Engineering Failure Analysis 66 (2016) 8898

Contents lists available at ScienceDirect

Engineering Failure Analysis

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

Failure analysis of tungsten based tool materials used in friction

stir welding of high strength low alloy steels
S. Ragu Nathan a,, S. Malarvizhi a, V. Balasubramanian b, A.G. Rao c
a
b
c

Department of Manufacturing Engineering, Annamalai University, Annamalai Nagar 608 002, Tamil Nadu, India
Department of Manufacturing Engineering, Annamalai University, Annamalai Nagar 608 002, India
Naval Materials Research Laboratory (NMRL), Ambernath, Mumbai 421 506, Maharastra, India

a r t i c l e

i n f o

Article history:
Received 6 November 2015
Received in revised form 11 April 2016
Accepted 12 April 2016
Available online 14 April 2016
Keywords:
Friction stir welding
HSLA steel
Tungsten based alloy
Tool degradation

a b s t r a c t
A non-consumable tool is a vital requirement for friction stir welding (FSW) of high melting
point alloys such as steel and titanium. In this investigation, an attempt was made to understand the pre-weld and post-weld microstructural characteristics of three tungsten based
alloy FSW tools viz. 90%W, 95%W and 99%W. High strength low alloy (HSLA) steel plates of
5 mm thickness were welded using the above tools with a tool rotational speed of 600 rpm
and welding speed of 30 mm/min. Microstructural characteristics of the FSW tools, before
and after welding, were analyzed using optical microscopy (OM), scanning electron microscopy
(SEM) and electron backscatter diffraction (EBSD) techniques. From this investigation, it is
found that the tool made of 99%W doped with 1% La2O3 exhibited microstructural stability
due to absence of FeCoNi phase formation at elevated temperatures during FSW process.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Friction stir welding (FSW) is a promising solid state joining technique for welding of high melting point alloys such as steel,
titanium and nickel based alloys. FSW is known for green, high productive welding technology with lower emission of harmful
gases as compared to fusion welding process [1]. Mechanical action in the form of frictional stirring on the base material modies
the coarse grain microstructures into ne grains due to plastic deformation and fast cooling rate [2]. Welding of steels will be
affected by both the temperature and composition which extensively affects the microstructure evolution. Tool wear and plastic
deformation are the two major problems encountered during FSW of high melting point alloys (L80 steel) [3]. The tool wear is
either due to mechanical damage or chemical afnity of the tool and work piece; however, the plastic deformation is associated
with the variation in stress, strain rate and temperature during FSW. Therefore, the tool must withstand high frictional and resultant forces experienced by the pin during initial plunge stage [4].
Most of the tool failures are reported during plunge stage, thus resulting in poor stirring and non-uniform grain renement of
the parent material in stir zone and thereby, violating the primary advantage of the FSW process itself. However, the tool should
withstand and counter various forces generated at the initial plunge stage along with other factors while tool is allowed to traverse in the weld direction. The tool pin is responsible for plasticizing the stir zone, excavating the softened material from advancing side to retreating side and consolidating beneath it so as to begin the next cycle [5]. Selection of tool material and its
conguration is the key features to avoid interaction gap between consecutive cycles, which eliminates defects like wormholes,
pin hole and tunnel defect at the advancing side [6].
Corresponding author.
E-mail addresses: ragucemajor@gmail.com (S.R. Nathan), jeejoo@rediffmail.com (S. Malarvizhi), visvabalu@yahoo.co.in (V. Balasubramanian),
gouravdrdo@gmail.com (A.G. Rao).

http://dx.doi.org/10.1016/j.engfailanal.2016.04.018
1350-6307/ 2016 Elsevier Ltd. All rights reserved.

