Materials and Design 32 (2011) 3036–3050

Contents lists available at ScienceDirect

Materials and Design

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

Technical Report

Investigation on AISI 304 austenitic stainless steel to AISI 4140 low alloy
steel dissimilar joints by gas tungsten arc, electron beam and friction welding
N. Arivazhagan a,⇑, Surendra Singh a,b, Satya Prakash a,b, G.M. Reddy a,c
a
School of Mechanical and Building Sciences, VIT University, Vellore 632 014, India
b
Department of Metallurgical and Materials Engineering, Indian Institute of Technology, Roorkee, India
c
Defence Metallurgical and Research Laboratory, Hyderabad, India

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

Article history: This paper presents the investigations carried out to study the microstructure and mechanical properties
Received 24 November 2010 of AISI 304 stainless steel and AISI 4140 low alloy steel joints by Gas Tungsten Arc Welding (GTAW), Elec-
Accepted 24 January 2011 tron Beam Welding (EBW) and Friction Welding (FRW). For each of the weldments, detailed analysis was
Available online 31 January 2011
conducted on the phase composition, microstructure characteristics and mechanical properties. The
results of the analysis shows that the joint made by EBW has the highest tensile strength (681 MPa) than
the joint made by GTAW (635 Mpa) and FRW (494 Mpa). From the fractographs, it could be observed that
the ductility of the EBW and GTA weldment were higher with an elongation of 32% and 25% respectively
when compared with friction weldment (19%). Moreover, the impact strength of weldment made by
GTAW is higher compared to EBW and FRW.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction steels. Since it is of great concern in steam generating/handling

Austenitic stainless steel (AISI 304) and low alloy steel (AISI
4140) possess a good combination of mechanical properties, form-
ability, weldability, and resistance to stress corrosion cracking and
other forms of corrosion [1,2]. Owing to these attributes at moder-
ately high temperatures, such combinations of metals are exten-
sively used in the power generation industry [3]. In a nuclear
water reactor, dissimilar metal welds are employed to connect
the low alloy steel reactor pressure vessel and stainless steel pipe
systems. The dissimilar metal weldment joining boiler water reac-
tor nozzles to safe ends is one of the more complex configurations
in the entire recirculation system [4]. The problem with the dis-
similar metal weld made between low alloy steel and austenitic
stainless steel with an austenitic stainless steel filler metal is the
carbide formation due to higher carbon content of low alloy steels
than that of austenitic stainless steel [5].
As the strength of the dissimilar weldments is generally inferior,
most of the in-service failures are reported to take place in the
weld region [6,7]. Such failures in the transition zone between fer-
ritic steel and austenitic stainless steel is a perennial problem in
fossil-fired steam plants [8]. Li et al. [9] have studied the effect of
post weld heat treatment on stress corrosion cracking of low alloy
steel to austenitic stainless steel transition weld made by manual
metal Arc welding. Raman et al. [10] discussed the prevalence of
in-service failures in the welds of chromium–molybdenum ferritic
⇑ Corresponding author. Tel.: +91 9443034794; fax: +91 416 2243092.
E-mail address: narivazhagan@vit.ac.in (N. Arivazhagan).
systems of power plants as well as the components of petro-
leum/petrochemical industries, studies are required to avoid such
failures. Mainly these failures occur either in the heat affected zone
or in the weld zone [11–13]. Similar and dissimilar joints involving
austenitic steels are susceptible to unexpected phase propagation.
As a result of this, a series of negative metallurgical changes such
as delta ferrite phase, grain boundary corrosion and sigma phase
occurs at the weld interface. Therefore, higher welding speeds
are necessary to avoid such effects. Sometimes extensive care
and precautions are needed such as pre and post heat treatment
processes [14]. The micro-segregation which occurs in weld fusion
zones of dissimilar metals leads to a situation where interdendritic
regions are enriched in Fe, Cr and C. This segregation within the
dendritic structure results in deterioration of the mechanical prop-
erties and corrosion resistance of the joints [15,16]. However, such
problems can be minimized by judicious selection of the welding
process and parameters. In view of the above, solidification during
EBW technique is less likely to result in precipitation of unwanted
intermetallic constituents. Joining dissimilar metals using EBW has
been a subject of interest in recent years. It has been reported that
the mechanical properties of dissimilar metals strongly depend on
the microstructure of the joints. Thus, studies on the relationship
between microstructure and mechanical properties are needed.
Notwithstanding the ability of EBW process, friction welding of
AISI 4140 and AISI 304 have also found to bring down the severity
of segregation associated with solidification and in cutting down
the volume fraction of the unwanted phases [17]. Roberto [18]
had studied the pulsed Nd:YAG laser welding of AISI 304–AISI

