A Review on the Out-of-Autoclave Process for Composite Manufacturing

The out-of-autoclave (OoA) process manufactures composites by applying a vacuum, pressure, and heat outside the autoclave [ 20 ]. The OoA process uses lower pressure than the autoclave and cures composites in an oven or heat blankets. Hence, a special resin system is developed to evacuate voids efficiently [ 21 ]. Though the OoA process is more cost-effective than autoclave curing, the quality of composites manufactured by this process is still inferior to those processed in the autoclave [ 21 ]. This review discusses the common out-of-autoclave processes, their merits and demerits, and their applications. Future research direction on some OoA processes is also presented.

Autoclave processing is typically used for fabricating fiber-reinforced plastic (FRP) composites for high structural applications [ 16 ]. Layers of fibers pre-impregnated with resin (known as prepreg) are stacked on a mold to form the desired component shape. The assembly is covered with different layers of bleeder and breather and then sealed with a vacuum bag. The bleeder helps absorb excess resin squeezed out from the laminate, and the breather creates a channel through which air and volatiles are ejected from the assembly [ 16 ]. The mold-laminate assembly is placed in the autoclave, a large, temperature, and pressure-controlled vessel. The bag is connected to the vacuum system, and a predetermined temperature and pressure (cure cycle) is applied to the laminate. The temperature initiates and sustains the chemical reaction to cure the resin. The pressure compacts the laminate to the desired fiber volume fraction and collapses any void present during curing. In addition, the pressure conforms the laminate to the tool surface. Several models have been developed to simulate autoclave curing for efficient processing [ 17 ]. For example, the cooling and reheating cure models were developed to prevent a thermal spike from an exotherm reaction, leading to partial degradation when curing a thick composite [ 18 ]. Furthermore, incorporating the smart cure monitoring model [ 19 ] has helped optimize the cure cycle in autoclave curing. An optimized autoclave curing is believed to reduce the cost of processing. Though the product of the autoclave process is a high-performance and reliable composite structure, many manufacturers are concerned with its numerous drawbacks. Some of its disadvantages are massive investments, excessive energy consumption, and costly tooling. Consequently, only the aerospace industries can conveniently afford the costs due to safety reasons. Most manufacturers are turning to other alternatives.

The composite material is formed by combining a fiber reinforcement and a binding matrix [ 1 ]. The resulting material is lightweight but has high strength and stiffness [ 2 ]. Composite materials offer exceptional properties such as high thermal stability, flexural strength, damping property, corrosion resistance, impact resistance, and fire resistance [ 3 ]. Due to their exceptional properties, composite materials are useful in various industries such as aerospace [ 4 ], space exploration [ 5 ], construction [ 6 ], automobile [ 7 ], biomedical [ 8 9 ], sports [ 10 ], and marine [ 11 ]. Based on fiber types, composites are categorized as particle reinforced composites, discontinuous fiber-reinforced composites, and continuous fiber-reinforced composites. Composites made of fibrous reinforcements are stronger and stiffer than those made from particulates and are referred to as fiber-reinforced plastic (FRP). In FRP systems, the fiber acts as the load-carrying member, and the matrix binds the fibers together, protects the fibers from abrasion and the environment, and acts as a load transfer medium. Fibers commonly used in FRP systems are glass, carbon, aramid/Kevlar, and boron fibers. These fibers are combined with the polymer matrix in either a chopped or a continuous form. Based on the matrix used, composites are categorized into polymer matrix composites (PMC), metal matrix composites (MMC), ceramic matrix composites (CMC) [ 12 ], or hybrid composite materials [ 13 ]. PMC-wide usage can be attributed to its flexibility in fabricating complex and large shapes. Thermosetting or thermoplastic polymers are used as matrix components. Thermoplastic polymers can be subjected to repeated heating and cooling cycles. In contrast, a thermosetting polymer cannot be reversed after curing [ 14 ]. Commonly used thermoplastic polymers are polyethylene (PE), polypropylene (PP), and polyvinyl chloride (PVC), and examples of thermosetting polymers are epoxy, unsaturated polyester, polyimides, and bismaleimides [ 15 ]. Thermosetting polymers are often used to fabricate polymer-based composites because of their ease of processing.

