Ind. Eng. Chem. Res. 2008, 47, 171-175
171
Mass Transfer Coefficient in Liquid-Liquid Extraction and the Influence of Aqueous Phase pH Javad Saien* and Shabnam Daliri Department of Applied Chemistry, UniVersity of Bu-Ali Sina, Hamedan 65174, Iran
The influence of aqueous phase pH on the rate of liquid-liquid extraction was investigated, using single drop experiments for a range of drop size and in either of the mass transfer directions. The chemical system of toluene-acetone-water was employed with the advantage of high accuracy and repeatability when analyzed with gas chromatography. The pH of continuous aqueous phase was adjusted within 5.5-8, a range in which most industrial waters are lain. An extraction column was used to introduce the drops into the aqueous phase. Results have demonstrated the impact of pH on the mass transfer performance giving a significant reduction in mass transfer with increasing pH and nearly independent of the kind of accompanied cation for two utilized bases. This matter can be attributed to the adsorption of hydroxyl ions from the aqueous phase which causes small drops to be more prone to this effect. The rate of mass transfer is greater for the dispersed to continuous phase direction under similar conditions. Introduction
Table 1. Physical Properties of Chemical System Toluene-Acetone-Water at 20 °C
Liquid-liquid extraction is one of the separation methods that have found a vast application in various fields such as the food industry, pharmaceuticals, nuclear fuels, and biochemistry. The design of contactors is often connected with expensive experimental investigations, and therefore, studies on the pilot plants and laboratory-scale apparatuses are inevitable. To date it is still difficult to describe the rate of mass transfer as a function of process and physical parameters. Additionally, in industrial applications, impurities or the solute to be transferred from one phase to another causes a change of ionic strength or pH. In the past this led to a lack of reproducibility as this parameter influences the drop size distribution significantly.1-3 In a number of so far recommended chemical systems for liquid-liquid extraction investigations,4,5 water has been selected as the continuous phase. Also, in many industrial processes, at least one of the phases in the liquid-liquid extraction systems is aqueous. This is because the choice of water could be accompanied by a range of pH values due to different sources. Recently, the influence of pH has been considered with respect to the drop size distribution in stirred liquid-liquid dispersions.1 The increase of pH can exert an interfacial barrier layer due to the adsorption of hydroxyl ions and lead to hydrodynamic revolutions in lowering interfacial mobilization, which in turn causes changes in the internal circulation of drops. The influence of some electrolytes and charged species and surface potential has also been demonstrated in many publications,6-8 and the strong influence of electrolytes on the partition behavior of charged species has been reported. The aim of the present work is to investigate the influence of the pH of the aqueous phase on the rate of mass transfer for single drops, moving in the extraction column. The results can help to find a suitable pH, in order to obtain higher efficiencies. Experimental Section Materials. The chemical system of toluene-acetone-water, a recommended system for liquid-liquid extraction studies,4 * To whom correspondence should be addressed. Tel. and fax: +98811-8257407. E-mail:
[email protected].
property F [kg/m3] µ [kg/m‚s] D [m2/s] γ [mN/m]
dispersed phase 866.15 0.628 × 10-3 2.55 × 10-9 29.98-32.79
continuous phase 998.15 (0.990-1.025) × 10-3 1.09 × 10-9 29.98-32.79
was chosen. The main specification of this system is its high interfacial tension. Toluene and acetone were Merck products with purities of more than 99% and 99.5%, respectively. Distilled water with an initial pH of 5.5 was produced from a special water source and was used as the aqueous continuous phase. To adjust the pH values, a 0.1 M solution of NaOH (supplied by Merck) was used to reach the desired pH in the aqueous phase. The physical properties of the chemical system, including the range of the continuous phase viscosity and interfacial tension (with respect to variation in pH) at 20 °C (ambient temperature) are given in Table 1. The physical properties were measured using a self-adjusting temperature density meter (Anton Paar DMA 4500) for measuring density, using an Ostwald viscometer for measuring viscosity, and employing a drop-weight method for measuring interfacial tension.9 Experiments for measuring interfacial tension at different pH values showed that the interfacial tension tends to decrease significantly by increasing the pH (Figure 1). Regarding the important role of viscosity in mass transfer, this property was measured for the continuous phase at different pH values. The variation in viscosity was negligible, since it decreased only 3% within the range used. The influence of pH variation on the density of the aqueous phase was also negligible. Organic and aqueous molecular diffusivities can be obtained from those reported by Baldauf and Knapp.10 The values of the equilibrium distribution of acetone between the phases at 20 °C and at different pH values were examined and correlated by
Cd* ) 0.839Cc
(1)
where Cd* and Cc are equilibrium dispersed phase (organic) and continuous phase (aqueous) concentrations of acetone (g/ L), respectively. The change of pH in the continuous phase, within the range used, shows no significant influence on the
10.1021/ie0704086 CCC: $40.75 © 2008 American Chemical Society Published on Web 12/01/2007
172
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008
Figure 1. Variation of interfacial tension with pH for toluene (containing a typical acetone concentration of 10 mg/L) + water at 20 °C. Table 2. Ranges of Drop Diameter (mm), Generated by Nozzles at Different pH Values nozzle no.
