Ab initiostudy of the diffusion and decomposition pathways ofSiHxspecies on Si(100)

Abstract

Diffusion and decomposition of SiH x species adsorbed on the clean Si͑100͒ surface are processes of relevance for the growth of crystalline silicon by plasma-enhanced chemical vapor deposition. In this work, we report an extensive search of diffusion and decomposition pathways for SiH 3 , SiH 2 , and SiH by means of combined ab initio metadynamics simulations and optimization of minimum-energy reactions paths. We find that on the clean surface SiH 3 undergoes stepwise decompositions into Si and H adatoms according to SiH 3 → SiH 2 +H→ SiH+ 2H → Si+ 3H with an overall reaction barrier of the order of 0.8 eV, consistent with the scenario inferred from secondary ion mass spectroscopy data. The lifetime of SiH 3 at room temperature calculated within transition state theory in the harmonic approximation is in agreement with experiments. The lifetime of SiH 2 turns out to be similar to that of SiH 3 . Possible trap states for SiH 2 are proposed, based on energetics and by comparing calculated scanning tunneling microscope images with experimental data.

Metadynamics

“…These included studies on small solvated amino acids,79 polyoxometalates,80 beta‐ d ‐xylose condensation reactions,81 ester82 and formamide hydrolysis,83 and solvation84 and isomerization in water 85. The area of reaction at surfaces has also been explored to study chemical vapor deposition,86 heterogeneous reactions on carbon surfaces,87,88 and on anatase 89. Studies also included reactions in the solid state 90,91.…”

Computational Study for Reactions of H Atoms with Adsorbed SiH3 and Si2H5 on H-Covered Si(100)-(2 × 1) Surface

“…The bottom three silicon layers are constrained at the bulk parameters, while the remaining layers are fully optimized. Previous studies have demonstrated that the supercell used in this work is sufficient for modeling the silicon surface. ,, It should be noted that we also performed test calculations for SiH 3 and Si 2 H 5 adsorption on six-, seven- and eight-layer Si(100) surfaces. The difference (less than 1.0 kcal/mol) of the adsorption energy can be neglected.…”

“…Kubo et al have used tight-binding quantum chemical molecular dynamics to simulate silicon crystal growth mechanisms for deposition of SiH 3 and SiH 2 radicals on the H-terminated Si(001) in a CVD process. , Maroudas et al reported several detailed studies on the hydrogen-induced crystallization mechanisms of amorphous silicon thin film via molecular dynamics (MD) simulations. − Montalenti also showed hydrogen atoms can promote growth of ordered and epitaxial Si thin films in PECVD at low temperature on the H-terminated Si(001)-(2 × 1) by employing Car–Parrinello simulations and static density functional theory (DFT) calculations . In addition, DFT calculations have provided a way to obtain the electronic adsorption structures and the silicon-related reaction mechanisms accurately. − In this work, we study the detailed mechanisms of the reactions of H atoms with adsorbed SiH x ( x = 1–3) and Si 2 H 5 species and predict the rate constants for the mentioned reactions by means of DFT calculations and transition-state theory (TST).…”

Well-Tempered Metadynamics as a Tool for Characterizing Multi-Component, Crystalline Molecular Machines

“…Its success and increasing popularity in the literature is due not only to the efficiency of the algorithm and the acceleration it offers in both classical MD and ab initio approaches but also to its versatility, as the choice of CVs that can be used is almost limitless. A wide array of CVs have been reported in the literature: structural order parameters to accelerate melting, clustering, − and binding-type processes; , phase transitions have been induced using the simulation cell parameters as CVs; − conformational changes of molecules have been accelerated using intramolecular bond angles or order parameters for CVs, − including many studies of protein folding − and other biological processes such as ion gating , and substrate binding; , and finally, simulations of chemical reactions have been accelerated using simple internuclear vectors, basic atomic coordinates, and coordination numbers as CVs. − This versatility would clearly be a major advantage to the simulation of molecular machines, where the motions of different molecular components in a complicated system could each be represented by a separate CV in the multidimensional CV space of a single metadynamics simulation. The efficiency of the algorithm would then ensure that this complicated space was fully sampled.…”