Paramagnetic intermediates in the photolysis of 7-silanorbornadiene studied by means of spin chemistry method

Article abstract of DOI:10.1016/S0022-328X(98)00605-6

The influence of an external magnetic field on the yield of photo-decomposition products of 7,7′-dimethyl-7-silanorbornadiene derivative has been detected in laser pulse photolysis experiments. The observations of magnetic field effects alterations in the presence of scavengers, O₂ and PPh₃, in combination with ¹H-CIDNP data form the basis for the identification of the structure of paramagnetic intermediates involved in the process. It has been shown that magnetic field effects originate in biradical intermediates. These species result from both endocyclic Si-C bond cleavage in the initial compound and the reaction of dimethylsilylene (in a singlet or a triplet excited state) with starting 7-silanorbornadiene. To explain the influence of O₂ upon the magnetic field effects, the reversible formation of oxygen complex with biradical species has been suggested.

Full text of DOI:10.1016/S0022-328X(98)00605-6

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of Organometallic Chemistry 566 (1998) 73–83
Paramagnetic intermediates in the photolysis of 7-silanorbornadiene
studied by means of spin chemistry method
M.B. Taraban a,*, O.S. Volkova a, A.I. Kruppa a, V.F. Plyusnin a, V.P. Grivin a, Yu.V. Ivanov a,
T.V. Leshina a, M.P. Egorov b, O.M. Nefedov b
a Institute of Chemical Kinetics and Combustion, Siberian Branch of the Russian Academy of Sciences, Institutskaya street 3, No6osibirsk-90,
630090, Russian Federation
b N.D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Moscow, 117913, Russian Federation
Received 9 February 1998
Abstract
The influence of an external magnetic field on the yield of photo-decomposition products of 7,7%-dimethyl-7-silanorbornadiene
derivative has been detected in laser pulse photolysis experiments. The observations of magnetic field effects alterations in the
1
presence of scavengers, O2 and PPh3, in combination with H-CIDNP data form the basis for the identification of the structure
of paramagnetic intermediates involved in the process. It has been shown that magnetic field effects originate in biradical
intermediates. These species result from both endocyclic SiꢀC bond cleavage in the initial compound and the reaction of
dimethylsilylene (in a singlet or a triplet excited state) with starting 7-silanorbornadiene. To explain the influence of O2 upon the
magnetic field effects, the reversible formation of oxygen complex with biradical species has been suggested. © 1998 Elsevier
Science S.A. All rights reserved.
Keywords: Magnetic field effect; CIDNP; Biradicals; Oxygen complexes; Dimethylsilylene
cles, as well as the mechanisms of their insertion and
cycloaddition reactions. The results of quantum–chem-
1. Introduction
ical calculations indicate that carbene analogs are
ground state singlets [1,2,13,14], but the experimental
data on the multiplicity of the reactive states of these
species are still scarce [7–12]. There is a lack of refer-
ence data on the reactivity of triplet excited silylenes
and germylenes.
Heavy carbene analogs, silylenes, germylenes and
stannylenes, are the important reactive intermediates in
a number of chemical reactions [1–4]. In the last
decade, these short-lived species have become a subject
of growing interest. The derivatives of 7-
sila(germa)norbornadienes play a significant role as
convenient precursors of carbene analogs generated
upon thermolysis or photolysis [1,2,4,5]. However, only
a few attempts were made to apply the physico-chemi-
cal methods, including laser pulse photolysis [6] and
chemically induced dynamic nuclear polarization
(CIDNP) techniques [7–12] to study the mechanisms of
silylene and germylene generation from these heterocy-
The following scheme of photo-decomposition of
7,7%-dimethyl-7-silanorbornadiene (I) has been proposed
1
on the basis of both H-CIDNP and laser pulse photol-
ysis studies [6] (Scheme 1).
The involvement of paramagnetic species (singlet and
triplet biradicals of I and triplet dimethylsilylene) in this
photoreaction is still a matter of dispute [2,15]. An
unambiguous conclusion on the participation of the
radical species in this process might be drawn based on
the findings of a magnetic field effect (MFE) study. In
* Corresponding author. Tel.: +7 3832 331405; fax: +7 3832
342350; e-mail: taraban@ns.kinetics.nsc.ru
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00605-6

