Materials

1. Contents lists available at ScienceDirect
Journal of Manufacturing Processes
journal homepage: www.elsevier.com/locate/manpro
Dieless friction stir lap joining of AA 5050-H32 with AA 6061-T6 at varying
pre-drilled hole diameters
Tinu P. Sajua
, R. Ganesh Narayananb,
*
a
Department of Mechanical Engineering, TKM College of Engineering, Kollam 691005, Kerala, India
b
Department of Mechanical Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, Assam, India
A R T I C L E I N F O
Keywords:
Friction stir forming
Friction stir spot welding
Dieless friction stir forming
Dieless friction stir extrusion
Microstructure
Aluminum alloy
Hardness
A B S T R A C T
The role of hole diameter on mechanical behavior and joint formation of dieless friction stir lap joints in dis-
similar alloy grades of aluminum is investigated in the present article. Concurrent occurrence of mechanical
interlocking along with metallurgical bonding reinforce the fabricated joints. Keyhole defect and hook defect of
conventional friction stir spot welding are apparently suppressed. Hole diameter range from 3 to 3.5 mm is found
to be favorable, while considering the macrostructure analysis, joint morphology analysis and load performance
analysis. Appreciable fracture load up to 7 kN is obtained in lap shear test throughout the hole diameters
employed. Zones present in the joint are identified, in addition the effect of size of the hole on mechanical
interlocking is exposed through macrostructure analysis. The extent of recrystallization induced by frictional
heat generation along with plastic deformation has affected the grain structure of the joint. Critical zones re-
sulting in the failure of the joints are stir spot circumference and neck of the pin. A tool diameter to hole
diameter ratio for successful DFSL joint fabrication is also established through the present work.
1. Introduction
Lightweight sheet metal fabrication has achieved at most im-
portance in aerospace and automobile industries. Use of aluminum al-
loys as body panel for automobiles yield fuel economy along with re-
duced greenhouse gas emissions, improved vehicle safety and better
stability. High coefficient of thermal expansion, high thermal con-
ductivity and surface oxide layer formation hinder the application of
fusion welding for aluminum alloys. Alternatives such as self-pierce
riveting adds extra mass, extra cost and create recyclability issues.
Clinching is not suitable for sliding surfaces and are susceptible to vi-
brations. However, friction stir dependent joining techniques namely,
Friction Stir Spot Welding (FSSW) and Friction Stir Welding (FSW) have
considerable significance in light-weight sheet metal assembly and
fabrication.
A few articles presenting the friction stir based joining techniques
are quoted here. A study on the impact of change in process parameters
upon the shearing resistance of FSSWed AA 6061 sheets with thickness
0.9 mm, revealed that rotational speed of the stir tool is the primary
variable affecting the joint strength followed by pin profile of tool and
dwell time [1]. Energy input for FSSW in AA 6061-T6 sheets was in-
vestigated by varying the tool rotational speed and dwell time [2]. The
energy developed linearly varies with the shear strength in tensile
loading. Optimum energy level was found to be from 4.2 kJ to 6.3 kJ for
sheets of 2 mm thickness. Though an open loop control system based on
energy generation was constructed, defects like pinhole defect and hook
defect hinder achieving significant improvement in FSSW joint
strength. Variants such as refill FSSW and pinless/ probless FSSW are
successful in eliminating these defects. Li et al. conducted pinless FSSW
in 2198-T8 aluminum alloy for eliminating pinhole defect [3]. How-
ever, the hook defect affected shear resistance and failure mode of the
joints. In another attempt to eliminate pinhole, 6111-T4 aluminum
alloy sheets of 0.93 mm thickness were spot joined with pinless FSSW
process [4]. Short welding time, negligible heat affected zone, pene-
tration of plastic zone into the lower sheet and considerable grain re-
finement were achieved with pinless stir tool. Elimination of pinhole
formation and hook defect using two stage FSSW process was proposed
by Li et al. [5]. Hook defect developed in the first stage was eliminated
by performing FSW over the hook formation with a pinless stir tool.
However, size of the stir spot nugget was affected. A two stage refilled
FSSW for aluminum alloys specifically, AA 1100 sheets was proposed
by Moosa et al. [6] for removing pinhole defect in FSSW. Strength of the
FSSW joints increased with pinless tool shoulder diameter as well as
reduction of tool rotational speed. Refill FSSWed AA 5754 and AlSi
coated steel sheets showed that eutectic phase coating on the steel
surface performed a vital role in the spot welding of dissimilar sheets
https://doi.org/10.1016/j.jmapro.2020.01.048
Received 10 September 2019; Received in revised form 25 January 2020; Accepted 27 January 2020
⁎
Corresponding author.
E-mail addresses: tinu.saju@iitg.ac.in (T.P. Saju), ganu@iitg.ac.in (R.G. Narayanan).
Journal of Manufacturing Processes 53 (2020) 21–33
1526-6125/ © 2020 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
T

2. [7]. Joint lap shear strength increased with the tool plunge depth. 2 mm
thick AA 6061-T6 sheets spot joined with a pinless FSSW tool and ro-
tating anvil set up improved the strength of the joint as well as reduced
the tool forces [8]. Pinhole defect was successfully eliminated; how-
ever, hook defect was retained. AA 6061 aluminum sheet with low
carbon sheet was spot joined by friction stir extrusion, for which 6 kN
joint strength was obtained [9]. Extrusion and interlocking of alu-
minum into a pre-formed groove in steel sheet created the joint. It is to
be noted that simultaneous elimination of pinhole defect and hook
defect was not achieved in most cases, and hence, a significant im-
provement in the joint strength was not reported.
A new variant of FSSW namely Friction Stir Form (FSF) joining was
developed for creating lap joints in aluminum alloy to automotive steel
sheets. There is no formation of pin hole and hook defect in this
method. Metallurgical bonding and mechanical interlocking by rivet
head formation are the major strengthening mechanisms of the spot
joint. In this method, instead of stir mixing the two sheets, the stirred
upper sheet is forged and force extruded into an anvil cavity through a
hole pre-fabricated in the lower sheet. Anvil cavity realizes the rivet
head formation. Strength of the spot joint made between AISI 5182 and
AA 6014, and zinc coated mild steel by FSF was found to be superior
than the strength of self-pierced riveting [10]. Nishihara and Ito pro-
posed cladding of S45C steel with AA 6061-T6 aluminum alloy em-
ploying FSF [11] at tool traverse speed of 150 mm/min and tool rota-
tional speed of 705 RPM. Lazarevic et al. also mentioned that tool
plunge depth, tool diameter as well as anvil cavity design have vital
influence on strength and evolution of FSF joints [12]. Refilling the
FSSW stir spot with FSF process was performed by Prakash et al. [13].
Second stage refilling of FSSWed joints on 3 mm thick AA 6061-T6
sheets was performed using another stir tool and die setup. Refilling
increased the nugget cross-sectional area and consequently increased
tensile shear strength and joint efficiency was recorded. In another
work, FSF process was performed for refilling of FSSW stir spot by
Venukumar et al. [14]. Refilled FSSW joints in 2 mm thick AA 6061-T6
sheets possessed better fatigue strength than conventional FSSW joints.
