Application of Two-Site/Two-Region Models for Studying Simultaneous Nonequilibrium Transport and Degradation of Pesticides

Quantitative laboratory study of pesticide sorption and degradation during transport can provide insight into the basic processes affecting pesticide fate in field soils. Accordingly, we demonstrate the application of analytical solutions of two-site/two-region transport models useful in studying simultaneous pesticide sorption and degradation. Soil-column displacement experiments involving H2O, Cl, and atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) were conducted during steady-state water flow at two pore-water velocities and two pesticide concentrations. The soil used is a Valois silry loam (coarse-loamy, mixed, mesic Typic Dystrochrept). Effluent data from these experiments were used to demonstrate the application of these analytical solutions, as well as a parameter-estimation computer program based on these solutions. The ability to use laboratory-derived estimates of equilibrium sorption parameters to describe sorption under flowing conditions was evaluated at each flow velocity. Significant correlation between soil sorption partitioning and degradation prevents the simultaneous determination of both processes using these solutions. However, estimates of degradation obtained from mass balances of the column data were useful in identifying the equilibrium sorption parameters. Data collected elsewhere for 2,4,5-T (2,4,5-trichlorophenoxyacetic acid) herbicide transport were used as an additional example of the application of the model with degradation. Strengths and weaknesses of the models, and suggestions for further study, are presented. A BREAKTHROUGH CURVES for SOlutCS leached through porous media have been observed for a variety of chemicals in a broad range of soils. Much of these data were obtained in laboratoryscale experiments and are reviewed elsewhere (Rao et al., 1979; van Genuchten and Wagenet, 1989; Brusseau and Rao, 1989a). Recently, Winters and Lee (1987) reported tailing of three trace organics during in situ experiments under field conditions. Similar observations have also been made for organic contaminant transport in groundwater systems (Goltz and Roberts, 1986). These results indicate a need for understanding the processes behind the tailing phenomenon as contaminant-transport models are being implemented in research and management. Several models have been proposed to describe asymmetric breakthrough curves where tailing is attributed to either chemical or physical nonequilibrium processes. One-site or two-site chemical-process models assume that sorption is not at equilibrium and has a time-dependent or kinetic component. On the other hand, physical-process two-region models assume that sorption is always at equilibrium, but that transfer to some sorption sites is diffusion controlled. A.P. Gamerdinger, Dep. of Soil Science, Univ. of Florida, Gainesville, FL 32611; R.J. Wagenet, Dep. of Soil, Crop, and Atmospheric Sciences, Cornell Univ., Ithaca, NY 14853; and M.Th. van Genuchten, USDA-ARS, U.S. Salinity Lab., 4500 Glenwood Drive, Riverside, CA 92501. Joint contribution from the Dep. of 891!, Crop, and Atmospheric Sciences, Cornell Univ., and the U.S. Salinity Lab. Received 29 June 1988. *Corresponding author. Published in Soil Sci. Soc. Am. J. 54:957-963 (1990). In these models, water is commonly divided into mobile and immobile regions. Brusseau and Rao (1989b) discussed intra-organic-matter diffusion as a third process that may contribute to the tailing of hydrophobic organic compounds. The desire to distinguish physical and chemical processes that result in asymmetric breakthrough curves has led experimentalists to use several solutes in soilcolumn studies. Tritiated water (H2O) is often used to determine if immobile water is present in the system (Rao et al., 1980; Nkedi-Kizza et al., 1983). This tracer is probably the most suited for measuring the macroscopic average pore-water velocity, v, and the dispersion coefficient, D. The Cl ion is also often used as a tracer, the advantage being that this tracer is safer to handle than radioactive tritium. However, the use of Cl is only effective in so far as this tracer truly behaves as a noninteracting solute, thereby mimicking the flow of water. If this is true, then it may be possible to define a retardation factor, R, of unity. Unfortunately, Cl often undergoes anion exclusion, especially in relatively fine-textured soils (Biggar and Nielsen, 1962; Corey et al., 1963; Jacobs, 1964), resulting in R < 1. Organic-chemical transport is simultaneously influenced by both sorption and degradation. Miscible-displacement techniques may be better suited to simultaneous study of sorption and degradation than are traditional batch techniques, as flowing conditions, even in the laboratory, may better represent dynamic, transient field conditions. Unfortunately, nonequilibrium sorption confounds the measurement of degradation in miscible-displacement studies. Likewise, chemically or biologically mediated transformations may complicate the measurement of sorption, particularly when sorption is time dependent. A survey of the literature reveals that, in most previous soil-column experiments, great efforts were taken to eliminate degradation during studies of sorption kinetics accomplished with pesticide-displacement experiments. Detailed study of asymmetric breakthrough curves may be accomplished through a combination of analytical solutions of candidate transport models and properly designed experiments. The theoretical development of the two-site sorption (equilibrium and kinetic) and two-region (mobile-immobile water) transport models under conditions in which the solute is being transformed has been presented by van Genuchten and Wagenet (1989). We presented a generic analytical solution that allows implementation of different hypotheses about the relative rates of degradation in the liquid and solid phases of a soil. Here, we show how the generic model can be used to advantage in the analysis of observed column effluent data for Cl, H2O, and atrazine, and for 2,4,5-T effluent data of van Genuchten (1974). We emphasize that the purpose of our study is not to provide a comprehensive analysis of atrazine movement, but rather to il-