3642 J. Org. Chem.

1997, 62, 3642-3649

Fullerides of Pyrrolidine-Functionalized C60

Yongping Sun,† Tatiana Drovetskaya,† Robert D. Bolskar,† Robert Bau,†

Peter D. W. Boyd,*,‡ and Christopher A. Reed*,†
Departments of Chemistry, University of Southern California, Los Angeles, California 90089-0744, and
The University of Auckland, Private Bag 92019, Auckland 1, New Zealand

Received February 25, 1997X

The spectral characteristics of the fulleride(1-) anions of pyrrolidine-functionalized C60 derivatives

have been investigated in order to assess the degree to which the parent C60 moiety retains its
distinctive redox and chromophoric properties upon functionalization. Such comparison data are
necessary if C60 is to fulfill its anticipated role as a versatile electron-accepting chromophore and
a multielectron reservoir. Three systems have been investigated: a simple, monomeric, N-
methylpyrrolidine derivative, 1, a pyrrolidine-linked tetraphenylporphyrin/C60 dyad, 2, and a pair
of bifullerenes, 3 and 4, with adjacent and remote dispositions of the C60 moieties, respectively.
Near infrared spectra of the fulleride(1-) ions of these derivatives are notably similar to C601-
with only small energy shifts of the band envelope. 1H and 13C NMR data on 11- are consistent
with unpaired spin density confined largely to the C60 moiety rather than delocalized onto the
pyrrolidine functionality. The EPR spectra of all of the pyrrolidine-functionalized fulleride(1-)
species are characterized by sharp doublets (∆H ) 1-2 G) and g values less than the free electron
value. Unlike the EPR spectrum of C601-, there is little temperature dependence of the line width.
EPR evidence for ball-to-ball spin-spin interactions are observed in the fulleride of 3 but not in 4
or in the CuII/C601- metalated porphyrin dyad. The X-ray crystal structure of the porphyrin-C60
dyad 2 has been determined. It is the first X-ray structural characterization of a pyrrolidine-
functionalized C60 species and confirms that the azamethine ylide addition has occurred across a
6:6 ring juncture. The crystal packing of the dyad reveals an intermolecular interaction of the C60
ball nestled in remarkably close approach to the porphyrin plane. The closest approach of a fullerene
carbon atom to the mean plane of the porphyrin is ca. 2.75 Å. This interaction is notably similar
to that envisaged for porphyrin-functionalized chromatographic supports used to separate fullerenes.
A novel donor/acceptor relationship is proposed to account for this close interaction.

Introduction lence, and hence, as a fulleride (C60n-), it becomes a very

attractive electron reservoir. Photoinduced charge sepa-
As a redox-active chromophore, C60 may eventually ration in dyads with a C60 moiety acting as the electron
rival ferrocene and porphyrins in the versatility of its acceptor is currently receiving considerable attention,8
application to processes involving charge separation and and an electroactive polymer based on (C60O)n has
storage. C60 is functionalizable1 and thus is readily recently been described.4j The hyperpolarizability of C601-
tagged to itself,2 to other chromophores,3 to polymers,4 is significantly greater than that of C60, suggesting that
and to biomolecules.5 It has strong electronic absorptions fullerides rather than fullerenes will have superior
across the UV and visible portions of the electromagnetic nonlinear optical applications.9
spectrum and associated with this are distinctive photo- A fundamental question underlying the application of
physical properties.6 C60 has a unique redox chemistry C60 to problems of charge separation and storage is the
in solution, adding up to six electrons with nearly equal extent to which the redox properties of C60 are retained
spacings of potential in successive reductions.7 This when it is functionalized. How are its redox processes
electronegativity makes it an electron acceptor par excel- and reduction potentials affected?10 Are the reduction
electrons and unpaired spin localized on the ball or are
† University of Southern California.
‡
they delocalized onto the connecting functionality? How
The University of Auckland.
X Abstract published in Advance ACS Abstracts, May 1, 1997. are electron transfer mechanisms affected by function-
(1) (a) Diederich, F.; Thilgen, C. Science 1996, 271, 317-323. (b) alization? Is the NIR region of the absorption spectrum
Hirsch, A. The Chemistry of the Fullerenes; Georg Thieme Verlag: as distinctive and useful for functionalized fullerides as
Stuttgart, 1994. (c) Taylor, R.; Walton, D. R. M. Nature (London) 1993,
363, 685-693. it is for C60n- fullerides? Does the proximity of fullerene
(2) (a) Suzuki, T; Li, Q.; Khemani, K. C.; Wudl, F. J. Am. Chem. units lead to redox-activated ball-to-ball chemical reac-
Soc. 1992, 114, 7300-7301. (b) Belik, P.; Gügel, A.; Spickermann, J.; tivity? What is the nature of ball-to-ball spin coupling?
Müllen, K. Angew. Chem., Int. Ed. Engl. 1993, 32, 78-80. (c) Paquette,
L. A.; Graham, R. J. J. Org. Chem. 1995, 60 , 2958-2959. (d) Smith, These are some of the questions which must be answered
A. B., III; Tokuyama, H.; Strongin, R. M.; Furst, G. T.; Romanow, W. before fullerenes can realize their full potential as unique
J.; Chait, B. T.; Mirza, U. A.; Haller, I. J. Am. Chem. Soc. 1995, 117, and useful redox-active chromophores.
9359-9360. (e) Kraus, A.; Gügel, A.; Belik, P.; Walter, M.; Müllen, K.
Tetrahedron 1995, 51, 9927-9940. (f) Lebedkin, S.; Ballenweg, S.; Our particular interest is centered on the electron
Gross, J.; Taylor, R.; Krätschmer, W. Tetrahedron Lett. 1995, 36, 4971- reservoir possibilities of C60. We have recently linked C60
4974. (g) Osawa, S.; Osawa, E.; Harada, M. J. Org. Chem. 1996, 61,
257-265. (h) Timmerman, P.; Witschel, L. E.; Diederich, F.; Boudon, covalently to tetraphenylporphyrin3b with the idea that
C.; Gisselbrecht, J.-P.; Gross, M. Helv. Chim. Acta 1996, 79, 6-20. (i) fulleride substituents on a metalloporphyrin will promote
Paquette, L. A.; Trego, W. E. J. Chem. Soc., Chem. Commun. 1996, multielectron reduction chemistry at an axial ligand
419-420. (j) Komatsu, K.; Takimoto, N.; Murata, Y.; Wan, T. S. M.;
Wong, T. Tetrahedron Lett. 1996, 37, 6153-6156. (k) Irngartinger, H.; binding site. We have chosen pyrrolidine derivatization
Weber, A. Tetrahedron Lett. 1996, 37, 4137-4140. of C60 because, of the known methodologies for function-
S0022-3263(97)00357-5 CCC: $14.00 © 1997 American Chemical Society
Fullerides of Pyrrolidine-Functionalized C60 J. Org. Chem., Vol. 62, No. 11, 1997 3643

