Microstructure evolution and abrasive wear behavior of D2
steelKritika Singh, Rajesh K. Khatirkar, Sanjay G.
SapatenDepartment of Metallurgical and Materials Engineering,
Visvesvaraya National Institute of Technology (VNIT), South
Ambazari Road, Nagpur 440010,Maharashtra, Indiaa rti cle in
foArticle history:Received 28 August 2014Received in revised form5
February 2015Accepted 9 February 2015Available online 17 February
2015Keywords:Two body abrasionHardnessSteelWear testingElectron
microscopyabstractThe effect of heat treatment on microstructure
and abrasive wear resistance of AISI D2 steel has
beeninvestigatedinthepresentwork.
Thestructuralcharacterizationofhardenedandmultipletemperedspecimens
was carried out using scanning electron microscopy (SEM), energy
dispersive spectrometer(EDS) and X-ray diffraction (XRD). Two body
abrasive wear tests were carried out using silicon carbideabrasive
with systematic and simultaneous variation of test parameters. The
abrasive wear volume
lossincreasedwithincreasingtemperingtemperatureandincreasingseverity
oftest conditions, althoughtheincreasewasnotproportionate.
Thewornoutsurface,
weardebrisandsubsurfacedamagewereexaminedusingSEM, EDSandXRD.
Theresultsof present workwererationalizedwithrespect
tomicrostructure and operating wear mechanisms as inuenced by
severity of wear test conditions.& 2015 Elsevier B.V. All
rights reserved.1.
IntroductionHigh-carbonhigh-chromium(HCHCr)steelswererstdevel-oped
as a substitute for high speed (HS) steels, but were found tobe of
limited use due to insufcient hot hardness and brittleness.However,
thesesteelsproveduseful inapplicationswherehighwear resistance and
non-deforming properties were required e.g.indiesandpunches[14].
Thehighwear resistanceinHCHCrsteels is
attributedtohighvolumefractionof hardchromiumcarbides[4]. D2steel
ndsapplicationsindrawingandformingdies, cold drawing punches,
blanking/stamping dies and extrusiondies. Die steels are usually
subjected to compressive-tensilestresses, shear stresses and hence
die steels require high strengthand toughness apart from good wear
resistance [5].In general, an increase in hardness of material
results in an increaseinitswearresistance.
Thiscorrelationistrueonlyincaseof puremetals in the annealed
condition and alloys of same family [6]. Theresearch investigations
in the past have focused on improvement ofthe
wearresistancemainlyby
alterationofmicrostructurebyusingconventionalheattreatmenttechniques.
Theabrasive wear
losshasbeenreportedtodecreasewithincreaseinthevolumefractionofmartensite.
It has also been reported that the hardness of martensitehas a
greater inuence on the abrasive wear resistance of steels thanits
volume fraction [79]. The martensitic microstructure with
carbidesexhibited better wear resistance as compared to bainitic or
pearliticmicrostructures [10]. Theabrasivewear insteels
withmultiphasemicrostructure is inuenced by morphology of the
carbides, abrasiveparticlepropertiesandmaterialproperties[1119].
Torkamanietal.[20] noted that bright hardened samples of D2 steel
showed highertensile strength, impact toughness and higher hardness
in comparisontooilquenchedsampleswithuniformdistributionof
necarbides.Tanget al. [21] concludedthat thedryslidingwearratesof
heattreated D2 steel decreased with the increase in hardness in the
slidingspeedrangeof0.050.50 m/s.
Themodeofwearwasadhesiveforspecimens with hardness 5571 and 6271
HRc, whereas the modewas abrasive for specimens with hardness 5171
and 5871 HRc. Acomparative study of D2 and O1 tool steels in
hardened and doubletemperedcondition withthesamehardness(60
HRc)revealedthattheir wear resistance was inuenced by wear mode and
mechanismofmaterial removal under different test conditions [22].
D2 steelexhibited two times better wear resistance than O1 tool
steel, whichwasattributedtoplatelikecarbidemorphology.
Thestudydidnotreport the effect of retained austenite on wear
behavior of D2 steel. Maet al. [23] noted the decrease in mass loss
of D2 steel with increase
inslidingspeedandappliedloadwhichwasattributedtofractureofabrasive
particles under the used test conditions. The effect ofcryogenic
treatment on abrasive wear resistance of D2 steel has
alsobeeninvestigatedinthe past [2427]. The results revealed
thatprecipitationof secondarycarbidesandthepercentageof
retainedaustenite after quenching and cryogenic treatment
signicantly inu-enced the wear behavior under diverse test
conditions.Xu et al. [28] reported that wear resistance was optimum
at 30%volumefractionofretainedausteniteinhighspeedsteels.
Atlowerloads, conventionally heat treated samples exhibited better
wearresistance as compared to laser melted specimens, whereas at
higherloads superior wear resistance of laser melted tool steel was
attributedto stress induced transformation of austenite into
martensite [29]. InContents lists available at
ScienceDirectjournalhomepage:www.elsevier.com/locate/wearWearhttp://dx.doi.org/10.1016/j.wear.2015.02.0190043-1648/&
2015 Elsevier B.V. All rights reserved.nCorresponding author. Tel.:
91 712 2801519; fax: 91 712 2223230.E-mail addresses:
[email protected] (K. Singh),[email protected],
[email protected] (R.K. Khatirkar),[email protected] (S.G.