FSF joining in aluminum alloys particularly AA 5052-H32 with AA
6061-T6 sheets attempted by Saju and Narayanan [15] revealed that
the rotational speed governs the hole closure beyond 3000 RPM, and
lack of pin formation restricts the joint strength. Similar effect was
observed when plunge depth of the stir tool was changed between 0.5
mm to 0.7 mm [16]. Even though high strength FSF joints can be
fabricated between sheets of dissimilar metals like automotive steel and
aluminum, the process stands as an unreliable option for spot joining
dissimilar aluminum alloy sheets mainly because of pre-drilled hole
closure and lack of mechanical interlocking.
An enhanced variant of FSF process called Dieless Friction Stir Lap
(DFSL) joining (Indian patent application No: 201741040528) mainly
for ductile metallic sheets such as alloys of aluminum is presented in
this work. Sheets of AA 5052-H32 and AA 6061-T6, each with 2 mm
thickness, are spot joined. Lack of pin interlocking in FSF process is
compensated by virtue of controlled deformation of lower sheet. Anvil
surface cavity employed in FSF process is not at all required in the new
process. Out of all parameters which influence the strength and for-
mation of the joint, effect of variation in the pre-fabricated Hole
Diameter (HD) is investigated through the present work. Various static
load performance analysis tests were performed to reveal the relation
between joint strength, its macrostructure and microstructure, hard-
ness, external morphological features and modes of failure. A study on
the impact of change in tool diameter on strength and formation of
DFSL joints was performed earlier [17]. A preliminary investigation on
the effect of HD during DFSL joining was also conducted [18] and a
comprehensive analysis is extended in the present work.
2. Methodology
2.1. Principle
The process sequence of DFSL process is schematically illustrated in
Fig. 1. The sheets are clamped in lap joint configuration with the lower
sheet possessing a pre-drilled hole at the desired stir spot location
(Stage 1). When the rotating stir tool is plunged over the upper sheet,
frictional heat flux generated at the stir spot plasticizes the upper sheet
metal. Plunge of the stir tool results in forging and extruding the upper
sheet metal through the pre-drilled hole in the lower sheet resulting in
pin formation (Stage 2). Meanwhile, the heat conduct to the lower
sheet also, resulting in metallurgical bonding at the interface between
the two sheets. When the stir tool is plunged further, the extruded pin is
locked by the controlled deformation of the lower sheet. The growth of
an inward collar (as shown in Fig. 1) into the pre-drilled hole leads to
the mechanical interlocking. In other words, the top of the pre-drilled
hole deforms into a collar shaped structure, which enables neck for-
mation across the extruded pin. This prevents the retraction of the pin
from the pre-drilled hole (Stage 3). Therefore, simultaneous mechanical
interlocking and metallurgical bonding are realized in a DFSL joint.
Pinless stir tool prevents the pinhole formation and hook defect for-
mation is also eliminated by preventing stir mixing of the upper and
lower sheets in the DFSL process. The process has the provision for
producing single/ multi-hole configurations. The stir spot of DFSL joint
has similar external appearance of pinless FSSWed joint, however, the
mode of joint formation is different.
2.2. DFSL experiments
A pinless flat faced stir tool and fixture with clamps form the basic
Fig. 1. Stages of DFSL process (Not to scale).
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
22

3. experimental set up for the DFSF process as shown in Fig. 2. The set up
was mounted on Kirloskar Viking KTM 40 CNC milling machine. The
mild steel fixture is capable of clamping various samples for mechanical
performance tests. The flat faced pinless friction stir tool is fabricated
on H13 tool steel with 14 mm shoulder diameter and 25 mm length.
Aluminum alloys namely AA 5052-H32 and AA 6061-T6 sheets of 2 mm
thickness are chosen as upper and lower sheets respectively. The me-
chanical properties of the sheet metals obtained through standard
procedures are given in Table. 1. The mechanical properties of the two
sheet metals measured with standard tensile tests are given in Table 1.
ASTM E8-09 [19] and ASTM E517 [20] standards are followed for the
tensile test. The measurement of grain size was conducted as per ASTM
E112 standard [21]. The chemical composition of the sheets obtained
through Energy Dispersive X-Ray analysis is provided in the Table 2.
The pre-drilled hole diameter, the primary variable of the present
work, is varied in three levels, namely the lower level at 2.5 mm and 3
mm, the medium level at 3.5 mm and 4 mm and the higher level at 4.5
mm and 5 mm. Minimum HD is kept at 2.5 mm with an intention to
avoid possible damage (closure) of the pre-drilled hole. Similarly, the
maximum HD is restricted to 5 mm to keep the metallurgically bonded
area unaffected. Other process parameters particularly tool plunge
depth, direction of tool rotation, tool rotational speed and tool plunge
rate are kept constant at 0.5 mm, clockwise direction, 500 RPM, and
0.002 mm/s, respectively from past experimental trials [17,18].
2.3. Macrostructure, microstructure, measurement of hardness and analysis
of joint morphology
Macrostructure analysis was conducted to identify the distinct zones
in the joint, joint evolution, metal flow pattern and critical weak zones
in the joint. Identification of grain structure, grain size and the identi-
fication of the boundary between various zones were conducted with
microstructure analysis. Rough finishing of cross-sections was done
with emery abrasive paper of 2000 grade as well as fine finishing was
achieved using a velvet cloth wetted with polishing liquid. Sample
preparation for macrostructure analysis as well as microstructure ana-
lysis were done as per ASTM E407 [22] standard. Kellers reagent (3 ml
HCl, 5 ml HNO3, 2 ml HF, 190 ml distilled water) was used in etching
treatment for a period of 20–40 sec. Microscopic images were obtained
at 200X magnification and macroscopic images were obtained at 50X
magnification using Zeiss Axiocam MR3 microscope.
Hardness measurements using Buehler MMT3-B Vickers micro
hardness tester and schematic of hardness measurements are as shown
in Fig. 3. Each indentations was performed with 500 gf load for 10 sec
duration as per the standard, ASTM E92 [23].
External morphological features such as upper sheet bulging, flash
width and flash height (Fig. 4), which affect the aesthetic appearance of
the joint were measured with Dinolite Dinocapture 2 USB digital mi-
croscope.
2.4. Load performance analysis tests
Fracture load and extensibility of DFSL samples were evaluated
through variable loading conditions induced by lap shear, cross-tension
and peel tests. Formability of DFSL samples were evaluated with uni-
axial tensile test. Load performance test were conducted on 250 kN
servo hydraulic universal testing machine (BISS Median 250), shown in
Fig. 5. Samples for load performance analysis tests were prepared as per
AWS D8.9 standard [24]. Fig. 6 shows the dimensions of samples with
rolling direction (RD) through the length of the samples. Axis of DFSL
joint passes through the stir spot center lying orthogonal to plane of the
two sheets. The extension rate was maintained at 1 mm/min. The joint
fracture load, the extension at fracture and the failure mode were re-
corded. Two samples were prepared for load performance tests and the
average output value was recorded.