Prepregs are sheets of unidirectional fibers pre-impregnated with a partially (B-stage) cured resin matrix [ 22 ]. Prepregs are produced by placing fibers between two resin sheets, usually epoxies, and passing the fibers through rollers to achieve complete wet-out. In order to prevent premature curing, the wetted prepreg is wound up and stored in a refrigerator (typically at −18 °C) [ 22 ]. Prepregs are available in varieties of widths ranging from 3 inches to 72 inches, depending on the dimension of the machine used. Their thickness ranges from 0.01 mm to 0.8 mm, depending on the type of fiber form used. Common fiber prepregs include unidirectional tapes and woven and prepreg tows [ 23 ]. Different types of resin used for prepreg manufacturing are epoxies, phenolics, and cyanate esters [ 24 ]. Prepregs are very flexible, which permits them to be shaped to fit a complex mold. Furthermore, prepregs have sticky surfaces due to the partially cured resin, which facilitates the prepreg layers’ stacking and prevents possible movement. Prepregs can be laid either by the manual lay-up process or by automation. A schematic of a typical prepreg is presented in Figure 1 25 ].

Conversely, Khan et al. [ 32 ] did not observe significant results when they applied the DVB technique in the quickstep process to squeeze out excess resin during curing. Cytec cycom prepreg containing epoxy resin was used in the study. Results demonstrated that DVB techniques could remove small resin patches when the application of external pressure is controlled.

Yasir et al. [ 30 ] compared the compatibility and performance of the DVB to the SVB methods for fabricating high-quality composites. The study used carbon prepreg with 58% fiber content. Their results show that laminates fabricated with DVB techniques had better performance than SVB fabricated laminates. The DVB laminates had surface porosity and through-thickness of 0.04% and 0.5%, respectively, while the SVB laminates had surface porosity and through-thickness of 0.1% and 0.7%, respectively. The DVB process provides a better pitting effect via the ballooning effect to allow the evacuation of volatiles during curing than the SVB process. The vacuum bag holds tightly to the laminates and leaves minimal space for entrapped air to be evacuated. Furthermore, it is observed that increasing laminate thickness increases the void content and hinders the evacuation of the entrapped air. In another study, Hou et al. [ 31 ] observed similar results assessing the void content of laminates fabricated by these two methods. Figure 4 illustrates the increased void content observed in laminates fabricated by SVB compared to those manufactured by DVB.

There are two types of vacuum bagging processes, i.e., single vacuum bagging (SVB) and double vacuum bagging (DVB). In the SVB technique, prepregs are stacked between the tool and caul then a vacuum bag is used to seal the assembly. The assembly is installed in an air-forced oven, and heat is applied. At low temperature, usually in stage B, a vacuum is pulled on the inside of the vacuum bag to consolidate the composite. On the contrary, the DVB uses a double vacuum bag to efficiently evacuate volatiles produced during curing. With the normal setup of the SVB, the DVB installs an outer vacuum bag over the assembly with a perforated steel tool in between the two bags [ 30 ]. As shown in Figure 3 , the perforated steel tool prevents the outer bag from collapsing into the inner bag when the vacuum is pulled. A full and partial vacuum is pulled on the outer and inner bags, respectively. The pressure differential results in a “ballooning effect” that aids the evacuation of entrapped air and volatiles. During the final curing stage of the laminates, the outer bag vacuum is purged to atmospheric pressure, and the inner bag is maximally vacuumed to consolidate laminates. Studies have demonstrated the escape of volatiles and entrapped air to be more efficient in the DVB technique than in the SVB technique.

