We have determined the physical properties of Na+ and Mg++ in water.

J. White, F. Lightstone, E. Schwegler, G. Galli, F. Gygi, J. Chem. Phys. (2000).

Bonding and solvation dynamics of Na+ and Mg++ are in agreement with experiment and show important differences with respect to classical models.

The solvation of ions is a fundamental problem encountered in a wide range of biological and chemical systems. In particular, the manner in which water solvates alkali metal cations is relevant to problems such as the mechanism of enzymatic catalysis, and the structural stability of DNA and RNA. In order to investigate the hydration structure of ions, a variety of experimental techniques have been applied. For example, x-ray and neutron diffraction methods have proven to be particularly valuable for determining the static structure of waters solvating a given ion. However, experimental measurements alone may in some cases give an ambiguous or incomplete description of ionic solvation. Problems can arise such as a lack of suitable isotopes, or difficulties in separating the various atomic correlations from the experimental data. In recent years, molecular dynamics simulations have emerged as a successful complement to experimental measurement, and have lead to a greater understanding of the solvation process. We have carried out first principles molecular dynamics simulations on aqueous solutions of sodium and magnesium cations at approximate infinite dilution. The calculations include a single ion and 53 water molecules in a cubic box at ambient temperature and pressure. By applying periodic boundary conditions, the bulk properties of the liquid are effectively modeled.

Figure 1. The radial distribution function gNaO determined by simulation (solid line) and experiment (long dashed line).

In Fig. 1, the radial distribution functions gNaO(r) is shown along with the gNaO(r) determined from an x-ray diffraction measurement of a solution of NaNO3 in water. The location of the first and second peaks in gNaO(r) is in good agreement between simulation and experiment, with only a slight shift towards larger separations in the simulated gNaO(r). Also shown in Fig. 1 is the running integration number, which indicate the number of water molecules in the first solvation shell. We find that there is on average 5.2 water molecules in the first solvation shell around the sodium ion.

Figure 2. The sodium-oxygen separation in angstroms as a function of time. The four solid lines represent waters that always stay in the first solvation shell, and the dashed lines are waters that make exchanges from the second to the first shell during the simulation.

The height of the first minimum of gNaO in Fig. 1 indicates that the first solvation shell around sodium is not completely separated from the second one, and that a significant amount of water exchanges between the shells takes place. The exchange of waters between the solvation shells of sodium can be monitored by tracking the sodium-oxygen separation, as in Fig. 2, for those waters that come within 3.2 angstroms of the ion. Although four waters always stay within the first solvation shell without leaving, there are two waters that make exchanges between the two shells. Within the limited amount of simulation that has been collected, the water exchanges proceed by an associative interchange mechanism.

UCRL-MI-
Date last modified: 9/27/00
Contact Miriam Rinnert for website information