Phosphine-catalyzed asymmetric synthesis of β-lactones from disubstituted ketenes and aldehydes

Article abstract of DOI:10.1021/jo500486e

In this article we describe a general catalytic procedure for the formation of β-lactones bearing two stereogenic centers, from disubstituted ketenes and achiral aldehydes. BINAPHANE was found to display excellent enantioselectivity (≤90% ee for eight examples) and good diastereoselectivity (≤90:10 for 13 examples) in catalyzing the formation of β-lactones bearing two stereogenic centers from achiral aldehydes (both aromatic and aliphatic) and alkylarylketenes or dialkylketenes. A preference for formation of the trans diastereomer was observed in these reactions. For those reactions where BINAPHANE failed as a catalyst, tri-n-butylphosphine was found to be an effective achiral nucleophilic catalyst, effecting good yield and diastereoselectivity in racemic β-lactone formation. Evidence for the involvement of phosphonium enolate intermediates in the reaction mechanism was obtained through reaction monitoring by ³¹P NMR spectroscopy and by comparison with previously characterized intermediates observed in the phosphine-catalyzed ketene homodimerization reaction.

Full text of DOI:10.1021/jo500486e

Article
pubs.acs.org/joc
Phosphine-Catalyzed Asymmetric Synthesis of β‑Lactones from
Disubstituted Ketenes and Aldehydes
Shi Chen,† Mukulesh Mondal,† Ahmad A. Ibrahim,† Kraig A. Wheeler,‡ and Nessan J. Kerrigan*,†
†Department of Chemistry, Oakland University, 2200 North Squirrel Road, Rochester, Michigan 48309-4477, United States
‡Department of Chemistry, Eastern Illinois University, 600 Lincoln Avenue, Charleston, Illinois 61920-3099, United States
S
* Supporting Information
ABSTRACT: In this article we describe a general catalytic procedure for the
formation of β-lactones bearing two stereogenic centers, from disubstituted ketenes
and achiral aldehydes. BINAPHANE was found to display excellent enantiose-
lectivity (≥90% ee for eight examples) and good diastereoselectivity (≥90:10 for
13 examples) in catalyzing the formation of β-lactones bearing two stereogenic
centers from achiral aldehydes (both aromatic and aliphatic) and alkylarylketenes
or dialkylketenes. A preference for formation of the trans diastereomer was
observed in these reactions. For those reactions where BINAPHANE failed as a
catalyst, tri-n-butylphosphine was found to be an effective achiral nucleophilic
catalyst, effecting good yield and diastereoselectivity in racemic β-lactone
formation. Evidence for the involvement of phosphonium enolate intermediates
in the reaction mechanism was obtained through reaction monitoring by 31P NMR
spectroscopy and by comparison with previously characterized intermediates
observed in the phosphine-catalyzed ketene homodimerization reaction.
trations of enolates). Fu’s group was the first to show that
enantioenriched β-lactones could be generated from disub-
stituted ketenes through a catalytic asymmetric approach. In
their system a planar chiral azaferrocene catalyst catalyzed the
enantioselective reaction of dialkylketenes with aromatic
aldehydes, albeit with moderate diastereoselectivity. Unfortu-
nately, Fu’s system was not successful with alkylarylketenes or
with aliphatic aldehydes.8 Ye’s group later reported a chiral N-
heterocyclic carbene catalyzed formal cycloaddition of alkylar-
ylketenes with highly activated 2-oxoaldehydes.9 Smith’s group
have also recently disclosed their results using a chiral N-
heterocyclic carbene catalyst to catalyze the reaction of
alkylarylketenes with highly activated nitrobenzaldehydes and
pyridinecarboxaldehydes.10 However, both Ye’s and Smith’s
groups found that more electron rich aromatic aldehydes were
not tolerated as substrates.9,10
For a number of years, we have pursued a program of
research investigating phosphines as nucleophilic catalysts for
reactions of ketenes, with the goal of developing new reactions
and improving the scope of existing reactions. For the synthesis
of highly substituted β-lactones, via a formal [2 + 2]
cycloaddition of disubstituted ketenes and aldehydes, we
anticipated that a phosphine catalytic system would compare
favorably or surpass other systems from a reactivity standpoint.
This would be expected due to the superior polarizability of the
phosphorus atom in phosphines relative to that of the nitrogen
atom in amines.11 We were also confident of developing a
INTRODUCTION
■
β-Lactones are regarded as highly prized small molecules due to
their potential for use as intermediates in synthetic activities
and because of their presence as integral components of many
important drug molecules.1,2 In recent years many catalytic
asymmetric synthetic approaches to β-lactones have been
introduced, with chiral Lewis acid and chiral nucleophile-
catalyzed approaches being the most prominent and most
successful.3 Wynberg’s group was the first to demonstrate that
β-lactones could be formed in very high enantiomeric excesses
(>90% ee) through the use of an organic nucleophilic catalyst.
In what was a seminal moment in the development of
organocatalysis, they discovered that cinchona alkaloid catalysts
gave the best results in promoting the formal [2 + 2]
cycloaddition of ketene with a highly activated aldehyde,
chloral.4 Over the following 25 years, a number of groups used
the basic template of an alkaloid catalyst to develop improved
methodologies for the asymmetric synthesis of β-lactones. In
particular, the groups of Romo, Nelson, and Calter made
impressive contributions to extend the utility of the formal [2 +
2] cycloaddition of a ketene with an aldehyde to include less
activated, and more synthetically useful, aldehydes.5−7 The
addition of mild Lewis acids, in combination with an alkaloid
catalyst, was shown by some of these groups to be beneficial
with regard to expansion of substrate scope.6,7 However,
alkaloid catalysis has clear limitations in that only certain ketene
substrates can be tolerated. Ketene and simple monosubstituted
alkylketenes (usually methylketene or alkoxyketenes) work best
as substrates, most likely due to the attenuated nucleophilicity
of derived ammonium enolates (or low equilibrium concen-
Received: February 28, 2014
Published: April 29, 2014
© 2014 American Chemical Society
4920
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
highly enantioselective variant of the reaction, given the success
that other groups had demonstrated in the phosphine-catalyzed
reactions of related cumulenes, such as allenoates.12
Table 1. Catalyst Screening for the Phosphine-Catalyzed
Formal [2 + 2] Cycloaddition of Ketoketenes and Aldehydes
In 2010 we reported that the axially chiral phosphine,
BINAPHANE, could catalyze the formal [2 + 2] cycloaddition
of alkylarylketenes with mainly aromatic aldehydes.13 Recently,
we broadened the scope of this study to investigate the
BINAPHANE-catalyzed reaction of alkylarylketenes with
aliphatic aldehydes and of dialkylketenes in reaction with
aromatic aldehydes, as well as completing our examination of
the reaction of alkylarylketenes with aromatic aldehydes. In
those cases where BINAPHANE failed as a catalyst, PBu3 was
investigated as an achiral nucleophilic catalyst for the
diastereoselective synthesis of racemic β-lactones. A compre-
hensive examination of the BINAPHANE/PBu3 system’s scope
with regard to aldehyde type (aromatic and aliphatic) and
ketene type (alkylarylketene and dialkylketene) was therefore
carried out and is described in this paper. A series of
mechanistic experiments was also designed, and on the basis
of the results of those studies, a discussion of the most likely
mechanism of the reaction is presented.
