Out-of-plane effects on dynamic pull-out of p -phenylene terephthalamide yarns

Abstract

In this study, the mechanical response of a single yarn pull-out from single layers of Kevlar® and Twaron® fabric under out-of-plane loading at dynamic rates was experimentally investigated, as ballistic applications typically occur at higher velocities and out-of-plane directions compared to previous literature. In order to perform the dynamic experiments, a pendulum impact setup was designed and constructed to pull out a single yarn dynamically. The pull-out load was measured directly by a load cell and the movement of the fabric was measured to portray the load–displacement history. The effects of fabric length, out-of-plane versus in-plane yarn pull-out, and constraining boundary conditions were compared.

Keywords

yarn pull-out dynamic out-of-plane pendulum impact

Aramid fiber-based woven fabrics have been in use in ballistic impact protective systems for a long time. Examples of these aramid-based materials include Kevlar® developed by DuPont, and Twaron® and Technora® produced by Teijin. These PPTA (p-phenylene terephthalamide) fibers are useful because of their low density, high strength, modulus, and flexibility, especially when applied in woven structures, and are therefore widely used in body armor and aircraft turbine engine fragment barrier applications. The ballistic impact resistance of these fabrics is known to be determined by various factors, such as the properties of a single fiber, the woven fabric structure, impacting projectile geometry and impact velocity, multiple ply interaction between the fabric layers, far field boundary condition effects from the impact zone, and inter-yarn friction.¹

It has been previously established that inter-yarn friction due to fabric weaving is an important mechanism when taking into consideration the kinetic energy dissipation from the projectile impact.² Kirkwood et al.³ have also demonstrated the significance of yarn pull-out analysis when predicting the ballistic performance of these bulletproof fabrics. This relation between yarn pull-out and inter-yarn friction suggests that the yarn pull-out mechanism is an important behavior to study the effect of inter-yarn friction in the woven structure. According to Dong et al.⁴ and Kirkwood et al.,⁵ yarn pull-out force in fabric has a correlation with the fabric’s performance of ballistic resistance, as fabrics with higher yarn pull-out force displayed better capabilities in resisting ballistic impact. The main factors influencing the yarn pull-out force have been identified as fabric weave style, material properties such as the yarn elastic modulus and ultimate strength and its constitutive fibers, fabric length, multiple-ply interaction, surface treatment, and pull-out loading rate.⁶

A great amount of literature exists on experimental, analytical, and numerical studies performed on the yarn pull-out mechanism on these PPTA materials; however, a large majority of these papers focus on in-plane pull-out and at quasi-static rates.^7,8 In real-life applications of these high-performance ballistic fabrics, projectile impact (and therefore yarn pull-out) occurs mainly in the out-of-plane direction rather than in-plane. The importance and significance of any potential out-of-plane effects is then immediately apparent when studying how projectile kinetic energy is being dissipated. In a previous study by Guo et al.,⁹ the rate effects of yarn pull-out on Kevlar® and Twaron® fabrics were investigated by pulling the warp and weft yarns from Kevlar® and Twaron® fabrics. In that study, a novel pendulum impact setup was designed to pull out the aramid yarns not only at a higher rate than previous studies, but also from an out-of-plane direction. An increase in the pull-out rates from 1 mm/s to 1 m/s resulted in an increase in peak uncrimping loads, as shown by the pull-out force histories. However, the previous paper did not focus on the out-of-plane effects, but rather, the comparison between the pull-out rates.

The main objective of this paper therefore is to investigate the inter-yarn frictional behavior of aramid plain-woven fabrics when subject to out-of-plane yarn pull-out. To achieve the objective of experimentally investigating the out-of-plane pull-out response of yarn at dynamic rates, the same pendulum impact setup was used. In particular, out-of-plane boundary conditions and potential geometry effects were investigated by varying the constrained width and principal fabric lengths, and then comparing the results to in-plane yarn pull-out at the same dynamic rates.

Experimental procedure

In this study, the fabric samples were prepared using a Point Blank Pathfinder Special® bulletproof vest manufactured in 2006 and 2008 by Point Blank Body Armor, and consisted of 600d Kevlar® and 500d Twaron® samples. No further specifications on the surface treatment were provided; therefore, the fabric samples were tested as-is. In these fabric layers, the warp direction yarns exhibit more crimp and were observed to have a thicker width than the weft direction yarns (Figure 1).

