedtoolsduringcutting.

3. Characterizationoftheinvestigateddiamondcoatings 3.1.Crystallinityofthediamondcoatings

For characterizing thecrystallinityof theproduceddiamond coatings, Raman spectroscopy was performed. The recorded

spectraarepresentedinFig.2.Bothcoatedinsertbatchesexhibit anarrowpeakataround1340cm 1_{confirmingthehighdiamond} crystalline quality of the coatings and the existenceof a sp3 -bondedphase[11].Twomorepeaksappearpronouncedatroughly 1150cm 1 _{and 1450cm} 1 _{in the Raman spectra of the T2} specimens. These extrafeatures originate fromco-existing sp2 -phases,calledastranspolyacetylene.Thelatterisassociatedwith the chemical elements of the applied gases during the film deposition [11]. The trans polyacetylene possesses obviously inferiorstrengthpropertiescomparedtothesp3-bondedphases. Moreover,since thedeveloped maximum cutting temperatures duringtheconductedmillinginvestigationsareroughlyequalto 300C[3],thepreparedcoatedinsertswereannealedat300Cfor checking a potential effect of such a temperature on their crystallinity.BasedontheresultspresentedinFig.2,theRaman spectraremaininvariablebyatemperatureincreaseupto300C, i.e. no effect on the _film crystallinity occurs. However, a decomposition of the sp2_{-bonded trans polyacetylene phases} maytakeplacewhenrepetitivethermalandmechanicalloadsboth leadingtoelasticfilmdeformationsaresimultaneouslyapplied. Thiscouldhappenforexampleduringtheimpacttestatelevated temperaturesoracuttingprocedureaswell.

3.2. Fatiguestrengthofthediamondcoatingsatroomandelevated temperatures

ThepreparedMCDcoatingsweresubjectedtorepetitiveimpactsat 25C and 300C using the inclined impact test. Characteristic imprints after106_{impactsgeneratedonT2coatedinsertsat25}_{Caftervarious}

impactloads,aredisplayedinFig.3.Theseimprintswerescannedby whitelightconfocalmicroscopy and theydepictthe developedsurface topomorphyafteronemillionimpactsattherelatedimpactloads.All over the test duration, the impact force, the specimen temperature and furthertestparametersarerecordedandcontrolledforavoidinga potentialdrift ofthe impactspot.Inthecaseofanimpactloadof900N, nofilmdamagecanbeobserved.Onthecontrary,theimpactload growthupto1050Nleadstoatotalcoatingremovalandsubstrate wear. Similar results arise in the case of T1 coated inserts. Consequently, thesp2-bondedphaseof theT2 coatedinserts did notaffectatambienttemperaturesthefilmfatiguestrength.Thelatter dependsinbothbatchcasesonlyontheimpactload.

Fig.4demonstratesfurtherimpactcratersformedat300Cand atanimpactloadof150N,aftervariousnumbersofimpacts.These imprintsweregeneratedatadjacentcoatedsurfacelocations,since the repositioning of the ball indenter after the conduct of a confocalmicroscopyscanningcannotbeperformedatanaccuracy of few micrometers. The coating damage evolution versus the numberofimpactsismoreintensiveonthecoatedspecimensof theT2batch.Comparedwiththis,inthecaseofT1inserts,the

Fig.1.(a)Themechanicalunitoftheemployedimpacttester;(b)characteristicdata oftheappliedforcesignals;(c)theworkpiecematerialusedinmillingexperiments.

Fig.2.Ramanspectraoftheinvestigatedcoatings.

Fig.3.Inclinedimpacttestresultsafter106

impactsat25_C.

G.Skordarisetal./CIRPAnnals-ManufacturingTechnology68(2019)65–68 66

diamondcoatingwithstandstherepetitiveimpactloadsafterone millionimpactswithoutafilmtotalremoval.Onthecontrary,an extensivecoatingfailureappearsatthesameloadandnumberof impactsinthecaseofT2coatedinserts.

Anoverviewofthedevelopedimprintdepthsduringtheimpact test at 25C and 300C is illustrated in Fig. 5. Although both coatingbatchesexhibitedthesamefatiguestrengthat ambient temperature, the coating batch T1 shows a superior wear resistanceagainsttherepetitiveimpactloadsat300Ccompared to the T2 batch. Hereupon, the impact load at the 300C is signi_ficantlylowercomparedtothatoneatambienttemperature. Hence,theelevatedtemperaturerestrictsthefilmfatiguestrength. Thisrestrictionismoreintensive,ifasinthecaseoftheT2films, transpolyacetylenephasesexistinthecoatingstructure.

