Synthesis of 18? annulenic fluorofullerenes from tertiary carbanions: size matters!

Article abstract of DOI:10.1039/b301820m

A range of tertiary carbanions XCH(CO2Et)2 of differing sizes have been reacted with C60F18 to assess the steric effect of X on the position of nucleophilic substitution. For X = CO2Et. NO2, P(O)(OMe)2, SO2CH2Ph, the all trans annulenes (trannulenes) were obtained as a result of extended S_N2' (i.e. S_N2 ) substitution; in the case of the phosphorus compound, with reduced amounts of base (DBU) dephosphonylation of one or more P(O)(OMe)2 groups by hydrogen occurred. Trannulene formation did not occur for X = F, CN due to the smaller size of the nucleophile, and in the latter case substitution was shown to take place by an S_N2' mechanism. resulting in the addend being adjacent to a fluorine addend. Trannulenes (X = CO2Et, Br, Cl) exhibited reversible one-electron reductions at potentials (-0.02 to -0.09 V) significantly more positive than for [60]fullerene. Trannulene (X = NO2) exhibited an irreversible one-electron reduction (0.08 V); the irreversibility may be associated with fluorine loss Conformational isomerism at temperatures below 298 K was observed for all trannulene derivatives as a result of eclipsing addend-addend interactions. Minimum energy conformations with a rotational energy barrier of 12-15 kcal mol^-1 were observed when these interactions arc calculated using molecular mechanics.

Full text of DOI:10.1039/b301820m

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of 18ꢀ annulenic fluorofullerenes from tertiary carbanions:
size matters!
Glenn A. Burley,*a Anthony G. Avent,a Olga V. Boltalina,b Thomas Drewello,c Ilya V. Goldt,b
Massimo Marcaccio,d Francesco Paolucci,d Demis Paolucci,d Joan M. Streete and
Roger Taylor*a
a School of Chemistry, Physics and Environmental Sciences, University of Sussex, Brighton,
UK BN1 9QJ. E-mail: G.Burley@sussex.ac.uk, R.Taylor@sussex.ac.uk
b Department of Chemistry, Moscow State University, Moscow 119899, Russia
c Department of Chemistry, University of Warwick, Coventry, UK CV4 7AL
d Dipartimento di Chimica, Università di Bologna, 40126 Bologna, Italy
e Department of Chemistry, University of Southampton, Southampton, UK SO17 1BJ
Received 14th February 2003, Accepted 11th April 2003
First published as an Advance Article on the web 30th April 2003
A range of tertiary carbanions XCH(CO2Et)2 of differing sizes have been reacted with C60F18 to assess the steric
effect of X on the position of nucleophilic substitution. For X = CO2Et, NO2, P(O)(OMe)2, SO2CH2Ph, the all
trans annulenes (trannulenes) were obtained as a result of extended S2Ј (i.e. S2Љ) substitution; in the case of the
phosphorus compound, with reduced amounts of base (DBU) dephosphonylation of one or more P(O)(OMe)2
groups by hydrogen occurred. Trannulene formation did not occur for X = F, CN due to the smaller size of the
nucleophile, and in the latter case substitution was shown to take place by an S2Ј mechanism, resulting in the
addend being adjacent to a fluorine addend. Trannulenes (X = CO2Et, Br, Cl) exhibited reversible one-electron
reductions at potentials (Ϫ0.02 to Ϫ0.09 V) significantly more positive than for [60]fullerene. Trannulene (X = NO2)
exhibited an irreversible one-electron reduction (0.08 V); the irreversibility may be associated with fluorine loss.
Conformational isomerism at temperatures below 298 K was observed for all trannulene derivatives as a result
of eclipsing addend–addend interactions. Minimum energy conformations with a rotational energy barrier of
12–15 kcal molϪ1 were observed when these interactions are calculated using molecular mechanics.
Introduction
Recently we reported the preparation of the first all-trans
18π annulenic fluorofullerene (named trannulene) via a triple
nucleophilic substitution of diethyl bromomalonate carbanion
in C60F18(Fig. 1, X = Br, RЈ = Et).1 This serendipitous discovery
Scheme 1 Generalisation of the nucleophilic attack of the α-halo-
arose because instead of formation of the expected methano-
fullerene, an extended S2Ј-derived nucleophilic substitution
(S2Љ) by the bromomalonate carbanion takes place in C60F18 at
a remote (δ) position relative to the departing fluoride anion,
giving the tris-substitution product (Scheme 1).
malonate carbanion with C60F18 (X = Br, Cl; R = Me, Et).
compounds still afforded the corresponding trannulenes. The
presence of oxide functions (ethers) in the fluorinated part of
the cage also did not affect trannulene formation.3
Three features are characteristic of trannulenes. First they
have a brilliant emerald-green colour. Secondly, they tend not
to show the parent ion in EI mass spectrometry, but rather the
C60F15ϩ fragmentation ion at 1005 amu. Thirdly, the 19F NMR
spectra show three peaks in a 1 : 2 : 2 ratio near δF Ϫ137 and
Ϫ144/Ϫ144 (coincident or nearly so).
The present work was undertaken with two aspects in mind.
First we wished to investigate the trannulation of C60F18 by
using a series of tertiary carbanions in which both the steric
bulk and the geometrical molecular arrangement close to the
tertiary nucleophilic centre are altered. This will provide further
evidence as to whether steric bulk of the nucleophile is the pre-
dominant factor in trannulene formation. Secondly, we wished
to examine the effect of the substituents on the spectroscopic
and electrochemical properties of the trannulenes.
The underlying interest in the physical properties of these
compounds is their potential for the preparation of multi-
component donor–acceptor assemblies, and molecular elec-
tronic devices. Relative to [60]fullerene, the fundamental
advantages of trannulenes lie in their improved chromophoric
features, and greater electron-acceptor properties. Strong visible
light absorption, with a maximum ca. 660–670 nm, renders
them good light-harvesting building blocks. Indeed an initial
Fig. 1 Schlegel diagrams of (a) C60F18 (᭹ = F), in which the outermost
fluorines (attached to carbons 10, 16 and 40) are nucleophilically
displaced; and (b) the general structure of [18]trannulenes (18π
annulene shown in green) (R = CO2RЈ, X = electron withdrawing group,
RЈ = Me, Et. ᭹ = F.
