Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://chem.msu.ru/rus/lab/phys/cryschem/liter/pap_18.pdf Äàòà èçìåíåíèÿ: Fri Jun 28 10:18:00 2002 Äàòà èíäåêñèðîâàíèÿ: Mon Apr 11 05:27:49 2016 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 27

COMMUNICATIONS
C60F18 , a Flattened Fullerene : Alias a Hexa-Substituted Benzene**
Ivan S. Neretin, Konstantin A. Lyssenko, Mikhail Yu. Antipin, Yuri L. Slovokhotov,* Olga V. Boltalina,* Pavel A. Troshin, Andrei Yu. Lukonin, Lev. N. Sidorov, and Roger Taylor*
The aromaticity of fullerenes[1] is restricted by the increase in (bond-shortening) strain within the pentagonal rings that arises from the degree of double-bond character they must acquire following electron delocalization.[2] Nevertheless, some degree of delocalization in the carbon cage exists, as shown by the sequential pattern of polyadditions to fullerenes. Addition to one 6þ6 ªdoubleº bond localizes the electrons in the remaining double bonds of the same hexagon and causes them to undergo preferential addition,[3] as exemplified by the products of hydrogenation,[4] and fluorination.[5] The strain caused by the presence of a delocalized double bond within a pentagon is diminished when any of the remaining three carbon atoms in the pentagon changes its hybridization from sp2 to sp3 ; this change facilitates delocalization in the adjacent hexagons. This principle was invoked when the tetrahedral (T) molecular structure for C60H36 was suggested[6] and accounts for the octahedral addition patterns to [60]fullerene,[7] for example, in the formation of [C60{Pt(PEt3)2}6].[8] In each case more-delocalized benzenoid rings are created. A novel D3 addition pattern resulting from the addition of tetramethylazomethine ylide to [60]fullerene was observed recently by Rubin and co-workers and is likewise a consequence of the creation of a combination of benzenoid and naphthalenoid rings.[9] Relatively few single-crystal X-ray structures have been determined for fullerenes, and only three are known for halogenofullerenes, namely, C60Br24 ,[10] C60Br8 , and C60Br6 .[11] Two X-ray structures have provided evidence for the additiondriven delocalization of electrons described above. The lengths of the 6þ6 and 6þ5 bonds are 1.39 and 1.42 , respectively, in the eight semi-benzenoid rings of C66(COOEt)12 , produced by the octahedral addition of six
[*] Dr. Yu. L. Slovokhotov, I. S. Neretin, K. A. Lyssenko, Prof. M. Yu. Antipin Institute of Organoelement Compounds, R.A.S. 28 Vavilov Street, Moscow 117813 (Russia) Fax : ( 7) 095-1355085 E-mail : slov@ineos.ac.ru Dr. O. V. Boltalina, P. A. Troshin, A. Yu. Lukonin, Prof. L. N. Sidorov Chemistry Department Moscow State University Moscow 119899 (Russia) Fax : ( 7) 095-9391240 E-mail : ovb@capital.ru Dr. R. Taylor School of Chemistry Physics & Environmental Sciences Sussex University, BN1 9QJ (UK) Fax : ( 44) 1273-677-196 E-mail : R.Taylor@sussex.ac.uk [**] This work was supported by the Royal Society, INTAS, the Russian Foundation for Fundamental Research (grant no. 99-03-32810), and the Russian Programme, ªFullerenes and Atomic Clustersº.
Angew. Chem. Int. Ed. 2000, 39, No. 18

diethyl malonyl moieties to [60]fullerene. Similar values have been obtained for a related tethered hexakis adduct.[12] The alternation is thus less than in [60]fullerene itself (about 1.40 and 1.45 ).[13] A few fully characterized fluorofullerenes have been produced, for example, C60F48 ,[14] C60F36 (both T and C3 isomers),[15] and C3v-C60F18 .[16] The high stability of the latter three derivatives is attributed to the presence of four-, three-, and one p-delocalized benzenoid rings, respectively. We now report the first single-crystal X-ray structure of a fluorinated fullerene, C60F18 (as a toluene solvate), which reveals that one half of the [60]fullerene cage is flattened (Figure 1). The

Figure 1. Two projections of C60F18 (ORTEP).

hexagonal face at the center of the fluorinated crown has equal bond lengths, which shows it is fully aromatic. It is the first truly benzenoid ring identified on a fullerene surface. The unique geometry of the molecule makes it also a hexasubstituted benzene (specifically here a ªpseudocyclophaneº with a nondistorted fullerene hemisphere opposite to the arene ring). The single-crystal X-ray crystallographic study confirms the conclusions reached previously from the 19F NMR spectrum,[16] namely, that all the fluorine atoms are bound to one hemisphere of the [60]fullerene cage. The ªupperº sixmembered ring of sp2 carbon atoms is isolated from the residual molecular p system by a ªbeltº of sp3-hybridized carbon atoms. The Schlegel diagram (Figure 2) shows the positions of the fluorine atoms, together with the notation for the symmetry-independent CþF and CþC bonds and the conformations of the nonplanar polygons. A similar addition pattern was earlier predicted for C60H18 by semiempirical AM1 quantum chemistry calculations,[17] and confirmed by 1 H NMR spectroscopy.[18] Although such calculations (and indeed simple molecular models) indicate substantial cage flattening, the X-ray data (Figure 1) now reveal the actual extent to which this occurs. The distances (Di) of the C atoms from the center of the carbon cage,[19] and the spherical excesses (fi) of the ªsurfaceº environments of the carbon atoms were used, respectively, to characterize the radial atomic displacements and surface puckering in C60F18 . The latter fi parameter was defined as 3608 minus the sum of three C-C-C bond angles at the ith
1433-7851/00/3918-3273 $ 17.50+.50/0

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000

3273


COMMUNICATIONS
The pentagons and hexagons adjacent to the planar face have envelope and boat conformations, respectively. Their neighboring pentagons and hexagons are flattened envelopes and half-chairs, respectively (Figure 2). Single C(sp3)þC(sp3) bonds in the fluorinated belt are elongated to 1.623 (bonds c) and 1.672 (bonds e). Such elongation is typical for fluoro-substituted hydrocarbons and polycyclic fluorocarbons.[22] The lengths of four symmetrically independent CþF bonds (1 ± 4) differ slightly, but significantly (1.361 ± 1.396 , Table 1). The averaged C-C-F bond angles have normal tetrahedral values of 106.9 ± 111.38 with a slight asymmetry relative to the C60 framework ; this observation reflects the stronger asymmetry of their opposite C-C-C angles in the puckered five-membered (100.4 ± 109.98) and six-membered rings (108.3 ± 122.68), composed of sp2 and sp3 carbon atoms. The alterations in the cage CþC bonds in C60F18 relative to those in C60 (mapped[23] in Figure 3) are : a) a small shortening in the aromatic ring from the standard value of 1.399 in arenes ;[24] b) a strong elongation of the C(sp3)þC(sp3) bonds in the Figure 3. Shortening (filled lines) and elongation (dotted fluorinated belt ; c) a smaller lines) of the CþC bonds in but significant (0.01 ± 0.03 ) C60F18 , averaged within a C3v 2þ 3 shortening of the C(sp ) C(sp ) symmetry, from their standard bonds in the adjacent ªflatlengths of 1.397 (sp2þsp2), 1.507 (sp2þsp3), and 1.530 tenedº belt ; and d) no signifi(sp3þsp3 ; see ref. [24] ). The cant bond redistribution in the thickness of the bonds is proªlowerº hemisphere. The distriportional to the extent of their bution of the charge mostly in shortening/elongation ; fluorithe vicinity of the fluorinated nated sp3 carbon atoms are marked by black circles. Bond belt agrees with the generally lengths in the opposite hemiobserved poor relay of electronsphere (not shown) coincide ic effects through the [60]fullwithin 0.02 with those in C60 . erene framework, which is necessitated by the need to avoid the energetically unfavorable reduction in bond lengths in the pentagons.[25] Thus, for example, the unpaired electrons resulting from radical additions are highly localized,[26] and electron addition produces only minor changes in the Raman spectrum since the general cage structure is retained.[27] The C60F18 molecules are linked by inversion centers in antiparallel zigzag chains along the [101] direction of the crystal lattice (Figure 4). The planar ªupperº arene face of each molecule in these chains has normal van der Waals contacts with the ªlowerº fullerene hemisphere of the neighboring chain, with the shortest center-to-center distance in the chain being 9.4 . The shortest intermolecular CF ´ ´ ´ C(cage) contacts between neighboring chains are 2.75 and 2.92 (from the type 4 CþF bond to the i-type C(sp2)þC(sp2) bond). From the observation of a lack of elongation of these bonds and the slightly lower spherical excess fi values of the surrounding C(sp2) (3.5 and 4.48) and C(sp3) (26.48) atoms we suggest that these shortened F ´ ´ ´ C contacts are forced and therefore
1433-7851/00/3918-3274 $ 17.50+.50/0 Angew. Chem. Int. Ed. 2000, 39, No. 18

