See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/270190029

Abrasive Wear Behaviour of Heat Treated En31 Steel

Article  in  ISIJ International · January 2013

DOI: 10.2355/isijinternational.53.1471

CITATIONS READS

7 260

4 authors, including:

Abin Baburaj Rajesh K. Khatirkar

Rashtriya Chemicals and Fertilizers Limited (RCF) Visvesvaraya National Institute of Technology
1 PUBLICATION   7 CITATIONS    51 PUBLICATIONS   347 CITATIONS   

SEE PROFILE SEE PROFILE

Sanjay Sapate
Visvesvaraya National Institute of Technology
47 PUBLICATIONS   367 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Welding of Duplex stainless steel View project

Iron & Steel View project

All content following this page was uploaded by Rajesh K. Khatirkar on 16 August 2017.

The user has requested enhancement of the downloaded file.

 ISIJ International, Vol. 53 (2013), No. 8, pp. 1471–1478

Abrasive Wear Behaviour of Heat Treated En31 Steel

Abin BABURAJ, K. B. Shishupal Singh CHAUDHARY, Rajesh K. KHATIRKAR* and Sanjay G. SAPATE

Department of Metallurgical and Materials Engineering, Visvesvaraya National Institute of Technology (VNIT), South Ambazari
Road, Nagpur – 440010, Maharashtra, India.
(Received on January 13, 2013; accepted on April 18, 2013)

Ni–Cr–Mo steels are widely used in machine part members, gears and shafts. Steels with higher carbon
content (~1%) are used for heavy machine parts and bearings. Abrasive wear resistance is often a very
important requirement for these high carbon steels, apart from sliding wear properties. In the present
study, En31 steel was subjected to varying heat treatments to generate different microstructures. An
attempt has been made to correlate the two body abrasive wear resistance with the bulk hardness and
microstructures. The microstructures were studied through a combination of scanning electron micros-
copy (SEM), energy dispersive spectrometer attached to SEM (i.e. SEM-EDS) and X-ray diffraction (XRD).
The bulk hardness decreased with increase in tempering temperature from 423 K to 848 K. The precipita-
tion of Cr7C3 after 598 K tempering did not cause an appreciable increase in the hardness. At higher tem-
pering temperatures (848 K), the martensite decomposed to give ferrite and cementite. The abrasive wear
tests were carried out on hardened and tempered specimens. The abrasive wear mass loss increased
with increase in the tempering temperature. Hardness had a direct correlation with the two body abrasive
wear behaviour in En31 steel – increase in hardness increased the abrasive wear resistance. The impor-
tant material removal mechanism were micro cutting and micro ploughing, the relative contribution of
each to total wear loss was influenced by abrasive wear test conditions.

KEY WORDS: Ni–Cr–Mo steel; En31 steel; abrasive wear; hardening; tempering; carbides; martensite; X-
ray diffraction; wear debris; alumina.

