Sage Journals: Discover world-class research

Abstract

This paper investigates the influence of thread stitching on the ballistic performance of plain weaves made of ultra-high-molecular-weight polyethylene multi-filament yarns. The inter-yarn friction is increased due to the constraint imparted by the sewing thread. The yarn pull-out test shows that the peak-load force of the sample with one stitching line is almost 10 times greater than that of the unstitched plain weave, and the maximum pull-out force increases with the loading rate. Ball-bearing impact tests are performed to characterize the ballistic performance of the stitched and the unstitched samples, and finite element simulation is used to study the underlying mechanisms of energy absorption. The ballistic penetration tests show that the stitched fabrics outperform the plain weave in terms of energy absorption. The most significant improvement in ballistic performance is observed in stitched panels where sample SL2T (a triple-ply plain fabric system stitched on every two yarns) exhibits a specific energy absorption over two times greater than that of multi-ply systems consisting of plain weaves. It is also found from the high-speed photography that thread stitching constrains the yarn displacement and therefore eliminates the possibility of yarn pull-out, enabling the primary yarns to be well-engaged with the projectile at low impact velocities and to be stretched to fail at high impact velocities. Numerical predictions show that thread stitching enlarges the area of stress distribution and widens the transverse deflection, making the stitched systems absorb more energy than the unstitched system shortly after impact.

Keywords

fabrics ballistic impact stitching energy absorption yarn pull-out

Introduction

Strong, low-density fibers have been favored materials for impact load-bearing applications.^1–3 Among the limited choices, ultra-high-molecular-weight polyethylene (UHMWPE) is one of the most successfully commercialized and developed fibers for designing protective systems that are flexible and lightweight. For ballistic applications, these fibers are mainly used in the form of cross-plied continuous filaments laminated by a soft matrix. The filament reinforcement serves as the load-bearing constituent, and the matrix acts as a load transfer medium to constrain the filament mobility and improve dimensional stability. There is no shortage of publications studying the impact responses of cross-plied UHMWPE laminates. Most of these researches focused on commercially available products such as Dyneema^®HB26, Dyneema^®HB210, and Dyneema®HB80.^4–8 Apart from the cross-plied laminates, the ballistic performance of needle-punched non-wovens was also investigated. Martínez-Hergueta et al. found that the ballistic limit of the needle-punched Dyneema^®SK 75 non-woven fabric is dependent on the fiber orientation. A more isotropic structure led to a more significant increase in ballistic limit and energy absorption capability. The performance deteriorated when the fibers were more oriented in the system.⁹

In comparison with non-woven laminates and needle-punched fabrics, flexible woven fabrics have attracted much less attention. Zhou et al. found that penetration is mainly accommodated by yarn pull-out and windowing due to the low yarn-yarn friction and projectile-fabric friction of UHMWPE woven fabrics.^10,11 It follows that producing appreciable friction between the fabric-forming yarns might prove beneficial for the energy absorption capability. The simplest method of increasing inter-yarn friction is to weave tight fabrics. Nevertheless, the performance of tightly woven fabrics deteriorates due to the increased crimp ratio. This is because that the propagation of the transverse wave is slowed by yarn undulation and the direction of loading force is not normal to the fiber axis, leading to a lower ballistic limit¹² and reduced energy absorption capability.¹³ Chu et al. attempted to use the plasma-enhanced chemical vapor deposition technique to increase the yarn–yarn friction. The results showed that the static friction was improved from 0.12 to 0.23, and kinetic friction rose from 0.11 to 0.19.¹⁴ Hasanzadeh et al. impregnated the UHMWPE fabrics with shear-thickening fluids (STFs) composed of fumed silica nanoparticles suspended in polyethylene glycol. The ballistic results confirmed the improved impact resistance of the treated fabrics and the contribution of frictional properties induced by STF impregnation in restriction of the yarns within the fabric.¹⁵ Wang et al. treated the fabrics with four different resin matrices and found that composites having flexible matrices performed much better in perforation resistance and energy absorption than those using rigid matrices. This is because that more rigid matrix restrained the laminate’s transverse deformation to a smaller area, resulting in a more localized strain area.¹⁶ Zhou et al. incorporated leno structures into the plain weave. This combination demonstrated an increase in yarn pull-out force but was insufficient to provide a noticeable improvement in ballistic performance.¹⁷ Some researchers used thread stitching to improve inter-ply reinforcement.^18–21 Zhou et al. used thread stitching to increase the yarn-yarn friction of aramid fabrics, and the results suggested a significant improvement in energy absorption capability upon ballistic impact.²² The main limitation of Zhou et al.’s work is that the experiments were performed in the vicinity of the ballistic limits only. In addition, the responses of the other types of stitched high-performance fiber assemblies upon ballistic impacts are still unknown. Since it has been demonstrated that thread stitching provides an appreciable increase in the energy absorption capability of the Kevlar plain weaves, it is interesting to put another commercially available material, UHMWPE fabric, as a subject of study over a range of impact velocities from the ballistic limits to 400 m/s. In this paper, the responses of stitched UHMWPE plain weave upon ballistic impact will be studied using a gas gun and high-speed photography. The finite element (FE) method will be used to investigate the underlying mechanism of the energy dissipation process. This work aims to improve the ballistic performance of UHMWPE plain weave and explore the possibility of using dry woven UHMWPE fabrics for the engineering design of flexible protective systems.

