Journal of Manufacturing Processes 16 (2014) 296–304

Contents lists available at ScienceDirect

Journal of Manufacturing Processes

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

Technical Paper

Comparison of mechanical properties of pure copper welded using

friction stir welding and tungsten inert gas welding
Jau-Wen Lin a,∗ , Hsi-Cherng Chang a , Ming-Hsiu Wu b
a
Department of Mechanical Engineering, National Kaohsiung University of Applied Sciences, 415 Chien-Kung Road, Kaohsiung 807, Taiwan, ROC
b
Metal Industries Research & Development Centre, 1001 Kaonan Highway, Kaohsiung 811, Taiwan, ROC

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

Article history: The objective of this research is to investigate the mechanical properties including bonding, tensile
Received 13 March 2013 strength, and impact resistance of pure copper welded using friction stir welding (FSW) method and
Received in revised form compare them with that of tungsten inert gas (TIG) welding. Micro-hardness tests are performed on pure
12 September 2013
copper, TIG welded copper and FSW welded copper to determine the effect of heat on the hardness of
Accepted 19 September 2013
welded coppers. Tensile strength tests and notch tensile strength tests are performed to determine the
Available online 15 November 2013
mechanical properties of different weld process.
In this experiment, it is found that the notch tensile strength and the notch strength ratio for FSW
Keywords:
Friction stir welding (FSW)
(212 MPa, 1.10) are significantly higher than those (190 MPa, 1.02) of TIG welding. For the impact tests,
Tungsten inert gas (TIG) welding the weld zone and heat-affected zone energy absorption values for FSW (2.87 J, 2.25 J) are higher than
Pure copper those (1.32 J, 0 J) of TIG welding. XRD tests are performed to determine components of copper before and
Impact toughness after welding process for TIG and FSW.
© 2013 Published by Elsevier Ltd on behalf of The Society of Manufacturing Engineers.

1. Introduction welding copper since 1940s. However, the welding of copper is usu-
ally difficult because copper has a high thermal diffusivity, which
Pure copper is a tough, malleable, and ductile metal. Copper is about 10–100 times higher than those of many steel and nickel
metal has been utilized to manufacture many things since the alloys. The heat dissipates faster than almost any other material
beginning of human civilization. Nowadays, the consumption of thus results in low welding speed. To overcome these problems,
copper in quantity is the second among all kinds of metal. Copper friction stir welding (FSW), a solid-state welding technique, has
is a good conductor of electricity and heat, thus it is widely used as become a better option to TIG [2,10].
the electric conducting and wiring material for electronic devices, Compared to traditional welding method or TIG, FSW has a
communication products and power transmission lines. Due to its lower heat input, less welding distortion, zero pollution, and higher
excellent heat conductivity, copper is also used in many appliances welding performance. Therefore, it is now widely used for welding
such as refrigerators, evaporators, and heating coils. It is a situation of aluminum, magnesium, copper, zinc, lead, low-carbon steel and
which occurs frequently in the product manufacturing processes to aluminum matrix composites [3,4]. The thickness of welds varies
weld pieces of copper together. from 1 to 50 mm and applications range from small cooling compo-
However, pure copper is a challenge for traditional welding pro- nents to ship construction [5,6]. Welds with a thickness of 50 mm
cess like brazing because oxygen is present during the extraction of for nuclear waste disposal canisters have been achieved [7]. Pre-
pure copper. At above 400 ◦ C, hydrogen atoms in the reducing gas vious studies [8–11] have investigated the effect of welding speed
rapidly diffuse into the solid metal and react with the Cu2 O to form on the microstructure and mechanical properties of pure copper. It
H2 O. Sufficient local pressure is built up by the vapor that is formed is found that the grains decreased and became finer with increas-
through the aforementioned reaction and creates internal holes; ing welding speed, resulting in enhanced mechanical properties.
these holes make a porous structure which reduces malleability Khodaverdizadeh et al. [12] used hardening and strain harden-
and strength [1] of the material. ing indices to explore the influence of stir revolutions and feed
Tungsten inert gas (TIG) welding is an inexpensive welding pro- speed on the strain hardening behavior of pure copper. He et al.
cess that produces good quality welds. It has been adopted for [13] used FSW and TIG welding to investigate the annealing of
cold-pressed and hot-pressed plates of the alloy Al–Mg–Mn–Sc–Zr.
A comparison of the welding properties and microstructure indi-
∗ Corresponding author. Tel.: +886 7 381 4526x5346; fax: +886 7 392 0392. cated that the hardness of the alloy with FSW was higher than
E-mail addresses: [email protected], [email protected], that of the alloy with TIG welding. Liu et al. [14] studied the TIG
[email protected] (J.-W. Lin). welding of magnesium alloys and investigated the characteristics

