A Comprehensive Technical Review of the Friction Stir Welding of Metal-to-Polymer Hybrid Structures

Metal–polymer hybrid structures are becoming desirable due to their wide range of applications in the automotive, aerospace, biomedical and construction industries. Properties such as a light weight, high specific strength, and design flexibility along with the low manufacturing costs of metal–polymer hybrid structures make them widely attractive in several applications. One of the main challenges that hinders the widespread utilization of metal–polymer hybrid structures is the challenging dissimilar joining of metals to polymers. Friction stir welding (FSW) shows a promising potential in overcoming most of the issues and limitations faced in the conventional joining methods of such structures. Several works in the literature have explored the FSW of different metal-to-polymer combinations. In some of the works, the joints are examined based on processing parameter optimization, microstructural characteristics, and mechanical performances. It is, therefore, important to summarize the findings of these works as a means of providing a reference to researchers to facilitate further research on the utilization of FSW in joining metals to polymers. Thus, this work aims to present a comprehensive technical review on the FSW technique for joining metals to polymers by reviewing the reported literature findings on the impact of materials, tools, process parameters, and defects on the strength and microstructure of the produced joints. In addition, this work reviews and presents the latest practices aiming to enhance the metal–polymer joint quality that have been reported in the literature.


Introduction
The demand on hybrid metal-plastic joints is on the rise due to their ability to create lightweight yet strong structures, making them attractive for the automobile, aerospace, electronics, and biomedical industries. For instance, the shift towards environmentally friendly lightweight materials in the automotive industry has led to an increase in the adoption rate of plastics [1]. Plastics consist of large molecules called polymers. Polymers have a high specific strength, low thermal expansion, and excellent fatigue and fracture strength. Moreover, polymers are light and easily formable materials. These properties make polymers an attractive material for the automobile and aerospace industries [2]. With that comes the need to be able to join metals to polymers. Such an assembly would produce a part that combines both the lightweight property of polymers with the properties of metals, such as strength and ductility, all at once. To further explore metal-to-polymer joining, Section 1.1 will present the different metal-polymer joining techniques that have been discussed in the literature. Later on, Section 1.2 will introduce friction stir welding (FSW) in the context of metal-to-polymer joining.

Metal-to-Polymer Joining Techniques
As mentioned earlier, plastics are joined with metals to form hybrid, strong yet light structures. At present, there are several methods used to join plastics to metals, such as adhesive bonding, mechanical fastening, and direct joining techniques. This section will briefly present the aforementioned metal-to-polymer joining techniques and will highlight the pros and cons of each method.
Firstly, adhesive bonding is a traditional joining method that uses a polymeric adhesive material between the joining pieces to create a joint. The strength of the joint mainly depends on the intermolecular forces between the joining pieces and the adhesive material. Thus, surface treatment for the pieces pre-bonding is crucial to enhance the joint durability and strength. Adhesive bonding is used in vast applications due to its ability to offer lightweight joints along with uniform stress distribution when subjected to load [3]. Many researchers investigated adhesive joint strength [4][5][6][7][8]. Yet, adhesive bonds have many limitations. Among these limitations are their sensitivity to environmental factors, especially temperature and humidity. Additionally, the bonded joints formed tend to fail instantaneously rather than degrade with time [3].
Another simple joining technique is mechanical fasting. While there are many different processes and members used in mechanical fastening, research is focused on riveting, as it has proven to produce reliable joints for metal-to-polymer joining [9][10][11][12]. The typical case of riveting is placing the metallic part at the bottom because it is exposed to the highest deformation while the polymer is placed on top of the metal; a rivet is punched through the polymer into the metal creating the joint. However, the main limitations of this process are the increased structure weight due to the use of an external rivet and the increased stress concentrations around the holes created for the rivets [3].
In recent decades, there has been a steady rise toward the research and development of alternative joining techniques in response to the growing demand for metal-polymer hybrid structures in industry and the limitations of conventional joining methods. For instance, ultrasonic welding (USW) is a process by which high-frequency, low-amplitude waves, typically in the range of 10-250 µm, are applied to the joint area while in the solid state to break the oxide layer and create the joint [13][14][15]. USW stands out for its low cost and short process time. However, for metal-polymer bonding, the difference in material performance is the main cause of USW formed joint high sensitivity to the applied waves vibrations amplitude which deteriorates joint strength [15]. Another metal-polymer welding process is laser welding, in which the joint is subjected to a laser beam, initiating bubbles of the plastic part to spread and diffuse in the molten solid interface forming the bond between the metal and the polymer. Laser welding produces strong joints due to the chemical bond between oxide film and carbon atoms, as well as mechanical bonds created as a result of van der Waals forces. The main drawbacks of laser welding are the utilization of shielding gas and sophisticated processes [3,16]. Injection molded joining is another metal-polymer joining process that injects the polymer into the mold through a nozzle. After the mold is filled, a pressure is applied to press the melt and avoid shrinkage during solidification [17][18][19][20]. This process stands out for producing complex shapes at a high production rate. However, low joint strength and additive pre-treatment requirements are considered among the major disadvantages of injection molding [19]. Friction-assisted joining is another category of metal-polymer joining where a tool plunges the metallic part into the polymer surface and rotates at a prescribed speed for a specific time [21][22][23][24][25][26]. An example is friction lap welding (FLW), in which the joint is created by applying pressure and heat using a non-consumable cylindrical rotating tool. The metal sheet is placed on the polymer. The tool is pressed against the metal sheet to ensure the desired pressure and is then rotated along the welding direction [27]. The heat generated due to friction between the metal and the tool melts the area of the plastic sheet adjacent to the metal creating the joint with the pressed metal. FLW produces strong joints. However, it is limited to creating overlapping joints in addition to non-uniform heat distribution across the weld line [27]. Table 1 summarizes the advantages and disadvantages for each method. However, because of the complexity of the machinery and the high production costs of the methods proposed thus far, new revolutionary joining methods are still required. Industries and researchers alike are investigating new environmentally friendly methods for joining polymers to metals that produce strong joints while satisfying the market demand of low cost and energy. A new promising method is friction stir welding (FSW). FSW is one of the most recently developed welding processes. It attracts the interest of the scientific and the industrial communities alike [28]. It is defined as a solid-state joining process that uses a non-consumable rotating tool to join two adjacent parts. The tool consists of two parts: shoulder and pin. The pin is immersed into the abutting faces of the clamped parts until the shoulder touches the surface of the parts. Afterwards, the tool rotates and moves in the welding direction at a prescribed speed. Simultaneously, frictional heat is generated softening the base material creating a joint along the weld direction [29].
Several studies have focused on identifying the impact of the different types of tool shoulders and pin profiles on the microstructure and the mechanical properties of the joint. Mojtaba et al. showed that the weld quality is highly affected by the tool pin shape, unlike the welding process condition which is independent of the pin shape [69]. A similar conclusion was drawn by Kumar et al. when different shapes, sizes and materials of the tool pin and shoulder were reviewed [91].
Researchers have studied several properties of the weld joint. Microstructural investigations revealed that the stir zone consists of fine-grain structures that were created due to dynamic recrystallization [61,64]. Several studies reported a high joint strength for dissimilar materials using the FSW technique. Geng et al. achieved a joint strength of 75% of Al5052 when using the FSW process to join Al5052 to DP590 steel [63]. In a similar study, Saravana et al. joined Al6061 to Ti-6Al-4V and produced a joint with a strength of 87% of Al6061 [67].
Over the past decade, researchers have been aiming to establish hybrid metal-topolymer joints using FSW. Several promising joint strengths were reported in the literature. Dalwadi et al. reported a hybrid relative joint strength of 20% of PMMA welded between PMMA and AA6061 using FSW [79]. In a similar study, Patel et al. joined PC and AA6061 using FSW with a joint strength of 34% of PC [78]. Khodabakhshi et al. produced a dissimilar FSW joint with a joint efficiency of 50% relative to high-density polyethylene (HDPE) while joining HDPE and AA5059 [73]. Moreover, Haghshenas and Khodabakhshi provided a review paper addressing dissimilar joining of aluminum and polymer using FSW in terms of joint quality, soundness, material mixing, interfacial bonding, flow pattern, microstructure, mechanical properties, and thermo-mechanical modeling [92]. However, there is still a need for further investigations to understand the joining of metals to polymers using FSW. It is important, therefore, to summarize and discuss the findings of these works as a means of providing a reference to researchers of this field to facilitate further research on the utilization of FSW in joining metals to polymers. Thus, this work presents a comprehensive technical review on the FSW technique for joining metals to polymers by reviewing the reported literature. In addition, this work reviews and presents the latest practices reported in the literature aimed at enhancing the joint quality. Figure 1 presents the topics covered in this review paper.
Polymers 2023, 15, 220 5 of 26 these works as a means of providing a reference to researchers of this field to facilitate further research on the utilization of FSW in joining metals to polymers. Thus, this work presents a comprehensive technical review on the FSW technique for joining metals to polymers by reviewing the reported literature. In addition, this work reviews and presents the latest practices reported in the literature aimed at enhancing the joint quality. Figure 1 presents the topics covered in this review paper. To collect the data presented in this work, a systematic literature review was conducted through the SCOPUS database. The search keywords used were ("Friction Stir Welding" AND "Metal" AND "Polymer"). By searching for these keywords, 137 results were obtained. All papers with technical aspects related to this work were included. To be specific, works that reported findings on the impact of materials, tools, process parameters, and defects on the strength and microstructure of the produced joint were included. Other works that focused on other non-technical aspects were excluded.
The next section of this paper reviews the materials utilized in FSW, followed by a section reviewing the tools and process parameters. Thereafter, the joint quality and defects are discussed, and the joint mechanical and thermal properties are reviewed. Finally, the review paper is concluded with a summary, future outlook, and research gaps.

Materials
Aluminum alloys are the most utilized materials in FSW. However, other non-ferrous and ferrous alloys have also been investigated. Several researchers have been investigating ways of joining metals to polymers. Yet, this has been challenging due to numerous reasons. Firstly, metallic and polymeric materials have widely different surface energies, which has an impact on adhesive bonding at their interfaces. Secondly, metals and polymers have distinct structures: metals have crystalline structures with very high cohesive energy, whereas polymers contain long molecules of covalently bound carbon atoms with To collect the data presented in this work, a systematic literature review was conducted through the SCOPUS database. The search keywords used were ("Friction Stir Welding" AND "Metal" AND "Polymer"). By searching for these keywords, 137 results were obtained. All papers with technical aspects related to this work were included. To be specific, works that reported findings on the impact of materials, tools, process parameters, and defects on the strength and microstructure of the produced joint were included. Other works that focused on other non-technical aspects were excluded.
The next section of this paper reviews the materials utilized in FSW, followed by a section reviewing the tools and process parameters. Thereafter, the joint quality and defects are discussed, and the joint mechanical and thermal properties are reviewed. Finally, the review paper is concluded with a summary, future outlook, and research gaps.

Materials
Aluminum alloys are the most utilized materials in FSW. However, other non-ferrous and ferrous alloys have also been investigated. Several researchers have been investigating ways of joining metals to polymers. Yet, this has been challenging due to numerous reasons. Firstly, metallic and polymeric materials have widely different surface energies, which has an impact on adhesive bonding at their interfaces. Secondly, metals and polymers have distinct structures: metals have crystalline structures with very high cohesive energy, whereas polymers contain long molecules of covalently bound carbon atoms with weak secondary forces between them. Furthermore, metals contained in polymers tend to form circular clusters rather than mixing, resulting in limited metal solubility in polymers. The deterioration of polymers is particularly critical in this case, especially when metalpolymer joining is achieved via welding. The major source of deterioration is the metals' high hot-working temperatures, which are often higher than polymer decomposition temperatures. This can lead to oxidation, molecular weight loss, or polymeric molecule fracture [73]. Therefore, to join the work pieces successfully, it is important to choose the materials based on their thermal properties, where the difference between the metals' hot working temperature range and the polymers' decomposition onset temperature is minimal. Different joint configurations have been reported in the literature, as presented in Table 3. In a butt joint, the work pieces are placed adjacent to one another with the metal on the advancing side (AS) (with tool rotating direction). This is mainly because the harder material is more convenient to transport compared to the softer material when positioned in AS [77]. Likewise, the metal is placed on the upper side of a lap joint.

