THEORETICALLY MODELING THE WATER BILAYER ON THE AL(111) SURFACE USINGCLUSTER CALCULATIONS

Citation
Md. Calvin et al., THEORETICALLY MODELING THE WATER BILAYER ON THE AL(111) SURFACE USINGCLUSTER CALCULATIONS, Surface science, 345(1-2), 1996, pp. 161-172
Citations number
22
Categorie Soggetti
Chemistry Physical
Journal title
ISSN journal
00396028
Volume
345
Issue
1-2
Year of publication
1996
Pages
161 - 172
Database
ISI
SICI code
0039-6028(1996)345:1-2<161:TMTWBO>2.0.ZU;2-#
Abstract
Cluster calculations involving two water molecules are used to theoret ically investigate the formation of the water bilayer on the Al(111) s urface. In previous calculations we found a single water to molecularl y adsorb at an on-top site vertically above an Al atom. We also comput ed vibrational frequencies which were consistent with data obtained at low water coverage by electron energy loss spectra (EELS). The EELS e xperiment initially has a peak at 3720 cm(-1) which becomes augmented and eventually dominated by a second peak at 3450 cm(-1) as the water coverage on the Al(111) surface is increased. In the present calculati ons we optimized the geometry of two water molecules on Al-10 and Al-1 5 clusters. The resulting structures appear to be prototypes for the f ormation of a water bilayer on the Al(111) surface: one of the water m olecules again adsorbs on-top of an At atom, and then the second water hydrogen bonds to the first water molecule, at a height corresponding to the second layer of a water bilayer, vertically above an Al atom n eighboring the Al atom bonded to the first layer water. The computed v ibrational frequencies for the chemisorbed water dimer have a number o f features in common with the EELS results obtained at higher water co verage. Especially striking is that we compute a similar to 300 cm(-1) blue shift in the H-bonded OH stretching mode which matches the split ting observed for the high frequency modes observed in EELS at higher water coverage. These dimer calculations add further support to our pr evious suggestions that the 3720 cm(-1) vibration is due to a symmetri c OH stretch in non H-bonded water and that water only molecularly ads orbs on Al(111) at low temperatures. Again we allow for adsorbate indu ced surface relaxation effects in;he cluster calculations.