Enantioselective 1,3-Dipolar Cycloaddition Reactions of C-Carboxy Ketonitrones and Enals with MacMillan Catalysts: Evidence of a Nonconcerted Mechanism

Article abstract of DOI:10.1002/ejoc.201701307

Highly diastereo- and enantioselective 1,3-dipolar cycloadditions between functional ketonitrones and β-substituted enals are promoted by MacMillan imidazolidinium organocatalysts. A study of the reaction scope shows that high selectivities are conserved if the N-protecting group or the ester function is varied. However, the reaction is sensitive to steric interactions with the C substituent of the nitrone. In all cases, the reaction proceeds with high exo selectivity. In most cases, a third diastereomer, incompatible with a concerted mechanism, was also observed, albeit in minute amounts. DFT calculations suggest that the cycloaddition proceeds in a nonconcerted fashion by an initial oxa-Michael-type addition of the nitrone to the double bond followed by a cyclization. This mechanism explains the formation of the observed minor diastereomers. In addition, the diastereo- and enantioselectivities of the reaction were shown to be intermediately thermodynamically controlled, and the diastereomeric ratio is modulated by the kinetics of iminium hydrolysis.

Full text of DOI:10.1002/ejoc.201701307

A Journal of
Accepted Article
Title: Enantioselective 1,3-DC reactions of C-Carboxy Ketonitrones
and Enals with MacMillan catalyst: Evidence of a non-concerted
mechanism
Authors: Kawther Ben Ayed, Mathieu Yves Laurent, Arnaud Martel,
Khalid Bashir Selim, Souhir Abid, and Gilles Dujardin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201701307
Link to VoR: http://dx.doi.org/10.1002/ejoc.201701307
Supported by

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
Enantioselective 1,3-DC reactions of C-Carboxy Ketonitrones and
Enals with MacMillan catalyst: Evidence of a non-concerted
mechanism
*
Kawther Ben Ayed,[a,b] Mathieu Y. Laurent,[a] Arnaud Martel,[a] Khalid B. Selim,[c] Souhir Abid,[b] Gilles
*
Dujardin[a]
cycloaddition (1,3-DC) reaction between nitrones and ethylenic
dipolarophiles.1,2,3 This method has a particular interest to obtain
polyfunctional quaternary-containing isoxazolidines, and to this
aim, the involvement of functional ketonitrones was recently
developed.4
Abstract: Highly diastereo- and enantioselective 1,3-dipolar
cycloadditions between functional ketonitrones and β-substituted
enals are promoted by organocatalysis with the imidazolidinium
catalyst of MacMillan. Study of the scope of the reaction shows
that high selectivities are conserved by varying the N-protecting
group or the ester function. However it is sensitive to sterical
interaction with the C-substituent of the nitrone. Reaction
proceeds in all cases with a high exo selectivity. In most cases,
In 2012, the enantioselective 1,3-DC reaction of (E)-
crotonaldehyde with N-benzyl C-ethoxycarbonyl ketonitrone 1a
using MacMillan organocatalyst5 I (Figure 1)6,7 was reported.
Good yields (up to 75%) were reached with high exo- and
enantio-selectivities (up to 96:4 dr, 95% ee) using anhydrous
conditions in nitromethane at –10 °C.6b The obtained
polyfunctional isoxazolidine 3aA possesses three contiguous
stereogenic centers, including a quaternary one, and its absolute
configuration was established.6b Such type of isoxazolidine has
a
third diastereomer, not compatible with
a
concerted
mechanism, was observed, although in minute amount. DFT
calculations evidence that the cycloaddition proceeds in a non-
concerted fashion by a first oxa Michael-type addition of the
nitrone to the double bond followed by a cyclization. This
mechanism explains the formation of the observed minor
a
high potential for synthetic applications, owing to the
numerous modifications that can be envisaged, either involving
the aldehyde functionality or based on the chemodivergent N-O
ring opening of the heterocycle. For these reasons and
according to these preliminary and partial results, it seemed
important (i) to study the scope and limitations to reach a
diversity of quaternary-containing 4-formyl isoxazolidines, (ii) to
investigate the mechanism of the organocatalytic process taking
into account all the experimental features. On the basis of the
experimental results and the theoretical calculations, we present
diastereomers.
enantioselectivities of the reaction were shown to be
intermediately thermodynamically controlled and the
In
addition,
the
diastereo-
and
diastereomeric ratio is modulated by the kinetics of iminium
hydrolysis.
here
a
new mechanistical proposal for the MacMillan
organocatalyzed cycloaddition between nitrones and enals
through a non-concerted pathway.
Introduction
Isoxazolidines are pivotal N-O bond-containing 5-membered ring
heterocycles, due to their ability to produce polyfunctional β-
amino acid derivatives (alcohols, aldehydes, esters, amides,
imides) via N-O ring opening reactions.1 These heterocycles are
also important scaffolds for medicinal chemistry as they can act
as natural building block mimetics (nucleosides, carbohydrates,
peptide nucleic acids, amino acids, steroids).1 A main pathway
to reach these valuable 5-membered ring compounds in
diastereo- and enantioselective manner is the 1,3-dipolar
O
N
·HX
Ph
N
H
I, X = Cl
II, X = ClO4
Figure 1: MacMillan catalysts
Results and Discussion
Variation on the Ketonitrone
[a]
IMMM UMR CNRS 6283
Faculté des Sciences-Université du Maine, Le Mans 72085, France
E-mail: gilles.dujardin@univ-lemans.fr, arnaud.martel@univ-
lemans.fr, http://immm.univ-lemans.fr
To study the scope and the limitations of the titled reaction, this
cycloaddition was first extended to different ketonitrones 1
towards (E)-crotonaldehyde 2A.
In a preliminary study, the cycloaddition between C-carboxy
ketonitrones 1b-g and (E)-crotonaldehyde 2A was conducted
under uncatalyzed thermal conditions. As already observed in
the case of 3aA,6b the thermal 1,3-DC reactions proceeded
under microwave activation (Table 1) to afford adducts in high
yields as a mixture of 3 isomers (3/4/5), with a selectivity for the
major isomer exo (3) ranging for 69 to 91% (Table 1).
