Recent developments in ultrasonic welding of similar and dissimilar joints of carbon fiber reinforcement thermoplastics with and without interlayer: A state-of-the-art review

1.1 Ultrasonic application and ultrasonic welding (USW) for polymers

USW is a speedy process of joining materials. It can be used for joining alloy/composites and composite/composite. USW is considered as a solid-state welding method that is defined as a sustainable welding process [1]. It is worth mentioning that the USW is one of the most promising fusion bonding processes. Nonetheless, the fusion bonding technologies consist of USW, induction welding (IW), and resistance welding (RW) techniques [2]. The basic concept of USW is based on the ultrasonic wave, which is same as any other ultrasonic process machine (based on ultrasonic wave) applications. Figure 1 illustrates the current applications of various ultrasonic machines for polymer industries. As shown in Figure 1, the major applications of ultrasonic for polymers can be classified into four main categories, which are fabrication process, welding, defect detection, and cleaning. However, each classification can be further classified as shown.

Figure 1

Applications of ultrasonic in polymers.

Furthermore, the USW main principle is based on very high frequency usually 20 kHz, associated with very low amplitude at the joining surface interface of the adherends to be welded. In the USW process, the vibrations are converted to heat between the two joining surfaces, which melt the thermoplastic composites (TPCs) and enable the joining [3,4,5,6], where the vibration energy leads to a surface asperity deformation, which disperses into heat. In consequence, the heat melts the surface asperities, which then flow, expanding the interfacial area, which permits the molecules of the polymer chains to diffuse [7]. High amount of frequency in kHz in the oscillation of the USW to the stacked specimens, frictional heat will be generated by the plastic boundaries, which melt and form the joint [4]. A schematic representation of USW machine with its components is given in Figure 2. The generated frictional heat is caused by the transmission of the mechanical vibration, which helps in melting the TPCs causing them to flow, which will cause the interfacial bond between the two specimens to join. This heating mechanism is classified as surface friction as well as viscoelastic friction [8,9]. Mainly, the viscoelastic heating became governing when the temperature reached the glass transition temperature, is the reason of providing the absolute heating during welding process [8]. As shown in Figure 2, the USW machine consists of generator, transducer, booster, sonotrode, and fixture. The generator converts the electrical power to high electrical energy at 20–40 kHz. The transducer works as a converter, which alters the high frequency electrical pulses to a mechanical vibration. The amplitude of this mechanical vibration is further increased or decreased by the booster. In consequence, the mechanical oscillations are transferred to the specimens by the sonotrode. However, the specimens remain stable by the fixture on the anvil [10]. Besides, the welding process consists of a solidification phase, which is a consequence of the vibrational phase, in which the heat is generated. Further, the vibration energy is transfererred to the sontrode, which creates the welded zone by heating the surface [11].

Figure 2

The schematic of USW [20].

The USW has the potential for extensive applications, particularly in lightweight vehicle structures, owing to its good strength and stiffness [12,13], as well as in the marine industry [14]. The global market scale of USW technology is demonstrated in Figure 3 [15,16]. As shown in Figure 3, the applications of joining polymers by USW are much greater than those of metals. Besides, a substantial increase in the global market is observed, where in 2022, the market size reached 550–560 million US dollar, while it is estimated to reach 825–840 million US dollar in 2030. However, the first publication regarding USW of TPCs was recorded in 1989 published by Benatar and Gutowski [17], where the processing information was highlighted, and the obtained outcomes were significant. Nonetheless, the USW has many advantages, e.g., it is a time-saving method and an economical technique for joining polymer composites [18], and disadvantages, e.g., it is mainly used for welding small areas and it can be used under some conditions for welding larger areas [19]. Further advantages and drawbacks of USW are illustrated in Figure 4.

Figure 3

The global market scale of USW technology: (a) The development of USW in polymers and metals and (b and c) the annual global market of USW in various application sectors [15,16].

Figure 4

The advantages and disadvantages of USW process.

From literature records, Sandeep and Natarajan [21] presented a review that focused on investigating the recent developed joining methodologies for carbon fiber (CF)-reinforced polymer (CFRP)/Al-joined hybrid structures. Liu et al. [22] presented a comprehensive review to examine the metal/fiber-reinforced thermoplastic (FRTP) joined by USW. Abbas et al. [23] explored the developments in USW of joining lightweight alloys. Furthermore, Pramanik et al. [24] studied the earlier available joining techniques that were utilized in joining CFRP/Al. Besides, Fan et al. [25] presented a review study focused on considering various joints joined by implementing the ultrasound vibration. The dissimilar joining of carbon fiber-reinforced thermoplastic (CFRTP) with metals by laser welding (LW) has been reviewed by Jiao et al. [26]. Additionally, Acherjee [27] reviewed the parameters, process monitoring, and quality attributes of laser transmission welding used for similar and dissimilar joining of polymer/polymer and polymer/metal. Moreover, Li and Palardy [28] reviewed the methodologies, which are currently used to monitor the damage caused by fiber-reinforced polymer (FRP) joints. However, Li et al. [29] accomplished a comprehensive review that focused on examining the mechanical features as well as the behavior of structural health monitoring of carbon nanotube (CNT)–FRP composites. An evaluation between different types of joining approaches used to join TPCs and thermoset composites (TSCs) was published earlier. As findings, the USW was found to be most suitable in some essential factors of joining performance, processing time, minimal surface preparation, in situ inspection production environment, and reproducibility [30]. However, a short review study has been accomplished by Forintos and Czigany [31], who investigated the applications of CFRPs and highlighted the electrical characterizations of the CF reinforcement. Zeng et al. [32] inspected the impact of fiber interfacial and the mechanical responses of CFRP reinforced with gelatin-CNTs. In addition, the welding methodologies that were implemented in joining polymers for biomedical applications were stated by Amanat et al. [33]. Nonetheless, Hamza and Jalal [34] accomplished a comprehensive review that covered the PCs joined by friction stir proceeding. Asmael et al. [35] performed a review study that investigated the enhancements in tensile features of the CFRP joined by friction welding. Meanwhile, El-Sayed et al. [36] presented a review study focused on friction stir welding (FSW) in joining metallic materials. Wang et al. [37] reviewed the TPCs reinforced with various types of fibers and joined by USW. Zhao et al. [38] presented a study that covers the recent progress in USW in joining various fiber-reinforced polymer composites. Nagarajan and Manoharan [39] presented a comprehensive review that studied various types of joining and welding used in joining metal with polymer in a hybrid structure. Besides, Ni and Ye [40] reviewed various Al alloys joined by USW where the macrostructure, microstructure, and mechanical characterizations of the weldment were addressed. Moreover, Bose et al. [41] accomplished a recent review study which focused on the latest improvements in the USW of polymers and polymeric composites that investigate specific types of TPCs reinforced with CF and glass fiber (GF). Besides, for obtaining superior mechanical properties, Mirzaahmadi et al. [42] added TiO₂ and CuO nanoparticles to the polyvinyl chloride interface of the ultrasonically welded GF-reinforced polymer (GFRP) at optimum process parameters. However, Mahato et al. [43] presented a review paper exploring fibrous PC materials’ mechanical response (static and dynamic) under various environmental circumstances and mainly to define the failure modes. Meschut et al. [44] accomplished a comprehensive review that focused on joining by plastic deformation. Additionally, Luckachan and Pillai [45] performed a comprehensive review for the purpose of investigating eco-friendly polymers and their developed perspective. Besides, the recent developments in eco-friendly sustainable materials such as bamboo and wood and their modification processes were investigated and comprehensively reviewed by Paul et al. [46]. Moreover, Alhijazi et al. [47] presented a comprehensive review investigating the latest improvements in palm fiber composites. In a further study, Odesanya et al. [48] revised several types of natural fiber-reinforced composite (NFRC) implemented in ballistic functions. Besides, based on high demands on natural fiber such as bamboo, it is essential to develop the bamboo material properties. Accordingly, Sun et al. [49] presented a comprehensive review that explored the nanotechnology applications in improving the bamboo material properties in different aspects. However, Gandini and Belgacem [50] presented a review on the preparation process for producing polymers generated from sustainable resources. In addition, to overcome the defects of NFRC, Hosseini et al. [51] presented a comprehensive review that focused on exploring the enhancements of fiber/natural-fiber treatment on the mechanical and physical characterizations of NFRC. Accordingly, several treatment processes were discussed, such as NaOH, polyvinyl alcohol, steam-exploded (SE) treatment, fungicide solutions, plasma treatment, distillate treatment, and chemical coupling agents. Nonetheless, Rafiee and Shahzadi [52] presented a comprehensive review on the investigation of the mechanical characterizations of polymer reinforced by nanoclay. Hosseini et al. [53] presented a comprehensive study that focused on exploring the behavior of fiber-reinforced composites (FRCs) laminated with CNT. Francisco et al. [54] presented a comprehensive review addressing the enhancement of various nanocomposites (NCs) (tubes, particles, and layers) on polyamide NCs. Gu and Gu [55] reviewed the methodologies utilized in investigating the microbiological degradation and the deterioration of numerous polymers with different grades of degradability. Anugrahwidya et al. [56] presented a comprehensive review that focused on investigating the performance of starch-based bioplastics embedded with fiber and nanoparticles. SJ and Natarajan [57] presented a review study that focused on recently developed joining techniques for joining Al to polymer/CFRP for hybrid lightweight structures. Furthermore, Xiao et al. [58] explored the latest improvements in the mechanical characterization of hybrid fiber metal laminates reinforced polymers and their applications in automotive industrial sector. A comprehensive review has been accomplished to investigate the current techniques for curing the deformation process of CF-reinforced resin composites and their challenges. Also, Zhang et al. observed that the significant parameters that affected the curing deformation were variable and varied by changing the process and environmental conditions [59]. However, Yu et al. [60] performed a novel review study that investigates the preparation processes as well as the regulations of cellulose materials for developing cellulose/epoxy (Ep) composite which overcome the disadvantages of CF/Ep composite such as high costs. Furthermore, due to the global high demand on CFRP as shown in Figure 5 where a yearly gradual increase can be observed, Zhang et al. [61] presented a comprehensive review that focused on the recyclability of CFRP. In addition, the global estimation of CFRP wastes in 2050 from the aeronautical sector by region is predicted and plotted in Figure 6, the maximum waste is estimated to be in Asia followed by Europe. While, the lowest is estimated to be in Oceania. Besides, a further analysis of capital investments in CFRP in various sectors is illustrated in Figure 7. It is seen that highest capital investment is recorded by aerospace and defense sectors, while the lowest is in marine sector.

Figure 5

The annual global demand on CFRP in kilo tons [61].

Figure 6

The global estimation of CFRP wastes in 2050 from the aeronautical sector by region [61].

Figure 7

The annual capital investments in CFRP in various sectors [62].

