PhD Studentship - Using Natural Language Processing and Molecular Simulations to Build Polymer Databases

How to apply: https://uom.link/pgr-apply-fap [uom.link]

This is a 3.5 year funded project open to UK students or EU students with settled status. You will receive a tax free stipend (depending on circumstances) set at the UKRI rate as a minimum (£19,237 for 2024/25).

A fully funded PhD position in multiscale modelling of polymer composites is available in the group of Professor Carbone at the University of Manchester from October 2024.

The use of machine learning methods and molecular simulations for polymer design and property prediction is the new frontier in polymer science. This project aims at using a mix of natural language processing (NLP) and high throughput simulations to build databases for polymer property which can then be used to aid down-selection and exploration of new material solutions. The aim of the project is to replace existing methodologies involving extensive experimental campaigns, which are generally expensive and time-consuming, even when supported by statistical design of experiments (DOE) analyses. The machine learning aspects will be targeted to identify regions of uncertainty in the parameter space, which are then to be supplemented by targeted and automated atomistic simulations techniques such as molecular dynamics and/or Monte Carlo simulations.

The project will initially be expected to focus on diffusion and solubility of gases in model polymers, though other material properties can also be explored using the same overall methodology. Of interest here is the development of the workflow and benchmarking against available data.

The project will involve the use of machine learning methods and molecular simulations of polymers and it is industrially sponsored.

Informal enquiries can be sent along with a CV to Dr Carbone at [email protected]

Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master’s in Physics, Chemistry, Chemical Engineering, Materials Science and other relevant STEM degrees. Pre-knowledge on molecular simulations, machine learning methods or programming is desirable but not necessary.

Before you apply please contact [email protected]