Scope and mechanism of tandem aza-michael reaction/enantioselective protonation using a Pd-Iμ-Hydroxo complex under mild conditions buffered with amine salts

Article abstract of DOI:10.1002/asia.201000740

The tandem aza-Michael reaction/enantioselective protonation of α-substituted α,β-unsaturated carbonyl compounds is described in detail. The key to success is the combined use of a Bronsted basic palladium-Iμ-hydroxo complex and amine salts, which allows for the controlled generation of active catalyst and nucleophilic free amines. This catalytic system was applicable to various acceptors and aromatic amines, and the desired β-amino acid derivatives with a chiral center at the α position were produced in good yield with excellent enantioselectivity (up to 98% ee). For electron-deficient amines, the introduction of free amine as an additive was effective in promoting the reaction. The results of mechanistic studies, including determination of the absolute configuration of the product, are discussed. Salt-controlled reaction: The tandem aza-Michael reaction/ enantioselective protonation of α-substituted α,β-unsaturated carbonyl compounds is described in detail. The combined use of chiral palladium complex 1 and amine salts was critical to obtain α-substituted β-amino carbonyl compounds with high enantioselectivity (see scheme). Copyright

Full text of DOI:10.1002/asia.201000740

FULL PAPERS
DOI: 10.1002/asia.201000740
Scope and Mechanism of Tandem Aza-Michael Reaction/Enantioselective
Protonation Using a Pd–m-Hydroxo Complex under Mild Conditions
Buffered with Amine Salts
Yoshitaka Hamashima,[a, b] Shoko Suzuki,[a] Toshihiro Tamura,[b] Hidenori Somei,[b] and
Mikiko Sodeoka*[a, b]
Dedicated to Professor Eiichi Nakamura on the occasion of his 60th birthday
Abstract: The tandem aza-Michael re-
action/enantioselective protonation of
a-substituted a,b-unsaturated carbonyl
compounds is described in detail. The
key to success is the combined use of a
Brønsted basic palladium–m-hydroxo
complex and amine salts, which allows
for the controlled generation of active
catalyst and nucleophilic free amines.
This catalytic system was applicable to
various acceptors and aromatic amines,
and the desired b-amino acid deriva-
tives with a chiral center at the a posi-
tion were produced in good yield with
excellent enantioselectivity (up to
98% ee). For electron-deficient amines,
the introduction of free amine as an
additive was effective in promoting the
reaction. The results of mechanistic
studies, including determination of the
absolute configuration of the product,
are discussed.
Keywords: amino acids
· asym-
metric catalysis · Michael addition ·
palladium · protonation
Introduction
1.[4a] Although there are many reports on the preparation of
chiral b-substituted b-amino acids, the number of reactions
that can afford a-substituted variants is quite limited. As
such compounds are of value in medicinal chemistry, we
became interested in the development of an efficient
method to synthesize optically active a-substituted b-amino
acid derivatives. For this purpose, we investigated the utility
of a tandem aza-Michael reaction/enantioselective protona-
tion of a-substituted a,b-unsaturated carbonyl compounds.
Enantioselective protonation of enolate equivalents has
been intensively investigated because it is a powerful
method to create a stereogenic carbon center at the a posi-
tion of carbonyl compounds.[5] Among several approaches,
enantioselective protonation of preformed metal enolate
with a chiral proton source has been investigated primarily.[6]
Alternatively, in situ generation of a chiral enolate followed
by enantioselective protonation is also an attractive ap-
proach.[7] Most importantly, enantioselective protonation of
enolates is directly relevant to the conjugate addition of
(pro)nucleophiles to a,b-unsaturated carbonyl compounds.
There are several reports of successful tandem conjugate ad-
dition–protonation sequences in the literature that involve
the conjugate addition of sulfur,[8] phosphorus,[8c] aryl boron-
ic acids,[9] substituted pyroles,[10] cyanide,[11] and malonates.[12]
However, reactions initiated by the conjugate addition of ni-
trogen nucleophiles are rare, and there are only a few exam-
Natural and unnatural optically active b-amino acids consti-
tute an important class of chiral building blocks in organic
synthesis,[1] and efficient methods for their preparation are
important. For this purpose, much effort has been focused
on the development of catalytic asymmetric Mannich-type
reactions[2] and conjugate addition with hydroxylamine and
azide as representative nitrogen nucleophiles.[3] As a part of
our program on the synthesis of b-amino acid derivatives,[4]
we previously reported a highly enantioselective aza-Mi-
chael reaction using chiral palladium–m-hydroxo complexes
[a] Dr. Y. Hamashima,+ Dr. S. Suzuki, Prof. Dr. M. Sodeoka
Synthetic Organic Chemistry Laboratory
Advanced Science Institute, RIKEN
2-1 Hirosawa, Wako, Saitama 351-10198 (Japan)
Fax : (+81)48-467-9373
E-mail: sodeoka@riken.jp
[b] Dr. Y. Hamashima,+ T. Tamura, Dr. H. Somei, Prof. Dr. M. Sodeoka
Institute of Multidisciplinary Research for Advanced Materials
Tohoku University
2-1-1 Katahira, Aoba-ku, Sendai 980-8577 (Japan)
[+] Present address:
School of Pharmaceutical Sciences
University of Shizuoka
52-1 Yada, Suruga-ku, Shizuoka 422-8526 (Japan)
658
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668

ples, including our original reports.[13,14] Hii et al. examined
enantioselective protonation during their investigation on a
catalytic asymmetric aza-Michael reaction using our cationic
tion mechanism. The results were consistent with our origi-
nal hypothesis (see below). Thirdly, the absolute configura-
tion of the products was determined to shed light on the
possible mechanism of the enolate protonation step.
palladium–diaqua complex [Pd{(R)-binap}
(binap=2,2’-bis(diphenylphosphino)-1,1’-binaphthyl)
their cationic palladium–monoaqua-monoacetonitrile com-
plex [Pd{(R)-binap}(H2O)(CH3CN)](TfO)2. They achieved
ACHUTGTNREN(NUG H2O)2]ACHTNUGTRENNNUG
A
E
U
Results and Discussion
reasonably high enantioselectivity when less basic aromatic
amines, such as para-chloroaniline were employed as nucle-
ophiles. But their catalytic system was difficult to apply to
reactions with electron-rich anilines. For example, low asym-
metric induction was observed in the case of para-methoxy
aniline (anisidine) as a reaction partner. Therefore, there is
scope to improve the reaction efficiency of the tandem aza-
Michael reaction–protonation sequence.
We recently reported preliminary results on the title
tandem sequence using a-substituted a,b-unsaturated car-
bonyl compounds 2. The corresponding b-amino acid deriva-
tives 3 were produced in a highly enantioselective manner
(Scheme 1).[13] As this reaction shows excellent enantioselec-
Working Hypothesis
It is difficult to employ strongly basic amines in metal- or
Brønsted acid catalyzed asymmetric reactions, because they
coordinate to and deactivate the catalyst, and they also
readily undergo uncatalyzed racemic reactions. However,
highly reactive amine nucleophiles, such as anisidine and
benzylamine could be successfully used in our previous
asymmetric aza-Michael reaction.[4a] We expected that our
reaction system would also be applicable to enantioselective
protonation initiated by an aza-Michael reaction, as illustrat-
ed in Scheme 2. Indeed, the key to success was the com-
Scheme 1. Tandem aza-Michael reaction/enantioselective protonation.
tivity of more than 90% with broad generality, we believe
that it will be a useful method for the preparation of novel
b-amino acids. In this paper, we present full details of our
study. Firstly, the generality of the reaction was further in-
vestigated. Secondly, NMR spectroscopy and ESI-MS ex-
periments were carried out to obtain insight into the reac-
Scheme 2. Working hypothesis.
bined use of chiral palladium–m-hydroxo complex 1 and the
trifluoromethanesulfonic acid salt of amines, which was es-
sential to achieve excellent enantioselectivity. Under such
conditions, the m-hydroxo complex reacts with amine salts,
and an acid–base reaction affords a catalytically active
Lewis acidic palladium complex 4 with concomitant genera-
tion of free amine (Scheme 2). Then, the activation of 2 by 4
promotes the conjugate addition of the free amine. We ex-
pected that controlled generation of the nucleophilic amine
under our reaction conditions would be effective to suppress
the above-mentioned undesired reactions. Since the putative
palladium enolate 5 is generated in a chiral environment, we
anticipated that protonation of 5 would occur stereoselec-
Abstract in Japanese:
Chem. Asian J. 2011, 6, 658 – 668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
659

M. Sodeoka et al.
FULL PAPERS
tively. In this step, the palladium enolate has two possible
protonation pathways: intermolecular protonation from an
excess amount of amine salt is likely to occur, but intramo-
lecular proton transfer from the ammonium moiety at the b
position cannot be ruled out. It is desirable that a single
pathway is operative to achieve high asymmetric induction.
Finally, to complete the catalytic cycle, the dissociation of
the product as a salt would regenerate the active palladium
catalyst 4.
plete consumption of the substrate (entry 7). The more steri-
cally demanding 2,2’-bisACHTUNTGRNEUNG[bis(3,5-dimethylphenyl)phosphino]-
1,1’-binaphthyl (DM-binap) complex 1c did not serve as an
efficient catalyst, and only a small amount of the product
was formed (entry 8). It was possible to reduce the catalyst
amount to as little as 1 mol%. Although the reaction did
not go to completion after 12 h, results comparable to those
in entry 1 were obtained after 24 h (entry 9). As our catalyt-
ic system could maintain a low concentration of the nucleo-
philic free amine during the reaction, slow addition of anisi-
dine was not necessary, and we could achieve a high catalyst
turnover number with high stereoselectivity.
Optimization of the Reaction Conditions
At the outset, we selected a methacrylic acid derivative as a
model substrate. The 2-oxazolidinone derivative did not
show sufficient reactivity, probably because intramolecular
steric interaction between the auxiliary and the a-substitu-
ent prevents the carbonyl group and the olefin from being
coplanar. We next examined N-benzyloxycarbonyl (cbz)-
protected methacrylimide 2a and expected that such a steric
interaction would be minimized.[15] As we had hoped, the re-
action of 2a with anisidine salt (6a) proceeded smoothly in
tetrahydrofuran (THF) at room temperature in the presence
of 5 mol% of the palladium complex 1a (Table 1, entry 1),
Scope of the Reaction
To examine the generality of this reaction, reactions with
various Michael acceptors and aromatic amines were carried
out (Table 2). Under the optimized conditions, the reaction
of ethyl-substituted compound 2b afforded 3ba in 54%
Table 2. Tandem aza-Michael reaction/enantioselective protonation using
various substrates.
Table 1. Optimization of aza-Michael reaction/enantioselective protona-
tion.
Entry
Acceptor
Salt
Product
t [h]
Yield [%]
ee [%]
1
2b
2b
2c
2a
2a
2a
2a
6a
6a
6a
6b
6c
6d
6e
3ba
3ba
3ca
3ab
3ac
3ad
3ae
24
24
24
24
24
24
50
54
78
79
66
73
65
11
88
84
93
96
Entry
Pd cat.
Solvent
T
t [h]
Yield[a] [%]
ee [%]
2[a]
3
1
2
3
4
5
6
7
1a
1a
1a
1a
1a
1a
1d
1c
1a
THF
RT
08C
RT
RT
RT
08C
RT
RT
RT
8
24
3.5
24
24
12
24
71
24
80
79
82
18
NR[b]
31
86
7
94
77
88
65
–
93
94
53
94
4
5
CH2Cl2
toluene
EtOH
DMF
THF
THF
THF
THF
96
6
97
7[b]
ND[c]
[a] THF/CMPE=1:1 (CPME=cyclopentyl methyl ether). [b] 408C.
[c] ND=Not determined.
8
9[c]
79
yield with 88% ee (entry 1). Further optimization revealed
that the chemical yield was improved to 78% without signif-
icant loss of enantioselectivity when a mixed solvent system
of THF/CPME (1:1) was used (CPME=cyclopentyl methyl
ether; entry 2). In addition to alkyl-substituted compounds,
phenyl-substituted compound 2c underwent the desired re-
action without difficulty to give 3ca in good yield with ex-
cellent enantioselectivity (79% yield, 93% ee; entry 3). Un-
fortunately, substrates with bulkier substituents, such as iPr
and Bn groups did not react well. Other electron-rich aro-
matic amines could be used as Michael donors. Both the di-
oxomethylene group and methyl thioether could tolerate the
reaction conditions, and an excellent enantioselectivity of
96% was achieved in both cases (entries 4 and 5). Aniline
salt 6d also reacted with 2a to give the corresponding prod-
uct 3ad with 97% ee (entry 6). However, aromatic amines
[a] Isolated yield. [b] NR=No reaction. [c] 1 mol% 1a.
and the ee value of product 3aa was determined to be 94%
by chiral HPLC analysis. Among the solvents tested, THF
was the best. Whereas less polar solvents, such as CH2Cl2
and toluene gave decreased enantioselectivity (entries 2 and
3), polar solvents, such as EtOH and N,N-dimethylforma-
mide (DMF) were detrimental in terms of the catalytic ac-
tivity (entries 4 and 5). Reaction temperature influenced the
reaction rate. As shown in entry 6, reaction at 08C resulted
in low conversion even after 12 h, and there was no im-
provement of the ee value. We also tested other chiral li-
gands. The 5,5’-bis(diphenylphosphino)-4,4’-bi-1,3-benzo-
dioxole (SEGPHOS) complex 1d promoted the reaction,
even though a longer reaction time was necessary for com-
660
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668

Aza-Michael Reaction
with electron-withdrawing groups, such as CF3 and Br, failed
to react smoothly. For example, the adduct 3ae was ob-
tained in only 11% yield at higher temperature (408C;
entry 7). This problem was overcome by modifying the reac-
tion conditions, as will be discussed later.
The low reactivity of bulkier substrates, as described
above, might be associated with the flexible conformation of
the substituent, thereby resulting in steric repulsion with the
incoming nucleophile. If this were the case, we envisaged
that the reaction of a conformationally constrained cyclic
compound might be more efficient. Under the standard con-
ditions, the reaction of six-membered exo-methylene com-
pound 7[15] was carried out by using 6d as a Michael donor.
At room temperature, the starting material was completely
consumed after 24 h, but the addition product 8d was isolat-
ed in only 48% yield with 47% ee (Table 3, entry 1). The
observation that compound 7 was prone to undergo poly-
negligible in the case of less nucleophilic amines. Therefore,
we next examined the addition of free p-CF3C6H4NH2 (9e)
as an additive with the expectation that the initial aza-Mi-
chael reaction would be accelerated. Gratifyingly, the chemi-
cal yield was improved to 55% as the amount of 9e was in-
creased to 0.5 equiv (entries 1 and 2). But further addition
of 9e was not effective: when 1 equiv of 9e was employed,
the chemical yield was decreased to only 20% (entry 3). In-
terestingly, the ee value was in the range of 96–97% in these
reactions. These results suggest that a higher concentration
of the parent amine is necessary in the case of less nucleo-
philic amines, but the ratio of the salt to the added amine in-
fluences the reaction rate. Thus, by keeping their ratio the
same as in entry 2, twice the amount of the reagents was
used in entry 4. As a result, the chemical yield was im-
proved, and the desired product 3ae was obtained in 76%
yield with 95% ee. When the concentration of the reaction
mixture was increased to 1m, comparable results were also
obtained under the same conditions as shown in entry 3
(entry 5).
Table 3. Reactions of cyclic substrate 7.
Additionally, we tested reactions with other less nucleo-
philic aromatic amines. As shown in Scheme 3, para-bromo-
and para-iodoanilines were subjected to the modified reac-
tion conditions. After 24 h, the desired products 3af and 3ag
Entry
Salt
Product
T
t [h]
Yield [%]
ee [%]
1[a]
2
3
4
5
6d
6d
6d
6d
6a
8d
8d
8d
8d
8a
RT
08C
ꢀ208C
ꢀ208C
ꢀ208C
24
24
24
48
48
48
67
63
77
73
47
91
98
98
94
[a] 0.5m.
Scheme 3. Reactions with halogen-substituted aromatic amines under the
modified reaction conditions.
merization during storage suggested that it is more reactive
than the acyclic substrates, thus accounting for the lower
chemical yield and enantioselectivity observed in entry 1.
Better results were obtained when the concentration and
temperature were decreased. But the reaction did not pro-
ceed at ꢀ408C. The best results were obtained at ꢀ208C,
and the desired product 8d was obtained in 77% yield with
98% ee (entry 4). Under the same reaction conditions, anisi-
dine also reacted to afford the product 8a without difficulty
(entry 5). These results suggest that an s-cis conformer of
were produced in a highly enantioselective manner (96% ee
in both cases). It is noteworthy that no detectable side reac-
tions of the bromo and iodo groups with palladium were ob-
served. Since bromo and iodo groups are useful functional
handles for transition-metal-catalyzed cross-coupling reac-
tions, the availability of these compounds should enhance
the synthetic utility of this reaction. However, the sterically
hindered ortho-bromoaniline failed to react efficiently
(<10% after 24 hours).
