5570
Langmuir 2006, 22, 5570-5571
Premicellization of Dimethyl Di-n-dodecylammonium Chloride Nicholas D. Gillitt,† Gianfranco Savelli,‡ and Clifford A. Bunton*,† Department of Chemistry and Biochemistry, UniVersity of California, Santa Barbara, California 93106-9510, and CEMIN, Departamento di Chimica, UniVersita` di Perugia, 06100 Perugia, Italy ReceiVed March 10, 2006. In Final Form: April 19, 2006 NCH3, CH2-1, and ω-CH3 1H NMR signals of 2.5 × 10-5 M dimethyl di-n-dodecylammonium chloride (DDDACl) are sharp in D2O with expected line shapes, but the line width of NCH3 increases markedly to ca. 10 Hz and the ω-CH3 triplet is distorted at 10-4M, where electrolytic conductance had earlier demonstrated premicellar formation. The N-CH3 signal sharpens, and the resolved ω-CH3 triplet signal reappears at ca. 2.5 × 10-4 M DDDACl where there was earlier evidence of micellization. All of the 1H NMR signals become broad and ill-defined at >10-2 M DDDACl where there is growth and a structural change of the association colloid.
Introduction Synthetic surfactants are amphiphiles with polar, ionic, or nonionic headgroups and extended apolar, organic residues.1-3 At concentrations above the so-called critical micelle concentration, cmc, self-association in water generates micelles, which are typically spheroidal but with increasing concentration grow and form rodlike assemblies, or other association colloids. In one model of micellization, surfactants are treated as monomers at concentrations below the cmc, but in the alternative massaction model, there is a distribution of monomers and premicelles in this region.4 Aqueous micelles control the rates of many reactions, often at concentrations below the cmc in water, which may be due to the formation of premicelles or to reactant-induced micellization.2,3,5 There is strong physical evidence for the premicellization of some gemini surfactants without added probes or other solutes6,7 because equivalent conductance goes through maxima and decreases as micelles form. This behavior had been observed with dimethyldidodecylammonium chloride, DDDACl, and similar twin-chain cationic surfactants by Ralston et al.,8 who ascribed it to the formation of premicelles that do not associate with Cl-. We have examined the 1H NMR spectrum of DDDACl at concentrations below the reported cmc of 0.176 mM,9 and our results are consistent with the earlier conductance work8 and also support the formation of larger assemblies at higher concentrations and other conditions.10 (CH3)2N+(C12H25)2Cl-: C12H25 ) CH2(1)CH2(2)(CH2)9CH3 DDDACl The 1H signals of the terminal methyl groups and CH2(1) and CH2(2) are separated from the other methylene signals that are overlapped. * Corresponding author. E-mail:
[email protected]. † University of California. ‡ Universita di Perugia. (1) (a) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.; Academic Press: New York, 1991. (b) Fendler, J. H. Membrane Mimetic Chemistry; Wiley: New York, 1986. (2) Tascioglu, S. Tetrahedron 1996, 52, 11113. (3) Savelli, G.; Germani, R.; Brinchi, L. In Reactions and Synthesis in Surfactant Systems; Texter, J., Ed; Dekker: New York, 2001; Chapter 8. (4) (a) Mukerjee, P. AdV. Colloid Interface Sci. 1967, 1, 241. (b) Elworthy, P. H.; Florence, H. T.; Macfarlane, C. B. Surfactants and Surface ActiVe Agents; Chapman and Hall: London, 1968; Chapter 1. (5) Bunton, C. A.; Nome, F.; Quina, F. H.; Romsted, L. S. Acc. Chem. Res. 1991, 24, 358. (6) Menger, F. M.; Littau, C. A. J. Am. Chem. Soc. 1993, 115, 10083.
Table 1. Chemical Shifts, δ, and Line Widths, B(Hz), of DDDACl in D2O at 25 °C and the Dioxane Reference unless Specified NCH3 B(Hz)
[DDDACl], mM
δ
0.025b 0.025 0.042b 0.061b 0.085b 0.101c 0.106b 0.132b 0.135 0.160 0.168b 0.250 0.260b 0.490 0.980 1.01b 2.89b 6.30 12.5b 12.5
3.022 3.022 3.025 3.035 3.051 3.043 3.060 3.069 3.069 3.079 3.068 3.091 3.088 3.100 3.105 3.102 3.115 3.140 3.148 3.148
1.7 2.1 4.5 10.2 10.8 10.1 10.4 7.3 8.5 5.4 6.1 3.7 4.1 3.1 3.5 2.7 3.5 5.4 7.2 7.2
ω-CH3 δ
δ
0.853 0.853 0.856 0.861 0.870 0.870 0.874 0.879 0.879 0.884 0.883 0.888 0.886 0.890 0.890 0.891 0.890 0.890 0.891 0.891
3.246 3.245 3.246 3.245 3.243 3.242 3.243 3.242 3.243 3.242 3.242 3.245 3.243 3.249 3.258 3.252 3.266 3.292 3.306 3.306
CH2(1) Na
La
17 17 17 19 17.5
7.0 7.0 8.0 7.5 8.0
17.0 17.5 17.5 15.5 17.0 17.0 17.0 16.5 17.5 16.5 16.5
7.0 8.5 7.0 7.5
CH2(2) δ 1.718 1.720 1.716 1.706 1.699 1.702 1.693 1.685 1.687 1.682 1.690 1.678 1.684 1.674 1.679 1.679 1.681 1.691 1.691 1.696
a Values in Hz between the outer and inner signal peaks defined as in ref 14. b No dioxane, ethanol reference. c Very dilute dioxane as reference as specified in ref 11.