1526-6125/$ – see front matter © 2013 Published by Elsevier Ltd on behalf of The Society of Manufacturing Engineers.
http://dx.doi.org/10.1016/j.jmapro.2013.09.006
 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304 297

of metal compounds at the joined surface. They showed that the

metal compounds become brittle after welding process. Squillac
et al. [15] compared the electrochemistry characteristics of welds
created using TIG welding and FSW. The present study primarily
focuses on investigating the mechanical properties of pure cop-
per with FSW and TIG welding, including the microstructure of the
welding joint, tensile strength, and impact resistance.

2. Experimental methods

The experiment apparatus layout for TIG is shown in Fig. 1a, the
FSW apparatus picture is shown in Fig. 1b. The operation parame-
ters for TIG are listed in Table 1. Fig. 1c is the schematic plot of the
FSW welding layout. For FSW, the ambient temperature of opera-
tion is 25 ◦ C. The operation parameters are described later in the
text.

Fig. 2. (a) The dimensions of V notch cut for TIG. (b) The dimensions of friction stir
tool head.

Pure copper (C11000) was processed to the desired dimen-

sions of 300 mm × 50 mm × 3 mm. A V shape notch was then made
between the joint of two pure copper plates for TIG welding, as
shown in Fig. 2a. In this experiment, the filler wire was fed man-
ually for the welding process. In order to increase the strength
of the weld, two-pass welding was used. To avoid distortion and
rapid cooling of the metal during the welding process, the copper
plates were fixed and pre-heated before welding. Oxygen-free cop-
per wires with a diameter of 2.4 mm were used as filling material
and high-purity argon gas was used as shielding gas. The welding
parameters are listed in Table 1. The stir tool used in this process is a
JIS SKH55 standard stir head, the dimension of it is shown in Fig. 2b.
The stir head is machined into a cone shape in order to obtain a
better welding result. The FSW process is affected by the rotational
speed and translational speed of the stir tool, and friction tool pres-
sure (axial force). In the FSW experiment, the rotational speed of
the stir head is 1100 rpm, the translational speed is 25 mm/min, the
axial force exerts on the stir head is 10 kN, and the backward incli-
nation angle of the stir head is 2.5◦ . The shoulder of the rotating
tool was pushed below the level of the copper surface for 0.2 mm.
After welding, three tensile test specimens were cut out from
each welded plates in accordance with ASTM E8M-9 [16] standards
by longitudinal metal production, as shown in Fig. 3a and b. Fig. 3c is
the specimen dimensions for notch tensile strength test. The num-
ber of specimens for tensile strength and notch tensile test are 3 for
base metal, 3 for TIG, and 3 for FSW, respectively. The base metal
was used to compare two specimens. The unnotched specimen was
used to analyze yielding strength, tensile strength, cross-sectional
Fig. 1. (a) Schematic drawing of TIG. (1) AC/DC welder, (2) argon gas supply, (3) elongation, and reduction. The notched specimen was used to ana-
water supply, (4) welding torch, (5) filler wire, (6) base metal, and (7) foot operated lyze the notch tensile strength (NTS) and notch strength ratio (NSR).
current control. (b) Photograph of friction stir welding. (c) The schematic plot of the
of FSW welding layout
A 100-kN tensile tester (Instron, 5582, USA) was used for the tensile
test.
298 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304

Table 1
Criteria of tungsten inert gas welding.

