See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/266148591

Micro-friction stir welding of titan zinc sheets

ARTICLE in JOURNAL OF MATERIALS PROCESSING TECHNOLOGY FEBRUARY 2015
Impact Factor: 2.24 DOI: 10.1016/j.jmatprotec.2014.08.029

READS

113

3 AUTHORS:
Spyros Papaefthymiou

Constantinos Goulas

National Technical University of Athens

Delft University of Technology

15 PUBLICATIONS 94 CITATIONS

3 PUBLICATIONS 0 CITATIONS

SEE PROFILE

SEE PROFILE

Evangelos Gavalas
RWTH Aachen University
1 PUBLICATION 0 CITATIONS
SEE PROFILE

Available from: Constantinos Goulas

Retrieved on: 27 October 2015

Journal of Materials Processing Technology 216 (2015) 133139

Contents lists available at ScienceDirect

Journal of Materials Processing Technology

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

Micro-friction stir welding of titan zinc sheets

S. Papaefthymiou a, , C. Goulas b,c , E. Gavalas d
a
National Technical University of Athens, School of Mining & Metallurgical Engineering, Laboratory of Physical Metallurgy, 9, Her. Polytechniou str.,
Zografos, GR-15780 Athens, Greece
b
Materials Innovation Institute (M2i), Mekelweg 2, 2628 CD Delft, The Netherlands
c
Delft University of Technology, Department of Materials Science and Engineering, Mekelweg 2, 2628 CD Delft, The Netherlands
d
ELKEME S.A., Piraeus str. 252, GR-17778 Tavros, Athens, Greece

a r t i c l e

i n f o

Article history:
Received 11 March 2014
Received in revised form 4 August 2014
Accepted 29 August 2014
Available online 6 September 2014
Keywords:
Friction stir welding
ZnTiCu
TiZn15

a b s t r a c t
Aim of this research is rst to evaluate the applicability of micro-friction stir welding (FSW) to wrought
zinc alloy sheets and then to improve the structural integrity of such joints. FSW tool design was based
on an algorithm that considers material and process limitations. Joining trials were performed at different
feed rates. It is proven that joining by FSW thin ZnTiCu sheets is possible and it offers extremely ne
microstructures and -phase distribution due to the mechanical fragmentation which is the outcome of
the stirring. The -phase particles were homogenized and precipitated inside deformed zinc grains and
not at the grain boundaries, where they used to be in fusion welds. Electron microscopy showed that
its size was limited to 150 nm, which is in average 13 times smaller than the size of the 2 m that they
get when sheets are TIG welded. Macroscopically, the FSW joint mechanical properties are comparable
with industrially fusion-welded material. The relative low elongation achieved, similar to fusion-welded
sheets, is explained by the occurrence of three main defects: root opening, thinning and kissing bond.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Friction stir welding (FSW) is, a solid state joining technique,
invented by The Welding Institute (TWI), United Kingdom in 1992
(Thomas et al., 1992). Although initially developed for Al-alloys,
soon it was applied successfully to many other metals and materials, especially to metals hard to weld using the given fusion welding
techniques.
In FSW frictional heat is generated between the wear-resistant
welding tool shoulder and pin, and the material of the work-piece.
The frictional heat and surrounding temperature cause the stirred
materials to be softened and mixed avoiding their melting. Thus,
the bonding is a solid state process. However, the grains are transformed and relocated. Material ow patterns under the shoulder
are similar to the forging process, while these around the tool pin
are like an extrusion process as stated by (Mishra and Ma, 2005).
Weld quality is strongly affected by the tool geometry as shown in
detail by Sued et al. (2014) and Groche et al. (2014).
Groche et al. (2014) in their review paper described in detail
methods of joining by forming, among others FSW, which seems

Corresponding author. Tel.: +30 210 772 4710.

E-mail addresses: spapaef@metal.ntua.gr, spyrospapaef@gmail.com
(S. Papaefthymiou), K.Goulas@tudelft.nl (C. Goulas).
http://dx.doi.org/10.1016/j.jmatprotec.2014.08.029
0924-0136/ 2014 Elsevier B.V. All rights reserved.