For explaining the latter statement, Raman spectroscopy was conductedontheremainingworncoatingintheimpactimprintand onanundamagedareaoutsideoftheimprintforbothfilmcases(see

Fig.6).Ononehand,theRamanspectraoftheT1coatedinsertsare similarinbothmeasurementpositionsAandB(seeFig.6a).Onthe otherhand,inthecaseoftheT2coating,intheimprintareaatthe measurement position D (seeFig. 6b), where thermaland impact loads aresimultaneously exercised, thespectral peaksare considerably weaker comparedto thecorrespondingonesoutsidetheimpactcrater atthemeasurementpositionC.Thisfactrevealsadecompositionof thesp2_{-bondedtranspolyacetylenephasesindicatedbythepeaksat} around1150cm 1and1450cm 1aswellasofthenanosp3-bonded phaseofhighcrystallinityatroughly1340cm 1_{.Inthisway,thefilm} fatiguestrengthworsens.

4. Cuttingperformanceofthediamondcoatedinsertsin millingandexplanationoftheobtainedresults

Thecuttingperformanceofthemanufactureddiamondcoated insertswasinvestigatedinmillingwithoutcoolantorlubricantfor

attaining a more intense wear evolution. After a prescribed numberofsuccessivecuts,thecuttinginsertwearwasrecorded.In thecaseoftheinsertsbatchT1,afterapproximately12,500cuts,a flank wear width of 0.15mm (NC0.15) developed (see Fig. 7). Comparedwiththis,thewearbehaviourofthecoatedinsertsofthe

Fig.4.Inclinedimpacttestresultsat300_{Cand150Naftervariousnumberof}

impactsinbothinvestigatedcoatingcases.

Fig.5.Anoverviewoftheobtainedimpacttestresultsat25_Cand300_Cforboth

coatingcases.

Fig.6.Ramanspectraonthecoatingintheimpactimprint(positionsBandD)and outsideofit(positionsAandC)inbothcoatingcases.

Fig.7.Flankweardevelopmentversusthenumberofcutsoftheinvestigated diamondcoatedinsertsandtheirwearstatusaftervariousnumberofcuts. G.Skordarisetal./CIRPAnnals-ManufacturingTechnology68(2019)65–68 67

T2batchdeterioratessignificantly,asit canbeobserved inthe samefigure.Thelattercoatedinserts,uptothesameflankwear width,cutonly~8000times.Hereupon,flankwearwidthsof70

m

m and150

m

mappearedafterthenumberofcuts,illustratedatthe bottompartofFig.7.TheobtainedlargernumberofcutswhenT1 insertsareusedexhibittheirsuperiorwearresistancecomparedto theT2ones.

In order to explain these results, Raman spectroscopy was conductedontheremainingcoatingsonthetoolrakeaftermilling (see Fig. 8). In this way, potential crystallinity changes of the diamondcoatingsdue totheexercisedthermaland mechanical loads could be captured. Practically, thesame Ramanpatterns appearedinallexamined diamondcoatedsamplesofthesame batchbeforeandaftertheimpacttestsaswellasonthetoolrake aftermilling.InthecaseofT2coatings,theweakerspectralpeaks inthediagramatthemiddleofFig.8indicateadecompositionof the trans polyacetylene sp2_{-bonded phases (peaks at around} 1150 cm 1_{) as well as of the nano sp}3_{-bonded phase of high} crystallinity,whenthermalandmechanicalloadsduringmilling are simultaneously applied. Consequently, a diamond coating strengthweakeningarisesresultinginamorerapidwearevolution

comparedtoT1coatingsduringthecuttingprocesses.Hence,the lackofatranspolyacetylenephase,byadjustingappropriatelythe deposition temperature, can lead to an improved MCD coated insertscuttingperformance.

