The aim of this study was to evaluate diuron sorption, desorption and degradation in two anthropogenic soils (Terra Preta de Índio - TPI) in contrast to a sandy soil (Quartzarenic Neosol - NQo). Sorption-desorption studies were performed by the batch equilibrium method and biodegradation in biometric bottles using radiolabeled diuron in 14C. Freundlich coefficient (Kf) values ranged from 13.50 to 50.41 µmol(1-1/n) L1/n kg-1 in TPI-2 and TPI-1, respectively, indicating very high diuron sorption in anthropogenic soils, following the order: TPI-1 ≥ TPI-2 > NQo (99.10, 98.95 and 60.8%, respectively). Diuron desorption was very low in anthropogenic soils, ranging from 1.36 (TPI-1) to 1.70% (TPI-2), and 24% to NQo. Accumulated diuron mineralization to 14C-CO2 was < 3% at 70 days after herbicide application, regardless of the assessed soil. Formation of 35 and 44% residue bound to TPI-2 and TPI-1 was observed, higher than to NQo (17%). In contrast, the residue extracted from NQo varied from 72 to 91%, ranging from 48 to 83% for TPI-1 and TPI-2 during the incubation period. The degradation half-life (DT50) of diuron in anthropogenic soils was of 66.65 and 68.63 days for TPI-1 and TPI-2, respectively, while a period of 88.86 days was observed for NQo. The formation of only one herbicide metabolite in all soils was evidenced. The application of diuron in arable areas in the presence of anthropogenic Amazonian soils may lead to inefficient chemical weed control, since these soils may reduce herbicide soil bioavailability due to high OC contents, where high sorption and low herbicide desorption are noted, as well as faster degradation compared to sandy soil.
bound residue; mineralization; sorption isotherms; amazonian soils