Photocatalytic Reactor for Indoor Air Purification: CFD Modeling and Experimental Evaluation

PASSALÍA, C.; ALFANO, O. M.; BRANDI, R. J.

Buenos Aires

Encuentro; 2011 ANSYS Conference & ESSS Users Meeting; 2011

Indoor air pollution poses a major threat to human health due to the prolonged exposure times to a wide variety of pollutants. Among them, formaldehyde (HCHO) is one of the most typical. A suitable technology for its elimination is the heterogeneous photocatalysis, an advanced oxidation technology that employs UV radiation and a photosensitive semiconductor. In this work we deal with the modeling and experimental evaluation of a continuous, single-pass, corrugated plate photocatalytic reactor for the elimination of gaseous formaldehyde in air. The reactor was constructed and operated in the laboratory. The reactor configuration consists of a corrugated stainless steel plate coated with titanium dioxide catalyst, irradiated from both sides with UV lamps. The stainless steel folded angle is 57°, giving a total of 17 triangular channels through which air with HCHO flows in a zig-zag pattern. The virtual geometry and meshing were made in GambitÒ software. Two types of volume elements were used in different zones of the reactor: wedges and tetrahedrons, reaching a total of 7 ´ 105 elements. The modeling of this corrugated wall photocatalytic reactor was achieved using ANSYS 12 suite, in particular, Fluent. Nonetheless, the radiative interchange between the reactor walls was modeled externally, using as a basis a superficial emission model for the lamps and the radiative interaction by the computation of geometrical configuration factors. The reaction kinetics was also imposed externally from previous experimentally determined parameters; being a superficial reaction, the kinetics was introduced into each catalytic wall.  Both the kinetics and the local configuration factors were introduced by means of user defined functions. The conditions of the experimental runs were replicated in the computational simulations. Model predictions of the formaldehyde overall conversion showed good agreement with experiments, with a root-mean-square error less than 4%.