Spin chemistry of organometallic compounds 5. Interaction of N-bromohexamethyl disilazane with substituted silyl hydrides

Article abstract of DOI:10.1016/j.jorganchem.2008.09.068

Reaction of N-bromohexamethyl disilazane (Me₃Si)₂NBr with substituted triorganyl silanes R¹R²R³SiH results in asymmetric disilazanes Me₃SiNHSiR¹R²R³ and bromination product, bromotrimethyl silane Me₃SiBr. The reaction has demonstrated an unusual dependence on specific solvation. In benzene, bromination occurs immediately after mixing of the reagents, while in cyclohexane, the reaction products are formed only under UV-irradiation. Application of photoinduced CIDNP method has shown that the mechanism of bromination of triorganyl silanes is comprised of a series of consecutive radical stages involving N-centered disilazanyl (Me₃Si)₂N^{{radical dot}} and Si-centered silyl R¹R²R³Si^{{radical dot}} radicals.

Full text of DOI:10.1016/j.jorganchem.2008.09.068

Journal of Organometallic Chemistry 693 (2008) 3815–3820
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Spin chemistry of organometallic compounds 5. Interaction of N-bromohexamethyl
disilazane with substituted silyl hydrides
b
Marc B. Taraban a, , Vladimir I. Rakhlin , Olga S. Volkova a,c, Tatyana A. Podgorbunskaya b,
*
Leonid V. Kuibida a, Rudolf G. Mirskov b, Tatyana V. Leshina a, Larisa V. Sherstyannikova b,
Mikhail G. Voronkov b
a Institute of Chemical Kinetics and Combustion, Institutskaya Street 3, Novosibirsk 630090, Russia
b A.E. Favorsky Irkutsk Institute of Chemistry, Favorsky Street 1, Irkutsk 664033, Russia
c Novosibirsk State University, Pirogova Street 2, Novosibirsk 630090, Russia
a r t i c l e i n f o
a b s t r a c t
Reaction of N-bromohexamethyl disilazane (Me3Si)2NBr with substituted triorganyl silanes R1R2R3SiH
results in asymmetric disilazanes Me3SiNHSiR1R2R3 and bromination product, bromotrimethyl silane
Me3SiBr. The reaction has demonstrated an unusual dependence on specific solvation. In benzene, bro-
mination occurs immediately after mixing of the reagents, while in cyclohexane, the reaction products
are formed only under UV-irradiation. Application of photoinduced CIDNP method has shown that the
mechanism of bromination of triorganyl silanes is comprised of a series of consecutive radical stages
involving N-centered disilazanyl (Me3Si)2NÅ and Si-centered silyl R1R2R3SiÅ radicals.
Article history:
Received 30 May 2008
Received in revised form 16 September
2008
Accepted 30 September 2008
Available online 8 October 2008
Keywords:
Ó 2008 Elsevier B.V. All rights reserved.
N-bromohexamethyl disilazane
Silyl hydrides
Silazanes
Free radicals
CIDNP
1. Introduction
analogy of the reaction between triphenyl silane and bromine or
N-bromosuccinimide [6,7], one might expect that bromotriorganyl
N-bromohexamethyl disilazane (Me3Si)2NBr (1) is well-known
as a very convenient and smooth brominating agent widely used
in the organic and organoelement syntheses. However, apart from
outdated and disperse data on the interactions of 1 with organic
derivatives of Group 14 elements – e.g., the reaction of 1 with N-
sodiumhexamethyl disilazane [1], trimethylstannyl lithium [2],
bivalent germanium species, and bis-(trimethylsilylamides) of tin
and lead [3] – modern literature lacks the detailed study of its
reactivity and reaction mechanisms when attacking the element-
hydrogen bond. In our previous papers, we have successfully ap-
plied 1H CIDNP (chemically induced dynamic nuclear polarization)
method to explore in details the structure of paramagnetic inter-
mediates of the photoinduced bromination of allyltrialkyl silanes,
germanes and stannanes CH2@CHCH2MR3 (M@Si, R = Me, Et,
OMe, OEt, Cl; M = Ge, Sn, R = Et) by N-bromohexamethyl disilazane
1 [4,5].
silanes would also be the major ultimate bromination products of
the reaction between 1 and 2.
