Short Communication

Effect of post-weld heat treatment on the microstructure and plastic

deformation behavior of friction stir welded 2024
Zhili Hu, Shijian Yuan, Xiaosong Wang

, Gang Liu, Yongxian Huang

School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, PR China
a r t i c l e i n f o
Article history:
Received 25 February 2011
Accepted 19 May 2011
Available online 25 May 2011
a b s t r a c t
To improve the plasticity of friction stir welded joints for plastic processing applications, post-weld heat
treatment (PWHT) of 2024-O aluminum alloy friction stir welding joints was carried out at annealing
temperatures from 250 C to 450 C with an interval of 50 C for 2 h, followed by cooling to 200 C in
the furnace. The effect of PWHT on the microstructure and plastic deformation behavior of the joints
was investigated. It was found that the ne-equiaxed grains are stable and retained in the nugget of
the joints even after annealing at 450 C for 2 h. However, the grains in the thermo-mechanically affected
zone (TMAZ) of the joints become coarse and equiaxed as annealing temperature increases. The plastic
deformation of as-welded joint is very heterogeneous. In contrast, the plastic deformation of PWHT joint
is relatively homogeneous by both the nugget and the base material showing large deformation. The
decrease in elongation of as-welded joints is completely recovered by PWHT. The high ductility of the
joint is mainly attributed to the retention of the ne-equiaxed grains in the nugget during PWHT.
2011 Elsevier Ltd. All rights reserved.
1. Introduction
Due to economical and ecological reasons as well as to improve
product properties, aluminum alloy welds, which show high pro-
duction efciency and exibility, have gained more interest in
manufacturing lightweight structures for automotives and aircrafts
[13]. The industrial application of aluminum alloy welds in light-
weight constructions is basically connected to one fundamental
manufacturing requirement. The formability of welds must be ade-
quate in order to shape the required geometries [4].
It is well known that high-strength aluminum alloys (2XXX ser-
ies) are widely used for aerospace structures such as fuselage, ns
and wings [5]. Unfortunately, such high-strength aluminum alloys
are difcult to join by conventional fusion welding due to the
occurrence of hot cracking during welding [6]. Friction stir welding
(FSW), as a solid state metal joining technique, do prove itself ideal
for creating better quality joints compared to fusion welds [68].
However, many studies on the transverse tensile properties of
FSW joints indicated that the elongations of as-welded joints are
only 2040% those of the BM for the heterogeneous microstructure
of the joint [6,911].
To improve the mechanical properties of FSW joints, one option
is to fully post-weld heat treatment (PWHT) of the welded compo-
nents. Currently the PWHT mainly consists of solution heat treat-
ment and precipitation or aging treatment, aiming to recover the
loss of tensile strength in the nugget zone [1215]. A few studies
on the PWHT of FSW aluminum alloys such as 7449 [14], 6061-O
[12], 2219-O [13] and 2024-T4 [15] had shown that the PWHT
recovered a large portion of the lost strength in the nugget, but
there was an accompanying loss in ductility. The joints have a
4077% decrease in elongation compared to that of base material
(BM). In addition, studies in this area have shown that the ne
grains in the nugget are not stable during PWHT, which results
in undesirable coarsened grain structures [1220]. This kind of
microstructural instability has been identied as abnormal grain
growth (AGG). Although in a few studies AGG has proved to be
benecial to recover the lost strength of joint, it is not at all desir-
able for enhanced plasticity [1719]. Ductility falls off sharply due
to extreme microstructural instability [18,19]. Accordingly, it is
necessary to nd out some effective remedial process to improve
the elongation of joints, and avoid AGG occurring in the nugget
at same time.
Normally the friction stir welds are used in the as-welded con-
dition or with stabilization aging when BM is in the hardened con-
ditions (T3, T4 and T6 tempers). However, there are some
advantages in carrying out the welding with BM in a soft condition.
