Mass Transfer in a Turbulent Contact Absorber

Abstract

Vast amount of discrepancies are found in literature reporting the hydrodynamic and mass transfer characteristics of the turbulent contact absorbers. Most of the literature available in this regard is for small diameter columns having wall effects. Therefore hydrodynamic and mass transfer characteristics of turbulent contact absorbers have been studied in a relatively large scale, 44.7 cm diameter Perspex column so that there are no wall effects. The data reported can now be used in the designing of these absorbers. Efforts have also been made to point out the reasons of discrepancies in the reported literature. Dimensionless correlations have been developed from the data obtained in this study based on the criteria that the ratio of the column diameter to packing diameter should be greater than 10, the column diameter should be greater than 15 cm, grid free area should be greater than 70%, the ratio of static bed height to column diameter should be less than 1 and the contributions of the gas and liquid distributors should be accounted for. The correlations developed were used to simulate the data of the earlier workers, which shows that the data of those workers who have fulfilled the above criteria, their data can also be represented with in ±30% error band, using the correlations developed in this study. To develop these correlations, necessary detailed pressure drop, liquid holdup, expanded bed height and the volumetric gas film mass transfer coefficients data was obtained in this study. To do this the variation of packing diameter (25mm, 38mm and 45mm), apparent packing density (160, 270, 354, 442, 547 kg/m3) and static bed height (15cm, 25cm and 35 cm) were studied. Gas and liquid velocities in the range of 1.8 to 3.6 m/s and 0 to 0.012 m/s were used respectively. For the hydrodynamic study water and air were used as working fluids. For the mass transfer coefficients the absorber was used in an adiabatic humidification mode. Violations of one or more of the above mentioned criteria are the main reasons for the vast amount of discrepancies in the literature data. These discrepancies fall within the experimental errors, only in the reported literature which has explicitly taken into consideration of the effect of gas and liquid distributors. It is also shown that all the reported data can be represented in a dimensionless form