H. Sternschulte

Growth rate enhancement by nitrogen in diamond chemical vapor deposition—a catalytic effect

The diamond growth rate enhancement factor A([N2],[CH4]) of nitrogen has been measured in situ by laser reflection interferometry using thin reflecting iridium interlayers on on-axis and off-axis single crystals. “A” shows a characteristic linear decrease with the methyl radical concentration in the gas phase. The resulting local maximum in the growth rate curve yields conditions for which growth is accelerated when the methane concentration is decreased. In a model that fits the measurements quantitatively nitrogen catalyzes growth and competes with the hydrocarbon growth species for adsorption sites. The data allow excluding of several alternative models for nitrogen induced growth rate enhancement.

Diamond growth with boron addition

Diamond coatings were produced on Si substrates by the hot-filament method, with B(C2H5)3 added to the gas phase. Ratios of B(C2H5)3: CH4 up to 0.01 (10000 ppm) were used which gave boron concentrations up to 3% in the layer according to secondary ion mass spectrometry (SIMS) and elastic recoil detection (ERD) measurements. The characteristic Raman peak of diamond at 1332 cm−1 decreases with increasing boron incorporation. Studying this effect in detail shows that on (100) facets the Raman peak still can be observed while on (111) it is already severely deteriorated. TEM and localized EELS spectra show high boron incorporation in the (111) growth sectors and low boron concentration in the (100) sectors. With cathodoluminescence spectroscopy measurements electronic properties were determined. The Mott-transition from semiconductor to metal-like conduction was found to occur at 0.11% B, which is in agreement with published Hall-measurements.

Growth and properties of CVD diamond films grown under H2S addition

b ¨ Abstract The influence of H S on the CVD diamond growth, the sulfur incorporation and the electronic properties of sulfur containing 2 homoepitaxial diamond films were studied. Laser reflection interferometry (LRI) in combination with mass spectroscopy (MS) showed that H S modifies the gas phase chemistry by reducing the concentration of CH species. As a consequence thereof, at 2 x high deposition temperatures the growth rate decreased. At lower substrate temperatures, the observed increase in the growth rate after sulfur addition indicates that these gas phase effects are overcompensated by processes at the growing diamond surface. The incorporation coefficient of sulfur into the definitely boron free diamond films was very low (less than 10 ). Incorporation y6 seems to be enhanced by a reduction of the substrate temperature, by the presence of Si and, most effectively, by addition of CO . For 0.5% CO in the gas mixture a maximum S concentration of 480 ppm (9=10 ycm ) corresponding to an incorporation 19 3 22 coefficient of 6=10 was attained. Even for the highest H S concentrations (nearly 1%) the deposited diamond films preserve y4 2 their excellent quality as judged from m-Raman measurements. The electrical properties were not changed by the S incorporation. The electrical conductivity is thermally activated with typically 1.4-1.5 eV independent from the S concentration in the films. No values below 1.0 eV have been measured which argues against doping. 2002 Elsevier Science B.V. All rights reserved.

In situ characterisation of CVD diamond growth under H2S addition by optical emission spectroscopy, mass spectroscopy and laser reflection interferometry

Abstract The diamond growth process under the addition of H2S as a precursor for sulfur doping was studied in situ by optical emission spectroscopy (OES) and mass spectroscopy (MS). The changes in gas composition observed were correlated with the growth rate, which was simultaneously measured by laser reflection interferometry (LRI). While MS allows monitoring of the sulfur addition by means of the signal from the original H2S molecule or several reaction products, such as CS, OES only yields a significant signal attributed to S2 molecules. At high substrate temperatures, H2S in the gas phase causes a reduction in the diamond growth rate. A proportional decrease in the concentration of the relevant hydrocarbon growth species was measured. At the same time, the morphology of the diamond films was only weakly influenced by sulfur. Both observations suggest that the influence of sulfur on diamond growth is restricted to modifications of the gas-phase chemistry. At lower substrate temperatures, the behaviour is changed, with the growth rate slightly increased by H2S addition.

Photoconductivity Study of Li Doped Homoepitaxially Grown CVD Diamond

Spectrally resolved photoconductivity (PC) experiments on homoepitaxially grown, Li doped chemical vapour deposited diamond layers are presented. Our measurements reveal two new photoconductive levels with absorption thresholds at photon energies of 0.9 eV and 1.5 eV. Due to metastable occupation, the spectral dependence of the photoconductivity exhibits a pronounced peak. Via photoconductivity excitation experiments it could be shown that both new levels can be filled by photoionisation of the N donor. No correlation between Li content and spectral weight of photoexcitation at 1.7 eV could be observed.

CONTROL OF LITHIUM-T-BUTOXIDE ADDITION DURING CHEMICAL VAPOUR DEPOSITION OF LI-DOPED DIAMOND FILMS BY OPTICAL EMISSION SPECTROSCOPY

The potential of in situ lithium doping of diamond during microwave plasma chemical vapour deposition (MWPCVD) using a source of solid lithium-t-butoxide has been studied. It is shown that atomic lithium emission lines can be easily detected in the plasma by optical emission spectroscopy (OES). It was found that for a fixed fraction of the Li precursor in the feed gas a variation of the experimental conditions in the CVD reactor can drastically change the Li concentration in the plasma. The experimental results demonstrate that an optical control of the Li concentration in the plasma is indispensible. Elastic recoil detection (ERD) measurements clearly established that Li was incorporated into the diamond films in concentrations ranging from 40 up to 300 ppm. Etching of plasma exposed steel components in the reactor due to Li addition and the subsequent incorporation of iron, cobalt, and nickel into the films could be strongly reduced by replacing these components by graphite parts.