An optimised scalable synthesis of H₂O@C₆₀ and a new synthesis of H₂@C₆₀

Abstract

New high-yielding synthetic routes to the small-molecule endofullerenes H₂O@C₆₀, D₂O@C₆₀ and H₂@C₆₀ are described. The use of high temperatures and pressures for the endohedral molecule incorporation are avoided. A new partial closure step using PPh₃, and final suturing using a novel Diels–Alder/retro-Diels–Alder sequence are amongst the advances reported.

---

The molecular structure of C₆₀ fullerene consists of a carbon cage enclosing a spherical cavity with a diameter of approximately 3.5 Å. Small-molecule endofullerenes encapsulate a single small molecule such as H₂ or H₂O within the cavity.^1,2 The encapsulated molecule behaves as a confined quantum rotor, displaying a rich energy level structure,³ which may be studied by neutron scattering,^4,5 infrared spectroscopy^5,6 and nuclear magnetic resonance (NMR).^3,5,7 The stability and homogeneity of endofullerenes such as H₂@C₆₀ and H₂O@C₆₀ allows the convenient and precise study of important physical phenomena such as nuclear spin isomer conversion.^5,7,8 It is possible to detect the endohedral interconversion of ortho and para-water, and to show that the conversion is a bimolecular process even under cryogenic conditions, with neighbouring pairs of ortho-water molecules interacting so as to generate a pair of para-water molecules.⁷ Apart from their intrinsic interest, such observations may underpin new routes to the generation of nuclear hyperpolarization, analogous to the chemical reactions of para-enriched hydrogen gas.⁹ The high demand for these unique substances has highlighted limitations of the current methods for their preparation. The synthetic route to endofullerenes has been described as “molecular surgery”¹⁰ and involves three stages; (i) opening and progressive widening of an orifice in the C₆₀ cage through exohedral reactions, (ii) introduction of a small molecule through the orifice into the cavity, (iii) further reactions leading to reclosure of the C₆₀ cage and encapsulation of the endohedral molecule. The final (reclosure) stage is the most difficult and has been accomplished only by Komatsu and Murata using two different routes leading to H₂@C₆₀ (8 steps, 8% overall yield),¹ and H₂O@C₆₀ (6 steps, 5% yield) respectively.² Both syntheses require specialised high pressure equipment for the endohedral molecule incorporation step. Herein we report more efficient and practical routes to these important molecular systems.

The key feature of the H₂O@C₆₀ synthesis is that the open-cage derivative 1, having a 13-membered orifice, can be converted in situ into its dehydrated form 2 that has a 16-membered orifice. Compound 2 is able to incorporate a water molecule but is easily re-hydrated, reducing once again the size of the orifice trapping the guest molecule inside the cage. Compound H₂O@1 is then reduced with (ⁱPrO)₃P to H₂O@3 which in turn is pyrolysed under vacuum to produce H₂O@C₆₀ (Scheme 1).

Scheme 1 H₂O@C₆₀ synthesis.²

The reported route² to the open fullerene 1 worked well but we lacked the equipment to apply the very high pressures (9000 atm) used to efficiently encapsulate the water molecule. It was already known that compound 1 could incorporate water (8%) under atmospheric pressure² and that higher filling factors could be achieved under similar conditions in other open fullerenes bearing larger orifices.¹¹ Published studies on the filling of hydrophobic cavities by water molecules suggested that an increased filling factor could be as well achieved by changing variables other than pressure.¹² On these premises we investigated the filling of 1 under low pressure conditions.

Heating a solution of 1 in toluene in an NMR tube in the presence of water (5.6 equiv.) at 120 °C reached a maximum of 23% incorporation after 36 h reflecting the equilibrium situation. Increasing reaction time further gave substantial decomposition. Changing to a sealed tube which could be uniformly heated to prevent condensation in cooler regions increased incorporation to 45%. Using a large excess of water did not significantly change incorporation. Changing to other aromatic solvents demonstrated increased incorporation (benzene, 39%; ortho-dichlorobenzene, 61%; 1-chloronapthalene, 67%). Using 1-chloronapthalene as the solvent and lowering the temperature to 100 °C gave 78% H₂O incorporation after 48 h providing a practical alternative to high pressure procedures. The same process can be extended to the synthesis of D₂O@1 but to avoid contamination with HDO@1 and H₂O@1 it was necessary to pre-form the dehydrated compound 2 by refluxing a toluene solution of 1 for one hour, passing the condensed solvent through a column of activated 3 Å molecular sieves. Subsequent removal of solvent to give 2, dissolution in 1-chloronapthalene, addition of D₂O and heating as above gave D₂O@1 with no HOD@1 or H₂O@1 visible by ¹H NMR spectroscopy. Compound 2 was recently isolated using a different protocol.¹³

In an effort to complete the synthesis of H₂O@C₆₀ our attention turned to suturing H₂O@1. Unfortunately when compound H₂O@1 was reacted with (ⁱPrO)₃P the expected product H₂O@3 was isolated in only 34% yield. The same reaction mixture yielded 31% of the novel compound H₂O@4. The structure of the latter can be deduced from the appearance in the ¹H NMR spectrum of a singlet at δ 7.80 ppm from the proton present on the rim of the orifice and a broad signal at δ 5.21 attributable to the single hydroxyl group. The DFT-GIAO calculated spectra suggest the formation of regioisomer 4. Compound 4 was the only product isolated (76% yield) when 1 was reacted with (ⁱPrO)₃P at room temperature. Compound 4 is inert to (ⁱPrO)₃P and H₂O@4 does not lose its endohedral water molecule upon heating. A somewhat similar reactivity towards alkyl phosphites has been reported for another open-cage fullerene, and the mechanism of the reduction studied theoretically.¹⁴ Pre-formed 2 reacted with (ⁱPrO)₃P at room temperature to afford compound 3 in 33% yield without contamination by 4. It is likely that under the reported closure conditions² it is a mixture of H₂O@1 and H₂O@2 formed in situ which are reacting with (ⁱPrO)₃P to give varying mixtures of H₂O@4 and H₂O@3 (Scheme 2).

Scheme 2 Reactivity of H₂O@1 and H₂O@2 towards triisopropyl phosphite.

In order to overcome this side reactivity of H₂O@1 towards alkyl phosphites we looked at alternative reagents and found that reaction with excess Ph₃P at 120 °C cleanly gave compound H₂O@5, isolated in 84% yield. Compound H₂O@4 was not formed and, even with longer reaction times, reduction to compound H₂O@3 was not observed. No endohedral water was lost in the formation of H₂O@5 for which this reaction provides the first synthesis. Compound H₂O@5 could be reacted with excess (ⁱPrO)₃P to give H₂O@3 which was isolated after chromatography in 84% yield. In comparison with the reported direct reaction with (ⁱPrO)₃P,² the yield of H₂O@3 afforded in the two step process, is higher overall (69%) (Scheme 3).

Scheme 3 Optimised suturing procedure of compound H₂O@1 to give H₂O@C₆₀.

Our attention finally turned to the last step of the closure – the conversion of H₂O@3 into H₂O@C₆₀. The reported yield of 29% was achieved by vacuum pyrolysis (360 °C) of H₂O@3 dispersed on dry neutral alumina.² Unfortunately in our hands we could isolate only traces of the product. A reasonable mechanism involves an initial [4+2] intramolecular cycloaddition to give strained intermediate 6 which rearranges via radical cleavage (formally a retro-[4+4] cycloaddition) of the strained C–C cyclopropane bonds generating intermediate 7. Finally, a retro-[2+2+2] cycloaddition leads to the extrusion of the side aromatic groups and formation of the C₆₀ structure (Scheme 4).

Scheme 4 Proposed mechanism for the conversion of 3 into C₆₀. The C₆₀ skeleton other than the orifice carbons (solid line) has been omitted for clarity.

The last step (7 to C₆₀) is likely to have the highest activation energy. We postulated that a lower energy reaction profile could be followed in presence of N-phenylmaleimide, a strong dienophile that was expected to react readily with the intermediate 7 in a [4+2] cycloaddition to afford adduct 8 which is a good substrate for a retro [4+2] cycloaddition to regenerate the C₆₀ structure. These predictions were confirmed experimentally by reacting compound H₂O@3 with N-phenylmaleimide in 1-chloronaphthalene under reflux conditions. After 20 hours a clean conversion to H₂O@C₆₀, was observed and column chromatography afforded the product in 90% yield. Further purification by sublimation gave pure H₂O@C₆₀ in 72% overall yield (Scheme 3). The same route has been followed to complete the synthesis of D₂O@C₆₀ in identical yield.