ꢀ
acyclic substrates may be preferentially involved in the C N
bond-forming event. Unfortunately, however, we could not
apply this reaction to the corresponding five-membered ring
compound, as it was extremely unstable.
Mechanistic Studies
Modification of the Reaction Conditions
In our initial working hypothesis (Scheme 2), we speculated
that the acid–base reaction of the Brønsted basic palladium–
m-hydroxo complex 1 and the amine salt 6 would give a
monomeric Lewis acidic palladium complex 4. To confirm
As mentioned above, our reaction was not applicable to
electron-deficient amines (Table 2, entry 7). Our original
conditions were optimized for electron-rich amines by keep-
ing the concentration low to avoid occurrence of the uncata-
lyzed reaction. In contrast, such a racemic pathway seems
1
this idea, we monitored the reaction by means of H NMR
1
spectroscopy and ESI-MS. In the H NMR spectroscopic ex-
Chem. Asian J. 2011, 6, 658 – 668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
661

M. Sodeoka et al.
FULL PAPERS
periments, we used 2,2’-bis(di-p-tolylphosphino)-1,1’-bi-
naphthyl (Tol-binap) as a chiral ligand of the catalyst be-
cause it has characteristic methyl groups that are convenient
for monitoring structural change of the complex. A selected
1
region of the H NMR spectrum of 1b is shown in Figure 1
(top). Two methyl groups of 1b were observed in this range.
When this solution was treated with anisidine salt 6a (Pd/
Figure 2. Observed ESI-MS spectra of Pd complexes. a.i.=arbitrary in-
tensity.
complex 11 (12). As this complex is not C2-symmetric, it is
possible that the eight methyl groups of the tolyl groups are
magnetically different. Therefore, it is likely that complex X
is the palladium–m-hydroxo m-amido complex 11.[16] Solu-
tion B also showed a signal with a different distribution pat-
tern. Its molecular weight corresponded to a monocation of
the palladium–diaqua complex 13 (4: L=H2O) (14), thereby
indicating that complex Y is a monomeric cationic palladium
complex like 4, although its neutral ligands are not identi-
fied. When an excess amount of the salt was added (solu-
tion C, bottom), monomeric 14 was mainly observed, where-
as dimeric 11 became a minor peak.
Figure 1. Selected region of the 1H NMR spectrum of the reaction mix-
ture of the Pd complex 1b and salt 6a.
Even though the dimeric palladium–m-hydroxo m-amido
complex 11 is a minor species under the present catalytic re-
action conditions, there is still a possibility that the palladi-
um–amide complex acts as the actual nucleophile. To exam-
ine this scenario, the following experiment was carried out
(Scheme 4). Treatment of the palladium complex 1b with
1 equiv of anisidine (Pd/amine=1:1) in [D8]THF led to
clean formation of the dimeric complex 11. This result is
consistent with findings from another laboratory.[16] Probably
because the pH of the solution was not exactly the same
and because the reaction did not go to completion, the
1H NMR spectrum was slightly different from that of solu-
tion B (1b+6a (1:1)), but an identical isotopic distribution
pattern was observed in ESI-MS measurements, which clear-
ly indicated that the same dimeric palladium–m-amido com-
plex 12 was formed in each solution. 1 equiv of 2a was
added at room temperature to this mixture, but no forma-
tion of the product was observed. Although we cannot rule
6a=1:1), a new complex X with one set of eight peaks (d=
1.85, 1.87, 1.90, 1.94, 2.12, 2.15, 2.16, 2.53 ppm) appeared
(middle). This was accompanied with the formation of an-
other new complex Y with one set of two peaks (d=2.01,
2.48 ppm). When an excess amount of the salt 6a was added
to this solution (Pd/6a=1:10), the complex Y became pre-
dominant, and the complex X almost disappeared (bottom).
To obtain further information on the nature of the com-
plexes X and Y, ESI-MS measurements were carried out
(Figure 2). From solution A (Figure 2, top left), a signal
characteristic of a molecule containing two palladium atoms
with six isotopes was observed, the molecular weight of
which corresponded to a monocation of 1b (10). A similar
isotope distribution pattern was also observed in solution B
(Figure 2, top right), and this signal corresponded to a
monocation of a dimeric palladium–m-hydroxo m-amido
662
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668

Aza-Michael Reaction
To achieve high reaction efficiency, a precise balance
between the salt (proton source) and the amine is impor-
tant, and the optimum amine/salt ratio varies depending
on the nature of the amine. It is considered that the salt
plays an important role in maintaining the active catalyst.
To summarize, a plausible catalytic cycle for the amine/
salt combined system is outlined in Scheme 6. In accord
with our initial hypothesis, the active catalyst is generated
from the reaction of 1a and the excess amount of salt.
When the reaction requires additional free amine, the
catalyst would be intercepted by the additional amine to
form 11, 15, and 16. Also, the substrate that coordinates
to the palladium complex could be deprotonated to give
the inactive imide complex 17. But the salt may allow for
the regeneration of the active catalyst by means of
proton-exchange reactions. Even though the product po-
tentially coordinates to the palladium complex, the
Scheme 4. Control experiment using the Pd–m-hydroxo m-amido complex
11.
out other possibilities completely, it is most likely that the
palladium–amide complex is not involved in the present re-
action, and that the monomeric palladium complex acts as
the active catalyst.
Consideration of the Mechanism
As discussed above, our reaction requires an appropriate
free amine/salt ratio, depending on the nature of the amine
nucleophile. In the case of anisidine, a low concentration of
free anisidine is necessary to achieve high enantioselectivity.
In contrast, less nucleophilic amines require a much higher
concentration of the parent amine to promote the reaction.
Nevertheless, too much free amine has a negative effect,
perhaps due to deactivation of the catalyst through the for-
mation of unreactive complexes, such as 11 and 15–17
(Scheme 6). Actually, we observed the formation of such
Scheme 5. Reaction with 13a catalyzed by Pd–aqua complex without the
amine salt.
Table 4. Modification of the reaction conditions in the reaction of 2a
with less reactive aromatic amines.
1
complexes by means of H NMR spectroscopic and ESI-MS
measurements of mixtures of the palladium–diaqua complex
4 (L=H2O) and amines. Hii et al. also reported stepwise
formation of palladium–monoamine and palladium–diamine
complexes upon mixing the palladium–aqua complex and
aniline, as confirmed by ESI-MS and UV/Vis spectrome-
try.[14a] In addition, deprotonation of the imide proton of the
substrate is likely to occur as a result of double activation of
the Lewis acidic palladium complex and the free amine, so
that the catalyst would be deactivated.[17] We considered the
idea that these complexes were responsible for the decrease
in the catalyst turnover observed in entry 3 of Table 4. This
idea was supported by the results illustrated in Scheme 5.
Without the amine salt, the reaction was slow even in the
presence of the palladium–diaqua complex 13a (10 mol%
Pd), and the product 3ae was formed in only 13% yield
with 58% ee, which strongly indicates the importance of the
coexisting amine salt.
Entry
Salt [equiv]
Amine [equiv]
Yield [%]
ee [%]
1
2
3
1.5
1.5
1.5
3.0
1.5
0.25
0.5
1.0
1.0
0.5
35
55
20
76
64
97
96
97
95
94
4
5[a]
[a] 1m.
proton exchange from the salt may facilitate dissociation of
the product as the ammonium salt. As a result, the desired
catalytic cycle operates smoothly.
Chem. Asian J. 2011, 6, 658 – 668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
663

M. Sodeoka et al.
FULL PAPERS
from commercially available methyl (R)-3-hydroxy-2-meth-
ylpropanoate (20). It was converted into amide 21 under
basic conditions, in which partial racemization occurred. The
obtained amide was reacted with LiAlH4, and the corre-
sponding amino alcohol 22 ((S)-19) was obtained in good
yield. The absolute configuration of 19 was determined to
be R by comparison of optical rotation with that of 22. This
result clearly indicates that the product obtained by our
method has R configuration. Similar conversion and HPLC
analyses also showed that the absolute configuration of 3ad
and 3ae is R.
The sense of enantioselection can be explained as depict-
ed in Scheme 8. Based on our previous work,[4a] it is likely
that the nucleophile approaches to the alkene from the
open space opposite the phenyl group of the ligand.[19] On
Scheme 6. Plausible catalytic cycle.
