OTIMIZATION OF DICLOFENAC SODIUM REMOVAL BY ADSORPTION IN GRAPHENE OXIDE, POWDERED AND GRANULATED ACTIVATED CARBON USING A CENTRAL COMPOSITE DESIGN
Micropollutants. Adsorption. Nanosheets. Graphene Oxide. Powdered Activated Carbon. Granulated Activated Carbon.
Recent studies developed in different countries reveal the presence of pharmaceuticals drugs in aquatic bodies. Among the main pharmaceuticals detected, the anti-inflammatory sodium diclofenac (SD) has greater ecotoxicity and was included in the list of potentially dangerous substances in Directive 39/2013 of the European Union. Therefore, in this work, DS adsorption by graphene oxide (GO), pulverized activated carbon (PAC) and granulated activated carbon (GAC) were compared. The adsorbent materials were subjected to characterization tests by X-ray diffraction, Fourier transform infrared spectroscopy and attenuated total reflectance, thermogravimetric analysis, differential scanning calorimetry and point of zero charge point (PZC). The DS adsorption was analyzed through a central composition design, where the influence of the concentration of sodium diclofenac (CSD), adsorbent concentration (ADSC), contact time (Ct) and pH were evaluated. The results supported the modeling of responses for adsorption capacity (mg.g-1), adsorbate removal (%), pseudo-first order adsorption kinetics, pseudo-second order (PSO) and intra-particle diffusion (IPD), as well as Langmuir isotherms and Freundlich. The modified Hummers method was effective in the chemical oxidation of graphite and showed a crystalline diffraction peak at 2θ = 10.64°. The synthesized GO showed oxidized functional groups, especially the carboxylic group that it characterizes with oxide as a hydrophilic and extremely acid material (PCZGO 1.65). PAC was characterized as a semi-crystalline structure and GAC was amorphous, both demonstrated to have the predominant hydroxyl group in its composition and were characterized as slightly basic adsorbents (PZCPAC 7.34 and PZCGAC 7.52). The maximum adsorption capacity for GO was 669.50 mg.g-1 for CSD of 450 mg.L-1, ADSC of 0.2 g.L-1, Ct of 34.3 min and pH 5. PAC obtained maximum capacity of 169.39 mg.g-1 for CSD of 331.64 mg.L-1, ADSC of 0.2 g.L-1, Ct of 40.6 min and pH 5. The GAC showed a capacity of 77.73 mg.g-1 for 450 mg.L-1 CSD, ADSC of 0.2 g.L-1 ADSC, 25 min Ct and pH 9. The adsorption capacity was obtained by duplicate confirmation batches in which the GO was 295% more efficient than PAC and 761% than GAC. The GO exhibited SD removal ranging from 97.59% to 99.95% and there was no statistically significant difference for ADSC above 1.4 and 5 gL-1, CSD from 50 to 450 mg.L-1, Tc between 5 and 45 min and pH between 5 and 9. The model developed for the PAC predicted removals of up to 100% for CSD from 50 to 150 mg.L-1. The GAC demonstrated to have a maximum removal of 43.50% for CSD of 50 mg.L-1. GO, PAC and GAC were better suited to the PSO model. IPD revealed that adsorption is controlled by the combination of intra-film and intra-pore mechanisms. The isotherms of the GO and GAC adapted to the Freundlich model and the PAC to Langmuir. DS adsorption by GO was characterized by chemisorption, while PAC and GAC were through physisorption. GO has shown to be promising in the adsorption of DS in aqueous solution and has interesting characteristics for implementation in water treatment units as it requires a short contact time, adapts to different pH and provides greater sanitary safety.