Material and methods

Materials

The plain fabrics were made of Tekmilon®multi-filament yarns, which were provided by Mitsui & Co. LTD. The UHMWPE mono-filament has a yield stress of 3.88 GPa, a longitudinal modulus of 135 GPa, and a transverse modulus of 1.34 GPa. The filament is 17.2 μm in diameter and 145°C in melting temperature. All of the fabrics have the same thread density and yarn linear density of 9 threads/cm and 1300 dtex, respectively. The plain weaves were flat-stitched on both of the warp and weft directions using a cotton sewing thread on an automatic embroidery machine (Figure 1(a)

Figure 1.

(a) Embroidery machine used to stitch the plain weaves; (b) plain weave with Stitching lines on every four yarns (SL4S).

). A stitch density of 20 stitches/cm was selected to ensure that each fabric-forming yarn was pinned by the sewing thread. The sewing thread has a linear density of 216 dtex and was manufactured by plying two cotton staple yarns. Table 1

Table 1.

Yarn and sewing thread specifications.

	Tekmilon®I	Sewing thread
Material	Ultra-high-molecular-weight polyethylene	Cotton
Yarn type	Multi-filament yarn	Staple yarn
Yarn breaking strength (cN/tex)	400	32.7
Yarn breaking elongation (%)	3.7	20
Yarn modulus (cN/tex)	14,000	133
Yarn count (dtex)	1300	216
Density (kg/m³)	970	1540

shows the specifications of the multi-filament yarn and the sewing thread in use. Table 2

Table 2.

Fabric specifications.

Sample code	The density of the stitching lines	Areal density (g/m²)	No of plies stitched
PW	Plain weave without thread stitching	240	—
SL2S	Stitching lines on every two yarns in a single-ply fabric system	371	—
SL4S	Stitching lines on every four yarns in a single-ply fabric system	303	One
SL6S	Stitching lines on every six yarn in a single-ply fabric system	283	—
SL2T	Stitching lines on every two yarns in a triple-ply plain fabric system	889	—
SL4T	Stitching lines on every four yarns a triple-ply plain fabric system	801	Three
SL6T	Stitching lines on every six yarns a triple-ply plain fabric system	766	—

shows the specifications of the different fabric samples studied in this research. In Table 2, PW is short for plain weave, and SL2S indicates that the stitching line was applied to every two yarns on a single-ply fabric. SL2T indicates that the stitching lines were applied to every two yarns on a triple-ply plain fabric system. Figure 1(b) shows a typical sample of SL4S.

Testing methods

The stiffness of fabric and yarn pull-out tests

Stiffness was tested to characterize the influence of thread stitching on fabric softness. This test was performed according to the standard of ASTM D1388-18 (cantilever method). Yarn pull-out tests were performed to characterize the influence of stitching on the frictional force between the fabric-forming yarns. Schematic diagrams of the yarn pull-out method are shown in Figure 2(a). Samples with zero, one, two, and three stitching lines 2 (b) were put into tests. Sample targets were cut into 6 × 12 cm. The length of the fabric zone is 6 cm. Yarn tails were left on the top (5 cm) and bottom (1 cm) sides. A slot was maintained for the very yarn to be pulled out on the bottom while the rest of the tails were clamped by a bottom jaw. Tests were performed at constant rates of 50 mm/min, 250 mm/min, and 500 mm/min, respectively.

Figure 2.

(a) A schematic diagram of the yarn pull-out test (b) Samples with a different number of stitching lines.

Ballistic test

The ballistic tests were performed on a set-up shown in Figure 3. In this ballistic apparatus, the spherical steel projectile in use is 2 g in weight and 8 mm in diameter. The projectile is propelled by compressed gas, and the impact velocity varies with the range of 0∼400 m/s. The impact and the residual velocities of the projectile were obtained by using two sets of infrared chronographs. The energy loss of the projectile can be determined by

Δ E = \frac{1}{2} m (v_{1}^{2} - v_{2}^{2})

(1)

where

Δ E

is the kinetic energy loss of the projectile, m is the mass of the projectile, and v₁ and v₂ are impact and residual velocities of the projectile, respectively. In this research, the effect of air drag on the projectile was considered as a vital factor of projectile kinetic energy loss, and the energy dissipated by air drag increases exponentially with the impact velocity. By removing this portion of energy loss

Δ E

, energy absorbed by the sample target can be calculated. This value was used as an indicator of its ballistic performance.

Figure 3.

(a) A schematic diagram of the set-up for the penetration test, (b) Ballistic apparatus.

A corner-clamped frame was designed to clamp the sample target (Figure 4). In this frame, the fabric was gripped at its corners to allow yarn pull-out during the ballistic event. Fixing bolts were through-bolted with the backplate and fastened by screw nuts. The resistance against yarn pull-out arises solely from the frictional force at the crossovers and the stitching lines. The sample was cut into 25 × 25 cm to fit the size of the frames. Both of the single-ply and triple-ply fabric systems were clamped on this set-up.

Figure 4.

Schematic diagram and photographs of the corner-clamped frame.

Finite element model

Commercial FE software ABAQUS^® 6.14 was used to model the ballistic event on fabric samples, aiming to study the mechanism of energy absorption evolution and stress distribution of different fabric systems. In this research, FE simulation is limited to colliding between a projectile of rigid material and single-ply PW and SL2S at impact velocities of 160 m/s and 400 m/s, respectively. The projectile model is spherical, with the diameter being 8 mm, and the mass of the projectile being 2 g, identical to the real projectile used for the penetration tests. Fabrics were simulated at the yarn level to construct a plain weave of 125 × 125 mm. The warp and weft yarns were modeled as 3D solid geometry with cross-sections and crimps according to their corresponding fabric sett. The yarn cross-section is considered to be lenticular^23,24 and is defined by two arcs having radii r. According to Equation (2), the parameter r can be calculated from the width w (0.985mm) and height h (0.16mm) in Figure 5.

r = (w^{2} + h^{2}) / 4 h

(2)

Figure 5.

A schematic diagram of the lenticular yarn cross-section.

The undulated yarn path is also formed by a series of arcs to match the geometry of the yarn cross-section. The width and height of the yarn crimp are 1.11 mm and 0.165 mm, respectively.

Eight-node hexahedron elements (C3D8R) and ten-node tetrahedral elements (C3D10m) were used for yarns and the projectile in the model, respectively. Encastre boundary conditions were assigned for fabric corners. Symmetric boundary conditions about the X and Z axes were applied to the edges, and therefore a model of 62.5×62.5 mm can be used to simulate the PW. SL2S can be simulated based on the plain weave model shown in Figure 6(a). In the real fabrics, threads were stitched to constrain the yarn mobility. The constraint on yarn mobility in the FE model was simulated by using the “mesh merge” module in ABAQUS Explicit to “weld” the elements at the nodes. The welded nodes were indicated by the pink points in Figure 6(b), and the warp and weft yarns have the same degree of freedom at the crossovers. The energy evolution history and stress distributions of the primary yarns were recorded and compared among different systems to identify the effect of thread stitching on the response of the constituent yarns during a ballistic event.

Figure 6.

The meshing of FE models for a single-ply (a) PW and (b) SL2S.

Yarns were modeled as homogenous continua to represent the filament-level architecture. The material is assumed to be orthotropic, and the relative parameters are listed in Table 3. The values of yield stress, breaking strain, and Young’s modulus were provided by the fabric manufacturer and were assigned to the yarn material, and the values of transverse modulus and shear modulus were obtained from references²⁵. Damage is initiated when the following condition is satisfied²⁶

\int \frac{d {\bar{ε}}^{p l}}{{\bar{ε}}_{D}^{p L} (η, {\bar{ε}}^{p L})} = 1

(3)

where

η

is the stress triaxiality,

{\bar{ε}}^{pL}

is the equivalent plastic strain rate,

{\bar{ε}}^{p l}

is the equivalent plastic strain. The model assumes that the equivalent plastic strain at the onset of damage,

{\bar{ε}}_{D}^{p L}

, is a function of stress triaxiality and strain rate:

{\bar{ε}}_{D}^{p L} (η, {\bar{ε}}^{p L})

. The evolution of damage can be specified in the exponential form²⁷

d = 1 - exp (- \int_{0}^{{\bar{μ}}^{p l}} \frac{{\bar{σ}}_{y} {\bar{μ}}^{p l}}{G_{f}})

(4)

where

{\bar{μ}}^{p l}

is the effective plastic displacement,

{\bar{μ}}^{p l}

is the plastic displacement at failure,

{\bar{σ}}_{y}

is the effective plastic stress, and

G_{f}

is the fracture energy per unit area. Once the damage initiation criterion is met, the damage variable, d, increases according to Equation (4). When the damage variable reaches a value of 1, failure occurs. The value of fracture energy per unit area was set to 500J.

