Gemini Surfactants with Acetylenic Spacers - Langmuir (ACS

---

2062

Langmuir 2000, 16, 2062-2067

Gemini Surfactants with Acetylenic Spacers F. M. Menger,* J. S. Keiper, and V. Azov Department of Chemistry, Emory University, 1515 Pierce Drive, Atlanta, Georgia 30322 Received August 5, 1999 A total of 24 dicationic geminis surfactants were synthesized, having a 2-butynyl spacer and chain lengths varying from 8 to 18 carbons. Geminis with more flexible butyl spacers were also synthesized for comparison purposes. Tensiometry gave critical micelle concentration (cmc) values, surface activity, and Gibbs areas. Film studies on the C18 geminis gave liftoff and collapse areas. The acetylenic geminis behave normally with the log cmc values decreasing linearly with chain length for all the dibromide salts. Anomalous behavior occurred, however, with the C18 gemini having a butyl spacer. It was found that cmc data correlate with spacer area, a relationship ascribed to packing constraints on curved surfaces and, possibly, to steric inhibition of ion binding. The effects are accentuated with the C18 surfactants. Geminis with heterocyclic headgroups were examined.

Introduction In the early 1970s Bunton and co-workers published a series of papers on micellar catalysis of nucleophilic substitution and decarboxylation reactions using dicationic detergentsscompounds containing two quaternary ammonium surfactant monomers connected at the headgroup level by an alkyl or 2-butynyl spacer.1-3 Nearly

three decades later, research into such surfactants, now commonly referred to as gemini4,5 (or dimeric)6 surfactants, has resurged as their exceptional surface activity and widening utility have become apparent.7 Additionally, the gemini format allows for expanded structural diversity in surfactant chemistry, as headgroups, hydrophobic units, spacers, and counterions can each be varied in a search for enhanced performance. In the present work, we revisit the 2-butynyl spacer for gemini surfactants, inspired in particular by the skeleton structures of two hypotensive agents:8 1,4-bis(pyrrolidino)2-butyne (3, commonly known as tremorine) and 1,4-bis(piperidino)-2-butyne (4). Preparing geminis based on

3

4

(1) Bunton, C. A.; Robinson, L.; Schaak, J.; Stam, M. F. J. Org. Chem. 1971, 36, 2346. (2) Bunton, C. A.; Kamego, A.; Minch, M. J. J. Org. Chem. 1972, 37, 1388. (3) Bunton, C. A.; Minch, M. J.; Hidalgo, J.; Sepulveda, L. J. Am. Chem. Soc. 1973, 95, 3262. (4) Menger, F. M.; Littau, C. A. J. Am. Chem. Soc. 1991, 113, 1451. (5) Menger, F. M.; Littau, C. A. J. Am. Chem. Soc. 1993, 115, 10083. (6) Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072. (7) Rosen, M. J.; Tracy, D. J. J. Surfactants Detergents 1998, 1, 547. (8) Biel, J. H.; DiPierro, F. J. Am. Chem. Soc. 1958, 80, 4609.