chemically assisted wear.8–13) Wear not only causes loss of

1. Introduction
material, but also includes costs involved in the replacement
Ni–Cr–Mo steels are widely used in applications like of engineering components and downtime of the equipment.
machine part member, bearings, shafts and gears. Steels This makes wear resistance a very important material selec-
with carbon content in the range 0.8–1.1 wt% and total tion parameter. Abrasive wear alone accounts for more than
substitutional solute content less than 3 wt% are particu- 50% of the total failures of the engineering components in
larly suitable for machine tools and bearings.1–3) These industries.7) The abrasive wear is generally classified as two
steels can be made martensitic by quenching in oil or water body abrasion (e.g. Pneumatic conveyors, grinding of ore
from the austenizing temperature.4) The martensite is then and minerals in ball/rod mill, agricultural implements) and
subjected to tempering treatment to obtain suitable combi- three body abrasion (e.g. jaw crushers).14) Abrasive wear
nation of hardness and ductility.4,5) The martensitic micro- resistance of a material depends on many factors like hard-
structure is rarely used in un-tempered condition due to ness of the material, size, shape, amount and distribution of
presence of large internal stresses associated with the mar- the second phase particles and properties of abrasive parti-
tensitic phase transformation and very low ductility.4,5) cles along with the service conditions.10,12,14–17) It has been
Low-temperature tempering is sufficient to reduce these reported18) that the abrasion resistance in pearlite-carbide
internal stresses considerably without essentially changing and ferrite-carbide microstructure depends on the volume
the basic features of the martensitic structure.4) On the oth- fraction of the carbides. The abrasion resistance in pearlite-
er hand, high temperature tempering causes a change in carbide microstructures increases with the increase in car-
internal structure and enhances the ductility.4,6) In many bide volume fraction up to 35%, while in ferrite-carbide
applications, adequate wear resistance is also required in microstructures, the carbide volume fraction does not seem to
addition to the strength and ductility of the steels. It is one have a significant effect on the abrasion resistance. The abra-
of the major factors responsible for pre-mature failures in sion resistance is also influenced by morphology of carbides
engineering components in industries.7) The primary in austenitic/martensitic matrix.18) The wear resistance of
modes of wear responsible for failures of engineering com- steels with martensitic microstructure was influenced by
ponents are sliding wear, abrasive wear, erosive wear and carbon content and volume fraction of martensite.7,19,20) The
dispersion of fine carbides in martensitic microstructure has
* Corresponding author: E-mail: [email protected] been reported to be beneficial for increasing abrasive wear
DOI: http://dx.doi.org/10.2355/isijinternational.53.1471 resistance.21)

