Wear, 155 (1992) 15-29 15

Performance studies on treated steels as substitutes for

conventional bearing materials

A. Ramamohana Rae’*, R. Marappanb and P. V. Mohanrama

‘Indian Institute of Technology, Madras 600036 (India)
bGovernment College of Engineering, Salem 636 011 {India)

(Received April 22, 1991; revised and accepted October 17, 1991)

Abstract

Sursulf and Tufftride treatments were employed on typical plain carbon steels and grey
iron. The treated materials were tested under mixed lubrication conditions in a bearing
test rig. Friction and wear characteristics were established and the results compared with
those obtained from conventional bearing materials, namely brass and bronze. The results
show conclusively that the treated ferrous materials perform satisfactorily in relation to
the untreated and conventional bearing materials, sometimes even showing improved
performance. Among the test situations investigated, on the basis of friction, wear and
economic considerations, Sursulf-treated like pairs appear to be the best.

1. Introduction

Wear resistance and cost are two important factors that go with the selection of
bearing materials for lubricated journal bearings. However, the fact that conventional
journal bearing materials are costly necessitates a continued search for cheaper materials.
It is evident that wear loss can be minimized by suitably selecting the bearing material
pair and the operational parameters such as load, speed, lubricant viscosity etc., so
as not to operate under conditions of metal-to-metal contact [l]. Even with well
designed bearings, occasional metal-to-metal contact due to unexpected variation in
the operational parameters and unavoidable slow speeds during starting and stopping
results in excessive wear compared with that of steady-state operational periods. Hence,
it is desirable to go for cheaper and easily available bearing materials to reduce the
overall cost including that of maintenance and inventory. Plain carbon steels and grey
iron are comparatively cheap and readily available but their wear properties are poor.
Sursulf-treatment and Tufftriding are two liquid nitridmg processes which are capable
of improving fatigue resistance, wear and scuffing properties of ferrous materials, which
are incidently some of the desirable characteristics for bearing materials as well. The
objective of this work was to determine the suitability of Sursulf-treated and Tufftrided
plain carbon steels of two different compositions and a particular grade of grey iron,
as wear-resistant bearing materials under mixed lubrication.
The Sursulf treatment bath (surface hardening sulphur catalyst) [2] operates at
a temperature of 570 “C and produces on the treated parts a compound zone of 2&30
pm thickness and a nitrogen diision zone of 300-500 pm thickness, depending on

*Present address: Ohio State University, Columbus, OH 43210, USA.

0043-1648/92/$5.00 0 1992 - Elsevier Sequoia. All rights reserved

16

the material, for an average treatment time of 90 min. The compound zone consists
of only epsilon iron nitride with a very shallow finely porous zone at the extreme
surface, containing sulphur compounds. This zone is supposed to help in reducing the
scuffing and wear loss. Diffusion of nitrogen produces a residual compressive stress
at the surface of treated parts [3]. This in turn confers a considerable increase in
fatigue strength both at the surface and in the substrate of the materials. The process
also improves the resistance to corrosion and cavitation erosion [4]. This treatment
can be applied to plain carbon steels, low alloy and tool steels, high-speed steels and
stainless steels containing up to 25 wt.% chromium. It can also be used for the
treatment of white irons, grey irons, and ductile irons.
Tufftriding (Tuffnitriding) is another liquid nitriding process designed to improve
the resistance of ferrous materials to wear and fatigue [S, 61. The Tufftride bath
operates at a temperature of 570 “C and produces a compound zone of approximately
20 pm thickness and a nitrogen diffusion zone of 500-800 pm thickness depending
on the material, for an average treatment time of 90 min. The compound zone consists
of carbide bearing epsilon iron nitrides. This zone is ductile in nature [7]. Diffusion
of nitrogen gives a considerable increase in fatigue strength [6]. This process also
improves the resistance to corrosion and cavitation erosion.

2. Experimental details

Details of the test specimens comprising the two parts, the bush and the journal,
are shown in Fig. 1. The various combinations of treated materials for the bush and
the journal studied in this work are listed in Table 1. The elemental composition of

Fig. 1. Test specimen: (a) bush; (b) journal.