Komatsu has reported the synthesis of H₂@C₆₀ in 8 steps and 8% overall yield via insertion of H₂ into a different cage-opened fullerene at 800 atm and 200 °C.¹ Given the efficiency of the opening/closure described above a more convenient synthesis seemed possible. Incorporation of molecular hydrogen into 2 (formed in situ from 1 using 3 Å molecular sieves) at 120 °C and 120 atm H₂ for 20 h gave H₂@1 after work-up with a 60% filling factor. The lower temperature (120 °C cf. 200 °C) allowed reasonable incorporation at moderate pressures. It is likely that we would obtain complete filling at the high pressures (800 atm) used previously.¹

In order to test if the system could be closed without the loss of the endohedral hydrogen molecule, a mixture of H₂O@1 and H₂@1 was reacted with (ⁱPrO)₃P in refluxing toluene. All the endohedral hydrogen was lost while all the water was retained inside. The problem was overcome by carrying out the phosphite reaction under hydrogen pressure. Thus compound 1 was heated to 120 °C in o-dichlorobenzene under hydrogen pressure (120 atm) in the presence of molecular sieves to give H₂@2 (60% H₂ incorporation). The bomb was then depressurised and (ⁱPrO)₃P or Ph₃P was added before pressurising with hydrogen again and heating to 120 °C overnight. The reactions afforded 60% filled H₂@3 in 50% yield and 60% filled H₂@5 in 84% yield respectively. Both H₂@3 and H₂@5 can be reacted in an analogous way as already described for H₂O@3 and H₂O@5 to isolate 60% filled H₂@C₆₀ (Scheme 5).

Scheme 5 Synthesis of H₂@C₆₀ from 1.

In conclusion a systematic study of the H₂O@C₆₀ synthesis developed by Murata² has been carried out. It has been demonstrated that 78% incorporation of H₂O and D₂O into compound 1 can be achieved without the need for high pressure equipment. Treatment of hemiacetal 1 with (ⁱPrO)₃P at room temperature gave the reduced product 4 and H₂O@4 was a major by product during the (ⁱPrO)₃P induce closure of H₂O@1 to give H₂O@3. Dehydration of hemi-acetal 1 to give tetraketone 2 before reaction with (ⁱPrO)₃P at room temperature gave 3 without formation of 4. A new partial closure procedure from H₂O@1 using PPh₃ allowed H₂O@5 to be isolated for the first time, and subsequent deoxygenation with (ⁱPrO)₃P gave H₂O@3 in improved overall yield. A new method for closing the fullerene H₂O@3 to afford H₂O@C₆₀ was developed using a Diels–Alder/retro-Diels–Alder sequence which gave substantially higher yields than the known pyrolysis method, and was readily scalable. Overall pure sublimed H₂O@C₆₀ (78% H₂O incorporation) was synthesised in 15% yield from C₆₀. D₂O@C₆₀ uncontaminated with HOD@C₆₀ was made in similar yield. The same synthetic route has been successfully applied to the formation of H₂@C₆₀, the large size of the orifice of compound 2 enabling a 60% incorporation of H₂ at 120 °C and 120 atm. These advances will lead to a greatly improved availability of the important molecular endofullerenes H₂O@C₆₀ and H₂@C₆₀, and their isotopologues.

We thank the EPSRC (EP/I029451/1) and ERDF (Interreg-IVB, MEET project) for funding, and acknowledge the use of the IRIDIS High Performance Computing Facility and associated support services at the University of Southampton. We wish to thank Prof. Koichi Komatsu, Prof. Yasujiro Murata, Prof. Nick Turro, Dr Lei Xuegong and Dr YongJun Li for sharing their detailed experimental procedures.

