Practical and highly selective sulfur ylide-mediated asymmetric epoxidations and aziridinations using a cheap and readily available chiral sulfide: Extensive studies to map out scope, limitations, and rationalization of diastereo- and enantioselectivities

Article abstract of DOI:10.1021/ja405073w

The chiral sulfide, isothiocineole, has been synthesized in one step from elemental sulfur, γ-terpinene, and limonene in 61% yield. A mechanism involving radical intermediates for this reaction is proposed based on experimental evidence. The application of isothiocineole to the asymmetric epoxidation of aldehydes and the aziridination of imines is described. Excellent enantioselectivities and diastereoselectivities have been obtained over a wide range of aromatic, aliphatic, and α,β-unsaturated aldehydes using simple protocols. In aziridinations, excellent enantioselectivities and good diastereoselectivities were obtained for a wide range of imines. Mechanistic models have been put forward to rationalize the high selectivities observed, which should enable the sulfide to be used with confidence in synthesis. In epoxidations, the degree of reversibility in betaine formation dominates both the diastereoselectivity and the enantioselectivity. Appropriate tuning of reaction conditions based on understanding the reaction mechanism enables high selectivities to be obtained in most cases. In aziridinations, betaine formation is nonreversible with semistabilized ylides and diastereoselectivities are determined in the betaine forming step and are more variable as a result.

Full text of DOI:10.1021/ja405073w

Article
pubs.acs.org/JACS
Practical and Highly Selective Sulfur Ylide-Mediated Asymmetric
Epoxidations and Aziridinations Using a Cheap and Readily Available
Chiral Sulfide: Extensive Studies To Map Out Scope, Limitations, and
Rationalization of Diastereo- and Enantioselectivities
Ona Illa,†,§ Mariam Namutebi,† Chandreyee Saha,† Mehrnoosh Ostovar,† C. Chun Chen,†
Mairi F. Haddow,† Sophie Nocquet-Thibault,† Matteo Lusi,† Eoghan M. McGarrigle,*,†,‡
and Varinder K. Aggarwal*,†
†School of Chemistry, University of Bristol, Cantock’s Close BS8 1TS, United Kingdom
‡Centre for Synthesis and Chemical Biology, UCD School of Chemistry and Chemical Biology, University College Dublin, Belfield
Dublin 4, Ireland
S
* Supporting Information
ABSTRACT: The chiral sulfide, isothiocineole, has been synthe-
sized in one step from elemental sulfur, γ-terpinene, and limonene in
61% yield. A mechanism involving radical intermediates for this
reaction is proposed based on experimental evidence. The
application of isothiocineole to the asymmetric epoxidation of
aldehydes and the aziridination of imines is described. Excellent
enantioselectivities and diastereoselectivities have been obtained over
a wide range of aromatic, aliphatic, and α,β-unsaturated aldehydes
using simple protocols. In aziridinations, excellent enantioselectivities
and good diastereoselectivities were obtained for a wide range of imines. Mechanistic models have been put forward to rationalize
the high selectivities observed, which should enable the sulfide to be used with confidence in synthesis. In epoxidations, the
degree of reversibility in betaine formation dominates both the diastereoselectivity and the enantioselectivity. Appropriate tuning
of reaction conditions based on understanding the reaction mechanism enables high selectivities to be obtained in most cases. In
aziridinations, betaine formation is nonreversible with semistabilized ylides and diastereoselectivities are determined in the
betaine forming step and are more variable as a result.
are not 1,2-diaryl epoxides (Figure 1).6−8 Table 1 shows the
INTRODUCTION
■
demonstrated ability of these 11 sulfides to deliver epoxides in
The direct asymmetric transformation of carbonyl compounds
>90% ee from different ylide/aldehyde combinations.6,7 Being
into epoxides using chiral sulfur ylides offers a complementary
able to also control diastereoselectivity is critical to the practical
and potentially advantageous method over the two-step
usage of the technology. To the best of our knowledge, only 3
sulfides (1, 2, and 10) have been shown to give >90:10
diastereoselectivity with >90% ee in epoxidations of aliphatic
aldehydes.
protocol of Wittig olefination followed by asymmetric
epoxidation.1−5 However, despite its appeal and over 30 years
of research, the methodology has rarely been used. Herein, we
detail results that make the sulfur ylide disconnection a genuine
alternative to alkene epoxidation for practical asymmetric
epoxidation, which can be incorporated into a synthetic plan
(ii) Sulfide availability. The sulfides that deliver high
enantioselectivity usually require multistep synthesis.1 The
number of steps required for each sulfide synthesis is shown in
with confidence.
Figure 1. Furthermore, in a number of cases, the chiral pool
The previous lack of use of the sulfur ylide disconnection can
starting material is only readily available in one enantiomeric
be attributed to two main factors:
(i) Limited demonstrated substrate scope. The majority of
form, which clearly limits the application of such sulfides. The
́
examples of 1 and 7 are illustrative. Solladie-Cavallo has
asymmetric, sulfur ylide-mediated epoxidations have been used
to prepare 1,2-diaryl epoxides, which have limited synthetic
utility. A survey of more than 80 publications with reports of
sulfur ylide asymmetric epoxidations found that just 22 chiral
sulfides (Chart 1, Supporting Information) show enantiose-
lectivities of >90% enantiomeric excess (ee) in the preparation
of 1,2-diaryl epoxides. However, in aldehyde epoxidations, only
11 sulfides have been shown to give >90% ee for epoxides that
reported many examples of asymmetric epoxidations using 7,7i,j
giving very high ee’s (>95%) over a range of substrates, but the
sulfide was derived in three steps9 from pulegone, which is only
readily available in one enantiomeric form. We are only aware
Received: May 21, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Sulfides that mediated asymmetric epoxidations of aldehydes giving >90% ee for epoxides other than 1,2-diarylepoxides.6,7 The number of
steps to synthesize the sulfides from commercially available precursors is given (sulfide 2 is now commercially available).17
Table 1. Scope of Sulfur Ylide-Mediated Asymmetric Epoxidation of Aldehydes−Sulfides with >90% ee in the Synthesis of
a
,
Epoxides Other than 1,2-Diaryl Epoxides6,7
ylide
benzyl
aldehyde
aliphatic
sulfide
1
2
3
4
5
6
7
8
9
10
11
Y*
Y*
Y*
Y*
Y
Y*
Y*
Y*
Y*
Y
Y
Y
Y
aromatic
Y*
Y*
Y*
Y*
Y
Y
Y
Y*
Y*
Y*
heteroaromatic
vinyl
Y*
Y
alkynyl
formaldehyde
aliphatic
Y
b
allyl
Y
Y*
Y*
aromatic
Y*
c
Y
Y
alkyl (intramol)
methyl
aliphatic
Y
aromatic
Y
Y
Y
heteroaromatic
aliphatic
amido
Y*
Y*
Y*
aromatic
Y*
heteroaromatic
a
b
An asterisk indicates a diastereomeric ratio (dr) >90:10 (favoring trans-epoxide). Some of the entries for sulfide 2 are reported in this paper. dr not
reported.6i dr n/a.
c
of two reports of its use in asymmetric epoxidation by a group
models to account for selectivity, now provide a genuine,
practical methodology that can be applied in synthesis.
other than the Solladie-
́
Cavallo group.10 Similarly, we reported
a sulfide, 1, which gave high ee’s (>95%) over a range of
substrates and reported its application to a range of synthetic
targets.6,11,12 However, it requires four synthetic steps from
camphorsulfonyl chloride (available in both enantiomeric
forms),13 and although we have reported its synthesis on
multigram-scale,14 we are only aware of two reports by groups
other than our own using this sulfide in asymmetric
epoxidations.6h,i,11,15,16
We recently reported a chiral sulfide, isothiocineole 2, which
simultaneously addressed both of these limitations.7a The
sulfide was easily prepared in one step from limonene and
elemental sulfur and delivered the highest combined outcome
in terms of enantioselectivity and diastereoselectivity in
epoxidations and aziridinations of any sulfide to date. In this
paper, we describe (i) substantial improvements in the
synthesis of the sulfide, (ii) enhanced scope of ylide reactions
in terms of the ylide substituents (aryl, alkenyl) and the
aldehyde (aromatic, heteroaromatic, α,β-unsaturated, and
aliphatic) and imine components, and (iii) models to account
for the diastereo- and enantioselectivity of the reactions. We
believe these significant improvements, underpinned by the
RESULTS AND DISCUSSION
■
Sulfide Synthesis. In the search for a suitable chiral sulfide,
we were attracted to the little-known bicyclic compound
isothiocineole 2, as it seemed to fulfill many of the criteria
established as desirable.1,2 In terms of enantioselectivity
(Scheme 1):
(i) Its rigid bicyclic structure would dictate the position of
the ylide substituent in relation to the sulfide scaffold
(lone pair selectivity);
(ii) Its rigid bicyclic structure would control the conforma-
tion of the ylide through nonbonded steric interactions;
(iii) One of the two gem-dimethyl groups should block one
face of the ylide leading to high enantioselectivity.
In terms of preparation, Weitkamp had reported a one-step
synthesis of isothiocineole from the simplest and cheapest of
reagents, elemental sulfur and limonene.18 Heating the two
components followed by distillation and separation of
isothiocineole 2 from dehydroisothiocineole 13 by thiourea
co-crystallization gave the target molecule in 20% yield and
B
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
hydrogen shifts, which would avoid losing limonene to various
aromatic byproducts.
Scheme 1. Design Features of Isothiocineole for
Enantioselectivity in Asymmetric Epoxidation
It was found that by adding 1.0 equiv of γ-terpinene 14, the
reaction between elemental sulfur and limonene could indeed
be conducted at 125 °C and the formation of 13 was
completely suppressed. Furthermore, simple distillation of the
crude reaction mixture furnished essentially pure isothiocineole
in much improved yield (36%), but more importantly, now
without racemization (99:1 er).7a Further optimization of
stoichiometry, time, and temperature has led to further
significant improvements including conducting the reaction at
125 °C and adding the sulfur (1.0 and 0.8 equiv) and γ-
terpinene (1.0 and 1.1 equiv) in two portions (the second after
8 h) gave >95% conversion of limonene after 24 h (Scheme 3).