1471 © 2013 ISIJ

 ISIJ International, Vol. 53 (2013), No. 8

The abrasion resistance of steels did not show consistent the samples was measured using Rockwell hardness tester
correlation with the bulk hardness. There are contradictory using a load of 150 kg. An average of five readings is report-
opinions in the reported literature – some authors22–25) sug- ed in the results. X-ray diffraction peak profiles were mea-
gested that increase in hardness increases abrasive wear sured on a PANalytical X’Pert Pro MPD diffractometer
resistance, while few others reported26,27) a decrease in abra- using CuKα radiation. The measurements were obtained
sive wear resistance with the increase in hardness, in partic- using a step size of 0.02° in the 2θ range 34° to 90° and an
ular, where abrasive wear by brittle fractures predominates. integration (measurement) time of 30 s at each step. The dif-
Relatively less data is published regarding the abrasive wear fractometer was used in the Bragg-Brentano configuration
behaviour of En31 steel. This has been the motivation for in line focus mode with 0.02 rad soller slits on both incident
the present study. The present study is an effort in obtaining and diffracted side. The diffracted radiation was measured
physical understanding and co-relation, if any, between the with a xenon filled point detector fitted with a curved crystal
wear behaviour and different microstructures in En31 steel. monochromator to remove the fluorescent anf Kβ radiation.
The microstructural and hardness variations in En31 steel The X-ray data was post processed using X’Pert Highscore
were obtained by using hardening and tempering heat software.29)
treatments. The microstructures were characterized using Pin-on-disc wear testing apparatus (model: TR20-LE,
scanning electron microscopy (SEM), energy dispersive Wear and Friction Monitor, Ducom, Bangalore, India) was
spectrometer attached to SEM (SEM-EDS) and X-ray used for conducting two body abrasive wear tests. The sur-
Diffraction (XRD); while the two body abrasive wear face of the pin shaped samples (30 mm length and 6 mm
behaviour was studied using a pin-on-disk type machine diameter) was ground and metallographically polished using
using abrasive paper. emery paper (up to 4/0) prior to each test. The specimen was
firmly fixed in the specimen holder, while the abrasive
paper (alumina Al2O3– 150 grit) was fixed on the circular
2. Experimental Methods
steel disc, which rotated about its axis. All the tests were
In the present investigation, En31 steel in the as-rolled carried out for a period of 600 s (constant) using track radius
and annealed initial condition was used to study the two of 40 mm (constant), 450 rpm (constant) and a sliding veloc-
body abrasive wear behaviour. The chemical composition of ity of 1.885 m/s. The abrasive paper (alumina - 150 grit) was
the as- received (i.e. as-rolled and annealed) steel as deter- firmly secured to the disc ensuring that there was no relative
mined by wet chemical analysis is given in Table 1. The movement between the disc and paper. The abrasive paper
specimens used were in the form of a pin of 30 mm length was changed after every 300 s to minimize fragmentation
with a diameter of 6 mm. For hardening treatments, speci- and fracture of abrasive particles due to repeated pass of the
mens were initially austenized at 1 123 K (± 5 K) for 3 600 specimen over same track. The other effects associated due
s followed by oil quenching. For tempering treatments, the to repeated pass of specimen over same path are capping
oil quenched samples were heated at temperatures of 423, (adhesion of wear debris particles to the tip of abrasive par-
598, 723 and 848 K respectively for 2 400 s followed by air ticles) which reduces its angularity, shelling (removal of
cooling to room temperature (298 K). It needs to be noted abrasive particles, particularly at higher applied loads and
that tempering was done immediately after hardening treat- rounding of tip of abrasive particles) and clogging (accumu-
ment. All the samples were vacuum sealed to avoid oxida- lation of wear debris particles between the spaces of abra-
tion during hardening and tempering treatments. After- sive particles thus carrying a part of the applied load). All
wards, all the samples were subjected to detailed these factors tend to decrease material removal rate with
microstructural and XRD characterizations. The samples passage of time due to decrease in number of cutting points,
were ground and polished with successively fine emery decreased angularity and reduction in effective load carried
papers (1/0, 2/0, 3/0 and 4/0). After polishing on emery by each particle. The rotation of specimen minimizes sharp
papers, samples were polished on velvet cloth using alumina edges at periphery of specimen and avoids tearing of abra-
slurry and finally using diamond paste (1 μ m) using Hiffin sive paper.14) The samples were cleaned with ethyl alcohol
chloride as lubricant. The mirror polished samples were then and weighed (up to an accuracy of 0.1 mg using microbalance)
deep etched with 2% Nital28) to observe the microstructure. prior to and after each test. The wear loss was calculated from
Microstructures of all the specimens (as-received, hardened the weight loss measurement and expressed in terms of mass
and tempered) were examined using a JEOL 6380A W- loss. An average of two tests was used to calculate mass loss
SEM. The SEM micrographs (secondary electron and back- after abrasive wear. The effect of load on mass loss (g) of
scatter electron) were obtained using an accelerating voltage En31 steel was studied for all the heat treated conditions at a
of 25 kV. For few selected tempered samples, high resolu- constant velocity (1.885 m/s) for two different loads (10 N &
tion i.e. field emission gun SEM (make FEI Quanta 3D) was 20 N). The abraded surfaces and wear debris particles were
used to obtain the secondary electron images and EDS further characterised by using SEM and XRD to study micro-
(make EDAX) attached to SEM was used to obtain the structural variations during abrasive wear.
chemical composition of the carbides. Bulk hardness of all
3. Results and Discussion
Table 1. Chemical composition (in wt% alloying elements) of
En31 steel used in the present study. 3.1. Hardness and Microstructures
Figure 1 shows the hardness of as-quenched, quenched &
C Mn Cr Si Fe
tempered and as-received En31 steel. The as-received mate-
En31 steel 1.00 0.50 1.40 0.20 balance rials has the lowest hardness (22 HRc), while the as-

© 2013 ISIJ 1472

 ISIJ International, Vol. 53 (2013), No. 8

quenched material has the highest hardness (60 HRc). After takes place from the supersaturated martensite.4) At temper-
tempering at 423 K, the drop was insignificant (2 HRc). It atures in excess of 473 K, the material softens.4,30) In the
has been reported that quenching from austenization tem- present study, there is a gradual drop in hardness with the
perature leads to a microstructure containing martensite, increase in tempering temperature from 423 K to 873 K.
about 6 vol% of retained austenite and 3–4% of cementite Such monotonous decrease in hardness with the increase in
particles which fail to dissolve during the austenizing treat- tempering temperature was also reported by Lee et al. and
ment.1,30) When the steel is tempered at low temperatures (< Leskovsek et al. for AISI 4340 and H11 tool steel respec-
473 K) to relieve the internal stresses, retained austenite tively.31,32) This decrease in hardness with tempering tem-
decomposes and precipitation of transition carbides of iron perature could be attributed to reduction in dislocation den-
sity and decomposition of martensite into ferrite and
cementite. The secondary electron micrographs of the as-
received, as-quenched and quenched + tempered En31 steel
samples are shown in Fig. 2. The as-received (i.e. as-rolled
and annealed – Fig. 2(a)) sample showed the presence of
pearlite along with pro-eutectoid cementite network along
the prior austenite grain boundaries. The approximate aver-
age interlamellar spacing of pearlite was observed to be
about 0.5 μ m. The morphology of the pearlite is related to
the initial grain size of austenite during the heat treatment;
coarser austenite leads to the formation of coarser pearlite
and consequently larger interlamellar spacing. The micro-
structure of the as-received material was pearlite with pro-
eutectoid grain boundary cementite with lower hardness of
22 HRc. The as-quenched structure (Fig. 2(b)) consisted of
un-dissolved iron carbides and martensite. The inset in Fig.
Fig. 1. Variation of bulk hardness (HRc) as a function of temper- 2(b) shows the microstructure at higher magnification,
ing temperature. The bulk hardness of the as-quenched and which clearly reveals the martensitic structure and un-dis-
as-received samples is also shown for comparison. solved iron carbides. These results were also confirmed by

