ESR study of reactivity of the complex (η2-C 60)Os(CO)(PPh3)2(CNBut) toward free radicals

Article abstract of DOI:10.1007/s11172-008-0048-1

Addition of the ^?P(O)(OPrⁱ)₂, Me ^?, Et^?, ^?Bu^t, and Cl₃C^? radicals to the (ν²-C ₆₀)Os(CO)-(PPh₃)₂(CNBu^t

Full text of DOI:10.1007/s11172-008-0048-1

Russian Chemical Bulletin, International Edition, Vol. 57, No. 2, pp. 309—315, February, 2008
309
ESR study of reactivity of the complex
2
(η ꢀC60)Os(CO)(PPh3)2(CNBut) toward free radicals
R. G. Gasanov,ꢀ M. V. Tsikalova, E. V. Martynova, and Yu. N. Novikov
A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences,
28 ul. Vavilova, 119991 Moscow, Russian Federation.
Fax: +7 (499) 135 5085
2
Addition of the •P(O)(OPri)2, Me•, Et•, •But, and Cl3C• radicals to the (η ꢀC60)Os(CO)ꢀ
(PPh3)2(CNBut) complex (1) was studied by ESR spectroscopy. The spectral parameters of the
spinꢀadducts of these radicals with complex 1 were determined. The predominant direction of
the attack by the •P(O)(OPri)2, •But, and Cl3C• radicals are the cisꢀ1 and cisꢀ2 bonds of the
fullerene molecule. The stability of the spinꢀadducts depends substantially on the nature of
the added radical. The addition rate constants of the•P(O)(OPri)2, •But, and Cl3C• radicals to
complex 1 and the dimerization rate constants for these spinꢀadducts were determined.
Key words: fullerenes, complexes, radical reactions, spinꢀadducts, electron paramagnetic
resonance, regiomers, addition and dimerization rate constants, competitive kinetics.
According to an analysis of structural data on diverse
fullerene C60 derivatives, the addition of various groups to
a C60 molecule results in deformation of the fullerene
carbon cage and distortion of the curvature of its surface.
As a rule, the most changes in the pyramidal character of
the carbon atoms are observed for the cisꢀn bonds (see
Refs 1—4). The distortion of the cage in the fullerene
derivatives, most likely, is a factor determining their
reactivity toward free radicals and the regioselectivity of
addition of the latter.
the SA toward free radicals. In addition, it is important to
elucidate whether these regularities are retained for
the formation and stability of the SA of complex 1 with
the Cꢀcentered radicals Me•, Et•, and •But, Cl3C•
and to determine reasons for the stability of the SA. Here
we report the results of these studies.
Experimental
The procedure of the synthesis of complex 1 and its structure
are given in Ref. 8. The compound Hg[P(O)(OPri)2]2 (3) was
synthesized according to a known procedure.5
For the fullerene complexes containing the metal atom
with different ligand environments, the electronic effect
of the metal fragment and steric hindrance created by the
ligand environment of the metal atom added to fullerene
can be a substantial factor favoring their reactivity toward
free radicals, the regioselectivity of radical addition, and
the stability of the spinꢀadducts (SA), along with a change
in the fullerene surface curvature. We have previously5
ESR spectra were recorded on a Varianꢀ12A spectrometer.
Samples placed in glass ampules were deaerated in several
“freezing—evacuation—thawing out” cycles. The ampules filled
with the reaction mixture were irradiated in the spectrometer
resonator with a DRShꢀ1000 lamp at room temperature. The
bands with λ = 366 and 640 nm were separated by the glass color
filters. The R1 radicals were generated by the photochemical
decomposition of compound 3, and the Ме•, Еt•, •But, and
•CCl3 radicals were formed by halogen atom abstraction from
MeI, EtBr, BrCMe3, and CCl4, respectively, by the •Re(CO)5
radical generated by the photochemical decomposition of
Re2(CO)10.9 The spin trap for the 2,3,5,6ꢀМе4C6HNO radicals
(ND) were introduced into the ampule as a solution in methylene
chloride. The radical concentration was calculated using
TEMPO (3•10–3 mol L–1) (Aldrich) as standard, and the
gꢀfactor was determined using the Varian reference standard.
