Reactions of fullerene C60 with organometallic azides

Article abstract of DOI:10.1016/j.tetlet.2014.05.070

Reactions of fullerene C₆₀ with organometallic azides [Et ₂AlN₃, EtAl(N₃)₂ and Bu ₃SnN₃] led to novel 1-azido-2-alkylfullerenes. The structures of the products were confirmed by

Full text of DOI:10.1016/j.tetlet.2014.05.070

Tetrahedron Letters 55 (2014) 3747–3749
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Reactions of fullerene C60 with organometallic azides
⇑
Arslan R. Akhmetov , Ildar R. Yarullin, Airat R. Tuktarov, Usein M. Dzhemilev
Institute of Petrochemistry and Catalysis, Russian Academy of Sciences, 141 Prospekt Oktyabrya, Ufa 450075, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Reactions of fullerene C60 with organometallic azides [Et2AlN3, EtAl(N3)2 and Bu3SnN3] led to novel 1-
azido-2-alkylfullerenes. The structures of the products were confirmed by IR, 1H and 13C NMR spectros-
copy and MALDI TOF mass spectrometry.
Received 28 January 2014
Revised 11 March 2014
Accepted 15 May 2014
Available online 23 May 2014
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Fullerene C60
Diethylaluminium azide
Tributyltin azide
1,2-Addition
1-Azido-2-alkylfullerene
Interest in the reactions of fullerenes with organic azides is
related to the possibility to obtain functionally substituted com-
pounds which might be useful for medicine and technology.1,2
The reactions of fullerene C60 with organic azides, resulting in
the formation of [2+3]-cycloadducts—triazolinofullerenes,3 or
[2+1]-cycloadducts—5,6- and 6,6-azahomofullerenes or 5,6- and
6,6-aziridinofullerenes,4 are covered extensively in the literature.
We previously demonstrated5 that the reaction of fullerene C60
with hydrazoic acid (HN3) gave unsubstituted aziridino- or triazo-
linofullerenes, depending on the reaction conditions, while the
reactions of C60 with halogen azides (IN3, BrN3) led to the forma-
tion of novel 1-halo-2-azidofullerenes.6
conditions9 (À20 °C, followed by heating to 40 °C over 3 h, in tolu-
ene as the solvent) to give 1-azido-2-ethyl(C60-Ih)[5,6]fullerene (1)
in ꢀ20% yield (Scheme 1). Similar results were obtained when Et2-
AlN3 was replaced by EtAl(N3)2. No reaction took place, with alu-
minium triazide [Al(N3)3].
An increase in the time for the reaction between C60 and Et2AlN3
from three to 24 hours had no effect on the yield or composition of
the target azide, and replacement of toluene with chlorobenzene or
dichlorobenzene as the solvent resulted in polyadducts.
Compound 110 was isolated by preparative HPLC and character-
ized by standard analytical methods (IR, 1H and 13C NMR spectros-
copy and MALDI TOF mass spectrometry).
At the beginning of our study, no reports were available on the
reactions of fullerene C60 with organometallic azides, which, in our
opinion, could produce more stable fullerene azide derivatives.
As a further development of the above-mentioned work, and to
elaborate an efficient method for the synthesis of previously
unknown azidofullerenes, we have performed the first study of
the reactions of C60 with organometallic azides including Et2AlN3,
EtAl(N3)2 and Bu3SnN3, which were synthesized via reported
procedures.7,8
As a model system, we chose the reaction of Et2AlN3 with fuller-
ene C60 and used it to study the effects of the reaction conditions
and reactant ratio on the yield and composition of the reaction
products.
The IR spectrum of adduct 1 exhibited strong absorption bands
at 2092, 1632, 1110, 754, 552 and 526 cmÀ1, which belong to
AN@NA bond vibrations characteristic of N3 and the fullerene
core. The MALDI TOF mass spectrum showed only one ion at m/z
749.040, due to the [C60C2H5]À fragment. Unfortunately, we did
not observe
a molecular ion corresponding to the formula
C62H5N3. Apparently, MALDI ionization of molecule 1 using a UV
laser induces elimination of the azide group, and thus we only
observe the corresponding [MÀN3]À ion.
Et2AlN3
or
EtAl(N3)2
N3
Et
-20 oC to 40 oC, 3 h
toluene
It was found that C60 fullerene reacted with a fivefold excess of
+
freshly prepared diethylaluminium azide7 under the developed
1
⇑
Corresponding author. Tel./fax: +7 3472842750.
