Engineering Science and Technology, an International Journal 21 (2018) 245–254

Contents lists available at ScienceDirect

Engineering Science and Technology,

an International Journal
journal homepage: www.elsevier.com/locate/jestch

Full Length Article

Friction stir welding of polypropylene sheet

Santosh K. Sahu a, Debasish Mishra a, Raju P. Mahto b, Vyas M. Sharma b, Surjya K. Pal b,⇑,
Kamal Pal a, Susanta Banerjee c, Padmanav Dash d
a
Department of Production Engineering, Veer Surendra Sai University of Technology, Burla, Odisha 768016, India
b
Department of Mechanical Engineering, Indian Institute of Technology Kharagpur, West Bengal 721302, India
c
Material Science Centre, Indian Institute of Technology Kharagpur, West Bengal 721302, India
d
Department of Mechanical Engineering, Veer Surendra Sai University of Technology, Burla, Odisha 768016, India

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

Article history: Friction stir welding (FSW) is a growing technology in the manufacturing industries because of its numer-
Received 9 June 2017 ous contributions involving the welding of light-weight materials such as aluminium and magnesium.
Revised 4 December 2017 The process has been recently implemented for joining thermoplastics. The present paper has been ded-
Accepted 1 March 2018
icated to investigate the application of the FSW process to join the thermoplastics. The article has been
Available online 17 March 2018
divided into two parts; the first part outlines the importance of the thermoplastics through their wide
applications and the principle of the FSW process has been described. A detailed literature study regard-
Keywords:
ing the effect of process parameters, pin profile, defects and weld zones during the FSW of thermoplastics
Friction stir welding
Light-weight materials
has been conducted. The second part has been focused on the experimental investigation carried out to
Thermoplastics analyse the effect of tool pin geometry on FSW of 6 mm thick polypropylene sheets. The variation of the
Process parameters Z-load with time has been utilised to understand the occurrence of various stages of the welding. Also, a
Tensile strength comparison between the thermoplastic and metal concerning the variation of Z-load with time has been
Tool geometry performed to emphasis the process difference between them. Further, the effect of tool rotational speed
and welding speed on the average force during FSW of thermoplastics has been shown. In order to inves-
tigate the weld quality, three different tools having cylindrical, square and conical pin geometry have
been employed. Three different tool rotational speeds and traverse speeds were selected for carrying
out the experiments. The effect of the shoulder diameter has also been studied. The tensile strength of
the welds has been determined and correlated with pin profile. Results obtained proved the potential
of square pin in achieving high quality welds. The process parameters have been found to have a signif-
icant effect on the tensile strength of the welds.
Ó 2018 Karabuk University. Publishing services by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
There exists a mismatch between the demand for natural
resources such as fuel and its supply which has created a pressure
on the manufacturers to involve efficient production processes that
would lead to resource conservation. As such various light-weight
materials and improved processes have been discovered. Alloys of
aluminium, magnesium, copper etc. are being widely used nowa-
days in automobiles, aerospace, shipbuilding and railway indus-
tries. Thermoplastics being light-weight with high specific
strength, design flexibility and low manufacturing cost have drawn
the attention of manufacturers to explore these properties in lead-
ing industries [1]. Some of the applications of thermoplastics are
⇑ Corresponding author.
E-mail address: skpal@mech.iitkgp.ernet.in (S.K. Pal).
Peer review under responsibility of Karabuk University.
the clutch pedal and centre console manufactured from polyamide
6 (PA 6) which has significantly reduced the weight of the compo-
nent [2], use of various fibre reinforced plastics (FRP) as structural
components and clips and brackets of aerospace structures made
of thermoplastics which aid to weight savings [3]. Adhesive bond-
ing, mechanical fastening and the fusion bonding methods are the
techniques which have been employed for joining of polymeric
materials [4]. Adhesive bonding method has been employed in
many industrial applications but has adverse environmental effects
because of the chemicals used; and the process needs surface
preparation. Moreover, the joints can neither be examined qualita-
tively nor the forecasting of failure. Mechanical fastening is another
alternative but it leads to further weight addition because of the
rivets, clamps etc. being used and moreover residual stresses are
developed [5]. Thus, the preferred method to join is welding.
Traditional welding methods degrade the material and thus we
need a method which would not only provide better efficiency with

