Article

Investigation on Ultrasonic Welding Attributes of

Novel Carbon/Elium® Composites
Somen K Bhudolia 1,2,*, Goram Gohel 1,2, Leong Kah Fai 1,2 and Robert J. Barsotti Jr 3
1 School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50, Nanyang
Avenue, Singapore 639798, Singapore; goram001@e.ntu.edu.sg (G.G.); MKFLEONG@ntu.edu.sg (L.K.F.)
2 Institute for Sports Research, Nanyang Technological University, 50, Nanyang Avenue,

Singapore 639798, Singapore

3 Arkema Pte Ltd., Singapore 117528, Singapore; robert.barsotti@arkema.com

* Correspondence: skbhudolia@ntu.edu.sg; Tel.: +65‐82305882

Received: 3 February 2020; Accepted: 2 March 2020; Published: 3 March 2020

Abstract: Joining large and complex polymer–matrix composite structures is becoming increasingly
important in industries such as automobiles, aerospace, sports, wind turbines, and others.
Ultrasonic welding is an ultra‐fast joining process and also provides excellent joint quality as a cost‐
effective alternative to other joining processes. This research aims at investigating the welding
characteristics of novel methyl methacrylate Elium®, a liquid thermoplastic resin. Elium® is the first
of its kind of thermoplastic resin, which is curable at room temperature and is suitable for mass
production processes. The welding characteristics of Elium® composites were investigated by
optimizing the welding parameters with specially designed integrated energy directors (ED) and
manufactured using the Resin transfer molding process. The results showed a 23% higher lap shear
strength for ultrasonically welded composite joints when compared to the adhesively bonded joints.
The optimized welding time for the ultrasonic welded joint was found to be 1.5 s whereas it was 10
min for the adhesively bonded joint. Fractographic analysis showed the significant plastic
deformation and shear cusps formation on the fractured surface, which are typical characteristics
for strong interfacial bonding.

Keywords: polymer–matrix composites (PMCs); thermoplastic resin; ultrasonics; joints/joining

1. Introduction
Thermoplastic (TP) composites are preferred due to their excellent damping, impact, fracture
toughness, recyclability properties and their ability to be fused or welded to itself or with other
materials. Thermoplastic resin has an inherent ability to soften once heated above the defined
temperature range and retain their properties once they are cooled down. Hence, the manufactured
TP composite is an attractive candidate for welding to TP composites and also with dissimilar
materials like thermoset (TS) composites and metals. There is growing call from a wide spectrum of
industries (aerospace, automotive, sports, and many more) to eradicate the classical ways of joining
the polymer composite parts viz. mechanical fastening and the usage of control adhesives. The major
drawback of using the former is that composites are susceptible to high‐stress concentration
generated due to the holes and its labor intensiveness whilst the latter requires an incredibly longer
curing time as well as the longer surface preparation [1,2]. Both of the conventional approaches of
joining hinders the realistic chances of achieving shorter production cycles and are not suitable for
automation processes. Welding attributes of thermoplastics aids in the cost‐effectiveness of the
composite part to be manufactured in an industrial environment from forming until the finishing
steps [3–6]. The most feasible welding techniques available for fusion bonding of thermoplastic
composites are resistance [7–9], induction [8,10–15] and ultrasonic welding [6,16–35]. These behave

Materials 2020, 13, 1117; doi:10.3390/ma13051117 www.mdpi.com/journal/materials

Materials 2020, 13, 1117 2 of 18

differently in the way the heat is generated at the welding interface. Ultrasonic welding is an ultrafast
process of joining thermoplastic composites and works on the principle of the application of high
frequency and low amplitude vibration at the interface of the joining surfaces of the adherends to be
welded.
Many researchers have carried out extensive studies on the welding attributes of different types
of thermoplastic composites [36]. Liu et al. [20,37] investigated the effect of different welding
parameters like weld time, weld pressure, the geometry of the energy director (ED), amplitude, hold
time, and hold pressure on the weld quality using the Taguchi method. The material used for the
study was polypropylene (PP) reinforced glass fiber composites and Nylon 6 reinforced glass fiber
composites. Both investigations showed that weld time, the amplitude of vibration, and ED geometry
had a significant effect on the weld quality. The energy director also has a significant effect on the
weld quality as it allows for energy concentration during the joining process [24,38,39]. Y.K. Chuah
et al. investigated the effect of ED by ultrasonically welding pure acrylonitrile butadiene styrene
(ABS) and a polyethylene (PE) thermoplastic with different energy director configurations viz. semi‐
circular, triangular, and rectangular [39]. The semi‐circular shape was found to be the most efficient
welding condition while the triangular ED showed the lowest result. A similar study was carried out
by Villegas on the effect of the weld quality by using different configurations, direction, and shape
of the energy director [24]. A polyetherimide matrix reinforced carbon fiber composite was used for
the investigation and the results were examined by carrying out a static lap shear test. Results showed
that using multiple ED allowed for maximum coverage of the overlap area, hence also provides the
maximum strength.
An energy control mode was used by Keita Goto et al. [34] and constant time mode was used by
Wang Tao et al [35] to investigate the welding efficiency using the flat ED. Wang Tao et al.
investigated the effect of different welding times on the welding strength of CF/PEEK composites
using a flat ED with a 0.45 mm thickness [35]. Results from this study showed that with the gradual
increase in time, the weld strength also increased. In contrast, after an optimum time, a further
increase in time resulted in larger cracks and voids and the weld strength was significantly reduced.
Aerospace, automotive industries, wind turbines, and other industries require lightweight material
to increase efficiency by reducing energy consumption. Hence, using a fastener or bolted joint will
add the weight. Due to the long curing cycle of adhesive joints, ultrasonic welding is the preferred
option for mass production in industries [40]. Palardy et al. [41] investigated and showed that
ultrasonic welding could be scaled up by sequential welding (i.e., a continuous line of spot welding)
will serve the same effect of a continuous weld.
A research study was carried out on the fusion bonding of the Elium® composite by Murray et
al. [42]. In this research, resistance welding with different heating elements and induction welding
techniques were used to weld the Glass fiber Elium® composites for wind turbine blade applications.
There was around a 30% improvement in lap shear test for the welded samples compared to the one
bonded with adhesives. At 10 million cycles (defined stress for no failure), the fatigue limit for a
fusion‐welded sample was found to be 5 MPa compared to 3 MPa in the case of the adhesively
bonded sample. A preliminary study was also carried out by Bhudolia et al. to demonstrate the
fatigue response of an ultrasonically welded Carbon/Elium® composite [43]. The results showed a 7–
12% higher fatigue strength of welded joints with integrated ED compared to adhesively bonded
samples.
Recently, a novel acrylic thermoplastic resin, Elium® has been developed by Arkema, which is a
first of its kind TP resin system to cure at room temperature and possesses the same in‐plane
mechanical properties compared to the high‐performance epoxy resin. It can be manufactured using
liquid injection processes like Resin Transfer Molding (RTM) and vacuum infusion processes as it
possesses a viscosity that can go as low as 50 cP. Recently, significant research has been reported in
the literature investigating the impact [44–48], fracture toughness [49–51], vibration [52], flexure [53–
55], tensile [55,56], and other mechanical attributes of this novel resin system with different fiber
reinforcements. Current research aims at investigating the welding attributes of this novel resin
system, which could pave an excellent way of joining Elium® composite parts. It could be of
Materials 2020, 13, 1117 3 of 18