A comparative analysis of lap shear fracture load obtained for DFSL
samples with that of samples with conventional joints, such as FSF and
FSSW, in same material combination is also provided in the present
work. DFSL sample fabricated with HD of 3 mm is chosen for com-
parison. For conventional FSF process, a pre-formed hole having 3 mm
diameter was employed [16]. A hemispherical shaped anvil cavity with
0.55 mm depth and 3.5 mm diameter was employed for the rivet head
formation. FSSW joints were fabricated with a pinless tool as well as
tool having pin. A tool with a pin having 1 mm length, 4 mm diameter
and 14 mm shoulder diameter was used for FSSW [25]. For pinless
FSSW process, the pinless stir tool used in DFSL process was employed.
For uniformity in process execution, all other process parameters were
kept same for all the four processes. Other process parameters parti-
cularly tool plunge depth, direction of tool rotation, tool rotational
speed and tool plunge rate are kept constant at 0.5 mm, clockwise di-
rection, 500 RPM, and 0.002 mm/s, respectively for all smaples.
Fig. 2. Experimental setup used for DFSL process.
Table 1
Parent sheet metal mechanical properties (along the 0° rolling direction).
Mechanical properties AA 5052-H32
(upper sheet)
AA 6061-T6
(lower sheet)
Yield resistance in tension (MPa) 156.18 ± 2 224.95 ± 3
Tensile fracture resistance (MPa) 212.10 ± 2 307.99 ± 5
Total sample elongation (%) 14.62 ± 7 22.51 ± 3
Vickers hardness (HV) 76.9 ± 4 98.9 ± 1
Strain hardening exponent, n 0.159 ± 0.005 0.16 ± 0.001
Strength coefficient, K (MPa) 353.12 ± 4 491.08 ± 2
Plastic strain ratio, R 0.561 0.745
Average Grain diameter (μm) 22.5 11.2
Table 2
Composition in parent sheet metals (% weight).
Material Fe Mn Cr Si Cu Mg Zn Ti Al
AA 5052-H32 0.22 0.14 0.24 ≤ 0.19 ≤ 0.20 3.60 ≤ 0.32 – Remaining
AA 6061-T6 0.31 ≤ 0.11 0.26 0.78 0.24 1.61 ≤ 0.28 ≤ 0.14 Remaining
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
23

4. 3. Results and discussion
A comprehensive discussion about the effect of HD on the evolution
of joint macrostructure and microstructure, hardness, external joint
features, load performance analysis and failure modes are presented in
the following sub-sections.
3.1. Joint formation analysis through macrostructure
DFSL samples fabricated at HD of 2.5 mm from lower level, 3.5 mm
from medium level and 5 mm from higher level are choosen in mac-
rostructure analysis. In addition, metal flow direction during the joint
formation are also identified. The macrostructure with distinct zones
marked and the schematic illustration of the metal flow directions are
given in Figs. 7 and 8, respectively. As said earlier, DFSL joints are
realized by concurrent occurrence of metallurgical bonding and me-
chanical interlocking. The effect of change in HD over these strength-
ening mechanisms is discussed here. The mechanical interlocking in
DFSL joints is effected by the formation of inward collar in the lower
sheet as shown in Figs. 7(b–c)–8 (b–c). It prevents the retraction of
extruded upper sheet metal. When upper sheet is subjected to plastic
deformation induced by the stirring motion of the tool and its down-
ward plunge, lower sheet undergoes plastic deformation by virtue of the
conducted frictional heat flux and the tool plunge force. The extruded
pin forms the central part of the DFSL joint cross-section. It is sur-
rounded by the interface bonded region of the two sheets.
The various zones in DFSL joint are Stir Zone (SZ), Heat Affected
Zone (HAZ), Plastically Deformed metal flow Zone (PDZ) and Thermo-
mechanically affected zone (TMAZ) as marked in Figs. 7 and 8. Similar
zones were also identified in FSSW joints and refill FSSW joints [7].
However, PDZ is unique for DFSL and conventional FSF joints [14].
The region, which undergoes stirring induced severe plastic de-
formation forms SZ. It is located at the central part of the stir spot in
upper sheet, where stir tool directly contacts the upper sheet. The ‘U’
shaped SZ has maximum depth at the center. Similar ‘basin’ shaped SZs,
which is symmetrical about the tool axis, was also reported during
probeless FSSW process [3] and refill FSSW process [7].
The PDZ is positioned below SZ as well as at the center of the pre-
fabricated hole in the lower sheet. A part of upper sheet, which un-
dergoes plastic deformation and extrusion through the this hole forms
the PDZ. Directional extrusion of PDZ into the pre-drilled hole is rea-
lized by plunging action of the stirring tool. The collar, which forms a
neck around this extruded PDZ, realizes the mechanical interlocking.
The PDZ receives frictional heat transferred from SZ. Unlike SZ, mate-
rial of the PDZ is not subjected to stirring.
The surrounding regions of the SZ as well as certain regions on the
sidewalls constitute the TMAZ region. The TMAZ is not undergoing
stirring induced severe plastic deformation, instead plunging of stir tool
has significant impact on its formation. In addition, ring-shaped TMAZ
region located around the stir spot side walls undergoes plastic de-
formation due to the constant rubbing over the lateral curved surface of
rotating tool. In addition, TMAZ regions are under the effect of fric-
tional heat generated from the SZ.
Regions of the two sheets surrounding the above three zones in the
DFSL joint constitute the HAZ. Significant plastic deformation is not
observed in the HAZ, instead frictional heat flux generated micro-
structural changes. Effect of change in HD over the formation of these
zones are discussed below. The internal joint features such as upward
deformation in the lower sheet, diameter of the neck of pin and collar
length are shown in Fig. 8c.
Lower HDs (2.5 mm and 3 mm): At lower HDs, the hole size is so
small that possibility of pre-drilled hole closure is high. Pre-drilled hole
closure at 2.5 mm HD is shown in Figs. 7a and 8 a. The deformed upper
sheet material is subjected to partial extrusion into the pre-drilled hole.
During extrusion, plastic deformation of the region around the hole has
resulted in its closure. Due to this, the extruded metal is isolated from
the upper sheet. Collar growth and subsequent pin formation is absent
at 2.5 mm HD. However, the DFSL samples with lower HDs are
strengthened by metallurgical bonding. With increase in HD, the pos-
sibility of closure of the hole is comparatively low. Closure of the pre-
drilled hole is absent for 3 mm HD. The hook defect formation is not
observed at lower HD levels, however, average upward deformation of
lower sheet of about 2.45 ± 0.012 mm is observed at 2.5 mm HD.
Fig. 3. Measurement locations of hardness along the joint cross-section (Dimension not to scale).
Fig. 4. Morphological features measured in the DFSL joints.
Fig. 5. Universal testing machine used for load performance tests.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
24

5. Fig. 6. Schematic of samples used for load performance analysis tests (all dimensions in mm).
Fig. 7. Macrostructure images of DFSL joints fabricated with (a) 2.5 mm HD (Lower level), (b) 3.5 mm HD (Medium level), (c) 5 mm HD (Higher level).