Scheme 3. Improved Synthesis of Isothiocineole 2 with
Added Equivalents of Reagents after 8 h
94:6 enantiomeric ratio (er), a reaction we were able to
reproduce (Scheme 2). Despite being known for over 50 years,
isothiocineole’s potential utility in synthesis was not recognized.
The reaction had been conducted on scales from 0.2 to 2300
mol limonene. Although the reaction had been operated on
>100 gal scale,18b,19 further improvements were required to
improve the yield and to avoid partial racemization and
formation of the side product 13, which was difficult to remove.
This was challenging because of the lack of mechanistic
information and because substantial optimization would have
been conducted prior to conducting the reaction on such a vast
scale. In our initial analysis, we noted that the conversion of
limonene into isothiocineole 2 requires the overall addition of
sulfur and two hydrogens (Scheme 2). Although sulfur is clearly
added, the source of the two hydrogens is actually limonene
itself, which, in the process of liberating the hydrogens,
becomes converted to various aromatic byproducts (e.g., p-
cymene). The generation of the two hydrogens from limonene
is evidently not very efficient because it requires high
temperatures and thus results in significant formation of the
unsaturated sulfide 13, which requires the addition of sulfur
only. Therefore, a more efficient source of the two hydrogens
was required to allow the reaction to run at lower temperatures
and to limit the formation of 13.
This ultimately gave isothiocineole 2 in an improved 61% yield,
even on a mole scale. With such inexpensive reagents, a simple
protocol and facile isolation, (+)-isothiocineole is now easily
obtained. The (−)-isomer can also be accessed with equal ease
by the same method but with lower er (90:10) because
(−)-limonene is only available commercially as a 90:10 mixture
of enantiomers. Nevertheless, low temperature recrystallization
(−50 °C) from pentane (twice) can be used to upgrade this
material to >98:2 er, if required.17,21
Mechanism for Formation. We propose the mechanism
shown in Scheme 4 for the formation of isothiocineole 2 from
limonene.22 First, elemental sulfur and γ-terpinene 14 interact
at elevated temperature and form a thiol radical.23 Then, the
sulfur-centered radical adds to the cyclic alkene of limonene to
give the more stable tertiary radical 16,24 which reacts with the
hydrogen atom donor, γ-terpinene, leading to the all-equatorial,
thermodynamic intermediate 17. Addition to the exocyclic
alkene may occur but does not lead to the desired product. In
the absence of a good hydrogen source (γ-terpinene), 16 loses
a hydrogen atom to give alkene 18. Both 17 and 18 suffer
reversible loss of sulfur to generate radicals 19 and 20,
respectively, which cyclize to give the products isothiocineole 2
and dehydroisothiocineole 13.
We believed that limonene undergoes a series of 1,3-
hydrogen shifts at high temperature leading to a key
intermediate, γ-terpinene 14 (Scheme 2). Because it is a 1,4-
cyclohexadiene, γ-terpinene should be a good hydrogen donor
and thus be able to contribute to the formation of
isothiocineole, itself being converted into p-cymene as a
byproduct. Therefore, we decided to add γ-terpinene20 directly
to the reaction mixture because we expected that this would
allow us to avoid the high temperatures required for 1,3-
Scheme 2. Weitkamp’s Synthesis of Isothiocineole18
C
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 4. Plausible Mechanism for Isothiocineole 2 Formation
Evidence for this proposal comes from the following
observations and literature examples:
(i) Radical cyclization of 1-p-menthene-8-thiol 21 (report-
edly the most powerful flavor compound ever found in
nature,25) gave sulfide 22 exclusively in which the methyl
group is oriented in an equatorial rather than an axial position
(Scheme 5).25 This indicates that in the formation of
isothiocineole, the order of events must be addition of the
thiol radical to the endocyclic alkene first, followed by
intramolecular cyclization, not initial addition to the exocyclic
alkene. Furthermore, it is hard to see how the thiol radical
could add to the exocyclic alkene of limonene to generate thiol
21 because it would be expected to add anti-Markovnikov
instead.26
(ii) In the presence of the hydrogen donor γ-terpinene, no
racemization occurred. We believe that the source of
racemization in the absence of γ-terpinene is thermal
isomerization of the alkenes in limonene. After the first 1,3-
hydrogen shift to give 15, a subsequent 1,3-hydrogen shift will
lead to γ-terpinene (Scheme 2). However, 15, which is achiral,
could undergo the reverse of the first 1,3-hydrogen shift and
give racemic limonene. This is the likely source of the small
amount of racemization observed at elevated temperature and
in the absence of γ-terpinene.
Scheme 5. Formation of Thiocineole 22 from Thiol 21 under
Radical Conditions25
Table 2. Reactions of Benzyl Sulfonium Salt 23a with Aldehydes
a
b
entry
aldehyde
benzaldehyde
method
yield (%)
dr
er
1
A
A
A
A
A
A
A
A
A
B
B
B
86
89
86
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
>95:5
93:7
99:1
98:2
99:1
99:1
97:3
98:2
98:2
98:2
99:1
99:1
99:1
2
p-methoxybenzaldehyde
p-cyanobenzaldehyde
3
c
4
(E)-cinnamaldehyde
88
c
5
(E)-crotonaldehyde
86
c
6
(E)-PhCHC(Me)CHO
2-pyridinecarboxaldehyde
3-pyridinecarboxaldehyde
furan-3-carbaldehyde
84
7
85
80
71
8
9
d
10
11
12
13
14
cyclohexanecarboxaldehyde (Cy)
valeraldehyde (Val, n-BuCHO)
pivaldehyde (t-BuCHO)
TIPS-propargyl aldehyde
TIPS-propargyl aldehyde
62
d
56
91:9
e
0
f
d
Neat MeCN
51
68:32
60:40
97:3
g
d
A
76
a
b
c
d
e
1
trans:cis. Determined by chiral HPLC. Determined by H NMR with an internal standard. Mixture of diastereomers. Formation of 24 was
f
g
observed. KOH, rt, 3 h. rt, 3 h.
D
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 3. Effect of Protic Solvent on Reactions of Benzyl Sulfonium Salt 23a with Cyclohexanecarboxaldehyde
a
b
entry
solvent
yield (%)
dr
er
2:24
1
2
3
4
5
MeCN
44
50
55
62
75
57
>95:5
94:6
99:1
99:1
99:1
99:1
99:1
99:1
56:44
63:37
67:33
70:30
87:13
100:0
MeCN/t-BuOH (50:1)
MeCN/t-BuOH (25:1)
MeCN/t-BuOH (15:1)
MeCN/t-BuOH (9:1)
MeCN/H2O (9:1)
94:6
93:7
91:9
6
73:27
a
b
trans:cis. Determined by chiral HPLC.
Table 4. Reactions of Electron-Rich Benzyl Sulfonium Salts with Aldehydes
b
c
entry
salt
R1
R2
R3
R
method
yield (%)
dr
er
1
23a
23a
23b
23b
23c
23c
23c
23d
23d
23e
H
H
H
H
H
H
H
H
H
H
H
Ph
Cy
Ph
Cy
Ph
Ph
Cy
Ph
Cy
Ph
Cy
A
86
62
>95:5
93:7
>99:1
99:1
99:1
98:2
94:6
96:4
98:2
98:2
99:1
98:2
92:8
2
H
H
B
a
3
OMe
OMe
H
H
A
66
>95:5
84:16
90:10
97:3
a
4
H
B
63
5
OMe
OMe
OMe
H
A
65
69
56
45
d
6
H
A (MeCN)
7
H
B
A
B
A
B
67:33
>95:5
72:28
>95:5
71:29
8
Me
Me
a
9
H
62 (43)
10
11
-(CH)4-
-(CH)4-
H
97
23e
H
62
a
b
c
1
Determined by H NMR with an internal standard (isolated yield is given in parentheses for entry 9). trans:cis. Determined by chiral HPLC.
d
Method A except MeCN was used in place of MeCN/H2O.
(iii) The mechanism for formation. The mechanism of the
reactions involving isothiocineole 2. We were especially mindful
of going beyond simple 1,2-diaryl epoxides that are commonly
evaluated, to the synthetically much more useful 1,2-alkylaryl
and α,β-unsaturated epoxides.
side product, dehydroisothiocineole 13, and in particular its
absolute stereochemistry, is consistent with the series of events
shown in Scheme 4 and does not require the occurrence of
some form of allylic shift when sulfur is added to the double
bond as previously suggested.18c
Epoxide Synthesis. Because both (+)- and (−)-isothioci-
neole can be easily prepared on large scale and have now
become commercially available, we explored the stoichiometric
epoxidation reactions of sulfur ylides as these show
considerably greater scope than the catalytic process.6b,27 For
example, the catalytic process usually leads to low yields, low
dr’s, or low er’s with aliphatic, α,β-unsaturated, heteroaromatic,
and acetylenic aldehydes.6c Further limitations of the catalytic
process were the low yields and limited substrate scope with
α,β-unsaturated hydrazones.6b Therefore, we set out to map the
scope and limitations of the stoichiometric epoxidation
Several benzyl sulfonium salts were prepared by the reaction
of benzyl bromide with isothiocineole in a two-phase mixture of
CH2Cl2 and aqueous solution of LiOTf or NaBF4. The
tetrafluoroborate salt was found to be rather insoluble in most
organic solvents and so subsequent studies focused on the
triflate salt 23a. The alkylations occurred exclusively on the exo
lone pair, which is presumably less hindered. X-ray analysis of
sulfonium salts 23a−d, f, and g confirmed their structure (see
the Supporting Information for crystal structure data).