Semi-empirical calculations however revealed that trannu-
lation of C60F18 will occur only if the incipient carbanion is of
sufficient steric bulk.2 Further work was undertaken to deter-
mine some aspects of the size boundary, using compounds with
X = Br, RЈ = Me; X = Cl, RЈ = Et (Fig. 1b),3 however these
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2015

View Article Online
Table 1 Products of reaction of carbon acids with C60F18
products. Although the 7.4 min fraction was only present in
minor amounts, mass spectrometric analysis (Fig. 2) revealed a
molecular ion at m/z 1220 corresponding to the monoaddition
product C60F17[CF(CO2Et)2] (2). The fragment ion at m/z 1201
most likely results from the loss of fluorine attached to the
addend. The base peak at m/z 1043 corresponds to the loss of
the entire methano addend resulting in the formation of the
stable C60F17ϩcation. Altering the reaction conditions (i.e.
increasing the amount of 1a and base to 1.6 and 1.5 equiv.
respectively) produced a dramatic increase in black precipitate.
Reactant
X
Product
XЈ
Yield (%)
1a
1b
1c
1d
1e
1f
F
CN
2a, d
NA
NA
NA
CO2Et
NO2
—
—
—
3a/3ba, c
b, cor d
COCH2CH2Ph
CO2Et
NO2
P(O)(OMe)2
SO2CH2Ph
4d
5d
6d
7c
29c
33d
14c
Tracec
H
1g
SO2CH2Ph
a Reaction afforded product other than trannulene b No product was
isolated c Reagents and conditions: (i) tertiary carbon acid (1.6 equiv.),
C60F18 (1.0 equiv.), DBU (1.5 equiv.), toluene d Reagents and condi-
tions: (i) tertiary carbon acid (1.0 equiv.),C60F18 (1.0 equiv.), DBU
(0.9 equiv.), toluene
foray into a trannulene-extended tetrathiafulvalene tetrad has
proved encouraging with an energetically low lying (0.54 eV)
and long-lived charge-separated state (870 ns), generated via a
rapid intramolecular electron-transfer process.4
Results and discussion
The tertiary carbon acids (1a–g, Scheme 2) were chosen for
this study. Diethyl fluoromalonate (1a) has a single electron-
withdrawing atom attached to the tertiary carbon centre, and
provides a direct comparison with the bromo- and chloro-
analogues, shown previously to afford trannulene derivatives.
Electron withdrawal by the cyano group in diethyl cyano-
malonate (1b) should similarly aid trannulene formation, but on
the other hand its linearity produces little steric bulk at the
tertiary nucleophilic centre, and similar in fact to the fluorine
substituent. Thus compounds 1a and 1b would presumably
react in a similar fashion.
Fig. 2 EI mass spectrum (70 eV) for compound 2.
HPLC analysis revealed a decrease in recovered C60F18 but
no change in the yield of 2. This result deviates from
previous experiments using halomalonate precursors, whereby
the diethyl chloro- and bromomalonate analogues afford the
corresponding trannulenes as the major product.2,3
Compounds 3a and 3b
HPLC analysis of products from reaction of C60F18 with 1b and
DBU (see Experimental section) revealed two major peaks at
6.4 (13%) and 12.7 min (25%) accompanied by a minor amount
of unreacted C60F18. The mass spectra (Fig. 3a,b) exhibited
molecular ions at m/z 1227 and 1392, corresponding to mono-
C60F17[CCN(CO2Et)2] and bisaddition (C60F16[CCN(CO2Et)2]2,
products, 3a and 3b, respectively. Compound 3a exhibited a
base peak at m/z1043 corresponding to the formation of
C60F17ϩresulting from the loss of the addend (cf. fragmentation
of compound 2). Compound 3b revealed fragment ions at m/z
1208 and 1024, corresponding to the loss of one and two
addends respectively, giving in the latter case, C60F16ϩ. 1H NMR
analysis of both 3a and 3b revealed broad AX coupled systems
corresponding to the ethyl ester resonances.
Scheme 2 Tertiary carbon acids utilised in the reaction with C60F18.
The trigonal planar geometry of the keto function of com-
pound 1c confers a significant increase in steric bulk compared
with compounds 1a and 1b. However the presence of keto–enol
tautomers5 could provide unique substitution patterns. By con-
trast the trigonal planar ester function of 1d prevents enolis-
ation5,6 and provides another stepwise increase in steric bulk
over 1c. Compounds 1e, 1f and 1g provide yet further step-
wise increases in both the atomic radius of atoms adjacent to
the nucleophilic centre, the steric bulk of atoms surrounding
the nucleophilic centre and the geometry of the substituent
(trigonal planar for the nitro substituent in 1e; tetrahedral for
the phosphonate and sulfonate in 1f and 1g).
The 19F NMR spectrum of 3a comprises eleven fluorine
resonances, seven of which are of single intensity, and two each
of double and triple intensity (Fig. 4). Thus a total of seventeen
fluorines are present, showing 3a to be non-symmetrical. Com-
parison with the 19F NMR spectrum of the corresponding
bromo monoadduct (i.e. C60F17[CBr(CO2Et)2]) revealed signifi-
cant differences. Most notably the NMR spectrum of the
bromo compound comprised a distinct downfield resonance
(δ Ϫ107.2) corresponding to one of the outermost fluorines.
Such a resonance is not observed in the 19F NMR spectrum of
3a in which the most downfield resonance is a double intensity
peak at Ϫ130.1 ppm. Such a considerable difference in chemical
shifts of these outermost fluorines indicates a different sub-
stitution pattern. A 2D F–F COSY experiment acquired in
both d8-toluene and CDCl3 yielded the assignments of 3a
shown in Table 2.
Table 1 summarises the products derived from the reaction of
compounds 1a–g with C60F18 under Bingel conditions. Each
reaction will be discussed in detail.
Compound 2
HPLC (high pressure liquid chromatography) analysis of the
products from reaction of C60F18, 1a and DBU (see Experi-
mental section) revealed a peak at 8.2 min (4.7 ml minϪ1), a
large amount of unreacted C60F18 (38 min) and a host of minor
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2016

View Article Online
Table 2 Chemical shifts and peak assignments of the F–F COSY
spectrum of 3a (d8 toluene)
Increasing the amounts of 1b and base relative to C60F18
was found not to afford a corresponding trannulene derivative.
In addition to 3a exhibiting a unique 19F NMR spectrum
when compared to a known δ–substituted monoadduct (i.e.
C60F17[CBr(CO2Et)2]), substitution of the carbanion of 1b at
the δ-position is unlikely. Direct S2 substitution is not steric-
ally possible, leaving as the only alternative S2Ј substitution of
3a at the β-position relative to the outgoing fluorine (Fig. 5).