Absolute Stereochemistry
To determine the absolute stereochemistry of 3aa, the fol-
lowing conversion into 1,3-amino alcohol 19 was carried out
(Scheme 7). Thus, benzylation of the nitrogen atom of 3aa
was conducted by using a modified Mitsunobu reaction[18] to
afford the corresponding product 18 in 91% yield. Reduc-
tion of 18 with NaBH4 proceeded without racemization to
give the 1,3-amino alcohol 19, for which the sign of the opti-
cal rotation was negative. An authentic sample was prepared
Scheme 8. Proposed transition-state model.
the assumption that the putative palladium enolate inter-
mediate is bidentate and that rotation of the s bond be-
tween the a-carbon and b-carbon atoms is restricted, proto-
nation should occur at the Si face of the enolate.[20] Even
though proton transfer possibly occurs intramolecularly
ꢀ
from the neighboring N H group, the intermolecular proto-
nation may be favored in the presence of a large excess
amount of the salt. If protonation occurs at the Re face of
the enolate (path A), the aminomethyl group at the devel-
oping sp3-carbon center (a position) would be exposed to
Scheme 7. Comparison of optical rotation after the conversion.
664
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668

Aza-Michael Reaction
General Procedure for Catalytic Tandem Aza-Michael Reaction/
Enantioselective Protonation
the equatorial phenyl group. Such a negative interaction
may disfavor the protonation step. In contrast, protonation
from the Si face would minimize such steric interaction.
Additionally, it appears that sufficient open space is avail-
able for a proton source to access the a position, as judged
from the X-ray structure of the complex of [Pd{(R)-
The starting material 2 (0.1 mmol), amine salts 6 (0.15 mmol), and the Pd
complex 1a (5 mol%) were dissolved in THF (0.2 mL). In the case of
less nucleophilic amines, the parent amine 9 (0.05 mmol) was included as
an additive. The resulting solution was stirred at ambient temperature for
the time given in the tables. For quenching, cold saturated aqueous
NaHCO3 (2 mL) was added under ice-bath cooling. The usual workup
followed by flash column chromatography (silica gel, hexane/ethyl ace-
tate system) gave the pure products.
binap}(N-benzoyl methacrylamide)]ACTHNUTRGENUG(N TfO)2 (Figure 4 in
Ref. [14a]). Furthermore, in path B, the aminomethyl group
located at the Re face could be a good shielding group as
well. Although other mechanisms, including O-protonation,
cannot be ruled out at present, we believe that the palladi-
um enolate is protonated from the Si face in a highly enan-
tioselective manner.
Benzyl (R)-3-(4-methoxyphenylamino)-2-methylpropionylcarbamate
(3aa)
Pale yellow oil; ½aꢁ2D5 =ꢀ26.0 (c=1.4, CHCl3) (94% ee); HPLC (Chiral-
pak AD-H, hexane/2-propanol=1:1, 1.0 mLminꢀ1, 254 nm): tr (minor)=
1
6.1 min, tr (major)=7.1 min; H NMR (400 MHz, CDCl3): d=1.23 (d, J=
6.8 Hz, 3H), 3.18 (dd, J=4.4, 12.5 Hz, 1H), 3.22–3.42 (m, 2H), 3.74 (s,
3H), 5.16 (s, 2H), 6.70 (dt, J=3.4, 9.0 Hz, 2H), 6.77 (dt, J=3.4, 9.0 Hz,
2H), 7.23–7.41 (m, 5H), 8.06 ppm (s, 1H); 13C NMR (100 MHz,
[D6]acetone): d=15.5, 40.8, 48.6, 55.8, 67.4, 115.1, 115.4, 128.9, 129.0,
129.3, 137.0, 143.0, 152.1, 153.1, 175.7 ppm; IR (neat): n˜ =3275, 2936,
1761, 1702, 1509, 1377, 1232, 1177, 1016 cmꢀ1; FAB-LRMS (m-NBA): m/
z: 342 [M]+; FAB-HRMS (PEG 400/m-NBA): m/z: calcd for
C19H22N2O4: 342.1580 [M]+, 343.1658 [M+1]+; found: 342.1588 [M+],
343.1659 [M+1]+.
Conclusion
In summary, we have succeeded in developing a tandem
aza-Michael reaction/enantioselective protonation reaction.
Our novel strategy using the palladium–m-hydroxo complex
and amine salts was the key to success. The obtained chiral
b-amino acid derivatives with a chiral center at the a posi-
tion are expected to be useful in the field of medicinal
chemistry. Because catalytic enantioselective aminomethyla-
tion—namely, classical Mannich reaction—is still difficult to
achieve,[21,22] the present work provides an important alter-
native method. Further studies to improve the catalytic ac-
tivity and to establish the scope of the reaction are under
way.
Benzyl (R)-3-(benzo[d]ACTHNUTRGNE[UNG 1,3]dioxol-5-ylamino)-2-
methylpropionylcarbamate (3ab)
Orange solid; m.p. 106.0–107.08C; ½aꢁ2D6 =ꢀ42.5 (c=0.89, CHCl3)
(96% ee);
HPLC
(Chiralcel
OD-H,
hexane/2-propanol=4:1,
1.0 mLminꢀ1
, 280 nm): tr (major)=27.6 min, tr (minor)=40.4 min;
1H NMR (CDCl3, 400 MHz): d=1.23 (d, J=6.6 Hz, 3H), 3.14–3.18 (m,
1H), 3.34–3.39 (m, 2H), 5.16 (s, 2H), 5.85 (s, 2H), 6.04 (dd, J=2.2,
8.3 Hz, 1H), 6.24 (d, J=2.2 Hz, 1H), 6.63 (d, J=8.3 Hz, 1H), 7.32–7.40
(m, 5H), 7.82 ppm (brs, 1H); 13C NMR (CDCl3, 100 MHz): d=15.1, 39.6,
48.1, 67.8, 96.4, 100.5, 104.9, 108.5, 128.2, 128.5, 134.7, 139.8, 143,0, 148.1,
150.9, 175.6 ppm; IR (solid): n˜ =3279, 2974, 2879, 1759, 1700, 1635, 1502,
1489, 1380, 1290, 1196, 1038, 1014, 928, 812, 751, 698 cmꢀ1; FAB-LRMS
(m-NBA): m/z: 356 [M]+; FAB-HRMS (m-NBA): m/z: calcd for
C19H21O5N2: 357.1450 [M+H]+; found 457.1448.
Experimental Section
General
NMR spectra were recorded with a JEOL JNM-LA 300 or 400 spectrom-
eter operating at 300 or 400 MHz for 1H NMR spectroscopy and 75 or
100.4 MHz for 13C NMR spectroscopy. Chemical shifts were reported
downfield from TMS (d=0 ppm) for 1H NMR spectroscopy. For
13C NMR spectroscopy, chemical shifts were reported in the scale relative
to CDCl3 as an internal reference. 19F NMR spectra were measured at
376 MHz, and CF3COOH (TFA) was used as an external standard. Fast
atom bombardment (FAB) low-resolution mass spectrometry (LRMS)
was taken with a JEOL JMS GCmate II by using meta-nitrobenzyl alco-
hol (m-NBA) as the matrix. FAB high-resolution (HR) MS was also
taken with a JEOL JMS GCmate II using m-NBA as the matrix and with
PEG 400 as an internal standard. ESI-MS experiments were carried out
with a Bruker Bio-TOF II. Optical rotations were measured with a
JASCO DIP-370 polarimeter. IR was measured with a Thermo Nicolet
AVATAR 370 FTIR equipped with a DuraScope accessory. Melting
points were measured with a Yanaco MP-J3. Short column chromatogra-
phy was performed using silica gel 60 (40–100 mm) purchased from Kanto
Chemical Co. Purification was carried out by using medium-pressure
liquid chromatography (MPLC). The enantiomeric excesses (ee values)
were determined by chiral HPLC. HPLC analysis was performed with a
Shimadzu HPLC system with the following equipment: pump, LC-10AD;
detector, SPD-10A set at 220 or 254 nm. A JASCO HPLC system was
also used: pump, PU-2080 Plus; detector, CD-2095 Plus set at 220 or
254 nm; column, DAICEL CHIRALPAK AD-H, AS-H, or DAICEL
CHIRALCEL OD-H, OJ-H, OF; mobile phase, hexane/2-propanol. De-
hydrated stabilizer-free tetrahydrofuran (THF) was purchased from
Kanto Chemical Co. and was used directly. Other reagents were purified
according to usual methods.