b
addition
conversn, %
(yield, %)
a
c
d
entry
catalyst
PBu3
PBu3
time of 1a
dr
ee, %
1
2
3
direct
4 h
20
91
90:10
95:5
PBu3/LiI (0.3
equiv)
direct
>99
60:40
4
5
6
7
Duanphos
Josiphos
(R,R)-4
4 h
4 h
4 h
4 h
0
(19)
6
56:44
71:29
93:7
90
(R)-
(94)
64
32
31
BINAPHANE
8
9
(R)-Ph-
4 h
4 h
(63)
(42)
92:8
phosphepine 5
(R)-t-Bu-
80:20
phosphepine 6
a
Structures of catalysts:
RESULTS AND DISCUSSION
■
For reaction optimization, we chose a moderately reactive
alkylarylketene, ethylphenylketene, and a weakly activated
aromatic aldehyde, 4-ClPhCHO, as test substrates (Table
1).13,14 Not surprisingly, it was found necessary to add the
ketene solution slowly to the phosphine solution in order to
obtain optimal yields/conversions of the desired β-lactone. This
is because phosphines, especially trialkylphosphines, act as
excellent nucleophilic catalysts for the homodimerization of
disubstituted ketenes, and so it was essential to keep the
concentration of the ketene low.14 When the reaction was
carried out in the presence of Lewis acids (e.g., LiI), poor
diastereoselectivity was observed, and so systems involving
Lewis acid additives were not evaluated further.6,7 A diverse
array of chiral phosphine nucleophilic catalysts that had
previously shown success in related reactions was then
evaluatedthis included those containing central chirality
(DUANPHOS, Duphos, and Josiphos), as well as examples of
molecules displaying axial chirality (BINAPHANE and related
phosphepines), and planar chirality (Josiphos).12,14,15 (R)-
BINAPHANE was determined to be the best catalyst when the
key factors of reactivity, diastereoselectivity, and enantioselec-
tivity were considered (Table 1, entry 7).13 Surprisingly, the
related monophosphepines 5 and 6 were found to give
significantly lower levels of asymmetric induction than
BINAPHANE. Therefore, it appears that steric interactions
imposed by an ortho-substituted aryl group on the
diphosphepine phosphorus atom are necessary for reasonably
high levels of enantioselection (>60% ee). Interestingly, a
completely different type of catalyst, Josiphos, was found to
catalyze the reaction with very good enantioselectivity (90%
ee), but unfortunately with poor diastereoselectivity, and so was
not investigated any further (entry 5).
b
conversn = conversion to 3a; yield is isolated yield for 3a.
c
Diastereomeric ratio (dr) determined by GC-MS or 1H NMR analysis
of crude product. ee determined by chiral HPLC analysis.
d
compounds.13 The relative stereochemistry of β-lactones 3 was
determined to be trans through X-ray crystallographic analysis
of ( )-3d.13 We (and others)8,9 have defined the “trans”
diastereomer of the trisubstituted β-lactone 3 to represent the
isomer with the highest priority groups at each stereogenic
center on opposite sides of the β-lactone ring, while the “cis”
diastereomer represents the isomer with the highest priority
groups at each stereogenic center on the same side of the β-
lactone ring. Given that most catalytic asymmetric cyclo-
addition methods for β-lactone synthesis favor formation of the
cis diastereomer, the phosphine-catalyzed methodology fulfills
an important role in providing direct access to the less readily
accessible trans diastereomer.6−10
A range of structurally diverse alkylarylketenes were
examined, with the best results in terms of enantioselectivity
and diastereoselectivity being obtained with methylphenylke-
tene and ethyl-N-methyl-3-indolylketene (entries 2, 3, and 17).
Ethyl-N-methyl-3-indolylketene proved to be a remarkable
substrate, with >90% ee being obtained for all β-lactone
examples derived from it (Tables 1 and 2). Moreover, it was
especially noteworthy that the system was found to be tolerant
of very stable ketene substrates, such as diphenylketene (entry
Having identified BINAPHANE as an appropriate asym-
metric catalyst for further study, we then proceeded to evaluate
the substrate scope of the chiral phosphine system (Table 2).
Due to incomplete conversion in many cases and to facilitate
compound isolation and characterization, many isolated yields
for examples in Table 2 were determined following ring
opening of crude β-lactones 3 with aqueous KOH. The derived
β-hydroxycarboxylic acids 7 were obtained as analytically pure
4921
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
Table 2. Substrate Scope of Phosphine-Catalyzed Formal [2 + 2] Cycloaddition: Alkylarylketenes and Aromatic Aldehydes
a
b
entry
R1
R2
R3
yield, %
dr
ee, %
79
lactone
1
2
3
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Me
Me
Me
Me
Et
Ph
87
65
63
96:4
3b
3c
3d
3e
3f
4-ClPh
95:5
90
92
4-NO2Ph
2-NO2Ph
2-FPh
94:6
c
d
4
35
69:31
65:35
62:38
>99:1
75:25
93:7
c
d
5
56
c
d
6
2-ClPh
39
3g
3h
3i
c
d
7
3-ClPh
49
d
8
2-ClPh
>99
nd
64
d
9
Et
4-ClPh
94
3a
3j
10
11
12
13
14
15
Et
4-NO2Ph
4-ClPh
51
58
99
72
72
55
80:20
92:8
87
n-Bu
n-Bu
Me
Me
Me
Me
Et
41/>99
61/96
54
3k
3l
4-NO2Ph
4-NO2Ph
4-ClPh
53:47
96:4
2-tolyl
4-tolyl
4-tolyl
4-tolyl
indolyl
Ph
3m
3n
3o
3p
3q
3r
92:8
84
4-NO2Ph
3-ClPh
90:10
>99:1
83:17
85
c
d
16
37
e
17
18
4-ClPh
75
>99
96
d
Ph
4-NO2Ph
62
a
b
Isolated yield for β-hydroxyacid 7 derived from β-lactone 3 unless stated otherwise. Diastereomeric ratio (dr) determined by GCMS and 1H NMR
c
d
e
analysis of crude product. 10 mol % PBu3 used as catalyst. Isolated yield for β-lactone 3. indolyl = N-methyl-3-indolyl.
Table 3. Substrate Scope of Phosphine-Catalyzed Formal [2 + 2] Cycloaddition: Alkylarylketenes and Aliphatic Aldehydes
a
b
entry
R1
Ph
R2
R3
yield, %
dr
ee, %
lactone
1
2
Me
Et
(CH2)3CH3
(CH2)3CH3
CH2CH2Ph
(CH2)3CH3
42
83
61
72:28
65:35
>99:1
>99:1
nd
nd
97
93
3s
3t
Ph
c
3
c
indolyl
indolyl
Et
3u
4
Et
59
3v
a
b
1
Yield is isolated yield for both diastereomers of 3 or derived β-hydroxy acid 7. Diastereomeric ratio (dr) determined by GCMS and H NMR
c
analysis of crude product. indolyl = N-methyl-3-indolyl.
18).16 Electron-deficient aromatic aldehydes, where the
electron-withdrawing substituent was Cl or NO2, gave optimal
levels of diastereoselectivity and enantioselectivity (e.g., entries
2, 3, 14, and15).16 However, even in the absence of an electron-
withdrawing group on the aromatic ring of the aldehyde (as in
the case of benzaldehyde), good diastereoselectivity and
enantioselectivity could be obtained, providing a ketene of
appropriate reactivity (methylphenylketene) was used as
reactant partner (entry 1). With a less reactive ketene, such
as ethylphenylketene, much lower enantioselectivity was
observed when it was subjected to reaction with benzaldehyde.
Poor to moderate diastereoselectivity was observed in cases
where an ortho-substituted aromatic aldehyde was used (Table
2, entries 4−6, and 8). In addition, diastereoselectivity was
found to decrease as the ketene alkyl substituent became
progressively longer, with n-butylphenylketene giving substan-
tially worse results than ethylphenylketene, which in turn gave
worse results than methylphenylketene (e.g., entry 12 vs entry
10 vs entry 3). This decrease in diastereoselectivity may be
attributed to competing reaction mechanisms (see Schemes 1
and 5), or more likely to increased reversibility of aldolate
formation (retro-aldol reaction; see Scheme 5). In some cases
low reactivity was shown by BINAPHANE, and so only racemic
product could be obtained through the PBu3-catalyzed reaction
(e.g., entry 4). 3-Thiophenecarboxaldehyde underwent a
smooth reaction with methylphenylketene to give the desired
β-lactone, but it underwent decarboxylation during silica gel
purification to give the corresponding alkene in good yield
(82%) and with high E selectivity (89:11). This was also the
case with α,β-unsaturated aldehydes, such as cinnamaldehyde.
The scope of the phosphine-catalyzed methodology with
respect to the reaction of alkylarylketenes with aliphatic
aldehydes was then investigated (Table 3). Aliphatic aldehydes
were investigated, as they represent an important class of
aldehydes with regard to the potential use of the methodology
in natural product synthesis.6 The level of enantioselectivity and
diastereoselectivity observed was strongly dependent upon the
ketene partner used in these reactions. When ethyl-N-methyl-3-
indolylketene was used as a reactant partner (entries 3 and 4,
Table 3), excellent levels of asymmetric induction (>90% ee)
4922
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
Table 4. Substrate Scope of Phosphine-Catalyzed Formal [2 + 2] Cycloaddition: Dialkylketenes and Aromatic Aldehydes
a
b
entry
cat.
