Organocatalyzed michael addition of aldehydes to nitro alkenes - Generally accepted mechanism revisited and revised

Article abstract of DOI:10.1002/hlca.201100122

The amine-catalyzed enantioselective Michael addition of aldehydes to nitro alkenes (Scheme 1) is known to be acid-catalyzed (Fig. 1). A mechanistic investigation of this reaction, catalyzed by diphenylprolinol trimethylsilyl ether is described. Of the 13 acids tested, 4-NO₂-C₆H ₄OH turned out to be the most effective additive, with which the amount of catalyst could be reduced to 1 mol-% (Tables 2-5). Fast formation of an amino-nitro-cyclobutane 12 was discovered by in situ NMR analysis of a reaction mixture. Enamines, preformed from the prolinol ether and aldehydes (benzene/molecular sieves), and nitroolefins underwent a stoichiometric reaction to give single all-trans-isomers of cyclobutanes (Fig. 3) in a [2+2] cycloaddition. This reaction was shown, in one case, to be acid-catalyzed (Fig. 4) and, in another case, to be thermally reversible (Fig. 5). Treatment of benzene solutions of the isolated amino-nitro-cyclobutanes with H₂O led to mixtures of 4-nitro aldehydes (the products 7 of overall Michael addition) and enamines 13 derived thereof (Figs. 6-9). From the results obtained with specific examples, the following tentative, general conclusions are drawn for the mechanism of the reaction (Schemes 2 and 3): enamine and cyclobutane formation are fast, as compared to product formation; the zwitterionic primary product 5 of C,C-bond formation is in equilibrium with the product of its collapse (the cyclobutane) and with its precursors (enamine and nitro alkene); when protonated at its nitronate anion moiety the zwitterion gives rise to an iminium ion 6, which is hydrolyzed to the desired nitro aldehyde 7 or deprotonated to an enamine 13. While the enantioselectivity of the reaction is generally very high (>97% ee), the diastereoselectivity depends upon the conditions, under which the reaction is carried out (Fig. 10 and Table 1-5). Various acid-catalyzed steps have been identified. The cyclobutanes 12 may be considered an off-cycle 'reservoir' of catalyst, and the zwitterions 5 the 'key players' of the process (bottom part of Scheme 2 and Scheme 3). Copyright

Full text of DOI:10.1002/hlca.201100122

Helvetica Chimica Acta – Vol. 94 (2011)
719
Organocatalyzed Michael Addition of Aldehydes to Nitro Alkenes –
Generally Accepted Mechanism Revisited and Revised
by Krystyna Patora-Komisarskaa)b), Meryem Benohouda), Hayato Ishikawaa), Dieter Seebach*b), and
Yujiro Hayashi*a)
a) Tokyo University of Science, Department of Industrial Chemistry, Faculty of Engineering,
Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan
(fax: þ 81-3-5261-4631; e-mail: hayashi@ci.kagu.tus.ac.jp)
b) Laboratorium fꢀr Organische Chemie, Departement fꢀr Chemie und Angewandte Biowissenschaften,
ETH-Zꢀrich, HCI Hçnggerberg, Wolfgang-Pauli-Strasse 10, CH-8093 Zꢀrich
The amine-catalyzed enantioselective Michael addition of aldehydes to nitro alkenes (Scheme 1) is
known to be acid-catalyzed (Fig. 1). A mechanistic investigation of this reaction, catalyzed by
diphenylprolinol trimethylsilyl ether is described. Of the 13 acids tested, 4-NO2ꢀC6H4OH turned out to
be the most effective additive, with which the amount of catalyst could be reduced to 1 mol-% (Tables 2 –
5). Fast formation of an amino-nitro-cyclobutane 12 was discovered by in situ NMR analysis of a reaction
mixture. Enamines, preformed from the prolinol ether and aldehydes (benzene/molecular sieves), and
nitroolefins underwent a stoichiometric reaction to give single all-trans-isomers of cyclobutanes (Fig. 3)
in a [2 þ 2] cycloaddition. This reaction was shown, in one case, to be acid-catalyzed (Fig. 4) and, in
another case, to be thermally reversible (Fig. 5). Treatment of benzene solutions of the isolated amino-
nitro-cyclobutanes with H2O led to mixtures of 4-nitro aldehydes (the products 7 of overall Michael
addition) and enamines 13 derived thereof (Figs. 6 – 9). From the results obtained with specific examples,
the following tentative, general conclusions are drawn for the mechanism of the reaction (Schemes 2 and
3): enamine and cyclobutane formation are fast, as compared to product formation; the zwitterionic
primary product 5 of C,C-bond formation is in equilibrium with the product of its collapse (the
cyclobutane) and with its precursors (enamine and nitro alkene); when protonated at its nitronate anion
moiety the zwitterion gives rise to an iminium ion 6, which is hydrolyzed to the desired nitro aldehyde 7
or deprotonated to an enamine 13. While the enantioselectivity of the reaction is generally very high
(> 97% ee), the diastereoselectivity depends upon the conditions, under which the reaction is carried out
(Fig. 10 and Tables 1 – 5). Various acid-catalyzed steps have been identified. The cyclobutanes 12 may be
considered an off-cycle ꢁreservoirꢂ of catalyst, and the zwitterions 5 the ꢁkey playersꢂ of the process
(bottom part of Scheme 2 and Scheme 3).
1. Introduction. – The Michael addition is widely recognized as one of the most
important C,C-bond-forming transformations in organic synthesis [1]. The reaction of a
nucleophile with a Michael acceptor can be catalyzed by small organic molecules via
enamines and iminium ions as reactive intermediates [2 – 8]. Among such reactions,
organocatalyzed Michael addition of carbonyl compounds to nitro alkenes, which
proceeds via an enamine, is particularly important, because it affords synthetically
useful g-nitro carbonyl compounds with excellent diastereoselectivities. In the early
1980s, Seebach and co-workers reported the reaction of achiral enamines with b-
nitrostyrene yielding the Michael adducts in good yields and with excellent
diastereoselectivities [9]. The same authors obtained Michael adducts in diastereo-
and enantiomerically pure form by the stoichiometric reaction of enamines, derived
from prolinol methyl ether, with nitro alkenes [10]. In 2001, the first catalytic version of
ꢃ 2011 Verlag Helvetica Chimica Acta AG, Zꢀrich

720
Helvetica Chimica Acta – Vol. 94 (2011)
this transformation was independently developed by List et al. [11], and Betancort and
Barbas [12]. Since these initial results, extensive efforts have been devoted to the
development of more selective and efficient catalytic systems.
In 2005, Jørgensen and co-workers [13], and Hayashi et al. [14], independently,
discovered that diphenylprolinol silyl ether is an effective organocatalyst [15][16].
Hayashi et al. reported its application for the highly enantioselective, catalytic Michael
addition of aldehydes to nitro alkenes. The ꢁacid testꢂ of new synthetic methodology is
still its applicability in challenging total syntheses of biologically active compounds.
Michael additions of aldehydes to nitro alkenes catalyzed by diphenylprolinol silyl
ether turned out to be a powerful method for the synthesis of compounds such as
oseltamivir [17][18] and ABT-341 [19]. During the course of these synthetic studies, we
had noticed that an acid additive improves remarkably reactivity and diastereoselec-
tivity of the Michael reaction. As this transformation consists of several steps, we have
now investigated the effect of acid in detail for each step. Furthermore, in situ NMR
studies have now revealed the presence of an intermediate cyclobutane formed from
the catalyst-derived enamine and nitro alkene by a [2 þ 2] cycloaddition. Herein, we
discuss the effect(s) of added acid in the diphenylprolinolsilyl ether-catalyzed Michael
addition of aldehydes to nitro alkenes, and we challenge the generally accepted mechanism.
2. Previous Results. – 2.1. Effect of Additives. The beneficial effect of additives, and
in particular, of acids, in the organocatalyzed Michael reaction of an aldehyde and a
nitro alkene is well-known and has been reported in the literature (see Fig. 1), but its
cause is still unclear. Alexakis and Andrey reported the use of catalytic amounts of
toluenesulfonic acid (TsOH) or HCl in the addition of aliphatic aldehydes (propanal,
butanal, pentanal, and isovaleraldehyde) to b-nitrostyrene in the presence of a diamine
catalyst [20]: TsOH gives rise to higher diastereoselectivities but lower enantioselec-
tivities, as compared to the reaction without acid additive; with HCl, the enantio- and
diastereoselectivity are higher, but reactions are slower than those without acid. The
combination chiral-diamine/TFA as catalyst for the Michael addition of a,a-disub-
stituted aldehydes to b-nitrostyrene was reported by Barbas and co-workers [21]; there
is a small increase of yields and enantioselectivities with concomitant decrease of
reaction rate, as compared to the reaction carried out in the absence of acid.
Wennemers and co-workers [22] also used the CF3COOH (TFA) salt of their tripeptide
catalyst in the Michael addition of aliphatic aldehydes (propanal, butanal, pentanal,
isovaleraldehyde, and 3-phenylpropanal) to nitro alkenes. Kotsuki and co-workers [23]
reported Michael additions of ketones and an aldehyde to b-nitrostyrenes catalyzed by
a pyrrolidineꢀpyridine conjugate base catalyst and 2,4-dinitrobenzenesulfonic acid as
an additive: although excellent results were obtained with this catalyst system in the
case of ketones, the product of addition of isovaleraldehyde to b-nitrostyrene was
obtained with a very low enantio- but still a good diastereoselectivity. Moorthy and co-
workers [24] have screened a wide range of conditions, including acid additives, for the
Michael addition of cyclohexanone to b-nitrostyrene catalyzed by functionalized
proline derivatives containing two H-bond donors; the best results were obtained when
brine was used as solvent and benzoic acid as additive; only one example of the addition
of an aldehyde (propanal) to b-nitrostyrene was reported, and the product was
obtained in excellent yield, and diastereo- and enantioselectivity (stereoselectivity).

Helvetica Chimica Acta – Vol. 94 (2011)
721
Finally, Ni and co-workers [25] also selected benzoic acid as the best additive for the
Michael addition of aliphatic aldehydes to nitro alkenes, catalyzed by a diarylprolinol
ether; the adducts were obtained in excellent yields and with high stereoselectivities;
with other acids, such as TFA, AcOH, or TsOH, only trace amounts of the desired
products were formed.
Fig. 1. Amine-catalysts/acids combinations for aldehyde or ketone additions to nitro olefins, as reported by
various research groups. For details, see accompanying text and references cited therein.
Why do these reactions proceed well in the presence of acid additives? Does the
acid effect reactivity, diastereo- and/or enantioselectivity depending upon the
particular reaction and amine catalyst used? It is a fact that formation of an enamine
from an aldehyde and a sec.-amine is catalyzed by – not too strong – acids. However, to
the best of our knowledge, there have been no detailed mechanistic investigations to
elucidate the role of acid in the context of the entire catalytic cycle involved in the
amine-catalyzed conjugate addition of aldehydes to nitro alkenes.
2.2. Generally Accepted Mechanism. The generally accepted mechanism (Scheme 1)
consists of the formation of an enamine 3 by condensation of the chiral amine 1 and an
aldehyde 2. The enamine 3 adds to the nitro alkene 4 to form the zwitterionic
intermediate 5. Protonation of the nitronate C-atom is followed by hydrolysis of the
iminium ion 6 to yield the Michael product 7 with regeneration of the chiral amine

722
Helvetica Chimica Acta – Vol. 94 (2011)
catalyst 1. Two roles of acid are proposed: 1) acid is essential for enamine formation:
there is acid-catalyzed addition of the amine 1 to the aldehyde group, followed by
likewise acid-catalyzed elimination of H2O (vide infra); 2) acid protonates the
nitronate-anion part of the zwitterionic intermediate 5 to give the iminium ion 6, thus
promoting the conversion to product 7.
Scheme 1. Generally Accepted Catalytic Cycle of the Diphenylprolinol Silyl Ether-Catalyzed Michael
Addition of Aldehydes to Nitro Olefins. For discussions of this catalytic cycle, see [11][12][14][20 – 25];
for [2 þ 2] cycloadditions occurring through zwitterionic intermediates, see Footnotes 5 and 6 below.
3. Results and Discussion. – 3.1. Preliminary Results. During our studies on the
synthesis of (ꢀ)-oseltamivir (11), with the aim of limiting the purification steps, an
extensive optimization study of the Michael addition of 2-(pentan-3-yloxy)acetalde-
hyde (8) to (E)-tert-butyl 3-nitroacrylate (9) was carried out. We found that
ClCH2COOH enhances rate and increases diastereoselectivity of the reaction
(Table 11)) [17][18].
1
)
The nomenclature like/unlike [26] is used here:

Helvetica Chimica Acta – Vol. 94 (2011)
723
Table 1. Beneficial Effect of Added Acids to the Reaction Mixture of 2-(Pentan-3-yloxy)acetaldehyde (8)
and Nitro-acrylate (9)
Entry
1
Additive
pKa
in H2O
Time
[h]
Yielda)
[%]
u/lb)1)
ee [%]
[mol-%]
(u, l)c)
1
2
3
20
20
5
–
–
7.1
2.86
60
24
1
> 99
> 99
> 99
1.7
1.7
6.3
ndd)
nd
96, 87
4-NO2ꢀC6H4OH
ClCH2CO2H
a) Yield of purified product. b) u/l Ratio was determined by NMR analysis of crude product. c) Enan-
tiomer excess (ee) was determined by HPLC analysis on chiral column material. d) nd ¼ Not determined.
3.2. Effect of Acid in the Amine-Catalyzed Reaction of Aldehydes with Nitro
Alkenes. Our investigations of the effects of acid began by screening acid additives in
the reaction of propanal (2a) with b-nitrostyrene (4a). The reactions were carried out
in toluene (1.0m) with 5-mol-% of both 2-[(diphenyl)(trimethylsilyloxy)methyl]pyrro-
lidine (1) and a selection of acids that covered a wide range of pKa values (Table 2).
The results of the screening of acids indicate a direct correlation between the pKa
value of the acid and reaction efficiency. The reaction was complete within 15 min in
the presence of 4-nitrophenol (Entry 10; pKa 7.15) with excellent diastereo- and
enantioselectivity. Strong acids with pKa < 3.0 inhibit the reaction, and a decrease of the
yields and diastereoselectivities is observed in the presence of chloro-, dichloro- and
trichloroacetic acid (Entries 1 – 3; pKa 0.7 – 2.85). These results suggest that the optimal
acid for Michael addition of propanal to b-nitrostyrene would have a pKa value in the
range of 6 – 8. Accordingly, the acid additive does not affect the enantioselectivity, but
increases the reaction rate, allowing the completion of the reaction in a few min compared
to 6 h in the absence of the acid additive (Entry 14).
To evaluate the scope of 4-nitrophenol as additive, we investigated a range of
aldehydes and nitro alkenes in the presence of 5 mol-% of both chiral amine 1 and 4-
nitrophenol (Tables 3 and 4). The additive 4-nitrophenol has a significant effect on the
rate of the Michael addition of propanal (2a) to b-nitrostyrenes 4a – 4e and to aliphatic
nitro alkenes 4f – 4i. The reaction is complete in less than 1 h2), giving access to the
desired Michael adduct without loss of selectivity, as compared to the transformation
carried out in the absence of additive. Nitro alkenes with sterically demanding
2
)
To avoid decrease of the diastereoselectivity, the reaction was quenched as soon as the conversion
was complete.

724
Helvetica Chimica Acta – Vol. 94 (2011)
Table 2. Effect of Various Acids (5 mol-%) in the Michael Addition of Propanal to b-Nitrostyrene (4a),
Catalyzed by Prolinol Ether 1a)
Entry
Acid
pKa
Reaction
time
Yieldc)
[%]
u/ld)
ee ue)
[%]
in H2Ob)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Cl3CCO2H
0.7
1.29
2.85
3.75
3.98
4.19
4.87
5.97
6.21
7.15
9.18
9.89
10.2
–
10 d
11 d
18 h
25
70
85
88
90
90
95
92
88
98
90
89
90
98
2.5
2.5
3
20
6
ndf)
nd
nd
99
99
99
98
98
99
99
99
99
99
99
Cl2CHCO2H
ClCH2CO2H
HCO2H
CH2ClCH2CO2H
PhCO2H
120 min
80 min
75 min
60 min
30 min
15 min
15 min
25 min
90 min
105 min
6 h
7
EtCO2H
17
10
18
15
20
16
18
14
2,4,6-Br3ꢀC6H2OH
2,4,6-Cl3ꢀC6H2OH
4-NO2ꢀC6H4OH
4-ClꢀC6H4OH
PhOH
4-MeꢀC6H4OH
–
a) Propanal (2a; 0.51 mmol), b-nitrostyrene (4a; 0.34 mmol, 100 mol-%), catalyst (0.017 mmol, 5 mol-
%), toluene (0.3 ml). b) pKa Values were exported from www.zirchrom.com/organic.htm and the
references therein. c) Yield of purified product. d) u/l Ratio was determined by NMR analysis of crude
product. e) ee was determined by HPLC analysis on chiral column material. f) nd ¼ Not determined.
substituents in the b-position, such as 3,3-dimethyl-1-nitrobut-1-ene (4j) did not react
even in the presence of acid additive (Entry 10, Table 3).
These results demonstrate that 4-nitrophenol acts as a general additive in the
addition of propanal (2a) to a series of different nitro alkenes, causing rate
accelerations, with retention of excellent stereoselectivity.
We next studied the reaction of b-nitrostyrene (4a) with various Michael donors
(Table 4). Propanal (2a), butanal (2b), isovaleraldehyde (2c), 2-(benzyloxy)acetalde-
hyde (2d) and 3-phenylpropanal (2e) led to the desired g-nitro aldehydes 7.
A clear increase of the rate is observed when reactions are performed in the
presence of 4-nitrophenol (Entries 1 – 5). In the case of aldehyde 2f, with a sterically
hindered CHO group, the reaction proceeded neither without nor with additive
(Entry 6, Table 4).
The beneficial effect of the additive on the reaction rate prompted us to examine
the possibility of catalyst-loading reduction for the addition of propanal (2a) to b-
nitrostyrene (4a) (Table 5). The reaction ran efficiently with 5 mol-% (Entry 1), 3 mol-
% (Entry 2), and even with 1 mol-% of catalyst (Entry 3, Table 5), with slightly
extended reaction times. Only when the amount of catalyst was further reduced to
0.1 mol-% did the reaction almost come to a halt (12% conversion after 7 d, Entry 4). It

Helvetica Chimica Acta – Vol. 94 (2011)
725
Table 3. Michael Reaction of Propanal (2a) with Various Nitro Alkenes with and without 4-Nitrophenol
Additive (5 mol-%)a)
Entry Nitro alkene
R
with 4-nitrophenol
without 4-nitrophenol
Time
Yieldb) eec) drd) Time Yieldb) eec) drd)
[%]
[%]
[%]
[%]
1
2
3
4
5
6
7
8
9
4a
4b
4c
4d
4e
4f
Ph
15 min 98
35 min 95
30 min 95
15 min 90
99
99
97
99
99
98
98
97
99
–
15
14
4
5
15
5
9
2
5
–
6 h
16 h
3 h
1 h
1.5 h 92
12 h
4 h
10 d
7 d
–
98
90
95
92
99
99
95
97
98
97
98
97
99
–
15
10
7
9
15
3
7
1.5
6.5
–
4-BrꢀC6H4
4-MeOꢀC6H4
4-CF3ꢀC6H4
3,4-(OCH2O)C6H3 30 min 95
Ph(CH2)2
tBuO2C
Cyclohexyl
iPr
90 min 92
45 min 90
88
92
43
79
–
4g
4h
4i
22 h
12 h
–
85
90
–
10
4je)
tBu
a) Nitro alkene (0.30 mmol), aldehyde (0.45 mmol), catalyst (0.015 mmol, 5 mol-%), toluene (0.3 ml).
b) Yield of purified product. c) ee was determined by HPLC analysis on chiral column material.
d) Diastereoisomer ratio (dr) was determined by NMR analysis of crude product. e) No product formed
in this case.
Table 4. Michael Reaction of Various Aldehydes with b-Nitrostyrene (4a) with or without 4-Nitrophenol
Additive (5 mol-%)a)
Entry Aldehyde
R
with 4-nitrophenol
without 4-nitrophenol
Time
Yieldb) eec) drd) Time Yieldb) eec) drd)
[%]
[%]
[%]
[%]
1
2
3
4
5
6
2a
2b
2c
Me
Et
iPr
BnO
Bn
Me(Ph)CH
15 min 98
45 min 95
99
96
99
nd
99
–
15
15
14
1
11
–
6 h
28 h
14 d
13 d
18 h
–
98
98
77
50
90
–
99
95
nd
nd
99
–
15
8
15
1
8
–
20 h
55 h
6 h
–
95
65
90
–
2d
2ee)
2ff)
a) Nitro alkene (0.30 mmol), aldehyde (0.45 mmol), catalyst (0.015 mmol, 5 mol-%), toluene (0.3 ml).
b) Yield of purified product. c) ee was determined by HPLC analysis on chiral column material. d) dr was
determined by NMR analysis of crude product. e) Experimental details not described herein. f) No
product formed in this case.