Base metal and size (mm) Filler material and Tungsten rod Welding Welding built-up Pre-heating Voltage (V) Current (A) Post-welding
dimension (mm) extrusion distance pass sequence temperature (◦ C) temperature (◦ C)
(mm) number

Copper 300 × 50 × 3 Oxygen-free 6 1 1 170 21.6 212 143

copper 2.4
Copper 300 × 50 × 3 Oxygen-free 6 1 1 230 21.6 212 182
copper 2.4

Welding length (mm) Welding time (s) Welding velocity Tungsten rod Tungsten rod Welding gas Welding gas flow Ceramic ferrule
(mm/min) materials diameter (mm) rate (l/min) number

300 110 163.64 Thorium 3.2 Ar 8 6

300 149 120.81 Tungsten 3.2 Ar 8 6

A part of the weld specimens were cut for micro-hardness test

and microstructure observation. The cutting plane is perpendicular
to the welding line.
The specimens are prepared using sandpaper with grit sizes
ranging from #200 to #1500. Subsequently, aluminum oxide par-
ticles (with sizes of 1, 0.5, and 0.3 ␮m, in sequence) were adopted
for polishing. A mixture of 5 g of Fe(NO3 )3 , 25 ml of HCl, and 75 ml
of H2 O was used for cleaning the specimen’s surface and to remove
the oxidation films on the surface.
A micro-hardness tester (Mitutoty, KK810-969K, Japan) was
adopted to measure the micro-hardness with a load of 100 g. The
micro-hardness specimens were cut off along the welding line (cen-
tral line of the specimen) from the welded plate with a width
of 40 mm. The micro-hardness measurement is performed on 29
points along the line perpendicular to the weld line on the speci-
men as shown in Fig. 4. The distance between each measuring point
is 1 mm.
An optical microscope (Nikon, MM-400/LM, Japan) was used to
observe the microstructure of the weld zone (stir zone), the heat-
affected zone and the base metal. The impact specimens were made
in accordance with ASTM E23-05 [17] standards and for each spec-
imen, the specimens can be divided into base metal, weld zone,
and heat-affected zone, as shown in Fig. 5. The energy absorption
Fig. 4. Microhardness measuring positions.
and toughness were compared and scanning electron microscopy

(SEM, Hitachi, S3000N, Japan) was adopted to observe the charac-

teristics of surface failure. The pendulum mass of the impact tester
(OTTO, D-6700, Germany) was W = 19.3 kg, the distance between
the axis center and sample center was R = 0.79791 m, and the drop
angle was ˛ = 160◦ .
Observations of the surface structure at different experimental
conditions were carried out using an X-ray diffractometer (XRD,
Rigaku, Japan). The exam parameters are: a Cu K␣ radiation with
 = 0.15418 nm, 2 range of 10–70◦ and scan rate of 3◦ /min.

3. Results and discussion

3.1. Microstructure

The hardness values of micro-hardness test are shown in Fig. 6.

The hardness of pure copper after FSW is higher than that after
TIG welding. The difference in hardness for FSW weld zone and
TIG weld zone is 20 HV (Vickers hardness). In the figure, there is a
small variation in hardness along the line, with the highest hardness
on two sides which are base metal zone; the hardness gradually
decreases toward the weld zone at center. Fig. 7a shows the pos-
itions of optical microscope observation on the FSW weld zone.
Fig. 7b–d shows microsturcure from top to bottom of the stir zone.
Evidence of considerable microstructural refinement of the top of
Fig. 3. Dimensions of tensile test specimens for (a) base metal, (b) TIG and FSW the stir zone was seen relative to microstructures at the center and
specimens, and (c) notched specimens. the bottom area. The changes in the microstructure, consistent with
 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304 299

Fig. 5. Positions base metal, weld zone, and heat-affected zone.

