2.1. Materials
The chosen laminate design aligns with the classic structural and material design of laminate materials used in inflatable structures such as airships, aerostats, inflatable antennas, and other applications. It is also suitable for assessing the adhesion strength between the fiber layer and the film layer. The structure of the laminate is shown in
Figure 1.
The laminate is made up of a polyethylene terephthalate (PET) film with vacuum-deposited aluminum (VDA) called Mylar (CS Hyde, Lake Villa, IL, USA) and two adhesive layers sandwiching the load-bearing layer. The Mylar film is a popular choice for FRL production, particularly in industries such as aerospace, military, and industrial applications, where environmental protection is a critical consideration as the VDA coating enhances the material’s barrier properties, providing excellent protection against moisture, gases, and light transmission. Woven fabrics from Kevlar (CST, Tehachapi, CA, USA) (ACP Composites, Livermore, CA, USA) and Ultra-High Molecular Weight Polyethylene (UHMWPE) (Zhejiang Mengtex Special Materials Technology Co., Ltd., Tongxiang, Zhejiang, China) fibers were selected as the core strength-bearing/fiber-reinforcement element for this study. These two high-performance fibers (high strength-to-weight ratio) are among the most popular fibers used in the industry, but they are also two of the more difficult to work with due to their chemical inertness and high crystallinity. Four plain woven fabrics from Kevlar and two from UHMWPE fibers, each with distinct structures, were sourced and utilized for this study. The fabric construction parameters include the areal density, the yarn linear density, yarn crimp due to weave as well as the fabric count warp thread density × weft thread density). A summary of the specifications of each fabric are depicted in
Table 1. Three different adhesives were used in order to confirm the trend across different adhesive types. The adhesive types sourced are as follows: ethylene-vinyl alcohol (EVOH) (EVAL™, Kuraray, Chiyoda-ku, Tokyo, Japan), ethylene-vinyl acetate (EVA) (KETAEBO, Suzhou, Jiangsu, China), and thermoplastic polyurethane (TPU) (KETAEBO, Suzhou, Jiangsu, China). The thickness of each adhesive was kept constant across all the samples, with 0.015 mm for EVOH and 0.025mm for TPU and EVA. These adhesives were chosen because they are the top readily used choices in the industry and they are also commercially accessible.
Kevlar has a structure primarily composed of long chains of polymeric aromatic amide groups. The amide groups (-CONH-) in Kevlar are known for their high polarity and strong hydrogen-bonding capacity. TPU contains urethane groups (-NHCOO-) that can form strong hydrogen bonds [
9]. These hydrogen bonds can interact effectively with the amide groups in Kevlar, leading to better adhesion, which can be seen in
Figure 2a. EVA has both ethylene (-CH
2-CH
2-) and acetate (-COOCH
3) groups as seen in
Figure 2b. While it contains polar acetate groups which can form some hydrogen bonds with Kevlar’s amide groups, the hydrogen bonds formed may be undermined by the repulson between the carbonyl groups, resulting in weaker adhesion [
10].
Furthermore, the presence of hydroxyl groups in EVOH originating from the vinyl alcohol units creates the possibility of hydrogen bonding with the amide groups present in Kevlar, potentially augmenting the adhesion between the two materials. The polar nature of EVOH, as seen in
Figure 2c, attributed to the oxygen in the vinyl alcohol units, further promotes intermolecular interactions, enhancing the likelihood of bonding. Additionally, EVOH is known for its high crystallinity, and the regular and ordered structure of its crystalline regions provides favorable sites for enhanced adhesion. The increased mechanical rigidity associated with crystallinity facilitates improved interaction with other materials, and the larger surface area of crystalline regions also presents additional opportunities for bonding [
11].
UHMWPE is renowned for its low friction, high wear resistance, and exceptional chemical resistance. This is attributed to its unique molecular structure characterized by high crystallinity and the absence of functional groups, which presents significant challenges for bonding with adhesives. The crystalline structure of UHMWPE is composed of long polyethylene chains (-CH
2-CH
2-), contributing to its nonpolar nature and making it difficult for many adhesives to form effective bonds [
12]. The carbon-carbon covalent bonds in UHMWPE are very strong and rigid, however, in the transverse directions, the bonds between hydrogen atoms in the molecules are much weaker as they rely on van der Waals interactions as seen in
Figure 3. The high molecular weight molecules align along the fiber axis which leads to significant strength due to the numerous albeit weak hydrogen bonds and high alignment [
12,
13].
