Selective functionalization on [60]fullerene governed by tether length

Article abstract of DOI:10.1021/ja960873m

In order to accomplish the selective synthesis of [60]fullerene bisadducts, the reactions of [60]fullerene with compounds in which two α,α'-dibromo-o-xylene moieties were connected by an oligomethylene chain (n = 2-5) were investigated. By this method, on

Full text of DOI:10.1021/ja960873m

926
J. Am. Chem. Soc. 1997, 119, 926-932
Selective Functionalization on [60]Fullerene Governed by
Tether Length
Masumi Taki,† Sachiko Sugita,† Yosuke Nakamura,† Eiji Kasashima,‡
Eiji Yashima,‡ Yoshio Okamoto,‡ and Jun Nishimura*,†
Contribution from the Department of Chemistry, Gunma UniVersity, 1-5-1 Tenjin-cho,
Kiryu 376, Japan, and Department of Applied Chemistry, Nagoya UniVersity,
Chikusa-ku, Nagoya 464-01, Japan
ReceiVed March 19, 1996. ReVised Manuscript ReceiVed NoVember 18, 1996X
Abstract: In order to accomplish the selective synthesis of [60]fullerene bisadducts, the reactions of [60]fullerene
with compounds in which two R,R′-dibromo-o-xylene moieties were connected by an oligomethylene chain (n )
2-5) were investigated. By this method, only two isomers (cis-2- and cis-3-isomers) were selectively obtained
when n ) 2 and 3, while another isomer (e-isomer) was obtained when n ) 5. When n ) 4, a complex mixture of
bisadducts was formed and has not been separated so far. cis-2-Bisadducts have been, for the first time, selectively
obtained in the fullerene chemistry. The structures of bisadducts were determined on the basis of 1H, 13C NMR, IR,
UV-vis, and mass spectroscopies. According to the NMR experiments, the symmetries of cis-2-, cis-3-, and e-isomers
were concluded to be Cs, C2, and C1, respectively. Chiral cis-3- and e-bisadducts were successfully resolved into
the respective enantiomers on a chiral HPLC column, although cis-2-bisadducts only gave a single peak. The UV-
vis spectra of cis-2-, cis-3-, and e-bisadducts were remarkably different from one another. Specifically, the e-bisadducts
showed a characteristic absorption peak around 420 nm. The cleavage of the oligomethylene chain produced the
corresponding [60]fullerene derivatives possessing two phenol moieties. These compounds are applicable to further
functionalization.
Introduction
Diederich and his co-workers6 invented a sophisticated
method, the so-called tether-directed remote functionalization.
They used tandem carbenoid and diene additions; i.e., one [6,6]
junction of fullerene was first modified by cyclopropanation,
and then two sites around the equator were selectively modified
by the Diels-Alder addition of the dienes attached at the end
of the two tethers stretched from the cyclopropane ring. They
exclusively got a headphone-shaped trisadduct.7
We are interested in a one-step tandem bisaddition occurring
within one hemisphere of [60]fullerene, using the tether-caused
constraint. Then a question arises. What tether should be
chosen in order to obtain a specific bisadduct selectively?
For the selective bisaddition, we designed R,ω-bis(3,4-bis-
(bromomethyl)phenoxy)alkanes 1, which generate o-quin-
odimethane species at both ends in situ by 1,4-elimination.2d
These reactive intermediates are expected to add stepwise to
two [6,6] junctions of fullerene and give specific bisadducts
depending on the tether length. In practice, we have successfully
achieved the selective addition of two o-quinodimethane species
by changing the tether length. In this paper, we would like to
report the selectivity, structural analysis, optical resolution, and
removal of the tether to afford fullerene derivatives possessing
phenolic hydroxy groups.
As many functionalizations on [60]fullerene have been
reported,1 the research is now strongly directed toward selective
multifunctionalization.2 Hirsch and his co-workers reported the
reactivity of a mono-cyclopropanated fullerene on the second
cyclopropanation by an R-bromocarbanion generated from
diethyl bromomalonate:2a,c i.e., the second addition is apt to
occur in the order of e, trans-3, and trans-2 or cis-3.3 Therefore,
under kinetically controlled conditions, e-bisadducts can be
obtained as major products, due to the bulkiness of reagents
and the reactivities of [6,6] junctions, although the selectivity
is not always high. For instance, when an o-quinodimethane
homologue was used for the modification of [60]fullerene,
bisadducts turned out to be a complex mixture which could not
be separated by a simple column chromatographic technique.4,5
† Gunma University.
‡ Nagoya University.
X Abstract published in AdVance ACS Abstracts, January 15, 1997.
(1) (a) Taylor, R.; Walton, D. R. M. Nature 1993, 363, 685. (b) Hirsch,
A. Angew. Chem., Int. Ed. Engl. 1993, 32, 1138. (c) Olah, G. A.; Bucsi, I.;
Aniszfeld, R.; Prakash, G. K. S. Carbon 1992, 30, 1203. (d) Schwarz, H.
Angew. Chem., Int. Ed. Engl. 1992, 31, 293. (e) Tago, T.; Minowa, T.;
Okada, Y.; Nishimura, J. Tetrahedron Lett. 1993, 34, 8461. (f) Hirsch, A.
Synthesis 1995, 895.
Results and Discussion
(2) (a) Hirsch, A.; Lamparth, I.; Gro¨sser, T. J. Am. Chem. Soc. 1994,
116, 9385. (b) Henderson, C. C.; Rohlfing, C. M.; Assink, R. A.; Cahill, P.
A. Angew. Chem., Int. Ed. Engl. 1994, 33, 786. (c) Hirsch, A.; Lamparth,
I.; Karfunkel, H. R. Angew. Chem., Int. Ed. Engl. 1994, 33, 437. (d) Belik,
P.; Gu¨gel, A.; Spickermann, J.; Mu¨llen, K. Angew. Chem., Int. Ed. Engl.
1993, 32, 78. (e) Tsuda, M.; Ishida, T.; Nogami, T.; Kurono, S.; Ohashi,
M. J. Chem. Soc., Chem. Commun. 1993, 1296. (f) Kra¨utler, B.; Maynollo,
J. Angew. Chem., Int. Ed. Engl. 1995, 34, 87.
(3) We use Hirsch’s edge labeling throughout this paper because of the
popularity, although we developed our own edge labeling. Our labeling
can express isomers more precisely than theirs, so that occasionally we
apply ours for the stereochemical descriptions of compounds in footnotes;
see: Nakamura, Y.; Taki, M.; Nishimura, J. Chem. Lett. 1995, 703.
(4) Nakamura, Y.; Minowa, T.; Tobita, S.; Shizuka, H.; Nishimura, J. J.
