Introduction

Composites are widely used in several engineering applications because of their beneficial features (like higher strength-to-weight ratio and good stiffness) and capacity to endure the inherent constraints encountered in conventional materials [1,2,3,4,5]. Strips of fiber are stacked in layers and bound together within a matrix to generate laminate composite structures in numerous applications [6, 7]. The majority of carbon fiber-reinforced polymer (FRP) composites are made up of a strong matrix material reinforcing with higher strength filaments and non-reinforcement fibers. The right mix of this kind of material yields mechanical characteristics that are both designable and exceptional. These composites typically have a higher strength-to-weight ratio and elastic constant, and also excellent geometric rigidity and corrosion resistance [8,9,10,11]. Carbon fiber-reinforced polymer (CFRP) composites are popular in the advanced industrial fields, such as aerospace, automotive, and medical applications, because of their practical advantages and reducing manufacturing costs of carbon fibers [12,13,14,15].

CFRP composites have less density, strong damping capability, better geometrical stabilities, and improved corrosion resistance, along with their advantageous properties of strength to weight ratio and stiffness. Molding, vacuum bagging, compression molding, filament winding, and hand layup laminating processes are all being used to manufacture CFRP parts that are ready to shape.

The first step is to add a releasing layer to the mold so that the completed boat and mold may be readily separated. The gelcoat stratification is the next step. By providing a clean outside surface, this covering protects the boat from outside environmental elements and extends its life. Normally, the thickness of the gelcoat is 0.5 mm. During the third process, the skin coat is applied. The skin coat is manually laminated. It is strongly recommended that this technique be carried out with attention to avoid print-through, an aesthetic fault that is undesirable for a recreational craft. It reduces the sharp angles by ensuring that there are no gaps between the first layer of the laminate and the gelcoat. In this phase, the layers of fibers, which are frequently built in the shape of a mat, woven cloth, or roving, are dispersed by hand over the skin coat, and the resin is injected, sprayed, or sprinkled physically over the previously stretched plies. The operator uses the hand roller to help the resin penetrate the plies and eliminate gas bubbles. A core adhesion agent is then applied to the core, and the final coatings are manually laminated over the core material. Following that, the curing phase started as a result of a response between the catalyst and the resin. Without generating heat, this process hardens the fiber-reinforced resin composite [16].

The vacuum bagging technique is an enclosed mold method for producing greater capacity, large-scale fiber-reinforcing polymer composites with minimal equipment cost. Vacuum bagging was developed to produce high-quality, large composite materials. It employs pressure to suit the particular requirements. The advantages of vacuum bagging include the capacity to produce large and complicated composite components with high standards, as well as customizable mold equipment design and mold material selection. Vacuum bagging, which is equivalent to the hand layup method's open mold, is easily adaptable to various part geometries. One disadvantage of vacuum bagging is that the vacuum bag, flow dispersing medium, peel ply, seal tape, and resin tubing cannot be reused. The possibility of air leaks is great, and it is heavily reliant on the laborer's competence. Between the environmental pressure (e.g., air pressure) and the vacuum, the resin injection pressure is limited [17].