E-mail address: TuktarovAR@gmail.com (A.R. Akhmetov).
Scheme 1.
http://dx.doi.org/10.1016/j.tetlet.2014.05.070
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

3748
A. R. Akhmetov et al. / Tetrahedron Letters 55 (2014) 3747–3749
The 13C NMR spectrum of compound 1 exhibited 29 signals due to
N=PPh3
Et
N3
Et
the sp2-hybridized carbon atoms of the fullerene cage at 134–
155 ppm, and two signals for the sp3-hybridized carbon atoms of
C60 linked to the azide and ethyl groups (79.91 and 65.93 ppm,
respectively).In theHMBCexperimentonmonoadduct1, crosspeaks
were observed between the methylene hydrogen atom of the ethyl
20 oC, 2 h
toluene
PPh3
+
2
1
Scheme 2.
group and the carbon atoms in the a- and b-environment (Fig. 1).
To prove more reliably the structure of 1-azido-2-ethyl
(C60-Ih)[5,6]fullerene (1), namely the presence of the azido group
in these molecule, we have further transformed the azido group.
Thus, the reaction of azidofullerene 1 with PPh3 in dry toluene at
room temperature led to the formation of phosphine imide 2 in
quantitative yield (Scheme 2).11
The MALDI TOF mass spectrum exhibited a molecular ion peak at
m/z 1025.208 (for C80H20NP, ca. m/z 1025.995) and a fragment ion
peak at m/z 996.137 due to [MÀEt]+. In the IR spectrum of the
adduct 2,12 no absorption band at ca. 2092 cmÀ1 characteristic of
the azido group was present. The 13C NMR spectrum displayed 29
signals due to the sp2 hybridized carbon atoms of the fullerene cage,
while the sp3 hybridized carbon atoms of C60 were responsible for
the signals at 79.89 and 65.93 ppm, being due to the carbon atoms
bearing the phosphine imide and the ethyl groups, respectively. The
presence of a phosphorus atom between the nitrogen atom and the
N3
Bu
Bu3SnN3
100 oC, 1 h
Cu(OTf)2
3
Scheme 3.
All our attempts to react metal azides such as Bu2Sn(N3)2 or
Na2Sn(N3)6 with fullerene C60 failed. Each experiment resulted in
recovery of the initial fullerene.
According to thermal stability measurements of azidofullerenes
and 3, refluxing these compounds in 1,2-dichlorobenzene
1
resulted in an insoluble black precipitate, which was difficult to
characterize. This precipitate is, in our opinion, an oligomer formed
via intermolecular reaction of the starting azidofullerenes. Mean-
while, refluxing compounds 1 and 3 in toluene or chlorobenzene,
even for a long period of time (10–15 h), did not induce thermal
decomposition or further transformations.
phenyl substituents caused doublet splitting signals for the
a, b and
c
carbon atoms of the phenyl substituents in the 13C NMR spectrum
1
3
with the corresponding constants JP–C,
2JP–C and JP–C. The maxi-
mum value of the spin–spin coupling constant (SSCC) for a fixed
a
carbon atom at the phosphorus atom was 1JP–C = 45.28 Hz, while
the SSCCs of the b and
c carbon atoms of the phenyl substituent
In conclusion, we have reported for the first time the one-step
synthesis of stable 1-azido-2-alkylfullerenes by the reaction of
organometallic azides [Et2AlN3, EtAl(N3)2, Bu3SnN3] with fullerene
C60 under thermal and catalytic reaction conditions.
were 2JP–C = 12.07 Hz and 3JP–C = 10.06 Hz, respectively.
In order to extend the scope of this method, we have developed
the synthesis of alkylazidofullerenes, via the reaction of C60 with
Sn-containing azides. In analogy with Et2AlN3 and EtAl(N3)2,
Bu3SnN3 reacted with C60 fullerene under the developed condi-
tions13 (100 °C, 1 h, chlorobenzene as the solvent) only in the
Acknowledgement
presence of
1-azido-2-butyl(C60-Ih)[5,6]fullerene (3) in
a
stoichiometric amount of Cu(OTf)2 to give
This work was supported financially by the Russian Foundation
for Basic Research (Project 12-03-31023).
a
yield of ꢀ30%
(Scheme 3). In the absence of Cu(OTf)2, the reaction did not occur.
