Ab initio derived kinetic Monte Carlo model ofH2sdesorption from Si(100)-2×1

Abstract

Using an ab initio derived kinetic Monte Carlo model, we show that within the framework of a mechanism involving the systematic conversion of monohydrides to dihydrides and the prepairing assumption, H 2 desorption from Si͑100͒-2ϫ1 via isolated dihydrides follows first-order kinetics. Our model predicts that the kinetic order is invariant with respect to coverage, temperature, and surface features ͑e.g., steps and defects͒. However, we show that the concentration of defects on the surface has a profound effect on the desorption rate constant. The dependence of the rate constant on surface quality might explain the wide range of experimental values reported for the desorption rate constant.

Density-functional study of hydrogen chemisorption on vicinal Si(001) surfaces

“…Thus the authors of these calculations favor models, in which the adsorption on and desorption from surface defects play a major role. [25][26][27][28] On the other hand, Pai and Doren [29] obtained barriers from density-functional cluster calculations that are consistent with the pre-pairing model.…”

Chemical vapor deposition of undoped and in-situ boron- and arsenic-doped epitaxial and polycrystalline silicon films grown using silane at reduced pressure

“…The growth rate will then be shifted from p H 2 Ϫ1 dependency to p H 2 Ϫ1/2 dependency. It has been reported 42 that the presence of surface defects increases the rate constant for hydrogen desorption. The surface of a polycrystalline Si film is apparently more defective than that of a Si ͑100͒ substrate; the rough surface of polycrystalline Si not only consists of numerous atomic steps extending into three dimensions but also contains grain boundaries.…”

“…The first-order kinetics has been interpreted using either a model invoking delocalization of adsorbed surface hydrogen 16,17 or a slightly different model assuming an isolated dihydride mechanism. 42 If the decomposition reactions of the surface species SiH 3 * , SiH 2 * and SiH* are considered to be faster than the dissociative adsorption of silane, 15,33,38 then it is possible to express the growth rate by…”

Silicon epitaxial growth on the Si(001)2×1 surface from silane using dynamic Monte Carlo simulations

“…Regarding this reaction called defect-induced isomerization, Radeke and Carter 15 determined rate constants for the forward and reverse reactions and predicted the rate of H 2 desorption from the Si͑001͒2ϫ1 surface using the DMC simulation through the defect-induced isomerization. 15,36,37 From these calculation results, the rate constants are determined as shown in Table I. Finally they suggested that any other processes such as monohydride-dihydride isomerization on a single dimer might increase the creation of dihydrides, especially in the case that the rate of defect migration is faster than that for dihydride reversion.…”