The gꢀfactors of the SA of the Cꢀcentered radicals with complex
1 were calculated by comparing the experimental spectra and
simulated spectra. The spectra were simulated by the Simfonia
program, and all kinetic results were processed using the
Microsoft Excel program. The accuracy of determination of
•
reported that, when the P(O)(OPri)2 phosphoryl radical
(R1) reacts with the (η2ꢀC60)Os(CO)(PPh3)2(CNBut)
complex (1) to form SA, the R1 radical adds to that hemiꢀ
sphere of the complex in which the metal fragment is
localized. It should be emphasized that during this reacꢀ
tion, unlike the С60M(PPh3)2 complexes (M = Pt (2a),
Pd (2b)),6,7 no demetallation of complex 1 occurs and the
identified SA are not dimerized for a prolong time after
the irradiation ceased.5
The properties observed for the SA of compound 1
with R1 make it possible to study both the kinetics of SA
formation and their recombination and to reveal the role
of the steric and electronic factors in the reactivity of
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 2, pp. 301—307, February, 2008.
1066ꢀ5285/08/5702ꢀ0309 © 2008 Springer Science+Business Media, Inc.

310
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
Gasanov et al.
the addition rate constants were 20%. Relative rate constants
for the formation of the SA of the •But or •CCl3 radicals with
complex 1 were determined from the simulated spectra by
comparing in pairs of the integral intensity of the signal with
g = 2.0028 or 2.0037 with the integral intensity of the particularly
SA
Aa
Ab
Ас
aP/G
62.0
59.5
55.7
gꢀFactor
2.0072
2.0073
2.0070
SA
Ad
Ae
Af
aP/G
gꢀFactor
2.0072
2.0075
2.0072
53.8
50.5
48.5
The line width ∆Н is 0.9 G for the five first regiomers, and
chosen singlet.
1•
When studying the reactions of the SA [R
C
Os(CO)ꢀ
that for the latter is 1.3 G (see Fig. 1).
60
(PPh3)2(CNBut)] (A) with the Me•, Et•, and •But radicals, we
used the complexes [RCo(DH)2py] (R = Me (4), Et (5); (DH)2
is bisdimethylglyoximate)10,11 and Me3CNO (6),12 whose
irradiation (λ = 640 nm) produces the corresponding radicals.
The procedure included the preliminary generation of the SA,
then the opening of the ampules, the introduction of a benzene
solution of compounds 4, 5, or 6, and the sealing of the ampules
followed by irradiation with λ = 640 nm in the resonator of the
ESR spectrometer.
Unlike the SA С60R1 (aР = 63.6 G, g = 2.0023)
•
and the SA of the R1 radicals with methanoꢀ or
pyrrolidinofullerenes, the signals from the Aa—f radicals
remain unchanged in intensity after the radiation ceased
and decrease approximately twofold for 2 days. This comꢀ
parison of the stability of the SA suggests that the carbon
atoms of the cisꢀ1 and cisꢀ2 bonds in compound 1 are
subjected to the attack by the R1 radical, because the
substituents at the metal atom and the added radical screen
the radical center in such a way that the SA recombinaꢀ
tion becomes hindered. In fact, if assuming that the most
pyramidal carbon atoms primarily undergo attack by the
free radicals,3,4 these atoms in complex 1 are С(5), С(8),
С(13), С(19), С(22), and С(28), of which the first two
atoms are attributed to the cisꢀ1 bond and the others refer
to the cisꢀ2 bond. Although the C(5) atom (346.5°) is
more pyramidal than the С(13) (347.6°), С(19) (347.8°),
С(22) (347.5°), and С(28) (347°) atoms,8 the bulky subꢀ
stituents at the osmium atom can prevent the addition of
the R1 radical to this atom. The Аc and Аf radicals are
assumed to form upon the addition of the R1 radicals to
the С(5) and С(8) atoms (347.7°), and the Аa, Ab, Ad, and
Ae radicals are formed by the addition to the С(13), С(19),
С(22), and С(28) atoms of compound 1.
Results and Discussion
Spinꢀadducts of metallofullerene 1 with the diisopropylꢀ
phosphoryl radicals. The irradiation with the light of the
mercury lamp5 with λ = 366 nm of a benzene solution
with [1] = 10–3 mol L–1 and [3] = 8.8•10–3 mol L–1
results in the appearance of the ESR spectral lines
(Fig. 1) from six regiomeric SA (Aa—f), whose formation
can be described by the reactions
nm
Hg[P(O)(OPri)2]2
•P(O)(OPri)2,
R1
(1)
3
R1 + (η2ꢀC60)Os(CO)(PPh3)2(CNBut)
1
[R1(η2ꢀ•С60)Os(CO)(PPh3)2(CNBut)].
