1,3-γ-Silyl-elimination in electron-deficient cationic systems

Article abstract of DOI:10.1039/c4sc01732c

Placement of an electron-withdrawing trifluoromethyl group (-CF ₃) at a putative cationic centre enhances γ-silyl neighbouring-group participation (NGP). In stark contrast to previously studied γ-silyl-substituted systems, the preferred reaction pathway is 1,3-γ-silyl elimination, giving ring closure over solvent substitution or alkene formation. The scope of this cyclopropanation reaction is explored for numerous cyclic and acyclic examples, proving this method to be a viable approach to preparing CF₃-substituted cyclopropanes and bicyclic systems, both containing quaternary centres. Rate-constants, kinetic isotope effects, and quantum mechanical calculations provided evidence for this enhancement and further elaborated the disparity in the reaction outcome between these systems and previously studied γ-silyl systems.

Full text of DOI:10.1039/c4sc01732c

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
1,3-g-Silyl-elimination in electron-deficient
cationic systems†
Cite this: DOI: 10.1039/c4sc01732c
b
b
b
Michael A. Mercadante, Christopher B. Kelly, Trevor A. Hamlin, Kayla R. Delle
a
a
a
a
a
Chiaie, Michael D. Drago, Katherine K. Duffy, Megan T. Dumas, Diana C. Fager,
Bryanna L. C. Glod, Katherine E. Hansen, Cameron R. Hill, Rebecca M. Leising,
Catherine L. Lynes, Allyson E. MacInnis, Madeline R. McGohey,
Stephanie A. Murray, Marc C. Piquette, Shaina L. Roy, Ryan M. Smith,
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
Katherine R. Sullivan, Bao H. Truong, Kristina M. Vailonis, Vitaliy Gorbatyuk,
Nicholas E. Leadbeater* and Leon J. Tilley*a
bc
3
Placement of an electron-withdrawing trifluoromethyl group (–CF ) at a putative cationic centre enhances
g-silyl neighbouring-group participation (NGP). In stark contrast to previously studied g-silyl-substituted
systems, the preferred reaction pathway is 1,3-g-silyl elimination, giving ring closure over solvent
substitution or alkene formation. The scope of this cyclopropanation reaction is explored for numerous
3
cyclic and acyclic examples, proving this method to be a viable approach to preparing CF -substituted
Received 11th June 2014
Accepted 23rd June 2014
cyclopropanes and bicyclic systems, both containing quaternary centres. Rate-constants, kinetic isotope
effects, and quantum mechanical calculations provided evidence for this enhancement and further
elaborated the disparity in the reaction outcome between these systems and previously studied g-silyl
systems.
DOI: 10.1039/c4sc01732c
www.rsc.org/chemicalscience
g-silyl substituted sulfonate esters were solvent substitution
with retention of stereochemistry (Fig. 1, Product IV) and rear-
Introduction
The construction of highly-strained ring systems and the crea- rangement to give b-silyl elimination (Product III), with 1,3-g-
tion of quaternary centers are arduous processes coveted by silyl elimination to give cyclopropanation (Product II) being a
2g
synthetic chemists. One unique unexplored route to simulta- minor path. Even cyclic systems, wherein the leaving group
neously perform these tasks exploits 1,3-g-silyl elimination. In and the g-silyl substituent are in the ideal cis “W” conformation
1
946, Sommer and Whitmore reported on the ability of a g-silyl for maximum percaudal interaction, gave only slightly higher
3a
substituent to donate electron density to the a-carbon, thereby percentages of cyclopropanation (Product II).
1
enhancing the reactivity of this centre. Subsequently, this type
One logical solution to enhance cyclopropanation would be
of interaction has been observed in a wide variety of g-silyl to increase the electron-donating character of the g-substituent.
2,3
scaffolds, particularly in carbocationic systems. Shiner and Indeed, Kuivila reported that the solvolysis of 1-aryl-3-trime-
others reported on homohyperconjugative stabilization of the thylstannyl 3,5-dinitrobenzoates led exclusively to cyclo-
4
developing p-orbital of a cationic centre by back-lobe (“per- propanation via 1,3-g-stannyl elimination. Elimination
caudal”) participation of the C–Si bond of the g-silyl substit-
2,3
uent. Unfortunately, this transient percaudal interaction has
not been effectively harnessed for ring closure to access
substituted cyclopropanes. Shiner and others have shown that
the major product-forming pathways in the solvolysis studies of
a
Department of Chemistry, Stonehill College, 320 Washington Street, Easton,
Massachusetts 02357, USA. E-mail: ltilley@stonehill.edu
b
Department of Chemistry, University of Connecticut, 55 North Eagleville Road, Storrs,
Connecticut 06269, USA. E-mail: nicholas.leadbeater@uconn.edu
c
Department of Community Medicine & Health Care, University of Connecticut Health
Center, The Exchange, 263 Farmington Ave, Farmington, CT 06030, USA
†
Electronic supplementary information (ESI) available: Experimental procedures,
characterization data, and spectra of the compounds. See DOI:
0.1039/c4sc01732c
1
Fig. 1 Possible pathways of g-silyl stabilized carbocations.
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
11g
products in Yoshida's competitive study of a dual g-stannyl and inammatory diseases as well as in potential pharmacons for
5
11h
11i
g-silyl system arose solely from the g-stannyl substituent.
management of cholesterol levels and pain. Likewise, CF3
However, potentially toxic organostannanes limit the synthetic cyclopropane-containing substances have found use as pestici-
12a,b
12c,d
utility of this solvolytic reaction. As a potential surrogate for a g- des
and parasiticides.
Recently, the CF3 cyclopropyl
stannyl group, we hypothesized that an umpolung strategy group has been found to be a suitable replacement for the tert-
13
would be to augment the g-silyl percaudal interaction by butyl group, but with enhanced metabolic stability.
installing a CF group at the a-carbon. Work by Gassman and Current methods for CF cyclopropane synthesis (Fig. 2)
others supports this hypothesis by suggesting that “the place- primarily rely on anionic or carbene/carbenoid approaches
3
3
14,15
ment of a strongly electron-withdrawing group at the incipient (Fig. 2 eqn (2)–(4)).
However, nearly all of these reactions
cationic centre can magnify the inuence of a related series of involve the use of potentially explosive and/or harsh reagents/
neighbouring groups” by generating an “electron-decient” conditions to facilitate ring formation. Therefore, we sought to
6
carbocation. This strategy is supported by the known delocal- bolster the limited repertoire available for constructing CF3
ization of electron density from adjacent p- and s-orbitals into cyclopropanes by utilizing 1,3-g-silyl elimination to effect ring
7
the C–F s*-orbital.
closure. We explored a broad scope of substrates and performed
We recently validated our strategy by synthesizing the highly comprehensive product, rate-constant, isotope effect, and
strained 1-(triuoromethyl)bicyclo[1.1.0]butane via exclusive computational studies. The results of our investigations are
8
1
,3-g-silyl elimination (Fig. 2 eqn (1)). This result was in stark presented here.
contrast to studies by Creary and co-workers on analogous
g-trimethylsilyl alkyl cyclobutyl systems in which no 1,3-g-silyl
elimination was observed. We therefore wished to explore the
generality of our method in the context of a potential synthetic Seeking to develop a generalized route for the synthesis of our
Synthetic results and discussion
2h
route to triuoromethylcyclopropanes (CF
containing a quaternary centre.
