Predicting the effective mechanical and transport properties of heterogeneous media

We develop homogenization techniques to determine the effective properties of heterogeneous media (composites, multiphase materials, porous materials). We are particularly interested in mean-field theories which rely on a simplified relationships between the microfield statistics and can provide estimates in a semi-analytical form. Alternatively, we also perform detailed Finite Element simulations on Representative Volume Elements of the microstructure. The effective constitutive relations are implemented in the UMAT format of Abaqus, enabling the simulation of composite structures at macroscopic scale.

Our research focuses on nonlinear mechanical responses, including plasticity, elasto-plasticity, and damage, in diverse material systems: (i) particle- and fibre-reinforced composites, (ii) Dual-Phase and Martensitic steels, (iii) Porous Titanium as biomedical scaffolds.

We also investigate the effective thermodynamic and transport properties in two-phase media. These properties are important for a variety of emerging applications, including energy-storage materials and soft materials, where they further couple with the mechanical properties.

    2d_pp    woven.png     DPsteel.png

Modelling chemo-mechanical couplings in materials

The study of species transport coupled to mechanics in solids has recently come into sharp focus as lithium-ion batteries attracts considerable interest.  In an electrode of Li-ion battery, Li atoms are cyclically inserted into – and extracted from – a host solid.  The long-range transport and reaction of the Li atoms with the host alters the atomic structure and cause a range of mechanical phenomena: large, possibly inelastic deformations, anisotropic behaviour, damage, fracture.  Our research focuses on the formulation and numerical implementation of continuum theories that couple large plastic deformation to diffusion and chemical reaction.   We use the proposed models to gain understanding of experimentally observed phenomena.

Our modelling framework is general and can be used to model other systems in which diffusion and inelastic deformation are intrinsically coupled.  In recent works we proposed a hydrodynamic theory of supercooled liquids, and a non-classical theory of mixing limited by creep in binary mixtures.


Constitutive modelling of soft materials

Soft materials are quickly growing as future materials for emerging technologies in engineering and medicine. Among them, hydrogels consisting of 3D polymer networks swollen in water offer unique opportunities owing to their outstanding characteristics, including biocompatibility and tunable physical properties. However, the development of predictive theories and simulation tools for hydrogels is still in its infancy.

In our group we study the mechanical behaviour of model gels having well-defined network architectures, both experimentally and theoretically.  Our aim is to develop micromechanics-based models that can relate the network topology to the observed mechanical and swelling properties.  Modelling tools include Monte Carlo simulations, discrete network models and rubber elasticity theory.