Figure 1.

Magnified view of plain-woven structure of (a) Kevlar® and (b) Twaron® fabrics as provided in the Point Blank Pathfinder Special® bulletproof vest.

As warp yarns are typically more crimped than weft yarns, the weave direction was also further verified by measuring the crimp ratios using the following equation:¹⁰

Crimp = \frac{L_{y} - L_{f}}{L_{f}} \times 100

(1)

where L_y and L_f are the yarn length and fabric length, respectively. The respective yarn and fiber properties are given in Table 1.

Table 1.

Specifications of extracted yarns and tested fabrics¹¹

Fiber	Kevlar®		Twaron®
Fiber diameter (µm)	11.22 ± 0.16		9.37 ± 0.12
Density (kg/m³)	1440		1450
Fiber modulus (GPa)	109.39 ± 5.45		152.49 ± 9.22
Ultimate yield strength (GPa)	4.83 ± 0.29		4.74 ± 0.32
Fiber failure strain (%)	4.42 ± 0.26		3.12 ± 0.22
Weave type	Plain		Plain
Yarn linear density (denier)	600		500
Yarn linear density (tex)	66.7		55.6
Ends per cm (warp)	12.0		11.5
Picks per cm (weft)	13.5		11.5
Crimp (%)	2.94 (warp)	1.01 (weft)	1.47 (warp)	1.07 (weft)

Dynamic experiments were performed using a pendulum setup inspired by the Charpy impact test (ASTM E23, 6110, ISO 148-1 standard) in order to achieve higher pull-out velocities compared to existing literature, replicating the same experimental setup as Guo et al.⁹ A schematic of this pendulum setup is shown in Figure 2. The fabric fixture is mounted on a linearly sliding cart (Edmund Optics EDM-37365A). The center yarn to be pulled out of the fabric is attached to a quartz force transducer (Kistler 9712B50) with a hook. The hook had a diameter of 1.0 mm, and therefore any pre-pull-out effects due to hooking of the principal yarn can be considered negligible. The quartz force transducer is then mounted onto a stationary post isolated from the pendulum fixture. The isolation of the force transducer from the rest of the setup eliminates the effects of inertia being recorded in the measured pull-out force histories. A piece of rubber is attached on the cart at the pendulum impact point in order to reduce damage to the setup as well as dampen mechanical vibrations from the impact. Figure 3 gives a full view of the pendulum impact experimental system.

Figure 2.

Pendulum setup with fabric fixture mounted on the linear sliding cart (a) before impact and (b) moving at velocity v after being impacted by the pendulum. The pull-out load is measured by the transducer, on which a hook is attached.

Figure 3.

(a) Top view and (b) side view of out-of-plane experimental setup.

For in-plane pull-out, a grip was designed to clamp only the principal yarn at one end, as in Figure 4. The grip clamps an approximate 3 mm length of principal yarn. The pendulum of mass 5.53 kg was raised to an initial vertical height of approximately 7 cm in order to achieve a theoretical post-impact cart velocity of 1.3 m/s. While higher velocities could be achieved theoretically, a higher pendulum impact velocity resulted in significant noise in the load signals and irregularity in the achieved cart velocities. Upon impact from the pendulum mass, the cart moves forward with a calculated velocity and the hook pulls out a single yarn. The cart displacement was measured using a magnetic linear band and encoder system (SIKO® MSK 230 and MB 320) mounted on the side of the linear cart system. Using the displacement–time history of the cart, the actual pull-out velocities achieved ranged between 0.9 and 1.2 m/s due to energy loss during impact and cart friction. A piece of rubber at the impact point between the linear slide and pendulum mass was used to reduce mechanical noise from shock.

Figure 4.

(a) Out-of-plane and (b) in-plane yarn fixtures for dynamic pull-out experiments. A hook was used for the out-of-plane pull-out, while a grip was designed to clamp on to the principal yarn in the in-plane direction.

For out-of-plane experiments, the opening width between the two aluminum plates was varied between widths of 2.5, 5.0, and 7.5 cm. This is achieved by having the screw holes on the aluminum fixtures appropriately placed to allow for variation in the constrained width.

