Brucine N-oxide-catalyzed Morita-Baylis-Hillman reaction of vinyl ketones: A mechanistic implication of dual catalyst system with proline

Article abstract of DOI:10.1039/c003667f

The brucine N-oxide promoted Morita-Baylis-Hillman (MBH) reaction of vinyl ketones with aldehydes has been achieved. The corresponding asymmetric version of MBH reaction was also investigated, and the electron-deficient aryl aldehydes have emerged as suit

Full text of DOI:10.1039/c003667f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Brucine N-oxide-catalyzed Morita–Baylis–Hillman reaction of vinyl ketones:
a mechanistic implication of dual catalyst system with proline†
Kyungsoo Oh,* Jian-Yuan Li and Jinhyang Ryu
Received 4th March 2010, Accepted 14th April 2010
First published as an Advance Article on the web 13th May 2010
DOI: 10.1039/c003667f
The brucine N-oxide promoted Morita–Baylis–Hillman (MBH) reaction of vinyl ketones with
aldehydes has been achieved. The corresponding asymmetric version of MBH reaction was also
investigated, and the electron-deficient aryl aldehydes have emerged as suitable reaction partners for
vinyl ketones; where proline was employed as a co-catalyst. In this dual catalyst system, proline is
believed to form iminium intermediates with electron-deficient aryl aldehydes, while the N-oxide
activates vinyl ketones to provide enolates through conjugate addition. Upon the combination of these
two intermediates, the MBH products with high enantioselectivities are obtained by controlling of the
rate-determining step through H-bridged chair-like transition state. Intrinsically, the resulting MBH
products, alcohols, are found to interfere with the formation of both intermediates, enolates and proline
iminium intermediates, thus the observed enantioselectivity of products attenuates upon further
reaction conversion, possibly due to autocatalysis. This current study sheds lights on the synthetic
utility of iminium species, derived from electron-deficient aryl aldehydes and proline.
mechanistic implication for the potential asymmetric origin was
Introduction
not established due to the univocal formation of racemic MBH
The Morita–Baylis–Hillman (MBH) reaction, an aldol condensa-
products. Herein, we report our mechanistic investigation of the
tion of electron-deficient alkenes with aldehydes in the presence
proline-catalyzed asymmetric MBH reactions of vinyl ketones in
of nucleophiles, is one of the most synthetically useful carbon–
the presence of brucine N-oxide as a co-catalyst. One particular
carbon bond forming reactions, resulting in highly functionalized
noteworthy aspect of the current study is that the asymmetric
carbonyl compounds with a newly created chiral center.1 Although
induction of MBH products in our dual catalyst system could
the recent development of asymmetric MBH reactions under-
be dictated by a judicious choice of proline with a specific
scores the importance of chiral nucleophilic catalysts, the substrate
configuration.
scope of both reacting partners, alkenes and aldehydes, is rather
limited, perhaps due to the complex nature of the mechanism of
MBH reactions. In particular, the asymmetric MBH reaction of
vinylketones still remains agreatchallenge.2 The state-of-the-art in
Results and discussion
In our continuing effort to utilize chiral tertiary amine N-oxides in
the enantioselective MBH reaction of methyl vinyl ketone involves
the use of cyclohexane-based aminothiourea, achieving the MBH
products in the range of 90–94% ee’s with electon-deficient
aromatic aldehydes.2b Prior to this work, the enantioselective MBH
reaction of methyl vinyl ketone with electron-deficient aromatic
aldehydes was reported in the range of 63–78% ee’s from the
laboratory of Miller et al. using a dual catalytic system of peptides
and proline.2g The use of proline in the asymmetric MBH reaction
of methyl vinyl ketone was intriguing since the asymmetric MBH
reactions utilizing additives as a co-catalyst were less successful
in the past.2h The co-catalyst, proline, was first introduced by the
Shi group in 2002;3 however, the exact nature of the co-catalyst
has not been understood. One possible reaction mechanism of the
proline/NaHCO3-catalyzed MBH reaction has been recently pro-
posed by Gruttadauria et al., where proline acts as a bifunctional
catalyst via a bicyclic enaminolactone species.4 Nevertheless, the
asymmetric reactions, we have recently reported the stereoselective
oxygen transfer reaction of brucine N-oxide (BNO) 2b with
chalcone derivatives 1, albeit with modest enantioselectivities
(Scheme 1).5 From our studies it was clearly evident that the
bridgehead amine N-oxide offers an intrinsic asymmetric environ-
ment as well as an enhanced nucleophilicity of the oxygen atom
of the N-oxide. Interestingly, our attempts to apply this oxygen
transfer protocol to other a,b-unsaturated carbonyl systems were
met with stringent resistance from several classes of substrates
under our optimized reaction conditions. In particular, methyl
vinyl ketone (MVK), a well known Michael acceptor,6 did not
participate in the proposed oxygen transfer reaction. We initially
postulated that the intermediate 3b from a conjugate addition
of N-oxides might not adopt the favorable conformation for a
subsequent nucleophilic attack of an a-carbanion to the oxygen
atom in 3b. Given the possibility of the asymmetric environment
created by the conjugate addition intermediate 3, we elected the
Morita–Baylis–Hillman reaction to further examine the synthetic
utility of such species.
