Many theoretical studies were devoted in the past to ion-specific effects trying to interpret a large body of experimental evidence, such as surface tension at air/water interfaces and force measurements between charged objects. Although several mechanisms were suggested to explain the results, such as dispersion forces and specific surface-ion interactions, we would like to suggest another source of ion-specificity originating from the local variations of the dielectric constant due to the presence of ions in the solution. We present a mean-field model to account for the heterogeneity of the dielectric constant caused by the ions. In particular, for ions that decrease the dielectric constant we find a depletion of ions from the vicinity of charged surfaces. For a two-plate system, the same effect leads to an increase of the pressure in between two surfaces. Our results suggest that the effect of ions on the local dielectric constant should be taken into account when interpreting experiments that address ion-specific effects.

In aqueous solutions, dissolved ions interact strongly with the surrounding water, thereby modifying the solution properties in an ion-specific manner. These ion-hydration interactions can be accounted for theoretically on a mean-field level by including phenomenological terms in the free energy that correspond to the most dominant ion-specific interactions. Minimizing this free energy leads to modified Poisson-Boltzmann equations with appropriate boundary conditions. Here, we review how this strategy has been used to predict some of the ways ion-specific effects can modify the forces acting within and between charged interfaces immersed in salt solutions.

The Poisson-Boltzmann mean-field description of ionic solutions has been successfully used in predicting charge distributions and interactions between charged macromolecules. While the electrostatic model of charged fluids, on which the Poisson-Boltzmann description rests, and its statistical mechanical consequences have been scrutinized in great detail, much less is understood about its probable shortcomings when dealing with various aspects of real physical, chemical and biological systems. These shortcomings are not only a consequence of the limitations of the mean-field approximation per se, but perhaps are primarily due to the fact that the purely Coulombic model Hamiltonian does not take into account various additional interactions that are not electrostatic in their origin. We explore several possible non-electrostatic contributions to the free energy of ions in confined aqueous solutions and investigate their ramifications and consequences on ionic profiles and interactions between charged surfaces and macromolecules.

The forces between charged macromolecules, usually given in terms of osmotic pressure, are highly affected by the intervening ionic solution. While in most theoretical studies the solution is treated as a homogeneous structureless dielectric medium, recent experimental studies concluded that, for a bathing solution composed of two solvents (binary mixture), the osmotic pressure between charged macromolecules is affected by the binary solvent composition. By adding local solvent composition terms to the free energy, we obtain a general expression for the osmotic pressure, in planar geometry and within the mean-field framework. The added effect is due to the permeability inhomogeneity and nonelectrostatic short-range interactions between the ions and solvents (preferential solvation). This effect is mostly pronounced at small distances and leads to a reduction in the osmotic pressure for macromolecular separations of the order 1-2 nm. Furthermore, it leads to a depletion of one of the two solvents from the charged macromolecules (modeled as planar interfaces). Lastly, by comparing the theoretical results with experimental ones, an explanation based on preferential solvation is offered for recent experiments on the osmotic pressure of DNA solutions.

Interactions in aqueous media between uniformly charged surfaces are well understood, but most real surfaces are heterogeneous and disordered. Here we show that two such heterogeneous surfaces covered with random charge domains experience a long-range attraction across water that is orders of magnitude stronger than van der Waals forces, even in the complete absence of any charge correlations between the opposing surfaces. We demonstrate that such strong attraction may arise generally, even for overall neutral surfaces, from the inherent interaction asymmetry between equally and between oppositely charged domains.