Specific Ion Effects

The following sections are included:

• Contents

• Foreword

• Preface

• Acknowledgements

• List of Contributors

In the present chapter, selected examples of specific ion effects are discussed. We start with experimental evidences in simple systems and end with ion effects at complex biological surfaces. Some general rules are extracted from these examples. The chapter also contains a short overview of current theories and simulations aiming at a more fundamental understanding of ion specificity. This chapter is intended to be a general introduction to the field and a basis for the following chapters that deal with more specific aspects.

The Langmuir monolayers and bilayer stacks of zwitterionic phospholipids have been recently used as model systems to understand specific anionic effects. Simple thermodynamic experiments coupled to theoretical considerations and advanced spectroscopic methods have provided considerable insights. Inorganic monovalent anions were found to affect only the disorganized liquid-like phases of lipid monolayers. The perturbation of the surface pressure–area isotherm was treated as a problem of electrostatic ionic adsorption using a range of models, and the separate effects of sodium and various anions were assessed. A chemical potential of transfer of anions to the interface was obtained from the fits, and was found to correlate well with a function of the ionic size of the anions, suggesting that ionic specificity is largely a matter of size, and reflects the difference between ion–water and water–water interactions.

Osmotic stress experiments on lipid bilayers in the presence of sodium salts yield a perpendicular and a lateral equation of state (EOS). The attempt to fit these EOS using standard theory failed in the presence of large (0.5M) concentrations of chaotropic anions, proving that the adsorption at the surface of the bilayers modifies their local structure considerably. Information about the changes at the interface propagates to the bulk, affecting the perpendicular EOS. Overall, these experiments revealed a strong need for better fundamental understanding of the effect of salts on the structure of lipid bilayers. In particular, more work is needed to understand the specific salt effect on the lipid headgroup region.

During the last decades, the application of electrolytes in many industrial and research fields has increasingly gained in importance. The modeling of aqueous electrolyte systems is crucial for the design and optimization of processes in biochemical engineering. It can provide component and solution properties which are substantial for the design of separation units. Liquid densities (pvT data), solubilities as well as vapour pressures (VLE data) are directly obtained by an equation of state (EOS). Water activity coefficients (WAC) and mean ionic activity coefficients (MIAC) of electrolyte solutions can be calculated by either an EOS or by an excess Gibbs Energy model (G^E model).

Here, we describe the application and typical modelling results for a G^E model (MSA-NRTL) as well as for an EOS (ePC-SAFT). In addition to strong electrolytes which are almost fully dissociated, we also consider some weak electrolytes (acids like HF or ion-paired electrolytes) that do only partially dissociate in aqueous solution. Here, ion pairing is accounted for by an association/dissociation equilibrium between the ion pair and the respective free ions in solution.

Aqueous ions at interfaces play a substantial role in numerous processes in our environment. The diversity of ion-specific phenomena has been extensively investigated both theoretically and experimentally, and has expanded our knowledge about ion-specific effects. In particular, the in situ application of novel experimental tools such as non-linear optical spectroscopic techniques has led to a more detailed picture of interfacial ion profiles. Among these, second-harmonic-generation (SHG) spectroscopy and sum-frequency-generation (SFG) spectroscopy have emerged in recent years as suitable methods for characterizing aqueous interfaces. The selection rules for SHG and SFG dictate that a signal can only be generated in a non-centrosymmetric environment, effectively resulting in the suppression of isotropic bulk signal. This feature makes non-linear spectroscopy inherently surface-specific with submonolayer resolution, allowing for the tracking of subtle modifications in the interfacial layer. Being all optical techniques, SHG and SFG are also non-invasive and applicable under aqueous conditions, making them suitable for in situ characterization of ions at the air-water interface. In this chapter, an overview is given about linear and especially non-linear optical techniques used to investigate ions at charged and uncharged air-water interfaces. The theoretical background of non-linear optics is briefly reviewed and some recent experimental results are critically examined.

There is a broad range of phenomena in biological, environmental, and physical sciences where ions of the same valency have a dramatically different effect. Since the seminal work of Hofmeister in 1888 on protein stability, this has been illustrated in many examples ranging from enzymatic activity and amyloidosis to humics stability and halide heterogeneous chemistry. These ‘ion specific’ effects are not completely understood yet and unraveling the short-range solvent-mediated couplings dominant at interfaces might provide a clue towards understanding them. This points out the need for quantitative and direct measurements on aqueous salt solutions to access the surface composition at the relevant nm or sub-nm lengthscale. In this chapter, we discuss how this can be achieved using X-rays, mainly X-ray reflectivity, anomalous X-ray reflectivity, grazing incidence X-ray fluorescence and X-ray standing waves.

The presence of ions in water constitutes a rich and varied environment in which many natural processes occur. This review provides results of structural studies of aqueous solutions derived from state-of-the-art neutron scattering methods. The enhanced resolution provided by methods such as Neutron Diffraction and Isotopic Substitution (NDIS) have given scientists new insights into the contrasting hydration structures of a variety of ions and small solute molecules, and crucially how these structures might affect the general properties of solutions. The discussion points out common features of ionic hydration within particular series such as the alkalis, halides and transition metals, and also indicates where significant differences in hydration structure appear. For polyatomic ions there is a clear need for additional information to interpret the neutron scattering results. Accordingly we show how computer simulation can assist in the particular case of hydration of complex ions such as guanadinium (Gdm⁺) or thiocyanate (SCN⁻), both of which have important consequences for protein denaturation in aqueous solution.