to be of paramount importance. In general, regarding the process,

during FSW the rotational movement of a tool is overlapped by
the feed speed. The tools are primarily machined from tool steels
such as H13 and subsequently hardened. In research settings, these
tools would typically exhibit little wear and would be considered
practically non-consumable. The side aligning the directions of
both speeds is called the advancing side, whereas the other side
is called the retreating side. Plastic ow and frictional heating create a characteristic stir zone. Friction stir welded microstructures
consist of four zones: the dynamically recrystallized zone (DXZ) or
nugget, the thermo-mechanical affected zone (TMAZ) and the heat
affected zone (HAZ). The DXZ or nugget is located where the pin is
in contact with the base material. This zone is characterized by sufcient values of thermal and strain energy for recrystallization to
occur leading to ne grains of equal size. The thermo-mechanically
affected zone (TMAZ) is located at the side of the nugget, is affected
by high strains and heat, but the energy provided is insufcient for
a complete recrystallization. The heat affected zone (HAZ) consists
of undeformed base metal and is thermally affected by the heat
produced by the stirring of the tool.
As Groche et al. explicitly point out during FSW the material
near the top of the work piece is stirred under by the threads
and is deposited in the weld nugget, if fusion welding, melting
and homogenization would take place. However, larger concentration differences can be found in the welding zone, with diffusion
couples at the edge. The joining mechanism can be described as

134

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

a combination of cold pressure welding, diffusion welding and

mechanical alloying at the same time (Groche et al., 2014).
Defects in FSW might be of any orientation, size, or shape.
Groche et al. describe and categorize very well the defects in
FSW. Like arc welding, the process moves in a linear fashion, usually at a constant rate along the joint line, and therefore has a
similar tendency to produce defects, which propagate for some
length and have their major dimensions parallel to the travel direction. Defects are divided into inclusions, volumetric defects, and
non-volumetric (laminar) defects. Flash is produced by displacement of material from the face (tool-side surface) of friction stir
welded components. Root aws are surface-breaking discontinuities that are present on the material surface which is opposite of
the tool. The mechanical effect of a root defect is dependent primarily on its through-wall height and degree of bonding or width.
Changes in root defect orientation also change mechanical properties. Especially in thin sheets root defect is detrimental for the
joints structural integrity.
According to Kim et al. (2006) the defects which occur during
FSW can be caused by an overheating or insufcient heating of partial areas. Overheating can lead to the formation of an excess ash.
Furthermore, the grain structure can be affected by exceeding a
certain temperature value. Cavities can be formed due to an inadequate material ow, which can be caused by a disadvantageous
rotational speed to travel speed of the tool.
Thus, from all the above it is clear that the FSW process can be
inuenced by the tool design, the travel speed and the rotational
speed, the vertical pressure of the tool, the tilt angle of the tool and
the material characteristics.
Mishra and Ma (2005) underline the difculty of creating high
integrity welds of aerospace aluminium alloys able to withstand
high-strength and fatigue loads, not to crack due to the poor solidication microstructure and porosity in the fusion zone and the
very signicant loss in mechanical properties as compared to the
base material. These aluminium alloys (e.g. highly alloyed 2XXX
and 7XXX series) are generally classied as non-weldable as these
factors make the joining of such alloys by conventional welding
processes unattractive. It is known that some aluminium alloys can
be resistance welded, but the surface preparation is expensive, with
surface oxide being a major problem. Thus, FSW is an alternative
joining process to fusion welding. It is a suitable joining technique
for sheet metals to be welded by butt, lap or llet joints. The rapid
development of FSW in aluminium and its successful commercial
implementation has motivated its application to other non-ferrous
materials (Mg, Cu, Ti, as well as their composites), steel, dissimilar alloys/metals, and even thermoplastics. As many metals are
welded using given fusion techniques, it is not only important to
show the feasibility of FSW, but also to delineate its advantages
over existing fusion techniques. Additionally, FSW gains potential
engineering importance due to the absence of problems associated
with conventional welding such as distortion, grain growth in the
heat affected zone and porosity.
As previously stated a dening characteristic of FSW is that
the joint is created by a non-consumed cylindrical rotating tool,
mechanically traversed through the materials. The pin though
that might be threaded or polished, can be consumed. Especially
threaded pins are usually altered (polished) during processing as
Gibson et al. (2014) describe in their review paper. Lorrain et al.
(2010) state that material ows during FSW are very complex
and not fully understood. Given the fact that most of the studies
describe the use of threaded pins and that the initially threaded
tools may become unthreaded due to tool wear (Gibson et al., 2014;
Lorrain et al., 2010), the material ow becomes even more complex.
Lorrain et al. have shown that the material ow with unthreaded
pin undergoes the same features with classical threaded pins;
material is deposited in the advancing side (AS) in the upper part

