Substituent effects on the reactivity of the silicon-carbon double bond. Resonance, inductive, and steric effects of substituents at silicon on the reactivity of simple 1-methylsilenes

Article abstract of DOI:10.1021/ja981435d

The reactivities of a series of substituted 1-methylsilenes RMeSi=CH₂ (R = H, methyl, ethyl, t-butyl, vinyl, ethynyl, phenyl, trimethylsilyl, and trimethylsilymethyl) in hydrocarbon solvents have been investigated by far- UV (193-nm) laser flash photolysis techniques, using the corresponding 1- methylsilacyclobutane derivatives as silene precursors. Each of these silacyclobutanes yields ethylene and the corresponding silene, which can be trapped as the alkoxysilane RSiMe₂OR' cleanly upon 193- or 214-nm photolysis in solution in the presence of aliphatic alcohols. UV absorption spectra and absolute rate constants for reaction of the silenes with methanol, ethanol, and t-butyl alcohol have been determined in hexane solution at 23°C. The rate constants vary from a low 3 x 10⁷ M^-1 s^-1 for reaction of 1- methyl-1-trimethylsilylsilene with t-BuOH to a high of 1 x 10¹⁰ M^-1 s^{- 1} for reaction of 1-ethynyl-1-methylsilene with MeOH. In several cases, rate constants have been determined for addition of the deuterated alcohols, and for addition of methanol over the 0-55°C range. Invariably, small primary deuterium kinetic isotope effects and negative Arrhenius activation energies are observed. These characteristics are consistent with a mechanism involving reversible formation of a silene-alcohol complex which collapses to alkoxysilane by unimolecular proton transfer from oxygen to carbon. Silene reactivity increases with increasing resonance electron-donating and inductive electron-withdrawing ability of the substituents at silicon and is significantly affected by steric effects within this series of compounds. This is suggested to be due to a combination of effects on both the degree of electrophilicity at silicon (affecting the rate constants for formation and reversion of the complex) and nucleophilicity at carbon (affecting the partitioning of the complex between product and free reactants). Two 1- methyl-1-alkoxysilacyclobutanes were also investigated, but proved to be inert to 193-nm photolysis.

Full text of DOI:10.1021/ja981435d

9504
J. Am. Chem. Soc. 1998, 120, 9504-9512
Substituent Effects on the Reactivity of the Silicon-Carbon Double
Bond. Resonance, Inductive, and Steric Effects of Substituents at
Silicon on the Reactivity of Simple 1-Methylsilenes
William J. Leigh,* Rabah Boukherroub, and Corinna Kerst‡
†
Department of Chemistry, McMaster UniVersity, 1280 Main Street West,
Hamilton, Ontario, Canada L8S 4M1
ReceiVed April 27, 1998
Abstract: The reactivities of a series of substituted 1-methylsilenes RMeSidCH2 (R ) H, methyl, ethyl, t-butyl,
vinyl, ethynyl, phenyl, trimethylsilyl, and trimethylsilylmethyl) in hydrocarbon solvents have been investigated
by far-UV (193-nm) laser flash photolysis techniques, using the corresponding 1-methylsilacyclobutane
derivatives as silene precursors. Each of these silacyclobutanes yields ethylene and the corresponding silene,
which can be trapped as the alkoxysilane RSiMe2OR′ cleanly upon 193- or 214-nm photolysis in solution in
the presence of aliphatic alcohols. UV absorption spectra and absolute rate constants for reaction of the silenes
with methanol, ethanol, and t-butyl alcohol have been determined in hexane solution at 23 °C. The rate constants
7
-1 -1
vary from a low of 3 × 10 M
s
for reaction of 1-methyl-1-trimethylsilylsilene with t-BuOH to a high of
1
0
-1 -1
1
× 10
M
s
for reaction of 1-ethynyl-1-methylsilene with MeOH. In several cases, rate constants have
been determined for addition of the deuterated alcohols, and for addition of methanol over the 0-55 °C range.
Invariably, small primary deuterium kinetic isotope effects and negative Arrhenius activation energies are
observed. These characteristics are consistent with a mechanism involving reversible formation of a silene-
alcohol complex which collapses to alkoxysilane by unimolecular proton transfer from oxygen to carbon.
Silene reactivity increases with increasing resonance electron-donating and inductive electron-withdrawing
ability of the substituents at silicon and is significantly affected by steric effects within this series of compounds.
This is suggested to be due to a combination of effects on both the degree of electrophilicity at silicon (affecting
the rate constants for formation and reversion of the complex) and nucleophilicity at carbon (affecting the
partitioning of the complex between product and free reactants). Two 1-methyl-1-alkoxysilacyclobutanes were
also investigated, but proved to be inert to 193-nm photolysis.
Introduction
who employed simple competition methods to determine relative
rate constants for reaction of the transient 1,1-dimethyl-2,2-bis-
While there has been considerable experimental and theoreti-
cal interest over the past thirty years in the properties of the
silicon-carbon double bond, there have been relatively few
quantitative studies of the reactivity of simple silenes in either
(trimethylsilyl)silene (1) with a large number of oxygen,
1
-4
5
-7
8-12
the gas phase
or solution.
Perhaps the best known of
the early attempts to quantify silene reactivity is that of Wiberg,
nitrogen, and carbon nucleophiles in ether solution at low
temperatures.
1
3
†
Present address: Steacie Institute for Molecular Sciences, National
More recently, several groups have reported absolute rate
constants for silene trapping reactions in solution, using laser
flash photolysis techniques to generate transient silenes under
Research Council of Canada, 100 Sussex Drive, Ottawa, ON Canada K1A
0
R6‡ .
Present address: Institut f u¨ r Physikalische Chemie der Universit a¨ t Kiel,
Olshausenstrasse 40, 24098 Kiel, Germany.
8-12,14-21
conditions where they can be detected directly.
These
(
(
(
(
(
1) Raabe, G.; Michl, J. Chem. ReV. 1985, 85, 419.
2) Brook, A. G.; Baines, K. M. AdV. Organomet. Chem. 1986, 25, 1.
3) Grev, R. S. AdV. Organomet. Chem. 1991, 33, 125.
studies have been useful in establishing the more quantitative
aspects of silene reactivity and have provided detailed informa-
4) Brook, A. G.; Brook, M. A. AdV. Organomet. Chem. 1996, 39, 71.
5) Davidson, I. M. T.; Dean, C. E.; Lawrence, F. T. J. Chem. Soc.,
(13) Wiberg, N. J. Organomet. Chem. 1984, 273, 141.
(14) Shizuka, H.; Okazaki, K.; Tanaka, M.; Ishikawa, M.; Sumitani, M.;
Yoshihara, K. Chem. Phys. Lett. 1985, 113, 89.
(15) Gaspar, P. P.; Holten, D.; Konieczny, S.; Corey, J. Y. Acc. Chem.
Res. 1987, 20, 329.
(16) Sluggett, G. W.; Leigh, W. J. J. Am. Chem. Soc. 1992, 114, 1195.
(17) Zhang, S.; Conlin, R. T.; McGarry, P. F.; Scaiano, J. C. Organo-
metallics 1992, 11, 2317.
(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.
Chem. Commun. 1981, 52.
6) Davidson, M. T.; Wood, I. T. J. Chem. Soc., Chem. Commun. 1982,
50.
