Project „OPUS-25"

Study of silicon diffusion phenomenon in different crystallographic directions of gallium nitride

Project is carried out from 2024-01-07  to 2026-01-07

Project Number

2023/49/B/ST5/03319

Allocated funding

1 388 800,00zł

Project Description

Gallium nitride (GaN) is considered one of the most promising candidates for future high-power electronic devices. Switches and transistors find applications everywhere. Power semiconductor devices are critical for future energy infrastructure around the world. GaN-based power devices are currently the alternative with the potential to meet future demands. Their full triumph depends on successful solutions to materials problems that are partially analyzed and solved in this project. Silicon (Si) is the main dopant for obtaining highly conductive (n+-type) GaN in epitaxial technologies, such as Metal Organic Vapor Phase Epitaxy (MOVPE) and Molecular Beam Epitaxy (MBE), as well as halide vapor phase epitaxy (HVPE). Silicon is also implanted into n-type and p-type GaN to obtain selected areas of higher free carrier concentration in electronic devices. However, the diffusion coefficient of Si in n-type and p-type GaN, as well as the activation energy and solubility limit of Si have not been determined yet. At epitaxial temperature, of the order of 1000-1100°C, strong diffusion effects never occur in GaN. As a rule, diffusion begins at 2/3 of the melting temperature of the compound in question. It was shown that the melting point of GaN exceeds 2300°C. Hence, diffusion processes begin at 1300-1400°C. At such high temperature, GaN decomposes under atmospheric pressure. If the process of diffusion of a given impurity in GaN is considered, high temperature and therefore also high nitrogen pressure, which prevents the GaN surface decomposition, are needed. Only thanks to ultra-high-pressure annealing (UHPA) technology it is possible to study the diffusion of various elements in GaN. The main goal of this work is to analyze the diffusion process of Si in GaN of the highest structural quality and purity. Gallium nitride crystals grown by HVPE on ammonothermal GaN substrates will be used. The diffusion phenomenon in four basic crystallographic directions for GaN: , , and will be examined. UHPA technology will be used to force the Si diffusion in HVPE-GaN. It provides stability of GaN at high temperatures up to 1650°C by applying hydrostatic nitrogen pressure of up to 2 GPa. Silicon diffusion in GaN will be studied by annealing SixNy-coated and Si-implanted GaN (both treated as infinite sources of the species) at high temperature. By studying the depth profiles of Si in GaN (obtained by Secondary Ion Mass Spectrometry; SIMS) the diffusion coefficients, activation energies and solubility limits will be determined. This will be performed for the four basic crystallographic directions in n-type GaN and for the direction of p-type GaN (there is no bulk p-type GaN and, therefore, no possibilities to fabricate non-polar samples). Fitting of SIMS profiles will be performed with the Finite Difference Method. The depth profile alone is not sufficient to identify the mechanisms of diffusion. A detailed analysis has to include a direct comparison of the experimental profiles with numerical solutions of the full partial-differential-equation system. When using this method it has to be kept in mind that the charge states of the point defects involved in the diffusion process are generally important when the solubility of the dopant at the diffusion temperature exceeds the intrinsic concentration of the carriers. The Fermi level then deviates from its intrinsic position and affects the formation of charged point defects. In addition, the diffusion of the dopant can produce an electric field, which also affects the diffusion by generating a drift of charged defects, which must be accounted for in the Fick's equations. The use of two Si sources for GaN will allow for a better understanding of the role of point defects, particularly gallium and nitrogen vacancies, in the diffusion process. It is well-known that ion implantation generates Frenkel and Schottky defects in the crystal structure. They are, in turn, not generated during diffusion of any element from a sputtered layer. During the proposed project such experiments will be carried out for the first time for GaN and Si. Attention should also be paid to using GaN of high structural quality and purity. Both threading dislocations (TDs) and impurities will not strongly interfere with the diffusion process. The influence of TDs and impurities will, therefore, not be taken into account, which should facilitate the analysis of the diffusion process itself.

Publications

Project results have been published in:

• in progress

Submitted patents

none

Professor

none

PhD Theses

none

Master theses

none

Lectures

Project results have been presented by: