Abstract
The aim of the research programme was to develop novel Kevlar®/polypropylene
composites for ballistic protection using a manufacturing process that enables
through-thickness reinforcement. The process involved two steps: (i)
needlepunching to form nonwoven Kevlar®/polypropylene substrates, and (ii)
compression moulding to manufacture the composites. Nonwoven substrates
were cut in 0°, 90°, -45⁰ and +45⁰ directions and moulded into unidirectional and
quasi-isotropic composites with 8, 16, 24, and 32 plies. Optimal moulding
parameters — time, temperature, and pressure — were determined using design
of experiments via Taguchi’s method. The test results showed that the throughthickness modulus of the composites was more than twice their tensile modulus,
underscoring the effectiveness of the through-thickness reinforcement process.
The Kevlar®/polypropylene composites were tested using spherical balls and
bullets. The ballistic limits, V50 for 8, 16, and 24-ply composites against 1.1g
stainless-steel balls were 352, 394, and 425 m/s, respectively. The 24-ply
composite met NIJ Level II requirements when tested with 9mm FMJ RN bullets
(398 m/s, 349 m/s) and .357 Magnum JSP bullets (413 m/s, 403 m/s, 357 m/s).
The 32-ply composite passed NIJ Level IIIA when tested with .357 SIG FMJ FN
bullets at 440 ± 10 m/s. In both 24- and 32-ply panels, bullets were embedded
after limited penetration, with low backface deformation due to strain distribution
over a large area of the composites.
The fractographic and failure analysis showed that tensile loading caused the
composites to split into ply groupings before final fibre failure. After ballistic
impact, shear-induced fibre failure was observed on the impact side, while ductile
deformation of reinforcements in a bulging mode occurred on the back. In fully
penetrated samples, ductile deformation progressed to tensile (tear) failure.
Additionally, 24- and 32-ply composites exhibited complex matrix/interface
cracking under ballistic impacts.
Analytical models based on experimental data predicted the ballistic limit of 24-
and 32-ply Kevlar®/polypropylene composites as proportional to area
density^0.18, indicating that further increases in area density would not improve
ballistic performance.
Finite element models for 24-ply composites under three-point bending and
ballistic impact showed good agreement with experimental results. Simulated
bending stiffness was slightly higher than experimental, as expected. Due to the
symmetrical configuration of the composite, natural axis i.e., zero bending stress
was placed in the middle of the composite, with approximately equal tensile and
compressive stresses at the top and bottom, respectively. Maximum S11
(longitudinal) stress occurred in topmost and bottommost 0° plies, while
maximum S22 (transverse) stress appeared in topmost and bottommost 90° plies.
Ballistic simulations closely matched experimental deformation at 349 m/s
(10.87 mm vs. 11 mm) but differed at 398 m/s (12.7 mm vs. 20 mm). Simulated
stress was localized and propagated through thickness. The ballistic simulation
agreed with the experimental observation in that strain distributed over a large
area of the composites, which contributed to low backface deformation.