 17

TABLE 1
Details of materiaIs tested

Bush material Journkl material

0.17% plain 0.25% plain

carbon steel carbon steel
Grey iron Grey iron

TABLE 2
imposition of carbon and manganese in the materials tested

Material Composition (wt.%)

Carbon Manganese
_ - ~~~
0.17% plain 0.17 0.65
carbon steel
0.25% plain 0.25 0.60
carbon steel
Grey iron Pearlitic grey iron grade 20 with A type
flake graphite

the above materials is given in Table 2. The untreated bearing materials studied for
comparison include conventional pairs such as brass and bronze against steel and
unconventional pairs such as steel against steel and grey iron against grey iron.
The two parts of the bearing specimens were first ground finished giving an
average diametral clearance of 30 JUII when assembled. The surface roughnesses of
bush and journal specimens were maintained between 0.2-0.4 and OS-O.7 pm centre-
line average (c.1.a.) value respectively. Different specimens prepared from the same
lot were segregated into two groups. Each group was treated in one of the two different
commercial salt baths, Sursulf or Tufftride. After treatment, the changes in dimensions
and roughnesses were found to be insi~ificant. However, the resulting surface hardness
change was sig~~nt, as shown in Table 3. The hardness gradually decreased with
increasing depth below the surface.
The experimental set-up used to test the specimens is shown in Fig. 2. The drive
shaft (7) which carries the test journal was supported by two self-aligning ball bearings
(12) and driven by a variable speed d.c. motor (13) through V-belts. The test bearing
(1) was a full journal bearing mounted in a floating bracket (2) which was loaded by
means of a loading arrangement (8), (9), (10). The frictional torque was measured
using strain gauges (11) mounted on a cantilever which engages with the floating
bracket. A gravity feed arrangement supplied lubricating oil to the bearing assembly
through an oil groove in the floating bracket and an oil hole in the bush located
diagonally opposite to the loading side. The oil film temperature was measured using
a thermocouple (4) located at approximately 2 mm from the friction interface in the
region of expected ~nimum film thickness. The oil collector (6) collected the oil
coming from the bearing assembly. The frictional force measured by the strain gauges
18

TABLE 3

Effect of treatment on surface hardness

Treatment Material Microhardness, VPHa

Before After
treatment treatment

Sursulf 0.17% 138 330

plain carbon
steel
0.25% 141 362
plain carbon
steel
Grey iron 185 320

Tufftride 0.17% 138 334

plain carbon
steel
0.25% 141 346
plain carbon
steel
Grey iron 185 308
‘Near the surface, load 0.5 N.

Fig. 2. Experimental set-up: (1) test specimen, (2) floating bracket, (3) oil supply hose, (4)
thermocouple, (5) revolution counter, (6) oil collector, (7) drive shaft, (8) connecting links, (9)
loading lever, (10) loading pan, (11) strain gauge, (12) support bearings, (13) d.c. motor, (14)
speed indicator probe.

was continuously recorded using an X-Y plotter. The wear in the bearing parts was
measured by considering the weight loss, to an accuracy of 0.1 mg.
It is well known [l] that, for a journal bearing, the Stribeck curve which relates
the coefficient of friction f with the bearing characteristic number I,+ (q&) has three
distinct regimes as shown in Fig. 3. Here r) is the dynamic viscosity of the lubricating
oil, o is the angular velocity of the journal and g is the load per unit projected area
of the bearing. f remains approximately constant in the boundary lubrication regime
and it reduces with increasing @values in the mixed tim regime. In the hydrodynamic
regime, f increases with increasing +.
 19

loundar y Hydrodynamic
ubrication

Bearing characteristic number Y

Fig. 3. The Stribeck curve for a journal bearing.

Bearing characteristic number Y X107

Fig. 4. Relationship between coefficient of friction and bearing characteristic number.