these motifs allowed us to address two issues. First, we could determine the viability among gemini surfactants of heterocyclic headgroups such as pyrrolidine and piperidine. Heterocycles have been commonly used as surfactant headgroups, but such surfactants are often nonionic9-11 and of low water solubility, or else they are asymmetrically quaternized at the positively charged nitrogen.12 The gemini structure, however, can lead to a symmetrical disposition of charged heterocyclic groups. Second is the issue of spacer flexibility or rigidity. It has been shown both experimentally13-15 and theoretically16-19 that the spacer occupies a major role in the colloidal properties of geminis. To date, however, most studies have involved fully saturated alkyl groups as flexible spacers6,13-15,20,21 or aryl groups as rigid spacers.4,5,22,23 Thus, the 2-butynyl group provides a spacer of “intermediate” bending stiffness. Using these geminis, we will show how the distinction between flexible and rigid spacers is not necessarily clear-cut in regard to micellization, especially at chain lengths where dicationic geminis deviate from “classical” critical micelle concentration (cmc) behavior. To pursue this study, we prepared five series of cationic gemini surfactants with four-carbon spacers and alkyl chains between 8 and 18 carbons in length (shown below). In addition to the two heterocyclic gemini forms (A and B), dibromide and dichloride versions of an acetylenic dicationic surfactant (C and D) were synthesized, along with so-called m-s-m geminis (E) (where m and s (9) Rosen, M. J.; Zhu, Z. H.; Gu, B.; Murphy, D. S. Langmuir 1988, 4, 1273. (10) Rosen, M. J.; Gu, B.; Murphy, D. S.; Zhu, Z. H. J. Colloid Interface Sci. 1989, 129, 468. (11) Zhu, Z. H.; Yang, D.; Rosen, M. J. J. Am. Oil. Chem. Soc. 1989, 66, 998. (12) Savelli, G.; Focher, B.; Bunton, C. A. Colloids Surf. 1990, 48, 29. (13) Hirata, H.; Hattori, N.; Ishida, M.; Okabayoshi, H.; Frusaka, M.; Zana, R. J. Phys. Chem. 1995, 99, 17778. (14) De, S.; Aswal, V. K.; Goyal, P. S.; Bhattacharya, S. J. Phys. Chem. 1996, 100, 11664. (15) Aswal, V. K.; De, S.; Goyal, P. S.; Bhattacharya, S.; Heenan, R. K. Phys. Rev. E 1998, 57, 776. (16) Diamant, H.; Andelman, D. Langmuir 1994, 10, 2910. (17) Diamant, H.; Andelman, D. Langmuir 1995, 11, 3605. (18) Maiti, P. K.; Chowdhury, D. Europhys. Lett. 1998, 41, 183. (19) Maiti, P. K.; Chowdhury, D. J. Chem. Phys. 1998, 109, 5126. (20) Alami, E.; Beinert, G.; Marie, P.; Zana, R. Langmuir 1993, 9, 1465. (21) Esumi, K.; Taguma, K.; Koide, Y. Langmuir 1996, 12, 4039. (22) Song, L. D.; Rosen, M. J. Langmuir 1996, 12, 1149. (23) Karthaus, O.; Shimomura, M.; Hioki, M.; Tahara, R.; Nakamura, H. J. Am. Chem. Soc. 1996, 118, 9174.

10.1021/la9910576 CCC: $19.00 © 2000 American Chemical Society Published on Web 12/10/1999

Gemini Surfactants with Acetylenic Spacers

Langmuir, Vol. 16, No. 5, 2000 2063 Scheme 1

represent alkyl chain and spacer carbon number, respectively) having completely saturated butyl spacers.

Results and Discussion Gemini surfactants A and B having 8-18 carbon chains were prepared in two-step procedures starting with 1,4dichloro-2-butyne as shown in Scheme 1. Surfactants C-E

having 8-18 carbon chains were obtained by reacting an appropriate 1,4-dihalo compound with N,N-dimethylalkylamines. Acetone was an optimal medium for the diquaternization step, as in most cases the dialkylated product precipitated from solution while the monoalkylated compound (the major side product) remained dissolved. Difficulty in the workup of the shorter chain analogues (8-12 carbons) was often encountered owing to their hygroscopic nature. Further details on the synthesis and characterization (1H and 13C NMR, highresolution mass spectra, elemental analyses) of these compounds can be found in the Supporting Information. Suffice it to mention here that all analytical methods indicated high levels of purity. Surface tensiometry on the gemini surfactants (carried out by the du Nuoy method) is profiled in Table 1 and Figure 1. Solubility properties allowed tensiometry to be performed at 23 °C for surfactants with C8-C14 chains, while 50 °C was necessary for C12-C18 chain lengths. As expected, the cationic gemini surfactants gave critical

2064

Langmuir, Vol. 16, No. 5, 2000

Menger et al.

