Direct detection and characterization of a transient 1-silaallene derivative in solution

Article abstract of DOI:10.1021/ja962591s

Direct irradiation of ((trimethylsilyl)ethynyl)pentamethyldisilane in hydrocarbon solution affords a mixture of reactive intermediates which have been detected and identified using laser flash photolysis techniques and trapped as the methanol adducts in steady state irradiation experiments. Flash photolysis of air-saturated hexane solutions of the disilane allows detection of transient species assigned to 1,1-dimethyl-3,3-bis(trimethylsilyl)-1-silaallene and dimethylsilylene, along with a non-decaying species assigned to 1,1-dimethyl-2,3-bis(trimethylsilyl)-1-silacyclopropene. These were identified on the basis of their UV absorption spectra and reactivity toward various reagents. The 1-silaallene is a minor photoproduct, but it is readily observable in transient absorption experiments because it is relatively long-lived (τ = 2.4 μs in air-saturated solution) and absorbs strongly, with absorption maxima considerably to the red of those of the 1-silacyclopropene. The 1-silaallene exhibits characteristic silene reactivity; it reacts with methanol, acetic acid, acetone (κ = 10⁶-10⁸ M^-1 s^-1) and oxygen (κ~10⁸ M^-1 s^-1), exhibits low reactivity toward aliphatic dienes and tert-butyl alcohol, and decays with second-order kinetics in the absence of quenchers. With methanol and acetic acid, it has been shown that reaction in hexane occurs with the monomeric ROH species only. Steady state irradiations in the presence of methanol afford methanol-addition products consistent with the formation of the silaallene, silacyclopropene, and silylene, along with bis(trimethylsilyl)acetylene as the major product.

Full text of DOI:10.1021/ja962591s

466
J. Am. Chem. Soc. 1997, 119, 466-471
Direct Detection and Characterization of a Transient
1-Silaallene Derivative in Solution
Corinna Kerst, Cerrie W. Rogers, Ralph Ruffolo, and William J. Leigh*
Contribution from the Department of Chemistry, McMaster UniVersity, 1280 Main Street West,
Hamilton, Ontario L8S 4M1, Canada
ReceiVed July 26, 1996X
Abstract: Direct irradiation of ((trimethylsilyl)ethynyl)pentamethyldisilane in hydrocarbon solution affords a mixture
of reactive intermediates which have been detected and identified using laser flash photolysis techniques and trapped
as the methanol adducts in steady state irradiation experiments. Flash photolysis of air-saturated hexane solutions
of the disilane allows detection of transient species assigned to 1,1-dimethyl-3,3-bis(trimethylsilyl)-1-silaallene and
dimethylsilylene, along with a non-decaying species assigned to 1,1-dimethyl-2,3-bis(trimethylsilyl)-1-silacyclopropene.
These were identified on the basis of their UV absorption spectra and reactivity toward various reagents. The
1-silaallene is a minor photoproduct, but it is readily observable in transient absorption experiments because it is
relatively long-lived (τ ) 2.4 µs in air-saturated solution) and absorbs strongly, with absorption maxima considerably
to the red of those of the 1-silacyclopropene. The 1-silaallene exhibits characteristic silene reactivity; it reacts with
methanol, acetic acid, acetone (k ) 106-108 M-1 s-1) and oxygen (k ∼ 108 M-1 s-1), exhibits low reactivity toward
aliphatic dienes and tert-butyl alcohol, and decays with second-order kinetics in the absence of quenchers. With
methanol and acetic acid, it has been shown that reaction in hexane occurs with the monomeric ROH species only.
Steady state irradiations in the presence of methanol afford methanol-addition products consistent with the formation
of the silaallene, silacyclopropene, and silylene, along with bis(trimethylsilyl)acetylene as the major product.
Introduction
have demonstrated that the yields of 1-silaallene- and silylene
(:SiR′2)-derived products are both significantly higher from
((trialkylsilyl)ethynyl)disilanes than from arylethynyl-deriva-
tives.4,5 This suggests that compounds of this type might prove
to be convenient precursors to reactive 1-silaallene derivatives
for study by transient spectroscopic methods, particularly if
silylene extrusion might be suppressed through appropriate
substitution. Thus, we have identified ((trimethylsilyl)ethynyl)-
pentamethyldisilane (1) as a particularly promising candidate
for study, and have examined its photorearrangement to the
isomeric silacyclopropene (2) and 1-silaallene (3) derivatives
in hydrocarbon solution.
1-Silaallenes are a class of silenes (compounds containing a
SidC bond) that are of fundamental interest in organosilicon
chemistry,1 but which have received scant attention in the
literature. West and co-workers have recently reported the first
example of an isolable 1-silaallene derivative,2 and a ruthenium
complex of a more reactive 1-silaallene has also been reported.3
However, no reactive 1-silaallene derivatives have yet been
characterized spectroscopically, either in solution or solid
matrices. In this paper, we report the first example of a transient
1-silaallene to be detected and characterized in solution at room
temperature.
