A novel approach to reaction modeling for photocatalytic oxidation processes

Abstract

Volatile organic compounds (VOCs) are one of the major concerns for indoor air quality. A new method for treating VOCs in the air is through an advance oxidation process (AOP) known as photocatalytic oxidation (PCO). This method uses Ultraviolet (UV) radiant power to activate the surface of a catalyst (e.g. Ti02). From the past experimental work that has been done, it can be concluded that PCO reactions follow the Langmuir-Hinshelwood (L-H) kinetic model. However, the complexity of the L-H kinetic model is difficult to simulate using existing Computation Fluid Dynamic (CFD) software. In this thesis, a new method for modeling the L-H surface reaction kinetics is proposed. The focus of this work is on the development of a novel approach to model the complex surface reaction rate expressions in order to define PCO reaction rates on the photocatalyst surfaces. A new approach is developed to adapt the overall experimental reaction rates, which are in terms of the total system volume. This adaptation will help in deriving the actual rate of reaction happening on the catalyst surface in terms of catalyst surface area. Two cases were studied in order to demonstrate how this new approach can be used to accurately model the complex reaction kinetics of PCO systems. In each case, an integrated CFD model was developed to accurately predict the rate ofVOC decomposition based on the work conducted by Shiraishi et al., (2005b) and Brosillon et al., (2008). In the first case, the experimental kinetic model for formaldehyde decomposition was adapted in order to describe a surface reaction based on formaldehyde concentration on the catalyst surface using three different approaches. It was determined that a two part polynomial rate expression was the most accurate one, as it was able to account for the higher initial rate of reaction. However, the exponential model did give reasonable results as well. In the second case, the reaction rate model was able to predict the rate of decomposition for butyric acid in the air for a variety of initial concentrations and UV irradiance levels at the catalyst surface. The developed CFD model results also discredited the assumptions made in a number of published papers that UV irradiance levels are uniform across a catalyst surface. Finally, a simple case study was developed in order to demonstrate how the novel approach to reaction modeling could be used to predict PCO system performance treating air in a close system. In this case, the PCO system was capable of treating air contaminated with butyric acid, as well as quickly reducing the concentration below the odor threshold.