A reactor model that can interpret the effects of multiple variables on photocatalytic performance is crucial to bridging the gap between lab- and commercial-scale reactors. However, there is no explicit expression to simultaneously describe the relationships between reaction rate and operating parameters including initial concentration, flow rate, light intensity, relative humidity, and catalyst dosage. Therefore, this work presents a multivariate reactor model that extensively describes the effects of these quantities on reaction rate. First, the model is developed by resolving the Navier-Stokes equations coupled with the convection-diffusion equations and an improved reaction kinetics expression. Then, the particle swarm optimization (PSO) algorithm is integrated with the numerical simulation of velocity and concentration fields to determine intrinsic kinetic parameters. Subsequently, the multivariate model for NO degradation is experimentally validated. In addition, the versatility of the proposed modeling method is evidenced by the experimental data of different photocatalyst-pollutant systems from literature. The model predictions agree with experimental results well with the R2 values of 0.949–0.995. Furthermore, a CFD simulation is executed to understand the interactive effects of fluid flow, mass transfer, and kinetic reaction to guide reactor optimization. Finally, the model for NO removal using UiO-66-NH2 in the current study outperforms several reported NO removal models with the smallest RMSE of 2.57% and MAPE of 2.74%. It is concluded that the proposed modeling method could successfully predict the performance of different gas-solid or liquid-solid heterogeneous reactions.
Read full abstract