Tool Parameters
The most important part of the FSW process development is the tool geometry. Material flow is governed by the tool geometry, which, in turn, governs the traverse rate at which FSW may be performed [93]. The primary geometry features of FSW tools are the shoulder diameter and feature, the pin diameter and feature, and the pin shape and length [94]. The tool is mainly fabricated from steel and serves two main purposes: localized heating and material flow. The primary cause of the tool plunge heating up in early stages is the friction between the pin and the workpiece. In addition, deformation of the material causes some extra heating [93]. The tool is plunged into the workpiece until the shoulder of the tool contacts the workpiece surface. The most significant source of heat is friction between the shoulder and the workpiece. The shoulder also accommodates the volume of the heated material. The standard type of shoulder design is the concave shoulder. The simple, easily machined concavity design is fabricated by a small angle between the edge of the shoulder and the pin. This shoulder type needs the tool to be tilted by 2 • to 4 • from the normal of the welding line away from the direction of weld travel [95]. Shoulders may contain features that can be machined onto any tool shoulder profile. These features increase the amount of material flow and lead deformed material from the edge of the shoulder to the pin, thus eliminating tool tilting requirement [95]. Scrolled and grooved are reported shoulder features in the literature used to enhance the weld quality in hybrid metal-polymer joints [83,86]. A large shoulder diameter increases the frictional heat input, which enhances Polymers 2023, 15, 220 7 of 26 material flow, obtaining high quality joints. On the contrary, a small shoulder diameter results in an insufficient heat input and material flow, leading to a defect at the stir zone (SZ) [94]. Therefore, the size of the pin and shoulder is crucial in terms of heating.
Moreover, tool pin geometry affects the weld appearance and determines the consistency of the microstructure which are important characteristics of a high-quality weld [96]. The most common used pin geometries to successfully join metals to polymers and obtain relatively strong welds were cylindrical threaded [72,73,79,83,86] and tapered pin geometries [74][75][76][77][78]85,87]. However, Huang et al. concluded that using a tapered thread pin with triple facets can produce more pulsating action. Thus, the material transfer is improved compared to a thread-tapered pin and a weld strength of 20.6% of the polymer's strength is achieved using the friction stir lap welding (FSLW) technique [82]. The pin length and diameter likewise affect the weld appearance. Large pin dimensions may fail to induce the plastic flow, creating a rough surface and large valley-like defects at the joint area [77]. The welding tools that have been used in the literature during conventional FSW of metals to polymers are listed in Table 4.

Process Parameters
The FSW process depends on four main parameters: tool rotational speed, traverse welding speed, plunge depth, and tilt angle. Each of these parameters are independent, though each has a significant effect on the joint quality and weld appearance.

Plunge Depth
By rotating, tilting, and plunging the permanent probe on the material's surface, the softened and plasticized material is stirred and moved away from the advancing side towards the retreating side (RS) and solidifies under the forging force of the probe [72]. Several published papers recorded that a defect-free joint is obtained at low (usually 0.2-0.5 mm) plunge depths; by increasing the plunge depth, the area of the SZ decreases [72][73][74][75][76][77][78]82,85]. At high depth rates, flashes may form on the material surface due to the excessive axial forces moving the material away from the welding line [74].

Tilt Angle
The tilt angle affects the material flow of the weld. A low tilt angle may lead to tunneland crack-like defects in the welds [96]. When increasing the tilt angle, the forging force to plasticize the polymer increases and fills the defects better while bonding with metal fragments [74,87,96]. However, by increasing the tilt angle, the area of the stir zone (SZ) decreases; thus, high tilt angles may cause material overflow from the SZ [75].

Traverse Speed
The traverse speed contributes to the size of the metal fragments within the polymer. At high traverse speeds, both length and thickness of the metal fragments are small, and as the traverse speed decreases, the fragments sizes grow, resulting in a better mechanical interlocking [75,80]. This is because decreasing the traverse speed increases the preheat temperature and duration of the weld process, leading to a better material transfer and, hence, larger metal fragments [94].

Rotational Speed
Likewise, increasing the rotational speed enhances the heat input; thus, longer and wider metal fragments are generated within the SZ [72]. In general, the sizes of the fragments in the advanced side of the material are larger than that at the retreating side due to the counterclockwise rotation of the tool. However, high rotational speeds cause defects, including micro-voids [72], wider gaps [85], and tunnels [73], which deteriorate the joint strength. On the other hand, low rotational speeds may not be adequate to stir the material due to low heat input in addition to the presence of wormhole defects because of the insufficient material flow [73,85]. Therefore, process parameters must be optimized to avoid defects and obtain high-quality welds. Table 5 shows optimum process parameters reported in the literature for the FSW process between metals and polymers.

Surface Appearance
To better understand the role of the FSW process parameters on the joint between polymers and metals, assessing the weld appearance and examining the visual surfaces and cross-sections of the joint are required. Rahmat et al. studied the effect of plunge depth optimization on the joint quality. At a plunge depth of 0.1 mm, a valley-like structure was observed on the joint surface. However, by slightly increasing the depth to 0.2 mm, a defect-free joint was created, increasing the value further leading to the presence of residual material on the advancing side (AS) [77].
Gao et al. [80] and Shahmiri et al. [85] obtained rougher joint surfaces by increasing the tool rotational speed and lowering the traverse speed, respectively, for different materials. This is because such a combination increases the thermal input, which causes larger metal fragments to mix with the polymer, producing a rougher surface. Nonetheless, operating at overly high rotation speeds causes the polymer to overflow due to its large thermal expansion [87]. Huang et al. reported a case of overflow in an attempt to join AA6061 to PEEK by the FSLW process. It was observed that the overflow gradually decreases as the traverse speed increases [82].
Contrary to common practice, Derazkola and Elyasi [58] and Derazkola et al. [88] performed the FSLW process while placing the polymer at the top, facing the tool probe, and the metal at the bottom to increase the cooling rate after removing the tool. Both studies observed the presence of grey joint lines at different process parameters, as illustrated in Nonetheless, operating at overly high rotation speeds causes the polymer to overflow due to its large thermal expansion [87]. Huang et al. reported a case of overflow in an attempt to join AA6061 to PEEK by the FSLW process. It was observed that the overflow gradually decreases as the traverse speed increases [82].
Contrary to common practice, Derazkola and Elyasi [58] and Derazkola et al. [88] performed the FSLW process while placing the polymer at the top, facing the tool probe, and the metal at the bottom to increase the cooling rate after removing the tool. Both studies observed the presence of grey joint lines at different process parameters, as illustrated in Figure 2. The gray color indicates the mixing of the metal alloy and the polymer sheet at the SZ. The weld appearance and material overflow were smooth.

Microstructure and Internal Defects
An excellent joint surface appearance may be obtained using different parameters. However, the performance and strength of the joint vary due to the joint's microstructural difference in the SZ. One of the most common techniques used to observe the microstructure is scanning electron microscopy (SEM). The dominant joining mechanism is mechanical interlocking between the metal and the polymer at the SZ [73,75,79,80]. The severely deformed metal penetrates the softened polymer and leads to the formation of metal interlocked fragments. In general, metal fragments improve the mechanical performance of the joint. Figure 3 illustrates the mechanical interlocking between the metal and the polymer matrix, where the bright gray contrast resembles the metal, and the dark gray continuous matrix is the melted and resolidified polymer [73].