In the present paper, 1H CIDNP method was used to study
the detailed mechanism of the reaction of 1 with various silyl
hydrides – dimethylphenyl silane PhMe2SiH (2a), methyldiphenyl
silane Ph2MeSiH (2b), triphenyl silane Ph3SiH (2c), trietoxysilane
(EtO)3SiH (2d), 1,1,3,3-tetramethyl disiloxane Me2HSiOSiHMe2
(2e), 1,1,3,3-tetramethyl disilazane Me2HSiN(H)SiHMe2, and (2f),
trichlorosilane CI3SiH (2g). We have also observed and discussed
the effect of media polarity on the reactivity of 1 in the reaction
with Si–H bond.
2. Results and discussion
Reactions of equimolar quantities of 1 and 2a–g result in high
yields of asymmetric 1,1,1-trimethyl-3,3,3-triorganyl disilazane
3a–g and bromotrimethyl silane 4 (see Table 1 and Scheme 1).
All reactions are strongly exothermic.
Two solvents were used as a reaction media when studying the
interaction of of 1 with 2a–g. We have found that solvent proper-
ties drastically affect the reactivity of 1 as a bromination agent. In
cyclohexane, 1 reacts with 2a–g only under UV-irradiation, while
To extend the synthetic potential of 1 and to get better insight
into the mechanistic details, we have studied the reaction of 1 and
a series of silyl hydrides R1R2R3SiH (2) with Si–H bond. On the
* Corresponding author. Tel.: +7 383 333 1405; fax: +7 383 330 7350.
E-mail address: taraban@kinetics.nsc.ru (M.B. Taraban).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.09.068

3816
M.B. Taraban et al. / Journal of Organometallic Chemistry 693 (2008) 3815–3820
Table 1
Major products and yields of the reaction of N-bromohexamethyl disilazane with silyl
hydrides.
Silyl hydride
Major reaction products
Yield (%)
PhMe2SiH (2a)
Ph2MeSiH (2b)
Ph3SiH (2c)
PhMe2SiN(H)SiMe3 (3a)
Ph2MeSiN(H)SiMe3 (3b)
Ph3SiN(H)SiMe3 (3c)
(EtO)3SiN(H)SiMe3 (3d)
Me2HSiOSiMe2N(H)SiMe3 (3e)
Me2HSiN(H)SiMe2N(H)SiMe3 (3f)
CI3SiN(H)SiMe3 (3g)
46.6
83.0
69.2
51.7
43.9
51.9
59.6
(EtO)3SiH (2d)
Me2HSiOSiHMe2 (2e)
Me2HSiN(H)SiHMe2 (2f)
CI3SiH (2g)
Fig. 1. 90 MHz 1H CIDNP spectra in photoinduced reaction of Ph3SiH (2c) and
(Me3Si)2NBr (1) in cyclo-C6D12 (A) initial spectrum; (B) spectrum under UV-
Scheme 1. Reaction of N-bromohexamethyl silazane (1) with silyl hydrides.
:
irradiation; (C) spectrum after photolysis.
in benzene, the reaction products are formed immediately after
mixing of the reagents.
effects are formed in the RPs originated from the random encoun-
ters of the radicals in the bulk (so-called diffusion or F-pairs). In the
reaction under study, the silazanyl radical (Me3Si)2NÅ and silyl rad-
ical R1R2R3SiÅ are the most sound candidates to form F-pair which
would be a precursor of principal reaction products and a source of
the observed polarization effects. Silyl radical R1R2R3SiÅ could re-
2.1. Photoinduced reaction of N-bromohexamethyl disilazane with
silyl hydrides
In our previous papers [4,5], we have used 1H CIDNP method to
demonstrate that photoinduced bromination of different sub-
strates by 1 is initialized by homolytic dissociation of N–Br bond
of 1, i.e., the photoinduced reaction follows a radical pathway dri-
ven by (Me3Si)2NÅ and ÅBr free radicals. 1H CIDNP method was also
used in the present paper to unravel the structure of paramagnetic
intermediates and to suggest the radical stages of the photolysis of
organosubstituted silyl hydrides 2a–g in the presence of N-bromo-
hexamethyl disilazane 1. To this end, the mixture of 1 and one of
2a–g in cyclohexane was irradiated directly in the probe of the
NMR-spectrometer.
