2822
Langmuir 2005, 21, 2822-2826
An Experimental and Theoretical Study of Competitive Adsorption at the n-Heptane/Water Interface Dallas B. Warren,*,†,‡ Franz Grieser,§ Jilska M. Perera,† and Geoff W. Stevens† Particulate Fluids Processing Centre, Department of Chemical and Biomolecular Engineering, The University of Melbourne, Melbourne, Victoria, 3010, Australia, and Particulate Fluids Processing Centre, School of Chemistry, The University of Melbourne, Melbourne, Victoria, 3010, Australia Received August 5, 2004. In Final Form: November 30, 2004 A model to calculate the interfacial concentration of competing surface active species in a two-phase oil/water system was developed. To enable the calculation of the surface excess of 2-hydroxy-5nonylacetophenone oxime (HNAPO, active ingredient of LIX 84) in the presence of surfactants competing for interfacial area, an interfacial adsorption competition model was derived for noninteracting surface active species in a n-heptane/aqueous system, assuming ideal enthalpy and entropy of mixing. The model was found to be valid for HNAPO in the presence of sodium dodecyl sulfate (SDS) or dodecyldimethyl(3-sulfopropyl)ammonium (DDSA). In the case of dodecyltrimethylammonium chloride (DTAC) or octa(ethylene glycol) mono-n-dodecyl ether (C12E8) as the competing surfactants with HNAPO, the predicted surface excess values from the model fit less favorably. The difference was shown to not be due to nonideal entropy of mixing.
Introduction Solvent extraction is an important separation process that is utilized in a variety of industries, ranging from the hydrometallurgical to the pharmaceutical. The process involves the contact of two immiscible liquid phases in order to move a particular desired compound, or solute, from one phase into the other. Since the phases are immiscible and some of the dissolved species present are surface active, the adsorption and interactions that occur within the interfacial zone are important to the overall process. The adsorption of the surfactant molecules at an oil/ water interface for a single surfactant system can be described by a Langmuir isotherm.1,2 By use of the Gibbs isotherm and the interfacial tension, the concentration of species located at the interface can be calculated.3,4 The Langmuir isotherm has been extended to treat a binary surfactant mixture using two techniques, assuming no interactions between the different surfactant molecules,5,6 minimum area differences between the surfactant molecules (nonideal entropy of mixing),5,7,8 or application of * Corresponding author: Phone: +61 3 9903 9083. Fax: +61 3 9903 9638. E-mail address:
[email protected] (D. B. Warren). † Particulate Fluids Processing Centre, Department of Chemical and Biomolecular Engineering, The University of Melbourne. ‡ Present address: Department of Pharmaceutical Biology and Pharmacology, Victorian College of Pharmacy, Monash University, Parkville, Victoria 3052, Australia. § Particulate Fluids Processing Centre, School of Chemistry, The University of Melbourne. (1) Fainerman, V. B.; Zholob, S. A.; Miller, R. Langmuir 1997, 13, 283. (2) Hansen, F. K.; Fagerheim, H. Colloids Surf., A 1998, 137, 217. (3) Danesi, P. R. Solvent extraction kinetics. Principles and practices of solvent extraction; Rydberg, J., Musikas, C., Choppin, G. R., Eds.; Marcel Dekker: New York, 1992; pp 157-207. (4) Prochaska, K.; Alejski, K.; Szymanowski, J. J. Chem. Technol. Biotechnol. 1994, 60, 195. (5) Lucassen-Reynders, E. H. Colloids Surf., A 1994, 91, 79. (6) Campanelli, J. R.; Wang, X. J. Colloid Interface Sci. 1999, 213, 340. (7) De Feijter, J. A.; Benjamins, J.; Tamboer, M. Colloids Surf. 1987, 27, 242.
