Mass transfer characteristics in a large-scale slurry bubble column reactor with organic liquid mixtures

https://doi.org/10.1016/S0009-2509(02)00201-4Get rights and content

Abstract

The mass transfer coefficients (kLa) and bubbles size distribution were measured in a large-scale (0.316-m inside diameter, 2.8-m high) slurry bubble column reactor (SBCR) for H2, CO, N2 and CH4 in two organic liquid mixtures (Isopar-M and a hexanes mixture) in the presence and absence of two solids (iron oxides catalyst and glass beads) in a wide range of pressure (0.17–0.8MPa), superficial gas velocity (0.08–0.2m/s), and solid concentration (0–36vol%).

The kLa values were found to intimately follow the behavior of the bubble size distribution under the operating conditions used, as they appeared to increase with pressure and gas velocity and dramatically decrease when operating with high solid concentrations. The effect of high solid concentrations in the reactor prevailed over the effect of pressure on the bubble size distribution by creating large gas bubbles with significantly small gas–liquid interfacial areas, leading to small kLa values. Thus, industrial SBCRs operating with high catalyst loading could likely operate in a mass transfer-controlled regime due to the expected low kLa values.

Introduction

Slurry bubble column reactors (SBCRs) are used in a wide spectrum of industrial processes, including syngas conversion to fuels and chemicals, heavy oil upgrading, and coal liquefaction (Stolojanu & Prakash, 1997; Fan, 1989; Ishibashi et al., 2001). These reactors are becoming more popular and gradually replacing fixed bed reactors in several important industrial applications, such as the Fischer–Tropsch synthesis. This is due to their numerous advantages, including ability of handling high slurry concentrations of fine particles, good interphase mass transfer rates at low energy input, high selectivity and conversion per pass, better temperature control (isothermal operation), high online factor and simple construction (Inga & Morsi, 1999; Gandhi, Prakash, & Bergougnou, 1999).

The design and scale-up of SBCRs require, among others, precise knowledge of the kinetics (reaction rate constants, order of the reaction with respect to all reactants and products), thermodynamics (solubilities of the gases into liquids, equilibrium rate constants), hydrodynamics (pressure drop, gas, liquid, and catalyst holdups and distributions), and heat as well as mass transfer characteristics (mass and heat transfer coefficients, interfacial areas) (Inga and Morsi 1997, Inga and Morsi 1999). Some experimental data on the kinetics and hydrodynamics of SBCRs can be found in the literature, however, only few studies are available on the mass transfer characteristics for organic liquids in large-scale reactors as can be seen in Table 1. In these studies, the effects of a number of operating variables, including solid concentration, system pressure, liquid viscosity, and liquid surface tension on the gas holdup (εG) and the mass transfer coefficient (kLa) were investigated. Krishna, de Swart, Ellenberg, Martina, and Maretto (1997), Kojima, Anjyo, and Mochizuki (1986), and Inga and Morsi (1999) reported that increasing solid particles concentration appeared to decrease the gas holdup by increasing the coalescence tendency of the gas bubbles. Quicker, Schumpe, and Deckwer (1984) and Schumpe, Saxena, and Fang (1987) found that kLa values increase with fine solid particles at low concentrations, whereas the gas–liquid interfacial area decreases with increasing solid concentrations. Dewes, Kuksal, and Schumpe (1995), observed an increase of kLa values with pressure (from 0.1 to 0.8MPa) due to the increase of the gas–liquid interfacial area. These authors reported that the gas holdup and the gas–liquid interfacial area slightly decrease with the addition of up to 2vol% glass particles and that kL at 0.8MPa was greater in the slurry system than in the gas/liquid system, containing no solids. Fukuma et al. (1987) used up to 50vol% glass beads and observed that kL values are proportional to the volume-surface mean bubble diameter and decrease with liquid viscosity, μL. Muller and Davidson (1995) studied the effect of surfactants on the mass transfer in a viscous liquid and found an increase of kLa values with the addition of surfactants. They attributed a 60–75% of this increase to the presence of small gas bubbles created in the system. Koide, Takazawa, Komura, and Matsunaga (1984) studied εG and kLa in the transition and heterogeneous flow regimes and reported that their values in both regimes decrease with increasing solid concentration. They also observed that this decrease was more pronounced in the transition regime than in the heterogeneous flow regime.

A number of correlations available in the literature for predicting kLa values under different conditions are listed in Table 2 and as can be seen most of these correlations were developed under ambient conditions, using air/water system in the absence of solid particles. These conditions, unfortunately, are not typical of those employed in various important industrial applications, where organic liquids and high solid concentrations are usually used to achieve high reaction rates. Thus, in order to understand the behavior of industrial SBCRs, the hydrodynamic and heat as well as mass transfer characteristics should be obtained in a wide range of operating variables typical to industrial applications, using large-scale reactors.

This study presents some experimental data on the volumetric mass transfer coefficient (kLa), and bubbles size distribution obtained for four gases (H2, CO, N2, and CH4) in two different organic liquid mixtures (hexanes mixture and Isopar-M). These data were measured in a large-scale SBCR (0.316m, 2.8-m high), in the presence and absence of two different solid particles (iron oxides catalyst and glass beads). The effects of pressure, superficial gas velocity, and solid concentration on kLa and bubbles size distribution were investigated and empirical as well as statistical correlations of the experimental data were developed.

