Experimental Investigation of Processing Temperature Effect on Adhesive Bond Strength Between Engineering Thermoplastics in the Plastic Injection Molding Process

The 2022 Research and Market report into injection molding indicates growth in the multicomponent injection molding market, with annual growth of 6.4% predicted by 2026 [1]. This technique enables the production of low-tolerance mass polymer parts unlike other manufacturing methods [2]. Therefore, industries such as automotive, electronics, telecommunication, consumer goods, power tools, and toys typically prefer multicomponent molding over traditional single-material injection molding methods [3]. Toothbrushes and screw drivers are examples of products designed with a high-strength inner core and soft-touch materials at the regions that come into contact with users. Molded interconnected represents another example, where multimaterial injection is utilized, allowing one of the plastics to be exposed to metal plating for fabricating electric circuits on complex shapes.

An advanced injection technique called multimaterial injection, which combines the injection of plastic with various materials, was initiated in the 1960s [4]. In this study, two subcategories of multimaterial injection, called bi-injection and overmolding, were investigated. In bi-injection molding, two molten plastics fill the mold simultaneously from two distinct gates, effectively combining multiple steps into a single operation, resulting in increased production efficiency and reduced cycle times [5]. However, the main drawback of bi-injection is its higher equipment and tooling costs compared to other molding methods. In overmolding, the substrate refers to the portion that has already been filled and solidified [6]. Once the substrate has cooled to below the glass transition temperature, it is placed into another cavity as an insert where merging occurs. Although it represents a more cost-effective and straightforward approach, the overmolding method entails longer cycle times as a result of the requirement for multiple injection and cooling stages.

The essence of the multimaterial molding process relies on the adhesion between the two polymers. Interfacial adhesion strength describes the performance of injection molding of various polymers. Therefore, predicting and developing adhesion between materials is one of the industry's main goals [7–9]. According to Kisslinger et al., a variety of manufacturers for injection machines and raw materials (compounds) provide compatibility tables and process recommendations for adhesion strength [10]. Nevertheless, this insight stems from industrial part production experience, and achieving enhanced performance necessitates further integration with systematic experiments.

The adhesion strength between two different polymers depends on the material, injection process type, and geometric factors [11]. To date, various studies have focused on the different steps of injection molding to predict the adhesion strength [12,13]. The first step of the injection molding process is the mold design. Manufacturing two separate molds for overmolding and bi-injection is an outdated and expensive practice for this process [14]. To address this issue, Islam utilized a cavity transfer system with one convertible mold that can be readily switched from conventional injection molding to two-shot molding [15]. They examined two-shot molding and selective metallization for micro-molded interconnected device (MID) production. Furthermore, Chen and Young combined micro-injection with transfer molding using polycarbonate (PC) and acrylonitrile butadiene styrene (ABS) polymers for MID fabrication [16]. Similarly, Kisslinger et al. produced tensile test parts using hard/hard and hard/soft two-shots with another flexible mold structure called the core-back mold [10]. Here, the terms “hard” and “soft” denote rigid engineering plastics and thermoplastic elastomers, respectively. Their study found that overmolding caused a cold interface, which resulted in weaker bonding. Alternatively, using rotary table-assisted two-shot molding can produce better-bonded ISO 527 type I tensile test specimens [7,17]. Six used ISO 527 test specimens with a removable insert-equipped mold for overmolding hard–hard and hard–soft combinations [18]. In addition to the mold design impact, the adhesive bond strength also increases as the overlap area and interface roughness increase [19].

