Overview
This project aims to bridge the gap between computational predictions and real-world synthesis. You’ll join a collaborative team, working at the cutting edge of materials discovery to make more realistic predictions of the stability of materials at the real-world synthesis temperatures by integrating machine learning, thermodynamics and disorder modelling into traditional computational chemistry methods.
About this opportunity
New materials are critical for technological advances, and this project tackles a major challenge in the field: the gap between computational predictions and real-world experimental synthesis. You will develop next generation methods by integrating machine learning and thermodynamic modelling to predict synthetically accessible structures with greater accuracy. Current crystal structure prediction workflows are useful in selecting target compounds but largely rely on computation of energies rather than free energies. Better descriptions of free energy can be achieved by capturing the finite temperature behaviour of materials and including disordered materials in stability assessment. This project will combine traditional computational chemistry tools with machine learning models to improve the accuracy of predictions. This builds on our achievements in digitally targeted discovery (Science 383 739 2024) and comprehensive description of disorder in crystalline materials (J. Appl. Cryst. 58 659 2025).
The student will join an integrated team of computational and experimental researchers, providing close collaboration and a feedback loop based on synthetic outcomes, allowing methodology refinement, including the use of explainable AI. The student will develop skills in teamwork, scientific communication, and expertise in programming, machine learning, solid state and computational chemistry techniques.
The supervisory team bridges digital and experimental materials chemistry with expertise in developing techniques for crystal structure prediction and high-throughput screening (Dr George Darling) and digitally-driven inorganic materials discovery, synthesis and characterisation (Prof Matthew Rosseinsky). The team has demonstrated integrated ML/computational chemistry pathways for materials discovery and defined a unique perspective on disorder in crystalline materials that forms the basis of this project by providing a previously unavailable route to calculating the entropy of a crystalline solid.
Dr Darling brings expertise in thermodynamics, crystal structure prediction and machine learning, ensuring robust development and refinement of the workflow and a track record demonstrating the integration of traditional simulation tools with modern AI approaches.
Prof Rosseinsky brings complementary expertise in disorder, the design and application of integrated workflows for materials discovery and feedback loops between theory and experiment. This setup enables iterative improvement of both the computational models and the materials design hypotheses. The student will work in a broader team of experts that will directly testi predictions experimentally and incorporate the new methods into explainable AI-driven discovery workflows.
This project is expected to start in October 2026 and is offered under the EPSRC Centre for Doctoral Training in Digital and Automated Materials Chemistry based in the Materials Innovation Factory at the University of Liverpool, the largest industry-academia colocation in UK physical science. The successful candidate will benefit from training in robotic, digital, chemical and physical thinking, which they will apply in their domain-specific research in materials design, discovery and processing. PhD training has been developed with 35 industrial partners and is designed to generate flexible, employable, enterprising researchers who can communicate across domains.
Further reading
Discovery of Crystalline Inorganic Solids in the Digital Age (https://pubs.acs.org/doi/10.1021/acs.accounts.4c00694)