PhD in Electronics & Nanoscale Engineering - Machine Learning-Enhanced Atomistic Simulations of Electrochemistry in Memristors

About the Project

Start date: From October 2026, flexible

Memristors, nanoscale structures with switchable resistance, are an emerging class of electronic devices which promise to enable next-generation computing. By co-locating memory and compute in the same device, memristors open the door to advanced paradigms like neuromorphic computing: Computers that mimic the human brain at the hardware level. This paves the way for reducing the extreme energy demands of contemporary AI/ML workloads, since the human brain is far more energy efficient than conventional CMOS-based hardware like GPUs.

However, memristors are still not ready for large-scale industrial applications due to insufficient performance and reliability characteristics. Although significant advances have been made in the experimental optimization of materials, the fundamental question of the precise mechanism leading to the resistive switching behaviour has not been addressed for a wide gamut of material stacks. This can only be achieved looking at these devices through the atomistic-scale lens.

In this 3.5 year PhD project, you will develop a machine learning- (ML-)enhanced framework to simulate memristor resistive switching at the atomistic level. You will perform simulations using established atomistic techniques (e.g., DFT, MD, NEGF) and develop a novel interatomic ML approach able to handle a crucial aspect of several memristor devices: Electrochemical/redox reactions. With these tools, you will aim to answer:

• What are the dominant vacancy/ion migration mechanisms enabling resistive switching?
• How accurate are ML potentials in reproducing DFT fidelity for large, electrochemically active systems?
• How do microstructure, stoichiometry, and temperature drive device-to-device and cycle-to cycle variability?
• Can we extract compact models which capture the long-term endurance characteristics of these devices?

Application details & further information

We are looking for a motivated student with considerable interest in computational nanoelectronics and existing skills in scientific computing. The ideal candidate will have:

• A first class or upper second class degree in Electronic Engineering, Physics, Chemistry, Computational Science and Engineering, Materials Science, or a related discipline.
• A relevant Master’s degree is desirable.
• Strong programming skills (e.g. Linux development, Python, C/C++).
• Experience with or interest in atomistic methods (DFT/MD).

We are committed to fostering and promoting an inclusive, supportive, and flexible working environment in all our activities. We particularly welcome applications from candidates from groups which have been historically under-represented in STEM subjects/research.

For more information, see the DeepNano Group and the Glasgow Computational Engineering Centre’s (GCEC) websites.

How to Apply:

To apply, please contact Luiz.Aguinsky@glasgow.ac.uk with:

• A short motivation statement outlining your interest and suitability.
• Your CV.
• A link to a technical writing sample in English of which you were the main author (ideally a published or preprint paper).

Funding Notes

The studentship funds tuition fees and provides a stipend at the UKRI rate (currently at £20,780 for the 2025-26 academic year) for 3.5 years.

Funding is prioritized for UK home students (including Republic of Ireland). Exceptional international students (i.e., having first class-equivalent degrees from highly ranked universities and published works) are encouraged to apply.