TABLE OF CONTENTS Heterostructure bipolar transistors and high electron mobility transistors (HEMTs) are among the most advanced semiconductor devices. In this section, the simulation of SiGe, III-V GaAs and InP based heterostructure semiconductor devices is considered. A common feature is the lack of a rigorous approach to III-V and IV-IV group semiconductor materials modelling. As an example, modelling of AlGaAs, InGaAs, or even InAlAs and InGaP is restricted to slight modifications of the GaAs material properties. Critical issues concerning simulation of heterostructures, such as interface modelling at heterojunctions and insulator surfaces as well as hydrodynamic and high field effects modelling, carrier energy relaxation, impact ionisation, gate current modelling and self-heating effects are mostly not considered. In addition, quantum mechanical effects are often neglected or accounted for only by simple models for quantum corrections [23, 24], as solving the Schrodinger or the Wigner equation is extremely expensive in terms of computational resources.
To enable predictive simulation of semiconductor devices better carrier transport models are required. The drift-diffusion (DD) transport model is the simplest and most popular model used for device simulation [25]. However, with downscaling the feature sizes, non-local effects become more pronounced and may be accounted for by using an energy-transport (ET) or hydrodynamic (HD) transport model.
During the past two decades Monte Carlo (MC) methods for solving the time-dependent Boltzmann equation have been developed [26, 27] and applied for device simulation [28 30]. Reduction of the demand on computational resources is still an issue for MC simulation. Also, the MC...
