Synthesis and Characterization of Optically Active Surfactants Derived from Phenylalanine and Leucine

Journal of Colloid and Interface Science 234, 122–126 (2001) doi:10.1006/jcis.2000.7285, available online at http://www.idealibrary.com on

Synthesis and Characterization of Optically Active Surfactants Derived from Phenylalanine and Leucine Michael John Diego-Castro,∗ Helen Claire Hailes,∗,1 and Margaret Jayne Lawrence† ∗ Department of Chemistry, University College London, London WC1H OAJ, United Kingdom; and †Department of Pharmacy, King’s College London, London SE1 8WA, United Kingdom Received April 18, 2000; accepted October 16, 2000

RATIONALE OF SURFACTANT DESIGN Two optically active cationic surfactants, (2S)-N-hexadecyl-N, N-dimethyl-(1-hydroxy-3-phenylpropyl)-2-ammonium chloride 1 and (2S)-N-hexadecyl-N,N-dimethyl-(1-hydroxy-4-methylpentyl)2-ammonium chloride 2, have been selected and synthesized for use as enantioselective micellar catalysts in aqueous media. Their surface and aggregation behavior has been investigated at 298 K using surface tension and light scattering studies, which revealed that both molecules associate at low concentrations to produce micellar aggregates. Interestingly, although the area per molecule occupied ˚ 2 for 1 and by the surfactants at the air–water interface (43.6 A 2 54.6 A˚ for 2) is similar to that of related cationic surfactants, their aggregation number (23 for 1 and 19 for 2) is much smaller, perhaps reflecting the influence of the size or homochiral nature of the head group in the packing of the micelle. °C 2001 Academic Press Key Words: physicochemical properties; enantioselective catalysis; micelles; critical micelle concentration; aggregation.

INTRODUCTION

Aqueous micellar media have the ability to confer special properties on reactions due to their ability to solubilize, concentrate, and orient reactants within the micelle (1). In addition, they provide an alternative reaction medium to volatile organic solvents (VOCs) for use in organic synthesis. Of particular interest in the present study is the potential of homochiral micellar solutions to confer enantioselectivities on organic reactions. While this field undoubtedly exhibits considerable promise, there are only a few investigations reporting the use of chiral micellar media for such purposes (2, 3). Indeed in comparison to the large amount of literature detailing the synthesis of surfactants, relatively few describe the synthesis of homochiral surfactants (4). The aim of our current study was to synthesise two novel, structurally related, homochiral surfactants, to measure their physicochemical properties and then examine their use as media for asymmetric catalytic organic applications (3).

1

To whom correspondence should be addressed. E-mail: [email protected].

C 2001 by Academic Press Copyright ° All rights of reproduction in any form reserved.

SYNTHESIS

The syntheses of 1 and 2 are outlined in Scheme 1. The Meyer (6) reduction of phenylalanine and leucine gave the amino alcohols in 99 and 92% yield which were N -dimethylated using formic acid and formaldehyde (7) in 98 and 80% yield, respectively. Subsequent N -alkylation using 1-chlorohexadecane gave the pure quaternary ammonium salts 1 and 2 in 24 and 22% yield after several recrystallizations.

SCHEME 1. The synthesis of 1 and 2. Reagents and conditions: (i) NaBH4 , I2 , THF, 99% (R = Ph), 92% (R = CHMe2 ); (ii) 37% CH2 O, 98% HCO2 H, H2 O, 98% (R = Ph), 80% (R = CHMe2 ); (iii) CH3 (CH2 )15 Cl, 24% (R = Ph), 22% (R = CHMe2 ).

uk. 0021-9797/01 $35.00

In addition to the basic requirement for chirality, other factors were considered to be important when designing the surfactants. First, for use in asymmetric synthetic applications we reasoned that, since the reaction substrates are most likely to reside in the head group region of the micelle, the chiral moiety should also be located in this area. Second, we required cationic surfactants, since in our applications (Diels–Alder cycloadditions (5)), the ability of a cationic surfactant to complex to the carbonyl moiety of the dienophile could facilitate the reaction. Third, the presence of an alcohol functionality was considered to be advantageous to assist in the complexation of the dienophile to the surfactant head group by way of hydrogen bonding. Finally, hexadecyl chains were used as the hydrophobe in an attempt to generate as large a micelle as possible, thereby ensuring that the aggregates would have a high solubilizing capacity for the reactants. Accordingly, surfactants 1 and 2 were selected and prepared from S-phenylalanine and S-leucine, respectively, bearing a hexadecyl chain on the nitrogen.

