Langmuir 2008, 24, 11431-11441
11431
Linear-Dendritic Supramolecular Complexes as Nanoscale Reaction Vessels for “Green” Chemistry. Diels-Alder Reactions between Fullerene C60 and Polycyclic Aromatic Hydrocarbons in Aqueous Medium Arsen Simonyan† and Ivan Gitsov*,†,‡ Department of Chemistry, College of EnVironmental Science and Forestry, State UniVersity of New York, Syracuse, New York 13210, and The Michael M. Szwarc Polymer Research Institute, State UniVersity of New York, Syracuse, New York 13210 ReceiVed May 23, 2008. ReVised Manuscript ReceiVed July 9, 2008 This study describes the first Diels-Alder (DA) reaction performed in aqueous medium with highly hydrophobic compoundssfullerene (C60) as the dienophile and anthracene (An) or tetracene (Tet) as the dienes, respectively. The reactions are performed in nanocontainers, constructed by self-assembly of linear-dendritic amphiphilic copolymers with poly(ethylene glycol), PEG or poly(ethylene oxide), PEO as the hydrophilic blocks and poly(benzyl ether) monodendrons as the hydrophobic fragments: G3PEO13k, dG3 and dG2. Comparative studies under identical conditions are carried out with an amphiphilic linear-linear copolymer, poly(styrene)1800-block-PEO2100, PSt-PEO, and the nonionic surfactant Igepal CO-720, IP720. The binding affinity of supermolecules built of these amphiphiles toward the DA reagents decreases in the following order: G3PEO13k > dG3 > PSt-PEO > dG2 > IP720. The kinetic constant of binding is evaluated for tetracene and decreases in a similar fashion: 5 × 10-7 M/min (G3PEO13k), through 4 × 10-7 M/min (PSt-PEO) down to 1.5 × 10-7 M/min for IP720. The mobility of substrates encapsulated in the micellar core, estimated by pyrene fluorescence decay, is 95-121 ns for the micelles of the linear-dendritic copolymers and notably higher for PSt-PEO (152 ns), revealing the much denser interior of the linear analogue. The apparent kinetic constant for the DA reaction of C60 and Tet within the G3PEO13k supermolecule in aqueous medium is markedly higher than in organic solvent (toluene), 208 vs 1.82 M/min. With G3PEO13k the conversions reach 49% for the DA reaction between C60 and An, and 55% for C60 and Tet. Besides the monoadduct (26.5% yield) the reaction with An produces exclusively increasing amounts of D2h-symmetric antipodal bis-adduct, whose yield reaches up to 22.5% after 48 h. In addition to the environmentally friendly conditions notable advantages of the synthetic strategy described are the extended stability of the linear-dendritic nanovessels, the easy collection of the products formed, and the recovery and reuse of unreacted reagents and linear-dendritic copolymers.
Introduction The Diels-Alder (DA) reaction has long been recognized as one of the essential synthetic1 and natural2 tools for the construction of complex molecules of great technological and biomedical importance. Fullerene (C60) represents a rather versatile dienophile for this reaction and has been extensively used in the construction of various interesting adducts and polymeric materials.3 Of particular interest is the reported biological activity of some of these fullerene derivatives.4 Therefore the development of facile synthetic routes to C60 adducts of specific chemical composition and molecular structure is currently the area of intensive research. Regrettably most of the reported DA reactions involving fullerene and linear polycyclic aromatic hydrocarbons (acenes) employ toxic organic solvents * To whom correspondence should be addressed. E-mail:
[email protected]. † College of Environmental Science and Forestry, State University of New York. ‡ The Michael M. Szwarc Polymer Research Institute, State University of New York. (1) Brocksom, T. J.; Nakamura, J.; Ferreira, M. L.; Brocksom, U. J. Braz. Chem. Soc. 2001, 12, 587. (2) (a) Hilvert, D.; Hill, K. W.; Nared, K. D.; Auditor, M. T. M. J. Am. Chem. 1989, 111, 9261. (b) Auclair, K.; Sutherland, A.; Kennedy, J.; Witter, D. J.; Van den Heever, J. P.; Hutchinson, C. R.; Vederas, J. C. J. Am. Chem. 2000, 122, 11519. (c) Stocking, E.; Williams, R. Angew. Chem., Int. Ed. 2003, 42, 3078. (d) Oikawa, H.; Tokiwano, T Nat. Prod. Rep. 2004, 21, 321, and references there in. (3) Yurovskaya, M. A.; Rushkov, I. V Russ. Chem. Bull., Int. Ed. 2002, 51, 367, and references therein. (4) Bosi, S.; Da Ros, T.; Spalluto, G.; Prato, M Eur. J. Med. Chem. 2003, 38, 913, and references therein.
and/or elevated temperatures - reaction conditions, which are not environmentally friendly.5 For example, the first DA reaction between C60 and anthracene was performed in refluxing toluene and after 3 days afforded the monoadduct in 13% yield.6 Syntheses in boiling benzene7 or refluxing naphthalene8 also required extended reaction times and produced the targeted molecule in moderate yields. Despite the fact that DA reactions have been performed with spectacular success in aqueous media and other environmentally friendly solvents,9 the process is not possible with C60 and acenes due to their practical insolubility in water. Early reports suggest that C60 could be solubilized in water by incorporation into micelles formed by nonionic surfactants,10 but toluene was still used to assist the binding, the amount of encapsulated C60 was low and the resulting complexes were ill-defined. In one of our previous studies we have discovered that micelles, formed by amphiphilic dendritic-linear-dendritic copolymers consisting of poly(ethylene glycol), PEG, and (5) Briggs, J. B.; Miller, G. P. C. R. Chimie 2006, 916. (6) Schlueter, J. A.; Seaman, J. M.; Taha, Sh.; Cohen, H.; Lykke, K. R.; Wang, H. H.; Williams, J. M J. Chem. Soc., Chem. Commun. 1993, 972. (7) Tsuda, M.; Ishida, T.; Nogami, T.; Kurono, S.; Ohashi, M. J. Chem. Soc., Chem. Commun. 1993, 1296. (8) Komatsu, K.; Murata, Y.; Sugita, N.; Takeuchi, K.; Wan, T. S. M. Tetrahedron Lett. 1993, 34, 8473. (9) (a) Otto, S.; Engberts, J. B. F. N. Pure Appl. Chem. 2000, 72, 1365. (b) Fringuelli, F.; Piermatti, O.; Pizzo, F.; Vaccaro, L Eur. J. Org. Chem. 2001, 439, and references therein. (c) Brummond, K. M.; Wach, C. K. Mini-ReV. Org. Chem. 2007, 4, 89, and references therein. (10) (a) Hungerbu¨hler, H.; Guldi, D. M.; Asmus, K.-D. J. Am. Chem. Soc. 1993, 115, 3386. (b) Beeby, A.; Eastoe, J.; Heenan, R. K. J. Chem. Soc., Chem. Commun. 1994, 173.