Ion specificity is important in all soft matter systems at high electrolyte concentrations. Specific ion effects that influence many biological processes fall within the Hofmeister paradigm, whereby the strength of action of the anions and cations follow a well-defined order, independent of the co-ion. In contrast, the ion specificity seen when salts inhibit bubble coalescence depends on the combination of ions present. Single electrolytes at 0.1M may inhibit bubble coalescence or have no effect, as determined by simple ion combining rules and empirical assignments (α or β) of the cation and the anion. The coalescence of mixed electrolyte systems can also be predicted by an extension of these rules. Bubble coalescence in some non-aqueous solvents can also show ion-specificity that depends upon ion combination in a way analogous to the combining rules observed in water — which suggests that this is a general property of the air–solution interface. Here, we summarise evidence that the contrasting specific ion effects observed in bubble coalescence and in other soft matter systems have the same origin: Ion specificity arises from differences in the affinity of ions for interfaces. This leads to the conclusion that combining rules seen in bubble coalescence inhibition arise from the dynamic nature of the air–solution interface.

We investigated specific anion binding to basic amino acid residues as well as to a range of patchy protein models. This microscopic information was subsequently used to probe protein–protein interactions for aqueous lysozyme solutions. Using computer simulations to study both atomistic and coarse grained protein molecules, it is shown that the ion–protein interaction mechanism as well as magnitude is largely controlled by the nature of the interfacial amino acid residues. Small anions interact with charged side-chains via ionpairing while larger, poorly hydrated anions are attracted to nonpolar residues due to a number of solvent-assisted mechanisms. Taking into account ion and surface specificity in a mesoscopic model for protein–protein interactions, we investigated the association of the protein lysozyme in aqueous solutions of sodium iodide and sodium chloride. As observed experimentally, it is found that ‘salting out’ of lysozyme follows the reverse Hofmeister series for pH below the iso-electric point and the direct series for pH above.

Molecular dynamics simulations based on classical force fields have become the standard tool for the modelling of ion specificity in bulk and at interfaces, but the choice of the non-bonding force parameters, the so-called force fields, is a subtle issue. We discuss how thermodynamic solvation properties in the infinite-dilute limit can be used to construct optimised non-polarisable force-fields for alkali metal cations and halide anions. Next, ion specificity in bulk and at interfaces is studied. In bulk, we obtain osmotic coefficients directly from molecular simulations and discuss the relation between ion–ion correlations and the resultant osmotic thermodynamics. At the air-water interface we determine the single-ion potential-of-mean-force in the infinite dilution limit. In agreement with previous polarisable force-field simulations, the larger halide ions are less repelled from the interface and iodide is even enhanced. We conclude that the surface activity of iodide is due to the ion hydrophobicity. Combining the simulation results with a generalized Poisson–Boltzmann approach, the interfacial tension increment at one molar salt concentration is predicted in quantitative agreement with experiments. The general trend, in agreement with the Hofmeister series, is: the smaller the ion, the larger the interfacial tension increment, indicative of stronger ion repulsion from the interface. On solid hydrophobic surfaces, the behaviour is similar to the air–water interface. On solid hydrophilic surfaces, on the other hand, the Hofmeister series is reversed and small ions show enhanced binding to polar surface groups.

The theoretical approach based on the HNC integral equation is described in the context of ionic specificity. Two levels of description of the water medium are considered.Within the Primitive Model (continuous solvent), ionic specificity is introduced via effective, solvent-averaged, dispersion forces. The agreement with experimental data in bulk or at air–water interfaces is only partial and illustrates the limits of that approach. Within the Born–Oppenheimer model, the molecular HNC equation is solved with an explicit description of the solvent molecules (SPC water). Ionic and solvent profiles in bulk and at interfaces are enriched by short-range oscillated structures. The ionic polarisability is introduced via the self-consistent mean-field theory, the polarisable ions carrying an effective, fixed dipole moment. The study of the air–water interface reveals the limits of the conventional HNC approach and the needs for improved integral equations.

In this chapter, a short history is given about the introduction of non-electrostatic interactions in the Poisson–Boltzmann equation, starting from simple ionic dispersion forces up to very recent approaches, in which both water profile and ion-surface potentials inferred from molecular dynamics simulations are used.

In its origins, the Hofmeister series is simply an empirical ranking of the extent to which added electrolytes modify protein solubility. The very wide range of systems and phenomena to which this concept has now been applied is amply demonstrated by the chapters in this volume. Almost all of the examples involve the adsorption of ions to surfaces. For those that do not, there is the adsorption of ions to each other (‘complex’ formation). Most of the applications also involve some reorganisation of the system, whether it is a phase change, a change in the molecular configuration (usually within a surface layer) or in the composition of the species present (e.g. a change in the number of water molecules bound to a solute). Occasionally there is a change in the chemical nature of the species involved. The obvious example is that of proteins in aqueous solution, where each different ionic structure is a different chemical. The concentrations of all of these change as the pH is altered. Hence the chemical species above and below pI are different. A less obvious example is the lipid lecithin. Here the alkyl chains can adopt hundreds of thousands of different conformations — each one a different compound. Fortunately, as far as ion-head group interactions are concerned, they are all broadly similar. However, the head group too can adopt different conformations, in this case the number being about ten. Some of these are likely to have very different interactions with ions. Indeed, ion adsorption may well change their relative populations. Important contributions to the understanding of ion-specific effects on phospholipids can be found in Chap. 2.

The following sections are included:

• Index