(401c) Process Modeling and Kinetic Parameter Estimation for the Design of Photocatalytic Wastewater Treatment Systems

Semiconductor photocatalysis in principle relies on the use of irradiation for catalyst activation and has received considerable attention over the past decades toward achieving efficient removal of organic toxic substances from wastewater effluents. Moreover, solar photocatalysis is considered as a cost-effective and sustainable water treatment technique, due to the utilization of the abundant solar energy and the guarantee of no toxic emissions. Several heterogeneous photocatalysts have been investigated under ultraviolet and/or visible light irradiation conditions. However, the most acute problem of slurry photocatalysis is the recognized need for a post-treatment catalyst recovery step. Therefore, research has focused on supported catalysts that can prevent particle migration into the treated water and reduce photocatalytic process costs even more, by eliminating post-treatment. 

The efficiency of solar photocatalytic removal of Bisphenol A (BPA) from aqueous samples has been previously studied by immobilizing two different photocatalytic materials (ZnO and TiO₂) via heat attachment onto a glass substrate and subsequently investigating the effect of several operating parameters known to affect this environmentally promising process: these include the immobilized catalyst amount, the initial BPA concentration, the treatment time, the water matrix, the addition of hydrogen peroxide and the simultaneous presence (or absence) of other organic substances.

A series of kinetic data (time-dependent concentration vectors) from sixteen (16) experimental campaigns has been employed in order to systematically extract a multiparametric kinetic model which can be efficiently used for photocatalytic wastewater purification design. The identification of the optimal model that can be formulated relies on the combined employment of (a) global optimization for kinetic parameter estimation, and (b) quantitative criteria for optimal model selection. The latter are namely the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC), which are measures of the relative quality of a statistical model (for a given set of experimental data) that rely on the likelihood function and can be used for model selection among a finite set of postulated models. As ad hoc parameter set expansion can easily increase the likelihood metric but also result in overfitting, both AIC and BIC resolve this problem by introducing a penalty term for the number of parameters in the model. The identification of the optimal multiparametric model is then implemented toward preliminary design of a photocatalytic wastewater purification system and subsequent sensitivity analysis.

REFERENCES

Donoso-Bravoa, A., Mailier, J., Martin, C., Rodríguez, J., Aceves-Larae, C.A., Vande Wouwer, A. (2011). Model selection, identification and validation in anaerobic digestion: A review, Water Research 45: 5347–5364.

Flores-Sánchez, A., Flores-Tlacuahuac, A., Pedraza-Segura, L.L. (2013), Model-based experimental design to estimate kinetic parameters of the enzymatic hydrolysis of lignocellulose, Ind. Eng. Chem. Res. 52(13): 4834–4850.

Miró A., Pozo C., Guillén-Gosálbez, G., Egea, J.A., Jiménez, L. (2012). Deterministic global optimization algorithm based on outer approximation for the parameter estimation of nonlinear dynamic biological systems, BMC Bioinformatics 13(90): 1–12.

Singer A.B., Taylor J.W., Green W.H., Barton, P.I. (2006). Global dynamic optimization for parameter estimation in chemical kinetics, J. Phys. Chem. A 110(3): 971–976.

Zacharakis, A., Chatzisymeon, E., Binas, V., Frontistis, Z., Venieri, D., Mantzavinos D. (2013). Solar photocatalytic degradation of Bisphenol A on immobilized ZnO or TiO₂ Int. J. Photoenergy 570587: 1–9.

Zhao, Y., Stadtherr M.A. (2011). Rigorous global optimization for dynamic systems subject to inequality path constraints, Ind. Eng. Chem. Res. 50(22): 12678–12693.