7) Vatsa, R. K.; Kumar, A.; Naik, P. D.; Upadhyaya, H. P.; Pavanaja,
U. B.; Saini, R. D.; Mittal, J. P.; Pola, J. Chem. Phys. Lett. 1996, 255, 129.
8) Conlin, R. T.; Zhang, S.; Namavari, M.; Bobbitt, K. L.; Fink, M. J.
Organometallics 1989, 8, 571.
9) Leigh, W. J.; Bradaric, C. J.; Sluggett, G. W. J. Am. Chem. Soc.
993, 115, 5332.
(
5
(
(
(
1
(
(
10) Bradaric, C. J.; Leigh, W. J. J. Am. Chem. Soc. 1996, 118, 8971.
11) Kerst, C.; Rogers, C. W.; Ruffolo, R.; Leigh, W. J. J. Am. Chem.
Soc. 1997, 119, 466.
12) Kerst, C.; Byloos, M.; Leigh, W. J. Can. J. Chem. 1997, 75, 975.
(21) Kerst, C.; Boukherroub, R.; Leigh, W. J. J. Photochem. Photobiol.,
A: 1997, 110, 243.
(
S0002-7863(98)01435-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/01/1998

Substituent Effects on ReactiVity of SidC Bonds
J. Am. Chem. Soc., Vol. 120, No. 37, 1998 9505
tion on the mechanisms of the reaction of silenes with
1
0,18,20,22
9,19,23,24
alcohols
and ketones,
two of the most widely
used reagent types for the trapping of reactive silenes in
solution.1
-4
Relatively few studies, however, have been directed at
developing an understanding of the effects of substituents at
2
2,24
either silicon or carbon on silene reactivity. Until recently,
the only systematic investigations of this type have been
2
5-27
theoretical.
The most comprehensive of these is that of
Apeloig and Karni, who employed ab initio (6-31G*//3-21G)
calculations to examine the geometries, charge distributions, and
frontier molecular orbital properties of a number of simple
silenes with substituents encompassing a wide range of donor/
with aliphatic alcohols as a gauge of their kinetic stabilities.
The substituents were chosen to span as wide a range in
inductive and resonance electronic properties as possible, within
the limitations defined by the ease of synthesis and hydrolytic
stability of the required silacyclobutane precursors.
25
acceptor strength at silicon and carbon. By considering the
structures of the few stable silenes that were known at the time
in light of the results of their calculations, they proposed that
the polarity of the SidC bond is the most important factor in
controlling its reactivity; π-electron acceptor and σ-electron
donor substituents at silicon and/or π-electron donors and
σ-electron acceptors at carbon were suggested to lead to
increased kinetic stability because these substituents act to
reduce both the degree of positive π-charge at silicon and the
degree of negative π-charge at carbon. Nagase and co-workers
came to similar conclusions on the basis of a theoretical
examination of the addition of water and HCl to H2SidCH2,
Me2SidCH2, F2SidCH2, and H2SidCHOH, using ab initio
calculations at a slightly higher level of theory.27
21
The first two members of the series, 1,1-dimethylsilene (3a)
2
4,33
and 1-methyl-1-phenylsilene (3b),
have been studied by us
previously by laser flash photolysis techniques, using the
corresponding silacyclobutanes (2a,b) as precursors. All of the
nine additional methylsilacyclobutanes chosen for study in the
present work can be expected to (and do) exhibit absorption
characteristics similar to 2a, necessitating the use of far UV
3
4
12,21
steady-state and time-resolved
photochemical techniques
for the study of their photochemistry and the UV absorption
spectra and reactivity of the resulting methylsilenes in hydro-
carbon solution. As will be shown below, all but the two
alkoxy-substituted derivatives (2j,k) yield the corresponding
silenes cleanly upon 193-nm photolysis in solution. The main
limitation of the far-UV flash photolysis technique arises from
the necessity for transient quenchers to be transparent at the
excitation wavelength, limiting the list to water, aliphatic
alcohols, or other nonabsorbing small molecules. Fortunately,
the addition of alcohols is mechanistically the best understood
of the many characteristic reactions of silicon-carbon double
bonds and thus provides an excellent quantitative tool for studies
of the reactivities of simple silenes in solution.
These ideas might be tested quantitatively through the use
of laser flash photolysis techniques to generate reactive silenes
in solution under conditions where they can be detected directly,
and absolute rate constants for their reactions with various silene
traps can be measured. Recently, we reported studies of the
effects of aryl substituents on the absolute rate constants and
Arrhenius parameters for reaction of 1,1-diphenylsilene with
alcohols, acetic acid and acetone in solution.22,24 These reactions
are all initiated by nucleophilic attack at silicon, with transfer
of a proton to the silenic carbon occurring in a subsequent
1
0,13,16,18,20,27-32
step.
Small positive Hammett F-values were
Results
observed, indicating that substitution with electron-withdrawing
groups on the phenyl rings attached to silicon increases
reactivity toward these reagents. This would appear to contra-
dict theoretical predictions, unless remote substituents at the
silicon end of the SidC bond in these compounds exert mainly
an inductive effect on the degree of positive charge character
The ultraviolet absorption spectra of 2c-k were recorded in
argon-saturated isooctane or hexane solutions. As expected,
the spectra of all but 2g,h show only edge absorption above
1
90 nm, while those of the vinyl- and ethynyl-substituted
compounds (2g,h) exhibit absorption maxima of 195 and 200
nm, respectively. Molar extinction coefficients at 193 nm,
determined from the slopes of plots of absorbance vs concentra-
2
5-27
at silicon.
With the intent of obtaining experimental data more directly
relevant to previous theoretical work in this area, we have
undertaken a study of the effects on silene reactivity of several
substituents directly attached to silicon, using a series of
-1
-1
tion, were found to vary between ∼500 and ∼2500 M cm ,
depending on the substituent.
Direct irradiation of deoxygenated, 0.01 M solutions of 2c-i
in argon-saturated pentane, hexane, or isooctane containing 0.1
M methanol (MeOH) with 185- or 214-nm light led to the
formation of ethylene and the corresponding methoxysilane (4;
eq 1) as the only detectable products (by GC analysis of the
crude photolysates) between 5 and 20% conversion of the
starting material. The methoxysilanes were identified in the
crude photolysates by GC/MS analysis and GC co-injections
with authentic samples. In contrast, the alkoxysilacyclobutanes
2j,k proved to be inert to 214-nm UV light after extensive
periods of photolysis under conditions similar to those employed
for 2c-i.
1
-substituted 1-methylsilacyclobutanes (2a-k; see eq 1) as
silene precursors and absolute rate constants for their reactions
(22) Bradaric, C. J.; Leigh, W. J. Can. J. Chem. 1997, 75, 1393.
(23) Toltl, N. P.; Leigh, W. J. Organometallics 1996, 15, 2554.
(24) Bradaric, C. J.; Leigh, W. J. Organometallics 1998, 17, 645.
(25) Apeloig, Y.; Karni, M. J. Am. Chem. Soc. 1984, 106, 6676.
(26) Trinquier, G.; Malrieu, J. P. J. Am. Chem. Soc. 1981, 103, 6313.
(27) Nagase, S.; Kudo, T.; Ito, K. In Applied Quantum Chemistry; Smith,
V. H. J., Schaefer, H. F., Morokuma, K., Eds.; D. Reidel: Dordrecht, 1986;
pp 249-267.