110
Hardness(HV)

90
FSW
TIG
70

50
-14 -12 -8 -4 0 4 8 12 14

Distance from the center (mm)

Fig. 6. Microhardness test values.

the development of a strain gradient, are a function of distance from testing points are on the plane perpendicular to the welding line.
the tool shoulder. The workpiece material closest to the tool shoul- Fig. 8b is the hardness values. The difference in hardness is due to
der is subjected to higher strain than that away from tool shoulder the difference in the grain size. Hardness gradually decreases from
[10]. From these images, the grain sizes at the top, the center, and top to bottom at these four points due to the variation in grain
the bottom areas can be estimated to be 110.5, 130, and 150 ␮m, size.
respectively. Fig. 9a–c shows the microstructures of the base metal, friction
The microhardness test of 4 points in the FSW weld zone is stir for FSW (a and b), and TIG welds (c). The base metal has rela-
presented in Fig. 8. Fig. 8a is the location of testing points. The tively coarse grains. Due to the effect of the stirring motion by the

Fig. 7. Micro-hardness test positions of FSW weld zone from the top to the bottom on the plane perpendicular to the welding line and the images of microstructure at
corresponding positions (indicated by characters).
300 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304

3.2. Tensile characteristics

Tensile testing was adopted to analyze the yielding strength,

tensile strength, elongation, area of reduction, NTS and NSR. Three
specimens were used to measure each property to obtain a mean
value. The results are shown in Table 2. The yielding strength and
tensile strength of the base metal were 68 and 212 MPa, respec-
tively, and those of the TIG weld were 53 and 168 MPa, respectively.
There is about a 20% reduction in these values for TIG welding. The
friction stir weld had a yielding strength of 70 MPa and a tensile
strength of 194 MPa, which are 24.3% and 13.4% higher than those
of the TIG welding. The values of elongation and area of reduction
for the base metal were 28.1% and 13%, those for the TIG weld were
12.3% and 5.7% (56% lower than those of base metal), and those for
the friction stir weld were 22.8% and 10.5% (19% lower than those of
base metal), respectively. The ductility of the friction stir weld was
37% higher than that of the TIG weld. For the notched tensile spec-
imens, the base metal, TIG weld, and friction stir weld had tensile
strengths of 247, 182, and 212 MPa, respectively. The notch tensile
test results indicate that the tensile strength of the friction stir weld
Fig. 8. Microhardness test positions on the plane perpendicular to the welding line
was 16% higher than that of the TIG weld.
(a) and their hardness values (b). The distance between adjacent measuring points
is 0.5 mm. The two welds had NSRs of above 1, indicating that the C11000
pure copper notched specimens were ductile materials. The NSRs
of the base metal, friction stir weld, and TIG weld were 1.17, 1.10,
and 1.02, respectively. Therefore, the NSR of the friction stir weld
stir tool, the metal plastic flow in FSW destroyed the coarse grains, is 8% higher than that of the TIG weld. The joint efficiency of the
creating a fine, isometric, and evenly distributed stir zone [18]. The friction stir weld is 13% higher than that of the TIG weld.
rapid heating and rapid cooling of molten metal caused TIG welding Fig. 10 shows the fracture locations of unnotched tensile spec-
to produce elongated grains. imens. The fracture line for the parent metal distribute in wide
The grain structure images show that for TIG welding, there is range. For FSW, the fracture lines are in the weld zone but the loca-
a clear boundary between the welded zone and the heat affected tion of fracture line is random suggesting the strength of material in
zone (Fig. 9d). The grain size in the weld zone is obvious large, this the welding zone is almost equal and is lower than that of unstirred
results in lower hardness and yielding strength in the weld zone. zone. For TIG weld, the fracture lines are located at the center of the
However, due to the fine grain size in the heat affected zone, the specimens where the coppers experience high temperature during
elongation of TIG welding is low (Table 2). welding process.

Fig. 9. Microstructure of the base metal, friction stir for FSW (a and b), TIG welds (c) and the boundary of the welded zone and the heat affected zone (lower left) for TIG (d).
 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304 301

Table 2
Mechanical properties of base metal and other welded metals.

Join type Yielding Ultimate Elongation (%) Reduction in Notch tensile Notch Impact Joint efficiency (%)
strength tensile cross sectional strength strength toughness
(MPa) strength area (%) (MPa) ratio (J)

Base metal 68 212 28.1 13 247 1.17 1.67

TIG 53 168 12.3 5.7 172 1.02 Weld zone Heat affect zone 79
1.32 0
FSW 70 194 22.8 10.5 212 1.10 2.87 2.55 92

the welding spot through the slot. The relative humidity of the air
was above 85%, this provides abundant source of hydrogen. Since
the temperature exceeded 400 ◦ C during the welding process of
pure copper, pores were generated due to the rapid diffusion of
hydrogen atoms, which reacted with Cu2 O to produce H2 O:

Cu2 O + H2 → 2Cu + H2 O (1)

When the reaction takes place at temperatures above the crit-

ical temperature of water, the copper oxide inclusion is replaced
with steam, which is virtually insoluble in copper. The resulting
steam pockets or bubbles exert pressure on the surrounding metal,
producing bubble or gas pockets. The oxide inclusions that reacted
with hydrogen to form steam are generally associated with grain
Fig. 10. Photograph of fracture locations for tensile test specimens. boundaries; thus, continued bubble development tends to force the
grains apart. Bubble coalescence ultimately causes grain boundary
fracture [20,21].
Figs. 11–13 show SEM images of the surface for the base metal, Comparing Fig. 12b–d with Fig. 11b–d, FSW produced smaller
friction stir weld, and TIG weld specimens, respectively. The failure and finer dimples because the direct contact of the stir tool with the
of the specimens is considered to be ductile failure under the effect base material caused the base material to undergo plastic deforma-
of a tensile load [19]. The size of dimples varied with the welding tion. The recystallization of grains was due to the energy produced
approach. TIG welding produced the largest dimples, which were from the stirring process. The base metal specimens in Fig. 11b–d
cone-shaped. In the welding process, there is a 0.5 mm gap between have the largest voids, which are material defects. This suggests
the two plates (Fig. 2a) at the welding line, the moisture air reached that copper contained considerable oxygen when it was cast. The

Fig. 11. (a) SEM image of surface failure for base metal tensile test at low magnitude, (b–d) are magnified images of regions 1–3 marked in (a). (b–d) Same magnification.
302 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304

Fig. 12. (a) SEM image of surface failure for stir weld tensile test at low magnitude, (b–d) are magnified images of regions 1–3 marked in (a). (b–d) Same magnification.

Fig. 13. (a) SEM image of surface failure for TIG tensile test at low magnitude, (b–d) are magnified images of regions 1–3 marked in (a). (b–d) Same magnification.
 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304 303

Fig. 15. Photograph of cracks for impact specimens. (1) Base metal, (2) weld zone
of FSW, (3) heat-affected zone of FSW, (4) heat-affected zone of TIG weld, and (5)
weld zone of TIG weld.