Sapate).Wear 328-329 (2015) 206216another study harmful effect of
retained austenite on wear resistancewasreported[30].
Theabrasiveweardataonconventionallyheattreated D2 steel is
relatively scarce. Previous studies did not report theeffect of
multipletemperingtreatment, phasetransformations
andretainedaustenite onabrasive wear of D2 steels. Inthe
presentinvestigation, the effect of multiple tempering treatment on
themicrostructure and two body abrasive wear behavior of D2 steel
hasbeenstudiedby
systematicandsimultaneousvariationoftestpara-meters.The detailed
structuralcharacterization of D2 steelhas beencarried out using
scanning electron microscopy (SEM), energy disper-sive spectrometer
(EDS) and X-ray diffraction (XRD). The mechanismofmaterial removal
hasbeensubstantiatedbyshortdurationabrasiveweartests. Theresultsof
thepresentworkwererationalizedwithrespect to hardness,
microstructure and operating wear mechanisms.2. Experimental
method2.1. Material and heat treatmentThe material used in the
present investigation (AISI D2 steel) wasobtained in form of a
forged bar of 100 mm diameter. The chemicalcompositionof AISI
D2steel was determinedbyoptical emissionspectrometer (OES) andis
giveninTable1. Theforgedbar wasannealed at 1158 K for 60 min to
remove the effect of prior thermo-mechanical processing history.
The specimens for further heat treat-ment were cut from the
annealed bar in the form of cylinder havinglength of 25 mm and
diameter of 10 mm. The specimens wererstaustenitizedat1303 Kfor45
minfollowedbyoilquenching(hard-ening).
Thetemperingtemperatureanddifferent temperingcycleswere selected to
obtain variations in microstructure; carbide morphol-ogy and
matrix, surface hardness and retained austenite content. Anattempt
is made to correlate abrasive wear of multiple tempered D2steel
with change in mechanism of material removal, associated withphase
transformations during multiple tempering treatments,underdifferent
abrasive wear test
conditions.Thehardenedspecimenswerethensubjectedtomultipletem-pering
treatments. The tempering temperatures selected were 523
K(designated as T523), 623 K (designated as T623), 723 K
(designatedTable 1Chemical composition (weight%) of D2
steel.Elements C Mn Si Cr Mo P S V FeWt% 1.5 0.45 0.3 12 0.9 0.03
0.03 1.0 BalanceFig. 1. Experimental set-up used for two-body
abrasion tests.Fig. 2. Secondary electron scanning electron
microscopy (SEM) images of the abrasive papers used for the wear
tests (a) 80 grit (320 m) (b) 150 grit (122 m) (c) 220 grit(92 m)
and (d) 320 grit (52 m).Table 2Summary of test parameters used for
abrasive wear of D2 steel in the present work.(Constant test
parameters: linear velocity 0.8 ms1andspecimenrotation50
rpm).Specimen designation Abrasive wear test conditionsLoad (N) SiC
particle size (m)AN 10,20,30 and 40 52AN 20 N 52, 92, 122 and 320OQ
10 5220 9230 12240 320T523 10 5220 9230 12240 320T723 10 5220 9230
12240 320T923 10 5220 9230 12240 320K. Singh et al. / Wear 328-329
(2015) 206216 207as T723), 823 K (designated as T823) and 923 K
(designated as
T923).TheannealedandasquenchedspecimensweredesignatedasANand OQ,
respectively. The specimens were single tempered, double,triple,
quadrupleandvestagestemperedattemperatures523 K,623 K, 723 K, 823 K
and 923 K respectively, with one tempering cycleof 60 min each. All
the heat treatments were carried out in an
inertatmospherefurnacetoavoidoxidationofthespecimens.
ThebulkhardnessofheattreatedspecimenswasmeasuredusingRockwellhardness
tester at 150 kgf load. An average of ve readings isreported in the
results.2.2. Microstructural characterizationThemicrostructural
examinationwas carriedout for all theheat treated specimens of D2
steel using SEM (JEOL 6380 A), EDS Fig. 3. Effect of tempering
temperature on the bulk hardness of D2 steel.Fig. 4. SEM
micrographs of heat treated specimens showing size, distribution
and morphology of carbides for different conditions (a) annealed at
1158 K (b) as-quenchedafter austenitizing at 1303 K (c) quenched
and tempered at 523 K (single stage) (d) quenched and tempered at
623 K (two stage) (e) quenched and tempered at 723 K (threestage)
(f) quenched and tempered at 823 K (four stage) (g) quenched and
tempered at 923 K (ve stage). All the samples were etched with 2%
Nital to reveal the matrix(dark) followed by etching with Vilella's
reagent to reveal the carbides (bright).K. Singh et al. / Wear
328-329 (2015) 206216 208(Bruker XFlash) attached to SEM and XRD to
study size, distribu-tion and morphology of carbides, to obtain
chemical compositionofthe carbides and to determine retained
austenite content. Thespecimensformicrostructural
observationswerepreparedusingstandard metallographic polishing
techniques followed by etchingwith 2% Nital and then with
Villella's reagent for 10 s each
[3132].XRDpatternsweremeasuredusingPANalytical XpertProMPDsystem
with Cu-K radiations and diffracted beam monochroma-tor. The XRD
patterns were then analyzed using XPert
Highscoresoftwaretoobtaintypeof
carbidesandretainedaustenite(RA)[3334]. Forthemeasurementof RA,
{220}and{200}peaksofaustenite and {200} peak of ferrite were
considered.2.3. Abrasive wear testsTwo body abrasive wear tests
were carried out on pin-on-plateabrasive wear test apparatus [35]
(DUCOM make, India) as shown inFig. 1. The commercially bonded
silicon carbides paper withdifferent particlesizes was usedas
abrasivemedia. Fig. 2(ad)shows secondary electron SEM micrographs
of silicon carbide (SiC)abrasive paper with average particle size
of 320 m, 122 m, 92 mand52 musedinthepresentstudy. Theshapeof
SiCabrasiveparticles was predominantly angular, although ner
particles exhib-ited relatively greater angularity. The specimens
for abrasive weartestswereintheformofpin(cylindrical
shape)withthelength25 mmanddiameterof10 mm.
Thespecimensweremetallogra-phically polished prior to each wear
test. The objective of simulta-neousvariation of abrasiveparticle
size andload was to examineregime of wear i.e. mild, moderate or
severe as reected in abrasivewear loss. The data on the effect of
simultaneous variation of testparametersonabrasivewearof dieandtool
steel isscarce. Theobjective was also to assess the change in
mechanism of materialremoval from the matrix and the carbides
associated with change intemperingtemperatureandtemperingcycles
(singlestagetovestage) with respect to severity of abrasive wear
test conditions. Twobody abrasive wear tests were conducted as per
the experimentalconditions mentioned in Table 2. The mass loss was
well distributedover the surface due to rotation of specimen, which
also avoids thepossibility of intense wear of the edges. The
lateral displacement
ofspecimenensuresthatthespecimenalwaystraversesnewtrack.The
specimens were cleanedwithethanol and weighed usingdigital
micro-balance (to an accuracy of 0.1 mg) prior to and aftereachwear
test. Thedifferenceintheinitial andnal weight ofthe sample was
recordedto calculate abrasive wear loss. TwoFig. 5. EDS spectrum of
as-quenched specimen of D2 steel.K. Singh et al. / Wear 328-329
(2015) 206216 209specimens were tested under similar abrasive wear
test conditionsand accordingly the scatter is shown in the test
result. The worn outsurfaces and wear debris particles were
examined using SEM andXRD. Theshort durationwear tests
werealsocarriedout withabrasiveparticlesizeof 122 mandat 10 Nand40
Nloadsforquenchedandtemperedat723
Ksampletogaininsightintothemechanism(s) of wear.3. Results and
discussion3.1. Hardness and microstructuresFig. 3showstheeffect of
temperingtemperatureonthebulkhardness of D2steel samples.
Theannealedsampleexhibitedthelowesthardness(19 HRc),
whereasas-quenchedsampleexhibitedthe highest hardness (65 HRc). It
can be observed from Fig. 3 that
anincreaseintemperingtemperaturefrom523 Kto723 Kresultedinmarginal
dropof hardness(3 HRc), whereasasharpdecreaseinhardness was
observed with the increase in tempering temperature to923 K (35
HRc). This sharp decrease in hardness was attributed
todecomposition of martensite into ferrite and carbides [36]
withincreaseintemperingtemperature. ThequenchedD2steel
hasthehighest hardness due to the formation of martensite, which
decreasedwithincreaseintemperingtemperatureandtime.
Similartrendofdecrease in hardness with increase in tempering
temperature has beenreported by Lee and Su. [37] and Leskovsek et
al. [38] for AISI 4340steel and H11 tool steel respectively as well
as other researchers [39].Fig.
4showsthesecondaryelectronSEMmicrographsof annealed,quenched and
quenched-tempered D2 steel samples. The higherFig. 6.
HighresolutionindexedX-raydiffractionpatternsof(a)as-quenchedsample(b)quenchedand523
Ktemperedsample(singlestage)(c)quenchedand723 Ktempered sample
(three stage) and (d) quenched and 923 K tempered sample (ve
stage).Table 4Retained austenite for as-quenched and quench and
tempered specimens ofD2 steel.Condition Retained
austenite(%)As-quenched at 1303 K 17.20Quenched and tempered at 523
K (single tempered) 17.00Quenched and tempered at 723 K (three
stagetempered)11.82Quenched and tempered at 923 K (ve stage
tempered) 6.39Table 3Chemical composition of primary and secondary
carbides (wt%) as determined byenergy dispersive spectrometer
(EDS).As-quenchedWeight %Fe Cr Si VPrimary carbides 88.25 11.06
0.69 o0.05Secondary carbides 51.19 44.67 0.22 o0.05Quenched and
tempered at 523 KPrimary carbides 90.54 8.82 0.49 o0.05Coarse
secondarycarbides93.88 6.21 0.37 o0.05Fine secondary carbides 51.83
48.03 0.07 o0.05Quenched and tempered at 923 KPrimary carbides
92.58 8.03 0.57 o0.05Coarse secondarycarbides51.05 46.84 0.29
0.11Fine secondary carbides 91.39 8.15 0.46 o0.05K. Singh et al. /
Wear 328-329 (2015) 206216 210amount of alloying elements in D2
steel caused a bi-modal distributionof
carbideparticlesintheannealedcondition(Fig. 4a). Thecoarseparticles
were observedto be M7C3carbides whichformduringsolidication[4].