122

SYNTHESIS AND CHARACTERIZATION OF SURFACTANTS

Materials Unless otherwise indicated, reagents were obtained from commercial suppliers and were used without further purification. Ether was freshly redistilled under anhydrous conditions from sodium-benzophenone ketal. Flash column chromatography (8) was carried out using silica gel (particle size 40–63 µm) purchased from BDH. The 1 H NMR spectra were recorded at 300 MHz on a Bruker AMX300 or at 400 MHz on a Varian VXR-400 instrument. The 13 C NMR spectra were recorded at 75.4 MHz on a Bruker AMX300 or at 100.6 MHz on a Varian VXR-400 instrument. Residual protic solvent was taken as internal standard, with ˚ molecular sieves and filtered through CDCl3 (stored over 4 A basic alumina prior to use) or the solvent stated. Mass spectra were taken on an Autospec Q, VG 7070, or VG 7070B instrument with sources for electron ionization (EI) or a ZAB-SE instrument for fast atom bombardment (FAB). Infrared spectra were recorded on a Perkin–Elmer FT-IR 1605 spectrometer as thin films on sodium plates, or as potassium bromide discs. Melting points were taken on a Reichert hot stage instrument and are uncorrected. Optical rotations were measured on a JASCO 600 spectrophotometer and an Optical Activity POLAAR 2000 polarimeter using sucrose as a standard in the solvent indicated. Synthesis of 1 (2S)-2-Amino-3-phenyl-1-propanol. This compound was prepared as previously described (6). (2S)-N,N-Dimethyl-2-amino-3- phenyl-1-propano1. To (2S)-2-amino-3-phenyl-1-propanol (1.51 g, 10 mol) in water (2.50 ml) was added 37% formaldehyde solution (3.00 ml, 34.1 mmol) followed by 88% aqueous formic acid (3.50 ml, 54.9 mmol). The reaction mixture was stirred at room temperature for 5 h. The temperature was increased to 95◦ C and the reaction was heated at reflux for 18 h. The mixture was cooled to room temperature and the pH adjusted to 12 with 2 M potassium hydroxide (100 ml) and extracted with diethyl ether (3 × 100 ml). The combined organic extracts were washed with distilled water (3 × 100 ml) and brine (100 ml) and then dried over sodium sulfate. The solvent was removed in vacuo. The crude product was purified by recrystallization from diethyl ether/hexane (1 : 3) to yield colorless crystals of (2S)-N,N-dimethyl-2-amino-3-phenyl-1-propanol (1.71 g, 98%). Alternatively it could be purified by flash chromatography (dichloromethane/methanol, 10 : 1), Mp 49–51◦ C (diethyl ether/hexane, 1 : 3) (9); [α]D = −18.4◦ (25◦ C), c 1.0 dichloromethane; νmax (KBr)/cm−1 3310br (OH), 2927s, 2865s, and 1077s; δH (400 MHz; CDCl3 ) 7.27 (2H, m, Ph-H ), 7.13 (3H, m, Ph-H ), 3.48 (1H, dd, J 10.2, and 5.0 Hz, 1-HH ), 3.38 (1H, dd, J 10.2, and 8.4 Hz, 1-H H), 3.10 (1H, m, 2-H), 2.81 (1H, dd, J 12.9, and 5.0 Hz, 3-H H), 2.62 (1H, dd, J 12.9, and 8.4 Hz, 3HH ), and 2.36 (6H, s, N(CH3 )2 ); δC (101 MHz; CDCl3 ) 139.04 (C-10 ), 129.32 (2 × C), 129.11 (2 × C), 126.24 (C-40 ), 63.00

