Langmuir 2004, 20, 5179-5181
5179
Apparent Equilibration Time Required for Surfactant-Oil-Water Systems to Emulsify into the Morphology Imposed by the Formulation. Part 2: Effect of sec-Butanol Concentration and Initial Location Gabriela Alvarez, Raquel Anto´n, Shirley Marfisi, Laura Ma´rquez, and Jean-Louis Salager* Laboratory FIRP, Ingenierı´a Quı´mica, Universidad de Los Andes, Me´ rida 5101, Venezuela Received January 31, 2004. In Final Form: May 1, 2004 Winsor type I equilibrated surfactant-oil-water (SOW) systems produce o/w emulsions upon stirring. However, if the surfactant is initially dissolved in the oil phase, the attained type after inmediate emulsification is usually w/o. If the SOW system is partially equilibrated, it could result in a normal o/w emulsion, as if it were fully equilibrated. The minimum contact time for that to happen, the so-called apparent equilibration time tAPE, was previously shown (Langmuir 2002, 18, 607) to strongly depend on formulation, surfactant molecular weight, and oil viscosity. The present report shows that it depends on alcohol concentration and location in the unequilibrated system.
Introduction The relationship between the physicochemical formulation and the emulsion type has been known for a century as Bancroft’s rule and experimentally corroborated whatever the formulation variable. If the case of complex morphologies is put aside for the sake of simplicity, it can be said that oil-in-water (o/w; respectively w/o) is associated with Winsor I (respectively II) phase behavior, provided that the surfactant-oil-water (SOW) system is at equilibrium before emulsification takes place and that the water-to-oil ratio (WOR) is not far from unity.1,2 However, if the system is not equilibrated, which is an important case as far as the applications are concerned, things can be quite different. Lin3 reported a long time ago that if the surfactant is initially introduced in the oil phase, the SOW system is likely to produce a w/o emulsion upon immediate stirring, that is, opposite to the expected morphology from the corresponding equilibrated system. This actually resulted in a case of abnormal emulsion, as reported by many other authors,4-7 but this time out of equilibrium, as far as the surfactant partitioning is concerned. In this case, it can be said that with stirred unequilibrated SOW systems the external phase of the emulsion is the one that temporarily contains the surfactant, which is some kind of transient Bancroft’rule. During the equilibration, the surfactant transfers from the oil to the water phase. The question is how long the unequilibrated SOW system should be left to evolve toward complete equilibration so that enough surfactant has transferred in the water phase to produce a normal o/w * Corresponding author. E-mail:
[email protected]. (1) Bourrel, M.; Graciaa, A.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci. 1979, 72, 161. (2) Salager, J. L.; Loaiza-Maldonado, I.; Min˜ana-Pe´rez, M.; Silva, F. J. Dispersion Sci. Technol. 1982, 3, 279. (3) Lin, T. J. J. Soc. Cosmet. Chem. 1970, 21, 365. (4) Salager, J. L.; Min˜ana-Pe´rez, M.; Ra´mirez-Gouveia, M.; Rojas, C. I. J. Dispersion Sci. Technol. 1983, 4, 313. (5) Brooks, B. W.; Richmond, H. N. Chem. Eng. Sci. 1994, 49, 1065. (6) Vaessen, G. E. J.; Stein, H. N. J. Colloid Interface Sci. 1995, 176, 378. (7) Zerfa, M.; Sajjadi, S.; Brooks, B. W. Colloids Surf., A 2001, 178, 41.
emulsion. The time span after which the emulsion type is the same as if the SOW system were completely equilibrated has been called the apparent equilibration time tAPE. This value is quite important in practice because it is some kind of minimum time the phases have to be in contact together to attain the expected emulsion in an industrial process, and it is, thus, directly related with the cost. In a previous report,8 it was shown that tAPE strongly depends on formulation and that it decreases as formulation approaches the boundary between Winsor I and Winsor III phase behavior at equilibrium. In some instances tAPE was found to be essentially 0, as if equilibration had taken place instantly. The apparent equilibration time tAPE was also found to depend on the surfactant molecular weight and on the oil viscosity. The present paper is dedicated to the strong effect the presence and location of alcohol cosurfactant has on tAPE. Experimental Procedures The surfactant is dissolved in the oil phase, although all studied SOW systems have a physicochemical formulation that corresponds to a Winsor type I phase behavior at equilibrium. Hence, the system is originally unequilibrated, because the surfactant is always introduced in the phase where it will not be when equilibrium is reached. Alcohol is introduced either in oil or in water, as indicated. sec-Butanol is selected because it does not shift the physicochemical formulation, but it tends to dilute the adsorbed surfactant layer and to speed up transfer processes. The physicochemical formulation is changed, either by changing the salinity of the aqueous phase with an anionic surfactant system as in the previous paper or by changing the composition of a mixture of two nonionic surfactants of different HLB values. The typical emulsion is prepared according to the following procedure, which is slightly more simple than the one previously used.8 First, 25 mL of aqueous phase, eventually containing sodium chloride and alcohol, is introduced in a 250-mL highprofile beaker. Then, 25 mL of a solution of surfactant dissolved in oil, eventually containing alcohol, is carefully poured on top of the aqueous phase. The WOR is unity on a volume basis, and the overall surfactant content is 1 wt %. The system is left to rest (8) Salager, J. L.; Moreno, N.; Anto´n, R. E.; Marfisi, S. Langmuir 2002, 18, 607.
10.1021/la049727h CCC: $27.50 © 2004 American Chemical Society Published on Web 05/27/2004
5180
Langmuir, Vol. 20, No. 13, 2004
Figure 1. Conductivity of emulsion containing nonionic surfactant mixtures with different amounts of sec-butanol. In equilibrated systems, the surfactant is mostly in the aqueous phase. In unequilibrated systems, the surfactant is initially dissolved in the oil phase and the alcohol is initially located either in oil or in water (as indicated), and the system is emulsified immediately (contact time tC is 0). at constant temperature (25 ( 1 °C) during a contact time tC, after which it is emulsified by stirring it in a beaker with an UltraTurrax turbine blender model 45S8 at 10 000 rpm during 15 s. The conductivity is then measured with a Cole-Parmer model 19101 conductimeter to estimate the emulsion morphology. If the emulsion type is found to be w/o, the apparent equilibration has not been reached; if it is o/w, that is, as it would be with a fully equilibrated system, it is said to apparently behave as if it were equilibrated. The apparent equilibration time tAPE is defined as the shorter value of tC for which the normal o/w morphology is exhibited. The experiment is repeated with systems containing different formulations spanning over the whole Winsor type I range at equilibrium, including some Winsor III borderline cases. The oil phase is a kerosen cut which behaves, as far as phase behavior is concerned, as an equimolar mixture of heptane and octane, hence, exhibiting an experimnetally determined equivalent alkane carbon number of 7.5.9 The aqueous phase used in nonionic systems contains 1 wt % sodium chloride. With ionic surfactant systems, the salinity of the aqueous phase is the formulation variable and ranges from 0.5 to 7 wt %. Nonionic systems contain mixtures of Tween 85 (sorbitan trioleate + 20 ethylene oxide (EO) groups) and Tween 80 (sorbitan monooleate + 20 EO) supplied by Sigma whose hydrophilic-lipophilic balance (HLB) numbers are 11 and 15, respectively. Winsor type I phase behavior takes place at HLB g 11.5. Ionic systems contain 1 wt % petroleum sulfonate from Stepan Chemical Petrostep series, which is labeled PS410 because its optimum salinity (6.5 wt % NaCl) indicates an average molecular weight at about 410. It is worth remarking that commercial surfactant mixtures of this type were found to behave as pure ones, and as far as applications are concerned, to often provide an even better performance.10 Alcohol is sec-butanol reagent grade from Scharlau. Its concentration is indicated as vol % with respect to the total volume of the system. The amount of alcohol and its initial location, that is, whether it is placed in the water phase or in the oil phase, are of paramount importance in the present study.
Results and Discussion Nonionic Systems. The optimum formulation for a HLB scan is found around HLB ) 11 with the boundary between the Winsor III phase behavior and Winsor I phase behavior at equilibrium at HLB ) 11.5. Figure 1 indicates (9) Cash, L.; Cayias, J. L.; Fournier, G.; MacAllister, D.; Shares, T.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci. 1977, 59, 39. (10) Cayias, J. L.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci. 1977, 59, 31.