Chem. Soc., Perkin Trans. 2 1995, 2351.
Synthesis of Precursors 1. The tether length (n) of 1 is
limited between 2 and 5, so that the bisaddition can occur within
one hemisphere of [60]fullerene, leading to cis-2-, cis-3-, and
e-bisaddition.8
The synthesis of precursor 1 was first attempted by the
reaction of N-bromosuccinimide (NBS) and compound 2, which
(5) Nakamura, Y.; Minowa, T.; Hayashida, Y.; Tobita, S.; Shizuka, H.;
Nishimura J. J. Chem. Soc., Faraday Trans. 1996, 92, 377.
(6) Isaacs, L.; Haldimann, R. F.; Diederich, F. Angew. Chem., Int. Ed.
Engl. 1994, 33, 2339.
(7) (A,2V)-Trisadduct (see ref 3).
S0002-7863(96)00873-6 CCC: $14.00 © 1997 American Chemical Society

SelectiVe Functionalization on [60]Fullerene
J. Am. Chem. Soc., Vol. 119, No. 5, 1997 927
Scheme 1
was readily prepared from 3,4-dimethylphenol by the William-
son ether synthesis, as shown in Scheme 1. However, the radical
bromination of 2 resulted in the meager yield of desired
compound 1 with many byproducts.9
Thus, we took another route. Tetraester 3, prepared from
dimethyl 4-hydroxyphthalate, was reduced into tetraol 4 by
lithium aluminum hydride (LAH). The bromination of 4 by
phosphorous tribromide successfully afforded the desired com-
pound 1.
Figure 1. Possible isomers for cis-2-, cis-3-, and e-bisadducts. Bold
lines and circles mean addition sites and positions of the oxygen-
functional group, respectively.
Scheme 2
Product Distribution. Precursors 1 and [60]fullerene were
refluxed in toluene overnight under high-dilution conditions (ca.
10-4 M), in order to prevent two fullerene molecules from being
bridged by the tether of 1. When n ) 2, 3, and 5, major
products were easily isolated by column chromatography (silica
gel, benzene/hexane), and their purities were monitored by TLC.
On the other hand, in the case of n ) 4, many products were
detected, each in a small amount, and could not be separated
by the simple procedure mentioned above.
In fact, when n ) 2, two products, major 5a and minor 6a,
were isolated in 10% and 8% yields, respectively, as shown in
1
Scheme 2. The crude reaction mixture was analyzed by H
NMR spectroscopy, and there were no significant products other
than the isolated materials.
In the same manner, two products were isolated when n )
3. In this case, major product 5b was obtained in rather high
20% isolated yield as well as minor 6b in 9% yield. Again we
could not detect any other significant isomers in the crude
1
reaction mixture by H NMR spectroscopy.
Only one product 7 was isolated in 30% yield when n ) 5.
There were several products in the crude reaction mixture, but
their total yield did not exceed two or three percent.
The linkages of all bisadducts could be cleaved quantitatively
by the treatment of excess boron tribromide in benzene at room
temperature,10 to give the corresponding bisphenols. Note that
two cyclic bisadducts 5a and 5b gave the identical bisphenol
8, and bisadducts 6a and 6b another single product 9, according
to NMR, IR, and mass spectroscopy, as described below.
Moreover, product 7 was also transformed into the third
bisphenol 10.
1
work, H NMR, 13C NMR, and UV-vis spectroscopy gave
decisive evidence on the structural determination.
There are eight patterns with respect to the positions of
bisaddition at [6,6] junctions of [60]fullerene. Among them,
only three (cis-2, cis-3, and e) seem to be possible for 5-10
because of the tether length. For each of the three bisadditions,
two or three isomers are possible in principle, due to the
presence of a substituent (alkoxy group) on the benzene rings.
Three isomers (m-, n-, and o-forms),11 in which the two
substituents are together, apart, and mixed, respectively, are
possible for cis-2-bisadducts, as schematically illustrated in
Figure 1. In a similar way, there are three isomers (m-, n-, and
o-forms)11 for cis-3-bisadducts and two isomers (m- and
n-forms)11 for e-bisadducts.
Characterization of Bisadducts. Products 5-10 gave
correct molecular ion peaks in mass spectroscopy. In the IR
spectra, we could not find any peaks characteristic of a certain
isomer. Among the spectroscopic methods employed in this
(8) It is hard to believe that cis-1-bisaddition takes place, because of the
severe steric interaction between two benzocyclohexene moieties. Therefore,
the addition was not taken into consideration.
(9) In this reaction, the bromination is considered to take place not only
at the benzyl positions but also on the tethers and benzene rings, due to the
presence of an electron-donating alkoxy group.
(10) Iyoda, M.; Sultana, F.; Sasaki, S.; Yoshida, M. J. Chem. Soc., Chem.
Commun. 1994, 1929.
(11) According to our nomenclature, the m-, n-, and o-forms of cis-2-
bisadducts are systematically and precisely described as (A1,C11), (A1, Cu12),
and (A1,Cu11r), the m-, n-, and o-forms of cis-3-bisadducts as (A1,Du11r),
(A1, D12r), and (A1,D11r), and the m- and n-forms of e-bisadducts as (A1,Hr1)
and (A1, Hr2) (see ref 3).

928 J. Am. Chem. Soc., Vol. 119, No. 5, 1997
Taki et al.
Table 1. Retention Time in HPLC and Optical Rotation [R]D of
Enantiomers of 6a, 7, 9, and 10
HPLC retention
bisadducts
time (min)a
[R]D
concn (g/mL)
6a
19
82
31
35
47
57
36b
78b
5.5 × 103
-5.8 × 103
5.7 × 102
1.0 × 10-4 c
9.0 × 10-5 c
6.5 × 10-5 c
1.2 × 10-4 c
1.3 × 10-4 d
1.1 × 10-4 d
1.8 × 10-4 d
4.1 × 10-5 d
7
9
-5.2 × 102
5.2 × 103
-5.6 × 103
4.4 × 101
10
-7.4 × 101
a See conditions in Figure 2 unless otherwise noted. b Hexane/2-
propanol (7:3 (v/v)). c Cyclohexane. d Cyclohexane/2-propanol (1:1
(v/v)).
Figure 2. Optical resolution of (a) 5a, (b) 6a, and (c) 7 by chiral HPLC.
See Table 1 for their optical rotation. Conditions: stationary phase,
CHIRALCEL OD; eluent, hexane/2-propanol (9:1 (v/v)); flow rate, 1.0
mL/min.
