Surface properties of AISI 4140 steel modified by pulse plasma technique

w w w . j m r t . c o m . b r

Availableonlineatwww.sciencedirect.com

Original Article

Surface properties of AISI 4140 steel modified by

pulse plasma technique

Yıldız Yaralı Özbek

SakaryaUniversity,MetallurgicalandMaterialsEngineeringDepartment,EsentepeCampus,54187,Sakarya,Turkey

a r t i c l e i n f o

Articlehistory:

Received25January2019 Accepted17December2019 Availableonline25December2019

Keywords:

Steel

Surfaceanalysis Weartesting AFM Pulseplasma

a bs t r a c t

Thepulseplasmaprocessisakindofsurfacemodificationtechnique.Inthisstudy,the microstructureandmechanicalpropertiesofpulseplasma-treatedAISI4140steelwere studied.Fourdifferentsample-plasmagunnozzledistancesandthreedifferentpulseswere chosenforthesurfacemodificationataconstantbatterycapacityof800mf.Thesamples weresubjectedtoopticalmicroscope,SEMandEDSanalyses,microhardnesstestingand X-raydiffraction(XRD)analysis.Thecolumnarandfinegrainedstructureswereformed inmodifiedlayer.Newandhardphaseswereformedonthemodifiedlayer.Hence,the hardnessincreasedfivetimesafterpulseplasmatreatment.Theamountofwearforall specimenswasevaluatedbyusingthereciprocatingwear(linearweartestmachine)test witha0.15m/sconstantslidingspeedunder5,7,and9Nloadsalonga200mslidingdis- tance.AWCball(with6mmdiameter)wasusedinthistest.Thefrictioncoefficientand wearratewerechangedinaccordancewiththeappliedload.Thefrictioncoefficientvalues decreasedandthewearresistanceincreasedinthesurface-modifiedspecimenscompared tothenon-modifiedones.Thewearrateandthefrictioncoefficientwerechangedwiththe weardebrisandload.Thedebriswasincreasedbytheresistancetowearofsurface.Theworn surfacesofthespecimenswerestudiedbyusingatomicforcemicroscopy(AFM),scanning electronmicroscopy(SEM)andelectrondispersivespectroscopy(EDS).Theabrasivewear wasshownonwornsurface.

©2019TheAuthor.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Themicrostructureoftoolsteelsandtheirsurfaceproperties canbemodifiedwithdifferentsurfacetreatmentprocessesby pulsedlaser,ionorplasmabeams[1–6].Asoneofthesurface

∗ Correspondingauthor.

E-mail:yyarali@sakarya.edu.tr

modificationtechniques,thepulseplasmaprocessiswidely usedtoimprovethesurfacepropertiesoftoolsteels.[4–7].

The pulseplasma system consists of;the elasto-plastic deformation,impactbysoundandthepulsedmagneticfield, heat,electric-pulsetreatment,deformationofthemetalsand alloys duringthe operation. Ajet havingenergydensity is formedviathedetonationcombustionoffuelgasmixtures.

Superpositionofanelectromagneticfieldontoadetonation wavetransformsthelatterintoaplasmapulseduetothepres- enceofextraenergypresent[4].Thesurfacelayerofthetarget https://doi.org/10.1016/j.jmrt.2019.12.048

2238-7854/©2019 The Author. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/4.0/).

Fig.1–Schematicpresentationofthepulsed-plasma modificationsystem:1-detonationchamber,2-central electrode,anode,3-conicalelectrode,cathode, 4-interelectrodegap,5-consumableelectrode,6-power supply,7-gapbetweentheelectrodes,8-pulsedplasma forming,9-worksurface.

exposetoarapidmeltandsolidificationwithheatingandcool- ingratestypicallyintherangeof10⁷–10¹⁰K/s.Afterthepulse plasmaprocess,thesurfacelayercanhavehighanti-friction propertiesandwearresistance.Thepulseplasmahasnosize limitationorresidualstressproblems,andthus,itisasuit- ablesurfacemodificationtechniquetotreatcomplex-shaped industrialcomponents[1–4,6].Comparedtolasertreatment, electronbeamtreatmentandconventionalionimplantation, Pulseplasmatreatmenthashighenergyconversionandpro- cessingefficiencyandfacileprocessing[6–8].

Inthiswork,thesubstratewaschosenasAISI4140.AISI 4140steel,whichiseasilyobtainableandextensivelyusedfor industrialsurfacemodificationapplications, wasstudiedto determinetheeffectsofpulseplasmatreatmentparameters onthestructure.Thechromium–molybdenumalloysteelsare knowntobewidelyusedinthestructural,automotiveand gasindustriesduetotheir superiorhardenabilityand high strength [7,8]. Inaddition, the effects of the pulseplasma parametersonthereciprocatingslidingwearpropertiesand therelatedwearmechanismswereinvestigated.

2. Experimental procedure

ThepulseplasmasystemisshowninFig.1.Theplasmatron consistsofadetonationchamberwherethefuelgasmixture isoccuredanditsdetonationcombustionisinitiated,acentral electrode-anode,aconicalelectrode-cathode,inter-electrode gap(4),aconsumableelectrode(5)andapowersupply(6).

Thecyclicthermaleffectwithafrequencyof3−5Hzleads periodicheatingofthesurfaceuptoameltingpoint,which altersthephasestateofthemetallayerandacceleratesthe transferofalloyingelements.Multipleheatingandcoolingof

Table1–Chemicalcompositionof4140steelusedfor pulseplasmasurfacemodification.

%W C Si Mn P S Cr Mo

AISI4140 0.40 0.30 0.70 0.035 0.035 0.98 0.27

Table2–Samplecodesandthepulseplasma parametersofthespecimens.

Sampleno Nozzle-samplespacing(mm) Numberofpulse

1 70 15

2 70 10

3 70 5

4 80 15

5 80 10

6 80 5

7 60 15

8 60 10

9 60 5

10 50 15

11 50 10

12 50 5

the surfacelayerwithhightemperaturegradients causeto periodicchangesinstressesandstrainsinthislayer,which leadstosubstantialchangesinthestructuralstateofametal alloy.Plasma-detonationtreatmentisparticipatedbyapulsed mechanical pressure of the high-velocity plasma jet. The energy(upto4kJ)offreegas-dynamicshocksisdetermined bymassandvelocityoftheplasmajet.Mechanicaleffectis considered tobe anefficientmeans on theaccelerationof chemicalandmass-exchangeprocessesoccurringinasolid body.Itactivatesoscillationprocessesinametalalloy,excites long-wave acousticphonons,thus accelerating coolingand crystallisation,andintensifiesmasstransferofalloyingele- mentsintheheatedlayer[1].

Theworkpiecesurfacesofthesamplesweresubjectedto detonationprocessaccompaniedunderaplasmaatmosphere containingalloyingelementsdissolvedfromthemetalelec- troderodininteractionwiththeconstituentsofpropaneand nitrogengasesusedfortheprocess.Thetungstenwaschosen asconsumableelectrodeinthiswork.Thisprocessprovided theworkpiecesurfaceswithalloyingofplasmacomponents andhardeningofsurfaces[1,4,7,8].

Becauseitiscosteffectiveandcommonlyusedinindus- trialapplications,AISI4140steelwasselectedasthesubstrate material. The chemical composition ofthe AISI 4140steel was giveninTable1. Thediameter ofthecylindrical sam- pleswas22mm, andtheirthicknesswas10mm.Theywere machinedbyaCNClathe.Themachinedsampleswerenot subjectedtoanytreatmentpriortothepulseplasma.Thepro- cessparametersthatwereappliedtothemachinedsamples werepresentedinTable2withtheirsamplecodes.

2.2. Surfacecharacterization

Afterthepulseplasmasurfacemodification,thecylindrical specimenswerecutintocross-sectionstomeasurethecase depth.Thespecimensweregrindedby120,240,320,400,600, 800,1000,2400and4000meshemerypapers.Then,thesam- pleswerepolishedwith3␮mdiamondpasteandetchedby

Fig.2–Micrographcross-sectionofmodifiedAISI4140 steelsamples(distance40mm,10pulse).

using3wt. %nitalsolution; thenthe microstructureswere observedbymeans ofanopticalmicroscope,SEMandEDS (JEOL-JSM6060LV).Thestructuresofpulse plasma-treated samples were investigated by X-ray diffraction (XRD) by a Rigaku D/MAX/2200/PC model device with Cu-K_␣ radiation (␭=0.154056nm).

Thecross-sectionalhardnessvaluesofthespecimenswere measuredataspacingof10␮mfromthemodifiedsurfaceat 5gloadsforadwelltimeof15s(LeicaVMHTdevice).Threesets ofcross-sectionalmicrohardnessmeasurementswerecarried outforeachsample.

2.3. Weartests

Wearandfrictiontestswereperformedinalinearweartest machinenamedCSMtribometer.Thetestswerecarriedoutat ambienttemperature(28^◦C±3^◦C)underacontrolledhumid- itythatwasbetween40and50%underdryslidingconditions.

Theslidingdistance(way)is12mmonwearsurface.Therecip- rocatingslidingmodewascalculatedwitha0.15m/sconstant slidingspeedunder5,7and9Nloadsina200mslidingdis- tance.Eachofthetestswasrepeatedatleastthreetimesto ensure the accuracyofthe obtainedwear andfriction val- ues Asphericalcounter-partwasoscillatingundertheload againstthe modifiedsteelsample. AWCball witha6mm diameterwasutilizedasthecounter-part.Thefrictionforce wascontinuouslyrecordedbythesensorsatthetestblockso thatthecoefficientoffrictioncouldbecalculatedwithrespect tonormalforce.Theresultsshownherewereselectedfrom numeroustestsperformedatdifferentsamples.Thefriction coefficientandthewearratevariedasafunctionofload.The systemenablesmeasuringfrictioncoefficientandtimedepen- dentdepthprofilesbyusingsensitivetransducers.Thedepth transducerwaslocatedverticallyontopofthesample.

