Oxygen transfer rates in a single concentric tube airlift reactor

The study of oxygen absorption in airlift reactors has encompassed various reactor types and design configurations as well as a wide range of operational parameters and fluid compositions. Recent investigations have concentrated on airlift reactor types with liquid recycle. Many workers report the volumetric mass transfer coefficient, k_la for oxygen in terms of superficial gas velocity. Although the works are for absorption in reactors of similar design, reports from independent sources are seemingly inconsistent. In an attempt to unify in one correlation mass transfer from different systems, one report in particular relates k_la to operational parameters and gassing rates on dimensionless coordinates for two reactor types. Three gassing rates each for three liquid compositions -- distilled water, five weight percent sucrose, and wild carrot media (pH = 4.5) -- were studied to determine the volumetric mass transfer coefficient as well as the specific interfacial area for oxygen in absorption and desorption operations of a single concentric tube airlift reactor (internal diameter = 14 cm) with sparging in the central draft tube (internal diameter = 6.3 cm). Preliminary experiments in a bubble column were performed to determine how bubble diameter related to orifice size, liquid composition and gassing rate. Air or nitrogen was sparged at relatively constant bubble size through one of three perforated plate spargers into 2300 milliliters of liquid at volumetric gas flow rates ranging up to 26.6 mls/sec. The average liquid-phase mass transfer coefficient, k_la, the specific interfacial area, a, and the liquid phase mass transfer coefficient, k_l were analysed relative to gassing rate and operational parameters. The results are compared to data and correlations produced by reporters working with reactors of similar design. The data correlated to the gassing rate alone produced results similar to other reports, but indicate that more operational parameters should be incorporated into the correlations. The data from the present work were also compared to those reported on dimensionless coordinates. The present work produced smaller dimensionless mass transfer data than the reported values such that a correlation on the data extrapolated below but parallel to the reported correlation. The disparities among data produced in the present work and data to which they were compared have been attributed to the differences in sparging configuration and bubble size and to the lower gas phase turbulence generated in the present work.