(
)
314
M. Fedurco et al.rChemical Physics Letters 319 2000 309–317
Ž
.
w
x
w x
9:1 mixture is close to that in toluene:acetonitrile
ppm 19 , C60 H4, ds5.06–5.50 ppm 25 , or ds
q
y
Ž
.
Ž
w
x
9:1 mixture containing 0.1 M TBA PF6 being
4.6–5.4 ppm 26 , and C60 H18, ds3.42–5.43 ppm
.
w
x.
Ž
.
only 16 mV more negative for the latter system .
Infrared spectra for biphenyls linked to the
fullerene core via cyclopropane, and via a methine
bridge are expected to differ from each other because
of differing symmetries of both molecules and due to
27 . The signals at 4.18 dd, Jf6 Hz, Jf2 Hz
and 4.23 ppm t, Jf7 Hz cannot be assigned to
protons directly connected to the C60 . The coupling
usually measured in hydrogenated C60 is )9 Hz in
all cases reported in the literature 19,25–27 . The
transition of protons on the methylene group in
diphenylmethane solvent is usually observed at ;
4.0 ppm. If one of the three signals in the range from
4.1 to 4.4 ppm be due to an additional hydrogen on
the C60 , it is only possible if it sits symmetrically
with respect to other addends the only way to
appear as a singlet . Interesting in this respect is that
Ž
.
w
x
Ž .
the presence of hydrogen s on the fullerene core in
the former case but not in the latter. As expected, the
Ž
.
Ž
.
FTIR spectrum not shown of samples of C60 in
DPM irradiated for 15 min showed high-intensity
stretching modes around 2920 cmy1, typical for
hydrogenated fullerene and a weak C–H stretching
mode for the methine bridge. Bands between 1300
Ž
.
and 1500 cmy1 deformations of aromatic C–H
Ž
hydrogen trotting on the fullerene core has been
detected in the case of the morpholine adduct of C60
28 .
Our preliminary experiments on the irradiation of
groups and strong signals in the 700–750 cmy1
.
Ž
w
x
region out-of-plane vibrations of aromatic hydro-
.
gens were also detected. Bands at 2850 and 2825
cmy1, typical for C60 H2 19 or 2912, 2847 and
w
x
Ž
.
C60 irradiation time was 1 h in phenylacetylene
2827 cmy1 characteristic of C60 H36 20 were absent
w
x
Ž
.
PA , followed by CI MS analysis of reaction prod-
ucts, revealed that up to 11 phenylethinyl radicals
add to C60 under similar conditions as utilized in the
w x
in the IR spectrum of diphenyl fulleroid 9 .
We have attempted to separate Ph2CH–C60–H
Ž
.
from the solvent following the photolysis of C60 in
case of fullerene photolysis in DPM Fig. 5 . We
assume, that in each case, the acidic C–H bond in
PA gets labilized due to a strong interaction of
photoexcited fullerene with the delocalized system of
p-electron over the phenyl ring. Here, two possibili-
ties can be considered. First, C–H bond scission, as
discussed above, would lead to a direct C[C bond
introduction onto the C60 core and one hydrogen
atom on an adjacent carbon on the fullerene. A
second alternative would rely on cycloaddition of PA
to C60 . The corresponding derivative formed under
such conditions would have a similar molecular mass
as in the previous case but the structure would
contain a double bond connected to fullerene via two
1
Ž
.
diphenylmethane ;20 min for the H NMR exper-
iment. However, in order to distil the solvent out
Ž
.
b.p.s2648C , solutions had to be heated to rela-
Ž
tively high temperatures ;1008C under the reduced
pressure . Under such conditions, some decomposi-
.
tion of fullerene derivative andror their further
chemical reactions could have occurred in a process.
1
Fig. 4 shows H NMR spectra for such a mixture of
reaction products recorded in the region from 4.0 to
8.5 ppm. In the region of aromatic protons, only the
signal at ;8.18 ppm is compatible with the chemi-
cal shift of ortho protons of phenyl rings that are
w
x
bent toward the surface of the C60 9,21–23 . The
aromatic signals occurring in the 6.85–7.75 ppm
region are difficult to assign without additional 13C
NMR experiments. They might be, at least in part,
due to the rotation of phenyl rings in Ph2CH–C60–H,
and also due to aromatic proton signals of more than
Ž
.
single bonds with no hydrogen directly on the
fullerene core . FTIR experiments performed on
products formed upon fullerene photolysis in pheny-
lacetylene and deuterated phenylacetylene suggest
the presence of H or D atoms on the fullerene core
and, therefore, the first possibility seems to be more
plausible. The identification of asymmetric and sym-
metric stretches typical for the C[C bond around
1600 cmy1 is not straightforward in this case since
these bands are often quite weak.
w
x
one kind of phenyl ring 24 . The signals at 6.03,
6.17 and 4.92 ppm should correspond to protons that
are directly bound to the C60 moiety. Note that all
the signals are singlets. They may be due to a proton
near the C60 –methine bond, or due to pairs of nearby
protons that are chemically equivalent since no cou-
As manifested in the present work, photolysis of
fullerene solutions in DPM allows one to attach
Ž
pling structures are observable: C60 H2 ds5.94