The use of Cu(OTf)2 in an equimolar amount with respect to C60
was optimal; therefore, even a slight decrease in its quantity
(for example, to 70–80 mol %) decreased sharply the yield of the
target adduct 3. An increase in the concentration of Cu(OTf)2 did
not improve the yield of 3 further.
References and notes
1. Wang, N.; Li, J.; Zhu, D. Tetrahedron Lett. 1995, 36, 431–434.
2. Yamakoshi, Y. N.; Yagami, T.; Sueyoshi, S.; Miyata, N. J. Org. Chem. 1996, 61,
7236–7237.
3. Gonzalez, S.; Martin, M.; Swartz, A.; Guldi, D. M. Org. Lett. 2003, 5, 557–560.
4. Sinyashin, O. G.; Romanova, I. P.; Yusupova, G. G.; Kovalenko, V. I.; Yanilkin, V.
V.; Azancheev, N. M. Mendeleev Commun. 1999, 96–98.
As in the case of 1, the structure of compound 314 was reliably
determined by standard analytical methods (IR, 1H and 13C NMR
spectroscopy and MALDI TOF MS).
5. Akhmetov, A. R.; Tuktarov, A. R.; Dzhemilev, U. M.; Yarullin, I. R.; Gabidullina, L.
A. Russ. Chem. Bull. 2011, 9, 1885–1887.
6. Dzhemilev, U. M.; Tuktarov, A. R.; Yarullin, I. R.; Akhmetov, A. R. Mendeleev
Commun. 2013, 23, 326–328.
7. Rawal, V. H.; Zhong, H. M. Tetrahedron Lett. 1994, 35, 4947–4950.
8. Saito, S.; Yamashita, S.; Nishikawa, T.; Yokoyama, Y.; Inaba, M.; Moriwake, T.
Tetrahedron Lett. 1989, 30, 4153–4156.
An increase in the time for the reaction between C60 and Bu3-
SnN3 from one to three hours led to the formation of polyadducts,
which were difficult to separate or identify. Replacement of chloro-
benzene by toluene, or a decrease in the reaction temperature to
40 °C resulted in yields of compound 3 not exceeding 5%.
9. Procedure for the synthesis of 1-azido-2-ethyl(C60-Ih)[5,6]fullerene (1). In a two-
necked glass reactor, C60 (20 mg, 0.0277 mmol) was dissolved in anhydrous
toluene (20 mL). The solution was cooled to À20 °C and a solution of Et2AlN3 or
EtAl(N3)2 (0.1385 mmol) in toluene (2 mL) was added with vigorous stirring.7
The cooling bath was removed and the mixture was heated to 40 °C, and
stirring was continued for an additional 3 h. All experiments were carried out
under a dry argon flow. After the reaction was complete, the mixture was
quenched with 5% HCl, and the organic phase was separated and passed
through a short silica gel layer. The reaction product 1 and unreacted C60 were
separated by preparative HPLC using toluene as the eluent. This gave 1-azido-
2-ethyl(C60-Ih)[5,6]fullerene (1) in 20% yield as a brown powder.
10. 1-Azido-2-ethyl(C60-Ih)[5,6]fullerene (1). IR: 526, 552, 754, 1110, 1632, 2092,
2853, 2924 cmÀ1. UV (CHCl3), kmax, nm: 255, 315, 428. 1H NMR (400.13 MHz,
CDCl3:CS2 1:5): d 3.66 (q, 2H, CH2, J = 7.2 Hz), 2.06 (t, 3H, CH3, J = 7.2 Hz). 13C
NMR (100.62 MHz, CDCl3:CS2 1:5): d 14.85, 34.92, 65.93, 79.91, 134.65, 136.62,
139.43, 139.78, 141.19, 141.28, 141.73, 142.05, 142.12, 142.14, 142.46, 142.58,
142.92, 144.12, 144.39, 144.68, 145.11, 145.16, 145.31, 145.46, 146.06, 146.09,
146.10, 146.27, 146.50, 147.58, 148.16, 148.99, 155.34. MALDI TOF, found: m/z
749.040 [MÀN3]À; C62H5; calculated: M = 791.048 m/z C62N3H5.
3.66
CH3
CH2
155.34
5
65.93
8
7
4
6
1
3
9
79.91
10
2
N3
Figure 1. Long-range interactions of the methylene hydrogen atoms with the
11. Procedure for the synthesis of 1-N,P,P,P-triphenylphosphine imide-2-ethyl
(C60-Ih)[5,6]fullerene (2).
A two-necked glass reactor was charged with a
fullerene cage carbon atoms in the HMBC experiment.

A. R. Akhmetov et al. / Tetrahedron Letters 55 (2014) 3747–3749
solution of 1 (10 mg, 0.012 mmol) in dry toluene (20 mL). Under a stream of chlorobenzene (4 mL), and
3749
a
solution of Bu3SnN3 (0.0554 mmol) in
dry argon at room temperature, and with vigorous stirring, PPh3 (4.3 mg,
0.018 mmol) was added. Stirring was continued for 2.5 h in the dark. After the
reaction was complete, the solvent was evaporated under reduced pressure
and the residue washed with dry Et2O (3 Â 2 mL) and dried to give compound
2 in a quantitative yield.
chlorobenzene (1 mL) and Cu(OTf)2 (10 mg, 0.0277 mmol) were added with
vigorous stirring. The mixture was heated to 100 °C over 1 h with stirring
under a dry argon flow. After the reaction was complete, the mixture was
quenched with 5% HCl, and the organic phase separated and passed through a
short silica gel layer. The reaction product 3 and C60 were separated by
12. 1-N,P,P,P-Triphenylphosphine imide-2-ethyl(C60-Ih)[5,6]fullerene (2). IR: 526, 542,
721, 1181, 1379, 1436, 2851, 2923 cmÀ1. UV (CHCl3), kmax, nm: 256, 314, 431.
1H NMR (400.13 MHz, CDCl3:CS2 1:5): d 2.06 (t, 3H, CH3, J = 7.2 Hz), 3.66 (q, 2H,
CH2, J = 7.2 Hz), 7.45–7.6 (m, 3H, 3CH), 7.6–7.7 (m, 6H, 6CH), 8.0–8.12 (m, 6H,
6CH). 13C NMR (100.62 MHz, CDCl3:CS2 1:5): d 14.90, 34.96, 65.93, 79.89, 128.6
preparative HPLC using toluene as the eluent. This gave 1-azido-2-butyl(C60-
Ih)[5,6]fullerene (3) in 30% yield as a brown powder.
14. 1-Azido-2-butyl(C60-Ih)[5,6]fullerene (3). IR: 527, 553, 574, 747, 1035, 1227,
2090, 2854, 2866, 2924, 2953 cmÀ1. UV (CHCl3), kmax, nm: 257, 319, 430. 1H
NMR (400.13 MHz, CDCl3:CS2 1:5): d 3.56 (t, 2H, CH2, J = 7.6 Hz), 2.51 (m, 2H,
CH2), 1.88 (m, 2H, CH2), 1.29 (t, 3H, CH3, J = 7.6 Hz). 13C NMR (100.62 MHz,
CDCl3:CS2 1:5): d 14.82, 24.40, 33.04, 41.75, 65.72, 80.16, 134.78, 136.92,
139.72, 140.12, 141.48, 141.56, 142.02, 142.34, 142.40, 142.44, 142.75, 142.88,
143.19, 144.43, 144.68, 144.97, 145.36, 145.38, 145.41, 145.60, 145.75, 146.36,
146.39, 146.54, 146.78, 147.86, 148.43, 149.27, 155.94. MALDI TOF, found: m/z
777.067 [MÀN3]À; C64H9; calculated: M = 819.776 m/z C64N3H9.
2
3
1
(d, JP–C = 12.07), 131.89 (d, JP–C = 10.06), 132.93 (d, JP–C = 45.28), 141.20,
141.74, 142.06, 142.12, 142.15, 142.47, 142.59, 142.90, 142.93, 144.12, 144.40,
144.68, 145.11, 145.17, 145.31, 145.46, 146.07, 146.09, 146.10, 146.27, 146.50,
148.99, 155.34. MALDI TOF, found: m/z 1025.208 [M]; calculated: m/z
1025.995; C80H20NP; m/z 996.137 [MÀC2H5]+; C78H15NP.
13. Procedure for the synthesis of 1-azido-2-butyl(C60-Ih)[5,6]fullerene (3). In a two-
necked glass reactor, C60 (20 mg, 0.0277 mmol) was dissolved in anhydrous