Hydrogen Plasma Treatment of Natural and Homoepitaxial Diamond

Abstract

As-polished and homoepitaxial diamond (100)- and (111)-oriented single crystals, natural diamond powders of 0.12-6 μm diameter, and high-pressure/hightemperature (HPHT) diamond powders of 50-100 μm diameter were treated in a microwave hydrogen plasma under four sets of conditions. Negligible changes in the weights (±10 μg) of the as-polished and homoepitaxial diamonds were observed.Post-treatment atomic force microscopy (AFM) showed a combination of smoothing and pit formation on (100) surfaces, the details of which were sensitive to the structure of the starting surface. Asymmetric pits formed on an as-polished (111) surface, but no qualitative changes in the structure of highly defective (100) and (111) films were observed. Natural diamond powders, which were quite irregular and rough prior to treatment, became markedly smoother and well-faceted, as observed by scanning electron microscopy (SEM). The degree of faceting was sensitive to plasma power level but was independent of H2 flow rate. The size and degree of faceting appeared to be the same after plasma treatment for isolated and closelypacked particles; however, the latter fused into a quasi-continuous film. “Regrown” crystallites were observed on the surfaces of synthetic type Ib diamond particles following plasma treatment. We argue that surface diffusion is the dominant mechanism for the observed morphological changes.

Etching effects during the chemical vapor deposition of (100) diamond

“…The simulated growth rates exhibit maxima around 1200 K, whereas the experimental measurements of single-crystal growth rates [16,17] do not show such a pronounced peak. On the other hand, measurements of polycrystalline growth rates do show maxima with temperature [60,6 1], but this might arise due to the deposition of non-diamond carbon at high temperatures in these experiments [60].…”

“…Experimental evidence [64] also suggests that much of the smooth, ordered (100)(2x1) material observed in the experiments might not be produced during growth, but rather by hydrogen-assisted surface diffusion or etching / regrowth in a H / H, plasma prior to scanning probe analysis. In addition, (100) faces on CVD diamond often show macrosteps [65,66], presumably from step bunching, which suggests that some mechanism (perhaps nitrogen-assisted) exists to pin (100) steps as they grow.…”

“…3a are (100) growth rates from Rawles et aL [17] (0.570 CH~in ?31fed into a hot filament reactor held at 25 TOIT).When the growth simulations are performed without etching, the (100) growth rate ( Fig. 3a) is highest, followed by (110) (Fig.…”

Surface Migration in Diamond Growth

“…Recent experiments systematically investigating this phenomenon led to a conclusion that surface migration is the more likely factor among several considered to explain the observations. 67,70 The examples presented in Figures 6 and 7 were computed for one-dimensional migration along dimer chains. Migration along dimer rows should be numerically close to that, because of the similar nature and thermochemistry of the reaction mechanism.…”

Point defect incorporation during diamond chemical vapor deposition

“…3b are comparatively low. The quality of hot filament CVD diamond is known to deteriorate due to sp2 bond formation above about 1% inlet CH~ [28,29] and, hence, inlet CH~concentrations in the 0.3% to 1.0% range are commonly employed [15,[28][29][30][31]. The predictions in Fig.…”