metals
Article
Microstructure and Mechanical Properties of 6061 Al/AZ31 Mg
Joints Friction Stir Lap Welded by a Tool with Variable-Pitch
Thread Pin
Changshu He 1,2, *, Chengpeng Qiu 1,2 , Zhiqiang Zhang 1,2 , Jingxun Wei 1,2 , Hao Zhang 1,2 , Ni Tian 1,2
and Gaowu Qin 1,2






Citation: He, C.; Qiu, C.; Zhang, Z.;
Wei, J.; Zhang, H.; Tian, N.; Qin, G.
Microstructure and Mechanical
Properties of 6061 Al/AZ31 Mg
Joints Friction Stir Lap Welded by a
Tool with Variable-Pitch Thread Pin.
Metals 2021, 11, 34.
https://dx.doi.org/10.3390/
met11010034
Received: 12 December 2020
Accepted: 23 December 2020
Published: 26 December 2020
Publisher’s Note: MDPI stays neu-
tral with regard to jurisdictional claims
in published maps and institutional
affiliations.
Copyright: © 2020 by the authors. Li-
censee MDPI, Basel, Switzerland. This
article is an open access article distributed
under the terms and conditions of the
Creative Commons Attribution (CC BY)
license (https://creativecommons.org/
licenses/by/4.0/).
Metals 2021, 11, 34. https://dx.doi.org/10.3390/met11010034
1 School of Materials Science and Engineering, Northeastern University, Shenyang 110819, China;
qiximi3478@163.com (C.Q.); lnkdzzq@126.com (Z.Z.); 18978226891@163.com (J.W.);
15734040978@163.com (H.Z.); tiann@atm.neu.edu.cn (N.T.); qingw@smm.neu.edu.cn (G.Q.)
2 Key Laboratory for Anisotropy and Texture of Materials, Northeastern University, Shenyang 110819, China
* Correspondence: changshuhe@mail.neu.edu.cn; Tel.: +86-024-83691573
Abstract: Sheets of 6061-T6 aluminum alloy (thickness = 3 mm) and AZ31 magnesium alloy were
friction stir lap welded by a tool with a variable-pitch thread pin (coarse-threaded in the upper part
and fine-threaded in the lower part). For the same rotation speed and welding speed, the heat input
was higher in joints with an upper Al alloy (Configuration Al-Mg) than in those with an upper Mg
alloy (Configuration Mg-Al). In Configuration Al-Mg, these two dissimilar metals were poorly mixed
and Al dominated the stirred zone (SZ). Many intermetallic compounds (IMCs) of Al3 Mg2 formed
inside the SZ. In Configuration Mg-Al, Mg alloy bands, flocculent Al12 Mg17 bands, and minor Al
alloy bands intersected in the SZ, forming a complex onion-ring structure. Moreover, a complex
mechanical interlocking structure developed at the bottom interface of the SZ. The maximum tensile
shear strengths of the Al-Mg and Mg-Al lap configurations were 160.3 and 217 N/mm, respectively,
at 700 rpm. The higher tensile shear strength of the Mg-Al configuration primarily represented less
IMCs and complex mechanical interlocking structures in the SZ.
Keywords: friction stir lap welding; variable-pitch thread pin; dissimilar metals; metal flow;
intermetallic compounds
1. Introduction
Aluminum and magnesium alloys are lightweight metallic materials with several desir-
able properties: low density, high specific strength, and stiffness. Therefore, they are widely
used in the automobile, aviation, aerospace, and other industries requiring lightweight
structures [1,2]. A hybrid structure formed by welding aluminum and magnesium alloys
is expected to inherit the performance advantages of both alloys, and is considered as an
alternative approach for structural weight-saving [3]. Friction stir welding (FSW) is a solid-
state joining process requiring less heat input than traditional fusion welding. Furthermore,
joints formed by FSW have small distortion, no porosity, and no crack defects [4,5]. FSW
has emerged as a promising method for joining dissimilar aluminum alloy and magnesium
alloy [6–8].
Friction stir lap welding (FSLW) performs FSW in a lap-joint configuration.
During FSLW, a high-speed rotating welding tool (composed of a shoulder and pin) is
plunged into the upper plate and enters the lower plate to a certain depth, before moving
forward along the lap centerline to form the lap weld [9]. However, during FSLW of Al and
Mg (two dissimilar metals), hard and brittle intermetallic compounds (IMCs) are formed
in the joint, which seriously deteriorate the property of the joint [10]. Moreover, hook and
cold lap defects are formed at the advancing side (AS) and the retreating side (RS) of the
joint, respectively. These defects are related to the squeeze effect of the metal in the stirred
https://www.mdpi.com/journal/metals
Metals 2021, 11, 34 2 of 15

zone (SZ) on the original lap interface [11]. During loading, the hook and cold lap defects
produce serious stress concentration, which decreases the bearing capacity of the joint [12].
In similar metals welded by FSLW, the peak temperature gradually decreases from
the top to the bottom of the SZ [13]. This vertical temperature gradient causes an appar-
ent difference in flow ability of the upper and lower metals, leading to voids and tunnel
defects [14]. When dissimilar metals are welded by FSLW, the flow ability difference
between the two metals is relatively large, which is detrimental to the mechanical interlock
formation and degrades the joint property as mentioned above [15]. Therefore, balancing
the heat generation from the upper and lower sheets and homogenizing the tempera-
ture distribution are important goals of FSLW. Using a stationary shoulder welding tool,
Ji et al. [16] fabricated Al/Mg dissimilar metal lap joints. They showed that the assisted
stationary shoulder reduces the heat generation and promotes material mixing in the SZ.
Kush et al. [17] installed a water-cooled nozzle behind the welding tool, which cools the
weld surface during the welding process. They demonstrated that water cooling not only
inhibits the fluidity of the weld surface material and smooths the weld surface, but also
promotes the mixing of Al/Mg and reduces the occur probability of welding defect.
Heat generation and material flow in the FSLW process largely depend on the ge-
ometry of the welding tool, including its shape, dimensions and thread features [18].
Numerous studies have shown that the thread features (thread pitch and thread orienta-
tion) of the tool pin [19–22] greatly affect the material flow (especially the vertical movement
of the material), in turn affecting the joint properties. Sun et al. [19] demonstrated that a
fine-threaded tool pin promotes the mixing of materials in both the horizontal and vertical
directions, and increases the heat generation. Alavi et al. [20] investigated the effect of pin
thread pitch on the material properties of friction stir processed AZ31. They demonstrated
that a pin with a 1.0-mm pitch thread obtains finer and more uniform grains than a pin
with a 3.0-mm pitch thread, and hence improves the mechanical properties of the joint.
Ji et al. [21] and Yue et al. [22] applied a reverse-threaded pin (in which the upper and lower
parts of the pin have left-handed and right-handed threads, respectively), which drives the
material flow toward the lap interface. Consequently, the effective lap width (ELW) and
tensile shear load of the joint are both increased.
In addition, when dissimilar metals are welded by FSLW, the material properties
differ between the top and bottom sheets of the joint, causing heterogeneous material
flows and heat generation [23]. The material position with respect to the welding tool
inevitably impacts the lap joint strength. Firouzdor et al. [24] investigated the effect of
material position on the heat input, material flow, and joint strength in the FSLW process
of 6061 Al to AZ31 Mg. They demonstrated that placing Mg above Al reduces the heat
input and enhances the lap joint strength. Ji et al. [14] showed that placing Mg above Al
can promote the mixing of Mg and Al and increase the width of the SZ, thus increasing
mechanical interlocking effects and improving joint performance. Generally, the heat input
and material flow behavior are greatly affected by the welding parameters (rotation speed
and welding speed). Rao et al. [25] demonstrated that low rotational speeds and high
welding speeds can cause insufficient heat generation by friction, resulting in insufficient
mixing of materials. High rotational speeds and low welding speeds result in higher
friction heat, causing flash and void defects in joints. Tan et al. [26] found that regardless of
whether the heat input was too high or too low, it will cause cracks in the joint, thereby
reducing its performance.
Although previous research has successfully regulated heat input and material flow in
the FSLW process, there have been no reports on the effect of the change of pin thread pitch
on the heat input and material flow in Al/Mg dissimilar metal FSLW processes. In the
present study, a dissimilar lap joint (6061 Al to AZ31 Mg) was fabricated by a welding
tool with a tapered pin. The upper and lower parts of the pin were coarse-threaded and
fine-threaded, respectively, and the fabrication was performed under different welding
parameters and material positions. By investigating the interfacial features, material flow
behavior and mechanical properties of the lap joints in detail, we can reveal the influence
 Metals2021,
Metals 2021,11,
11,34
x FOR PEER REVIEW 3 of 163 of 15
Metals 2021, 11, x FOR PEER REVIEW 3 of 16

