Fig. ETL.1. Expansion of Electronics by the Development of Hard Electronics.
Date Visited: 10 June 1998
TTEC Attendees:
P. M. Stipan (report author)
T. P. Chow
S. DenBaars
C. Uyehara
Hosts:
Dr. Kazuo Arai, Director, Materials Science Division, ETL
Dr. Hideyo Okushi, Assistant Director, Materials Science Division
Dr. Sadafumi Yoshida, Leader, Wide Bandgap Semiconductor Lab., Materials Science Division.
Dr. Hajime Okumura, Senior Researcher, Wide Bandgap Semiconductor Lab., Materials Science Division.
Dr. Shiro Hara, Interface Science on Semiconductors Group, Materials Science Division
Dr. Naoto Kobayashi, Director, Quantum Radiation Division
Dr. Toshihiro Sekigawa, Electron Devices Division
INTRODUCTION
Researchers at the Electrotechnical Lab (ETL) believe an expansion of electronics is possible through the development of "hard electronics," otherwise known as electronics based on wide bandgap semiconductors. A long-term project has started investigating 3 materials primarily: silicon carbide, nitrides, and diamond. ETL could not state the project details but did discuss the project stages, timeline, initial budget, and number of companies and universities wishing to join this project.
Presentations included ETL work on metal electrodes, SiC interfaces, hot ion implants, III-nitrides, and diamond films. ETL provided many paper reprints.
When asked about packaging concerns for these materials, Dr. Kazuo Arai commented, "One cannot make dinner when still preparing the plates - but this activity is very important and will be included in the next stage of the project (in 5 to 8 years)."
STATUS OF ACTIVITIES
ETL's own Dr. Arai is responsible for the "Hard Electronics" project sponsored by the Japanese government. He provided Figure ETL.1, which shows pictorially the application demands and the required specifications driving this project.
He believes that SiC is the material front-runner, but ETL will include the development of nitride and diamond. Substrate processing is key because of crystal growth defects, micropipe dislocations, and sublimation. Channel resistance issues and electrode issues are also being addressed using substrate processing. For example, he discussed FET enhancements, including JFET (ion implantation), MOSFET (gate oxide), and MESFET (ohmic contact).
Fig. ETL.1. Expansion of Electronics by the Development of Hard Electronics.
The project target has 3 stages:
Stage 1 research may divide into two groups:
This project started in October 1998 with its half-year budget set at ¥320 million and an anticipated project budget set at nearly ¥7.5 billion. Presently 10 companies wish to join. Six to 8 universities collaborate with ETL, but the exact number depends on the budget.
LPCVD
ETL researchers have found that LPCVD improves the quality of 3C-SiC epilayers on Si with atomically smooth surfaces as compared to APCVD. They plan to demonstrate the quality by fabricating a Schottky device. They use LPCVD-grown 3C-SiC epilayers as the substrates for the growth of cubic III-nitrides.
SiC Interfaces
Dr. Shiro Hara presented SiC interfaces and metal electrodes for 6H-SiC (0001). A graph showed the Schottky barrier height versus metal work function for titanium, molybdenum, and nickel. A 5% HF etch is used for cleaning along with boiling water (new process with oxygen dissolved in pure water). This lowers the contact resistivity.
Hot Ion Implant
Dr. Kobayashi presented information on hot ion implant. For 2-3 years ETL has focused on SiC process and conduction control (device processes). He said that it is difficult for proper p-types (use Al or B) in SiC, but the n-type is easier and N2 is appropriate. Aluminum sublimation occurs in a high dose implant. He discussed Poisson annihilation for defect control and stated that no implant has been used at ETL for GaN.
III-Nitrides
ETL grows crystals by Molecular Beam Epitaxy (MBE), but this technique has not been the best. ETL uses a new plasma to improve the growth rate. It uses surface science for hexagonal epilayers on sapphire and cubic layers on GaAs and 3C-SiC. The optical properties of cubic III-nitrides are used for characterization. The h-GaN (hexagonal-GaN) is 0.2eV larger than the c-GaN (cubic-GaN). The future plan is to evaluate the function of III-nitride heterostructures, wide bandgap materials with large band offset, large saturation drift velocity, a small dielectric constant, and chemical and thermal stability.
Diamond Films
High pressure and high temperature synthesize diamonds. Impurity problems over large areas are a concern. The two methods ETL evaluated for epi (crystalline) are hetero-epi growth and homo-epi growth. ETL has found that homo-epi growth produces a high quality, atomically flat film. For clean epitaxy, ETL researchers recommend the following:
ETL presented dependence of CH4 concentration (CH4 /H2). It compared concentration percentages at 6 hours' deposition time from 0.05%, 0.15%, 0.3%, 0.5%, 1% (defect loaded), and 2.0%. At 0.05%, the homogeneous material is clean. Going to 0.025% reveals an atomic force image of a 200-nm x 200-nm film with an atomically flat surface in the whole region of the substrate-but this process takes 42 hours' deposition time.
ETL showed current density versus voltage curves for an Al/diamond (001) Schottky barrier diode with F b = 1.58eV.
For a B-doped CVD diamond (using B[CH3]3), ETL presented the temperature dependence of Hall mobility compared to natural diamond.
REFERENCES
Arai, K. et al. 1997. Prospect of hard electronics: what and how approach. (paper). Oct .
Balakrishnan, K., et al. 1997. Structural analysis of cubic GaN through X-ray pole figure generation. (paper). October.
Balakrishnan, K., et al. 1998. Study on the initial stages of heteroepitaxial growth of hexagonal GaN on sapphire by plasma assisted MBE. (paper). March.
Feuillet, G., et al. 1997. Surface reconstructions and III-V stoichiometry: the case of cubic and hexagonal GaN. (paper).
Feuillet, G., et al. 1997. Arsenic mediated reconstructions on cubic (001) GaN. Feb.
Hara, S., et al. 1997. Control of Schottky and ohmic interfaces by unpinning Fermi level. (paper).
Ishida, Y. et al. 1997. Atomically flat 3C-SiC epilayers by low pressure chemical vapor deposition. (paper).
Ishida, Y. et al. CVD growth mechanism of 3C-SiC on Si substrates. (paper).
Kobayashi, N., et al. 1997. Ion-beam-induced epitaxial crystallization (IBIEC) and solid phase epitaxial growth (SPEG) of Si1-xCx layers in Si fabricated by C-ion implantation. (paper).
Okumura, H. 1998. Arsenic surfactant effects and arsenic mediated molecular beam epitaxial growth for cubic GaN. (paper). June.
Okumura, H. et al. 1994. Epitaxial growth of cubic and hexagonal GaN by gas source molecular beam epitaxy using a microwave plasma nitrogen source. (paper).
Okumura, H. et al. 1997. Growth and characterization of cubic GaN. (paper).
Okumura, H., et al. 1997. Bandgap energy of cubic GaN. (paper).
Okumura, H., et al. 1998. Analysis of MBE growth mode for GaN epilayers by RHEED. (paper). March.
Okumura, H., et al. 1998. Growth of cubic III-nitrides by gas source MBE using atomic nitrogen plasma: GaN, AlGaN and AlN. (March)
Reddy, C., et al. The origin of persistent photoconductivity and its relationship with yellow luminescence in molecular beam epitaxy grown undoped GaN. (to be published in APL).
Takahashi, T., et al. Surface morphology of 3C-SiC heteroepitaxial layers grown by LPCVD on Si substrates. (paper).
Tanaka, Y., et al. Hot implantation of Ga+ ion in SiC. (paper). Dec.
Teraji, T., et al. 1997. Ideal ohmic contact to n-type 6H-SiC by reduction of Schottky barrier height. (paper). May.