Table 1. Values of the cmc, γcmc, and C20 for Geminis A-E surfactant

cmc (M)

γcmc (mN/m)

C20 (M)

temperature (°C)

A-8 A-10 A-12

3.45 × 10-2 5.31 × 10-3 9.10 × 10-4 1.27 × 10-3 1.30 × 10-4 2.68 × 10-4 4.40 × 10-5 1.05 × 10-5

42.8 40.0 39.7 39.9 40.4 39.5 39.4 46.2

1.18 × 10-2 1.59 × 10-3 2.69 × 10-4 6.15 × 10-4 5.21 × 10-5 1.26 × 10-4 2.18 × 10-5 8.78 × 10-6

23 23 23 50 23 50 50 50

B-14 B-16 B-18

4.78 × 10-2 5.96 × 10-3 8.92 × 10-4 1.14 × 10-3 2.07 × 10-4 5.97 × 10-5 1.66 × 10-5

43.1 39.6 38.5 39.1 35.0 37.9 41.5

1.75 × 10-2 1.52 × 10-3 2.08 × 10-4 4.75 × 10-4 5.62 × 10-5 2.46 × 10-5 8.50 × 10-6

23 23 23 50 50 50 50

C-8 C-10 C-12 C-14 C-16 C-18

2.55 × 10-2 1.39 × 10-3 1.13 × 10-3 2.50 × 10-4 4.78 × 10-5 1.71 × 10-5

41.8 41.1 38.7 38.8 44.7 45.8

1.08 × 10-2 6.01 × 10-4 4.64 × 10-4 1.09 × 10-4 3.97 × 10-5 1.36 × 10-5

23 23 50 50 50 50

D-8 D-10 D-12

7.40 × 10-2 9.38 × 10-3 1.57 × 10-3 1.43 × 10-3 2.97 × 10-4 3.93 × 10-4 7.59 × 10-5 4.33 × 10-5

46.2 45.9 44.6 42.6 45.7 42.5 43.0 46.8

3.69 × 10-2 4.99 × 10-3 7.05 × 10-3 8.57 × 10-3 1.49 × 10-4 1.94 × 10-4 4.76 × 10-5 3.72 × 10-5

23 23 23 50 23 50 50 50

5.61 × 10-2 8.78 × 10-3 9.31 × 10-4 1.36 × 10-3 1.35 × 10-4 2.78 × 10-4 4.22 × 10-5 4.19 × 10-5

41.6 40.2 39.9 39.1 40.5 39.8 41.8 43.4

1.98 × 10-2 2.49 × 10-3 2.62 × 10-4 5.56 × 10-4 4.69 × 10-5 1.28 × 10-4 2.44 × 10-5 2.46 × 10-5

23 23 23 50 23 50 50 50

A-14 A-16 A-18 B-8 B-10 B-12

D-14 D-16 D-18 E-8 E10 E-12 E-14 E-16 E-18

Figure 1. Surface tension vs log concentration for C12-chain geminis at 50 °C. (9) A-12; (0) B-12; (b) C-12; (O) D-12; (2) E-12.

micelle concentration (cmc) values significantly lower than their single-chained cousins.7 For corresponding C8-C16 chain-length geminis, the surface activity among the five series proved quite similar; two relevant points will be mentioned here. (1) Solubilities for the heterocyclic A and B series surfactants exceeded those of the acyclic C series. For example, among the C12 surfactants, A-12 has a solubility in water at 23 °C nearly 400 times that of C-12 (approximate values: A-12, 0.38 M; B-12, 0.12 M; C-12, 0.00097 M; D-12, 0.61 M; E-12, 0.16 M). (2) Surface activities can be judged by the γcmc values (the surface tension at the cmc) and the C20 values (the concentration that lowers the surface tension by 20 mN/m) in Table 1.

