AI-Powered Microscope Revolutionizes Material Analysis

Scientists in Manchester are building a revolutionary microscope called AutomaTEM. This microscope uses AI and automation to analyze materials much faster than current technology, which will accelerate innovation in fields like quantum computing and clean energy.

Although current TEMs can image atomic-scale structure and chemistry, the typical regions of interest (ROI), or portions of the sample chosen for additional analysis, are very small due to the laborious nature of the technique.

This will be fixed by the AutomaTEM, which will enhance the capacity to locate and analyze, saving time and boosting return on investment. Consequently, it will spur innovation in low-power electronics, new catalysts to support the energy transition, and materials applications for quantum computing, all currently hampered by the constraints of existing technology.

A £9.5 million project, in partnership with manufacturer Thermo Fisher Scientific, is funding the development of AutomaTEM. The project is supported by The University of Manchester, The Henry Royce Institute, bp, and EPSRC. 

To create the AutomaTEM, an instrument that can acquire massive data sets of local chemical information in days rather than years, the Manchester team, led by Professor Sarah Haigh, will combine TEM's current atomic-scale elemental and chemical mapping capabilities with newly developed automation and data analysis technologies. 

Understanding atomic detail at the micrometer or millimeter scale is crucial for developing materials for various applications, from catalysis and quantum technologies to nuclear energy and pharmaceuticals. This system is not simply another TEM instrument. It will provide new opportunities for atomic-scale investigation of materials with less human intervention.

Sarah Haigh, Professor and Director, Electron Microscopy Centre, School of Materials Characterisation, The University of Manchester

Sarah Haigh adds, “For the first time, we will be able to enable atomic resolution analysis of hundreds of regions of interest in a matter of hours, providing unprecedented insights into sparse defects and heterogeneous materials.”

With automated workflows and artificial intelligence at its heart, the AutomaTEM has many cutting-edge features, such as:

• With computer control, precise regions of interest can be addressed by automatically adjusting the sample stage and beam. This allows for diffraction-based analysis and detailed high-resolution imaging without requiring constant operator interaction.
• Using machine learning integration to create functional relationships between experimental results and segment data at lower resolution can improve the identification of new features.
• A cutting-edge Energy Dispersive X-Ray Spectroscopy (EDS) system that offers accurate compositional analysis and remarkable collection efficiency.
• An innovative, high-performance Electron Energy Loss Spectrometer (EELS) designed for complex systems chemical analysis of various species.

It is being developed in partnership with Thermo Fisher Scientific and will be custom-built. It is expected to arrive in the summer of 2025. The multinational lab equipment manufacturer has given Professor Haigh's team access to the required API control and will provide an energy-dispersive X-Ray spectroscopy (EDS) system with a collection efficiency of 4.5 srad, the best in the world.

The state-of-the-art Electron Microscopy Centre (EMC) at The University of Manchester, one of the biggest in the UK, will house the AutomaTEM. Six Transmission Electron Microscopes (TEMs), 13 Scanning Electron Microscopes (SEMs), and six focused ion beam (FIB) instruments are currently available at the EMC.

It welcomes users from institutions worldwide, including Cardiff, Durham, Queen Mary, and Manchester Metropolitan universities, the University of Cape Town (SA), Ceres Power, Nexperia, Nanoco, bp, Johnson Matthey, Oxford Instruments, and UKAEA, and supports over 500 internal users from 12 different UoM departments.

 To promote collaboration and enhance research capabilities, AutomaTEM will be made available to external users for free proof of principle academic projects for a maximum of 30 % of its total usage during the first three years.

 The faster, more accurate analysis capabilities of AutomaTEM represent a significant leap forward in materials science research. With the potential to impact various industries, including aerospace, automotive, and semiconductor, AutomaTEM aims to support the UK’s position at the forefront of materials science innovation.

Dr. Alexander Eggeman, Royal Society University Research, The University of Manchester

Dr. Alexander Eggeman is the leading co-investigator on the project.