(224d) Optimal Operation of Plasma Enhanced Atomic Layer Deposition Via Machine Learning

Over the past decade, plasma enhanced atomic layer deposition (PEALD) has been extensively studied in experimental labs and gradually become a popular deposition method in the semiconductor industry. Its distinguished ability to deposit ultra-thin high-k dielectric films makes it the perfect candidate in the production of contemporary high aspect ratio designs like fin field-effect transistor (FinFET) [1]. With the introduced high-energy plasma species, PEALD ideally requires much less thermal energy input, while enabling high throughput with temperature-sensitive precursors. Despite the advantages of PEALD, the complicated combination of remote plasma chamber, the showerhead reactor geometry and the plasma induced surface reaction makes the optimal operation hard to locate [2]. In addition, the process operation still remains costly, which discourages the extensive experimental exploration [3].

In light of this, a comprehensive multiscale computational fluid dynamics (CFD) model has been proposed to capture the integrated dynamic profile of a remote plasma PEALD process with Tetrakis-dimethylamino-Hafnium (TDMAHf) and oxygen plasma as precursors. Based on the simulation model, in this work, the operation of a PEALD reactor under different operating conditions is discussed, and the optimal operating regime of this reactor is analyzed. Specifically, an operational database is constructed using the multiscale CFD model to map a variety of reactor operation input combinations to their resulting film qualities and deposition profile. This database is then processed through machine learning analysis to explore the feasible operating domain, and within which, the optimal operating decision can be identified according to the desired production throughput and economic demand.

[1] Ishikawa, K., Karahashi, K., Ichiki, T., Chang, J.P., George, S.M., Kessels, W., Lee, H.J., Tinck,S., Um, J.H., Kinoshita, K., 2017. Progress and prospects in nanoscale dry processes: How can we control atomic layer reactions? Japanese Journal of Applied Physics 56, 06HA02.

[2] Profijt, H., Potts, S., Van de Sanden, M., Kessels, W., 2011. Plasma-assisted atomic layer deposition:basics, opportunities, and challenges. Journal of Vacuum Science & Technology A: Vacuum,Surfaces, and Films 29, 050801.

[3] Joo, J., Rossnagel, S.M., 2009. Plasma modeling of a PEALD system for the deposition of TiO2and HfO2. Journal of Korean Physical Society 54, 1048.