Enantioselective trifunctional organocatalysts for rate-enhanced aza-Morita-Baylis-Hillman reactions at room temperature

Article abstract of DOI:10.1002/adsc.200800679

A Bronsted acid-activated trifunctional organocatalyst, based on the BINAP scaffold, was used for the first time to catalyze aza-Morita-Baylis- Hillman reactions between N-tosylimines and methyl vinyl ketone with fast reaction rates and good enantioselect

Full text of DOI:10.1002/adsc.200800679

COMMUNICATIONS
DOI: 10.1002/adsc.200800679
Enantioselective Trifunctional Organocatalysts for Rate-
Enhanced Aza-Morita–Baylis–Hillman Reactions at Room
Temperature
Jean-Marc Garnier,a Christopher Anstiss,a and Fei Liua,*
a
Department of Chemistry & Biomolecular Sciences, Macquarie University, Sydney, NSW 2109, Australia
Fax : (+61)-2-9850-8313; e-mail: fliu@alchemist.chem.mq.edu.au
Received: November 3, 2008; Published online: January 27, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200800679.
Abstract: A Brønsted acid-activated trifunctional
organocatalyst, based on the BINAP scaffold, was
used for the first time to catalyze aza-Morita-
Baylis–Hillman reactions between N-tosylimines
and methyl vinyl ketone with fast reaction rates and
good enantioselectivity at room temperature. This
trifunctional catalyst, containing a Lewis base, a
Brønsted base, and a Brønsted acid, required acid
activation to confer its enantioselectivity and rate
improvement for both electron-rich and electron-
deficient imine substrates. The role of the amino
Lewis base of 1a was investigated and found to be
the activity switch in response to an acid additive.
The counterion of the acid additive was found to in-
fluence not only the excess ratio but also the sense
of asymmetric induction.
generic MBH or azaMBH proceeds via a Michael-
aldol-proton transfer sequence. The current limita-
tions of these reactions in asymmetric organocatalysis
have manifested in their slow reaction rate of days
and capricious substrate scope, primarily due to the
complex mechanistic intricacies typical of multi-step
reactions.[4]
Key advances have recently been reported for cata-
lyzing enantioselective azaMBH reactions using a va-
riety of strategies,[5] two of which appealed to the tri-
functional approach here. One is a bifunctional ap-
proach to catalyzing the MBH between imines and
methyl vinyl ketone (MVK). A BINAP-phosphine
serves as the Lewis base to initiate the Michael step,
with a phenolic Brønsted acid for polar H-bonding in-
teractions to confer enantioselecivity at low tempera-
tures.[6] Several variations on this theme, in which the
required H-bonding interactions could be relayed by
multiple Brønsted acids[6j,k] or enabled by additives,[6l]
have led to good or high enantioselectivities for some
substrates even at room temperature. The completion
time of these azaMBH reactions is typically around
10 h for most of the imine substrates, while for some
selected substrates, this strategy has led to highly
Keywords: asymmetric catalysis; aza-Morita–
Baylis–Hillman reaction; chiral amine protonation;
cooperative effects; ion pairs
Asymmetric organocatalysis has in recent years seen enantioselective azaMBH reactions that reached com-
accelerated growth in its application in synthesis.[1] pletion within a day at 08C with as little as 1 mol%
Given the existing conceptual framework, bifunction- catalyst loading.[6k] The other strategy extended
al catalysts have evolved to promote highly enantiose- beyond the bifunctional framework and employed a
lective catalytic reactions and underpin further elabo- co-catalytic, trifunctional system.[7] This approach uses
ration in catalyst developement.[2] Herein reported is an alkylphosphine as the Lewis base to initiate the
a trifunctional approach, developed from known bi- Michael step and a separate bifunctional chiral borate
functional leads, for catalyzing aza-Morita-Baylis– salt as the medium to confer good enantioselectivity
Hillman (azaMBH) reactions with good enantioselec- via a counterion strategy[8] and H-bonding interac-
tivity and rate enhancement at room temperature.
tions.[9] These two catalytic models taken together
The azaMBH reaction, similar to its generic coun- raised the possibility of developing a trifunctional or-
terpart, is an atom-economic carbon-carbon bond ganocatalyst[10] for the azaMBH reaction with motifs
forming reaction between an imine and an enone with of Lewis nucleophilicity, a Brønsted base for ion pair-
great potential in diversity oriented and convergent ing, and a Brønsted acid for H-bonding interactions
synthesis.[3] A generally accepted mechanism of the all incorporated into one chiral platform (Figure 1 a).
Adv. Synth. Catal. 2009, 351, 331 – 338
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
331

Jean-Marc Garnier et al.
