Materials and Design 195 (2020) 109039

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Synthesis, processing and characterization of impact

hardening gel (IHG) reinforced Kevlar fabric composites
Shen Li a, Qianqian Fu a, Kun Qian a, Kejing Yu a,⁎, Hongfu Zhou b,⁎, Yunxuan Weng b, Zhongwei Zhang c
a
Key laboratory of Eco-Textiles, Ministry of Education, Jiangnan University, Wuxi, Jiangsu 214122, China
b
Beijing Key Laboratory of Quality Evaluation Technology for Hygiene and Safety of Plastics, Beijing Technology and Business University, Beijing 100048, China
c
State Key Laboratory of Explosion & Impact and Disaster Prevention & Mitigation, Peoples Liberation Army Engineering University, Nanjing 210007, China

H I G H L I G H T S

• With the unique needle-like structure, T-ZnOw could assist the construction of multiple parallel networks in IHG.
• The mechanical performance of Kevlar/IHG was improved significantly by the B-O dynamic cross-linking junction and particles.
• The Kevlar/IHG/IHGcore composites were customizable and had potential applications in wearable impact protection materials.

a r t i c l e i n f o a b s t r a c t

Article history: The impact hardening gel (IHG) reinforced Kevlar fabric composite is a new type of protective material prepared
Received 20 May 2020 by Kevlar fabric as dispersed phase and IHG as matrix. In this paper, IHG was strengthened by SiO2 and T-ZnOw
Received in revised form 3 August 2020 particles, and then the Kevlar fabric was compounded by dipping and sandwich lamination. With the help of
Accepted 4 August 2020
rheological testing and yarn pull-out testing, the relationship between microscopic shear rate and the macro-
Available online 7 August 2020
scopic linear velocity was established. To verify the enhancement effect of the material, the low velocity impact
Keywords:
testing was used to analyze the impact response and damage morphology. And the energy dissipation mecha-
Impact hardening gel (IHG) nism of the material was obtained. As a result, the reinforcing particles significantly increased the energy storage
Kevlar fabric modulus of IHG, and the friction between the fibers of Kevlar/IHG impregnated composites. The structure-
Reinforcing particles optimized sandwich material had a significant energy absorption effect, which was 41.9% higher than that of
Impact behavior the untreated Kevlar fabric. This provided the theoretical basis and reference for the optimal development of
lightweight anti-impact flexible protection materials.
© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction In recent years, modified IHG high-performance fiber materials, such

as Kevlar fiber [5], ultra-high molecular weight polyethylene fiber, PBO
In the process of law enforcement by the military and police, there fiber and other high-performance fibers have been developed by re-
will usually be encountered various types of dangerous supplies such searchers. These materials have been widely used in protective clothing
as knives, hammers and other external forces on the human body, as due to their well flexibility, wearing comfort, high specific strength, high
well as accidental impacts during training, especially flexible parts specific modulus and excellent textile properties [6–9]. In the case of the
such as joints. Although the emergence of high-performance materials, bullet-proof vests, they are usually composed of multi-layers of Kevlar
such as high-strength ceramics and high-strength alloys, have improved fabric, and the compact structures enable the external force energy to
the protective performance of body armors to a certain extent, these be quickly diffused and absorbed, achieving reliable protection of the
materials are generally rigid [1–3]. For rigid protective equipment, it is human body. However, they were unable to make the appropriate re-
difficult to effectively protect the flexible parts of the human body, sponse to various external stimuli. In addition, dozens of fabric layers
which greatly limits the movement of human body [4]. Therefore, the makes body armor heavy and bulky. To solve this problem, people
development of lightweight, flexible, low-cost materials with effective have been looking for new kinds of shielding materials with lighter
mechanical protection has become a research hotspot currently. weight and better performance.
Impact hardening material (IHM) [10], as an emerging material, has
⁎ Corresponding authors.
been widely known for the excellent mechanical properties, which is a
E-mail addresses: yukejing@jiangnan.edu.cn (K. Yu), zhouhongfu@th.btbu.edu.cn type of intelligent materials associated with strain rate. One of the ex-
(H. Zhou). amples is shear thickening fluid (STF). When STF was stimulated by

https://doi.org/10.1016/j.matdes.2020.109039
0264-1275/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2 S. Li et al. / Materials and Design 195 (2020) 109039

external forces (such as shear, impact, etc.), the apparent viscosity and IHG was dispersed into isopropyl alcohol impregnated Kevlar fabric
modulus would show a sharp increase, which indicates that it can rap- (200 g/m2) by ultrasonic agitation for 2 h (MIHG: M isopropyl alcohol =
idly change from a fluid state to a solid state reversibly. Since the end 1:2). Then, the obtained impregnation was dried in a vacuum oven at
of the 1980s, STF has been explored in the fields of damping, vibration 60 °C for 48 h to remove excess solvent. The obtained materials were
control and human body protection [6,11]. Some researchers proved named as Kevlar/IHG. The other method was making sandwich struc-
that the Kevlar fabrics could be enhanced after impregnating with STF ture composite [21]. The front and back pieces were both single layer
[12–15]. However, the defects with STF limit its application, such as of Kevlar/IHG fabric and the edges were stitched by sewing threads to
the instability, fluidity and long-term precipitation. Because of that, a re- maintain IHG inside. Different components of IHG used for core were
cent material, impact hardening gel (IHG) materials, was introduced. all weighed 20 g (Fig. 1 (a)). As a control, we used the method above
IHG is a boron-containing polymer viscoelastic material, which has a to replace IHG with RTV. For simplicity, we named the sandwich struc-
sensitive strain rate effect and stable performance. When suffering an tures as Kevlar/IHG/IHGcore or Kevlar/IHG/RTVcore.
external impact, it behaves like a solid but becomes very soft in nature
state [16–19]. The earliest reported IHG was a silicon boron copolymer 2.4. Characterization
flow-type material system prepared by DOW Corning under high tem-
perature conditions using a catalyst. This system shows a certain impact Fourier transform infrared (FTIR) spectroscopy was performed with
hardening performance. At present, researches mainly focus on the de- Nicolet Nexus 470 series FTIR spectrometer. It was used to characterize
sign of the material's macromolecular structure [16,18,20]. Few works the infrared spectra of IHG in the range of 4000–500 cm−1.
regarding the protective performance of particle modified IHG/fabric Dynamic rheological properties of IHG were tested using a rheometer
composites from the microscopic aspect has been reported. (Physica MCR 301, Anton Paar Co., Austria). Both steady shear experi-
Hence, this study aims to investigate the effect of different enhanced ments and dynamic oscillatory shear experiments were performed
particles (SiO2 and T-ZnOw)on the impact resistance performance of with a rotational rheometer. The storage modulus and loss modulus of
IHG reinforced Kevlar fabric composites microscopically. The effects of IHG and the composites were measured in the frequency range from
the particles morphology and content on the mechanical properties of 0.1 to 100 Hz under a deformation amplitude of 1% in the range of linear
the IHG and Kevlar/IHG were investigated. The rheological properties viscoelastic response. The viscosity was measured in the shear rate
of IHG were analyzed by means of steady shear experiments and range from 0.1 to 10 s−1. The geometry of parallel plates was employed
dynamic oscillatory shear experiments. And the relationship between with a plate diameter of 25 mm and the sample thickness of about
the shear rate and the linear velocity response was established. Addi- 1 mm.
tionally, the yarn pull-out testing and the low velocity impact experi- Scanning electron microscope (SEM) (SU1510, Hitachi Co., Japan) was
ment were used to validate the protection performance of Kevlar/IHG/ used to observe the morphologies of samples at different magnifica-
IHGcore. The macroscopic and microscopic action mechanism of IHG tions. The test conditions were vacuum conditions, and the working
modified Kevlar were revealed to provide insight into the optimization voltage was 5 KV.
of lightweight anti-impact flexible protection materials. Air permeability testing, referring to GB/T 5435–1997, was carried out
on a full-automatic air permeability meter (YG461E, Ningbo textile in-
2. Experiment strument factory, China). The test condition were from 0 to 100 Pa.
The samples were tested after being conditioned for 24 h in an environ-
2.1. Materials ment with the temperature of 25 °C ± 2 °C and the relative humidity of
65% ± 5%.
The materials included dimethyl silicone polymer (PDMS), boric Yarn pull-out testing, performed on MTS Criterion™ Model 43,
acid (BA), isopropyl alcohol, oleic acid, ethyl alcohol (from Sinopharm was used to examine the effect of IHG additives on the friction between
Chemical Reagent Co. Ltd., Shanghai, China), silicon dioxide (from Kevlar yarns by universal material testing machine (3385H, Instron Co.,
Evonik Industries AG, Shanghai, China) and four needles of zinc oxide America). The fabrics were cut into pieces in a size of 80 × 50 mm2
(T-ZnOw) (from Zhenyi Industry Co. Ltd., Shanghai, China). The di- (Fig. 1 (b)). The bottom was clamped. The upper yarn was pulled out
methyl silicone polymer, boric acid and ethyl alcohol were raw mate- by the grip at the speed of 5, 50 and 100 mm/min in constant.
rials to prepare IHG matrix. The isopropyl alcohol and oleic acid were
taken as dispersants and plasticizers respectively. SiO2 and T-ZnOw 2.5. Low velocity impact testing
were used as reinforcing fillers. Besides, RTV silicone rubber (type HY-
E615 from Hongyejie technology co. Ltd., Shenzhen, China) was the con- The low velocity impact testing [12,22] was conducted by the drop
trol group in contrast with IHG. Kevlar fabric (from Taihe New Material hammer impact machine (LC-2, Hengsi Shengda Instrument Co.,
Co. Ltd., Shandong, China) was a type of plain-woven high-performance China). The fabric (15 × 15 cm2) was securely clamped between two
aramid. steel plates. Only a circular area with the diameter of 76 mm of the fabric
was subjected to the impact (Fig. 1 (d, e)).The impactor hammer with a
2.2. Synthesis of IHG diameter of 16 mm weighs 5.5 kg (Fig. 1 (c)). The Kevlar fabrics and the
composites were tested. In addition, the IHG was tested by using the
BA was heated at 160 °C in a vacuum oven for 1.5 h to gain other fixture (Fig. 10 (a)). During the impact test, the accelerometer col-
pyroboric acid before use. PDMS and BA were premixed by a kneading lected the acceleration signals of drop tower and transformed them into
machine (Rugao xiyun machinery manufacturing co. Ltd., Jiangsu, electrical signals. The electrical signal amplified by the charge amplifier
China) at 210 °C for 4 h. Then, 0.5 ml of oleinic acid and different was transmitted to the computer for processing.
fillers of T-ZnOw or SiO2 (0, 1, 3 and 5 vol%) were added and reacted
for 1 h. The mixture was dried for 20 h at 60 °C to obtain IHG. For sim- 3. Result and discussion
plicity, the IHG with T-ZnOw or SiO2 fillers are named as IHG-T-ZnOw
(or SiO2) -X%. 3.1. Characterization of IHG and IHG composites

2.3. Finishing of anti-impact fabric Infrared spectroscopy was used as a method for chemical structure
analysis. The infrared spectra of IHG is shown in Fig. 2. The absorption
Two kinds of anti-impact fabric were finished with the IHG prepared peak at 3230 cm−1 was caused by –OH. The characteristic peak at
by the above steps. One finishing method was dipping process that 2960 cm−1 and 1340 cm−1 was caused by the stretching vibration of
 S. Li et al. / Materials and Design 195 (2020) 109039 3

Fig. 1. Schematic illustration of sandwich structure sample and experimental equipment. a) Sandwich structure (Kevlar/IHG/IHGcore). b) Yarn pull-out test. Low velocity impact testing:
c) device configuration, d) left view of the sample holder, e) up view of the sample holder.