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
74
M.B. Taraban et al. / Journal of Organometallic Chemistry 566 (1998) 73–83
principle, the analysis of the character of MFE-depen-
dence and the sign of the magnetic field effect provide
the means for independent definitions of both the mul-
tiplicity of the reactive state and the structure of para-
magnetic precursors of the reaction products. The
present paper discusses the results from the application
of the MFE method to the investigations of the photol-
ysis of 7-silanorbornadiene, I.
The external magnetic field was applied by means of
a specially designed permanent Sm–Co magnet, with
variable distances between the poles. The geomagnetic
field was assumed to be close to a zero level. The MFE
was defined as the ratio of TPN yield under the action
of the external magnetic field (DDH) to that under the
geomagnetic one (DD0):
DD0−DD
MFE=
H×100%
DD0
2. Experimental
This expression shows that MFE\0, if the yield of
TPN decreases in the external magnetic field (DDHB
DD0); accordingly, MFEB0 reflects the growth of TPN
yield in the external magnetic field (DDH\DD0).
Data points of the MFE dependencies show the
results of standard statistical processing of 7–10 kinetic
traces. The error bars represent the S.D.
7,7%-Dimethyl-1,4,5,6-tetraphenyl-2,3-benzo-7-
silanorbornadiene I was synthesized according to an
earlier procedure [16]. PPh3 (99%, Aldrich) was used
without additional purification The solvents (n-hexane,
cyclo-C6D12, C6D6 and CD2Cl2) were dried and purified
by conventional techniques [17]. All experiments were
carried out at ambient temperature.
NMR spectra were taken using a JEOL JNM FX90Q
spectrometer (1H operating frequency 90 MHz)
equipped with a photochemical device for sample irra-
diation directly in the probe of the spectrometer. The
light source employed was a high pressure Hg lamp
(DRSh-1000 of 1 kW power) with a thermal filter. The
concentration of I was varied in the range from 10−3 to
10−1 M. Samples with equilibrium concentrations of
O2 were used; PPh3 was taken in 2–5-fold excess.
Attempts to study CIDNP in low magnetic fields
were carried out using a Bruker AM 300 NMR spec-
trometer (1H operating frequency 300 MHz) equipped
with a specially designed continuous flow system. The
solution of I (5×10−3 M in C6D6) has been irradiated
by a sequence of laser pulses (excimer laser, u=308
nm, pulse frequency is equal to 20 Hz, pulse duration
was 20 ns, pulse energy amounts to 40 mJ) in a separate
magnet with a variable magnetic field strength ranging
from 100 to 2100 G. After irradiation, the solution was
transferred to the probe of the NMR spectrometer by
means of the flow system. The shortest time of transfer
was ca. 1 s. Apparently, failure to observe CIDNP
effects within this magnetic field range was related to
very small relaxation times of the methyl protons of I.
Independent measurements have shown that this time is
equal to 0.8 s, which is even smaller than the character-
istic time of sample transfer by means of a flow system.
The set-up for laser flash photolysis experiments in-
cludes excimer XeCl laser (u=308 nm) with a pulse
energy ca. 10 mJ, pulse duration 15 ns, delay time 50
ns. Angle between the excitation and probing light
beams is ca. 2°. A detailed description of the set-up can
be found elsewhere [18].
Experiments were performed in 1 cm-optical quartz
cells. The concentration of I (typically ca. 1×10−3 M)
was controlled spectrophotometrically (m308=2×103
M
−1 cm−1) by a Specord UV-VIS spectrophotometer.
In all experiments, the wavelength of probing light was
334 nm. With the exceptions mentioned below, samples
were deaerated by 15 min of Ar bubbling. Photo-de-
composition of I was monitored by the yield of a stable
end product of the reaction, 1,2,3,4-tetraphenylnaph-
thalene (TPN). Below, the change of its optical density
is designated as DD. All experimental results were
normalized with respect to laser pulse intensity.
3. Results and discussion
In laser pulse photolysis experiments, the yield of
TPN formed upon photo-decomposition of I was found
to depend on the external magnetic field (MF) strength
and the presence of the scavengers, O2 and PPh3, in the
reaction mixture. Fig. 1 presents the typical kinetic
traces of TPN absorbance observed in MF of different
strengths.
Scheme 1.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
M.B. Taraban et al. / Journal of Organometallic Chemistry 566 (1998) 73–83
75
for the protons of I, II and TPN (Fig. 4). Chemical
shifts of the initial compounds and the reaction prod-
1
ucts as well as the signs of H-CIDNP effects are given
in Table 2.
According to modern understanding, the influence of
the external MF on the reaction rate or the yield of the
reaction products is a reliable evidence of the prevalent
participation of paramagnetic precursors. Thus, MFEs
detected in the formation of TPN during the photolysis
of deaerated solutions of I confirm the earlier proposed
reaction mechanism (Scheme 1) [6,7,21].
Indeed, in the framework of the existing theories [22]
it is very difficult to explain the extremes of MFE
dependencies located at several hundred Gauss (see
Figs. 2 and 3) without the biradical species taken into
account. In a general case, the position of the extremes
correlates with the value of mixing parameter which
affects the process of S–T conversion, and virtually
determines the MFE magnitude. Typically, the mixing
efficiency is defined by effective hyperfine interaction
(HFI) in the radical pair (Aeff=ꢀꢁi A2i ), by DgH0 term
(Dg is the difference of g-factors of the radicals in a
