A team of MIT researchers has unveiled a groundbreaking AI-powered robotic system that could revolutionize semiconductor analysis and accelerate the development of next-generation solar panels.
The fully autonomous system, detailed in a July 4 publication in Science Advances, measures photoconductance—a critical electrical property that determines how materials respond to light—with unprecedented speed and precision. During a 24-hour test, the system performed more than 3,000 unique measurements, operating at a rate exceeding 125 readings per hour.
"Not every important property of a material can be measured in a contactless way. If you need to make contact with your sample, you want it to be fast and you want to maximize the amount of information that you gain," explains Professor Tonio Buonassisi, senior author of the study.
The innovation combines three critical technologies: a robotic probe that physically contacts semiconductor samples, a self-supervised neural network that identifies optimal measurement points, and a specialized path-planning algorithm that determines the most efficient routes between contact points. By injecting materials science domain knowledge into the AI system, the researchers enabled it to make expert-level decisions about where and how to test samples.
This breakthrough addresses a fundamental bottleneck in materials discovery. While researchers can rapidly synthesize new semiconductor candidates, manually measuring their properties has remained slow and labor-intensive. The MIT system dramatically accelerates this process, enabling faster identification of promising materials for solar cells and other applications.
The detailed measurements revealed performance hotspots and early signs of material degradation that might be missed in conventional testing. Lead author Alexander Siemenn notes, "Being able to gather such rich data that can be captured at such fast rates, without the need for human guidance, starts to open up doors to be able to discover and develop new high-performance semiconductors."
The project, funded by the U.S. Department of Energy, National Science Foundation, First Solar, and other partners, represents a significant step toward MIT's vision of a fully autonomous materials discovery laboratory. The team aims to extend the system's capabilities to create a complete automated lab that combines synthesis, imaging, and measurement—potentially transforming how we discover and develop new materials for clean energy applications.