of the weld and in the retreating side (RS) in the lower part of the
weld; a rotating layer appears around the tool.
Nishihara and Nagasaka (2004) successfully adapted the FSW
process to materials with thicknesses of 1000 m or less. The
authors named this modied technique micro-friction stir welding
(FSW). FSW could be applied as well in thin walled structures,
electrical, electronic and micro-mechanical assemblies.
Signicant challenges related to tool design and critical procedure parameters are to be overcome when FSW is applied.
Different materials with various tool geometries and process
parameters are to be tested before excellent joints with optimized conditions can exist. Teh et al. (2012) achieved homogeneous
microstructure distribution after FSW of thin aluminium sheets;
even 180 bend was carried out without cracking. This technique
could be appropriate for special welding uses requiring properties
such as high exural strength, ductility and durability, combined
with smooth joint appearance.
Br et al. (2000) indicated the importance of the c/a ratio in
relation to formability. Zinc with its hexagonal close packed crystal structure and with lattice parameter c/a ratio 1.856, which is
higher than the ideal 1.633, has a very limited formability, an
important factor that needs to be taken into consideration when
designing new techniques that involve high degree of deformation. FSW up to now has mostly been applied in easily deformable
metals like aluminium and steel. But when materials with limited
formability (such as metals with hexagonal crystal structure e.g.
magnesium, zinc) are to be joined, literature and industrial experience are limited. Unfortunately, especially for welding of zinc alloys
the available scientic literature, even with current fusion welding
techniques, is very limited. Furthermore, FSW of zinc alloys has
not been industrially applied yet and experimental results are not
published (if available).
Copper (Cu) and titanium (Ti) are the main alloying elements in
the ZnTiCu alloy, from which rolled zinc obtains its desired mechanical properties. Their chemical composition is dened by the
standard EN 988:1997. ZnTiCu alloy contains usually 0.060.20%
Ti. Titanium has very little solubility in Zn (0.02%), and it also reacts
with Zn and forms a hard intermetallic phase with chemical composition TiZn15 called -phase. Copper is in solid solution, forming
-Zn solution.
The presence of the intermetallic -phase is of high importance,
as it renes the cast Zn grains and prevents grain growth during hot
rolling. The -phase precipitates very close to zincs melting point
(m.p. = 421 C; -phase forms 23 C below the m.p.), and forms
stringers along the rolling direction in the -Zn matrix. Their size
and distribution are crucial for the alloys mechanical properties
as they can be responsible for precipitation hardening based on
their size, but also they can act as stress raisers and crack initiation
points, especially when they are enlarged and placed at the grain
boundaries.
Zinc is easily joined with fusion techniques such as TIG/MIG
(Squillace et al., 2004) and/or ultrasonic welding (Patel et al., 2014)
but the structural integrity of such joints is easily deteriorated.
Especially, when very thin zinc sheets are to be welded for construction purposes, such as gutters, industrial roong systems and
other architectural applications, both suppliers and applicants face
severe brittle fracture occurrences even at low loading conditions
leading to cracking within the weld zone.
After fusion welding, -phase redistributes and is totally concentrated at the grain boundaries. The precipitation hardening
effect is thus cancelled and the overall structural integrity of the
welded part can be compromised.
Pantazopoulos and Sampani (2007), in a case study of a zinc
tube failure, found that zinc wrought alloys can suffer from failure
caused by crack formation initiated at such hard intermetallic phase precipitates.