The direct irradiation of alkynyldisilanes in solution is known
to lead to the formation of products ([2 + 2]-dimers or in the
presence of alcohols, alkoxysilanes) consistent with the forma-
tion of 1-silaallene derivatives in low to moderate yields.4-6 In
most of these cases, the major product is the isomeric 1-sila-
cyclopropene derivative (eq 1), which is generally isolable in
Results and Discussion
Compound 1 was prepared using a similar method to that
reported by Ishikawa and co-workers for the synthesis of related
alkynyldisilanes4,5 and was obtained in >99% purity after
chromatographic purification. This compound has been reported
previously,9 but no spectroscopic or analytical data were
reported; those determined here are consistent with its proposed
structure.
Irradiation of a deoxygenated hexane solution containing 1
(0.02 M) and MeOH (0.5 M) with the light from a 16-W Cd
resonance lamp (228 nm) was carried to ca. 50% conversion of
spite of possessing considerable reactivity toward oxygen,
alcohols, and carbonyl compounds.5-8 Ishikawa and co-workers
X Abstract published in AdVance ACS Abstracts, January 1, 1997.
(1) Gordon, M. S.; Koob, R. D. J. Am. Chem. Soc. 1981, 103, 2939.
(2) Miracle, G. E.; Ball, J. L.; Powell, D. R.; West, R. J. Am. Chem.
Soc. 1993, 115, 11598.
(3) Yin, J.; Klosin, J.; Abboud, K. A.; Jones, W. M. J. Am. Chem. Soc.
1995, 117, 3298.
(4) Ishikawa, M.; Fuchikami, T.; Kumada, M. J. Am. Chem. Soc. 1977,
99, 245.
(5) Ishikawa, M.; Nishimura, K.; Sugisawa, H.; Kumada, M. J. Orga-
nomet. Chem. 1980, 194, 147.
(6) Ishikawa, M.; Sugisawa, H.; Fuchikami, T.; Kumada, M.; Yamabe,
T.; Kawakami, H.; Fukui, K.; Ueki, Y.; Shizuka, H. J. Am. Chem. Soc.
1982, 104, 2872.
(7) Seyferth, D.; Annarelli, D. C.; Vick, S. C. J. Am. Chem. Soc. 1976,
98, 6382.
(8) Seyferth, D.; Annarelli, D. C.; Vick, S. C. J. Organomet. Chem. 1984,
272, 123.
(9) Sakurai, H.; Fujii, T.; Sakamoto, K. Chem. Lett. 1992, 339.
S0002-7863(96)02591-7 CCC: $14.00 © 1997 American Chemical Society

Characterization of a Transient 1-Silaallene DeriVatiVe
J. Am. Chem. Soc., Vol. 119, No. 3, 1997 467
the alkynyldisilane. This resulted in the formation of four major
products, according to GC analysis of the photolysate; these
were isolated by radial chromatography and identified as 4-7
on the basis of spectroscopic data (see eq 2). Product yields
pulse level, while those recorded over the λmon ) 250-270 nm
range showed an initial decay (of identical temporal behavior
to those recorded at higher wavelengths) to a residual absorption
level that was stable over the full 5-ms time scale of our system.
Saturation of the solution with air led to a reduction in the
lifetime of the transient absorption and a change to pseudo-
first-order decay kinetics (τ ∼ 2.4 µs) but had no effect on the
lifetime or yield of the non-decaying residual absorption.
Experiments carried out with a nitrogen-saturated solution using
a shorter time scale revealed the presence of an additional
transient species which absorbs in the 400-600 nm range and
decays with a ∼50 ns lifetime. This component could not be
detected in experiments with air-saturated solutions. Thus, three
primary products are detectable from photolysis of 1: an
extremely short-lived one which absorbs in the visible and is
very reactive toward oxygen (k > 109 M-1 s-1), a moderately
short-lived one which absorbs in the UV and is moderately
reactive toward oxygen (k ∼ 108 M-1 s-1), and an apparently
stable one (τ > 50 ms) which absorbs in the deep UV and is
not detectably reactive toward oxygen within the time resolution
of our system (k < 103 M-1 s-1).
Time-resolved UV absorption spectra, recorded for air-
saturated solutions 0.2-1 and 22-33 µs after the laser pulse,
are shown in Figure 1a, along with a spectrum recorded for a
nitrogen-saturated solution 15-55 ns after the pulse. The Inserts
in Figure 1a show decay traces recorded at 270 nm (air-
saturated), where both the long-lived transient and “stable”
product can be detected, and at 450 nm over the shorter time
scale. Figure 1b shows the isolated spectrum of the long-lived
transient species, calculated by simple subtraction of the two
air-saturated spectra in Figure 1a. The short-lived transient with
λmax ) 460 nm can be assigned to dimethylsilylene, on the basis
of its characteristic absorption spectrum and sensitivity to
oxygen and alcohols.10,11
Addition of O2, MeOH, MeOD, acetic acid (HOAc), or
acetone to the hexane solutions of 1 resulted in shortening of
the lifetime of the 325-nm transient absorption, but had no effect
on its initial “yield” (i.e. the initial ∆-OD of the transient) or
on the yield or stability of the residual absorption over a 5-ms
time scale. This indicates that these reagents all react with the
transient but not with the non-decaying species, and furthermore,
that they do not react efficiently with the photochemical
precursor(s) to these species. A plot of the transient decay rate
constant (kdecay) versus quencher (Q) concentration according
to eq 3 was linear for acetone, affording a value of kq ) (1.78
( 0.05) × 106 M-1 s-1. The rate constant for quenching by
O2 was estimated to be kq ) (1.1 ( 0.2) × 108 M-1 s-1 from
the lifetimes of the transient in air- (τ ) 2.4 µs) and O2-saturated
(τ ) 640 ns) hexane solution. 1,3-Octadiene and tert-butyl
alcohol were ineffective transient quenchers; lifetimes measured
in the absence and presence of relatively high concentrations
of added reagent afforded upper limits of kq e 4 × 104 M-1
s-1 for the quenching rate constants in these cases.
were determined by GC analysis as a function of irradiation
time. Monitoring of the photolysate between 5 and 60%
conversion indicated compound 5 to be a secondary product,
and also revealed the presence of dimethylmethoxysilane (8)
in the reaction mixture.
The methanol-adducts 4-6 were identified on the basis of
one- and two-dimensional 1H, 13C, and 29Si NMR experiments.
1H-29Si heteronuclear multiple bond correlation (HMBC)
1
experiments allowed determination of H-29Si coupling con-
stants between the vinylic hydrogen and the three silicon atoms
in the three compounds; those involving the alkoxysilyl silicon
atom (JH-SiO) were the most useful for structural assignments,
assuming the normal ordering Jtrans > Jcis . Jgeminal. The three
compounds exhibited JH-SiO values of 17.6, 21.6, and 6.8 Hz,
and were thus assigned as 4, 5, and 6, respectively.
Compound 4 has been reported previously, as the product of
addition of MeOH to silacyclopropene 2,7,8 and is most likely
responsible for the formation of 5 by secondary photoisomer-
ization at higher conversions.7,8 Compound 6 can be attributed
to reaction of MeOH with 1,1-dimethyl-3,3-bis(trimethylsilyl)-
1-silaallene (3).4-6
Corroboration of the structural assignments for 4-6 and the
proposed pathways by which they arise was derived from an
experiment in which an argon-saturated hexane solution of 1
was irradiated to ca. 20% conversion in the absence of methanol.
The photolysate in this case contained only two major products,
7 (ca. 40%) and a compound identifiable as 2 on the basis of
its GC retention time and mass spectrum.8 The photolysate also
contained at least ten minor products of relatively high molecular
weight which were not present in the reaction mixture from
photolysis in the presence of methanol. Addition of MeOH to
the photolysate resulted in the rapid disappearance of the
compound identified as 2 and the formation of 4, but had no
effect on the yields of the other products of photolysis. Most
significantly, compound 6 was not produced in this experiment.
This verifies that the product arises from reaction of methanol
with a relatively short-lived reactive intermediate, as silaallene
3 is expected to be. Presumably, the numerous minor products
formed in this experiment can be assigned as due to reaction of
silaallene 3 with itself and/or 1.
Laser flash photolysis of continuously flowing, deoxygenated
0.028 M hexane solutions of 1 (ꢀ248 ) 85 M-1 cm-1) with the
248-nm pulses from a Kr/F2 excimer laser produced readily
detectable, long-lived absorptions throughout the 250-350 nm
spectral range. The appearance of the traces varied markedly
depending on the monitoring wavelength (λmon). Those recorded
over the λmon ) 300-350 nm range decayed with complex
kinetics over several tens of microseconds to nearly the pre-
kdecay ) (1/τo) + kq[Q]
(3)
On the basis of these trends in reactivity, the transient giving
rise to the absorption spectrum shown in Figure 1b (λmax
275, 325 nm) is assigned to 1-silaallene 3. The spectrum of
the stable residual component, with its weak band at λmax
)
)
345 nm and prominent absorption below 250 nm, is assigned
(10) Levin, G.; Das, P. K.; Bilgrien, C.; Lee, C. L. Organometallics 1989,
8, 1206.
