The Yukawa-Tsuno relationship for the β-silicon effect in the solvolysis rates of 2-(aryldimethylsilyl)ethyl chlorides

Article abstract of DOI:10.1246/bcsj.78.1834

Solvolysis rates of 2-(aryldimethylsilyl)ethyl chlorides were determined conductimetrically in 60% (v/v) aqueous ethanol. The effects of aryl substituents at the silyl atom on the solvolysis rates at 50°C were correlated with essentially nonresonant σ parameters of r = 0:10 in terms of Yukawa-Tsuno (Y-T) Eq. 1, giving a ρ value of -1:75. Such a high ρ value may be regarded as the effect of aryl ring on the bridged Si in the rate-determining step. The Si-bridging is consistent with the fact that the solvolysis of the unsubstituted substrate with d₂-labeled ethylene moiety gave substitution products with label scrambling. The arylsilyl substituent effects were likewise analyzed for several relevant sets of β-silyl systems in order to ascertain significant variations of ρ values from system to system; Y-T Eq. 1 correlated quite excellently with essentially nonresonant sigmas of negligible resonance demand (r ? 0:10), to exhibit significant variations of ρ from -1:75 to -0:95.

Full text of DOI:10.1246/bcsj.78.1834

1834 Bull. Chem. Soc. Jpn., 78, 1834–1842 (2005)
ꢀ 2005 The Chemical Society of Japan
The Yukawa–Tsuno Relationship for the ꢀ-Silicon Effect
in the Solvolysis Rates of 2-(Aryldimethylsilyl)ethyl Chlorides
ꢀ
Mizue Fujio, Mai Uchida, Ayumi Okada, Md. Ashadul Alam,
Ryoji Fujiyama,1 Hans-Ullrich Siehl,2 and Yuho Tsuno
Institute for Materials Chemistry and Engineering, Kyushu University, Hakozaki, Higashi-ku, Fukuoka 812-8581
1Department of Material Science, Faculty of Science, Kochi University, Akebono-cho, Kochi 780-8520
2Division of Organic Chemistry 1, University of Ulm, D-89069 Ulm, Germany
Received March 3, 2005; E-mail: fujio@ms.ifoc.kyushu-u.ac.jp
Solvolysis rates of 2-(aryldimethylsilyl)ethyl chlorides were determined conductimetrically in 60% (v/v) aqueous
ethanol. The effects of aryl substituents at the silyl atom on the solvolysis rates at 50 ꢁC were correlated with essentially
ꢀ
nonresonant ꢀ parameters of r ¼ 0:10 in terms of Yukawa–Tsuno (Y–T) Eq. 1, giving a ꢁ value of ꢂ1:75. Such a high ꢁ
value may be regarded as the effect of aryl ring on the bridged Si in the rate-determining step. The Si-bridging is con-
sistent with the fact that the solvolysis of the unsubstituted substrate with d2-labeled ethylene moiety gave substitution
products with label scrambling. The arylsilyl substituent effects were likewise analyzed for several relevant sets of ꢂ-
silyl systems in order to ascertain significant variations of ꢁ values from system to system; Y–T Eq. 1 correlated quite
ꢃ
excellently with essentially nonresonant sigmas of negligible resonance demand (r 0:10), to exhibit significant varia-
tions of ꢁ from ꢂ1:75 to ꢂ0:95.
¼
It is well known that the ꢂ-silicon substituent greatly facil-
itates the heterolysis of the C–Lg bond by 1012-fold. This im-
provement is caused by trimethylsilyl group in an antiperipla-
nar conformation.1 The so-called ꢂ-silicon effect is manifested
primarily in an E1-like mechanism, in which departure of the
nucleofuge is rate-determining (Scheme 1). Stabilization of
this electron deficiency was interpreted in terms of either the
open form (1a+) stabilized by ꢀ–ꢃ hyperconjugation or by
the Si-bridged form (1b+) of the cationic intermediate,1–7 as
shown in Scheme 1. Advanced MO calculations indicated that
the two forms (1a+ and 1b+) have similar energies generally
in the secondary cations (R1{R4 ¼ H or Me in 1)6 but the
bridged structure (1b+) is somewhat more stable in the pri-
mary case, while the predictions have seldom been found ex-
perimental support in solvolytic reactions.5b,9–13 The confor-
mational dependences of the ꢂ-silicon effect on the dihedral
evidence for the Si-bridged pathway (1b). Thus the relative
importance of these two pathways in the primary system re-
mains substantially unresolved. Numerous results have re-
vealed the necessity of the antiperiplanar stereochemistry
and can be explained satisfactorily by either the non-classical
(bridged) or the classical (open cation) transition state.
On the other hand, the bridged structure (1b+ in Scheme 1)
has been invoked by Eaborn et al., who observed reversible
ionization and migration of the SiMe3 group between the
two carbon atoms in the solvolysis of Me3SiCH2CD2Br.13
The Si-bridged pathway (1b) is highly plausible in the primary
carbocationic system, while the question of which pathway, 1a
or 1b, should be favored still remains open. At least the ꢂ-
silyl-group bridged (silyl-participation) pathway appeared to
be analogous to the neighboring aryl-group assistance in the
ꢂ-arylalkyl solvolyses. The distinction between the pathways
(1a) and (1b) in Scheme 1 may be made most practically by
the difference in the relative position of Si atom in the transi-
tion state.
angles of C –Si bond and on the cation p-orbital in the solvol-
ꢂ
ysis of cyclic secondary systems,1,3b were taken as convincing
evidence for the Si–C hyperconjugation mechanism. Never-
ꢂ
theless, this conformational dependence may also be taken as
The correlation analysis of substituent effects should be an
effective tool of investigation of reaction mechanisms. The
Yukawa–Tsuno (Y–T) Eq. 1 has been successfully applied to
the analysis of various ꢂ-aryl-assisted solvolyses; the ꢁ and
r values permit one to assess the structure of the transition
state:14
R
MeSiMe
R4
R3
R1
R2
R
MeSiMe
R4
R3
R1
R2
(1a+)
0
ꢀ þ
logðk=k0Þ ¼ ꢁðꢀ þ rꢁꢀR Þ;
ð1Þ
R
E1 or SN1 products
Lg
MeSiMe
where k is the rate constant for a given reaction of a ring-
substituted derivative and k0 is the corresponding value of
the unsubstituted one. ꢀ0 is the normal substituent constant
(1)
R1
R2
R4
R3
(1b+)
ꢀ þ
that excludes additional ꢃ-electronic interaction, and ꢁꢀR
is the resonance substituent constant, defined by ꢀþ ꢂ ꢀ0,
Scheme 1. ꢂ-Silicon effect in the carbocationic solvolysis.
Published on the web October 4, 2005; DOI 10.1246/bcsj.78.1834

M. Fujio et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1835
Table 1. Solvolysis Rates of 2-(Aryldimethylsilyl)ethyl Chlorides in 60% (v/v) aq EtOH
Subst.
105 ꢄ k/sꢂ1
55 C
ꢁH5z0
ꢁSz50
ꢁC
ꢁC
/kcal molꢂ1aÞ
/e.u.aÞ
ꢁ
ꢁ
ꢁ
ꢁ
25 C
45 C
50 C
114.3bÞ
p-MeO
p-Me
3,5-Me2
m-Me
6.45
66.7
21.4
ꢂ6:0
94.8cÞ
79.2cÞ
71.9cÞ
H
3.43
34.7
59.1,bÞ 59.5cÞ
29.2bÞ
21.2
20.8
20.7
22.8
ꢂ8:0
ꢂ10:5
ꢂ11:3
ꢂ8:0
p-MeO-m-Cl
p-Cl
m-Cl
m-CF3
3,5-(CF3)2
1.773
1.597
0.575
17.28
15.74
6.431
24.9,bÞ 25.9cÞ
11.33,bÞ 11.77cÞ
7.1cÞ
19.59
1.99
0.661
1.157bÞ
22.2
ꢂ12:5
a) 1 cal = 4.184 J. b) Calculated from the rate data at other temperatures. c) Ref. 17.
measuring the capability of ꢃ-delocalization of p–ꢃ-donor
substituent.