Aa—f
(2)
The parameters of the formed regiomeric SA are given
below.
a
A study of the reactivity of methanoꢀ and pyrroꢀ
lidino[60]fullerenes on the basis of the calculated data
found3,4 that in each of the cisꢀn bonds involved in free
radical addition one carbon atom becomes flatter, whereas
the pyramidal character of another atom increases. It is
most likely that, in this case, the С(6), С(7), С(11), С(12),
С(20), and С(24) atoms of the С(5)—С(6), С(8)—С(7),
С(28)—С(11), С(13)—С(12), С(19)—С(20), and
С(23)—С(24) bonds become flatter, i.e., the radical cenꢀ
ter in radicals Ai is localized at the listed above carbon
atoms, namely, at the atoms arranged closely to the
osmium atom. This assumption is consistent with the
following facts. First, the gꢀfactors of all isomers
exhibit noticeable positive shifts relative to the
*
*
*
*
*
*
I•103 (arb. units)
10.0
b
5.0
0
–5.0
–10.0
3440
3460
3480
3500
H/G
Fig. 1. Experimental (a) and simulated (b) ESR spectra of the
spinꢀadducts of the •P(O)(OPri)2 radicals with complex 1 and
nitrosodurene obtained upon the irradiation with λ = 366 nm of
benzene solutions of compounds ND ([ND]0 = 5•10–3 mol L–1),
1 ([1]0 = 2•10–3 mol L–1), and [P(O)(OPri)2]2Hg (8•10–3 mol L–1).
The spinꢀadducts with ND are marked with asterisk.
•
gꢀfactor of C60R1 and also to the gꢀfactors of the SA of

Spinꢀadducts of (η2ꢀC60)Os(CO)(PPh3)2(CNBut)
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
311
the R1 radicals with methanoꢀ and pyrrolidino[60]fullerenes
(g = 2.0027).3,4,13 This is due to the electronic effect of the
metal fragment, because the pyramidal character of the
carbon atoms at the cisꢀn bonds in complex 1 does not
differ from that for methanoꢀ or pyrrolidinofullerenes.
Second, the HFC constants of an unpaired electron with
the phosphorus nucleus of the Ai regiomers are lower than
analogous constants for the SA of the R1 radicals with С60
and methanoꢀ or pyrrolidinofullerenes, also indicating the
influence of the metal fragment on the unpaired electron.
Thus, the stability of the A radicals is determined by
steric hindrance created by the ligands at the osmium
atom and the added R1 radical. In fact, for the interaction
of the A radicals with the Me•, Et•, and •But radicals, the
intensities of the signals from the А radicals decrease
according to the logarithmic law
nonꢀfiltered light of the mercury lamp, the signals appear
from the unstable regiomers formed due to the addition of
the R1 radicals to the carbon atoms of other bonds in
addition to the signals from the stable SA.
It can be assumed that only the stable SA are formed
under the chosen experimental conditions. However, for
the generation of the reaction even by the nonꢀfiltered
light of the mercury lamp, the ESR spectrum does not
differ from that obtained when the light with λ = 366 nm
was used. Therefore, this can be related to the rate conꢀ
stants for the addition of the R1 radicals to the carbon
atoms of the cisꢀ1 and cisꢀ2 bonds, which should be someꢀ
what higher than the corresponding rate constant for the
addition to the carbon atoms of the eꢀ and transꢀn bonds.
For this purpose, we determined the rate constants of R1
radical addition to complex 1.
•
Rate constant of addition of the P(O)(OPri)2 radicals
logAa = –0.0264t + 1.6442 (r2 = 0.8621),
logAb = –0.0261t + 1.8788 (r2 = 0.9242),
logAd = –0.024t + 1.7872 (r2 = 0.8934),
logAe = –0.0205t + 1.736 (r2 = 0.8699),
to compound 1. The rate constant for the addition of the
R1 radicals to compound 1 (k2) was determined by the
competitive kinetics method. The addition of R1 to
nitrosodurene (ND) was chosen as the reaction competꢀ
ing with step (2).