(11) Baggott, J. E.; Blitz, M. A.; Frey, H. M.; Lightfoot, P. D.; Walsh,
R. Int. J. Chem. Kinet. 1992, 24, 127.

468 J. Am. Chem. Soc., Vol. 119, No. 3, 1997
Kerst et al.
As is the case for acetone, the reaction of 3 with MeOH,
MeOD, and HOAc proved to be significantly slower than those
reported previously for other silenes.16-20 Plots of kdecay Vs [Q]
for these three reagents in hexane solution (see Figure 2) exhibit
negative curvature, which is barely detectable for MeOH and
MeOD, but distinct in the case of acetic acid. Treatment of
the data for MeOH(D) according to the simple expression given
in eq 3 (i.e., ignoring the slight curvature) affords apparent
quenching rate constants of kq ) (1.7 ( 0.1) × 106 M-1 s-1
for MeOH and kq ) (7.4 ( 0.8) × 105 M-1 s-1 for MeOD, and
the indication of a sizable primary deuterium kinetic isotope
effect (kH/kD ) 2.3 ( 0.2) on the addition of MeOH to 3.
A rigorous analysis of the data for MeOH and HOAc must
take into account the fact that both reagents undergo significant
oligomerization in hexane at bulk concentrations greater than
ca. 0.01 M.21,22 Oligomerization will have significant effects
on the kinetics of reaction, since it will cause the actual
concentration of potentially reactive species to be lower than
the nominal bulk ROH concentration. These effects will be
even greater if the reactivity of ROH oligomers differs from
that of the monomer owing to differences in nucleophilicity and/
or acidity. Such effects have been reported previously for
singlet carbene insertion reactions with alcohols, where the rate
determining step is protonation of the carbene and is faster for
oligomeric than monomeric alcohol.23 In contrast to the
situation with carbenes, the negative curvature in the quenching
plots for 3 (Figure 2) suggests that oligomerization reduces the
overall reactivity of MeOH and HOAc toward addition to the
SidC bond in 3.
Addition of alcohols and acetic acid to silenes is thought to
occur by a two-step mechanism involving initial formation of
a zwitterionic silene-alcohol complex, which collapses to
product by proton transfer from oxygen to the silenic carbon
(see Scheme 1).16,18,20,24-26 With methanol, the proton-transfer
step is the slow one in the sequence, and occurs by two
competing pathways: one which predominates at low alcohol
concentrations, involving unimolecular transfer within the
complex, and a higher order pathway involving a second
molecule of alcohol.16,18,20,25 The latter has been identified as
general base catalysis (i.e., deprotonation at oxygen followed
by protonation at carbon) on the basis of kinetic results.18 The
overall rate of reaction in this case depends on alcohol
nucleophilicity, since the equilibrium constant for complex
formation is contained in the expression for the overall rate
constant. Reaction of silenes with HOAc differs in that
Figure 1. (a) Transient UV absorption spectra recorded by nanosecond
laser flash photolysis of 0.028 M solutions of 1 in hexane at 23 °C:
0.2-1.0 (b) and 22-33 µs (9) after excitation of an air-saturated
solution and 15-55 ns ([) after excitation of a nitrogen-saturated
solution. (b) Difference spectrum calculated from the two air-saturated
spectra of part a. Typical decay traces, recorded at monitoring
wavelengths of 270 and 450 nm, are shown as Inserts in part a.
to 1-silacyclopropene 2 on the basis of comparison to the
published UV absorption spectrum.7,8 As might be expected,
its known reactivity toward oxygen and alcohols7,8 is not high
enough to be detected under the conditions of our experiments
(kO < 103 M-1 s-1; kMeOH < 10 M-1 s-1).
2
The UV absorption spectrum of 3 can be compared to that
of the simple silenic model compound 9 (λmax ) 275, 300 nm
(sh)) in the same solvent.12 The ∼30 nm bathochromic shift
of the spectral maximum for 9 relative to that of 1,1-
dimethylsilene (λmax ) 244-nm13) can be ascribed to the effects
of hyperconjugative interactions between the â-trimethylsilyl
group and the SidC bond,12 based on analogies in alkene
spectroscopy.14,15 More substantial spectroscopic effects are
anticipated for 3, since it contains two trimethylsilyl groups in
the required orientation for hyperconjugation with the SidC
bond.
(16) Sluggett, G. W.; Leigh, W. J. J. Am. Chem. Soc. 1992, 114, 1195.
(17) Leigh, W. J.; Bradaric, C. J.; Sluggett, G. W. J. Am. Chem. Soc.
1993, 115, 5332.
(18) Leigh, W. J.; Sluggett, G. W. J. Am. Chem. Soc. 1994, 116, 10468.
(19) Leigh, W. J.; Sluggett, G. W. Organometallics 1994, 13, 269.
(20) Leigh, W. J.; Bradaric, C. J.; Kerst, C.; Banisch, J. H. Organome-
tallics 1996, 15, 2246.
(21) Landeck, H.; Wolff, H.; Gotz, R. J. Phys. Chem. 1977, 81, 718.
(22) Fujii, Y.; Yamada, H.; Mizuta, M. J. Phys. Chem. 1988, 92, 6768.
(23) Griller, D.; Liu, M. T. H.; Scaiano, J. C. J. Am. Chem. Soc. 1982,
104, 5549.
(24) Wiberg, N. J. Organomet. Chem. 1984, 273, 141.
(25) Kira, M.; Maruyama, T.; Sakurai, H. J. Am. Chem. Soc. 1991, 113,
3986.
(12) Conlin, R. T.; Ezhova, M. B.; Leigh, W. J.; Venneri, P. C.; Bradaric,
C. J. To be submitted for publication.
(13) Maier, G.; Mihm, G.; Reisenauer, H. P. Chem. Ber. 1984, 117, 2351.
(14) Jarvie, A. W. P. Organomet. Chem. ReV. A 1970, 6, 153.
(15) Ponec, R.; Chvalovsky, V.; Cernysev, E. A.; Komarenkova, N. G.;
Baskirova, S.A. Collect. Czech. Chem. Commun. 1974, 39, 1177.
(26) The structures of the intermediates in Scheme 1 are those thought
to be involved in alcohol additions to silenes. Addition of HOAc probably
involves (rate-determining) attack by the carbonyl oxygen, followed by or
in concert with proton transfer. See: Bradaric, C. J.; Leigh, W. J. J. Am.
Chem. Soc. 1996, 118, 8971.

Characterization of a Transient 1-Silaallene DeriVatiVe
J. Am. Chem. Soc., Vol. 119, No. 3, 1997 469
Figure 3. Plots of kdecay versus monomeric HOAc concentration for
quenching of 1-silaallene 3 in air-saturated hexane solution at 23 °C.
Scheme 1
Figure 2. Plots of kdecay versus bulk ROH concentration for quenching
of 1-silaallene 3 in air-saturated hexane solution at 23 °C by (a) HOAc
and (b) MeOH (b) and MeOD (O).
nucleophilic attack is rate determining, followed by fast proton
transfer.16,18,20 In both cases, it would be predicted that ROH
oligomers should be substantially less reactive than the mono-
mers, because of the dependence of the rate on nucleophilicity.
The data for acetic acid are the simplest to treat quantitatively,
since the mechanism is kinetically simpler (see Scheme 126),
and it is known that oligomers higher than the dimer are not
formed to a significant extent.22 The data of Figure 2a are
replotted Versus monomeric HOAc concentration in Figure 3,
using the reported equilibrium constant for dimeric HOAc
formation in hexane (K2 ) 3200 ( 500 M-1 at 25 °C)22 to
calculate the monomer concentrations as a function of the bulk
concentration. The linearity in the plot is excellent, and the
slope affords a value of kq ) (9.8 ( 0.6) × 107 M-1 s-1 for the
bimolecular rate constant for quenching of 3 by monomeric
acetic acid in hexane solution.
Treatment of the data for MeOH(D) quenching in analogous
fashion, employing the oligomerization data of Wolff and co-
workers for the alcohols in n-hexane,21 leads to less satisfactory
results because the data are of poorer quality than those for
HOAc. Nevertheless, plots of kdecay Vs monomeric MeOH(D)
concentration exhibit strong positive curvature, suggesting a
quadratic concentration dependence analogous to that reported
previously for the reaction of other reactive silenes with this
alcohol.16,20 This would be consistent with the addition mech-
anism shown in Scheme 1, with a particularly strong contribution
from the pathway which is overall second order in methanol. It
is interesting to note that the reported equilibrium constants for
oligomerization of MeOD in n-hexane are somewhat higher than
those for MeOH,21 suggesting that the true isotope effect on
the rate constant for methanol addition to 3 may be substantially
smaller than that suggested by the data in Figure 2b.
The absolute rate constants for reaction of silaallene 1 with
the reagents studied in this work are collected in Table 1. In
all cases, these values were obtained by analysis of kdecay vs
concentration data according to eq 3 (and for HOAc, using
calculated values of the monomeric acid concentration).
The results of steady state photolysis of 1 suggest that silylene
extrusion (yielding Me2Si: and 7) competes with trimethylsilyl
migration (yielding 2 and 3), to an extent comparable to that
observed for 1-((trimethylsilyl)ethynyl)-1,1-diaryl-2,2,2-tri-
methyldisilanes (10).5,27 The predominance of this product has
no effect on the spectra of the longer-lived species or the rate

470 J. Am. Chem. Soc., Vol. 119, No. 3, 1997
Kerst et al.
