An annular flow pressure drop model has been developed and compared with experimental data obtained on a pilot-plant-size Pease-Anthony-type venturi scrubber. Droplet and film accelerations as well as wall friction, based on the two-phase Lockhart-Martinelli correlation, are considered. Excellent agreement with experimental data is demonstrated for a wide range of throat gas velocities, liquid to gas ratios, and film flow rates for a single scrubber geometry. Venturi scrubbers have been recognized widely for their high particulate matter collection efficiencies. Fine-particle collection generally requires elevated pressure drops. The primary objective of this study was to develop a realistic mathematical model for prediction of pressure drop in a PeaseAnthony venturi scrubber.Flow losses are predicted by accounting for frictional effects and acceleration of liquid drops and the liquid films flowing on the walls. Recovery in the diffuser is also considered. Model validation involved measurement of pressure gradients, film flow rates, liquid to gas ratios, and throat gas velocities for a pilot-plant-scale venturi. A comparison is made among three widely used correlations using experimental data taken from the pilot-plant-scale venturi scrubber.
CONCLUSIONS AND SIGNIFICANCEAn annular flow model is developed for accurate prediction of pressure drops in Pease-Anthony venturi scrubbers. This model-which considers the primary design parameters of liquid to gas ratio, throat gas velocity, venturi geometry, and liquid film flow rate-accurately predicted the measured pressure gradients and overall energy losses.The Hesketh (1974) correlation underestimated pressure drops at all liquid to gas ratios and throat gas velocities tested. The Calvert modified model (Yung et al., 1977a) predicted overall pressure drops lower than those experimentally measured for liquid to gas ratios below 8.0 X lod4 m3 liquid/m3 air, and greater magnitudes for liquid to gas ratios exceeding 1.3 xThe region of good agreement was achieved in spite of the neglect of wall friction, converging section losses, and diffuser pressure recovery when these effects compensated each other.Boll 's model (1973) consistently overpredicted the experimental pressure drops. The deviation increased with increasing liquid to gas ratios. Since film flow was not considered, the droplet acceleration component was always overestimated. The frictional losses, which were predicted using a homogeneous model (Wallis, 1969), were overestimated.