10.1021/la801593y CCC: $40.75 2008 American Chemical Society Published on Web 09/10/2008
11432 Langmuir, Vol. 24, No. 20, 2008
Simonyan and GitsoV
Materials. Fullerene, C60 (99.5%) was purchased from SES Research, Ltd. (Houston). Anthracene (An), Tetracene (Tet), and Pyrene (Py) (all of 98% purity) were purchased from Aldrich (Saint Louis) and used without further purification. The nonionic surfactants: Triton X-100 (TX100) and Igepal CO-720 (IP720) (both from Aldrich) were used as received. The linear-dendritic copolymers used in this study were poly(G3-benzyl ether)-block-PEG10800block-poly(G3-benzyl ether), dG3, poly(G2-benzyl ether)-blockPEG5000-block-poly(G2-benzyl ether), dG2, and poly(G3-benzyl ether)-block-poly(ethylene oxide)13000, G3PEO13k, where the number signifies the molecular weight of the linear block and G3/ G2 is the generation of the dendritic block. They were synthesized by previously described procedures.12,13 The linear-linear copolymer analogue poly(styrene)1800-block-poly(ethylene oxide)2100, PStPEO, was used as obtained from Polymer Source, Inc. (Montreal). Toluene (ACS grade) was purchased from Fisher Scientific (Fair Lawn, NJ), tetrachloromethane and carbon sulfide (both ACS grade) were acquired from Baker (Phillipsburg, NJ). All aqueous solutions were made with deionized (DI) water of 18.2 MΩ. Instrumentation. The preparation of micellar solutions and the solubilization of all hydrophobic substances in the micelles were induced by sonication, performed on a 75D Aquasonic Sonication Bath from VWR Scientific at 22 °C for different periods of time. UV-vis spectra and kinetic measurements were taken on a Beckman DU-640B spectrophotometer. Dynamic measurements of Py fluorescence decay in the copolymer micelles were carried out on a Photochemical Research Associates Inc. System 2000 using a previously published procedure.14 Size-exclusion chromatography (SEC) in organic solvents was performed in tetrahydrofuran (THF) on a line consisting of Waters M510 pump, Waters U6K injector, an Applied Biosystems 785A UV-vis detector and a Viscotek dual refractive index and viscometric detector Model 250. Three 5 µm PLgel columns with pore sizes 50 and 500 Å and a mixed C (Polymer Laboratories, Amherst) were calibrated with 20 poly(styrene) standards (PSS-USA, Warwick). The separations were accomplished at 40 °C and eluent flow rate 1 mL/min. Data were collected and analyzed by Omni-SEC (Viscotek) software package, version 3.1. SEC analyses in aqueous medium were performed on a Waters 2695 Alliance module equipped with three 8 µm Polymer Laboratories Aquagel-OH columns (30, 40 and mixed), Waters M991 UV photodiode array (PDA) detector and an Optilab DSP refractive index detector, from Wyatt Technology, Inc. (Santa Barbara). The solvent was deionized water, stabilized with 0.02% NaN3, the
separation temperature was 40 °C at eluent flow rate of 1 mL/min. Separation data was collected and analyzed by Millenium32 (Waters). 1H NMR spectra were taken at room temperature on Bruker Avance spectrometers at 300 or 600 MHz in CS2 or CS2/CDCl3 1:4 as solvent and tetramethylsilane (TMS) as internal standard where needed. Methods. Solution Preparation. The supramolecular assembly of G3PEO13k and d2 was achieved by stirring for 1 h in DI water with polymer concentrations at 1 × 10-4 M. The same concentration was used for PSt-PEO assuming cmc value similar to previously published data for other PSt-PEO micelles.15 The micellar solutions of the nonionic surfactants TX100 and IP720 were prepared using 10-2 M concentrations (critical micelle concentration, cmc, for TX100 is 2.1 × 10-4 M;16 cmc for IP720 is 4.9 × 10-4 M17a). In the case of dG3, the solubilization in water required four additional sonication/ stirring sequences of 1 h each, until a clear solution was obtained. Solubilization of all hydrophobic probes was best achieved by sonication sequences of 1 h each, repeated at least four times. Dynamic Fluorescence Measurements. The micellar core characteristics were investigated for the 10-4 M solutions of G3PEO13k, dG3, dG2 and PSt-PEO, saturated with pyrene according to a known method.14 Briefly, the procedure consisted of adding ∼2 mg of pyrene powder per milliliter of micellar copolymer solution under nitrogen and the mixture was then subjected to 4 one hour sonication/ stirring cycles. The equilibrated mixtures were filtered through 0.2 µm PTFE Target filters (National Scientific Company) and Py fluorescence decay curves of the clear yellow-colored micellar solutions were recorded by time-correlated single-photon counting.18 Acquired data were used to estimate and compare the core densities of the micelles formed by copolymers with different architecture. UV-Vis studies. The binding affinity of the hydrophobic probe tetracene to three representative micellar systems (G3PEO13k, IP720 and PSt-PEO) was measured in the following fashion: (a) Standard solutions of the polymers with aforementioned concentrations were tested toward predetermined initial loads of C60, tetracene and anthracene in the range of 0.25-1.2 mg/mL, for equal time of 4 h under sonication and centrifuged at 500 g before the UV-vis spectrum of the clear solutions was taken. Solubilized anthracene, tetracene and C60 molarities were obtained by UV-vis using the respective predetermined extinction coefficients: εAn(381 nm) ) 1300 M-1 · cm-1, εTet(474 nm) ) 8900 M-1 · cm-1 and εC60(350 nm) ) 48000 M-1 · cm-1. The latter were determined by building the respective calibration graph for each probe in toluene. Graphs of final solubilized concentration (Cin) versus initial solid amount A0 of the probe were built and the dimensionless value of the slope c ) dCin/dA0 was calculated and used as the descriptor of the binding affinity. (b) Micellar solutions of G3PEO13k, PSt-PEO and IP720 of aforementioned concentrations were subjected to sonication with an excess amount of solid tetracene (1.5 mg/mL). At regular periods of time, the solutions were centrifuged at 500 g for 30 min and UV-vis was taken of the clear layer. The concentration of tetracene in the respective solution was calculated by Beer-Lambert law. A graph for the amount of entrapped tetracene versus time was built. (c) All solutions obtained in (a) as well as the related empty micelle solutions were tested by aqueous SEC with UV PDA detection, to confirm the presence of the hydrophobic probe inside the micellar nanocontainer and to estimate any changes in the size and size distribution of the respective micelle occurring upon sonication and encapsulation. UV Light-Induced Dimerization of An and Tet. Concentrated solutions of the acenes were prepared in 1 mL CS2 (0.12 mg Tet and 0.42 mg An). They were transferred into NMR tubes and irradiated for 1 h with a Multiband Mineralight lamp, model UVGL-
(11) Gitsov, I.; Lambrych, K. R.; Remnant, V. A.; Pracitto, R. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2711. (12) Gitsov, I.; Wooley, K.; Frechet, J. Angew. Chem., Int. Ed. Engl. 1992, 31, 1200. (13) Gitsov, I.; Simonyan, A.; Vladimirov, N. G. J. Polym. Sci.: Part A: Polym. Chem. 2007, 45, 5136. (14) Claracq, J.; Santos, S. F. C. R.; Duhamel, J.; Dumousseaux, C.; Corpart, J.-M. Langmuir 2002, 18, 3829.
(15) Wilhelm, M.; Zhao, C.; Wang, Y.; Xu, R.; Winnik, M.; Mura, J.; Riess, G.; Croucher, M. Macromolecules 1991, 24, 1033. (16) Boquet, P; Silverman, M.; Pappenheimer, A.; Vernon, W. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 4449. (17) (a) Banerjee, P.; Chatterjee, S.; Pramanik, S.; Bhattacharya, S. Colloids Surf., A 2007, 302, 44. (b) Ghosh, S. K.; Khatua, P. K.; Bhattacharya, S. C. Int. J. Mol. Sci. 2003, 4, 562. (18) Mathew, A.; Siu, H.; Duhamel, J. Macromolecules 1999, 32, 7100.
poly(benzyl ether) monodendrons, can encapsulate substantial amounts of C60 and other polycyclic aromatic hydrocarbons.11 In addition to a large binding capacity, the environmentally friendly supramolecular nanocontainer should possess adequate interior space enabling sufficient mobility of the reagents, their proper orientation and, ultimately, their efficient interaction. In addition to the environmentally benign solvent, a “green chemistry” process requires low energy consumption, as well. Thus the main goal of this study is to test the ability of the hybrid supermolecules built of amphiphilic linear-dendritic copolymers to meet the abovementioned requirements and serve as nanocontainers for the DA reaction of highly hydrophobic substances (C60, anthracene and tetracene) under “green chemistry” conditions (in water and at room temperature).
Experimental Section
Supramolecular Complexes as Vessels for “Green” Chemistry 58 at λmax ) 365 nm. After the irradiation, CDCl3 was added (to 25% v/v) and the obtained solutions were analyzed by 1H NMR. In a separate experiment with similar quantities of the acenes in THF, the obtained (4 + 4) dimers were analyzed by SEC. An (4 + 4)-dimer-(TLC Rf ) 0.23). 1H NMR (CS2/CDCl3 ) 4:1) δ 7.36 (AA′BB′, 1J ) 3.3 Hz, 8H), 7.26 (AA′BB′, 1J ) 3.3 Hz, 8H), 5.88 (s, 4H); SEC (THF) retention time: 26.2 min (Retention time of pure An - 27.2 min) Tet (4 + 4)-dimer-(TLC Rf ) 0.14). 1H NMR (CS2/CDCl3 ) 4:1) δ 7.81 (AA′BB′, 1J ) 2.5 Hz, 2J ) 6.0 Hz, 2H), 7.77 (s, 2H), 7.48 (AA′BB′, 1J ) 3.3 Hz, 2J ) 6.2 Hz, 2H), 7.41 (AA′BB′, 1J ) 3.1 Hz, 2J ) 5.2 Hz, 2H), 7.28 (AA′BB′, 1J ) 3.1 Hz, 2J ) 5.8 Hz, 2H), 6.02 (s, 2H); SEC (THF) retention time: 25.8 min (Retention time of pure Tet - 26.9 min) DA Reactions of C60 and Tet in Homogeneous Phase. The procedure used here was similar to the one reported by Sarova et al.19 with the exception that the acene was not used in excess. Instead, saturated solutions of tetracene and C60 were prepared in toluene and their molarity was calculated based on UV-vis data and Beer-Lambert law. Equimolar amounts of the diene and dienophile solutions were mixed in the spectrometric cell at room temperature and were left to react inside the UV-vis spectrophotometer. Spectra were collected every 20 min for 72 h. The adduct of Tet with C60 has two characteristic, but faint peaks at 424 and 706 nm,19 which are not suitable for process monitoring at the conditions of this study. Therefore the kinetics of the reaction was followed by the disappearance of the peak of the starting tetracene at 474 nm. At the end, the reaction mixture was tested by TLC on Sorbent Technologies 200 µm Silica G TLC plates using hexane/CS2 ) 1:1 (v/v) mixture as eluent. Only the single spot of the product was observed at Rf ) 0.38. DA Reactions of C60 with An and Tet in Aqueous Media. A mixture of the respective two solid reactants was prepared in a ratio enabling their equimolar concentrations in the micelle. A typical reaction protocol used 42 mg of C60 and 1.0 mg of An in 6 mL of 10-4 M degassed micellar copolymer solution, or 43.5 mg of C60 and 2.04 mg of Tet in the same amount of micellar solution, in both cases under nitrogen. The obtained biphasic mixture was subjected to sonication in a dark water bath for 12 h. In another set of experiments involving G3PEO13k as the nanovessel building block, the reaction time was increased to 24 h and further to 48 h. The temperature of the bath was kept at 22 °C, by periodically cooling with roomtemperature water. When kinetics was followed for the Tet-C60 reaction, 3 mL of the turbid mixture was transferred to a vial at regular times and centrifuged at 500 g for 1 h. The clear layer was then placed into a UV cell and the respective UV-vis spectrum was taken. The workup procedure is depicted in Scheme 1. At the end of the reaction, the aqueous layer was evaporated at 30 °C under vacuum. The solid matter was weighed and extracted three times with 3 mL of THF. The combined THF extracts, which contain the DA adducts, unreacted acenes and the copolymer together with negligible amounts of C60 (saturated solutions contain 1.7 × 10-5 M C60), were evaporated under vacuum and the solid residue was extracted twice with 3 mL of MeOH and once with a mixture MeOH/H2O 1:1 (v/v) to separate and recover the copolymer. When Tet was involved, EtOH was used instead because of the poor solubility of Tet in MeOH. The alcohol solutions were removed after centrifugation; the remaining solids were washed with 3 mL alcohol and dried. The DA adducts were isolated in yields ranging from 46% (C60-An) to 37% (C60-Tet). Thin layer chromatography using hexane/CS2 ) 1:1 (v/v) mixture was performed on Sorbent Technologies 200 µm Silica G TLC plates, delivering two spots for each reaction pair with Rf values as follows: 0.11 and 0.32 for C60An and 0.14 and 0.38 for C60-Tet. Isolation and purification of the reaction products was achieved by eluting a CS2 solution of (19) Sarova, G. H.; Berberan-Santos, Mario N. Chem. Phys. Lett. 2004, 397, 402. (20) Murata, Y.; Kato, N.; Fujiwara, K.; Komatsu, K. J. Org. Chem. 1999, 64, 3483.
Langmuir, Vol. 24, No. 20, 2008 11433 the reaction products through a 20 cm silica gel flash column with the same CS2/hexane 1:1 eluent similarly to the procedure described in ref 20. The products of C60-An reaction were isolated in the following quantities: 0.66 mg of the product with Rf 0.34 and 0.68 mg of product with Rf 0.11. This separation procedure yielded 1.4 mg of the C60-Tet product with Rf 0.38 and 1.03 mg of the product with Rf 0.14 after washing the column with pure CS2. The structures and the molecular sizes of all products isolated were elucidated by 1H NMR, UV-vis spectroscopy and SEC in THF. C60-An adduct 1 (TLC Rf ) 0.32). 1H NMR (CS2/CDCl3 ) 4:1) δ 7.77 (AA′BB′, 1J ) 3 Hz, 4H), 7.47 (AA′BB′, 1J ) 3 Hz, 4H), 6.02 (s, 2H); UV-vis (CS2) λ1max ) 439 nm, λ2max ) 713 nm; SEC (THF) Retention time: 24.4 min. C60-An adduct 2 (TLC Rf ) 0.11). 1H NMR (CS2/CDCl3 ) 4:1) δ 7.71 (AA′BB′, 1J ) 3 Hz, 4H), 7.44 (AA′BB′, 1J ) 3 Hz, 4H), 5.74 (s, 2H); UV-vis (CS2) λ1max )503 nm, λ2max ) 729 nm; SEC (THF) retention time: 23.6 min. C60-Tet adduct 1 (Rf ) 0.38). 1H NMR (CS2/CDCl3 ) 4:1) δ 8.17 (s, 2H), 7.92 (AA′BB′, 1J ) 3.3 Hz, 2J ) 6.2 Hz, 2H), 7.77 (AA′BB′, 1J ) 4.5 Hz, 2J ) 7.7 Hz, 2H), 7.51 (AA′BB′, 1J ) 3.3 Hz, 2J ) 6.2 Hz, 2H), 7.47 (AA′BB′, 1J ) 1.9 Hz, 2J ) 5.4 Hz, 2H), 5.87 (s, 2H); UV-vis (CS2) λ1max ) 436 nm, λ2max ) 703 nm; SEC (THF) retention time: 24.5 min. C60-Tet product 2 (Rf ) 0.14). 1H NMR (CS2/CDCl3 ) 4:1) δ 7.81 (AA′BB′, 1J ) 2.5 Hz, 2J ) 6.0 Hz, 2H), 7.77 (s, 2H), 7.48 (AA′BB′, 1J ) 3.3 Hz, 2J ) 6.2 Hz, 2H), 7.41 (AA′BB′, 1J ) 3.1 Hz, 2J ) 5.2 Hz, 2H), 7.28 (AA′BB′, 1J ) 3.1 Hz, 2J ) 5.8 Hz, 2H), 6.02 (s, 2H). SEC (THF) retention time: 25.8 min.
Results and Discussion Any supramolecular structure considered as nanovessel for a chemical reaction should meet several criteria: (a) it should be formed through the self-assembly process at low cmc as a single population and should retain its size during the entire reaction process; (b) it should possess high affinity toward the reagents involved and bind amounts, which would be sufficient to trigger and maintain a continuous reaction flow, and (c) it should have relatively low core microviscosity to facilitate the proper orientation of the reagents and their efficient interaction. Ideally the reaction should yield only the desired product, which should be isolated by simple laboratory procedures (precipitation, filtration and/or open column chromatography). The current exploration was performed along these requirements and directions. Stability and Binding Studies. Preliminary QualitatiVe Comparison between Various Amphiphiles. Our previous studies indicated that some of the linear-dendritic block copolymers (LDBC), containing PEG segments, were able to form monodisperse micelles in aqueous medium, which showed considerable binding capabilities for hydrophobic substances.11-13 In this investigation the binding abilities of the LDBC are compared with those of common amphiphiles (low-molecular weight or polymeric) under identical conditions. Standard micellar solutions of each of them are saturated by 4 h sonication. The binding of the probes and the changes in the micellar size distribution are investigated by aqueous SEC with UV PDA detection, Figure 1. Despite its qualitative character, this analysis enables the evaluation of the stability of the micelles over extended periods of sonication and confirms that the binding of the substrates occurs within the micelles studied. Table 1 summarizes the peak data for all materials initially tested. The nonionic surfactants of the Igepal and Triton series are not stable upon sonication as evidenced by the appearance of lower molecular weight species in their SEC traces (Figure 1 A). As a result their binding ability is low (the loaded/unloaded signal ratio in the chromatogram is less than 1, Figure 1 A, Table 1). Therefore these surfactants are
11434 Langmuir, Vol. 24, No. 20, 2008 Scheme 1. Sequence of Actions in the Work-up Procedurea
a 1: Reaction vessel charged with An or Tet, C60 and copolymer, i. Add DI H2O, sonicate 24-48 h; 2: Micellar suspension of products, unreacted reagents, ii. Vacuum evaporate H2O, 30°C; 3. Solid mixture of products, unreacted reagents and copolymer, iii. Extract 3 times with 3 mL THF; remove THF layer, recycle leftover C60 back to step 1; 4. THF solution of products, acene and copolymer, iv. Vacuum evaporate THF, 30°C; 5. Solid products, acene and copolymer, v. Extract three times with MeOH or EtOH, remove alcohol layers, evaporate them, recycle leftover acene and copolymer back to step 1; wash and dry undissolved DA adducts; 6. Solid adduct(s) of C60 and acene.