We established two sets of conditions, Method A
(MeCN:H2O (9:1)) and Method B (MeCN:tBuOH (15:1))
for reactions with aromatic and aliphatic aldehydes, respec-
tively, which gave moderate-to-high yields and high diastereo-
E
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 5. Reaction of Electron-Deficient Benzyl Sulfonium Salts with Aldehydes
b
c
entry
salt
R1
R
method
yield (%)
94
dr
er
1
23f
23f
23f
23f
23f
23f
23g
23g
23g
23g
23g
23g
23h
23h
CF3
CF3
CF3
CF3
CF3
CF3
CN
CN
CN
CN
CN
CN
Cl
Ph
A
B
>99:1
97:3
92:8
90:10
97:3
99:1
>99:1
>99:1
95:5
92:8
97:3
99:1
98:2
92:8
64:36
87:13
92:8
a
2
Cy
36
a
3
Cy
C
D
C
D
A
B
42
4
Cy
45
99:1
a
5
n-Bu
n-Bu
Ph
63
94:6
6
58
77
98:2
7
67:33
58:42
73:27
78:22
85:15
90:10
95:5
a
8
Cy
35
a
9
Cy
C
D
C
D
A
B
40
10
11
12
13
14
Cy
43
54
69
n-Bu
n-Bu
Ph
a
76
a
Cl
Cy
49
95:5
a
b
c
1
Determined by H NMR with an internal standard. trans:cis. Determined by chiral HPLC.
and enantioselectivities (Table 2). Method A was successfully
applied to electron-rich and electron-deficient aromatics
(entries 2 and 3), α,β-unsaturated aldehydes (entries 4−6),
and heteroaromatics (entries 7−9) all leading to high yields,
and very high diastereomeric ratio (dr) and er. Method B was
successfully applied to α-branched and unbranched aliphatic
aldehydes, again with moderate-to-high dr and very high er
(entries 10 and 11). However, the more hindered pivaldehyde
(t-BuCHO, entry 12) was not successful in epoxidation. In
general, we found that, with slower reacting electrophiles, a
competing elimination reaction of the ylide occurred leading to
sulfide 24. In fact, we were unable to extend this chemistry to
cyclopropanation reactions with Michael acceptors (e.g.,
chalcone),28 which are inherently less electrophilic, again
because of competing elimination. Acetylenic aldehydes could
also be employed and led to high enantioselectivity but low
diastereoselectivity (entries 13 and 14). In this case we found
that neat MeCN gave the best selectivities. To maximize
diastereoselectivity with unhindered aldehydes, conditions are
required that maximize the extent of reversibility in betaine
formation, which requires aprotic conditions (see later for a
discussion).
In our optimization studies for aliphatic aldehydes, we found
that higher dr was obtained in less protic media, but higher
yield was obtained in more protic media. A representative set of
results, illustrating the effect of protic solvent on the reaction
with cyclohexanecarboxaldehyde, is shown in Table 3.
Particularly instructive is the ratio of sulfides 2:24 formed in
the reaction, which is a measure of the ratio of two competing
processes, epoxidation and elimination, which occurred under
the reaction conditions. With increasing protic solvent, the
yield of epoxide increased (increase in ratio of epoxidation/
elimination 2:24), but the diastereoselectivity decreased. The
use of MeCN:t-BuOH (15:1) offered the optimum balance of
yield and dr (entry 4). In fact, the dr obtained for the aliphatic
aldehydes (Table 2) represent the highest to date.
Extension of the methodology to a range of electron-rich
benzyl sulfonium salts 23a−e was evaluated and the results are
summarized in Table 4. Once again, all reactions were tested
with a representative aromatic (PhCHO) and aliphatic
(CyCHO) aldehyde. In all cases, essentially perfect enantiose-
lectivity was observed but the dr was more variable. The dr was
dependent on the electronic and steric properties of the benzyl
group and the aldehyde. In all reactions with PhCHO, high dr
was observed although the electron-rich and unhindered aryl
substrate 23c required aprotic conditions to achieve this
(entries 5 and 6). Reactions with aliphatic aldehydes led to
lower dr. This aspect is discussed in detail later.
Electron-deficient benzyl sulfonium salts 23f−h were also
explored (Table 5) and, in contrast to the results with electron-
rich salts, this time high dr but variable levels of er were
observed. To maximize enantioselectivity, reversibility in
betaine formation had to be minimized and so more protic
conditions (Method C) and low temperature with a
coordinating metal counterion (Method D) were also explored
with certain substrates. Reactions with the highly stabilized
sulfur ylides derived from 23f and g were expected to give low
er with all aldehydes, especially aromatic ones. Therefore, we
explored aliphatic aldehydes in more detail and extended our
study to include valeraldehyde (n-BuCHO), which, being the
least hindered of aldehydes, was expected to show the lowest
degree of betaine reversibility and, thus, maximum enantiose-
lectivity. In practice, reactions with aromatic aldehydes gave low
er with the highly electron-deficient benzyl sulfonium salts 23f
F
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 6. Reaction of α,β-Unsaturated Sulfonium Salts with Aldehydes
a
b
entry
salt
R1
R2
R
method
yield (%)
dr
er
c
d
d
1
2
3
4
5
6
26a
26b
26c
26d
26c
26d
H
H
Ph
Ph
Ph
Ph
Cy
Cy
A
A
A
A
B
B
57
65
97
80
77
77
75:25
80:20
>95:5
>95:5
>95:5
>95:5
70:30
85:15
e
H
Ph
Ph
H
e
c
e
c
d
Me
Me
Me
Me
99:1
99:1
98:2
d
d
Ph
H
f
97:3
a
b
c
d
e
f
1
Determined by H NMR with an internal standard. trans:cis. X = OTf. Determined by chiral HPLC. X = BF4 Determined by chiral GC.
and g as expected, but high er was observed with the less
electron-deficient benzyl sulfonium salt 23h (compare entries 1
and 7 vs 13). In contrast, even with the highly stabilized ylides
23f and g we were able to obtain both high diastereoselectivity
and high enantioselectivity with both valeraldehyde (entries 6
and 12) and cyclohexanecarboxaldehyde (entries 4 and 10)
using method D. Again, the factors that affect both dr and er are
discussed later.
Table 7. Reaction of Benzyl Sulfonium Salt with N-Ts and N-
BOC Imines
The process was also extended to α,β-unsaturated sulfonium
salts 26a−d, which were prepared either by the reaction of the
sulfide with the corresponding allylic alcohol and HBF4 or by
alkylation with the appropriate allylic bromide. Although the α-
unsubstituted allylic sulfonium salts 26a and b only gave
moderate dr and er (Table 6, entries 1 and 2), the α-substituted
allylic sulfonium salts 26c and d gave very high dr and er even
with cyclohexanecarboxaldehyde (entries 3−6). The prepara-
tion of synthetically useful vinyl epoxides in high ee and high dr
by this simple sulfur ylide disconnection is especially note-
worthy.
yield
a
b
entry
R1
Ts
R2
condition
dr
er
(%)
1
2
3
4
5
6
7
8
Ph
A
A
A
A
A
A
B
C
85:15
86:14
75:25
83:17
>99:1
87:13
97:3
99:1
99:1
99:1
99:1
98:2
99:1
98:2
99:1
72
63
65
80
78
78
52
68
Ts
p-MeC6H4
p-ClC6H4
Ts
Ts
p-MeOC6H4
(E)-PhCHCH
(E)-TMSCHCH
Ph
Ts
Ts
Aziridination.29,30 The benzyl sulfur ylide reaction was
initially tested with a range of imines bearing different
substituents and different activating groups on nitrogen (p-
toluenesulfonyl (Ts) and tert-butyloxycarbonyl (BOC)) (Table
7). In all cases essentially complete enantioselectivity was
observed although diastereoselectivity was, as expected,1,2l more
variable. With N-Ts imines derived from aromatic aldehydes,
moderate diastereoselectivity was obtained (entries 1−4),
whereas the N-BOC imine gave very high dr (entry 7).
Extension to unsaturated imines was also explored and this time
both very high er and high dr (from 83:17 to >95:5) were
observed (entries 5 and 6). The imine derived from
pivaldehyde (t-BuCHNTs) also worked (entry 8) and gave
the aziridine with high trans selectivity and again perfect er.
BOC
Ts
b
t-Bu
89:11
a
trans:cis Determined by chiral HPLC.
(Ts, P(O)Ph2; note that BOC-imines were not successful)
(Table 8). Essentially perfect er was obtained with α-
substituted allyl sulfonium salts (entries 1, 2, and 5), but
surprisingly, very high er was also observed with the α-
unsubstituted allyl sulfonium salts, in contrast to epoxidation
(entries 3, 4, and 6). Both N-Ts and N-P(O)Ph2 imines showed
similar levels of dr.
Origin of Diastereoselectivity in Epoxidation. Sulfur
ylides react with carbonyl compounds via betaine intermediates
to give epoxides. We have previously reported that the reaction
of a benzyl sulfonium ylide with an aldehyde or ketone was
remarkably finely balanced.6b,32 In reactions with benzaldehyde,
the trans-epoxide was derived from nonreversible formation of
the anti-betaine, followed by bond rotation and ring closure
(Scheme 6).33
̌
Interestingly, Hamersak obtained the cis-aziridine exclusively
with this imine using the benzyl sulfonium ylide derived from
Eliel’s oxathiane 7,31 opposite to what we observed with
isothiocineole 2. It should be noted that pivaldehyde itself
could not be employed in epoxidations because it was too
unreactive and led to competing elimination of the sulfonium
salt, indicating the higher reactivity of the N-Ts imines relative
to aldehydes.