Benzyl (R)-3-[4-(methythio)phenylamino]-2-methylpropionylcarbamate
(3ac)
Pale yellow solid; m.p. 96.0–97.28C; ½aꢁ2D6 =ꢀ61.1 (c=0.87, CHCl3)
(97% ee);
HPLC
(Chiralcel
OD-H,
hexane/2-propanol=4:1,
1.0 mLminꢀ1
, 280 nm): tr (major)=18.1 min, tr (minor)=21.4 min;
1H NMR (CDCl3, 400 MHz): d=1.24 (d, J=6.6 Hz, 3H), 2.40 (s, 3H),
3.19–3.23 (m, 1H), 3.42–3.47 (m, 2H), 4.12 (brs, 1H), 5.15 (s, 2H), 6.54
(d, J=8.5 Hz, 2H), 7.19 (d, J=8.5 Hz, 2H), 7.32–7.39 (m, 5H), 7.67 ppm
(brs, 1H); 13C NMR (CDCl3, 100 MHz): d=15.2, 19.1, 39.5, 46.8, 67.9,
113.5, 124.6, 128.3, 128.5, 128.6, 131.2, 134,6, 146.2, 150.9, 175.7 ppm; IR
(solid): n˜ =3380, 3276, 3292, 3062, 3030, 2975, 2918, 2877, 1759, 1701,
1598, 1499, 1455, 1377, 1313, 1289, 1219, 1175, 1063, 1013, 816, 752,
697 cmꢀ1; FAB-LRMS (m-NBA): m/z: 358 [M]+; FAB-HRMS (m-NBA):
m/z: calcd for C19H22O3N2S: 358.1351 [M]+; found: 358.135.
Benzyl (R)-3-(phenylamino)-2-methylpropionylcarbamate (3ad)
Pale yellow oil; ½aꢁ2D5 =ꢀ43.9 (c=1.1, CHCl3) (97% ee); HPLC (Chiral-
pak AD-H, hexane/2-propanol=1:1, 0.5 mLminꢀ1, 254 nm): tr (major)=
12.2 min, tr (minor)=14.2 min; 1H NMR (400 MHz, CDCl3): d=1.23 (d,
J=6.6 Hz, 3H), 3.21–3.24 (m, 1H), 3.43–3.48 (m, 2H), 4.01 (s, 1H), 5.15
(s, 2H), 6.58–6.60 (m, 2H), 6.69–6.73 (m, 1H), 7.14–7.19 (m, 2H), 7.32–
7.41 (m, 5H), 7.82 ppm (s, 1H); 13C NMR (100 MHz, CDCl3): d=15.2,
39.7, 46.9, 67.9, 112.9, 117.7, 128.3, 128.5, 128.6, 129.2, 134.7, 147.4, 150.8,
175.6 ppm; IR (neat): n˜ =3272, 2971, 1762, 1704, 1603, 1506, 1377, 1223,
1178, 1014 cmꢀ1; FAB-LRMS (m-NBA): m/z: 313 [M+H]+; FAB-HRMS
(m-NBA): m/z: calcd for C18H20O3N2: 312.1474 [M]+; found: 312.1477.
Chem. Asian J. 2011, 6, 658 – 668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
665

M. Sodeoka et al.
FULL PAPERS
Benzyl (R)-3-(4-trifluoromethylphenylamino)-2-
methylpropionylcarbamate (3ae)
7.26–7.39 (m, 10H), 7.55 ppm (brs, 1H); 13C NMR (100 MHz, CDCl3):
d=48.0, 51.2, 55.7, 67.9, 114.7, 115.0, 128.0, 128.4, 128.5, 128.6, 128.7,
129.1, 134.7, 136.0, 141.2, 150.6, 152.5, 172.4 ppm; IR (neat): n˜ =3268,
3030, 2950, 2932, 2831, 1760, 1699, 1510, 1454, 1375, 1291, 1233, 1191,
1177, 1035, 912, 820, 751, 698 cmꢀ1; FAB-LRMS (m-NBA): m/z: 404 [M+
], 405 [M+H]+; FAB-HRMS (PEG 400/m-NBA): m/z: calcd for
C24H24N2O4: 404.1736 [M]+; found: 404.1739.
White solid; m.p. 82.2–83.28C; ½aꢁ2D4 =ꢀ41.5 (c=0.95, CHCl3) (97% ee);
HPLC (Chiralpak AS-H, hexane/EtOH=95:5, 1.0 mLminꢀ1, 280 nm): tr
(major)=20.7 min, tr (minor)=26.0 min; 1H NMR (CDCl3, 400 MHz):
d=1.25 (d, J=6.8 Hz, 3H), 3.26 (dd, J=4.4, 12.9 Hz, 1H), 3.52 (dd, J=
8.5, 12.9 Hz, 1H), 3.60 (brs, 1H), 4.40 (brs, 1H), 5.16 (s, 2H), 6.57 (d, J=
8.5 Hz, 2H), 7.32–7.39 (m, 7H), 7.59 ppm (brs, 1H); 13C NMR (CDCl3,
100 MHz): d=15.3, 39.3, 46.1, 68.1, 111.8, 118.8 (q, JC,F =32.5 Hz), 124.8
(q, JC,F =269.4 Hz), 126.5 (q, JC,F =3.3 Hz), 128.3, 128.6, 128.6, 134.5,
150.0, 151.0, 176.0 ppm; 19F NMR (CDCl3, 372.4 MHz, std: TFA): d=
14.9 ppm; IR (solid): n˜ =3380, 3275, 3185, 3036, 2977, 2933, 2875, 2846,
1760, 1700, 1616, 1498, 1456, 1418, 1377, 1324, 1263, 1224, 1186, 1104,
Benzyl (R)-2-oxo-3-[(4-methoxyphenylamino)methyl]piperidine-1-
carboxylate (8a)
Pale yellow oil; ½aꢁ2D6 =ꢀ29.0 (c=0.61, CHCl3) (93% ee); HPLC (Chiral-
pak AD-H, hexane/2-propanol=1:1, 1.0 mLminꢀ1, 254 nm): tr (minor)=
1
10.1 min, tr (major)=12.4 min; H NMR (CDCl3, 400 MHz): d=1.58–1.68
1064, 1013, 940, 825, 779 cmꢀ1
; FAB-LRMS (m-NBA): m/z: 403
(m, 1H), 1.81–1.94 (m, 2H), 2.00–2.07 (m, 1H), 2.68–2.80 (m, 1H), 3.29
(dd, J=4.6, 13.4 Hz, 1H), 3.45 (dd, J=7.1, 13.4 Hz, 1H), 3.63–3.69 (m,
1H), 3.74 (s, 3H), 3.81–3.92 (m, 1H), 5.28 (s, 2H), 6.65 (d, J=8.8 Hz,
2H), 6.78 (d, J=8.8 Hz, 2H), 7.29–7.44 ppm (m, 5H); 13C NMR (CDCl3,
100 MHz): d=21.6, 24.5, 43.1, 45.7, 46.3, 55.8, 68.6, 114.7, 114.9, 128.2,
128.4, 128.6, 135.3, 141.9, 152.3, 153.7, 173.7 ppm; IR (neat): n˜ =3272,
2936, 2826, 1776, 1764, 1708, 1692, 1558, 1540, 1512, 1502, 1231, 1173,
1011, 819, 735, 696 cmꢀ1; FAB-LRMS (m-NBA): m/z: 368 [M+]; FAB-
HRMS (PEG 400/m-NBA): m/z: calcd for C21H24O4N2: 368.1736 [M]+;
found: 368.1741.
[M+Na]+; FAB-HRMS (m-NBA): m/z: calcd for C18H19O3N2F3Na:
403.1245 [M+Na]+; found: 403.1241.
Benzyl (R)-3-(4-bromophenylamino)-2-methylpropionylcarbamate (3af)
White solid; m.p. 110.0–111.08C; ½aꢁ2D6 =ꢀ45.0 (c=1.03, CHCl3)
(96% ee); HPLC (Chiralpak AS-H, hexane/EtOH=4:1, 1.0 mLminꢀ1
,
280 nm): tr (major)=7.9 min, tr (minor)=17.0 min; 1H NMR (CDCl3,
400 MHz): d=1.23 (d, J=6.8 Hz, 3H), 3.18 (dd, J=4.4, 12.9 Hz, 1H),
3.43 (dd, J=8.8, 12.9 Hz, 1H), 3.51 (brs, 1H), 4.07 (brs, 1H), 5.15 (s,
2H), 6.43–6.45 (m, 2H), 7.20–7.23 (m, 2H), 7.32–7.40 (m, 5H), 7.70 ppm
(brs, 1H); 13C NMR (CDCl3, 100 MHz): d=15.2, 39.4, 46.7, 68.0, 109.1,
114.4, 128.3, 128.5, 128.6, 131.8, 134,6, 146.4, 150.9, 175.9 ppm; IR (solid):
n˜ =3397, 3265, 3177, 3089, 3063, 3030, 2877, 2935, 2879, 2848, 1757, 1700,
1595, 1496, 1455, 1396, 1376, 1324, 1226, 1177, 1071, 1036, 1014, 811, 776,
739, 695 cmꢀ1; FAB-LRMS (m-NBA): m/z: 393 [C18H19O3N281Br+H]+,
392 [C18H19O3N281Br]+, 391 [C18H19O3N279Br+H]+, 390 [C18H19O3N279Br]+;
FAB-HRMS (m-NBA): m/z: calcd for C18H19O3N2Br: 390.0579 [M]+;
found: 390.0583.