mechanical behavior of the welds and the influence of heat loss

should be studied [23]. The present study adopted two approaches
for the weld zone and the heat-affected zone for the specimens, as
shown in Fig. 15. The experimental data represent the mean value
of three specimens. The energy absorption data for each group
of impact specimens are: base metal (1.65 J, 2.05 J, 1.3 J); weld
zone of friction stir weld (2.55 J, 2.6 J, 2.5 J); heat-affected zone of
friction stir weld (2.55 J, 2.6 J, 2.5 J); weld zone of TIG weld (1.2 J,
1.65 J, 1.1 J); and heat-affected zone of TIG weld (0 J, 0 J, 0 J). The
mean values are listed in Table 2. The non-welded specimens had
an impact toughness value of 1.67 J. The impact toughness of the
TIG weld was 1.32 J, a significant reduction of 21% as compared
to the non-welded specimens. The mean value of impact ductility
for the friction stir weld was 2.87 J, which is approximately 72%
higher than that of the non-welded specimens. During the friction
stir process, the recrystallization process produces finer grains;
hence, higher impact energy is absorbed. Fig. 12b–d shows more
dimples with a larger size in the interior of the TIG specimens. The
C11000 copper welding process involves hydrogen. The formation
of water vapor affects the welding strength by reducing the impact
Fig. 14. XRD patterns of the copper at (a) base metal, (b) FSW and (c) TIG weld.

oxygen is present in the form of Cu2 O, which forms a eutectic struc-

ture with Cu. Hot rolling breaks up the eutectic structure, making
Cu2 O discrete [1,22].
Fig. 14a–c shows the XRD pattern measured in the pure copper
FSW and TIG weld. As shown in Fig. 14a, sharp peaks at around
2 = 43◦ and 51◦ shows copper exist in the base metal, the intensity
of Cu2 O is very low due to the low concentration of oxygen (0.04%);
however, Fig. 14b and c shows sharp peaks not only copper and
2 = 37◦ have a few Cu2 O exist in the weld zone. The larger intensity
of Cu2 O is due to more oxygen is introduced in the welding process.
Although for TIG welding, the welding spot is shielded by argon
gas, however, some moisture could reach the welding spot from
beneath the gap between two plates.

3.3. Impact toughness

Although tensile testing can determine the ductility and brittle-

ness of materials, it is considered a static test. Defects on the surface
of a material are subjected to an axial force during the tensile
Fig. 16. (a) SEM cross-sectional image of impact at weld zone for FSW, position 1 is
process, which reduces the toughness of the material. Therefore,
the upper surface where the circular tool marks from stir tool can be seen, position
the results underestimate the actual impact values. In order to 2 is lower surface; (b) SEM cross-sectional image of impact test at weld zone for TIG
obtain the expected strength and toughness of the material, the weld, the upper side is the weld surface.
304 J.-W. Lin et al. / Journal of Manufacturing Processes 16 (2014) 296–304