alizing C60,1 the addition of an azamethine ylide across

a 6:6 ring juncture is synthetically versatile and the
derivatized products have excellent thermal stability.11
In addition, the reduction potentials of pyrrolidine-
derivatized C60 moieties typically remain fairly close to
those of the corresponding reductions of free C60.3f,10b For
example, Prato and co-workers3f have shown that the first
two reduction potentials of the N-methylpyrrolidine
derivative 1 are cathodically shifted by 110 mV relative
to that of C60. A total of five reductions can be observed
in the voltammetry of 1 with increasing differences in
potential relative to that of C60 as the negative charge tion that C60 retains its fulleride-like properties in its
increases. The retention of a similar number of voltam- reduced derivatives.10 In this paper, we examine this
etric reductions with relatively similar values for their expectation in more detail by isolating the fulleride(1-)
reduction potentials is the basis of the general expecta- ion of 1 for NMR, EPR and NIR characterization.
As a prototype for studying spin-spin coupling phe-
(3) Porphyrins: (a) Liddell, P. A.; Sumida, J. P.; MacPherson, A.
N.; Noss, L.; Seely, G. R.; Clark, K. N.; Moore, A. L.; Moore, T. A.; nomena, we have put a paramagnetic metal ion (Cu2+)
Gust, D. Photochem. Photobiol. 1994, 60, 537-541. (b) Drovetskaya, into the porphyrin-C60 dyad 2 and used EPR and NIR
T.; Reed, C. A.; Boyd, P. D. W. Tetrahedron Lett. 1995, 36, 7971-7974. spectroscopy to probe the interaction between copper(II)
(c) Imahori, H.; Hagiwara, K.; Akiyama, T.; Taniguchi, S.; Okada, T.;
Sakata, Y. Chem. Lett. 1995, 265-266. (d) Imahori, H.; Sakata, Y. and the appended fulleride(1-) radical anion.
Chem. Lett. 1996, 199-200. (e) Ranasinghe, M. G.; Oliver, A. M.;
Rothenfluh, D. F.; Salek, A.; Paddon-Row, M. N. Tetrahedron Lett.
1996, 37, 4997-4800. Ferrocene: (f) Maggini, M.; Karlsson, A.;
Scorrano, G.; Sandonà, G.; Farnia, G.; Prato, M. J. Chem. Soc., Chem.
Commun. 1994, 589-590. (g) Prato, M.; Maggini, M.; Giaometti, C.;
Scorrano, G.; Sandonà, G.; Farnia, G. Tetrahedron 1996, 52, 5221-
5234. Phthalocyanine: (h) Linssen, T. G.; Dürr, K.; Hanack, M.; Hirsch,
A. J. Chem. Soc., Chem. Commun. 1995, 103-104. Tris-
(bipyridine)ruthenium: (i) Maggini, M.; Donò, A.; Scorrano, G.; Prato,
M. J. Chem. Soc., Chem. Commun. 1995, 845-846. (j) Sariciftci, N.
S.; Wudl, F.; Heeger, A. J.; Maggini, M.; Scorrano, G.; Prato, M.;
Bourassa, J.; Ford, P. C. Chem. Phys. Lett. 1995, 247, 510-514. (k)
Armspach, D.; Constable, E. C.; Diederich, F; Housecroft, C. E.;
Nierengarten, J.-F. Chem. Commun. 1996, 2009-2010. Electron-rich
moieties: (l) Khan, S. I.; Oliver, A. M.; Paddon-Row, M. N.; Rubin, Y.
J. Am. Chem. Soc. 1993, 115, 4919-4920. (m) Diederich, F.; Jonas,
U.; Gramlich, V.; Herrmann, A.; Ringsdorf, H.; Thilgen, C. Helv. Chim.
Acta 1993, 76, 2445-2453. (n) Williams, R. M.; Zwier, J. M.; Verhoeven,
J. W. J. Am. Chem. Soc. 1995, 117, 4093-4099. (o) Imahori, H.;
Cardosa, S.; Tatman, D.; Lin, S.; Macpherson, A. N.; Noss, L.; Seely, We also report the synthesis and characterization of
G. R.; Sereno, L.; Chessa de Silber, J.; Moore, T. A.; Moore, A. L.; Gust,
D. Photochem. Photobiol. 1995, 62, 1009-1014. (p) Diederich, F.; bifullerenes 3 and 4, linked meta and para to benzene
Dietrich-Buchecker, C.; Nierengarten, J.-F.; Sauvage, J.-P. J. Chem. via pyrrolidine amides. Molecular modeling suggests
Soc., Chem. Commun. 1995, 781-782. (q) Warrener, R. N.; Elsey, G. that the C60 moieties in 3 are held in van der Waals
M.; Houghton, M. A. Ibid. 1995, 1417-1418. (r) Murata, Y.; Komatsu,
K.; Wan, T. S. M. Tetrahedron Lett. 1996, 37, 7061-7064. (s) Fernan- contact whereas those in 4 are held remote from each
dez-Paniagua, U. M.; Illescas, B. M.; Martin, N.; Scoane, C. J. Chem. other.
Soc., Perkins Trans. 1 1996, 1077-1079. (t) Lawson, J. M.; Oliver, A.
M.; Rothenfluh, D. F.; An, Y.-Z.; Ellis, G. A.; Ranasinghe, M. G.; Khan, The idea behind the synthesis of 3 was to probe