R1
Me
R2
R3
yield, %
dr
ee, %
lactone
1
2
3
4
5
6
7
8
PBu3
Me
Me
Me
Me
Me
Et
Ph
Ph
>99
47
3aa
3aa
3bb
3bb
3cc
3dd
3ee
3ff
(R)-BINAPHANE
PBu3
Me
22
3
Me
4-NO2Ph
4-NO2Ph
4-MePh
Ph
46
(R)-BINAPHANE
PBu3
Me
43
Me
49
PBu3
Et
>99
61
(R)-BINAPHANE
(R)-BINAPHANE
c-Hex
c-Hex
Me
Me
4-ClPh
4-NO2Ph
54:46
31/3
nd
64
>99:1
a
b
1
Yield is isolated yield for both diastereomers of 3. Diastereomeric ratio (dr) determined by GCMS and H NMR analysis of crude product.
Scheme 1. Previously Proposed Mechanism of Phosphine-Catalyzed Formal [2 + 2] Cycloaddition Reaction
Scheme 2. Reaction Monitoring by 31P NMR
and diastereoselection (dr >99:1) were observed. In contrast,
moderate levels of diastereoselectivity were observed in these
reactions when methylphenylketene and ethylphenylketene
were used as ketene substrates (entries 1 and 2, Table 3).
Moreover, isobutyraldehyde was investigated as an aliphatic
aldehyde substrate in reaction with methylphenylketene but the
corresponding β-lactone was obtained in low yield (ca. 35%, dr
= 96:4) and low enantiomeric excess (7%).
The reaction of dialkylketenes with aromatic aldehydes was
also investigated (Table 4). In general, lower levels of
enantioselectivity were observed with dialkylketenes than with
alkylarylketenes, with a maximum of 31% ee being obtained
(entry 7). Great variability in diastereoselectivity was also
observed, with optimal diastereoselectivity being observed with
a strongly deactivating substituent in comparison to a weakly
deactivating substituent on the aromatic ring of the aldehyde
(entry 8 vs entry 7). In some cases low reactivity was displayed
by BINAPHANE, and so only racemic product could be
obtained through the PBu3-catalyzed reaction variant (entries 5
and 6). Fu’s azaferrocene catalytic system performs consid-
erably better (ee up to 91%) with dialkylketenes and provides a
practical alternative to the phosphine catalytic system described
in this paper.8
4923
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
Scheme 3. Reactions of Phosphonium Enolate I
It is important to note that in situ ketene generation
previous studies on the PBu3-catalyzed homodimerization of
diphenylketene we characterized the product of the reaction of
PBu3 and diphenylketene and determined that it had the
structure of phosphonium enolate I (Scheme 2, eq 2).19
Phosphonium enolate I, generated through reaction of 1 equiv
of PBu3 with 1 equiv of diphenylketene, gave rise to a signal at
13.4 ppm in the 31P NMR spectrum, and from this we deduced
that the signal observed at 13.4 ppm in the PBu3-catalyzed
reaction of diphenylketene with 4-NO2PhCHO was due to
phosphonium enolate I (Scheme 2, eq 2).19 We speculate that
the signal observed at 29.9 ppm (2%) is due to the aldehyde-
derived phosphonium adduct (9; Scheme 1) or due to an
acylphosphonium intermediate derived from phosphonium
enolate I (Scheme 2), as acylphosphonium species are known
to give signals in the 28−32 ppm range of the 31P NMR
spectrum.18−21
(mediated by Hunig’s base) for the PBu -catalyzed reaction of
̈
3
methylphenylketene with 4-ClPhCHO was attempted but
failed. This was in contrast to our recent findings with an
alkaloid-catalyzed heterodimerization of ketenes, where two in
situ generated ketenes were cross-dimerized.17 We speculate
that this failure is due to stabilization, and hence lowered
reactivity, of phosphonium enolate intermediates by a Bronsted
̈
acid (Hunig’s base salt) or due to reversible protonation of
̈
phosphonium enolate intermediates by Hunig’s base-derived
̈
ammonium salt. Regardless of which process is operative, this
has implications for the use of phosphine catalysis for reactions
of unstable ketenes. If in situ generated monosubstituted
ketenes are required, then the use of an alkaloid catalytic
system, which has been demonstrated to be compatible with in
situ ketene generation, is recommended.3,17
To investigate whether phosphonium enolate I could
undergo an aldol-type reaction with an aldehyde to form β-
lactone 3r, we subjected a stoichiometric amount of
phosphonium enolate I to reaction with 4-NO2PhCHO
(Scheme 3, eq 1). Under these conditions, no β-lactone (3r)
was formed (Scheme 3, eq 1). This was in contrast to the result
of the phosphine-catalyzed reaction (Table 2, entry 18 and
Scheme 2, eq 1). Interestingly, when another 1 equiv of
diphenylketene was added to the former reaction (that is,
added to eq 1 of Scheme 3, after 1 equiv of 4-NO2PhCHO had
been previously added), β-lactone 3r was formed in a yield of
43%. From these results, we inferred that a more complex
mechanism than had initially been conceived was responsible
for the formation of β-lactones.13
MECHANISM
■
We originally speculated that the mechanism for formation of
trans-3 involved initial attack of the phosphine catalyst on
aldehyde 2 to give phosphonium alkoxide 8 (Scheme 1,
mechanism A).13,18 Phosphonium alkoxide 8 would add to
another molecule of ketoketene 1 to generate enolate 9.
Intramolecular SN2 (4-exo-tet) would provide trans-3 as the
major product.13 However, in light of our subsequent studies
on the phosphine-catalyzed homodimerization of ketoketenes,
where we showed that the reaction involved phosphonium
enolate intermediates, and recent studies described herein, we
have reevaluated the reaction mechanism of the phosphine-
catalyzed cycloaddition of ketenes with aldehydes.19
Careful 31P NMR monitoring at room temperature (and at
−78 °C) of the PBu3-catalyzed reaction of diphenylketene with
4-NO2PhCHO revealed signals at 13.4 ppm (55%), 29.9 ppm
(2%), and 36.0 ppm (31%) (Scheme 2, eq 1). During our
In an attempt to mimic the formation of intermediates in the
reaction, the reaction was also run with 4-NO2PhCHO added
subsequent to the reaction of 1 equiv of phosphonium enolate I
with 1 equiv of diphenylketene (Scheme 3, eq 3). In this
4924
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
Scheme 4. Additional Mechanistic Experiments
iteration of the reaction, a total of 2 equiv of diphenylketene
was added sequentially, 20 min apart, and finally 4-
NO2PhCHO was added, during which time the reaction was
continuously monitored by 31P NMR spectroscopy. Ultimately,
under these reaction conditions, 3r was formed in a yield of
39% (Scheme 3, eq 3). After the reaction of PBu3 with 2 equiv
of diphenylketene, a significant new signal was formed at 36
ppm (12% of all P species). Given that C-acylation of
phosphonium enolate I would give an acylphosphonium
species, which are known to appear in the range 28−32 ppm,
we favor O-acylation instead and have assigned structure II to
the intermediate at 36 ppm.19−21
Scheme 5. Proposed Mechanism for Phosphine-Catalyzed
Formal [2 + 2] Cycloaddition Reaction
Further support for this assignment was obtained from our
phosphine-catalyzed homodimerization studies,19 where an
analogous phosphonium intermediate IIb derived from ethyl-
phenylketene was isolated and characterized after trapping with
TMSCl (Scheme 4, eq 1); IIb gave a signal at 34 ppm in the
31P NMR spectrum which, given the close proximity of the
resonance signal of II, strongly suggests that II resembles it
structurally.20,21 Finally, when a catalytic amount of prepre-
pared phosphonium enolate I (derived from diphenylketene
and PBu3) was used as a catalyst for the reaction of
methylphenylketene with 4-ClPhCHO, the desired β-lactone
was formed in a yield of 79% (Scheme 4, eq 2), clearly
implicating phosphonium enolate I as a catalyst for this and
possibly for other phosphine-catalyzed [2 + 2] formal
cycloadditions.13,14,19 All of these 31P NMR monitoring
experiments were repeated with ethylphenylketene as the
ketene component (instead of diphenylketene), and similar
results were observed. This mechanism of catalysis may be
operative in other Lewis base catalytic systems, such as N-
heterocyclic carbenes.22 Indeed, Delaude and co-workers have
recently demonstrated that enolates derived from ketenes and
N-heterocyclic carbenes are important intermediates in NHC-
catalyzed β-lactam formation.22
β-lactone product, along with regeneration of the phosphonium
enolate I.