Microstructure and Internal Defects
An excellent joint surface appearance may be obtained using different parameters. However, the performance and strength of the joint vary due to the joint's microstructural difference in the SZ. One of the most common techniques used to observe the microstructure is scanning electron microscopy (SEM). The dominant joining mechanism is mechanical interlocking between the metal and the polymer at the SZ [73,75,79,80]. The severely deformed metal penetrates the softened polymer and leads to the formation of metal interlocked fragments. In general, metal fragments improve the mechanical performance of the joint. Figure 3 illustrates the mechanical interlocking between the metal and the polymer matrix, where the bright gray contrast resembles the metal, and the dark gray continuous matrix is the melted and resolidified polymer [73].
Heat input affects the size of the interaction layer that contributes to the adhesion bonding between the metal and the polymer. Derazkola and Simchi studied the effect of the process parameters on the thickness of the interaction layer in both advancing and retreating sides [74]. Aluminum fragments emerge and spread in the contact layer. Smaller aluminum fragments are distributed in the RS (labeled as "A") compared to the AS (labeled as "B"). Furthermore, uniform flow patterns, such as onion rings, tunnels, and kissing boundaries, are not observed. During the FSW process, low-temperature severe plastic deformation results in the creation of aluminum pieces in the SZ, which strengthen the joint area. Stretched aluminum ramus creates a wavy boundary at the Al-polymer contact, as shown in Figure 4c. This wavy border line functions as a mechanical interlock between the aluminum alloy and the polymer sheet. Figure 4d shows a magnified image of the contact between the aluminum ramus and the polymer matrix (labeled as "C"). It appears that a contact layer forms between the outside portion of the aluminum ramus and the polymer. The interaction layer thickness is larger at the AS due to a greater heat input. Likewise, the thickness increases with incrementing process parameters such as rotational velocity, tilt angle, and plunge depth, as shown in Figure 4e,f. However, as discussed in the previous section, an excessive generation of heat input (a high rotation speed and/or a low traverse speed) increases the amount and size of metal fragments at the SZ, and the more metal fragments, the less the contribution from the polymer towards the joint formation, leading to less mechanical interlocking, thus decrementing the joint strength [74]. Heat input affects the size of the interaction layer that contributes to the adhesion bonding between the metal and the polymer. Derazkola and Simchi studied the effect of the process parameters on the thickness of the interaction layer in both advancing and retreating sides [74]. Aluminum fragments emerge and spread in the contact layer. Smaller aluminum fragments are distributed in the RS (labeled as "A") compared to the AS (labeled as "B"). Furthermore, uniform flow patterns, such as onion rings, tunnels, and kissing boundaries, are not observed. During the FSW process, low-temperature severe plastic deformation results in the creation of aluminum pieces in the SZ, which strengthen the joint area. Stretched aluminum ramus creates a wavy boundary at the Al-polymer contact, as shown in Figure 4c. This wavy border line functions as a mechanical interlock between the aluminum alloy and the polymer sheet. Figure 4d shows a magnified image where bright gray contrast is Aluminum layers, dark gray matrix is melted and re-solidified polymer, dark spots through polymer matrix are shrinkage voids, and mixed gray-white contrast at interface presents the solid state modified polymer via FSW. Reprinted with permission from Ref. [73]. 2014, F. Khodabakhshi et al. input. Likewise, the thickness increases with incrementing process parameters such as rotational velocity, tilt angle, and plunge depth, as shown in Figure 4e,f. However, as discussed in the previous section, an excessive generation of heat input (a high rotation speed and/or a low traverse speed) increases the amount and size of metal fragments at the SZ, and the more metal fragments, the less the contribution from the polymer towards the joint formation, leading to less mechanical interlocking, thus decrementing the joint strength [74]. Shahmiri et al. reported the formation of a gap between the interaction layer and the polymer matrix in the stir zone [85]. The gap was formed due to the significant differences between the metal's and the polymer's coefficients of thermal expansion. During the cooling stage, polymers shrink faster than metals, leaving a gap at the interaction layer, as shown in Figure 5b-e. By comparing the size of the interaction layer in Figure 5d and Figure 5e, which were formed at rotation speeds of 800 and 1200 rpm, respectively, it can be noticed that the thickness of the interaction layer in Figure 5e is larger. This confirms the previous results obtained in [74]. However, the gap formed in Figure 5e is wider than that of Figure 5d; thus, the gap width increases with more generated heat input. This showed that the chemical adhesive bonding at the interaction layer between the polymer and the metal is weak and the mechanical interlocking is the dominant bond. Shahmiri et al. reported the formation of a gap between the interaction layer and the polymer matrix in the stir zone [85]. The gap was formed due to the significant differences between the metal's and the polymer's coefficients of thermal expansion. During the cooling stage, polymers shrink faster than metals, leaving a gap at the interaction layer, as shown in Figure 5b-e. By comparing the size of the interaction layer in Figure 5d,e, which were formed at rotation speeds of 800 and 1200 rpm, respectively, it can be noticed that the thickness of the interaction layer in Figure 5e is larger. This confirms the previous results obtained in [74]. However, the gap formed in Figure 5e is wider than that of Figure 5d; thus, the gap width increases with more generated heat input. This showed that the chemical adhesive bonding at the interaction layer between the polymer and the metal is weak and the mechanical interlocking is the dominant bond.
Energy dispersive spectroscopy (EDS) examinations of the white contrast areas in Figure 6B,C show the presence of carbon on the polymer side. As a result, these regions may have some polymer attached to the aluminum alloy, providing additional evidence for bonding at the aluminum-polymer contact during the FSW process. This adds credibility to the theory that the bonding was accommodated by a connection between aluminum oxide and the polymer, which may have been aided by mechanical interlocking via polymer penetration into nano-scale holes, and perhaps molecular bonding, which is a mixed regime of Al-O-C-Mg components. It also showed that aluminum and magnesium oxidation occurred, which may have contributed to the strength of the created contact between the polymer and the metal [73]. Therefore, it is discovered that the FSW parameters had an impact on the chemical composition of the contact zone. At larger heat inputs, more oxygen is present in the interaction layer, indicating a more extensive breakdown of the polymer and in situ reactivity with the base metal to generate oxides. It is also concluded that any linking between the polymer and the surface aluminum oxide is either secondary or van der Waals bonding [74]. oxidation occurred, which may have contributed to the strength of the created contact between the polymer and the metal [73]. Therefore, it is discovered that the FSW parameters had an impact on the chemical composition of the contact zone. At larger heat inputs, more oxygen is present in the interaction layer, indicating a more extensive breakdown of the polymer and in situ reactivity with the base metal to generate oxides. It is also concluded that any linking between the polymer and the surface aluminum oxide is either secondary or van der Waals bonding [74].  [73,[75][76][77]80]. Voids appear as dark spots in the polymer matrix, as shown in Figure 7. The main cause of void formation is the incompatible thermal deformation of dissimilar materials [87]. Polymers lose their molecular weight as a result of melting and re-solidification. This phenomenon intensifies their shrinkage properties, creating voids [75].  [73,[75][76][77]80]. Voids appear as dark spots in the polymer matrix, as shown in Figure 7. The main cause of void formation is the incompatible thermal deformation of dissimilar materials [87]. Polymers lose their molecular weight as a result of melting and re-solidification. This phenomenon intensifies their shrinkage properties, creating voids [75].
Bubbles and pores are also common defects found at the interface between the metal and the polymer. Their generation indicates the presence of structural water and/or trapped air in the polymer [82]. During the FSW process, the melted polymer absorbs humid air from the surroundings. Due to high temperature, the moisture expands as it turns to vapor, creating the bubbles [75]. Although internal defects deteriorate the joint strength, it must be noted that the causes of the mentioned defects are not yet conclusive. However, internal defects may be evaded by optimizing the heat input in the SZ area, choosing the right weld tools, and operating at reasonable process parameters, resulting in a defect-free joint with decent mechanical performance [74,75].
Under the thermal cycle and mechanical stirring of the FSW tool, it is important to analyze the grain structure and the changes in properties of the metal. Shahmiri et al. observed the grain structure of the joint at different areas, as shown in Figure 8 [85]. The base metal has an elongated grain structure due to the rolling manufacturing process. The grain structure remains unaffected in the HAZ. In the TMAZ, only partial recrystallization occurs, as the temperature is not adequate to induce dynamic recrystallization. Moreover, the grains were deformed and extended towards the direction of stirring action of the rotating pin. In the SZ, the metal was exposed to high strain and elevated temperatures. As it can be seen in Figure 8c, the extended grains were renovated into very fine equiaxed grains in the SZ because of dynamic recovery/recrystallization. Similar results were achieved by Huang et al., who reported the microstructure of the aluminum anchor in the SZ to be equiaxed, with an average grain size of approximately 35 µm [82]. MirHashemi et al. studied the effects of rotational and traverse speeds on the grain structure of aluminum fragments inside the SZ for dissimilar LDPE-AA7075 weldment, as illustrated in Figure 9 [72]. As it can be found, by increasing the rotation speed and/or decreasing the traverse speed, the mixing action of the FSW treatment was increased, which led to the formation of more aluminum fragments inside the polymer matrix with a more refined and recrystallized grain structure. Bubbles and pores are also common defects found at the interface between the metal and the polymer. Their generation indicates the presence of structural water and/or trapped air in the polymer [82]. During the FSW process, the melted polymer absorbs humid air from the surroundings. Due to high temperature, the moisture expands as it turns to vapor, creating the bubbles [75]. Although internal defects deteriorate the joint strength, it must be noted that the causes of the mentioned defects are not yet conclusive. However, internal defects may be evaded by optimizing the heat input in the SZ area, choosing the right weld tools, and operating at reasonable process parameters, resulting in a defect-free joint with decent mechanical performance [74,75].
Under the thermal cycle and mechanical stirring of the FSW tool, it is important to analyze the grain structure and the changes in properties of the metal. Shahmiri et al. observed the grain structure of the joint at different areas, as shown in Figure 8 [85]. The base metal has an elongated grain structure due to the rolling manufacturing process. The grain structure remains unaffected in the HAZ. In the TMAZ, only partial recrystallization occurs, as the temperature is not adequate to induce dynamic recrystallization. Moreover, the grains were deformed and extended towards the direction of stirring action of the rotating pin. In the SZ, the metal was exposed to high strain and elevated temperatures. SZ to be equiaxed, with an average grain size of approximately 35 µm [82]. MirHashemi et al. studied the effects of rotational and traverse speeds on the grain structure of aluminum fragments inside the SZ for dissimilar LDPE-AA7075 weldment, as illustrated in Figure 9 [72]. As it can be found, by increasing the rotation speed and/or decreasing the traverse speed, the mixing action of the FSW treatment was increased, which led to the formation of more aluminum fragments inside the polymer matrix with a more refined and recrystallized grain structure.

