Computational fluid dynamics modeling of contaminant transport with adsorption filtration inside planar-shaped air-purifying respirator canister

• Published in: Chemical engineering research & design Vol. 196; pp. 171 - 183
• Main Authors: Oh, Geunwoo, Hyun, Yesol, Choi, Jung-Il, Lee, Jaeheon, Kim, Min-Kun, Jung, Heesoo
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2023
• Subjects: Adsorption modeling; Air age; Air-purifying respirator canister; Bayesian inference; Computational fluid dynamics; Porous media modeling; Air age; Computational fluid dynamics; Bayesian inference; Porous media modeling; Adsorption modeling; Air-purifying respirator canister
• ISSN: 0263-8762
• DOI: 10.1016/j.cherd.2023.06.020

We performed computational fluid dynamics (CFD) simulations of the contaminant transport with adsorption filtration inside a planar-shaped air-purifying respirator (APR) canister and predicted the breakthrough times of the canisters with various internal shapes. The numerical modeling of the adsorption process based on the backward differentiation formulas, which comprise the implicit method implemented in COMSOL Multiphysics, demonstrated the validity through the simulations of the fixed-bed column breakthrough test. Model parameters were estimated by applying Bayesian inference to the breakthrough test and pressure drop measurement data. Previous CFD studies assessed the filtering efficiency based on the local flow speed and mean air age; this approach replaced adsorption modeling, which is computationally expensive. We tested six planar-shaped APR canisters and observed that the dead zone near the walls in the air age field does not appear in the contaminant transport simulation with adsorption modeling. We identified that the velocity fields inside the filter determined the breakthrough distributions. Further, we confirmed that the breakthrough time was more related to the maximum flow speed inside the filter and less with local mean air age. We demonstrated the importance of simulating contaminant transport with adsorption modeling for the APR canister and reliability of the alternative indices. [Display omitted] •Computational fluid dynamics enabled adsorption modeling in respirator canister.•Model coefficients of filter were estimated via inverse uncertainty quantification.•Dead zone of gas transport with adsorption was different from that of the air age.•Internal structures of the canister improve the biased flow and breakthrough time.•Breakthrough was more related to maximum flow speed and less to local mean air age.