Å
sult from the interaction of silyl hydride 2a–g with Br radical
and/or silazanyl radical (Me3Si)2NÅ with the synchronous formation
of HBr and (Me3Si)2NH. Indeed, hydrogen abstraction from olefin
by (Me3Si)2NÅ is postulated as one of its basic reactions [8,9]. All
radical and nonradical reaction steps are summarized in Scheme 2.
To verify the proposed scheme we have analyzed the observed
CIDNP effects based on the following magnetic resonance parame-
ters of the involved free radicals. The sign of hyperfine interaction
(HFI) constants of methyl protons in (Me3Si)2NÅ radical is positive;
g-factor of the (Me3Si)2NÅ is suggested to be greater than that of its
silicon-centered partner, R1R2R3SiÅ, in radical F-pair (Scheme 2),
since g-factors of R2NÅ radicals typically amount to 2.0040 and in
the case of two adjacent Me3Si-groups this value would be even
greater, whereas g-factors of R1R2R3SiÅ radicals are about 2.0030
[10].
As an example of characteristic 1H CIDNP effects, let us consider
Fig. 1 which shows NMR spectra of the reaction mixture of 1 and 2c
before, under UV-irradiation, and after the photolysis. As seen from
Fig. 1, we observe CIDNP effects of the protons of initial disilazane
1 (d 0.25 ppm, emission, E) and the reaction product (d 0.1 ppm,
absorption, A). Similar CIDNP effects have been detected in the
photoinduced reactions of 1 with other silyl hydrides 2a–g. There-
fore, the conclusion was made that photoinduced bromination of
organosubstituted silyl hydrides 2a–g by 1 follow similar radical
reaction pathway.
Observation of CIDNP effects testify to the fact that polarized
reaction products are generated through the elementary step
involving the formation of the radical pair (RP). As we have shown
in our previous papers [4,5], the first step of the photodecomposi-
tion of 1 is the cleavage of N–Br bond of (Me3Si)2NBr resulting in
(Me3Si)2NÅ and ÅBr radicals. However, high spin-orbit coupling con-
stant of Br radical would seriously affect spin-correlation in the ini-
tial RP, effectively mixing S- and T-states of initial radical pair
composed of (Me3Si)2NÅ and ÅBr radicals and precluding the forma-
tion of CIDNP effects. Therefore, similar to the inference made in
our previous papers [4,5], we have concluded that observed CIDNP
The radicals forming F-pair might recombine or escape into the
bulk. In that case, the products of in-cage recombination and those
of the reactions in the bulk would demonstrate CIDNP effects of the
opposite signs. According to Scheme 2, the in-cage recombination
of F-pair results in trisilazane 5. At the same time, it is known that
the addition of HBr promotes the dissociation of Si–N bonds [11].
Therefore, the alternative cleavage of two different Si–N bonds in
trisilazane 5, (Me3Si)2N-SiR1R2R3, in the presence of HBr (Scheme
2) would result in silylbromides 4 and 6 along with the asymmetric
(3) and the symmetric (7) disilazanes. Under the experimental con-
ditions used for CIDNP observations, major decomposition product
of trisilazane 5 is hexamethyldisilazane 7. This was also confirmed
by GS/MS analysis of the reaction mixture after irradiation. Note
that CIDNP effects were observed only for Me3Si-protons of
hexamethyldisilazane 7 (absorption (A), Fig. 1 and Scheme 2),
while one would otherwise expect to see the polarizations of trisi-

M.B. Taraban et al. / Journal of Organometallic Chemistry 693 (2008) 3815–3820
3817
ν
Scheme 2. Elementary steps of photoinduced bromination of silyl hydrides by N-bromohexamethyl disilazane (1) (polarized Me-groups are underlined; (A) is enhanced
absorption, (E) is emission).