The forming operation after FSW can be much more easily per-
formed if the welding is conducted under O condition. Further-
more, the solution and aging heat treatment can restore the
mechanical properties of the joints successfully after the forming
operation [1215,20].
The present letter analyzes the microstructure and plastic
deformation behavior of FSW aluminum alloys to evaluate if, with
the right selection of post-weld annealing treatment parameters,
their plasticity and strength could be optimized.
0261-3069/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2011.05.035

Corresponding author. Tel.: +86 451 86414761; fax: +86 451 86414751.
E-mail address: hitxswang@hit.edu.cn (X. Wang).
Materials and Design 32 (2011) 50555060
Contents lists available at ScienceDirect
Materials and Design
j our nal homepage: www. el sevi er . com/ l ocat e/ mat des
2. Experimental procedures
Some 5 mm-thick 2024 aluminum alloy rolled plates were FSW
with a travel speed, a rotational speed and a shoulder diameter of
300 mm/min, 600 rpm and 14 mm, respectively. Then the plates
were cut into two parts, one part for PWHT and the other for as-
welded examinations. The PWHT included annealing at different
temperature from 250 C to 450 C with an interval of 50 C for
2 h, followed by cooling to 200 C in the furnace.
Following FSW and PWHT, The orientation-imaging microscope
(OIM) was attached to an S-570/HITACHI-4700 scanning electron
microscope (SEM) operating at 25 kV. The OIM was used to evalu-
ate grain size distribution with step size of 1 lm. The Vickers
micro-hardness test was carried out along the centerlines of the
cross-sections with a distance between neighboring measured
points of 1 mm under a load of 100 g for 10 s. The transverse ten-
sile specimens were prepared with reference to China National
Standard GB/T2651-2008 [21] (equivalent to ASTM B557-02
[22]). The room temperature tensile tests were performed at a
crosshead speed of 2 mm/min using an Instron-1186 testing
machine. An axial extensometer with 25 mm gage length was at-
tached to the test specimens at the gage section. The strain analysis
of each specimen was made by the ASAME

automatic strain mea-

suring system, and the tensile properties of the joints were evalu-
ated using three tensile specimens cut from the same joint.
3. Results and discussion
3.1. Microstructures
The OIM images of joints in the transition zone from the nugget
to thermo-mechanically affected zone (TMAZ) are shown in Fig. 1.
The nugget of as-welded joint is composed of ne-equiaxed recrys-
tallized grains with average grain size of 1.7 lm as seen in Fig. 1a.
After PWHT, average grain size in the nugget of the 300 C an-
nealed joint and the 450 C annealed joint are1.8 lm and 2.0 lm,
respectively. It indicates that the ne-equiaxed grains are stable
in the nugget during PWHT (Fig. 1b and c) and no AGG is observed
even the annealing temperature up to 450 C. However, the grain
Fig. 1. EBSD showing the microstructure of the FSW joints: (a) as-welded, (b) 300 C, (c) 450 C.
Fig. 2. Back scattered SEM images: (a) as-welded joint, (b) 450 C annealed joint, (c) EDS spectra of the A particle and (d) EDS spectra of the B particle.
5056 Z. Hu et al. / Materials and Design 32 (2011) 50555060
size and shape in the TMAZ vary signicantly with the annealing
temperature. The initial grains in the TMAZ are rotated and elon-
gated in an upward owing pattern around the nugget before
PWHT (Fig. 1a). Such phenomenon can also be observed in the
TMAZ of the 300 C annealed joint (Fig. 1b), but nearly all the
grains in the TMAZ of the joint become coarse and equiaxed after
annealing at 450 C (Fig. 1c). It indicates that the rotated and elon-
gated grains in the TMAZ of joints tend to be coarse and equiaxed
with the increasing annealing temperature.