https://doi.org/10.1016/j.jestch.2018.03.002
2215-0986/Ó 2018 Karabuk University. Publishing services by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
246 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254
reduced cost but also be environmentally friendly. One of the
solutions to this issue can be FSW. The major significance of FSW
is that it does not involve melting of the base material.
This innovative joining process takes place in solid state with-
out generating any toxic fumes and requires no filler material. It
was patented by The Welding Institute (TWI), UK in 1991 for join-
ing aluminium alloys [6]. It utilises a rotating tool with a specially
designed pin which plunges into the substrate to be welded main-
taining zero root gap and travels along the joint line. The schematic
image of the process has been shown in Fig. 1. The ease with which
the tool can travel over the joint line is controlled by the parameter
termed tilt angle. The tool is normally tilted towards the trailing
edge and as such the front edge of the tool is a little above the sur-
face of the work-piece. It helps to ensure the contact between the
shoulder and the work-piece and also ensures better flow of mate-
rial across the edges of the tool. Literature shows that the butt joint
and lap joint configurations are the need of the hour. This process
can be employed for other geometries as well. Continuous research
on this technology has been carried out which has contributed sig-
nificantly to the manufacturing industries. Since a few years, the
process has been employed to weld thermoplastic materials [7].
TWI reported the successful welding of Polypropylene plates
(PP) with a reciprocating tool (‘‘Reciprocating friction stir welding
joins thermoplastics.‘‘, 2000) [8]. The scope of FSW against other
joining methods for plastics was analysed and studied (Strand,
2003) [9]. The author also presented the comparative data among
various plastic joining methods in terms of energy requirement
and the cost incurred which further leverages FSW. Various tools
for FSW of thermoplastics have been developed and the degrada-
tion of top surface of the material caused by the traditional FSW
tools has been reported [10]. Nevertheless, a wide range of works
have been carried out using traditional tools. These works serve
as the foundation and have inspired other researchers to con-
tribute more to this field [7,11–29]. Table 1 approximately shows
the various thermoplastic materials in percentage which have been
the interest of researchers till date.
It has been observed that as the tool plunges into the plates to
be welded, the soft material gets deformed. The shoulder rotation
over the top helps to throw out this softened material. A new vari-
ant of the traditional FSW tool called the ‘‘heated shoe” was used to
weld the PP plates [11]. It consists of a stationary shoulder having a
rectangular cross sectional area and is referred as the shoe. A heat-
ing coil is placed inside the shoe which was held responsible for
providing the necessary heat to the base metals. The long shoe
exerted force on a large area and restricted the material to come
out from the joint line. Stationary shoulder was employed to weld
ABS plates and superficial finished joints without any external
Fig. 1. Schematic view of friction stir butt welding process.
Table 1
Thermoplastic materials used in research.
Sl no Thermoplastic Use of the thermoplastic in FSW (%)
1 HDPE 36
2 ABS 24
3 PP 20
4 PC 8
5 PMMA 4
6 UHMW-PE 4
7 NYLON 6 4
(PP- Polypropylene, PC- polycarbonate, PMMA- polymethyl methacrylate, HDPE-
high density polyethylene, UHMW-PE- ultra high molecular weight polyethylene,
ABS- acrylonitrile butadiene styrene)
heating were obtained [21]. A tool with two stationery shoulders
to weld ABS plates with a convex pin and a simple pin was devel-
oped [23]. The use of stationary shoulder aids in achieving good
welds in case of polymeric materials because the material present
at the top surface are not destroyed while the softened materials
inside the abutting edges solidify firmly without any loss of
strength. Further, the loss of the material from the work-piece as
in case of the traditional tool was not only reduced with the use
of two stationary shoulders but also the root defects arising during
the welding process were eliminated. In case of traditional FSW
tools, threaded pin profile has proven to be better for welding of
polymeric materials. Two tools with left hand threaded pin to weld
Nylon-6 were developed which were fed in clockwise and anti-
clockwise directions separately [24]. They observed that the
threads in the pin are creating a path for the soft material to flow
along the edges of the tool. Thus, feeding the tool in clockwise
direction resulted in loss of material. However, good welds were
created in case of anti-clockwise direction. A study on welding of
PP composites in a lap configuration using four different tools
was performed [18]. They reported that the conical pin is one of
the poorest designs in welding of the polymeric materials because
its profile failed in achieving the homogenization of the material.
Thermoplastic materials have properties very different from
that of the metals. One of the major features being the low thermal
conductivity, because of which the heat developed in the advanc-
ing side and retreating side is not uniform which leads to defects.
Tool rotational speed, traverse speed and tilt angle are the effects
that have been studied. The welding parameters are found to have
significant effects on the mechanical properties of the welds. Lower
rotational speed created cavities as it was unable to generate the
required heat to deform the material while higher rotational speed
was constantly degrading the top surface of the work-piece. In the
case, where traverse speed of the tool was at lower values, it pre-
sented good appearance, rough surfaces otherwise. The tool tilt
angle effect has also been studied by several researchers
[12,16,25]. They observed that a tilt in the tool is assisting in
obtaining higher strength. This is because the tilting helps in
plunging and better mixing of the material. But a higher tilt in
the tool will result in plunging of tool more inside and cannot
ensure mixing. The UHMW-PE sheets were pre-heated prior to
welding and this technique helped to achieve adequate heat and
also reduced the need of high tool rotational speeds [14]. More-
over, it is suggested that it is wise to remove the samples after they
reach room temperature so that they do not distort thereby ensur-
ing better results [25]. The FSW process succeeded in eliminating
the defects occurring in a traditional welding method but has given
rise to other defects. The occurrence of root defect was found while
welding PP plates with a heated shoe with conical pin [11]. A
longer pin will result in melting of the material while a shorter
pin length will result in root defects. The peeling defect was also
observed while welding PP plates with a traditional FSW tool
[26]. This defect was observed by other researchers as well while
 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254