The preference of the reaction for the trans diastereomer can
be rationalized by invoking an anti-periplanar transition state
(rather than a gauche transition state) in the reaction between
phosphonium enolate II and the aldehyde (II to III, as depicted
in Scheme 5, and Scheme 6).7,13 In the anti-periplanar TS1,
where steric interactions between the two large substituents (Ar
and OR4) and dipole−dipole repulsions (C−O and CO) are
Scheme 6. Model for Diastereoselection in Phosphine-
Catalyzed Formal [2 + 2] Cycloaddition Reactions
On the basis of the results of reaction monitoring and control
experiments (Schemes 2−4) we propose the following revised
mechanism (Scheme 5). Stereoselective addition of the
phosphine catalyst to the ketene would afford phosphonium
enolate I, which acts as the true catalyst (and resting state of the
catalyst) in the catalytic cycle.23 Enolate I then undergoes O-
acylation by reaction with another molecule of the ketene to
give phosphonium enolate II in a stereoselective fashion.23
Phosphonium enolate II adds to the aldehyde to generate
aldolate III, which subsequently undergoes cyclization to give a
4925
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
nm; details given for each compound). Authentic racemic samples for
chiral HPLC analysis were generated through the PBu3-catalyzed
reaction. Compounds 3a−d,i,j,l−o,q,r,u,v were prepared and charac-
terized (as their carboxylic derivatives in many cases) as previously
described by our group.13 Tributyl[2-phenyl-1-(2-phenyl-1-trimethyl-
silanyloxy-but-1-enyloxy)-but-1-enyl]phosphonium (IIb) (Scheme 4,
eq 1) was prepared and characterized as previously described.19
Procedure A for β-Lactone Synthesis. To a stirred solution of
aldehyde (1 equiv or multiple equivalents) and phosphine catalyst
(PBu3 or BINAPHANE) (0.1 equiv) in dichloromethane (amount
specified for each example), at −78 °C under a nitrogen atmosphere,
was added a solution of disubstituted ketene (1 equiv) in
dichloromethane (amount specified for each example) over a period
of 4 h using a syringe pump. The reaction solution was stirred at −78
°C for another 4 h. The reaction solution was then warmed to room
temperature gradually over 4−12 h in the cooling bath (total reaction
time 12−20 h). The crude solution was passed through a plug column
(Iatrobeads, 2.5 × 3.0 cm, 6 g) (50 × weight of product mixture). The
plug column was eluted with an EtOAc/hexane solvent system (100−
250 mL), and the solvent was removed under vacuum to furnish the
desired β-lactone 3 in high purity as determined by 1H NMR and GC-
MS analysis.
Procedure B for β-Lactone Synthesis. To a stirred solution of
aldehyde (1 equiv) and phosphine catalyst (PBu3 or BINAPHANE)
(0.1 equiv) in THF (amount specified for each example), at −78 °C
under a nitrogen atmosphere, was added a solution of disubstituted
ketene (3 equiv) in THF (amount specified for each example) in one
portion. The reaction solution was stirred at −78 °C for another 8 h.
The reaction solution was then warmed to room temperature gradually
over 12 h in the cooling bath (total reaction time 20 h). Purification
was as for procedure A.
minimized, formation of the trans diastereomer would be
predicted (Scheme 6).7
CONCLUSION
■
In summary, we have developed a versatile phosphine-catalyzed
asymmetric reaction of ketenes (alkylaryl and dialkyl) with
achiral aldehydes (aromatic and aliphatic) that provides access
to β-lactones bearing two stereogenic centers with excellent
enantioselectivity (≥85% ee for 10 examples) and good
diastereoselectivity (≥90:10 for 13 examples) in many cases.
A purely diastereoselective variant of the reaction employing
PBu3 as the nucleophilic catalyst promoted the formation of
racemic β-lactones bearing two stereogenic centers with
moderate to excellent diastereoselectivity. An analysis of the
possible reaction pathways, supported by 31P NMR analysis of
reactions and intermediate trapping studies, suggests that the
reaction involves phosphonium enolate intermediates, with one
of these acting as the true catalyst of the reaction. Current
studies involve an exploration of the substrate scope with
respect to chiral aldehydes, while future studies will focus on
elucidating a model for enantioselection.
EXPERIMENTAL SECTION
■
General Considerations. THF was freshly distilled from
benzophenone ketyl radical under nitrogen prior to use. N,N-
Dimethylethylamine was distilled from calcium hydride under
nitrogen.24 Dichloromethane and diethyl ether were dried by passing
through activated alumina columns on a solvent purification system.
Tri-n-butylphosphine, (R)-BINAPHANE, lithium iodide, isobutyralde-
hyde, benzaldehyde, 4-methylbenzaldehyde, 2-fluorobenzaldehyde, 2-
chlorobenzaldehyde, 3-chlorobenzaldehyde, 4-chlorobenzaldehyde, 4-
nitrobenzaldehyde, pentanal, hydrocinnamaldehyde, cinnamaldehyde,
3-thiophenecarboxaldehyde, and potassium hydroxide were purchased
and distilled in some cases.24 Iatrobeads (60 μM particle size) and
TLC plates (UV254, 250 μM) were used as received. Methylphe-
nylketene, ethylphenylketene, n-butylphenylketene, cyclohexylmethyl-
ketene, methyl-4-tolylketene, methyl-2-tolylketene, diphenylketene,
and ethyl-N-methylindolylketene were prepared through amine-
mediated dehydrohalogenation. Dimethylketene and diethylketene
were prepared through zinc-mediated dehalogenation of the
appropriate α-bromoacyl bromide precursor.25−27
Procedure for β-Hydroxy Acid Synthesis (from 3b−d,j−
3o,u,3v).13 The following procedure was used to facilitate purification
and characterization of those β-lactones not obtained in high purity
(>95%) through use of procedure A or B. A stirred mixture of crude β-
lactone 3 (0.30 mmol, 1.0 equiv) and aqueous KOH (1.0 N; 0.60 mL,
0.60 mmol, 2.0 equiv) in THF (1.2 mL) was heated to 60 °C in a
sealed tube for 6−12 h. After it was cooled, the reaction mixture was
diluted with water (2 mL) and extracted with dichloromethane (30
mL) to remove undesired product. The aqueous layer was acidified
with HCl (10%) and extracted with dichloromethane (20 mL × 3),
and the combined extracts were washed with brine and dried over
Na2SO4. Removal of the solvent under vacuum followed by column
chromatographic purification using a hexane/EtOAc solvent system
furnished the desired β-hydroxy acid 7. Isolated yields were
determined for two steps from the relevant disubstituted ketene.13
( )-3-Methyl-4-(2-nitrophenyl)-3-phenyloxetan-2-one (3e). Fol-
lowing procedure A, methylphenylketene (31 mg, 0.23 mmol) in
CH2Cl2 (0.4 mL) was added to 2-nitrobenzaldehyde (35 mg, 0.23
mmol) and n-Bu3P (5 mg, 0.025 mmol) in CH2Cl2 (0.2 mL). Elution
with 10% EtOAc/hexane through a plug column of neutral silica gel
afforded 3e as a white solid (23 mg, 35% yield): dr = 69:31 (by GC-
NMR spectra were recorded on a 400 MHz spectrometer (400
MHz for 1H and 100 MHz for 13C). NMR chemical shifts were
reported relative to TMS (0 ppm) for 1H and to CDCl3 (77.23 ppm)
for 13C spectra. High-resolution mass spectra were obtained on an
Accurate Mass Q-TOF LC-MS instrument (with ESI as the ionization
method) at Oakland University, or from the College of Sciences Major
Instrumentation Cluster at Old Dominion University. Low-resolution
mass spectra were recorded on a GC/MS instrument with a mass-
selective detector and using a Restek Rtx-CL Pesticides2 GC column
(30 m, 0.25 mm i.d.). Optical rotations were measured on an
automatic polarimeter. IR spectra were recorded on an IR
spectrometer.
Analytical high-performance liquid chromatography (HPLC) was
performed using an AD column (0.46 cm × 25 cm), an OD-H column
(0.46 cm × 25 cm), or an AS-H column (0.46 cm × 25 cm) on an
HPLC instrument attached with a diode array detector (deuterium
lamp, 190−600 nm) with HPLC-grade isopropyl alcohol and hexanes
as the eluting solvents.