parameters
parametersand andmaterial
material positions.
positions. ByByinvestigating
investigating the
theinterfacial
interfacial features,
features, material
materialflow
flowand
mechanism of the tool with the variable-pitch thread pin on the heat generation
behavior
behavior and
and mechanical
mechanical properties
properties ofofthe lap
the joints
lap jointsin detail,
in detail, wewe can
canreveal
revealthe
theinfluence
influence
material flow
mechanism during FSLW the of the two dissimilar alloys (6061 Al to AZ31 Mg). The designed
mechanismof
pin-thread ofthe
form
tool
toolwith
theand with thevariable-pitch
optimized variable-pitch
lap
thread
threadpin
configuration pinon
are onthetheheat
expected heatgeneration
to generation
enhance
andandma-
the ma-
vertical
terial flow
terial flowduring
during FSLW
FSLW ofofthe two
the two dissimilar
dissimilar alloys
alloys (6061
(6061 AlAltotoAZ31
AZ31 Mg).
Mg). The
Thedesigned
designed
material flow,
pin-thread promote the mixing of dissimilar expected
metals, and hencethe improve the joint
pin-threadform
strength. formandandoptimized
optimizedlap lapconfiguration
configurationare are expectedtotoenhance
enhance thevertical
verticalma-ma-
terial
terialflow,
flow,promote
promotethe themixing
mixingofofdissimilar
dissimilarmetals,
metals,andandhence
henceimprove
improvethethejoint
jointstrength.
strength.
2. Materials
2. Materials andMethods Methods
2. Materialsand and Methods
Thebase
The basematerials
materialswerewere3-mm-thick
3-mm-thick 6061-T6
6061-T6 aluminum
aluminum alloy
alloy and and AZ31 magnesium
alloy The base
sheets. Two materials
(300 × weremm
150) 3-mm-thick 6061-T6
2 sheets were aluminum
welded into analloy andAZ31
overlap AZ31magnesium
joint magnesium
with an overlap
alloy
alloysheets.
sheets. Two
Two(300
(300× 150)
150)mm
×welding
2 sheets
mmdirection were
2 sheets were welded
welded into
intoanan overlap
overlap joint with
joint withananoverlap
overlap
width
width of 56 mm.
ofof5656mm. The
The welding direction waswas parallel
parallel to
totothe the rolling
rolling direction
direction of
ofofthe the base
base
width
material. Beforemm. The
welding,welding
the direction
overlapping was parallel
sheet surfaces the rolling
were direction
cleaned with the base
sandpapers
material.
material. Before
Before welding,
welding, thetheoverlapping
overlapping sheet
sheetsurfaces
surfaces were
were cleaned
cleaned with
withsandpapers
sandpapers toto
to remove
remove the the oxidation
oxidation and and organic
organic residues.
residues. The The
sheets sheets
were were
then then
fixed fixed
on a on a worktable,
worktable,
remove the oxidation and organic residues. The sheets were then fixed on a worktable,
and the
and the lap joints
joints wereweldedwelded witha aFSW FSW machine. As shown in Figure 1, two weld
and thelap lap jointswerewere weldedwith with a FSWmachine.
machine.As Asshown
shownininFigureFigure1,1,two
twoweld
weld
configurationswere
configurations were investigated:
investigated: an an Al-Mg
Al-Mg configuration
configuration with with
Al as Alupper
the as thesheet,
upperand sheet,
configurations were investigated: an Al-Mg configuration with Al as the upper sheet, and
andMg-Al
an an Mg-Al configuration with Mg as the upper The sheet. The welding speed was fixed at
an Mg-Alconfiguration
configurationwith withMg Mgasasthe theupper
uppersheet.
sheet. Thewelding weldingspeed speedwaswasfixed
fixedatat100
100
100 mm/min,
mm/min, and and
the the rotation
rotation speed speed
was was as
varied varied
700, as 700,
1100, and 1100,
1500 and
rpm. 1500
The rpm.
tiltingThe tilting
angle
mm/min, and the rotation speed was varied as 700, 1100, and 1500 rpm. The tilting angle
◦ and 0.2 mm, respectively.
angle
of of the tool and the shoulder-plungewere depth were set to 2.5
ofthe
thetool
toolandandthetheshoulder-plunge
shoulder-plungedepth depth wereset settoto2.5°
2.5°andand0.20.2mm,
mm,respectively.
respectively.The The
The tool
tool was outfitted with a variable-pitch thread pin with a length ofmm
3.8 mm and a thread
tool was outfitted with a variable-pitch thread pin with a length of 3.8 mmand
was outfitted with a variable-pitch thread pin with a length of 3.8 anda athread
thread
depth
depth of 0.45
depthofof0.45
mm.
0.45mm.mm.The
The other
Theother
otherpin
pin dimensions
dimensions
pindimensions arearedetailed
detailed
aredetailed
inFigure
Figure
ininFigure 2.2. 2.

Figure 1. Schematic of lap joint modes: (a) Al-Mg and (b) Mg-Al configurations and (c) dimensions of the tensile shear
Figure
Figure 1. Schematic
1. Schematic of joint
of lap lap joint modes:
modes: (a) Al-Mg
(a) Al-Mg andMg-Al
and (b) (b) Mg-Al configurations
configurations and (c)and (c) dimensions
dimensions of the shear
of the tensile tensilespecimen.
shear
specimen.
specimen.

Figure 2. Schematic ofofthe tool with a variable-pitch thread pin.pin.

Figure2.2.Schematic
Figure Schematic of the
the tool with
tool with aa variable-pitch
variable-pitch thread
thread pin.

After
Afterwelding,
After welding,cross-section
welding, cross-sectionsamples
cross-section samples
samples for microstructural
for
for observation
microstructural
microstructural were
observation
observation cut
cutper-
were
were cut
per-per-
pendicularly
pendicularly to
tothe
thewelding
welding direction,
direction, and
andthen mechanically
then mechanicallyground and
ground etched
and
pendicularly to the welding direction, and then mechanically ground and etched with with
etched with
picric
picric acid solution and 20% NaOH solution. The metallographic sample was analyzed
picric acid solution and 20% NaOH solution. The metallographic sample was analyzedun-
acid solution and 20% NaOH solution. The metallographic sample was analyzed
under
der
under optical
optical
opticalmicroscopes
microscopes (OLYMPUS-DSX-WZ
(OLYMPUS-DSX-WZ
microscopes (OLYMPUS-DSX-WZ and
and OLYMPUS-BX53MRF-S,
OLYMPUS-BX53MRF-S,Tokyo,
OLYMPUS-BX53MRF-S,
and Tokyo,Ja-
Tokyo, Japan)
Ja-
pan)
and
pan) and
a anda scanning
scanning electron
electron
a scanning microscope
microscope
electron microscope (JSM-7001F,
(JSM-7001F, JEOL,
JEOL,
(JSM-7001F, Tokyo,
Tokyo,
JEOL, Tokyo,Japan)
Japan) equipped
Japan)equipped with
equippedwith
with an
energy dispersive X ray spectrometer (EDS), and IMCs was identified according to EDS
data. Microhardness tests of representative specimens were performed using a Vickers
microhardness tester (Wolpert Wilson-401MVD, Worcester, MA, USA) with a load of 100 g
Metals 2021, 11, 34 4 of 15

and dwell time of 15 s for measuring the Vickers microhardness (HV); the distance between
each hardness indent was 0.5 mm. The tensile shear properties of the lap joints were deter-
mined in lap shear tests following the ASTM D1002-10 standard. Tensile shear tests were
performed in a Shimadzu AG-X Plus universal testing machine (Tokyo, Japan). During the
tensile process, the load was applied to the AS of the upper sheet. Test specimens (of width
15 mm) were cut from the lap joints perpendicularly to the welding direction, as shown
in Figure 1c. The tensile shear strength of each specimen (in N/mm) was calculated by
dividing the fracture load by the specimen width. The reported tensile shear strengths
are the average values of three trials for each configuration. After the lap shear tests,
the fracture positions of each lap joint were recorded. The fracture morphologies were
observed using a scanning electron microscope (JEOL-JSM6501, JEOL, Tokyo, Japan).