EVA adhesive contains polar acetate groups (-COOCH
3) which can potentially form weak hydrogen bonds with UHMWPE. Although UHMWPE’s surface is generally resistant to adhesion due to its lack of polar groups, the presence of surface irregularities might allow EVA to establish some degree of interaction. The compatibility between EVA and UHMWPE can also be enhanced due to the ethylene (-CH
2-CH
2-) units present in both materials, providing common segments for entanglement of extremely lengthy polymer chains [
14]. TPU adhesive contains polar urethane groups (-NHCOO-) which can also form weak hydrogen bonds. However, the bonding between TPU and UHMWPE may not be as strong as that of EVA and UHMWPE. This is because EVA offers more sites for weak hydrogen bonding due to the presence of acetate groups. However, despite these interactions, the overall adhesion remains relatively weak due to the highly crystalline and nonpolar nature of UHMWPE, necessitating the need of specialized surface treatments or primers to enhance bonding with UHMWPE [
15].
2.2. Fabric Structure Parameters
Understanding fabric structure parameters is essential for comprehending the mechanical, physical, and adhesive properties of textiles. Fabric structure is influenced by various factors, including yarn type, yarn linear density, fabric count, and weave pattern. Among the parameters used to characterize fabric structure are fabric cover factor and fabric tightness. Fabric cover and tightness play pivotal roles in determining the adhesive bonding characteristics of fabrics. The fabric cover factor, denoted by C, represents the proportion of the fabric area covered by actual yarn [
16,
17]. It is a crucial parameter in assessing the density and arrangement of yarns within the fabric structure. The calculation of fabric cover factor typically considers the yarn diameter and spacing, providing insights into the packing density of yarns and the overall tightness of the fabric weave.
The fabric cover factor provides valuable insights into the density and arrangement of yarns within the fabric structure. A lower cover factor suggests a more open and loosely woven fabric (
Figure 4a), whereas a higher cover factor indicates a denser fabric with closely packed yarns (
Figure 4b). This distinction in fabric tightness can significantly influence the adhesive bonding characteristics of fabrics. Higher cover fabrics with closely packed yarns offer more contact points and surface area for adhesive bonding. This promotes strong interfacial adhesion and bonding. However, the reduced spacing between yarns minimizes adhesive penetration into the fabric structure, inhibiting proper wetting of the fibers. Conversely, in looser woven fabrics with lower fabric cover factors, greater spacing between yarns may limit contact points available; however, the increased porosity allows the adhesive to wet the fibers effectively, creating strong cohesion between adhesive layers above and below the fabric. This enhances overall adhesion strength, despite there being lesser fabric surface available for bonding. Moreover, the surface bonding is primarily hydrogen bonds, and they are inherently weaker than the cohesion between the adhesive layers. Therefore, fabric cover factor and tightness are crucial considerations in adhesive bonding, and the relationship between fabric structure and adhesion is nuanced. Adjusting fabric tightness and cover must be carefully considered alongside adhesive characteristics to achieve the desired adhesive performance.
In practice, cover factors are calculated independently for the warp and weft directions (Warp cover,
C1) and (Weft cover,
C2), reflecting the coverage of yarns along each axis. The fractional warp cover factor (
C1) is determined by the ratio of the yarn diameter (d) to the thread spacing (p) in the warp direction. The yarn diameter was determined by dividing the thickness of the fabric by 2, assuming that the thickness is the distance at the intersection of the warp and weft yarns. The yarn spacing was obtained by averaging 10 measurements under microscope. Finally, the fabric cover was calculated using Equation (1) below.
On the other hand, when determining the tightness of a fabric, consideration is given to the weave factor, which is determined by the design of the weave [
18]. However, as all fabrics utilized in this investigation were of a plain weave (weave factor of 1), the assessment of tightness was omitted from subsequent analyses. Nevertheless, in future research endeavors examining various weave designs, the tightness of each fabric will be of paramount significance in elucidating the impact of diverse weave patterns on the adhesive properties of woven fabrics. The cover of the fabrics used in this investigation are summarized in
Table 2 below.
2.3. Laminates Formation
A Pratix OK-12L Seamless Teflon Belt Drum Laminator was used for the laminating procedure. All the laminates’ components were fed through a heated roller using a one-pass procedure. The feed speed was set at 0.32 m/min and the pressure at 413.685 kPa (60 psi). The temperature was set to a predetermined level depending on the fiber type and adhesive type in order to obtain the laminate samples without sacrificing their own properties. Higher temperatures lower viscosity, improving flow for application and enhancing adhesion strength, while lower temperatures increase viscosity, extending the open time. Proper temperature control is vital to ensure optimal bonding, substrate compatibility, equipment functionality, and safety during application and subsequent use [
5]. The temperature for each adhesive was chosen based on the recommendation of each manufacturer. The polymeric film chosen was VDA-coated PET film. Before lamination, a very thin coating of VDA is deposited on the films to reinforce their ability to retain gas while adding the least amount of weight possible. While it is not necessary for this study, in real-life applications, the function of the VDA coatings placed on the exterior of the laminate is to effectively reflect heat and UV rays and protect the fiber and adhesive from deterioration.