Figure 2. Schlegel diagram for C60F18 showing the symmetry-independent fluorine atoms (1 ± 4), CþC bonds (a ± s) and ring conformations (see text) within the idealized C3v symmetry.

carbon atom. Standard values in [60]fullerene itself are D 3.57 and f 128 for all atoms. Using these parameters four regions on the surface of the distorted cage may be distinguished in C60F18 : 1) a planar six-membered ªupperº cycle with D 2.97 and f < 0.18 ; 2) a puckered fluorinated belt of eighteen C(sp3) atoms (3.8 < D < 4.0 ; 26.4 < f < 29.98) ; 3) a flattened equatorial belt of twelve C(sp2) atoms bonded directly to C(sp3) atoms (3.5 < f < 5.28) ; 4) a residual bottom hemisphere made up of 24 C(sp2) atoms which is close to the normal fullerene geometry (7.8 < f < 13.38). The Di parameter, which is strongly dependent on the choice of the molecule center, is not informative in the last two regions because of the small radial distortions (2 ± 3 pm).[19] Table 1 shows the bond lengths in C60F18 averaged within the ideal C3v symmetry (the lengths of symmetry-related bonds mostly agree within a few pm). The ªupperº hexagonal ring, planar within 0.003 , shows no alternation in the 6þ5
Table 1. Bond lengths [] in C60F18 (for notations, see Figure 2). CþC a,a' b c d e f g h i 1.372 1.476 1.623 1.557 1.672 1.558 1.500 1.524 1.363 j k l m n p q r s 1.428 1.435 1.386 1.437 1.387 1.436 1.453 1.387 1.447 1 2 3 4 CþF 1.396 1.361 1.385 1.377

and 6þ6 carbon ± carbon bond lengths and is thus fully aromatic. Neighboring C(sp3) atoms deviate from the mean plane of this ring towards the rest of the molecule by 0.055 ± 0.075 . The average CþC bond lengths in the arene ring (1.372 ) and those attached to it (1.476 ) are noticeably shorter than in either the related fluorocarbon species, perfluorinated indacene C15F18 (1.389 and 1.499 ),[20] or hexa(trifluoromethyl)benzene (1.405 and 1.541 ).[21] The bond lengths of the ring are essentially the same as the average CSD value in hexafluorobenzene (1.368 ), which is indicative of the strong electron-withdrawing effect of the eighteen fluorine atoms in the ªupperº hemisphere. 3274
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000


COMMUNICATIONS
[1] R. C. Haddon, Science 1993, 261, 1545. [2] R. Taylor, Tetrahedron Lett. 1991, 3731. [3] R. Taylor, Lecture Notes on Fullerene Chemistry : A Handbook for Chemists, Imperial College Press, London, 1999, chap. 3. [4] A. G. Avent, A. D. Darwish, D. K Heimbach, H. W. Kroto, M. F. Meidine, J. P. Parsons, C. Remars, R. Roers, O. Ohashi, R. Taylor, D. R. M. Walton, J. Chem. Soc. Perkin Trans. 2 1994, 15 ; C. C. Henderson, C. M. Rohlfing, R. A. Assink, P. A. Cahill, Angew. Chem. 1994, 109, 793 ; Angew. Chem. Int. Ed. Engl. 1994, 33, 786. [5] O. V. Boltalina, A. Yu. Lukonin, A. G. Avent, J. M. Street, R. Taylor, J. Chem. Soc. Perkin Trans. 2 2000, 683. [6] R. Taylor, Philos. Trans. R. Soc. London A 1993, 343, 87. [7] R. Taylor, J. Chem. Soc. Perkin Trans. 2 1992, 1667; P. W. Fowler, D. J. Collins, S. J. Austin, J. Chem. Soc. Perkin Trans 2 1993, 275. [8] P. J. Fagan, J. C. Calabrese, B. Malone, Science 1991, 252, 1160. [9] G. Schick, M. Levitus, L. D. Kvetko, B. A. Johnson, I. Lamparth, R. Lunkwitz, B. Ma, S. I. Khan, M. A. Garcia-Garibay, Y. Rubin, J. Am. Chem. Soc. 1999, 121, 3246 ; see also ref. [3] . [10] F. N. Tebbe, R. L. Harlow, D. B. Chase, D. L. Thorn, G. C. Campbell, J. C. Calabrese, N. Herron, R. J. Young, E. Wasserman, Science 1992, 256, 822. [11] P. R. Birkett, P. B. Hitchcock, H. W. Kroto, R. Taylor, D. R. M. Walton, Nature 1992, 357, 479. [12] I. Lamparth, C. Maichle-MÆssmer, A. Hirsch, Angew Chem. 1995, 107, 1755 ; Angew. Chem. Int. Ed. Engl. 1995, 34, 1607; P. Seiler, L. Isaacs, F. Diederich, Helv. Chim. Acta 1996, 79, 1047. [13] C. S. Yannoni, P. P. Bernier, D. S. Bethune, G. Meijer, J. R. Salem, J. Am. Chem. Soc. 1991, 113, 437; W. I. F. David, R. M. Ibberson, J. C. Matthewman, K. Prassides, T. J. S. Dennis, J. P. Hare, H. W. Kroto, R. Taylor, D. R. M. Walton, Nature 1991, 353, 147; K. Hedberg, L. Hedberg, D. S. Bethune, C. A. Brown, H. C. Dorn, R. D. Johnson, M. de Vries, Science 1991, 254, 410. [14] A. A. Gakh, A. A. Tuinman, J. L. Adcock, R. A. Sachleben, R. A. Compton, J. Am. Chem. Soc. 1994, 116, 819 ; O. V. Boltalina, L. N. Sidorov, V. F. Bagryantsev, V. A. Seredenko, A. S. Zapolskii, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1996, 2275. [15] O. V. Boltalina, J. M. Street, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1998, 649 ; O. V. Boltalina, M. BÝhl, A. Khong, M. Saunders, J. M. Street, R. Taylor, J. Chem. Soc. Perkin Trans 2. 1999, 1475. [16] O. V. Boltalina, V. Yu. Markov, R. Taylor, M. P. Waugh, Chem. Commun. 1996, 2549. [17] B. W. Clare, D. L. Kepert, J. Mol. Struct. (THEOCHEM) 1994, 303, 1. [18] A. D. Darwish, A. G. Avent, R. Taylor, D. R. M. Walton, J. Chem. Soc. Perkin Trans. 2 1996, 2051. [19] The position of the cage centre in C60F18 was calculated by two methods : 1) as the center of an ideal C60 moiety fitted to the experimental one by the least-squares minimization of the discrepancies in the respective atomic positions ; and 2) as the center of C60 fitted with the unaddended hemisphere only. [20] F. W. B. Einstein, T. Jones, Acta Crystallogr. Sect. B 1982, 38, 328. [21] M. H. Couldwell, B. R. Penfold, J. Cryst. Mol. Struct. 1976, 6, 59. [22] B. M. Neil, J. Mol. Struct. (THEOCHEM) 1976, 31, 65 ; R. T. Thummel, J. O. Korp, J. Bernal, J. Am. Chem. Soc. 1977, 99, 6916 ; L. Golic, L. Leban, Cryst Struct. Commun. 1978, 53 ; V. R. Polischuk, M. Yu. Antipin, V. I. Bakhmutov, Dokl. Akad. Nauk SSSR 1979, 249, 1125. [23] Yu. L. Slovokhotov, I. V. Moskaleva, V. I. Shilnikov, E. F. Valeev, Yu. N. Novikov, A. I. Yanovsky, Yu. T. Struchkov, Mol. Mat. 1996, 8, 117. [24] K. Tamagawa, T. Iijima, M. Kimura, J. Mol. Struct. 1976, 30, 243 ; F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1987, S1. [25] R. Taylor, Tetrahedron Lett. 1991, 3731. [26] P. J. Krusic, E. Wasserman, P. N. Keizer, J. R. Morton, K. F. Preston, Science 1991, 254, 1183. [27] M. L. McGlashen, M. E. Blackwood, T. G. Spiro, J. Am. Chem. Soc. 1993, 115, 2074. [28] J. H. Williams, Acc. Chem. Res. 1993, 26, 593 ; U. H. F. Bunz, V. Enkelmann, Chem. Eur. J. 1999, 5, 263, and references therein. [29] MNDO and density functional calculations predict the values of 6.9 and 6.3 D, respectively : J. P. B. Sandall, M. Heggie, private communication to R. Taylor.
1433-7851/00/3918-3275 $ 17.50+.50/0

Figure 4. Molecules of C60F18 in the crystal lattice.

repulsive ones. Attractive electrostatic interactions between the strongly polarized C60F18 molecules probably account for the closeness of the molecules and the unusually high degree of molecular ordering in C60F18 ´C6H5CH3 . Such electrostatic forces, expressed in multipole interactions, are believed to be a reason for a weak specific bonding to donors in molecular complexes of hexafluorobenzene.[28] The distorted geometry of the C60F18 molecule is quantitatively reproduced by quantum chemistry calculations, even on a semiempirical PM-3 level. Ab initio HF calculations (GAUSSIAN-94, 3-21G* and 6-31G basis sets, geometry optimized) convergently predict a substantial polarization of the CþF bonds withcharges of þ 0.34 to þ 0.39 on the fluorine atoms and charges of 0.25 to 0.40 on the sp3 carbon atoms. The calculated charges are mainly located in the ªfluorinatedº belt of the C60 moiety, which is in agreement with the experimental redistribution of the CþC bond lengths (Figure 3). The unique arene ring carries a small positive charge per carbon atom ( 0.04 in 3-21G* and 0.03 in 6-31G). Both HF approximations used ascribe a large dipole moment to the molecule (12.4 D in 3-21G* and 15.7 D in 6-31G) ; however, smaller values of 6 ± 7 D have been predicted independently.[29] DFT calculations (B3LYP/6-31G) give similar values to those obtained by HF, but with the CþF bond distance overestimated by 0.04 , a slightly more uniform charge distribution, and a predicted dipole moment of 13.3 D. The very strongly electron-deficient planar region suggests that this molecule may find important applications in intramolecular electron-transfer complexes, whilst h6 coordination with transition metal atoms may also be possible. These aspects are to be investigated.

Experimental Section
C60F18 was prepared and purified as described previously.[16] Light-yellow diamond-shaped plates of C60F18 ´C6H5Me with a good X-ray quality were grown from toluene. The structure was determined by X-ray diffraction.[30] Errors (esd) for the CþF and CþC bond lengths were 0.002 ± 0.003 and 0.003 ± 0.004 , respectively. Received : May 4, 2000 [Z 15081]
Angew. Chem. Int. Ed. 2000, 39, No. 18

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000

3275


COMMUNICATIONS
[30] Crystal data for C60F18 ´(C7H8), Mr 1154, monoclinic, space group P2l/n, a 11.532(2), b 21.501(3), c 16.261(2) , b 101.798(5)8, V 3947 3, Z 4, m(MoKa) 0.17 mmþ1, T 100 K. 49 263 Reflections measured (Siemens SMART diffractometer, 2qmax 608) 11 565 unique, refined on F 2 using SHELXTL, R1 0.049 (for 6426 reflections with I > 2s > (I), wR2 0.118 (for all reflections). Crystallographic data (excluding structure factors) for the structure reported in this paper have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication no. CCDC-142291. Copies of the data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax : ( 44) 1223336-033 ; e-mail : deposit@ccdc.cam.ac.uk).

was to explore the possibility of preparing porous cerium silicates with potential for use in optoelectronics. The crystal structure of montregianite (and AV-5) consists of two different types of layers alternating along the [010] direction (Figure 1): a) a double silicate sheet, in which the

The First Microporous Framework Cerium Silicate**
ä Joao Rocha,* Paula Ferreira, LuÌs D. Carlos, and Artur Ferreira
Recently, much research has been carried out aimed at preparing inorganic microporous zeotype solids composed of interconnected octahedral- and tetrahedral-oxide polyhedra. One such family of materials of considerable interest embraces microporous titanosilicates and their derivatives which contain TiIV usually in octahedral coordination.[1, 2] As part of a systematic search for novel microporous framework inorganic materials we have prepared (among others) AV-1, a synthetic analogue of the rare sodium yttrium silicate mineral montregianite (Na4K2Y2Si16O38 ´10 H2O).[3, 4] In nature, montregianite usually occurs withframework substitution of some yttrium by cerium and this raised the possibility of obtaining a purely cereous form of AV-1. Here we report the synthesis and characterization of AV-5, a microporous framework sodium cerium silicate with the structure of montregianite. It is well known that CeIII-containing materials present potential for applications in optoelectronics. For example, CeIII in a LiYF6 matrix has been shown to lase.[5] This system may find important applications since CeIII lasers emit in the high-energy region of the spectrum (blue or UV) and they exhibit tunability due to the inherently broad emission lines. Glass-supported cerium is used in Faraday rotators, which protect lasers from back-reflected light.[6] As for other lanthanides, cerium has been used in transducers designed to detect or modify IR, UV, X-ray, and gamma radiation.[6] The luminescence of CeIII/silica systems prepared by sol ± gel methods has also been studied.[7] Our main aim withthis work
[*] Prof. J. Rocha, Dr. P. Ferreira, A. Ferreira Department of Chemistry University of Aveiro 3810 Aveiro (Portugal) Fax : ( 351) 234-370084 E-mail : rocha@dq.ua.pt Prof. L. D. Carlos Department of Physics, University of Aveiro 3810 Aveiro (Portugal) [**] This work was supported by PRAXIS XXI, FEDER, and FCT.

Figure 1. Polyhedral representation of the montregianite structure : a) viewed along [100] showing the alternating octahedral sheet and double silicate sheet along [010] ; b) open octahedral sheet, viewed along [010] , consisting of Y and three different Na octahedra (unfilled octahedra).

single silicate sheet is of the apophyllite type with four- and eight-membered rings, and b) an open octahedral sheet composed of [YO6] (or [CeO6] in AV-5) and three distinct [NaO4(H2O)2] octahedra.[8] The layers are parallel to the (010) plane. The potassium ions are ten-coordinate and the six water molecules (not shown) are located within large channels formed by the planar eight-membered silicate rings. Figure 2 shows the experimental and simulated powder X-ray diffraction (XRD) patterns of AV-5. The differences observed are due to the presence of a small amount of an, asyet, unknown impurity. The unit cell parameters have been refined by assuming a monoclinic unit cell, space group P21/n, and cell dimensions a 9.6906(9), b 24.055(2), c 9.5759(8) , and b 93.683(7)8 and are similar to those reported for montregianite, dimensions a 9.512, b 23.956, c 9.617 , and b 93.858).[9] Presumably owing to the presence of paramagnetic CeIII centers we were unable to record a 29Si magic-angle spinning (MAS) NMR spectrum of AV-5. It is, however, possible to oxidize CeIII to CeIV by calcining AV-5 in air at 300 8C for 3 h.
1433-7851/00/3918-3276 $ 17.50+.50/0 Angew. Chem. Int. Ed. 2000, 39, No. 18

3276

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000