X-ray photoelectron diffraction of (100)-oriented chemical vapor deposited diamond films on silicon (100)

Abstract

͑100͒-oriented diamond films have been grown on silicon ͑100͒ in a microwave plasma assisted chemical vapor deposition ͑CVD͒ tubular system. X-ray photoelectron diffraction ͑XPD͒ has been used to study such oriented polycrystalline films. Comparing the diffractograms of a natural diamond ͑100͒ surface and of polycrystalline ͑100͒-oriented CVD diamond films quite similar features are observed. XPD measurements after 8 min of bias treatment show that the tiny crystals are already preferentially oriented at deposition parameters required for ͑100͒-oriented film growth. Our measurements indicate a strong need to control the growth parameters very carefully during the first minutes of growth to get an orientation.Presently, polycrystalline chemical vapor deposition ͑CVD͒ diamond films oriented toward the substrate are routinely grown on silicon 1 using microwave plasma enhanced deposition. No single-crystal thin films have so far been deposited on nondiamond substrates. However, true heteroepitaxy remains as a goal for further research activities due to the expected performance of such films in electronic applications. 2 In a first step, it is important to understand the physical and chemical mechanisms which are responsible for the oriented growth on silicon. Various studies on oriented diamond films have been done using scanning and transmission electron microscopy ͑SEM͒, ͑TEM͒, 3,4 x-ray photoelectron spectroscopy ͑XPS͒, 5 or x-ray diffraction ͑XRD͒. 6,7 While most diagnostic techniques are sensitive to m thick films only, TEM was used to reveal the structure of the silicon-diamond interface. 4 Nevertheless, there is a lack of diagnostic tools to investigate the interface in order to understand oriented growth. Moreover, it is not clear whether orientation already occurs during the first minutes of bias treatment or later during the deposition process.In this letter, we report on x-ray induced photoelectron diffraction measurements for the characterization of the very early stage of oriented diamond growth on silicon ͑100͒. This technique was used to study diamond films after 8 min of bias treatment.Low pressure diamond growth was performed on silicon ͑100͒ substrates via microwave plasma CVD in a tubular deposition system. Silicon substrates were cleaned in acetone, introduced in the plasma system, and the deposition was started after the pressure in the chamber reached 10 Ϫ6 mbar. A first run during 3 min in pure hydrogen was used to remove the native oxide layer on the substrate and to adjust the deposition temperature. Nucleation was induced by applying a dc bias of Ϫ225 V to the substrate during 8 min at 810°C and under 20 mbar of a 2% CH 4 /H 2 gas mixture. The parameters for the subsequent deposition were 870°C at 40 mbar with 1% CH 4 in H 2 . High purity hydrogen was used ͑6.0 H 2 ͒. Figure 1 shows a 15 h deposited polycrystalline CVD diamond film with 80% of ͑100͒ oriented crystals. The tilting angle is less than 8°as measured by XRD.XPD is a well-established technique in surface science for studying surf...

Comparative electron diffraction study of the diamond nucleation layer on Ir(001)

“…However, the fine structure is much less pronounced compared with the pattern of the heteroepitaxial diamond film on Ir/YSZ/Si(001). The latter yields an excellent single crystal reference since it is identical to the pattern of a diamond single crystal reported in literature [28].…”

“…In contrast, the fine structure of the present C 1s pattern shows a striking similarity to the XPD pattern taken by Schaller et al [28] from highly oriented diamond films heteroepitaxially grown on silicon. Layers of this type and thickness typically show a mosaic spread of 5-10°.…”

Local heteroepitaxy of diamond on silicon (100):mA study of the interface structure

“…Diamond nucleation was then enhanced by 8 min of dc bias treatment in a 2% CH 4 /H 2 gas mixture at a pressure of 20 mbar and a temperature of 780°C. 21 Oriented diamond growth was performed during ten additional minutes at a pressure of 40 mbar.…”

“…7͑a͔͒ are due to the weak scattering power of C atoms 20 ͑see below͒ and the weak preferential orientation of the diamond crystallites. 21 Hence, the existence of a tilt between the growing diamond lattice and the silicon substrate cannot be addressed. 27,28 However, our measurements absolutely rule out the experimental existence of the R45°theo-retical model proposed by Verwoerd 29 consisting in the epitaxy of diamond on silicon ͑100͒ through a 45°rotation of the diamond lattice relative to the substrate.…”

A chemical state resolved x-ray photoelectron diffraction study: Initial stages in diamondlike carbon film deposition

“…Usually, the films have an amorphous structure and different phases coexist by giving rise to a layered structure in which the stoichiometry varies continuously from the pure Si substrate to a hydrogenated C layer at the surface. [9][10][11][12][13][14] There are two main problems when applying the XPD technique to the investigation of the initial states of DLC growth: the emission from atoms in an amorphous phase results in a nonstructured XPD pattern and both the C and Si photoemission signals will carry information from different phases, i.e., Si:Si, Si:C, C:C, C:H.…”