cfd
dfc
1 2 3 4 5
2.77-3.32 2.89-3.46 3.07-3.61 3.36-3.94 3.45-4.21
2.57-3.19 2.79-3.40 3.03-3.51 3.14-3.90 3.32-4.02
equilibrium distribution. The equilibrium data are in agreement with those reported by Brodkorb et al.,11 which include data in either the presence or absence of phenol. Setup. A Pyrex glass column (11.4 cm diameter and 51 cm height) was used as the contactor. The details are given in our previous work.12 Drop forming was provided, using a variety of glass nozzles located at the bottom of this column. The toluene phase was held in a glass syringe, conducted by an adjustable syringe pump (Phoenix M-CP) and flowed through a rigid tube to the glass nozzle. A small inverted glass funnel attached to a pipet and vacuum bulb was used to catch a sample of about 1 mL of dispersed phase at the top of the column with 33 cm distance from the initial point. The interfacial area in the funnel was minimized at the level of the pipet inlet by occasionally pulling toluene into the pipet. Three samples were prepared for each concentration and were kept in closed sample tubes at a cold medium until analysis by gas chromatography. In order to omit the influence of unsteady mass transfer during the drop formation and its transient velocity, the initial drop concentration was considered for a location of moving drops to be at 6.5 cm above the nozzles’ tips. Drop motion was observed to reach steady movement after about 40 mm travel. Slater et al.13 have reported a distance of about 30 mm for the chemical system of cumene-acetic acid-water. To determine the initial concentrations at this location, an empty small column equipped with the same nozzle was used. Drops were collected at the same distance of 6.5 cm and under the same conditions and drop sizes as the main column. Procedure and Analysis. The syringe and the connection tube to the nozzle tip were first filled with toluene or toluene + acetone (depending on the mass transfer direction) to produce drops. The column was then filled with distilled water, adjusted to the appropriate pH, as the continuous phase. The measurement of pH values was performed with a CORNING-M140 pH meter, having an uncertainty of (0.1. The syringe pump was initially calibrated with respect to the specified volume scale on the calibrated syringe. Knowing the flow rate and the number of generated drops, during a particular period, the drop size can easily be determined. Toluene drop
Figure 2. Variation of generated drop sizes with pH for different nozzles in both directions.
