See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/297845740

Effect of Surface Hardening Technique and Case

Depth on Rolling Contact Fatigue Behavior of
Alloy Steels

Data · March 2016

CITATIONS READS

0 136

5 authors, including:

Lechun Xie David Palmer

Deakin University BRP US, Inc.
32 PUBLICATIONS 318 CITATIONS 8 PUBLICATIONS 2 CITATIONS

SEE PROFILE SEE PROFILE

Qian Wang
Northwestern University
246 PUBLICATIONS 4,384 CITATIONS

SEE PROFILE

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

Experimental and Numerical Investigations of Stribeck Curves View project

Microstructure and mechanical properties of additive manufactured titanium alloys View project

All content following this page was uploaded by Qian Wang on 12 March 2016.

The user has requested enhancement of the downloaded file.

This article was downloaded by: [Northwestern University]
On: 30 December 2014, At: 13:49
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Tribology Transactions
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/utrb20

Effect of Surface Hardening Technique and Case Depth

on Rolling Contact Fatigue Behavior of Alloy Steels
ab c d ae a
Lechun Xie , David Palmer , Frederick Otto , Zhanjiang Wang & Q. Jane Wang
a
Center for Surface Engineering and Tribology, Northwestern University, Evanston, Illinois
60208
b
State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai
200240, China
c
Marine Propulsion Systems Division, BRP US, Inc., Sturtevant, Wisconsin
d
Midwest Thermal-Vac, Kenosha, Wisconsin 53144
e
State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400030,
Click for updates China
Accepted author version posted online: 25 Sep 2014.Published online: 15 Dec 2014.

To cite this article: Lechun Xie, David Palmer, Frederick Otto, Zhanjiang Wang & Q. Jane Wang (2015) Effect of Surface
Hardening Technique and Case Depth on Rolling Contact Fatigue Behavior of Alloy Steels, Tribology Transactions, 58:2,
215-224, DOI: 10.1080/10402004.2014.960957

To link to this article: http://dx.doi.org/10.1080/10402004.2014.960957

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Tribology Transactions, 58: 215-224, 2015
Copyright C Society of Tribologists and Lubrication Engineers

ISSN: 1040-2004 print / 1547-397X online

DOI: 10.1080/10402004.2014.960957

Effect of Surface Hardening Technique and Case Depth

on Rolling Contact Fatigue Behavior of Alloy Steels
LECHUN XIE,1,2 DAVID PALMER,3 FREDERICK OTTO,4 ZHANJIANG WANG,1,5 and Q. JANE WANG1
1
Center for Surface Engineering and Tribology, Northwestern University, Evanston, Illinois 60208
2
State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai 200240, China
3
Marine Propulsion Systems Division, BRP US, Inc., Sturtevant, Wisconsin 53177
4
Midwest Thermal-Vac, Kenosha, Wisconsin 53144
5
State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400030, China
Downloaded by [Northwestern University] at 13:49 30 December 2014

Surface hardening techniques are widely used to improve
the rolling contact fatigue (RCF) resistance of materials. This
study investigated the RCF resistance of hardened, ground
steel rods made from three different aircraft-quality alloy steels
(AISI 8620, 9310, and 4140) and hardened using different tech-
niques (atmosphere carburizing, vacuum carburizing, and in-
duction hardening) at different case depths. The RCF life of
the rods was determined using a three ball-on-rod rolling con-
tact fatigue test machine. After testing, the microstructures of
the rods were examined using metallographic techniques. The
stress distributions and plastic deformation zones for the spec-
imens under RCF were calculated using an elastoplastic model
for plastically graded materials. Relationships between sur-
face hardness, case depth, and RCF life were investigated. The
longest lives were observed for the vacuum-carburized AISI
9310 specimens, and the shortest lives were observed for the
induction-hardened AISI 4140 specimens. It is concluded that
the most important factors in determining the RCF lives of
high-cleanliness surface-hardened alloy steels are (1) the hard-
ness in the region of highest octahedral shear stress (in this
case, ∼0.13 mm beneath the surface) and (2) the depth of high
hardness (>613 HV), which determines the plastic deformation
zone size.
KEY WORDS
Ferrous Alloys, Steel; Carburizing; Case Depth; Hardening;
Hardness; Metallurgical Analysis; Optical Microscopy; Rolling
Contact Fatigue
INTRODUCTION
Rolling contact fatigue (RCF) is a failure mode that is com-
monly observed in bearings, gears, and shafts, among other com-
ponents. In order to improve the RCF life of these components,
Manuscript received February 8, 2014
Manuscript accepted August 13, 2014
Review led by Daniel Nelias
Color versions of one or more of the figures in the article can be
found online at www.tandfonline.com/utrb.
they are typically heat treated. However, through-hardening heat
treatments may induce tensile residual stresses, which are disad-
vantageous to fatigue properties (Stickels and Janotik (1); Stick-
els, et al. (2)). Furthermore, through-hardened components may
have low toughness relative to their strength (Parrish (3)). There-
fore, surface hardening processes may be used. Usually, the main
effect of such processes is to improve the hardness and wear re-
sistance of the surface layer.
Surface hardening processes include diffusion methods (such
as carburizing and nitriding), applied energy methods (such as
induction hardening and flame hardening), and coating methods
(such as electroplating, thermal spray, and physical vapor depo-
sition; Davis (4)). Diffusion and applied energy methods have an
advantage over coating processes in that the surface layers pro-
duced are integral with the substrate. Thus, these processes are
well suited for precision or high-tolerance components, especially
bearings, gears, and shafts.
Carburizing and nitriding treatments have been shown to sig-
nificantly improve the RCF lives of bearing steels (Murakami,
et al. (5); Murakami and Mitamura (6)). Over the past 40 years,
various researchers have tested the effects of carburizing and ni-
trocarburizing on the RCF lives of several types of important
bearing and gear steels, including AISI M50, M50NiL, 52100, and
9310, as well as the effects on sliding wear and spall propagation
characteristics (Townsend and Bamberger (7); Braza (8); Forster,
et al. (9)). The effect of surface hardening on RCF of other steels,
such as chromium steel, molybdenum alloyed steel, and so on, has
also been studied (Muro and Tsushima (10); Sundquist (11)).
In recent years, vacuum carburizing has become an increas-
ingly popular alternative to atmosphere carburizing. First devel-
oped in the late 1960s (Gronquist, et al. (12)), the process has
gained widespread industrial acceptance starting in the mid-1990s
(Otto and Herring (13)). Claimed advantages of the vacuum car-
burizing process include lower distortion (particularly when used
in combination with high-pressure gas quenching) and greater
control of microstructural variables such as case depth, surface
hardness, and hardness profile (Otto and Herring (14)). In a com-
parison of the effects of atmosphere carburizing and vacuum car-
burizing on AISI 8620 steel, vacuum carburizing was found to re-
sult in higher surface hardness both before and after grinding, a