3. Results
3.1. Joint Formations
Figure 3 shows the macro cross-sections of the FSLW joints of dissimilar metals
(Al/Mg) welded in different lap configurations at different rotation speeds. In Configu-
ration Al-Mg at 700 rpm, the SZ boundary was straight and smooth and the thick IMCs
bands in the SZ were distributed in the Al matrix (Figure 3a). The original lap interface of
the thermo-mechanically affected zone near the AS and the RS migrated upward, forming
a hook and cold lap defect, respectively. However, in Configuration Mg-Al at the same
rotation speed, the SZ boundary was complicated and tortuous, entangling a large number
of Al and Mg alloy bands. The complex onion-ring structure indicates that an effective me-
chanical interlock formed in the SZ [27]. Such complex mechanical interlocking improves
the external-load resistance of a joint [28,29]. In addition, the hook and cold lap formed in
Configuration Mg-Al migrated upward over a certain distance before migrating into the SZ.
The lap configuration altered the characteristics not only of the hook and cold lap, but also
of the SZ; in particular, the degree of mechanical locking was higher in Configuration
Metals 2021, 11, x FOR PEER REVIEW 5 of 16
Mg-Al than in Configuration Al-Mg.

Configuration Al-Mg Configuration Mg-Al

(a)

(b)

(c)

FigureFigure 3. Macro
3. Macro morphologies
morphologies of the
of the weldcross
weld crosssections
sections in
indifferent
differentlap configurations
lap formed
configurations at different
formed rotation
at different speeds.speeds.
rotation
(a) 700 rpm; (b) 1100 rpm; (c) 1500 rpm.
(a) 700 rpm; (b) 1100 rpm; (c) 1500 rpm.
In general, increasing the rotation speed of the tool increased the welding heat input,
thereby improving the metal flow ability in the SZ [30]. In Configuration Al-Mg, the SZ
exhibited no obvious onion-ring structures at 700 rpm, but onion-ring characteristics were
observed at 1100 rpm. Meanwhile, onion ring structures were evidenced in the SZ of Con-
figuration Mg-Al at 700 rpm, and became denser as the rotation speed increased. This
onion-ring structure occurred by violent mixing of the two alloys, indicating that the mix-
Metals 2021, 11, 34 5 of 15

In general, increasing the rotation speed of the tool increased the welding heat input,
thereby improving the metal flow ability in the SZ [30]. In Configuration Al-Mg, the SZ
exhibited no obvious onion-ring structures at 700 rpm, but onion-ring characteristics were
observed at 1100 rpm. Meanwhile, onion ring structures were evidenced in the SZ of
Configuration Mg-Al at 700 rpm, and became denser as the rotation speed increased. This
onion-ring structure occurred by violent mixing of the two alloys, indicating that the mixing
degree of the two alloys in the SZ was higher in Configuration Mg-Al than in Configuration
Al-Mg.
In FSLW joints, the ELW is an important determiner of the shear strength of a joint,
which defines the horizontal distance between the hook tip at the AS and the cold lap tip at
the RS [31,32]. Its value is mainly determined by the migration behavior of the interface
materials in the SZ (see Figure 4). In both lap configurations, increasing the rotation speed
from 700 to 1100 rpm decreased the ELW, but when the rotation speed was further increased
to 1500 rpm, the ELW recovered and reached its maximum. The high rotation speed of the
tool promoted material flow in the SZ, increasing the ELW [33]. For any set of welding
Metals 2021, 11, x FOR PEER REVIEW
parameters, the ELW was always greater in the Configuration Al-Mg joint than in6 of 16
the
Configuration Mg-Al joint.

Effectivelap
Figure 4. Effective lap width
width of joints
of joints welded
welded in different
in different lap configurations
lap configurations at different
at different rota-
rotation
speeds.
tion speeds.

3.2. Tensile Shear

3.2. Tensile Shear Strength
Strength
Figure
Figure 5 comparesthe
5 compares thetensile
tensileshear
shearstrengths
strengthsofofthe
thejoints
jointswith
withdifferent
differentlap
lapconfigura-
configu-
tions welded at different rotation speeds. For the same parameter settings, the
rations welded at different rotation speeds. For the same parameter settings, the tensile tensile shear
strength of the joint was stronger in Configuration Mg-Al than in Configuration
shear strength of the joint was stronger in Configuration Mg-Al than in Configuration Al- Al-Mg.
In
Mg. both lap configurations,
In both the tensile
lap configurations, shear strength
the tensile of the joint
shear strength wasjoint
of the maximized at 700 rpm
was maximized at
(160.3 and 217 N/mm in the Al-Mg and Mg-Al configurations, respectively).
700 rpm (160.3 and 217 N/mm in the Al-Mg and Mg-Al configurations, respectively). The The tensile
shear performance of the joint welded in configuration Mg-Al was comparable to that of
tensile shear performance of the joint welded in configuration Mg-Al was comparable to
Ji et al. [15] after adding Zn foil (220 N/mm). As the rotation speed increased, the tensile
that of Ji et al. [15] after adding Zn foil (220 N/mm). As the rotation speed increased, the
shear strengths of both joints trended similarly to the ELW. Enlarging the ELW enhanced the
tensile shear strengths of both joints trended similarly to the ELW. Enlarging the ELW
tensile shear strength by increasing the crack propagation path [34]. However, the strength
enhanced the tensile shear strength by increasing the crack propagation path [34]. How-
of the joint also depends on the amount of IMCs and the microstructure characteristics in
ever, the strength of the joint also depends on the amount of IMCs and the microstructure
the SZ. In general, the heat input is related to the welding parameters (rotation speed and
characteristics in the SZ. In general, the heat input is related to the welding parameters
welding speed) through a parameter called the heat index (HI) factor [4], defined as follows:
(rotation speed and welding speed) through a parameter called the heat index (HI) factor
[4], defined as follows: ω2
HI = (1)
v𝜔
𝐻𝐼 = (1)
𝑣
Here, ω and v represent the rotation speed and welding speed of the tool, respec-
tively. As the rotation speed increased from 700 to 1500 rpm, the HI increased and more
IMCs formed in the SZ, deteriorating the joint performance [34]. Although the ELWs in
the joints formed at 700 rpm were sub-maximal, the low heat input inhibited the emer-
gence of IMCs, so the tensile shear strength was highest at this speed.
 𝐻𝐼 = (1)
𝑣
Here, ω and v represent the rotation speed and welding speed of the tool, respec-
tively. As the rotation speed increased from 700 to 1500 rpm, the HI increased and more
Metals 2021, 11, 34 IMCs formed in the SZ, deteriorating the joint performance [34]. Although the ELWs in
6 of 15
the joints formed at 700 rpm were sub-maximal, the low heat input inhibited the emer-
gence of IMCs, so the tensile shear strength was highest at this speed.

Figure 5.
Figure 5. Tensile
Tensileshear
shearstrengths
strengthsofofthe
thejoints formed
joints in in
formed different laplap
different configurations at different
configurations at different
rotation speeds.
rotation speeds.

Here, ω and v represent the rotation speed and welding speed of the tool, respectively.
As the rotation speed increased from 700 to 1500 rpm, the HI increased and more IMCs
formed in the SZ, deteriorating the joint performance [34]. Although the ELWs in the joints
formed at 700 rpm were sub-maximal, the low heat input inhibited the emergence of IMCs,
so the tensile shear strength was highest at this speed.
For the same parameter settings, the tensile shear strength of the joint was found to be
stronger in the Mg-Al configuration than in the Al-Mg configuration. For dissimilar metals
such as Al and Mg, the heat input is also affected by the lap configuration of the welded
structure. For instance, the material properties (friction coefficient, liquation sensitivity,
and deformability) of the materials [8,35,36], and the shoulder is the main point of heat
generation. For the Al-Mg configuration, the shoulder of the welding tool is completely
in contact with the Al alloy. Both friction heat and plastic deformation heat increased
during the FSLW process because of the higher coefficient of friction and deformability,
as well as the lower liquation sensitivity of the Al alloy [15]. For the joint welded in the
Mg-Al configuration, a liquid-film forms more easily between the shoulder and the Mg
alloy (which has high liquation sensitivity) [36]. This film decreases the friction coefficient
and deformation resistance of materials in the SZ and thus decreases the friction heat and
plastic deformation heat. The lower heat input is beneficial for decreasing the formation of
IMCs and improving joint strength. In addition, the variation of the standard deviation of
the joint welded in the Mg-Al configuration is smaller than that in the Al-Mg configuration
at different rotating speeds. This may be a result of the formation of a large amount
of IMCs, the distribution of which affects the fracture behavior of the welded Al-Mg
joints, causing fluctuations in their tensile shear strength. For joints welded in the Mg-Al
configuration, Al and Mg alloys form an effective mechanical interlocking structure, which
has relatively little influence on the fracture behavior of the joint, such that the tensile shear
strength of the joint is relatively stable under different parameters.
In subsequent experiments, the microstructure, IMCs characteristics, and microhard-
ness of the SZ were investigated in the joints obtained at 700 rpm. The aim was to reveal the
mechanism underlying the different tensile shear strengths of the joints formed in different
lap configurations.