Department of Chemistry and Chemical Biology, Indiana University Purdue
University Indianapolis, Indianapolis, Indiana 46202, USA. E-mail: ohk@
iupui.edu
† Electronic supplementary information (ESI) available: Further experi-
mental details and NMR spectra. See DOI: 10.1039/c003667f
We first examined the MBH reaction of methyl vinyl ketone 6
with o-nitrobenzaldehyde 7a in the presence of catalytic amounts
of chiral nucleophiles (Scheme 2). While our initial investigation
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3015–3024 | 3015
©

View Article Online
Scheme 1 Asymmetric oxygen transfer reaction using brucine N-oxide 2b.
providing racemic 8a with <10% yields (entry 1–2). However, to
our surprise, the reaction conversion drastically improved beyond
36 h reaction time in the absence of solvent, thus racemic 8a was
isolated in 90% yield after a total of 110–120 h reaction time
(entry 6). This optimal reaction condition was identified by the
use of 3 equivalents of methyl vinyl ketone 6, since some of 6
was consumed in a self dimerization.9 We interpreted the current
MBH reaction with two plausible reaction mechanisms; (1) the
BNO-promoted mechanism at the beginning of the reaction, (2)
the reaction mechanism involving a synergistic action of BNO
and 8a to facilitate the product formation beyond 10% reaction
conversion, typically after 36 h reaction time. We ruled out the
possibility of autocatalysis of 8a by performing the experiment
using 20 mol% of 8a in the absence of BNO, where a negligible
reaction conversion was observed after a prolonged reaction time.
Furthermore, we added 20 mol% of 8a from the beginning of
reaction and observed the total reaction time could be reduced to
about 72 h, which clearly indicated the potential synergistic action
of amine N-oxide and 8a.
Having established the reaction conditions for the amine
N-oxide-catalyzed MBH reaction, we further explored the reactiv-
ity of other aldehydes under our optimized conditions (Table 2).
Electron-deficient aldehydes, in particular nitro group-containing
aldehydes, readily reacted to afford the MBH products in good to
excellent yields (entries 1–6). Halogen-substituted benzaldehydes
were less efficient, providing modest yields of the MBH products
(entries 7–9). Heteroaromatic aldehydes were also suitable sub-
strates for our amine N-oxide-catalyzed MBH reaction (entries
10–11); however, the isolated yields of products were somewhat
low, possibly due to the instability of the products during the
reaction (entry 11) and upon purification on silica column
chromatography (entries 10–11). The MBH reactions employing
electron-rich arylaldehydes such as 4-methylbenzaldehyde and
aliphatic aldehyde such as cyclohexanecarboxaldehyde proceeded
less efficiently (entries 12–14). Our attempts to improve the
reaction conversion or to reduce the reaction time by increas-
ing the amounts of amine N-oxide catalyst were unsuccessful.
Surprisingly, the varied amount of amine N-oxides did not affect
the reaction yields and the reaction times for the electron-rich
aldehydes, not only at the initial stage of reaction but also the
overall reaction time.
Scheme 2 The tertiary amine-catalyzed MBH reaction.
examined the tertiary amine-catalyzed MBH reactions, such as
brucine and strychnine, the results were rather disappointing.
The reaction conversions remained very low, providing 8a in at
best 10–30% yields, even with prolonged reaction times (typically
4–10 days). Furthermore, the observed enantioselectivities of 8a
were negligible, in the range of 0–5% ee’s, regardless of reaction
conditions involving various solvents and varied amounts of
reaction components (catalysts, 6 and 7a). Our experimental
results closely mirrored the previous findings by Drewes7 and Shi,2g
respectively.
Thus, we turn our attention to the tertiary amine N-oxides
of the alkaloids due to their weakened basicity (Table 1).8 Our
initial experimental results employing 15 mol% of 2a or 2b in a
variety of solvents were similar to those of the parent alkaloids,
Table 1 The amine N-oxide catalyzed MBH reactiona
Reaction
time/h
8a yields
Entry
Catalystb
(%)c
Comments
1
2
3
4
5
6
7
8
2a
2b
2b
2b
2b
2b
2b
2b
120
120
36
120
120
120
120
36
<5
<10
<10
41
60
90
Protic/aprotic solvents
Protic/aprotic solvents
No solvent
No solvent
No solvent, 6 (2 eq)
No solvent, 6 (3 eq)
No solvent, 6 (5 eq)
No solvent, 6 (3 eq)
88
<10
a Unless stated otherwise, reactions were performed with methyl vinyl
ketone 6 (0.65 mmol) and o-nitrobenzaldehyde 7a (0.65 mmol) at ambient
temperature. b 15 mol% of catalysts 2 (0.098 mmol) were used. c Isolated
yields after column chromatography.
3016 | Org. Biomol. Chem., 2010, 8, 3015–3024
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 2 Substrate scope of the brucine N-oxide-catalyzed MBH reactiona
Table 3 Asymmetric MBH reaction of methyl vinyl ketonea
Reaction Yield
Reaction Yield
time/d b(%)
Entry Aldehyde
1
time/d b(%) Entry Aldehyde
5
3
5
90
96
77
8
5
6
5
55
55
63
Entry
N-oxide (eq)
Additive (eq)
ee (%)b
2
3
9
1
2
3
4
5
6
7
8
2a (0.1)
2b (0.1)
2b (0.1)
2b (0.1)
2a (0.1)
2b (0.1)
2c c(0.1)
2d d(0.1)
2e e(0.1)
—
2b (0.1)
2b (0.1)
2b (0.1)
2b (0.1)
2b (0.2)
2b (0.3)
2b (0.4)
2b (0.5)
—
—
nr
8 (R)
8 (R)
nr
39 (R)
57 (R)
nr
16 (R)
29 (R)
nr
40 (R)
34 (R)
27 (R)
32 (R)
62 (R)
82 (R)
83 (R)
85 (R)
Imidazole (0.1)
LiClO4 (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.2)
(L)-Pro (0.3)
(L)-Pro (0.4)
(L)-Pro (0.5)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
(L)-Pro (0.1)
10
9
10
11
12
13
14
15
16
17
18
4
5
5
4
61
80
11
12
5
6
37
47
6
7
4
5
82
44
13
14
6
8
25
26
a Reaction performed with methyl vinyl ketone 6 (0.65 mmol) and
o-nitrobenzaldehyde 7a (0.65 mmol) in 1,4-dioxane (0.77 mL/0.1 mmol
of amine N-oxide) in the presence of varied amounts of 2 and additives
at ambient temperature for 18 h (0–20% conversion by 1H NMR).
b Determined by chiral HPLC (absolute configuration of 8a was deter-
mined by comparison with the HPLC retention times in ref. 13). c 2c (4-
phenyl pyridine N-oxide). d 2d (4-methylmorpholine N-oxide). e 2e (N,N-
dimethylundecylamine N-oxide).
a Reactions were performed with methyl vinyl ketone 6 (1.95 mmol)
and aldehydes 7 (0.65 mmol) using 15 mol% (0.098 mmol) 2b in the
absence of solvent at ambient temperature. b Isolated yields after column
chromatography.
turned our attention to systems employing a co-catalyst, using
2b in combination with imidazole, lithium perchlorate, and (L)-
proline (entry 3–6).2h Although imidazole and LiClO4 delivered
no obvious improvements in enantioselectivity, the presence of (L)-
proline markedly enhanced the observed enantioselectivity to 57%
in combination with BNO 2b (entry 6). Interestingly, no reaction
was observed with 4-phenyl pyridine N-oxide (entry 7), while other
tertiary amine N-oxides were significantly less selective (entries
8–9). After confirming that (L)-proline alone did not catalyze the
reaction (entry 10), further reaction optimization was conducted
with varied amounts of both catalysts; 2b and (L)-proline (entry
11–18). While increasing the amount of (L)-proline was detrimen-
tal to the observed enantioselectivity, the excess amount of 2b led
to further improvements in the observed enantioselectivities up to
85%. Considering the cost benefit of amounts of catalysts used
against a minimal difference in the observed enantioselectivities
(entry 16 vs. 18), the optimal catalyst combination was chosen as
a 3 : 1 mixture of 2b and (L)-proline. Although at this juncture we
also investigated potential effects of various solvents and molar
ratios of the reagents, no further improvement of enantioselectivity
was obtained (see the Supplementary Information†).
Although our initial postulation regarding the conjugate ad-
dition of amine N-oxides to methyl vinyl ketone (MVK) was
vindicated by the facile MBH reactions with a variety of aldehydes,
the asymmetric induction using such intermediate species was
not apparent from our BNO-catalyzed MBH reactions. The poor
asymmetric induction was first attributed to the slow and non-
selective BNO-catalyzed MBH reaction, which is believed to be
a primary reaction pathway at the early stage of reaction in the
presence or absence of solvents. Secondly, the BNO-MBH product
catalyzed reaction mechanism is also believed to be non-selective,
which might be a major reaction pathway for beyond 10–20%
reaction conversion. Fortunately, our experiments implied that
the BNO-MBH product catalyzed reaction pathway could slow
down, if not completely shut down, in the presence of solvents,
such as 1,4-dioxane. Therefore, we further examined the possibility
of asymmetric induction in the BNO-catalyzed MBH reaction
(Table 3). Similar to our previous studies, strychinine N-oxide
2a did not catalyze the reaction (<5%); however, to our delight,
employing BNO 2b as a chiral nucleophilic catalyst led to the slow
formation of MBH product 8a with low ee (entry 2). Although
further effort was made to improve enantioselectivity, due to the
complex nature of MBH reaction mechanisms, our optimization
attempts were unsuccessful regardless of changing reaction param-
eters: varied temperatures, solvents, and amounts of 2b. We next
Interestingly, during our investigation, we found that the enan-
tiomeric excess of the MBH products 8a was inversely proportional
to the overall reaction conversion (Fig. 1). Indeed, our studies
on the variation of enantiomeric excess against the reaction
conversion clearly indicated that the observed enantiomeric excess
was maximized at the beginning of the reaction and gradually
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3015–3024 | 3017
©

View Article Online
Fig. 1 Proline-catalyzed asymmetric MBH reaction: a, c, e refer to the reaction conversion versus time and b, d, f refer to the enantiomeric excess versus
time.