Table 2. Solvolysis Rates of 2-(Dimethylphenylsilyl)ethyl
ꢁ
Chloride at 25 C
The present ꢂ-silyl group participation reactions (Scheme 1)
were similarly subjected to correlation analysis by introducing
X-substituted phenyls to the silyl function (R ¼ X-C6H4 in
1).15–17 The final goal of our investigation is to establish the
use of a selectivity parameter (ꢁ-value) of the arylsilyl sub-
stituent effects as a criterion to distinguish the mechanisms
1a and 1b.
In this paper, we have carried out determination of the sol-
volysis rates of a wide range of ring substituents to rationalize
the aryl substituent effect on the ꢂ-Si participation in the
solvolysis of the simplest ꢂ-silyl system 2.
Solv.aÞ
105 ꢄ k/sꢂ1
Solv.aÞ
105 ꢄ k/sꢂ1
EtOH
90E
80E
70E
60E
50E
40E
30E
80A
70A
60A
50A
40A
0.0094bÞ
0.111bÞ
0.312
1.097
3.40
30A
MeOH
80M
60M
50M
224
0.102bÞ
1.819
18.90
59.9
38.5
49.4
37.9
59.1
0.0338bÞ
0.1279
0.807
5.99
11.85
55.0
249
0.107bÞ
0.329
1.677
5.63
TFE
97TcÞ
80T
50T
80ET
60ET
40ET
20ET
Results
2-(Aryldimethylsilyl)ethyl chlorides (2) were synthesized
according to the procedure of Vencl et al.16 The solvolysis
rates for these aryl substituents in 60% (v/v) aqueous ethanol
(60E) were measured conductimetrically at appropriate tem-
peratures in the same manner as reported before17,18 and were
32.7
a) Volume percent (v/v) of first named organic component un-
less otherwise noted. Abbreviations; E ¼ EtOH, M ¼ MeOH,
A ¼ Acetone, T ¼ TFE (2,2,2-trifluoroethanol), ET ¼
EtOH{TFE (e.g., 20ET = 20% vol EtOH:80% vol TFE mix-
ture). b) Extrapolated from the data at other temperatures. c)
Weight % (w/w).
ꢁ
extrapolated to those at 50 C to compare with the previous
data.17 The solvolysis rates of the unsubstituted derivative 2
(X ¼ H) were also determined in various solvents at 25 ꢁC
to analyze the solvent effect. All the solvolyses followed accu-
rately a first-order rate law until 2.5 half-lives. Kinetic data are
listed in Tables 1 and 2 together with the previous data.
The solvolysis rates were mostly in good agreement with
those reported before, except for p-MeO and p-MeO-m-Cl de-
rivatives, for which more than 10 times higher rate constants
were reported earlier;17 we could not reproduce these rate con-
stants in repeated measurements in this laboratory in Kyushu
University.
either with dry HCl gas or SOCl2, the complete label scram-
bling of the ethyl moiety occurred, suggesting some involve-
ment of Si-bridged species at least in the chlorination process.
The label scrambling analysis for the solvolyses with three
isotopomers (protio, 1,1-d2, and 2,2-d2 derivatives) of 2-(di-
methylphenylsilyl)ethyl 3,5-dinitrobenzoate was carried out
in CF3CD2OD in the presence of excess 2,6-lutidine at 125
ꢁC, by using H NMR method. The products of these solvoly-
1
ses were found to be ethylene accompanied by the substitution
products of the label scrambling, as summarized in Table 3.
The substitution products, 2,2,2-trifluoroethyl ether, were ob-
tained in 32–40% yield for the 3,5-dinitrobenzoate solvolysis
in this less nucleophilic solvent, while the chloride solvolysis
gave exclusively the elimination product, ethylene, but no sub-
stitution product even in 60E. No d2 scrambling was observed
in the reactant 3,5-dinitrobenzoates but such scrambling was
observed in the product 2,2,2-trifluoroethyl ether, as shown
in Table 3. The unrearranged ether from both the deuterium
compounds was slightly larger than the rearranged product.
Solvent Effect. The solvent effect on the solvolyses of
2-(dimethylphenylsilyl)ethyl chloride at 25 ꢁC (in Table 2)
The solvolysis product of the chlorides (2) was ethylene that
is given by the E1-desilylation from incipient ꢂ-silyl carbo-
cation (Scheme 2) in all cases examined here, as reported in
the literature.16
The solvolysis rates were faster than that of the non-silylat-
ed derivative, ethyl chloride, by ca. 109-fold acceleration by
PhSiMe2 substitution. In this simple primary system (2), there
seems to be the largest ꢂ-silicon effect ever reported.1 Such a
remarkable cation-stabilizing ꢂ-Si effect makes it possible for
the system to undergo solvolysis by the E1 kC mechanism.
Deuterium Scrambling Analysis. In the ordinary chlori-
nation of the ꢄ or ꢂ-d2-labeled 2-(dimethylphenylsilyl)ethanol

1836 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
The ꢂ-Silicon Effect on Solvolysis Rates
X
X
X
X
MeSiMe
MeSiMe
MeSiMe
MeSiMe
or
+
H
H
H
H
H
H
H
H
H
CH2=CH2
H
H
H
Cl
(2)
(2a+)
(2b+)
Scheme 2. Solvolysis of 2-(aryldimethylsilyl)ethyl chlorides 2.
Table 3. Product Analyses on the Solvolyses of 2-(Dimethylphenylsilyl)ethyl 3,5-Dinitrobenzoate and Its
ꢁ
Deuterium Compounds in CF3CD2OD at 125 CaÞ
b)
c)
D d)
OS
e)
(D)H
(D)H
H
H
D
Si
Si
Compounds
Si
OS
OS
D
D
Si
68%
32%
—
—
—
ODNBz
D
D
60%
62%
22%
16%
18%
22%
Si
Si
ODNBz
—
ODNBz
D
D
a) S ¼ CD2CF3 and ODNBz = 3,5-dinitrobenzoate. b) ꢆ 5.47 ppm (quintet, JHD ¼ 2:4 Hz for
CH2=CD2). c) ꢆ 1.27–1.38 ppm (2H, m, SiCH2), 3.80–3.85 ppm (2H, m, CH2). d) ꢆ 1.38 ppm
(2H, s, SiCH2). e) ꢆ 3.78–3.94 ppm (2H, m, CH2).
was analyzed with the Winstein–Grunwald–Schleyer–Bentley
Eq. 2:19
3
2
30E
Me
Si
50Tv
40E
Cl
logðk=k80EÞ ¼ mYCl;
ð2Þ
80Tv
Me
TFE
97Tw
m = 0.73
where YCl is the solvent ionizing power parameter based on
solvolyses of 1-adamantyl chloride, and m is the susceptibility.
Figure 1 shows the linear solvent-effects correlation against
YCl with an m value of 0:73 ꢅ 0:02 without any serious devia-
tion of particular solvent series (R ¼ 0:990 and SD ¼ ꢅ0:18).
