Adsorção de Prednisona em carvões ativados – modelagem matemática

Abstract

The contamination of water resources by pharmaceuticals has been generating concern and encouraging research and improvement of treatment techniques. The objective of the work was to verify the efficiency of removal of Prednisone using two adsorbents: Mineral Activated Carbon (CAM) and Vegetal Activated Carbon (CAV) and perform the mathematical modeling of the adsorption process in column for the most efficient adsorbent. In order to verify the influence of adsorbent dose, drug concentration and pH on the adsorption process, an experimental planning (DCCR) was performed, the surface of the CAM and CAV were characterized (pHPCZ, BET and FTIR) and the kinetics and adsorption isotherms were performed in batch. For the most efficient adsorbent, the breakthrough curves were obtained in a continuous system by evaluating the effect of the variation of the operating flow and the initial concentration of the drug in the adsorption process. With DCCR it was verified that the CAM is more efficient in the removal of prednisone than the CAV and the variable that most influences the adsorption of prednisone in the adsorbents was the dose of adsorbent, being the pH significant only for the CAV at pH<4 or pH> 10. In the batch adsorption kinetics the values of qeq obtained were 22.44 mg g-1 and 9.79 mg g-1 for CAM (t: 3h) and CAV (t: 4h), respectively, and the pseudo-first order model represented the process better. The Isoterm models that best represented the batch data were the Freundlich model and the Langmuir model, for the CAM and CAV (qmax 18.04 mg g -1), respectively. The rupture curves obtained for the CAM in the continuous system indicated the flow of 1.1 mL min-1 as the best operating flow. The Langmuir Isotherm model presented the best fit to the data (qmax 109.27 mg g -1) and the modeling of the process showed that the step that controls the adsorption of prednisone in CAM is the intraparticle diffusion of adsorbent, presenting R² greater than 0.995.