COMMUNICATIONS
Figure 1. a) A trifunctional catalytic scheme proposed from existing bifunctional systems. b) Structures of catalysts 1a–e. c)
Structures of bifunctional catalysts as controls.
All of these themes are mechanistically relevant to lysts (Figure 1 c), 1e, 1f and 1g, in which either the
the azaMBH and may lead to improved catalytic amino Brønsted base or the phenolic Brønsted acid
scope and outcomes in practical terms.
was removed, were also prepared as controls (see
To answer the question if a trifunctional system, Supporting Information). By comparing directly the
rooted in existing bifunctional leads, can be engi- catalytic activities of 1a–g, with or without a Brønsted
neered to enhance the catalytic rate as well as enan- acid activator, it would be possible to investigate if
tioselectivity in the azaMBH reaction, a new series of the addition of a Brønsted base could allow this
BINAP-based phosphine catalysts were prepared system to switch from bifunctional to trifunctional as
(Figure 1 b). Consistent with the parent bifunctional reflected by changes in reaction rates and enantiose-
examples, the Lewis nucleophile used for initiation of lectivity.
the azaMBH was a BINAP-based arylphosphine, and
The synthesis of 1a–d revolved around a facile re-
the H-bonding functionality an aromatic phenol. In ductive amination reaction to rapidly assemble the re-
addition, an amino group was installed as the Brønst- quired functional groups onto the BINAP chiral scaf-
ed base for protonation with a Brønsted acid in order fold from a known phosphine amine, 1e (Scheme 1).
to provide the chiral secondary ammonium salt.[11] An The preparation of 1e, following a well-established
ancillary advantage, in addition to a chiral ion pair, is route from commercially available starting materi-
electrostatic interaction or ion H-bonding interactions als,[13] was accomplished on a gram scale. The reduc-
that may help with stabilizing zwitterionic intermedi- tive amination reaction between 1e and an aldehyde
ates of the MBH reaction.[12] Three bifunctional cata- ensued to furnish the final catalysts in one-pot in very
332
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 331 – 338

Enantioselective Trifunctional Organocatalysts
COMMUNICATIONS
Scheme 1. Synthesis of catalysts 1a–d.
good yield. This sequence was amenable to all exam- Table 1. Aza-MBH reactions of N-(arylmethylidene)arylsul-
fonamide 2a (1.0 equiv.) with MVK (2.0 equiv.) in the pres-
ence of catalysts PPh3 and 1a–1e (10 mol%) with and with-
out benzoic acid.
ples except 1d, in which case the phosphine oxide of
1e was subjected to the reductive amination first, fol-
lowed by chemospecific reduction of the phosphine
oxide with phenylsilane. The stability of these cata-
lysts was sufficient to resist oxidation during purifica-
tion except for 1c, which was partially oxidized during
column chromatography.
Phosphines 1a–d were first tested using a known
azaMBH reaction between an imine and MVK at
room temperature (25–278C) in dichloromethane
(DCM) with 10 mol% catalyst loading in the presence
or absence of benzoic acid (Table 1). For comparison,
triphenylphosphine, 1e’, and 1e were also tested. Tri-
phenylphosphine resulted in complex mixtures with
or without benzoic acid (entries 1 and 2). The absence
of any catalytic activity without a Lewis nucleophile,
as seen in the case of 1e’, confirmed the existing
mechanistic interpretation of the MBH in requiring a
Lewis nucleophile for initiation (entries 2 and 3).
With the necessary phosphine nucleophile, 1e was
able to catalyze the formation of 3a, independent of
benzoic acid activation, although with little enantiose-
lectivity in both cases (entries 5 and 6).
Entry Catalyst Benzoic acid Time Yield 3a
ee
[%]
[h]
[%][a]
[%][b]
[c]
1
2
3
4
5
6
7
8
PPh3
PPh3
1e’
1e’
1e
1e
1a
1b
1c
1d
1a
0
10
0
10
0
10
0
0
0
0
3
3
3
3
3
3
3
15
3
3
3
15
3
–
–
–
–
–
–
rac.
10
rac.
24
69
rac.
82
38
77
[c]
–
–
63
60
16
63
39
13
>95
77
>95
9
10
11
12
13
10
10
10
1b
1d
For 1a, 1b and 1d, both the rate of reaction and
enantioselectivity were very poor in the absence of
benzoic acid (entries 7, 8 and 10). Catalyst 1c was no-
table in that both the rate of reaction and enantiose-
lectivity were considerably higher even without
[a]
[b]
[c]
Calculated by 1H NMR spectroscopy.
Determined by chiral HPLC analysis.
Complex mixture of products without 3a.
