Electron-acceptor-induced isomerization of aryl [6,5] open fulleroids to [6,6] closed methanofullerenes and the electrochemical evaluation of their free energy difference

Article abstract of DOI:10.1021/jo0521706

The thermal and photochemical rearrangements of a series of aryl substituted [6,5] open fulleroids to [6,6] closed methanofullerenes are accelerated in the presence in of electron acceptors. These [6,5] open fulleroids, facilitated by electron acceptors, rearrange thermally by a zwitteronic-type intermediate, while the photochemical reactions proceed via an excited-state electron-transfer process. The oxidation potentials of these [6,5] open fulleroids and their corresponding [6,6] closed methanofullerenes isomers have been evaluated. The free energy difference between the [6,5] open fulleroids and their corresponding [6,6] closed isomers have been estimated from the difference in their oxidation potentials.

Full text of DOI:10.1021/jo0521706

Electron-Acceptor-Induced Isomerization of Aryl [6,5] Open
Fulleroids to [6,6] Closed Methanofullerenes and the
Electrochemical Evaluation of Their Free Energy Difference
Melvin H. Hall*
Department of Chemistry and Physics, LaGrange College, LaGrange, Georgia 30240
Philip Shevlin,* Huajun Lu, Anthony Gichuhi, and Curtis Shannon
Department of Chemistry, Auburn UniVersity, Auburn, Alabama 36849
mhall@lagrange.edu
ReceiVed October 18, 2005
The thermal and photochemical rearrangements of a series of aryl substituted [6,5] open fulleroids to
[6,6] closed methanofullerenes are accelerated in the presence in of electron acceptors. These [6,5] open
fulleroids, facilitated by electron acceptors, rearrange thermally by a zwitteronic-type intermediate, while
the photochemical reactions proceed via an excited-state electron-transfer process. The oxidation potentials
of these [6,5] open fulleroids and their corresponding [6,6] closed methanofullerenes isomers have been
evaluated. The free energy difference between the [6,5] open fulleroids and their corresponding [6,6]
closed isomers have been estimated from the difference in their oxidation potentials.
Introduction
functionalized fullerene derivatives with electronic structures
which largely preserve the C60 chromophore.7,8 Although several
workers have observed zero-order kinetics for the thermal [6,5]
open to [6,6] closed rearrangement,3,9 we have demonstrated
that a photochemical step in the thermal rearrangement is
The addition of diazo compounds to C60 generates fullerene
derivatives with a one-carbon bridge across a carbon-carbon
bond.1 Although there are hypothetically four possible ways to
add a one carbon bridge to C60, only the [6,5] open fulleroids,
1, and the [6,6] closed methanofullerenes, 2, have been observed
experimentally. The [6,5] open fulleroids 1 are the kinetically
controlled product, whereas the [6,6] closed methanofullerenes
2 are the products of thermodynamic control.2 The [6,5] open
isomers have been reported to rearrange thermally,3 photo-
chemically,4 electrochemically,5 and under acid catalysis6 to the
methanofullerenes, which have attracted much attention as useful
(3) Smith, A. B., III; Strongin, R. M.; Brard, L.; Furst, G. T.; Romanow,
W. J.; Owens, K. G.; Goldschmidt, R. J.; King, R. C. J. Am. Chem. Soc.
1995, 117, 5492-5502.
(4) (a) Janssen, R. A. J.; Hummelen, J. C.; Wudl, F. J. Am. Chem. Soc.
1995, 117, 544-545. (b) Ceroni, P.; Conti, F.; Corvaja, C.; Maggini, M.;
Paolucci, F.; Roffia, S.; Scorrano, G.; Toffoletti, A. J. Phys. Chem. A 2000,
104, 156-163.
(5) Eiermann, M.; Wudl, F.; Prato, M.; Maggini, M. J. Am. Chem. Soc.
1994, 116, 8364-8365.
(6) Gonzalez, R.; Hummelen, J. C.; Wudl, F. J. Org. Chem. 1995, 60,
2618-2620.
(7) (a) Suzuki, T.; Maruyama, Y.; Akasaka, T.; Ando, W.; Kobayashi,
K.; Nagase, S. J. Am. Chem. Soc. 1994, 116, 1359-1363. (b) Arias, F.;
Echegoyen, L.; Wilson, S. R.; Lu, Q. J. Am. Chem. Soc. 1995, 117, 1422-
1427.
