Adsorption and desorption of selenium by two non-living biomasses of aquatic weeds at dynamic conditions

The adsorption and desorption of selenium by non-living biomasses of Eichhornia crassipes (Ec) and Lemna minor (Lm) at dynamic conditions were evaluated, in terms of: pH, flow direction, mass loading rate, and theoretical speciation. These biomasses are worldwide present in watersheds high in nutrients. The experimental adsorption data were fitted to Thomas Model to obtain the parameters which describe the dynamic process. The Se removal capacity of Ec was 0.3489 µg g⁻¹ and for Lm 0.1855 µg g⁻¹ at pH of 6 and initial selenium concentration of 0.02 mg L⁻¹. For both systems, the vertical flow results are more efficient to remove Se and the horizontal flow is more efficient to recover Se from the Ec packed columns. The highest Se adsorption capacity of non-living biomass of Ec was when the mass loading rate (MLR) is 2.85 mL min⁻¹ g⁻¹. For Lm, a MLR of 1.33 mL min⁻¹ g⁻¹ was more efficient to adsorb and the less efficient to desorb Se, attributed to its natural swelling physical characteristic and the strong bounding of Se. Both biomasses have the capacity to buffer the pH of the solution, which promotes a species change from selenate (\({\text{SeO}}_{4}^{ - }\)) to selenite (\({{{\text{SeO}}_{3}^{2 - } } \mathord{\left/ {\vphantom {{{\text{SeO}}_{3}^{2 - } } {{\text{HSeO}}_{3}^{ - } }}} \right. \kern-0pt} {{\text{HSeO}}_{3}^{ - } }}\)) during the adsorption process. The data for Ec packed columns are in accordance with the Thomas Model, suggesting that the adsorption process is by ion exchange due to the hydroxide groups naturally present in the non-living biomasses.