The principal fabric length was also varied for both in- and out-of-plane pull-out experiments to investigate the effects of a longer yarn on the peak pull-out load. For out-of-plane experiments, the fabric length (and therefore the principal yarn length) was varied between 8 and 16 cm in 1 cm intervals. For in-plane experiments, the fabric length was varied at 1 cm intervals from 4 to 8 cm. The opening constrained width for the experiments was kept at a constant 2.5 cm. These dimensions of the in-plane and out-of-plane fabric samples were chosen so as to compare the effects of out-of-plane yarn pull-out. The fabric lengths of the in-plane samples are half of the out-of-plane fabric lengths since the principal yarn is ideally pulled symmetrically from the center of the out-of-plane fabrics, as schematically shown in Figure 5.

Figure 5.

The in-plane fabric lengths are half of the out-of-plane fabric samples as the pull-out mechanism is assumed to be symmetrical about the center of the out-of-plane fabric.

While a larger number of samples is typically preferred for statistical significance, the availability of fabric samples was limited. Therefore, for all parameters tested in this study, each experiment was performed three times to ensure repeatability. All cut samples have 1 cm wide tails on each side to reduce the effects of irregular fabric widths after cutting. The fabric sample was clamped between aluminum fixtures, with a 1-mm thick sheet of Teflon on either side to ensure full contact of the fabric with the fixture, and to reduce possible friction with the aluminum fixture.

Results and discussion

The yarn pull-out tests for these different parameters were performed, and the results for the yarn pull-out forces are presented in the following sections. For purposes of discussion regarding the frictional behavior during the yarn pull-out process, we qualitatively use the classical friction law relations in Equation (2):¹²

F_{friction} = f (μ, N, L_{f}, A_{contact})

(2{\rm a})

A_{contact} = f (F_{normal})

(2{\rm b})

In Equation (2a), the frictional force is a function of several parameters as shown above, where μ is the coefficient of friction, N is the yarn density (ends or picks per unit length in the principal direction), L_f is the fabric length in the principal direction, and A_contact is the real contact area between the yarns at each crossing point. Newer theories of friction attempt to explain the dependence of friction on the real contact area between the two surfaces by taking into account the microscopic asperities, which is in turn a function of the total normal force (Equation (2b)).

The behavior of yarn pull-out is usually portrayed in the form of a pull-out load versus displacement curve. Figures 6(a) and (b) show typical out-of-plane yarn pull-out responses under dynamic loading for a single layer of Kevlar® fabric at both in- and out-of-plane directions. At the initial stage of pull-out of the yarn, the originally crimped yarn undergoes the process of uncrimping and the fabric holding the yarn is simultaneously deformed toward the direction of pull-out and loading. After the fabric is fully deformed, the yarn is further uncrimped until it reaches the point where the pull-out load is maximum. Once maximum uncrimping is achieved and the yarn is fully straightened, the yarn being pulled out starts translating. In the translation stage, stick–slip response is clearly observed via an oscillatory behavior, as the yarn being pulled out is sticking when it passes under the yarn and slipping when it passes over the crossover points. The local maxima and minima are representative of the sticking and slipping motion in the yarn translation region, respectively. The mean value of the oscillation in the kinetic friction region also decreases due to the reduction in the number of the crossover points as the yarn is being pulled out. The same trend is observed regardless of pull-out rate or pull-out direction.

Figure 6.

Typical Kevlar® yarn pull-out curves for (a) in-plane and (b) out-of-plane directions.

Owing to the existence of high-frequency background noise produced by mechanical shock from the impact, the Savitzky–Golay smoothing filter was implemented to preserve local maxima and minima in the oscillatory motion. For in-plane experiments that were previously performed, the fabric deformation is in the form of a triangular extension symmetrical about the principal yarn.⁸ For out-of-plane yarn pull-out, this fabric deformation results in a pyramidal fabric deformation, as shown in Figure 7, which also plays a role in the force–displacement history of the yarn pull-out process, as the fabric deforms during the yarn uncrimping phase and contributes to the fabric displacement in the out-of-plane direction. This will be further elaborated upon.

