Metal–Polymer Joining by Additive Manufacturing: Effect of Printing Parameters and Interlocking Design

Abstract

Additive manufacturing has a strong potential to produce sound metal–polymer joints using controlled polymer deposition on a metallic substrate. In this way, this study aimed to explore the morphological and mechanical properties of metal–polymer joints produced through material-extrusion-based AM using a pin-based macro-mechanical interlocking mechanism. Joints were fabricated with polylactic acid deposited onto a heated aluminium alloy substrate to form the connection. The optimisation process was focused on improving the printing parameters and pin geometries to reduce voids and enhance joint integrity. The results indicate that optimised samples exhibit superior mechanical resistance, achieving a maximum load improvement with an overall strength increase of 368.97% compared to non-optimised joints. A combined pin geometry (50% cylindrical, 50% conical) was found to be the most effective. Morphological analysis confirmed uniform polymer deposition, ensuring reliable joint performance. These findings underscore the critical role of geometric optimisation in enhancing the strength and durability of metal–polymer joints in AM applications.

1. Introduction

Currently, manufacturing processes are undergoing the need to produce multi-material components. These parts combine the unique properties of different materials, enabling the creation of components with superior properties compared to those produced with a single material [1]. This approach has gained prominence in various industries, from automotive to aerospace, and is transforming the way products are designed and produced [2,3,4]. By combining materials with complementary properties, a wide range of performance requirements for components can be met.

Among the various material combinations, the metal–polymer combination is one of the most interesting and highly desirable due to the mechanical strength of metals and the lightness and corrosion resistance of polymers [5]. As expected, hybrid structures have emerged as a solution to take advantage of the unique properties of different materials. The connection between these two materials results in components that are not only stronger and more durable but also lighter. Furthermore, the addition of polymer to the metallic parts can be more sustainable and economical than the production of exclusively metallic parts, mainly in terms of raw material costs, and can also increase the performance of the component [6].

Traditional methods of connecting dissimilar materials may involve mechanical fastening, riveting, adhesive bonding, and welding; however, these methods have their limitations and requirements, often increasing the weight of structures or making them susceptible to thermal and environmental degradation [7,8]. Mechanical fastening with screws or rivets, while robust and effective, tends to add extra weight to the structure due to the additional materials required and may introduce weak points in the structure [9]. Adhesive bonding is known for its ability to create strong and uniform bonds without adding significant weight. However, its strength may be affected by temperature variations and exposure to aggressive environments such as humidity and chemicals [10]. Despite welding allowing the creation of a stronger and more durable connection between metals and polymers with greater resistance to adverse environmental conditions, it is a complex and slower process, which makes it less productive. These advancements are still in the developmental stages and more studies need to be carried out to effectively understand the feasibility and durability of the joints [11].

In line with this, the evolution of AM technologies has gained great attention, as they provide many advantages over conventional methods. AM is the technology of making parts from 3D digital model data by layering materials together, as opposed to subtractive manufacturing [12]. In the 1980s, early AM materials and equipment were developed and, since then, the way people think about manufacturing processes has been revolutionised and recognised throughout all industrial sectors, becoming increasingly present in society [13].

Chen et al. [14] reported that the advantages of AM technology include being a direct manufacturing process without moulds, being unrestricted by the degree of structural complexity, allowing more freedom for innovative design, providing high utilisation of materials, and being environmentally friendly [14,15]. AM can reduce the consumption of raw materials by up to 75%, making it an economical and sustainable technology [15].

Fused deposition modelling or fused filament fabrication (FFF) is an AM technique that creates 3D items directly from a computer-aided design model by selectively depositing material through a nozzle or orifice that follows the cross-sectional geometry of the object and resolidifies or acquires consistency [16]. This process, which is included in the category of material-extrusion-based AM [17], utilises a wide range of materials and does not require a large initial investment in equipment. This is also a new process in research for joining dissimilar materials, facilitating the production of complex multi-material parts and allowing manufacturers to create geometries and structures at reasonable costs and with less raw material [14,18]. It is also characterised by its inherent simplicity, high joint strength, short joining time, and high automation and standardisation viability [19]. However, the creation of metal–polymer hybrid structures through AM also presents some technical challenges, such as ensuring the connection between materials of different natures due to their poor compatibility. This approach has applications in a wide range of sectors. For example, in the automotive industry, the use of metal–polymer hybrids can result in lighter and more efficient vehicles, contributing to the reduction in carbon emissions, increased energy efficiency, and reduced vehicle overall wear [2,3,4].

The studies believe in the potential of this method and argue that investment in research is worthwhile. Previous research has demonstrated significant advancements in the joining of hybrid structures using FFF. For instance, various configurations and material combinations have been explored, and research involving overlapping joints with mechanical interlocking, as well as single-lap joint configurations without mechanical interlocking, has showcased the versatility of FFF in creating robust connections between diverse materials. For example, Ozlati et al. [20] explored an overlapping joint with mechanical interlocking between polypropylene (PP) and AA5083, in which a cylindrical pin connection was utilised. In turn, Falck et al. [21] investigated a single-lap joint configuration without any macro-mechanical interlocking mechanism, where polyamide-6 (PA6) and carbon-fibre-reinforced polyamide-6 (PA6/CF) were deposited onto an AA24024-T3 sheet. More recently, Belei et al. [22] examined single-lap joints between sandblasted Ti-6Al-4V substrates and polyamide/carbon fibre (PA/CF) sheets, and, although there was no macro-mechanical interlocking, the joint still exhibited good performance.

The recurring theme across these studies is the consistent improvement in joint integrity and strength, highlighting the immense possibilities of this technique. However, based on what was described in previous research and looking at the techniques currently used, it can be realised that this approach is still in its early phases of investigation compared to other techniques. Knowledge gaps have been detected, suggesting that there is much space for innovation and advancement in this area.

So, with the aim of advancing these studies and demonstrating the potential of AM to join different materials using controlled polymer deposition on a metallic substrate, this study aims to explore the morphological and mechanical properties of metal–polymer joints produced through material-extrusion-based AM using a pin-based macro-mechanical interlocking mechanism. This includes optimising the deposition parameters to improve the morphological properties of the connection, ensuring its structural integrity, and optimising the geometry of the joint.

2. Experimental Procedure

Dissimilar joints were produced by FFF, and the materials used were PLA and the AA5754-H111 aluminium alloy (Al). For the joining experiments, a 2 mm-thick AA5754-H111 sheet was used as the metal component of the joint, and the polylactic acid (PLA) was chosen as the polymer component, the filament (PLA Galaxy Silver, Prusa, Prague, Czech Republic) has a diameter of 1.75 mm [23]. These materials were chosen by considering the target application of the structures and their properties [24,25].

The research focuses on a joint concept based on an interlocking mechanism promoted by a deposited pin [26]. Strong differences in chemical, thermal, and mechanical properties of materials pose barriers to producing dissimilar material joints and the mechanical connection plays a crucial role in overcoming these barriers, so it is important to have an anchoring system. Self-constraining movements through interlocking mechanisms have proven to be effective in improving the mechanical behaviour of joints, minimising the impact of the differences in material properties across interfaces [7]. Figure 1 presents the geometry of the interlocking mechanism of the anchoring system, showing three configurations: 100% cylindrical (CL), 50% cylindrical and 50% conical (CL50-CN50), and 100% conical (CN), each influencing the way the system performs. All the designs have the same contact area, with a diameter of 10 mm, between the top surface of the pin and the PLA part.

The joints were produced using an Original Prusa I3 MK3 3D printer. A description of the joint production process is shown in Figure 2. After heating the printing table to the desired temperature, the PLA filament was extruded onto the heated table to create the pin (I). Then, the printing process was temporarily interrupted (II) to allow the integration of the metal substrate, drilled for a precise fit on the pin. When printing resumed, the PLA part was printed, completing the joint (III).

Figure 3 shows the final configuration of the metal–polymer connection from the bottom to make it easier to understand the pin position. The metal part is designed to have a hole with dimensions matching the pin, ensuring a precise fit. In turn, the polymer part is a solid parallelepiped without details. It should be noted that all the polymeric components were produced by material-extrusion-based AM.