Notes and references

1. K. Komatsu, M. Murata and Y. Murata, Science, 2005, 307, 238–240.
2. K. Kurotobi and Y. Murata, Science, 2011, 333, 613–616.
3. M. H. Levitt, Philos. Trans. R. Soc., A, 2013, 371, 20120429.
4. (a) A. J. Horsewill, K. Goh, S. Rols, J. Ollivier, M. R. Johnson, M. H. Levitt, M. Carravetta, S. Mamone, Y. Murata, J. Y.-C. Chen, J. A. Johnson, X. Lei and N. J. Turro, Philos. Trans. R. Soc., A, 2013, 371, 20110627; (b) A. J. Horsewill, K. S. Panesar, S. Rols, M. R. Johnson, Y. Murata, K. Komatsu, S. Mamone, A. Danquigny, F. Cuda, S. Maltsev, M. C. Grossel, M. Carravetta and M. H. Levitt, Phys. Rev. Lett., 2009, 102, 013001.
5. C. Beduz, M. Carravetta, J. Y.-C. Chen, M. Concistré, M. Denning, M. Frunzi, A. J. Horsewill, O. G. Johannessen, R. Lawler, X. Lei, M. H. Levitt, Y. Li, S. Mamone, Y. Murata, U. Nagel, T. Nishida, J. Ollivier, S. Rols, T. Rõõm, R. Sarkar, N. J. Turro and Y. Yang, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 12894–12898.
6. (a) S. Mamone, M. Ge, D. Hüvonen, U. Nagel, A. Danquigny, F. Cuda, M. C. Grossel, Y. Murata, K. Komatsu, M. H. Levitt, T. Rõõm and M. Carravetta, J. Chem. Phys., 2009, 130, 081103; (b) M. Ge, U. Nagel, D. Hüvonen, T. Rõõm, S. Mamone, M. H. Levitt, M. Carravetta, Y. Murata, K. Komatsu, X. Lei and N. J. Turro, J. Chem. Phys., 2011, 135, 114511; (c) T. Rõõm, L. Peedu, M. Ge, D. Hüvonen, U. Nagel, S. Ye, M. Xu, Z. Bačić, S. Mamone, M. H. Levitt, M. Carravetta, J. Y.-C. Chen, X. Lei, N. J. Turro, Y. Murata and K. Komatsu, Philos. Trans. R. Soc., A, 2013, 371, 20110631.
7. S. Mamone, M. Concistrè, E. Carignani, B. Meier, A. Krachmalnicoff, O. G. Johannessen, X. Lei, Y. Li, M. Denning, M. Carravetta, K. Goh, A. J. Horsewill, R. J. Whitby and M. H. Levitt, J. Chem. Phys., 2014, 140, 194306.
8. (a) M. Frunzi, S. Jockusch, J. Y.-C. Chen, R. M. K. Calderon, X. Lei, Y. Murata, K. Komatsu, D. M. Guldi, R. G. Lawler and N. J. Turro, J. Am. Chem. Soc., 2011, 133, 14232–14235; (b) Y. Li, X. Lei, S. Jockusch, J. Y.-C. Chen, M. Frunzi, J. A. Johnson, R. G. Lawler, Y. Murata, M. Murata, K. Komatsu and N. J. Turro, J. Am. Chem. Soc., 2010, 132, 4042–4043.
9. (a) C. R. Bowers and D. P. Weitekamp, J. Am. Chem. Soc., 1987, 109, 5541–5542; (b) R. W. Adams, J. A. Aguilar, K. D. Atkinson, M. J. Cowley, P. I. P. Elliott, S. B. Duckett, G. G. R. Green, I. G. Khazal, J. López-Serrano and D. C. Williamson, Science, 2009, 323, 1708–1711.
10. J. C. Hummelen, M. Prato and F. Wudl, J. Am. Chem. Soc., 1995, 117, 7003–7004.
11. (a) S. Iwamatsu, T. Uozaki, K. Kobayashi, S. Re, S. Nagase and S. Murata, J. Am. Chem. Soc., 2004, 126, 2668–2669; (b) Q. Zhang, Z. Jia, S. Liu, G. Zhang, Z. Xiao, D. Yang, L. Gan, Z. Wang and Y. Li, Org. Lett., 2009, 11, 2772–2774.
12. J. C. Rasaiah, S. Garde and G. Hummer, Annu. Rev. Phys. Chem., 2008, 59, 713–740.
13. T. Futagoishi, M. Murata, A. Wakamiya, T. Sasamori and Y. Murata, Org. Lett., 2013, 15, 2750–2753.
14. (a) Q. Zhang, T. Pankewitz, S. Liu, W. Klopper and L. Gan, Angew. Chem., Int. Ed., 2010, 49, 9935–9938; (b) Y. Guo, J. Yan and N. M. Khashab, ChemPhysChem, 2012, 13, 751–755.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, compound characterization data, and details of calculations. See DOI: 10.1039/c4cc06198e

---