employing traditional tool because of the shoulder rotation which
throws out the softened material. Flash was observed while
welding PP composites with 20% carbon fibre in a lap configuration
[18]. Flash is a typical FSW defect which is a result of the excessive
heating. The welds produced with high tool rotational speed and
low tool traverse speed generally experience high level of material
softening. As a result of this high heat, the softened material flows
out from the shoulder zone and does not goes into the weld zone,
rather it gets accumulated in the retreating side. This accumulated
material is termed as the flash. Tunneling defect was observed
while welding PP composites filled with 30% glass fibres [16]. Tun-
nel defect also known as wormhole defect is a result of improper
material mixing and the loss of material takes place in the form
of flash. It occurs because of high or low tool traverse speed. In
FSW, the tool extrudes the material from the advancing side and
makes it flow through the retreating side across its edges and
forges it back in the advancing side again. When the heat input
to the weld is high, the materials to be welded are highly plasti-
cized and the loss of material occurs as flash as discussed above.
With increase in the tool traverse speed, the heat input to the weld
decreases. As such, the materials to be welded are not sufficiently
plasticized which makes them unable to flow and hence the result
in improper material mixing. It is mainly found in the advancing
side of the weld where the pin region and shoulder flow regions
intersect. The defect occurs in materials with low thermal
conductivity.
The flow regions during FSW of thermoplastics can be divided
into four types namely; advancing interface, retreating interface,
central region and bottom disturbance [11]. The flow lines in the
advancing side and the retreating regions seem to be parallel to
the welding direction. The central region refers to the weld nugget
which actually shows the concentric rings of material flow. The
bottom disturbance region formed due to rotation of pin generates
flow lines having no particular structure probably because of the
high turbulence in this region. Due to the difference in the heat
developed between advancing side and retreating side, lot of
cracks and voids are observed. The stir zone looks separated from
the retreating side while welding with traditional FSW tools. The
flow pattern resembles the onion ring structures just like the con-
centric rings observed during welding with a stationery shoulder
[11,21].
The present study investigated various experimental works
which have been done regarding FSW of thermoplastic. Joining of
thermoplastic is a gruelling task and requires high precision and
accuracy with respect to parameters opted in the joining technique.
The achieved joints’ ultimate tensile strength in some previous
works is although very close to the ultimate tensile strength of
the base material, the commercialisation of FSW process for ther-
moplastic materials requires substantial research. The selection of
tool is an important feature since the blending of material is neces-
sary. A proper tool design with ability of mixing the softened ther-
moplastic material at wide range of parameters will improve the
process capability and efficiency. It can be inferred from the litera-
ture survey and from Table 1 that HDPE has been the most explored
material for joining with FSW till date but the present work chooses
PP as the testing material. Irrespective of the fact that PP and HDPE
have similar physical properties, PP is more advantageous than
HDPE because it can withstand high temperatures due to its higher
Table 2
Physical properties of PP.
Density (gm/cc) Ultimate tensile strength (MPa) Co-efficient of linear expansion (per °C)10
0.91 33 160
247
melting point. It is stiffer and more rigid than HDPE. Also, PP is less
brittle than HDPE.
The purpose of this research is to study the variation of axial
force with time during FSW of PP and also the effect of welding
parameters on average force. The effect of tool geometry on the
quality of welds has been studied. Also, a comparative study of
variation of axial force with time in case of metals and thermoplas-
tics has been carried out.
2. Experimental procedure
The base material is a commercial grade PP sheet of 6 mm
thickness. PP is an all-purpose polymer synthesised from Propy-
lene monomer serving the dual purpose of plastic and fibre. It
has a wide range of applications as it doesn’t melt below 160 °C.
It is a linear hydrocarbon with a methyl group attached to every
other carbon atom in the backbone chain. They provide excellent
resistance to stress and cracking. It is non-toxic and moisture resis-
tant [30,31]. The physical properties of the base material as pro-
vided by the vendor have been listed in Table 2. For the
investigation, samples of size 120 mm  100 mm  6 mm were
prepared. Square butt welding was done with a 2 Tonne; 3-axis
CNC controlled FSW Machine (ETA, Bangalore).
The tools were machined from H13 tool steel. Three different
profiled pin tools were employed to carry out the welding process
as shown in Fig. 2 namely conical pin (a), square pin (b) and cylin-
drical pin (c). The dimensions such as shoulder diameter, pin diam-
eter and pin length were kept same for all the tools and has been
shown in the figure.
During trial experiments, it was observed that with rotational
speed lower than 500 rpm, insufficient heat was generated which
resulted in poor welds and with rotational speed higher than
1000 rpm, material degraded and resulted in low weld strength.
Also, if the welding speed was lower than 5 mm/min, high fric-
tional heat was generated which led to defects. Moreover, it
increases the welding time and lowers the process efficiency.
When the welding speed was kept higher than 25 mm/min it
resulted in lot of voids and poor welds due to insufficient heat. A
constant tilt angle of 10 towards the trailing edge is used through-
out the process. The arrangement in the machine used for tilting
the tool has been depicted in Fig. 3(a) and the cross sectional view
of the tool with tilt inside the work-pieces has been shown in Fig. 3
(b). The selected parameters for the proposed experiment have
been listed in Table 3. The sample after experiment was left for
2 min to cool and then was removed from the fixture. The fixture
arrangement used for the present experiment has been shown in
Fig. 4.
To study the effect of shoulder diameter, two cylindrical tools
were chosen. Out of the two, one was made with 16 mm diameter
and other tool with 12 mm diameter. Both of them have same pin
diameter of 6 mm and pin length of 5.6 mm. Same set of parame-
ters were selected for the study. For determining the strength of
the welds, tensile test specimens were cut from the welded sam-
ples as per ASTM: D638 standard (ASTM D638 Standard Test Method
for Tensile Testing of Plastics) [32] and were tested in a Universal
Tensile Test Machine (Instron, 1344). The dimensions of the tensile
specimen have been shown in Fig. 5.
6
Melting temperature (°C) Thermal conductivity (W/mK)
160 0.22
248 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254

