Materials and Design 32 (2011) 1888–1894

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Fatigue behaviour of friction welded medium carbon steel and austenitic

stainless steel dissimilar joints
R. Paventhan a,1, P.R. Lakshminarayanan a, V. Balasubramanian b,⇑
a
Department of Manufacturing Engineering, Annamalai University, Annamalainagar 608 002, Chidambaram, Tamil Nadu, India
b
Center for Materials Joining Research (CEMAJOR), Department of Manufacturing Engineering, Annamalai University, Annamalainagar 608 002, Tamil Nadu, India

a r t i c l e i n f o a b s t r a c t

Article history: This paper reports the fatigue behaviour of friction welded medium carbon steel–austenitic stainless
Received 27 May 2010 steel (MCS–ASS) dissimilar joints. Commercial grade medium carbon steel rods of 12 mm diameter
Accepted 4 December 2010 and AISI 304 grade austenitic stainless steel rods of 12 mm diameter were used to fabricate the joints.
Available online 13 December 2010
A constant speed, continuous drive friction welding machine was used to fabricate the joints. Fatigue life
of the joints was evaluated conducting the experiments using rotary bending fatigue testing machine
Keywords: (R = 1). Applied stress vs. number of cycles to failure (S–N) curve was plotted for unnotched and
A. Ferrous metals and alloys
notched specimens. Basquin constants, fatigue strength, fatigue notch factor and notch sensitivity factor
D. Welding
E. Fatigue
were evaluated for the dissimilar joints. Fatigue strength of the joints is correlated with microstructure,
microhardness and tensile properties of the joints.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction
Joints of dissimilar metal combinations are employed in differ-
ent applications requiring certain special combination of proper-
ties as well as to save cost incurred towards costly and scarce
materials [1]. Conventional fusion welding of many such dissimilar
metal combinations is not feasible owing to the formation of brittle
and low melting intermetallics due to metallurgical incompatibil-
ity, wide difference in melting point, thermal mismatch, etc. So-
lid-state welding processes that limit extent of intermixing are
generally employed in such situations. Friction welding is one such
solid-state welding process widely employed in such situations [2].
Dobrovidov [3] investigated selection of optimum conditions
for the friction welding of high speed steel to carbon steel. Ishib-
ashi et al. [4] chose stainless steel and high speed steel as represen-
tative materials with an appreciably difficult weldability, and their
adequate welding conditions were established. The distributions of
the alloying elements at and near the weld interface with sufficient
strength were analysed using X-ray microanalyser. Sahin [5] has
analysed the variations in hardness and microstructure at the
interfaces of friction welded steel joints. While using austenitic
stainless steel, negative metallurgical changes like delta ferrite
formation and chromium carbide precipitation between grain
⇑ Corresponding author. Tel.: +91 4144 239734 (O), +91 4144 241147 (R), mobile:
+91 9443412249; fax: +91 4144 238080/238275.
E-mail addresses: mr_paventhan@yahoo.com (R. Paventhan), prlnau_1992@
yahoo.co.in (P.R. Lakshminarayanan), balasubramanian.v.2784@annamalaiuniversity.
ac.in, visvabalu@yahoo.com (V. Balasubramanian).
1
Tel.: +91 09944013323; fax: +91 4144 238080/238275.
boundaries took place during fusion welding. These changes are
eliminated by friction welding. The effect of friction time on the
fully plastically deformed region in the vicinity of the weld has
been investigated by Sathiya et al. [6].
Ananthapadmanaban et al. [7] have reported the effect of fric-
tion welding parameters on tensile properties of steel. Satyanara-
yana et al. [8] joined austenitic–ferritic stainless steel (AISI 304
and AISI 430) using continuous drive friction welding and investi-
gated optimum parameters, microstructures-mechanical property
and fracture behaviours. Yılmaz [9] investigated variations in hard-
ness and microstructures in the welding zone of friction welded
dissimilar materials The effect of friction pressure on the proper-
ties of hot rolled iron based super alloy has been investigated by
Afes et al. [10]. Meshram et al. [11] investigated the influence of
interaction time on microstructure and tensile properties of the
friction welding of dissimilar metal combinations.
Hiizu et al. [12] evaluated tensile and fatigue behaviour of fric-
tion welded SUS304 stainless steel joints. They found that the fati-
gue strength of sound joints was higher than that of SUS304 base
material. This is probably because the material near the weld inter-
face was hardened due to the work hardening. Lee et al. [13] inves-
tigated the fatigue crack propagation characteristics of dissimilar
friction welding zone of two kinds of heat-resisting steels, STR3
and STR35, commonly used as valve materials for vehicles. A small
circular artificial defect was machined to induce fatigue crack on
weld interface, heat affected zone and base metal to compare fati-
gue lives according to the notch position. Sahin [14] investigated
fatigue properties of friction welded austenitic stainless steel (AISI
304) joints. The fatigue strength of welded joints is close to those of
base metal.

