SciELO - Scientific Electronic Library Online

Home Pagealphabetic serial listing  

Services on Demand




Related links


Journal of the Chilean Chemical Society

On-line version ISSN 0717-9707

J. Chil. Chem. Soc. vol.48 no.1 Concepción Mar. 2003 



Laboratorio de Operaciones Unitarias e Hidrometalurgia
Facultad de Ciencias Químicas y Farmacéuticas,
Universidad de Chile, C.C. 233, Santiago 1, Chile
(Received: June 27, 2002 - Accepted: December 10, 2002)


the influence of a nonionic surfactant compound (sorbitan monooleate) on the coupled transport of Cu(II) ions in a hollow fiber-type solid supported liquid membrane extractor was studied. The acid compound 5-nonylsalicylaldoxime was used as carrier extractant. The experimental results indicate that the surfactant presents a remarkable interfacial activity, higher than those shown by the carrier extractant which would adsorb at the inteface too. From the experimental results it were calculated for the surfactant the interfacial area occupied by mol, the adsorption equilibrium constant and their critical micelle concentration.

Results of extraction equilibrium experiments indicate that Span-80 itself did not extract copper and that Cu(II) extraction is enhanced as the acidity of feed aqueous solution decreases.

It was found that the surfactant accelerates the permeability of metal in runs carried out using a higher concentration of Span-80 and when a pH value of 2.5 was adjusted for feed solution. In such condition transport rate of metal through liquid membrane would be controlled by the diffusion of chemical species in it. Furthermore, apparent permeability of copper is enhanced when the content of extractant at liquid membrane is augmented. A preliminary simplified mechanism of metal permeation through the liquid membrane is proposed. The mechanism is based on an heterogeneous interfacial reaction accelerated in the presence of the nonionic surfactant.

Key words : Liquid membranes, copper, LIX-860 NIC, surfactant


Among the different practical methods to achieve the separation of dissolved compounds from aqueous solutions, the use of a permeation process through "liquid membranes" (LM) have reached an special interest during the last years. Among other fields of application, this technology has come being used with much success by different investigators in environmental sciences, ecology, control of water pollution, analytical chemistry and particularly in hydrometalurgy in processes of separation and concentration of metallic ions 1-3). Practically, LM correspond to an technological advance of the conventional liquid-liquid extraction process (solvent extraction in mixer-settler extractor, industrially known as SX process), as the extraction reaction between the valuable metal and the chosen-selective extractant occurs "on" or "in" a liquid membrane. The fundamental characteristics of LM would be its ability to concentrate traces of metals existent in aqueous solution towards the interior of another one, by means the employment of an organic phase which contains very small concentrations of extractant agents or "carriers" 4,5). Two kinds of liquid membranes are known, both of them presenting a huge potential on industrial applications: surfactant liquid membrane (SLM) also called emulsion liquid membrane and the solid-supported liquid membrane (SSLM). Both types of LM present advantages with respect to conventional SX process, particularly huge savings in inventory of expensive solvent, which it has a positive effect on process economy. SLM would present a bigger contact surface between organic and aqueous phases which assures a better transport of species through the membrane. In turn, SSLM would offer a higher stability of operation.

In recent years we have come developing a number of research works in order to study the extraction and recovery of metals existent in diluted acid aqueous solutions like mine waters and leach residual solution of ores and concentrates, by means of hollow fiber-type solid supported liquid membrane extractors. In these studies we have made some progress not only on the process and mechanism of metals transport on the LM but also on achieving high grade of separation and selective extraction of metallic ions present in treated solutions 6-9).

Nevertheless, this hollow fiber-type membrane presents under certain operation condition a short lifetime. The main reason would be the degradation or unstableness undergone by the organic solution that contains the active carrier extractant of valuable metal, due to its dissolution or entrainment into the feed aqueous solution or eventually to the strip liquor. This difficulty can be easily overcome by choosing appropriate organic solvents highly insoluble at aqueous solutions and those diluents with a low W/O interfacial tension. Other way to avoid this problem is by taking a careful control of volumetric rates of both aqueous phases, i.e., feed and strip liquor, in order to assure a suitable pressure balance at each side of membrane.