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

Zinc alloys industrially are welded using highly productive

fusion welding techniques. In a previous work of the authors
(Goulas et al., 2013) it was indicated that these welds are susceptible to cracking due to their inhomogeneous microstructure with
coarse columnar grains and segregated intermetallic precipitates
at the grain boundaries.
The application of FSW is of great interest for such applications,
as it will prevent the -phase enlargement that is caused by high
input energy of the fusion technique and will contribute to mechanical fragmentation of the precipitates caused by the stirring tool.
The production of welds by plastic deformation, at temperatures
below the melting temperature of the base materials, can be an
effective way for reducing the formation of brittle intermetallic
phases and consequently to create joints with higher integrity and
lack of cracking in the weld zone.
In our case where brittle, elongated -phase particles can cause
stress concentration at grain boundaries and can lead to early
failure of the zinc sheets, it is important to dene the proper
processing conditions to avoid overloading and grain degradation
due to overheating-melting and/or mechanical stirring. Similar to
zincs hard -phase, Simoncini and Forcellese (2012) pointed out
that the formation of brittle inter-metallic compounds near the
interface between Al and Mg alloys, when these two metals are
fusion welded deteriorate the integrity of the joints. Presence of
such inter-metallic phases in the fused zone is the main reason
causing brittleness and low strength of the joints, as observed by
Ben-Artzy et al. on joints in Mg-Al alloys obtained by means of gas
tungsten arc and electron beam welding techniques, and by Liu et al.
(2013) on laser weld bonded joints in MgAl alloys. Such defect,
that strongly reduces the mechanical properties of the assembly,
also occurs in unconventional welding, such as high energy density laser welding. An emerging and very attractive technology to
overcome such drawback is the friction stir welding (FSW).
The problem of the second phase brittleness is even more
intense and has severe impact in the structural integrity of the joints
when thin and ultra-thin sheets are welded.
Lately spot FSW is applied to successfully joint very thin sheets.
For micro-electronic, medical, micro devices, soldering and other
applications micro-FSW and spot-FSW are effective joining methods as dissimilar metals can be joined together (Wang et al., 2010).
Wang et al. (2010) characterized numerous spot-FSW of 300 m
ultra-thin aluminium 1050 sheets investigating also the failure
causes that were found to be related to the processing parameters as failure initiated near the possible original notch tip in the
stir zone and the failure propagated along the circumference of
the nugget to cause nal fracture. The interface between the heat
affected zone and the thermal-mechanical affected zone was the
softest region, where the cracks of friction stir spot microwelds in
the lap-shear specimens under the loadings initiated and lead to
fracture of the specimens.
Yin et al. (2010) correlated FSW processing parameters to failure
of AZ31 Magnesium alloy after spot welds. Tool shoulder penetration and dwell time are crucial for material mixing and the
formation of hook regions. Different grain sizes contributed to the
nal material-joint integrity. Buffa et al. (2009) showed how partially bonded regions with its curved proles, referred to as hook
regions, create defects (hooking defects).
Many researchers applied FSW in very thin sheets, most of
them joined aluminium, copper and/or dissimilar alloys and
metals with each other. Galvo et al. (2013) successfully joined
1 mm-thick dissimilar copper sheet to 6 mm thick aluminium
plates (aluminiumcopper/different Al alloys copper) proving
the importance of the processing parameters in combination to
the alloy (heat-treatable or non-heat-treatable alloy). Barcellona
et al. (2006) discussed the critical microstructural phenomena that
inuence the mechanical properties of different aluminium alloys

135

when FSW is applied and even post-weld heat treatments are

applied.
Understanding the efforts of researchers and the technology
drive to join thin sheets for applications of a broad scope, it is a
tremendous challenge of FSW technology to be coupled together
with material characteristics that will enable a successful widening of the process application spectrum. In this framework, the aim
of this work is to create joints of thin titanium zinc sheets with
ner structure and improved ductility in order to overcome the
brittleness of current fusion welded sheets for structural and architectural applications. Additionally, as zinc has a similar lattice with
magnesium and thus shows similar limitations in forming, FSW
was one more challenge to overcome including process parameter
determination and successful joining.
For this purpose FSW was applied. Using FSW, thin sheets
could be joined avoiding the columnar microstructure and the
stringer like morphology of intermetallic phases while creating
joints with good integrity and smooth appearance, comparable to
industrially welded sheets.

2. Experimental procedure
The usual FSW tool designs are mostly empirical and difcult
to optimize, as time consuming series of experiments are required.
For this reason, Gratecap et al. (2008) developed a model to dene
the tool geometry and dimensions, based on a selected working
temperature (0.70.8Tm ). In their study, this model was applied
to design the tool to join 5 mm thick aluminium plates. A major
challenge in our work was to apply the same algorithm aiming to
produce sound joints of thin (700 m thick) ZnTiCu sheets. Apart
from the signicant thickness difference in the study of Gratecap
et al., our material, ZnTiCu, has remarkably different physical properties than aluminium.
According to the model of Gratecap et al., the temperature developed during the joining process is a function of tool geometry
and dimensions, physical properties of the material and the joining parameters, such as tool rotation speed and feed rate. In our
case, the working temperature was selected to be 320 C, corresponding to the 0.75Tm . The used experimental setup faced the
following limitations (that were considered in the tool design):
maximum rotational speed and feed rate were 1000 rpm and
318 mm/s respectively. Additionally, the tool geometry was tuned
by the model with aim to minimize material loss and at the same
time to have sufcient material displacement around the pin and
under the shoulder.
The nal geometry and dimensions of the used tool are shown in
Fig. 1. The FSW tool was made of X40CrMoV5-1 tool steel tempered
at 500 C, with the following dimensions: shoulder diameter 5 mm,
shoulder height 5 mm, pin diameter 0.5 mm and pin height 0.7 mm.
The tool was applied to the sheets under a 3 tilt angle.
For a range of different feed rates, two of the above sheets were
butt welded longitudinally to rolling direction. The joining length
was 200 mm on 250 mm total sample length. The tool penetration
depth was manually adjusted to achieve maximum stirring combined with the lowest possible weld ash under the given process
conditions. Trials were executed in a Mill with an adapted FSW
tool. Additionally, according to the practice of Teh et al. (2012) special hold-downs in combination with a low thermal conductivity
back plate were used as shown in Fig. 2. The presence of the back
plate assisted the process by reducing heat extraction from the weld
zone (Teh et al., 2012).
Metallographic, tensile and three point bending samples
were cut from FSW welded ZnTiCu pieces with original size
0.7 mm 250 mm 50 mm. The chemical composition of the used
ZnTiCu alloy is shown in Table 1.

