This fully funded PhD studentship is expected to commence in 2026 and will be for a duration of up to four years.

The successful candidate will develop and apply machine learning, molecular simulation, and statistical mechanical methods to investigate polymorphism and crystallisation in molecular crystals. The role will involve computational modelling, data analysis, and collaboration with researchers working at the interface of chemical engineering, chemistry, materials science, and artificial intelligence.

Studentship description

The ability to predict and control crystal polymorphism remains one of the grand challenges in chemical engineering, pharmaceutical development, and materials science. Different crystal forms of the same molecule can exhibit markedly different physical properties, including solubility, stability, manufacturability, and bioavailability. Recent advances in computational chemistry have raised the prospect of a “digital design” framework in which molecular simulations can be used to predict crystal structures, thermodynamic stability, crystallisation pathways, and ultimately process outcomes directly from molecular structure.

This PhD project will contribute to the development of next-generation computational approaches for understanding and predicting polymorphism in molecular crystals. The student will combine statistical mechanics, molecular simulation, and machine learning to address fundamental questions concerning the thermodynamic and kinetic origins of crystal form selection.

One research direction will focus on developing advanced free-energy methods for molecular crystals, leveraging recent advances in machine-learning potentials and generative models to enable accurate, computationally efficient predictions of polymorph stability at finite temperatures.

A second direction will explore the use of machine learning to identify and accelerate the slow collective processes that govern crystal nucleation and polymorphic transformations. In particular, the project will investigate graph-based and neural-network approaches to discovering collective variables that describe the emergence of crystalline order and rare-event dynamics in complex molecular systems.

The successful candidate will gain expertise in molecular dynamics, enhanced sampling, free-energy calculations, statistical thermodynamics, machine learning, and scientific computing. The project will be highly interdisciplinary and situated at the interface of chemical engineering, chemistry, physics, and artificial intelligence, with applications to pharmaceutical solids and advanced molecular materials.

Person specification

Applicants should hold, or be expected to obtain, a first-class or upper second-class degree (or equivalent) in Chemical Engineering, Chemistry, Physics, Materials Science, Mathematics, Computer Science, or a related discipline. The successful candidate will have strong quantitative and analytical skills, an interest in molecular modelling and computational science, and a willingness to work across disciplinary boundaries. Experience in one or more of the following areas would be advantageous: molecular simulation, statistical mechanics, thermodynamics, machine learning, scientific programming (e.g. Python), data analysis, or applied mathematics. Candidates should demonstrate the ability to work independently while contributing effectively within a collaborative research environment, together with excellent communication, problem-solving, and organisational skills.

Eligibility

This studentship is funded at the UK/Home fee rate and is therefore primarily intended for applicants eligible for UK/Home tuition fee status. Outstanding international applicants are encouraged to express their interest, as they may be considered should additional funding opportunities become available.

How to apply

Applications should be submitted through the provided link. Please nominate Prof. Matteo Salvalaglio as supervisor and include a statement of interest.

For informal enquiries please contact Prof. Matteo Salvalaglio at: m.salvalaglio@ucl.ac.uk