J Mater Sci (2017) 52:13977–13991

Polymers
POLYMERS

Effects of fabric folding and thickness on the impact

behaviour of multi-ply UHMWPE woven fabrics
Hongxu Wang1,* , Paul J. Hazell1, Krishna Shankar1, Evgeny V. Morozov1, Juan P. Escobedo1,
and Caizheng Wang1

1
School of Engineering and Information Technology, The University of New South Wales, Canberra, ACT 2600, Australia

Received: 30 June 2017 ABSTRACT

Accepted: 14 August 2017 This paper presents an experimental study on the dynamic response of multi-
Published online: ply woven fabrics subjected to impact by a spherical steel projectile in a velocity
22 August 2017 regime ranging from 120 to 200 m/s. A plain weave fabric made of SpectraÒ
1000 fibres was used to fabricate test samples having different numbers of layers
Ó Springer Science+Business and fold patterns. The effect of fabric folding was investigated by comparing the
Media, LLC 2017 impact performance of samples consisting of separate unfolded fabrics, con-
tinuous accordion-fold fabrics, and continuous roll-fold fabrics, respectively. It
was found that the perforation resistance and energy absorption capacity were
significantly improved by folding a fabric into multiple plies compared to the
unfolded counterparts, and the roll-fold fabrics performed best in this study.
Fabric folding seemed to have only a negligible effect on the back-face deflec-
tions of woven fabrics. The effect of thickness (or the number of layers) on
energy absorption efficiency was different among the tested fabrics, due to the
different inter-layer interactions depending on fold patterns.

Introduction ballistic fabrics has been greatly enhanced by the

improvements in fibre materials over the past few
Flexible woven fabrics made of high-performance decades. Currently, the most commonly used fibres
polymeric fibres are increasingly employed for pro- for impact protection are aramid fibres (e.g., KevlarÒ
tection against high-velocity impact threats, such as and TwaronÒ) and ultra-high molecular weight
in ballistic body armour [1–4] and propulsion engine polyethylene (UHMWPE) fibres (e.g., SpectraÒ and
containment systems [5–7], where high penetration DyneemaÒ) [8].
resistance and energy absorption are required. These Considerable research efforts have been put into
woven fabrics resist impact loading through a com- increasing the impact resistance and energy absorp-
bination of various energy absorption mechanisms, tion capacity of woven fabrics since the 1960s when
which include the decrimping, stretching, breakage, the first high-strength and high-modulus fibres were
and pull-out of constituent yarns within the fabrics. developed [9]. Cunniff [10–13] conducted a series of
The protection capability provided by modern impact studies on nylon, aramid, and UHMWPE

Address correspondence to E-mail: hongxu.wang@adfa.edu.au

DOI 10.1007/s10853-017-1482-y

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13978 J Mater Sci (2017) 52:13977–13991

textile-based armour systems with various yarn Yarn pull-out from woven fabrics during impact
deniers and weave types, which built an original process has been regularly observed in a number of
database for future work. Early research on the experimental and numerical studies [18, 23–26], and it
impact behaviour of woven fabrics has been well acts as a significant energy dissipation mechanism to
reviewed by David et al. [2], Cheeseman and Bogetti decelerate the impacting projectile [24, 26–28]. This type
[14], and Tabiei and Nilakantan [15]. These reviews of failure can only occur when yarn ends are free or
summarised a number of factors that influence the loosely gripped. In general, the greater the extent of free
impact performance of woven fabrics, specifically edges in a fabric, the more likely the occurrence of yarn
mechanical properties of constituent fibres, fabric pull-out. It has been found that both the force and energy
weave structure, projectile geometry, impact velocity, required to pull a yarn out of a fabric mesh increases
interaction of multiple plies, boundary condition, with yarn length [29–31]. This leads to a hypothesis that
friction, etc. As demand continues to increase for the longer length of continuous yarns in a folded fabric
more reliable and enhanced ballistic body armour, would result in more resistance to pulling and thus
recent research work starts to focus on optimising the better performance under impact loading. The currently
impact performance of woven fabrics via new con- used soft armour inserts for protection against handgun
cepts and ideas. Chen et al. [16] found that placing bullets consist of multiple layers of separate fabrics, the
woven fabrics in the front and unidirectional material edges of which are typically stabilised by sewing, gluing
in the rear of a hybrid panel provided better ballistic or seam sealing, in order to restrict yarn pull-out from
performance than the reverse sequence. Yang et al. fabric boundaries. An impact that occurs close to an edge
[17] investigated the effect of weave structure on the of the fabric armour insert usually results in lower bal-
impact behaviour of single-ply and multi-ply fabrics listic resistance [32]. The folded fabrics have potential to
with plain, 2/2 basket, 2/2 twill, and 4-harness satin improve this issue because the yarns are continuous
weaves; it was found that the plain weave fabric had over the fold edges between layers. Once the enhanced
the highest energy absorption capability among sin- impact performance of folded fabrics is confirmed, they
gle-ply fabrics, while weave structure was less can be used for soft armour inserts and stand-alone
influential on the impact behaviour of multi-ply ballistic vests. However, to the best of the authors’
fabrics. Nilakantan et al. [18] demonstrated a corre- knowledge, there has been little previous work con-
lation between the probabilistic impact behaviour of ducted to examine the effect of fabric folding on the
single-ply woven fabrics and the variability in inter- impact behaviour of multi-ply woven fabrics.
yarn friction by conducting yarn pull-out tests and In the present study, projectile impact tests have
ballistic impact tests on KevlarÒ fabrics. Wang et al. been carried out on unfolded and folded fabrics to
[19, 20] investigated the effect of ply orientation on investigate whether the impact performance of
the impact response of multi-ply aramid fabrics both woven fabrics can be improved by fabric folding.
experimentally and numerically, and they found Two fold patterns, accordion fold and roll fold, were
angle-laid fabrics led to a significant improvement in selected to fabricate multi-ply folded fabric speci-
energy absorption compared to conventional mens. The effect and mechanism of fabric folding
orthogonally aligned fabrics. Min et al. [21] carried were studied in detail by analysing the high-speed
out a similar study on the mechanisms of angle-laid videos of impact process and by examining the fabric
UHMWPE fabrics, and the results showed that angle- specimens after impact. The influence of thickness (or
laid fabrics were more energy absorbent than their the number of plies) on the impact behaviour of
aligned counterparts until the impact velocity woven fabrics has also been studied.
reached a certain level, after which this advantage
appeared to diminish. Tapie et al. [22] studied the
effects of pre-tension and incidence angle on the Experimental methodology
penetration resistance of single-ply woven fabrics,
and fabric performance was found to be governed Specimen preparation
primarily by four factors: the number of deformed
yarns, the strain energy absorption capacity of yarns, A plain weave fabric material made of SpectraÒ
the sliding of projectile against fabric, and yarn 1000 fibres was used to fabricate specimens for
mobility. impact testing. The mechanical properties of