lowered upon further reaction conversion (Fig. 1A-b, d and f). It
should be noted that our experimental data sets involved frequent
aliquot removals from a reaction, which are filtered using a short
pad of silica gel and analyzed using 1H NMR and HPLC. In doing
so, the observed conversions were slightly higher than experiments
involving direct isolation of MBH products. Subsequently, the
observed ee values were slightly lower. We believe that this is
due to the uneven removal of solid components, N-oxides and
proline, upon taking out frequent aliquots. Nevertheless, the
trend of reaction conversion versus enantioselectivity persisted
throughout our experiments. The effect of catalyst loading showed
that the increment of dual catalyst loading positively influenced
the reaction rate; however, the optimal condition was achieved
upon the employment of 1.5 equivalents of 2b and 0.5 equivalents
3018 | Org. Biomol. Chem., 2010, 8, 3015–3024
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
of (L)-proline (Fig. 1A-a, c and e).10 Next, we investigated the
potential effect of added protic sources (Fig. 1B). The presence
of 1 equivalent of MeOH slightly lowered the reaction rate and
enantioselectivity; 61% conversion with 46% ee, compared to 74%
conversion with 54% ee in the absence of added protic source after
72 h at ambient temperature (Fig. 1B-a and b).11 Furthermore,
we performed the reaction with added MBH product (rac)-8a
and found that the reaction was slower as in the case of MeOH,
and the corrected enantiomeric excess of the reaction products
suggested that a similar decrease of enantioselectivity persisted
upon further reaction conversion (Fig. 1B-c and d). In the presence
of enantiomerically enriched (R)-8, the MBH reaction significantly
slowed down (Fig. 1B-e and f), for example, it took 72 h for 43%
conversion, while it took only 36 h in the absence of added protic
source. However, the corrected enantiomeric excess of reaction
products indicated a high level of enantioselectivity (81% ee
at 43% conversion). Interestingly, the reaction was completely
shut down in the presence of 1 equivalent of enantiomerically
enriched (S)-8. In order to probe the nature of the dual catalyst
system, we examined the catalyst combination of 2b with (D)-
proline for the cooperative catalyst activity as noted in classical
double stereodifferentiation (Fig. 1C).12 Consistent with the results
reported by Miller et al.,2g the MBH products with opposite
configuration, (S)-8, were observed with a slightly lower reaction
rate, possibly a mis-matched case (Fig. 1C-a and b). Once again,
the presence of enantiomerically enriched (S)-8 displayed a similar
reactivity and enantioselectivity as in combination of (D)-proline
(Fig. 1C-c and d). The reversal of enantioselectivity clearly
suggested that the stereochemistry of MBH products was dictated
by the proline stereochemistry.
While proline and BNO 2b are capable of undergoing conjugate
addition to MVK,13 our preliminary NMR studies (in 1,4-dioxane-
D8) did not show any evidence of such species. However, since the
MBH reaction typically shows extremely slow reaction kinetics,
the NMR time scale might not be suitable for detecting a low
concentration of the intermediates from the conjugate addition
of proline or BNO. The possibility of ionic interaction between
catalysts has been ruled out by our NMR study, where no
noticeable change in chemical shifts was observed in a mixture of
2b and proline.14 Although further research is required to establish
the identity of the catalyst, one possible mechanistic implication
has been derived from the NMR studies of proline and aldehyde
1
7a (Fig. 2). The H NMR analysis of a mixture of proline and
o-nitro benzaldehyde 7a showed an extremely low concentration
of oxazolidinone species, exo-9a, at ambient temperature, while
diastereomeric endo-9a was only observable upon raising the
Fig. 2 1H NMR studies of a mixture of proline and aldehyde 7a.
The Royal Society of Chemistry 2010 Org. Biomol. Chem., 2010, 8, 3015–3024 | 3019
This journal is
©

View Article Online
temperature.15 The observation of oxazolidinones is very interest-
ing, since this strongly indicates the presence of iminium species
in the proline-catalyzed MBH reactions. Although our synthetic
effort was made to isolate such oxazolidinone intermediates
as possible catalyst precursors, in our hands, the isolation of
oxazolidines from the reaction of proline and aromatic aldehydes
was not successful. This was not surprising, since the only known
proline-derived oxazolidinones in literature were derived from
aliphatic aldehydes such as pivaldehyde,16 trichloroacetaldehyde
and 2-methylbutanal,17 subsequently, oxazolidinones derived from
aromatic aldehydes have only been NMR-detectable species in the
course of proline-catalyzed transformations.18 The formation of
single diastereomeric 1-oxapyrrolizidine 10a from our synthetic
studies, as an isolable chemical entity, illustrates the potential
thermodynamic sink for oxazolidinones and the transient nature
of iminium species using aromatic aldehydes. 1-Oxapyrrolizidines
10 were typically obtained either from a prolonged exposure of
proline and aromatic aldehydes at ambient temperature or a brief
mixing of proline and aromatic aldehydes at elevated temperatures.
