1964 Korean J. Chem. Eng., 31(11), 1964-1972 (2014) DOI: 10.1007/s11814-014-0121-4 INVITED REVIEW PAPER pISSN: 0256-1115 eISSN: 1975-7220 INVITED REVIEW PAPER † To whom correspondence should be addressed. E-mail: akhanna@iitk.ac.in Copyright by The Korean Institute of Chemical Engineers. Experimental analysis and development of correlations for gas holdup in high pressure slurry co-current bubble columns Shyam Kumar and Ashok Khanna † Department of Chemical Engineering, IIT Kanpur, UP-208016, India (Received 20 September 2013 • accepted 21 April 2014) Abstract -The effect of liquid and gas velocities, solid concentrations, and operating pressure has been studied ex- perimentally in a 15 cm diameter air-water-glass beads bubble column. The superficial gas and liquid velocities varied from 1.0 to 40.00 cm/ s and 0 to 16.04 cm/ s, respectively, while the solid loading varied from 1 to 9%. The gas holdup in the column was reduced sharply as we switched from batch to co-current mode of operation. At low gas velocity, the effect of liquid velocity was insignificant; while at high gas velocity, increasing liquid velocity decreased the gas holdup. Drift flux approach was applied to quantify the combined effect of liquid and gas velocities over gas holdup. For co-current three phase flows, the gas holdup decreased with increase in solid loading for all pressures. But for batch operations, when solid loading was 5% or more, settling started leading to higher gas holdup. Increasing pressure from atmospheric conditions increased the gas holdup significantly, flattening asymptotically. Keywords: Multi Phase Flows, High Pressure, Gas Holdup, Drift Flux, Solid Concentration INTRODUCTION A bubble column is a multiphase reactor in which the gas flows in the form of bubbles through the cylindrical reactor while liquid remains either stagnant (batch case) or also flows in the column. If the direction of the gas and liquid is the same, it is co-current flow. When solid particles are also present in the system, it is three phase flow. These types of reactors are used in chemical, petrochemical, biochemical and metallurgical industries [1]. Some of the exam- ples include chemical processes involving reactions such as oxida- tion, chlorination, alkylation, polymerization and hydrogenation, biochemical processes like fermentation and biological wastewater treatment [2,3]. The advantages of this reactor over other multiphase reactors include less maintenance and low operating costs due to lack of moving parts, higher values of interfacial area and mass transfer coefficients, ability to handle solids, and high liquid residence time [2]. However, back mixing and complex hydrodynamics are the disadvantages. Average gas holdup is a dimensionless key parame- ter that characterizes the hydrodynamics inside the bubble column needed for design and scale up [4]. It is defined as the volume frac- tion of the column occupied by the gas bubbles. Variables affecting the gas holdup in a bubble column are liquid and gas velocities, solid concentration, operating pressure and column diameter [2,3]. The gas holdup does not depend upon column diameter for up-flow sys- tems, if the diameter is greater than 15 cm [5]. The average gas holdup in a column has been estimated by many invasive and non-invasive techniques [6,7]. These include ultrasound, PIV, γ -ray, X-ray, laser Doppler anemometry, pressure drop meas- urements. Measuring the pressure drop using differential pressure transducer (DPT) is a low cost non-invasive technique, hence does not interrupt bubble column operation. Also, this technique does not necessarily need a transparent fluid, or electrolytic liquid [8]. It is used to measure the axial variation of the holdup in a column as well as the overall average gas holdup. Previous researchers have used it to calculate the gas holdup in two- and three-phase bubble columns [9,10]. Tang and Heindel [11] studied the effect of liquid velocity in a 15.24 cm diameter column [11]. They observed a decrease in holdup for co-current flows compared to batch experiments in a two-phase column. The effect of solid loading has been studied by various authors for a batch bubble column [12-17]. One group found a decreasing effect of solid loading over gas holdup [12,14-17]. The other group says that increasing solid loading first increases the gas holdup, and then after a certain value it decreases the holdup [13,18,19]. The presence of solid increases the mixture viscosity, which in turn in- creases the bubble size [20,21]. Further, the reduction of bubble break- up allows larger bubbles to increase in the column [22,23]. These large bubbles rise faster than small bubbles [12,17]. This decrease in the residence time leads to lower gas holdup. In the presence of solid, the increment in bubble population from small to large bub- bles was also observed by Swart et al. [15]. Krishna et al. attributed the observed decrease in gas holdup to enhanced bubble coales- cence with increasing slurry concentrations [24]. Gandhi et al. studied the effect of fine glass beads loading (35 μm) in water up to 40 vol % in discrete step of 10 % on the gas holdup in a batch column [22]. They found that the average gas holdup initially decreases fast with increasing slurry concentration and the rate sub- sequently decreases for higher slurry concentrations. Kumar et al. performed experiments for three-phase flows at atmospheric con- ditions up to 16 cm/s gas velocity and found that the gas holdup remains nearly constant from batch case to up to 3% solid loading [25]. Afterwards it decreases when solid loading is increased. Mena et al. discussed the effect of solid loading (2.1 mm size with neu- trally buoyant particle) varying from 1 to 30% in a batch column