In contrast, crossover experiments showed that the syn-
betaine, which would lead to the cis-epoxide, was formed
reversibly.33 This indicated that bond rotation and ring closure
had a higher activation barrier than that for reversion to starting
materials (relative rates: k5 < k−4). DFT calculations under-
Allylic sulfonium salts were also explored with benzaldehyde-
derived imines bearing a range of activating groups on nitrogen
G
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 8. Reaction of Allyl Sulfonium Salts with Benzaldehyde-Derived Imines
a
b
b
d
entry
R1
R2
R3
Ts
method
dr
er of trans
er of cis
yield (%)
1
2
3
4
5
6
Me
Me
H
Ph
H
D
D
D
D
E
78:22
83:17
85:15
80:20
84:16
86:14
99:1
99:1
94:6
97:3
76
72
73
81
84
83
Ts
>99:1
85:15
H
Ts
c
e
H
Ph
H
Ts
95:5
nd
Me
H
Ph2PO
Ph2PO
99:1
91:9
>99:1
90:10
H
E
a
b
c
d
trans/cis Determined by chiral HPLC on the crude mixture. Determined by chiral HPLC on the pure product. Yield of combined cis and trans
e
1
isomers. Determined by H NMR with an internal standard. Not determined.
Scheme 6. Rationalization of Diastereoselectivity in Epoxidations
pinned these experimental observations, producing the same
relative activation barriers (relative rates: k2 > k−1; k−4 > k5).32
It was found that the highest activation barrier along the two
reaction pathways was for the torsional rotation step from the
gauche to the trans conformation of the syn-betaine. Thus, the
formation of the syn-betaine is nonproductive under appro-
priate conditions; it is formed but reverts back to the aldehyde
and ylide, as subsequent rotation from the gauche to the trans
conformation has a higher activation barrier than that for
reversion to starting materials. Hence, the high trans selectivity
observed with benzaldehyde is a result of nonproductive
formation of the syn-betaine and productive formation of the
anti-betaine, not as a result of which betaine is preferentially
formed. In general, providing syn-betaine formation is reversible
and is nonproductive, high diastereoselectivity should result.
The degree of reversibility in syn-betaine formation therefore
determines the dr of the reaction and is thus critical. The
degree of reversibility is influenced in the following ways: (i) an
increase in the thermodynamic stability of the starting materials
of the torsional rotation barrier (decrease in k5) and thus
reduced reversibility leading to lower diastereoselectivity.
Of course, the factors that increase the reversibility in the syn-
betaine formation also impact on the anti-betaine formation,
and this process can, therefore, also be partially reversible.
Although this tends not to have any effect on the
diastereoselectivity, it does have important consequences for
the enantioselectivity (vide infra).
These factors can now be used to account for the selectivity
observed in the many examples provided and are discussed
below.
1. Stability of the Carbonyl Group. Aromatic aldehydes give
high trans selectivity because reversion of the syn-betaine yields
a carbonyl group that is in conjugation with an aromatic ring.
Such conjugation is not available to aliphatic aldehydes, thus,
resulting in reduced reversibility, and therefore lower dr. On the
basis of this analysis, the results in Table 2 can be broadly
understood. Aromatic (entries 1−3), heteroaromatic (entries
7−9), and unsaturated aldehydes (entries 4−6) gave high
diastereocontrol, whereas aliphatic aldehydes (entries 10 and
11) gave lower diastereoselectivities.
2. Steric Hindrance of the Ylide/Aldehyde. Reduced steric
bulk of the ylide/aldehyde allows more facile bond rotation
from the gauche to the trans conformation of the betaine,
leading to reduced reversibility in betaine formation thereby
resulting in a decrease in diastereoselectivity. Conversely, an
increase in steric hindrance of the ylide/aldehyde leads to an
(ylide and aldehyde) will lead to greater reversibility in betaine
formation (increase in k−4) and thus higher diastereoselectivity,
(ii) increasing the steric bulk of the ylide or aldehyde will give
rise to an increase in the torsional rotation barrier (increase in
k5) and thus render betaine formation more reversible, resulting
in increased diastereoselectivity, (iii) increased solvation of the
alkoxide by metals or a protic solvent will result in the lowering
H
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Influence of steric properties of sulfide and aldehyde on dr (trans:cis) of epoxidation reaction with benzyl sulfur ylides.6a,c,7a,d
Figure 3. Influence of steric properties of sulfide and substitution of allyl moiety on dr (trans:cis) and er of epoxidation of benzaldehyde.6b,c,7a,c
Figure 4. Comparison of different aryl stabilized ylides on dr of reactions with aromatic and aliphatic aldehydes (all results obtained with KOH as
base at 0 °C).
increase in diastereoselectivity. Thus, propargylic aldehydes
give low dr, whereas aliphatic aldehydes of increasing steric bulk
showed increasing levels of diastereocontrol (Figure 2).
A comparison of different sulfides of increasing steric
hindrance, employed in epoxidations with CyCHO is also
shown in Figure 2. Isothiocineole 2 is clearly a hindered sulfide.
Its steric bulk leads to an increase in the barrier to bond
rotation of the intermediate betaines and a decrease in the
barrier to reversion to its constituents. Figure 2 illustrates the
record levels of dr obtained with isothiocineole 2.
The α-substituted allylic sulfonium ylides also gave very high
diastereoselectivity, presumably because they show similar
steric properties to an aromatic group. In the absence of the
α-substituent, lower dr was observed. Once again, in
I
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
comparison to other sulfides, isothiocineole provides record
levels of combined diastereo- and enantiocontrol, most likely
because of its steric bulk (Figure 3).
anti-betaine intermediate, followed by bond rotation and
subsequent ring closure (Scheme 7).35
In contrast, the cis-aziridine was formed from a transoid
addition of the ylide to the imine to give the syn-betaine
intermediate, followed by direct ring closure. However, the
differences between the energies of the barriers of the key TSs
leading to the syn and anti-betaines and therefore the cis- and
trans-aziridines in the model systems used in the calculations
were relatively small, reflecting the low diastereoselectivity
generally observed. Clearly, these systems are finely balanced,
and it is difficult to predict what the outcome will be for a given
substrate. The moderate trans selectivity observed with N-Ts
imines derived from aryl aldehydes and the unsaturated imine is
a reflection of the energy differences between the two addition
TSs for formation of the anti and syn-betaines (Table 7, entries
1−4 and 6). It is difficult to explain why the unsaturated imine
derived from cinnamaldehyde gave such high diastereoselectiv-
ity (Table 7, entry 5). The stark contrast between the high trans
selectivity obtained with pivaldehyde-derived imine (t-BuCH
NTs) compared to the high cis selectivity obtained by
3. Reduced Stability of the Ylide. On the basis of the
principles described above, the selectivity with different benzyl
sulfonium salts can also be rationalized. Clearly, syn-betaine
formation will be more reversible with more stable ylides,
resulting in increased trans selectivity. Indeed, electron-deficient
benzyl substrates all gave very high diastereoselectivities, even
with aliphatic aldehydes (Table 5). Conversely, betaine
formation is less reversible with less stable ylides (electron-
rich benzyl sulfonium salts) and so lower dr was obtained
(Table 4; compare the dr observed for p-CN, p-H, p-MeO-
substituted salts; Figure 4).
Interestingly, electron-rich substrates bearing an ortho-
methoxy substituent also showed higher stereocontrol than
that of the para-methoxy isomer (Figure 4) reflecting increased
reversibility due to increased steric hindrance. Clearly, the
selectivity will be dependent upon the nature and position of
the substituents attached to the aromatic ring.
4. Solvation of Charge. The charges on the betaine are
separated during the bond rotation step (Scheme 6), and so
solvents that can solvate the charges (e.g., protic solvents) will
lower the barrier to bond-rotation making syn-betaine
formation less reversible, which in turn will lower diaster-
eoselectivity. As illustrated in Figure 5, increased amounts of
Hamersa
̌
k31 is not something we can rationalize either at the
present time.
The high trans selectivity observed for the N-BOC imine
compared to N-Ts imines may be associated with its reduced
steric properties coupled with its lower anion-stabilizing ability.
The latter will result in a later addition TS. In turn this will
increase the importance of steric factors but, maybe more
importantly, of Coulombic interactions. The addition TS
leading to the trans-aziridine has a cisoid TS where the anion
and cation are gauche to each other and so will be favored. In
contrast, the addition TS leading to the cis-aziridine has a
transoid TS where the anion and cation are anti to each other
and so will be disfavored.
Origin of Enantioselectivity in Epoxidation and
Aziridination Reactions. The model for the origin of
enantioselectivity is shown in Scheme 8. Enantioselectivity is
governed by three main factors: (i) ylide conformation, (ii)
facial selectivity of the ylide reaction, and (iii) the degree of
reversibility in betaine formation.
Analysis of space-filling models for sulfonium ylides derived
from 2 shows that complete facial selectivity can be expected as
a result of the Me group blocking reaction from one face. X-ray
structures of several of the corresponding salts have been
obtained (see Supporting Information) and one is illustrated
below (Figure 6). The salt is closely related to the ylide
intermediate and shows that one face is essentially completely
blocked, whereas the other face is open and therefore accessible
to substrates.
Figure 5. Influence of protic solvent on dr of reaction.
protic solvents lowered the dr of the reaction (see also Table
3). This ultimately led to the use of method B for reaction with
aliphatic aldehydes and to the use of neat MeCN as solvent for
the reaction of p-methoxy substituted benzyl ylide with
benzaldehyde (Figure 4 and Table 4).
Diastereoselectivity in Aziridinations. In contrast to
reactions with aldehydes, the addition of benzyl-stabilized sulfur
ylides to N-Ts imines is nonreversible, and therefore, the
selectivity is determined by the relative rates of formation of the
anti and syn-betaines.34 From computational studies, Robiette
found that the lowest energy pathway to the trans-aziridine
occurred via cisoid addition of the ylide to the imine to give the
Scheme 7. Proposed Reaction Pathways in Aziridinations with Semistabilized Sulfur Ylides
J
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 8. Origin of Enantioselectivity for Epoxidation with Sulfur Ylide 12
enantioselectivity with all of these substrates is ylide
conformation (12A:12B ratio), rather than the difference in
reactivity of the two ylide conformers.36 As stated above
(Scheme 1), the ylide can adopt conformations 12A or 12B,
but 12A should be strongly favored as 12B suffers from
nonbonded 1,4 steric interactions (Scheme 8).