There were 18 combinations of the 6 woven fabrics with the 3 adhesives; however, because the melting temperature for EVOH is too high for the UHMWPE fabrics, a total of 16 different laminates were made. Each laminate with their corresponding laminate IDs are summarized in
Table 3. The properties of the fiber-reinforced laminate, including the interfacial adhesion, were assessed by T-peel test (20 replicates in each direction) and yarn pullout in laminate test in both weft and warp fabric directions (10 replicates in each direction).
2.4. Testing and Characterization
T-peel test: T-peel test and analysis were performed following the procedure outlined in ASTM standard D1876 [
19]. By using a T-shape specimen, this technique is primarily designed to assess the relative peel resistance of adhesive bindings between flexible adherents.
Figure 5 shows a schematic of the peel test samples. For the T-peel test, an Instron machine 34TM-5 with a 5 kN load cell was used.
The T-peel test result profile of FRLs is very important to understand the delamination behavior. As the woven fabric-reinforced laminates are made with woven fabric, the behavior of delamination will depend heavily on the properties of the material used to make the reinforcement [
20]. Every T-peel test profile is unique; some have an initial maximum peak load which is a measure of the force required to initial failure and to break the rim-seal contact at the start of peeling, but such behavior is not present in FRLs [
21,
22]. However, there are commonly large fluctuations on the graph which indicate inconsistencies in the peel strength. The peak-to-peak distance represents the interval of bond separation i.e., the maximum variation in force during the test, indicating the range of forces experienced during the peeling process. The peak-to-peak distance can also provide insights into the strength and durability of the bond between the materials being tested. A larger peak-to-peak distance may indicate greater variability or instability in the bond, whereas a smaller peak-to-peak distance may suggest a more consistent and reliable bond.
Figure 6 shows the typical load-displacement curve of FRLs. According to the testing procedure, the specimen’s peel strength/resistance will be determined by averaging the load values over a bond line that is at least 12.7 cm (5 inches) long after the initial peak which is represented by the red line [
19].
Yarn pullout in laminate test: For the yarn pullout in laminate test, a new schematic for the testing and the test samples has to be developed. The specifications shown in
Figure 7 were determined to be adequate and consistent for testing the force to pull out a yarn from the laminate based on extensive preliminary work [
23]. A test speed of 0.212 mm/s (0.5 inch/minute) was selected. To prepare the sample for testing, the film backing should be separated from the laminate by at least about 50 mm. Next, ten strips, each 15 mm wide and aligned parallel to the yarn, are cut from the material. Among these strips, one yarn located near the center is selected while the surrounding yarns in the same direction are removed from the laminate area. The laminate length is then marked, and a small slit is made in the middle, enclosing the selected yarn along with at least one yarn on either side which completes the preparation process. For precision, a magnifying glass is used to make sure the yarn as well as one yarn on either side is cut to ensure that the entire pullout yarn is separated from the rest of the laminate. The final step before testing involves attaching tabs to the ends using a material that facilitates gripping by the tensile machine’s jaw. It is important to ensure that the glue used for the tabs does not cover the slit. Leaving a small gap between the slit and the bottom jaw can prevent any obstruction during testing, as illustrated in
Figure 7.
In
Figure 8, an exemplary load-displacement curve from a yarn pullout in laminate test is depicted, which encompasses many aspects of the plain fabric pullout response [
24]. When a yarn is pulled, the pullout force will reach a peak value. This peak pullout force is reported as an indication of the quality of the adhesion. The area under the curve preceding the dotted line is the energy required to cause the pullout yarn to break from the matrix and is dependent on the extent of adhesion between the fiber and the adhesive matrix.
After it reaches the peak, there are drag frictional forces of the pullout yarn still in contact with the adhesive and crossing yarns in the laminate. The less contact the pullout yarn has in the laminate, the lower those drag forces will be. The drag forces will only reach zero once the yarn is fully out of the laminate. For stiff and rigid adhesives, it was observed that the drag forces are much lower than those for more elastic adhesives. Because rigid and stiff adhesives tend to ‘snap’ or break suddenly once peak pullout force has been reached, elastic matrixes tend to yield or bend before finally detaching. Additionally, in rare cases, when the yarn breaks instead of being pulled out, the curve reflects a single-yarn tensile property. In this case, the datum is considered an outlier and discarded.