An advanced model for self-heating effects in power semiconductor devices is derived from principles of irreversible thermodynamics. The importance of the entropy balance equation is emphasized. The governing equations for the coupled transport of charge carriers and heat are valid in both the stationary and transient regimes. Four characteristic effects contributing to the heat generation can be identified: Joule heating, recombination heating, Thomson heating and carrier source heating. Bandgap narrowing effects are included. Hot carrier effects are neglected. Numerical methods to solve the governing equations for the coupled transport of charge carriers and heat are described. Finally, results obtained in simulating latch-up in an IGT are discussed.

For tracking electromigration induced evolution of voids a diffuse interface model is applied. We assume an interconnect as two-dimensional electrically conducting via which contains initially a circular void. The diffuse interface governing equation was solved applying a finite element scheme with a robust local grid adaptation algorithm. Simulations were carried out for voids exposed to high current. An influence of the void dynamics on the resistance of interconnect is investigated. In the case of the interconnect via it was shown that a migrating void exactly follows the current flow, retaining its stability, but due to change of shape and position causes significant fluctuations in interconnect resistance.

After a brief discussion of the demands in meshing for semiconductor process and device simulation, we present a three-dimensional Delaunay refinement technique combined with a modified advancing front algorithm.

We present a simulation system which meets the requirements for practical application of inverse modeling in a professional environment. A tool interface for the integration of arbitrary simulation tools at the user level is introduced and a methodology for the formation of simulation networks is described. A Levenberg-Marquardt optimizer automates the inverse modeling procedure. Strategies for the efficient execution of simulation tools are discussed. An example demonstrates the extraction of doping profile information on the basis of electrical measurements.

An approach for analysis of the small signal response of the carriers in semiconductors is presented. The integro-differential equation, describing the phenomenon in the time domain is transformed into a Fredholm integral equation of the second kind. The response of the carrier system to a small signal of a general time dependence can be calculated by the knowledge of the response to an impulse signal, defined by a delta function in time. For an impulse signal, the obtained integral equation resembles the basic structure of the integral form of the time dependent (evolution) Boltzmann equation. Due to this similarity a physical model of the impulse response process is developed. The model explains the response to an impulse signal in terms of a relaxation process of two carrier ensembles, governed by a Boltzmann equation. A Monte-Carlo method is developed which consists of algorithms for modeling the initial distribution of the two ensembles. The numerical Monte-Carlo theory for evaluation of integrals is applied. The subsequent relaxation process can be simulated by the standard algorithms for solving the Boltzmann equation. The presented simulation results for Si and GaAs electrons serve as a test of the Monte-Carlo method and demonstrate that the physical model can be used for explanation of the small signal response process.

Monte Carlo simulation of the low field electron mobility of strained Si and SiGe active layers on Si and SiGe substrates is considered. The Ge mole fractions of both the active layer and the substrate are varied in a wide range. The linear deformation potential theory is used to calculate the shifts of the conduction band minima due to uniaxial strain along [001]. The energy shifts and the effective masses are assumed to be functions of the Ge mole fraction. It is shown that in spite of the fact that the L-valleys remain degenerate under strain conditions considered here, they play an important role at very high Ge compositions especially when SiGe as substrate is used. We found that in this case the repopulation effects of the X-valleys affect electron mobility much stronger than the alloy scattering. We also generalize the ionized impurity scattering rate to include strain effects for doped materials and show that some of the important parameters such as effective density of states, inverse screening length, and the screening function are split due to strain and must be properly modified. Finally, we perform several simulations for undoped and doped materials using Si and SiGe substrates.

The high accuracy which is necessary for modern process simulation often requires the use of Monte-Carlo ion implantation simulation methods with the disadvantage of very long simulation times especially for three-dimensional applications. In this work a new method for an accurate and CPU time efficient three-dimensional simulation of ion implantation is suggested. The approach is based on a combination of the algorithmic capabilities of a fast analytical and the Monte-Carlo simulation method.

A new method for simulation of etching and deposition processes has been developed. This method is based on fundamental morphological operations derived from image and signal processing. As the material surface during simulation moves in time, the geometry either increases or decreases. If the simulation geometry is considered as a two-valued image (material or vacuum), etching and deposition processes can be simulated by means of the erosion and dilation operation. Together with a cellular material representation this method allows an accurate and stable simulation of three-dimensional arbitrary structures. Simulation results for several etching and deposition problems demonstrate accuracy and generality of our method.

A fully three-dimensional simulation tool for modeling the ion implantation in arbitrarily complex three-dimensional structures is described. The calculation is based on the Monte Carlo (MC) method. For MC simulations of realistic three-dimensional structures the key problem is the CPU-time consumption which is primarily caused by two facts. (1) A large number of ion trajectories (about 107) has to be simulated to get results with reasonable low statistical noise. (2) The point location problem is very complex in the three-dimensional space. Solutions for these problems are given in this paper. To reduce the CPU-time for calculating the numerous ion trajectories a superposition method is applied. For the point location (geometry checks) different possibilities are presented. Advantages and disadvantages of the conventional intersection method and a newly introduced octree method are discussed. The octree method was found to be suited best for three-dimensional simulation. Using the octree the CPU-time required for the simulation of one ion trajectory could be reduced so that it only needs approximately the same time as the intersection method in the two-dimensional case. Additionally, the data structure of the octree simplifies the coupling of this simulation tool with topography simulators based on a cellular method. Simulation results for a three-dimensional trench structure are presented.