The controlled polymer deposition on a metallic substrate was difficult to perform, due to the poor adhesion and incompatibility between the two materials. To meet this challenge, various experimental approaches were tried to modify the surface of the Al, with the aim of facilitating the connection and improving the deposition of the polymer on the metal. These approaches involved techniques such as painting, crimping, and sanding. Painting the Al surface proved to be the method that produced the best PLA deposition results, creating a layer of adhesion between the Al and PLA, and achieving a precise and uniform deposition of the polymer. Black aqueous spray paint was used as the paint resistance had a negligible effect on joint strength. Although it introduces an additional step in the process, since this occurs during the joint preparation phase and before actual production, it does not present a significant inconvenience.

After production, the deposited samples underwent longitudinal cutting and were prepared for morphological analysis using conventional metallographic practices, according to the ASTM E3-11 standard. For the morphological inspection, an optical microscope, Leica DM4000M LED (Leica, Wetzlar, Germany), was used. Tomography was employed to characterise the joints by using a Bruker Skyscan X-ray apparatus (Bruker, Billerica, MA, USA). The mechanical behaviour and tensile strength of the joints were analysed and evaluated by performing tensile-shear tests on 3 specimens using a universal testing machine, 100 kN Shimadzu AGS-X (Shimadzu, Kyoto, Japan), with 3 mm/min of deformation speed.

3. Selection of Printing Parameters

Due to the difficulty of adhesion between materials of different natures, the interface between the pin and the PLA part is responsible for guaranteeing a large part of the connection, which makes it mandatory to optimise the printing parameters of the polymeric components. In fact, defects and voids are important concerns in 3D printing as they can compromise the mechanical strength of printed parts. The presence of defects and voids in a 3D-printed object can result in a reduction in its final density and properties, depending on the printing parameters used [27]. It is crucial to minimise these defects and voids in order to improve the integrity and performance of the connecting joint [28], making it necessary to manufacture the polymer component with a density close to 100%. This way, prior to the fabrication of the joints, the printing settings were optimised, which corresponded to the first stage of the study of the joining conditions.

The analysis of voids in 3D-printed structures using the FFF technique reveals that, theoretically, these gaps do not appear randomly but repeat layer by layer during printing, as can be observed in Figure 4a. The types of voids include gaps in vertical shells (I), between contour lines (II), between infill lines (III), and empty areas resulting from the geometry of the infill pattern. For the first print test, a 100% infill with a rectilinear style was used. The analysis of a printed sample confirms the presence of the described voids, although their geometry is not perfectly regular, as illustrated in Figure 4b.

This study of void patterns in 3D prints helps to understand how the infill influences the structural integrity of the piece. To explore deeper into this understanding, various printing parameters were examined. As shown in Figure 5, the parameters can be separated into four groups [28]. For this study, the influence of parameters such as temperature and humidity of the filament, bed and nozzle temperature, printing speed, raster angle, layer thickness, and number of contours was analysed.

The selection of the test parameters, and their variation, was based on the literature [29,30,31,32,33]. The optimisation of the printing parameters was started by applying all the suggested parameters provided by the slicing software (PrusaSlicer, version 2.6.1). The analysis was focused on the optimised production of the final joint design for a sample density close to 100%. Figure 6 shows the process of evolution of the voids in a PLA sample under the influence of the parameters mentioned above.

Table 1. Parameters obtained for optimised density conditions.

Parameters		Optimised Values
Environmental Factors	Filament temperature	50 °C
	Filament humidity	20–25%
Nozzle temperature	First layer	215 °C
	Infill	210 °C
Printing speed	Contour	36 mm/s
	First layer	20 mm/s
	Infill	64 mm/s
	Top layer	32 mm/s
Bed temperature		60 °C
Raster angle		45°
Layer thickness		0.1 mm
Number of contours		3

shows a summary of the final values of the parameters used to optimise the printing conditions. The best results were obtained by introducing control of the filament conditions, the filament temperature was set to a constant value of 50 °C, and then the filament humidity was varied between 20 and 25%. To control the properties of the filament, a FilaDryer S2 from Sunlu^TM (Zhuhai Sunlu Industrial Co., Ltd., Zhuhai Guangdong, China) drying cabinet was utilised. The printing bed was heated to a temperature of 60 °C and the printer was calibrated frequently to ensure good printing quality.