(1) (a) Suzuki, T.; Li, Q.; Khemani, K. C.; Wudl, F.; Almarsson, O.
Science 1991, 245, 1186-1888. (b) Vasella, A.; Uhlmann, P.; Waldraff,
C. A. A.; Diederich, F.; Thilgen, C. Angew. Chem., Int. Ed. Engl. 1992,
31, 1388-1390. (c) Wudl, F. Acc. Chem. Res. 1992, 25, 157-161.
(2) For reviews, see: (a) Diederich, F.; Isaacs, L.; Philip, D. J. Chem.
Soc. ReV. 1994, 243-255. (b) Hirsch, A. The Chemistry of the Fullerenes;
Thieme: Stuttgart, 1994. (c) Diederich, F.; Thilgen, C. Science 1996, 271,
317-323. (d) Taylor, R. Lecture Notes on Fullerene Chemistry; Imperial
College Press: London, 1999. (e) Yurovskaya, M. A.; Trushkov, I. V. Russ.
Chem. Bull. 2002, 51, 367-443.
(8) (a) Martin, N.; Sa´nchez, L.; Illescas, B.; Pe´rez, I. Chem. ReV. 1998,
98, 2527-2547. (b) Echegoyen, L.; Echegoyen, L. E. Acc. Chem. Res. 1998,
31, 593-601.
(9) Hummelen, J. C.; Knight, B. W.; LePeq, F.; Wudl, F.; Yao, J.;
Wilkins, C. L. J. Org. Chem. 1995, 60, 532-538.
10.1021/jo0521706 CCC: $33.50 © 2006 American Chemical Society
Published on Web 03/31/2006
J. Org. Chem. 2006, 71, 3357-3363
3357

Hall et al.
SCHEME 1
SCHEME 2
responsible for the zero-order kinetics.10 Thus, while the
rearrangement of [6,5] open fulleroid 1 to [6,6] closed metha-
nofullerene 2 occurs in 8 h under ambient light at 153 °C with
zero-order kinetics, no rearrangement was observed when 1 was
heated for 24 h in the dark at 153 °C. The inhibitory effect of
oxygen and quenching by triplet sensitizers led us to propose
in the oxidation potentials of the corresponding isomers to
estimate the free energy difference between them.
Results and Discussion
Thermal Rearrangement of Aryl-Substituted Fulleroids
in the Presence of Electron Acceptors. When a degassed
solution of 4a in o-dichlorobenzene-d4 (ODCB-d4) was heated
in the dark at 170 °C, it rearranged to 5a in 2 h. However,
when a degassed solution of equal molar amounts of 4a and
tetracyanoethylene (TCNE) in OCDB-d4 was heated in the dark
at 170 °C, the rearrangement of 4a was too fast to conveniently
measure. This TCNE-promoted reaction shows first-order kinet-
ics between 60 and 140 °C with a ∆Hq of 6.7 ( 0.4 kcal/mol
and a ∆Sq ) -58.4 ( 0.6 cal/mol-deg (Figure 1). The reaction
rate doubles when the amount of TCNE is doubled. The thermal
isomerization of 4a in the absence TCNE has a ∆Hq of 23.0 (
0.3 kcal/mol and a ∆Sq of -24.4 ( 0.8 cal/mol-deg. Thus,
TCNE lowers the enthalpy barrier by about 16 kcal/mol but
3
the intermediacy of 1* in which opening to biradical 3 is the
step requiring thermal activation (Scheme 1).
We recently reported that [6,5] open fulleroids with radical
stabilizing groups, 4, rearrange to the corresponding [6,6] closed
isomers, 5, both by a zero-order photochemical process and by
a higher energy first-order thermal pathway.11 For example, the
rearrangement of the [6,5] open fulleroid 4a to the [6,6] closed
fullerene 5a occurs in 40 min in the presence of ambient light
at 35 °C with zero-order kinetics (k0 ) 1.38 × 10-6 M s-1
)
with an activation energy of 9.5 kcal/mol by the mechanism in
Scheme 1. However, the rearrangement of 4a occurs over 27 h
in the dark at 130 °C with first-order kinetics (k1 ) 1.63 ×
10-5 s-1) and an activation energy of 23 kcal/mol. We have
proposed that the higher energy unimolecular thermal rear-
rangement of 4 involves a disrotatory ring closure to the [6,5]
closed isomer 6 which subsequently rearranges to the [6,6]
closed methanofullerene 5 via a biradical-like intermediate, 7
(Scheme 2).