215
 216 L. XIE ET AL.
TABLE 1—LADLE CHEMISTRY OF AIRCRAFT-QUALITY STEELS USED FOR RCF SPECIMENS
Material C Mn Si Ni Cr
8620 0.21 0.83 0.24 0.48 0.54
9310 0.11 0.62 0.26 3.06 1.16
4140 0.41 0.90 0.25 0.06 0.99
greater depth of high hardness (>652 HV) at the same overall
case depth, and higher surface residual compressive stresses than
atmosphere carburizing (Lindell, et al. (15)).
In addition to carburizing and nitriding, induction hardening is
an effective method of surface hardening bearing steels. A met-
allographic study was carried out on induction-hardened 0.44%
C–1.71% Mn bearing steels after rolling contact fatigue (Choi,
et al. (16)). The microstructure and its effect on rolling contact fa-
Downloaded by [Northwestern University] at 13:49 30 December 2014
tigue of induction hardened medium carbon bearing steels (Choi
and Bahng (17)) were also investigated.
Several comparisons of induction hardening to carburizing
have been performed. A study comparing atmosphere-carburized
AISI 9310 and induction-hardened AISI 4340 straight-bevel
gears concluded that induction hardening minimized distortion,
whereas carburizing maximized mechanical properties (Jones,
et al. (18)). In contrast, a study comparing atmosphere-carburized
AISI 9310 and induction-hardened AISI 1552 spur gears found
that the induction-hardened gears had a significantly longer pit-
ting fatigue life (Townsend, et al. (19)).
Although many studies on the effect of surface hardening on
RCF have been carried out over the past several decades, more
work is still needed. Methods of surface hardening for various
materials are constantly being improved, and new applications
are constantly demanding higher and higher Hertzian contact
pressures. In addition, there has been relatively little published
work on the effects of case depth on RCF life and on the mi-
crostructural variations due to RCF. Therefore, there is a need
for an in-depth study about the effects of surface hardening tech-
nique and case depth on RCF resistance of bearing steels.
This work explores the RCF resistance of hardened, ground
steel rods made from three different aircraft-quality alloy steels
(AISI 8620, 9310, and 4140) and hardened using different tech-
niques (atmosphere carburizing, vacuum carburizing, and induc-
tion hardening) at different case depths. The RCF life of the rods
was determined using a three ball-on-rod rolling contact fatigue
test machine. After testing, the microstructures of the rods were
examined using metallographic techniques. The stress distribu-
tions and plastic deformation zones in the specimens under RCF
were estimated using an elastoplastic model for plastically graded
materials. Relationships between surface hardness, case depth,
microstructure, and fatigue life were investigated.
EXPERIMENTAL
Preparation of Materials
Three types of vacuum-degassed aircraft-quality steel (AISI
8620, 9310, and 4140), certified per SAE AMS 2301 (20) require-
ments based on magnetic particle testing, were obtained in the
form of cold-drawn bars with diameter d = 12.7 mm (1/2 in.) and
length of about L = 3 m (10 ft). The AISI 8620 material was
Mo Cu P S Al Fe
0.21 0.12 0.008 0.010 0.031 Balance
0.10 0.15 0.007 0.020 0.033 Balance
0.22 0.07 0.007 0.019 0.030 Balance
also certified bearing quality per ASTM A534 (21) requirements
based on microetch testing. The ladle chemistry, in weight per-
centage, for each of the three steels is shown in Table 1. The 10-ft
bars were cut into specimens of length of 76.2 ± 1.3 mm (3.00 ±
0.05 in.) and turned down to a diameter of 10.16 ± 0.13 mm (0.400
± 0.005 in.).
The cut specimens were heat treated to three different case
depth levels (low: 0.50–0.75 mm; medium: 0.75–1.00 mm; high:
1.00–1.25 mm) using the surface hardening techniques described
below.
The 8620 specimens were divided into two groups. The first
group was atmosphere carburized, and the second group was vac-
uum carburized. The first group was atmosphere carburized at
954◦ C (1750◦ F), then cooled to 815◦ C (1500◦ F), quenched in agi-
tated 80◦ C (175◦ F) oil, and tempered at 177◦ C (350◦ F). The sec-
ond group was vacuum carburized at 940◦ C (1740◦ F) for 194 min
(low case depth), 287 min (medium case depth), or 445 min (high
case depth) and then cooled to 815◦ C (1500◦ F), held at tempera-
ture for 20 min, quenched in agitated 80◦ C (175◦ F) oil, and tem-
pered at 177◦ C (350◦ F) for 3 h.
The 9310 specimens were vacuum carburized at 940◦ C
(1740◦ F) for 261 min (low case depth), 334 min (medium case
depth), or 510 min (high case depth) and then cooled to 815◦ C
(1500◦ F), held at temperature for 20 min, quenched with nitro-
gen gas at 1,500 kPa, and double tempered at 177◦ C (350◦ F) for
3 h each.
The 4140 specimens were induction hardened by a commer-
cial heat treater using proprietary parameters. The purpose of the
carburizing heat treatments was to produce a microstructure con-
sisting primarily of tempered martensite, with a surface carbon
concentration of 0.80–0.95%. The purpose of the induction hard-
ening treatment was to produce a martensitic microstructure at
the surface, gradually transitioning to a bainitic and finally to a
ferritic/pearlitic microstructure with increasing distance from the
surface. Time–temperature transformation curves for the alloys
can be found in multiple reference works (for example, Chandler
(22)).
After heat treatment, all of the specimens were centerless
ground to a final diameter of 9.53 mm (0.375 in.), with a surface
finish of 0.05–0.10 μm (2–4 μin.).
RCF Testing
The RCF testing was conducted at the Center for Surface En-
gineering and Tribology at Northwestern University. A schematic
and photo of RCF tester are shown in Fig. 1. The test speci-
mens were held vertically by a precision collet. Three AISI 52100
steel balls (d = 12.7 mm) were loaded against the specimen. The
balls were held at equal spacing around the rod by a brass re-
tainer. The radial load was applied through two tapered races
 Rolling Contact Fatigue Behavior of Alloy Steels 217
Downloaded by [Northwestern University] at 13:49 30 December 2014

