Mass transfer in mechanically agitated gas—liquid contactors

doi:10.1016/0009-2509(71)83019-1

Chemical Engineering Science

Volume 26, Issue 3, March 1971, Pages 461-479

https://doi.org/10.1016/0009-2509(71)83019-1 Get rights and content

Abstract

The theory of absorption with pseudo mth order irreversible reaction was used to obtain effective interfacial area, a, in mechanically agitated vessels. The liquid-side mass transfer coefficient, k_La was found by using the theory of absorption with slow chemical reaction. Eleven different systems were used to obtain a and four different systems were used to obtain k_La. The diameter of the contactor was varied from 12.5 cm to 70 cm.

The effect of the following variables on a and K_La was investigated: speed, type and diameter of the agitator; tank diameter, height of the agitator and height of the clear liquid from the base of the contactor; ionic strength, nature of ions, viscosity and surface tension of the liquid. It appears that values of a and k_La cannot be predicted a priori, as system properties have a profound effect on a and k_La. However, experiments can be done in a small agitated contactor (say, 20 cm dia. contactor) and scale-up can be done on the basis of equal tip speed divided by the square root of the contactor diameter, possibly with a fair degree of confidence.

Résumé

La théorie de l'absorption avec une pseudo-réaction irréversible de mième ordre a été utilisée pour obtenir une surface interfaciale effective, a, dans des récipients agités mécaniquement. On a trouvé le coefficient de transfert de masse du côté liquide, k_La, au moyen de la théorie de l'absorption avec une réaction chimique lente. Onze systèmes différents on été utilisés pour obtenir a et quatre pour obtenir k_La. Le diamétre du contacteur variait de 12,5 à 70 cm.

L'effet des variables suivantes sur a et k_La a été étudié: vitesse, type et diamétre de l'agitateur; diamétre du réservoir, hauteur de l'agitateur et hauteur du liquide transparent depuis la base du contacteur; résistance ionique, nature des ions, viscosité et tension en surface du liquide. Il apparaĭt que les valeurs et a et k_La ne peuvent ĕtre prévues a priori, a car les propriétés du système affectant profondément a et k_La. Toutefois, les expériences peuvent ĕtre effectuées dans un peit contacteur agité (par example, un contacteur de 20 cm de dia.) et l'augmentation à l'échelle peut ĕtre faite, avec une certaine assurance, sur la base d'une vitesse de pointe égale divisée par la racine carrée du diamètre du contacteur.

Zusammenfassung

Es wurde von der Theorie der Absorption mit nicht umkehrbare Reaktion pseudo mter Ordnung Gebrauch gemecht, um die wirksame Grenzfläche a in mechanisch gerührten Gefässen zu errechnen. k_La, der Übertragungskoeffizient Flüssigkeit/Seitenmasse, wurde unter Anwendung der Absorptionstheorie bei langsamer chemischer Reaktion bestimmt. Es wurden 11 verschiedene Systeme benutzt, um a zu bestimmen, and 4 verschiedene Systeme, um k_La festzustellen. Der Durchmesser des Kontaktgefässes wurde zwischen 12,5 cm und 70 cm variiert.

Es wurde die Wirkung der folgenden Veränderlichen auf a und k_La untersucht: Drehzahl, Typ und Durchmesser des Rührers; Gefäßdurchmesser, Höhe des Rührers und der klaren Flüssigkeit vom Boden des Kontaktgefässes; Ionenstärke, Beschaffenheit der Ionen, Viskosität und Oberflächenspannung der Flüssigkeit. Es scheint, daß Werte von a und k_La nicht a priori vorausbestimmt werden können, da die Merkmale des Systems eine erhebliche Wirkung auf a und k_La ausüben. Es können jedoch Versuche in einem kleinen gerührten Kontaktgefäß (z.B einem Kontaktgefäß mit 20 cm Durchmesser) durchgeführt werden, und die Umrechnung auf größere Maßstäbe kann möglicherweise mit erheblicher Erfolgsaussicht aufgrund gleicher Kippgeschwindigkeit (?) geteilt durch die Quadratwurzel des Kontaktgefäß-Durchmessers durchgeführt werden.

References (25)

K.R. Westerterp et al.
Chem. Engng Sci.
(1963)
V. Linek et al.
Chem. Engng Sci.
(1969)
H. Boerma et al.
Chem. Engng Sci.
(1968)
M.M. Sharma et al.
Br. Chem. Engng
(1970)
A.S. Jhaveri et al.
Chem. Engng Sci.
(1967)
P.V. Danckwerts
Trans. Faraday Soc.
(1950)
G. Marrucci et al.
Chem. Engng Sci.
(1967)
H.K. Abdel-Aal et al.
A.I.E.Ch.Jl
(1966)
J.A. Wesselingh et al.
The oxidation of Aqueous sulphite solutions: A model reaction for measurements in gas—liquid dispersions
(211969)
A.S. Jhaveria et al.
Chem. Engng Sci.
(1969)

M.M. Sharma et al.

