On the mechanism of ylide-mediated cyclopropanations: Evidence for a proton-transfer step and its effect on stereoselectivity

Article abstract of DOI:10.1021/ja910631u

In this paper, we describe studies on the cyclopropanation of Michael acceptors with chiral sulfur ylides. It had previously been found that semi-stabilized sulfonium ylides (e.g., Ph-stabilized) reacted with cyclic and acyclic enones and substituted acrylates with high ee and that stabilized sulfonium ylides (e.g., ester-stabilized) reacted with cyclic enones again with high ee. The current study has focused on the reactions of stabilized sulfonium ylides with acyclic enones which unexpectedly gave low ee. Furthermore, a clear correlation of ee with ylide stability was observed in reactions with methyl vinyl ketone (MVK): ketone-stabilized ylide gave 25% ee, ester-stabilized ylide gave 46% ee, and amide-stabilized ylide gave 89% ee. It is believed that following betaine formation an unusual proton transfer step intervenes which compromises the enantioselectivity of the process. Thus, following addition of a stabilized ylide to the Michael acceptor, rapid and reversible intramolecular proton transfer within the betaine intermediate, prior to ring closure, results in an erosion of ee. Proton transfer occurred to the greatest extent with the most stabilized ylide (ketone). When the same reactions were carried out with deuterium-labeled sulfonium ylides, higher ees were observed in all cases since proton/deuteron transfer was slowed down. The competing proton transfer or direct ring-closure pathways that are open to the betaine intermediate apply not only to all sulfur ylides but potentially to all ylides. By applying this model to S-, N-, and P-ylides we have been able to rationalize the outcome of different ylide reactions bearing a variety of substituents in terms of chemo- and enantioselectivity.

Full text of DOI:10.1021/ja910631u

Published on Web 05/18/2010
On the Mechanism of Ylide-Mediated Cyclopropanations:
Evidence for a Proton-Transfer Step and Its Effect on
Stereoselectivity
Samantha L. Riches, Chandreyee Saha, Noelia Fonta´n Filgueira, Emma Grange,
Eoghan M. McGarrigle, and Varinder K. Aggarwal*
School of Chemistry, UniVersity of Bristol, Cantock’s Close, Bristol BS8 1TS, U.K.
Received December 17, 2009; E-mail: v.aggarwal@bristol.ac.uk
Abstract: In this paper, we describe studies on the cyclopropanation of Michael acceptors with chiral sulfur
ylides. It had previously been found that semi-stabilized sulfonium ylides (e.g., Ph-stabilized) reacted with
cyclic and acyclic enones and substituted acrylates with high ee and that stabilized sulfonium ylides (e.g.,
ester-stabilized) reacted with cyclic enones again with high ee. The current study has focused on the
reactions of stabilized sulfonium ylides with acyclic enones which unexpectedly gave low ee. Furthermore,
a clear correlation of ee with ylide stability was observed in reactions with methyl vinyl ketone (MVK):
ketone-stabilized ylide gave 25% ee, ester-stabilized ylide gave 46% ee, and amide-stabilized ylide gave
89% ee. It is believed that following betaine formation an unusual proton transfer step intervenes which
compromises the enantioselectivity of the process. Thus, following addition of a stabilized ylide to the Michael
acceptor, rapid and reversible intramolecular proton transfer within the betaine intermediate, prior to ring
closure, results in an erosion of ee. Proton transfer occurred to the greatest extent with the most stabilized
ylide (ketone). When the same reactions were carried out with deuterium-labeled sulfonium ylides, higher
ee’s were observed in all cases since proton/deuteron transfer was slowed down. The competing proton
transfer or direct ring-closure pathways that are open to the betaine intermediate apply not only to all
sulfur ylides but potentially to all ylides. By applying this model to S-, N-, and P-ylides we have been able
to rationalize the outcome of different ylide reactions bearing a variety of substituents in terms of chemo-
and enantioselectivity.