In order to determine the structures of 5-10, their symmetries
were examined first. The symmetries of [60]fullerene bis-
adducts are sensitively reflected by the number of signals in
1
the H and 13C NMR spectra.12 If a bisadduct has only C1
symmetry, 6 signals for the aromatic protons and more than 60
signals for the fullerene and aromatic sp2 carbons will be
1
observed in the H and 13C NMR spectra, respectively. In a
bisadduct with C2 or Cs symmetry, however, each number will
be decreased by about half.13 The symmetries of the adducts
are also associated with their chirality; bisadducts with C1 or
C2 symmetry are chiral, while those with Cs symmetry are
achiral. Thus, the optical resolution of the enantiomers seems
to be possible for the former.
In practice, only one set (3 signals) of aromatic proton peaks
1
was observed in the H NMR spectra of bisadducts 5a,b and
6a,b. In the 13C NMR spectra, these compounds gave 30-36
peaks for the sp2 carbons of the fullerene moiety and the benzene
rings. Therefore, both 5a,b and 6a,b were concluded to have
Cs or C2 symmetry. At this stage, four isomers (m- and n-forms
of cis-2 (Cs) and cis-3 (C2)) are candidates for 5a,b and 6a,b.
The distinction between Cs and C2 symmetry, however, could
not be accomplished by the number of signals in the NMR
spectra.13
The optical resolution of 5a,b and 6a,b was attempted on a
chiral HPLC column (see the Experimental Section). As shown
in Figure 2 and Table 1, adduct 6a exhibited two peaks, which
were found to correspond to a pair of enantiomers according to
the sign and value of optical rotation.14 These results apparently
indicate that 6a is chiral, i.e., a cis-3 isomer (m- or n-form)
with C2 symmetry. On the contrary, adduct 5a afforded only
one peak in the chromatogram. Even the front and tail parts of
the peak did not give any optical rotation in the range of 200-
800 nm. This evidence clearly suggests that 5a is achiral, i.e.,
a cis-2 isomer (m- or n-form) with Cs symmetry. Adducts 5b
and 6b provided results similar to those of 5a and 6a,
respectively. A remarkable difference in retention for 5a-7
toward the chiral stationary phase (CHIRALCEL OD) consisting
of cellulose tris(3,5-dimethylphenyl carbamate)-coated silica
Figure 3. Absorption spectra of (a) 5a, (b) 6a, and (c) 7 in cyclohexane
at room temperature.
gel15 is notable. The retention in chromatography is the
consequence of the interaction between a solute and a stationary
phase, and is sensitive to the structure of a solute. The
arrangement of the two phenyl residues of 5a must be highly
suitable to induce its tight interaction with the chiral stationary.16
The electronic absorption spectra of these adducts partially
support the above assignments. As shown in Figure 3, the
spectra of 5a and 6a are quite different from each other; the
(12) Balch, A. L.; Costa, D. A.; Noll, B. C.; Olmstead, M. M. J. Am.
Chem. Soc. 1995, 117, 8926.
(13) Theoretically, the total number of sp2 carbon signals of the fullerene
moiety and the benzene rings should be different in the case of C2 (34
signals) and Cs (36 signals) symmetry. Furthermore, all the fullerene sp2
carbons should have the same intensity in bisadducts with C2, while four
fullerene sp2 carbon signals should be half as intense as the others in
bisadducts with Cs. It was, however, difficult to distinguish between C2
and Cs for 5a,b and 6a,b due to the overlapping of some signals.
(14) Their mirror-imaged CD spectra have finally proved the enantiomeric
relationship. These spectra will be reported elsewhere.
(15) Yashima, E.; Okamoto, Y. Bull. Chem. Soc. Jpn. 1995, 68, 3289.
(16) Yashima, E.; Yamamoto, C.; Okamoto, Y. J. Am. Chem. Soc. 1996,
118, 4036.

SelectiVe Functionalization on [60]Fullerene
J. Am. Chem. Soc., Vol. 119, No. 5, 1997 929
Table 2. 1H NMR Spectroscopic Data of Bisadducts 5-10 and
Monoadducts 11 and 12 (Aromatic Part Only)a
at C3b
at C5b
at C6b
adduct
(δ, d, J in Hz)c
(δ, dd, J in Hz)c
(δ, d, J in Hz)c
5a
5b
6a
6b
7
7.84 (2.5)
7.44 (2.2)
6.94 (2.5)
6.85 (2.4)
6.63 (2.8)
6.78 (2.8)
7.22 (2.2)
7.20 (2.1)
6.88 (2.3)
6.90 (2.4)
6.99 (2.4)
6.70 (2.5)
7.02 (2.5, 7.9)
6.96 (2.3, 8.1)
7.02 (2.5, 8.0)
6.99 (2.4, 8.0)
6.84 (2.6, 8.1)
6.94 (2.7, 8.2)
7.06 (2.5, 8.3)
6.78 (2.3, 8.1)
6.85 (2.5, 8.0)
6.82 (2.5, 8.3)
6.85 (2.6, 8.1)
6.57 (2.6, 8.1)
7.30 (7.9)
7.27 (8.2)
7.26 (8.0)
7.27 (7.6)
7.27 (7.9)
7.35 (8.2)
7.56 (8.2)
7.18 (7.9)
7.29 (8.0)
7.33 (8.4)
7.34 (8.4)
7.16 (8.3)
Figure 4. Possible conformers for each of the cis-2-, cis-3-, and
e-bisadducts.
11
8
9
the case of n ) 2 and 3, the steric energies of the m-form are
less than those of the n-form, especially in n ) 2 as shown in
Table 3.
For 6a,b, the distinction between the m- and n-form of cis-3
could not be achieved by NMR spectroscopy. The MM2
calculations, however, clearly indicate that the m-form11 is much
more favorable than the n-form by more than 30 kcal/mol.
Therefore, bisadducts 6a,b were concluded to be the m-form
of cis-3 as depicted in Scheme 2.
10
12
a Recorded at room temperature in CDCl3 for 5a, 6a,b, and 7 in
CDCl3/CS2 for 5b and 11, and at 75 °C in toluene-d8 for 12, and at
120 °C in DMSO-d6 for the others. b Carbon positions are shown in
structure 11. c Coupling constants are listed in parentheses.
longest absorption band extends up to more than 750 nm in 6a.
Such a remarkable difference should be brought about by the
difference in the addition sites rather than the difference in the
positional relationship of the substituents; it seems unlikely that
the three isomers of cis-2 or cis-3 exhibit quite different
absorption spectra between them. These results are consistent
with the above assignment that 5a,b and 6a,b are cis-2 and cis-
3, respectively.