Figure 7.

Out-of-plane fabric deformation when the center yarn is being pulled out.

Constrained width

The constrained width (i.e. the width between the two aluminum plates) was varied from 2.5 to 7.5 cm at 2.5 cm intervals. This parameter basically imposes a boundary condition on the perpendicular direction of the principal yarn. The constrained width parameter was tested for a fabric sample dimension of 10 cm × 10 cm for both Kevlar® and Twaron® yarns.

In this batch of fabric samples provided, the peak pull-out load for the Kevlar® fabric samples is larger than that for the Twaron® fabric samples due to a higher yarn density in the Kevlar® fabrics (i.e. more ends and picks per cm). The higher yarn density results in a tighter weave, which in turn results in a larger peak load; the effects of these geometric properties on the peak pull-out load are detailed by Pan and Yoon¹³ and Zhu et al.¹⁴ A more in-depth analysis on the same fabric samples comparing Kevlar® and Twaron® yarn pull-out trends is given by Guo et al.⁹ in a previous study.

As shown in Figure 8, it is interesting to note that the constrained width boundary condition does not affect the peak pull-out load significantly, if at all. This implies that a constrained out-of-plane displacement boundary condition in the opposing weave direction has no significant effect on the peak uncrimping load during yarn pull-out, but rather, the peak uncrimping force (and therefore peak pull-out energy) depends largely on the principal yarn. This observation is consistent with experimental results by Hudspeth et al.,¹⁵ who demonstrated that the change in yarn strength is minimal if the yarn is taken from a piece of impacted fabric just away from the impact point.

Figure 8.

Variation of peak uncrimping load with an increase in constrained boundary width.

Pull-out direction and fabric length

In order to compare the effects of pulling a yarn in the out-of-plane direction compared to pulling it in the in-plane direction, both directions were performed at dynamic pull-out rates. The fabric length was varied for both directions and the variation peak pull-out trends are depicted with the range of fabric length, as shown in Figure 9.

Figure 9.

Peak uncrimping force at dynamic pull-out rates against fabric length in the principal pull-out direction for (a) in-plane experiments and (b) out-of-plane experiments.

For the out-of-plane dynamic pull-out experiments, the peak loads were observed to increase with increasing principal fabric length, which is expected as the total friction acting on the principal yarn increases with a larger yarn density in the pull-out direction. However, the peak pull-out force for Kevlar® warp yarns shows a slowed rate of growth, that is, the pull-out force per end of the Kevlar® fabric begins to decrease. In order to study this trend in particular, the peak uncrimping force per end/pick for Kevlar® and Twaron® in the warp and weft directions was determined. The total number of yarns was calculated using the given yarn densities as in Table 1. These normalized pull-out trends are summarized in Figures 10(a) and (b).

Figure 10.

Uncrimping force per yarn for Kevlar® and Twaron® fabric for (a) in-plane pull-out and (b) out-of-plane pull-out.

From Equation (2), since the total friction acting on the principal yarn is a function of the total contact area, the normalized peak uncrimping force per crossover point should be expected to remain relatively constant, but the data depicted in Figure 10 (with trendlines to highlight the general pattern) show otherwise. For in-plane peak uncrimping forces, the normalized force is observed to decrease as the length of the yarn being pulled out increases. This trend is the most pronounced for the Kevlar® yarns in both weave directions; however, the Twaron® yarns seem to exhibit very slight or close to no discernible decrease in normalized peak uncrimping force with increased fabric length. In the out-of-plane direction, the normalized peak uncrimping force for Kevlar warp is observed to decrease the most drastically across the range of principal yarn lengths. For a more quantitative comparison, the linear fit curve slopes and the corresponding R² values are shown in Table 2.

Table 2.