(
(
28) Nagase, S.; Kudo, T. J. Chem. Soc., Chem. Commun. 1983, 363.
29) Brook, A. G.; Safa, K. D.; Lickiss, P. D.; Baines, K. M. J. Am.
Chem. Soc. 1985, 107, 4338.
Laser flash photolysis experiments were carried out on
continuously flowing, deoxygenated solutions of the silacy-
(
(
30) Jones, P. R.; Bates, T. F. J. Am. Chem. Soc. 1987, 109, 913.
31) Steinmetz, M. G.; Udayakumar, B. S.; Gordon, M. S. Organome-
tallics 1989, 8, 530.
32) Kira, M.; Maruyama, T.; Sakurai, H. J. Am. Chem. Soc. 1991, 113,
986.
(33) Leigh, W. J.; Bradaric, C. J.; Boukherroub, R.; Cserti, C.; Schmeis-
ser, J. M., to be published.
(34) Leigh, W. J. Chem. ReV. 1993, 93, 487.
(
3

9506 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Leigh et al.
Figure 2. Plots of kdecay versus [ROH] for the quenching of 2-methyl-
1,3-(2-sila)butadiene (3g) by MeOH (O), EtOH (0), and t-BuOH (∆)
in deoxygenated isooctane solution at 23 °C.
Figure 1. Time-resolved UV absorption spectrum of 1-methyl-1-
trimethylsilylsilene (3i) in deoxygenated hexane solution at 23 °C, from
1
93-nm laser flash photolysis of silacyclobutane 2i. The inset shows a
decay trace recorded at the wavelength of maximum absorbance (285
nm). UV absorption spectra of silenes 3a-h are available as Supporting
Information.
the presence of alcohol throughout the concentrations ranges
studied, and the resulting plots of kdecay versus [ROH] were linear
in every case. Figure 2 shows typical plots of this type for
quenching of silene 3g by MeOH, EtOH, and t-BuOH.
Quenching by MeOD and t-BuOD was also studied in several
cases; it was consistently slower than that by the protiated
alcohols, but the plots of kdecay vs [ROD] were similarly linear
over the concentration ranges studied. Table 1 lists the absolute
rate constants determined in these experiments along with the
absorption maxima from the transient UV absorption spectra
of the silenes. The rate constant for reaction of 1,1-dimethyl-
silene (3a) with EtOH in isooctane at 23 °C was also determined
and is listed in Table 1 along with the previously reported rate
-3
-4
clobutanes (10 -10 M) in isooctane (2c,g) or hexane, using
the pulses (193 nm; 20-40 mJ; ∼20 ns) from an ArF excimer
laser. In all cases except those of 2j and 2k, readily detectable
transient absorptions were observed, which decayed over 1-10
µs with mixed pseudo-first- and -second-order kinetics at normal
laser intensities. The decay kinetics could usually be simplified
to pseudo-first-order by reducing the laser intensity with neutral
density filters. Time-resolved UV absorption spectra were
recorded in point-by-point fashion over several time windows
after the laser pulse. Figure 1 shows a representative transient
spectrum, recorded 0.5-1.0 µs after the pulse from a deoxy-
genated hexane solution of the trimethylsilyl derivative 2i, along
with a typical decay trace recorded at the wavelength of
maximum transient absorption. The transient spectra from
photolysis of all nine of the methylsilacyclobutanes 2a-i are
very similar in appearance, with only the absorption maximum
varying between 255 and 315 nm depending on the substituent.
In all cases, addition of aliphatic alcohols to the solutions
shortens the transient lifetime and results in a change to clean
pseudo-first-order decay kinetics. The common sensitivities of
these transients to alcohols and the similarities in their absorption
spectra are consistent with their assignments to the correspond-
ing silenes 3. Flash photolysis of the two alkoxysilacyclo-
butanes 2j,k led to the formation of very weakly absorbing
transient species which could not be reliably identified. These
results are consistent with the indication from the steady-state
experiments that these two derivatives are essentially photoinert.
The UV absorption spectra of these two compounds show only
edge absorption above 190 nm, like that of 2a (for example),
and offer no clue as to the reasons for this substituent effect on
silacyclobutane photoreactivity.
21
constants for reaction of this silene with MeOH and t-BuOH.
That for reaction of the least reactive silene (3i) with t-BuOH
was difficult to determine because of screening of the excitation
light by the alcohol at the high concentrations required for
significant quenching.
For compound 2c, a few steady-state competition experiments
were attempted in order to check the accuracy of the absolute
rate constants for reaction of 3c with alcohols. For example,
argon-saturated isooctane solutions of 2c (0.01 M) in the
presence of equimolar amounts of MeOH and EtOH (total
concentration ∼ 0.05 M) were irradiated (193 nm) to ca. 20%
conversion of silacyclobutane. The photolyses yielded mixtures
of methoxy- and ethoxydimethylsilane (4c), in the ratio
[EtOSiMe2H]/[MeOSiMe2H] ) 0.96 ( 0.02, in excellent
agreement with the ratio of absolute rate constants measured
by flash photolysis. Similar experiments using MeOH and
t-BuOH as the quenchers could not be carried out because the
t-butoxysilane coelutes with the solvent under our GC condi-
tions.
Absolute rate constants for the quenching of the transients
by methanol (MeOH), ethanol (EtOH), and t-butyl alcohol (t-
BuOH) were determined from plots of the pseudo-first-order
rate constant for silene decay (kdecay) versus ROH concentration
according to eq 2
Arrhenius parameters were estimated for the reaction of
methanol with 3c (R ) H), 3f (R ) CH2SiMe3), 3g (R ) CHd
CH2), 3h (R ) CtCH), and 3i (R ) SiMe3) in isooctane or
hexane solution, by determining decay rate constants at several
temperatures between 2 and 54 °C, for deoxygenated solutions
of the silenes containing enough MeOH to reduce the lifetime
of the silene to less than 10-15% of its value in the absence of
the alcohol at room temperature. Thus, the kMeOH values can
be approximated directly from the decay rate constants at each
temperature (i.e., kMeOH ≈ kdecay/[MeOH]). It should be noted,
however, that these values have not been corrected for decay
processes due to other than methanol quenching, so the
kdecay ) k + kROH[ROH]
(2)
o
in which ko is the pseudo-first-order rate constant for decay in
the absence of quencher and kROH is the bimolecular rate
constant for quenching by the added alcohol. The silene
absorptions decayed with clean pseudo-first-order kinetics in

Substituent Effects on ReactiVity of SidC Bonds
J. Am. Chem. Soc., Vol. 120, No. 37, 1998 9507
Table 1. UV Absorption Maxima and Bimolecular Rate Constants for Reaction of Substituted 1-Methylsilenes (3a-i) with MeOH(D), EtOH,
and t-BuOH(D) in Deoxygenated Isooctane (3a,c,g) or Hexane (3b,d-f,h,i) Solution at 23 °Ca
MeOH
t-BuOH
/10 M s-1
9
-1
k
EtOH k
/10 M s-1
9
-1
k
/10 M s-1
9
-1
R, R′
UV λmax (nm)
k
q
H
/k
D
q
q
H D
k /k
Me, Me (3a)b
Me, C H (3b)
6 5
Me, H (3c)
Me,Et (3d)
Me, t-Bu (3e)
255
315
265
255
260
260
305
290
285
4.9 ( 0.2
3.2 ( 0.2
1.5 ( 0.1
1.4 ( 0.1
1.3 ( 0.1
4.5 ( 0.2
1.8 ( 0.1
1.2 ( 0.1
1.6 ( 0.2
1.1 ( 0.1
1.2 ( 0.1
1.8 ( 0.2
c
4.41 ( 0.14
4.2 ( 0.5
4.4 ( 0.2
3.6 ( 0.3
2.4 ( 0.2
2.57 ( 0.12
1.3 ( 0.1
3.7 ( 0.1
0.33 ( 0.03
0.71 ( 0.11
0.81 ( 0.04
5.4 ( 0.3
Me, CH
Me, CHdCH
Me, CtCH (3h)
2
SiMe
3
(3f)
1.86 ( 0.09
1.75 ( 0.10
10.0 ( 0.1
0.18 ( 0.01
1.80 ( 0.05
1.86 ( 0.08
9.8 ( 0.2
2
(3g)
1.6 ( 0.2
1.1 ( 0.1
1.6 ( 0.2
Me, SiMe
3
(3i)
0.19 ( 0.01
0.03 ( 0.01
a
Errors are listed as twice the standard deviation from least-squares analysis of kdecay-concentration data according to eq 2. b Data from ref 21,
c
except for that for EtOH. Data from ref 33.
