Please tell us about your research subject.
Deura : Since my student days, I have been researching crystal growth and characterization of semiconductors. Currently, my research focuses on group-III nitride semiconductors.
When it comes to semiconductors, silicon (Si) is the most widely used material. However, Group-III nitride semiconductors including gallium nitride (GaN) as well as III-V compound semiconductors such as gallium arsenide (GaAs)are utilized in different fields from Si.
Semiconductors reveal their function in the form of crystals in which atoms are regularly arranged. Nitride semiconductors are expected to have various applications, including blue and white LEDs, light-receiving devices, and electronic devices such as transistors. However, further improvement in crystal quality, and progress in crystal growth technology essential. I have been studying the crystal growth mechanisms of group-III nitride semiconductors and crystal growth techniques to control their properties.
Could you share some of your research results to date?
Deura : Semiconductor thin films used in devices are fabricated by growing crystals on substrates. The substrate is preferable to the same material as the film. However, in the case of group-III nitride semiconductors, "heteroepitaxial growth (or heteroepitaxy)" is generally used, in which nitride crystals are grown on different types of substrates, such as sapphire, silicon (Si), or silicon carbide (SiC). It is extremely challenging to obtain high-quality crystals by heteroepitaxy, and thus research and development are actively pursued worldwide.
I have been working on developing a technique to grow GaN crystals on Si substrates. In heteroepitaxy, a buffer (intermediate) layer is usually grown on the substrate to mitigate the mismatch in properties between the substrate and the epilayer. However, the thick buffer layer of approximately 3 μm thick has been an issue. The thicker the buffer layer, the higher the cost of crystal growth. Therefore, I have proposed SiC/Si substrates, in which the Si surface is covered with a SiC thin film. It can be fabricated by a simple “Si surface carbonization”, in which a carbon source is supplied on the heated Si substrate. When GaN was grown on this SiC/Si substrate, a flat continuous GaN layer of GaN with single crystal orientation was obtained, demonstrating the potential for a low-cost, high-quality heteroepitaxial growth technique.
Could you tell us about the research you are currently focusing on?
Deura : As a new challenge, I am advancing a project to apply nitride semiconductors as materials for thermoelectric devices, which have not been utilized for this purpose before.
Heat is inevitably generated wherever energy is used, such as mobile phones, computers, home appliances, factories, and power plants. The inception of this research project was the idea of reusing the waste heat, which is currently mostly discarded, as an energy source. However, converting heat into other forms of energy is inherently very difficult. Therefore, we are trying to overcome this challenge by applying thermoelectric conversion technology.
In optical and electronic devices, the layer required to exhibit functionality (the active layer) is generally thin as approximately 10 nm at most. In contrast, for thermoelectric materials, a film thickness of more than 100 nm is necessary. This is disadvantageous for nitride semiconductors, for which heteroepitaxy is essential. On the contrary, inhomogeneous crystals may be more advantageous for thermoelectric materials, while homogeneity and uniformity are crucial to enhance the performance of optical and electronic devices. As a matter of fact, nitride semiconductors possess many features advantageous for thermoelectric materials, such as the ease of growing crystals with disordered structures. Therefore, I may be able to fabricate nitride semiconductor crystals suitable for thermoelectric materials by leveraging my knowledge and techniques in crystal growth. We are currently working on elucidating the inherent thermoelectric properties of the material and exploring ways to enhance thermoelectric conversion performance through structural control. Our goal is to increase the thermoelectric conversion efficiency of low-temperature waste heat below 100 °C, which has the lowest utilization efficiency, to a practical level. By combining this technology with batteries, the practical application of waste heat utilization is becoming more tangible.
Could you share the progress on crystal growth technology development?
Deura : Recently, we have also been working on developing nitride semiconductors crystal growth technology using new substrates. In heteroepitaxy, due to the mismatch in lattice constants, which represent the size of the atomic arrangement per crystal cell unit, defects frequently occur in the epilayer during growth. Furthermore, the crystal growth of nitride semiconductors requires high temperatures between approximately 500-1000 °C. Due to the mismatch in the thermal expansion coefficients of the substrate and the epilayer, the substrate can warp or crack during cooled to room temperature.
In our lab, we are focusing on ScAlMgO4 (SAM) as a new substrate material. SAM and indium gallium nitride (InGaN) have the same lattice constant, and the mismatch in their thermal expansion coefficients is also small. Because of the minimal mismatch in physical properties that typically pose challenges in heteroepitaxy, high-quality InGaN crystals can be grown on SAM substrates. If we can establish a technique for growing high-quality GaN or InGaN crystals, not only blue but also efficient yellow and red light-emitting devices can be expected. This means that the three primary colors can be realized using only nitride semiconductors. Furthermore, SAM has a property similar to mica, which makes it easy to peel off. If the substrate can be detached after growth and reused for the next growth, it also contributes to cost reduction. We are collaborating with other universities and companies with looking towards future practical applications.