Single and multiple pull-out properties of a nano-processed para-aramid fabric structure were investigated. The nano pull-out behavior exhibited three distinct regions, namely crimp extension, interlacement rupture, and stick-slip. Multiple yarn pull-out tests demonstrated a significantly higher pull-out force compared to single-yarn pull-out, primarily attributed to the incorporation of nanoparticles. Furthermore, it was observed that an increase in fabric length resulted in an approximately linear increase in both yarn crimp extension and pull-out force. The highest pull-out force was obtained in the nano-hexagonal boron carbide (nh-B₄C, 0.3%) para-aramid structure, followed by multiwalled carbon nanotube (MWCNT, 0.3%) para-aramids. This is because of the enhancement of filament-to-filament friction, especially in the interlacement zone of fabric, alongside the cumulative frictional interactions among the nanoparticles. Additionally, the findings highlight an improvement in crimp extension energy absorption facilitated by nanoparticle incorporation in soft fabric. Notably, the improvement in the energy absorption capacity of yarns within the fabric, without disintegration, is considered significant at this stage. These results indicate a promising potential for performance enhancement in prospective soft ballistic applications.

Carbon nanotubes (MWCNTs) were synthesized in situ on para-aramid fabrics (gCPO) via a low-temperature (450 °C) chemical vapor deposition (CVD) process to enhance interfacial pull-out, frictional, and fracture toughness characteristics. FESEM analysis confirmed CNT coverage on fiber surfaces, while FTIR, Raman, and XRD results indicated limited structural modification without significant polymer degradation. The CNT-functionalized fabrics exhibited a 66.19% increase in maximum pull-out force, 55.32% improvement in interlacement rupture strength, and a three-fold rise in intra-yarn shear resistance compared with control fabrics (KPO). The static and kinetic friction coefficients increased by 26.67% and 16.67%, respectively, due to CNT-induced surface roughness, enhancing inter-fiber load transfer and reducing slippage. Single-yarn pull-out tests revealed notable gains in energy dissipation and fracture toughness (up to 1769 J/m²), whereas multi-yarn pull-out performance decreased due to excessive friction surpassing filament strength. The study demonstrates that low-temperature MWCNT growth enables effective interfacial reinforcement of soft para-aramid fabrics, establishing a novel framework for meso-scale mechanical screening of flexible nano-ballistic composites.

Interfacial yarn pull-out plays a critical role in load transfer and energy dissipation in soft ballistic materials; however, its multistage and friction-dominated nature makes comprehensive experimental characterization time-consuming and experimentally demanding. In this study, a stage-wise physics-guided machine learning (PG-HML) framework is proposed to predict the pull-out behavior of nano hexagonal boron carbide (nh-B₄C)-functionalized para-aramid fabrics using an experimentally constrained dataset. The pull-out response was decomposed into three physically meaningful deformation stages, namely crimp extension, initial interlacement rupture, and stick–slip sliding. Physics-based continuity and admissibility constraints were incorporated into the learning framework to preserve mechanical consistency across stage transitions and improve prediction robustness under limited data conditions. Comparative analyses demonstrated that the proposed PG-HML framework achieved superior predictive capability, particularly in capturing rupture transitions and post-peak stick–slip evolution, with R² values exceeding 0.98 during the crimp extension and rupture stages. Increasing nh-B₄C content enhanced interfacial friction, rupture resistance, and pull-out energy dissipation, while displacement responses gradually approached saturation under force-dominated extraction conditions. Therefore, interfacial pull-out behavior in nh-B₄C/aramid materials can be predicted with high fidelity using limited experimental input, providing a surrogate modeling strategy that enables virtual material screening and significantly reduces the experimental effort required for preliminary design and optimization of advanced soft ballistic materials.