[b]
[c]
Laboratoire de Chimie Appliquée
Faculté des Sciences - Université de Sfax, Sfax 3000, Tunisia
Department of Pharmaceutical Organic Chemistry
Faculty of Pharmacy-Mansoura University, Mansoura 35516, Egypt
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
Having identified in each case the isomer exo 3 as the major
isomer and the isomer endo 4 as the second isomer by
comparative NMR data, we had to define the nature of the
intriguing third isomer 5 observed in various amounts (1 to 11%
of the isomeric mixture), and characterized by its highly
deshielded (9.89 ppm) aldehydic proton in NMR spectra. This
third adduct 5 could be attributed either to a regioisomer (3,5-
formyl adduct) or to an isomer resulting from the 1,3-DC reaction
of (Z)-crotonaldehyde. Fortunately, the NMR data of 3,5-formyl
adducts (exo and endo) could be both deduced from previous
studies of our group by Ru-catalyzed 1,3-DC reaction between
ketonitrone 1a and (E)-crotonaldehyde 2A.4a The aldehyde
protons of the 3,5-formyl regioisomers are significantly less
deshielded: 9.78 ppm for the exo isomer and 9.69 ppm for the
endo isomer. On the basis of these results, we have assumed
that the third adduct here observed displayed a cis-relationship
On the basis of these results, the organocatalytic asymmetric
version of the 1,3-DC between ketonitrones 1a-g and (E)-
crotonaldehyde 2A was then investigated, using chiral catalyst I
(10 mol %) in dry nitromethane (Table 2).6b
Table 2. Asymmetric organocatalytic 1,3-DC reaction between ketonitrones 1a-1g
and (E)-crotonaldehyde 2A.
CO2R2
CO2R2
R3
CO2R2
R3
R1
CO2R2
R3
R3
O
R1
R1
N
O
R1
CHO
N
O
N
O
Cat. I (10 mol%)
+
+
CHO
CHO
N
O
+
MeNO2
Me
Me
5aA-5gA
Me
Me
Conditions
1a-1g
3aA-3gA
4aA-4gA
2A
3,4-exo
3,4-endo
°C/
Yield
(%)[a]
exo/
endo[c]
ee[d]
%
Entry
1
3
3:4:5[b]
days
CO2Et
Me
Bn
Bn
Bn
Bn
Bn
Me
Me
between the substituents at C-4 and C-5, and
a trans-
1[6b]
-10/5
3aA
75
96:4:0
96:4
95
N
O
configuration between the oxygenated substituents at C-3 and
C-4. This unexpected isomer 5 might result from in situ E/Z
isomerization of (E)-crotonaldehyde under the used thermal
conditions.
1a
CO2Me
2
3
-10/5
-20/7
3bA
3bA
57
58
94:3:3
96:1:3
97:3
99:1
-
N
O
Me
93
1b
CO2t-Bu
Me
Table 1. Uncatalyzed thermal 1,3-DC reaction between ketonitrones 1a-1g
and (E)- crotonaldehyde 2A.
4
-15/7
3cA
60
96:4:0
96:4
96[e]
N
O
CO2R2
R3
CO2R2
R3
CO2R2
R3
CO2R2
R3
R1
R1
O
R1
1c
N
O
N
O
R1
N
O
MW
CHO
CHO
+
+
CHO
N
O
+
CO2Et
3 h , 90°C
neat
5
6
-10/7
-15/5
3dA
3dA
54
56
72:15:8:5[f]
87:10:3
83:17
90:10
-
Me
Me
Me
Me
5aA-5gA
N
O
Et
1a-1g
3aA-3gA
4aA-4gA
2A
92
1d
3,4-exo
3,4-endo
CO2Me
Yield
(%)[a]
7
-15/7
3eA
33
81:15:4
84:16
90[e]
1
3
3:4:5[b]
exo/endo[c]
N
O
n-Pr
Entry
1[6b]
1e
CO2Et
Me
Bn
CO2Et
N
3aA
3bA
3cA
3dA
3eA
3fA
78
84
68
81
77
73
56
82:10:8
90:4:6
88:12
8
9
-10/5
-20/7
3fA
3fA
56
57
90:3:7
94:1:5
97:3
99:1
-
O
1a[6b]
N
O
Me
92
CO2Me
1f
Bn
N
O
Me
2
3
4
5
6
7
96:4
90:10
78:22
77:23
92:8
CO2Me
Me
1b[8]
N
O
10
-18/7
3gA
100[g]
98:1.5:0.5
98:2
92
CO2t-Bu
1g
Bn
89:11:<2[d]
69:20:11
70:20:10[e]
91:8:1
N
O
Me
[a] Isolated combined yield. [b] Determined by 1H NMR after purification by
column chromatography. [c] 3/4 ratio. [d] Determined by 1H NMR after
transformation of the adduct into imine (Figure 2). [e] Determined by HPLC
after reduction of the adduct into alcohol (Figure 2). [f] The fourth isomer was
tentatively attributed to the epimer at C-3 of the isomer 5dA. [g] Conversion.
1c[4a]
CO2Et
Bn
N
Et
O
1d
CO2Me
Bn
N
O
n-Pr
CO2R2
R3
CO2R2
R3
CO2R2
R3
R1
R1
R1
1e[4a]
NaBH4, THF
0°C, 1 h
H2N Ph
4Å MS
N
Ph
N
O
N
O
N
O
CHO
CO2Et
OH
Me
Me
Me
Me
solvent, 24 h
N
Me
O
7
3
8
1f[4a]
CO2Me
Me
N
O
Me
3gA
84:12:4
87:13
Figure 2: Determination of enantiomeric excess of 3 from derivatives 7 or 8.