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

89

Jiye et al. [4] explained the tool wear mechanism of WLa tool and advocated the use of conical pin with large pin length as
compared to smaller one for better stability [7]. Hence, the taper conical larger pin was mostly recommended. During the tool
transverse motion, the tool shoulder should support the pin by generating optimum heat to plasticize the parent material
which in-turn reduce the ow stresses. This supportive heat generation should be uniform throughout the weld length. The
variation in ow stresses during welding will lead to pin damage resulting in poor weld quality. Consolidations of the material
extruded are secondary function of the shoulder [8,9].
The tool distortion such as expansion or contraction, rubbing wear and if any one of the situation prevails will lead to poor
weld quality, loss of tool pin, and severe plastic deformation [10]. Much of the tool degradation may be attributed to the high
heat (temperature around 1200 C) and stresses generated during FSW of steel. The brittle tendency of the poly-cubic boron nitride (PCBN) tool used for FSW of steel and titanium alloys can cause tool breakage due to sudden spike in load or vibration during tool plunging and traversing [11]. Apart from the tool material selection, researchers had also investigated other possibilities
to reduce tool wear and breakage such as pilot-hole and partial penetration. However, the problems are not solved yet completely. Recent developments in the tool material, tool design and processing strategies fetters feasibility of friction stir welding of high
melting temperature materials.
Tool cost is also considered as an important factor, which restricts the application of FSW technology for steels. However, the
tungsten base alloys are cheaper than WRe/WReHfC/PCBN tools. Therefore, tough tungsten base alloys were selected as FSW
tool materials for this investigation. Tungsten has a high energy threshold for physical sputtering and offer distractive physical
properties which include the highest melting point of all metal, the lowest vapor pressure, good thermal conductivity and high
temperature strength and toughness [1214]. Also, these alloys contain a small amount of impurities which are known to increase
the high temperature creep resistance. As the tool experiences high temperature and large forces during FSW, it is appropriate to
investigate the microstructural characteristics of tools before and after welding to quantify the tool degradation. Hence, in this investigation, high strength low alloy (HSLA) steel was friction stir welded using tungsten based tool materials to check the feasibility and employability.
2. Experimental
The rolled plates of 5 mm thick high strength low alloy (HSLA) steel were used as the base material. The microstructural features of the base metal are shown in Fig. 1 and it primarily consists of ferrite with a small amount of pearlite. The three different
grades of tungsten based alloy were used as the tool materials in the present investigation to weld HSLA steel. The tool was produced by powder metallurgy route and supplied by M/s. Heavy Alloys Penetrator Project (HAPP), Tiruchirappalli, India. The chemical composition and the mechanical properties of base metal and tool materials are presented in Tables 1a, 1b and 2, respectively.
Tool materials are designated as W90, W95 and W99 for the purpose of convenience.
The tools were machined to the dimensions and conguration as shown in Fig. 2. Using these three tools, HSLA steel plates
were joined by FSW process. The process was done in worst-case set-up that is, without using shielding gas and without
pre-heating the base metal. Tool rotational speed of 600 rpm, welding speed of 30 mm/min, plunge rate of 2.5 mm/min and
with an axial force of 15 kN welding parameters were used to fabricate the joints. To measure the temperature elds, four Ktype thermocouples of 1 mm diameter were embedded on the advancing side of the plate at a depth of approximately 1.5 mm
from the surface. Thermocouples were placed at 14 mm from the weld center as shown in Fig. 4. Six thermocouples were connected to the rst six channels of the DAQ to collect temperature. Thermocouples were attached to a DAQ system which could
measure the data at 15 Hz. Data collection was done with DAQ system that was attached to the computer running Lab VIEW
software.
After 1 m length of welding, the tools were sliced for post-weld metallographic examination. Specimens were extracted and
sectioned to the required sizes from the tool materials before and after the weld and then polished using different grades of
emery papers. Final polishing was done using the diamond compound (0.1 m of particle size) in the disc polishing machine.

Fig. 1. SEM micrograph of base metal.

90

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

Table 1a
Chemical composition (wt.%) of the base metal.
C

Mn

Si

Mo

Ti

Nb

Ni

Cu

Al

0.08

1.42

0.015

0.19

0.02

0.016

0.032

0.035

0.69

0.126

0.032

0.04

Polished specimens were then etched with a solution of ammonia mixed with a small amount of hydrogen peroxide to reveal the
microstructure of the tool specimens. Microstructural examination was carried out using an optical microscopy (OM) and electron
backscattered diffraction (EBSD) to reveal the micrograph and to detect phases present in the tool material. The phases present in
the pre- and post-weld tool material were conrmed by X-ray diffraction and the results are presented in the following sections.
3. Results
3.1. Thermo-mechanical analysis
The plastic deformation of the FSW tool pin is controlled by the ow stresses and temperature. In Fig. 3, effects of tool plunge
on axial load and temperature are presented. In Fig. 3a it can be understood that the lower axial load was required during plunge
which might be due to the larger contact area and resultant higher frictional heating and vice versa. The peak temperature recorded during FSW of HSLA steel is shown in Fig. 3b. From this observation, it is clear that the increase in tool plunge rate will reduce
the frictional heat generated during FSW and in turn it affects the material ow around pin. Once material ow is affected, it leads
to a rise in stress to the pin which causes severe plastic deformation.
3.2. Microscopic observations
Fig. 4 displays the photographs and macrographs of all the three tungsten based alloy tools used in this investigation after friction stir welding 1 m length of HSLA steel plates of 5 mm thickness. W90 tool experienced maximum deformation and thus
ended in mushrooming effect. The deformation and tool wear were observed in pin and shoulder regions. This suggests that recrystallization occurred in the tool material well before its actual recrystallization temperature due to ow stresses generated in
the base material during FSW. On the other hand, W95 tool showed better performance compared to W90 tool but the damage
on the top region of the tool pin was observed. Pin damage was clearly shown in Fig. 4. In contrast, severe deformation, tool
wears and pin damage were not observed in the W99 tool. The dimensional stability of the W99 tool and no wear or deformation
were observed.
Optical micrographs of the pin and the shoulder regions of the tools before and after welding are depicted in Fig. 5. Grain
growth and grain coarsening were observed in all the tools after FSW. Grain coarsening, W/matrix separations were observed
at the pin region (Fig. 5b) and the compressed grains were observed at the shoulder region (Fig. 5c). In Fig. 5ac, it is clear
that recrystallization occurred in the tool material well before its actual recrystallization temperature. This was due to the instability of the tungsten matrix of the W90 tool to withstand the thermo-mechanical stresses. Fig. 5d shows the W95 tool raw stock
microstructure (before FSW) consisting of ne equiaxed tungsten grains merged with the matrix. W-cleavage failure was observed in the tool pin region (Fig. 5e) and a conned amount of compression of W/W grains was also witnessed at the shoulder
region (Fig. 5f). The raw microstructure (before FSW) of W99 tool is shown in Fig. 5g and it consists of elongated grains with
uniformly dispersed oxide at the grains and grain boundaries. The lanthanum oxide particles pins the grain boundary during
high temperature deformation and thus restrict the grain growth during FSW. The better thermo mechanical stability of W99
tool material at elevated temperatures is evident in Fig. 5h & i. The microstructure of the tool pin and the shoulder region
after FSW is more or less similar to raw stock microstructure (before FSW).
3.3. EBSD analysis
3.3.1. Image quality mapping
Fig. 6 presents the electron backscattered diffraction (EBSD) analysis of pre- and post-weld tool pin top region. Fig. 6a and b
represent the microstructure of W90 tool pin top region and reveals the severe deformation of the tool. Coarsened tungsten grains
are clearly visible in the quality map (Fig. 6b) and it illustrates that the W90 tool had undergone hot deformation at peak temperature and ow stresses. W95 tool pin sustained the forces and temperature generated during the FSW and it is evident from
Table 1b
Chemical composition (wt.%) of the tool materials.
Tool Designation

Ni

Fe

Mo

Co

La2O3

W90
W95
W99

90.11%
95.09%
99%

5.96%
3.01%

3.56%
1.73%

0.22%
0.11%

0.15%
0.06%

1%

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

91

Table 2
Mechanical properties of the base metal and tool materials.
Material

0.2% offset yield strength

(MPa)

Ultimate tensile strength

(MPa)

Elongation in 50 mm
gauge length (%)

Hardness of base
metal @ 0.5 kg load (HV)