Benzyl (R)-2-oxo-3-[(phenylamino)methyl]piperidine-1-carboxylate (8d)
Pale yellow oil; ½aꢁ2D6 =ꢀ29.4 (c=0.65, CHCl3) (98% ee); HPLC (Chiral-
pak AD-H, hexane/2-propanol=4:1, 1.0 mLminꢀ1, 254 nm): tr (minor)=
1
12.4 min, tr (major)=14.1 min; H NMR (CDCl3, 400 MHz): d=1.59–1.68
(m, 1H), 1.80–1.96 (m, 2H), 2.02–2.10 (m, 1H), 2.70–2.78 (m, 1H), 3.34
(dd, J=4.6, 13.7 Hz, 1H), 3.51 (dd, J=7.1, 13.7 Hz, 1H), 3.62–3.68 (m,
1H), 3.85–3.91 (m, 1H), 4.44 (brs, 1H), 5.28 (s, 2H), 6.62 (d, J=8.6 Hz,
2H), 6.70 (dt, J=0.76, 7.3 Hz, 1H), 7.16 (t, J=7.3 Hz, 2H), 7.31–
7.44 ppm (m, 5H); 13C NMR (CDCl3, 100 MHz): d=21.6, 24.4, 43.2, 45.0,
45.7, 68.6, 113.1, 117.6, 128.2, 128.4, 128.6, 129.3, 135.3, 147.7, 153.7,
173.6 ppm; IR (neat): n˜ =3385, 2950, 1767, 1707, 1601, 1502, 1454, 1376,
1248, 1164, 1087, 985, 748, 692 cmꢀ1; FAB-LRMS (m-NBA): m/z: 338
[M+]; FAB-HRMS (PEG 400/m-NBA): m/z: calcd for C20H22O3N2:
338.1631 [M]+; found: 338.1632.
Benzyl (R)-3-(4-iodophenylamino)-2-methylpropionylcarbamate (3ag)
White solid; m.p. 99.0–100.08C; ½aꢁ2D7 =ꢀ50.6 (c=0.72, CHCl3) (96% ee);
HPLC (Chiralpak AS-H, hexane/EtOH=4:1, 1.0 mLminꢀ1, 280 nm): tr
(major)=8.0 min, tr (minor)=15.4 min; 1H NMR (CDCl3, 400 MHz): d=
1.22 (d, J=6.8 Hz, 3H), 3.18 (dd, J=13.2, 4.6 Hz, 1H), 3.43 (dd, J=13.2,
8.5 Hz, 1H), 3.51 (brs, 1H), 4.09 (brs, 1H), 5.15 (s, 2H), 6.36 (d, J=
8.1 Hz, 2H), 7.33–7.40 (m, 7H), 7.64 ppm (brs, 1H); 13C NMR (CDCl3,
100 MHz): d=15.2, 39.4, 46.5, 68.0, 78.1, 115.0, 128.3, 128.5, 128.6, 134.6,
137.6, 147.0, 150.9, 175.9 ppm; IR (solid): n˜ =3379, 3269, 3189, 3059,
3030, 2970, 2931, 2874, 1757, 1699, 1590, 1494, 1455, 1376, 1318, 1293,
1249, 1220, 1174, 1080, 1059, 1012, 992, 917, 810, 750, 695 cmꢀ1; FAB-
LRMS (m-NBA): m/z: 461 [M+Na]+; FAB-HRMS (m-NBA): m/z: calcd
for C18H19O3N2NaI: 461.0338 [M+Na]+; found: 461.0325.
Compound 18 (N-Benzylated 3aa)
A solution of 1,1’-azobis-N,N’-dimethylformamide (40 mg, 0.22 mmol) in
THF (1 mL) was slowly added to a solution of 3aa (50 mg, 0.146 mmol,
93% ee), benzyl alcohol (23 mL, 0.22 mmol), and tributylphosphine
(54 mL, 0.22 mmol) in THF (0.5 mL) at 08C. The resulting mixture was
stirred for 30 min at the same temperature and then overnight at room
temperature. Saturated aqueous NaHCO3 was added, and the mixture
was extracted with ethyl acetate (2ꢁ10 mL). The combined organic
layers were dried over Na2SO4 and concentrated under reduced pressure.
The obtained residue was purified by flash column chromatography
(hexane/ethyl acetate 20:1) to give the benzylated product 18 in 91%
yield (57.2 mg, 0.13 mmol, 93% ee) as a colorless oil. HPLC (Chiralpak
Benzyl (R)-2-[(4-methoxyphenylamino)methyl]butanoylcarbamate (3ba)
Yellow oil; ½aꢁ2D7 =ꢀ35.5 (c=0.6, CHCl3) (84% ee); HPLC (Chiralpak
AD-H, hexane/2-propanol=3:1, 1.0 mLminꢀ1
, 254 nm): tr (minor)=
11.0 min, tr (major)=14.2 min; 1H NMR (CDCl3, 400 MHz): d=0.98 (t,
J=7.6 Hz, 3H), 1.56–1.66 (m, 1H), 1.74–1.84 (m, 1H), 3.23–3.27 (m,
2H), 3.40 (dd, J=9.3, 13.0 Hz 1H), 3.74 (s, 3H), 5.16 (s, 2H), 6.58 (d, J=
8.8 Hz, 2H), 6.74–6.76 (d, J=8.8 Hz, 2H), 7.32–7.40 (m, 5H), 7.92 ppm
(brs, 1H); 13C NMR (CDCl3, 100 MHz): d=11.7, 23.2, 46.5, 46.9, 55.8,
67.8, 114.6, 114.8, 128.2, 128.4, 128.5, 134.7, 141.6, 150.9, 152.6,
175.2 ppm; IR (neat): n˜ =3266, 2962, 1759, 1695, 1508, 1459, 1382, 1232,
AS-H, hexane/2-propanol=7:1, 0.5 mLminꢀ1
, 254 nm): tr (major)=
24.0 min, tr (minor)=27.8 min; 1H NMR (CDCl3, 300 MHz): d=1.23 (d,
J=6.6 Hz, 3H), 3.18 (dd, J=5.0, 12.5 Hz, 1H), 3.46 (dd, J=8.4, 12.5 Hz,
1H), 3.74 (s, 3H), 3.98–4.12 (m, 1H), 4.92 (d, J=15.0 Hz, 2H), 5.14 (s,
2H), 6.51 (d, J=8.8 Hz, 2H), 6.77 (d, J=8.8 Hz, 2H), 7.10–7.40 ppm (m,
10H); 13C NMR (CDCl3, 75 MHz): d=15.9, 40.2, 47.5, 48.8, 55.8, 68.7,
114.2, 114.8, 127.2, 127.5, 128.3, 128.4, 128.56, 128.58, 134.7, 137.6, 142.1,
152.0, 154.3, 178.9 ppm; IR (neat): n˜ =3387, 2933, 1732, 1693, 1514, 1455,
1385, 1236, 1186, 999 cmꢀ1; FAB-LRMS (m-NBA): m/z: 432 [M]+.
1173, 1027, 914 cmꢀ1 FAB-LRMS (m-NBA): m/z: 356 [M]+; FAB-
;
HRMS (PEG-400/m-NBA): m/z: calcd for C20H24O4N2: 356.1736 [M]+;
found: 356.1738.
Benzyl (R)-3-(4-methoxyphenylamino)-2-phenylpropionylcarbamate
(3ca)
(R)-3-(4-Methoxyphenylamino)-2-methylpropan-1-ol ((R)-19)
NaBH4 (45 mg, 10 equiv) was added in one portion to a solution of 18
(54 mg, 0.126 mmol, 93% ee) in EtOH (2.0 mL) at 08C. The resulting
mixture was stirred for 30 min at the same temperature and then for 4 h
at room temperature. Saturated aqueous NH4Cl was added for quenching
and the mixture was extracted with ethyl acetate (2ꢁ10 mL). The com-
bined organic layers were washed with brine and dried over Na2SO4.