Brønsted acid activation (entry 9). This characteristic 1a and 1d, the rate of reaction dramatically improved,
is reminiscent of that from prior bifunctional exam- and the yield changed from under 20% to nearly
ples where multiple phenol groups (multivalent H- quantitative (>95% given the detection limit of
bonding interactions from Brønsted acids) showed en- proton NMR spectroscopy) in three hours at room
hanced catalytic performance.[6j,k] However, due to temperature.[14] The enantioselectivity in these two
the low stability of this catalyst toward oxidation, its cases also significantly improved from none to good
investigation was not pursued further. Catalysts 1a, (77–82% ee). For 1b, the catalytic rate and enantiose-
1b, and 1d were then subjected to the same test reac- lectivity were also consistently low without benzoic
tion with Brønsted acid activation (entries 11–13). For acid, and the reaction time was extended to 15 h
Adv. Synth. Catal. 2009, 351, 331 – 338
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
333

Jean-Marc Garnier et al.
COMMUNICATIONS
vestigated next by comparing the activities of 1a, 1f,
and 1g in the absence or presence of benzoic acid
(Table 2). Catalyst 1f, in which the phenolic proton is
replaced with a methyl group, exhibited rate improve-
ment upon acid activation, as seen with 1a, but the
enantioselectivity had only a marginal response to
acid activation (entries 1 and 2). Catalyst 1g, in which
the Brønsted base is replaced by an isoelectronic
oxygen, exhibited the standard bifunctional catalysis
where acid activation is not required. Furthermore,
the addition of benzoic acid reduced the level of
enantioselectivity from 71% to 42% (entries 3 and 4).
Only in the case of 1a was the improvement of both
the rate of catalysis and enantioselectivity observed in
a coordinated manner (entries 5 and 6). This suggests
that the amino Brønsted base provides the switch
point from bifunctional to trifunctional catalysis in re-
sponse to acid activation. Given the essential role of
the phosphine Lewis base in reaction initiation, and
the importance of an appropriately positioned phenol-
ic Brønsted acid, all three functional groups are re-
quired for the activity of 1a.
Figure 2. Effect of benzoic acid loading on the azaMBH re-
actions of N-(4-bromophenylmethylidene)arylsulfonamide
2b (1.0 equiv.) with MVK (2.0 equiv.) in the presence of cat-
alyst 1a (10 mol%).
The next question then became the catalytic signifi-
cance of a protonated Brønsted base. An immediate
consequence is the formation of a secondary ammoni-
(entry 8). However, the addition of benzoic acid had um salt that may help with stabilizing zwitterionic in-
a minor improvement in rate and enantioselectivity
(entry 12). The position of the Brønsted acid phenol
appeared to be a determinant in whether the enabling
effect of the acid activation on the catalyst could
Table 2. Aza-MBH reactions of N-(arylmethylidene)arylsul-
fonamide 2b (1.0 equiv.) with MVK (2.0 equiv.) in the pres-
ence of catalysts 1a, 1f, and 1g (10 mol%) with or without
eventuate. This suggests that the catalytically produc-
acid additives.
tive form by acid activation was influenced by the
phenol through likely H-bonding interactions.
The effect of benzoic acid equivalence was next ex-
amined using 1a and a test azaMBH reaction between
a bromoaryl imine and MVK (Figure 2). The loading
of catalyst 1a was maintained at 10 mol% while that
of benzoic acid was varied from 5 mol% to 100
Entry Catalyst Acid additive
Time Yield
[h]
ee
mol%. When the benzoic acid loading was 5 mol%,
the reaction rate was slightly lower, leading to 85%
yield in three hours. The reaction rate appeared to
saturate at 10 mol% benzoic acid loading, with all
yields being quantitative in three hours for the re-
maining loading variations. The ee continued to im-
prove with more benzoic acid and peaked at 50 mol%
benzoic acid loading, after which point the ee was
lower. The reaction rate was not affected by the
amount of benzoic acid after the saturation point of
10 mol%. This characteristic is to the contrary of that
seen in bifunctional catalysts where the reaction yield
was severely compromised with more than 20 mol%
acid additive.[15] The tolerance of excess benzoic acid
here suggests that the Brønsted base, upon protona-
tion, may have prevented the protonation of the
Lewis base in the presence of excess benzoic acid.
With the optimized activation conditions, the roles
of the amino group and the phenolic group were in-
[%][a]
[%][b]
[c]
1
2
3
4
5
6
7
8
9
1f
1f
1g
1g
1a
1a
1a
1a
1a
–
6
49
17
46
71
42
benzoic acid
4
81
[c]
–
6
66
benzoic acid
6
78
[c]
–
3
10
rac.
