ADVANCED MATERIALS & PROCESSES | MARCH 2024 10 MACHINE LEARNING | AI AI SHEDS LIGHT ON CRYSTAL GROWTH Researchers at Nagoya University, Japan, recently used artificial intelligence to analyze the image data of polycrystalline silicon, a material widely used in solar panels. The AI created a 3D model in virtual space, helping the team identify areas where dislocation clusters were affecting the material’s performance. The scientists then used electron microscopy and theoretical calculations to understand how these areas formed. These techniques revealed stress distribution in the crystal lattice and found staircase-like structures at crystal grain boundaries that appear to cause dislocations during crystal growth. In addition to understanding how to improve solar cells, the study could have important implications for the science of crystal growth and deformation. For starters, the new research could add to the Haasen-Alexander- Sumino (HAS) model—a theoretical framework used to understand the behavior of dislocations in materials. Researcher Noritaka Usami believes his team found dislocations that the HAS model missed. Another discovery occurred when the team calculated the arrangement of atoms in the polycrystalline structures. They found unexpectedly large tensile bond strains along the edge of the staircase-like struc- tures that triggered dislocation generation. “As experts who have been studying this for years, we were amazed and excited to finally see proof of the presence of dislocations in these structures. It suggests that we can control the formation of dislocation clusters by controlling the direction in which the boundary spreads,” says Usami. www.nagoya-u.ac.jp. SPEEDY SOLAR CELL PRODUCTION Scientists at RMIT University along with colleagues at Monash University and CSIRO, all in Australia, are using AI to produce solar cells from perovskite in a matter of weeks, saving years of human labor and potential errors. Lead researcher Nastaran Meftahi said scientific teams worldwide have been racing to make perovskite cells, which are less expensive than silicon and now stable enough for longterm commercial use. “Until now, the process of creating perovskite cells has been more like alchemy than science—record efficiencies have been reached, but positive results are notoriously difficult to reproduce,” 3D model generated by AI helps scientists explore complex polycrystalline materials. she said. “What we have achieved is the development of a method for rapidly and reproducibly making and testing new solar cells, where each generation learns from and improves upon the previous.” Using a multimillion-dollar automated system for solar cell manufacturing being built by Adam Surmiak at Monash University, the model will be able to predict huge volumes of promising chemical recipes for new perovskite solar cells. The facility is currently under construction. So far, the scientists’ work has resulted in reproducible perovskite solar cells with power-conversion efficiencies of 16.9%, the best-known efficiency manufactured without human intervention. Surmiak’s team designed and characterized 16 new solar cells never seen before using his innovative setup, and Meftahi employed these cells to predict the properties of 256 new solar cell recipes. “At Monash, they’ll soon be able to make 2000 unique solar cells per day. We’re quickly getting to the stage where we’ll be able to predict the properties of millions of different cells,” says Meftahi. www.rmit.edu.au. Adam Surmiak working with automatic device characterization equipment. Courtesy of Australian Research Council.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==