diameters from 2.57 to 4.21 mm were generated. The range of drop sizes, generated by each nozzle when pH varies, is listed in Table 2 for both directions of mass transfer. Figure 2 shows that the size of each drop, formed at the nozzles’ tips, was influenced by increasing pH, mainly due to the decrease of interfacial tension.9 Larger drop sizes were obtained when the direction of mass transfer was from the continuous to the dispersed phase and variation in drop size was achieved by using different glass nozzles. The outside diameter of the nozzles was typically on the order of 0.3 mm. The contact time of drops from the initial to the collection point was measured several times with a stopwatch, and its average was considered. The terminal velocity of drops was obtained with respect to the average times. Drops were spaced more than 60 mm apart typically with a range of flow rate between 20.54 and 120.23 mL/h. Skelland and Vasti14 have shown that interactions are negligible for this distance. Acetone concentration in the collected samples was measured using a gas chromatograph (Shimadzu, 14B) with a flame ionization detector, calibrated with reference substances of toluene and acetone (Merck) for gas chromatography. The concentration of acetone in the known samples was within the range of 1.5-44.7 g/L, and the number of known samples, used in the calibration curve experiments, was 14. All the equipments and glassware were cleaned with Decon 90 solution, followed by several rinses with distilled water, prior to experiments. The ambient temperature was within 20 ( 2 °C. Results and Discussion The range of drops used in this work is within the conditions of circulating drops since the values of the dimensionless group H defined in the model by Grace et al.15 are in the range of 13.9-35.5 (2 < H < 59.3), corresponding to circulating drops. The range of the drop Weber number is also within 0.4 < We < 2 (less than 3.58) and the range of the ratio Re/NPG0.15 (NPG is the inverse of the Morton number) is 4.93-14.28 (less than 20).16 Experiments were conducted in both mass transfer directions of dispersed phase to continuous phase (d f c) and vice versa (c f d). Acetone was dissolved in the aqueous or the organic phase with initial concentrations of 43.3 g/L (5%) and 29.9 g/L (3%) for the d f c and c f d directions, respectively. The pH of the aqueous phase was adjusted to 5.5, 6, 6.5, 7, 7.5, and 8 around its neutral value, a range in which most natural and industrial waters are lain. For each series of experimental data, initial and final concentrations, drop size, and contact time were
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 173
Figure 3. Variation of overall mass transfer coefficient vs pH for different nozzles in both directions.
Figure 4. Influence of NaOH and KOH reagents in the reduction of overall mass transfer coefficient, nozzle 1, mass transfer direction of d f c.
obtained. The experimental overall mass transfer coefficient values were calculated using
Kod ) -
d ln(1 - E) 6t
(2)
where t is the contact time and E is the extraction fraction, defined by
E)
Cdf - Cdi Cd* - Cdi
(3)
where Cdi, Cdf, and Cd* are the initial, final, and equilibrium solute concentrations in the drop phase (obtained from eq 1), respectively. For the d f c direction Cd* is zero, since the solute concentration in the aqueous phase is zero for this case. The continuous phase is considered uniform in the bulk phase and thus has a unique bulk concentration. The overall mass transfer coefficient obtained from eq 2 is in fact a time-averaged mass transfer coefficient. In several studies, authors have reported that drops of an organic phase, in contact with water, carry an electrical charge which is usually negative.17,18 It has been presumed that the origin of the negative charge at the organic-water interface is the adsorbed hydroxyl ions from the aqueous phase. With increasing pH, adsorption of hydroxyl ions increases, widening the surface charge layer around the drops.1,18-20 The reduction of interfacial tension with pH, as mentioned in the Experimental Section, can be attributed to this problem. This in turn leads to the reduction of internal circulation for drops of similar size, and as a consequence, the mass transfer rate will be decreased. Figure 3 shows the variation of overall mass transfer with pH of the aqueous phase. The overall mass transfer finds a significant reduction (within 62.7%-82.7%) as the pH increases. The reason can be attributed to the interfacial barrier layer due to the adsorption of hydroxyl ions and to the hydrodynamic revolutions in lowering interfacial mobilization, which in turn cause changes in the internal circulation of drops.17 The acidic pH in the aqueous phase provides a significantly higher rate of mass transfer for a specified drop size, due to the lack of many hydroxyl ions. When the hydroxyl ions are adsorbed directly from the aqueous phase, the number of ions per unit area will depend on the hydroxyl ion concentration in the bulk of the aqueous phase, i.e., on the pH of this phase. As the pH increases, the surface becomes saturated with hydroxyl ions and it would be expected that the thickness of the adsorbed layer is maintained almost constant. A relatively constant mass transfer coefficient could be expected under these conditions; however, that is not
Figure 5. Variation of overall mass transfer coefficient with drop size at different pH values in both directions.