Compound Characterization and Determination of Diaster-
eomeric Ratios and Enantiomeric Excesses. The β-lactones 3
were purified by plug column chromatography through neutral silica to
provide pure samples for full characterization. Diastereomeric ratios
were determined for the crude β-lactones 3 by integrating the tertiary
CH resonances on the β-lactones in 1H NMR spectra or by integration
of peaks in GC-MS spectra. Enantiomeric excesses were determined by
assaying the β-lactones 3 using chiral HPLC analysis (at λ 225 or 254
1
MS); IR (CH2Cl2) 2977, 2924, 1832, 1526, 1347, 1132 cm−1; H
NMR (400 MHz, CDCl3, TMS, major diastereomer) δ 8.27 (dd, J =
8.2, 1.2 Hz, 1H), 7.97 (d, J = 7.8 Hz, 1H), 7.86 (dt, J = 7.7, 1.0 Hz,
1H), 7.64 (dt, J = 7.8, 1.1 Hz, 1H), 7.58−7.53 (m, 2H), 7.53−7.46 (m,
2H), 7.45−7.39 (m, 1H), 6.16 (s, 1H), 1.36 (s, 3H); 13C NMR (100
MHz, CDCl3, major diastereomer) δ 173.1, 146.9, 137.7, 135.0, 132.5,
130.0, 129.2, 128.6, 128.5, 126.2, 125.6, 82.1, 66.0, 15.6; (M + H)+
HRMS m/z calcd for (C16H14NO4)+ 284.0923, found 284.0924.
( )-4-(2-Fluorophenyl)-3-methyl-3-phenyloxetan-2-one (3f). Fol-
lowing procedure A, methylphenylketene (31 mg, 0.23 mmol) in
CH2Cl2 (0.3 mL) was added to 2-fluorobenzaldehyde (29 mg, 0.23
mmol) and n-Bu3P (5 mg, 0.025 mmol) in CH2Cl2 (0.2 mL). Elution
with 10% EtOAc/hexane through a plug column of neutral silica gel
afforded 3f as a colorless gel-like liquid (33 mg, 56% yield): dr = 65:35
1
(by H NMR); IR (CH2Cl2) 3064, 2976, 2930, 1830, 1234, 1093
cm−1; 1H NMR (400 MHz, CDCl3, TMS, major diastereomer) δ 7.60
(d, J = 7.5 Hz, 2H), 7.52−7.29 (m, 3H), 7.24−7.07 (m, 3H), 6.98−
6.86 (m, 1H), 5.98 (s, 1H), 1.33 (s, 3H); 13C NMR (100 MHz,
4926
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
CDCl3, major diastereomer) δ 172.6, 159.9 (d, J = 244.1 Hz), 139.5,
135.7, 130.6 (d, J = 8.1 Hz), 129.3, 128.2, 127.5 (d, J = 3.7 Hz), 126.4,
125.9 (d, J = 3.4 Hz), 115.6 (d, J = 20.4 Hz), 78.4 (d, J = 3.6 Hz), 65.3,
20.3; (M + H)+ HRMS m/z calcd for (C16H14O2F)+ 257.0978, found
257.0969.
1.54−1.15 (m, 4H), 1.00 (t, J = 7.2 Hz, 3H); 13C NMR (100 MHz,
CDCl3, major diastereomer) δ 173.4, 129.3, 128.0, 126.8, 125.7, 83.4,
61.7, 30.5, 28.0, 22.8, 19.2, 14.1; (M + H)+ HRMS m/z calcd
(C14H19O2)+ 219.1385, found 219.1384.
(3R,4R)-4-Butyl-3-ethyl-3-phenyloxetan-2-one (3t). Following
procedure A, ethylphenylketene (51 mg, 0.35 mmol) in CH2Cl2 (0.6
mL) was added to pentanal (90 mg, 1.04 mmol) and (R)-
BINAPHANE (23 mg, 0.033 mmol) in CH2Cl2 (0.3 mL). Elution
with 5% and then 10% EtOAc/hexane through a plug column of
neutral silica gel afforded 3t as a colorless gel (68 mg, 83% yield): dr =
65:35 (by 1H NMR); [α]D24 = 61.2° (c = 0.31, CH2Cl2); IR (CH2Cl2)
2960, 2933, 2873, 1810, 1262, 1117, 1098, 1067 cm−1; 1H NMR (400
MHz, CDCl3, TMS, major diastereomer) δ 7.57−7.21 (m, 5H), 4.51
(dd, J = 10.0, 2.8 Hz, 1H), 2.29−2.15 (m, 1H), 2.03−1.87 (m, 1H),
1.55−1.36 (m, 2H), 1.36−1.19 (m, 3H), 1.06−0.91 (m, 4H), 0.83 (t, J
= 7.2 Hz, 3H); 13C NMR (100 MHz, CDCl3, major diastereomer) δ
172.8, 128.9, 127.9, 127.1, 126.5, 82.9, 67.5, 32.4, 30.8, 27.7, 22.5, 14.0,
9.2; (M + H)+ HRMS m/z calcd (C15H21O2)+ 233.1542, found
233.1539.
( )-4-(2-Chlorophenyl)-3-methyl-3-phenyloxetan-2-one (3g).
Following procedure A, methylphenylketene (31 mg, 0.23 mmol) in
CH2Cl2 (0.3 mL) was added to 2-chlorobenzaldehyde (33 mg, 0.23
mmol) and n-Bu3P (5 mg, 0.025 mmol) in CH2Cl2 (0.2 mL). Elution
with 1%, 2%, and then 5% EtOAc/hexane through a plug column of
neutral silica gel afforded 3g as a white solid (25 mg, 39% yield): dr =
1
62:38 (by H NMR); IR (CH2Cl2) 2977, 2930, 1833, 1216, 1135,
1097 cm−1; 1H NMR (400 MHz, CDCl3, TMS, major isomer) δ 7.68−
7.59 (m, 1H), 7.52−7.35 (m, 3H), 7.26−7.19 (m, 1H), 7.19−7.06 (m,
4H), 5.95 (s, 1H), 1.35 (s, 3H); 13C NMR (100 MHz, CDCl3, major
isomer) δ 173.0, 138.5, 135.3, 133.7, 131.9, 129.9, 129.2, 128.5, 127.6,
126.9, 126.6, 81.8, 67.0, 23.1; (M + H)+ HRMS m/z calcd
(C16H14O2Cl)+ 273.0682, found 273.0680.
( )-4-(3-Chlorophenyl)-3-methyl-3-phenyloxetan-2-one (3h).
Following procedure A, methylphenylketene (34 mg, 0.26 mmol) in
CH2Cl2 (0.4 mL) was added to 3-chlorobenzaldehyde (33 mg, 0.23
mmol) and n-Bu3P (5 mg, 0.025 mmol) in CH2Cl2 (0.3 mL). Elution
with 5% and then 10% EtOAc/hexane through a plug column of
neutral silica gel afforded 3h as a colorless gel-like liquid (34 mg, 49%
yield): dr >99:1 (by 1H NMR); IR (CH2Cl2) 3064, 2976, 2929, 1830,
(R)-3,3-Dimethyl-4-phenyloxetan-2-one (3aa).8 Following proce-
dure B, dimethylketene (36 mg, 0.51 mmol) in THF (1.0 mL) was
added to benzaldehyde (18 mg, 0.17 mmol) and (R)-BINAPHANE
(12 mg, 0.017 mmol) in THF (0.8 mL). Elution with 10% EtOAc/
hexane through a plug column of neutral silica gel afforded 3aa as a
colorless liquid (14 mg, 47% yield): HPLC analysis 22% ee (Daicel
Chiralpak AS-H column; 1.0 mL/min; solvent system 10% isopropyl
alcohol in hexane; retention time 5.7 min (major), 6.3 min (minor));
1
1261, 1137, 1089 cm−1; H NMR (400 MHz, CDCl3, TMS) δ 7.51−
7.46 (m, 4H), 7.46−7.38 (m, 4H), 7.36−7.31 (m, 1H), 5.71 (s, 1H),
1.30 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 172.4, 139.4, 137.2,
135.4, 130.5, 129.6, 129.2, 128.4, 125.9, 125.7, 123.9, 82.1, 65.0, 20.5;
(M + H)+ HRMS m/z calcd (C16H14O2Cl)+ 273.0682, found
273.0682.
24
[α]D = 2.5° (c = 0.08, CH2Cl2); IR (CH2Cl2): 2974, 2934, 1827,
1
1183, 1101, 1076 cm−1; H NMR (400 MHz, CDCl3, TMS) δ 7.48−
7.35 (m, 4H), 7.33−7.29 (m, 1H), 5.34 (s, 1H), 1.61 (s, 3H), 0.92 (s,
3H); 13C NMR (100 MHz, CDCl3) δ 175.2, 135.6, 128.9, 128.7,
125.4, 83.1, 57.0, 22.8, 18.2; (M + H)+ HRMS m/z calcd for
(C11H13O2)+ 177.0916, found 177.0921.