Fig. 2. View of the tools (a) Conical pin (b) Square pin (c) Cylindrical pin.

Fig. 3. (a) The tool tilting arrangement available in the machine, (b) cross-sectional view of the sample.

3. Results and discussion
3.1. Variation of Z-load with time in FSW process
Fig. 6 shows the Z-load versus time plot during the entire weld-
ing process of a particular sample (tool rotational speed 750 rpm,
traverse speed 25 mm/min). The plot has been divided into five
stages. The start of the machine resembles stage 1. The tool catches
the rotational speed and approaches the substrate surface. As the
pin makes a contact with the surface, certain amount of force is
provided by the tool as it is plunging into the plates. This is
depicted as stage 2. In general, the positive load is known to be ten-
 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254 249

Table 3
Process parameters.

Pin profile 1: Cylindrical pin

Process parameters
Experiment No. x (rpm) v (mm/min)
1 500 5
2 500 15
3 500 25
4 750 5
5 750 15
6 750 25
7 1000 5
8 1000 15
9 1000 25
Pin profile 2: Square pin
Process parameters
Experiment No. x (rpm) v (mm/min)
10 500 5
11 500 15
12 500 25
13 750 5 Fig. 5. Dimension of the tensile specimen.
14 750 15
15 750 25
16 1000 5
17 1000 15
18 1000 25
Pin profile 3: Conical pin
Process parameters
Experiment No. x (rpm) v (mm/min)
19 500 5
20 500 15
21 500 25
22 750 5
23 750 15
24 750 25
25 1000 5
26 1000 15
27 1000 25

(x- tool rotational speed, v- tool traverse speed)

Fig. 6. Variation of Z-load with time in FSW.

from which the force has become almost constant is nearly 5 s.

Fig. 4. Fixture arrangement used for the experiment.
sile in nature but here it should be noted that the nature of the pos-
itive load is compressive in nature. The force value started to rise as
the surface material was deformed by the tool. As a result of defor-
mation, the material from the work-piece surface was removed in
the form of fine particles. From the Fig. 6, a slight decrease in the
force can be observed at this point. This was because of the soften-
ing of the material. With further insertion of pin and contact of
shoulder with the plates, the force rises up. The peak force value
obtained during the whole process was about 1467 N. After this,
a sudden decrease in the plot can be seen which is because of
dwelling, as the material below the tool surface was deformed.
The time difference between the peak force value and the stage
This time is devoted for dwelling which is a fixed characteristic
of the machine used. Thermoplastic materials are soft in nature
with low binding energy. Also, the intermolecular force of
attraction between the particles of the material is low. Thus, they
highly differ from metals. Therefore, force value decreased con-
stantly which is depicted as stage 3. Once the welding started,
the force value became approximately constant which is depicted
as stage 4. During this stage, the little variation in force was due
to the shoulder rotation over the surface. At last, the tool retracted
out of the welded sample and the process was completed which is
stage 5 where the force dropped down.
Fig. 7 shows the difference in variation of Z-load obtained in
case of metal and thermoplastic. For this, two 6 mm thick alu-
minium sheets were butt welded. From the two plots it can be
observed that the load in case of metals is much higher than that
of the thermoplastics. Metals and thermoplastics differ signifi-
cantly with respect to their various properties such as thermal con-
ductivity, strength, hardness, melting point etc. The thermoplastic
materials have weak Van Der Waals bonds as compared to the
primary bonds in metals. In addition to this, as the pin plunged into
the material, the deformation initiated, as a result a drop in the
force value was observed. This in the case of metals is not true,
because it can be seen that the pin plunging part seems to be
250 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254

Fig. 7. Comparison of Z-load variation with time between metal and thermoplastic.

constant. The amount of energy required to initiate the deforma-

tion is reduced in thermoplastic materials. It is evident that with
increase in heat generation in the process, the force will reduce
eventually. The metals have high values of thermal conductivity
which makes uniform heat distribution. But with thermoplastics,
the thermal conductivity is almost negligible which makes the heat
locally concentrated in the joint line. Thus, FSW of metals revolves
around the factors which will contribute to heat generation mak-
ing the material to flow around the tool edges during welding.
But thermoplastics need careful handling so as to reduce the loss
of material from its top surface due to high heat generation.