In the boundary lubrication regime, the hydrodynamic effect of the lubricant is

completely absent and the load is transmitted by shaft and bearing metals in direct
contact. In the mixed lubrication regime the load is shared by the contacting materials
and the fluid film. The load carried by the fluid film increases with increasing I) value
in this regime. In the hydrodynamic regime, the mating surfaces are completely separated
by the fluid film which completely carries the load.
A trial test was conducted to obtain the Stribeck curve for the test specimens.
In this test, a load of 375 N was applied to the bearing and it was run for 2 h at a
speed of 50 rev min-‘. At the end of this period, the speed was varied keeping the
load on the bearing constant. At each speed the specimen was run for 5 min and at
the end the frictional force and the bearing temperature were measured. FTOXII this
frictional force, f was calculated by dividing the frictional force by the normal load
on the bearing. The coefficient of friction was then plotted against the bearing
characteristic number. A typical Stribeck curve obtained for a Sursulf-treated plain
carbon steel pair is illustrated in Fig. 4.
20

From the Stribeck curve, the test values of w and p required for the whole range
of experiments were selected such that the operating conditions fell in the mixed film
region. With these loads and speeds, the experiments were conducted in two phases,
A and B. In phase A, the speed was kept constant and the load varied. In phase B,
the load was kept constant and the speed varied. Speed and load values in the two
phases of the experiments are given below:
phase A, load (N), 375, 571, 767, 963, 1160; speed (rev min-‘), 50;
phase B, load (N), 375; speed (rev min-‘), 10, 20, 40, 50, 70.
In determining the friction and wear characteristics, the following procedure was
adopted. The test specimens were assembled after thorough cleaning and run under
no load at the required test speed for 5 min. Then the experimental load was gradually
applied and the test nm for 4 h at full load. Weight loss and surface roughness
measurements were taken by dismantling the arrangement at 1 h intervals. The roughness
was measured using a perthometer. Lubricating oil was plain mineral oil-Cabol 32
(77=45 x 10d3 N s mm2 at 30 “C) in all cases.
The experiments were repeated and the results were found to vary within + 10%.
The average values were used for plotting.

3. Results and discussion

3.1. Friction characteristics

It is desirable that a good bearing material exhibits a minimum coefficient of
friction over a range of operational regimes. The frictional behaviour of a few sliding
pairs as obtained from experimental results is illustrated in Fig. 5. In this figure, the
coefficient of friction after 1 h of operation is plotted as a function of load and speed
in (a) and (b) respectively for different sliding pairs as described in Section 2.
The plot shown in Fig. 5(a) indicates that over the load range tested, the conventional
pair, i.e. bronze sliding against untreated steel, exhibited a slight increase in the
coefficient of friction with increasing load. Bronze sliding over Sursulf-treated steel
showed a similar trend. For steels sliding over steels, the coefficient of friction showed
an increase with increasing load (increasing metal-to-metal contact) up to some load
and then it reduced. This increase in coefficient of friction may be due to the increase
in resistance to motion caused by the increase in elastic deformation of the surface
asperities in contact. A further increase in load may induce plastic deformation of
the material below the peaks, leading to a decrease in the coefficient of friction.
Similar trends were also noticed in Sursulf-treated and Tu&ided steel specimens,
with the difference that the coefficient of friction again increased in the case of
Tufftrided steel journals paired with Ttitrided steel bushes. Although a similar change
was observed in the case of TufRrided grey iron journals paired with TuEtrided grey
iron bushes, and Sursulf-treated grey iron journals against Sursulf-treated grey iron
bushes, it was not significant in these cases. Unlike the other pairs, in the load range
tested, grey iron journals against grey iron bushes showed a decrease in the coefficient
of friction with increasing load and then it increased. As seen in the plot in Fig. 5(b),
the coefficient of friction in general showed a decreasing trend with increasing speed
up to some speed and increased thereafter. This indicates that mixed lubrication
conditions only existed up to a certain speed under a given load.
From the plots of Fig. 5, it is also noticed that the conventional bearing pairs
exhibited good friction characteristics with a low coefficient of friction over a considerable
load range. The frictional behaviour of bronze sliding against Sursulf-treated steel was
 21

G
5
- 0.04
.s
t
;E
6
_ 0.02
E
._
.u
z
:
u 0
200 LOO 600 800 100 0

(4 Load F,N Speed n, rev. rnin-’

Fig. 5. Variation in f (after 1 h of operation) with (a) bearing load (speed 50 rev min-‘) and
(b) journal speed (load 375 N).