This method uses a vacuum pump to extract the air inside the vacuum bag and then compresses the part under atmospheric pressure [ 27 ]. The resin is squeezed and sucked from the wet laminate into the bleeder (woven polyester fabric). Materials used in the vacuum bagging process are cheap, yet the parts fabricated with this process yield better mechanical properties than hand lay-up. Furthermore, the applied pressure is evenly distributed over the entire surface regardless of the quantity and type of material processed. The effect of evenly applied pressure is a thinner laminate with fewer voids [ 28 ]. Therefore, the process effectively controls excess resin in the laminate that increases the fiber volume fraction. Furthermore, it is a simple process that can use a variety of molds. However, some of the disadvantages of using this process are that with a bigger and more complex lay-up comes more support which increases labor. The process needs to be completed once started without having a break in between. Fiber volume fraction cannot be effectively calculated as other methods, mainly when over-bleeding occurs. The materials used for production in the vacuum bagging process are listed in Table 1 . The vacuum bag technique can be used to fabricate yachts, primary structures such as decks, hulls, superstructures, bulkheads, and secondary structures such as partition panels and interior joint work [ 29 ]. The vacuum bagging process has shown considerable improvements in the mechanical properties of fabricated parts compared to hand lay-up processing. However, hand lay-up parts are inferior to parts manufactured by the vacuum infusion process. The vacuum infusion process will be considered later in this review.

Porosity is formed due to incomplete resin flow into dry areas of the reinforcement. It is commonly caused by increased resin viscosity due to exposure to ambient conditions and low-temperature cure cycles, resulting in an insufficient flow. Three types of voids are identified in the VBO prepreg; these are spherical voids in resin-rich regions, interlaminar voids, and voids within fiber bundles. However, a “super-ambient” cure cycle that dwells at 50 °C to 60 °C for four hours has been found to be effective in achieving low-level porosity [ 48 ]. Dong et al. [ 49 ] minimized the cure cycle time of the VBO process and studied how heating rate, initial cure temperature, dwelling time, and post-cure time affect the final quality of the composite. The authors developed an independent OoA epoxy resin for the study. They observed that a faster heating rate of 5 °C/min results in lower resin viscosity and saves 35% of cure time than a heating rate of 1.5 °C/min. The optimized dwell time of one hour at 120 °C allows enough resin flow time for fibers to be adequately infiltrated by low viscosity resin. Hence, improving fiber/matrix interfacial bonding and leading to enhanced mechanical properties. Furthermore, the authors concluded that an optimal post-cure could be carried out at 180 °C for two hours. Below or above this time duration would reduce mechanical properties or result in extra manufacturing costs. Some of the OoA prepregs are listed in Table 2

The OoA prepreg is consolidated by applying a vacuum at room temperature to evacuate the vacuum bag, compact the part and push out voids in the laminate toward the vacuum source, as illustrated in Figure 6 . As a result, the fiber volume fraction increases, and the in-plane permeability decreases in the prepreg. During this time, the resin flow is limited due to the high viscosity of the matrix. When the part’s temperature is increased, the resin viscosity decreases such that there is a progressive infiltration of the fiber bed by the resin. Resin flows into the dry fiber tows and saturates the interlaminar spaces. Impregnation for OoA prepregs is usually completed at the end of the first temperature ramp. Based on the impregnation rate, the dry evacuated channels are saturated once the dwell temperature is reached. In the last stage of consolidation, the resin undergoes gelation and vitrification, and then the cure is complete.

Several studies have demonstrated the traditional cure kinetic behavior of OoA resins. The resin is cured with an initial increase rate, followed by a decrease due to the diffusion-based cure mechanism. Gelation occurs, and then vitrification of resin takes place during complete polymerization. The OoA resin is more reactive and more viscous than the autoclave resin for a cure cycle of 121 °C dwell time [ 47 ]. The OoA resin systems undergo a freestanding post-cure at 177 °C.