Table 3.

Material properties.

Material properties	Projectile	Yarn
Yield stress (GPa)	/	3.88
E₁₁ (GPa)	/	135.8
E₂₂ and E₃₃ (GPa)	/	1.34
G₁₃ and G₁₃ (GPa)	/	3.28
G₂₃ (GPa)	/	0.504
Poisson’s ratio	/	0.3
Mass density (kg/m³)	7800	970

The projectile was modeled as a rigid body and was not deformed during the impacting process. The material properties are listed in Table 2. A general contact interaction was used to define the contact between the projectile and the sample target and between the interlaced warp and weft yarns. The coefficient of friction was set to 0.14.²⁰

Results and discussions

Yarn pull-out behavior and fabric stiffness

Inter-yarn friction is characterized by performing the yarn pull-out test, measuring the force required to pull a single yarn out from the stitched and unstitched plain weaves. Figure 7 displays the force-displacement curves and peak-load forces of the yarn pull-out test performed at different loading rates. The peak-load force increases with the loading rates, and the stitched samples are more sensitive to the loading rate than the unstitched samples. Since the rate of pull-out is far lower than that of a ballistic event, a greater resistance against yarn pull-out is expected when impacted by a projectile. It was also found that thread stitching contributes to a significant increase in pull-out force, and the maximum pull-out force increases with the number of stitching lines. Sample with four lines exhibits a peak-load force of approximately 55 N at a pull-out rate of 500 mm/min, which is more than 15 times greater than that of PW. It must be noted that the load-displacement curves of stitched samples increase rapidly beyond the displacement of 10 mm, indicating that the constraint mechanism of thread stitching is activated at a later stage of the pull-out process. Therefore, the effect of stitching is “delayed” to some extent. When the projectile impacts the plain weave region, stitching plays a limited role in resisting yarn pull-out until and unless the primary yarns are pulled to a certain extent. The pulled yarns provide additional length for the projectile to push the yarns aside and consequently increase the probability of “windowing.”

Figure 7.

Pull-out force as a function of displacement for a plain weave with (a) zero, (b) one, (c) two, and (d) three stitching lines.

Figure 8 displays the results of the fabric stiffness test according to ASTM D1388-18 (cantilever method). The value shown on the vertical axis is the fabric flexural rigidity in μJ/m. PW exhibits the least flexural rigidity. Stitching stiffens the originally soft plain weave, and the magnitude of stiffness increases exponentially with the density of the stitching lines. The constraint provided by the stitching lines increases the friction between the warp and weft yarns, restricts the yarn displacement, and impedes the inter-fiber and inter-yarn motion. This mechanism inevitably increases the rigidity of the plain weave. A detailed explanation of the relationship between friction and stiffness was provided by McNeil and Standard in their investigation of the bending rigidity of sol-gel coated wool fabrics.²⁷

Figure 8.

The flexural rigidity of the stitched and the unstitched fabrics.

The ballistic performance of the stitched and unstitched UHMWPE fabrics

Single-ply fabrics

Figure 9 shows the plot of the residual velocity of the spherical projectile against its impact velocity for different samples. The impact velocity varies from the ballistic limit to approximately 400 m/s. The dots representing the stitched and unstitched sample targets exhibit a concave upward increase in residual velocity near the ballistic limit and a linear increase thereafter, indicating that the flexible systems absorb more kinetic energy of the projectile at low impact velocities and less at high velocities. This phenomenon was explained by Guo et al. their investigations of the responses of fabric upon ballistic impact.²⁸ They suggested that fabrics behave elastically when the impact velocity is near the system’s ballistic limit, and the majority of the kinetic energy of the impacting projectile is absorbed in the form of the strain and the kinetic energy of the impacted fabric; at higher impact velocities, the elastic response converted into an inelastic one. This results in a decrease in the strain and the kinetic energy absorbed by the fabric, making the system contribute less in slowing the projectile. It can be seen in Figure 9 that the ballistic limit of PW is approximately 111 m/s, which is lower than those of the stitched fabrics. In terms of the stitched fabrics, the ballistic limit increases with the density of the stitching lines, with SL2S being the most protective (a 57% increase in the ballistic limit when compared with PW).