lazane 5 which is promary recombination product of the precursor
F-pair in Scheme 2. This is quite possible if the rate of Si–N bond
cleavage in trisilazane 5 in the presence of HBr is competing with
the rate of spin-lattice relaxation of Me3Si-protons in 5, and the
product resulting from such cleavage – hexamethyldisilazane 7 –
will demonstrate in-cage CIDNP effect. For F-pair the multiplicity
2.2. Smooth bromination of silyl hydrides by N-bromohexamethyl
disilazane
As opposed to the reaction in cyclohexane, the interaction in ben-
zene starts after mixing of reagents (1 and 2a, b, d–g) and completes
in 15–30 min. Higher yields of asymmetric disilazanes 3a, b, d–g
(see Table 1) could be explained by the assumption that the reaction
under study is bimolecular and involves the formation of four-cen-
tered activated complex (Scheme 3) [12]. Decay of the complex re-
sults in HBr and the substituted trisilazane 5, however, as it has
been mentioned earlier, the latter is unstable in the presence of
HBr [11]. Interaction of HBr with asymmetric trisilazane 5 is depen-
dent on the mutual orientation of the reagents and should result in
disilazanes3 and7andsubstitutedbromosilanes4and6 (Scheme3).
Of note, an attempt of the fractionated separation of the reac-
tion products resulted in isolation of asymmetric disilazanes 3 a–
g only. Neither bromotriorganosilanes 6 a–g, nor bromotrimeth-
ylsilane 4 were isolated from the reaction mixture. It is reasonable
to assume that reaction products of two pathways of the decompo-
sition of two four-centered activated complexes formed by 5 and
HBr (mixture of 4 and 3 a–g, and 6 a–g and 7; Scheme 3) are in
equilibrium, and isolation of the most volatile Me3SiBr (4) (b.p.
79 °C) would shift this equilibrium towards asymmetric disila-
zanes 3 a–g.
factor
l
is positive (
l = (+)); the in-cage recombination factor e is
positive (
e
= (+)); and as it has been suggested above, the relation-
ship between the g-factors of the partner radicals obeys the
inequality g((Me3Si)2NÅ) > g(R1R2R3SiÅ), so the difference of radical
g-factors,
stant A of Me3Si-protons in (Me3Si)2NÅ radical is also positive
(A = (+)), therefore, the product of these factors
Dg, is also positive (Dg = (+)); and the sign of HFI con-
l
ꢀ
e
ꢀ
D
g ꢀ A = (+) ꢀ (+) ꢀ (+) ꢀ (+) is positive and corresponds
to the enhanced absorption (A) (see Section 3 for more details).
Thus, experimentally observed positive polarization of Me3Si-pro-
tons of 7 (d 0.1 ppm, Fig. 1) is in agreement with the proposed
Scheme 2.
Radicals escaping the recombination in the F-pair in Scheme 2
diffuse into the bulk and bear negatively polarized Me3Si-protons.
Indeed, as above, the multiplicity factor
g-factors, g, and the sign of HFI constant A of Me3Si-protons are
all positive, whereas the escape recombination factor
= (ꢁ)), therefore, the product of these factors
l, the difference of radical
D
e
e
is negative
g ꢀ A =
(e
l
ꢀ
ꢀ
D
(+) ꢀ (ꢁ) ꢀ (+) ꢀ (+) is negative and corresponds to emission (E)
of Me3Si-protons (see Section 3 for more details). These negatively
polarized (Me3Si)2NÅ radicals are capable to react with initial N-
bromohexamethyl disilazane 1 and abstract the bromine atom.
Therefore, chemical exchange between (Me3Si)2NÅ and 1 (Scheme
2) is a most likely explanation of negative polarization transfer to
the Me3Si-protons of 1 (d 0.25 ppm, Fig. 1). Thus, the above analy-
sis of CIDNP effects fully confirms the sequence of the radical steps
suggested in Scheme 2. It is necessary to emphasize that observed
CIDNP effects favor fully radical mechanism of the photoinduced
interaction of 1 and 2a–g in cyclohexane since both (Me3Si)2NÅ
and ÅBr free radicals are formed only under the photoinitiation. This
conclusion is also supported by the observation that no products
are formed when the reaction mixture in cyclohexane is kept in
the dark.