-CH3 and B\\O bond, respectively. The absorption peak at 1260 cm−1 decreased sharply. This indicated that IHG and its derivatives had strong
corresponded to the stretch vibration absorption of Si-CH3 bond [20]. sensitivity to shear rate. Under high shear rate conditions, the entangle-
Therefore, the result could be concluded that IHG contains Si\\O, ment between macromolecules increased, and the viscosity of the
Si-CH3, B\\O and Si-O-B bonds. This was due to the formation of the system decreased [24]. Based on the rheological properties of the mate-
Si-O-B covalent bond at the terminal hydroxyl group of PDMS-OH rials, the effective range of resistance to shearing at high viscosity could
chain. The B-OH at the end of different molecular chains was further be estimated.
cross-linked by hydrolysis condensation reaction [23]. Then, the dynamic oscillatory shear experiments were conducted.
To explore the influence of enhanced particles in IHG, the rotational Fig. 3 (b) shows the response of dynamic storage modulus (G') to a
rheometer was used for the test. Firstly, we conducted dynamic oscilla- change in angular frequency for pure IHG and IHG-SiO2. The average
tory shear experiments. As shown in Fig.3 (a), when the shear rate was size of SiO2 particles was 1 μm, 2 μm and 5 μm. When the angular fre-
0.1 s−1, the viscosity values of the samples all reached 4 orders of mag- quency was increased, the shear rate on the materials also increased.
nitude. At the same time, the viscosity of IHG composites added en- Under the condition of high frequency, G' gradually tended to reach a
hancement particles were significantly greater than that of the pure platform value, which indicated that both of pure IHG and IHG-SiO2 sys-
IHG. Compared with SiO2 particles, the specific surface area of T-ZnOw tem have the tendency from viscosity to elasticity. When the filler vol-
particles was larger, and the viscosity of the system increased with the ume fraction was the same, the energy storage modulus of the IHG
adsorption amount of the polymer chain on the surface. With the in- with 1 μm was larger than 5 μm, which was related to the half-spacing
crease of shear rate, the viscosity of the three samples decreased slowly. between particles. According to Eq. (1), we could find that the half-
When the shear rate reached 4 s−1, the viscosity of the samples spacing of particles decreases with the decrease of particle size, which
causes the relaxation time of the molecular chain to become smaller
and the material rigidity to increase [25].
"
1 #
ϕm 3
h¼ −1  α ð1Þ
ϕ

where h is the half spacing between particles (m), α is the partical size
(m), ϕm is the maximum packed volume fraction of particles and ϕ is
the volume fraction of particles. By comparing the data in Table 1, it
can be also seen that the shear effect (RSTe%) of IHG was about 2 times
smaller than IHG-SiO2. The sample IHG-SiO2(b)-3% exhibited distin-
guished stiffening behavior. The calculation formula is defined in
Eq. (2). That may be caused by the adsorption and agglomeration be-
tween particles, which affected the dispersibility of the materials. As a
result, the dynamic response of the material was inhibited [26].

G0max −G0min
RSTe% ¼  100% ð2Þ
G0min

where the G'max and G'min storage modulus caused by the maximum and
initial shear frequency, respectively. The higher the numerical value of
Fig. 2. FTIR spectrum of pure IHG measured at room temperature. RSTe%, the better the impact hardening property.
4 S. Li et al. / Materials and Design 195 (2020) 109039

Fig. 3. a) Shear rate versus viscosity for IHG and its composites. b) Frequency sweeping tests of IHGpure and IHG-SiO2 in different silica particle sizes, c) frequency sweeping tests of
IHG-SiO2 in different volume fraction and d) frequency sweeping tests of IHG-T-ZnOw in different volume fraction.

To further study the influence of particle concentration and particle plastic characteristic. Once the angular frequency reached to 100 1/s,
structure on the rheological behaviors between SiO2 and T-ZnOw, we the G'max increased to 1.58 × 105 Pa, which exhibited remarkable rigid-
conducted the following experiments. Firstly, the rheological properties ity. We could calculate its relative shear effect as 5.52 × 105%, which was
of modified IHG materials with filler contents of 1, 3 and 5 vol% were in- the largest one among the frequency sweeping tests (Table 1). Fig. 4
vestigated. With the increase of the angular frequency, the storage mod- shows the SEM images of IHG modified by SiO2 and T-ZnOw. The SiO2
ulus of two materials showed the same rising trend. The range of change with the average diameter of 1 μm was agglomerate in IHG. However,
was greater than that of pure IHG. The IHG with 3 vol% SiO2 showed a the SiO2(5) and the T-ZnOw particles free-agglomerated. In Fig. 4 (e),
higher relative shear effect than the IHG with 1 vol% SiO2. The relative T-ZnOw showed a unique single-crystal microfiber structure. The
shear effect was the lowest when the content of SiO2 was 5 vol%. This angle between any two spicules were 109°, which made it easy to
was due to the fact that the half-spacing between particles decreased. achieve the uniform distribution in the matrix material [27]. According
When the concentration reached a certain degree, the remaining free to the research of Kourki [28], the 5% of T-ZnOw particle high-filling sys-
flowing dispersion medium between particles was less, and the fluidity tem contained three kinds of parallel networks namely: packing net-
was poor, which inhibited its dynamic response. In addition, the rheo- work formed between particles, adsorption network formed between
logical properties of SiO2(5) (Fig. 3 (c)) and T-ZnOw (Fig. 3 (d)) were in- particles and polymer chains, and polymer network formed between
vestigated. When excited by stress with the angular frequency at 0.1 1/s, polymer molecules (Fig. 4 (g)). This was equivalent to countless
the G'min of IHG-T-ZnOw-5% was 28.6 Pa, presenting a soft, fluidity and “quasi-crosslinking points”, which could improve the apparent
crosslinking density of IHG system.