radical pair, H0 is the external MF strength), or by the
energy of the effective electron exchange interaction (J).
Fig. 1. The time profile of the absorbance of TPN measured during
the photolysis I in the presence of O2. (1) in geomagnetic field; (2) at
540 G.
Figs. 2 and 3 show the dependencies of the MFE on
the external MF strength detected under various exper-
imental conditions. Thus, Fig. 2 shows MFE dependen-
cies observed in the photolysis of I in aerated and
deaerated solutions. MFE dependencies presented in
Fig. 3 were obtained in the presence of a Lewis base,
PPh3. Lewis bases (e.g. ethers, amines, phosphines,
sulfides) are known to form complexes with silylenes;
such complexes were previously studied by experimen-
tal [19] and theoretical [20] methods. Therefore, in the
system under study, PPh3 plays a role of the trapping
agent for the dimethylsilylene, which is another product
of the photolysis of I. The relative yields of TPN
formed during the photolysis of I in the presence of
various scavengers are listed in Table 1.
The photolysis of I was also studied by 1H-NMR and
1H-CIDNP techniques in cyclo-C6D12, C6D6 and
1
CD2Cl2 solutions. H-NMR spectral data clearly testify
to the formation of 7,7%,8,8%-tetramethyl-1,4,5,6-te-
traphenyl-2,3-benzo-7,8-disilabicyclooctadiene (II) re-
sulting from the photolysis of I. The compound II is a
product of Me2Si insertion into the endocyclic SiꢀC
bond of starting I (Scheme 2):
The maximum yield of II was detected in CD2Cl2
under the irradiation with the full light of high pressure
Hg lamp. In this case, the CIDNP effects were detected
Fig. 2. Magnetic field effect on TPN yield observed in the photolysis
of I in aerated and deaerated solutions: (1) in aerated solution; (2) in
the absence of O2.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
76
M.B. Taraban et al. / Journal of Organometallic Chemistry 566 (1998) 73–83
Scheme 2.
istic for the electron exchange interaction values of J,
which are quite typical for biradicals (250–500 G) [22].
The manifestation of S–T0 mechanism of CIDNP
formation in the reaction under study [7] by no means
is at variance with the assumption on the formation of
biradical intermediates. It is known that in high mag-
netic fields polarizations in rigid biradicals are gener-
ated only via S–T0 mechanism [22]. Unfortunately, due
to short longitudinal relaxation times of the methyl
protons of I (see Section 2) it was not possible to
measure the dependence of CIDNP effects versus exter-
nal MF strength, which is commonly accepted criterion
for the detection of biradical species.
However, the observed influence of the scavengers on
the yield of TPN points to a more complex process of
TPN formation than that proposed in Scheme 1. For
example, the dependence of TPN yield on PPh3 concen-
tration (Fig. 5, Table 1) suggests that TPN could result
not only from the decomposition of the biradical III
shown in Scheme 1, but also after the reactions of
dimethylsilylene in the bulk, since it is Me2Si that forms
a donor–acceptor complex with PPh3 [19,20].
Fig. 3. Magnetic field effect on TPN yield observed in the photolysis
of I in aerated and deaerated solutions in the presence of 2×10−3 M
of Ph3P: (1) in deaerated solution; (2) in the presence of O2.
As it has been already mentioned, laser pulse photol-
ysis experiments [6] as well as 1H-NMR spectral data of
the present paper have unraveled the reaction between
Me2Si and the initial I leading to disilabicyclooctadiene
derivative II (Scheme 2). The dependence of TPN yield
on Ph3P concentration (Fig. 5) allows one to suggest
that TPN also results from the interaction of I with
Me2Si. The higher the concentration of PPh3 is, the
lower the fraction of ‘free’ Me2Si that will attack initial
I, leading ultimately to the formation of TPN. CIDNP
effects for methyl and phenyl protons of II (Fig. 4 and
Table 2) indicate that the reaction of I with Me2Si also
includes the stages with paramagnetic precursors. Simi-
lar to Scheme 1, the process could be written as below
(Scheme 3).
In the reaction under study, for any feasible radical pair
comprised of silicon- and carbon-centered free radicals,
the values of HFI and DgH0 terms are not greater than
several Gauss. Thus, the field dependence of the MFE,
if only formed via HFI- and Dg-mechanisms, should
not demonstrate any extremes within the external mag-
netic field range under study.
On the other hand, CIDNP effects detected for the
methyl protons of the starting I suggest the formation
of biradical intermediate which is a paramagnetic pre-
cursor of the polarized species. Note also that the
positions of the extremes of the experimental MFE
dependencies (Figs. 2 and 3) are in the range character-
Table 1
Scheme 3 shows the case of the reaction between I
and the excited triplet state of Me2Si. Indeed, the rate
constant of the reaction between Me2Si and I measured
earlier by laser pulse photolysis method [6] approaches
The relative yield of TPN in the photolysis of 7-silanorbornadiene I
in the presence of various scavengersa
Scavenger
TPN yield (%)
85
the diffusion control limit and equals 5.7×109 M−1 s−1
.
O2 (equilibrium concentration, ca. 2×10−3
The value is quite characteristic for the reactions of
1
M)
triplet carbenes [23]. H-CIDNP technique has allowed
O2+2×10−3 M PPh3
O2+0.1 M CHBr3
75
57
the detection of several reactions of singlet Me2E (E=
Si, Ge) species [7–10]; however, two cases of the excited
triplet dimethylsilylene reactions with thiacycloheptyne
[11] and carbon tetrachloride [12] were also reported.
a The yield of TPN in the photolysis of I in the absence of scavengers
was set to 100%.