Sorption kinetics and equilibria of a mixture of atrazine and diuron for selected tropical soils from Uganda

Abstract

The purpose of this study was to measure and model the sorption behavior of a mixture of pesticides in selected tropical soils, using atrazine and diuron as probe compounds. This information is important since it can be a basis for recommending the appropriate application rates of the pesticide mixture to farmers so as to address the intended use and to protect the environment. The soil organic carbon fractions for soils from Arua, Kabale and Lira were 0.023, 0.009, and 0.013 respectively, with their pH values respectively being 8.03, 7.75 and 7.77. Single and mixture α values for Arua, Kabale and Lira soils were respectively (0.19, 0.2, 0.16) and (0.44, 0.51, 0.44), (0.33, 0.19, 0.21) and (0.49, 0.49, 0.44), (0.30, 0.31, 0.29) and (0.45, 0.44, 0.44) yet their respective k2 values were (0.10, 0.010, 0.10) and (0.12, 0.12, 0.11), (0.05, 0.06, 0.06) and (0.05, 0.05, 0.05), (0.08, 0.11, 0.11) and (0.17, 0.17, 0.21). This showed that in the mixture, atrazine obtains equilibrium faster compared to when it is single yet when diuron is in mixture, it behaves the same way as when in single meaning the equilibrium time remains the same for diuron regardless of single or mixture. The equilibrium time of atrazine, diuron and atrazine-diuron mixture was determined using the standard batch kinetics experiment. For atrazine, equilibrium was established by the 20th hour when single yet due to the increase in the mass transfer coefficient (α) when in mixture with diuron, equilibrium was attained as early at the 12th hour in all the three soils (Arua, Kabale and Lira). The batch one-site mass transfer model was used to optimize kinetic parameters α and Kd for atrazine data where non-linear regression procedures on C/Co data were used. The value of the mass transfer coefficient (α) both singly and in mixture was not affected by initial concentration (Co) and thus all C/Co data were plotted together and simulated with the one-site non-equilibrium model (OSNE). Diuron’s equilibrium time was 24 h. in all soils for both single and mixture except in Kabale where equilibrium stretched to 48 h. The two-site non-equilibrium model (TSNE) described the C/Co data as a function of time at each initial solution concentration in all the three soils for diuron both singly and in mixture with atrazine. All C/Co data were plotted together and simulated with the TSNE model. The batch equilibration method was used to measure sorption isotherms at time intervals determined by the kinetics experiments for atrazine and diuron both singly and in mixture. The adsorbed amount at equilibrium (Se), as from the kinetics experiment were plotted against their respective equilibrium solution concentration (Ce) for each initial concentration (Co) so as to yield an isotherm for a given soil and pesticide for either single or mixture. All sorption equilibrium experimental data for atrazine and diuron irrespective of single or mixture, fit the linear model of sorption. Kd values between single and mixture were significantly different for atrazine in all the three soils with the mixture having higher values, yet diuron’s Kd values had no significant difference between single and mixture in all the three soils except for Kabale where mixture had a significantly higher value. For both atrazine and diuron, singly and in mixture, all Kd values were linearly correlated with fraction of organic carbon content in each soil.