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
25

6. Medium HDs (3.5 mm and 4 mm): The possibility of pre-drilled hole
closure is low at medium HDs. At medium HDs such as 3.5 mm, the
collar growth is observed inside the joint as shown in Fig. 7b as well as
in Fig. 8b, which interlocks the pin extruded from the upper sheet
metal. In addition, these joints are also metallurgical bonded. Similar
observation is seen at 4 mm HD. With increase in the HD value, the PDZ
size is increased. Neck diameter of the pin and collar length about
4.15 ± 0.001 mm and 0.83 ± 0.017 mm are obtained, respectively at
3.5 mm HD. Upward deformation of lower sheet is about 2.48 ± 0.004
mm at 3.5 mm HD. However, no hook defect is observed.
Higher HDs (4.5 mm and 5 mm): With increase in HD to higher level
such as 5 mm, major improvement in the mechanical interlocking is not
observed. However, the collar growth is perfect and the samples are
metallurgically bonded. The PDZ size is increased due to larger HD.
Neck diameter of the pin is increased to 5.51 ± 0.007 mm at 5 mm HD.
A significant improvement in the collar length is not observed. It is
about 0.83 ± 0.069 mm at 5 mm HD. The surrounding metallurgically
bonded area has been decreased due to larger pin size (Figs. 7c and 8 c).
Further, upward deformation of the lower sheet remains more or less
same as lower and medium HDs. However, the hook defect is absent.
Similar situation exists at 4.5 mm HD.
To summarize, HD range from 3 mm to 5 mm can be considered to
be optimum from macrostructure analysis on conditions such as the
probability of closure of the hole is less and significant metallurgically
bonded area is present in the joint interface. Increase in diameter of the
neck of the extruded pin is observed with increase in the HD. However,
with increase in HD, a significant improvement in strength is not
achieved by the growth of these internal joint features. The upward
deformation in the lower sheet remains same with increase in HD,
which shows that it has not reached up to such severity to act as hook
defect in DFSL joints.
A brief comparison about SZ in DFSL and FSSW processes is given
below. For DFSL joints, the SZ size remains unchanged throughout the
entire HD range and the SZ region is constrained to the upper sheet
only. SZ has no significant influence in joint formation other than
generating sufficient heat for plastic deformation and extrusion of
upper sheet metal. Similar shallow deformation zone was also reported
during FSSW process with pinless flat stir tool [4]. Unlike FSSW, the
intermixing of the two sheet metals by stirring is absent in DFSL joints
[17]. As a result, the hook defect is absent in DFSL joints.
3.2. Microstructure characterization
Fig. 9 depicts the microstructures of SZ, PDZ, HAZ and TMAZ at
200X magnification for the joint fabricated at 3.5 mm HD. The corre-
sponding regions on the macrostructure of DFSL joint is shown in boxes
marked with alphabets A to H in Fig. 8b. Literature reports that the
intense plastic deformation as well as high frictional heat generation
inside SZ can bring recrystallization in and around the SZ [26]. Various
alloys of aluminum can easily undergo dynamic recrystallization due to
its considerable stacking fault energy, which leads to high recovery
rate. Similar observations are also observed in DFSL joints.
For DFSL joints, the PDZ and HAZ have undergone recrystallization
with further grain growth due to frictional heat conducted from the stir
spot. SZ has undergone recrystallization due to frictional heat flux along
with severe plastic deformation. TMAZ has undergone grain recovery
without recrystallization. Similar recrystallization behavior of alu-
minum alloys was also reported during refill FSSW process [7]. The
lower sheet is mounted over the top of the backing plate so that the heat
dissipation becomes faster than that of the upper sheet. This could also
contribute to variable microstructural changes for the two sheets. The
comparison of grain size of various zones of the DFSL joint is given in
Table 3.
HAZ of upper sheet, Fig. 9a, possesses coarse grains (with grain size
of 44.9 μm) than that of the HAZ of the lower sheet (with grain size 22.5
μm), Fig. 9b. Recrystallization, and grain growth of about 200 % is
observed in the HAZ of the two sheets, while comparing with corre-
sponding grain size of the parent sheet (Table 1). This is because of stir
friction heat conducted from the joint spot. However, only 41 % growth
in the grain size is observed in HAZ of lower sheet just under the joint
Fig. 8. Schematic illustration of the cross-section of DFSL joints at (a) 2.5 mm HD, (b) 3.5 mm HD, (c) 5 mm HD with direction of metal flow indicated.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
26

7. spot as shown in Fig. 9c and d. These regions have more tendency to-
wards prolonged recrystallization because of its location just under the
stir spot. The grain growth in these regions might be arrested by the
plastic deformation created by the plunging of the stir tool.
Equi-axed fine grains are observed in SZ region (Fig. 9e) with grain
size about 5.6 μm. The stirring induced plastic deformation and heat
generation by friction have brought about dynamic recrystallization of
the SZ. Identical observation was also reported in conventional FSSW
process [26]. Fine equi-axed grains represents more intensive stirring.
Fine grains in the SZ and coarse grains in the HAZ show that other than
heat flux, the grain growth is dictated by the severity of plastic de-
formation. Stirring induced plastic deformation arrests the SZ grain
growth, while HAZ with no plastic deformation possesses larger grains.
Transition from HAZ grains to TMAZ grains is shown in Fig. 9f. TMAZ
grains are also smaller than upper sheet HAZ grains, with grain size
about 18.9 μm. The rubbing contact with the curved lateral surface of
stir tool and heat flux generated highly deformed grains in the TMAZ.
Recovered grains without recrystallization is observed in the TMAZ.
The strain induced by plastic deformation is insufficient to create
Fig. 9. Microstructure images of (a) the upper sheet, AA 5052-H32 HAZ, (b) the lower sheet, AA 6061-T6 HAZ, (c) AA 6061-T6 HAZ (below stir spot side wall), (d)
AA 6061-T6 HAZ (below SZ), (e) SZ, (f) TMAZ- HAZ boundary, (g) PDZ and (h) SZ-PDZ boundary with 200X magnification (The regions from which microscopic
images obtained are marked with boxes in the Fig. 8b).
Table 3
Grain size comparison of various zones of the DFSL joint.
Zone Average grain size (μm)
SZ 5.6
AA 6061-T6 HAZ (below stir spot side wall) 15.9
AA 6061-T6 HAZ (below SZ) 15.9
TMAZ 18.9
AA 6061-T6 HAZ (lower sheet) 22.5
PDZ 26.7
AA 5052-H32 HAZ (upper sheet) 44.9
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
27

8. recrystallization in the TMAZ. Similar recovered grains were also ob-
tained in the TMAZ of FSWed AA 5083-O sheets [26,27].
The PDZ grains are statically recrystallized (with grain size of 26.7
μm), whose size is almost comparable with that of the parent sheet
metal (Fig. 9g). Unlike SZ, the PDZ is not subjected to severe plastic
deformation caused by stirring. However, frictional heat and directional
extrusion towards the pre-drilled hole during the plunging of the tool
result in static recrystallization followed by further grain growth. SZ
possess circular motion along with the rotation of the tool, while PDZ
undergoes radial inward movement and extrusion into the lower sheet
hole during the tool plunge. Therefore, extent of grain refinement of
these two regions are entirely different. A clear boundary is observed
between the grain structures of the two region as shown in Fig. 9h.