deemed of unacceptable stability as nanocontainers for the DA reactions in this study and are used as reference materials only. Peak broadening toward higher elution volumes (smaller micellar sizes) is observed with PSt-PEO, as well, Figure 1 C. The substrates reside within the eluting micellar species as evidenced by the UV-vis spectra taken across the corresponding micellar peak in the aqueous SEC. An example with G3PEO13k built micelles is shown in Figure 2. The 2-D chromatograms unequivocally confirm the encapsulation of the substrates across the entire population of micelles. Interestingly, in the spectrum of the Tet-loaded micelles a prominent bimodal peak appears at 520 nm (Figure 2), which differs from the characteristic Tet absorption maxima that are usually seen in the toluene solution, Figure 3. It was shown that tetracene22 and anthracene23 reversibly form aggregates from frozen iso-octane/methylcyclohexane solutions. Our previous investigation11 confirmed that the cores of the micelles built of dG3 and dG2 have polarities that closely resemble those nonpolar solvents. Katul et al.22 discovered that unlike An, the Tet dimer spectrum is very similar to the spectrum of the Tet crystal with absorption maximum located near 520 nm. The observed similarity suggests that the Tet molecules are able to form dimers in the core of the LDBC micelles due most probably to their preferred accumulation in the larger voids between the monodendrons. As for An, no changes were observed in the UV-vis spectra of the An-loaded micelles, indicating that they might be predominantly compartmentalized in the “clefts” of the assembled monodendrons. The assumed placement of the two acenes is shown in Figure 4. In a similar fashion, the increased absorbance in the 400-540 nm region of the encapsulated C60 (Figures 2 and 3) suggests that this molecule is also predominantly bound between the monodendrons in the core, a positioning also supported by the mechanistic investigations of the same complex in our previous publication.11 In a recent elegant study Gauthier et al. have found similar dependencies for hydrophobic substrates bound in unimolecular micelles.24 (21) Malvern Instruments, Ltd.; Dynamic Light Scattering - Application Note No. MRK809-01 available at http://www.malvern.co.uk/common/downloads/ campaign/MRK809-01.pdf. (22) Katul, J.; Zahlan, A. J. Chem. Phys. 1967, 47, 1012. (23) Chandross, E; Ferguson, E.; McRae, E. J. Chem. Phys. 1966, 45, 3546.
Simonyan and GitsoV
QuantitatiVe Binding Study. Using the results in the previous section, the binding characteristics of five micellar systems are investigated in more detailsthe three linear-dendritic copolymers, their linear analogue PSt-PEO, and the nonionic surfactant IP720 for comparison. The first series of experiments investigate the influence of the solid probe initial concentration on the amount of substrate bound for a constant time, while the second set of experiments follows the kinetics of Tet binding at constant concentration of the solid added. To evaluate the mass pressure effect on solubilization a set of four initial load amounts, A0, of Tet, An or C60 are added to four identical micellar solutions of each of the five amphiphiles. Each mixture is sonicated and treated under identical conditions before the corresponding UV-vis spectra are taken. Therewith the concentration (in g/L) of the encapsulated material, Cin, is calculated and is found to be directly proportional to A0 (also expressed in g/L). The relationships Cin ) f(A0) for three representative types of amphiphiles and Tet as the encapsulated probe are shown in Figure 5. Similar diagrams are constructed with C60 and An. The slopes in these graphs could be regarded as indicators of the binding affinity of the respective micellar system toward the given probe. Therefore the dimensionless slope parameter could be used as a characteristic feature for each amphiphile-probe pair. All measured values of the binding affinity are shown in Table 2. Second, the kinetics of Tet binding for three representative micellar systems is comparatively evaluated, namely for the linearlinear copolymer PSt-PEO, the linear-dendritic G3PEO13k and the nonionic surfactant IP720. Using the extinction coefficient of tetracene at 474 nm, the values of bound substrate Cin are calculated and the respective Cin ) f(time) diagrams are built, Figure 6. The binding process could be described with the following simple equation: kb
tetracene(solid) f tetracene(bound in micelle) where kb is the kinetic constant of binding. It can be found from the increase in bound substrate with time:
dCin ) kb dt
(1)
The solution of eq 1 assuming Cin ) 0 at time t ) 0, can be derived trivially by integration and yields
Cin ) kbt
(2)
The slopes of the linear fits shown in Figure 6 for the initial stages of the encapsulation yield the kinetic constants of the respective binding process and could be regarded as descriptors of each pair. Their values are 5.0 × 10-7 M/min for G3PEO13k, 4.0 × 10-7 M/min for PSt-PEO and 1.5 × 10-7 M/min for IP720. As it can be seen, however, later in the binding process, saturation leads to formation of a plateau with different maximum Cin for the three amphiphiles, which is a characteristic for their binding abilities. In this set of experiments, the highest Cin(max) (calculated as the mean average of the values beyond the linear fit) is detected for G3PEO13k: 1.01 × 10-4 M, then for PSt-PEO with Cin(max) )4.01 × 10-5 M, and finally for IP720 with Cin(max) )1.5 × 10-5 M. It should be noted though that IP720 shows a decrease in Cin of about 10% after the second hour in accord with the previously mentioned micelle dissociation (Figure 1 A). (24) Njikang, G.; Gauthier, M.; Li, J. Polymer 2008, 49.
Supramolecular Complexes as Vessels for “Green” Chemistry
Langmuir, Vol. 24, No. 20, 2008 11435
Figure 1. Aqueous SEC eluograms of micelles formed by self-assembly of (A) nonionic surfactant IP720; (B) linear-dendritic block copolymer G3PEO13k and (C) linear-linear copolymer PSt-PEO. 1, empty micelle; 2, micelle loaded with Tet. UV PDA detector trace at 254 nm. Table 1. Aqueous SEC Data for All Amphiphiles in This Study amphiphile
concn, M
elution time, min
Nagg/Rg (nm)
area ratiob C60/empty
area ratioc Tet/empty
G3PEG11kG3 G2PEG5kG2 G3PEO13k TX100 IP720 PSt-PEO
1 × 10-4 1 × 10-4 1 × 10-4 1 × 10-2 1 × 10-2 1 × 10-4
16.24 16.35 16.63 16.11 16.11 16.12
953/4411 547/3211 138/5813 121/3.221 a 54/2.117 a,b,c N/A
1.14 1.91 2.43 0.51 0.44 1.22
3.74 6.23 3.62 0.84 0.39 1.55
a The values shown correspond to the hydrodynamic volume RH rather than Rg. b Area of elution peak of C60-loaded micelle recorded at 254 nm vs area of elution peak of the same micelle before binding. c Area of elution peak of Tet-loaded micelle recorded at 254 nm vs area of elution peak of the same micelle before binding.
At this stage it is established that the LDBC and their linearlinear analogue PSt-PEO are able to form micelles of relatively uniform size, which are stable over extended sonication periods and are able to bind substantial amounts of the DA reagents
chosen. Their ability to serve as reaction vessels is estimated in the second stage of this study. Fluorescence Studies. The mobility of a substrate within the micellar core is influenced by the core microviscosity and
11436 Langmuir, Vol. 24, No. 20, 2008
Simonyan and GitsoV
Figure 2. UV-vis spectra of empty and loaded supramolecular G3PEO13k aggregates eluting at 16.6 min in aqueous SEC (Figure 1.B). UV PDA detection at identical concentrations. Inset, 2-D chromatogram of Tet-loaded micelle.
Figure 3. UV-vis spectra of Tet and C60 in toluene.
ultimately affects the ability of the compound to participate in core-bound reactions. The suitability of a particular polymeric material as a nanocontainer building block will therefore be judged by its potentially high binding ability and low microviscosity of the core. Pyrene encapsulation and fluorescence decay patterns have long been employed for estimation of the latter parameter for various amphiphilic materials.15 This technique is also used here to provide an additional insight on the nanoenvironment in the supermolecules studied. The calculated decay times, , of the investigated materials and comparative literature values are shown in Table 3.
As reported earlier by Winnik et al.15and Duhamel et al.,14 the typical linear-linear copolymers form in aqueous solution supramolecular structures with densely packed cores showing increased Py lifetimes due to the restricted mobility of the probe therein. Therefore, it is assumable that in the case of associating linear-dendritic copolymers, the value of will depend on the number and size/generation of monodendrons in the core. From previous studies it is known that the aggregation numbers (Nagg) of the micelles of ABA dendritic-linear-dendritic copolymers are 953 for dG3 and 547 for dG2,11 and 138 for the linear(25) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039.
Supramolecular Complexes as Vessels for “Green” Chemistry
Langmuir, Vol. 24, No. 20, 2008 11437
Figure 4. Schematic presentation of the probable accommodation sites for An and Tet within the micellar core. Table 2. Binding Affinity Values for Linear-Dendritic, Linear-Linear and Low-Molecular-Weight Amphiphiles binding affinitya/no. of bound substrate molecules per micelleb amphiphile G3PEO13k G3PEG11kG3 G2PEG5kG2 PSt-PEO IP720
anthracene 0.0271/775 0.028/1499 4.2 × 10-3/248 8.8 × 10-3/n/a 9.0 × 10-4/∼0.1-1
tetracene
C60
0.0161/174 9 × 10-3/751 3.5 × 10-3/84 6.1 × 10-3/n/a 4.0 × 10-4/∼ 0.1-1
1.3 × 10-3/10 5.9 × 10-3/78 5.9 × 10-4/11 1.3 × 10-3/n/a 5.2 × 10-4/∼0.05
a As determined from the slope of the linear fit of Cin ) f(A0), shown in Figure 5. b As determined from C(amphiphile) and Nagg shown in Table 1, and the maximum C(bound substrate).
dendritic type (Table 1).13 The determined values here indicate that despite the large aggregation number dG2 has a loose core probably due to the inherently much lower number of benzyl ether moieties per G2 dendron. However, its binding affinity values from Table 2 are also lowest among the lineardendritic members, suggesting that the generation number is crucial in core architecture and binding capabilities. The increase in generation and Nagg, contributes to the highest microviscosity observed with the cores of dG3 micelles. The value for G3PEO13k comes in between the ABA copolymers, due to the lower aggregation number of its micelles and the presence of only one G3 block per copolymer molecule. In comparison, the linearlinear PSt-PEO has value, which is higher than any lineardendritic material. Although the aggregation number for this species is not known, the distinct increase in core microviscosity can be assumed to arise from the linear nature of the hydrophobic block and its ability to pack more densely. The well-known detergent SDS is reported to have even higher pyrene lifetime
values than PSt-PEO probably due to the small volume and flexibility of the dodecyl hydrocarbon tails enabling their compact packing.25 The poly(styrene)-poly(maleic anhydride) copolymers, reported in ref 14, were able to pack extremely densely, not allowing transport of encapsulated probe. The overall conclusion that can be drawn from both binding and fluorescence studies is that while the dG3 copolymer offers high affinity to probe molecules, it has a relatively dense core and thus less space and mobility for the DA reagents. The pyrene value of the dG2 copolymer is the lowest measured, but the affinity parameters are also much lower than those of the other linear-dendritic copolymers. The highest binding affinity was shown by the G3PEO13k micelles, which also have lowmicroviscosity core thus enabling the needed mobility for the reagents. Diels-Alder Reactions. Kinetics. The ability of the best micellar candidates to mediate a DA reaction in aqueous biphasic solution is investigated in this study. Binding and core flexibility
11438 Langmuir, Vol. 24, No. 20, 2008
Simonyan and GitsoV
Figure 5. Dependence of Tet amount bound in the micelle, Tet (Cin), on the amount of solid Tet added to the system, A0. The solid lines are linear fits. A0 ) 0.25-1.0 mg/mL for G3PEO13k (9), PSt-PEO (2) and IP720 (•).
Figure 6. Kinetic profile of Tet binding to micelles of G3PEO13k (9), PSt-PEO (2) and IP720 (•) in aqueous medium. The solid lines are linear fits. A0 (Tet) is 1.5 mg/mL in all systems. Table 3. Quality of the Exponential Fit (χ2) and Lifetimes () Quality of Monomeric Py in Different Micelles sample
χ2
, ns
G3PEO13K PSt-PEO dG3 dG2 SDSa SMA 3-1.0b
1.20 1.06 1.21 1.08 n/a 1.0914
114 152 121 95 45025 31514
a SDS,sodiumdodecylsulfate. b SMA,copolymerpoly(styrene)-poly(maleic anhydride).14
data from the previous three sections suggest that the lineardendritic copolymer G3PEO13k most closely fulfills the requirements for a nanovessel building block outlined in Section 1. It offers high binding capability for the chosen reagent molecules and possesses rather loose core, which does not hamper their mobility. At the same time its supermolecules are stable enough
to hold the reagents tightly and prevent disassembly or uncontrolled migration of materials in and out of them. The other two linear-dendritic copolymers are also superior in stability to the low molecular weight surfactants and exhibit higher binding affinity and binding kinetic constants toward the hydrophobic probes. The linear PSt-PEO shows similarly favorable binding parameters although the relatively high density of its core could overall hamper the proceeding of the model reactions, due to limited mobility. In this section kinetic results from a DA reaction of C60 with Tet are reported for G3PEO13k, and PSt-PEO. The DA monoadducts were previously characterized7,8,19 and they do not exhibit any characteristic peaks in their UV-vis spectra that could be used effectively for reaction monitoring in our systems. Both adducts show a very faint broad peak (ε ) 100 M-1 · cm-1) at 703-715 nm, which is fairly difficult to be quantitatively measured in the rather diluted micellar solutions. Another adduct-
Supramolecular Complexes as Vessels for “Green” Chemistry
Langmuir, Vol. 24, No. 20, 2008 11439 Scheme 2. Mass Transfer in and out of the Micellar Nanoreactor Entitya
Figure 7. Second-order kinetic profile for the DA reaction between Tet and C60 in aqueous micellar solutions of G3PEO13k (9) and PSt-PEO (2).The solid lines are linear fits. [Polymer] ) 1 × 10-4 M; room temperature.
related peak is positioned at 436 nm, but in this region it overlaps with major peaks from the starting materials and is difficult to be quantified, as well. To monitor C60-Tet DA addition, the disappearance of the 474 nm characteristic peak of tetracene in the Vis spectrum, is followed. Initially, saturated Tet and C60 solutions are prepared in toluene with the respective concentrations being [Tet] ) 2.26 × 10-4 M and [C60] ) 2.40 × 10-4 M. Equimolar amounts of both solutions are mixed in a UV cuvette and UV-vis spectra are taken every 30 min for 2 days. The kinetic profiles at two temperatures can be found in the Supporting Information. At room temperature (22 °C) a value of 1.82 M-1 · min-1 is obtained for the second order rate constant, which increases to 3.90 M-1 · min-1 at 31 °C. Previously reported rates obtained by different monitoring method19 are in close agreement. In the next step C60-Tet DA reactions are performed in G3PEO13k aqueous solution and in PSt-PEO solution of the same concentration for comparison. Monitoring the disappearance of the Tet 474 nm peak by UV-vis spectrometry affords respectively apparent kinetic constants of 208.2 M-1 · min-1 for G3PEO13k and 144.9 M-1 · min-1 for PSt-PEO, as seen on their kinetic profiles in Figure 7. The comparison of the processes performed in organic solvent and in water shows that the DA reaction in the nanocontainers proceeds between 80 (PSt-PEO) and 114 (G3PEO13k) times faster. While this rate enhancement has been noted before, the differences recorded in this study are markedly higher than the previously reported 15-25-fold acceleration of similar DA syntheses induced by a change from organic solvents to SDS micelles in water.26 The commonly encountered retardation of DA reactions in water upon addition of micelles27 is not taking place at all, since the reaction of C60 with both acenes does not proceed in pure water. The observed phenomena could be explained by the superior stability and binding capability of the micelles formed by the linear-dendritic copolymers and the relatively high degree of freedom experienced by the encapsulated reagents in their cores. These specific features positively enhance the “micellar” effect of these systems on the apparent rate constants of the DA reaction. (26) (a) Braun, R.; Schuster, F.; Sauer, J. Tetrahedron Lett. 1986, 27, 1285. (b) van der Wel, G. K.; Wijnen, J. W.; Engberts, J.B.F.N. J. Org. Chem. 1996, 61, 9001. (27) (a) Breslow, R.; Maitra, U.; Redeout, D. Tetrahedron Lett. 1983, 24, 1901. (b) Sangwan, N. K.; Schneider, H. J. J. Chem. Soc., Perkin Trans. 2 1989, 1223. (c) Hunt, I.; Johnson, D. C. J. Chem. Soc., Perkin Trans. 2 1991, 1051.
a Note that when tetracene is involved the DA reaction is effectively irreversible.
Isolation and Characterization of DA Adduct. Considering the previously evaluated binding and kinetic parameters the micelles, constructed by G3PEO13k seem to be the most suitable nanocontainers for the practical synthesis of fullerene DA adducts. In a parallel set of 12 h experiments with both acenes the solid material is separated by centrifugation from the aqueous solution. After drying in vacuum and dissolution in 1 mL CS2 (containing 20% CDCl3) the 1H NMR of this solid precipitate clearly reveals the presence of reaction products besides unreacted dienes. The fact that a reaction product could be found outside the micelles is somewhat surprising since the large hydrophobic species are expected to remain encapsulated in the linear-dendritic nanocontainers even after the reaction is completed. This phenomenon can only be accounted for by diffusion mass transfer through the micellar corona. On one side, there is a large amount of reagent moleculessC60 and the respective acene outside the micelles. They exert mass pressure and certain amount of them migrates into the micellar cores obeying the binding affinity parameter and binding kinetic constant discussed previously. On the other side, it should be expected that the mass pressure exerted by the solid unreacted material outside the micelle will displace some of its contents out. Consequently there must be a partitioning competition between the reagents and the products outside the micelle (i.e., for re-entry). The ones of high affinities will partition more easily and the concentration of the low binding substance(s) in the solid matter outside of the nanocontainers would increase. It can be seen from Table 2 that
11440 Langmuir, Vol. 24, No. 20, 2008
Figure 8. (a) 1H NMR spectra in CS2/CDCl3 4:1 of (A) pure anthracene, (B) products of DA reaction between An and C60 in G3PEO13k micelles after 24 h, (C) products of the same reaction after 48 h, and (D) anthracene irradiated for 1 h by a Multiband Mineralight lamp, model UVGL-58 at 365 nm. (b) 1H NMR spectra in CS2/CDCl3 4:1 of chromatographically separated (A) C60-anthracene mono-adduct and (B) C60-anthracene D2h symmetric trans-1-bis-adduct.
there is an inverse proportionality between binding affinity and molecular size. It could be qualitatively assumed that for molecules of similar polarity the binding affinity will decrease with the size of the molecule. Due to their large size, the DA adducts would be expected to have smaller affinity as compared to their parent molecules and hence smaller driving force for re-entry. Their displacement into the extra-micellar medium will lead to their precipitation due to their negligible solubility in water. Similar picture is observed when Tet is used as the diene. This reasoning can certainly be extended to higher adducts, as well. The processes presumably taking place in the reaction nanovessel are schematically shown in Scheme 2 with An as the DA diene. The conversions achieved in the linear-dendritic nanovessels after 12-, 24- and 48-h periods are ∼10%, 49% and 45% for the C60/An pair and ∼3%, 37% and 55% for the C60/Tet pair. The results at shorter reaction times seem to contradict the higher apparent rate constant of the DA reaction with tetracene, but are most probably affected by the lower diffusion rate of Tet into the linear-dendritic supermolecules and their lower binding capability toward this reagent. A typical workup involves evaporation of the water under vacuum and then extraction of the remaining solids with THF, where C60 is only negligibly soluble. The combined THF extracts,
Simonyan and GitsoV
Figure 9. (a) 1H NMR spectra in CS2/CDCl3 4:1 of (A) pure Tet, (B) products of DA reaction between Tet and C60 in G3PEO13k micelles after 24 h, (C) products of the same reaction after 48 h, and (D) Tet irradiated for 1 h with a Multiband Mineralight lamp, model UVGL-58, at 365 nm. (b) 1H NMR spectra in CS2/CDCl3 4:1 of (A) chromatographically pure C60-tetracene mono-adduct and (B) soluble tetracene [4 + 4] dimer.