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13979

constituent fibres [33, 34] and the basic parameters Impact testing
of fabric are summarised in Table 1. The micro-
graphs of fabric weave structure and yarn paths High-velocity impact tests were conducted using a
and cross sections are shown in Fig. 1, from which single-stage gas gun, which was powered by a com-
the geometric properties of fabric material, such as pressed gas reservoir connected to the gun barrel
yarn spacing, yarn path, and cross-sectional area, having a length of 2 m, as shown in Fig. 3a. The
were measured. The yarn spacing between the details of impact chamber at which the target fixture
centrelines of adjacent warp yarns was the same as was situated are shown in Fig. 3b. A spherical steel
that of weft yarns, and the measured average yarn projectile, with a diameter of 12 mm and a mass of
spacing was 1.50 mm. Both the warp and weft 7.05 g, was propelled to a predetermined velocity to
yarns have apparent crimp which refers to the yarn impact the specimen centre at normal incidence. The
undulation and is by virtue of fabric weaving. It projectile’s impact velocity was controlled by the
was found that the centreline profile of yarn path charged pressure in the reservoir. Two pairs of pho-
closely approximated to a sinusoid with amplitude toelectric sensors (OMRON E32-T16PR) were placed
of 80 lm and wavelength of 3000 lm, so the degree
of yarn crimp was 0.70%. The cross-sectional area (a)
of individual yarns was 0.074 mm2 calculated by
dividing the yarn’s linear density by the bulk
density of its constituent fibres.
Fabric specimens consisting of 4, 8, and 16 layers
were fabricated. In the modern multi-ply body
armour, fabric layers are usually orthogonally
aligned, primarily due to the simplicity of layering
process and low material consumption [19]. There-
fore, in this study, the fabric layers were aligned to
have the same warp/weft orientation. Three different
layups were used for making fabric specimens: sep-
arate unfolded fabrics, continuous fabric with accor-
dion fold (also called zigzag fold), and continuous
fabric with roll fold. Taking the four-ply fabric
specimens as an example, Fig. 2 illustrates the
unfolded and folded layups. For the separate unfol-
ded specimens, the fabric layers were sectioned into a
(b)
size of 250 mm 9 150 mm and hand-stacked ply by
ply to the desired number of layers. For the folded
specimens, a continuous fabric strip with dimensions
of 250 mm 9 (150 mm 9 number of layers) was cut
from the raw fabric material and then folded to the
designed pattern. Then, all the fabric specimens were
sewed using soft and weak cotton threads along the Figure 1 Optical micrographs showing: a the plain weave
short edges, in order to facilitate mounting them to structure of SpectraÒ fabric and b the paths and cross sections
the target fixture. of warp and weft yarns.

Table 1 Properties of SpectraÒ woven fabric

Fibre Tensile strength Tensile Elongation to Number of fibres Yarn denier (warp Weave Areal density
material (GPa) modulus (GPa) break (%) in a yarn and weft) pattern (g/m2)

SpectraÒ 2.9–3.3 97–113 2.9–3.5 120 650 Plain 95

1000

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13980 J Mater Sci (2017) 52:13977–13991

(a)

(b)

Figure 2 Schematic diagram of the unfolded and folded four-ply

fabric panels.