Table 2. Gibbs Equation-Derived Molecular Areas, MM2-Estimated Fully Extended, “All-Anti” Conformation Areas, and Fully Compressed, “Vertical Tails” Areas for C12-Chain Geminis

surfactant A-12 B-12 C-12 D-12 E-12

Gibbs area (Å2/molecule) 23 °C 50 °C 121 132 132 127

121 131 128 127 135

“all-anti” area (Å2/molecule)

“vertical tails” area (Å2/molecule)

163 163 163 163 168

94.7 111 76.9 76.9 81.3

It is seen that heterocyclic geminis A and B possess surface activities comparable to or better than the acyclic dibromide surfactants C and E. Dipiperidinium geminis B have γcmc values modestly lower than the other geminis. This may be a manifestation of the additional hydrocarbon content at the interface provided by the rings. Molecular areas of C8-C16 geminis, determined by applying the Gibbs equation to the tensiometric data, revealed few differences among geminis of equivalent chain length (Table 2). Thus, little variance was detected regardless of headgroup, spacer type, or temperature. As also seen from Table 2, the empirical surface areas fall between areas estimated by molecular mechanics (MM2) for two possible conformational extremes of these geminis: (a) a fully extended, “all-anti” conformation and (b) a fully compressed, “vertical tails” configuration reflecting mainly the spacer/headgroup area. Both possibilities are shown for E-12 (Chart 1). While the models exclude factors such as counterions, electrostatic headgroup repulsion, and hydration, they are useful for establishing reasonable limits. For example, some previously published Gibbs-derived molecular areas for p-xylyl diammonium geminis by Song and Rosen22 gave areas (68, 47, 46, and 45 Å2/molecule for di-C12, C14 ,C16, and C18 analogues, respectively, in 0.1 N NaCl at 50 °C) well below the MM2-estimated value of 87 Å2/molecule for the “vertical tails” configuration. In such a case, an unconventional packing mode must be present. Conventional single-chain surfactants undergo linear decreases in log cmc with increasing chain lengths.24 Some gemini surfactants, however, have shown departures from “classical” linear trends as cmcs level off or increase with increasing chain lengths.5,22,25 Geminis with rigid aromatic or hydroxylated spacers have been reported to behave in this unconventional manner, but it has been vaguely proposed by Zana and Le´vy26 that m-s-m gemini surfactants do not deviate from linearity. They based their suggestion on three m-2-m dicationic geminis (as well as one system estimated by extrapolation) with chain lengths from C8 to C16. Having on hand a particularly rich array of compounds, we performed measurements on geminis A-E from C12 to C18 at 50 °C to test for possible deviations. The results are given in Table 1 and Figure 2. Geminis A-C, with 2-butynyl spacers, show linear decreases in cmc with increasing chain length (correlation coefficients: A series, 0.998; B series, 0.994; C series, 0.991), while series D slightly deviates from linearity at its C18 analogue. A more pronounced deviation is shown with the m-s-m gemini E series, where the cmc for E-16 (4.22 × 10-5 M) is equivalent to that of the longer-chained E-18 (4.19 × 10-5 M). It would seem, therefore, that nonlinear behavior is not simply a matter of geminis (24) Myers, D. Surfactant Science and Technology, VCH: New York, 1988; p 107. (25) Rosen, M. J.; Liu, L. J. Am. Oil Chem. Soc. 1996, 73, 885. (26) Zana, R.; Le´vy, H. Colloids Surf. A 1997, 127, 229.

Gemini Surfactants with Acetylenic Spacers

Langmuir, Vol. 16, No. 5, 2000 2065 Chart 1

Table 3. Liftoff Area, Collapse Area, and Πcollapse for C18-Chain Surfactants Determined with Kibron µTrough S at 23 °C, and Gibbs-Derived Molecular Areas Determined at 50 °C surfliftoff areaa collapse areaa actant (Å2/molecule) (Å2/molecule) A-18 B-18 C-18 D-18 E-18 a

164 ( 8 168 ( 4 166 ( 7 187 ( 11 181 ( 3

51.4 ( 3.3 46.3 ( 2.9 42.2 ( 3.1 52.0 ( 7.8 57.5 ( 6.1

Πcollapsea (mN/m)

Gibbs area (Å2/molecule)

42.7 ( 0.7 39.9 ( 0.6 42.6 ( 1.0 42.1 ( 0.5 40.5 ( 0.6

133 138 138 171 154

Average of 10 repeat runs.