BM
W90
W95
W99

540
850
1190
1210

610
1100
1400
1475

24
20
14
11

270
570
610
506

the presence of a dense microstructure (Fig. 6c) compared to raw stock microstructure (Fig. 6d). W99 tool has sufcient microstructural stability to restrict the grain coarsening at higher working temperature and it is clear in Fig. 6e & f.
3.3.2. Phase identication
FeCoNi phase formation in W90 tool pin top region is clearly evident in phase analysis (Fig. 7a). But, a higher amount of Fe
CoNi phase formation was observed in the W95 tool pin top region (Fig. 7b) compared to W90 tool pin. Since W99 tool is doped
with 1% La2O3, FeCoNi phase was not observed in the phase analysis (Fig. 7c). Due to the reasons, the W90 and W95 tools are
excluded from the analysis of grain boundary misorientation and elemental mapping but both the analyses were carried out on
W99 tool.
3.3.3. Grain boundary misorientation
Fig. 8 displays the grain boundary misorientation of the pre- and post-weld W99 tool. High angle grain boundaries and low
angle grain are observed in the tool. Fig. 8a and b present the grain boundary mapping. In that, green color represents low
angle grain boundaries (LAGB) and black color represents high angle grain boundaries (HAGB). Presence of HAGBs was observed
away from the pin and as well as near the pin and it is evident for substructure stability of the material during FSW. A fraction of
HAGB's was not reduced drastically compared with observation made away from the pin (Fig. 8c). It is attributed to tool material
substructure stability at higher forces and temperatures involved during FSW process on steels.
3.4. Evaluation of transfer layer
Adhesion between tool and the base metal is high in W90 and W95 tools compared to W99 tool (Fig. 4). Thickness of adhered
layer increases the stress concentration at the tool pin corner and lead to cracking, followed by pin failure. Wear resistivity greatly
reduced due to this stress induced crack tip. Whereas no adhesion of workpiece material was observed on W99 tool surface.
Instead of adhesion, tool and workpiece interaction zone named as reaction zone (a thin layer on tool pin) is evident and it inuences the tool life by reducing severity of grain deformation as shown in Fig. 9a. Deformation is not observed in W99 tool
pin and this may be due to the acceptable range of deformed and un-deformed grains at the pin top corner to sustain the tool
conguration. To identify the formation of new phases or layer present on the tool pin top region, backscattered SEM analysis
was performed at tool cross-section. The analysis revealed the presence of transfer layer (Fig. 9a). To reveal the details of
phase formation on the tool surface, X-ray diffraction (XRD) analysis was carried out and the results are presented in Fig. 9b & c.
Black dots on the pin top surface are observed as work piece material. It was veried and conrmed through XRD analysis
(Fig. 9b & c). XRD result conrmed the presence of iron and nickel on the W99 tool pin top surface. From these analyses, it is
understood that though the lanthanated tungsten tool (W99) is capable of withstanding high forces and high temperatures during
FSW of HSLA steel, there is a possibility of transfer of elements from the workpiece to tool by diffusion process.

Fig. 2. Photograph of tool and tool conguration.

92

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

Fig. 3. Axial load and temperature prole of the FSW joint.

4. Discussion
The selected tool material should possess good mechanical properties at elevated temperature, durable at high torque, high
hardness to resist wear and good thermal conductivity. The favorable material ow around the tool pin can be catered by a suitable tool material selected particularly by considering the base metal properties [15]. It may inuence the tool life, weld quality
and cost of the tool material too. Ability of the tool in FSW process is to lead the material to root of the weld and to plasticize the
material under the tool without change in dimension/conguration in the tool prole. In fact, as pointed out by Thompson and

Fig. 4. Photographs of tungsten based alloy tool before and after welding.

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

93

Fig. 5. Optical micrographs of tungsten based alloy tools before and after FSW process.

Babu [16], tool degradation in FSW of hard metals is inuenced by ow stresses and welding temperatures that were generated
during frictional stirring. Inuence of axial load (Fig. 3) at a particular period of time should not affect the pin conguration before
plasticizing the parent metal. It may have inuence on tool life, weld quality and cost of the tool material also.
The formation of mushroom effect, severe deformation and adhesive wear shown in the macrostructure (Fig. 4) and microstructure (Fig. 5b), could be attributed to insufcient and inadequate thermal stability of the W90 tool. This is due to mismatch
in thermal coefcients of W-grains and the matrix phase. It was also reported by Kim et al. [5], that intermediate temperatures
lead to reduce the contiguity of W-grains mainly due to thermal coefcient variations of W-grain and the matrix phase. From
the post-weld W90 tool micrographs, the W/matrix decohesion, matrix tearing and W-cleavage mechanism were revealed and
this could be the reasons for the inferior performance of W90 tool at elevated temperatures. The results displayed in Table 2 indicate that the W90 tool material has superior mechanical properties at room temperature. However, the high temperature during
the FSW (nearly 1200 C) along with high forces could have melted the matrix, due to the presence of softer FeNiCo phase. It
was also investigated and reported by Mabuchi et al. [10] that a smaller amount of grain boundary segments does not necessarily
lead to high strength at higher temperatures. The inferior mechanical properties of W90 tool material operating at higher temperatures were reported by Jiten et al. [3] and Yuvan et al. (2013). They reported that application of tungsten heavy alloy as a tool
material is restricted mainly due to ductile to brittle transition temperature (DBTT) and recrystallization temperature of the matrix phase.
Compared to W90 tool material, W95 tool has superior strength, toughness, elongation and hardness (Table 2b). Because of
the increased mechanical properties, W95 tool pin has the better load bearing ability but galling occurred between the tool
and the work piece interaction (Fig. 4). This is mainly due to high chemical afnity between the base metal and the tool material.
The reason for failure of W95 tool is the increased sticking torque and local deformation caused by adhered material on the tool.
The thick layer formation around the tool pin resulted in the increase of local compressive stresses on the pin and led to failure.
However, microstructural result shows that the failure occurred due to W/matrix decohesion and matrix tearing. The local stress
generated by the increase in sticking torque could have increased the stress concentration at the corners of the tool pin and hence
failure occurred. At the same time, due to the compression of vertical force and resultant forces from the base metal are also likely
to reduce the densication of the matrix and hence crack propagated through the matrix phase. Densication was also greatly
reduced in the W95 tool revealed by pore formation (Fig. 6c).

