Joining of stainless steel and copper materials

with friction welding

Mumin Sahin
Mechanical Engineering Department, Faculty of Engineering and Architecture, Trakya University, Edirne, Turkey

Abstract
Purpose – The purpose of this paper is to investigate mechanical and metallurgical variations at interfaces of commercial austenitic-stainless steel and
copper materials welded by friction welding.
Design/methodology/approach – In this paper, austenitic-stainless commercial steel and copper materials are welded using the friction welding
method. The optimum parameters are obtained for the joints. The joints are applied to the tensile and micro-hardness tests. Then, micro- and
macro-photos of the joints are examined.
Findings – It is found that some of the welds show poor strength depending on some accumulation of alloying elements at the interface result of
temperature rise and the existence of intermetallic layers.
Research limitations/implications – It would be interesting to search about the toughness values and fatigue behaviour of the joints. It could be a
good idea for future work to concentrate on the friction welding of these materials.
Practical implications – Friction welding can be achieved at high-production rates and therefore is economical in operation. In applications where
friction welding has replaced other joining processes, the production rate has been increased substantially.
Originality/value – The main value of this paper is to contribute to the literature on friction welding of dissimilar materials.

Keywords Friction welding, Stainless steel, Copper

Paper type Research paper

1. Introduction and rotation speed are the principal welding variables in the
friction welding method (Vill, 1962).
Friction welding can be achieved at high-production rates and In practice, friction welding is divided into two ways as
therefore is economical in operation. In applications where continuous drive friction welding and inertia friction welding
friction welding has replaced other joining processes, the (Kinley, 1979; Fomichev, 1980). In the continuous drive friction
production rate has been increased substantially. method shown in Figure 1, one of the components is held
When joining dissimilar metals such as copper and steel, stationary while the other is rotated at a constant speed (s). The
friction welding is of great importance in applications. two components are brought together under axial pressure (Pf)
Welding is possible within the limited range of the welding for a certain friction time (tf). Then, the clutch is separated from
conditions although problems for welding exist because of the drive, and the rotary component is brought to a stop within
brittle intermetallic compounds and high-thermal the braking time while the axial pressure on the stationary part is
conductivity. To minimize the problems, the friction welding increased to a higher upset pressure (Pu) for a predetermined
parameters must be taken into consideration for strong welds. upset time (tu). Parameters of the method are shown in Figure 2.
Heat in friction welding is generated by conversion of Various studies on the effects of mechanical, thermal and
mechanical energy both into heat energy and visco-plastic metallurgical properties of welded parts are obtained from
deformation work (energy) at the interface of the work pieces literature on the friction welding method.
during rotation under pressure. This visco-plastic deformation Vill (1962), Kinley (1979) and Fomichev (1980) have
work (energy) also produces heat energy via internal viscous written studies on the friction welding set-ups and the
dissipation. strength of the joints, and Imshennik (1973), Rich and
Various ferrous and non-ferrous alloys, which have circular Roberts (1971) and Sluzalec (1990) wrote studies on the
or non-circular cross-sections and, having different thermal thermal effects of welded parts in friction welding. Ellis
and mechanical properties, can easily be joined by the friction (1977) examined relations between “friction time-workpiece
welding method. Friction welding is classified as a solid-state diameter”, “loss of length-forging pressure” and “carbon
welding process where metallic bonding is produced at equivalent-hardness variation”. Yılbaş et al. (1995) and Şahin
temperatures lower than the melting point of the base metals. et al. (1998) investigated the mechanical, metallurgical and
Friction time, friction pressure, forging time, forging pressure thermal properties of friction welded steel-aluminium,
steel-copper and aluminium-copper bars. Murti and
The current issue and full text archive of this journal is available at Sundaresan (1983) directed a study about parameter
www.emeraldinsight.com/0036-8792.htm optimisation in friction welding of dissimilar materials.

The author would like to thank Hema Industry/Çerkezköy and

Industrial Lubrication and Tribology
61/6 (2009) 319– 324 Mechanical Engineering Department of Trakya University, Edirne,
q Emerald Group Publishing Limited [ISSN 0036-8792] Turkey for their assistance in experimental and microstructure part of
[DOI 10.1108/00368790910988435] the study.

