Systematic repression of β-silyl carboeation stabilization

Article abstract of DOI:10.1021/jo802722z

Solvolysis of 1-(trimethylsilylmethyl)cyclopropyl mesylate in CD ₃CO₂D gives ring-opened products as well as methylenecyclopropane. The rate enhancement due to the β-trimethylsilyl group is a factor of about 10⁶. The large

Full text of DOI:10.1021/jo802722z

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
Xavier Creary* and Elizabeth D. Kochly
Department of Chemistry and Biochemistry, UniVersity of Notre Dame, Notre Dame, Indiana 46556
creary.1@nd.edu
ReceiVed December 11, 2008
Solvolysis of 1-(trimethylsilylmethyl)cyclopropyl mesylate in CD3CO2D gives ring-opened products as
well as methylenecyclopropane. The rate enhancement due to the ꢀ-trimethylsilyl group is a factor of
about 106. The large stabilizing effect of a ꢀ-silyl group (which can cause rate enhancements of up to
1012) on the intermediate cation has been repressed. B3LYP/6-31G* computational studies indicate a
carbocation stabilization energy of 16.6 kcal/mol. Rates of solvolyses of 1-phenyl-2-trimethylsilylcyclo-
propyl chlorides are enhanced by a factor of 103-104. The intermediate cyclopropyl cation undergoes
substantial ring opening since ꢀ-silyl stabilization is not large (calculated stabilization energy of 12 kcal/
mol). Solvolysis rates of 2-trimethylsilylbenzocyclobutyl derivatives are not significantly enhanced by
the ꢀ-trimethylsilyl group. ꢀ-Silyl stabilization of benzocyclobutenyl carbocations generated in solution
has been effectively eliminated due to antiaromatic considerations (calculated stabilization energy of 3.7
kcal/mol when R ) Ph). While computational studies parallel solvolytic rate studies, they overestimate
the extent of ꢀ-trimethylsilyl stabilization of solvolytically generated carbocations.
Introduction
that are destabilized due to bond angle effects.2 Both of these
types of carbocations have been studied in much detail. On the
other side of the general stability spectrum lie carbocations of
general structure 3, where stabilization by adjacent silicon is
quite dramatic.3,4 This phenomenon has also been extensively
Features that stabilize and destabilize reactive intermediates
have fascinated chemists over the years. We have been heavily
involved in the study of carbocation reactive intermediates that
are both stabilized and destabilized by electronic factors.
Carbocations 1, where the group E is formally electron-
withdrawing, represent so-called “destabilized carbocations”,1
while cyclopropyl cations 2 represent a class of carbocations
(2) (a) Roberts, J. D.; Chambers, V. C. J. Am. Chem. Soc. 1951, 73, 5034.
(b) Foote, C. S. J. Am. Chem. Soc. 1964, 86, 1853. (c) Schleyer, P. v. R. J. Am.
Chem. Soc. 1964, 86, 1854. (d) Schleyer, P. v. R. J. Am. Chem. Soc, 1964, 86,
1856. (e) Woodward, R. B.; Hoffman, R. J. Am. Chem. Soc. 1965, 87, 395. (f)
Schleyer, P. v. R.; Su, T. M.; Saunders, M.; Rosenfeld, J. C. J. Am. Chem. Soc.
1969, 91, 5174. (g) Schleyer, P. v. R.; Sliwinski, W. F.; Van Dine, G. W.;
Schoellkopf, U.; Paust, J.; Fellenberger, K. J. Am. Chem. Soc. 1972, 94, 125.
(h) DePuy, C. H.; Schnack, L. G.; Hausser, J. W.; Wiedemann, W. J. Am. Chem.
Soc. 1965, 87, 4006. (i) Creary, X. J. Org. Chem. 1976, 41, 3740. (j) DePuy,
C. H.; Schnack, L. G.; Hausser, J. W. J. Am. Chem. Soc. 1966, 88, 3343. (k)
Brown, H. C.; Rao, C. G.; Ravindranathan, M. J. Am. Chem. Soc. 1977, 99,
7663.
(1) (a) Be´gue´, J.-P.; Charpentier-Morize, M. Acc. Chem. Res. 1980, 13, 207.
(b) Gassman, P. G.; Tidwell, T. T. Acc. Chem. Res. 1983, 16, 279. (c) Creary,
X. Acc. Chem. Res. 1985, 18, 3. (d) Allen, A. D.; Tidwell, T. T. In AdVances in
Carbocation Chemistry; Creary, X., Ed.; JAI Press Inc.: Greenwich, CT, 1989,
p 1. (e) Creary, X. AdVances in Carbocation Chemistry; Creary, X., Ed.; JAI
Press Inc.: Greenwich, CT, 1989, p 45. (f) Charpentier-Morize, M.; Bonnet-
Delpon, D. In AdVances in Carbocation Chemistry; Creary, X., Ed.; JAI Press
Inc.: Greenwich, CT, 1989, p 219. (g) Takeuchi, K.; Yoshida, M. J. Org. Chem.
1989, 54, 3772. (h) Creary, X. Chem. ReV. 1991, 91, 1625.
(3) For reviews and early studies, see: (a) Colvin, E. W. Silicon in Organic
Synthesis; Butterworths: London, 1981. (b) White, J. M. Aust. J. Chem. 1995,
48, 1227.
2134 J. Org. Chem. 2009, 74, 2134–2144
10.1021/jo802722z CCC: $40.75 2009 American Chemical Society
Published on Web 02/09/2009

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
studied quantitatively by Lambert.5 Stabilization of carbocations
by ꢀ-silyl groups is due to a favorable interaction between the
cation vacant p-orbital with the adjacent C-Si σ-bond. A recent
X-ray crystallographic and computational study on a cyclopro-
pylcarbinyl type of ꢀ-silyl cation supports this stabilization
mechanism.6 Stabilization of aryl cations by ꢀ-silyl groups has
also recently been reported.7 A similar type of stabilization
occurs in carbenes 4 and this interaction leads to facile silyl
migration to the carbenic center of ꢀ-silyl carbenes 4.8
SCHEME 1. Synthesis and Solvolysis of
1-(Trimethylsilylmethyl)cyclopropyl Mesylate
TABLE 1. Solvolysis Rates for Substrates in CD3CO2D
It was our desire to combine these two contrasting themes.
What kind of reactivity will a ꢀ-silylcyclopropyl carbocation
have? Is the carbocation-stabilizing ability of a ꢀ-silyl group
enhanced due to the extraordinary stabilization demands of
cyclopropyl cations? Is this ꢀ-silyl effect powerful enough to
stabilize cyclopropyl cations enough to prevent ring opening?
Or is the ꢀ-silyl effect repressed due to hyperconjugation
problems in small rings? To what extent will these cations
undergo desilylation, the usual fate of ꢀ-silyl carbocations? We
now report on the chemistry of carbocations 5 and 6. It was
hoped that potential cations 5 and 6 would give further insight
into both cyclopropyl and ꢀ-trimethylsilyl carbocations. Finally,
the benzocyclobutenyl carbocations 7 were also investigated.
This system represents one in which the large demand for
ꢀ-trimethylsilyl stabilization is potentially repressed due to
antiaromatic considerations.
Results and Discussion
Solvolytic Generation of Carbocation 5. Synthesis of the
mesylate 9 (a precursor to carbocation 5) from the alcohol 8
a Solvent is nondeuterated. b Extrapolated from data at higher
temperatures.
(4) (a) Ushakav, S. N.; Itenberg, A. M. Zh. Obshch. Khim. 1937, 7, 2495.
(b) Sommer, L. H.; Dorfman, E.; Goldberg, G. M.; Whitmore, F. C. J. Am.
Chem. Soc. 1946, 68, 488. (c) Sommer, L. H.; Bailey, D. L.; Whitmore, F. C.
J. Am. Chem. Soc. 1948, 70, 2869. (d) Sommer, L. H.; Baughman, G. A. J. Am.
Chem. Soc. 1961, 83, 3346. (e) Davis, D. D.; Jacocks, H. M., III. J. Organomet.
Chem. 1981, 206, 33.
(5) (a) Lambert, J. B.; Wang, G.-t.; Finzel, R. B.; Teramura, D. H. J. Am.
Chem. Soc. 1987, 109, 7838. (b) Lambert, J. B.; Chelius, E. C. J. Am. Chem.
Soc. 1990, 112, 8120. (c) Lambert, J. B. Tetrahedron 1990, 46, 2677. (d) Lambert,
J. B.; Liu, X. J. Organomet. Chem. 1996, 521, 203. (e) Lambert, J. B.; Zhao,
Y.; Emblidge, R. W.; Salvador, L. A.; Liu, X.; So, J. H.; Chelius, E. C. Acc.
Chem. Res. 1999, 32, 183.
(6) Klaer, A.; Saak, W.; Haase, D.; Mu¨ller, T. J. Am. Chem. Soc. 2008, 130,
14956.
(7) (a) Dichiarante, V.; Salvaneschi, A.; Protti, S.; Dondi, D.; Fagnoni, M.;
Albini, A. J. Am. Chem. Soc. 2007, 129, 15919. (b) Himeshima, Y.; Kobayashi,
H.; Sonada, T. J. Am. Chem. Soc. 1985, 107, 5286.
was straightforward (Scheme 1). The mesylate 9 reacted readily
in CD3CO2D at room temperature to give methylenecyclopro-
pane 10 (30%), the ring-opened acetate 11 (26%), and the ring-
opened mesylate 12 (44%). These products are consistent with
the involvement of cation 5 in the acetolysis of 9. Loss of the
trimethylsilyl group from 5 would generate 10, while competing
ring opening to allylic cation 13 and solvent capture gives 11.
Internal return is often seen in cationic rearrangements, and
mesylate 12 is derived from this process. This mesylate 12 is
unreactive under the acetolysis conditions.
Kinetic studies provide evidence for the role of the trimeth-
ylsilyl group in the solvolysis of 9. Rate data for 9 in CD3CO2D
and previously studied substrates are summarized in Table 1.
Note that in terms of solvolysis rates, tosylate and mesylate
(8) (a) Creary, X.; Wang, Y.-X. Tetrahedron Lett. 1989, 19, 2493. (b) Creary,
X.; Wang, Y.-X. J. Org. Chem. 1994, 59, 1604. (c) Creary, X.; Butchko, M. A.
J. Org. Chem. 2001, 66, 1115. (d) Creary, X.; Butchko, M. A. J. Org. Chem.
2002, 67, 112.
J. Org. Chem. Vol. 74, No. 5, 2009 2135

Creary and Kochly
SCHEME 2. Isodesmic Calculations Evaluating
leaving groups are comparable.9 The mesylate 9 is the most
reactive substrate in Table 1. It is 103 more reactive than the
phenyl analogue 16 and 106 more reactive than the “desilylated
analogue” 15. How does this 106 rate enhancement relative to
15 compare with other cases of ꢀ-silyl rate enhancements?10 A
classic example is Lambert’s ꢀ-trimethylsilylcyclohexyl trif-
luoroacetate 18, where the rate enhancement relative to the
“desilylated” cyclohexyl trifluoroacetate 17 is a factor of 1012.5b
In view of this enormous rate-enhancing ability of a ꢀ-trim-
ethylsilyl group in 18, the rate enhancement of 106 in 9 appears
to have been “repressed” significantly relative to this maximum
value.
ꢀ-Trimethylsilyl Cation Stabilization
Computational Studies on Cation 5. Molecular orbital
calculations can also be used to gain insight into the structure
of carbocation intermediates. Figure 1 shows the B3LYP/6-
31G*-calculated structure of cation 5, which is an energy
minimum, as well as the corresponding neutral molecule 19.
The indicated carbon-carbon bond has contracted from 1.522
to 1.368 Å in the cation. This implies some double-bond
character in 5 and, thus, a stabilizing effect by the trimethylsilyl
group. Furthermore, the calculated structure of 5 shows a tilting
of the trimethylsilyl group toward the cationic center, which is
also indicative of significant ꢀ-silyl stabilization.
this calculation, cation 21 was therefore constrained by fixing
the internal carbocation bond angle to prevent ring opening.
The resulting calculation suggests that the ꢀ-silyl cation 5 is
stabilized by 16.6 kcal/mol relative to the desilylated analogue
21. Silyl stabilization of the acyclic cation 22 is quantified by
the second isodesmic calculation in Scheme 2. This cation 22
is stabilized by 27.6 kcal/mol relative to 25. The ꢀ-silyl
stabilization of cyclopropyl cation 5, at least computationally,
is 11 kcal/mol less than stabilization of the acyclic analogue
22. The third isodesmic calculation in Scheme 2 suggests that
the ꢀ-silyl cation 26 involved in the Lambert rate study5b is
stabilized by 34.9 kcal/mol relative to the cyclohexyl cation 29.
By any computational measure, stabilization of cation 5 by the
ꢀ-silyl group does not approach that of model systems.
These computational studies are in line with our kinetic
studies that suggest that there is significant stabilization of cation
5 by the ꢀ-silyl group. However, this stabilization is substantially
smaller than one might expect given the large demands of
cyclopropyl cations. This repression of ꢀ-silyl stabilization in
5 is a manifestation of repressed overlap between the
carbon-silicon bond and the vacant p-orbital of the carbocation.
This can be rationalized using a valence bond approach. The
high strain in methylenecyclopropane (40 kcal/mol)13 makes
contributions of form 5b less important than in analogous acyclic
forms. In qualitative terms, 5b is less prominent, i.e., less “bold”
than 5a.
FIGURE 1. B3LYP/6-31G*-calculated structures of cation 5 and
(trimethylsilylmethyl)cyclopropane, 19.
How does the stabilization of carbocation 5 compare to
calculated measures of ꢀ-silyl stabilization? There have been
previous ab initio calculations of ꢀ-SiH3 stabilization of other
carbocations,10 as well as estimations of ꢀ-silyl stabilization
based on gas-phase studies.11 The isodesmic calculations
(B3LYP/6-31G*) shown in Scheme 2 represent an attempt to
quantify the stabilizing effect of the ꢀ-trimethylsilyl group in
5. Comparison of the energy of 5 with that of the 1-methylcy-
clopropyl cation 21 is complicated by the fact that 21 is not an
energy minimum at the B3LYP/6-31G* level but opens without
barrier to the corresponding allylic cation. For the purpose of
(9) Robertson, R. E. Prog. Phys. Org. Chem. 1967, 4, 213.
(10) Enhancement due to trimethylsilyl assistance in 9 may in fact be larger
than a factor of 106. This is because the model system 15 is itself rate-enhanced2i
due to the k∆ process by which it solvolyzes.
(11) (a) Wierschke, S. G.; Chandrasekhar, J.; Jorgensen, W. L. J. Am. Chem.
Soc. 1985, 107, 1496. (b) Ibrahim, M. R.; Jorgensen, W. L. J. Am. Chem. Soc.
1989, 111, 819.
Solvolytic Generation of Carbocation 6. Attention was next
focused on systems where the trimethylsilyl group was directly
(12) (a) Hajdasz, D.; Squires, R. J. Chem. Soc., Chem. Commun. 1988, 1212.