The formation of 1-oxapyrrolizidines 10 could be explained by
spontaneous decarboxylation of 9 to generate azomethine ylides
followed by cycloaddition.19 Upon a close inspection of 1H NMR
of our MBH reaction (in 1,4-dioxane-D8), considerably higher
concentrations of oxazolidinones, 9, seem to be present at ambient
temperature, similar to the mixture of proline and 7a at elevated
temperatures, while the formation of 1-oxapyrrolizidine 10a was
completely suppressed. Two isomeric oxazolidinones 9a in the
MBH reaction mixture were assigned by using chemical shifts,
since the observed coupling constants for 9a in the reaction
mixture were somewhat different from that of 9a in a mixture
of 7a and (L)-proline. This difference in coupling constants could
be attributed to the conformational variation of 9a due to the
presence of a large amount of 2b and (L)-proline, as well as the
MBH product.20
Although there is plenty of room for alternative interpretations
of our experimental findings, the formation of oxazolidinone
9a might provide valuable mechanistic insights for the proline-
catalyzed MBH reaction.21 Recently, McQuade et al. proposed
the mechanism of MBH reactions, in which the rate-determining
step (RDS) is the elimination of the a-proton by a hemiacetal
intermediate, by kinetic isotope studies (Fig. 3 (a)).22 Moreover, the
kinetic studies by Aggarwal and Lloyd-Jones concluded that the
a-proton-transfer (or RDS) could be facilitated in the presence of
protic species (Fig. 3 (b)),23 thus the MBH products are dominant
species for autocatalysis beyond 20% conversion. These two
mechanistic pathways are consistent with our experimental results,
where iminium intermediate 11 can be proposed to participate
in the formation of the N,O-hemiacetal intermediate (Fig. 3 (c))
giving rise to the MBH products with high enantioselectivity after
preferential a-H elimination (via H-bridged chair-like transition
Fig. 3 Proposed reaction mechanism of the MBH reaction.
3020 | Org. Biomol. Chem., 2010, 8, 3015–3024 This journal is
The Royal Society of Chemistry 2010
©

View Article Online
state 13) at the initial stage of reaction. The presence of three
stereogenic centers in the transition state 13 renders 8 possible di-
astereomeric species. However, considering the most stable chair-
like transition state, where proline, two aromatic substituents,
and –CH2O–NR3 groups24 occupy equatorial positions, the pro-
posed transition state should effectively discriminate all possible
diastereomers for the one shown in Fig. 3. Furthermore, the
stereochemistry of the iminium species 11 is expected to influence
the stereochemical outcome of the N,O-acetal 13, possibly through
a preferential dissociative ring opening of exo-oxazolidinone exo-
9a.15 Thus, our proline-catalyzed MBH reaction appears to control
the stereoselectivity of the proton-transfer step.
However, as the reaction progresses, the accumulation of
MBH products 8 will compete with alkoxide intermediate 12 on
the consumption of the iminium species 11 to give new N,O-
acetals 14. Our observation of slow MBH reaction rates in the
presence of added protic species coincides with this interpretation.
Aggarwal’s alcohol-catalyzed autocatalysis could also compete
with our proline-catalyzed reaction; however, the fact that our
MBH reactions showed a high level of enantioselectivity in the
presence of a small amount of protic species (MeOH or (rac)-
8) suggests that the rate of autocatalysis might be lower than
that of the proline-catalyzed reaction. Alternatively, it can be
viewed that Aggarwal’s autocatalysis phenomenon is somewhat
selective, since there is a matched case (Fig. 1A and 1B) and a
mis-matched case (in the presence of (S)-8a and (L)-Pro). The
total shutdown of the mis-matched MBH reaction is not expected
from our proposed role of MBH products in consuming the
iminium intermediate since the N,O-acetals 14 should be able
to revert back to re-generate the iminium intermediate 11. This
result, therefore, implies that alternative mechanisms for either
proline or aldehyde consumption by the MBH products might
Scheme 3 The proposed roles of brucine N-oxide.
reaction; the BNO-promoted MBH reactions proceed with very
low reaction rates and enantioselectivities in 1,4-dioxane (~10%
conversion with 10 mol% 2b after 30 h at 23 ◦C).
Next, we examined the scope of aldehyde substrates with
different vinyl ketones (Table 4). As expected, the reaction at
the 24 h mark appeared to be highly dependent on the aldehyde
substrates in terms of observed reactivity and enantioselectivity.
For example, 2-nitro-substituted aromatic aldehydes collectively
showed good enantioselectivities with reasonable reactivities (en-
try 1–5), while other less electron-deficient aldehydes showed
significantly diminished reactivities and enantioselectivities (entry
6–10). In addition, there was no reaction upon the use of
aliphatic aldehydes and acrylates, as well as N-tosyl imine28
derived from 2-nitrobenzaldehyde in place of vinyl ketones. These
results were not so surprising, since no iminium intermediates
are possible with acrylates and imines, and the oxazolidinones
derived from aliphatic aldehydes are known to be stable, in
fact isolable, thus, the concentration of iminium intermediates
could be significantly low.16,19 The substitution pattern on aryl
aldehydes also influences the enantioselectivity of MBH products,
and this could be attributed to the different activation energies of
proline-catalyzed and alcohol-catalyzed (autocatalysis) processes.
However, the general trend of the dual catalyst system regarding
the formation of MBH products with opposite configuration
in the presence of an appropriate proline persisted throughout
substrates. Furthermore, the asymmetric MBH reactions of ethyl
vinyl ketone with 2-nitro-substituted aromatic aldehydes also
confirmed the generality of our reaction, providing the MBH
products with good enantioselectivities (entry 11–15).