Tensile Properties
Various studies were aimed at understanding the effects of different process parameter variations on the joint strength [74][75][76]82,83,85,86]. Table 6 illustrates the tensile strength for metal-polymer hybrid joints at different process parameters obtained from the literature [78]. The highest tensile strength recorded was at 14.9 MPa for specimen E at 500 rpm and 40 mm/min. By comparing specimen E with K, it is observed that a slight change in the feed rate severely deteriorated the tensile strength. A similar observation is made for the rotational speed by comparing specimens E and T. Hence, it can be concluded that FSW is a parameter-sensitive technique. Various papers in the literature concluded that the joint strength increases with rotational speed and then decreases after reaching a certain point [75,78,79].
The tensile strength deteriorates at high rotational speeds because of the excessive heat input from friction between the material and the tool, which results in defects, as discussed in previous sections. Likewise, the tensile strength increases to reach a peak and then decreases as the traverse speed increases [73,85]. Additionally, the traverse speeds are usually set at relatively low rates (30-100 mm min −1 ) to provide sufficient time for stirring. When increasing the tilt angle within the range of 0° to 2°, the intermixing of dissimilar materials was improved. The tool plunge depth optimization in the range of 0.1 to 0.4 mm possessed an optimal effect on the joint strength [88]. However, increasing the tool plunge depth and tilt angle beyond such ranges forces more material to be ejected from the SZ. This is due to the hefty heat generated that causes material overflow and surface defects, such as flashes, leading to a lower joint strength [74]. Moreover, increasing the tilt angle along with decreasing the plunge depth leads to intensifying air bubble