To unravel the factors which control the solvent influence on
the reaction mechanism of bromination of silyl hydrides 2 a–g
by N-bromohexamethyl disilazane 1 we have explored the capabil-
ity of benzene to form the complex with one of the reagents. How-
ever, this assumption appears to be ungrounded, since the
maximum of the UV-absorption band of 1 (342–345 nm) remains
unchanged in benzene, cyclohexane, or methylene chloride. Simi-
larly, Si–H vibration frequency, m(Si–H) of 2 a–f is also independent
of the solvent variation. But the value of the stretching vibration’s
frequency of Si–H in 2g is sensitive to variations of the dielectric
constant and other relevant characteristics of the solvents
[13,14]. In solvents with different polarity, measurable high-fre-
quency shift of m(Si–H) has been observed, as compared to cyclo-
hexane, e.g., in benzene, this high-frequency shift of m(Si–H) in 2g
is 5 cmꢁ1 [14].

3818
M.B. Taraban et al. / Journal of Organometallic Chemistry 693 (2008) 3815–3820
Scheme 3. Formation and decay of four-centered activated complexes in smooth bromination of silyl hydrides by N-bromohexamethyl disilazane.
To clarify the nature of the interactions underlying the observed
effects we have compared the IR-spectra (in the region from 2000
to 2300 cmꢁ1) of 2g in solvents of different polarity/polarizability,
including in benzene and in cyclohexane. As a quantitative charac-
teristic of solvent properties, we used the Kamlet–Taft parameter
p* which reflects the capability of a solvent to participate in non-
specific interactions [15,16]. Table 2 shows the correlation be-
2258
2256
2254
2252
2250
tween Kamlet–Taft parameter p* and Si–H vibration frequency
(Si–H) of Cl3SiH in various solvents.
m
Fig. 2 shows the linear fit of
m(Si–H) in Cl3SiH as a function of
Kamlet–Taft parameter p* which follows the correlation Eq. (1):
m
ðSi—HÞ ¼ ð2250:8 ꢂ 0:8Þ þ ð9:2 ꢂ 0:8Þ ꢀ pꢃ
ð1Þ
(Si–
This allows to suggest that measurable high-frequency shift of
m
0.0
0.2
0.4
0.6
0.8
H) in the IR-spectrum of Cl3SiH 2g in benzene is stipulated by non-
specific intermolecular interaction between 2g and the solvent.
If the solvent influence on the reaction mechanism in the case of
other silyl hydrides 2a–f is of the same nature, i.e., it is due to the
variations of solvent’s capability to enter nonspecific interactions,
one might expect that the reaction of 2a–f with 1 in a series of sol-
vents would follow different pathways reflecting the changes in
*
π
Fig. 2. Linear fit (R2 = 0.99, SD = 0.48) of the dependence of
m
(Si–H) in Cl3SiH on
*
Kamlet–Taft parameter
p
in a series of solvents (data taken from Table 2).
series of solvents somewhere between cyclo-C6H12 and benzene,
only a weak warming-up of the reaction mixture is observed, and
the formation of measurable amounts of the reaction products re-
quires the UV-irradiation.
the Kamlet–Taft parameter
the reaction of 1 and dimethylphenylsilane PhMe2SiH (2a) in
p
*. To this end, we have compared
These results confirm the assumption that in the present case,
nonspecific interactions of the reagents with a solvent affect the
reaction mechanism. One of the possible explanations could in-
volve solvent-mediated differences in stabilization effects for sug-
gested four-centered activation complexes. As for the possible
cyclohexane, carbon tetrachloride, benzene, and methylene chlo-
*
ride. Indeed, in cyclohexane (
p
= 0), the reaction occurs only un-
*
(p = 0.59) and in methylene
der UV-irradiation, in benzene
*
chloride (
p
= 0.82) the products are formed after the mixing of re-
*
agents. However, in CCl4
(p
= 0.28), which is intermediate in this
Table 2
*
Correlation between Kamlet–Taft parameter
p
and Si–H vibration frequency
m
(Si–H) of Cl3SiH in various solvents.