The lack of any serious deviation depending on the solvent
nucleophilicity, l ¼ 0, indicates the absence of nucleophilic
attack to either the Si atom or the carbon center in the rate-
determining transition state.
While the systematic dispersion between the respective bi-
nary solvent series in the plot against YCl was slightly observed
in Fig. 1, such dispersal attributable to charge delocalization is
not very significant in this reaction, compared with the serious
dispersion caused by the aryl-ꢃ-delocalization as in benzylic
chloride solvolyses.20 The reduced m value of 0.73 suggests
for this solvolysis a Si-bridged transition state that can be
referred to as the intramolecular nucleophilic participation to
the carbocation center, as observed similarly in the ꢅ-silyl
solvolyses of percaudal mechanism.18
50E
20ET
1
60E
70E
40ET
0
80E
60ET
90E
MeOH
aq.EtOH
aq.Acetone
aq.MeOH
aq.TFE
-1
-2
80ET
EtOH
EtOH-TFE
Cl
-4
-2
0
2
4
YCl
Fig. 1. The mYCl plots for the solvolysis of 2-(dimethyl-
ꢁ
phenylsilyl)ethyl chloride at 25 C.
the nonresonant ꢀ0 or ꢀ reactivity, as indicated in the result
of Vencl.16 In Fig. 2, the logarithms of the relative rates at
50 C were plotted against ꢀ0 (closed circles) and ꢀþ (open
ꢁ
circles). An application of Y–T Eq. 1 affords a good correla-
tion with r ¼ 0:10 (R ¼ 0:997 and SD ¼ ꢅ0:05) for the whole
range of substituents:
Substituent Effect Analysis. The substituent effects on
the solvolysis of 2-(aryldimethylsilyl)ethyl chlorides (2) were
earlier reported to give the aryl substituent effects of ꢁ ¼
ꢂ1:4{ ꢂ2:3 against ꢀ constant.15,16
ꢀ þ
0
logðk=k0Þ ¼ ðꢂ1:75 ꢅ 0:05Þ½ꢀ þ ð0:10 ꢅ 0:01ÞꢁꢀR ꢆ: ð3Þ
The r value of 0.10 assigned for this system may be explic-
able if the phenyl ꢃ-system on the ꢂ-silicon atom interacts
through negligibly small ꢃ-delocalization with the carbocat-
ionic reaction center in the transition state. Employment of
the higher k values for p-MeO and p-MeO-m-Cl substituents
reported earlier17 resulted in a ridiculously high r value of
The rate change with ring substituents in Table 1 was ca.
100-fold for the range of substituent changes from the p-
MeO to 3,5-(CF3)2. The rate of the p-MeO derivative is com-
parable with that of the p-Me derivative, indicating that the
substituent effect on this solvolysis can be characterized as

M. Fujio et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1837
1.5 with unsatisfactory precision.
By the way, as far as we are aware of the ꢂ-aryl-silyl sub-
stituent effects, the ꢂ-silyl systems are, without exception, cor-
The arylsilyl substituent effects were likewise analyzed for
several relevant sets of ꢂ-silyl systems in order to ascertain
significant variations of ꢁ values from system to system; the
correlation results are summarized in Table 4. The rate data
of ꢂ-silyl systems were quoted from the unpublished results
in these laboratories, since no data were available in the liter-
ature other than our file of unpublished results. For all the ꢂ-
silyl systems, Y–T Eq. 1 correlated quite excellently with es-
sentially nonresonant sigmas of negligible resonance demand
related precisely linearly against essentially nonresonant ꢀ0
ꢃ
with r 0:10. This is the strongest support for the validity
¼
of the present rate data for p-MeO and p-MeO-m-Cl deriva-
tives (in Table 1).
Discussion
The so-called ꢂ-silicon effect, ca. 1012-fold rate enhance-
ment, arises because the high polarizability and electropositiv-
ity of silicon enable it to stabilize the electron-deficient inter-
mediate. The overall reaction pathway is a ꢂ-elimination (for-
mation of an olefin by loss of the electrofuge Me3Si group and
nucleofuge chloride ion). The ꢂ-silyl effects due to the tri-
methylsilyl group were observed as remarkable rate accelera-
tion for systems in an antiperiplanar conformation,1 and this
was manifested primarily in an E1-like mechanism, in which
departure of the nucleofuge is rate-determining (Scheme 1).
Overall stabilization of this electron deficiency occurs by ꢀ–
ꢃ hyperconjugation between the highly polarizable C–Si bond
and the empty p-orbital (1a+).
On the other hand, the silyl-bridged pathway was invoked
by Eaborn et al.13 especially for primary carbocation systems
1 (R1, R2 ¼ H). In the silyl-bridged pathway (1b), the silicon
atom forms a partial C–Si bond to the carbon atom from which
the nucleofuge departs. The bridging may be in concert with
nucleofuge departure, so that the reaction is analogous to
neighboring group participation (anchimeric assistance). The
antiperiplanar arrangement is required for the bridged pathway
(1b) to place the internal nucleophile (silicon) backside to the
breaking bond (C–Lg) during formation of the three-members
ring (1b+). Thus, two modes of stabilization differing in the
geometry of the intermediate, the unbridged (vertical) and
the bridged (nonvertical) structures, have been considered:
Both bridged and unbridged models are consistent with the
whole set of evidence.
ꢃ
(r 0:10), to exhibit significant variations of ꢁ from ꢂ1:75
¼
to ꢂ0:96.
0.5
p-MeO
p-Me
m-Me
3,5-Me
2
0.0
-0.5
-1.0
-1.5
-2.0
H
p-MeO-m-Cl
p-Cl
Me
X
m-Cl
Si
CH2CH2Cl
Me
m-CF
3
60E, 50 °C
ρ = -1.75
r = 0.10
+
0
σ
σ
3,5-(CF
)
3 2
σ (r = 0.1)
-1.0
-0.5
0.0
0.5
1.0
σ
Fig. 2. Substituent effect on the solvolysis of 2-(aryldi-
methylsilyl)ethyl chlorides (2) in 60% (v/v) aq EtOH at
50 ꢁC: Closed circles for ꢀ0, open circles for ꢀþ, and open
ꢀ
squares for ꢀ with r ¼ 0:10.
Table 4. Correlation Analysis for 2-(Aryldimethylsilyl)ethyl Solvolyses in 60E Using Y–T Eq. 1
ꢁ
SubstratesaÞ
Temp/ꢁC
ꢁ
r
k(ODNBz,50 C)bÞ
Me
Si
2cÞ
50
ꢂ1:75
0.10
9:55 ꢄ 10ꢂ9
Cl
Me
X
Me
Si
11dÞ
70
70
ꢂ1:77
ꢂ1:54
0.17
0.08
9:95 ꢄ 10ꢂ7
1:63 ꢄ 10ꢂ5
ODNBz
Me
Me
X
X
X
Me
Si
6dÞ
ODNBz
ODNBz
Me
Me
Ph
Me
Me
5dÞ
50
50
ꢂ0:96
ꢂ1:08
0.05
0.15
1:13 ꢄ 10ꢂ2
8:9 ꢄ 10ꢂ1
Si
Me
Me
Si
Me Me
OPNB
4dÞ
Me
X
a) Cl ¼ chloride, ODNBz = 3,5-dinitrobenzoate, and OPNB = p-nitrobenzoate. b) Rate constants
(/sꢂ1) for the 3,5-dinitrobenzoates of X ¼ H derivatives at 50 ꢁC in 60E. c) Present study. d) To
be published in separate papers.