Mass transfer with chemical reaction

R.A. Mashelkar et al.

Trans. Instn chem. Engrs

(1970)

Cited by (146)

Power consumption and gas–liquid mass transfer in a hot-sparged three-phase stirred reactor
2019, Powder Technology
The volumetric mass transfer coefficient (k_La) is one of the most important parameters for design and scale-up of gas–liquid–solid three-phase mechanically stirred tank reactors. In this work, the effects of agitation speed, superficial gas velocity, temperature, and volumetric solid concentration (C_V) on the power consumption and k_La were discussed and quantitatively analyzed. In an air-water-glass beads system, the gassed power number (N_PG) and k_La increase remarkably with the temperature. At ambient temperature, k_La decreases significantly with increasing C_V with a power law exponent of 3.10, but this decreasing effect becomes weaker at higher temperatures. In this hot-sparged three-phase system, the addition of solids slows down the increasing tendency of N_PG with the temperature, but enhances the increasing tendency of k_La. The increase in temperature weakens the decreasing effect of C_V on k_La. Compared with the effect of temperature on N_PG and k_La, the effect of C_V is nearly negligible. As a practical guide to industrial applications, it is more effective to enhance the gas–liquid mass transfer by increasing the temperature or power consumption than by changing C_V or increasing superficial gas velocity. The empirical correlations of power consumption and k_La obtained in this work can provide helpful guidance for the industrial design and operation of hot-sparged three-phase stirred tank reactors.
Experimental study on gas-liquid-liquid macro-mixing in a stirred tank
2012, Chemical Engineering Science
Citation Excerpt :
Typical applications include hydroformylation, hydrogenation, microorganism fermentation, liquid–liquid extraction, etc. Previous investigations concerning the gas–liquid–liquid dispersions in stirred tanks were mainly focused on the minimum agitation speed for complete liquid–liquid dispersion (Armenante and Abu-Hekmeh, 1994; Kamil et al., 2001), the effect of a dispersed second inert liquid phase on gas–liquid interfacial area (Das et al., 1985; Mehta and Sharma, 1971), gas absorption rate (Bruining et al., 1986; Cents et al., 2001; Dumont and Delmas, 2003; Lekhal et al., 1997; Linek and Benes, 1976; Van der Meer et al., 1992; Van Ede et al., 1995) and homogeneous biphasic catalytic aspects (Purwanto and Delmas, 1995). The knowledge of macro-mixing time is essential when the rate of mixing is comparable with or slower than that of chemical reaction or mass transfer (Raghav Rao and Joshi, 1988).
In this paper, experimental data on the mixing time of the continuous phase and power consumption of gas–liquid–liquid dispersions in a mechanically agitated baffled tank are presented. The electrical conductivity method is taken for the measurement of mixing time and the shaft-torque method for power consumption measurement. Tap water is used as the continuous phase, and kerosene and air as the dispersed ones. The effects of probe/tracer injection position, agitation speed, type of impeller, clearance of impeller off tank bottom, oil volume fraction, gas holdup and physical properties of the dispersed liquids on the macro-mixing of the gas–liquid–liquid system have been investigated. The phenomenon of gas–liquid–liquid macro-mixing in a stirred tank is largely similar to that of liquid–liquid and gas–liquid stirred tanks. Our experiments indicate that the gas–liquid–liquid macro-mixing can be enhanced at higher gas holdups while damped at low gas holdups. Contrary to gas effect, the dispersed oil phase at low holdups increases the macro-mixing intensity but at higher holdups decreases the macro-mixing intensity of the continuous phase. The experimental results show that axial impellers are more energy efficient for gas–liquid–liquid macro-mixing than radial impellers. A simple correlation is developed for predicting the mixing time in gas–liquid–liquid three-phase systems and satisfactory agreement with experimental data is observed.
Hydrodynamics of a carbon dioxide/water/silicone oil bubble column
2011, Chemical Engineering Journal
Citation Excerpt :
Several studies [5,6] have analysed the effect of different organic liquids (i.e. toluene, dodecane, heptane, etc.) on the oxygen absorption process; however, different experimental results have been obtained showing contradictory conclusions. On the one hand, several studies have found positive effects upon the global absorption process [7,8] but, on the other hand, the opposite behaviour has also been obtained [9]. This variety of results could be due to the lack of experimental data in these studies: those about the analysis of the interfacial area produced between gas–liquid–liquid systems, as well as the influence of the organic substances upon the gas hold-up, the bubbles size and the interfacial area.
