Prediction and Control of Transistor Threshold Voltage by Virtual Metrology (Virtual PCM) Using Equipment Data

This paper is a description of how to predict and control the transistor threshold voltage ( Vth) for an advanced system on chip (SoC) by virtual metrology (VM), which we call virtual process control module (PCM), by using equipment data. Impact analysis for Vth variation by using a Virtual PCM model indicates that the impact of source and drain (S/D) resistance and extension resistance are comparable with that of the shape factor (i.e., gate length, sidewall width). Virtual metrology models for the resistance were developed to control and predict Vth. As a result of the VM control, Vth variation was reduced by 28%. Moreover, in-process prediction of Vth was put into practical use.