The adhesion bonding strength of injection-molded parts is strongly influenced by various injection process parameters, including the injection sequence, melt temperature of the polymer, mold temperature, heat dissipation rate, cooling, and injection speed [20]. Among these parameters, the dominant process variables are the melt temperature, melt pressure, heat dissipation rate, and cooling [21]. The tensile strength of the molded part increases as the melt temperature increases until it reaches the highest melting point [22]. In addition, Giusti and Lucchetta stated that high molten polymer temperature improves the adhesive bond strength via enhanced macromolecular diffusion while overmolding polypropylene parts [23]. Candal et al. demonstrated that the melt temperature exerts a greater influence on the adhesion strength compared to the mold temperature [24]. However, the mold temperature is still a significant factor as it determines the internal form and residual stress of the molten plastic by influencing the fountain flow structure [25]. Furthermore, elevating the holding pressure can escalate the interface residual stress, leading to a reduction in adhesion strength [24]. Therefore, Islam highlighted the melting and molding temperatures as dominant parameters [15]. Results showed that bonding is not possible without a sufficiently high interface temperature. Likewise, the weld line temperature exerts the most significant effect on the weld line strength [26]. Therefore, the shot sequence is critical for the overmolding process because the second shot polymer should be sufficiently hot to melt the first shot [11]. Changing the injection order decreases the re-melt zone, which in turn directly affects the adhesion [17]. However, excessive temperatures may be harmful to the properties of the resulting polymers. For the hard–soft combination, Hudacek examined the adhesion of thermoplastic elastomer (TPE) with 65 Shore A hardness overmolded on a PC substrate [27]. The adhesive bond strength exhibited an increase from the TPE melt temperature of 188 °C–204 °C, after which it began to decrease, attributed to the impact of thermal degradation. Goth et al. similarly confirmed that excessive temperatures harm the adhesive strength of metal coatings on MIDs [28].

Selecting a process variable range to achieve optimum results while minimizing the effects of invariable parameters effects on the process quality is critical. Hence, high-end plastic injection molding CAE simulation can be used for detailed analysis and parameter optimization of specimens [17,29]. In addition, Banerjee et al. employed algorithms to automatically identify potential manufacturing issues and produce design recommendations for the design process of injection-molded multimaterial objects [30].

In addition to the injection parameters, process control and monitoring have also been explored to observe parameter variations with high sensitivity during the injection molding process. As a basic-level process control, sensors are deployed for monitoring followed by adjustments to the process parameters by a skilled operator based on the quality of the parts [31]. For example, Chen and Turng constructed a process control system using temperature and pressure sensors on three distinct controlling parameters, namely, the barrel temperature, back pressure, and the velocity/pressure switchover point [22]. Mahshid et al. utilized displacement sensors on the mold to gauge part warpage, aiming to establish dimensional control for precision injection molding in multicavity applications [21]. Zhang et al. introduced a statistically derived model predictive control algorithm focused on coolant flowrate and coolant temperature to enhance part quality [32]. Park et al. subsequently employed pressure and temperature sensors for real-time monitoring, alongside an artificial intelligence machine learning algorithm for parameter optimization [33].

Thoroughly understanding the process parameters' effects on part quality requires an understanding of the adhesion mechanisms. Diffusion is one of the primary elements of adhesion [23,24], and prior research has established a direct correlation between the interface temperature and diffusion in polymers [15]. Clements conducted numerical analyses using matlab, which indicated that when thermoplastic polyurethane (TPU) is injected into a PC, the solid–melt interface temperature is insufficient for facilitating diffusion while reverse order provides good bonding [20]. Kisslinger et al. studied interfaces and polymer diffusion of two-shot molded parts for TPE-S and PP materials [10]. They analyzed the interface using Raman spectroscopy along a 12 μm length and provided evidence of changes in the material and the interdiffusion relation. In addition to the interface temperature, material compatibility also influences the diffusion between polymers. For example, PC and ABS are commonly chosen as a hard–hard combination of amorphous thermoplastics due to their advanced mechanical properties and chemical affinity [15,16].

To characterize the adhesive bonding mechanism of two polymers, weld line healing along the interface of two polymers can be examined. The strength loss can be explained by three mechanisms: insufficient molecular chain interdiffusion across the interface, parallel molecular orientation around the weld line, and stress concentration in a V-notch at the component surface [34–37]. For PS polymers, Onken et al. developed analytical healing models to estimate the strength of the weld line [38]. Their findings emphasized that high melt temperature and glass transition temperature difference result in faster interdiffusion. In addition, they claimed that interdiffusion is minimal below the glass transition temperature.

As evidenced by the literature review, the bonding between two different injected polymers relies on various process parameters and their impact on the diffusion mechanisms. The main goal of the present study is to characterize and understand the process parameter effects on two-shot injected parts. A specific injection mold with a pneumatic configuration and removable insert was designed and built to clearly distinguish the adhesive mechanisms. Tension tests were performed on molded specimens derived from two distinct types of PC (one with siloxane added), a polycarbonate–acrylonitrile butadiene styrene (PC/ABS) blend, ABS, and styrene ethylene butylene styrene (SEBS). A comparison between the overmolding and bi-injection methods is provided, with a focus on detailed control of process parameters such as melt temperature, mold temperature, and interface temperature. To the best of the authors' knowledge, no study has comprehensively demonstrated the effects of these parameters on bonding strength for bi-injection and overmolding at this level of detail. The resulting work provides an injection parameter guideline to achieve high adhesion strength.

