5666
Langmuir 2004, 20, 5666-5668
Partial Phase Behavior and Micellar Properties of Tetrabutylammonium Salts of Fatty Acids: Unusual Solubility in Water and Formation of Unexpectedly Small Micelles Raoul Zana Institut C. Sadron (CNRS), 6 rue Boussingault, 67000 Strasbourg, France Received February 24, 2004 The tetrabutylammonium (TBA) salts of fatty acids, from dodecanoic acid (C12) to octacosanoic acid (C28), have been prepared by direct neutralization of the fatty acid by TBA hydroxide. Unexpectedly, all of these surfactants have been found to be soluble in water under the form of micelles at a sufficiently high temperature. For instance, the solubility of TBA octacosanoate in water is of about 7 wt % at 46 °C. Starting from TBA docosanoate, the aqueous solutions of the surfactants gelled below a certain temperature. The gelling temperature increased linearly with the fatty acid carbon number. Upon increasing temperature, the TBA octocosanoate showed a relatively complex phase behavior that has been investigated. The micellar solutions of these surfactants did not cloud at high temperatures, up to 98 °C, contrary to TBA alkylsulfates. The aggregation numbers of micelles of the various TBA alkylcarboxylates have been measured and found to be smaller than those for the maximum spherical micelle that a surfactant with the same alkyl chain length can form. The micelle micropolarity and microviscosity (as sensed by fluorescent probes) decreased and increased, respectively, with the fatty acid carbon number.
Introduction The tetrabutylammonium (TBA) alkyl sulfates (alkyl ranging from dodecyl to octadecyl) are unusual ionic surfactants. Indeed, their aqueous solutions show clouding and phase separation upon increasing temperature, resulting in a very concentrated surfactant phase of nearly constant composition and a more dilute surfactant solution.1-3 Besides, they form micelles that have an aggregation number that increases with temperature when the surfactant concentration is, say, above 50 mM.4,5 These two properties are usually observed with the nonionic poly(ethylene glycol) monoalkyl ether surfactants.6 In contradistinction to TBA alkyl sulfates, the TBA dodecanoate shows no clouding or Krafft temperature in the range 0-100 °C and its phase diagram (temperature versus composition) shows only a very restricted range of mesophases at a very high concentration.7 This enormous difference between TBA alkyl sulfates and TBA dodecanoate led us to prepare a series of TBA alkylcarboxylates (TBACm) of increasing alkyl chain length, dodecanoate (TBAC12), tetradecanoate (TBAC14), hexadecanoate (TBAC16), octadecanoate (TBAC18), eicosanoate (TBAC20), docosanoate (TBAC22), tetracosanoate (TBAC24), and octacosanoate (TBAC28), and to investigate their solution properties. The last three surfactants belong to what Laughlin8a categorizes as ultralong-chain surfactants. To our surprise, all these (1) Yu, Z.-J.; Xu, G. J. Phys. Chem. 1989, 93, 7441. (2) Yu, Z.-J.; Zhou, Z.; Xu, G. J. Phys. Chem. 1989, 93, 7446. (3) Yu, Z.-J.; Zhang, X.; Zhou, Z.; Xu, G. J. Phys. Chem. 1990, 94, 3675. (4) Benrraou, M.; Bales, B. L.; Zana, R. J. Phys. Chem. B 2003, 107, 13432. (5) Benrraou, M.; Bales, B. L.; Talmon, Y.; Schmidt, J.; Zana, R. Manuscript in preparation. (6) Degiorgio, V. In Physics of Amphiphiles: Micelles, Vesicles and Microemulsions; Degiorgio, V., Corti, M., Eds.; North-Holland: Amsterdam, 1985; p 303. (7) Jansson, M.; Jo¨nsson, A.; Li, P.; Stilbs, P. Colloids Surf. 1991, 59, 387. (8) Laughlin, R. L. The Aqueous Phase Behavior of Surfactants; Academic Press: London, 1994; pp (a) 285, (b) 271.
surfactants were found to be soluble in water, forming a low viscous micellar solution, even the TBAC28 (at above 41 °C, though). This letter is to report on the partial phase behavior of these surfactants and on the properties of their micelles. A more detailed account of the micellar properties of TBACm surfactants and of the microstructure of their solutions will be given in a forthcoming publication. Surfactant Preparation The TBACm surfactants were prepared by direct neutralization of tetrabutylammonium hydroxide (TBAOH) by the corresponding fatty acid in water. TBAOH was purchased from Fluka as a 1.0 M aqueous solution (normality given by the manufacturer). The molality of this solution was redetermined by potentiometric titration with hydrochloric acid (Titrisol). The density of the solution was also determined and found to be 0.990 g/mL. All fatty acids were purchased from Fluka with the best available grade. A weighted amount of fatty acid was put in contact with water, and the weighted stoichiometric amount of TBAOH was added. The flask was sealed, and the mixture was agitated for 1 night. For the C12, C14, C18, C20, and C22 fatty acids, the stirring resulted in clear solutions at room temperature. Stirring at room temperature resulted in gelling for the C24 and C28 fatty acids. To our surprise, homogeneous solutions resulted when stirring of the mixtures was performed above 30 and 60 °C, respectively. The stock solutions were stored in the dark at 2-3 °C. Partial Phase Behavior The stock solutions of TBAC12 (0.31 M, or about 14 wt %), TBAC14 (0.31 M, or about 15 wt %), TBAC18 (0.30 M, or about 16 wt %), TBAC20 (0.36 M, or about 20 wt %), TBAC22 (0.17 M, or about 10 wt %), TBAC24 (0.31 M, or about 18 wt %), and TBAC28 (0.12 M, or about 8.0 wt %) showed no clouding when heated to 98 °C. No precipitation occurred when the stock solutions of TBAC12, TBAC14, TBAC18, and TBAC20 were maintained at 2-3 °C for periods of weeks. Thus, the Krafft temperature of these four surfactants is considerably lower
10.1021/la040033i CCC: $27.50 © 2004 American Chemical Society Published on Web 06/11/2004
Letters
than that of alkali metal salts of the same fatty acids. For instance, the Krafft temperatures of sodium dodecanoate and tetradecanoate are around 20 and 40 °C, respectively.9 Thus, the substitution of the sodium ion by the TBA ion brought about a considerable increase of the solubility of the surfactant in water. The stock TBAC22 solution turned into a translucent, homogeneous, and hard gel phase at temperatures below about 12 °C. The gel did not flow when turning the test tube upside down. More concentrated or more dilute TBAC22 solutions were prepared by evaporation of part of the water or dilution by addition of water. The gelling temperature was found to be independent of concentration in the range 5-21 wt %. The 21 wt % gel became turbid before melting, but the turbid range was quite narrow, less than 1 °C wide. The turbid range was also observed at a concentration of 14.8 wt % but not at lower concentration, probably because turbidity takes a relatively long time to occur. The observation of gel melting became very difficult at 27 wt % because the resulting solution was very viscous. It is likely that the turbid range corresponds to a range of coexistence of two phases: the gel phase and the micellar phase that is stable at above, say, 12 °C for TBAC22. This is supported by the results for the longer TBACm surfactants presented in the following. A similar behavior was observed with the 18 wt % stock solution of the TBAC24 surfactant. This system was under the form of a clear, hard, and homogeneous gel at T < 24 °C. This gel slowly turned turbid at 24.3 °C after 2 h. Two phases were clearly observed at 25 °C: a lower clear solution phase and an upper phase that is probably the remainder of the gel phase. The system became a clear homogeneous solution at above 25.8 °C. The more dilute 5.7 wt % system showed the same behavior with a softer gel phase at 24 °C and a turbid system at 24.3 °C. The separation into two phases at a higher temperature was not observed because the experiments were not performed slowly enough. The gel phase remained apparently homogeneous when maintained at around 2-3 °C even for 7 weeks, suggesting the absence of a precipitation of the solid surfactant. The phase behavior of the 8.0 wt % stock solution of TBAC28 was more complex. The solution was clear at 60 °C. When left at 2-3 °C for some time, the solution gave rise to a two-phase system with a clear and low viscous liquid phase and a bulky white precipitate floating over it, occupying much more volume than the solid surfactant should. The dry content of the liquid phase was found to be very low (below 0.5 wt % and this included some precipitate that was carried with the solution used for the dry content measurement). This result indicates that the precipitate is probably made up of hydrated surfactant. The temperature was progressively raised, and no change was observed until 27.3 °C, the temperature at which the liquid phase turned into a gel in which the white solid surfactant was dispersed, giving rise to a macroscopically heterogeneous system. The transformation of the liquid phase into a gel resulted from the solubilization or fusion of the solid white phase. This could be observed visually: the outer part of the white solid became transparent and was slowly dispersed in the liquid phase upon agitation. Upon increasing temperature, the volume of the white solid phase progressively decreased until its complete disappearance at 40 °C, giving rise to a single translucent gel phase. Note that the magnetic stirrer placed in the system was completely blocked in the range between 28 (9) Madelmont, C.; Perron, R. Colloid Polym. Sci. 1976, 254, 581.
Langmuir, Vol. 20, No. 14, 2004 5667
Figure 1. Partial phase diagram of the TABCm/water mixtures showing the succession of phases for the TBAC22, TBAC24, and TBAC28 surfactants at the concentration of the stock solutions of these three surfactants.
and 40 °C, revealing the hardness of the gel. Also, after 1 day at 39 °C some solid surfactant was still visible. It took 36 h at 40 °C for the system to turn into a homogeneous gel phase. That phase remained unchanged until about 45 °C, the temperature at which the system became turbid and separated into two phases, a lowviscosity lower liquid phase and a very turbid upper phase. The concentration the liquid phase was found to be 7.2 wt % at 46 °C (concentration of the initial stock solution: 8.0 wt %), indicating that it is a micellar solution. The upper phase is probably a mixture of the gel phase with a dispersed micellar solution in its bulk. At about 52 °C, the system turned into a nearly transparent one-phase micellar solution. The phase transition temperatures of 40, 46, and 52 °C may be slightly overestimated because, even though the temperature was increased at a relatively slow rate (less than 4 °C per day), this rate was still perhaps too fast for the equilibrium state of the system to be reached. This last part of the phase behavior of the TBAC28 system is very similar to that of the TBAC22 and TBAC24 systems. The two-phase-one-phase transition temperature T21 increases with the surfactant chain length and so does the range of coexistence of the gel and micellar phases. Figure 1 shows that the temperature T12 above which the gel phase starts expelling a micellar solution and turns into two phases also increases nearly linearly with the fatty acid carbon number m (alkyl chain carbon number ) m - 1). It is noteworthy that the extrapolation of the T12 line to m ) 20 yields a value of about 0 °C. This is in agreement with the fact that the TBAC20 showed no gelling down to a temperature of 2-3 °C. The main difference between the TBAC28 system and the TBAC22 and TBAC24 systems is the occurrence for the former of the transition from a two-phase system (gel phase + hydrated surfactant) into a gel phase. The most striking result is probably the extraordinary solubility of the TBAC22, TBAC24, and TBAC28 surfactants (ultralong-chain surfactants)8a in water under the form of micelles. Indeed, the above results show that TBAC24 is soluble in water to a minimum concentration of 18 wt % at above 25 °C. The same is true for TBAC28 with a solubility of about of 7 wt % at 46 °C. To the best of our knowledge, solubilization of such ultralong-chain surfactants in water has not been reported thus far.8 Recall that the solubility of a compound depends on the difference between the chemical potentials of the compound in the crystal state and in the solubilized state. It is likely that the free energy level of TBA alkylcarboxylates and alkyl sulfates is quite high in the crystal state. Indeed, the TBA dodecyl sulfate remains liquid at 3 °C4 and the TBA tetradecyl sulfate is a liquid at room temperature.5 Presumably a similar behavior would be found for the TBACm surfactants. This behavior does indicate a higher
5668
Langmuir, Vol. 20, No. 14, 2004
Letters
Table 1. Values of the Pyrene Polarity Ratio I1/I3, of the Micelle Aggregation Number N, of the Pyrene Lifetime (τ and 1/A2), and of the Intramicellar Quenching Rate Constant kQ of the Investigated Surfactants C surfactant (mM) TBAC12a TBAC14a TBAC18a TBAC22a TBAC24b b
N
NS
286 35 47 169 52 64 216 87 105 169 106 155 49.5 113 185
τ 1/A2 (ns) (ns) 351 352 355 364 353
335 354 361 363 358
kQ (s-1)
NkQ (s-1)
2.7 × 107 1.7 × 107 7.3 × 106 3.8 × 106 4.5 × 106
94 × 107 88 × 107 64 × 107 40 × 107
I1/I3 1.39 1.34 1.32 1.29 1.23
a Quencher: dodecylpyridinium chloride. Results at 25 °C. Quencher: hexadecylpyridinium chloride. Results at 35 °C.