Meanwhile, this state of art review is considered as the first comprehensive review that covers the joining process of similar and dissimilar CFRP weldment produced by USW, considering single, double, multi, and continuous welding. In this context, the enhancement of process parameters on the microstructure and mechanical characterizations, and welding quality is discussed. Besides, it highlights the influence of using energy director (ED) or interlayer and the consequence of preheating and heat treatment on the weldment strength and quality. The various fabrication processes and other welding processes are also briefly addressed, besides highlighting the challenges faced by USW.

1.2 Comparison of USW and other joining approaches

The USW has become more attractive to use in different industrial applications due to its several advantages over regular welding processes such as FSW, shield metal arc welding, etc. Apart from the USW advantages mentioned in Figure 4, it also requires a very short welding time (a few seconds) and does not require the usage of another material as gas tungsten arc welding or between the interface as required in other welding techniques such as RW, as well as the welding strength is very close to the neat polymer strength [63]. Additionally, the most unique advantage of USW over other welding processes is that the potential of in situ observation is automated through process data [64], e.g., a strong relationship among the response of a microprocessor-controlled welder and the physical variations at the welding interface for USW using flat energy director (FED) [65]. Furthermore, joining TPCs by USW has the advantage of the absence of forging material on the welding interface, such as metal mesh, and other welding features like significant welding joint quality [64]. Accordingly, a comparison study was performed between the strength joint by adhesive bonding and by USW, the joint was in between C/Elium^® [66]. The results showed that ultrasonically welded composite joints had a 23% higher lap shear strength (LSS), as shown in Figure 8. It has been found that the optimal welding time for an ultrasonic welded junction was 1.5 s, compared to 10 min for an adhesively bonded joint. Besides, it has been reported that it is necessary to perform a co-curing process to coat the TSCs with a rich layer of TPCs for the purpose of achieving a sufficient adhesion in between TPC and TSC polymers for the welding process, such as USW [67,68]. In addition, by implementing the adhesive joining technique, numerous bonding defects may occur in the bonded joints, which consequently affects the bonding quality, such as porosity, inadequate adhesive, and cracking [69]. However, the basic steps of adhesive bonding and types of adhesive joints are shown in Figure 9. Nevertheless, the quality of adhesive bonding can be established either by physical breaking or by observing the number of adhesive-bonded joints [70]. Additionally, by generating a hybrid joining technique that consists of adhesive joining followed by USW, the uncured adhesive bonding in the USW process is uninspected due to the high damping property, which will dramatically reduce the ultrasonic signal, and subsequently, the inspection sensitivity [71]. However, this effect can be mitigated by squeezing the melted adhesive out by using the pressure of the sonotrode in the USW. The formatted weld nugget may use to overhaul the partially adhesive bonded joints under the appropriate USW process parameters [70]. Furthermore, Zhang et al. [70] implemented the USW to repair a previous joint of CFPA6 by adhesive bonding.

Figure 8

A comparison of the LSS between welded and adhesively bonded laminate joints for C/Elium^®, under USW conditions of 1.5 s, 3 bar, and 75% amplitude (48 μm) [66].

Figure 9

(a) Steps of adhesive joining and (b) types of adhesive joints.

Moreover, an evaluation between the strength and stiffness of CFRTP and CFRTP joints by USW spot welding and by mechanical fasteners (MFs) with double-lap shear (DLS) and pull-through test had been carried out [72]. However, on comparing the results of LSS in MPa of APC-2 laminates (polyetherimide, PEI film) joined by USW, RW, IW, MFs, adhesive bonding, and bulk heating, the results of USW show to be the most significant joining process and the most studied method on average compared to other methods [73,74,75,76]. A recent review shows that, given the capacity to transmit loads and create a homogenous junction that maintains structural integrity, adhesive bonding is mostly preferred to mechanical joining due to the lower cost and less assembly weight. However, fusion bonding approaches such as USW, IW, and RW have great potential for applications requiring quick processing times, such as wind turbine blades, automotive bulkheads, aerospace fuselages, and surfboards, which require large volumes of material. The advantages of fusion bonding methods include lower surface preparation necessities, recyclability, reprocessing, and enhanced integrity/durability over other joining processes [76]. Furthermore, a comparison between the capital investment costs of arc welding and USW is shown in Figure 10. Besides, the USW consider as one of the sustainable welding process. The sustainability criteria in welding is shown in Figure 11.

Figure 10

A comparison between capital investments cost of arc welding and USW technologies.

Figure 11

Sustainability in welding technology.

1.3 Polymers and polymer matrix fabrication

1.3.1 TP polymer

The advantages of TPCs over other polymers are that they are cost-effective in manufacturing, and have excellent impact resistance and recyclability [72,77,78,79]. Moreover, consolidation under elevated/high pressure and heat after welding through melting is a unique advantage for TPCs [80]. Table 1 presents the properties of TPs used in the USW process, followed by the most favorable types of TPs to fabricate with CF and joined by USW. As observed from Table 1, there is a considerable variation in the values of density, Young’s modulus, processing temperature, melting and glass transient temperatures of the TPCs, which are based on the unique structure of each TPC. All mentioned properties are classified as the most important properties that influence the USW process since they directly impact weldability, melting, crystallization, phase transformation, and the thermal cycle. The importance of maximum operational temperature during USW is based on the purpose and the nature of using the TPCs as a reinforcement of the CF or as an ED or interlayer coupling, considering that the melting must be generated in the ED or interlayer before the adherent, which means that the melting temperature of the interlayer is supposed to be less than that of CFRTPs. Besides, the maximum operational temperature significantly enhanced the material flow in the welding zone. However, considering the density and Young’s modulus is essential because it may affect the USW process factors, such as required welding power and welding time. Moreover, the glass transition temperature gives an indication of where the chains of polymer starts to move [81]. In addition, the glass transition temperature is directly affected by the crystallinity in semi-crystalline polymers [82]. Based on these, the material properties of TPCs must be studied intensely before performing the USW process. Furthermore, Elium is a novel acrylic TP resin developed by ARKEMA. The advantage of Elium over other kinds of TPs is that it can be cured at room temperature [63]. The mechanical features of Elium^® with various fiber reinforcements studied in various literature are fracture toughness [83], tensile [84], flexure [85,86], impact [87,88], and vibration [89]. Also, their mechanical properties are similar to TSCs’ mechanical properties. Shogren et al. [90] explored the enhancement of orientation on the microstructure and mechanical characterizations of the polylactic acid (PLA)/starch composite filaments produced by the extrusion (twin screw) methodology. Chen et al. [91] inspected the ability to enhance the toughness of polyethylene terephthalate (PET), where the outcomes showed a significant enhancement in the mechanical properties. Additionally, the results of that study are considered as an initial step for future application of the PET. However, by reinforcing the TPCs with CF to develop the CFRTPs, the CF was found to develop the mechanical behavior of polymers [92]. Likewise, Kord et al. [93] considered the enhancement of multi-walled CNT (MWCNT) on the dynamic behavior of PCs made of PP/reed flour (RF). The outcomes show that the MWCNT has a negative effect on some properties and a positive effect on other properties simultaneously. However, the treated polyetheretherketone (PEEK) by ultraviolet irradiation with carbon/Ep prepregs has been used as a direct co-curing process between the interfaces of TSCs/TPCs [94]. Conversely, Talbott et al. [95] recorded that due to reducing the PEEK’s crystallinity, the tensile strength, shear strength, and modulus of elasticity will all decrease. Besides, a review has examined the mechanical behavior of CF/PEEK in biomechanical applications [96]. By using the PEI as a coupling layer for carbon/PEEK, there may be a high-performance PEI-PEEK exploit caused by the molecular inter-diffusion between the adhesive and the adherend as a consequence of the total miscibility of PEI and PEEK over the melting temperature of the layer [97].

Table 1

Properties of TP polymers employed in USW processing

Types of polymers	Abbreviation	Density (kg·m−3)	Young’s modulus at room temperature (GPa)	Glass-transition temperature (°C)	Processing temperature (°C)	Melting temperature (°C)	Ref.
Elium		1181.33 ± 20.43	0.0033	97.9	85	NA	[98]
High-density polyethylene	HDPE	940–970	0.6–1.1	−100	90	130	[99,100]
Low-density polyethylene	LDPE	910–955	0.1–0.3	−100	88	110	[100,101]
Acrylonitrile butadiene styrene copolymer	ABS	1,080	1.9–2.0	100	75	125	[100,102]
Polyamide (Nylon 6)	PA6	1,084–1,230	1.5–3.0	45	80–160	221	[103,104]
Polycarbonate	PC	1,196	2.2–2.4	149	100–140	149	[100,105]
Polyetheretherketone	PEEK	1,260–1,300	3.7	143	250	343	[106,107]
Polyetherimide	PEI	1.27	3.3	215	170	340–360	[108,109,110]
Polyethylene terephthalate	PET	1,333–1,365	2.7–4.1	70	150	240	[100,111]
Polymethyl methacrylate	PMMA	1,170–1,230	2.2–3.8	105	80–100	130–160	[112,113]
Polypropylene	PP	900–920	2.55	(−25)	85–120	170	[100,114,115,116]
Polyphenylene sulfide	PPS	1,350	3.8–4.2	85–90	218	280–285	[117,118,119]
Polyvinyl chloride	PVC	1,385–1,440	2.5–4.0	80	60	160	[120,121,122]
Polyvinylidene fluoride	PVDF	1,750–1,800	1.2–1.6	–35	148–190	100	[123,124]
Polystyrene	PS	1,050–1,080	2.4–3.5	95	105	240	[100,125]
Styrene-acrylonitrile copolymer	SAN	1,070–1,080	3.9	115	85	160–260	[126,127]

Remarkably the TPCs became highly attractive over TSCs in different manufacturing sectors, owing to their high damage tolerance, cost-effective manufacturing, and welding process [2,80,128]. Numerous fusion-bonding techniques that are founded on altered heating mechanisms such as ultrasonic, microwaves, laser, IW, FSW, and hot gas and plates can be used to join the TPCs [80,129,130,131]. Anaç [132] investigated the mechanical characterization of similar/dissimilar polymers joined by FSW. Based on the literature, the TPCs are considered a semi-crystalline natural structure that enhances the mechanical features, such as shear strength, stiffness, fracture toughness, the interfacial strength of fiber/matrix, and chemical resistance. Besides, the industrial applications of polymers and polymer matrix are becoming widespread in industrial manufacturing, such as railway [133], automotive [134], aerospace [135,136], elastomeric shape memory, and its application [137,138], as well as biomedical [139]. Popp and Drummer [140] performed joining process of carbon fiber-reinforced thermoplastic composites (CFRTSCs)/steel through infrared heating and undercutting pin structure. Hussen et al. [141] presented an experimental and developed analytical model to investigate the peeling strength of a composite structure consist of multilayer hybrid textiles and coated with PVC, that was joined by using USW process. However, the most commonly used PC matrix-reinforced fibers are CF, GF, and aramid fiber (AF), as shown in Figure 12 with their specifications and compared with TPCs reinforced with TPC polymer fiber-reinforced polymers (PFRPs) and their specifications [135]. Researchers and industries show more interest in CF than other fibers due to their superior mechanical characteristics. For modern and next-generation structures, the CFRTPs are a significant candidate and have higher productivity than the CFRTSCs due to their fast formation by injection molding and press molding. However, Wang et al. [142] investigated the thermal characterizations and the fusion performance at the interface of CFRTP (CFPEEK) enhanced by heat input and joined through IW with CF subsector. Besides, Liu et al. [143] studied the compression behavior and impact resistance of CFPEEK laminates experimentally, followed by repairing with hot-press fusion with various stacking arrangements. The authors observed that 20–30% of development was achieved in compression residual strength by the repairing process.