R1 + ND
R1N(O•)Ar
(3)
where t is time (min).
Significant stability of the A regiomers made it posꢀ
sible to measure the rates of their formation
B
Ar = 2,3,5,6ꢀМе4C6H
When benzene solutions containing ND and 3
([3]0 = 10–2 mol L–1) were irradiated at λ = 366 nm, the
ESR spectra exhibit the triplet of doublets of the same
intensity from the nitroxy radicals В, which are characꢀ
terized by the parameters aN = 9.25 G, aP = 14.75 G, and
dAa/dt = 0.49t (r2 = 0.98),
dAb/dt = 0.33t (r2 = 0.96),
dAd/dt = 0.45t (r2 = 0.97),
dAe/dt = 0.43t (r2 = 0.99),
dAf/dt = 0.17t (r2 = 0.99),
g = 2.007. If the benzene solution of ND (5•10–3 mol L–1
)
contains compounds 3 and 1 (2•10–3 mol L–1), the sigꢀ
nals from the A and В radicals are simultaneously
detected in the ESR spectra upon photochemical initiaꢀ
tion (see Fig. 1). The rates of radical accumulation
d[Аа + Аb]/dt and d[B]/dt obtained by the leastꢀsquares
processing of the kinetic curves are
where t is time (min). It follows from these data that four
isomers, viz., Aa, Ab, Ad, and Ae, have approximately the
same accumulation rates, which, in turn, are 2—3 times
higher than the accumulation rates for the Af isomers. We
failed to measure dAс/dt even on the basis of the highꢀ
field spectral region, where the HFS for this isomer is
more pronounced (see Fig. 1). Note that in the reaction
of the R1 radicals with complex 1, unlike the Cꢀcentered
radicals (see below), we observed no SA formed by the
addition of the R1 radicals to the carbon atoms of another
hemisphere of metallofullerene, which are characterized
by the gꢀfactors ∼2.0025 and aP ≈ 63 G.
d[Аа + Аb]/dt = 0.81t (r = 0.99)
and
1/6d[B]/dt = 0.29t (r = 0.98).
Accepting that the addition rate constant is k3 = 108 L mol–1 s–1
(see below) and using the expression
d[Аа + Аb]/d[B] = k2[1]0/k3[ND]0,
we determined the rate constant k2 = 1.3•108 L mol–1 s–1
.
The study of the SA of the R1 radicals with the iridium
complexes (η2ꢀC60)IrH(CO)(CNBut)2(oꢀHCB10H9CCH2ꢀ
PPh2ꢀB,P) and (η2ꢀC60)IrH(CO)(DIOP) (DIOP is (4R,5R)ꢀ
(+)ꢀ4,5ꢀbis(diphenylphosphinomethyl)ꢀ2,2ꢀdimethylꢀ
1,3ꢀdioxolane) showed14 that the use of the light with
λ = 366 nm generated only stable SA due to the addition
of the R1 radicals to the carbon atoms of the cisꢀ1 and
cisꢀ2 bonds. In the case of R1 radical generation by the
The rate constant for the addition of the R1 radicals to
ND was determined for the competition of reaction (3)
with the addition of the R1 radicals to hexꢀ1ꢀene
R1 + CH2=CHBu
R1CH2•CHBu
ArN(•O)(R1CH2)CH(Bu),
(4)
C

312
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
Gasanov et al.
Table 1. gꢀFactors of regiomers of the •But and •CCl3 radicals
with complex 1
where k4 = 1.8•105 L mol–1 –1 15 The nitroxy radical C is
s .
characterized by the parameters aN = 13.5 G, aβꢀH = 7.5 G,
and g = 2.0059.
Thus, the rate constant for the addition of the R1
radicals to the carbon atoms of the cisꢀ2 bonds in complex
1 is ∼108 L mol–1 s–1. Based on the data on the rate of Ai
radical accumulation, we can assume that the rate constants for
the addition of the R1 radicals to the C(5) and C(8)
Regiomer gꢀFactor
Intensity*
Assignment**
1••But
2.0091
2.0085
2.0073
2.0068
2.0062
2.0050
2.0034
2.0028
0.25
0.15
0.2
0.2
0.1
0.2
0.4
1.0
cisꢀ1
cisꢀ1
cisꢀ2
cisꢀ2
cisꢀ2
cisꢀ2
cisꢀ3, e
transꢀn
atoms of the cisꢀ1 bond are about (3—5)•107 L mol–1 s–1
.
