Artificial carrier heating due to the introduction of ab initio Coulomb scattering in Monte Carlo simulations

Accurately treating ionised impurity scattering in a way suitable to describe the influence of random dopant fluctuations on device characteristics is important in next generation MOSFETs. Statistical variations are unobservable using a continuous treatment of the doping, requiring a discrete representation of impurities. In particle-based simulations the P³M method, which resolves the Coulomb interaction into long- and short-range components, is in principle capable of describing Coulomb scattering through propagation in this accurately resolved potential. However, numerically the integration of the equations of motion is inaccurate and controlling the errors in practical simulations is vital. In this paper we investigate the effect of the choice of short-range correction strategy and integration time step on accuracy in a 3D self-consistent ensemble Monte Carlo simulations featuring random discrete dopants. We illustrate the importance of the ‘ab initio’ Coulomb scattering comparing the effect of a single trapped charge in drift-diffusion and Monte Carlo simulations.