To summarize, the frictional heat and induced plastic deformation
have introduced microstructural changes such as recrystallization and
grain growth in the DFSL joint. This leads to the distinction of various
zones of the DFSL joint, which are identical to the zones observed in
other friction stir based joining processes.
3.3. Hardness variation
Fig. 10 shows the variation of hardness in the DFSL joint. The effect
of frictional heat and plastic deformation on DFSL samples are evident
through the notable decrease in hardness than that of the parent metal.
The hardness variation is not influenced by the change in HD as uni-
form profiles are observed in Fig. 10a,b.
The upper array of indentations possess an increase in hardness
towards the SZ. This is due to strain hardening during extrusion and
further forging of upper sheet metal through pre-drilled hole. However,
Kesharwani et al. [28] reported that the increase in hardness of the SZ
due to fine equi-axed grain formation. The decrease in hardness to-
wards the periphery of the joint is attributed to softening of the HAZ
due to recrystallization and grain growth.
The lower array of indentations possess uniform hardness. However,
the slight increase in hardness at the PDZ occurs due to the forging
action of the stir tool. The lower array of indentations are mostly lo-
cated in the lower sheet HAZ region.
Nature of change in hardness of the two sheets are different because
upper sheet is a solid solution-hardened alloy (of type non-heat
treatable) and the lower sheet is a precipitation hardened alloy (of type
heat treatable). It is to be noted that the parent sheet possessing coarse
grains shows similar hardness values as the SZ possessing fine grains in
the upper sheet. Therefore, the grain structure is not only the factor
controlling hardness of the SZ. It is evident from literature that in ad-
dition to grain boundary strengthening, factors such as dislocation
density, precipitation hardening and solid solution hardening play
major role in controlling the hardness of AA 5XXX alloys [29]. The
hardness of AA 5XXX series alloys is dictated by Orowan hardening; the
homogenous distribution of particles (such as compounds of iron, alu-
minum and manganese) leads to strengthening of the alloy [26]. Small
particles, with size less than 0.5 μm, distributed in the grain interiors
acted as obstacles to mobility of dislocations in AA 5083 alloy [27].
Furthermore, Huskins et al. reported that the precipitates such as Mn
strengthen the aluminum alloy namely Al-5083 H-131 by acting as
obstacles to dislocation motions. Pritam et al. [30] reported that the
fragmentation of second phase particles and the dislocation density
affected the hardness in the SZ in FSSWed AA 5052-H32. Hence, it can
be understood that the hardness profile of the upper sheet of DFSL
joints is determined by dislocation density rather than grain size.
Literature reports that hardness of the AA 6061-T6 is greatly in-
fluenced by the distribution of precipitates rather than grain structure
[30]. The softening of the HAZ is created by the coarsening and dis-
solution of strengthening precipitates under frictional heat flux. Re-
duction in hardness than that of the parent sheet due to coarsening of
the precipitates was also reported in FSSW of AA 6061-T6 [31].
3.4. Morphology analysis of the joint
Effect of HD on the size of morphological features of the joint such
as width and height of flash and bulging of upper sheet, which affect the
aesthetic appearance of DFSL joint is shown in Fig. 11. The outward
material flow direction beneath the shoulder of a pinless stir tool leads
to weld flash formation [8]. Slight decrease in flash width of about 1.15
mm is observed when HD increases from 2.5 mm to 5 mm (Fig. 11a).
However, the flash height remains same throughout the change in HD
(Fig. 11b). The small reduction in the overall flash size is contributed by
the increase in HD. At 2.5 mm HD, the closure of the hole results in
outward material flow in DFSL sample. With increase in HD, the plunge
Fig. 10. Comparison of hardness along the cross-section of DFSL joints. (a) Indentation along the upper array, (b) Indentation along the lower array (Hardness
variation at each indentation is approximately ± 2.1).
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
28

9. of the tool results in extruding more upper sheet material towards the
hole in the lower sheet, instead of pushing material of upper sheet to
the periphery of the stir spot as flash.
A decrease in upper sheet bulging is observed between 2.5 mm and
4.5 mm HD (Fig. 11c). At lower HDs such as 2.5 mm, because of closure
of hole in the lower sheet, the upper sheet is bulged outwards during
the plunging of the stir tool. At higher HD (say at 4.5 mm) the extrusion
of upper sheet is easy and there is no closure of hole reducing the upper
sheet bulging.
To summarize, with increase in HD to medium and higher levels, the
size of the external morphological features, which impact the aesthetic
appearance of the joint decreased. The pre-drilled hole closure followed
by the extrusion of upper sheet into the hole have opposite effect on
development of external morphological features. HD range from 3 mm
to 5 mm can be fixed as optimum from the morphology analysis.
3.5. Load performance analysis tests
Influence of HD on joint fracture load of DFSL samples is shown in
Fig. 12. The HD for which the maximum fracture load obtained varies
for different tests. Lap shear samples exhibit almost identical fracture
load throughout the HD range. Maximum lap shear fracture load is
recorded for 4.5 mm HD and minimum for 4 mm HD. Maximum cross-
tensile fracture load is obtained at 2.5 mm HD and minimum at 5 mm
HD. For peel test, the fracture load is maximum for 3.5 mm HD and
minimum for 5 mm HD. However, it is observed that DFSL samples at 3
mm and 3.5 mm HDs show appreciable overall performance in all the
load performance analysis tests. DFSL samples at 3 mm HD show 95 %
of maximum achieved lap shear fracture load (7.42 kN), 83 % of
maximum achieved cross-tension fracture load (2.89 kN) and 90 % of
the maximum achieved peel fracture load (1.05 kN). Samples at 3.5 mm
HD show 92 % of the maximum achieved lap shear fracture load (7.18
kN), 88 % of the maximum achieved cross-tension fracture load (3.07
kN) and it also possess the maximum achieved peel fracture load (1.17
kN). Therefore, 3 mm and 3.5 mm HDs can be chosen as the optimum
range for better load performance of the present material combination.
At higher HDs like 4.5 mm and 5 mm, samples show minimum fracture
load in cross-tension and peel test. 56 % decrease of cross-tension
fracture load and 44 % decrease of peel fracture load is obtained be-
tween 3.5 mm and 5 mm. For any particular HD, lap shear samples
exhibit maximum fracture load, cross-tension samples exhibit inter-
mediate fracture load, while peel test samples show minimum fracture
load. During friction stir riveting process by extrusion, it was reported
by Evans et al. that strong correlation between joint strength and hole
size cannot be established during load performance analysis tests [32].
However, the joint strength of DFSL samples show notable variation
with change in HD.
Sajed et al. reported that larger displacement/extensibility can be
treated as indication of energy absorption before failure and further
accounts to its safety consideration of the joints [6]. In DFSL samples,
the extensibility is about 1 mm in lap shear samples, 13 mm in cross-
tension samples, while it is 26 mm in peel test samples. DFSL samples
show identical fracture load in uniaxial tensile tests. Fracture load of
21.46 kN and average extension about 4.24 mm are recorded
throughout the HD range.
To summarize, the HD at which maximum fracture load obtained
for each of the load performance analysis test is not unique and it de-
pends on the nature of loading. The effect of HD is also significant.
However, 3 mm and 3.5 mm HDs can be treated as optimum for the
present material combination considering mild loss in overall load
performance.
