Atomistic modelling of adsorption and segregation at inorganic solid interfaces

Abstract

Recent work using atomistic simulations on a number of different oxide and mineral interfaces is described. Static simulation techniques have been applied to gadolinium doped ceria grain boundaries and show that there is marked variation in oxygen vacancy and dopant segregation with depth and orientation of a number of tilt boundaries. These methods have also been used to model the carbonation of magnesium and calcium hydroxide surfaces and predict that the calcium hydroxide is more reactive, particularly {1 0 1} and {1 0 2} surfaces. Another important interface studied is the solid–water interface and we report a number of recent molecular dynamics simulations which show how the water ordering is affected by structure and composition. These include showing that calcium oxide–water interfaces show a range of water ordering including the appearance of ice-like structures, and on carbonation the water structure is totally disrupted. Simulations on the water ordering at silica–water interfaces predict that {11.0} quartz surfaces are more hydrophobic than {10.0} leading in turn to a preference for organic adsorption on {11.0}, while preliminary results for a siliceous porous surface suggest that the water structure influences the transport properties at the surface, particularly by extremes of pH.

Keywords:

• energy minimisation
• molecular dynamics
• LTA
• calcium hydroxide
• calcium oxide
• quartz
• ceria

Acknowledgements

We would like to thank EPSRC (Grant Number EP/D053897/2), AWE, and a NERC thematic programme for funding the various projects detailed in this work. In addition, acknowledgment is given to D.W. Price and M. S. D. Read (AWE), A. Benge for initial work on the carbonation of hydroxide surface and U. Anselmi-Tamburini for useful and informative discussions relating to the ceria grain boundaries. The eMinerals project, which is part of eScience, is also acknowledged along with the following members of the Nanogrowth consortium: M. Anderson, B. Slater and D.W. Lewis.

Computer resources used for the calculations above include those provided by the MOTT2 facility (EPSRC Grant GR/S84415/01) run by the STFC e-Science Centre.