The α-substituted allylic sulfonium ylides also gave very high
enantioselectivity, presumably because they show similar steric
properties to an aromatic group. In the absence of the α-
substituent, lower er was observed in epoxidation presumably
because conformer 12B was now less disfavored (reduced steric
hindrance in conformer 12B). The higher er observed in
aziridination with α-unsubstituted allylic sulfonium ylides is
intriguing and suggests that in this case the reactivity of the two
ylide conformers is markedly different in reactions with imines
compared to aldehydes (Curtin-Hammet).36
2. Hindered, Electron-Deficient, Aryl-Stabilized Ylides.
Lower enantioselectivity was generally observed with electron
deficient, aryl-stabilized ylides and particularly in their reactions
with aromatic aldehydes (Table 5). As with the neutral/
electron-rich substrates, ylide conformation should also be well
controlled in favor of conformer 12A. In these cases, formation
of the syn-betaine is reversible and nonproductive, but
formation of the anti-betaine is also likely to be partially
reversible (see section on diastereoselectivity). This has
consequences for enantioselectivity because the degree of
reversibility in anti-betaine formation is likely to be different for
the different conformers 12A and 12B (Scheme 9). Because
ylide conformer 12B is less stable (higher in energy) than
Figure 6. Two space-filling representations of crystal structure of 23d.
The enantioselectivity observed with different ylides is
therefore influenced by factors (i) and (iii) and these are
discussed, according to ylide type, in more detail below.
1. Electron Rich/Neutral Aryl-Stabilized and Alkenyl-
Stabilized Ylides. Phenyl-stabilized sulfonium ylide gave high
and uniform enantioselectivity, not only with different
aldehydes (Table 2, 94−98% ee) but also in the aziridination
of imines (96−98% ee; vide infra). Indeed, all electron-rich and
neutral, aryl-stabilized ylides gave very high enantioselectivities
with all of the aldehydes and imines studied (Tables 4 and 7).
This suggests that the dominant factor responsible for
Scheme 9. Effect of anti-Betaine Reversibility on Enantioselectivity
K
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
conformer 12A, it will react less reversibly (ylides of increasing
stability react with increasing reversibility in betaine formation)
with aldehydes resulting in an increased proportion of the
product being derived from conformer 12B, leading to low ee
(Scheme 9) (Curtin-Hammett). Conditions that reduce
reversibility in anti-betaine formation by promoting the bond
rotation step (e.g., increased protic solvent, method C), or by
inhibiting the breakdown of the betaine to its constituents
(reduced entropic driving force for converting one molecule
back to two molecules at low temperature, method D) increase
the enantioselectivity (Table 5, entries 8−10).
reaction to a narrow set of substrates, from an understanding of
the factors that influence reversibility (temperature, protic
solvent, and metal counterion), we have in fact been able to
find conditions that lead to high diastereo- and enantiose-
lectivity for a broad range of epoxides and aziridines. Figure 7
3. Alternative Diastereomeric Sulfide 22. The benzyl
sulfonium salt of 22, differing only in the stereochemistry of
the methyl substituent, was also tested in epoxidation with
benzaldehyde. This gave lower er than isothiocineole (90:10 vs
99:1), most likely because the methyl group points into the
space occupied by the aldehyde and it inhibits bond rotation
from the gauche to the trans-betaine (Scheme 10). The methyl
Scheme 10. Asymmetric Epoxidation of Benzaldehyde with
Sulfide 22
Figure 7. Epoxides available with >90:10 dr and >95:5 er using the
sulfur-ylide disconnection.
shows the different classes of epoxides that can be made in
good yield and with synthetically useful levels of stereocontrol.
This analysis shows that diaryl, heteroaryl−aryl, aryl−alkyl and
α,β-unsaturated epoxides can all be prepared with good levels
of selectivity (>90:10 dr, >95:5 er).
The methodology is now a viable alternative to alkene
epoxidation and offers a strategically different disconnection.
Table 9 shows selected comparative data on results for the
synthesis of epoxides using asymmetric alkene epoxidation
versus the method described herein. To the best of our
knowledge for aryl−alkyl-substituted trans-epoxides, Shi
dioxirane epoxidations,37 Mn(salen),38 Ru(salen)39 epoxida-
tions, and biotransformations40 are other leading alternatives.
For vinyl epoxides, alternatives are alkene epoxidation by
Mn(salen)41 or dioxirane42 catalysts. Of course the final
decision on which methodology to use will come down to
the individual requirements in a particular case.
In aziridination, betaine formation is largely nonreversible for
the reactions of semistabilized ylides. The diastereoselectivity is
therefore determined by the relative energies of the TSs
involved in their formation, which in turn is influenced largely
by the nature of the substituents on the imine and ylide.
Although lower diastereoselectivity is often observed, the levels
achieved are still synthetically useful. Figure 8 shows the
different classes of aziridines that can be made in good yield
and with >80:20 dr and >95:5 er. For the synthesis of vinyl−
alkyl substituted aziridines, this sulfur ylide methodology is a
viable alternative to the use of nitrido Mn(salen) complexes
(Table 10 shows comparative data).43 To the best of our
knowledge, the enantioselectivities reported here for the
synthesis of the types of unfunctionalized trans-disubstituted
aryl/aryl and aryl/vinyl aziridines have not been matched by
enantioselective alkene aziridinations to date.29 It should be
noted that other classes of alkenes such as α,β-unsaturated, cis
and terminal alkenes can be aziridinated with high enantiose-
lectivity.29
group behaves like a stick inserted into the spokes of a wheel,
inhibiting bond rotation, resulting in increased reversibility.
This sulfide is effectively more hindered than 2. Fortuitously,
the easier-to-access isothiocineole gave considerably higher
enantioselectivity. Once again, this highlights the importance of
understanding the factors responsible for selectivity, because
the stereochemistry of the remote methyl group would not
have been expected to influence enantioselectivity at the outset
based on a more simplistic model.
CONCLUSIONS
■
We have described a simple protocol for the large-scale, one-
step preparation of isothiocineole 2 from the simplest of
reagents, limonene, elemental sulfur and γ-terpinene. This
sulfide gives the highest selectivity (combined enantioselectivity
and diastereoselectivity) to date in asymmetric epoxidations
and aziridinations because of its rigidity, position of appropriate
substituents, and its steric properties.
Interestingly, one issue dominates the selectivity observed in
epoxidations with this sulfide and that is the degree of
reversibility in betaine formation. If betaine formation is highly
reversible, then high diastereoselectivity but low enantioselec-
tivity will result. If betaine formation is essentially non-
reversible, then low diastereoselectivity but high enantioselec-
tivity will result. To achieve both high diastereoselectivity and
high enantioselectivity, reversible formation of the syn-betaine
and nonreversible formation of the anti-betaine are required.
Although this scenario may seem on the surface to limit this
We have already applied sulfide 2 in the context of total
synthesis: the asymmetric epoxidation methodology was
L
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 9. Synthesis of trans-Epoxides by Asymmetric Alkene Epoxidation versus Aldehyde Epoxidation with Sulfide 2
a
r.r. = regioisomeric ratio.
utilized in the synthesis of quinine and quinidine,7a,44,45 and the
asymmetric aziridination methodology was utilized in the
synthesis of kainic acid.46 In these incidences, we demonstrated
that the methodology could be applied on a multigram scale
and that after the reaction the sulfide could be recovered in
good yield by distillation or chromatography for reuse. Further
applications in total synthesis are ongoing as they provide the
ultimate litmus test for the methodology. We envisage that the
ready availability of isothiocineole 2 combined with the
mechanistic picture presented here will enable widespread use
of the sulfur ylide disconnection in asymmetric epoxidations
and aziridinations.
ASSOCIATED CONTENT
* Supporting Information
■
S
Full experimental details and characterization of compounds
including NMR spectra, HPLC/GC chromatograms, and X-ray
text files are provided. This material is available free of charge
via the Internet at http://pubs.acs.org.
Figure 8. Aziridines available with >80:20 dr and >95:5 er using the
sulfur-ylide disconnection.
Table 10. Synthesis of Aryl,Alkyl-Substituted trans-Aziridines
by Asymmetric Alkene Aziridination vs Imine Aziridination
with Sulfide 2
AUTHOR INFORMATION
Corresponding Author
V.Aggarwal@bristol.ac.uk; eoghan.mcgarrigle@ucd.ie
■
Present Address
Departament de Quimica, Universitat Auton
na, 08193 Cerdanyola del Valles
§
́
̀
oma de Barcelo-
̀
, Spain.
yield
R1
R2
Ts
method
Sulfide 2
er
99:1
(%)
dr
ref.
Notes
t-Bu
Cy
68
62
66
53
20
89:11
The authors declare no competing financial interest.
SES
Ts
Ts
H
Mn(salen)N
Mn(salen)N
Mn(salen)N
Mn(salen)N
96.5:3.5
95:5
43b
43a
43a
43c
n-Pr
i-Pr
Me
ACKNOWLEDGMENTS
■
97:3
We thank EPSRC for support of this work. C.S. thanks the
ORS and the University of Bristol; C.C.C. thanks the
University of Bristol for a Centenary Scholarship; O.I. thanks
95.5:4.5
M
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(6) Sulfide 1: (a) Aggarwal, V. K.; Alonso, E.; Hynd, G.; Lydon, K.
M.; Palmer, M. J.; Porcelloni, M.; Studley, J. R. Angew. Chem., Int. Ed.