Fig. 2. Secondary electron scanning electron microscopy (SEM) images of En31 steel in (a) as-received (annealed – 1 123
K followed by furnace cooling to room temperature) and (b) as-quenched (1 123 K for 3 600 s followed by oil
quenching) condition and after tempering at (c) 423 K (d) 598 K (e) 723 K (f) 848 K. All the samples were aus-
tenized at 1 123 K for 3 600 s followed by oil quenching prior to tempering treatments. (f) shows the high resolu-
tion SEM image (field emission gun - SEM) of the tempered En31 steel.

1473 © 2013 ISIJ

 ISIJ International, Vol. 53 (2013), No. 8

XRD (see Fig. 3). The XRD pattern of the as-received of the as-quenched sample (Fig. 3(b)) showed the presence
sample (Fig. 3(a)) showed ferrite (α-Fe) and cementite of martensite, retained austenite and un-dissolved Fe3C.
(Fe3C) peaks. The Fe3C peaks were very low in intensity, The martensite peaks were indexed as α-Fe peaks, since
in-spite of its greater proportion in the annealed structure the resolution of the diffractometer at smaller 2θ angles
due to the fact that the atomic scattering factor of carbon was not sufficient to resolve the tetragonality of the
is very low as compared to Fe.33) The α-Fe peaks in the martensite lattice.4,33) It has been reported4,10,34) that the
diffraction pattern were also very narrow, since the struc- alloying additions lower the martensitic start temperature
ture was completely in the annealed condition. The broad- (Ms). The role of alloying elements on the Ms temperature
ening of the X-ray peaks is related to the size and/or is given by the empirical relationship due to Andrew’s4,34)
micro-strain present in the sample.33) Reduction in the size (concentration in weight%):
of the crystallites/domains and increase in the micro-strain Ms (K) = 539 − 423(%C) − 30.4(%Mn)
leads to broadening of the X-ray peaks. The XRD pattern
−17.7(%Ni) − 12.1(%Cr) − 7.5(%Mo) + 273 ....... (1)
The Ms temperature obtained by using Andrew’s relation-
ship (Eq. (1)) for the present En31 steel was found to be
356.36 K. If the steel is cooled sufficiently rapidly from the
austenizing temperature, some amount of austenite is
retained in the microstructure (usually known as retained
austenite). The XRD pattern of the as-quenched steel (Fig.
3(b)) showed the presence of retained austenite (approxi-
mately 5–6%). It was observed that the retained austenite
persists in the microstructure even after tempering at 598 K
for 3 600 s. It may be possible that the tempering time used
was not sufficient to convert all the retained austenite into
martensite. The diffraction peaks are also substantially
broader in the as-quenched steel as compared to the as-
received and quenched + tempered steel indicating the pres-
ence of large micro-strain (due to dislocations).10) This is
typical for the as-quenched structure. An earlier investiga-
tion on En24 steel also reported the broadening of X-ray
peaks due to presence of dislocations in the quenched
microstructure.10) The broadening reduced after tempering
the martensite. Figures 2(c) to 2(f) show the microstructures
of tempered martensite in En31 steel. It can be observed
from Figs. 2(c) to 2(e) that the un-dissolved carbides
became coarser with increase in tempering temperature.
Also there is precipitation of new finer carbides. The carbide
morphology was different after tempering at 598 K and 848
K. Therefore, it was decided to use SEM-EDS for selected
heat treatment conditions to investigate/substantiate the
microstructures further. The microstructure after tempering
at the highest temperature (848 K) was not resolved proper-
ly even after using FEG-SEM. Table 2 shows the SEM-EDS
results for samples tempered at 598 K and 848 K, while Fig.
3(c) shows the XRD patterns for the quenched and 598 K
tempered samples. It can be observed from Table 2 that the