1
exist. To investigate this notion further, we examined H NMR
spectra of a mixture of (L)-proline, aldehyde, and (S)-8, a mis-
matched case, for possible molecular recognition between them.
Although no significant information could be obtained in this
regard, we observed a slow formation of 1-oxapyrrolizidine 10a
from the MBH reaction mixture of this mis-matched case, a
potential inhibitory pathway during the MBH reaction (see the
Supplementary Information†). We believe that Miller’s highly
successful asymmetric MBH reaction effectively controls this
unproductive (S)-8-promoted consumption of (L)-proline and
aldehydes in the presence of chiral peptide.25
The exact role of brucine N-oxide in the asymmetric induction
is not clear at this time; however, it is possible to speculate that
2b could be involved in stabilizing the iminium species 11 for a
subsequent N,O-acetal formation 13,26 possibly through either a
contact ion pair 15 or a N,O-acetal formation 16 (Scheme 3). To
substantiate the positive role of 2b, we conducted our proline-
catalyzed reactions at 30 and 35 ◦C in the absence of 2b. No MBH
product was observed at either temperature, while the formation
of 1-oxapyrrolizidine 10 was observed. The formation of 10 was
significantly suppressed in the presence of 2b at both temperatures,
resulting in the exclusive formation of the MBH product 8a with
diminished enantioselectivity. Furthermore, it is reasonable to
assume that the amine N-oxide 2b undergoes conjugate addition
to MVK to generate enolates 3b (Fig. 3) as evidenced in our
BNO-promoted MBH reactions (see Table 2–3),27 although this
might not be a major role of BNO in the proline-catalyzed MBH
Conclusion
Our studies have shown that brucine N-oxide is a nucleophilic
catalyst for the Morita–Baylis–Hillman reaction of methyl vinyl
ketone. In particular, the presence of co-catalyst, brucine N-oxide,
proline catalyzes the asymmetric Morita–Baylis–Hillman reac-
tions of electron-deficient aryl aldehydes through iminium inter-
mediates. Although the proline-catalyzed MBH reaction appears
to control the proton-transfer step with high stereoselectivies, the
observed enantioselectivity of the products varies on the nature of
aldehyde substrates. The alcohol-catalyzed (autocatalysis) process
is believed to become a competing reaction pathway as the reaction
progresses; however, various MBH products with modest to good
enantioselectivities can be achieved with electron-deficient aryl
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3015–3024 | 3021
©

View Article Online
Table 4 Asymmetric MBH reaction of vinyl ketonesa
Entry
1
8
Additive
ee (%) at 24 h
Reaction time/d
Yield b(%)
ee (%)c
(L)-Pro
(D)-Pro
74 (R)
56 (S)
4
5
42
30
63 (R)
40 (S)
2
3
4
5
(L)-Pro
(D)-Pro
78 (R)
81 (S)
3
4
45
49
44 (R)
39 (S)
(L)-Pro
(D)-Pro
75 (R)
45 (S)
8
4
49
51
49 (R)
21 (S)
(L)-Pro
(D)-Pro
35 (R)
44 (S)
4
3
72
67
55 (R)
32 (S)
(L)-Pro
(D)-Pro
80 (R)
81 (S)
7
7
20
27
45 (R)
43 (S)
6
7
(L)-Pro
(D)-Pro
60 (R)
84 (S)
6
6
16
16
59 (R)
37 (S)
(L)-Pro
(D)-Pro
44 (R)
33 (S)
6
6
21
22
50 (R)
29 (S)
8
(L)-Pro
(D)-Pro
29 (R)
36 (S)
7
6
38
43
48 (R)
16 (S)
9
(L)-Pro
(D)-Pro
20 (R)
49 (S)
5
4
34
49
26 (R)
42 (S)
10
(L)-Pro
(D)-Pro
10 (R)
19 (S)
7
7
12
16
8 (R)
11 (S)
11
12
(L)-Pro
(D)-Pro
74 (R)
74 (S)
5
5
38
30
58 (R)
54 (S)
(L)-Pro
(D)-Pro
79 (R)
82 (S)
5
5
30
39
54 (R)
61 (S)
13
(L)-Pro
(D)-Pro
67 (R)
78 (S)
5
5
54
62
72 (R)
64 (S)
3022 | Org. Biomol. Chem., 2010, 8, 3015–3024
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 4 (Contd.)
Entry
14
8
Additive
ee (%) at 24 h
Reaction time/d
Yield b(%)
ee (%)c
(L)-Pro
(D)-Pro
66 (R)
69 (S)
3
3
47
61
57 (R)
65 (S)
15
(L)-Pro
(D)-Pro
80 (R)
87 (S)
8
8
55
40
60 (R)
56 (S)
a Reactions were performed with vinyl ketone 6 (0.65 mmol) and aldehydes 7 (0.65 mmol) in 1,4-dioxane (5.0 mL) in the presence of N-oxide 2b (0.98 mmol)
and (L)/(D)-proline (0.32 mmol) at ambient temperature. b Isolated after flash chromatography. c Determined by chiral HPLC (absolute configuration of
8 was determined by comparison of the HPLC retention times with known data).
aldehydes. While the synthetic potential of iminium intermediates
derived from aryl aldehydes and proline in proline catalysis has
not been well recognized, our studies clearly showed a high
potential of such species, where a judicious choice of proline and
electron-deficient aryl aldehydes could lead to the formation of
both enantiomerically enriched MBH products. We are currently
investigating the generality of our dual catalyst system in other
asymmetric reactions, and our result will be reported in due course.