3
cyclopropanes) substrates, we modeled our strategy on our previous prepara-
8
tion of cyclobutyl tosylates. Thus, we would rst prepare a
CF
3
cyclopropanes are highly attractive synthetic targets for carbonyl compound containing a g-silyl moiety (–SiMe
) and subsequently introduce the CF group via the
This small-ring motif can be Ruppert–Prakash reagent (TMS–CF3). The resulting a-CF3
3
,
medicinal chemists due to their compact, rigid structure and –SiPhMe
their lipophilic properties.
2
3
9,10
16
found in a variety of recent candidates for the treatment of carbinol could then be converted to the desired leaving group.
11a
11b
11c
11d,e
11e,f
cancer,
hepatitis,
HIV,
obesity,
diabetes,
and
Drawing on the work of Fleming and coworkers, we found
that a variety of substrates could be synthesized using Route A
in good overall yield (Scheme 1). Alkylation of cyclohexyl
2c,i
imines with Me Si–CH I or PhMe Si–CH I, reagents that are
3
2
2
2
easily prepared from the corresponding commercially available
chlorides by a Finkelstein reaction with NaI, proved to be a
2k
facile and rapid methodology for silyl group introduction. To
prepare other non-aryl substituted systems, we elected to utilize
3 2
either Michael additions of Me Si–Li or the cuprate of PhMe Si–
17
Li to a,b-unsaturated carbonyl derivatives (Route B). Another
route involving alkylation of b-ketoesters with Me Si–CH I or
3
2
PhMe Si–CH I followed by decarboxylation (Route C) was also
2
2
2b
used to access straight chain alkyl systems. Lastly, we devised
a modied route to prepare an allyl system via a series of
alkylations and oxidations (Route D). Subsequent tri-

uoromethylation of the intermediate ketones using TMS–CF
3
6c
followed by deprotonation with KH and functionalization of
the alkoxide with the appropriate anhydride gave us the
8
desired substrates for solvolysis. The results of our solvolytic
studies utilizing these compounds are presented in Table 1.
Due in part to the relative ease in their preparation, we rst
examined straight-chain systems with an a-phenyl ring. We
were initially concerned that the stabilization afforded by the
aromatic system would deter 1,3-g-silyl participation. However,
the solvolytic reaction of 2a proceeded smoothly in 97 T [97%
2
2,2,2-triuoroethanol (TFE)/3% H O (w/w)] giving 1,3-g-silyl
elimination as the major product of solvolysis. But, even using
TFE as the sole solvating agent, minor amounts of substitution
Fig.
2
Overview
of
current
methods
for
effecting products were detected. To mitigate formation of unwanted
trifluoromethylcyclopropanation.
substitution products, we conducted the reaction in the even
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
Scheme 1 Generalized routes to access potential cyclopropane precursor solvolytic substrates.
less nucleophilic solvent hexauoroisopropanol (HFIP) at a substitution, E1 elimination, and b-silyl elimination (via a
relatively high substrate concentration (1 mmol per mL of Wagner–Meerwein rearrangement).
HFIP). This led to cyclopropanation as the exclusive reaction
pathway for 2a (Table 1, Entry 1). We explored the –SiMe
We next explored 1,3-g-silyl elimination in cyclic systems to
Ph determine if ring closure was possible in systems other than
2
group as an alternative g-silyl substituent (Entry 2) and obtained cyclobutyl. Both six- and seven-membered ring systems (Entries
identical yield. Thus, it is likely that a wide range of silyl groups 14 and 15) exhibited bridging. In contrast, the cyclopentyl
can be used so long as they do not lead to unfavorable confor- system (Entry 13) gave mostly elimination (80% E1 elimination,
mational restriction or reduction in electron availability on 20% b-silyl elimination following rearrangement) with trace
silicon. Moreover, having a less volatile silyl byproduct may be amounts of substitution products. We suspect none of the low-
advantageous when isolating lower molecular weight CF3 energy cyclopentyl conformers permit the “W” orbital confor-
2h,8
cyclopropanes.
mation to facilitate percaudal participation. In the case of the
To probe the effect of electronic changes on 1,3-g-silyl cyclohexyl system, 2h, which adopts the correct “W” con-
3a
ꢀ
participation we screened a series of additional aromatic formation, mild heating (40 C), yielded the highly strained
systems. Both electron-decient (Entries 3, 4, 8) and moderately bicyclo[3.1.0]hexane as the sole product of solvolysis. The
electron-rich (Entries 5 & 7) arenes underwent facile g-silyl increased degrees of freedom in the larger cycloheptyl ring
elimination to afford the corresponding CF cyclopropanes in system, 2m, provides access to a favorable conformation for
3
good to excellent yield. Of particular note are the pyridyl bridging. This exibility still did not, however, permit the trans
0
systems (Entries 9 & 10). Because of the disparity in electronics isomer, 2m , to bridge.
between the 2- and 3-positions on the pyridine rings, heating
In an attempt to circumvent the conformational issues
0
was needed to facilitate the ionization of the 2-pyridyl system arising from the cyclopentyl system, we solvolysed 2j and 2j
while the 3-pyridyl system reacted, albeit slowly, at room (Entries 11 & 12). Rather than obtaining the desired housane,
temperature. We were also pleased to nd that the pyridyl we serendipitously obtained 1j. A mechanistic rationale for this
substituent itself could successfully serve as the base for the result can be seen in Fig. 3. Due to the rapidity of this reaction
reaction (Entry 9), yielding the pyridinium salt aer solvent and its highly selectivity towards ring contraction, we suggest
removal. The salt facilitated product isolation as it decreased that, in addition to the known NGP by C–C s-bonds in cyclo-
volatility (compare yield of Entry 9 to Entry 10 or 1 and 2). The butyl systems, there is a transient cation-stabilizing interac-
tosylate of the 3-pyridyl system could not be obtained as a pure tion between the g-silyl group and the forming cation. This is
material as it was contaminated with z20% of its alcohol evidenced by a substantial rate difference (z12 times) between
precursor. However, the solvolysis reaction could still be con- the two diastereomers, indicating that the putative carbocation
18
0
ducted with impure material because the cyclopropane product formed from isomer 2j is beneting from additional NGP
was easily separated from this impurity. This demonstrates the stabilization. However, ultimately the product determining
overall robustness of this protocol. Likewise of note are the pathway is dictated by the C–C s-bond NGP which yields a
4-bromo and 4-chloro systems (Entries 3 & 4) as they could highly stabilized b-silyl cyclopropylcarbinyl cation. Subsequent
potentially be further utilized in transition-metal-mediated b-silyl elimination gives the alkenyl CF cyclopropane in both
3
coupling reactions to prepare more complex systems. Not all cases.