726
Helvetica Chimica Acta – Vol. 94 (2011)
is needless to say that it is a great synthetic advantage when such an asymmetric
Michael addition proceeds in the presence of only 1 mol-% of diphenylprolinol
trimethylsilyl ether 1 within a reasonable period of time (95 min) and with excellent
stereoselectivity.
Table 5. Catalyst Loading in the Michael Reaction of Propanal (2a) with b-Nitrostyrene (4a)a)
b
Entry
Loading
cat. 1 [mol-%]
Reaction
time
Conversionb)
[%]
u/l
)
eec)
[%]
1
2
3
4
5
3
1
15 min
20 min
95 min
7 d
98
98
98
12
15
14
19
23
99
97
99
99
0.1d)
a) 4a (0.33 mmol), 2a (0.50 mmol), 1 (x-mol-%), 4-nitrophenol (0.016 mmol), and toluene (0.3 ml).
b
) Determined by H-NMR spectroscopy. c) ee were determined by HPLC analysis on chiral column
1
material. d) Reaction performed on a 2-mmol scale; 0.5 mol-% of additive was used.
3.3. Studies towards Elucidation of the Reaction Mechanism. To determine the role
of the acid in the Michael reaction, which consists of several steps, we decided to
investigate the effect of acid in individual steps of the sequence.
a) Enamine Formation. The formation of enamine from equimolar amounts of
aldehyde and amine 1 was examined in C6D6. In less than 5 min, ca. 30% of an enamine
3 are detected by 1H-NMR analysis. The amount of enamine present in the mixture is
unchanged with time, indicating that the reaction had reached an equilibrium. When
the reaction was carried out in the presence of molecular sieves, the conversion to
enamine 3 was over 90% in 5 min, indicating that the enamine formation is rapid, and
that this step is not rate-limiting in the catalytic cycle involving an acid additive.
1
The effect of acid additives on enamine formation was also studied by H-NMR
spectroscopy in C6D6, with equimolar amounts of butanal (2b), catalyst 1, and four
different additives: 4-nitrophenol (pKa 7.1), propanoic acid (pKa 4.7), ClCH2COOH
(pKa 2.8) [17][18], and Cl3COOH (pKa 0.8). With the strongest acid, the reaction
scarcely proceeded3). With the weaker acids (ClCH2COOH and propionic acid, or 4-
nitrophenol), the enamine 3a is generated in smaller amounts, with simultaneous
formation of substantial amounts of self-aldolization product 2-ethylhex-2-enal
(Fig. 2). The quantitative formation of this self-aldolization product within 6 h in the
presence of 4-nitrophenol is compatible with the well-known fact that acid catalyzes
enamine formation from an aldehyde and an amine, addition of an enamine to an
3
)
The peaks corresponding to the catalyst are shifted to low field when a strong acid is used,
indicating protonation of the amino group with formation of an ammonium salt.

Helvetica Chimica Acta – Vol. 94 (2011)
727
aldehyde (cf. (ii) in Footnote 8 below), and dehydration of an aldol adduct to an a,b-
unsaturated carbonyl compound4).
Fig. 2. NMR Analysis of the effect of acid on the reaction of butanal (2b) with diphenylprolinol
trimethylsilyl ether (1). An 0.1-mmol amount of each component (2b, 4-nitrophenol, 1) was dissolved in
0.6 ml of C6D6 in an NMR tube at room temperature. In the absence of 4-nitrophenol, ca. 30% enamine
is formed, and the amount of self-aldol condensation product slowly increases. In the presence of 4-
nitrophenol, the small amount of enamine formed initially disappears quickly, and the unsaturated
aldehyde is formed quantitatively.
b) Discovery of Cyclobutane Formation. Next, the addition of enamine to b-
nitrostyrene was investigated, in the absence of H2O, i.e., not like in the catalytic
reaction where H2O is being formed (cf. enamine formation) and consumed (cf.
iminium-ion hydrolysis; vide supra, Scheme 1). b-Nitrostyrene (4a) was added to a
solution containing a preformed enamine 3 in the presence of molecular sieves. Under
these conditions, the spontaneous and very fast formation of cyclobutanes 12 as single
isomers was observed, the all-trans-configurations of which were deduced from their
NMR analysis (see Fig. 3). With the NMR spectra of stoichiometrically generated
cyclobutanes available, we returned to the reaction under catalytic conditions and
searched for the corresponding signals: in every case studied – by catalytic NMR-tube
experiments – we were able to detect the tiny peaks stemming from the cyclobutane.
The formation and preparation of cyclobutanes from reactions of isolated
enamines, derived from ketones or aldehydes, and alkenes with electron-withdrawing
substituents [27 – 33], e.g., nitro alkenes [27][28][30 – 32], acrylates [27][29], vinyl
4
)
See textbooks of Organic Chemistry.

728
Helvetica Chimica Acta – Vol. 94 (2011)
Fig. 3. Cyclobutanes 12 prepared from preformed enamines 3 and nitroolefins 4, and assignment of the
relative configuration of 12c from NOE in the NMR spectrum. Note that the cyclobutane 12f is formed in
the stoichiometric reaction of the isovaleraldehyde-derived enamine with the tBu-substituted nitroolefin
4j, while there is no catalytic conversion of propanal and 4j to the corresponding nitroaldehyde! Cf.
Table 3, Entry 10.
sulfones [27], and fumarates [27][29][33], has been described a long time ago5)6).
However, to the best of our knowledge, a cyclobutane has not been reported as being
involved in the organocatalytic additions of aldehydes to nitro alkenes or to other
Michael acceptors, nor has its possible role in the corresponding catalytic cycles been
considered.
To find out whether cyclobutanes might be intermediates on the way to the Michael
adducts in the amine-catalyzed reaction of aldehydes with nitro alkenes, cyclobutane
5
)
For masterful discussions by Huisgen of [2 þ 2] cycloadditions via zwitterionic intermediates, see
[34].
The three types of products identified from reactions of enamines with nitro alkenes are a product-
derived enamine, a [2 þ 2] and a [4 þ 2] cycloadduct (see discussion in [9][27][28][30 – 32]):
6
)

Helvetica Chimica Acta – Vol. 94 (2011)
729
formation, and cyclobutane transformation to the products of Michael addition, as well
as the effect of acid in these two reaction steps were first investigated separately.
c) [2 þ 2] Cycloaddition of Enamines 3 to Nitro Alkenes 4. The formation of
cyclobutanes by [2 þ 2] cycloaddition of diphenylprolinol silyl ether-derived enamines
to nitro alkenes was found to be general (Fig. 3): solutions af cyclobutanes 12a – 12f
were prepared from the propanal-, butanal-, or 3-methylbutanal-derived enamines and
b-nitrostyrene, 3-methyl-1-nitrobut-1-ene or 3,3-dimethyl-1-nitrobut-1-ene. The cyclo-
butanes formed from the substrates with more bulky substituents (cf. 12f) are stable
enough for isolation and characterization.
In most cases, cyclobutane formation is a rapid process: in situ NMR studies show
that the preformed enamine derived from propanal (2a) reacts with b-nitrostyrene (4a)
within 5 min to afford the cyclobutane 12a almost quantitatively. When the enamine
and the nitro alkene carry bulky substituents, however, the cyclobutane formation is
1
slow and can be monitored by H-NMR analysis. Thus, we chose the reaction of the
enamine (3b) derived from 3-methylbutanal with 3-methyl-1-nitrobut-1-ene (4i; !
12e) for a more detailed investigation. Without acid, the reaction was slow, affording
the cyclobutane 12e to the extent of 73% conversion after 80 min at room temperature.
On the other hand, 90% of product 12e was obtained within 12 min in the presence of 1
equiv. of 4-nitrophenol (Fig. 4). These results clearly show that the acid accelerates the
cyclobutane formation.
Fig. 4. Formation cyclobutane 12e without and with 4-nitrophenol. The reaction was carried out in an
NMR tube; an equimolar amount of 4-nitrophenol was added to a 1:1 mixture of preformed enamine
and the nitro alkene.

730
Helvetica Chimica Acta – Vol. 94 (2011)
d) From Cyclobutane to the Michael Adduct. We first determined the thermal
1
stability of an amino-nitro-cyclobutane by H-NMR spectroscopy and chose 12c as a
representative example, varying the temperature from 30 to 708 in C6D6. At room
temperature, a constant 95 :5 ratio of cyclobutane 12c and enamine plus b-nitrostyrene,
12c/(3b þ 4a), was observed. The ratio changed gradually from 95 :5 to 60 :40 when the
reaction mixture was heated to 508 within 1 min, followed by heating to 708 within
1 min and keeping the probe at 708 for 7 min. When the solution was allowed to cool to
room temperature, the ratio 12c/(3b þ 4a) returned to 95 :5 (Fig. 5). Thus, the
cyclobutane 12c was in equilibrium with its precursors, and it prevailed at room
temperature in benzene. For previous observations of equilibration between amino-
nitro-cyclobutanes and their precursors, see [31], and for (E/Z)-equilibrations in
mixtures of enamines and nitroolefins, see [9b].
Fig. 5. A thermal equilibration of an amino-nitro-cyclobutane with its precursors enamine and nitro
alkene. After mixing the aldehyde 2c and prolinol ether 1 (in C6D6, in the presence of molecular sieves)
to form the enamine 3b, an equimolar amount of b-nitrostyrene (4a) was added at room temperature;
there was 95% conversion to the cyclobutane 12c; heating in an NMR tube decreased the percentage of
cyclobutane; the effect of temperature was completely reversible.
We next studied the reaction of the cyclobutane 12b with H2O to see whether this
would lead to the Michael adduct; addition of an equimolar amount of H2O in C6D6 led