Table 1
Gmin′、Gmax′ and RSTe(%) of samples in frequency sweeping tests.
3.2. Characterization of Kevlar/IHG composites

Sample Gmin′(MPa) Gmax′(MPa) RSTe(%) The Kevlar/IHG composites were prepared by dipping processes. In
IHGpure 7.33 × 10−5 0.146 1.99 × 105 order to explore the wear ability of Kevlar/IHG composite materials,
IHG-SiO2(1)-3% 9.51 × 10−5 0.312 3.28 × 105 the air permeability tests were conducted. Air permeability is one of
IHG-SiO2(2)-3% 8.89 × 10−5 0.247 2.78 × 105
the important indicators of fabric wear ability. As shown in Fig. 6, the
IHG-SiO2(5)-3% 6.17 × 10−5 0.294 4.76 × 105
IHG-SiO2(5)-1% 9.26 × 10−5 0.332 3.58 × 105 air permeability of the IHG-treated fabric was smaller than that of the
IHG-SiO2(5)-5% 1.18 × 10−4 0.320 2.71 × 105 neat Kevlar fabric. When the content of reinforcing phase particles in-
IHG-T-ZnOw-1% 3.30 × 10−4 0.265 8.02 × 104 creased, the air permeability decreased. The particles aggregating on
IHG-T-ZnOw-3% 1.86 × 10−4 0.247 1.33 × 105 the fabric surface resulted in the decrease of the fabric porosity and
IHG-T-ZnOw-5% 2.86 × 10−5 0.158 5.52 × 105
the flow rate of the passing gas per unit time. Fig. 5 shows the SEM
 S. Li et al. / Materials and Design 195 (2020) 109039 5

Fig. 4. SEM images of particles and IHG: a) The SiO2 particles with the average diameter of 1 μm, b) IHG-SiO2(1)-5%, c) the SiO2 particles with the average diameter of 5 μm, d) IHG-SiO2(5)-
5%, e) the T-ZnOw particles, f) IHG-T-ZnOw-5%. g) Three parallel network structures in IHG-T-ZnOw system.

images of the untreated Kevlar fabric and the composites. It was obvious
: Dω
that the random fibers on the fabric surface were significantly reduced γ¼ ð3Þ
after treating with IHG. The gap on the fabric surface was small (Fig. 5 2H
(c) and (d)). Due to the roughness of fabric was improved by surface en-
_
D γ̇ H
hanced phase, the Kevlar/IHG was fully filled with internal gaps. That v≈ω ¼ ð4Þ
made the composite structure more compact [27]. Moreover, the bind- 4 2
ing force inside and between fiber bundles increases, which was condu- :
cive to the diffusion and propagation of stress (Fig. 5 (e) and (f)). Where γ is the shear rate, ω is the angle speed, D is the diameter of the
parallel plate, H is the clearance of the parallel plate, and v is the pull-out
speed of the test. The linear velocity of the flat plate rotor had a corre-
3.3. Yarn pull-out testing sponding relation with the pull-out speed and according to the rheolog-
ical properties of the IHG, the range of pull-out speed which could
The dynamic mechanical properties of fabrics is highly related with trigger the shear thickening was about order of the magnitudes of
the friction between yarns. Therefore, we carried out the verification 5– 300 mm/min [12].
through yarn pull-out test. To explore the inter-yarn friction perfor- The typical tensile curve of the fabric during the yarn pull-out test is
mance of the fabric, we calculated the relationship between the rheo- shown in Fig. 7. The force vs. displacement curve assumed the oscilla-
logical shear rate and the linear velocity at a radius of 1/2 of the flat tion phenomenon until the warp was pulled out of the fabric
plate rotor to obtain an effective range for the experimental pull-out completely. The number of oscillations corresponded to the number of
speed [12]. The calculation formula was defined in Eq. (3) and (4): weft yarns through which the pulled warp yarn passed. When the

