New types of electronics and clean energy technologies are emerging every day, from flexible phone screens to bendable wind turbine blades. Some of the biggest advances occur at a scale too small for the average person to notice. Scientists consistently work to create new nanomaterials with unique properties that make the larger technologies possible.

Developing nanomaterials is a painstaking process. Scientists have to hand-make each material before testing its capabilities. Many of the new materials are built from stacking layers of 2D materials that are only a few atoms thick. The delicate nature makes it difficult to automate the process. But a team including researchers from the University of Chicago and the University of Michigan developed a robot to speed up nanomaterial manufacturing. 

The innovation

“This process is fully automated — you can program it and walk away,” Andrew Mannix, a former University of Chicago postdoctoral fellow who is now an assistant professor at Stanford University, said in a news release. “Previously, if you wanted to try 10 different permutations of materials, it would all be done by hand, which is weeks of labor. We can now do this in an hour.”

The research team had to figure out how to precisely cut the sheets of material into exact shapes without damaging them. They also had to make a robot “hand” that could gently pick up and handle the fragile material sheets.

They created the hand from soft polymers that fall apart if exposed to heat or ultraviolet light. Once the device correctly positions a material sheet, the hand dissolves and the sheet falls into place. The robots, when strung together, create an assembly line.

The implications

The system can be customized for different applications. For instance, each material sheet can be rotated to different angles. The assembly line can be programmed to create a material with dozens of different layers.

Scientists can leave the automated system to build the material on its own, then return minutes later to test the finished product.

“When we started looking at this problem, it seemed unimaginable to automate it,” said UChicago professor Jiwoong Park. “This should speed up the pace of discovery considerably. It’s kind of like the difference between handwriting a book letter by letter versus using a printing press.”