3.2. Effect of welding parameters on force

During the experiment, force values were recorded and anal-

ysed to study the effect of welding parameters. The cylindrical
pin tool was used for the analysis. Fig. 8(a) depicts the effect of tool
rotational speed on average force keeping the traverse speed con-
stant. It can be observed that the force decreased with an increase
in tool rotational speed. As the tool rotation increases, heat gener-
ation increases and thus the average force imparted by the tool to
the substrate decreases. Fig. 8(b) shows the effect of traverse speed
on average force keeping the tool rotational speed constant. The
force value increases with an increase in traverse speed because
the heat input to the substrate is reduced. As the welding speed
is increased, more power is required as compared to slower weld-
ing speed which eventually increases the force during the process.
From the figure it can be inferred that the variation of average force
due to increase in welding speed was found to be significant at low
tool rotational speed (500 rpm). Similarly, the variation can be
seen at higher traverse speed (25 mm/min) with low tool rota-
tional speed.
3.3. Appearance of the welded sample
The experiment was successfully carried out on PP sheets with
the selected set of parameters. A typical weld appearance of a
welded sample is shown in Fig. 9 (rotational speed 750 rpm, tra-
verse speed 15 mm/min and square pin).
All the samples welded using conical pin employing different
set of parameters showed negligible material mixing. The stir zone
showed a complete void. The welds were so weak that they could
have been broken by hand. Thus, conical tool pin was unable to
blend the thermoplastic material irrespective of the welding
parameters employed. This is because of the simple design of the
pin which has failed to blend the softened material during the
Fig. 8. (a) Effect of rotational speed on average force, (b) Effect of welding speed on
average force.
welding process. Moreover, the rotating shoulder over the surface
peels off the material. A view of a sample welded using conical pin
is shown in Fig. 10. Similar results have been reported while weld-
ing carbon fibre reinforced PP composite in a lap configuration
using a traditional tool with conical pin [18]. But these results
are predominant to traditional tools which employ the rotation
of shoulder. The use of conical pin with stationary shoulder pro-
duced welds with good finish. Thus, stationary shoulder is more
efficient for welding thermoplastics with a conical pin than a tradi-
tional tool in FSW [21]. Hence, no further experiments were carried
out with conical pin.
The welding scenario differs from thermoplastics to metals
because of the differences in their physical characteristics. Since,
thermoplastics are poor conductors of heat; it is difficult to heat
them uniformly. As a consequence of their poor conductivity, with
incorrect heat application, the surface may get charred. However,
the material immediately below the surface remains unchanged.
Square pin successfully welded the PP plates. Out of 9 experiments
conducted using square pin, most of the welded samples showed
satisfactory appearance. At lower rotational speed some cracks
are observed on the retreating side. Moreover, a straight line dis-
tinguishing the retreating side from the stir zone is seen which
 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254
Fig. 9. Welded sample.
Fig. 10. Sample welded using conical pin.
shows lack of bonding and improper material mixing. With
increasing rotational speed, voids are decreased. The formation of
cavities in the retreating side has also been observed during FSW
of ABS plates using a stationery shoulder [19,21]. The visual exam-
ination of samples welded with cylindrical pin showed bonding at
the top and bottom region only. The mixing of the material and its
transportation occurred due to the shoulder rotation at the top sur-
face. The stirring of the pin generated heat which was responsible
for bonding at the bottom. Moreover, the samples welded at the
lowest rotational speed, 500 rpm with cylindrical pin showed a
lack of bonding at the advancing side of the weld. Thus, the soft-
ened material got accumulated at the retreating side. Fig. 11 shows
a sample welded with 500 rpm, 5 mm/min and cylindrical pin.
Fig. 11. Sample welded with cylindrical pin.
251
Fig. 12. Material peeling.
Fig. 13. Material peeling taking the shape of flower like structure.
Comparing a sample welded with same welding parameters but
employing square pin showed uniform material transportation
than the case of cylindrical pin. The stir zone of square pin welded
sample had good surface finish which proves the efficiency of
square pin to blend the softened polymer material.
Material peeling out of the samples during welding was
observed as depicted in Fig. 12. This kind of defect was seen in
all samples during the welding process. The rubbing of shoulder
over the material surface removed the softened material in the
form of tiny particles. It was observed that with increasing rota-
tional speed, the frictional heat increases. At high welding speeds,
the plasticized material is peeled out which appears to be flower
like structures surrounding the stir zone as shown in Fig. 13. At
higher rotational speed, more heat was generated which deformed
the material; and at this stage with high welding speed when the
tool runs over the joint line, it peels off the material at its edges.
The peeling defect was also observed in other research works as
well [26].
3.4. Effect of shoulder diameter
To determine the effect of shoulder diameter, two tools with
different shoulder diameters were employed. The other dimen-
sions such as pin diameter (6 mm), pin length (5.6 mm) and pin
profile (cylindrical) were kept same for both the tools. The param-
eters for the welding were kept constant (rotational speed of 750
rpm, traverse speed of 5 mm/min). At first, the cylindrical tool with
16 mm shoulder diameter was employed to weld the polymer
sheets. A lot of distortion was found in the sample as shown in
Fig. 14. The rotation of shoulder in case of polymer has been proved
to be damaging. The high force and heat generated tends to
252 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254

Fig. 14. Distorted sample.

squeeze the material at the abutting edges lifting it from the fixed
ends. Secondly, we employed another tool with shoulder diameter
of 12 mm and cylindrical pin to weld the sheets by using the same
welding parameters. The reduction in diameter produced samples
which were almost flat and with less shrinkage as compared to the
samples welded with 16 mm shoulder diameter. However, there
was a minimal area of contact between the shoulder and the
work-piece in case of 12 mm shoulder diameter. The stir zone area
narrowed down with small shoulder diameter whereas thermal
shrinkage occurred with higher value of diameter in case of ther- Fig. 15. Effect of process parameters on tensile strength.

moplastics. This observation suggests avoiding the contact of rotat-

ing shoulder and the work-piece.