Five different polymers compatible with each other were studied, as shown in Table 1 including material properties obtained from technical sheets. Lexan EXL 9330 resin with 130% elongation at break was selected to represent PC in this study. For the PC–PC control group tests, similar Lexan 943A resin, a transparent plastic, was selected to visually observe the merging of the polymers. For the PC/ABS blend, commercially available Triloy resin was selected, while Terluran resin was purchased for ABS grading. SEBS resin ENSOFT, a thermoplastic elastomer developed by Ravago, was also chosen due to its advanced compatibility with PC and ABS. PC, PC/ABS, and ABS are highly chosen engineering thermoplastics known for their high mechanical strength.

A cutting-edge tie-bar-less injection molding machine (ENGEL Victory VC160H) was used for sample manufacturing. This apparatus enables two-shot molding with its second barrel, which was placed 160 mm above the first barrel. The nozzle of the second barrel is horizontal and parallel to that of the first barrel, and the barrels can be operated simultaneously or individually. To maintain dry conditions and prevent humidity-related defects during the process, two desiccants (Moretto Hopper Loader F1F30K) were incorporated. The machine generates graphs illustrating actual injection speed and pressure with a resolution of 0.1 s. This feature helps to precisely control the process. An injection molding machine with a 160-ton clamping force capacity also defined our mold design limits.

A dog-bone tensile sample design was selected to characterize the adhesion strength. Initially, we considered the ASTM D638 standard, which is the most widely used testing standard for determining the tensile properties of plastics. However, the ASTM D638 Type-A specimen is 165 mm long, while the height difference between the barrels on the injection machine in this study is 160 mm. Thus, the tensile specimen design was modified to a 140 mm long geometry, as shown in Fig. 1 avoiding the need to use an excessively high-cost hot runner system.

A multi-functioned mold was designed and manufactured for the tensile specimens, as shown in Figs. 2(a) and 2(b). Separator, used for overmolding, was not located at the center of the dog-bone cavity to keep the barrel time equal for both material injections. Design was studied according to ENGEL Victory VC160H machine's barrels volumes. That allows optimum cycle time for production. There are two coupling halves for each nozzle, and the molten plastic directly goes to the mold core via a cold runner. The mold is equipped with a pneumatic system to separate the two plastic melts to create a controlled merging. After the first shot, the separator is removed and the second shot can be performed with an adjusted time delay. This results in attachment to the first shot polymer, thus generating a single multimaterial rigid part. In this study, first shot samples were manufactured and allowed to equilibrate at room temperature within the container. Subsequently, the separator went down, and the samples were overmolded.

The gate size for the parts typically changes for each material. If an improper gate width and height are selected (according to the material's viscosity value), flow changes may create residual stress, sink marks, jetting defects, etc. Therefore, in this study, PC and ABS with close melt flowrate values were selected for our single mold design due to their compatibility.

The plastic injection simulation was performed using cadmould software, which utilizes a finite element method solver for its analyses. Filling, packing, and warpage analyses of the injection-molded parts were performed. The simulation results gave initial insights into the selection of injection parameters, optimization of the final part, and mold design. In addition, the interface temperature between the polymers for adhesion analysis was predicted using these simulations. A structured mesh of approximately 0.8 mm was found to give converged results compared to finer mesh that took three times longer to run. The simulation was performed for 80 groups, where each group has a unique parameter set for three different mold temperatures, three different melt temperature, three different plastic combinations, and three different injection methods. In the overmolding scenario, the first shot plastic material was placed as an insert at ambient temperature to replicate the experimental conditions.

The investigation involved the analysis of machine monitoring data to examine the melt temperature, injection pressure as a function of time, and injection rate over time. Typically, mold-filling simulations furnish these parameters. However, variations between the experimental and simulated values exist due to the machine specifications such as its length-to-diameter (L/D) ratio and compression ratio.

Mold temperature has typically been measured using coolant temperature or IR thermometers, which can provide misleading information. Coolant temperature serves as input, whereas the mold temperature is an output parameter, underscoring their fundamental distinction. Additionally, IR thermometers exhibit high surface sensitivity and necessitate pausing the cycle for measuring the mold core. Hence, in this study, a k-type thermocouple was positioned approximately 5 mm from the base of the part, as shown in Fig. 2(a), and secured by a screw. The thermocouple was connected to a thermometer (Testo 925), and measurements were recorded every second. With this setup, the mold temperature was maintained within a sensitivity of 1 °C.