Such a deviation in the substitution process previously observed
for diethyl chloro and bromomalonate is due to the reduced
steric bulk of the cyano subsituent, so that substitution at the
β-position becomes favourable. We assume that the substitution
pattern of compound 2 would also be in the β-position due to
its comparable steric bulk around the tertiary nucleophilic
centre.
Chemical Shift (δF)
Fluorine Type
Connectivities
Ϫ130.1
Ϫ136.1
Ϫ136.5
Ϫ136.7
Ϫ140.4
Ϫ141.1
Ϫ142.0
Ϫ142.7
Ϫ145.8
Ϫ157.5
Ϫ157.7
A1/A2
C2/C3
C4/C5/C6
C1
D1
D3
D4
D5/D6/E
D2
B2
B1
B1, B2, C1, C4, C5, C6
D2. D3, E
B1, B2, D4, D5/D6
B1, D1
C1, D2
C3, D4
C4, D3
C2/C3, C5/C6
D1, C2
A2, C4/C5
A1, C1, C6
Fig. 5 Schlegel diagram of C60F17[CCN(CO2Et)2] 3a.
Use of the keto ester 1c
No discernable product was obtained using the keto diester 1c
in the presence of C60F18 and DBU. The presence of keto–enol
equilibria is presumed to be responsible for this.
Compound 4
Treatment of a toluene solution of 1d and C60F18 with base
afforded the desired trannulene 4 in moderate yield (29%) after
HPLC purification (Table 1). The EI mass spectrum of 4
showed a molecular ion (Mϩ) at m/z 1698 and fragment ions at
m/z 1467 and 1236 corresponding to the loss of one and two
methane tricarboxylate addends respectively. The base peak
ϩ
1
at m/z 1005 corresponds to the C60F15 cation. The H NMR
spectrum of 4 comprised an AX coupled system [δ 1.47 (t,
J = 7.1 Hz), 4.50 (br q, J = 7.1 Hz)] corresponding to the
ethyl ester of the methanetricarboxylate moiety. The 19F NMR
spectrum of 4 (Fig. 6a) showed three resonances (1 : 2 : 2
intensity), indicative of C3v symmetry, as observed in previous
trannulene derivatives.1–3 It contrasts with that of C60F18 by the
Fig. 3 EI mass spectra (70 eV) for compounds (a) 3a and (b) 3b.
Fig. 4 19F NMR (376 MHz, d8-toluene) spectrum of compound 3.
Three regiochemical possibilities arise for 3a as follows: (i)
direct substitution of one of the outermost fluorines (i.e. either
fluorine 10, 16, 40 in Fig. 1); (ii) substitution at the β-position
relative to the departing fluorine; (iii) substitution at the
δ-position relative to the departing fluorine (Scheme 1).
Fig. 6 19F NMR spectra of compounds (a) 4, (b) 5, (c) 6 and (d) 7 at
298 K.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2017

View Article Online
notable absence of the most upfield resonance (ca. Ϫ158 ppm)
which is associated with the fluorine attached to a fullerene
sp3 carbon surrounded by three fluorinated sp3 carbons (see
Fig. 1a).7 This fluorine resonance exhibits a downfield shift
(from ca. Ϫ158 to ca. Ϫ142 ppm) due to the loss of the neigh-
bouring (conjugatively electron donating) fluorine atoms.8
The lack of a 31P resonance confirmed the loss of the dimethyl
phosphonate moiety. A 1 D NOE difference experiment of 6
using a long pulse delay revealed an NOE from a singlet reson-
ance (3 H, δ 4.08) to the ABX3 multiplet of methylene protons
(12 H, δ 4.55) associated with the malonate functionality. Thus
the three phosphonate moieties have been replaced by three
hydrogen substituents. The 19F NMR spectrum of 5 comprised
a resolved set of three fluorine resonances at δF Ϫ142.9, Ϫ141.9
and Ϫ135.5 (Fig. 6c).
Compound 5
The highest yield of a trannulene product was obtained using
the commercially available diethyl nitromalonate (1e). Equi-
molar amounts of 1e, C60F18 and base, gave the trannulene 5 in
33% yield. EI mass spectrometric analysis showed only a peak
at m/z 720 due to C60ϩ. MALDI-TOF revealed an extremely
weak fragment ion at 1574 corresponding to [M Ϫ NO2]ϩ.
Stronger fragment ions at m/z 1415, 1323 and 1166 corre-
spond to C60F15[C(CO2Et)2NO2]2ϩ, C60F15[C(CO2Et)2]2ϩ, and
C60F15[C(CO2Et)2]ϩ respectively. No peak corresponding to
C60F15ϩwas observed in the positive mode, however negative
mode revealed [C60F15]Ϫ (m/z 1005) as the base peak.
A possible mechanism for the formation of 6 from the ter-
tiary carbon acid 1f is shown in Scheme 3. The first step is
presumed to be nucleophilic attack of the fullerene carbon δ to
the departing fluorine via 8 to form the phosphonate-contain-
ing trannulene 9. Dephosphonylation of quaternary phos-
phonates using potassium fluoride in triglyme was reported
by Thanappan and Burton.9 Since each nucleophilic addition
of 1f to C60F18 produces an equimolar amount of fluoride, the
fluoride is able to attack the pentavalent phosphorus centre to
produce a strong P–F bond (11). Protonation of 10 by the
conjugate acid of DBU affords 6. The low observed yields
compared to 4 and 5 can be explained by the presence of
excess DBU as a consequence of the protonation of 10.
Consistent with this explanation is the fact that when the
amount of base is reduced (0.5 equivalents compared with one
equivalent of C60F18 and 1f ) the yield of the 8.0 min fraction
rises sharply. 1H NMR of the 8.0 min fraction revealed a com-
plex multiplet (δ 4.58) corresponding to the methylene reson-
ances of the ethyl ester. At least four unique resonances in the
methyl region (δ 1.5–1.6) confirmed the presence of several
products; a doublet (δ 4.18, J = 11.3 Hz) is consistent with one
or two phosphonate moieties, whilst a singlet at δ 4.06 revealed
the dephosphorylation of at least one of them. 19F NMR analy-
sis revealed the presence of at least three trannulene products,
with one set of resonances corresponding to compound 6. Thus
reducing the amount of base present in the reaction system is a
critical factor in the formation of 6 via 9.