3.5. Effect of process parameters
Table 4 represents the tensile strength values obtained employ-
ing different parameters both in case of a cylindrical pin and a
square pin. A maximum of 59.82% joint strength efficiency with
respect to the base material is obtained with the square pin.
Fig. 15 shows the effect of rotational speed and traverse speed on
the tensile strength of the welds in case of cylindrical pin and
square pin. From the figure it can be inferred that tensile strength
increases with increase in rotational speed up to an extent and
then again decreases. At 500 rpm tool rotational speed, heat is
insufficient and thus the strength values obtained are low. At
750 rpm, the stirring of the pin increases and thus the heat gener-
ated is high which helps in better mixing of the material. But, with
further increase of tool rotational speed to 1000 rpm, the strength
of the weld decreases. This was due to the high amount of heat
generated which highly deformed the material. The effect of tra-
verse speed is also similar to that of the effect of rotational speed
on the tensile strength values. Sample welded with 500 rpm and
15 mm/min parameters is producing better result than 500 rpm
and 5 mm/min. This is because of decrease in formation of voids
and little enhancement in material flow which occurred due to
increase in welding speed. With constant rotational speed 500 rpm
and an increase in welding speed to 25 mm/min, strength
decreases due to lack of bonding. Thus, we can say, there exists
an optimum parameter set with which we can obtain high tensile
strength welds.
An interesting observation which can be found from Fig. 15 is
the trend with higher tool rotational speed. The strength values
obtained with cylindrical pin at tool rotational speed of 1000
rpm are higher than that of the square pin which stands in contrast
to rest of the results. At tool rotational speed of 1000 rpm and
welding speed of 5 mm/min, the heat generated is very high and
thus, it completely degrades the soft material and lowers the
strength. The frictional heat developed with higher tool rotation
increases and thus, the fluidity of thermoplastic material increases.
At this point, square pin is unable to make it flow along the edges
rather it throws the material out from the zone. The cylindrical
pin’s ability to blend the material with increase in fluidity
increases and thus higher strength is achieved. From Table 4, it
can be seen that the change in joint strength efficiency in case of
cylindrical pin (tool rotational speed 500 rpm; traverse speed 5,
15, 25 mm/min) is only 2.97% (16.97–14%). Whereas with same
pin (tool rotational speed 1000 rpm; traverse speed 5, 15, 25

Table 4
Tensile test results.

Sl No. x (rpm) v (mm/min) Pin profile UTS (MPa) Joint strength efficiency (%)
1 500 5 Cylindrical 4.62 14
2 500 15 Cylindrical 8.47 25.67
3 500 25 Cylindrical 5.6 16.97
4 750 5 Cylindrical 12.02 36.43
5 750 15 Cylindrical 18.47 55.97
6 750 25 Cylindrical 13.07 39.61
7 1000 5 Cylindrical 8.33 25.25
8 1000 15 Cylindrical 17.56 53.22
9 1000 25 Cylindrical 14.48 43.88
10 500 5 Square 11.2 33.94
11 500 15 Square 17.18 52.06
12 500 25 Square 16.16 48.97
13 750 5 Square 12.3 37.27
14 750 15 Square 19.74 59.82
15 750 25 Square 17.29 52.39
16 1000 5 Square 3.8 11.52
17 1000 15 Square 10.65 32.27
18 1000 25 Square 5.65 17.12

