Hydrodynamic Modeling of Impact Ionization Based on High- Energy Electron Subpopulation

ECS-9710463 Ting-wei Tang 碰撞电离是限制当今先进半导体器件性能的最重要因素之一。 碰撞电离是热电子引起的器件退化的来源，是晶体管雪崩击穿的原因。 碰撞电离的主要问题是碰撞电离系数的建模。 常规地，将其建模为电场或总电子布居的平均能量的函数。 不幸的是，这些模型没有正确地描述碰撞电离的非局部性。 在这项研究中，我们提出了一种创新的方法来模型（作为一个独特的功能，只有那些高能量的平均能量超过碰撞电离阈值能量。 这个平均宏观变量比电场或总电子布居的平均能量更准确地描述了（的行为。 本研究的目的是找到一个系统的流体动力学输运方程，当解决，将提供有关这个宏观变量的信息。 初步研究表明，高能电子亚布居的输运方程组可以从三个简化为一个输运方程。 这使我们能够很容易地解决输运方程，并使用我们新开发的新型碰撞电离模型作为后处理器，任何现有的设备模拟器，没有一个碰撞电离模型。 可以预见，这一研究对半导体器件的计算机辅助设计将产生重大影响。 该模型可以很容易地纳入现有的器件模拟工具，更准确地预测碰撞电离相关的现象发生在晶体管中，包括预测的雪崩击穿电压。 额外的计算成本非常小。

ECS-9710463 Ting -wei Tang Impact ionization is one of the most important factors limiting the performance of today's advanced semiconductor devices. The impact ionization is the source of hot-electron-induced device degradation and is responsible for avalanche breakdown of transistors. The main issue of impact ionization is the modeling of the impact ionization coefficient, (. Conventionally ( is modeled as a function of electric field or the average energy of total electron population. Unfortunately, these models do not properly describe the non-local of impact ionization. In this research, we propose an innovative approach to model ( as a unique function of the average energy of only those high-energy exceed the impact ionization threshold energy. This average macroscopic variable describes the behavior of ( more accurately than the electric field or the average energy of total electron population. The objective of this research is then to find a system of hydrodynamic transport equations, when solved, will provide information about this macroscopic variable. A preliminary investigation indicates that the system of transport equations for the high-energy electron subpopulation can be reduced from three to a single transport equation. This enables us to solve the transport equation easily and use our newly developed novel impact ionization model as a post processor to any existing device simulator which does not have an impact ionization model. A great impact of this research on the computer-aided design of semiconductor devices can be anticipated. The model can be readily incorporated in existing device simulation tools for more accurate prediction of impact ionization related phenomena occurring in the transistors, including the prediction of the avalanche breakdown voltage. The additional computational cost is very minimal.