Equilibrium surface hydrogen coverage during silicon epitaxy using SiH4

Abstract

Epitaxial silicon has been grown on Si(100) wafers using SiH4 in a rapid thermal chemical vapor deposition reactor in the temperature regime from 450–700 °C. Gas analysis during growth and thermal desorption spectra (TDS) after growth were measured with a differentially pumped mass spectrometer. We have attempted to estimate the surface population of hydrogen during epitaxial growth by ‘‘freezing out’’ the surface hydrogen with a rapid cool down and pump down followed by a temperature programmed desorption taken in the reactor. SiH is found as the majority species in equilibrium during growth, with the surface population decreasing from one monolayer around 550–600 °C, at a pressure of three mTorr SiH4. Molecular hydrogen does not interfere with silane adsorption in this pressure regime.

Atomically Controlled Processing for Group IV Semiconductors by Chemical Vapor Deposition

“…25,26) It is found that the SiH 4 reaction rate constant, k Si on Si n 0 , agrees with the SiH constant estimated from ref. 27 within 50%. The equilibrium surface SiH 4 coverage, ðk 1 on Si =k Si on Si ÞP SiH4 =½1 þ ðk 1 on Si = k Si on Si ÞP SiH4 calculated from the fitting parameters shown in the figure caption, is also in good agreement with the data of equilibrium surface hydrogen coverage in ref.…”

“…In the substrate temperature range in our experiments, it is considered that the deposited surface could always be hydrogen-terminated by SiH 4 or GeH 4 flow. 27,32,44) It has been reported that, in the SiH 4 adsorption process, one monolayer of SiH 4 dissociation products is adsorbed on the clean Si surface in the temperature range of 25 -125 C, and one monolayer of SiH 4 is adsorbed on a surface almost fully covered with SiH (1)- (6) in Table I and the fitting parameters in Table II in the temperature range of 200 -350 C. 45) Based on these reported results, it is considered that the lower value of k MHx in our experiments is caused by SiH 4 or GeH 4 adsorption on the surface terminated with hydrogen, and the surface before SiH 4 or GeH 4 adsorption but just after the reaction by flash heating is still terminated by hydrogen, although the adsorbed species are not clear at present. This consideration supports single-atomic-layer growth in our experiments, since SiH 4 or GeH 4 adsorption on the hydrogen-terminated surface cannot be prevented by adsorption of hydrogen atoms formed by SiH 4 or GeH 4 decomposition.…”

D2 dissociative adsorption on and associative desorption from Si(100): Dynamic consequences of an ab initio potential energy surface

“…Combining these two experiments implies that the barrier to adsorption should be quite small. 15 In contrast, all measurements of dissociative chemisorption of H 2 and D 2 on Si͑100͒ find that the magnitude for this sticking is very small, [15][16][17] consistent with a high barrier to dissociation. Since the barrier to dissociation and desorption should be equivalent, the incompatibility in the two barrier determinations has been discussed in terms of an apparent violation of detailed balance.…”

Reaction dynamics of molecular hydrogen on silicon surfaces

“…Lattice excitations provide the solution to the so-called ''barrier puzzle'' resulting from recently performed state resolved experiments of hydrogen desorption from silicon surfaces: 9,11,24 It had been known for many years that the room-temperature sticking coefficient of molecular H 2 on silicon surfaces is very low, less than 10 Ϫ6 -10 Ϫ8 . [40][41][42] This demands the presence of a high adsorption barrier of V ads Ͼ0.5 eV. However, the energetics of desorbing molecules, in particular the absence of translational heating, 11 indicates the absence of a substantial potential drop during desorption.…”