*
Solvent
Stretching vibration frequency
m
(Si–H) in 2g, cmꢁ1
Kamlet–Taft parameter, p
Cyclohexane
Carbon tetrachloride
Chloroform
Benzene
Methylene chloride
2250
2253
2255
2255
2258
0.00
0.28
0.58
0.59
0.82

M.B. Taraban et al. / Journal of Organometallic Chemistry 693 (2008) 3815–3820
3819
involvement of free radicals, we failed to detect any CIDNP effects
when the reagents are mixed in benzene, and this supports the
conclusion that reaction of 1 and 2a–g in benzene is nonradical.
Still, it is necessary to emphasize that unusual drastic solvent ef-
fects in the brominations using N-bromohexamethyl disilazane 1
require further in-depth exploration.
was carried out in standard Pyrex NMR tubes with the full light
of high-pressure Hg lamp (DRSh-1000, 1 kW). A thermal filter
was used to prevent the heating of the sample. GS/MS analysis of
the irradiated reaction mixture was carried out by means of Agilent
Technologies HP6890N/HP5973N mass-spectrometer using the
capillary columns HP-5MS with step-gradient heating from 50 °C
to 150 °C.
3. Experimental
The analysis of CIDNP effects was carried out using existing
empirical rules [19]. Sign of net CIDNP effect (C) observed in high
Well-established and approved standard approaches were used
to synthesize N-bromohexamethyl disilazane (Me3Si)2NBr (1) [17]
and silyl hydrides 2a–g [18]. Their measured physical constants
(b.p., n2D0 and d (ppm) in the NMR spectra, etc.) were in full agree-
ment with the reference data, and GLC-determined purity grade
was not lower than 99.5%.
For GLC analyses, we employed Tsvet 500 gas-liquid chromato-
graph (heat conductivity detector, He flow, glass column
3 mm ꢀ 4 mm, 10% PMS-1000 on a Chromaton N-AW-HMDS,
0.20–0.25 mm grain). NMR spectra of the initial reagents and the
reaction products were recorded in CDCl3 (‘‘Izotop”) using BRUKER
DPX-400 NMR spectrometer (400 MHz 1H operating frequency)
with TMS or HMDS added as an internal standard. Spectral charac-
teristics of the compounds under study were measured using IR-
spectrometer Specord 75 and UV-spectrometer Specord UV–Vis.
magnetic fields is defined by the product of multiplication of the
following parameters:
C
¼
l
ꢀ
ꢀ
ꢀ
D
g ꢀ A
where
l
reflects the multiplicity of the radical pair precursor, (+) for
triplet and uncorrelated F-pair, and (ꢁ) for singlet precursor,
e
is (+)
for in-cage and (ꢁ) for escape recombination products,
Dg is the
sign of the difference between g-factor of the radical with polarized
nucleus (proton in our case) and g-factor of the partner radical in a
radical pair, A defines the sign of hyperfine interaction constant of
the polarized nucleus (proton, in our case) in the radical under
study. The sign of
C reflects the phase of the NMR signal of nucleus
under study: (+) for enhanced absorption (A) and (ꢁ) for the emis-
sion (E). For example, if one consider certain group of protons in the
product resulting from recombination (
ical pair (F-pair, is (+)), and if this group belonged to a radical with
the g-factor which is smaller than that of the partner radical in the
radical pair ( g is (+)), and if the sign of hyperfine interaction for
e is (+)) of uncorrelated rad-
l
3.1. Bromination reaction and product analysis
D
Equimolar mixture of N-bromohexamethyl disilazane (Me3-
Si)2NBr (1) and silyl hydride R1R2R3SiH (2a–g) after degassing in
accordance with ‘‘freeze-pump-thaw” technique was irradiated
by the full light of UV lamp (DRT-400, 400 W, Hg) during 10 h in
a sealed glass tube. Reaction product was isolated by fractional dis-
tillation and identified by means of NMR spectra recorded using
BRUKER DPX-400 NMR-spectrometer. Following reaction products
were isolated and characterized.