Using a hand layup process followed by the vacuum bagging methodology, the research article of Muralidhar et al. [18] based on bidirectional carbon fiber-reinforced composites (CFRPs) reveal the effect of fiber design on mechanical characteristics. The mechanical characteristics were studied using 3 different fiber configurations: T800CFRP, T700CFRP, and T300CFRP. All of the composite structures were made using the same amount of carbon fiber and the same consolidating method. To understand how the mechanical behavior of composites varies with fiber design, researchers looked at tensile properties, flexural modulus, interfacial shear strength, impact resistance, and hardness. In terms of mechanical characteristics, the impact of fiber weaving design is more apparent in CFRP composite structures than in pure ones, according to findings. T800CFRP composite structures have a very strong strengthening impact on both tensile and bending load responses of CFRP composite structures, according to tensile and flexural characteristics. The fiber design of the CFRP composite structure has a minor impact on density but a significant impact on strength and ductility. The effectiveness of composite structures was determined by the inherent characteristics and weaving architecture of carbon fiber. When compared to the hand layup approach, the vacuum bagging technique enhanced tensile strength, elastic/shear modulus, impact resistance, and hardness, but the shear strength values were almost equal. The work Sang-yong Kim et al. [19] intends to examine the mechanical characteristics and fracture causes of glass fiber-reinforced polymer composites created by employing hand layup-treated outside and vacuum infusion-treated inner layups, which provide structural integrity manufacturing simplicity. 1, 2, and 3 vacuum infusion treated layers’ sets are combined with consistent sets of hand layup processed layers in the glass fiber reinforcing polymer composites. Tension, compression, and also shear qualities are among the mechanical characteristics examined in this study. The tensile mechanical qualities of composite materials with 3 pairs of vacuum infusion layers were the best, whereas those with 2 pairs had the best compressive mechanical properties. The batch uniformity in the glass fiber reinforcement composite production method is assessed using experimentally determined mechanical characteristics. The multilayer hybrid fiber-reinforced composite components were efficiently designed and produced, showing the mechanical characteristics of tension, shear, impact, and hardness by Karthik et al. [20] and finding the scientific proof that is followed was an interesting discovery. The multilayer composite structures stacked sequence C-C-K-C-C specimen S4 showed good mechanical characteristics with tensile resistance of 170 MPa and shear resistance 46 MPa, the standard pure carbon fiber has 180 MPa and shear resistance of 64 MPa. The impact resistance of 18 J and hardness of 48 HV of the hybrid composite C-C-K-C-C stacked sequence are disclosed, and the pure carbon fiber standard specimen S6 is generated at 20 J. Surface morphology and cracked mechanisms of samples were investigated using hybrid composite multilayer microstructural characterization. Various fiber-stacked sequences and adhesion strength were investigated in order to determine them. Bharath et al. [21] studied tensile characteristics and erosive nature of calcinated aluminum oxide on glass fiber reinforcement composites made using a vacuum bagging route and discovered the following key factors. When the vacuum pressure is exerted while composite manufacture was increased, the composite rose. The tensile characteristics and erosive behavior of the polymer composite parts have been altered by the inclusion of aluminum oxide particles. With a small sacrifice in tensile properties, a considerable portion of erosion resistance was detected. Using scanning electron microstructures, it was discovered that the specimens made with a mechanical stirrer had a considerable amount of particle retention. The quasi-static mechanical performance of 3 innovative Bouligand-based composites was compared to a typical aviation laminate using a vacuum bag infusion production process by Amorim et al. [22]. For all combinations, a microstructural and physical examination was undertaken initially, followed by evaluation and discussion of their mechanical response under tensile, 3-point bending, and interlaminar shear resistance testing. Although no substantial cavities were found in the laminates, both Bouligand-kind designs had a lumpy and poorly defined interlaminar region. Bioinspired designs showed more same to eighteen percent greater elastic and shear constant when compared to standard layup during quasi-static mechanical stimuli. The progressing and translaminar breakdown mechanisms that are common in these systems, however, resulted in a decreased load-bearing capacity. Despite having identical interfacial resistivity to the standard, finite element models demonstrated a high connection between the cross-sectional stress field and the empirically observed mode of failure. Vinay et al. [23] used the hand layup technique for making glass and carbon fiber-reinforced polymer composites and studied the shear strength, impact resistance, and hardness. They stated that the carbon fiber reinforcement composites showed better mechanical properties than glass fiber reinforcement composites. They also reported that because of higher mechanical strength, hardness, and deflection resistance, the shear strength and energy absorption capacity of carbon fiber-reinforced composites were found higher than the shear strength of glass fiber reinforcement composites. But, as compared to carbon fiber-reinforced composites, the hardness was found higher in glass fiber reinforcement composites due to strong binding between the laminates. In the study of Geston et al. [24], unidirectional flax fiber composite components were made using simply vacuum bagging technique and vacuum bagging technique accompanied by compressive molding under various compressive stresses. The composites' surface finish was considerably improved by applying external forces, which reduced the area of superficial holes from thirty percent to between three and five percent, relying on the operating pressure. This permeability was shown to diminish the composites' flexural strength marginally, although this influence was not substantial when compared to the impact of modifying the fiber volume fraction by applying various pressures. In a broad sense, as the external pressure that was utilized in the treatment and prevention of the laminates increased, the mechanical properties (shear and tensile) enhanced, and that was primarily due to the improved fiber-to-resin ratio as well as to the enhanced microstructure with fewer cavities acquired because when compressive molding phase was conducted on the vacuum bagging laminates. However, in contrast to the impact generated by modifying the fiber content, the impact of these faults on the characteristics was not as severe. As a result, the enhancement in fiber volume fraction appeared to be the primary driver of the improved characteristics.

From the literature studies, it was observed that many studies focused on the microstructural and mechanical properties of fiber reinforcement composites fabricated by hand layup method and vacuum bagging method separately. However, the comparison between the fiber reinforcement composites made by the hand layup method and the vacuum bagging method was given less attention. The present there focused on the preparation of carbon fiber-reinforced composites by using vacuum bagging mold technique and hand layup technique and comparing the microstructural features and tensile properties between them.

Experimental Methodology

Materials

Reinforcing agent: The woven bidirectional (2 × 2) twill carbon fabric was used as a reinforcing agent. Carbon fiber was chosen as the principal reinforcing material due to its unique mechanical qualities, such as high ultimate tensile strength, elastic modulus, and percentage elongation. Carbon fiber's qualities make it ideal for structural components that require strength and rigidity.

Because of its better mechanical qualities, notably in terms of tensile strength and modulus, carbon fiber was chosen above other alternative fibers. The thickness of the carbon fabric was estimated to be 0.45 mm. The fibers were supplied by R. V materials & alloys, Bengaluru, India.

Matrix: Huntsman’s epoxy resins, which are known to be Araldite LY-1564 (appearance—clear yellowish liquid, viscosity-1150 MPa at room temperature, density-1.15 g/cc, and mix ratio-100 wt. %) were utilized as matrix materials in the current study, and the hardener used was Aradur-22962 (appearance—colorless with little yellowish liquid, viscosity—6–19 MPa at room temperature, density—0.85 g/cc and mix ratio-25 wt %). It was supplied by R. V materials & alloys, Bengaluru, India.

Releasing agent: QZ-13 was preferred as the mold release agent that acts as a protective coating on the mold, and it was supplied by Huntsman’s Petro-Chemical industries.

Preparation of Composites

The resin-to-fiber ratio has been maintained at a 1:1 ratio, while the resin-to-hardener ratio has been maintained at a 10:1 ratio by weight. Various stacking sequences of carbon/epoxy composites (0/90) are fabricated using two different processes. The composites were fabricated using the manual layup method and the vacuum bagging mold process. Every material system is composed of lamina, which is considered a set. This method of shipping and flexible stitching enables the fabric simpler for the fiber producers to manage and takes a lesser duration to layup. The hand layup specimens were combined using the traditional hand layup technique. Manually placing carbon fibers into a mold, then properly distributing matrix material across the fiber layers. Squeegees and rollers were then used to release any trapped air. Then, vacuum was used to disperse the resin uniformly throughout the fiber and assist in removing air bubbles to lessen gaps and structural incompatibilities. All specimens underwent room temperature hardening of the fiber or matrix system.

The weight percentage ratio for combining epoxy resin and hardener was 100:25. The entire layup was vacuum-bagged and given a 24 h cure at ambient temperatures. Following the completion of the mold, the cure was carried out in an air-circulated oven at a temperature between 90 and 100 °C for 1 hour. The prepared laminate has a dimension of (300 mm × 215 mm × 3 mm).

The present research concentrates on the construction of the fiber weave while taking ostensibly same wt% and similar consolidating techniques into consideration. The reinforcements to matrix ratio were kept at 60:40 for all composite materials. It was necessary to build a new carbon fibrous structure having enhanced flexural capabilities because polymer matrix composites utilized in constructions frequently fracture in flexure. Almost all of the time, fiber construction and interface quality control the stress that is transmitted, the elasticity from the matrix to the fiber, and the stress that can resist the capability of the fiber. The photographs of the carbon fiber-reinforced epoxy polymer composites made by vacuum bagging mold technique and hand layup technique are shown in Fig. 1.