Figure 9.

The ballistic performance of the stitched and unstitched plain weaves.

Figure 10 shows the energy absorption capacity of different samples at impact velocities of approximately 160 m/s, 230 m/s, 300 m/s, 350 m/s, and 400 m/s. In Figure 10(a), the vertical axis indicates the kinetic energy loss of the projectile after removing the portion dissipated by air drag. The energy absorbed by the sample target decreases as the impact velocity of the projectile increases. The most dramatic reduction in energy absorption was found on sample SL2S, being approximately 63.3% from 25.6 J at an impact velocity of 160 m/s to 9.4 J at an impact velocity of 397 m/s. Among single-ply fabric systems, SL2S exhibits the best ballistic performance at all velocities. PW exhibits inferior energy absorption capability when compared with the stitched samples. For instance, at impact velocities of approximately 160 m/s, the energy absorption of SL2S is around 150.7% greater than that of PW. At impact velocities of approximately 400 m/s, the improvement in energy absorption is 133.8%. SL4S and SL6S exhibit similar performance. It also appears that the influence of stitching on ballistic performance is more pronounced when the fabric is densely stitched. Figure 10(b) shows the specific energy absorption (SEA) of different sample targets. When the energy absorption is normalized by areal density, the SEA of SL2S is approximately 64.2% and 51.4% greater than that of PW at impact velocities of 160 m/s and 400 m/s, respectively.

Figure 10.

(a) Energy absorption as a function of impact velocity; (b) Specific energy absorption as a function of impact velocity.

Figure 11 shows the images of PW and SL2S at impact velocities of approximately 400 m/s. The penetration of PW was largely accommodated by yarn pull-out. The impacting projectile was loaded by a primary weft yarn and primary warp yarn at the initial stage of the ballistic event. The warp yarn slipped aside from the projectile surface and was disengaged with the projectile at 160 μs after impact, resulting in a half-pulled yarn in Figure 12(a). The yarn pull-out marker in the weft direction indicates that the primary weft yarn was entirely removed by the projectile. In addition, yarn windowing was also observed at the impact site of PW. It can be concluded that the majority of the projective kinetic energy loss was dissipated by the frictional sliding between the primary yarns and their crossover secondary yarns. Compared with PW, SL2S exhibited a more localized penetration mode where fiber failure contributes significantly to energy dissipation. It can be seen in Figure 12(b) that fabric penetration was mainly driven by yarn windowing and damage. Although the primary warp yarn displaces during the ballistic event, the extent of yarn displacement was limited to a minimal extent by the stitching thread. Thread stitching reduced the yarn mobility and prevented the primary yarns from being pulled out by the projectile. Therefore, the primary yarns were either damaged by the impact loading or pushed aside due to the damage of sewing thread, allowing the projectile to “wedge-through” the fabric.

Figure 11.

Fabric deformation of (a) PW and (b) SL2S at an impact velocity of approximately 400 m/s.

Figure 12.

Post-impact close-ups of (a) PW and (b) SL2S at an impact velocity of approximately 400 m/s.

At an impact velocity of approximately 160 m/s, PW was penetrated by the projectile in Figure 13(a), leaving two half-pulled primary yarns in Figure 14(a). SL2S stopped the impacting projectile and did not show any markers of yarn displacement and fiber damage. The width of the transverse deflection is more defined and wider in SL2S than that in PW. It is interesting to find that the velocity of the transverse wave decreases as time elapses, for example, the average velocity in SL2S is approximately 233.5 m/s from 10 μs to 110 µs and is 13 m/s from 210 μs to 310 µs, respectively. This is probably because that the kinetic energy of the projectile was progressively absorbed and dissipated by the fabric system, slowing the propagation of the transverse wave. This can be supported by Smith et al., who correlates the propagation of the transverse wave with projectile impact velocity in their investigations of a single yarn subjected to rapid impact loading (Stress–Strain Relationships in Yarns Subjected to Rapid Impact Loading).

Figure 13.

Fabric deformation of PW and (b) SL2S at an impact velocity of approximately 160 m/s.

Figure 14.

Post-impact close-ups of (a) PW and (b) SL2S at an impact velocity of approximately 160 m/s.

Multi-ply fabrics

In order to further investigate the ballistic performance of the stitched and unstitched plain weaves, penetration tests were performed on multi-ply fabric systems at impact velocities of approximately 400 m/s. The results are shown in Figure 15. It seems that the SEA of multi-ply systems is not sensitive to the number of plies. The SEA of PW keeps constant as the number of fabric plies increases, being approximately 17 J/kg.m². Similar trends were also observed on SL2S, SL4S, and SL6S. The energy absorption capability increases with the density of stitching lines for multi-ply systems. Nevertheless, it can be concluded that the stitched systems exhibited better performance over PW in a panel, for example, the improvement in SEA ranges from 31.8% to 64.4% when comparing SL2S and PW.