92
benzoic acid
acetic acid
phosphoric acid
pyridinium
chloride
3
3
2
23
>95
20
À60[e]
31 (71[d]) À28[e]
[f]
–
–
–
[f]
10
1a
pyridinium bro- 23
mide
–
[a]
[b]
[c]
[d]
[e]
[f]
Calculated by 1H NMR spectroscopy.
Determined by chiral HPLC analysis.
No additive.
Obtained after 6 h.
Opposite enantioselectivity observed.
Starting imine recovered.
334
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 331 – 338

Enantioselective Trifunctional Organocatalysts
COMMUNICATIONS
termediates of the MBH during the reversible Mi- 7–9). Large substituents at the meta position on the
chael and aldol steps. The role of the counterion of acid additive resulted in loss of catalysis (entries 5
this chiral ammonium salt as a conjugate base could and 6). This is consistent with earlier observations
also be catalytically relevant and therefore was inves- where the counterion greatly influenced the catalytic
tigated. In addition to benzoic acid, other acids such rate and enantioselectivity. In this series of aryl car-
as acetic acid, phosphoric acid, pyridinium chloride boxylic acid additives, no change in the the sense of
and pyridinium bromide were also examined (Table 2, asymmetric induction was observed.
entries 7–10). Changing the counterion changed the
The same test reaction was next performed in vari-
rate of catalysis significantly. Acetic acid did not pro- ous solvents (Table 4). It may be difficult to deconvo-
vide much activation (entry 7). The activation by lute the solvent effect on catalysis, as many factors
phosphoric acid was stronger yet still lower than that would be altered concurrently by changing the sol-
from benzoic acid (entry 8). In addition, the sense of vent, such as acidity, solubility, conformation, etc.
asymmetric induction was reversed in both cases to However, in general, organic ammonium ion pairing
prefer the opposite enantiomer. Catalysis was lost and H-bonding interactions would be better preserved
with pyridinium chloride or pyridinium bromide (en- in non-polar aprotic solvents. Consistent with this
tries 9 and 10). These observations, taken together, premise, the rate of catalysis and enantioselectivity
suggest that the counterion of the ammonium salt were found to be generally better in non-polar and
may play a critical role in the irreversible proton- aprotic environments (entries 1–5), while not evident
transfer step, as reflected by the dependence of the was a strict correlation between catalytic proficiency
rate and the the sense of asymmetric induction on the and apparent polarity as reflected in solvent permis-
conjugate base.
sivities. When a more polar solvent such as acetoni-
The performance dependence of 1a on benzoic acid trile was used, the rate of catalysis along with enantio-
additive prompted further investigation of other aryl selectivity decreased significantly (entry 6). When a
carboxylic acid additives with varying pKa values and protic polar solvent, such as methanol, was used, cat-
shapes. Both the reaction rate and enantioselectivity alysis was completely inhibited (entry 7). This con-
were adversely affected compared to the case with firms that tight ion pairing is essential for catalysis ac-
benzoic acid (Table 3). In some cases where the pKa tivation of 1a.
of the acid additive was varied, with approximate
The substrate scope was investigated last using 1a
shape conservation, the reduction in reaction rate and and benzoic acid (Table 5). Both electron-deficient
enantioselectivity was not as severe (entries 2–4 and and electron-rich aryl N-tosylimines were amenable
to this trifunctional approach. As expected, electron-
rich imines required longer reaction times, although
Table 3. Aza-MBH reactions of N-(4-bromophenylmethyl-
ACHTUNGTRENNUNG
Table 4. Aza-MBH reactions of N-(4-bromophenylmethyli-
dene)arylsulfonamide 2b (1.0 equiv.) with MVK (2.0 equiv.)
in various solvents with catalyst 1a (10 mol%) and benzoic
acid (50 mol%).
Entry
Additive
Time
[h]
Yield
[%][a]
ee
[%][b]
1
2
3
4
5
6
7
8
9
benzoic acid
3
5.5
3
24
3.5
3.5
3
15
15
>95
57
7
92
77
n.d.
80
–
Entry
Solvent
Time [h]
Yield [%][a]
ee [%][b]
p-FC6H4COOH
o-FC6H4COOH
o-FC6H4COOH
G
T
1-naphthoic acid
1-naphthoic acid
2-naphthoic acid
1
2
3
4
5
6
7
THF
Toluene
Et2O
CHCl3
CH2Cl2
CH3CN
MeOH
3
2
4
3
32
63
>95
51
70
70
61
78
92
53
–
87
[c]
–
–
[c]
–
60
92
>95
n.d.
72
75
3
>95
39
21[c]
3
[d]
–
[a]
[a]
Calculated by 1H NMR spectroscopy.