The back scattered SEM micrograph in the nugget of FSW joint
is shown in Fig. 2. There are two different particles in the as-weld
joint, marked by A and B (Fig. 2a). The Energy Dispersive Spectrom-
etry (EDS) analysis showed that these particles are compounds of
AlCuMg and AlCuFeMn (Fig. 2c and d). According to the lit-
eratures these compounds are most likely Al
2
CuMg and (Cu, Fe,
Mn) Al
6
[23,24]. As seen in Fig. 2a, the grain size of the strengthen-
ing precipitates such as rod or lath shaped Al
2
CuMg are 0.2
0.5 lm, and also it is shown that some of the particles present in
the as-welded joint dissolve and slightly coarsen during PWHT.
As seen in Fig. 2b, a much-reduced density of particles with the
grain size of 0.52 lm is observed after annealing at 450 C.
It is known that strengthening precipitates such as rod or lath
shaped Al
2
CuMg can coarsen or dissolve during high temperature
heat treatment [6], while particles, including Fe and Mn, are insol-
uble in aluminum alloys [25]. Therefore all particles, which can be
seen in SEM image after PWHT, are undissolved Al
2
CuMg and
insoluble (Cu, Fe, Mn) Al
6
particles. It is reported that a dispersed,
thermally stable precipitates may help to control the AGG behavior
[16]. Consequently joints do not show any sign of AGG during
PWHT and the ne-equiaxed grains are retained in the nugget.
3.2. Micro-hardness
The micro-hardness proles of the joints before and after PWHT
can be seen in Fig. 3. The maximum micro-hardness value of the
as-welded joint can reach two times that of the BM. It is known
that the hardness in the precipitation-hardening aluminum alloys
greatly depends on the grain size and precipitate distribution
[26]. The nugget of the as-welded joint is composed of the recrys-
tallized ne-equiaxed grains (Fig. 1a) for experiencing intense
plastic deformation and frictional heating during FSW [7]. There-
fore, the joint is strengthened by FSW process.
The micro-hardness values of PWHT joints vary depending on
the annealing temperature. The micro-hardness value in the nug-
get of PWHT joint is slightly higher than that of the BM after
annealing at 300 C, but when the annealing temperature is above
300 C, the micro-hardness values of the PWHT joints are almost
same with that of the BM. It indicates that the micro-harnesses val-
ues in the nugget of the PWHT joints decrease with the increasing
annealing temperature. It is mainly due to the dissolving and the
coarsening of precipitates resulted from the thermal history during
PWHT (Fig. 2).
3.3. Tensile properties and fracture surfaces
The room temperature tensile properties are summarized in Ta-
ble 1. The as-welded joint possesses the same tensile strength as
that of the BM. However, the elongation of the joint is 12.5%, which
shows a 44% reduction in elongation compared with that of the
BM. After PWHT, the elongation of the joints has been signicantly
improved, and increases with decreasing annealing temperature.
The maximum elongation of the joints is up to 20.2% after anneal-
ing at 300 C, equivalent to 1.6 times that of the as-welded joint.
Furthermore, all joints show nearly the same tensile strength as
that of the BM at different annealing temperature, as shown in
Table 1. It can be attributed to the presence of a dispersed, stable
Fig. 3. Micro-hardness prole at mid thickness of the joints before and after PWHT.
Table 1
Tensile properties of the joints (BM: base material).
Samples Elongation
(%)
Tensile
strength
(MPa)
Yield
strength
(MPa)
Fracture
locations
of joints
Annealed joints
450 C 10.7 197 86 Nugget/TMAZ
400 C 14.7 197 81 Nugget/TMAZ
350 C 18.9 198 80 Nugget/TMAZ
300 C 20.2 206 85 BM
250 C 18.7 206 91 BM
As-welded joints 12.5 206 92 BM
BM(O-temper) 22.2 207 88
Z. Hu et al. / Materials and Design 32 (2011) 50555060 5057
precipitates, which increases strength by pinning grain boundaries
and by also interfering with dislocation motion [27]. These results
indicate that the plasticity of as-welded joint can be improved by
PWHT effectively with no loss of tensile strength.
