Multiscale Modelling of Composite Materials

Lead Research Organisation: University of Manchester
Department Name: Mathematics

Abstract

Objectives:
Build computational and mathematical multiscale models that can be employed to predict the mechanical properties (mainly linear elastic moduli) of particulate composite materials, initially focusing on simplified composites incorporating spherical inclusions (e.g. silica or microspheres) inside a homogeneous polymer matrix medium.
Extend models to more complex systems (multiple phases, and more complex fillers and matrix media)
Develop models that can incorporate high filler content, accommodating strong filler-filler interactions
Compare and iterate models with real world data

In achieving these objectives the project will answer the key question of how to model the effective mechanical behavior of complex composites incorporating effects at both the molecular and microscale. In particular a key difficulty is in incorporating length scale effects in the composite: the standard homogenization process leads to effective properties that are independent of the size of the particulate phase in the composite material. The methods developed, although focused on mechanical properties, could also be employed to determine other material properties, e.g. thermal conductivity.

Approach - three areas:
(i) computational (molecular dynamics) modelling at molecular length scales;
(ii) analytical (micromechanical) modelling at the microscale - homogenization techniques;
(iii) comparing predictions with real world data and feedback - are models capturing the observed effects? If not, why not?


Novelty:

Coupling (i) and (ii) is difficult and is frequently not done at all. Hence the two supervisors with respective expertise.

Length scale effects in (ii) are frequently not incorporated and it is this (e.g. when particles get close together) that needs to be modelled here, reflecting the need for molecular modelling techniques given the small length scales involved.

Methods developed will also contribute to broader questions with regard to materials optimization and digital design of materials.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/T517823/1 01/10/2020 30/09/2025
2481424 Studentship EP/T517823/1 21/09/2020 31/03/2024 Mark Mesbur