123

(C-1), 51.78 (C-2), 42.28, and 42.12 (2 × Me), and 39.30 (C-3); m/z (EI) 180.1386 (MH+ , 83%. C11 H18 NO requires 180.1388); m/z (FAB) 162 (M+ -OH, 21), 148 (M+ -H-Me2 , 8). (2S)-N-Hexadecyl-N,N-dimethyl-(1-hydroxy-3-phenylpropyl)-2-ammonium chloride 1. (2S)-N ,N -Dimethyl-2-amino3-phenyl-1-propanol (5.74 g, 32.0 mmol) and 1-chlorohexadecane (8.36 g, 32.0 mmol) were stirred under anhydrous conditions at 90◦ C for 40 h, then cooled to room temperature. Diethyl ether (100 ml) was added to precipitate the quaternary amine. The precipitate of (2S)-N-hexadecyl-N,N-dimethyl-(1hydroxy-3-phenylpropyl)-2-ammonium chloride 1 was filtered under gravity and the was solid air-dried. The crude product was recrystallized several times from ethyl acetate/hexane (1 : 5) to yield colorless crystals of the desired product (3.31 g, 24%), Mp 119–120◦ C (ethyl acetate/hexane, 1 : 5); cmc (surface tension) 42 mg/L; [α]D = −54.0◦ (25◦ C), c 1.1 ethanol; νmax (KBr)/cm−1 3296s (OH), 2952s, 2851s, 1470s, 1454s, and 1425m; δH (400 MHz; d6 -DMSO) 7.32 (5H, m, Ph), 4.97 (1H, br t, J 4.4 Hz, s, OH), 3.30 (1H, m, CH HOH), 3.10 (1H, m, CHH OH), 3.00 (1H, m, 2-H), 2.80 (2H, m, 3-HH), 2.68 (6H, s, N(CH3 )2 ), 2.65 (2H, t, J 8.0 Hz, N-CH2 CH2 ), 1.25 (2H, m, NCH2 CH2 ), 0.80 (26H, m, alkyl chain), and 0.35 (3H, m, CH2 CH3 ); δH (300 MHz; D2 O) 6.99 (5H, m, Ph), 3.60 (1H, d, J 12.4 Hz, CH HOH), 3.41 (1H, br d, J 12.4 Hz, CHH OH), 3.00 (3H, m, 2-H, and NCH2 ), 2.81 (6H, s, N(CH3 )2 , and 1H, m, 3-H H), 2.72 (1H, m, 3-HH ), 1.30 (2H, m, NCH2 CH2 ), 0.99 (26H, m, alkyl chain), and 0.60 (3H, t, J 6.2 Hz, CH2 CH3 ); δC (75 MHz; d6 -DMSO) 137.73 (C-10 ), 129.66 (2 × C), 128.77 (2 × C), 127.63 (C-40 ), 73.52 (C-1), 63.32 (C-2), 56.29 (N-CH2 ), 49.39, and 49.25 (2 × NCH3 ), 31.49, 30.43, 30.26, 29.26—28.56 (9 × CH2 ), 26.06, 22.29, 22.08, and 14.14 (CH2 CH3 ); m/z (EI) 404.3891 (M+ - Cl; C27 H50 NO requires 404.3892); m/z (FAB) 404 (M+ -Cl, 100%), 307 (MH+ -Cl-C7 H14 , 30), and 136 (MH+ -Cl-C18 H39 N, 19). Synthesis of 2 (2S)-2-Amino-4-methyl-1-pentanol. This compound was prepared as previously described (6). (2S)-N,N-Dimethyl-2-amino-4-methyl-1-pentanol(7). To (2S)-2-amino-4-methyl-1-pentanol (4.04 g, 35.0 mmol) in water (8.70 ml, 0.50 mol) was added 37% formaldehyde solution (3.90 ml, 0.14 mmol) followed by 88% aqueous formic acid (12.3 ml, 0.30 mol). The reaction mixture was stirred continually at room temperature for 5 h, then the temperature was increased to 95◦ C and the mixture was heated at reflux for 18 h. The mixture was cooled to room temperature and the pH was adjusted to 12 with 2 M potassium hydroxide (100 ml) and extracted with diethyl ether (3 × 100 ml). The combined organic extracts were washed with distilled water (3 × 100 ml) and brine (100 ml), then dried over sodium sulfate. The solvent was removed in vacuo to yield an oil (2S)-N,N-dimethyl-2amino-4-methyl-1-pentanol (4.00 g, 80%); [α]D = +0.052◦ , c 1.05 ethanol; νmax (film)/cm−1 3405s (OH), 2954s, 1462s, 1262s, 1367m, and 1046s; δH (300 MHz; CDCl3 ) 3.46 (1H, br d,