Since C60 contains 60 equivalent carbon atoms to which
the R1 radicals are added and the rate constant of addition
of these radicals to С60 is 6•108 L mol–1 s–1 16
we can
,
1••CCl3
2.0114
2.0107
2.0101
2.0087
2.0080
2.0071
2.0061
2.0053
2.0043
2.0037
0.05
0.45
0.35
0.4
0.3
0.2
0.2
0.4
0.65
1.0
cisꢀ1
cisꢀ1
cisꢀ1
cisꢀ1
cisꢀ2
cisꢀ2
cisꢀ2
cisꢀ2
cisꢀ3, e
transꢀn
conclude that the spatial effects caused by the substituꢀ
ents at the osmium atom exert no substantial effect on
the rate of R1 radical addition to the carbon atoms of the
cisꢀn bond. On the contrary, the polar effect related, most
likely, to the partial electron density transfer to the carꢀ
bon atoms of the cisꢀ1 and cisꢀ2 bonds favors an increase
in the reactivity of the carbon atoms of these bonds toward
the R1 radicals compared to those in C60.
Thus, the metal fragment in complex 1 has a dual
character. This fragment in 1 manifests the donor properꢀ
ties by the electron density transfer to the nearby bonds,
while in the SA this fragment exhibits the acceptor propꢀ
erties, which follows from a decrease in the HFC conꢀ
stants with the phosphorus nucleus and the positive shift
of the gꢀfactor (see above).
Spinꢀadducts of the Cꢀcentered radicals with comꢀ
pound 1. When the Ме• and Et• radicals are added to
compound 1, the ESR spectrum exhibits a superposition
of two singlets with g = 2.0022 and 2.0028. The gꢀfactors
were calculated from the simulated spectra. After the irraꢀ
diation of the reaction mixture ceased, the dimerization
rate of these radicals can be well described by the secꢀ
ondꢀorder equation and the dimerization rate constant
kd ≈ 102 L mol–1 s–1 at room temperature. These signals
cannot be assigned to particular regiomers, because the
gꢀfactors coincide with those for the SA of the Et• radiꢀ
cals with С60 and agree with the data on the stability of
radicals formed by the addition of the Ме• and Et• radiꢀ
cals to metallofullerenes. However, the results obtained
for the stability of the SA of the phosphoryl and methyl
radicals with compound 1 additionally confirm the deꢀ
pendence of the properties of the SA of this compound on
the polar and steric properties of the added radicals.
The signals from eight to ten regiomers are observed in
the ESR spectra when the •But and •CCl3 radicals are
added to complex 1 (Table 1, Figs 2 and 3).
* Relative intensities of the signals calculated from the simulated
spectra.
** See the text.
fragment on the magnetic resonance parameters of the
SA. By analogy to the data for the R1 signals, the six and
eight signals with the maximum values of the gꢀfactors
and maximum intensities were assigned to six regiomers
formed due to the addition of the •But and Cl3C• radicals
to the carbon atoms of the cisꢀ1 and cisꢀ2 bonds in the
fullerene cage of complex 1. The rest signals were asꢀ
signed to the regiomers formed due to the addition of the
•But and Cl3C• radicals to the carbon atoms of other
bonds in complex 1. After the irradiation was switched
a
I•103 (arb. units)
b
200
100
0
The results obtained suggest the electronic influence
of the metal fragment on the magnetic resonance paramꢀ
eters of the SA formed, because only one singlet (g =
2.0023) is observed in the ESR spectra upon the addition
of the •But radicals to methanoꢀ3 or pyrrolidinofullerenes.4
It is most likely that the closer to the metal atom is the
radical center, the greater is the influence of the metal
–100
3460
3470
3480
3490
H/G
Fig. 2. Experimental (a) and simulated (b) ESR spectra of the
radicals generated by the irradiation with λ = 366 nm of benzene
solutions of compounds 1, Re2(CO)10, and BrCMe3.

