At the University of Nottingham (UK), researchers have developed an automated laboratory that combines robotics and artificial intelligence. The system measures and mixes chemical components itself, processes samples in a high-temperature furnace and retrieves them. The platform was created at the Centre of Excellence in Coatings and Surface Engineering (CE-CSE).
Developing materials for extreme conditions usually requires many repetitive experiments: researchers manually prepare samples, process them and test their properties. This is slow and labor-intensive. The Nottingham team combined these stages into a single automated process with two robotic arms, a special liquid dispenser, laboratory equipment and its own software.
A key role is played by an AI-based approach called "active learning." It uses data from previous experiments to suggest which compositions should be investigated next. Using this method, the scientists identified and prepared several candidates with a special crystal structure - pyrochlore. They are being considered as potential materials for thermal barrier coatings.
Thermal barrier coatings work as heat-resistant screens: they reduce the heating of parts beneath them. This is important for aircraft engines, where components are exposed to very high temperatures. Finding materials with low thermal conductivity - that is, those that conduct heat poorly - is an important step toward creating more efficient coatings.
Although robots increasingly automate routine laboratory operations, combining chemical preparation, sample handling and high-temperature processing into one reliable cycle is technically difficult. The researchers solved this task by combining different stages in a single system. Calibration experiments showed precise dispensing of different liquids. The laboratory then automatically prepared a precursor solution (the starting substance for subsequent processing), transferred the material to the furnace, ensured controlled cooling and retrieved the finished sample.
"Self-driving laboratories are capable of transforming the development of advanced materials. The key problem is getting laboratory robots to work reliably at different stages of an experiment, especially when liquid handling must be combined with high-temperature processing. Our work solves an important part of this task," the university's press service quoted Professor Tanvir Hussain as saying.
"We wanted to link laboratory operations that are often performed separately into one coordinated robotic process. By combining liquid dispensing, robotic handling and high-temperature processing, we showed how complex materials preparation can be automated from start to finish," added Dr. Mohammadreza Laleghani Dezaki.
The work demonstrates a practical way to combine robotics, laboratory equipment and AI-driven materials discovery. By reducing the need for manual intervention in repetitive operations, the platform can improve the stability of experiments and the reproducibility of research. The results were published in the journal Engineering Applications of Artificial Intelligence and in a working paper on ChemRxiv.