Helvetica Chimica Acta – Vol. 94 (2011)
731
to two compounds, the enamine 13a and the Michael adduct 7k (Fig. 6). The new
enamine7) is the product of deprotonation of the intermediate iminium ion of type 6
(Scheme 1) and, by way of hydrolysis, a possible precursor to the product 7k of the
overall Michael addition. This result is compatible with capture of the zwitterion of type
5, formed upon cyclobutane-ring opening, by H2O, whereby OHꢀ would either add as a
nucleophile to the iminium C-atom (! 7k) or acts as a base to deprotonate the iminium
ion (!13a)8). The result of the hydrolysis experiment shows that cyclobutane 12b is
converted to the Michael adduct 7k in the presence of H2O9).
The effect of added acid in this reaction of a cyclobutane with H2O was next
examined, using the cyclobutane 12e with two bulky i-Pr substituents as an example,
since the reaction is slow in this case, and can be easily monitored. Without acid,
cyclobutane 12e hardly reacted with H2O: the starting material was recovered after
12 h. However, the Michael adduct 7l was obtained with 80% conversion in the same
period of time in the presence of 1 equiv. of 4-nitrophenol (Fig. 7). In this case, the
corresponding enamine 13b was not detected. This indicates that acid accelerates the
hydrolytic conversion of cyclobutane 12e to the product 7l of Michael addition. Neither
the zwitterionic intermediate of type 5 nor the iminium-ion intermediate of type 6
1
(Scheme 1) has been detected in any of these H-NMR analyses (vide infra).
The in situ formation of cyclobutane and Michael adduct was studied by using
equimolar amounts of butanal (2b), b-nitrostyrene (4a), and amine 1 in toluene. Three
series of reactions were performed: i) in the absence of additive, ii) in the presence of
1
10 mol-% propanoic acid, iii) in the presence of 10 mol-% 4-nitrophenol. H-NMR
Experiments (with aliquots withdrawn from the reaction mixture and diluted with
C6D6) led to the identification of three compounds: cyclobutane 12b, enamine 13a of
Michael adduct, and Michael adduct 7k (Fig. 8). We realize that the sampling process
for NMR analysis could actually have led to changes of the composition of the reaction
mixture10), but there were clear-cut differences of results between the three series of
7
)
Product-derived enamines such as 13a are commonly observed in the above-mentioned [9][10]
stoichiometric reactions of enamines with nitro alkenes under anhydrous conditions.
We should keep in mind that H2O (pKa ca. 15) is not the ꢁidealꢂ acid to protonate a nitronate anion
(i) and that the cleavage of an a-amino alcohol to a carbonyl compound and an amine is actually
acid catalyzed (ii) (cf. Footnote 4).
8
)
9
)
)
The solubility of H2O in benzene is, of course, very low; the reaction mixture is heterogeneous.
Due to the dilution with C6D6, due to contact with the atmosphere (moisture) during sampling and
diluting, and due to the period of time elapsing between the actual sample withdrawal and the NMR
measurement.
10

732
Helvetica Chimica Acta – Vol. 94 (2011)
Fig. 6. NMR Analysis of the reaction of the cyclobutane 12b with H2O. To a solution of 12b in C6D6, an
equimolar amount of H2O was added, and withdrawn samples were subjected to NMR analysis. The
main product of the sluggish reaction is the enamine 13a derived from the nitro aldehyde 7k and 1. For
previous reports on hydrolytic cleavages of 1-amino-2-nitrocyclobutanes to g-nitro-aldehydes and
-ketones, see [27][28][31].
experiments, which we consider significant in view of the role of the acid in the overall
process.
In absence of additive, the initial major product formed was the cyclobutane 12b
(40% maximum). The emergence of enamine 13a and Michael adduct 7k was slow;
these two compounds showed similar profiles, indicating they were generated from the
same precursor. The curve in Fig. 8, a, corresponding to ꢁenamine 13a þ Michael
product 7kꢂ indicates that the rate of formation of the Michael adduct increases as the
reaction progresses, which could be rationalized by considering two different pathways:
one would be the direct formation of 13a and 7k via Michael addition of butanal (2b) to
b-nitrostyrene (4a) catalyzed by diphenylprolinol silyl ether 1, without intervention of
the four-membered ring, the other one would be the ring-opening reaction of
cyclobutane 12b generated by prior [2 þ 2] cycloaddition (vide infra).
When 10 mol-% of propanoic acid was added (Fig. 8,b), a smaller amount of
cyclobutane 12b was detected, which gradually decreased with time. The rate, at which
enamine 13a plus Michael adduct 7k were formed, was higher than in the absence of the
additive (cf. Fig. 8,a and b); the combined yield of 13a þ 7k was 90% after 3 h.

Helvetica Chimica Acta – Vol. 94 (2011)
733
Fig. 7. Hydrolytic cleavage experiments with the cyclobutane 12e in the absence and in the presence of 4-
nitrophenol additive. With this cyclobutane which carries more bulky substituents than 12b, there was
essentially no reaction in the absence of 4-nitrophenol; in its presence, only the nitro aldehyde 7l was
detected. For previous reports on hydrolytic cleavages of 1-amino-2-nitrocyclobutanes to g-nitro-
aldehydes and -ketones, see [27][28][31].
The reaction in the presence of 4-nitrophenol (Fig. 8, c) as an additive had a
distinctly different profile: the cyclobutane 12b formed initially was immediately and
completely converted to enamine 13a plus Michael adduct 7k. The formation of these
two compounds was fast and reached a stationary stage within 30 min. These results are
an indication that the acid additive plays a role in both the formation of cyclobutane
12b and its conversion to the Michael product 7k (cf. Figs. 4 and 7).
While the reaction (in benzene) of cyclobutane 12b with H2O was shown to be slow
in the absence of an acid additive, affording enamine 13a and Michael adduct 7k in 24%
yield within 4 h (Fig. 6), cyclobutane 12b was formed (in toluene) to the extent of 40%
in the first 10 min after equimolar amounts of aldehyde, nitro alkene, and amine had
been mixed, again without acid additive; the subsequent conversion of the cyclobutane
to enamine 13a and Michael adduct 7k occurred within 3 h (Fig. 8, a); thus, the
conversion of the cyclobutane to open-chain products was much faster in the reaction
mixture (Fig. 8) than that observed with the cyclobutane alone (Fig. 6). To disclose the
origin of this difference, the cyclobutane 12b was treated in C6D6 with an equimolar
amount of H2O and butanal. Under these conditions, the ring opening was indeed faster

734
Helvetica Chimica Acta – Vol. 94 (2011)

Helvetica Chimica Acta – Vol. 94 (2011)
735
Fig. 9. Reaction of cyclobutane 12b with H2O in C6D6 in the presence of butanal. Equimolar amounts of
the three components were used; the reaction was carried out in an NMR tube and followed by peak
integrations. The reaction is complete in 6 h, while it has proceeded only to the extent of 25% after the
same period of time in the absence of butanal (see Fig. 6).
than in the absence of butanal (cf. Fig. 6 with Fig. 9). We could imagine that the hydrate
of butanal is involved in this accelerating effect11).
e) Configurational Stability of Product Nitro Aldehydes 7. The observed decrease of
diastereoselectivity, when the reaction is left ꢁrunningꢂ after completion2), prompted us
to investigate the diastereoisomer ratio of an isolated Michael product 7 in the presence
of i) propanoic acid alone, ii) catalyst amine 1 þ propanoic acid, and iii) amine 1 alone.
The results of experiments carried out with the nitro aldehyde 7k in C6H6 showed
(Fig. 10) that the acid alone does not affect the diastereoisomer ratio (i.e. no
enolization); however, in the presence of the amine 1, and even more so when 1 plus
propanoic acid were present, formation of the enamine 13a (derived from the aldehyde
7k) was observed, and the dr value of the recovered nitro aldehyde 7k decreased
significantly (i.e., epimerization in the a-carbonyl position).
11
)
This hydrate of butanal, C3H7ꢀCH(OH)2, could possibly help to carry H2O into the organic phase
(like in a phase-transfer catalysis; cf. Footnote 9), or, otherwise, have an accelerating effect similar to
that of the weak acid 4-nitrophenol (cf. Scheme 3, below); aldehyde hydrates are known to be more
acidic than alcohols, the pKa being close to that of phenol [35].

736
Helvetica Chimica Acta – Vol. 94 (2011)
Fig. 10. Epimerization, monitored by NMR analysis, of the nitro aldehyde 7k under the influence of the
catalyst amine 1 in C6D6. The configuration is stable in the presence of propanoic acid alone, but
decreases especially rapidly when, in addition, the catalyst amine 1 is also present. All components were
employed in equimolar amounts.
4. Revised Mechanism of the Michael Addition of Aldehydes to Nitro Alkenes
Catalyzed by Diphenylprolinol Silyl Ether. The isolation and characterization of
intermediates in organic reactions is crucial for discussing reaction mechanisms. The
detection of cyclobutane 12 by in situ NMR spectroscopy of a reaction mixture
containing an aldehyde 2, a nitro alkene 4, and the diphenylprolinol silyl ether 1 led us
to an investigation with the aim to find out whether these cyclobutane derivatives are
reactive intermediates or ꢁunproductive, parasitic speciesꢂ12). Each reaction step has
been investigated separately under non-catalytic conditions. As a result of these
investigations, we should like to propose a modified reaction mechanism, which takes
into account some of our observations (see Scheme 2; compare with Scheme 1).
a) The Modified Mechanism. The catalyst 1 reacts with the aldehyde 2 to afford the
enamine 3, which reacts with the nitro alkene 4 to generate the zwitterionic intermediate
5; two pathways are possible from 5: 1) cyclization to a cyclobutane derivative 12, and 2)
protonation leading to an iminium ion 6, which, in turn, is either hydrolyzed to the
product 7 of overall-Michael addition – with regeneration of the catalyst 1 – or
deprotonated to an enamine 13 of the Michael adduct; the cyclobutane 12 undergoes ring
opening back to the zwitterion 5, and thus provides access to the product-forming route,
but also to an equilibrium with its precursors, the enamine 3 and the nitro alkene 4;
12
)
The term ꢁparasiticꢂ was first used in connection with organocatalysis by List et al. [36], when they
detected – by in-situ NMR measurements – oxazolidinones in proline-catalyzed aldol additions; for
recent discussions of the role of oxazolidinones in proline catalysis, see [37 – 45].

Helvetica Chimica Acta – Vol. 94 (2011)
737
Scheme 2. Revised Mechanism of the Amine-Catalyzed Michael Addition of Aldehydes to Nitro Olefins
(cf. Scheme 1). The cyclobutanes 12 are off-cycle species, in which the catalyst is taken out of the catalytic
ꢁtrafficꢂ. With the nitrostyrene/3-methylbutanal-derived cyclobutane 12c, reversibility of the [2 þ 2]
cycloaddition was observed (cf. Fig. 5). For the observations leading to inclusion of the nitro aldehyde-
derived enamine 13 in this scheme and of the epimerization step 7 ! epi-7, see accompanying text.
finally, the product nitro aldehyde 7 can be epimerized in the a-carbonyl position by the
catalyst amine 1.
b) Comments on Particular Steps. i) The enamine formation is very fast in the
presence of a weak acid such as 4-nitrophenol. ii) In the catalytic reaction mixture, a
cyclobutane 12 is formed in the first minutes of reaction; in the stoichiometric reaction
under anhydrous conditions, there is complete formation of single isomers of
cyclobutanes 12 from enamines 3 and nitro alkenes 4; this step is especially fast in
the presence of an equivalent amount of 4-nitrophenol. iii) As documented by one
example, amino-nitro-cyclobutanes 12 can be in thermal equilibrium with the
corresponding enamines 3 and nitro alkenes 4, in a reversible [2 þ 2] cycloaddition.
iv) The transformation of cyclobutanes 12 into the corresponding Michael adducts 7 by
treatment with equimolar amounts of H2O in C6D6 was studied in two cases; it is a
process that is also accelerated by 4-nitrophenol13). v) The two reactive intermediates 5
13
)
... and, surprisingly, also by butanal (see Footnote 11)!

738
Helvetica Chimica Acta – Vol. 94 (2011)
1
and 6 were not detected by H-NMR spectroscopy, neither in the catalytic nor in
stoichiometric version of the reaction (i.e., they are short-lived species on the NMR
time scale). vi) As shown in one case, a product 7 of Michael addition is configura-
tionally stable in the presence of acid but may epimerize when catalyst base 1 is present.
c) Discussion of the Role of Acid Additives. When reactions were performed in the
presence of acid additives, several effects of the additive in various steps have been
identified – under catalytic and non-catalytic conditions. Most of the experiments were
performed with 4-nitrophenol (pKa in H2O ca. 7).
1) Acid accelerates the generation of the enamine 3, a well-known effect (cf. the
self-aldolization product, which is formed rapidly in the absence of nitro alkene, Fig. 2).
2) Acid accelerates the addition of enamine 3 to nitro olefin 4. This is demonstrated
by the increase of the rate of the [2 þ 2] cycloaddition in the presence of an equimolar
amount of 4-nitrophenol (Fig. 4). The low pKa value of a protonated nitro group (pKa
ca. ꢀ 12) [46] rules out the possibility of protonation of the nitro alkene NO2 group
and, thus, of activation of the nitro alkene, as 4-nitrophenol is too weak an acid (pKa ca.
7). The accelerating effect of 4-nitrophenol in this step might be explained by its ability
to interact with an O-atom of the developing nitronate anion in the transition state A of
15
C,C-bond formation (3 þ 4 > 5; Scheme 3)14
) )
16). Thus, decrease in the transition-
state energy would cause the acceleration by nitrophenol, and not a ground-state
activation of the nitro alkene.
3) The acid additive accelerates the conversion of the cyclobutanes 12 with H2O to
the products of Michael addition, the nitro aldehyde of type 7 and its enamine 13
(compare Figs. 6 and 8, a, with Figs. 7 and 8, c). As outlined in Scheme 3, the ring
opening to the zwitterion, 12 ! 5, could profit from interaction with the acid 4-
nitrophenol (see transition state B17)); furthermore, the zwitterion must be protonated
on the nitronate moiety in the course of the product-forming process.
4) An acid additive does actually not only affect the rate but also the
diastereoselectivity of the reaction (see Tables 1 – 4, and Figs. 6 and 8), while the
enantioselectivity is not affected. The enantiomer-differentiating step is the C,C-bond
formation between the b-C-atoms of enamine and nitro alkene (see the trajectory C
14
)
We expect this kind of H-bonding to be strongest when the pKa values of the species involved are
similar: the pKa value of nitromethane (HꢀCH2ꢀNO2) is ca. 10 and the pKa value of the tautomeric
nitronic acid (H2C¼NO2ꢀH), the aci-nitro form of nitromethane, is ca. 4.5 [46]; intriguingly, the pKa
of 7 of what turned out to be the best acid (4-nitrophenol) for the process discussed herein is almost
exactly in the middle between these two values.
The arrangement A in Scheme 3 depicts a transition state of a termolecular reaction, which might
occur under the stoichiometric conditions of our experiments (Fig. 4), but may not be possible
under the catalytic conditions employing as little as 5 mol-% of 4-nitrophenol. The transition state
B, on the other hand, corresponds to a bimolecular reaction.
15
)
16
17
)
)
This kind of effect of an acid additive has been discussed by Aggarwal et al. for the BaylisꢀHillman
reaction [47].
The drawing B in Scheme 3 may also be considered as a rationale for the observed stabilization of
the cyclobutanes 12 by bulky R1 and R2 groups. To bring the heteroatom substituents in close
proximity (to reduce charge separation in the transition state, an attractive effect), the four-
membered ring has to attain a so-called roof conformation, which simultaneously leads to a
repulsive approach of the two R groups. In extreme cases, there appears to be ꢁno way backꢂ from
the cyclobutane (cf. Entry 10 in Table 3 and isolation of 4f (Fig. 3)).

Helvetica Chimica Acta – Vol. 94 (2011)
739
Scheme 3. Transition States A and B of C,C-Bond Formation and of Cyclobutane Ring Opening,
Respectively, and Intermediate Zwitterion 5. The indicated interaction with 4-NO2ꢀC6H4OH is
compatible with the observed catalysis by this additive (under stoichiometric conditions). Only when
protonated atꢀCH¼NOꢀ2 does the zwitterion eventually lead to product and catalyst ꢁrecoveryꢂ. In C, the
trajectory of approach of the two trigonal centers is shown (cf. A and the discussions in [9]).
originally proposed by Seebach and Golinski [9] in Scheme 3). This process occurs from
the Si-face of the nucleophilic s-trans-enamine C-atom, since the bulky (Me3SiO)Ph2C
moiety covers the Re-face, securing essentially exclusive Si,Si-coupling of the two
trigonal centers, and thus excellent enantioselectivity. After the C,C-bond formation,
the resulting zwitterionic species undergoes cyclization to the cyclobutane in a fast,
intramolecular step (cf. the formation of the cyclobutanes 12 as single stereoisomers).
The zwitterion can also be protonated (!6), followed by hydrolysis (! 7), or undergo
proton shift to the product-derived enamine (!13). Whenever this enamine forms,
diastereoselectivity is at stake: protonation of the enamine C¼C bond, back to the

740
Helvetica Chimica Acta – Vol. 94 (2011)
iminium ion, can lead to diastereoisomers. The acid additive is expected to suppress
deprotonation of the iminium ion to the enamine.
5) An effect of acid additive has also been detected in the epimerization in the a-
carbonyl position of the final product, the nitro aldehyde 7 (Fig. 10 and Footnote 2).
d) The Importance of Amino-nitro-cyclobutanes 12 and of the Zwitterionic Species 5
(Schemes 2 and 3). The cyclobutane 12 contains the amine catalyst moiety, and it is a
kind of a resting state of the catalytic cycle. Thus, its formation and occurrence decrease
the effective catalyst concentration, ꢁinhibitingꢂ Michael-adduct formation and catalyst
turnover. The cyclobutane is in equilibrium with its precursors, the enamine 3 and the
nitro alkene 4. In this equilibrium, the zwitterion 5 is an intermediate that is constantly
formed and consumed. The ꢁbottle neckꢂ for product formation appears to be the
protonation of the zwitterion 5 to the iminium ion 6, i.e., preventing the zwitterion to
cyclize to cyclobutane and/or the zwitterion dissociation to the olefinic precursors.
Thus, of all the discussed effects of acid, this zwitterion protonation appears to be the
most important one.
4. Conclusions. – Optimization of the asymmetric Michael addition of aldehydes to
nitro alkenes catalyzed by diphenylprolinol silyl ether for application in the synthesis of
biologically important molecules was our original goal. We have demonstrated the
importance of acid in the reaction mixture and identified 4-nitrophenol as a potent
additive for a range of aldehydes and nitro alkenes. In some cases, the reaction times
were shortened by a factor of more than 20, without affecting the excellent enantio- and
diastereoselectivities of the process. Further, the use of catalytic amounts of 4-
nitrophenol allows for reduction of catalyst loading all the way down to 1-mol-% in the
case of the propanal-to-b-nitrostyrene addition.
In situ NMR studies led, for the first time, to the identification of cyclobutanes in
this amine-catalyzed Michael addition of aldehydes to nitro alkenes. The cyclobutanes
may be regarded as ꢁparasiticꢂ components, i.e., an off-cycle resting state of the catalyst,
by which the zwitterion intermediate 5 is removed from the catalytic cycle. The acid
additive has been shown to influence various steps of the overall process, the most
important one of which may be the protonation of the intermediate zwitterion 5, which,
however, our investigation – in the classical organic chemistꢂs way – could not establish.
Much more elaborate kinetic investigations, such as Blackmondꢂs reaction
calorimetry18), and NMR-exchange spectroscopy (EXSY) and real-time NMR studies
of Gschwind and co-workers [39] of the actual catalytic reaction will be necessary to
determine the rate-determining step of these amine-catalyzed Michael additions of
aldehydes to nitro alkenes.
It is likely that other organocatalyzed Michael additions also involve cyclobutanes,
formed from enamines and olefins with electron-withdrawing substituents, and that
there may be cases in which the cyclobutanes are so stable [27][29][33] that their
formation actually prevents catalysis (cf. Entry 10 in Table 3 and caption of Fig. 3).
18
)
Blackmond and co-workers have studied the reaction of propanal with b-nitrostyrene under
diphenylprolinol silyl ether catalysis independently, and their results will be published elsewhere
[48].