250 mm ahead of the fabric specimen to measure the

impact velocity. The impact tests were carried out at
three reservoir pressures which resulted in nominal
impact energies of 53, 83, and 139 J (or nominal
impact velocities of 122, 153, and 198 m/s). The
actual impact energies and velocities varied slightly
from test to test at the same set pressure. A box filled
with clay was placed behind the specimen to catch Figure 3 Experimental set-up of impact tests: a single-stage gas
the projectile after perforating it. During the impact gun and b impact chamber.
process, the impact chamber was fully sealed only
leaving a small hole on the front side to allow the horizontally parallel to the clamped edges (and the
projectile to pass through, so as to protect the sur- weft yarns vertically) in all the test specimens as
rounding equipment. shown in Fig. 4b. The clamping pressure was applied
A steel target fixture was designed to mount and by tightening four M10 bolts on each side, and a
secure fabric specimens in place, as shown in Fig. 4a. torque wrench was used to tighten each bolt to
When subjected to transverse impact loading, the 10 N m.
clamped edges of fabric specimen are highly prone to A high-speed camera (Phantom v710 at 18000
slippage due to the presence of in-plane membrane frames-per-second with an exposure time of 10 ls)
tension [35]. In order to increase friction and protect was employed to capture the projectile’s trajectory
the fabric material from being damaged by high and the transient impact response of fabric speci-
clamping pressure, a 1.8-mm-thick rubber lining was mens. The projectile’s impact and residual velocities
bonded to the inner surfaces of clamp using a high- in each test were determined by image analysis of the
strength epoxy adhesive. The fabric specimen was video data. The impact velocity calculated from the
gripped along the short edges at top and bottom over video matched closely with that measured by the
a width of 50 mm while the left and right edges sensors for every test. After impact, the performance
remained free, exposing a square target area of of each fabric specimen was assessed by calculating
150 mm 9 150 mm in the middle. For the folded the energy absorption and by measuring the dynamic
fabrics, the folds were along the vertical free edges of back-face deflection. For the perforated fabric speci-
specimens. The warp yarns were oriented mens, it was assumed that the loss of projectile’s

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13981

(a) (b)

Figure 4 Design of the target fixture: a schematic diagram and b a fabric specimen clamped at two opposite edges in the target fixture.

kinetic energy equalled the energy absorption by perforated fabric specimen was measured by tracking
fabric target (Et), which was calculated from the movement of its back-face centre (the impact
1
location) from the calibrated video data.
Et ¼ mp v2i  v2r ; ð1Þ
2
where mp is the mass of projectile and vi and vr are
the impact and residual velocities, respectively. For Results and discussion
the non-perforated fabric specimens, the projectile
was either stuck in the fabrics or rebounded with a Perforation status of tested specimens
negligible velocity, so the energy absorbed by fabric
target was taken as equal to the impact energy, given A total of 27 multi-ply fabric specimens were tested
by under the three nominal impact energy levels. In the
designing of armour systems, the foremost concern is
1
Et ¼ mp v2i : ð2Þ the prevention of material perforation by the projectile.
2 The perforation status of the tested specimens is listed
In addition to the perforation resistance and energy in Table 2. In this table and the following discussion,
absorption, the back-face deformation is an important the fabric samples are represented using a code ‘NX/
consideration in the engineering design of body Y’, where ‘N’ indicates the number of layers, ‘X’
armour [36], because it is critical to prevent behind denotes the fabric material (‘S’ for SpectraÒ), and ‘Y’
armour blunt trauma (BABT) which is a non-pene- denotes the fold pattern (‘U’ for Unfolded, ‘A’ for
trating injury resulting from the rapid deformation of Accordion fold, and ‘R’ for Roll fold). For example, the
armour material coving the human body [37]. eight-ply accordion-fold SpectraÒ fabric specimen is
Therefore, the dynamic deflection of each non- represented by 8S/A. As given in Table 2, none of the

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13982 J Mater Sci (2017) 52:13977–13991

Table 2 Perforation status of fabric specimens tested 160

Fabric sample Nominal impact energy 140 4S/U

4S/A
120

Absorbed energy (J)

53 J 83 J 139 J
4S/R
100
4S/U P P P 8S/U
4S/A P P P 80 8S/A
4S/R P P P 8S/R
60
8S/U NP P P 16S/U
8S/A NP P P 40
16S/A
8S/R NP NP P
20 16S/R
16S/U NP NP NP
16S/A NP NP NP 0
0 20 40 60 80 100 120 140 160
16S/R NP NP NP
Impact energy (J)
P perforated; NP non-perforated
Figure 5 Absorbed energy versus impact energy for all the fabric
16-ply specimens were perforated under the three specimens tested.
nominal impact energies. All the four-ply specimens
were perforated even under the lowest impact energy specimens in Fig. 6. The non-perforated specimens
level of 53 J. The eight-ply specimens remained are indicated by the check mark ‘4’, while all the
unperforated under the lowest impact energy level of other specimens in this figure are perforated. As can
53 J but developed perforations under the two higher be seen, the amount of absorbed energy increases
energy levels, except for 8S/R which did not fail under with impact energy for each type of fabric sample. It
the 83 J impact. is clear that the unfolded fabric specimens had the
lowest energy absorption capacity while the roll-fold
Energy absorption characteristics fabric specimens had the highest under all the impact
energy levels. In particular, the 8S/R specimen
The energies absorbed by fabric specimens are plotted absorbed about 30 J (about 20% of the impact energy)
against the actual measured impact energies in Fig. 5. more than 8S/U and 8S/A under the highest impact
The data points along the 45° dashed line represent the energy level of 139 J. Figure 7 shows the back-face
non-perforated specimens for which the absorbed central deflection histories of all the 16-ply fabric
energy equalled the impact energy, while the data specimens which were not perforated under the three
points from the perforated specimens lie below it.
Some trends are immediately apparent from this fig-
ure. Not surprisingly, as the areal density or the 4S/U 4S/A 4S/R 8S/U 8S/A 8S/R
number of layers increased, the fabric specimens pos- 120
111
sessed higher perforation resistance and energy
absorption capacity. It is worth noting that the energies 100
absorbed by fabric specimens increased with impact 86
83
79
Absorbed energy (J)

energy even when they were perforated. These results 80 74

indicate that the roll-fold fabric specimens performed 68

best, because the 8S/R specimen successfully resisted 60 56 55

50
83 J impact but 8S/U and 8S/A failed when subjected 47 47
to the same nominal impact energy level. The effects of 39
40 34 32 34 34
fabric folding and thickness are discussed in detail in 30 30
terms of energy absorption and back-face deflection in
the following subsections. 20

