Improvement of TCAD Augmented Machine Learning Using Autoencoder for Semiconductor Variation Identification and Inverse Design

Kashyap Mehta,Yuhao Zhang,Hiu Yung Wong,Ming Xiao,Sophia Susan Raju,Boyan Wang

doi:10.1109/access.2020.3014470

Kashyap Mehta, Yuhao Zhang + Show 4 more

Open Access

https://doi.org/10.1109/access.2020.3014470

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2020
Citations: 37	License type: CC BY 4.0

Affiliation: San Jose State University, Virginia Tech

Abstract

A machine learning (ML) model by combing two autoencoders and one linear regression model is proposed to avoid overfitting and to improve the accuracy of Technology Computer-Aided Design (TCAD)-augmented ML for semiconductor structural variation identification and inverse design, without using domain expertise. TCAD-augmented ML utilizes TCAD simulations to generate sufficient data for ML model development when experimental data are inadequate. The ML model can then be used to identify semiconductor structural variation for given experimental electrical measurements. In this study, the variation of layer thicknesses in the p-i-n diode is used as a demonstration. An ML model is developed to predict the diode layer thicknesses based on a given Current-Voltage (IV) curve. Although the variations of interest can be incorporated easily in TCAD simulations to generate ML training data, the TCAD-augmented ML model generally is overfitted and cannot predict the variations in experiment well due to hidden variables which also alters the IV curves. We show that by using an autoencoder, this problem can be solved. To verify the effectiveness, another set of TCAD simulation data is generated with hidden variables (dopant concentration variation) to emulate experimental data. Testing on the second set of data shows that the proposed model can avoid overfitting and has up to 15 times improvement in accuracy in thickness prediction. Moreover, this model is used successfully to perform inverse design and can capture an underlying physics that cannot be described by a simple physical parameter.

Highlights

Machine learning (ML) has been widely used in semiconductor fabrication to improve manufacturing yields at various manufacturing stages [1]-[5]
The result can be confirmed with physical failure analysis such as Transmission Electron Microscopy (TEM), which is very expensive in terms of cost and turn-around time
In [8], we showed that the Technology Computer-Aided Design (TCAD) generated ML model cannot be used on experimental data directly because experimental data contains more variations than in the TCAD data used to train the ML model

Summary

INTRODUCTION

Machine learning (ML) has been widely used in semiconductor fabrication to improve manufacturing yields at various manufacturing stages [1]-[5]. To the best of our knowledge, this ideal strategy has not been successfully deployed as can be seen in the lack of relevant publications We believe this is because of the lack of data to train a successful ML model. We propose a new model based on autoencoder and linear regression to make TCAD trained ML model more robust when it is presented with data with hidden variations. TCAD dataset is used for validation because the additional variations can be more precisely controlled and many more data can be generated compared to experiments. This new model requires no domain expertise.

TCAD SIMULATION AND DATASETS DISCUSSION

Data Preparation

Linear Regression and Overfitting

Autoencoder for Variation Identification

Autoencoder for Reverse Engineering

Findings

CONCLUSION