Tensile Properties
Various studies were aimed at understanding the effects of different process parameter variations on the joint strength [74][75][76]82,83,85,86]. Table 6 illustrates the tensile strength for metal-polymer hybrid joints at different process parameters obtained from the literature [78]. The highest tensile strength recorded was at 14.9 MPa for specimen E at 500 rpm and 40 mm/min. By comparing specimen E with K, it is observed that a slight change in the feed rate severely deteriorated the tensile strength. A similar observation is made for the rotational speed by comparing specimens E and T. Hence, it can be concluded that FSW is a parameter-sensitive technique. Various papers in the literature concluded that the joint strength increases with rotational speed and then decreases after reaching a certain point [75,78,79]. The tensile strength deteriorates at high rotational speeds because of the excessive heat input from friction between the material and the tool, which results in defects, as discussed in previous sections. Likewise, the tensile strength increases to reach a peak and then decreases as the traverse speed increases [73,85]. Additionally, the traverse speeds are usually set at relatively low rates (30-100 mm min −1 ) to provide sufficient time for stirring. When increasing the tilt angle within the range of 0 • to 2 • , the intermixing of dissimilar materials was improved. The tool plunge depth optimization in the range of 0.1 to 0.4 mm possessed an optimal effect on the joint strength [88]. However, increasing the tool plunge depth and tilt angle beyond such ranges forces more material to be ejected from the SZ. This is due to the hefty heat generated that causes material overflow and surface defects, such as flashes, leading to a lower joint strength [74]. Moreover, increasing the tilt angle along with decreasing the plunge depth leads to intensifying air bubble formation within the solidified polymer. This can depreciate the mixed zone characteristics and quality [88].
To enhance the joint quality, eliminate weld defects, and increase weld strength, several researchers introduced new procedures to the FSW process for welding dissimilar materials. For example, Wang et al. introduced groove micro-textures through a laser ablation pre-treatment on the aluminum alloy surface prior to joining with PA6 using FSW [81]. For joints with V-shaped grooved aluminum plates, the bonding mechanism was attributed to the large mechanical interlocking from the geometric grooves in the micro-textures and the development of a C-O-Al bond at the interfaces. Upon fracture, the PA6 plate underwent severe plastic deformation and subsequently fractured with a maximum shear-tensile load of 1194 N, which exceeded the yield strength of the PA6 base material with a cross-sectional area of 20 × 1.5 mm 2 . MirHashemi et al. placed silicon carbide (SiC) nanoparticles on the dissimilar FSW of low-density polyethylene (LDPE) and AA7075 aluminum [72]. The primary purpose was strengthening the polymer side during the joining process by pre-placing the particle in the polymer side which enhanced the dissimilar intermixing process and the material flow profile. The dissimilar joining strength ratio was significantly enhanced up to around 70% of the LDPE strength.
A similar approach was applied by Derazkola and Simchi, in which alumina nanoparticles were fed in the joining line during the process of obtaining AA6062/PMMA joints. However, there was no need to preplace the particles at the welding interface by machining a cavity. The process is introduced as friction fed stir welding (FFSW). The strength of the joint using conventional FSW was around 49 MPa, which is 64% of PMMA's tensile strength. The addition of Al 2 O 3 nanoparticles increased the tensile strength to 61 MPa (∼30% improvement) [76]. Li et al. developed a new technique named top thermic friction stir lap welding (TT-FSLW) [87]. Based on thermal tensioning, two heating strips of different resistance are set up in parallel on both sides of the weld (see Figure 10a) to increase the compressive plastic strain during the FSLW process. This is to minimize the stress during the process and residual stress after welding. Figure 10b illustrates the TT-FSLW joint strength at 400 and 500 rpm, which are 37.0 MPa and 59.9 MPa, respectively. These values present an increase of 106.7% and 52.4% compared with 17.9 MPa and 39.3 MPa of the FSLW joint at 500 and 700 rpm, respectively. Moreover, the effect of rotation velocity is obvious, which increases to reach a peak at 39.3 MPa for FSLW, then decreases at higher speeds, which agrees with the results obtained from the literature [75,78,79].

Microhardness
In order to test the hardness of the welded joints, the hybrid joints are indented at different regions. On the polymer side, researchers reported the lowest hardness value obtained for melted-resolidified polymers at the SZ area [72,73,75,82,85], presumably due to the loss of molecular weight of the polymer and the reduction of crystallinity. However, the average hardness value of the SZ is greater than that of the base polymer because of the existence of embedded metal fragments at the metal-polymer interface [73,82]. The metal pieces had a fine-grain structure with high hardness, as they were exposed to high strains before being cut using the FSW tool. Thus, process parameters such as tool rotational and traverse speeds, that contribute to the size and number of metal pieces at the SZ will impact the overall hardness [74]. On the metal side of the joint, the hardness decreases at the SZ, TMAZ, and HAZ areas due to the recovery process [85]. During the FSW process, the metal softens and loses its pre-cold-worked strength after recrystallization, due to the applied heat input during joining. Moreover, the hardness at the SZ is higher than that at the HAZ, due to the small size of grains at the SZ resulting from the Hall-Petch effect [85]. Huang et al. and Shahmiri et al. measured the microhardness for different zone areas using the Vickers test [82,85]. Both studies recorded the minimum hardness at the HAZ on the metal side. This reduction in hardness was due to the decrease in dislocation density and the growth of sub-grains during the recovery process, as well as partial recrystallization. However, on the polymer side, the HAZ hardness remained unchanged. This is due to the polymer's low thermal conductivity. Moreover, increasing the tool rotational speed reduced the weld hardness, regardless of the zone area, as seen in Figure 11. On the contrary, Figure 12 shows that increasing the traverse speed increases the weld hardness. The hardness value of the metal fragments increases with the increase in the welding speed, which may be advantageous to the tensile bond properties of the hybrid joint [82]. In general, the effect of the process parameters on hybrid metal-to-polymer joint hardness is inversely proportional. High heat input (high rotational speed, tilt angle, and plunge depth) softens the metal and decomposes the polymer.

Microhardness
In order to test the hardness of the welded joints, the hybrid joints are indented at different regions. On the polymer side, researchers reported the lowest hardness value obtained for melted-resolidified polymers at the SZ area [72,73,75,82,85], presumably due to the loss of molecular weight of the polymer and the reduction of crystallinity. However, the average hardness value of the SZ is greater than that of the base polymer because of the existence of embedded metal fragments at the metal-polymer interface [73,82]. The metal pieces had a fine-grain structure with high hardness, as they were exposed to high strains before being cut using the FSW tool. Thus, process parameters such as tool rotational and traverse speeds, that contribute to the size and number of metal pieces at the SZ will impact the overall hardness [74]. On the metal side of the joint, the hardness decreases at the SZ, TMAZ, and HAZ areas due to the recovery process [85]. During the FSW process, the metal softens and loses its pre-cold-worked strength after recrystallization, due to the applied heat input during joining. Moreover, the hardness at the SZ is higher than that at the HAZ, due to the small size of grains at the SZ resulting from the Hall-Petch effect [85]. Huang et al. and Shahmiri et al. measured the microhardness for different zone areas using the Vickers test [82,85]. Both studies recorded the minimum hardness at the HAZ on the metal side. This reduction in hardness was due to the decrease in dislocation density and the growth of sub-grains during the recovery process, as well as partial recrystallization. However, on the polymer side, the HAZ hardness remained unchanged. This is due to the polymer's low thermal conductivity. Moreover, increasing the tool rotational speed reduced the weld hardness, regardless of the zone area, as seen in Figure 11. On the contrary, Figure 12 shows that increasing the traverse speed increases the weld hardness. The hardness value of the metal fragments increases with the increase in the welding speed, which may be advantageous to the tensile bond properties of the hybrid joint [82]. In general, the effect of the process parameters on hybrid metal-to-polymer joint hardness is inversely proportional. High heat input (high rotational speed, tilt angle, and plunge depth) softens the metal and decomposes the polymer.

