Odd−Even Effects in Azobenzene−Urea Amphiphile Assemblies

Langmuir 2003, 19, 9297-9304

9297

Odd-Even Effects in Azobenzene-Urea Amphiphile Assemblies Toru Kobayashi† and Takahiro Seki*,‡ Chemical Resources Laboratory, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama 226-8503, Japan, and Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Chikusa, Nagoya 464-8603, Japan Received June 16, 2003. In Final Form: August 24, 2003 Azobenzene (Az) derivatives with a urea headgroup, n-{4-[(4-hexylphenyl)azo]phenoxy}alkylurea (6Aznurea, n ) 3, 4, 5, 6, and 7), were synthesized. The characteristics of mono- and multilayers of a series of 6Azn-urea compounds were investigated by means of surface pressure-area isotherm and compressibility measurements, UV-visible and infrared absorption spectroscopy, and X-ray diffraction measurements. Various types of odd-even effects were observed in the spreading behavior and rheological properties in the Langmuir monolayer on the water surface, in the transfer behavior from the water surface to a solid substrate, in the packing state in terms of molecular tilt and Az aggregation in the deposited mono- and multilayers, and in the nature of hydrogen bonding in the multilayers. The carbon parity nature in the 6Azn-urea monolayer should result from the firm fixation at the terminal urea headgroup via intermolecular bifurcated hydrogen bonding, which resembles the situation of self-assembled monolayers attached covalently onto a solid substrate.

1. Introduction Urea, synthesized for the first time artificially with organic compounds, is one of the indispensable compounds in the history of organic chemistry. Hydrogen bonds play fundamental roles in biological and chemical systems. Also, urea is widely recognized as an important building block in supramolecular architectures,1 and the assemblies supply inclusion cavities.2 As urea derivatives form a hydrogen bonding network between a carbonyl group of one urea unit and two hydrogen atoms of a neighboring one (bifurcated hydrogen bonding), this dual hydrogen bond strongly fixes the molecular packing (Chart 1). Langmuir monolayers of long-chain urea derivatives also have attracted considerable attention due to their characteristic structure-forming abilities derived from this hydrogen bonding. Alkylureas show a distinctive area contraction at a phase transition temperature in the heating process, which can be ascribed to the disruption of the bifurcated hydrogen bonding.3-5 Furthermore, an unusual type of Langmuir monolayer is formed from a urea compound. Huo et al.6 reported on the formation of a Langmuir monolayer of a disubstituted urea lipid molecule in which the polar urea part is not in contact with the water surface but is positioned at the middle of the molecular array. Our recent efforts have been directed to the functionalization of urea amphiphiles with a photochromic azoben* To whom correspondence should be addressed. E-mail: [email protected]. † Tokyo Institute of Technology. ‡ Nagoya University. (1) MacDonald, J. C.; Whitesides, G. M. Chem. Rev. 1994, 94, 2383. (2) Takemoto, K.; Sonoda, N. Inclusion Compounds; Atwood, J. L., Davies, J. E. D., MacNicol, D. D., Eds.; Academic Press: London, 1984; Vol. 2, p 47. (3) Glazer, J.; Alexander, A. E. Trans. Faraday Soc. 1951, 47, 401. (4) Kato, T.; Akiyama, H.; Yoshida, M. Chem. Lett. 1992, 565. (5) Urai, Y.; Ohe, C.; Itoh, K.; Yoshida, M.; Iimura, K.; Kato, T. Langmuir 2000, 16, 3920. (6) Huo, Q.; Russev, S.; Hasegawa, T.; Nishijo, J.; Umemura, J.; Puccetti, G.; Russell, K. C.; Leblanc, R. M. J. Am. Chem. Soc. 2000, 122, 7890.

Chart 1

zene (Az) unit (6Az10-urea, Chart 1).7-9 This approach revealed many characteristic features originating from the nature of urea, for example, a complete hindrance of the trans-to-cis photoisomerization in the Langmuir monolayer7 and reversible alternations of the packing state of a monolayer on a substrate by atmospheric humidity changes, leading to a reversible on/off switching of the photoreactivity.8 The important role of the bifurcated hydrogen bond is confirmed from the fact that the above features do not appear when two hydrogen atoms at the terminal nitrogen are substituted by two methyl groups.9 This paper reports on the marked odd-even effects observed in the Az-urea layer assemblies, where the carbon number of the methylene spacer between the urea head and the Az moiety is varied (6Azn-urea, n being 3, 4, 5, 6, and 7 in Chart 1).10 The parity effects are observed for the packing state in a Langmuir monolayer floating on water, the deposition behavior in the Langmuir-Blodgett (LB) process, and further the structure of the mono- and (7) Seki, T.; Fukuchi, T.; Ichimura, K. Bull. Chem. Soc. Jpn. 1998, 71, 2807. (8) Seki, T.; Fukuchi, T.; Ichimura, K. Langmuir 2000, 16, 3564. (9) Seki, T.; Fukuchi, T.; Ichimura, K. Langmuir 2002, 18, 5462. (10) A preliminary communication paper has been already published: Kobayashi, T.; Seki, T.; Ichimura, K. Chem. Commun. 2000, 1193.

