Hydrogen Etching

As-received porous SiC samples are not ideal as substrates for MBE growth because of polishing damage on the surface. Flydrogen etching is often used to remove such polishing damage [14-18]. This method 

---

Figure 5.5 Plan-view SEM image of the hydrogen etched surface of porous 4H-SiC. Pores have opened up after 60 s of hydrogen etching at 1680 °C. The average pore size is about 100 nm and the surface porosity is about 3.5 %. Reproduced from A. Sagar et al.,J. Appl. Phys. 92, 4070. Copyright (2002), with permission from the American Institute of Physics...

C. Hallin, A. S. Baskin, F. Owman, P. Martensson, O. Kordina and E. Jangen, Study of the hydrogen etching of silicon carbide substrates , Silicon Carbide and Related Materials 1995, Kyoto, Japan, Inst. Phys. Conf. Ser. No. 142 (IOP, Bristol, 1996), Ch. 3. 

V. Ramachandran, M. F. Brady, A. R. Smith and R. M. Feenstra, Preparation of atomically flat surfaces on silicon carbide using hydrogen etching ,/. Electron. Mater., 27, 308 (1998). 

Figure 8.9 Cross-sectional transmission electron micrograph of GaN grown on a columnar SiC substrate. The SiC surface was annealed in hydrogen at 1200 °C for 10 min. A thin GaN buffer layer was grown first at 850 °C, followed by a thick GaN overlayer at 1030 °C. Surface contaminants not removed by hydrogen etching were found to produce small voids (v) in the buffer layer...

Kim et al.f studied the effect of gas pressure on the nucleation behavior of diamond on a Si(lOO) substrate in HFCVD. The pressure was varied from 2 to 50 torr, while a filament temperature of2200°C, a substrate temperature of 850°C, a total flow rate of 20 seem and a CH4 concentration of 0.8 vol.% were used. The characterization of diamond deposits using micro-Raman spectroscopy, SEM and OM revealed that the maximum nucleation density of diamond parades on the unscratched Si substrate occurred at a pressure of 5 torr. The pressure dependence of the nucleation density was explained by the competition effect between P-SiC formation, which increases the diamond nucleation density, and atomic-hydrogen etching, which decreases the nmnber of nucleation sites. On the basis of this finding, a new fabrication approach for high-quality diamond films without... 

Atomic hydrogen etches graphite faster than diamond. 

Hydrogen [1] and HC1 [10] at high temperatures, above 1350°C, etch SiC. Hydrogen etching is useful in cleaning the surface prior to epitaxial growth and appears to be non-preferential. 

The formation of step bunches and/or facets on hydrogen-etched Si- and C-faces of 6H-SiC has been studied by Nie et al. [38], using both nominally on-axis and intentionally miscut (i.e., vicinal) substrates. For nominally on-axis substrates, H2-etching produced uniformly distributed step-terrace arrays on both Si- and C-faces, as shown in Figure 4.10. The step arrays form because of the miscut of the surface (the average miscut values for the substrates of Figure 4.10a and b are... 

By analogy with Si (001) epitaxy we can expect formation of antiphase domains. A diamond structure reconstructed to a 2 x 1 dimer structure allows two positions for new dimer nucleation (34). If, after nucleation, two islands grow and interact, there is a 50% probability that they will have opposite phases. At the intersection, two kinds of antiphase boundaries exist (34). Such boundaries have been revealed on hydrogen-etched CVD diamond surfaces by STM. In Fig. 10 we see an orthogonal network of dimer rows. The distance between rows is 5 A. In addition to antiphase boundaries, there are narrower rows that correspond to 3 X 1 reconsfiuction. In such reconstruction we have one dihydride row (35). 

---