HETEROCYCLES

Short Paper | Regular issue | Vol. 83, No. 7, 2011, pp. 1615-1620
Received, 10th February, 2011, Accepted, 11th April, 2011, Published online, 19th April, 2011.
DOI: 10.3987/COM-11-12170

■ Reaction of [60]Fullerene with Epoxides under Photo-irradiation: Synthesis of C60-Fused Tetrahydrofuran Derivatives

Chun-Bao Miao,* Zong-Yong Tian, Xiao-Jiao Ruan, Xiao-Qiang Sun, and Hai-Tao Yang*

School of Petrochemical Engineering, Changzhou University, Changzhou 213164, China

Abstract

The photoreactions of [60]fullerene (C60) with various epoxides have been investigated. Trans 2-aroyl-3-aryloxiranes reacted with [60]fullerene giving both the major cis products and the minor trans adducts. The influence of substituent group on aryl ring on the photoreaction is entirely different from that of preliminary reported thermal reaction. In addition, the photo condition could avoid the decomposition of some original epoxides compared to the thermal condition at high temperature. It provides a simple and efficient route to synthesize C60-fused tetrahydrofuran derivatives.

Fullerene derivatives possess potential applications in material science and medicinal chemistry.1 The 1,3-dipolar cycloaddition reaction is one of the most useful methodologies for the functionalization of fullerenes. Various 1,3-dipoles including azomethine ylides, diazo compounds, azides, nitrile oxides, nitrile ylides, nitrile imine, pyrazolinium ylides have been reported to react with fullerenes.2 Epoxides undergoing thermal or photochemical 1,3-dipolar cycloaddition via carbonyl ylides with alkenes,3 alkynes,4 benzylidene anilines,5 thioketones,6 phosphaalkynes7 have been well explored. However, carbonyl ylides as 1, 3-dipoles applied to the functionalization of C60 are seldom investigated. Jagerovic first reported the TCNE oxide serving as a carbonyl ylide to react with C60.8 Nair and coworkers have shown that cyclic carbonyl ylides generated in situ from diazo ketones in the presence of Rh2(OAc)4 undergo facile dipolar cycloaddition with C60 affording novel organofullerenes.9 In a preliminary report two of us have investigated the reaction of [60]fullerene (C60) with the carbonyl ylides generated in situ from trans epoxides under thermal condition.10 The thermal reaction of C60 with trans epoxides afforded exclusively or predominantly the cis isomers of C60-fused tetrahydrofuran derivatives containing functional groups such as ketone, ester and cyano groups. According to Woodward-Hoffmann rules for electrocyclic ring opening in four-electron systems, C-C bond cleavage of the trans oxiranes under thermal or photo condition will give different 1,3-dipoles which may result in the difference in the isomeric distribution of addition product.11,12 We here present the reaction of [60]fullerene with trans epoxides under photo-irradiation.

We firstly investigated the photoreaction of C60 with typical trans oxiranes 1a-h, which were prepared according to the reported procedure11 (Scheme 1). The mixture of C60 with trans oxiranes 1a-h in toluene were photoirradiated with 250 W high-voltage mercury lamp at 12 oC, respectively. The reaction time and yields were outlined in Table 1. Theoretically, trans product 3 would be the only product because photochemical disrotatory opening of the trans epoxides leads to a trans carbonyl ylide whose geometry would be preserved after cycloaddition with C60. Surprisingly, for substrates 1a-d and 1f-g, not only trans products 3a-d and 3f-g but also unexpected cis products 2a-d and 2f-g were afforded while in the thermal reaction only the cis products could be obtained. Moreover, the cis isomers were the major products. It could be explained by that the C-O bond rotation of trans carbonyl ylide would lead to cis carbonyl ylide, which reacted with C60 to give more stable cis product because the aryl and benzoyl groups are both in the equatorial position. The results were entirely different from our forecast. The isomeric distribution of 2a-d, 2f-g and 3a-d, 3f-g was seldom affected by the electronic property of the substituent on the phenyl ring. However, the reaction rate was dramatically affected by the substituents R1 and R2 on the phenyl ring. R1 had great influence on the reaction rate whereas R2 had significant influence on the yield. When R2 was H, whether R1 was weak electron-withdrawing or electron-donating group the reaction gave satisfactory yields. Electron-donating group R1 on the aryl ring such as OMe accelerated the reaction greatly (Table 1, entry 1). On the contrary, no product was observed under thermal reaction when R1 was OMe due to the decomposition of corresponding epoxide. It should be noted that no reaction occurred when R1 or R2 was strong withdrawing group NO2 even after prolonging the reaction time to 48 h (Table

