Self Consistent Field method with damped oscillation for nano device

As CMOS technology progresses, device dimensions have been scaled down into the nanometer regime and classical physics failing to characterize the nanoscale devices as quantum mechanics start to play an important role in their characteristics. Modeling Quantum transport is being very significant for rigorous performance analysis of nano devices and simulation plays an important role in the field of nano device modeling. The most widely accepted method for simulation of nano devices is Self Consistent Field (SCF) method which uses Non-Equilibrium Green's Function (NEGF) formalism self consistently with Poisson solver equation. The NEGF formalism provides a rigorous description of quantum transport in nanoscale devices but heavy in computation. Now to reduce the computational burden a supervised algorithm in the updating procedure of SCF method is proposed in this paper. It reduces the computational burden drastically and to show the superiority a comparison in simulating DGMOSFET by traditional SCF and our proposed algorithm is provided.

I. Introduction

Rapid device scaling pushes the dimensions of the field effect transistors to the nanometer regime [1]. Quantum effects play an important role in determining the device characteristics in this regime of operation. The main challenges arise with this advancement in feature size is Modeling of the complex physical phenomena which can't be rigorously explained by classical physics analysis based models where Non Equilibrium Green Function(NEGF) is required to rigorously describe the quantum phenomena that control the device behavior in the so extremely small dimensions [2]. To offer an additional insight into transport phenomena in these deeply scaled devices, simulation tools that consider quantum effect is highly appreciated [3]. In the field of Quantum Simulation Self Consistent Field (SCF) method is the most widely accepted method. It is an iterative process which solves the NEGF and Poisson's equation self consistently. As NEGF formalism is heavy in computation, simulation time of performance analysis has become an important factor [4]. We have provided a supervised algorithm in the updating procedure of SCF method to reduce iteration which in turn reduces the simulation time and we have considered Computational efficiency which is needed to make the self-consistent method suitable for device design and characteristic prediction. Our proposed supervised algorithm doesn't hamper the computational efficiency it just reduces the no. of iteration. For establishing the algorithm we have analyzed Double Gate (DG) MOSFET as it has emerged as promising devices for Very Large Scale Integration (VLSI) circuits due to their better scalability compared to bulk CMOS [5]. DG-MOSFETs characteristics are greatly affected by the quantum effects. These effects can be accurately predicted only using quantum mechanical based device simulation and the characteristics found by proposed algorithm shows exactly the same characteristics found by traditional SCF which proves the acceptance of the proposed method.