Kinetics and mechanism of the addition of aliphatic amines to transient silenes

Article abstract of DOI:10.1021/jp026267x

Primary and secondary amines such as n-butyl-, t-butyl-, and N,N-diethylamine add across the Si=C bond of transient silenes such as 1,1-diphenylsilene (1a) and 1,1-bis(4-trifluoromethylpenyl)silene (1b) to yield the corresponding amino(methyl)diarysilanes as the only products of reaction. The kinetics and mechanism of reaction of these three amines with the two 1,1-diarylsilene derivatives have been studied in hexane, acetonitrile, and tetrahydrofuran (THF) solution by laser flash photolysis techniques, using the corresponding 1,1-diarylsilacyclobutanes as photochemical precursors to the silenes. The reactions proceed with clean second-order kinetics and bimolecular rare constants in excess of 5 × 10⁸ M^-1 s^-1 in hexane and MeCN, with 1b being up to four times more reactive than 1a. Arhenius plots for reaction in hexane and/or acetonitrile solution show strong curvature over the 0-60°C temperature range, consistent with an addition mechanism involving the intermediacy of a zwitterionic silene-amine addition complex, which collapses to product by intramolecular proton transfer from nitrogen of the silenic carbon. The reactions are substantially slower in THF, where rate reductions on the order of 5-10-fold and 50-70-fold are observed for 1a and 1b, respectively, as compared to MeCN solution. This is due to the effects of complexation of the silenes with the ether solvent, the equilibrium constant for which enters the expression for the reaction rate constant in the complexing solvent. In contrast to the situation in hexane and MeCN. Arrhenius plots for reaction of n-BuNH₂ wth 1a,b in THF solution are linear and lead to positive activation energies. Addition of amines in THF solution is proposed to occur predominantly via the free silenes but with minor contributions from a pathway involving nucleophilic displacement of the solvent from the silene-solvent complex.

Full text of DOI:10.1021/jp026267x

J. Phys. Chem. A 2003, 107, 1517-1524
1517
Kinetics and Mechanism of the Addition of Aliphatic Amines to Transient Silenes
William J. Leigh* and Xiaojing Li
Department of Chemistry, McMaster UniVersity, 1280 Main Street West, Hamilton, Ontario, Canada L8S 4M1
ReceiVed: June 8, 2002; In Final Form: NoVember 14, 2002
Primary and secondary amines such as n-butyl-, t-butyl-, and N,N-diethylamine add across the SidC bond of
transient silenes such as 1,1-diphenylsilene (1a) and 1,1-bis(4-trifluoromethylphenyl)silene (1b) to yield the
corresponding amino(methyl)diarylsilanes as the only products of reaction. The kinetics and mechanism of
reaction of these three amines with the two 1,1-diarylsilene derivatives have been studied in hexane, acetonitrile,
and tetrahydrofuran (THF) solution by laser flash photolysis techniques, using the corresponding 1,1-
diarylsilacyclobutanes as photochemical precursors to the silenes. The reactions proceed with clean second-
order kinetics and bimolecular rate constants in excess of 5 × 108 M-1 s-1 in hexane and MeCN, with 1b
being up to four times more reactive than 1a. Arrhenius plots for reaction in hexane and/or acetonitrile solution
show strong curvature over the 0-60 °C temperature range, consistent with an addition mechanism involving
the intermediacy of a zwitterionic silene-amine addition complex, which collapses to product by intramolecular
proton transfer from nitrogen to the silenic carbon. The reactions are substantially slower in THF, where rate
reductions on the order of 5-10-fold and 50-70-fold are observed for 1a and 1b, respectively, as compared
to MeCN solution. This is due to the effects of complexation of the silenes with the ether solvent, the equilibrium
constant for which enters the expression for the reaction rate constant in the complexing solvent. In contrast
to the situation in hexane and MeCN, Arrhenius plots for reaction of n-BuNH2 with 1a,b in THF solution are
linear and lead to positive activation energies. Addition of amines in THF solution is proposed to occur
predominantly via the free silenes but with minor contributions from a pathway involving nucleophilic
displacement of the solvent from the silene-solvent complex.
Introduction
toward alcohols18,19 and various other oxygen-centered nucleo-
philes.9
Silenes react regiospecifically with nucleophiles such as
water, alcohols, carboxylic acids, alkoxysilanes, and amines by
(1,2)-addition across the MdC bond.1-9 The mechanism of
alcohol addition has been extensively studied and is particularly
well-understood (eq 1).10,11 It is initiated by reversible nucleo-
philic attack at silicon to form a zwitterionic complex,1,12 which
proceeds to product by two competing, rate-limiting proton
transfer pathwayssone intramolecular and one involving a
second molecule of alcohol (see eq 1).13-15 The two proton
transfer pathways have different stereo- and regiochemical
consequences; the intramolecular pathway leads to stereoselec-
tive syn (1,2)-addition, while the bimolecular pathway leads to
overall trans addition or the formation of (1,4)- and (1,6)-
addition products if a polyenic substituent is present on the
silenic carbon.14 Time-resolved spectroscopic studies have
provided kinetic evidence for the two proton transfer pathways
in selected cases, in the form of a mixed first- and second-
order dependence of the rate on alcohol concentration.14,16,17
However, most of the transient silenes that have been studied
to date are sufficiently reactive toward alcohols and water that
only the second-order reaction pathway can be detected by these
methods.9 The 1,1-diarylsilene derivatives 1a-e are examples
of silenes that show such behavior and have proven extremely
useful in delineating the quantitative aspects of silene reactivity
The reaction of silenes with nitrogen-centered nucleophiles
is also well-known,1,3,8 but no detailed absolute kinetic studies
have yet been reported.7 More than 15 years ago, Wiberg and
co-workers published the results of competition experiments on
the addition of a comprehensive series of reagents to silene 2
and showed that primary amines are slightly more reactive than
aliphatic alcohols toward addition to the SidC bond of this
compound.1 Preliminary kinetic experiments in our laboratory
* To whom correspondence should be addressed. E-mail: leigh@
mcmaster.ca.
