A major review of electrolytes and the electric double layer was completed this year. Its scope was limited to the equilibrium regime, with the emphasis on the formally exact behaviour that could be deduced within the primitive model. The unifying theme was the close relation between the properties of the bulk electrolyte and those of isolated and interacting double layers. It was emphasised that there was a unique decay length for the bulk correlation functions, and that this was the same as for the double layer. Similarly, the ions of the bulk electrolyte could be recast with an effective charge so that the correlations had Debye-HŸckel form, just as there was an effective surface charge that made the Poisson-Boltzmann approximation for the double layer formally exact. The common properties of the electrolyte and the double layer are reflected in the similar theoretical techniques used to treat the two, and integral equation computational methods were covered in detail, with specific numerical comparisons for the effective surface charge and the double layer interaction. The review also briefly covered the influence of the solvent and the relevance for experiment of these newer notions of the double layer.
Conversion tables for the effective and actual surface charges were calculated for various ionic strengths. Since the effective surface charge would be obtained by fitting Poisson-Boltzmann approximation to measured double layer data, as is the common practice, these tables should be useful to experimentalists since they allow the actual amount of surface dissociation or ion binding to be determined. The practical implications of the saturation and reversal of the effective surface charge was also discussed, and it proved possible to relate these two unusual phenomena to changes in the structure of the double layer.
Foundations of Statistical Mechanics
The maximum entropy formulation of statistical mechanics is arguably the most satisfactory since it frees the discipline from the impediment of thermodynamics. Jaynes' original presentation makes it clear that one needs a density of states for an invariant measure of entropy, yet it seems that this observation has not been taken as seriously as it might have been. In the isobaric ensemble, (fluctuating volume, constant pressure), it becomes clear that one needs a density of volume states, if only to make the partition function dimensionless. The correct density is the a priori probability for a scale parameter, which in this case is the reciprocal of the volume. We have shown that this restores the consistency between the canonical and isobaric ensembles for the equation of state for an ideal gas and for hard-rods. The correction, which is negligable in the thermodynamic limit, is important for finite-sized simulations, for accurate equations of state, and for systems near criticality or the spinodal line.
Together with Dan Gordon, we have developed a computer model of the interaction of real atoms with evanescent light. The major innovation is the incorporation of real atomic level structures. We have been able to quantitatively model observations of dopplerons (ANU experiment) and of atom diffraction (Bonn experiment). This work is reported in D. Gordon's Honours thesis "Multi-level evanescent wave atom optics".
We have investigated the effect of environmental decoherence on quantum reflection of wave packets by potential wells. This is relevant to the reflection of wave packets by potential at surfaces. This work is in progress and reported by Glenn Moy.
In collaboration with Kylie Catchpole, Neil Manson and other members of the solid state spectroscopy group in the Laser Physics Center we have studied multiphoton transitions between atomic dressed states. Excellent agreement between numerical solutions of the Bloch equations and experimental absorption spectra was obtained. This work is reported in the Honours thesis of K. Catchpole, "Probe absorption spectra of a three-level system with two strong fields".