Molecular Beam Epitaxial Growth And Characterization Of InN Films And Nanostructures

Ⅲnitrides as the third generation of semiconductor materials are becoming a hotspot. InN is an important component of theⅢ-nitrides, and can be used not only in light-emitting diodes, high electron mobility transistors, but also in solar cells and terahertz lasers, since its band gap was revised to 0.7eV in 2003. However, there are large mismatch of lattice and thermal expansion coefficient between InN and common substrates, such as sapphire, Si (111), GaN. Moreover, the decomposition temperature of In-polar InN is about 200℃lower than that of N-polar InN. Therefore, to grow high-quality InN, especially, In-polar InN becomes a worldwide difficulty.InN nano-structures can increase the surface area and improve the extraction efficiency of the light emitting devices. Currently, there are many reports on InN nanostructures grown on sapphire or Si substrates, but few on GaN substrates. As InN usually has a high electron concentration, preparation of p-type InN, and even the p-type InGaN with a high In component (≥30%) is very difficult. Therefore, InN devices with p-n junctions develop slowly.In this thesis, I have developed different growth techniques to grow high-quality InN thin films and InN nanocolumns on GaN. The main results are summarized as follows:1. I have determinded the temperature window for the growth of In-polar InN on GaN substrates with an InGaN buffer layer. The experimental data showed that the surface morphology and crystalline quality of InN epitaxial layers have been significantly improved by using the InGaN buffer layer.2. In phase separation in InGaN with a high In-concentration assists the growth of InN nanocolumns on GaN substrates. According to the results of transmission electron microscopy and scanning electron microscopy, I proposed a growth mechanism for InN nanocolumns. It is possible to adjust the size and density of the InN nanocolumns by changing the ratio of indium and gallium in InGaN buffer, or controlling the stoichiometry of In and N in the growth process.3. I have prepared single crystalline epitaxial Cu2O films on InN and GaN, and studied their structural and electrical properties. Cu2O on InN shows a better crystalline quality than that on GaN, due to a 30°in-plane rotation between Cu2O and InN which leads to a smaller lattice mismatch of only 0.7%. The as oxidized Cu2O on GaN shows a n-type conducting behavior, while that on InN presents a higher resistance with a lower electron concentration. Compared to Cu2O/GaN, Cu2O/InN shows an improvement of both structural and electrical properties. Moreover, a transition from n- to p-type was found after the Cu2O thin films were annealed at 500℃in vacuum. The reduction of Cu2+ accounts for this phenomenon.4. I have designedⅢ-nitride solar cells using p-type GaN as the p-layer, InGaN with a variety of In components as the i-layer, and n-type InN as the n-layer. This model avoids the difficulty to realize p-InN or p-InGaN with a high In concentration. It is expected that this kind of solar cells has high conversion efficiencies.