News

A team at Queen Mary University of London has made a major advance in materials science with the publication of "Accelerated discovery of perovskite solid solutions through automated materials synthesis and characterization" in Nature Communications. The study introduces an integrated platform that combines machine-learning screening, robotic synthesis, and high-throughput characterization to rapidly explore and identify new perovskite solid-solution materials.

Among the central contributors from the Centre for Electronics are Dr Mojan Omidvar, Dr Achintha Avin Ihalage, Dr Henry Giddens, and Professor Yang Hao. Dr Omidvar and Dr Ihalage drove the implementation of the automated workflows and machine-learning driven composition screening. Dr Giddens provided expertise in high-frequency characterization and ensured that the discovered materials could meet the exacting demands of electronic and electromagnetic applications. Professor Hao led the project, shaping the vision of a self-driving materials discovery platform and supervising the multidisciplinary collaboration that combined materials chemistry, electronics engineering and robotics.

The novelty of this work lies in its truly integrated pipeline: rather than the traditional manual approach to synthesizing and testing each new composition, the workflow can process synthesis, structure and property measurements in a fraction of the time. For example, the team demonstrates that some solid solutions in the barium/strontium/cerium perovskite family can be generated and screened in minutes rather than hours or days.

Scientifically, this opens a new frontier in materials discovery, particularly for functional perovskites whose dielectric and electromagnetic properties are critical for applications in wireless communication, sensors and high-frequency electronics. Societally, the impact could be significant: faster discovery means quicker deployment of advanced materials for next-generation devices; from 6G wireless components to efficient biosensors and energy-efficient electronics.

In short, this research exemplifies how automation, machine learning and cross-disciplinary engineering can accelerate discovery; not just incrementally but by orders of magnitude. With the Centre for Electronics at the heart of the effort, Queen Mary continues to broaden its influence at the intersection of materials science and electronics.

Read the paper here: www.nature.com/articles/s41467-024-50884-y