10.1021/jp026267x CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/12/2003

1518 J. Phys. Chem. A, Vol. 107, No. 10, 2003
Leigh and Li
on the addition of n-butylamine (n-BuNH2) to 1,1-diphenylsilene
(1a) have shown this to be true of this silene as well.7 This and
Wiberg and co-workers’ characterization of metastable com-
plexes of tertiary aliphatic amines with 22 and the related silene
34 suggest that as with alcohol addition, the reaction with
primary amines is initiated by nucleophilic attack at silicon,
followed by proton transfer from nitrogen to carbon.
In this paper, we report the results of a laser flash photolysis
study of the kinetics of addition of primary and secondary
amines to the silicon-carbon double bonds of 1a and 1,1-bis-
(4-trifluoromethylphenyl)silene (1b) in hexane, acetonitrile
(MeCN), and tetrahydrofuran (THF) solution, with a view of
fully defining the kinetics and mechanism for the addition of
aliphatic amines to transient silenes.
Results
Steady state photolysis (254 nm) of deoxygenated 0.02-0.08
M solutions of the 1,1-diarylsilacyclobutanes 4a,b in hexane
or cyclohexane-d12 containing a ca. 10% molar excess of
n-BuNH2, tert-butylamine (t-BuNH2), or N,N-diethylamine (Et2-
NH) afforded (along with ethylene) the corresponding aminosi-
lanes 5-7 as the major nitrogen-containing products (eq 2).
These are the products expected from (1,2)-addition of the
amines to silenes 1a and 1b, respectively. The secondary
aminosilanes formed from n-BuNH2 and t-BuNH2 (5a,b and
6a,b) are evidently much less reactive than the primary amines
from which they are derived, as the photolyses could be carried
out to quite high conversions, with little or no evidence for the
formation of secondary products corresponding to addition of
5 or 6 to 1.
Figure 1. Plots of (kdecay - k0) vs [amine] for quenching of 1a by
n-BuNH2 (O), t-BuNH2 (0), and Et2NH (]) in (a) hexane and (b)
acetonitrile solution at 23 °C.
for the more electrophilic silene (1b), which results in a greater
spectral shift (λmax ) 375 nm), only slight variations in the
spectrum with temperatures between 0 and 60 °C, and a more
pronounced lengthening in lifetime as compared to 1a. Under
these conditions, the lifetimes of 1a and 1b are limited by the
rates of dimerization and reaction with trace amounts of water.
The lifetimes of 1a,b were shortened upon addition of
n-BuNH2, t-BuNH2, or Et2NH to the solutions. The decays
followed clean pseudo-first-order kinetics in the presence of
amine, with individual lifetimes that were independent of
monitoring wavelength throughout the 310-400 nm range in
each solvent. The decay rate constants (kdecay) varied linearly
with amine concentration according to eq 3, where kamine is the
second-order rate constant for reaction of 1a,b with amine and
k0 is the pseudo-first-order rate constant for silene decay in the
absence of amine. For example, Figure 1 shows plots of the
corrected pseudo-first-order rate constants for decay of 1a (kdecay
- k0) as a function of amine concentration in hexane and
acetonitrile solution (the corresponding plots for silene 1b are
available as Supporting Information). Table 1 lists the absolute
second-order rate constants obtained from these experiments.