Effect of fabric folding 0

53 J 83 J 139 J

The absorbed energy is plotted against the nominal Figure 6 Energies absorbed by four-ply and eight-ply fabric
impact energy for the four-ply and eight-ply fabric specimens with different fold patterns.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13983

16S/U_53 J 16S/U_83 J 16S/U_139 J can clearly be observed emanating from the impact
16S/A_53 J 16S/A_83 J 16S/A_139 J point and travelling towards the clamped edges. At
16S/R_53 J 16S/R_83 J 16S/R_139 J the beginning of impact process, the primary yarns
45 are transversely deflected first and the initial defor-
40
mation shape is close to the curvature of projectile
(see the frames at 56 ls). Then, the remaining yarns
B ack-face central defl ection (mm)

35 within the fabric, namely secondary yarns, are driven

30 out of the original plane and engage in the impact
process by means of the yarn–yarn interactions at the
25
crossovers. As the fabric specimen is further
20 deformed, the vertical primary yarns are subjected to
15 high tensile stress, while the unrestrained horizontal
primary yarns become prone to slippage from their
10
original positions when the tension force exceeds the
5 frictional resistance. The decrimping, stretching, and
0 slippage of primary yarns decrease the weave tight-
0 250 500 750 1000 1250 1500 1750 2000 ness and therefore increase the yarn mobility in the
Time (µs) vicinity of projectile. Then, most of the primary yarns
Figure 7 Back-face central deflection histories of 16-ply fabric slide off from the projectile (see ‘A’ and ‘B’ in Fig. 8)
specimens with different fold patterns. due to the surface curvature and are pushed aside as
the projectile wedges through. In the unfolded fabric
impact energy levels. The peak back-face deflection specimen 4S/U, several horizontal primary yarns are
increases with impact energy until the fabric speci- pulled out from the weave (see ‘C’), and some of the
men is perforated, as is to be expected. The deflection vertical yarns are highly stretched (see ‘D’). In the
histories suggest that fabric folding has a negligible folded fabric specimen 4S/A, a lesser number of
effect on back-face deflection, since the deflection horizontal yarns are also stretched to a large extent
histories of fabric specimens with different fold pat- (see ‘E’) as are the vertical yarns (see ‘F’). Finally, all
terns are quite close to each other under the same these yarns slide off from the projectile and the pro-
impact energy level. jectile exits the target. The thicker fabric specimen
8S/U successfully resists the projectile impact under
Penetration mechanism the same energy level, as seen in the frame at 556 ls
in Fig. 8c when this specimen reaches the maximum
The transient response of woven fabrics undergoing deformation and the projectile is completely stopped.
impact can be studied by analysing high-speed It would appear that the transverse deformation and
videos and post-impact damage. Figure 8 illustrates yarn mobility of the front plies are constrained by the
the side profiles of three fabric specimens, 4S/U, 4S/ subsequent plies in a multi-ply fabric specimen [10].
A, and 8S/U, subjected to the same nominal impact The inter-layer frictional forces inhibit the sideways
energy at different instants during impact. Because motion of yarns in the front plies, so the primary
the woven fabrics have nearly zero bending stiffness, yarns are more restrained and not loose enough to
they behave as membranes and sustain the transverse allow the projectile to penetrate.
impact loading with in-plane tensile stresses. When a To understand the penetration mechanism further,
projectile impacts upon a woven fabric, longitudinal a single-ply fabric was tested under impact of 83 J.
and transverse waves are generated from the impact The post-impact state of this fabric is shown in Fig. 9.
point and propagate along the primary yarns, i.e. the It can be seen that there is a clear hole formed due to
yarns that come into contact with the projectile the projectile’s passage through the fabric. There is
[10, 14, 38]. The longitudinal tensile wave, travelling intense misalignment of orthogonal yarns due to
along the fibre axis at the speed of sound in the their relative motion around the impact region, and
material, cannot be observed in the recorded videos. fabric creasing is apparent over the entire exposed
The fabric material deflects in the direction of pro- area. Two vertical primary yarns were highly stret-
jectile’s trajectory forming a transverse wave, which ched and finally slid off from the projectile (see ‘A’),

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13984 J Mater Sci (2017) 52:13977–13991

(a)

(b)

(c)

Figure 8 Selected high-speed video frames showing the impact processes of a 4S/U, b 4S/A, and c 8S/U fabric specimens subjected to
nominal impact of 53 J (time unit: ls).

while the other vertical primary yarns were pushed ravelling can be observed at the two vertical free
aside by the projectile (see ‘B’). The decrimping edges (see the coloured areas), which is due to the
process is irreversible, and once stretched and inward movement of unconstrained horizontal yarns.
straightened, the yarn does not return to its original The degree of yarn ravelling diminishes gradually
length and crimped shape (see ‘C’). One horizontal from the middle to the top and bottom clamps along
primary yarn was completely pulled out from the the free edges.
fabric mesh, which left a gap in its original position Figure 10 displays the front and back views of
(see ‘D’), and two adjacent horizontal yarns were eight-ply fabric specimens with different fold pat-
partially pulled out by the projectile (see ‘E’). Yarn terns subjected to nominal impact energy of 83 J. Of

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13985

Figure 9 The back view of a

C
single-ply fabric subjected to
83 J impact.