0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2011.01.037
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3037

420 stainless steels and Anawa and Olabi [19] optimized the ten-
sile strength of ferritic/austenitic laser-welded components. How-
B5
ever, the mechanical properties of dissimilar welding of austenitic
stainless steel to low alloy steel by GTAW, EBW and FRW have 4140 304
been reported seldom. Especially, the comparative studies of the
effect of welding methods on mechanical properties of AISI 304
and AISI 4140 dissimilar metals has received scant attention in
the past. This paper presents the studies on AISI 304 and AISI
4140 dissimilar metals welded by GTAW, EBW and FRW. For each B7
of the weldments, studies were conducted on the phase composi-
tion, microstructure characteristics and mechanical properties
4140 304
using X-ray Diffration (XRD), Scanning electron Microscopy (SEM).

2. Experimental procedures
B9
In this study experiments were conducted using a conventional
DCEN GTAW, Low kV Electron beam welding machine (TECH-
META) and 150 KN capacities continuous drive friction welding 4140 304
machine (ETA Pvt, Ltd). The parent metals employed in this study
are AISI 304 stainless steel, AISI 4140 low alloy steel and their
chemical composition shown in table 1. The parameters used for
autogenious weldments are shown in table 2. Specimens welded
by FRW were cut to the shape of the cylinder of diameter and B12
length 25 mm and 55 mm respectively. The base metals used for
EBW and GTAW were taken from the same lot (25 mm diameter 4140 304
and 200 mm length) used for FRW. These were hot forged to thick-
ness of 6 mm sheets which has been square butt joint configura-
tion used for EBW and GTAW respectively and the weld Flash
specimens are shown in Figs. 1a–c.
In order to observe the microstructure under the optical micro- Fig. 1a. Friction welded specimen (Burn-off length B5, B7, B9 and B12).
scope, specimens were cut from the welds, and then prepared
according to the standard procedures. Micro-hardness tests were
carried out using a Vickers digital micro-hardness tester along
the weld joint. A load of l00 gm was applied for duration of 10 s.
Tensile test (ASTM E8M, [20]) and impact toughness test (ASTM

AISI 304 AISI 4140

Table 1
Chemical composition of AISI 304 and AISI 4140.

Parent Metal C Cr Mn Ni Si Mo Fe
AISI 304 0.06 18.4 1.38 8.17 0.32 – Balance
AISI 4140 0.40 1.1 0.75 – 0.31 0.28 Balance

Table 2
TIG, EBW and FRW welding parameters.
Weld

Type of weld Welding parameter Value

Electron beam Current (DCSP) (Amp) 300
welding
Voltage (volt) 23
Argon pressure (bar) 2.1
Torch traveling speed (cm/s) 0.2
Tungsten inert gas Work distance (mm) 275
welding
Accelerating voltage (kV) 55
Beam current (mA)
(Beam focus slightly above 50
the surface)
Travel speed (m/min) 1
Vacuum level (mbar) 104
Friction welding Speed 1500 rpm
Friction force 37.5 KN
Upset force of 50 KN
Bum-off length ‘B’ 5 mm, 7 mm, 9 mm
and 12 mm

Fig. 1b. Photograph of electron beam welded specimen.

3038 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

AISI 4140
Weld interface

AISI 304

Fig. 1c. Photograph of GTA welded specimen.

Flash

304 4140
AISI 304

Fig. 2a. Tensile specimen preparation by EDM wire cut.

Weld

AISI 4140
B5

B12
B7

EBW TIG
B9
AISI 304

Fig. 2c. Tensile specimen for EBW and TIG welded sample.

Figs. 2a–c. The fractured morphology of the specimens was ana-

Friction weld interface

lyzed by Scanning Electron Microscopy (make JEOL, JSM-5800)

AISI 4140

with EDAX attachment (Oxford, Model-6841, England). The joints

were analyzed by a Philips X-ray diffractometer (XRD) for studying
the phase composition.