Linear fit curve slopes and corresponding R² values for peak uncrimping load variation with increasing fabric length

	In-plane		Out-of-plane
	Slope (×10^–3 N/yarn)	R ²	Slope (×10^–3 N/yarn)	R ²
Kevlar® warp	–6.4	0.9765	–3.1	0.8759
Kevlar® weft	–5.6	0.9486	–0.3	0.0891
Twaron® warp	–1.8	0.9806	–0.5	0.1423
Twaron® weft	–1.9	0.2999	–0.7	0.3067

Given the numerical values of the curve fit slope, it is obvious that the decrease in uncrimping force is more pronounced for in-plane than out-of-plane pull-out. In particular, except for Kevlar warp yarns, the out-of-plane pull-out uncrimping force is relatively constant with increasing fabric length. This suggests that, within the range of fabric lengths tested, the uncrimping force is still increasing relatively linearly with respect to an increase in fabric length. From the trends observed in Figures 9 and 10, it is possible that there is a critical fabric length after which the peak uncrimping loads do not increase significantly, that is, the normalized peak uncrimping loads per yarn decrease asymptotically to zero. However, due to the limitations in allowable sample dimensions imposed by the provided fabric samples, further experimentation is required to adequately conclude the existence of such a phenomenon.

The effect of increasing the fabric length on the out-of-plane cart displacement at which the peak pull-out force occurs is also compared across the same fabric length range for out-of-plane directions, and these trends are summarized in Figure 11.

Figure 11.

Cart displacement at peak uncrimping load for out-of-plane pull-out tests.

It is apparent that the cart displacement at peak load increases with an increase in the fabric length from which the yarn is being pulled out. For a certain fabric length, we can easily calculate the theoretical fabric displacement based purely on the yarn uncrimping via basic trigonometry using the fabric length, yarn length, and crimp ratios, as shown schematically in Figure 12.

Figure 12.

Trigonometric representation of maximum out-of-plane fabric displacement height h due purely to uncrimping of yarn from fabric.

Using this simplification, we can then differentiate and estimate the fabric displacement compared to the cart displacement at peak uncrimping load, which also includes the out-of-plane deformation of the fabric. The theoretical maximum height due to uncrimping of the yarn is then given by the following equation:

h = \frac{L_{f}}{2} \sqrt{(1 + x)^{2} - 1}

(3)

where x is the crimp ratio as calculated using Equation (1). Since the maximum cart displacement at which the peak uncrimping load occurs includes the movement of the fabric in the out-of-plane direction, we can estimate the amount of energy transferred from the kinetic energy of the projectile to the kinetic energy of the fabric as it moves and warps. The yarn uncrimping energy is typically determined by finding the area under the region of the yarn pull-out curve up to the peak pull-out force, and this value was obtained by numerically integrating the obtained curves. In this case, the amount of uncrimping energy is estimated by integrating the curve up to the peak uncrimping height given by Equation (3), and the remaining energy is assumed to be used to warp the fabric in the out-of-plane direction until the principal yarn itself starts translating within the fabric. The ratios of the theoretical uncrimping energy to the fabric displacement kinetic energy are given in Figure 13.

Figure 13.

Average total peak energy variation with increasing fabric length. The trends of the proportion of fabric displacement energy to uncrimping energy are also plotted for (a) Kevlar® warp direction; (b) Kevlar® weft direction; (c) Twaron® warp direction; and (d) Twaron® weft direction.

Despite some slight fluctuations, the general trend of the peak pull-out energy increased with pull-out fabric length, which is expected as the total area in contact is larger, and therefore by taking into account the effects from Equation (2), the total friction force during the uncrimping phase is greater. It is also worth noting that the yarn uncrimping force becomes a more significant mechanism compared to the displacement of the fabric in the out-of-plane direction as the fabric length is increased. The crimp ratio of the fabric also seems to be a determining factor in whether the uncrimping energy percentage increases significantly. The warp yarns for both Kevlar® and Twaron®, having higher crimp ratios of 2.94% and 1.47% respectively, were observed to have larger uncrimping energy proportions compared to their respective weft yarn directions.

Finally, since we know that there is a huge complexity of the different mechanisms interacting with each other, we compare the peak pull-out energies of the out-of-plane samples to the in-plane samples. As per Figure 5, if there is no out-of-plane yarn pull-out effect, the out-of-plane peak pull-out energies should theoretically be exactly twice that of the in-plane peak energies. In view of this assumption, the in-plane peak energies were multiplied by two and compared to the out-of-plane peak energies, as shown in Figure 14.

Figure 14.

Ratio of peak out-of-plane pull-out energies to twice the in-plane peak pull-out energies.