Aftertheweartest,thewornsurfaceswerecharacterized bySEM,EDSandAFManalysis.Thecalculationsofthewear ratewere performedbyusingaPerthometerMAHRsurface roughnessapparatusaftertheweartest.Theamountofwear onthe surfaceaftereach testwascalculatedbymeasuring thewearwidth,depthandtheamountofsurfaceroughness byusingasurfaceprofilometerandlowmagnificationoptical micrographs.

Fig.3–SEMmicrographof(a)sample11(50mm–10pulse),(b)sample7(60mm–15pulse).

Fig.4–XRDresultsof(a)untreatedsample,(b)sample1,sample2andsample3(70mmdistance,15-10-5pulsenumber), (c)sample4,sample5andsample6(80mmdistance,15-10-5pulsenumber).

3. Results and discussions

Fig.2showstheopticalmicrographofmodifiedsample.As expected,forclassicalsurfacemodificationtreatments,asur- facelayerwascomposedofanexternalcompoundlayerwith anunderlyingdiffusionlayer[9–13].

Thecompoundlayer,thediffusionlayerandthesubstrate wereeasilyobservedduetothecontrastincolour.Nitrogen thatwasdiffusedintothesteelsurfacewascombinedwith analloyingelementalongtheelectrodetoformfinedispersed nitrides.Asaresult,athinironnitridelayer,whichcomposes ofnitridephasesthatismentionedtothewhitelayerdueto thelackofetchant,wasproducedonthesurface[16].There werehigh concentrationofnitrogenionsandthe presence ofweightynon-ionizednitrogenmolecules(compoundlayer) [11].Thecompoundlayerhasaporousstructure.Thesefind- ingsaresimilartotheexperimentalresultsreportedbySirin etal.[10].StahliandSturzeneggerinvestigatedtheformation ofathinmodifiedlayerofanewstructureandsameresults wereobserved[12].

Dependingonthenitrogenconcentrationinthegasmix- tureutilizedduringpulseplasmaprocess,thecompoundlayer maybeconstitutedby␧-Fe2–3Nor␥Fe-Fe4Nphasesoramix- tureofboth.

Fig. 3(a and b) shows SEM micrograph of sample 11 (50mm–10pulse)andsample7(60mm–15pulse).Inthesefig- ures,plastic deformation presents in the modified surface

Fig.5–Relationshipbetweenpulseplasmaparametersand microhardness.

layer,wherethemicrostructuralmorphologydiffersfromthe matrix.Existence ofa veryfine microstructurewasassoci- ated with the plastic deformation, which was believed to occur because ofthe repeated heatingand coolingdue to theincreasingnumberofpulses.Thedeformationofmetallic materialscanenhancethenitrogendiffusionintoiron,leading toadecreasedgrainsizeanddislocationdensity[16].

Thethicknessofthemodifiedlayerswaschangedasafunc- tionofthedifferentprocessparameters.When thenumber

Fig.6–Therelationshipbetweenthewearrateandapplied loadformodifiedsamplesfor5pulse.

Fig.7–Therelationshipbetweenthewearrateandapplied loadformodifiedsamplesfor15pulse.

ofpulseswasincreased,thecompound anddiffusionlayer thicknesseswerealsoincreasedshowninFig.3(aandb)[11].

Thiseffectisduetothehightemperatureofthesurfacedur- ingtheplasmatreatmentwhichresultsinthemeltingsurface layerandthesubsequentliquidphasemixingoftheionized gases,whichweredepositedearlier,withthesubstratemate- rial[12,15].Thus,anincreaseinthepulsenumberresultsin adeeperpenetrationoftungstenandnitrogenintothebulk, andtungsten andnitrogen-rich layersareproduced onthe steelsurface.YangLietal.[9]alsoreportedthesamefindings for4140steel.

Additionally,thenozzle-samplespacingisanotherimpor- tantparameter.Increasingthenozzledistanceresultedina decreaseinthicknessofthemodificationlayerduetothepro- jectileeffectoftheplasma.Anotherreasonforthisdecrease isthattheionizedgasesthatwereexhaustedfromthenoz- zlecouldnotreachthesurfaceofthespecimenclearlyand homogeneously.Whenthenozzle-sampledistancedecreases, the surfaceisoverheated.The fusedareas can beseenon thesurfaceinthiscase.Theseresultsareinagreementwith theliteratureas[10,14]Tyurinandfriendsreportedthesame results[8].

Fig.8–Thefrictioncoefficientvaluesfordifferentload, nozzlespacingandpulsenumbers.

ThisprocesswasalsoappliedtoM2steelsurface[20].The M2steel(highspeedsteel)ishighalloysteel.However,the same morphology was observed in both steels [20]. These worksshowthat,thepulseplasmaprocesscanbeappliedto thesteelswhosecarboncontentishigherthan0.4wt. %C.

Highamountsofalloyingelementsinsteelarenotnecessary forthisprocess.

