The interface morphology of diamond film grown on Si

Abstract

The interfaces between diamond films and silicon substrate have been investigated to obtain a better understanding of the characteristics of diamond-substrate interface. In order to investigate the interface morphology, the backside surface of diamond film and the substrate surface have been observed after the deposited diamond film is separated from the silicon substrate. The vacant space between particles is found on the backside surface of the separated film and the central part of each particle caves in. Also, many of the hillocks are present on the Si substrate surface after the film is removed. From the observations, it is found that the interfaces between the diamond film and the silicon substrate is very rough. It is suggested that such interface morphology is attributed to the etching of the silicon substrate which takes place at the early stage of the synthesis. For the energy dispersive spectroscopy experiment of the backside surface of the separated film, more silicon is detected at the central part of each particle. This result implies that strong Si—C bonds are localized at the center of each particle, which would result in the poor adhesion.

Interface study of diamond films grown on (100) silicon

“…As the deposition proceeds, the sp 2 structure reduces and transforms to sp 3 bonded diamond as shown in the Raman spectrum. This observation is consistent with that reported by Yugo et al [12]. Fig.…”

“…As they come into contact with each other, the gaps between the particles were formed. This observation is similar to that reported by Kweon et al [12]. In their study, the diamond films were directly separated from the Si substrate with the help of the poor adhesion between the diamond film and the substrate.…”

“…Moreover, a small broad hump centered at about 1500 cm À 1 was seen in the Raman spectrum, which indicates the existing of the sp 2 type carbon. Owing to the sensitivity of sp 2 bonding phase of carbon in Raman technique is about two orders of magnitude larger than the sp 3 bonding of diamond [12], the small peaks at 1450-1600 cm À 1 represents merely a small quantity of sp 2 phase in the deposit. Comparing the films prepared at different distance between the hot filament and the substrate, the deposit prepared at the distance of 10 mm has a higher intensity of peak at 1332 wave number.…”

“…Mitsuda et al [11] observed that a vacant space between diamond particles existed at the interfaces due to the well-defined crystal habits of regular polyhedral. Kweon et al [12] reported that, on the backside of the separated film, the vacant space between particles was found and the central part of each particle caves in. Many hillocks were present on the Si substrate surface, which is attributed to the etching of the silicon substrate during diamond growth condition.…”

Space filling by nucleation and growth in chemical vapor deposition of diamond

“…On the other hand, formation of group 3 particles seems to be common in MWCVD, as nearly continuous films produced by the hot filament technique generally exhibit more uniform particle distributions and more porosity at the interface between film and substrate. 21 It would be premature to infer an electronic effect from this observation as simply the supersaturation with reactive species might be different. At present, more experimental data are required in order to gain insight into the avalanche effect.…”