the regular solution theory.9 A molecular thermodynamic theory has also been developed for a binary system that will determine the interfacial concentrations below the critical micelle concentration.10 The aim of this study was to obtain a model to enable the calculation of the interfacial concentration of competing surface active components in a two-phase oil/water system. 2-Hydroxy-5-nonylacetophenone oxime (HNAPO), a complexing agent commonly used to extract nickel(II),11 has been utilized to observe the effects of various measurable interfacial properties on metal ion extraction kinetics in a n-heptane/aqueous system.12 This model enables the determination of the concentration of the extractant, HNAPO, in the presence of surfactants competing for interfacial area.12 Adsorption Model To calculate the surface excess when there are two competing surface active species present, an equation describing surface competition was derived for a twosurfactant system based on the kinetics of adsorption and desorption of species at an interface.7 The rate of adsorption of surfactant 1, υ1(abs), at the interface is proportional to the volume fraction of surfactant (φ ) surfactant volume, from the density and mass, divided by the total bulk phase volume) and the surface fraction (θ ) surface excess (Γ) divided by the Γmax, available for the surfactant to adsorb onto) as given by eq 1
υ1(abs) ) a1φ1(1 - θ)
(1)
where a1 is a proportionality constant and 1 - θ is the surface fraction unoccupied by surfactants 1 and 2. The (8) Fainerman, V. B.; Lucassen-Reynders, E. H.; Miller, R. Colloids Surf., A 1998, 143, 141. (9) Staples, E.; Thompson, L.; Tucker, I.; Penfold, J.; Thomas, R. K.; Lu, J. R. Langmuir 1993, 9, 1651. (10) Mulqueen, M.; Blankschtein, D. Langmuir 2001, 18, 365. (11) Hosking, J. W.; Rice, N. M. Hydrometallurgy 1978, 3, 217. (12) Warren, D. B.; Grieser, F.; Perera, J. M.; Stevens, G. W. In preparation.
10.1021/la048014w CCC: $30.25 © 2005 American Chemical Society Published on Web 02/24/2005
Competitive Absorption
Langmuir, Vol. 21, No. 7, 2005 2823
Table 1. Interfacial Data for Hydroxyoxime Extractants and Some Metal Complexes in Oil/Water Systems species
organic phase
aqueous phase
HNAPO HNAPO HNAPO HNAPO HNAPO HNAPO HNBPOb HNBPO Ni(NAPO)2 Ni(NAPO)2 Cu(NAPO)2 Cu(NBPO)2
n-heptane n-heptane n-heptane MSB 210 Dispersol Benzene n-heptane n-heptane n-heptane MSB 210 MSB 210 n-heptane
0.01 M HClO4/triethanol amine NaClO4/acetate buffer HClO4/NaClO4 NH3/NH4NO3 H2SO4/Na2SO4 HClO4/NaClO4 HClO4/NaClO4
a
0.01M HClO4/triethanol amine NH3/NH4NO3 NH3/NH4NO3
Γmax × 106 (mol m-2)
min area per molecule (Å2)
4.0 2.0 4.0 2.1 3.3 1.8 2.5 4.0 1.6 1.4 1.5 1.2
42 80 42 79 50 92 66 42 104 120 110 140
K
ref
1.5 × 104
a 19 20 21 22 23 24 21 a 21 21 21
8.0 × 103
This work. b 2-Hydroxy-5-nonylbenzophenone oxime.
rate of desorption, υ1(des), at the interface is proportional to the surface fraction of surfactant and the volume fraction available for the surfactant to desorb into, as represented by eq 2
υ1(des) ) b1θ1(1 - φ1)
(2)
where b1 is a proportionality constant and 1 - φ1 is the volume fraction of the bulk phase unoccupied by surfactant 1. Under equilibrium conditions the rate of adsorption is equal to the rate of desorption; therefore eqs 1 and 2 can be equated. This results in eq 3, a relationship that provides the surface fraction for each surfactant
θ1 ) K1φ1
(1 - θ) ; (1 - φ1)
θ2 ) K2φ2
(1 - θ) (1 - φ2)
(3)
where K1 is a1/b1 and K2 is a2/b2. The total surface fraction is the sum of the surface fraction for each surfactant, eq 3, which becomes eq 4 for a two-surfactant system.
θ)
K1φ1(1 - φ2) + K2φ2(1 - φ1) (1 - φ1)(1 - φ2) + K1φ1(1 - φ2) + K2φ2(1 - φ1) (4)
If only one surfactant is present in the system and the limit of small volume fraction is taken, i.e., 1 . φ, then eq 4 reduces to the Langmuir isotherm. Substituting eq 4 into eq 3 and solving for θ1 results in eq 5, allowing the prediction of the surface fraction of surfactant 1 when competing with surfactant 2. The same equation can be derived for surfactant 2, eq 6.