(x - tool rotational speed, v- tool traverse speed, UTS- ultimate tensile strength) (Base material UTS = 33 MPa)
 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254
Table 5
Effect of pin profile.
Pin profile Profile in Area Length of pin Static
static mode (mm2) (mm) volume
mm3
Square 4.252 = 5.6 101.15
18.0625
Cylindrical 3.14  32 = 5.6 158.256
28.26
(rps- rotations per second; calculated for 750 rpm = 12.5 rps)
mm/min) the change is 18.63%. This change in the values is due to
the variation in force during the welding process as explained in
Section 3.2.
The base material tensile specimen was tested first to deter-
mine its tensile strength. The thermoplastic material showed duc-
tile nature while all the welded samples were brittle. The
cylindrical pin welded samples at lower rotational speed were very
weak, hence got separated easily. The square pin welded samples
at higher rotational speed exhibited lower strength because of
material degradation. The samples broke from the retreating side
of the weld suggesting that it was the weakest zone.
3.6. Effect of pin profile
From Table 4, it can be observed that higher tensile strength is
achieved with square pin. Keeping the process parameters con-
stant, square pin develops more heat. The eccentricity of square
pin allows the plasticized material to flow easily around the probe.
Thus, it links the dynamic volume with static volume which deci-
des the flow path for the molten material from leading edge to the
trailing edge [33]. Moreover, square pin generates a pulsating
action because of the flat faces which helps in material blending.
The results can be validated from Table 5. The cylindrical pin being
simple in design has lower potential in blending the softened ther-
moplastic material and thus, the strength values obtained are les-
ser than square pin.
4. Conclusions
In this paper, the current state-of-art knowledge of FSW of ther-
moplastics has been presented and along with that an experimen-
tal investigation was carried out on PP sheets of 6 mm thickness.
From the experiment, the following conclusions have been drawn-
 Conical pin has failed to blend the thermoplastic material.
 Square pin has successfully welded PP sheets, obtaining a max-
imum weld strength efficiency of 59.82% with respect to the
base material.
 It was observed that there is an optimum set of tool rotational
speed and tool traverse speed with which one can obtain high
strength welds.
 At higher temperature, the fluidity of thermoplastic material
increases which then easily flows around the tool edges of
cylindrical pin and produces sound welds. At the same time,
the material with high fluidity gets thrown out with square pin.
 Tool with high shoulder diameter constantly degrades the
material, which distorts the sample lifting it from the fixed
ends. Avoiding rotation of shoulder will eliminate the fibre
breakage of the plastic material and thus surface defect such
as material peeling can be avoided.
 Square pin with higher ratio of dynamic volume to static
volume has produced better results than cylindrical pin.
253
Profile in Dynamic Ratio (Dynamic volume/ Pulses per second
dynamic mode volume (mm3) Static volume) (rps  faces)
158.256 1.56 50
158.256 1 Nil
Acknowledgement
This research was carried out by the support of FSW Lab, IIT-
Kharagpur. Sincere gratitude is expressed to our colleagues from
VSSUT and IIT-Kharagpur who have provided insights and inspira-
tion that has greatly assisted the research.
References
[1] P. Kah, R. Suoranta, J. Martikainen, C. Magnus, Techniques for joining dissimilar
materials: metals and polymers, Rev. Adv. Mater. Sci. 36 (2014) 152–164.
[2] S.T. Raithel, T. Kunstoffe, International, Lightweight and Innovative, (2015). akro-
plastic.com/file/i-pdf_akro_kuint_2015_09_.pdf (accessed March 28, 2016).
[3] Growing Importance of Thermoplastics in Aerospace, http://www.tencate.
com/news/2014/Growing-importance-of-thermoplastics-in-aerospace.aspx
(accessed March 8, 2016) (2014).
[4] A. Yousefpour, M. Hojjati, J.-P. Immarigeon, Fusion Bonding/Welding of
Thermoplastic Composites, J. Thermoplast. Compos. Mater. 17 (2004) 303–
341, https://doi.org/10.1177/0892705704045187.
[5] S.T. Amancio-Filho, J.F. dos Santos, Joining of polymers and polymer-metal
hybrid structures: recent developments and trends, Polym. Eng. Sci. 49 (2009)
1461–1476, https://doi.org/10.1002/pen.21424.
[6] W.M.W. Thomas, I. Norris, E.D. Nicholas, J.C. Needham, M.G. Murch, P. Temple-
Smith, C.J. Dawes, Friction stir welding—process developments and variant
techniques, SME Summit. (1991) 1–21. http://www.google.com/patents/
US5460317[sps]nhttp://scholar.google.com/scholar?
hl=en&btnG=Search&q=intitle:5,460,317#3[sps]nhttp://scholar.google.com/
scholar?hl=en&btnG=Search&q=intitle:5460317#3[sps]nhttp://
hegesztesportal.hu/hegesztes/tudastar/wt_fsw.pdf.
[7] Z. Kiss, T. Czigány, Applicability of friction stir welding in polymeric materials,
Period. Polytech. Mech. Eng. 51 (2007) 15, https://doi.org/10.3311/pp.
me.2007-1.02.
[8] Reciprocating friction stir welding joins thermoplastics.‘‘, Adv. Mater. Process.