3.6. Failure mode analysis during load performance tests
The modes of failure observed in load performance analysis tests are
shown in Fig. 13 and summarized in Table 4. Different failure modes
such as combined pin pull-out and bond delamination, combined pin
shear and bond delamination, and tear off are observed for DFSL
samples.
Tear off: Characterized by fracture around the circumference of the
stir spot (Fig. 13a). Pin interlocking and metallurgical bonding remain
unaffected. Shear failure mode in two stage refilled FSSW joints is
Fig. 11. Variation of (a) Width of flash, (b) Height of flash, (c) Upper sheet bulging, with change in HD.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
29

10. analogous to tear off failure in DFSL samples [6].
Combined pin pull-out and bond delamination: Characterized by
retraction of pin due to damage of the collar associated with the dela-
mination of metallurgical bonding (Fig. 13b). It is analogous to nugget
pull-out observed during friction stir riveting by extrusion [32].
Combined pin shear and bond delamination: Characterized by shear
fracture of extruded pin across the neck along with delamination across
metallurgical bonding (Fig. 13c). The pin shearing results in retaining
Fig. 12. Fracture loads at various HDs during load performance analysis tests of DFSL samples.
Fig. 13. Different failure modes. (a) Tear-off failure, (b) Combine pin pull-out and bond delamination failure, (c) Combined pin shear and bond delamination failure,
(d) Base metal fracture and (e) Stir spot fracture.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
30

11. the extruded pin inside the lower sheet hole. Similar failure mode was
also observed in friction stir riveting by extrusion [32].
Base metal fracture: Occurs rarely due to base metal surrounding
the stir spot act as critical weak zone of failure. Transverse fracture
occurs in the base metal (Fig. 13d), while the stir spot is not affected.
The annealing induced by frictional heat flux leads to the base metal
softening and further fracture.
Stir spot fracture: A common type of failure mode, which is char-
acterized by fracture along a transverse section of the stir spot
(Fig. 13e). Joint spot act as a critical weak zone in the gauge length
region leading to failure.
Critical zones which are responsible for initiating failure in DFSL
samples during load performance analysis tests are neck of the extruded
pin (CR1), stir spot circumference (CR2), as shown in Fig. 8a and b.
Shearing of the pin and pull-out failures are initiated from the neck of
the pin, CR1, which act as weakest zone susceptible to failure. The bond
delamination and tear-off failures originate from stir spot cir-
cumference, CR2. At maximum load, the fracture along the cir-
cumference of the stir spot lead to tear off failure and bond delamina-
tion failure initiates from partially bonded regions below the stir spot
circumference.
The failure modes occur randomly throughout the HD range for
various load performance analysis tests. However, combined pin pull-
out and bond delamination failure is the common failure mode. This
shows that the two critical zones are equally active in the DFSL joint.
3.7. Comparison between DFSL, CFSF and FSSW processes
Table 5 compares the shear fracture load obtained from lap shear
tests of joints made with DFSL, conventional FSF, FSSW with a tool
having pin, and without pin. The lap shear fracture load of DFSL
samples is 42 % higher than corresponding values of conventional FSF
samples, 86 % higher than corresponding values of FSSW samples made
with a tool having pin and 25 % higher than corresponding values of
pinless FSSW samples. The superior joint strength of DFSL samples is
evident from this comparison.
The cross-sectional views of DFSL, conventional FSF and FSSW
joints are shown in Fig. 14. Flat bottom surface and stir spot without pin
hole distinguish DFSL joints from conventional FSF and FSSW joints
made with a tool having pin. Therefore, the strength and overall aes-
thetic appearance of DFSL joints are superior than that of the conven-
tional FSF and the FSSW joints made with a tool having pin. Unlike in
conventional FSF joints, the rivet head formation is absent in DFSL
joints so that flat surface on the bottom make it suitable for sliding
surfaces.
The external appearance of the cross-sections of pinless FSSW joint
and DFSL joint are similar and the joint strength of pinless FSSW joints
is comparable with that of DFSL joints. However, for pinless FSSW
joints, the hook defect is not eliminated and the flow of upper sheet
metal in the thickness direction is comparatively limited [3]. The
drawback is resolved in DFSL joints by the controlled extrusion of the
upper sheet material through the lower sheet hole. Employing pinless
tool in FSSW of thicker sheets is ineffective and pinless tool induces
severe axial load on the FSSW tool holding set up [8]. However, the
pinless stir tool is sufficient for DFSL process irrespective of the sheet
Table
4
Failure
modes
of
DFSL
samples.
Hole
diameter
(mm)
Lap
shear
test
Cross-tension
test
Peel
test
Tensile
test
Trial
1
Trial
2
Trial
1
Trial
2
Trial
1
Trial
2
Trial
1
Trial
2
2.5
Pin
shear
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Tear-off
Pin
shear
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Stir
spot
fracture
Stir
spot
fracture
3
Pin
shear
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Tear-off
Pin
shear
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Base
metal
fracture
Stir
spot
fracture
3.5
Pin
pull
out
+
Bond
delamination
Tear-off
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Stir
spot
fracture
Stir
spot
fracture
4
Pin
shear
+
Bond
delamination
Pin
shear
+
Bond
delamination
Tear-off
Tear-off
Pin
pull
out
+
Bond
delamination
Pin
shear
+
Bond
delamination
Stir
spot
fracture
Stir
spot
fracture
4.5
Pin
shear
+
Bond
delamination
Tear-off
Tear-off
Pin
shear
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
shear
+
Bond
delamination
Stir
spot
fracture
Stir
spot
fracture
5
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Pin
pull
out
+
Bond
delamination
Stir
spot
fracture
Stir
spot
fracture
Table 5
Lap shear fracture load obtained for DFSL, conventional FSF and FSSW
samples.
Joint fabrication process Fracture load (kN)
DFSL 7.42 ± 0.21
Conventional FSF 5.23 ± 0.21
FSSW with stir tool having pin 3.99 ± 0.23
Pinless FSSW 5.93 ± 0.27
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
31

12. thickness.
It is understood that drilling a hole in DFSL process needs extra
fabrication time and fabrication cost. This is taken care of in DFSL
process by obtaining better joint strength through concurrent formation
of mechanical interlocking as well as metallurgical bonding. However,
dwell stage of stir tool is consciously avoided in DFSL process since it
adds extra time after the tool plunge. It may also eventually damage the
mechanical interlocking and metallurgical bonding in DFSL joints due
to acute plastic deformation and excessive frictional heat generation.
3.8. Tool diameter to hole diameter ratio for DFSL joints
A tool diameter to hole diameter ratio (TD-HD ratio) is obtained
from the analysis of the results of present work and the previous lit-
erature published by the authors. A study on the impact of change in TD
on formation and strength of DFSL joints was performed in the same set
of aluminum alloys and reported in literature [17].
It was observed that larger tool diameters were not effective due to
severe frictional heat generation and damage of the pre-fabricated hole.
Smaller tool diameters resulted in insufficient metallurgical bonding
and imperfect mechanical interlocking, and further, the closure of pre-
drilled hole was observed. Present work shows that larger HDs result in
reduction of metallurgically bonded area. Smaller HDs lead to closure
of the lower sheet hole. So there exist a relation between the tool dia-
meter and hole diameter employed for DFSL joint fabrication.