Laser flash photolysis of continuously flowing solutions of
4a and 4b (∼0.004 M) in dry hexane or MeCN ([H2O] e 10-4
M) gave rise to transient absorptions (λmax ) 325 nm) assignable
to 1a and 1b, respectively, as previously reported.18,20,21 The
two silenes exhibit lifetimes of >5 µs in hexane and 1-4 µs in
acetonitrile under these conditions and decay with mixed first-
and second-order kinetics, with 1a being the longer-lived of
the two. Much different behavior is observed in THF solution,
where the lifetimes of the two silenes are longer, 1b is the
longer-lived of the two, and their spectra are broadened and/or
red-shifted as compared to hexane solution because of com-
plexation with the ether solvent.19 As reported previously,19 the
spectrum of 1a varies over the 0-60 °C temperature range,
exhibiting λmax ) 325 and 350 nm at the high and low ends of
this temperature range, respectively. Complexation is stronger
kdecay ) k0 + kamine[amine]
(3)
Addition of up to ca. 0.01 M triethylamine caused no change
in the lifetime or absorption spectrum of either silene in hexane
solution. Only the initial transient ∆OD values varied as a
function of amine concentration, becoming increasingly smaller

Addition of Aliphatic Amines to Transient Silenes
J. Phys. Chem. A, Vol. 107, No. 10, 2003 1519
TABLE 1: Absolute Rate Constants (in Units of 109 M-1
s-1) for Reaction of Aliphatic Amines with 1a and 1b in
Hexane, MeCN, and THF Solution at 23 °Ca
kamine (109 M-1 s-1
)
silene
amine
hexane MeCN
THF
Ph2SidCH2
(1a)
n-BuNH2 6.49 ( 0.12 2.81 ( 0.06 0.32 ( 0.04
t-BuNH2 0.54 ( 0.01 1.59 ( 0.07 0.150 ( 0.004
Et2NH
0.77 ( 0.02 1.48 ( 0.06 0.148 ( 0.005
(4-CF3C6H4)2- n-BuNH2 10.6 ( 0.1 3.28 ( 0.06 0.094 ( 0.003
SidCH2
(1b)
t-BuNH2 2.03 ( 0.04 2.06 ( 0.11 0.032 ( 0.001
Et2NH 3.33 ( 0.10 1.46 ( 0.11 0.020 ( 0.001
a Errors are reported as (2σ from least squares analyses of kdecay vs
[amine] data according to eq 3.
Figure 3. Arrhenius plots for reaction of silenes 1a (O) and 1b (0)
with n-butylamine (a), tert-butylamine (b), and N,N-diethylamine (c)
in dry acetonitrile solution.
Figure 2. Arrhenius plots for reaction of silenes 1a (a) and 1b (b)
with n-BuNH2 in hexane (0), acetonitrile (O), and THF (]) solution.
Arrhenius plots for reaction of 1a,b with the three amines in
hexane and/or acetonitrile solution all show distinct curvature
over the 0-60 °C temperature range. In contrast, those for
reaction of n-BuNH2 in THF are linear; least squares analyses
afforded the Arrhenius parameters Ea ) 7.3 ( 0.6 kcal/mol and
log(A/M-1 s-1) ) 14.0 ( 0.5 (r2 ) 0.9914) in the case of 1a
and Ea ) 4.9 ( 1.3 kcal/mol and log(A/M-1 s-1) ) 11.6 ( 1.0
(r2 ) 0.9476) in the case of 1b.
as the amine concentration was increased due to competing
absorption of the excitation light by the amine. It was thus not
possible to carry out experiments at amine concentrations high
enough to enable discrete spectroscopic observation of silene-
amine complexes.
Absolute rate constants for reaction of 1a,b with n-BuNH2
were determined at several temperatures between 0 and 60 °C
in the three solvents and are shown as Arrhenius plots in Figure
2. Similar experiments were carried out for the reactions of 1a,b
with t-BuNH2 and Et2NH in acetonitrile, the results of which
are shown in Figure 3 along with those for n-BuNH2 from Figure
2. For comparison, the temperature dependencies of the diffu-
sional rate constants in acetonitrile (calculated using the standard
Debye equation (kdiff ) 8RT/3000η) and published temperature-
viscosity data22) are also shown in the plots of Figure 3. The
Discussion
The 1,1-diarylsilenes 1a,b are both exceedingly electrophilic
and react rapidly with the three amines studied in this work.
The nonhindered primary amine n-BuNH2 reacts somewhat
faster than its more hindered isomers t-BuNH2 and Et2NH,
especially in hexane solution. This can probably be ascribed to

1520 J. Phys. Chem. A, Vol. 107, No. 10, 2003
Leigh and Li
steric effects, given that the (nonaqueous) pKa values of the
three amines are essentially identical.23 Silene 1b is consistently
a factor of ca. 2-4 times more reactive than 1a in hexane
solution toward each amine studied, as we have found previously
for the reactions of these two silenes with oxygen-centered
nucleophiles.18 This indicates that the Hammett F-value for
amine addition to 1,1-diarylsilenes is small but positive,
consistent with a reaction mechanism analogous to that for
oxygen-centered nucleophiles.9 This mechanism is shown in eq
4, while eq 5 gives the expression for the overall rate constant
for reaction (kamine), derived using the steady state assumption
for the intermediate complex.