Figure 12

The most commonly used PC matrix reinforced fibers with their specifications and compared with (a–c) TPCs reinforced with TPC polymer fiber (PFRPs) specifications [135].

1.3.2 Polymer matrix’s fabrication

The CFRP, TPCs and TSCs which joined by USW can be fabricated in various approaches. In this context, a brief discussion about the fabrication process implemented in the preparation of CFRP and polymer matrix is presented in this section. There are several methods to fabricate the CFRPs, such as vacuum-assisted resin transfer molding (VaRTM), which is considered as one of the fabrication techniques with the lowest cost [144,145]. Moreover, Francis et al. [146] presented a comprehensive review that focused on enhancing disparity in the composition of material characteristics and the performance of PEEK and then exposed the association with additive manufacturing (AM) processability of such composites. Besides, the study highlighted the challenges faced in the production of PEEK by AM. However, despite the remarkable advantages of fabricating PCs by AM over traditional manufacturing processes in terms of mechanical behavior, it still has some drawbacks. Darji et al. [147] presented an inclusive review that addressed the essentials and sorts of current AM PCs. It also investigated the present cutting-edge studies as well as progress. Besides, the gaps in the mechanical properties of PCs fabricated over several AM methods, such as powder bed fusion, material jetting, material extrusion, sheet lamination, etc., were also highlighted. Furthermore, in the prospective of producing polymer matrix, which is created by embedding fibers, Figure 13 illustrates a summary of the most utilized fabrication processes for TPCs [135]. Furthermore, Iwata et al. [148] employed a microstructure that includes a protrusion and is produced by AM to enhance the CFRTPs joined with Al by solid-state approaches. Pinto et al. [149] recorded that due to the adaptability and remarkable inherent qualities, two-dimensional (2D) materials have become a highly potential class of additives in the field of PCs, which in consequence allows the researchers to develop various NCs that might be used in a wide range of multipurpose implementations. The effectiveness of these NCs strongly depends on the integrity of the 2D materials, the relationship with the polymer matrix, distribution, and the form when implanted in the polymer. Additionally, one of the most essential key factors for achieving multipurpose applications in PC NCs is how the nanofillers (NFs) are embedded in the polymer matrix. These NCs are usually created through three various technologies, as shown in Figure 14, where red, green, and yellow signify the advantages and disadvantages of each process. However, Azizli et al. [150] presented a novel PA6/PLA nanocomposite by utilizing the graphene oxide and poly ethylene-butyl acrylate-glycidyl methacrylate with superior mechanical characterizations. While a PLA reinforced with halloysite nanotube (HNT) through melt-blending the PLA matrix with the HNT was studied by Murariu et al. [151].

Figure 13

The most utilized fabrication processes for TPCs: (a) Film stacking, (b) hot compaction utilizing polymer fabrics only, (c) jot compaction utilizing composite films; (d) powder impregnation, (e) solution impregnation, (f) injection molding, (g) matrix melt impregnation or infusion, (h) AM via fused deposition molding; and (i) fiber intermingling [135].

Figure 14

The various technologies for producing polymer NCs, addressing their major advantages (green) and disadvantages (red): (a) in situ polymerization, (b) solvent casting, and (c) melt compounding [149].

Furthermore, the generated polymer matrix’s mechanical, physical, and microstructural characterizations were investigated. The production of the polymer matrix composites (PMCs) by utilizing microwave energy has been revised by Naik et al. [152]. Meanwhile, Melentiev et al. [153] reviewed the capability of employing polymer metallization, counted as one of the AM process techniques for producing polymers and PCs. Nonetheless, Gupta et al. [154] presented a review study for exploring the benefits of producing CFRPs by utilizing the near-surface mounted (NSM) methodology. Additionally, it has been observed that the NSM method influences the shear strength, fatigue resistance, durability, and flexural capacity of the CFRPs. Besides, it highlighted the bonding performance and failure mode of the CFRPs. However, the most commonly utilized geometrical structure is the sandwich composite structure owing to its advanced energy absorption capabilities and mechanical properties, which depend on the core and face sheet material characteristics as well as the strength between the core-inner face sheet as recorded by Patekar and Kale [155].

Furthermore, various reviews were published in the field of investigating the fabrication process utilized for producing polymer matrix. For instance, Miranda Campos et al. [156] investigated the TPC matrix formed by implementing the resin transfer molding (RTM) method. Moreover, Eratbeni and Rostamiyan [155] developed a novel model of sandwich panels made of CFRPs with rhombus cores. Also, the achieved outcomes of the vibrational analysis performance of the novel sandwich structure were compared and validated arithmetically and experimentally with the traditional elliptical sandwich structure. Besides, the CFRTPs are appropriate for mass-produced products and high-end production [6,157,158,159]. Additionally, by comparing the CFRPs with conservative metals, it is observed that CFRPs have comparatively higher tensile strength and stiffness, outstanding corrosion properties, impact as well as lower densities [160]. Sawpan [161] stated that while embedding the FRP into the water, the composite thermal and mechanical characterizations are exposed to vary owing to the diffusivity of water molecules and ions into the polymer matrix. An occurrence of hydrolytic reaction, a reaction between sodium ions, which is one of the solution components, and hydroxyl ions, will be generated for electrical charge balancing, which will cause an osmotic condition, as a consequence, a considerable pressure will be generated that will cause degradation in the mechanical features of CFRP [162] as well as GFRPs [163]. Moreover, Cheng et al. [164] analyzed a chain reaction of a cross-linked network in CFRP by reaction kinetics models of CFRP to define the degradation of CFRP immersed in supercritical fluids. It is observed that the CFRP deterioration was caused mainly by the scission of the C–C, C–O, and –O– bonds in the longitudinal chain region and the C–N bond in the cross-linked segment of an Ep resin curing system.

1.4 USW for TPC and TSC

In spite of all welding techniques that may be used in joining TPCs, such as FSW, LW, RW, and electromagnetic welding [165,166,167,168], the USW accomplished advanced benefits for welding the TPCs including reliability; cleanliness; harmlessness, which allows in situ monitoring; and less welding time [169,170]. Additionally, in certain circumstances, it is possible to perform the joint by USW at a temperature below the melting point of the weldment while maintaining a high welding quality distinct from other welding methods [171]. Joining polymers by the USW has been prevalent and used in various industrial applications for many decades [172,173,174,175]. Furthermore, there are several differences between the joining process of TPCs by USW and the joining process of metals by USW [176,177]. For instance, Abbas et al. [178] performed a dissimilar metal joint Al/Cu by USW while considering increasing the welding contact area. Besides, a study investigated the USW of plastic optical fibers (POFs), which are made of polymethyl methacrylate polymer (PMMA) joined with CFRPs fabricated by using VaRTM (Ep resin)/thermoset plastic. It has been noticed that the POFs melted and firmly welded to the CFRP, and it experienced the deformation of the CFRP. While no changes were noticed in the CFRP, it remained in the same state without melting [179]. Nevertheless, the USW compared with other fusion bonding techniques, such as adhesive bonding and MFs, by using the APC-2 laminates (PEI film) system has been examined [30,180]. Bonding TPCs by TPC films as hot melt adhesive by fusion bonding has been examined; with high interest in USW technology, the bonding was accomplished for similar and dissimilar materials [2]. Meanwhile, Biswal et al. [181] utilized the USW to join vitrimer composites with GF and CF. Sadeghi et al. [182] joined GF/PA6 by USW through using FED to concentrate the interface heat, where the optimum welding time was found to be 1.8 s at which the welding strength reached 24.46 MPa. Moreover, the welding process of dissimilar materials is more sensitive than welding similar materials, as a primary necessity, the melting temperature for both materials must be identical with a temperature variation less than 20°C [183]. The interest in dissimilar welding TPCs to TSCs and TPCs to metal has been widely studied by many researchers [184,185,186,187,188,189,190]. The dissimilar USW joins metal to TPCs, and TPCs to TSCs [185]. However, as a demonstration in USW, the strength of the composite interface is correlated with the molten polymer flow [17]. A rich coupling layer of TPCs must be included in the un-cured TSCs, owing to the fact that TSCs by itself is not weldable; so the coupling layer must be added to achieve weldability [5]. As mentioned in the literature, the curing reaction can be done only for TSC resins, but the process was referred to as (co-curing) [191]. Furthermore, several publications focused on finding a strong method to connect the TSCs with the coupling layer during the co-curing process [94,191–193]. The polyethersulfone, PEI, and polysulfone are examples of TP resins and share a typical amorphous nature. Besides, they have also been reported as a compatible common Ep system [193–195]. Elsewhere, at appropriate process parameters, the interface between ultrasonic oscillation and TPCs will result in vibration energy that disperses as heat, which elevates the temperature of the plastic to a suitable level to endorse fusion bonds between the workpieces [9]. On the other hand, a novel study on hybrid welding of CFRTP established that the location of failure and the final thickness of the weld line were significantly influenced by the welding technique used as well as the initial thickness of the TPC film. The welding procedures were found to be made more accessible by thicker TPC films [196]. Liu et al. [197] investigated the nonlinear (NL) friction performance of CFPA6/Al joined by USW. Lionetto et al. [198] performed a joining process of CF/Ep/Al alloys by using hybrid USW with single-spot. Besides, Shi et al. [199] performed a dissimilar joint by using the USW through joining CF/PA66 to 6061Al by using ED made of PA6, where the effect of various process parameters and the impact of ED crystallinity on the welding joint strength were examined. Similarly, Kalyan Kumar and Omkumar [200] joined the CF/PA6 by USW while implementing interfacial coating instead of single ED, where the interfacial coating layer behave as an insulator, which avoids the galvanic corrosion in the joint and the fretting damage. Furthermore, Huang et al. [201] employed a hybrid joining technique, combining the plastic deformation and chemical bonding to join CFRTSs with thin metal sheets. Reisgen et al. [202] implemented a novel technique for embedding metallic elements through the PMCs to join metals to polymers with FRC using traditional joining methods. Furthermore, by investigating the welding performance of the semi-crystalline polymers, Xue et al. [203] revealed that the chain mobility of semi-crystalline polymers remarkably decreased, since in autohesion process, the interdiffusion of polymer chain and the crystalline structures’ presence are considerably limited. On the other hand, Lamèthe et al. [204] explored the polymer dynamics subjected to the PEEK matrix composite welding. The obtained results show that the essential factors that influence the interface strengthening are the crystallization and the interdiffusion. Additionally, Kurtz and Devine [205] reported that when the polymer surface cool rapidly by the injection molding components, a thin layer of amorphous structure will form at the surface, which is further explained by Awaja [206] to have a remarkable influence on the self-bonding progressions, since the crystallinity content of the surface layer affects the inter-diffusion of molecules. For example, the graphical representation of the interface of polymer–polymer bonding processes is illustrated in Figure 15. Moreover, it has been reported that the crystallinity highly influenced the interfacial shear strength of CF/polyphenylene sulfide (PPS) and CF/PEEK [207,208].