When the tool diameter was varied with constant HD of 3 mm,
successful joints were fabricated for tool diameter ranging from 10 mm
to 14 mm. Therefore, window of sound joint fabrication for these DFSL
samples lies in the TD-HD ratio of about 3.33–4.66. Similarly in the
present work, when the HD is varied with constant tool diameter of 14
mm, successful joints are fabricated for HDs ranging from 3 mm to 5
mm. Therefore, the window of sound joint fabrication for these DFSL
samples lies in the TD-HD ratio about 2.8 to 4.66. Therefore, it can be
summarized that the TD-HD ratio for successful DFSL joint fabrication
in aluminum alloy sheets of 2 mm thickness should lie in the range of
3.33–4.66.
4. Conclusion
The main aim of the present work is to explore the effect of HD upon
load performance and joint formation during DFSL joining. The con-
clusions derived from the present work are as follows.
• As obtained in typical friction stir-based spot joints, the stir zone,
thermo-mechanically affected zone and heat affected zone are
identified in DFSL joints also. Plastically deformed metal flow zone
is observed exclusively in DFSL joints and in conventional FSF
joints. Unlike in FSSW joints, the symmetrical ‘basin’ shaped stir
zone is limited only to upper sheet, hence the stir mixing of the two
sheets are absent in DFSL joints. Consequently, the hook defect is
eliminated in DFSL joints. Pinhole formation has been prevented
through the usage of a pinless stir tool.
• Macrostructure analysis revealed that for lower HDs like 2.5 mm,
the likelihood of pre-drilled hole closure is more and the mechanical
pin interlocking in not realized always. DFSL samples fabricated at
HD range from 3−5 mm are reinforced by concurrent mechanical
interlocking as well as metallurgical bonding. Increase in HD to
higher levels such as 5 mm leads to large pin size, but with a cor-
responding decrease in the metallurgically bonded area. The size of
internal joint features such as collar length and neck diameter of the
pin are appreciable at medium and higher HDs. The upward de-
formation of the lower sheet remains same with increase in HD;
however, it has not grown up to severity of hook defect.
• The heat generation by friction and the severe plastic deformation
have contributed microstructural changes such as dynamic re-
crystallization in the SZ, and static recrystallization and further
grain growth in HAZ and PDZ. SZ possess fine equi-axed grains with
5.6 μm grain diameter, while the PDZ grain diameter about 26.7 μm
is almost comparable with parent sheet metal grain diameter. The
TMAZ regions possess recovered grains without recrystallization.
• Considerable decrease in the hardness is noticed in stir spot than the
corresponding values of the parent sheet metal due to micro-
structural changes. The hardness of the joint spot is independent of
change in the HD so that uniform pattern of hardness profiles is
observed in both upper and lower array of indentations. Forging and
further extrusion of the upper sheet material into the pre-fabricated
hole results in higher hardness in the stir spot center.
• It is revealed through joint morphology study that for HD range of 3
mm–5 mm, the size of external morphological features namely flash
and upper sheet bulging are less. The extrusion of upper sheet metal
into the pre-drilled hole and the closure of the pre-drilled hole have
opposite effect on the development of external morphological fea-
tures.
• From load performance analysis tests, the overall performance is
found to be appreciable at 3 mm and 3.5 mm hole diameters with
lap shear fracture load about 7.18–7.42 kN, cross-tension fracture
load about 2.89–3.07 kN and peel test fracture load range about
1.05–1.17 kN. DFSL samples show appreciable extensibility
throughout the HD range. The uniform formability of DFSL samples
irrespective of the change in HD is revealed through uniaxial tensile
test.
• Considering the macrostructure analysis, joint morphology analysis
and load performance analysis, HD range of 3 mm–3.5 mm is fa-
vorable for joining the selected material combination of aluminum
alloys with the DFSL process.
• The failure modes of DFSL samples such as tear off, combined pin
pull-out and bond delamination, and combined pin shear and bond
delamination are different from that of conventional FSSW samples
and their occurrence is random in nature during load performance
analysis tests. The neck of extruded pin and the stir spot cir-
cumference play as critical weak zones initiating the failure in DFSL
samples. Stir spot fracture occurs commonly during uniaxial tensile
test.
• The DFSL samples exhibit superior lap shear fracture load ac-
counting about 42 % higher than conventional FSF samples, 86 %
higher than pinned FSSW samples and 25 % higher than pinless
FSSW samples. The superior lap shear fracture load in DFSL samples
is effected by concurrent occurrence mechanical interlocking as well
as metallurgical bonding. Unlike its counterparts, the pinhole defect
and hook defect are successfully eliminated in DFSL process.
• The ‘TD-HD’ ratio for successful DFSL joint fabrication in aluminum
alloy sheets of 2 mm thickness specifically, AA 5052-H32 and AA
Fig. 14. Cross sectional views of (a) DFSL, (b) conventional FSF and (c) FSSW
samples.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
32

13. 6061-T6 should lie within the range of 3.33–4.66.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
The authors thank the facilities extended by Central Instruments
Facility and by Department of Mechanical Engineering of IIT Guwahati
for conducting the load performance analysis tests and microscopic
analysis. This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
References
[1] Garg A, Bhattacharya A. On lap shear strength of friction stir spot welded AA6061
alloy. J Manuf Process 2017;26:203–15. https://doi.org/10.1016/j.jmapro.2017.
02.019.
[2] Cox CD, Gibson BT, Strauss AM, Cook GE. Energy input during friction stir spot
welding. J Manuf Process 2014;16:479–84. https://doi.org/10.1016/j.jmapro.
2014.05.007.
[3] Li WY, Chu Q, Yang XW, Shen JJ, Vairis A, Wang WB. Microstructure and mor-
phology evolution of probeless friction stir spot welded joints of aluminum alloy. J
Mater Process Technol 2018;252:69–80. https://doi.org/10.1016/j.jmatprotec.
2017.09.003.
[4] Bakavos D, Chen Y, Babout L, Prangnell P. Material interactions in a novel pinless
tool approach to friction stir spot welding thin aluminum sheet. Metall Mater Trans
A Phys Metall Mater Sci 2011;42:1266–82. https://doi.org/10.1007/s11661-010-
0514-x.
[5] Li W, Li J, Zhang Z, Gao D, Wang W, Dong C. Improving mechanical properties of
pinless friction stir spot welded joints by eliminating hook defect. Mater Des
2014;62:247–54. https://doi.org/10.1016/j.matdes.2014.05.028.
[6] Sajed M. Parametric study of two-stage refilled friction stir spot welding. J Manuf
Process 2016;24:307–17. https://doi.org/10.1016/j.jmapro.2016.09.011.
[7] Ding Y, Shen Z, Gerlich AP. Refill friction stir spot welding of dissimilar aluminum
alloy and AlSi coated steel. J Manuf Process 2017;30:353–60.
[8] Cox CD, Gibson BT, Delapp DR, Strauss AM, Cook GE. Technical paper A method for
double-sided friction stir spot welding. J Manuf Process 2014;16:241–7. https://doi.
org/10.1016/j.jmapro.2013.10.006.