1g[4a]
As shown in Table 2, the 1,3-DC reaction of ketonitrone 1b8 with
(E)-crotonaldehyde at –10 °C affords the adducts 3bA with
complete regiocontrol and excellent exo-selectivity in 57% yield
(entry 2). Lowering the temperature slightly improved the exo-
[a] Isolated combined yield. [b] Determined by 1H NMR after purification by
column chromatography. [c] 3/4 ratio. [d] 46% yield of isolated pure 3cA. [e]
30% yield of isolated pure 3eA.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
selectivity with high enantioselectivity (93%) of the exo product
3bA (entry 3). Increasing the size of the ester group (t-Bu) in 1c
gave adduct 3cA with 96% ee in 60% yield (entry 4). Increasing
the size of C-substituent and ester chain of nitrone 1d at –10 °C
decreased the exo-selectivity (entry 5). However, at –15 °C the
exo-selectivity for 3dA was improved together with high
enantioselectivity (92% ee) (entry 6). Using n-Pr group as C-
substituent in nitrone 1e gave moderate exo selectivity
(exo:endo = 84:16) with a high enantioselectivity (90% ee) (entry
7). The decrease in the exo selectivity is probably due to the
steric interaction between the C-substituent and the geminal
dimethyl group of the transient iminium. Using nitrones equipped
with C-functionalized side chain (R3 = CH2OAc and CH2CO2Me)
led to no reaction. At last, decreasing the size of the N-
protecting group (Me) of nitrones 1f and 1g afforded adducts
3fA and 3gA with high exo- and enantioselectivities (>97% dr,
92% ee, entries 8-10).
solvent. The imaginary frequencies were checked to exhibit the
expected motion and transition states were verified using IRC
calculations. The enthalpies (ΔH) and Gibbs free energies (ΔG)
were calculated at 256 K. In order to limit the possible
conformations, all the calculations were performed on the N-
methyl nitrone 1g (Scheme 1).
The first information gained from the DFT study of the
cycloaddition of the iminium is that, contrary to the initial thought,
the reaction proceeds via a non-concerted mechanism involving
in a first instance the addition of the oxygen of the nitrone to the
eniminium, and then, the cyclization by addition of the enamine
to the N-alcoxy-iminium (Scheme 1). All attempts to identify a
transition state resulting from a concerted mechanism were
unsuccessful and converged to the stepwise mechanism.
O
Me
O
Me
N
O
Me
N
Me
Me
Me
Me
HCl
N
Ph
CO2Me
N Me
Me
Me
N
N
H
Ph
H
Me
H
N
H
Cl
Intriguingly, even under optimized conditions, the adducts were
mostly not obtained as a couple of exo/endo diastereomers but
as a mixture of 3 diastereomers. Indeed, apart from the
expected exo (always major) and endo isomers, a third isomer
was readily observed for most adducts in minor amounts (not
more than 4% of the isomeric mixture when the reaction was
conducted at –15° or below). However, due to its characteristic
NMR data (vide supra), this minor adduct could be
unambiguously identified in each case as the isomer 5 already
identified as the minor product of the thermal cycloaddition (see
Table 1). The relative configuration of diastereomers 3/4/5 could
be confirmed by characteristic NOESY correlations of the
aldehydic protons of 3gA-5gA. For both 3/4 isomers, a strong
correlation between aldehydic proton and H-5 indicates the 4,5-
trans relationship on the contrary to the 5 isomer for which a
correlation between aldehydic proton and Me-5 confirms a 4,5-
cis relationship. The 3,4-trans relationship for isomer 5 is shown
by the correlation between aldehydic proton and Me-3 and the
absence of correlation with the methyl ester as it is observed for
major isomer 3 (see Supporting Information).
+
O
O
O
N
Cl
2'A
1g
Me
Me
CO2Me
2A
Intermediate 6'gA
Cl
O
Me
N
O
Me
Me
Me
HCl
Si Face
Cl
CO2Me
N Me
N
H2O
N
H
Me
Me
Me
RO
H
Ph
N
H
Me
MeO2C
Me
MeO2C
Me
Re Face
Me
CHO
N
N
O
6'gA
Me
6gA
O
3gA
Me
Scheme 1: Proposed reaction mechanism
The energy profile calculated for the major diastereoisomers and
displayed in figure 3 shows the energy profile of the electronic
energy, the enthalpy and the Gibbs free energy at 256K in
kJ/mol. It appears from this energy profile that the intermediate
6’gA (Figure 3) constitutes a step toward the real transition state
of the cycloaddition corresponding to the cyclization and the
formation of the C-C bond. Interestingly, a look at the electronic
energy and the enthalpy shows that the intermediate is more
stable than the starting materials. Accounting for this specific
feature, the transition state corresponding to the first step of the
process (i.e. the addition of the oxygen of the nitrone to the
eneiminium) could not be located due, most likely, to the very
flat and low energy gap. However, probably the most surprising
result is related to the energy profile of the Gibbs free energy
and to the importance of the entropy term on the course of the
reaction. The Gibbs free energy profile at 256K shows a
continuous rise up to the transition state corresponding to the
formation of the C-C bond. It should also be noted that, at this
stage, the cycloadduct 6gA is more stable than the starting
material by only approximately 20 kJ/mol.
Computational studies on the mechanism of the
organocatalyzed 1,3-DC reaction
An exo-selectivity and a (Re)-facial selectivity relative to the
nitrone (Scheme 1) favouring the formation of adduct 3-exo is in
full agreement with the mechanism postulated by D. MacMillan
for similar [3+2] cycloadditions.5 However, this mechanistic
proposal based on a concerted mechanism can hardly explain
all the results and particularly the formation of the third adduct 5
under reaction conditions that should exclude the thermal
isomerization of the dipolarophile. In order to get a better
understanding of the real course of the reaction and the
parameters that influence the stereochemical outcome of the
reaction, a DFT study of the cycloaddition from the eneiminium
to the corresponding cycloadduct was considered.
Interestingly, due to the negative effect of the temperature on
the entropic term, the energy gap of the transition state increase
with the temperature explaining the weak influence of the
temperature on the kinetic of the reaction.
The DFT calculations were performed using Gaussian 099 at the
M06-2X/6-311+G(d,p) level. 10 Solvent effects were included
using IEFPCM model for the description of nitromethane as the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
Surprisingly, the transition state of lower energy (TS6gA “Endo
E” Re) does not correspond to the major diastereomer. On the
contrary the relative position of the cycloadducts 6gA “Exo E”
Re and 6gA “Endo E” Re corresponds to the experimental
observation. Considering this energy profile and the
experimental results giving the exo adduct 3gA as the major
isomer, it should be assumed that the cycloaddition is not under
kinetic control. Contrary to the most natural guess of the step
influencing the diastereoselectivity of the reaction, the hydrolysis
of the iminium and the ratio existing between the
diastereoisomers of the cycloadducts 6gA (still linked to the
catalyst) seem to both play a key role in the stereochemical
outcome of the cycloaddition process.
Figure 3: Energy profile from the eniminium to the [3+2] cycloadduct
The stepwise reaction mechanism can explain the (minor)
formation of the cycloadduct 5 and means that up to eight
diastereomers of type 6 could be formed during the reaction.
Although, the reaction does not proceed via a concerted
mechanism, the cycloadducts obtained can be seen as the
formal products resulting from the cycloaddition of the Z or the E
enal from the favoured Re or the unfavoured Si face of the
aldehyde olefin. In order to name each diastereoisomer of 6gA,
the corresponding cycloadducts 6gA will be named following this
“formal” nomenclature (Figure 4).
Figure 5: Energy profiles of ΔΔH256 and ΔΔG256 in kJ/mol leading to the
cycloadducts 6gA.
The relative Gibbs free energies of the eight expected
diastereomers represented in Figure 4 were calculated at the
same level with a reference energy corresponding to the starting
materials (i.e. the catalyst in its protonated form, the nitrone and
the crotonaldehyde). In order to take into account the liberation
of water during the formation of the iminium, the Gibbs free
energy of a free molecule of water was added to the energy of
the cycloadduct 6gA. The results are given in Table 3.
O
Me
Cl
Ph
Me
MeO2C
Me N
O
Me
O
Me
O
Me
Cl
Ph
Me
MeO2C
Me N
Cl
Ph
Me
MeO2C
Me N
Cl
Ph
Me
MeO2C
Me N
N
N
N
N
Me
Me
Me
Me
Me
Me
Me
Me
N
N
N
N
H
H
H
H
O
Me
6gA "Exo E" Re
O
O
O
Me
Me
6gA "Exo E" Si
Me
6gA "Endo E" Re
6gA "Endo E" Si
O
Me
Cl
Ph
Me
MeO2C
Me N
O
Me
O
Me
O
Me
Cl
Ph
Me
MeO2C
Me N
Cl
Ph
Me
MeO2C
Me N
Cl
Ph
Me
MeO2C
Me N
N
N
N
N
Me
Me
Me
Me
Me
Me
Me
Me
N
N
N
N
Table 3. Relative Gibbs free energies at 256K of cycloadducts 6gA compared
to 2A+1g+ catalyst I in kJ/mol.
H
H
H
H
O
Me
O
O
O
Me
Me
Me
Cycloadduct
6gA
ΔΔG256
Cycloadduct
6gA
ΔΔG256
6gA "Exo Z" Re
6gA "Exo Z" Si
6gA "Endo Z" Re
6gA "Endo Z" Si
“Exo E” Re
“Exo E” Si
“Endo E” Re
“Endo E” Si
–22.5
–7.9
“Exo Z” Re
“Exo Z” Si
“Endo Z” Re
“Endo Z” Si
–18.3
–6.9
-3.0
7.9
Figure 4: Structure of the diastereomers 6gA possibly formed during the
reaction
–19.5
–11.7
The reaction path was then modelled for the four cycloadducts
resulting from the formal (E)-crotonaldehyde 2A. The
corresponding energy profiles are represented in Figure 5.
It appears that the relative Gibbs free energies of the
cycloadduct 6gA follows exactly the same trend as the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
diastereoselectivity experimentally observed. These theoretical
results show that the three diastereomers significantly more
stable than the others correspond exactly to the three
diastereomers observed experimentally (Table 2 and Table 3).
Due to the adverse contribution of the entropic term on the
transition state and on the stability of 6gA, the kinetic of the
reaction is very poorly affected by the variation of temperature
and particularly its decrease.
The diastereoselectivity of the sequence can therefore be
explained as the result of two factors. First, due to the
reversibility of the cycloaddition step, the cycloadduct 6gA still
linked to the catalyst is obtained as a thermodynamic mixture of
diastereoisomers in a ratio in favor to the “exo E” Re 6gA. Then,
this selectivity is increased during the hydrolysis. The figure 6
describes an intermediate “endo E” Re 9gA higher than the
“exo E” Re 9gA by 9 kJ/mol indicating that the transition states
corresponding to the addition of water certainly follow the same
trend. Due to the second order kinetic of the addition of water
dependent of the concentration of water in the reaction mixture,
this step is most probably the rate limiting step of the all
sequence. This means that the diastereoselectivity of the
reaction is the consequence of the relative stability of the
iminium 6gA modulated by the kinetic of the hydrolysis and
particularly the addition of water on the iminium. It should be
noted that, contrary to usual Macmillan’s conditions, no water is
added to the reaction mixture in order to limit the degradation of
the nitrone leading to a low concentration of water in the reaction
medium. The diastereoselectivity is also influenced by the nature
of the counteranion. While the influence of the counterion was
difficult to view in the single cycloaddition step, however, in the
intermediate 9gA, the counterion stands very close to the water
molecule added on the iminium explaining the importance of the
anion during the addition and therefore in the selectivity of the
sequence. This can give a new sight on previous calculations on
Macmillan’s catalyzed reactions.12
However,
a comparison of the relative energies of the
cycloadducts fails to explain alone the large diastereomeric
excess observed in the experimental results. At this stage, our
hypothesis was that the hydrolysis also plays a key role in the
diastereoselectivity of the reaction. In order to confirm this guess,
an attempt to model the hydrolysis steps for the two major
diastereomers was attempted. However, all our efforts to find the
four transition states involved in the hydrolysis process at the
same DFT level of theory that the previous study were
unsuccessful. Nonetheless, the three intermediates were
calculated and the energy profile is described in Figure 6. This
figure describes the energy profile starting from the starting
materials 1g, 2A and the catalyst used as a reference for the
relative Gibbs free energy. For the transition state and 6gA, the
Gibbs free energy of a free molecule of water was then added.
We should then keep in mind that the kinetics of the
cycloaddition step and the addition of water are of different
orders, the second one depending on the concentration of water
in the reaction mixture. A simple comparison of the energy levels
can therefore not describe alone the kinetic of each step of the
sequence. In both cases described in figure 6, the highest
intermediate in energy corresponds to the intermediate 9gA.
This strongly suggest that the rate limiting step of the hydrolysis
sequence corresponds to the addition of a molecule of water to
the iminium.11
Contrary to the cycloaddition, the hydrolysis step is not
reversible. The minor adduct 4gA was submitted to the reaction
conditions (catalyst I, MeNO2, –10 C, 5 days). However, no
change in the diastereomers ratio was observed confirming the
data given by the energy profile described in figure 6. Indeed,
the energy gap to overcome from the final products 3gA and
4gA to the intermediate 6gA are too high to be compatible with
the reaction conditions.
Figure 6: Energy profiles leading to 3gA and 4gA
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
Variation on the dipolarophile
selectivity (entries 6–7). The use of citral 2I which has β-
disubstitution led to no reactivity over 10 h (entry 8).
To extend the scope of the reaction, different β-substituted enals
2B-2I were used with ketonitrone 1a under thermal conditions.
Most promising β-substituted enals were then tested in
organocatalytic
conditions
in
the
presence
of
imidazolidinone·HCl I at –15 °C for 7 days (Table 5, entries 1, 3,
7-8). The enals 2B and 2C gave the exo adducts 3aB and 3aC
with high diastereoselectivities in low yields (31% and 34%) and
moderate enantioselectivities (56% ee and 54% ee),
respectively (entries 1, 3). Enal 2D having longer aliphatic side
chain gave sluggish conversion as previously mentioned under
thermal condition. The energy profiles described previously
show a weak energy stabilization of the iminium cycloadduct
6gA compared to the starting materials (nitrone + eniminium).
The modest conversion observed with enals bearing an aliphatic
Table 4. Uncatalyzed thermal 1,3-DC reaction between ketonitrone 1a and
enals 2B-2I.
CO2Et
Me
CO2Et
Me
CO2Et
Me
O
CO2Et
Me
Bn
Bn
Bn
MW
3 h, 90°C
neat
N
O
N
O
N
O
Bn
CHO
CHO
+
+
CHO
N
O
+
R
R
R
R
1a
2B-2I
3aB-4aI
4aB-4aI
5aB-5aI
3,4-exo
3,4-endo
Yield
(%)[a]
Entry
2
3
3:4:5[b]
exo/endo
chain might be due to
a relative destabilization of the
O
corresponding cycloadduct 6 compared to the starting materials.
84
78
17
88:9:3
90:8:2
91:9
92:8
This exposed the nitrone to hydrolysis in a larger extent.
1
2
3
2B
2C
2D
3aB
3aC
3aD
Et
O
The cycloaddition of functional dipolarophiles 2E and 2F
proceeds with low yields however diastereo- and
enantioselectivities remain at high levels (entries 6-7). While, β-
aryl substituted enals 2G and 2H and citral 2I did not react with
nitrone 1a under these conditions.
n-Pr
O
84:13:3
87:13
H3C(H2C)6
O
In order to enhance the reactivity and the selectivity of these
reactions, we screened different reaction parameters. As shown
in Table 5, performing the reaction using 2B or 2C at 0 °C or at
room temperature led to higher conversion but in lower isolated
yield even after an increase in the load of the catalyst (entries 2,
4–6). The low yield may be attributed to a partial degradation of
nitrone 1a under the reaction condition in addition to the
dimerization of the enal that gives an aromatic aldehyde 9
(Scheme 2). The dimerized product was obtained in 58% yield
and its structure was identified by NMR analysis. The formation
of this aromatized product is due to iminium ion-enamine
transformation of 2C at room temperature.15
82
92:8:0
92:8
4
2E
3aE
OAc
O
28
26
86:14:0
80:20:0
86:14
80:20
5
6
2F
3aF
3aG
AcO
O
2G
Ph
O
< 5[d]
< 5[d]
-
-
-
-
7
8
2H[c]
2I
3aH
3aI
O2NPh
O
O
CO2Et
Me
CO2Et
Me
EtO2C
Bn
O
Bn
Bn
+
Me
N
O
N
O
cat. I (20 mol%)
CHO
CHO
+
N
+
O
CH3NO2, rt.
4 days
[a] Isolated combined yield. [b] Determined by 1H NMR after purification by
column chromatography. [c] Reaction was performed in nitromethane at 90 °C
for 10 h. [d] No conversion of 1a after 10 h.
n-Pr
3aC exo
n-Pr
4aC endo
n-Pr
1a
2C
9
Scheme 2. Aromatic aldehyde 9 formation through dimerization of enal.
Enals 2B and 2C having Et and Pr groups at β-positions gave
the expected adducts 3aB and 3aC with high exo selectivity
(exo:endo ≥ 90:10) in good yields along with formation of isomer
5 in a trace amount (Table 4, entries 1–2). Examining longer
alkyl group ((CH2)6Me) in 2D slightly decreased the exo-
selectivity (exo:endo = 87:13) in low yield (17%) (entry 3).
Fortunately, the use of functionalized enal (CH2OAc) 2E 13
affords the exo product in high ratio (92:8) in 82% yield without
formation of the isomer 5 (entry 4). Using 2F 14 which has
CH2CH2OAc substituent gave an exo/endo mixture in 86:14 ratio
in low yield (entry 5). β-Aryl substitution in 2G and 2H induced
slow reaction rate over 10 h and without improvement in the
Surprisingly when imidazolidinone.HClO4 II catalyst (10 mol%)
was used with 2C, the selectivity increased to give 94% of exo
product with 88% ee in an improved 62% yield (entry 10). In
contrast cycloaddition with enal 2B did not proceed efficiently
with this catalyst (entry 9). Despite of the lack of reactivity of 2E
under the catalytic effect of HCl salt I, a dramatic change
occurred in the presence of HClO4 salt II as 2E gave the major
adduct 3aE in 61% yield with 82% ee in diastereoisomeric ratio
83:15:2 (entry 11). 2F gave good diastereo- and
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
counterion. This last effect could be seen as a direct influence of
the nature of the counterion on the relative energies of iminium
intermediates 6 (Figure 4) which was shown previously by
theoretical calculations to monitor the selectivity.
enantioselectivities however in lower yield (10%) with formation
of the isomer 5 (entry 12).