Fig. 1
figure 1

Photographs of the carbon fiber-reinforced epoxy polymer composites made by vacuum bagging mold technique and hand layup technique

Scanning Electron Microscope

The specimens required for microstructural analysis were reduced to the necessary dimensions. Using the standard metallography procedure commonly employed for PLA-CRPF composites, the extracted specimens were subsequently prepared for microstructural analysis. The preparation steps for SEM analysis involve cutting the sample, mounting it, and applying coatings. Subsequently, the samples underwent examination using a scanning electron microscope (SEM). Subsequently, the scanning electron microscope (SEM) was employed to analyze the fracture samples.

Tensile Testing

The tensile testing for both the composites was done as per the ASTM-D3039 standards (the standards for composites) and the testing was carried out using a 100 kN capacity computer-controlled universal testing machine (UTM) with a crosshead speed of 10 mm/min and 5 mm/min, respectively. The load and elongation curves obtained from tensile testing were transferred to the stress–strain plot, and tensile properties were extracted from those plots. For tensile testing, three samples were extracted from each condition and averaged the tensile properties excluding the properties of defect specimens. Figure 2 shows the dimensions of the tensile test specimen.

Fig. 2
figure 2

Dimensions of tensile test specimens

Results and Discussion

Microstructural Comparison

A microstructural examination of both processing samples was carried out to help understand the discrepancies in the evaluated characteristics between the laminates and the local changes in the observed changes. The microstructures of the samples processed with both the vacuum bagging mold technique and hand layup technique observed under the scanning electron microscope are shown in Fig. 3. It was identified from the microstructures that the majority of the resin in the hand layup specimens was found to be on the surfaces, but the resin was not present on the fiber portion. The lack of composites between fiber and resin in the sample's center has an impact on its tensile characteristics. It influences the tensile properties of the specimens as the center has not created a composite structure between the fiber and the resins. It seems that the glue and fiber have already been bound together during the vacuum bagging mold technique. The gaps between the fibers are still clearly evident.

Fig. 3
figure 3

Cross-sectional SEM images of carbon fiber-reinforced epoxy polymer composites prepared by vacuum bagging mold technique and hand layup technique

The fabric-based composite created using the vacuum bagging mold process exhibits better quality upon visual inspection. It is completely transparent, with no discernible opacification. More precise microscopy reveals air-filled "compartments" that are nearly closed. They appear at the intersection of the fiber bands, right adjacent to the surface. They develop as a result of the gelatin resin "collapsing" under the influence of shrinking. The resin layer that was initially attached to the bag is still there, attaching to the preforms surfaces and curing there to close the "compartment." Such flaws obviously have a negative impact on the laminate's mechanical characteristics and can act as a source of crack initiation. However, because the "compartments" are dispersed consistently across the entire surface of the device, they have an evenly negative impact across the entire sheet surface and so do not regionally affect the material characteristics or the overall dispersion of results. The microstructure of the vacuum bagging mold technique-made fabric lamina in cross section displayed great layers’ adherence without discernible under-saturated patches. Compare hand layup microstructures to see the structural differences between samples taken from the panel’s segment closer to but far from the suction edges. As you go closer to the suction edges, there is a clear decline in layer thickness. In addition, the examples cut far from the suction edges exhibit pools of resin that are discernible at the laminate's top surface, in spaces between the fiber filaments.

Comparison of Tensile Properties Between Vacuum Bagging Mold Technique And Hand Layup Technique

The stress–strain plots of the two different samples made by the vacuum bagging mold method and hand layup technique are displayed in Fig. 4. The extracted tensile data such as ultimate tensile strength (UTS), elongation (EL), and elastic constant are documented in the graphs. From the tensile date, it was identified that as compared to the samples processed with the hand layup technique, the samples processed with the vacuum bagging mold technique resulted in better tensile strength characteristics. Within the vacuum bagging mold process, the sample-2 has highest tensile strength and less ductility as compared to sample-2, on the other hand, within the hand layup technique, the sample-1 has highest tensile strength and less ductility as compared to sample 2, which could be attributed to the resin fiber ratio and resin ratio. The comparison of tensile properties between vacuum bagging mold technique and hand mold technique are explained as follow.