Figure 2. Log cmc vs chain carbon number for C12-C18-chain geminis at 50 °C. (9) A; (0) B; (b) C; (O) D; (2) E.

having rigid or hydroxylated spacers. Indeed, surfactants A-C contain a rigid acetylenic unit but behave normally, while surfactants E are flexible and yet ultimately become nonlinear. Information on the packing abilities of surfactants can be obtained from monolayer film studies.27,28 Consequently, pressure-area isotherms were determined for C18-chain surfactants at 23 °C where the geminis are sufficiently water-insoluble to carry out the experiments. Results of 10 averaged repetitions per surfactant are shown in Figure 3 and Table 3. Liftoff areas for geminis A-18, B-18, and C-18 are smaller than those of D-18 and E-18, a comparison also true for the Gibbs areas (determined at 50 °C where C18 geminis are more water-soluble). All geminis collapsed at surface pressures of 40-43 mN/ m. Two points stand out from these data. First, the larger liftoff areas were found for D-18 and E-18, which are also the surfactants with unusually high cmcs. In other words, those C18 geminis that have difficulty packing at an air/ water interface also possess higher cmcs. Second, there are clear differences in cmc, liftoff area, and Gibbs area between C-18 and D-18, geminis with bromide and chloride counterions, respectively. For D-18, headgroup(27) Menger, F. M.; Wood, M. G., Jr.; Richardson, S.; Zhou, Q.; Elrington, A. R.; Sherrod, M. J. J. Am. Chem. Soc. 1988, 110, 6797. (28) Sumida, Y.; Masuyama, A.; Oki, T.; Kida, T.; Nakatsuji, Y.; Ikeda, I.; Nojima, M. Langmuir 1996, 12, 3986.

Figure 3. Pressure-area isotherms at 23 °C for C18-chain geminis A-E.

headgroup repulsion is greater due to its more loosely bound Cl- counterions, leading to higher cmcs and interfacial area parameters (Tables 1 and 3). Such behavior is well-known for conventional surfactants,29 and our bromide/chloride data show that ion binding effects cannot be ignored with geminis as well. In Table 4 are compiled cmcs (at 50 °C), MM2 energyminimized N+-N+ distances, and van der Waals spacer surface areas for C18 geminis. Included in the table are 18-p-xylyl-18 from our previous study5 and m-s-m geminis 18-3-18 and 18-5-18. It is seen that all geminis have N+-N+ distances below the approximate equilibrium (29) Rosen, M. J Surfactants and Interfacial Phenomena, 2nd ed.; Wiley: New York, 1989; p 135.

2066

Langmuir, Vol. 16, No. 5, 2000

Menger et al.

Table 4. Critical Micelle Concentrations at 50 °C, N+-N+ Distances, and van der Waals Surface Areas for C18-Chain Gemini Surfactants

surfactant

cmc (M)

N+-N+ distancea (Å)

A-18 B-18 C-18 18-3-18 18-4-18 (E-18) 18-5-18 18-p-xylyl-18

1.05 × 10-5 1.66 × 10-5 1.71 × 10-5 5.49 × 10-5 4.19 × 10-5 2.22 × 10-4 3.70 × 10-4 c

6.15 6.16 6.16 5.23 6.51 7.76 7.80

a

spacer surface areab (cm2/mol × 109) 3.68 3.68 3.68 4.05 5.40 6.75 8.71

MM2-calculated values. b Estimated from ref 30. c From ref 5.