free energy of the crystal state of TBACm surfactants with respect to the corresponding NaCm surfactants, probably associated with the large size of the TBA ions that hinders the alkyl chain packing in the crystal structure. This would favor the solubilized state with respect to the solid state in the presence of water. Another fact that may favor the solubility of the TBA carboxylates in water is the strongly hydrophilic character of the carboxylate group, much discussed by Laughlin.8b Properties of TBACm Micelles The aggregation number N of the micelles of all TBACm surfactants, except TBAC28, have been measured using the time-resolved fluorescence quenching method with pyrene as a fluorescent probe and dodecyl or hexadecyl pyridinium chlorides as quenchers of the pyrene fluorescence.10-12 In this letter, only the values of N at 25 °C and a single surfactant concentration are given in Table 1. The values of NS in Table 1 are those calculated for the maximum spherical micelle formed by a surfactant having the same alkyl chain carbon number (m - 1) as the TBACm surfactant. These values are calculated from NS ) 4πl3/ 3v, where v is the volume of the alkyl chain and l is its length (assuming a fully extended conformation). Both v and l were calculated using the expressions reported by Tanford.13 It is seen that NS is always well above N, particularly for the surfactants with the longest chain. For the TBA alkyl sulfate surfactants, the micelle aggregation number is about equal to NS at room temperature and low concentration.4,5 However, the growth of TBA alkyl sulfate micelles upon increasing surfactant concentration is rather weak, much weaker than expected on the basis of the low critical micelle concentration value of these surfactants.4,5 This effect has been attributed to the large size of the TBA counterion. Table 1 also lists the values of the pyrene lifetime, τ, in the micellar environment, measured in the absence of quencher and of the apparent pyrene lifetime 1/A2, (10) Tachiya, M. Chem. Phys. Lett. 1975, 33, 289. (11) Infelta, P. P. Chem. Phys. Lett. 1979, 61, 88. (12) Zana, R. In Surfactant Solutions. New Methods of Investigation; Zana, R., Ed.; M. Dekker, Inc.: New York, 1987; p 241. (13) Tanford, C. J. Phys. Chem. 1974, 78, 2469.
obtained from the analysis of the biphasic fluorescence decay curve recorded in the presence of quencher.11,12 In all instances but TBAC12, τ is very close to 1/A2. This indicates that in the experimental conditions used (surfactant concentration, temperature, and nature of the quencher) the quencher distribution among micelles is frozen during a time long with respect to the fluorescence lifetime.12,13 The values of the intramicellar quenching rate constant kQ, obtained from the analysis of the biphasic fluorescence decay curve recorded in the presence of quencher10-12 are also listed in Table 1. It is seen that kQ decreases upon increasing length of the surfactant alkyl chain, i.e., increasing micelle aggregation number. In view of their small N values the TBACm micelles are likely to be close to spherical. If such is indeed the case the product NkQ is proportional to the reciprocal of the micelle microviscosity (microviscosity at the site of solubilization of the probe and quencher).14 The results listed in Table 1 show that the value of this product decreases regularly from about 95 to 40 in going from TBAC12 to TBAC22 (TBAC24 was not included because the experiments were performed at a higher temperature for this surfactant). This decrease reveals that the micelle microviscosity is increasing with the surfactant chain length, a result already noted for other surfactants.14 We have also determined the pyrene polarity ratio I1/I3 for the TBACm surfactant micelles.12 The values listed in Table 1 show that I1/I3 decreases as the alkyl chain length increases, revealing that pyrene senses a progressively less polar micellar environment as the surfactant chain length becomes longer. A similar behavior has been noted for other series of surfactants.15 Table 1 also shows that the pyrene fluorescence lifetime τ increases with m. This increase is an additional indication that the TBACm micelle microenvironment becomes less polar as m increases.15 Conclusions The TBACm surfactants have been found to be highly soluble in water at room temperature up to m ) 22, at above 25 °C for m ) 24, and at above about 46 °C for m ) 28. The TBACm surfactants with m up to 20 show no Krafft phenomenon and no clouding in the temperature range between 3 and 98 °C. The partial phase behavior of the TBACm surfactants with m ) 22, 24, and 28 has been investigated. The most prominent feature is the occurrence of a gel phase which melts at a temperature increasing with m and gives rise to a micellar solution. In aqueous solution, the TBACm surfactants form micelles which are much smaller than the maximum spherical micelles that can be formed by a surfactant having the same alkyl chain length. LA040033I (14) Zana, R. J. Phys. Chem. 1999, 103, 9177. (15) Lianos, P.; Zana, R. J. Colloid Interface Sci. 1981, 84, 100.