124

DIEGO-CASTRO, HAILES, AND LAWRENCE

J 10.4 Hz, 1-H H), 3.10 (1H, dd, J 10.4, and 10.4 Hz, 1-HH ), 2.62 (1H, m, 2-H), 2.21 (6H, s, NMe2 ), 1.44 (3H, m, 3-H2 , and 4-H) and 0.86 (6H, d, J 6.5 Hz, CH(CH3 )2 ); δC (101 MHz; CDCl3 ) 62.24 (C-1), 51.10 (C-2), 39.65 (2 × NCH3 ), 32.65 (C-3), 23.64 (C-4), 21.90 (CH(CH3 )2 ); m/z (EI) 144.1380 (M+ -H, 56%; C8 H18 NO requires 144.1388), 98 (C6 H12 N+ , 8), 72 (M+ -C4 H9 O, 76), 57 (C4 H9 + , 35). (2S)- N -Hexadecyl- N ,N -dimethyl-(1-hydroxy-4-methylpentyl)-2-ammonium chloride 2. (2S)-N ,N -Dimethyl-2amino-4-methyl-1-pentanol (4.00 g, 30.0 mmol) and 1-chlorohexadecane (8.40 g, 28.0 mmol) was stirred under anhydrous conditions at 90◦ C for 40 h, then cooled to room temperature. Diethyl ether (100 ml) was added to precipitate the quaternary amine. The precipitate of (2S)-N-hexadecyl-N,N-dimethyl-(1hydroxy-4-methylpentyl)-2-ammonium chloride was filtered under gravity and the white solid was air-dried. The crude product was recrystallized several times from ethyl acetate and hexane, 1 : 5, to yield colorless crystals of the desired product (2.70 g, 22%), Mp 72–74◦ C (ethyl acetate/hexane, 1 : 5); cmc (surface tension) 160 mg/L; [α]D = +0.070◦ (25◦ C), c 0.025 ethanol; νmax (film)/cm−1 3237br (OH), 2910s, 1632w, 1469s, 1376m, 1039s cm−1 ; δH (400 MHz; D2 O) 4.04 (1H, d, J 14.0 Hz, CH HOH), 3.91 (1H, dd, J 14.0, and 4.5 Hz, CHH OH), 3.35 (3H, m, 2-H, and NCH2 ), 3.14 (3H, s, NMe), 3.05 (3H, s, NMe), 1.85 (2H, m), 1.64 (2H, m), 1.29 (27H, m, 4H, and 13 × CH2 ), 0.97 (3H, d, J 7.5 Hz, CHCH3 ), 0.92 (3H, d, J 7.5 Hz, CHCH3 ), 0.84 (3H, t, J 7.0 Hz, CH2 CH3 ); δC (75 MHz; d6 -DMSO) 70.85 (C-1), 63.20 (C-2), 55.25 (N-CH2 ), 49.04, and 48.66 (2× NCH3 ), 31.19, 28.95–28.43 (9 × CH2 ), 25.74, 25.12, 24.97, 23.09, 21.99, 21.86, 20.53, 17.69, and 13.85 (CH2 CH3 ); m/z (EI) 370.4106 (M+ - Cl; C24 H52 NO requires 370.4149); m/z (FAB) 370 (M+ -Cl, 100%), 72 (M+ -Cl-C16 H33 -C4 H9 O, 5), 57 (C4 H+ 9 , 19).