Debye equation and published viscosities.22 This indicates that
the initial complexation of the two silenes with the amine
proceeds with rate constants (kC) that are similar to one another
and close or equal to the diffusion-controlled limit. Second, the
maximum rate constant for reaction of the more reactive
derivative (1b) in hexane solution is attained at a higher
temperature (40-50 °C) than that for the less reactive derivative
(1a; 15-25 °C). This suggests that at any given temperature
within the range studied, collapse of the intermediate complex
to product is more efficient (and hence less kinetically signifi-
cant) for the more reactive derivative 1b, assuming similar
activation parameters for the complexation step in the two
derivatives.25 On the basis of our earlier studies of the
complexation of these two compounds with THF,19 we suspect
that this is due mainly to k-C being smaller for the more reactive
derivative. Unfortunately, attempts to detect complexes of 1a,b
with triethylamine were unsuccessful due to light absorption
problems with solutions containing more than 0.01 M amine,
so we are unable to compare the quantitative details of tertiary
amine complexation with those of an oxygen-centered nucleo-
phile such as THF. Nevertheless, the trends observed here in
the temperature dependencies of the reactions of 1a,b with
aliphatic amines are similar to those reported previously by us
for the reactions of these silenes with aliphatic alcohols,18 acetic
acid,18 acetone,21 and methoxytrimethylsilane.20 In the latter
case, the electrophilic component of the trapping agent is the
trimethylsilyl group, rather than a proton.
kamine ) kC(kH/[kH + k-C])
(5)
More evidence for this mechanism is provided by the
temperature dependencies of the absolute rate constants for
reaction of the two silenes with n-BuNH2 in hexane and MeCN
solution (Figure 2). The nonlinear variations in log kamine with
T-1 result from the superposition of a positive temperature
dependence of kC on a negative temperature dependence of the
complex partitioning ratio, kH/(kH + k-C), which occurs because
the reaction is strongly exothermic and the rate constant for
the product-forming step (kH) is dominated by entropic factors.9
The temperature dependence of kamine, as defined by this
mechanism, is given explicitly by eq 6, where AC and EC are
the Arrhenius preexponential factor and activation energy for
formation of the intermediate complex, A-C and AH are the
preexponential factors for reversion of the complex to starting
materials and collapse to product, respectively, and ∆E-C,H
() E-C - EH) is the difference in activation energies for the
two processes. Integration of the logarithmic form of this
Figure 4 shows a comparison of the Arrhenius data for
reaction of 1a,b with n-BuNH2 in acetonitrile to our previously
reported data for addition of MeOH, AcOH, and acetone to the
same two silenes in the same solvent.18,21 It is interesting to
note that on a qualitative level, the kinetics of amine addition
exhibit characteristics most like those of acetic acid addition;
both reactions exhibit positive temperature dependencies over
the 0-60 °C temperature range, relatively small variations in
rate with ring substituents in the silene, and overall rate constants
in excess of 109 M-1 s-1. The positive temperature dependencies
indicate that the kinetics of both reactions are dominated by
the initial complexation step and thus that the partitioning of
the intermediate complexes strongly favors product formation
in both cases (i.e., kH > k-C in eq 5). The data for 1a suggest
that the rate constant for the initial complexation step is
somewhat higher for the amine, as would be expected consider-
ing its higher nucleophilicity as compared to the carboxylic acid.
The difference is small, however, and narrows further as the
electrophilicity of the silene is increased through CF3 substitution
in the aryl rings. While substantial differences would be
predicted in the absolute magnitudes of the rate constants for
reversion of the intermediate complex to reactants and collapse
to product (k-C and kH, respectively) for amine vs carboxylic
acid addition, because of the dramatically different nucleophi-
licities and proton acidities of the two reagents, these differences
are masked by the fact that it is their relative magnitudes that
determine the overall rate constant for reaction and the form of
its temperature dependence.
expression with respect to T-1 leads to the prediction that kamine
overall
goes through a maximum value (i.e., Ea
) 0) at the
temperature where the partitioning ratio is equal to the quantity
(1 - EC/∆E-C,H).24 The Arrhenius plot exhibits a positive slope
(negative Ea) over the temperature range where the partitioning
ratio is less than this value and a negative slope (positive Ea)
over the range where it exceeds it.
kamine ) [ACexp(-EC/RT)]/[1 + (A-C/AH)exp(-∆E-C,H/RT)] (6)
While the solid lines in the plots for quenching in hexane
and acetonitrile solution shown in Figures 2 and 3 represent
the best fits of the experimental data to eq 6, the four Arrhenius
parameters obtained from the analyses have extremely large
errors associated with them, not unexpectedly considering the
small number of data points obtained and the relatively narrow
temperature range over which they were determined. Accord-
ingly, we can offer only a few qualitative comments on the data
at the present time. First, the data show clearly that the overall
rate constants for reaction (kamine) of 1a and 1b with n-BuNH2
in hexane take on maximum values of (6-7) × 109 M-1 s-1
and ca. 1.2 × 1010 M-1 s-1, respectively, values that are within
a factor of 3 and 2 (respectively) of the diffusional rate constants,
calculated at the corresponding temperatures using the standard
The data of Figure 2 show that in acetonitrile solution, the
reactivities of 1a,b with n-BuNH2 are lower and the apexes of
the Arrhenius plots are shifted toward higher temperatures as
compared to the situation in hexane solution. This behavior is
also analogous to what has been observed previously for the
addition of oxygen-centered nucleophiles to these two
silenes.15,18,21 In the case of oxygen-centered nucleophiles, we
have recently shown that these trends are due, at least to some
extent, to the effects of weak complexation of the free silene

Addition of Aliphatic Amines to Transient Silenes
J. Phys. Chem. A, Vol. 107, No. 10, 2003 1521
second-order rate constant for reaction in the complexing solvent
(kNu-ESïlv) is given by the equivalent expressions of eqs 8 and
9, where F1 and F1-Solv are, respectively, the fractional
concentrations of free and complexed silene present in solution.
If the complexed form of the silene is substantially less reactive
than the uncomplexed form, as might be expected for particu-
larly nonacidic nucleophilic reagents, then the expression for
the overall rate constant for reaction simplifies to that of eq 10.
This limiting situation is the one that will lead to the largest
(positive) shift in the observed Ea for reaction in a complexing
solvent, as compared to its value in a noncomplexing solvent
of similar bulk polarity.19 For reactions that exhibit very small
or negative intrinsic activation energies in noncomplexing
solvents, the temperature dependence for reaction in a com-
plexing medium will be dominated by the complexation
phenomenon itself; this probably accounts for the observation
of (apparent) preexponential factors that approach or exceed
those of diffusion. For two silenes of differing intrinsic
electrophilicities, and hence different equilibrium constants for
complexation with a nucleophilic solvent, this can also lead to
an inversion in their relative reactivities in a strongly complexing
solvent as compared to the situation in non- or weakly
complexing solvents.
Figure 4. Arrhenius plots for reaction of 1a (a) and 1b (b) with
n-BuNH2 (0), MeOH (O), acetone (]), and AcOH (∆) in acetonitrile
solution. Data for MeOH and AcOH: Reprinted from ref 18. Copyright
1997 National Research Council of Canada. Data for acetone: Reprinted
from ref 21. Copyright 1998 American Chemical Society.
kNu-ESolv ) F1k1 + F1-Solvk1-Solv
(8)
K
Solv[Solv]
1
Solv
kNu-E
)
k1 +
k1-Solv
(1 + KSolv[Solv])
(1 + KSolv[Solv])
with the nitrile solvent.19 Such complexation reduces overall
reactivity by lowering the concentration of free silene available
for reaction with the nucleophile; furthermore, because the
equilibrium constant for solvent complexation increases with
decreasing temperature, the overall Ea values for reaction in the
nitrile solvent are shifted to more positive values relative to
hexane solution because of the superposition of the temperature
dependence of this (nonproductive) phenomenon on the true
temperature dependence of the reaction.19 This effect provides
an explanation for the small but nonetheless interesting varia-
tions in the relative reactivities of 1a,b with the three amines
in acetonitrile solution as a function of temperature (Figure 3);
1b is the more reactive of the two silenes at high temperatures,
but the difference decreases or even inverts as the temperature
is lowered. Even larger changes in the relative reactivity of the
two silenes are effected in THF solution, where 1b is substan-
tially less reactive than 1a throughout the entire temperature
range studied. The ether solvent is much less polar than
acetonitrile but more strongly complexing, particularly with
silicon electrophiles.
(9)
1
Solv
kNu-E ∼ F1 ‚ k1 )
‚ k1
(10)
(1 + KSolv[Solv])
The kinetic data for the reactions of 1a,b with the three
amines in THF solution are fully consistent with this analysis
and show trends similar to those reported previously for reaction
of 1a,b with acetone, another nucleophilic silene trap of
especially low Bronsted acidity.19 As with acetone addition, the
activation energies for reaction of 1a,b with n-BuNH2 are shifted
to substantially more positive values and the relative reactivities
of the two silenes are inverted in THF solution as compared to
hexane or MeCN. Moreover, the magnitude of the differences
in reactivity of 1a,b in THF relative to MeCN is similar to that
reported previously for acetone addition,19 consistent with the
fact that amines also fit into the general category of weakly
acidic nucleophiles that in THF solution, react exclusively (or
almost so; vide infra) via the free silene. This is illustrated more
clearly in Figure 5, which shows the Arrhenius plots for the
addition of n-BuNH2 to 1a,b in THF solution, along with our
previously reported data for acetone and MeOH addition.19 As
can be seen in the figure, both acetone and n-BuNH2 addition
exhibit similarly large, positive activation energies in THF,
which has been shown in the former case19 to mainly reflect
the temperature dependence of KSolv (or F1 in eq 10). On the
other hand, reaction with MeOH proceeds with similar absolute
Solvent complexation alters the mechanism for reaction of
silenes relative to those in noncomplexing solvents as described
in eq 7, where KSolv is the equilibrium constant for solvent
complexation, and k1 and k1-Solv are the rate constants for
reaction of the nucleophile (Nu-E) with the free and complexed
forms of the silene, respectively.19,26 In the general situation
where both the free and the complexed forms react, the overall

1522 J. Phys. Chem. A, Vol. 107, No. 10, 2003
Leigh and Li
to which the relative reactivities of 1a,b are inverted in the
complexing solvent as compared to noncomplexing solvents.
It does appear to be more dominant with reasonably acidic
nucleophilic traps such as acetic acid, and as a result, there is
very little net change in the relative reactivities of 1a,b toward
this reagent in THF solution as compared to the more weakly
complexing solvent MeCN.