The sample shown in Figure 7a, which is established theoretically for a density of 100%, adopting the suggested parameters of the slicing software, corresponds to the starting stage of optimisation. This sample presents a significant number of voids distributed relatively across all the object volume, although they are more prevalent in the centre and at the contour, as can be seen from both the tomography results (red zones) and the transverse section. On the other hand, the outcome of optimising the printing parameters, which is illustrated in Figure 7b, shows some gaps in the contours, but they are not significant when compared to the first results. An almost filled volume without voids can be observed, which contributes to improving the mechanical behaviour of the components.

Table 2. Properties obtained experimentally for the PLA samples before and after parameter optimisation.

Properties	Before Optimisation	After Optimisation
Average Maximum Load [kN]	1.23	1.39
Standard deviation [kN]	0.00	0.02
Average Maximum Normal Stress [MPa]	48.44	53.43
Ratio of voids in the transverse section [%]	7.57	1.21

shows the values of the average maximum in tensile testing and the ratio of voids in the transverse section, before and after the optimisation of the printing parameters. An increase in the mechanical strength resulting from the optimisation of the printing parameters can be realised. The average maximum load after optimisation is 13% higher than before optimisation. Printing parameter optimisation was validated by the quantitative analysis of the morphology of the FFF printed samples. The results revealed a considerable percentage of voids in the samples printed with the parameters suggested by the slicing software. In turn, a decrease in voids was achieved with parameter optimisation, suggesting a more stable and uniform internal structure. The outcomes demonstrate the benefits of the printing parameter optimisation that was performed, which, although it may seem like a small difference, has an impact on the resistance of the polymeric components in the dissimilar joints.

The morphologies of the longitudinal sections of two Al-PLA samples, with CN geometry, which was initially considered the geometry with the highest capacity to block movement and resist load, before and after the optimisation of printing parameters, are illustrated in Figure 8. Important differences in the morphology of the interface of samples are clearly discernible in the macrographs. In fact, as observed in Figure 8a, before optimisation, the sample shows a significant number of voids, visible as black spots in the areas within the printed material, PLA. These irregularly distributed voids indicate imperfections in the printing process. After optimisation, in Figure 8b, the sample shows a notable reduction in the number of voids, with a smoother and more uniform interface. In summary, the morphological analysis indicates that the dissimilar Al-PLA samples present a sounder and more reliable material interface.

Al-PLA joints were also tested for tensile-shear strength to study the effect of depositing parameters optimisation on the mechanical properties of the dissimilar samples. The optimised samples of the Al-PLA joint show greater mechanical strength compared to the non-optimised samples for the same joint, reinforcing that the optimisation of the printing parameters was effective.

Table 3. Properties obtained experimentally by tensile tests for dissimilar CN Al-PLA joints.

Properties	Before Optimisation	After Optimisation
Average Maximum Load [kN]	0.63	0.85
Standard deviation [kN]	0.18	0.23

shows the mechanical behaviour of the joint with a conical pin, from which it can be observed that the adjustments made have contributed to improving the resistance of the joints, reducing the voids that could compromise their integrity.

4. Discussion

For the dissimilar material joints, as mentioned before, various pin geometry concepts were tested, i.e., with no macro-mechanical interlocking, with a cylindrical and a conical pin-based interlocking, and with a combination of both geometries. The detailed geometry of each pin can be seen in

Table 4. Geometry of the different mechanisms of interlocking. Dimensions in mm.

Mechanical Interlocking Abbreviation		Geometry
Without pin	-	-	-
100% Cylindrical	CL
50% Cylindrical 50% Conical	CL50-CN50
100% Conical	CN

.

Various geometries of the interlocking mechanism were analysed by FEM using Nastran software (Inventor Nastran 2025), an extension included in Autodesk^® Inventor (San Francisco, CA, USA), in order to improve the interlocking mechanism by predicting weak points and consequent failure for each pin geometry studied. The simulation aimed to replicate a tensile test, with the applied displacement corresponding to the conditions experienced by the specimen under the tensile-shear tests. Simulations were performed through a nonlinear finite element analysis controlled by displacement, iterated to meet a convergence criterion. As can be seen in Figure 9, where the FEM model is schematically represented by half of the specimen studied, taking advantage of its symmetry, constraints were applied to the tip of the specimen, restricting all movements and rotations. In addition, a constant displacement was applied to the end of the PLA part, which contains the pin, and shims were introduced to ensure the alignment of the displacement along the specimen. The boundary conditions of the model included separation between all the contact surfaces of the components. The mesh used in the simulation was refined in the areas of greatest interest, particularly along the faces and edges of the pin and in the regions adjacent to the hole, guaranteeing a more precise analysis in these critical areas. The polymeric components, the pin and PLA parts, were considered as a single part in the simulation, with no separation at the interface between them or between the layers formed by the material-extrusion-based AM, simplifying the model.

