| dc.description.abstract | The decaffeination processes present as final result a caffeine solution in aqueous phase. Caffeine is removed from the solution by using activated carbon or other adsorbents. In search of selective adsorbents for caffeine adsorption, spent coffee grounds were considered a low cost waste, with potential for activated carbon production. In this work, we evaluated different methods of activated carbon production from spent coffee grounds for caffeine adsorption. It was initially performed a pretreatment in coffee grounds for caffeine removal and then, the sample was subjected to different chemical and thermal treatments. The chemical agents used were KOH, K2CO3, H3PO4 e HNO3, and the thermal treatments were carried out in muffle furnace or conventional microwave. Adsorption experiments were conducted with the aim of identifying the adsorbent with better performance on caffeine adsorption in aqueous solutions in the batch. Evaluations were carried out in terms of adsorption capacity and production yield. The best adsorbent was activated with H3PO4 in microwave oven, for which studies of kinetic, equilibrium and the influence of adsorption parameters in batch were performed. In the adsorption kinetic study, models of pseudo firstorder and pseudo second-order were tested. The equation that best fitted the experimental data was the pseudo second-order. The models applied to equilibrium data were Langmuir, Freundlich and Sips, the maximum adsorption capacity of the material was 117.65 m.g-1, in an equilibrium time of 24 hours. The best fit was given by Freundlich model, indicating that the adsorption occurred in multiple layers. In the study of adsorption conditions influence, the effects of the initial concentration of adsorbate, the dosage and particle size of the adsorbent, and the pH of the adsorbate solution was evaluated. It was observed that there is an equilibrium point between the capacity and the percentage of adsorption in the initial concentration of 600 mg.L-1. The optimum conditions found for the process were: dosage of 5 g.L-1, particle size in the range of 60-100 mesh, and a neutral pH of 7. As a complement to results obtained in the study of adsorption in batch, adsorption studies in fixed-bed columns were performed. Some operating parameters were evaluated, such as flow rate and initial concentration. A flow rate of 23 mL.min-1 was found as optimal and the increase of the initial concentration did not significantly alter the rate of useful adsorption. The breaks curves were modeled by non-linear regressions for the Bohart-Adams, Yoon-Nelson, Thomas and DoseResponse models, being the last model, the one with the best fit. For comparison purposes, adsorbents were produced from bone char treated thermally and chemically with the same chemical agents used in the spent coffee grounds. The results showed that only the adsorbent activated with HNO3 in microwave oven, obtained an adsorption capacity similar to the spent coffee grounds adsorbent. However, the bone char showed some disadvantages such as lower percentage yield and lower adsorption velocity in the studies in batch and fixed-bed columns. The adsorbents were characterized by determining the zero-charge point, thermogravimetric analysis, X-ray diffraction, X-ray Fluorescence Spectroscopy and infrared spectroscopy (FTIR). Finally, spent coffee grounds demonstrated a good potential as adsorbent in the removal of caffeine in the aqueous phase. | |
