Welding analysis and optimization of ultra- sonic welding in HDPE-5%PBI composite by CODAS decision-making approach

Engineering plastics are increasingly being used in structural and non-structural applications. As the need for plastics grows, so do the joining requirements. Ultrasonic welding is one of the favoured processes for polymer composites among numerous approaches, especially for bonding plates, automotive panel boards and couplings due to its joint quality and low cost. The High-density polyethylene (HDPE) matrix was reinforced with 5 wt% of polybenzimidazole (PBI) fibre by twin screw extrusion process and welding samples were prepared using injection moulding. In this study, Ultrasonic welding was used to connect 4 mm-thick lapped HDPE-5wt% PBI composites. Taguchi-based L27 orthogonal array was employed for experiments to study the lateral shear strength (LSS) and Shore-D-Hardness for the input parameters (i) weld duration (30,45 and 60 ms), (ii) frequency (19,20 and 21 Hz) and (iii) horn pressure (1.5, 2 and 2.5 MPa). The relevance of the parameters affecting shear strength and hardness of weld was studied using analysis of variance (ANOVA). ANOVA results shows the frequency (Hz) and weld time (ms) had contributed 32.4% and 31.6% in determining shear strength and weld time (ms) contributes 45% for shore D hardness. The ultrasonic welding process was carried out, and the multi-criteria decision-making (MCDM) approach was used to optimise the process parameters. From CODAS assessment the relative evaluation score of 2.444 was achieved by the twenty fifth experiment having the input parameters (IP) at 60 ms welding time, frequency of 21 Hz and pressure at 2.5 MPa considering each output factor with an entropy of 0.5. Fractography and weld morphology of welded joints were investigated using scanning electron microscope (SEM).


Introduction
In recent decades, industries have generally embraced ultrasonic welding of thermoplastics and it is still widely recognised as a very efficient technology that results in no polymer degradation. Ultrasonic welding is utilised in the packaging industry, for automotive panels, textiles, appliance manufacturing, etc [1][2][3][4]. The process of joining the composites is crucial in the manufacturing process of fibre-reinforced thermoplastic composites for structural purposes. While conventional joining solutions such as mechanical fastening and adhesive bonding are conceivable, they are not optimal since they might result in a range of structural abnormalities, such as stress concentration and fibre breaking in the case of mechanical fastening, as well as increased susceptibility to temperature, moisture, and other environmental factors in the case of adhesive bonding. These irregularities could ultimately lead to a reduction in the structural performance of the structure [5][6][7][8].
Fusion welding may be accomplished by the use of a variety of plastic welding processes. These welding procedures are classified according to the method used to provide heating elements at the anticipated bond point. Friction welding, which includes vibration, spin, stir, and ultrasonic welding techniques, is a well-known kind of fusion bonding. To solve the disadvantages of adhesive bonding, particularly the complexity of the heating process, alternative heating and curing technologies such as vibration, resistance, microwave, induction, and ultrasonic welding may be used.
When it comes to connecting thermoplastics, ultrasonic welding has been shown to be a viable option among all of the fusion methods tested. The intermolecular interaction between the thermoplastics induced by ultrasonic vibrations causes intermolecular diffusion and trapping of the molecular chain within the liquid state on the interfaces of the weld polymer in this approach, which is a distinguishing property [9][10][11][12][13][14][15][16][17].
When it comes to ultrasonic welding polymeric systems, semi-crystalline and amorphous thermoplastics are considered superior raw materials to crystalline thermoplastics. Nevertheless, energy directors (EDs) are often made by utilising compression or injection moulding techniques on a single mated thermoplastic specimen in order to localise ultrasonic vibrations to the bonding zone. However, thermosets are difficult to weld ultrasonically [18,19]. Yousef Pour et al offered an overview of several thermoplastic joining procedures [20]. They discussed several fusion bonding processes, as well as associated equipment, manufacturing methodologies, the influence of technical parameters on welded joints, and their applications. Nonhof et al [21] proposed a four-step procedure for doing ultrasonic welding. These stages have resulted in a variety of phenomena when polymers are ultrasonically welded. An effort to mimic thermoplastic ultrasonic welding was made by Suresh et al [22]. When ultrasonically welding thermoplastics, they employed viscoelastic heating as the major heating technique. The simulated findings are consistent with the temperatures measured during ultrasonic welding. In comparison to other energy directors, triangular energy director exhibits high temperature zones and higher ultimate tensile strengths (UTS). In spite of this, the number of experiments appears to be insufficient in depth and breadth in terms of ultrasonic welding thermoplastics. Villegas et al [23] investigated the impact of various energy directors' orientation and arrangement on their performance. While TRI ED has a significant impact on weld quality, this study only uses polyetherimide as an example of the material. Tsujino et al [24,25] investigated the welding properties of ultrasonic welding machines employing a range of frequency settings. The increased joint strength is achieved by employing 67 kHz welding equipment rather than 20 or 40 kHz welding equipment. Again, the effects of other weld parameters (ED, frequency, pressure, etc.) on weld strength were not explored in this study. In addition, many ultrasonic welding techniques have been used to execute plastic welding, however the thickness of the specimens appears to be incredibly thin. In particular, the energy director has never been used for thermoplastic welding. In order to have a solid understanding of the ultrasonic welding process for plastics, it is vital to consider all weld characteristics.
The objective of the energy director is to direct and concentrate the ultrasonic weld energy on to a smaller contact area, hence enhancing the welding rate and quality. While the energy director can increase the robustness of the process, it may complicate component assembly since the horn may not be able to identify the energy director readily in a production setting. Therefore, the process complexity would be reduced if ultrasonic welding could be used to connect lapping workpieces without an energy director. Research conducted by A V Aaken and C Hopmann [26,27] sought to assess the impact of moisture on the process-material interaction during ultrasonic welding. They conducted their investigation using polyamide, which is extremely hygroscopic. Despite what the authors say is the best way to study ultrasonic welding of plastics, this study only looked at one weld parameter (welding time) for how it changed with moisture levels.
The design of parameters is critical to achieving both good quality and affordable cost. We may achieve a large number of results from a small number of studies using the Design of Experiments (DOE). The ANOVA approach is commonly used to determine the significance of each factor [28][29][30][31][32]. The analysis of variance reveals that the most important weld component affecting the welding quality is the hardening time. Chang et al [33] employed experimental design to gain a thorough grasp of vibration and ultrasonic welding of thermoplastic polyolefins. If two thermoplastic polymers are used, ultrasonic welding may appear to have a greater weld strength than vibration welding.
As part of their prior research, Liu and his colleagues [34][35][36][37] looked at how to improve ultrasonically welded thermoplastics' bond strength. When linking virgin polypropylene and 10% glass fibre reinforced polypropylene composites, triangular energy directors were found to have the best strength, while when joining 30% glass fiber-filled composites, a semi-circular energy director had the highest strength. The researchers discovered that SEMI ED provided the strongest welds. Welding thermoplastics like HDPE and acrylonitrile butadiene styrene (ABS) can be improved by using fractional factorial design. They discovered that welding duration and pressure had a substantial effect on joint strength. Although Liu and Rani investigated how to optimise weld strength when thermoplastics are employed, they used the same Philips sample shape. As a result, weld strength optimization may be accomplished through the use of enhanced and novel sample shapes [38].
In addition to butt joints, ultrasonic welding was employed to link the lapping workpieces with energy directors. Ultrasonic weld energy is directed by an energy director to a smaller region of contact, which increases weld quality and speed. While the energy director may increase the resilience of the process, it is possible that the horn was unable to identify the energy director in the manufacturing environment, making component assembly more difficult. There will be less complexity in the process if an energy director isn't needed for ultrasonic welding of parts that are lapped [39].
With a few exceptions, the majority of the studies employed standard optimization techniques such as GRA and RSM for multi-objective optimization. Until now, the MCDM method of the combinative distance-based assessment (CODAS) technique has not been used to optimise the weldability of HDPE/PBI composites on a multi-objective basis. According to the industries, CODAS is typically used in the aerospace and automotive sectors, as well as in applications that demand a high-quality weld but are difficult to build using conventional techniques. Certain geometrical quality characteristics must be addressed in order to achieve a stronger weld connection [40].
From the literature above, there is no study related to ultrasonic welding of HDPE/PBI composite. The present study was undertaken to examine the weldability in ultrasonic welding of 4 mm-thick lapped HDPE/ PBI composites without energy director. A parametric analysis was carried out to identify the influence of various welding control factors on the quality metrics of HDPE/PBI composites, such as lap shear strength (LSS) and Shore D hardness. Multi-objective optimization was examined using the MCDM methodology, which is a component of the CODAS method. The fractography of the sheared joint was experimentally studied. This study provides valuable guidelines on the application of ultrasonic welding to join HDPE/PBI composites for electronic component manufacturing.