Figure 15

A graphical representation of the interface of the polymer–polymer bonding processes: (a) molecular bonding, (b) chain interdiffusion, and (c) chain entanglements [206].

1.5 Recyclability of CFRPs

As the global market demand increases, capital investments as well as the wastes of CFRP are increasing annually, which cause the necessity of recyclability of CFRPs to become essential. The waste of CFRP is considerably important due to challenges and difficulties of it is huge volume [209]. Besides, the maximum amount of CFRP wastes is from aerospace industries [210]. In addition, it is necessary to recycle this type of waste considering the cost-effectiveness and environmental impact [211]. The CFRP wastes can be classified into two groups, which are new scraps and old scraps [212]. However, the necessity of recyclability of CFRPs is also due to the great amount of consumed energy that is required to fabricate 1 kg of CFRP which is around 28 MJ [213]. The life cycle and the most used recycling method for CFRPs are shown in Figure 16. Also, further classifications of CFRC waste and scrap managements are shown in Figure 17. The mechanical recycling technique considered efficient in managing the increasingly great volume of CFRP waste based on grinding the CFRCs into minor pieces [214]. Nonetheless, the amount of materials, which are processed by the mechanical recycling, can be utilized as a partial reinforcements in further productions, which in consequence will restrict the combination in new materials [209]. Meanwhile, it is recommended to use the chemical recycling for long fibers [215]. However, with the recent progress in recycling manufacture field, it was noticed that the recycled carbon fibers (rCFs), achieved by implementing the solution degradation or pyrolysis methodologies, has lower resin content and good adhesion to the resin which subsequently augment the mechanical characterizations of the CFRP [216]. Additionally, researchers show interests in studying mechanical and electrical properties and the environmental impact of cementitious composites reinforced with rCFs. Recently, Vidal et al. [217] proposed the enhancement of binders made of ionic dynamic networks on the processing, properties as well as recyclability of CF/Ep. Though each recycling process has it is own advantages and limitations and its own specific properties such as energy demand in mechanical recycling [218], fiber recovery and their retain characterizations [211], global impact and cost analysis must be considered before starting the recycling process [212,215].

Figure 16

(a) The CFRP life cycle and (b) the most used recycling methods for CFRP [215].

Figure 17

The waste of CFRC and dry CF scrap management routes [214].

There are many factors that restrict the recyclability of CFRP, such as the initial steps before recycling process, namely, identification, assembly, and separation are a considerable challenge [219], the difficulty of having the same significant mechanical properties of CFRP after recycling process [220], the cross-linked combination between the CF and TSs lead to more difficulties in treating the waste [221]. Further limitations facing the CFRP recyclability are illustrated in Figure 18. However, the recycling approaches of CFRP are promising in terms of reducing the greenhouse gas emissions, resource depletion, and energy usage [214]. Besides, the financial feasibility is directly associated with the consumed energy cost and raw material cost.

Figure 18

The limitations of CFRP recycling.

2.1 Types of USW machines and welding tools

Commercially, a variety of USW machines are available; the main differences in USW machines are energy output, maximum force, amplitude, and range of frequency, although all these parameters are adjustable. The USW machines generally consist of the generator, transducer, booster, and sonotrode [223–226]. Figure 21 illustrates the USW machine components’ structure and schematic [227]. Mainly, all USW machines provide the same variable adjustable process parameters, although there are some differences in their range. However, the USW machines are classified into two types based on the oscillation direction as shown in Figure 22 [228]. As realized from Figure 22(a), the oscillation direction in polymers joining is perpendicular to the welding zone, where the high vibrational frequency transmission causes a heat generation at the interface which melts the TPCs and makes the TPCs to flow and form the bond between the two specimens to be welded [185]. Besides, from Figure 22(b), the oscillation direction in metal joining (similar joint or dissimilar metal to TPC) is parallel, which influences the interaction area between the sonotrode and the specimens, in which the friction action at the surface makes the solid-state bond to form the joint without melting the metal material [229,230]. Nonetheless, a tabular classification based on the type of USW machine used in joining polymers is presented in Table 2 without using an interlayer between the two welded specimens.

Figure 21

Structure and the schematic of USW machine component [227].

Figure 22

Types of USW machine based on oscillation direction: (a) USW machine for joining polymers and (b) USW machine for joining metals (similar/dissimilar) [228].

Furthermore, the welding tool in USW process, known as sonotrode or horn, and it has an impact on the performance of USW processes, welding parameters, and welding quality. However, owing to the high expense of the sonotrode, only a few studies focused on examining the effect of the welding tool. The welding quality and the welding process are affected by the distance between the sonotrode and the interface of the working piece; owing to these joints categorized into two types: (1) Direct ultrasonic (near-field), the distance of a maximum of 6 mm and (2) indirect (far field the distance is greater than 6 mm, which may reduce the amount of energy reaching the interface [175]. Besides, a unique study has investigated the enhancement of changing welding tools on process parameters, overall welded area, LSS, and fracture of CF/PPS [231]. The samples were separated into two groups: single spot welding (SSW) welded the first group samples by using three different sonotrodes with 10, 20, and 40 mm diameters, as shown in Figure 23. The first and second group samples were welded by double spot welding (DSW) with three different sonotrodes 10, 20 and 40 mm diameter as shown in Figure 24 [231]. Furthermore, Yang et al. [232] joined CF/PEEK by USW using sonotrodes with 10 and 20 mm diameters with different ranges of spherical radius. Besides, the joint’s microstructure, fracture characteristics, and tensile-shear properties were examined. The findings demonstrate that joining with a spherical surface anvil can efficiently concentrate welding energy. As a result, the weld scatter was observed to decrease. The influence of spherical radius on the weld formation and the area of scattered welds are shown in Figure 25. For both large and small sonotrodes, a remarkable quantity of scattered welds was observed on the fracture surface by using a flat anvil, and these results were significantly validated. Figure 25 demonstrates the welding energy and spherical radius enhancement on the joint failure load. A significant variability in the failure load for flat anvils has been revealed, which suggests that the weld quality is quite unsteady. A lower quantity of energy is required to produce over welds, and the energy concentration effect is superior for spherical surface anvils with smaller radii, which implies that the stability time will be shorter for the smaller radius. The spherical surface anvil has increased stability and mechanical qualities compared to the flat anvil [232]. Additionally, the sonotrode displacement and duration directly impacted the weld quality. As a result, an in situ inspection approach based on target sonotrode displacement and duration has been developed to assess the weld quality [233].

Figure 23

Schematic of SSW joints using three dissimilar Sonotrodes. The grey semi-transparent parts specify the cross-section area of each sonotrode: (a) D = 10 mm, (b) D = 20 mm, and (c) D = 40 mm [231].

Figure 24

Schematic of DSW joints using two dissimilar sonotrodes. The grey semi-transparent parts specify the cross-section area of each sonotrode: (a) a sequential spot welding with sonotrode D = 10 mm; and (b) an instantaneous spot welding [231].

Figure 25

The Measured scattered weld area: (a, c) sonotrode with small diameter 10 mm, (b, d) sonotrode with large diameter 20 mm and the joint failure load at dissimilar welding energies and spherical radii [232].

Moreover, an investigation of the microstructure (fracture morphologies) of short CF/PEEK joined without ED by using different sonotrodes at various welding energies (600–1,600) J, concluded with remarkable results, has been carried [232]. The edge of the sonotrode may cause a stress concentration which caused an approximately annular weld shape and the weld quality was not good. As shown in Figure 26, the fracture morphology is consistent over the whole weld. For a standard weld, the primary fracture mechanisms were adhesive fracture and fiber-matrix debonding. However, as a result of the LSS test and ultimate failure (UF) of CF/PPS joints by USW under different diameter sizes of sonotrode, the ultimate failure load (UFL) was considerably improved by around 86% by using a sonotrode with D = 20 mm, and further improvement has been obtained by using a sonotrode with D = 40 mm. More extraordinarily, the LSS observed reduced by around 22% for sonotrode with D = 20 mm and D = 40 mm for SSW samples. Besides, the averaged LSS achieved by using 20 and 40 mm diameter displayed superficially greater scatter with 13 and 11%, in contrast, sonotorde with 100 mm diameter achieved 3%. Furthermore, the reduction in LSS denotes a decline in the load-carrying efficiency of the SSW joints. This indicates that the total welding quality obtained by utilizing the sonotrode with D = 20 mm and D = 40 mm SSW joint was lower than that welded by using the sonotrode with D = 10 mm counterparts, as shown in Figure 27 [231]. Nevertheless, the sonotrode with D = 40 mm provides more significant energy efficiency due to its additional inspiringly enhanced magnitude, almost 140%, of the spot-welding area. Furthermore, the SEM micrographs have been employed to examine CF/PPS SSW welded failure modes using different sonotrode sizes to evaluate the welding quality. It was observed that the deep fiber imprints on polymer resins and the bare fiber bundles depend on the fiber matrix featured [231], which indicates that high welding quality was achieved [3]. Figure 28 presents the SEM details of the fracture surface welded by a sonotrode with 20 and 40 mm diameter at 1,500 N, 1,000 N·s⁻¹. In Figure 28(a), the fiber was damaged directly and reduced the load-bearing ability of the welded joints. In Figure 28(b), the voids that appeared by using sonotrode with D = 40 mm SSW joints are thought to be one of the features that adequately describe overheated welds as stated by Gao et al. and Zhao et al. [170,234]. Besides, that considerably reduces the bonding region which could effectively support the load. Jongbloed et al. [235] studied the impact of sonotrode with rounded geometry on joining TPCs by USW. It is revealed that significant welding quality can be achieved by using rounded sonotrode, while reducing the required heat at the welding interface since the direct local exposure of the weldment to sonotrode is less.