1
The H NMR spectrum of 5 comprised an AX coupled sys-
tem [δ 1.57 (t, J = 7.3 Hz), 4.67 (br q, J = 7.3 Hz)] associated
with the ethyl ester of the diethyl nitromalonate moiety. The
presence of the nitro function causes a downfield shift of
the methylene resonance compared with those found in the
corresponding ethyl ester (compound 4, δ 4.50) and bromo
trannulenes (δ 4.13).1–3 The effect of the nitro group on the
trannulene structure is more significant in the 19F NMR spec-
trum (Fig. 6b), which comprises three resonances at δ Ϫ142.5,
Ϫ141.9 and Ϫ135.5. The fluorines in closest proximity to the
site of functionalisation (set C in Fig. 7) exhibit a downfield
shift of over 2 ppm compared with the corresponding fluorines
of compound 4; likewise set B fluorines exhibit a downfield (but
smaller) shift (ca.1.7 ppm). By contrast the fluorines that are
further away from the functional group (set A, Fig. 7) are
shifted relatively upfield.
Compound 7
Upon addition of base to a toluene solution of C60F18 and 1g, a
colour change from lemon-yellow to pale emerald-green was
accompanied by formation of a large amount of black precipi-
tate. HPLC purification revealed only trace amounts of the cor-
responding trannulene 7. The MALDI-TOF spectrum afforded
peaks at m/z 1481, 1323 and 1163 corresponding to [C60F15-
[C(CO2Et)2]3]ϩ, [C60F15[C(CO2Et)2]2]ϩ and C60F15[C(CO2Et)2]ϩ
respectively. No [C60F15]ϩ was observed in the positive mode,
however negative mode afforded m/z 1005 [C60F15]Ϫ exclusively.
In contrast to the reaction of 1f with C60F18 and DBU, no
desulfonylation was observed, confirmed by the presence of the
benzylsulfonate function in the 1H NMR spectra.
The methylene resonances associated with the ethyl ester
afforded an ABX3 coupled system [δ 1.35 (t, 18 H, J = 7.1 Hz),
4.34 (qd, 12 H, J = 7.1, 1.1 Hz)] rather than a broad quartet
in previous (room temperature) examples. The 19F NMR
spectrum of 7 (Fig. 6d) exhibited three fluorine resonances
(δ Ϫ135.5, Ϫ142.4, Ϫ143.3) characteristic of trannulation. The
presence of the sulfonate group causes an upfield shift of the
fluorines close to the site of functionalisation (i.e. set B and C in
Fig. 7).
Fig. 7 Schlegel diagram (R = CO2Et, X = CO2Et, NO2, P(O)(OMe)2,
SO2CH2Ph; ᭹ = F) showing three sets of fluorines labelled A, B, C, in
C3v-symmetrical [18]trannulenes.
Compound 6
An unexpected result arose from the reaction of the corre-
sponding carbanion of the phosphonate diester 1f (1.0 equiv.)
with C60F18 (1.0 equiv.). Upon addition of base (0.9 equiv.), the
solution changed colour from the lemon yellow of C60F18 to the
characteristic emerald-green of a trannulene. HPLC purifi-
cation of the emerald-green solution in toluene afforded a
major peak at 7.6 min with a minor (emerald-green) fraction
eluting at 8.0 min. The MALDI-TOF spectra of neither frac-
tion showed a molecular ion corresponding to a phosphonate-
containing trannulene. Both exhibited a base peak in the
negative mode corresponding to [C60F15]Ϫ and a minor frag-
ment ion at 1163 attributed to C60F15[C(CO2Et)2]Ϫ.
Electrochemistry
Cyclic voltammetry of 4 displayed two reduction peaks,
denoted as I and II (Ep)(Fig. 8). Peak I was found to be revers-
ible (E1/2 = Ϫ0.09 V) and corresponds to a one-electron transfer,
whereas peak II was irreversible (Ep = Ϫ0.70 V). When the scan
potential is reversed after peak II, also peak I becomes less
reversible (i.e. the corresponding anodic-to-cathodic peak
current ratio is significantly less than one). Furthermore, a
broad anodic peak (peak A at ca. 1.5 V) appears as the result of
1H NMR analysis of 6 (7.6 min fraction) revealed the absence
of bis-methoxy resonances δ 3.82 and δ 3.85 (for 1f ) associated
with the phosphonate ester; suggestive of dephosphonylation.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2018

View Article Online
Scheme 3 Conjectured mechanism for dephosphonylation of compound 1f.
is probably associated with fast follow-up catalytic reactions
coupled to the charge-transfer processes.†
In the region of positive potentials, two oxidation peaks are
observed, the first (reversible) with E1/2 = 1.85 V and the second
(irreversible) with Ep = 2.15 V (at 0.5 V sϪ1). By comparison
with the one-electron reversible reduction peak, the reversible
anodic process at 1.85 V should correspond to the exchange of
two electrons.
Similar electrochemical behaviour was observed for com-
pounds 12 and 13 (Table 3). Slight differences in the E1/2 values
were observed, while the irreversible multi-electron reduction
peak remained unchanged. By contrast, significantly different
behaviour was observed for compound 5 (Fig. 9). The first
one-electron peak I, occurring at 0.08 V, was irreversible at
scan rates up to 10 V sϪ1. Reversibility of this peak was only
obtained at scan rates greater than 10 V sϪ1 (Fig. 9, inset).
Assuming that the irreversibility is due to the occurrence of a
first-order follow-up reaction (EC mechanism), a kinetic rate
Fig. 8 Cyclic voltammetric curves for a 0.5 mM solution of 4, 0.05 M
tetrabutylammonium tetrafluoroborate, tetrachloroethane solution.
Scan rate: 0.5 V sϪ1. Working electrode: Pt disc (62 µm radius). T = 25
ЊC; (a) negative switching potential: Ϫ0.45 V (——) and Ϫ0.85 V (---);
(b) negative switching potential: Ϫ1.33 V.
Fig. 9 Cyclic voltammetric curves for a 0.5 mM solution of 5, 0.05 M
tetrabutylammonium tetrafluoroborate, tetrachloroethane solution.
Scan rate: 0.5 V sϪ1. Working electrode: Pt disc (62 µm radius);
(inset) scan rate: 10 V sϪ1. Working electrode: Pt disc (12 µm radius).
T = 25 ЊC.
II and is attributed to the oxidation of a follow-up product of
the latter irreversible reduction (electrochemical–chemical –
EC-mechanism10).
