Date Visited: 12 June 1998
TTEC Attendees:
J. H. Maurice (report author)
T. P. Chow
S. DenBaars
P. M. Stipan
C. Uyehara
Hosts:
Dr. Isamu Akasaki, Professor,
INTRODUCTION
Meijo University, a private university, is one of the largest universities in central Japan. It has 6 faculties and 14 departments covering all the fields of the social and natural sciences, as well as interdisciplinary fields. It has a graduate school and a junior college division and also provides special course study. It began in 1926-28 as Nagoya Advanced Science and Technology School; in 1951 it was established as Meijo University. It has 3 campuses in the Nagoya area, and its slogan is "Towards the Age of Human Renaissance." In 1996, the High Tech Research Center at Meijo was the first of its kind in Japan to focus on HTE.
Professor Akasaki is an important pioneer in the field of semiconducting GaN and related devices, a well-recognized and highly accomplished figure in his field. The focus of his research activities is group-III nitride semiconductors, their crystal growth, and their electrical and optical properties.
Status of Activities
There was a brief discussion of the research activities of Prof. Akasaki's team. According to the materials provided, Meijo's focus is on the structural and optical properties of strained quantum wells in III-V nitrides and their application to short wavelength lasers. Some recent projects and work include the following:
Main Points of Discussion: Questions and Answers
Q.How is the vision of HTE defined in Japan?
A.As rad-hard, high temperature, high frequency, high switching voltage/power for engines (automotive, aerospace), the high saturation velocity of electrons. Also what Akasaki named in 1988 "Frontier Electronics." The future of photonic devices, for example, is in TV screens or in monitoring (e.g., atmospheric pollution monitoring).
Q.Is your organization involved in any government HTE programs?
A.Yes, under the Ministry of Education, Science, Sports and Culture of Japan, this High Tech Research Center (HTRC, see above, established in 1996) is one of several such centers, its focus being HTE. There is also involvement with projects under the Japan Society for the Promotion of Science (JSPS) and the New Energy and Industrial Technology Organization (NEDO, a MITI member), which funds research for combustion control systems for energy conservation. The 3 big Japanese projects for nitrides are the HTRC, JSPS, and NEDO.
Q.How long will it take to develop the materials?
A.It's hard to say, as there are bigger (more important) improvements occurring, but there is still no substrate.
Q.What are the major issues? What is the most difficult problem?
A.Material growth.
Q.What temperature range would you consider for HTE operation?
A.The metallization is limited: At 600°C, the metallization becomes unstable, and at 800°C, it sublimates.
Q.What power level would you consider for high-temperature devices?
A.100 W at several tens of GHz. The power level is always frequency dependent, and everyone is always hungry for frequency and power.
Q.What wide bandgap semiconductor material has the best potential for wide bandgap electronics?
A.Diamond, nitrides, and SiC. The n-type doping is difficult to prepare in diamond, though the p-type is easy. SiC has MOS capability, yet using it is difficult because it cannot make heterostructures. There is also the SiC substrate problem. Nitrides can be grown on sapphire, much more cheaply than SiC at present. If SiC substrate becomes cheaper and more available-if Cree cut its price, say-this might boost research. Dr. Akasaki said that (at the time of the panel's visit) nitride looks best. Judging from the past few specialty conferences, GaN work constitutes 55% of all semiconductor research.
Q.What are the most important material characteristics for substrate material for wide bandgap electronics?
A.Clearly defect density.
Q.What are the most important material characteristics for epitaxial material for wide bandgap, high temperature, high power electronics?
A.Of doping uniformity, thickness uniformity, growth rate, interfaces, defect density, electrical properties, surface properties, and cost-none of the above. The most important characteristic of the material must be its uniformity, reproducibility, and consistency in producibility. Of course defect density at the interfaces and doping uniformity fall under this.
Q.What are the main issues in manufacturing materials for HTE?
A.Clearly, defect control.
Q.Does Meijo University grow material for HTE? If so, by what methods?
A.Yes, nitride, by MOCVD.
Q.Is MOCVD superior to MBE?
A.Yes, as it's easier to do (the vapor pressure of nitride compared to As).
Q.What device building blocks (e.g., ohmic contacts, Schottky barriers, p-n junctions, device isolations, etc.) do you consider to be bottlenecks for high temperature devices fabricated in SiC and group-III nitrides?
A.Ohmic contacts, but also device isolation is not easy.
Q.Is ion implantation for doping possible in GaN?
A.Doubtful. Prof. Akasaki attempted 20 years ago both p- and n-type implantation-both with no good results at that time.
Q.What are the specific features of interconnects, passive components, and packaging for high temperature operations? What are the barriers for the development of passive components for HTE?
A.Resistors and capacitors. Interconnects suffer reliability problems and so are not that easy. The key parameters for SiC and group-III nitrides for the design of high temperature, high power electronics are breakdown voltage, thermal-conductivity, saturation velocity, and turn-on resistivity.
Perhaps as in the United States, there is a three-way split among industry, academia, and government regarding GaN work. For SiC, the split is perhaps 40% government and 35% industry.
Comments
Materials
Issues
REFERENCES
Oyo-Butori International reprint of an interview with Prof. Akasaki
High-Tech Research Center Annual Report (volume of consolidated recent research, written in Japanese)
Lab tours (from window):
Phillips Expert System for X-ray diffraction characterization
MOCVD systems for nitrides
MBE
Analysis Room (buffer layer work, FIB, Hitachi TEM)
Clean Room
Lab Facility: approx. 500 m2