Helvetica Chimica Acta – Vol. 94 (2011)
741
The results presented herein are yet another demonstration that a ꢁcloser lookꢂ at
organocatalyzed reactions can provide reasons for revisions of the generally accepted
mechanisms19).
We thank D. Blackmond for a critical discussion of our experimental results and of the conclusions to
be drawn for the catalytic cycle of this Michael addition. Financial support by the Japanese Society for the
Promotion of Science (PE 10021) to enable K. P.-K.ꢂs work in the Hayashi group (May – December 2010)
is gratefully acknowledged. We thank Albert K. Beck for his invaluable help in preparing the manuscript.
Experimental Part
1. General. Abbreviations. MS: molecular sieves, TsOH: p-toluenesulfonic acid, TFA: trifluoroacetic
acid. Aldehydes were purchased from TCI, Japan; nitro alkenes and diphenylprolinol trimethylsilyl ether
were prepared in house. All reactions were carried out under Ar and monitored by TLC with Merck 60
F254 precoated silica-gel plates (0.25 mm thickness). Prep. TLC: Wakogel B-5F purchased from Wako
Pure Chemical Industries, Tokyo, Japan. Flash chromatography (FC): silica gel 60N of Kanto Chemical
Co. Int., Tokyo, Japan. Specific optical rotations: JASCO P-1020 polarimeter. FT-IR Spectra: Horiba FT-
1
720 spectrometer. H- and 13C-NMR spectra: Brucker AM-400 (400 MHz) instrument; chemical shifts
(d) in ppm rel. to internal standard Me4Si; coupling constants J in Hz; assignments on a routine basis by a
combination of 1D and 2D experiments (COSY, HSCQ, HMBC). High-resolution MS: Bruker ESI-
TOF-MS.
2. General Procedure (GP) for Michael Addition of Aldehydes to Nitro Alkenes. To a mixture of nitro
alkene 4 (0.3 mmol) and aldehyde 2 (0.45 mmol) in dry toluene (c(nitro alkene) ¼ 1.0m) was added
diphenylprolinol trimethylsilyl ether 1 (0.015 mmol, 5 mol-%) and, if applicable, an additive
(0.015 mmol, 5 mol-%), and the mixture was stirred at r.t., followed by TLC. After completion, the
reaction was quenched by the addition of aq. 1m HCl. The org. material was extracted with AcOEt (3ꢁ).
The combined org. layers were dried (Na2SO4), filtered, concentrated, and purified by prep. TLC
(hexane/AcOEt 6 :1) to afford the pure Michael adduct.
(2R,3S)-2-Methyl-4-nitro-3-phenylbutanal (7a). Prepared from propanal (2a) and b-nitrostyrene
(¼(2-nitroethenyl)benzene; 4a) according to the GP. NMR Data correspond to those published in [12].
1H-NMR (400 MHz, CDCl3): 9.76 (d, J ¼ 1.2, 1 H); 7.32 – 7.41 (m, 3 H); 7.22 (d, J ¼ 6.8, 2 H); 4.85 (dd, J ¼
5.6, 13.2, 1 H); 4.74 (dd, J ¼ 10.0, 12.8, 1 H); 3.85 (ddd, J ¼ 5.6, 9.6, 9.6, 1 H); 2.83 (m, 1 H); 1.05 (d, J ¼
6.8, 2 H). The enantiomeric excess (ee) was determined by HPLC with Chiralpak OD-H column
(hexane/iPrOH 10 :1; 258; 1 ml minꢀ1; 212 nm): tR(syn major) 21.1 min, tR(syn minor) 16.5 min.
(2R,3S)-3-(4-Bromophenyl)-2-methyl-4-nitrobutanal (7b). Prepared from 2a and 1-bromo-4-[(E)-2-
nitroethenyl]benzene (4b) according to GP. NMR Data correspond to those published in [14]. 1H-NMR
(400 MHz, CDCl3): 9.68 (s, 1 H); 7.46 (d, J ¼ 8.0, 2 H); 7.05 (d, J ¼ 8.4, 2 H); 4.78 (dd, J ¼ 5.2, 12.8, 1 H);
4.64 (dd, J ¼ 9.6, 12.4, 1 H); 3.80 – 3.75 (m, 1 H); 2.78 – 2.70 (m, 1 H); 0.99 (d, J ¼ 7.2, 3 H). The ee was
determined by HPLC with Chiralpak AD-H column (hexane/iPrOH 20 :1; 258; 1 ml minꢀ1; 250 nm):
tR(syn major) 12.3 min, tR(syn minor) 9.6 min.
(2R,3S)-3-(4-Methoxyphenyl)-2-methyl-4-nitrobutanal (7c). Prepared from 2a and 1-methoxy-4-
[(E)-2-nitroethenyl]benzene (4c) according to the GP. NMR Data correspond to those published in [51].
1H-NMR (400 MHz, CDCl3): 9.68 (d, J ¼ 1.6, 1 H); 7.07 (d, J ¼ 8.4, 2 H); 6.85 (d, J ¼ 8.8, 2 H); 4.75 (dd,
J ¼ 5.2, 12.8, 1 H); 4.62 (dd, J ¼ 9.2, 12.4, 1 H); 3.78 – 3.70 (m, 1 H); 3.77 (s, 3 H); 2.75 – 2.68 (m, 1 H);
0.98 (d, J ¼ 7.2, 3 H). The ee was determined by HPLC with Chiralpak AS-H column (hexane/iPrOH
20 :1; 258; 1 ml minꢀ1; 231 nm): tR(syn major) 24.5 min, tR(syn minor) 18.9 min.
(2R,3S)-2-Methyl-4-nitro-3-[4-(trifluoromethyl)phenyl]butanal (7d). Prepared from 2a and 1-[(E)-
1
2-nitroethenyl]-4-(trifluoromethyl)benzene (4d) according to the GP. H-NMR (400 MHz, CDCl3): 9.69
19
)
See also the proline catalysis referred to in Footnote 12, above, and the organocatalysis involving
iminium ions derived from diarylprolinol ethers and from imidazolidinones as reactive
intermediates [49][50].

742
Helvetica Chimica Acta – Vol. 94 (2011)
(s, 1 H); 7.60 (d, J ¼ 7.6, 2 H); 7.32 (d, J ¼ 7.6, 2 H); 4.84 (dd, J ¼ 5.6, 13.6, 1 H); 4.70 (dd, J ¼ 10.0, 12.8,
1 H); 3.94 – 3.86 (m, 1 H); 2.84 – 2.77 (m, 1 H); 1.85 – 1.81 (m, 1 H); 1.00 (d, J ¼ 7.2, 3 H). 13C-NMR
(100 MHz, CDCl3): 202.5; 141.1; 130.8; 128.8; 126.2; 101.6; 77.8; 48.3; 43.9; 12.4. HR-MS: 298.0653
(C12H12F3NNaOþ3 ; calc. 298.0661). The ee was determined by HPLC with Chiralpak AD-H column
(hexane/iPrOH 20 :1; 258; 1 ml minꢀ1; 227 nm): tR(syn major) 10.4 min, tR(syn minor) 14.0 min.
(2R,3S)-3-(Benzo[d][1,3]dioxol-5-yl)-2-methyl-4-nitrobutanal (7e). Prepared from 2a and [(E)-2-
nitroethenyl]benzo[d][1,3]dioxole (4e) according to the GP. Colorless oil. 1H-NMR (400 MHz, CDCl3):
9.69 (d, J ¼ 1.6, 1 H); 6.61 – 6.76 (m, 4 H); 5.95 (s, 2 H); 4.74 (dd, J ¼ 6.8, 12.0, 1 H); 4.60 (dd, J ¼ 9.6, 12.8,
1 H); 3.71 (m, 1 H); 2.70 (m, 1 H); 1.01 (d, J ¼ 7.2, 3 H). 13C-NMR (100 MHz, CDCl3): 202.5; 148.6;
130.7; 122.0; 109.0; 108.6; 108.3; 101.6; 78.6; 48.9; 44.2; 12.5. HR-MS: 274.668 (C12H13NNaOþ5 ; calc.
274.0686). The ee was determined by HPLC with Chiralpak AS-H column (hexane/iPrOH 10 :1; 258;
1 ml minꢀ1; 206 nm): tR(syn major) 26.2 min, tR(syn minor) 20.6 min.
(2R,3R)-2-Methyl-3-(nitromethyl)-5-phenylpentanal (7f). Prepared from 2a and [(E)-(4-nitrobut-3-
en-1-yl]benzene (4f) according to GP. NMR Data correspond to those published in [52]. 1H-NMR
(400 MHz, CDCl3, 258): 9.68 (s, 1 H); 7.20 – 7.37 (m, 5 H); 4.60 (dd, J ¼ 6.0, 12.4, 1 H); 4.51 (dd, J ¼ 7.6,
12.4, 1 H); 2.90 – 2.79 (m, 1 H); 2.79 – 2.61 (m, 3 H); 1.79 – 1.65 (m, 2 H); 1.22 (d, J ¼ 7.2, 3 H). The ee
was determined by HPLC with Chiralpak OJ-H column (hexane/iPrOH 10 :1; 258; 1 ml minꢀ1; 209 nm):
tR(syn major) 28.1 min, tR(syn minor) 35.6 min.
tert-Butyl (2R,3R)-3-Methyl-2-(nitromethyl)-4-oxobutanoate (7g). Prepared from 2a and tert-butyl
3-nitroprop-2-enoate (4g) according to GP. 1H-NMR (400 MHz, CDCl3): 9.68 (s, 1 H); 4.85 (dd, J ¼ 8.8,
14.4, 1 H); 4.41 (dd, J ¼ 5.6, 14.4, 1 H); 3.69 (ddd, J ¼ 5.2, 8.8, 10.4, 1 H); 2.71 – 2.66 (m, 1 H); 1.43 (s,
3 H); 0.98 (d, J ¼ 7.4, 3 H). 13C-NMR (100 MHz, CDCl3): 200.6; 169.0; 126.3; 115.9; 83.6; 73.5; 45.7; 43.9;
27.9; 9.9. HR-MS: 254.0974 (C10H17NNaOþ5 ; calc. 254.0999). The ee was determined by HPLC with
Chiralpak OD-H column (hexane/iPrOH 100 :1; 258; 1 ml minꢀ1; 216 nm): tR(syn major) 29.5 min, tR(syn
minor) 31.0 min.
(2R,3R)-3-Cyclohexyl-2-methyl-4-nitrobutanal (7h). Prepared from 2a and [(E)-2-nitroethenyl]cy-
clohexane (4h) according to GP. NMR Data correspond to those published in [51]. 1H-NMR (400 MHz,
CDCl3): 9.68 (s, 1 H); 4.60 (dd, J ¼ 5.6, 13.2, 1 H); 4.42 (dd, J ¼ 6.7, 13.4, 1 H); 2.53 – 2.62 (m, 2 H); 1.53 –
1.87 (m, 5 H); 1.53 – 1.40 (m, 1 H); 1.25 – 0.84 (m, 5 H); 1.22 (d, J ¼ 7.5, 3 H). The ee was determined by
HPLC with Chiralpak AS-H column (hexane/iPrOH 40 :1; 258; 1 ml minꢀ1; 209 nm): tR(syn major)
9.2 min, tR(syn minor) 8.7 min.
2,4-Dimethyl-3-(nitromethyl)pentanal (7i). Prepared from 2a and (E)-3-methyl-1-nitrobut-1-ene (4i)
according to GP. NMR Data correspond to those published in [53]. 1H-NMR (400 MHz, CDCl3): 9.76 (d,
J ¼ 1.2, 1 H); 4.85 (dd, J ¼ 5.6, 13.2, 1 H); 4.72 (dd, J ¼ 10.0, 12.8, 1 H); 3.90 – 3.84 (m, 1 H); 2.85 – 2.81
(m, 1 H); 1.21 (d, J ¼ 6.8, 3 H); 0.96 (d, J ¼ 6.8, 3 H); 0.94 (d, J ¼ 6.8, 3 H). The ee was determined by
HPLC with Chiralpak AD-H column (hexane/iPrOH 200 :1; 258; 1 ml minꢀ1; 213 nm): tR(syn major)
21.9 min, tR(syn minor) 23.8 min.
(2R,3S)-2-Ethyl-4-nitro-3-phenylbutanal (7k). Prepared from butanal (2b) and 4a according to GP.
1
NMR Data correspond to those published in [12]. H-NMR (400 MHz, CDCl3): 9.78 (d, J ¼ 2.4, 1 H);
7.33 – 7.43 (m, 3 H); 7.25 (d, J ¼ 7.2, 2 H); 4.78 (dd, J ¼ 7.6, 12.4, 1 H); 4.70 (dd, J ¼ 9.6, 12.8, 1 H); 3.87
(ddd, J ¼ 4.8, 9.6, 9.6, 1 H); 2.72 – 2.78 (m, 1 H); 1.57 – 1.60 (m, 2 H); 0.90 (t, J ¼ 7.2, 3 H). The ee was
determined by HPLC with Chiralpak OD-H column (hexane/iPrOH 20 :1; 258; 1 ml minꢀ1; 220 nm);
tR(syn major) 23.7 min, tR(syn minor) 19.6 min.
(2R,3S)-2-Isopropyl-4-nitro-3-phenylbutanal (7l). Prepared from isovalerylaldehyde (¼ 3-methyl-
butanal; 2c) and 4a according to GP. NMR Data correspond to those published in [12]. 1H-NMR
(400 MHz, CDCl3): 10.03 (d, J ¼ 2.4, 1 H); 7.47 – 7.37 (m, 3 H); 7.31 – 7.29 (m, 2 H); 4.77 (dd, J ¼ 4.4,
12.8, 1 H); 4.68 (dd, J ¼ 10.0, 12.8, 1 H); 4.01 (ddd, J ¼ 4.4, 10.4, 10.2, 1 H); 2.90 – 2.87 (ddd, J ¼ 2.4, 4.1,
10.7, 1 H); 1.85 – 1.81 (m, 1 H); 1.20 (d, J ¼ 7.2, 3 H); 0.99 (d, J ¼ 6.8, 3 H). The ee was determined by
HPLC with Chiralpak AD-H column (hexane/iPrOH 20 :1; 258; 1 ml minꢀ1; 217 nm): tR(syn major)
7.1 min, tR(syn minor) 6.7 min.
(2R,3S)-2-Benzyl-4-nitro-3-phenylbutanal (7n). Prepared from 3-phenylpropanal (2e) and 4a
1
according to GP. NMR Data correspond to those published in [54]. H-NMR (400 MHz, CDCl3): 9.88
(d, J ¼ 2.4, 1 H); 7.57 – 7.29 (m, 8 H); 7.20 (d, J ¼ 7.2, 2 H); 5.05 (dd, J ¼ 6.4, 13.2, 1 H); 4.97 (dd, J ¼ 8.8,