TOTAL INTENSITY LIGHT SCATTERING STUDIES

Total intensity light scattering (TILS) measurements were performed on aqueous solutions of 1 and 2 at 298 ± 0.1 K using a Malvern 4700c digital autocorrelator equipped with a 75-mW Argon ion laser operating at 488 nm. The sample was clarifed via ultrafiltration through a 0.1-µm filter (Millipore, cellulose acetate) and its scattering was determined at 45◦ , 90◦ , and 135◦ to the incident beam. As the ratio of scattering intensities at angles of 45◦ and 135◦ were in the range 1.00 ± 0.05, further measurements were made at 90◦ only. The instrument was calibrated with a dust-free sample of benzene (AnalaR, BDH Chemicals, UK) using an R90 for benzene (10) at 488 nm and 298 K of 32.00 × 10−6 cm−1 and a refractive index for benzene (at 488 nm and 298 K) of 1.50993. Specific refractive-index measurements (δn/δc) required for analysis of the total intensity light scattering data were determined using an Abbe 60 ED precision refractometer (Bellingham and Stanely Ltd.) and a sodium lamp. The experimentally determined specific refractive index incre-

ment values were 0.0542 kg mol−1 for 1 and 0.0464 kg mol−1 for 2. SURFACE TENSION MEASUREMENTS

The surface tension (γ ) of the aqueous surfactant solutions was measured at 298 ± 0.1 K using the Wilhelmy plate method by the detachment of a roughened glass plate suspended from a microforce balance (CI electronics, UK) attached to an intelligent digital multimeter (Thurly 1905a, Radio Spares, UK). Prior to any measurements the microforce balance was calibrated with a series of known weights. Throughout a surface tension determination, the surface tension of pure water was frequently checked. The usual precautions were taken to ensure clean glassware and to prevent the adsorption of surfactant to the glassware. The results are the mean of two determinations. The critical micelle concentration (cmc) was determined from the break in the γ verses log10 concentration curve, with all the data below the cmc being fitted to a polynomial in order to determine the area per molecule at the cmc. CHARACTERIZATION OF SURFACTANTS

The surface and aggregation behavior of the surfactants in aqueous solution were established using a combination of surface tension and laser light scattering experiments. When dispersed in water at concentrations of up to several percent, the surfactants formed clear, foaming solutions. Surface tension experiments performed at 298 K using the Wilhelmy plate method showed a clear break in the surface tension against the surfactant concentration curve for both surfactants (Fig. 1), suggesting the presence of a cmc. The surface properties of the two surfactants are given in Table 1: values of 0.095 mmol/L (42 mg/L) and 0.39 mmol/L (160 mg/L) were obtained for 1 and 2, respectively, indicating, perhaps not surprisingly because of the increased lipophilic nature of its head group, that 1 was the more hydrophobic of the two surfactants. Although the cmc values of the surfactants 1 and 2 were very similar, they are lower than the values recorded for other C16 cationic surfactants, namely, 1.3 mmol/L at 303 K (11) for hexadecyltrimethylammonium chloride and 0.94 and 0.97 mmol/L at 298 K (12, 13) and 0.91 mmol/L at 298 K (14) for hexadecyltrimethylammonium bromide, suggesting that, not surprisingly, the head groups of the novel surfactants are less hydrophilic.

TABLE 1 Surface Properties Surfactant

cmc (mmol/L)

γcmc mNm−1

Area per molecule (A˚ 2 )

1 2

0.095 0.39

40 39

43.6 54.6

SYNTHESIS AND CHARACTERIZATION OF SURFACTANTS

125

method) have been discussed by Adamson (15) and Chattoraj and Birdi (16). These errors include the complete wetting of the plate. Indeed, it has been recorded that there are particular problems with the use of glass and platinum Wilhemy plates in obtaining an estimate of the area per molecule for cationic surfactants (17). Interestingly however, the head group area of the two surfactants was in the same range as the area per molecule at the air–water interface obtained for hexadecyltrimethyl am˚ 2, monium bromide using neutron reflection (14), namely, 43 A 2 ˚ , using and surface tension measurements (18), namely, 63 A the maximum ring pull method. Figure 2 shows the variation in S90 (the ratio of the scattering intensity of the micelles corrected for that of the solvent to the scattering intensity of benzene at 90◦ ) against surfactant concentration for the two surfactants at 298 K. The total intensity light scattering experiments gave a relationship between scattering intensity and surfactant concentration, over the concentration range studied, that was typical of a charged surfactant. Extrapolation of the light scattering data to zero scattering yielded cmcs of 0.26 and 0.50 mmol/L for surfactants 1 and 2, respectively. Although the values obtained for the cmc from the light scattering experiments are higher than those obtained from surface tension measurements, they are of a similar rank order. It is worth commenting that discrepancies between the cmc values obtained via different techniques are frequently noted (18). The aggregation numbers obtained for the surfactants using the analysis of Anacker and Westwell (19) were 23 for 1 and 19 for 2, indicating that the replacement of the leucine head group with a phenylalanine moiety leads to only very subtle changes in the aggregation properties of the molecules. Surprisingly, however, the values obtained for the micelles were much smaller

FIG. 1. Surface tension, γ , as a function of log concentration for the surfactants 1 (d) and 2 (j) in water at 298 K.

Calculations of the area per molecule occupied by the surfac˚2 tant just below the cmc showed that 1 occupied an area of 43.6 A ˚ 2 . This result was surprising, since based and 2 an area of 54.6 A upon volume calculations of the head groups, it was anticipated that surfactant 1 would exhibit the larger area per molecule. This discrepancy may reflect the different conformations adopted by the head groups; indeed, initial investigations into their conformations by NMR suggest this to be the case.2 However, it should be noted that the error associated with the calculation of the area per molecule using the Wilhelmy plate method is high and the differences observed should not be relied upon too heavily. The sources of error in the area per molecule (determined by this 2

Thanks to Dr. P. Sandor and Dr. E. Curzon for running NOE experiments.

FIG. 2. Variation in scattering ratio, S90 , with concentration for the surfactants 1 (d) and 2 (j) in water at 298 K. The long dashes give the predicted monomer line for surfactant 1, and the dotted line gives the predicted monomer line for surfactant 2.

126

DIEGO-CASTRO, HAILES, AND LAWRENCE

state is positioned close to the chiral head group for asymmetric induction to occur, but with 2 it is the endo transition state which is preferentially positioned in the head group area (3). In addition, the endo transition state is more sterically compact than that for the exo, which may result in a different locus for the transition state depending on whether 1 or 2 is used. The degree of penetration of water into the head group region of the micelle may also be important since when the cycloaddition reaction between nonyl acrylate and cyclopentadiene is performed in aqueous media the endo product predominates while, by comparison, the proportion of exo adduct increases when the reaction is carried out in lipophilic media (5). Overall, these studies demonstrate that the nature of the head group can influence the enantioselectivity observed, and they suggest that ultimately it may be possible to tailor the surfactant type to the required product outcome. ACKNOWLEDGMENTS SCHEME 2.

Degree of enantioselectivity observed when using 1 and 2.

than anticipated from a knowledge of the aggregation behavior of similar surfactant structures; for example, the aggregation number for hexadedcyltrimethylammonium bromide is 90 (20), suggesting that the presence of the hydroxyl group has a significant influence on aggregation behavior and/or the homochiral nature of the head group affects packing in this region of the micelle. DISCUSSION

