Sodium-Calcium Exchange with Anion Exclusion and Weathering Corrections

In order to accurately model soil reclamation with concentrated electrolytes such as CaCl2 or sea water, it must be determined if there is an effect of salinity on the Na-Ca exchange selectivity. A new procedure for determining Na-Ca selectivity in calcareous and gypsiferous soils was used to study the effects of ionic strength and mineralogy on Na-Ca exchange. Four specimen clay minerals and three whole soils were equilibrated with solutions ranging in concentration from 10 to 1000 mmolc L' and at Na adsorption ratios from 1 to SO (mmol L~'). Exchangeable-cation values were corrected for calcite and gypsum dissolution as well as anion exclusion. The selectivity data were compared with other reported values for Na-Ca exchange in order to determine if there were any trends in selectivity with mineralogy or surface charge density. Generally, NaCa selectivity was independent of ionic strength. Vermiculite exhibited near-ideal exchange when compared with the nonpreference isotherm. There was no effect of mineralogy on the Na-Ca selectivity, even though the soils and minerals had various surface charge densities. The Gapon selectivity coefficients (KG) for the calcareous Many Farms soil (a mesic Torrifluvent) and the gypsiferous Shiprock soil (a mesic Torriorthent) averaged 0.011 and 0.013 (mmol L~')~i respectively. Failure to correct for anion exclusion and minC. Amrhein, Dep. of Soil and Environmental Sciences, Univ. of California, Riverside, CA 92521; and D.L. Suarez, U.S. Salinity Lab., USDA-ARS, 4500 Glenwood Dr., Riverside, CA 92501. Submitted by the U.S. Salinity Lab. Received 26 Jan. 1990. *Corresponding author. Published in Soil Sci. Soc. Am. J. 55:698-706 (1991). eral weathering lowered these average A"G values to 0.0085 and 0.0064, respectively. The data were compared with the Vanselow (Kv = 4.3), nonpreference (Ay = 1.0), Gapon (JiTG = 0.015), and the diffuse double-layer (DDL) models. After the review of the problems associated with the various methods of determining exchangeable cations, we recommend that the alcohol rinse method to remove soluble salts not be used. M SALT-AFFECTED SOILS require the addition of chemical amendments prior to leaching to remove soluble salts and replace adsorbed Na. In order to calculate the amount of amendment needed to reclaim a sodic soil, knowledge is required of the equilibrium relationship between the solution and exchanger phase composition of the soil. Traditionally, this relationship has been described using the Gapon convention and conveniently expressed by the linear equation ESR = (Ac SAR) + x [I] where ESR (exchangeable sodium ratio) is equal to ^Na/E(ca+Mg)> where Ef is the equivalent fraction of cation i on the exchanger phase, SAR is the sodium adsorption ratio of the solution phase denned as Na*/ V^(Ca + Mg) with concentrations in mmol L". Since x is a small intercept value, often assumed to AMRHEIN & SUAREZ: SODIUM-CALCIUM EXCHANGE 699 be zero (U.S. Salinity Laboratory Staff, 1954), the value of KG can be equated to ESR/SAR and is generally assumed to be a constant over a wide range of soil types, salinity, and exchangeable Na. The U.S. Salinity Laboratory Staff (1954) reported an average value for Ka of 0.015 (mmol L-')-' on 59 soil samples from the western USA. Many workers have reported variations in KG, and it is felt by many to be site specific (Levy and Hillel, 1968; Doering and Willis, 1980; Nadler and Magaritz, 1981). Two factors that are known to affect the NaCa selectivity relationship are organic-matter content (organic matter tends to increase Ca preference [Pratt and Grover, 1964; Fletcher et al., 1984b]) and pH (particularly in soils containing large amounts of variablecharge materials [Pratt et al., 1962]). In addition, the DDL theory for cation adsorption predicts an increasing Ca preference with increasing ionic strength and decreasing exchangeable Na (Babcock, 1963). The reported changes in Na-Ca selectivity as a function of salinity and exchanger phase composition have been quite variable. Bower (1959) found that the Kc of a montmorillonitic clay soil was not significantly affected by solution concentration (50-200 mmolc L~') but increased with increasing exchangeable Na; that is, Na preference increased with increasing ENa. Similar results were found by Pratt et al. (1962) on a soil high in amorphous clays and kaolinite and equilibrated with solutions that ranged in concentration from 50 to 330 mmolc L". They also found that the selectivity coefficient of this soil was strongly pH dependent. On the other hand, Doering and Willis (1980) reported that the KG of a strip-mine spoil material (montmorillonitic) was constant with variations in £Na but showed a strong dependence on ionic strength, decreasing from 0.0163 to 0.0085 (mmol L-')~ at ionic strengths from 0.04 to 1.06 molc L", respectively. Jurinak et al. (1984) reported that KG was a constant for a montmorillonitic soil from the Central Valley of California and for a surface overburden material from Montana at all concentrations and SAR values studied (0.01-0.5 molc Land 5-80 (mmol L-'), respectively). On a sample of kaolinitic deep overburden material, KG varied from 0.0138 to 0.0047 (mmol L-i)-i/2 across the concentration range 0.01 to 0.5 molc L-, respectively. They suggested that variations in clay mineralogy were a prime factor determining the ESRSAR relationship. They also noted that correction for anion exclusion was necessary at concentrations > 100 mmolc Lfor accurate exchangeable-Na values. Another convention for describing cation exchange was proposed by Vanselow (1932), in which the activity of the exchangeable cations are equated to their mole fraction (N,) on the exchanger and activity is used for ions in solution. This is in contrast to the Gapon convention, in which solution concentration is used. The mass-action expression for Na-Ca exchange and the Vanselow selectivity coefficient (Kv) for the reaction are as follows: