View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/B704924B (Paper) New J. Chem., 2008, 32, 89-93

Regioselective synthesis and crystal structure of C70(CF3)10[C(CO2Et)2]†

Nataliya S. Ovchinnikova a, Daria V. Ignat’eva a, Nadezhda B. Tamm a, Stanislav M. Avdoshenko a, Alexey A. Goryunkov *a, Ilya N. Ioffe a, Vitaliy Yu. Markov a, Sergey I. Troyanov a, Lev N. Sidorov a, Marina A. Yurovskaya a and Erhard Kemnitz b
aChemistry Department, Moscow State University, Leninskie Gory, 119992 Moscow, Russia. E-mail: aag@thermo.chem.msu.ru; Fax: +7 495-939-1240; Tel: +7 495-939-5373
bInstitute of Chemistry, Humboldt University Berlin, Brook-Taylor-Straße 2, 12489 Berlin, Germany

Received (in Durham, UK) 3rd April 2007 , Accepted 8th August 2007

First published on 28th August 2007

Abstract

The Bingel reaction of poly(trifluoromethyl)fullerene p7mp-C70(CF3)10 with diethyl malonate and CBr4 in the presence of bases yields the C70(CF3)10[C(CO2Et)2] cycloadduct as a major product, along with two C70(CF3)10[CH(CO2Et)] isomers. An XRD study of the main compound demonstrates that a [2 + 1] cycloaddition occurs at the unoccupied pole of the p7mp-C70(CF3)10 molecule. The observed regiochemical selectivity of the [2 + 1] cycloaddition is shown to be favored from both energetic and orbital reactivity viewpoints.

Introduction

The recent development of synthetic techniques for the trifluoromethylation of C60 and C70 has provided a considerable number of individual structurally-characterized C60/70(CF3)n compounds with n = 2–18.1–6 Enhanced electron-withdrawing properties,7,8 chemical stability against nucleophilic substitution and high solubility in commonly used organic solvents make it possible to regard trifluoromethylated fullerenes as promising building blocks for light-harvesting systems. The design of such systems requires the linking of an acceptor fullerene derivative to an organic donor. There is a need for further reliable techniques for the organic functionalization of trifluoromethylated compounds. The approach to this problem that looks the most natural is to apply the reactions known for bare C60 and C70, such as the Bingel reaction with stabilized α-halocarbanions,9,10 an extremely popular fullerene chemistry workhorse.

Here we report the first example of the functionalization of a fluoroalkylated fullerene, p7mp-C70(CF3)10(1,4,10,19,25,41,49,60,66,69)-C70(CF3)1011 according to IUPAC recommendation12) via the Bingel reaction.

Experimental section

General information

Mass spectrometric analysis. Negative ion MALDI mass spectrometry was applied to analyze the crude product and isolated HPLC fractions. The spectra were recorded using a Bruker AutoFlex II (Germany) Rf-TOF mass spectrometer equipped with an N2 laser (337 nm, 1 ns pulse), both in reflectron and linear modes to filter off the signals due to metastable ions formed via post source decay. The 2-[(2E)-3-(4-tert-butylphenyl)-2-methylprop-2-enylidene]malononitrile (DCTB, ≥99%, Fluka) matrix was used with matrix-to-analyte ratios of 1000–4000.
High performance liquid chromatography . HPLC analyses were carried out using an Agilent Technologies System Model 1100 equipped with a diode array detector (DAD) and a temperature-controlled Cosmosil Buckyprep 4.6 mm × 25 cm column (Nacalai Tesque, Inc.). Further HPLC separation was carried out using a Waters 1500 chromatographic system equipped with a dual wavelength UV/vis detector and 10 mm × 25 cm column (Nacalai Tesque, Inc.). The eluents used included toluene (99.8%, Khimmed, Russia), hexane (99.7%, Khimmed, Russia) and their mixtures. The UV/vis spectra of toluene solutions given below were obtained via a DAD in the 290–950 nm range with 2 nm resolution.

Synthesis

p 7 mp-C70(CF3)10. Starting C70 (Term-USA, 99.8%; 42 mg) was placed into a glass ampoule and an excess of CF3I (Apollo, 98%; ca. 0.5 mL) was then condensed into it under cooling with liquid nitrogen. The sealed ampoule was placed into a gradient furnace so that the fullerene was heated to 400 °C for 70 h while the liquid CF3I remained at room temperature and thus provided a vapor pressure of ca. 5 bar. After removal of I2viasublimation, the synthesized mixture of C70(CF3)n with n between 10 and 18, as evidenced by MALDI mass spectroscopy, was additionally heated in an evacuated glass ampoule at 400 °C for 40 h. The final product was found to be dominated by C70(CF3)10 with small admixtures of C70(CF3)8 and C70(CF3)12. HPLC purification of p7mp-C70(CF3)10 (retention time 3.46 min) was carried out using a Cosmosil Buckyprep (10 mm x 25 cm) column with toluene as the eluent (4.6 mL min–1). This yielded the p7mp-C70(CF3)10 isomer (24.5 mg, 32%).
Bingel reaction. Diethyl malonate (Acros Organics, 99+%; 1.5 µL, 9.9 µmol in 0.2 mL of toluene) and carbon tetrabromide (Acros Organics; 3.3 mg, 10.0 µmol in 0.2 mL of toluene) were added to the purified C70(CF3)10 (10 mg, 6.5 µmol in 5 mL of toluene). 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU; Acros Organics, 98%; 3 µL, 19.9 µmol in 0.2 mL of toluene) was then added dropwise. The reaction mixture was stirred at room temperature for 2 h. HPLC separation provided three products with retention times of 1.35 (main fraction, 6.8 mg, 62%), 1.57 (ca. 0.5 mg, 4.3%) and 1.76 (ca. 0.4 mg, 3.6%) min (Cosmosil Buckyprep 4.6 mm × 25 cm, toluene, 2 mL min–1), shown by means of MALDI mass spectrometry to be C70(CF3)10[C(CO2Et)2] and the two isomers of C70(CF3)10[CH(CO2Et)], respectively.

X-Ray crystallography

Diffraction measurements for the C70(CF3)10[C(CO2Et)2]·2.5C7H8 crystals were carried out at 100 K using synchrotron radiation at the BESSY storage ring (λ = 0.9100 Å, BL14.2, PSF of the Free University of Berlin, Germany) and a MAR345 image plate detector. The structure was solved using SHELXS-97 and refined using SHELXL-97. Crystal data: C104.5H30F30O4, M = 1919.29, triclinic, P[1 with combining macron], a = 12.434(1), b = 14.692(1), c = 20.783(1) Å, α = 100.088(3), β = 93.337(2), γ = 103.419(3)°, V = 3616.5(4) Å3, Z = 2, µ = 0.160 mm–1, Rint = 0.0759. All atoms, except for those of the ethyl groups and toluene molecules, were refined anisotropically; wR2 = 0.323 (for 9014 reflections and 1210 parameters), R1 = 0.120 (for 6640 reflections with I > 2σ(I)). CCDC 642153. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b704924b

Quantum chemical calculations

The range of theoretically possible isomers of C70(CF3)10[CR′R″], where CR′R″ was CH2, CH(CO2Et) and C(CO2Et)2, was considered in the present study. Preliminary geometry optimization for these molecules was carried out at the AM1 level of theory using PC-GAMESS software.13 Final optimizations at the DFT level of theory and MO calculations were performed using PRIRODA software14 that employed an original TZ2P basis set and a PBE exchange–correlation functional.15 Calculation details, and relative energies at the AM1 and DFT levels of theory are available in the ESI.†

Results and discussion

The described technique for the preparation of p7mp-C70(CF3)10 differs from that previously reported.11,16 The reaction of CF3I with C70 in a sealed glass ampoule in a gradient furnace at the 400 °C resulted in the accumulation of highly trifluoromethylated products C70(CF3)n (n = 12–20) in the cold zone of the ampoule.2,17,18 Further prolonged heating of the products in the evacuated ampoule at 400 °C led to the partial loss of CF3groups (that, perhaps, recombine into C2F6) and the formation of a C70(CF3)n (n = 8–12) mixture dominated by p7mp-C70(CF3)10. This can be regarded as evidence for the higher thermodynamic stability of the latter molecule or, in other words, the higher average C–CF3 binding energy due, perhaps, to the lower amount of steric strain in the less densely CF3-populated latter molecule.

A room temperature Bingel reaction of the HPLC-purified p7mp-C70(CF3)10 with equimolar amounts of diethyl malonate and CBr4, in the presence of 2 equivalents of DBU as a base, was observed to proceed smoothly, and its completion took 2 h. HPLC separation of the crude reaction mixture (see Fig. 1a) afforded one major fraction (1.35 min, ca. 90% HPLC-monitored relative abundance) and two minor fractions (1.57 and 1.76 min, ca. 5% each) but no traces of the starting p7mp-C70(CF3)10. In Fig. 1, we provide MALDI mass spectra of the major and one of the minor fractions, the other minor fraction having shown an identical spectrum. The most prominent signal in the spectrum of the major fraction corresponded to the C70(CF3)10[C(CO2Et)2]– molecular ion, while the minor fractions showed the C70(CF3)10[CHCO2Et]– ion. Fragment ions due to the loss of one CF3group were also observed in all fractions. Besides this, there were signals at irregular m/z values separated from the fragment ion peaks by ca. +13 Da. These signals, not detectable in the linear mode of spectra acquisition, were attributed to metastable ions formed during the post source decay via a similar CF3-loss process.


The negative ion MALDI mass spectra (reflectron mode) of the separated adducts C70(CF3)10[C(CO2Et)2] (top) and C70(CF3)10[CH(CO2Et)] (bottom). The HPLC traces of the crude product mixture (a), the separated fractions of the major product (b) and the by-products (c and d) are shown as insets.
Fig. 1 The negative ion MALDI mass spectra (reflectron mode) of the separated adducts C70(CF3)10[C(CO2Et)2] (top) and C70(CF3)10[CH(CO2Et)] (bottom). The HPLC traces of the crude product mixture (a), the separated fractions of the major product (b) and the by-products (c and d) are shown as insets.

The above HPLC and MS data demonstrate that trifluoromethylated fullerenes are likely to show standard Bingel behavior. This constitutes an important distinction from polyfluorinated fullerenes, which are known to lose fluorinevia SN2′ reactions with intermediate bromomalonate anions.19 The observed isomers of the C70(CF3)10[CHCO2Et] by-product are likely to be due to the reaction of C70(CF3)10 with trace admixture of monoethyl malonate (detected by NMR in the starting reagent) and subsequent decarboxylation. It should be noted that the synthesis also yields trace amounts of the C70(CF3)10[C(CO2Et)2]2 bis-adduct, as revealed by MALDI-MS analysis of the solution eluted prior to the C70(CF3)10[C(CO2Et)2] chromatographic fraction.

Slow evaporation of the toluene solution of C70(CF3)10[C(CO2Et)2] gave a yellow crystalline material suitable for X-ray single-crystal study. The obtained X-ray structure, shown in Fig. 2, demonstrates that cycloaddition involves one of the five [6,6]-bonds that radiate from the unoccupied polar pentagon of p7mp-C70(CF3)10. This bond becomes strongly elongated from 1.386(3) Å in the parent p7mp-C70(CF3)1011 to 1.61(1) Å. Two other C–C bonds in the cyclopropane ring are of a more common 1.51(1) Å length. Four adjacent C–C cage bonds show a minor increase in their average length, from 1.446 Å in the starting molecule to 1.483 Å, probably due to a non-perfect sp3 hybridization of the carbon atoms shared by the C70 cage and the cyclopropane ring.


Two perspective views (a) and Schlegel diagram (b) of the C70(CF3)10[C(CO2Et)2] adduct (black circles denote the CF3groups).
Fig. 2 Two perspective views (a) and Schlegel diagram (b) of the C70(CF3)10[C(CO2Et)2] adduct (black circles denote the CF3groups).

It is known that the [2 + 1] cycloaddition of diethyl bromomalonate to C70 in presence of DBU yields a single mono-adduct, with its organic moiety attached to the same C70 cage sites as in C70(CF3)10[C(CO2Et)2] reported here.9 The suggested reaction mechanism involves nucleophilic addition of the diethyl bromomalonate anion, followed by ring closure via the intramolecular nucleophilic substitution of bromine;9 it can be shown that such nucleophilic cycloaddition to the most pyramidalized carbon atoms of the C70 cage is favored by both energetic (most pronounced relief of steric strain)20 and orbital reactivity (largest contributions to the LUMO)21 factors. Assuming that cyclopropanation of the more electron deficient p7mp-C70(CF3)10 occurs in a similar way, one can already provide a possible explanation of the observed selective formation of only a single isomer of C70(CF3)10[C(CO2Et)2]. Indeed, among the ten [6,6]-bonds of interest, only bond “a” (see Fig. 2b) has no CF3groups sharing a common pentagon or hexagons with either of its two atoms. This makes the least shielded bond “a” the most accessible site for a nucleophilic attack by the rather bulky diethyl bromomalonate anion. It is worth mentioning, on the contrary, that sequential Bingel addition to pristine fullerenes is known to yield isomeric mixtures,22,23 which may be due to the larger accessible area of the cage in the products of the initial stages (except the regiospecific bis-functionalization of C70 by covalently-tethered malonate motifs).24 Nevertheless, the increasing role of steric and certain electronic factors upon the further addition can cause the selective formation of products like tetrahedral hexaadducts of C60.25

In spite of the above arguments, which seem to apply, we felt it important to go beyond the above simplistic considerations and analyze in more detail the energetic and electronic regiochemical aspects. Since such cycloaddition reactions of fullerenes involving addition and elimination steps are known to yield [6,6]-closed methanofullerenes only,22 any [5,6]-cycloadducts were excluded from consideration. Further restrictions to those [6,6]-bonds that radiate from polar pentagons left only nine possible cycloaddition sites, marked a–e and a′–d′ in Fig. 2b. The relative DFT energies of the respective isomers of C70(CF3)10[CR2], where CR2 = C(CO2Et)2, CH(CO2Et) and CH2, are given in Table 1. It can be clearly seen that, from an energetic point of view, only sites a′ and d′ are strongly sterically hindered, whereas the remaining seven sites exhibit very similar trends, irrespective of the size of the addends at the cyclopropane, though site accessibility issues might, nevertheless, still be important in other cases. Hence, the differences in stability between isomers b′, c′ and a–e are likely to be due to the effect of the addition pattern on the carbon cage distortion and π-electronic structure rather than direct interactions of the addends. Anyway, isomer a is obviously favored thermodynamically and might be favored kinetically as well, provided that the addition kinetics correlate with the thermodynamics. It is worth mentioning that the AM1 relative energies of isomers a′ and d′ were found to be unreliable, with ca. 31 and 25 kJ mol–1 deviations from the DFT results, thus perhaps manifesting poor treatment of steric interactions and planar aromatic fragments by the semi-empirical methods.

Table 1 Relative stability of the possible isomers of C70(CF3)10[C(CO2Et)2], C70(CF3)10[CHCO2Et] and C70(CF3)10[CH2]
Sitea Relative DFT energies of C70(CF3)10[CR′R″] isomers/kJ mol–1, where [CR′R″] =
C[CO2Et]2 C[H(CO2Et)] #1b C[H(CO2Et)] #2b CH2
a See Fig. 2b. b Results for the two possible diastereomers.
a 0.0 0.1 0.0 0.0
b 17.1 14.9 15.5 11.1
c 6.5 10.5 10.8 8.9
d 12.7 11.2 10.9 10.8
e 15.4 15.3 15.4 12.6
a′ 49.6 –14.7 34.2 –23.1
b′ 18.2 21.6 19.1 19.3
c′ 24.9 25.0 25.5 23.9
d′ 23.1 2.8 19.4 0.5

The relative energy values of the C70(CF3)10[CHCO2Et] diastereomeric cycloadducts indicate, not surprisingly, that significant differences can only be observed in the same sterically problematic cases a′ and d′, where orientation of the carboxyethyl group towards the already CF3-occupied hexagon should, apparently, cause repulsive interactions. Taking into account the stability data, the two (or more HPLC-unresolved) reported isomers of C70(CF3)10[CHCO2Et] are likely to be either isomers a#1 and a#2, given that site accessibility issues are important, or, in the opposite case, diastereoisomer a′#1 and some or all isomers selected among a#1, a#2 and d′#1. The currently available experimental data most likely supports the first alternative. The UV/vis spectra of C70(CF3)10[C(CO2Et)2] and the two isomers of C70(CF3)10[CHCO2Et] presented in Fig. 3 reveal striking similarities, with all their features being almost coincident in position. Anticipating that the electronic spectra of fullerene derivatives would provide sufficiently distinctive fingerprints of the π-electron system topology, we believe that the similarity of the spectra implies the isostructural nature of the respective compounds. However, it would still be very desirable, particularly for the purposes of understanding the reactivity, to elucidate the structures of the isolated isomers and to further investigate whether the addition of CH2, despite being of non-Bingel type, would, perhaps, afford the single isomer a′.


UV/vis spectra (in toluene) of C70(CF3)10 and its cycloadducts.
Fig. 3 UV/vis spectra (in toluene) of C70(CF3)10 and its cycloadducts.

An alternative approach to regiochemical predictions is based on different types of reactivity indices that reflect various chemically relevant aspects of the molecular electronic structure. When applied to fullerene derivatives, this approach should again be corrected for the steric effects of large addends on closely located sites. Since the above-mentioned commonly accepted pathway of the Bingel reaction involves the nucleophilic addition of the diethyl bromomalonate anion, two widely used types of criteria may be used to predict the most reactive addition sites, namely the orbital reactivity criteria, based on the maximum density of LUMO or quasi-degenerate LUMOs, and charge-based criteria, based on positive atomic charges or Fukui functions. However, the charge variations in p7mp-C70(CF3)10 were typically found to be quite negligible for large delocalized fullerene-based systems. On the contrary, the structure of the frontier orbitals appeared to be more instructive. Our calculations demonstrate that three lower LUMOs of C70(CF3)10 span a range of only 0.14 eV, while the gap between LUMO + 2 and LUMO + 3 is approximately twice the size. Therefore, LUMO, LUMO + 1 and LUMO + 2 can be regarded as comparable and predominantly relevant. These three orbitals are plotted in Fig. 4, where sterically inaccessible sites hindered by the 1,2- or 1,3-adjacency of CF3groups are marked by a gray filling. Although, without steric constrains, the most probable Bingel addition site would be site f, steric factors clearly favor the least eclipsed bond a, which exhibits significant contributions to LUMO + 1 and LUMO + 2. Thus, orbital considerations, again, bring us to the experimental structure of C70(CF3)10[C(CO2Et)2].


Localization of LUMO, LUMO + 1 and LUMO + 2 in C70(CF3)10 (orbitals are shown as pairs of black and white circles, the most important sites being marked with letters). The gray filling denotes those regions of the molecule where addition is sterically hindered by CF3groups (the latter are depicted as triangles of circles).
Fig. 4 Localization of LUMO, LUMO + 1 and LUMO + 2 in C70(CF3)10 (orbitals are shown as pairs of black and white circles, the most important sites being marked with letters). The gray filling denotes those regions of the molecule where addition is sterically hindered by CF3groups (the latter are depicted as triangles of circles).

Conclusions

Polytrifluoromethylated fullerenep7mp-C70(CF3)10 has been demonstrated to be prone to enter a standard Bingel reaction with diethyl bromomalonate to afford a single isomer of C70(CF3)10[C(CO2Et)2]. The regioselectivity of the reaction is governed by the addition motif of the CF3groupsvia the energetic and orbital reactivity differences of various fullerene sites, induced by the said groups. The reported compound provides the first example of the organic modification of trifluoromethylated fullerenes and thus contributes to the poorly explored research direction of utilizing the enhanced acceptor properties of halogenated fullerene derivatives for prospective photochemical and photoelectric systems. It is currently planned by our group to expand such organic functionalization studies over broader classes of acceptor derivatives, as well as over other types of fullerene reactions.

Acknowledgements

This work was supported by the Federal Agency for Science and Innovations (02.120.11.80.2), the Russian Foundation for Basic Research (06-03-329420 and 05-03-04006), INTAS (YS-2004-83-3316) and the Deutsche Forschungsgemeinschaft (Ke 489/26-1).

References

  1. I. E. Kareev, N. B. Shustova, I. V. Kuvychko, S. F. Lebedkin, S. M. Miller, O. P. Anderson, A. A. Popov, S. H. Strauss and O. V. Boltalina, J. Am. Chem. Soc., 2006, 128, 12268 CrossRef CAS.
  2. S. M. Avdoshenko, A. A. Goryunkov, I. N. Ioffe, D. V. Ignat’eva, L. N. Sidorov, P. Pattison, E. Kemnitz and S. I. Troyanov, Chem. Commun., 2006, 2463 RSC.
  3. A. A. Goryunkov, E. I. Dorozhkin, N. B. Tamm, D. V. Ignat’eva, S. M. Avdoshenko, L. N. Sidorov and S. I. Troyanov, Mendeleev Commun., 2007, 110 Search PubMed.
  4. S. I. Troyanov, A. A. Goryunkov, E. I. Dorozhkin, D. V. Ignat’eva, N. B. Tamm, S. M. Avdoshenko, I. N. Ioffe, V. Yu. Markov, L. N. Sidorov, K. Scheurel and E. Kemnitz, J. Fluorine Chem., 2007, 128, 545 CrossRef CAS.
  5. E. I. Dorozhkin, D. V. Ignat’eva, N. B. Tamm, A. A. Goryunkov, P. A. Khavrel, I. N. Ioffe, A. A. Popov, I. V. Kuvychko, A. V. Streletskiy, V. Yu. Markov, J. Spandal, S. H. Strauss and O. V. Boltalina, Chem.–Eur. J., 2006, 12, 3876 CrossRef.
  6. E. I. Dorozhkin, A. A. Goryunkov, I. N. Ioffe, S. M. Avdoshenko, V. Yu. Markov, N. B. Tamm, D. V. Ignat’eva, L. N. Sidorov and S. I. Troyanov, Eur. J. Org. Chem., 2007 DOI:10.1002/ejoc.200700487.
  7. V. Yu. Markov, V. E. Aleshina, A. Y. Borschevskiy, R. V. Khatymov, R. F. Tuktarov, A. V. Pogulay, A. L. Maximov, S. V. Kardashev, I. N. Ioffe, S. M. Avdoshenko, E. I. Dorozhkin, A. A. Goryunkov, D. V. Ignat’eva, N. I. Gruzinskaya and L. N. Sidorov, Int. J. Mass Spectrom., 2006, 251, 16 CrossRef CAS.
  8. A. A. Popov, J. Tarábek, I. E. Kareev, S. F. Lebedkin, S. H. Strauss, O. V. Boltalina and L. Dunsch, J. Phys. Chem. A, 2005, 109, 9709 CrossRef CAS.
  9. C. Bingel, Chem. Ber., 1993, 126, 1957 CrossRef CAS.
  10. X. Camps and A. Hirsch, J. Chem. Soc., Perkin Trans. 1, 1997, 1595 RSC.
  11. I. E. Kareev, I. V. Kuvychko, A. A. Popov, S. F. Lebedkin, S. M. Miller, O. P. Anderson, S. H. Strauss and O. V. Boltalina, Angew. Chem., Int. Ed., 2005, 44, 7984 CrossRef CAS.
  12. W. Powell, F. Cozzi, G. Moss, C. C. Thilgen, R. J. Hwu and A. Yerin, Pure Appl. Chem., 2002, 74, 629 CrossRef CAS.
  13. http://classic.chem.msu.su/gran/gamess/index.html; A. A. Granovsky, M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. J. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M. Dupuis and J. A. Montgomery, J. Comput. Chem., 1993, 14, 1347 Search PubMed.
  14. D. N. Laikov, Chem. Phys. Lett., 1997, 281, 151 CrossRef CAS.
  15. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865 CrossRef CAS.
  16. I. E. Kareev, I. V. Kuvychko, S. F. Lebedkin, S. M. Miller, O. P. Anderson, K. Seppelt, S. H. Strauss and O. V. Boltalina, J. Am. Chem. Soc., 2005, 127, 8362 CrossRef CAS.
  17. D. V. Ignat’eva, A. A. Goryunkov, N. B. Tamm, I. N. Ioffe, S. M. Avdoshenko, L. N. Sidorov, A. Dimitrov, E. Kemnitz and S. I. Troyanov, Chem. Commun., 2006, 1778 RSC.
  18. A. A. Goryunkov, D. V. Ignat’eva, N. B. Tamm, N. N. Moiseeva, I. N. Ioffe, S. M. Avdoshenko, V. Yu. Markov, L. N. Sidorov, E. Kemnitz and S. I. Troyanov, Eur. J. Org. Chem., 2006, 2508 CrossRef CAS.
  19. X. W. Wei, A. D. Darwish, O. V. Boltalina, P. B. Hitchcock, J. M. Street and R. Taylor, Angew. Chem., Int. Ed., 2001, 40, 2989 CrossRef CAS.
  20. R. C. Haddon, Science, 1993, 261, 1545 CAS.
  21. K. H. Lee, C. Lee, J. Kang, S. S. Park, J. Lee, S. K. Lee and D. K. Bohme, J. Phys. Chem. A, 2006, 110, 11730 CrossRef CAS.
  22. F. Diederich, L. Isaacs and D. Philp, Chem. Soc. Rev., 1994, 23, 243 RSC.
  23. A. Herrmann, M. Rüttimann, C. Thilgen and F. Diederich, Helv. Chim. Acta, 1995, 78, 1673 CrossRef CAS.
  24. M. J. van Eis, P. Seiler, L. A. Muslinkina, M. Badertscher, E. Pretsch, F. Diederich, R. J. Alvarado, L. Echegoyen and I. P. Núñez, Helv. Chim. Acta, 2002, 85, 2009 CrossRef CAS.
  25. A. Hirsch, I. Lamparth, T. Grösser and H. R. Karfunkel, J. Am. Chem. Soc., 1994, 116, 9385 CrossRef CAS.

Footnote

† Electronic supplementary information (ESI) available: Calculation details and relative energies at the AM1 and DFT levels of theory. See DOI: 10.1039/b704924b.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
Click here to see how this site uses Cookies. View our privacy policy here.