Yarn ravelling
D E
C
D
A

B
Vertical
A E
Horizontal

Figure 10 Front and back views of 8S/U, 8S/A, and 8S/R fabric specimens subjected to nominal impact of 83 J.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13986 J Mater Sci (2017) 52:13977–13991

these three fabric specimens, only 8S/R was not fact, the yarns (or the constituent fibres) would fail in
perforated under this impact energy level. Clearly, tension when they elongate beyond their failure
the post-impact states of multi-ply unfolded fabrics strain. Because the friction coefficient of UHMWPE
are analogous to that of single-ply fabric. Yarn fibres against steel is about 0.1 and the fibre–fibre
translation and pull-out are prominent in each layer friction coefficient is only about 0.05–0.07 [34, 39], the
of the unfolded fabrics where the length of horizontal yarns are prone to lateral movement from their
yarns is short. In the folded fabric specimens, the original positions and slipping from the curved sur-
middle portion narrows due the pulling of horizontal face of steel sphere during the impact process.
primary yarns like wearing a girdle, and there is only Figure 12 shows the comparison of the post-impact
yarn ravelling at the two layers with free edges. The states of 8S/R fabric specimens subjected to different
details of yarn pulling at the edges of unfolded 8S/U impact energies. This fabric assembly successfully
and folded 8S/R fabric specimens are shown in resisted the impacts with nominal energies of 53 and
Fig. 11. The pulling of horizontal yarns towards the 83 J, but was perforated under the 139 J impact. It can
impact location leads to severe yarn ravelling and be seen that as the impact energy increases, both the
weave loosing at the edges of 8S/U specimen, vertical and horizontal primary yarns are stretched
whereas in the 8S/R specimen, the pulling of hori- more, leading to more severe narrowing in the mid-
zontal yarns tightens the fabric weave, which results dle and a larger extent of fabric creasing. It has been
in the increase in frictional resistance to yarn pulling. reported that the severity of creases in post-impact
Therefore, the inter-yarn energy dissipation within fabrics gives an indication of the degree of material
the same layer and between the adjacent layers of deformation and the amount of energy absorbed [23].
folded fabric specimens is higher. This is consistent with the results here, as the values
The clamped regions of fabric specimens are also of energy absorption were 53, 83, and 111 J under the
shown in Fig. 10, and it can be seen that the slippage three impact levels, respectively. A closer examina-
of fabric from the clamps is negligible using the tion of the impact sites (as shown in the lower row of
designed target fixture. No complete yarn breakage Fig. 12) reveals that fabric perforation is related to the
was observed in any of the perforated and non-per- local yarn mobility around the projectile. When sub-
forated specimens in this study; rather the primary jected to 53 J impact, the 8S/R specimen deformed
yarns were pushed apart laterally by the projectile to gently and the fabric weave at impact site is still very
accommodate its ‘wedge-through’ mechanism. In tight. Slight frictional sliding of vertical primary

(a) (b)

Figure 11 Pulling of primary yarns at the edges of 8S/U and 8S/R fabric specimens subjected to nominal impact of 139 J.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13987

50 mm

10 mm
53 J 83 J 139 J

Figure 12 Back views of 8S/R fabric specimens subjected to different nominal impact energies.

yarns away from the impact point is visible, resulting fabrics [18, 23, 40, 41]. The amount of absorbed strain
in the enlargement of gaps between the primary energy depends on the amount of fabric material that
yarns. Under the higher impact energy of 83 J, the is stretched during impact and the induced stress
projectile was able to push aside the primary yarns level. For the unfolded fabrics, strain energy is
more to create a small hole in the fabric specimen but mainly absorbed by the vertical yarns especially the
did not have enough kinetic energy to slip through. primary ones, whereas strain energy in horizontal
When the impact energy was further increased to yarns is small since they are pulled laterally inward
139 J, the primary yarns became slack enough to let and stretch much less, as shown in Fig. 10. The longer
the projectile slip through. The hole on the back-face continuous horizontal yarns increase their resistance
of perforated specimen is somewhat obscured by the to slippage and therefore are stretched more, facili-
stretched yarns. The results here suggest that using a tating more strain energy absorption by the hori-
fabric material with tighter weave and/or higher zontal yarns in the folded fabrics. Due to the two-
coefficient of friction where the yarn mobility is more sided clamping set-up, the energy dissipation asso-
constrained could provide higher penetration ciated with the pull-out of horizontal yarns accounts
resistance. for the major portion of absorbed frictional energy. It
In terms of energy absorbing components, woven was observed that the extent of horizontal yarn
fabrics dissipate the projectile’s kinetic energy by a movement and pull-out was much smaller in the
combination of several mechanisms. These mecha- folded fabrics than in the unfolded counterparts (see
nisms mainly include strain energy due to the Figs. 7, 9). However, it is not necessary that the fric-
stretching of yarns, frictional energy dissipation tional energy dissipation was also smaller in the fol-
associated with the frictional sliding interactions ded fabrics. It has been confirmed by previous
between yarns at their crossovers and between pro- studies [29–31] that the force required to pull a yarn
jectile and primary yarns, and kinetic energy arising out of a fabric mesh increases with the yarn length;
from the momentum transfer between projectile and therefore, the related frictional energy dissipation

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13988 J Mater Sci (2017) 52:13977–13991

may be higher even with a smaller pull-out distance 16

in the folded fabrics. Besides, the frictional interaction 13.9
14
between projectile and primary yarns during the 12.6

Energy absorption per ply (J)

wedging process induces energy dissipation as well. 11.7
12
For more precise quantification of the frictional 10.4
9.8 9.9
energy dissipated in the unfolded and folded fabrics, 10
an in-depth investigation by finite element modelling
8
must be undertaken. At the instant of maximum
deflection for a non-perforated fabric specimen (e.g., 6
556 ls in Fig. 8c), the projectile was completely
stopped, so its kinetic energy was dissipated by strain 4
energy and frictional energy as the fabric’s in-plane
2
and out-of-plane motions are negligible at this
instant. The amount of energy absorbed as the kinetic 0
energy of fabric specimen is a function of the mass of 4S/U→8S/U 4S/A→8S/A 4S/R→8S/R
fabric material that is deflected and the velocity at Specimen designation
which this deflection takes place [23]. Fabric speci- Figure 13 Average energy absorbed per ply for perforated fabric
mens of the same type, which get perforated under specimens subjected to nominal impact energy of 139 J.
impact, show increasing magnitudes of energy
absorption with increasing impact energies, as shown specimens showed an opposite tendency. This dif-
in the plots in Figs. 4 and 5. This may be attributed to ference in the effect of thickness comes from the inter-
the higher kinetic energy within the fabrics (due to layer interaction of fabrics which depends on fold
in-plane movement and out-of-plane deformation) at pattern. For the unfolded fabrics, each layer is sepa-
the moment of perforation imparted by the projectile rate and behaves similarly under transverse impact
at higher impact energies. The test results suggest loading, which is evident from the post-impact status
that the transfer of kinetic energy plays an important of the first and last layers of 8S/U specimen shown in
role in the absorption of projectile’s impact energy Fig. 10. Because the penetration of projectile into each
and its contribution continually increases with layer is dominated by the same wedge-through
impact velocity. mechanism, the energy absorbed per ply remains
nearly unchanged with the increase in thickness. In
Effect of thickness the roll-fold fabrics, the horizontal yarns are contin-
uous throughout all the layers, making their inter-
There is no doubt that the penetration resistance and layer interaction mechanism different from the
energy absorption capacity of woven fabric speci- unfolded fabrics. When the projectile impacts on the
mens improve with their thickness or number of plies front layers, the induced longitudinal tensile stress
due to more material being involved in energy propagates along the continuous yarns to the back
absorption, but the increment may not be necessarily layers, which generates additional tension in the back
linear with the thickness. It is of interest to find out layers. It has been shown that the energy absorption
how the thickness affects the energy absorption effi- capacity of woven fabrics increases with pre-tension
ciency of woven fabrics. Hence, the average energy over a certain range of low stress [22]. As the number
absorbed per ply of the perforated fabric specimens of plies increases in the roll-fold fabrics, it takes
subjected to nominal impact of 139 J is compared in longer time for the projectile to perforate the front
Fig. 13. It can be seen that the effects of thickness on plies, which gives rise to higher additional tension in
the energy absorption efficiency are different among the back layers before the projectile perforates them
the tested fabrics. Specifically, the energy absorbed and therefore enhances the energy absorption effi-
per ply of unfolded fabrics remained virtually unaf- ciency. In addition, fold pattern also plays an
fected when the thickness increased from four to important role in the inter-layer interaction since the
eight plies. The roll-fold fabric specimens exhibited accordion-fold fabric specimens show reduced
enhanced energy absorption per ply as the number of energy absorption efficiency as the thickness increa-
plies increased, whereas the accordion-fold fabric ses. The phenomenon of inter-layer interaction of

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13989

multi-ply fabric systems has not been clearly under- failure mechanisms of woven fabrics are dependent
stood, and there are controversial observations in the on the velocity and nose shape of the impacting
literature [10, 42]. Therefore, the effect of thickness on projectile. Therefore, further studies need to be car-
the energy absorption is not conclusive here, and the ried out on folded fabrics subjected to higher impact
interaction mechanism of multi-ply unfolded and velocities and impacts by projectiles with other
folded fabric systems requires further investigation. geometries such as right circular cylinder (RCC) and
fragment simulating projectile (FSP), before the
superiority of folded fabric panels can be fully
Conclusions established.