319
 Joining of stainless steel and copper materials Industrial Lubrication and Tribology
Mumin Sahin Volume 61 · Number 6 · 2009 · 319 –324

Figure 1 Layout of continuous drive friction welding 2. The experimental procedure

2.1 Material
In the experiments, AISI 304 austenitic-stainless steel and
copper materials were used. The composition of materials is
given in Tables I and II (Wegst, 1995).
5 6
1 2 2.2 Geometry of parts
3 4 7
The experiment specimens were machined from materials
Notes: 1. Motor; 2. Brake; 3. Rotating chuck; 4. Non rotating chuck; 5. Rotating according to geometry in Figure 3.
workpiece; 6. Non rotating workpiece; 7. Hydraulic cylinder

2.3 The experiment set-up

Figure 2 Parameters for continuous drive friction welding In the present study, an experimental set-up was designed and
constructed as a continuous drive method. Experimental
parts of the presented study were conducted using an
experimental set-up that was designed and constructed as a
Braking
Friction time (tf) continuous drive type. In addition, a solenoid valve and
electrical control circuit was designed and constructed to
Upsetting control of friction time and pressure in the set-up. Thus, these
Parameters

(tu) supplied the control possibility of the experiments easily. The

Rotation speed (s) Upsetting friction welding set-up is shown in Figure 4.
pressure(Pu)
Friction pressure (Pf) 3. Selection of welding parameters using
Torque statistical approach
Shortening
When the statistician can suggest the experimental runs to be
made in a series of tests, data analysis can be quite
Welding Welding
Time completed
straightforward. Often, however, an analysis must be carried
started out on the results that have been accumulated as part of a test
program or as a series of related test programs and an
Nentwig (1996) investigated the effect of cross-section empirical predictive model must be constructed from the
differences of the components upon the joint quality of results obtained. Terms for possible inclusion in the model
friction welding and offered that friction pressure, upset might involve not only the principal variables but also
pressure and rotation speed must be changed in friction variables such as cross products, squares, or other
welding of different cross-sections. Fabritsiev et al. (1998) combinations, or transformations, of the principal variables
examined the irradiation resistance of DS copper stainless (Sahin and Akata, 2003; Draper, 1981).
steel joints fabricated by friction welding methods. Lee and The basis of this approach is the assumption of a
Jung (1993) investigated the effect of microstructure variation simplified linear model for the optimisation parameter
on the Cu/CK45 carbon steel friction weld joint. h given by h ¼ b0 þ b1 x1 þ b2 x2 þ · · ·; where x1, x2, . . . , etc.
Then, Sahin and Akata (2003) made a study on joining of are the factors on which h depends and b0, b1, b2 . . . , etc.
plastically deformed steel (carburising steel) with friction represent the “true” values of the corresponding unknowns.
welding. Later, Sahin and Akata (2004) directed an From the results of an experiment comprising a finite number
experimental study on friction welding of medium-carbon of trials, one can arrive at sample estimates of the coefficients,
steel and austenitic-stainless steel components. The same b, which are then usually fitted into a linear regression
author also directed various studies on friction welding of equation of the type y ¼ b0 þ b1x1 þ b2x2 þ · · ·, where y is the
austenitic-stainless steels (Sahin, 2005; 2007). Fu and Du response function and the bs are the “estimated” values of the
(2006) examined the effects of external electric field on bs. In simple terms, each coefficient represents the influence
of the corresponding factor on the quality of the weld
microstructure and property of friction welded joint between
expressed by the optimisation parameter.
copper and stainless steel. Jayabharath et al. (2007) directed a
The statistical analysis involves two steps: first, the
study on the continuous drive friction welding of sintered
adequacy of the model is tested. A suitable method is based
powder metallurgical steel and wrought copper parts.
Surface cleanliness in terms of contaminants, especially
grease, reduces the quality of joints. The cleanliness of the Table I Chemical compositions of austenitic-stainless steel used in the
parts must be considered as important as well. Therefore, experiment
the ends of the parts were cleaned prior to the welding Tensile
process to minimize the effect of organic contamination in strength
the welding zone by acetone (Yılbaş et al., 1995). Then, the Material %C %P %S %Mn %Si %Cr %Ni (MPa)
present work was done in the light of previous studies. In this
study, austenitic-stainless steel and copper materials were AISI 304
welded using the friction welding method. The optimum (X5CrNi
parameters were obtained for the joints. The joints were then 1810) , 0.07 , 0.045 ,0.030 ,2.0 ,1.0 17-19 8.5-10.5 825
applied to the tensile and micro-hardness tests. Following Source: Wegst (1995)
this, micro- and macro-photos of the joints were examined.

