Sublimation Growth of 6H-SiC Boule on Various a-Plane Substrates
- PDF / 149,609 Bytes
- 5 Pages / 595 x 842 pts (A4) Page_size
- 75 Downloads / 184 Views
SUBLIMATION GROWTH of 6H-SiC BOULE on VARIOUS a-PLANE SUBSTRATES S.Nishino, T.Nishiguchi, Y.Masuda, M.Sasaki and S.Ohshima Department of Electronics and Information Science, Faculty of Engineering and Design, Kyoto Institute of Technology, Matsugasaki, Sakyo, Kyoto 606-8585, JAPAN Phone: +81-75-724-7415, Fax: +81-75-724-7400, E-mail: [email protected], nishig5t@dj,kit.ac.jp
Abstract Sublimation growth of 6H-SiC was performed on {1100} and {1120} substrates. The difference between the growth on {1100} plane and {1120} plane was observed. {1100} facet was almost flat and there were grooves oriented toward direction. The step bunching was observed on {1100} plane 5° off-axis. A lot of pits were introduced on {1120} plane of the crystal grown both on {1100} and {1120} substrates. Step flow growth toward direction created the pits on {1120} plane. It was important to grow crystal by layer by layer growth on {1120} plane. By changing the growth mode from step flow growth to layer by layer growth, pit on the {1120} plane may be reduced as same as CVD growth on {1120} plane. Growth temperature and C/Si ratio should be optimized to keep layer by layer growth. 1. Introduction SiC can be thermally oxidized, so conventional Si process can be applied. This feature is a major advantage of SiC. However, low channel mobility has been one of the most serious problems. Recently, high channel mobility of MOSFET on {1120} plane was reported [1]. So, the crystal growth on {1120} plane is strongly focused. Conventional boule growth of SiC has been done on {0001} plane, and the research for the boule growth on {1120} and {1100} planes is limited [2, 3]. The crystal grown on {1120} plane and {1100} plane (a-plane) has different feature from the crystal grown on {0001} plane (c-plane). Though micropipes never appear in the growth on a-plane, stacking faults with a density several magnitudes higher than that of the growth on c-plane are a serious problem. Several mechanisms have been reported [3], however, the stacking fault generation has not yet been cleared well. In this paper, we focus on the sublimation growth of SiC on {1120} and {1100} substrates to investigate the growth mechanism and characteristic features of the crystal grown on these planes. 2. Experiment SiC does not have its liquid phase at normal pressure. So, the crystal growth is being conducted by sublimating the source material. The crucible was set in the water-cooled quartz tube, and it was heated by RF induction with a frequency of 30 kHz. Seed crystal was cut from the boule grown on {0001} plane. Having several degrees off {1120} and {1100} substrates were prepared. {1120} and {1100} substrates were attached on the graphite lid side by side, and grown together at same condition. The abrasive SiC powder was charged in the crucible as a source material. Sublimation growth was carried out at following conditions; substrate temperature was 1900~2100°C, source temperature was 2300~2400°C and argon pressure was 20~100 Torr. Grown crystals were observed by optical micros
Data Loading...