Figure 15.

The energy absorption capability of multi-ply systems at impact velocities of approximately 400 m/s.

Stitched fabric panels

When stitching a fabric, the additional weight of the sewing thread is non-negligible and must be considered when designing a ballistic panel with an improved performance-to-weight ratio. It was found from the ballistic results the improvement in energy absorption capability is reduced on mass-normalized metrics. This is because that the use of sewing thread resulted in a 100% increase in the mass of the fabric-forming yarns. In order to reduce the mass of the sewing thread involved in the ballistic fabric, single-ply plain fabrics were stacked up and stitched together to form stitched panels. In this paper, triple-ply fabric systems were selected to investigate the ballistic protection of stitched panels and plain weaves, and the results are shown in Figure 16. Figure 16 reveals the plot of the residual velocity of the projectile against its impact velocity for fabric systems. The ballistic limit of SL2T is approximately 50% greater than that of triple-ply PW, being 234 m/s. The difference between the stitched and unstitched fabrics in residual velocity is also noticeable at higher impact velocities. Figure 17 shows the SEA of the stitched panels and panels consisting of three layers of stitched single-ply fabrics at different impact velocities. It is interesting to find that the stitched panels show better performance than the stacked single-ply fabrics at all velocities. At an impact velocity of approximately 230 m/s, for instance, the SEA of SL2T is approximately one and half times greater than that of SL2S×3 and more than double that of PW × 3. The increase in SEA of the stitched panels is less pronounced at an impact velocity of approximately 400 m/s when compared with plain wave panels. The improvement is less noticeable in sparsely stitched samples, for example, the SEAs of SL4T are 63.8% and 23.9% higher than those of SL4S×3 at impact velocities of 240 m/s and 400 m/s, respectively.

Figure 16.

The ballistic performance of the stitched and unstitched triple-ply fabric systems.

Figure 17.

The energy absorption capability of multi-ply systems at impact velocities of approximately (a) 230 m/s, (b) 305 m/s, (c) 350 m/s, and (d) 400 m/s.

The ballistic results obtained from the penetration tests show that thread stitching provides an appreciable increase in the energy absorption capability of the UHMWPE plain weaves. Similar results have been reported by Bilisik²¹ and Zhou et al.²² in their investigations of the performance of stitched Kevlar plain weaves. Bilisik found that the stitched panels exhibited a more defined back face signature than the unstitched panels. Zhou et al. narrowed the stitching lines and discovered that stitching the fabric on every other yarn yields a 146% improvement in SEA when compared with the unstitched samples. Both the publications corroborate the benefit of thread stitching on fabric performance, encouraging us to further explore the penitential applications of dry woven fabrics for ballistic applications.

Finite element Results

For both the stitched and unstitched fabrics, simulations were performed at different impact velocities. The numerical predictions were compared with the deformation of the numerical model with images in Figures 11 and 13. Figure 18 compares the FE and experimental results for the variation of energy absorption as a function of the impact velocity for single-ply PW and SL2S. The numerical predictions and experimental results exhibit a similar trend, that is, the energy absorbed by the fabric decreases with an increase in the impact velocity. The energy absorption of the experimental work was greater than those of numerical predictions. The underestimation of fabric performance by the simulation can be attributed to the reduced size of the model, and probably less material was involved in slowing the projectile.

Figure 18.

Comparison of experimental results and FE predictions.

Figure 19 and Figure 20 show the fabric deformation and stress distribution of different systems at impact velocities of 160 m/s and 400 m/s. It can be seen in Figure 19(a) that the responses of PW in FE simulation corroborate with that in Figure 11(a), where yarn pull-out was evident, and yarn failure was not observed. SL2S successfully stopped the impacting projectile without yarn displacement and fiber damage (Figure 19(b)), which enhances the coupling between the primary and secondary yarns and enlarges the stress distribution area. Therefore, more material was involved in the energy dissipation process, and the SL2S absorbed more strain energy than the PW in Figure 21 b. Moreover, the transverse deflection of PW was less defined than SL2S in the numerical predictions, resulting in less amount of kinetic energy transmitted to the system (Figure 21(c)). Additionally, as there is less frictional sliding between the primary and secondary yarns, less amount of energy was dissipated by frictional effect in the stitched system than that in the unstitched system (Figure 21(d)). At an impact velocity of 400 m/s, the penetration of PW was still accommodated by yarn pull-out (Figure 20(a)), whereas SL2S was completely penetrated by the projectile at an early stage of the ballistic event (Figure 20(b)). Despite the early failure of SL2S, the stitched system exhibits a greater amount of energy absorbed during the impact process, with the exception of energy dissipated by friction. It is also interesting to find that both the systems responded more rapidly at higher impact velocities than at lower impact velocities.