320
 Joining of stainless steel and copper materials Industrial Lubrication and Tribology
Mumin Sahin Volume 61 · Number 6 · 2009 · 319 –324

Table II Chemical compositions obtained using chemical analysis of copper used in the experiment
Tensile
strength
Copper %Sn %Pb %Zn %P %Mn %Fe %Ni %Si %Mg %Al %Bi %S %Sb %Cu (MPa)
0.00222 ,0.00200 ,0.00100 0.00137 ,0.00050 0.0381 ,0.00100 0.00745 0.00376 0.00500 ,0.00050 0.00251 ,0.00200 99.93 300

Figure 3 Parts used in the experiments

y ¼ 2346:079 þ 6:545 £ x1 þ 6:026 £ x2
d1 = 10 (mm) d2 = 10 (mm)
T, s (rpm) Later, parameters having the least error using the method of
Pf (MPa), tf (sec)
least squares were taken as the optimum welding parameters.
The optimum welding conditions are given in Table IV.
Subsequently, various tests were applied to the welded
parts. The tests are given as tensile tests, micro-hardness tests
Lathe
and metallurgical tests below.
chuck
Copper Austenitic-stainless steel
4. The experimental results and discussion
Figure 4 Continuous drive friction welding set-up
4.1 The tensile tests
Work-pieces Sliding guide
Hydraulic First, the optimum parameters such as the ones shown in
Drive group Table IV were found using statistical analysis for the welded
1 parts. Later, many parts were machined and these parts, using
the optimum parameters were welded. The effects of friction
2 time and friction pressure on the strength of the joints were
Chuck Clamps
examined in welding of equal diameter parts. In experiments,
Clutch Bearings T upset time was kept constant. The strengths of the joints were
Friction 3 P T
pressure determined by tensile tests, and the results were compared
control with those of fully machined joints. The tensile strengths of
Upset the joints were estimated as dividing the ultimate loads by an
M: Motor
P: Pressure pressure area of 10 mm’s diameter. Relation obtained between the
P T
control tensile strength versus the friction time and friction pressure is
T: Tank exit
1: Return valve M
graphically shown in Figures 5 and 6.
2: Pressure indicator
3: Solenoid valve Table IV Parameters used in the friction welding experiments
Oil tank
Friction Friction Upset Upset
Aluminium, AISI 304, time, pressure, time, pressure,
on the Fischer or F ratio, which can be used to confirm if the d1 (mm) d2 (mm) tf (s) Pf (MPa) tu (s) Pu (MPa)
terms in the assumed linear function are statistically
significant. The second step is to obtain optimal estimates 10 10 8.5 75 20 160
of the regression coefficients for the significant factors, which
may be done using the method of least squares (Murti and
Sundaresan, 1983; Sahin and Akata, 2003). Figure 5 Relationship between tensile strength versus friction pressure
Parameter optimisation was carried out using factorial design
Stainless-steel and copper joints
of experiments. The two factors chosen were friction time and
(d = 10 mm, – d = 10 mm)
friction pressure. The other parameters such as upset time, (tf = 8.5 sec, Pu = 150 MPa, tu = 20 sec)
upset pressure and rotational speed were maintained constant.
250
Experimental results for factors are given in Table III.
First, optimal estimates of regression coefficients were
obtained using the method of Fisher ratio. The resulting 200
Tensile strength (MPa)

equation is also given below:

150
Table III Experimental results for factors
Friction pressure, X1 Friction time, X2 Tensile strength
100
Trial No. (MPa) (s) (MPa)
1 54 8.5 30 50
2 66 8.5 203.8
3 75 8.5 223
4 75 6 143 0
0 20 40 60 80 100
5 75 12 190
Friction pressure (Pf-MPa)

321
 Joining of stainless steel and copper materials Industrial Lubrication and Tribology
Mumin Sahin Volume 61 · Number 6 · 2009 · 319 –324

Figure 6 Relationship between tensile strength versus friction time in the parent metals and interface region of the joints are
shown in Figures 9-12.
Stainless-steel and copper joints As it is shown in Figures 7 and 8, we see that the axial
(d = 10 mm, – d = 10 mm)
shortening in copper side is much more than that in the
(Pf = 75 MPa, Pu = 150 MPa, tu = 20 sec)
stainless-steel side. The axial shortening values of the joints
250

Figure 9 Micro-photo of stainless-steel

200
Tensile strength (MPa)

150

100

50

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Friction time (tf-sec)

As the friction time and pressure for the joints are increased as
it is shown in Figures 5 and 6, the tensile strength of the joints
increases. But, the strength of the joints raises a peak in
Figures 5 and 6, and later, when the friction time and
pressure for the joints are increased, the tensile strength of the ×1,000
joints decreases. Here, we can see that the maximum strength
obtained in the joints is about 75 per cent that of copper parts
having the weakest strength (ASM, 1983). Figure 10 Micro-photo of copper