Thermal Studies
Experimental and simulation results have shown that friction produces more than 90% of the heat generated during the FSW process [76]. The welding temperature in the FSW process affects the material flow and the welding stress [87]. Thus, all FSW parameters that influence the heat input to weld the specimens should be optimized to provide the necessary heat input. Thus, it is important to examine the thermal profile of the joint forming. Figure 13a shows the trend of the joint area temperature throughout the process. The temperature of the joint area increases rapidly to reach a peak above the polymer's decomposing temperature for a short period of time. Subsequently, the cooling stage begins at a slow rate due to the low heat transfer coefficient of the polymer; hence, the temperature fields remain high at the weld line [75]. The peak temperature always exceeds the melting point and may reach a higher value than the thermal decomposition point of the polymer, resulting in local chemical and physical changes to the material [74][75][76]. However, the peak temperature steady-state time is relatively short. Thus, complete polymer degradation is unlikely to occur.
Moreover, increasing the traverse speed reduces the peak temperature of the FSW process and enhances the cooling rate that contributes to better joint properties. Figure  13b illustrates the change of peak temperature with rotational speed. Increasing the rotational speed generates more heat, resulting in higher elevated temperatures. Similar behavior was reported for other FSW process parameters, such as tilt angle and plunge depth [74]. Dong et al. analyzed the thermal characteristics of CF-PEEK, as illustrated in Figure 14 [84]. The sample weight of the PEEK remained constant until the thermal degradation point of 570 °C. Afterwards, the polymer lost a significant amount of its molecular weight, which deteriorated its properties. Moreover, the interfacial peak temperature

Thermal Studies
Experimental and simulation results have shown that friction produces more than 90% of the heat generated during the FSW process [76]. The welding temperature in the FSW process affects the material flow and the welding stress [87]. Thus, all FSW parameters that influence the heat input to weld the specimens should be optimized to provide the necessary heat input. Thus, it is important to examine the thermal profile of the joint forming. Figure 13a shows the trend of the joint area temperature throughout the process. The temperature of the joint area increases rapidly to reach a peak above the polymer's decomposing temperature for a short period of time. Subsequently, the cooling stage begins at a slow rate due to the low heat transfer coefficient of the polymer; hence, the temperature fields remain high at the weld line [75]. The peak temperature always exceeds the melting point and may reach a higher value than the thermal decomposition point of the polymer, resulting in local chemical and physical changes to the material [74][75][76]. However, the peak temperature steady-state time is relatively short. Thus, complete polymer degradation is unlikely to occur.
Moreover, increasing the traverse speed reduces the peak temperature of the FSW process and enhances the cooling rate that contributes to better joint properties. Figure  13b illustrates the change of peak temperature with rotational speed. Increasing the rotational speed generates more heat, resulting in higher elevated temperatures. Similar behavior was reported for other FSW process parameters, such as tilt angle and plunge depth [74]. Dong et al. analyzed the thermal characteristics of CF-PEEK, as illustrated in Figure 14 [84]. The sample weight of the PEEK remained constant until the thermal degradation point of 570 °C. Afterwards, the polymer lost a significant amount of its molecular weight, which deteriorated its properties. Moreover, the interfacial peak temperature

Thermal Studies
Experimental and simulation results have shown that friction produces more than 90% of the heat generated during the FSW process [76]. The welding temperature in the FSW process affects the material flow and the welding stress [87]. Thus, all FSW parameters that influence the heat input to weld the specimens should be optimized to provide the necessary heat input. Thus, it is important to examine the thermal profile of the joint forming. Figure 13a shows the trend of the joint area temperature throughout the process. The temperature of the joint area increases rapidly to reach a peak above the polymer's decomposing temperature for a short period of time. Subsequently, the cooling stage begins at a slow rate due to the low heat transfer coefficient of the polymer; hence, the temperature fields remain high at the weld line [75]. The peak temperature always exceeds the melting point and may reach a higher value than the thermal decomposition point of the polymer, resulting in local chemical and physical changes to the material [74][75][76]. However, the peak temperature steady-state time is relatively short. Thus, complete polymer degradation is unlikely to occur. of aluminum-polymer joints is estimated to be around 0.5-0.6 of the melting point of aluminum alloys. Therefore, the risk of thermal degradation may be eliminated by choosing materials that are thermally compatible, where around half the metal temperature is lower than the thermal decomposition temperature of the polymer.  At present, the welding temperature at the interface is difficult to be obtained experimentally. Thus, a numerical simulation method may be used to predict the thermal profile of the joint. Li et al. simulated the temperature distribution for FSLW and TT-FSLW to understand the TT-FLSW process of the aluminum alloy and polymer hybrid weld, as shown in Figure 15 [87]. At 700 rpm for FSLW and 500 rpm for TT-FSLW, the center temperatures reached 457.1 and 448.6 °C, respectively. FSLW at high rotational speed and TT-FSLW at low rotational speed both produced about the same welding temperature. The high temperature zone had a H-shaped distribution, and the parallel top-thermic regions reduced the temperature gradient surrounding the welding heat source. During TT-FSLW, the weld center experienced both preheating and post-heating impacts. As a result, when compared to FSLW, TT-FSLW had lower heating and cooling rates. Meanwhile, the preheating and post-heating temperatures in the weld center were both below 200 °C due to the heat conduction distance between the resistance heating strips and the weld for TT-FSLW. Moreover, increasing the traverse speed reduces the peak temperature of the FSW process and enhances the cooling rate that contributes to better joint properties. Figure 13b illustrates the change of peak temperature with rotational speed. Increasing the rotational speed generates more heat, resulting in higher elevated temperatures. Similar behavior was reported for other FSW process parameters, such as tilt angle and plunge depth [74]. Dong et al. analyzed the thermal characteristics of CF-PEEK, as illustrated in Figure 14 [84]. The sample weight of the PEEK remained constant until the thermal degradation point of 570 • C. Afterwards, the polymer lost a significant amount of its molecular weight, which deteriorated its properties. Moreover, the interfacial peak temperature of aluminumpolymer joints is estimated to be around 0.5-0.6 of the melting point of aluminum alloys. Therefore, the risk of thermal degradation may be eliminated by choosing materials that are thermally compatible, where around half the metal temperature is lower than the thermal decomposition temperature of the polymer. of aluminum-polymer joints is estimated to be around 0.5-0.6 of the melting point of aluminum alloys. Therefore, the risk of thermal degradation may be eliminated by choosing materials that are thermally compatible, where around half the metal temperature is lower than the thermal decomposition temperature of the polymer.  At present, the welding temperature at the interface is difficult to be obtained experimentally. Thus, a numerical simulation method may be used to predict the thermal profile of the joint. Li et al. simulated the temperature distribution for FSLW and TT-FSLW to understand the TT-FLSW process of the aluminum alloy and polymer hybrid weld, as shown in Figure 15 [87]. At 700 rpm for FSLW and 500 rpm for TT-FSLW, the center temperatures reached 457.1 and 448.6 °C, respectively. FSLW at high rotational speed and TT-FSLW at low rotational speed both produced about the same welding temperature. The high temperature zone had a H-shaped distribution, and the parallel top-thermic regions reduced the temperature gradient surrounding the welding heat source. During TT- At present, the welding temperature at the interface is difficult to be obtained experimentally. Thus, a numerical simulation method may be used to predict the thermal profile of the joint. Li et al. simulated the temperature distribution for FSLW and TT-FSLW to understand the TT-FLSW process of the aluminum alloy and polymer hybrid weld, as shown in Figure 15 [87]. At 700 rpm for FSLW and 500 rpm for TT-FSLW, the center temperatures reached 457.1 and 448.6 • C, respectively. FSLW at high rotational speed and TT-FSLW at low rotational speed both produced about the same welding temperature. The high temperature zone had a H-shaped distribution, and the parallel top-thermic regions reduced the temperature gradient surrounding the welding heat source. During TT-FSLW, the weld center experienced both preheating and post-heating impacts. As a result, when compared to FSLW, TT-FSLW had lower heating and cooling rates. Meanwhile, the preheating and post-heating temperatures in the weld center were both below 200 • C due to the heat conduction distance between the resistance heating strips and the weld for TT-FSLW.