Materials
HDPE granules with grade name REPOL HD50MA180 was purchased from Reliance India, Ltd and used as a matrix in the current study. Merck, India Ltd supplied the PBI particles (90-130 nm) with the grade name of GAZOLE TM 5000 having 485°C glass transition temperature. HDPE −5wt% PBI was found with enhanced mechanical properties [41].

Fabrication of hdpe/pbi composite
In this work, HDPE/PBI composites were made and evaluated. Preheating the PBI (Poly (2, 5 benzimidazole) powder for 6 h and the high-density polyethylene (HDPE) pellets for 2 h in an 80°C hot air oven. Preheating was used to eliminate moisture from the raw ingredients, as moisture impairs the performance of plastics. A volumetric form feeder is used to combine HDPE and PBI nanoparticles into the twin screw extruder. The HDPE pellets and PBI powder are interlaced using the Flytech Engineering FUE-2EA twin screw extruder at a length to diameter ratio of 40:1. To achieve constant distribution, the screw speed was adjusted at 65 rpm and the volumetric feeder speed to 6 rpm. The Twin screw extruder temperature profiles ranging from 130°C to 195°C for this investigation. To mesh the PBI powder with HDPE granules uniformly, various screw pitches and equipment such as kneading blocks were used. The filament is fed through a pelletizer machine, which converts the HDPE/PBI composite into 3 mm granules. The HDPE-5% PBI test specimens for this welding study were manufactured using an injection moulding machine. Ultra-sonic welding was done on the manufactured samples with various parameters to obtain the best result for the weld as shown in figure 1.

Experimental details
In this current research, 4 mm thick HDPE-5%PBI composite specimens were manufactured by injection moulding process. The samples were of 130×20 mm 2 . The properties of HDPE-5% PBI composite had been shown in table 1. The lap area of weld is about 20 mm without surface cleaning. The vibration of the horn was in direction perpendicular to the clamped specimen. Using a computerised tensometer (Kalpak instruments and controls) with a repeatability of three readings per specimen, the lap shear strength was determined. The Shore D hardness was measured at the cross section of the weld joint, with a repeatability of five measurements at different locations along the sides of the joint.

Characterization of USW processed HDPE-5%PBI joints
In accordance with ASTM D638-14, tensile tests on the USW joints were performed utilising a computerised tensometer. The shore D hardness tester was used to conduct the test, which was conducted in accordance with ASTM D2240-15. Each test was carried out five times for each set of parameters, with the average of the results being taken into consideration. The metallurgical characterization of USW joints was investigated using a scanning electron microscope (SEM).

Results and discussions
3.1. Microstructure examination Analyses of produced specimens showed high-intensity peaks of carbon peaks and low-intensity ones for other elements including silicon, oxides and calcium. EDS investigation revealed that PBI fibres contain silica with a low-intensity peak in their chemical composition. It was found that Si, C, and O had high intensity peaks in figure 2. Twin screw extruder was used to evenly spread secondary reinforcement particles throughout the material's manufacture, as confirmed by EDS. Element mapping of HDPE-5% PBI composite is seen in the SEM images shown. The component base metal has larger concentrations of some elements, such as C, as revealed by the EDS study and tabulated in table 2. The spectra showed that HDPE's Matrix material contained Si and Ca. Secondary particles might be seen in the form of low-intensity peaks.
The input energy is calculated by the physical principles of ultrasonic technology, as given by equation (1).
where, E is the Input Energy (J), F the Welding Force (N), f represents Frequency (Hz), l denotes Peak to peak amplitude and t is the Vibration Time (s).