136

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

Fig. 2. The FSW set-up used for the current experiments.

All welds produced were examined visually for their integrity.

The macroscopically inspected and approved welds were then
cross-sectioned for detailed examination by means of light optical microscopy (LOM) using an inverted microscope and scanning
electron microscopy (SEM) equipped with secondary electron and
back scattered electron sensors. Energy dispersive spectroscopy
was carried out for local elemental chemical composition analysis and for the identication of the -phase intermetallic particles.
Finally, from each weld, specimens for tensile and 3-point-bending
tests were also prepared in sets of three (3).
The cross sections were prepared according to the metallographical standards; chemical etching with Nital 2% (2% HNO3 98%
ethanol) was applied to reveal -phase distribution. For the tensile and the 3-point bending tests an electromechanical testing
machine with 30 kN load cell was used.
3. Results
Metallographic evaluation proved successful joining at different
feed rates (travel speed). Optical and electron micrographs showing
the microstructure of ZnTiCu FSW welds are illustrated in Figs. 37.
LOM examination discerned the characteristic FSW wineglass and bowl morphologies as described by Khaled (2005).
Fig. 1. Geometry and dimensions of the used FSW tool.

Table 1
Chemical composition of ZnTiCu sheets used for FSW experiments (wt%).
Zn

Pb

Al

Cu

Sn

Fe

Ti

99.7665

0.0039

0.0093

0.1498

0.0001

0.0009

0.0690

Fig. 4. FSW with 318 mm/min (at 1000 rpm). As-polished micrograph showing joint
area in optical microscope under polarized light. Characteristic bowl morphology
is observed. Intense weld ash is evident affecting the overall performance of the
weld.

Fig. 3. FSW with 127 mm/min (at 1000 rpm). As-polished micrograph showing joint area in optical microscope under polarized light. Characteristic wineglass morphology
is observed. The characteristics of FSW (stir zone, onion rings, weld ash) are indicated.

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

137

Fig. 7. Electron micrograph. FSW performed at 318 mm/min. Details of the submicroscopic dispersoids of the -phase in the stir zone. The -phase is no more in
stringers, but in nely dispersed round particles.

Fig. 5. Electron micrograph (SEM) illustrating the distribution of the -phase (secondary electron detector). FSW was performed at 127 mm/min. (a) Stir zone and
base metal. (b) Details of stir zone showing the sub-microscopic dispersoids that
were formed due to the extended mechanical fragmentation. The difference in the
-phase particles in comparison to (a) is clear.

The characteristic FSW regions i.e. Nugget, stir zone, thermomechanically affected zone and onion ring patterns are identied in
all joints (Figs. 3 and 4). Weld ash was observed as well. The existence of heat affected zones (HAZ) was not able to be distinguished

Fig. 6. Electron micrograph showing -phase stringers in the base metal.

as no microstructural changes were observed near the stir zone;

instead base metal structure was present, neighbouring with the
thermo-mechanically affected zone (Fig. 3).
Base metal microstructure consisted of highly elongated grains
towards rolling direction and -phase stringers, with average grain
size of 1000 nm.
The microstructure of the stir zone was very ne. The -phase
particles were found in the SEM to be signicantly smaller in the
stir zone than in the base metal (Fig. 5). They exhibited a size variation from 50 nm to 250 nm, with an average size at about 170 nm.
For comparison, note that the base metals -phase particle size
was approximately 2000 nm. The -phase distribution in the welds
regions was not in stringers as it was in the as-rolled material
(Fig. 6), but in spherical sub-microscopic particles, evenly dispersed
in the -Zn matrix (Fig. 7). The average grain size was also reduced
at about 400 nm in average within the stir zone (Fig. 7).
Mechanical properties obtained for FSW samples are shown
below. The mechanical properties of industrial TIG welded zinc
sheets, approved by quality inspection, are used for reference/comparison reasons (Figs. 8 and 9). Specimens were tested
parallel to the rolling direction.
Macroscopically, the mechanical properties of FSW and fusion
welded sheets are comparable. TIG welded samples achieved
slightly greater UTS, but with less elongation. From the 3-point
bending test of FSW samples, fusion-welded and as-received samples, stress-exure extension curves were obtained. Specimens

Fig. 8. Tensile test results of FSW-welded ZnTiCu sheets; industrial TIG sample is
used as reference.

138

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

Both these facts explain the reason for the FSW 3-point bending samples not being able to reach the strain level of the reference
TIG samples. Flexure extension though is superior. This is attributed
to the very ne microstructure in the joint area (Rodrigues et al.,
2009).
At the retreating side (Fig. 3), the microstructure inconsistency
between stir zone and base metal is clear. An increase of the tool
rotation speed or shoulder diameter would increase the heat generation. This would lead to smoother and more homogeneous joints.
Heat caused by friction is dependent on the vertical load. Due to
the fact that load varies because of:
1.
2.
3.
4.

inhomogeneities in the base metal;

variation of the sheets cross section;
disturbances in the base metal or tool drives;
dynamic loads generated by acceleration/deceleration of moving
components;
5. vibration transmitted by the environment;
6. self-excited vibration generated by the process or friction, in
order to achieve uniform welds throughout their length, automation of the process with force control system is essential. In this
way, over or under loading would be avoided and, thus, better
control of the process and higher precision could be achieved as
well as better repeatability.

Fig. 9. (a) Bending test results of FSW ZnTiCu sheets. (b) Statistical evaluation of
the results and scatter plots.

were tested transverse to rolling direction. Fusion welds exhibited

higher exural strength, but they reached lower extension (up to
10 mm). On the other hand, FSW joints showed more consistent
behaviour even at extensions of about 20 mm.
4. Discussion
Grain boundary embrittlement of fusion welded TiZnCu sheets
due to hard -phase precipitates compromises the structural
integrity of very thin welded sheets. As -phase particles precipitate at temperature just below the solidus, there is no heat
treatment to bring those into solution and then to allow their reprecipitation in a ne, homogeneous dispersed manner. Friction
stir welding, through the high deformation in the joining region,
offers a ne microstructure with evenly distributed intermetallics,
which could give an effective solution to the embrittlement issue
of very thin TiZnCu sheet joints.
In order to achieve joints of high integrity, all possible aws
related to the FSW process parameters must be rst determined
and then the occurrence of aws must be limited if not directly
eliminated. Despite the successful FSW joints, three main defects
related to process parameters were evident: root opening, thinning
and kissing bond. Occurrence of kissing bond (Fig. 3) indicates low
heat input, usually barely enough to create a joint. The kissing bond
phenomenon was observed and analysed by Sato et al. (2005). Poor
performance in the 3-point bending test of the 318 mm/min sample
in comparison to the 127 mm/min can be attributed to the following
issues:
1. Thinning by 15% further weakened the materials resistance, as
it reduced signicantly the area to which the tension is applied.
2. The kissing bond weakened the material as it acted like a precrack in the 45 direction to the loading.

Workpiece alignment could also benet signicantly by process

automation. Manual workpiece alignment can lead to asymmetric
heat transfer, material ow and properties of both sides of the weld
as Nandan et al. (2007) well indicate. Of course in terms of materials
that are not easily formable and/or have low heat transfer coefcient, as in the case investigated, such characteristics can play a
signicant role in successful joining and will be determinative to
the process conditions.
Process parameters, indeed, were found to be directly affected
by the weld morphology as presented by Cavaliere et al. (2006).
Specically, tool penetration depth as well as travel speed affected
the joint prole. A shallower penetration combined with a lower
travel speed produced a wineglass-like joint, while a deeper penetration with higher travel speed produced a bowl-like joint. Weld
ash was evident leading to thinning of the joint area by average
15% affecting the overall quality of the weld. The integrity of the
weld was signicantly affected by the occasional root opening. This
was a common defect that seems to be not travel speed dependent
as it appeared to both low and high speeds. It can be attributed to
the variant distance between the bottom of the sheet and the pin
and the manually controlled process parameters.
As Groche et al. (2014) explicitly point out lack of penetration (LOP), lack of consolidation (LOC), and kissing bond type
root defects are caused by excessive penetration ligament (distance between bottom of probe and root-side surface of material),
inadequate tool-joint alignment (missing the joint), or inadequate
disruption (poor stirring) of the abutted parent material surfaces
near the root respectively. Both LOP and kissing bonds imply that
there is some portion of the abutted joint surfaces which remain
unbonded or inadequately disrupted. The kissing bond is generally
less severe and can involve signicant distortion of the jointline.
According to the literature, depending on location and extent,
LOP defects can have signicant effects on mechanical properties.
A distinct difference in the microstructure between stir zone
and base metal was revealed in the SEM. Due to extended stirring,
the -phase particles were mechanically fragmented, thus becoming signicantly smaller in the stir zone than in the base metal
(Fig. 5). The intensity of the stirring affected also the average grain
size, which is between 170 nm (feed rate 318 mm/min) and 400 nm
(128 mm/min) in the stir zone (Fig. 7) compared to 2000 nm in the