a-aryl substrates underwent cyclopropanation successfully. No We next turned our attention to linear aliphatic systems
cyclopropane products could be obtained in the case of the (Entries 16–20), which required triuoromethylsulfonate (OTf)
19
4
-methoxy substituted system (Entry 6). Even when using a less- leaving groups to facilitate solvolysis. In general, these reac-
reactive heptauorobutyrate (–OHFB) leaving group, solvolysis tions proceeded smoothly, with cyclopropanation being the sole
occurred rapidly and gave complex mixture of S 1 reaction pathway. NMR yields were high, but due to the
a
N
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
a
Table 1 Solvolysis studies of various g-trimethylsilyl-a-trifluoromethyl systems
b
b
Entry
Substrate
Product
Yield (%)
Entry
Substrate
Product
Yield (%)
c
c,i
1
58
11
12
(84)
(87)
c
c,i
2
58
76
c
f,g
3
13
14
15
—
c
f,g,h
d,e,f
4
74
72
(95)
(92)
c
5
c
f,h
e,f
j
6
—
74
95
16
17
18
29 (91)
64 (95)
(91)
c
7
c,d
8
c,d,e
j
9
100
65
19
20
35 (98)
c
k
l
1
0
(68)
a
Reaction conditions unless otherwise noted: hexauoroisopropanol (HFIP), cyclopropane precursor (1 equiv., 1 M in HFIP), pyridine (2 equiv.);
b
OTs: p-toluenesulfonate, OTf: triuoromethanesulfonate, OPf: pentauorobenzenesulfonate, OHFB: heptauorobutyrate. Isolated yield; yields
in parentheses indicate yields determined by H NMR spectroscopy. Prepared via Route A. Reaction was performed at 50 C. No pyridine
was used. Solvolysis was performed in TFE. Prepared via Route B. Reaction was performed at 40 C. While both substrates gave the same
1
c
d
ꢀ
e
f
g
h
ꢀ
i
0
ꢂ4 ꢂ1
0
ꢂ3 ꢂ1
j
k
product, 2j reacted z12 times faster than 2j (2j: k ¼ 1.49 ꢁ 10
s
; 2j : k ¼ 1.83 ꢁ 10
s ). Prepared via Route C. Prepared via Route D.
l
0
2
0% yield of S
N
1 product, 12% yield of the S
N
1 product.
volatile systems (by raising the boiling point of silyl ether by-
products) and led to improved yields. NMR yields were obtained
exclusively in some cases as the desired cyclopropanes were
nearly inseparable from either silyl-ether byproducts or extrac-
tion solvents. As an alternative route to the vinyl cyclopropane 1j
0
obtained from 2j and 2j , we screened the corresponding
straight-chain allyl system 2p. This system gave a mixture of
Fig. 3 Possible pathways of g-silyl stabilized carbocations.
cyclopropanation and substitution [both at the cationic center
0
(S
N
1) and g-carbon on the olenic moiety (S
N
1 )]. Increased
volatility of the products, isolation and purication were substitution is likely a combination of the enhanced stability of
extremely difficult. Altering the silyl group (Me Si to Me PhSi), the resonance-stabilized ion, and decreased participation of the
3
2
minimized complications in product purication of several
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
silyl group due to reduced conformational restriction in the nucleophilicity (as would be expected for extensive participa-
23
straight-chain system.
tion) whereas 2t exhibits nucleophilic solvent assistance. In
contrast, Raber–Harris plots of 2u and 2v are very similar,
showing very little if any nucleophilic solvent assistance for
either system, as might be expected during formation of a
Kinetic results and discussion
To determine the extent of g-silyl participation during the benzylic cation (see ESI†).
course of these reactions, we obtained rst-order solvolytic rate- Secondary b-d KIE data are also consistent with greater
constants (using conductometric methods) for several repre- participation via a “nonclassical” cation type or concerted
sentative systems and corresponding carbon analogs. We also mechanism for straight chain aliphaitc CF systems and
prepared b-d
labeled analogs of some of these systems to probe reduced (but not absent) participation in straight-chain a-aryl
secondary b-deuterium kinetic isotope effects (KIE). These systems. The magnitude of the b-d KIE depends on both the
2
3
2
2
secondary KIEs measure rate changes due to isotopic substitu- extent of positive charge at the potential cationic centre and the
tions at a site other than the bond breaking/formation site in dihedral angle between the incipient p-orbital and the adjacent
24
the rate determining step of the reaction, providing transition- C–H/C–D bonds. In cases with large percaudal participation
state conformational and charge distribution information. The this dihedral angle is such that virtually no hyperconjugative
results of these studies and values of relevant literature stabilization exists. The absence of hyperconjugation results in
20,21
compounds are shown in Table 2.
little or no difference between isotopically-labelled and -unla-
The greatly enhanced rate effects seen for the silyl systems belled compounds, giving secondary KIEs of one or less. Inverse
provide strong evidence of percaudal participation during the KIEs (i.e. less than one) result from inductive effects of deute-
3
a,24
transition state for this reaction. The most striking enhance- rium, as Shiner has reported previously (Entries 8 & 14).
8
ment is for our a-CF
rate acceleration of more than 6 ꢁ 10 represents the largest g- small secondary b-d
silyl enhancement measured to date. The largest rate via a concerted process in which the C–Si bond acts as an
3
cyclobutyl system (Table 2, Entry 2). The Thus extensive percaudal participation would result in in very
6
2
KIEs. This participation can occur either
2h
N
enhancement previously measured was reported by Creary to internal nucleophile in an S 2-like fashion or a stepwise process
be 1 ꢁ 10 for a secondary g-silyl cyclobutyl system. Our extreme leading to a conformationally-bridged ion with extensive charge
5
case results from a combination of increased demand for silyl delocalization.
participation and ideal orbital overlap due to the locked “W”
0
Our systems 2n/2n exhibit virtually no KIE (Entry 10). The
large KIE seen for the carbon analog in system 2t/2t is consis-
Shiner reported similar (although less drastic) effects for tent with literature values. In the case of a-aryl systems, the
the less rigid cyclohexyl systems lacking a CF group (Entries isotope effects for both systems 2u/2u and 2v/2v , are quite
0
conformation.
2
k
0
0
3
3a
8
& 9). Rate enhancements for the straight chain systems in small. The small normal KIE for the carbon analog is in line
this work also seem to correspond to the stability of the putative with the resonance delocalization of the positive charge into the
carbocation and therefore inversely to the demand being placed aromatic system expected as a result of the greatly increased
on the silyl group for stabilization. The rate increase of over 200 demand for electron density created by the a-CF in that system
3
6,25
for 2n over 2t is the largest ever observed for any acyclic g-silyl (Entry 15). The inverse (i.e. non-unity) isotope effect for the
system to date (Entries 9 & 10). An analogous comparison using silyl system likely indicates the presence of an intermediate ion
systems lacking an a-CF
3
group (Entries 11 & 12) showed only a that is stabilized both by percaudal participation (requiring
mere doubling in rate in the silyl system. Presumably, this is conformational restriction and reduced hyperconjugation) and
because the inherent stability of the tertiary carbocation places resonance with the phenyl group. Thus, while the phenyl group
2
k
very little demand on the silyl group. The marked decrease in deters participation, it is not completely removed. A similar
rate enhancement for the more stable phenyl systems (compare deterrence in silyl participation (and thereby a diminished rate
Entries 15 & 16) seems to initially be indicative of greatly enhancement) has also been observed in a-phenyl-b-silyl
26
diminished silyl participation during the ionization step. cationic systems.
However, the major product from the silyl-substituted phenyl
Our isotope effects provide a nice comparison to reports by
system was still the cyclopropane (see Table 1 Entries 1 & 2), Kuivila and by Davis who observed similarly small secondary
4
27
although as noted above for these systems (as well as the allylic b-d
2
KIEs for deoxystannylation of both aliphatic and aryl
one) substitution was also observed in TFE.
straight chain systems. Both reports concluded the mecha-
4,27
Consequently, we envisioned an electron-demand depen- nisms to be concerted. Because in our case only the aliphatic
dent mechanistic spectrum for these reactions. This ranges systems show extensive participation, the stronger effect of the
from either a concerted or “nonclassical” ion-containing stannyl group must be such as to preclude the formation of an
process in systems analogous to 2n where strong g-silyl demand intermediate benzyl cation. Thus we suggest that the electronic
exists to a stepwise process with a clearly extant “classical” inuence on percaudal participation in our unique g-silyl a-CF
carbocation intermediate with reduced demand in systems system lies between those of g-silyl and g-stannyl groups alone
3
22
analogous to 2v. To further substantiate this concept, we and results from a combined push-pull effect based on the
examined Raber–Harris plots and secondary b-d KIEs. Raber– electronics of the a-carbon.
2
Harris plots for 2n and 2t (see ESI†) show a clear difference in
3
The inuence of the CF group itself may be further extracted
mechanism for these systems; 2n is not dependent upon solvent utilizing rate data from the cyclobutyl and straight-chain
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Table 2 First order rate constants and secondary b-d
analogues at 25 C in 97 T
2
KIE data for solvolytic studies of representative g-silyl sulfonic esters and their carbon
ꢀ
ꢂ1 a
b
Entry
1
Compound
Rate (s
)
b-d
2
KIE (k
H
/k
D
)
g-Silyl group acceleration
1.00 ꢁ 10ꢂ5
—
—
6.12 ꢁ 10ꢂ4
3.38 ꢁ 10ꢂ9e
1.55 ꢁ 10ꢂ4e
—
—
—
6.0 ꢁ 10
6c
2
d
3
—
d
4
4
4.6 ꢁ 10
5
6
2.00 ꢁ 10ꢂ1
4.40 ꢁ 10ꢂ2
—
—
—
4f
2.2 ꢁ 10
g
1.64 ꢁ 10ꢂ5
1.76 ꢁ 10ꢂ3
1.39 ꢁ 10ꢂ5
7
2.06
—
g
8
0.972
1.85
107.5
—
9
1
1
0
1
2.90 ꢁ 10ꢂ3
5.58 ꢁ 10ꢂ4
1.00
208
h
1.452
—
h
1.31 ꢁ 10ꢂ3
8.44 ꢁ 10ꢂ4
6.54 ꢁ 10ꢂ6
1
1
1
2
3
4
1.159
—
2.35
—
i
i
0.975
129
1
1
5
6
2.48 ꢁ 10ꢂ3
9.43 ꢁ 10ꢂ3
1.02
0.98
—
3.8
a
c
b
Values in parentheses indicate percent error of the obtained rate constant. Ratio of rates between the silyl sulfonate ester and its carbon analog.
5
d
e
f
5
Assuming kOTf/kOTs ¼ 1 ꢁ 10 (conversion factor obtained from ref. 20). Ref. 2h. Solvent was CD
3
CO
2
D. Assuming kOTs/kCl ¼ 2 ꢁ 10 and kOHFB
/
g
h
i
k
Cl ¼ 2 (ref. 20 and 21). Ref. 3a. Ref. 2k. Ref. 2g.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
aliphatic systems. Creary reported a g-silyl rate acceleration of
Chemical Science
2h
4
4.6 ꢁ 10 for tertiary cyclobutyl systems (Entries 3 & 4) lacking a
cation-destabilizing a-CF group. We supplemented Creary's
3
data by preparing and solvolyzing the more sterically-analogous
1
-methyl-3-tert-butyl systems (Entries 5 and 6), which showed a
4
similar enhancement of 2.2 ꢁ 10 based purely on conforma-
tional locking. Consequently, the rate enhancement ratio for
6
4
Scheme 2 Isodesmic calculations evaluating g-silyl stabilization of
cations IIIa–g.
Entries 2 and 1 to Entries 6 and 5 (6.0 ꢁ 10 /2.2 ꢁ 10 ¼ 130)
allows the demand created by the a-CF for homo-
3
hyperconjugative stabilization by the g-silyl group to be
2
extracted and inferred to be roughly a factor of 10 . Application
the degree of g-silyl stabilization and the available electron
density of the aryl ring. IIIa and IIIb are only slightly more
energetically stable than IIa and IIb, respectively, due to more
available ring electron density and, in the case of IIIa, an
additional resonance contributor. This stabilization thereby
of this approach to the corresponding acyclic aliphatic systems,
wherein the added variable of conformational restriction is
removed, gives a remarkably similar result. The rate enhance-
ment ratio for Entries 11 and 10 to Entries 13 and 12 (208/2.35 ¼
8
a-CF
8.5) shows a demand for silyl participation created by the
mitigates the need for g-silyl cation stabilization. However, IIId
2
3
group that is identical in order of magnitude (z10 ).
ꢂ1
and IIIe are ꢃ2 kcal mol
more stable than IId and IIe,
Thus, ne tuning of the electronics by modication of the
substitution pattern at the a-carbon can induce dramatic
mechanistic changes. However, there must still be sufficient
percaudal interaction to allow cyclopropanation to be the lowest
energy product-forming pathway. This intermediate reactivity
level between traditional g-silyl and g-stannyl systems (lacking
respectively, due to diminished available ring electron density
thereby necessitating enhanced g-silyl participation. Compar-
ison to the corresponding a-CH
g-stannyl systems IIIf and IIIg leads further credence to the
electronic intermediacy of our a-CF g-silyl system in terms of
3
substituted g-silyl and
3
the g-substituents ability to participate in cation stabilization.
Little difference between IIIf and IIf conrms minimal partic-
3
a-CF groups) allows for preparation of relatively stable, tin-free
substrates while still providing regioselective product control.
2k
ꢀ
ipation of the g-silyl group in the 3 system while the g-stannyl
ꢂ1
4,27
system is ꢃ2 kcal mol more stabilizing and is consistent
Computational results and discussion
with its ability to contribute signicantly via bridging in the
aromatic systems even without the added demand of the a-CF
3
The aryl and vinyl systems clearly proceed via a carbocation
intermediate, because solvent substitution was observed and
secondary KIEs supported reduced participation. We used
computational modeling to predict secondary KIEs for 2v and
group. Energetically, the contribution of the g-stannyl substit-
uent appears to be equivalent to that of a g-silyl system with
both an a-CF and a 4-CF -substituted-a-aryl substituent.
3 3
To further probe these electronic effects, the relative degree
of aromaticity of the a-aryl ring as a function of cation structure
2n and found them to be in excellent accord with experiment
(
computed KIE for 2n ¼ 0.98; 2v ¼ 0.96). Density functional
was assessed using the Harmonic Oscillator Model of Aromaticity
28
theory (DFT) calculations were performed with Gaussian 09 at
the B3LYP level of theory with the 6-31+g(d) basis set. The
33
(
HOMA; eqn (1)) for IIa, IIIa, IIb, IIIb, etc. In eqn (1), a is a
29
30
normalization constant (a ¼ 257.7 for C–C bonds), n is the
31
polarizable continuum model (PCM) was used to better eval-
uate solvated cationic species. TFE was chosen as the solvent
used for our solvation model as it was representative of the
solvents used experimentally. Further details on the calcula-
tions can be found in the ESI.†
number of C–C bonds included in the summation, Ropt
¼
˚
1
.388 A, and R
i
¼ length of successive C–C bonds. Using HOMA,
the degree of aromatic character of a system relative to benzene
can be assessed numerically. A HOMA value of 0 corresponds to
a completely nonaromatic system, while a HOMA of 1 is equal in
aromatic character to benzene.
Although our mechanistic (and computed KIEs) results on
the aliphatic and carbocyclic systems indicate extensive partic-
ipation, the lack of substitution does not provide sufficient
evidence to unambiguously conclude a mechanistic pathway.
Exclusive 1,3-g-silyl elimination could result from either poor
solvent nucleophilicity preventing substitution on a bridged ion
intermediate, or because of a concerted pathway. To further
glean mechanistic insight, the electronic and conformational
effects on g-silyl participation (and subsequent 1,3-elimination)
Analysis of the DHOMA values conrmed that in all cases the
aryl ring had more aromatic character in g-silyl systems than in
their carbon analogues. This implies that the extent of aryl
stabilization is diminished in these systems because of
compensatory g-silyl participation. As dispersal of charge into
the aryl ring becomes less favourable, g-silyl stabilization is
enhanced and allows the ring system to retain a higher degree
of aromaticity. With the removal of the CF decreasing elec-
3
3
in a-CF cationic systems were investigated by additional
tronic demand, IIIf and IIIg both show little difference from the
carbon analogue IIf; with the slightly higher degree of aroma-
ticity IIIg consistent with participation from the tin.
n
computational modeling.
32
An isodesmic study of a-aryl systems (Scheme 2) was con-
ducted to elucidate the balance between phenyl resonance- and
g-silyl-mediated carbocation stabilization. In all cases g-silyl
cations were slightly more stable than their carbon analogs
X ꢀ
ꢁ
a
2
HOMA ¼ 1 ꢂ
R
opt ꢂ R
i
(1)
n
i¼1
(Table 3). A clear inverse correlation can be observed between
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
a
Table 3 Relative stabilization energies and HOMA values for g-metalloidal carbenium ions and the corresponding carbon analogues
b
ꢂ1
c
c
d
Entry
Eqn in Scheme 2
DH kcal mol
HOMA IIa–g
HOMA IIIa–g
DHOMA (III–II)
a
b
c
d
e
f
(1)
(2)
(3)
(4)
(5)
(6)
(7)
ꢂ0.37
ꢂ0.68
ꢂ1.00
ꢂ1.67
ꢂ2.18
ꢂ0.53
ꢂ1.60
0.487
0.633
0.714
0.781
0.812
0.834
0.834
0.511
0.663
0.747
0.825
0.872
0.846
0.860
0.0242
0.0296
0.0335
0.0448
0.0609
0.0111
0.0257
g
a
ꢂ1
b
In kcal mol at 298 K. A negative value indicates that the g-silyl carbenium ion IIa–e is more stabilized than the corresponding carbon analog
c
carbenium ion IIIa–e. HOMA values range from 0–1. HOMA ¼ 0 for completely nonarmoatic system and HOMA ¼ 1 for fully aromatic system (all
d
˚
C–C bonds Ropt ¼ 1.388 A). Difference between HOMA IIa–e and HOMA IIIa–e. Positive values indicate a higher degree of aromaticity.
For the 4-CF , 4-CH , and 4-H systems, thermodynamically energies), but is energetically unfavorable for Vd and Ve. This
3
3
favorable isodesmic values correlated with empirical cyclo- explains the need for heating in the solvolysis study of Vd. In the
4
propanation as did the g-stannyl system IIIg. However, exper- case of Va, however, the extremely electron rich nature of the
imental solvolysis of the 4-OMe system resulted in predominant methoxy substituent indicates a potentially competing rate-
solvent substitution, suggesting some thermodynamic determining second step, which can further explain the pref-
threshold had been crossed. To better understand the relative erential solvent substitution in this system.
energetics of the cyclization and solvolytic pathways, we popu-
In all observed cases the energetics of the second step
lated the free energy surface (Fig. 4). Initial attempts to locate controls product formation. In this “product-determining” step
transition state structures involved chemical intuition and or of the solvolysis, the cyclization is exergonic for VIIb–e. Cycli-
serendipity. If these attempts proved fruitless, QST3 calcula- zation of cationic intermediates VIIc–e proceeds via TS-VIIIc–e,
tions were employed where the reactant, product, and initial the energy barrier for which decreases systematically as the aryl
guess of the transition state were inputted. Transition states substituent becomes more deactivating. Once the 4-Me system
34
were conrmed by IRC analysis. The ionization step of Va–c is is reached, the activation energy increases slightly compared to
an exergonic process (summing cation and leaving-group anion VIIc. This is likely due to the increased stability of the
Fig. 4 Calculated energy profile for the cyclopropanation of Va–e. Figures of the optimized transition state geometries for both the ionization
and cyclization steps are provided in the ESI.†
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
Chemical Science
intermediate VIIb starting to disfavor the cyclization pathway. to TS-XIIIc-e). This disparity in energy likely stems from
Indeed, in the case of VIIa, the predicted cyclization step conformational restriction and is consistent with experimental
becomes rate determining. This change in energetics suggests requirements for less reactive leaving groups in the cyclic
either a longer-lived cation for this pathway (allowing for systems. In a select case we also explored a concerted b-hydride
competing substitution) or simply other potentially lower elimination reaction pathway and compared it to the g-silyl
energy “product-determining” pathways. These alternative elimination path. Comparing the two pathways from straight
reaction pathways are consistent with our experimental obser- chain systems leading to TS-XIa and TS-XIIIa, it is clear that the
vation that solvent substitution predominates. Efforts to locate controlling factor for elimination in g-silyl a-CF
3
cationic
a transition state for a substitution pathway were unsuccessful. systems is almost certainly kinetic. While the cyclopropane XIVa
Attempts to perform analogous isodesmic studies of the is ultimately higher in energy, its activation barrier is signi-
‡
straight-chain aliphatic and carbocyclic systems did not provide cantly lower than the barrier leading to XIIa (DDG ¼ 3.0 kcal
ꢂ1
the expected results based on the assumption of a stable mol ), allowing this pathway to predominate. Additionally, the
bridged ion-intermediate. Optimization of the ground state so-called “peruoroalkyl effect” initially described by Lemal,
structures of the g-silyl cations of these systems resulted in may provide some kinetic stability to this strained system by
spontaneous cyclization. No stable carbocationic intermediate preventing electrophilic degradation and raising activation
35
36
could be located. Since the optimized ground state-structures energies of isomerization. In the case of the cyclopentyl system
of the silyl systems had different bond arrangements than their we could not locate a transition-state for cyclopropanation. The
carbon analogs (which were located as the expected carboca- structure of the system is such that the preferred stabilizing
tions), isodesmic studies could not be performed. The inability interaction was hyperconjugation of the b C–H bond, leading to
to locate a bridged ion pointed toward a concerted mechanism concerted elimination (XIb). This pathway is consistent with our
or at least an ionization step with a very early transition state. prediction of unfavorable ring conformation forbidding cycli-
Furthermore, in our attempts to locate a transition state cor- zation and preferring elimination as seen experimentally.
responding to an ionization step, we discovered a transition
state corresponding to both ionization and cyclization.
While treatment of most cyclic systems with assumption that
both steps were concerted was viable, the cyclobutyl system
We therefore explored the potential energy surfaces of the proved to be an exception (Fig. 6, XVII). Indeed, even unsub-
straight-chain and carbocyclic systems with the premise that stituted cyclobutyl systems are known to receive anchimeric
18
there is a single transition state involving both ionization and assistance from the C–C s-bonds in the ring. Such assistance
cyclization (Fig. 5). In all cases the ring-closing reaction was would diminish the dependency on direct g-silyl-promoted
exergonic and a marked difference in the activation barrier was elimination and lead to a longer-lived ion. However, analysis of
observed between cyclic and acyclic systems (compare TS-XIIIa the ground state structure XVII still showed the hallmarks of
Fig. 5 Calculated energy profile for the possible pathways for straight chain and carbocyclic g-silyl a-CF
transition state geometries for the cyclization steps are provided in the ESI.† Numeric values indicate energies in kcal mol
3
cations. Figures of the optimized
ꢂ1
.
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
reaction to be permissible from a thermodynamic standpoint;
(4) the “peruoroalkyl effect,” may provide some kinetic
stability to this resulting strained system thereby trapping it in a
36
potential energy well.
Conclusions
These studies evince that placement of a cation-destabilizing
group adjacent to the a-carbon in a g-silyl systems leads to a
much stronger g-silyl interaction. This enhanced interaction
Fig. 6 Calculated energy profile for the ring closure of XV. Figures of
the optimized transition state geometries for the cyclization steps are allows the 1,3-g-silyl elimination pathway to predominate,
ꢂ1
provided in the ESI.† Numeric values indicate energies in kcal mol
.
yielding a method for cyclopropanation in most systems. A
strategy for the preparation of CF cyclopropanes was developed
3
to demonstrate the synthetic value of this enhancement. Using
g-silyl stabilization (e.g. ring contraction, elongation of the C–Si kinetic studies and hybrid density functional theory calcula-
bond). To quantify the degree of g-silyl stabilization in this tions, the entire mechanistic spectrum was elucidated. Solvol-
system, we conducted an isodesmic study in a manner similar yses of the straight chain aliphatic, cyclohexyl and cycloheptyl
to our previous study of the a-phenyl system (Fig. 7). This study systems appear to be concerted with a transition state involving
revealed that not only was g-silyl stabilization present, but it was signicant charge delocalization by percaudal participation,
quite substantial as compared to the straight-chain a-phenyl whereas the cyclobutyl system proceeds through a percaudally-
ꢂ1
ꢂ1
systems (ꢂ15.4 kcal mol vs. ꢂ0.37 to ꢂ2.18 kcal mol ) and stabilized bridged ion intermediate. The difference between
signicantly more than the corresponding a-CH system. To these systems can be accounted for by the high demand for silyl
3
give a comparison of disparate stabilizing effects (resonance vs. participation in the former without the added stabilizing factors
anchimeric assistance), an isodesmic study was conducted of maximal conformational restriction and neighbouring
using an a-vinyl substituent. This study produced stabilization s-bond participation found in the latter. The stabilization
energy more in line with the a-phenyl systems (ꢂ2.25 kcal provided by the adjacent p system in both the a-aryl and a-vinyl
ꢂ1
mol ). Additionally, we were able to quantify the amount of systems begins to dominate enough to allow for a more classical
energetic stabilization afford by the a-CF
by comparing it to its a-CH
3
group in this system ion. However, the complementary directing effects of the g-silyl
ꢂ1
3
congener (5.58 kcal mol ). and a-CF substituents are still sufficient (excepting the most
3
Therefore, while stabilizing, the anchimeric assistance of C–C electron-donating p-methoxy system, 2e) to favour 1,3-silyl
s-bonds is not nearly as competitive to g-silyl homo- elimination over substitution.
hyperconjugative stabilization as resonance stabilization.
Finally, to address the discrepancy between the endothermic
nature of the potential energy surface (Fig. 6) for the cyclobutyl
Acknowledgements
system and the highly successful bridging under laboratory We would like to dedicate this article to Dr Vernon J. “Jack”
conditions, we offer the following rationales: (1) the high Baeyer Shiner, Jr for inspiring this work. This work was supported by
strain of XIX in the resulting bicyclobutyl system dramatically Stonehill College, the Office of Naval Research (Award Number:
increases the energy of this system; (2) the g-silyl homo- N00014-11-1-0921) (KRDC, MDD, KKD, MTD, DCF, KEH, CRH,
hyperconjugative stabilization is very signicant in this system RML, CLL, AEM, MRM, SAM, MCP, KMV), and the National
leading to a highly electropositive silicon; (3) the formation of Science Foundation (CAREER Award CHE-0847262) (CBK, TAH,
the bond between the electropositive silicon and the solvent, MAM, NEL). We thank Dr Martha Morton of the University of
which was not accounted for in these computations, may offset Connecticut for her assistance with NMR characterization and
the energetic cost of forming this system and would allow this Dr You-Jun Fu and Adam McShane for their assistance in
obtaining mass spectra of various compounds. We would also
like to thank Barbara Anzivino, Dr Louis Liotta, and Dr Pamela
Lombardi of Stonehill College; Dr Christian Hamann of
Albright College; Dr James Bobbitt, Dr Nicholas Eddy, Dr Robert
Birge, and DiAndra Rudzinski of the University of Connecticut
for helpful discussions.
Notes and references
1
L. H. Sommer, E. Dorfman, G. M. Goldberg and
F. C. Whitmore, J. Am. Chem. Soc., 1946, 68, 488.
2
(a) L. H. Sommer, R. E. Van Strien and F. C. Whitmore, J. Am.
Chem. Soc., 1949, 71, 3056; (b) L. H. Sommer and
N. S. Marans, J. Am. Chem. Soc., 1950, 72, 1935; (c)
Fig. 7 Isodesmic calculations evaluating g-silyl stabilization of cations
XXIIa/b and XXVI.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Edge Article
I. Fleming and J. Goldhill, J. Chem. Soc., Perkin Trans. 1, 1980,
Chemical Science
P. R. Guzzo, A. J. Henderson, M. L. Reitman,
R. P. Nargund, M. J. Wyvratt and L. S. Lin, ACS Med. Chem.
Lett., 2011, 2, 43; (e) US Pat., 015199A1, 2011; (f) US Pat.,
317645A1, 2010; (g) US Pat., 205773A1, 2006; (h) US Pat.,
161685A1, 2007; (i) US Pat., 105306A1, 2009.
1
1
493; (d) J. B. Lambert and R. B. Finzel, J. Am. Chem. Soc.,
982, 104, 2020; (e) V. J. Shiner Jr and E. R. Davidson, J.
Am. Chem. Soc., 1986, 108, 3135; (f) J. Coope, V. J. Shiner Jr
and M. W. Ensinger, J. Am. Chem. Soc., 1990, 112, 2834; (g)
V. J. Shiner Jr, M. W. Ensinger and R. D. Rutkowske, J. Am. 12 (a) US Pat., 148649A1, 2005; (b) US Pat., 120884A1, 2010; (c)
Chem. Soc., 1987, 109, 804; (h) X. Creary and E. D. Kochly, WO Pat. 023773A1, 2005; (d) WO Pat., 090313A1, 2006.
J. Org. Chem., 2009, 74, 9044; (i) I. Fleming, S. K. Patel and 13 D. Barnes-Seeman, M. Jain, L. Bell, S. Ferreira, S. Cohen,
C. Urch, J. Chem. Soc., Perkin Trans. 1, 1989, 115; (j)
M. Fujio, T. Nakashima, R. Fujiyama, H.-J. Kim and
X.-H. Chen, J. Amin, B. Snodgrass and P. Hatsis, ACS Med.
Chem. Lett., 2013, 4, 514.
Y. Tsuno, J. Phys. Org. Chem., 2000, 13, 612; (k) L. J. Tilley 14 Carbene/Carbenoid
Approaches:
(a)
B.
Morandi,
and V. J. Shiner Jr, J. Phys. Org. Chem., 1999, 12, 564; (l)
J. Coope and V. J. Shiner Jr, Org. Chem., 1989, 54, 4270.
(a) V. J. Shiner Jr, G. S. Kriz, K. A. Halley and M. W. Ensinger,
J. Org. Chem., 1990, 55, 653; (b) W. T. Bentley, W. Kirmse,
G. Llewellyn and F. Sollenbohmer, J. Org. Chem., 1990, 55,
B. Mariampillai and E. M. Carreira, Angew. Chem., Int. Ed.,
2011, 50, 110; (b) I. R. del Villar, A. Gradillas and J. P ´e rez-
Castells, Eur. J. Org. Chem., 2010, 5850; (c)
P. K. Mykhailiuk, S. Afonin, G. V. Palamarchuk,
O. V. Shishkin, A. S. Ulrich and I. V. Komarov, Angew.
Chem., Int. Ed., 2008, 47, 5765; (d) B. Morandi and
E. M. Carreira, Angew. Chem., Int. Ed., 2010, 49, 938; (e)
B. Morandi, J. Cheang and E. M. Carreira, Org. Lett., 2011,
13, 3080; (f) J. R. Denton, D. Sukumaran and
H. M. L. Davies, Org. Lett., 2007, 9, 2625; (g) B. Morandi
and E. M. Carreira, Angew. Chem., Int. Ed., 2010, 49, 4294;
(h) P. M u¨ ller, S. Grass, S. P. Shahi and G. Bernardinelli,
Tetrahedron, 2004, 60, 4755; (i) J. Risse, M. A. Fernandez-
Zumel, Y. Cudre and K. Severin, Org. Lett., 2012, 14, 3060.
3
4
1536; (c) W. Kirmse and F. Sollenbohmer, J. Am. Chem.
Soc., 1989, 111, 4127.
(a) D. C. McWilliam, T. R. Balasubramanian and
H. G. Kuivila, J. Am. Chem. Soc., 1978, 100, 6407; (b)
J. A. Verdone, J. A. Mangravite, N. M. Scarpa and
H. G. Kuivila, J. Am. Chem. Soc., 1975, 97, 843; (c)
H. G. Kuivila and N. M. Scarpa, J. Am. Chem. Soc., 1970, 92,
6990.
5
6
J. Yoshida and M. Sugawara, J. Org. Chem., 2000, 65, 3135.
(a) P. G. Gassman and J. B. Hall, J. Am. Chem. Soc., 1984, 106, 15 Anionic Ring Closures: (a) B. Jiang, F. Zhang and W. Xiong,
4
1
267; (b) P. G. Gassman and M. M. Doherty, J. Am. Chem. Soc.,
982, 104, 3742; (c) D. W. Nelson, N. J. O'Reilly, J. Speier and
Chem. Commun., 2003, 536; (b) Y. Wang, X. Zhao, Y. Li and
L. Lu, Tetrahedron Lett., 2004, 45, 7775; (c) N. Shinohara,
J. Haga, T. Yamazaki, T. Kitazume and S. Nakamura, J.
Org. Chem., 1995, 60, 4363; (d) T. Katagiri, M. Irie and
K. Uneyama, Org. Lett., 2000, 2, 2423; (e) T. Katagiri,
M. Irie and K. Uneyama, Tetrahedron: Asymmetry, 1999, 10,
2583; (f) T. Katagiri, S. Yamaji, M. Handa, M. Irie and
K. Uneyama, Chem. Commun., 2001, 2054; (g) H. Mack and
M. Hanack, Liebigs Ann. Chem., 1989, 83.
P. G. Gassman, J. Org. Chem., 1994, 59, 8157; (d)
P. G. Gassman and T. T. Tidwell, Acc. Chem. Res., 1983, 16,
2
(a) W. T. Borden, Chem. Commun., 1998, 1919; (b)
A. Greenberg, J. Liebman, W. R. Dolbier Jr, K. S. Medinger
and A. Skancke, Tetrahedron, 1983, 39, 1533.