Fig. 4
figure 4

Stress–strain plots of the two different samples made by the vacuum bagging mold method and hand layup technique

Ultimate Tensile Strength Comparison

The tensile strength values extracted from stress-strains curves of both vacuum bagging mold technique and hand layup technique are compared in Fig. 5. From Fig. 5, it was identified that the samples processed with the vacuum bagging mold technique have higher tensile strength than the samples processed with the hand layup technique due to the following reasons. There are many benefits to carrying out the vacuum bagging mold technique over the traditional hand layup technique for making composite components. The primary is that the finished product will yield improved ultimate tensile strength values and be light dense. Comments that are strengthened with less density could have serious competitive benefits among the others, particularly because they require fewer resources and materials, and are more durable. This is especially clear when examining the fiber to resin ratio, which is improved by the vacuum bag molding method. Another advantage of the vacuum bagging mold technique over the hand layup technique is that the technique is fairly simple to convert to, and materials for fundamental components are readily available and reasonably priced.

Fig. 5
figure 5

Ultimate tensile strength of carbon fiber composites prepared by vacuum bagging mold technique and hand layup technique

Experts in the composites industry and seasoned veterans are well aware of vacuum bagging mold technique advantages, but to others outside the industry, it can seem like a complicated procedure with an odd name. In a unique procedure called the vacuum bagging mold technique, the composite's fabric and resin are kept precisely wherever you want them to be while everything cures and dries. The method ensures consistent pressure over the cloth and forces it to follow the desired shape, making it perfect for forcing composite materials to follow patterns or molds. Additionally, the technique' consistent pressure guarantees that the resin impregnates the fabric fibers deeply and uniformly, leading to a finished product with better strength. Imagine it as a vacuum pump for resin infusion Because of the above-mentioned reasons, the samples processed with the vacuum bagging mold technique resulted in better ultimate tensile strength properties than the samples processed with the hand layup composite making technique [25,26,27].

Ductility Comparison

The ductility or elongation values extracted from stress-strains curves of both vacuum bagging mold technique and hand layup technique are compared in Fig. 6.

Fig. 6
figure 6

Ductility of carbon fiber composites prepared by vacuum bagging mold technique and hand layup technique

From Fig. 6, it was observed that the laminar composites made with the vacuum bagging mold technique have lesser ductility values than the laminar composites made with the hand layup technique due to the following drawbacks of the vacuum bagging mold technique. (i) Making an oversaturated laminate is the first step in the vacuum bagging mold technique. The surplus resin is subsequently removed using vacuum pressure, although the quantity removed will rely on a number of important factors that may be challenging to regulate, (ii) it is obviously a waste of effort and resources to remove more resin, which was initially brought in, (iii) since the resin pot-life is the limiting element in laminated composite parts, the vacuum bagging mold method must also be used a few times, (iii) the duration of the vacuuming pressure that is applied can have a significant impact on how much resin is removed from one section to the next, (iv) If a leak in the vacuum seal develops and cannot be found right once, the vacuum bagging mold method may become hurried, increasing the chance of a mistake, (v) when making hand layups, a significant amount of air bubbles must be introduced because they will eventually become trapped in the reinforcing fabric. The idea that using a vacuum will allow you to remove embedded air is a misconception. The vacuum bagging method, on the other hand, will encourage the formation of those air bubbles, creating voids in the laminate and surface, (vi) unfortunately, bagging makes it impossible to accurately determine the fiber to volume ratio as is possible with other methods, and over-bleeding or dry laminates can be a major worry, (vii) another immediate drawback of vacuum bagging mold technique is that there is no way to pause or back up once the operation has begun and (viii) with a race against the resin pot-life and putting all of the ingredients in place, wettability and scraping methods clearly have an economic and tolerance disadvantage. The above-mentioned disadvantages of the composites made with the vacuum bagging mold technique over the composites made with the hand layup technique resulted in less ductility [17, 28,29,30].