2001, 40, 1430. (b) 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. (c) Aggarwal, V. K.; Bae, I.; Lee, H.-Y.; Richardson,
J.; Williams, D. T. Angew. Chem., Int. Ed. 2003, 42, 3274−3278.
(d) Aggarwal, V. K.; Bae, I.; Hee-Yoon, L. Tetrahedron 2004, 60, 9725.
(e) Aggarwal, V. K.; Bi, J. Beilstein J. Org. Chem. 2005, 1, 4. http://
www.beilstein-journals.org/bjoc/single/articleFullText.htm?publicId=
1860-5397-1-4 (f) Kokotos, C. G.; Aggarwal, V. K. Chem. Commun.
2006, 2156. (g) Unthank, M. G.; Hussain, N.; Aggarwal, V. K. Angew.
Chem., Int. Ed. 2006, 45, 7066. (h) Kotoku, N.; Narumi, F.; Kato, T.;
Yamaguchi, M.; Kobayashi, M. Tetrahedron Lett. 2007, 48, 7147.
the Departament d’Educacio I Universitats de la Generalitat de
Catalunya for a Fellowship; E.M.M. thanks Science Foundation
Ireland and Marie-Curie COFUND for a SIRG award (Grant
Number 11/SIRG/B2154). We thank Maziar Mohiti for
carrying out the catalytic reactions described in reference 27.
REFERENCES
■
(1) McGarrigle, E. M.; Myers, E. L.; Illa, O.; Shaw, M. A.; Riches, S.
L.; Aggarwal, V. K. Chem. Rev. 2007, 107, 5841−5883.
(2) For reviews of sulfur ylide epoxidation see: (a) Li, A.-H.; Dai, L.-
X.; Aggarwal, V. K. Chem. Rev. 1997, 97, 2341. (b) Aggarwal, V. K.
Synlett 1998, 329. (c) Dai, L.-X.; Hou, X.-L.; Zhou, Y.-G. Pure Appl.
Chem. 1999, 71, 369. (d) Aggarwal, V. K.; Winn, C. L. Acc. Chem. Res.
(i) Morales-Serna, J. A.; Llaveria, J.; Díaz, Y.; Matheu, M. Y.; Castillon
́
,
̀ ̀
2004, 37, 611. (e) Blot, V.; Briere, J.-F.; Davoust, M.; Miniere, S.;
S. Org. Biomol. Chem. 2008, 6, 4502. (j) Arshad, M.; Fernandez, M. A.;
́
Reboul, V.; Metzner, P. Phosphorus, Sulfur Silicon Relat. Elem. 2005,
180, 1171. (f) Aggarwal, V. K. In Comprehensive Asymmetric Catalysis
II; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: New
York, 1999; pp 679−693. (g) Aggarwal, V. K.; Richardson, J. Science of
Synthesis; George Thieme Verlag: Stuttgart, Germany, 2004; Vol. 27,
pp 21−104. (h) Aggarwal, V. K.; Badine, D. M.; Moorthie, V. A. In
Aziridines and Epoxides in Asymmetric Synthesis; Yudin, A. K., Ed.;
Wiley-VCH: Weinheim, Germany, 2006; Chapter 1. (i) McGarrigle, E.
M.; Aggarwal, V. K. In Enantioselective Organocatalysis: Reactions and
Experimental Procedures; Dalko, P. I., Ed.; Wiley-VCH: Weinheim,
McGarrigle, E. M.; Aggarwal, V. K. Tetrahedron: Asymmetry 2010, 21,
1771.
(7) Sulfide 2: (a) Illa, O.; Arshad, M.; Ros, A.; McGarrigle, E. M.;
Aggarwal, V. K. J. Am. Chem. Soc. 2010, 132, 1828. Sulfide 3:
(b) Badine, D. M.; Hebach, C.; Aggarwal, V. K. Chem. Asian J. 2006, 1,
̀
438. Sulfide 4: (c) Zanardi, J.; Lamazure, D.; Miniere, S.; Reboul, V.;
Metzner, P. J. Org. Chem. 2002, 67, 9083. (d) Julienne, K.; Metzner, P.;
Henryon, V. J. Chem. Soc., Perkin Trans. 1 1999, 731−735. (e) Julienne,
K.; Metzner, P. J. Org. Chem. 1998, 63, 4532. Sulfide 5: (f) Aggarwal,
V. K.; Ford, J. G.; Fonquerna, S.; Adams, H.; Jones, R. V. H.;
Fieldhouse, R. J. Am. Chem. Soc. 1998, 120, 8328. (g) Aggarwal, V. K.;
Ford, J. G.; Thompson, A.; Jones, R. V. H.; Standen, M. C. H. J. Am.
̀
Germany, 2007; Chapter 10. (j) Briere, J.-F.; Metzner, P. In
Organosulfur Chemistry in Asymmetric Synthesis; Wiley-VCH: Wein-
heim, Germany, 2008; Chapter 5. (k) Aggarwal, V. K.; Crimmin, M.;
Riches, S. In Science of Synthesis; George Thieme Verlag: Stuttgart,
Germany, 2008; Vol. 37, p 321. (l) Aggarwal, V. K.; McGarrigle, E. M.;
Shaw, M. A. In Stereoselective Synthesis 2: Stereoselective Reactions of
Carbonyl and Imino Groups; Molander, G. A., Ed.; Science of Synthesis,
Georg-Thieme-Verlag KG: Stuttgart, 2011. Chapter 2.6, pp 311−347.
(m) Aggarwal, V. K.; Richardson, J.; Winn, C. L. In Science of Synthesis;
Charette, A. B., Ed.; George Thieme Verlag:Stuttgart, Germany, 2005;
Vol.22, 11−74. (n) Illa, O., McGarrigle, E. M.; Aggarwal, V. K. In
Encyclopedia of Reagents for Organic Synthesis; Paquette, L. A., Crich,
D., Fuchs, P. L., Molander, G. A., Eds.; eEROS John Wiley & Sons
Ltd.: Hoboken, New Jersey, 2012. DOI: 10.1002/
047084289X.rn01477.
Chem. Soc. 1996, 118, 7004. Sulfide 6: (h) Minier
Metzner, P.; Fochi, M.; Bonini, B. F. Tetrahedron: Asymmetry 2004, 15,
3275. Sulfide 7: (i) Solladie-Cavallo, A.; Diep-Vohuule, A. J. Org.
Chem. 1995, 60, 3494. (j) Solladie-Cavallo, A.; L. Bouerat, L.; Roje, M.
̀
e, S.; Reboul, V.;
́
́
́
Tetrahedron Lett. 2000, 7309. Sulfide 8 and 9: (k) Piccinini, A.;
Kavanagh, S. A.; Connon, S. J. Chem. Commun. 2012, 48, 7814.
Sulfide 10: (l) Aggarwal, V. K.; Hynd, G.; Picoul, W.; Vasse, J.-L. J. Am.
Chem. Soc. 2002, 124, 9964. (m) Aggarwal, V. K.; Charmant, J. P. H.;
Fuentes, D.; Harvey, J. N.; Hynd, G.; Ohara, D.; Picoul, W.; Smith, C.;
Vasse, J.-L.; Winn, C. L. J. Am. Chem. Soc. 2006, 128, 2105. (n) Li, A.-
H.; Dai, L.-X.; Hou, X.-L.; Huang, Y.-Z.; Li, F.-W. J. Org. Chem. 1996,
61, 489−493. Sulfide 11: ref 7m, n and (o) 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. For a new method for the synthesis of
chiral tetrahydrothiophenes using a non-hazardous H2S surrogate see:
Robertson, F. J.; Wu, J. J. Am. Chem. Soc. 2012, 134, 2775−2780.
(8) For examples of high diastereoselectivities in sulfur ylide
epoxidations in reactions using chiral aldehydes, chiral auxiliaries and
cyclic chiral sulfides where the sulfide is retained in the product
epoxide see: (a) Sarabia, F.; Chammaa, S.; García-Castro, M.; Martín-
(3) Reviews of alkene epoxidation: (a) Li, Z.; Yamamoto, H. Acc.
Chem. Res. 2013, 46, 506−518. (b) Wong, O. A.; Shi, Y. Chem. Rev.
2008, 108, 3958−3987. (c) Xia, Q.-H.; Ge, H.-Q.; Ye, C.-P.; Liu, Z.-
M.; Su, K.-X. Chem. Rev. 2005, 105, 1603−1662. (d) McGarrigle, E.
M.; Gilheany, D. G. Chem. Rev. 2005, 105, 1563−1602. (e) De Faveri,
G.; Ilyashenko, G.; Watkinson, M. Chem. Soc. Rev. 2011, 40, 1722−
1760. (f) Rose, E.; Andrioletti, B.; Zrig, S.; Quelquejeu-Etheve, M.
Chem. Soc. Rev. 2005, 34, 573−583. (g) Jacobsen, E. N.; Wu, M. H. In
Comprehensive Asymmetric Catalysis I-III; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds;Springer-Verlag: Berlin Heidelberg,1999; Vol.2,
649. (h) Porter, M. J.; Skidmore, J. Org. React. 2009, 74, 425−672.
(i) Johnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric Synthesis,
2nd ed.; Ojima, I., Ed.; Wiley-VCH: New York, 2000; pp231−280.
(j) Katsuki, T. In Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.;
Wiley-VCH: New York, 2000; p287. (k) Matsumoto, K.; Sawada, Y.;
Katsuki, T. Pure Appl. Chem. 2008, 80, 1071.
(4) For early reports of asymmetric epoxidations using sulfur ylides
see: (a) Furukawa, N.; Sugihara, Y.; Fujihara, H. J. Org. Chem. 1989,
54, 4222. (b) Breau, L.; Durst, T. Tetrahedron: Asymmetry 1991, 2,
367. For a very early unsuccessful attempted asymmetric epoxidation
with a chiral sulfide see: (c) Trost, B. M.; Hammem, R. F. J. Am. Chem.