Hu et al. [ 41 ] observed that vacuum quality (between 80–100%) does not affect the final porosity of VBO laminates. However, with 80% vacuum, the bubbles were observed to expand at the intermediate cure. Water molecules were reported to be responsible for this expansion, and they dissolved into the resin before the cure cycle ended. However, increased moisture content increases porosity and prevents the evacuation of air. Porosity between inter-ply is determined by entrapped air that is not evacuated prior to consolidation. The application of an appropriate cure cycle was shown to mitigate defect formation in VBO processing. Park et al. [ 42 ] optimized the cure cycle for VBO prepreg to minimize defect formations such as surface porosity and void content. The authors used a carbon-fiber-toughened epoxy prepreg system for the investigation. They observed that an isothermal dwell at 130 °C for 30 min would cause adequate infiltration of resin into the dry areas of the prepreg. Infiltration occurs when resin viscosity is reduced, and fibers are thoroughly wet, resulting in improved fiber/matrix interfacial bonding. Laminates produced with an optimized cure cycle have similar mechanical properties to autoclave cure laminates. Conversely, Yoozbashizadch et al. [ 43 ] held the isothermal cure and post-cure temperatures for eight hours at 181 °C and 211 °C, respectively, of a carbon fiber BMI prepreg system and attained an optimal ILSS value for a VBO prepreg. However, using an eight-hour cycle each for a cure and a post-cure will amount to a long cure cycle, increasing the cost of manufacturing. Therefore, a shortened cure cycle would be economically competitive for VBO processing. Hyun et al. [ 44 ] reduced the cure cycle duration for VBO prepreg processing and still maintained improved part quality. Cycom 5320-1 epoxy system was used for the experiments. The study showed that applying an isothermal dwell at 60 °C for two hours will produce parts with no wrinkles and less porosity than a 16 h RT vacuum hold [ 45 ]. Entrapped air is efficiently removed at a moderate temperature when evacuation channels are not yet collapsed. In addition, they observed that eliminating the intermediate dwell at 121 °C for two hours shortened the cure cycle yet improved the mechanical properties. The question then becomes if a shortened cure cycle will improve mechanical properties for variously shaped parts. In this light, Mujahid et al. [ 46 ] investigated how different curing profiles interact with other VBO processing parameters, such as bagging techniques and laminate structures, to minimize defects. A 58% fiber-rich OoA epoxy prepreg was used in the study. The modified single vacuum-bag-only technique exhibited fewer defects and thickness variations in the fabricated part than other bagging methods. The laminates fabricated with the convex mold had better quality than the concave mold. However, concave parts can be improved by increasing the local curvature angles and corner radii of the mold.

Vacuum bag only (VBO) curing is an out-of-autoclave (OoA) technique for processing composite laminates. It is performed in a contemporary oven without external pressure, such as the autoclave, to consolidate the laminate. In the absence of elevated pressure, it is important to consider the OoA resin property, fiber bed architecture, and prepreg system. The OoA resin is a slow cure kinetics and low cure temperature matrix system. Figure 5 shows the manufacturing assembly of a vacuum bag only composite, with its consumables. The OoA prepregs are characterized by a partially impregnated microstructure that presents in-plane permeability, which permits air evacuation and aids the manufacturing of low-porosity parts without using autoclave pressure [ 33 ]. The partially impregnated microstructure includes dry spots and resin-rich regions. Low pressure of 0.1 MPa is available for consolidation during cure, and it is insufficient to prevent void formation [ 34 ]. Therefore, the entrapped air, moisture, and other volatiles in the laminate must be evacuated before the resin gels. As a result, the dry regions in the partially impregnated microstructure form an internal network that facilitates gas exit during the initial low-temperature stage of cure. At high temperatures, the dry areas are infiltrated by resin from the resin-rich region. Repecka and Boyd [ 35 ] reported that partially impregnated prepregs resulted in a void-free panel, while fully impregnated prepregs led to over 5% void content. An impregnation level of 60% has been found to produce void-free panels. However, Ridgard [ 36 ] highlighted that the degree of impregnation should be considered regarding resin viscosity, cure cycles, and laminate quality. Yang and Young [ 37 ] demonstrated that the degree of saturation of a VBO prepreg affects the mechanical properties of laminates. The laminates were made with epoxy resin. Fully impregnated carbon fiber and dry fibers were assembled as hybrid laminates, and different degrees of saturation were defined; over-saturation, saturation, and undersaturation. Laminates with over-saturation exhibited similar mechanical properties as those fabricated with the autoclave. For over-saturation to occur during VBO processing, conditions should favor the impregnation rate. Centea et al. [ 38 ] demonstrated that the thermal gradient of a partially impregnated prepreg affects the rate of impregnation and gas transport during consolidation. The Cycom 5320-1 epoxy system was used for the investigation. Porosity distribution is shown to be influenced by the thermal gradient. Areas with hotter-than-average temperatures prevented air from evacuating the laminates. The study reported that resin flow, permeability, bubble transport, and temperature evolution affected air evacuation. Other parameters such as prepreg formats may affect laminates produced with VBO prepregs. Maguire et al. [ 39 ] investigated the importance of prepreg formats and the manufacturing method for VBO prepregs. Manually applying epoxy powder may lead to non-uniform powder distribution, which could produce better laminate uniformity. The study confirmed that epoxy powder prevents an exotherm reaction in thick composites. However, the temperature cycle and latency of the epoxy powder need to be optimized for the best results. How the epoxy powder propagates heat within the VBO prepreg was unclear to the authors; hence further investigation is required. In another study, Edward et al. [ 40 ] designed a unidirectional semi-prepreg that improved the robustness of VBO processing. A toughened epoxy resin was used. The semi-prepreg was customized to discontinue resin distribution. As a result, through-thickness permeability was improved, which facilitated gas evacuation. Laminates produced by the semi-prepreg had fewer defects than those produced by conventional VBO prepregs. The resin feature morphology was observed to be critical in defect formation.

