I-Corps: Translation Potential of Chemical Process Simulation for Advanced Semiconductor Device Manufacturing

Award ID: 2534684

This I-Corps project investigates the commercial potential of a software package and database system to simulate and predict chemical reactions during semiconductor manufacturing. In current semiconductor manufacturing, forming nanoscale patterns requires hundreds of consecutive, precisely controlled chemical reactions to deposit and etch thin films. As devices shrink and evolve in structure, developing new processes is time intensive and costly. Process optimization often proceeds through trial and error, consuming chemicals and generating waste from scrapped material. A rapid, accurate simulation procedure for these reactions would save considerable time and money and accelerate the development of new products. Providing engineers with real-time simulation data will alleviate many of the most restrictive pain points of the process development burden, providing a significant economic benefit. Furthermore, the understanding gained from accurately programming these chemical mechanisms will generate new insights into nanoscale reaction engineering, promoting the progress of science. Microchips manufacturing is crucial for civilian and military technologies, so the advantage provided by computer simulations will have uses in both commercial and national defense applications. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of computational tools to integrate cutting-edge, vapor-phase, thin film deposition techniques into manufacturing, such as area-selective atomic layer deposition and chemical vapor deposition. Traditionally, patterned features are formed using a repeated sequence of uniform thin film deposition, image generation by photolithography, thin film etching, surface cleaning, and chemical planarization. Lithographic patterning has been a staple of semiconductor manufacturing for decades, but the process is energy intensive, costly, and generates large amounts of chemical waste. Recently, area-selective deposition is being explored to augment or replace some lithographic steps. The approach uses vapor-phase reactants to directly grow patterned thin films, where the pattern is generated by controlled surface chemical reactions. Achieving the necessary control during these reactions is an ongoing challenge. This software tool should allow engineers to accelerate integration of vapor-phase patterning procedures into manufacturing, saving time and reducing waste. This software uses an atomistic algorithm informed by the mechanistic effects of surface chemistry to rigorously simulate area-selective deposition, differentiating it from existing chemical process simulations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman