The direct, enantioselective, one-pot, three-component, cross-Mannich reaction of aldehydes: The reason for the higher reactivity of aldimine versus aldehyde in proline-mediated Mannich and aldol reactions

Article abstract of DOI:10.1002/adsc.200505190

In the proline-mediated Mannich and aldol reactions of propanal as a nucleophile, the aldimine prepared from benzaldehyde and p-anisidine is about 7 times more reactive than the corresponding aldehyde, benzaldehyde, as an electrophile. This higher reactivity of aldimine over aldehyde is attributed to the carboxylic acid of proline protonating the basic nitrogen atom of the aldimine more effectively than the oxygen atom of the aldehyde, an explanation which has been both experimentally and theoretically verified.

Full text of DOI:10.1002/adsc.200505190

FULL PAPERS
DOI: 10.1002/adsc.200505190
The Direct, Enantioselective, One-Pot, Three-Component,
Cross-Mannich Reaction of Aldehydes: The Reason for the
Higher Reactivity of Aldimine versus Aldehyde in
Proline-Mediated Mannich and Aldol Reactions
Yujiro Hayashi,a,* Tatsuya Urushima,a Mitsuru Shoji,a Tadafumi Uchimaru,b
Isamu Shiinac
a
Department of Industrial Chemistry, Faculty of Engineering, Tokyo University of Science, Kagurazaka, Shinjuku-ku,
Tokyo 162-8601, Japan
Fax: (þ81)-3-5261-4631, e-mail: hayashi@ci.kagu.tus.ac.jp
b
Research Institute for Computational Sciences, National Institute of Advanced Industrial Science and Technology,
Tsukuba, Ibaraki 305-8568, Japan
Department of Applied Chemistry, Faculty of Science, Tokyo University of Science, Kagurazaka, Shinjuku-ku, Tokyo
c
162-8601, Japan
Received: May 8, 2005; Accepted: July 27, 2005
Abstract: In the proline-mediated Mannich and aldol trogen atom of the aldimine more effectively than the
reactions of propanal as a nucleophile, the aldimine oxygen atom of the aldehyde, an explanation which
prepared from benzaldehyde and p-anisidine is about has been both experimentally and theoretically veri-
7 times more reactive than the corresponding alde- fied.
hyde, benzaldehyde, as an electrophile. This higher
reactivity of aldimine over aldehyde is attributed to Keywords: asymmetric synthesis; imines; Mannich re-
the carboxylic acid of proline protonating the basic ni- action; organocatalysis; proline
Introduction
the combination of BF3 ·OEt and water.[4] In reductive
amination using NaBH3CN, the imine is more reactive
2
Aldehydes and aldimines are both important electro- than the corresponding carbonyl compound.[5] Howev-
philes in organic synthesis, and addition reactions to er, as far as we are aware, these have been the only exam-
these electrophiles constitute some of the most useful ples in which the aldimine is more reactive than alde-
and fundamental organic transformations. It is generally hyde.
accepted that an aldimine is less reactive toward nucle-
The catalytic, asymmetric Mannich reaction is one of
ophilic addition than its corresponding aldehyde owing the most powerful methods for the construction of chiral
to the difference in electronegativity between O and nitrogen-containing molecules. Recently several excel-
N, and the steric hindrance present in the aldimine.[1] Re- lent results have been reported in this area, some of
cently, however, an aldimine has been found to be more which are based on catalytic asymmetric additions of a
reactive than the corresponding aldehyde in the follow- preformed enolate to aldimines.[6] In addition to such
ing reactions. In the Mukaiyama aldol and Mannich re- Mannich reactions of preformed enolates, direct, cata-
actions, Kobayashi and co-workers have reported that a lytic asymmetric Mannich reactions have also been de-
silyl enol ether reacts preferentially with aldimines over veloped recently.[7] Chiral organometallic catalysts of
aldehydes in the presence of Yb(OTf)3,[2] which is in this reaction have been developed by Shibasaki,[8] Jør-
marked contrast to conventional Lewis acids such as gensen,[9] and Trost,[10] while a chiral Brønsted acid cata-
TiCl4 or TMSOTf, which afford the aldol product selec- lyst was devised by Terada.[11] Listꢀs[12] and Barbasꢀ[13]
tively. In the presence of PdCl2(PPh3)2 the reactivity of groups have independently developed the asymmetric,
allylstannanes with aldimines is higher than that with al- direct three-component Mannich reaction catalyzed by
dehydes.[3] Akiyama and co-workers have reported that proline,[14] and now organocatalysis-mediated asymmet-
the aldimine was activated chemoselectively by HBF4, ric Mannich reactions have been investigated by the sev-
or a small loading of the conventional Lewis acid or eral groups.[15]
Adv. Synth. Catal. 2005, 347, 1595 – 1604
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1595

Yujiro Hayashi et al.
FULL PAPERS
We have been studying organic reactions under high Results and Discussion
pressure induced by water-freezing,[16] and recently
found that this high pressure effectively promotes the
List–Barbas Mannich reaction between aldehyde, ani-
sidine and proline, affording the Mannich adducts in
good yield and excellent enantioselectivities, even for
Optimization of the Conditions of the Mannich
Reaction
À
substrates which cannot be obtained at ambient pres- As a typical three-component Mannich reaction, we se-
sure.[17] During the application of high pressure induced lected that involving propanal, benzaldehyde and p-
by water-freezing to the three-component cross-Man- anisidine as the nucleophilic aldehyde, electrophilic al-
nich reaction of two different aldehydes, we found that dehyde and amine, respectively. The reaction was per-
the desired Mannich product was obtained in good yield formed as follows [Eq. (2)]: To the aldimine, generated
with excellent enantioselectivity at À208C, and soon by stirring p-anisidine (1.1 equivs.), benzaldehyde
discovered that high pressure is not necessary.[18] That (1.0 equiv.), and proline (10 mol %) for 2 h in a given
is, a direct and enantioselective, one-pot, three-compo- solvent at room temperature, was added propanal
nent cross-Mannich reaction of two different aldehydes (3.0 equivs.) at a given temperature. The reaction was
proceeded at ambient pressure. A Mannich reaction in quenched after 20 h by the addition of NaBH4 in order
which one aldehyde is employed as the Mannich donor, to reduce the b-amino aldehyde initially formed to the
and the other aldehyde is utilized as a component of the corresponding b-amino alcohol 1 because of the for-
Mannich acceptor, affording a synthetically versatile in- merꢀs instability. Optimization of the reaction condi-
termediate, a b-amino aldehyde. A similar reaction was tions was performed. Screening of solvents revealed
reported by Barbasꢀ[13g] and Cordovaꢀs[19] groups inde- that DMF and NMP (N-methyl-2-pyrrolidinone) were
pendently.
suitable, affording the product in good yield with high
The corresponding cross-aldol reaction of two differ- syn- and enantio-selectivities, while the reaction
ent aldehydes has been reported to proceed at 48C by scarcely proceeded in CH3CN, CH2Cl2, THF and tol-
Northrup and MacMillan.[20] The fact that the Mannich uene. The reaction was examined in NMP at three dif-
reaction proceeds at lower temperature (À208C) than ferent temperatures, 08C, À108C and À208C with
the aldol reaction (48C) indicates that the aldimine is varying reaction times, the results being summarized
more electrophilic than the parent aldehyde.
in Figure 1. Although a high yield (86%) was obtained
The self-Mannich reaction of propanal, p-anisidine after 5 h at 08C, the yield decreased after 5 h owing to
and a catalytic amount of l-proline also proceeds side-reactions of the initially generated Mannich ad-
when all three components are mixed together at duct with propanal (vide infra). At À108C and
À208C without preformation of the aldimine
À208C, however, good yields (85%) were obtained af-
[Eq. (1)].[18] The success of this procedure also indicates ter 20 h and the yield did not decrease beyond a certain
that the self-Mannich reaction is faster than the self-al- time. The following conclusions were drawn from these
dol reaction.
results. The progress of the reaction should be moni-
tored carefully when it is carried out at 08C, because
of the rapid over-reaction, and the best reaction time
is dependent on the substrate. At À108C or À208C,
however, careful monitoring of the reaction progress
is not necessary and good yields are obtained even after
long reaction times.
ð1Þ
These results are contrary to the generally accepted no-
tion mentioned above that an aldimine is less reactive
toward nucleophilic addition than its parent aldehyde.
In order to clarify the reason for the reactivity difference
between aldimine and aldehyde in the proline-catalyzed
Mannich and aldol reactions,[21] a systematic comparison
of aldimine and aldehyde has been made, and theoreti-
cal calculations on these reactions performed, which
will be disclosed in this full paper.
ð2Þ
1596
asc.wiley-vch.de
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1595 – 1604

One-Pot, Three-Component, Cross-Mannich Reaction of Aldehydes
FULL PAPERS
Figure 1. Effects of temperature and reaction time on the
proline-mediated Mannich reaction.
Figure 2. Relationship between yield and reaction time in the
proline-mediated cross-aldol reaction of benzaldehyde and
propanal and self aldol reaction of propanal at À208C.
Experimental Comparison of the Reactivity between
Aldimine and Aldehyde
(1.0 M) and p-anisidine (0.50 M) were stirred in NMP
for 2 h in the presence of proline (0.1 M), affording N-
Next, the cross-aldol reaction of benzaldehyde and benzylidene-p-anisidine (0.50 M) and benzaldehyde
1
propanal was examined using NMP as solvent (0.50 M), as checked by H NMR. To this mixture was
[Eq. (3)]. When the reaction was performed at 08C, sev- added propanal (3.0 M) at À208C, and the reaction
eral products were obtained; the cross-aldol product, the was quenched after a certain period of time by the addi-
self-aldol product of propanal, and the aldol product of tion of NaBH4 or LiAlH4 [Eq. (4)]. Yields of the Man-
the initially generated cross- and self-aldol products nich and aldol products were determined after isolation
with propanal. On the other hand, when the reaction of b-amino alcohol 1 and 1,3-diol 2. The results are sum-
was carried out at À208C, the reaction was rather clean, marized in Figure 3. From this in situ comparison, N-
though its progress was slow. After 20 h at À208C, the benzylidene-p-anisidine is found tobe 6.8times more re-
anti-aldol product was obtained in 12% yield with active than benzaldehyde, which is in good agreement
good diastereoselectivity (anti:syn¼10:1) and excellent with the reactivity difference determined by the indivi-
enantioselectivity (90% ee). The yield of the cross-aldol dual experiments.
product versus reaction time is summarized in Figure 2.
We also performed another experiment in which the
self-aldol reaction of propanal was investigated at
À208C, the results of which are also summarized in Fig-
ure 2. Figure 2 shows that reaction rates of the cross- and
self-aldol reactions are similar.
ð3Þ
ð4Þ
From Figures 1 and 2, it was found that N-benzylidene-
p-anisidine is about 7 times more reactive than the pa-
rent aldehyde, benzaldehyde, in the proline-mediated
Theoretical Calculations
reaction with propanal at À208C.
The reactivity difference between aldimine and alde-
hyde was also investigated by direct in situ comparison. In order to explain the reactivity difference between al-
The reaction was performed as follows: benzaldehyde dimine and aldehyde as electrophiles in the proline-
Adv. Synth. Catal. 2005, 347, 1595 – 1604
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1597

Yujiro Hayashi et al.
FULL PAPERS
Figure 3. Reactivity difference of proline-mediated Mannich
and aldol reactions by direct in situ comparison at À208C.
Figure 5. Potential energy profile for the aldol reaction.
hyde and the enamine interact to generate a stable inter-
mediate 4, which is stabilized by 7.6 kcal/mol.[25] From
this intermediate 4, the aldol reaction proceeds via TS-
2, with a TS energy of 22.3 kcal/mol.
This theoretical comparison of the Mannich and aldol
reactions revealed the following phenomena: the timing
of the carboxylic acid proton transfer from proline to al-
dimine and aldehyde is completely different, which was
originally observed in the Houkꢀs calculation of the sim-
pler system.[23] In the Mannich reaction, the carboxylic
acid proton is mostly transferred to the aldimine, with
À
the newly formed C C single bond length is 2.160 Š,
À
the newly formed N C iminium bond length is
À
1.325 Š, the newly formed N H bond length is
1.050 Š and the cleaved RCO2 H bond length is
1.646 Š (TS-1). In the aldol reaction, however, proton
transfer from the carboxylic acid to aldehyde and C C
single-bond formation occur nearly simultaneously
with the newly forming O H bond length being
1.184 Š and the cleaving RCO2 H bond length is
1.230 Š in the transition state (TS-2). In other words,
À
À
À
Figure 4. Potential energy profile for the Mannich reaction.
À
mediated reaction with propanal, we have calculated the in the Mannich reaction initially protonation occurs on
Mannich reaction of N-benzylidene-p-anisidine and the basic nitrogen of the aldimine, which strongly acti-
propanal catalyzed by proline, and the aldol reaction vates it. This activation is much more effective than
of benzaldehyde and propanal catalyzed by proline us- the activation by protonation of the less basic aldehyde.
ing density functional theory (B3LYP/6-31G*).[22] As the aldimine is more effectively activated by carbox-
Houk and Bahmanyar have already calculated the ylic acid protonation than the aldehyde, it is more elec-
much simpler systems such as the Mannich reaction of trophilic than the aldehyde in proline-mediated reac-
N-ethylideneaniline and acetone[23] and the aldol reac- tions.
tion of acetone.[24]
Moreover, as the proton was mostly transferred on the
In the Mannich reaction, the enamine interacts with nitrogen of the imine in the transition state, the charge
the aldimine to generate a stable intermediate 3, which density at the transition state is more localized in the
is stabilized by 6.9 kcal/mol.[25] From this intermediate Mannich reaction than in the aldol reaction. Because
3, the Mannich reaction proceeds via TS-1 with a TS en- the solvent stabilizes the more charge-separated transi-
ergy of 22.4 kcal/mol. In the aldol reaction, benzalde- tion state, decreasing the activation energy, it would be
1598
asc.wiley-vch.de
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1595 – 1604

One-Pot, Three-Component, Cross-Mannich Reaction of Aldehydes
FULL PAPERS
Figure 6. Transition state structure for the proline-catalyzed Mannich and aldol reactions.
Figure 7. The structures of the complexes in the Mannich and aldol reactions.
reasonable that the Mannich reaction is faster than the
aldol reaction in a polar solvent such as NMP.[26]
1H and 13C NMR Study of Aldimine and Aldehyde in
the Presence of Acid
1
Figure 8. A proton examined in H NMR.
To confirm this conclusion further, 1H and 13C NMR ex-
periments[27] were carried out on the three samples; N-
benzylidene-p-anisidine (sample A), benzaldehyde
(sample B), and an equimolar mixture of N-benzyli-
dene-p-anisidine and benzaldehyde (sample C). A cer-
tain amount of CF3CO2H was added little by little to
(CD3)2SO solutions of these three samples at 300 K.
The variations in chemical shift of the aldimine and/or
aldehyde proton and carbon (Figures 8 and 9) versus
Figure 9. A carbon examined in 13C NMR.
the amount of CF3CO2H were measured, with results sults. Shifts to lower field were observed for the aldimine
summarized in Figures 10 and 11, respectively.
carbon and hydrogen in accordance with the amount of
There is a good correlation between the 1H NMR and CF3CO2H. In the case of the aldehyde, however, only
13C NMR data, and moreover the experiments on alde- small down field shifts were observed. These results
hyde and aldimine separately (samples A and B), and on clearly indicate that the aldimine is protonated effec-
direct in situ comparison (sample C), gave the same re- tively, while the aldehyde is not.
Adv. Synth. Catal. 2005, 347, 1595 – 1604
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1599

Yujiro Hayashi et al.
FULL PAPERS
ments. This concept should be applied not only to the
proline-mediated Mannich and aldol reactions but also
to other Brønsted acid-mediated reactions. Thus, we in-
vestigated the reactivity difference between aldimine
and aldehyde in the presence of Brønsted acid in the ad-
dition reaction of silyl enol ether, and allyltin and reduc-
tion by the use of Bu3SnH. The equimolar amounts of al-
dimine and aldehyde were treated with 1-phenyl-1-tri-
methylsiloxyethene,[4d] allyltributhyltin,[4b] or Bu3SnH
in the presence of an equimolar amount of CF3CO2H
with the results summarized in Table 1. All these nucle-
ophiles selectively add to aldimine, affording the corre-
sponding amine with a trace amount of alcohol generat-
ed by the reaction with aldehyde. These results clearly
indicate that Brønsted acid selectively activates imine
over aldehyde.[4]
Figure 10. Correlation between the amount of CF3CO2H and
1H NMR chemical shift.
ð5Þ
Conclusion
The reason for the higher reactivity of aldimine over al-
dehyde in the asymmetric, direct, one-pot, three-com-
ponent, cross-Mannich reaction of two different alde-
hydes has been experimentally and theoretically investi-
gated with the conclusion that the effective protonation
by acid of the aldimine basic nitrogen is the key. The gen-
erally accepted notion that aldehyde is more reactive
than aldimine would be applicable to nucleophilic addi-
tion reactions under basic conditions (Figure 12). In the
Brønsted acid-mediated reaction, however, aldimine is
more reactive than aldehyde. A related phenomenon
was reported by Yamamoto and co-workers in the rever-
sal of the reactivity between electron-rich and electron-
deficient aldehydes by the use of Lewis acid:[28] that is,
Figure 11. Correlation between the amount of CF3CO2H and
13C NMR chemical shift.
The higher reactivity of aldimine over aldehyde is at- electron-deficient aldehydes were more reactive in the
tributed to the effective protonation by acid of the aldi- reaction with basic nucleophiles, while nucleophiles re-
mine basic nitrogen, which is supported by NMR experi- acted preferentially with electron-rich aldehydes in the
Table 1. Reactivity difference between aldimine and aldehyde in the presence of Brønsted acid.[a]
Entry
Ar
Nucleophile
T [8C]
Yield [%][b]
Amine :Alcohol
Amine
Alcohol
¼
1
2
3
Ph
Ph
2-Naph
H2C C(OTMS)Ph
À40 to À25
À25
95
80
95
trace
trace
trace
>95: 5
>95: 5
>95: 5
¼
Bu3SnCH2CH CH2
Bu3SnH
À25
[a]
The reaction was performed in the presence of equimolar amount of aldehyde, aldimine and CF3CO2H.
Yield of isolated product.
[b]
1600
asc.wiley-vch.de
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1595 – 1604

One-Pot, Three-Component, Cross-Mannich Reaction of Aldehydes
FULL PAPERS
dried over Na2SO4. After removal of the volatile materials un-
der reduced pressure, the residue was purified by thin-layer
chromatography and afforded (2S,3S)-3-(p-anisidino)-2-meth-
yl-3-phenylpropan-1-ol (1).[19b]
Experimental Procedures for the Cross- and Self-
Aldol Reactions
Cross-aldol reaction (Figure 2): To an NMP (1.0 mL) solution
of benzaldehyde (106.0 mL, 1.0 mmol) and l-proline (11.5 mg,
0.1 mmol) was added propanal (216.0 mL, 3.0 mmol) at
À208C, and stirring was continued for a certain period at the
same temperature. The reaction was quenched with phosphate
buffer (pH 7.0), and the organic materials were extracted with
ethyl acetate (ꢀ3), and the organic phases were combined and
washed with brine, and dried over Na2SO4. After removal of
the volatile materials under reduced pressure, the residue
was dissolved in THF (3.0 mL), and then LiAlH4 (114.0 mg,
3.0 mmol) was added at 08C. After stirring the mixture for 30
minutes, the reaction was quenched with phosphate buffer
(pH 7.0), and the organic materials were extracted with ethyl
acetate (ꢀ3). The organic phases were then combined and
dried over Na2SO4. After removal of the volatile materials un-
der reduced pressure, the residue was purified by thin-layer
chromatography, which afforded (1R,2R)-2-methyl-1-phenyl-
propane-1,3-diol (2).[20]
Figure 12. Reactivity difference between aldimine and alde-
hyde.
presence of Lewis acid owing to the facile coordination
of the electron-rich aldehydes to the Lewis acid. The
idea of effective aldimine activation by Brønsted acid
is important, and should find wide use and application
in synthetic organic chemistry. Indeed it has already
been utilized in the recent elegant asymmetric catalytic
reaction of aldimine promoted by chiral Brønsted acid
described independently by Akiyamaꢀs[6p] and Tera-
daꢀs[11,29] groups.
Self-aldol reaction (Figure 2): To an NMP (1.0 mL) solu-
tion of l-proline (11.5 mg, 0.1 mmol) was added propanal
(216.0 mL, 3.0 mmol) at À208C, and stirring was continued
for a certain period at the same temperature. The reaction
was quenched with phosphate buffer (pH 7.0), and the organic
materials were extracted with ethyl acetate (ꢀ3), and the or-
ganic phases were combined and washed with brine, and dried
over Na2SO4. After removal of the volatile materials under re-
duced pressure, the residue was dissolved in MeOH (3.0 mL),
and then NaBH4 (113.0 mg, 3.0 mmol) was added at 08C. After
stirring the mixture for 30 minutes, the reaction was quenched
with phosphate buffer (pH 7.0), and the organic materials were
extracted with ethyl acetate (ꢀ3). The organic phases were
then combined and dried over Na2SO4. After removal of the
volatile materials under reduced pressure, the residue was pu-
rified by thin-layer chromatography, which afforded 2-methyl-
pentane-1,3-diol (2).[20]
Experimental Section
General Remarks
All reactions were carried out under an argon atmosphere and
monitored by thin-layer chromatography covered with silica
gel 60 F254. Preparative thin-layer chromatography was per-
formed using silica gel 60 F254 and B-5F. Flash chromatography
was performed using silica gel 60 and silica gel 60N. 1H and 13
C
NMR spectra were obtained as solutions in deuterated sol-
vents, referred to undeuterated solvents for 1H NMR and deu-
terated solvents for 13C NMR.
Experimental Procedure for the Investigation of
Temperature Effect in the Mannich Reaction
(Figure 1)
Experimental Procedure for the in situ Comparison of
the Reactivity Difference between Aldimine and
Aldehyde (Figure 3)
After stirring a solution of benzaldehyde (510 mL, 5.0 mmol),
p-anisidine (677.0 mg, 5.5 mmol) and l-proline (57.6 mg,
0.1 mmol) in NMP (5.0 mL) for 2 h at room temperature,
propanal (1.1 mL, 15.0 mmol) was added to the reaction mix-
ture at À208C, and stirring was continued for a certain period
at the same temperature. The reaction was quenched with
phosphate buffer (pH 7.0), and the organic materials were ex-
tracted with ethyl acetate (ꢀ3), and the organic phases were
combined and washed with brine, and dried over Na2SO4. After
removal of the volatile materials under reduced pressure, the
residue was dissolved in MeOH (3.0 mL), and then NaBH4
(3.0 mmol) was added at 08C. After stirring the mixture for
30 minutes, the reaction was quenched with phosphate buffer
(pH 7.0), and the organic materials were extracted with ethyl
acetate (ꢀ3). The organic phases were then combined and
After stirring a solution of benzaldehyde (106.0 mL, 1.0 mmol),
p-anisidine (62.0 mg, 0.5 mmol) and l-proline (11.5 mg,
0.1 mmol) in NMP (1.0 mL) for 2 h at room temperature,
propanal (216.0 mL, 3.0 mmol) was added to the reaction mix-
ture at À208C, and stirring was continued for 20 h at this tem-
perature. The reaction was quenched with phosphate buffer
(pH 7.0), and the organic materials were extracted with ethyl
acetate (ꢀ3), and the organic phases were combined and
washed with brine, and dried over Na2SO4. After removal of
the volatile materials under reduced pressure, the residue
was dissolved in MeOH (3.0 mL), and then NaBH4
(3.0 mmol) was added at 08C. After stirring the mixture for
Adv. Synth. Catal. 2005, 347, 1595 – 1604
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1601

Yujiro Hayashi et al.
FULL PAPERS
30 minutes, the reaction was quenched with phosphate buffer
(pH 7.0), and the organic materials were extracted with ethyl
acetate (ꢀ3). The organic phases were then combined and
dried over Na2SO4. After removal of the volatile materials un-
der reduced pressure, the residue was purified by thin-layer
chromatography and afforded the b-amino alcohol 1 and the
1,3-diol 2.
Experimental Procedure of Table 1, Entry 3
To an NMP (0.6 mL) solution of N-naphthylidene-4-methoxy-
aniline (78.4 mg, 0.3 mmol), naphthaldehyde (46.9 mg,
0.3 mmol) and CF3CO2H (23.1 mL, 0.3 mmol) was added
Bu3SnH (81 mL, 0.3 mmol) at À258C. After stirring the mix-
ture for 0.5 h at that temperature, the reaction was quenched
with 5% KF solution, and the organic materials were extracted
with ethyl acetate (ꢀ3), and washed with brine (ꢀ3). The or-
ganic phases were combined and dried over Na2SO4. After re-
moval of the volatile materials under reduced pressure, the res-
idue was purified by thin-layer chromatography
(AcOEt:hexane¼1:3), affording N-(2-naphthyl)methyl-p-
Experimental Procedure of NMR Study
1
A: The NMR sample was prepared as follows: N-benzylidene-
p-anisidine was synthesized according to the literature proce-
dure.[30] To a (CD3)2SO (0.5 mL) solution of N-benzylidene-
p-anisidine (53.0 mg, 0.25 mmol) was added (CD3)2SO solu-
tion containing CF3CO2H (5 M, 10 mL, 0.05 mmol) little by lit-
tle.
B: The NMR sample was prepared as follows: to a (CD3)2SO
(0.5 mL) solution of benzaldehyde (25.5 mL, 0.25 mmol) was
added (CD3)2SO solution containing CF3CO2H (5 M, 10 mL,
0.05 mmol) little by little.
anisidine; yield: 75.1 mg (95%). H NMR (600 MHz, CDCl3):
d¼3.75 (3H, s), 4.44 (2H, s), 6.65 (2H, d, J¼2.8 Hz), 6.78–
6.85 (2H, m), 7.45–7.53 (3H, m), 7.80–7.88 (3H, m); 13C
NMR (150 MHz, CDCl3): d¼49.3, 55.7, 114.1, 114.8, 125.6,
125.8, 125.9, 126.1, 127.6, 127.7, 128.2, 132.6, 133.4, 137.1,
142.3, 152.1; IR (neat): n¼2832, 1516, 1464, 1296, 1240,
1126, 1119, 1038, 823, 748 cmÀ1; HR-MS (FAB): calcd. for
C18H17NO: 263.1310; found: 263.1282.
C: The NMR sample was prepared as follows: to a (CD3)2SO
(0.5 mL) solution of N-benzylidene-p-anisidine (26.0 mg,
0.125 mmol) and benzaldehyde (13 mL, 0.125 mmol) was add-
ed (CD3)2SO solution containing CF3CO2H (2.5 M, 10 mL,
0.025 mmol) little by little.
Acknowledgements
This work was partially supported by a Grand-in-Aid for Scien-
tific Research on Priority Areas (A) “Creation of Biologically
Functional Molecules” from the Ministry of Education, Culture,
Sports, Science and Technology, Japan.
References and Notes
Experimental Procedure of Table 1, Entry 1
To an NMP (0.6 mL) solution of N-benzylidene-p-anisidine
(63.4 mg, 0.3 mmol), benzaldehyde (30.5 mL, 0.3 mmol) and
CF3CO2H (23.1 mL, 0.3 mmol) was added 1-phenyl-1-trime-
thylsiloxyethene (61.5 mL, 0.3 mmol) at À408C. After stirring
the mixture at from À408C to À258C for 3 h, the reaction was
quenched with phosphate buffer (pH 7.0), and the organic ma-
terials were extracted with ethyl acetate (ꢀ3). The organic
phases were combined and dried over Na2SO4. After removal
of the volatile materials under reduced pressure, the residue
[1] a) N. B. Ambhaikar, J. P. Snyder, D. C. Liotta, J. Am.
Chem. Soc. 2003, 125, 3690; b) F. Chemla, V. Hebbe, J.
Normant, Synthesis 2000, 75; c) M. Yamaguchi, in: Com-
prehensive Organic Synthesis, (Ed.: B. M. Trost, I. Flem-
ing), Pergamon Press, Oxford, 1991, Vol. 1, p. 325.
[2] a) S. Kobayashi, S. Nagayama, J. Org. Chem. 1997, 62,
232; b) S. Kobayashi, S. Nagayama, J. Am. Chem. Soc.
1997, 119, 10049.
[3] H. Nakamura, H. Iwama, Y. Yamamoto, J. Am. Chem.
Soc. 1996, 118, 6641.
was
purified
by
thin-layer
chromatography
(AcOEt:hexane¼1:3), affording 3-(p-anisidino)-1,3-diphe-
nylpropan-1-one;[4d] yield: 95 mg (95%).
[4] a) T. Akiyama, J. Takaya, H. Kagoshima, Chem. Lett.
1999, 947; b) T. Akiyama, J. Iwai, Y. Onuma, H. Kagosh-
ima, Chem. Commun. 1999, 2191; c) T. Akiyama, J. Ta-
kaya, H. Kagoshima, Synlett 1999, 1045; d) T. Akiyama,
J. Takaya, H. Kagoshima, Adv. Synth. Catal. 2002, 344,
338.
Experimental Procedure of Table 1, Entry 2
[5] R. F. Borch, M. D. Bernstein, H. D. Durst, J. Am. Chem.
Soc. 1971, 93, 2897.
To an NMP (0.6 mL) solution of N-benzylidene-p-anisidine
(63.4 mg, 0.3 mmol), benzaldehyde (30.5 mL, 0.3 mmol) and
CF3CO2H (23.1 mL, 0.3 mmol) was added allyltributyltin
(91.4 mL, 0.3 mmol) at À258C. After stirring the mixture for
1 h at that temperature, the reaction was quenched with 5%
KF solution, and the organic materials were extracted with eth-
yl acetate (ꢀ3), and washed with brine (ꢀ3). The organic
phases were combined and dried over Na2SO4. After removal
of the volatile materials under reduced pressure, the residue
[6] a) H. Fujieda, M. Kanai, T. Kambara, A. Iida, K. Tomio-
ka, J. Am. Chem. Soc. 1997, 119, 2060; b) H. Ishitani, M.
Ueno, S. Kobayashi, J. Am. Chem. Soc. 1997, 119, 7153;
c) S. Kobayashi, H. Ishitani, M. Ueno, J. Am. Chem.
Soc. 1998, 120, 431; d) E. Hagiwara, A. Fujii, M. Sodeo-
ka, J. Am. Chem. Soc. 1998, 120, 2474; e) D. Ferraris, B.
Young, T. Dudding, T. Lectka, J. Am. Chem. Soc. 1998,
120, 4548; f) D. Ferraris, B. Young, C. Cox, W. J. Drury
III, T. Dudding, T. Lectka, J. Org. Chem. 1998, 63,
6090; g) D. Ferraris, T. Dudding, B. Young, W. J. Drury
was
purified
by
thin-layer
chromatography
(AcOEt:hexane¼1:3), affording N-(1-phenylbut-3-enyl)-p-
anisidine;[31] yield: 60.8 mg (80%).
1602
asc.wiley-vch.de
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1595 – 1604

One-Pot, Three-Component, Cross-Mannich Reaction of Aldehydes
FULL PAPERS
III, T. Lectka, J. Org. Chem. 1999, 64, 2168; h) A. Fujii,
E. Hagiwara, M. Sodeoka, J. Am. Chem. Soc. 1999,
121, 5450; i) H. Ishitani, M. Ueno, S. Kobayashi, J. Am.
Barbas III, Org. Lett. 2004, 6, 2507; k) Review, see, W.
Notz, F. Tanaka, C. F. Barbas III, Acc. Chem. Res.
2004, 37, 580.
Chem. Soc. 2000, 122, 8180; j) D. Ferraris, B. Young, C.
Cox, T. Dudding, W. J. Drury III, L. Ryzhkov, A. E. Tag-
gi, T. Lectka, J. Am. Chem. Soc. 2002, 124, 67; k) Y. Ya-
mashita, M. Ueno, Y. Kuriyama, S. Kobayashi, Adv.
Synth. Catal. 2002, 344, 929; l) S. Kobayashi, J. Kobaya-
shi, H. Ishitani, M. Ueno, Chem. Eur. J. 2002, 8, 4185;
m) A. G. Wenzel, E. N. Jacobsen, J. Am. Chem. Soc.
2002, 124, 12964; n) Y. Nakamura, R. Matsubara, H.
Kiyohara, S. Kobayashi, Org. Lett. 2003, 5, 2481; o) S.
Kobayashi, R. Matsubara, Y. Nakamura, H. Kitagawa,
M. Sugiura, J. Am. Chem. Soc. 2003, 125, 2507; p) T.
Akiyama, J. Itoh, K. Yokota, K. Fuchibe, Angew.
Chem. 2004, 116, 1592; Angew. Chem. Int. Ed. 2004, 43,
1566; q) T. Hamada, K. Manabe, S. Kobayashi, J. Am.
Chem. Soc. 2004, 126, 7768; r) S. Kobayashi, M. Ueno,
S. Saito, Y. Mizuki, H. Ishitani, Y. Yamashita, Proc.
Natl. Acad. Sci. USA 2004, 101, 5476; s) S. Hata, T. Iwa-
sawa, M. Iguchi, K. Yamada, K. Tomioka, Synthesis 2004,
1471.
[14] Reviews of organocatalysis, see: a) B. List, Synlett 2001,
1675; b) M. Movassaghi, E. N. Jacobsen, Science, 2002,
298, 1904; c) B. List, Tetrahedron 2002, 58, 5573; d) P. I.
Dalko, L. Moisan, Angew. Chem. 2001, 113, 3840; An-
gew. Chem. Int. Ed. 2001, 40, 3726; e) B. List, Acc.
Chem. Res. 2004, 37, 548; f) P. I. Dalko, L. Moisan, An-
gew. Chem. 2004, 116, 5248; Angew. Chem. Int. Ed.
2004, 43, 5138; g) A. Berkessel, H. Groger, Asymmetric
Organocatalysis, Wiley-VCH, Weinheim, 2005.
[15] a) A. J. A. Cobb, D. M. Shaw, S. V. Ley, Synlett 2004, 558;
b) W. Zhuang, S. Saaby, K. A. Jørgensen, Angew. Chem.
2004, 116, 4576; Angew. Chem. Int. Ed. 2004, 43, 4476;
c) K. Funabiki, M. Nagamori, S. Goushi, M. Matsui,
Chem. Commun. 2004, 1928; d) W. Wang, J. Wang, H.
Li, Tetrahdron Lett. 2004, 45, 7243; e) Y. Hayashi, J. Ya-
maguchi, K. Hibino, T. Sumiya, T. Urushima, M. Shoji,
D. Hashizume, H. Koshino, Adv. Synth. Catal. 2004,
346, 1435; f) T. Ooi, M. Kameda, J. Fujii, K. Maruoka,
Org. Lett. 2004, 6, 2397; g) A. J. A. Cobb, D. M. Shaw,
D. A. Longbottom, J. B. Gold, S. V. Ley, Org. Biomol.
Chem. 2005, 3, 84.
[16] a) Y. Hayashi, K. Nishimura, Chem. Lett. 2002, 296; b) Y.
Hayashi, K. Okado, I. Ashimine, M. Shoji, Tetrahedron
Lett. 2002, 43, 8683; c) Y. Hayashi, W. Tsuboi, M. Shoji,
N. Suzuki, Tetrahedron Lett. 2004, 45, 4353.
[17] Y. Hayashi, W. Tsuboi, M. Shoji, N. Suzuki, J. Am. Chem.
Soc. 2003, 125, 11208.
[18] Y. Hayashi, W. Tsuboi, I. Ashimine, T. Urushima, M.
Shoji, K. Sakai, Angew. Chem. 2003, 115, 3805; Angew.
Chem. Int. Ed. 2003, 42, 3677.
[19] a) A. Cordova, Synlett 2003, 1651; b) A. Cordova, Chem.
Eur. J. 2004, 10, 1987.
[7] Review, see; A. Cordova, Acc. Chem. Res. 2004, 37, 102.
[8] a) S. Yamasaki, T. Iida, M. Shibasaki, Tetrahedron Lett.
1999, 40, 307; b) S. Yamasaki, T. Iida, M. Shibasaki, Tet-
rahedron 1999, 55, 8857; c) S. Matsunaga, N. Kumagai, S.
Harada, M. Shibasaki, J. Am. Chem. Soc. 2003, 125, 4712;
d) S. Matsunaga, T. Yoshida, H. Morimoto, N. Kumagai,
M. Shibasaki, J. Am. Chem. Soc. 2004, 126, 8777.
[9] a) K. Juhl, N. Gathergood, K. A. Jørgensen, Angew.
Chem. 2001, 113, 3083; Angew. Chem. Int. Ed. 2001, 40,
2995; b) L. Bernardi, A. S. Gothelf, R. G. Hazell, K. A.
Jørgensen, J. Org. Chem. 2003, 68, 2583; c) M. Marigo,
A. Kjarsgaard, K. Juhl, N. Gathergood, K. A. Jørgensen,
Chem. Eur. J. 2003, 9, 2359.
[10] B. M. Trost, L. R. Terrell, J. Am. Chem. Soc. 2003, 125,
338.
[20] A. B. Northrup, D. W. C. MacMillan, J. Am. Chem. Soc.
2002, 124, 6798.
[11] D. Uraguchi, M. Terada, J. Am. Chem. Soc. 2004, 126,
5356.
[12] a) B. List, J. Am. Chem. Soc. 2000, 122, 9336; b) B. List, P.
Pojarliev, W. T. Biller, H. J. Martin, J. Am. Chem. Soc.
2002, 124, 827; c) P. Pojarliev, W. T. Biller, H. J. Martin,
B. List, Synlett 2003, 1903.
[21] For selected studies of proline-catalyzed aldol reaction,
see: a) B. List, R. A. Lerner, C. F. Barbas III. J. Am.
Chem. Soc. 2000, 122, 2395; b) W. Notz, B. List, J. Am.
Chem. Soc. 2000, 122, 7386; c) K. Sakthivel, W. Notz, T.
Bui, C. F. Barbas III. J. Am. Chem. Soc. 2001, 123,
5260; d) A. Bøgevig, N. Kumaragurubaran, K. A. Jørgen-
sen, Chem. Commun. 2002, 620; e) C. Pidathala, L.
Hoang, N. Vignola, B. List, Angew. Chem. 2003, 115,
2891; Angew. Chem. Int. Ed. 2003, 42, 2785; f) S. Saito,
H. Yamamoto, Acc. Chem. Res. 2004, 37, 570; g) A. Ber-
kessel, B. Koch, J. Lex, Adv. Synth. Catal. 2004, 346, 1141;
h) N. Mase, F. Tanaka, C. F. Barbas III, Angew. Chem.
2004, 116, 2474; Angew. Chem. Int. Ed. 2004, 43, 2420;
i) A. B. Nothrup, I. K. Mangion, F. Hettche, D. W. C.
MacMillan, Angew. Chem. 2004, 116, 2204; Angew.
Chem. Int. Ed. 2004, 43, 2152; j) Z. Tang, F. Jiang, X.
Cui, L. Gong, A. Mi, Y. Jiang, Y. Wu, Proc. Natl.
Acad. Sci. USA 2004, 101, 5755; k) B. List, L. Hoang,
H. J. Martin, Proc. Natl. Acad. Sci. USA 2004, 101,
5839; l) I. K. Mangion, A. B. Northrup, D. W. C. MacMil-
lan, Angew. Chem. 2004, 116, 6890; Angew. Chem. Int.
Ed. 2004, 43, 6722; m) Review, see; B. List, in: Modern
[13] a) W. Notz, K. Sakthivel, T. Bui, G. Zhong, C. F. Barbas
III, Tetrahedron Lett. 2001, 42, 199; b) S. Watanabe, A.
Cordova, F. Tanaka, C. F. Barbas III, Org. Lett. 2002, 4,
4519; c) A. Cordova, C. F. Barbas III, Tetrahedron Lett.
2002, 43, 7749; d) A. Cordova, W. Notz, G. Zhong,
J. M. Betancort, C. F. Barbas III, J. Am. Chem. Soc.
2002, 124, 1842; e) A. Cordova, S. Watanabe, F. Tanaka,
W. Notz, C. F. Barbas III, J. Am. Chem. Soc. 2002, 124,
1866; f) A. Cordova, C. F. Barbas III, Tetrahedron Lett.
2003, 44, 1923; g) W. Notz, F. Tanaka, S. Watanabe,
N. S. Chowdari, J. M. Turner, R. Thayumanavan, C. F.
Barbas III, J. Org. Chem. 2003, 68, 9624; h) N. S. Chow-
dari, D. B. Ramachary, C. F. Barbas III, Synlett 2003,
1906; i) W. Notz, S. Watanabe, N. S. Chowdari, G. Zhong,
J. M. Betancort, F. Tanaka, C. F. Barbas III, Adv. Synth.
Catal. 2004, 346, 1131; j) N. S. Chowdari, J. T. Suri, C. F.
Adv. Synth. Catal. 2005, 347, 1595 – 1604
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1603

Yujiro Hayashi et al.
FULL PAPERS
Aldol Reactions, Vol. 1, (Ed.: R. Mahrwald), Wiley-
VCH, Weinheim, 2004, p. 161.
BSSEs into account (3.7 and 3.2 kcal/mol for 3 and 4, re-
spectively), we obtained stabilization energies of 6.9 and
7.6 kcal/mol for the reactant complexes 3 and 4.
[26] The Mannich reaction scarcely proceeded in CH3CN,
CH2Cl2 and toluene, while good yield was obtained in
NMP and DMF, see ref.[18]
[27] Kobayashi and co-workers have investigated the NMR
studies of aldimine and aldehyde in the presence of
Yb(OTf)3, see ref.[2]
[28] N. Asao, T. Asano, Y. Yamamoto, Angew. Chem. 2001,
113, 3306; Angew. Chem. Int. Ed. 2001, 40, 3206.
[29] D. Uraguchi, K. Sorimachi, M. Terada, J. Am. Chem. Soc.
2004, 126, 11804.
[30] K. Neuvonen, F. Fülçp, A. Koch, E. Kleinpeter, K. Pih-
laja, J. Org. Chem. 2003, 68, 2151.
[22] a) A. D. Becke, J. Chem. Phys. 1993, 98, 1372; b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; c) R.
Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys.
1971, 54, 724; d) W. J. Hehre, R. Ditchfield, J. A. Pople,
J. Chem. Phys. 1972, 56, 2257; e) P. C. Hariharan, J. A.
Pople, Theor. Chim. Acta 1973, 28, 213.
[23] S. Bahmanyar, K. N. Houk, Org. Lett. 2003, 5, 1249.
[24] a) S. Bahmanyar, K. N. Houk, J. Am. Chem. Soc. 2001,
123, 12911; b) S. Bahmanyar, K. N. Houk, H. J. Martin,
B. List, J. Am. Chem. Soc. 2003, 125, 2475; c) Review,
see: C. Allemann, R. Goldillo, F. R. Clemente, P. H.
Cheong, K. N. Houk, Acc. Chem. Res. 2004, 37, 558.
[25] By using the counterpoise method (S. F. Boys, F. Bernar-
di, Mol. Phys. 1970, 19, 553), we estimated the basis set
superposition errors (BSSEs) for the reactant complexes
3 and 4 in the Mannich and aldol reactions. After taking
[31] H. C. Aspinall, J. S. Bissett, N. Greeves, D. Levin, Tetra-
hedron Lett. 2002, 43, 323.
1604
asc.wiley-vch.de
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2005, 347, 1595 – 1604