Letters
the resulting emulsion conductivity as a function of formulation for different cases. In the equilibrated systems, the surfactant is found in the aqueous phase. In presence of 1 vol % alcohol (black squares), the conductivity curve exhibits the o/w expected behavior.11 The lower conductivity observed in the region of HLB from 12 to 13 in the case of the no-alcohol emulsion (white squares) is attributed to a w/o/w multiple emulsion with the incorporation of water droplets in oil drops, as corroborated by microscopic observation. This may be due to a change in partitioning of the different species as discussed elsewhere.12 The point of main interest in Figure 1 is the remaining cases, that is, the unequilibrated emulsified systems made with 1 vol % alcohol introduced either in oil (black circles) or in water (white circles), whose emulsification is carried out at once, that is, after a contact time tC ) 0, which is not really zero but the minimum time to handle the pouring and mixing process, say 20-30 s. The figure shows a striking difference between the last two cases. The unequilibrated emulsified systems in which the alcohol has been introduced in the aqueous phase (white circles) essentially behave as the equilibrated emulsified systems with the same amount of alcohol (black squares). Emulsification results in a normal o/w morphology at once, despite the surfactant being in the oil, that is, in the “wrong” phase. On the other hand, the unequilibrated emulsified system in which the alcohol has been introduced in the oil phase (black circles) exhibits an abnormal low conductivity w/o morphology over the whole range of formulation. These results indicate that the alcohol initial location in the system is an important issue, even more than its very presence. This aspect will be studied here as before8 in a map of the emulsion type versus formulation and contact time. The Figure 2 upper graph indicates the meaning of the different regions for the case of an unequilibrated emulsified system in which 1 vol % alcohol was originally introduced in the oil phase. The inversion line between the abnormal (shaded) w/o and normal o/w emulsions indicates the apparent equilibration time tAPE, which is decreasing as formulation approaches optimum (and becomes zero at HLB ) 11.5), hence, corroborating the previously reported behavior for ionic systems.8,13 The Figure 2 lower graph indicates only the tAPE variation or value for different systems. Taking as a reference the case of 1% alcohol initially in oil (black squares), it is easily seen that decreasing the alcohol concentration in oil produces a shift of the inversion line to the right, i.e., it results in a longer apparent equilibration time tAPE, with a maximum delay for the no-alcohol case (black triangles). On the other hand, an increase from 1 to 3% alcohol in oil produces (at HLB ) 14) a 5-fold reduction of tAPE, which becomes essentially 0 beyond 3.5 vol % alcohol (not shown). The Figure 2 lower graph also shows that the apparent equilibration is much quicker when the alcohol is initially introduced in the aqueous phase. For an alcohol concentration higher or equal to 0.1% in water, tAPE is essentially zero over the whole HLB range (white squares). It is only when the initial alcohol concentration in water is reduced (11) Salager, J. L.; Loaiza-Maldonado, I.; Min˜ana-Pe´rez, M.; Silva, F. J. Dispersion Sci. Technol. 1982, 3, 279. (12) Graciaa, A.; Lachaise, J.; Sayous, J. G.; Grenier, P.; Yiv, S.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci. 1983, 93, 474. (13) Fillous, L.; Cardenas, A.; Rouvie`re, J.; Salager, J. L. J. Surfactants Deterg. 1999, 3, 303. (14) Chiang, M. Y.; Shah, D. O. Presented at the 5th International Symposium on Oilfield and Geothermal Chemistry, Stanford, CA, May 1980, Paper SPE 8988.
Letters
Langmuir, Vol. 20, No. 13, 2004 5181
Figure 3. Apparent equilibration time tAPE versus formulation (salinity of the aqueous phase), with variable amount and initial location of sec-butanol.
equilibration time tAPE with respect to the nonalcohol case (black triangles), whereas its initial introduction in the aqueous phase (white squares) turns it to essentially 0. Hence, the effect of the alcohol appears to be the same in ionic and nonionic systems. Conclusions
Figure 2. Upper graph: apparent equilibration time tAPE versus formulation (HLB of nonionic surfactant mixture). Lower graph: variation of tAPE with the amount and initial location of sec-butanol.
down to 0.04% (white circle) that a nonzero tAPE is found. It is worth noting that even a very low 0.02%, that is, 200 ppm (white triangle) of alcohol initially placed in the aqueous phase, is sufficient to half the value of tAPE with respect to the no-alcohol case (black triangles). Anionic Systems. The same kind of experiment is carried out with a system containing an anionic surfactant of the petroleum sulfonate type (Figure 3). In this case, the formulation variable is taken as the aqueous phase salinity. It is found that the equilibrated system with 1 vol % sec-butanol exhibits the Winsor I-Winsor III transition for a salinity of 6 wt % NaCl, with the emulsion inversion at a salinity close to 6.5 wt % NaCl. Threephase behavior is not exhibited by this system in the absence of alcohol, but the inversion of the equilibrated emulsified system also takes place at 6.5 wt % NaCl, corroborating that sec-butanol does not alter the formulation. It is seen that the initial introduction of 1% alcohol in the oil phase (black squares) slightly reduces the apparent
The apparent preequilibration time tAPE is found to considerably depend on sec-butanol amount and location. sec-Butanol is an alcohol which does not alter formulation, and its essentially unique role as a cosurfactant is to adsorb at the interface and to introduce disorder in any kind of surfactant organization. The presence of alcohol adsorbed at the interface is known to reduce the interfacial viscosity14 and, hence, to reduce the barrier, facilitate the transfer, and speed up equilibration. This reasonning seems to apply differently on the two sides of the interface, with much less influence on the oil side, where the surfactant is initially located, than on the water side where the alcohol is the only amphiphile to adsorb from the bulk phase, with an effect at an extremely low concentration. As far as applications are concerned, this indicates that the introduction of a very small amount of alcohol in the aqueous phase can speed up considerably the transfer if the alternative of shifting the formulation toward optimum is not feasible. Acknowledgment. The authors would like to express their appreciation to the Venezuelan Ministry of Science and Technology (FONACIT AP-1997-3719, F-2000-1629, and S1-2001-1156 Grants) and to their University Research Council (CDCHT-ULA) for sponsoring the Lab. FIRP research program in emulsion science. LA049727H
