Film quality in relation to deposition conditions of a-SI:H films deposited by the ‘‘hot wire’’ method using highly diluted silane

Abstract

The deposition parameter space has been extensively explored using the hot wire technique with 1% SiH4 in He as a source gas. To achieve reasonable deposition rates despite the high dilution, the filament was positioned at 1–2 cm from the substrate. This short distance introduced a large nonuniformity across the substrate in deposition rate as well as in film properties. These spatial variations were used to analyze which factors in the deposition determine film quality. Radiation from the filament as well as deposition rate cannot explain the large variation in film properties, leaving gas-phase reactions of Si and H from the hot filament as the primary cause. It is clear that radicals evaporated from the filament must undergo gas-phase reactions with SiH4 before deposition in order to produce high-quality material. Thus, conditions such as increasing the chamber pressure or going to a heavier carrier gas increase the fraction of radicals that can react before reaching the substrate and, therefore, improve the film quality. However, such conditions also enhance multiple radical reactions before such radicals reach the substrate and this can have a negative effect on film quality: this is attributed to gas-phase nucleation with incorporation of conglomerates. The gas-phase chemistry is quite different from that of plasma-enhanced decomposition in that no disilane or trisilane is formed in significant quantities. This, and the dependence on pressure, indicates that the pathway for formation of these heavier particles is radical–radical reactions.

The aging of tungsten filaments and its effect on wire surface kinetics in hot-wire chemical vapor deposition

“…In the low pressure, collisionless regime, these wire-desorbed radicals may act as the primary film deposition precursors, while at higher pressure, they may react with other species in the gas phase to produce the precursors. In particular, the study by Molenbroek et al 1 suggested a direct correlation between amorphous film quality and the degree of reaction of wiregenerated Si in the gas phase.…”

Hydrogenated amorphous silicon deposited at very high growth rates by an expanding Ar–H2–SiH4 plasma

“…The probability for these reactions, however, is limited due to the relatively low SiH 4 pressure and small distance between filament and substrate ͑typically р5 cm͒. It is expected that a considerable fraction of the ͑hot͒ Si atoms can react with SiH 4 ͑to mainly H 2 SiSiH 2 ), 63,78 but this is less evident for H. It is rather improbable that H reacts to SiH 3 in the gas phase and the proposed scheme that SiH 3 is created at the a-Si:H surface by H ͑i.e., H creates a surface dangling bond by H abstraction and this surface dangling bond abstracts subsequently a H atom from SiH 4 ) 78 is also unlikely because of the very low reactivity of SiH 4 with pristine crystalline Si surfaces. 79,80 It can therefore be excluded that in HWCVD SiH 3 contributes dominantly to film growth.…”

Pressure-dependent rate coefficients of chemical reactions involving Si2H4 isomerization from QRRK calculations

“…2). Hence, it is clear that triplet silylsilylene will cross sooner or later to singlet state, which easily isomerizes to singlet disilene H 2 SiSiH 2 ( 1 A g ) [51,54]. If the crossing between the two systems is fast, vibrationally excited H 3 SiSiH( 3 A 00 ) produced from collisions between SiH 4 and Si( 3 P) can be converted to excited singlet and then may dissociate to Si(H 2 )Si( 1 A 1 ) + H 2 at low pressure.…”