To evaluate the mechanical strength of the interface, tensile specimens as shown in Fig. 2(b) were tested at 2 mm/min speed using a Lloyd Instruments LR5K Plus machine. Each material combination was tested with ten samples to confirm whether the elongation at break and tensile strength analyses were statistically significant. To investigate the diffusion region, an optical microscope (Nikon Eclipse MA100) was also used to visualize the interfaces before and after testing.

To verify the mold design not causing any defects into the part, control tensile test specimens were fabricated from PC-A (Lexan EXL 9330)–PC-B (Lexan 943A). Figures 3(a) and 3(b) present tensile strength results as a function of injection melt temperature for bi-injection and overmold, respectively. Bi-injection samples’ strength values reach up to 59.43 MPa that is almost on the dot of single-material reference strength lines for PC as shown in Fig. 3(a). With the overmolding process, strength drops to 53.99 MPa. However, these control group tests prove that the setup is suitable to sensitively test the process parameter effects.

In addition to the experimental parameters, the interface between the two polymers should be as flat as possible to measure all the specimens over the same bond area. However, due to the nature of the bi-injection method, some interference can occur. This interference was minimized by decreasing the injection speed and equalizing the injection pressure.

During the study, tensile tests were performed at least 1 day after the injection due to the shrinkage and residual stress change over a day. A total of 15 tensile test specimens were injected for each group, with 10 being utilized for tensile testing, and the remainder was reserved for characterization studies. The average tensile strength values of the ten specimens are presented in the results. The maximum standard deviation of the tensile test results within each test group was determined to be 11% of the mean tensile strength. In total, 1260 tensile test specimens were injected with 84 distinct parameter settings. Eight hundred forty tensile tests were conducted.

Another method for characterizing mechanical strength is microhardness testing, which has been previously used by researchers to elucidate changes in the microstructure, molecular orientation, and micromechanical characteristics of injection-molded polymers. In our study, we conducted microhardness testing using a Shimadzu HMV-2T instrument with a Vickers square-based pyramidal diamond indenter to investigate polymer diffusion at the surface boundary of the selected samples. Samples were positioned under a glass magnifier, and positioning measurements were taken using calipers. Starting from the interface point, measurements were conducted at 0.2 mm intervals. At each location, three data points were recorded.

Another highly sensitive indicator for interface adhesion is based on Fourier-transform infrared (FTIR) spectroscopy, which has previously been used to study the surface energy of PP/PP 12 blend virgin and recycled carbon fiber-reinforced materials [44]. In our study, the interfaces before and after destructive testing were examined using an FTIR spectrometer (Bruker Alpha) to understand the material's interface variation in terms of the crystal structure variation. Samples were cut from their interface region at 10 mm × 20 mm surface area. Subsequently, the surface was cleaned using isopropyl alcohol. FTIR system has the capability to analyze a 2 mm × 2 mm area. Multiple analyses were conducted on the sample by gradually sliding it starting from the interface region.

Injection molding parameters optimization for the highest adhesion strength study focused on three major parameters: the injection method, the melt temperature $(Tmelt)$⁠, and the mold temperature $(Tmold)$⁠. Bi-injection and overmolding methods were compared. The melt temperatures of four different polymer grades (PC, ABS, PC/ABS, and SEBS) were studied. Unlike semi-crystalline polymers, amorphous PC and ABS exhibit a relatively wide range (∼20 °C) of melting temperatures. As a result, two melt temperatures for each polymer, with a 15 °C deviation, were selected, as listed in Table 2. In Table 2, the injection sequence defines the process method. For instance, material_A – material_B injected sample is manufactured by first injecting material B and then material A for overmold-1 (OM-1). The sequence is reversed for overmold-2 (OM-2). Bi-injection method (BI) injects the polymers simultaneously into the same cavity from different gates.