Introduction
the cyclopropanation of cyclic enones, albeit with low diaste-
reoselectivity in this case (Scheme 2).4,5 In all cases, high ee’s
Cyclopropanes are common motifs to a large number of
natural products and biologically active pharmaceutical agents.1
They are usually prepared by addition of metal carbenoids to
electron-rich/-neutral alkenes or by addition of ylides to electron-
deficient alkenes.2 We, and others, have shown that chiral semi-
stabilized sulfur ylides are particularly effective at promoting
asymmetric cyclopropanation of acyclic electron-deficient alk-
enes; with unsaturated esters, in particular, high diastereose-
lectivity is also observed (Scheme 1).3 Furthermore, we have
also shown that ester-stabilized sulfur ylides are effective in
were obtained, and the absolute sense of asymmetric induction
was in keeping with the model described for asymmetric
epoxidation.6 Based on these positive results, the reaction of
an ester-stabilized ylide with an acyclic enone was expected to
give a cyclopropane with both high dr and ee. We envisaged
that such methodology could be effectively utilized in the
synthesis of the cyclopropyl-containing natural products hali-
(3) For aryl-stabilized ylide cyclopropanations, see: (a) McGarrigle, E. M.;
Aggarwal, V. K. In EnantioselectiVe Organocatalysis: Reactions and
Experimental Procedures; Dalko, P. I., Ed.; Wiley-VCH: Weinheim,
2007; Chapter 10. (b) Aggarwal, V. K.; Alonso, E.; Fang, G.; Ferrara,
M.; Hynd, G.; Porcelloni, M. Angew. Chem., Int. Ed. 2001, 40, 1433–
1436. (c) Aggarwal, V. K.; Smith, H. W.; Jones, R. V. H.; Fieldhouse,
R. Chem. Commun. 1997, 1785–1786. (d) Aggarwal, V. K.; Smith,
H. W.; Hynd, G.; Jones, R. V. H.; Fieldhouse, R.; Spey, S. E. J. Chem.
Soc., Perkin Trans. 1 2000, 3267–3276. (e) Huang, K.; Huang, Z.-Z.
Synlett 2005, 1621–1623. (f) Solladie´-Cavallo, A.; Diep-Vohuule, A.;
Isarno, T. Angew. Chem., Int. Ed. 1998, 37, 1689–1691. Allyl-
stabilized ylide cyclopropanations: (g) Tang, Y.; Ye, S.; Sun, X.-L.
Synlett 2005, 2720–2730. (h) Deng, X.-M.; Cai, P.; Ye, S.; Sun, X.-
L.; Liao, W.-W.; Li, K.; Tang, Y.; Wu, Y.-D.; Dai, L.-X. J. Am. Chem.
Soc. 2006, 128, 9730–9740. (i) Ye, S.; Huang, Z.-Z.; Xia, C.-A.; Tang,
Y.; Dai, L.-X. J. Am. Chem. Soc. 2002, 124, 2432–2433. (j) Zhu, C.-
Y.; Cao, X.-Y.; Zhu, B.-H.; Deng, C.; Sun, X.-L.; Wang, B.-Q.; Shen,
Q.; Tang, Y. Chem.sEur. J. 2009, 15, 11465–11468.
(1) (a) Donaldson, W. A. Tetrahedron 2001, 57, 8589–8627. (b) Pi-
etruszka, J. Chem. ReV. 2003, 103, 1051–1070. (c) Yeung, K.-S.;
Paterson, I. Chem. ReV. 2005, 105, 4237–4313. (d) Faust, R. Angew.
Chem., Int. Ed. 2001, 40, 2251–2253. (e) Nishii, Y.; Maruyama, N.;
Wakasugi, K.; Tanabe, Y. Bioorg. Med. Chem. 2001, 9, 33–39.
(2) For reviews on the stereoselective synthesis of cyclopropanes, see:
(a) Pellissier, H. Tetrahedron 2008, 64, 7041–7095. (b) Lebel, H.;
Marcoux, J.-F.; Molinaro, C.; Charette, A. B. Chem. ReV. 2003, 103,
977–1050. (c) Nicolas, I.; Le Maux, P.; Simonneaux, G. Coord. Chem.