1838 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
The ꢂ-Silicon Effect on Solvolysis Rates
with ꢁ ¼ ꢂ1:75 against essentially nonresonant ꢀ0 constant
(r ¼ 0:10). The precise linearity (R ¼ 0:997) of the correlation
Eq. 3 indicates a constant mechanism over the full range of
substituents. For all the ꢂ-silyl systems in Table 4, linear cor-
relations of high quality (R > 0:994) against essentially non-
Me3
Me3Si
Me3Si
2.20Å
98°
1.37Å
+
Si
2.44Å
Me
Me
+
Me
H
+
2.12Å
H
H
74°
106°
1.36Å
1.39Å
(4Ma+)
0
ꢃ
resonant ꢀ (r 0) are obtained irrespective of structures of
(2Mb+)
(3Ma+)
¼
the ethylene chain, and no significant mechanistic change
should be the case in any system with the silyl substituent
perturbation.
Scheme 3. The ab initio calculation for ꢂ-silyl cations where
R ¼ Me in 1a+ or 1b+.
The substituent effect of 2-(aryldimethylsilyl)ethyl system
bearing 2,2-dimethyl groups (6) has a large ꢁ value of
ꢂ1:54. Most significant, all the primary systems in Table 4
provide essentially the same ꢁ value as that for 2. On the other
hand, the ꢄ-phenyl derivative 5 as well as 1,1-dimethyl sub-
strate 4, where the silyl group assistance is unnecessary be-
cause of the strong stabilization with phenyl or Me2 gave a dis-
tinctly small ꢁ value of ca. ꢂ1:0, as shown in Table 4. The
large ꢁ value of ꢂ1:75 points to the silyl bridged transition
state, while the small ꢁ value of ꢂ1:0 points to the open (un-
bridged) cation transition state.
It is of particular interest to compare such a significant
change in the ꢁ values in the ꢂ-(aryldimethylsilyl)alkyl system
1 (R ¼ X-C6H4) with the change in ꢁ in the corresponding ꢂ-
arylethyl systems, e.g., neophyl tosylate 7 and 2-aryl-1,1-di-
methylethyl chlorides 8. Whereas tertiary reactant (8) under-
goes the kC solvolysis via a stable open carbocation (8a+) as
a single intermediate, the neophyl one (7) does not undergo
the kC solvolysis to form an unstable intermediate 7a+. The
extremely unstable primary cation 7a+ does not compete to
act as an appropriate intermediate with other stable cations,
7ꢁ and 8a+. As the solvolysis of 7 gave exclusively the rear-
ranged product(s) that is/are identical to the kC solvolysis
product(s) from 8a+, an appropriate intermediate of the solvol-
ysis of 7 should be a rearranged intermediate 8a+. Thus the
solvolysis of 7 does not proceed through the kC pathway to-
ward an unstable intermediate 7a+ but proceeds by the aryl-
migration (kꢁ) pathway to give a pseudo-intermediate cation
7ꢁ that immediately cascades down to the more stable isomer-
ic open cation (8a+) in Scheme 4.
It was well documented that the aryl-assisted correlation
(ꢁ ¼ ꢂ3:8; r ¼ 0:6) could be observed for 714b and that the
unassisted correlation (ꢁ ¼ ꢂ2:1; r ¼ 0) was obtained for
8,21 respectively. High ꢁꢁ values of ꢂ3:5–ꢂ3:8 were general-
ly observed for the kꢁ processes, but lower values of ca. ꢂ2
for the aryl-unassisted kC processes. The enhanced ꢁ value
for the kꢁ pathway is ascribed to the bond-contraction by di-
rect bonding between aryl ring and the reaction center in the
transition state.14b
In the corresponding ꢂ-silyl systems, ArSiMe2–CH2CMe2-
(4) solvolyzes undoubtedly by the vertical pathway (4a) to
form an open cation (4a+). In the case of ArSiMe2–CMe2CH2-
(6), on the other hand, MO calculation indicates significantly
The ab initio calculation gave the following structures
(Scheme 3) optimized at MP2/6-31G(d).8 The progressive
substitution with Me-groups at the reaction center carbon re-
sults in the increase of the C –C distance, decrease of C –
ꢄ
ꢂ
ꢂ
Si distance, and increase of C –C –Si angle. Furthermore,
ꢄ
ꢂ
the optimized structure of ꢄ-phenyl cation (5Ma+) shows
the larger C –C –Si angle, indicating the most open cation
ꢄ
form among the calculated systems.
ꢂ
Jorgensen’s calculations found that the bridged structure
(1b+) was more stable than the open structure (1a+) for the
primary case, and that the open form was not an energy mini-
mum.4b In the secondary case, however, the open form was
found to be slightly more stable than the bridged form, and
in this case the bridged form was not an energy minimum.
Nevertheless, for 1,2-dimethyl-2-silylethyl cation, the bridged
form is slightly more stable.8 Thus, direct calculation did not
provide a clear answer. Moreover, theory addresses only the
fully developed carbocation, whereas the actual transition state
has only a partial positive charge and an incompletely departed
leaving group. Anyhow, the actual involvement of structure
(1a+) vs (1b+) is hard to prove.
For the most fundamental case, the primary ꢂ-SiMe3-ethyl
system (2M), MO calculation indicates that the unbridged cat-
ion (2Ma+) is not an energy minimum in the potential surface
but immediately leaks to the bridged intermediate (2Mb+). It
appears therefore highly plausible that the reaction in Scheme 2
proceeds by the bridged pathway (2b) toward the more stable
bridged cation (2b+) but not by the unbridged pathway (2a) to-
ward the corresponding vertical (unbridged) cation (2a+). At
this symmetrical intermediate (2b+), the substrate loses the
distinction between ꢄ- and ꢂ-carbons. The solvolyses (2M:
Lg = bromide) in acetic acid accompanied the return to the re-
actant from the carbocation intermediate with the d2 scram-
bling in the ethyl moiety.13 This means that the departure of
the nucleofuge occurs before the departure of the electrofuge
silyl moiety. It should be noted that the solvolysis of 2 yields
the substitution products besides the commonly occurring
silyl-eliminated product (ethylene). The solvolysis in TFE
(Lg = 3,5-dinitrobenzoate) resulted in the substitution product
2,2,2-trifluoroethyl ether with d2 scrambling in Table 3. The d2
scrambling in the ethyl chain is consistent with the bridged cat-
ion (2b+) as the key intermediate in the primary cation system.
The solvent effects indicate that the nucleophilic solvent attack
on either the silyl or the cationic center carbon does not take
place at the rate-determining transition state. The migration
of the silyl moiety had to exceed the solvent capture of the
cation center.
different stabilities of the isomeric silyl-cations; 6a+
ꢇ
6b+ < 4a+. The vertical solvolysis of 6 should not proceed to-
ward an extremely unstable primary cation (6a+) that is pre-
sumably incapable of existing as an intermediate. The solvol-
ysis of 6 may be rationalized to proceed through the nonvert-
ical pathway (6b) to form a Si-bridged cation (6b+), which im-
mediately cascades downhill toward the more stable tertiary
The substituent effect on the solvolysis of 2-(aryldimethyl-
silyl)ethyl chlorides (2) in Fig. 2 gave a linear Y–T correlation

M. Fujio et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1839
X
X
Me CH C
CMe CH
2
2
2
2
H
H
X
X
OTs
Me
Me
Cl
Me
Me
H
H
+
7
7
∆
8a
8
Scheme 4. Solvolysis of 2-phenylethyl system bearing 2,2- or 1,1-dimethyl substituents.