Discussion
The utility of silacyclobutane derivatives as photochemical
or thermal precursors of reactive silenes is well estab-
1
,4,36,37
lished,
but only a few of the silacyclobutane derivatives
2
1,38
24
7
39,40
studied in this work (2a,
2b, 2c, and 2g
) have been
previously reported to yield products consistent with the
intermediacy of the corresponding silenes 3 upon direct pho-
tolysis in solution or the gas phase. Indeed, photolysis of all
nine compounds in the presence of methanol or t-butyl alcohol
leads to the alkoxysilane expected from trapping of the
corresponding silene by the alcohol. It is particularly important
to note that the photolyses are all clean at conversions of <20%,
producing only ethylene and the corresponding alkoxysilane
within the limits of our GC detection method. There is nothing
to indicate that any reactive intermediates other than the silenes
Figure 3. Arrhenius plots for the quenching of substituted 1-methyl-
silenes 3c (R ) H), 3a (R ) Me), 3f (R ) CH SiMe ), 3g (R ) CHd
CH ), 3h (R ) CtCH) and 3i (R ) SiMe ) by methanol in
2
3
3
a-i are produced upon photolysis of any of the corresponding
2
3
silacyclobutanes in solution, on the basis of these experiments.
deoxygenated hexane or isooctane solution. The dotted line shows the
temperature dependence of the rate constant for the diffusion of hexane,
calculated using the modified Debye equation (kdiff ) 8RT/3000η) and
published viscosities.3
Like that of 2a21 and 2b,24,33 laser flash photolysis of 2c-i
gives rise in each case to what appears to be a single transient
species, judging from the appearance of the spectra and the fact
that the decay kinetics of each transient are the same throughout
the observed transient absorption band both in the absence and
presence of alcohols. The transient spectra obtained from laser
photolysis of all nine of these silacyclobutanes are characterized
by a single long wavelength absorption band, which varies only
in position as a function of substituent. This, along with the
fact that they decay with second-order kinetics at high laser
intensities (indicative of dimerization) and exhibit high reactivity
toward alcohols, allows the transients to be assigned to the
corresponding silenes with a high level of certainty. The
transient absorption spectrum observed from flash photolysis
of 2c in isooctane solution agrees well with that reported from
gas-phase flash photolysis of this compound, and which was
5
Table 2. Arrhenius Parameters for Reaction of Representative
a
Methylsilenes with MeOH in Hydrocarbon Solution
log(A/M-1 s-1
)
E
a
(kcal/mol)
H (3c)
Me (3a)
-1.9 ( 0.3
-2.6 ( 0.3
-2.0 ( 0.5
-2.4 ( 0.1
-0.8 ( 0.1
-3.6 ( 0.6
8.3 ( 0.3
7.7 ( 0.3
7.8 ( 0.6
7.5 ( 0.3
9.3 ( 0.2
5.5 ( 0.5
CH
CHdCH
CtCH (3h)
SiMe (3i)
2
SiMe
3
(3f)
(3g)
2
3
a
Estimates based on the determination of the silene decay rate
constant as a function of temperature between 2-55 °C, using solutions
of the corresponding silacyclobutane in hexane or isooctane (3a,c,g)
containing enough methanol to ensure that >85% of the decay is due
to the bimolecular reaction with the alcohol. Errors are reported as twice
the standard deviation from least-squares analysis of the data shown
in Figure 3.
7
assigned to 1-methylsilene (3c). The solution spectrum of 3c
also agrees well with the reported low-temperature matrix
spectrum, for which the silene was generated by flash vacuum
pyrolysis of a silabicyclo[2.2.2]octadiene precursor.41 As es-
pected, the UV absorption spectra of the ethyl and t-butyl
derivatives (3d,e) are more or less indistinguishable from that
of 3a. There is a slight red-shift in the absorption maximum
of 1-methyl-1-trimethylsilylmethylsilene (3f; λmax ) 260 nm)
corresponding Arrhenius plots (Figure 3) will be slightly
distorted from their true form. The approximate Arrhenius
parameters calculated from these data are listed in Table 2. Table
2
and Figure 3 also contain the previously reported Arrhenius
2
1
(36) Jutzi, P.; Langer, P. J. Organomet. Chem. 1980, 202, 401.
data for reaction of 3a with methanol in isooctane.
The
(
37) Steinmetz, M. G. Chem. ReV. 1995, 95, 1527.
temperature dependence of the diffusional rate constant in
hexane, calculated from the modified Debye equation and
published viscosities,35 is also included in Figure 3.
(38) Brix, T.; Arthur, N. L.; Potzinger, P. J. Phys. Chem. 1989, 93, 8193.
(39) Bertrand, G.; Manuel, G.; Mazerolles, P. Tetrahedron Lett. 1978,
2149.
40) Bertrand, G.; Manuel, G.; Mazerolles, P.; Trinquier, G. Tetrahedron
(
(
35) CRC Handbook of Chemistry and Physics; CRC Press: Boca Raton,
1981, 37, 2875.
1
995; pp 6-241.
(41) Maier, G.; Mihm, G.; Reisenauer, H. P. Chem. Ber. 1984, 117, 2351.

9508 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Leigh et al.
compared to that of 3a (λmax ) 255 nm), indicative of a small
hyperconjugative effect of the â-SiMe3 group on the silenic π,π*
absorption. The effect is smaller than when the -CH2SiMe3
group is attached to the silenic carbon, which results in a ∼15-
nm red-shift of the long-wavelength shoulder absorption in the
UV spectrum of 1,1-dimethyl-2-trimethylsilylmethylsilene com-
pared to that of 1,1,2-trimethylsilene.42 The absorption maxima
of the vinyl (3g), ethynyl (3h), phenyl (3b), and trimethylsilyl
(
2
3i) derivatives are all red-shifted significantly from the 255-
60 nm absorption maxima of the dialkylsilenes, and they follow
a similar trend as a function of subsituent as in the spectra of
the corresponding alkene analogues.43
All of the transient silenes studied here exhibit the expected
high degree of reactivity toward aliphatic alcohols, with relative
reactivities following the order kMeOH ≈ kEtOH > kt-BuOH. For
3
h, the most reactive silene yet investigated in solution, the
bimolecular rate constant for reaction with methanol is only
about a factor of 2 lower than the diffusion rate in hexane at 23
°
C. The other silenes are less reactive than this, with the rate
constants varying by about a factor of 50 throughout the series.