Mechanical properties of the PLA were obtained through experimentation following the optimisation of printing parameters, ensuring that the material behaviour in the simulation accurately reflects the joint performance. The properties of both materials for the correct characterisation of their behaviour in the software are listed in

Table 5. Properties obtained experimentally for the AA5754 and PLA.

Material	Density [g/cm3]	Young’s Modulus [GPa]	Yield Strength [MPa]
AA5754-H111	2.66 [34]	68	110
PLA	1.24 [23]	37.32	70

.

During the simulations, the CL geometry proved to be very ineffective at blocking the movement between components perpendicularly to the displacement direction since this requirement was ensured by the existence of the conical portion in the pin. This pin geometry, CL, exhibited a pronounced behaviour of detaching from the hole where it was fitted, and consequently, separation between components was found to occur. This behaviour and the ineffectiveness of the interlocking mechanism at blocking the movement are shown in Figure 10 with the analysis of the displacement in the direction perpendicular to the imposed displacement.

The simulation provided significant insights into the behaviour of different pin geometries regarding movement restriction between components. It was observed that the conical portion of the pin played a crucial role in effectively constraining motion. In contrast, the CL geometry exhibited a tendency to detach from the insertion hole, leading to the separation of components. On the other hand, the CN and CL50-CN50 geometries proved to be the most effective at preventing movement, ensuring a more robust interlocking mechanism between the materials. These findings offer a clearer understanding of how pin geometry can influence the efficiency of mechanical interlocking systems.

5. Results

Figure 11 shows the macrographs of the longitudinal section of the dissimilar Al-PLA joints, and looking at the figure, it can be seen that the precise and uniform deposition of the polymer was achieved. This good deposition was repeated for all the different geometries of the anchoring mechanism. Although the samples exhibit small voids on the internal structure, specifically, in the pin, they are not relevant to the overall integrity of the joint, and it can also be seen that a defect-free pin and PLA part interface is presented, which is essential to guaranteeing good performance of the joining mechanism.

Figure 12 shows the average maximum load values for the dissimilar Al-PLA joints after the printing parameters optimisation. The fracture mode of the joints is also illustrated in the figure. Starting by analysing the most conventional method, which connects the two parts without mechanical interlocking, it can be seen that this joint offers a very low resistance, with an average maximum load of 0.29 kN, highlighting the need for a macro-mechanical interlocking mechanism. It is possible to observe that the introduction of a macro-mechanical interlocking mechanism increases the maximum average load to 0.56 kN. However, this pin design, CL, has a tendency to be projected out of the hole. In turn, the CN pin restricts the movement more efficiently, which increases the strength of the joint to 0.85 kN. From the image of the fracture of this joint, it can be observed that it failed because the pin started to be cut by the metal substrate due to the joint geometry.

Based on numerical simulation results, to optimise and further improve the mechanical performance of the joint, a mixed pin design was implemented, which was 50% cylindrical and 50% conical, called CL50-CN50. The CL50-CN50 mechanical interlocking achieved the highest average maximum load of 1.36 kN, representing an overall improvement of 368.97% compared to the joint without mechanical interlocking. These results highlight the critical impact of pin geometry on the mechanical strength of dissimilar material joints and the effectiveness of combined mechanical interlocking to maximise the joint performance.

Figure 13 shows the typical behaviour of the joints with CL50-CN50 pin design in tensile-shear testing. Although this load–displacement curve corresponds to the joint with the best mechanical behaviour, all the joints presented the same load–displacement evolution. The figure shows that different loading stages can be identified via the tensile-shear test.