Me Ar
2
Si
SiMe Ar
2
Me CH C
ArMe Si
2
SiMe Ar
2
CMe CH
2
2
2
2
H
H
Me
Me
H
H
OPNB
Me
Me
ODNBz
+
+
6
6b
4a
4
Scheme 5. Solvolysis of 2-silylethyl system bearing 2,2- or 1,1-dimethyl substituents.
SiMe
3
OTFA
SiMe
Me Si
3
3
Ar
H
CHCH
CHCH
Cl
2
2
H
H
H
H
X
H
X
+
SiMe
3
Ar
5Ma
+
9b
9
5M
X
SiMe
3
+
X
SO
Scheme 6. Solvolyses of ꢂ-silylethyl system bearing ꢄ- and ꢂ-phenyl groups.
cation (4a+) (Scheme 5).
ate, the structures of transition state may be manifested by the
continuous shift of the ꢂ-Si moiety from sp3 position (open
cation form) through the nonvertical (bridged) to the rear-
ranged vertical position within the antiperiplanar phase. The
kinetic parameters will be promising as a probe for discrimi-
nating the structures of intermediates. Indeed, significantly dif-
ferent ꢁ values for ꢂ-silyl substituent effects on solvolyses
were found for substrates of varying ethylene moieties. We
then believe that the ꢁ value should be different for systems
that proceed through the rate-determining transition states of
vertical (open) ion structure or the nonvertical (bridged) ꢂ-sil-
yl cation structure. It is worthy of note that the substituent ef-
fect on the solvolysis through the percaudal interaction of ꢅ-
silyl-ester (10) also gave a correlation with ꢁ ¼ ꢂ1:0 and
We showed before that the ꢄ- and ꢂ-phenyl derivatives of
ꢂ-trimethylsilyl series 1 (where R ¼ Me) gave on solvolysis
the same products as those which were derived from the open
cation intermediate (5Ma+) in the vertical solvolysis of the ꢄ-
phenyl system (5M).5b,5c
The products indicated that the solvolysis of 9 does not pro-
ceed by the unbridged pathway toward the open primary cation
9a+ but by the bridged pathway toward the silyl-migrated cat-
ion (5Ma+) as an appropriate intermediate. Thus, the rate-de-
termining transition state (9z) has to be a nonvertical (bridged)
structure (Scheme 6).5b
Unfortunately, however, both 6 and 4 yielded 2-methylpro-
pene as a sole product that is incapable of discriminating rear-
ranged and unrearranged intermediates. Most of ꢂ-silylalkyl
substrates likewise yielded only desilylated olefins. The same
is also true for the case of the secondary substrate that is under-
stood to proceed by the vertical mechanism. Besides MO cal-
culations, there is no chemical evidence to discriminate an ap-
propriate intermediate among three possible cations; vertical
(nonmigrated) cation (2a+), nonvertical (bridged) one (2b+),
and vertical isomeric (migrated) cation. It is therefore remark-
able that a quite significant difference was observed in the ꢁ
value for ꢂ-silyl substituent effect between the primary 6
and tertiary substrate 4.
As briefly discussed above, the significant enhancement in
the ꢂ-silicon effects observed in solvolysis are essentially ki-
netic in nature, whereas the remarkable ꢂ-silyl participations
in most of preceding studies were related exclusively to the
thermodynamic stabilities of the intermediates. While the
structure of the transition state resembles that of the intermedi-
18
ꢃ
r
0, that is expected for the transition state where the
¼
aryl-silyl function is attached at the ꢂ-position to the carbo-
cation center.
In order to establish the use of selectivity (ꢁ) parameters as
a sensitive probe for the ꢂ-silyl participation, we have to ob-
tain more data about ꢂ-(arylsilyl) substituent effects in a series
of relevant ꢂ-silylalkyl systems bearing substituents in the
ethyl chain (R1–R4 in 1).
Experimental
Column chromatography was performed by using SiO2 (Silica
gel 60 (230–400 mesh, Merck) for flash column chromatography
or LiChroprep Si 60 (25–40 mm, Merck) for middle-pressure liq-
1
uid chromatography). The H NMR spectra were taken in CDCl3
on a JEOL JNM-A500 FT-NMR spectrometer operating at 500
MHz and the chemical shifts were recorded in ppm (ꢆ) downfield
from TMS as an internal standard. All air- and moisture-sensitive

1840 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
The ꢂ-Silicon Effect on Solvolysis Rates
reactions were carried out under nitrogen or argon. Ether and
tetrahydrofuran were distilled from sodium/benzophenone under
nitrogen.
0.40 (6H, s, SiCH3), 1.16 (3H, t, J ¼ 7:0 Hz, CH3), 2.10 (2H, s,
CH2), 4.04 (2H, q, J ¼ 7:0 Hz, OCH2), 7.28–7.41 (4H, m, Ar-H).
Ethyl [3,5-Bis(trifluoromethyl)phenyldimethylsilyl]acetate:
1H NMR ꢆ 0.48 (6H, s, SiCH3), 1.14 (3H, t, J ¼ 7:0 Hz, CH3),
2.16 (2H, s, CH2), 4.03 (2H, q, J ¼ 7:0 Hz, OCH2), 7.89 (1H, s,
Ar-H), 7.95 (2H, s, Ar-H).
2-(Dimethylphenylsilyl)ethanol. Ethyl (dimethylphenylsilyl)-
acetate was reduced by LiAlH4 in ether with reflux overnight. Af-
ter the usual work-up, the crude alcohol was purified by column
chromatography. Colorless oil: 1H NMR ꢆ 0.30 (6H, s, SiCH3),
1.20 (2H, t, J ¼ 8:5 Hz, SiCH2), 3.74 (2H, t, J ¼ 8:5 Hz, CH2),
7.33–7.52 (5H, m, Ph-H). Found: C, 66.54; H, 9.01%. Calcd for
C10H16OSi: C, 66.61; H, 8.94%. The following ring substituted
derivatives were obtained in the same way.
Material
[(3,5-Bis(trifluoromethyl)phenyl)dimethylsilyl]-
methyl Chloride. A solution of 3,5-bis(trifluoromethyl)phenyl-
magnesium bromide prepared from the 3,5-bis(trifluoromethyl)-
bromobenzene and Mg in THF was added dropwise to chloro-
(chloromethyl)dimethylsilane in THF with stirring at 0 ꢁC and
the reaction mixture was heated under reflux with stirring over-
night under nitrogen atmosphere, according to the method of
Vencl et al.16 After cooling, the reaction mixture was treated with
aqueous NH4Cl, and extracted with ether. The ethereal extract was
washed with aqueous sodium hydrogencarbonate and aqueous sat-
urated sodium chloride, and then dried over anhydrous magnesi-
um sulfate. The crude chloride was purified by vacuum distillation
(bp 71–74 ꢁC/1.5 mmHg (1 mmHg = 133.322 Pa)) to give a
colorless oil (yield, 50%): 1H NMR ꢆ 0.50 (6H, s, SiCH3), 2.98
(2H, s, CH2), 7.90 (1H, s, Ar-H), 7.95 (2H, s, Ar-H).
2-(p-Methoxyphenyldimethylsilyl)ethanol:
Colorless oil:
1H NMR ꢆ 0.28 (6H, s, SiCH3), 1.17 (2H, t, J ¼ 8:5 Hz, SiCH2),
3.73 (2H, t, J ¼ 8:5 Hz, CH2), 3.81 (3H, s, OCH3), 6.91 (2H, d,
J ¼ 8:5 Hz, Ar-H), 7.43 (2H, d, J ¼ 8:5 Hz, Ar-H). Found: C,
62.64; H, 8.62%. Calcd for C11H18O2Si: C, 62.81; H, 8.63%.
