[paper] Teaching Traditional TCAD New Tricks

Sanghoon Myung¹, Wonik Jang¹, Seonghoon Jin²

Myung Choe¹, Changwook Jeong¹, and Dae Sin Kim¹

Restructuring TCAD System:

Teaching Traditional TCAD New Tricks

DOI: 10.1109/IEDM19574.2021.9720616

1Data and Information Technology Center, Samsung Electronics.
2Device Lab, Samsung Semiconductor Inc.

Abstract : Traditional TCAD simulation has succeeded in predicting and optimizing the device performance; however, it still faces a massive challenge - a high computational cost. There have been many attempts to replace TCAD with deep learning, but it has not yet been completely replaced. This paper presents a novel algorithm restructuring the traditional TCAD system. The proposed algorithm predicts three-dimensional (3D) TCAD simulation in real-time while capturing a variance, enables deep learning and TCAD to complement each other, and fully resolves convergence errors.

Fig: (a) A TCAD process simulation result. (b) A prediction result of RTT process model.

(c) 1D doping concentration plot in the horizontal direction below the gate.

(d) 1D doping concentration plot in the vertical direction at the center of drain.