S. I.; Franz, A. G.; Ganapathi, P. S.; Shephard, M. J.; Paddon-Row, potential ball-to-ball chemical and spectroscopic interac-
M. N.; Rubin, Y. J. Org. Chem. 1996, 61, 5032-5054. (u) Williams, R. tions. The possibility of ball-to-ball addition chemistry
M.; Koeberg, M.; Lawson, J. M.; An, Y.-Z.; Rubin, Y.; Paddon-Row, M.
N.; Verhoeven, J. W. J. Org. Chem. 1996, 61, 5055-5062. (v) Martin, must be considered in light of the polymer12 and dimer13
N.; Sanchez, L.; Seoane, C.; Andreu, R.; Garin J.; Orduna, J. Tetra- phases of KC60 and RbC60. We also wanted to investigate
hedron Lett. 1996, 37, 5979-5982. Electron-poor moieties: (w) Shu, the EPR spectral characteristics of the anions of 1, 3, and
L. H.; Wang, G. W.; Wu, S. H.; Wu, H. M. Ibid. 1995, 367-368. (x)
Ohno, T.; Martin, N.; Knight, B.; Wudl, F; Suzuki, T.; Yu, H. J. Org. 4 to see if they would shed light on the the origin of
Chem. 1996, 61, 1306-1309. (y) Iyoda, M.; Sasaki, S.; Sultana, F.; unexplained peaks in the EPR spectra of C60n- fullerides.
Yoshida, M.; Kuwatani, Y.; Nagase, S. Tetrahedron Lett. 1996, 37,
7987-7988.
(4) (a) Loy, D. A.; Assink, R. A. J. Am. Chem. Soc. 1992, 114, 3977- (5) (a) Boutorine, A. S.; Tokuyama, H.; Takasugi, M.; Isobe, H.;
3978. (b) Sariciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Science Nakamura, E.; Hélène, C. Angew. Chem., Int. Ed. Engl. 1994, 33,
1992, 258, 1474-1476. (c) Shi, S.; Khemani, K. C.; Li, Q. C.; Wudl, F. 2462-2465. (b) Toniolo, C; Bianco, A.; Maggini, M.; Scorrano, G.; Prato,
J. Am. Chem. Soc. 1992, 114, 10656-10657. (d) Samulski, E. T.; M.; Marastoni, M.; Tomatis, R.; Spisani, S.; Palú, G.; Blair, E. D. J.
DeSimone, J. M.; Hunt, M. O., Jr.; Menceloglu, Y. Z.; Jarnagin, R. C.; Med. Chem. 1994, 37, 4558-4562. (c) Jensen, A. W.; Wilson, S. R.;
York, G. A.; Labat, K. B.; Wang, H. Chem. Mater. 1992, 4, 1153-1157. Schuster, D. I. Bioorg. Med. Chem. In press.
(e) Geckeler, K. E.; Hirsch, A. J. Am. Chem. Soc. 1993, 115, 3850- (6) Foote, C. S. in Top. Curr. Chem. 1994, 169, 347-363.
3851. (f) Patil, A. O.; Schrivre, G. W.; Carstensen, B.; Lundberg, R. D. (7) (a) Xie, Q.; Pèrez-Cordero, E.; Echegoyen, L. J. Am. Chem. Soc.
Polym. Bull. 1993, 30, 187-190. (g) Berrada, M.; Hashimoto, Y.; 1992, 114, 3978-3980. (b) Ohsawa, Y.; Saji, T. J. Chem. Soc., Chem.
Miyata, S. Chem. Mater. 1994, 6, 2023-2025. (h) Thomas, R. N. J. Commun. 1992, 781-782.
Polym. Sci. A 1994, 32, 2727-2737. (i) Geckeler, K. E. Trends Polym. (8) See refs 3a,c,j,m,t, and Kuciauski, D.; Lin, S.; Seely, G. R.; Moore,
Sci. 1994, 2, 355-360. (j) Fedurco, M.; Costa, D. A.; Balch, A. L.; A. L.; Moore, T. A.; Gust, D.; Drovetskaya, T.; Reed, C. A.; Boyd, P. D.
Fawcett, W. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 194-196. (k) W. J. Phys. Chem. 1996, 100, 15926-15932.
Camp, A. G.; Lary, A.; Ford, W. T. Macromolecules 1995, 28, 7959- (9) Lascola, R.; Wright, J. C. Chem. Phys. Lett. In press.
7961. (l) Zhang, N.; Schricker, S. R.; Wudl, F. Chem. Mater. 1995, 7, (10) (a) Wudl, F. Acc. Chem. Res. 1992, 25, 157-161. (b) Suzuki,
441-442. (m) Cao, T.; Webber, S. E. Macromolecules 1995, 28, 3741- T.; Muruyama, Y.; Akasaka, T.; Ando, W.; Kobayashi, K.; Nagase, S.
3743. (n) Bunker, C. E.; Lawson, G. E.; Sun, Y. P. Macromolecules J. Am. Chem. Soc. 1994, 116, 1359-1363.
1995, 28, 3744-3746. (o) Chen, Y.; Cai, R.-F.; Huang, Z.- E.; Bai, X.; (11) Maggini, M.; Scorrano, G.; Prato, M. J. Am. Chem. Soc. 1993,
Chen, S.-M.; Yan, X. M. Solid State Commun. 1996, 97, 239-242. (p) 115, 9798-9799.
Chen, Y.; Huang, Z.-E.; Cai, R.-F. J. Polym. Sci. B 1996, 34, 631-640. (12) Pekker, S.; Jánossy, A.; Mihaly, L.; Chauvet, O.; Carrard, M.;
(q) Liu, B.; Bunker, C. E.; Sun, Y.-P. Chem. Commun. 1996, 1241- Forró, L. Science 1994, 265, 1077-1078.
1242. (r) Cloutet, E.; Gnanou, Y.; Fillaut, J.-L; Astruc, D. Chem. (13) Zhu, Q.; Cox, D. E.; Fischer, J. E. Phys. Rev. B 1995, 51, 3966-
Commun. 1996, 1565-1566. 3969.
3644 J. Org. Chem., Vol. 62, No. 11, 1997 Sun et al.

NIR (benzonitrile) λmax() ) 1002 (7.8 × 103), 882 (3.6 × 103),

780 (3.0 × 103) nm.
Bifullerene 3. To a suspension of C60(CH2)2NCPh3 (150
mg, 0.149 mmol) in CH2Cl2 (30 mL) was added triflic acid (0.4
mL), and the mixture was stirred for 2 h. After the solution
volume was reduced, Et2O (10 mL) was added and the
precipitate was centrifuged and washed several times with
Et2O, benzene, and Et2O again. The brownish solid was
pumped to dryness, taken into the glovebox, and suspended
in o-dichlorobenzene (15 mL). To this were added 1,3-C6H4-
(COCl)2 (15.1 mg, 0.0744 mmol) and pyridine (2.3 mL), and
after 8 h of stirring, the reaction mixture was removed from
the glovebox and the product was filtered off and washed with
methanol and hexane several times (119 mg, 96%): 1H NMR
(CS2, CDCl3) δ 5.56 (br, 8 H, CH2), 7.80 (t, J ) 7.7 Hz, 1H),
8.10 (dd, J ) 7.6 Hz, J′ ) 1.6 Hz, 2H), 8.36 (t, J ) 1.6 Hz, 1H);
13C NMR (CS , acetone-d as insert) δ phenyl ring 129.0, 129.4,
2 6
131.3, 135.6; fullerene peaks 136.1, 140.5, 142.1, 142.3, 142.4,
142.9, 143.3, 144.6, 145.5, 145.6, 145.7, 145.9, 146.3, 146.5,
147.5, 153.3; CdO 168.2.
Bifullerene 4. This was prepared in a similar manner to
3 using 1,4-C6H4(COCl)2. 4: 1H NMR (CS2, acetone-d6 as
insert) δ 4.84 (br, 8H, CH2), 7.32 (s, 4H); 13C NMR (CS2,
The existence of a “spike” in the C601- spectrum and acetone-d6 as insert) δ phenyl ring 129.15, 137.61; fullerene
various triplets in the C602- spectrum have no wholly peaks 136.1, 140.5, 142.2, 142.3, 142.4, 142.9, 143.3, 144.7,
145.5, 145.6, 145.7, 145.9, 146.3, 146.5, 147.5, 153.3.
satisfactory explanation,14 and we are considering the EPR Sample of 31-. A deficit of reducing agent was used
possibility that they might arise from (C60)21- and (C60)22-, to ensure that no reduction to the (2-) state occurred. 3 (5.0
respectively, or from derivatized impurities similar to mg, 3.0 × 10-3 mmol) and 2.5 × 10-3 M cobaltocene PhCN
those of 1. solution (0.50 mL, 1.3 × 10-3 mmol) were mixed in PhCN (2
mL) and stirred for several hours. The suspension was
Experimental Section filtered, and the supernatant was sealed in an EPR tube.
EPR Sample of 32-. This was prepared in a similar
All manipulations of air-sensitive compounds were carried manner using 0.275 mL of the cobaltocene solution (6.88 ×
out in an inert atmosphere in a Vaccum Atmospheres drybox 10-3 mmol). EPR samples of 41- and 42- were prepared in an
under helium (H2O, O2 < 1 ppm). C60 was purchased from identical manner.
MER Corp. and used without any further purification. 1 and Copper(II)-Metalated 2. This was prepared in the same
C60(CH2)2NCPh3 were prepared by the published method,11 and manner as for 23b except that Cu(â-CHO-TPP) replaced Ni(â-
all other chemicals were purchased from Aldrich. Solvents CHO-TPP). Copper(II)-metalated 21- was prepared by addi-
were purified and dried by being heated under reflux overnight tion of cobaltocene to a solution of copper(II)-metalated 2 in
with the agent indicated below, distilled or vacuum distilled o-dichlorobenzene.
both outside and inside the glovebox, passed through a short X-ray Structure of 2. Well-shaped black prisms of 23b
column of activiated neutral alumina, and stored over 3A suitable for X-ray diffraction were obtained by slow diffusion
molecular sieves. Pyridine was distilled from sodium metal, of hexane into a solution of 2 in CHCl3/CS2(1:1).
toluene from sodium benzophenone ketyl, benzonitrile and Compound 2 (C108H36Cl3N5, Mr ) 1509.77) crystallized in
o-dichlorobenzene from a small piece of sodium metal, and the monoclinic space group P2(1)/n with cell dimensions of a
dimethyl sulfoxide (DMSO) from CaH2. Deuterated DMSO ) 14.2520 (10) Å, b ) 22.412 (2) Å, c ) 21.238(2) Å, â ) 91.760
was dried by passage through activated alumina and stored (10), V ) 6780.6 (10) Å3, Z ) 4. Data were collected at -110
over 3A molecular sieves. Deuterated nitrobenzene was first °C on a SIEMENS P4/RA diffractometer using graphite-
vacuum distilled and then dried similarly. monochromated Cu KR radiation (λ ) 1.541 78 Å) to a
EPR spectra were recorded in frozen solution near liquid maximum 2θ ) 102.5°. The structure was solved by direct
He temperatures on a Bruker ER 200D-SRC spectrometer methods (SHELXTL IRIS) and refined against F2 (SHELX93)
equipped with an Oxford Instruments ESR 900 cryostat, and yielding R1 ) 0.0801 for 5605 reflections with I > 2σ(I) and
g values were determined by calibration to DPPH powder. R1 ) 0.0967 for all data (7301 independent reflections, Rint )
Vis-NIR spectra were recorded on an Ocean Optics spectrom- 0.0349).
eter. NMR spectra were recorded at room temperature on a Refinement showed that the position of the C60-N-methyl-
Bruker AC-250, AM-360, or AMX-500-MHz spectrometer, and pyrrolidine part of the molecule could be resolved into two
the chemical shifts were calibrated against internal solvent orientations with major/minor relative populations of 0.58/0.42.
or a sealed acetone-d6 capillary. Reduction potentials for 2, The same was true for the CHCl3 solvate molecule which is
3, and 4 were obtained as previously described.8 located near the pyrrolidine N atom. Additionally, the phenyl
[Cp2Co+][11-]. 1 (25 mg, 0.032 mmol) was stirred with groups of the TPP fragment showed orientational disorder and
cobaltocene (6.2 mg, 0.033 mmol) in benzonitrile (2 mL) for were resolved into three or two orientational sites each.
several hours. Half of the solvent was removed under vacuum, During the refinement it became obvious from the observation
and toluene was added to effect precipitation of the product. of excess electron density at the center of the porphyrin that
This solid was filtered off, washed with toluene several times, it was partially metalated. Final refinement included 19%
and pumped to dryness (30 mg, 97%): 1H NMR (DMSO-d6) δ cobalt occupation. This must have arisen during recrystalli-
1.4 (br, 4H, CH2), 4.1 (br, 3H, CH3), 5.8 (10H, Cp2Co+ ); 13C zation whereby traces of the cobalt-metalated species, a
NMR (DMSO-d6) δ 85.1 (Cp2Co+), 187 (C60-pyrrolidine anion); residual from incomplete demetalation in the synthesis
reaction,3b,15 were concentrated in the single crystal.37
(14) (a) Stinchcombe, J.; Pénicaud, A.; Bhyrappa, P.; Boyd, P. D.
W.; Reed, C. A. J. Am. Chem. Soc. 1993, 115, 5212-5217. (b) Khaled, Results and Discussion
M. M.; Carlin, R. T.; Trulove, P. C.; Eaton, G. R.; Eaton, S. S. J. Am.
Chem. Soc. 1994, 116, 3465-3474. (c) Boyd, P. D. W.; Bhyrappa, P.;
Properties of the N-Methylfulleropyrrolidine(1-)
Paul, P.; Stinchcombe, J.; Bolskar, R.; Sun, Y.; Reed, C. A. J. Am. Anion, 11-. Reduction of 1 to its fulleride(1-) ion was
Chem. Soc. 1995, 117, 2907-2914. (d) Brezová, V.; Stasko, A.; Rapta,
P.; Domschke, G.; Bartl, A.; Dunsch, L. J. Phys. Chem. 1995, 99, (15) Ponomarev, G. V.; Maravin, G. B. Chem. Heterocycl. Compd.
16234-16241 and references within these papers. 1982, 18, 50-55.
Fullerides of Pyrrolidine-Functionalized C60 J. Org. Chem., Vol. 62, No. 11, 1997 3645

Figure 2. 1H NMR spectrum (DMSO-d6) of 11- (s denotes

peak due to solvent).