At more negative potentials, an unresolved multi-electron
irreversible peak is observed (peak III Ep = Ϫ1.08 V; Fig. 8b).
This corresponds to the exchange of a much larger charge than
in previous processes (i.e. ca. 20 times compared with peak I
based on convolutive analysis9). The CV behaviour of 4 was
investigated at scan rates up to 100 V sϪ1 without any significant
change of the CV pattern, suggesting that the peak at Ϫ1.08 V
† Similar multielectron irreversible reduction processes were also
11
observed in other polyfluorinated fullerenes e.g. C60F48, C60F36
investigated under the same conditions used in this work. This would
suggest that the fluorine atoms may be involved in the mechanism of
the reactions coupled to the electron transfer processes, analogous to
the catalytic process observed in C60F18.12
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2019

View Article Online
Table 3 Electronic absorption and electrochemical data for trannulenes C60F15[C(CO2Et)2X]3, in tetrachloroethane at 25 ЊC
λ/nm (103 ε/MϪ1 cmϪ1
)
Compound
X
E1/2/V
4
5
CO2Et
NO2
Br
666
(9.1)
660
(7.3)
665
(7.7)
663
(7.9)
617a
(5.5)
613
(4.9)
618a
(4.9)
617
(5.4)
438
(8.7)
434
(8.8)
437
(8.8)
397
(24.4)
396
(24.0)
397
(24.0)
395
339
279
Ϫ0.09
0.08
(24.7)
315a
(23.5)
335
(25.0)
332a
(31.4)
(23.7)
278a
(30.7)
276
(33.2)
278a
(41.5)
12
13
Ϫ0.03
Ϫ0.02
Cl
435
(10.8)
(25.4)
a Shoulder.
constant of ca. 10 sϪ1 was estimated by digital simulation of the
CV curves. Furthermore a significant shift of the irreversible
reduction peak II (Ep = Ϫ0.56 V) and of the irreversible oxid-
ation peak A (Ep = 1.10 V) in the reverse scan was observed
(Fig. 9). The multi-electron peak III (Ep = Ϫ1.08 V) largely
replicated that of the previous species in compounds 4, 12
and 13, and is analogously attributed to an electro-catalytic
reduction process likely involving fluorine loss.
Spectroelectrochemical experiments were carried out in order
to further characterize the radical anions generated by the first
one-electron reduction of the above species. Fig. 10b shows the
absorption spectrum obtained for a 3 mM solution of 4 in tetra-
chloroethane after in-situ bulk reduction at Ϫ0.45 V, corre-
sponding to the first reversible one-electron reduction of 4.
After electrolysis, the intense absorption bands at 666 and
397 nm disappeared while those at 617, 438 and 339 nm greatly
diminished. The most interesting feature of the absorption
Ϫ
ؒ
spectrum of 4 is however the appearance of the intense and
broad band in the NIR region, centred at 952 nm. Analogous to
C60 and several of its derivatives, the NIR triplet–triplet transi-
tion represents the fingerprint of the fullerene-centred radical
anion and was recently used for the identification of the charge-
separated state generated by photoexcitation of a trannulene-
based photoactive dyad.4 Similar spectral changes were
observed for 5, 12 and 13.
In line with the observed 170 mV positive shifts of the first
reduction in 5 with respect to 4, the NIR absorption band was
also shifted in the former to 989 nm (Fig. 10c). The pristine
absorption spectra were recovered in the case of 4, 11 and 13
upon re-oxidation of the electrolysed solutions. In the case of 5,
by contrast, and in line with the irreversible CV behaviour, the
original spectrum was not recovered.
Dynamic behaviour of trannulenes as observed by NMR
A previous variable temperature NMR study on a trannulene
derivative (X = Br, 12)1 revealed broadening of the ethyl ester
resonances at temperatures below 298 K. This unusual feature
was shown to be reversible, an observation also observed for
compounds 4–7. Fig. 11 illustrates the methylene region of 4 at
various temperatures. At 298 K a single methylene quartet
(δ 4.50) is observed (Fig. 11a). As the temperature is lowered the
methylene quartet broadens significantly (Figs. 11b,c) and
moves downfield (δ 4.76, Fig. 11d). At 213 K (Fig. 11e) other
minor signals (δ 5.17, δ 4.21 and δ 3.38) appear in the spectrum
at chemical shifts unique to the methylene quartet at 298 K, and
three major signals are also present. The reversibility of this
variable temperature behaviour is indicative of conformational
isomerism. Although the addends comprise a number of bonds
that could contribute to various conformations, restricted
rotation of the carbon–carbon bond connecting the addend to
the fluorofullerene core (Fig. 12b) would most likely induce sig-
nificant perturbations in chemical shifts. Molecular mechanics
calculations (pcff forcefield)13 were performed on compound 12
(a crystal structure being available)1 in order to evaluate the
relationship of the potential energy and the dihedral angle (Φ)
Fig. 10 UV-Vis electronic absorption spectra in tetrachloroethane
soluions, 25 ЊC for (a) compounds 4 (——), 5 (- - -), 12 (- -) and
ؒ
13 (---); (b) UV-Vis-NIR spectra of 4 Ϫ(——) and 4 (---); (c) UV-Vis-
ؒ
NIR spectra of 5 Ϫ (——) and 5 (---).
ؒ
of the addends relative to each other. Due to the flexible nature
of the addends, calculations were performed with the positions
of the addend atoms fixed. The dihedral angle of a single
addend–fluorofullerene carbon bond was rotated at 15 degree
intervals and an optimisation calculation performed.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2020

View Article Online
The three most stable conformers are very similar in energy
(3–4 kcal molϪ1) with a rotational energy barrier ca. 12–15 kcal
molϪ1between each minimum energy conformation. With no
restrictions imposed on each of the respective conformations, a
molecular mechanics optimisation produces a conformation
identical to the starting crystal structure. This confirms that the
crystal structure is a true energy minimum and the rotational
energy barrier is small enough for conformations to interchange
until the respective minimum is established. A rotational energy
barrier of 12–15 kcal molϪ1 for 12 was found to be significantly
smaller than conformers found in a pyrrolidinofullerene–
calix[4]arene14 derivative (15–20 kcal molϪ1). The smaller
rotational energy barrier is reflected in the lower conformer
coalescence temperature of 12 (298 K) compared with that
(393 K) for the pyrrolidinofullerene–calix[4]arene derivative.