From the summary of the hardness and tensile results, it was observed a trend that the UTS values are inversely proportional to ductility values in all cases irrespective samples’ condition and content of the reinforcement. It is well known that in most of the cases the strength values are inversely proportional to ductility values. If the researchers want to improve the strength of any component, they have to sacrifice the elongation/ ductility and if they want to improve the elongation in the component, they have to sacrifice the strength. This is the major drawback in most of industries and aerospace applications. But, as per the application point of view, a component requires both strengths (to resist the loads) and elongation (to store the strain energy). In order to get the advantages of both strength and ductility, the combined effect of the strength and ductility should be assessed. Hence, keeping this point in view, most of the steel industries use one term called strain hardening parameter (e). The strain hardening parameter is used to measure the combined effect of both strength and ductility. The strain hardening parameter (e) is the product of strength and ductility and is given by;

Strain hardening parameter (e) = Strength (i.e., UTS) × Ductility (i.e., % elongation) [31,32,33,34]

Where UTS in MPa, elongation is in %, and strain hardening parameter in MPa-%. The strain hardening index values are documented along with the tensile properties. It was identified that the samples fabricated with the vacuum bagging mold technique have the highest product of strength and elongation as compared to the samples processed with the hand layup technique. From these results, it is suggested to choose the optimum content of reinforcements and resin that were added to the matrix instead of adding higher or lower reinforcement contents. Finally, the ductility results are coinciding with the ultimate tensile strength data and microstructural results.

Elastic Modulus Comparison

The elastic modulus values extracted from stress-strains curves of both the vacuum bagging mold technique and hand layup technique are compared in Fig. 7. From Fig. 6, it was identified that the composites made with the vacuum bagging mold technique have a higher elastic modulus than the composites made with the hand layup technique due to the following drawback of the vacuum layup technique. The hand layup resins may be more damaging than goods with greater molecular weights due to their lower molecular weights. The resins' decreased viscosity also suggests that they have a greater propensity to pierce garments. It is well-known that ever the material is having a lesser elongation and higher strength will definitely have a lesser elastic constant because the strain and young’s modulus are inverses proportional to each other and stress and young’s modulus are proportional to each other. Finally, the values of ultimate tensile strength, ductility, and elastic modulus are in line with each other and supported the results of the microstructural analysis. Abdurahman et al. [35] prepared fiber-reinforced composites in their study using three different techniques such as vacuum infusion technique, vacuum bagging mold technique, and hand layup technique. And they stated that the specimens manufactured with the vacuum infusion technique have greater ultimate tensile strength as compared to the vacuum bagging mold technique and hand layup technique, also the samples made with the vacuum infusion technique have greater elastic modulus than the vacuum bagging mold technique and hand layup technique and it could be attributed to the manufacturing route for making fiber-reinforced composites. Due to the compressive stresses, it experiences during the procedure, the vacuum infusion technique exhibits better tensile characteristics and can prolong the time required for the resin to fully cure. The procedure is also carried out in a vacuum environment to get rid of the gases within the molds. The vacancies or spaces within the specimen would be reduced indirectly by air pressure. The tensile characteristics can be enhanced by reducing vacancies or empty spaces inside the specimen [36]. The likelihood of air leakages during the vacuum bagging mold technique is substantial and highly dependent on the knowledge and expertise of the operator. The ambient pressure, such as air pressure, and the vacuum have a restriction on the resin injection pressure.

Fig. 7
figure 7

Elastic modulus of carbon fiber composites prepared by vacuum bagging mold technique and hand layup technique

This procedure is still ineffective, though, since this curing which began well before the vacuum phase prevents the vacuum from being capable of removing the gases and vacancies which have already become caught in the lamina. It however offers improved tensile characteristics than manual layup since many of the gases and vacancies on the surfaces could be eliminated with a vacuum. To evacuate gas from the layer of fiber and the matrix during the manual layup procedure, rubber is the only material that is physically compressed. This could also remove gaps between the layers, although the procedure is less efficient since the lamina is not under concurrent stress [37].

Fracture Analysis Comparison

Following the tensile testing, the fracture tensile samples were examined under a scanning electron microscope to examine the behavior of the fractures and failures. In Figs. 8 and 9, respectively, scanning electron micrographs of samples processed with vacuum bagging mold technique and hand layup technique, subjected to tensile tests under the predefined processing conditions.

Fig. 8
figure 8

Tensile fractured scanning electron micrographs of samples processed with vacuum bagging mold technique

Fig. 9
figure 9

Tensile fractured scanning electron micrographs of samples processed with hand layup technique

Fracture morphologies of the samples processed with the hand layup technique show that matrix fracturing is the primary cause of test sample fracture. During transversal tensile testing, the strain was mostly supported by the matrix since the forcing was transverse to the fibers. When compared to samples not exposed to hot, wet circumstances, the matrix breaking was accelerated by the wetness that was absorbed, leading to a loss in strength properties. The scanning electron micrographs in Fig. 8 depict matrix fracturing and debonding brought on by the tension testing stress. At high resolution, delamination between the fiber and matrix brought on by the imposed load can be detected. Additionally, the debonding and matrix fracturing are clearly visible in these higher-resolution pictures. Samples were tested that have been conditioned in high resin content, humid environment shows higher delamination and matrix fracking under these two study circumstances, compared with each other. It indicates unequivocally that the quantity of humidity received by carbon fiber-reinforced composite leads to increased crack formation and delamination, which lowers tensile strength.

The tensile test of samples processed with vacuum bagging mold technique mostly led to fracture because of both matrix fracturing and fiber breaking, according to scanning electron microstructures of samples from the control group. Furthermore, the force was primarily supported by the fiber due to tension stress in that orientation. Filament fracture and matrix delamination are visible in scanning electron microstructures. Comparing samples not exposed to high resin content, wet conditions with those that were exposed to absorption of moisture accelerated delamination, which led to higher tensile strength. The scanning electron microstructures are shown in Fig. 9 demonstrate debonding and fiber breaking as a consequence of tensile testing's load. This is much more obvious when magnified more. Debonding at the fiber-matrix junction, fiber fracture, and matrix fracturing are all clearly visible. In contrast to testing samples that have not been treated, high resin content environment conditions result in higher delamination, matrix fracturing, and fiber breaking. Therefore, absorbed humidity always causes an increase in tensile strength.

In a fiber-reinforced composite, humidity predominantly occurs through diffusion, which is influenced by heat and relative concentration. Other phenomena that might be involved comprise fiber/matrix boundaries, capillary activity at the fiber intersection, and transfer through micro-cracks [38,39,40]. Composites are negatively impacted by humidity since it frequently results in swelling and deterioration [41]. Fiber/matrix intersection mechanical strength decreases as a result of polymeric matrix plasticization and absorption humidity in composites [41]. Conventional observations show that moisture content follows Fick's law, and that mobility is driven by the difference in humidity concentrations between the operational environment and the composite structure as well as the atmosphere's temperatures, leading to continued absorption until saturating. The layup order affects the rate of humidity and maximal humidity levels in carbon fiber-reinforced composites. It was found that developed a semi-laminated absorbed humidity more gradually than fully laminate [42]. The degree of deterioration depends on the reinforcing chemical, the matrix, and the exposure period. Differing types of composite materials have different levels of responsiveness to varying environmental conditions.

Conclusions

In the current investigation, the carbon fiber-reinforced composites were successfully prepared by using the vacuum bagging mold technique and hand layup technique and compared the microstructural and mechanical characteristics between them, and the following conclusions were drawn.

  1. (1)

    The hand layup specimens had a prominent accumulation of resin on the surfaces, but the fiber part in the core of the sample was noticeably devoid of resin. The absence of integration between the fiber and resin in the core has a substantial impact on the tensile properties of the composite material.

  2. (2)

    The adhesion between the adhesive and the fiber seemed to be firmly achieved during the vacuum bagging molding procedure. Nevertheless, noticeable spaces between fibers persist, indicating that, despite the bonding, there are regions where the composite structure may lack complete cohesion.

  3. (3)

    The findings suggest that carbon fiber-reinforced composites produced using the vacuum bagging mold method demonstrate higher ultimate tensile strength and flexibility in comparison to those manufactured using the hand layup technique. This can be ascribed to the successful removal of gases and empty spaces on the surfaces of the composite material by using a vacuum during the molding process.

  4. (4)

    In contrast, composites produced by the hand layup method demonstrate higher ductility values compared to those made utilizing the vacuum bagging approach. The variation can be ascribed to the substantial probability of gas leakages occurring during the vacuum bag molding procedure, which strongly depends on the operator's experience. Moreover, the interplay between vacuum and ambient pressures imposes limitations on the flow rates of resin throughout the molding procedure.