Figure 1. Comparison of NIR spectra in benzonitrile solu-

tion: (a) the N-methylpyrrolidine-derivatized fulleride 11-, (b)
C601- as CoCp2+ salts.
Figure 3. 13C NMR spectrum (DMSO-d6) of 11- (s denotes
readily achieved with cobaltocene or with the sodium/ peaks due to a trace of benzonitrile).
crown ether (1 equiv) method as developed earlier for
C60.14a Unlike C60, where the cobaltocene/cobaltocenium N-trityl species whose methylene protons shift from 4.17
redox potential is sufficiently negative that further to 1.2 ppm upon reduction (in nitrobenzene-d5). The
reduction can occur to give some C602-, the pyrrolidine- direction and magnitude of these shifts can be understood
functionalized C60 species are reduced selectively to the by considering the nature of the SOMO (singly occupied
monoanion by cobaltocene. This reflects the 110 mV molecular orbital)ssee below.
more negative reduction potential of 1 relative to that of
The 13C NMR spectrum of [CoCp2][1] shows only two
C60.11
detectable peaks, that of the cation at 85.1 ppm and a
Near infrared (NIR) spectroscopy offers distinctive
broad (ca. 10 ppm) fulleride peak centered at 187 ppm
fingerprints for fullerides because the LUMO/LUMO+1
(see Figure 3). The close similarity with that of unfunc-
energy gap of C60 is about 1 eV.16 Notably, the distinctive
tionalized C601- whose 2-3 ppm wide signal appears
fingerprint of C601- is retained in 11- although, as shown
centered at 186 ppm is notable.14c In diamagnetic 1, the
in Figure 1, there is a small blue shift of λmax (by ca. 70
sp2 buckyball 13C signals span the range 136-154 ppm,
nm). Similar observations have been reported in a
although the majority cluster around 143 ( 3 ppm.11
Diels-Alder adduct.17 These findings are consistent with
Upon reduction to paramagnetic 11-, they experience
a LUMO electronic structure that has only minor per-
almost precisely the same average downfield shift and
turbations from C60. It would be premature to suggest
broadening as does C60. This is compelling evidence for
assignments for the observed transitions given that the
the view that the unpaired electron in 11- is delocalized
NIR spectrum of C601- is incompletely understood.18
over the buckyball in much the same way as it is in C601-
We next explored the locale of electron spin density in
and that the average spin density seen by the majority
11- as reflected in the response of its NMR active nuclei
of the sp2 carbon atoms is the same. The sp3 carbon
to the presence of paramagnetism. The 1H NMR spec-
atoms on the buckyball, as well as those on the pyrroli-
trum of [CoCp2][1] in dimethyl-d6 sulfoxide solution is
dine ring, are not observed, presumably due to the
shown in Figure 2. The CoCp2+ resonance appears as a
combination of low relative abundance and paramagnetic
singlet at 5.8 ppm. The three N-CH3 protons on the
broadening.
pyrrolidine appear as a broad singlet at 4.1 ppm, some-
what downfield shifted from their diamagnetic position The third experimental technique we have used to
at 3.00. The four methylene protons appear as a broad probe the electronic nature of pyrrolidine-functionalized
resonance at 1.4 ppm, upfield shifted from their diamag- fullerides is EPR spectroscopy. Typical spectra arising
netic position at 4.39 ppm. The assignments were from 11- are shown in Figure 4. Narrow isotropic
consistent with integration of the areas under each peak doublets are seen at all temperatures although there are
and were verified by comparison to the corresponding hints of anisotropy at the lowest temperatures (ca. 4 K).
There is a detectable effect of the cation, cobaltocenium
(16) (a) Kato, T.; Kodama, T.; Oyama, M.; Okazaki, S.; Shida, T.; versus Na(crown)+, but this is much less dramatic than
Nakagawa, T.; Matsui, Y.; Suzuki, S.; Shiromaru, H.; Yamauchi, K.; that seen in C601-.14a Apparently, the inherently lower
Achiba, Y. Chem. Phys. Lett. 1991, 186, 35-39. (b) Lawson, D. R.; symmetry and the presence of the addends in these
Feldheim, D. L.; Foss, C. A.; Dorhout, P. K.; Elliott, C. M.; Martin, C.
R.; Parkinson, B. J. Electrochem. Soc. 1992, 139, L68-71. functionalized fullerides makes them much less sensitive
(17) Baumgarten, M.; Gherghel, L. Progress in Fullerene Research; to the effects of ion pairing. The g values are, however,
World Scientific: Singapore, 1994; pp 384-388. quite similar to those of C601- and, although a little
(18) (a) Kondo, H.; Momose, T.; Shida, T. Chem. Phys. Lett. 1995,
237, 111-114. (b) Bolskar, R. D. ; Gallagher, S. H.; Armstrong, R. S.; higher, are diagnostically below the free electron value.
Lay, P. A.; Reed, C. A. Chem. Phys. Lett. 1995, 247, 57-62. The greatest difference between the EPR spectrum of
3646 J. Org. Chem., Vol. 62, No. 11, 1997 Sun et al.

Figure 4. Typical EPR spectra of the fulleride(1-) ion of a

pyrrolidine-functionalized C60, 11-, in frozen benzonitrile (g )
2.0006).