Although there is no clear trend in the peak energy ratios with an increase in fabric length, it is obvious that the peak pull-out energies for both fabrics and both weave directions is much greater than the in-plane peak pull-out energies, with an overall series average of about 2.67. This suggests that the out-of-plane yarn pull-out mechanism is considerably significant and should be taken into greater consideration for future studies.

Conclusions

In this study, the dynamic mechanical response of the single yarn pull-out from both Kevlar® and Twaron® fabrics was experimentally investigated. A novel pendulum impact setup was developed for in- and out-of-plane yarn pull-out at targeted dynamic rates of 1 m/s. It was found that the boundary conditions imposed on the yarn being pulled out are much more significant than an imposed boundary condition away from the principal yarn. As the principal fabric length is increased, the total peak energy increases, but the proportion of this total peak energy also increases in the form of the peak uncrimping energy. The remaining projectile kinetic energy is dissipated in the form of kinetic energy of the fabric as it warps in the out-of-plane direction.

As a projectile impacts a woven fabric structure, some of the projectile kinetic energy is absorbed and dissipated by the inter-yarn friction at these crossing points through mechanisms such as yarn uncrimping, yarn extension, out-of-plane fabric displacement, and yarn translation. While the ballistic limit of bullet-resistant fabrics far exceeds the pull-out velocities examined in this study, the results of the dynamic pull-out for both Kevlar® and Twaron® fabrics show a marked increase in the uncrimping load and uncrimping energy, which suggests that a larger portion of the projectile kinetic energy is being absorbed through the yarn pull-out mechanism, at least for velocities below the ballistic limit. Further experimentation is required to determine the significance of yarn pull-out as an effective mechanism for energy absorption for velocities near or higher than the ballistic limit, although yarn breakage at these velocities would potentially present difficulties when isolating the effects of out-of-plane yarn pull-out.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Cheeseman

Bogetti

. Ballistic impact into fabric and compliant composite laminates. Compos Struct 2003; 61: 161–173.

Duan

. A numerical investigation of the influence of friction on energy absorption by a high-strength fabric subjected to ballistic impact. Int J Impact Eng 2006; 32: 1299–1312.

Kirkwood

. Yarn pull-out as a mechanism for dissipating ballistic impact energy in Kevlar® KM-2 fabric, Part 1: quasi-static characterization of yarn pull-out. Textil Res J 2004; 77: 920–928.

Dong

Manimala

Sun

. Mechanical behavior of silica nanoparticle-Impregnated Kevlar fabrics. J Mech Mater Struct 2009; 5: 529–548.

Kirkwood

. Yarn pull-out as a mechanism for dissipating ballistic impact energy in Kevlar® KM-2 fabric, Part II: predicting ballistic performance. Textil Res J 2004; 74: 939–948.

Sebastian

SARD

. Extension, displacements and forces associated with pulling a single yarn from a fabric. J Phys D Appl Phys 1987; 20: 130–139.

Gawandi

Thostenson

Gilllespie

JW Jr

. Tow pullout behavior of polymer-coated Kevlar fabric. J Mater Sci 2011; 46: 77–89.

Bilisik

Korkmaz

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

Guo

. Loading rate effects on dynamic out-of-plane yarn pull-out. Textil Res J. Epub ahead of print 27 March 2014. DOI: 10.1177/0040517514527376.

10.

Adanur

. Handbook of weaving, Boca Raton, FL: CRC Press, 2001, pp. 363–364.

11.

Lim

. Effects of gage length, loading rates, and damage on the strength of PPTA fibers. Int J Impact Eng 2011; 38: 219–227.

12.

Myshkin

Petrokovets

Kovalev

. Tribology of polymers: adhesion, friction, wear, and mass-transfer. Tribol Int 2005; 38: 910–921.

13.

Pan

Yoon

. Behavior of yarn pullout from woven fabrics: theoretical and experimental. Textil Res J 1993; 63: 629–637.

14.

Zhu

. Experimental study and modeling of single yarn pull-out behavior of Kevlar® 49 fabric. Composites Part A 2011; 42: 868–879.

15.

Hudspeth

. Degradation of yarns recovered from soft-armor targets subjected to multiple ballistic impacts. Composites Part A 2014; 58: 98–106.