10.1021/la035056w CCC: $25.00 © 2003 American Chemical Society Published on Web 10/03/2003

9298

Langmuir, Vol. 19, No. 22, 2003 Scheme 1

multilayers on substrates. Odd-even effects affecting the thermal properties have long been investigated in liquid crystals11,12 and polymers.13 In recent studies, the carbon parity effects are frequently discussed for supramolecular assemblies and self-assembled monolayers (SAMs) on a substrate. For example, Shimizu et al.14 demonstrated a significant odd-even dependence for the assemblies of bolaamphiphiles showing alternations between supramolecular fibers and planar crystals. In two-dimensional monolayer and bilayer systems, various surface-mediated phenomena such as wetting of a liquid,15,16 ice nucleation,17 liquid crystal alignment,18 electrochemical properties,19 and polymerization20 are altered by the carbon parity. We emphasize here the unique carbon parity effects which are most likely attributable to the bifurcated hydrogen bond formation in the assemblies of urea derivatives. 2. Experimental Section 2.1. Materials. Starting compounds for syntheses (Scheme 1) were purchased from Tokyo Kasei Kogyo (TCI Inc.) and used without further purification. The structures of the final products were confirmed by 1H NMR, Fourier transform infrared (FT-IR) spectroscopy, and elemental analysis. The melting points were measured with a Yanaco MP-S3 melting-point apparatus and are uncorrected. 1H NMR spectra were recorded on a Bruker AC-200 spectrometer using Me4Si as the internal standard for the chloroform solutions. IR spectra were taken on a JASCO FT/IR-300. Elemental analysis was performed on a Yanaco MT-5 CHN CORDER. 2.1.1. Synthesis of N-(n-{4-[(4-Hexylphenyl)azo]phenoxy})alkylurea (6Azn-urea, n ) 3-7). The synthetic procedure for the compound having the 10 methylene spacer (6Az10(11) For example: Demus, D. In Handbook of Liquid Crystals; Demus, D., Goodgy, J., Gray, G. W., Spiess, H.-W., Vill, V., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, pp 133-187 and references therein. (12) Seo, D.-S.; Kobayashi, S.; Nishikawa, M.; Kim, J.-H.; Yabe, Y. Appl. Phys. Lett. 1995, 66, 1334. (13) Francisca, M. L. J.; Kannan, P. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 1755. (14) Shimizu, T.; Masuda, M. J. Am. Chem. Soc. 1997, 119, 2812. (15) Colorado, R., Jr.; Villazana, R. J.; Lee, T. R. Langmuir 1998, 14, 6337. (16) Graupe, M.; Takenaga, M.; Koini, T.; Colorado, R., Jr.; Lee, T. R. J. Am. Chem. Soc. 1999, 121, 3222. (17) Gavush, M.; Popovitz-Biro, R.; Laahav, M.; Leiserowitz, L. Science 1990, 250, 972. (18) Gupta, V. K.; Abbot, N. L. Phys. Rev. E 1996, 54, R4541. (19) Shimomura, M.; Utsugi, K.; Horikoshi, J.; Okuyama, K.; Hotozaki, O.; Oyama, N. Langmuir 1991, 7, 760. (20) Menzel, H.; Hoestmenn, S.; Mowery, M. D.; Cai, M.; Evans, C. E. Polymer 2000, 41, 8113.

Kobayashi and Seki urea) was already described in ref 7. The compounds employed in this work were also synthesized in the same manner. 2.1.2. 4-[(4-Hexylphenyl)azo]phenol (6AzOH). 4-Hexylaniline (25.4 g, 143 mmol) was dissolved in a mixture of concentrated hydrochloric acid (25 mL) and water (150 mL). Sodium nitrate (11.9 g, 179 mmol) in water (15 mL) was added dropwise to the above solution at 5 °C. An aqueous solution (800 mL) dissolving phenol (16.5 g, 175 mmol), sodium hydroxide (8.80 g, 220 mmol), and urea was then added dropwise under vigorous stirring. The solution was stirred for 2 h at 5 °C and 3 h at room temperature. After this solution was neutralized with a diluted hydrochloric acid aqueous solution, the precipitate was filtered off and dissolved in ethyl acetate. The solution was washed with a NaCl aqueous solution and then dried over anhydrous magnesium sulfate. The precipitate was recrystallized from hexane twice to give yellow platelet crystals. Yield, 24.5 g (61%); mp, 71-72 °C. 1H NMR (δ [ppm], CDCl3): 0.87 (3H, t, J ) 7 Hz, CH3-), 1.32-1.65 (8H, m, -CH2-), 2.67 (2H, t, J ) 8 Hz, -CH2Ph), 5.27 (1H, s, -OH), 6.94 (2H, d, J ) 9 Hz, Ph-H), 7.30 (2H, d, J ) 9 Hz, Ph-H), 7.77-7.88 (4H, m, Ph-H). 2.1.3. 2-Tetrahydropyranyl-4-bromobutanate (1). A mixture of 4-bromotetranoic acid (6.00 g, 35.2 mmol) and 3,4-dihydro2H-pyrane (3.04 g, 35.9 mmol) in dichloromethane (40 mL) was prepared, and pyridinium p-toluenesulfonate (0.91, 3.59 mmol) was added at 0 °C. The reaction mixture was stirred for 30 min at 0 °C and at room temperature for an additional 4 h. After washing with 4% sodium bicarbonate aqueous solution, water, and brine, the solution was dried over anhydrous magnesium sulfate. The solvent was evaporated, and the product was obtained as an oily material. Yield (crude), 5.12 g (57%). 2.1.4. 2-Tetrahydropyranyl-5-bromopentanate (2). This compound was synthesized as described for 1, starting from 5-bromopentanoic acid (10.0 g, 55.2 mmol). The product was obtained as an oily material. Yield (crude), 9.75 g (67%). 2.1.5. 2-Tetrahydropyranyl-6-bromohexanate (3). This compound was synthesized as described for 1, starting from 6-bromohexanoic acid (13.0 g, 66.7 mmol). The product was obtained as an oily material. Yield (crude), 15.0 g (80%). 2.1.6. 2-Tetrahydropyranyl-8-bromooctanate (4). This compound was synthesized as described for 1, starting from 5-bromooctanoic acid (5.00 g, 22.4 mmol). The product was obtained as an oily material. Yield (crude), 5.05 g (73%). 2.1.7. 4-{4-[(4-Hexylphenyl)azo]phenoxy}butanoic Acid (6Az3-COOH). To a mixed solution of 6AzOH (3.00 g, 10.6 mmol) and 1 (2.67 g, 10.6 mmol) in dry N,N-dimethylformamide (80 mL), potassium carbonate (2.95 g, 21.3 mmol) and a catalytic amount of potassium iodide were added and stirred at 70 °C for 45 h. The reaction mixture was extracted with ether and washed with water. After evaporation of the organic solvent, tetrahydrofuran (30 mL) and concentrated hydrochloric acid (5.0 mL) were added. The solution was stirred for 1 h at room temperature. After extraction with chloroform, the organic layer was washed with water and a NaCl aqueous solution and dried over anhydrous magnesium sulfate. The solvent was evaporated, and the residue was recrystallized from hexane. Yield, 1.63 g (42%); mp, 150152 °C. 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.33-1.65 (6H, m, -CH2-), 1.68 (2H, m, -CH2-CH2-Ph), 2.17 (2H, quint, J ) 7 Hz, -CH2-CH2-COOH), 2.65 (4H, m, -CH2Ph, -CH2-COOH), 4.11 (2H, t, J ) 6 Hz, -O-CH2-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.30 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 71.66; H, 7.43; N, 7.75%. Calcd for C22H28N2O3: C, 71.71; H, 7.66; N, 7.60%. 2.1.8. 5-{4-[(4-Hexylphenyl)azo]phenoxy}pentanoic Acid (6Az4-COOH). This compound was synthesized as described for 6Az3-COOH, starting from 6AzOH (3.01 g, 10.7 mmol) and 2 (3.10 g, 11.7 mmol). Yield, 3.01 g (74%); mp, 110-111 °C. 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.32-1.90 (12H, m, -CH2-), 2.48 (2H, t, J ) 8 Hz, -CH2-COOH), 2.67 (2H, t, J ) 7 Hz, -CH2-Ph), 4.06 (2H, t, J ) 6 Hz, -O-CH2-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.30 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 72.16; H, 7.69; N, 7.21%. Calcd for C23H30N2O3: C, 72.22; H, 7.91; N, 7.32%. 2.1.9. 6-{4-[(4-Hexylphenyl)azo]phenoxy}hexanoic Acid (6Az5-COOH). This compound was synthesized as described for 6Az3-COOH, starting from 6AzOH (5.00 g, 17.7 mmol) and 3 (5.94 g, 21.3 mmol). Yield, 6.25 g (89%); mp, 137-139 °C. 1H