C601- and its derivatives is the temperature dependence

of the line width.17 In C601-, the line width increases from
a few gauss at low temperatures to tens of gauss at room
temperature.14a A dynamic Jahn-Teller effect (pseudo-
rotation) and lifetime broadening from thermal popula-
tion of a low-lying excited state are among the proposals
Figure 5. Contour surface of the LUMO of 1 calculated from
to account for this. By contrast, the line width of 11- the PM3 wave function at a value of 0.002 e au-3.
increases by no more than 1-2 G over a similar temper-
ature range. Obviously, the functionalization of C60 contact interaction. An approximate estimate of the spin
eliminates the possibility of significant pseudorotation density assuming a McConnell type relationship gives a
(and of real rotation in frozen solution). It may also value of 0.001-0.002 (0.1-0.2%) at the methylene carbon
separate the ground state further from the first excited atoms. This suggests an upper limit of 0.5% for the
state inasmuch as permanent functionalization may delocalization of the unpaired electron from the fullerene
remove the (t1u)1 triple degeneracy of C601- more ef- cage onto the N-methylpyrrolidine ring.
fectively than a Jahn-Teller distortion. Finally, we note X-ray Structure of the Porphyrin-C60 Dyad, 2.
that there is no evidence for hyperfine coupling to N or The synthesis, the ground state perturbation of the
H. This is consistent with very low spin density on the electronic spectrum of the porphyrin by the C60 substitu-
pyrrolidine functionality. ent, the redox potentials, and the photoinduced electron
The overall close spectroscopic similarity of 11- to C601-, transfer of 2 have been reported.3b,8 We now report on
particularly with techniques that probe unpaired electron its crystal structure as a chloroform solvate. We have
spin density, has led us to seek calculational support for been fortunate in growing X-ray diffraction quality
the emerging picture of spin delocalized over the bucky- crystals, but as often seems to be the case with C60
ball but not onto the pyrrolidine functionality. Semi- structures,21 there is crystallographic disorder which, in
empirical molecular orbital calculations of the electronic the present case, defeats the goal of obtaining useful
and molecular structure19,20 of the monoanion using the molecular bond lengths and angles from the data. Never-
PM3 method show that the SOMO is principally localized theless, the disposition of the porphyrin and C60 moieties,
on the fullerene cage. Examination of spin density maps both intra- and intermolecularly, are well-resolved, as are
indicates that the majority of the unpaired electron is the modes of linkage.
localized on the fullerene carbon atoms. Figure 5 shows Perhaps the most interesting feature of the structure
a surface representation of the LUMO for the neutral is the intermolecular packing relationship of C60 to the
molecule at the contour level of 0.001 e au-3. It shows porphyrin. Figure 6 illustrates the pairwise, ball-on-
that only very small components of this orbital are hand manner in which the buckyball makes intermo-
delocalized from the fullerene cage. The nodal sequence lecular contact with the porphyrin. There are two
of this molecular orbital in the pyrrolidine ring leads to crystallographically distinct contacts. The distances of
an orthogonality of the carbon p orbital with respect to the closest C60 carbon atoms to the mean plane of the
the methylene protons and delocalization of the orbital 16-atom inner core of the porphyrin ring are very short:
onto the methyl protons. The proton isotropic hyperfine 2.78 Å to one ring and 2.79 Å to the other. (The
coupling constants calculated from the observed shifts corresponding distances in the minor orientation, which
are -0.042 and +0.014 G for the methylene and N- arises from crystallographic disorder, are 2.75 and 2.70
methyl protons, respectively. These coupling constants Å.) The shortness of these distances is apparent when
are in keeping with the calculated form of the SOMO. comparisons are made to the inter-layer separation of
Spin density at the methylene protons arises from a spin graphite (3.35 Å), to interfacial porphyrin/porphyrin
polarization mechanism whilst that at the N-methyl separations (>3.2 Å),22 to the separations of C60 from
protons arises from direct delocalization, leading to a
(21) Balch, A. L.; Lee, J. W.; Noll, B. C.; Olmstead, M. M. J. Chem.
(19) Calculations were performed using Spartan version 4.1, Wave- Soc., Chem. Commun. 1993, 56-58.
function Inc., Irvine, CA 92715. (22) Scheidt, W. R.; Lee, Y. J. Struct. Bonding (Berlin) 1987, 64,
(20) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209-220. 1-70.
Fullerides of Pyrrolidine-Functionalized C60 J. Org. Chem., Vol. 62, No. 11, 1997 3647

Figure 6. Perspective drawing of the unit cell packing in the

crystal structure of 2 (major orientation) showing close ball/
porphyrin intermolecular interactions. Phenyl substituents on
the porphyrins and the chloroform solvate are omitted for
clarity.