(3R,4R)-3-Butyl-4-(4-chlorophenyl)-3-phenyloxetan-2-one (3k).
Following procedure A, n-butylphenylketene (62 mg, 0.36 mmol) in
CH2Cl2 (0.6 mL) was added to 4-chlorobenzaldehyde (50 mg, 0.36
mmol) and (R)-BINAPHANE (24 mg, 0.034 mmol) in CH2Cl2 (0.3
mL). Elution with 5% and then 10% EtOAc/hexane through a plug
column of neutral silica gel afforded 3k as a gel-like liquid (64 mg, 58%
yield): dr = 92:8 (by 1H NMR); HPLC analysis 41% ee (major
diastereomer), >99% ee (minor diastereomer) (Daicel Chiralpak AS-H
column; 0.5 mL/min; solvent system 2% isopropyl alcohol in hexane;
retention times 14.7 and 21.3 min (major diastereomer), 18.4 and 23.4
min (minor diastereomer)); IR (CH2Cl2) 2957, 2936, 2871, 1825,
1114, 1092 cm−1; 1H NMR (400 MHz, CDCl3, TMS, major
diastereomer) δ 7.54−7.36 (m, 9H), 5.66 (s, 1H), 1.71−1.59 (m,
1H), 1.52−1.40 (m, 1H), 1.39−1.19 (m, 2H), 1.15−0.75 (m, 2H),
0.67 (t, J = 7.3 Hz, 3H); 13C NMR (100 MHz, CDCl3, major
diastereomer) δ 171.9, 138.0, 135.0, 133.5, 129.3, 129.3, 128.1, 127.4,
126.4, 82.4, 68.4, 33.8, 26.1, 22.8, 13.8; (M + H)+ HRMS m/z calcd
(C19H20O2Cl)+ 315.1152, found 315.1144.
(R)-3,3-Dimethyl-4-(4-nitrophenyl)oxetan-2-one (3bb). Following
procedure B, dimethylketene (34 mg, 0.49 mmol) in THF (0.7 mL)
was added to 4-nitrobenzaldehyde (24 mg, 0.16 mmol) and (R)-
BINAPHANE (11 mg, 0.016 mmol) in THF (0.9 mL). Elution with
10% EtOAc/hexane through a plug column of neutral silica gel
afforded 3bb as a slightly yellow solid (15 mg, 43% yield): HPLC
analysis 3% ee (Daicel Chiralpak AD column; 1.0 mL/min; solvent
system 10% isopropyl alcohol in hexane; retention time 10.4 min
(major), 11.6 min (minor)); IR (CH2Cl2) 2972, 2933, 1829, 1521,
1347, 1182, 1094 cm−1; 1H NMR (400 MHz, CDCl3, TMS) δ 8.33 (d,
J = 8.8 Hz, 2H), 7.49 (d, J = 8.4 Hz, 2H), 5.42 (s, 1H), 1.66 (s, 3H),
0.93 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 173.8, 148.3, 142.8,
126.4, 124.3, 81.8, 58.0, 22.7, 18.4; (M + H)+ HRMS m/z calcd for
(C11H12NO4)+ 222.0766, found 222.0763.
( )-3,3-Dimethyl-4-p-tolyloxetan-2-one (3cc). Following proce-
dure B, dimethylketene (68 mg, 0.97 mmol) in THF (0.8 mL) was
added to 4-methylbenzaldehyde (40 mg, 0.33 mmol) and n-Bu3P (7
mg, 0.034 mmol) in THF (0.4 mL). Elution with 5% and then 10%
EtOAc/hexane through a plug column of neutral silica gel afforded 3cc
as a white solid (31 mg, 49% yield); IR (CH2Cl2) 2971, 2927, 2873,
1828, 1178, 1098 cm−1; 1H NMR (400 MHz, CDCl3, TMS) δ 7.24 (d,
J = 8.0 Hz, 2H), 7.17 (d, J = 8.1 Hz, 2H), 5.31 (s, 1H), 2.39 (s, 3H),
1.59 (s, 3H), 0.92 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 175.4,
138.6, 132.5, 129.6, 125.4, 83.3, 56.8, 22.7, 21.4, 18.2; (M + H)+
HRMS m/z calcd for (C12H15O2)+ 191.1072, found 191.1070.
( )-3,3-Diethyl-4-phenyloxetan-2-one (3dd).8 Following proce-
dure B, diethylketene (51 mg, 0.52 mmol) in THF (1.2 mL) was
added to benzaldehyde (18 mg, 0.17 mmol) and n-Bu3P (4 mg, 0.020
mmol) in THF (0.6 mL). Elution with 10% EtOAc/hexane through a
plug column of neutral silica gel afforded 3dd as a colorless liquid (35
mg, >99% yield): IR (CH2Cl2): 2972, 2943, 1823, 1246, 1146, 1102,
1076 cm−1; 1H NMR (400 MHz, CDCl3, TMS) δ 7.47−7.40 (m, 2H),
7.40−7.34 (m, 1H), 7.34−7.28 (m, 2H), 5.39 (s, 1H), 1.99 (q, J = 7.6
Hz, 2H), 1.50−1.38 (m, 1H), 1.33−1.20 (m, 1H), 1.14 (t, J = 7.5 Hz,
3H), 0.78 (t, J = 7.5 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 174.3,
135.5, 128.7, 128.6, 125.8, 81.0, 64.7, 24.8, 22.0, 8.8, 8.0; (M + H)+
HRMS m/z calcd for (C13H17O2)+ 205.1229, found 205.1227.
( )-4-(3-Chlorophenyl)-3-methyl-3-p-tolyloxetan-2-one (3p).
Following procedure A, methyl-4-tolylketene (31 mg, 0.23 mmol) in
CH2Cl2 (0.4 mL) was added to 3-chlorobenzaldehyde (33 mg, 0.23
mmol) and n-Bu3P (5 mg, 0.025 mmol) in CH2Cl2 (0.2 mL). Elution
with 5% and then 10% EtOAc/hexane through a plug column of
neutral silica gel afforded 3p as a colorless liquid (25 mg, 37% yield):
1
dr >99:1 (by H NMR); IR (CH2Cl2) 2976, 2927, 1830, 1212, 1137,
1
1115, 1083 cm−1; H NMR (400 MHz, CDCl3, TMS) δ 7.47−7.39
(m, 3H), 7.38−7.26 (m, 5H), 5.68 (s, 1H), 2.41 (s, 3H), 1.28 (s, 3H);
13C NMR (100 MHz, CDCl3) δ 172.6, 138.2, 137.3, 136.4, 135.3,
130.4, 130.2, 129.2, 125.9, 125.6, 123.9, 82.3, 64.7, 21.3, 20.3; (M +
H)+ HRMS m/z calcd (C17H16O2Cl)+ 287.0839, found 287.0837.
(3R,4R)-4-Butyl-3-methyl-3-phenyloxetan-2-one (3s). Following
procedure A, methylphenylketene (53 mg, 0.40 mmol) in CH2Cl2
(0.6 mL) was added to pentanal (103 mg, 1.20 mmol) and (R)-
BINAPHANE (28 mg, 0.040 mmol) in CH2Cl2 (0.4 mL). Elution with
5% EtOAc/hexane through a plug column of neutral silica gel afforded
3s as a colorless liquid (37 mg, 42% yield): dr = 72:28 (by 1H NMR);
1
IR (CH2Cl2) 2958, 2932, 2872, 1820, 1142, 1095, 1061 cm−1; H
NMR (400 MHz, CDCl3, TMS, major diastereomer) δ 7.47−7.27 (m,
5H), 4.67 (dd, J = 8.5, 5.4 Hz, 1H), 2.05−1.86 (m, 2H), 1.64 (s, 3H),
4927
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
(3R,4R)-4-(4-Chlorophenyl)-3-cyclohexyl-3-methyloxetan-2-one
(3ee). Following procedure A, cyclohexylmethylketene (51 mg, 0.37
mmol) in CH2Cl2 (0.6 mL) was added to 4-chlorobenzaldehyde (51
mg, 0.36 mmol) and (R)-BINAPHANE (24 mg, 0.034 mmol) in
CH2Cl2 (0.3 mL). Elution with 2.5% EtOAc/hexane through a plug
column of neutral silica gel afforded 3ee as a white solid (62 mg, 61%
solution, and the reaction solution was stirred at −78 °C for 3.5 h. 31
P
NMR (162 MHz, 85% H3PO4) analysis of the reaction solution: δ 46.8
(19%), 37.7 (9%), 35.4 (4%), 13.5 (53%), −31.1 (16%) ppm. Then
diphenylketene (125 mg, 0.64 mmol, l equiv) in CH2Cl2 (0.5 mL) was
added to the reaction solution. The reaction solution was warmed
slowly to room temperature overnight. 3r was obtained as a yellowish
solid (95 mg, 43%). Characterization data for isolated 3r agreed with
those previously reported by us.13
1
yield): dr = 54:46 (by H NMR); HPLC analysis 31% ee (major
diastereomer), 3% ee (minor diastereomer) (Daicel Chiralpak OD-H
column; 1.0 mL/min; solvent system: 10% isopropyl alcohol in
hexane; retention times 4.1 and 5.3 min (major diastereomer), 4.5 and
4.8 min (minor diastereomer)); IR (CH2Cl2) 2928, 2853, 1825, 1267,
Sequential Reaction of Phosphonium Enolate I with Diphenyl-
ketene and 4-Nitrobenzaldehyde (Scheme 3, eq 3). To diphenylke-
tene 1r (115 mg, 0.59 mmol, l equiv) in CH2Cl2 (0.5 mL, 1.2 M) was
added Bu3P (0.15 mL, 0.59 mmol, 1 equiv) dropwise over 5 min at
−78 °C, and the resulting phosphonium enolate solution was stirred at
−78 °C for 20 min. 31P NMR (162 MHz, 85% H3PO4) analysis of the
reaction solution: δ 35.9 (3%), 13.4 (83%), −31.0 (13%) ppm.