(b) Li, X.; Stone, J. A. J. Am. Chem. Soc. 1989, 111, 5586.
(13) Johnson, W. T. G.; Borden, W. T. J. Am. Chem. Soc. 1997, 119, 5930.
2136 J. Org. Chem. Vol. 74, No. 5, 2009

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
FIGURE 2. 1H NMR spectrum (upfield region) of products from reaction of 32 and 33 in CF3CH2OH.
SCHEME 3. Synthesis of Chlorides 32 and 33
SCHEME 4. Solvolysis Products from 32 and 32 in
CF3CH2OH
attached to the cyclopropyl ring. Chlorides 32 and 33 were
prepared as an inseparable mixture by photolysis of a solution
of phenylchlorodiazarine 3114 in vinyltrimethylsilane 30 (Scheme
3). Stereochemical assignments were based on NMR shifts of
the trimethylsilyl groups and the corresponding R-protons of
32 and 33. The trimethylsilyl group of 33 is in the shielding
region of the cis-phenyl ring and therefore has a distinct upfield
shift (-0.24 ppm). For similar reasons, the cis-hydrogen in 32
(0.46 ppm) is further upfield than the corresponding trans-
hydrogen in 33 (0.86 ppm). These observed chemical shifts
correspond well with calculated chemical shifts based on the
B3LYP/6-31G* structures of 32 and 33.
The mixture of chlorides 32 and 33 was solvolyzed in
CF3CH2OH containing 2,6-lutidine as a buffering base. Under
these conditions, a complex mixture of products was formed
as evidenced by the trimethylsilyl region of the 1H NMR
spectrum of the products (Figure 2). Despite the challenge
presented by this mixture, the eight different products were all
identified using a variety of techniques that included independent
syntheses, and comparison of calculated H NMR chemical
shifts to differentiate isomers. Structures are shown in Scheme
4 and trimethylsilyl signals are all appropriately labeled in Figure
2. This mixture of rearranged and unrearranged products is
completely consistent with the intermediacy of a cyclopropyl
1
(14) (a) Graham, W. H. J. Am. Chem. Soc. 1965, 87, 4396. (b) Padwa, A.;
Pulwer, M. J.; Blacklock, T. J. Org. Synth., Coll. Vol. VII 1990, 203.
J. Org. Chem. Vol. 74, No. 5, 2009 2137

Creary and Kochly
TABLE 2. Solvolysis Rates for Substrates in CF3CH2OD and
CH3OD
cationic intermediate which partitions between solvent capture
and ring-opening processes.
in CF3CH2OD
in CH3OD
T (°C) )
k (s-1
Kinetic Studies on 32, 33, and Aryl Analogues. In order to
further probe the nature of the cationic intermediate produced
on solvolysis of 32 and 33, the aryl-substituted cyclopropyl
chlorides 42 and 43 were synthesized using the same general
method involving arylchlorodiazirine photolyses in vinyltrim-
ethylsilane. As before, inseparable mixtures of 42 and 43 were
formed.
compd
T (°C)
k (s-1
)
42 (m-Cl)
32 (p-H)
80.0 6.85 × 10-5
60.0 1.11 × 10-5
25.0 2.61 × 10-7
25.0 1.70 × 10-5
a
80.0 1.60 × 10-4
25.0 2.58 × 10-7
42 (p-CH3)
42 (p-OCH3)
43 (m-Cl)
25.0 7.38 × 10-4 50.0 6.26 × 10-6
a
25.0 2.25 × 10-1 25.0 7.86 × 10-5
b
60.0 9.81 × 10-5
40.0 1.17 × 10-5
25.0 1.97 × 10-4
a
33 (p-H)
25.0 1.40 × 10-4
25.0 4.12 × 10-3 25.0 1.44 × 10-6
b
43 (p-CH3)
43 (p-OCH3)
b
25.0 1.20 × 100
25.0 4.18 × 10-4
1-chlorophenylcyclopropane 95.0 2.47 × 10-5
75.0 4.25 × 10-6
25.0 1.87 × 10-8
a
a Extrapolated from data at higher temperatures. b Calculated using a
CF3CH2OD/CH3OD rate ratio of 2860.
Solvolytic studies were carried out in either CF3CH2OD or
CH3OD for the very reactive p-CH3- and p-CH3O-containing
substrates. Mixtures of ring-retained trifluoroethyl ethers 44 and
ring-opened allylic products analogous to those formed in
Scheme 4 were produced, and these are summarized in Scheme
5. As the substituent became more electron donating, more of
the products derived from solvent capture of the closed
cyclopropyl carbocation were formed. In the case of p-OCH3
substitution, only ring-closed products 44 were formed. It is
important to note that none of these solvolyses gave products
where the trimethylsilyl group was lost.
The rate data in Table 2 as well as the product studies in
Schemes 4 and 5 are in line with the involvement of cyclopropyl
cations 6 as discrete intermediates that can capture solvent or
undergo competing ring-opening processes. Comparison of rates
of solvolyses of 32 and 33 with that of 1-chlorophenylcyclo-
propane gives trimethylsilyl rate-enhancing effects of about 103
and 104, respectively. These values are even smaller than the
trimethylsilyl rate-enhancing effect of 106 in cation 5 and does
not come close to the 1012 value observed in cation 26. The
slightly higher reactivity of 32 and 42 relative to the isomeric
substrates 33 and 43 might be a ground-state effect, where there
is relief of steric interactions between the cis-aryl and trimeth-
ylsilyl groups of 33 and 43 as the ionization process occurs.
Alternatively, the faster rates of 33 and 43 could be an angular
effect, where there is better transition-state interaction of the
developing cation with the C-Si σ-bond when the trimethylsilyl
group is trans to the leaving group. This angular effect on
trimethylsilyl stabilization has been discussed in detail by
Lambert.4
SCHEME 5. Solvolysis Products from 42 and 43 in
CF3CH2OD and CH3OD
From the kinetic data in Table 2, Hammett plots were
constructed, and these are shown in Figure 3. The large negative
F+ values of -4.9 and -5.1 verify that carbocations of type 6
are involved in the rate-determining step of these solvolyses.
Of interest is a comparison of these F+ values with those of
analogous desilylated systems. Previous Hammett studies on
tosylates 46 and dinitrobenzoates 47 by DePuy2j and Brown2k
gave F+ values of -5.15 and -5.19, respectively. These F+
values are very similar to those observed for chlorides 42 and
43. This similarity indicates that both carbocation 6 and the
unsilylated arylcyclopropyl analogues rely on stabilization by
the aryl group to a similar extent. Furthermore, this suggests
that the trimethylsilyl group in 6 does little to offset the demand
for aryl group stabilization.
Rates of reaction of 32 and 33, as well as substituted
analogues 42 and 43, were determined by monitoring the
disappearance of the 1H NMR signal corresponding to the
trimethylsilyl groups of the starting chlorides. There were no
problems encountered by using isomeric mixtures for kinetic
studies since signals corresponding to 32 and 33 are cleanly
separated from each other as well as from product signals. Rate
data are summarized in Table 2. For comparison purposes, rate
data for 1-chlorophenylcyclopropane, 45, the desilylated ana-
logue of 32 and 33, are also given. Solvolyses of 42 (p-OCH3)
and 43 (p-OCH3) in CF3CH2OD were too fast to monitor
accurately by 1H NMR. Therefore, these reactions were
monitored in CH3OD. The rate of solvolysis of 42 (p-CH3) was
measured in both solvents and was determined to be 2860 times
faster in CF3CH2OD. Using this CF3CH2OD/CH3OD solvent
rate ratio, rates of 43 (p-CH3), 42 (p-OCH3), and 43 (p-OCH3)
in CF3CH2OD were calculated.