In contrast with 5a,b and 6a,b, product 7 gave 59 signals of
sp2 carbons of the fullerene moiety and the benzene rings in
the 13C NMR spectrum and two sets of aromatic signals (6
signals) in the 1H NMR spectrum. Therefore, bisadduct 7
should have C1 symmetry. This is also compatible with the
fact that 7 was resolved into two peaks on the chiral HPLC
(Figure 2c and Table 1). Among the four bisadducts (o-form
of cis-2, o-form of cis-3, and m- and n-forms of e) with C1
symmetry (Figure 1), an e-isomer (m- or n-form) seems to be
the most appropriate for 7, according to the absorption spectrum.
As shown in Figure 3, the spectrum of 7 is obviously different
from those of 5a and 6a, indicating that 7 should be neither
cis-2 nor cis-3, but e-isomer. Furthermore, the rather sharp band
around 420 nm in 7, in agreement with that of the e-bisadducts
reported by Diederich et al.,6 supports the e-bisaddition. Among
the two forms (m or n) of the e-isomer, the m-form is much
more suitable than the n-form because of the extremely larger
steric energy of the latter.
Further characterization of each bisadduct was carried out
1
on the basis of H NMR spectroscopy and MM2 calculations.
1H NMR spectroscopic data are summarized in Table 2. It is
noteworthy that bisadducts 5-7 recorded sharp peaks at room
temperature in the 1H NMR spectra, apparently indicating that
the flipping motion around the cyclohexene rings is highly
restricted by the tether. Such a situation enables the presence
of three conformers (syn,syn, syn,anti, and anti,anti)17 for each
of the cis-2- and cis-3-bisadducts, and two (syn and anti) for
each of the e-bisadduct, as depicted in Figure 4. The most stable
conformation for each isomer of cis-2-, cis-3-, and e-bisadducts
was calculated by the MM2 method. In all cases, the most
stable conformation was found to be syn,syn or syn, whose steric
energies are listed in Table 3. The value for the n-form of
e-isomer was omitted due to the extremely large steric energy.
The determination of whether 5a,b are the m- or n-form of
cis-2 is as follows. The two protons at C3 (see structure 11 for
numbering) of 5a,b were resonated at lower fields than those
of other bisadducts 6a,b and 7 and monoadduct 11, suggesting
that the C3 protons in 5a,b suffer from steric compression. Such
interaction seems to be possible only in the m-form, as
demonstrated by the MM2 calculations showing that the distance
between the two C3 protons is 2.1-2.2 Å in this particular
isomer. Therefore, products 5a,b can be reasonably assigned
as the m-form11 of cis-2 as shown in Scheme 2. This assignment
is also supported by the calculations of the steric energies. In
The two isomers (o-form of cis-2 and cis-3) that were
excluded above according to the absorption spectra seem to be
1
inappropriate also on the basis of the H NMR spectra. Since
the C3 and C6 protons of 7 hardly suffer from steric compression
unlike 5a,b, the o-form of the cis-2-isomer is not suitable for
7. In the o-form of cis-3-isomer, there should be two methylene
protons which face the opposite benzene ring and resonate at
higher fields than the other methylene protons, as recognized
in 6a,b (at δ 3.40 and 3.49, compared with those of 5a,b at δ
3.78 and 3.82). In fact, only one proton was found to be shifted
to a considerably high field at δ 3.11, in agreement with the
structure of the m-form of the e-isomer: i.e., the calculated
structure of the m-form indicates that the proton is located just
above the center of the benzene ring apart by ca. 4 Å (see
structure 7 in Scheme 2). It suggests a high-field shift by the
extent of 0.4 ppm.18 Consequently, product 7 is concluded to
be the m-form of the e-bisadduct, depicted in Scheme 2.
(17) The syn and anti conformations of bisadducts are defined as
follows: when one cyclohexene ring attached at a [6,6] junction is directed
toward another addition site, the stereochemical situation is defined as syn,
and when it faces against another, this is defined as anti.
(18) Bovey, F. A.; Jelinski, L.; Mirau, P. A. Nuclear Magnetic Resonance
Spectroscopy, 2nd ed.; Academic Press: San Diego, 1988; p 110.

930 J. Am. Chem. Soc., Vol. 119, No. 5, 1997
Taki et al.
Table 3. Steric Energies (kcal/mol) in syn,syn- or syn-Conformation of Each Form of cis-2-, cis-3-, and e-Isomers Obtained by Addition of 1
(n ) 2-5)a
cis-2-isomer
n-form (Cs)
cis-3-isomer
n-form (C2)
e-isomer
n
m-form (Cs)
o-form (C1)
m-form (C2)
o-form (C1)
m-form (C1)
2
3
4
5
233
229
236
238
258
236
238
245
242
239
236
240
230
227
232
235
269
261
253
255
240
240
235
234
242
242
234
234
a Steric energies for the most stable conformation. Values with lower steric energies are underlined.
Moreover, e-bisadducts seem to be kinetically much more
accessible than the cis-3 ones, according to the experiments and
calculations by Hirsch et al.2c
Meyer, and Nambu reported chiral trans-2 and trans-3 osmium
bisadducts with achiral ligands.21 Recently, Nakamura and his
co-workers reported the synthesis of one chiral cis-3-bisadduct
enantiomer with an achiral substituent.22 They reported its
optical rotation of -1872°, which is as large as those of 6.
Further detailed comparison of circular dichroism spectra among
independently synthesized chiral bisadducts can determine the
absolute configuration of these interesting bisadducts.
These stereochemically well-defined [60]fullerene bisphenol
adducts obtained are applicable to further transformations,23 by
taking advantage of their high reactivity. For example, it seems
to be of great interest and importance to transform them into
bioactive compounds,24,25 electrolytes or functional dyes,26 pearl-
necklace shaped [60]fullerene polymers,19 and so forth.
In contrast to the cyclic fullerene adducts 5a,b, 6a,b, and 7,
the 1H NMR spectra of acyclic products 8, 9, and 10 were found
to be extremely broadened at room temperature, indicating the
flipping motion of the cyclohexene rings brought about by the
loss of the chain, though the motion is relatively slow compared
with the NMR time scale. Thus, their NMR spectra were
measured above room temperature. At 120 °C in DMSO-d6,
1
both H and 13C NMR spectra gave sufficiently sharp signals.
The number of signals in the 13C NMR spectra of 8, 9, and 10
was almost the same as those of 5, 6, and 7, respectively, except
for the disappearance of the sp3 carbons of the tether. This is
1
also the case for the H NMR spectra. Therefore, the sym-
metries in 5-7 are apparently maintained in 8-10. The signals
of the fullerene sp2 carbons in 8-10 exhibited no appreciable
shifts relative to those of 5-7. These observations undoubtedly
indicate that, in the transformation of 5-7 to 8-10, only the
cleavage of the tether took place without any rearrangement.