Fig. 5. SEM images of Kevlar fabrics and Kevlar/IHG composites. a-b) Neat Kevlar fabric, c-d) Kevlar/IHGpure, e-f) Kevlar/IHG(T-ZnOw-5%) and g-h) Kevlar/IHG(SiO2–5%).
6 S. Li et al. / Materials and Design 195 (2020) 109039

friction between the yarns had a significant increased. The particles

had a positive effect on the improvement of cohesion between the
yarns. The pull-out energy dissipations for untreated Kevlar fabric
(0.233 J) and IHG-treated one (1.11 J) were calculated and the average
energy dissipation of IHG-treated Kevlar fabric was about 376% higher
than the neat one [12,31]. Besides, the maximum pull-out force of
Kevlar/IHG(T-ZnOw) increased to 89.1 N, which was 39% larger than
that of the pure IHG-treated fabric and 5.4% larger than that of the
Kevlar/IHG(SiO2). This result demonstrated that the fabric with T-
ZnOw exhibited larger friction, so macroscopic pull-out force increased
with the microscopic particle-induced resistance. However, after passing
through three pull-out cycles, the peak force of Kevlar/IHG(T-ZnOw-5%)
showed slight decrease. The pull-out energy was also smaller than other
groups. The reason was that the filler network of T-ZnOw was damaged
and could not recover in a short time. And a larger part of mechanical en-
ergy was converted to thermal energy. As shown in Fig. 5, the additive
particles adhered to the surface of the Kevlar fibers. The SiO2 particles
were agglomerated, and the interaction between the yarns was relatively
small (Fig. 5 (d)). In Fig. 5 (c), the T-ZnOw reinforced IHG surrounded the
Fig. 6. Air permeability of Kevlar and Kevlar/IHG composites. yarns uniformly. When the yarns were pulled from the Kevlar fabrics, the
shear stiffening of IHG caused a hindering effect. The adhesion between
IHG and yarns was benefit from the irregular shape of T-ZnOw particles.
drawn yarn passed through the warp and weft intersections, a sine It could be concluded that the particle modified IHG limited the sliding of
curve was generated on the force vs. displacement curve, which the Kevlar yarns. The shear thickening effect and the particle friction
corresponded to the “slip-in-slip-out” phase of the yarns [29,30]. must be responded for the force increment in Kevlar/IHG.
After calculation based on Eq.(3) and Eq.(4), the speed of the upper
grip was set to 5, 50, 100 mm/min in the yarn pull-out test. Fig. 8
(a) shows the speed vs. force curve of the untreated fabric. At different 3.4. Low velocity impact testing
pull-out speeds, the force was kept at a low level (Fig. 8 (a)). However,
when the fabric was doped with IHG, the friction was increased, and In the low velocity impact testing, the hammer was launched by the
the pull-out force was changed significantly. When the pull-out speed energy storage device to impact the samples. The acceleration-time re-
was 100 mm/min, the peak of pull-out force was 65% larger than that lationship was recorded during the penetration process. Fig. 9 (a) shows
of the low-speed state (Fig. 8 (b)). This is because that the surface of the force vs. time curve of Kevlar and Kevlar/IHG(T-ZnOw-5%) under
the IHG-treated yarns was relatively rough. It could be seen that the the impact energy of 24.8 J. During the impact process, the growth
pull-out force of neat Kevlar yarns was not related on the pull speed, rate of the energy absorption value increased with time. The curves
and the IHG effectively worked in the dynamic process. On this basis, could be divided into three stages. In the first stage (I), the energy ab-
the pull-out property of different IHG-treated Kevlar fabrics was further sorption value was slow, which could be attributed to the elastic elonga-
studied. Fig. 8 (c) was the Kevlar/IHG composites with T-ZnOw particles. tion of Kevlar yarns. In the second stage (II), both the Kevlar and Kevlar/
At the speed of 5 mm/min, the peak force was 30.1 N, which was similar IHG(T-ZnOw-5%) exhibited a similar slope. The energy absorption was
to the peak force of Kevlar/IHGpure. However, at the speed of 100 mm/ mainly attributed to the large area contact between the samples and
min, the maximum pull-out force was about 89.1 N, which was 5 times the impact hammer. In the third stage (III), the IHG shortened the
larger than that of the untreated fabric and 1.3 times more than that of break time, and the unloading start time changed from 13.5 ms (Kevlar)
the Kevlar/IHG. Compare to the research of He [29] and Wang [12], the to 12.3 ms (Kevlar/IHG). As shown in Fig. 9 (b-e), the degree of yarn
pull-out force in this work was 3 times more than that of Kevlar/STF, slippage deformation of untreated Kevlar fabric was more severe than
which verified the perfect shear resistance and cohesiveness of IHG. that of Kevlar/IHG(T-ZnOw-5%). As described by He [29], the IHG in-
Fig. 8 (d) and Table 2 show the force-displacement curves of each creased the sliding resistance and limited the relative movement of
sample at the high speed of 100 mm/min. After the IHG was added, the the yarns, so more Kevlar yarns were loaded under impact.

Fig. 7. a) Principle of yarn pull-out testing: typical tensile curve. b) Schematic diagram of yarn pull-out process.
 S. Li et al. / Materials and Design 195 (2020) 109039 7

Fig. 8. Yarn pull-out testing at a constant speed of 5, 50, 100 mm/min. a) The pull-out force (N) vs. displacement (mm) curves of untreated Kevlar, b) Kevlar/IHGpure, c) Kevlar/IHG(T-
ZnOw-5%) and d) three types of composites at the speed of 100 mm/min.