the case for the experiments in this work, since the aqueous phase pH was adjusted just around the neutral water pH. An important observation, revealed from Figure 3, is that Kod reduction (with respect to its highest value at pH 5.5) is not the same for the nozzles. Small drops are more prone to the retarding effect of pH. The higher contact times for small drops which in turn provide higher adsorption of hydroxyl ions can be a reason for this effect. The influence of the accompanied cation in the aqueous phase was also investigated, using KOH reagent instead of NaOH. As presented in Figure 4, no significant change of the overall mass transfer coefficient, within the used pH range, is observed except for a minor shift to lower values. The role of hydroxyl ions is confirmed here as the main influencing parameter. Recently, a nearly close influence of the above used reagents on stability of extracting emulsions toward coalescence has been reported.21 As Figure 5 shows, the mass transfer coefficient increases with drop size. Drops tend to higher internal circulation as the size increases, leading to enhancement in mass transfer. Meanwhile, large drops move faster with lower contact times. It is obvious from Figures 3 and 5 that the mass transfer coefficient is greater in the d f c direction under the same conditions. Similar results have been reported for the influence of different surfactants.12 The range of Kod values is about 24.43-151.83 µm/s for the d f c direction and about 12.27134.39 µm/s for the c f d direction. To explain the difference in mass transfer rate with respect to directions, investigations in contact time (hydrodynamic term) and the extraction fraction (mass transfer term), included in eq 2, become useful. Figures 6 and 7 present the variations of
174
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008
H ) a dimensionless group defined in the Grace et al.15 model K ) overall mass transfer coefficient (µm/s) NPG ) inverse of Morton number (Fc2γ3/gµc4∆F) Re ) drop Reynolds number (FcVtd/µc) t ) contact time (s) Vt ) terminal velocity of a drop (m/s) We ) drop Weber number (FcVt2d/γ) Greek Symbols γ ) interfacial tension (mN/m) µ ) viscosity (kg/m‚s) F ) density (kg/m3) Subscripts Figure 6. Variation of contact time of drops with pH for different nozzles in both directions.
c ) continuous phase d ) dispersed phase i ) initial value f ) final value o ) overall value Superscript * ) equilibrium Literature Cited
Figure 7. Variation of extraction fraction with pH for different nozzles in both directions.
contact time and extraction fraction with pH. As is obvious, the extraction fraction is higher for the d f c direction, and this term, as a dominant parameter, would provide the higher mass transfer coefficient. The kind of variation of contact time with pH, presented in Figure 6, is due to the reduction of terminal velocity which is influenced by the reduction in drop size (Figure 2). Conclusions The overall mass transfer coefficient decreases when the aqueous phase pH increases within the important range of 5.58. The drop size also decreases with increasing pH, which can be attributed to the reduction of interfacial tension. Considering a specified drop size, an acidic continuous phase provides better mass transfer results. These observations confirm the adsorption of hydroxyl ions onto the interface and, as a result, retarding the solute mass transfer. The direction of mass transfer itself influences the rate of mass transfer, no matter the variation of pH; however, the trend of variation of the overall mass transfer coefficient is similar. The mass transfer in drops of different size is not equally influenced by pH, and the percentage of reduction is higher for small drops. Nomenclature C ) solute concentration (g/L) d ) drop diameter (mm) D ) diffusivity (m2/s) E ) extraction fraction defined by eq 3
