The kinetics of sorption by retarded diffusion into soil aggregate pores.

This study investigates time-dependent sorption of pesticides in soil aggregates. We tested if the sorption kinetics of pesticides in soil aggregates can be described by modeling diffusion into aggregates for a range of soils and pesticides. Our hypothesis is that the rate of sorption is negatively related to sorption strength due to retardated diffusion. Natural aggregates of 3-5 mm diameter were separated from three soils: a clay, a silty clay loam, and a clay loam. The aggregates were stabilized with alginate gel, and adsorption of azoxystrobin, chlorotoluron, and atrazine was measured in batch experiments with eight equilibration times up to 28 days. Equilibrium sorption appeared to be reached within the 28-day period for each pesticide. An intra-aggregate diffusion model was employed to describe the increase of sorption with time. The model describes diffusion of the dissolved pesticides through the pore space inside the aggregates and sorption on internal surfaces. Sorption could be described by pore diffusion into the aggregates with diffusion coefficients between 0.5 x 10(-10) and 1.5 x 10(-10) m(2) s(-1). The model fits support the theory that pore diffusion is the rate-limiting process for sorption of pesticides in aggregates, although the diffusion coefficients were a factor 3-10 smaller than the theoretical diffusion coefficient for diffusion in water. Comparing the results from the different pesticide-soil combinations showed that the extent of nonequilibrium increased with increasing sorption strength. This confirmed that sorption takes longer to reach equilibrium for pesticides and soils with stronger sorption. The differences between the different pesticides and soils were fully accounted for in the model by stronger retardation of the more strongly sorbed pesticides. The results imply that diffusion into aggregates may be the major time-limiting process for sorption of pesticides in structured soils. Commonly performed sorption experiments with sieved soil fail to account for this process.