Anion Exchange Chemistry of Middle Atlantic Soils: Charge Properties and Nitrate Retention Kinetics

The negative impact of nitrate (NO3) on groundwater supplies has sparked a great deal of interest and concern in recent years, particularly in areas where coarse-textured soils abound. In light of this concern, the anion exchange chemistry of eight soils from the Middle Atlantic region was studied with particular emphasis on NO3 retention and kinetics. The soils were chosen to encompass a range of physicochemical and mineralogical properties and were extensively characterized. Anion exchange capacity (AEC) was determined on Cl-saturated samples by desorption of Cl with SO4. Anion exchange capacity ranged from 0 to 1.35 cmolc kg' for the eight soils and was found to parallel increases in clay and Fe oxide contents in the soil profiles. Point of zero salt effect (PZSE) values were determined by potentiometric titration with 0.001, 0.01, and 0.1 M NaCl as the indifferent electrolyte. These were of little value in predicting the development of AEC for the soils. The kinetics of NO, adsorption and desorption were studied using a stirred-flow reaction chamber and a first-order reaction best described the data. Nitrate adsorption was found to be completely reversible, indicating a simple electrostatic retention mechanism. The effect of pH and NO3 concentration on cumulative NO, adsorbed (CNA) and on NO3 adsorption kinetics was also studied. The CNA was found to increase with a decrease in pH below 5.5 and to increase with increasing NO, concentration. The latter indicated an increase in competitiveness by NO3 for positively charged sites. C.V. Toner, IV and D.L. Sparks, Delaware Agric. Exp. Stn., Dep. of Plant Science, College of Agric. Sci., Univ. of Delaware, Newark, DE 19717-1303; and T.H. Carski, E.I. DuPont de Nemours Corp., Stine-Haskell Laboratories, Newark, DE. Published as Miscellaneous Paper no. 1275 of the Delaware Agric. Exp. Stn. Contribution no. 247 of the Dep. of Plant Science, Univ. of Delaware. Received 12 Aug. 1988. *Corresponding author. Published in Soil Sci. Soc. Am. J. 53:1061-1067 (1989). I RESPONSE TO the growing alarm about ground and surface water degradation, there has been increased concern about NO3 mobility and retention in soils. Excess applied N from various sources has been shown to negatively impact ground water under sandy soils of the Delmarva (Delaware-Maryland-Virginia) Peninsula (Liebhart et al., 1979; Ritter and Chirnside, 1982; Robertson, 1977a,b). At present, however, little attention has been focused on anion exchange chemistry and, in particular, on the retention of NO3 in soils of the eastern USA. Nitrate is usually assumed to be weakly held by soils and would not be extensively retained in the root zone. Recently, some researchers have reported retention of NO3 in Middle Atlantic soils in quantities worthy of consideration when formulating N-fertilizer recommendations. Meisinger et al. (1982,1983) found large quantities of NO3 from fertilizer applied the previous year remained in the root zone in spring, even though winter rainfall had been sufficient for percolation of NO3 through the soil profile. The amount of residual NO3 was influenced by the quantity of NO3 fertilizer applied. No definitive mechanism was offered for the observed NO3 retention. It could be ascribed to physical retention in occluded micropores or to chemical retention by adsorption. With the presence of variable-charged soil constituents, development of positive charge would be expected when the soil pH is less than the zero point of charge (ZPC) of the constituents due to protonation of surface functional groups. Negatively charged NO3 should be adsorbed to these protonated sites provided it can com1062 SOIL SCI. SOC. AM. J., VOL. 53, JULY-AUGUST 1989 pete effectively with other anions present in the soil solution. The presence of variable-charged components such as Al and Fe oxides and clay minerals such as kaolinite in soils of the Middle Atlantic region has been well established (Elliott and Sparks, 1981). Nitrate retention was observed in weathered tropical soils high in Al and Fe oxides from Chile, Mexico and Hawaii (Singh and Kanehiro, 1969; Kinjo and Pratt, 1971a,b; Kinjo et al., 1971; Espinozo et al., 1975). Retention of NO3 increased with decreasing pH, indicating pH-dependent variable charge was operative. Kingston et al. (1972) proposed that protonated surfaces of kaolinite and Al and Fe oxides were important sites for NO3 adsorption. Adsorption was considered electrostatic in nature and would therefore only occur at pH < ZPC. By determining the ZPC of a given soil as well as the native pH, one may be able to predict the sign of net surface charge and the likelihood of NO3 adsorption. In this study, several common soil types, sampled from the Middle Atlantic region and representing a wide range of physicochemical and mineralogical properties, were characterized and their AEC and nitrate adsorption potential (NAP) determined at native pH. Point of zero salt effect was determined to assess its potential for predicting AEC. Cumulative nitrate adsorbed was determined as a function of pH and adsorptive solution concentration using a kinetic approach to further substantiate retention mechanisms involved. Any attempt to predict the fate of NO3 amendments should consider the rate and mechanism of adsorption. MATERIALS AND METHODS Eight major soil types from the Middle Atlantic region were chosen for study. The soils were air dried, crushed and passed through a 1-mm mesh screen. Soil pH was determined in a water slurry (1:1 w/v) using an Orion 901 lonalyzer and an Orion combination electrode (Orion Research, Cambridge, MA). Organic matter was determined using a modified Walkley-Black procedure (Nelson and Sommers, 1982). Iron oxide content was determined using the Na-citrate-dithionite-bicarbonate extracting method (Mehra and Jackson, 1960). Amorphous Al oxides were measured using the NH4-oxalate extracting method (lyengar et al., 1981). The mineralogy of the <2 jim clay fraction was determined by x-ray diffraction (XRD) using fractionation procedures outlined by Sparks and Jardine (1984). Particle size distribution was determined using the pipet method (Gee and Bauder, 1986).