significant importance for wind turbines, automotive, sporting, and other applications where the
parts are currently joined using long cure and sophisticated control adhesives.

2. Materials and Manufacturing

2.1. Materials
For the manufacturing of thermoplastic composite laminates, FOE sized 12K 2 × 2 twill weave
dry carbon fibers supplied from CHOMARAT were used as the reinforcement of 380 gsm [57] and
Elium® 150 thermoplastic resin was used (Arkema, France) as the matrix material. Elium® 150 resin
undergoes radical polymerization to form high molecular weight acrylic co‐polymers with the
addition of a benzoyl peroxide initiator at a mixture ratio of resin to hardener 100:3 at room
temperature (RT) [4,46,49,58]. For the adhesive bonding study of composite materials, SAF 30 5
adhesives were used, provided by Bostik [59]. The important properties of the Elium® 150 resin and
the bonding properties for SAF 30 5 are shown in Table 1.

Table 1. Mechanical, curing, and bonding characteristics of the resin and adhesives used in the current
project.

Tg (Glass
Mixing Shear
Density Viscosity transition
Matrix Hardener/Initiator Ratio by Strength
(g/cm3) (cP) temperature)
weight (MPa)
°C
Elium®
Peroxide 100:3 1.2 100 @ 25 °C 22 110
150
Lap
Open Time Fixture time shear Curing
Adhesive Mixing ratio
(min) (min) strength temperature
(MPa)
SAF 30 5 1:10 2–3 8 22 RT

2.2. Manufacturing of Composite Laminates

2.2.1. Manufacturing of Flat Composite Laminates

To manufacture a flat composite laminate, a 3‐part (top, bottom, and a frame) RTM mold was
used as shown in Figure 1a with a frame thickness of 2 mm. Mold surface preparation was carried
out with a release agent to ease the de‐molding after manufacturing. Five layers of twill weave carbon
fibers of 270 × 270 mm2 were used in order to achieve the fiber volume fraction of 54%. Each layer of
the fibers was bonded using a binder and a heat gun was used to activate the binder. All of the layers
were placed into the mold along with the frame and the top mold was closed. Elium® resin was
prepared by mixing it with benzoyl peroxide at a weight ratio of 100:3 and was injected into the mold
at an inlet pressure of 2 bar and the outlet was set at atmospheric pressure in room temperature
condition.
Figure 1b shows the flow trend of the resin. Resin flows circumferentially in the mold and is
then introduced into the dry fibers. Once the part is completely filled and the excess resin comes out
through the outlet, both inlet and outlet ports were clamped, and the laminate was allowed to cure
at room temperature. Post‐curing of the laminate was performed in the heat plate at 65 °C
temperature for 45 min and later cooled to room temperature before demolding.
Figure 2 refers to the general steps for the manufacturing of the composite laminate using the
RTM manufacturing technique in the current research.
Materials 2020, 13, 1117 4 of 18

Figure 1. (a) Mold design for flat composite laminate manufacturing; (b) circumferential resin
strategy.

Figure 2. Manufacturing steps for the Resin Transfer Moulding (RTM) process to manufacture
composite laminate.

2.2.2. Manufacturing of Energy Director (ED) Integrated Laminates

For the welding study, thermoplastic composite laminates were also manufactured with the
integrated ED of semi‐circular configuration, as recommended by many researchers in the literature
[24,34,60]. The below subsections will give further details regarding the entire manufacturing
process.
Mould Design:
A four‐part (top mold, bottom mold, frame, and ED plate) mold was conceptualized for use in
the manufacturing of an ED integrated composite laminate. The ED integrated laminates that were
manufactured were also of 2 mm thickness. A frame of 6 mm thickness was used wherein a 4 mm ED
mold plate with semi‐circular grove (refer Figure 3) was tight fitted to give the laminate a final
Materials 2020, 13, 1117 5 of 18

thickness of 2 mm. The mold design of the ED integrated composite laminate manufacturing is shown
in Figure 3.
For the manufacturing of the ED integrated thermoplastic laminate, a similar method, as in the
case of manufacturing flat laminates was used (refer Figure 2). After mold preparation, the preform
was placed on the ED grooved plates so that when the resin is injected, it will flow into the grove and
become cured. The final consolidated part will have a neat cured resin ED, acting as a concentrated
energy source during the welding process. The injection parameters were different, which were
obtained by different manufacturing trials and the optimization study. The optimized injection
strategy was to keep the outlet at the vacuum of −0.91 bars and the resin was injected at an inlet
pressure of 2 bars. This resulted in a clean and almost bubble‐free ED integrated panel. It should be
noted that it was also mandatory to apply the full vacuum before the resin injection to remove the
entrapped air from the mold. Figure 4 shows the final bubble‐free integrated ED laminate for
welding.