Summary
In this work, the friction stir welding (FSW) process for solid-state dissimilar joining of hybrid metal-polymer structures was reviewed and discussed. This was achieved by reviewing different works in the literature that have studied the effects of the different materials utilized in the process, the used tool types of the pin and shoulder profile, and the different process parameters on the joint's quality, strength, hardness, microstructure, and thermal profile. It can be concluded from the review conducted that in order to obtain the best weld quality, the following points should be considered: • Joining metals to polymers is challenging due to huge differences in their properties. However, choosing soft metals, such as aluminum and magnesium, along with high thermal decomposition polymers would successfully form a successful joint using wide range of processing parameters.
• A large shoulder diameter with a concave feature and a short, tapered pin enhances the material flow through the weld joint that directly improves the joint strength.
• Increasing the heat input through the weld joint enhances the strength. This can be achieved by increasing the tool rotational speed, tilt angle and plunge depth, or by decreasing the traverse speed. However, excessive heat input can cause polymer degradation and lead to internal defects decreasing the joint quality. Hence, it is critically important to optimize the process parameters to provide the optimum heat input.
• The hardness of the stir zone reduced as the heat input increased and varied among locations. The hardness was altered by thermal degradation of the polymer and variations in crystallization degree.
• Internal defects such as voids, bubbles, and pores, as well as the contact layer between the metal and polymer, were discovered to have a substantial impact on the mechanical strength of joints.
• Embedded metal pieces in a solidified polymer matrix are found in the microstructure of the FSW joint cross-section.

•
The bonding mechanisms throughout the joining process are major mechanical interlocking via the creation of micro-and macro-constraints in addition to a minor

Summary
In this work, the friction stir welding (FSW) process for solid-state dissimilar joining of hybrid metal-polymer structures was reviewed and discussed. This was achieved by reviewing different works in the literature that have studied the effects of the different materials utilized in the process, the used tool types of the pin and shoulder profile, and the different process parameters on the joint's quality, strength, hardness, microstructure, and thermal profile. It can be concluded from the review conducted that in order to obtain the best weld quality, the following points should be considered:

•
Joining metals to polymers is challenging due to huge differences in their properties. However, choosing soft metals, such as aluminum and magnesium, along with high thermal decomposition polymers would successfully form a successful joint using wide range of processing parameters. • A large shoulder diameter with a concave feature and a short, tapered pin enhances the material flow through the weld joint that directly improves the joint strength.

•
Increasing the heat input through the weld joint enhances the strength. This can be achieved by increasing the tool rotational speed, tilt angle and plunge depth, or by decreasing the traverse speed. However, excessive heat input can cause polymer degradation and lead to internal defects decreasing the joint quality. Hence, it is critically important to optimize the process parameters to provide the optimum heat input.

•
The hardness of the stir zone reduced as the heat input increased and varied among locations. The hardness was altered by thermal degradation of the polymer and variations in crystallization degree.

•
Internal defects such as voids, bubbles, and pores, as well as the contact layer between the metal and polymer, were discovered to have a substantial impact on the mechanical strength of joints.
• Embedded metal pieces in a solidified polymer matrix are found in the microstructure of the FSW joint cross-section.

•
The bonding mechanisms throughout the joining process are major mechanical interlocking via the creation of micro-and macro-constraints in addition to a minor interfacial chemical adhesion between the metal and consolidated polymeric layers.

Research Gaps and Future Outlook
FSW researchers are interested in metal-polymer hybrid structures. However, several study areas of FSW of metal-to-polymer joints must be addressed in order to better understand the joint properties and increase the utilization of FSW technology in joining metals to polymers. These areas include the following: • In depth examination of the flow mechanism of deformed metal under the impact of stirring tools during welding, with the aim of increasing weld metal flowability and the service life of FSW tools.

•
Research could be extended to better comprehend the principles of thermo-mechanical interactions and material flow characteristics during welding to produce strong joints by experimenting with different combinations of input parameters to increase the efficiency of the FSW of dissimilar materials.

•
Extensive investigation of the binding between metals and polymers at the microscale is needed to manage and eliminate micro-flaws in order to increase joint strength and create joints with excellent overall properties. • Evaluation of joint fatigue, bending, and toughness to better understand the feasibility of FSW hybrid joints in applications involving dynamic conditions.