The kinematic drape model [ 59 ] is used to predict fiber shear deformation and its effect on fiber volume fractions. Commercial software such as FiberSIMand laminateTool [ 60 ] provide the draping angle for the lay-up; this helps adjust the flow pattern and predict the time to fill due to the draping fabric. Macro void occurs when resin flow arrives at the vents before the preforms impregnation is completed. Macro void is likely to happen when air is present in the preform and resin pressure is not high enough to collapse the void. Macro voids can be eliminated by bleeding the resin via the vent by allowing sufficient time or by using process control monitoring. RTM process parameters such as resin characteristics, fiber preform, resin preheated temperature, injection pressure, gating method, mold geometry, mold temperature, and vacuum assistance have been studied and identified to influence the quality of parts molded using RTM. Combining the peripheral gating system and the vacuum resin transfer method results in a shorter injection time, minimizes void content, and increases fiber volume fraction and flexural strength compared to combining the positive pressure injection method and the peripheral gating system [ 61 ]. The radial gating arrangement decreases permeability which resists the resin flow resulting in a differential macroflow and microflow, leading to void formation [ 62 ]. Cevdet et al. [ 63 ] found 2 atm to be the optimum injection pressure and resulting in the highest flexural strength and modulus for injecting resin in the RTM technique. However, increasing injection pressure beyond 2 atm decreased the mechanical properties of the panels. The decreased mechanical properties were attributed to the difference in microflow and macroflow, leading to non-uniform resin flow and void formation. Increased injection pressure leads to fiber misorientation, reduced fiber volume fraction, and mold filling time. Moderate resin flow rate, low binder concentration on both sides of the fabric, and vacuum assistance have been reported as the most favorable parameters in fabricating composites for both injection and compression high-pressure RTM [ 64 ]. Compression pressure and increased temperature effectively reduce the void content [ 65 ].