(1) Gabler, A.; Wegener, M.; Paschedag, A. R.; Kraume, M. The Effect of pH on Experimental and Simulation Results of Transient Drop Size Distributions in Stirred Liquid-Liquid Dispersions. Chem. Eng. Sci. 2006, 61, 3018. (2) Kraume, M.; Gabler, A.; Schulze, K. Influence of Physical Properties on Drop Size Distribution of Stirred Liquid-Liquid Dispersions. Chem. Eng. Technol. 2004, 27, 330. (3) Tobin, T.; Ramkrishna, D. Coalescence of Charged Droplets in Agitated Liquid-Liquid Dispersions. AIChE J. 1992, 40, 395. (4) Misek, T.; Berger, R.; Schroter, J. Standard Test Systems for Liquid Extraction Studies. EFCE Publ. Ser. 1985, 46. (5) Bailes, P. J.; Godfrey, J. C.; Slater, M. J. Liquid-Liquid Extraction Test Systems. Chem. Eng. Res. Des. 1983, 61, 321. (6) Czapla, C.; Bart, H. J. Influence of the Surface Potential on Interfacial Kinetics. Ind. Eng. Chem. Res. 2001, 40, 2525. (7) Pfennig, A.; Schwerin, A. Influence of Electrolytes on Liquid-Liquid Extraction. Ind. Eng. Chem. Res. 1998, 37, 3180. (8) Pfennig, A.; Schwerin, A.; Gaube, J. Consistent View of Electrolytes in Aqueous Two-Phase Systems. J. Chromatogr., B 1998, 711, 45. (9) Saien, J.; Akbari, S. Interfacial Tension of Toluene + Water + Sodium Dodecyl Sulfate from (20 to 50) °C and pH between 4 and 9. J. Chem. Eng. Data 2006, 51, 1832. (10) Baldauf, W.; Knapp, H. Experimental Determination of Diffusion Coefficients, Viscosities, Densities and Refractive Indexes of 11 Binary Liquid Systems. Ber. Bunsen-Ges. Phys. Chem. 1983, 87, 304. (11) Brodkorb, M. J.; Bosse, D.; von Reden, C.; Gorak, A.; Slater, M. J. Single Drop Mass Transfer in Ternary and Quaternary Liquid-Liquid Extraction Systems. Chem. Eng. Process. 2003, 42, 825. (12) Saien, J.; Riazikhah, M.; Ashrafizadeh, S. N. Comparative Investigation on the Effect of Contamination and Mass Transfer Direction in Liquid-Liquid Extraction. Ind. Eng. Chem. Res. 2006, 45, 1434. (13) Slater, M. J.; Baird, M. H. I.; Liang, T.-B. Drop Phase Mass Transfer Coefficients for Liquid-Liquid Systems and the Influence of Packings. Chem. Eng. Sci. 1988, 43, 233. (14) Skelland, A. H. P.; Vasti, N. C. Effects of Interaction Between Circulating or Oscillating Droplets on Drop Formation, Free Fall and Mass Transfer. Can. J. Chem. Eng. 1985, 63, 390. (15) Grace, J. R.; Wairegi, T.; Nguyen, T. H. Shapes and Velocities of Single Drops and Bubbles Moving Freely Through Immiscible Liquids. Trans. Inst. Chem. Eng. 1976, 54, 167. (16) Skelland, A. H. P. Interphase Mass Transfer. In Science and Practice of Liquid-Liquid Extraction; Thornton, J. D., Ed.; Clarendon Press: Oxford, 1992. (17) Marinova, K. G.; Alargova, R. G.; Denkov, N. D.; Velev, O. D.; Petsev, D. N.; Ivanov, I. B.; Borwankar, R. P. Charging of Oil-Water Interfaces Due to Spontaneous Adsorption of Hydroxyl Ions. Langmuir 1996, 12, 2045.
Ind. Eng. Chem. Res., Vol. 47, No. 1, 2008 175 (18) Dickinson, W. The Effect of pH upon the Electrophoretic Mobility of Emulsions of Certain Hydrocarbons and Aliphatic Halides. Trans. Faraday Soc. 1941, 37, 140. (19) Taylor, A. J.; Wood, F. W. The Electrophoresis of Hydrocarbon Droplets in Dilute Solutions of Electrolytes. Trans. Faraday Soc. 1957, 53, 523. (20) Maab, S.; Gabler, A.; Paschedag, A. R.; Kraume, M. Influence of Electrolytes and Turbulence Parameters on Drop Breakage and Drop Size Distributions in Stirred Liquid/Liquid Dispersions. Proceedings of International Conference on Multiphase Flow, ICMF, Leipzig, Germany, 2007; in press.
(21) Koroleva, M. Y.; Yurtov, E. V. Effect of Electrolytes on Stability of Extracting Emulsions. In Proceedings of International SolVent Extraction Conference ISEC 05, Chinese Academic Journal, Electronic Publishing House: Beijing, 2005.
ReceiVed for reView March 20, 2007 ReVised manuscript receiVed September 26, 2007 Accepted October 2, 2007 IE0704086