Determined by chiral HPLC analysis.
Starting imine recovered.
Calculated by 1H NMR spectroscopy.
Determined by chiral HPLC analysis.
Very low conversion in 3 h, and reaction was extended to
[b]
[c]
[b]
[c]
21 h.
[d]
Starting materials recovered.
Adv. Synth. Catal. 2009, 351, 331 – 338
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
335

Jean-Marc Garnier et al.
COMMUNICATIONS
Table 5. Aza-Morita-Baylis–Hillman reactions of N-(arylmethylidene)arylsulfonamide 2 (1.0 equiv.) with MVK (2.0 equiv.)
in DCM with catalyst 1a (10 mol%) and benzoic acid (50 mol%).
Entry
R
Time [h]
Yield [%][a] of 3
Isolated Yield [%] of 3
ee [%]b
1
2
3
4
5
6
7
8
9
10
m-NO2C6H4, 2a
p-BrC6H4, 2b
p-ClC6H4, 2c
o-ClC6H4, 2d
p-FC6H4, 2e
p-NO2C6H4, 2f
o-NO2C6H4, 2g
p-MeC6H4, 2h
o-MeOC6H4, 2i
m-MeOC6H4, 2j
3
3
3
2.5
3.5
3
3
5
3a, >95
93
96
92
86
92
94
90
89
91
90
82
3b, >95 (39[c])
3c, 68 (>95[d])
3d, 37 (>95[d])
3e, 55 (>95[d])
3f, >95
92 (90[c])
80 (79[d])
59 (59[d])
81 (80[d])
80
3g, >95
80
3h, 19 (>95[d])
3i, 39 (>95[d])
3j, 29 (>95[d])
82 (82[d])
79 (78[d])
87 (84[d])
2.5
5
[a]
Calculated by 1H NMR spectroscopy.
Determined by chiral HPLC analysis.
Reaction performed at 08C.
Obtained after 24 h.
[b]
[c]
[d]
their ee values are comparable to those of the elec- the strategy of multivalent H-bond donors for further
tron-deficient imines. Notably, electron-rich aryl- catalysis improvement.[18]
A
In summary, a Brønsted acid-activated trifunctional
able to react to completion within a day.[16] Lowering organocatalyst 1a was used for the first time to cata-
the reaction temperature only slowed the reaction lyze azaMBH reactions between N-tosylimines and
rate with little effect on enantioselectivity (entry 2). MVK with fast rates and good enantioselectivity at
This characteristic is distinctly different from that of room temperature. The extension of bifunctional cat-
bifunctional catalysts that require low temperatures alysis to trifunctional catalysis was achieved by includ-
to restrict conformation for high enantioselectivity. ing a Brønsted base that, upon protonation, provides
Reaction time exhibited no influence on enantioselec- a secondary ammonium ion pair that positively coop-
tivity (entries 3–6 and 8–10), suggesting a catalytic erates with the phenolic Brønsted acid after nucleo-
path that is independent of product formation. The philic initiation of the reaction. This improved the
isolated yields are comparable to calculated conver- rate of catalysis for these azaMBH reactions in aprot-
sion yields for substrates that exhibited fast comple- ic solvents with good enantioselectivity without the
tion rates. When the reactions rates are slower, as need to lower the reaction temperature. For the more
seen with 3d and 3h (entries 4 and 8), the isolated difficult generic MBH counterpart, this catalyst was
yields are 5–10% lower than the conversion yields. able to catalyze a reaction between an aryl aldehyde
This suggests that that as the catalysis of the desired and MVK in a non-polar aprotic solvent although
pathway becomes less efficient, there may be more with low rate and enantioselectivity at this point. A
conversion loss to minor competing side reactions.
working hypothesis is that activation by benzoic acid
This trifunctional system was also applied to a ge- after protonation at the nitrogen center, in conjunc-
neric MBH reaction between MVK and 4-nitroben- tion with the phenol group, provided an H-bonding
zaldehyde, and furnished the desired MBH product in network and counterion that are required for the ob-
3 days in 86% yield, although with only 26% ee. Triar- served catalysis (Figure 3, proposed transition struc-
ylphosphines in their monomeric form usually exhibit tures). In the disfavoured transition structure, the
no catalytic activity in generic MBH reactions.[17] This proton-transfer step would be slow, and the rate of
raises the possibility that tuning the acidity and H- this pathway does not depend on the presence or ab-
bonding network of this system may lead to improved sence of an ion pair. The pathway through the fav-
rate and enantioselectivity for the more difficult ge- oured transition structure, however, is expedited by
neric MBH reactions with MVK. It is also possible 1a only after protonation with benzoic acid to form
that this trifunctional approach can be combined with the required chiral ion pair. The kinetic advantage
336
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 331 – 338

Enantioselective Trifunctional Organocatalysts
COMMUNICATIONS
(S)-2-({1-[2-(Diphenylphosphino)naphthalen-1-yl]-
naphthalen-2-ylamino}methyl)-4-tert-butylphenol
oxide (1d’)
To a solution of (S)-(+)-2-(diphenylphosphine oxide)-1,1’-bi-
naphthyl-2’-amine (200 mg, 0.442 mmol) 1e’ in absolute eth-
anol (8 mL) was added 2-hydroxy-5-tert-butylbenzaldehyde
(83 mg, 0.464 mmol) at room temperature and the reaction
mixture was refluxed for 2 h. After cooling, CH3CN (2 mL)
was added to solubilize the imine. The reaction mixture was
cooled to 08C and then NaBH4 (50 mg, 1.32 mmol) was
added in one portion and stirred at room temperature for
1 h. The mixture was quenched with H2O and the organic
phase was extracted with EtOAc (3 times). The combined
organic phases were washed with brine, dried over anhy-
drous magnesium sulfate and concentrated under vacuum.
The residue was purified by flash chromatography on silica
gel (eluent petroleum ether/EtOAc: 85/15) to furnish 1d’ as
an off-white solid; yield: 230 mg (characterization data in
the Supporting Information).
Figure 3. Proposed transition structures for the formation of
3.
from a counterion-facilitated proton transfer allows
the substrates to pass through this pathway faster
than other competing pathways, which is consistent
with the observation that the rate enhancement and
enantioselectivity arise jointly upon acid activation.
Without benzoic acid, catalysis by 1a was much less
effective and unable to provide enantioselectivity. As
the MBH reaction, particularly the generic version, is
a multi-step, complex reaction with many challenges
such as autocatalysis, reversibility and racemization, it
may be a prerequisite that a generally productive cat-
alytic system will need to negotiate fast turn-over as
well as enantioselectivity to out-compete intrinsic
complications. This trifunctional approach, when com-
bined with known strategies in the bifunctional frame-
work, may offer new opportunities in addressing the
rate and scope problem of this very complex reaction.
(S)-2-({1-[2-(Diphenylphosphino)naphthalen-1-yl]-
naphthalen-2-ylamino}methyl)-4-tert-butylphenol (1d)
Phosphine oxide 1e’ (207 mg, 0.328 mmol) was combined
with PhSiH3 (2 mL) in a sealed tube. The reaction mixture
was heated at 1208C for 15 h and then cooled to room tem-
perature. The PhSiH3 was flushed with N2 for 10 min and
the residue was quickly purified by flash chromatography on
silica gel (eluent petroleum ether/EtOAc: 90/10) to afford
1e as an off-white solid; yield: 171 mg (characterization data
in the Supporting Information).
Typical Reaction Procedure for Phosphines 1a–e-
Catalyzed Aza-Baylis–Hillman Reaction of N-
Tosylimines 2a–j with MVK
N-Tosylimine (0.5 mmol), phosphine (10 mol%, 0.05 mmol)
and benzoic acid (50 mol%, 0.25 mmol) were combined
under N2. DCM (0.1 mL per mg of catalyst) was added fol-
lowed by distilled methyl vinyl ketone (MVK) (1.0 mmol)
dropwise. The reaction mixture was stirred until completion.
The solvent was evaporated and the crude mixture was sub-
jected to chiral HPLC for the ee analysis and purified by
silica gel flash column chromatography.
Experimental Section
General Procedure for the Preparation of Catalysts
1a–c
To
a solution of (S)-(+)-2-(diphenylphosphino)-1,1’-bi-
naphthyl-2’-amine 1e (0.5 mmol) in absolute ethanol (4 mL)
was added salicylaldehyde (0.5 mmol) at room temperature,
and the reaction mixture was refluxed until completion (fol-
lowed by TLC). After cooling, CH3CN (2 mL) was added to
solubilize the imine and the reaction mixture was cooled to
08C and then NaBH4 (1.5 mmol) was added in one portion
and stirred at room temperature for 1 h. The mixture was
quenched with H2O, and the organic phase was extracted
with EtOAc (3 times). The combined organic phases were
washed with brine, dried over anhydrous magnesium sulfate
and concentrated under vacuum. The residue was purified
by flash chromatography on silica gel (eluent petroleum
ether/EtOAc) to afford catalysts 1a–c as off-white solids
(characterization data in the Supporting Information).
Acknowledgements
This work is supported by an Australian Research Council
Discovery Grant to F. L. (ARC-DP055068). C. A. is support-
ed by an Australian Postgraduate Award. Felix Cleeman
(Macquarie University) is thanked for his helpful discussions.