Section snippets

Experimental setup

A schematic diagram of the SBCR used in this study is shown in Fig. 1. The reactor is 0.316-m diameter and 2.8-m height similar to that used by Inga and Morsi (1999). Several pressure transducers and thermocouples are provided on both the supply vessel and reactor in order to allow building a mass balance on the entire system. The gas is recycled through the system using a double-acting, single-stage, air-driven compressor, (model 8 AGD-1), manufactured by Haskel Inc., USA. A damper vessel is

Calculation procedures

The following assumptions were made for the gas solubility, C and kLa calculations:

  • 1.

    In the reactor, the behavior of the gas–liquid systems used was ideal. The operating conditions used justified such an assumption, since the temperature was low and the pressure was not very high (0.8MPa maximum). Thus, the ideal gas law was applicable to calculate the number of moles of gas absorbed in the liquid or slurry under the operating conditions employed.

  • 2.

    In the supply vessel, the behavior of the gas was

Gas solubility, C

The solubilities of H2, N2, CO and CH4 in the SBCR were calculated using the material balance and the equilibrium thermodynamic conditions after the absorption had been completed. In this study, these values were obtained in the absence and presence of solid particles under room temperature in a stirred reactor to insure better temperature and pressure control and more accurate liquid volume. A comparison between C values obtained in the stirred reactor and those in the SBCR were in very good

Conclusions

The kLa and bubble size distribution were measured in a large-scale SBCR for H2, CO, N2 and CH4 in two organic liquid mixtures in the presence and absence of iron oxides catalyst and glass beads. The effects of superficial gas velocity, pressure, and solid concentration on the mass transfer parameters were investigated and kLa values were empirically as well as statistically correlated.

The data showed that kLa values increase with pressure and gas velocity and dramatically decrease with

Notation

agas–liquid interfacial area per unit liquid (solid free), m−1
aiconstants of the statistical correlations, ,
Csolubility of the gas at equilibrium, kmol/m3
CLconcentration of the gas in the liquid bulk, kmol/m3
CSaverage solid concentration in gas-free slurry, ρSεS/(εSL), kg/m3
CVvolumetric solid concentration, vol%
CWconcentration of solid in the slurry, wt%
d32Sauter-mean bubble diameter, m
DABdiffusivity of gas in the liquid, m2/s
dbbubble diameter, m
DCdiameter of the column, m
DLliquid

Acknowledgements

Partial support of this research by Exxon Research and Engineering Company is gratefully acknowledged. The authors wish to thank SASTECH, Sasol Group of Companies (South Africa) for providing the iron oxides catalyst.

References (51)

  • J.O. Kim et al.

    Bubble breakage phenomena, phase holdups and mass transfer in three-phase fluidized beds with floating bubble breakers

    Chemical Engineering and Processing

    (1990)
  • H. Kojima et al.

    Effect of pressure on volumetric mass transfer coefficient and gas holdup in bubble column

    Chemical Engineering Science

    (1997)
  • H.M. Letzel et al.

    Gas holdup and mass transfer in bubble column reactors operated at elevated pressure

    Chemical Engineering Science

    (1999)
  • S.S. Ozturk et al.

    Influence of suspended solids on oxygen transfer to organic liquids in a bubble column

    Chemical Engineering Science

    (1987)
  • G. Quicker et al.

    Gas–liquid interfacial areas in a bubble column with suspended solids

    Chemical Engineering Science

    (1984)
  • E. Sada et al.

    Chemical absorption in a bubble column loading concentrated slurry

    Chemical Engineering Science

    (1983)
  • A. Schumpe et al.

    Gas/liquid mass transfer in a slurry bubble column

    Chemical Engineering Science

    (1987)
  • V. Stolojanu et al.

    Hydrodynamic measurements in a slurry bubble column using ultrasonic techniques

    Chemical Engineering Science

    (1997)
  • P.M. Wilkinson et al.

    Mass transfer and bubble size in a bubble column under pressure

    Chemical Engineering Science

    (1994)
  • K. Akita et al.

    Gas holdup and volumetric mass transfer coefficient in bubble columns

    Industrial and Engineering Chemistry Process Design and Development

    (1973)
  • Behkish, A., Lemoine, R., & Morsi, B. I. (2002). Use of neural network to predict gas holdup in slurry bubble column...
  • M.Y. Chisti et al.

    Hydrodynamics and oxygen transfer in pneumatic bioreactor devices

    Biotechnology and Bioengineering

    (1988)
  • A. Cockx et al.

    Modeling of mass transfer coefficient kL in bubble columns

    Chemical Engineering Research and Design

    (1995)
  • I. Dewes et al.

    Gas density effect on mass transfer in three-phase sparged reactors

    Chemical Engineering Research and Design

    (1995)
  • L.S. Fan

    Gas–liquid–solid fluidization engineering

    (1989)
  • Cited by (108)

    • Bubble collisions on parallel arranged fibers

      2019, International Journal of Multiphase Flow
    View all citing articles on Scopus
    View full text