An examination of the tensile fracture surfaces does provide
useful information pertaining to the role and contribution of the
intrinsic microstructural features on strength and ductility of the
joints [9,15]. The fracture surfaces of the tensile tested specimens
are characterized using SEM. The SEM photographs are taken from
the center regions of the tested specimens, and the pictures are
presented in Fig. 4. The fractures of joints are predominantly trans-
granular. As seen in Fig. 4a, an array of randomly ne microscopic
cracks and a population of voids of varying size and shape are dis-
tributed through the fracture surface of as-welded joint. Cracked
second-phase particles and shallow dimples are found covering
the transgranular fracture region.
Such phenomena can also be observed in fracture faces of
PWHT joints. However, The depth and width of the observed dim-
ples in fracture face of the 300 C annealed joint (Fig. 4b) are more,
compared with the as-welded joint (Fig. 4a) and the PWHT 450 C
joint (Fig. 4c). It is reported that the size of the dimples in the frac-
ture surfaces could be indicative of the ductility of FSW joints [9].
The size of the dimples in the fracture surface of joint is large. It
suggests that a large stretch zone is present at the tip of the crack.
This led to a large plastic zone ahead of the crack and rather high
ductility [10]. Consequently, the 300 C annealed joint shows high-
er ductility than the as-welded joint and the 450 C annealed joint.
As the dimples increase in size (Fig. 4), the elongation of the joints
also increases (Table 1).
3.4. Plastic deformation behavior
The microstructure variation occurring through the weld results
in a heterogeneous strain distribution when a tensile test is carried
out normal to the welding direction, as shown in Fig. 5. The retreat-
ing side and advancing side of each joint are denoted by RS and AS,
respectively. As shown in Fig. 5a, the BM on the advancing side lo-
cally elongated to high levels of strain (over 30%), eventually
resulting in necking and fracture, whereas the strain of the nugget
and the BM on the retreating side experience only 4% and 10%,
Fig. 4. SEM micrographs showing fracture surface features of joints: (a) as-welded joint, (b) 300 C annealed joint and (c) 450 C annealed joint.
Fig. 5. Strain maps for the as-welded and annealed joints: (a) as-welded, (b) 300 C and (c) 450 C.
5058 Z. Hu et al. / Materials and Design 32 (2011) 50555060
respectively. It shows that regional deformation of as-welded joint
is highly inhomogeneous. It is mainly because the strength of the
nugget is higher than that of the BM for the grain size strengthen-
ing (average grain size in the nugget is 1.7 lm). Thus deformation
in the nugget is small during tensile test. It can be concluded that
the highly inhomogeneous deformation throughout the joint,
caused by heterogeneous microstructure, is the reason for the
44% decrease in elongation of the as-welded joint compared with
that of the BM.
The strain map of the 300 C annealed joint is shown in Fig. 5b.
The plastic deformation concentrates in the BM on the advancing
side with the local strain exceeding 30%. The average strain of
the nugget and the BM on the retreating side is about 8% and
12%, respectively. It indicates that the deformation inhomogeneity
of the joint has been improved by both the nugget and the BM
showing large strain, especially, the average strain in the nugget
is two times that of the as-welded joints. The slightly harder nature
of the nugget promotes plastic deformation of the BMduring trans-
verse tensile tests, up to fracture in this zone. Therefore, the joint
shows relatively homogeneous deformation and the elongation of
the joint increases. The high ductility of the PWHT joint is mainly
attributed to the retention of the ne-equiaxed grains in the nug-
get during PWHT [28].