Like the melt temperature, there is a wide range of acceptable mold temperature values for PC and ABS [39–43,45]. Although ABS requires a lower mold temperature relative to PC due to its material composition, both materials are exposed to the same mold temperature in the bi-injection method. Thus, mold temperatures were selected from the top, bottom, and middle of the technical data sheet value ranges, as presented in Table 2. For a hard–soft injection, the mold temperature difference is larger. Six used two temperatures, such as 20 °C and 50 °C, in his study of hard–soft injection [19]. If two such extreme points are determined, intermediate values should also be added to the series. Therefore, three mold temperatures with a 15 °C difference (35/50/65 °C) were added to the experimental design in the present study. Only 210 °C was selected as a melting temperature parameter for SEBS because it has only a 10 °C melting range.

Tensile tests revealed that failures occurred at the interface. In comparison to the yield strength values of single materials, the failure strengths of multimaterial samples were lower than the yield strength value of the weaker constituent. Following the tensile tests, the sample length was measured by coordinate-measuring machine. The total length of the broken two pieces and the untested portion was nearly equal, with a deviation of approximately 100 µm. This observation suggests that the interface failure occurred within the elastic region of the injected materials. This analysis was performed to enhance confidence in our experiments given that the tensile test was carried out without an extensometer. Representative stress–strain curves for different injection methods and plastic combinations are shown in Fig. 4. Strain values were calculated by dividing the extension of the sample by the gauge length of 84.42 mm. PC–PC bi-injected sample exhibits the highest ductility and best adhesion due to the diffusion of the same material at the interface, as seen with a black dashed line in Fig. 4(a). PC/ABS–PC bi-injected curve, presented in green dashed line, shows high strength with reduced 7% elongation at break. For the same material combination, OM-1 and OM-2 values decrease. However, in each condition, the linear elastic slopes are similar, and the curves align on top of each other. Figure 4(b) demonstrates the PC–SEBS curve with a much larger elongation at break.

In this study, average tensile strength values were focused and reported for different material combinations and process parameters. Section 3.1 presents and discusses the effect of process parameters on strength. Section 3.2 focuses on the interpretation of the adhesion mechanisms.

There were two different injection parameters with two different injection methods for a total of four different parameter combinations.

The surface temperature of both the male and female mold cores can be referred to as the mold temperature. These two mold temperatures are usually equal in commercial settings. The mold temperatures for both materials are identical during bi-injection in this study. However, in overmolding, the mold temperature only relates to the second component that is injected later. A tensile bar is employed to examine the effects of the mold surface temperature on the interphase strength for hard–hard and hard–soft material combinations.

The hard–hard combinations included PC (Lexan EXL 9330)–PC/ABS and PC–ABS combinations. For the PC–PC/ABS injections, 11 different parameter groups were tested with three different mold temperatures (60/70/80 °C). Figure 5(a) shows the tensile strength values of each configuration. All the results are slightly lower relative to the strength of single-shot PC reference value (61 MPa). Increasing mold temperatures result in ∼5% higher strength values for bi-injection parts. However, the strength of overmolded parts tends to decrease slightly as the mold temperature increases from 60 °C to 80 °C. The mold temperature has only a minimal effect on the strength of the PC parts. Therefore, to reduce the overall number of specimens and associated labor required, only two mold temperatures (60 and 75 °C) were selected for the remaining injections with other plastics experiments.

For PC (Lexan EXL 9330)–ABS injections, 12 different parameter groups were tested with two different mold temperatures (60 and 75 °C). Figure 5(b) shows the tensile strength values of each specimen. Increasing mold temperatures plays an insignificant role in strength values for bi-injection. The mold temperature correlation is not clear due to different trends seen.

Figure 6 shows the PC–SEBS (hard–soft) tensile strength values of six different parameter groups with three different mold temperatures (35/50/65 °C). Bi-injected parts again do not show any significant change. Increased mold temperatures were found to result in slightly higher strength values for OM-2 parts, and it drops for the OM-1 case. The largest strength increment for bi-injection and overmolding were 5.79% and 25.50%, respectively, while the greatest decrease of 16.06% was recorded for OM-1. The mold temperature effect is not the dominant parameter for the adhesive strength [24], and it plays a more important role in controlling residual stresses for complex features.

Melt temperature is an important factor in determining the interface strength between amorphous polymers. Reducing the viscosity of the material enhances uniform contact. Additionally, increasing the interface temperature promotes greater polymer chain mobility, facilitating interdiffusion.

