Interactions and Transport of Charged Species in Bulk and at Interfaces
The Asymmetric Restricted Primitive Model of Molten Salts and Ionic LiquidsJan Forsman
Coauthor(s) : Honduo Lu  Bin Li  Clifford E. Woodward  Sture Nordholm 
 Theoretical Chemistry, Lund University, 221 00 Lund, Sweden  PEMS, ADFA, UNSW, Canberra, ACT 2600, Australia  Department of Chemistry, Gothenburg University, 41296 Gothenburg, Sweden
Recall the restricted primitive model (RPM), where ions are modelled as hard spheres, all of diameter d, carrying a charge at their centres. In the simplest charge-symmetric case, this charge has the same absolute value. The presence of any dielectric medium (solvent) is simply represented by its relative permittivity, ε. The RPM has been very useful for statistical mechanical analyses of electrolyte solutions, molten salts, and plasmas. Here we study an extension to the RPM, namely the asymmetric restricted primitive model (ARPM), which allows a new mechanism to enter, controlled by a single parameter b. This parameter measures a displacement of the charge from the centre of the hard sphere. Such a displacement, when uniformly applied to all particles, leaves the model very simple. Nevertheless, b introduces a new mechanism that we believe is of crucial importance to many systems, such as ionic liquids. A similar system was studied by Ganzenmuller and Camp , although their work had a different focus. The presence of a non-vanishing displacement b means that the ions now have an internal structure, generating orientational correlations, with an increased propensity (for a sufficiently large value of b) to form effective dimers (ion pairs) with oppositely charged ions. This is expected to significantly alter the properties of the fluid, including:
lower freezing temperature
reduced efficiency of ionic screening, i.e. a longer screening length
reduced conductivity (already confirmed by Ganzenmuller and Camp)
dielectric response, brought about by orientational correlations of dipolar ion pairs
The ARPM is likely to have properties that are reminiscent to a mixture of an RPM fluid and a dipolar fluid, where the dipoles are generated by ion pairs. We believe that the puzzle of room temperature ionic liquid (RTIL) melting points to a large extent can be rationalized in terms of dimerization, as described by the ARPM. There are recent surface force measurements by Gebbie et al. [2,3], that suggest an extremely high degree of ion pairing (more than 99 %) in a typical RTIL, as indicated by measured very long-ranged double-layer interactions.
We have performed simulation studies that confirm the conjecture of a much lower melting temperature of ARPM systems, relative to the corresponding RPM. Simulation works focusing on the remaining conjectures are underway. An interesting structural observation is that radial distribution functions tend to vary with the displacement b in a non-monotonic fashion. This can be rationalized in terms of rotational entropy and orientational correlations.
 G. C. Ganzenmuller and P.J. Camp, Cond. Mat. Phys. 14, 33602 (2012))
 Gebbie A. M., Valtiner M., Banquy X., et al., PNAS 110, 9674 (2013).
 Gebbie A. M., Bobbs A.H., Valtiner,M, et al., PNAS 112, 7432 (2015).