arene rings in various X-ray structures (ca. 3.0-3.5

Å),21,23-25 and to ball-to-ball separations in C60 and its Figure 7. Plane projections of portions of the C60 moieties
derivatives (>3.2 Å).21,23,24 This suggests that the usual (open bonds) onto the porphyrin cores (filled bonds) for the
and expected π/π interaction is substantially augmented two intermolecular contacts of the major orientation. Closest
by a strong donor/acceptor interaction. There is a close approaches to the porphyrin planes are (a) C(4la) 2.79 Å,
structural analogy to the interaction of arenes with the C(42a) 2.82 Å, C(40a) 3.02 Å, C(23a) 3.06 Å, C(22a) 3.28 Å, (b)
(tetraphenylporphinato)iron(III) cation, Fe(TPP)+. In the C(50a) 2.78 Å, C(32a) 2.97 Å, C(49a) 3.00 Å, C(51a) 3.19 Å.
crystal structure of [Fe(TPP)‚p-xylene]+, two arene car-
bon atoms approach to within 2.89 Å of the porphyrin the other, it is the second closest atom by a mere 0.07 Å.
plane.26 In the sterically less-hindered toluene analog, This centering of the buckyball over the porphyrin is
the closest approach is even shorter (2.71 Å).27 With its consistent with the proposed donor/acceptor relationship
curved surface, C60 is sterically suited to a similarly close because the two hydrogen atoms (or a metal)28 at the
approach and the electroneutrality of the porphyrin in 2 center of the porphyrin are the locus of acidity in this
vis-à-vis the cationic nature of Fe(TPP)+ serves only to otherwise relatively electron-rich macrocycle. It could be
emphasize the closeness of the approach. In contrast to argued that the centering is simply the result of crystal
the usual view of C60 as an acceptor, the sense of the packing. However, the closeness of the approach of the
donor/acceptor relationship might be with C60 as the C60 moiety to the porphyrin suggests there is a strong
donor and the porphyrin (or metalloporphyrin)28 as the electronic effect as well.
acceptor. Figure 7 shows plane projections of the closest With C60 as the donor in this interaction, one might
carbon atoms of C60 onto the porphyrin planes for the expect that the more electron-rich bonds of the buckyball
two contacts in their major orientations. In one, the C60 would be closest to the porphyrin, i.e., the more electron
carbon atom closest to the porphyrin plane is also the rich “double” bonds at 6:6 ring junctures might interact
carbon atom closest to the center of the porphyrin. In with the porphyrin in preference to the less electron-rich
“single” bonds at 5:6 ring junctures. Figure 8 shows that
(23) Meidine, M. F.; Hitchcock, P. B.; Kroto, H. W.; Taylor, R.; this expectation is borne out for one of the interactions
Walton, D. R. M. J. Chem. Soc., Chem. Commun. 1992, 1534-1537. but not for the other. This probably means that the
(24) Crane, J. D.; Hitchcock, P. B.; Kroto, H. W.; Taylor, R.; Walton,
D. R. M. J. Chem. Soc., Chem. Commun. 1992, 1764-1765. energy difference is not very great, and the near identi-
(25) (a) The shortest known arene-C60 interaction is 2.94 Å in cality of the ball/porphyrin plane separations would seem
C60‚hexamethylbenzene (Sun, Y.-P.; Drovetskaya, T.; Reed, C. A. to support this. On the other hand, there are crystal
Unpublished results). In this structure, C6Me6 is sandwiched between
C60 molecules which are separated by a center-to-center distance of packing considerations and, of course, the functionalized
12.96 Å. The C‚‚‚C diameter of C60 is taken as 7.08 Å.24,25g (b) Ermer, C60 moieties are not free to rotate. These features of the
O. Helv. Chim. Acta 1991 74, 1339-1351. (c) Balch, A. L.; Catalano, solid state may prevent the buckyballs from adopting
V. J.; Lee, J. W.; Olmstead, M. M. J. Am. Chem. Soc. 1992, 114, 5455-
5457. (d) Chiang, L. Y.; Swirczewski, J. W.; Liang, K.; Millar, J. Chem. their truly optimal orientation with respect to the por-
Lett. 1994, 981-984. (e) Osterodt, J.; Nieger, M.; Vögtle, F. J. Chem. phyrin.
Soc., Chem. Commun. 1994, 1609-1608. (f) Steed, J. W.; Junk, P. C.;
Atwood, J. L.; Barnes, M. J.; Raston, C. L. J. Am. Chem. Soc. 1994, The binding of fullerenes to tetraphenylporphyrin
116, 10346-10347. (g) Fedurco, M.; Olmstead, M. M.; Fawcett, W. R. moieties is believed to be the basis by which chromato-
Inorg. Chem. 1995, 34, 390-392. (h) Atwood, J. L.; Barnes, M. J.; graphic separation of fullerenes is achieved with a
Gardiner, M. G.; Raston, C. L. Chem. Commun. 1996, 1449-1450.
(26) Xie, Z.; Bau, R.; Reed C. A. Angew. Chem., Int. Ed. Engl. 1994, porphyrin-modified silica gel column.29 Indeed, the
33, 2433-2434. graphical representation envisaged by Meyerhoff and co-
(27) Xie, Z.; Reed, C. A. Unpublished results.
(28) The present structure shows partial metalation of the porphyrin
with cobalt(II), a residue from the synthesis. Independently, Balch and (29) (a) Xiao, J.; Savina, M. R.; Martin, G. B.; Francis, A. H.;
co-workers have isolated cobalt octaethylporphyrin/fullerene species Meyerhoff, M. E. J. Am. Chem. Soc. 1994, 116, 9341-9342. (b) Kibbey,
showing similar close interactions: reported at the Electrochemical C. E.; Savina, M. R.; Parseghian, B. K.; Francis, A. H.; Meyerhoff, M.
Society Meeting, Los Angeles, May 6-10, 1996. E. Anal. Chem. 1993, 65, 3717-3719.
3648 J. Org. Chem., Vol. 62, No. 11, 1997 Sun et al.

a molecular mechanics calculation.31 The equatorial/

axial preference in this calculation, however, was only 1
kcal and was reversed in a semiempirical calculation.19
This suggests that the equatorial substituent preference
is real, but not particularly strong.
Chromophore/Chromophore Interactions. Com-
pounds 2, 3, and 4 provide opportunities to observe the
perturbation of the electronic structure of a functionalized
C60 moiety by a nearby porphyrin or C60 chromophore.
Since each of these species has the same pyrrolidine
functionality, direct comparisons can be made to 1. As
the connectivities in 2-4 involve sp3 carbon linkages to
remote chromophores, the mutual influence of the chro-
mophores will be largely through space, at van der Waals,
or solvent-separated contact. Our studies have been
directed at ground state electronic pertubation,3b excited
state phenomena,8 redox potential influences,8 and, now,
spin coupling effects.
The ground state perturbation of the NIR spectrum of
Figure 8. Perspective drawing of 2 showing the addition
across a 6:6 ring juncture and a trans (equatorial-equatorial) the C601- moiety is indicated by shifts from λmax in 11-
disposition of the methyl and porphyrin substituents with where it is 1002 nm. The shift is to higher energy for
respect to the C-N bond of the pyrrolidine ring. the porphyrin species and increases from Cu(II)-meta-
lated 2 (993 nm) to free base 2 (986 nm) in o-dichloroben-
workers for the interaction of C60 with zinc tetraphenyl- zene solution. The shift is to lower energy in 3 and 4 by
porphyrin (see Figure 3 in ref 29a) is remarkably close the same amount (to 1015 nm) in benzonitrile solution.
to that observed in the present work. They emphasize These effects are relatively small and probably reflect the
π/π interactions, overall fullerene basicity, and shape proximity and solvation influences of the appendage. The
selectivity as important factors in understanding the same is true for the redox potentials of 2, 3, and 4. Under
chemical basis of the chromatographic discrimination identical conditions (benzonitrile, Bu4N+PF6-), the three
mechanism. The present work suggests that a fairly reduction waves of 1 are shifted by no more than 20 mV
specific donor/acceptor interaction is of major importance. in 2,8 3, and 4.
In particular, the retention differences between fullerenes Spin coupling effects were investigated in Cu(II)-
and endohedral metallofullerenes, which have dipole metalated 21-, but no significant evidence of coupling was
moments, can be readily understood in terms of a donor/ found. The EPR spectrum of 21- was essentially the
acceptor relationship. same as that of 11-. The EPR spectrum of Cu(II)-
Despite the widespread use of pyrrolidine functional- metalated 21- was essentially the summation of 21- and
ization of C60, this is the first pyrrolidine derivative to Cu(II)-metalated 2. This can be understood in terms of
be characterized by X-ray crystallography. It confirms the nature of the magnetic orbitals. The Cu(II) unpaired
that the azamethine ylide addition occurs across a 6:6 electron is in a dx2-y2 orbital and would be delocalized
ring juncture, as previously deduced from 13C NMR into the σ*Cu-N orbitals of the porphyrin. On the other
spectroscopy.11 Two further, more subtle aspects of bond hand, the unpaired electron on the C601- moiety is in a
connectivity and conformation are revealed by this π* orbital. These orbitals are not cospatial via van der
structure. The first concerns isomer preference in 2-sub- Waals contact, nor do they interact via the pyrrolidine
stituted pyrrolidines. There are cis and trans disposi- linkage. This contrasts with the UV-vis and NIR
tions possible for the N-methyl and 2-porphyrin substit- mutual pertubations of electronic spectra where π/π van
uents on the C-N bond of the five-membered pyrrolidine der Waals effects are observed.
ring. As shown in Figure 8, the X-ray structure reveals The most interesting spin coupling effect is observed
the trans disposition. This indicates a substantial bias in 32- where two C601- moieties are held in van der Waals
toward trans addition in the synthetic reaction3b because contact by virtue of the 1,3-phthalamido linkage. As
the material used in this work is the major product. A shown in Figure 9, triplet “wings” (∆H ) 29 G) are
minor product was observed in the chromatographic evident in 32- in addition to the dominant central doublet
purification. Its 1H NMR spectrum is consistent with a of spin-isolated C601- moieties. These triplet features are
nonequilibrating cis isomer, but there was insufficient absent in 11-, 21-, 31-, and 41- and essentially absent in
material to achieve a complete characterization. The 42-. We believe that the triplet arises from intramolecu-
second aspect concerns the conformational preference of lar ball-to-ball spin-spin interactions. An investigation
the N-methyl substituent. Inspection of Figure 8 reveals of possible conformations of the neutral form of 3 using
its equatorial orientation, i.e., the lone pair on the N atom molecular mechanics31 has found that the most stable
is axial. This is consistent with data on other N- structure, by more than 10 kcal/mol, is that with the two
methylpyrrolidines whose structures have been deter- balls in near van der Waals contact. The enhanced
mined by X-ray crystallography30 and with the result of stability of this conformation for this model arises from