Figure 3. Mold design of the Energy Director (ED) composite laminate manufacturing.

Figure 4. Manufactured Energy Director (ED) laminate with optimized parameters.

3. Experiments

3.1. Welding Methodology

The standard lap‐shear specimens [61] were manufactured to carry out the weld strength study
(Figure 5). An ultrasonic welding machine with a maximum power output of 3000 W, generator of
20 kHz frequency, and AWC‐6 microprocessor controller was used to weld the specimens in the
current research as shown in Figure 6. To weld the laminates, a constant time mode was used and in
the constant time mode, four parameters need to be set welding: weld time, hold time, weld pressure,
and the amplitude. At 100% of the amplitude, the value transferred on to the surface of the adherend
through the sonotrode was 65 μm.
Materials 2020, 13, 1117 6 of 18

Figure 5. Schematic of Semi‐circular ED Elium® composite_ Flat Elium® Composite (SC‐ELC_FL‐

ELC) welding configuration.

Figure 6 shows the ultrasonic welding machine with the fixture designed to weld the lap joint
samples. A step of 2 mm was milled to balance the offset with a specimen thickness of 2 mm. After
placing the specimen into the fixture, it was fixed by the tightening mechanism to ensure that it did
not move during the welding process. The SC‐ELC_FL‐ELC (Semi‐circular ED Elium® composite_
Flat Elium® Composite) welding configuration was used in the current research. Note that the semi‐
circular ED Elium® composite was kept on the top during welding. The schematic of the welding
configuration is shown in Figure 5.

Figure 6. Ultrasonic welding machine test setup.

3.2. Adhesive Sample Preparation

Adhesive and hardener were mixed properly and then applied to the adherend surfaces. The
mixed adhesive was applied to both the adherends in the required area and bound together. The
adhesive thickness was found to be around 0.65 mm.

3.3. Lap Shear Test

A standard Lap shear test was performed to evaluate the static shear strength of the bonded
joints due to its ease of usage as well as excellent reproducibility in the results [62]. The lap shear test
was performed according to the ASTM D5868‐01 standard [61]. A universal testing machine of 50 kN
Materials 2020, 13, 1117 7 of 18

load capacity INSTRON was used for performing this test and the test was performed with a
crosshead speed of 13 mm/min and was carried out at ambient conditions. Two different lap shear
strengths (LSS) were calculated as suggested by Villegas et. al [24] due to a difference in the welded
area, LSS1 and LSS2. LSS1 ids calculated as the peak load divided by the total overlap area and LSS2
is calculated as the peak load divided by the effective welded area. Thus, LSS1 defines the
effectiveness of the joint and LSS2 defines the weld quality. The effective welded area was calculated
by observing the actual welded area of the fracture surface after the specimen was statically tested
with ImageJ software (1.8.0_112). Table 2 depicts the technical specification used for lap shear testing
of the specimens.

Table 2. Lap shear testing specification.

ASTM D5868−01 [61]

Data measured Force (N), Displacement (mm) or Strain (με)
Mechanical Properties
Shear strength (MPa)
Calculated
Length = 101.6 ± 0.2 mm, Width = 25.4 ± 0.1 mm, Thickness =
Specimen dimensions
2 ± 0.01 mm, overlap area = 25.4 × 25.4 mm2
Feed rate 13 mm/min
LSS1 Peak load/ total overlap area; Weld efficiency
LSS2 Peak load/ actual welded area; Weld quality

3.4. Fractographic Investigation

The microscopic investigation was carried out on the fractured welded samples under static lap
shear tests to check the failure mechanisms. A stereotype optical microscope (OLYMPUS S Z X7) was
used to check the quality of the weld. Scanning electron microscope (SEM), JEOL SEM 5 600 LV was
used to carry out the fractographic investigation and to check the associated failure mechanisms on
the top and bottom surfaces of the adherends.

4. Results and Discussions

4.1. Initial Welding Trials

Initially, all the welding trials were carried out at 100% amplitude with varying weld time and
weld pressure. However, the weld resulted in (1) excess matrix flowing out of the interface and (2)
massive delamination of the top adherend. Samples were also welded at the lower weld time and
lower pressure at 100% amplitude, but the same phenomenon was observed. This can be attributed
to the higher energy transfer at the weld interface. Therefore, the amplitude was further reduced to
75% and the welded samples were statically tested in lap shear. It showed more promising results in
lap shear tests compared to the one with 100% weld condition. Additionally, from the literature, it is
advised to weld the acrylic‐based matrix in the range of 40–70 μm amplitude [63]. The maximum
amplitude in the current welding machine is 65 μm and 75% of this amplitude (48.75 μm) falls within
the desirable limit of the ultrasonic welding for acrylic polymers. Few samples were welded with a
50% amplitude, but the energy director was not melted and a visible gap between the adherends was
observed. In most of the available research [34,35], the amplitude is kept constant and other
parameters are varied, so in the current investigation, for the ED integrated samples, 75% (48.75 μm)
amplitude was fixed in the welding optimization study.
Materials 2020, 13, 1117 8 of 18

Figure 7. Visual and microscopic pictures of initial welding trials (a–c) over‐bonded (d) un‐bonded
joints.

In order to fix the weld time range, initially, the samples were welded in the time range from 0.5
s to 3 s and the joint strength was evaluated visually and by the lap shear test. At a lower time of 0.5
s and high‐pressure of 4–5 bars, satisfactory results were achieved, but when samples were welded
at higher welding times above 2 s, it resulted in significant damage to the adherend. At higher weld
time and even at lower pressure, the top adherend was damaged by delamination and fiber crushing.
The adherend failed by minimal manual force (pulling by hand) as it was over‐bonded and damaged.
Thus, the range of welding time for the current research was selected to be from 0.5s to 2 s with a 0.5
s interval. Considering hold time is not a significant weld parameter [35,37], so in the current
investigation, it was kept constant at 2 s. Weld pressures selected for the current research were 3, 4,
and 5 bar. Weld pressure of higher than the 5 bar showed delamination of the upper adherend as
noted in the case with a higher weld time and amplitude. While at a pressure lower than 2 bar, the
bond was very weak where it could be fractured with manual force. In the ultrasonic welding, the
energy is transferred to the interface from the top adherend, hence in the current study, the semi‐
circular ED Elium® adherend was kept on the top. Figure 7 shows the under welded and over bonded
specimen during the initial trials. Figure 8 shows a summary of the initial trials and the effect of the
welding parameters (weld time and weld pressure) on the bonding conditions.

Figure 8. Effect of near and far‐field weld parameters on the bonding of ED integrated panels.

4.2. Design of Experiments

In the current research, full factorial design (FFD) was used for the design of the experiments
[64]. From the initial trails above‐mentioned, the welding parameters selected for the welding of the
adherends are shown in Table 3.
Materials 2020, 13, 1117 9 of 18

Table 3. Design of the experiment for the current research.

Configuration Weld time (s) Weld pressure (Bar) Hold time (s) Amplitude (%)
SHED_FLAT 0.5, 1, 1.5, 2 3, 4, 5 2 75

4.3. Adhesively Bonded Joints

Figure 9 shows the load vs. displacement graph of the adhesively bonded Elium® composite
laminate. Five samples were tested in this study. The kinks in the curve can be attributed to the
difference in the distribution of the adhesive thickness during the adhesion process. This resulted in
a loss of adherence due to the difference in crack propagation at some points of the adhesively bonded
laminate. The average shear stress value obtained for the ELC_ELC (Elium® composite_ Elium®
composite) was 14.2 MPa.
Figure 10 shows the fractured interface of the adhesively bonded ELC_ELC joint tested under
lap shear testing. Adhesives can be observed on both the adherend surfaces representing a pure
cohesive failure. It should be noted that there was no surface preparation carried out on the adherend
surfaces. As for the adhesive bond, given that the adhesive was applied to the whole overlap area,
the LLS1 and LSS2 values will be similar.

Figure 9. Load vs. displacement curves for the adhesively bonded Elium® composites.

Figure 10. Failure surfaces of adhesively bonded ELC_ELC composites.

Materials 2020, 13, 1117 10 of 18

4.4. Welding of Elium® Composites with Integrated EDs

Figure 11. Load vs. displacement curves of SC‐ELC_FL‐ELC configuration at different welding
conditions.

Figure 11 shows the load vs. displacement curves of all welding conditions at three different
weld pressures (Wp), corresponding to four weld times (Wt) for the SC‐ELC_FL‐ELC configuration.
It should be noted that the graphs show the best representative curve of the average values of the
three trials carried out at the same welding condition. A small non‐linearity at the start of the curve
was observed, which is usually attributed to the backlash in the testing machine and fixture, at
around 0.1 mm of displacement. Later, the graphs showed a linear behavior until it reached the
maximum load, followed by a drastic load drop, showing the complete failure of the bond in the
laminated specimen.
As seen in Figure 11, at constant pressure, the maximum load value increased with an increase
in the weld time up to an instance after which a further increase in the weld time reduced the load
value or the corresponding weld strength of the sample. The weld time after which the load started
reducing represents the maximum/optimal weld strength condition corresponding to a specific
pressure condition. A similar effect was observed by Wang Tao et. al. where they investigated the
ultrasonic welding of CF/PEEK composites at different weld times with and without ED [35]. An
increase in weld time led to an increase in energy at the interface. Higher energy at the interface helps
in increasing the melting and flowability characteristics of the energy directors and adherends can be
welded more efficiently, increasing the weld strength. At higher energy, the matrix of the bottom
adherend also melts and significantly adds strength during the fusion bonding. Once an optimal weld
strength has been reached, any further increase in weld energy (i.e., weld time) at the interface will
reverse the phenomenon and there will be excessive resin flow out of the interface. The insufficient
amount of resin results in less deformation and the sample fails easily without taking a significant
load. In contrast, at a lower weld time, the amount of energy transferred to the interface is insufficient
to melt the resin and fusion bond it to the bottom adherend. From Figure 12, it can be clearly seen
that the weld area increased with the increase in weld time for each weld pressure condition. This
can be explained as with the increase in weld time, the energy concentration at the interface increases
and in turn elevates the temperature at the interface. This results in more melting of the resin and it
flows over a larger area. At a higher pressure of 5 bar, the graph shows slightly abnormal behavior.
This can be explained by the fact that at this pressure, there is more squeeze out of the resin, which
Materials 2020, 13, 1117 11 of 18

consequently results in a lower weld area compared to 4 bar. At an increased weld time and at 5 bar
pressure, this phenomenon starts dominating and leads to a further reduction in the weld area.

Figure 12. Welded area under all the welding conditions for the SC‐ELC_FL‐ELC configuration.

Figure 13. LSS1 and LSS2 graphs for SC‐ELC_FL‐ELC configuration.

Figure 13 shows the LSS1 and LSS2 values of the specimens at different welding conditions for
SC‐ELC_FL‐ELC. The LSS value showed a similar trend as explained for the load vs. displacement
curve. An increase in weld strength was seen with an increase in the weld time at specific pressure,
but after the optimal weld time had been reached, it starts to decrease. At higher energy fiber
breakage, delamination and other phenomenon were observed. The reason can be explained as the
resin being squeezed out of the interface at the higher weld time and pressure, which also damages
the adherend, so reducing the weld strength. This phenomenon will be explained further in Section
4.5.
Table 4 shows all the welding combinations carried out for SC‐ELC_FL‐ELC and its
corresponding LSS1 and LSS2 values. As seen from Table 3, the maximum LSS2 value was obtained
at the weld time of 1.5 s and 3 bar weld pressure while the minimum LSS2 value was obtained at a 2
s weld time and 3 bar weld pressure.
Materials 2020, 13, 1117 12 of 18

Table 4. LSS1 and LSS2 test results for the SC‐ELC_FL‐ELC configuration.

