Refine
Document Type
- Doctoral thesis (3)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Language
- English (3) (remove)
Keywords
- Galliumnitrid (3) (remove)
Although GaN HEMTs are regarded as one of the most promising RF power transistor technologies thanks to their high-voltage high-speed characteristics, they are still known to be prone to trapping effects, which hamper achievable output power and linearity. Hence, accurately and efficiently modeling the trapping effects is crucial in nonlinear large-signal modeling for GaN HEMTs.
This work proposes a trap model based on an industry standard large-signal model, named Chalmers model. Instead of a complex nonlinear trap description, only four constant parameters of the proposed trap model need to be determined to accurately describe the significant impacts of the trapping effects, e.g., drain-source current slump, typical kink observed in pulsed I/V characteristics, and degradation of the output power. Moreover, the extraction procedure of the trap model parameters is based on pulsed S-parameter measurements, which allow to freeze traps and isolate the trapping effects from self-heating. The model validity is tested through small- and large-signal model verification procedures. Particularly, it is shown that the use of this trap model enables a dramatical improvement of the large-signal simulation results.
Gallium nitride (GaN) is a III-V semiconductor, characterized by direct, wide band gap of 3.4 eV at RT. As a material of particular interest for opto- and power electronics applications, it has been thoroughly studied in recent years. Utilization of GaN homoepitaxy in manufacturing of laser diodes (LDs), light-emitting diodes (LEDs), power devices, etc. would be beneficial in terms of reducing defect density, thus improving their lifetime and performance. Yet cost-effective process for providing native GaN substrates has not been established so far.
The focus of this work is put on development of a new method to grow single crystalline GaN layers from Ga vapour. Our approach exploits microwave (MW) plasma as a source of excited nitrogen species, in contrast to classical physical vapour transport (PVT)-based technique, in which ammonia (NH3) serves as a source of reactive nitrogen. Novelty of MW plasma enhanced growth of GaN from vapour lies in MW nitrogen plasma formation in the vicinity of the seed, at moderate pressure (200 – 800 mbar range), and concurrent physical vapour transport of Ga to the growth zone. Simulations of the growth setup (HEpiGaN software) and of the MW plasma source (CST Microwave software) have followed the extensive investigations of material properties. The growth setup and the MW plasma source, with the resonance cavity being its crucial part, have been constructed and implemented into the existing growth reactor.
The stability of MW plasma in function of temperature and pressure has been studied along with its influence on the seed temperature, and thus on the growth conditions. Furthermore, optical emission spectroscopy (OES) has been utilized for in-situ characterization of the growth atmosphere. Studies on the interaction of Ga vapour with the nitrogen discharge were interpreted on the basis of the level structure of lower excited states of Ga.
Deposition experiments have been conducted, using sapphire seeds, GaN, AlN and AlGaN templates, while GaN single crystalline layers have been grown on sapphire and GaN templates. Characterization of GaN layers have been done by various methods, i.e. structure of layers by scanning electron microscopy (SEM), their composition by energy dispersive X-ray spectroscopy (EDX) and secondary ion mass spectrometry (SIMS), and crystal quality by high resolution X-ray diffraction (HRXRD). Results of the characterization together with outcome of OES measurements revealed importance of carbon for the sub-atmospheric MW plasma enhanced growth of GaN from vapour. In addition, this fact was confirmed by experiments in the setup with reduced carbon content. Possible routes for GaN synthesis have been discussed, with the most probable being CN-assisted GaN formation. While CN was detected in the plasma spectra, there was no evidence for the existence of GaN molecules in vapour phase.
Novel oxide buffer approach for GaN integration on Si(111) platform through Sc₂O₃/Y₂O₃ bi-layer
(2012)
Motivation: Preparation of GaN virtual substrates on large-scale Si wafers is intensively pursued as a cost-effective approach for high power/high frequency electronics (HEMT's etc.) and optoelectronic applications (LED, LASER). However, the growth of high quality GaN layers on Si is hampered by several difficulties mainly related to a large lattice mismatch (-17%) and a huge difference in the thermal expansion coefficient (56%). As a consequence, GaN epitaxial layers grown on Si substrates show a high number of defects (threading dislocations etc.), which severely deteriorate the overall quality of the GaN films. Additionally, due to the different thermal expansion coefficients of the substrate and the film, µm-thick GaN layers crack during post-growth cooling. To solve these integration problems, different semiconducting (e.g. AlN, GaAs, ZnO, HfN) and insulating (e.g. Al₂O₃, MgO, LiGaO₂) buffer layers, separating the Si substrate from the GaN film, are applied. Goal: In this thesis, a novel buffer approach for the integration of GaN on Si is proposed and investigated. The new approach employs Sc₂O₃/ Y₂O₃ bilayer templates as a step-graded buffer to reduce the lattice mismatch between GaN and the Si(111) substrate. According to the bulk crystal lattices, since the Y₂O₃ has an in-plane lattice misfit of -2% to Si, Sc₂O₃ -7% to Y₂O₃, the lattice misfit between GaN and the substrate can be theoretically reduced by about 50% from -17% (GaN/Si) to -8% (GaN/Sc₂O₃). Experimental: The GaN/Sc₂O₃/ Y₂O₃/Si(111) heterostructures are prepared in a multichamber molecular beam epitaxy system on 4 inch Si(111) wafers. In order to obtain complete information on the structural quality of the oxide buffer as well as the GaN layer, synchrotron- and laboratory-based x-ray diffraction, transmission electron microscopy and photoluminescence measurements are performed. The topography of the films is characterized by scanning electron microscopy and chemical inter-diffusion is investigated by energy-dispersive x-ray spectroscopy. The nucleation processes of the GaN onSc₂O₃ buff er are followed in-situ by reflection high energy electron diffraction and the interface chemistry is analyzed by means of x-ray photoelectron spectroscopy. Results: It is found, that the Sc₂O₃/ Y₂O₃ buffer approach provides a template of high structural quality for GaN overgrowth. The bi-layer buffer plays a lattice match mediator role between GaN and Si and acts as a barrier against impurity diffusion. GaN grown on Sc₂O₃/ Y₂O₃/Si(111) templates is single crystalline with a wurtzite structure and (0001) oriented. Due to the -8% lattice mismatch between GaN and Sc₂O₃, GaN growth proceeds by the nucleation of 3D islands. The size of the islands, coalescence time and the relaxation process depend on the GaN growth conditions and have a strong influence on the topography of closed layers, crystalline quality (defect density) as well as optical properties. The best GaN material parameters are obtained for the layers grown in Ga-rich regime when the Ga/N ratio is slightly higher than unity. The main three defects found in the µm-thick GaN layers are a) threading dislocation, with density in the order of 10^10 cm-2, b) stacking faults, resulting in cubic inclusions in the hexagonal matrix and c) inversion domain boundaries causing Ga-polar regions in the mainly N-polar film. A theoretical GaN/Sc₂O₃ interface model is discussed to explain these experimental findings. Despite the relatively large number of structural defects, photoluminescence shows sharp and strong donor-bound exciton transition and very low intensity yellow emission, which indicate that GaN layers grown on Sc₂O₃/ Y₂O₃/Si(111) are promising for future optoelectronic applications. Outlook: Future growth strategies will focus on interface engineering approach to further reduce the lattice mismatch between GaN(0001) and Sc₂O₃ (111) surfaces, enable growth of unipolar GaN films and trigger the occurrence of an early 2D-like growth mode to avoid cubic GaN inclusions and guarantee low threading dislocation densities.