2-[(3-Chloro-4-methoxyphenyl)dimethylsilyl]ethanol: Col-
orless oil: 1H NMR ꢆ 0.28 (6H, s, SiCH3), 1.17 (2H, t, J ¼ 8:5
Hz, SiCH2), 3.72 (2H, t, J ¼ 8:5 Hz, CH2), 3.90 (3H, s, OCH3),
6.93 (1H, d, J ¼ 7:5 Hz, Ar-H), 7.34 (1H, dd, J ¼ 7:5 Hz, J ¼
1:5 Hz, Ar-H), 7.46 (1H, d, J ¼ 1:5 Hz, Ar-H). Found: C,
53.98; H, 7.04%. Calcd for C11H17ClO2Si: C, 53.97; H, 7.00%.
Other (aryldimethylsilyl)methyl chlorides were synthesized
from arylmagnesium bromides and chloro(chloromethyl)dimeth-
ylsilane in ether as reported before.18
Ethyl (Dimethylphenylsilyl)acetate. Ethyl (dimethylphenyl-
silyl)acetate was synthesized by the Reformatsky reaction from
the chlorodimethylphenylsilane and ethyl bromoacetate, accord-
ing to the method of Vencl et al.16 The mixture of chlorodimethyl-
phenylsilane (5.16 g, 30.2 mmol) and ethyl bromoacetate (4.95 g,
29.6 mmol) in THF (10 mL) solution was added dropwise to Zn
(1.94 g, 29.7 mmol) in THF 30 mL under reflux and the mixture
was maintained under reflux until Zn completely dissolved. Then,
10% HCl 20 mL was added to the cold reaction mixture, which
was extracted with ether. Ethereal extract was washed with aq
sat NH4Cl, aq NaHCO3, and aq sat NaCl, and then dried over an-
hydrous MgSO4. Purification by column chromatography afforded
3.35 g (yield, 50%) of the acetate as a colorless oil: 1H NMR ꢆ
0.40 (6H, s, SiCH3), 1.15 (3H, t, J ¼ 7:0 Hz, CH3), 2.10 (2H, s,
CH2), 4.03 (2H, q, J ¼ 7:0 Hz, OCH2), 7.34–7.54 (5H, m, Ph-H).
2-(p-Chlorophenyldimethylsilyl)ethanol:
Colorless oil:
1H NMR ꢆ 0.30 (6H, s, SiCH3), 1.18 (2H, t, J ¼ 8:0 Hz, SiCH2),
3.73 (2H, t, J ¼ 8:0 Hz, CH2), 7.33 (2H, d, J ¼ 8:0 Hz, Ar-H),
7.43 (2H, d, J ¼ 8:0 Hz, Ar-H). Anal. Found: C, 56.28; H,
7.04%. Calcd for C10H15ClOSi: C, 55.93; H, 7.04%.
2-(m-Chlorophenyldimethylsilyl)ethanol:
Colorless oil:
1H NMR ꢆ 0.31 (6H, s, SiCH3), 1.19 (2H, t, J ¼ 8:0 Hz, SiCH2),
3.74 (2H, t, J ¼ 8:0 Hz, CH2), 7.29–7.52 (4H, m, Ar-H). Found:
C, 55.86; H, 6.93%.
2-[3,5-Bis(trifluoromethyl)phenyldimethylsilyl]ethanol: Col-
orless oil: 1H NMR ꢆ 0.34 (6H, s, SiCH3), 1.21 (2H, t, J ¼ 8:0 Hz,
SiCH2), 3.78 (2H, t, J ¼ 8:0 Hz, CH2), 7.84 (1H, s, Ar-H), 7.91
(2H, s, Ar-H). Found: C, 45.75; H, 4.59%. Calcd for
C12H14F6OSi: C, 45.56; H, 4.46%.
Ethyl (p-Methoxyphenyldimethylsilyl)acetate.
To a
Grignard solution prepared from (p-methoxyphenyldimethyl-
silyl)methyl chloride (11 g, 51 mmol) and Mg (1.24 g, 51 mmol)
in 70 cm3 of ether was added dropwise a solution of ethyl chloro-
formate (5.53 g, 51 mmol) in 60 cm3 of ether with stirring at 0 ꢁC,
according to the method of Vencl et al.16 After stirring overnight,
the reaction mixture was treated with aq NH4Cl, and extracted
with ether. The ethereal extract was washed with aq NaHCO3
and aq sat NaCl, and then dried over anhydrous MgSO4. Purifica-
tion by column chromatography afforded 3.57 g (yield, 27%) of
2-(Dimethylphenylsilyl)ethan-1,1-d2-ol.
Ethyl (dimethyl-
phenylsilyl)acetate was reduced by LiAlD4 (98%D) in ether in
the same way as the protio isotopomer. After the usual work-up,
the crude alcohol was purified by column chromatograpy. Color-
less oil: 1H NMR ꢆ 0.30 (6H, s, SiCH3), 1.18 (2H, s, SiCH2),
7.33–7.52 (5H, m, Ph-H).
2-(Dimethylphenylsilyl)ethan-2,2-d2-ol. To a stirred solution
of lithium diisopropylamide prepared from diisopropylamine (1.9
cm3, 13.5 mmol) and 1.6 M (M = mol dmꢂ3) butyllithium hexane
solution (9.0 cm3, 13.5 mmol) in 15 cm3 of THF at 0 ꢁC was add-
ed dropwise a solution of ethyl 2-(dimethylphenylsilyl)acetate
1
the ester as a colorless oil: H NMR ꢆ 0.37 (6H, s, SiCH3), 1.17
(3H, t, J ¼ 7:5 Hz, CH3), 2.08 (2H, s, CH2), 3.81 (3H, s, OCH3),
4.03 (2H, q, J ¼ 7:5 Hz, OCH2), 6.92 (2H, d, J ¼ 8:5 Hz, Ar-H),
7.46 (2H, d, J ¼ 8:5 Hz, Ar-H). By the same procedure, the fol-
lowing acetates were prepared from the corresponding (aryldi-
methylsilyl)methyl chlorides.
ꢁ
(2.00 g, 9.0 mmol) in 5 cm3 of THF at ꢂ70 C. After stirring at
ꢁ
ꢂ70 C for 10 min, D2O (0.37 g, 18.0 mmol) in 5 cm3 of THF
ꢁ
Ethyl [(3-Chloro-4-methoxyphenyl)dimethylsilyl]acetate:
1H NMR ꢆ 0.38 (6H, s, SiCH3), 1.17 (3H, t, J ¼ 7:5 Hz, CH3),
2.08 (2H, s, CH2), 3.91 (3H, s, OCH3), 4.04 (2H, q, J ¼ 7:5 Hz,
OCH2), 6.94 (1H, d, J ¼ 8:0 Hz, Ar-H), 7.37 (1H, dd, J ¼ 8:0
Hz, J ¼ 1:5 Hz, Ar-H), 7.48 (1H, d, J ¼ 1:5 Hz, Ar-H).
was added quickly. After stirring at ꢂ70 C for 1 h, the reaction
mixture was kept at room temperature with stirring for 2 h, and
then quenched with 40 cm3 of water. The aqueous layer was sep-
arated from the organic layer, and then extracted with ether. The
combined organic layer was washed with aq NH4Cl, aq NaHCO3,
and sat NaCl, and dried over anhydrous MgSO4. Purification by
column chromatography afforded 1.0 g (yield, 83%) of the 2-d
ester as a colorless oil: 1H NMR ꢆ 0.40 (6H, s, SiCH3), 1.16
(3H, t, J ¼ 7:0 Hz, CH3), 2.11 (1H, s, CHD), 4.03 (2H, q, J ¼
7:0 Hz, CH2), 7.34–7.54 (5H, m, Ph-H). By the same procedure,
Ethyl (p-Chlorophenyldimethylsilyl)acetate:
1H NMR ꢆ
0.39 (6H, s, SiCH3), 1.16 (3H, t, J ¼ 7:0 Hz, CH3), 2.09 (2H, s,
CH2), 4.03 (2H, q, J ¼ 7:0 Hz, OCH2), 7.35 (2H, d, J ¼ 8:5
Hz, Ar-H), 7.46 (2H, d, J ¼ 8:5 Hz, Ar-H).