toughness for TIG welding. The low ductility of copper was also [3] Sun YF, Fujii H. Investigation of the welding parameter dependent microstruc-
attributed to occluded hydrogen gas bubbles existing at high ture and mechanical properties of friction stir welded pure copper. Mater Sci
Eng A 2010;527:6879–86.
pressure [24]. [4] Mishra RS, Ma ZY. Friction stir welding and processing. Mater Sci Eng R
Fig. 16a is the SEM image of fracture surface of FSW specimen 2005;50:1–78.
of impact test, necking at the cross section was observed. The angle [5] Shtrikman MM. Current state and development of friction stir welding. Part 3.
Industrial application of friction stir welding. Weld Int 2008;22:806–15.
between the fracture surface and the side surface is about 45◦ shows [6] Ericsson M, Sandstrom R. Influence of welding speed on the fatigue of friction
the copper in the weld zone is ductile. Fig. 16b shows shrinkage stir welds, and comparison with MIG and TIG. Int J Fatigue 2003;25:1379–87.
holes and cracks exist in the TIG specimens. In the test, the material [7] Cederqvist L, Oberg T. Reliability study of friction stir welded copper canisters
containing Sweden’s nuclear waste. Reliab Eng Syst Safe 2008;93:1491–9.
became torn due to uneven cooling. The uneven cooling was a result
[8] Shen JJ, Liu HJ, Cui F. Effect of welding speed on microstructure and mechanical
of the unidirectional flow of air used to carry away the heat. In the properties of friction stir welded copper. Mater Des 2010;31:3937–42.
heat-affected zone, the impact toughness values for the FSW and [9] Xue P, Xie GM, Xiao BL, Ma ZY, Geng L. Effect of heat input conditions on
microstructure and mechanical properties of friction stir welded pure copper.
TIG specimens were 2.55 and 0 J, respectively, indicating that the
Metall Mater Trans A 2010;41A:2010–21.
TIG weld could not absorb the impact energy in the heat-affected [10] Lee WB, Jung SB. The joint properties of copper by friction stir welding. Mater
zone. Lett 2004;(58):1041–6.
[11] Wang XJ, DA CB, Li J, Zhang ZK. Processing of friction stir welding of red copper
and analysis of its joint properties. J Lanzhou Univ Technol 2006;32:25–8.
4. Conclusion [12] Khodaverdizadeh H, Mahmoudi A, Heidarzadeh A, NazarE I. Effect of friction
stirwelding parameters on strain hardening behavior of pure copper joint.
The mechanical properties of pure copper welded with FSW and Mater Des 2012;35:330–4.
[13] He ZB, Peng YY, Yin Zin ZM, Lei XF. Comparison of FSW and TIG welded joint in
TIG welding were investigated. Several important conclusions were Al-Mg-Mn-Sc-Zr alloy plates. Trans Nonferr Metals Soc 2011;21:1685–91.
obtained: [14] Liu L, Wang S, Zhao L. Study on the dissimilar magnesium alloy and copper lap
joint by TIG welding. Mater Sci Eng A 2008;476:206–9.
[15] Squillac A, De Fenzo A, Giorleo G, Bellucci F. A comparison between FSW and
1. The tensile strength of a friction stir weld is 13.4% higher than TIG welding techniques: modifications of microstructure and pitting corrosion
that of a TIG weld. resistance in AA2024-T3 butt joints. J Mater Process Technol 2004;152:97–105.
2. The hardness of a friction stir weld is 30 HV higher than that of [16] Standard test methods for tension testing of metallic materials, ASTM. E8M;
2009.
a TIG weld. [17] Standard test methods for notched bar impact testing of metallic materials,
3. For FSW, the grain size at stir zone is smallest due to the rapid ASTM. E23; 2005.
dissipation of heat. The grain size at heat affected zone is finer [18] Kwon YJ, Shigematsu I, Saito N. Mechanical properties of fine-grained alu-
minum alloy produced by friction stir process. Scr Mater 2003;49:785–9.
than that of base metal but large than the stir zone. The dimple
[19] Lakshminarayanan AK, Balasubramanian V, Elangovan K. Effect of welding pro-
sizes of FSW weld zone are small than that of TIG. cesses on tensile properties of AA6061 aluminium alloy joint. Int J Adv Manuf
4. In this experiment, the weld efficiency for TIG is 79%, while it is Technol 2009;40:286–96.
[20] Liu P, Li YJ, Geng HR, Wang J. Microstructure characteristics in TIG welded joint
92% for FSW.
of Mg/Al dissimilar materials. Mater Lett 2007;61:1288–91.
[21] Louthan Jr MR. Hydrogen embrittlement of metals: a primer for the failure
References analyst. J Fail Anal Prevent 2008;8:289–307.
[22] Smith WF. Principle of materials science and engineering. 3rd ed. New York:
McGraw Hill; 1998. p. 542–3.
[1] Henkel D, Pense AL. Structure and properties of engineering materials. 5th ed.
[23] Rodrigues DM, Menezes LF, Loureiro A. The influence of the HAZ softening on
New York: McGraw Hill; 2001. p. 304–7.
the mechanical behaviour of welded joints containing crack in the weld metal.
[2] Scutelnicu E, Birsan D, Cojocaru R. Research on friction stir welding and tung-
Eng Fract Mech 2004;71:2053–64.
sten inert gas assisted friction stir welding of copper. In: 4th international
[24] Nakahara S, Okinaka Y. The hydrogen effect in copper. Mater Sci Eng A
conference on manufacturing engineering, quality and production systems.
1988;1014:206227–30.
2011. p. 97–102.