Researchers at the University of Nottingham have developed a self-driving laboratory combining robotics and artificial intelligence (AI) to automate the preparation and processing of materials, potentially accelerating the discovery of coatings for extreme operating environments.

The system, developed at the university's Centre of Excellence in Coating and Surface Engineering (CE-CSE), integrates robotic handling, chemical dispensing and high-temperature furnace processing into a coordinated workflow. It is designed to support the discovery and optimisation of advanced materials for applications including aerospace, energy and high-temperature engineering.

Two robotic arms work alongside a specially designed liquid-dispensing tool, laboratory equipment and purpose-built software to carry out tasks that would traditionally require several stages of manual intervention.

The laboratory also uses the AI active learning technique, which draws on the results of previous experiments to identify promising material compositions for further investigation. Using this approach, the researchers identified and prepared candidate materials with a crystal structure known as pyrochlore, which is being investigated for potential use in thermal barrier coatings.

The coatings act as heat shields, protecting underlying components from extreme temperatures. They are particularly important in aerospace applications, where engine parts operate under demanding thermal conditions. Materials with low thermal conductivity, which transfer heat less readily, are therefore of considerable interest.

Integrating the different stages of materials preparation presented a significant technical challenge. The team demonstrated accurate liquid dispensing using different liquids before automating the preparation of a precursor solution, its transfer for furnace processing, controlled cooling and retrieval of the processed sample.

Professor Tanvir Hussain, professor of coatings and surface engineering at Nottingham, said a key challenge for self-driving laboratories was enabling robotic systems to operate reliably across different experimental stages, particularly when liquid handling was combined with high-temperature processing.

Dr Mohammadreza Lalegani Dezaki, research fellow in mechanical and manufacturing engineering, said the aim was to connect operations normally performed separately into a single robotic workflow.

The researchers say the platform provides a foundation for more autonomous and reproducible materials research, with the potential to reduce manual intervention and support higher-throughput experimentation.

The research is published in Engineering Applications of Artificial Intelligence.