In order to verify the impact resistance of IHG, we used a maximum penetration depths enlarged. As shown in Fig. 10 (b),
simple container to fix the IHG for test (Fig. 10 (a)). The force- the energy dissipation rate was increased from 36% to 55%. This re-
displacement curves were obtained by mathematical integration, sult provided the basis for the material structure design in this work.
and the energy dissipation was calculated. Fig. 10 (a) shows the im- In consideration of optimizing the wearability and verifing the
pact load at various initial impact energy ranging from 9.7 to 62.8 J. mechanical properties, we prepared two types of sandwich compos-
With the increase of initial impact energy, both peak forces and ites by using sandwich structure. The RTV rubber was used as a

Fig. 9. Performance characterization before and after treatment with IHG at the speed of 3 m/s with a loading of 5.5 kg. a) The energy-force-time curves of Kevlar and Kevlar/IHG(T-ZnOw-
5%). (b, c) Damage morphology of untreated Kevlar fabrics and (d, e) damage morphology of Kevlar/IHG(T-ZnOw-5%).
8 S. Li et al. / Materials and Design 195 (2020) 109039

Fig. 10. Impact testing results of IHG: a) the force-displacement curves of IHG and the image of the special impact installation and b) the energy dissipation. Impact testing result of
sandwich structure composites at the speed of 2, 3 and 4 m/s with a loading of 5.5 kg: c) the energy-force-time curves of Kevlar/IHG/IHGcore (T-ZnOw-5%), d) the force-displacement
curves of Kevlar/IHG/IHGcore (T-ZnOw-5%), e) the energy-force-time curves of Kevlar/IHG/RTVcore and f) the force-displacement curves of Kevlar/IHG/RTVcore (T-ZnOw-5%).

control. The preparation form is shown in Fig. 1 (a), and the impact the energy dissipation of Kevlar/IHG/IHGcore(T-ZnOw-5%) was signifi-
location is shown in Fig. 12 (a). In Fig. 10 (c), the peak force was cantly improved, which was 41.9% higher than that of the untreated
ranging from 1.52 KN (2 m/s) to 2.36 KN (4 m/s). This trend was sim- Kevlar fabric. It was obviously better than the silica-based sandwich ma-
ilar to the group of RTV (Fig. 10 (e)). However, under the different terials [29]. The above result further validated that IHG had a stronger
impact energy, all impact curves of Kevlar/IHG/RTVcore showed dras- impact resistance, and the impact resistance of IHG reinforced compos-
tic fluctuation, especially at 4 m/s, and the peak force of was smaller ites could be greatly improved.
than that of IHG core. That was due to the efficient energy dissipation
property of Kevlar/IHG/IHGcore. 3.5. Energy dissipation mechanism
To calculate the energy dissipation accurately, we integrated the force
vs. displacement curves. The slopes of Kevlar/IHG/IHGcore at the initial Different samples in the experiment had different response charac-
stage were significantly higher than that of RTV core (Fig. 10 (d) and teristics when subjected to impact. Due to the anisotropy characteristic
(f)). As shown in Fig. 11, the energy dissipation was 14.8 J of Kevlar, of the plain weave fabric, the untreated Kevlar fabrics dissipated energy
16.7 J of Kevlar/IHG, 21 J of Kevlar/IHG/IHGcore(T-ZnOw-5%) and 19.8 J mainly by slipping between yarns. If the warp or weft direction was sep-
of Kevlar/IHG/RTVcore(T-ZnOw-5%) in 4 m/s. After introducing T-ZnOw, arately stretched, the strain of the fabric was small. However, the
 S. Li et al. / Materials and Design 195 (2020) 109039 9

Table 2
Composite fabric static tensile test results.

Sample content Weight Maximum Displacement at Pull-out

gain (%) pull-out force maximum pull-out force energy
(N) (mm) (J)

Kevlar – 18.1 2.83 0.233

Kevlar/IHGpure 65.1 64.1 4.33 1.11
Kevlar/IHG 66.3 84.5 6.16 1.26
(SiO2–5%)
Kevlar/IHG 65.1 89.1 6.06 0.953
(T-ZnOw-5%)

time to disentanglement and displayed the fluid behavior macroscopi-

cally. When subjected to a large impact force, the B\\O bond failed to
broken instantaneously, which caused the entanglement point and the
cross-linking point locked with each other tightly. That was greatly hin-
dered the free movement of the molecular chain. Therefore, when the
Fig. 11. The energy dissipation during impacting process and the peak force.
longitudinal stress impacted to the Kevlar/IHG/IHGcore sandwich com-
posites, the IHG turned into solid-liked state immediately. During the
process, the stress was diffused and attenuated.
oblique tensile strain was large, and the maximum tensile strain existed
near the 45° and 135°. When the impact force applied to the surface of
the fabric, the 45° and 135° areas converged the stress propagation in 4. Conclusion
the warp and weft direction through the changes in the interlacing
angle of the yarns [32,33]. Since the warp and weft direction was In this work, the effects of different types of reinforcing particles and
bound by deformation, the yarn reduced the dissipated energy through particles content on the performance of Kevlar/IHG composites were in-
buckling degree firstly. When the limit was reached, the warp and weft vestigated. SiO2 and T-ZnOw particles were used to improve the
yarns were pulled out and bound by deformation, which was consistent strength and modulus of IHG. With the increase of the angular fre-
with the experimental phenomenon in Fig. 8 (b). After impregnation, quency, the storage modulus of IHG-T-ZnOw-5% increased by 4 orders
IHG inhibited the movement and buckling of the Kevlar yarns, but the of magnitude indicating the shear-hardening property, which benefit
energy propagation was unstable, and the impact hardening perfor- from the inherent attribute of IHG and the unique single-crystal micro-
mance was still not given full play. For the Kevlar/IHG/IHGcore sandwich fiber structure of T-ZnOw. Besides, the Kevlar/IHG/IHGcore prepared by
composites (Figs.12 (b–c)), IHG adhered to the fabric and strengthened combining modified IHG with Kevlar fabrics had the significant impact
the friction between the yarns, which was beneficial to the transverse protection property. The IHG effectively incorporated into the gap of
stress propagates to the surroundings and reduce stress concentration the Kevlar yarns, and increased the friction when the yarn sliding. The
[21]. According to existing research [23,26,34], the O atom contained low velocity impact testing showed that filling the IHG core on the
in the Si\\O structure in IHG could share the valence shell electrons basis of impregnated fabric could effectively change the way of stress
with the p-orbital of B, forming “B-O crosslinks”. The bond was dynam- diffusion, which could indicate the energy dissipation effect of
ically reversible, weaker than general chemical bonds, and similar in na- “1 + 1 > 2”. For Kevlar/IHG/IHGcore(T-ZnOw-5%), the energy absorption
ture to hydrogen bonds (Fig. 12 (d)). Fig. 12 (e) shows the multiple effect was 41.9% higher than that of untreated fabric. It was considered
coexistence forms and evolution processes of B\\O bonds in IHG. All that the O atom contained in the Si\\O structure in IHG could share the
the different forms connected and entangled IHG polymer molecular valence shell electrons with the p-orbital of B atom, forming a dynamic
chains together to form microscopic three-dimensional networks. reversible “B-O crosslink” with impact-hardening response characteris-
When the interaction rate of IHG was low, the B\\O bond had enough tics. The 5% of T-ZnOw particle high-filling system contained countless