The absorption spectra of 8, 9, and 10 are in good agreement
with those of 5, 6, and 7, respectively, demonstrating the
preservation of each fullerene skeleton. The formation of
acyclic 8-10 was also confirmed by the appearance of a broad
band around 3400 cm-1 in the IR spectra, corresponding to the
Experimental Section
General. FAB mass spectra were taken by a Jeol JMS-MStation.
The elemental analysis was performed at the Technical Research Center
of Instrumental Analysis, Gunma University. NMR spectra were
recorded on a Jeol R-500 FT NMR spectrometer with tetramethylsilane
as an internal standard. IR spectra were taken on a JASCO FT-IR
5300 spectrophotometer. Chiral HPLC was carried out by using a
Shimadzu LC-6A HPLC apparatus (CHIRALCEL OD from Daicel
Chemical Ind. Co., 15 µm, hexane/2-propanol). Absorption spectra
were recorded on a JASCO Ubest-50 or a Hitachi U3210 spectropho-
tometer. The optical rotation [R]D was measured by a JASCO DIP370
polarimeter. The MM2 calculation was performed by Chem3D.
Benzene, toluene, and tetrahydrofuran (THF) were distilled over sodium
after prolonged heating. Other materials and reagents were com-
mercially available and used without further purification. Fullerene
was isolated from soot and purified by the reported method.27
1
O-H stretching. Naturally, the H NMR, IR, and UV-vis
spectra of 8 (or 9) obtained from 5a and 5b (or 6a and 6b)
with the different tether length were superimposable on each
other, suggesting the formation of the identical compounds from
different precursors.
Preparation of Precursor 1b (General Procedure). A mixture
of dimethyl 4-hydroxyphthalate (10.5 g, 50 mmol), 1,3-propanediol
ditosylate28 (6.5 g, 17 mmol), and potassium fluoride (3.1 g, 53 mmol)
in acetonitrile (300 mL) was refluxed for a week under a nitrogen
atmosphere. It was concentrated under reduced pressure to give a slurry
residue, which was extracted with benzene. The benzene extracts were
washed with water, dried over sodium sulfate, and concentrated under
reduced pressure. Purification by column chromatography (hexane/
AcOEt ) 2:1) gave 4.1 g (53% yield) of desired phenoxy ether 3b.
Conclusion
The selective synthesis of [60]fullerene bisadducts was
successfully carried out by the reaction of [60]fullerene and the
compounds, where two R,R′-dibromo-o-xylene moieties were
connected by an oligomethylene chain (-(CH2)n-, n ) 2-5).
By this method, cis-2- and cis-3-bisadducts were selectively
obtained when n ) 2 and 3 but only one e-bisadduct with n )
5. The selectivities are concluded to result from the subtle
balance of stereochemical and electronic effects, because such
high selectivity is usually not accomplished in the absence of
the chain.19 The cleavage of the oligomethylene chain in the
obtained bisadducts gave the corresponding [60]fullerene de-
rivatives possessing two phenol moieties. The structures of
(20) For example; Bianco, A.; Maggini, M.; Scorrano, G.; Toniolo, C.;
Marconi, G.; Villani, C.; Prato, M. J. Am. Chem. Soc. 1996, 118, 4072.
(21) Hawkins, J. M.; Meyer, A.; Nambu, M. J. Am. Chem. Soc. 1993,
115, 9844.
(22) Nakamura, E.; Isobe, H.; Tokuyama, H.; Sawamura, M. J. Chem.
Soc., Chem. Commun. 1996, 1747.
(23) Bidell, W.; Douthwaite, R. E.; Green, M. L. H.; Stephens, A. H.
H.; Turner, J. F. C. J. Chem. Soc., Chem. Commun. 1994, 1641.
(24) Tokuyama, H.; Yamago, S.; Nakamura, E. J. Am. Chem. Soc. 1993,
115, 7918.
(25) Friedman, S. H.; DeCamp, D. L.; Sijbesma, R. P.; Srdanov, G.;
Wudl, F.; Kenyon, G. L. J. Am. Chem. Soc. 1993, 115, 6506.
(26) (a) Linssen, T. G.; Du¨rr, K.; Hanack, M.; Hirsch, A. J. Chem. Soc.,
Chem. Commun. 1995, 103. (b) Khan, S. I.; Oliver, A. M.; Paddon-Row,
M. N.; Rubin, Y. J. Am. Chem. Soc. 1993, 115, 4919. (c) Iyoda, M.; Sultana,
F.; Sasaki, S.; Butenscho¨n, H. Tetrahedron Lett. 1995, 36, 579.
(27) Scrivens, W. A.; Bedworth, P. V.; Tour, J. M. J. Am. Chem. Soc.
1992, 114, 7917.
1
these [60]fullerene derivatives were determined by H NMR,
13C NMR, IR, UV-vis, and mass spectroscopy. The UV-vis
spectra of cis-2-, cis-3-, and e-bisadducts were remarkably
different from each other. Thus, the positions of bisaddition
within one hemisphere are readily discernable, unless other large
chromophores are attached to the fullerene.
Many chiral fullerene derivatives have been reported with
chiral substituents.20 There are, however, only a few examples
of chiral derivatives with achiral substituents; i.e., Hawkins,
(19) Kraus, A.; Gu¨gel, A.; Belik, P.; Walter, M.; Mu¨llen, K. Tetrahedron
1995, 36, 9927.
(28) Reinhoudt, D. N.; Jong, F. D.; Tomassen, H. P. M. Tetrahedron
Lett. 1979, 22, 2067.

SelectiVe Functionalization on [60]Fullerene
J. Am. Chem. Soc., Vol. 119, No. 5, 1997 931
To a suspension of LAH (1.3 g, 34 mmol) in dry THF (400 mL)
was added 3b (1.8 g, 3.9 mmol) dissolved in dry THF (200 mL) under
reflux. The mixture was refluxed overnight under a nitrogen atmo-
sphere. After the mixture was cooled to 0 °C, ice water was carefully
added dropwise. After gas evolution had ceased, the solution was
poured onto iced 10% hydrochloric acid (200 mL) and extracted with
ether. The ether extracts were washed with water, dried over sodium
sulfate, and concentrated under reduced pressure, giving 1.1 g (78%
yield) of white fine powdered 1,3-bis(3,4-bis(hydroxymethyl)phenoxy)-
propane (4b), which was subjected to the next reaction without further
purification.