Journal Bush

Sursulf-treated steel Sursulf-treated steel

Tufftrided steel Tufftrided steel
Bronze Sursulf-treated steel
Tuiftrided grey iron Tufftrided grey iron
Sursulf-treated grey iron Sursulf-treated grey iron
Bronze Steel
Brass Sursulf-treated steel
Brass Steel
Steel Steel
Grey iron Grey iron

better than that of the conventional pair. Tufftrided and Sursulf-treated grey iron
pairs as well as the untreated grey iron pair showed comparable frictional behaviour
over the load range with respect to the inventions pair. All the other pairs tested
showed inferior frictional behaviour. Although brass sliding against plain carbon steel
is another conventional pair, in a single test it showed a very high frictional coefficient
in relation to other pairs but the friction value was considerably lowered when brass
was slid against Sursulf-treated steel.
The individual sliding pairs exhibited their lowest coefficient of friction at different
load and speed combinations. These details are given in Table 4.
3.2. Wear characteristics
It is desirable that a good bearing material has a very low specific wear rate.
The specific wear rate k is calculated as the wear loss in cubic millimetres per unit
load and unit sliding distance of the bearing [S]. The wear rate curves as obtained
from experiments for different material combinations are shown in Fig. 6(a) and 6(b)
for the bushes and journals respectively. These plots correspond to wear rates at the
end of 4 h of operation.
22

TABLE 4
Details of load and speed values that resulted in lowest coefficient of friction

Material pair Lowest Load Speed

coefficient (N) (rev min-‘)
Journal Bush of friction

Bronze Steel 0.0125 375 50

Bronze Sursulf-treated 0.0025 375 50
steel
Steel Steel 0.0025 375 50
Grey iron Grey iron 0.0025 571 50
Tufftrided Tufftrided 0.0025 375 70
grey iron grey iron
Sursulf-treated Sursulf-treated 0.0210 375 50
steel steel
Sursulf-treated Sursulf-treated 0.0025 963 50
grey iron grey iron
Tufftrided Tufftrided 0.0025 375 50
steel steel

400 800 1200

(4 Load on bearing F,N @I Load on bearing F,N

Fig. 6. Variation in k with bearing load (speed 50 rev min-‘) in (a) bush and (b) journal.

Symbol Journal Bush

A Sursulf-treated steel Sursulf-treated steel

0 TutTtrided grey iron Tufkided grey iron
V Sursulf-treated grey iron Sursulf-treated grey iron
0 Bronze Steel
0 Bronze Sursulf-treated steel
X TutRrided Tufftrided
steel steel
 23

From Fig. 6 it is seen that the specific wear rates decreased with increased loads
with the exception that the Sursulf-treated steel bushes showed a slightly increased
wear rate at medium loads. Also, the wear rates of bushes are seen to be slightly
lower than those of the mating journals which might be due to the better initial surface
finish of the bushes. Referring to Fig. 6(b), it is seen that the specific wear rate of
the bronze journal was relatively low. The specific wear rate of the mating steel bush
was also extremely low. The Sursulf-treated steel journals paired with Sursulf-treated
steel bushes exhibited lower journal wear rates and higher bush wear rates compared
with corresponding values for the conventional bearing paris, i.e. bronze journals against
steel bushes. Sursulf-treated grey iron journals against Sursulf-treated grey iron bushes
exhibited comparable journal wear rates but slightly higher bush wear rates in relation
to the above bearings.
The specific wear rates of Tufftrided steel journals were close to those obtained
from the conventional bronze journals but the wear rates of mating Tufftrided steel
bushes were high compared with the wear rates of the corresponding steel bushes in
the conventional pair. TufItrided grey iron journals against Tufftrided grey iron bushes
exhibited poor wear characteristics in relation to the other bushes tested. The wear
rates of the journals and the bushes of the conventional pair showed an increase
when the plain steel bushes were Sursulf treated. Slight metal transfer was also noticed
in this case.
Figure 7(a) and 7(b) show the plots of specific wear rates as a function of speed
for the bush and journal respectively. In general the specific wear rates decreased
with increasing speed. Comparing these plots with those in Fig. 6 it is observed that
the specific wear rates at low speeds (in spite of the fact that the load was low) were
very much higher than those at high loads. It should be noted that the coefficient of
friction was also high at low speeds.
When plain carbon steel was sliding over plain carbon steel, or when grey iron
was sliding over grey iron, considerable metal transfer took place as established by
scanning electron microscopy and hence they have not been included in the figure.
Details of the load and speed values at which the lowest specific wear rates k
were observed for different bearing materials are given in Table 5.