Weld time (s) Weld Press (Bar) LSS2 (MPa) Std. Dev LSS1 (MPa) Std. Dev
0.5 13.43 1.14 4.29 0.36
1 14.83 1.13 8.85 0.09
3
1.5 17.50 1.24 12.37 1.28
2 8.42 1.14 6.92 0.84
0.5 12.69 0.86 4.75 0.57
1 13.54 0.51 9.6 1.32
4
1.5 13.32 1.07 10.45 0.65
2 10.04 1.22 8.36 1.53
0.5 10.41 1.14 5.53 0.89
1 11.15 0.32 6.87 0.13
5
1.5 14.39 0.96 10.19 1.7
2 10.60 0.60 8.55 1.8

4.5. Microscopic Investigation and Surface Morphology

Figures 14 and 15 show the microscopic images of the failed surfaces of both the top and bottom
adherend of the SC‐ELC_FL‐ELC configuration at a maximum and minimum lap shear strength
conditions, respectively. As seen in Figure 1, at points A and B, features such as fiber pull out, melted
resin, and a chunk of resin can be observed in the bottom adherend. Whereas in Figure 15, at points
A and B, the fibers were extensively damaged, and a bundle of fibers had been pulled out, and voids
could also be observed. As seen from the side view of Figure 14c, lesser delamination of the top
adherend was observed compared to the top adherend shown in Figure 15c, where extensive
delamination of the adherend could be observed. Many other bonding features such as fiber imprints,
shear cups, matrix deformation, and others that represent an excellent bonding quality were also
studied. These features were investigated in‐depth by scanning electron microscopy (SEM)
observations.

Figure 14: Fracture surfaces of SC‐ELC_FL‐ELC specimen with maximum LSS value
Materials 2020, 13, 1117 13 of 18

Figure 15. Fracture surface of SC‐ELC_FL‐ELC specimen with minimum LSS value.

Figure 16 shows the surface morphology of the SC‐ELC_FL‐ELC composite laminate with the
highest lap shear strength of 17.5 ± 1.24 MPa achieved at the following welding parameters: 1.5 s, 3
bar, and 75% amplitude (48 μm). Figures 16a and 15b show the top adherends with shear cusps,
plastic deformation sites, good surface adhesion, and the fiber pullout with the adhered resin. All
these fractographic features tended to increase the lap shear performance of the welded laminate [65].
The surface morphology study of the bottom adherend as depicted in Figure 16c shows the cohesive
failure attribute with fiber rupture and pullout, which was recuperated by the matrix, indicating the
damage propagation in the vicinity of the carbon fiber/Elium® interface. Figure 16d shows the clear
fiber impingement and the cusps sites near the ED integrated zone. It also shows clear evidence of
the melting of the bottom adherend, which is essentially required to achieve excellent weld
properties.
Materials 2020, 13, 1117 14 of 18

Figure 16. Scanning Electron Microscope (SEM) fractography of SC‐ELC_FL‐ELC at maximum LSS.

Figure 17 shows the fractography of the SC‐ELC_FL‐ELC composite laminate with the lowest
shear strength of 8.42 ± 1.14 MPa achieved at the welding parameters of 2 s, 3 bar, and 75% amplitude
(48 μm). With only a 0.5 s increase in the weld time and keeping the other weld parameters constant,
there was significant over melting of the resin and excessive fiber damage occurred on the top
adherend (Figure 17a). There were also no signs of cusps near the deformed matrix sites as well as
no melting of the resin on the bottom adherend, which was evident by the presence of the bare fibers
and cleaner fiber rupture and pullouts.

Figure 17. SEM fractography of the SC‐ELC_FL‐ELC composite for the minimum LSS.

4.6. Comparison of Lap Shear Bonding Strength: Adhesives vs. Welding

Elium® composite was bonded to the Elium® composite by adhesive bonding and by ultrasonic
welding. It is to be noted that the comparison for welding was done using the LSS2 value, which is
the measure of the weld quality. While comparing the welding results to the adhesive results, it could
be clearly seen that the SC‐ELC_FL‐ELC composite showed a 23% higher LSS value when compared
to the adhesive bond strength (SAF 30 5) (Figure 18). Along with the significantly higher bonding
strength, the welding time was only 1.5 s as opposed to 10 min of curing time for the adhesive joints.
Materials 2020, 13, 1117 15 of 18

Figure 18. Comparison of the Lap shear stress (LSS) for the laminate joints with adhesively bonded
and welded.

5. Conclusions
Novel carbon Elium® composites with integrated energy directors were successfully
manufactured and an experimental study on the influence of different parameters such as weld time,
weld pressure, amplitude, ED type, etc. on the weld strength was conducted. The following are the
salient findings from the research:
(1) Elium® composites with energy directors can be efficiently welded with the optimized
welding parameters of a weld time of 1.5 s and weld pressure of 3 bar. The SC‐ELC_FL‐ELC welded
laminate configuration showed the maximum LSS2 value of 17.5 MPa
(2) The maximum lap shear strength of the welded laminate (SC‐ELC_FL‐ELC) was found to be
23.2% higher than the adhesively bonded Elium® laminates.
(3) SEM analysis showed the significant plastic deformation of Elium® resin and the shear cusp
formation near the resin‐rich sites. These observations were typical of the optimized weld condition
and had a direct relationship with strong interfacial bonding.
This research presents an excellent solution to reduce the joining time of Elium® composites in
various applications such as automotive, wind turbines, sports, and others with tremendous potential
for industrial automation with continuous welding.

Author Contributions: Conceptualization, S.K.B. and G.G.; Methodology, S.K.B. and G.G.; Investigation, S.K.B.
and G.G.; Writing—original draft preparation, G.G.; Writing—review and editing, S.K.B. and R.J.B.J.;
Supervision, L.K.F.; Project administration, R.J.B.J.; Funding acquisition, L.K.F. All authors have read and agreed
to the published version of the manuscript.