Helvetica Chimica Acta – Vol. 94 (2011)
743
12.8, 1 H); 4.0 (dd, J ¼ 6.0, 8.4, 1 H); 3.27 (ddd, J ¼ 6.0, 8.4, 14.8, 1 H); 2.93 (d, J ¼ 5.6, 2 H). The ee was
determined by HPLC with Chiralpak OD-H column (hexane/iPrOH 10 :1; 258; 1 ml minꢀ1; 220 nm):
tR(syn major) 32.7 min, tR(syn minor) 30.4 min.
4. Preparation of Cyclobutane 12. (2S)-1-[(1S,2S,3S,4R)-3-(tert-Butyl)-2-nitro-4-(propan-2-yl)cy-
clobutyl]-2-{diphenyl[(trimethylsilyl)oxy]methyl}pyrrolidine (12f). To a mixture of 1 (97.5 mg,
0.3 mmol) and 2c (33 ml, 0.3 mmol) in benzene (0.3 ml) with MS (4 ꢄ) was added (E)-3,3-dimethyl-1-
nitrobut-1-ene (4j; 38.75 mg, 0.3 mmol), and the mixture was left to react for 48 h. The material was
purified by prep. TLC (hexane/AcOEt, 8 :1) to afford 12f (34%). Yellow solid. [a]2D5 ¼ ꢀ6.7 (c ¼ 19.53,
CHCl3). IR (CHCl3): 3056w, 2959s, 2871w, 2815w, 2359m, 2337m, 1729w, 1537s, 1468w, 1367m, 1249s,
1092m, 1064s, 877m, 840s, 704s. 1H-NMR (400 MHz, C6D6): 7.55 – 7.49 (m, 2 H); 7.45 – 7.40 (m, 2 H);
7.33 – 7.25 (m, 4 H); 4.86 (dd, J ¼ 7.2, 8.0, 1 H); 4.06 (dd, J ¼ 2.8, 9.2, 1 H); 3.90 (m, 1 H); 2.62 – 2.54 (m,
1 H); 2.54 – 2.47 (m, 1 H); 2.20 (t, J ¼ 9.2, 1 H); 1.94 – 1.72 (m, 3 H); 1.71 – 1.59 (m, 1 H); 1.24 (m, 1 H);
0.84 (s, 9 H); 0.78 (d, 3 H, J ¼ 8.0); 0.67 (br. d, 3 H); 0.325 (m, 1 H); 0.28 (s, 9 H). 13C-NMR (100 MHz,
CDCl3): 143.6; 142.9; 130.32; 130.14; 128.6; 78.8; 69.5; 62.7; 49.4; 47.5; 42.3; 31.86; 29.4; 29.1; 23.8; 22.3;
2.23. HR-MS: 521.3160 ([M ꢀ H]ꢀ, C31H45N2O3Siꢀ; calc. 521.32).
(2S)-2-{Diphenyl[(trimethylsilyl]oxy]methyl}-1-[(1S,2S,3S,4R)-2-nitro-3-phenyl-4-(propan-2-yl)cy-
clobutyl]pyrrolidine (12c). To a mixture of 1 (32.5 mg, 0.1 mmol, 1 equiv.) and 2c (10.8 ml, 0.1 mmol,
1 equiv.) in benzene with MS (4 ꢄ) was added 4a (38.75 mg, 0.3 mmol, 1 equiv.), and the mixture was
stirred at r.t. for 2 h. The crude material was analyzed by NMR; the purification failed due to
decomposition towards Michael product. 1H-NMR (400 MHz, C6D6): 7.67 – 7.61 (m, 4 H); 7.22 – 6.95 (m,
11 H); 4.96 (t, J ¼ 7.6, 1 H); 4.50 (dd, J ¼ 4.4, 8.4, 1 H); 4.26 (br. m, 1 H); 3.46 (t, J ¼ 9.2, 1 H); 2.46 – 2.53
(m, 2 H); 1.90 – 1.85 (m, 2 H); 1.89 (m, 1 H); 1.67 – 1.56 (m, 1 H); 1.16 (m, 1 H); 0.89 (br. d, 3 H); 0.65 (d,
J ¼ 6.8, 3 H); 0.53 (br. m, 1 H). 13C-NMR (100 MHz, C6D6): 140.69; 140.36; 137.5; 134.9; 128.4; 127.2;
127.0; 126.0; 125.97; 124.3; 81.9; 79.97; 65.75; 63.27; 46.34; 45.01; 40.89; 26.23; 20.90; 17.25; ꢀ 0.85.
5. Catalyst Loading. To a mixture of 4a (50 mg, 0.3 mmol, 1 equiv.) and 2a (36 ml, 0.45 mmol,
1.5 equiv.) in dry toluene (c(4a) ¼ 1.0m) were added 1 (x mol-%) and 4-nitrophenol (x mol-%) at r.t. The
1
reaction was quenched according to GP. The product conversion was determined by H-NMR of the
crude material (cf. Table 5).
6. Studies towards the Elucidation of the Mechanism. Enamine Formation. Five experiments were
performed in 1H-NMR tubes with 1 (0.1 mmol, 1m stock soln. in C6D6) and 2b (0.1 mmol, 1m stock soln. in
C6D6) in 600 ml of C6D6.
Cyclobutane Formation. These experiments were performed using preformed enamine (from
equimolar amounts of 1 and 2c in C6D6 with MS (4 ꢄ)) and 4i in NMR tube. One equiv. of acid (if
applicable) was added to the NMR tube.
Thermal Stability. To a mixture of preformed enamine from 1 (32.5 mg, 0.1 mmol, 1 equiv.) and 2c
(10.8 ml, 0.1 mmol, 1 equiv.) in the presence of MS (4 ꢄ) in 1 ml of C6D6 was added 4a (14.9 mg, 0.1 mmol,
1 equiv.) to form 12c. From this reaction mixture, 0.6 ml were transferred into an NMR tube, and
1H-NMR recordings were performed at the given temp.: starting from 258, heating up to 508, and then
708, followed by cooling down to 258.
Cyclobutane Hydrolysis. Effect of H2O and Butanal (2b). These experiments were performed using
preformed enamine (from equimolar amounts of 1 and 2b in C6D6 with MS (4 ꢄ)) and 4a. MS were
removed, 1 equiv. of H2O and, if applicable, 1 equiv. of 2b were added. Aliquots (20 ml) were taken at
time T and diluted in 500 ml C6D6 for 1H-NMR analysis.
Effect of H2O and Acid Additive. These experiments were performed using preformed enamine
(from equimolar amounts of 1 and 2c in C6D6 with MS) and 4i. The mixture was transferred to an NMR
tube, and 1 equiv. of H2O and, if applicable, 1 equiv. of 4-nitrophenol were added. The evolution of the
1
cyclobutane hydrolysis was followed by H-NMR.
Observation of Cyclobutane Formation and Ring Opening under Reaction Conditions. b-Nitrostyrene
(4a) (0.2 mmol), 2b (0.2 mmol), and 1 (0.2 mmol) in 200 ml of dry toluene (c ¼ 1.0m). If applicable,
additive (propanoic acid or 4-nitrophenol) was added. Aliquots (20 ml) were taken at time T and diluted
in 500 ml C6D6 for 1H-NMR analysis.
Isomerization of Michael Product. Michael adduct 7k of dr ratio 11:1 (0.3 mmol) was dissolved in
C6D6 (1.8 ml) and distributed into three NMR tubes (0.6 ml each). To one NMR tube, 1 equiv. of

744
Helvetica Chimica Acta – Vol. 94 (2011)
propanoic acid (0.1 mmol) was added, to the second one, 1 equiv. of 1 (0.1 mmol) was added, and to third
one, 1 equiv. of propanoic acid (0.1 mmol) þ 1 equiv. of 1 (0.1 mmol) were added. The dr was determined
by integration of the CHO signals of the two diastereoisomers in the 1H-NMR spectrum.
REFERENCES
[1] P. Perlmutter, ꢁConjugate Addition Reactions in Organic Synthesisꢂ, Oxford Press, Oxford, 1992.
[2] S. Mukherjee, J. W. Yang, S. Hoffmann, B. List, Chem. Rev. 2007, 107, 5471 – 5569.
[3] P. M. Pihko, I. Majander, A. Erkkila, in ꢁAsymmetric Organocatalysisꢂ, Springer-Verlag, Berlin,
2010, pp. 29 – 75.
[4] A. Erkkila, I. Majander, P. M. Pihko, Chem. Rev. 2007, 107, 5416 – 5470.
[5] D. Almasi, D. A. Alonso, C. Najera, Tetrahedron: Asymmetry 2007, 18, 299 – 365.
[6] S. B. Tsogoeva, Eur. J. Org. Chem. 2007, 1701.
´
[7] S. Sulzer-Mosse, A. Alexakis, Chem. Commun. 2007, 3123.
[8] O. M. Berner, L. Tedeschi, D. Enders, Eur. J. Org. Chem. 2002, 1877.
´
´
[9] a) D. Seebach, J. Golinski, Helv. Chim. Acta 1981, 64, 1413; b) D. Seebach, A. K. Beck, J. Golinski,
J. N. Hay, T. Laube, Helv. Chim. Acta 1985, 68, 162.
[10] S. J. Blarer, W. B. Schweizer, D. Seebach, Helv. Chim. Acta 1982, 65, 1637.
[11] B. List, P. Pojarliev, H. J. Martin, Org. Lett. 2001, 3, 2423.
[12] J. M. Betancort, C. F. Barbas, Org. Lett. 2001, 3, 3737.
[13] J. Franzen, M. Marigo, D. Fielenbach, T. C. Wabnitz, K. A. Jørgensen, J. Am. Chem. Soc. 2005, 127,
18296.
[14] Y. Hayashi, H. Gotoh, T. Hayashi, M. Shoji, Angew. Chem. 2005, 117, 4284; Angew. Chem., Int. Ed.
2005, 44, 4212.
[15] C. Palomo, A. Mielgo, Angew. Chem. 2006, 118, 8042; Angew. Chem., Int. Ed. 2006, 45, 7876.
[16] A. Mielgo, C. Palomo, Chem. – Asian J. 2008, 3, 922.
[17] H. Ishikawa, T. Suzuki, Y. Hayashi, Angew. Chem. 2009, 121, 1330; Angew. Chem., Int. Ed. 2009, 48,
1304.
[18] H. Ishikawa, T. Suzuki, H. Orita, T. Uchimaru, Y. Hayashi, Chem. – Eur. J. 2010, 16, 12616.
[19] H. Ishikawa, M. Honma, Y. Hayashi, Angew. Chem. 2011, 123, 2700; Angew. Chem., Int. Ed. 2011,
50, 2824.
[20] A. Alexakis, O. Andrey, Org. Lett. 2002, 4, 3611.
[21] N. Mase, R. Thayumanavan, F. Tanaka, C. F. Barbas, Org. Lett. 2004, 6, 2527.
[22] M. Wiesner, G. Upert, G. Angelici, H. Wennemers, J. Am. Chem. Soc. 2009, 132, 6.
[23] T. Ishii, S. Fujioka, Y. Sekiguchi, H. Kotsuki, J. Am. Chem. Soc. 2004, 126, 9558.
[24] S. Saha, S. Seth, J. N. Moorthy, Tetrahedron Lett. 2010, 51, 5281.
[25] Z. Zheng, B. L. Perkins, B. Ni, J. Am. Chem. Soc. 2010, 132, 50.
[26] D. Seebach, V. Prelog, Angew. Chem. 1982, 94, 696; Angew. Chem., Int. Ed. 1982, 21, 654.
[27] K. C. Brannock, A. Bell, R. D. Burpitt, C. A. Kelly, J. Org. Chem. 1964, 29, 801.
[28] M. E. Kuehne, L. Foley, J. Org. Chem. 1965, 30, 4280.
[29] K. C. Brannock, A. Bell, R. D. Burpitt, C. A. Kelly, J. Org. Chem. 1961, 26, 625.
[30] F. P. Colonna, E. Valentin, G. Pitacco, A. Risaliti, Tetrahedron 1973, 29, 3011.
[31] F. Felluga, P. Nitti, G. Pitacco, E. Valentin, Tetrahedron 1989, 45, 5667.
[32] F. Felluga, P. Nitti, G. Pitacco, E. Valentin, J. Chem. Soc., Perkin Trans. 1 1992, 2331.
[33] F. D. Lewis, T.-I. Ho, R. J. DeVoe, J. Org. Chem. 1980, 45, 5283.
[34] R. Huisgen, Acc. Chem. Res. 1977, 10, 117, 199.
[35] J. Hine, J. G. Houston, J. H. Jensen, J. Org. Chem. 1965, 30, 1184.
[36] B. List, L. Hoang, H. J. Martin, Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5839.
[37] D. Seebach, A. K. Beck, D. M. Badine, M. Limbach, A. Eschenmoser, A. M. Treasurywala, R. Hobi,
W. Prikoszovich, B. Linder, Helv. Chim. Acta 2007, 90, 425.
´
[38] C. Isart, J Bures, J. Vilarrasa, Tetrahedron Lett. 2008, 49, 5414.

Helvetica Chimica Acta – Vol. 94 (2011)
745
[39] M. B. Schmid, K. Zeitler, R. M. Gschwind, Angew. Chem. 2010, 122, 5117; Angew. Chem., Int. Ed.
2010, 49, 4997.
[40] A. K. Sharma, R. B. Sunoj, Angew. Chem. 2010, 122, 6517; Angew. Chem., Int. Ed. 2010, 49, 6373.
[41] T. Kanzian, S. Lakhdar, H. Mayr, Angew. Chem. 2010, 122, 9717; Angew. Chem., Int. Ed. 2010, 49,
9526.
[42] D. Blackmond, A. Moran, M. Hughes, A. Armstrong, J. Am. Chem. Soc. 2010, 132, 7598.
[43] I. A. Konstantinov, L. J. Broadbelt, Top. Catal. 2010, 53, 1031.
[44] P. D. de Maria, P. Bracco, L. F. Castelhano, G. Bargeman, ACS Catal. 2011, 1, 70.
[45] D. A. Bock, C. W. Lehmann, B. List, Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20636.
[46] D. Turnbull, S. H. Maron, J. Am. Chem. Soc. 1943, 65, 212; I. Erden, J. R. Keeffe, F. P. Xu, J. B.
Zheng, J. Am. Chem. Soc. 1993, 115, 9834; T. R. Kelly, M. H. Kim, J. Am. Chem. Soc. 1995, 116, 7072;
B. R. Linton, M. S. Goodman, A. D. Hamilton, Chem. – Eur. J. 2000, 6, 2449, and refs. cit. therein.
[47] V. K. Aggarwal, S. Y. Fulford, G. C. Lloyd-Jones, Angew. Chem. 2005, 117, 1734; Angew. Chem., Int.
Ed. 2005, 44, 1706.
´
[48] Jordi Bures, Alan Armstrong, Donna G. Blackmond, J. Am. Chem. Soc. 2011, submitted.
ˇ
[49] D. Seebach, U. Groselj, D. M. Badine, W. B. Schweizer, A. K. Beck, Helv. Chim. Acta 2008, 91, 1999;
ˇ
ˇ
U. Groselj, W. B. Schweizer, M.-O. Ebert, D. Seebach, Helv. Chim. Acta 2009, 92, 1; U. Groselj, D.
Seebach, D. M. Badine, W. B. Schweizer, A. K. Beck, I. Krossing, P. Klose, Y. Hayashi, T. Uchimaru,
ˇ
Helv. Chim. Acta 2009, 92, 1225; D. Seebach, U. Groselj, W. B. Schweizer, S. Grimme, C. Mꢀck-
Lichtenfeld, Helv. Chim. Acta 2010, 93, 1.
[50] J. B. Brazier, G. Evans, T. J. K. Gibbs, S. J. Coles, M. B. Hursthouse, J. A. Platts, N. C. O. Tompkinson
Org. Lett. 2009, 11, 133.
[51] O. Andrey, A. Alexakis, A. Tomassini, G. Bernardinelli, Adv. Synth. Catal. 2004, 346, 1147.
[52] L. S. Zu, H. Li, J. Wang, X. H. Yu, W. Wang, Tetrahedron Lett. 2006, 47, 5131.
[53] T. Mandal, C. G. Zhao, Angew. Chem. 2008, 120, 7828; Angew. Chem., Int. Ed. 2008, 47, 7714.
[54] P. Kotrusz, S. Toma, H. G. Schmalz, A. Adler, Eur. J. Org. Chem. 2004, 1577.
Received March 21, 2011