Table 2. Chemical composition of the carbides in the quenched

and tempered condition as obtained by energy dispersive
spectroscopy (in weight%).

Tempering temperature Cr Fe Si
598 K 52.48 47.52 –
848 K 03.37 96.00 0.66

Fig. 3. High resolution indexed X-ray diffraction (XRD) patterns Table 3. Bulk hardness (HRc) of heat treated En31 steel samples.
of (a) as-received (annealed – 1 123 K followed by furnace
quenched quenched quenched quenched
cooling to room temperature) sample, (b) as-quenched as- as-
and tempered and tempered and tempered and tempered
(1 123 K for 3 600 s) sample and (c) quenched and tem- received quenched
at 423 K at 598 K at 723 K at 848 K
pered sample (tempering temperature – 598 K and temper-
ing time - 3 600 s). 22 60 58 53 47 38

© 2013 ISIJ 1474

 ISIJ International, Vol. 53 (2013), No. 8

carbides after 598 K tempering were complex carbides. The The as-quenched sample consisting of martensitic structure
composition was almost 50%Cr and 50%Fe. This also with un-dissolved iron carbides showed the lowest mass
explains the different morphology of the carbides. XRD pat- loss. The as-received sample consisting of pearlite and pro-
terns (Fig. 3(c)) also substantiates the SEM-EDS results as eutectoid cementite showed highest mass loss, while the
these carbides were indexed to be of type Cr7C3. The car- quenched + tempered samples showed intermediate mass
bides after 848 K tempering were predominantly iron car- loss. For quenched + tempered samples, there was a gradual
bides (Fe3C), which were slightly enriched with Cr and Si increase in mass loss with the increase in tempering temper-
(see Table 2). The published data on the microstructural ature. For all the heat treated conditions, increase in load
evolution during tempering of En31 steel is very limited. In from 10 N to 20 N caused a corresponding increase in the
Cr containing steels, two types of Cr carbides (Cr7C3 and wear mass loss. As explained in section 3.1, after tempering
Cr23C6) are often encountered and the normal carbide at 598 K, the microstructure consisted of alloy carbides
sequence during tempering is:35) (Cr7C3) and after tempering at 848 K, the carbides were
Fe3C. The wear mass loss seems to depend only on the hard-
Matrix → (FeCr)3 C → Cr7 C3 → Cr23 C6 ness. Figure 5 shows the correlation of bulk hardness with
Cr23C6 precipitation does not occur until the Cr concen- the wear mass loss and load. It can be clearly seen from the
tration in steel exceeds about 7%. The transformation of Fig. 5 that the wear mass loss decreases with the increase in
Fe3C to Cr7C3 occurs mainly by nucleation at the Fe3C/fer- the hardness. The wear is much more pronounced at higher
rite interfaces. Cr diffuses more rapidly in steels as com- load (20 N) as compared to lower load (10 N) – about 1.85
pared to other alloying elements and hence Cr7C3 is detected times. The wear mass loss and hardness has a very good lin-
during tempering even at temperature less than 773 K and ear correlation at both the loads (the regression coefficients
it coarsens very rapidly (598 K tempering in the present are greater than 0.90 in both the cases). The increase in the
case). The sluggish microstructure after tempering at 848 K wear mass loss at higher load (20 N) could be attributed to
and presence of Fe3C indicates the decomposition of mar- the increased depth of cut by the abrasive particles on the
tensite into ferrite and Fe3C (also confirmed by the SEM- surface of the specimen with increased load.12) The abrasive
EDS results – see Table 2). wear resistance of the heat treated samples was calculated
by considering the annealed sample as reference. The abra-
3.2. Abrasive Wear Behaviour sive wear resistance of the samples tempered at 423 K and
The mass loss (gm) during two body abrasive wear of 598 K was about 2.3–2.6 times when compared with the as-
En31 steel for as-quenched, quenched & tempered and as- received sample, while it was 1.1–1.8 times for samples
received condition as a function of load is shown in Fig. 4.