3. Experimental results

3.1. Macrostructure and microstructure

Full penetration joints were produced by GTAW, EBW and FRW.

Fig. 2b. Tensile specimen for friction weldment, Burn-off length ‘B0 5, 7, 9 and
12 mm.
E 23, [21]) specimens were prepared by cutting the welded speci-
mens. The welded samples were machined by EDM wire cut to the
required size keeping weld interface at center, and are shown in
Macrostructure of the three joints is shown in Figs. 3a–c. The width
and fusion area are different for joints made by the above three
welding techniques. This may be attributed to the variation in
welding speed and heat input used in the different welding
methods.
The XRD analysis of the three joints is given in Figs. 4a–d. Inves-
tigations of the results of the XRD shows high intensity peaks of
FeNi and Ni3C have for 5 mm burn-off length before the specimens
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050
Fig. 3a. Macro structure of GTA welded AISI 4140 and AISI 304 dissimilar metals.
Weld
AISI 304 AISI 4140
Fig. 3b. Macro structure of electron beam welded AISI 4140 and AISI 304 dissimilar
metals.
are tested in corrosive environment. Similarly, at 7 mm burn-off
length, many intermetallics (FeNi, Cr–Ni–Mo and CrNiFe) and car-
bides (Ni3C, Cr–Si–C etc.,) were indicated. Whereas for the EBW
welded specimen, (Cr,Fe)7C3 and Cr–Ni–Mo have been observed
with high intensity peaks and Ni3C, FeNi as low intensity peaks.
XRD studies for GTA welded specimen, (Cr,Fe)7C3 has been indicat-
ing with high intensity peak whereas Cr–Ni–Mo, CrNiFe, Ni3C, FeNi
etc., are having low intensity peaks.
The EPMA elemental mapping results are shown in Figs. 5a–f.
From these figures it is observed that chromium and nickel diffuse
towards low alloy steel from the stainless steel and diffusion of
iron from low alloy steel side towards stainless steel. The extent
of diffusion is more in case of 12 mm burn-off length welds. A dis-
tinct weld region with enrichment of chromium, nickel and iron
develops in 12 mm burn-off length welds. Similar trends have been
observed for EBW and GTA weldments.
The microstructures of AISI 4140 are found adjacent to the FRW
weld interface and coarser martensitic features were observed.
Moreover, extensive grain growths with well developed grain
boundaries are noted adjacent to the weld interface and near this
region predominant precipitation were observed (Fig. 6).
B5 B7
Flash
Weld
2000µm 2000µm
Fig. 3c. Effect of burn – off length on flash formation (Material on left side in the macrographs is AISI 304 stainless steel.) burn-off length: B5, B7, B9 and B12.
3039
Prominent deformation bands can be observed in AISI 304 stainless
steel side adjacent to the interface.
The microstructure of the EBW dissimilar weld is depicted in
Fig. 7. It can be observed that ferrite mixed with pearlite structure
is predominant near weld interface on AISI 4140 side. Due to large
amounts of Cr and Ni, twin structure with mixtures of austenite
and some ferrite is observed on AISI 304 side. During the welding,
the region close to the fusion zone undergoes rapid heating at a
temperature above the AC3 point. The microstructure of the
austenitic steel close to the weld interface shows evidence of the
small amount of grain recrystallization with little carbide precipi-
tation. In addition, there was a lot of variation in grain size in the
HAZ of the low alloy steel, caused by variations in cooling rate.
The microstructures in the interface are extremely heterogeneous
and most of them are dendrite structures. Even though the excess
phases exist along the fusion line on both sides of weld metal, they
are not predominant.
The microstructure of GTA welded low alloy steel (AISI 4140)
and austenitic stainless steel (AISI 304) dissimilar weldment is het-
erogeneous (Fig. 8). At the weld interface on the low alloy steel side
the microstructure was coarse, contained malformed lamellae of
pearlite. Further some directionality has been observed near fusion
boundary. This may be due to local variation in solidification mode
and growth velocity at fusion boundary. The fusion zone presents
typical metallograph in GTAW and EBW joints. These joints are
composed of dark d-Fe dendritic structure in austenite matrix.
However, different dendrite sizes can be observed. Furthermore,
the d-Fe in GTAW and EBW joint exists as columnar structure.
3.2. Mechanical properties
The tensile, impact toughness and hardness data of weldment
made by FRW, EBW and GTAW joints are presented in Table 3.
The tensile strength of all joints made by above mentioned pro-
cesses had been evaluated and are shown in Figs. 9a–c. In each con-
dition, three specimens have been tested and the average tensile
strength is obtained to ensure repeatability. The tensile strength
of EBW joint is 681 MPa, while those of GTAW and FRW joints
B9 B12
2000µm 2000µm
3040 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

Fig. 4a. X-ray diffraction patterns for friction weldment as welded condition burn-off length B5 and B7.

