Radiation Modeling and Degradation Kinetics of a Contaminant of Emerging Concern in Packed-Bed Photocatalytic Reactors

A. MANASSERO; O. M. ALFANO; M.L. SATUF

Buenos Aires

Congreso; 11th World Congress of Chemical Engineering; 2023

Heterogeneous photocatalysis has proven to be an effective technology to degrade contaminants of emerging concern like pharmaceuticals and personal care products. Nevertheless, photocatalytic applications at real scale are still scarce due the lack of a systematic approach for reactor design and scale-up. Proper reactor design involves rigorous mathematical modeling, including the derivation of kinetic expressions, the evaluation of the rate of photon absorption in the reactor, and the resolution of mass balances of the chemical species involved. In this work, a strategy to obtain intrinsic kinetic parameters in a simple, cylindrical packed-bed photocatalytic reactor is presented. This information was used to simulate the performance of a more complex, annular packed-bed reactor, with different dimensions, type of lamps, and illumination arrangement. Clofibric acid (CA), the active metabolite of many pharmaceuticals employed as blood lipid regulators, was the pollutant chosen to validate the proposal.Firstly, intrinsic kinetic parameters were obtained in a cylindrical reactor (CR) irradiated from one end by a mercury lamp. (emission range: 350 nm-410 nm). The reactor has a volume of 54 mL, and it was filled with 310 TiO2-coated glass rings. The catalyst was immobilized over the surface of the rings by the dip-coating technique [1]. A kinetic model to represent the photocatalytic degradation of the pollutant CA was developed. The expression of the CA degradation rate includes the value of the local surface rate of photon absorption (LSRPA). A 1-D radiation model was solved to obtain the LSRPA distribution inside the reactor. Then, the mass balance of CA was set in order to predict the evolution of the pollutant in the system, and the values of the intrinsic kinetic parameters were obtained by applying an optimization algorithm to adjust model simulations to experimental data. The second stage implicates the use of the kinetic parameters found in the first reactor to predict the performance of a second one: an annular reactor (AR) of 214 mL, illuminated internally and externally by 40 UV-LEDs with maximum emission at 365 nm, and filled with 900 TiO2-coated rings. This task involved the resolution of the mass balance for CA, using the kinetic expression obtained in the CR, and the calculation of the LSRPA. Due to its particular configuration and illumination arrangement, a 3-D radiation model was developed. Finally, simulation results were compared with experimental data. With only two kinetic parameters, and no adjustable factors, the evolution of the concentration of CA in the annular reactor could be predicted. The percentage root mean square error of the estimations was 4.6 %. The presented results demonstrate that kinetic parameters calculated with this methodology are independent of the reactor geometry, reactor size and irradiation conditions, and that they can be employed to design, optimize and scale-up photocatalytic reactors.