79.
7
8
C. B. Kelly, A. M. Colthart, B. D. Constant, S. R. Corning,
L. N. E. Dubois, J. T. Genovese, J. L. Radziewicz, 16 R. Krishnamurti, D. R. Bellew and G. K. S. Prakash, J. Org.
E. M. Sletten, K. R. Whitaker and L. J. Tilley, Org. Lett.,
011, 13, 1646.
For information on the triuoromethyl group and its use in
medicinal chemistry see: (a) B. E. Smart, J. Fluorine Chem.,
Chem., 1991, 56, 984.
2
17 (a) L. Chow, M. McClure and J. White, Org. Biomol. Chem.,
2004, 2, 648; (b) D. J. Ager, I. Fleming and S. K. Patel, J.
Chem. Soc., Perkin Trans. 1, 1981, 2520.
9
2
3
001, 109, 3; (b) K. L. Kirk, Org. Process Res. Dev., 2008, 12, 18 D. J. Roberts, J. Org. Chem., 1971, 36, 1913.
05; (c) K. L. Kirk, J. Fluorine Chem., 2006, 127, 1013; (d) 19 While the p-toluenesulfonic esters did react, they reacted far
K. M u¨ ller, C. Faeh and F. Diederich, Science, 2007, 317,
too slowly to be practical. Hence, we utilized a more reactive
sulfonate ester leaving group.
1
5
881; (e) M. Schlosser, Angew. Chem., Int. Ed., 2006, 45,
432; (f) S. Purser, P. R. Moore, S. Swallow and 20 T. H. Lowery and K. S. Richardson, Mechanism and Theory in
V. Gouverneur, Chem. Soc. Rev., 2008, 37, 320; (g) J.-A. Ma
and D. Cahard, Chem. Rev., 2008, 108, PR1–PR43.
0 For information about cyclopropanes and their use in
Organic Chemistry, Harper & Row, New York, 3rd edn, 1987.
21 D. F ˘a rcasiu, J. J ¨a hme and C. R u¨ chardt, J. Am. Chem. Soc.,
1985, 107, 5717.
1
medicinal or natural product chemistry see: (a) 22 (a) G. A. Olah, J. Am. Chem. Soc., 1972, 94, 808; (b) R. Reddy
W. A. Donaldson, Tetrahedron, 2001, 57, 8589; (b) R. Faust, and D. A. Klumpp, Chem. Rev., 2013, 113, 6905.
Angew. Chem., Int. Ed., 2001, 40, 2251; (c) J. Sala u¨ n, Top. 23 D. J. Raber, W. C. Neal, M. D. Dukes, J. M. Harris and
Curr. Chem., 2000, 207, 1. D. L. Mount, J. Am. Chem. Soc., 1978, 100, 8137.
1 (a) US Pat., 003818A1, 2011; (b) WO Pat., 082725A1, 2008; (c) 24 (a) V. J. Shiner Jr and J. G. Jewett, J. Am. Chem. Soc., 1964, 86,
1
WO Pat., 034731A1, 2008; (d) I. K. Sebhat, C. F. Franklin,
M. M.-C. Lo, D. Chen, J. P. Jewell, R. Miller, J. Pang,
945; (b) V. J. Shiner Jr, B. L. Murr and G. Heinemann, J. Am.
Chem. Soc., 1963, 85, 2413.
O. Palyha, Y. Kan, T. M. Kelly, X.-M. Guan, D. J. Marsh, 25 X. Creary, Chem. Rev., 1991, 91, 1625.
J. A. Kosinski, J. M. Metzger, K. Lyons, J. Dragovic, 26 X. Creary and E. D. Kochly, J. Org. Chem., 2009, 74, 2134.
This journal is © The Royal Society of Chemistry 2014
Chem. Sci.

View Article Online
Chemical Science
Edge Article
27 (a) D. D. Davis and R. H. Black, J. Organomet. Chem., 1974, 82, 34 (a) H. P. Hratchian and H. B. Schlegel, J. Chem. Phys., 2004,
C30; (b) D. D. Davis and H. T. Johnson, J. Am. Chem. Soc.,
974, 96, 7576.
8 M. J. Frisch, et al., Gaussian-09, Gaussian, Inc., Pittsburgh
PA, 1998.
120, 9918; (b) H. P. Hratchian and H. B. Schlegel, J. Chem.
Theory Comput., 2005, 1, 61; (c) C. E. Dykstra, G. Frenking,
K. S. Kim and G. Scuseria, Theory and Applications of
Computational Chemistry: The First 40 Years, Elsevier
Science, Amsterdam, 1st edn, 2005.
1
2
2
9 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) A. D. Becke,
J. Chem. Phys., 1993, 98, 1372; (c) C. Lee, W. Yang and 35 Consistent with these results was location of a uncyclized
R. G. Parr, Phys. Rev. B, 1988, 37, 785.
3
intermediate cation for the a-CH g-silyl system analogous
3
0 (a) R. Ditcheld, W. J. Hehre and J. A. Pople, J. Chem. Phys.,
to Xa but only a cyclized ground state for the a-CH
g-stannyl analogue.
3
1971, 54, 724; (b) W. J. Hehre, R. Ditcheld and J. A. Pople, J.
Chem. Phys., 1972, 56, 2257; (c) P. C. Hariharan and 36 (a) D. M. Lemal and L. H. Dunlap Jr, J. Am. Chem. Soc., 1972,
J. A. Pople, Theor. Chim. Acta, 1973, 28, 213.
1 (a) S. Miertu ˇs , E. Scrocco and J. Tomasi, Chem. Phys., 1981,
94, 6562; (b) M. W. Grayston and D. M. Lemal, J. Am. Chem.
Soc., 1976, 98, 1278; (c) D. Wirth and D. M. Lemal, J. Am.
Chem. Soc., 1982, 104, 847; (d) D. M. Lemal, J. V. Staros
and V. Austel, J. Am. Chem. Soc., 1969, 91, 3373; (e)
A. Greenberg, J. F. Liebman and D. V. Vechten,
Tetrahedron, 1980, 36, 1161.
3
5
1
5, 117; (b) R. Cammi and J. Tomasi, J. Chem. Phys., 1994,
00, 7495.
3
2 W. J. Hehre, R. Ditcheld, L. Radom and J. A. Pople, J. Am.
Chem. Soc., 1970, 92, 4796.
33 (a) J. Kruszewski and T. M. Krygowski, Tetrahedron Lett.,
1
972, 13, 3839; (b) A. Kostenko, B. M u¨ ller, F.-P. Kaufmann,
Y. Apeloig and H.-U. Siehl, Eur. J. Org. Chem., 2012, 1730.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2014