Soc. 1973, 95, 962−964.
(5) For an example of asymmetric sulfur ylide epoxidations using
achiral sulfides see: (a) Sone, T.; Yamaguchi, A.; Matsunaga, S.;
Shibasaki, M. J. Am. Chem. Soc. 2008, 130, 10078−10079. A report on
using “chiral micelles” in sulfur ylide asymmetric epoxidations was later
shown to be incorrect: Kavanagh, S. A.; Connon, S. J. Tetrahedron:
Asymmetry 2008, 19, 1414−1417. Zhang, Y.; Wu, W. Tetrahedron:
Asymmetry 1997, 8, 2723−2725.
Gal
́
vez, F. Chem. Commun. 2009, 5763. (b) Sarabia, F.; Martín-Gal
́
vez,
F.; Chammaa, S.; Martín-Ortiz, L.; San
2010, 75, 5530−5532. (c) Sarabia, F.; Chammaa, S.; García-Ruiz, C. J.
Org. Chem. 2011, 76, 2132. (d) Sarabia, F.; Vivar-García, C.; García-
́
chez-Ruiz, A. J. Org. Chem.
́ ́
Castro, M.; García-Ruiz, C.; Martín-Galvez, F.; Sanchez-Ruiz, A.;
Chammaa, S. Chem.Eur. J. 2012, 18, 15190−15201. (e) Bellenie, B.
R.; Goodman, J. M. Chem. Commun. 2004, 1076. (f) Midura, W. H.
Synlett 2006, 733. (g) Aparicio, D. M.; Gnecco, D.; Suar
M. L.; Mendoza, A.; Waksman, N.; Salazar, R.; Fores-Alamo, M.;
Teran
́
ez, J. R.; Orea,
́
, J. L. Tetrahedron 2012, 68, 10252−10256.
(9) (a) Lynch, J. E.; Eliel, E. L. J. Am. Chem. Soc. 1984, 106, 2943.
(b) Eliel, E. L.; Lynch, J. E.; Kume, F.; Frye, S. V. Org. Synth. 1987, 65,
215.
(10) (a) Bonini, C.; Chiummiento, L.; Funicello, M.; Lopardo, M. T.;
Lupattelli, P.; Laurita, A.; Cornia, A. J. Org. Chem. 2008, 73, 4233.
(b) Di Blasio, N.; Lopardo, M. T.; Lupattelli, P. Eur. J. Org. Chem.
2009, 938.
(11) For examples of our own applications in epoxidations see
references 6c−e, g, j and: (a) Bi, J.; Aggarwal, V. K. Chem. Commun.
2008, 120. (b) Unthank, M. G.; Tavassoli, B.; Aggarwal, V. K., Org.
Lett. 2008, 10, 1501. For aziridinations see: (c) Aggarwal, V. K.; Vasse,
N
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
J. L. Org. Lett. 2003, 5, 3987. For cyclopropanation see: (d) Aggarwal,
V. K.; Grange, E. Chem.Eur. J. 2006, 12, 568. (e) Riches, S. L.; Saha,
of the products he proposed were not correct: (d) Nakatsuchi, A. J.
Soc. Chem. Ind. Jpn. 1930, 33, 408. (e) Nakatsuchi, A. J. Soc. Chem. Ind.
Jpn. 1932, 35, 376.
́
C.; Fontan Filgueira, N.; Grange, E.; McGarrigle, E. M.; Aggarwal, V.
K. J. Am. Chem. Soc. 2010, 132, 7626. For use in borane chemistry see:
(f) Aggarwal, V. K.; Fang, G. Y.; Schmidt, A. T. J. Am. Chem. Soc. 2005,
127, 1642−1643. (g) Fang, G. Y.; Wallner, O. A.; Di Blasio, N.;
Ginesta, X.; Harvey, J. N.; Aggarwal, V. K. J. Am. Chem. Soc. 2007, 129,
14632. (h) Fang, G. Y.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2007,
46, 359−362. (i) Howells, D.; Robiette, R.; Fang, G. Y.; Knowles, L.
S.; Woodrow, M. D.; Harvey, J. N.; Aggarwal, V. K. Org. Biomol. Chem.
2008, 6, 1185. For Baylis-Hillman reactions see: (j) Myers, E. L.; de
Vries, J. G.; Aggarwal, V. K. Angew. Chem., Int. Ed. 2007, 46, 1893.
(12) See reference 6a and Aggarwal, V. K.; Alonso, E.; Fang, G.;
Ferrara, M.; Hynd, G.; Porcelloni, M. Angew. Chem., Int. Ed. 2001, 40,
1433.
(13) (S)-Camphor sulfonyl chloride is sold in 100 g quantities by
Sigma-Aldrich, whereas the (R)-enantiomer is sold in 5 g quantities.
The (R)-enantiomer is approximately three times more expensive but
remains relatively affordable (May 2013).
(14) (a) Aggarwal, V. K.; Fang, G.; Kokotos, C. G.; Richardson, J.;
Unthank, M. G. Tetrahedron 2006, 62, 11297. (b) Aggarwal, V. K.;
Aragoncillo, C.; Winn, C. L. Synthesis 2005, 1378−1382.
(15) Metzner’s sulfide 4 was used in azirdinations by Aggarwal with
some high enantioselectivities: (a) Aggarwal, V. K.; Ferrara, M.;
O’Brien, C. J.; Thompson, A.; Jones, R. V. H.; Fieldhouse, R. J. Chem.
Soc., Perkin Trans. 1 2001, 1635−1643. It was also used in
epoxidations by Aggarwal, in halocyclisations by Snyder, and in
sulfenylations by Denmark; in these cases low er’s were obtained with
sulfide 4: (b) Aggarwal, V. K.; Angelaud, R.; Bihan, D.; Blackburn, P.;
Fieldhouse, R.; Fonquerna, S. J.; Ford, G. D.; Hynd, G.; Jones, E.;
Jones, R. V. H.; Jubault, P.; Palmer, M. J.; Ratcliffe, P. D.; Adams, H. J.
Chem. Soc., Perkin Trans. 1 2001, 2604−2622. (c) Aggarwal, V. K.;
Coogan, M. P.; Stenson, R. A.; Jones, R. V. H.; Fieldhouse, R.; Blacker,
J. Eur. J. Org. Chem. 2002, 319−326. (d) Snyder, S. A.; Treitler, D. S.;
Brucks, A. P. J. Am. Chem. Soc. 2010, 132, 14303−14314. (e) Denmark,
S. E.; Kornfilt, D. J. P.; Vogler, T. J. Am. Chem. Soc. 2011, 133, 15308−
15311.
(19) Reference 18b describes 100 gallon of distillate being
crystallized with 800 pounds of thiourea, thus >100 gal of limonene
would have been used.
(20) 1,4-Cyclohexadiene can also be used, but it is much more
expensive than γ-terpinene and generates carcinogenic benzene as a
byproduct.
(21) Weitkamp reported crystallizing isothiocineole 2 to constant
optical rotation with isopentane in reference 18a.
(22) Moore, C. G.; Porter, M. Tetrahedron 1959, 6, 10−15.
(23) Purely thermal homolysis of sulfur−sulfur bonds in elemental
sulfur, at temperatures below 140 °C, has not been conclusively
demonstrated: (a) Bateman, L.; Moore, C. G.; Porter, M. J. Chem. Soc.
1958, 2866. (b) Parker, A. J.; Kharasch, N. Chem. Rev. 1959, 59, 583−
628.
(24) The proposed regiochemistry is consistent with literature
precedent (e.g., radical addition of mercaptoacetic acid to limonene
was reported to give 2:1 adducts, with sulfur forming bonds to the less-
substituted carbons of the alkenes). See: Buess, C. M.; Yiannios, C. N.;
Fitzgerald, W. T. J. Org. Chem. 1957, 22, 197−200.
(25) Demole, E.; Enggist, P.; Ohloff, G. Helv. Chim. Acta 1982, 65,
1785−1794.
(26) (a) Crich, D.; Brebion, F.; Suk, D.-H. Top. in Curr. Chem. 2006,
263, 1−38. (b) Crich, D. Helv. Chim. Acta 2006, 89, 2167−2182.
(27) Isothiocineole performs poorly in the catalytic cycle. For
example, under the standard catalytic cycle conditions but using 1
equiv of isothiocineole and PhCHO, stilbene oxide was formed in 80%
yield and 30% ee. In general, sulfides with high steric bulk perform
poorly in the catalytic cycle compared to the stoichiometric process
because it operates at elevated temperatures (40 °C). At these higher
temperatures, anti-betaine formation becomes reversible (see dis-
cussion), leading to lower enantioselectivities.
(28) For sulfur ylide cyclopropanations see refs 7o, 11d, e, 12b,
16d−g, and: (a) Solladie-Cavallo, A.; Diep-Vohuule, A.; Isarno, T.
́
(16) Sulfide 10/sulfide 11 have been used in azirdinations: (a) Li, A.-
H.; Dai, L.-X.; Hou, X.-L. J. Chem. Soc., Perkin Trans. 1 1996, 867−869.
(b) Kim, K.-H.; Metobo, S.; Jimenez, L. S. Phosphorus, Sulfur Silicon
Relat. Elem. 2001, 176, 29−47. (c) Kim, K.-H.; Metobo, S.; Jimenez, L.
S. Tetrahedron: Asymmetry 2001, 12, 999−1005. Cyclopropanations:
ref 7o and (d) Ye, S.; Huang, Z.-Z.; Xia, C.-A.; Tang, Y.; Dai, L.-X. J.
Am. Chem. Soc. 2002, 124, 2432. (e) Huang, K.; Huang, Z.-Z. Synlett
2005, 1621. (f) Zhu, C. Y.; Cao, X. Y.; Zhu, B. H.; Deng, C.; Sun, X.-
L.; Wang, B.-Q.; Shen, Q.; Tang, Y. Chem.Eur. J. 2009, 15, 11465.
