Structure, composition and stability of water and mixed OH/Water films at the single crystal interface : a surface science study

• Main Author: Clay, Christopher M.
• Published: University of Liverpool 2006
• Subjects: 541
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.486096

Pure water and mixed OH/H20 overlayers have been studied using a combination of TPD, TPRS, LEED and RAIRS and compared to DFT and kinetic Monte Carlo simulations carried out by collaborations with Georg Held, Gustav Garlberg and George Darling. On Pt(1l1) a mixed OH/H20 overlayer forms as an intermediate in the water oxygen reaction, with a (V3 x V3)R30° periodicity. T~e stability of the overlayer is dependent on the degree of hydrogen bonding, and reaches a maximum at an OH/H20 ratio of one. The optimal structure is found to be a mixed (OH+H20) phase containing a hexagonal lattice of alternating OH and H20 with a (3x3) superstructure, caused by ordering of the hydrogen bonds. The mixed overlayer can accommodate a range of H20/OH compositions, but becomes less stable as the water content is reduced. A LEED I-V analysis indicates that the (3x3)-3(OH+H20) overlayer is coplanar and the periodicity results from a lateral distortion of the 0 atoms away from the atop adsorption site. TPD of water from a (3x3)-3(OH+H20) overlayer on Pt(111) shows pseudo zero order desorption kinetics. However, leading edge and Polanyi-Wigner analysis of various mixed OH/water overlayers find an activation barrier to desorption that depends on the composition, coverage and heating rate of the film, indicating complex desorption kinetics. To understand these kinetics a new model based on direct H20 desorption, proton transfer, and OH recombination is proposed, and is compared and contrasted with kinetic Monte Carlo simulations. Water and oxygen also react on Pd(111) at low temperatures to form a mixed OH/H20 layer with a (V3 x V3)R30° registry. As on Pt(1l1), the most stable structure is formed when the OH/H20 ratio reaches unity. The (OH+H20) phase cannot be formed by O/H reaction and is distinct from the (V3 x V3)R30° structure formed by O/H coadsorption below 200 K. Mixed OH/water structures do not react with coadsorbed H below 190 K on Pd(l11), preventing this phase catalysing the low temperature H2/02 reaction. Adsorption of H20 and D20 on Ru(OOOl) shows a unique isotope effect. Above 150 K H20 dissociates to form a mixed (OH/H20/H) overlayer, whereas D20 desorbs intact. Dissociation of H20 to form the mixed overlayer competes with H20 desorption and this branching can be enhanced or suppressed by preadsorbing 0 or H. D20, and below 150 K H20, adsorbs intact on Ru(OOOl) with a diffuse (V3 x y'3)R30° tEED pattern visible at low coverage, which sharpens as the coverage reaches 0.6 ML, becoming diffuse again as the coverage exceeds 0.67 ML. Water initially adsorbs flat at Ru(OOOl), forming small clusters, which buckle as the coverage increases to accommodate an extended, hydrogen bonded overlayer at saturation. Only as the overlayer starts to complete does the free OH/OD stretch appear in RAIRS, indicating that water is embedded either flat or in a H-down geometry, up to 0.67 ML. DFT calculations find that water favours formation of flat lying small hexamer clusters at low coverage, adsorbed atop the metal. At 0.67 ML, calculations find stable water structures containing flat lying chains, linked by upright waters bonded H-down, with a hexagonal backbone of oxygen atoms. This structure is found to be more than 20% more stable than the conventional H-up or H-down bilayer and is expected to wet Ru(OOOl)