3.3. Microstructure and Microhardness

Figure 6 shows the macro- and micro-structures of the joint welded in the Al-Mg
configuration at 700 rpm, and Figure 7 shows the SEM morphologies and EDS results in
different micro regions. The IMCs generated in the joint appear as bright white regions
in Figure 6b–f. The IMCs bands inside the SZ (Figure 6c) and cold lap (Figure 6d) were
more than 60-µm wide. As shown in Figure 7d–e, the dark grey IMCs bands in the SZ were
mainly Al3 Mg2 (point 3 in Figure 7), indicating that constitutional liquation occurred during
Metals 2021, 11, 34 7 of 15

welding [37]. The IMCs was identified by calculating composition ratio of atoms from EDS
data and reference to the published results [6,24]. Many studies have shown that IMCs
cannot grow to such large sizes by solid-state diffusion alone [23,24,38]. During the welding
process of Configuration Al-Mg, the liquation phase was generated at temperatures above
the eutectic temperature (450 ◦ C), and the eutectic reaction L→Al+Al3 Mg2 occurred during
Metals 2021, 11, x FOR PEER REVIEW 8 of 16
cooling. In addition, the black structure (point 4 in Figure 7) around the Al3 Mg2 bands was
an Al-based solid solution.
Metals 2021, 11, x FOR PEER REVIEW 8 of 16

Figure 6. Macrostructure and microstructure of the joint welded in Configuration Al-Mg at 700 rpm: (a) (a) macro
macro structure;
structure;
Figure 6. Macrostructure and microstructure of the joint welded in Configuration Al-Mg at 700 rpm: (a) macro structure;
(b) hook;
(b) hook; (c)
(c) SZ;
SZ; (d)
(d) cold
cold lap;
lap; (e)
(e) SZ
SZ bottom
bottom at
at the
the AS;
AS; (f)
(f) SZ
SZ bottom
bottom at
at the
the RS.
RS.
(b) hook; (c) SZ; (d) cold lap; (e) SZ bottom at the AS; (f) SZ bottom at the RS.

FigureFigure
7. SEM morphology
7. SEM morphology andandEDS results
EDS atatvarious
results variouspositions ofthe
positions of the jointwelded
welded in Configuration Al–Mg
at 700at 700 rpm:
Figure 7. SEM morphology and EDS results at various positions of thejoint
joint welded in Configuration Al–Mg
in Configuration Al–Mg rpm:
at 700 rpm:
(a) hook(a) tip;
hook(b)tip;interface of the
(b) interface SZ SZ
of the bottom
bottomatat
the
theAS;
AS;(c)
(c)EDS
EDS results
results atatpoints
points1 and
1 and 2; (d)
2; (d) SZ;SZ; (e) magnified
(e) magnified micrograph
micrograph of of
(a)
(d);hook
(f)(d);
EDStip;results
(b) interface of the
of points SZ 4.
3 and bottom at the AS; (c) EDS results at points 1 and 2; (d) SZ; (e) magnified micrograph of
(f) EDS results of points 3 and 4.
(d); (f) EDS results of points 3 and 4.
Figure
Figure 8 8shows
showsthethe macro-
macro- andand micro-structures
micro-structuresof of
thethe
jointjoint
welded in theinMg-Al
welded the Mg-Al
configuration at 700 rpm, and Figure 9 shows the SEM morphology
configuration at 700 rpm, and Figure 9 shows the SEM morphology and EDS and EDS results in
results in
different micro regions. In this sample, thin IMC layers appeared around the hook
different micro regions. In this sample, thin IMC layers appeared around the hook (Figure (Figure
8b)8b)
andand cold lap (Figure 8d). During the FSLW process, the Al and Mg alloys in the SZ
cold lap (Figure 8d). During the FSLW process, the Al and Mg alloys in the SZ
were largely plastically deformed by strong thermal-mechanical coupling, and intersected
were largely plastically deformed by strong thermal-mechanical coupling, and intersected
Metals
Metals 2021, 11, 34
2021, 11, x FOR PEER REVIEW 98 of
of 16
15

EDS results (Figure

At the tip 9f),
of the the regions
hook of deepest
defect (Figure 6b), and lightest
a typical contrast
eutectic constituted
structure (FigureMg7a)andwas
Al
alloy, respectively. The EDS result of Figure 9h confirmed the banded
observed. According to the EDS results (Figure 7c), the eutectic structure (point 1 in flocculent structure
(point
Figure5) 7)as Al12
was Mg17. When
mainly Al12 Mg the +Mg,
local temperature inside the
further confirming the SZ reached the
occurrence eutectic tem-
of constitutional
17
perature
liquation. During the tensile shear test, those hard and brittle IMCs acceleratedtothe
(437 °C), the contact interface between the Al and Mg bands began liquefy,
crack
forming band liquation films which precipitated flocculent Al 12Mg17 during subsequent
propagation, thereby deteriorating the joint properties. The bottom interface of the SZ at
cooling.
the AS wasTherefore,
flat andthe Mg alloy
smooth bands,
(Figure 6e), flocculent
indicatingAl 12Mg17 bands, and a small proportion
failure to form an effective mechanical
of
interlocking structure. Examining panels b and c ofthe
the Al alloy bands intersected in the SZ. Note that bottom
Figure interface
7, the IMCs of (Althe SZ at the
3 Mg2 ) were
RS was very tortuous, forming a complex mechanical interlocking
distributed as a 2.4-µm thick layer along the interface. The IMCs were concentrated around structure (Figure 8f).
The Al alloy and bulk IMCs at the interface bit into each other (Figure
the tunnel defect observed at the bottom of the SZ at the RS (Figure 6f). This effect is mainly 9a,b), thereby in-
creasing the strength of the joint [39]. As seen in Figure 9c, the bulk
attributable to blockage of the plastic flow by the IMCs, which prevented the plastic metal IMCs generated near
the
fromAlfilling
side were mainly
the cavity at Al
the3Mg 2, and
bottom ofthe
theflocculent structure
SZ in sufficient time.generated near the Mg side
was eutectic
Figure 8Alshows
12Mg17+Mg, consistent
the macro- and with a previous study
micro-structures of the[24]. Figure
joint welded9g isina the
magnified
Mg-Al
observation of the hook structure in Figure 8e, showing fine
configuration at 700 rpm, and Figure 9 shows the SEM morphology and EDS results in broken and dispersed IMCs
dif-
through
ferent micro regions. In this sample, thin IMC layers appeared around the hook (Figurewas
the structure. As shown in the EDS results (Figure 9h), this black structure 8b)
an
andAl-based
cold lap solid
(Figuresolution
8d). Duringwith athe MgFSLWcontent of 4.3the
process, at.%.
Al The
and Mg6061alloys
Al alloy (point
in the 7 in
SZ were
Figure
largely 9) containeddeformed
plastically only 1.9 by at.% Mg, indicating
strong that the black
thermal-mechanical structures
coupling, were mainly
and intersected to
formed by solid-state
form a complex diffusion
onion-ring during
structure the welding process.
[16].

(a) macro
Figure 8. Macrostructure and microstructure of the joint welded in Configuration Mg-Al at 700 rpm: (a) macro structure;
structure;
at the
(b) hook; (c) SZ; (d) cold lap; (e) SZ bottom at the AS;
AS; (f)
(f) SZ
SZ bottom
bottom at
at the
the RS.
RS.