While the idea that silene-THF complexes should be prone
to reaction with particularly acidic nucleophilic traps is a
reasonable one that has some experimental support, one might
also expect them to be reactive toward particularly strong
nucleophiles, yielding the same nucleophile-silene addition
complex as involved in the reaction with the free silene but by
a mechanism involving SN2 displacement of the solvent in the
first step of the reaction. This could account for the fact that
the activation energy for reaction of 1b with n-BuNH2 in THF
is smaller than that for reaction of 1a, in contrast to the opposite
trend in activation energies that is observed for reaction of the
two silenes with acetone in this solvent (Figure 5). Because
reaction by this pathway can be expected to be intrinsically much
slower than that involving the free silenes, it can be expected
to exhibit a substantial substituent effect, with a reasonably large,
positive Hammett F-value. Thus, it is expected to be more
important with 1b than with 1a, and because the extent to which
it contributes to the overall rate constant for reaction will
increase as the fractional concentration of complexed silene
increases (i.e., with decreasing temperature), it could account
for 1b exhibiting a smaller positive activation energy than 1a.
Summary and Conclusions
The transient 1,1-diarylsilenes 1a,b react with aliphatic amines
by regiospecific (1,2)-addition, with absolute rate constants
within a factor of 10 of the diffusional rate constants in hexane
and acetonitrile solution over the 0-60 °C temperature range.
Similar reactivities are observed toward more hindered primary
amines and secondary amines under similar conditions. The
Arrhenius plots for these reactions show strong curvature over
this temperature range, which is consistent with a stepwise
reaction mechanism analogous to that for addition of oxygen-
centered nucleophiles such as methanol: reaction begins by
reversible attack of the neutral nucleophile at silicon to yield a
zwitterionic addition complex, which proceeds to product by
intramolecular proton transfer from the nucleophilic site to the
silenic carbon.
Positive activation energies for reaction are observed for the
reaction of n-BuNH2 with 1a,b in THF solution, and the relative
reactivities of the two silenes are inverted as compared to the
situation in hexane or MeCN solution. These effects are due to
complexation of the silenes with the ether solvent, the equilib-
rium constant for which enters the expression for the overall
rate constant for reaction. The temperature dependencies of the
rate constants in THF are qualitatively similar to those observed
previously for reaction of 1a,b with acetone, another nucleo-
philic silene trapping agent of relatively low Bronsted acidity.
The relatively large effects of the solvent-silene complexation
equilibrium for nonacidic silene traps, as compared to more
acidic ones such as alcohols or acetic acid, are due to the fact
that the reaction proceeds almost exclusively via the free silene,
and the solvent complexes are relatively unreactive. More
detailed comparisons suggest that in fact, reaction via the
silene-THF complex may contribute in a minor way to the
kinetics of n-butylamine addition to 1b, the more highly
electrophilic (and hence more strongly complexed) of the two
silenes studied in this work. The possible intervention of this
Figure 5. Arrhenius plots for reaction of silenes 1a (O) and 1b (0)
with acetone (a), MeOH (b), and n-butylamine (c) in dry THF solution.
Panels a and b: Reprinted from ref 19. Copyright 2002 American
Chemical Society.
rate constants to that with n-BuNH2 but exhibits almost no
variation in rate with temperature. This has been proposed to
be due to the intervention of a second reaction pathway,
involving reaction of the alcohol with the complexed form of
the silene via initial protonation at carbon, that competes with
the “normal” mechanism involving reaction with the free silene
via initial nucleophilic attack at silicon.19 The smaller positive
activation energies result from the fact that this pathway should
increase in importance with decreasing temperature, as the
solvent complexation equilibrium increasingly favors the silene-
solvent complex over the free silene. It should be especially
important in the case of 1b, which complexes more strongly
with the ether solvent. This second pathway is sufficiently minor
in the case of alcohol addition that it does not alter the extent

Addition of Aliphatic Amines to Transient Silenes
J. Phys. Chem. A, Vol. 107, No. 10, 2003 1523
pathway is manifested by a slight rate acceleration at the low
end of the temperature range studied, where the fractional
concentration of the silene-THF complex is highest and results
in a smaller overall activation energy for reaction of this silene
with the amine as compared to that for 1a. The opposite trend
in activation energy is observed for the reactions of 1a,b in
THF with the substantially weaker nucleophile, acetone. Evi-
dence for reaction of silene-THF complexes has been reported
previously for the reaction of 1a,b with alcohols and acetic acid,
whose higher proton acidities can lead to reaction of the silene-
THF complex via initial protonation at the silenic carbon. Future
work will further explore the effects of silene electrophilicity,
solvent complexation, and trapping agent nucleophilicity and
acidity on the kinetics and mechanisms of the reactions of
silicon-carbon double bonds.
A solution of n-butyllithium (20.5 mmol) in anhydrous ether
(50 mL) was placed in a 250 mL round bottom flask fitted with
a condenser, addition funnel, nitrogen inlet, and magnetic stirrer
and cooled in an ice bath. A solution of N,N-diethylamine (1.5
g, 20.5 mmol) in anhydrous diethyl ether (30 mL) was then
added dropwise with stirring over ca. 30 min. The ice bath was
removed, and the solution was stirred for 1 h, after which a
solution of diphenylmethylchlorosilane (4.77 g, 20.5 mmol) in
anhydrous ether (50 mL) was added dropwise over 30 min,
causing the formation of a colorless precipitate. The mixture
was stirred for a further 3 h and filtered, and the solvent was
removed by distillation. Pentane (20 mL) was added to the
resulting yellow oily solid, the mixture was filtered, and the
solvent was removed again by distillation to yield a yellow oil,
which was then distilled under vacuum. The product was
collected as a colorless oil (bp 97-98 °C (0.03 mmHg), 4.85
g, 87.5%) and identified as (N,N-diethylamino)methyldiphe-
nylsilane (7a) on the basis of the following spectroscopic data.