Figure 19.

Fabric deformation and stress distribution of single-ply (a) PW and (b) SL2S at an impact velocity of 160 m/s.

Figure 20.

Fabric deformation and stress distribution of single-ply (a) PW and (b) SL2S at an impact velocity of 400 m/s.

Figure 21.

Evolution of (a) projectile kinetic energy loss, (b) fabric strain energy, and (c) fabric kinetic energy and (d) frictional energy at the impact velocities of 160 m/s and 400 m/s.

Conclusions

In this paper, the single and multi-plied UHMWPE woven fabrics were fabricated using the stitching technique to increase the yarn-yarn friction. Yarn pull-out tests were performed to characterize the influence of thread stitching, and the ballistic impact tests were performed at the velocity ranging from 0 ∼ 400 m/s to study the energy absorption capability of different fabric systems. It was found that the constraint provided by the stitching lines significantly increased the frictional force between the weft and warp yarns. The peak-load force of the sample with one stitching line is almost 10 times higher than that of the unstitched plain weave, and the maximum pull-out force increases with the number of yarns stitched. Nevertheless, stitching stiffens the woven fabric, making it less comfortable to wear when used in soft body armor interior systems. More importantly, thread stitching significantly improves the ballistic performance of plain woven fabrics. For instance, single-ply SL2S exhibited a ballistic limit that is 57% higher than that of PW. The SEA of the stitched fabric is over 50% greater than that of PW. High-speed photography showed that thread stitching constrains the yarn displacement and therefore eliminates the possibility of yarn pull-out. This enables the primary yarns to be well-engaged with the projectile at low impact velocities and to be stretched to fail at high impact velocities, increasing the amount of energy dissipated by the systems. The superior performance of SL2S was less pronounced in multi-ply systems. When the multi-ply plain weave systems were stitched together, the energy absorption capability was found to be better than layered up multi-ply systems, for example, the SEAs of SL4T are 63.8% and 23.9% higher than those of SL4S × 3 at impact velocities of 240 m/s and 400 m/s, respectively. In FE models, the element was merged at the nodes to simulate the yarn-constraining effect of thread stitching. The numerical predictions showed that thread stitching enlarges the area of stress distribution and transverse deflection and consequently enables the quilted systems to absorb more strain and kinetic energy than the PW system. In the stitched fabrics, the energy dissipated by yarn friction was much lower than the unstitched fabric due to the elimination of yarn pull-out during a ballistic event.

This work provides an alternative to improve the ballistic performance of plain weaves by increasing the friction between the crossover-forming yarns. Compared with conventional chemically related treatment, thread stitching produces limited mass and rigidity penalty on the fabrics. In addition, the performance of the stitched fabrics is not sensitive to service time, that is, this system can serve for a couple of years without significant property degradation. Those benefits offer potential as to the use of the stitched woven fabrics in flexible protective systems, for example, the engineering design of soft personal body armor.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the Hubei Province Science and Technology Project (2020BED004 and 2021BAA069), and National Natural Science Foundation of China (51708553).

ORCID iD

Yi Zhou

References

Yong

Peng

, et al. Predicting the tensile and compressive failure behavior of angle-ply spread tow woven composites. Compos Struct 2020; 234: 111701.

Dang

Liu

Zhang

, et al. Theoretical prediction for effective properties and progressive failure of textile composites: a generalized multi-scale approach. Acta Mech Sin 2021; 37: 1222–1244.

Strungar

Yankin

Zubova

, et al. Experimental study of shear properties of 3D woven composite using digital image correlation and acoustic emission. Acta Mech Sin 2020; 36: 448–459.

Karthikeyan

Russell

. Polyethylene ballistic laminates: failure mechanics and interface effect. Mater Des 2014; 63: 115–125.

Greenhalgh

Bloodworth

Iannucci

, et al. Fractographic observations on Dyneema composites under ballistic impact. Compos Part A Appl Sci Manuf 2013; 44: 51–62.

Julia

Bryan

. The effect of in-plane shear properties on the ballistic performance of polyethylene composites. Int J Impact Eng 2020; 143: 304592.

Academy

Boussinesq

, et al. On analysis of deformation and damage mechanisms of DYNEEMA composite under ballistic impact. Compos Struct 2020; 253: 112791.