On the other hand, it is clear that the permeability rate of metals through the membrane can be improved. A form to obtain it is by means of the addition of substances that promote the extraction reaction, additive that must be soluble in organic phase because its possible recovery from aqueous raffinates is extremely difficult. Some works published by different authors exist up to date. In these studies it is indicated that the use of certain concentrations of some surfactants substances in the preparation of the liquid membrane would increase the transport of metallic ions from the feed aqueous solution towards the receiving one of stripping 10-11). Nevertheless, the great majority of these works are concerned about studies made with pure and ideal solutions prepared at laboratory and in many cases using very pure surfactants compounds synthesized in laboratory and therefore very expensive, impossible to use in practical processes of separation, as for example the use of this membranes technology in hydrometallurgical processes or industrial waste water decontamination treatment.

In this work it is studied the influence to use a commercial nonionic surfactant substance in the preparation of the liquid membrane, on the rate of permeación of Cu(II) through it. A hollow fiber-type membrane extractor is employed and a mine water obtained in the metallurgical mining activity of Chile is used as copper-containing aqueous solution. It is proposed a possible mechanism of action of the surfactant on the coupled transport of the metal with the carrier extractant.



The organic solutions used in this study were prepared by dissolving in the diluent -commercial aviation kerosene (Esso Chile) - the extractant LIX 860-NIC (5-nonilsalicylaldoxime) supplied by Chile Harting-Henkel and the commercial surfactant substance, Span 80 (monooleate of sorbitan), provided by Munnich Pharm. Co, which was utilized such as it was received. This compound corresponds to a yellow liquid that presents a hydrophilics/lipophilics balance of 4.3 12). In Figure 1 the molecular structures of both compounds are shown. As copper-donnor feed solution was used a mine water of El Teniente Division of Codelco-Chile whose composition average is the following one: 0.8 g/L Cu(II), 0.2 g/L Fe, 0.2 g/L Al (III) and minor amounts of other metals. It corresponds to a slightly blue color solution with a pH 2.8, which was filtered previously in order to separate solid particles of greater size than it contained suspended. As stripping agent were used concentrated solutions of H2SO4.

Interfacial Tension Measurements

The measurements of the interface tension, g, between the organic solutions with and without the surfactant (Span-80) and extractant (Lix-860-NIC) compounds and the copper aqueous solutions were made at 30ºC by means of a Dunoy Precision Tensiometer . Diverse determinations were made varying so much the concentration of the surfactant substance like the one of the extractant in the organic phase, looking for to obtain data with respect to the interfacial adsorption equilibrium of both compounds, particularly the type of interaction between the metallic ion Cu(II), the noionic surfactant and the chelate-forming acid extractant.

Equilibrium and Kinetics Extraction Experiments

The experiments of extraction equilibrium were carried out by contacting until to let attain equilibrium, equal volumes of copper-donor mine water solution and the organic solutions. Later was accomplished the separation of phases to determine its copper content by atomic absorption spectrophotometry in an equipment Perkin Elmer 3110. The metal content in aqueous phases were determined directly, however the concentration of copper in the organic phases was determined making previously an "analytical stripping" with HCl 2N. The experiments of copper extraction kinetic were performed in a polytetrafluoroethylene hollow fiber-type liquid membrane extractor. In Figure 2 is shown a scheme of the extractor, whose technical specifications have been described in detail in previous works 6,7,13,14). Previously it was impregnated the porous structure of hollow fibers with the organic solvent made up for the solution of the extractant and the surfactant substance in kerosene. Then, using Masterflex microtube pumps, the feed aqueous solution and the striping aqueous solution were circulated cocurrently through the outer and the inner sides of the hollow fibers, respectively. The volumetric flows of both phases were regulated properly to assure suitable pressures of transport in each side of the liquid membrane to insure by capillarity the retention of the organic solution in the porous structure of hollow fibers used as solid support. Once steady state was attained, samples of both diffusates (raffinate and strip liquor) were taken at intervals to determine their copper concentration by atomic absorption spectrophotometry and to measure the average flow of copper transported in the membrane reactor, JCu, expressed in [ mol Cu/sec.cm2 ].

Fig. 1. Extractant and surfactant molecular structures

Fig. 2. Schematic diagram of liquid membrane extractor


Interfacial activity

The first experiments were designed to measure the effects of the participant substances in the process, (metal and extractant) on the interfacial activity of the surfactant compound. In Figure 3 is observed the relation between the interfacial tension, g, and the concentration of the surfactant compound Span-80, confirming itself from these results, that the sorbitan monooleate presents a remarkable interfacial activity, indicated in the figure by the diminution of the interfacial tension when increasing the concentration of the emusilfier. The strong diminution of g to higher concentrations of the surfactant could be attributed to the formation of inverted micelles when molecules of surfactant are collected or aggregated by itself. From the plot a 3.16.10-4 mol/L (0.135 g/L) critical micelle concentration can be estimated for Span-80. The relation between γ and the amount of surfactant adsorbed in the interface is given by the Gibb´s adsorption equation. If a Langmuir adsorption isotherm is considered between the amount of adsorbed Span-80 and the concentration of this compound in the organic solution, CST, the following equation can be established:

g = go - ( RT / SST ) ln ( 1 + KSTCST) (1)

where g o is the interfacial tension between the diluent kerosene (CST = 0) and the aqueous solution, SST the interfacial area occupied by mol of tensoactive or surfactant (ST) expressed in cm2/mol and KST the adsorption equilibrium constant for the surfactant, measured in cm3/mol. The values of KST and CST were obtained fitting by nonlinear regression the experimental results of interfacial tension by means of equation 1. The following values were determined: KST = 1.15.108 cm3/mol and SST = 3.43.109 cm2/mol. In Figure 3 the continuous solid line was calculated with equation 1 and the obtained values for KST and SST. As the content of the surfactant compound increases in the organic solution, as much the stability of the liquid membrane as its acelerative effect potential on the permeación of the metal must grow. This fact would occur because it should increase the adsorption of the emulsifier in the interface by an increase in the interface film force of surfactant molecules.

Fig. 3. Effect of surfactant concentration on interfacial tension between Kerosene (diluent) and an aqueous solution at pH 2.0.

In Figure 4 the dependency of the interfacial tension of the organic solution based on the concentration of the surfactant and the composition of the participant phases appears. It is observed that the extractant also presents interfacial activity but to a lesser extent that the surfactant substance, which corroborates the results obtained in previous studies of our group with other chelating-type acid extractants 15,16). When coexisting in the organic phase the surfactant substance and the carrier extractant, it obvious happens a natural competence by the adsorption sites. It is clear that a high concentration of the extractant in the organic phase will prevent the adsorption of the emulsifier, reducing or annulling the critical micelle concentration of the surfactant or moving it to greater values. This fact could in theory to be compensated using a greater concentration of the surfactant in the organic film that assured a higher enough interfacial activity to cause the formation of micelles. Nevertheless, this solution could lead to an excessive increase of the liquid membrane viscosity, affecting the diffusion of the species in it, resulting finally a smaller metal permeability. From the figure it is also deduced that the interfacial tension between the solutions is not affected mainly by the presence of the ion Cu(II) in the aqueous phase. In experiments carried out with aqueous solutions that contained higher concentrations of metal, it was observed that the interfacial tension increases slightly with the increase of the copper concentration which would be suggesting a weak interaction between the ion Cu(II) and the noionic surfactant.

Fig. 4.Dependence of interfacial tension on surfactant substance concentration and the composition of aqueous phase.
%Organic phase (Span-80), Aqueous phase (water pH 2.0)
O Organic phase (Span-80) , Aqueous phase (Cu 1 g/L, pH 2.0)
 Organic phase (Span-80; LIX-860 NIC 0.01 M), Aqueous phase (water pH 2,0)

Extraction equlibrium

Previous to the study of the effect of the surfactant on the permeability rate of the metal in the liquid membrane, it require to get information with respect to the influence of this compound on the copper liquid-liquid extraction equilibrium with the extractant LIX-860 NIC. With such purpose a series of experiments was designed and whose results are delivered from now. In previous studies we have indicated that the extraction equilibrium of extraction of metal with the chelating acid extractant can be expressed by means of the following equation 17):

