3D Monte Carlo simulation of the programming dynamics and their statistical variability in nanoscale charge-trap memories

Abstract

We present a comprehensive investigation of the programming dynamics of nanoscale charge-trap memories, based on 3D Monte Carlo simulations accounting for: 1) true 3D electro-statics during programming and read; 2) atomistic substrate doping; 3) discrete traps, fluctuating in number and position, with localized electron storage; 4) discrete electron injection into traps. The model allows to clarify several key issues affecting the program operation of charge-trap memories, most notably the reduced slope of the ISPP transients exhibited by scaled cells, the programming variability, and the width of the final programmed threshold-voltage distribution. Results are of utmost importance for the assessment of the true programming performance of nanoscale charge-trap memory technologies.