In this study we report that both the thermal and photochemi-
cal rearrangements of these aryl-substituted fulleroids can be
facilitated by the addition of electron acceptors to the reaction
mixture. To evaluate the mechanism of this unusual reaction,
we have measured the kinetics of these thermal and photo-
chemical processes. In addition, we have evaluated the oxidation
potentials of a series of [6,5] open fulleroids and their [6,6]
closed methanofullerenes isomers and have used the difference
FIGURE 1. Eyring plot of the first-order thermal rearrangement of
4a to 5a with TCNE present at 60, 80, 100, and 140 °C in the absence
of light. An activation enthalpy of 6.7 ( 0.4 kcal/mol and an activation
entropy of -58.3 ( 0.6 cal/mol-deg were measured.
(10) Li, Z.; Shevlin, P. B. J. Am. Chem. Soc. 1997, 119, 1149-1150.
(11) Hall, M. H.; Lu, H.; Shevlin, P. B. J. Am. Chem. Soc. 2001, 123,
1349-1354.
3358 J. Org. Chem., Vol. 71, No. 9, 2006

Isomerization of Aryl [6,5] Open Fulleroids
TABLE 1. Potential Values (E°) for All Observed Redox Processes
measured at 25 °Ca
places a greater entropic demand on the transition state. Since
TCNE is a good electron acceptor (Ered ) -0.25 V),12 there is
the possibility that this reaction proceeds by electron transfer.
Thus, we have evaluated the rearrangement of 4a in the presence
of other electron acceptors including chloranil (Ered ) -0.44
E° (V vs Fc/Fc+)
oxidation
reduction
first
compound
first
second
V),13 C60 (Ered ) -0.99 V),14 and p-dinitrobenzene (Ered
)
C60
4a
5a
4b
5b
4c
5c
4d
5d
8
+1.26b
+1.00
+1.20
+1.06
+1.28
+1.11
+1.29
+1.11
+1.34
+0.52
+0.74
-0.99c
-1.28
-1.17
-1.28
-1.25
-1.16
-1.02
-1.22
-1.13
-1.15 V),15 with the hope of observing a thermal electron-
transfer process. Chloranil and p-dinitrobenzene show no effect
on the rate of rearrangement of 4a, whereas C60 increases the
rearrangement rate by a factor of 1.5 at 170 °C. Since C60 is
more difficult to reduce than chloranil, it is clear that the
effectiveness of the catalysis does not parallel the electrochemi-
cal reduction potential of the acceptor.
The rate-accelerating effect of TCNE is not only observed
on the fulleroids with aryl substituents on the methano bridge
carbon but also manifests itself in the rearrangement of 8 to 9.
We have previously reported that the thermal rearrangement of
8 to 9 is rapid as a result of the favorable prealignment of the
π-orbitals of the double bond for stabilization of the allylic
radical that develops on the methano bridge carbon.11 At 80
°C, TCNE increases the rate of rearrangement of 8 to 9 by a
factor of 3.
+1.23
+1.42
+1.29
+1.48
+1.37
+1.53
+1.35
+1.60
+0.79
+0.93
9
a Values were measured in 1.0 mM ODCB/0.15 M TBAPF6 with Ag/
Ag+ quasi-reference electrode and reported in volts relative to the potential
for internal Fc/Fc+. The scan rate was 40 mV/s. b The oxidation potential
for C60 is reported in ref 16. c The reduction potentials for C60 are reported
in ref 14.
between -37 to -45 cal/mol-deg.18 The unfavorable thermo-
dynamics coupled with the fact that chloranil (E1/2 -0.44 V),13
a better electron acceptor than C60 (E1/2 -0.99 V),14 has no effect
on reaction rate, whereas C60 does, argues against an electron
transfer in the mechanism of the TCNE-promoted isomeriza-
tions. Reactions in which the rate parallels the electrochemical
reduction potential of the acceptors are usually thought to
proceed by an electron transfer.19
Dilute solutions of 4a and TCNE in ODCB, when mixed,
did not exhibit any color change that is usually observed upon
the formation of charge-transfer complexes of aromatic com-
pounds and TCNE.20 The absorption spectrum of a mixture of
the equal molar solutions of TCNE and 4a is identical with the
sum of its individual components at the same concentrations.