(30) (a) Murray-Rust, P.; Murray-Rust, J.; Middlemiss, D. Acta (31) Calculations were performed using Cerius 2, BIOSYM/Molec-
Crystallogr. B 1980, 36 1678-1680. (b) Greco, M. N.; Rasmussen, C. ular Simulations, San Diego, CA. The model employed the Universal
R. J. Org. Chem. 1992, 57, 5532-5535. (c) Tetzlaff, C.; Butz, V.; Force Field32 with the Charge Equilibration method.33
Vilsmaier, E.; Wagemann, R.; Maas, G.; Ritter von Onciul, A.; Clark, (32) Rappé, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.;
T. J. Chem. Soc., Perkin Trans. 2 1993, 1901-1905. (d) Seibel, J.; Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024-10035.
Vilsmaier, E.; Frohlich, K.; Maas, G.; Wagemann, R. Tetrahedron 1994, (33) Rappé, A. K.; Goddard, W. A. J. Phys. Chem. 1991, 95, 3358-
50, 715-730. 3363.
Fullerides of Pyrrolidine-Functionalized C60 J. Org. Chem., Vol. 62, No. 11, 1997 3649

impurities, or aggregation effects.14 The present work

supports the notion that the presence of a functionalized
fulleride(1-) as a minor impurity in C60 could give rise
to the minor EPR signal in C601-. On the other hand, it
is hard to believe that everyone’s C60 is contaminated
with similar amounts of a functionalized impurity that
gives a sharp EPR signal when reduced.35

Conclusion
Upon reduction to the fulleride(1-) state, monofunc-
tionalized C60 retains much of its parent C601- character.
Figure 9. Frozen solution EPR spectra of 32- at 83 K showing The frontier electronic levels have lower symmetry, and
central doublet features (g ) 2.0006, ∆H ) 1.1 G) and triplet the LUMO is typically shifted to somewhat higher energy
“wings” (∆H ) 29 G, g ) 2.0007). but the delocalized π character is retained. The interac-
tion of functionalized C60 with itself or with other
an attractive van der Waals interaction between the balls chromophores via this π system shows considerable
of about 10 kcal. The close proximity of the two balls potential in applications to problems of electron transfer36
(nearest C-C contact 3.2 Å) in 32- should give rise to a and spin coupling. Although the data in the present work
magnetic dipole/dipole interaction between the neighbor- were gathered on pyrrolidine-functionalized species, it
ing unpaired spins. The magnitude of the dipolar seems likely that the observed features will be seen in
interaction is reflected in the size of the zero field other monofunctionalized C60 derivatives. It remains to
splitting of the triplet state spectrum, D ) 29 G (Figure be seen what the range and extent of these features will
9). Occasionally, these triplet features are observed at be since presently available data are limited mainly to
high spectrometer gain settings in species other than 32-, reduction potential measurements and ground state UV-
such as 42-; we take this to mean that aggregation effects vis spectral perturbations.
can, under certain circumstances, bring about small
amounts of intermolecular spin-spin interaction. This Acknowledgment. We thank Dr. Zuowei Xie for
observation may be important in understanding the bulk help with the X-ray structure determination and the
ferromagnetism observed in [TDAE+][C601-].34 National Institutes of Health (GM 23851 to C.A.R.) and
On the Origin of “Spurious” EPR Signals in C60n-. the Marsden Fund of New Zealand (UOA613 to
As discussed earlier, the EPR signals of pyrrolidine- P.D.W.B.) for support.
functionalized C601- moieties are characterized by sharp
doublets with g values less than the free electron value, JO970357U
and line widths of about 1 G that are not particularly
(35) The observation of triplet features from the side-by-side 1-
temperature dependent. This line width behavior con- buckyballs in 32- has led us to entertain the notion that dimers are
trasts markedly with the predominant EPR signal of responsible for the observation of minor EPR signals in C60n-. Interest-
C601- which, in most reports, increases from several gauss ingly, the 29 G splitting in 32- corresponds exactly to the perpendicular
portion of the “high temperature triplet” (2D ) 58 G) seen as a minor
at low temperature to tens of gauss at room temperature. signal in the EPR spectrum of C602-.14c In both cases, the signals are
The features of the pyrrolidine-functionalized C601- moi- not observable at 4K and their g values are very similar (ca. 2.0005).
eties are, however, very similar to the minor signal or This raises the possibility that aggregation of C602- leads to a range of
disproportionation products, one of which may be (C60)22-. Whether
“spike” in the EPR spectrum of C601-. This minor signal such species have a specific, weak C-C bond is very difficult to say.
has been variously ascribed to ion pairing effects, dis- More experiments on this tricky problem are needed.
proportionation, a thermally accessible excited state, (36) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science
1995, 270, 1789-1791.
(37) The author has deposited atomic coordinates for this structure
(34) (a) Allemand, P.-M.; Khemani, K. C.; Koch, A.; Holczer, K.; with the Cambridge Crystallographic Data Centre. The coordinates
Donovan, S.; Gruner, G.; Thompson, J. D.; Wudl, F. Science 1991, 253, can be obtained, on request, from the Director, Cambridge Crystal-
301-303. (b) Gotschy, B Phys. Rev. B 1995, 52, 7378-7383. lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.