4.2 Microstructure of welded parts

As regards joints, the photo and the macro-photo of the joint
is shown in Figures 7 and 8. Then, the microstructure-photos

Figure 7 Photo of joint

Stainless-
steel Copper

×1,000
×3

Figure 11 Micro-photo of interface region in joints

Figure 8 Macro-photo of joint

Copper Stainless- Stainless-

steel steel
side
Copper
side

×3 ×50

322
 Joining of stainless steel and copper materials Industrial Lubrication and Tribology
Mumin Sahin Volume 61 · Number 6 · 2009 · 319 –324

Figure 12 Micro-photo of interface region in joints Figure 14 Hardness distribution on the horizontal distance of joints
Stainless steel (AISI 304) - copper
(d = 10 mm, –d = 10 mm)
Stainless-
steel Welding interface
Copper side 300
side Steel Copper
275
side side
250
225

Vickers hardness (HV03)

200
175
150
125
x 1,000
100
change between 8.0 and 8.3 mm. However, the stainless steel 75
was almost never deformed because the melting temperature of Base 50 Base
copper material is lower than that of stainless-steel. Therefore, material HAZ zone material
25
the weld flash consists of copper material at the interface.
0
The microstructure of the base metal consists of austenitic
grain structure. There are in relatively small melting and –4 –3 –2 –1 0 1 2 3 4
heat-affected zones developing at the interface of joints. The Horizontal distance (mm)
widths of the melted zone and heat affected zones are about
0.2 mm and 4.5 mm, respectively. The heat-affected zone in
copper side having a higher thermal diffusivity is wider than Figure 15 Hardness distributions along the vertical distance for
that in steel side. different horizontal distances of joints
While no significant variation is observed in the grain size
on the steel side, small grains in the copper side indicate Stainless-steel and copper (d = 10 mm, – d = 10 mm)
hardening. Copper exhibited equalized grains and dispersion Horizontal distance 0.5 mm from joint centre
of Cu2O particles. Constituent elements of both materials had Horizontal distance 1 mm from joint centre
Horizontal distance 1.5 mm from joint centre
interdiffused through the weld interface, and some Horizontal distance 2 mm from joint centre
intermetallic compounds were formed at the weld interface. Horizontal distance 2.5 mm from joint centre
Horizontal distance 3 mm from joint centre

4.3 Hardness variations of welded parts 300

The strength of the joints is related to the hardness variation Steel side 280
within the HAZ. Hardness variation was obtained using 300 g 260
load by Vickers micro-hardness testing and measuring 240
locations are shown in Figure 13. The hardness variations 220
Vickers hardness (HV03)

in horizontal distance and vertical distance to the centre in the 200

welding zone of the joints are shown in Figures 14 and 15.
180
Micro-hardness test results on the horizontal distance of
160
joints are shown in Figure 14. It can be seen that around the
interface, the hardness of the copper increases slightly. 140
As it is shown in Figure 15, the hardness values along 120
100 Copper side
the vertical distance for different horizontal distances on the
stainless-steel side of the joints is higher according to the 80
copper side as it is advanced from the interface both to the end 60
of the parts and the base metals. It can be considered that 40
different thermal diffusivity of materials and intermetallic 20
layer existing at the interface cause hardness variations.
0
–5 –4 –3 –2 –1 0 1 2 3 4 5
Figure 13 Hardness test orientation
Vertical distance to centre of joints (mm)
Joint centre
AISI 304
Copper (midaxis) Vertical distance
to the centre 5. Conclusions
y .
The continuous drive friction welding method can be
suitably adopted for the production of different steels and
x Horizontal distance metals having different thermo physical properties and
to the centre strong welds can be achieved.
.
For optimising welding parameters, statistical analysis is
Welding flash
an economical method.

323
 Joining of stainless steel and copper materials Industrial Lubrication and Tribology
Mumin Sahin Volume 61 · Number 6 · 2009 · 319 –324