Figure 8c shows the onion ring structure inside the SZ. Three kinds of banded struc-
tures with different contrasts appeared in the onion ring (Figure 9d,e). As shown in the
EDS results (Figure 9f), the regions of deepest and lightest contrast constituted Mg and Al
alloy, respectively. The EDS result of Figure 9h confirmed the banded flocculent structure
(point 5) as Al12 Mg17 . When the local temperature inside the SZ reached the eutectic
temperature (437 ◦ C), the contact interface between the Al and Mg bands began to liquefy,
forming band liquation films which precipitated flocculent Al12 Mg17 during subsequent
cooling. Therefore, the Mg alloy bands, flocculent Al12 Mg17 bands, and a small proportion
of the Al alloy bands intersected in the SZ. Note that the bottom interface of the SZ at
the RS was very tortuous, forming a complex mechanical interlocking structure (Figure
8f). The Al alloy and bulk IMCs at the interface bit into each other (Figure 9a,b), thereby
increasing the strength of the joint [39]. As seen in Figure 9c, the bulk IMCs generated near
the Al side were mainly Al3 Mg2 , and the flocculent structure generated near the Mg side
was eutectic Al12 Mg17 +Mg, consistent with a previous study [24]. Figure 9g is a magnified
observation of the hook structure in Figure 8e, showing fine broken and dispersed IMCs
Metals 2021, 11, 34 9 of 15

through the structure. As shown in the EDS results (Figure 9h), this black structure was an
Al-based solid solution with a Mg content of 4.3 at.%. The 6061 Al alloy (point 7 in Figure 9)
contained only 1.9 at.% Mg, indicating that the black structures were mainly10formed
Metals 2021, 11, x FOR PEER REVIEW of 16 by
solid-state diffusion during the welding process.

FigureFigure
9. SEM 9. SEM morphology
morphology andand
EDSEDS results
results ofofthe
thejoint
jointwelded
welded in
in Configuration
Configuration Mg-Al
Mg-Alat 700 rpm:
at 700 (a) interface
rpm: of SZ of SZ
(a) interface
bottom at the RS; (b) magnified micrograph of (a); (c) EDS results of points 1 and 2; (d) SZ; (e) magnified micrograph of
bottom(d);
at (f)
theEDS
RS; results
(b) magnified micrograph of (a); (c) EDS results of points 1 and 2; (d) SZ; (e) magnified micrograph of (d);
of points 3 and 4; (g) hook structure at SZ bottom at the AS; (h) EDS results of points 5, 6 and 7.
(f) EDS results of points 3 and 4; (g) hook structure at SZ bottom at the AS; (h) EDS results of points 5, 6 and 7.
Figures 10 and 11 show the results of microhardness tests of joints welded in the Al-
MgFigures
and Mg-Al10 and 11 show the
configurations results
at 700 rpm. of
Themicrohardness
figures show that tests
theof joints welded
hardness values ofin the
Al-Mg and
the SZ ofMg-Al configurations
two joints at 70060
are approximately rpm.
HV,The figures
whereas theshow thatofthe
hardness SZhardness
in the Al-Mgvalues of
theconfiguration
SZ of two joints are lower
is much approximately
than that of60
theHV, whereas
6061-T6 the hardness
aluminum of SZ in the97Al-Mg
alloy (approximately
configuration
HV); this may is be
much lowertothan
attributed that of the
the softening 6061-T6
of the aluminumaluminum alloy
alloy during (approximately
welding. No-
tably, this
97 HV); the microhardness
may be attributed profiles
to reveal large fluctuations
the softening at the Al12alloy
of the aluminum Mg17+Mgduringeutectic
welding.
structure
Notably, theofmicrohardness
the hook tip (Figure 10c)reveal
profiles and Allarge
3Mg2 bands inside the
fluctuations at SZ
the(Figure
Al12 Mg 10d),
17
which
+Mg eutectic
are 177.3 HV and 130.8 HV, respectively. For the joint welded in the Mg-Al
structure of the hook tip (Figure 10c) and Al3 Mg2 bands inside the SZ (Figure 10d), which configuration,
arelarge
177.3fluctuations
HV and 130.8 in hardness were also observed.
HV, respectively. For the The hardness
joint welded of the mechanical
in the Mg-Al inter-
configura-
locking structure composed of Al12Mg17 bands and Mg alloy bands around the hook (Fig-
tion, large fluctuations in hardness were also observed. The hardness of the mechanical
ure 11c) was up to 132.2 HV, whereas the hardness of the flocculent Al12Mg17 inside the SZ
interlocking structure composed of Al12 Mg17 bands and Mg alloy bands around the hook
(Figure 11d) was up to 118.1 HV. A large amount of hard and brittle Al3Mg2 bands were
(Figure
formed11c) wasSZ
in the upofto
the132.2
Al-Mg HV, whereas the
configuration hardness
at 700 of thetoflocculent
rpm, leading Al12 Mg
the preferential 17 inside
initia-
thetion
SZ (Figure 11d) was up to 118.1 HV. A large amount of hard and brittle
of cracks at the interface during tensile shear test. As a result, the mechanical proper- Al 3 Mg 2 bands
were formed in the SZ of the Al-Mg configuration at 700 rpm, leading
ties of joints welded in the Al-Mg configuration are weaker than those welded in the Mg- to the preferential
initiation of cracks at the interface during tensile shear test. As a result, the mechanical
Al configuration.
 Metals 2021, 11, 34 10 of 15

als 2021, 11, x FOR PEER REVIEW 11 of 16

als 2021, 11, x FOR PEER REVIEW 11 of 16
properties of joints welded in the Al-Mg configuration are weaker than those welded in the
Mg-Al configuration.

Figure 10. Results of microhardness tests of the joint welded in Configuration Al-Mg at 700 rpm:
Figure 10. Results
(a) of microhardness
Figure 10. Results
schematic tests of the joint
of microhardness
diagram of hardness welded
tests
test of theinjoint
location; Configuration
(b) welded in Al-Mg at 700(c)
Configuration
microhardness profiles; rpm: (a) at
Al-Mg
opticalschematic
700 rpm:diagram of
micrograph
(a)indent
at schematic
hardness test location; (b) diagram of hardness
microhardness
1; (d) test
profiles;at
optical micrograph location;
(c)indent
optical (b) microhardness
2. micrograph at indentprofiles; (c) optical
1; (d) optical micrograph
micrograph at indent 2.
at indent 1; (d) optical micrograph at indent 2.

Figure
Figure 11. Results 11. Results of microhardness
of microhardness tests
tests of the joint of theinjoint
welded welded in Mg-Al
Configuration Configuration Mg-Al
at 700 rpm: (a) at 700 rpm:diagram of
schematic
Figure
(a)
hardness test location;11.
(b)Results
schematic of microhardness
diagram of hardness
microhardness tests
profiles;test
(c) of themicrograph
location;
optical joint welded in Configuration
(b) microhardness
at indentprofiles; (c) Mg-Al
optical
1; (d) optical at 700 rpm:
micrograph
micrograph at indent 2.
(a)indent
at schematic
1; (d)diagram of hardness at
optical micrograph test location;
indent 2. (b) microhardness profiles; (c) optical micrograph
at indent 1; (d) optical micrograph at indent 2.
3.4. Fracture 3.4. Fracture Morphology
Morphology
3.4. Fracture Morphology
Figures 12 and
Figures 12 and 13 show the13 showand
macro- the macro- and micro-structures
micro-structures of the cross of the cross
sections sections of the
of the
failure joints welded in the Al-Mg and Mg-Al configurations at 700 rpm. Rao et al. [25]et al. [25] re-
Figures failure
12 and joints
13 welded
show the in the
macro- Al-Mg
and and Mg-Al configurations
micro-structures of the at
cross 700 rpm.
sections ofRao
the
reported theported
failure joints welded thein
formation formation
themain
of Al-Mg of main
and and andcracks
Mg-Al
minor minor cracks during
configurations
during at the
700loading
the loading rpm. process.
Rao
process. The
[25] main cracks
et al.main
The
reported
cracks ledthe formation
to final failureofof main andDuring
the joint. minor the
cracks during
loading the loading
process process.
of the present Theminor
joint, main
cracks led to final failure of the joint. During the loading process of the present
cracks were generated at the hook defect and propagated upward (Figures 12a and 13a). joint, minor
cracks
As the were generated the
load increased, at the hook defectbending
strengthening and propagated
momentupward
initiated(Figures
the main12a andat13a).
crack the
As the load increased, the strengthening bending moment initiated the main crack at the
Metals 2021, 11, 34 11 of 15

Metals
Metals2021,
2021,11,
11,xxFOR
FORPEER
PEERREVIEW
REVIEW 12 of 16
12 of 16
led to final failure of the joint. During the loading process of the present joint, minor cracks
were generated at the hook defect and propagated upward (Figures 12a and 13a). As the
load increased, the strengthening bending moment initiated the main crack at the cold lap
cold
cold lap
defect. defect.
lapThis crack
defect. This crack
crack propagated
propagated
This along thealong
propagated interface
along the interface
the at the bottom
interface at
at theof bottom
the the of
SZ, and
bottom of the SZ,
SZ, and
eventually
the and
eventually
fractured fractured
the lap the
joint. Inlap joint. In
Configuration Configuration
Al-Mg, cracksAl-Mg,
were cracks
eventually fractured the lap joint. In Configuration Al-Mg, cracks were initiatedaround
initiated were
in initiated
the IMCs in
in the
the
IMCs
IMCs around
the hook the
the hook
and cold
around and
and cold
lap defects
hook lap
lap defects
(Figure
cold 12b,c). In
defects (Figure 12b,c).
12b,c). In
Configuration
(Figure Configuration
InMg-Al, the mainMg-Al,
Configuration cracks the
Mg-Al, did
the
main
main cracks
cracks did
not converge with
did not converge
notthe with
with the
minor cracks
converge minor
along
the minorthe cracks
cracks along
contour of thethe
along SZ,
the contour of
of the
but extended
contour the SZ,
intobut
SZ, theex-
but Al
ex-
tended
matrix into
tended into the
the Al
(Figure Al matrix
13b). (Figure
(Figure 13b).
Meanwhile,
matrix cracks
13b). Meanwhile,
at the RS cracks
Meanwhile, cracks atat the
propagated RS
RS propagated
thealong the cold lap
propagated along
along the
defect
the
cold
cold lap
lap defect
interface, defect interface,
increasing increasing
the crack
interface, the
the crack
propagation
increasing path
crack propagation
and therebypath
propagation and
and thereby
improving
path improving
the performance
thereby improving
the performance
(Figure 13c). As (Figure
shown 13c).
in As
Figures shown
12d in
and Figures
13d, 12d
the and
crack 13d, the
propagation
the performance (Figure 13c). As shown in Figures 12d and 13d, the crack propagation crack propagation
path along the
path
path along the bottom interface of the SZ in Configuration Al-Mg was relatively smooth
along
bottom the
interface bottom
of the interface
SZ in of the
ConfigurationSZ in Configuration
Al-Mg was Al-Mg
relatively was
smooth relatively
and straight.
smooth In
and straight.
contrast, the In contrast,
main crack the main
propagation crack propagation
path along the path
bottom alongof
and straight. In contrast, the main crack propagation path along the bottom of the SZ in the
the bottom
SZ in of the
Configuration SZ in
Mg-Al was branched
Configuration
Configuration Mg-Al
Mg-Al was withbranched
was many secondary
branched with
with many
many cracks. The complex
secondary
secondary cracks.
cracks. The interlocking
The complex structure
complex interlock-
interlock-
hindered
ing the propagation of the main crack, creating a curved and
ing structure hindered the propagation of the main crack, creating a curved and tortuous
structure hindered the propagation of the main crack, creating a tortuous
curved propagation
and tortuous
path.
propagation
propagation path.
path.

Figure
Figure 12.
12. Macro-
Macro- and
and micro-structures
micro-structures of of the
the cross
cross section
section of
of the
the failure
failure joint
joint welded
weldedin
in Configuration
Configuration Al-Mg
Al-Mg at
at 700
700 rpm:
rpm:
rpm:
(a) macro-structure; (b)
macro-structure; (b)
(a) macro-structure; hook;
(b) hook; (c)
hook; (c) cold
(c) cold lap;
cold lap; (d)
lap; (d) bottom
(d) bottom
bottom ofof the
of the SZ.
the SZ.
SZ.

Figure
Figure 13.
13.Macro-
Macro-and
andmicro-structures
micro-structuresof ofthe
thecross
crosssection
sectionofofthe
thefailure
failurejoints
jointswelded
weldedin
inConfiguration
ConfigurationMg-Al
Configuration Mg-Alat
at700
700rpm:
700 rpm:
rpm:
(a) macro-structure;
(a) macro-structure; (b)
macro-structure; (b) hook;
(b) hook; (c)
hook; (c) cold
(c) cold lap;
cold lap; (d)
lap; (d) bottom
(d) bottom of
bottom of the
of the SZ.
the SZ.
SZ.
(a)

Figure
Figure 14
14 shows
shows thethe fracture
fracture morphologies
morphologiesof
morphologies of both
both joint
joint configurations.
configurations.In
configurations. InConfigu-
In Configu-
Configu-
ration Al-Mg,
ration Al-Mg,
ration the
Al-Mg, the fracture
the fracture surface
fracture surface was
surface was relatively
was relatively flat,
relatively flat, showing
flat, showing
showing an an arc-like
an arc-like material
materialflow.
arc-like material flow.
flow.
The
The cracks
cracks propagated
propagated rapidly
rapidly through
through the
the smooth
smooth IMCsIMCs layer
layer atat the
the bottom
bottom surface
surface of
of
the
the SZ
SZ (Figure
(Figure 14b),
14b), which
which mainly
mainly explains
explains the
the poor
poor performance
performance of of the
the joint
joint welded
welded inin
this configuration. In contrast, Configuration Mg-Al presented a relatively
this configuration. In contrast, Configuration Mg-Al presented a relatively rough fracturerough fracture
surface
surface (Figure
(Figure 14c)
14c) with
with aa multitude
multitude of
of secondary
secondary cracks
cracks (Figure
(Figure 14d).
14d). The
The secondary
secondary
Metals 2021, 11, 34 12 of 15

Metals 2021, 11, x FOR PEER REVIEW

The cracks propagated rapidly through the smooth IMCs layer at the bottom surface 13 ofofthe
16

SZ (Figure 14b), which mainly explains the poor performance of the joint welded in this con-
figuration. In contrast, Configuration Mg-Al presented a relatively rough fracture surface
(Figure 14c) enforced
cracks were with a multitude of secondary
by the complex cracks (Figure
interlocking 14d).
structure Thebottom
at the secondary
of thecracks were
SZ, which
enforced by the complex interlocking structure at the bottom of the SZ, which hindered
hindered the propagation of the main cracks. By this mechanism, the joint performance the
propagation
was improved. of the main cracks. By this mechanism, the joint performance was improved.

Figure 14. Fracture morphologies of the joints welded in different lap configurations at 700 rpm: (a)
(a) Configuration
Configuration Al-Mg;
Al-Mg;
(b) magnified
(b) magnified micrograph
micrograph of
of (a);
(a); (c)
(c) Configuration
Configuration Mg-Al;
Mg-Al; (d)
(d) magnified
magnifiedmicrograph
micrographof of(c).
(c).

4. Discussion
4. Discussion
To further reveal the material flow behaviors in the SZs of the two lap configurations,
To further
the FSLW reveal
process wasthesimplified
material flow behaviors
to three stagesin the
(seeSZs of the15).
Figure twoInlap configurations,
Stage 1, the pin
the FSLW process was simplified to three stages (see Figure 15). In
plunged into the upper sheet but did not penetrate the lower sheet. In Stage Stage 1, the pin2,plunged
the pin
into the the
entered upper sheet
lower butbut
sheet didthe
notshoulder
penetrate thenot
did lower
touch sheet. In Stage
the upper 2, the
sheet. In pin
Stageentered
3, the
the lower sheet but the shoulder did not touch the
shoulder touched the upper sheet, and stable welding began. upper sheet. In Stage 3, the shoulder
touched the upper sheet,
In Configuration and stable
Al-Mg, the pinwelding began.the downward migration of the Al al-
thread drove
loy during the first stage. As the plunge depthdrove
In Configuration Al-Mg, the pin thread of thethe
pindownward migration
increased, the upper (Alof the Al
alloy)
alloy during the first stage. As the plunge depth of the pin increased, the upper
and lower (Mg alloy) sheets contacted the coarse and fine threads of the pin, respectively (Al alloy)
and lower (Mg alloy) sheets contacted the coarse and fine threads of the pin, respectively
(Stage 2). As the fine thread possessed a larger surface area than the coarse thread, it gen-
(Stage 2). As the fine thread possessed a larger surface area than the coarse thread, it gen-
erated a large frictional heat when contacting the Mg alloy. However, owing to its high
erated a large frictional heat when contacting the Mg alloy. However, owing to its high
friction coefficient and good deformability, Al flows more easily than Mg. The rapid
friction coefficient and good deformability, Al flows more easily than Mg. The rapid down-
downward flow of Al alloy squeezed the Mg alloy, inhibiting its upward migration; con-
ward flow of Al alloy squeezed the Mg alloy, inhibiting its upward migration; consequently,
sequently, the original lap interface arched upward. During the Stage 3, the shoulder
the original lap interface arched upward. During the Stage 3, the shoulder plunged into the
plunged into the upper sheet of the Al alloy, further increasing the frictional heat genera-
upper sheet of the Al alloy, further increasing the frictional heat generation and improving
tion and improving the flow ability of the Al alloy. When the SZ was almost filled with Al
the flow ability of the Al alloy. When the SZ was almost filled with Al alloy, the mixing
alloy, the mixing degree of Al/Mg was reduced and an effective interlocking structure
could not be formed (Figure 15a).
As the plunge depth of the pin increased in Configuration Mg-Al, the fine threaded
part of the pin eventually contacted the lower Al sheet. The increased contact area be-
tween the Al alloy and the fine-threaded pin part improved the flow ability of the Al alloy