Figure 26

(a) Unweld zone at 600 J, (b) under weld zone at 600 J, (c) and (d) normal weld zone at 800 J, (e) weld edge zone at 1000 J, (f) normal weld zone at 1000 J, (g) normal weld zone at 1200 J, (h) over weld zone at 1200 J, (i) normal weld zone at 1400 J, (j) over weld zone at 1400 J, (k) and (l) over weld zone at 1600 J. [232].

Figure 27

UFL and LLS of the SSW joints formed by diverse sonotrodes at the optimal displacements [231].

Figure 28

SEM details and fracture surfaces of CF/PPS SSW of (a) sonotrode with D = 20 mm and (b) sonotrode with D = 40 mm welded at the corresponding optimum displacements [231].

2.2 Enhancement of spot number in USW (single, double, and multi)

There are three main types of spot joints in USW: SSW, multi-spot welding (MSW), and DSW. The TPCs joined by SSW have similar characteristics of shear-strength joints with single MF of identical size [72]. An example of the difference between the SSW and DSW in the USW process is demonstrated in Figure 29(a) and (b) [170]. However, in the MSW in USW, consecutive welding has the potential to be composite-friendly and has fast-processing viable alternatives to MFs for TPC structure and CFRTP [170]. The main challenge of this type of USW is that each single spot has its unique boundary conditions. Also, the number of spots has an impact on the distribution of vibration among the joining interface. Figure 29(c) depicts a multi-spot USW schematic [170].

Figure 29

A schematic of an example of welded joints; (a) The single spot, (b) the double spot, and (c) the schematic of multi-spot USW [170].

Comparison of the fracture surface through SEM of the SSW with specimens joined by the DSW for optimum displacement after the mechanical testing is illustrated in Figure 30. In the SSW joints for both optimum displacement and optimum energy control, bare fiber, deep fiber imprints on the matrix, and torn fibers were detected as the key features on the fracture surface, corresponding to the primary failure mode, which is fiber-matrix debonding [170]. A previous study claimed that the weld qualityhighly affected by the fiber matrix debonding [3]. Furthermore, for DSW, the features of the joints were the same as for SSW: deep fiber imprints, torn fibers, and bare fibers. Moreover, Zhao et al. [236] provided a comparison between SSW, DSW and MSW in USW as an efficient substitution of single, double, and multi-MFs. As shown in Figure 31, the LSS has been implemented to obtain the failure and load capacity for each type of joint. After performing the LSS, the samples were examined, and it was determined that all the spot-welded joints had a first-ply failure, as shown in Figure 32. It is noticed that the average welded area in DSW was less in the further examples in the four-row welded joints because of the small spacing between the spots. This can be owing to the contact between melted and unmelted ED during the welding process since it is identical to the double-row welded joints with the least inter-row distance [236]. The study was followed by a further finite element (FE) investigation to analyze the mechanical performance of single-lap TPC joints welded together by the USW [237]. Besides, cohesive zone elements were utilized to estimate the welded areas. A comparison among the numerical outcomes and the actual data served as the initial step in validating the provided FE model. The numerical outcomes offer prospective directions for the structural design of TPC joints by USW and are used in airplane construction.

Figure 30

The surface of fracture and the SEM for optimum displacement controlling for SSW and DSW of CF/PPS; (a) the optimal displacement controlled in SSW, (b) the optimal energy controlled, (c) the DSW under optimal displacement controlling, (d) the spot 1 and (e) the spot 2. The scales are 5 mm (for stereo-microscopy) and 10 μm (for SEM) [170].

Figure 31

The number of rows that make up the load-bearing capacity of multi-row spot-welded and mechanically fastened junctions with 70 mm overlap length is present in the samples with 2, 3, and 4 rows [236].

Figure 32

The SEM details of welded CF/PPS show the first-ply failure of fracture surfaces for 2, 3, and 4 SW joints [236].

The CF/PA6 joined by adhesive bonding was further repaired by USW and studied by Zhang et al. [70]. The fracture microstructures of the adhesive-bonded joints were observed to comprehend the reason for the scatter in joint strength. The fracture surfaces of the examined joints together with different adhesive coverage range at the overlap area are shown in Figure 33, which shows that each joint displayed an adhesive failure mode, demonstrating the weak bond amongst the adhesive and the adherend. Additionally, after curing, the adhesive coverage area transformed from a rectangular form to an irregular form, changing the actual bonding area. As a consequence, the strength of the joints varies. Besides, Figure 34 roves that the weld area increased dramatically as the sonotrode force increased with an oscillation duration of 0.9 s. The expansion in the welded area has considerably improved the joint strength. However, when the joints were subjected to a sonotrode force below 935 N over an oscillation time of 1.2 s, considerable melting and outflow of molten materials were seen at the faying surfaces (Figure 34d), otherwise, the top of the healed joint developed a significant sonotrode indentation.

Figure 33

The adhesive effect coverage on the tensile test adhesive-bonded failure modes of 2.3 mm thickness CF/PA6 with 30 wt% fiber: (a) 25%, (b) 50%, (c) 75%, and (d) 100% adhesive coverage [70].

Figure 34

The impact of various process factors used in USW on failure modes of restored adhesive-bonded 2.3 mm thick CF/PA6 composite with an adhesive coverage of 50% at the overlap area, (a–c) welding time of 0.9 s and sonotrode force of (a) 312 N, (b) 623 N, (c) 935 N, and (d) 935 N for 1.2 s.

3.1 Welding energy

By investigating the CF/PEEK joined with SSW USW with respect to modes I and II, the results show that by using constant welding energy, the supreme critical strain energy release rates (GIC and GIIC) were achieved [269]. Likewise, while investigating samples with dissimilar surface roughness, a significant correlation has been noticed between welding energy and welding strength [270]. On the other hand, the adherents’ stiffness and thickness considerably impact the amount of required welding energy [64]. Furthermore, it has been indicated that using the welding energy to control parameters is more suitable than using vibration time [64,269]. Zhi et al. [271] found that the welding energy affects the weld indentation, joint strength, tensile strength, and welded area. The results of these relationships are presented in Figure 35. In a further investigation, an FE model has been generated to study the stress and welding indentation, and it has been observed that the welding energy directly affects the vibrational stages of the welding process [269]. Besides, Zhang et al. [272] found that increasing the welding energy above the optimum value will have the opposite effect on the weld strength and will no more be beneficial. Since designing the welding joint in USW of CFRT is considered to be one of the essential steps, Wang et al. [273] explored the microstructure and the fracture characterizations after performing the tensile test for welded CF/PA66. Besides, the obtained results reveal that the structure surface behaves as an ED, which in consequence significantly decreases the dispersion and arbitrariness of the weld dispersal, which emphasizes the weld quality and guarantees more efficient weldment compared with those of unstructured surface.

Figure 35

The impact of welding energy on the joint strength, weld indentation, and welded area of the CF/PA66 [271].

Furthermore, the LSS has been obtained by two approaches, LSS1 (maximum load over the entire overlap area) and LSS2 (maximum load over the actual welded area), which were measured for cross-ply laminate, and the outcomes show that even when the energy increased over 450 J without using ED, the welding quality was unaffected. Meanwhile, this increase only impacts the welding efficiency. Figure 36 presents the results of welded areas with and without ED under various welding energies [6]. In addition, by measuring the relation between the UFL in the LSS test and the welded area, it is noticeable that in SSW and DSW, increasing the welded area causes a remarkable increase in the UFL and vice versa [170].

Figure 36

Welded area of (a) cross-ply laminate joints for LSS and (b) twill woven laminate joints for LSS [6].

According to the literature, the fracture surface can be used to forecast the welding quality following the LSS of joined CF/PA6, the partially melted joint bead of the over-welding was examined, whereas the melted material has been ejected [255]. Because the polymer–polymer interface curing of CFRTP for the under weld was the predominant finding in the microstructure analysis, the interfacial matrix fracture has been observed individually, as shown in Figure 37(a). More fibers shift into the weld region as they are welded, which results in more pullout fibers, as presented in Figure 37(b). However, excessive welding caused part of the polymer chains to break into gases, leaving the weldment with significant porosity, as shown in Figure 37(c). Figure 38 shows the supreme shear load with diverse preheating temperatures and welding conditions, and the outcomes without preheating are involved as a baseline. The extreme shear load initially rises with rising welding energy when the preheating state is assumed to be constant.

Figure 37

The evaluation of the welded area of CF/PA6 at preheating temperature of 50°C combined with the fracture surfaces at various welded stages; (a) under-welded stage (400 J), (b) good-welded stage (800 J), and (c) over-welded stage (1,200 J) after LSS [255].

Figure 38

Experimental results of maximum shear load with various preheating temperatures and welding conditions, also the outcomes without preheating are included as a baseline [255].

Nevertheless, at a critical point, it reduces (1,000 J in this case), as presented in Figure 38 [255]. The welded joints go through stages which are under-welded, good-welded, and over-welded by rising the welding energy [274]. However, by studying the influence of welding energy on the microstructure of CF/PA6 with 40 wt%, joined by USW, the microstructure investigation of the welded part shows a close relationship between the mobility of the fiber and matrix inter-diffusion. Furthermore, under welding occurred at low energies (Figure 39) at 200 J. The weld zone is dominated by melted polymer, with hardly any visible fibers. According to Figure 39 at 400 J, when energy rises, the weld region expands, and as polymer chains pass across the welded region, more random fibers are observed. Referring to Figure 39 at 600 J, when the joint is over melted with excessive energy, some pores were created that encroach the area amongst the polymer and the fibers, changing the distribution and arrangement of the fibers from the original material [274].

Figure 39

The microstructures cross sectional evolution through the welded area of CF/PA6 [274].

tWeld area and performance features like LSS are two factors that can be utilized to quantify weld quality. Figure 40 depicts the correlation between weld energy, welded area, and welding strength. Three weld quality zones can be established by defining the desired weld strength to three regions: an under weld (region I), a normal weld (region II), and an over weld (region III). Figure 40 demonstrates that when weld energy increases, the weld area asymptotically grows, and LSS has increased before decreasing. Any definition of a “normal-weld” region must include the location of the maximum shear strength. Nevertheless, it is necessary to decide where to position the boundaries of regions II and III [250].

Figure 40

The LSS and weld area variation with weld energy determined by microstructure investigation [250].

The USW has been utilized to join the TC910 (CF/PA6) with and without implementing the ED by using the controlling mode. Besides, at each welding energy value the welded area and the indentation of the sonotrode were explored. As an outcome, as shown in Figure 41, the twill fiber loads were instantly decreased, followed by a rise to the values mentioned in Figure 41. The same trend can be observed in all different values of welding energy with/without FED. Because, the failure progressively spread from the borders of the joints, identical fracture surfaces were detected with/without FED [6]. By replacing the cross-ply fiber with twill fiber with the PA6-3KT1 (CF/PA6) shows that the maximum load is more significant when using FED. The obtained results of both fibers were similar for LSS and the fracture results were similar to the results that cross-ply fiber identical fracture surfaces obtained were detected whether with/without FED, as shown in Figures 41 and 42 [6].