Table 1. Absolute Rate Constants for Quenching of
1,1-Dimethyl-3,3-bis(trimethylsilyl)-1-silaallene (3) in Air-Saturated
Hexane Solution at 23.0 ( 0.2 °Ca
Hexane (BDH Omnisolv) was used as received from the supplier.
Quenchers were of the highest purity available and were used as
received from Aldrich Chemical Co. Chloropentamethyldisilane was
prepared by the method of Ishikawa and co-workers.28 It was used in
the synthesis of 1 as a ∼80:20 mixture with hexamethyldisilane.
quencher
kq/106 M-1 s-1
MeOH
1.7 ( 0.1
0.74 ( 0.08
∼0.02b
((Trimethylsilyl)ethynyl)pentamethyldisilane (1) was prepared by
a modification of the procedure reported by Ishikawa and co-workers
for related alkynyldisilanes.6 An oven-dried, 2-neck, 250-mL, round-
bottom flask fitted with magnetic stirrer, rubber septum, and nitrogen
inlet was placed in a dry ice-2-propanol bath. Anhydrous ether (60
mL) and (trimethylsilyl)acetylene (13.0 g, 0.13 mol) were introduced
by syringe. After stirring for ∼15-min, n-butyllithium (85 mL of a
1.6 M hexane solution; 0.14 mol) was added over several minutes.
The dry ice bath was removed, the mixture was allowed to warm for
20 min, and chloropentamethyldisilane (0.13 mol) was added over 1 h
via syringe after cooling the reaction mixture back down to -78 °C.
The resulting mixture was stirred at room temperature under nitrogen
for 12 h, quenched by slow addition of distilled water (100 mL), and
extracted with ether (5 × 50 mL). The combined ether extracts were
washed with water (2 × 50 mL), dried over anhydrous magnesium
sulfate, and filtered. Evaporation of the solvent on the rotary evaporator
yielded a yellow liquid (∼50 mL), which was purified in 1-mL portions
by radial chromatography using a 4-mm silica plate and hexane as the
eluant. Fractions containing >99.8% of the major component of the
reaction mixture (as determined by GC analysis) were combined and
stripped of solvent, resulting in an isolated yield of 8.2 g (0.036 mol,
27%) of ((trimethylsilyl)ethynyl)pentamethyldisilane (1; bp 48 °C (0.3
mmHg)). The compound was identified on the basis of the following
data: 1H NMR, δ 0.095 (s, 9H), 0.143 (s, 9H), 0.167 (s, 6H); 13C NMR,
δ -3.12, -2.68, 0.03, 112.92, 116.5; 29Si NMR, δ -37.55 (SiMe3),
-19.41 (SiMe3), -18.95 (SiMe2); MS, m/e (I) 228 (49), 213 (70), 155
(100), 140 (55), 119 (71); exact mass, Calcd for C10H24Si3, 228.1186,
found 228.1198; UV (hexane), λmax (ꢀ/M-1 cm-1) ) 195 (17750), 217
nm (7100).
MeOD
t-BuOHb
HOAcc
acetone
O2
98 ( 6
1.78 ( 0.05
110 ( 20
e0.04b
1,3-octadieneb
a From analysis of kdecay vs concentration data according to eq 3.
Errors are quoted as twice the standard deviation obtained from linear
least-squares analysis of these data. b Estimates obtained from kdecay
values obtained in the absence and presence of quencher at a single
(high) concentration. c For quenching by monomeric HOAc (the slope
of the plot shown in Figure 3).
constants for quenching of 3, because it is over two orders of
magnitude shorter lived than 3 in air-saturated solutions (even
more so in the presence of additional quenchers). Quenching
of Me2Si: by oxygen, alcohols, etc. is not accompanied by a
reduction in the initial ∆-OD of the absorption assigned to 3,
allowing the conclusion that the silaallene is formed directly
by photorearrangement of 1. The relative yield of 2 and 3 from
photolysis of 1 is also similar to that obtained from photolysis
of 10. In spite of the fact that 3 is formed in only ∼15%
chemical yield, it is readily detectable by flash photolysis
because its UV absorption spectrum is relatively intense and is
well-shifted from those of the precursor and the other products
formed.
Qualitatively, 1-silaallene 3 exhibits reactivity which is
characteristic of silicon-carbon double bonds. On a more
quantitative level, the absolute rate constants for reaction toward
the characteristic silene traps studied here (see Table 1) are
significantly slower than those determined for other transient
silenes which have been studied. It is premature to speculate
as to the origins of these differences; fortunately, the manipula-
tion of both the alkynyl and silyl substituents in 1 is synthetically
straightforward and affords convenient photochemical precursors
to a variety of transient 1-silaallene derivatives. The spectro-
scopic characterization and more detailed studies of the bimo-
lecular reactivity of these organosilicon reactive intermediates
are the subject of continued investigation in our laboratory.