Fig. 4. Mass loss (gm) as a function of tempering temperature for

10 N and 20 N loads. Mass loss of the as-received
(annealed) sample and as-quenched (hardened) sample is
also shown for reference.

Fig. 6. Secondary electron scanning electron microscopy (SEM)

images of worn out En31 steel in (a) as-received (annealed
– 1 123 K followed by furnace cooling to room tempera-
ture), (b) as-quenched (austenized at 1 123 K for 3 600 s fol-
lowed by oil quenching) condition and after tempering at
(c) 423 K (d) 598 K (e) 723 K (f) 848 K. Prior to tempering
treatments, all the samples were austenized at 1 123 K for
Fig. 5. Mass loss (gm) as a function of hardness (HRc) for 10 N 3 600 s followed by oil quenching. The load used during the
and 20 N loads. wear tests was 10 N.

1475 © 2013 ISIJ

 ISIJ International, Vol. 53 (2013), No. 8

tempered at 723 K and 848 K. Similar observations were harder were able to cause deep plastic indentation in the
reported by many investigators in the past.36,37) In the pres- annealed as-received material.41) If the ratio of hardness of
ent investigation, the tempering time was kept constant and abrasive particle to that of surface is more than 1.2 the abra-
tempering temperature was varied. It is possible that by sive particles were able to plastically indent the surface.14,42)
keeping the tempering temperature constant and varying the The width and depth of the grooves was considerably large
time may result in different microstructural features/hard- in the annealed sample (Fig. 6(a)) as compared to the as-
ness and consequently different abrasive wear resistance.38) quenched sample (Fig. 6(b)). There was no substantial dif-
However, in the present investigation, the hardness seems to ference in the morphology of abraded surfaces of as-
have a direct correlation with the abrasive wear resistance. quenched (Fig. 6(b)) and 423 K tempered sample (Fig. 6(c)).
To find the micro-mechanisms of two body abrasive wear, The width and depth of the grooves became progressively
SEM of worn surfaces (all conditions) and SEM & XRD of wider and deeper with the increase in tempering tempera-
the wear debris (few selected conditions) was carried out. ture. Figure 7 shows the worn out surface of heat treated
The SEM micrographs of worn out as-received, as- En31 steel at a load of 20 N. In Fig. 7 (20 N load), the abrad-
quenched and quenched + tempered En31 steel specimens ed surfaces have much deeper and wider grooves as com-
at a load of 10 N are shown in Fig. 6. Figure 6(a) shows long pared to Fig. 6 (10 N load). With the increase in load the
deep grooves, which were formed as the abrasive particles relative contribution of micro-cutting to total wear loss
plough across the surface and eventually removing or push- increases. These observations are consistent with the mea-
ing material into ridges along sides of the grooves formed sured mass loss values (Figs. 4 and 5). The SEM micro-
by Al2O3 abrasive particles.39,40) The abrasive particles being graphs of the wear debris of as-received, as-quenched and
quenched + 598 K tempered samples are shown in Fig. 8 at
both 10 N and 20 N loads. The wear debris particles in gen-
eral exhibited flaky morphology. The wear debris particles
were relatively finer at a load of 10 N and for the as-
quenched heat treated condition as compared to other con-
ditions. Figures 9, 10 and 11 show the XRD pattern of the
wear debris for as-received, as-quenched and quenched +
598 K tempered samples at both the loads (10 N and 20 N).
The XRD patterns did not show the presence of iron oxide
peaks for any of the heat treated condition. This indicates
that the wear mechanism was not oxidative at both the loads.

