Optics and Lasers in Engineering 50 (2012) 1621–1627

Contents lists available at SciVerse ScienceDirect

Optics and Lasers in Engineering

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

Fusion welding studies using laser on Ti–SS dissimilar combination

B. Shanmugarajan n, G. Padmanabham
Centre for Laser Processing of Materials (CLPM), International Advanced Research Centre for Powder Metallurgy and New Materials (ARCI), Balapur P.O., Hyderabad 500005, India

a r t i c l e i n f o abstract

Article history: Laser welding investigations were carried out on dissimilar Ti–SS combination. The study is aimed to
Received 6 January 2012 improve the weld strength and ductility by minimizing harmful intermetallics and taking advantage of
Received in revised form high cooling rates in laser welding. Results of continuous wave 3.5 kW CO2 laser welding of totally
24 April 2012
dissimilar combination of Titanium and stainless steel (304) have been discussed. Bead on plate
Accepted 22 May 2012
Available online 22 June 2012
welding experiments were conducted to identify the laser welding parameters using depth of
penetration as criteria. The welding of dissimilar combination has been attempted both autogenously
Keywords: and with interlayers such as Vanadium (V) and Tantalum (Ta) in the form of laser cladding as well as
Laser welding strip. Autogenous welds were carried out by varying the laser power, welding speed and position of the
Titanium
laser beam with respect to the joint centre. The resultant welds are characterized by macrostructure
Stainless steel
analysis, SEM/EDAX and XRD and as welded tensile test in UTM. The autogenous welds have exhibited
Dissimilar materials
extensive cracking even when welded at high speeds or by manipulating the beam position with
respect to the joint. Similarly Vandaium as interlayer could not achieve crack free joint. A joint with
40 MPa strength could be made with Ta as interlayer. Results and analysis of these variants of laser
welded joints are reported and discussed.
& 2012 Elsevier Ltd. All rights reserved.

1. Introduction the size of the intermetallics can be maintained below 10 mm [1,2],

the intermetallics will not embrittle the weld. Such a control in size
Joining of dissimilar materials finds applications in variety of should be feasible with processes which give high cooling rates.
fields such as thermal power station, nuclear industries, petro- Wang et al. [3] studied the feasibility of electron beam welding
chemical industries, cryogenic vessels, micro electronics, medical, of TA15 Ti alloy to SS using a Cu interlayer. They have found that
etc. due to the need to tailor the location of materials according to with dual pass electron beam welding with first welding between
design/performance. However, the difference in thermo physical Ti and Cu sheet and second welding between Cu and SS sheet,
properties of materials being joined can increase the residual cracking could be avoided. However, the intermetallics could not
stress on one side of the joint making the heat affected zone be avoided completely which has led to brittle fracture when
(HAZ) of low expansion material weaker and the difference in stretched. Mousavi and Sartangi [4] have arrived at a suitable
chemistry can lead to formation of brittle intermetallic phases. parametric window both analytically and experimentally for
The difficulty in joining increases with increase in dissimilarity. explosive welding of CP Ti–SS 304. They have found that explo-
For example, the boiler heat exchanger tubes require joining of sive load had a severe impact on both the interface morphology
austenitic to ferritic grade of steels. The materials are not differing in and intermetallic formation and concluded that at low loads
chemistry and metallurgy much but there are thermal properties formation of intermetallics could be totally avoided. Cola et al.
mismatch which can result in cracking during high temperature [5] studied the microstructural features of inertial friction welded
service. Similarly, there are applications requiring totally different Ti–316L SS joint using Nb as interlayer and were able to achieve
materials to be joined like Al–Steel, Titanium–stainless steel, etc. for the desired microstructure and strength for cryogenic application.
variety of applications which differ both in metallurgy and thermo Kamachi Mudali et al. [6] studied the corrosion and bend ductility
physical properties. One such dissimilar combination of Ti–SS is of joint between Ti dissolver and 304L SS tubing by friction and
applied in nuclear industries, cryogenic vessels, etc. Fusion welding explosive welding. They have found out that friction welded
of this totally dissimilar combination will require careful considera- joints have poor bend ductility but good corrosion resistance
tions of the difference in properties. However, it is understood that if and vice versa in explosive bonded joints. Kaharaman et al. [7]
have studied the explosive bonding of Ti to SS with respect to
explosive ratio concentrating on the interface appearance. Lee and
n
Corresponding author. Jung [8] has attempted friction welding of Ti–SS321 and they
E-mail address: [email protected] (B. Shanmugarajan). were able to achieve a joint strength of 400 MPa by altering the