94

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

Fig. 6. Image quality maps of FSW tools.

Therefore, it is expected that W99 tool (WLa2O3) has a great potential for carrying out FSW due to higher tungsten content
with lanthanum oxide particles pinning the grain boundaries and higher toughness. The microstructural features before and after
FSW are almost similar, conrming the microstructural stability during FSW. The stability of the microstructure is mainly due to
the presence of lanthanum oxide particles at the grain boundaries. The ultimate tensile strength of W99 (W1%La2O3) material at
1000 C was approximately 500 MPa, reported by Ghosh et al. (2014). Moreover, presence of lanthanum oxide particles acts as a
site for dislocation generation as well as impeding the dislocation movement thus having good strength and toughness. Mabuchi
et al. [10] investigated and reported that dislocation movement was greatly inuenced by particle interaction and substructure
formed in the grains due to uniform distribution of La2O3. Yinzhong et al. [17] also reported that secondary grain growth and

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

95

Fig. 7. Phase mapping of FSW tools.

increase in recrystallization temperature are impeded due to ne dispersion of lanthanum oxide particles in the matrix and thereby result in high strength and toughness at high working temperatures.
Deformation of the grains in the pin region of FSW tool is vital in preventing the tool wear and degradation. The same was
achieved by the formation of a protective transfer layer on both sides of the pin tip as shown in Fig. 9a. The protective transfer
layer was formed from the base metal to the tool pin surface during interaction of the tool with the base metal at high temperature and load. The protective transfer layer conrmed the presence of W, Fe, Ni and C through pre- and post-weld tool analyses

96

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

Fig. 8. Grain boundary misorientation results of W99 tool.

by XRD (Fig. 9b and c). This thin protective transfer layer acts as a good thermal barrier around the tool pin region and thus assists in reducing the tool wear and deformation. Furthermore the recrystallization temperature of W99 tool is around 1600 C
which is more than the temperature experienced by the tool during FSW of HSLA steel. Therefore, as compared with other W
based tools (W90 and W95), W99 tool has better stability at elevated temperature (Fig. 3) and can be used for FSW of HSLA
steel plates of 5 mm thickness.
5. Conclusions
In this investigation, an attempt was made to analyze the microstructural changes that occurred in tungsten base tool materials during friction stir welding of HSLA steels. The important conclusions are given below.
i. Of the three tungsten base tool materials used in this investigation, W99 (W1%La2O3) tool exhibited better microstructural
stability without undergoing physical (dimensional) changes in tool conguration.
ii. The W/matrix de-cohesion, matrix tearing and W-cleavage in the W90 and W95 tools are mainly due to formation of FeCo
Ni phase (softer phase) in the matrix.

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

Fig. 9. Micrograph and XRD results of W99 tool.

97

98

S.R. Nathan et al. / Engineering Failure Analysis 66 (2016) 8898

iii. It was found that the tool made of 99% W and 1% La2O3 withstood high strain rate, temperature and ow stresses generated
during FSW of HSLA steel. This is mainly due to the microstructural stability of the tool materials and formation of protective
transfer layer during FSW of HSLA steel.