θ1 )
Γ1 ) Γmax1 K1φ1(1 - φ2)
(1 - φ1)(1 - φ2) + K1φ1(1 - φ2) + K2φ2(1 - φ1) θ2 )
Γ2 Γmax2
(5)
(SDS).13 The constants K and Γmax are determined for each surface active species in a single-component system using a linearized Langmuir isotherm, as indicated by eq 7.
1 1 K-1 ) + Γ KΓmaxφ KΓmax
(7)
Experimental Section Chemicals. Dodecyl dimethyl(3-sulfopropyl)ammonium (DDSA) and nickel perchlorate hexahydrate (analytical reagent grade) were obtained from Aldrich Chemical Co. Inc., USA. Sodium dodecyl sulfate (SDS) and perchloric acid of analytical reagent grade were from BDH Limited, England. Dodecyltrimethylammonium chloride (DTAC) was from TCI, Japan, and octa(ethylene glycol) mono-n-dodecyl ether (C12E8) was from Nikko Chemicals Company Ltd., Japan. HNAPO was obtained by purification from an industrial sample of LIX 84i (courtesy of Cognis Corporation), as described previously.13 The aqueous phase was buffered with 0.01 M perchloric acid/triethanol amine (analytical reagent grade, Fisons AAG Pty. Ltd, Australia), as described previously.14 All chemicals were used as supplied, with the exception of HNAPO. Interfacial Tension. The interfacial tension between the n-heptane and aqueous phase was determined using a FTA° 200 Interfacial Tension Measurement Device (First Ten A° ngstroms, USA). The instrument software finds the edge of a drop from an CCD image, determines the drop profile, and calculates the interfacial tension ((0.5 mN m-1) using the Bashforth-Adams technique.15,16 Interfacial tension measurements were performed in a quartz cell with n-heptane (0.684 g cm-3 17) as the continuous phase and the aqueous phase (0.998 g cm-3 18) forming a pendent drop from the syringe needle tip (diameter 0.635 mm). All equipment exposed to either phase was thoroughly washed in hot water and then acetone to remove any possible source of surface contamination. The diameter of the pendent drop was made as large as possible (>3 mm) to provide an exaggerated profile that permits improved profile fitting. Due to the fact that surface active species were present, time was allowed for the species to diffuse to the interface and reach an equilibrium concentration. A period of 400 s was found to be suitable to allow equilibrium to be reached. All measurements were carried out at 20 °C.
Results and Discussion
) K2φ2(1 - φ1)
(1 - φ1)(1 - φ2) + K1φ1(1 - φ2) + K2φ2(1 - φ1)
From the interfacial tension measurements of HNAPO and Ni(NAPO)2 in a n-heptane/aqueous two-phase system,
(6)
This surface excess model assumes that the enthalpy and entropy of mixing are ideal; i.e., there is no interaction between the two competing species. Previously, no interaction was evident between HNAPO and the surfactants utilized in this study using NMR: n-dodecyldimethyl3-ammonio-1-propanesulfonate (DDSA), dodecyltrimethylammonium chloride (DTAC), and sodium dodecyl sulfate
(13) Warren, D. B.; Dyson, G.; Grieser, F.; Perera, J. M.; Stevens, G. W.; Rizzacasa, M. A. Colloids Surf., A 2003, 227, 49. (14) Warren, D. B.; Grieser, F.; Perera, J. M.; Stevens, G. W. Submitted for publication in Colloids Surf., A. (15) Bashforth, F.; Adams, J. C. An attempt to test the theory of capillary action; Cambridge University Press: London, 1883. (16) Paddy, J. F. Surface and colloid science; Matijevic, E., Ed.; Wiley: New York, 1969; Vol. 1. (17) Perry’s chemical engineers’ handbook, 6th ed.; Perry, R. H., Green, D. W., Maloney, J. O., Eds.; McGraw-Hill: Sydney, 1984. (18) Rogers, G. F. C.; Mayhew, Y. R. Thermodynamic transport properties of fluids; Basil Blackwell: Cambridge, 1991; SI units.
2824
Langmuir, Vol. 21, No. 7, 2005
Warren et al.