ASM Int. (n.d.). https://www.highbeam.com/doc/1G1-65775078.html.
[9] S. Strand, Joining plastics - can friction stir welding compete?, in: Proc. Electr.
Insul. Conf. Electr. Manuf. Coil Wind. Technol. Conf. (Cat. No.03CH37480), IEEE,
2003: pp. 321–326. doi:10.1109/EICEMC.2003.1247904.
[10] TW Nelson, CD Sorensen, CJ John, Friction Stir Welding of Polymeric Materials,
US 6,811,632 B2, 2009.
[11] S.R. Strand, Effects of Friction Stir Welding on Polymer Microstructure, 2004.
[12] A. Arici, S. Selale, Effects of tool tilt angle on tensile strength and fracture
locations of friction stir welding of polyethylene, Sci. Technol. Weld. Join. 12
(2007) 536–539, https://doi.org/10.1179/174329307X173706.
[13] E.A. Squeo, G. Bruno, A. Guglielmotti, F. Quadrini, Friction stir welding of
polyethylene sheets, Ann. ‘‘DUNĂREA JOS” Univ. Galati Fascicle V,, Technol.
Mach. Build. (2009) 241–246.
[14] M. Aydin, Effects of Welding Parameters and Pre-Heating on the Friction Stir
Welding of UHMW-Polyethylene, Polym. Plast. Technol. Eng. 49 (2010) 595–
601, https://doi.org/10.1080/03602551003664503.
[15] S. Saeedy, M.K.B. Givi, Proceedings of the Institution of Mechanical Engineers,
Part B: J. Eng. Manuf. Investig. Effects Crit. Process Parameters (2011), https://
doi.org/10.1243/09544054JEM1989.
[16] G.H. Payganeh, N.B.M. Arab, Y.D. Asl, F.a. Ghasemi, M.S. Boroujeni, Effects of
friction stir welding process parameters on appearance and strength of
polypropylene composite welds, Int. J. Phys. Sci. 6 (2011) 4595–4601, https://
doi.org/10.5897/IJPS11.866.
[17] Y. Bozkurt, The optimization of friction stir welding process parameters to
achieve maximum tensile strength in polyethylene sheets, Mater. Des. 35
(2012) 440–445, https://doi.org/10.1016/j.matdes.2011.09.008.
[18] H. Ahmadi, N.B.M. Arab, F.A. Ghasemi, R.E. Farsani, Influence of Pin Profile on
Quality of Friction Stir Lap Welds in Carbon Fiber Reinforced Polypropylene
Composite, Int. J. Mech. Appl. 2 (2012) 24–28, https://doi.org/10.5923/j.
mechanics.20120203.01.
[19] A. Bagheri, T. Azdast, A. Doniavi, An experimental study on mechanical
properties of friction stir welded ABS sheets, Mater. Des. 43 (2013) 402–409,
https://doi.org/10.1016/j.matdes.2012.06.059.
254 S.K. Sahu et al. / Engineering Science and Technology, an International Journal 21 (2018) 245–254
[20] E. Azarsa, A. Mostafapour, Experimental investigation on flexural behavior of
friction stir welded high density polyethylene sheets, J. Manuf. Process. 16
(2014) 149–155, https://doi.org/10.1016/j.jmapro.2013.12.003.
[21] N. Mendes, A. Loureiro, C. Martins, P. Neto, J.N. Pires, Effect of friction stir
welding parameters on morphology and strength of acrylonitrile butadiene
styrene plate welds, Mater. Des. 58 (2014) 457–464, https://doi.org/10.1016/
j.matdes.2014.02.036.
[22] F. Simões, D.M. Rodrigues, Material flow and thermo-mechanical conditions
during Friction Stir Welding of polymers: Literature review, experimental
results and empirical analysis, Mater. Des. 59 (2014) 344–351, https://doi.org/
10.1016/j.matdes.2013.12.038.
[23] M. Pirizadeh, T. Azdast, S. Rash Ahmadi, S. Mamaghani Shishavan, A. Bagheri,
Friction stir welding of thermoplastics using a newly designed tool, Mater.
Des. 54 (2014) 342–347, https://doi.org/10.1016/j.matdes.2013.08.053.
[24] K. Panneerselvam, K. Lenin, Joining of Nylon 6 plate by friction stir welding
process using threaded pin profile, Mater. Des. 53 (2014) 302–307, https://doi.
org/10.1016/j.matdes.2013.07.017.
[25] M. Shazly, M. El-raey, Friction stir welding of polycarbonate sheets (2014)
555–563.
[26] V. Jaiganesh, B. Maruthu, E. Gopinath, Optimization of process parameters on
friction stir welding of high density polypropylene plate, Procedia Eng. 97
(2014) 1957–1965, https://doi.org/10.1016/j.proeng.2014.12.350.
[27] J. Gao, C. Li, U. Shilpakar, Y. Shen, Improvements of mechanical properties in
dissimilar joints of HDPE and ABS via carbon nanotubes during friction stir
welding process, Mater. Des. 86 (2015) 289–296, https://doi.org/10.1016/
j.matdes.2015.07.095.
[28] S. Hoseinlaghab, S.S. Mirjavadi, N. Sadeghian, I. Jalili, M. Azarbarmas, M.K.
Besharati, Givi, Influences of welding parameters on the quality and creep
properties of friction stir welded polyethylene plates, Mater. Des. 67 (2015)
369–378, https://doi.org/10.1016/j.matdes.2014.11.039.
[29] N. Sadeghian, M.K. Besharati, Givi, Experimental optimization of the mechanical
properties of friction stir welded Acrylonitrile Butadiene Styrene sheets, Mater.
Des. 67 (2015) 145–153, https://doi.org/10.1016/j.matdes.2014.11.032.
[30] Basic of Plastic Resin-Polypropylene, http://composite.about.com/od/Plastics/
a/What-Is-Polypropylene.htm (accessed July 7, 2016) (2016).
[31] Creative Mechanisms, https://www.creativemechanisms.com/blog/all-about-
polypropylene-pp-plastic (accessed July 7, 2016) (2016).
[32] ASTM D638 Standard Test Method for Tensile Testing of Plastics, n.d.
[33] W. Thomas, E. Nicholas, Friction stir welding for the transportation industries,
Mater. Des. 18 (1997) 269–273, https://doi.org/10.1016/S0261-3069(97)
00062-9.