1,1,1,3,3-Pentamethyl-3-phenyl disilazane, PhMe2SiN(H)SiMe3
(3a): b.p. 87 °C/2 mm, n2D0 1.4861; d (1H, CDCl3, TMS): 0.03 (s, 9H,
SiMe3), 0.32 (s, 6H, SiMe2), 7.3–7.35, 7.55–7.6 (m, 5H, SiPh).
1,1,1,3-Tetramethyl-3,3-diphenyl disilazane, Ph2MeSiN(H)-
SiMe3 (3b): b.p. 163 °C/9 mm, n2D0 1.5435; d (1H, CDCl3, TMS):
0.06 (s, 9H, SiMe3), 0.66 (s, 3H, SiMe), 7.36–7.43, 7.58–7.66 (m,
10H, SiPh2).
1,1,1-Trimethyl-3,3,3-triphenyl disilazane, Ph3SiN(H)SiMe3
(3c): b.p. 187 °C/1 mm, nD20 1.5823; d (1H, CDCl3, TMS): ꢁ0.04 (s,
9H, SiMe3), 7.23–7.30, 7.62–7.67 (m, 15H, SiPh3).
1,1,1-Trimethyl-3,3,3-triethoxy disilazane, (EtO)3SiN(H)SiMe3
(3d): b.p. 68 °C/2 mm, n2D0 1.4078; d (1H, CDCl3, TMS): 0.10 (s, 9H,
SiMe3), 1.22 (t, 7 Hz, 9H, CH3–CH2–), 3.81 (q, 7 Hz, 6H, CH3–CH2–).
1,1,1,3,3-Pentamethyl-3-dimethylsiloxy disilazane, Me2HSiO-
SiMe2N(H)SiMe3 (3e): b.p. 46 °C/12 mm, n2D0 1.4185; d (1H, CDCl3,
TMS): 0.08 (s, 9H, SiMe3), 0.1 (s, 6H, SiMe2), 0.18 (d, 3 Hz, 6H,
SiMe2), 4,71 (m, 3 Hz, 1H, SiH).
this particular group in the radical is positive (A is (+)), then the
multiplication gives
l
ꢀ
ꢀ
ꢀ
D
g ꢀ A ¼ ðþÞ ꢀ ðþÞ ꢀ ðþÞ ꢀ ðþÞ
¼ ðþÞ that is; absorption ðAÞ;
and one should observe the enhanced absorption of the NMR signal
of this group.
4. Conclusion
Reaction of N-bromohexamethyl disilazane (Me3Si)2NBr (1)
with substituted silyl hydrides R1R2R3SiH (2a–g) results in asym-
metric disilazanes Me3SiNHSiR1R2R3 (3a–g) and bromotrimethyl
silane Me3SiBr (4). The simplicity of the reaction under study and
high yield of the final reaction products favour this reaction as a
convenient preparative method for asymmetric disilazanes.
Photoinitiated interaction of 1 with silyl hydrides in cyclohex-
ane leads to major reaction product, the symmetric silazane 7.
Using 1H CIDNP method, it has been demonstrated that photoiniti-
ated reaction between 1 and 2a–g in cyclohexane includes a series
of consecutive radical steps involving disilazanyl (Me3Si)2NÅ and si-
lyl R1R2R3SiÅ radicals.
Important result of the present paper is the observation of two
formation mechanisms of asymmetric silazanes in different sol-
vents – photoinduced radical reaction in cyclohexane and soft bro-
mination when the reagents are mixed in benzene. Thus, we have
found very convenient reaction which allows to perform a compar-
ative study of chirality effects in radical and nonradical reactions
resulting in identical products. This issue will be a subject of our
further investigations.
1,1,1,3,3-Pentamethyl-3-dimethylsilylamino disilazane, Me2H-
SiN(H)SiMe2N(H)SiMe3 (3f): b.p. 92 °C/2 mm, n2D0 1.4566; d (1H,
CDCl3, TMS): 0.16 (s, 9H, SiMe3), 0.19 (d, 3.5 Hz, 6H, SiMe2), 0.29
(s, 6H, SiMe2), 4,71 (m, 3.5 Hz 1H, SiH).