Cu2+ aq + 2 HX org = CuX2 org + 2 H+ aq (2)

where HX represents the acid extractant species, CuX2 its complex with copper and "aq" and "org" denote aqueous and organic phase respectively. From this expression the extraction equilibrium constant, Kex, and the distribution ratio of copper between organic and aqueous solutions, DCu, are defined according to:

Kex = { [CuX2] [H+]2 } / { [Cu2+][HX]2 }
DCu = [CuX2] / [Cu2+]

In Figure 5 is showed the effect of the surfactant substance concentration (Span-80) on the copper extraction equilibrium when LIX-860 NIC is used as extractant. It is clearly appreciated that the surfactant by itself does not extract copper and that even in the presence of the extractant under different conditions of acidity, Span-80 does not have influence on the extraction equilibrium of metal. It is also observed that the extractability by Cu(II) of LIX-860 NIC is greater as the acidity of the metal feed solution decreases. This, in spite of being a strong extractant for copper, theoretically able to extract this metal from strongly acid solutions. These experiments were accomplished with an feed solution that contained 100 mg/L of Cu(II) (1.57.10-3 M), quite similar to the level of content of metal in many industrial waste waters. In Figure 6 it is presented the variation of copper extraction extent as a function of the extractant concentration in an organic phase that contains a Span-80 concentration of 1.10-4 M. In these experiments, the acidity of the aqueous solution was adjusted to a appropriate pH value. It is observed that a slight excess of extractant content with respect to the stoichiometric value assures a good extraction.

Fig. 5. Influence of Span-80 concentration on the distribution ratio of copper.
Initial concentration, Cu(II) = 100 mg/L
 Organic phase (Span-80; LIX-860 NIC 0.01 M), Aqueous phase ( pH 2.5)
0 Organic phase (Span-80; LIX-860 NIC 0.01 M), Aqueous phase ( pH 1.0)
%Organic phase (Span-80), Aqueous phase (water pH 2.5)

Fig. 6. Effect of LIX-860 NIC concentration on the extraction extent of metal.
Aqueous phase: 100 mg/L Cu(II), pH 2.5, Organic phase: 1.10-4 M Span-80.

Metal permeability rate

Several experiments were carried out to study the permeability rate of metal through the liquid membrane by coupled transport. The transported copper flow, JCu, was measured and expressed in [mol Cu(II)/cm2.sec]. From these values the apparent permeability, PCu, was calculated and expressed in [cm/sec], since this parameter reflects in a better form the metal transport liquid membrane. It can be defined as follows:

PCu = JCu / [Cu]o (5)

where the suffix "o" denotes initial concentration of metal. Figure 7 shows the effect of the variation of surfactant concentration on the apparent permeability of metal in the liquid membrane, calculated on the basis of mol of metal effectively transported towards the strip liquor. From the figure it is observed that the metal permeability increases in the experiments carried out at an initial pH value of 2.5 where the rate of metal transport would be controlled by the diffusion of the chemical species 16,18). This effect is particularly appreciated when a higher level of Span-80 concentration is employed in the organic phase. As well, from Figure 7 it is observed that there would not be effect of use of this noionic surfactant on the permeability rate of the metal, in experimental runs brought about at a pH value around 1.0, a range where the carrier extractant does not present its better capacity of extraction of metal. It would be possible to think that in this condition of acidity, the surfactant substance could have an accelerative effect on the mass transference of Cu(II) ions. This, since in this pH range, the extraction rate would be controlled by the interfacial reaction between the metal and the carrier extractant, as it has been described previously. In Figure 8 is showed the effect of the initial concentration of the extractant LIX-860 NIC in the preparation of the liquid membrane on the apparent copper permeability. A positive effect of a higher concentration of the extractant in the liquid membrane is observed when the experiments were carried out under recommended conditions of acidity of fed aqueous phase (pH 2.5). This effect also was observed although in slighter form, when higher contents of the surfactant were used in the organic solution. These results are coherent with the obtained ones in the experiments of extraction equilibrium. It would be possible therefore to think that the surfactant compound, under certain and specific conditions would play a catalytic roll on the permeability rate of metal in the membrane.