Theprecipitationof carbides inannealedsteel isdependent on the
ratio of chromium to carbon. When the ratio is lessthan three, the
only carbide which forms is the alloyed cementite. If theratio is
greater than three, chromium carbides (both Cr7C3 and Cr23C6)are
formed. InD2 steels only Cr7C3has beenreported to form(observed in
the present work also) which is usually enriched by Fe[4]. Fig. 4b
shows SEM micrograph of as-quenched specimen showingun-dissolved
carbides as well as small secondary carbides. Thesecondary carbides
in as-quenched specimen might have formed duetotheverysmall
delayinquenchingthesamples. Inas-quenchedspecimen, the martensitic
lathes are veryne and cannot be resolved.Fig.4cg shows
SEMmicrographs of hardened andtempered speci-mens. It can be
observed that with increase in tempering temperatureamount ofne
secondary carbides increased along with coarsening ofcarbides.
ThecarbidesasseenfromFig. 4gappeartoberelativelycoarser at the
maximum tempering temperature of 923 K.Thenatureof
theprimaryandsecondarycarbidesprecipitatedduring tempering was
determined by using EDS. Secondary electronSEM image along with
representative EDS spectrum of primary andsecondary carbideisshown
inFig.5.
ThechemicalcompositionofthecarbidesasdeterminedbyEDSissummarizedinTable3.
Theprimary carbides (bigger Fig. 4b) were rich in Fe and lean in Cr
inthe as-quenched condition. The secondary carbides (smaller Fig.
4b) which might have precipitated due to slight delay inquenching
were having 50% Fe and 35% Cr. After tempering at523 K, the
composition of the primary and secondary carbideschanged only
marginally. Although,there was precipitation of
newsecondarycarbides (other thanthat
precipitatedduringdelayinquenching referred to asne secondary
carbides). After temperingat 923 K, the primary and coarse
secondary carbides were rich in Feand lean in Cr, the coarse
secondary carbides had almost 50% Fe and50% Cr, whereas the ne
secondary carbides which precipitated werevery rich in Fe. In
chromium containing tool steels, two types of Crcarbides (Cr7C3 and
Cr23C6) are often observed and these are
usuallyenrichedwithalloyingelementslikeFe,
Moetcdependingonthechemical composition of the steel. The
precipitation sequence
withincreasingtemperingtemperatureintoolsteelshasbeenreported[36]
to be M3C-M7C3-M23C6 where M stands for metal atoms.
Thetransformation of M3C (Fe3C) to M7C3 (Cr7C3) occurs by the
nuclea-tion at the Fe3C (M3C)/ferrite interface [4]. The Cr carbide
precipita-tion seems to occur at lower and intermediate
temperingtemperatures. At higher temperingtemperatures,
precipitationofironrichcarbidesindicatesthedecompositionof
martensiteintoferrite and Fe3C, which is slightly enriched with Cr.
A similarsequencehasbeenreportedfor astudyonEn31steel [40]. TheXRD
patterns of the as-quenched, quenched-tempered 523
K,quenched-tempered 723 K andquenched-tempered 923 K D2
steelsamples are showninFig. 6. The XRDdata supplements
SEMobservationsonmicrostructural
changesassociatedwithmultipletempering treatment of D2 steel. The
as-quenched D2 sampleshowedthepresenceof
martensite(indexedasferrite, sincetheresolutionof the diffractomter
was not sufcient toresolve thetetragonality of the martensite),
alloyed cementite (indexed as Fe3C),Fig. 7. (a) Abrasive wear
volume loss (cm3) as a function of load for annealed specimen for
abrasive particle size of 52 m. (b) Abrasive wear volume loss (cm3)
as a functionof abrasive particle size for annealed samples at a
load of 20 N and at a velocity of 0.8 cm/s and at 50 rpm.Fig. 8.
Abrasivewearvolumeloss(cm3)vs. simultaneousvariationof
loadandabrasive particle size at a velocity of 0.8 m/s and rpm of
50.Fig. 9. Variation of abrasive wear volume loss (cm3) as a
function of hardness (HRc)at a load of 10 N and 40 N.K. Singh et
al. / Wear 328-329 (2015) 206216 211RA and M7C3 (indexed as Cr7C3).
After tempering at 523 K and 723
K,theprimarychromiumcarbides(Cr7C3)persistedalongwithalloycementiteandRA,
althoughtheir
proportionvaried(asindicatedrelativelybythepeakintensities).
Atthehighesttemperingtem-peratureof 923 K,
theXRDpatternshowedveryweakvanadiumcarbide (V8C7) peak. The
variation of RA with tempering temperatureis given in Table 4.
There was almost no change in RA after temperingat 523 K which
gradually decreased with increase in the temperingtemperature.
Evenafter vestagesof temperingat923 K, theRAcontent was not zero.
The diffraction peaks (martensite, indexed asferrite) are also much
broader in the as-quenched D2 steel sample.The martensitic
structure is always associated with large
dislocationdensityinthestructure. Thebroadeningof
X-raypeakprolesisrelated to size of the crystallites and/or
micro-strain present in thesample. Reduction in the size of the
crystallites/coherently diffractingdomains and increase in the
micro-strain lead to the broadening ofthe peak proles. The presence
of dislocations inthe structureinduces micro-strains, which results
in the broadening of thediffraction peaks [40,41]. After tempering
at 523 K, 723 K and923 K,
thebroadeningreducedduetoreductioninthedislocationdensity as well
as decomposition of martensite into ferrite.3.2. Two body abrasive
wear testingFig. 7a shows the variation of abrasive wear volume
loss withload for annealed D2 steel. The abrasive wear volume
lossFig. 10. (a,b): SEM micrographs of worn out surface of quenched
and tempered specimens of D2 steel at a load of 40 N. (a) tempered
at 723 K (triple tempering) and (b)vestage tempered at 923 K.Fig.