Fig. 11 1H NMR (500 MHz, CD2Cl2–CDCl3) spectra of the
methylene region of compound 4 at (a) 298 K, (b) 283 K, (c) 243 K,
(d) 213 K and (e) at 205 K.
Conclusions
We have demonstrated the synthetic versatility of trannulene-
formation beyond chloro- and bromomalonate carbanion addi-
tions to C60F18. This provides a range of novel trannulenes
comprising functions amenable for further chemical manipu-
lation. The steric bulk of the tertiary carbanion was confirmed
to be the predominant factor in trannulation with the fluoro-
and cyano- functions being sterically insufficient for S2Љ
subsitution. In the case of carbanion additions of diethyl cyano-
malonate (1b) the first/second example of an S2Ј subsitution
was observed giving the corresponding monoadduct C60F17-
[CCN(CO2Et)2].
All trannulenes exhibit a one-electron reduction potential
anodically shifted by nearly one volt compared withg C60. The
nitro function of compound 5 causes an irreversibility, arising
most likely from fluorine loss.
We have also demonstrated that trannnulene derivatives
exhibit conformational isomerism arising most likely from the
C–C bond connecting the addends to the fluorofullerene
nucleus. At 298 K the addends rotate with relative freedom, the
conformational freedom being reduced markedly at temper-
atures below 298 K due to eclipsing addend–addend inter-
actions. Possibilities of utilising these restricted conformations
in a functional sense such as molecular propellers and switches
will be investigated in due course.
Experimental
Toluene was distilled from sodium benzophenone ketyl.
Other sovents used were purchased from Aldrich. All reac-
tions were performed in standard glassware under an inert
atmosphere of argon. Evaporation and concentration in
vacuo utilised water-aspirator pressure, and compounds were
dried at 10Ϫ1 Torr. Flash column chromatography was per-
formed using silica 60 (230–400 mesh, 0.040–0.063 mm,
Aldrich). MALDI-TOF spectra were recorded on a Kratos
Kompact MALDI IV (Kratos Inc.) mass spectrometer in posi-
tive and negative ion modes using 2-[(2E)-3-(4-tert-butyl-
phenyl)-2-methylprop-2-enylidene]malononitrile as matrix.
Electrospray mass spectra were recorded on a Bruker
FT-MS APEX-III. 1H, 19F and 13C NMR spectra were acquired
at 500, 376.5 and 75 MHz, respectively. 19F NMR spectra used
either CDCl3 or d8-toluene as solvent. HPLC separations
employed a 10 × 250 mm Cosmosil Buckyprep column oper-
ated at a flow rate of either 4.7 or 2.0 mL minϪ1 using toluene as
eluent.
Fig. 12 (a) Calculated potential energy profile (pcff ) as a function
of dihedral angle of the C–C bond between the addend and the
fluorofullerene. (b) Schlegel diagram of 12 (R = CO2Et; X = Br)
illustrating the rotation of the addend–fluorofullerene C–C bond.
Fig. 12a illustrates the energy profile of 12 as a function of
Φ. The lowest energy minimum was found to correspond to
the starting crystal structure. Two other energy minima were
found at dihedral angles resulting from a 120Њ and 240Њ
rotation from the starting crystal structure conformation. The
corresponding three energy maxima resulted from a 60Њ
rotation from an energy minima. Inspection of the conform-
ations adopted at the respective energy maxima and minima
revealed the potential energy of a conformation reaches a
maximum when eclipsing interactions (at dihedral angles of
Ϫ125Њ and Ϫ6Њ) of the ethyl ester addends occurs. The lower
energy maximum conformation (at dihedral angle 131Њ) cor-
responds to an eclipsing interaction between the bromo and
ethyl ester substituents. Thus, the mimimum energy conform-
ations correspond to such conformations whereby these
eclipsing interactions are minimised (at dihedral angles 4Њ, 55Њ
and Ϫ65Њ).
Electrochemical instrumentation and measurements
The one-compartment electrochemical cell was of airtight
design with high-vacuum glass stopcocks fitted with either
Teflon or Viton O-rings in order to prevent contamination
by grease. The connections to the high-vacuum line and to
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2021

View Article Online
the Schlenck flask containing the solvent were obtained by
spherical joints also fitted with Viton O-rings. The pressure
measured in the electrochemical cell prior to performing the
trap-to-trap distillation of the solvent was typically 1.0 to 2.0 ×
10Ϫ5 mbar. The working electrode consisted either of a Pt disc
ultramicroelectrode (r = 62 or 12 µm) sealed in glass. The
counter electrode consisted of a platinum spiral and the quasi-
reference electrode was a silver spiral. The quasi-reference
electrode drift was negligible for the time required by a single
experiment. Both the counter and the reference electrode were
separated from the working electrode by ∼0.5 cm.
Potentials were measured with the ferrocene standard and
are referred to saturated calomel electrode (SCE). E1/2 values
correspond to (Epc ϩ Epa)/2 from CV. For irreversible peaks, the
peak potential, Ep, is given, measured at 0.5 V sϪ1. Ferrocene
was also used as an internal standard for checking the electro-
chemical reversibility of a redox couple. The temperature
dependence of the ferrocinium–ferrocene couple standard
potential was measured with respect to SCE by a non-
isothermal arrangement.
Voltammograms were recorded with a AMEL Model 552
potentiostat or a custom-made fast potentiostat controlled by
either an AMEL Model 568 or ELCHEMA Model FG-206F
function generator. Data acquisition was performed by a
Nicolet Model 3091 digital oscilloscope interfaced to a PC.
Temperature control was accomplished within 0.1 ЊC with a
Lauda RL6 CS thermostat. The DigiSim 3.03 software by Bio-
analytical Systems Inc., or the Antigona software developed by
Dr L. Mottier (http://www.ciam.unibo.it/electrochem.html/),
were used for the simulation of the CV curves. The latter pro-
gram was also used for the acquisition of the CV curves and
their convolutive analysis.
The spectroelectrochemical experiments were carried out
using a quartz OTTLE cell with a 0.03 cm path length. Temper-
ature control was achieved by a special cell holder with quartz
windows, in which two nitrogen fluxes (one at room temper-
ature and the other at low temperature) are regulated by two
needle valves. All the spectra have been recorded by a Varian
Cary 5 spectrophotometer.
chloride (20 mL), and brine (3 × 20 mL) to neutrality. The
organic layer was dried (MgSO4), and concentrated in vacuo.