1H NMR (CDCl3): δ (ppm) 7.56 (m, 4H), 7.36 (m, 6H), 2.91
(q, 4H), 0.99 (t, 6H), 0.58 (s, 3H). 13C NMR (CDCl3): δ (ppm)
138.1, 134.8, 129.1, 127.6, 40.1, 15.5, -2.2. IR (neat): 3069
(m), 2965 (s), 2866 (m), 1428 (m), 1375 (m), 1253 (m), 1172
(s), 1110 (s), 1028 (s), 927 (m), 789 (s), 724 (m), 700 (m).
EIMS: m/z (I) 269 (7), 255 (14), 254 (62), 198 (19) 197 (100),
183 (8), 181 (9), 121 (15), 105 (20). HRMS: m/z calcd for
C17H23NSi (M+), 269.1600; found, 269.1586. The 13C NMR
spectrum of the compound is similar to that reported previously
by Filleux-Blanchard and An.27 Spectroscopic data for the other
aminosilanes (5a, 6a, 5b-7b) are listed in the Supporting
Information.
Experimental Section
1H NMR spectra were recorded in deuteriochloroform or
cyclohexane-d12 on Bruker AC200 or DRX500 spectrometers
and are referenced to tetramethylsilane. Infrared spectra were
recorded on a BioRad FTS-40 FT/IR spectrometer and are
reported in wavenumbers (cm-1). Ultraviolet absorption spectra
were recorded on a Varian Cary 50 spectrophotometer. Low-
resolution mass spectra were determined by gas chromatography
(GC)/mass spectrometry (MS), using a Hewlett-Packard 5890II
gas chromatograph equipped with a HP-5971 mass selective
detector and a DB-5 fused silica capillary column (30 m × 0.25
mm; 0.25 µm film; Agilent Technologies). High-resolution
electron impact mass spectra and exact masses were determined
using a VGH ZABE mass spectrometer.
Analytical gas chromatographic analyses were carried out
using a Hewlett-Packard 5890II+ gas chromatograph equipped
with a conventional heated splitless injector, flame ionization
detector, HP3396A integrator, and DB1 or DB1701 megabore
capillary columns (15 m × 0.53 mm; Chromatographic Special-
ties, Inc.).
Acetonitrile (Caledon Reagent) was refluxed over calcium
hydride (Fisher) for several days, distilled under dry nitrogen,
and then cycled three times through a 1 in × 6 in column of
neutral alumina (Aldrich), which had been activated by heating
under vacuum (ca. 0.05 Torr) at 320 °C for 10 h with periodic
shaking. THF (Caledon Reagent) was refluxed over molten
potassium in a nitrogen atmosphere and distilled. n-Butylamine,
tert-butylamine, N,N-diethylamine, and N,N,N-triethylamine
(Aldrich) were distilled over solid potassium hydroxide. The
1,1-diarylsilacyclobutanes (4a,b) were prepared according to the
published methods.18
Nanosecond laser flash photolysis experiments employed the
pulses (248 nm; 15-20 ns; 70-120 mJ) from a Lambda Physik
Compex 100 excimer laser, filled with F2/Kr/Ne mixtures, and
a Luzchem Research mLFP-111 laser flash photolysis system.
Solutions were prepared at concentrations such that the absor-
bance at the excitation wavelength (248 nm) was ca. 0.7 (0.003-
0.004 M) and was flowed continuously through a 7 × 7 Suprasil
flow cell connected to a calibrated 100 mL reservoir. Oxygen
has no discernible effect on the lifetimes of 1a,b,18 so the
solutions were not deoxygenated. Solution temperatures were
measured with a Teflon-coated copper/constantan thermocouple
inserted directly into the flow cell. Reagents were added to the
reservoir by microliter syringe as aliquots of standard solutions.
Rate constants were calculated by linear least-squares analysis
of decay rate-concentration data (6-10 points) that spanned
at least a factor of 5 (usually more than 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.
Steady state photolysis (254 nm) of deoxygenated 0.02-0.08
M solutions of the 1,1-diarylsilacyclobutanes 4a,b (10-20 mg)
in hexane or cyclohexane-d12 containing a ca. 10% molar excess
of n-BuNH2, t-BuNH2, or Et2NH was carried out using a
Rayonet photochemical reactor. Photolyses were carried to 50-
95% conversion, as determined by GC or 1H NMR spectroscopic
analysis, and yielded ethylene and a single nitrogen-containing
product in each case. The products were isolated in impure form
as colorless oils in yields of 40-80%, by microdistillation of
the crude, yellow residues obtained from evaporation of solvent
and excess amine. They were not purified further because of a
sensitivity to hydrolysis and prolonged heating. The products
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for financial support,
Cameron R. Harrington for the preparation of the authentic
sample of 7a, and a referee for insightful comments.
