Published in J. Chem. Soc., Chem. Commun., 1994, 1567.
(c) Royal Societiy of Chemistry, 1994.

An SCF-MO Study of the Dimerisation Reaction of Hemifullerene (C30H12) to the Potential Fullerene precursor C60H24.

M. John Plater,[a] Henry S. Rzepa,[b]and Stefan Stossel[b]

[a]Department of Chemistry, Meston Walk, University of Aberdeen,. AB9, 2UE. [b]Department of Chemistry, Imperial College of Science, Technology and Medicine, London, UK SW7 2AY.

AM1 and PM3 SCF-MO calculations suggest the dimerisation of Hemifullerene 1 to 2 by a mechanism involving six concurrent [pi]2s+[pi]4s additions corresponds to a stationary point with six negative force constants; the first stepwise [pi]2s+[pi]4s transition state is found to be highly unsymmetrical with a large barrier to reaction.

Whilst Triindenotriphenylene 1 (C30H12) has attracted recent attention[1] as a rational synthetic precursor to the C60 fullerene skeleton, and the overall dimerisation to 2 (C60H24) is estimated to be thermoneutral,[2 ]no estimate of the dimerisation barrier and hence the feasibility of this reaction is available. This system also represents an unusual example of a multibond reaction involving potentially twelve bonds, where the timing of the bond formations is of some interest.[3 ]Whilst the simple [pi]2s+[pi]4s cycloaddition of ethene and butadiene is thought to be synchronous,[4] both strain[5] and ring size[6] may induce asynchronous or even stepwise behaviour. We report here semi-empirical SCF-MO calculations at the closed shell AM1 and PM3 levels for the reaction of 1 and various model compounds, which suggest that the concerted [pi]2s+[pi]4s dimerisation reaction may be both asynchronous and have a large barrier to reaction.
12 345 67
The transition state for the initial single [pi]2s+[pi]4s cycloaddition of 1 (to its mirror image) proved unusually difficult to locate using conventional methods. We adopted a stepwise strategy of combining 1 with smaller precursors such as ethene or 3-7 acting as the [pi]2s component or with butadiene acting as the [pi]4s component, and using the located transition states as the starting geometry for the next optimisation (Table). We did not attempt to calculate any of the subsequent five [pi]2s+[pi]4s cycloadditions, and hence the first barrier establishes only a lower limit for the overall reaction rate limiting step. The results reveal that the barriers to these reactions are significantly higher than for the simple reaction of ethene and butadiene (Table), due to both loss of aromaticity and strain. The increase in barrier derives more from 1 acting as a diene than as an alkene (Table).We also note that for symmetrical [pi]2s+[pi]4s reactions at least, the closed shell RHF approximation compares well with MCSCF results.[4] For the series 1 + 3, 4, 5, 6, 7 and finally 1 itself, the calculated asymmetry in the two forming bonds increases (Table).[5] Although both the AM1 and PM3 methods concur structurally with ab initio methods for the ethene+butadiene reaction,[4] asymmetric distorsion due to strain has not hitherto been demonstrated at the ab initio level.[ ]The asymmetry for the [pi]2s+[pi]4s cycloaddition of 1+1 also implies that significant biradical character may be present in the transition state, and that the reaction may be easily diverted from the full dimerisation pathway by radical or hydrogen elimination reactions. Although the closed shell SCF derived barrier for such an asynchronous and biradical like reaction is certainly overestimated, that for the more symmetrical reaction of 1 + 3 is also high. The size of the system currently precludes a properly correlated calculation at an ab initio level.

The stationary point corresponding to concurrent six fold [pi]2s+[pi]4s cycloaddition was readily characterised, and was shown to have either 6 (PM3) or 7 (AM1) negative force constants, with a very high energy barrier (Table). The near degeneracy of the first six vibrations indicates that the genuine transition states are likely to correspond to six successive rather than to six concurrent [pi]2s+[pi]4s additions. Our previous finding that metal ions as templates do not appear to enhance the dimerisation, coupled with the present findings that the first [pi]2s+[pi]4s step may have a higher than normal barrier, indicate that high temperatures may still be neccessary for attempts to form the C60 nucleus from 1.

Acknowledgements. We thank the Erasmus scheme for a studentship to SS and the SERC (EPSRC) and Wolfson Foundation for equipment grants.

Theoretical calculations were carried out at the restricted Hartree-Fock level (RHF) AM1 or PM3 semi-empirical methods, as implemented in the MOPAC 93 program.[ ]All structures were optimised using the eigenvector following algorithm, followed by a vibrational analysis to characterise the stationary points.