Diphenylketene (122 mg, 0.64 mmol, 1.1 equiv) in CH2Cl2 (0.5 mL)
was added to the phosphonium enolate solution, and the mixture was
stirred at −78 °C for 3.5 h. 31P NMR (162 MHz, 85% H3PO4) analysis
of the reaction solution: δ 13.4 (87%), 36.1(12%) ppm. After the
solution was stirred at −78 °C for 25 min, 4-nitrobenzaldehyde (90
mg, 0.60 mmol, 1 equiv) in CH2Cl2 (0.5 mL) was added to the
phosphonium enolate solution. The reaction solution was warmed
slowly to room temperature overnight. 3r was obtained as a yellowish
solid (80 mg, 39%). Characterization data for isolated 3r agreed with
those previously reported by us.13
1
1242, 1157, 1091 cm−1; H NMR (400 MHz, CDCl3, TMS, major
diastereomer) δ 7.41 (d, J = 8.5 Hz, 2H), 7.33 (d, J = 8.5 Hz, 2H), 5.21
(s, 1H), 2.03−1.94 (m, 1H), 1.82−1.69 (m, 1H), 1.65−1.55 (m, 2H),
1.53 (s, 3H), 1.49−1.41 (m, 1H), 1.34−1.24 (m, 1H), 1.22−1.09 (m,
2H), 1.09−0.95 (m, 1H), 0.95−0.79 (m, 2H); 13C NMR (100 MHz,
CDCl3, major diastereomer) δ 174.3, 135.0, 133.7, 128.8, 128.4, 83.5,
63.8, 38.2, 27.8, 26.4, 26.3, 26.2, 26.2, 16.3; (M + H)+ HRMS m/z
calcd (C16H20O2Cl)+ 279.1152, found 279.1151.
(3R,4R)-3-Cyclohexyl-3-methyl-4-(4-nitrophenyl)oxetan-2-one
(3ff). Following procedure A, cyclohexylmethylketene (34 mg, 0.25
mmol) in CH2Cl2 (0.4 mL) was added to 4-nitrobenzaldehyde (38
mg, 0.25 mmol) and (R)-BINAPHANE (16 mg, 0.023 mmol) in
CH2Cl2 (0.2 mL). Elution with 5% and then 10% EtOAc/hexane
through a plug column of neutral silica gel afforded 3ff as a white solid
(47 mg, 64% yield): dr >99:1 (by 1H NMR); [α]D24 = 31.9° (c = 0.80,
CH2Cl2); IR (CH2Cl2) 2932, 2856, 1822, 1516, 1449, 1346, 1159,
Phosphonium Enolate I Catalyzed Reaction of Methylphenylke-
tene with 4-Chlorobenzaldehyde (Scheme 4, eq 2). Following
procedure A, methylphenylketene (59 mg, 0.45 mmol) was added to
4-chlorobenzaldehyde (62 mg, 0.44 mmol) and phosphonium enolate
I (94 μL, 0.044 mmol, 0.47 M solution in CH2Cl2). 3c was obtained as
1
1105, 1081 cm−1; H NMR (400 MHz, CDCl3, TMS) δ 8.30 (d, J =
8.8 Hz, 2H), 7.47 (d, J = 8.6 Hz, 2H), 5.50 (s, 1H), 2.10−1.74 (m,
6H), 1.49−1.09 (m, 5H), 0.89 (s, 3H); 13C NMR (100 MHz, CDCl3)
δ 173.3, 148.1, 143.2, 126.6, 124.2, 78.6, 65.7, 42.9, 28.6, 28.1, 26.4,
13.7; (M + H)+ HRMS m/z calcd (C16H20NO4)+ 290.1392, found
290.1383.
Mechanistic Studies. Bu3P-Catalyzed Reaction of Diphenylke-
tene with 4-Nitrobenzaldehyde (Scheme 2, eq 1). Following
procedure A, diphenylketene (98 mg, 0.50 mmol) in CH2Cl2 (0.8
mL) was added to 4-nitrobenzaldehyde (76 mg, 0.50 mmol) and Bu3P
(10 mg, 0.050 mmol) in CH2Cl2 (0.5 mL). After the syringe pump
addition of the ketene solution, the reaction mixture was stirred at −78
°C for 1 h. 31P NMR (162 MHz, 85% H3PO4) analysis of the reaction
solution was then performed: δ 36.0 (31%), 29.9 (2%), 13.4 (55%)
ppm. The remaining solution in the reaction flask was treated as per
procedure A. 4-(4-Nitrophenyl)-3,3-diphenyloxetan-2-one (3r) was
obtained as a yellowish solid (68 mg, 73%). Characterization data for
isolated 3r agreed with those previously reported by us.13
Quantitative Generation of Phosphonium Enolate I from
Diphenylketene (Scheme 2, eq 2).19 To diphenylketene 1r (109
mg, 0.56 mmol, l equiv) in CH2Cl2 (1.2 mL, 0.5 M) was added Bu3P
(0.14 mL, 0.56 mmol, 1 equiv) dropwise over 5 min at −78 °C, and
the resulting solution was stirred at −78 °C for 20 min. 31P NMR
analysis of the solution (at room temperature and at −78 °C)
indicated the formation of phosphonium enolate I with >98% purity:
31P NMR (162 MHz, 85% H3PO4) δ 35.9 (1%), 13.4 (99%) ppm.19
Attempted Reaction of Phosphonium Enolate I with 4-Nitro-
benzaldehyde (Scheme 3, eq 1). To diphenylketene 1r (132 mg, 0.68
mmol, l equiv) in CH2Cl2 (0.9 mL, 0.7 M) was added Bu3P (0.17 mL,
0.68 mmol, 1 equiv) dropwise over 5 min at −78 °C, and the resulting
phosphonium enolate solution was stirred at −78 °C for 20 min. 4-
Nitrobenzaldehyde (103 mg, 0.68 mmol) in CH2Cl2 (0.8 mL) was
added to the phosphonium enolate solution, and the mixture was
stirred at −78 °C for 3.5 h. The reaction solution was warmed slowly
to room temperature overnight. GC-MS analysis of the reaction
solution confirmed that 3r was not formed.
1
1
a white solid (95 mg, 79%): dr = 97:3 (by H NMR); H NMR (400
MHz, CDCl3, TMS) δ 7.52−7.45 (m, 6H), 7.43−7.37 (m, 3H), 5.72
(s, 1H), 1.28 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 172.5, 139.5,
134.9, 133.6, 129.5, 129.3, 128.3, 127.1, 125.6, 82.3, 64.8, 20.4; (M +
H)+ HRMS m/z calcd (C16H14O2Cl) 273.0682, found 273.0675.
ASSOCIATED CONTENT
■
S
* Supporting Information
Figures giving 1H and 13C NMR spectra and HPLC
chromatograms for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail for N.J.K.: kerrigan@oakland.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Support has been provided by the National Science
Foundation: Grant Nos. CHE-0911483 and CHE-1213638 to
N.J.K., CHE-0722547 to K.A.W., CHE-0821487 for NMR
facilities at Oakland University, and CHE-1048719 for LC-MS
facilities at Oakland University.