Funding: This research was funded by Nanyang Technological University, Singapore and ARKEMA, France
under RCA‐18/46 and RIE2020 Advanced Manufacturing and Engineering (AME) domain – Industry Alignment
Fund – Pre‐positioning polymer matrix composites program.

Acknowledgments: The authors would like to acknowledge the technical support rendered by students at the
Institute for Sports Research, Nanyang Technological University, Singapore and CHOMARAT, France for their
in‐kind support for providing the FOE sized fibers.

Conflicts of Interest: The authors declare no conflicts of interest.

Materials 2020, 13, 1117 16 of 18

References
1. Gilmore, C.M. Advanced Composites Manufacturing By Timothy, G. Gutowski. Mater. Manuf. Process.
1998, 13, 626, doi:10.1080/10426919808935286.
2. Niu, C.; Niu, M.C.Y. Airframe Structural Design: Practical Design Information and Data on Aircraft Structures;
Adaso Adastra Engineering Center: Hong Kong, 1999.
3. Offringa, A.R. Thermoplastic composites—rapid processing applications. Compos. Part A: Appl. Sci. Manuf.
1996, 27, 329–336, doi:10.1016/1359‐835X(95)00048‐7.
4. van Rijswijk, K.; Bersee, H.E.N. Reactive processing of textile fiber‐reinforced thermoplastic composites –
An overview. Compos. Part A: Appl. Sci. Manuf. 2007, 38, 666–681, doi:10.1016/j.compositesa.2006.05.007.
5. Advani, S.G.; Hsiao, K.T. 1 ‐ Introduction to composites and manufacturing processes. In Manufacturing
Techniques for Polymer Matrix Composites (PMCs), Advani, S.G., Hsiao, K.‐T., Eds.; Woodhead Publishing:
United Kingdom, 2012; pp. 1–12. doi:10.1533/9780857096258.1.1
6. Raza, S.F. Ultrasonic welding of thermoplastics. Ph.D. Thesis, University of Sheffield, Sheffield, UK, 2015.
7. Stavrov, D.; Bersee, H.E.N. Resistance welding of thermoplastic composites‐an overview. Compos. Part A:
Appl. Sci. Manuf. 2005, 36, 39–54, doi:10.1016/j.compositesa.2004.06.030.
8. Villegas, I.F.; Moser, L.; Yousefpour, A.; Mitschang, P.; Bersee, H.E. Process and performance evaluation of
ultrasonic, induction and resistance welding of advanced thermoplastic composites. J. Thermoplast. Compos.
Mater. 2013, 26, 1007–1024, doi:10.1177/0892705712456031.
9. Brassard, D.; Dubé, M.; Tavares, J.R. Resistance welding of thermoplastic composites with a nanocomposite
heating element. Compos. Part B: Eng. 2019, 165, 779–784, doi:10.1016/j.compositesb.2019.02.038.
10. Wedgewood, R.a.H., P.E. Induction Welding of Thermoset Composite Adherends using Thermoplastic
Interlayers and Susceptors. In Proceedings of 28th International technical conference, Society for the
Advancement of Material and Process Engineering; 1996, Seattle; WA; pp. 850–861.
11. Mahdi, S.; Kim, H.J.; Gama, B.A.; Yarlagadda, S.; Gillespie, J.W. A comparison of oven‐cured and induction‐
cured adhesively bonded composite joints. J. Compos. Mater. 2003, 37, 519–542,
doi:10.1106/002199803031776.
12. Ahmed, T.J.; Stavrov, D.; Bersee, H.E.N.; Beukers, A. Induction welding of thermoplastic composites—an
overview. Compos. Part A Appl. Sci. Manuf. 2006, 37, 1638–1651, doi:10.1016/j.compositesa.2005.10.009.
13. Chapter 11 ‐ Induction Welding. In Handbook of Plastics Joining (Second Edition), Troughton, M.J., Ed.;
William Andrew Publishing: Boston, MA, USA, 2009; pp. 113–120. doi:10.1016/B978‐0‐8155‐1581‐4.50013‐
5.
14. Pappadà, S.; Salomi, A.; Montanaro, J.; Passaro, A.; Caruso, A.; Maffezzoli, A. Fabrication of a thermoplastic
matrix composite stiffened panel by induction welding. Aerosp. Sci. Technol. 2015, 43, 314–320,
doi:10.1016/j.ast.2015.03.013.
15. Vijendra, B.; Sharma, A. Induction heated tool assisted friction‐stir welding (i‐FSW): A novel hybrid process
for joining of thermoplastics. J. Manuf. Process. 2015, 20, 234–244, doi:10.1016/j.jmapro.2015.07.005.
16. Tolunay, M.N.; Dawson, P.R.; Wang, K.K. Heating and bonding mechanisms in ultrasonic welding of
thermoplastics. Polym. Eng. Sci. 1983, 23, 726–733, doi:10.1002/pen.760231307.
17. Harras, B.; Cole, K.C.; Vu‐Khanh, T. Optimization of the Ultrasonic Welding of PEEK‐Carbon Composites.
J. Reinf. Plast. Compos. 1996, 15, 174–182, doi:10.1177/073168449601500203.
18. Devine, J. Ultrasonic Plastics Welding Basics. Ed. New York: Weld. J. 2001, 80, 29–33.
19. Gutnik, V.G.; Gorbach, N.V.; Dashkov, A.V. Some Characteristics of Ultrasonic Welding of Polymers. Fibre
Chem. 2002, 34, 426–432, doi:10.1023/a:1022912325343.
20. Liu, S.‐J.; Chang, I.T. Optimizing the Weld Strength of Ultrasonically Welded Nylon Composites. J. Compos.
Mater. 2002, 36, 611–624, doi:10.1177/0021998302036005476.
21. Tsujino, J.H.M.; Yoshikuni, M.; Miura, H.; Ueoka, T. Frequency Characteristics of Ultrasonic Plastic
Welding. Jsme Int. J. 2006, 49, 634–641.
22. Balle, F.; Wagner, G.; Eifler, D. Ultrasonic Metal Welding of Aluminium Sheets to Carbon Fibre Reinforced
Thermoplastic Composites. Adv. Eng. Mater. 2009, 11, 35–39, doi:10.1002/adem.200800271.
23. Zhang, Z.; Wang, X.; Luo, Y.; Zhang, Z.; Wang, L. Study on heating process of ultrasonic welding for
thermoplastics. J. Compos Mater 2010, 23, 647–664.
24. Villegas, I.F.; Bersee, H.E.N. Ultrasonic welding of advanced thermoplastic composites: An investigation
on energy‐directing surfaces. Adv. Polym. Technol. 2010, 29, 112–121, doi:10.1002/adv.20178.
Materials 2020, 13, 1117 17 of 18