0143-8166/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.optlaseng.2012.05.008
1622 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627

microstructure by varying the friction force. Fuji et al. [9] were for welding and then the butt welding has been carried out
able to increase the bend ductility of friction welded joints by using the optimized parameter obtained from BOP. The clad-
stress relieving by PWHT at 500 1C and rapid cooling. Vollertsen ding was done by preplacing the tantalum powder on both SS
and Grupp [10] were able to braze a thin sheet of titanium to and titanium with PVA as the binder. This was executed with
stainless steel using laser. Still, there is no known or prior report 2 kW power at a speed of 1 m/min and a defocus of 125 mm.
on laser fusion welding of Ti–SS combination. High power density EDS has been done on samples after cladding and polishing to
laser welding will enable welding at high speeds with a cooling check the level of dilution.
rate 410001 C/s creating favorable conditions in dissimilar III. Butt welding with Ta interlayer in the form of a strip with
materials joining through reduction in number and sizes of inter 3.5 kW, 6 m/min and gauss mode. The welding was carried
metallic phases in the fusion zone. out in two steps due to the thickness of the Ta (0.5 mm)
In this paper some of the results of CW CO2 laser welding of interlayer. First the Ta strip was placed and welded to
titanium (Ti) to AISI 304 stainless steel (SS 304) of 3 mm thickness titanium and then the surface was polished and cleaned and
have been presented. In order to evaluate the efficacy of laser again it was welded to SS employing the same parameter.
fusion welding at high speeds, investigations have been under-
taken to join 3 mm thick Ti plate to similar thickness SS304 after The butt welded joints with interlayers have been subjected to
identifying the usable laser power and welding speed through dye penetration testing (DPT) and tensile testing in the as welded
bead on plate studies on stainless steel (bead on plate studies on condition by pulling it in a UTM. The fractured surfaces were
Ti were not done as fusion depth etc. are likely to be more than SS subjected to EDS and XRD to analyze the type of fracture and
304 for the same welding conditions). The results of laser welding composition of the fractured surface of welds.
in both autogenous mode and with interlayers are discussed.

2. Experimental details 3. Results and discussion

The welding experiments were carried out on 3 mm thick CP Ti 3.1. Bead on plate experiments
and SS 304. The base material composition and the strength
characteristics of the material are given in Table 1. The beads on plate experiments were conducted to optimize
Laser welding experiments were carried out using a 3.5 kW the parameter for the maximum cooling rate possible for the
CW CO2 (Rofin Sinar DC035) laser. This laser has a very high beam required depth of penetration of 3 mm. The results have shown
quality (K 40.9) enabling it to be focused to a very fine spot of that the laser power of 3.5 kW and welding speeds of 5 and 6 m/
180 mm using a focusing mirror of 300 mm focal length. All the min (Fig. 1) have yielded the weld with minimum melting for the
experiments were carried out using argon shielding at a flow rate required depth of penetration. The high speed could be achieved
of 30 l/min for both plasma suppression and shielding, in trailing because of the very high beam quality (K40.9) of the laser. Hence,
configuration. Before welding the surfaces to be joined were the same parameter has been frozen for carrying out butt welding.
cleaned using acetone to remove any dirt, oil, grease, etc. Initially
bead on plate (BOP) experiments were conducted on 3 mm thick
3.2. Autogenous butt joints
SS 304 to identify welding parameters to get full penetration. BOP
experiments were not conducted on titanium as the fusion
3.2.1. Autogenous joints with CW output
characteristics such as depth of penetration is more for titanium.
The autogenous butt joints carried out with CW laser at 3.5 kW
Autogenous welding has been carried out using optimized para-
and 6 m/min have resulted in extensive cracking. The formed
meters in both continuous wave (CW) and pulsed mode to obtain
joint broke even while taking the pieces from the welding fixture.
higher cooling rates in the fusion zone. Welding experiment in
The resultant fractured surface has been analyzed by XRD for the
CW mode is also carried out with 0.15 mm beam offset towards
the titanium side and the welding head being tilted 151 towards
titanium side. Butt welding has also been attempted with inter-
layers. The methodology and the interlayers adopted are

