Simulation and machine learning based analytical study of single electron transistor (SET)

In recent years, the requirement for greater scalability of transistor technology for the continuation of Moore’s law has led researchers toward the investigations of several innovative advanced semiconductor device as potentially superior alternatives to conventional transistors. Among them, single-electron transistors (SETs) have shown considerable promise in terms of performance and reliability with significant device dimension scaling. However, realistic modeling and simulation are the primary steps toward the practical implementation of SET designs. In this work, a technology computer-aided design simulation-based analytical study of silicon quantum dot SETs is developed to improve the electrical characteristics of the devices through optimization of different device parameters. Further, the investigation is extended to explore the temperature dependency of quantum tunneling by analysis of the characteristic plots of such quantum dot-based nano-devices. Moreover, a machine learning (ML)-based approach has been implemented and validated through development and testing of ML models predicting SET device performance by examining dependence of relevant design parameters on device performance. Hence, the proposed model of SETs provides the analytical understanding for a sustainable and realistic design of SETs allowing approaches to future nano-device-based IC design.