1, entry 5, 8) whereas the same reaction proceeded well when R1 was NO2 under thermal condition. When R2 on the aryl ring changed from H to OMe or Cl the yields decreased notably (Table 1, entry 6, 7).

The structures of 2a, 2d, 2f, 2g, 3a-d, 3f and 3g were fully identified by their MS, 1H NMR, 13C NMR, FT-IR and UV-vis spectra. The cis stereochemistry of 2 was substantiated by the NOESY spectrum.10 The 1H NMR spectral patterns of isomeric 3 and 2 were very similar, the only noticeable difference was a downfield shift of 0.68-0.88 ppm for the two methine hydrogen of 3 (3a, 3d, 3f, 3f vs. 2a, 2d, 2f, 2g) in the 1H NMR spectrum. For example, 2a and 3a allowed the assignment of all chemical shifts in their 1H NMR and 13C NMR spectra in CS2-CDCl3, as shown in Figure 1. Hydrogen bond must be existed between the Ha and carbonyl oxygen, which caused the downfield shifts13 of 0.87 ppm for Ha and 3.19 ppm for the carbonyl group in the 1H NMR spectrum and 13C NMR spectrum of trans isomers relative to the cis isomers, respectively. The observed red shift of 6 cm-1 for the absorption of the carbonyl group in compounds 3a relative to 2a (1684 cm-1 for 3a vs 1690 cm-1 for 2a) also proved the existence of hydrogen bond. The downfield shifts of 0.68 ppm for Hb in the 1H NMR spectrum may be induced by the change of Hb from axial position to equatorial position. Thus, adducts 2a-d are cis isomers, and compounds 3a-d have trans structures. Similar phenomena for the 1H and 13C NMR downfield shifts was also observed in previous report.10,14

In order to study the application universality of the photoreaction we next examined many other substrates 4-9 (Figure 2). We firstly took the trans-2-cyano-2-ethoxycarbonyl-3-phenyl oxirane 4 as a try. Although oxirane 4 reacted well with C60 under thermal condition, no reaction was observed under photo-irradiation. It was conjectured that the strong electron-withdrawing group (CN) may be an unfavourable factor. Therefore, 5 was chosen to react with C60 under photo condition. To our disappointment, 5 also did not react with C60. Furthermore, we investigated the photoreaction of oxiranes 6-9 with C60. No any desired cycloaddition reactions occurred yet. From these results, it was concluded that the two aryl rings and carbonyl group in the structure of 1 are essential to the photoreaction.

In summary, the photoreaction of C60 with trans epoxides to give C60-fused tetrahydrofuran derivatives has been investigated. Not only the major cis products but also the minor trans products could be obtained from the photoreaction while only cis product could be obtained in the preliminary reported thermal reaction. The isomer distribution was seldom affected by the substituent on the phenyl ring. The influence of substituent on phenyl ring on the photoreaction was entirely different from the thermal reaction. In addition, the photo condition could avoid the decomposition of some original epoxides compared to the thermal condition at high temperature. It provided a simple and efficient methodology to synthesize C60-fused tetrahydrofuran derivatives.