Figure 4. Log cmc vs estimated spacer volume for C18-chain geminis. Points correspond to data in Table 4.

headgroup separation for CTAB micelles (7.9 Å).14 It is interesting to note that 18-5-18 and 18-p-xylyl-18, whose spacers possess N+-N+ distances closest to the equilibrium value, have cmc values most similar to that of octadecyltrimethylammonium bromide (2.5 × 10-4 M at 25 °C).31 For these two C18 surfactants, the “gemini effect” on cmcs is essentially eliminated. A clear dependence of cmc on N+-N+ distance (i.e., spacer length) is not present for this set of seven geminis. Log cmc values plotted against spacer area, however, create a reasonably smooth curve of increasing cmc with larger spacer area (Figure 4). Since issues of micelle structure and properties are almost always complex and multifaceted, we can only speculate as to why cmcs get larger as the C18 gemini spacers increase in area from 2-butynyl to pentyl in Figure 4 and Table 4. Note that this runs counter to what one would expect on a purely hydrophobic basis, where larger spacers should enhance the propensity to self-assemble. It is relevant in this regard that previous studies on C12H25NRMe2+Br- revealed that an increase in R’s chain length diminishes the cmc, an effect explained in terms of actual entry of the R groups into the micellar interior.32 A similar phenomenon for the geminis is not possible because spacers (unless they are long and flexible) are more-or-less confined to the micelle surface. In the absence of a compensating internalization, therefore, a larger spacer will impede micellization owing to the difficulties of packing the spacer on a curved surface. By way of analogy, it is easier to cover a ballroom light reflector with small mirrored squares than with large rectangular ones. As a consequence, the cmcs rise with spacer area in Figure 4. Indeed, as already mentioned, the cmc of 185-18 is similar to that of its single-chained analogue. We can use the same rationale to explain the abnormally high cmcs of the 18-4-18 gemini E-18 in Figure 2. Its (30) Bondi, A.; J. Phys. Chem. 1964, 68, 441. (31) Ouni, S.; Hafiane, A.; Dhahbi, M. J. Chim. Phys. 1998, 95, 911. (32) Zana, R. J. Colloid Interface Sci. 1980, 78, 330.

large surface-bound spacer resists the tight packing “desired” by the long C18 chains. Shorter chains, on the other hand, mean looser packing and easier accommodation of the spatial requirements of the spacers. Mention should be made of the potential role that ion binding may have in the anomalous cmc behavior of the C18 surfactants. The differences in cmc and pressurearea isotherm data for C-18 and D-18 demonstrate the sensitivity to counterion identity. In the work with C12H25NRMe2+Br- discussed above, an increase in ion dissociation (R) was found with lengthening of R, implying that the variable alkyl group sterically impedes N+/Brassociation.32 It is possible that a spacer-induced steric effect also operates with the C18 geminis (Figure 4). Thus, the cmcs should rise as the spacers increase in area, the counterions have greater difficulty in binding, and the micelles experience an overall higher cationic charge. Why should a presumed steric effect on ion-binding manifest itself only with chain lengths of 18? With increasing chain lengths, headgroup charge density becomes greater at the micelle surface.32 This serves to accentuate the sensitivity to ion binding as affected by the spacer area. Such speculation is preliminary and clearly in need of further data, particularly those from conductivity and interfacial chemical trapping studies that measure the degree of micelle/counterion association.33 Also badly needed in the gemini arena is NMR information on spacer conformation and on micelle morphology.13-15 In summary, five sets of C4-spacer dicationic gemini surfactants of variable headgroups, spacers, and counterions were prepared and studied for their surface activities. Cmcs were quite similar for dibromide surfactants from C8 to C16, proving heterocyclic surfactants A and B to be viable geminis. At C18 chain-length, m-s-m gemini E-18 deviated from the expected logarithmic decrease with increasing chain length, contrary to previous assumptions of m-s-m geminis. Comparison of cmc values for seven C18 dibromide gemini surfactants indicated a possible relationship between this deviatory behavior and spacer area, an effect explained in terms of packing constraints on the micelles. Experimental Section General Methods. 1H NMR and 13C NMR spectra were recorded on Varian INOVA 400, Varian Mercury 300, or General Electric QE 300 spectrophotometers. Mass spectral data were obtained from the Emory University Mass Spectrometry Center. Atlantic Microlabs (Norcross, GA) performed elemental analyses. Melting points were taken on a Thomas-Hoover apparatus and are uncorrected. Product yields were not optimized. Materials. All solvents and reagents were purchased from Aldrich except for N,N′-dimethyltetradecylamine, N,N′-dimethylhexadecylamine, and N,N′-dimethyloctadecylamine (Pfaltz and Bauer). Surface Tensiometry. Surface tension measurements were performed on a Fisher surface tensiomat du Nuoy apparatus with a 6 cm Pt/Ir ring. Milli-Q purified water (18 MΩ‚cm resistance) was used for all preparations. For measurements at 23 °C, 10 measurements were taken for each concentration (at least eight concentrations per sample to acquire an adequate cmc). Measurements at 50 °C were taken with a specially designed water-jacketed flask attached to a Lauda M3 circulating bath to control temperatures to (0.5 °C. For C18-chained surfactants, solutions were equilibrated for 1 h before single measurements were taken for each concentration.5 Extra care was taken to avoid disturbing the interface during measurements. (33) Chaudhuri, A.; Romsted, L. S.; Yao, J. J. Am. Chem. Soc. 1993, 115, 8362.

