Heteroepitaxial Growth of GaN on Diamond Substrates with High Thermal Conductivity

自从20世纪80年代单晶GaN薄膜的生长取得突破性进展以来，直接带隙半导体GaN已经成为最重要的工业应用半导体之一。没有它，如今无处不在的发光二极管的发展是不可行的。但GaN不仅在发光二极管方面有很大的应用价值，而且在电力电子器件，特别是高频器件方面也有很大的应用价值。然而，在高功率密度下，这些设备正在达到一定的限制。来自有源器件区域的热传输不足限制了它们的最大功率密度。为了提高这些器件的功率密度或可靠性，改善热传输将是有利的。该项目的目标是通过在钻石衬底上直接生长GaN来实现这一点。在该项目的第一部分，将在单晶金刚石衬底上生长GaN薄膜，以确定最佳的生长条件。由于人造金刚石单晶价格昂贵，因此只对特定应用感兴趣，因此该项目的第二个也是最重要的部分旨在在多晶金刚石薄膜上生长单晶GaN薄膜。为了实现这一目标，需要通过对金刚石薄膜进行预处理和选择合适的生长条件来影响GaN的成核，使GaN核只在相同取向的金刚石晶体上形成。随后，GaN原子核将结合成一个封闭的单晶膜。在项目的第三部分，也是最后一部分，将演示开发过程的可伸缩性。这一过程将转移到6英寸或8英寸的硅晶片上，这些硅片上覆盖着多晶金刚石薄膜。

Since breakthroughs in the growth of single crystal GaN films in the 1980s, the direct bandgap semiconductor GaN has become one of the most important semiconductors for industrial applications. Without it, the development of the nowadays ubiquitous light-emitting diode would have been unfeasible. But GaN is not only interesting for the application in light emitting diodes but also in power electronic devices, especially for devices with high operating frequency. However, at high power densities these devices are reaching certain limitations. The insufficient heat transport form the active device region limits their maximum power density. To increase the power density or the reliability of these devices it would be favourable to improve the heat transport. The aim of this project is to achieve this by directly growing GaN onto diamond substrates. In the first part of the project GaN films will be grown on single crystal diamond substrates to determine the optimum growth conditions. As even synthetic diamond single crystals are expensive, and thus only of interest for specific applications, the second and most important part of the project aims at growing single crystal GaN films on polycrystalline diamond films. Achieving this goal will require to influence the nucleation of the GaN in such a way that GaN nuclei only form on diamond crystallites with the same orientation by pretreating the diamond films and choosing appropriate growth conditions. Subsequently the GaN nuclei will coalesce into a closed single crystal film. In the third, and last, part of the project the scalability of the developed process will be demonstrated. The process will be transferred to 6” or 8” Si-wafers that are coated with polycrystalline diamond films.