Figure 41

TC910 (CF/PA6) cross-ply fiber load vs displacement with/without FED and fracture surface with FED [6].

Figure 42

PA6-3KT1 (CF/PA6) twill fiber load vs displacement with/without FED and fracture surface with FED [6].

A recent study found that the shear strength in the welded joint at very high welding energy decreased due to the generation of pores [274]. An FE model following the experimental exploration using ABAQUS was accomplished to apply uniaxial pulling forces to the welded coupons, where a connector comprises two reference points. The CF/PA6 joined by USW under seven various levels of welding energy for examining the assessment of welding attributes without ED and the polymer morphology in the cross-section shown in Figure 43(a). Also, the evaluated fracture surface of the weld area is presented in Figure 43(b). It proves that particular pores were presented in the bonding layer following critical welding energy at 1,000 J. Additionally, Figure 43(c) illustrates the expansion of the pores at 1,600 J of welding energy. It is observed that the pores begin to form within the bonding layer and in the coupon volume, further confirming that overheating during USW is the main reason for the pores to form and not the trapped gas. The bonding effectiveness of the joint decreased because the pores appear in the bonding layer [252]. More information about variations in USW parameters and their enhancements on mechanical properties, welding quality, and welding strength are presented in Tables 2 and 4.

Figure 43

(a) The microstructure cross-sectional area of the weld zone of CF/PA6 at 1,000 J, (b) the resultant fracture surface, and (c) the dispersal of the pores in the welded coupon of CF/PA6 at welding energy of 1,600 J [252].

3.2 Welding time

It has been defined that the structure of the welded specimens has a remarkable effect on welding time [269]. By examining the LSS and fracture failure of static welding by USW SSW to examine the similarity between SSW and continuous USW, Figure 44 shows the results of LSS values consistent with the SSW process. For 80 µm and 500 N, the strength steadily increased to a high interval, i.e., 36 MPa for a vibration time of 415–565 ms. By maintaining the vibrational amplitude at 80 µm and rising the welding force to 1,500 N, the same strength is obtained in less vibrational time of 260–435 ms [165].

Figure 44

Average LSS OF SSW at different welding/vibration times at various welding forces and welding amplitudes. The bars denote plus and minus [165].

Upon investigating the CF/PEEK and the enhancement of welding time on the welding strength of the joint with and without ED, the results show that gradually rising welding time will increase the welding strength until the welding time reaches its optimum value. However, the welding strength decreased after a further increase in welding time, followed by voids and large cracks [256]. Above all, it has been shown that by studying the relation between welding time and the LSS with and without using the ED, the LSS drops to half of its value without using ED. Figures 45 and 46 demonstrate the joint morphology without ED vibration times of 0.7, 0.8, 0.9, 1.0, and 1.1 s, respectively. An incessant gap at the interface of the joints was noticed because the 0.7 s ultrasonic time was insufficient, and the heat generated in the interface was insufficient to melt the interface entirely, as illustrated in Figure 46(a). The gap length at the interface gradually reduced as the ultrasonic action time extended to 0.8–0.9 s, as illustrated in Figure 46(b) and (c). Nevertheless, as demonstrated in Figure 46(d) and (e), the excessive ultrasonic duration can also result in cracks and voids at the joint’s HAZ. The width size of the HAZ rose from 0.2 to 0.42 mm once the welding time extended from 1.0 to 1.1 s. However, an interface with strong bonding was achieved at 1.1 s ultrasonic action time. Additionally, throughout the HAZ, more substantial voids and cracks were scattered, also the largest porosity area only reached 0.003 mm².

Figure 45

LSS of samples with and without ED [256].

Figure 46

Morphology of the welded CF/PEEK without implementing ED at various vibrational time: (a) 0.7 s, (b) 0.8 s, (c) 0.9 s, (d) 1.0 s, and (e) 1.1 s [256].

The energy dissipation in the TPCs has been enhanced by longer welding time [272], where longer welding time cause to reduce the joints’ void formation [275]. However, at an elevated temperature, a degradation in the TPCs and fiber distortion in its orientation may occur due to a very long welding time [276]. Meanwhile, longer holding time is reported to enhance the welded joints’ strength by allowing the part to cool down fully [276]. Welding the polyoxymethylene (POM) to PMMA through a coupling layer of PLA in a very short time generated weak friction heat for various welding pressures. Besides, all welded strengths were low at a welding pressure of 0.1 MPa. The highest welding strength was 38 MPa and measured at a welding time of 4 s and a welding pressure of 0.2 MPa. The ideal welding strength at 0.3 MP and 4 s of welding duration was 47 MPa. However, the strengths were first improved and then reduced similarly at 2, 3, and 4 s, as shown in Figure 47(a) [277]. Additionally, after welding PMMA to POM by USW by using the PLA layer, the tensile test was accomplished, and as an outcome, shown in Figure 47(b), the minimal stresses and strains of POM, PMMA, and interposed sheet (IPS), with POM and PMMA recording yield strengths of 61 MPa and 65 MPa, respectively. POM showed ductile performance with an 85% minimal strain, while PMMA displayed brittle features with a 10% nominal strain. Only 43 MPa and 5% were the nominal strength and strain of the IPS, respectively. The rupture surface image of the IPS is shown in Figure 47(c), where it is possible to realize the contact between the transparent PLA and opaque POM. In other words, the sime-crystallinity might be more interesting as compared to the color. However, due to their transparency, the interface of PLA and PMMA was difficult to observe from the photos. There is no evidence of phase separation, which would indicate that PMMA and PLA were also compatible with POM. All of the materials were suitable for welding, and the IPS made of the three polymers successfully joined POM and PMMA [277].

Figure 47

(a) The welding strength at different welding pressures and various welding times; (b) nominal stress vs nominal strain for PMMA, POM, and IPS; and (c) the SEM images of the IPS fracture surface after the tensile test [277].

The relation between vibration time and LSS of joined CF/PPS similar welded by USW with triangular ED has been tested. The results indicate that the LSS as a function of the vibrational period increased until the optimum value of vibrational time, as shown in Figure 48. Also, at an approximation vibrational time equal to 217 ms the LSS reached around 30 MPa, and by increasing the vibration time to 520 ± 58 ms, the LSS increased as well to 37.1 ± 1.3 MPa. It is worth mentioning that the required vibration time to reach the extreme strength by using triangular ED is lower than the required vibrational time when using FED [278]. The numbers in the graph designate the stage in the welding process in which the specimens for mechanical investigation were achieved.

Figure 48

The LSS as a function of vibrational time [278].

Considering the welded interface of POM to PLA joined by USW as a microstructure observation through the polarizing microscope at 4 s welding time and 0.2 MPa welding pressure, as presented in Figure 49 [279]. There are differences in the crystalline forms of PLA and POM; the crystalline form of PLA is smaller than that of POM. However, the POM crystalline is identical to the IPS crystalline. Additionally, the two sides of the interfaces melt and produce molecular inter-diffusion due to IPS ultrasonic vibrations. According to Figure 49(a), the pressure employed during welding causes the melted polymers to force out of the interfaces. As a result, if the welding pressure was significantly high, some of the melted IPS polymers were forced out of the interfaces, breaking the IPS and stopping the friction among the two weld matrixes. The IPS thickness was altered to be between 40 and 50 µm, as shown in Figure 49(b). The IPS was separated into two shapes, particularly along the interfacial direction, as shown in Figure 49(c) and (d). The crystalline forms of PLA are more substantial for the weld line on the PLA side referred to as the mixed layer of PLA and IPS. This inter-diffusion layer surrounds the weld for PLA and IPS molecules. The crystalline morphologies were comparable, and the interface is fuzzy for the POM and IPS weld lines, as shown in Figure 49(d). In contrast to the other blend layer, the POM and IPS blend cannot be easily noticed by the polarized microscopy [279]. Furthermore, the relation between joint strength and welding time for the similar joined CF/PA66 without ED with 30 wt% CF has been examined, and the microstructure characterization was defined [127]. It has been observed that the pores found in the areas between the fusion zone, the HAZ, and the severe weld indentation on the surface of the upper workpieces reduced the loading capacity of the USW joints and led to the early fracture of the welded carbon/nylon 66 composite. The balance between the beneficial impacts of the weld area and the detrimental effects of the weld indentation and porosity at the fusion zone determined the strengths of the ultrasonic welds. Figure 50 presents the impact of the sonotrode pressure and the welding time on joint strength. Meanwhile, Figure 51 shows the microstructure observation [234]. Zhi et al. [243] studied the impact of welding time on peak load and welded area; both values show a linear increase until welding time reaches 2 s. After that, both values remained constant. This outcome is accurate in case that circular conventional USW and hollow fixture USW (HFUSW) were utilized. Additionally, if the square conventional USW and HFUSW were used instead, the peak load and welding area would increase to the optimum welding time, which is almost 1.75 s, followed by a decrease in both values. Furthermore, Alexenko et al. [280] joined PEI/CF by USW without using ED at various welding time. The obtained results show that by rising the welding time from 0.4 to 0.8 s, has no significant changes in LSS were it reached 42 and 48 MPa, respectively, although the microstructure of the fusion zone has been noticed to fundamentally change. Besides, the experimental test was followed with numerical simulation, which shows that the thickness of prepreg and the ratio of PEI/CF did not cause a significant impact on the tensile strength stress–strain diagram. However, Calabrese et al. [281] performed LSS to explore the welding strength of CF/Ep joints by USW through optimizing process parameters, especially welding time and sonotrode pressure. Meanwhile, the welding process was accomplished by including electrospun veils which consist of either nylon or polyether into the outer layers of the laminated composite. A remarkable impact of the TPCs interlayer was achieved, although by implementing the nylon, greater mechanical strength was obtained. Additionally, this type of joints has promising application in sports automotive, in which the reliable and rapid welding part is fundamental. More information about variations in USW parameters and their enhancements on mechanical properties, welding quality, and welding strength are presented in Tables 2 and 4.

Figure 49

Polarizing microscopy images of the welded interfaces of POM/PLA joints [279].

Figure 50

The observation effect of sonotrode pressure and welding time on the strength of the CF/PA66 without ED [234].

Figure 51

The observation of the effect of welding time (a) 1.7 s, (b) 2.1 s, and (c) 2.5 s on the microstructure of the CF/PA66 without ED [234].

3.3 Welding force and pressure

In a displacement-controlled USW of complete overlaps, a plateau in displacement has been obtained due to employing a constant welding force, where it occurred just before the optimum stage [64,65,238]. In contrast, some studies hold the welding force constant [72]. However, it has been proven that varying the welding force and vibrational amplitude affects the vibrational time and dissipated power significantly [65]. Furthermore, increasing welding force or vibrational amplitude leads to greater dissipated power combined with more significant and quicker heating rates at the welding interface consequently, the vibrational time decreases for a travel value [64]. Overall, the combination of high vibrational amplitude and force limits the supreme size of the welded area due to increased power in a short welding time [65]. A further investigation studied the relationship between different travel, welding force, and amplitude and their impact on the LSS presented in Figure 52 [64].

