Uso de Resíduos Lignocelulósicos Ativados como Adsorventes na Remoção de Íons Metálicos Tóxicos: Estudos de Batelada e Coluna Utilizando Planejamento Experimental

Abstract

The tururi and buriti fibers, and the castor bean stalks are extremely efficient to sorption of metal ions from aqueous solutions. In this study, the adsorbents were activated with different concentrations of alkali solutions (5, 7, 10 and 15% w/v). The materials were characterized by analytical techniques as infrared, X-ray Diffraction (XRD) and Thermogravimetry analysis. The infrared spectra revealed that associated groups to macromolecules of hemicellulose and lignin after alkaline activation have less intensity compared to the initial samples. The XRD revealed, that after castor bean stalks alkaline activation, the presence of the cellulose type II peaks, which did not occur with tururi and buriti fibers. The fluorescence X-ray data showed that the cations present in the initial adsorbents, K+ and Ca2 + are exchanged by Cu (II), Ni (II), Cd (II) and Pb (II) in the sorption process. Studies of the influences of variables: mass of adsorbent; agitation rate; the initial pH and concentration using a fractional factorial design 24-1 demonstrated that all variables affect the response (adsorption capacity mg g-1). As a result for a larger value of qtotal, conditions were observed while maintaining the pH at 5.5; mass of the adsorbent 50 mg; agitation at 200 rpm and the initial concentration: 500 mg L-1. The adsorption kinetics revealed fast adsorption process, about 30 minutes, and good fitting to pseudo-second-order theoretical model to all adsorption process studied. Intraparticle diffusion models as Weber-Morris and Boyd were testes in order to study the limiting steps of the process.The results showed that for tururi and buriti fibers (with the exception of nickel ions) the rate-limiting step is not the intraparticle diffusion, while for the castor bean stalks, it was found that the rate-limiting step of the process is the intraparticle diffusion. The study of monoelement and multielement systems were performed at pH 5.5, initial concentrations from 20-500 mg L-1, which the experimental data were fitted to the Langmuir, Freundlich and Sips theoretical models. The tururi fibers adsorption capacities (mg g-1) in monoelementar and multielement system follows the order Pb (188.79)> Cd (92.20)> Cu (32.82)> Ni (22.23) and Cd (77.53)> Pb (43.93)> Cu (24.99)> Ni (19.51), respectively. Buriti fibers adsorption capacities (g-1 mg) in monoelementar and multielement systems follows the order of Cu (143.1)> Pb (112.1)> Ni (103.7)> Cd (86.33) and Pb (69.12)> Cu (49.28)> Ni (45.10)> Cd (24.95), respectively. Castor bean stalks adsorption capacities (g-1 mg) in in monoelementar and multielement systems follows the order of Pb (175.1)> Cd (124.8)> Ni (111.1)> Cu (89.23) and Cu (56.78)> Pb (55.82)> Cd (44.72)> Ni (43.48), respectively. The results showed a better fit for the Sips model, relating to a heterogeneous adsorption. Fixed bed studies using castor bean stalks checking the influence of variables adsorbent flow (1, 2 and 3 mL min-1), the height of the column (5, 7 and 10 cm) and initial concentration (100 , 200, 300 mg L-1) by the Box-Behnken planning revealed that there were no influence between the variables in the studied. The breakthrough curves were well fitted to the Thomas model. The study in real effluent with Cu (II) (galvanoplastic sector) was carried out using the optimized condition: flow (1 ml min-1); bed height (10 cm) initial concentration: (245.5 mg L-1) and it was found adsorption capacity of 32.42 mg g-1. The mamoneira stalks adsorbent was used for five cycles to verify their potential reuse, and it was found no significant efficiency losses.