S. Papaefthymiou et al. / Journal of Materials Processing Technology 216 (2015) 133139

base metal. This is of high importance as the -phase is mostly

related to embrittlement issues found in fusion welds. Fragmentation of the -phase particles is caused mechanically, as a result of
the high deformation applied in the stir zone. This contributes also
to weld strength increase as proven through tensile and bending
tests.
The FSW samples exhibited 15% lower strength than the reference TIG samples. This is mainly caused due to the 1015% thinning
and the occasional existence of root opening. Root opening represents a notch perpendicular to the loading direction. However, the
ner -phase distribution in the stir zone of the FSW contributed
to increase of total elongation.
The solid-state joining prevented intermetallic phases from
coarsening. FSW samples managed to achieve bending angles above
90 without breaking (in contrast with the fusion welded samples
that broke at much smaller angles). FSW joints showed greater
formability potential than fusion welded pieces.
The metallographic analysis revealed onion ring patterns within
the Nugget zone. The onion ring pattern is a banded microstructure
usually caused by threaded pins.
Most probably the Nugget zones ring pattern is an effect of
periodic differences in the crystallographic orientation of grains,
or varying relative orientation in adjacent grains as described by
Larsson et al. (2000). The used pin though was not threaded, so it is
likely that the pin had poor surface quality, wear, or possible ZnTiCu
pickup. This could be avoided by ner tool nishing as well as by
higher rotational speeds.
5. Conclusions
ZnTiCu sheets 700 m thick were successfully joined by applying
micro-friction stir welding (FSW) with a tool specially designed
for this purpose.
The FSW parameters of 1000 rpm and 318 mm/min resulted in
the best joint integrity.
The FSW joints were comparable to the sheets welded by processes of present state-of-the-art industrial practice TIG welds.
The -phase (TiZn15 ) particles are fragmented through the FSW
process, nely dispersed and their size is reduced in the stir zone.
The grain structure of the FSW welds consists of small, equiaxed
grains, which differ signicantly from the large columnar grains
of the fusion welds.
The FSW resulted in better formability.
Acknowledgments
The authors express their gratitude to Prof. Dr. D. Manolakos
for providing his lab and to Mr. N. Melissas and Mr. G. Michas for
carrying out numerous FSW trials. Special thanks to Mrs. V. Panteleakou and Mr. M. Mariou, vital members of the team. Thanks to
Mr. A. Vazdirvanidis and Mr. A. Toulfatzis for their support in LOM
and mechanical testing respectively.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.jmatprotec.
2014.08.029.