Table 2. Interfacial Data for the Surfactants DTAC, SDS, C12E8, and DDSA at the Air/Water and n-Heptane/Water Interface
a
surfactant (aqueous phase)
second phase
Γmax × 106 (mol m-2)
min area per molecule (Å2)
DTAC DTAC SDS SDS SDS C12E8 G12A8 DDSA DDCAb
n-heptane air n-heptane n-heptane air n-heptane air n-heptane air
2.4 4.3 2.9 3.1 3.1 3.2 2.5 2.9 3.5
70 38 58 53 53 51 66 57 46
K × 10-5
ref
3.5
a 25 (T ) 25 °C) a 26 (T ) 20 °C) 27 (T ) 25 °C) a 28 (T ) 25 °C) a 29 (T ) 23 °C)
1.5 7.1 3.8
This study. b C12H25N+(CH3)2CH2COO-.
Figure 1. Plot of 1/Γ versus 1/φ for the n-heptane/aqueous system with the addition of HNAPO for the determination of the adsorption isotherm constants, K and Γmax. Insert indicates the interfacial tension data for the n-heptane/aqueous system with the addition of HNAPO. Aqueous phase of 0.01 M HClO4/ triethanol amine at pH 7.4.
Γmax and K were determined using eq 7 as 4.0 × 10-6 mol m-2 and 1.5 × 104 for HNAPO and 1.6 × 10-6 mol m-2 and 8.0 × 103 for Ni(NAPO)2, respectively. The interfacial tension variation with the addition of HNAPO is shown in Figure 1. The results obtained, along with other published values, are shown in Table 1. Γmax for HNAPO determined in this study is at the upper range of published values, with a reasonable area per molecule. Harada et al.20 justified the validity of an area per molecule of 40 Å2 from theoretical calculations of the cross-sectional area of an o-hydroxy aryl group. By use of the Chorey-PaulingKoltun molecular model, a cross-sectional area in the plane of the aryl ring was calculated as 45 Å2, and perpendicular to the plane, as 35 Å2.20 Interfacial tension measurements were also performed for the surfactants DTAC, SDS, C12E8, and DDSA in a n-heptane/aqueous two-phase system. The values of Γmax and K determined are summarized in Table 2, along with published data for comparison. (19) Watarai, H.; Takahashi, M.; Shibata, K. Bull. Chem. Soc. Jpn. 1986, 59, 3469. (20) Harada, M.; Miyake, Y.; Kayahara, Y. J. Chem. Eng. Soc. Jpn. 1989, 22, 168.
Figure 2. Surface excess of SDS, experimental (O) and calculated (s), versus bulk aqueous SDS concentration with HNAPO present. Organic phase of 7.5 × 10-5 M (φ ) 2.8 × 10-5) HNAPO in n-heptane and aqueous phase of 0.01 M HClO4/ triethanol amine at pH 7.4.
A consequence of the Gibbs adsorption equation is that the surface excess is proportional to dγ/d ln c, where c is the bulk phase concentration of the species. Therefore, even in a system with two surface active species present, it is possible to calculate the surface excess of one species if the total concentration of the second is constant.30 This enables a check on the validity of the two-surfactant (21) Inoue, K.; Tsunomachi, H.; Maruuchi, T. J. Chem. Eng. Jpn. 1986, 19, 131. (22) Miyake, Y.; Takenoshita, Y.; Teramoto, M. J. Chem. Eng. Jpn. 1983, 16, 203. (23) Miyake, Y.; Imanishi, Y.; Katayama, Y.; Hamatani, T.; Teramoto, M.; J. Chem. Eng. Jpn. 1986, 19, 117. (24) Watarai, H.; Satoh, K. Langmuir 1994, 10, 3913. (25) Caskey, J.; Barlage, W. B., Jr. J. Colloid Interface Sci. 1971, 35, 46. (26) van Voorst Vader, F. Trans. Faraday Soc. 1960, 56, 1067. (27) Dahanayake, M.; Cohen, A. W.; Rosen, M. J. J. Phys. Chem. 1986, 90, 2413. (28) Rosen, M. J.; Cohen, A. W.; Dahanayake, M.; Hua, X. Y. J. Phys. Chem. 1982, 86, 541. (29) Beckett, A. H.; Woodward, R. J. J. Pharm. Pharmacol. 1963, 15, 422. (30) Rosen, M. J. Surfactants and interfacial phenomena, 2nd ed, Wiley: Brisbane, 1989.
Competitive Absorption
Figure 3. Surface excess of DTAC, experimental (O) and calculated (s), as a function of the bulk aqueous DTAC concentration with HNAPO present. Organic phase of 7.5 × 10-5 M (φ ) 2.8 × 10-5) HNAPO in n-heptane and aqueous phase of 0.01 M HClO4/triethanol amine at pH 7.4.