The interaction of input variables on the behaviours of the output
Typical welding frequencies range from the 40 KHz range to the 15 KHz range. The various parameters of the application will determine the best equipment and frequency to achieve an optimal weld for the parts. The 20 KHz frequency is the most commonly used ultrasonic frequency for plastics assembly and offers maximum flexibility, as it is suitable for a wide range of applications and thermoplastic components.  Investigations had been carried out to determine whether the hold time is a factor to be considered for the weld strength of composite joints. Increased weld strength was seen as a result of extending the hold duration, which was shown in table 3, to be beneficial until optimal values were obtained. When the two pieces of material are in contact under ultrasonic welding process the molecular contact occurs at the melting state at fusion zone. The inter diffusion of the polymeric chains during the vibration resulting in the formation of fusion bond between the two pieces. The samples are placed under certain pressure for a period of time. This procedure involved freezing the molten polymers just before they contacted each other on each side of the joint. It is natural for a semi-crystalline polymer to recrystallize as it cools. When you apply sustained pressure to a melt for an extended length of time, the polymer's molecular chains are capable of penetrating the joint interface and become intertwined with chains on the opposite side of the contact as shown in image 3 b & 3 c. This is called entanglement. This gives molecular chain formation and the joint strength will both greatly increase in magnitude as a result of this development.

Welding pressure influence on LSS & Shore D Hardness
As a result of the welding process, the samples are subjected to a weld pressure, which causes the energy to flow in one direction and the components fuse together. When the weld pressure is applied over an extended period of time, the polymer's molecular chains are able to pierce the joint contact and intertwine with chains on the opposite side of the joint. As a consequence of this, both the connectivity of the molecular chains and the shear strength of the components are greatly enhanced. As a result of the hold pressure, the tangling of molecular chains at the weld junction increases, resulting in increased joint strength as shown in figure 3(b). Too low of a hold pressure results in improper flow of molten material. Molecular alignment occurs transverse to the weld  surface as a result of an excessive amount of pressure, resulting in a weakening of the weld. The value of the shore D hardness has showed very little rise as a consequence of the increased pressure. This is due to the fact that the higher pressure causes the molecules to connect more strongly with one another. The molecular chains at the weld zone are enhanced as a result of the weld pressure. The hardness value at the weld junction should be good since it represents the stability of the specimen in response to changes in the welding circumstances. The voids that will be formed during the changes in flow and pressure results in low shore D hardness value.

Frequency influence on LSS & Shore D Hardness
For HDPE/PBI composites, the most essential factor was frequency, which provided a greater proportion to the final result than weld length or horn pressure did separately. The frequency, time period and frequency of an ultrasonic vibration all have a role in the significance of the vibration. The frequency of ultrasonic vibrations is  increased, which causes the molecules that make up a polymer to vibrate more vigorously. Frictional heat is generated as a result of molecules rubbing against one another during molecular vibration. The amount of frictional heat produced rises as the frequency of the vibration increases. Because of frictional heat, thermoplastics may melt more quickly. Increased vibration frequency results in a considerable increase in the quantity of energy taken in as well as the strength of the weld. Because crystalline materials need a larger amount of energy to melt and weld than non-crystalline materials, as previously stated in the introduction, this is most likely owing to the higher quantities of energy required for melting and welding. The quantity of energy wasted by a vibration is proportional to the square of the frequency of the wave. Additional benefits of lengthening your welding process include increased energy dissipation, which should result in an improvement in the strength of your weld. The frequency and duration of an ultrasonic vibration are well-known constants. The frequency was adjusted to 19, 20, and 21 Hz in this study. The Shore D Hardness value increased nearly 17% for higher frequency than the lower frequency joints. The joints at the HAZ with higher pressure shown enhanced hardness value due to tight packing of molecules from molten state to solid state transition. A constant and enough supply of molten material at the anticipated weld contact is essential for an optimal LSS value and Shore D Hardness value. Because of the large frequency's, a significant amount of molten material is utilised to start the weld. An equal distribution of molten material under pressure results in an appropriate heat affected zone (HAZ) and an improved life span. As a result, the supply of molten material (HDPE composite) may be tightly controlled by applying varied frequency and pressure levels. The SEM images shown in figures 3(a), (b), and (c) also reveals the changes in surface due to different frequency vibrations.