Flash column chromatography (silica gel), eluting with
DCM : petroleum spirit (80 : 20) provided the desired product
1
(0.408 g, 30%) as a pale yellow solid. H NMR: δ 1.31 (t, 6 H,
J 6.9 Hz), 4.30 (q, 4 H, J 6.9 Hz), 4.71 (s, 2 H), 4.72 (s, 1 H), 7.40
(m, 5 H). 13C NMR δ 13.8, 58.8, 63.4, 70.6, 126.0, 128.9, 129.3,
131.4, 161.2. MS(ES): m/z 337.0721 [M ϩ Na]ϩ.
Preparation of Trannulenes
Compound 2. Treatment of a toluene solution of equimolar
amounts of C60F18 (3.0 mg, 2.82 µmol) and 1a (0.8 mg, 4.52
µmol) with DBU (0.6 mg, 4.24 µmol) at room temperature
produced a lemon yellow solution which gradually turned light
brown with the formation of large amounts of black precipi-
tate. The reaction mixture was stirred for a further 10 min and
filtered. HPLC (4.7 mL minϪ1) afforded fractions at 8.2 min and
38.0 min (C60F18). Concentration of fractions gave trace
amounts of 2 and recovered C60F18 (1.0 mg). MS(EI): m/z 1220
[M]ϩ, 1201 [M Ϫ F]ϩ, 1043 [C60F17]ϩ (100%).
Compounds 3a/3b. Using 1b under similar reaction condi-
tions to the reaction of 1a and C60F18 with base, gave a pale
yellow–brown solution with only a minimal amount of black
precipitate. DBU (1.5 mg, 10.2 µmol) was added to a solution
of C60F18 (12.0 mg, 11.30 µmol) and 1b (2.1 mg, 11.30 µmol)
in toluene at room temperature. The solution changed colour
from lemon yellow to light brown. The reaction mixture
was stirred for a further 10 min and filtered. HPLC (flow rate
4.7 mL minϪ1) gave fractions at 6.4 min, 12.7 min and 38.0 min
(C60F18). Concentration of these fractions afforded 3b (2.0 mg,
13%), 3a (4.0 mg, 25%) and C60F18 (2.0 mg) respectively.
3a: UV/vis (DCM): 343 nm, 442, 505. 1H NMR: δ 4.42 (br q,
4 H), 1.45 (br t, 6 H, J 7.3 Hz). 19F NMR (d8-toluene): δ Ϫ157.7
(m, 1 F), Ϫ157.5 (m, 1 F), Ϫ145.9 (d, 1 F, J 30.1 Hz), Ϫ142.7
(m, 3 F), Ϫ142.0 (d, J 26.4 Hz, 1 F), Ϫ141.1 (d, J 26.4 Hz, 1 F),
Ϫ140.4 (d, J 30.1 Hz, 1 F), Ϫ136.7 (br s, 1 F), Ϫ136.45 (br s,
3 F), Ϫ136.15 (br s, 2 F), Ϫ130.15 (br s, 2 F). MS(EI): 1227
[M]ϩ, 1043 [C60F17]ϩ (100%).
3b: UV/vis (DCM): 344 nm, 444, 513. 1H NMR: δ 4.41 (br m,
8 H), 1.43 (br t, 6 H, J7.3 Hz), 1.41 (br t, 6 H, J 7.3 Hz).
MS(EI): 1392 [M]ϩ, 1208 [C60F16[CCN(CO2Et)2]]ϩ, 1024
[C60F16]ϩ (100%).
Preparation of C60F18
Most of the C60F18used in this study was prepared by fluorin-
15
ating [60]fullerene with K2PtF6 and was supplemented by
material obtained by fluorinating with a mixture of MnF3–
K2NiF6, (a lower yield process). Fuller details together with
relative and absolute yields, and description of the large-scale
HPLC purification process will be described elsewhere.16
Compound 4. DBU (1.9 mg, 12.7 µmol) was added to a solu-
tion of C60F18 (15.0 mg, 14.1 µmol) and 1d (3.3 mg, 14.1 µmol)
in toluene at room temperature. The solution changed colour
from lemon-yellow to emerald-green. The reaction mixture was
stirred for a further 10 min and filtered. HPLC (2.0 mL minϪ1
)
Preparation of addends
gave a single fraction at 7.7 min. Concentration of the toluene
solution gave the product as a green solid (7.0 mg, 29%). UV/vis
[DCM, (ε)]: 440(3690) nm, 614(2590), 666(3760). 1H NMR:
δ 1.47 (t, 27 H, J 7.1 Hz), 4.50 (br q, H, J 7.1 Hz). 19F NMR:
δ Ϫ144.24 (s, 6 F), Ϫ144.23 (s, 6 F), Ϫ136.7 (s, 3 F). MS(EI):
m/z 1698 [M]ϩ, 1467 [C60F15[C(CO2Et)3]2]ϩ, 1236 [C60F15-
[C(CO2Et)3]]ϩ, 1005 [C60F15]ϩ (100%).
Compounds 1a, 1d and 1e were obtained from Aldrich, and 1b
was prepared by reaction of the enolate of ethyl cyanoacetate
with ethyl chloroformate.17 Compound 1c was prepared by
reaction of the enolate of diethyl malonate with 3-phenyl-
propionyl chloride,18 and 1f was prepared by reaction of
the enolate of ethyl dimethylphosphonoacetate with ethyl
chloroformate.9
Compound 5. DBU (0.8 mg, 5.08 µmol) was added to a solu-
tion of C60F18 (6.0 mg, 5.65 µmol) and 1e (1.2 mg, 5.65 µmol) in
toluene at room temperature. The solution changed colour
from lemon-yellow to emerald-green. The reaction mixture was
Diethyl 2-(2-phenylmethanesulfonyl)malonate (1g)
A suspension of NaH (0.209 g, 5.21 mmol, 60% dispersion
in oil) was added to a solution of diethyl malonate (0.696 g,
4.35 mmol) in DMF (5 mL) at Ϫ20 ЊC. The reaction mixture
was allowed to warm to room temperature and stirred for
a further 30 min. The reaction mixture was then recooled
to Ϫ20 ЊC and 2-phenylmethanesulfonyl chloride (0.995 g,
5.22 mmol) was added. The reaction mixture was allowed to
warm to room temperature, stirred for a further 3 h, then
quenched with saturated ammonium chloride (20 mL), diluted
with DCM (200 mL) and washed with saturated ammonium
stirred for a further 10 min and filtered. HPLC (4.7 mL minϪ1
)
gave fractions at 3.2 min and 38.0 min. Concentration of the
toluene solution provided the desired product as a green solid
(3.0 mg, 33%) and recovered C60F18 (1.0 mg). UV/vis (DCM):
1
435 nm, 609, 659. H NMR: δ 1.57 (t, 18 H, J 7.3 Hz), 4.67
(br q, 12 H, J 7.3 Hz). 19F NMR: δ Ϫ142.5 (s, 6 F), Ϫ141.9
(s, 6 F), Ϫ135.5 (s, 3 F). MALDI-TOF (ϩve ion mode): m/z
1574 [M Ϫ NO2]ϩ, 1415 [C60F15[C(CO2Et)2NO2]2]ϩ, 1323
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2022

View Article Online
[C60F15[C(CO2Et)2]2]ϩ, 1166 [C60F15[C(CO2Et)2]]ϩ, 1005 (Ϫve
References
mode only, [C60F15]Ϫ).