Figure 52

Average LSS vs travel for the different sets of amplitude and force [64].

The USW technique has been used to create a dissimilar joint of CF/PEEK-CF/Ep by implementing a coupling layer of PEI. The relation between welding force and vibration into the other welding process parameters and welding quality was examined. The obtained outcomes reveal that reducing the vibrational amplitude or the welding force will rise the essential heating time which is required to achieve the optimal welding strength, since the heat generation rates reduced. Besides, the reduction in amplitude or welding force increases the exposed temperature of CF/Ep in the welding process. Nevertheless, during the welding circumstances which produced the highest welding strength, local symptoms of thermal degradation were discovered in the CF/Ep adherend. Nevertheless, these changes were not substantial enough to be shown as impacting the welded joints’ maximum LSS. Figure 53 demonstrates the impact of sonotrode displacement on the LSS for arrangements welded at various welding force. As indicated in Figure 53, reducing welding forces caused the process to move toward minor displacement values. This change had a substantial impact on the 400/86 setup. Meanwhile, the reduction in LSS in the 1,200/86 configuration was less noticeable than the values recorded in 800/86 and 400/86 at the maximum points. Hence, the maximum points are 0.28 mm displacement for 800/86 and 0.18 mm for 400/86. Furthermore, Figure 54 presents the concerning variations in the welding force in 1,200/86 and 400/86 and the impact of reducing welding force on the generated temperature at the welding interface. The generated temperature improved at a similar step, and the 800/86 configuration temperature improved expressively [282]. Nonetheless, that study shows that both ED as well as the coupling layer expert a squeeze flow, as verified through the reduction in thickness and a slightly wavy edge among both the polymers as shown in Figure 55 which presents the cross-section micrographs. Since the heat generation rates were identical at all conditions, and the temperature curves were overlapping until almost 250 ms into the welding process Figure 54 [282].

Figure 53

The impact of sonotrode displacement on the LSS evolution in the 1,200/86, 800/86, and 400/86 configurations [282].

Figure 54

The enhancement of reducing force on temperature progress at the interface between the PEI coupling layer and CF/Ep adherend (0.28 mm displacement in the 1,200/86 and 800/86 cases, up to 0.18 mm in the 400/86 case). The diagonal arrows specify the end of the vibration. While, the vertical arrows specify the end of the USW welding process [282].

Figure 55

The cross-sectional micrographs of illustrative samples of CF/PEEK with PEI ED at diverse force/amplitude patterns and diverse displacement caused in supreme LSS. The displacements values are (a) at 0.24 mm, (b) at 0.24 mm, (c) at 0.16 mm, (d) at 0.24 mm, and (e) at 0.24 mm. The arrows specify the resin-rich weld line [282].

Koutras et al. [245] examined the degree of crystallinity at the welding interface of USW CF/PPS joints by exploring the effect of welding amplitude and force. The results prove that the cooling rates at the center of the overlap of the joints dropped from 41 to 16.9°C·s⁻¹ with moderately low welding force of 300 N and vibrational amplitude of 51.8 μm, contrary to great welding force of 1,000 N and vibrational amplitude of 86.2 μm. The decelerate heat generation below 300 N and 51.8 μm was attributed to the lower cooling rates. In addition, at lower welding force of 300 N and vibrational amplitude of 51.8 μm increased the crystallinity to moderate levels and the crystal perfection. However, great welding force of 1,000 N and vibrational amplitude of 86.2 μm primarily formed amorphous PPS and imperfect crystals. Besides, reduction in cooling rates and longer melting time with lower welding force and vibrational amplitude, which favored the circumstances for strain-induced crystallization to take place, were guaranteed. Moreover, the combined static (welding force) and dynamic (vibration amplitude) strains throughout the USW process led to the cyclic strain, ε₀. In addition, it can be assumed that only the dynamic strain will significantly affect heat generation because it is numerous orders of magnitude lower than the static strain [283].

By studying the factors that affect joint strength, Liu et al. [259] reported that the welding pressure is less crucial than other parameters on welding strength. However, Zhang et al. [272] testified that the significant welding pressure cause to have acceptable interfacial contact for joining which accordingly reduce the welding strength significantly. Another investigation reported that a drop in the overall strength was detected when increasing the welding pressure significantly, which destroys the TPC substrates and crowds out the molten matrix from the contact area, which becomes delicate [276]. Nevertheless, the holding time of pressure has no effect on the welding efficiency of high-stiffness material [12].

In addition to the parameters mentioned earlier, travel replaces vibrational time with the displacement of the sonotrode, which is considered one of the parameters of the vibrational phase. In travel, as a function in a range of 0%–100%, the thickness of ED will decrease, though the complete melt and maximum squeeze of ED occurred before reaching 100% travel. Figure 56 presents a comparison by SEM of welded CF/PEI between different percentages of travels: 20, 40, 60, and 80% [64]. At a travel of 20%, it is observed that there is a deep fiber imprint on the post-welded ED, and there were no patches of intact ED. Besides, at 40% of travel, broken fiber bundles were observed on the fracture surface. Furthermore, it is noticeable that there were resin flashes and kinking of the uppermost layers of the lower substrate at the edge of the overlap, combined with porosity at 60 and 80% travel. Likewise, at short travels, significant patches of intact ED and shallow fiber imprints on the post-welded ED were observed [64].

Figure 56

SEM detail of CF/PEI joint at 300 N; 86.2 µm; and 20, 40, 60, and 80% travel [64].

For the purpose of examining the fatigue behavior of the carbon/Elium^®, a stress ratio (R) = 0.1 was adjusted at elevated temperature and 5 Hz, while considering the optimum value of LSS. Figure 57 presents the results of LSS and the S–N curve for several cycles of failure for both adhesive and the welded formation [63]. Figure 58 shows the microstructure after mechanical failure with the naked eye and SEM test formation [63].

Figure 57

Adhesive and welded formation results: (a) LSS vs elongation and (b) fatigue test results [63].

Figure 58

The adherends failure at 5.68 MPa: (a) ELC_IED, (b) ELC_FED, (c) adhesively bonded (SAF 30 5), and (d) the SEM of ELC_IED top adherend [63].

Additionally, Figure 59 displays maximum LSS values and load–displacement curves for various welded setups of weldment CF/Elium^®-CF/Ep; with the absence of a coupling layer, a carbon/Ep laminate shows a supreme LSS of 5.02 MPa at a welding condition of 3 s and 5 bars of weld pressure. The weaker link between the adherents, which led to the pure adhesive failure, is the reason for the reduction in the welding strength. Besides, Elium^® composite welded to Ep composites with co-cured Elium^® film (ELF) has revealed significantly lower LSS readings of 3.16 MPa. The failure modes justify the drastic reduction in bonding strength. The film was entirely de-bonded from the Elium^® particle (ELP) rather than being cohesive or between adherents; the failure occurred between the ELF and the ELP adherend [284].

Figure 59

(a) Load vs displacement and (b) the maximum LSS of all the welded configurations [284].

However, based on the observation from this review, the shortage in results obtained by investigating the impact of welding force on welding strength and quality may be due to the reason stated in literature studies. More information about variations in USW parameters and their enhancements on mechanical properties, welding quality, and welding strength are presented in Tables 2 and 4.

3.4 Vibrational amplitude

The impact of vibrational amplitude and welding force on CF/PPS welded by USW has been studied. The results show that a reduction in the cooling rate was caused by setting the low vibrational amplitude and welding force, which leads to slowing down the heat generation. As a result, a substantial HAZ is generated [245]. Moreover, a considerably high welding strength may result by employing high vibrational amplitude, which produced further energy on the weldment, leading to increased strength [285]. Additionally, it has been revealed that the deformation of fibers at the fracture surface decreases with the reduction in the amount of vibrational plausible amplitude [64].

By investigating the fracture surface through SEM after the DLS of CF/PEEK joined with USW by using spot ED, a post visual examination of the welded specimens revealed that the fracture surfaces contained a circular welded junction on the overlap. As illustrated in Figure 60(a), there was no additional damage visible, and the remaining overlap was undamaged. Besides, the SEM investigation confirmed intralaminar failure in light of the excellent bond produced by the welding technique. Such failure mechanism is illustrated in Figure 60(b) and (c) by tearing the outermost laminate ply’s fiber bundles and debonding the fiber matrix. Figure 60(d) shows a further magnification of (c) to provide a more explicit observation of the separation of the fibers from the TPC resin, which denotes the debonding of the fiber matrix [72].

Figure 60

The fracture surface of CF/PEEK at (a) optical and SEM detail, (b–d) for USW spot welding after DLS tests. Welding parameters: 600 J energy, 1,500 N welding force, and 60.8 μm peak-to-peak amplitude [72].

The exploration of crystallinity on the interface of ultrasonically welded CF/PPS joints has been reported by examining the vibration amplitude and welding force effects on the level of crystallinity on the welding interface. The outcomes show that the USW process parameters significantly influenced the degree of PPS’s crystallinity and crystal perfection. Besides, low welding force of 300 N and vibrational amplitude of 51.8 µm increased the crystallinity to an adequate level along with the crystal perfection. However, great welding force of 1,000 N and vibrational amplitude of 86.2 µm primarily formed amorphous PPS and defective crystals. In addition, from the DSC thermograms and the WAXD diffractogram, it was evident that a semi-crystalline structure was achieved in the PPS, even if the crystal perfection and crystallinity degree were not as great as in the original conditions of the PPS EDs (i.e., before USW) Figure 61 shows the impact of USW process parameters on the crystallinity of the PPS ED [245].

Figure 61

The standard DSC measurement of the calculated crystallinity degree of PPS ED films. The ED film with moderate crystallinity processed by low vibration amplitude and low welding force. But predominantly amorphous ED films were produced by the high vibration amplitude and high welding force [245].

The microstructure of CF/PA6 welded by USW without ED using a servo-driven welder has been examined. Besides, the SEM results before and after the welding process are presented in Figure 62. The cross-section of a weldment while being welded under standard conditions using a servo-driven welder is illustrated. Intimate contact between the two surfaces exists after the two workpieces proximity to joint, as shown in Figure 62(a). The fibers were still securely encircled by the matrix during that time because the components at the interface had not yet started to melt. The polymer chains spread over the interface as the temperature rises, and the surface becomes less smooth (shown in the yellow box). As the joint was created under the ultrasonic, as depicted in Figure 62(b), the visible interface among the workpieces vanished, and more fibers were exposed as the interface material melts [248]. In addition, this welding process was run at various conditions of process parameters; the relation between parameters and the weld characteristics using a servo-driven welder is presented in Figure 62(c). Furthermore, research focused on studying the welding quality of CF/Ep to CF/PEEK joined by USW with and without a coupling layer of joining PEEK, taking into consideration two welding parameters, which are vibrational amplitude and welding force, was carried out. The results show that these combination of parameters have an impact on preventing the thermal degradation of Ep resin owing to the short heating time [190]. In the case of TPCs/TSCs welding, a delay in the degradation in TSC adherents may be caused by a very short heating time. The heat which generates adhesion was caused by the reduction in heat transformation from TPC coating to TSCs. Brito et al. [247] found that the peak load increased linearly with the increase in amplitude in joining the CF/PA66 under the effect of moisture on the USW. However, by studying the effect of increasing moisture absorption with increasing amplitude, the welded area and peak load were decreased. According to the authors, the vibrational amplitude may be affected by using misaligned adherents as well as other USW process parameters. However, Villegas [65] stated that the magnitude and the duration of power consumption were affected by the vibration amplitude during all USW process stages; either low/high welding force was utilized.

Figure 62

SEM of CF/PA6 of the weldment cross-section under servo-driven welder: (a) The initial unmelted interface, (b) the welded joint, and (c) under typical circumstances with the following characteristics: 600 J of welding energy, 90% of the maximum amplitude, 0.5 mm·s⁻¹ of plunging speed, and 200 N of trigger force. Positive displacement value indicates that the sonotrode is moving downward [248].

Three welding samples per alignment, up to 0.28 mm, have been examined by Tsiangou et al. [282], to determine the impact of reducing amplitude on temperature at the interface amongst the PEI coupling layer and CF/Ep adherend. The variation in temperature at the interface in between the CF/Ep adherend and the PEI coupling layer, in welding setups with various welding forces and vibrational amplitudes is shown in Figure 63. The temperature raised at an almost identical pace in the 1,200/86 and 1,200/70 configurations regardless of variations in the vibrational amplitude. The temperature generally rose more slowly in the 1,200/60 setup than in the other two configurations. In contrast, the maximum average temperature value was recorded at 800/86 equal to 339 ± 27°C. In addition, it should be noted that the most extraordinary temperatures were mainly attained just after the vibrations were turned off, which is compatible with the theory that the temperatures were caused by heat conduction through the PEI coupling layer. Furthermore, Wang et al. [286] considered the enhancement of welding amplitudes on CF/PEI joint properties, and the temperature measurement field has been analyzed. Acceptable welding quality has been obtained at the optimum amplitude value. However, the PEI resin of the joints’ interface melts significantly when the amplitude increases excessively. More information about variations in USW parameters and their enhancements on mechanical properties, welding quality, and welding strength are presented in Tables 2 and 4.

Figure 63

The variation in temperature at the interface in between the CF/Ep adherend and PEI coupling layer affected by the reduction in amplitude (three welding samples per configuration, up to 0.28 mm in all cases). The vibration’s end is indicated by diagonal arrows. the end of the vibration is indicated by vertical arrows as the end of the welding process [282].

3.5 Impact of pretreatment process (heating time and heat treatment)

The preheating process enhances welding quality and joint quality. Besides, it has an extraordinary benefit whereby preheating for a longer time causes the maximum shear load followed by the reduction due to the porosity generated by the over-welded welding layer [255]. Furthermore, the decomposition in the CFRTP can be eliminated, and the temperature gradient can be reduced by the preheating process for the inter-substrates [239]. Likewise, the fatigue performance and the thermal stress in the joint produced by welding were improved by performing a preheating process before joining with USW [287]. Hargou et al. [254] performed a fractographic study of the laminates of the CF/Ep/mendable polymer, EMMA joined by USW. The results indicate that the EMAA changed from solid to porous filaments, as shown in Figure 64(a). Besides, the shown porosity indicates that the raised temperature in the welding of laminates was sufficient to start the condensation reaction between both the Ep and EMAA phases, which leads to the development of gas-filled pores inside the filaments. However, USW produced substantially fewer and smaller pores compared to oven heating in the EMAA (Figure 64(b)). The reason for this discrepancy was that there was a shorter accessible time through welding for the reaction process to generate volatiles, which permeate from the Ep-EMAA interface (where the reaction happens) into the filaments, where it is essential to condense the volatiles to produce the pores.

Figure 64

(a) Various magnifications of the delamination fracture surface of a mendable laminate (from left to right: 100× – 1,000× – 10,000×). An EMAA filament is visible in the magnified region, and the porosity is shown clearly (Right side) caused by volatile formation. (b) Porous structure in the EMAA phase following the curing by oven heating [254].

As a significant part of the relationship between viscoelasticity and temperature, the preheating treatment directly affects the USW parameters. Consequently, it enhances a superior microstructure characterization and mechanical properties such as tensile test and failure. The preheated CF/PA66 has been investigated, and the impact of various preheating temperatures on the welded area and peak load of the joints in various welding times under 0.15 MPa was inspected and illustrated in Figure 65. As the welding time changed, the preheated and as-welded (i.e., non-preheated) welds both followed a similar pattern. In particular, the welding area first increased and then began to plateau, whereas the peak load first improved with the welding time and then reduced. It anticipated that as the preheating temperature increased, the ideal welding time for the joint with the highest ultimate tensile strength dropped. For workpieces prepared at different temperatures of 25, 75, 125, and 175°C, the ideal welding time was found to be 2.1, 2.1, 1.5, and 1.3 s, respectively [239]. In addition, various preheating temperatures were employed to evaluate the fracture surfaces of the broken workpieces CF/PA66 joined by USW. The outcomes are displayed in Figure 66. The macrostructure and microstructure of the cracked workpiece surface heated to 25 and 125°C were identically compacted. In contrast, some pores were dispersed throughout the workpieces, which were heated to 75 and 175°C. Besides, it has been revealed that the occurrence of porosity in the workpiece reduces the mechanical properties of CF/PA66. As a result, joints with preheating temperatures of 75 and 175°C obtained relatively lower tensile strengths than workpieces with no pores. These properties might be associated with the joints’ temperature history during welding [239]. Moreover, small voids were noticed in the faying interfaces of the workpieces preheated to 25 and 125°C. However, for those treated at 75 and 175°C, the small voids expanded into huge pores marked as porous regions. Many techniques assist in reducing and eliminating the fracture toughness. For instance, the annealing heat treatment process for CFRPPS tapes to CFRPP laminates has been shown to substantially reduce the welded interface’s fracture toughness [288].

Figure 65

The relationship between welding time on peak load and joints welded area of fabricated preheated workpieces under different temperatures [239].

Figure 66

Morphology of CF/PA66 of the fractured surface for upper work pieces with various preheating temperatures after the fatigue test [239].

In a recent study by Qu et al. [289], the uttermost appropriate composite post-treatment method routed between the ultrasonic-assisted reinforcement and annealing methodology has been determined. The study’s primary objective was to enhance crystallinity, prevent and decrease the inescapable voids within the printing procedure, and improve its general mechanical characteristics. Furthermore, increasing the crystallinity may decrease the PEEK fracture toughness [290]. Besides, it has been presented that the fracture toughness of PEEK increased by a drop in crystallinity [291,292]. The tensile test has been performed to explore the moisture impact on the mechanical properties of CF/PA66 joined by USW [239]. The welded interface between the two welded workpieces was examined for dry and wet specimens to compare the moisture effect. It has been revealed that in the tensile test results, the quantity of voids in the weld joint decreased the loading capacity when the water absorption reached a specific point (i.e., above 1.7 wt%). Moisture absorption damaged the weld microstructure together with the composites’ mechanical qualities. As an outcome, the joints created using specimens that absorbed a partially small amount of moisture exhibited workpiece breaking, while the junctions created using specimens that ingested a considerable amount of moisture fractured during the tensile test due to the degraded weld joint.

On the other hand, the microstructure at the faying interface became loose, and the mechanical characteristics declined, even though the weld area of the wet joint fell less than the peak load with the increased water absorption. Consequently, the change in the weld area was independent of peak load [239]. Li et al. [293] explored the enhancement of quantum heating on the USW for joining polymers with double-vibrator parameter. Furthermore, Barkley et al. [294] explored the correspondence of interfacial and surface temperature through disparity spot USW for joining TPCs. More information about variations in USW parameters and their enhancements on mechanical properties, welding quality, and welding strength are presented in Tables 2 and 4.

3.6 General contribution of process parameter in USW

From the literature, many factors play a key role in controlling the welding quality of USW such as preheated or pretreatment methods and process parameters during both vibration and solidification stages. The impact of these factors is given concisely in Table 3 [37].

Table 3

Summary of factors that affect the USW

Note: The (+) sign means high value and (−) sign means low value.

By studying the welding quality of CF/PA6 welded by USW without ED utilizing a servo-driven welder, the process parameters were found to have a direct relation with bonding efficiency and the overall welded area. Figure 67 illustrates the connection between the welding area and bonding effectiveness. It is observed that the strongest bond is produced when the welding region is near the sonotrode contact area. At vibration amplitude of 50–70% and minor amount of welding energy (i.e., 200 J), materials were insufficiently melted when the welding area is small compared with the contact area. The increase in welding force brought on by great trigger force of 1,000 N and plunging speed of 1.2 and 1.5 mm·s⁻¹ would squeeze roughly some molten materials out of the weld region. Besides, when the welding area is larger than the contact area, it results in reducing the bonding effectiveness. Additionally, the welding time and the quantity of melted materials are influenced mainly by amplitude. However, only the first three stages and fewer melted components were visible at low energy and vibration amplitudes, resulting in a relatively limited weld area. In comparison, great energy can melt additional materials. However, because welding takes a long time, excessive melted materials might be squeezed out [248].

Figure 67

The relation between bonding efficiency with welding area and process parameters effects [248].

Welding process factors, weld characteristics, and joint behavior are essential to evaluate the weld quality and are all related. The degree to which a weldment satisfies its practical necessities is the definition of weld quality. Weld attributes are a weldment’s visible traits, while weld performance is a weldment’s internal qualities. The weld quality can be circuitously defined by weld qualities and joint behavior, such as weld area, the morphology of the weld zone, and extreme LSS, founded on the relationship of the previous four sets of variables [250]. Figure 68 shows the correlation between weld quality, characteristics, and joint performance. Overall, the ideal heating time and the optimum frequency in USW were found to be 3–4 s and 20–40 kHz, respectively, in large-scale consideration [295]. An increase in welding strength was reported due to high vibrational frequency, which also increases material vibration [296,297]. Moreover, it is anticipated that USW will always have a very high instantaneous strain rate due to the nature of the process, which involves a very high strain due to the high vibrational frequency. It has been established that a substantial strain rate affects the crystallization of the polymer [298–301], as a further enhancement of the strain rate.

Figure 68

Relationship between welding quality, welding characteristics, and joint performance.