Weld time influence on LSS & Shore D Hardness
Semi-crystalline thermoplastic HDPE/PBI composite has a melting point of 135 degrees Celsius. The specimens are ultrasonically welded with enough ultrasonic energy. The HDPE composite's characteristics will decline if the weld duration is too long, therefore kept it to a minimum. The less time it takes to weld, the fewer cycles of vibration the specimens are subjected to. It takes a specific length of time for the vibrations to develop frictional heat, which is necessary for the weld connection to be formed. When ultrasonic vibrations are given to specimens, heat energy dissipates, and this is reflected in the loss modulus. The out-of-phase component of a complex modulus is another name for the loss modulus. The Shore D hardness value is enhanced mainly because of the weld time, the higher weld time helps the molten to solid state transition and allows the molecules to settle along with the pressure given. The contribution of weld time for hardness property is 45% as the heat is considered as a major issue for the plastic. A high amount of welding time, then, ensures that the best LSS and shore D hardness may be achieved. The majority of the time, plastics become tangled up in each other. The topological constraints (TCs) that regulate the viscoelastic and plastic changes are called entanglements. In most cases, these interfacial entanglements increase the bond strength and reduce the likelihood of a chain pull-out failure mode.

Analysis of variance for ultra-sonic welding joints
Using ANOVA, researchers may examine the correlation between input variables and output variables. The Fisher test is used to determine the rate of significance of the input variables, and it is utilised to do so (F-test).
The workpiece feed rate, tool profile, and vibrating tool frequency were all examined as input factors. According to table 4, the most important parameters for determining shear strength (MPa), the analysis of variance shows that weld time (ms), Frequency (Hz), and pressure (MPa) are crucial because experimental outcomes that provide F-values that are greater than those obtained using statistical values, are F (2, 8, 0.05)=4.46. Vibrating tool frequency contributes 32.39% of the total weight of a vibrating tool. There is a 31.6% and 29.3% contribution from the Welding time (ms) and Pressure (MPa) to this %age. Interaction is significant only when weld duration (ms) x frequency (Hz) is compared to the F-table value of F (4, 8, 0.05)=3.84, since the F-values estimated for interaction is 8.21, which is greater than the F-table value. The R-squared score is 98.82%. The interaction plot graph as shown in figure 4 reveals the parameter interaction of weld time (ms), frequency (Hz), and pressure (MPa) for improvement of shear strength. Table 5 shows the ANOVA for the shore D hardness of ultrasonic welding joints. The Weld time (ms),      Evaluation of weights for MCDM issues is carried out using the entropy approach, which is employed when the information of the decision matrix is available. C E Shannon introduced the Entropy theoretical concepts in 1948, and it has been in use ever since. The three primary weighing methods are subjective, objective, and integral weighting methods, respectively. When evaluating the weights for MCDM issues, the objective Entropy weighting method is utilised. Process parameters were determined using MCDM approaches, which involved evaluating the alternatives based on how well they performed or how well they produced the desired results. Schematic approach of the study is shown in figure 6. The following equation describes the entropy weights that are included with the CODAS system.

