TY - JOUR
T1 - Theoretical study of the Si(100) surface reconstruction
AU - Ramstad, A.
AU - Brocks, G.
AU - Kelly, P. J.
PY - 1995/1/1
Y1 - 1995/1/1
N2 - We present the results of a systematic study of the reconstruction of the Si(100) surface based upon total energies calculated within the framework of the local-density approximation. We focus on the extent to which total energy differences may be calculated reliably by examining these differences for the ideal surface and four proposed reconstructions: p(2×1) symmetric, p(2×1) asymmetric, p(2×2), and c(4×2). The calculations were performed using norm-conserving pseudopotentials and a plane-wave basis. The convergence of the total energy differences was assessed by varying the energy cutoff used to truncate the plane-wave basis and the number of sampling points used to perform Brillouin zone (BZ) integrals over a large range. The effect of optimizing atomic geometries as a function of the energy cutoff and density of BZ sampling points was determined. With the exception of the p(2×2) and c(4×2) reconstructions, whose energies only differ by 3 meV per dimer, we are able to unambiguously determine the energy ordering of the five systems studied. Disagreements between previous calculations can be largely understood in terms of the different energy cutoffs and BZ samplings used. The electronic structures of the different reconstructions are calculated and compared.
AB - We present the results of a systematic study of the reconstruction of the Si(100) surface based upon total energies calculated within the framework of the local-density approximation. We focus on the extent to which total energy differences may be calculated reliably by examining these differences for the ideal surface and four proposed reconstructions: p(2×1) symmetric, p(2×1) asymmetric, p(2×2), and c(4×2). The calculations were performed using norm-conserving pseudopotentials and a plane-wave basis. The convergence of the total energy differences was assessed by varying the energy cutoff used to truncate the plane-wave basis and the number of sampling points used to perform Brillouin zone (BZ) integrals over a large range. The effect of optimizing atomic geometries as a function of the energy cutoff and density of BZ sampling points was determined. With the exception of the p(2×2) and c(4×2) reconstructions, whose energies only differ by 3 meV per dimer, we are able to unambiguously determine the energy ordering of the five systems studied. Disagreements between previous calculations can be largely understood in terms of the different energy cutoffs and BZ samplings used. The electronic structures of the different reconstructions are calculated and compared.
UR - http://www.scopus.com/inward/record.url?scp=0000282851&partnerID=8YFLogxK
U2 - 10.1103/PhysRevB.51.14504
DO - 10.1103/PhysRevB.51.14504
M3 - Article
AN - SCOPUS:0000282851
VL - 51
SP - 14504
EP - 14523
JO - Physical review B: Covering condensed matter and materials physics
JF - Physical review B: Covering condensed matter and materials physics
SN - 2469-9950
IS - 20
ER -