.
Optimum welding parameters should be properly selected Jayabharath, K., Ashfaq, M., Venugopal, P. and Achar, D.R.G.
in friction welding of similar and dissimilar materials. (2007), “Investigations on the continuous drive friction
. Tensile strengths for austenitic-stainless steel and copper welding of sintered powder metallurgical (P/M) steel and
parts were considered as positive results when compared wrought copper parts”, Materials Science & Engineering
with those of the base metals. Joint strength increased and A-Structural Materials Properties Microstructure and Processing,
reached a maximum, and then decreased again as the Vol. 454, pp. 114-23.
friction time and friction pressure increased. Sufficient Kinley, W. (1979), “Inertia welding: simple in principle and
heat to obtain a strong joint could not be generated with a application”, Welding and Metal Fabrication, October,
shorter friction time. A longer friction time causes pp. 585-9.
the excess formation of an intermetallic layer. However, Lee, W.B. and Jung, S.B. (1993), “Effect of microstructural
some of the welds show poor strength depending on some variation on the Cu/CK45 carbon steel friction weld joint”,
accumulation of alloying elements at the interface result of Zeitschrift Fur Metallkunde, Vol. 94 No. 12, pp. 1300-6.
temperature rise and the existence of intermetallic layers. Murti, K.G.K. and Sundaresan, S. (1983), “Parameter
.
As it is shown in the micro-photos, it can be seen from the optimisation in friction welding dissimilar materials”,
figures that a maximum mixing of Fe into the Cu region Metal Construction, June, pp. 331-5.
occurs not at the centre of the specimen. The absence of Nentwig, A.W.E. (1996), “Friction welding of cross section of
an intermetallic layer at the centre is consistent with the different sizes”, Schweissen und Schneiden (Welding &
low-relative velocity in that region. It should be noted that Cutting), Vol. 48 No. 12, pp. 236-7.
when friction welding, it is highly probable that the copper Rich, T. and Roberts, R. (1971), “Thermal analysis for basic
oxide film is broken up, allowing the exposure of friction welding”, Metal Construction and British Welding
uncontaminated surfaces as a result large amounts of Journal, pp. 93-8, March.
deformation at the interface occurring. Şahin, A.Z., Yılbaş, B.S., Ahmed, M. and Nickel, J. (1998),
.
As it is shown from hardness figures, while the hardness of “Analysis of the friction welding process in relation to the
the copper according to horizontal distance increases welding of copper and steel bars”, Journal of Materials
slightly, that of the steel slight decreases. This is due to Processing Technology, Vol. 82, pp. 127-36.
the steel reaching the annealing temperature during the Sahin, M. (2005), “An investigation into joining of
welding process, which in turn reduces hardness on the austenitic-stainless steels (AISI 304) with friction
steel side. On the other hand, copper exhibits hardening welding”, Assembly Automation, Vol. 25 No. 2, pp. 140-5.
due to its high-thermal conductivity and fast cooling Sahin, M. (2007), “Evaluation of the joint-interface
behaviour. properties of austenitic-stainless steels (AISI 304) joined
by friction welding”, Materials & Design, Vol. 28 No. 7,
References pp. 2244-50.
ASM (1983), Welding and Brazing, Metals Handbook, 8th ed., Sahin, M. and Akata, H.E. (2003), “Joining with friction
ASM, Metals Park, OH, pp. 507-18. welding of plastically deformed steel”, Journal of Materials
Draper, N.R. and Smith, H. (1981), Applied Regression Processing Technology, Vol. 142 No. 1, pp. 239-46.
Analysis, 2nd ed., Wiley, New York, NY. Sahin, M. and Akata, H.E. (2004), “An experimental study
Ellis, C.R.G. (1977), “Friction welding, some recent on friction welding of medium carbon and austenitic
applications of friction welding”, Welding and Metal stainless steel components”, Industial Lubrication &
Fabrication, May, pp. 207-13. Tribology, Vol. 56 No. 2, pp. 122-9.
Fabritsiev, S.A., Pokrovsky, A.S., Nakamichi, M. and Sluzalec, A. (1990), “Thermal effects in friction welding”,
Kawamura, H. (1998), “Irradiation resistance of DS International Journal Mechanical Science, pp. 467-78.
copper stainless steel joints fabricated by friction welding Vill, V.I. (1962), Friction Welding of Metals, AWS, New York, NY.
methods”, Journal of Nuclear Materials, Vol. 258, October- Wegst, C.W. (1995), Stahlschlüssel, Verlag Stahlschlüssel
Part B, pp. 2030-5. Wegst GmbH, Marbach.
Fomichev, N.I. (1980), “The friction welding of new high speed Yılbaş, B.S., Şahin, A.Z., Kahraman, N. and Al-Garni
tool steels to structural steels”, Welding Production, pp. 35-8. (1995), “Friction welding of St – Al and Al – Cu materials”,
Fu, L. and Du, S.G. (2006), “Effects of external electric field Journal of Materials Processing Technology, Vol. 49 Nos 3/4,
on microstructure and property of friction welded joint pp. 431-43.
between copper and stainless steel”, Journal of Materials
Science, Vol. 41 No. 13, pp. 4137-42.
Corresponding author
Imshennik, K.P. (1973), “Heating in friction welding”,
Welding Production, pp. 76-9. Mumin Sahin can be contacted at: mumins@trakya.edu.tr

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints

324