11. (a,b): SEM micrographs of heat treated D2 steel after short
duration abrasive wear test for quenched and tempered sample at 723
K (a) 10 N (b) 40 N. (c,d ): Sub-surface SEM micrographs of worn
out samples at 40 N load (a) as-quenched, (b) quenched and tempered
at 523 K (single stage).K. Singh et al. / Wear 328-329 (2015)
206216 212increased linearly and the volume loss at highest load
(40 N) wasapproximatelytwotimes as comparedtothat observedat
thelowest load (10 N).The abrasive wear volume loss increased
withtheincreaseintheparticlesizeasshowninFig. 7b. Withtheincrease
in particle size from 52 m to 122m, the increase inthe volume loss
was approximately 1.6 times, whereas for
increaseinparticlesizefrom122 mto320 m, theincreasewasonlymarginal.
These results can be understood in terms of particle sizeeffect,
whichcanbe attributedtoclogging, adhesionof weardebris tothetipof
abrasiveparticlethus reducingits cuttingefciency, fracture of
abrasive particles and increase in local owstress withdecreasing
scale of deformationfor ner
abrasiveparticlesleadingtoreducedwearloss[6,42,43].
Intheannealedcondition, the matrix is verysoft andthe increase
inloadisexpectedtoresult inhigher depthof cut
resultinginmaterialremoval proportional totheappliednormal load.
SiCparticleshave higher hardness as compared to the Cr7C3 carbides
and henceareexpectedtocauseindentationandfractureof
thecarbidessince ratioof Ha/Hs( ratioof hardness of abrasive tothat
ofcarbide) was greater than 1.2 [6]. Fig. 8 shows the effect
ofsimultaneous variationof loadandparticlesizeof
abrasiveonabrasivewearvolumeof heattreatedD2steel.
Thedatapointsweretted by bestt line and regression coefcients were
in therangeof0.930.95.
Itcanbeobservedthatwiththeincreaseinseverityofwearconditionfrom(10
N, 52 mm)to(40 N, 320 mm)abrasivewearvolumelossincreasedforD2steel
irrespectiveofthe heat treatment condition. The abrasive wear
volume lossincreased more than two times for specimens tempered at
923 Kwhereasasquenchedspecimenexhibitedanincreaseof nearly1.25
times with increase in severity of test condition ( from 10 N,52
mmto40 N, 320 mm). Itcanalsobeobservedthatspecimentempered at 923 K
exhibited nearly 2.4 times more abrasive wearvolume loss at a
loadof 40 Nandparticle size of 320 mmascompared to oil quenched
specimen whereas this increase was notmore than 1.6 times at a load
of 10 N and particles size of 52 mm.As-quenched specimen exhibited
relatively higher volumefractionof retainedaustenite,
asshowninTable4, whichwasnearly three times as compared to specimen
tempered at 923 K asaresult ofmultipletemperingtreatment, whereas
nosignicantdifferencewas observedfor specimentemperedat 523 K.
Therelatively lower abrasive wear loss exhibited by
as-quenchedspecimen can be attributed to ner morphology of
carbides,martensiticmicrostructureandstressinducedtransformationofaustenite
to martensite during abrasive wear. At higher temperingtemperature
the contribution of retained austenite in reduction ofabrasive wear
loss associated with stress induced transformationwas
relativelyinsignicant duetoits lower percentage. Fig. 9shows
abrasive wear volume loss plotted vs bulk hardness of D2steel at 10
N and 40 N loads. The data points weretted by powerlawrelationship
and the dependence of volume loss (V) onhardness (H) was expressed
by the power law equation:V K H m1Fig. 12. SEM micrographs of wear
debris of (a) and (b) as-quenched samples, (c) and (d) quenched and
523 K tempered samples (single stage), (e) and (f) quenched and923
K tempered samples (ve stage). For (a), (c) and (e) the load used
was 10 N, while for (b), (d) and (f), the load used was 40 N.K.
Singh et al. / Wear 328-329 (2015) 206216 213where Vvolume loss,
Hbulk hardness, Kconstant and mhardnessexponent.
Itwasobservedthatonlyamarginal
differ-encewasobservedonthedependenceofabrasivewearvolumeloss on
hardness as revealed by the hardness exponents (0.69 and0.66at
aloadof 10 Nand40 N, respectively). Asexpected,
thewearvolumelosswasgreaterataloadof40 Nascomparedto10 N. The
increase in volume loss can be attributed to thestructural
transformations that occurredduringheat treatmentand due to the
greater depth of cut by coarseparticles at higherloads. These
observations are consistent with previous studies byMoore
[44]andUedaetal. [45]. Italsosuggeststhatthefactorsother than
hardness, e.g. microstructure and morphology ofcarbides are
expected to exhibit considerable inuence on abrasivewear volume
loss of D2 steel. Hence SEM and XRD studies of wornout surfaces and
wear debris particles were focused to substanti-ate the results of
the wear tests.Fig. 10 (a and b) shows abraded surfaces for
quenched andtempered samples at 723 K and 923 K respectively. As
the temperingtemperatureincreased,
theabrasionresistancereducedduetothecoarsening of carbides coupled
with decomposition of martensite [36].With increase in tempering
temperature the softer matrix was easilyplowedandcut bythe abrasive
particles as
comparedtospecimenstemperedatlowertemperatures[6,40,41,46].