Computer readable files for Unix, Apple Macintosh or Microsoft Windows systems in MPEG video animation format illustrating the three dimensional properties of these stationary points are available for general access using the world-wide-webb uniform resource locator http://www.ch.ic.ac.uk/rzepa/RSC/CC/4_02038C.html for a period of at least two years from the publication of this paper. A description of how to visualise such material, together with appropriate programs is available from the same sources.

MIME Types: If you have a WWW browser and you wish to acquire and visualise the MOPAC files corresponding to the transition states described here, you will have to define some additional MIME types. Click here for a description of how this is done. If you click on single 2+4 or six fold 2+4 you should be able to download the MOPAC data file and visualise it locally.
References

1. M. L. McKee and W. C. Herndon, J. Mol. Struct. (Theochem), 1987, 153, 75; M. J. Plater, Synlett., 1993, 6, 405; R. Faust and K. P. C. Vollhardt, J. Chem. Soc., Chem. Commun., 1993, 1471. See also R. Taylor, G. J. Langley, H. W. Kroto and D. R. M. Walton, Nature, 1993, 336, 728 for formation from naphthalene.

2. M. J. Plater, H. S. Rzepa, F. Stoppa and S. Stossel, J. Chem. Soc., Perkin Trans. 2, 1994, 399.

3. R. Huisgen, Pure Appl. Chem., 1981, 53, 171; A. Firestone, Heterocycles, 1987, 25, 61; W. T. Borden, R. J. Loncharich and K. N. Houk, Ann. Rev. Phys. Chem., 1988, 39, 213; M. J. S. Dewar and C. X. Jie, Acc. Chem. Res., 1992, 25, 537.; K. N. Houk, Y. Li, J. D. Evanseck, Angew. Chemie (Int Ed), 1992, 31, 682. D. A. Hrovat, K. Morokuma, W. T. Borden, J. Am. Chem. Soc., 1994, 116, 1072.

4. F. Bernardi, A. Bottoni, M. J. Field, M. F. Guest, I. H. Hillier, M. A. Robb and A. Venturini, J. Am. Chem. Soc., 1988, 110, 3050; J. H. W. McDouall, M. A. Robb, V. Niazi, F. Bernardi and H. B. Schlegel, J. Am. Chem. Soc., 1987, 109, 4642; M. J. S. Dewar, J. J. P. Stewart and S. Olivella, J. Am. Chem. Soc., 1986, 108, 5771.

5. H. S. Rzepa, P. Molina, M. Alajarin, A. Vidal, Tetrahedron, 1992, 48, 7425.

6. J. W. McIver, Acc. Chem. Res, 1974, 7, 73, J. W. McIver and A. Komornicki, J. Am. Chem. Soc., 1972, 94, 2625. See also H. S. Rzepa and W. Wylie, J. Chem. Soc., Perkin Trans. 2, 1991, 939.

Table. Calculated PM3 Enthalpies for the Reaction between 1 and 3-7, in kJ mol[-1]. 
Structure     [Delta]H    [Delta]H    [Delta]H#   r1, r2 (Å)    [nu]i       
              Reactants     ts                                        (cm-1)        
1+ethene      1135          1309          174           2.06, 2.10    1066          
1+butadiene   1198          1357          159           2.10, 2.15    966           
1+3           1838          2108          270           1.99, 2.19    1017          
1+4           1898          2169          271           1.98, 2.21    1003          
1+5           2016          2295          279           1.97, 2.23    985           
1+6           2506          2778          272           1.99, 2.22    1003          
1+7           2558          2831          273           1.97, 2.25    975           
1+1a          2131          2388          257           1.69, 3.03    205           
1+1b          2441          2710          268           1.74, 2.91    497           
1+1c          2131          3754          1623          2.14, 2.10    e             
1+1b,c        2441          4120          1680          2.17, 2.10    f             
d             202           312           110           2.14, 2.14    928           


[a]
Energies of intermediates corresponding to the product of successive [pi]2s+[pi]4s additions; 2206, 2121, 2310, 2316, 2189 kJ mol-[1.] [b ]AM1 calculation.[ c ]Stationary point for six concurrent cycloadditions. [d] Reaction between ethene and butadiene. [e] Values of six imaginary modes, 1274, 1189, 1189, 1060, 1060, 977 cm[-1]. [f] Values of seven imaginary modes: 1199, 1116, 1116, 982, 982, 904, 78.