ReV. 2008, 252, 727–735. For selected recent examples of asymmetric
cyclopropanations, see: (d) Goudreau, S. R.; Charette, A. B. J. Am.
Chem. Soc. 2009, 131, 15633–15635. (e) Liu, W.; Chen, D.; Zhu,
X.-Z.; Wan, X.-L.; Hou, X.-L. J. Am. Chem. Soc. 2009, 131, 8734–
8735. (f) Ichinose, M.; Suematsu, H.; Katsuki, T. Angew. Chem., Int.
Ed. 2009, 48, 3121–3123. (g) Zhu, S.; Perman, J. A.; Zhang, X. P.
Angew. Chem., Int. Ed. 2008, 47, 8460–8463. (h) Wang, H.-Y.; Yang,
F.; Li, X.-L.; Yan, X.-M.; Huang, Z.-Z. Chem.sEur. J. 2009, 15,
3784–3789.
(4) Aggarwal, V. K.; Grange, E. Chem.sEur. J. 2006, 12, 568–575.
(5) For cyclopropanations using achiral, stabilized sulfur ylides, see ref
3d and references therein; see also: Payne, G. B. J. Org. Chem. 1967,
32, 3351–3355.
9
7626 J. AM. CHEM. SOC. 2010, 132, 7626–7630
10.1021/ja910631u  2010 American Chemical Society

Mechanism of Ylide-Mediated Cyclopropanations
A R T I C L E S
Scheme 1. Asymmetric Cyclopropanations with Semi-stabilized Ylides3
Scheme 3. Planned Starting Point for Syntheses of Halicholactone
and (+)-trans-Chrysanthemic Acid
Scheme 4. Asymmetric Cyclopropanations of Acyclic Enones with
Ester-Stabilized Ylide 1b
ing cyclopropanes but with alarmingly low enantioselectivity
(Scheme 4).7 The differences in enantioselectivities in reactions
of the chiral ester-stabilized ylides 1b (R1 ) OEt) and 1c (R1
) O-t-Bu) with cyclopentenone (81% and 95% ee, respectively)4
and enone 3 (24% ee) were especially striking. Some clues to
the origin of this difference in selectivity were illuminated from
our earlier study on the cyclopropanation of cyclopentenone.4
Previously, we had found that the enantioselectivity was
dependent on concentration: high concentration gave low
enantioselectivity, whereas low concentration gave high enan-
tioselectivity. These observations were accounted for by con-
sidering the fate of the betaine intermediate, which can undergo
either direct (unimolecular) ring closure leading to high enan-
tioselectivity (pathway A, Scheme 2) or bimolecular deproto-
nation/protonation by base or ylide (pathway B), which leads
to epimerization and consequently a substantial erosion in
enantioselectivity. At low concentration, the bimolecular process
B is slowed down considerably, and unimolecular ring-closure
via pathway A dominates leading to high enantioselectivity.
However, the reaction of ylide 1b with enone 3 did not
respond to changes in concentration: the same levels of selec-
tivity (24% ee) were observed at 1 and 0.01 M. If epimerization
of the betaine intermediate was still the cause of the erosion of
ee, the above experiments showed that this process was now
also unimolecular. In the case of the acyclic enone reaction,
the Z-enolate could act as the base and remove the acidic proton
intramolecularly, generating an ylide intermediate 7. Reproto-
nation on either face of the ylide and ring closure would lead
then to racemic cyclopropane (Scheme 5). In the case of cyclic
Scheme 2. Asymmetric Cyclopropanation of Cyclopentenone with
Ester-Stabilized Ylide ent-1c4
cholactone and chrysanthemic acid (Scheme 3). However, our
synthetic plans were thwarted as further methodological studies
produced surprisingly poor results. Through detailed analysis
of the reaction, we are now able to account for the factors
responsible for the chemo- and enantioselectivity in this and in
fact all reactions of stabilized ylides (S, N, and P) with enones.