The span in the rate constants is somewhat greater for t-BuOH,
reflecting its lower reactivity compared to MeOH.
The addition of aliphatic alcohols to transient silenes is
thought to proceed by a mechanism initiated by reversible
nucleophilic attack at silicon to form a silene-alcohol complex,
which collapses to the product by rate-controlling unimolecular
proton transfer from oxygen to carbon (see eq 3).1
0,12,16,18,20-22,32
The latter can also be catalyzed by a second molecule of
1
1,16,18,20,32
alcohol,
but the linear dependence of kdecay on alcohol
Figure 4. Substituent parameter correlations for the quenching of
concentration observed for the silenes studied here indicates that
this is unimportant in these cases, at least over the small alcohol
concentration ranges employed for our kinetic experiments. In
general, reversion of the complex to the free reactants is faster
than proton transfer, giving rise to small primary deuterium
substituted 1-methylsilenes 3a-i by methanol in isooctane or hexane
0
solution at 23 °C: (a) log kMeOH vs (-3.0 σ
R
I
+ 3.6 σ ); (b) log kMeOH
0
vs (-3.6 σ
R
+ 3.1σ
I
s
+ 0.21E ). The solid line represents the fit of the
data to the specified functions (r ) 0.659 and 0.965 for (a) and (b),
2
respectively).
12,16,18,20-22
isotope effects
and negative Arrhenius activation
for the overall rate constants for reaction. With
silicon, increasing in the order SiMe3 , CHdCH2, CH2SiMe3
1
0,12,21,22
energies
<
Ph < H, Me, Et, t-Bu < CtCH. It became evident rather
the present silenes, the rate constants for reaction with MeOH
and t-BuOH exhibit deuterium isotope effects in the range 1.1-
.8, which can, in most cases, be clearly identified as a primary
early in this study that the rate constants for these reactions do
not correlate in a straightforward way with either inductive or
1
o
resonance substituent parameters alone (σI and σR ,
respectively ), and thus to allow for the possibility of a
kinetic isotope effect. Furthermore, the rate constants for
reaction with MeOH all follow similar (negative) temperature
dependences, with a trend toward increasingly negative Ea and
a smaller preexponential factor with decreasing reactivity
throughout the series (see Table 2). Thus, all aspects of their
reactivity toward aliphatic alcohols are consistent with the
mechanism shown in eq 3. With the application of the steady
state approximation for the complex, the experimentally mea-
sured second-order rate constant for the reaction can be
expressed in terms of the individual rate constants for formation
and collapse of the complex, as shown in eq 4.
4
4a
4
5
multiparameter analysis, the series was expanded to include
as wide a range in substituent electronic and steric properties
as could be accommodated by the conditions necessary for the
kinetic measurements. The poor fit of the data to a single
substituent parameter is evident, for example, from the similar
reactivities of the 1-methyl (3a; σI ) 0.10)- and 1-phenyl (3b;
σI ) -0.10)-substituted derivatives, and from the widely
o
different reactivities of the 1-trimethylsilyl (3i; σR ) +0.06)-
o
and 1-ethynyl (3h; σR ) +0.08)-substituted derivatives. Fitting
the data to a two-parameter function incorporating both reso-
4
5
nance and inductive parameters leads to the relatively poor
correlation shown in Figure 4a (r ) 0.659). A significant
improvement is obtained by including the steric substituent
k k
1
2
2
kROH
)
(4)
k-1 + k2
The rate constants for reaction of 3a-i with MeOH and EtOH
(44) (a) Wayner, D. D. M. In CRC Handbook of organic photochemistry;
Scaiano, J. C. Ed.; CRC Press: Boca Raton, 1989; pp 395-400. (b) Unger,
S. H.; Hansch, C., Prog. Phys. Org. Chem. 1976, 12, 91. The Es value
employed for the trimethylsilylmethyl substituent (Es ≈ -3.56) was
approximated from the value for SiMe3, corrected by the difference between
the Es values for CMe3 and CH2CMe3.
vary regularly as a function of the single variable substituent at
(
42) Leigh, W. J.; Bradaric, C. J.; Sluggett, G. W.; Venneri, P. C.; Conlin,
R. T.; Dhurjati, M. S. K.; Ezhova, M. B. J. Organomet. Chem. 1998, 561,
9.
1
(
43) Robin, M. B. Higher excited states of polyatomic molecules, Vol.
(45) Ehrenson, S.; Brownlee, R. T. C.; Taft, R. W. Prog. Phys. Org.
Chem. 1973, 10, 1.
II; Academic Press: New York, 1975; pp 1-68.

Substituent Effects on ReactiVity of SidC Bonds
J. Am. Chem. Soc., Vol. 120, No. 37, 1998 9509
parameter Es,44b as shown in Figure 4b. The least-squares fit
to the nine data points (r ) 0.965) affords the coefficients, FR
of the trimethylsilyl derivative 3i, resulting in a substantial
decrease in reactivity compared to the similarly sized t-butyl
and trimethylsilylmethyl derivatives 3e and 3f, respectively.
Interestingly, the vinyl substituent in 3g is approximately
electronically neutral, suggesting that its reduced reactivity
compared to that of 3c is almost entirely due to steric effects.
Finally, our data suggest that phenyl substitution at silicon
produces roughly twice the electronic destabilization that
methyl-substitution does. This is compensated for by steric
effects, which are 2-3 times greater for phenyl than for methyl
2
)
-3.6 ( 1.2, FI ) 3.1 ( 1.0, and Fs ) 0.21 ( 0.08, where
the quoted errors represent the 95% confidence limits of the
analysis. The results of this fit must obviously be interpreted
qualitatively and with a good deal of caution, not just because
of the use of three independent variables in the analysis, but
also because the intrinsic reactivity of the ethynyl derivative
relative to the others might very well be underestimated because
it reacts with MeOH with a rate constant within a factor of ca.
4
4b
2
of the diffusional rate. Nevertheless, the negative FR and
substitution; the result is a modest reduction in reactivity for
1-methyl-1-phenyl- (3b) compared to 1,1-dimethylsilene (3a).
This trend is continued with 1,1-diphenylsilene, which exhibits
slightly lower reactivity than 3b with alcohols under similar
positive FI values suggest that the intrinsic reactivity of the
SidC double bond in these compounds is enhanced by
π-electron donor and/or σ-electron acceptor substituents at
silicon, respectively, and the positive Fs value indicates that it
is reduced slightly (but significantly) by increasing the steric
bulk of the one variable substituent in this series of compounds.
Clearly, the steric effect would be expected to contribute more
significantly if the common substituent in the homologous series
of silenes was larger than methyl or if both substituents at silicon
were varied.
3
3
conditions. Wiberg and Link have recently reached a similar
conclusion from a comparison of the reactivities of the transient
silenes Ph2SidC(SiMe3)2 and Me2SidC(SiMe3)2 toward a wide
4
6
range of silene trapping agents.