Odd-Even Effects in Amphiphile Assemblies NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.32-1.92 (14H, m, -CH2-), 2.42 (2H, t, J ) 8 Hz, -CH2-COOH), 2.67 (2H, t, J ) 7 Hz, -CH2-Ph), 4.05 (2H, t, J ) 6 Hz, -O-CH2-), 6.98 (2H, d, J ) 9 Hz, Ph-H), 7.29 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.90 (4H, m, Ph-H). Found: C, 72.53; H, 8.27; N, 7.03%. Calcd for C24H32N2O3: C, 72.73; H, 8.08; N, 7.07%. 2.1.10. 7-{4-[(4-Hexylphenyl)azo]phenoxy}heptanoic Acid (6Az6-COOH). This compound was synthesized as described for 6Az3-COOH, starting from 6AzOH (4.00 g, 14.2 mmol) and 7-bromoheptanoic acid ethyl ester (4.03 g, 17.0 mmol). Yield, 4.87 g (84%); mp, 113-114 °C. 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.26-1.87 (16H, m, -CH2-), 2.39 (2H, t, J ) 7 Hz, -CH2-COOH), 2.67 (2H, t, J ) 7 Hz, -CH2-Ph), 4.04 (2H, t, J ) 6 Hz, -O-CH2-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.29 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 73.10; H, 8.16; N, 6.78%. Calcd for C25H34N2O3: C, 73.14; H, 8.35; N, 6.82%. 2.1.11. 8-{4-[(4-Hexylphenyl)azo]phenoxy}octanoic Acid (6Az7-COOH). This compound was synthesized as described for 6Az3-COOH, starting from 6AzOH (3.00 g, 10.6 mmol) and 4 (4.57 g, 14.9 mmol). Yield, 3.25 g (72%); mp, 104-105 °C. 1H NMR (δ [ppm], CDCl3): 0.88 (3H, t, J ) 7 Hz, CH3-), 1.32-1.87 (18H, m, -CH2-), 2.37 (2H, t, J ) 7 Hz, -CH2-COOH), 2.67 (2H, t, J ) 8 Hz, -CH2-Ph), 4.03 (2H, t, J ) 6 Hz, -O-CH2-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.29 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 73.23; H, 8.30; N, 6.41%. Calcd for C26H36N2O3: C, 73.55; H, 8.55; N, 6.60%. 2.1.12. 3-{4-[(4-Hexylphenyl)azo]phenoxy}propylurea (6Az3-urea). To a solution of 6Az3-COOH (1.00 g, 2.71 mmol) in dry benzene (40 mL) were added diphenylphosphoryl azide (DPPA) (850 mg, 3.05 mmol) and triethylamine (310 mg, 3.06 mmol), and the mixture was stirred at refluxing temperature for 5 h. After the carboxylic acid was converted to isocyanate in the above procedure,21 a dry ammonia gas was passed into the solution for 1 h at room temperature.22 After a pale yellow solid was precipitated, the mixture was vigorously stirred overnight. The solid was filtered off and recrystallized from chloroform. Yield, 660 mg (63%); mp, 163-165 °C. IR (KBr, cm-1): 3413 (νNH, -NH2), 3326 (νNH, -NH-), 1638 (νCdO), 1587 (δNH, -NH-). 1H NMR (δ [ppm], CDCl ): 0.89 (3H, t, J ) 7 Hz, CH -), 1.323 3 1.65 (8H, m, -CH2-), 2.05 (2H, quint, J ) 6 Hz, -CH2-CH2NH-), 2.68 (2H, t, J ) 7 Hz, -CH2-Ph), 3.42 (2H, q, J ) 7 Hz, -CH2-NH-), 4.13 (2H, t, J ) 6 Hz, -O-CH2-), 4.29 (2H, br, J ) 7 Hz, -NH2), 4.66 (1H, br, -NH-), 7.00 (2H, d, J ) 9 Hz, Ph-H), 7.30 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 68.70; H, 7.80; N, 14.57%. Calcd for C22H30N4O2: C, 69.08; H, 7.91; N, 14.65%. 2.1.13. 4-{4-[(4-Hexylphenyl)azo]phenoxy}butylurea (6Az4-urea). This compound was synthesized as described for 6Az3-urea, starting from 6Az4-COOH (1.01 g, 2.64 mmol). Yield, 860 mg (81%); mp, 154-156 °C. IR (KBr, cm-1): 3452 (νNH, -NH2), 3323 (νNH, -NH-), 1630 (νCdO), 1566 (δNH, -NH-). 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.32-1.92 (12H, m, -CH2-), 2.67 (2H, t, J ) 8 Hz, -CH2-Ph), 3.28 (2H, q, J ) 6 Hz, -CH2-NH-), 4.07 (2H, t, J ) 6 Hz, -O-CH2-), 4.30 (2H, br, -NH2), 4.54 (1H, br, -NH-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.30 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 69.80; H, 7.85; N, 14.30%. Calcd for C23H32N4O2: C, 69.67; H, 8.13; N, 14.13%. 2.1.14. 5-{4-[(4-Hexylphenyl)azo]phenoxy}pentylurea (6Az5-urea). This compound was synthesized as described for 6Az3-urea, starting from 6Az5-COOH (3.00 g, 7.56 mmol). Yield, 2.64 g (85%); mp, 148-149 °C. IR (KBr, cm-1): 3431 (νNH, -NH2), 3336 (νNH, -NH-), 1640 (νCdO), 1560 (δNH, -NH-). 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.33-1.88 (14H, m, -CH2-), 2.67 (2H, t, J ) 8 Hz, -CH2-Ph), 3.18 (2H, t, J ) 7 Hz, -CH2-NH-), 4.02 (2H, t, J ) 6 Hz, -O-CH2-), 4.35 (2H, br, -NH2), 4.65 (1H, br, -NH-), 6.96 (2H, d, J ) 9 Hz, Ph-H), 7.26 (2H, d, J ) 8 Hz, Ph-H), 7.75-7.88 (4H, m, Ph-H). Found: C, 70.09; H, 8.19; N, 13.80%. Calcd for C24H34N4O2: C, 70.21; H, 8.35; N, 13.65%. 2.1.15. 6-{4-[(4-Hexylphenyl)azo]phenoxy}hexylurea (6Az6-urea). This compound was synthesized as described for (21) Shioiri, T.; Ninomiya, K.; Yamada, S. J. Am. Chem. Soc. 1972, 94, 6203. (22) Jones, L. W.; Mason, J. P. J. Am. Chem. Soc. 1927, 49, 2528.

Langmuir, Vol. 19, No. 22, 2003 9299 6Az3-urea, starting from 6Az6-COOH (1.02 g, 2.48 mmol). Yield, 720 mg (69%); mp, 154-155 °C. IR (KBr, cm-1): 3434 (νNH, -NH2), 3333 (νNH, -NH-), 1633 (νCdO), 1560 (δNH, -NH-). 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.32-1.86 (16H, m, -CH2-), 2.67 (2H, t, J ) 8 Hz, -CH2-Ph), 3.19 (2H, m, -CH2NH-), 4.04 (2H, t, J ) 6 Hz, -O-CH2-), 4.24 (2H, br, -NH2), 4.46 (1H, br, -NH-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.29 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 70.64; H, 8.69; N, 13.22%. Calcd for C25H36N4O2: C, 70.72; H, 8.55; N, 13.20%. 2.1.16. 7-{4-[(4-Hexylphenyl)azo]phenoxy}heptylurea (6Az7-urea). This compound was synthesized as described for 6Az3-urea, starting from 6Az7-COOH (1.00 g, 2.36 mmol). Yield, 630 mg (61%); mp, 141-143 °C. IR (KBr, cm-1): 3431 (νNH, -NH2), 3339 (νNH, -NH-), 1639 (νCdO), 1559 (δNH, -NH-). 1H NMR (δ [ppm], CDCl3): 0.89 (3H, t, J ) 7 Hz, CH3-), 1.32-1.85 (18H, m, -CH2-), 2.67 (2H, t, J ) 8 Hz, -CH2-Ph), 3.17 (2H, q, J ) 6 Hz, -CH2-NH-), 4.03 (2H, t, J ) 6 Hz, -O-CH2-), 4.27 (2H, br, -NH2), 4.43 (1H, br, -NH-), 6.99 (2H, d, J ) 9 Hz, Ph-H), 7.29 (2H, d, J ) 8 Hz, Ph-H), 7.77-7.91 (4H, m, Ph-H). Found: C, 71.05; H, 8.67; N, 12.81%. Calcd for C26H38N4O2: C, 71.20; H, 8.73; N, 12.77%. 2.2. Methods. 2.2.1. Surface Pressure-Area Isotherms. The spreading behavior of the Az-containing amphiphiles was evaluated on pure water (Milli-Q grade, 18 MΩ cm-1, pH ) 5.8) using a Lauda FW1 film balance in subdued red light. The temperature of the subphase was maintained at 20 ( 0.5 °C by water circulation. The Az amphiphile (urea or carboxylic acid derivative) was spread from a chloroform solution (1.0 × 10-3 mol dm-3). After evaporation of chloroform, the monolayer was compressed at a speed of 30 cm2 min-1. The surface pressure was recorded versus molecular area. UV-visible absorption spectra for the Langmuir monolayer were taken using an Ohtsuka Electronics MCPD-2000 system at room temperature. In compressibility, preparations of 6Azn-urea Langmuir monolayers were previously described. Each Langmuir monolayer was compressed at a surface pressure of 20 mN m-1 at 20 ( 0.5 °C. The variation of molecular area was recorded versus time. 2.2.2. Langmuir-Blodgett Deposition. Quartz plates were washed with a saturated sodium hydroxide ethanol solution, ethanol, flowing water, and finally pure water under ultrasonic wave treatment. The static contact angle of water on this surface measured with a FACE CA-X was 8 ( 2° (hydrophilic plates). Hydrophobized quartz plates were prepared by treatment with a vapor of 1,1,1,3,3,3-hexamethyldisilazane. The contact angle of water on the hydrophobic surface was 81 ( 1°. The LB films were prepared by the Langmuir-Blodgett (vertical dipping) or Langmuir-Schaefer (horizontal lifting) method. In the vertical dipping procedure, the monolayer was transferred on a hydrophilic quartz plate at 20 mN m-1 and a dipping speed of 5 mm min-1 at 20 ( 0.5 °C. For the horizontal lifting procedure, a homemade frame of Teflon sheets (1 mm thickness) was prepared for compartmentalization. The deposition was performed onto the hydrophobic quartz plates prepared in the above manner. 2.2.3. X-ray Diffraction Measurements. Spectra of all LB films of 30 layers were taken on a PW3050 X’Pert system (Philips) using Cu KR radiation (40 kV, 30 mA). The diffraction intensity profiles were measured between 1° and 40° in the 2θ/θ-scan mode with steps at room temperature. 2.2.4. UV-Visible Absorption Measurements. UV-visible absorption spectra for the LB films were taken with a JASCO MAC-1 spectrophotometer for monolayers or a HP8452A for multilayers at room temperature. 2.2.5. Infrared Absorption Measurements. FT-IR spectra were recorded on a Biorad FTS6000 spectrometer equipped with a deuterated triglycine sulfate (DTGS) detector. Transmission spectra of KBr disk pellets and LB films were measured. The LB films of 30-layered 6Azn-urea were prepared by the horizontal lifting method on a CaF2 plate. The spectra were obtained by integration of 64-1000 scans at a resolution of 2 cm-1.

3. Results and Discussion 3.1. Langmuir Monolayers. 3.1.1. Spreading Behavior and Molecular Orientation. Figure 1 represents the surface pressure (π)-area (A) isotherms of Az-urea

9300

Langmuir, Vol. 19, No. 22, 2003

Figure 1. Surface pressure-area isotherms of the 6Azn-urea monolayer on pure water at 20 °C. In the inset, the limiting occupying area is plotted against the carbon number of the alkylene spacer.

monolayers upon a subphase at a temperature of 20 °C. As indicated, the molecular occupying area per molecule, which was estimated by extrapolating the steepest slope to zero pressure, indicated systematic alternation with the carbon number of the spacer. For the compounds with odd-numbered carbon spacers (n ) 3, 5, and 7), the molecular occupying areas were 0.29, 0.30, and 0.30 nm2 for 6Az3-urea, 6Az5-urea, and 6Az7-urea, respectively. In contrast, the monolayers with even-numbered spacers (n ) 4, 6, and 10) indicated significant area expansions, giving the molecular occupying area of 0.34, 0.35, and 0.32 nm2 for 6Az4-urea, 6Az6-urea, and 6Az10-urea, respectively. The inset displays the limiting molecular area versus carbon number of the spacer. The odd-even effect is obvious. Since the cross section of vertically aligned Az is 0.25 nm2 as known from the X-ray diffraction data of crystals,23 it can be thus interpreted that the molecules in these monolayers are tilted with respect to the surface normal. The degree of molecular tilt for the even-numbered carbon spacers should be greater than that of the oddnumbered ones. The shape of the π-A curves also showed alternate changes with the carbon number of the spacer. The monolayers with the odd-numbered carbon series (n ) 3, 5, and 7) showed a sharper pressure increase with surface pressure in the curve, indicative of the lower compressibility. In contrast, the even-numbered carbon series (n ) 4 and 6) compounds gave a continuous smoother slope in the π-A isotherms showing the higher compressibility. Furthermore, a characteristic inflecting region around 33 mN m-1 was observed only for the odd-numbered series. This seems to be a phase transition point; however, we have not yet clarified the process involved in this phase transition. As the control experiments, π-A curves of the monolayers were taken for the corresponding precursor carboxylic acid derivatives (6Azn-COOH, Chart 1). π-A isotherms of this series gave completely different results.24 In these cases, odd-even regular changes in the molecular occupying area were not admitted, indicating that the odd-even alternation is the characteristic spreading behavior for the monolayers of Az derivatives having the urea headgroup. 3.1.2. Rheological Aspect. The compressibility of Langmuir monolayers is directly related to the molecular packing state in two dimensions and should be strongly (23) Brown, C. J. Acta Crystallogr. 1966, 21, 146. (24) Kobayashi, T.; Seki, T.; Ichimura, K. Trans. Mater. Res. Soc. Jpn. 2000, 26, 487.

Kobayashi and Seki

Figure 2. Time courses of the area decrease of the 6Azn-urea monolayer on pure water upon storage at a constant pressure of 20 mN m-1. In the inset, the area decrease after 60 min [(area after 60 min)/(initial area)] is plotted against the carbon number of the alkylene spacer.

coupled with the deposition behavior onto a solid surface as will be described in section 3.2.1. The area changes of 6Azn-urea monolayers with time at a constant pressure (20 mN m-1) are shown in Figure 2. The decrease in area with time was obvious for the even-numbered spacers (n ) 4 and 6). For odd-numbered spacers (n ) 3, 5, and 7), the area reduction is highly suppressed. Thus, the rheological properties are altered by the carbon number. The inset shows the ratio of area decrease at 60 min to that at the initial stage as a function of carbon number of the spacer. The monolayers with the odd-numbered spacers showed only a small area decrease, showing a close-packing state which was already attained at the initial stage of film spreading. These results were consistent with the alternate behavior in the lift-off area in the π-A curve measurements (Figure 1). 3.2. Monolayers on Solid Substrates. 3.2.1. Transfer Process. Figure 3 displays the transfer ratios of 6Aznurea monolayers on the hydrophilic (a) and hydrophobic (b) quartz plates when the vertical dipping transfer was attempted in the upstroke direction. The 6Azn-urea monolayers could be successfully transferred onto the hydrophilic quartz plate at 20 mN m-1 with transfer ratios at 1.0 ( 0.1 (a) in all cases. In contrast, the transfer behavior of the 6Azn-urea monolayers on the hydrophobic quartz plate was definitely dependent on the carbon number of the spacer (b). The monolayers with oddnumbered spacers gave the ideal transfer ratios of nearly unity, and in sharp contrast, those with even-numbered spacers were not transferred to the substrate at all. The deposition behavior is, in general, very sensitive to many factors such as the rheological properties of the monolayer, the packing state of the molecules, the hydrophobic/ hydrophilic balance, the interaction modes between the substrate and the molecule, and so forth. On the hydrophilic surface, the monolayer should be strongly adhered to the substrate through polar interactions, most dominantly via hydrogen bonding, which can lead to successful deposition for all monolayers. In the case of the hydrophobic surface, the contribution of the polar interactions between the monolayer and the substrate becomes minor, and thus the rheological factor (rigidity and fluidity) of the monolayer should influence the deposition behavior. The reason for the sharp carbon parity effect in the transfer behavior for the hydrophobic surface is not clearly elucidated yet; however, this is, to our knowledge, the first example that the transfer behavior is definitely switched by the carbon number of the molecule.

Odd-Even Effects in Amphiphile Assemblies

Figure 3. Transfer ratio of the 6Azn-urea monolayer on a hydrophilic (panel a, clean quartz) and hydrophobic (panel b, hydrophobilized with hexemethydisilazane) quartz plate. The error bars are based on 4 measurements.

Figure 4. UV-visible absorption spectra of the 6Azn-urea monolayer transferred onto both sides of a quartz plate at 20 mN m-1. The inset shows λmax (filled squares) and A245/Along (open circles) plotted against the carbon number of the alkylene spacer.

3.2.2. UV-Visible Absorption Spectroscopy. Figure 4 shows absorption spectra of the 6Azn-urea monolayers on a hydrophilic quartz plate. Two intense bands were observed in all spectra. From these spectral features, the odd-even effect of the spacer length affecting the packing state of Az can be pointed out as follows. The peak position of the π-π* band attributed to the long axis transition of Az (λmax), which indicates the aggregation state of the chromophore,25 is an effective tool to discuss the molecular packing state. In the Langmuir monolayers, UV-visible absorption spectra of 6Azn-urea on a water surface did not show significant changes for odd and even series, (25) Shimomura, M.; Ando, R.; Kunitake, T. Ber. Bunsen-Ges. Phys. Chem. 1983, 87, 1134.

Langmuir, Vol. 19, No. 22, 2003 9301

indicating that the molecular packing state of these films on water was not significantly affected by the spacer length. In the transferred monolayers of 6Azn-urea, the Az unit formed H-aggregates in all cases judging from the considerable hypsochromic shifts of λmax (317-327 nm) from that in a chloroform solution (352 nm) and that in the Langmuir monolayers (355 nm) in a dry atmosphere.8 The spectra of the transferred monolayers showed zigzag alternations in λmax with an increase of the carbon spacer (closed squares in the inset of Figure 4). Generally, the longer the aliphatic chains, the stronger the van der Waals interactions among the molecules. Accordingly, the λmax is shifted to shorter wavelengths with increasing spacer length, as a consequence of the formation of stronger H-type aggregation. The alternation in the molecular orientation in these monolayers was also suggested from the spectra. The ratio of the absorption intensities of two peaks, A242/Along, where A242 and Along are the absorbances at 242 nm and λmax, respectively, is plotted against the carbon number of the spacer (open circles in the inset of Figure 4). The A242/Along ratios for the odd-numbered series exceeded 1.0, and those for the even-numbered series were below 1.0. This absorbance ratio can be a measure of molecular tilt. A smaller value of A242/Along indicated a larger molecular tilt from the surface normal in the transmission mode measurements. The results obtained here indicate that the monolayers of even carbon numbers are tilted to a greater extent and are consistent with the data for molecular occupying area in the spreading behavior shown in Figure 1. 3.3. Multilayers on Substrates. 3.3.1. Multilayer Deposition. For precise structural evaluation of the monolayer by X-ray analysis and IR spectroscopy, preparation of LB films in the multilayered state is desirable. First, the vertical dipping method onto the hydrophilic plate was attempted for the preparation of multilayers. In this procedure, the first layer was satisfactorily transferred as stated in section 3.2.1; however, the second layer was readily peeled off in the downstroke dipping process. The multilayer deposition was unsuccessful. We next applied the horizontal lifting procedure (Langmuir-Schaefer method). The 6Azn-urea monolayer was compressed to a surface pressure of 20 mN m-1, the film was compartmentalized by a Tefron frame, and then a hydrophobic quartz plate was attached to the monolayer and lifted. In this method, multilayer deposition could be performed at least up to 30 layers. UV-visible absorption spectra were taken at each step in order to confirm the transfer state, and representative data are depicted in Figure 5 for 6Az5-urea. In the multilayered films, the aggregation state of Az became somewhat multifarious. The π-π* absorption band gave multiple peaks around 320-350 nm. Also, the peaks at shorter wavelengths became more dominant with the increase in deposition number. As shown in the inset, the absorbances at the band peaks (340 and 242 nm) were proportional to the deposition number, indicative of successful deposition at good transfer ratios and proper multilayer formation. The absorbance of a 2-layered film prepared by the horizontal lifting method exactly agreed with that of two singlelayered films on both sides obtained by the vertical lifting procedure in which the transfer ratio can be correctly evaluated (see Figure 3a). This fact further justifies the successful deposition in the horizontal lifting method. 3.3.2. X-ray Diffraction Measurements. The characteristic layer spacing and ordering of the multilayers were examined by X-ray diffraction analysis. Small-angle

9302

Langmuir, Vol. 19, No. 22, 2003

Figure 5. UV-visible absorption spectra of multilayers of 6Az5-urea at various deposition numbers. The inset shows absorbances at the two band peaks as a function of deposition number. Essentially the same data were obtained for the other series (n ) 3, 4, 6, and 7).

Kobayashi and Seki

Figure 7. Layer spacing of the multilayers evaluated from the X-ray diffraction patterns (filled circles) and the molecular tilt angle in the multilayers (filled squares) of 6Azn-urea as a function of the carbon number of the alkylene spacer. The tilt angles were determined by comparisons between the layer spacing (X-ray) and molecular length (CPK model).

Figure 8. Transmission infrared spectra of multilayers of 6Aznurea (30 layers) on a CaF2 plate. The spectral region of 17001200 cm-1 is indicated.

Figure 6. X-ray diffraction pattern profiles of multilayers (30 layers) of 6Azn-urea on a quartz plate.

regions of the X-ray diagrams for 30-layered 6Azn-urea LB films gave a series of sharp reflection peaks up to fifth order, as shown in Figure 6. These results indicate that the highly ordered multilayered structure is obtained in all cases. The layer spacings in all cases were longer than the length of each molecule, suggestive of the formation of the Y-type (head to tail) built-up structure. The turnover process should have occurred during the deposition process or sample storage; otherwise the Y-type layer structure would not be obtained. The static contact angles of water on the 6Azn-urea samples possessing 1-8 layers all exhibited the common values above 90°. These facts indicate that the alkyl chains are positioned at the topmost surface and the bilayer structure is formed in the inner part. The tilt angle of the 6Azn-urea molecules in the layer was evaluated from the layer spacing and the molecular length estimated from the Corey-Pauling-Koltun (CPK) model. The layer spacings and estimated tilt angles at

each carbon number are summarized in Figure 7. The increase in the layer spacing was not proportional with the carbon number of the spacer, but the increment was altered. This situation is clearly reflected in the change in the molecular tilt angle (closed circles). Since the tilt angles from the normal were smaller for the odd-numbered series, the 6Azn-urea molecules having the odd-numbered carbon spacer adopt a more upright orientation with respect to the substrate plane. This tendency is again in agreement with the data for molecular occupying area in the Langmuir monolayers (section 3.1.1) and also with those of the absorption spectra (section 3.2.2). 3.3.3. Infrared Spectroscopy. FT-IR spectroscopy provides useful information on the state of hydrogen bonds. Figure 8 exhibits the IR spectra of 30-layered multilayers on a CaF2 substrate for the five films in the range of 17001200 cm-1. The CdO stretching vibration band and the N-H (-NH-) deformation band were observed at 16451635 and 1545-1570 cm-1, respectively. Figure 9 shows the absorption frequencies of these bands versus the spacer chain length (n). As indicated, clear odd-even alternations were obtained. Because the bond energy of CdO is weakened by hydrogen bond formation, the absorption frequencies of the stretching vibration of these functional groups (νCdO) should shift to lower wavenumbers when a stronger hydrogen bond is formed. On the other hand, the absorption frequency of the N-H deformation band (δN-H) increases with the hydrogen bond formation due to the larger energy required for the deformation of the N-H bond when it is strained hard by the hydrogen bonding.

Odd-Even Effects in Amphiphile Assemblies

Figure 9. Band frequencies of typical infrared absorption bands as a function of the carbon number of the alkylene spacer of 6Azn-urea multilayers deposited on a CaF2 plate. The two figures show the peak positions corresponding to the CdO stretching (a) and N-H deformation (b) modes.

The IR spectral data show systematic shifts to relatively lower frequencies for the νCdO band and higher frequencies for δN-H when the even-numbered derivatives are employed. Hence, the intermolecular hydrogen bonding becomes stronger for the 6Azn-urea derivatives having an even-numbered carbon spacer than for those having an odd-numbered one. Similar carbon parity on the amide absorption bands has been reported for the cast films of Az-containing amphiphiles.26 3.4. Features of the Odd-Even Effects in the 6Aznurea Assemblies. In this section, we summarize and revisit the features of the effects observed here. The various types of odd-even effects found in the present system are summarized in Table 1. The effects are observed in various assembly types including bulk microcrystals, monolayers on both water and a solid substrate, and multilayers. An unusual feature is the clear effect observed in the deposition process. The deposition onto a hydrophobic substrate by the vertical dipping method is successful only for the odd-numbered spacers, and this alternation effect is drastic, appearing in an all-or-none fashion (see Figure 3b). The 6Azn-urea amphiphiles having an even-numbered spacer gave ca. 1.3-fold larger lift-off areas than those with an odd-numbered one in the spread monolayer. In other words, the amphiphiles with an even-numbered carbon spacer show a looser packing state. This behavior seems inconsistent with that generally recognized for liquid crystal11 and polymer12 assemblies in which the even-numbered chain shows closer packing properties due (26) Yamada, N.; Okuyama, K.; Serizawa, T.; Kawasaki, M.; Oshima, S. J. Chem. Soc., Perkin Trans. 2 1996, 2707.

Langmuir, Vol. 19, No. 22, 2003 9303

to the higher molecular symmetry. We assume that in our system, the ethereal oxygen atom connecting the Az moiety and the carbon chain contributes to the packing state, namely, an even-numbered carbon spacer provides an oddnumbered spacer as the total number of elements. Precise understanding requires further structural characterizations in the monolayer state. A powerful candidate for this purpose can be grazing incidence X-ray reflectivity measurements. In general, the appearance of carbon parity effects strongly depends on the degrees of freedom within the assembly. For example, polydiacetylene LB films of carboxylic acid derivatives indicate no discernible odd/ even effect,27 in contrast with the clear odd/even nature of the methylene spacer in SAMs formed by disulfide compounds on gold.20 It is interpreted that some motional freedom is allowed in the LB monolayer and multilayers, which can conceal the parity effect. On the other hand, the molecular motions in the self-assembled monolayers are highly hindered through covalent bonding onto a solid surface, resulting in the appearance of the parity effects. We actually confirmed that no discernible odd-even effects were observed in the monolayer of carboxylic acid derivatives having the homologous structure (6Azn-COOH).24 Therefore, we conclude that the obvious appearance of the odd-even effects on the Langmuir monolayers floating on the highly fluid water surface is of a nature peculiar to the present 6Azn-urea monolayer systems. Most probably, this nature is derived from the strong fixation and motional restriction of the headgroups via the bifurcated hydrogen bonds like self-assembled monolayers attached on a solid substrate. The highly restricted environment is evident from the fact that the trans-to-cis photoisomerization is fully hindered in the Langmuir monolayer of 6Az10-urea.7 Yamada et al. reported that the systematic periodic change with carbon number in the packing state of a family of azobenzene-containing amphiphiles is observed for cast films.26 However, the appearance of odd/even effects in Langmuir monolayers on water observed in the present systems is quite unique and is observed for the first time with the urea derivatives. The C-H antisymmetric stretching peaks are positioned at 2926-2934 cm-1. For this band, no odd-even nature was observed; instead, the molecule having the longer spacer showed an absorption band at the higher wavenumber. In any case, the alkyl chains are not in a highly crystalline trans-zigzag state but are relatively disordered.28,29 As discussed above, the appearance of the oddeven nature is frequently explained by the trans-zigzag packing of the alkyl chain, namely, the two terminal bonds of a methylene chain adopt parallel (even number) or bent (odd number) directions. The disordered conformation of the alkyl chain observed here may be somewhat conflicting. Two explanations can be possible in this regard. (i) Only the tail part of Az is disordered, whereas the spacer part is highly crystallized. Each molecule has both alkyl tail and alkylene spacer, and the IR spectrum provides the overall signals. (ii) The periodic effects are attained as the result of interdependence between the urea and Az parts even if the spacer chain is in such a disordered state. This situation may be similar to that in fluid liquid crystal materials where the periodic alternations are often observed in the nematic to isotropic phase transition temperature.11 If the latter is the case, the effects are (27) Tieke, B.; Lieser, G. J. Colloid Interface Sci. 1982, 88, 471. (28) Kajiyama, T.; Oishi, Y.; Uchida, M.; Tanimoto, Y.; Kozuru, H. Langmuir 1992, 8, 1563. (29) Sapper, H.; Cameron, D. G.; Mantsch, H. H. Can. J. Chem. 1981, 59, 2543.

9304

Langmuir, Vol. 19, No. 22, 2003

Kobayashi and Seki

Table 1. Various Odd-Even Effects in 6Azn-urea Assemblies n assembled state microcrystal monolayer on water

method π-A isotherm

LB transfer monolayer on quartza

UV-visible spectra

multilayerb

X-ray diffraction infrared spectra

measurement

3

4

5

6

7

melting point (°C) occupying area (nm2) collapse pressure (mN m-1) inflecting point slope compressibility (10-2)c LB depositiond λmax (nm) A242/Along tilt angle (deg)e hydrogen bondingf

163-165 0.29 52 O steeper 0.90 O 325 1.07 44.6 weaker

154-156 0.33 48 × smoother 4.32 × 327 0.92 43.4 stronger

148-149 0.30 51 O steeper 1.51 O 321 1.08 43.7 weaker

154-155 0.34 48 × smoother 5.87 × 325 0.86 43.1 stronger

141-143 0.30 54 O steeper 0.76 O 317 1.08 43.6 weaker

a Vertical dipping method onto a hydrophilic quartz plate. b 30 layers by the horizontal lifting method. c The ratio of area decrease at 60 min to that at the initial stage. d Vertical dipping method onto hydrophobized quartz at 20 mN m-1. e Estimated from the layer spacing and molecular length from the CPK model. f Judged from the peak positions of νCdO and δN-H bands.

derived from a strong cooperativity of the molecular assemblies. 4. Conclusion This paper proposed unique odd-even effects observed in 6Azn-urea assemblies, where n denotes the carbon number of the methylene spacer between the urea and Az moieties (n ) 3-7). A number of alternating effects with increasing n are observed in various assembly types such as bulk microcrystals, monolayers both on water and on a solid substrate, and multilayers on a substrate. Among the above properties, the effects observed for the spreading behavior of Langmuir monolayers on water and for the dynamic LB deposition process (compressibility and deposition behavior) are particularly unusual because the odd-even effects are apt to appear in a restricted environment with suppressed molecular motions. To our knowledge, this is the first evident observation of the odd-

even effects in Langmuir monolayers. The effects should be the consequence of the formation of bifurcated hydrogen bonds between urea headgroups since the other hydrophilic headgroup, carboxylic acid, does not lead to such effects. Acknowledgment. We thank Professor K. Ichimura of the Tokyo University of Science, Dr. Nakagawa of the Chemical Resources Laboratory at the Tokyo Institute of Technology, and Shusaku Nagano of Nagoya University for helpful discussions. We also thank Dr. Saiki of the Tokyo Institute of Technology for X-ray measurements. This work was in part supported by a Grant-in-Aid for Scientific Research on the Priority Areas (No. 417) from the Ministry of Education, Culture, Sports, Science and Technology and the CREST Program of Japan Science and Technology Corporation. LA035056W