Thematerialfromthematrixwasremovedbytheprocessof microcutting,
plowinganddelamination as shown in Fig.10 (a and b), whereas
indentation andgross fracture and carbide pull out was observed at
higher load andcoarser particle size (40 N load and 320 m particle
size). Withnerabrasive particles, the contact zone size was
relatively small and
onlysmallscalechippingofcarbideswasobservedresultinginrelativelylower
abrasive wear volume loss under these conditions. Fig. 11 showsSEM
micrographs of quenched and tempered (723 K) D2 steel sampleafter
short duration abrasion test at a load of 10 Nand 40 Nrespectively.
Fig. 11(a) showsmaterial removal fromthematrixbyplowing and some of
thener carbides carried with plowed
materialalsocanbeseenatthecenterof theimage. Fig.
11(b)showsSEMphotograph at a load of 40 N. The important mechanisms
of materialremoval from the carbides were observed to be large
scale chipping,edge fracture and gross fracture of carbides shown
in the SEMphotographs.
Asthematrixiswornoutbytheprocessofplowing/micro cutting, it offers
reduced support to the carbide particles leadingto carbide pull out
resulting in increased abrasive wear volume loss
athigherloadandcoarserparticlesize. [47].ThisisalsoevidentfromFig.
11(candd)whichshowssub-surfacemorphology ofwornoutspecimens
as-quenched and quenched-523 K tempered specimens.Fig. 12 shows the
SEM micrographs of wear debris of as-quenched,quenched and tempered
523 K sample and quenched and tempered923 K sample at load of 10 N
and 40 N. Fig. 12(a, c and e) shows weardebris at load of 10 N and
SiC particle size of 52 mwhereas Fig. 12(b,dandf) showsweardebrisat
loadof 40 Nwithparticlesizeof320 m. Fig. 13 shows the XRDpatterns
of wear debris of as-quenched, quenched-523 K, quenched-723
Kandquenched-923 Ktempered specimens at 40 N load and 320 m SiC
particle size. Theweardebrisparticlesconsistedof
fracturedSiCabrasiveparticles,carbideparticlesremovedbytheprocessof
fractureandcarbidesremoved by pull out (Mo2C, Cr7C3 and Fe3C) along
with ferrite
andmartensitematrixremovedbytheprocessofcuttingasshowninFig.
12.These observations were conrmed by XRD patterns of weardebris
particles at a load of 40 N and 320 m SiC particle size (Fig.
13(ad)). It can also be observed from Fig. 13 that at a load of 10
N thefragmented and fractured SiC carbides were relatively ner
ascompared to that at higher load of 40 N. XRD patterns in Fig. 13
alsoindicated presence of iron oxide (Fe2O3) peaks with
signicantFig. 13. High resolution X-ray diffraction (XRD) pattern
of wear debris of D2 steel for (a) as-quenched (b) quenched and 523
K tempered (single stage) (c) quenched and723 K tempered (three
stage) and (d) quenched and 923 K tempered (ve stage) samples at a
load of 40 N.K. Singh et al. / Wear 328-329 (2015) 206216
214intensityfor alltheheattreatedconditionswhichimpliedthatthewear
mechanismwas oxidative. The ironoxide lmprovides alubricating
lmwith lowshear strength reducing the friction
betweensurfaceandabrasiveparticlessubsequentlysubsidingtheeffectsofcapping
and clogging [6]. At higher load and coarser particle size,
ironoxidelm is penetrated by abrasive particles with exposure to
freshmetalsurface withincreasedwear loss. There
wasnoformationofwhite layer for very high loads for D2 steel as
reported in a previousstudy[25]. Theaboveobservations
alsoindicatethat D2steel isrelatively more abrasive wear resistant
under mildto moderateabrasive wear conditions. From the above it
can be concluded that acombination of wear mechanisms was
responsible for
materialremovalandtherelativecontributionofeach(e.g. Plowing,
micro-cutting, edge fracture, chipping, gross fracture of carbides,
carbide pullout) was decided by structural transformations
associated withtemperingtreatmentandmorphology ofcarbides.
Thiscanexplaintheabrasivewearlosstrendsobtainedinthepresentinvestigationwith
increase in tempering temperature and with change in abrasivewear
test conditions from benign to relatively severe one. From
theresults of the present investigation, it can be also pointed out
that it isdifcult to pin point the effect of retained austenite
(617%) in heattreated D2 steel on the improvement of abrasive wear
resistance.