Supporting Information Available: Spectroscopic data for
aminosilanes 5-7, plots of (kdecay - k0) vs [amine] for the
reaction of 1b with n-BuNH2, t-BuNH2, and Et2NH in hexane
and MeCN solution at 23 °C, and tables of absolute rate
constants for reaction of 1a,b with the three amines in hexane,
MeCN, and THF at various temperatures between 0 and 60 °C.
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
were identified by a combination of GC/MS, IR, H and 13C
NMR spectroscopy, and high-resolution exact mass determina-
tions on the M+ or (M+-15) ions, the latter being used in the
cases of 5b and 6b because the M+ ions are extremely weak.
The identity of aminosilane 7a was verified by GC coinjection
with an authentic sample, which was prepared as follows.

1524 J. Phys. Chem. A, Vol. 107, No. 10, 2003
Leigh and Li
(16) Sluggett, G. W.; Leigh, W. J. J. Am. Chem. Soc. 1992, 114, 1195-
1201.
References and Notes
(1) Wiberg, N. J. Organomet. Chem. 1984, 273, 141-177.
(2) Wiberg, N.; Kopf, H. J. Organomet. Chem. 1986, 315, 9-18.
(3) Wiberg, N.; Kim, C.-K. Chem. Ber. 1986, 119, 2980-2994.
(4) Wiberg, N.; Joo, K. S.; Polborn, K. Chem. Ber. 1993, 126, 67-69.
(5) Baines, K. M.; Stibbs, W. G. AdV. Organomet. Chem. 1996, 39,
275-324.
(17) Leigh, W. J.; Bradaric, C. J.; Sluggett, G. W.; Venneri, P.; Conlin,
R. T.; Dhurjati, M. S. K.; Ezhova, M. B. J. Organomet. Chem. 1998, 561,
19-27.
(18) Bradaric, C. J.; Leigh, W. J. Can. J. Chem. 1997, 75, 1393-1402.
(19) Leigh, W. J.; Li, X. Organometallics 2002, 21, 1197-1207.
(20) Leigh, W. J.; Bradaric, C. J.; Morkin, T. L.; Li, X. Organometallics
2001, 20, 932-936.
(21) Bradaric, C. J.; Leigh, W. J. Organometallics 1998, 17, 645-651.
(22) CRC Handbook of Chemistry and Physics; CRC Press: Boca Raton,
1995; pp 6-241.
(23) Izutsu, K. Acid-Base Dissociation Constants in Dipolar Aprotic
SolVents; IUPAC Chemical Data Series No. 35; Blackwell Scientific
Publications: Oxford, 1990; pp 17-35.
(24) We have previously stated that the maximum rate constant is
obtained at the temperature where the partitioning ratio is approximately
equal to 0.5 (i.e., when k-C ≈ k-H).9,11,21 This may be the case for the
addition of methanol to silene 1b in acetonitrile solution18 but is not
necessarily so in general.
(25) The other extreme situation where such behavior would be obtained
is one in which the complex partitioning ratios for the two derivatives are
similar and vary similarly with temperature, but the activation energy for
the initial step is higher for the more reactive compound (1b). This seems
unlikely but nevertheless cannot be ruled out on the basis of the data at
hand.
(6) Brook, A. G.; Brook, M. A. AdV. Organomet. Chem. 1996, 39,
71-158.
(7) Leigh, W. J. Pure Appl. Chem. 1999, 71, 453-462.
(8) Brook, A. G.; Yu, Z. Organometallics 2000, 19, 1859-1863.
(9) Morkin, T. L.; Owens, T. R.; Leigh, W. J. Kinetic studies of the
reactions of SidC and SidSi double bonds. In The Chemistry of Organo-
silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; John Wiley and
Sons: New York, 2001; Vol. 3, pp 949-1026.
(10) Sakurai, H. Mechanism and structures in alcohol addition reactions
of disilenes and silenes. In The Chemistry of Organic Silicon Compounds;
Rappoport, Z., Apeloig, Y., Eds.; Wiley and Sons Ltd: New York, 1998;
pp 827-855.
(11) Morkin, T. L.; Leigh, W. J. Acc. Chem. Res. 2001, 34, 129-136.
(12) Wiberg, N.; Wagner, G.; Muller, G.; Riede, J. J. Organomet. Chem.
1984, 271, 381-391.
(13) Kira, M.; Maruyama, T.; Sakurai, H. J. Am. Chem. Soc. 1991, 113,
3986-3987.
(14) Leigh, W. J.; Sluggett, G. W. J. Am. Chem. Soc. 1994, 116, 10468-
10476.
(15) Leigh, W. J.; Bradaric, C. J.; Kerst, C.; Banisch, J. H. Organome-
tallics 1996, 15, 2246-2253.
(26) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill: New York, 1995; pp 15-45.
(27) Filleux-Blanchard, M. L.; An, N. D. Org. Magn. Res. 1979, 12,
12-16.