1 X.-W. Wei, A. D. Darwish, O. V. Boltalina, P. B. Hitchcock,
J. M. Street and R. Taylor, Angew. Chem., Int. Ed., 2001, 40, 2989.
2 X.-W. Wei, A. G. Avent, O. V. Boltalina, A. D. Darwish,
P. W. Fowler, J. P. B. Sandall, J. M. Street and R. Taylor, J. Chem.
Soc., Perkin Trans. 2, 2002, 41.
3 A. D. Darwish, I. V. Kouvytchko, A. G. Avent, I. V. Gol’dt,
A. K. Abdul-Sada, O. V. Boltalina and R. Taylor, J. Chem. Soc.,
Perkin Trans. 2, 2002, 1118.
4 G. A. Burley, A. G. Avent, O. V. Boltalina, I. V. Gol’dt, D. M. Guldi,
M. Marcaccio, F. Paolucci, D. Paolucci and R. Taylor, Chem.
Commun., 2003, 148.
5 J. P. Guthrie, Can. J. Chem., 1979, 57, 1177.
6 G. R. Newkome and G. R. Baker, Org. Prep. Proc. Int., 1986, 18,
117.
Compound 6. DBU (0.6 mg, 4.24 µmol) was added to a solu-
tion of C60F18 (5.0 mg, 4.71 µmol) and 1f (1.3 mg, 4.71 µmol) in
toluene at room temperature. The solution changed colour
from lemon yellow to emerald green. The reaction mixture was
stirred for a further 10 min and filtered. HPLC (flow rate
2.0 mL minϪ1) gave fractions at 7.6 min and 8.0 min. UV/vis
1
(DCM): 392 nm, 443 (sh), 613, 666. H NMR: δ 4.55 (m, AB
part of ABX3 spin system, 12 H, J 10.4, 7.2 Hz), 4.08 (s, 3 H),
1.53 (t, 18 H, J 7.3 Hz). 19F NMR: δ Ϫ142.9 (s, F), Ϫ141.9
(s, F), Ϫ135.5 (s, F). MALDI-TOF (Ϫve ion mode): m/z 1163
[C60F15[C(CO2Et)2]]Ϫ, 1007 [C60F15]Ϫ (100%).
7 A. D. Darwish, A. K. Abdul-Sada, A. G. Avent, J. M. Street and
R. Taylor, J. Fluorine Chem., 2003, 121, in press.
8 P. W. Fowler, private communication to R. Taylor.
9 A. Thanappan and D. J. Burton, J Org. Chem., 1991, 56, 273.
10 A. J. Bard and L. R. Faulkner, Electrochemical Methods:
Fundamentals and Applications. Wiley, New York, 2001, ch. 6.
11 F. Zhou, G. J. Van Berkel and B. T. Donovan, J. Am. Chem. Soc.,
1994, 116, 5485; N. Liu, Y. Morio, F. Okino, H. Touhara,
O. V. Boltalina and V. K. Pavlovich, Synth. Met., 1997, 86, 2289.
12 K. Okhubo, R. Taylor, O. V. Boltalina, S. Ogo and S. Fukuzumi,
Chem. Commun., 2002, 1952.
13 Molecular mechanics calculations (pcff forcefield) were performed
on a Silicon Graphics O2 Workstation using the Cerius 2 software
package (www.accelrys.com).
14 T. Gu, C. Bourgogne and J.-F. Nierengarten, Tetrahedron Lett.,
2001, 42, 7249.
Compound 7. DBU (0.6 mg, 4.24 µmol) was added to a solu-
tion of C60F18 (3.0 mg, 2.82 µmol) and 1g (1.4 mg, 4.52 µmol) in
toluene at room temperature. The solution changed colour
from lemon-yellow to a pale green. The reaction mixture was
stirred for a further 10 min and filtered. HPLC (flow rate
2.0 mL minϪ1) afforded a major fraction at 7.7 min. Concen-
tration of the toluene solution provided trace amounts of the
1
product as a green solid. UV/vis (DCM): H NMR: δ 1.35 (t,
18 H, J 7.1 Hz), 4.34 (qd, 1 H, J 7.1, 1.1 Hz), 4.75 (s, 6 H), 7.41
(m, 15 H), 7.47 (m, 6 H); 19F NMR: δ Ϫ143.3 (s, 6 F), Ϫ142.4 (s,
6 F), Ϫ135.5 (s, 3 F). MALDI-TOF (ϩve ion mode): m/z 1481
[C60F15[C(CO2Et)2]3 ϩ H]ϩ, 1323 [C60F15[C(CO2Et)2]2]ϩ, 1163
[C60F15[C(CO2Et)2]]ϩ, 1005 (Ϫve mode only, [C60F15]Ϫ).
15 O. V. Boltalina, V. Yu Markov, R. Taylor and M. P. Waugh, Chem.
Commun., 1996, 2549.
16 A. D. Darwish, A. G. Avent, A. K. Abdul-Sada, O. V. Boltalina,
I. Kouvytchko, P. B. Hitchcock and R. Taylor, Org. Biol. Chem.,
in preparation.
17 J. I. G. Cadogan, D. H. Hey and J. T. Sharp, J. Chem. Soc. C, 1966,
1743.
Acknowledgements
We thank the EPSRC, the Volkswagen Foundation (grant I-77/
855) and the Russian Foundation for Basic Research (grant
03-03-32855) for financial support.
18 M. W. Rathke and P. J. Cowan, J. Org. Chem., 1985, 50, 2622.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 0 1 5 – 2 0 2 3
2023