R. Skanji et al. / Tetrahedron 68 (2012) 2713e2718
2717
3
. Conclusions
chromatographic conditions, a third very large yellow fraction
containing a mixture of poly-adducts as well as a pale yellow de-
posit probably containing higher poly-adducts remains in the col-
umn and at its top, respectively.
The results of the present study show that photo-addition of
GME to C60 can only occur in the presence of oxygen and can only
proceed through the correlated additions of two GME radicals. This
sequence easily leads to successive poly-additions through a hith-
erto un-described cyclizationedeamination mechanism. This se-
quence is similar to that previously described for the photo-
addition of morpholine to C60
presence of oxygen is absolutely necessary to initiate the addition
of amines to C60
4.3.2. Characterization of
the fulleropyrrolidine mono-ad-
duct. 4.3.2.1. NMR data. H NMR (200 MHz, CDCl ): (ppm): 3.93
(s, 6H, 2ꢃCH ), 4.54 (t, 1H, J ¼ 13.5 Hz, NH), 5.59 (d, 2H, J ¼ 13.6 Hz,
): (ppm): 52.9 (2C, OCH ); 74.02
1
3
d
3
1
9
13
.
According to these authors the
2ꢃCH), C NMR (100 MHz, CDCl
3
d
3
(1C, HCeN); (135.5 (1C); 136.7(2C); 139.6(1C); 139.8 (2C); 139.9
(2C); 140.0 (1C); 141.7(2C); 141.9 (2C); 142.0(2C); 142.14(2C);
142.19(1C); 142.2(4C); 142.6(4C); 143.0(1C); 143.17(1C); 144.2(2C);
1
9
.
They proposed a mechanism for amine addition
to C60 involving singlet oxygen, formed via the fullerene triplet,
reacting with the amine to lead to a neutral amine radical, which
144.3(2C);
144.9(1C);
145.2(1C);
145.3(2C);
145.43(3C);
19
can react with the fullerene. We checked that tri and di-
ethylamines as well as several amino-acid esters (leucine, lysine
and aspartic acid) do not react with C60 in the absence of oxygen
even under irradiation or heating (up to 353 K) under our experi-
mental conditions. We conclude that the mechanism of radical
145.49(2C); 145.5(1C); 145.7(2C); 146.1(4C); 146.2(2C); 146.30(2C);
146.38(2C); 147.1(2C); 149.81(2C); 149.84(1C); 152.1(1C), C60);
ꢁ
1
169.2 (2C, C]O), FT-IR (cm ): 3288(
nNH), 2944(nS(CH3)),
1740( ),1428( CeN),1255( CeO). Mass spectrometry (MS) (m/z):
n
C
]
O
n
n
þ
880 (MþH ). UVevis: 335, 407, and 430 nm.
1
9
addition proposed for amines can be generalized to amino-acids.
Thus, in theory, it is necessary to irradiate a mixture of mono-
adduct and GME (GME/C60¼2) to obtain mainly pure bis-adducts.
In the same way, it would be possible to obtain almost pure tris-
adducts from a bis-adduct or pure tetrakis-adducts from a tris-
adduct and so on, under the same conditions. The reaction rate
would be very slow if the initial GME/C60 molar ratio is stoichio-
metric (Fig. 4) and the yield would be very weak (Table 1). So, to
prepare bis-adducts with a good yield it is better to use GME in
excess and to stop the irradiation when the maximum yield is
attained. Fig. 2 represents the irradiation of a mixture of C60-GME
mono-adduct/GME (1/10, equiv/equiv) after 1 h of irradiation. In
the same way, to prepare tris-adducts with a good yield it is nec-
essary to irradiate a mixture of bis-adducts and an excess of GME
and so on for higher adducts.
4.4. HPLCeMS analysis
Basic studies of HPLC separations were performed using a P4000
multi-solvent delivery system coupled with a UV6000LP photodi-
ode array detector (Thermo Separation Products, Les Ulis, France).