Figure 4. Surface excess of C12E8, experimental (O) and calculated (s), as a function of the bulk aqueous C12E8 concentration with HNAPO present. Organic phase of 7.5 × 10-5 M (φ ) 2.8 × 10-5) HNAPO in n-heptane and aqueous phase of 0.01 M HClO4/triethanol amine at pH 7.4.
adsorption equation developed above, eq 5, with experimental data. The calculated surface excess for each surfactant in the presence of HNAPO was ascertained using eq 5, along with the appropriate values of Γmax and K, and compared to the experimental values. The calculated values are plotted (denoted as a line) alongside the
Langmuir, Vol. 21, No. 7, 2005 2825
Figure 5. Surface excess of DDSA, experimental (O) and calculated (s), as a function of the bulk aqueous DDSA concentration with HNAPO present. Organic phase of 7.5 × 10-5 M (φ ) 2.8 × 10-5) HNAPO in n-heptane and aqueous phase of 0.01 M HClO4/triethanol amine at pH 7.4.
Figure 6. Surface fraction of DTAC, calculated using (s) eq 5, (- - -) eq 8 with water as the replaced solvent, and (‚ ‚ ‚) eq 8 with heptane as the replaced solvent, as a function of the bulk aqueous DTAC concentration with HNAPO present. Organic phase of 7.5 × 10-5 M HNAPO (φ ) 2.8 × 10-5).
experimental data (denoted by open circles) in Figures 2, 3, 4, and 5 for SDS, DTAC, C12E8, and DDSA, respectively. For SDS and DDSA the experimental surface excess is satisfactorily predicted, with some underestimation at low concentrations in the case of SDS. However, the competing
2826
Langmuir, Vol. 21, No. 7, 2005
Warren et al.
surfactants DTAC and C12E8 display behavior that does not strictly follow the predicted values as well. DTAC’s calculated surface excess has the correct profile, except the values are higher across the entire concentration range indicating less DTAC is absorbed to the interface. C12E8, on the other hand, exhibits a slightly flatter curve shape than experimentally determined, with the C12E8 molecules being absorbed to the n-heptane/water interface more strongly. As noted above, previous work with HNAPO in a micellar system in the presence of these same surfactants displayed no indication of any interactions.13 A check on the validity of the model’s assumption of ideal entropy of mixing was performed by comparing it with the model presented by Fainerman et al.,8 shown in eq 8
Kiφie(ni-1) )
θi (1 - θ1 - θ2)ni
(8)
where ni is the ratio of the surfactant molar surface area to that of the replaced solvent. The molar surface areas of the surfactants were taken from the results obtained in this study, Tables 1 and 2, and 11 and 20 Å2 for water and n-heptane, respectively. For a liquid/gas system, the definition of the solvent molecules being replaced at the surface is simple. However, for a two-liquid system there are two different solvent molecules available at the interface to be replaced by the surfactant molecules. In reality, it will be some combination of the two due to the irregular nature of the liquid/
liquid interface on the molecular scale, therefore, the two extremes of water or n-heptane as the solvent was modeled. Equation 8 was solved numerically for the surface excess of the competing molecules, with the results for DTAC shown in Figure 6. Similar curves were also obtained for the four other surfactant systems. The adsorption isotherm curve when incorporating nonideal entropy of mixing, or different surfactant molecular size, fits the experimental data poorly for the two extremes of water or n-heptane being replaced at the interface. Entropy of mixing appears to play a very minor role in competitive adsorption of the systems studied. This was also found to be the case for surfactants and alcohols on the surface of microbubbles in an aqueous solution.31 Conclusions An interfacial adsorption competition model was derived for noninteracting surface active species in a n-heptane/ aqueous system. The model was found to be valid for HNAPO in the presence of SDS or DDSA at the concentration ranges studied. In the case of HNAPO with DTAC or C12E8, the fit is less favorable; however, this is not due to nonideal entropy of mixing. Acknowledgment. Support from the Australian Research Council and Particulate Fluids Processing Centre is gratefully acknowledged. We also thank Cognis Corporation for the LIX 84i sample. LA048014W (31) Tronson, R.; Ashokkumar, M.; Grieser, F. J. Phys. Chem. B 2003, 107, 7307.