In RTM, some manufacturing issues include race tracks, deformation of fiber structure, and macro void formation. The region close to the fiber walls has higher permeability; therefore, the resin race along the path of high permeability leading to race tracking [ 52 ]. Race tracking occurs along the mold edges, along the ribs of bends [ 53 ], and on the lines of injection gates. In RTM, when the fiber is draped over the mold surface, fiber orientation in the preform is altered; hence permeability to resin flow is changed. The extent of deformation and permeability change is defined by fabric type and the radius of the mold curvature. There are four fiber deformation mechanisms: inter-fiber (intra-ply) shear, inter-fiber slip, fiber buckling, and fiber extension [ 54 ]. Deformation can cause fiber misalignment that leads to mechanical properties degradation. Vallon et al. [ 55 ] observed a 9% and 22% reduction in stiffness and strength, respectively, for a fiber misalignment of 5°. Li et al. [ 56 ] also found a 4% reduction in elastic modulus for every 1° of fiber misalignment up to 20°. Hsiao and Daniel [ 57 ] observed a 50% and 70% reduction in compression stiffness and strength, respectively, in S-glass/epoxy composites for a wrinkle value of 0.2. Kugler and Moon [ 58 ] observed a 20% reduction in a carbon/polysulfone laminate wrinkling with a lowering cooling rate from 20 °C/min to 2 °C/min.

The RTM process involves using a closed mold to fabricate a composite part. Figure 7 presents the various steps in the RTM process. Fiber preform is cut according to the mold shape and placed in a closed mold cavity [ 50 ]. A low-viscosity thermoset resin is injected through the injection port into the mold cavity, usually with a 3.5–7 bar pressure. The injected resin impregnates the preform evacuating entrapped air bubbles until complete wetting is reached. Once the resin starts exiting from the vent ports, the resin injection is stopped, and vent ports are closed. The resin is allowed to cure by heating the mold or the initial addition of inhibitors to the resin system. After the resin is cured, the mold is opened, and the part is de-molded. Some variants of the RTM process are VIPR, FASTRAC, light RTM (LRTM), structural reaction injection molding (S-RIM), and co-injection resin transfer molding. Some advantages of RTM are that the process can produce parts with close dimensional tolerance and an improved surface finish. Parts made by RTM have a high-volume fraction of about 60–70%. RTM can manufacture complex-shaped composite parts. Consistent reproducibility of composite parts can be achieved using the RTM process. Due to high resin pressure and faster mold opening and closing, a fast-manufacturing cycle is reached, further improved by process control. Some drawbacks of the RTM process are the limited size of parts that can be manufactured. Fiber wash can occur due to high resin pressure and loose fiber compaction. Furthermore, improper location of injection gates and vents can lead to a macro void in the composite [ 51 ].

6. Vacuum Assisted Resin Transfer Molding (VARTM)

In the VARTM method, the reinforcement is placed on a one-sided mold and sealed with a vacuum bag to form a closed mold. A vacuum is applied at the vent, which drives the resin under atmospheric pressure to impregnate the reinforcement while evacuating the air bubbles and compacting the fiber preform ( Figure 8 ). The resin flows through the porous preform and arrives at the vent. The injection is closed, but the vacuum is maintained until the part is completely cured and de-molded. The VARTM process is used to produce large composite parts at a low cost with a low production volume [ 66 ]. This process is widely used in the energy, aerospace, marine, defense, and infrastructure building industries [ 67 ]. Variations of VARTM have been invented to cater to the manufacturing of complex parts with better quality at a reduced cost. The VARTM process has some advantages: flexibility of mold tooling and selection of mold materials [ 68 ], resin and catalyst can be stored separately and mixed before infusion, low emission of volatile organic compound (VOC), and visible inspection of the process to identify and manage dry spot occurrence [ 69 ]. However, some drawbacks of this process are that consumables such as sealing tape, peel-ply, and vacuum bags may not be reusable. The low resin injection pressure can limit void compressibility resulting in high void content and low fiber volume fraction. The process may be susceptible to high chances of air leakage, depending on the operator’s skill level [ 70 ].

u = − K μ ∇ P

(1)

u

is the Darcy velocity,

μ

is the dynamic viscosity of the fluid,

P

is the fluid pressure, and

K

is the permeability of the stationary porous media. Equation (1) quantifies the relation between the Darcy velocity and resin-saturated pressure in a porous medium. Liquid resin in a solid porous medium requires the mass flow continuity for incompressible fluid and the solid, which is stated as

∇ · U ¯ D = 0

(2)