0261-3069/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2010.12.011
 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894 1889

Many of the structural components in pressure vessels, trans-
port vehicles, earthmoving equipment, spacecraft, etc. are made
of welded joints. The dissimilar joints are the most common ones
in the fabrication and construction of many structures. The wide
application of dissimilar joints in various structures, including off-
shore, nuclear, and spacecraft, gives large scope for the researchers
to analyze the behaviour under different types of loading condi-
tions. Failure analysis of the weldments indicated that fatigue
alone is to be considered to account for most of the disruptive fail-
ures [15].
From the literature review, it is understood that the most of the
published information [3–11] on friction welded dissimilar joints
focused on microstructural characteristics, microhardness varia-
tions, phase formation at the interface and tensile properties. But
the published information on fatigue characteristics of friction
welded dissimilar joints is very scanty. Hence, the present investi-
gation was carried out to study the fatigue behaviour of friction
welded dissimilar joints of medium carbon steels and austenitic
stainless steel joints and the results are presented in this paper.
parameters were optimized to attain maximum tensile strength
using response surface methodology [16]. The optimized friction
welding parameters, presented in Table 2, were used to fabricate
the joints. Two different fatigue specimens were prepared to eval-
uate the fatigue properties. Unnotched (smooth) specimens were
prepared (as shown in Fig. 1b) to evaluate fatigue strength and Bas-
quin constants. Notched specimens were prepared (as shown in
Fig. 1c) to evaluate the fatigue notch factor and notch sensitivity
factor. The rotary bending fatigue testing machine (0.5 kN capac-
ity) was used to evaluate fatigue behaviour of the base metals
and friction welded joints. Fatigue experiments were conducted
at five different stress levels under completely reversed bending
load conditions, where mean stress was zero and stress ratio was
1. At each conditions three specimens were tested and the aver-
age of three fatigue life (number of cycles to failure) values was
used to draw the S (stress)–N (fatigue life) curves.
Tensile specimens were prepared as shown in Fig. 1b to evalu-
ate yield strength, tensile strength and elongation. Tensile test was
carried out in 100 kN, electro-mechanical controlled Universal
Testing Machine. The specimen was loaded at the rate of 1.5 kN/

2. Materials and methods

Table 2
The base materials, AISI 1040 grade medium carbon steel (MCS) Optimised friction welding parameters used to
and AISI 304 austenitic stainless steels (ASS), used in this investiga- fabricate the joints.

tion were cylindrical rods of 12 mm diameter and 75 mm length. Rotational speed (rpm) 1500
The chemical composition of the base materials is presented in Friction pressure (MPa) 90
Table 1. Hydraulic controlled, continuous drive friction welding Forging pressure (MPa) 90
Friction time (s) 2
machine (20 kN capacity) was used to fabricate the joints. The joint Forging time (s) 2
dimensions are shown in Fig. 1a. The friction welding process

Table 1
Chemical composition of base metals.