3.3. Comparison with standard bearing materials

The friction and specific wear rate values of different standard bearing materials
as obtained by different investigators [8, 91 are given in Table 6, along with the results
of the present investigation to facilitate comparison. The specific wear rate values
given in the table from the present investigation are the maximum values obtained
during the 4 h test. Specific wear rates of negligible magnitude were measured under
hydrodynamic lubrication conditions. Normally, it should be expected that hydrodynamic
conditions lead to zero wear. The values indicated are caused by periodic starting
and stopping of the bearing test rig for inspection of bearing surface topography,
which led to metal-to-metal contact. The values given by Begelinger and De Gee [S]
correspond to tests conducted under boundary lubrication conditions and these values
could be expected to be a little less if tested under mixed lubrication conditions. The
values given by Habig et al. [9] correspond to tests conducted under near boundary
lubrication conditions (mixed lubrication conditions) and hence they can be directly
compared with the results of the present investigation.
From the tabulated values it can be seen that the friction and wear rate values
of treated grey iron and steel bearings compare favourably with those of standard
24

Bush
12- i

0 20 40 60 0 20 40 60
Journal speed n, rev rnin-’ Journal speed n, rev. mm-’
(4 PI
Fig. 7. Variation in k with journal speed (load 375 N) in (a) bush and (b) journal.

Symbol Journal Bush

SursuKtreated steel Sursulf-treated steel

TutTtrided steel Tufftrided steel
Tufftrided grey iron Tufftrided grey iron
Sursulf-treated grey iron Sursulf-treated grey iron
Bronze Steel
Brass Sursulf-treated steel
Brass Steel
Bronze Sursulf-treated steel

TABLE 5
Details of load and speed values that gave lowest specific wear rates of journals

Material pair Lowest Load Speed

value of k WI (rev min-r)
Journal Bush (mm3 N-r m-r)

Bronze Steel 0.8 x lO+ 767 50

TutBrided TutTtrided 1.5 x 10-6 963 50
grey iron grey iron
Sursulf-treated Sursulf-treated 0.4x 10-6 963 50
steel steel
Sursulf-treated Sursulf-treated 0.5 x 10-6 571 50
grey iron grey iron
TufBrided Tutftrided 1.2x 10-6 963 50
steel steel
TABLE 6
Coefficient of friction and specific wear rates for different materials

Reference Materials tested Test Test rig Test results

conditions used
Coefficient of Specific wear rate
friction kx106

8 AlSn20, AlSn6, SnSb12 Boundary Pin and Not given 0.5-150

CuAl13Fe4, CuPb20 lubrication ring
ZnAl38Cu2, PbSblSSnlO 0.3-85.1
CuSn6(P), CuSn6
9 PbSbl4Sn9CuAs, AlSn20 Near Different 0.07-0.18 0.1-36
SnSb8Cu4Cd, CuSn8 boundary types
CuPblOSnlO, CuSnl2Pb lubrication
Present Sursulf-treated plain Mixed Bearing 0.02-0.17 0.4-9
investigation carbon steel lubrication test rig
(0.17% carbon Hydrodynamic 0.001-0.003 0.004-0.007
VS. 0.25% carbon) lubrication
Ttitrided plain carbon Mixed Bearing 0.001-0.12 0.4-14
steel (0.17% lubrication test rig
carbon vs. 0.25% Hydrodynamic 0.001~).003 -
carbon) lubrication
Sursulf-treated Mixed Bearing 0.005-0.04 1.5-10
grey iron lubrication test rig
Hydrodynamic 0.001~.003 0.009-0.011
lubrication