IUPUI. The Bruker 500 MHz NMR was purchased via a NSF-
MRI award (CHE-0619254).
References
1 For recent reviews, see: (a) G.-N. Ma, J.-J. Jiang, M. Shi and Y. Wei,
Chem. Commun., 2009, 5496–5514; (b) V. Carrasco-Sanchez, M. J.
Simirgiotis and L. S. Santos, Molecules, 2009, 14, 3989–4021; (c) V.
Singh and S. Batra, Tetrahedron, 2008, 64, 4511–4574; (d) Y-L. Shi and
M. Shi, Eur. J. Org. Chem., 2007, 2905–2916; (e) D. Basavaiah, K. V. Rao
and R. J. Reddy, Chem. Soc. Rev., 2007, 36, 1581–1588; (f) G. Masson,
C. Housseman and J. Zhu, Angew. Chem., Int. Ed., 2007, 46, 4614–4628;
(g) P. R. Krishna and G. V. M. Sharma, Mini-Rev. Org. Chem., 2006,
3, 137–153; (h) D. Basavaiah, A. J. Rao and T. Satyanarayana, Chem.
Rev., 2003, 103, 811–892.
2 (a) Z.-Y. Lei, X.-G. Liu, M. Shi and M. Zhao, Tetrahedron: Asymmetry,
2008, 19, 2058–2062; (b) K. Yuan, L. Zhang, H.-L. Song, Y. Hu and
X.-Y. Wu, Tetrahedron Lett., 2008, 49, 6262–6264; (c) M. Shi, Y.-H. Liu
and L.-H. Chen, Chirality, 2007, 19, 124–128; (d) H. Tang, G. Zhao,
Z. Zhou, Q. Zhou and C. Tang, Tetrahedron Lett., 2006, 47, 5717–
5721; (e) Y. Hayashi, T. Tamura and M. Shoji, Adv. Synth. Catal.,
2004, 346, 1106–1110; (f) P. R. Krishna, V. Kannan and P. V. N. Reddy,
Adv. Synth. Catal., 2004, 346, 603–606; (g) J. E. Imbriglio, M. M.
Vasbinder and S. J. Miller, Org. Lett., 2003, 5, 3741–3743; (h) M. Shi and
J.-K. Jiang, Tetrahedron: Asymmetry, 2002, 13, 1941–1947; (i) A. G. M.
Barrett, A. S. Cook and A. Kamimura, Chem. Commun., 1998, 2533–
2534; (j) I. E. Marko´, P. G. Giles and N. J. Hindley, Tetrahedron, 1997,
53, 1015–1024; (k) T. Oishi, H. Oguri and M. Hirama, Tetrahedron:
Asymmetry, 1995, 6, 1241–1244.
3 (a) M. Shi, J.-K. Jiang and C.-Q. Li, Tetrahedron Lett., 2002, 43, 127–
130. For the use of pipecolic acid in the intramolecular MBH reaction,
see: (b) C. E. Aroyan, M. M. Vasbinder and S. J. Miller, Org. Lett.,
2005, 7, 3849–3851.
4 M. Gruttadauria, F. Giacalone, P. L. Meo, A. M. Marculescu, S. Riela
and R. Noto, Eur. J. Org. Chem., 2008, 1589–1596.
5 K. Oh and J. Ryu, Tetrahedron Lett., 2008, 49, 1935–1938.
6 For reviews, see: (a) N. Krause and A. Hoffmann-Ro¨der, Synthesis,
2001, 171–196; (b) M. P. Sibi and S. Manyem, Tetrahedron, 2000, 56,
8033–8061.
7 S. E. Drewes and G. H. P. Roos, Tetrahedron, 1988, 44, 4653–4670.
8 1-Methylimidazole 3-N-oxide in MBH reaction, see: Y-S. Lin, C-W.
Liu and T. Y. R. Tsai, Tetrahedron Lett., 2005, 46, 1859–1861.
9 K.-S. Park, J. Kim, H. Choo and Y. Chong, Synlett, 2007, 395–398.
10 Proline and brucine N-oxide are not particularly soluble in 1,4-dioxane,
thus it is difficult to gauge the concentration of both catalysts.
11 In protic solvents, low enantioselectivity was observed (0–10% ee).
Experimental section
Typical experimental procedure for the enantioselective MBH
reaction of aldehydes
To a stirred solution of 2-nitroaldehyde 7a (100 mg, 0.65 mmol),
brucine N-oxide 2b (404 mg, 0.98 mmol), and (L)-proline (37 mg,
0.32 mmol) in dry 1,4-dioxane (5.0 mL) at ambient temperature,
was added ethyl vinyl ketone 6b (63 ml, 0.65 mmol) in one portion.
The resulting suspension was stirred at 25 ◦C for 5 days, after
which the mixture was directly loaded to silica gel for flash
column chromatography (eluent 33 : 67 diethyl ether–hexanes) to
give Morita–Baylis–Hillman product 8k29 (58 mg, 38% (58% ee)).