(a)

The initial part of the curve, where the load begins to be applied, shows a sudden drop in joint resistance. This moment was ensured by the paint, which acted as an adhesive and at that moment lost its effect, resulting in the detachment between the Al and PLA parts of the specimen. This reinforces that the paint is solely to ensure proper deposition of the polymer on the metallic substrate, not providing resistance to the joint.

(b)

Here, the pin tends to rotate out of the plane parallel to the load direction, leading to a reduction in load after reaching the maximum load that the specimen can withstand before failing.

(c)

The rotation of the pin led to a reduction in load, corresponding to the tearing/cutting of the conical portion of the anchoring pin near the surface of the metal. The load is then supported by the pin itself, specifically the conical portion, which is braced against the internal wall of the hole in the metallic substrate. This moment, where the load starts to decrease, indicates the onset of failure of the joint. At this stage, the deformation of the PLA part becomes more visible as it cracks and propagates through the thickness of the part.

(d)

As deformation continued, the parts slowly separated from each other, decreasing the load supported until the total loss of mechanical resistance occurred due to the fracture of the PLA part.

In general, according to the literature, and summarised in

Table 6. Summary of the literature results on hybrid structures produced by FFF.

Reference	Method	Geometry	Materials	Failure
Present Research	Overlapping joint with mechanical interlocking between the additive part and the pre-drilled Al sheet	Pin connection, cylindrical and conical geometry, with 10 mm diameter contact area; PLA and Al sheets with 40 × 20 × 2 mm3	PLA and AA5754-H111	Fracture of the PLA part near the joint area
Ozlati et al., 2019 [20]	Overlapping joint with mechanical interlocking between the additive part and the pre-drilled Al sheet	Cylindrical pin connection, with a 13 mm diameter; PP and Al sheets with 75 × 25 × 1 mm3	Polypropylene (PP) and AA5083	Fracture of the joint interface, between additive parts
Falck et al., 2018 [21]	Single-lap joint configuration by deposition of the PA6/CF-PA6 on the Al	Without a mechanical interlocking, Al sheet with 101.6 × 25.5 × 2 mm3 with an overlap area of 12.5 × 25.5 mm2	Polyamide-6 and carbon-fibre-reinforced polyamide-6 (PA6/CF-PA6) and AA2024-T3	Fracture of the joint within the fibre-reinforced printed layer
Belei et al., 2022 [22]	Single-lap joint configuration between sandblasted Ti-6Al-4V substrate and PA-CF	Without a mechanical interlocking, Ti-6Al-4V and PA/CF sheets with 100 × 25.4 × 0.6 mm3 and 100 × 25.4 × 2.2 mm3, respectively	PA/CF and Ti-6Al-4V	Adhesive failure between metal and coating layer
Alhmoudi et al. 2022 [35]	Single-lap joint configuration by depositing PLA on textured Al	Without a mechanical interlocking, Al and ABS sheets with 2 mm thickness and 100 × 25 mm2 and 100 × 30 mm2, respectively	PLA and AA5052	Fracture of the PLA part near the joint area
Oliveira et al., 2023 [36]	Single-lap joint configuration by depositing PC on the AlSi10Mg substrate with structures printed by laser powder bed fusion	Without a mechanical interlocking, AlSi10Mg and PC sheets with 56.5 × 24.8 mm2 and, 1.6 mm and 4 mm thickness, respectively, with an overlap area of 12.7 × 24.8 mm2	Polycarbonate (PC) and AlSi10Mg	Fracture and displacement of the PC plate in the joint area

, the types of failures observed in these hybrid joints can be categorised into interface failures, cohesive failures within the printed layers, and adhesive failures between the materials. Interface failures typically occur when there is insufficient connection between the different materials, despite the existence of an interlocking mechanism, leading to fracture at the joint interface. Cohesive failures happen within the printed material itself, where the internal structure of the additive material cannot withstand the applied load, causing it to fracture in areas close to the joint. Adhesive failures are seen when the connection between the metal substrate and the coating or printed layer is not strong enough, and the joint fails due to the separation of the materials without fracture of either.

The use of a pin or other mechanical interlocking can significantly influence the failure mode. Without mechanical interlocking, as seen in some studies, the joints are more likely to fail between the interface of the materials, known as adhesive failure [22]. This contrasts with joints that employ mechanical interlocking, such as pins, which tend to enhance the load-bearing capacity of the joint by distributing stress more effectively. However, even with interlocking, failure can still occur if the materials are not sufficiently compatible or if the mechanical design does not fully address the stress concentrations; usually, failure occurs due to a fracture in the polymer component near the interlocking region.

As described above, the connection is established with the addition of a pin (CL, CL50-CN50, CN). A main crack nucleates at the pin near the surface of the Al part, propagating towards the thickness of the PLA part, as illustrated in Figure 14 (I). Additionally, the rotation induces a radial crack at the top of the pin when it has a conical-shaped portion, as shown in Figure 14 (II), separating the material volume of the conical portion from the rest of the pin. The continuous rotation of the pin causes these two cracking modes to occur simultaneously.

Therefore, analysing Figure 14 can help to understand the failure mode of the three geometries:

(a)

CL50-CN50—When the pin is a combination of both previous geometries, CL50-CN50, the failure mechanism is determined by which part fractures first. As previously mentioned, these two modes, (I) and (II), occur simultaneously, leading to failure either by a radial crack at the conical portion of the pin or by a crack at the PLA part.

(b)

CL—When the pin consists solely of a cylindrical part, as in the specific case of the CL design, due to the lack of movement restriction in the direction perpendicular to the load, ensured by the conical portion, the pin tends to “jump” and become attached to the hole, and simultaneously, the failure occurs due to the fracture of the PLA plate. So, in CL geometry, only the behaviour of (I) is observed.

(c)

CN—On the other hand, when the pin is completely conical, CN, failure occurs by cutting the entire conical portion of the pin before the PLA part fractures. Here, just the action noted in (II) happens; once the pin is totally conical, all the material is supported by the cavity of the hole and ends up being cut off due to the pulling direction and the sharp geometry of the Al part.

To better understand the result presented, several techniques used to obtain dissimilar metal–polymer-based joints were compared and summarised in Table 6, based on findings from the literature that match this study. This analysis allows us to comprehend the outcomes of previous research, giving us a strong foundation for interpreting current results.

When comparing the present study with the others listed in Table 6, it can be seen that the present study is promising, particularly due to the cohesive failure fracture mode, showing the strength of the mechanical connection, which exceeds the strength of the printed material itself. Thus, the mechanical interlocking configuration and the materials used in this study show good mechanical behaviour and prove to be effective in creating a robust joint.

6. Conclusions

This investigation successfully demonstrates the feasibility and benefits of using material-extrusion-based AM to create metal–polymer joints. This work presents a study that uses material-extrusion-based AM to optimise the production of dissimilar material joints between metal and polymer. The main conclusions are presented below:

• By optimising the printing parameters and joint geometries, the mechanical performance of these joints was significantly improved, underscoring the importance of optimising printing parameters to achieve a nearly 100% density at the interface between the pin and the PLA part. This optimisation is crucial for establishing a robust attachment, ensuring the integrity of the joint in the primary area responsible for its resistance, thus increasing the durability and efficacy of the joints.
• While the metal–polymer joints inherently present challenges in adhering dissimilar materials, the optimised CL50-CN50 Al-PLA joints exhibited a substantial increase in mechanical strength, with an improvement in the maximum average load of 368.97% compared to joints without interlocking. This improvement is attributed to the effective stress distribution and the reduction in voids achieved through precise and uniform polymer deposition.

The study highlights the importance of geometric considerations in the design of metal–polymer joints and suggests that further exploration of mixed CL-CN mechanical interlock designs could lead to even greater advancements in this field.

This concept of integrating materials through material-extrusion-based AM opens new possibilities for complex part design, reducing assembly complexity. Additionally, this joint concept offers a design that is both simple and aesthetically appealing, resulting in products that are not only functional but also visually captivating. This approach paves the way for innovative designs and applications in various industries.

These results provide a solid foundation for future experimental studies and potential industrial applications, demonstrating the versatility and effectiveness of material-extrusion-based AM in producing reliable dissimilar material joints. In future work, we expect the influence of thermal gradients on the performance of these joints and their ageing to be studied, providing further insights into the long-term durability and reliability of metal–polymer interfaces.