For the PC–PC/ABS hard–hard injections, 18 different parameter groups were tested with two different melt temperatures, each pair separated by 15 °C. Figures 7(a) and 7(b) depict the tensile strength values of each parameter set. The strength of bi-injected parts increased with the PC/ABS melting temperature, rising from 240 °C to 255 °C, as shown in Fig. 7(a). The circle and square markers correspond to an increase in PC melting temperature from 295 °C to 310 °C. All the square markers exhibit higher strength than circles, except for one data point. We can conclude that bi-injected samples get stronger as the melting temperature of the materials increased. This phenomenon can be interpreted as improved molecular diffusion between the mating polymers with the increasing temperature. The results for overmolded PC–PC/ABS are presented in Fig. 7(b). Similarly, the strength of OM2 parts increases as the PC/ABS melt temperature rises. In this case, PC parts were first injected, followed by PC/ABS overmolding. The overmolding melt temperature corresponds to the PC/ABS melt temperature, and its effects on better fusion at the interface can be observed. However, mold temperature and PC melt temperature effects are hard to distinguish for OM2 results. Concerning the OM1 samples, PC/ABS parts were initially injected and allowed to cool at room temperature before PC overmolding. Hence, PC melt temperature was investigated for its effect on adhesion strength. Notably, at melting temperature of 310 °C, square markers exhibit higher strength values compared to circle markers at 295 °C, with the exception of a single data point. The largest strength increases for bi-injection and overmolding are 23.04% and 18.82%, respectively.

In PC (Lexan EXL 9330)–ABS injections, the impact of melting temperature appears to be minor as shown in Fig. 5(b). For bi-injected samples, there is an observed increase in strength as the melting temperature increases, particularly when comparing the first two columns of data and the last two columns of data set presented with circle markers. Similar behavior is observed for OM1 parts where PC melting temperature serves as the overmolding temperature, bonding to an allocated ABS half. However, OM2 parts, in general, failed to achieve adhesive strength. Overmolding sequence importance has been also presented in the literature [11].

In contrast, the material combination of PC–SEBS does not exhibit a melting temperature effect in the results shown in Fig. 6.

We studied bi-injection, OM-1, and OM-2 injection methods effects. For the hard–hard plastic combinations, a significant strength decrease was observed between bi-injection and overmolding, as depicted in Figs. 6–8. Figure 5(a) clearly presents an arrangement of PC–PC/ABS from high strength to low based on average strength values, BI (48.78 MPa) > OM1 (30.13 MPa) > OM2 (13.30 MPa). PC–ABS injections also show a similar trend as shown in Fig.5(b). Average tensile strength values of 41.07 MPa, 25.07 MPa, and 0.88 MPa were recorded for BI, OM-1, and OM-2, respectively. Bonding of PC–ABS through OM-2 injection was not achieved for six out of the eight sample groups. To improve the molecular diffusion especially for the overmolding method, thin films can be tailor designed and implemented at the interface as studied in metal-polymer joining [46]. Plasma cleaning also is an alternative to improve the adhesive strength [47], but it might be a challenge to implement it without reducing cycle times.

Figure 6 shows tensile strength results as a function of PC temperature for PC–SEBS (hard–soft) combinations. Compared to the reference value of 14 MPa for SEBS, any injection method yields less than half of its performance when bonded with PC. In contrast to the hard–hard combinations, OM-1 produced a greater interface strength in hard–soft combinations where PC injected onto the positioned SEBS part in the mold. During bi-injection, PC is processed at temperatures either 295 °C or 310 °C, while SEBS is maintained at 210 °C. That may lead thermal degradation at the interface. Specification sheet of ENSOFT AG3138S(46) SEBS also suggests that the maximum melt temperature for the process is 240 °C. Consequently, a decrease in tensile strength may occur, as evidenced in a previous study [27]. Average tensile strength values of 3.71, 5.67, and 2.66 MPa were recorded for BI, OM-1, and OM-2 processes, respectively.

Although the injection method exerts a significant influence on the tensile strength of the polymer interface, the main mechanism determining the interfacial bond strength is the polymers' interface temperature. To characterize the interface temperature and tensile strength relationship, numerical simulations were conducted.