References
[1] For a general overview, see A. Berkessel, H. Grçger,
Asymmetric Organocatalysis, VCH, Weinheim, 2005.
[2] For a general definition of acid and base organocataly-
sis, see: J. Seayad, B. List, Org. Biomol. Chem. 2005, 3,
719–724. For a recent review on Lewis base catalysis,
Adv. Synth. Catal. 2009, 351, 331 – 338
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
337

Jean-Marc Garnier et al.
COMMUNICATIONS
see: S. E. Denmark, G. L. Beunter. Angew. Chem.
2008, 120, 2–82; Angew. Chem. Int. Ed. 2008, 47, 2–81.
[3] For a recent review on the MBH in general, see: a) D.
Basavaiah, A. J. Rao, T. Satyanarayana, Chem. Rev.
2003, 103, 811–891. For a recent review on the
azaMBH, see: b) Y.-L. Shi, M. Shi, Eur. J. Org. Chem.
2007, 18, 2905–2916.
[8] a) S. Mayer, B. List, Angew. Chem. 2006, 118, 4299–
4301; Angew. Chem. Int. Ed. 2006, 45, 4193–4196. For
a recent highlight on counterion catalysis, see: b) J.
Lacour, D. Linder, Science 2007, 317, 462–463.
[9] For recent reviews: a) M. S. Taylor, E. N. Jacobsen,
Angew. Chem. 2006, 118, 1550–1573; Angew. Chem.
Int. Ed. 2006, 45, 1520–1543; b) A. G. Doyle, E. N. Ja-
cobsen, Chem. Rev. 2007, 107, 5713–5743.
[10] For the first biomimetic trifunctional organocatalyst,
see: a) T. Ema, D. Tanida, T. Matsukawa, T. Sakai,
Chem. Commun. 2008, 957–958. For a recent example
of acid activated, enantioselective trifuntional organo-
catalyst for a Michael addition, see b) D.-Q. Xu, L.-P.
Wang, S.-P. Luo, Y.-F. Wang, S. Zhang, Z.-Y. Xu, Eur. J.
Org. Chem. 2008, 12, 1049–1053.
[11] For representative examples of asymmetric ammonium
salt catalysis, see: epoxidation-a) M. F. A. Adamo, V. K.
Aggarwal, J. Am. Chem. Soc. 2000, 122, 8317–8318;
b) V. K. Aggarwal, C. Lopin, F. Sandrinelli, J. Am.
Chem. Soc. 2003, 125, 7596–7601; c) X. Wang, B. List,
Angew. Chem. 2008, 120, 1135–1138; Angew. Chem.
Int. Ed. 2008, 47, 1119–1122. Michael addition: d) Y.
Xiong, Y. Wen, F. Wang, B. Gao, X. Liu, X. Huang, X.
Feng, Adv. Synth. Catal. 2007, 349, 2156–2166. Aldol
reactions: e) S. Luo, H. Xu, L. Zhang, J. Li, J.-P. Cheng,
Org Lett. 2008, 10, 653–656. For a recent review on
protonated chiral catalysts, see: f) C. Bolm, T. Ranta-
nen, I. Schiffers, L. Zani, Angew. Chem. 2005, 117,
1788–1793; Angew. Chem. Int. Ed. 2005, 44, 1758–
1763.
[4] For initial mechanistic work: a) J. S. Hill, N. S. Issacs, J.
Phys. Org. Chem. 1990, 3, 285–288. For recent mecha-
nistic investigation, see: b) L. S. Santos, C. H. Pavam,
W. P. Almeida, F. Coelho, M. N. Eberlin, Angew. Chem.
2004, 116, 4430–4433; Angew. Chem. Int. Ed. 2004, 43,
4330–4333; c) K. E. Price, S. J. Broadwater, H. M.
Jung, D. T McQuade, Org. Lett. 2005, 7, 147–150;
d) K. E. Price, S. J. Broadwater, D. T. McQuade, J. Org.
Chem. 2005, 70, 3980–3987; e) P. Buskens, J. Klanker-
mayer, W. Leitner, J. Am. Chem. Soc. 2005, 127,
16762–16763; f) V. K. Aggarwal, S. Y. Fulford, G. C.
Lloyd-Jones, Angew. Chem. 2005, 117, 1734–1736;
Angew. Chem. Int. Ed. 2005, 44, 1706–1708; g) I. T.
Raheem, E. N. Jacobsen, Adv. Synth. Catal. 2005, 347,
1701–1708; h) R. Robiette, V. K. Aggarwal, J. N.