139

References
Barcellona, A., Buffa, G., Fratini, L., Palmeri, D., 2006. On microstructural phenomena
occurring in friction stir welding of aluminium alloys. J. Mater. Process. Technol.
177, 340343.
Br, A., Chen, J., Hairie, A., Nouet, G., Paumier, E., 2000. Determination of the high c/a
ratio of hexagonal metals with a semi-empirical tight-binding method. Comput.
Mater. Sci. 17, 249254.
Buffa, G., Campanile, G., Fratini, L., Prisco, A., 2009. Friction stir welding of lap joints:
inuence of process parameters on the metallurgical and mechanical properties.
Mater. Sci. Eng. A 519, 1926.
Cavaliere, P., Campanile, G., Panella, F., Squillace, A., 2006. Effect of welding parameters on mechanical and microstructural properties of AA6056 joints produced
by friction stir welding. J. Mater. Process. Technol. 180, 263270.
Galvo, I., Leal, R.M., Rodrigues, D.M., Loureiro, A., 2013. Inuence of tool shoulder
geometry on properties of friction stir welds in thin copper sheets. J. Mater.
Process. Technol. 213, 129135.
Gibson, B.T., Lammlein, D.H., Prater, T.J., Longhurst, W.R., Cox, C.D., Ballun, M.C.,
Dharmaraj, K.J., Cook, G.E., Strauss, A.M., 2014. Friction stir welding: process,
automation, and control. J. Manuf. Process. 16, 5673.
Goulas, C., Papaefthymiou, S.A., Pantazopoulos, G., Panagopoulos, C., 2013. Welding
of thin titanium zinc sheets. In: 5th Hellenic Conference on Metallic Materials,
Volos, Greece.
Gratecap, F., Racineux, G., Marya, S., 2008. A simple methodology to dene conical
tool geometry and welding parameters in friction stir welding. Int. J. Mater.
Form. 1, 143158.
Groche, P., Wohletz, S., Brenneis, M., Pabst, C., Resch, F., 2014. Joining by forminga
review on joint mechanisms, applications and future trends. J. Mater. Process.
Technol. 214, 19721994.
Khaled, T., 2005. An Outsider Look at Friction Stir Welding. Federal Aviation Administration.
Kim, Y.G., Fujii, H., Tsumura, T., Komazaki, T., Nakata, K., 2006. Three defect types
in friction stir welding of aluminum die casting alloy. Mater. Sci. Eng. A 415,
250254.
Larsson, H., Karlsson, L., Svensson, L.E., 2000. Friction stir welding of aluminium alloy
5083 and aluminium alloy 6082. In: Svetsaren, p. 54.
Liu, H.J., Hou, J.C., Guo, H., 2013. Effect of welding speed on microstructure and
mechanical properties of self-reacting friction stir welded 6061-T6 aluminum
alloy. Mater. Des. 50, 872878.
Lorrain, O., Favier, V., Zahrouni, H., Lawrjaniec, D., 2010. Understanding the material ow path of friction stir welding process using unthreaded tools. J. Mater.
Process. Technol. 210, 603609.
Mishra, R.S., Ma, Z.Y., 2005. Friction stir welding and processing. Mater. Sci. Eng. R:
Rep. 50, 178.
Nandan, R., Roy, G.G., Lienert, T.J., Debroy, T., 2007. Three-dimensional heat and
material ow during friction stir welding of mild steel. Acta Mater. 55, 883895.
Nishihara, T., Nagasaka, Y., 2004. Development of micro-FSW. In: 5th International
Friction Stir Selding Symposium, Metz.
Pantazopoulos, G., Sampani, A., 2007. Analysis of a weld failure of a rolled Zn-alloy
strip a case study. Eng. Fail. Anal. 14, 642651.
Patel, V.K., Bhole, S.D., Chen, D.L., 2014. Characterization of ultrasonic spot welded
joints of Mg-to-galvanized and ungalvanized steel with a tin interlayer. J. Mater.
Process. Technol. 214, 811817.
Rodrigues, D.M., Loureiro, A., Leitao, C., Leal, R.M., Chaparro, B.M., Vilaca, P.,
2009. Inuence of friction stir welding parameters on the microstructural and
mechanical properties of AA 6016-T4 thin welds. Mater. Des. 30, 19131921.
Sato, Y.S., Takauchi, H., Park, S.H.C., Kokawa, H., 2005. Characteristics of the kissingbond in friction stir welded Al alloy 1050. Mater. Sci. Eng. A 405, 333338.
Simoncini, M., Forcellese, A., 2012. Effect of the welding parameters and tool conguration on micro- and macro-mechanical properties of similar and dissimilar
FSWed joints in AA5754 and AZ31 thin sheets. Mater. Des. 41, 5060.
Squillace, A., De Fenzo, A., Giorleo, G., Bellucci, F., 2004. A comparison between
FSW and TIG welding techniques: modications of microstructure and pitting
corrosion resistance in AA 2024-T3 butt joints. J. Mater. Process. Technol. 152,
97105.
Sued, M.K., Pons, D., Lavroff, J., Wong, E.H., 2014. Design features for bobbin friction
stir welding tools: development of a conceptual model linking the underlying
physics to the production process. Mater. Des. 54, 632643.
Teh, N.J., Goddin, H., Whitaker, A., 2012. In: Institute, T.W. (Ed.), Developments in
micro applications of friction stir welding.
Thomas, W.M., Nicholas, E.D., Needham, J.D., Murch, M.G., Templesmith, P., Dawes,
C.S., 1992. Friction Stir Butt Welding.
Wang, D.-A., Chao, C.-W., Lin, P.-C., Uan, J.-Y., 2010. Mechanical characterization of
friction stir spot microwelds. J. Mater. Process. Technol. 210, 19421948.
Yin, Y.H., Sun, N., North, T.H., Hu, S.S., 2010. Hook formation and mechanical properties in AZ31 friction stir spot welds. J. Mater. Process. Technol. 210, 20622070.