The present work analyses the hydrodynamic behaviour of a gas phase in a bubble column using different emulsions as liquid phase. The hydrodynamic behaviour was analysed on the basis of gas hold-up, bubble diameter and gas–liquid interfacial area; the different variables employed in this study [gas flow-rate (between 18 and 40 L h⁻¹), as well as the organic phase nature, the organic phase concentration (between 0 and 1% (vol)) and the surfactant presence (between 0 and 0.1% (vol))] have been taken into account. The results obtained have been explained on the basis of the liquid phase physical properties and the special characteristics of the gas–liquid–liquid systems.
Kinetics of absorption of CO<inf>2</inf> in amino-functionalized ionic liquids
2011, Chemical Engineering Journal
Citation Excerpt :
The CO2 physical absorption rate is dependent on the agitation rate. In the ionic liquid, the volumetric mass transfer coefficient increased with an increase in the agitation rate as is observed in aqueous systems, where kLa increases with an increase in the agitation speed [47,48]. The mass transfer coefficient (kL) increased with an increase in temperature.
The kinetics of the reaction of CO₂ with a primary amine functionalized ionic liquid (NH₂-RTIL) at 303 K and 333 K are studied in this work. The absorption of CO₂ was carried out by the decreasing pressure method in a gas–liquid stirred cell reactor. The volumetric mass transfer coefficient of the liquid phase was determined from the experiments using [bmim]BF₄ as liquid phase and the kinetics of the reaction was studied based on the experiments carried out with a liquid phase containing [bmim]BF₄ and 1-(3-aminopropyl)-3-methylimidazolium tetrafluoroborate ([APmim]BF₄) at concentrations between 45 and 253 mol m⁻³. The enhancement factor due to the chemical reaction was calculated from the fluxes of CO₂ absorbed. The kinetic results were analyzed using the general solution given by Van Krevelen and Hoftijzer based on the film model and the approximate solution of DeCoursey based on the Danckwerts surface renewal model. The results indicate that the reaction takes place in an intermediate regime limited by diffusion of the amine functionalized ionic liquid. The reaction was assumed first order in both CO₂ and [APmim]BF₄ and the calculated kinetic constants (k_1,1) are of the same order of magnitude as the ones available for primary amines and CO₂ in viscous media.
Oxygen mass transfer in a biphasic medium: Influence on the biotransformation of methyl ricinoleate into γ-decalactone by the yeast Yarrowia lipolytica
2007, Biochemical Engineering Journal
In this work, an empirical correlation is proposed to describe k_La as a function of operating conditions (agitation and aeration rates) and of oil and surfactant volumetric fractions in a biotransformation medium, an oil-in-water dispersion. An interaction effect between the oil and the surfactant effects was found, since oil presence increased k_La in the absence of the surfactant but had an opposite effect when Tween 80 was available in the medium. The biotransformation of methyl ricinoleate (MR) into γ-decalactone (an aroma compound of industrial interest), by the yeast Yarrowia lipolytica, was carried out at different conditions of operation, to evaluate the influence of k_La on the production of the aroma. It was demonstrated that k_La had an influence on the aroma production; however, for the low hydrophobic substrate concentration used (1.08% v/v) and cellular density of 2.0 × 10⁷ cells mL⁻¹, a minimal k_La value of 70 h⁻¹ was necessary to attain the maximal aroma production.
Validation of the Danckwerts-plot technique by simultaneous chemical absorption of CO<inf>2</inf> and physical desorption of O<inf>2</inf>
2005, Chemical Engineering Science
The Danckwerts-plot technique [Danckwerts et al., 1963. Chemical Engineering Science 18, 63–72] is used in chemical engineering to simultaneously obtain the mass transfer parameters, $k_{L}$ and a, from mass transfer experiments. This method requires variation of the reaction kinetics by adding different amounts of catalyst. Although the method is known for several decades, it was never verified that the variation of the amount of catalyst does not affect the hydrodynamics of the system under investigation. To study this, absorption of ${CO}_{2}$ in a carbonate/bicarbonate buffer solution was performed simultaneously with desorption of oxygen from this solution, after verification that absorption and desorption are processes, taken place at identical rates, but in a different direction. It was shown that the addition of catalyst did not affect the desorption rate of oxygen. The obtained $k_{L} a$ for oxygen was, however, 64% higher compared to the $k_{L} a$ of carbon dioxide. This was probably due to a lower effective interfacial area, caused by the complete depletion of small bubbles containing ${CO}_{2}$ . Mass transfer experiments with oxygen, with a low gas phase conversion, are therefore to be preferred, as the measured mass transfer parameters are less affected by the gas phase RTD and the shape of the bubble size distribution.

View all citing articles on Scopus

View full text

Mass transfer in mechanically agitated gas—liquid contactors

Abstract

Résumé

Zusammenfassung

Chem. Engng Sci.

Chem. Engng Sci.

Chem. Engng Sci.

Br. Chem. Engng

Chem. Engng Sci.

Trans. Faraday Soc.

Chem. Engng Sci.

A.I.E.Ch.Jl

The oxidation of Aqueous sulphite solutions: A model reaction for measurements in gas—liquid dispersions

Chem. Engng Sci.

Mass transfer with chemical reaction

Trans. Instn chem. Engrs