Preparative scale irradiations of 1 were carried out using a cadmium
resonance lamp (228 nm; Philips 93107E) and a water-cooled Pyrex
immersion well with a Suprasil inner sleeve. A solution of 1 (0.32 g,
1.4 mmol), n-dodecane (0.027 g), and anhydrous methanol (1.12 g, 35
mmol) in anhydrous hexane (70 mL) was stirred vigorously and
continuously degassed with a stream of dry nitrogen throughout
photolysis to 48% conversion (1.5 h), with periodic monitoring of the
course of the photolysis by GC. Three major products were formed,
in approximate yields (determined from the slopes of concentration vs
time plots, with “concentrations” determined from the relative areas
of the GC peaks due to product and n-dodecane) of 42% (7), 12% (6),
and 23% (4) (listed in order of increasing GC retention time).
Compound 5 (elutes before 6; ca. 4% after 48% conversion) increased
in yield throughout the course of the photolysis, allowing it to be
identified as a secondary photoproduct. Compound 7 was identified
by GC/MS and by coinjection of the photolysate with an authentic
sample of bis(trimethylsilyl)acetylene (Aldrich). The other three
products were isolated from the photolysate as colorless liquids by radial
chromatography (using a 2-mm silica gel plate and hexane as the eluant)
after evaporation of the solvent and volatiles. Coinjection of each of
the purified components with a portion of the original photolysate
verified that they survived purification intact. They were identified
on the basis of the following spectroscopic data (JSi-H refer to coupling
constants between silicon atoms and the single vinylic hydrogen in
each of the adducts):
Experimental Section
1H and 13C NMR spectra were recorded on a Bruker DRX500
spectrometer in deuteriochloroform solution and are referenced to
tetramethylsilane. 29Si NMR spectra were obtained from 1H-29Si
gradient heteronuclear multiple bond correlation (HMBC) experiments,
also in deuteriochloroform. High-resolution mass spectra and exact
masses were determined on a VGH ZABE mass spectrometer. Low-
resolution spectra were recorded on a Hewlett-Packard 5890II gas
chromatograph equipped with an HP-5971A mass selective detector
and a DB5 fused silica capillary column (30 m × 0.25 mm;
Chromatographic Specialties, Inc.). Ultraviolet absorption spectra were
recorded on a Perkin Elmer Lambda 9 spectrometer interfaced to an
IBM PS/2-286 microcomputer, or on a Hewlett-Packard HP8451 UV
spectrometer. Gas chromatographic (GC) analyses were carried out
using a Hewlett-Packard 5890II+ gas chromatograph equipped with a
conventional heated injector, a flame ionization detector, a Hewlett-
Packard 3396A integrator, and a DB1701 megabore capillary column
(15 m × 0.53 mm; Chromatographic Specialties, Inc). Radial chro-
matography was carried out using a Chromatotron (Harrison Research)
and 2- or 4-mm silica gel 60 thick layer plates.
1-(Methoxydimethylsilyl)-(E)-1,2-bis(trimethylsilyl)ethene (4). 1H
NMR, δ 0.159 (s, 9H), 0.166 (s, 6H), 0.169 (s, 9H), 3.364 (s, 3H),
7.426 (s, 1H); 13C NMR, δ -1.344 (SiMe2), 0.765 (SiMe3), 1.629
(SiMe3), 50.259 (OMe), 165.868 (dCH), 166.867 (dC); 29Si NMR, δ
-10.85 (SiMe3, JSi-H ) 5.1 Hz), -9.04 (SiMe3, JSi-H ) 21.5 Hz),
10.29 (SiMe2, JSi-H ) 17.6 Hz); MS, m/e (I) 260 (0.5), 245 (20), 187
(33), 157 (27), 156 (28), 147 (11), 141 (32), 89 (62), 73 (100), 59
(49).
1-(Methoxydimethylsilyl)-(Z)-1,2-bis(trimethylsilyl)ethene (5). 1H
NMR, δ 0.065 (s, 9H), 0.109 (s, 9H), 0.189 (s, 6H), 3.373 (s, 3H),
(27) (a) Ishikawa, M.; Kovar, D.; Fuchikami, T.; Nishimura, K.; Kumada,
M.; Higuchi, T.; Miyamoto S. J. Am. Chem. Soc. 1981, 103, 2324. (b)
Ishikawa, M.; Matsuzawa, S.; Sugisawa, H.; Yano, F.; Kamitori, S.; Higuchi,
T. J. Am. Chem. Soc. 1985, 107, 7706.
(28) Ishikawa, M.; Kumada, M.; Sakurai, H. J. Organomet. Chem. 1970,
23, 63.