Fig. 2—Surface hardness vs. average case depth for alloy steels after
different surface hardening treatments.

measured until RCF testing was complete. Therefore, after the

Fig. 1—Schematic and photo of the RCF tester: (a) top view; (b) side
cross-sectional view; and (c) photo.
RCF testing, each of the RCF specimens was sectioned perpen-
dicular to its longitudinal axis at a distance of approximately
12.7 mm (1/2 in.) from the contact zone. This distance was chosen
in order to ensure that the measurement would not be affected
that were clamped together by the force of three calibrated com-
by strain hardening or microstructural changes that occurred in
pression springs, each individually adjusted to provide a balanced
the contact zone during RCF. The sections were mounted and
radial load for each ball. The spring load used in this work was
polished. Effective case depth was determined by measuring mi-
319 N. This load was selected in order to obtain failures in within
crohardness at 0.127-mm (0.005-in.) increments from the surface,
approximately 106–107 cycles.
using a Vickers indenter with a 500 g load. In this article, in ac-
New balls were used for each test, and the races were re-
cordance with standard industry practice, the effective case depth
placed after 10 tests. Before each test, the equipment was cali-
is defined as the depth at which the hardness is equal to 513 HV,
brated based on the external load. The rods, balls, and retainer
linearly interpolating between successive measurements. The mi-
were cleansed of any debris in an ultrasonic bath. Once the balls
crohardness measurements were carried out at Midwest Thermal
were loaded against the test rod, the rod was rotated at a con-
Vac using a Buehler microhardness tester with OmniMet image
stant speed of 3,600 rpm. The contact area was continuously lu-
analysis software.
bricated by a commercial lubricant (NMMA TCW-3) through a
In order to account for the fact that the case depth may not
gravity drip system at a rate of 8–10 drops per minute (about
be uniform at all points around the diameter of the specimen, mi-
0.3 mL/min). An accelerometer was used to monitor the vibra-
crohardness traverses were made at three locations, located 120◦
tion amplitude of the ball retainer mechanism. When the vibra-
apart. The diameter and center of the circumscribed circle were
tion amplitude exceeded a preset threshold, the motor used to
rotate the test rod was automatically shut off and the test was
complete. The duration of the test was used to calculate the num-
ber of contact cycles until the contact fatigue failure. In this work,
the vibration amplitude was set as 0.5 cm at the machine-set ref-
erence point.

