This section provides the scientific motivation to write this thesis, followed by a brief description of the overall structure of this work.
Despite their superior material properties, wide-bandagap semiconductor devices often suffer from severe reliability issues. Especially charge trapping phenomena prevent researchers and engineers from a further exploitation of the theoretical capabilities of the material systems.
This work primarily focuses on charge trapping in GaN devices, with a special focus on BTI degradation. Unlike silicon technology, which is a very mature technology and thus has a relatively low degradation at nominal operating voltages, comparable bias conditions in GaN devices often already lead to large instabilities of the device parameters. This work aims to give an overview of the current understanding of the physical mechanisms behind those instabilities. Furthermore, some pitfalls and peculiarities of well-established methods for reliability characterization and defect modeling specific to GaN devices are highlighted.
Specifically when using TCAD simulations, one very important phenomenon which is usually not considered for silicon is the charge feedback of the trapped charges. The effects of this mechanism and highlighting its importance for defect modeling of GaN devices is one of the main goals of this work. The second goal is the development of more robust methods for the extraction of different defect parameters from measurements. The focus here is to reliably extract the characteristic time constants of single-defects from RTN signals, which are then used to calculate different parameters like the defect structure, the trap-levels, and the vertical defect positions. Although exclusively tested for GaN devices, the proposed methods are formulated universally enough to be also transferable to other technologies.
Chapter 1 gives a brief explanation about the recent interest in wide-bandgap semiconductors and compares their most important material properties to those of silicon using different figures of merit. Furthermore, the scientific motivation of this work and a brief overview on the structure of this work is given.
Chapter 2 discusses the fundamental properties of III-N based devices. Special attention is paid to the modeling of important electronic material parameters for device simulation. The current understanding of the formation of the native 2DEG in GaN based devices together with state-of-the-art concepts for normally-on and normally-off devices is also introduced in this chapter.
Chapter 3 delivers an overview of the reliability issues in GaN devices. After a short introduction of common defects in the bulk material and the interfaces, three very important degradation mechanisms in GaN devices, namely current collapse, threshold voltage drift, and hot carrier degradation are presented.
Chapter 4 summarizes the experimental characterization of threshold voltage drift phenomena. Different measurement methods for large-area and nano-scale devices are introduced and discussed with respect to their applicability to the characterization of GaN technology.
Chapter 5 deals with the physical defect models used to simulate charge trapping throughout this work. After the description of CET maps, which are a clever way to visualize BTI degradation, the NMP model to calculate phonon assisted charge transitions in insulators and semiconductors will be discussed in more detail.
Chapter 6 is dedicated to the extraction of the characteristic time constants from stochastic charge capture and emission events. After a mathematical description of Markov processes and their application to different types of single defects, different methods to reliably estimate the expectation values of the capture and emission times are developed. First, the most common methods for extraction are compared to a new method based on spectral maps. Afterwards, a HMM able to reliably extract the time constants from complex RTN signals is put forward. After a detailed description of all parts of the algorithms, different benchmarks are used to test the robustness of the presented model.
Chapter 7 is split in two main parts. The first part investigates the relevance of charge feedback mechanisms on the 
drift of large-area GaN devices. By comparing TCAD simulations to measurement data for different voltages, the impact of these effects on defect modeling, the observed time constants and the simulations itself are discussed in detail. The second part is about single-defect characterization in nano-scale
GaN devices. It uses both extraction methods developed in the previous chapter to a) derive different defect structures from the observed RTN signals, b) extract their characteristic time constants, c) calculate different important defect parameters from these results and d) try to deduce the most likely defect candidate.
Chapter 8 gives a brief summary of the achievements of this thesis and some suggestions on future improvements of the presented methods.