Spinꢀadducts of (η2ꢀC60)Os(CO)(PPh3)2(CNBut)
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
313
ND (reaction (7)) was chosen as reactions competing
with the addition of the •But and Cl3C• radicals to 1
(reactions (5) and (6))
a
•But + 1
[But(η2ꢀ•С60)Os(CO)(PPh3)2(CNBut)],
(5)
D
I•103 (arb. units)
•CCl3 + 1
30
b
[CCl3(η2ꢀ•С60)Os(CO)(PPh3)2(CNBut)],
E
(6)
(7)
20
10
•But + ND
ButN(O•)Ar,
0
F
–10
–20
–30
whereas for the Cl3C• radicals the addition to PhCH=N(O)But
(7) was chosen
•CCl3+ 7
N(O•)(CBut)CH(CCl3)Ph.
3460
3470
3480
3490
H/G
G
Fig. 3. Experimental (a) and simulated (b) ESR spectra of the
radicals generated by the irradiation with λ = 366 nm of benzene
solutions of compounds 1, Re2(CO)10, and CCl4.
Upon the irradiation with λ = 366 nm of reaction mixtures
c
ontaining BrCMe3, Re2(CO)10, ND ([ND]0 = 5•10–4 mol L–1),
off, the intensity of the signals with g = 2.0028 for the SA
of the •But radicals and with g = 2.0037 for the SA of the
Cl3C• radicals with complex 1 decreases with the rate
corresponding to the secondꢀorder equation with the
dimerization rate constant kd ≈ 102 L mol–1 s–1 at room
temperature. The intensities of other signals from the SA
of the •But radicals with complex 1 decrease several times
more slowly. The intensities of the signals from the SA of
the Cl3C• radicals with 1, except for the singlet with g =
2.0037, remain unchanged for several hours. These results
additionally show that the stability of the SA of the radiꢀ
cals with complex 1 depends considerably on the mutual
arrangement of the metal fragment and the added radical.
The closer is the radical center to the metal fragment, the
more hindered are the bimolecular reactions of the SA.
Thus, the difference in particular regiomers formed by
the addition of the carbonꢀcentered radicals to substiꢀ
tuted fullerenes was shown for complex 1. The results
obtained make it possible to estimate the reactivity of
particular carbon atoms in complex 1 toward the phosꢀ
phorusꢀcontaining and Cꢀcentered radicals. For this purꢀ
pose, we determined the rate constants for the addition of
the •But and Cl3C• radicals to the carbon atoms of parꢀ
ticular bonds in compound 1, which allows one to comꢀ
pare its reactivity with fullerenes С60 and С70, thus revealꢀ
ing the electronic properties of the metal fragment. The
rate constants for the addition of the •But and Cl3C•
radicals to С60 have been determined earlier.3,16,17
a
I•103 (arb. units)
b
200
100
0
–100
–200
3460
3470
3480
3490
H/G
Fig. 4. Experimental (a) and simulated (b) ESR spectra of the
radicals generated by the irradiation with λ = 366 nm of benzene
solutions of a mixture of complex 1 (2•10–3 mol L–1),
Re2(CO)10, BrCMe3, and ND (5•10–4 mol L–1).
Rate constants for the addition of the •But and Cl3C•
radicals to complex 1. The addition of the •But radicals to

314
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
Gasanov et al.
and 1 ([1]0 = 2•10–3 mol L–1), the ESR spectra contain
the signal from the F radical: the triplet with aN = 13.2 G
(g = 2.0062),18 which is characteristic of the SA of the
•But radicals with ND (Fig. 4). The rates of formation of
the SA of the •But radicals with 1, which are described by
signals with the gꢀfactors 2.0034 and 2.0028, and with
ND until they achieve the stationary concentration are
dD/dt = 19.9t (r2 = 0.94) and 1/3dF/dt = 30.5t (r2 = 0.99),
respectively. Accepting for the addition rate constant of
the•But radicals to ND that k7 = 2•108 L mol–1 s–1 (see
Ref. 19) and using the expression
a
I•103 (arb. units)
*
b
40
0
d[F]/d[D] = k8[ND]0/k6[1],
which follows from reactions (5) and (7), we determined
the rate constant for the addition of the •But radicals to
the carbon atoms of the cisꢀ3, eꢀ, and transꢀn bonds in 1,
which is 107 L mol–1 s–1. The value determined for the
rate constant coincides with the corresponding rate conꢀ
stant for the addition of these radicals to С60, which was
determined by the same procedure in Ref. 16. Based on
the comparison of the rates of formation of particular SA
–40
3460
3470
3480
H/G
Fig. 5. Experimental (a) and simulated (b) ESR spectra of the
radicals generated by the irradiation with λ = 366 nm of benzene
•
for the addition of the But radicals to 1 (see Fig. 2) and
solutions of a mixture of compounds 1 (10–3 mol L–1), Re2(CO)10
,
on the simulated spectra (see Table 1), we can assume
that the rate constants of •But radical addition to the
carbon atoms of the cisꢀ1 and cisꢀ2 bonds in 1 would be
equal to ∼106 L mol–1 s–1, i.e., they are somewhat higher
than the rate constants of addition of these radicals to the
carbon atoms of the cisꢀ3, eꢀ, and transꢀn bonds of comꢀ
plex 1 and to the carbon atoms of С60, if taking into
account the number of C atoms.
