Chemometrics for optimization and modeling of Cu (II) continuous adsorption onto carboxymethylcellulose-alginate encapsulated graphene oxide hydrogel beads

The discharge of liquid effluents bearing heavy metals such as Cu (II) into the environment constitutes a source of pollution of both surface water and groundwater. Here, the boundaries of traditional adsorption methods have been challenged using a graphene oxide encapsulated in carboxymethyl cellulose-alginate beads in a fixed-bed column for continuous Cu (II) removal. Through the combination of response surface methodology (RSM) and artificial neural network (ANN), this work showcases a novel and cutting-edge approach to modeling adsorption processes. Experiments were conducted according to the Doehlert design by studying the interactive effects of three independent operational parameters, namely the bed height, the initial Cu (II) concentration and the flow rate on breakthrough time (t_b, min) and total bed adsorption capacity of Cu (II) (q_eq, mg/g). Under the following optimal conditions using RSM model: flow rate: 0.45 min/mL, initial Cu (II) concentration: 23 mg/L and bed height: 11 cm, the t_b and q_eq were 224.34 min and 39.22 mg/g, respectively. The best ANN training models obtained for t_b and q_eq has a maximum coefficient of determination of 0.996 and 0.952, respectively. It comprises six neurons in one hidden layer. The highest t_b and q_eq obtained by ANN model were 227.37 min and 38.26 mg/g, respectively. The outstanding adsorption capacity of the hydrogel beads adsorbent has been successfully demonstrated for Cu (II) removal on a fixed-bed column. These outcomes proved that the fixed-bed adsorber could be scale-up to treat effluent with low Cu (II) concentrations as a tertiary treatment plant.