Tufftrided Mixed Bearing 0.0054.06 0.8-22

grey iron lubrication test rig
Hydrodynamic 0.001-0.003 -
lubrication
26

bearing materials. The wear rates go up to 36x 10M6 mm3 N-’ m-l for standard
bearing materials as given by Habig et al. The present investigation shows that the
maximum specific wear rate for the Sursulf-treated steel pair was 9X 10P6 mm3 N-’
m-l and that for the Tufftrided steel pair was 14x 10W6 mm3 N-l m-r. The maximum
specific wear rates for the Sursulf-treated grey iron pair and Tufftrided grey iron pair
were 10X lop6 and 22x 10m6 mm3 N-’ m-l respectively which are also less than the
corresponding values for standard bearing materials. The coefficient of friction for
standard bearing materials has been given as varying from 0.07 to 0.18. The present
investigation shows that the coefficient of friction for the tested treated bearings ranges
between 0.001 and 0.17.
The wear coefficient is calculated by multiplying the specific wear rate by the
hardness of the softer of the two sliding materials [lo]. The wear coefficient values
for the treated bearings calculated using this relation vary from 1.412 X lo-” to 77x 10m6
under mixed lubrication conditions. These values fall under moderate wear conditions
[lo, 111. The results indicate that the friction and wear characteristics of treated
bearing materials, especially Sursulf-treated plain carbon steels, are satisfactory.

3.4. Material rating

The following properties are usually considered in the selection of materials for
use in lubricated journal bearings [S]: (i) mechanical properties, (ii) thermal properties,
(iii) corrosion resistance, (iv) friction and wear properties, (v) cost, (vi) behaviour
under conditions of abrasion, (vii) resistance under cavitation, (viii) resistance under
electrical pitting.
The criteria (i)-(v) are generally applicable and thus must be considered each
time a material is selected for a new application. Treated bearings, especially steel
bearings, have good mechanical properties such as yield strength. Their thermal properties
such as conductivity are also reasonably good. As a result of the treatment they have
good corrosion resistance [4, 73 which is an added advantage. The friction and wear
resistance are comparable with those of conventional bearing materials such as brass
and bronze. The relative costs of conventional and treated bearings are compared in
Table 7. The relative unit cost data for different materials including the treatment
cost were rationalized with respect to the recent prevailing cost per unit kilogram of
bronze in India. From the table it is clearly seen that the treated bearing materials
are 2-3 times cheaper than the conventional bearing materials, namely brass and
bronze. However, it should be pointed out that treated steels have inferior conformability
and embeddability characteristics compared with conventional bearing materials. This
aspect should be considered when selecting treated steels as substitutes for conventional
bearing materials.

TABLE 7
Relative unit cost comparison of materials

Materials Relative Relative Relative

material treatment total cost
cost per kg cost per kg per kg

Brass 0.57 Nil 0.57

Bronze 1.00 Nil 1.00
Treated steel 0.14 0.14 0.28
Treated grey iron 0.11 0.14 0.25
 27

3.5. Mechanism of wear process

The worn surfaces of treated bearings were viewed under a scanning electron
microscope and found to have ploughing tracks (Fig. 8). For Sursulf-treated steel
bearings, we noticed a thin homogeneous layer at the top of the compound zone.
This homogeneous zone had peeled off (Fig. 9), exposing the ductile porous zone.
Subsequently, the ductile porous zone had worn off by plastic shear flow (Fig. 10)
combined with spalling by fatigue (Fig. 11).
For TufBrided steel bearings, subsurface cracks could be noticed (Fig. 12) which,
when developed, led to the generation of flake-like wear debris. Severe plastic deformation
of the surface and subsurface cracks leading to the generation of flake-like sheets of
wear debris suggest that the wear mechanism is delamination fYPe, as proposed by
Suh [12]. In the test situations involving conventional untreated bearings, i.e. brass or
bronze journals paired against steel bushes, metal transfer was observed. A comparative
study of the above mechanisms of failure with that of conventional bearing materials
has been presented elsewhere [13-151.

Fig. 8. Scanning electron micrographs of wear tracks showing ploughing marks on treated bearings:
(a) Sursulf-treated steel (original magnification x780); (b) TuEtrided steel (original magnification
x 1000).