CCl4, and 7 (8.65•10–2 mol L–1). The signal for which the rate
of SA formation of the •CCl3 radicals with 1 is marked with
asterisk.
and cisꢀ2 bonds in 1 are ∼107 L mol–1 s–1. If taking into
account the rate constant for the addition of the •CCl3
radicals to С60 (108 L mol–1 s–1 20 and the number of carꢀ
)
For the irradiation with λ = 366 nm of benzene
solutions containing complex 1 ([1]0 = 10–3 mol L–1), 7
([7]0 = 8.65•10–2 mol L–1), Re2(CO)10, and CCl4, their
IR spectra simultaneously contain signals from the SA of
the G and E radicals (Fig. 5). After measuring the rates
of formation of the SA of the •CCl3 radicals with 7
(d[G]/dt = 2t) and with complex 1 (d[E]/dt = 12t) (the
signal is marked with asterisk), taking into account the
line width (∆Н ≈ 0.8 and 1.2 G for the SA of the •СCl3
radicals with 7 and 1, respectively) and multiplicity of the
SA of the G radicals (six lines of equal intensity), acceptꢀ
ing for the rate constant of •CCl3 radical addition to 7
that k8 = 7•104 L mol–1 s–1 (see Ref. 19), and using the
expression
bon atoms in С60, it becomes clear why the addition of the
•CCl3 radicals to complex 1, unlike other substituted deꢀ
rivatives of fullerene С60, results in the appearance of
signals from several regiomers: the rate constants of •CCl3
radical addition to the carbon atoms of the cisꢀ1 and cisꢀ2
bonds in 1 approximately an order of magnitude higher
than the rate constants of addition to the carbon atoms of
other bonds.
Thus, the results obtained show that the rate constants
for the addition of the electrophilic •CCl3 radicals to
complex 1 is approximately an order of magnitude higher
than the corresponding rate constant for the addition of
the •But radicals. This additionally indicates the elecꢀ
tronic influence of the metal fragment on the carbon atꢀ
oms of the bonds arranged closely to the metal atom. It
also follows from the data obtained that the complex conꢀ
sidered can be used as a spin trap for the identification of
bulky carbonꢀcentered radicals.
dE/dG = k7[1]/k8[7]0,
we determined the rate constant for the addition of the
•CCl3 radicals to the carbon atoms of the cisꢀ1 bond in
complex 1, which is 2•107 L mol–1 s–1. It follows from
the simulated spectra that the intensities of the signals
attributed to particular regiomers cisꢀ1 and cisꢀ2 differ
insignificantly, except for the singlet with g = 2.0114.
Therefore, it can be assumed that the rate constants for
•CCl3 radical addition to the carbon atoms of the cisꢀ1
This work was financially supported by the Federal
Agency on Science and Innovations at the Ministry of
Education and Sciences of the Russian Federation (Contract
Nos 02.513.11.3207 and 02.513.11.3209) and the Presidium
of the Russian Academy of Sciences (Program No. 8).

Spinꢀadducts of (η2ꢀC60)Os(CO)(PPh3)2(CNBut)
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 2, February, 2008
315
10. R. C. Haddon, J. Am. Chem. Soc., 1997, 119, 1797.
11. G. N. Schranzer, Inorg. Synth., 1968, 11, 61.
References
12. Th. A. J. W. Wajir, A. Mackor, Th. J. de Boer, J. D. W.
van Voorst, Tetrahedron, 1967, 23, 4021.
13. B. L. Tumanskii, V. V. Bashilov, N. N. Bubnov, V. P.
Solodovnikov, V. I. Sokolov, Izv. Akad. Nauk, Ser. Khim.,
1992, 1938 [Bull. Russ. Acad. Sci., Div. Chem. Sci., 1992, 41,
1521 (Engl. Transl.)].
14. R. G. Gasanov, M. V. Tsikalova, E. V. Martynova, A. L.
Chistyakov, I. V. Stankevich, Yu. N. Novikov. Izv. Akad.
Nauk, Ser. Khim., 2007, 1302 [Russ. Chem. Bull., Int. Ed.,
2007, 56, 1351].
15. D. Griller, P. R. Marriott, D. C. Nonhebel, M. J. Perkin,
P. C. Wong, J. Am. Chem. Soc., 1981, 103, 7761.
16. R. G. Gasanov, V. V. Bashilov, O. G. Kalina, B. L. Tumanskii,
Izv. Akad. Nauk, Ser. Khim., 2000, 1652 [Russ. Chem. Bull.,
Int. Ed., 2000, 49, 1642].
17. R. G. Gasanov, O. G. Kalina, V. V. Bashilov, B. L.
Tumanskii, Izv. Akad. Nauk, Ser. Khim., 1999, 2360 [Russ.
Chem. Bull., 1999, 48, 2344 (Engl. Transl.)].
18. S. Terabe, K. Kuruma, R. Konaka, J. Chem. Soc., Perkin
Trans. 2, 1973, 1252.
19. R. G. Gasanov, R. Kh. Freidlina, Usp. Khim., 1987, 56, 447
[Russ. Chem. Rev., 1987, 56 (Engl. Transl.)].
20. N. M. Dimitrijevic, P. V. Kamat, R. W. Fessnden, J. Phys.
Chem., 1993, 97, 615.
1. L. N. Sidorov, M. A. Yurovskaya, A. Ya. Borshchevskii, I. V.
Trushkov, I. N. Ioffe, Fullereny [Fullerenes], Ekzamen,
Moscow, 2005, 687 (in Russian).
2. I. S. Neretin, Yu. L. Slovokhotov, Usp. Khim., 2004, 73, 492
[Russ. Chem. Rev., 2004, 73 (Engl. Transl.)].
3. R. G. Gasanov, V. V. Bashilov, B. L. Tumanskii, V. I.
Sokolov, I. A. Nuretdinov, V. V. Zverev, V. P. Gubskaya,
L. Sh. Berezhnaya, Izv. Akad. Nauk, Ser. Khim., 2003, 364
[Russ. Chem. Bull., Int. Ed., 2006, 52, 380].
4. R. G. Gasanov, B. L. Tumanskii, M. V. Tsikalova, I. A.
Nuretdinov, V. P. Gubskaya, V. V. Zverev, G. M. Fazleeva,
Izv. Akad. Nauk, Ser. Khim., 2003, 2531 [Russ. Chem. Bull.,
Int. Ed., 2003, 52, 2675].
5. B. L. Tumanskii, R. G. Gasanov, M. V. Tsikalova, A. V.
Usatov, E. V. Martynova, Yu. N. Novikov, Izv. Akad. Nauk,
Ser. Khim., 2004, 1968 [Russ. Chem. Bull., Int. Ed., 2004, 53,
2051].
6. B. L. Tumanskii, V. V. Bashilov, S. P. Solodovnikov, N. N.
Bubnov, V. I. Sokolov, Izv. Akad. Nauk, Ser. Khim., 1994,
938 [Russ. Chem. Bull., 1994, 43, 884 (Engl. Transl.)].
7. B. L. Tumanskii, V. V. Bashilov, O. G. Kalina, V. I. Sokolov,
J. Organomet. Chem., 2000, 599, 28.
8. A. V. Usatov, E. V. Martynova, I. S. Neretin, Yu. L.
Slovokhotov, A. S. Peregudov, Yu. N. Novikov, Eur. J. Inorg.
Chem., 2003, 2041.
9. R. G. Gasanov, E. G. Sadykhov, R. Kh. Freidlina, Izv.
Akad. Nauk SSSR, Ser. Khim., 1986, 1031 [Bull. Acad. Sci.
USSR, Div. Chem. Sci., 1986, 35, 936 (Engl. Transl.)].
Received May 22, 2007;
in revised form July 5, 2007