Elsevier

Composite Structures

Volume 276, 15 November 2021, 114506
Composite Structures

Numerical investigation on the hybridization effect in inter-ply S2-glass and aramid woven composites subjected to ballistic impacts

https://doi.org/10.1016/j.compstruct.2021.114506Get rights and content

Abstract

Hybrid composites may be used to improve the ballistic properties of structures. These composites may be produced by enhanced manufacturing process that allow obtaining desired protection level and reducing costs. However, large variability in the material selection and architecture makes the correct choice a complex task that, in the past, has been mainly faced by experimental technique.

To reduce experimental efforts, the use of predictive modelling approaches is a current attractive possibility. This work focuses on numerical simulations of the ballistic impact on different combinations of inter-ply hybrid composites of woven Kevlar and S2-glass. The numerical analyses are carried out employing LS-Dyna MAT_162, which accounts for progressive inter- and intra-laminar failure modes. Some materials mechanical properties are obtained through tests and the parameters accounting for strain rate and softening are calibrated. This is crucial to obtain an accurate model, avoiding to address the calibration task in a pure numerical and parametric way. The effect of the stacking sequence and target thickness on the ballistic curves, the ballistic limit velocity and the absorbed energy by the target, are discussed. The numerical models show high accuracy in predicting the residual velocity, ballistic limit and damage features.

Introduction

Composite materials are used both in multilayer and monolithic ballistic shields and are often chosen when lightweight solutions are required, as they combine high strength with low density. Hybrid composites may be particularly interesting for ballistic protections since they combine different materials and constituents and can provide a better combination of properties that cannot be obtained in composites with a single reinforcement. More specifically, the coupling of glass and aramid fibres potentially adds high strength and stiffness to the structure, with an improved damage tolerance and impact resistance; while conceivably reducing costs of the protection [1].

Researchers try to obtain a beneficial effect from the hybridization of fibres in composites, when better properties than expected from the consideration of the rule of mixtures are achieved [2]. A positive or negative hybrid effect is defined by factors [3] like the loading configuration, the fibre arrangement and the relative content of the two fibres.

Some work has been reported on hybrid composites for ballistic protection. Cheon et al. [4] performed low velocity tests on different Kevlar®/glass fibre composite and found that hybridization with Kevlar improves the overall absorbed energy, which is maximized when the Kevlar layers are at the centre of the target. Park and Jang [1] analysed the effect of aramid layers on the impact properties of hybrid composites with glass by using dart impact tests. They reported that when the aramid layer was located at the back surface, the composite exhibited higher impact energy and larger delamination area. Ahmed et al. [5] investigated the failure under high-velocity impact of a hybrid three-dimensional orthogonal woven composite, reporting outstanding energy absorption capability for the hybridization of carbon and Kevlar fibres compared to the non-hybrid composites. In addition, the asymmetric hybrid laminates performed better when impacted on the stiffer carbon outer layer. The hybridization effect on energy absorption of Kevlar, glass and carbon inter-ply composites was explored by Randjbaran et al. [6] who found better ballistic performance when the glass fabric was the first impacted layer. However, the ballistic limit velocities were not determined, the tests were carried out with a cylindrical projectile and damage analysis was not carried out, which is vital for the understanding of the projectile penetration condition and the mechanisms governing energy absorption.

Numerical models and simulations of the ballistic impact on hybrid structures are scarce in the scientific literature and lack a comprehensive evaluation of all aspects involved. First, the material model needs to be based on reliable material characterization and validated parameters used in the analyses; secondly, the projectile needs to be reliably modelled, especially if real threats (i.e., standard bullets) are considered; and the ballistic curves for the studied structures are needed to set the ballistic limit velocity and to describe the panel response in a wide range of velocity.