Characterization of a Transient 1-Silaallene DeriVatiVe
J. Am. Chem. Soc., Vol. 119, No. 3, 1997 471
7.335 (s, 1H); 13C NMR, δ -0.169 (SiMe3), -0.028 (SiMe2), 0.529
(SiMe3), 49.322 (OMe), 165.886 (dCH); 29Si NMR, δ -11.01 (SiMe3,
JSi-H ) 6.2 Hz), -3.09 (SiMe3, JSi-H ) 16.6 Hz), 6.19 (SiMe2, JSi-H
) 21.6 Hz); MS, m/e (I) 260 (0.5), 246 (19), 187 (32), 157 (34), 156
(26), 147 (11), 141 (32), 89 (62), 83 (11), 73 (100), 59 (48).
1-(Methoxydimethylsilyl)-2,2-bis(trimethylsilyl)ethene (6). 1H
NMR, δ 0.088 (s, 9H), 0.161 (s, 9H), 0.224 (s, 6H), 3.404 (s, 3H),
7.160 (s, 1H); 13C NMR, δ -0.591 (SiMe2), 0.225 (SiMe3), 1.601
(SiMe3), 50.117 (OMe), 159.524 (dCH), 172.301 (dC); 29Si NMR, δ
-7.87 (SiMe3, JSi-H ) 16.8 Hz), -0.79 (SiMe3, JSi-H ) 21.3 Hz),
2.95 (SiMe2, JSi-H ) 6.8 Hz); MS, m/e (I) 260 (0.7), 245 (19), 187
(30), 157 (35), 156 (24), 147 (11), 141 (28), 89 (57), 83 (11), 73 (100),
59 (55).
Analytical-scale photolyses were carried out using the same lamp,
but using solutions contained in quartz tubes (50 mm × 5 mm) which
were clamped ∼0.5 in. from the light source. Photolysis of 1 in the
absence of methanol employed a 0.5-mL aliquot of a 0.02 M solution
of 1 in anhydrous hexane solution, which was sealed in a 50 × 5 mm
tube with a rubber septum and deoxygenated with a stream of dry argon.
The contents were photolyzed with periodic monitoring by GC to check
the course of the photolysis. Two major volatile products and ∼10
minor ones of considerably longer retention time than 1 could be
detected in the photolysis mixture. One of the major products was
identified as 7 (∼40%), while the other was identified as 1,1-dimethyl-
2,3-bis(trimethylsilyl)-1-silacyclopropene (2, ∼20%) by comparison
of its mass spectrum (recorded by GC/MS) to literature data:8 MS, m/e
(I) 228 (25), 213 (34), 156 (11), 155 (61), 140 (32), 125 (16), 97 (18),
83 (17), 73 (100). Addition of an excess of anhydrous methanol to
the photolysate resulted in the disappearance of 2 and concomitant
formation of 4.
filled with F2/Kr/He mixtures, and a microcomputer-controlled detection
system.29 The system incorporates a brass sample holder whose
temperature is controlled to within 0.1 °C by a VWR 1166 constant
temperature circulating bath. Solutions were prepared at concentrations
such that the absorbance at the excitation wavelength (248 nm) was
ca. 0.7 (0.028 M), and were flowed continuously through a 3 × 7 mm
Suprasil flow cell connected to a calibrated 100-mL reservoir. Where
required, nitrogen was bubbled continuously through the reservoir
throughout the experiments. Solution temperatures were measured with
a Teflon-coated copper/constantan thermocouple which was inserted
directly into the flow cell. Quenchers were added directly to the
reservoir by microlitre syringe as aliquots of standard solutions. Rate
constants were calculated by linear least-squares analysis of decay rate-
concentration data (6 or more points) which spanned at least one order
of magnitude in the transient decay rate. Errors are quoted as twice
the standard deviation obtained from the least squares analysis in each
case.
Acknowledgment. We gratefully acknowledge the assistance
of D. Hughes with high-resolution NMR experiments and the
McMaster Regional Centre for Mass Spectrometry for mass
spectra and exact mass determinations. We wish to thank the
Natural Sciences and Engineering Research Council of Canada
for financial support and the Max-Planck-Gesellschaft zur
Fo¨rderung der Wissenschaften for an Otto Hahn Fellowship to
C.K.
JA962591S
Nanosecond laser flash photolysis experiments employed the pulses
(248 nm; ca.16 ns; 70-120 mJ) from a Lumonics 510 excimer laser
(29) Leigh, W. J.; Workentin, M. S.; Andrew, D. J. Photochem.
Photobiol. A: Chem. 1991, 57, 97.