Computational Studies on Cation 6. The structure of cation
6 (Ar ) Ph) was also calculated at the B3LYP/6-31G* level.
2138 J. Org. Chem. Vol. 74, No. 5, 2009

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
SCHEME 6. Isodesmic Calculations Evaluating
ꢀ-Trimethylsilyl Cation Stabilization
SCHEME 7. Synthesis of Precursors to Cation 7
FIGURE 3. Hammett plots for solvolyses of 42 and 43 in CF3CH2OD.
In order to quantify the stabilization provided by the
trimethylsilyl group in cation 6 (Ar ) Ph), isodesmic calcula-
tions (B3LYP/6-31G*) were again carried out (Scheme 6). The
stabilization amounts to 12 kcal/mol relative to the 1-phenyl-
cyclopropyl cation 48. In the unconstrained system 49, the
ꢀ-trimethylsilyl group provides 17.9 kcal/mol of stabilization
to this benzylic carbocation. This computational study again
agrees with conclusions based on kinetic studies. While there
is clearly some stabilization by the ꢀ-trimethylsilyl group in
cation 6 (Ar ) Ph), it is less than in cation 5 or in models cations
such as 22, 26, or 49.
The relative rate data, the Hammett F+ values, the lack of
desilylated products, and the computational data all point to a
greatly repressed interaction of the trimethylsilyl group with
the cationic center in 6. A valence bond rationalization is again
of value. Contribution of forms such as 6b impart cyclopropene
character to the cation. Keeping in mind that the strain energy
of cyclopropene is 50 kcal/mol,13 forms such as 6b have
decreased importance. In other words, stabilization of cation 6
by the ꢀ-trimethylsilyl group is beginning to fade away (as is
structure 6b).
FIGURE 4. B3LYP/6-31G* structures of cation 6 (Ar ) Ph), cation
48, and chloride 32.
As in cation 5, the carbocation 6 (Ar ) Ph) shows a contraction
of the cyclopropane carbon-carbon bond (1.415 Å) relative to
that in the chloride 32 (1.510 Å). The cyclopropane bond in 6
(Ar ) Ph) is also shorter than the analogous bond in the
phenylcyclopropyl cation 48 (1.454 Å) (Figure 4). These
contractions point to some stabilizing effect by the trimethylsilyl
group in 6 (Ar ) Ph). In the cation 5, a contraction of about
0.15 Å between the neutral and cation was observed. In the
present case, however, the contraction is about 0.10 Å. This
suggests a smaller stabilizing interaction of the trimethylsilyl
group in 6 (Ar ) Ph).
J. Org. Chem. Vol. 74, No. 5, 2009 2139

Creary and Kochly
SCHEME 8. Solvolysis Products from 55 in CD3CO2D
TABLE 3. Solvolysis Rates for Substrates in CD3CO2D
SCHEME 9. Solvolysis Products from 56, 58, and 59 in
CD3CO2D
ꢀ-silyl cations are generated under solvolytic conditions. The
64:36 product ratio does not remain constant after 55 has
completely disappeared. Instead, 59 and 60 continue to react,
eventually equilibrating to produce 5:95 ratio of products. This
presumably occurs via reionization of acetates 59 and 60 to the
cationic intermediate 7 (R ) Ph) and eventual equilibration
favoring the more stable acetate 60. This information allows
an approximation of the acetolysis rate of trifluoroacetate 61,
which is not available from ketone 53. From the equilibrium
constant Keq ) k2/k3 )19, acetate 59 ionizes 19 times faster
than acetate 60. Assuming trifluoroacetate ionization rates
parallel acetate ionization rates, 61 should be 19 times more
reactive than 55.
Rates of 55 and 61 are the same order of magnitude as that
of the desilylated model compound 62. The solvolysis rate of
55 is actually slower than that of the model 62. As before, the
slightly faster rate of 61 may be an angular effect due to the
trimethylsilyl group being trans to the leaving group or it may
be a ground-state effect involving relief of strain between the
syn-phenyl and -trimethylsilyl groups as the ionization proceeds.
By any measure, trimethylsilyl rate enhancements in solvolyses
of 55 and 61 are negligible.
The tertiary trifluoroacetate 56 reacted in CD3CO2D to form
the elimination product 62 (60%) along with the acetates 63
(40%; mixture of stereoisomers). The secondary mesylates 57
and 58 both gave mixtures of the acetates 64 (Scheme 9).
Comparison of rates of 56, 57, and 58 with those of desilylated
analogues 65 and 66 showed only minor rate differences. Rate
enhancements due to trimethylsilyl stabilization or cationic
intermediates again appear to be minimal.
a Calculated value from Keq. b Calculated from data at higher
temperatures.
Solvolytic Generation of Carbocation 7. The next carboca-
tions to be investigated were the trimethylsilyl-substituted
benzocyclobutenyl cations 7. Precursors to these systems were
prepared from the ketone 53 by standard methods (Scheme 7).
The tertiary alcohols formed by addition of PhMgBr and
CH3MgBr to 53 were converted to trifluoroacetate derivatives
55 and 56 for solvolytic study, while the secondary alcohols
formed from NaBH4 reduction of 53 were converted to
mesylates 57 and 58.
The trifluoroacetate 55 was solvolyzed in CD3CO2D, where
reaction readily occurred at room temperature. Kinetic data for
55 and related substrates are given in Table 3. As shown in
Scheme 8, acetates 59 and 60 were produced from 55 in a 64:
36 ratio. There was no desilylation as usually occurs when
Computational Studies on Cation 7. As before, computa-
tional studies were used to shed further light on the ꢀ-silyl effect
on benzocyclobutenyl cations 7. The isodesmic calculations
(B3LYP/6-31G*) in Scheme 10 show a decreasing amount of
trimethylsilyl stabilization as the R group in 7 changes from H
2140 J. Org. Chem. Vol. 74, No. 5, 2009

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
FIGURE 5. B3LYP/6-31G* structures of cation 7 (R ) Ph), 7 (R ) CH3), 7 (R ) H), and cation 75.
SCHEME 10. Isodesmic Calculations Evaluating ꢀ-Trimethylsilyl Cation Stabilization
to CH3 to Ph. The cation 7 (R ) Ph) is stabilized by only 3.7
kcal/mol relative to the desilylated analogue 75. This represents
the smallest calculated value for ꢀ-silyl stabilization in our
studies.
that would necessarily come along with ꢀ-silyl stabilization.
Contribution of benzocyclobutadienyl forms15 such as 7b has
essentially faded away (just as structure 7b has faded).
The calculated structures of cations 7 are also revealing
(Figure 5). The bond contraction in 7 (R ) Ph), relative to the
desilylated cation 75, is only 0.014 Å and is indicative of
minimal trimethylsilyl stabilization. There is a systematic bond
contraction in 7 as the R group becomes less cation stabilizing.
The implication is that silyl stabilization increases in importance
in the series 7 (R ) Ph), 7 (R ) CH3), and 7 (R ) H).