Materials C Si Mn Ni Cr Cu Fe
Medium carbon steel (MCS) 0.18 0.11 0.39 0.016 0.02 0.02 99.10
Austenitic stainless steel (ASS) 0.09 0.29 1.42 8.413 14.08 1.24 74.53

(a)

(b)

(c)

All the dimensions are in mm

Fig. 1. Dimensions of the specimens. (a) Dimensions of the base materials before welding. (b) Dimensions of the unnotched fatigue/tensile specimen. (c) Dimensions of the
notched fatigue/tensile specimen.
1890 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894
min as per the ASTM specifications, so that tensile specimen
undergoes uniform deformation. The specimen finally fails after
the necking and the load vs. displacement was recorded. The
0.2% offset yield strength was derived from the diagram. Vicker’s
microhardness testing machine (Make: Zwick and MODEL: 3212)
was employed for measuring the hardness across the joint with
0.05 kg load. Fig. 2a–e displays the photographs of friction welded
joints, fatigue and tensile specimens. Microstructural examination
was carried out using a light optical microscope (Make: MEIJI,
Japan, Model: ML7100) incorporated with an image analyzing soft-
ware (Clemex-Vision). The specimen for microstructural analysis
was sectioned to the required size from the joint comprising weld
metal, HAZ and base metal regions and were polished using differ-
ent grades of emery papers. Final polishing was done using the dia-
mond compound (1 lm particle size) in the disc-polishing
machine. Specimen was etched with 2% Nital solution in the mild
steel side and oxalic acid in the stainless steel side to reveal the
microstructure. Microstructural examination was also carried out
using a Scanning Electron Microscope (SEM). X-ray Diffraction
Analysis was also done to identify the phases present at the
interface
3. Result and discussion
The fatigue test results obtained from the unnotched specimens
are plotted in the form of S–N curves as shown in Fig. 3a. The S–N
curve in the high cycle fatigue region is usually represented by the
Basquin equation [17]
(a)
(b)
(d)
Fig. 2. Photographs of welded joints and fatigue specimens. (a) Friction welded joints. (b) Unnotched fatigue specimen (before testing). (c) Notched fatigue specimen (before
testing). (d) Unnotched fatigue specimen (after testing). (e) Notched fatigue specimen (after testing).
Sn N ¼ A ð1Þ
where ‘S’ is the stress amplitude, ‘N’ is the number of cycles to fail-
ure and ‘n’ and ‘A’ are empirical constants. Each S–N curves shown
in Fig. 3a can be represented by the above equation. From those
equations, the empirical constants ‘n’ (slope of the curve) and ‘A’
(intercept of the curve) were evaluated and they are presented in
Table 3. Comparing the fatigue strengths of welded joints subjected
to similar loading, it is convenient to express fatigue strength in
terms of the stresses corresponding to particular lives, for example,
105, 106 and 107 cycles on the mean S–N curve. The choice of refer-
ence life is quite arbitrary. Traditionally, 2  106 cycles was used,
and indeed some design codes refer to their S–N curves in terms
of the corresponding stress range [18]. For these reasons, in this
investigation, fatigue strength of welded joints at 2  106 cycles
was taken as a basis for comparison. The stress corresponding to
2  106 cycles was taken as an indication of the fatigue strength
and it was evaluated for the joints and base metals and they are pre-
sented in Table 3.
The fatigue strength of medium carbon steel is 100 MPa and the
fatigue strength of austenitic stainless steel is 120 MPa. But the fa-
tigue strength of the dissimilar joints is 70 MPa which is 30% lower
than medium carbon steel and 40% lower than austenitic stainless
steel. Slope of the S–N curve (Basquin constant) is also a fatigue
parameter, which is lower for dissimilar joints compared to base
metal. If the slope of the curve is higher, then the fatigue life will
be higher and vice versa. Slope of the S–N curve for austenitic
stainless steel is higher and hence the fatigue life also higher com-
pared to medium carbon steel and the dissimilar joints.
(c)
(e)
 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894 1891

Fig. 3. Fatigue test results. (a) Fatigue life of unnotched specimens. (b) Fatigue life of notched specimens. (c) Fatigue life of the joints.

Table 3
Fatigue properties of welded joints.

Materials Slope of the Intercept of the S–N Fatigue strength of Fatigue strength of Fatigue notch Notch sensitivity
S–N curve (n) curve (A) unnotched specimens notched specimens factor (Kf) factor (q)
at 2  106 cycles (MPa) at 2  106 cycles (MPa)
MCS 0.42 34,978 100 60 1.66 0.43
ASS 0.50 143,087 120 80 1.50 0.33
MCS–ASS joint 0.39 17,301 70 30 2.33 0.88

Table 4
Mechanical properties of the base metals and friction welded joints.

Materials Yield strength Tensile strength Elongation Notch tensile strength Notch strength ratio Impact toughness @RT
(MPa) (MPa) (%) (MPa) (NSR) (J)
MCS 300 350 25 425 1.21 30
ASS 400 460 30 575 1.25 50
MCS–ASS 330 410 15 450 1.09 20
joint

Hasalik and Ozdemir [19] investigated the fatigue behaviour of
friction welded AISI 304 stainless steel and AISI 4340 steel and they
found that the fatigue endurance decreased with increasing the
rotational speed as a result of formation of chromium carbide
precipitation at the weld interface. The carbides at the grain
boundaries provide preferential sites for cavity nucleation in
austenitic stainless steels under fatigue conditions thereby reduc-
ing the fatigue life. Sahin [20] evaluated the fatigue behaviour of
friction welded high speed steel and medium carbon steel and they
found that the fatigue strengths of welded joints are slightly lower
than those of medium carbon steel parts having the weakest of the
materials .
The effect of notches on fatigue strength was determined by
comparing the S–N curves of notched and unnotched specimens
as shown in Fig. 3c. The data for notched specimens are usually
plotted in terms of nominal stress based on the net section of the
specimen. The effectiveness of the notch in decreasing the fatigue
limit is expressed by the fatigue strength reduction factor or fati-
gue notch factor, Kf. The fatigue notch factor was evaluated using
the following expression [21] and they are given in Table 3.
K f ¼ Fatigue strength of unnotched specimen=
fatigue strength of notched specimen ð2Þ
The notch sensitivity of a material in fatigue is expressed by a
notch sensitivity factor ‘q’ and it can be evaluated using the follow-
ing expression [21]:
q ¼ ðK f  1Þ=ðK t  1Þ ð3Þ
where ‘K t0 ’ is the theoretical stress concentration factor and is the
ratio of maximum stress to nominal stress. Using the above expres-
sion fatigue notch sensitivity factor ‘q’ was evaluated and they are
presented in Table 3.
The fatigue notch factor is 1.66 for medium carbon steel and
1.50 for austenitic stainless steel but it is 2.33 for the dissimilar
joint. The fatigue notch factor of the dissimilar joint is 70% higher
compared to medium carbon steel and 50% higher compared to
austenitic stainless steel. If the fatigue notch factor is higher, then
the fatigue life will be lower and hence the dissimilar joints are
showing very low fatigue strength with the notches. Similar trend
was observed in notch sensitivity factor also since these values are
1892 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894

derived from fatigue notch factor values. From the notch fatigue
test results, it is inferred that the dissimilar joints are more sensi-
tive to notches than the base materials. In the open literature, no
information is available on the effect of notches on fatigue strength
of friction welded dissimilar joints.
Fig. 4. Microhardness across the joint.
The mechanical properties of the base metals and the dissimilar
joints are presented in Table 4. From the results, it is inferred that
the tensile strength, elongation and Charpy impact toughness of
austenitic stainless steel base metal are higher compared to med-
ium carbon steel. This may be the one of the reasons for superior
fatigue performance of austenitic stainless steel compared to med-
ium carbon steel. The tensile strength, elongation and Charpy im-
pact toughness of dissimilar joints are lower compared to the
base metals. This may be one of the reasons for inferior fatigue per-
formance of the dissimilar joints. Microhardness was measured
across the weld region and the values are plotted as shown in
Fig. 4. The microhardness is maximum at the interface and this
may be due to the formation of brittle intermetallics and it is one
of the reasons for lower tensile strength, elongation, impact tough-
ness and fatigue strength of the dissimilar joints. However, Anan-
thapadmanaban et al. [7] reported that the hardness values in
the weld zone of mild steel to stainless steel are not too high. They
opined that the temperatures reached during friction welding
could have also contributed to some softening of the weld and
the higher hardness at the stainless steel, HAZ side is probably
due to the presence of chromium carbide.
Fig. 5a–f displays the optical micrographs of base metals and
welded joint. Fig. 5a shows the microstructure of medium carbon
steel base metal. It contains fine ferrite and small amount of pearl-
ite grains. Fig. 5b shows the microstructure of austenitic stainless
steel base metal. It contains full of austenitic grains. Fig. 5c and d
show the weld cross-section of dissimilar joints of MCS–ASS at

(a) (b)

(c) (d)

(e) (f)
Fig. 5. Optical micrographs of dissimilar joint. (a) Optical micrograph of MCS. (b) Optical micrograph of ASS. (c) Macrostructure of the joint. (d) Optical micrograph of
interface region at 50. (e) Optical micrograph of HAZ region at MCS side. (f) Optical micrograph of HAZ region at ASS region.
 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894
Fig. 6. SEM micrographs of base metal and weld region. (a) SEM micrograph of medium carbon steel. (b) SEM micrograph of austenitic stainless steel. (c) SEM micrograph of
friction welded cross-section. (d) XRD results.
two different magnifications. The formation of weld interface re-
gion is not so clear in these optical micrographs. However, highly
deformed region, due to thermomechanical action, is observed on
both the base metals nearer to the joint line and they are displayed
in Fig. 5e and f. Behnken and Hauk [22] investigated the micro-
residual stresses caused by deformation, heat or their combination
during friction welding of two phase materials. They found that the
materials at the very weld plane are recrystallized and in the de-
formed region, which is within the heat affected zone the fibres
are bent due to the friction welding process.
To reveal the formation of weld interface region between MCS
and ASS, SEM was used. The microstructure captured under SEM
is displayed in Figs. 6a–d. The formation of weld interface region
is evident from Fig. 6c. The interface region was further analysed
using X-ray Diffraction (XRD) and it is confirmed that the presence
of austenite, cementite, Cr23C6 and Cr7C3 (Fig. 6d). The formation of
intermetallics such as Cr23 C6 and Cr7C3 is responsible for higher
hardness at the interface (Fig. 4). The presences of these brittle
intermetallics at the interface are responsible for lower fatigue,
tensile, impact strength and higher hardness.
4. Conclusions
From this investigation following important conclusions are
derived.
(1) Fatigue strength of friction welded dissimilar joints of med-
ium carbon steel and austenitic stainless is lower compared
to the base metals. It is 30% lower than medium carbon steel
and 40% lower than austenitic stainless steel.
(2) Fatigue notch factor of friction welded dissimilar joints of
medium carbon steel and austenitic stainless steel are
higher compared to base metals. It is 50% higher than
austenitic stainless steel and 70% higher than medium
carbon steel.
1893
(3) The formation of intermetallics such as Cr23C6 and Cr7 C3 at
the weld interface are responsible for higher hardness and
lower tensile, impact and fatigue strength of the friction
welded MCS–ASS dissimilar joints.
Acknowledgements
The authors are grateful to the Department of Manufacturing
Engineering, Annamalai University, for extending the facilities of
Material Testing Laboratory to carry out this investigation. The
authors also wish to record their sincere thanks to the University
Grant Commission (UGC), New Delhi, for the Research Fellowship
in Science for Meritorious Students Scheme.
References
[1] Ozdemir N. Investigation of mechanical properties of friction – welded joints
between AISI 304 L and AISI 4340 steel as a function of rotational speed. Mater
Lett 2005;59:2504–9.
[2] Han-Ki Yoon, Yu-Sik Kong, Seon-Jin Kim, Akira Kohyama. Mechanical
properties of friction welds of RAFs (JLF-1) to SUS 304 steels as measured by