Metals 2021, 11, 34
Figure 15. Schematic of material flows in different lap configurations: (a) Configuration Al-Mg; (b) Configuration
liquation sensitivity) [36] and the shoulder. The lubricating effect of the liquation film re-
duced the frictional heat generated between the shoulder and the Mg alloy, weakening
the flow ability of the Mg alloy. This weakening effect reduced the difference in flow abil-
ities between the Al and Mg alloys, enabling their violent mixture in the SZ to form the
onion-ring structure. Within the vortex layer, the hook defect and cold lap defect migrated
13 of 15
into the SZ, and the bottom interface of the SZ became more tortuous (Figure 15b). The
high tensile shear strength of the joint welded in the Mg-Al configuration was attributed
to the mechanical interlocking structure. From the research result above, the variable-
degreetool
pitch of Al/Mg
proposedwasinreduced andcan
this paper an effective interlocking
effectively enhance thestructure
verticalcould not be
material formed
flow, pro-
(Figure 15a).
mote the mixing of dissimilar metals, and hence improve the joint strength.
Configuration Mg-Al.
Mg-Al.
As the plunge depth of the pin increased in Configuration Mg-Al, the fine threaded
5. Conclusions
part of
In the
thispin eventually
study, contacted
a 3-mm-thick the lower
6061-T6 Al sheet.
aluminum Thewas
alloy increased
weldedcontact
with aarea between
sheet (AZ31
the Al alloy and the fine-threaded pin part improved the flow ability of the
magnesium alloy) by friction stir lap welding (FSLW) using a tool with a variable-pitch Al alloy at
the bottom
thread of the SZ. The upper
pin (coarse-threaded in theMg sheet
upper then
part; squeezed the
fine-threaded inlap
theinterface andThis
lower part). migrated
paper
downward, forcing the upward migration of the lap interface bordering the lower Al sheet.
investigated the effects of the tool with a variable-pitch thread pin on the material flow
This mechanism caused the curvature of the original lap interface in Stage 2. During the
behavior, IMCs distribution and tensile shear strength of joints welded in different con-
Stage 3, a liquation film was easily formed between the Mg alloy (which has high liquation
figurations under different welding parameters. The significant conclusions are given be-
sensitivity) [36] and the shoulder. The lubricating effect of the liquation film reduced the
low.
frictional heat generated between the shoulder and the Mg alloy, weakening the flow ability
(1) In Configuration Al-Mg, the Al alloy around the coarse-threaded part of the pin
of the Mg alloy. This weakening effect reduced the difference in flow abilities between the
near the shoulder was easily driven downward, and the stirred zone was primarily com-
Al and Mg alloys, enabling their violent mixture in the SZ to form the onion-ring structure.
posed of Al alloy. In this case, the dissimilar metals were poorly mixed and the bottom
Within the vortex layer, the hook defect and cold lap defect migrated into the SZ, and the
bottom interface of the SZ became more tortuous (Figure 15b). The high tensile shear
strength of the joint welded in the Mg-Al configuration was attributed to the mechanical
interlocking structure. From the research result above, the variable-pitch tool proposed
in this paper can effectively enhance the vertical material flow, promote the mixing of
dissimilar metals, and hence improve the joint strength.
5. Conclusions
In this study, a 3-mm-thick 6061-T6 aluminum alloy was welded with a sheet (AZ31
magnesium alloy) by friction stir lap welding (FSLW) using a tool with a variable-pitch
thread pin (coarse-threaded in the upper part; fine-threaded in the lower part). This pa-
per investigated the effects of the tool with a variable-pitch thread pin on the material
Metals 2021, 11, 34 14 of 15

flow behavior, IMCs distribution and tensile shear strength of joints welded in different
configurations under different welding parameters. The significant conclusions are given
below.
(1) In Configuration Al-Mg, the Al alloy around the coarse-threaded part of the
pin near the shoulder was easily driven downward, and the stirred zone was primarily
composed of Al alloy. In this case, the dissimilar metals were poorly mixed and the
bottom interface remained smooth and straight. In Configuration Mg-Al, the Al alloy was
driven downward under the action of the fine-threaded pin part, then reversed direction
at the bottom of the stirred zone. Finally, it mixed violently with the Mg alloy migrating
downward under the action of the coarse-threaded pin part. The enhanced mixing degree
of Al/Mg formed a complex mechanical interlocking structure.
(2) The frictional heat generation was higher in Configuration Al-Mg than in Con-
figuration Mg-Al; consequently, large amounts of IMCs were formed inside the SZ of
Configuration Al-Mg. The IMCs (mainly constituting Al3 Mg2 ) were continuously dis-
tributed along the bottom interface of the SZ. In Configuration Mg-Al, the flocculent
Al12 Mg17 bands, Mg alloy bands, and minor Al alloy bands intersected in the SZ, form-
ing a complex onion-ring structure, and the IMCs (mainly constituting Al12 Mg17 ) were
intermittently distributed along the bottom interface of the SZ.
(3) The maximum tensile shear strengths of the joints welded in Configurations Al-Mg
and Mg-Al at 700 rpm were 160.3 and 217 N/mm, respectively. The higher tensile shear
strength in Configuration Mg-Al was mainly attributable to the less IMCs and the complex
mechanical interlocking structure in the SZ of this configuration.

Author Contributions: Conceptualization, C.H., C.Q. and Z.Z.; methodology, C.Q., N.T. and Z.Z.;
software, J.W. and H.Z.; validation, C.Q. and Z.Z.; investigation, C.Q. and Z.Z.; writing—original
draft preparation, C.H., C.Q. and Z.Z.; writing—review and editing, C.H., C.Q. and Z.Z.; supervision,
C.H. and G.Q.; project administration, N.T.; funding acquisition, G.Q. All authors have read and
agreed to the published version of the manuscript.
Funding: This work was funded by the Project of Promoting Talents in Liaoning Province
(No. XLYC1808038) and the National Key Research and Development Program of China
(No. 2016YFB1200506-12).
Conflicts of Interest: The authors declare no conflict of interest in this work.

References
1. Heinz, A.; Haszler, A.; Keidel, C.; Moldenhauer, S.; Miller, W.S. Recent development in aluminum alloys for aerospace application.
Mater. Sci. Eng. A 2000, 280, 102–107. [CrossRef]
2. Mordike, B.L.; Ebert, T. Magnesium properties-Applications-Potential. Mater. Sci. Eng. A 2001, 302, 37–45. [CrossRef]
3. Liu, H.; Chen, Y.; Yao, Z.; Lou, F. Effect of Tool Offset on the Microstructure and Properties of AA6061/AZ31B Friction Stir
Welding Joints. Metals 2020, 10, 546. [CrossRef]
4. Mishra, R.S.; Ma, Z.Y. Friction stir welding and processing. Mater. Sci. Eng. R 2010, 50, 1–278. [CrossRef]
5. Zhang, Z.; He, C.; Li, Y.; Yu, L.; Zhao, S.; Zhao, X. Effects of ultrasonic assisted friction stir welding on flow behavior, microstructure
and mechanical properties of 7N01-T4 aluminum alloy joints. J. Mater. Sci. Technol. 2020, 43, 1–13. [CrossRef]
6. Lv, X.Q.; Wu, C.S.; Padhy, G.K. Diminishing intermetallic compound layer in ultrasonic vibration enhanced friction stir welding
of aluminum alloy to magnesium alloy. Mater. Lett. 2017, 203, 81–84. [CrossRef]
7. Yan, J.; Xu, Z.; Li, Z.; Li, L.; Yang, S. Microstructure characteristics and performance of dissimilar welds between magnesium alloy
and aluminum formed by friction stirring. Scr. Mater. 2005, 53, 585–589. [CrossRef]
8. Fu, B.; Qin, G.; Li, F.; Meng, X.; Zhang, J.; Wu, C. Friction stir welding process of dissimilar metals of 6061-T6 aluminum alloy to
AZ31B magnesium alloy. J. Mater. Process. Technol. 2015, 218, 38–47. [CrossRef]
9. Liu, H.; Zhao, Y.; Hu, Y.; Chen, S.; Lin, Z. Microstructural characteristics and mechanical properties of friction stir lap welding
joint of Alclad 7B04-T74 aluminum alloy. Int. J. Adv. Manuf. Technol. 2015, 78, 1415–1425. [CrossRef]
10. Chen, Y.C.; Nakata, K. Friction stir lap joining aluminum and magnesium alloys. Scr. Mater. 2008, 58, 433–436. [CrossRef]
11. Liu, H.; Hu, Y.; Peng, Y.; Dou, C.; Wang, Z. The effect of interface defect on mechanical properties and its formation mechanism in
friction stir lap welded joints of aluminum alloys. J. Mater. Process. Technol. 2016, 238, 244–254. [CrossRef]
12. Lee, C.; Lee, W.; Kim, J.; Choi, D.; Yeon, Y.; Jung, S. Lap joint properties of FSWed dissimilar formed 5052 Al and 6061 Al alloys
with different thickness. J. Mater. Sci. 2008, 43, 3296–3304. [CrossRef]
Metals 2021, 11, 34 15 of 15