Bandaru et al. [7] carried out hydrocode simulations and showed that the stacking sequence significantly affects the ballistic performance of carbon, glass and Kevlar inter-ply hybrid laminates, for a given thickness, but a detailed description of the panel damage and a ballistic curve for the panels was not included. Goda and Girardot [8] proposed a computational framework using the Abaqus software for modelling the ballistic impact on woven laminates with a cylindrical rigid projectile. They used continuum shell for modelling each ply with a user-defined material model and delamination was accounted for with a surface-based cohesive zone method. Their study involved pure carbon, pure glass and their hybrids; although the hybrids were only studied in the validation phase while the results phase only involved pure glass and carbon composites. Muñoz et al. [9] studied the performance of hybrid 3D carbon-glass woven composites under ballistic impact using X-ray failure micro-mechanisms analysis. The through-thickness yarns were modelled with truss linear embedded element and the method was considered validated and effective. They reported that energy dissipation was not significantly improved by the presence of the z-yarns although they enhanced energy dissipation for some mechanisms such as matrix damage by crushing. The specimens were asymmetrical hybrids, and it was found that energy dissipation enhanced if they were impacted on the stiffer carbon face. Okhawilai et al. [10] analysed glass and aramid fibre-reinforced polybenzoxazine/polyurethane composites impacted with a 7.62 × 51 mm projectile at 847 ± 9.1 m/s. The target was also shot at different positions to evaluate multi-hit protection characteristics. The results show good ballistic performance for the panel with S-glass plies on the entrance surface and Kevlar on the back and their numerical model showed good agreement with experimental tests. The ballistic impact defined by the authors is based on the number of hits and differs from the current work. In another work of Ahmed et al. [11], the ballistic performance of hybrid 3D orthogonal woven composites was numerically investigated with continuum shell elements to model the laminates and connector elements for the z-yarns, similarly to [9]. They used the Hashin criterium for intra-laminar damage and cohesive contact for delamination. The impactor had a hemispherical tip and was modelled with rigid elements. The numerical framework was considered effective in representing the failure mechanisms and the effects of the different constituents on the hybrid laminate.

Based on these reports, hybridization in composites can improve the impact resistance and performance in general, and the layup sequence significantly affects the results. Plies with higher energy-absorption potential seem to perform better when on the exit side of the laminate while stiffer plies are more effective on the entrance side. This work focuses on validating the capabilities of the numerical model to predict a ballistic impact against inter-ply laminates. An effective framework is presented in which the numerical model is calibrated and validated, resulting in models able to replicate ballistic performance and damage mechanism of hybrid composites. Validation is performed by comparison with numerical and previous experimental results of the group [12], covering the complete ballistic curve. Thus, the reliability of the model is assessed for different impact velocities covering complete perforation and projectile defeat. The studied threat is a standardized bullet making the simulation significant for designing ballistic protection since projectile deformation is also accounted for.

Section snippets

Experimental tests

The experimental tests used as the background and benchmarking for this work are taken from a series of ballistic tests previously reported by the group [12], where a detailed description of the manufacturing and the ballistic experimental campaign can be found. The laminates were manufactured using a vacuum infusion technique, detailed in da Silva et al. [13].

In brief, plain-weave Kevlar®29 (440 g/m2, 0.62 mm) and Hexcel® 8 harness satin S2-glass (302 g/m2, 0.24 mm) fabrics and epoxy resin

Material model

For the ballistic impacts analyses, a built-in material model identified as MAT-162 is used in the LS-Dyna software [17]. This model has been employed before for studying impact damage on composite structures [18], [19], [20]. Among its advantages, delamination damage is automatically considered without the need for adding features such as cohesive zone modelling (CZM) for inter-ply properties, but the model requires 34 parameters for material description.

Table 1 compiles the input parameters

Results

The comparison between the simulated numerical ballistic curves and the experimental ones for the pure laminates, in Fig. 6, shows the reliability of the model in reproducing the behaviour of the panels in terms of residual velocities, and validates the followed calibration procedure for the material properties. Good representation of the residual velocities in the linear part of the ballistic curve is observed for all panels, while the zone close to the ballistic limit shows a small

Conclusion

This work investigated the ballistic impact on composite inter-ply hybrid panels whose experimental behaviour is replicated in a numerical framework. The panels are manufactured as a single laminate with Kevlar29/epoxy and S2-glass/epoxy layers, using different lay-ups. The experimental results are used as a benchmark to check and calibrate the numerical model. The numerical model accuracy is assessed comparing the experimental and the numerical ballistic curves, reporting the deviation on the

Data availability

The raw/processed data required to reproduce these findings cannot be shared at this time due to technical or time limitations.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

The Italian Ministry of Education, University and Research is acknowledged for the support provided through the Project “Department of Excellence LIS4.0 - Lightweight and Smart Structures for Industry 4.0”.

References (36)

Cited by (24)

View all citing articles on Scopus
View full text