The combination of surface tension and light scattering experiments demonstrates that these novel surfactants do aggregate into micellar structures although the aggregation numbers of the surfactants were surprisingly low, especially considering the low values of cmc obtained for these surfactants. The use of the two surfactants in Diels–Alder cycloaddition reactions with cyclopentadiene and nonyl acrylate revealed that when using surfactant 1, enantioselectivities were observed in the exo adduct (up to 13% e.e. and up to 18% e.e. when the bromide counterion was present) but that when using the leucinederived surfactant, 2, enantioselectivities (up to 15% e.e.s) were observed in the endo adduct (3) only (Scheme 2).3 In addition, preliminary NMR experiments have indicated that in 1 the phenyl ring is directed into the micellar core, while the isopropyl group is directed outward from the micellar core. This observation, in conjunction with the surface tension measurements of head group area (1 is smaller than 2), suggests that the conformation of the surfactant head group influences the relative orientation or positioning of substrates within the micelle during the reaction. In other words, when using 1, the exo transition

3 Enantioselectivities were monitored using a Chiralcel OD column with 0.1% isopropanol/hexane.

We are grateful to University College London (Access Funds for MJD-C), Bush Boake Allen, The Royal Society, The Nuffield Foundation, and the Central Research Fund, University of London, for supporting this work.

REFERENCES 1. Tascioglu, T., Tetrahedron 52, 11113 (1996). 2. Zhang, Y. M., Fan, W., Lu, P., and Wang, W., Synth. Commun. 18, 1495, (1988); Zhang, Y. M., and Li, W., Synth. Commun. 18, 1685 (1988); Zhang, Y. M., Qu, S., and Bao, W., Synth. Commun. 24, 2437 (1994); Zhang, Y. M., and Wu, W. D., Tetrahedron Asymm. 8, 3573 (1997); Zhang, Y. M., and Wu, W. D., Tetrahedron Asymm. 8, 2723 (1997); Andriamanampisoa, R., Boyer, B., Lamaty, G., and Roque, J. P., Tetrahedron 43, 77 (1987). 3. Diego-Castro, M. J., and Hailes, H. C., J. Chem. Soc. Chem. Commun., 1549 (1998). 4. Lee, C., O’Rear, E. A., Harwell, J. H., and Sheffield, J. A., J. Colloid Interface Sci. 137, 296 (1991). 5. Diego-Castro, M. J., and Hailes, H. C., Tetrahedron Lett. 39, 2211 (1998). 6. Meyer, A. I., and McKennon, M. J., J. Org. Chem. 58, 3568 (1993). 7. Toda, F., and Tanaka, F., J. Chem. Soc. Chem. Commun., 1087 (1997). 8. Stille, W. C., Kahn, M., and Mitra, A., J. Org. Chem. 43, 2923 (1978). 9. Corey, E. J., McCaully, R. J., and Sachdev, H. S., J. Am. Chem. Soc. 92, 2476 (1970). 10. Gulari, E., Chu, B., and Lui, T. Y., Biopolymers 18, 2943 (1979). 11. Ralston, A. W., Eggenberger, D. N., and Harwood, H. J., J. Am. Chem. Soc. 69, 2095 (1947). 12. Czernaiwski, M., Roczn. Chem. 40, 1935 (1966). 13. Aveyard, R., Cooper, P., and Fletcher, P. D. I., J. Chem. Soc. Faraday Trans. 86, 3623 (1990). 14. Lu, J. R., Thomas, R. K., Binks, B. P., Fletcher, P. D., and Penfold, J., J. Phys. Chem. 99, 4113 (1995). 15. Adamson, A. W., “Physical Chemistry of Surfaces,” 4th ed. Wiley, New York, 1986. 16. Chattoraj, D. K., and Birdi, K. S., “Adsorption and the Gibbs Surface Excess.” Plenum, New York, 1986. 17. Okada, H., Ozeki, S., and Ikeda, S., Bull. Chem. Soc. Jpn. 57, 1321 (1984). 18. Simister, E. A., Thomas, R. K., Penfold, J., Aveyard, R., Binks, B. P., Cooper, P., Fletcher, P. D., Lu, J. R., and Sokolowski, A., J. Phys. Chem. 96, 1383 (1992). 19. Anacker, E. W., and Westwall, A. E., J. Phys. Chem. 68, 3490 (1964). 20. Perche, T., Auvray, X., Petipas, C., Anthore, R., Rico, I., Lattes, A., and Bellissent, M. C., J. Phys I 2, 923 (1992).