These results argue against the existence of significant amounts
of a charge-transfer complex between 4a and TCNE under these
conditions.
To further consider the possibility that this reaction proceeds
by an initial electron transfer from the fulleroid to TCNE, we
have measured the electrochemical oxidation potentials of 4a-
d, 5a-d, 8, and 9. In these cyclic voltammetric studies, double
irreversible oxidation peaks were observed for all of the
substituted C60 adducts (Table 1). However, C60 itself shows
only a single reversible oxidation peak (+1.26 V).16
Since the data are not consistent with an electron transfer
and the consequent intermediacy of radical ions, we must
consider the possibility that the TCNE assists the reaction by
first forming a covalently bound adduct that subsequently
generates product with the elimination of the TCNE. The first
step in the generation of an adduct is likely to be formation of
the zwitterion 10, which could close to 11 or loose TCNE to
generate 5 as shown in Scheme 4. Although NMR and MS
analysis do not reveal the presence of 11, it is possible that it
is formed reversibly in small quantities during the reaction.
However, the simplest explanation for the effect of TCNE is
the formation of zwitterion 10, which provides a low energy
pathway connecting 4 and 5. The anti-Bredt strain energy
associated with double bonds at the bridgehead carbons in 4
could provide a driving force for the formation of 10. The fact
that the TCNE-assisted rearrangement of 4a has a more negative
∆Sq than thermal rearrangement in the absence of TCNE is
indicative of the associative nature of the former process.
Cycloadducts similar to 11 have been observed in the thermal
The first oxidation potentials of 4a-d and 8 are +1.00,
+1.06, +1.11, +1.11, and +0.52 V (vs Fc/Fc+), respectively.
Thus, the single electron transfers from 4a-d and 8 to TCNE
(Scheme 3) are all thermodynamically unfavorable with the
calculated free energy changes of +28.50, +30.20, +31.36,
+31.36, and +17.74 kcal/mol for 4a, 4b, 4c, 4d, and 8,
respectively. The fact that the activation free energy for the
TCNE-promoted rearrangement of 4a is 6.7 kcal/mol, whereas
the free energy change for the electron-transfer process requires
28 kcal/mol, provides evidence against an electron transfer in
this reaction. Furthermore, the ∆Sq of -58.4 ( 0.6 cal/mol-
deg is far more negative than that for a typical electron transfer,
which is usually about -11 to -16 cal/mol-deg17 and seems to
be more characteristic of an addition reaction. For example, the
entropy of activation of several Diels-Alder reactions falls
(12) (a) All potentials given in this report are determined to be, or
corrected to be, referenced to Fc/Fc+ (b) Peover, M. Trans. Faraday Soc.
1962, 58, 2370-2374.
(13) Jackman, A. AdV. Org. Chem. 1960, 2, 329-366.
(14) Arias, F.; Xie, Q.; Wu, Y.; Lu, Q.; Wilson, S. R.; Echegoyen, L. J.
Am. Chem. Soc. 1994, 116, 6388-6394.
(18) Kiselev, V. D.; Kashaeva, E. A.; Iskhakova, G. G.; Shihab, M.
(15) Mann, C. K.; Barnes, K. K. Electrochemical Reactions in Non-
aqueous Systems; Marcel Dekker: New York, 1970.
(16) Xie, Q.; Arias, F.; Echegoyen, L. J. Am. Chem. Soc. 1993, 115,
9818-9819.
Tetrahedron 1999, 55, 12201-12210.
(19) Peacock, N. J.; Schuster, G. B. J. Am. Chem. Soc. 1983, 1056, 3632-
2638.
(20) Kim, E.; Christl, M.; Kochi, J. K. Chem. Ber. 1990, 123, 1209-
1218.
(17) Borchardt, D.; Wherland, S. Inorg. Chem. 1986, 25, 901-905.
J. Org. Chem, Vol. 71, No. 9, 2006 3359

Hall et al.
SCHEME 3
SCHEME 4
TABLE 2. Rate Constants (s-1) for the Thermal (170 °C)
First-Order Rearrangement of [6,5] Open Fulleroids to [6,6] Closed
Methanofullerenes
further downfield than the corresponding protons in 12 and 13.22
Compounds 14, 15, and 16 were formed in a 70:20:10 ratio.