Thisisfurthersupportedbytheobservationthatabrasivewearvolumeloss
ofheattreatedD2steelshowednearly similar dependence onsurface
hardness at both lower and higher loads as indicated by thehardness
exponents. In fact, the contribution of change in mechanismof
material removal toabrasivewear resistanceof D2steel withincreasing
severity of abrasive wear test conditions was moresignicant as
comparedtothe effect of retainedaustenite. It
isdifculttocomparetheresultsof thepresentworkwiththoseofprevious
studies due to different processing conditions, mode of
wear,conguration of the test apparatus and the test conditions, as
far asthe effect of retained austenite on abrasive wear resistance
isconcerned [28,47].4. Conclusions1. The morphology of carbides
became coarser with increasing temper-ing temperature during
multiple tempering treatment of D2 steel.2. D2 steel in the
as-quenched condition exhibited highest abra-sive wear resistance
which was attributed to martensiticmicrostructure withner
morphology of carbides.3. Thesimultaneous variationof
loadandabrasiveparticlesizeincreased abrasive wear volume loss with
increasing temperature.4. The dependence of abrasive wear
resistance of heat treated D2steel on hardness was almost similar
irrespective of the load.5. The relatively higher abrasives wear
rate exhibited by multipletempered D2 steel at higher tempering
temperature was due tosofter matrix and coarser morphology of
carbides.6. The abrasive wear resistance of tempered D2 steels
waspredominantly inuenced by morphology of the carbides,nature of
the matrix and the change in mechanism of materialremoval
frombenigntorelativelyseveretestconditions. Theretained austenite
had relatively minor effect on abrasive wearbehavior of heat
treated D2 steel.AcknowledgmentThe authors are grateful to
Director, VNIT Nagpur for providingnecessary facilities in carrying
out this investigation. The authorsarealsothankfultoMr. S. L.
GadgeandMr. AvishkarRathodfortheir help during heat treatment and
SEM work.References[1] W.H. Wills, Practical observations on
high-carbon high chromium tool steels,Trans. ASM 23 (1935)
469475.[2] W.H. Wills,
Furtherstudyonhigh-carbonhighchromiumtoolsteels, Trans.ASM 25
(1937) 10131020.[3] J.P. Gill, High-carbon high chromium steels,
Trans. ASST 15 (1929) 387400.[4] G.A. Roberts, R.A. Cary, Tool
Steels, Fourth ed., ASM, Ohio, 1980.[5] MarcMasen, Abrasivetool
wearinmetal formingprocesses(Ph.D. thesis),University of Twente,
The Netherlands, 2004.[6] I.M. Hutchings,
TribologyFrictionandWearof EngineeringMaterials, CRCPress, Arnold,
London, 1992.[7] S.G. Sapate, A.D. Chopde, P.M. Nimbalkar, D.M.
Chandrakar,
Effectofmicro-structureonslurryabrasionresponseofEn-31steel, Mater.
Des. 29(2008)613621.[8] K.H. Zum Gahr, Abrasive wear of two phase
metallic materials with a coarsemicostructure, in: K.C. Ludema
(Ed), in: Proceedings of the InternationalConference on Wear of
Materials, Vancouver, 1985, ASME, pp. 4557.[9] S. Tekeli, A. Gural,
Dryslidingbehaviorof heat
treatedironbasedpowdermetallurgysteelswith0.3Graphite2%Niadditions,
Mater. Des. 28(2007)19231927.[10] M.X. Wei, S.Q. Wang, L. Wang,
X.H. Cui, K.M. Chen, Effect of temperingconditions onwear
resistanceinvarious wear mechanisms of H13steel,Tribol. Int. 44
(2011) 898905.[11] S.G. Sapate, A. Selokar, N. Garg, Experimental
investigation of hardfacedmartensitic steel under slurry
abrasionconditions, Mater. Des. 31(2010)40014006.[12] C.J. Zhang He
Lin, Aninvestigationof the role of secondary carbides inmartensitic
steel during three body abrasion wear, Wear 16 (1994) 103109.[13]
O.P Modi, P. Pandit, D.P. Mondal, B.K. Prasad, A.H. Yegneswaran,A
Chrysanthou, High-stress abrasive wear response of 0.2% carbon dual
phasesteel: effects of microstructuralfeatures and experimental
conditions,MaterSci. Eng. 458A (2007) 303311.[14] H. Saghaan, S.
Kheirandish, Correlatingmicrostructuralfeatureswithwearresistance
of dual phase steel, Mater. Lett. 61 (2007) 30593063.[15] E.
Badisch, C. Mitterer, Abrasive wear of high speed steels inuence of
abrasiveparticles andprimarycarbides onwear resistance, Tribol.
Int. 36(2003)765770.[16] F. Katsuki, K. Watari, H. Tahira, M.
Umino, Abrasive wear behavior of a pearlitic(0.4% C) steel
microalloyed with vanadium, Wear 264 (2008) 331336.[17] I. Sevim,
I.B. Eryurek, Effect of
abrasiveparticlesizeonwearresistanceinsteels, Mater. Des. 27 (2006)
173181.[18] A.K. Jha, B.K.Prasad, O.P. Modi, S.Das, A.H.
Yegneswaran, Correlating micro-structural featuresandmechanical
propertieswithabrasionresistanceof ahigh strength low alloy steel,
Wear 254 (2003) 120128.[19] P.K. Reponen, J. Liimatainen, S.
Ala-Kleme, J. Hellman, S.P. Hannula, Thecorrelationof material
characteristicsandwearinalaboratoryscaleconecrusher, Wear 267
(2009) 568575.[20] H. Torkamani, Sh. Raygan, J. Rassizadehghani,
Comparing microstructure andmechanical properties of AISI D2 steel
after bright hardening and oil quench-ing, Mater. Des. 54 (2014)
10491055.[21] Linhu Tang, Chengxiu Gao, Jianlong Huang, Hongyan
Zhang, Wenchun Chang,Dryslidingfrictionandwearbehaviorof
hardenedAISI D2tool steel withdifferent hardness levels, Tribol.