1,1,1-Trimethyl-3,3,3-trichloro disilazane, Me3SiN(H)SiCI3 (3g):
b.p. 40 °C/2 mm; d (1H, C6D6, TMS): 0.09 (s, 9H, SiMe3), 2.23 (s, 1H,
NH).
3.2. 1H CIDNP effects and their analyses
Acknowledgements
1H CIDNP spectra were detected using JEOL JNM FX90Q high-
resolution NMR-spectrometer (90 MHz 1H operating frequency)
equipped with special in-house device for the irradiation of the
samples directly in the probe of the spectrometer. Irradiation
Financial support of the Russian Foundation for Basic Research
(Project No. 04-03-32277), Foundation of President of Russian Fed-
eration (Science Schools’ Grant, NSh 4575.2006.3) and Federal Pro-

3820
M.B. Taraban et al. / Journal of Organometallic Chemistry 693 (2008) 3815–3820
[9] M.D. Cook, B.P. Roberts, K. Singh, J. Chem. Soc., Perkin Trans. II 5 (1983)
635.
[10] H. Fischer, K.-H. Hellwege (Eds.), Landolt–Boernstein New Series. Numerical
Data and Functional Relationship in Science and Technology, Springer-Verlag,
Berlin, 1979. Parts B and D.
[11] N. Wiberg, F. Raschig, K.H. Schmid, J. Organomet. Chem. 10 (1967) 15.
[12] S.V. Basenko, I.A. Gebel’, V.Yu. Vitkovskii, D.D. Toriashinova, R.G. Mirskov, M.V.
Voronkov, Russ. Chem. Bull. (1991) 1039.
[13] J. Devaure, Pham van Huong, J. Loscjmbe, J. Chem. Phys. 65 (1968)
1686.
[14] M.G. Voronkov, G.I. Lebedeva, Russ. Chem. Bull. 22 (1973) 2417.
[15] M.J. Kamlet, J.-L.M. Abboud, M.H. Abraham, R.W. Taft, J. Org. Chem. 48 (1983)
2877.
[16] M.H. Abraham, P.L. Grellier, J.-L.M. Abboud, R.M. Doherty, R.W. Taft, Can. J.
Chem. 66 (1988) 2673.
[17] V.I. Rakhlin, S.V. Grigor’ev, R.G. Mirskov, T.A. Podgorbunskaya, M.G. Voronkov,
D.V. Gendin, Russ. J. Gen. Chem. 73 (2003) 1952.
[18] A. Ladenburg, Ber. 5 (1872) 505.
gram of Fundamental Research ‘‘Highly Reactive Intermediates” is
gratefully acknowledged. We thank Dr. Takayuki Uno and Shinichi
Otsu (Analytical Instruments Division, JEOL Ltd.) who helped us to
maintain JEOL NMR Spectrometer up and running.
References
[1] R.E. Bailey, R. West, J. Organomet. Chem. 4 (1965) 430.
[2] N. Wiberg, F. Raschig, K.H. Schmid, J. Organomet. Chem. 10 (1967) 29.
[3] J. Lorberth, J. Organomet. Chem. 19 (1969) 435.
[4] M.B. Taraban, A.I. Kruppa, V.I. Rakhlin, S.V. Grigor’ev, O.S. Volkova, R.G.
Mirskov, T.V. Leshina, J. Organomet. Chem. 636 (2001) 12.
[5] M.B. Taraban, A.I. Kruppa, M.G. Voronkov, V.I. Rakhlin, S.V. Grigor’ev, O.S.
Volkova, R.G. Mirskov, T.V. Leshina, J. Organomet. Chem. 658 (2002) 228.
[6] A. Ladenburg, Ber. 40 (1907) 2274.
[7] W. Gee, R.A. Shaw, B.C. Smith, J. Chem Soc. (1964) 2845.
[8] B.P. Roberts, C. Wilson, J. Chem. Soc., Chem. Commun. 17 (1978) 752.
[19] R. Kaptein, Chem. Commun. (1971) 732.