Several materials have shown promise for coating graphite susceptors used in epitaxial growth of 4H-SiC. The development of appropriate susceptor-coating materials is an essential part of continuing efforts to produce high-quality, thick, undoped and lightly doped epilayers of 4H-SiC for use as semiconductors in future electronic devices that will operate at power levels and temperatures beyond the limits of Si- and GaAs-based devices.

A 4H-SiC Epilayer was grown by CVD, using a carbon-based coating on a graphite susceptor. An 11.5-µm thick epilayer showed a smooth surface morphology, and n-type doping control via site competition epitaxy was demonstrated as shown.

SiC grows in many different crystalline structures called "polytypes." 4H-SiC is one of these polytypes and is the preferred one for electronic devices because in comparison with other SiC polytypes, its charge-carrier mobility is greater and the anisotropy of its mobility is smaller. Epitaxial films of SiC are made by chemical vapor deposition (CVD) from carbon- and silicon-containing gas mixtures onto radio-heated susceptors. Heretofore, the most common susceptor materials have been graphite and SiC-coated graphite.

The input flows of Si- and C-containing gases into an SiC CVD chamber are well controlled by use of mass-flow controllers, but this does not always suffice to achieve the needed precise control of the proportion of C to Si: Both bare graphite and SiC-coated graphite susceptors are vulnerable to etching at the growth temperature (typically >1,350 °C) in the presence of H2, which is one of the gases in a typical CVD gas mixture. Depending on the susceptor material and the particular CVD process, such etching releases uncontrolled amounts of C and/or Si into the CVD gas mixture, thereby giving rise to undesired changes in the structure, density of defects, and doping of the epilayer. Etching of a graphite susceptor can also release B, Al, and other impurities that can become incorporated into the epilayer as undesired dopants. Moreover, a bare graphite susceptor is porous and can trap a desired dopant during a CVD run, then release the dopant during a subsequent CVD run, in which the dopant could be undesired.

Thus, a graphite-susceptor-coating material that is much less vulnerable to etching is needed for growing high-quality 4H-SiC. A suitable material must have a coefficient of thermal expansion close to that of the underlying graphite, must adhere to the graphite without flaking or chipping off during CVD, must be sufficiently thermally conductive to transfer heat from the susceptor to the growing epilayer, must be amenable to deposition in a layer thin enough not to obstruct coupling of radio-frequency energy to the susceptor, should have low porosity to prevent trapping of impurities, and should be as chemically inert as possible in the presence of hot H2.

Four materials that satisfy these criteria to various degrees were selected as candidates for experimental evaluation. One of these materials was SiC, which as reported above, was already known to be vulnerable to etching. The other materials were carbon-based coatings, denoted as C1, C2, and C3. The C1 coating was used to establish an epitaxial-growth baseline for comparisons. The films grown with SiC-coated graphite susceptors were of poor quality in comparison with those grown on the C1-coated susceptors, and the SiC-coated susceptors were severely degraded, leading to the conclusion that SiC should be excluded from further consideration. On the other hand, the SiC films produced with the C2- and C3-coated susceptors were of high quality, and the doping control was demonstrated (see figure). The C2-coated susceptors exhibited no obvious degradation and appeared to have operational lifetimes longer than those of the baseline C1-coated susceptors. The C3-coated susceptors exhibited some chipping. These results suggest that these carbon-based coatings could be suitable for epitaxial growth of high-quality 4H-SiC.

This work was done by Barbara Landini of Advanced Technology Materials, Inc., for Glenn Research Center. No further documentation is available.

Inquiries concerning rights for the commercial use of this invention should be addressed to

NASA Glenn Research Center,
Commercial Technology Office,
Attn: Steve Fedor,
Mail Stop 4 —8,
21000 Brookpark Road,
Cleveland, Ohio 44135.

Refer to LEW-16699.