Table 5. Asymmetric organocatalytic 1,3-DC reaction between ketonitrone 1a
and enals 2B – 2I
CO2Et
Me
CO2Et
Me
CO2Et
Me
O
CO2Et
Bn
Bn
Bn
N
O
Cat. I or II
N
O
N
O
Bn
Conclusions
+
CHO
CHO
CHO
+
N
O
Me
+
MeNO2
conditions
R
R
R
R
The 1,3-dipolar cycloaddition between C-carboxy ketonitrones
and β-substituted enals organocatalyzed by MacMillan
imidazolidinone salts was extended in the present work to a
1a
2B-2I
4aB-4aI
3aB-3aI
5aB-5aI
3,4-endo
3,4-exo
variety of cycloreactants, affording
a range of quaternary
Conditions
(°C/ days/
cat mol %)
Yield
exo/
endo
ee[d]
%
Entry
2
3
3:4:5[c]
containing polyfunctional 4-formyl isoxazolidines with good
yields and high diastereo- and enantioselectivities. High
selectivities were homogeneously obtained by varying the N-
protecting group or the ester function. However the cycloaddition
process proved to be significantly influenced by the nature of the
C-substituent of the nitrone. Variation on the enal led to some
limitation that could be partially overcome by modulating the
counterion of the catalyst.
(%) [a,b]
With Catalyst I
O
1
-15/7/I 10
rt/4/I 20
31
96:4:0
-
96:4
97:3
56
-
3aB
3aC
2
16 (58)
Et
2B
3
4
-15/7/I 10
0/5/I 10
rt/3/I 10
rt/4/I 20
34
93:5:2
95:5
91:1
91:1
92:8
54
-
O
13 (58)
- (44)
-
-
-
Reaction proceeds with the selective formation of the exo
product, obtained in most cases with better than 90% ee. Apart
5
-
n-Pr
2C
6
27 (45)
-
from the endo adduct, the formation of
a third (minor)
O
diastereomer, was found incompatible with the concerted
mechanism that had been previously postulated for the [3+2]
nitrone/enal cycloaddition under iminium activation. DFT
calculations evidenced that the cycloaddition proceeds in a non-
concerted fashion by a first oxa Michael-type addition of the
nitrone to the double bond followed by a cyclization. This
mechanism is compatible with the formation of the observed
minor diastereomers (formally deduced from a dipolarophile of
opposite geometry). In addition the diastereo- and enantio-
selectivities of the reaction are believed to be under partial
thermodynamic control and influenced by the kinetics of the final
hydrolysis step.
89:15:
2
7
3aE
3aF
-15/7/I 10
18 (20)
3 (16)
92:8
76
2E
OAc
O
8
-15/7/I 10
93:3:4
97:3
-
2F
AcO
With Catalyst II
O
9
3aB
3aC
-15/7/II 10
-15/7/II 10
18
62
87:9:4
93:6:1
91:9
94:6
-
Et
2B
O
10
11
88
n-Pr
2C
O
83:15:
2
Experimental Section
3aE
3aF
-15/7/II 10
-17/5/II 10
61
85:15
89:11
82
85
2E
OAc
(E)-N-(1-Ethoxy-1-oxobutan-2-ylidene)-1-phenylmethanamine oxide
1d
O
A mixture of ethyl 2-oxobutanoate (1.123 g, 8.63 mmol) and BnNHOH
(1.171 g, 9.49 mmol) in CH2Cl2 (4 mL) was stirred at rt for 20 h and then
concentrated under vacuum to give crude yellow oil. Purification by
column chromatography on silica gel (cyclohexane/EtOAc = 9:1) afforded
nitrone (1.153 g, 58 %) as yellow oil, Rf = 0.54 (cyclohexane/EtOAc =
8:2). IR (NaCl) ν = 3033, 2977, 2938, 1708, 1525, 1251, 1171, 1130,
83:10:
7
12
10 (37)
2F
AcO
[a] Isolated combined yield. [b] Conversions are in parentheses. [c]
Determined by 1H NMR after purification by column chromatography. [d]
Determined by HPLC after reduction of the adduct into alcohol.
1
1024, 706 cm-1. H NMR (400 MHz, CDCl3) δ = 1.09 (t, 3H, J = 7.4 Hz),
1.32 (t, 3H, J = 7.1 Hz), 2.75 (q, 2H, J = 7.4 Hz), 4.28 (q, 2H, J = 7.1 Hz),
5.63 (s, 2H), 7.31-7.48 (m, 5H, H-arom) ppm. 13C NMR (100 MHz,
CDCl3) δ = 9.0, 14.1, 22.1, 61.8, 67.4, 128.4, 128.6, 128.70, 134.2, 142.7,
162.6 ppm. HRMS: (ESI+) m/z: calcd for C13H17NNaO3 [M+Na]+:
258.1101, found: 258.1103.
To explain the improvement of reactivity during the use of
imidazolidinium perchlorate II, two hypotheses are proposed (i)
the improvement in the yield could be due to the increase of the
catalytic efficiency, since the kinetics of the catalytic
cycloaddition might be faster than the kinetics of nitrone
degradation, (ii) the increase in the diastereo- and
–
5-Acetate-pent-2-enal 2F
enantioselectivities could be due to the presence of the ClO4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
A solution of crotonaldehyde 2A (3.8 mL, 40 mmol), but-3-en-1-ol (1 mL,
11.6 mmol, 1 equiv) and 2nd generation Grubbs catalyst (80 mg, 0.8
mol %) was stirred at rt for 24 h. Direct purification by column
chromatography on silica gel (ether) and concentration under vacuum
4.46 (m, 1H), 7.22-7.39 (m, 5H, H-arom), 9.81 (d, 1H, J = 2.9 Hz) ppm.