25. Ultrasonics: Fundamentals, Technology and Applications, 3rd ed.; Bond, D.E.a.L.J., Ed.; Taylor & Francis:
Abingdon, UK, 2011.
26. Ishii, Y.; Biwa, S. Ultrasonic evaluation of interlayer interfacial stiffness of multilayered structures. J. Appl
Phys 2012, 111, doi:10.1063/1.4704692.
27. Wagner, G.; Balle, F.; Eifler, D. Ultrasonic Welding of Hybrid Joints. JOM 2012, 64, 401–406,
doi:10.1007/s11837‐012‐0269‐5.
28. Ultrasonics, Third ed.; Bond, D.E.a.L.J., Ed.; Boca Raton London New York: New York, NY, USA; CRC Press
(Taylor & Francis Group): Abingdon, UK, 2012.
29. Al‐Sarraf, Z.S. A Study of Ultrasonic Metal Welding. University of Glasgow: Glasgow, UK, 2013.
30. Benatar, A. 12 ‐ Ultrasonic welding of plastics and polymeric composites. In Power Ultrasonics; Gallego‐
Juárez, J.A., Graff, K.F., Eds.; Woodhead Publishing: Oxford, UK, 2015; doi:10.1016/B978‐1‐78242‐028‐
6.00012‐0pp. 295‐312.
31. Zhang, H.; Fernandes, H.; Yu, L.; Hassler, U.; Genest, M.; Robitaille, F.; Joncas, S.; Sheng, Y.; Maldague, X.
A comparative study of experimental and finite element analysis on submillimeter flaws by laser and
ultrasonic excited thermography; SPIE: Maryland, United States, 2016; Vol. 9861.
32. Wang, K.; Shriver, D.; Li, Y.; Banu, M.; Hu, S.J.; Xiao, G.; Arinez, J.; Fan, H.‐T. Characterization of weld
attributes in ultrasonic welding of short carbon fiber reinforced thermoplastic composites. J. Manuf. Process.
2017, 29, 124–132, doi:10.1016/j.jmapro.2017.07.024.
33. Villegas, I.F.; van Moorleghem, R. Ultrasonic welding of carbon/epoxy and carbon/PEEK composites
through a PEI thermoplastic coupling layer. Compos. Part A: Appl. Sci. Manuf. 2018, 109, 75–83,
doi:10.1016/j.compositesa.2018.02.022.
34. Goto, K.; Imai, K.; Arai, M.; Ishikawa, T. Shear and tensile joint strengths of carbon fiber‐reinforced
thermoplastics using ultrasonic welding. Compos. Part A: Appl. Sci. Manuf. 2019, 116, 126–137,
doi:10.1016/j.compositesa.2018.10.032.
35. Tao, W.; Su, X.; Wang, H.; Zhang, Z.; Li, H.; Chen, J. Influence mechanism of welding time and energy
director to the thermoplastic composite joints by ultrasonic welding. J. Manuf. Process. 2019, 37, 196–202,
doi:10.1016/j.jmapro.2018.11.002.
36. Ageorges, C.; Ye, L.; Hou, M. Advances in fusion bonding techniques for joining thermoplastic matrix
composites: A review. Compos. Part A: Appl. Sci. Manuf. 2001, 32, 839–857, doi:10.1016/S1359‐835X(00)00166‐
4.
37. Liu, S.‐J.; Chang, I.‐T.; Hung, S.‐W. Factors affecting the joint strength of ultrasonically welded
polypropylene composites. Polym. Compos. 2001, 22, 132–141, doi:10.1002/pc.10525.
38. Rashli, R.; Bakar, E.A.; Kamaruddin, S. Determination of ultrasonic welding optimal parameters for
thermoplastic material of manufacturing products. Sci. Eng 2013, 64, 19–24.
39. Chuah, Y.K.; Chien, L.‐H.; Chang, B.C.; Liu, S.‐J. Effects of the shape of the energy director on far‐field
ultrasonic welding of thermoplastics. Polym. Eng. Sci. 2000, 40, 157–167, doi:10.1002/pen.11149.
40. Wagner, G.; Balle, F.; Eifler, D. Ultrasonic Welding of Aluminum Alloys to Fiber Reinforced Polymers. Adv.
Eng. Mater. 2013, 15, 792–803, doi:10.1002/adem.201300043.
41. Palardy, G.F.V. Irene. Smart ultrasonic welding of thermoplastic composites. In Proceedings of the
American Society for Composites ‐ 31st Technical Conference, Netherlands, 2016.
42. Murray, R.E.; Roadman, J.; Beach, R. Fusion joining of thermoplastic composite wind turbine blades: Lap‐
shear bond characterization. Renew. Energy 2019, 140, 501–512, doi:10.1016/j.renene.2019.03.085.
43. Bhudolia, S.K.; Gohel, G.; Fai, L.K.; Barsotti, R.J. Fatigue response of ultrasonically welded carbon/Elium®
thermoplastic composites. Mater. Lett. 2020, 264, 127362, doi:10.1016/j.matlet.2020.127362.
44. Taillemite, S. Arkema Gains Ground in Composites and Launches A Revolutionary Range of Elium Liquid
Resins. Availabe online: http://www.arkema.com/en/media/news/news‐details/Arkema‐gains‐ground‐in‐
composites‐and‐launches‐a‐revolutionary‐range‐of‐Elium‐liquid‐resins/ (accessed on (Accessed on 12
February 2017)).
45. Matadi Boumbimba, R.; Coulibaly, M.; Khabouchi, A.; Kinvi‐Dossou, G.; Bonfoh, N.; Gerard, P. Glass fibres
reinforced acrylic thermoplastic resin‐based tri‐block copolymers composites: Low velocity impact
response at various temperatures. Compos. Struct. 2017, 160, 939–951,
doi:http://doi.org/10.1016/j.compstruct.2016.10.127.
Materials 2020, 13, 1117 18 of 18