The thermal rearrangements of 12 to 13, and 14 and 15 to 16
were not observed upon heating both solutions in the dark at
170 °C for 48 h in the presence or absence of TCNE.
[6,5] open fulleroid
no catalysta
catalyst (TCNE)
4a; R1 ) OCH3, R2 ) p-C6H4OCH3
4b; R1 ) OCH3, R2 ) CH3
4c; R1 ) OCH3, R2 ) H
2.27 × 10-4
2.29 × 10-5
1.37 × 10-5
1.56 × 10-5
1.03 × 10-5
very fast
3.47 × 10-3
2.10 × 10-4
2.97 × 10-5
3.77 × 10-5
4d; R1 ) NO2, R2 ) CH3
4e; R1 ) H, R2 ) CH3
a Values for the noncatalyzed reaction are reported in ref 11.
reaction of 1,1-(p-methoxyphenyl)spiropentanes assisted by
TCNE, and a zwitteronic intermediate similar to 10 has been
proposed.21
Table 2 demonstrates that the rate of rearrangement of
fulleroids 4a-e in the presence of TCNE decreases with
decreasing ability of the substituents to stabilize a positive charge
on the bridging carbon. It is interesting to note that 4d, with its
electron-withdrawing p-NO2 group rearranges faster than 4e,
with a p-H, in the absence of TCNE but slower in its presence.
This is consistent with our postulate that the bridging carbon
develops radical character in the absence of TCNE but has
cationic character in the lower energy TCNE-assisted pathway.
The fact that compounds 8 and 9 are easier to oxidize than 4
and 5, despite the fact that they bear electron-withdrawing CO2-
Et groups, may be a result of initial stabilization of the radical
cation by the prealigned π-orbitals.
The failure of these reactions to proceed is a result of the
fact that 12, 14, and 15 do not have substituents on the bridging
carbon that provide enough stabilization of either a radical or a
positive charge. It is interesting to note that the zero-order
photochemical rearrangements of both 12 and 14 were observed
in 1,2-dichlorobenzene at 170 °C under ambient light.
We have used PM3 computed heats of formation to evaluate
the involvement of the [6,5] closed isomer, 6a, as a viable
intermediate in the TCNE-catalyzed rearrangement of 4a to 5a.
These PM3 calculations, which predict that the [6,5] closed
isomer 6a has a ∆Hf 14.4 kcal/mol higher than that of 4a,11
appear to rule out the intermediacy of 6a in the TCNE-promoted
rearrangement, which has an activation enthalpy of only 6.7 (
0.4 kcal/mol.
To further investigate the need for positive charge stabilization
on the methano bridge in these reactions, we have examined
the thermolysis of 12, 14, and 15 in the presence and absence
of TCNE. Addition of the corresponding tosylhydrazone lithium
salt to C60 generated 12 with a trace of 13. There was no
evidence for a [6,5] open isomer with the proton over the five-
1
membered ring. This isomer would have a H NMR signal
(21) Takahashi, Y.; Ohaku, H.; Morishima, S.; Suzuki, T.; Ikeda, H.;
Miyashi, T. J. Chem. Soc., Perkin Trans. 1 1995, 319-325.
(22) Isaacs, L.; Wehrsig, A.; Diederich, F. HelV. Chim. Acta 1993, 76,
1231-1250.
3360 J. Org. Chem., Vol. 71, No. 9, 2006

Isomerization of Aryl [6,5] Open Fulleroids
TABLE 3. Free Energy Difference between the [6,5] Open Fulleroids and the Corresponding [6,6] Closed Isomer as Calculated by PM3 and
Estimated from Experimental Oxidation Potentials
PM3 values
experimental value
isomer
∆Hf (kcal/mol)
∆S° (kcal/mol)
∆G° (kcal/mol)
∆∆G° (kcal/mol)
E°, ox (V)
∆E° (V)
∆∆G° (kcal/mol)
4a
5a
4b
5b
4c
5c
4d
5d
8
799.48
791.98
799.98
793.72
801.30
798.77
830.01
824.03
730.84
724.28
0.206
0.208
0.178
0.178
0.172
0.172
0.178
0.180
0.192
0.193
737.68
729.83
746.86
740.50
750.02
747.37
776.72
770.30
673.43
666.63
+1.00
+1.20
+1.06
+1.28
+1.11
+1.29
+1.11
+1.34
+0.52
+0.74
7.85
6.36
2.65
6.42
6.80
0.20
0.22
0.18
0.23
0.22
4.72
5.19
4.25
5.42
5.19
9
SCHEME 5
The reason for the modest increase in the rate of rearrange-
ment of 4a in the presence of C60 is not clear. It may be that a
zwitterion analogous to 10, with a negative charge on C60, is
an intermediate. The fact that such an intermediate is rather
sterically unfavorable may account for the small rate increase
in this system.
isomers depend on the electron-donating properties of groups
on the methano bridge, the difference in oxidation potential
between 4 and its corresponding [6,6] closed isomer 5 is always
about 0.2 V (Table 3). This observation coupled with the fact
that both 4 and 5 develop a positive charge on their methano
carbon upon oxidation raises the possibility that oxidation of
both 4 and 5 generate the same radical cation 17 (Scheme 5).
In this scenario, 4 loses an electron from the π system with
concurrent bond formation to give 17, while removal of an
electron from 5 breaks a cyclopropyl bond to form 17. An
analogous opening of a cyclopropyl ring upon electrochemical
oxidation followed by rearrangement has been reported in 1,1-
(p-methoxyphenyl)spiropentene (+0.71 V).21
The second oxidation peaks in the cyclic voltammograms of
4 and 5 (Figure 2) are probably due to the loss of an electron
from the C60 cage in 17 to form the dication 18. If both 4 and
5 form the same radical cation 17 upon oxidation, the difference
between the first and second oxidation peak for a given isomer
pair should be the same. Table 1 demonstrates that this
difference, which varies from 0.01 to 0.03 V for the isomer
pairs 4, 5a-d, is in fact the same within the experimental error
of our measurements.
Electrochemical Studies of the Methanofullerenes and
Fulleroids. An examination of the oxidation potentials of the
fulleroids and methanofullerenes 4 and 5 in Table 1 reveals two
interesting trends: (i) As the electron-donating ability of
substituents on the phenyl groups of both 4 and 5 decreases,
the compounds become harder to oxidize. (ii) The [6,6] closed
isomers are harder to oxidize than the corresponding [6,5] open
isomers but show the same substituent effect. This behavior is
not consistent with the removal of an electron from the benzene
moiety in which case the [6,5] open fulleroids and the [6,6]
closed methanofullerene would have about the same oxidation
potential. Furthermore, a substituted methoxybenzene by itself
would probably be slightly harder to oxidize than 4a. For
example, the oxidization potential of 1-methoxy-2,4,6-triethyl-
benzene has been reported at +1.24 V.23 Since the ease of
oxidation of both 4 and 5 parallels the ability of the groups on
the methano bridge to stabilize a positive charge, our results
are consistent with formation of a positive charge on the
methano carbon upon oxidation in both 4 and 5. Although the
oxidation potentials of the [6,5] open and the [6,6] closed
If our postulate that both the [6,5] open 4 and the [6,6] closed
5 isomers are oxidized to the same radical cation intermediate
17 is correct, the free energy difference between 4 and 5 closed
isomers is simply the difference in their oxidation potentials
and can be electrochemically measured. To evaluate the validity
of this assumption, we have used the PM3 method to estimate
(23) Hammerich, O.; Parker, V. D.; Ronlan, A. Acta Chem. Scand. B 30
1976, 1, 89-90.
J. Org. Chem, Vol. 71, No. 9, 2006 3361

Hall et al.
SCHEME 6
the free energies of the [6,5] open and the [6,6] closed isomers
(Table 3). These semiempirical calculations agree with the
experimental observation that the [6,6] closed isomers are more
stable than the corresponding[6,5] open isomers. Table 3
demonstrates that calculated free energy differences between 4
and 5 and those derived experimentally by assuming that both
isomers are converted electrochemically to 17 are in reasonable
agreement. If both 4 and 5 are electrochemically converted to
17, there is the possibility for electrochemical conversion of 4
to 5 under conditions in which 17 is generated. Accordingly,
we have attempted a bulk electrochemical oxidation of 4a to
generate 5a by holding the potential of a cell containing a 1.0
mmol solution of 4a at its first oxidation value (1.00 V vs Fc/
Fc+) in the absence of light. However, after maintaining
potential of the system at 1.00 V for 30 min, the potential
abruptly fell to 0.00 V with the development of polymer film
on the surface of the glassy carbon electrode that blocked the
flow of current. No 5a was detected. Apparently the radical
cation 17 is reactive toward polymerization on the surface of
the glassy carbon electrode.
