Multi-Modal AI Models for Designer Ferroelectric Devices (PhD Studentship, Materials 4.0 CDT) The University of Manchester in United Kingdom

This fully-funded PhD studentship at The University of Manchester is part of cohort 3 of the EPSRC Centre for Doctoral Training (CDT) in Developing National Capability for Materials 4.0, in partnership with the Henry Royce Institute. The project focuses on the development of multi-modal AI models for the design and realisation of advanced ferroelectric devices, leveraging cutting-edge computational and machine learning techniques to explore the rich physics of interfacial ferroelectrics.

Interfacial ferroelectrics are a novel class of materials formed by stacking two-dimensional van der Waals crystals, creating symmetry-breaking interfaces that enable switchable polarisation without conventional lattice distortions. These heterostructures are highly resilient to charge trapping, mechanical strain, and long-term degradation, and their multi-level, dynamically switchable polarisation states open new possibilities for field-tuneable devices, including reconfigurable optoelectronics, ferroelectric tunnel junctions (fTJs) as analogue synapses, and two-dimensional multiferroics.

The project will investigate multilayer functional devices built from stacked monolayer semiconductors and magnetic materials. High-throughput ab-initio Density Functional Theory (DFT) will be used to determine layer-resolved properties such as band edges, dipoles, and magnetic moments. Machine learning interatomic potentials (MLIPs), including Atomic Cluster Expansion (ACE) models, will efficiently scan the chemistry, stacking, and twist space to identify structures with strong ferroelectric responses. The student will develop multi-modal, charge-aware MLIPs trained on the DFT corpus to predict energies, forces, and local charge densities, quantifying the impact of defects and impurities on collective polarisation and switching pathways. These models have broader applications in nano-catalysis and orbital-free DFT.

As a PhD researcher, you will join a vibrant community within the Materials 4.0 CDT and the National Graphene Institute, participating in seminars, reading groups, and hands-on training in modern theory of polarisation, DFT/ASE with QE or VASP, workflow automation, and MLIP development with ACE and MACE. You will gain expertise in vdW-inclusive DFT/MD, NEB for switching barriers, LAMMPS/MACE simulations, HPC best practices, and FAIR-data management.

The studentship covers full tuition fees (home & international), a tax-free stipend of at least £20,780 plus London allowance if applicable, and a research training support grant. All nationalities are welcome to apply, with up to 30% of studentships available for outstanding international candidates. The CDT is committed to Equality, Diversity, and Inclusion, and encourages applications from underrepresented groups.

Applicants should have a strong academic background in Physics, Materials Science, Engineering, Computer Science, or a related field, with experience in computational modeling, DFT, or machine learning highly desirable. English language proficiency is required. The application deadline is March 3, 2026.

For general enquiries, contact [email protected]. For technical questions, reach out to Dr. James McHugh ([email protected]). Apply online via the University of Manchester portal, selecting Postgraduate Research, the 2026/27 academic year, and CDT in Materials 4.0.