Flexible or soft armours though advantageous to conventional metallic armours are made of about 20-40 layers of para-Aramid fabrics such as Kevlar®. The large number of layers makes the armour heavy, rigid and expensive, which limits its use to high rank army personnel only. I embarked on a investigation into a probable method of minimizing this disadvantage, by replacing one or more layers of neat fabric with fabric treated with a suspension of microsilica in ethylene glycol. A part of this investigation was the formulation of a simple yet accurate yarn-level ABAQUS-FEM model which simulates the penetration of a neat fabric. The model is then analysed for various conditions of nose geometry and boundary condition.

Modelling viscoelastic deformation of amorphous polymers under varied environmental and loading conditions has been of immense interest both from scientific and industrial standpoints. There is a need of continuum scale constitutive theories can predict the myriad responses of polymers of different compositions to loads at varying temperatures, strain rates and moisture conditions. To this end I proposed unified, thermodynamically consistent, visco-elastic continuum scale models for different classes of polymers. The initial formulation was to predict the intrinsic mechanical responses of general thermoplastic polymers and block copolymers, particularly to model various features of their mechanical behaviour across glass transition. Then, I adapted the basic framework of these models to the mechanical response of hydrogels for varied moisture content by introducing a composition dependent glass transition temperature. The formulation for hydrogel predicts the kinetics of the water entrainment in the material, the swelling and moisture induced transition from visco-elastic to hyperelastic behaviour. Also, the effect of loading rate and changes in ambient temperature are also examined. All of the above models exploit the inter-molecular and intra-molecular mechanics of the polymeric network at the mesoscopic scale. These mesoscopic mechanisms possess different characteristic timescales of relaxation and all these timescales need to be captured as well as macroscopically bridged. This bridging of timescales and capturing of the relaxation is facilitated through an effective temperature thermodynamic framework. In the effective temperature based framework, the thermodynamic system is split into kinetic-vibration (K-V) and configurational subsystems and separate thermodynamic states (energies, entropies, temperatures and internal variables) for each subsystem are defined. This is followed by an exploitation of the laws of thermodynamics to establish restrictions on the forces and the fluxes. By implementing free energies (specific to the polymer) in the constitutive forms, I define the constitutive relations for stresses and evolution equations for temperatures as well as internal variables explicitly. The multi-temperature framework assumes a weak interaction between each configurational subsystem and the K-V subsystem, which forms the basis of structural relaxation. The models are validated against uniaxial compression or tension experiments for a range of temperatures, strain rates and moisture contents. In certain cases, such as for thermoplastic elastomers, we also carry out simulations involving cyclic loads to demonstrate Mullin’s effect. The models not only predict the intrinsic behaviour with accuracy but also physical ageing and shape memory behaviour. Being driven by physically inspired concepts, they should be far more generic than the existing approaches. Moreover, since the material parameters and variables involved have physical relevance, they may be experimentally obtained, in addition to enabling an exploitation of these formulations for response optimisation. In order to get some insight into differential geometry inspired constitutive models, I also proposed a Riemannian geometry based regularised- continuum damage model for compressible hyperelastic solids. The model was validated using a tensile experiment on a double notched plate with different notch sizes. In comparison with a quadratic phase-field approach this model showed faster convergence and could predict ideal brittle damage with less computational expense. 

This project was a part of my postdoctoral research at Weizmann Institute of Science under Prof. Eran Bouchbinder where we worked on dynamic fracture in polymeric gels and glasses. To this end, we build 3D numerical models that predict different features observed during fracture of gels and glasses. Some features are crack front waves, facets, microbranching and oscillatory instabilities. Our models would be able to determine the conditions that trigger the onset of these features and would be subsequently used to reveal the physics underlying the phenomena. Such models are typically used for understanding multi-modal fracture during earthquakes .