Fig. 7. Relationship between apparent permeability of metal and the surfactant concentration in the liquid membrane. Organic phases: 0.01 M LIX-860 NIC
(Aqueous phase 300 mg/L Cu(II), pH 2.5.
!Aqueous phase 1000 mg/L Cu(II), pH 2.5.
%Aqueous phase300 mg/L Cu(II), pH 1.0.
O Aqueous phase 1000 mg/L Cu(II), pH 1.0.

Fig. 8. Dependence of metal apparent permeability on the extractant concentration in the liquid membrane. Cu(II) = 300 mg/L
( Aqueous phase pH 2.5. Organic phase: 0.01 M Span-80
! Aqueous phase pH 2.5. Organic phase: 0.0001 M Span-80
% Aqueous phase pH 1.0. Organic phase: 0.01 M Span-80
O Aqueous phase pH 1.0. Organic phase: 0.0001 M Span-80

Extraction mechanism

In previous studies 15-17) we have proposed an extraction mechanism of metal by acid extractants in hollow fibers-type solid supported liquid membrane reactors based on an heterogenous interfacial reaction model, that assumes like rate-controlling step of the process, the interfacial reaction shown by Eqn. 8 i.e. the formation of the neutral complex CuX2 at membrane/feed aqueous solution interface, due to the hydrophobic character of polytetrafluorothylene support.

HX org = HX ad

Cu2+aq + HX ad = CuX+ ad + H+ aq
CuX+ ad + HX ad = CuX2 org + H+ aq (8)

where the suffix "ad" it indicates adsorbed species at interface. Necessarily it must be included to this mechanism when any surfactant substances is employed, some equation that considers the concentration of Cu(II) ions in the interface due to some kind of interaction of them with the surfactant (ST), according to:

Cu2+ aq + ST ad = Cu - ST2+ ad (9)

An excessive increase of the surfactant in the conformation of the liquid membrane could affect the rate of permeability, probably ought to an indiscriminate enlarge of the organic film viscosity. This increase of viscosity would repercuss in a slower diffusion of metal in LM, what is important if it is considered that the accelerative effect of the surfactant would be produced specially in experimental conditions controlled by the diffusion of species (pH 2.5). However the coexistence of a high concentration of surfactant and the acid extractant at the adsorption interface, would not affect the permeability of the metal, as it is observed in Figure 8, in spite of the preferential adsorption of the surfactant. This suggests that it would be produced a high extractant concentration at the interface due to the interaction between this carrier compound and the surfactant, by means of the formation of micelle systems 19) based on hydrophobics effects caused by hydrogen bonding and dipole-dipole interactions. Then it would be possible to state the following equations that would complement the above proposed transport mechanism:

HXorg + ST ad = ST - HXad (10)
2 ST - HXad + Cu - ST2+ ad = Cu-X2 org + 2 H+ aq + 3 ST ad (11)

The strong noticeable interfacial e activity of Span-80 can be understood if the nature of its hydrophobics "tail" is considered. It corresponds to a unsaturated structure (see Figure 1). Since the diluent (kerosene) is a saturated solvent, the surfactant will be more weakly solvated and therefore more interfacially active. Furthermore, the polar head of the surfactant, presents an important number of hydrophilics groups, which would assure a good emulsifier molecules transport from the organic phase towards the interface with the aqueous solution 12).

The experimental results obtained and the proposed preliminary analysis allow us to indicate that in fact, under certain experimental conditions related to the acidity of the aqueous phase, the noionic surfactant substance would presents an accelerative effect of the metal transport in the membrane. This would probably be caused due to electrostatic or solvatation-type interactions between the metal ions, the acid carrier extractant and the noionic surfactant.

The probable effects of the different types of surfactants (anionics, cationics, noionic) and its concentration in the organic liquid membrane, on the diffusion rate of the species through the membrane is being studied at the moment, and will be matter of future communications. Nevertheless, the experimental information obtained and analysed in this work will be of great value for such purposes.


The nonionic surfactant substance Span-80 presented a remarkable interfacial activity, being obtained values of interfacial area occupied by mol of surfactant, SST = 3.43.109 cm2/mol, an adsorption equilibrium constant for the surfactant, KST = 1.15.108 cm3/mol and a critical micelle concentration value of 0.135 g/L. This compound presented a null capacity of extraction of Cu(II).