Zhang

Satapathy

Vargas-Gonzalez

, et al. Ballistic impact response of ultra-high-molecular-weight polyethylene (UHMWPE). Compos Structures 2015; 133: 191–201.

Martínez-Hergueta

Ridruejo

Galvez

, et al. Influence of fiber orientation on the ballistic performance of needlepunched nonwoven fabrics. Mater Des 2016; 94: 106–116.

10.

Zhou

Xiong

, et al. The structural effects on the impact response of ultra-high-molecular-weight polyethylene plain weaves. Textile Res J 2021; 91: 911–924.

11.

Chen

Zhou

Wells

. Numerical and experimental investigations into ballistic performance of hybrid fabric panels. Compos B Eng 2014; 58: 35–42.

12.

US5935881 A . Bulletproof fabric and process for its production. Abiru, Shigeo, Iizuka, Kenji 1999.

13.

Calvin

Lisa

Edward

, et al. Optimization of soft armor: the response of single-ply para-aramid and ultra-high molecular weight polyethylene fabrics under ballistic impact. Text Res J 2020; 90: 15–16.

14.

Chu

Chen

Tian

. Modifying friction between ultra-high molecular weight polyethylene (UHMWPE) yarns with plasma enhanced chemical vapour deposition (PCVD). Appl Surf Sci 2017; 406: 77–83.

15.

Hasanzadeh

Mottaghitalab

Rezaei

, et al. Numerical and experimental investigations into the response of STF-treated fabric composites undergoing ballistic impact. Thin-Walled Struct 2017; 119: 700–706.

16.

Wang

Hazell

Shankar

, et al. Impact behaviour of Dyneema® fabric-reinforced composites with different resin matrices. Polym Test 2017; 61: 17–26.

17.

Zhou

Chen

Wells

. Influence of yarn gripping on the ballistic performance of woven fabrics from ultra-high molecular weight polyethylene fibre. Compos B Eng 2014; 62: 198–204.

18.

Mawkhlieng

Majumdar

. Designing of hybrid soft body armour using high-performance unidirectional and woven fabrics impregnated with shear thickening fluid. Compos Struct 2020; 253: 112776.

19.

Majumdar

Laha

Bhattacharjee

, et al. Tuning the structure of 3D woven aramid fabrics reinforced with shear thickening fluid for developing soft body armour. Compos Struct 2017; 178: 415–425.

20.

Sanchi

Abhijit

Bhupendra

. Soft armour design by angular stacking of shear thickening fluid impregnated high-performance fabrics for quasi-isotropic ballistic response. Compos Struct 2019; 233: 111720.

21.

Bilisik

Korkmaz

. Multilayered and multidirectionally-stitched aramid woven fabric structures: experimental characterization of ballistic performance by considering the yarn pull-out test. Text Res J 2010; 80: 1697–1720.

22.

Zhou

Zhang

, et al. Ballistic response of stitched woven fabrics with superior energy absorption capacity experimental and numerical investigation. Compos Struct 2021; 261: 111328.

23.

Martin

. Geometric and mechanical modelling of textiles. Nottingham, UK: University of Nottingham, 2007.

24.

Hearle

JWS

Shanahan

. 11-An energy method for calculations in fabric mechanics part I: principles of the method. J Textile Inst 1978; 69: 81–91.

25.

Emre

Howie

. On a multi-scale finite element model for evaluating ballistic performance of multi-ply woven fabrics. Compos Struct 2019; 207: 488–508.

26.

Documentation

AIA

. Abaqus analysis user’s manual, http://dsk.ippt.pan.pl/docs/abaqus/v6.13/books/usb/default.htm 2014. (accessed on 16 November 2021).

27.

McNeil

Standard

. Increased bending rigidity of wool fabric imparted by hybrid organic-inorganic sol-gel coatings. Textile Res J 2017; 87: 607–616.

28.

Guo

Chen

Zheng

. A semi-empirical design parameter for determining the inelastic strike-face mass fraction of soft armor targetsan analysis of system effect. Int J Impact Eng 2019; 125: 83–92.

The responses of stitched ultra-high-molecular-weight polyethylene woven fabrics upon ballistic impact

Abstract

Keywords

Introduction

Material and methods

Materials

Testing methods

The stiffness of fabric and yarn pull-out tests

Ballistic test

Finite element model

Results and discussions

Yarn pull-out behavior and fabric stiffness

The ballistic performance of the stitched and unstitched UHMWPE fabrics

Single-ply fabrics

Multi-ply fabrics

Stitched fabric panels

Finite element Results

Conclusions

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

References