The work undertaken in this paper provides insights Compliance with ethical standards
into how fabric folding influences the perforation
resistance, energy absorption, and back-face deflec- Conflict of interest The authors have no conflicts of
tion in woven fabric systems under impact loading. interest to declare.
Impact tests were conducted on unfolded, accordion-
fold, and roll-fold multi-ply woven fabric specimens;
References
the penetration process and various energy absorp-
tion mechanisms including strain energy, frictional [1] Jacobs MJN, Van Dingenen JLJ (2001) Ballistic protection
energy, and kinetic energy were analysed in detail. mechanisms in personal armour. J Mater Sci
The experiment results have for the first time 36(13):3137–3142. doi:10.1023/a:1017922000090
demonstrated that the perforation resistance and [2] David NV, Gao XL, Zheng JQ (2009) Ballistic resistant body
energy absorption capacity of woven fabrics can be armor: contemporary and prospective materials and related
significantly influenced by fabric folding and the protection mechanisms. Appl Mech Rev 62(5):050802.
pattern of folding employed. The folded fabrics doi:10.1115/1.3124644
possessed enhanced energy absorption capacity, [3] Cavallaro PV (2011) Soft body armor: an overview of
compared with the separate unfolded fabrics, materials, manufacturing, testing, and ballistic impact
increasing by as much as 20% of the impact energy, dynamics. DTIC Document, Fort Belvoir
depending on fold pattern, thickness, and impact [4] Loverro KL, Brown TN, Coyne ME, Schiffman JM (2015)
energy. This is attributed to increasing the effective Use of body armor protection with fighting load impacts
yarn length and reducing the number of yarns that soldier performance and kinematics. Appl Ergon 46(Part
can be pulled out. But fabric folding seemed to have A):168–175. doi:10.1016/j.apergo.2014.07.015
only a negligible effect on the back-face deflections of [5] Naik D, Sankaran S, Mobasher B, Rajan SD, Pereira JM
fabric specimens. In addition, it was found that the (2009) Development of reliable modeling methodologies for
effect of thickness on energy absorption efficiency fan blade out containment analysis—part I: experimental
was different among the tested fabric specimens, studies. Int J Impact Eng 36(1):1–11. doi:10.1016/j.ijimpeng.
because of the different inter-layer interactions 2008.03.007
depending on fold pattern. Due to the spherical steel [6] Stahlecker Z, Mobasher B, Rajan SD, Pereira JM (2009)
projectile employed, penetration into UHMWPE Development of reliable modeling methodologies for engine
woven fabrics was dominated by a ‘wedge-through’ fan blade out containment analysis. Part II: finite element
mechanism in the range of impact velocities tested. analysis. Int J Impact Eng 36(3):447–459. doi:10.1016/j.
This study seems promising for the engineering ijimpeng.2008.08.004
design of soft body armour comprising woven fabrics [7] Bansal S, Mobasher B, Rajan SD, Vintilescu I (2009)
because it provides a potential way to improve the Development of fabric constitutive behavior for use in
ballistic performance by fabric folding without any modeling engine fan blade-out events. J Aerosp Eng
increase in weight. Since this concept is new, it 22(3):249–259. doi:10.1061/(ASCE)0893-1321(2009)22:
requires further investigations to thoroughly under- 3(249)
stand the inter-layer interaction mechanisms of fol- [8] Hazell PJ (2016) Armour: materials, theory, and design.
ded fabrics and therefore to optimise their impact CRC Press, Boca Raton
performance. Besides, no yarn breakage was [9] Hearle JWS (2001) High-performance fibres. Woodhead
observed in this study. The impact resistance and Publishing, Cambridge