5. Conclusions

Inthispaper,theeffectofthecrystallinityofMCDcoatingson their fatigue strength and wear behaviour in milling was presented.Inthis context,MCDcoatingsweredepositedonthe samecementedcarbideinsertsbyvaryingthesubstrate tempera-ture during the deposition process for attaining different film crystallinities.Basedontheattainedimpactand millingresults, whenmechanicalandthermalloadsaresimultaneouslyexercised, thedecomposition ofa sp2 _{bonded transpolyacetylene phases,} potentiallyexistinginthediamondcoatingstructure,leadstoa moreintensewearcomparedtoatranspolyacetylenefreeMCD film.Asaresult,thefatiguestrengthandthecuttingperformance ofthelatterdiamondcoatedinsertscanbeimproved.

References

[1]Dumpala Ravikumar, Chandran Maneesh, Madhavan S, Ramamoorthy B, RamachandraRaoMS(2015)HighWearPerformanceoftheDual-layerGraded CompositeDiamondCoatedCuttingTools.InternationalJournalofRefractory MetalsandHardMaterials48:24–30.

[2]ZhangJianguo,YuanYigao,Zhang Jinjiang(2018) CuttingPerformance of Microcrystalline,NanocrystallineandDual-layerCompositeDiamondCoated Tools in DrillingCarbon Fiber Reinforced Plastics. Applied Sciences 2018 (8):1642.http://dx.doi.org/10.3390/app8091642.

[3]SkordarisG,BouzakisK-D,CharalampousP,KotsanisT,BouzakisE,LemmerO (2016)EffectofStructureandResidualStressesofDiamondCoatedCemented CarbideToolsontheFilmAdhesionandDevelopedWearMechanismsin Milling.CIRPAnnalsManufacturingTechnology65:101–104.

[4]GomezHumberto,DurhamaDelcie,XiaoXingcheng,LukitschMichael,LuPing, ChouKevin,SachdevAnil,KumarAshok(2012)AdhesionAnalysisandDry MachiningPerformance of CVDDiamond Coatings Deposited onSurface ModifiedWC–CoTurningInserts.JournalofMaterialsProcessingTechnology 212:523–533.

[5]DumpalaRavikumar,KumarN,KumaranCR,DashS,RamamoorthyB, Rama-chandraRaoMS(2014)AdhesionCharacteristicsofNano-and Micro-crystal-lineDiamondCoatings:RamanStressMappingoftheScratchTracks.Diamond andRelatedMaterials44:71–77.

[6]DumpalaRavikumar,ChandranManeesh,KumarN,DashS,RamamoorthyB, RamachandraRaoMS(2013)GrowthandCharacterizationofIntegrated Nano-andMicrocrystallineDualLayerCompositeDiamondCoatingsonWC–Co Sub-strates.InternationalJournalofRefractoryMetalsandHardMaterials37:127–133.

[7]FragaMA,ContinA,RodríguezLAA,VieiraJ,CamposRA,CoratEJ,TravaAiroldi VJ(2016)Nano-andMicrocrystallineDiamondDepositiononPretreatedWC– CoSubstrates:StructuralPropertiesandAdhesion.MaterialsResearchExpress3 (2):025601.

[8]ZhangJian-guo,WangXin-chang,ShenBin,SunFang-hong(2014)Effectof Deposition Parameters on Micro- and Nano-Crystalline Diamond Films GrowthonWC-CoSubstratesbyHFCVD.TransactionsofNonferrousMetals SocietyofChina24:3181–3188.

[9]BouzakisK-D,AsimakopoulosA,SkordarisG,PavlidouE,ErkensG(2007)The InclinedImpactTest:aNovelMethodfortheQuantificationoftheAdhesion PropertiesofPVDFilms.Wear262:1471–1478.

[10]BouzakisK-D,SkordarisG,BouzakisE,CharalampousP,KotsanisT,TasoulasD, KombogiannisS,LemmerO(2015)EffectoftheInterfaceFatigueStrengthof NCD CoatedHardmetal InsertsonTheir CuttingPerformance in Milling. DiamondandRelatedMaterials59:80–89.

[11]Ferrari AC,Robertson J(2001) Originof the1150-cm-1 RamanMode in NanocrystallineDiamond.PhysicalReviewB. http://dx.doi.org/10.1103/Phys-RevB.63.121405.

Fig.8.Ramanspectraoftherakeandoftheremainingcoatingaftermillingforboth coatingcases.

G.Skordarisetal./CIRPAnnals-ManufacturingTechnology68(2019)65–68 68