Excess PBr3 (2.2 mL, 23 mmol) in toluene (80 mL) was added
dropwise to a solution of 4b (0.50 g, 1.4 mmol) in toluene (180 mL)
under a nitrogen atmosphere. The mixture was stirred overnight at
room temperature, poured onto ice water (300 mL), and extracted with
toluene. The toluene extracts were washed with water, dried over
sodium sulfate, and concentrated under reduced pressure to give 0.73
g (84% yield) of 1b as a white powder, which was subjected to the
next reaction without further purification.
Spectroscopic data of tetraols 4 and tetrabromides 1 are given below:
4a: 1H NMR (acetone-d6) δ 7.34 (2H, d, J ) 8.2 Hz), 7.06 (2H, d,
J ) 2.8 Hz), 6.84 (2H, dd, J ) 8.3 and 2.8 Hz), 4.67 (4H, s), 4.59
(4H, s), 4.36 (4H, s). Anal. Calcd for C18H22O4: C, 64.66; H, 6.63.
Found: C, 64.44; H, 6.65.
4b: 1H NMR (DMSO-d6) δ 7.22 (2H, d, J ) 8 Hz), 6.99 (2H, d, J
) 3 Hz), 6.77 (2H, dd, J ) 8 and 3 Hz), 5.07 (2H, t, J ) 5 Hz), 4.90
(2H, t, J ) 5 Hz), 4.50 (4H, d, J ) 5 Hz), 4.42 (4H, d, J ) 5 Hz), 4.10
(4H, t, J ) 6 Hz), 2.15 (2H, m). Anal. Calcd for C19H24O4: C, 65.50;
H, 6.94. Found: C, 65.50; H, 6.91.
142.66, 141.68, 141.31, 139.27, 138.01, 133.53, 132.63, 132.03, 131.68,
128.68, 128.33, 118.92, 116.47, 70.05, 63.75, 63.01, 43.76, 43.38; FAB/
FD MS (m/z) 986 (M+), 987 (M + H)+; UV-vis (cyclohexane, λmax
,
nm (ꢀ, M-1 cm-1)) 445.6 (4400), 372.4 (1400), 309.2 (40 000), 253.6
(87 000).
1
5b: Rf 0.48 (benzene); H NMR (CDCl3/CS2) δ 7.44 (2H, d, J )
2.2 Hz), 7.27 (2H, d, J ) 8.2 Hz), 6.96 (2H, dd, J ) 8.1 and 2.3 Hz),
4.63 (2H, m), 4.61 (2H, d, J ) 14.4 Hz), 4.33 (2H, m), 4.02 (2H, d, J
) 14.3 Hz), 3.99 (2H, d, J ) 14.7 Hz), 3.80 (2H, d, J ) 14.0 Hz),
2.49 (1H, m), 1.90 (1H, m); 13C NMR (CDCl3/CS2, 125 MHz) δ 160.64,
158.97, 156.80, 149.49, 149.15, 148.86, 147.51, 147.25, 147.16, 147.02,
146.72, 146.23, 145.86, 145.56, 145.53, 145.17, 145.03, 144.89, 144.80,
144.77, 144.55, 144.33, 144.24, 143.64, 142.58, 141.51, 141.19, 138.45,
138.43, 133.26, 132.56, 131.26, 130.24, 128.66, 117.89, 113.67, 65.28,
63.19, 62.40, 44.11, 43.50, 26.59; FAB/FD MS (m/z) 1000 (M+), 1001
(M + H)+.
1
6a: Rf 0.81 (benzene); H NMR (CDCl3) δ 7.26 (2H, d, J ) 8.0
Hz), 7.02 (2H, dd, J ) 8.0 and 2.5 Hz), 6.94 (2H, d, J ) 2.5 Hz), 4.70
(2H, m), 4.60 (2H, m), 4.39 (2H, d, J ) 13.4 Hz), 4.02 (2H, d, J )
13.4 Hz), 3.95 (2H, d, J ) 13.3 Hz), 3.41 (2H, d, J ) 13.4 Hz); 13C
NMR (CDCl3, 125 MHz) δ 158.01, 154.25, 150.87, 149.53, 149.18,
149.02, 148.60, 148.07, 147.46, 146.56, 146.19, 146.15, 145.94, 145.39,
144.86, 144.61, 143.09, 142.29, 141.96, 141.89, 141.72, 141.46, 140.00,
139.14, 138.08, 136.17, 135.02, 133.29, 131.46, 128.80 (bd), 128.42,
117.15 (bd), 114.66 (bd), 72.20, 65.11, 60.93, 43.72, 42.95; FAB MS
(m/z) 986 (M+), 987 (M + H)+; UV-vis (cyclohexane, λmax, nm (ꢀ,
M-1 cm-1)) 740.0 (680), 728.4 (580), 667.2 (730), 394.4 (8300), 260.4
(100 000).
1
6b: Rf 0.72 (benzene); H NMR (CDCl3) δ 7.27 (2H, d, J ) 7.6
4c: 1H NMR (DMSO-d6) δ 7.21 (2H, d, J ) 8.2 Hz), 6.97 (2H, d,
J ) 2.8 Hz), 6.75 (2H, dd, J ) 8.2 and 2.8 Hz), 5.06 (2H, t, J ) 5
Hz), 4.91 (2H, t, J ) 5 Hz), 4.51 (4H, d, J ) 5 Hz), 4.42 (4H, d, J )
5 Hz), 4.00 (4H, m), 1.85 (4H, m). Anal. Calcd for C20H26O4‚
0.25H2O: C, 65.51; H, 7.22. Found: C, 65.58; H, 7.28.
4d: 1H NMR (DMSO-d6) δ 7.21 (2H, d, J ) 8 Hz), 6.96 (2H, d, J
) 3 Hz), 6.74 (2H, dd, J ) 8 and 3 Hz), 5.06 (2H, t, J ) 5 Hz), 4.90
(2H, t, J ) 5 Hz), 4.51 (4H, d, J ) 5 Hz), 4.42 (4H, d, J ) 5 Hz), 3.96
(4H, t, J ) 6 Hz), 1.76 (4H, m), 1.55 (2H, m). Anal. Calcd for
C21H28O4: C, 67.00; H, 7.50. Found: C, 67.06; H, 7.37.
1a: 1H NMR (CDCl3) δ 7.30 (2H, d, J ) 8 Hz), 6.96 (2H, d, J )
3 Hz), 6.87 (2H, dd, J ) 8 and 3 Hz), 4.66 (4H, s), 4.62 (4H, s), 4.33
(4H, s); MS (m/z) 585 (M+), 583, 508, 506, 504, 502.