Fig. 4b. X-ray diffraction patterns for friction weldment as welded condition Burn-off length B9 and B12.
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3041

Fig. 4c. X-ray diffraction patterns for Electron Beam weldment as welded condition.

Fig. 4d. X-ray diffraction patterns for TIG weldment as welded condition.

are 635 and 494 MPa respectively. The fractured surface of tensile
specimens is analyzed using SEM. SEM fractograph macrostructure
of all joints shown in Figs. 10a and 10b. Different fracture modes
are observed for joints by three welding processes. Particularly,
Surface slip and necking down region can be observed from FRW
joint, which is the typical characteristic of cup–cone shaped frac-
ture. While for the EBW and GTAW joints, it is just the pure-shear
fracture mode, resulting from the dislocation slip. Different micro-
zones showed in Fig. 11(i–vi) are investigated in detail to establish
the nature of fracture. The fractograph shows that relatively minor
size dimples which are surrounded by coarse dimples and a small
quantity of tearing ridge in EBW and GTAW joint. There exists
appreciable difference among EBW, GTAW and FRW joints in the
appearance of dimples. The EBW joint invariably consists of fine
and uniform dimples, which indicate that the specimen fails in a
ductile manner under the action of tensile loading. For the FRW
joint, the fractograph invariably contains of much smaller dimples
with some coarse dimples distributed among the fine dimples.
Micro-hardness analysis was carried out across the weld joint
in the axial direction using a Vickers digital micro-hardness tester
(Figs. 12i–iii). In addition the peak hardness of FRW weldment in-
creases with increase in burn-off length, whereas the hardness
values an adjacent low alloy steel further falls with an increase
in burn-off length. It should be noted that with increase in
burn-off length a soft region appears on the stainless steel adja-
cent to the interface. In general, the width of the hardened region
is said to decrease with increase in burn-off length. The formation
of soft region on the low alloy steel side can be attributed to
decarburization while on the stainless steel it is due to softening
of the material. This occurs by the buildup of heat as the thermal
conductivity of the material is relatively lower. The reduction in
the width of hard zones is attributed to the reason that loss of
material in the form of flash has increased with burn-off length.
The study shows that overall hardness as well as soft zone hard-
ness on the low alloy steel side of the interface occurs mainly
due to the increase in hardness at the interface. This behavior
can be attributed to carbon migration from the low alloy steel side
to stainless steel side. The hardness of the EBW weld region is in
the range 600 Hv (min) to 700 Hv (max), whereas the hardness of
the GTAW weld region is in the range 473 Hv (min) to 513 Hv
(max). Highest hardness (698 Hv) observed at interface of AISI
304 side.
3042 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

CK
CK

Cr K
Cr K

Fe K
Fe K

Ni K
Ni K

a b

CK CK

OK OK
Si K Si K
Cr K Cr K

Mn K Mn K
Fe K Fe K
Ni K Ni K

c d

CK CK

Cr K Cr K

Fe K Fe K

Ni K Ni K

e f
Fig. 5. Elemental distribution across the weld centre in 4140-304 (a) FRW-B5, (b) FRW-B-12, (c and d) EBW, (e and f) GTAW.

In general the impact toughness values are lower than those of
low alloy steel in the as received condition (65 J). At the lowest
burn-off length (5 mm) the impact toughness is around 40% of
the parent metal whereas at the highest burn-off length (12 mm)
it is as low as 11 J, which is l/6th of the parent low alloy steel. Frac-
tographs of the impact tested specimens of FRW, EBW and GTAW
are shown in Fig 13i–vi. For friction welded condition, predomi-
nantly cleavage fracture is observed which increases with burn-
off length. For larger burn-off length welds (12 mm), long facets
with micro void features are observed. There is a marked increase
in facet size with increase in burn-off length and it indicates low
energy fracture. This aids the decrease in toughness which is found
to increase with burn-off length (Fig 13iv). Fractographs of the im-
pact tested specimens of EBW welds shows dimple structure (duc-
tile fracture). In the case of GTA welded joints, it has been observed
that the structure exhibit primarily the cleavage fracture with
occasional appearance of micro void features (Fig. 13v, vi).
4. Discussions
Welding of dissimilar metals of AISI 4140 and AISI 304 by FRW,
EBW and GTAW techniques can produce satisfactory joints in
terms of weldability and room temperature mechanical properties.
The cumulative mechanical properties of AISI 4140 and AISI 304
dissimilar metals made by FRW, EBW and GTAW are shown in
Fig. 14.
Investigation on the effect of burn-off length on mechanical
properties of AISI 4140AISI 304 dissimilar welds revealed that
the toughness of these welds is very low while the hardness is
high. The reduction in toughness and increase in hardness with in-
crease in burn-off length may be attributed to increased carbon
migration from the low alloy steel side towards stainless steel. In-
creased burn-off length leads to increased friction time that aids in
raising the temperature of the surroundings of the interface
[22,23]. The measured value of hardness is in good agreement with
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3043

304 4140

304 304 4140

Inter mixing zone
Precipitation Deformation band

Coarser martensitic

Developed grain
boundaries
50 µm 50 µm

Fig. 6. Microstructures in different regions of AISI 304 in burn-off length 12.

AISI 304
AISI 4140

HAZ
Weld HAZ

Weld
Interface Weld
Interface
100µm 100µm

Carbid

40µm 40µm

AISI 4140

Weld zone

100µm 100µm

Fig. 7. Microstructure of electron beam welded dissimilar AISI 4140 and AISI 304 metals.
3044 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050
40µm
Weld Zone
40µm
AISI 4140
40µm
Fig. 8. Microstructure of TIG welded AISI 4140 and AISI 304 dissimilar metals.
previous findings by other researchers [24–26]. The hardness at
the weld interface obtained in the present study is less than one re-
ported for laser beam welding [27]. This may be attributed to lesser
tendency of carbide precipitation and intermetallics formation in
the weld region of FRW weldments. Moreover the soft zone ob-
served on the low alloy steel side can be ascribed to carbon deple-
tion whereas enrichment in the stainless steel side results leads to
formation of hard and brittle carbide phases formation (Fig. 12i),
which is also supported by Kuchuk and Lippold and Odegaard
[28,29]. The higher hardness at weld interface can be due to
enrichment of this zone with Fe, Ni and Cr and subsequent forma-
tion of intermetallic FeNi/CrNiFe phases (Figs. 4a–d, Fig 5) which
also supported by Murti and Sundarsean [30]. Elemental distribu-
tion across the interface and predominantly cleavage fracture fea-
tures suggest that inter-diffusion of elements has lead to creation
of low ductility quassi cleavage fracture (Fig. 13iv). Similar obser-
vation has been made by Cheng and Wang [22,23]. At lower
burn-off length, the weldment attains to lower interface tempera-
ture at which metal starts getting extruded resulting in less
susceptibility to formation of intermetallics and carbide precipita-
tion. Whereas with increase in the burn-off length, the welding
time is increased thereby allowing more time for precipitation of
carbide and intermetallics compound which may lead to reduced
AISI 304 (HAZ)
Carbide Weld
40µm
Weld
AISI 4140 (HAZ)
40µm
corrosion resistance. The micro-hardness curves indicate the
EBW and GTAW joints, which also follow the same trend as FRW
shown in Figs. 12ii and iii.
The HAZ in GTAW joint is attributed to the low welding speed
and high heat input used in the experiment. It causes to the fusion
pool longer residence in high temperature zone and slower cooling
rate, while for the EBW, higher welding speed and lower heat-in-
put are adopted. The size of fusion pool is small and the cooling
rate is very fast. Moreover in the cooling process, the grains be-
tween parent material and fusion zone have no time to grow up.
It should be noted that the HAZ is absent in EBW joints which re-
sults the higher tensile strength as compared to GTAW and FRW
joints (Figs. 6–8). Furthermore, high heat-input was likely to in-
duce the micro-segregation of alloying elements and formation of
Cr-depleted zones, resulting in the degradation in mechanical
properties for the GTAW joint [31]. Since the formation of the Cr
carbides is caused by exposure to the temperature range of 773–
1073 K, rapid cooling rate and shorter exposure time to the sensi-
tization temperatures in EBW joints are preferable for prevention
of sensitization and reduce the amount of Cr-rich carbides formed
on the grain boundaries [32,33]. Therefore, a significant improve-
ment in mechanical properties can be achieved in EBW joints when
compared with GTAW and FRW joints. In FRW joints volume of vis-
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3045

Table 3
The cumulative mechanical properties of AISI 4140 and AISI 304 dissimilar welment made by FRW, EBW and TIG.