Fig. 12. Schematic illustration of impacting energy dissipation. a) Impacting point, b) the untreated Kevlar fabric and c) the IHG treated Kevlar composite. d) The dynamic change of boron
cross-linking junction and e) the reversible change of B\ \O in different ways of combining.
10 S. Li et al. / Materials and Design 195 (2020) 109039

“quasi-crosslinking points”, which could improve the apparent [13] M. Liu, S.S. Zhang, S. Liu, S.S. Cao, S. Wang, L.F. Bai, M. Sang, S.H. Xuan, W.Q. Jiang, X.L.
Gong, CNT/STF/Kevlar-based wearable electronic textile with excellent anti-impact
crosslinking density of IHG system. Combining light-weight, flexible and sensing performance, Compos. Part A-Appl. Sci. 126 (2019) 105612.
and impact-resisitant performance, the Kevlar/IHG/IHGcore(T-ZnOw- [14] E.S.G. Güler, Rheological behaviours of silica/water, silica/PEG systems and mechan-
5%) is a promising candidate for body armors and other protective ical properties of shear thickening fluid impregnated Kevlar composites, Ind. Acad.
Sci. 41 (4) (2018) 112–119.
equipment. [15] T.A. Hassan, V.K. Rangari, S. Jeelani, Synthesis, processing and characterization of
shear thickening fluid (STF) impregnated fabric composites, Mater. Sci. Eng. A
Declaration of Competing Interest Struct. 527 (12) (2010) 2892–2899.
[16] R. Cross, Elastic and viscous properties of silly putty, Am. J. Phys. 80 (10) (2012)
870–875.
The authors declare that they have no known competing financial [17] W.F. Jiang, X.L. Gong, S. Wang, Q. Chen, H. Zhou, W.Q. Jiang, S.H. Xuan, Strain rate-
interests or personal relationships that could have appeared to influ- induced phase transitions in an impact-hardening polymer composite, Appl. Phys.
ence the work reported in this paper. Lett. 104 (12) (2014) 121915.1–121915.5.
[18] H.Y. Tan, Q.Q. Lyu, Z.J. Xie, M.M. Li, K. Wang, K. Wang, B.J. Xiong, L.B. Zhang, J.T. Zhu,
Metallosupramolecular photonic elastomers with self-healing capability and angle-
Acknowledgement independent color, Adv. Mater. 31 (6) (2019), e1805496.
[19] Y.P. Wang, S. Wang, C.H. Xu, S.H. Xuan, W.Q. Jiang, X.L. Gong, Dynamic behavior of
magnetically responsive shear-stiffening gel under high strain rate, Compos. Sci.
The authors gratefully acknowledge the financial support from the Technol. 127 (2016) 169–176.
National Key R&D Program of China (2018YFC0810300), Beijing Key [20] Z. Liu, S.J. Picken, N.A.M. Besseling, Polyborosiloxanes (PBSs), synthetic kinetics, and
Laboratory of Quality Evaluation Technology for Hygiene and Safety of characterization, Macromolecules 47 (14) (2014) 4531–4537.
[21] C.H. Xu, Y. Wang, J. Wu, S.C. Song, S.S. Cao, S.H. Xuan, W.Q. Jiang, X.L. Gong, Anti-
Plastics (PQETGP2019004), the Fundamental Research Funds for the impact response of Kevlar sandwich structure with silly putty core, Compos. Sci.
Central Universities (JUSRP51718A，JUSRP51907A). Technol. 153 (2017) 168–177.
[22] S. Arora, A. Majumdar, B.S. Butola, Structure induced effectiveness of shear thicken-
ing fluid for modulating impact resistance of UHMWPE fabrics, Compos. Struct. 210
References
(2019) 41–48.
[23] T.F. Wu, B.Q. Chen, Synthesis of multiwalled carbon nanotube-reinforced
[1] N. Prat, F. Rongieras, J.C. Sarron, A. Miras, E. Voiglio, Contemporary body armor: polyborosiloxane nanocomposites with mechanically adaptive and self-healing ca-
technical data, injuries, and limits, Eur. J. Trauma Emerg. Surg. 38 (2) (2012) pabilities for flexible conductors, ACS Appl. Mater. Inter. 8 (36) (2016)
95–105. 24071–24078.
[2] U.O. Costa, L.F.C. Nascimento, J.M. Garcia, S.N. Monteiro, F.S. da Luz, W.A. Pinheiro, [24] T.F. Tian, W.H. Li, J. Ding, G. Alici, H. Du, Study of shear-stiffened elastomers, Smart
F.D. Garcia, Effect of graphene oxide coating on natural fiber composite for multilay- Mater. Struct. 21 (12) (2012) 125009.
ered ballistic armor, Polymers-Basel 11 (8) (2019) 1356. [25] C.Q. Wu, X. Lin, D.Y. Ren, J. Zhang, Research progress in rheological behaviors of