The extractant molecule LIX-860 NIC presented a lower interfacial activity than that of Span-80, generating a competence with the surfactant by the adsorption sites. The capacity of LIX-860 NIC for copper extraction resulted to be superior when using feed aqueous solutions at pH 2.5. The permeability rate of metal in the liquid membrane increased with the increase of the extractant content in the organic phase. It is concluded that the surfactant substance causes a degree of acceleration in the metal permeability in the LM under experimental conditions in which the metal transport rate would be controlled by the diffusion of the chemical species, i.e. at an initial pH of 2.5 and when a higher concentration of Span-80 is utilized. This accelarative effect was not observed under pH conditions where the carrier extractant present a lower copper extraction capacity.


The authors express their gratefulness to FONDECYT by the economic support to this study through the Research Projects Nº 1011021 and Nº 7980004. The authors also thank to Japan Gore Tex and Química Harting Chile Co. by the provision of hollow fibers and the extractant LIX-860 NIC respectively. The authors give thanks Dr. Fumiyuki Nakashio from The Kumamoto Institute of Technology of Japan, for his valuable comments and contribution to our research line


1. S.W. Park, G.W. Kim, S.S. Kim and I.J. Sohn. Sep. Sci.Technol., 36, 2309 (2001).        [ Links ]

2. A. K. Chakravarti, S.B. Chowdhury and D.C. Mukherjee. Colloids and Surfaces, 166, 7 (2000).        [ Links ]

3. G.O. Yahaga. Sep. Sci.Technol., 36, 3563 (2001).        [ Links ]

4. Y. Sato, K. Kondo and F. Nakashio. J. Chem. Eng. Jpn, 23, 23 (1990).        [ Links ]

5. F. Nakashio. J. Chem. Eng. Jpn, 26, 123 (1993).        [ Links ]

6. F. Valenzuela, C. Basualto, C. Tapia and J. Sapag. J. Membrane Sci., 155, 163 (1999).        [ Links ]

7. F. Valenzuela, H. Aravena, C. Basualto, J. Sapag and C. Tapia. Sep. Sci.Technol., 35, 1409 (2000).        [ Links ]

8. F. Valenzuela, F. Pizarro, C. Basualto, J. Sapag y C. Tapia. Bol. Soc. Chil. Quím., 46, 187 (2001).        [ Links ]

9. F. Valenzuela, M.A. Vega, M.F. Yánez, C. Basualto. J. Membrane Sci., 204, 385 (2002).         [ Links ]

10. M. Szapakowska and O. Nagy, J. Membrane Sci., 76, 27 (1993).        [ Links ]

11. H. Dingsheng and M. Ming. Hydrometallurgy, 56, 157 (2000).        [ Links ]

12. B. A. Mikucki and K. Osseo-Asare, Solv. Extr. Ion Exch., 4, 503 (1986).        [ Links ]

13. F. Valenzuela, C. Basualto, J, Sapag and C. Tapia. Minerals Engineering, 10, 1421 (1997).        [ Links ]

14. A. Acosta, F. Valenzuela, C. Basualto, J. Marchese and M. Campderrós. Bol. Soc. Chil. Quím., 43, 401 (1998).        [ Links ]

15. F. Valenzuela and C. Basualto. Journal Lat. Applied Res., 24, 175 (1994).        [ Links ]

16. F. Valenzuela, F. Nakashio and K. Kondo. Journal Lat. Applied Res., 24, 33 (1994).        [ Links ]

17. F. Valenzuela, C. Basualto, C. Tapia, J. Sapag and C. Paratori. Minerals Engineering, 9, 15 (1996).        [ Links ]

18. K. Yoshizuka, K. Kondo and F. Nakashio. J. Chem. Eng. Jpn, 19, 396 (1986).        [ Links ]

19. L.S. Romsted, Micellar effects on reaction rates and equilibria, in K.L. Miller, B. Lindman Editors, Surfactant in Solution, Vol.2, Plenum Press, NY, 1984.        [ Links ](*) To whom correspondence should be addressed

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License