No abstract

The dynamic gas disengagement technique is discussed comprehensively in this paper. Also outlined is the procedure for estimating holdup structures and bubble rise velocities in a bubble column, which is extended to include the methodology for estimating the Sauter mean bubble diameter and therefore the specific gas-liquid interfacial area. The error limits for the estimated quantities are established using two extreme cases to describe the disengagement process: constant rate disengagement and interactive disengagement of bubbles. The analysis is done assuming a bimodal bubble size distribution; however, generalized equations for a multimodal bubble size distribution are also presented. Sensitivity of results to the two cases is illustrated using results obtained from experiments conducted with the air-tap water system in two bubble columns (0.05 and 0.23 m in diameter, 3 m tall). Sauter mean bubble diameters and specific interfacial areas estimated using the two approaches provide lower and upper limits for values reported in the literature.

In this study a γ-ray densitometry was used to measure radial and axial volume fraction distributions in a 0.21 m diameter bubble column, and flow regime transitions in a 0.05 m and 0.21 m diameter bubble columns. Experiments were conducted with two- and three-phase systems (nitrogen−Fischer−Tropsch derived waxes−iron oxide or silica particles) in both batch and continuous modes of operation at 265 °C and ambient pressure. Average radial gas holdups in the large diameter column showed the existence of a maximum at the center of the column, in both two- and three-phase systems. The maximum was more pronounced at higher superficial gas velocities (churn-turbulent flow regime). The axial and average gas holdup measurements in two-phase systems were in good agreement with results obtained by a conventional method (pressure measurements along the column height). An independent treatment of all three phases did not yield satisfactory results, due to large impact of small experimental errors on calculated holdups. Alternative methods based on measured values of solids concentration in the slurry yielded satisfactory results. Instantaneous signals from the nuclear density gauges were used successfully to determine transitions from homogeneous flow regime to slug flow (small diameter column) or churn-turbulent flow regime (large diameter column) in two-phase systems.

Bubble sizes were measured for molten wax-nitrogen systems using photography and dynamic gas disengagement. The effects of operating conditions, system geometry and wax type were studied in 0.05-and 0.23-m-diameter by 3-m-tall bubble columns. Both techniques were used with FT-300 wax, while only the dynamic gas disengagement technique could be used with reactor waxes due to their dark color. For FT-300 wax, Sauter mean diameters obtained from photographs taken near the column wall were significantly lower than those obtained from photographs taken near the center. The d, values obtained from dynamic gas disengagement and photographic (near the column center) methods, for this noncoalescing medium, were in the range 0.5-1.6 mm in the largediameter column. For reactor waxes, d, values were significantly higher (1 -2 mm for Sasol and 1-5.5 mm for Mobil's reactor wax) and are in agreement with results reported in earlier studies with similar waxes (d. = 2-4 mm), where different experimental techniques (light transmission or hot wire anemometry) were employed. Snehal

No abstract