Ethyl (m-Chlorophenyldimethylsilyl)acetate:
1H NMR ꢆ

M. Fujio et al.
Bull. Chem. Soc. Jpn., 78, No. 10 (2005) 1841
2-d ester was converted to 2,2-d2 ester, which has 94%D content:
1H NMR ꢆ 0.40 (6H, s, SiCH3), 1.16 (3H, t, J ¼ 7:0 Hz, CH3),
4.03 (2H, q, J ¼ 7:0 Hz, OCH2), 7.34–7.54 (5H, m, Ph-H). The
2,2-d2 ester was reduced with LiAlH4 in ether to 2-(dimethyl-
phenylsilyl)ethanol-2,2-d2 as the same way as used for protio
derivative: 1H NMR ꢆ 0.30 (6H, s, SiCH3), 3.74 (2H, s, CH2),
7.35–7.52 (5H, m, Ph-H).
Kinetic Measurements. The solvolysis rates of 2-(dimethyl-
phenylsilyl)ethyl chlorides were determined conductimetrically as
reported previously18 at initial concentrations of 10ꢂ5–10ꢂ4 M.
Using conductivity meters (Toa Electronics Ltd: CM-50AT,
CM-40S, and CM-60S, equipped with an interval time unit and
printer, or CM-60V, CM-40V, and CM-40G, connected to a per-
sonal computer) we followedꢁthe solvolyses in a thermostatted
bath controlled within ꢅ0:01 C by taking at least 100 readings
at appropriate intervals during 2.5 half-lives, and an infinity read-
ing after 10 half-lives. The experimental errors in respective runs
were generally less than 1.0% and the reproducibility of the rate
constants was within ꢅ1:5%.
2-(Dimethylphenylsilyl)ethyl Chloride. The alcohol (1.9 g,
10 mmol) in 15 cm3 of ether was chlorinated with SOCl2 (7.7
ꢁ
cm3, 105 mmol) at 0 C. Excess SOCl2 was completely removed
by the distillation with fresh dry benzene added under the reduced
pressure. The crude chloride was purified by distillation using
Kugelrohr distillation (bp 79–81 ꢁC/2 mmHg). Colorless oil:
1H NMR ꢆ 0.33 (6H, s, SiCH3), 1.42–1.46 (2H, m, SiCH2),
3.60–3.64 (2H, m, CH2), 7.34–7.54 (5H, m, Ph-H).
2-(p-Methoxyphenyldimethylsilyl)ethyl Chloride: Colorless
oil (bp 120–125 ꢁC/4 mmHg): 1H NMR ꢆ 0.26 (6H, s, SiCH3),
1.37–1.40 (2H, m, SiCH2), 3.56–3.60 (2H, m, CH2Cl), 3.77
(3H, s, OCH3), 6.85–7.34 (4H, m, Ar-H).
2-[(3-Chloro-4-methoxyphenyl)dimethylsilyl]ethyl Chloride:
Colorless oil (bp 115–117 ꢁC/2 mmHg): 1H NMR ꢆ 0.31 (6H,
s, SiCH3), 1.40–1.44 (2H, m, SiCH2), 3.59–3.62 (2H, m, CH2),
3.91 (3H, s, OCH3), 6.90–7.43 (3H, m, Ar-H).
Product Analyses. Product analyses for solvolyses of the
chlorides under the buffered condition with excess 2,6-lutidine
were carried out using 1H NMR in situ according to the same pro-
cedure as described before,18 providing the exclusive formation of
CH2=CH2 in the deuterated solvents of aq ethanol, aq acetone,
and 2,2,2-trifluoroethanol.
A solution (6 ꢄ 10ꢂ3 M) of the 3,5-dinitrobenzoate buffered
with 2 equiv 2,6-lutidine in deuterated 2,2,2-trifluoroethanol
(CF3CD2OD) was allowed to react at 125 ꢁC in a fused NMR tube
and the solvolysis products at ten half-lives were identified by
the 1H NMR spectra. The relative amounts of reaction products,
ethylene and 2-(dimethylphenylsilyl)ethyl 20,20,20-trifluoroethyl-
10,10-d2 ether, were determined from the integral areas of the cor-
responding peaks, as summarized in Table 3.
2-(p-Chlorophenyldimethylsilyl)ethyl Chloride: Colorless
oil (bp 110–113 ꢁC/3 mmHg): 1H NMR ꢆ 0.32 (6H, s, SiCH3),
1.41–1.45 (2H, m, SiCH2), 3.58–3.62 (2H, m, CH2), 7.35 (2H,
d, J ¼ 8:5 Hz, Ar-H), 7.41 (2H, d, J ¼ 8:5 Hz, Ar-H).
2-(m-Chlorophenyldimethylsilyl)ethyl Chloride: Colorless
oil: 1H NMR ꢆ 0.34 (6H, s, SiCH3), 1.43–1.46 (2H, m, SiCH2),
3.59–3.63 (2H, m, CH2), 7.32–7.54 (4H, m, Ar-H).
References
1
a) J. B. Lambert, G. T. Wang, R. B. Finzel, and D. H.
2-[3,5-Bis(trifluoromethyl)phenyldimethylsilyl]ethyl Chlo-
1
Teramura, J. Am. Chem. Soc., 109, 7838 (1987). b) J. B. Lambert
and E. C. Chelius, J. Am. Chem. Soc., 112, 8120 (1990); J. B.
Lambert, R. W. Emblidge, and S. Malany, J. Am. Chem. Soc.,
115, 1317 (1993).
ride: Colorless oil: H NMR ꢆ 0.43 (6H, s, SiCH3), 1.50 (2H,
t, J ¼ 8:5 Hz, SiCH2), 3.63 (2H, t, J ¼ 8:5 Hz, CH2), 7.88–7.95
(3H, m, Ar-H).
2-(Dimethylphenylsilyl)ethyl 3,5-Dinitrobenzoate. To a stir-
red solution of 2-(dimethylphenylsilyl)ethanol (0.300 g, 1.66
mmol) in 3 cm3 of ether with pyridine (2.0 cm3) was slowly added
3,5-dinitrobenzoyl chloride (0.384 g, 1.66 mmol) in 10 cm3 ether
2
Y. Apeloig, ‘‘The Chemistry of Organic Silicon Com-
pounds,’’ ed by S. Patai and Z. Rappoport, John Wiley & Sons,
Chichester (1989), Chap. 2, p. 57, and references cited therin.
3
a) J. B. Lambert, Tetrahedron, 46, 2677 (1990); N.