Fig. 7. Scanning electron microscopy (SEM) images of worn out

En31 steel in (a) as-received (annealed – 1 123 K followed
by furnace cooling to room temperature) and (b) as-
quenched (austenized at 1 123 K for 3 600 s followed by oil Fig. 8. Secondary electron scanning electron microscopy (SEM)
quenching) condition and after tempering at (c) 423 K (d) images of wear debris of En31 steel. (a) and (b) in as-
598 K (e) 723 K (f) 848 K. Prior to tempering treatments, received (annealed – 1 123 K followed by furnace cooling
all the samples were austenized at 1 123 K for 3 600 s fol- to room temperature) condition, (c) and (d) in as-quenched
lowed by oil quenching. The load used during the wear tests (austenized at 1 123 K for 3 600 s followed by oil quench-
was 20 N. All the images are secondary electron images ing) condition, (e) and (f) after tempering at 598 K for 3 600
except (e), in which one backscattered electron (BSE) s. The left hand side images are for load of 10 N, while the
image is also shown. right hand side images are for load of 20 N.

© 2013 ISIJ 1476

 ISIJ International, Vol. 53 (2013), No. 8

Fig. 11. High resolution XRD patterns of wear debris of En31

Fig. 9. High resolution XRD patterns of wear debris of En31 steel steel in the tempered condition (598 K – 3 600 s) for (a) 10
in the as-received (annealed) condition for (a) 10 N load N load and (b) 20 N load.
and (b) 20 N load.

For as-received and the as-quenched samples, the XRD pat-

tern showed the presence of ferrite (or martensite), cement-
ite and alumina (Al2O3). For the quenched + 598 K tem-
pered sample, along with ferrite (or martensite), cementite
and alumina (Al2O3) peaks, peaks of Cr7C3 were also pres-
ent. This indicates that along with micro-cutting and micro-
ploughing, pull out of carbide particles from the matrix also
occurred during abrasive wear at 20 N load resulting in
increased wear loss as compared to that at 10 N.

4. Conclusions
In the present study, the microstructures of En31 steel
after hardening and tempering were analyzed and the effect
of microstructures and bulk hardness on the two body abra-
sive wear behaviour was studied along with characterisation
of worn out surfaces and wear debris particles.
The as-received microstructure of En31 steel consisted of
pro-eutectoid ferrite along prior austenite grain boundaries
and pearlite. Martensite was formed after hardening, while
the un-dissolved cementite which remained after hardening
coarsened after low temperature tempering (423 K). After
tempering at 598 K, the microstructure consisted of alloy
carbides (Cr7C3) along with tempered martensite. At higher
tempering temperatures (848 K), the martensite decomposed
to give ferrite and cementite. With increase in tempering
temperature the morphology of the carbides became coarser.
Fig. 10. High resolution XRD patterns of wear debris of En31 The two body abrasive wear resistance of heat treated En31
steel in the as-quenched condition for (a) 10 N load and steel exhibited good correlation with bulk hardness; an
(b) 20 N load. increase in hardness increased the abrasive wear resistance.

1477 © 2013 ISIJ

 ISIJ International, Vol. 53 (2013), No. 8

The important material removal mechanisms during abra- 308.