The basics of the VARTM process are the resin flow phenomenon, fiber preform compaction, and resin viscosity [ 71 ]. The resin flow in the VARTM process is treated as flow-through anisotropic porous media and can be modeled by Darcy’s law, represented as Equation (1).whereis the Darcy velocity,is the dynamic viscosity of the fluid,is the fluid pressure, andis the permeability of the stationary porous media. Equation (1) quantifies the relation between the Darcy velocity and resin-saturated pressure in a porous medium. Liquid resin in a solid porous medium requires the mass flow continuity for incompressible fluid and the solid, which is stated as

0 = ∇ · k μ · Δ P

(3)

Equations (1) and (2) are combined to create a resin-saturated porous medium, which is stated as

When a boundary condition is assigned to the resin-filled porous domain, the pressure distribution inside the resin-filled domain can be solved by Equation (3). Equation (1) can then be used to solve the Darcy velocity distribution in the resin-saturated porous medium domain.

In the VARTM process, predicting the mold filling process is useful for determining major processing parameters and design windows. Hsiao et al. [ 72 ] proposed a two-dimensional analytical solution that uses dimensionless analysis to divide the resin-saturated porous medium domain into the saturation and flow front regions. The authors observed that flow front velocity in the VARTM process decreased significantly as the saturated region length increased. However, this effect can be minimized using a thicker flow distribution medium with higher permeability and preforms with higher permeability. The flow process is designed by determining the location of the injection gates and vacuum ports, sizes, and locations of the distribution lines, the number of layers, types and locations of flow distribution mediums, and finally, the timing to open and close the gates and vents. In VARTM, the preform compaction pressure is the difference between the atmospheric and vacuum pressure (local pressure) inside the fiber preform. The fiber preform compaction affects the resin infusion process due to the change in preform permeability, thickness, and porosity. However, the influence on the thickness of the final part may not be too obvious if adequate relaxation time is not allowed for compaction pressure to be evenly distributed in the vacuum bag after the injection gate is closed. Longer duration between the resin filling and the resin gelation point will allow a complete relaxation process, thus improving thickness uniformity in the VARTM part. Bekir et al. [ 73 ] investigated the effect of compaction pressure and resin flow on part thickness variation in the vacuum infusion process. A polyester polipol matrix system was used. The study demonstrates that the preform is compacted effectively due to the lubrication effect with the initial resin injection. However, as the flow front advances, the compaction pressure is reduced, increasing laminate thickness. The authors, therefore, reported the duration of initial vacuuming and gelation, resin pressure, compaction pressure, and resin shrinkage ratio to determine the part thickness. Yacinkaya et al. [ 74 ] studied the effect of compaction pressure (CP) and infusion pressure (IP) in the fabrication of laminate panels using the pressurized infusion (PI) technique illustrated in Figure 9 . The authors used an epoxy matrix in the study and observed that an increase in compaction pressure (CP) reduces the porosity and permeability of the fiber preform. While increasing the infusion pressure (IP) increase the porosity and permeability. Both CP and IP increase the fiber volume fraction and interlaminar shear strength and decrease the void content. Increasing the CP reduces laminate thickness and increases the fiber volume significantly. Increasing the CP and IP together reduces the void content.