Fig.4showstheX-raydiffractionprofilesoftheuntreated sampleandthesamplesmodifiedbythepulseplasma.Sam- ples1,2and3weretreatedin70mmspacingwith15,10,5 pulses,respectively;samples4,5and6weretreatedin80mm spacingwith15,10,5pulses,respectively.Thesamplescon- tainedstrongdiffractionpeaksforthe␣-Fephase,aswellas weakdiffractionpeaksfor␥-Fe,Fe2-3N(hcp),W3Oand FeN [16,20].Thenitrideandtungsten-richphasescanbeformed duringshortpulseheating,theinteractionofthephaseshave introducedimpurities,andfurtherrapidcooling[9–11].Addi- tionally,theintensityofdiffractionpeaksdecreasedcompared to theuntreated specimen,which suggests thatthe grains ofthepulseplasma-treatedspecimenwererefined andthe meanlatticemicrostrainwasincreased.Theamountoftung- stenoxidephaseincreaseswithdecreasingthenozzle-sample spacingasobservedinsample1.Lietal.suggestedthatthe Fe₂-₃NandFeNnitridephasesincreasedduetotheincreasing pulsenumber[13].

TheXRDpatternofthetreatedspecimenshowsthatthe fullwidthathalfmaximum(FWHM)ofthediffractionpeak wasbroadened.IthasbeenwellknownthattheFWHMbroad- eningoccursasaresultofdecreasedgrainsize[21–23].The residualstressoccursafterrapidpulse-treatment.Thecom- pressiveresidualstressonthesurfaceleadtoabroadeningof thediffractionpeaks[21].

Fig.5displaysthemicrohardnessvaluesofthesamplesas afunctionofdifferentparameters.Thehighestsurfacemicro- hardnessvalueofthemodifiedAISI4140steelwasmeasured as950HV0.05 ata15cyclepulseand50mmnozzlespacing, whereasthelowestmicrohardnessvaluewasmeasuredas700 HV0.05at5-cyclepulseand80mmnozzlespacing.Thehard- nessincreased5timeswithrespecttooriginal,untreatedbase metalthatwasmeasuredas180HV.

Fig.9–SEManalysesofsamplesafterweartesting(a),untreatmentsample,(b)weartrackofsample2for5N,(c)sample2 for7N.

Thehardnessgradually decreases dueto the alloycon- centration nearthe core, resulting in a diffused case–core interface.Thefastcoolingandsolidificationofsurfacemelting materialcouldproduceveryfinemicrostructure.Thesurface layerconsistedofnano-crystallinegrainsandaplasticlayer thatwasseverelydeformed,whichiscomposedofsub-grains withmany defects, suchas dislocations and grain bound- ariesthatcan beformedinthe pulseplasmamethod. The ultra-finegrainsareoneofthereasonsforsurfacehardening.

Themicrostructureofmodifiedsurfacewillbecomefinerwith increasingnumberofpulses[11].

Therelationshipbetweenamaterial’smicro-hardnessand itsgrainsizecanbeexplainedbytheHall–Petchformula;

␴y= ␴0+ky/√d

Where␴yistheyieldstrengthofmaterialwhichcanusuallybe replacedbymicro-hardness(HV),␴0isthelatticefrictionresis- tance,kyisaconstant,anddistheaveragegraindiameter.

Accordingtothisequation,themicro-hardnessofamaterial isinverselyproportionaltoitsaveragegrainsize.Inaddition, thenanostructuredsurfacelayeroftheAISI4140steelsignifi- cantlypromotesthediffusionofgaseswhenmanypulsesare used,resultingintheformationofnewhardphases[24].We cansafelyproposethatthehardphasesandnewstructure improvedthesurfacehardness[25–27].

Thepulseplasmaprocesscanbeappliedtodifferentsteel groups[28].Notonlythe4140steelbutalsoM2steelhardness valueswereincreasedafterthepulseplasmatreatment.The hardnessvaluesofhighspeedsteel(i.e.M2)wasincreased4 times[20].Thispropertyisveryimportanttosteelindustries.

During the modification process, the rapid heating and coolinggeneratessmallgrainsresultinginbetterwearresis- tance [13,14]. Thehard compound layercould significantly increasethewearresistanceoftheplasma-modified4140steel comparedtotheuntreatedzone[28–30].Thecracksarealways initiated from the contactsurfacesfor bothuntreated and pulse-plasmatreatedsamples.Theoccurrenceofcompound layeraffectsthecrackinitiationandthecolumnarstructure occurredindiffusionzoneretardsthecrackthepropagation onsurface[13].Therefore,thecompoundlayeronthesurface shouldhaveasignificantinfluenceonthesurfacewearbehav- ior.Intraditionalnitridingprocess,thecompoundlayercan bebrittle. Inpulseplasmatreatment, the surfacebecomes much moreductile than traditionally nitrided surface [13], due to the sputteringeffect during the pulseplasma pro- cess. Thecompoundlayerhasacertain degreeofductility despiteitshighhardness.Thiswillbringadditionalbenefits tothemodifiedcompoundlayerwithrespecttowearprotec- tion.However,Gualtieriet al.usedtheHall–Petch equation [19],andtheyfoundthattherelationshipbetweenthegrain sizereductionandhardnessincrementobeystheHall–Petch

Fig.10–SEMandEDSresultsofafterweartest(a)sample3(at5Nload),(b)sample3(at9Nload),(c)sample7(at5Nload), (d)sample7(at9Nload).