14) I. M. Hutchings: Tribology: Friction and Wear of Engineering Mate-
sive wear were ploughing and cutting and pull out of car- rials, CRC Press, Arnold, London, (1992).
bides. The finer morphology of carbides was observed to be 15) K. Luo and B. Bai: Mater. Design., 31 (2010), 2510.
beneficial in increasing the abrasive wear resistance. The 16) E. Atik, U. Yunker and C. Meric: Tribo. Int., 36 (2003), 155.
17) S. G. Sapate, C. Gurijala, A. Ratho and A. Singh: Int. Conf. on Adv.
wear debris particles exhibited flaky morphology and were Mech. Eng., IDES-CPS, New York, (2011).
relatively finer at lower loads. 18) J. Suchanek and V. Kuklik: Wear, 267 (2009), 2100.
19) Y. C. Lin, S. W. Wang and T. M. Cheng: J. Mater. Process. Technol.,
120 (2002), 126.
Acknowledgement 20) A. Bahrami, S. H. Mousavi Anijdan, M. A. Golozar, M. Shamanian
The authors would like to thank Director, VNIT Nagpur and V. Varahram: Wear, 258 (2005), 846.
21) G. H. Yang and W. M. Garrison, Jr.: Wear, 129 (1989), 93.
for providing necessary facilities for the present investiga- 22) I. Sevim and I. B. Eryurek: Mater. Design, 27 (2006), 173.
tion and constant encouragement to publish this paper. The 23) G. Balachandran, A. Vadiraj, M. Kamraj and E. Kazuya: Mater.
Design, 32 (2011), 4042.
authors would like to acknowledge the use of “National 24) P. L. Hurricks: Wear, 26 (1973), 285.
Facility of Texture & OIM (a DST-IRPHA facility)” for 25) A. K. Bhakat, A. K. Mishra and N. S. Mishra: Wear, 263 (2007), 228.
26) A. K. Jha, B. K. Prasad, O. P. Modi, S. Das and A. H. Yegneswaran:
SEM-EBSD-EDS measurements. Wear, 254 (2003), 120.
27) X. Ma, R. Liu and D. Y. Li: Wear, 241 (2000), 79.
REFERENCES 28) ASM Handbbok, Metallography and Microstructures, Vol. 9, ASM
International, OH, USA, (2004).
1) H. K . D. H. Bhadeshia: Prog. Mater. Sci., 57 (2012), 268. 29) X’Pert Highscore, Version 2.1b (2.1.2), PANaltical B. V., The
2) E. N. Bamberger: Bearing Design – Historical Aspects, Present Tech- Netherlands, (2005).
nology and Future Problems, ASME, New York, (1980), 1. 30) G. R. Speich and W. C. Leslie: Metall. Trans. A, 3 (1972), 1043.
3) J. Cappel, M. Wienberg and R. Flender: Steel Grips, 2 (2004), 261. 31) W. S. Lee and T. T. Su: J. Mater. Process. Technol., 87 (1999), 198.
4) R. W. K. Honeycombe and H. K. D. H. Bhadeshia: Steels – Micro- 32) V. Leskovšek, B. Sustarsic and G. Jutrisa: J. Mater. Process Technol.,
structure and Properties, Butterworth-Heinemann, Oxford, (1995). 178 (2006), 328.
5) M. F. Carlson, B. V. Narasimha and G. Thomas: Metall. Trans., 10A 33) B. D. Cuillity: Elements of X-ray Diffraction, Addison-Wesley Pub-
(1979), 1273. lishing Company, USA, (1978).
6) D. H. Huang and G. Thomas: Metall. Trans., 2A (1971), 1587. 34) K. W. Andrews: J. Iron Steel Inst., 203 (1965), 271.
7) K. Holmberg and A. Mathews: Coatings, Tribology, Property, Tech- 35) H. K. D. H. Bhadeshia and R. W. K. Honeycombe and Steels –
niques and Applications in Surface Engineering, Elsevier, Great Britain, Microstructure and Properties, Butterworth-Heinemann, Oxford,
(1994). (2006).
8) M. H. Shaeri, H. Saghafian and S. G. Shabestari: Mater. Design, 34 36) M. A. Moore: Wear, 28 (1974), 59.
(2012), 192. 37) K. Ueda, N. Ishikawa and Y. Murota: Mater. Sci. Forum, 706–709
9) F. Katsuki, K. Watari, H. Tahira and M. Umino: Wear, 264 (2008), (2012), 2342.
331. 38) Y. Sahin, M. Erdogan and M. Cerah: Wear, 265 (2008), 196.
10) R. K. Khatirkar, P. Yadav and S. G. Sapate: ISIJ Int., 52 (2012), 39) W. M. Garrison, Jr.: Wear, 114 (1987), 239.
1370. 40) W. M. Garrison, Jr. and R. A. Garriga: Wear, 85 (1983), 347.
11) A. Sundström, J. Rendon and M. Olsson: Wear, 250 (2001), 7441. 41) M. M. Khruschov: Wear, 28 (1974), 69.
12) S. G. Sapate, A. D. Chopde, P. M. Nimbalkar and D. K. Chandrakar: 42) K. H. Zum Ghar: Microstructure and Wear of Materials, Elsevier,
Mater. Design, 29 (2008), 613. Amesterdam, (1987).
13) W. Osterle, G. Nozle and P. X. Li: Mater. Sci. Eng. A, 262 (1999),

© 2013 ISIJ 1478

View publication stats