Optimization using CODAS Technique
The CODAS technique is used to determine the attractiveness of qualities using a ranking system. Calculation of the Euclidean and Taxicab distances between qualities with higher distances from the negative-ideal solution. There might be two distinct metrics. Primary distance is the Euclidean distance; secondary distance is the Taxicab distance. There are n choices and m criteria.
The CODAS approach was used to apply MCDM concepts to the selection of material handling equipment.
The following equation illustrates the phase for CODAS techniques. Step 1. The formation of a decision matrix (X).  Step 3. Constructing a decision matrix with normalisation weights.
The weights requirement, wj, must be between 0 and 1. w j -weight criterion for jth alternative, ).
Step 4. Establishment of a counter-intuitive solution.
[ ] ( ) =ńs ns 9 j m 1 ( ) = ns min r 10 j i ij Step 5. Calculation of the Euclidean and Taxicab distances between characteristics using the counterintuitive solution. Step 6. Preparation of the matrix of relative assessment (Ra).
ψ -Euclidean distances threshold function for two alternatives, τthreshold parameter value which lies between 0.01 and 0.05. τ = 0.02 is taken in this present study.
Step 7. Estimation of relative assessment score Step 8. By decreasing values, the assessment score (Zi) is ranked. The highest Zi value represents the ideal score.
The following table 6 summarises the weights for Optimization by Entropy that are included with CODAS (Combinative Distance based Assessment).
For all experiments, the highest Relative Assessment score resulted in the greatest output responses. According to the COADS-Relative Assessment score, the input parameters (IP) at 60 ms welding time, Frequency of 21 Hz and Pressure at 2.5 MPa had the best output performance characteristics across all tests. According to the results, experiment twenty-five demonstrates that a higher CODAS-Relative Evaluation score of 2.444 is the appropriate parameter for assessment.

Fractography of welded joints
The 4 mm thickness lapped HDPE composite with 5% PBI reinforced was used to get a better understanding of the impact of weld failure mode on joints. The joints were welded using the parameters stated above. The fractured surface of weld is vastly seen at the HAZ and not at the joints. The interlinking of materials gives more strength.
The SEM image 7(a) were taken at the fractured surface of the Heat affected zone, to show the voids formed. The small porosity is formed due to difference in the solid/melt state at the welding zone. The small pores were the air spots that occurred due to the difference in viscosity of the materials at fusion zone and heat affected zone. Due to high frequency vibration, small air pockets formed as the heat inside creates an entanglement between the materials to form weld joint. These air pockets form the void during the solidifying state under the pressure. Figures 7(b), (c) and (d) shows the lap sheared surface of different parameters. Figure 7(b) shows the weld shear of specimen 1 at low weld time, hold time, pressure and frequency. the sheared surface reveals the less pressure and the fusion was not proper. Figure 7(c) shows the specimen of L25 shear. The weld strength was high and with high pressure, high frequency and more weld time. Figure 7(d) shows the L16 sample shear fracture. The surface shows the deep grooves and tearing of joints.
The weld area, weld time and microstructure define the ultra-sonic weld strength. The microstructures formed were different at the fusion zone and heat affected zone. The excessive weld energy due to the higher welding time and Frequency at the faying surface resulted in double fusion zones. The flashes were formed at the sides of the weld joint. The softening of material is found at the HAZ and resulted in the reduction of Shore D Hardness value. The flow direction of the weld is changed due to higher welding time and horn pressure. The thermo-oxidative and thermal degradation of HDPE composite created some pores inside the material and

Conclusion
A twin-screw machine was used to create filaments by intercalating the HDPE pellets and 5% PBI fiber and making them into pellets. Injection moulding methodology was used to fabricate the HDPE-5%PBI composite specimens. The prepared composite was welded using an ultrasonic welding machine. An L-27 orthogonal array based on the Taguchi method was adopted to find the best output strength of a weld joint with varying input parameters. Utilizing entropy weight techniques, which were included in the CODAS approach, the output performance characteristics of the system were optimised, and the following conclusions were reached: 1. ANOVA studies revealed that the most important factors were frequency (Hz) and weld time (ms), with a contribution factor of 32.39% and 31.6% in determining the lap shear strength of the ultra-sonic welded HDPE-5%PBI composite.
2. The investigation of Shore D Hardness shows that the weld time and pressure are the major contributors, with 45.11% and 27.46%.
3. The interaction of weld time and frequency gave the best interaction for both the outputs, with a contribution factor of 4.83 for shear strength and 5.81 for shore D hardness.
4. The microstructure observation reveals that the rise in frequency (Hz) leads to an increase in flowability and an increase in entanglement. Because of the existence of voids and porosities, the shear strength and hardness of the material were lowered.
5. CODAS show the relative assessment score for optimum conditions was achieved at L25 for weld time of 60 ms, frequency of 21 Hz and pressure of 2.5 MPa with a score of 2.444.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).