ꢁ
at 0 C. After stirring overnight, the reaction mixture was treated
Shimizu, Rev. Heteroat. Chem., 9, 155 (1993). b) J. B. Lambert,
Y. Zhao, R. W. Emblidge, L. A. Salvador, X. Liu, J.-H. So, and
E. C. Chelius, Acc. Chem. Res., 32, 183 (1999).
with 10 cm3 of cold water under cooling and then extracted with
ether. After the material was washed with dil. HCl, aq NaHCO3,
and sat. NaCl, and dehydrated over anhydrous MgSO4, the solvent
was evaporated under reduced pressure to give the crude 3,5-dini-
trobenzoate, which was purified by column chromatography and
recrystallized with hexane–AcOEt to yield 0.42 g (yield 67%)
as needles, mp 79–80.5 ꢁC: 1H NMR ꢆ 0.39 (6H, s, SiCH3),
1.45 (2H, t, J ¼ 8:5 Hz, SiCH2), 4.55 (2H, t, J ¼ 8:5 Hz, CH2),
7.37–7.52 (5H, m, Ph-H), 9.03 (2H, d, J ¼ 2:0 Hz, Ar-H), 9.18
(1H, t, J ¼ 2:0 Hz, Ar-H). Found: C, 54.78; H, 4.94; N, 7.49%.
Calcd for C17H18N2O6Si: C, 54.53; H, 4.85; N, 7.48%. In the
same way, the following deuterated esters were prepared.
4
a) J. M. White and G. B. Robertson, J. Org. Chem., 57,
4638 (1992); Y. Apeloig, M. Karme, A. Stanger, H. Schwarz,
T. Drewellow, and G. Czekay, J. Chem. Soc., Chem. Commun.,
1987, 989. b) S. G. Wierschke, J. Chandrasekhar, and W. L.
Jorgensen, J. Am. Chem. Soc., 107, 1496 (1985); K. A. Nguyen,
M. S. Gordon, G. Wang, and J. B. Lambert, Organometallics,
10, 2798 (1991).
5
a) N. Shimizu, S. Watanabe, and Y. Tsuno, Bull. Chem.
Soc. Jpn., 64, 2249 (1991); N. Shimizu, C. Kinoshita, E. Osajima,
F. Hayakawa, and Y. Tsuno, Bull. Chem. Soc. Jpn., 64, 3280
(1991). b) F. Hayakawa, S. Watanabe, N. Shimizu, and Y.
Tsuno, Bull. Chem. Soc. Jpn., 66, 153 (1993). c) N. Shimizu, S.
Watanabe, F. Hayakawa, S. Yasuhara, Y. Tsuno, and T. Inazu,
Bull. Chem. Soc. Jpn., 67, 500 (1994).
2-(Dimethylphenylsilyl)ethyl-1,1-d2
3,5-Dinitrobenzoate:
mp 78–80 ꢁC: 1H NMR ꢆ 0.38 (6H, s, SiCH3), 1.43 (2H, s,
SiCH2), 7.27–7.69 (5H, m, Ph-H), 9.03 (2H, d, J ¼ 2:0 Hz, Ar-
H), 9.18 (1H, t, J ¼ 2:0 Hz, Ar-H).
2-(Dimethylphenylsilyl)ethyl-2,2-d2
3,5-Dinitrobenzoate:
6
2729 (1999).
J. B. Lambert, Y. Zhao, and H. Wu, J. Org. Chem., 64,
ꢁ
1
mp 77–78.5 C: H NMR ꢆ 0.38 (6H, s, SiCH3), 4.53–4.56 (2H,
m, CH2), 7.27–7.69 (5H, m, Ph-H), 9.03 (2H, d, J ¼ 2:0 Hz,
Ar-H), 9.18 (1H, t, J ¼ 2:0 Hz, Ar-H).
7
8
X. Li and J. A. Stone, J. Am. Chem. Soc., 111, 5586 (1989).
H.-U. Siehl, et al., unpublished results from these laborato-
Solvents. Solvents were purified and binary solvents prepared
by mixing appropriate volumes or weights of pure solvents at 25
ꢁC, as previously described.18
ries. Full data will be reported in a separate paper.
M. Mishima, C. H. Kang, M. Fujio, and Y. Tsuno, Chem.
Lett., 1992, 2439.
9

1842 Bull. Chem. Soc. Jpn., 78, No. 10 (2005)
The ꢂ-Silicon Effect on Solvolysis Rates
10 B. Chiavarino, M. E. Crestoni, and S. Fornarini, J. Am.
Chem. Soc., 120, 1523 (1998).
11 R. A. Cox, ‘‘Organic Reaction Mechanisms 1997,’’ ed by
A. C. Knipe and W. E. Watts, John Wiley (2001), Chap. 9,
pp. 279–281.
12 A. W. P. Jarvie, A. Holt, and J. Thompson, J. Chem. Soc.
B, 1969, 852; A. W. P. Jarvie, A. Holt, and J. Thompson, J. Chem.
Soc. B, 1970, 746.
13 M. A. Cook, C. Eaborn, and D. R. M. Walton, J. Organo-
met. Chem., 24, 301 (1970).
14 a) Y. Yukawa and Y. Tsuno, Bull. Chem. Soc. Jpn., 32, 971
(1959); Y. Yukawa, Y. Tsuno, and M. Sawada, Bull. Chem. Soc.
Jpn., 39, 2274 (1966). b) Y. Tsuno and M. Fujio, Chem. Soc. Rev.,
25, 129 (1996); Y. Tsuno and M. Fujio, Adv. Phys. Org. Chem.,
32, 267 (1999).
15 L. H. Sommer and G. A. Baughman, J. Am. Chem. Soc.,
83, 3346 (1961).
16 J. Vencl, K. J. Hetflejs, J. Cermak, and V. Chvalovsky,
Collect. Czech. Chem. Commun., 38, 1248 (1973); J. Vencl, K.
J. Hetflejs, P. Kucera, J. Cermak, and V. Chvalovsky, Collect.
Czech. Chem. Commun., 38, 1256 (1973).
5907 (1993).
18 T. Nakashima, R. Fujiyama, M. Fujio, and Y. Tsuno, Bull.
Chem. Soc. Jpn., 72, 741 (1999); T. Nakashima, R. Fujiyama, M.
Fujio, and Y. Tsuno, Bull. Chem. Soc. Jpn., 72, 1043 (1999); T.
Nakashima, R. Fujiyama, H.-J. Kim, M. Fujio, and Y. Tsuno, Bull.
Chem. Soc. Jpn., 73, 429 (2000).
19 E. Grunwald and S. Winstein, J. Am. Chem. Soc., 70, 846
(1948); F. L. Schadt, T. W. Bentley, and P. v. R. Schleyer, J. Am.
Chem. Soc., 98, 7667 (1976); T. W. Bentley and G. E. Carter,
J. Am. Chem. Soc., 104, 5741 (1982); T. W. Bentley and G.
Llewellyn, Prog. Phys. Org. Chem., 17, 121 (1990).
20 A. H. Fainberg and S. Winstein, J. Am. Chem. Soc., 79,
1597 (1957); A. H. Fainberg and S. Winstein, J. Am. Chem.
Soc., 79, 1608 (1957); S. Winstein, A. H. Fainberg, and E.
Grunwald, J. Am. Chem. Soc., 79, 4146 (1957); C. A. Bunton,
M. M. Mhala, and J. R. Moffat, J. Org. Chem., 49, 3639
(1984); M. Fujio, M. Goto, T. Yoshino, K. Funatsu, Y. Tsuji, S.
Ouchi, and Y. Tsuno, Mem. Fac. Sci., Kyushu Univ., Ser. C, 16,
85 (1987); T. W. Bentley, I. S. Koo, and J. Norman, J. Org.
Chem., 56, 1604 (1991).
21 M. Fujio, et al., unpublished results.
17 R. Fujiyama and T. Munechika, Tetrahedron Lett., 34,