Int. 66 (2013) 165173.[22] L. Bourithis,G.D. Papadimitriou,
J.Sideris,Comparisonof wear properties oftool steelsAISI
D2andO1withthesamehardness, Tribol. Int. 39(2006)479489.[23] X. Ma,
R. Liu, D.Y. Li, Abrasive wear behavior of D2 tool steel with
respect toload and sliding speed under dry sandrrubber wheel
abrasion condition, Wear241 (2000) 7985.[24] Debdulal Das, Apurba
Kishore Dutta, Kalyan Kumar Ray, Sub-zero treatmentsof AISI D2
steel Part I: Microstructure and hardness, Mater.Sci.Engg.A
527(2010) 21822193.[25] Debdulal Das, Apurba Kishore Dutta, Kalyan
Kumar Ray, Sub-zero treatmentsof AISI D2 steel Part II. Wear
behavior, Mater. Sci. Eng. A 527 (2010)21942206.[26] D. Das, K.K.
Ray, A.K. Dutta, Inuence of temperature of sub-zero treatmentson
the wear behavior of die steel, Wear 267 (2009) 13611370.[27]
Kamran Amini, Amin Akhbarizadeh, Sirus Javadpour, Effect of deep
cryogenictreatment on the formation of nano-sized carbides and the
wear behavior ofD2 tool steel, Int. J. Miner. Metall. Matls 19
(2012) 795.[28] L. Xu, J. Xing, S. Wei, Y. Zhang, R. Long, Study on
relative wear resistance andwearstabilityof high-speedsteel
withhighvanadiumcontent, Wear 262(2007) 253261.[29] R. Colaco, R.
Villar, On the inuence of retained austenite in the abrasive
wearbehavior of a laser surface melted tool steel, Wear 258 (2005)
225231.[30] X.H. Fan, L. He, Q.D. Zhou, Study of high Cr cast iron
on abrasion resistance andimpact fatigue resistance. in:
Proceedings of the International Conference onWear of Materials,
Denver, CO, ASME, NY, 1989, pp. 913.[31] KatharineB.Small,David
A.Englehart, Todd A.Christman,Etchingspecialtyalloys, Adv. Mater.
Process. 166 (2008) 3237.[32] ASM Handbook, Metallography and
microstructures, ASM, Ohio, 2004.[33] B.D. Cuillity, Elements of
X-ray Diffraction, Addison-Wesley Publishing Com-pany, USA, 1978.K.
Singh et al. / Wear 328-329 (2015) 206216 215[34] C.Y. Wang, J.
Shi, W.Q. Cao, H. Dong, Characterization of microstructureobtained
by quenching and partitioning process in low alloy martensitic
steel,Mater. Sci. Eng. A 527 (2010) 34423449.[35] ASTM G132-01,
Standard test method for pin abrasion testing, 2001.[36] H.K.D.H.
Bhadeshia, R.W.K. Honeycombe, Steels-MicrostructureandProper-ties,
Butterworth-Heinemann, Oxford, 2006.[37] W.S. Lee, T.T. Su,
Mechanical properties and microstructural of AISI4340 highstrength
alloy steel underquenchedandtemperedcondition, J. Mater.
Proc.Technol. 87 (1999) 198206.[38] V. Leskovek, B. Sustarsic, G.
Jutrisa, The inuence of austenitizing
andtemperingtemperatureonhardnessandfracturetoughnessof
hotworkedH11 tool steel, J. Mater. Proc. Technol. 178 (2006)
328334.[39] G.R. Speich, W.C. Leslie, Tempering of Steel, Metall.
Trans. A 3 (1972)10431054.[40] AbinBaburaj, K.B. Shishupal
SinghChaudhary, RajeshK. Khatirkar, SanjayG. Sapate, Abrasive wear
behavior of heat treated En31 steel, ISIJ Int. 53
(2013)14711478.[41] R.K. Khatirkar, P. Yadav, S.G. Sapate,
Structuralandwearcharacterizationofheat treated En24 steel, ISIJ
Int. 52 (2012) 13701376.[42] E. Rabinowicz, Friction and wear of
materials, 2nd Edition, Wiley, 1995.[43] A. Misra,I. Finnie,
Onthesizeeffectinabrasiveanderosive wear, Wear65(1981) 359373.[44]
M.A. Moore, The relationship between abrasive wear resistance,
hardness andmicrostructure of ferritic steels, Wear 28 (1974)
5968.[45] K. Ueda, N. Ishikawa, Y. Murota, Inuence of matrix
structure and hard carbideon abrasive resistance of steel plates,
Mater. Sci. Forum 706709 (2012) 2342.[46] Y. Sahin, M. Erdogan, M.
Cerah, Effect of
martensitevolumefractionandtemperingtimeonabrasivewearof
ferriticductileironwithdual matrix,Wear 265 (2008) 196202.[47] Y.
Sano, T. Hattori, M. Haga,
Characteristicsofhigh-carbonhigh-speedsteelrolls for hot strip
mill, ISIJ Int. 32 (1992) 11941201.K. Singh et al. / Wear 328-329
(2015) 206216 216