Intelligent Modeling of Electrical Characteristics of Multi-Channel Gate All Around Silicon Nanosheet MOSFETs Induced by Work Function Fluctuation

Intelligent model (IM) of device and performance prediction is an emerging methodology in the IC industry. This paper for the first time presents an optimized intelligent model approach based on an artificial neural network (ANN) to study the effects of a crucial source of variation, i.e., work function fluctuation (WKF) of multi-channel (MC) gate-all-around (GAA) silicon (Si) nanosheet (NS) MOSFET. Various fluctuated devices are simulated and their ID-VG curves, as well as their extracted electrical properties, are investigated using IM based ANN algorithm. These predicted electrical properties are estimated accurately and show that on-current (ION) is decreased from top contacts to the bottom source/drain contact as the electrostatic potential (V) is decreasing in MC GAA Si NS MOSFET. IM-based ANN methodology has negligible computational cost as compared to 3D device simulation. Therefore, this intelligent modeling approach can be utilized to accelerate the device simulation for advanced semiconductor nano-devices.