Photochemical Rearrangement of Aryl-Substituted Ful-
leroids in the Presence of Electron Acceptors. We have
reported that 4a undergoes a photochemical, zero-order rear-
rangement to 5a under ambient light. At 25 °C under ambient
light, a degassed solution of 4a in ODCB rearranges (ko ) 1.94
× 10-7 M s-1) to 5a in 3.5 h (Figure 3). However, when 1
molar equiv of TCNE is added to the solution of 4a under the
same conditions at 25 °C, 4a rearranges (ko ) 2.10 × 10-5
M
s-1) to 5a in 2 min with zero-order kinetics (Figure 4). No
cycloaddition products were observed. A degassed solution of
4a with chloranil (1:1 mole ratio) in ODCB-d4, when exposed
to ambient light, rearranged at 25 °C with a zero-order rate
constant of 6.44 × 10-6 M s-1 in 20 min. These results appear
to be consistent with an excited-state electron-transfer process.
It is known that electronic excitation causes an increase in both
the oxidizing and reducing potential of a molecule that give
rise to electron-transfer reactions.24 Scheme 6 shows two
possible mechanisms in which oxidation of 4 by TCNE leads
to ring opening to form 17 and the back electron transfer from
TCNE forms the [6,6] closed isomer 5 (pathway a). Our current
data does not preclude the involvement of a radical chain pro-
cess (pathway b). An electron-transfer mechanism has been
proposed in the photochemical reactions of 1,1-(p-methoxy-
phenyl)spiropentanes and TCNE to form methylidenecyclo-
butane.21
FIGURE 3. Kinetics of the light-promoted rearrangement of 4a at 25
°C under ambient light; ko ) 1.94 × 10-7 M/s.
FIGURE 2. Cyclic voltammograms of 4a (A) and 5a (B) recorded in
0.1 mM ODCB/0.15 M TBAFP6 at 25 °C on a glassy carbon electrode
with Fc/Fc+ as internal standard. The potential scan rate was 40 mV/s.
(24) Masnovi, J. M.; Kochi, J. K. J. Org. Chem. 1985, 50, 5245-5255.
3362 J. Org. Chem., Vol. 71, No. 9, 2006

Isomerization of Aryl [6,5] Open Fulleroids
1
[0.039 g, 37% yield]. H NMR (ODCB-d4, 250 MHz): δ 2.49 (s,
1H), 1.56 (s, 9H). 13C NMR (CDCl3, 100 MHz): δ 149.0, 147.3,
146.8, 146.0, 145.0, 144.9, 144.2, 144.0, 143.8, 143.5, 143.2, 142.9,
142.8, 141.9, 141.0, 140.8, 140.3, 140.2, 139.9, 139.1, 138.6, 137.7,
136.6, 134.4, 132.1, 131.1, 130.9, 130.0, 129.7, 128.9, 61.4, 36.7,
30.8. FABMS (NBA): m/s 791 ((M + 1)+), 720 (C60+).
1
Compound 13. H NMR (ODCB-d4, 250 MHz): [6,6] closed
isomer δ 3.84 (s, 1H) 1.61 (s, 9H). 13C NMR (ODCB-d4, 100
MHz): δ 148.6, 146.8, 145.8, 145.4, 145.2, 145.0, 144.9, 144.8,
144.5, 144.1, 143.6, 143.5, 143.2, 142.8, 142.6, 142.0, 141.4, 140.5,
140.1, 74.0, 53.6, 36.8, 29.7. FABMS (NBA): m/s 791 ((M + 1)+),
720 (C60+). Although all of the signals for the ring carbons in 12
and 13 cannot be resolved, signals for the carbons of the bridging
groups and the bridgehead carbons are consistent with the structures.
The electrochemical synthesis of compound 13 has been reported.25
Compounds 14, 15, and 16 were prepared as a mixture as
previously described, and the NMR spectra data agree with those
in the literature.22
FIGURE 4. Kinetics of the light-promoted rearrangement of 4a in
the presence of TCNE at 25 °C under ambient light; ko ) 2.10 × 10-5
M/s.
