Structure and Bonding of Alkanethiols on Cu(111) and Cu(100)

Abstract

The local structure of the sulfur atom of methanethiolate and ethanethiolate on the Cu(111) and Cu(100) surfaces was investigated from first principles employing the periodic supercell approach in the framework of density functional theory. On the 111 surface, we investigated the (square root 3 x square root 3)R30 degrees and (2 x 2) structures, whereas on the 100 surface, we investigated the p(2 x 2) and c(2 x 2) structures. The landscape of the potential energy surface on each metal surface presents distinctive features that explain the local adsorption structure of thiolates found experimentally. On the Cu(111) surface, the energy difference between the hollow and bridge sites is only 3 kcal/mol, and consequently, adsorption sites ranging from the hollow to the bridge site were observed for increasing surface coverages. On the Cu(100) surface, there is a large energy difference of 12 kcal/mol between the hollow and bridge sites, and therefore, only the 4-fold coordination was observed. The high stabilization of thiolates on the hollow site of Cu(100) may be the driving force for the pseudosquare reconstruction observed experimentally on Cu(111). Density of states analysis and density difference plots were employed to characterize the bonding on different surface sites. Upon interaction with the metal d bands, the pi* orbital of methanethiolate splits into several peaks. The two most prominent peaks are located on either edge of the metal d band. They correspond to bonding and antibonding S-Cu interactions. In the case of ethanethiolate, all the back-bonds are affected by the surface bonding, leading to alternating regions of depletion and accumulation of charge in the successive bonds.

Adsorbate-induced surface stress, surface strain and surface reconstruction: CH3S on Cu(100) and Cu(111)

“…In both phases the S headgroup atom lies in an off-hollow site, displaced towards the bridging site, the C-S bond being tilted relative to the surface normal by 33° in the (3x3)R30° phase and by 47° in the (2x2) phase. These conclusions are in excellent agreement with the results of Feral et al [16] who investigated only these unreconstructed models. There is one report of the observation of a Cu(111)(3x3)R30°-CH 3 S phase at low temperatures (110-140 K) [6], but at room temperature this transforms to the 4 3 1 3  L N M O Q P phase, suggesting that the (3x3)R30° phase is metastable and reconstructs with the aid of thermal activation.…”

“…This qualitative effect is consistent with the experimental results of the MEIS study [15], although the experimental value of 0.12±0.04 Å is significantly larger. The earlier DFT study of Ferral et al [16] led to the same value as in the present calculations.…”

“…Ferral et al [16] reported a DFT study of methylthiolate and ethylthiolate on both Cu(100) and Cu(111) in 2006, but considered only possible adsorption structures that involved no substrate reconstruction; they did, however, report some relevant results on the structure (and local strain) in the Cu(100)(2x2)-CH 3 S phase. More recently, during the rather extended period of our own investigation, there have been two independent reports of DFT calculations relating to the methylthiolate-induced reconstruction of Cu(111).…”

Molecular shape and the energetics of chemisorption: From simple to complex energy landscapes

“…Second, the model does not account for different orientations of the molecule relative to the surface normal. Density functional calculations of CH 3 S on metal (111) surfaces have found that, although the absolute binding energy is sensitive to the angle between the C-S bond and the metal surface [11], the energy barrier to translational motion of a molecule bound to the surface varies only slightly with this angle [12]. Third, although a more realistic molecular model would have included bond flexibility, only rigid molecules (i.e., molecules in which the positions of the interaction sites with respect to one another were kept constant) were considered in this paper.…”

“…Fourth, only the binding of a single molecule on the surface is considered. Thus, interactions between adsorbate molecules, which have been shown to have a strong effect on the binding energies of alkanethiols as a function of coverage on metal surfaces [3,11], have been ignored. The results of this paper are, therefore, relevant to the low-coverage regime.…”