From an analysis of the effects of substituents on the ab initio
SidC bond lengths and charge distributions in silenes of the
type H(R)SidCH2, Apeloig and Karni hypothesized that the
electronic effects of substituents on the reactivity of silenes
should result largely from their ability to reduce or enhance
Similar F values are obtained from a fit of the EtOH and
t-BuOH data to the three substituent parameters, FR ) -3.0 (
25
2
the natural polarity of the SidC bond. Thus, π-electron donor
and σ-electron acceptor substituents at silicon were predicted
to lead to enhanced reactivity toward both electrophiles and
nucleophiles because such substituents simultaneously increase
both the negative charge density at carbon and the positive
charge density at silicon. Exactly the opposite effects were
predicted for substituents of the same type at the silenic carbon.
With steric effects worked in, this rationale provided a very
satisfying explanation for the remarkable stabilities of the few
1
.7, FI ) 2.5 ( 1.6, and Fs ) 0.24 ( 0.14 (r ) 0.947) for
EtOH, and FR ) -3.3 ( 3.7, FI ) 3.3 ( 3.1, and Fs ) 0.28 (
.24 (r ) 0.789) for t-BuOH, where again the errors are listed
2
0
as the 95% confidence limits (the plots for these two alcohols
are available as Supporting Information). The poor fit for the
t-BuOH data appears to be mostly due to the t-butyl silene 3e,
whose reactivity shows a much greater sensitivity to alcohol
structure than is observed with the other silenes (see Table 1).
We can think of no ready mechanistic explanation for this, short
of the possibility that the steric effect of the t-butyl group might
be under-represented, compared to the other substituents, by
the Es substituent parameter. There is nothing in the product
studies, the silene decay kinetics in the presence of alcohols,
or the kinetic isotope effects to suggest a difference in the
mechanism for the addition of alcohols to this silene compared
to that of the others.
“
persistent” silenes that were known at the time. The kinetic
results reported here for the addition of alcohols to the series
of transient 1-methylsilenes are, for the most part, consistent
with these ideas. In this reaction, like many others involving
1
2,20,22,24,33
addition of oxygen- and nitrogen-nucleophiles,
both
the nucleophilic and electrophilic stages of the reaction are
kinetically significant.
It is difficult to provide a more detailed analysis of the
substituent effects on silene reactivity toward alcohols, in terms
of the primary rate constant for formation of the complex (k1)
and the partitioning ratio for its reversion to free reactants and
collapse to alkoxysilane (k2/(k-1 + k2)), the product of which
yields the overall rate constant for reaction (see eq 4). For the
ethynyl derivative 3h, the magnitude of kMeOH and the form of
its temperature dependence indicate that k1 is approximately
equal to the diffusional rate constant in hexane, and that the
partitioning ratio is very close to (but slightly less than) 0.5.
The latter must be true in order for the Arrhenius activation
energy to be negative according to this mechanism; a dominant
entropic component to the free energy barrier for intracomplex
proton transfer is also required. The trends in the temperature
dependences for MeOH addition (toward progressively more
negative Ea’s and preexponential factors; see Figure 3) and the
kinetic isotope effects are consistent with reductions in both k1
and k2/(k-1+k2) as overall reactivity decreases throughout the
series. The effect on k1 can be ascribed to decreasing electro-
philicity at silicon with decreasing π-electron donor/σ-electron
acceptor power and/or increasing steric bulk of the substituents,
while that on the partitioning ratio can be ascribed to increasing
entropic demands in the intracomplex proton transfer step with
decreasing nucleophilicity at carbon and increasing steric effects
The relative importance of steric and electronic effects on
the kinetic stabilities of the silenes in the series can be
understood more clearly by considering the correlations shown
in Figure 4 in more detail. The FR and FI values suggest that
resonance and inductive factors oppose one another but are of
similar importance in affecting the kinetic stability of the Sid
C bond in these compounds. Calculation of the net electronic
effect of the R substituent for each member of the series (as
o
FRσR + FIσI, where the F values are those obtained from the
least-squares fit of the data for methanol addition), suggests that
the trimethylsilyl substituent (a weak π-electron acceptor and
weak σ-electron donor) stabilizes the silene relative to RdH,
while all of the other substituents destabilize. Because reso-
nance and inductive effects are of similar importance, the net
electronic effects of the phenyl (a moderate π-electron donor
and σ-electron acceptor), t-butyl (a strong hyperconjugative
π-electron donor and very weak σ-electron acceptor), and
ethynyl (a moderate π-electron acceptor and strong σ-electron
acceptor) substituents are roughly equally destabilizing relative
to RdH (3c). The spread in reactivity for these three com-
pounds thus appears to be due primarily to steric effects, which
are minimal in the ethynyl derivative (resulting in nearly
diffusion-controlled reactivity; see Figure 3), but roughly cancel
the electronic destabilization in the phenyl and t-butyl deriva-
tives. Steric and electronic effects are cooperative in the case
(46) Wiberg, N.; Link, M. Chem. Ber. 1995, 128, 1231.

9510 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Leigh et al.
at silicon. We are presently investigating these details of the
mechanism of addition of alcohols to silenes and the effects of
substituents on the process, using computational methods.
corded on a Perkin-Elmer Lambda 9 spectrometer. When
absorbance data below 200 nm were desired, the compartment
and sample solution were flushed with argon or nitrogen prior
to measuring the spectra. Low-resolution mass spectra were
determined by GC/MS, using a Hewlett-Packard 5890II gas
chromatograph equipped with a HP-5971A mass selective
detector and a DB-5 fused silica capillary column (30 m × 0.25
mm; Chromatographic Specialties, Inc). Infrared spectra were
recorded as thin films on a BioRad FTS-40 FTIR spectrometer
Summary and Conclusions
The 193-nm photolysis of substituted alkylsilacyclobutanes
provides a convenient means of generating the corresponding
silenes in hydrocarbon solution, under conditions where they
can be detected directly and where absolute rate constants for
their reactions with aliphatic alcohols can be measured. Pho-
tocycloreversion proceeds cleanly in alkylsilacyclobutanes bear-
ing substituents such as hydrido, alkyl, vinyl, ethynyl, trime-
thylsilyl, and phenyl, but alkoxysilacyclobutanes appear to be
inert to photolysis with 193-214-nm light. Although it has
-
1
and are reported in wavenumbers (cm ).
Analytical gas chromatographic analyses were carried out
using a Hewlett-Packard 5890II+ gas chromatograph equipped
with conventional heated splitless injector, a flame ionization
detector, a Hewlett-Packard 3396A integrator, and DB1 or
DB1701 megabore capillary columns (15m × 0.53 mm;
Chromatographic Specialties, Inc.). Semipreparative gas chro-
matographic separations were carried out using a Varian 3300
gas chromatograph equipped with a thermal conductivity
detector and a stainless steel column (3% OV 101 on Chro-
mosorb W, HP 80/100; 6’ x 0.25”; Chromatographic Specialties,
Inc.).
been explicitly proven only for 3a2
1,47
20
and 3g, cycloreversion
is thought to involve the lowest excited singlet state, in which
the ring Si-C bonds are weakened through population of a σ*
4
7a
molecular orbital.