Characteristic signals for diastereoisomers: endo 9.74 (d, J = 1.9 Hz) and
C-5 epimer 9.88 (d, J = 1.8 Hz). 13C NMR (100 MHz, CDCl3) δ = 16.5,
18.8, 28.1, 55.0, 66.6, 72.0, 72.3, 82.8, 127.1, 128.2, 128.3, 138.1, 170.4,
200.2. HRMS (Cl+NH3CH4) m/z: calcd for C18H26NO4 [M+H]+: 320.1862,
found 320.1860.
afforded intermediate (863 mg, 8.6 mmol) as an oil, Rf
= 0.36
(Cyclohexane/EtOAc = 1:1). 1H NMR (200 MHz, CDCl3) δ = 2.63 (dd, 2H,
J = 2.0, 6.0 Hz), 3.84 (t, 2H, J = 6.0 Hz), 6.21 (tdd, 1H, J = 2.0, 8.0, 16.0
Hz), 6.92 (m, 1H), 9.53 (d, 1H, J = 8.0 Hz) ppm. To the residue dissolved
in anhydrous CH2Cl2 (20 mL) was added pyridine (0.85 mL, 10.4 mmol,
1.2 equiv) and acetic anhydride (0.82 mL, 8.7 mmol, 1 equiv). The
solution was stirred at rt during 24 h. After concentration under reduced
pressure, purification by column chromatography on silica gel
(Cyclohexane/EtOAc 7:3) and concentration under vacuum afforded enal
(3R,4S,5R)-Ethyl 2-benzyl-3-ethyl-4-formyl-5-methylisoxazolidine-3-
carboxylate 3dA
From Table 2, entry 6. Purification by column chromatography
(cyclohexane/ EtOAc = 19:1) gave colorless oil (41 mg, 56%), Rf = 0.54
(cyclohexane/ EtOAc
= 8:2). Diastereomeric ratio: exo:endo: 90:10
(crude), > 99:1 (purified). [α]22 = – 84.4 (c 0.615 , CH2Cl2). IR (NaCl) ν =
D
2978, 2937, 2732, 1724, 1455, 1383, 1218, 1144, 1024, 735 cm-1. 1H
NMR (400 MHz, CDCl3 ) δ = 0.94 (t, 3H, J = 7.4 Hz), 1.33 (d, 3H, J = 6.2
Hz), 1.37 (t. 3H, J = 7.1 Hz), 1.81 (dq, 1H, J = 7.4 Hz; 14.7 Hz), 2.04 (m,
1H), 3.57 (dd, 1H, J = 4.4 Hz, 5.3 Hz), 3.81 (d, 1H, J = 14.8 Hz), 4.10 (d,
1H, J = 14.8 Hz), 4.29 (m, 2H), 4.37 (m, 1H), 7.24-7.38 (m, 5H, H-arom),
9.86 (d, 1H, J = 4.3 Hz) ppm. Characteristic signals for diastereoisomers:
endo 9.79 (d, J = 1.1 Hz) and C-5 epimer 9.90 (d, J = 1.9 Hz) and 9.69 (d,
J = 2.9 Hz). 13C NMR (100 MHz, CDCl3) δ = 8.9, 14.4, 17.8, 23.2, 55.0,
61.6, 64.1, 72.5, 76.7, 127.1, 128.2, 128.3, 138.0, 170.5, 200.4 ppm.
HRMS (Cl+NH3CH4) m/z: calcd for C17H24NO4 [M+H]+: 306.1705, found
306.1704.
2F (1.06 g, 7.5 mmol, 65 % global yield) as an oil, Rf
= 0.44
(Cyclohexane/EtOAc = 1:1). IR (NaCl) ν = 2923, 1681, 1039 cm-1. 1H
NMR (400 MHz, CDCl3) δ = 2.06 (s, 3H), 2.66 (dd, 2H, J = 2.0, 6.0 Hz),
4.25 (t, 2H, J = 6.0 Hz), 6.21 (tdd, 1H, J = 2.0, 8.0, 16.0 Hz), 6.82 (m, 1H),
9.53 (d,1H, J = 4.0 Hz) ppm. 13C NMR (100 MHz, CDCl3) δ = 20.7, 31.8,
61.9, 134.5, 153.4, 171.0, 193.6 ppm. HRMS (ESI+) m/z calcd for
C6H12NaO2 [M+Na]+: 143.0708, found 143.0703.
General experimental procedure for the preparation of racemic
cycloadducts under thermal conditions
A mixture of nitrone 1a-g (1 equiv) and aldehyde 2A-I (1.5 - 6 equiv) was
heated under microwave irradiation at 90 °C. After the indicated time, the
residue was purified by column chromatography to give the racemic
adducts.
(3R,4S,5R)-Methyl 2-benzyl-4-formyl-5-methyl-3-propylisoxazolidine-
3-carboxylate 3eA
Purification by column chromatography (cyclohexane/ EtOAc = 19:1)
gave yellow oil (25 mg, 33%), Rf = 0.58 (cyclohexane/ EtOAc = 8:2).
General experimental procedure for the preparation of optically
active cycloadducts under organocatalytic conditions
Diastereomeric ratio: exo:endo: 84:16 (crude), 98:2 (purified). [α]20 = –
D
In a 10 mL test tube, catalyst I or II (0.025 mmol) and nitrone (0.25 mmol)
were charged and suspended in CH3NO2 (1 mL). The mixture was cooled
to the indicated temperature and stirred for 10 min before aldehyde (1
mmol) was added, followed by the addition of (0.75 mmol over 24 h
intervals) over the course of 5 or 7 days. The reaction mixture was
passed through a small plug of silica gel, washed with EtOAc, and
concentrated in vacuum to give a crude oil. The diasteroselectivity was
determined by 1H NMR. Purification by column chromatography gave the
expected product.