Gemini Surfactants with Acetylenic Spacers

Langmuir, Vol. 16, No. 5, 2000 2067

Gibbs molecular areas were determined with tensiometry data using

Γ)-

1 dγ 2.30nRT d log C

(

area ) (NAΓ)-1

)

T

(1) (2)

where Γ ) surface excess concentration, n ) 3 (constant used in previous Gibbs analyses of geminis with flexible spacers),20,21 R ) 8.32 J mol-1 K-1, T ) temperature in kelvins (here, 296.15 or 323.15 K), γ ) surface tension in newtons per meter, C ) concentration in moles per liter, and NA ) 6.022 × 1023 (Avogadro’s number).34 Pressure-Area Isotherms. A Kibron µTrough was used for pressure-area isotherm measurements. The general procedure was as follows: The trough and Teflon barriers were thoroughly cleaned with ethanol and deionized water and dried with compressed air. The conducting pin was also rinsed in ethanol and flame-dried. On a level marble table, 18 mL of Milli-Q purified water (18 MΩ‚cm resistance) was applied as the subphase and evened manually with the barriers. The instrument was then initialized and the air/water interface was calibrated (∆V ∼ 4000 mV). The barriers were adjusted to their starting positions and the “zero” pressure was set. The monolayer of surfactant was formed by application of 10 µL of CHCl3/MeOH solution [9/1 or 7/3 (v/v) for B-18] via a Hamilton syringe to the surface of the (34) Li, Z. X.; Dong, C. C.; Thomas, R. K. Langmuir 1999, 15, 4392.

water. The solution was applied ∼3 cm from the conducting pin in 10-12 droplets. The film was subsequently allowed to dry for 10 min. At that time, the barriers were compressed at a rate of 4 Å2 chain-1 min-1 and the data were recorded. After completion, the apparatus was vigorously cleaned and the protocol was repeated (10 times for each surfactant). MM2 Calculations. Calculations were carried out on a Micron PC using CS Chem 3D Pro software. Molecules were energy minimized in the gas phase using the MM2 force field with a minimum root-mean-square gradient of 0.025. To estimate “allanti” areas, 180° 1,4-dihedral angles were imposed on the carbon chains. For “vertical tails” spacer-headgroups area estimations, the N-methylenes of the carbon chains were forced perpendicular to the plane of the spacer-headgroups. Area estimations were then based on rectangular cross-sections of the molecules, with widths and lengths based upon furthest atomic distances plus van der Waals radii (2 × 1.20 Å), except for the “all-anti” areas, with which the width of geminal hydrogens on the chains plus van der Waals radii (4.2 Å) was used as the “average” width of the fully extended molecule.

Acknowledgment. This work was supported by the Army Research Office. Supporting Information Available: Details of synthesis and characterization of 3, 4, and gemini surfactants A-E. Supporting Information is available free of charge via the Internet at http://pubs.acs.org. LA9910576