Hz), 6.99 (2H, dd, J ) 8.0 and 2.4 Hz), 6.85 (2H, d, J ) 2.4 Hz), 4.72
(2H, m), 4.34 (2H, d, J ) 13.4 Hz), 4.11 (2H, m), 4.00 (2H, d, J )
13.7 Hz), 3.99 (2H, d, J ) 13.7 Hz), 3.50 (2H, d, J ) 13.7 Hz), 2.28
(2H, m); 13C NMR (CDCl3, 125 MHz) δ 158.49, 154.39, 151.13,
149.55, 149.19, 148.74, 148.03, 146.23, 145.99, 145.52, 145.46, 144.98,
144.76, 142.29, 142.05, 141.92, 141.76, 141.15, 139.96, 138.91, 138.13,
136.44, 134.32, 130.17, 129.18, 128.77, 128.32, 117.93, 116.80, 112.40,
64.64, 63.13, 60.57, 43.47; FAB/FD MS (m/z) 1000 (M+), 1001 (M +
H)+.
7: Rf 0.64 (benzene); 1H NMR (CDCl3) δ 7.35 (1H, d, J ) 8.2 Hz),
7.27 (1H, d, J ) 7.9 Hz), 6.94 (1H, dd, J ) 8.2 and 2.7 Hz), 6.84 (1H,
dd, J ) 8.1 and 2.6 Hz), 6.78 (1H, d, J ) 2.8 Hz), 6.63 (1H, d, J ) 2.8
Hz), 4.50 (1H, d, J ) 12.8 Hz), 4.45 (1H, d, J ) 12.8 Hz), 4.24 (2H,
m), 4.20 (1H, d, J ) 13.4 Hz), 4.17 (1H, d, J ) 13.4 Hz), 4.07 (2H,
m), 3.90 (1H, d, J ) 13.1 Hz), 3.87 (1H, d, J ) 13.4 Hz), 3.86 (1H,
d, J ) 12.8 Hz), 3.10 (1H, d, J ) 13.4 Hz), 1.89 (1H, m), 1.76 (1H,
m), 1.62 (1H, m), 1.25-1.48 (3H, m); 13C NMR (CDCl3, 125 MHz) δ
161.90, 160.65, 156.94, 156.57, 155.55, 155.09, 154.98, 154.70, 154.56,
154.25, 150.08, 149.11, 149.02, 148.32, 148.05, 147.94, 147.90, 147.78,
147.33, 147.12, 147.09, 146.41, 145.86, 145.78, 145.60, 145.44, 145.12,
144.79, 144.70, 144.55, 144.49, 144.43, 144.40, 144.06, 143.46, 143.39,
142.95, 142.90, 142.76, 142.48, 142.38, 142.03, 141.98, 141.95, 141.56,
141.04, 140.84, 140.13, 140.10, 139.93, 139.77, 139.28, 136.27, 136.13,
136.02, 135.68, 135.55, 130.38, 129.73, 129.70, 128.15, 116.80, 116.19,
110.77, 110.19, 66.92, 65.95, 65.76, 64.73, 64.43, 48.18, 45.99, 43.34,
42.24, 27.73, 25.69, 20.66; FAB MS (m/z) 1028 (M+), 1029 (M +
H)+; UV-vis (cyclohexane, λmax, nm (ꢀ, M-1 cm-1)) 423.2 (4500),
396.4 (5000), 318.0 (32 000).
1b: 1H NMR (CDCl3) δ 7.28 (2H, d, J ) 8 Hz), 6.91 (2H, d, J )
3 Hz), 6.83 (2H, dd, J ) 8 and 3 Hz), 4.65 (4H, s), 4.60 (4H, s), 4.16
(4H, t, J ) 6), 2.26 (2H, m).
1c: 1H NMR (CDCl3) δ 7.28 (2H, d, J ) 8 Hz), 6.89 (2H, d, J )
3 Hz), 6.81 (2H, dd, J ) 8 and 3 Hz), 4.66 (4H, s), 4.61 (4H, s), 4.04
(4H, m), 1.97 (4H, m).
1d: 1H NMR (CDCl3) δ 7.27 (2H, d, J ) 8 Hz), 6.90 (2H, d, J )
3 Hz), 6.81 (2H, dd, J ) 8 and 3 Hz), 4.65 (4H, s), 4.61 (4H, s), 3.99
(4H, t, J ) 6 Hz), 1.84 (4H, m), 1.65 (2H, m); MS (m/z) 550, 548,
546, 481, 469, 467, 465.
Formation of Bisadduct 7 (General Procedure). To a boiling
toluene solution (350 mL) of [60]fullerene (0.145 g, 0.202 mmol),
potassium iodide (0.33 g, 2.0 mmol), and 18-crown-6 (2.1 g, 7.9 mmol)
was added dropwise 1d (0.140 g, 0.223 mmol) dissolved in toluene
(300 mL) under a nitrogen atmosphere. The mixture was refluxed for
15 h, cooled to room temperature, poured onto iced 10% NaOH aqueous
solution (500 mL), and extracted with toluene. The extracts were
washed with water and dried over sodium sulfate. After insoluble
materials were removed by filtration, the solvent was removed under
reduced pressure. The purification of the reaction mixture by column
chromatography on silica gel (benzene/hexane) afforded desired bis-
adduct 7 (61.5 mg, 30% isolated yield).
11: Rf 0.88 (benzene), 0.52 (benzene:hexane ) 1:1); 1H NMR
(CDCl3/CS2) δ 7.56 (1H, d, J ) 8.2 Hz), 7.22 (1H, bs, J ) 2.2 Hz),
7.06 (1H, dd, J ) 8.3 and 2.5 Hz), 4.80 (2H, d, broad), 4.76 (2H, d,
broad), 4.38 (4H, d, broad), 3.95 (3H, s); 13C NMR (CDCl3/CS2, 125
MHz) δ 159.40, 156.48 (broad), 147.58, 147.57, 146.39, 146.37, 146.14
(broad), 145.40 (broad), 145.36, 144.63, 144.61, 142.99, 142.49, 142.15,
142.11, 141.98 (broad), 141.60 (broad), 140.1 (broad), 139.11, 129.98,
128.74, 113.75, 112.95, 66.05, 65.69, 55.23, 45.41, 44.38; FAB MS
(m/z) 854 (M+).
Spectroscopic data are given below:
1
5a: Rf 0.59 (benzene); H NMR (CDCl3) δ 7.84 (2H, d, J ) 2.5
Hz), 7.30 (2H, d, J ) 7.9 Hz), 7.02 (2H, dd, J ) 7.9 and 2.5 Hz), 4.87
(2H, m), 4.62 (2H, d, J ) 14.3 Hz), 4.31 (2H, m), 4.09 (2H, d, J )
13.2 Hz), 4.06 (2H, d, J ) 13.4 Hz), 3.79 (2H, d, J ) 13.8 Hz); 13C
NMR (CDCl3, 125 MHz) δ 160.98, 158.84, 158.79, 149.70, 149.34,
149.06, 147.67, 147.25, 147.16, 146.90, 146.47, 145.81, 145.78, 145.72,
145.35, 145.27, 145.21, 145.02, 144.97, 144.56, 144.42, 144.38, 143.59,
Synthesis of 10 (General Procedure). Bisadduct 7 (61.5 mg,
0.0598 mmol) was dissolved in dry benzene (20 mL) in a flask, which
was sealed with a septum under a nitrogen atmosphere. The system
was allowed to stand at ice-bath temperature, and then excess boron
tribromide (0.2 mL, 2.3 mmol) was added through the septum by a
hypodermic syringe. The mixture was stirred overnight, brought from

932 J. Am. Chem. Soc., Vol. 119, No. 5, 1997
Taki et al.
0 °C to room temperature, and carefully quenched by the addition of
10 mL of ice water. It was then extracted with ethyl acetate. The
extracts were washed with water, dried over sodium sulfate, and
concentrated under reduced pressure to give the target molecule 10 in
quantitative yield.
J ) 2.4 Hz), 6.90 (1H, d, J ) 2.4 Hz), 6.85 (1H, dd, J ) 8.1 and 2.6
Hz), 6.82 (1H, dd, J ) 8.3 and 2.5 Hz), 4.18 (6H, broad), 3.91 (2H,
broad); 13C NMR (DMSO-d6 at 120 °C, 125 MHz) δ 161.63, 161.61,
156.28, 156.24, 155.66, 155.52, 155.38, 155.37, 155.31, 154.35, 154.25,
153.59, 149.48, 148.08, 148.06, 147.15, 147.13, 147.09, 147.04, 147.01,
146.26, 146.24, 145.94, 145.65, 145.63, 145.41, 144.86, 144.10, 143.88,
143.78, 143.73, 143.56, 143.50, 142.65, 141.95, 141.53, 140.79, 140.74,
140.45, 140.43, 139.98, 139.84, 139.82, 138.45, 137.97, 136.72, 136.66,
135.18, 135.16, 133.84, 133.80, 127.85, 127.76, 127.57, 114.41, 114.37,
113.88, 113.79, 64.68, 64.40, 64.24, 63.76, 46.16, 44.26, 43.79, 42.88,
42.52; FD MS (m/z) 960 (M+); UV-vis (cyclohexane:2-propanol )
1:1, λmax, nm (ꢀ, M-1 cm-1)) 422.8 (3800), 396.0 (4500), 315.6 (27 000).
12: Rf 0.61 (benzene:AcOEt ) 9:1), 1H NMR (toluene-d8 at 75 °C)
δ 7.16 (1H, d, J ) 8.3 Hz), 6.70 (1H, d, J ) 2.5 Hz), 6.57 (1H, dd, J
) 8.1 and 2.6 Hz), 4.18 (4H, broad); 13C NMR (toluene-d8 at 75 °C,
125 MHz) δ 157.05, 156.30, 148.02, 148.01, 146.84, 146.83, 146.59,
146.13, 145.81, 145.71 (broad), 145.08, 145.06, 143.48, 142.95, 142.66,
142.62, 142.43, 141.98, 140.54, 139.76, 130.36, ca.128, 115.46, 114.93,
66.63, 66.23, 45.59, 44.73; FAB MS (m/z) 840 (M+).
Spectroscopic data are given below:
8: Rf 0.13 (benzene:AcOEt ) 9:1); 1H NMR (DMSO-d6 at 120 °C)
δ 7.20 (2H, d, J ) 2.1 Hz), 7.18 (2H, d, J ) 7.9 Hz), 6.78 (2H, dd, J
) 8.1 and 2.3 Hz), 4.36 (4H, s, broad), 4.11 (2H, d, broad), 3.91 (2H,
d, broad); 13C NMR (DMSO-d6 at 120 °C, 125 MHz) δ 161.06, 159.06,
156.16, 148.51, 148.17, 147.85, 147.60, 146.71, 146.29, 146.14, 145.79,
145.67, 145.27, 144.66, 144.49, 144.41, 144.17, 144.01, 143.73, 143.41,
143.35, 143.05, 142.79, 142.00, 140.67, 140.00, 137.25, 132.43, 131.69,
127.73, 127.55, 127.04, 115.13, 113.87, 63.07, 61.40, 42.84, 41.52;
FD MS (m/z) 960 (M+); UV-vis (cyclohexane:2-propanol ) 1:1, λmax
,
nm (ꢀ, M-1 cm-1)) 445.2 (4200), 371.6 (11 000), 308.4 (30 000), 252.8
(67 000).
9: Rf 0.50 (benzene:AcOEt ) 9:1); 1H NMR (DMSO-d6 at 120 °C)
δ 7.29 (2H, d, J ) 8.0 Hz), 6.88 (2H, d, J ) 2.3 Hz), 6.85 (2H, dd, J
) 8.0 and 2.5 Hz), 4.26 (2H, d, broad), 4.16 (2H, d, broad), 3.99 (2H,
d, broad), 3.77 (2H, d, broad); 13C NMR (DMSO-d6 at 120 °C, 125
MHz) δ 156.95, 154.68, 152.28, 150.64, 149.46, 149.21, 148.71, 147.86,
147.46, 146.41, 146.05, 146.03, 145.87, 145.77, 145.61, 145.21, 145.18,
144.73, 142.36, 142.02, 141.72, 141.62, 141.22, 139.73, 138.63, 138.02,
136.73, 134.79, 133.49, 129.28, 128.75, 116.06, 114.79, 65.29, 61.07,
43.24, 42.85; FD MS (m/z) 960 (M+); UV-vis (cyclohexane:2-propanol
) 1:1, λmax, nm (ꢀ, M-1 cm-1)) 738.8 (460), 666.0 (570), 394.0 (7000),
260.0 (74 000).
Acknowledgment. Mass spectra were cordially taken by
Prof. Masahiko Iyoda, Tokyo Metropolitan University. This
work was partly supported by a special grant of the Ministry of
Education, Science, Sports, and Culture, Japan and the Nishida
Research Fund for Fundamental Organic Chemistry. One of
the authors (M.T.) is supported by Research Fellowships (DC1)
of the Japan Society for the Promotion of Science for Young
Scientists.
1
10: Rf 0.28 (benzene:AcOEt ) 9:1); H NMR (DMSO-d6 at 120
°C) δ 7.34 (1H, d, J ) 8.4 Hz), 7.33 (1H, d, J ) 8.4 Hz), 6.99 (1H, d,
JA960873M