Pale yellow oil; ½aꢁ2D5 =+64.6 (c=0.82, CHCl3) (93% ee); HPLC (Chiral-
pak AD-H, hexane/2-propanol=3:1, 1.0 mLminꢀ1, 254 nm): tr (minor)=
1
17.2 min, tr (major)=20.9 min; H NMR (400 MHz, CDCl3): d=3.38 (dd,
J=5.5, 13.0 Hz, 1H), 3.74 (s, 3H), 3.84 (dd, J=8.7, 13.0 Hz, 1H), 4.48
(brs, 1H), 5.08 (s, 2H), 6.58 (d, J=8.8 Hz, 2H), 6.78 (d, J=8.8 Hz, 2H),
666
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668

Aza-Michael Reaction
After evaporation, the obtained residue was purified by flash column
chromatography (hexane/acetone=9:1) to give the desired amino alcohol
in 80% yield (19.7 mg, 0.1 mmol, 93% ee) as a colorless oil. ½aꢁ2D8 =ꢀ5.2
(c=0.80, CHCl3) (93% ee); HPLC (Chiralpak AS-H, hexane/2-propa-
Angew. Chem. 2003, 115, 4426; Angew. Chem. Int. Ed. 2003, 42,
4290.
[2] Reviews on catalytic asymmetric Mannich-type reactions: a) S. Ko-
bayashi, M. Ueno in Comprehensive Asymmetric Catalysis (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Supplement 1, chap-
ter 29.5, Springer, Berlin, 2004; b) A. Cꢂrdova, Acc. Chem. Res.
2004, 37, 102; c) S. Matsunaga, M. Shibasaki, Bull. Chem. Soc. Jpn.
2008, 81, 60.
[3] A short review on catalytic asymmetric aza-Michael reactions: a) L.-
W. Xu, C.-G. Xia, Eur. J. Org. Chem. 2005, 633; for more recent ex-
amples, see: b) N. Yamagiwa, N. Qin, S. Matsunaga, M. Shibasaki, J.
Am. Chem. Soc. 2005, 127, 13419; c) Y. K. Chen, M. Yoshida,
D. W. C. MacMillan, J. Am. Chem. Soc. 2006, 128, 9328; d) M. P.
Sibi, K. Itoh, J. Am. Chem. Soc. 2007, 129, 8064; e) X. Lu, L. Deng,
Angew. Chem. 2008, 120, 7824; Angew. Chem. Int. Ed. 2008, 47,
7710; f) D. Uraguchi, D. Nakashima, T. Ooi, J. Am. Chem. Soc.
2009, 131, 7242.
[4] a) Y. Hamashima, H. Somei, Y. Shimura, T. Tamura, M. Sodeoka,
Org. Lett. 2004, 6, 1861; b) Y. Hamashima, N. Sasamoto, D. Hotta,
H. Somei, N. Umebayashi, M. Sodeoka, Angew. Chem. 2005, 117,
1549; Angew. Chem. Int. Ed. 2005, 44, 1525; c) N. Sasamoto, C.
Dubs, Y. Hamashima, M. Sodeoka, J. Am. Chem. Soc. 2006, 128,
14010; d) C. Dubs, Y. Hamashima, N. Sasamoto, T. M. Seidel, S.
Suzuki, D. Hashizume, M. Sodeoka, J. Org. Chem. 2008, 73, 5859;
e) Y. Hamashima, N. Sasamoto, N. Umebayashi, M. Sodoeka, Chem.
Asian J. 2008, 3, 1443; for our review articles: f) M. Sodeoka, Y. Ha-
mashima, Bull. Chem. Soc. Jpn. 2005, 78, 941; g) M. Sodeoka, Y. Ha-
mashima, Pure Appl. Chem. 2006, 78, 477; h) M. Sodeoka, Y. Hama-
shima, Chem. Commun. 2009, 5787.
[5] Reviews on enantioselective protonation: a) L. Duhamel, P. Duha-
mel, J.-C. Plaquevent, Tetrahedron: Asymmetry 2004, 15, 3653; b) A.
Yanagisawa, H. Yamamoto in Comprehensive Asymmetric Catalysis,
Vol. 3 (Eds.: E. N. Jacobsen; A. Pfaltz, H. Yamamoto), Springer,
Berlin, 1999; c) A. Yanagisawa in Comprehensive Asymmetric Catal-
ysis, Supplement 2 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
Springer, Berlin, 2004; d) J. T. Morh, A. Y. Hong, B. M. Stoltz, Nat.
Chem. 2009, 1, 359.
[6] For recent examples of protonation of preformed enolates, see: a) S.
Nakamura, M. Kaneeda, K. Ishihara, H. Yamamoto, J. Am. Chem.
Soc. 2000, 122, 8120; b) E. Vedejs, A. W. Kruger, N. Lee, S. T.
Sakata, M. Stec, E. Suna, J. Am. Chem. Soc. 2000, 122, 4602; c) A.
Yanagisawa, T. Touge, T. Arai, Angew. Chem. 2005, 117, 1570;
Angew. Chem. Int. Ed. 2005, 44, 1546; d) C. H. Cheon, H. Yamamo-
to, J. Am. Chem. Soc. 2008, 130, 9246; e) T. Poisson, V. Dalla, F.
Marsais, G. Dupas, S. Oudeyer, V. Levacher, Angew. Chem. 2007,
119, 7220; Angew. Chem. Int. Ed. 2007, 46, 7090; f) C. Fehr, Angew.
Chem. 2007, 119, 7249; Angew. Chem. Int. Ed. 2007, 46, 7119; g) J.
Martin, J.-C. Plaquevent, J. Maddaluno, J. Rouden, M.-C. Lasne,
Eur. J. Org. Chem. 2009, 5414; h) D. Uraguchi, N. Kinoshita, T. Ooi,
J. Am. Chem. Soc. 2010, 132, 12240.
[7] Reactions starting from ketenes: a) B. L. Hodous, G. C. Fu, J. Am.
Chem. Soc. 2002, 124, 10006; b) S. L. Wiskur, G. C. Fu, J. Am.
Chem. Soc. 2005, 127, 6176; c) C. Schaefer, G. C. Fu, Angew. Chem.
2005, 117, 4682; Angew. Chem. Int. Ed. 2005, 44, 4606; decarboxyla-
tive protonation starting from b-ketoesters: d) M. A. Baur, A. Riahi,
F. Hꢃnin, J. Muzart, Tetrahedron: Asymmetry 2003, 14, 2755; e) P.
Kukula, V. Matousek, T. Mallat, A. Baiker, Chem. Eur. J. 2008, 14,
2699; f) S. C. Marinescu, T. Nishitama, J. T. Mohr, B. M. Stoltz, Org.
nol=1:1, 0.5 mLminꢀ1
, 254 nm): tr (minor)=16.6 min, tr (major)=
26.6 min; 1H NMR (CDCl3, 300 MHz): d=0.96 (d, J=6.9 Hz, 3H), 1.95–
2.10 (m, 1H), 2.69 (bs, 2H), 3.08–3.16 (m, 2H), 3.62 (dd, J=7.2, 10.5 Hz,
1H), 3.68 (dd, J=4.8, 10.5 Hz, 1H), 3.75 (s, 3H), 6.63 (d, J=8.8 Hz, 2H),
6.79 ppm (d, J=8.8 Hz, 2H); 13C NMR (CDCl3, 75 MHz): d=15.5, 35.4,
50.6, 55.8, 68.1, 114.8, 114.9, 142.3, 152.4 ppm; IR (neat): n˜ =3367, 2928,
1514, 1463, 1236, 1036, 821; FAB-LRMS (m-NBA): m/z: 195 [M]+.
(R)-2-(Hydroxymethyl)-N-(4-methoxyphenyl)propanamide (21)
nBuLi (2.5m solution in hexane, 450 mL) was added to a solution of p-
anisidine (139 mg, 2.5 equiv) in THF (5.0 mL) at ꢀ788C. The resulting
mixture was stirred for 1 h at the same temperature, and then methyl
(R)-3-hydroxy-2-methylpropionate (50 mL, 0.45 mmol, 1 equiv) was
slowly introduced at ꢀ788C. The mixture was stirred for an additional
2 h at the same temperature. After quenching with saturated aqueous
NH4Cl, the mixture was extracted with ethyl acetate (2ꢁ10 mL) and the
combined organic phase was dried over Na2SO4. Further purification was
carried out using flash column chromatography (hexane/acetone=6:1) to
give the desired amide 21 in 53% yield (53 mg, 0.25 mmol) as a white
solid. The ee value of the product was determined to be 77% by chiral
HPLC analysis. HPLC (Chiralpak AS-H, hexane/2-propanol=7:1,
1.0 mLminꢀ1
, 254 nm): tr (minor)=11.9 min, tr (major)=17.9 min;
1H NMR (CDCl3, 400 MHz): d=1.26 (d, J=6.9 Hz, 3H), 2.50–2.65 (m,
2H), 3.78–3.82 (m, 5H), 6.83 (d, J=8.8 Hz, 2H), 7.41 (d, J=8.8 Hz, 2H),
7.65 ppm (s, 1H); 13C NMR (CDCl3, 100 MHz): d=14.1, 43.2, 55.5, 65.0,
114.1, 121.7, 130.7, 156.3, 173.4 ppm; IR (solid): n˜ =3270, 1648, 1534,
1512, 1242, 1018, 824 cmꢀ1; FAB-LRMS (m-NBA): m/z: 209 [M]+, 210
[M+1]+.