13. Xu, W.; Liu, J.; Luan, G.; Dong, C. Temperature evolution, microstructure and mechanical properties of friction stir welded thick
2219-O aluminum alloy joints. Mater. Des. 2009, 30, 1886–1893. [CrossRef]
14. Zhu, R.; Gong, W.; Cui, H. Temperature evolution, microstructure, and properties of friction stir welded ultra-thick 6082
aluminum alloy joints. Int. J. Adv. Manuf. Technol. 2020, 108, 331–343. [CrossRef]
15. Ji, S.; Niu, S.; Liu, J.; Meng, X. Friction stir lap welding of Al to Mg assisted by ultrasound and a Zn interlayer.
J. Mater. Process. Technol. 2018, 267, 141–151. [CrossRef]
16. Ji, S.; Li, Z.; Zhang, L.; Zhou, Z.; Chai, P. Effect of lap configuration on magnesium to aluminum friction stir lap welding assisted
by external stationary shoulder. Mater. Des. 2016, 100, 160–170. [CrossRef]
17. Mehta, K.; Carlone, P.; Astarita, A.; Scherillo, F.; Rubino, F. Conventional and cooling assisted friction stir welding of AA6061 and
AZ31B alloys. Mater. Sci. Eng. A 2019, 759, 252–261. [CrossRef]
18. Thomas, W.M.; Staines, D.G.; Norris, I.M. Friction stir welding tools and developments. Weld. World 2003, 47, 10–17. [CrossRef]
19. Sun, Z.; Wu, C. Influence of tool thread pitch on material flow and thermal process in friction stir welding. J. Mater. Process. Technol.
2020, 275, 116281. [CrossRef]
20. Nia, A.A.; Omidvar, H.; Nourbakhsh, S.H. Investigation of the effects of thread pitch and water cooling action on the mechanical
strength and microstructure of friction stir processed AZ31. Mater. Des. 2013, 52, 615–620.
21. Ji, S.; Li, Z.; Zhou, Z.; Wu, B. Effect of Thread and Rotating Speed on Material Flow Behavior and Mechanical Properties of
Friction Stir Lap Welding Joints. J. Mater. Eng. Perform. 2017, 26, 5085–5096. [CrossRef]
22. Yue, Y.; Li, Z.; Ji, S.; Huang, Y.; Zhou, Z. Effect of Reverse-threaded Pin on Mechanical Properties of Friction Stir Lap Welded
Alclad 2024 Aluminum Alloy. J. Mater. Sci. Technol. 2016, 32, 671–675. [CrossRef]
23. Firouzdor, V.; Kou, S. Al-to-Mg Friction Stir Welding: Effect of Positions of Al and Mg with Respect to the Welding Tool. Weld. J.
2009, 88, 213–224.
24. Firouzdor, V.; Kou, S. Formation of Liquid and Intermetallics in Al-to-Mg Friction Stir Welding. Met. Mater. Trans. A
2010, 41, 3238–3251. [CrossRef]
25. Rao, H.M.; Ghaffari, B.; Yuan, W.; Jordon, J.B.; Badarinarayan, H. Effect of process parameters on microstructure and mechanical
behaviors of friction stir linear welded aluminum to magnesium. Mater. Sci. Eng. A 2016, 651, 27–36. [CrossRef]
26. Tan, S.; Zheng, F.; Chen, J.; Han, J.; Wu, Y.; Peng, L. Effects of process parameters on microstructure and mechanical properties of
friction stir lap linear welded 6061 aluminum alloy to NZ30K magnesium alloy. J. Magnes. Alloy. 2017, 5, 56–63. [CrossRef]
27. Liu, Z.; Yang, K.; Ji, S. Reducing Intermetallic Compounds of Mg/Al Joint in Friction Stir Lap Welding. J. Mater. Eng. Perform.
2018, 27, 5605–5612. [CrossRef]
28. Yamamoto, N.; Liao, J.; Watanabe, S.; Nakata, K. Effect of Intermetallic Compound Layer on Tensile Strength of Dissimilar
Friction-Stir Weld of a High Strength Mg Alloy and Al Alloy. Mater. Trans. 2009, 50, 2833–2838. [CrossRef]
29. Murr, L.E.; Li, Y.; Flores, R.D.; Trillo, E.; McClure, J.C. Intercalation vortices and related microstructural features in the friction-stir
welding of dissimilar metals. Mater. Res. Innov. 1998, 2, 150–163. [CrossRef]
30. Rhodes, C.G.; Mahoney, M.W.; Bingel, W.H.; Spurling, R.A.; Bampton, C.C. Effects of friction stir welding on microstructure of
7075 aluminum. Scr. Mater. 1997, 36, 69–75. [CrossRef]
31. He, C.; Zhang, Z.; Li, Y.; Wei, J.; Zhai, M.; Zhao, S.; Zhao, X. Interface Characteristics and Mechanical Properties of Ultrasonic-
Assisted Friction Stir Lap Welded 7075-T6 Aluminium Alloy. Materials 2020, 13, 5335. [CrossRef] [PubMed]
32. Park, S.; Yoon, T.; Kang, C. Effects of the shoulder diameter and weld pitch on the tensile shear load in friction-stir welding of
AA6111/AA5023 aluminum alloys. J. Mater. Process. Technol. 2017, 241, 112–119. [CrossRef]
33. Yazdanian, S.; Chen, Z.W.; Littlefair, G. Effects of friction stir lap welding parameters on weld features on advancing side and
fracture strength of AA6060-T5 welds. J. Mater. Sci. Technol. 2012, 47, 1251–1261. [CrossRef]
34. Niu, S.; Ji, S.; Yan, D.; Meng, C.; Xiong, X. AZ31B/7075-T6 alloys friction stir lap welding with a zinc interlayer.
J. Mater. Process. Technol. 2018, 263, 82–90. [CrossRef]
35. Zettler, R.; Silva, A.; Rodrigues, S.; Blanco, A.; Santos, J. Dissimilar Al to Mg Alloy Friction Stir Welds. Adv. Eng. Mater.
2006, 8, 415–421. [CrossRef]
36. Yang, Y.K.; Dong, H.; Kou, S. Liquation tendency and liquid-film formation in friction stir spot welding. Weld. J. 2008, 87, 202–211.
37. Sato, Y.S.; Park, S.H.C.; Michiuchi, M.; Kokawa, H. Constitutional liquation during dissimilar friction stir welding of Al and Mg
alloys. Scr. Mater. 2004, 50, 1233–1236. [CrossRef]
38. Kostka, A.; Coelho, R.S.; Santos, J.D.; Pyzalla, A.R. Microstructure of friction stir welding of aluminium alloy to magnesium alloy.
Scr. Mater. 2009, 60, 953–956. [CrossRef]
39. Wei, Y.; Li, J.; Xiong, J.; Huang, F.; Zhang, F. Microstructures and mechanical properties of magnesium alloy and stainless steel
weld-joint made by friction stir lap welding. Mater. Des. 2012, 33, 111–114. [CrossRef]