(6) (a) McGarrigle, E. M.; Myers, E. L.; Illa, O.; Shaw, M. A.; Riches,
S. L.; Aggarwal, V. K. Chem. ReV. 2007, 107, 5841–5883. (b)
Aggarwal, V. K.; Winn, C. L. Acc. Chem. Res. 2004, 37, 611–620.
Aggarwal, V. K. In ComprehensiVe Asymmetric Catalysis II; Jacobsen,
E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New York, 1999; pp
679-693. (c) Aggarwal, V. K.; Richardson, J. Chem. Commun. 2003,
2644–2651. (d) Aggarwal, V. K.; Alonso, E.; Bae, I.; Hynd, G.; Lydon,
K. M.; Palmer, M. J.; Patel, M.; Porcelloni, M.; Richardson, J.;
Stenson, R. A.; Studley, J. R.; Vasse, J.-L.; Winn, C. L. J. Am. Chem.
Soc. 2003, 125, 10926–10940.
Results and Discussion
Our surprise came in reactions of the ester-stabilized ylide
1b with the acyclic enones 2 and 3, which gave the correspond-
9
J. AM. CHEM. SOC. VOL. 132, NO. 22, 2010 7627

A R T I C L E S
Riches et al.
Scheme 5. Proposed Pathways Leading to Cyclopropanes from
Enones
Table 1. Enantioselectivities Obtained in the Reaction of 2 with
Different Stabilized Ylides under Various Conditionsa
% ee
R
NaH
NaOH/H2O
NaOD/D2Ob
1
2
3
4
5
Ph
OEt
O-t-Bu
NEt2
N-i-Pr2
26
37
46
78
85
49
c
68
c
70d
88
85
92
enones, the enolate is constrained to the E geometry and so
cannot participate in intramolecular proton-transfer events.8
This proposed rationale could be easily tested by simply
changing the nature of the ylide substituent: a more anion-
stabilizing group, e.g., a ketone, should reduce the rate of direct
ring closure (pathway A) and facilitate proton transfer (pathway
B) leading to low enantioselectivity.9 In contrast, a less anion-
stabilizing group should slow the proton transfer (pathway B)
and promote ring closure (pathway A) and thus lead to high
enantioselectivity. A series of sulfonium salts bearing different
carbonyl functionalities were therefore prepared and reacted with
methyl vinyl ketone (MVK) 2 (Table 1). As expected, a marked
increase in enantioselectivity was observed with decreasing
electron-withdrawing capacity of the carbonyl group (entries
1-4).
Further confirmation that proton transfer via pathway B was
responsible for the erosion in enantioselectivity came from
kinetic isotope effects. Substituting a proton on the ylidic carbon
for a deuteron should result in a slower rate of H/D transfer
from carbon to oxygen and thus slow down pathway B. The
rate of direct ring closure (pathway A) is not expected to be as
significantly affected by the isotope change, and so an overall
increase in ee can be expected with the change in isotope.
Unfortunately, attempts to use D2-8a with NaH resulted in
an unexpected loss of D during the cyclopropanation process.
Since base-promoted H/D-exchange of sulfonium salts was
known to be rapid,10 we employed a two-phase system consist-
ing of water/CH3CN in the presence of sulfonium salt 8a,d,e,
NaOH, and MVK 2 to effect cyclopropanations. The experi-
ments were repeated using D2O/NaOD in place of H2O/NaOH,
a Yields were in the range 52-91%; see the Supporting Information
for details. b Using NaOD/D2O, the isolated product had deuterium
incorporated on the cyclopropyl carbon R to C(O)R and on the CH3
group; see the Supporting Information for details. c These standard
conditions were not effective for cyclopropanations with ester-stabilized
ylides. d Reaction time: 18 h.
Scheme 6. Asymmetric Cyclopropanation of Ethyl Acrylate 911
and the results are summarized in Table 1. Control experiments
firmly established that the protons adjacent to the sulfonium
ion and carbonyl group exchanged much more rapidly than
formation of the cyclopropane and that no exchange of any of
the cyclopropyl protons occurred under the reaction conditions
(see the Supporting Information for details).