I. Welding with V strip of 0.7 mm thickness as interlayer with

laser power of 3.5 kW, speed of 3 m/min and donut mode
(360 mm spot size) to melt both the edges along with the
interlayer.
II. Welding experiments carried out by cladding the abutting
surfaces of SS and Ti with tantalum which was subsequently
polished using a rough emery sheet to have a smooth surface Fig. 1. BOP results on SS304.

Table 1
Chemical composition and mechanical properties of base materials.

Material O2 N2 H2 C Fe Others Ti YS (MPa) UTS (MPa) % El

Cp Ti 0.25 0.028 0.01 0.08 0.25 0.1 Bal 335–545 510–605 21–29

Material Cr Ni Mn C Si S Fe YS (MPa) UTS (MPa) % El

304 19 9 1.8 0.08 o 0.045 0.03 Bal 210 515 40

 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627 1623

phases that formed while welding and the XRD results are shown 3.2.2. Autogenous joints with pulsing the laser beam output
in Fig. 2. The autogenous joints with pulsing have been attempted with
The XRD analysis have shown that there was predominantly a the aim to further enhance the cooling rates. The laser output has
complex phase containing iron, titanium and nickel, i.e. Fe0.2Ni4.8Ti5 been pulsed with the duty cycles of 75% and 60% at an average
on both titanium and SS side with a considerable amount of Cr2Ti power of 2.8 kW and 2.3 kW respectively with the peak power
phase on the SS side and on the titanium side the next predominant remaining at 3.5 kW. Welding with pulsed parameters has
phase was NiTi. However other phases were also present but they yielded the joint which remained intact for quite sometime after
were negligible. The analysis has shown that even such a high removing from the fixture. This can be because of the reduction in
cooling rate cannot alleviate the brittle intermetallic formation overall heat input which would have increased the cooling rates
completely. However the quantification of the size of the inter- further resulting in a marginal reduction in the size of the
metallics formed could not be accomplished but since it is brittle intermetallics. XRD analysis has shown similar phases as dis-
the size of the intermetallics should be above 10 mm. cussed above. EDS analyses (Fig. 3) have also been done on the
fractured joint surfaces at the places which were holding the joint
together and failed at the last. The EDS analyses have shown that
the holding points contained more amount of titanium. However,
the brittle failure have indicated that the enhanced cooling rate
pulsed welding also could not control the size of the intermetal-
lics to less than 10 mm.

3.2.3. Autogenous welds with beam offset

Based on the above results, the welding experiments were
conducted by offsetting the beam by 0.15 mm towards titanium
with the welding head being tilted towards titanium side by 151
to induce more titanium into the welds. Welding has been done
with both CW and pulsed welding parameters as stated above.
However, the results obtained were similar to the welds with
pulsed parameters. The picture of the joint is shown in Fig. 4.
Hence, the hypothesis of producing crack free welds by control-
ling the size of the intermetallics just by enhanced cooling rates
could not be achieved for this combination.

Fig. 2. (a) XRD analyses results on Ti side. (b) XRD analyses results on SS side. Fig. 4. Autogenous joint with cracks.

Fig. 3. EDS analysis of fractured surface.

1624 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627

3.3. Butt welding with interlayers 3.3.1. Butt welds with V strip as interlayer
The joint has been intact while taking out of the fixture and the
Since the autogenous welds have resulted in cracking irre- joint was having sufficient holding strength for quite sometime.
spective of the parameters and the method used to obtain a joint, However, when examined under the microscope the surfaces have
the use of interlayers has become inevitable. The interlayers have shown both longitudinal and transverse cracking in the weld.
been selected based on their solubility with the major elements When the joint was subjected to tensile test in UTM, the joint was
present in both the materials of the combination. Accordingly, the forgiving at a very minimal load applied. EDS analysis of fractured
interlayers which had sufficient solubility in Fe, Ti, Cr and Ni have surface on titanium side has indicated heavy dilution of titanium
been chosen for the experimentation. Vanadium with 100%
solubility in Fe, Cr, Ti and more than 40% solubility in Ni and
tantalum with 100% solubility in titanium and sufficiently good
solubility with the other elements have been chosen as the
interlayers for the study. The results of welding with interlayer
in both cladded form and in the form of a strip are presented
below.

Fig. 5. (a) Joint surface indicating cracks. (b) Cleavage fracture at SS side.
Fig. 7. (a) XRD results with ‘V’ strip interlayer on SS side. (b) XRD results with ‘V’
strip interlayer on Ti side.

Fig. 6. EDS results on the SS fractured surface. Fig. 8. EDS spectrum of cladded and polished Ti surface.
 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627 1625

by iron and vice versa. Also the fractured surfaces have indicated 3.3.2. Cladded Ta interlayer
the clear cleavage type of fracture indicating the presence of The welds formed failed similar to autogenous welds. This can
brittle intermetallic phases. The joint and the fractured surface be due to the dilution of the clad metal by the base metal which
are given in Fig. 5. EDS results are shown in Fig. 6. The XRD results was also indicated by EDS. Even though tantalum was present in
of the fractured surface are given in Fig. 7. The XRD analysis of the quantities of more than 40% in titanium and 25% in SS side after
fractured surfaces has also indicated the presence of brittle polishing, it was not sufficient to avoid the formation of brittle
intermetallic phases indicating V cannot be the preferred intermetallic phases. Figs. 8 and 9 indicate the EDS analysis after
interlayer. cladding and polishing on both Ti and SS side respectively.
The XRD results are indicated in Fig. 10. The XRD analysis of
the fractured surface of the welds on both Ti and SS side revealed
the presence of the brittle intermetallic phases as present in the
autogenous joints. Also the XRD results have indicated the
presence of various oxides indicating the presence of excess
oxygen in the welds. The oxides could have formed while
cladding as both Ti and Ta are very reactive elements. The
presence of Ti, Fe and Ni intermetallic phases also indicated that
the amount of interlayer applied in the form of powder may not
be sufficient to avoid embrittlement. To confirm if Ta is a
preferred interlayer, Ta has been supplied in the form of strip
for carrying out welding.

3.3.3. Butt welds with Ta strip as interlayer

The dye penetrant test has indicated that the surface cracks
were very minimal and from the intensity of the dye wetting the
developer it can be seen that cracks were superficial. The joint
Fig. 9. EDS spectrum of cladded and polished SS surface.
and dye penetrant test results are indicated in Fig. 11.

Fig. 10. (a) XRD analysis of fractured surface on SS side. (b) XRD analysis of
fractured surface on Ti side. Fig. 11. (a) Crack free joint surface. (b) Dye penetrant test indicating no defects.
1626 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627

Fig. 12. (a) Fractured surface on Ti side. (b) Fractured surface on SS side. Fig. 13. (a) XRD results of fractured surface on SS side. (b) XRD results of fractured
surface on Ti side.

