IONIC INTERACTIONS OF DIVALENT METALS IN NATURAL-WATERS

The interactions of the ionic components of natural waters have been examined using the ion pairing and specific interaction models. The present paper reviews the recent developments made in using these models to estimate the activity and speciation of divalent metals in natural waters as a function of ionic strength. Since the ion pairing studies of most trace metals have been made in NaClO4, the activity coefficients of cations and anions have been determined in this media. The activity coefficients for free or uncomplexed ions have been estimated from the equation ln gamma(i) = Z(i)2f + B(i)0I + B(i)1f1 + C(i)I2 where f and f1 are functions of ionic strength; Z(i) is the charge on ion i; B(i)0, B(i)1 and C(i) are ionic Pitzer parameters derived from the values for Na+ and ClO4 salts using the mean salt approximation (gamma(K) = gamma(Cl)). The Pitzer coefficients were determined for a number of cations and anions. By combining the estimated activity coefficients for metals and anions with measured stability constants at a given ionic strength, it was possible to estimate the activity coefficient of various ion pairs and extrapolate the stability constants to infinite dilution. The results showed that ion pairs of the same charge type had similar values at a given ionic strength. This allowed us to estimate the effect of ionic strength on the stability constants for a number of divalent metals (Mg2+, Ca2+, Sr2+, Ba2+, Mn2+, Fe2+, Co2+, Pb2+, Ni2+, Zn2+) with a number of inorganic ligands (Cl-, OH-, HCO3-, CO32-, SO42-). The equations representing the ionic strength dependence of the stability constants are of the form ln K(MX)* = ln K(MX) + Z(MX)2f + B(MX)0I + B(MX)1f1 + C(MX)I2 where the values of Z(MX), B(MX)0, B(MX)1, and C(MX) are obtained from the Pitzer parameters for the free ions and the ion pairs (MX). These equations can be used to determine the speciation of divalent metal in natural waters over a wide range of ionic strength. Results of speciation calculations at seawater ionic strength show the same patterns as recorded in the literature, with some differences in the proportions of some metal ion pairs. The limitations of using this approach to account for the ionic interactions of metals and needed improvements to the model are discussed.