1H NMR (CDCl3, 500 MHz): d 7.95 (dd, 8.0, 1.0 Hz, 1H), 7.77
(dd, 8.0, 1.5 Hz, 1H), 7.64 (td, 7.7, 1.0 Hz, 1H), 7.44 (td, 7.7
1.5 Hz, 1H), 6.20 (s, 1H), 6.14 (s, 1H), 5.72 (d, 1.0 Hz, 1H), 3.61
(br s, 1H), 2.77–2.71 (m, 2H), 1.07 (t, 7.5 Hz, 3H); 13C NMR
(CDCl3, 125 MHz): d 202.7, 148.3, 147.9, 136.4, 133.4, 128.8,
128.4, 125.1, 124.6, 67.7, 31.1, 8.0; IR (film, cm-1) 3431, 1675, 1525,
1350; HRMS calcd for C12H13NO4Na 258.0742, found 258.0729
[MNa]+. HPLC (CHIRALPAK OD-H, hexane–2-propanol 95 : 5,
0.75 mL min-1) tR(minor) = 23.60 min, tR(major) = 26.21 min.
Acknowledgements
This work was financially supported by the School of Science
and the Department of Chemistry and Chemical Biology at
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 3015–3024 | 3023
©

View Article Online
12 S. Masamune, W. Choy, J. S. Petersen and L. R. Sita, Angew. Chem.,
Int. Ed. Engl., 1985, 24, 1–31.
13 Miller reported that proline underwent efficient conjugate addition of
the amine to MVK in CHCl3, see: M. M. Vasbinder, J. E. Imbriglio and
S. J. Miller, Tetrahedron, 2006, 62, 11450–11459.
14 For ionic interactions of proline, see: T. Mandal and C.-G. Zhao, Angew.
Chem., Int. Ed., 2008, 47, 7714–7717.
product with low enantioselectivity was observed (17% ee) with a
similar reactivity (10% conversion at 24 h at ambient temperature).
22 (a) K. E. Price, S. J. Broadwater, B. J. Walker and D. T. McQuade,
J. Org. Chem., 2005, 70, 3980–3987; (b) K. E. Price, S. J. Broadwater,
H. M. Jung and D. T. McQuade, Org. Lett., 2005, 7, 147–150. For ab
initio and DFT studies to support the McQuade mechanism, see: (c) D.
Roy and R. B. Sunoj, Org. Lett., 2007, 9, 4873–4876.
15 We assigned stereoisomers of oxazolidinone based on the stability
issue of oxazolidinones, see: D. Seebach, A. K. Beck, D. M.
Badine, M. Limbach, A. Eschenmoser, A. M. Treasurywala, R. Hobi,
W. Prikoszovich and B. Linder, Helv. Chim. Acta, 2007, 90, 425–
471.
23 (a) V. K. Aggarwal, S. Y. Fulford and G. C. Lloyd-Jones, Angew. Chem.,
Int. Ed., 2005, 44, 1706–1708. For recent computation studies, see: (b) R.
Robiette, V. K. Aggarwal and J. N. Harvey, J. Am. Chem. Soc., 2007,
129, 15513–15525.
-1
=
24 The A value for C( O)Me is estimated as 1.17 kcal mol , while
16 D. Seebach, M. Boes, R. Naef and W. B. Schweizer, J. Am. Chem. Soc.,
CH2OTs is estimated as 1.75 kcal mol-1. For a comprehensive
list of A-values, see: http://www.chem.wisc.edu/areas/organic/index-
chem.htm.
1983, 105, 5390–5398.
17 F. Orsini, F. Pelizzoni, M. Forte, R. Destro and F. Gariboldi,
Tetrahedron, 1988, 44, 519–541.
25 A more pronounced mis-matched case with (D)-proline and Miller’s
peptide catalyst has been evidenced, see: ref. 2g and 13.
26 Addition of water is known to suppress the formation of oxazolidinone
9 in proline-catalyzed aldol reactions, see: ref. 18.
27 The solubility of brucine N-oxide improved in the presence of methyl
vinyl ketone and proline, see the Supporting Information†.
28 Racemic MBH product, 8a, was obtained after 5 days in 10–20% yields
due to the hydrolysis of imines to 2-nitrobenzaldehyde.
29 A. G. M. Barrett, P. Dozzo, A. J. P. White and D. J. Williams,
Tetrahedron, 2002, 58, 7303–7313.
18 For 1H NMR studies for oxazolidinones derived from aromatic
aldehydes, see: N. Zotova, A. Franzke, A. Armstrong and D. G.
Blackmond, J. Am. Chem. Soc., 2007, 129, 15100–15101.
19 F. Orsini, F. Pelizzoni, M. Forte, R. Destro and F. Gariboldi,
Tetrahedron, 1988, 44, 519–541.
20 A. Aguirre-Valderrama and J. A. Dobado, J. Carbohydr. Chem., 2006,
25, 557–594.
21 To increase the concentration of oxazolidinone 9, we examined 10 mol
% of (S)-2-(5-tetrazolyl)pyrrolidine instead of (L)-proline, but the MBH
3024 | Org. Biomol. Chem., 2010, 8, 3015–3024
This journal is
The Royal Society of Chemistry 2010
©