Harvey, J. Am. Chem. Soc. 2007, 129, 15513–15525.
[5] For a review on enantioselective MBH reactions, see:
G. Masson, C. Housseman, J. Zhu, Angew. Chem. 2007,
119, 4698–4712; Angew. Chem. Int. Ed. 2007, 46, 4614–
4628.
[6] Representative BINAP phosphine-based bifunctional
organocatalysis of the azaMBH: a) M. Shi, L.-H. Chen,
Chem. Commun. 2003, 1310–1311; b) M. Shi, G.-L.
Zhao, Adv. Synth. Catal. 2004, 346, 1205–1219; c) M.
Shi, L.-H. Chen, J. Am. Chem. Soc. 2005, 127, 3790–
3800; d) K. Matsui, S. Takizawa, H. Sasai, J. Am.
Chem. Soc. 2005, 127, 3680–3681; e) M. Shi, C.-Q. Li,
Tetrahedron: Asymmetry 2005, 16, 1385–1391; f) M.
Shi, L.-H. Chen, W-D. Teng, Adv. Synth. Catal. 2005,
347, 1781–1789; g) M. Shi, L.-H. Chen, Pure Appl.
Chem. 2005, 77, 2105–2110; h) K. Matsui, K. Tanaka,
A. Horri, S. Takizawa, H. Sasai. Tetrahedron: Asym-
metry 2006, 17, 578–583; i) K. Matsui, S. Takizawa, H.
Sasai, Synlett 2006, 5, 761–765; j) Y-H. Liu, L.-H.
Chen, M. Shi, Adv. Synth. Catal. 2006, 348, 973–979;
k) K. Ito, K. Nishida, T. Gotanda, Tetrahedron Lett.
2007, 48, 6147–6149; l) Y-L. Shi, M. Shi, Adv. Synth.
Catal. 2007, 349, 2129–2135; m) M. Shi, G.-N. Ma, J.
Gao, J. Org. Chem. 2007, 72, 9779–9781; n) M.-J. Qi, T.
Ai, M. Shi, G. Li, Tetrahedron 2008, 64, 1181–1186;
o) Y-H. Liu, M. Shi, Adv. Synth. Catal. 2008, 350, 122–
128; p) X.-Y. Guan, Y.-Q. Jiang, M. Shi, Eur. J. Org.
Chem. 2008, 12, 2150–2155.
[12] For a recent review on organic ammonium salt and ion
hydrogen bonding, see: K. Sada, T. Tani, S. Shinkai,
Synlett 2006, 15, 2364–2374.
[13] a) P. Kocovsky, S. Vyskocil, M, Smrcina, Chem. Rev.
2003, 103, 3213–3245; b) S. Vyskocil, M. Smrcina, V.
Hanus, M. Polasek, P. Kocovsky, J. Org. Chem. 1998,
63, 7738–7748; c) M. Smrcina, M. Lorenc, V. Hanus, P.
Sedmera, P. Kocovsky, J. Org. Chem. 1992, 57, 1917–
1920.
[14] For catalyst 1a, the yields of product 3a in one and two
hours were 65% and 90%, respectively.
[15] 20–50 mol% benzoic acid additive could result in yield
reduction to as low as 11%. See ref.[6l]
[16] Reaction times for electron rich imine substrates under
comparable conditions can be as long as 3 days, see
ref.[6l]
[17] Polymeric triarylphosphines can catalyze generic MBH
reactions in moderate to good yield in 3–5 days at
room or elevated temperatures; see: C. Kwong, R.
Huang, M. Zhang, M. Shi, P. H. Toy, Chem. Eur. J.
2007, 13, 2369–2376.
[7] Phosphine-based trifunctional co-catalysis of the
azaMBH: a) S. Luo, X. Mi, L. Zhang, S. Liu, H. Xu, J-
P. Cheng, Angew. Chem. 2006, 118, 3165–3169; Angew.
Chem. Int. Ed. 2006, 45, 3093–3097; b) R. Gausepohl,
P. Buskens, J. Kleinen, A. Bruckmann, C. W. Lehmann,
J. Klankermayer, W. Leitner, Angew. Chem. 2006, 118,
3772–3775; Angew. Chem. Int. Ed. 2006, 45, 3689–
3692.
[18] It has just been reported that a partially racimized
chiral alkylACTHNUGTRNEUNG(diaryl)phosphine-based on the BINAP scaf-
fold can lead to completion of azaMBH reactions in 1
to 5 h between electron-rich and -deficient imines and
MVK; see: Z. Lei, G. Ma, M. Shi, Eur. J. Org. Chem.
2008, 22, 3817–3820.
338
asc.wiley-vch.de
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 331 – 338