The experiments showed that in all cases the reactions of
the deuterium-labeled ylides gave higher enantioselectivities than
their protonated analogues. This strongly supports the proposal
that deprotonation-reprotonation (probably by the enolate;
pathway B, Scheme 5) is responsible for the erosion in ee
observed.
Reaction of the amide-stabilized ylide 1e with ethyl acrylate
9 was also tested (Scheme 6), and this gave slightly lower ee
compared to MVK (compare entry 5 in Table 1). This result
also fits with our model: in this case the ester enolate is more
basic and so pathway B is enhanced.11
The competing pathways open to betaine 6 that we have
identified here now allow us to rationalize many of the sulfur
ylide reactions reported to date. Reactions of aryl-stabilized
ylides give high ee because cyclization (pathway A) is enhanced,
and proton transfer (pathway B) is strongly disfavored since an
aryl group is not able to stabilize the ylide sufficiently. Reaction
of the stabilized sulfur ylide with enone 10 gave intermediate
levels of enantioselectivity compared to MVK 2 (Scheme 7).
In this case, the gem-dimethyl group in the chain of the betaine
(7) Huang reported two examples of low ee’s in cyclopropanations of
acyclic substrates using a stabilized ylide: ref 3e.
(8) As part of a computational study on sulfur ylide cyclopropanations,
in partial geometry optimizations of betaine intermediates, Sunoj noted
the enolate oxygen could deprotonate ylidic hydrogen atoms but there
was no further discussion of mechanistic possibilities: Janardanan, D.;
Sunoj, R. B. J. Org. Chem. 2007, 72, 331–341.
(9) Calculations on transition states for ring-closure of betaines derived
from sulfur ylides and aldehydes and for analogous intermolecular
substitution reactions show a greater degree of SN1 character in the
intramolecular case, explaining the lower ring-closing reactivity with
EWGs adjacent to the leaving group; see: Aggarwal, V. K.; Charmant,
J. P. H.; Fuentes, D.; Harvey, J. N.; Hynd, G.; Ohara, D.; Picoul, W.;
Robiette, R.; Smith, C.; Vasse, J.-L.; Winn, C. L. J. Am. Chem. Soc.
2006, 128, 2105–2114.
(11) Of course, the ester enolate should also be more nucleophilic than the
ketone enolate, but evidently the difference in enolate reactivity is
more pronounced in proton transfer than in alkylation.
(10) Greenberg, F. H.; Schulman, E. M. J. Org. Chem. 1993, 58, 5853–
5854.
9
7628 J. AM. CHEM. SOC. VOL. 132, NO. 22, 2010

Mechanism of Ylide-Mediated Cyclopropanations
A R T I C L E S
Scheme 7. Cyclopropanation of Enone 10 with Ester-Stabilized
Ylide 1b
Table 2. Asymmetric Cyclopropanation with Ammonium Ylides14c
Scheme 8. Comparison of the Reactivity of Stabilized and
Nonstabilized Phosphorus Ylides with R,ꢀ-Unsaturated Esters13
entrya
R1
R2
% ee
1b
2
3
O-t-Bu
OBn
Ph
Ph
71
80
86
97
O-t-Bu
O-t-Bu
NEt2
4
Ph
a Yields: 75-96%. b 1 equiv of 11 was used.
Scheme 9. Ammonium Ylide Cyclopropanation with 12 and D2-12
intermediate. Indeed, compared to sulfonium salts, ammonium
salts are at least 6 orders of magnitude less acidic.15 Neverthe-
less, a subtle effect on ee was observed according to the nature
of the Michael acceptor and ylide substituent: higher ee was
observed with the less anion-stabilizing group (Table 2).