The kinetic studies in Table 3 indicate essentially no
significant rate enhancements when cations 7 are solvolytically
generated. The ꢀ-trimethylsilyl stabilizing effect has therefore
been essentially eliminated when cations 7 are generated in
solution. This is attributed to the antiaromatic character of 7
Computational studies, on the other hand, indicate a small
ꢀ-silyl stabilizing effect on cations 7. It is important to realize,
however, that all of these calculations describe gas-phase
properties and that these do not always correspond precisely to
solution-phase chemistry. Stabilization of transition states in
solvolysis reactions is not expected to be as large as in fully
formed gas-phase cations. In the gas phase, however, where
J. Org. Chem. Vol. 74, No. 5, 2009 2141

Creary and Kochly
is highly reactive and was stored in ether solution at -20 °C prior
to use: 1H NMR of 9 (CDCl3) δ 2.96 (s, 3 H), 1.31 (m, 4 H), 0.66
(m, 2 H), 0.11 (s, 9 H); 13C NMR of 9 (CDCl3) δ 65.8, 40.1, 25.5,
13.4, -0.3.
demands for carbocation stabilization are greater, a minimal
amount of ꢀ-trimethylsilyl stabilization of 7 is still apparent.
Mesylate 1618 was prepared from 1-phenylcyclopropanol16 using
Conclusions
1
an analogous procedure: H NMR of 16 (CDCl3) δ 7.57 (d, J )
Although the ꢀ-silyl effect is one of the largest neighboring
group effects in carbocation chemistry, it is greatly attenuated
when incorporated into cyclopropyl cations. Rate enhancements
are reduced to a factor of about 106 in the solvolysis of
1-(trimethylsilylmethyl)cyclopropyl mesylate and further re-
duced to only a factor of 103-104 in the solvolyses of 1-phenyl-
2-trimethylsilylcyclopropyl chlorides. Despite the extraordinary
stabilization demands of cyclopropyl cations, these rate en-
hancements are substantially repressed relative to maximum
enhancements previously observed in the literature. In addition,
the ꢀ-silyl effect can be effectively quashed upon incorporation
of a ꢀ-trimethylsilyl group into benzocyclobutyl cations. ꢀ-Silyl
stabilization in these cations would lead to antiaromatic
character, and hence, rate enhancements observed in solvolysis
reactions involving these cations are minimal.
6.9, 2 H), 7.40 (t, J ) 7.2, 2 H), 7.36 (t, J ) 7.3, 1 H) 2.54 (s, 3
H), 1.66 (m, 2 H), 1.21 (m, 2 H); 13C NMR of 16 (CDCl3) δ 137.4,
129.3, 129.1, 128.9, 67.6, 39.9, 13.5.
Preparation of 1-Phenyl-2-trimethylsilylcyclopropyl Chlo-
rides 32 and 33. General Procedure for Preparation of 42 and
43. Approximately 100 mg of the appropriate arylchlorodiazirine14
was dissolved in approximately 2.5 mL of vinyltrimethylsilane, and
the solution was sealed in a Pyrex tube. The tube was irradiated
for approximately 2 h using a 450 W Hanovia lamp. Excess
vinyltrimethylsilane was removed using a rotary evaporator, and
the crude products were filtered through 0.5 g of silica gel and
eluted with hexanes. The solvent was removed using a rotary
evaporator to yield an inseparable mixture of the corresponding
chlorides 42 and 43. Chlorides 42 and 43 were typically formed in
approximatelya60:40ratio.Thefollowingprocedureisrepresentative.
Irradiation of 80 mg of phenylchlorodiazarine 3114 dissolved in
2.6 mL of vinyltrimethylsilane for 2.25 h yielded 91 mg (77% yield)
1
of chlorides 32 and 33 in a 60:40 ratio: H NMR of 32 (C6D6) δ
Experimental Section
7.36 (m, 2 H), 7.08 (m, 2 H), 7.02 (m, 1 H), 1.27 (d of d, J ) 4.9,
11.7 Hz 1 H), 1.24 (d of d, J ) 4.9, 9.5 Hz, 1H) 0.25 (d of d, J )
9.5, 11.8 Hz, 1 H), 0.20 (s, 9 H); 13C NMR of 32 (C6D6) δ 144.8,
129.0, 128.1, 128.1, 51.0, 20.7, 17.7, -0.4. 1H NMR of 33 (C6D6)
δ 7.39 (m, 2 H), 7.02 (m, 2 H), 6.96 (m, 1 H), 1.45 (d of d, J )
5.1, 12.0 Hz, 1 H), 1.13 (d of d, J ) 5.1, 9.1 Hz, 1 H), 0.83 (d of
d, J ) 9.1, 12.0 Hz, 1 H), -0.34 (s, 9 H); 13C NMR of 33 (C6D6)
δ 141.7, 130.0, 128.8, 128.6, 48.9, 19.7, 17.5, -1.6; exact mass
(EI) calcd for C12H17ClSi (M+) 224.0788, found 224.0811.
Preparation of 2-(Trimethylsilyl)benzocyclobutanone, 53.
Following the general procedure developed by Suzuki and Tsujiya-
ma,19 o-iodophenyl triflate (2.925 g) was dissolved in 30 mL of
tetrahydrofuran, and 2.444 g of trimethylsilylketene diethylacetal20
was added. The solution was cooled to -78 °C, and 5.2 mL of 1.6
M n-butyllithium in hexanes was added dropwise. After being
stirred for 1 h, the solution was warmed to -50 °C over 30 min.
Water (15 mL) was added, and the solution was then warmed to
room temperature. An aqueous workup ensued with ether extraction.
The ether extract was washed with water and saturated salt solution
and dried over a mixture of Na2CO3 and MgSO4. After filtration,
the solvent was removed using a rotary evaporator. The residue
was distilled, and after a lower boiling forerun, 884 mg of distillate
was collected at 56-62 °C (0.25 mm). This distillate contained
1,1-diethoxy-2-(trimethylsilyl)benzocyclobutanone, 2-iodo-n-bu-
tylbenzene, and biphenylene in a 1.0:0.81:0.49 ratio, along with
smaller amounts of other impurities. The distillate was dissolved
in 10 mL of tetrahydrofuran and a solution of 160 mg of sulfuric
acid in 6 mL of water was added. After stirring vigorously for 2 h
at room temperature, NaHCO3 was added followed 15 mL of ether
and NaCl until the aqueous phase was saturated. The organic phase
was separated and dried over a mixture of Na2SO4 and MgSO4.
After filtration, the solvent was removed using a rotary evaporator
and the residue was chromatographed on 10.5 g of silica gel. The
column was eluted with increasing amounts of ether in hexanes.
Ketone 5321 (206 mg, 13% yield) eluted with 4% ether in hexanes:
1H NMR of 53 (CDCl3) δ 7.44 (t of d, J ) 7, 1 Hz, 1 H), 7.33 (m,
1 H), 7.29 (t, J ) 7 Hz, 1 H), 7.26 (t of t, J ) 7, 1 Hz, 1 H), 3.87
(br s, 1 H), 0.08 (s, 9 H); 13C NMR of 53 (CDCl3) δ 191.3, 153.5,
146.7, 134.9, 127.2, 122.1, 120.2, 59.4, -2.7.
Preparation of 1-(Trimethylsilylmethyl)cyclopropanol), 8. The
general cyclopropanol synthesis procedure of DePuy16 was em-
ployed. Trimethylsilylmethylmagnesium chloride (30 mL of a 1.0
M solution in ether) was cooled to 0 °C, and a solution of 3.206 g
of 1,3-dichloroacetone in 18 mL of ether was added dropwise. The
reaction was allowed to warm to room temperature and then
recooled to 0 °C. Solutions of 723 mg of FeCl3 in 25 mL of ether
and ethylmagnesium bromide (prepared from 5.0 g of magnesium
and 18.7 g of ethyl bromide in 60 mL of ether) were simultaneous-
ly added dropwise to the reaction mixture. After the mixture was
allowed to stand at room temperature for 16 h, the reaction was
recooled to 0 °C and an aqueous solution of ammonium chloride
was added. The organic layer was then separated, washed with water
and saturated salt solution, and dried over a mixture of Na2SO4
and MgSO4. After filtration, the solvent was removed using a rotary
evaporator. The crude product was distilled to give 1.65 g of a
mixture of alcohol 817 (83% of the mixture) and 2-butanone (17%
1
of the mixture): bp 63-65 °C (15 mm); H NMR of 8 (CDCl3) δ
1.66 (br s, 1 H), 0.98 (s, 2 H), 0.74 (m, 2 H), 0.41 (m, 2 H), 0.09
(s, 9 H); 13C NMR of 8 (CDCl3) δ 54.4, 27.4, 15.3, -0.3.