General Procedure for the Catalyzed Rearrangement of [6,5]
Open to the [6,6] Closed Isomer. A 5-mm NMR tube, wrapped
with aluminum foil, was charged with the [6,5] open isomer and
0.5 mL of ODCB-d4 (ca. 3.5 mmol) at room temperature. An equal
molar amount of the potential catalyst (TCNE, TNM, C60, Chloranil,
p-DNB) was added. The NMR tube was degassed, sealed and heated
in a temperature controlled jacketed flask. The kinetics were
monitored by NMR spectroscopy.
Conclusions
These results demonstrate that both the thermal and photo-
chemical rearrangements of aryl-substituted [6,5] open fulleroids
are catalyzed in the presence of some electron acceptors. The
thermal rearrangement appears to be catalyzed via an zwitter-
onic-type intermediate, whereas the catalyzed photochemical
reaction proceeds via an excited-state electron-transfer process.
We have proposed that both [6,5] open and the [6,6] closed
isomers are oxidized electrochemically to the same radical cation
intermediate. If our proposal is correct, then the free energy
difference between the [6,5] open and the [6,6] closed isomers
is simply the difference in their oxidation potentials and can be
electrochemically measured. In all cases studied, the free energy
difference between the corresponding isomers is only about 5
kcal/mol.
Electrochemistry. Voltammetric experiments were performed
on a potentiostat, electroanalytical system equipped with a three-
electrode cell, interfaced with a plotter. The glassy carbon working
electrode, 3 mm in diameter, was used after being polished with
Alpha micropolished alumina (1.0 µm). A silver wire served as a
pseudo-reference electrode. The platinum wire, used as the counter
electrode, was cleaned by heating in a flame for approximately 30
s. The supporting electrolyte (TBAPF6,) was recrystallized from
an ethanol/H2O (95:5) mixture and dried in vacuo prior to use. The
electrochemical measurements were made by dissolving the fullerene
sample (1.0 mmol) in 15 mL of ODCB with 0.15 M TBAPF6. The
solution was protected from light and placed in the three-electrode
electrochemical cell. The cell was degassed for 20 min by bubbling
argon gas through the solution. Ferrocene was added as an internal
standard. The measurements were made at a scan rate of 40 mV/s.
Computational Studies. Calculations were carried out on a
Silicon Graphics Personal Iris workstation. Initial input structures
were generated and minimized. Semiempirical PM3 calculations
were then preformed using the Spartan 5.0 program to optimize
all stationary points. Geometry optimizations of the [6,5] open and
the [6,6] closed isomers were carried out on all input structures.
Experimental Section
We have recently reported the synthesis and characterization of
4a-e, 8, and 9.11 Compound 4a was synthesized by the reaction
of C60 with the diazo compound. Compounds 4b-e, 8, and 9 were
synthesized from the general procedure for the reaction of C60 with
tosylhydrazone lithium salts.
Preparation of Compound 12. The tosylhydrazone of 2,2-
dimethylpropanal was prepared by standard procedures. To a
solution of the tosylhydrazone (0.1 mmol) in 3 mL of hexane at 0
°C under N2 in a 250-mL flask was added MeLi (1.4 M, 0.107
mL, 0.15 mmol) using a syringe. A previously prepared solution
of C60 in toluene (72 mL, 1.4 mM, 0.1 mmol) was added to the
flask. The flask was protected from light and heated while the
solution refluxed for 30 min. The reaction mixture was monitored
by TLC in toluene or hexane/toluene (4:1). The solvent was reduced
in vacuo to 20 mL, which was then loaded on a silica gel column
and eluted with hexane/toluene. The product was compound 12
with a trace of 13. There was no evidence for the [6,5] open isomer
in which the methine proton is over the five-membered ring
Supporting Information Available: Kinetic data for the
rearrangements of 4a-d and 8; cyclic voltammograms of 4a-d,
1
5a-d, 8, and 9; and the H and 13C spectra of 12 and 13. This
material is available free of charge via the Internet at http://
pubs.acs.org.
JO0521706
(25) Boulas, P. L.; Zuo, Y.; Echegoyen, L. Chem. Commun. 1996, 1547-
1548.
J. Org. Chem, Vol. 71, No. 9, 2006 3363