However, this was not due to competing proton transfer in the
intermediate betaine since we have found that both 12 and D2-
12 gave cyclopropanes 13 and D-13 with essentially the same
ee (Scheme 9). The small differences in ee observed between
the different substrates (Table 2) could be due to one or more
of the following: (i) the varying degrees of background
cyclopropanation (Darzens-type reaction of the halocarbonyl
compound with the Michael acceptor)16 (ii) the varying degrees
of competing alkylation on the quinoline nitrogen and subse-
promotes ring closure (pathway A) through the Thorpe-Ingold
effect, accounting for the higher levels of selectivity observed.
The analysis of the different competing pathways open to
the betaine intermediate (proton transfer versus ring-closure)
can be extended to other ylides and nicely accounts for the
observed results. For example, stabilized phosphorus ylides do
not react with Michael acceptors to give cyclopropanes (Scheme
8). Following conjugate addition, proton transfer occurs instead
to give a new stabilized phosphorus ylide.13 This observation
can be rationalized according to the competing pathways open
to the intermediate betaine. The extremely poor leaving group
ability of the phosphonium ion12 coupled with its ability to
stabilize negative charge results in very slow ring closure
(pathway A) and rapid proton transfer (pathway B). In contrast,
nonstabilized phosphorus ylides do react with Michael acceptors
to give cyclopropanes.13 In this case, the absence of an EWG
serves to inhibit proton transfer (pathway B) and the EDG
promotes ring-closure (pathway A), thus leading to cyclization.
At the opposite end of ylide reactivity, Gaunt and Ley have
shown that ammonium ylides derived from cinchona alkaloids
(e.g., 11) are highly effective in asymmetric cyclopropanations
of Michael acceptors.14 In this case, proton transfer in the betaine
intermediates is expected to be much less favored than ring
closure due to the lower acidity of the C-H protons as a result
of the poorer ability of the ammonium ion to stabilize the ylide
(14) (a) Papageorgiou, C. D.; Ley, S. V.; Gaunt, M. J. Angew. Chem., Int.
Ed. 2003, 42, 828–831. (b) Bremeyer, N.; Smith, S. C.; Ley, S. V.;
Gaunt, M. J. Angew. Chem., Int. Ed. 2004, 43, 2681–2684. (c)
Papageorgiou, C. D.; Cubillo de Dios, M. A.; Ley, S. V.; Gaunt, M. J.
Angew. Chem., Int. Ed. 2004, 43, 4641–4644. (d) Johansson, C. C. C.;
Bremeyer, N.; Ley, S. V.; Owen, D. R.; Smith, S. C.; Gaunt, M. J.
Angew. Chem., Int. Ed. 2006, 45, 6024–6028. (e) Gaunt, M. J.;
Johansson, C. C. C. Chem. ReV. 2007, 107, 5596–5605.
(15) pKa values for stabilized ylides of nitrogen, sulfur, and phosphorus
(in DMSO): (a) Cheng, J.-P.; Liu, B.; Zhao, Y.; Sun, Y.; Zhang, X.-
M.; Lu, Y. J. Org. Chem. 1999, 64, 604–610. (b) Zhang, X. M.;
Bordwell, F. G.; Van Der Puy, M.; Fried, H. E. J. Org. Chem. 1993,
58, 3060–3066. (c) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456–
463. (d) For the predicted pKa in DMSO for a range of ylides, see:
Fu, Y.; Wang, H.-J.; Chong, S.-S.; Guo, Q.-X.; Liu, L. J. Org. Chem.
2009, 74, 810-819.
(12) Aggarwal, V. K.; Harvey, J. N.; Robiette, R. Angew. Chem., Int. Ed.
2005, 44, 5468–5471.
(13) (a) Bestmann, H. J.; Seng, F. Angew. Chem., Int. Ed. 1962, 1, 116.
See also: (b) Redon, S.; Leleu, S.; Pannecoucke, X.; Franck, X.;
Outurquin, F. Tetrahedron 2008, 64, 9293–9304. (c) Kondratov, I. S.;
Gerus, I. I.; Furmanova, M. V.; Vdovenko, S. I.; Kukhar, V. P.