Preparation of 1-(Trimethylsilylmethyl)cyclopropyl Mesy-
late, 9, and 1-Phenylcyclopropyl Mesylate, 16. A solution of 287
mg (2.0 mmol) of the alcohol 8 obtained above (containing 17%
2-butanone) and 368 mg (3.2 mmol) of mesyl chloride in 2.5 mL
of CH2Cl2 was cooled to -10 °C, and a solution of 393 mg of
triethylamine in 1.5 mL of CH2Cl2 was added dropwise. The
reaction was allowed to warm to 20 °C for 7 min and then recooled
to -5 °C. Cold water (2 mL) was then added along with 5 mL of
ether. A fast aqueous workup ensued with ether extraction. The
organic layer was separated and washed with ice-cold aliquots of
water, dilute HCl solution, water, aqueous NaHCO3 solution, and
saturated NaCl solution. The ether extract was then dried over a
mixture of Na2SO4 and MgSO4 and filtered, and solvent was
removed using a rotary evaporator to yield mesylate 9. Mesylate 9
(15) For a discussion of antiaromaticity in benzocyclobutadiene and related
compounds, see: (a) Cava, M. P.; Napier, D. R. J. Am. Chem. Soc. 1957, 79,
1701. (b) Nenitzescu, C. D.; Avram, M. Chem. Ber. 1957, 90, 2541. (c) Pettit,
R.; Merk, W. J. Am. Chem. Soc. 1967, 80, 4788. (d) Schleyer, P. v. R.; Maerker,
C.; Dransfeld, A.; Jiao, H.; Hommes, N. J. R. v. E. J. Am. Chem. Soc. 1996,
118, 6317. (e) Nagorski, R. W.; DeAtley, A. D.; Broadus, K. M. J. Am. Chem.
Soc. 2001, 123, 8428. (f) Fu, N.; Allen, A. D.; Kobayashi, S.; Sequeira, P. A.;
Shang, M.; Tidwell, T. T.; Mishima, M. J. Am. Chem. Soc. 2007, 129, 6210.
(16) DePuy, C. H.; Dappen, G. M.; Eilers, K. L.; Klein, R. A. J. Org. Chem.
1964, 29, 2813.
Preparation of 1-Phenyl-cis-2-(Trimethylsilyl)benzocyclobu-
tanol, 54. A solution of 29 mg of ketone 53 in 1.5 mL of ether
was added dropwise to 0.3 mL of 1 M phenylmagnesium bromide
in ether at 0 °C. After 10 min, the solution was warmed to room
(17) Fukuzawa, S.; Furuya, H.; Tsuchimoto, T. Tetrahedron 1996, 52, 1953.
(18) Kozyrkov, Y. Y.; Kulinkovich, O. G. Synlett 2002, 3, 443.
2142 J. Org. Chem. Vol. 74, No. 5, 2009

Systematic Repression of ꢀ-Silyl Carbocation Stabilization
temperature for 1.5 h. The solution was then warmed to 35 °C for
15 min and then cooled to 0 °C. Aqueous NH4Br solution was
added. The ether phase was then separated, washed with saturated
salt solution, and dried over a mixture of Na2SO4 and MgSO4. After
filtration, the solvent was removed using a rotary evaporator. The
residue was chromatographed on 2.6 g of silica gel. The column
was eluted with increasing amounts of ether in pentane. Alcohol
5422 (36 mg, 88% yield) eluted with 2% ether in pentane: 1H NMR
of 54 (CDCl3) δ 7.44 (m, 2 H), 7.32 (m, 3 H), 7.26 (m, 3 H), 7.10
(d of q, J ) 1.0, 7.2 Hz, 1 H) 3.30 (t, J ) 0.9, 1 H), 2.39 (s, 1 H)
0.15 (s, 9 H); 13C NMR of 54 (CDCl3) δ 148.0, 145.6, 145.5, 129.8,
128.6, 127.5, 126.4, 125.7, 122.5, 121.5, 83.7, 53.3, -1.4; exact
mass (FAB) calcd for C17H21OSi 269.1362, found 269.1347 (M +
1).
General Procedure for the Preparation of Trifluoroacetates.
A solution of the appropriate alcohol in ether along with 2 equiv
of 2,6-lutidine was cooled to 0 °C, and 1.5 equiv of trifluoroacetic
anhydride was added dropwise. After 15 min at 0 °C, cold water
was added, and a cold aqueous workup ensued with ether extraction.
In the case of trifluoroacetate 55, the solution was warmed to room
temperature for 1 h before the aqueous workup. The ether layer
was rapidly washed with cold aliquots of water, dilute HCl solution,
water, aqueous NaHCO3 solution, and saturated salt solution and
dried over a mixture of Na2SO4 and MgSO4. Solvent removal using
a rotary evaporator gave the corresponding trifluoroacetates in
∼95% yield. This general procedure was used for the preparation
of trifluoroacetates 55, 56, 57, 62, and 65 from the corresponding
alcohols. These trifluoroacetate esters are prone to decomposition
on standing as neat liquids and were therefore stored at -20 °C in
ether solution.
1-Phenyl-cis-2-(Trimethylsilyl)benzocyclobutyl trifluoroac-
etate, 55: mp 84-86 °C; 1H NMR of 55 (CDCl3) δ 7.59 (d of d of
t, J ) 7, 1, 0.3 Hz, 2 H), 7.35 (m, 7 H), 7.34 (d of q, J ) 7, 1 Hz,
2 H), 3.30 (t, J ) 1 Hz, 1 H), 0.18 (s, 9 H); 13C NMR of 55 (CDCl3)
δ 156.5 (q, J ) 42 Hz), 145.1, 141.4, 139.9, 131.4, 128.7, 128.6,
127.0, 126.3, 126.0, 122.3, 114.6 (q, J ) 287 Hz), 90.2, 52.2, -1.9;
exact mass (FAB) calcd for C19H19F3O2Si 364.1106, found 364.1103.
1-Phenylbenzocyclobutyl Trifluoroacetate, 62. This substrate
was prepared from 1-phenylbenzocyclobutanol:23 1H NMR of 62
(CDCl3) δ 7.62 (d of t, J ) 7.4, 1 Hz, 1 H), 7.45 (m, 3 H), 7.36
(m, 4 H), 7.25 (d of quin, J ) 7, 1 Hz, 1 H), 3.99 (d of d, J ) 14,
0.4 Hz, 1 H), 3.64 (d, J ) 14 Hz, 1 H); 13C NMR of 62 (CDCl3)
δ 156.7 (q, J ) 42 Hz), 142.8, 141.9, 138.5, 131.5, 128.9, 128.7,
128.4, 126.8, 125.9, 124.2, 114.6 (q, J ) 286 Hz), 89.0, 47.0; exact
mass (FAB) calcd for C16H11F3O2 292.0711, found 292.0728.
1-Methyl-cis-2-(Trimethylsilyl)benzocyclobutyl Trifluoroac-
etate, 56. Methylmagnesium iodide (0.6 mL of 1.0 M in ether)
was added to 1 mL of ether, and the solution was cooled to -10
°C in an ice/acetone bath. A solution of 56 mg of the ketone 53 in
2 mL of ether was added dropwise. The solution was warmed to
room temperature and then recooled to -10 °C. Aqueous NH4Br
solution was then added dropwise to the stirred solution, and the
ether phase was then separated and washed with saturated NaCl
solution. The ether extract was dried over MgSO4 and filtered, and
the solvent was removed using a rotary evaporator to give 57 mg
(95% yield) of 1-methyl-cis-2-(trimethylsilyl)benzocyclobutanol:
1H NMR of 1-methyl-cis-2-(trimethylsilyl)benzocyclobutanol
(CDCl3) δ 7.23 (m, 1 H), 7.16-7.11 (m, 2 H), 7.03 (d of q, J )
7.2, 1.0 Hz, 1 H), 3.03 (br s, 1 H), 2.00 (s, 1 H), 1.69 (s, 3 H), 0.08
(s, 9 H); 13C NMR of 1-methyl-cis-2-(trimethylsilyl)benzocyclobu-
tanol (CDCl3) δ 150.0, 144.5, 129.2, 125.8, 122.2, 119.9, 80.3, 50.9,
27.9, -1.7.
Trifluoroacetate 56 was prepared from 1-methyl-cis-2-(trimeth-
ylsilyl)benzocyclobutanol by the general procedure described above:
1H NMR of 56 (CDCl3) δ 7.39 (d of t, J ) 7, 1 Hz, 1 H), 7.30 (t
of d, J ) 7.5, 1 Hz, 1 H), 7.18 (t of t, J ) 7.5, 1 Hz, 1 H), 7.04 (d
of q, J ) 7, 1 Hz, 1 H), 3.10 (t, J ) 1 Hz, 1 H), 1.92 (s, 3 H), 0.08
(s, 9 H); 13C NMR of 56 (CDCl3) δ 156.7 (q, J ) 42 Hz), 144.20,
144.17, 130.6, 126.5, 124.3, 121.7, 114.5 (q, J ) 287 Hz), 89.3,
49.8, 23.6, -2.2; exact mass (FAB) calcd for C14H17F3O2Si
302.0950, found 302.0958.
1-Methylbenzocyclobutyl Trifluoroacetate, 65. This substrate
was prepared from 1-methylbenzocyclobutanol.23 1H NMR of 65
(CDCl3) δ 7.39 (d of t, J ) 7, 1 Hz, 1 H), 7.36 (t of d, J ) 7, 1 Hz,
1 H), 7.28 (t of q, J ) 7, 1 Hz, 1 H), 7.19 (d of quin, J ) 7, 1 Hz,
1 H), 3.63 (d, J ) 14 Hz, 1 H), 3.42 (d, J ) 14 Hz, 1 H), 1.91 (d,
J ) 0.3 Hz, 3 H); 13C NMR of 65 (CDCl3) δ 156.8 (q, J ) 42 Hz),
145.2, 140.8, 130.8, 127.9, 123.74, 123.70, 114.4 (q, J ) 286 Hz),
86.9, 45.2, 22.1; exact mass (FAB) calcd for C11H10F3O2 231.0633,
found 231.0636 (M + 1).
cis-2-(Trimethylsilyl)benzocyclobutyl Mesylate, 57. A solution
of 59 mg of ketone 53 in 2 mL of CH3OH was cooled in an ice/
salt bath at -10 °C, and 17.8 mg of NaBH4 was added. The mixture
was warmed to room temperature and then recooled to -10 °C. A
solution of 40 mg of acetic acid in 0.5 mL of methanol was added
dropwise with stirring, and then 3 mL of water was added. The
mixture was then extracted with 6 mL of ether. Pentane (5 mL)
was added to the ether extract, and the solution was then washed
with water and saturated NaCl solution. The mixture was dried over
Na2SO4, and the solvents were removed using a rotary evaporator
to give 58 mg (97% yield) of cis-2-(trimethylsilyl)benzocyclobu-
tanol and trans-2-(trimethylsilyl)benzocyclobutanol in a 94:6 ratio:
1H NMR of cis-2-(trimethylsilyl)benzocyclobutanol (CDCl3) δ 7.26
(t of m, J ) 7.4 Hz, 1H), 7.19 (d of m, J ) 7.1 Hz, 1 H), 7.17 (d
of t, J ) 7.4, 0.9 Hz, 1 H), 7.03 (d of m, J ) 7.3 Hz, 1 H), 5.39
(d, J ) 4.9 Hz, 1 H), 3.27 (t of d, J ) 4.9, 0.9 Hz, 1 H), 2.05 (br,
1 H), 0.07 (s, 9 H); 13C NMR of cis-2-(trimethylsilyl)benzocy-
clobutanol (CDCl3) δ 146.7, 145.0, 129.3, 125.8, 121.8, 1231.7,
72.7, 44.3, -1.7.
A solution of 46 mg of the alcohols prepared above and 40 mg
of CH3SO2Cl in 2 mL of methylene chloride was cooled to -10
°C as 50 mg of Et3N in 0.5 mL of CH2Cl2 was added dropwise.
The mixture was warmed to room temperature and taken up into
10 mL of ether. The mixture was then washed successively with
cold water, cold dilute HCl solution, NaHCO3 solution, and
saturated NaCl solution. The organic phase was then dried over a
mixture of Na2SO4 and MgSO4. After filtration, the solvents were
removed using a rotary evaporator to give 61 mg (94% yield) of
1
mesylates 57 and 58 (94% of 57 and 6% of 58): H NMR of 57
(CDCl3) δ 7.33 (t of t, J ) 7.5, 0.8 Hz, 1 H), 7.29 (d of m, J ) 7.5
Hz, 1 H), 7.21 (t of t, J ) 7.5, 0.9 Hz, 1 H), 7.05 (d of m, J ) 7.4
Hz, 1 H), 5.96 (d of m, J ) 5.2 Hz, 1 H), 3.43 (d of t, J ) 5.2,
0.9 Hz, 1 H), 3.09 (s, 3 H), 0.08 (s, 9 H); 13C NMR of 57 (CDCl3)
δ 144.6, 141.6, 130.7, 126.5, 123.5, 121.5, 76.8, 42.3, 38.6, -2.1;
exact mass (FAB) calcd for C12H18O3SSi 270.0746, found 270.0721.
Benzocyclobutyl Mesylate, 66. This substrate was prepared from
benzocyclobutanol:24 1H NMR of 66 (CDCl3) δ 7.39 (t of d, J )
7.4, 1.4 Hz, 1 H), 7.34-7.28 (m, 2 H), 7.18 (d of m, J ) 7.4 Hz,
1 H), 5.92 (d of d, J ) 4.5, 1.9 Hz, 1 H), 3.73 (d of d, J ) 14.5,
4.5 Hz, 1 H), 3.44 (d of m, J ) 14.5 Hz, 1 H), 3.10 (s, 3 H); 13C
NMR of 66 (CDCl3) δ 142.2, 142.1, 131.0, 128.0, 123.7, 123.5,
75.1, 39.6, 38.6; exact mass (FAB) calcd for C9H10O3S 198.0351,
found 198.0365.
(19) Tsujiyama, S.; Suzuki, K. Org. Synth. 2007, 84, 272.
(20) (a) Schlosser, M.; Wei, H. Tetrahedron 1997, 53, 1735. (b) Wei, H.;
Schlosser, M. Tetrahedron Lett. 1996, 37, 2771.
Solvolyses of Mesylates, Chlorides, and Trifluoroacetates.