REFERENCES
■
(1) (a) Pommier, A.; Pons, J.-M. Synthesis 1995, 729−744. (b) Yang,
H. W.; Romo, D. J. Org. Chem. 1997, 62, 4−5. (c) Dymock, B. W.;
Kocienski, P. J.; Pons, J.-M. Synthesis 1998, 1655−1661. (d) Reddy, L.
R.; Saravanan, P.; Corey, E. J. J. Am. Chem. Soc. 2004, 126, 6230−6231.
(2) (a) Taunton, J.; Collins, J. L.; Schreiber, S. L. J. Am. Chem. Soc.
1996, 118, 10412−10422. (b) Wan, Z.; Nelson, S. G. J. Am. Chem. Soc.
2000, 122, 10470−10471. (c) Nelson, S. G.; Cheung, W. S.; Kassick,
A. J.; Hilfiker, M. A. J. Am. Chem. Soc. 2002, 124, 13654−13655.
(d) Shen, X.; Wasmuth, A. S.; Zhao, J.; Zhu, C.; Nelson, S. G. J. Am.
Chem. Soc. 2006, 128, 7438−7439. (e) Wang, Y.; Tennyson, R.; Romo,
D. Heterocycles 2004, 64, 605−658.
Sequential Reaction of Phosphonium Enolate I with 4-Nitro-
benzaldehyde and Diphenylketene (Scheme 3, eq 2). To
diphenylketene 1r (125 mg, 0.64 mmol, l equiv) in CH2Cl2 (0.7
mL, 0.9 M) was added Bu3P (0.16 mL, 0.64 mmol, 1 equiv) dropwise
over 5 min at −78 °C, and the resulting phosphonium enolate solution
was stirred at −78 °C for 20 min. 4-Nitrobenzaldehyde (97 mg, 0.64
mmol) in CH2Cl2 (0.4 mL) was added to the phosphonium enolate
4928
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929

The Journal of Organic Chemistry
Article
(3) (a) Paull, D. H.; Weatherwax, A.; Lectka, T. Tetrahedron 2009,
65, 6771−6803. (b) Orr, R. K.; Calter, M. A. Tetrahedron 2003, 59,
3545−3565. (c) Schneider, C. Angew. Chem., Int. Ed. 2002, 41, 744−
746. (d) Yang, H. W.; Romo, D. Tetrahedron 1999, 55, 6403−6434.
(e) Nelson, S. G.; Zhu, C.; Shen, X. J. Am. Chem. Soc. 2004, 126, 14−
15. (f) Kull, T.; Peters, R. Angew. Chem., Int. Ed. 2008, 47, 5461−5464.
(4) (a) Wynberg, H.; Staring, E. G. J. J. Am. Chem. Soc. 1982, 104,
166−168. (b) Wynberg, H.; Staring, E. G. J. J. Org. Chem. 1985, 50,
1977−1979. (c) Wynberg, H. Top. Stereochem. 1986, 16, 87−130.
(5) (a) Tennyson, R.; Romo, D. J. Org. Chem. 2000, 65, 7248−7252.
(b) Cortez, G. S.; Tennyson, R. L.; Romo, D. J. Am. Chem. Soc. 2001,
123, 7945−7946.
(25) Hodous, B. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 10006−
10007.
(26) Wiskur, S. L.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 6176−
6177.
(27) Allen, A. D.; Baigrie, L. M.; Gong, L.; Tidwell, T. T. Can. J.
Chem. 1991, 69, 138−145.
(6) Zhu, C.; Shen, X.; Nelson, S. G. J. Am. Chem. Soc. 2004, 126,
5352−5353.
(7) Calter, M. A.; Tretyak, O. A.; Flaschenriem, C. Org. Lett. 2005, 7,
1809−1812.
(8) Wilson, J. E.; Fu, G. C. Angew. Chem., Int. Ed. 2004, 43, 6358−
6360.
(9) He, L.; Lv, H.; Zhang, Y.-R.; Ye, S. J. Org. Chem. 2008, 73, 8101−
8103.
(10) Douglas, J.; Taylor, J. E.; Churchill, G.; Slawin, A. M. Z.; Smith,
A. D. J. Org. Chem. 2013, 78, 3925−3938.
(11) (a) Pearson, R. G.; Songstad, J. J. Am. Chem. Soc. 1967, 89,
1827−1836. (b) Methot, J. L.; Roush, W. R. Adv. Synth. Catal. 2004,
346, 1035−1050. (c) Wei, Y.; Sastry, G. N.; Zipse, H. J. Am. Chem. Soc.
2008, 130, 3473−3477.
(12) Wurz, R. P.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 12234−
12235.
(13) Mondal, M.; Ibrahim, A. A.; Wheeler, K. A.; Kerrigan, N. J. Org.
Lett. 2010, 12, 1664−1667.
(14) (a) Ibrahim, A. A.; Harzmann, G. D.; Kerrigan, N. J. J. Org.
Chem. 2009, 74, 1777−1780. (b) Ibrahim, A. A.; Wei, P.-H.;
Harzmann, G. D.; Kerrigan, N. J. J. Org. Chem. 2010, 75, 7901−
7904. (c) Chen, S.; Salo, E. C.; Wheeler, K. A.; Kerrigan, N. J. Org.
Lett. 2012, 14, 1784−1787.
(15) (a) Vedejs, E.; Daugulis, O.; MacKay, J. A.; Rozners, E. Synlett
2001, 1499−1505. (b) Wilson, J. E.; Fu, G. C. Angew. Chem., Int. Ed.
2006, 45, 1426−1429.
(16) The absolute configuration of 3r (entry 18, Table 2) was
determined to be R, through X-ray crystallographic analysis (see ref 13
for further details). The absolute configuration of most β-lactones 3a−
3ff was assigned to be R,R by analogy.
(17) Ibrahim, A. A.; Nalla, D.; Van Raaphorst, M.; Kerrigan, N. J. J.
Am. Chem. Soc. 2012, 134, 2942−2945.
(18) Lee, S. W.; Trogler, W. C. J. Org. Chem. 1990, 55, 2644−2648.
(19) Wei, P.-H.; Ibrahim, A. A.; Mondal, M.; Nalla, D.; Harzmann, G.
D.; Tedeschi, F. A.; Wheeler, K. A.; Kerrigan, N. J. Tetrahedron Lett.
2010, 51, 6690−6694.
(20) (a) Vedejs, E.; Diver, S. T. J. Am. Chem. Soc. 1993, 115, 3358−
3359. (b) Fan, R.-H.; Hou, X.-L. Tetrahedron Lett. 2003, 44, 4411−
4413. (c) Zhu, X.-F.; Henry, C. E.; Kwon, O. J. Am. Chem. Soc. 2007,
129, 6722−6723.
(21) (a) Hudson, H. R.; Dillon, K. B.; Walker, B. J. 31P NMR Data of
Four Coordinate Phosphonium Salts and Betaines.In Handbook of
Phosphorus-31 Nuclear Magnetic Resonance Data; Tebby, J. C., Ed.;
CRC Press: Boca Raton, FL, 1991; pp 181−226. (b) Dennis, L. W.;
Bartuska, V. J.; Maciel, G. E. J. Am. Chem. Soc. 1982, 104, 230−235.
(c) Chestnut, D. B.; Quin, L. D. Tetrahedron 2005, 61, 12343−12349.
(22) (a) Hans, M.; Wouters, J.; Demonceau, A.; Delaude, L. Chem.
Eur. J. 2013, 19, 9668−9676. (b) Maji, B.; Mayr, H. Angew. Chem., Int.
Ed. 2013, 52, 11163−11167. (c) Duguet, N.; Campbell, C. D.; Slawin,
A. M. Z.; Smith, A. D. Org. Biomol. Chem. 2008, 6, 1108−1113.
(d) Zhang, Y.-R.; He, L.; Wu, X.; Shao, P.-L.; Ye, S. Org. Lett. 2008, 10,
277−280.
(23) (a) Cannizzaro, C. E.; Strassner, T.; Houk, K. N. J. Am. Chem.
Soc. 2001, 123, 2668−2669. (b) Cannizzaro, C. E.; Houk, K. N. J. Am.
Chem. Soc. 2004, 126, 10992−11008.
(24) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals, 4th ed.; Butterworth Heinemann: Oxford, U.K., 2002.
4929
dx.doi.org/10.1021/jo500486e | J. Org. Chem. 2014, 79, 4920−4929