Charge transfer between isomer domains on n⁺-doped Si(111)-2 × 1: energetic stabilization

Abstract

Domains of different surface reconstruction -negatively-or positively-buckled isomershave been previously observed on highly n-doped Si(111)-2×1 surfaces by angle-resolved ultraviolet photoemission spectroscopy and scanning tunneling microscopy/spectroscopy. At low temperature, separate domains of the two isomer types are apparent in the data. It was argued in the prior work that the negative isomers have lower energy of their empty surface states than the positive isomers, providing a driving force for the formation of the negative isomers. In this work we show that the relative abundance of these two isomers shows considerable variation from sample to sample, and it is argued that the size of isomer domains is likely related to this variation. A model is introduced in which the electrostatic effects of charge transfer between the domains is computed, yielding total energy differences between the two types of isomers. It is found that transfer of electrons from domains of positive isomers to negative ones leads to an energetic stabilization of the negative isomers. The model predicts a dependence of the isomer populations on doping that is in agreement with most experimental results. Furthermore, it accounts, at least qualitatively, for the marked lineshape variation from sample to sample observed in photoemission spectra.

In-situ infrared study of silicon in KOH electrolyte: Surface hydrogenation and hydrogen penetration

“…13 [42]. Interestingly, while positive buckling is the normally stable configuration, it has been found that negative-buckling regions may coexist with positive buckling at high electron concentrations [43,44], such as those present at our strongly accumulated surface.…”

Confinement effects in π-bonded chains at group IV semiconductor (111) surfaces

“…It is worth recalling that Ge(111) reconstructed surfaces, similarly to Si(111), display the symmetry described by the Pandey model, with surface chains running along the [0 11] crystallographic direction [13]. Buckling distortion of the chains is believed to explain the gap between dangling-bond states for both Si(111)-2 × 1 and Ge(111)-2 × 1 surfaces, as established by several consistent experimental and theoretical studies [9,[14][15][16][17][18][19][20][21]. The π -bonded chain model is also commonly accepted for the geometry of the reconstructed C(111)-2 × 1 surface [22,23], though some details of the relaxation (such as the presence of interchain dimerization, buckling at the surface atoms and deeper layer distortions) are not clear yet.…”

Electronic disorder of P- and B-doped Si at the metal–insulator transition investigated by scanning tunnelling microscopy and electronic transport

“…as well as at RT [18,19]. It has been argued that the relative abundances of the isomers depend on the size of the isomer domains and on the doping level due to a charge transfer between the domains [20]. For the two types of buckling, the relative positions of the surface bands in the band gap of bulk Si were determined by STS with high lateral resolution [19].…”