Fig.11–TheAFMresultsofwearsurfacesample3for(a)5Nload,(b)9Nload.

behavior[13,14].Withareductioningrainsize,thedislocation activitybecomeslimited, preventingcrackpropagationand grainboundaryslidingoccurs(i.eHall–Petcheffect).Also,the presenceofnitrogenandtungstenbasedphasescanexplain theimprovementintribologicalproperties[24,27,28].Because ofthehighnitrogenconcentrationinironlattice,thesignif- icantcompressivestressesoccurredwhichhasanimportant roleinhardnessincrementandimprovesthewearresistance [6–15].C.X.lietal.andGuanetal.indicatedthatthenitrogen basedphasesimprovedthewearpropertiesofsurface,sharply [12–16].

Theparametersofpulseplasmaaffectedthegrainsizeand columnarstructure.Inthemaintime,theincreaseinenergy densityandthenumberofpulsesledtolowersurfacerough- ness[14].Thesmallertheratiooftemperaturegradient(G) tocrystallizationrate(R),thegreaterprobabilityoffinecrys- tallinestructureformation,whilealargerG/Rratiopromotes theformationofcolumnarcrystalstructures.Thecolumnar structurecontributestomechanicalproperties.Especially,this structurecanbepreventscrackpropagation.

Afterpulseplasmatreatment,themodifiedsurfacelayer providedabetterperformancebyuntreatmentsurface.The

Fig.12–AFMresultsofwearsurfacesample5for(a)5Nload,(b)9Nload.

modifiedlayershavebeen usedtoimprove the tribological behaviour[13–15].Theeffectofthenozzledistance(mm)on wearrate(mm³/n.m)forfivepulseswasdisplayedinFig.6.

When the nozzle increases, the wear rate increases for 5 pulses.InFig.7,thewearratechangeswithnozzlespacing when15pulseswereused.

The wear rate decreased from 2.1×10⁻⁷ to 1×10⁻⁷mm³/n.m after surface modification treatment.

Thesimilarratios were calculated byOzbek etal. [15]and Guana et al. [16]. Formation of higher amount of cracks causesfineweardebrisbecauseofsmallsizeddelamination layersformedbycrackpropagation,breakageand adhesion oftheweardebris.Thecolumnarstructureimprovethewear properties.Especially,thisstructurecanbepreventedcrack propagation[17–24].

Theoptimumwearratewasobservedin60mmdistances.

Thewearratewaschangedwithloadandpulsenumber.When the load was increased, the wear rate also was increased [15].The largest wearrates were obtainedunder9Nloads forallsamples.Thewearresistanceofthemodifiedsurface increasedwhenthenumberofpulseswasincreased.Thecom- poundlayerswerequiteeasytobebrittleafterpulseplasma process.Onepossiblesolutionwastoenhancethenumberof pulsesonthesurface.Whenthenumberofpulsesincreased the wear properties of surface was improved due to high energyandalloyelements(nitrideandtungsten)[10,23,25].

There fast heating and melting procedures, where the dissolutionofprecipitationhardeningparticlesandinterdif- fusionwiththeencircledmaterialwillbegreatlypromoted.

Thelocalconcentrationofalloyingelementscanbedestroyed effectivelyinthemodificationlayer.Duetotheshortduration of pulse plasma effect of modification layer, the high- temperaturesurfacelayerundergoessuperfastsolidification and cooling processes. Thus, the homogenous distribution andsuper-saturateddissolutionofalloyingelementscanbe retainedalongwiththeformationofremeltedlayeroriented preferentiallyalongdirection.

Withtheincreasingpulsesthemicrostructureindepthof severalmicrometerswasrefined[26].Intheprocessofcooling, therateoftemperaturechangecanbereachedto10⁸K/s.On onehand,thiskindofextremeprocessingpromotedprecip- itationofagreatnumberofsmallsizegrainandamorphous nitridealloys[12–17]withhighbrittlenesstherewasthether- malstress concentratedonmodified surface.Thesefactors

wereaffectedthewearproperties.Thewearpropertieswere improvedthesurfaceproperties.Thepulsenumberwaspos- itivelyaffected[18–20].

Thecoefficientsoffrictionobtainedfrom theweartests forthesamplesunder5,7and9NwereshowninFig.8.The frictioncoefficientoftheuntreatedsamplewas0.49.Thecoef- ficient offrictionforthe untreated samplekeptincreasing during thefirst 2−3hand then becamerelatively stableat ahighvalue.However,thefrictioncoefficientsofthemodi- fiedsamplesonlychangedmerelybetween0.12and0.19.The distanceaffectedthefrictioncoefficient.AsshowninFig.8, thereisprimarilyaseparatelowfrictioncoefficientwindow, whichbeganatthe70mmnozzledistanceandextendedtothe 50mmdistance.Asdecreasingthenozzledistanceto50mm resultedinanincreasedhardnessandanincreasedwearresis- tance,theshearforcesdecreasedandcausedtolowfriction coefficient.Oneinterestingresultisthatforthe70mmnozzle distanceunderalloftheappliedloadconditions,thefriction coefficientislow.

Despite oftheincreasedhardnessforthe50mm nozzle distance,thefrictioncoefficientonlyslightlyincreasedunder low load (5N) conditions. This isbelieved to berelated to the wear mechanisms and the microcrack formation, and therelatedabrasivemechanismisresponsibleforthehigher frictioncoefficient.Undertheincreasedloadconditions,the surfacetemperaturerisesandleadstomoreoxidation,which isbeneficialindecreasingthefrictioncoefficient.

Qualitativeassessmentofthewearscarswascarriedout using SEM. Fig. 12 shows a plan view secondary-electron (SE) micrograph. The analysis has shown the presence of the compound layer because of its compact and closely packedhexagonalstructureandhighernitrogencontent.The presence ofa compound layerthat has very good friction characteristicsproducedevenlowerfrictioncoefficients.This resultisinagreementwithFattah’sstudy[30].Thecompound layerprovidesthetribologicalcharacteristics,whilethedif- fusion zonedetermines thestrength ofthe modified layer.

TheFig.9(a)showsthewornsurfaceofuntreatedsample.The wornsurfacesexhibitedaseizureoftheweardebrisonthe slidingsurfaceandthesubsequentplasticdeformationhard- ening,asevidencedbythemicro-cracksonthesurface.The abrasivewearoccurredonthewornsurfacesandwasasso- ciatedwithvariousgrooves.Thus,micro-cuttingisthemain wearmechanisminthesesamples[25].

Itisalsopossibletoservetheformationoffinewearparti- cles,intheshapeofplateletsthatdevelopedinthesurfaceof thediffusionlayer.Therepeatedloadingactionbytheharder asperitiesand/orweardebrisonthe softersurfaceinduces plasticdeformation,whichincreaseswiththeprogressofwear test. This behaviour can be explained by the complete or almostcompletedestructionofthesuperficialmodifiedlayer.

Inadditiontothesefreeparticlesactingasathirdbody, therearealsohardasperitiesonthecounter-bodythatcause abrasionbetweenthetwobodies.Itisnecessarytoconsider thatthetypeandconfigurationusedinthesetests,thatis,the reciprocating andsphere-plane, contribute tomaintainthe weardebrisinsidethesystemduetothebidirectionalmove- mentandinexistenceofcentrifugalforces.Inthistypeoftest, thefreeweardebrisistransportedforwardsandbackwards, andbeingeliminatedfromthesystemintheperpendicular directionofthesamplemovement.

Theweartracksofsample2wereshowninFig.9(bandc).

Theweartrackwidthis497␮munder5Nload,whileaverage is580␮munder7Nloads.Thewidthofweartrackincreases withincreasingwearload.Duringtheinitialperiodofslid- ing,thecompoundlayerwithahighstressfracturedandthen transformedtheabrasiveparticles.Afterremovalofthecom- poundlayer,theweartrackexhibitedplasticdeformationand deepgrooves.Thewidespreadingwearmechanismcombines theadhesivewearandabrasiveness,whichwasfoundforallof thepulseplasma-treatedspecimens.However,thepresenceof acompoundlayercausestheappearanceofamoreabrasion- wearcomponentasthecompoundlayerbreaksdownduring slidingandbecomeshard[28].

Fig. 10 shows the SEM-EDS morphologies of the worn surfaces for sample 3 (Fig.10a,b) and sample 7 (Fig.10c,d) demonstratingthe different wear mechanisms under both 5Nand9Nloads.TheEDSresultsshowoxygenrichareas, which suggests oxidative wear on the modified surfaces.

When load increases, the characterizationof wearmecha- nismhaschangedandseverewearrateintensityisseenasin Fig.10(bandd).Thewornsurfacesexhibitedaseizureofthe weardebrisontheslidingsurfaceandthesubsequentabrasive deformationisevidencedbythemicro-cracksonthesurface.

Thenitridesphasesonsurfacewasaffectedtocharacteriza- tionofwear.Thebrittlelayerwasformeddelaminationzone [16].Cracks,weardebrisandsurfacelayerdelaminationwere observedatthemodifiedsurface.Parallelgrovesinthedirec- tionoftheintendeddisplacementwereformedattheworn area[25–28,31].

Fig.11(aandb)displaytheAFMresultsofthesurfacefor sample 3after wear test. The surface morphology is usu- ally flat and compact; the structural defects includesmall holes or caves. Theabrasivewear wasformed on thesur- face. Surfaceruptures formedduringthe wear teston the surface,andtrenchesformedduetomaterialremovalaswell asweardebris nearthe 2000nm×20nmarea.Theregions wherehigherasperitiesoccurred were maintainedbytheir highwearresistance.Theweardepthisnothigh.Thesurface topographywaswelldescribedevenwhentheloadincreases [31,32].

Thecharacterofthe surfacelayeraftertheweartest is clearly visible by AFM analysis. Fig. 12(a and b) show the specimen-5alsosufferedfromacombinationofadhesiveand

abrasivewearwithasmootherweartrackwasgained.Thedif- ferenceofelevationonsurfacescanbeseeneasily,whilethe loadchanges.Theheightofyellowcolourareaishigherthan thatofredarea.Thereiscleavagefailurebetweenmodificated surfacelayersduetowearprocess.

4. Conclusions

• The microstructure of the surface layer in all plasma- modifiedspecimensconsistedofacompoundlayeranda diffusionzonefromthesurfacetoinward.

• The rapid heating - cooling and diffusion leading to a decreaseingrainsizeonthesurface.

• Theprocessparameters,suchasthenozzledistanceandthe numberofpulses,werenecessarytoimproveandchange the surface mechanical properties. Thethickness of the modifiedlayerincreasedwithincreasingnumberofpulses andattheoptimumnozzledistance.

• Theconsumableelectrodecauses theionizationofTung- sten(W)andNitrogen(N)atomswhereandtheseionsare dopedintothesurfacebymeansofdiffusionmechanism.

• The surfacemodification process forthe AISI 4140steel produced asurfacelayerwithmorefavorableproperties.

Tungsten and nitride-rich phases such as Fe2-3N, W3O, W, FeN and ␥-Fe were formed; due tothe pulseplasma treatment. The intensity of peak which stands for the relevantphaseswasincreasedwithincreasednumberof pulse

• Thehardnessvalueof4140steelcanbeincreasedupto4–6 timeshighercomparedtountreatedsampleinaveryshort time(e.g.1min.)bymeansofpulseplasmatreatment.The newhardphasesandstructurewereimprovedthemicro- hardnessvalue. Theincreaseinpulse numberofenergy absorbedbythesurfaceleadstotheincreaseinmicrohard- nessvalue

• Themodifiedlayerprovidedwearandanti-scuffingprop- ertiestothesurface.Thewearresistanceofthemodified surfacewasapproximately2–3timesgreaterthanunmodi- fiedsurfaceduetothechangeinthemodifiedsurface.The hardphasesandstructureplayakeyroleinimprovingthe wearresistancebyincreasingthesurfaceproperties.

• The friction coefficients were decreased after the pulse plasmatreatment.

• Thevaluesofwearresistanceandfrictioncoefficientsare changedwithprocessparameters.

• Theprocessparametersaffectedthewearpropertiessuch asnumberofpulseanddistance.

• Whentheincreasethenumberofpulseincreased,thewear ratewasdecreased.

• Wearrateandcoefficientoffrictionweredecreasedwhile thenozzledistancedecreased

• Abrasive and adhesive wear were occurred on surface.

Thereweresmallcracksonsurfaceafterweartest.

Acknowledgment

ThisworkwassupportedbydepartmentofGovernmentPlan- ningwithgrantnumber2003K1209701

Appendix A. Supplementary data

Supplementary material related to this arti- cle can be found, in the online version, at doi:https://doi.org/10.1016/j.jmrt.2019.12.048.

references

[1]ByrkaaOV,ChebotarevaVV,DerepovskiaNT,MüllerbG, SchumacherbG,etal.Propertiesofmodifiedsurfacelayers ofindustrialsteelsamplesprocessedbypulsedplasma streams.Vacuum2000;58:195–201.

[2]SartowskaaB,PiekoszewskiaJ,Wali´saL,StanisławskiJ, NowickiL,RatajczakR,etal.Thermalstabilityofthephases formedinthenearsurfacelayersofunalloyedsteelsby nitrogenpulsedplasmatreatment.Vacuum2007;81:

1188–90.

[3]PodgornikB,Vi ˇzintinJ.Slidingandpittingwearresistanceof plasmaandpulseplasmanitridedsteel.SurfEng

2001;17(4):300–4.

[4]ÖzbekYY,DurmanM,AkbulutH.Wearbehaviourofsteel modıfıedbypulseplasmatechnıque.TribolTrans 2009;52:213–22.

[5]BoltonM,CockremJ,FarinottiA.Pulseplasmanitridingfor plasmavapourdepositioncoatingofhighspeedsteels.Surf Eng2004;20(3):167–73.

[6]UglovVV,AnishchikVM,CherendaNN,SveshnikovYuV, AstashynskiVM,KostyukevichEA,etal.Theformationofa tungstencontainingsurfacelayerinacarbonsteelby compressionplasmaflow.SurfCoatTechnol

2008;202:2439–42.

[7]PodgornikB,HogmarkS,SandbergO,LeskovsekV.Wear resistanceandanti-stickingpropertiesofduplextreated formingtoolsteel.Wear2003;254:1113–21.

[8]TyurinNY,KolısnıchenkoOV,TsygankovNG.Thepulse plasmatechnology.PatonWeldJ2004;1:38–43.

[9]LangneraJ,PiekoszewskiaJ,WerneraZ,TereshinVI, ChebotarevVV,GarkushaI,etal.Surfacemodificationof constructionalsteelsbyirradiationwithhighintensity pulsednitrogenplasmabeams.SurfCoatTechnol 2000;128-129:105–11.

[10]UlutanM,CelikON,GasanH,ErU.Effectofdifferentsurface treatmentmethodsonthefriction,wearbehaviorofAISI 4140steel.JMaterSciTechnol2010;26:251–7.

[11]DewanaMW,LiangaJ,WahabaMA,OkeAM.Effectof post-weldheattreatmentandelectrolyticplasmaprocessing ontungsteninertgasweldedAISI4140alloysteel.MaterDes 2014;54:6–13.

[12]SartowskaB,PiekoszewskiJ,WaliL,KopcewiczM,WernerZ, StanislawskiJ,etal.Phasechangesinsteelsirradiatedwith intensepulsedplasmabeams.Vacuum2003;70:285–91.

[13]LiCX,SunbY,BellT.Factorsinfluencingfrettingfatigue propertiesofplasma-nitridedlowalloysteel.MaterSciEng 2000;A292:18–259.

[14]SirinSY,SirinK,KalucE.Effectoftheionnitridingsurface hardeningprocessonfatiguebehaviorofAISI4340steel.

MaterCharact2008;59:351–8.

[15]OzbekYY,SariogluC,DurmanM.Theeffectofplasma detonationparametersonresidualstressesdevelopedinthe plasmamodifiedlayer.Vacuum2014;106:11–5.

[16]GuanaQF,ZouaH,ZouaGT,WubAM,HaobSZ,ZoubJX,etal.

Surfacenanostructureandamorphousstateofalowcarbon steelinducedbyhigh-currentpulsedelectronbeam.Surf CoatTechnol2005;196:145–9.

[17]AnishchikaVM,UglovaVV,AstashynskiaVV,Astashynski VM,AnaninbSI,KostyukevichbEA,etal.Compressive plasmaflowsinteractionwithsteelsurface:structureand mechanicalpropertiesofmodifiedlayer.Vacuum 2003;70:269–74.

[18]AmanovA,SikPyunY,SasakiS.Effectsofultrasonic nanocrystallinesurfacemodification(UNSM)techniqueon thetribologicalbehaviorofsinteredCu-basedalloy.TribolInt 2014;72:187–97.

[19]YanaMF,WuaYQ,LiRL.Grainandgrainboundary charactersinsurfacelayerofuntreatedandplasma nitrocarburized18Nimaragingsteelwithnanocrystalline structure.ApplSurfSci2013;273:520–6.

[20]ÖzbekYY,AkbulutH,DurmanM.SurfacepropertiesofM2 steeltreatedbypulseplasmatechnique.Vacuum 2015;122:90–5.

[21]GualtieriE,BorghiA,CalabriL,PugnoN,ValeriS.Increasing nanohardnessandreducingfrictionofnitridesteelbylaser surfacetexturing.TribolInt2009;42:699–705.

[22]WangYX,YanMF,LiB,GuoLX,ZhangCS,ZhangYX,etal.

Surfacepropertiesoflowalloysteeltreatedbyplasma nitrocarburizingpriortolaserquenchingprocess.OptLaser Technol2015;67:57–64.

[23]SartowskaB,PiekoszewskiJ,Wali´sL,SenatorskiJ,

StanisławskiJ,RatajczakR,etal.Structuralandtribological propertiesofcarbonsteelsmodifiedbyplasmapulses containinginertandactiveions.SurfCoatTechnol 2007;201:8295–8.

[24]DingshunS,WenY,ZhiqiangF,YanhongG,ChengbiaoW, JiajunL.Theeffectofnitridingtemperatureonhardnessand microstructureofdiesteelpre-treatedbyultrasoniccold forgingtechnology.MaterDes2013;49:392–9.

[25]ZhangaCS,YanaMF,SunaZ,WangaYX,YouaY,BaibB,etal.

OptimizingthemechanicalpropertiesofM50NiLsteelby plasmanitrocarburizing.ApplSurfSci2014;315:28–35.

[26]YanqiuX,ShijieW,FengZ,HaizhongW,YiminL,TaoX.

Tribologicalpropertiesofplasmanitridedstainlesssteel againstSAE52100steelunderionicliquidlubrication condition.TribolInt2006;39:635–40.

[27]YangL,LiangW,DandanZ,ShenLie.Theeffectofsurface nanocrystallizationonplasmanitridingbehaviourofAISI 4140steel.ApplSurfSci2010;257:979–84.

[28]FattahM,MahboubiF.Comparisonofferriticandaustenitic plasmanitridingandnitrocarburizingbehaviorofAISI4140 lowalloysteel.MaterDes2010:3915–21.

[29]YanMF,WangaYX,ChenXT,GuoLX,ZhangCS,YouY,etal.

Laserquenchingofplasmanitrided30CrMnSiAsteel.Mater Des2014;58:154–60.

[30]GangopadhyayS,AcharyaR,ChattopadhyayAK,PaulS.

Effectofsubstratebiasvoltageonstructuralandmechanical propertiesofpulsedDCmagnetronsputteredTiN–MoSx compositecoatings.Vacuum2010;84:843–50.

[31]ShahaMS,SaleemaM,AhmadaR,ZakaullahM,QayyumaA, MurtazaG.Langmuirprobe,characterizationofnitrogen plasmaforjournalofmaterialsprocessingtechnology surfacenitridingofAISI-4140steel.JMaterProcessTechnol 2008;199:363–8.

[32]JradiK,SchmittM,BistacS.Influenceofthesurface chemistryonthenanotribologicalbehaviourof(AFM tip/graphite)couples.ApplSurfSci2012;258:4687–97.