Ions in solids and liquids and at interfaces (ISLI2000). Report of meeting.

5-6th January 2000
Florence Boot Hall, University of Nottingham

List of Talks

• Jim Rustad (Pacific North Western National Laboratory)
  Molecular simulation of chemical processes at mineral-water interfaces

• Philip Lindan (CLRC Daresbury)
  Understanding oxide-aqueous solution interfaces from first principles

• Michiel Sprik (Cambridge)
  First principles determination of pK, electronic chemical potential and hardness in aqueous solution

• Don Jenkins (Warwick)
  New developments in the estimation of lattice potential energies: implications for chemistry

• Patrick Fowler (Exeter)
  Polarisabilities of ions in solids

• Mark Wilson (Oxford)
  Networks, chains, rings and holes: intermediate-range order in ionic systems

• Steve Parker (Bath)
  Atomistic modelling of the interaction of water with mineral interfaces

• John Harding (UCL)
  Modelling interfaces at a variety of lengthscales

Report of meeting

Four of the speakers (Rustad, Lindan, Sprik, Parker) addressed the problem of the aqueous interface. A basic issue emerged that was discussed but not solved: how can we get a decent model of dissociating water? When modelling water, one possibility is to turn to quantum calculations, and Lindan discussed progress on modelling the aqueous interface for TiO₂. Sprik discussed the importance of the electronic structure of the solvent and the different behaviour of 'hard' and 'soft' ions. ('Hard' ions are simple Coulombic systems; 'soft' ions have electronic states that are close to the electronic states of the solvent). However, large-scale simulations are still (as yet) going to be classical and so the problem of constructing classical potentials remains. Parker discussed what could be done with classical models on the growth and dissolution of minerals.

The problem of building a usable and useful ionic model was also addressed by Fowler and Wilson. Fowler discussed current progress in calculating polarisabilities of ions in solids. The problems encountered by density functional calculations in calculating accurate polarisabilities were aired. Wilson gave a number of examples to show how the correct inclusion of polarisability effects could enable ionic models to reproduce effects traditionally considered to be 'covalent' (a brief argument was held on what, if anything, is meant by 'covalent'). Wilson showed that ionic models can predict bond bending, chains and networks as a result of polarisation screening of cation-cation interactions.

Harding gave a brief survey of mesoscale modelling at interfaces. This was heard with the kind of patience shown by an audience that knows that lunch will follow when the speaker finishes.