the acoustic emission technique. Fusion Eng Des 2006;81:945–50.
[3] Dobrovidov AN. Selection of optimum conditions for the friction welding of
high speed steel 45. Weld Prod 1975;22(3):22–6.
[4] Ishibashi A, Ezde S, Tanaka S. Studies on friction welding of carbon and alloy–
steels. Bull JSME 1983;26:1080–7.
[5] Sahin Mumim. Evaluation of the joint interface properties of austenitic
stainless steel joined by friction welding. Mater Des 2007;28:2244–50.
[6] Sathiya P, Aravindan S, Noorul Haq A. Mechanical and metallurgical properties
of friction welded AISI 304 austenitic stainless steel. Int J Adv Manuf Technol
2005;26:505–11.
[7] Ananthapadmanaban D. A study of mechanical properties of friction welded
mild steel to stainless steel joints. Mater Des 2009;30:2642–6.
[8] Sathyanarayana VV, Madhusudhan Reddy G, Mohandas T. Dissimilar metal
friction welding of austenitic–ferritic stainless steels. J Mater Process Technol
2005:128–37.
[9] Yılmaz M. Investigation of the welding zone in friction welding of different
tool steels. Technical University of Yıldız, Doctoral Dissertation, Turkey;
1993.
1894 R. Paventhan et al. / Materials and Design 32 (2011) 1888–1894
[10] Afes Hakan, Turker Mehmet, Kurt Adem. Effect of friction pressure on the
properties of friction welded MA956 iron based super alloy. Mater Des
2007;28:948–53.
[11] Meshram SD, Mohandas T, Madhusudhan Reddy G. Friction welding of
dissimilar pure metals. J Mater Process Technol 2007;184:330–7.
[12] Hiizu Ochi, Koichi Ogawa, Takeshi Sawai, Yoshiaki Yamamoto, Ryoji Tsujino,
Yasuo Suga. Evaluation of tensile strength and fatigue strength of sus304
stainless steel friction welded joints. In: Proceedings of the thirteenth (2003)
international offshore and polar engineering conference, Honolulu, Hawaii,
USA; May, 2003. p. 25–30.
[13] Lee DG, Jang KC, Kuk JM, Kimb IS. Fatigue properties of inertia dissimilar
friction-welded stainless steels. J Mater Process Technol 2004;155–
156:1402–7.
[14] Mumin Sahin. Characterization of mechanical properties in AISI 1040 parts
welded by friction welding. Mater Charact 2007;58:1033–8.
[15] Malarvizhi S, Balasubramanian V. Effect of welding processes and post weld
aging treatment on fatigue behavior of AA2219 aluminium alloy joints. J Mater
Eng Perform 2010. doi:10.1007/5 11665-010-9682-5.
[16] Paventhan R, Lakshminarayanan PR, Balasubramanian V. Predictions of the
optimized friction welding process parameters for joining medium carbon
steel–austenitic stainless steel (MCS–ASS) dissimilar joints using RSM. Int J
Adv Manuf Technol; submitted for publication.
[17] Dieter GE. Mechanical metallurgy. 4th ed. New York: Tata McGraw Hill; 1988.
[18] Jaccard R. Fatigue crack propagation in aluminium. IIW Doc XIII; 1990. p.
1377–90.
[19] Hasalik A, Ozdemir N. Fatigue behavior of AISI 304 steel to AISI 4340 steel
welded by friction welding. J Mater Sci 2006;41:3233–9.
[20] Mumin Sahin. Joining with friction welding of high speed steel and medium-
carbon steel. J Mater Process Technol 2005;168:202–10.
[21] Sonsino CM. Fatigue assessment of welded joints in Al–Mg–4.5 Mn aluminium
alloy AA 5083 by local approaches. Int J Fatigue 1999;21:985–99.
[22] Behnken H, Hauk V. Micro-residual stresses caused by deformation, heat, or
their combination during friction welding. Mater Sci Eng 2000;A289:60–9.