An efficient large-scale transient electro-thermal field simulator for power devices

With ever-decreasing device size and extensive use of energy-consuming smart devices, heat generated within devices easily leads to extremely high temperature. In return, high temperature influences electrical operational characteristics of the semiconductor devices. Therefore, it is essential for designers to predict accurate temperature and voltage/current distribution and its impact on various devices. For this purpose, coupled electro-thermal (ET) simulation is indispensable. Another concern lies in the number of matrix elements for computation, possibly millions of elements, resulting in days of heavy computation. Therefore, a fast yet accurate modeling framework of overcoming the simulation difficulty is required. In this dissertation, a new transient electro-thermal simulation method for fast 3D chip-level analysis of power devices with field solver accuracy is proposed. The metallization stack and substrate are meshed and solved with 3D field solver using nonlinear temperature-dependent electrical and thermal parameters, and the active transistors are modeled with table models to avoid time-consuming TCAD simulation. Three main contributions are made to enhance physical relevance and computational performance. First, both implicit loose and tight coupling schemes are introduced to compare their computational performances under different coupling degrees. Also, their complexity analysis is presented. Second, the capacitive effects, including interconnect parasitic capacitance and gate capacitance of power devices with nonlinear dependence on bias and temperature, are explicitly accounted for. The inclusion of capacitive effects allows accurate modeling of devices with large numbers of transistor fingers and high frequency application. Third, a specialized nonlinear exponential integrator (EI) method is developed to address the considerably different time scales between electrical and thermal sectors. The EI-based transient solver allows the electrical system to step with much larger time step size than in conventional methods, thus the time step size gap between the electrical and the thermal simulation is largely reduced. Its benefits of scalability, adaptivity and accuracy are also demonstrated in the dissertation.published_or_final_versionElectrical and Electronic EngineeringMasterMaster of Philosoph