Interface temperatures for 80 combinations of parameters were analyzed. First, the relationship between interface temperature and strength was inspected, and a linear correlation was identified between these two parameters. This indicates that a greater temperature difference between the polymers' interfaces generates greater heat transfer, which is followed by mass transfer and diffusion. Subsequently, the temperature difference between the interface temperature and glass transition temperature (T_IF − T_G) was analyzed as illustrated in Fig. 8(a). T_G is set at 150 °C for PC, 125 °C for PC/ABS, and 105 °C for ABS. Red markers in the figure correspond to strength data in the right y-axis, and T_IF − T_G values are seen with circle markers and square markers for PC/ABS and PC, respectively. Three different colors for mold temperature variation although the effect is minimal, and they are on top of each other. Figure 8(a) illustrates why BI provides the highest strength for hard–hard plastics, as the temperature difference at the injection interface is at higher values compared to overmolding. OM2 cannot achieve glass transition temperature for PC at the interface. Thus, the strength of the part is at a minimum, also known as a cold interface leading to weak bonding [10].

Here, x and y represent the interface temperature and the glass transition temperature differences of material A (i.e., the one with a lower melting temperature) and material B, respectively. The coefficient values are listed in Table 3. The coefficients of the formula vary for different material combinations. The A and B values for the hard–hard combinations are quite similar. However, A has a negative value for the PC–SEBS pair because excessive temperatures can cause degradation in SEBS.

R² values of 0.98 and 0.97 were obtained for Eq. (1). Figure 8(b) shows that the interface's tensile strength correlates directly with the temperature difference between the interface and glass transition.

Interfaces, interferences, and crack surfaces were examined with both the naked eye and an optical microscope after tensile testing. Photographs of these surfaces were additionally captured, as illustrated in Fig. 9. Failure at the interface is attributed to the interference between the polymer components, resulting in the inclusion of both materials at the interface. In contrast, the overmolding method results in a distinctive flat interface shown in Fig. 9(b) indicate weak bonding between the pair of plastics.

Excessive interference between two polymers is a well-known issue that can occur during bi-injection. This phenomenon was detected by cutting the part with an electric fret saw directly after the injection. The corresponding side cross-sectional view is shown in Fig. 10. In bi-injection, interference occurred along ∼3 mm length as shown in Fig. 10. In contrast, overmolding does not provide sufficient heat during the second stage to melt the first shot resulting in weak interference. The same part was also examined under an optical microscope, as shown in Fig. 11. The interface line depicted in Fig. 11 delineates the boundary between two injected polymers: ABS on the left and PC on the right. For bi-injection, the transition between the polymers appears to be smoother when examining the OM images shown in Fig. 11. The findings regarding the interface and interference suggest that diffusion in bi-injected parts exceed that observed in overmolded parts.

FTIR spectroscopy is a valuable tool for identifying molecular bonds and detecting distinguishable peaks. Saviello et al. used FTIR micro spectroscopy to investigate the effects of photo-oxidation and additives on ABS [48]. Through their analysis, they identified peaks between 3000 and 3100 cm⁻¹ produced by C–H bonding, with a significant rise between 2220 cm⁻¹ and 2260 cm⁻¹ attributed to C–N bonding. This feature forms a crucial distinction between PC and ABS, as PC lacks nitrile groups while ABS contains them, thus resulting in a distinct peak in ABS. Additionally, the C–O bond produces peaks at 1696 and 1765 cm⁻¹, the aromatic ring in styrene produces a detectable peak at 1453 cm⁻¹, and CH₂ has a corresponding peak at 1494 cm⁻¹. The CH₂ and aromatic rings also caused several peaks to appear around 1300 cm⁻¹. In a study by Ferreira et al., FTIR methodology (with transmission and universal attenuated total reflection) was utilized to quantify automotive systems and investigate various PC/ABS blend mix ratios [49]. The peaks in the 3000–3100 cm⁻¹ and 1696–1765 cm⁻¹ ranges were found to be crucial for understanding mix ratios. These peaks have a near-attenuated total reflection value for 100% ABS material. However, as the PC ratio increased, the peak intensity near 1700 cm⁻¹ became stronger, while the peak at 3000 cm⁻¹ became weaker. Additionally, the peak at 2250 cm⁻¹, which represents a nitrile component, disappeared as the PC ratio increased. These findings are similar to those reported by Bano et al. from an analysis of the surface mechanical properties of ABS/PC blends [50].