(S)-3-(4-Methoxyphenylamino)-2-methylpropan-1-ol (22)
A solution of 21 (50 mg, 0.238 mmol) in THF (2 mL) was slowly added
under ice-bath cooling to a suspension of LiAlH4 (40 mg, 5 equiv) in
THF (2 mL). The resulting mixture was stirred for 30 min at room tem-
perature and then heated to reflux for 2 h. After quenching with saturat-
ed aqueous NH4Cl under ice-bath cooling, the precipitated white solid
was filtered off. The obtained organic phase was concentrated under re-
duced pressure. Further purification was performed by flash column
chromatography (hexane/acetone=6:1), and colorless amino alcohol 22
was obtained in 85% yield (39 mg, 0.20 mmol). The ee value of the prod-
uct was determined to be 73% by chiral HPLC analysis. Colorless oil;
½aꢁ2D6 =4.7 (c=0.89, CHCl3) (73% ee); HPLC (Chiralpak AS-H, hexane/
2-propanol=1:1, 0.5 mLminꢀ1
,
254 nm): tr (major)=18.8 min, tr
(minor)=31.4 min; 1H NMR (CDCl3, 300 MHz): d=0.96 (d, J=6.9 Hz,
3H), 1.95–2.10 (m, 1H), 2.69 (brs, 2H), 3.08–3.16 (m, 2H), 3.62 (dd, J=
7.2, 10.5 Hz, 1H), 3.68 (dd, J=4.8, 10.5 Hz, 1H), 3.75 (s, 3H), 6.63 (d, J=
8.8 Hz, 2H), 6.79 ppm (d, J=8.8 Hz, 2H); 13C NMR (CDCl3, 75 MHz):
d=15.5, 35.4, 50.6, 55.8, 68.1, 114.8, 114.9, 142.3, 152.4 ppm; IR (neat):
n˜ =3367, 2928, 1514, 1463, 1236, 1036, 821 cmꢀ1; FAB-LRMS (m-NBA):
m/z: 195 [M]+.
Acknowledgements
Lett. 2008, 10, 1039; reactions using
a chiral carbene catalyst:
This work was supported by a Grant-in-Aid for the Encouragement of
Young Scientists (B) (no. 18790005) and a Grant-in-Aid for Scientific Re-
search on Priority Areas (no. 19028065, “Chemistry of Concerto Cataly-
sis”), and Project Funding from RIKEN. We also thank Nippon Zeon
Corporation for the generous gift of CPME.
g) N. T. Reynolds, T. Rovis, J. Am. Chem. Soc. 2005, 127, 16406;
Proton transfer reaction: h) Y. Saga, R. Motoki, Y. Shimizu, M.
Kanai, M. Shibasaki, J. Am. Chem. Soc. 2010, 132, 7905.
[8] a) E. Emori, T. Arai, H. Sasai, M. Shibasaki, J. Am. Chem. Soc.
1998, 120, 4043; b) K. Nishimura, M. Ono, Y. Nagaoka, K. Tomioka,
Angew. Chem. 2001, 113, 454; Angew. Chem. Int. Ed. 2001, 40, 440;
c) D. Leow, S. Lin, S. K. Chittimalla, X. Fu, C.-H. Tan, Angew.
Chem. 2008, 120, 5723; Angew. Chem. Int. Ed. 2008, 47, 5641.
[1] Reviews on the synthesis of b-amino acids: a) Enantioselective Syn-
thesis of b-Amino Acids (Ed.: E. Juaristi), Wiley-VCH, New York,
1997; b) M. Liu, M. P. Sibi, Tetrahedron 2002, 58, 7991; c) J.-A. Ma,
[9] a) L. Navarre, S. Darses, J.-P. Genet, Angew. Chem. 2004, 116, 737;
Angew. Chem. Int. Ed. 2004, 43, 719; b) R. J. Moss, K. J. Wadsworth,
Chem. Asian J. 2011, 6, 658 – 668
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
667

M. Sodeoka et al.
FULL PAPERS
C. J. Chapman, C. G. Frost, Chem. Commun. 2004, 1984; c) M. P.
Sibi, H. Tatamidani, K. Patil, Org. Lett. 2005, 7, 2571; d) T. Nishi-
mura, S. Hirabayashi, Y. Yasuhara, T. Hayashi, J. Am. Chem. Soc.
2006, 128, 2556; e) C. G. Frost, S. D. Penrose, K. Lambshead, P. R.
Raithby, J. E. Warren, R. Gleave, Org. Lett. 2007, 9, 2119; f) L. Nav-
arre, R. Martinez, J.-P. Genet, S. Darses, J. Am. Chem. Soc. 2008,
130, 6159.
[18] T. Tsunoda, J. Otsuka, Y. Yamamiya, S. Ito, Chem. Lett. 1994, 539.
[19] We believe that the substrate is activated in s-cis conformation.
Since the sign of the optical rotation of 3ba suggests that it also has
R configuration, the reaction is considered to proceed with the same
conformation of the alkene part. When s-trans conformation of 2b
is taken in coordinating to the Pd complex, severe steric repulsion
would occur compared with the reaction of 2a. This idea is also sup-
ported by the results obtained in the case of the cyclic substrate 7.
[20] Hii and her co-workers reported similar reactions in reference [14a].
In analogy to the reaction of N-benzoyl methacrylamide with anisi-
dine, they assigned the absolute configuration of the N-methoxycar-
bamate derivative to be S, which suggests that the sense of enantio-
selection is different from ours. Although the substrate structure,
preparation method of the catalysts, and the reaction solvent are dif-
ferent factors to be considered, it is somewhat surprising that oppo-
site enantiomers were produced even though (R)-binap was used in
both cases. We think that this may be attributed to the use of the
amine salt, and thus we need a distinct model to explain our result.
[21] Several practical aminomethylation reactions of aldehydes using or-
ganocatalysts have been reported: a) Y. Chi, S. H. Gellman, J. Am.
Chem. Soc. 2006, 128, 6804; b) I. Ibrahem, P. Dziedzic, A. Cꢂrdova,
Synthesis 2006, 4060.
[10] M. P. Sibi, J. Coulomb, L. M. Stanley, Angew. Chem. 2008, 120,
10061; Angew. Chem. Int. Ed. 2008, 47, 9913.
[11] M. Morita, L. Drouin, R. Motoki, Y. Kimura, I. Fujimori, M. Kanai,
M. Shibasaki, J. Am. Chem. Soc. 2009, 131, 3858.
[12] T. Poisson, Y. Yamashita, S. Kobayashi, J. Am. Chem. Soc. 2010,
132, 7890.
[13] We communicated our preliminary results in reference [4a] and the
following paper: Y. Hamashima, T. Tamura, S. Suzuki, M. Sodeoka,
Synlett 2009, 1631.
[14] a) P. H. Phua, S. P. Mathew, A. J. P. White, J. G. de Vries, D. G.
Blackmond, K. K. (Mimi) Hii, Chem. Eur. J. 2007, 13, 4602; b) M. P.
Sibi, N. Prabagaran, S. G. Ghorpade, C. P. Jasperse, J. Am. Chem.
Soc. 2003, 125, 11796.
[15] The synthesis of the starting material was carried out based on the
reported procedure; a) X. Liu, X. E. Hu, X. Tian, A. Mazur, F. H.
Ebetino, J. Organomet. Chem. 2002, 646, 212; b) T. Tsujimoto, J. Ish-
ihara, M. Horie, A. Murai, Synlett 2002, 0399.
[22] Asymmetric three-component classical Mannich reactions: a) I.
Ibrahem, J. Casa, A. Cꢂrdova, Angew. Chem. 2004, 116, 6690;
Angew. Chem. Int. Ed. 2004, 43, 652. See also Ref. [4e].
[16] A similar Pd complex was previously reported. D. Nama, D. Schott,
P. S. Pregosin, L. F. Veiros, M. J. Calhorda, Organometallics 2006, 25,
4596.
[17] D. Nama, P. S. Pregosin, A. Albinati, S. Rizzato, Organometallics
2007, 26, 2111.
Received: October 9, 2010
Published online: December 16, 2010
668
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 658 – 668