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

13990 J Mater Sci (2017) 52:13977–13991

[10] Cunniff PM (1992) An analysis of the system effects in Impact Eng 28(2):207–222. doi:10.1016/S0734-743X(02)
woven fabrics under ballistic impact. Text Res J 00055-6
62(9):495–509. doi:10.1177/004051759206200902 [24] Kirkwood JE, Kirkwood KM, Lee YS, Egres RG, Wagner
[11] Cunniff PM (1996) A semiempirical model for the ballistic NJ, Wetzel ED (2004) Yarn pull-out as a mechanism for
impact performance of textile-based personnel armor. Text dissipating ballistic impact energy in KevlarÒ KM-2 fabric:
Res J 66(1):45–58. doi:10.1177/004051759606600107 part II: predicting ballistic performance. Text Res J
[12] Cunniff PM (1999) Dimensionless parameters for opti- 74(11):939–948. doi:10.1177/004051750407401101
mization of textile-based body armor systems. In: Proceed- [25] Ha-Minh C, Imad A, Kanit T, Boussu F (2013) Numerical
ings of the 18th international symposium on ballistics, San analysis of a ballistic impact on textile fabric. Int J Mech Sci
Antonio, Texas, pp 1303–1310 69:32–39. doi:10.1016/j.ijmecsci.2013.01.014
[13] Cunniff PM (1999) A design tool for the development of [26] Nilakantan G, Nutt S (2014) Effects of clamping design on
fragmentation protective body armor. In: Proceedings of the the ballistic impact response of soft body armor. Compos
18th International Symposium on Ballistics, San Antonio, Struct 108:137–150. doi:10.1016/j.compstruct.2013.09.017
Texas, pp 1295–1302 [27] Gawandi A, Thostenson ET, Gilllespie JW (2011) Tow
[14] Cheeseman BA, Bogetti TA (2003) Ballistic impact into pullout behavior of polymer-coated Kevlar fabric. J Mater
fabric and compliant composite laminates. Compos Struct Sci 46(1):77–89. doi:10.1007/s10853-010-4819-3
61(1–2):161–173. doi:10.1016/S0263-8223(03)00029-1 [28] LaBarre ED, Calderon-Colon X, Morris M, Tiffany J, Wetzel
[15] Tabiei A, Nilakantan G (2008) Ballistic impact of dry woven E, Merkle A, Trexler M (2015) Effect of a carbon nanotube
fabric composites: a review. Appl Mech Rev 61(1):010801. coating on friction and impact performance of Kevlar. J Mater
doi:10.1115/1.2821711 Sci 50(16):5431–5442. doi:10.1007/s10853-015-9088-8
[16] Chen X, Zhou Y, Wells G (2014) Numerical and experi- [29] Bazhenov S (1997) Dissipation of energy by bulletproof
mental investigations into ballistic performance of hybrid aramid fabric. J Mater Sci 32(15):4167–4173. doi:10.1023/
fabric panels. Compos B Eng 58:35–42. doi:10.1016/j.com A:1018674528993
positesb.2013.10.019 [30] Kirkwood KM, Kirkwood JE, Lee Young Sil, Egres RG,
[17] Yang C-C, Ngo T, Tran P (2015) Influences of weaving Wagner NJ, Wetzel ED (2004) Yarn pull-out as a mechanism
architectures on the impact resistance of multi-layer fabrics. for dissipating ballistic impact energy in KevlarÒ KM-2
Mater Des 85:282–295. doi:10.1016/j.matdes.2015.07.014 fabric: part I: quasi-static characterization of yarn pull-out.
[18] Nilakantan G, Merrill RL, Keefe M, Gillespie JW Jr, Wetzel Text Res J 74(10):920–928. doi:10.1177/
ED (2015) Experimental investigation of the role of fric- 004051750407401012
tional yarn pull-out and windowing on the probabilistic [31] Zhu D, Soranakom C, Mobasher B, Rajan SD (2011)
impact response of Kevlar fabrics. Compos B Eng Experimental study and modeling of single yarn pull-out
68:215–229. doi:10.1016/j.compositesb.2014.08.033 behavior of KevlarÒ 49 fabric. Compos A Appl Sci Manuf
[19] Wang Y, Chen X, Young R, Kinloch I, Wells G (2015) A 42(7):868–879. doi:10.1016/j.compositesa.2011.03.017
numerical study of ply orientation on ballistic impact resis- [32] Crouch IG, Arnold L, Pierlot A, Billon H (2017) Fibres,
tance of multi-ply fabric panels. Compos B Eng 68:259–265. textiles and protective apparel. In: Crouch IG (ed) The sci-
doi:10.1016/j.compositesb.2014.08.049 ence of armour materials. Woodhead Publishing, Melbourne,
[20] Wang Y, Chen X, Young R, Kinloch I, Garry W (2016) An pp 269–330. doi:10.1016/B978-0-08-100704-4.00006-2
experimental study of the effect of ply orientation on ballistic [33] Honeywell (2007) SpectraÒ fiber product information sheets.
impact performance of multi-ply fabric panels. Text Res J Honeywell International Inc., Morristown
86(1):34–43. doi:10.1177/0040517514566110 [34] Vlasblom MP, Van Dingenen JLJ (2009) The manufacture,
[21] Min S, Chu Y, Chen X (2016) Numerical study on mecha- properties and applications of high strength, high modulus
nisms of angle-plied panels for ballistic protection. Mater polyethylene fibers. In: Bunsell AR (ed) Handbook of tensile
Des 90:896–905. doi:10.1016/j.matdes.2015.11.019 properties of textile and technical fibres. Woodhead Pub-
[22] Tapie E, Tan ESL, Guo YB, Shim VPW (2017) Effects of lishing, Cambridge, pp 437–485. doi:10.1533/
pre-tension and impact angle on penetration resistance of 9781845696801.2.437
woven fabric. Int J Impact Eng 106:171–190. doi:10.1016/j. [35] Lee BL, Walsh TF, Won ST, Patts HM, Song JW, Mayer AH
ijimpeng.2017.03.022 (2001) Penetration failure mechanisms of armor-grade fiber
[23] Tan VBC, Lim CT, Cheong CH (2003) Perforation of high- composites under impact. J Compos Mater 35(18):
strength fabric by projectiles of different geometry. Int J 1605–1633. doi:10.1106/yrbh-jgt9-u6pt-l555

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

J Mater Sci (2017) 52:13977–13991 13991

[36] NIJ Standard–0101.04 (2000) Ballistic resistance of personal [40] Shim VPW, Guo YB, Tan VBC (2012) Response of woven
body armor. U.S. Department of Justice, Office of Justice and laminated high-strength fabric to oblique impact. Int J
Programs, National Institute of Justice, Washington Impact Eng 48:87–97. doi:10.1016/j.ijimpeng.2011.06.008
[37] Cannon L (2001) Behind armour blunt trauma—an emerging [41] Grujicic M, Bell WC, He T, Cheeseman BA (2008) Devel-
problem. J R Army Med Corps 147(1):87–96. doi:10.1136/ opment and verification of a meso-scale based dynamic
jramc-147-01-09 material model for plain-woven single-ply ballistic fabric.
[38] Miao M (2016) Dynamic modulus and strain wave velocity J Mater Sci 43(18):6301. doi:10.1007/s10853-008-2893-6
in ballistic fibre strands. J Mater Sci 51(12):5939–5947. [42] Lim CT, Tan VBC, Cheong CH (2002) Perforation of high-
doi:10.1007/s10853-016-9895-6 strength double-ply fabric system by varying shaped pro-
[39] Marissen R (2011) Design with ultra strong polyethylene jectiles. Int J Impact Eng 27(6):577–591. doi:10.1016/
fibers. Mater Sci Appl 2(05):319–330 S0734-743X(02)00004-0

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

onlineservice@springernature.com