Instrument monitoring and data acquisition were performed using
ChromQuest Software (ThermoQuest, Les Ulis, France). Separations
were carried out with a 4.0 mmꢃ125 mm Hypersil 120-5 ODS
cartridge (MachereyeNagel, Hoerdt, France) protected with
a 4.0 mmꢃ10 mm pre-column packed with the same stationary
ꢀ
phase. Separations of C60 and derivatives were performed at 30 C
with a flow rate set at 0.8 mL/min either under isocratic conditions
with a mixture of toluene/acetonitrile (45/55, v/v) or with a gradi-
ent elution under the following conditions: toluene/acetonitrile
(10/90, v/v) for the first 5 min, at which time the toluene was in-
4
4
. Experimental section
creased to 50% for 100 min and then hold constant for the
remaining 20 min of each sample run. At least five column volumes
of the initial composition were flushed through the column prior to
injecting the sample.
.1. General
All reagents and solvents were purchased from Across Organics
and Aldrich and used without further purification.
Mass spectra were obtained in positive linear mode on a mass
spectrometer Esquire-LC Brucker. The HPLCeAPCIeMSeMS studies
were performed with the same mass spectrometer coupled with
a HP 1100 HPLC system (HewlettePackard, Les Ulis, France). Sep-
arations were performed under the same chromatographic condi-
tions as for HPLCePDA.
4
.2. Synthesis of C60-GME products
Photochemical reactions were carried out as previously de-
scribed17 with the following modifications. Briefly, GME hydro-
chloride is dissolveddneutralized in 2 mL of methanol containing
4.5. Characterization of H
2 2 3
O and NH in the reaction mixture
1
equiv of potassium hydroxide. In order to avoid a possible trans-
esterification, we used ethanol instead of methanol in the case of
GEE. The mixture is then gently added under stirring to a solution of
Determination of singlet oxygen involvement was performed as
described previously.
Liquideliquid extraction (LLE) of GME, H O and NH was per-
2 2 3
17
C60 in toluene (1 g/L). The number of GME equivalents was varied
from 0.5 to 200. The resulting mixture was irradiated with a lumi-
formed by stirring the reaction mixture with distilled water (1/2, v/
v) during 30 min at ambient temperature. For each experiment,
a control extract was performed by subjecting the initial mixture to
LLE before irradiation.
nescent light source (500 W, distance¼30 cm) and stirred at a fixed
ꢀ
temperature (from 0 to 80 C). All reactions were carried out under
ambient atmosphere without any special caution to exclude air, as
17
oxygen is necessary for the reaction to occur.
2 2
H O characterization was performed through oxidation of io-
ꢁ
þ
dides (2I þH
2
O
2
þ2H
3
O /I
2
þ4H
2
O) monitored by direct UVevi-
in the
4
.3. Purification and characterization of C60-GME mono-
sible spectrometry. After checking the absence of H O
2 2
adduct
control extract, the specificity of the redox reaction was checked by
its enzymatic inhibition with catalase (Catalase from bovine liver,
4
.3.1. Purification. The purification of C60-GME mono-adduct was
EC No. 2325771, Fluka), which catalyzes the dismutation of H
into water and oxygen (2H /O O).
þ2H
NH and un-reacted GME were determined in the aqueous ex-
tracts obtained before and at the end of the irradiation with a JLC-
500/V amino-acid analyzer (JEOL Ltd.). The concentration of NH
2 2
O
performed on a PuriFlash 430 Evo system (Interchim, Montlu c¸ on,
France) with a 50 STD (200 g) Puriflash column and a mixture of
toluene/acetonitrile (30/70, v/v) as a mobile phase (flow rate
2
2
O
2
2
2
3
ꢀ
0 mL/min at 20 C). After discarding a first purple fraction con-
3
taining the unreacted C60, a second brownish fraction containing
the 60-GME mono-adduct is collected. Under these
was also determined by a specific enzymatic method in an auto-
mated analyzer (Modular, RochedHitachi). In this enzymatic
C