The strain map of the 450 C annealed joints on the retreating
side is shown in Fig. 5c. The distribution of the strain for the
450 C annealed joints is completely different. The deformation
localizes in the nugget and TMAZ. This localization extends until
fracture, which happens at the nugget/TMAZ boundary. The frac-
ture locations of joints change with the annealing temperature,
which can be explained by the grain coarsening in the TMAZ after
annealing at 450 C (Fig. 1c). The interface between the nugget and
the TMAZ is more clearly visible and becomes a weaker region
resulting in the fracture along the interface as seen in Fig. 5c. When
fracture happens, local strain in the interface is only 17%. However,
the fracture strain of the as-welded joint and the 300 C annealed
joint are over 30%, indicating that the fracture toughness of the
450 C annealed joint is low. The decrease in fracture toughness
can be correlated to the grain coarsening in the TMAZ and the de-
crease of precipitates in the nugget after annealing at 450 C
[12,15]. This low fracture toughness results in a much lower duc-
tility of the joint.
4. Conclusions
In this study, the effect of PWHT on the microstructure and
plastic deformation behavior of the joints was investigated, and
the following conclusions are obtained.
(1) The decrease in elongation of as-welded joints is completely
recovered by PWHT. The maximum elongation of the PWHT
joint is 1.6 times that of the as-welded joints with no loss of
tensile strength.
(2) The AGG are not observed in the joints during PWHT at 250
450 C, and the ne-equiaxed grains are retained in the nug-
get of the joints. However, the grains in the TMAZ of the
joints coarsen with increasing annealing temperature. The
interface between the nugget and TMAZ become more
clearly visible and results in the fracture along the interface.
(3) Tensile fractures of PWHT joints are predominantly trans-
granular, and fracture surface is covered with dimples of
varying size and shape, reminiscent of locally ductile failure.
The depth and width of the observed dimples in fracture face
of the 300 C annealed joint are more, compared with the as-
welded joint. As the dimples increase in size, the elongation
of the joints also increases.
(4) The micro-hardness value in the nugget of PWHT joint is
slight higher than that of the BM after annealing at 300 C.
When annealing temperature is above 300 C, the micro-
hardness values of PWHT joints are almost same with that
of the BM. The micro-hardness of the PWHT joints decreases
with the increase of annealing temperature.
(5) The plastic deformation of as-welded joint is very heteroge-
neous, but the plastic deformations of the PWHT joints are
relatively homogeneous by both the nugget and the BM
showing large strain. The slightly harder nature of the nug-
get promotes plastic deformation of the BM during trans-
verse tensile tests, up to fracture in the BM.
(6) A follow on work is planned to study the effect of post-weld
heat treatment on mechanical properties of FSW joints in
order to improve joint strength.
References
[1] Yuan SJ, Han C, Wang XS. Hydroforming of automotive structural components
with rectangular-sections. Int J Mach Tool Manuf 2006;46:12016.
[2] Garware M, Kridli GT, Mallick PK. Tensile and fatigue behavior of friction-stir
welded tailor-welded blank of aluminum alloy 5754. J Mater Eng Perform
2010;19:116171.
[3] Miles MP, Decker BJ, Nelson TW. Formability and strength of friction-stir-
welded aluminum sheets. Metall Mater Trans A 2004;35A:34618.
[4] Merklein M, Wieland M, Staud D. Friction stir welds made out of precipitation
hardenable aluminum alloys-experimental investigations of formability
potential. Int J Mater Form 2009;2:3236.
[5] Heinz A, Hasler A, Keidel C, Moldenhauer S, Benedictus R, Miller WS. Recent
development in aluminum alloy for aerospace applications. Mater Sci Eng A
2000;280:1027.
[6] Genevois C, Deschamps A, Denquin A, Doisneau-cottignies B. Quantitative
investigation of precipitation and mechanical behaviour for AA2024 friction
stir welds. Acta Mater 2005;53:244758.
[7] Nandan R, DebRoy T, Bhadeshia HKDH. Recent advances in friction-stir
welding-process, weldment structure and properties. Prog Mater Sci
2008;53:9801023.
[8] Mishra RS, Ma ZY. Friction stir welding and processing. Mater Sci Eng R
2005;50:178.
[9] Srivatsan TS, Vasudevan S, Park L. The tensile deformation and fracture
behavior of friction stir welded aluminum alloy 2024. Mater Sci Eng A
2007;466:23545.
[10] Aydn H, Bayram A, Uguz A, Akay SK. Tensile properties of friction stir welded
joints of 2024 aluminum alloys in different heat-treated-state. Mater Des
2009;30:221121.
[11] Xu WF, Liu JH, Luan GH, Dong CL. Temperature evolution, microstructure and
mechanical properties of friction stir welded thick 2219-O aluminum alloy
joints. Mater Des 2009;30:34607.
[12] Krishnan KN. The effect of post weld heat treatment on the properties of 6061
friction stir welded joints. J Mater Sci 2002;37:47380.
[13] Chen YC, Liu HJ, Feng JC. Effect of post-weld heat treatment on the mechanical
properties of 2219-O friction stir welded joints. J Mater Sci Lett
2006;41:2979.
[14] Sullivan A, Robson JD. Microstructural properties of friction stir welded and
post-weld heat-treated 7449 aluminum alloy thick plate. Mater Sci Eng A
2008;478:35160.
[15] Aydn H, Bayram A, Durgun I. The effect of post-weld heat treatment on the
mechanical properties of 2024-T4 friction stir-welded joints. Mater Des
2010;31:256877.
[16] Charit I, Mishra RS. Abnormal grain growth in friction stir processed alloys. Scr
Mater 2008;58:36771.
[17] Chen YC, Feng JC, Liu HJ. Stability of the grain structure in 2219-O aluminum
alloy friction stir welds during solution treatment. Mater Charact
2007;58:1748.
[18] Charit I, Mishra RS. High strain rate superplasticity in a commercial 2024 Al
alloy via friction stir processing. Mater Sci Eng A 2003;359:2906.
[19] Hassan KAA, Norman AF, Price DA, Prangnell PB. Stability of nugget zone grain
structures in high strength Al alloy friction stir welds during solution
treatment. Acta Mater 2003;51:192336.
[20] Feng JC, Chen YC, Liu HJ. Effects of post-weld heat treatment on microstructure
and mechanical properties of friction stir welded joints of 2219-O aluminium
alloy. Mater Sci Technol 2006;22:8690.
[21] China National Standard, GB/T2651-2008. Tensile test method on welded
joints. China; 2008.
[22] ASTM Standards, ASTM B557-02. Standard test methods of tension testing
wrought and cast aluminum- and magnesium-alloy products, USA; 2002.
[23] Mondolfo LF. Aluminumalloys: structure and properties. London: Butterworth;
1976.
[24] Brooks CR. Heat treatment, structure and properties of nonferrous alloys.
Metals Park (OH, USA): ASM 1984.
Z. Hu et al. / Materials and Design 32 (2011) 50555060 5059
[25] Cravetchi IV, Taschuk M, Tsui YY, Fedosejevs R. Spectrochemical microanalysis
of aluminum alloys by laser-induced breakdown spectroscopy. Spectrochim
Acta B 2004;59:143950.
[26] Sato YS, Park SHC, Kokawa H. Microstructural factors governing hardness in
friction-stir welds of solid-solution-hardened Al alloys. Metall Mater Trans A
2001;32A:303342.
[27] Starink MJ, Deschamps A, Wang SC. The strength of friction stir welded and
friction stir processed aluminium alloys. Scr Mater 2008;58:37782.
[28] Mishra RS, Sharma SR, Mara NA, Mahoney MW. Mechanical properties of
friction stir welded aluminum alloys. in: Proceedings of the international
conference on jointing of advanced and specialty materials III. ASM
International; 2000.
5060 Z. Hu et al. / Materials and Design 32 (2011) 50555060