[9] Evans WT, Gibson BT, Reynolds JT, Strauss AM, Cook GE. Friction stir Extrusion: a
new process for joining dissimilar materials. Manuf Lett 2015;5:25–8. https://doi.
org/10.1016/j.mfglet.2015.07.001.
[10] Lazarevic S, Miller SF, Li J, Carlson BE. Experimental analysis of friction stir
forming for dissimilar material joining application. J Manuf Process
2013;15:616–24. https://doi.org/10.1016/j.jmapro.2013.05.004.
[11] Nishihara T, Ito A. Measurement of die temperature during friction stir forming.
Weld World 2005;49:22–6.
[12] Lazarevic S, Ogata KA, Miller SF, Kruger GH, Carlson BE. Formation and structure
of work material in the friction stir forming process. J Manuf Sci Eng
2015;137(51018):1–9. https://doi.org/10.1115/1.4030641.
[13] Prakash SJ, Muthukumaran S. Refilling probe hole of friction spot joints by friction
forming. Mater Manuf Process 2011;26:1539–45. https://doi.org/10.1080/
10426914.2011.551959.
[14] Venukumar S, Muthukumaran S, Yalagi SG, Kailas SV. Failure modes and fatigue
behavior of conventional and refilled friction stir spot welds in AA 6061-T6 sheets.
Int J Fatigue 2014;61:93–100. https://doi.org/10.1016/j.ijfatigue.2013.12.009.
[15] Saju TP, Narayanan RG. Friction stir forming of dissimilar grade aluminum alloys:
influence of tool rotational speed on the joint evolution, mechanical performance,
and failure modes. Int J Adv Manuf Technol 2018;95:1377–97.
[16] Saju TP, Narayanan RG. Effect of tool plunge depth on joint formation and
mechanical performance of friction stir forming joints made between AA 5052-H32
and AA 6061-T6 sheet metals. Trans Nonferrous Met Soc China 2018;28:613–28.
https://doi.org/10.1016/S1003-6326(18)64694-1.
[17] Saju TP, Narayanan RG, Roy BS. Effect of pinless tool shoulder diameter on dieless
friction stir extrusion joining of AA 5052-H32 and AA 6061-T6 aluminum alloy
sheets. J Mech Sci Technol 2019;33(18):3981–97. https://doi.org/10.1007/
s12206-019-0136-1.
[18] Saju TP, Narayanan RG. Effect of hole diameter on joint strength and joint forma-
tion in Dieless friction stir form joints between dissimilar aluminum alloy sheets.
Procedia Manuf 2018;26:255–66. https://doi.org/10.1016/j.promfg.2018.07.034.
[19] Standard test methods for tension testing of metallic materials, ASTM E8/E8M-09.
2009.
[20] Standard test method for plastic strain ratio r for sheet metal, ASTM E517. 2010.
[21] Standard test methods for determining average grain size, ASTM E112-13. 2013.
[22] Standard practice for microetching metals and alloys, ASTM E407-07. 2007.
[23] Standard test methods for Vickers hardness and Knoop hardness of metallic mate-
rials, ASTM E92-16. 2016.
[24] Recommended practices for test methods for evaluating the resistance spot welding
behavior of automotive sheet steel materials. 1997. p. 9–97. ANSI/AWS D8.
[25] Paidar M, Khodabandeh A, Najafi H, Rouh-aghdam AS. Effects of the tool rotational
speed and shoulder penetration depth on mechanical properties and failure modes
of friction stir spot welds of aluminum 2024-T3 sheets. J Mech Sci Technol
2014;28:4893–8. https://doi.org/10.1007/s12206-014-1108-0.
[26] Mishra RS, Ma ZY. Friction stir welding and processing. Mater Sci Eng R Rep
2005;50:1–78. https://doi.org/10.1016/j.mser.2005.07.001.
[27] Sato YS, Park SHC, Kokawa H. Microstructural factors governing hardness in fric-
tion-stir welds of solid-solution-Hardened Al alloys. Metall Mater Trans A
2001;32:3033–42.
[28] Kesharwani RK, Panda SK, Pal SK. Experimental investigations on formability of
aluminum tailor friction stir welded blanks in deep drawing process. J Mater Eng
Perform 2015;24(2):1038–49. https://doi.org/10.1007/s11665-014-1361-5.
[29] Huskins EL, Cao B, Ramesh KT. Strengthening mechanisms in an Al–Mg alloy. Mater
Sci Eng A 2010;527:1292–8. https://doi.org/10.1016/j.msea.2009.11.056.
[30] Rana PK, Narayanan RG, Kailas SV. Effect of rotational speed on friction stir spot
welding of AA5052-H32/HDPE/AA5052-H32 sandwich sheets. J Mater Process
Technol 2018;252:511–23. https://doi.org/10.1016/j.jmatprotec.2017.10.016.
[31] Cao JY, Wang M, Kong L, Guo LJ. Hook formation and mechanical properties of
friction spot welding in alloy 6061-T6. J Mater Process Technol 2016;230:254–62.
[32] Evans WT, Cox C, Gibson BT, Strauss AM, Cook GE. Two-sided friction stir riveting
by extrusion: a process for joining dissimilar materials. J Manuf Process
2016;23:115–21. https://doi.org/10.1016/j.jmapro.2016.06.001.
Dr.Tinu P.Saju is an Assistant Professor in TKM College of Engineering, Kollam, Kerala.
He completed Ph.D. from Indian Institute of Technology, Guwahati, India. His area of
research includes novel friction stir based joining techniques such as friction stir spot
welding, friction stir forming and formability of lightweight sheet metals such as alu-
minum alloys. He has introduced dieless friction stir forming and authored five research
papers in international journals and presented papers in reputed conferences such as
NAMRC 2018.
Dr.R. Ganesh Narayanan is a Professor at the Department of Mechanical Engineering,
Indian Institute of Technology (IIT) Guwahati, India. He received his Ph.D. from the IIT
Bombay, India. His research areas of interest include Metal Forming and Joining. He has
contributed many research articles in reputed journals and international conferences. He
has edited few books including ‘Sustainable Material Forming and Joining’ published by
CRC press, ‘Strengthening and Joining by Plastic Deformation’ published by Springer
Singapore, ‘Advances in Material Forming and Joining’ published by Springer India, and
‘Metal Forming Technology and Process Modeling’ published by McGraw Hill Education,
India. He has also edited special issues of journals including ‘Advances in Computational
Methods in Manufacturing’ in the International Journal of Mechatronics and
Manufacturing Systems, and ‘Numerical Simulations in Manufacturing’ in the Journal of
Machining and Forming Technologies. He has organized three international conferences
at IIT Guwahati namely 1st
International Conference on Computational Methods in
Manufacturing (ICCMM) in 2011, the 5th International and 26th All India Manufacturing
Technology, Design and Research (AIMTDR) Conference in 2014, and 2nd
International
Conference on Computational Methods in Manufacturing (ICCMM) in 2019. He has also
organized a GIAN course on ‘Green Material Forming and Joining’ at IIT Guwahati in
2016.
T.P. Saju and R.G. Narayanan Journal of Manufacturing Processes 53 (2020) 21–33
33