Furthermore, Yacinkaya et al. [ 75 ] investigated the synergetic effect of the external pressure coupled with resin flushing to enhance the quality of fabricated composites in a heated-VARTM system. A glass/epoxy system was used in the study. Applying external pressure alone decreases the average void content from 24% to 1.4%. However, coupling the external pressure with resin flushing further reduces the void content to 0.86%. Apart from pressurized air, permanent magnets have been used to generate compaction pressure. Maya et al. [ 76 ] used a magnet-assisted composite manufacturing technique to generate consolidation pressure during cure. The authors observed increased flexural properties, fiber volume fraction, and low void content. Kedari et al. [ 77 ] studied the effect of vacuum pressure, inlet pressure, and mold temperature on the VARTM process. Mold temperature, inlet pressure, and vent vacuum pressure were varied, fabricating different polyester/E-glass fiber composite system samples. Experimental results suggest that increased mold temperature and vacuum pressure at the vent increase the fiber volume fraction in a VARTM system. However, void content can considerately increase if the inlet pressure of 1.013 bar is not appropriately modified when mold temperature is increased. The authors identified vent pressure, inlet pressure, and mold temperature as the major factors influencing void formation. Kedari et al. [ 77 ] reported the possibility of minimizing the microvoid content and increasing the fiber volume fraction by controlling mold temperature and resin inlet pressure. Controlling laminate thickness is critical in the mold and post-mold filling stages. The post-filling compaction relaxation process is influenced by the preform and fiber system, resin viscosity and cure kinetics, mold temperature, and the type and arrangement of the flow distribution network. Extending the resin processing window and keeping injection gates closed during the post-filling stage can improve uniformity in part thickness. Furthermore, increasing the mold filling speed and using resins with longer gel times can help control part thickness. Yacinkaya et al. [ 78 ] compared the compaction pressure between the VARTM and RTM methods. Increasing compaction pressure was found to decrease part thickness. However, changes in resin viscosity affected how compaction pressure changed. Resin viscosity, usually less than 1000 cP, is critical for mold filling, fiber preform compaction, and curing in the VARTM process. Mold temperature can be effectively used to control the resin viscosity of a high-performance resin system. Mold temperature selection is influenced by the type of mold material, resin gel time control, resin viscosity, resin curing management, flow medium distribution material, and peel-ply [ 79 ]. The mold is heated during the mold filling process to reduce resin viscosity and increase the processing window of the resin.

The challenges of the VARTM process are air entrapment, thickness and fiber volume fraction uniformity, curing and thermal management, and spring-in. Low vacuum can lead to the entrapment of air inside the composite part when resin flow fails to displace air; hence dry spot formation occurs. Optimizing the mold filling design can solve the problem of air entrapment. Dry spots also occur due to the slow mold filling process. Increasing the number of flow distribution layers, injection ports, and vents can be used to mitigate this problem. Hsiao et al. [ 80 ] designed and proposed a distribution media layout to manipulate the flow front, reducing race track. Reducing the resin curing rate or increasing the mold temperature can also be helpful. Leakage in the vacuum bag, resin supply line and sealing tapes can cause dry spots. Incompatible dual scale flow behavior between resin flow in the fiber tow and between the fiber tow can result in microvoid, another form of air entrapment. VARTM curing can lead to a potential thermal degradation in the thick parts. To avoid thermal spiking in a VARTM, White and Kim [ 81 ] proposed multi-stage curing (MSC). The MSC technique prevents thermal degradation, affecting the interlaminar fracture toughness and the interlaminar shear strength of composite parts.

The MSC process uses a combination of various “stage VARTM processes” that can only cure manageable layers of composites at a time to fabricate the thick laminate. The final composite thickness is different due to resin thermal contraction and volumetric shrinkage. Residual stress or strain occurs in the laminate due to resin cross-linking shrinkage and fiber and matrix mismatched thermal contraction. This residual stress–strain can cause dimensional problems called spring-in. Spring-in is the inward bending of the curve-shaped laminate part caused by curing. Laminates shrink in the thickness direction due to thermal contraction and cross-linking shrinkage, depending on the matrix. The fiber maintains the in-plane dimensions during the curing process. During the curing of the curve-shaped laminate part, the non-isotropic dimensional changes in the in-plane and thickness direction will result in the further inward bend of the laminate after de-molding.

VARTM is used to manufacture bio-based composites consisting of cellulose fiber mats and oil-based resins. The VARTM process has been identified as a promising method for manufacturing nano-enhanced FRP. The VARTM process reveals a unique, through-thickness, mold-filling flow pattern. The through-thickness flow reduces the nano-modified resin’s traveling distance. Furthermore, the change of nanoparticles is filled by the fiber preform. Fan et al. [ 82 ] reported a 0.5 wt. % of MWCNT composite made with the VARTM process had CNTs aligned in the through-thickness direction, which improves the laminates’ mechanical properties.