Tetrahedron 2007, 63, 7246–7255.
(16) Kojima, S.; Suzuki, M.; Watanabe, A.; Ohkata, K. Tetrahedron Lett.
2006, 47, 9061–9065.
9
J. AM. CHEM. SOC. VOL. 132, NO. 22, 2010 7629

A R T I C L E S
Riches et al.
Scheme 10. Mechanistic Investigations by Crudden Demonstrating
That Proton Transfer Events Can Intervene in Sulfur
Ylide-Mediated Epoxidations
Scheme 11. Factors Affecting the Outcome of Ylide
Cyclopropanation
Michael acceptors which compromises the enantioselectivity of
the process. From ketone- to amide-stabilized ylides, the
enantioselectivity varies from 25-89% ee. Thus, following
addition of a ketone-stabilized ylide to the Michael acceptor,
rapid and reversible intramolecular proton transfer within the
betaine intermediate, prior to ring closure, results in an erosion
of ee. With the less acidic amide-stabilized ylides, the rate of
intramolecular proton transfer is reduced and so higher ee’s are
observed. The competing proton-transfer or direct ring-closure
pathways that are open to the betaine intermediate apply not
only to all sulfur ylides but potentially to all ylides. Phosphorus
ylides suffer from considerably higher barriers to ring closure
(leaving group ability of the onium ion falls in the order S > N
. P),12 and as such, stabilized phosphorus ylides undergo proton
transfer only. In contrast, ammonium ylides, which are least
able to undergo proton transfer (ability of the onium ion to
stabilize negative charge falls in the order S ∼ P . N), are
much better able to deliver high enantioselectivities than sulfur
ylides, although their scope is more limited. Through consid-
eration of the pathways open to the betaine intermediates and
their relative rates according to the leaving group ability of the
onium ion and its ability to stabilize negative charge (pKa) allows
us to rationalize and now predict the outcome of different ylide
reactions in terms of chemo- and enantioselectivity (Scheme
11).
quent cyclopropanation17 or (iii) the inherent enantioselectivities
with the substrates in question.
That proton transfer events can intervene in sulfur-ylide
mediated epoxidations under certain conditions has also been
demonstrated by Crudden and co-workers (Scheme 10).18 They
showed that a syn-R-hydroxysulfonium salt 14 could be depro-
tonated to form the corresponding ylide 15 (as well as the syn-
betaine 16 which led to cis-stilbene oxide). Reprotonation of
the ylidic carbon then mostly led to the anti-betaine 17 (giving
trans-epoxide) as well as giving a small amount of syn-betaine
(giving cis-epoxide). The anti-R-hydroxysulfonium salts only
gave trans-epoxides. Thus syn-betaines can lead to trans-
epoxides without reversion to ylides (e.g., 18) if proton-transfer
processes are in operation.
Acknowledgment. S.L.R. thanks EPSRC, C.S. thanks the ORS
and the University of Bristol, and N.F.F. thanks the Ministerio de
Educacio´n y Ciencia (MEC) for an FPU fellowship. V.K.A. thanks
EPSRC for a Senior Research Fellowship and Merck for research
support.
Conclusions
In summary we have identified an unexpected proton transfer
step in the sulfur ylide cyclopropanation reaction of acyclic
(17) Gaunt has found that placing a Me substituent at the 2-position of the
quinoline ring enhances the enantioselectivity and yield in intramo-
lecular cyclopropanation reactions. See ref 14d. Attempts to prepare
a pure ammonium salt derived from 11 and phenacyl bromide were
not successful, suggesting that competing alkylation at the quinoline
ring was indeed a competing process.
Supporting Information Available: Experimental procedures,
full characterization, and additional experiments. This material
is available free of charge via the Internet at http://pubs.acs.org.
(18) Edwards, D. R.; Du, J.; Crudden, C. M. Org. Lett. 2007, 9, 2397–
2400.
JA910631U
9
7630 J. AM. CHEM. SOC. VOL. 132, NO. 22, 2010