Surface Hardness Testing

The surface hardness of each of the RCF specimens was tested
using a Vickers indenter with a 200 g load. The dwell time with
loading for each test was 5 s. The hardness tests were carried out
in the Optical Microscopy and Metallography Facility of the De-
partment of Materials Science and Engineering at Northwestern
University.

Case Depth Measurement and Metallography

Because it was desired to relate the case depth of each indi-
vidual specimen to its RCF life, and because case depth mea- Fig. 3—Cycles to failure vs. surface hardness for alloy steels after differ-
surement is a destructive test, the case depth could not be ent surface hardening treatments.
 218 L. XIE ET AL.
Downloaded by [Northwestern University] at 13:49 30 December 2014

Fig. 4—Microhardness profile for (a) vacuum-carburized AISI 8620, 24,105,600 cycles to failure; (b) atmosphere-carburized AISI 8620, 5,248,800 cycles
to failure; (c) vacuum-carburized AISI 9310, 42,832,800 cycles to failure; and (d) induction-hardened AISI 4140 rolling contact fatigue specimen,
453,600 cycles to failure.

calculated from these three points. In turn, this was used to cal-
culate the overall case depth and case eccentricity for each spec-
imen. For the purposes of this article, the average case depth is
defined as one-half of the difference between the outer diame-
ter of the specimen and the diameter of the circumscribed cir-
cle. (For three equally spaced case depth measurements—that is,
120◦ apart—this is identically equal to the average of the three
measurements.) The case eccentricity is defined as the distance
between the center of the specimen to the center of the circum-
scribed circle. The minimum case depth is defined as the overall
case depth minus the case eccentricity.
In addition to the sections used for microhardness testing, lon-
gitudinal sections were cut, perpendicular to the contact groove,
for four specimens with approximately the same case depth, one
from each group. The longitudinal sections were mounted, pol-
ished, and etched with 5% nital. The etched microstructures were
examined with an optical microscope at the Product Develop-
ment Center of BRP US, Inc.
MODELING
The stress distributions and plastic deformation zones of the
specimens under RCF were estimated using an elastoplastic
model for plastically graded materials. This model has been de-
scribed in detail in a recent paper (Wang, et al. (23)). The yield
strength as a function of radial position was modeled based on

TABLE 2—AVERAGE RCF LIVES FOR ALLOY STEELS WITH DIFFERENT SURFACE TREATMENTS AND CASE DEPTHS
Surface Average Life Average Case Eccentricity Depth to Hardness at
Material Treatment (Cycles) Depth (mm) (mm) 613 HV (mm) 0.127 mm (HV)

8620 Vacuum 2,300,400 0.813 0.066 0.484 777

8620 Vacuum 29,095,200 1.083 0.075 0.718 812
8620 Vacuum 37,778,400 1.500 0.066 0.942 866
8620 Atmosphere 2,049,300 0.578 0.067 0.287 693
8620 Atmosphere 8,812,800 1.009 0.093 0.545 716
8620 Atmosphere 21,772,800 1.332 0.141 0.560 673
9310 Vacuum 26,568,000 0.677 0.030 0.470 749
9310 Vacuum 43,372,800 1.078 0.033 0.793 782
9310 Vacuum 52,401,600 1.257 0.027 0.931 793
4140 Induction 237,600 0.502 0.041 0.080 605
4140 Induction 410,400 0.720 0.060 0.342 631
4140 Induction 604,800 1.115 0.086 0.465 635
 Rolling Contact Fatigue Behavior of Alloy Steels 219

the microhardness testing results, using the approximation H =

3σY0 , where H is the Vickers hardness and σY0 is the initial yield
strength in millipascals (Tabor (24)). Power-law strain hardening
behavior was assumed, and an isotropic hardening law was used,
as shown in the following equation:

σY = σY0 (1 + ε̄np ), [1]

where σY0 is the yield stress, ε̄p is the equivalent plastic strain, and
n is the strain hardening exponent. Following Branch, et al. (25),
the strain hardening exponent was assumed to be constant with
respect to hardness. Strain hardening exponents from the AISI
Bar Steel Fatigue database (n = 0.140 for AISI 8620, n = 0.035
for AISI 4140, and n = 0.113 for AISI 9310) were used (American
Iron and Steel Institute (26)). A uniform 128 × 128 × 128 grid was
used to cover the computational domain.
Downloaded by [Northwestern University] at 13:49 30 December 2014

RESULTS AND DISCUSSION

Surface Hardness
The surface hardness of the specimens is plotted against the
average case depth in Fig. 2. Best-fit lines are shown for illustra-
tion purposes. As Fig. 2 shows, the surface hardness increased
with increasing case depth, with the notable exception of the
atmosphere-carburized AISI 8620 specimens. However, the re-
lationship is different for each material and surface hardening
method.
Although increasing surface hardness is often observed with
increasing case depth (see, for example, Ahmad, et al. (27)), this
is not always the case. Decreasing surface hardness with increas-
ing case depth has been reported for atmosphere-carburized AISI
8620 (Zhang (28)). This is most likely due to a lower carbon con-
centration at the surface of specimens with higher case depth,
which in turn is due to the longer diffusion time needed to ob-
tain a higher case depth.
Cycles to failure are plotted against surface hardness in Fig. 3.
The number of cycles to failure increased with increasing surface
hardness, again with the exception of the atmosphere-carburized
AISI 8620 specimens. Once again, different relationships were
found for each material and surface hardening method.

Microhardness Profiles
The microhardness profiles for one specimen from each
group, each with a case depth around 1 mm, are shown in Fig. 4.
Though the case depth, defined using the traditional industry def-
inition of the depth to a hardness of 513 HV, is roughly the same
for each specimen, the microhardness profiles differ significantly.
The carburized specimens, particularly the vacuum-carburized
AISI 9310, maintain a high hardness to a greater depth. Each
of the carburized specimens has a hardness exceeding 613 HV
over more than 50% of the effective case depth; for the vacuum-
carburized AISI 9310 specimen, the hardness exceeds 613 HV
over more than 75% of the effective case depth. In contrast, the
hardness of the induction-hardened specimen exceeds 613 HV
over less than 25% of the effective case depth. It is hypothesized
that the greater depth of high hardness results in a smaller plas-
tic deformation zone, which, in turn, is responsible for the longer
rolling contact fatigue lives of these specimens.
Fig. 5—Cycles to failure vs. (a) average case depth; (b) average depth to
613 HV; and (c) hardness at depth of 0.127 mm.
RCF Life
The average RCF lives for each group of the specimens are
shown in Table 2, along with the average case depth, eccentricity,
depth to 613 HV, and hardness at a distance of 0.127 mm from
the surface.
The RCF life is plotted against the average case depth for each
specimen in Fig. 5a. For all materials and surface treatments, the
RCF life increases with increasing case depth. At the same case
depth, the RCF life for the vacuum-carburized AISI 9310 speci-
mens is roughly two orders of magnitude greater than that for the
 220 L. XIE ET AL.
Downloaded by [Northwestern University] at 13:49 30 December 2014

Fig. 6—Optical micrographs after RCF: (a) Vacuum-carburized AISI 8620 specimen. Dark etching area, shown at left, is approximately 0.49 mm wide,
0.15 mm deep, and begins at a depth of 0.024 mm from the surface. Case microstructure is shown at right. Case depth is 1.14 mm; 29,872,800
cycles to failure. (b) Atmosphere-carburized AISI 8620 specimen. Dark etching area, shown at left, is approximately 0.99 mm wide, 0.35 mm deep,
and begins at a depth of 0.020 mm from the surface. Case microstructure is shown at right. Case depth is 1.03 mm; 5,248,800 cycles to failure. (c)
Vacuum-carburized AISI 9310 specimen. Dark etching area, shown at left, is approximately 0.44 mm wide, 0.16 mm deep, and begins at a depth
of 0.020 mm from the surface. Case microstructure is shown at right. Case depth is 1.10 mm; 51,256,800 cycles to failure. (d) Induction-hardened
AISI 4140 specimen. Light etching area, shown at left, is approximately 0.78 mm wide, 0.22 mm deep, and begins at a depth of 0.028 mm from the
surface. Case microstructure is shown at right. Case depth is 1.17 mm; 874,800 cycles to failure.
 Rolling Contact Fatigue Behavior of Alloy Steels
Downloaded by [Northwestern University] at 13:49 30 December 2014
Fig. 7—Von Mises stress distributions in AISI 9310 specimens: (a) low case depth; (b) medium case depth; and (c) high case depth.
induction-hardened AISI 4140 specimens. The RCF lives for the
atmosphere-carburized and vacuum-carburized AISI 8620 spec-
imens fall somewhere between these two extremes. There is a
great deal of scatter in the data for the AISI 8620 specimens, but
the atmosphere-carburized and vacuum-carburized AISI 8620
specimens appear to follow the same general trend.
In Fig. 5b, the RCF life is plotted against the average depth to
613 HV for each specimen. It is observed that this measurement
provides an improved correlation to RCF life compared to the
Fig. 8—Plastic deformation zones in AISI 9310 with different case depths.
221
case depth measured according to the traditional definition (i.e.,
depth to 513 HV).
In Fig. 5c, the RCF life is plotted against the hardness at a
depth of 0.127 mm (0.005 in.) for each specimen. Compared to
the surface hardness, this measurement (i.e., subsurface hard-
ness) provides an improved correlation to RCF life. Notably, the
apparently anomalous behavior for the atmosphere carburized
AISI 8620 specimens (i.e., increasing RCF life with decreasing
hardness) disappears.
In this study, the longest RCF lives were observed for the
vacuum-carburized AISI 8620 and AISI 9310 specimens, whereas
the shortest RCF lives were observed for the induction-hardened
AISI 4140 specimens. At the same case depth, the RCF life of
the vacuum-carburized AISI 9310 specimens exceeded that of
the induction-hardened AISI 4140 specimens by two orders of
magnitude or more. This result differs from that of Townsend,
et al. (19), who found longer RCF lives for induction-hardened
specimens than for carburized specimens. This difference may
be explained by the fact that Townsend, et al. (19) ground the
carburized specimens after heat treatment but did not grind
the induction-hardened specimens after heat treatment. In
contrast, in this study, all of the specimens were ground after
heat treatment. This means that the immediate surface layer,
which generally has the highest hardness and highest residual
compressive stresses, was removed.
The need for grinding is largely a result of distortion occur-
ring during the heat treatment process. Therefore, less grinding
stock would be necessary, and more of the surface layer could
be preserved, if distortion can be minimized. This might result in
longer RCF lives. In this study, the amount of distortion resulting
 222 L. XIE ET AL.

from the heat treatment process was measured indirectly by
means of the eccentricity of the hardened case with respect to the
ground outer diameter of the specimen (see Table 2). The low-
est eccentricities were observed for the vacuum-carburized and
gas-quenched AISI 9310 specimens.
For each group of specimens, with the exception of the
atmosphere-carburized AISI 8620 specimens, the RCF life in-
creased with increasing surface hardness (see Fig. 3). This may
be explained by the fact, unlike the other specimen groups, the
atmosphere-carburized AISI 8620 specimens had decreasing sur-
face hardness with increasing case depth (see Fig. 2). When the
RCF lives are compared on the basis of hardness at a depth of
0.127 mm from the surface, rather than the hardness at the imme-
diate surface, all of the specimen groups show an increase in RCF
life with increasing hardness (see Fig. 5c).
Downloaded by [Northwestern University] at 13:49 30 December 2014
small amounts of retained austenite (<5%). The case microstruc-
ture of the induction-hardened AISI 4140 specimens consisted
of a mixture of martensite and bainite. For the AISI 8620 and
AISI 9310 specimens, dark etching areas were observed below
the surface in the area of rolling contact. For the AISI 8620 spec-
imens, light etching bands were observed within the dark etching
area. No such bands were observed for the AISI 9310 specimen.
For the induction-hardened AISI 4140 specimen, light etching
bands were observed, but no dark etching area was observed. The
microstructures are shown in Fig. 6.
Dark etching areas and light etching bands have been ob-
served in steels subject to RCF since at least 1945 (Jones (29)) and
are known to result from microstructural changes due to cyclic
stresses (Swahn, et al. (30)). The shapes and locations of these
areas indicate the region of maximum subsurface stress.

All of the specimen groups show an increase in RCF life with
increasing case depth (Fig. 5a). However, the relationship ap-
pears to be different for each material. Comparing the RCF lives
on the basis of the depth to a hardness of 613 HV, rather than the
traditional industry definition of case depth, appears to provide a
more straightforward relationship with more general applicabil-
ity (see Fig. 5b). The results of this study are in agreement with
those of Lindell, et al. (15), in that vacuum carburizing was found
to result in a greater depth of high hardness than other surface
hardening methods.
Metallography
The case microstructure of the atmosphere-carburized AISI
8620, vacuum-carburized AISI 8620, and vacuum-carburized
AISI 9310 specimens consisted of fine tempered martensite with
For all of the specimens, the dark etching areas and/or light
etching bands extended to a depth of 0.15–0.35 mm from the sur-
face. This observation helps to explain why the hardness at the
depth of 0.127 mm (0.005 in.) from the surface showed a closer
correlation to the RCF life than did the surface hardness.
Stress and Plastic Deformation Zones
The numerically estimated stress distributions and plastic de-
formation zones for the AISI 9310 material with three different
case depths (low, medium, and high) are shown in Figs. 7 and
8. Similar results are expected for the other materials. The mi-
crohardness profiles used in the numerical analyses, determined
by averaging the experimentally determined microhardness val-
ues for each specimen type, are given in Table 3. A comparison of
Figs. 7b and Fig. 6c reveals that the calculated plastic deformation

TABLE 3—MICROHARDNESS PROFILES USED IN NUMERICAL ANALYSES

Hardness (HV)

Depth Depth 8620, 8620, 8620, 8620, 8620, 8620,

(in) (mm) Vacuum, Low Vacuum, Medium Vacuum, High Atmosphere, Low Atmosphere, Medium Atmosphere, High

0.005 0.127 794 811 850 692 714 676

0.010 0.254 748 782 834 634 684 654
0.015 0.381 678 721 813 563 651 648
0.020 0.508 610 692 767 526 620 626
0.025 0.635 563 635 718 505 582 600
0.030 0.762 516 584 665 496 545 586
0.040 1.016 494 522 588 493 505 544
0.050 1.270 483 497 544 499 489 516
0.060 1.524 479 483 519 502 490 502

9310, 9310, 9310, 4140, 4140, 4140,

Vacuum, Low Vacuum, Medium Vacuum, High Induction, Low Induction, Medium Induction, High

0.005 0.127 746 784 794 605 657 621

0.01 0.254 711 784 778 585 623 645
0.015 0.381 656 761 739 594 673 611
0.020 0.508 590 721 741 533 623 600
0.025 0.635 524 681 705 397 572 599
0.030 0.762 479 637 665 299 514 582
0.040 1.016 434 537 589 278 323 554
0.050 1.270 415 463 508 278 260 489
0.060 1.524 416 434 458 276 274 381
 Rolling Contact Fatigue Behavior of Alloy Steels 223

TABLE 4—RESULTS OF THE ELASTOPLASTIC SIMULATION

Surface Case Maximum von Mises Depth of Maximum Volume of Plastic
Material Treatment Depth Stress (MPa) von Mises Stress (mm) Deformation Zone (mm3)

8620 Vacuum Low 3,566 0.138 0.065

8620 Vacuum Medium 3,603 0.148 0.056
8620 Vacuum High 3,653 0.148 0.042
8620 Atmosphere Low 3,279 0.128 0.113
8620 Atmosphere Medium 3,356 0.138 0.085
8620 Atmosphere High 3,229 0.128 0.093
9310 Vacuum Low 3,556 0.138 0.078
9310 Vacuum Medium 3,638 0.148 0.057
9310 Vacuum High 3,647 0.148 0.057
4140 Induction Low 3,531 0.128 0.120
4140 Induction Medium 3,665 0.148 0.094
4140 Induction High 3,636 0.189 0.110
Downloaded by [Northwestern University] at 13:49 30 December 2014

zones coincide well with the observed dark etching area. For all
three case depths, the maximum stress was approximately 3.6
GPa and occurred at a distance of approximately 0.13 mm below
the surface. As shown in Fig. 5c, the experimental results indicate
that the hardness at this depth correlates well to the fatigue life.
Note that in Fig.7, the z direction represents the depth of radial
direction of the shaft.
RCF life is known to be related to the plastic zone size.
Figure 8 indicates that the size of the plastic zone decreases with
increasing case depth. This agrees well with the experimental re-
sult, shown in Figs. 5a and 5b, that the RCF life increases with
increasing case depth. This comparison supports the theory that
RCF life increases with decreasing plastic zone size. The calcu-
lated von Mises stresses, depths, and plasticity volumes are listed
in Table 4.
CONCLUSIONS
The RCF behaviors of four types of alloy steels were inves-
tigated. The results show that both the surface hardening tech-
niques and the case depth influence the hardness, RCF lives, and
microstructure.
approximate depth of the highest subsurface von Mises stress.
Thus, the hardness in the region of highest subsurface stress
is more important in determining RCF life than the surface
hardness.
4. All of the specimen groups showed an increase in RCF life
with increasing case depth. However, a different relationship
was found for each material. Comparing the RCF lives on the
basis of the depth to a hardness of 613 HV, rather than the
traditional industry definition of case depth, provides a more
straightforward relationship with more general applicability.
5. Vacuum carburizing was found to result in a greater depth of
high hardness than other surface hardening methods. The test
data indicate that the depth of high hardness is a key factor in
determining RCF life. A greater depth of high hardness results
in a smaller plastic deformation zone, which leads to a longer
RCF life.
6. Metallographically dark etching and light etching areas ob-
served in the specimens after RCF coincided well with the
plastic deformation zones predicted by elastoplastic model-
ing. This indicates that metallography is a valuable tool for
determination of the plastic deformation zone size.