The tensile testing of the as welded joint has shown that the effective shielding could not be achieved. Hence, the presence
joint failed at a load of 6.8 kN indicating a strength of 44 MPa of oxides has led to the brittle failure with cladded interlayers.
which confirms that the joint has considerable strength. The SEM The use of interlayers in the form of strip has gradually
analyses of the fractured surfaces (Fig. 12) have shown a mixed improved the joint strength with the Tantalum strip giving
mode of fracture with micro pores indicating that the joint has the best weldability. The joints were showing very less defects
considerable ductility. in the as welded condition and could be subjected to tensile
However, the fracture surface of Ti at the top side has shown testing. However, further work will have to be carried out to
cracks with an enhanced oxygen level. The XRD results of the improve the ductility of the welded joints.
fractured surface are shown in Fig. 13. The XRD results of the
fractured surface have shown that the brittle intermetallic phases
could be eliminated. However, it has also indicated the presence Acknowledgment
of oxide phases. Hence the loss of ductility can be attributed to
the oxygen content at the top surface promoting a brittle fracture I sincerely thank Dr. G. Sundararajan, Director, ARCI for
which slowly changed to a ductile mode of fracture from top to permitting me to carry out this work. I also wish to extend my
bottom. gratitude to Dr. G. Madhusudhan Reddy, DMRL for the support.

References
4. Conclusions
[1] Dharmendra C, Rao KP, Wilden J, Reich S. Study on laser welding-brazing of
1. Autogenous welds of Ti–SS combination could not be achieved zinc coated steel to aluminium alloy with a zinc based filler. Mater Sci Eng A
for all the parameter combinations experimented indicating 2011;58:1497–503.
[2] Lin SB, Song JL, Yang CL, Fan CL, Wang DH. Brazability of dissimilar metals
that cooling rate alone will not avoid the embrittlement of the
tungsten inert gas butt welding between aluminium alloy and stainless steel
weld. This indicates that autogenous fusion welding of Ti–SS with Al–Cu filler metal. Mater Des 2010;31:2637–42.
dissimilar combinations is not feasible. [3] Wang T, Zhang BG, Chen GQ, Feng JC, Tang Q. Electron beam welding of Ti-15-
2. The use of interlayers in the form of laser cladding has not 3 titanium alloy to 304 stainless steel with copper interlayer sheet. Trans
Non-Ferrous Met Soc China 2010;20:1829–34.
given the best results. The interlayer elements are very [4] Mousavi SAA, Sartangi P Farhadi. Experimental investigation of explosive
reactive and hence clads were prone to oxidation as the welding of co-titanium/AISI 304 stainless steel. Mater Des 2009;30:459–68.
 B. Shanmugarajan, G. Padmanabham / Optics and Lasers in Engineering 50 (2012) 1621–1627 1627

[5] Cola MJ, Dickerson RM, Gentzlinger RC, Teter DF. Dissimilar metal joints for [8] Lee WB, Jung SB. Effect of microstructure on mechanical properties of friction
the apt superconducting cavity’s cryogenic plumbing system. In: Proceedings welded joints between Ti and AISI 321 stainless steel. Mater Trans 2004;5(9):
of the particle accelerator conference. New York; 1999. 2805–11.
[6] Kamachi Mudali U, Ananda Rao BM, Shanmugam K, Natarajan R, Raj Baldev. [9] Fuji A, North TH, Ameyama K, Futamata M. Improving tensile strength and
Corrosion and microstructural aspects of dissimilar joints of titanium and bend ductility of titanium/AlSl 304L stainless steel friction welds. Mater Sci
type 304L stainless steel. J Nucl Mater 2003;321:40–8. Technol 1992;8(3):219–35.
[7] Kaharaman N, Gulenc B, Findik F. Joining of titanium/stainless steel by explosive [10] Vollertsen F, Grupp M. Laser beam joining of dissimilar thin sheet materials.
welding and effect on interface. J Mater Process Technol 2005;169:127–33. Steel Res Int 2005;76:240–4.