Results and Discussion Chemical shifts, δ, and line widths, B(Hz), of identifiable signals as a function of [DDDACl] are in Table 1, and examples of signals are in Figure 1. The CH2(2) signal is a broad singlet, and other overlapped signals of the methylene chain are not shown.11 The broadening and subsequent sharpening of the NCH3 signal and distortion and recovery of the ω-CH3 triplet occur (7) Zana, R. J. Colloid Interface Sci. 2002, 246, 182. (8) Ralston, A. W.; Eggenburger, D. N.; DuBrow, P. L. J. Am. Chem. Soc. 1948, 70, 977. (9) Lang, J. J. Phys. Chem. 1982, 86, 992. (10) (a) Kunitake, T.; Okahata, Y.; Tamaki, K.; Kumamuro, F.; Yakaynagi, M. Chem. Lett. 1977, 387. (b) Carmona-Ribeiro, A. M. Chem. Soc. ReV. 1992, 21, 209. (c) Menger, F. M.; Keiper, J. S. Curr. Opin. Chem. Biol. 1998, 2, 726. (11) The preparation of DDDACC was as described in ref 12. The 1H NMR spectra were monitored on a Varian 500 MHz instrument with an ID probe at 25 °C and 1 s delay. Overnight accumulations were required for the most dilute solutions, and Varian LB software was used. References were dioxane, δ ) 3.75 ppm, or ethanol, CH3, δ ) 1.17 ppm.13 [Dioxane] was either such that its 1H signal was similar in magnitude to that of NCH3 or was just detectable and with no dioxane [ethanol] was sufficient to detect the signal with baseline resolution.
10.1021/la0606695 CCC: $33.50 © 2006 American Chemical Society Published on Web 05/26/2006
Letters
Figure 1. Effect of DDDACl concentration on the chemical shift and line width at (A) 0.025, (B) 0.106, (C) 0.26, and (D) 12.5 mM DDDACl.
within the same concentration range of DDDACl that had been associated with the formation of premicelles and their subsequent conversion to micelles.8 The spin-spin coupling pattern of CH2(1) is as expected for methylene groups in chains attached to a quaternary ammonium ion. N and L are the frequency differences in Hz between the outer and inner peaks, respectively, and for long-chain ammonium ions, their values indicate the balance between the trans and gauche orientations of the first two methylene groups of a chain.14 This general signal shape changes little with increasing [DDDACl] (Table 1 and Figure 1) up to concentrations where the growth of association colloids generates new structures that may be bilayered,10 as had been found at higher surfactant concentrations. We did not examine these conditions. The sphere-to-rod micellar (12) Cipiciani, A.; Germani, R.; Savelli, G.; Bunton, C. A.; Mhala, M.; Moffatt, J. R. J. Chem. Soc., Perkin Trans. 2 1987, 541. (13) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512. (14) Lichtenberg, D.; Kroon, P. H.; Chen. S. J. Am. Chem. Soc. 1974, 96, 5934.
Langmuir, Vol. 22, No. 13, 2006 5571
transition is seen with many surfactants as their concentrations are increased,1-4 but our solutions did not become viscous. The 1H NMR signals of 2.5 × 10-5 M DDDACl are as expected for monomers, but the broadening or distortion of the methyl signals at the ends of the chains at ca. 10-4 M (Table 1 and Figure 1) indicates that self-association, possibly, although not necessarily, as a dimer, sharply restricts local mobilities at these positions. The distortion of the ω-CH3 triplet signal between 0.042 and 0.135 mM DDDACl corresponds closely to the broadening of the N-CH3 signal (Table 1), although within the uncertainties of the measurement (Figure 1) the splitting did not change. The 1H chemical shifts of the terminal methyl groups increase modestly in the concentration region where line shapes change but then become insensitive to increases in [DDDACl] (Table 1). Mobilities, as indicated by line widths and signal shapes (Figure 1), apparently increase on micellization8,9 because micelles are fluidlike assemblies1-5 and both N-CH3 in the water-like surface region15 and ω-CH3 in the hydrocarbon-like interior could have considerable freedom of rotation. The flexibility of the methylene chain is probably less sensitive to these changes in morphology, but as noted, flexibility appears to be lost at all positions at higher [DDDACl], which is consistent with the formation of layered structures.10 The long chain is probably coiled in the monomer and the premicelles to minimize water-hydrocarbon interactions but becomes more extended upon the formation of association colloids. The data in Table 1 and Figure 1 show that changes in surfactant concentration initially have very different effects upon signals of the terminal CH3 groups and of the CH2 groups of the long chain because, where observable, the latter change little until the expected growth of the association colloid. The 1H NMR signals are insensitive to the choice of a reference11 because they are unaffected by a change from moderately concentrated to very dilute dioxane or its elimination and replacement by very dilute ethanol (Table 1). There are a number of surfactant-accelerated reactions, including decarboxylation and dephosphorylation,16,17 whose rates go through maxima in very dilute surfactant solution because of the formation of premicelles of surfactant and substrate rather than substrate-induced micellization. Tight packing between the substrate and surfactant in a premicelle probably excludes water molecules, which in solution or in normal micelles inhibit these reactions. Acknowledgment. We are grateful to Dr. A. Shirazi for very valuable technical advice. LA0606695 (15) Geng, Y.; Romsted, L. S.; Froehner, S.; Zanette, D.; Magid, L. J.; Cuccovia, I. M.; Chaimovich, H. Langmuir 2005, 21, 562. (16) Cuenca, A. Langmuir 2000, 16, 72. (17) Brinchi, L.; Di Profio, P.; Germani, R.; Giacomini, V.; Savelli, G.; Bunton, C. A. Langmuir 2000, 16, 222 and references therein.
