Semiconductor materials contain numerous ingredients in various mixing ratios, and high-performance materials are obtained by optimizing the formulation. However, more than 1,050 theoretical combinations of ingredients and mixing ratios need to be analyzed, so it would take more than dozens of years to explore all possible combinations of these ingredients and their mixing ratios with conventional artificial intelligence (AI) methods, according to Showa Denko K.K. (Tokyo, Japan; www.sdk.co.jp).
Optimization of semiconductor materials formulation (Source: Showa Denko)[/caption]
To reduce the time required for the exploration, the company used high-performance computing technology, Digital Annealer, a domain-specific computer architecture developed by Fujitsu Ltd. (Kawasaki, Japan: www.fujitsu.com) that is inspired by quantum technology (but not directly using quantum effects). Showa Denko developed an AI model for predicting the properties of semiconductor materials. To make the AI model computable on Digital Annealer, Showa Denko expressed the AI model as an Ising model, a statistical mechanical method. By simulating the Ising model on Digital Annealer, the company says it has reduced the exploration time to dozens of seconds,…
Chemical Engineering publishes FREE eletters that bring our original content to our readers
in an easily accessible email format about once a week.
Subscribe Now