Fig. 9. Scanning electron micrograph showing peeling off of the thin superficial layer of Sursulf-
treated steel bearings (original magnification x 2250).
Fig. 10. Scanning electron micrograph of a cut section of Sursulf-treated steel journal showing
the formation of wear particles by plastic shear flow (original magnification x750).
Fig. 11. Scanning electron micrograph showing the spalled spot on the Sursulf-treated steel
journal (original magnification X 960).

Fig. 12. Scanning electron micrograph showing subsurface cracks in Tufftrided steel bearings
(original magnification X 1000).

4. Conclusions

The results of an experimental investigation of the friction and wear characteristics

of Sursulf-treated and Tufftrided plain carbon steels of two different compositions and
a particular grade of grey iron, used as bearing materials operating under mixed
lubrication conditions, indicate that these bearings can perform satisfactorily in service.
Of the treated bearings Sursulf-treated plain carbon steel bearings appear to be the
best, substituting for bronze and giving an economic advantage.
Both Sursulf-treated and Tufftride-treated bearings exhibited maximum friction
force during the initial period which decreased with time. For Sursulf-treated steel
bearings the friction force reduced at a faster rate with increased load and the coefficient
of friction measured after 1 h of operation decreased with increasing load. However,
for Tufftrided steel bearings it was observed that the friction force increased with
load and decreased with speed; the coefficient of friction increased with increasing
load and decreased with increasing speed. The wear characteristics of treated bearings
indicate that the wear increased with increasing sliding distance and load.
The wear mechanism of Sursulf-treated steel bearings may be explained by peeling
off of the superficial thin homogeneous zone followed by plastic shear flow of the
porous zone combined with localized spalling by fatigue. In the case of Tufftrided
steel bearings, the mechanism of wear was one of delamination, leading to the generation
of flake-like wear debris.

References

A. W. J. De Gee, Selection of materials for lubricated journal bearings, Wear, 36 (1976)

38-61.
Sursulf - A nitriding treatment in a non-polluting salt bath accelerated with sulphur,
Information Bulletin de ZSyaVomehmique et Fmttement, France, July 1980.
A. Gaucher, G. Guilhot and C. Amsahem, The Sursulf process - a surface treatment for
iron and steel in a non-toxic salt bath which improves the resistance to scuffing wear and
fatigue, Tribol. Znt., 9 (1976) 131-137.
J. C. Gregory, Sursulf - improving the resistance of ferrous materials to sculling, wear,
fretting and fatigue by treatment in a non-toxic salt bath, Heat Treat. Met., 2 (1975) 5.5-64.
 29

5 Degussa, Tufftride manuals and information, Durferrit Technical Information, D-6450 Hanau,
FRG.
6 P. Astley, Tufftride - a new development reduces treatment costs and process toxicity, Heat
Treat. Met., 2 (1975) 51-54.
7 T. Oda and T. Daikoku, Fundamental investigation on sulphurising and Tufftride processes,
Mitsubirhe Tech. Bull., 42 (1967) l-9.
8 A. Begelinger and A. W. J. De Gee, Wear in lubricated journal bearings, J. Lubr. Technol.,
100 (1978) 104-109.
9 K. H. Habig, E. Broszeit and A. W. J. De Gee, Friction and wear tests on metallic bearing
materials for oil lubricated bearings, Wear, 69 (1981) 43-54.
10 E. Rabinowicz, Friction and Wear of Materials, Wiley, New York, 1965.
11 E. Rabinowicz, The least wear, Wear, 100 (1984) 533-541.
12 N. P. Suh, The delamination theory of wear, Wear, 25 (1973) 111-124.
13 R. Marappan, Ph.D. Thati, Indian Institute of Technology, Madras, 1987.
14 R. Marappan and A. Ramamohanrao, Performance studies on Sursulf-treated plain carbon
steel journal bearings under mixed lubrication conditions, Wear, 121 (1988) 239-253.
15 R. Marappan and A. Ramamohanarao, Performance studies on Tufftried plain carbon steel
journal bearings under mixed lubrication conditions, Tribal. Int., 21 (1988) 155-162.