Kinetics Procedures. Rate constants reported in Tables 1-3 were
(21) (a) Chenard, B. L.; Slapak, C.; Anderson, D. K.; Swenton, J. S. Chem.
Commun. 1981, 4, 179. (b) Swenton, J. S.; Anderson, D. K.; Jackson, D. K.;
Narasimhan, L. J. Org. Chem. 1981, 46, 4825.
(22) Saito, M.; Saito, A.; Ishikawa, Y.; Yoshioka, M. Org. Lett. 2005, 7,
3139.
(23) Adam, G.; Andrieux, J.; Plat, M. Tetrahedron 1985, 41, 399.
(24) Dhawan, K. L.; Gowland, B. D.; Durst, T. J. Org. Chem. 1980, 45,
922.
J. Org. Chem. Vol. 74, No. 5, 2009 2143

Creary and Kochly
1
all determined using 600 MHz H NMR spectroscopy. A solution
Solvolysis of Chlorides 32 and 33 in CF3CH2OH. Product
Study. A solution of the 31 mg of chlorides 32 and 33 and 22 mg
of 2,6-lutidine in 2.0 mL of CF3CH2OH was heated at 40 °C for
24 h. The solvent was removed using a rotary evaporator, the
residue was taken up into ether, and the ether solution was washed
with water and saturated NaCl solution and then dried over MgSO4.
After solvent removal using a rotary evaporator, the residue was
analyzed by 600 MHz 1H NMR. The upfield region of the spectrum
is shown in Figure 2, and the downfield region is shown as
Supporting Information. Products 34-36, 38, 39, and 41 were
identified by 1H NMR spectral comparison with independently
was prepared by dissolving approximately 5 mg of the appropriate
substrate in 400 mg of CD3CO2D, CF3CH2OD, or CH3OD contain-
ing approximately 1.5 equiv of 2,6-lutidine. The sample was sealed
in an NMR tube. For runs at 25 °C, the tube was placed in the
probe of an NMR set at 25.0 °C. At appropriate time intervals, the
1
tube was analyzed by H NMR, and relative areas due to starting
mesylate, chloride, or trifluoroacetate were measured. In most
instances, the 2,6-lutidine singlet at δ 2.74 was used as an internal
standard. For runs at higher temperature, the tube was placed in a
constant temperature bath at the appropriate temperature. At
appropriate time intervals, the tube was then rapidly quenched in
a water bath at 15 °C and rapidly analyzed by 1H NMR at ambient
temperature.
1
prepared samples. Products 37 and 40 were identified using H
NMR spectral data. Independent syntheses and spectral data for
these compounds are given as Supporting Information.
In the case of mesylates 9 and 16 in CD3CO2D, the rates of
disappearance of the singlets due to the ROSO2CH3 group at δ
2.99 and 3.02, respectively, were monitored. For mesylates 57, 58,
and 66 in CD3CO2D, the disappearance of singlets at δ 3.16, 3.13,
and 3.14, respectively, was monitored.
Solvolysis of Trifluoroacetate 55 in CD3CO2D. Product
Study. A solution of the 4.6 mg of trifluoroacetate 55 and 3.8 mg
of 2,6-lutidine in 360 mg of CD3CO2D was placed in an NMR
tube, and the tube was kept at 25 °C for 52 min. Analysis by 600
1
MHz H NMR showed products 59 and 60 in a 64:36 ratio. The
For trifluoroacetate 55 in CD3CO2D, the rate of disappearance
of the TMS singlet at δ 0.20 was monitored. For trifluoroacetate
56, the TMS singlet at δ 0.09 was monitored. For trifluoroacetate
65, the CH3 singlet at δ 1.90 was monitored. The reaction of
trifluoroacetate 62 (half-life ) 63 s) was monitored by quenching
CD3CO2D solutions of 62 in C6D6 before measuring the relative
amount of unreacted 62 (doublet at δ 3.67).
For chlorides 32 and 33 in CF3CH2OD, the rates of disappearance
of the TMS singlets at δ 0.21 and -0.27 respectively, were
monitored. For chlorides 42 and 43 in CF3CH2OD and in CH3OD,
the TMS singlets at approximately δ 0.21 and -0.26 were also
monitored.
Typical data illustrating reaction of 9 in CD3CO2D, 42 and 43
(Ar ) p-CH3OC6H4) in CH3OD, and 55 in CD3CO2D are given as
Supporting Information. First-order rate constants for disappearance
of substrates were calculated by standard least-squares procedures.
Correlation coefficients were all greater than 0.9997. The maximum
standard deviation in duplicate runs was (2%.
structure of 60 was verified by independent synthesis of the protio
analogue by acetylation of alcohol 54 by reaction with acetic
anhydride and dimethylamino pyridine in CH2Cl2.
The NMR tube was then placed in a water bath at 25.0 °C and
periodically monitored by 1H NMR over a period of 32 days. During
this time, acetate 59 epimerized to 60, and eventually an equilibrium
position consisting of 59 and 60 in a 5:95 ratio was attained. The
half-life for approach to equilibrium was 73 h. Corresponding
spectra are shown as Supporting Information.
Computational Studies. Ab initio molecular orbital calculations
were performed using the Gaussian 03 series of programs.28 All
structures, except the constrained system 21, were characterized
as minima via frequency calculations that showed no negative
frequencies.
Supporting Information Available: Complete ref 28; the
B3LYP/6-31G*-calculated structures, energies, and Cartesian
coordinates of 5, 6 (Ar ) Ph), 7 (R ) Ph), 7 (R ) CH3), 7 (R
) H), 19, 32, 48, and 75; 1H and 13C NMR spectra of
compounds 9, 32 and 33, 42 and 43 (Ar ) m-ClC6H4), 42 and
43 (Ar ) p-CH3C6H4), 42 and 43 (Ar ) p-CH3OC6H4), 55, 56,
57 and 58, 62, 65, and 66, as well as evolving 1H NMR spectra
during reaction of 9 in CD3CO2D, 42 and 43 (Ar ) CH3OC6H4)
in CH3OD, and 55 in CD3CO2D; experimental procedures for
independent syntheses of 34, 35, 37-39, and 41. This material
is available free of charge via the Internet at http://pubs.acs.
org.
Solvolysis of Mesylate 9 in CD3CO2D. Product Study. A
solution of the 3.3 mg of mesylate 9 and 2.6 mg of 2,6-lutidine in
350 mg of CD3CO2D was sealed in an NMR tube, and the tube
was kept at 25 °C for 4 h. The mixture was then analyzed by 600
25
1
MHz H NMR. The products (methylenecyclopropane 10, the
protio analog of acetate 11,26 and mesylate 1227) are known
compounds. The relative amount of each product was determined
1
by integration of the downfield region of the H NMR spectrum
and is shown as Supporting Information.
(25) (a) Gragson, J. T.; Greenlee, K. W.; Derfer, J. M.; Boord, C. E. J. Am.
Chem. Soc. 1953, 75, 3344. (b) LeFevre, G. N.; Crawford, R. J. J. Org. Chem.
1986, 51, 747. (c) Dolbier, W. R., Jr.; Alonso, J. H. J. Chem. Soc., Chem.
Commun. 1973, 394.
(26) Trost, B. M.; Chan, D. M. T. J. Am. Chem. Soc. 1983, 105, 2315.
(27) Trost, B. M.; Vincent, J. E. J. Am. Chem. Soc. 1980, 102, 5680.
JO802722Z
(28) Frisch, M. J. et al. Gaussian 03, ReVision C.01; Gaussian, Inc.:
Wallingford, CT, 2004.
2144 J. Org. Chem. Vol. 74, No. 5, 2009