which should contain unreacted acenes, the linear-dendritic copolymer and the eventual reaction products, are evaporated and the solids obtained are further extracted once with methanol for An/C60 or ethanol for Tet/C60, and again with 50% aqueous solutions of the same alcohols to fully remove any remaining copolymer. The alcohols dissolve well any unreacted acene and the copolymer, but not the respective DA adducts. Hence the residue should be composed of DA adducts uncontaminated by starting materials or nanocontainer building copolymer. Indeed, TLC confirms the removal of the starting reagents and reveals the presence of two spots for both reactions pairs. Along with the spots of the presumed mono-adducts (Rf values are 0.32 and 0.38 for C60-An and C60-Tet, respectively) second spots with much lower Rf are observed in both TLC plates (Rf ) 0.11 for C60-An and Rf ) 0.14 for C60-Tet). Subsequent flash chromatography enables the separation and yields roughly equal amounts of the C60-An reaction products. The C60-Tet compound with the higher Rf. is isolated in 1.4 mg yield, while the product with the lower Rf requires a wash-out with pure CS2 to afford 1.03 mg. The Vis spectra of the adducts with the higher Rf values from both reactions contain peaks at 439/713 nm (C60-An) and 436/ 703 nm, which are consistent with the previously reported
Supramolecular Complexes as Vessels for “Green” Chemistry
absorbencies of the corresponding mono-adducts.19,28 The 1H NMR spectra show the presence of peaks at 6.02 ppm (C60-An) and 5.85 ppm (C60-Tet) along with the typical AA′BB′ signals, that correlate well with the spectra of the corresponding fullerene mono-adducts of the two acenes, Figures 8a,b and 9a,b.7,20,28 When analyzed by SEC in THF the C60-An compound with the lower Rf value elutes 0.8 min before the mono-adduct, an indication for species with larger hydrodynamic volume. The vis spectrum in CS2 has a notable peak at 498 nm and a weak broad peak at 724 nm, both differing significantly from the corresponding 439 and 713 nm peaks of the mono-adduct. The signals in the 1H NMR spectrum are grouped similarly, but are shifted upfield, most noticeably for the bridge protons -5.74 vs 6.02 ppm, Figure 8b. They also differ from the spectral characteristics of the products (An dimer) obtained after UV irradiation, Figure 8a, D. The results obtained indicate that the compound with the lower Rf value is most probably a bis-adduct. The Vis and NMR spectra resemble closely the spectral characteristics of the symmetrical C60-An antipodal trans-1 bis-adduct, initially reported and characterized by Kra¨utler et al.29 The amount of this bisadduct increases with time, reaching ∼22.5% yield after 48 h as evidenced by the 1H NMR spectra of the product mixtures, Figure 8a, B and C. Notably, the characterized bis-adduct forms in measurable quantities within the linear-dendritic nanocontainers at room temperature and in 48 h, in contrast to the previously used conditions for the synthesis: 240 °C/1 h/∼44% yield29 or room temperature/26 days/1.3% yield.30 The observed regioselectivity of the second intramicellar DA reaction, which leads exclusively to the corresponding symmetrical antipodal bis-adducts, could probably be attributed to a favorable arrangement of the An molecules. Completely random orientation of the entrapped reagent molecules would conversely produce bis-adducts as a mixture of positional isomers as it was seen in the solution reaction.30 With Tet as the diene, the DA reaction yields increasing amounts of a mono-adduct, but the compound with Rf ) 0.14 (28) Kra¨utler, B.; Muller, T; Duarte-Ruiz, A. Chem.-Eur. J. 2001, 7, 3223. (29) Duarte-Ruiz, A.; Wurst, K.; Kra¨utler, B. HelV. Chim. Acta 2001, 84, 2167. (30) Duarte-Ruiz, A.; Mu¨ller, T.; Wurst, K.; Kra¨utler, B. Tetrahedron 2001, 57, 3709.
Langmuir, Vol. 24, No. 20, 2008 11441
seems to be the major reaction product after 48 h, Figure 9a. The logical assumption that this is also a bis-adduct in analogy to the C60-An reaction is disproved by a blank experiment, where the irradiation of pure tetracene at 365 nm in toluene solution yields a compound with identical Rf and spectral footprint, Figure 9a, D and B. Thus, the low Rf product could be identified as Tet dimer, whose formation is probably due to the suggested pairlike stacking of the acene molecules in the micellar core (Figure 4). The seemingly higher integral intensity of the benzylic protons in this compound (Figure 9a) is therefore caused not by the increased yield, but by their relatively higher content in the molecule.
Conclusions The results obtained show that in aqueous medium lineardendritic copolymers form supramolecular assemblies with interiors of relatively lower microviscosity, higher stability and higher binding capability than similar linear-linear copolymers and comparable nonionic surfactants. These favorable features enable for the first time the formation of C60 adducts with anthracene and tetracene under typical “green chemistry” conditionssin water and at room temperature. Other useful advantages of the linear-dendritic nanocontainers is their regioselective effect leading to the exclusive formation of symmetrical antipodal trans-1-bis-adducts and the substantial increase in the rate constants of the corresponding DA processes. Acknowledgment. We thank Dr. Jean Duhamel (University of Waterloo, Ontario, Canada) for his assistance in the dynamic fluorescence study and Dr. Johannes Smid of SUNY ESF for helpful discussions. Thanks are also due to Michael Brondon (Le Moyne College) for his participation in the initial binding measurements. Partial funding of this research, provided by the National Scientific Foundation (CHE-0243959 and MCB0315663) is acknowledged with thanks. Supporting Information Available: Aqueous SEC profiles of empty and substrate-loaded micelles; binding affinity graphs; dynamic fluorescence decay, kinetic experiment graphs of the DA reaction in organic solvent and 1H NMR spectra of reaction mixtures. This material is available free of charge via the Internet at http://pubs.acs.org. LA801593Y
