David K. Friday

PSA for Air Purification: Experiments and Modeling

Abstract The design of a PSA system for air purification applications is investigated with emphasis on experimental results. PSA experiments are performed using a test system consisting of a single adsorbable component, 1,1,2 trichloro, 1,2,2 trifluoroethane (CFC-113) and BPL carbon to examine the effects of (1) pressure ratio, (2) purge to product ratio, and (3) cycle time on system performance, where system performance is determined by the product air contaminant concentration. In-bed, vapor-phase concentration profiles are measured after a selected numbers of cycles in order to provide insight into the transient behavior of the system. An isothermal, trace component PSA model is developed which includes axial dispersion, an internal particle resistance, and non-linear adsorption equilibria. Results of the mathematical model are compared to experimental data.

Temperature Measurements to Characterize Dispersion Within PSA Beds

PSA experiments using a two-bed, laboratory-scale (100 Standard Liters per Minute [SLPM]) system equipped with in-bed temperature and vapor-phase concentration probes were performed using trichlorofluoromethane (R-11) as the feed contaminant and air as the carrier. In-bed, vapor-phase concentrations at each axial position were measured and correlated to the magnitude of the temperature swings during a cycle. PSA experiments were then conducted using a full-scale (5,660 SLPM) system with difluorochloromethane (R-22) and 13X molecular sieve. Fourteen temperature probes were placed within a bed at different radial and axial positions. Using the temperature-concentration correlation, it was possible to construct a visual representation of the dispersion characteristics of the larger beds.