The absolute rate constants for the reaction of aliphatic
alcohols with the corresponding silenes in hexane solution vary
by about 2 orders of magnitude depending on the alcohol, with
2
,2,4-Trimethylpentane (isooctane; BDH Omnisolv) and
1
1
-ethynyl-1-methylsilene (3h) reacting the fastest and 1-methyl-
-trimethylsilylsilene (3i) the slowest. The former reacts with
hexane (Caledon HPLC grade) exhibited absorbances of <0.2
at 193 nm in a 3-mm cell and were used without further
purification. Methanol, ethanol, and t-butyl alcohol were of
the highest purity available from Aldrich Chemical Co. and were
dried over activated 4 Å molecular sieves prior to use.
Methanol-Od, ethanol-Od, and t-butyl alcohol-Od were used
as received from Aldrich.
methanol at very close to the diffusion-controlled rate and is
the most reactive silene studied to date. Methanol addition
proceeds with a negative Arrhenius activation energy in all cases
examined and exhibits a small primary deuterium kinetic isotope
effect in all cases but that of the ethynyl-substituted derivative.
These characteristics indicate that the silenes react by a common
mechanism, involving reversible formation of a silene-alcohol
complex which collapses to alkoxysilane by rate-limiting,
entropy-controlled proton transfer from oxygen to carbon.
The rate constants for the reaction of nine silenes of the type
R(Me)SidCH2 with methanol, ethanol, and t-butyl alcohol
correlate with a three parameter function incorporating reso-
nance, inductive, and steric substituent parameters. The signs
and magnitudes of the coefficients indicate that resonance and
inductive effects act in opposing but roughly equal fashion on
the kinetic stabilities of transient 1-methylsilenes toward addition
of these reagents. Silene reactivity is reduced by π-electron
acceptor and/or σ-electron donor substituents at silicon, most
likely through an effect on the degree of electrophilicity at
silicon, which affects the complexation step, and on the degree
of nucleophilicity at carbon, which affects the proton-transfer
step. Steric effects can act to reinforce, cancel, or even reverse
the electronic effects of substituents at silicon on the intrinsic
kinetic stability of the SidC bond. These conclusions are in
good agreement with those made several years ago on the basis
of ab initio theoretical calculations of the structures, Mulliken
1
,1-Dimethylsilacyclobutane (2a) was available from a previ-
21
ous study, while the remaining silacyclobutanes were prepared
by methods similar to those published. They exhibited spec-
troscopic data and boiling points similar to those reported, 2c,
48
49
50
6
2
(
5 °C (760 mm); 2d, 72 °C (110 mm); 2f, 95 °C (97 mm);
4
0
51
g, 46 °C (110 mm); 2i, 68-69 °C (72 mm); 2k, 38-40 °C
5
2
52
760 mm); 2l, 61-62 °C (60 mm). 1-Ethynyl-1-methylsi-
-
1
lacyclobutane (2h, λmax 200 nm, ꢀ193 nm ) 2583 ( 300 M
-
1
cm ) codistilled with THF from its reaction mixture and was
thus separated by silica gel column chromatography using
pentane as eluant and identified by comparison to reported data
5
3
for this compound. After being distilled the silacyclobutanes
were isolated in >99% purity (as estimated by capillary GC
analysis) by semipreparative gas chromatography.
5
4
1
-Methyl-t-butylsilacyclobutane (2e). To a solution of
1
-chloro-1-methylsilacyclobutane (1.5 g, 0.0124 mol) in dry
diethyl ether (40 mL) cooled at -78 °C was added a 1.6 M
pentane solution of tert-butyllithium (8.0 mL, 0.0128 mol) with
stirring. A white precipitate formed immediately. The reaction
mixture was stirred for 14 h while the dry ice/acetone bath
warmed to room temperature. The reaction was carefully
quenched with water (20 mL). The aqueous layer was separated
and extracted with pentane (3 × 10 mL). The organic fractions
were combined, dried over anhydrous magnesium sulfate,
filtered, and the solvent was removed on the rotary evaporator
charge distributions, and frontier molecular orbital properties
of a series of simple 1- and 2-substituted silenes.25
For reactions initiated by nucleophilic attack at silicon, steric
effects should be less important when the substituents at carbon
are varied. A study of this type requires a more extensive list
of silene precursors than can be provided by silacyclobutane
derivatives and is in progress in our laboratory.
(48) Auner, N.; Binnewies, M.; Grobe, J. J. Organomet. Chem. 1984,
277, 311.
(49) Vdovin, V. M.; Nametkin, N. S.; Grinberg, P. L. Dokl. Akad. Nauk
SSSR 1963, 150, 799.
Experimental Section
(
50) Barton, T. J.; Burns, S. A.; Burns, G. T. Organometallics 1982, 1,
10.
(51) Barton, T. J.; Burns, G. T.; Gschneidner, D. Organometallics 1983,
, 8.
52) Dubac, J.; Mazerolles, P.; Lesbre, M.; Joly, M. J. Organomet. Chem.
1970, 25, 367.
2
NMR spectra were recorded in deuteriochloroform and are
referenced to tetramethylsilane. 1H and 13C NMR spectra were
recorded on a Bruker AC200 spectrometer and are referenced
to tetramethylsilane. Ultraviolet absorption spectra were re-
2
(
(53) Voronkov, M. G.; Yarosh, O. G.; Roman, V. K.; Albanov, A. I.
(
47) (a) Steinmetz, M. G.; Bai, H. Organometallics 1989, 8, 1112. (b)
Metalloorg. Khim. 1990, 3, 1423.
(54) Auner, N. J. prakt. Chem./Chem.-Ztg. 1995, 337, 79.
Dube, G.; Gey, E.; Koehler, P. Z. Anorg. Allg. Chem. 1974, 405, 46.

Substituent Effects on ReactiVity of SidC Bonds
J. Am. Chem. Soc., Vol. 120, No. 37, 1998 9511
to yield a colorless oil which was distilled under reduced
pressure to yield the product as a colorless liquid (bp 72-73
photolyzates; in the cases of the methoxysilanes, these were
verified by GC- and GC/MS-co-injection with the authentic
samples. The formation of ethylene as a coproduct from
photolysis of 2a,c was evident in experiments carried out in
isooctane solution but was not verified in the cases of 2d-h.
Competition experiments with 2c were carried out in argon-
saturated isooctane solutions containing 0.01 M 2c, 0.025 M
methanol, and 0.025 M ethanol. Solutions were contained in
Suprasil cuvettes (10 mm × 10 mm) and stirred vigorously with
a small magnetic stirrer during irradiation with an ArF excimer
laser (193 nm; ∼20 mJ; 5-Hz repetition rate) to a maximum of
20% conversion of the silacyclobutane. Product ratios were
determined from the relative peak areas without correction for
FID response and are the average of triplicate determinations.
Nanosecond laser flash photolysis experiments employed the
pulses from Lumonics 510 or Lambda Physik Compex 120
excimer lasers, filled with Ar/F2/He (193 nm, 12-25 ns, 30-
°
2
C, 98 mmHg; 1.1 g, 0.0077 mol, 62%). It was identified as
e on the basis of the following spectroscopic data: H NMR,
1
δ ) 0.19 (s, 3H), 0.93 (s, 9H), 0.8-1.2 (m, 4H), 2.01 (tt, 2H);
1
3
-1
C NMR, δ ) -4.5, 11.4, 17.6, 25.8, 32.0; IR (neat, cm ),
2
953, 2857, 1468, 1362, 1250, 1119, 1050, 865, 766; UV, λmax
-
1
-1
<
190 nm, ꢀ193 nm ) 2303 ( 200 M cm ; MS m/z ) 143
(
4), 142 (27), 114 (28), 100 (39), 99 (90), 86 (44), 85 (100), 73
63), 72 (58), 59 (93), 58 (53).
