Sugar-Integrated Gelators of Organic Solvents

S. Shinkai and O. Gronwald Sugar-Integrated Gelators 4328  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4328 $ 17.50+.50/0 Chem. Eur. J. 2001, 7, No. 20 CONCEPTS Sugar-Integrated Gelators of Organic Solvents Oliver Gronwald[a] and Seiji Shinkai*[a, b] Abstract: Some methyl 4,6-O-benzylidene monosaccharides can act as strong low molecular weight gelators for various organic solvents. As they are accessible in a variety of homologues, each with a unique molecular architecture, they can be used for systematic studies of gelation phenomena. Structural details of their hydrogenbond-based fiber network in the gel phase can be resolved by small angle X-ray scattering (SAXS). Analysis of the molecular arrangement in a single crystal can be a valuable tool for the prediction of gelation ability presupposing that the elongated shape of the gel fibers arises from an anisotropic assembly of the gelator molecules into one-dimensional aggregates. It is found that some derivatives act as ªsupergelatorsº, which can gelate hydrocarbon solvents with 0.03 ± 0.05 wt %. The recent results emerging from these investigations will be outlined in this article. Keywords: carbohydrates ´ hydrogen bonds ´ sol ± gel processes ´ X-ray diffraction Introduction Motivated by the numerous applications for gels, formed by dilute solutions of polymers, proteins, and inorganic substances,[1] the development of new low molecular weight gelators for organic solvents and investigation of their particular selfassembly properties have recently received much attention. They not only gelate various organic solvents but also create novel networks with fibrous superstructures, which can be characterised by scanning electron microscopy (SEM) pictures of xerogels.[2±13] The self-assembly of these gelling agents to fiber-like structures, which entangle to form a three[a] Prof. Dr. S. Shinkai, Dr. O. Gronwald Chemotransfiguration Project Japan Science and Technology Corporation 2432 Aikawa, Kurume, Fukuoka 839-0861 (Japan) [b] Prof. Dr. S. Shinkai Department of Chemistry and Biochemistry Graduate School of Engineering, Kyushu University Hakozaki, Hisgashi-ku, Fukuoka 812-8581 (Japan) Fax: (‡ 81) 92-642-3611 E-mail: seijitcm@mbox.nc.kyushu-u.ac.jp Chem. Eur. J. 2001, 7, No. 20 dimensional (3D) network, prevents the solvent from flowing similar to their macromolecular and inorganic counterparts.[14] Gelators can be classified according to their driving forces for molecular aggregation into two categories: nonhydrogenbond-based gelators and hydrogen-bond-based gelators. Cholesterol derivatives[7±10] are typical examples of the former group whereas aliphatic amide derivatives[2±5] and saccharidecontaining gelators[9, 15±17] are the main representatives of the latter group. As general guidelines for the design principles are accepted: i) the presence of strong self-complementary and unidirectional interactions to enforce one-dimensional self-assembly; ii) control of the fiber-solvent interfacial energy to control solubility and to prevent crystallisation; and iii) some factor to induce fiber cross-linking for network formation.[18] Despite the recent achievements elucidating the molecular prerequisites for gelation ability, the control of this aggregation phenomena is still a challenging goal. Recent studies demonstrated that methyl 4,6-O-benzylidene derivatives of monosaccharides are well-suited to study the structural prerequisites for gelation ability.[17] In gels these monomers establish rigid, strong, and highly directional hydrogen bonds. Their unique and well-defined molecular architecture added to the easy accessibility of a wide variety of isomers, each of which can be obtained as a single enantiomer and allows systematic studies to connect monomer structure and gelation ability. No other gelator discovered so far shows such a variety in its homologues: undoubtely, this mechanistic view utilising a rich carbohydrate library is one of the largest merits of sugar-integrated gelators. Here we give a brief description of their gelation properties and recent results of structural studies in gel state by small angle X-ray scattering (SAXS). The concept of unidirectional interactions as prerequisite for gelation ability is strengthened by correlation of the monomer structure with the molecular arrangement in single crystal and gel. Results and Discussion Gelation properties: Methyl 4,6-O-benzylidene derivatives of monosaccharides belong to a well-established class of compounds, though they have yet not been fully explored as gelators. Also well known to form strong and highly directional hydrogen bonds, this type of compounds meets the requirements for systematic studies. Their features are i) un-  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4329 $ 17.50+.50/0 4329 CONCEPTS S. Shinkai and O. Gronwald modified 2-OH and 3-OH group; ii) protection of the 1-OH group by a methyl group; and iii) protected 4-OH and 6-OH group with a benzylidene group, whereas their different abilities to gelate solvents are solely due to configurational isomerism. So far, eleven methyl 4,6-benzylidene derivatives of the monosaccharides d-glucose, d-mannose, d-allose, daltrose, d-galactose, and a-d-idose (1 ± 11) have been investigated (Scheme 1). Among the criteria that have to be taken into account when the monosaccharides derivatives should be classified according their ªqualityº as gelators are: i) versatility of gelating solvents, ii) stability of the gel, including Tgel and other physico-chemical properties, iii) minimum gelator concentration (Cmin). A comparison of the gelation potential for 34 different solvents reveals the versatility of gelated solvents for the different methyl 4,6-benzylidene derivatives (Table 1):[17b, d, e] Among the a-monosaccharides 1, 2, 3, 4, 9, and 11 only the gluco 1, manno 2, and galacto 9 isomers act as gelators. The widest variety of solvents is gelled by the a-galacto isomer 9. The range of gelated solvents covers apolar hydrocarbon and aromatic solvents (entries 1 ± 8), carbon tetrachloride, carbon disulfide, diethyl ether, diphenyl ether, n-octanol, triethylamine, triethylsilane, and tetraethoxysilane. The amanno 2 isomer is able to gelate the similar range with the exception of cyclohexane, benzene, carbon tetrachloride, diethyl ether, n-octanol, triethylamine, and tetraethoxysilane. Additionally, 2 can gelate water at 3 wt % which is very uncommon for gelators. Compared with a-galacto Scheme 1. Methyl 4,6-O-benzylidene monosaccharides as gelators. 4330 and a-manno isomers, the a-gluco 1 shows a reduced gelation ability since it gels only benzene, toluene, p-xylene, carbon tetrachloride, diphenyl ether and tetraethoxysilane. In contrast, the a-allo- 3, a-altro- 4 and a-idopyranoside 11 tend to be insoluble in or precipitate out of solvents in group I. Most a-compounds dissolve in the more polar solvents of group II. On the other hand, among the b-isomers only the manno 6 and galacto 10 isomers can be considered as gelators. Both gelate nearly a similar range of solvents as their anomeric counterparts. In contrast to its a-anomer 9, the banomer 10 forms a partial gel (Gp) in nitrobenzene, ethyl formate, and methyl acetate but precipitates in diethyl ether and tetraethoxysilane (9: both G*). Compared with its opposite anomer, b-manno 6 exhibits additional gelation ability for benzene and carbon tetrachloride but no gelation potential for water. The b-gluco- and allopyranosides (5, 7) tend to form mainly gel-like solids as denoted by the numerous ªPSº marks. Similarly, the b-altropyranoside 17 exhibits a low solubility in the major part of the solvents tested, reflected by the ªPº and ªPSº marks. In order to compare the gel qualities, the gel ± sol phase transition temperatures (Tgel) of gels of 1, 2, 6, 9, and 10 in p-xylene estimated by the oil-bath method are plotted against the gelator concentration (Figure 1). For the same concentration the Tgel values