Effect of Operating Variables on the Gas Holdup in a Large-Scale Slurry Bubble Column Reactor Operating with an Organic Liquid Mixture

1999 ◽  
Vol 38 (3) ◽  
pp. 928-937 ◽  
Author(s):  
Juan R. Inga ◽  
Badie I. Morsi
2016 ◽  
Vol 14 (2) ◽  
pp. 599-619 ◽  
Author(s):  
Omar M. Basha ◽  
Li Weng ◽  
Zhuowu Men ◽  
Badie I. Morsi

Abstract A multiphase-Eulerian, three-dimensional (3-D), computational fluid dynamics (CFD) model was built to investigate the local hydrodynamics of a pilot-scale (0.29 m ID, 3 m height) Slurry Bubble Column Reactor (SBCR). The model was first validated against the gas holdup radial profiles in an air-water-glass beads system obtained in a 0.254 m ID and 2.5 m height column under ambient conditions at various superficial gas velocities by Yu and Kim (Bubble characteristics in the radial direction of three-phase fluidized beds. AIChE Journal 34, 2069–2072, 1988). The model was next validated against the gas holdup radial profile data for N2-Drakeol-glass beads system obtained in a 0.44 m ID and 2.44 m height reactor, including internals, operating under ambient conditions at various superficial gas velocities by Chen et al. (Fluid dynamic parameters in bubble columns with internals. Chemical Engineering Science 54, 2187–2197, 1999). The model was also validated against experimental data obtained in our lab for N2-Fischer Tropsch (F-T) reactor wax-Fe catalyst system obtained in a pilot-scale, Slurry Bubble column Reactor, SBCR (0.29 m ID, 3 m height) under pressures and temperatures up to 25.9 bar and 490 K, respectively. These three validations led to the selection of the turbulence and interphase drag coefficient models, and the optimization of the solution method, mesh size and structure and the step size. Moreover, the inclusion of RNG k-ε turbulence model coupled with the Wen-Yu (Mechanics of Fluidization. Chemical Engineering Progress Symposium Series 62, 100–111, 1966) / Schiller-Naumann (A drag coefficient correlation. Zeitung Ver. Deutsch. Ing 77, 318–320, 1935) drag correlations, and the mass transfer coefficients were found to provide the most accurate predictions of the experimental data. The CFD model was then used to investigate local gas holdup, liquid recirculation, local turbulence intensities, bubble diameters, and solids distribution throughout our pilot-scale SBCR, operating under typical F-T process conditions. The model predictions showed strong liquid recirculation and backmixing near the walls of the reactor, and the solid-phase velocity vectors closely followed those of the liquid-phase. A relatively high liquid turbulence intensities were observed in the vicinity of the sparger upon startup, however, after reaching a steady state, the liquid turbulence intensities became more evenly distributed throughout the reactor. The liquid turbulence intensities were slightly higher near the center of the reactor, and closely resembled the velocity vectors. Also, the Sauter mean bubble diameters increased, whereas the solids distribution decreased with reactor height above the gas distributor.


2013 ◽  
Vol 11 (1) ◽  
pp. 83-102 ◽  
Author(s):  
Laurent Sehabiague ◽  
Badie I. Morsi

Abstract The hydrodynamics (gas holdup, Sauter mean bubble diameter, d 32) and the overall volumetric liquid-side mass transfer coefficients (kLa) were measured in a large-scale (0.29 m ID, 3 m high) slurry bubble column reactor (SBCR) for He/N2 gaseous mixtures, as surrogates for syngas, in three different Fisher–Tropsch (F-T) products (liquid paraffins mixture, light F-T cut and heavy F-T cut) in the presence and absence of three different solids (spent iron oxides catalyst, alumina powder and Puralox alumina). The effects of pressure (10–30 bar), temperature (up to 500 K), superficial gas velocity (0.14–0.26 m/s), solid concentration (0–20 vol.%) and gas density on these design parameters were investigated. The experimental data revealed that increasing the reactor pressure or gas density increased the gas holdup and decreased d 32, by increasing the population of the small gas bubbles, which increased the overall kLa values for all the gas mixtures used in the three F-T cuts under most of the operating conditions employed. Increasing temperature increased the gas holdup in the three F-T cuts, except for N2-light F-T cut, where the gas holdup values remained almost constant from 400 to 500 K. Increasing the slurry concentration decreased the gas holdup and increased d 32, mainly for gaseous mixtures with high He mole fractions, which decreased the overall kLa under all conditions used. Increasing the gas superficial velocities (UG ) increased the gas holdup and kLa values, even though d 32 was found to increase or decrease with increasing UG . Increasing the He mole fraction in the He/N2 gaseous mixture at constant pressure led to low gas holdup and high d 32 which decreased kLa values, and under similar operating conditions, kLa values of He as a single gas were always lower than those of N2 as a single gas. Increasing the He mole fraction in the He/N2 gaseous mixture at constant density, however, was found to have negligible effect on the gas holdup, d 32 and subsequently on the overall kLa. The gas holdup, the overall kLa and the population of the small gas bubbles for N2 in the liquid paraffins mixture were greater than those in the light F-T cut. Operating the SBCR with the heavy F-T cut resulted in the lowest gas holdup and the largest gas bubbles size which led to the lowest gas–liquid interfacial area and consequently, the lowest kLa values. Also, under the operating conditions investigated, the behavior of overall kLa for the gases used in the three F-T cuts in the presence and absence of the three solids employed was controlled by that of the gas–liquid interfacial area (a). Using the data obtained, two novel empirical correlations for predicting the gas holdup and the overall kLa for gases specifically in F-T cuts are proposed.


Sign in / Sign up

Export Citation Format

Share Document