(g) Zhou, R.; Deng, X.; Zheng, J.; Shen, Q.; Sun, X.; Tang, Y. Chin. J.
Chem. 2011, 29, 995−1000. Tandem Michael Addition/Intra-
molecular ylide epoxidation of ketones (h) Wang, Q.-G.; Deng, X.-
M.; Zhu, B.-H.; Ye, L.-W.; Sun, X.-L.; Li, C.-Y.; Zhu, C.-Y.; Shen, Q.;
Tang, Y. J. Am. Chem. Soc. 2008, 130, 5408. (i) Zhu, B.-H.; Zhou, R.;
Zheng, J.-C.; Deng, X.-M.; Sun, X.-L.; Shen, Q.; Tang, Y. J. Org. Chem.
2010, 75, 3454−3457. Isoxazoline N-oxides: (j) Zhu, C.-Y.; Sun, X.-
L.; Deng, X.-M.; Zheng, J.-C.; Tang, Y. Tetrahedron 2008, 64, 5583−
5589. (k) Zhong, C.; Gautam, L. N. S.; Petersen, J. L.; Akhmedov, N.
G.; Shi, X. Chem.Eur. J. 2010, 16, 8605−8609. (l) Dihydrofurans:
Zheng, J.-C.; Zhu, C.-Y.; Sun, X.-L.; Tang, Y.; Dai, L.-X. J. Org. Chem.
2008, 73, 6909−6912. Tandem conjugate addition/cyclopropanation
(m) Chen, Z.; Zhang, J. Chem. Asian J. 2009, 4, 1527−1529. Morita-
Baylis-Hillman reaction (n) Suarez del Villar, I.; Gradillas, A.;
Angew. Chem., Int. Ed. 1998, 37, 1690. (b) Tang, Y.; Ye, S.; Sun, X. L.
Synlett 2005, 2720. (c) Kunz, R. K.; MacMillan, D. W. C. J. Am. Chem.
Soc. 2005, 127, 3240−3241. (d) Appel, R.; Hartmann, N.; Mayr, H. J.
Am. Chem. Soc. 2010, 132, 17894.
(29) For reviews see refs 2h, l and: (a) Muller, P.; Fruit, C. Chem.
̈
Rev. 2003, 103, 2905. (b) Sweeney, J. B. In Aziridines and Epoxides in
Asymmetric Synthesis; Yudin, A. K., Ed.; Wiley-VCH: Weinheim,
Germany, 2006; Chapter 4. (c) Pellissier, H. Tetrahedron 2010, 66,
1509−1555.
(30) For examples of highly enantioselective sulfur ylide aziridina-
tions see reference 12 and: (a) Saito, T.; Sakairi, M.; Akiba, D.
Tetrahedron Lett. 2001, 42, 5451. (b) Aggarwal, V. K.; Thompson, A.;
Jones, R. V. H.; Standen, M. C. H. J. Org. Chem. 1996, 61, 8368. (c) Li,
A.-H.; Zhou, Y.-G.; Dai, L.-X.; Hou, X.-L.; Xia, L.-J.; Lin, L. Angew.
Chem., Int. Ed. 1997, 36, 1317. (d) Li, A.-H.; Zhou, Y.-G.; Dai, X.-L.;
Hou, X.-L.; Xia, L.-J.; Lin, L. J. Org. Chem. 1998, 63, 4338. (e) Li, A.-
H.; Dai, L.-X.; Hou, X.-L.; Chen, M.-B. J. Org. Chem. 1996, 61, 4641.
̌
(f) Solladie-
2004, 69, 1409.
(31) (a) Dokli, I.; Matanovic, I.; Hamersa
̌
16, 11744. (b) Stipetic, I.; Roje, M.; Hamersak, Z. Synlett 2008, 3149.
́
Cavallo, A.; Roje, M.; Welter, R.; Sunjic,
̧
V. J. Org. Chem.
̌
k, Z. Chem.Eur. J. 2010,
́
(32) Aggarwal, V. K.; Harvey, J. N.; Richardson, J. J. Am. Chem. Soc.
2002, 124, 5747−5756.
́
Domínguez, G.; Perez-Castells, J. Org. Lett. 2010, 12, 2418−2421.
(33) (a) Aggarwal, V. K.; Calamai, S.; Ford, J. G. J. Chem. Soc., Perkin
Trans. 1 1997, 593−600. (b) Yoshimine, M.; Hatch, M. J. J. Am. Chem.
Soc. 1967, 89, 5831. (c) Edwards, D. R.; Montoya-Peleaz, P.; Crudden,
C. M. Org. Lett. 2007, 9, 5481−5484. (d) Edwards, D. R.; Du, J.;
Crudden, C. M. Org. Lett. 2007, 9, 2397−2400.
(34) Aggarwal, V. K; Charmant, J. P. H; Ciampi, C.; Hornby, J. M.;
O’Brien, C. J.; Hynd, G.; Parsons, R. J. Chem. Soc., Perkin Trans. 1
2001, 3159.
(17) Since our original report isothiocineole 2 has become
commercially available in both enantiomeric forms through TCI and
Aldrich.
(18) (a) Weitkamp, A. W. J. Am. Chem. Soc. 1959, 81, 3430.
(b) Weitkamp, A. W.U.S. Patent 3,026,315, Mar 20, 1962 (Chem. Abs.
1962, 57, 4706g). (c) Weitkamp, A. W. J. Am. Chem. Soc. 1959, 81,
3437−3439. Nakatsuchi previously reported the isolation of
thioethers from the reaction of sulfur and limonene but the structures
O
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(35) (a) Robiette, R. J. Org. Chem. 2006, 71, 2726−2734. See also:
(b) Janardanan, D.; Sunoj, R. B. J. Org. Chem. 2008, 73, 8163−8174.
(c) Janardanan, D.; Sunoj, R. B. Chem.Eur. J. 2007, 13, 4805.
(36) In contrast, the enantioselectivities observed in reactions of the
ylides derived from C-2 symmetric sulfides are believed to be
dependent not on ylide conformation but on ylide reactivity (Curtin-
Hammett control): Silva, M. A.; Bellenie, B. R.; Goodman, J. M. Org.
Lett. 2004, 6, 2559.
(37) Ref 3b and (a) Tu, Y.; Wang, Z.-X.; Shi, Y. J. Am. Chem. Soc.
1996, 118, 9806−9807. (b) Wang, Z.-X.; Tu, Y.; Frohn, M.; Zhang, J.-
R.; Shi, Y. J. Am. Chem. Soc. 1997, 119, 11224−11235.
(38) Nishikori, H.; Ohta, C.; Katsuki, T. Synlett 2000, 1557−1560.
(39) Koya, S.; Nishioka, Y.; Mizoguchi, H.; Uchida, T.; Katsuki, T.
Angew. Chem., Int. Ed. 2012, 51, 8243−8246.
(40) (a) Schmid, A.; Hofstetter, K.; Feiten, H.-J.; Hollmann, F.;
Witholt, B. Adv. Synth. Catal. 2001, 343, 732−737. (b) Hollmann, F.;
Lin, P.-C.; Witholt, B.; Schmid, A. J. Am. Chem. Soc. 2003, 125, 8209−
8217. (c) Sello, G.; Orsini, F.; Bernasconi, S.; Di Gennaro, P.
Tetrahedron: Asymmetry 2006, 17, 372−376. (d) Lin, H.; Liu, Y.; Wu,
Z.-L. Tetrahedron: Asymmetry 2011, 22, 134−137. (e) Toda, H.; Imae,
R.; Itoh, N. Tetrahedron: Asymmetry 2012, 23, 1542−1549. (f) Ozaki,
S.-i.; Yang, H.-J.; Matsui, T.; Goto, Y.; Watanabe, Y. Tetrahedron:
Asymmetry 1999, 10, 183−192. (g) Bernasconi, S.; Orsini, F.; Sello, G.;
Colmegna, A.; Galli, E.; Bestetti, G. Tetrahedron Lett. 2000, 41, 9157−
9161.
(41) Chang, S.; Lee, N. H.; Jacobsen, E. N. J. Org. Chem. 1993, 58,
6939−6941.
(42) (a) Frohn, M.; Dalkiewicz, M.; Tu, Y.; Wang, Z.-X.; Shi, Y. J.
Org. Chem. 1998, 63, 2948−2953. (b) Olofsson, B.; Somfai, P. J. Org.
Chem. 2003, 68, 2514−2517.
(43) (a) Minakata, S.; Ando, T.; Nishimura, M.; Ryu, I.; Komatsu, M.
Angew. Chem., Int. Ed. 1998, 37, 3392−3394. (b) Nishimura, M.;
Minakata, S.; Takahashi, T.; Oderaotoshi, Y.; Komatsu, M. J. Org.
Chem. 2002, 67, 2101−2110. (c) Ho, C.-M.; Lau, T.-C.; Kwong, H.-L.;
Wong, W.-T. J. Chem. Soc., Dalton Trans. 1999, 2411−2413.
(44) For a report of an application by others see: Brichacek, M.;
Navarro Villalobos, M.; Plichta, A.; Njardarson, J. T. Org. Lett. 2011,
13, 1110.
(45) For use of ylides derived from sulfide 2 in intramolecular
epoxidations of aldehydes and ketones see: Fritz, S. P.; Ali, Z.;
Unthank, M. G.; McGarrigle, E. M.; Aggarwal, V. K. Helv. Chim. Acta
2012, 95, 2384−2398.
(46) Lowe, M. A.; Ostovar, M.; Ferrini, S.; Chen, C. C.; Lawrence, P.
G.; Fontana, F.; Aggarwal, V. K.; Calabrese, A. A. Angew. Chem., Int.
Ed. 2011, 50, 6370−6374; Angew. Chem. 2011, 123, 6494−6498.
P
dx.doi.org/10.1021/ja405073w | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX