Ionic van der Waals Interactions with Surfaces and with Water Molecules

At present, two parallel and complementary lines of work demonstrate that both surface effects and bulk effects are important for the understanding of Hofmeister effects. We will first briefly discuss surface-related effects, focusing on the van der Waals interaction between ions and interface. Boström et al. demonstrated, e.g. that this is one important reason why the charge of lysozyme (hen-egg-white protein) at pH = 4.5 is higher in 0.1 M NaSCN than in 0.1 M NaCl.

The negatively charged thiocyanate anions (SCN^-) are more attracted toward the protein, by larger van der Waals forces, than are the chloride ions. A higher local anion concentration with SCN^- ions near the protein surface gives rise to a larger local concentration of hydronium ions (H₃O+) at the protein surface, which, in turn, leads to more hydrogen ions being bound to the protein charge groups (i.e. a larger protein charge) in a thiocyanate (or iodide) salt compared with a chloride salt.

Remarkably, as far as the pH of the solution or the charge of hen-egg-white protein is concerned, one can often to a good approximation replace a certain concentration of NaCl with a smaller concentration of NaSCN and find the same final result. A similar result has been observed in experiments on enzymatic activity by Bauduin et al. and in pH measurements in protein (cytochrome c) solutions by Baglioni et al.

This kind of surface-related effects has also offered a partial explanation for the experimental observation by Bonnete et al. that the interaction between two lysozyme proteins in salt solutions follows a Hofmeister series (the interaction is, e.g. more attractive in NaNO₃ solutions than in NaCl solutions).

The other line of work has focused on interactions in the bulk solution between ions and between ion and water molecules. It is important to stress that van der Waals forces are only one consequence of the ion polarizability. Other consequences are dipole-induced dipole and induced dipole-induced dipole forces between ions and between ions and water molecules. These forces have been in focus in molecular simulations.

It has been repeatedly demonstrated that the static polarizability of ions must be included in computer molecular simulations to get reasonable results for the solvation energy of ions in water. This, however, is not the entire story. To get correct results for the activity coefficients and solvation energies of ions in water, strong evidence exists that one has to include the entire dynamic ionic polarizability.

In several important contributions, attempts have been made to include forces that originate from the static polarizability correctly in computer simulations that include water molecules. These ion-ion and ion-water interactions are also important for the air-water interface. The air-water interface, and the surface tension of salt solutions, is more complicated than one might initially have guessed and not yet fully understood.

Promising simulations seem to give the right trend in Hofmeister sequence for the surface tension change with added salt. But these simulations give results that differ greatly from experimental surface tension changes with added salt. This is also true for pure water where simulations without fitting parameters give incorrect results for the surface tension and freezing temperature. One reason for these problems must be that simulations so far have not included many body dispersion forces properly. But simulations together with experiments are still useful tools for further understanding.

As demonstrated by Jungwirth and Tobias, some highly polarizable anions, such as bromide, may, contrary to conventional wisdom (which is based on electrostatic theories), go to the air-water interface. It has also been demonstrated that one important effect is that the water molecules near an air-water interface are not isotropically ordered, which effectively gives an anisotropic film near the interface that strongly influences the interaction between ions and the air-water interface.

These interactions (ion-ion, ion-water molecule, ion interface) influence each other in a self-consistent manner, which makes this a complicated but solvable problem. In two landmark papers, Schwierz et al. demonstrated how Hofmeister series can be reversed depending on the polarity, hydrophobicity or the charge of surfaces.

Parsons and Ninham later introduced the use of quantum chemical calculations to determine the dynamic ion polarisability spectra required to evaluate van der Waals interactions. This enabled the gap between theory and experiment to start to close, in the first instance demonstrating that nonelectrostatic physisorption due to ionic van der Waals forces charge reversal could lead to charge reversal, where the electrostatic surface potential has the opposite sign to the surface charge, at mica and protein surfaces, matching and predicting experiment. Depending on the material of the surface, ions may form a reverse Hofmeister series.

Just as Hofmeister effects are observed in protein interactions, so too they play a role in the stability of other kinds of nanoparticles such as gold nanoparticles. Synthetic nanoparticles are interesting because they enable the optical spectrum to be tuned through surface plasmon resonance by controlling the size of the particle. Given the role that the optical spectrum plays in determining van der Waals interactions, this suggests that nanoparticles may be engineered with selectivity to adsorb specific ions. Nanohematite, for example has been used in this way to adsorb lead ions.