EXPERIMENTAL
General procedure for the reaction of C60 with trans 2-benzoyl-3-aryl oxiranes 1a-d. A mixture of C60 (54.0 mg, 0.075 mmol) and 1a-1d, (0.3 mmol) was dissolved in toluene (60 mL) in a big glass tube (25×250 mm), then the tube was immersed in flowing water and photo-irradiated with 250 W high-voltage mercury lamp at 12 oC for the designated time. The solvent was then evaporated in vacuo, and the residue was separated on a silica gel column eluted with CS2 or CS2-toluene mixture to afford unreacted C60, 2a-d and 3a-d. The spectra data and spectra could be seen in supporting information.

ACKNOWLEDGEMENTS
The authors are grateful for the financial support from the National Natural Science Foundation of China (Nos. 20902039 and 20872051) and Natural Science Foundation of Jiangsu Province (BK2009543).

References

1. (a) M. Prato, Top. Curr. Chem., 1999, 199, 173; CrossRef (b) F. Diederich and M. Gómez-López, Chem. Soc. Rev., 1999, 28, 263; CrossRef (c) E. Nakamura and H. Isobe, Acc. Chem. Res., 2003, 36, 807; CrossRef (d) D. M. Guldi, F. Zerbetto, V. Georgakilas, and M. Prato, Acc. Chem. Res., 2005, 38, 38. CrossRef
2. For reviews, see: (a) R. Taylor and D. R. M. Walton, Nature, 1993, 363, 685; CrossRef (b) A. Hirsch, Synthesis, 1995, 895; CrossRef (c) F. Diederich and C. Thilgen, Science, 1996, 271, 317; CrossRef (d) A. Hirsch, Top. Curr. Chem., 1999, 199, 1; CrossRef (e) C. Thilgen and F. Diederich, Top. Curr. Chem., 1999, 199, 135; CrossRef (f) M. A. Yurovskaya and I. V. Trushkov, Russ. Chem. Bull., Int. Ed., 2002, 51, 367.
3. (a) P. Clawson, P. M. Lunn, and D. A. Whiting, J. Chem. Soc. Perkin. Trans. 1, 1990, 159; CrossRef (b) C. Yoakim, N. Goudreau, G. A. Mcgibbon, J. Meara, P. W. White, and W. W. Ogilvie, Helv. Chim. Acta, 2003, 86, 3427. CrossRef
4. D. Ramaiah, S. Rajadurai, P. K. Das, and M. V. George, J. Org. Chem., 1987, 52, 1082. CrossRef
5. A. Robert, J. J. Pommeret, E. Marchand, and A. Foucaud, Tetrahedron, 1973, 29, 463. CrossRef
6. K.-R. Meier, A. Linden, G. Mlostoń, and H. Heimgartner, Helv. Chim. Acta, 1997, 80, 1190. CrossRef
7. S. G. Ruf, J. Dietz, and M. Regitz, Tetrahedron, 2000, 56, 6259. CrossRef
8. N. Jagerovic, J. Elguero, and J.-L. Aubagnac, J. Chem. Soc., Perkin Trans. 1, 1996, 499. CrossRef
9. (a) V. Nair, D. Sethumadhavan, K. C. Sheela, and G. K. Eigendorf, Tetrahedron Lett., 1999, 40, 5087; CrossRef (b) V. Nair, D. Sethumadhavan, K. C. Sheela, S. M. Nair, and G. K. Eigendorf, Tetrahedron, 2002, 58, 3009. CrossRef
10. G.-W. Wang, H.-T. Yang, P. Wu, C.-B. Miao, and Y. Xu, J. Org. Chem., 2006, 71, 4346. CrossRef
11. C. V. Kumar, D. Ramaiah, P. K. Das, and M. V. George, J. Org. Chem., 1985, 50, 2818. CrossRef
12. M. Lipson, B. C. Noll, and K. S. Peters, J. Org. Chem., 1997, 62, 2409. CrossRef
13. J. L. Adcock and H. Zhang, J. Org. Chem., 1995, 60, 1999. CrossRef
14. L.-H. Shu, G.-W. Wang, S.-H. Wu, H.-M. Wu, and X.-F. Lao, Tetrahedron Lett., 1995, 36, 3871. CrossRef