46. Bhudolia, S.K.; Joshi, S.C. Low‐velocity impact response of carbon fibre composites with novel liquid
Methylmethacrylate thermoplastic matrix. Compos. Struct. 2018, 203, 696–708,
doi:10.1016/j.compstruct.2018.07.066.
47. Obande, W.; Ray, D.; Ó Brádaigh, C.M. Viscoelastic and drop‐weight impact properties of an acrylic‐matrix
composite and a conventional thermoset composite – A comparative study. Mater. Lett. 2019, 238, 38–41,
doi:10.1016/j.matlet.2018.11.137.
48. Kinvi‐Dossou, G.; Matadi Boumbimba, R.; Bonfoh, N.; Garzon‐Hernandez, S.; Garcia‐Gonzalez, D.; Gerard,
P.; Arias, A. Innovative acrylic thermoplastic composites versus conventional composites: Improving the
impact performances. Compos. Struct. 2019, 217, 1–13, doi:10.1016/j.compstruct.2019.02.090.
49. Bhudolia, S.K.; Perrotey, P.; Joshi, S.C. Mode I fracture toughness and fractographic investigation of carbon
fibre composites with liquid Methylmethacrylate thermoplastic matrix. Compos. Part B: Eng. 2018, 134, 246–
253, doi:10.1016/j.compositesb.2017.09.057.
50. Barbosa, L.C.M.; Bortoluzzi, D.B.; Ancelotti, A.C. Analysis of fracture toughness in mode II and
fractographic study of composites based on Elium® 150 thermoplastic matrix. Compos. Part B: Eng. 2019,
175, 107082, doi:10.1016/j.compositesb.2019.107082.
51. Shanmugam, L.; Kazemi, M.E.; Rao, Z.; Lu, D.; Wang, X.; Wang, B.; Yang, L.; Yang, J. Enhanced mode I
fracture toughness of UHMWPE fabric/thermoplastic laminates with combined surface treatments of
polydopamine and functionalized carbon nanotubes. Compos. Part B: Eng. 2019,
doi:10.1016/j.compositesb.2019.107450.
52. Bhudolia, S.K.; Perrotey, P.; Joshi, S.C. Enhanced Vibration damping and dynamic mechanical
characteristics of composites with novel pseudo‐thermoset matrix system. Compos. Struct. 2017,
doi:10.1016/j.compstruct.2017.07.093.
53. Bhudolia, S.K.; Perrotey, P.; Joshi, S.C. Experimental investigation on suitability of carbon fibre thin plies
for racquets. Proc. Inst. Mech. Eng. Part P: J. Sports Eng. Technol. 2015, 1754337115598489.
54. Bhudolia, S.K.; Joshi, S.C.; Bert, A.; Yi Di, B.; Makam, R.; Gohel, G. Flexural characteristics of novel carbon
methylmethacrylate composites. Compos. Commun. 2019, 13, 129–133, doi:10.1016/j.coco.2019.04.007.
55. Kazemi, M.E.; Shanmugam, L.; Lu, D.; Wang, X.; Wang, B.; Yang, J. Mechanical properties and failure
modes of hybrid fiber reinforced polymer composites with a novel liquid thermoplastic resin, Elium®.
Compos. Part A: Appl. Sci. Manuf. 2019, 125, 105523, doi:10.1016/j.compositesa.2019.105523.
56. Bhudolia, S.K.; Joshi, S.C.; Boon, Y.D. Experimental and Microscopic Investigation on Mechanical
Performance of Textile Spread‐tow Thin Ply Composites. Fibers Polym. 2019, 20, 1036–1045,
doi:10.1007/s12221‐019‐1017‐z.
57. CHOMARAT: C‐WEAVE™ 380T 12K HS. https://composites.chomarat.com/wp‐
content/uploads/sites/2/2017/03/CW037‐2.pdf
58. Bhudolia, S.K.; Perrotey, P.; Joshi, S.C. Optimizing Polymer Infusion Process for Thin Ply Textile
Composites with Novel Matrix System. Materials 2017, 10, 293.
59. Bostik SAF 30‐5. https://www.bostik.com/globalassets/aec/tds/saf‐30.pdf
60. Villegas, I.F.; Palardy, G. Ultrasonic welding of CF/PPS composites with integrated triangular energy
directors: Melting, flow and weld strength development. Compos. Interfaces 2017, 24, 515–528,
doi:10.1080/09276440.2017.1236626.
61. ASTM D5868‐01(2014). Standard Test Method for Lap Shear Adhesion for Fiber Reinforced Plastic (FRP) Bonding;
ASTM: West Conshohocken, PA, USA, 2014.
62. Kairouz, K.C.; Matthews, F.L. Strength and failure modes of bonded single lap joints between cross‐ply
adherends. Composites 1993, 24, 475–484, doi:10.1016/0010‐4361(93)90017‐3.
63. Guide to Ultrasonic Plastics Assembly; Dukane Intelligent Assembly Solutions: USA, 2011.
64. R. L. Mason, R.F.G., and J. L. Hess. Statistical Design and Analysis of Experiments; John Wiley & Sons, Inc.,
Hoboken, New Jersey: United States of America, 2003.
65. Barbosa, L.C.M.; de Souza, S.D.B.; Botelho, E.C.; Cândido, G.M.; Rezende, M.C. Fractographic evaluation
of welded joints of PPS/glass fiber thermoplastic composites. Eng. Fail. Anal. 2019, 102, 60–68,
doi:10.1016/j.engfailanal.2019.04.032.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).