Nonlinear inverse problems in semiconductor manufacturing

Industrial Problems Seminar

Christopher A. Wong

Abstract:

Ongoing advances in computer technology, such as the AI boom of recent years, have been made possible by the incredible developments in semiconductor manufacturing at the nanometer scale. With each subsequent generation, improvements to transistor density and power efficiency allow to us perform more expensive computational tasks. However, at the nanoscale, manufacturing integrated circuits reliably becomes very challenging and it becomes increasingly necessary to tune the process by solving nonlinear inverse problems to determine optimal parameters for a given desired circuit pattern. In this talk, I will provide a broad overview of the engineering technologies involved in semiconductor manufacturing. Then, I'll focus the attention on inverse problems, especially the photolithography inverse problem, and discuss some of their mathematical and computational aspects. Lastly, I will describe some of my personal experiences in both academia and industry, and my transition from one to the other. By the conclusion of this talk, the audience will hopefully gain some appreciation for the immense amount of effort and multidisciplinary collaboration across a wide variety of scientific, engineering, and mathematical fields that has enabled our modern-day information age.