[3] D.J. Luo, Y.W. Wang, F.C. Wang, H.W. Cheng, Y. Zhu, Ballistic behavior of oblique ce- highly filled polymers beased nanocomposites, China Plastics 32 (2) (2018) 1–9.
ramic composite structure against long-rod tungsten projectiles, Materials 12 (18) [26] S. Wang, W.Q. Jiang, W.F. Jiang, F. Ye, Y. Mao, S.H. Xuan, X.L. Gong, Multifunctional
(2019) 2946. polymer composite with excellent shear stiffening performance and
[4] Y.S. Lee, E.D. Wetzel, N.J. Wagner, The ballistic impact characteristics of Kevlar magnetorheological effect, J. Mater. Chem. C 2 (34) (2014) 7133–7140.
(R) woven fabrics impregnated with a colloidal shear thickening fluid, J. Mater. [27] X. Jin, M. Deng, S.R. Kaps, X.W. Zhu, I. Holken, K. Mess, R. Adelung, Y.K. Mishra, Study
Sci. 38 (13) (2003) 2825–2833. of tetrapodal ZnO-PDMS composites: A comparison of fillers shapes in stiffness and
[5] R. Roy, A. Majumdar, B.S. Butola, Comparative study of P-aramid based soft and stiff hydrophobicity improvements, PLoS One 9 (9) (2014), e106991.
composite panels for protective application, Fiber Polym. 20 (2) (2019) 406–412. [28] H. Kourki, M. Mortezaei, M.H.N. Famili, Prediction of the viscoelastic response of
[6] W.H. Boersma, J. Laven, H.N. Stein, Viscoelastic properties of concentrated shear- filler network in highly nanofilled polymer composites, J. Compos. Mater. 49
thickening dispersions, J. Colloid Interf. Sci. 149 (1) (1992) 10–22. (2015) 3799–3807.
[7] M. Grujicic, B. Pandurangan, J.S. Snipes, C.F. Yen, B.A. Cheeseman, Multi-length scale- [29] Q.Y. He, S.S. Cao, Y.P. Wang, S.H. Xuan, P.F. Wang, X.L. Gong, Impact resistance of
enriched continuum-level material model for Kevlar(a (R))-fiber-reinforced shear thickening fluid/Kevlar composite treated with shear-stiffening gel, Compos.
polymer-matrix composites, J. Mater. Eng. Perform. 22 (3) (2013) 681–695. Part A Appl. Sci. 106 (2018) 82–90.
[30] D.J. Zhu, C. Soranakom, B. Mobasher, S.D. Rajan, Experimental study and modeling of
[8] J.B. Qin, B.R. Guo, L. Zhang, T.W. Wang, G.C. Zhang, X.T. Shi, Soft armor materials con-
single yarn pull-out behavior of kevlar (R) 49 fabric, Compos. Part A-Appl. Sci. 42 (7)
structed with Kevlar fabric and a novel shear thickening fluid, Compos. Part B-Eng.
(2011) 868–879.
183 (2020) 107686.
[31] R.X. Bai, W.K. Li, Z.K. Lei, Y. Ma, F.Y. Qin, Q.C. Fang, X.G. Chen, Y. Chen, Experimental
[9] X.D. Zu, Z.X. Huang, W. Zhai, Performance of kevlar fibre-reinforced rubber compos- study of yarn friction slip and fabric shear deformation in yarn pull-out test,
ite armour against shaped-charge jet penetration, Lat. Am. J. Solids Struct. 12 (3) Compos. Part A-Appl. Sci. 107 (2018) 529–535.
(2015) 507–519. [32] K. Dapkūnienė, E. Strazdiene, Influence of layer orientation upon textile systems
[10] Y.K. Deng, X.C. Liang, X.P. Pei, K.K. Zhai, C. Wang, B.C. Zhang, Y.G. Bai, Y.M. Zhang, P.X. tensile properties. Part 2. Investigation of tensile energy and linearity, Mater. Sci.
Wang, Y. Tan, K. Xu, Self-healing ability and application of impact hardening poly- 12 (3) (2006) 247–252.
mers, Polym. Test. 76 (2019) 43–53. [33] K. Dapkūnienė, E. Strazdiene, Influence of layer orientation upon textile systems
[11] Y.T. Hu, S.Q. Wang, A.M. Jamieson, Rheological and flow birefringence studies of a tensile properties. Part 1. Investigation of tensile strain and resilience, Mater. Sci.
shear-thickening complex fluid - A surfactant model system, J. Rheol. 37 (3) 12 (1) (2006) 73–78.
(1993) 531–546. [34] N. Seetapan, A. Fuongfuchat, D. Sirikittikul, N. Limparyoon, Unimodal and bimodal
[12] Q.S. Wang, R.J. Sun, M. Yao, M.Y. Chen, Y. Feng, The influence of temperature on networks of physically crosslinked polyborodimethylsiloxane: viscoelastic and
inter-yarns fictional properties of shear thickening fluids treated Kevlar fabrics, equibiaxial extension behaviors, J. Polym. Res. 20 (7) (2013) 183.
Compos. Part A-Appl. Sci. 116 (2019) 46–53.