Welding B.L XRD Max. Hard. At Yield % of Tensile Impact Impact Remark
(mm) weld (HV) strength Elongation fractography toughness fractography
(MPa) (J)
Friction welding 5 FeNi, Ni3C 305 485.2 18.15 Failure at 26 Low burn-off Diffusion of Fe, Cr
low alloy length indicates Ni, C high at higher
steel side. ductile fracture burn-off length. At
micro void and higher burn- lower burn good
ductile off length shows mechanical
fracture brittle fracture. property noted.
7 (Cr,Fe)7C3, Cr–Ni– 322 493.7 18.89 22
Mo, Cr–Si–C, FeNi,
Cr–Ni–Fe
9 Cr–Ni–Fe, 351 474.35 17.15 16
(Cr,Fe)7C3, Cr–Ni–
Mo, Ni,Fe, Ni3C,
FeNi
12 Cr–Ni–Mo, Cr–Ni– 386 476.4 18.30 11
Fe, Ni3C, FeNi
Electron beam – (Cr,Fe)7C3, Ni,Fe, 400 681 31.97 Failure at 21 Ductile fracture Excellent
welding Cr–Ni–Mo, Ni3C, HAZ of 4140 mechanical
Cr–Ni–Fe side. micro property
void ductile
fracture
Tungsten inert gas – (Cr,Fe)7C3, Cr–Ni– 400 634.5 24.96 Failure at 25.5 Cleavage fracture Excellent
welding Fe, Ni3C, FeNi, HAZ of 4140 mechanical
Ni,Fe, Cr–Ni–Mo side. micro property also
void ductile micro cracks noted
fracture at weld

B5 B7 B9 B12
AISI 304

Weld

AISI 4140

Fig. 9a. Photograph for fractured tensile specimen of friction welded specimen.

AISI 304 AISI 4140

AISI 304 AISI 4140

Weld
Weld

Fig. 9b. Photograph of tensile fractured weldment made by EBW. Fig. 9c. Photograph for tensile fractured GTA welded AISI 4140 and AISI 304.
3046 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

Fig. 10a. Tensile fractographs of friction welded AISI 304 and AISI 4140 by keeping burn-off length 5,7,9 and 12 mm.

Fig. 10b. SEM fractographs for tensile specimen for EBW (i) and GTAW (ii) welded AISI 4140 and AISI 304.

cous material transferred at the weld interface decreases as a result
of more mass discarded from the welding interface, which resulted
in lower tensile strength.
The fractograph of tensile tested specimen shows that relatively
minor sized imples surround coarse dimples and a small quantity
of tearing ridge (Fig. 13vi) can be observed in GTAW joint. Fracto-
graph of impact test GTAW weldment shows that the fracture dur-
ing impact loading propagated in the weld metal near the toe of
the weld. It should be noted that the fractography of EBW shows
ductile fracture both tensile and impact toughness test. The EBW
joint invariably consists of fine and uniform dimples (Fig. 13v),
which indicate that the specimen fails in a ductile manner under
the action of tensile loading. Since the failure of EBW tensile spec-
imen at the HAZ on AISI 4140 side, the weld strength is equal or
more than 681 (MPa) and the maximum impact toughness (21 J)
were noted. The strength of the EBW is higher as compared to
GTAW (635 MPa) and FRW (494 MPa).
5. Conclusions
On the basis of this investigation, the following conclusions can
be drawn.
(1) Sound welds could be obtained between AISI 304 austenitic
stainless steel and AISI 4140 low alloy steel by all the three
welding techniques.
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3047

Fig. 11. Tensile Fractographs of welded AISI 304 and AISI 4140 by FRW (i–iv), EBW (v) and GTAW (vi).

Fig. 12i. Effect of burn-off length (B 5, B7, B9 and B12) on hardness distribution across the weldment.
3048 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

800

700

600
Hardness (HV)
500

400

300

200

100

0
0 1 2 3 4 5
Distance across the weld (mm)
Fig. 12ii. Hardness distributions across the EBW weldment of AISI 4140 and AISI 304.

800

700

600
Hardness (HV)

500

400

300

200

100

0
0 2 4 6 8 10 12
Distance across the weld (mm)
Fig. 12iii. Hardness distributions across the TIG weldment of AISI 4140 and AISI 304.

(2) In friction welded AISI 304 and AISI 4140 dissimilar metals, the
reduction in toughness and increase in hardness across the
weld with increase in burn-off length is related to increased
carbon migration from the low alloy steel side towards stain-
less steel due to higher temperatures that prevail around the
interface region with increase in burn-off length.
(3) In friction weldment that brittle fracture was observed as we
increase the burn-off length which further supporting the
formation of carbides and intermetallic compounds. From
the SEM fractogaphic of impact tested dissimilar weldment
made by EBW and GTAW shows cleavage fracture.
(4) The yield strength of dissimilar weldment made by EBW is
higher as compared to GTAW and FRW. The failure was
observed on AISI 4140 side for friction weldement, where
as in the case of weldment made by EBW and GTAW, the
failure occurred on HAZ of AISI 4140.
(5) From EPMA analysis on the dissimilar weldment it is
observed that chromium and nickel diffuse towards AISI
4140 from the AISI 304 and diffusion of iron from AISI
4140 side towards AISI 304. The extent of diffusion is more
in case of EBW welded dissimilar metals as compared to
the weldment made by GTAW and FRW.
 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050 3049

Fig. 13. Fractographs of 304-4140 weld made by FRW (i–iv), EBW (v) and GTAW (vi).

800
FRW-B5
700 FRW-B7
FRW-B9
FRW-B12
600
EBW

500
TIG

400

300

200

100

0
Maximum weld hardness (HV) Yield strength (Mpa) Impact toughness (J)

Fig. 14. Bar charts showing cumulative mechanical properties of AISI 4140 and AISI 304 dissimilar metals made by FRW, EBW and TIG.

(6) A distinct weld region with enrichment of chromium, nickel, (7) The Electron beam welding is suitable for welding AISI 304
iron and carbon is prominent the weldment made by EBW as with AISI 4140 in industry application owing to high weld-
compared to GTAW and FRW. ing speed and excellent mechanical properties.
3050 N. Arivazhagan et al. / Materials and Design 32 (2011) 3036–3050

References [17] Hasçalik A, Ünal E, Özdemir N. Fatigue behaviour of AISI 304 steel to AISI 4340
steel welded by friction welding. J Mater Sci 2006;41:3233–9.
[18] Roberto Berretta Jose, de Rossi Wagner, Martins das Neves, Mauricio David, de
[1] Nascimento MP, Souza RC, Pigatin WL, Voorwald HJC. Effects of surface
Almeida Ivan Alves, de Rossi Wagner, et al. Pulsed Nd: YAG laser welding of
treatments on the fatigue strength of AISI 4340 aeronautical steel. Int J Fatigue
AISI 304 to AISI 420 stainless steels. Opt Laser Eng 2007;45:960–6.
2001;23:607–18.
[19] Anawa EM, Olabi AG. Optimization of tensile strength of ferritic/austenitic
[2] International Stainless Steel Forum Rue Colonel Bourg 120 B-1140 Brussels
laser-welded components. Opt Laser Eng 2008;46:571–7.
Belgium. Stainless Steel Applications – Aerospace. 2007/05/23. p. 1–2
[20] ASTM E8M-03, Standard test methods for tension testing of metallic materials,
[3] Shushan SM, Charles EA, Congleton J. The environment assisted cracking of
vol 01.02., 2003.
diffusion bonded stainless steel to carbon steel joints in an aqueous chloride
[21] ASTM E23-04, Standard Test Methods for Notched Bar Impact Testing of
solution. Corros Sci 1996;38(5):673.
Metallic Materials, Vol 03.01., 2004.
[4] Smith Richard E, Hanford Ray, Cheng SC. Pressure vessel nozzle repair. Nucl
[22] Cheng CJ. Transient temperature distribution during friction welding of two
Eng Des 1990;124(1–2):79–89.
dissimilar materials in tabular form. Weld J 1963;54(5):233s–40s.
[5] Welding Handbook. 8th ed., vol. 2, Miami, Florida: American Welding Society;
[23] Wang KK. Friction Welding. WRC Bull 204 1975.
1991.
[24] Celik Sare, Ersozlu Ismail. Investigation of the mechanical properties and
[6] Arivazhagan N, Singh Surendra, Prakash Satya, Reddy GM. High temperature
microstructure of friction welded joints between AISI 4140 and AISI 1050
corrosion studies on friction welded dissimilar metals. Mater Sci Eng: B J
steels. Mater Des 2009;30:970–6.
2006;132:222–7.
[25] Yenni C, Erim S, Cam G, Kocak M. Symposium on Microstructural Feature and
[7] ASME, ASME Boiler and Pressure Vessel Code, Code Case N-47. New York:
Fracture Behaviour of Laser Welded Similar and Dissimilar Steel
ASME; 1986.
Joints. Instanbul, Turkey: Gedik Education Foundation; 1996. p. 235.
[8] Tucker JT, Eberle F. Development of a ferritic-austenitic weld joint for steam
[26] Cam G, Yenni C, Erim S, Ventzke V, Kocak M. Progress in joining of advanced
plant application. Weld J 1956;35(11):529–40.
materials: part 1: solid state joining, fusion joining, and joining of
[9] Li GF, Charles EA, Congleton J. Effect of post weld heat treatment on stress
intermetallics. J Sci Tech weld Join 1998;3:3.
corrosion cracking of a low alloy steel to stainless steel transition weld. Corros
[27] Cam G, Erim S, Yenni C, Kocak M. Determination of mechanical and fracture
Sci 2001;43(10):1963–83.
properties of laser beam welded steel joints. Weld J 1999;78(6):193–201.
[10] Raman Singh RK, Muddle BC. High temperature oxidation in the context of life
[28] Kuchuk-Yatsenko SI. Flash welding of large high-strength steel pipe. Weld J
assessment and microstructural degradation of weldments of 2.25Cr–1Mo
1990;67(7):47–50.
steel. Int J Press Vess Pip 2002;79:585–90.
[29] Lippold JC, Odegaard BC. Microstructural evolution during inertia friction
[11] Tsay LW, Liu YC, Young MC, Lin DY. Fatigue crack growth of AISI 304 stainless
welding of austenitic stainless steels. Weld J 1984;63(1):35–8.
steel welds in air and hydrogen. Mater Sci Eng 2004;A374:204.
[30] Murti KGK, Sundarsean S. Thermal behavior of Austenitic–Ferritic transition
[12] Johan Singh P, Guha B, Achar DRG. Fatigue life prediction for stainless steel
joints made by friction welding. Weld J 1985;64(12):327s–34s.
welded CCT geometry based on Lawrance’s local-stress approach. Eng Failure
[31] Yan Jun, Gao Ming, Zeng Xiaoyan. Study on microstructure and mechanical
Anal 2003;10:655.
properties of 304 stainless steel joints by TIG laser and laser-TIG hybrid
[13] Wahap MA, Sakano M. Corrosion and biaxial fatigue of welded structures. J
welding. Opt Lasers Eng 2010;48:512–7.
Mater Process Technol 2003;410:143–4.
[32] Park Seung Hwan C, Kokawa Hiroyuki, Okamoto Kazutaka, Hirano Satoshi,
[14] Özdemira N, Sarsılmaza F, Hasçalıkb A. Effect of rotational speed on the
Inagaki Masahisa. Corrosion resistance of friction stir welded 304 stainless
interface properties of friction-welded AISI 304L to 4340 steel. Mater Des
steel. Scripta Mater 2004;51:101–5.
2007;28(1):301–7.
[33] McPherson NA, Li Y, Baker TN. Microstructure and properties of as welded
[15] Lima AS, Nascimento AM, Abreu HFG, de Lima-Neto P. Sensitization evaluation
duplex stainless steel. Sci Technol Weld Join 2000;5(4):235–44.
of the austenitic stainless steel AISI304L, 316L, 321 and 347. J Mater Sci
2005;40:143.
[16] Muthupandi V, Bala Srinivasan P, Seshadri SK, Sundaresan S. Effect of weld
metal chemistry and heat input on the structure and properties of duplex
stainless steel welds. Mater Sci Eng A 2003;358:9–16.