Acknowledgments
The authors are grateful to The Director, Naval Materials Research Laboratory (NMRL), Ambernath for the nancial support
through CARS project No: G8/15250/2011 dated 29.02.2012 and providing the base material for this investigation. The authors
would like to express their sincere thanks to Mr. V. K. Iyyer, Works Manager and Mr. S. K. Selvam, Junior Works Manager,
Heavy Alloy Penetrator Project (HAPP), Trichy for providing tool materials to carry out this investigation. The authors are thankful
to Dr. A. K. Lakshminarayanan, Associate Professor, S.S.N. Engineering College, Chennai for his valuable suggestions and support
throughout this investigation.

References
[1] W. Gan, Z.T. Li, S. Khurana, Tool materials selection for friction stir welding of L80 steel, Sci. Technol. Weld. Join. 12 (2007) 610613.
[2] M. Gosh, K. Kumar, R.S. Mishra, Analysis of microstructural evolution during friction stir welding of ultrahigh-strength steel, Scr. Mater. 63 (2010) 851854.
[3] A.N. Jiten Das, G. AppaRao, S.K. Pabi, M. Sankaranarayana, T.K. Nandy, Thermo-mechanical processing, microstructure and tensile properties of a tungsten heavy
alloy, Mater. Sci. Eng. A 613 (2014) 4859.
[4] Jiye Wang, Su Jianqing, Rajiv S. Mishra, Ray Xu, John A. Baumann, Tool wear mechanisms in friction stir welding of Ti6Al4V alloy, Wear 321 (2014) 2532.
[5] D.K. Kim, S. Lee, J.W. Noh, Dynamic and quasi-static torsional behavior of tungsten heavy alloy specimens fabricated through sintering, heat-treatment, swaging
and aging, Mater. Sci. Eng. A 247 (1998) 285294.
[6] K. Kumar, Satish V. Kailas, The role of friction stir welding tool on material flow and weld formation, Mater. Sci. Eng. A 485 (2008) 367374.
[7] K. Lakshminarayanan, V. Balasubramanian, An assessment of microstructure, hardness, tensile and impact strength of friction stir welded ferritic stainless steel
joints, Mater. Des. 31 (2010) 45924600.
[8] S. Malarvizhi, V. Balasubramanian, Influences of tool shoulder diameter to plate thickness ratio (D/T) on stir zone formation and tensile properties of friction stir
welded dissimilar joints of AA6061 aluminumAZ31B magnesium alloys, Mater. Des. 40 (2012) 453460.
[9] E. Novillo, D. Hernandez, I. Gutierrez, B.L. opez, Analysis of ferrite grain growth mechanisms during - transformation in a niobium alloyed steel using EBSD,
Mater. Sci. Eng. A 385 (2004) 8390.
[10] M. Mabuchi, K. Okamoto, N. Saito, T. Asahina, T. Igarashi, Deformation behavior and strengthening mechanisms at intermediate temperatures in WLa2O3, Mater.
Sci. Eng. A 237 (1997) 241249.
[11] G. Madhusudhan Reddy, T. Mohandas, Ballistic performance of high-strength low-alloy steel weldments, J. Mater. Process. Technol. 57 (1996) 2330.
[12] S. Mandal, J. Rice, A.A. Elmustaf, Experimental and numerical investigation of the plunge stage in friction stir welding, J. Mater. Process. Technol. 203 (2008)
411419.
[13] G. Padmanaban, V. Balasubramanian, Selection of FSW tool pin profile, shoulder diameter and material for joining AZ31B magnesium alloy an experimental
approach, Mater. Des. 30 (2009) 26472656.
[14] Sachin L. Shrestha, Andrew J. Breen, Patrick Trimby, Gwnalle Proust, Simon P. Ringer, Julie M. Cairney. An automated method of quantifying ferrite microstructures using electron backscatter diffraction (EBSD) data, Ultramicroscopy 137 (2014) 4047.
[15] Preetish Sinhaa, S. Muthukumaran, S.K. Mukherjeeb, Analysis of first mode of metal transfer in friction stir welded plates by image processing technique, J. Mater.
Process. Technol. 197 (2008) 1721.
[16] B. Thompson, S.S. Babu, Tool degradation characterization in the friction stir welding of hard metals, Weld. Res. 89 (2010) 256261.
[17] Yinzhong Shen, Zhiqiang Xu, Kai Cui, Jie Yu, Microstructure of a commercial W1% La2O3 alloy, J. Nucl. Mater. 455 (2014) 234241.