We approached our samples by utilizing the characteristic peaks provided by the literature. As noted by Ferreira et al., particular attention was given to the peaks in the 1700 and 3000 cm⁻¹ bands [49]. The FTIR results from the PC–ABS and PC–PC/ABS bi-injected samples are shown in Figs. 12(a) and 12(b), where the measured sample is shown by a red line in each figure. The peak at 1700 cm⁻¹ is much stronger than the peak at 3000 cm⁻¹ for PC–PC/ABS injection. This suggests that ABS content for PC–PC/ABS injection amounts to approximately 20% based on the scale presented by Ferreira et al. [49]. However, the PC–ABS injection sample exhibits peaks suggesting that the ABS ratio in the sample is approximately 80%, which is higher than the expected result for PC–ABS injections. Analyses were also conducted on overmolded injections, but no changes were observed in the results. Our FTIR analysis was set to measure the area of 2 mm × 2 mm. However, as seen under optical microscopy, the interface region is much smaller on a micron-scale. Kisslinger et al. showed a 12 µm interface width in their study by utilizing Raman spectroscopy [10]. Therefore, to capture the details and process parameter effects at the interface, atomic force microscopy analysis, scanning electron microscopy analysis by gradually moving along the interface utilizing focused ion beam drilling, or Raman microscopy techniques can be used. These detailed analyses are reserved for a future study.

The variation in microhardness can serve as an indicator of polymer diffusion. For example, García Gutiérrez et al. demonstrated the influence of diffusion and entanglement in the weld line region of PC and PS samples, revealing a V-shaped trend in the hardness profile relative to the interface position [51]. Their study identified the minimum hardness at the center of the interface, indicating a weakened area approximately 0.2 mm from the interface. Likewise, Boyanova et al. studied PC and PS samples [52] and also found a similar behavior. Furthermore, Lopez established a correlation between the diffusion coefficient and microhardness in polymethyl methacrylate (PMMA) [53].

In this study, hardness data were measured with a 200 µm spacing, and the results for PC–ABS and PC–PC/ABS are presented in Figs. 13(a) and 13(b), respectively. Plots show a visible V-shaped reduction at the interface in the overmolded parts for both multimaterial combinations. At a distance of 0.2 mm from the interface on the PC side, the hardness measures 13.8 MPa, decreasing to 11.9 MPa at the interface, and then increasing to 13.5 MPa on the ABS side. In this instance, molecular diffusion and entanglement are not highly advanced, resulting in a significant hardness decrease at $z=0μm$ location. Similar behavior is applicable for PC–PC/ABS as shown in Fig. 13(b) with an approximate ∼1 MPa negative offset in values. For BI, polymers have a higher interface temperature at the point of merging, and chain diffusion and entanglement formation are promoted. Therefore, the hardness of the weld line remains consistently stable across the surface, as indicated by the red lines in Fig. 13, demonstrating successful healing of the weld line through increased diffusion. The high hardness values observed at the interface are consistent with the high tensile strength values of bi-injected parts. In contrast, single-material proximity hardness values are comparatively low for bi-injected parts when compared to overmolded counterparts. This may occur due to uniformly aligned polymer chains at higher process temperature during bi-injection, suggesting a lesser degree of entanglement.

By investigating bonding problems in multimaterial injection molding of polymers, this study identified several key factors that influence the adhesive bond strength between various engineering plastics. The effects of mold and melt temperature were studied with bi-injection and overmolding. Different blends of polymers and various mold parameter effects on mechanical strength were presented.

The mold temperature variation was found to have a minor effect on the tensile strength for both injection methods and for different plastic combinations. However, melt temperature plays a more significant role to improve the adhesive strength as the temperature gets higher. The most dramatic effect on part strength is the injection method. Out of three methods (bi-injection, OM-1, and OM-2), bi-injection provides the highest mechanical bonding strength. Significant reductions in strength were observed when transitioning from bi-injection to overmolding-1 and overmolding-2, 39% and 97%, respectively. In addition, deviations from the average values were smaller in bi-injection parts but larger in overmolding parts. Bi-injection high-strength performance was explained by the interface temperature difference between mating polymers. Specifically, the interface tensile strength was found to be directly correlated with the temperature difference between the interface and glass transition temperature of the injected polymer. This is directly associated with diffusion backed up with the microhardness and FTIR analyses.

This study suggests that optimizing multi-injection molding process parameters can lead to results nearly as strong as a single-material injected part. Accurate numerical modeling of the injection process can be used to predict the interface strength based on the derived equation from this study. This can guide designing multi-injection molds and selecting appropriate process parameters. Future studies could focus on analyzing the correlation between the interface temperature and tensile strength using a reptation bonding model to describe interdiffusion at the polymer interface for crack healing [39,42].

• ASELSAN.

There are no conflicts of interest.

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.