(
Authentic samples of the methoxysilanes 4 were either
obtained from commercial sources (4a), or prepared by treating
the appropriate chlorosilane with the appropriate alcohol and
4
0,55-57
triethylamine in anhydrous diethyl or dibutyl ether,
or
from dimethoxydimethylsilane and the appropriate Grignard
reagent. They exhibited spectroscopic data and/or boiling points
57
similar to those reported, 4b, 75-80 °C (9 mm); 4c, 31 °C
5
9
47
12,62
(
(
760 mm); 4d (not distilled; identified by ms ); 4e, 58 °C
70 mJ), and a microcomputer-controlled detection system.
40
110 mm); 4g, 68-70 °C (760 mm); 4h, 78-80 °C (760
mm); 4i, 50 °C (85 mm).
Methoxydimethyl(trimethylsilylmethyl)silane (4f) was pre-
The intensity of the beam was reduced to 2-6 mJ at the cell
using a series of stainless steel wire meshes as neutral density
filters. Solutions were prepared at concentrations such that the
5
8
60
50
-4
absorbance at the excitation wavelength was ca. 0.7 (∼6 × 10
pared by treating dimethyl(trimethylsilylmethyl)silane (1.0 g,
-
3
to 4 × 10 M). They were flowed continuously from a
calibrated 100 mL reservoir through a 3 × 7 mm Suprasil flow
cell, which was contained in a brass sample holder whose
temperature was controlled to within 0.1 °C by a VWR 1166
constant-temperature circulating bath. Solution temperatures
were measured with a Teflon-coated copper/constantan ther-
mocouple which was inserted directly into the flow cell.
Solutions of the appropriate silacyclobutane were deoxygenated
with a stream of dry nitrogen in the reservoir for 30-60 min
prior to flash photolysis experiments. A slightly different
procedure was used for 2a,c, to prevent sample loss due to
evaporation. In these two cases, the reservoir was filled with
solvent, sealed with a rubber septum bearing a syringe needle
vent, and deoxygenated with a stream of dry nitrogen for 1 h.
The nitrogen supply was turned off, and the appropriate amount
of silacyclobutane was then added by a microliter syringe. The
nitrogen supply was either kept off or opened only slightly
during flash photolysis experiments to avoid loss of substrate
or quencher through evaporation. Quenchers were added
directly to the reservoir by a microliter syringe as aliquots of
standard solutions. Rate constants were calculated by linear
least-squares analysis of decay rate concentration data (6-10
points) which spanned at least a factor of 5 (usually more than
6.84 mmol) with methanol (0.28 mL, 6.84 mmol) in the presence
of a catalytic quantity (ca. 5 mg) of palladium(II) chloride. The
mixture was stirred at room temperature for 15 min until the
vigorous evolution of hydrogen ceased. The resulting liquid
was removed from insoluble palladium salts with a filtering
pipet, and excess methanol was removed on the rotary evapora-
tor to yield the product as a colorless liquid (1.0 g, 6.2 mmol,
6
1
9
1%), whose spectroscopic data matched those reported.
Dimethyl(trimethylsilylmethyl)silane was prepared from chlo-
rodimethylsilane and trimethylsilylmethylmagnesium chloride
o
in anhydrous ether. The product (bp 88 C, 760 mm) was
1
identified on the basis of the following spectroscopic data: H
NMR, δ ) 0.25 (d, 2H), 0.02 (s, 9H), 0.08 (d, 6H), 3.95 (m,
H); 1 C NMR, δ ) -1.5, 0.8, 1.5; IR (neat) 2956 (s), 2901
3
1
(
m), 2112 (s), 1425 (w), 1253 (s), 1048 (s), 896 (s), 837 (s);
MS m/z ) 146 (0.1), 145 (3), 133 (8), 132 (16), 131 (100), 73
(43), 59 (13).
Analytical-scale photolyses were carried out using a low-
pressure mercury lamp (185 + 254 nm; Osram HNS 10W/UOZ)
for 2a,c or a zinc resonance lamp (214 nm; Philips 93106E)
for 2d-k. The photolyses were carried out in pentane, hexane,
or isooctane, depending on the alcohol and silacyclobutane, to
allow baseline separation of the alkoxysilane products from the
solvent; thus, pentane was used for all irradiations using t-butyl
alcohol as the silene trap and for irradiation of 2h, while
isooctane was employed as solvent for irradiations with the
lighter alcohols. Photolyses of 2d-g were carried out in hexane
solution. Aliquots of the solutions of silacyclobutane (0.01 M)
and an alcohol (0.1 M) were placed in Suprasil quartz cuvettes
1
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.
Lifetimes were determined as a function of temperature
between ∼0 and 55 °C for silenes 3c,f-i, in deoxygenated
hexane solutions containing an appropriate concentration of
methanol to reduce the lifetime to 10-15% of its value in the
pure, deoxygenated solvent. The solutions were deoxygenated
for 1-2 h prior to the addition of alcohol, and then the gas
flow was reduced to a trickle in order to avoid evaporation of
the quencher during the 3-4-h period required for the experi-
ment. Quenching rate constants (kMeOH) were estimated at each
temperature as kdecay/[MeOH], without applying corrections for
thermal expansion of the solvent. For 3c, Arrhenius parameters
were also determined from individual absolute quenching rate
constants at several temperatures over the same range and were
found to be identical to the estimated values within the quoted
error limits ((2σ).
(
3 mm × 10 mm), sealed with rubber septa, and saturated with
dry argon. Photolysis of the solutions, with monitoring by GC
between 5 and 20% conversion, gave rise to the formation of
the corresponding alkoxysilane as the only detectable product;
their identities were determined by GC/MS analysis of the crude
(
(
(
55) Gerrard, W.; Kilburn, K. D. J. Chem. Soc. 1956, 1536.
56) Brook, A. G.; Dillon, P. J. Can. J. Chem. 1969, 47, 4347.
57) Rakita, P. E.; Srebro, J. P.; Worsham, L. S. J. Organomet. Chem.
1
976, 104, 27.
58) Komarov, N. V.; Yarosh, O. G.; Ivanova, Z. G. IzV. Akad. Nauk,
Ser. Khim. 1971, 872.
(
(
(
(
59) Viswanathan, N.; Van Dyke, C. H. J. Chem. Soc. A 1968, 487.
60) Hengge, E.; Holtschmidt, N. Monatsh. Chem. 1968, 99, 340.
61) Bain, S.; Ijadi-Maghsoodi, S.; Barton, T. J. J. Am. Chem. Soc. 1988,
(62) Leigh, W. J.; Workentin, M. S.; Andrew, D. J. Photochem.
Photobiol. A: Chem. 1991, 57, 97.
1
10, 2611.

9512 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Leigh et al.
Acknowledgment. We wish to thank the Natural Sciences
3a-i with EtOH and t-BuOH (12 pages, print/PDF). See any
current masthead page for ordering information and Web access
instructions.
and Engineering Research Council of Canada for financial
support and the Deutsche Forschungsgemeinschaft for a fel-
lowship to C.K.
Supporting Information Available: UV absorption spectra
and 3-parameter substituent correlation data for quenching of
JA981435D