2,4-Dichlorophenoxyacetic acid removal from aqueous solutions via adsorption in the presence of biological contamination

• Adsorption efficiency of GAC decreases when biological contaminants are present.
• Rather than linearization, non-linear method is preferred to obtain model parameters.
• Langmuir isotherm is a special case of Redlich–Peterson, Sips and Toth.
• The column process represents more realistic application of the adsorption technique.
• Thomas and Yoon–Nelson kinetic models showed a better fit than Bohart–Adams.

In this study, the adsorption of the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D), a chemical agricultural pollutant, onto granular activated carbon (GAC) was accomplished in the presence of biological contaminants in a batch and continuous system. In the batch studies, the maximum sorption capacities (mg/g) exhibited by GAC were found to be 5.9, 76.8, 124.0, 173.1, and 177.6 in the presence of Acidovorax avenae subsp. avenae LMG 17238, Gracilaria verrucosa, a group of aquarium-isolated microorganisms, Spirulina platensis, and in the absence of microorganisms, respectively. Two and three-parameter non-linear equilibrium models—Langmuir, Freundlich, Redlich–Peterson, Sips, and Toth—were applied to describe the batch sorption process. In the continuous-flow column studies, breakthrough curves were plotted as a function of influent 2,4-D concentration (50–200 mg/L), flow rate (0.2–0.4 mL/min), GAC mass (0.75–1.5 g), and microorganism load. The highest bed capacity was obtained by using 200 mg/L inlet 2,4-D concentration, 0.2 mL/min flow rate and 1.5 g GAC mass. In the presence of biological contaminants, the order of adsorption of 2,4-D in terms of the maximum adsorption capacity (mg/g) from the least to the greatest was as follows: no microorganism < S. platensis < the aquarium-isolated group of microorganisms < G. verrucosa < LMG 17238. Among the kinetic models applied to the fixed-bed column data, the Thomas and Yoon–Nelson models showed a better fit than the Bohart–Adams.