1. The longest RCF lives were observed for the vacuum-
carburized AISI 8620 and AISI 9310 specimens, and the short-
est RCF lives were observed for the induction-hardened AISI
4140 specimens. At the same case depth, the RCF life of the
vacuum-carburized AISI 9310 specimens exceeded that of the
induction-hardened AISI 4140 specimens by two orders of
magnitude or more.
2. For each group of specimens, with the exception of the
atmosphere carburized AISI 8620 specimens, RCF life in-
creased with increasing surface hardness. The exception may
be explained by the fact that, unlike the other groups, the
atmosphere-carburized AISI 8620 specimens showed decreas-
ing surface hardness with increasing case depth. This effect
has been reported in the literature before and is believed to
result from a lower carbon concentration at the surface due to
a longer diffusion time.
3. The hardness at the depth of 0.127 mm (0.005 in.) from the sur-
face showed a closer correlation to the RCF life than did the
surface hardness. Based on elastoplastic modeling, this is the
FUNDING
This work was supported by BRP US, Inc., and the Center for
Surface Engineering and Tribology at Northwestern University,
with technical assistance from Midwest Thermal Vac. Lechun Xie
acknowledges the scholarships from the China Scholarship Coun-
cil (No. 2011623074) and State Key Laboratory of Metal Matrix
Composites, Shanghai Jiao Tong University, Shanghai, China.
REFERENCES
(1) Stickels, C. and Janotik, A. (1980), “Controlling Residual Stresses in 52100
Bearing Steel by Heat Treatment,” Metallurgical Transactions A, 11(3),
pp 467–473.
(2) Stickels, C. A., Anderson, A. T., and Janotik, A. M. (1977), “Heat Treat-
ment for Ball Bearing Steel to Improve Resistance to Rolling Contact Fa-
tigue,” U.S. Patent 4,023,988 A.
(3) Parrish, G. (1999), Carburizing: Microstructures and Properties, 2nd ed.,
ASM International: Materials Park, OH.
(4) Davis, J. R. (2002), Surface Hardening of Steels: Understanding the Basics,
ASM International: Materials Park, OH.
(5) Murakami, Y., Matsumoto, Y., and Uemura, K. (1990), “Rolling Bearing,”
U.S. Patent 4,930,909.
 224
(6) Murakami, Y. and Mitamura, N. (1991), “Rolling Bearing,” U.S. Patent
5,030,017.
(7) Townsend, D. P. and Bamberger, E. N. (1991), “Surface Fatigue Life
of M50NiL and AISI 9310 Gears and Rolling-Contact Bars,” Journal of
Propulsion and Power, 7(4), pp 642–649.
(8) Braza, J. (1992), “Rolling Contact Fatigue and Sliding Wear Performance
of Ferritic Nitrocarburized M50 Steel,” Tribology Transactions, 35(1),
pp 89–97.
(9) Forster, N. H., Rosado, L., Ogden, W. P., and Trivedi, H. K. (2009),
“Rolling Contact Fatigue Life and Spall Propagation Characteristics of
AISI M50, M50 NiL, and AISI 52100, Part III: Metallurgical Examina-
tion,” Tribology Transactions, 53(1), pp 52–59.
(10) Muro, H. and Tsushima, N. (1970), “Microstructural, Microhardness
and Residual Stress Changes Due to Rolling Contact,” Wear, 15(5),
pp 309–330.
(11) Sundquist, H. (1981), “Rolling Contact Fatigue of Case-Hardened
Chromium Steel,” Wear, 66(1), pp 111–123.
(12) Gronquist, E., Scotto, V., Taft, D., and Western, H. (1974), “Method of
Vacuum Carburizing,” U.S. Patent 3,796,615.
(13) Otto, F. and Herring, D. (2007), “Low-Pressure Vacuum Carburizing,”
Downloaded by [Northwestern University] at 13:49 30 December 2014
Advanced Materials and Processes, 165(3), pp 31–33.
(14) Otto, F. and Herring, D. (2007), “Advancements in Precision Carburizing
of Aerospace and Motorsports Materials,” Heat Treating Progress, 7(3),
pp 35–42.
(15) Lindell, G. D., Breuer, D. J., and Herring, D. H. (2002), “Selecting the Best
Carburizing Method for the Heat Treatment of Gears,” AGMA Technical
Paper 02FTM7.
(16) Choi, B.-Y., Shin, J.-H., Bahng, G.-W., and Yoon, K.-B. (1996),
“Metallographic Study on Rolling Contact Fatigue of 0.44%
C–1.71% Mn Induction-Hardened Bearing Steels,” Wear, 192(1),
pp 1–5.
(17) Choi, B. and Bahng, G. (1998), “Characterisation of Microstructure
and Its Effect on Rolling Contact Fatigue of Induction Hardened
Medium Carbon Bearing Steels,” Materials Science and Technology, 14(8),
pp 816–821.
View publication stats
L. XIE ET AL.
(18) Jones, K. T., Newsome, M. R., and Carter, M. D. (2010), “Gas Carburizing
vs. Contour Induction Hardening in Bevel Gears,” Gear Solutions, 8(82),
pp 38–44, 51, 54.
(19) Townsend, D. P., Turza, A., and Chaplin, M. (1995), “The Surface Fatigue
Life of Induction Hardened AISI 1552 Gears,” AGMA Technical Paper
95FTM5.
(20) Steel Cleanliness, Aircraft Quality Magnetic Particle Inspection Proce-
dure. (2010), SAE AMS 2301K, August 2010.
(21) Standard Specification for Carburizing Steels for Anti-Friction Bearings.
(1994), ASTM A534, January 1994.
(22) Chandler, H. (Ed.). (1995), Heat Treaters’ Guide: Practices and Procedures
for Irons and Steels, 2nd ed., ASM International: Materials Park, OH.
(23) Wang, Z. J., Jin, X. Q., Zhou, Q. H., Ai, X. L., Keer, L. M., and Wang,
Q. (2013), “An Efficient Numerical Method with a Parallel Computational
Strategy for Solving Arbitrarily Shaped Inclusions in Elasto-Plastic Con-
tact Problems,” Journal of Tribology, 135(3), p 031401.
(24) Tabor, D. (1970), “The Hardness of Solids,” Review of Physics in Technol-
ogy, 1(3), pp 145.
(25) Branch, N. A., Arakere, N. K., Subhash, G., and Klecka, M. A. (2010),
“Determination of Constitutive Response of Plastically Graded Materi-
als,” International Journal of Plasticity, 27(5), pp 728–738.
(26) American Iron and Steel Institute. “Bar Steel Fatigue Database,” Avail-
able at: http://barsteelfatigue.autosteel.org/ (accessed July 8, 2014).
(27) Ahmad, T., Yousaf, O., Salam, A., and Ahmad, J. (2009), “Relationship
between Case Depth and Hardness in an Induction Hardened Medium
Carbon Steel,” Journal of the Pakistan Institute of Chemical Engineers, 37,
pp 31–33.
(28) Zhang, C. (2009), Assessment of Depth of Case-Hardening in Steel Rods by
Electromagnetic Methods, M.S. Thesis, Iowa State University, Iowa, USA.
(29) Jones, A. B. (1946), “Metallographic Observations of Ball Bearing Fatigue
Phenomena,” Symposium on Testing of Bearings: 49th Annual Meeting,
pp 35–52, American Society for Testing of Materials: Buffalo, NY,
(30) Swahn, H., Becker, P. C., and Vingsbo, O. (1976), “Martensite Decay dur-
ing Rolling Contact Fatigue in Ball Bearings,” Metallurgical Transactions
A, 7(8), pp 1099–1110.