always appear in the order of: 10 b 6 b 9 b 1 b 2. In other solvents similar results are obtained. Together with the results from Table 1 this confirms the conclusion that the galacto derivative tends to be more efficient than the manno derivative and that the a-anomer is inferior to the b-anomer. Consequently, a-gluco occupies an intermediate position between gelators and non-gelators because it forms Ps as well as gels. Therefore, the optimal requirements for a compound to be classified as a gelator are fulfilled only by methyl 4,6-Obenzylidene derivatives of dmannose and d-galactose. Compared with other low molecular weight gelators these saccharides gelate a broader range of solvents. Especially, a-manno 2 reveals highly flexible gelation properties. Few other gelators are capable to gelate a broad variety of organic solvents and water at the same time.[19] Not only the variety of solvents but also the Cmin reveals a further characteristic of sugar-based-gelators. Methyl 4,6-O-p-nitrobenzylidene-a-d-galactopyranoside (12) and methyl 4,6-benzylidene-a-d-mannopyranoside  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4330 $ 17.50+.50/0 Chem. Eur. J. 2001, 7, No. 20 Sugar-Integrated Gelators 4328 ± 4334 Table 1. Organic solvents tested for gelation by 1 ± 11.[a] Organic solvent group I 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 group II 21 22 23 24 25 26 27 28 29 30 31 32 33 34 1 2 3 n-hexane[b] n-heptane[b] n-octane[b] cyclohexane[b] methylcyclohexane[b] benzene[b] toluene[b] p-xylene[b] nitrobenzene[b] carbon tetrachloride[b] carbon disulfide[c] diethyl ether[b] diphenyl ether[b] ethyl formate[b] methyl acetate[b] n-octanol[b] triethylamine triethylsilane tetraethoxysilane water PS* PS* PS* PS* PS* G G* G* S G* P* S* G S S S S* PS* G* PS G* G* G* PS* G* PS* G* G* S P G* S* G S S S S G* S G P* P* P P P PPS PPS PPS S P P* P PPS S S S S* P P S 1,2-dichloroethane dichloromethane[b] chloroform ethyl acetate[b] ethyl malonate[b] acetone[b] methyl ethyl ketone[b] acetonitrile[b] ethanol[b] n-propanol[b] n-butanol[b] hexanoic acid acetic anhydride glycerol S S S S S S S S S P P S S S S S S S S S S S S S S S S S S S S S S S S S S S S S S S 4 a-series 9 11 5 6 I* I* P* P* PPS* PPS PPS PPS S P* P* P* PPS S S S S* S* PPS* S G* G* G* G* G* G* G* G* S G* G* G* G* P S G G Gp G* S P* P* P P* P S S S S P PPS S* S S S S P P* P P P S* P S* P S* P S* P S* P S* P S* P S* PS P S* PPS* P S* P S* P* PS PS P* P S* P* P* G* G* G* G* G* G* G* G* S G* G* S* G S S P S G* P S S S S P S S S S S S S S S S S S S P S S S S S P S P S S S S S S S S S S S S S S S S PS * PS PS PS PS P PS P P PS PS PS PS PS S S S S S S S S P P PS P S S 7 b-series 8 10 P S* P S* P S* P S* P S* PS P S* P S* S P S* P S* P* PS S S PS S P S* P PS P* P* P* PS* P* PPS PS PS S P PS P* PPS S P P P P* P P G* G* G* G* G* G* G* G* Gp G* G* P G* Gp Gp G G G P S S S S P S S P P P P PS PS S S P P S P S S P P S S S P S S S S S S S S S S S P P P S S [a] 3.0 wt/vol %, * ˆ 1.0 wt/vol %, G ˆ gel, PS ˆ self-supporting precipitate (gel-like solid which is unstable to inversion), PPS ˆ partial self supporting precipitate, P ˆ precipitation, S ˆ solution, I ˆ insoluble. [b] Dried over molecular sieves 4 Š. [c] Dried over anhydrous magnesium sulfate. 0.05 wt/vol %.[17c] Compared with the typical gelator concentration range of 3 ± 15 mm [12c] the Cmin of 0.9 ± 1.22 mm for 12 and 1.77 ± 2.48 mm for 2 represent the lowest concentrations reported for organic solvents so far. Especially, the drastically lowered minimum gelation concentration of 12 compared with its non-p-nitro-analogue 9 opens new perspectives for certain solvents. Through preparation of further p-nitroanalogues, lower Cmin might be obtained, which can facilitate further SAXS investigations since higher diluted systems provide clearer scattering pattern. Figure 1. Plots of Tgel against gelator concentration in p-xylene. (2) act as ªsupergelatorsº for apolar solvents. Both saccharides can gelate hydrocarbons (n-hexane, n-heptane, n-octane, n-decane, and cyclohexane) in concentrations around 0.03 ± Chem. Eur. J. 2001, 7, No. 20 Mode of aggregation: Methyl glycosides of 4,6-O-benzylidene derivatives of monosaccharides aggregate the solvents through formation of a hydrogen-bond-based gel network. The evidence is given by FT-IR and temperature dependent 1 H NMR spectroscopy.[17a, b] Due to intermolecular and intramolecular hydrogen-bonding interactions, no nOH peak for a Ä free OH group (around 3600 cmÀ1) could be detected for the solid samples (KBr) of all monosaccharides. In the gel state all signals are more broadened and the nOH values for the OH Ä groups appear in two groups between 3220 ± 3475 cmÀ1 and 3573 ± 3588 cmÀ1 (0.15 ± 0.60 wt % in toluene) and can there-  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4331 $ 17.50+.50/0 4331 CONCEPTS S. Shinkai and O. Gronwald fore be assigned to the intermolecular hydrogen bonds and free OH groups, respectively. The peak intensity ratio (R) of hydrogen-bonded OH to free OH abruptly decreases at the gel ± sol transition temperature (Tgel), indicating that the gel network is primarily stabilised by intermolecular hydrogen bonding.[17b] Additionally, the gels provided by sugar-integrated gelators offer the unique possibility to probe their thermal stability by monitoring the change of intermolecular aggregation near the Tgel region by temperature dependent 1H NMR spectroscopy[20] by observing the chemical shift values for dOH .[17b] Due to the formation of strong intermolecular bonds at Tgel , the chemical shifts have their maximum downfield values at Tgel . Since the molecular motion of gelators drastically changes at Tgel , this phenomenon is reflected by the line broadening effect in the 1H NMR spectrum.[17b] In the gel phase the mobility of gelator molecules is significantly suppressed, whereas in the sol phase it is comparable to that of a homogeneous solution. Therefore the width of the peak at half its height, d1/2 of the PhCH methine proton is nearly constant above Tgel , while it increases with falling temperature below Tgel . Molecular arrangement in single crystal and gel state: The above-mentioned results highlight that variations in the saccharide configuration result in a drastical change of the gelation properties. In an attempt to elucidate the origin of these differences, we turned our attention to the molecular arrangement in a single crystal. The question whether solidstate properties actually reflect solution properties has lead to contradictory results in the history of gel research. Crystal structures of gelator molecules have been scarcely reported up to now. Weiss et al.[6] proved that the gel fiber morphology of cholesterylanthrachion-2-carboxylate differs from the molecular packing of single crystals. For one urea-based hydrogen-bonded gelator Feringa[12b] demonstrated that its crystal structure does not account for low-angle reflections observed in toluene gels of this compound. Both studies suffer the disadvantage that each of them is based on only one example. Moreover, the fact that both investigated compounds showing polymorphism, limits the significance of the obtained results. In contrast, a very recent study demonstrated the successful design of efficient aryl-l-cystine hydrogelators based on the analysis of the crystalline ªfibrousº molecular orientation of the non gelator-analogue di(p-toluoyl)-l-cystine.[19a] Since methyl 4,6-O-benzylidene monosaccharides are not prone to polymorphism and their organogel fibers are not so ªwetº with solvent molecules,[6, 8, 10, 21] we expected them to display more or less crystal-like character, although the organogel fiber structure is somewhat more disordered than the crystal structure.[10] This implies, that the X-ray structure can make an important contribution to explain the mechanism of the organogel fiber formation. Since the examined saccharide gelators tend to grow as needles, only some single crystals suitable for X-ray analysis have been isolated so far. The crystal structures of 1,[17d] 3,[22] 4,[23] and 11[24] have been determined or are available from the Cambridge Structural Database. As shown by Hanabusa et al.[2] and Feringa et al.,[12] amide-based- and urea-basedgelators tend to form one-dimensional hydrogen-bond arrays. 4332 In the solid state 1 forms one-dimensional zigzag chains, in which molecules are connected by two hydrogen-bonds using 2-OH and 3-OH (Figure 2). In addition, phenyl groups positioned at the edge of this one-dimensional chain can show a p ± p interaction, with the phenyl groups arranged in other one-dimensional chains. Although this interaction is weaker than the hydrogen-bonding interaction, it may play an important role when chains grow up as a bundle. Since these characteristics of 1 seem to satisfy the prerequisites for a gelator, these results match with the observed gelation ability in aromatic solvents, carbon tetrachloride, and tetraethoxysilane. Figure 2. Molecular packing in methyl 4,6-O-benzylidene-a-d-glucopyranoside (1). Compound 3 (Figure 3) also exhibits a one-dimensional chain structure. In this case, however, the molecular packing is supported by only one intermolecular hydrogen bond and the second one is used for the intramolecular interaction between the 3-OH and the 1-OMe. Hence, 3 basically satisfies the primary prerequisite to be one-dimensional. In the gel phase Figure 3. Molecular packing in methyl 4,6-O-benzylidene-a-d-allopyranoside (3).  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4332 $ 17.50+.50/0 Chem. Eur. J. 2001, 7, No. 20 Sugar-Integrated Gelators where the intermolecular hydrogen-bonding interaction must compete with solvation, however, the fibrous structure cannot be as stabilised as that of 1. The instability gives rise to the disordered structure including free OH groups, the hydrogen bonds of which eventually result in the formation of threedimensional, insoluble aggregates. This situation is reflected by many P and P* marks for which 1 has G and G* marks. The crystal structure of compound 4 features saccharide molecules connected by hydrogen bonds between two OH groups and the 5-ether oxygen into two-dimensional layers (Figure 4). As a result, 4, cannot construct a one-dimensional hydrogen-bonding array. In the crystal of compound 11, on the other hand, two OH groups are both used to form the intramolecular hydrogen bonds and no significant intermolecular hydrogen bonding is found in the crystal structure (Figure 5). This may be called a zero-dimensional hydrogenbonding array. As expected, compound 11 is very soluble in many organic solvents. 4328 ± 4334 investigated by SEM.[25] Although in general the observed three-dimensional fiber network has been ascribed to the gel structure, the shrinking step induced by the freeze-drying procedure can result in collapses of the frail three-dimensional network. Thus, SEM might focus on general shapes and morphologies rather than on absolute quantities such as diameters, lengths, or topologies. To overcome these problems, synchrotron small angle X-ray scattering (SAXS), a powerful method to explore directly the supramolecular structure, can be used. Since the synchrotron X-ray is almost 106 times stronger than conventional X-rays, it has great a advantage for diluted systems such as organogels. Terech et al.[26] analyzed SAXS from different non-sugar-based organogelators. We measured SAXS with a BL-15A SAXS station at the Photon Factory High Energy Research Organization in Japan[27] from the sugar-based organogelators for the first time.[17e] The typical temperature dependent scattering profile of 6 (1.5 wt % in p-xylene) displays at 50 8C two broad peaks at q ˆ 0.018 and 0.11 ŠÀ1 (Figure 6). With increasing temperature (60 8C) the position of these peaks remains unchanged, however, the intensity of the peak at q ˆ 0.018 ŠÀ1 decreases about 50 % and it disappears completely at 70 8C. Because the sol ± gel transition temperature (Tgel) of this system was estimated by the oil-bath method to be 70 8C, this result supports the assumption that this peak can be assigned to the supramolecular structure of the gel. Figure 4. Molecular packing in methyl 4,6-O-benzylidene-a-d-altropyranoside (4). Figure 5. Molecular packing in methyl 4,6-O-benzylidene-a-d-idopyranoside (11). Although not fully clear, these results underline that the molecular arrangement in the single crystal can be relevant for the ability of monomers to assemble in one-dimensional aggregates. The concept of unidirectional interactions as prerequisites for gelation ability is therefore strengthened as derived from investigations of the affluent saccharide library. Investigation of the gel structure: To obtain visual insight in these structures, dry samples of organic gel fibers have been Chem. Eur. J. 2001, 7, No. 20 Figure 6. Temperature dependence of the SAXS profiles for 6 (1.5 wt % in p-xylene). Recently, further investigations carried out at the BL45XU biophysics beam line at SPring8 in Japan[28] gave an improved scattering quality for gels of 2, 6, 9, and 10 in p-xylene due to the higher S/N ratio. The fit of the scattering profiles with a  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4333 $ 17.50+.50/0 4333 CONCEPTS S. Shinkai and O. Gronwald solid cylinder model indicated that all gels consist of fibers with an approximate diameter of 60 Š. For a-manno-saccharide 2 the SAXS investigation provided four peaks. The ratio of their position gives strong evidence that the cylinders assemble in a hexagonal packing mode.[29] These results clearly suggest that organogels of methyl 4,6-O-benzylidene derivatives of glycosides can be resolved at nanoscopic scale by SAXS. Current investigations are ongoing to correlate the molecular structure with the SAXS and SEM results. Outlook The aforementioned results clearly demonstrate that methyl 4,6-O-benzylidene monosaccharides serve as excellent library compounds for the investigation of the gelation phenomenon based on the formation of hydrogen bonds. Preliminary results suggest that these compounds provide a possible correlation of the molecular structure to the different gelation properties by analysis of the molecular arrangement in single crystal. Furthermore, structural details of their gel network can be deduced from the analysis of SAXS data. Both strategies to approach the gelation phenomenon from a microscopic and macroscopic viewpoint contribute to a better understanding of the process how small molecules gelate solvents. Additional studies of the sugar-integrated gelators must be useful for the discovery and design of new gelators. We believe that the saccharide library provided by nature can be applied further, in particular to the design of molecular assemblies, such as macrocycles, DNA mimics, monolayers, bilayer membranes, liquid crystals. [1] a) Polymer Gels: Fundamentals and Biomedical Applications (Eds.: D. Derossi, K. Kajiwara, Y. Osada, A. Yamauchi), Plenum, New York, 1991; b) J.-M. Guenet, Thermoreverible Gelation of Polymers and Biopolymers, Academic Press, London, 1992; c) ªgelsº: Prog. Colloid Polym. Sci. (Eds.: Kremer, G. Lagaly), 1996, 102; d) R. J. P. Corriu, D. Leclercq, Angew. Chem. 1996, 108, 1524 ± 1540; Angew. Chem. Int. Ed. Engl. 1996, 35, 1420 ± 1436; e) M. T. Reetz, Adv. Mater. 1997, 9, 943 ± 954. [2] a) K. Hanabusa, K. Okui, K. Karaki, H. Shirai, J. Chem. Soc. Chem. Commun. 1992, 1371 ± 1373, and references therein; b) K. Hanabusa, Y. Yamada, M. Kimura, H. Shirai, Angew. Chem. 1996, 108, 2086 ± 2088; Angew. Chem. Int. Ed. 1996, 35, 1949 ± 1950; c) K. Hanabusa, K. Shimura, K. Hirose, M. Kimura, H. Shirai, Chem. Lett. 1996, 885 ± 886; d) K. Hanabusa, A. Kawakami, M. Kimura, H. Shirai, Chem. Lett. 1997, 191 ± 192. [3] E. J. de Vries, R. M. Kellog, J. Chem. Soc. Chem. Commun. 1993, 238 ± 240. [4] M. Takafujii, H. Ihara, C. Hirayama, H. Hachisoko, K. Yamada, Liq. Cryst. 1995, 18, 97 ± 100. [5] J.-E. S. Sohna, F. Frages, J. Chem. Soc. Chem. Commun. 1997, 327 ± 328. [6] E. Otsumi, P. Kamasas, R. G. Weiss, Angew. Chem. 1996, 108, 1423 ± 1425; Angew. Chem. Int. Ed. Engl. 1996, 35, 1324 ± 1326, and references therein. 4334 [7] P. Terech, I. Furman, R. G. Weiss, J. Phys. Chem. 1995, 99, 9558 ± 9566, and references therein. [8] K. Murata, M. Aoki, T. Suzuki, T. Hanada, H. Kawabata, T. Komori, F. Oseto, K. Ueda, S. Shinkai, J. Am. Chem. Soc. 1994, 116, 6664 ± 6674, and references therein. [9] T. D. James, K. Murata, T. Harada, K. Ueda, S. Shinkai, Chem. Lett. 1994, 273 ± 276. [10] a) S. W. Jeong, K. Murata, S. Shinkai, Supramol. Sci. 1996, 3, 83 ± 86; b) S. W. Jeong, S. Shinkai, Nanotechnology 1997, 8, 179 ± 183. [11] T. Brotin, R. Utermöhlen, F. Fagles, H. Bouas-Laurent, J.-P. Desvergne, J. Am. Chem. Soc. Chem. Commun. 1991, 416 ± 418. [12] a) J. van Esch, S. de Feyter, R. M. Kellogg, F. de Schryver, B. L. Feringa, Chem. Eur. J. 1997, 3, 1238 ± 1243; b) J. van Esch, F. Schoonbeek, M. de Loos, H. Kooijman, A. L. Spek, R. M. Kellogg, B. L. Feringa, Chem. Eur. J. 1999, 5, 937 ± 950; c) F. S. Schoonbeek, J. H. van Esch, R. Hulst, R. M. Kellogg, B. L. Feringa, Chem. Eur. J. 2000, 6, 2633 ± 2643. [13] For recent comprehensive reviews, see: a) P. Terech, R. G. Weiss, Chem. Rev. 1997, 97, 3133 ± 3159; b) S. Shinkai, K. Murata, J. Mater. Chem. 1998, 8, 485 ± 495; c) J. van Esch, F. Schoonbeeck, M. de Loos, E. M. Veen, R. M. Kellogg, B. L. Feringa, Nato ASI Ser. C 1999, 527, 233 ± 259. [14] a) P. J. Flory, Faraday Discuss. R. Soc. Chem. 1974, 57, 8; b) T. Tanaka, Am. Sci. 1981, 244, 110 ± 123. [15] S. Yamasaki, H. Tsutsumi, Bull. Chem. Soc. Jpn. 1996, 69, 561 ± 564 and references therein. [16] a) N. Amanokura, K. Yoza, H. Shinmori, S. Shinkai, J. Chem. Soc. Perkin Trans. 2 1998, 2585 ± 2591; b) N. Amanokura, Y. Kanekyio, S. Shinkai, D. N. Reinhoudt, J. Chem. Soc. Perkin Trans. 2 1999, 1995 ± 2000. [17] a) K. Yoza, Y. Ono, K. Yoshihara, T. Akao, H. Shinmori, M. Takeuchi, S. Shinkai, D. N. Reinhoudt, J. Chem. Soc. Chem. Commun. 1998, 907 ± 908; b) K. Yoza, N. Amanokura, Y. Ono, T. Akao, H. Shinmori, M. Takeuchi, S. Shinkai, D. N. Reinhoudt, Chem. Eur. J. 1999, 5, 2722 ± 2729; c) R. Luboradzki, O. Gronwald, A. Ikeda, S. Shinkai, Chem. Lett. 2000, 1148 ± 1149; d) R. Luboradzki, O. Gronwald, M. Ikeda, S. Shinkai, D. N. Reinhoudt, Tetrahedron 2000, 56, 8697 ± 8701; e) O. Gronwald, K. Sakurai, R. Luboradzki, T. Kimura, S. Shinkai, Carbohydr. Res. 2001, 331, 307 ± 318. [18] J. H. van Esch, B. L. Feringa, Angew. Chem. 2000, 112, 2351 ± 2354; Angew. Chem. Int. Ed. 2000, 39, 2263 ± 2266. [19] a) F. M. Menger, K. L. Caran, J. Am. Chem. Soc. 2000, 122, 11 679 ± 11 691; b) L. A. Estroff, A. D. Hamiltion, Angew. Chem. 2000, 112, 3593 ± 3595; Angew. Chem. Int. Ed. 2000, 39, 3447 ± 3450, and references therein. [20] D. C. Duncan, D. G. Whitten, Langmuir 2000, 16, 6445 ± 6452. [21] a) R. Mukkamala, R. G. Weiss, J. Chem. Soc. Chem. Commun. 1995, 375 ± 376; b) L. Lu, M. Cocker, R. E. Bachmann, R. G. Weiss, Langmuir 2000, 16, 20 ± 34. [22] P. R. Muddasani, B. Bernet, A. Vasella, Helv. Chim. Acta 1994, 77, 334 ± 350. [23] E. Bozo, A. Vasella, Helv. Chim. Acta 1992, 75, 2613 ± 2633. [24] H.-M. Liu, Y. Sato, Y. Tsuda, Chem. Pharm. Bull. 1993, 41, 491 ± 495. [25] For the preparation of dry samples for SEM observations, see ref. [8] and S. W. Jeong, S. Shinkai, Nanotechnology 1997, 8, 179 ± 183. [26] a) P. Terech, E. Ostuni, R. G. Weiss, J. Phys. Chem. 1996, 3759 ± 3766; b) P. Terech, I. Furman, R. G. Weiss, H. B. Laurent, J. P. Desvergne, R. Ramasseul, Faraday Discuss. 1995, 345 ± 358; c) P. Terech, J. J. Allegraud, C. M. Garner, Langmuir 1998, 14, 3991 ± 3998. [27] Y. Amemiya, K. Wakabayashi, T. Hamanaka, T. Wakabayashi, T. Matsushita, H. Hashizume, Nucl. Instrum. Method. 1983, 208, 471 ± 472. [28] T. Fujisawa, J. Synchrotron Radiat. 1999, 12, 194 ± 198. [29] K. Sakurai, T. Kimura, O. Gronwald, K. Inoue, S. Shinkai, Chem. Lett., in press.  WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001 0947-6539/01/0720-4334 $ 17.50+.50/0 Chem. Eur. J. 2001, 7, No. 20