62.86 (c 0.43, CHCl3). IR (NaCl) ν = 2961, 2874, 2735, 1725, 1605, 1496,
1454, 1382, 1217, 1144, 1029, 1004, 735, 697 cm-1. 1H NMR (400 MHz,
CDCl3 ) δ = 0.91 (t, 3H, J = 7.3 Hz), 1.30 (m, 2H), 1.31 (d, 3H, J = 6.2 Hz),
1.73 (ddd, 1H, J = 4.4 Hz, 12.2 Hz, 13.7 Hz), 1.95 (ddd, 1H, J = 4.9 Hz,
12.2 Hz, 13.7 Hz), 3.57 (dd, 1H, J = 4.3 Hz, 5.2 Hz), 3.78 (d, 1H, J = 14.7
Hz), 3.83 (s, 3H), 4.07 (d, 1H, J = 14.7 Hz), 4.40 (m, 1H), 7.23-7.38 (m,
5H, H-arom), 9.86 (d, 1H, J = 4.1 Hz) ppm. Characteristic signals for
diastereoisomers: endo 9.78 (d, J = 1.1 Hz) and C-5 epimer 9.90 (d, J =
2.0 Hz). 13C NMR (100 MHz, CDCl3) δ = 14.4, 17.8, 18.1, 32.4, 52.4, 54.9,
64.3, 67.6, 72.6, 127.2, 128.2, 128.3, 137.9, 171.4, 200.3. HRMS (ESI)
m/z : calcd for C17H23KNO4 [M+K]+: 344.1259, found 344.1236.
(3R,4S,5R)-Methyl
2-benzyl-4-formyl-3,5-dimethylisoxazolidine-3-
carboxylate 3bA
From Table 2, entry 3. Purification by column chromatography
(3R,4S,5R)-Ethyl 4-formyl-2,3,5-trimethylisoxazolidine-3-carboxylate
3fA
(cyclohexane/ EtOAc = 9:1) gave colorless oil (38 mg, 58%), Rf = 0. 33
(cyclohexane/ EtOAc = 8:2). Diastereomeric ratio: exo:endo: 99:1. [α]20
:
From Table 2, entry 9. Purification by column chromatography
(cyclohexane/ EtOAc = 6:1) gave colorless oil (29 mg, 57%), Rf = 0.21
D
– 75.7 (c 0.43, CH2Cl2). IR (NaCl) ν = 2981, 2952, 1773, 1728, 1605,
1455, 1380, 1237, 1145, 1062, 781 cm-1. 1H NMR (400 MHz, CDCl3 ) δ =
1.33 (d, 3H, J = 6.2 Hz), 1.48 (s, 3H), 3.49 (dd, 1H, J = 2.8 Hz, 6.4 Hz),
3.82 (s, 3H), 3.84 (d, 1H, J = 14.5 Hz), 3.96 (d, 1H, J = 14.5 Hz), 4.50 (m,
1H), 7.23 – 7.38 (m, 5H, H-arom), 9.82 (d, 1H, J = 2.8 Hz) ppm.
Characteristic signals for diastereoisomers: endo 9.74 (d, J = 1.6 Hz) and
C-5 epimer 9.89 (d, J = 1.5 Hz). 13C NMR (100 MHz, CDCl3) δ = 16.6,
18.5, 52.6, 55, 66.3, 72.1, 72.2, 127.2, 128.2, 128.3, 137.5, 171.8, 199.9
ppm. HRMS (Cl+NH3CH4) m/z: calcd for C15H20NO4 [M+H]+: 278.1392,
found 278.1380.
(cyclohexane/ EtOAc = 8:2). Diastereomeric ratio: exo:endo: 99:1. [α]22
D
= – 65.35 (c 0.635, CH2Cl2). IR (NaCl) ν = 2981, 2953, 2810, 1737, 1496,
1455, 1375, 1152, 1030, 745 cm-1. 1H NMR (400 MHz, CDCl3) δ = 1.32
(m, 6H), 1.40 (s, 3H), 2.64 (s, 3H), 3.53 (m, 1H), 4.27 (m, 2H), 4.58 (m,
1H), 9.81 (d, 1H,
J = 2.4 Hz) ppm. Characteristic signals for
diastereoisomers: endo 9.73 (d, J = 1.6 Hz) and C-5 epimer 9.90 (d, J =
1.4 Hz). 13C (100 MHz, CDCl3) δ = 14.2, 16.2,18.7, 38.6, 61.9, 66.1, 72.3,
72.9, 171.3, 199.8 ppm. HRMS (Cl+NH3) m/z: calcd for C10H18NO4
[M+H]+: 216.1236, found 216.1236.
(3R,4S,5R)-tert-Butyl 2-benzyl-4-formyl-3,5-dimethylisoxazolidine-3-
carboxylate 3cA
(3R,4S,5R)-Methyl 4-formyl-2,3,5-trimethylisoxazolidine-3-carboxy-
late 3gA
Purification by column chromatography (cyclohexane/ EtOAc = 19:1)
gave colorless oil (48 mg, 60%), Rf = 0, 52 (cyclohexane/ EtOAc = 8:2).
Diastereomeric ratio: exo:endo: 96:4. [α]22D = – 56.41 (c 0.34, CH2Cl2). IR
(NaCl) ν = 2978, 2933, 1723, 1606, 1455, 1392, 1369, 1251, 1145, 1065,
Purification by column chromatography (cyclohexane/ EtOAc = 6:1) gave
colorless oil (conversion >98%), Rf = 0.34 (cyclohexane/ EtOAc = 3:2).
Diastereomeric ratio: exo:endo: 98:2. 1H NMR (400 MHz, CDCl3) δ =
1.35 (d, 3H, J = 6.1 Hz),1.41 (s, 3H), 2.63 (s, 3H), 3.54 (dd, 1H, J = 2.3
Hz, 6.4 Hz), 3.81 (s, 3H), 4.60 (dq, 1H, J = 6.1 Hz, 6.1 Hz), 9.82 (d, 1H, J
= 2.3 Hz) ppm. Characteristic signals for diastereoisomers: endo 9.73 (d,
1
846, 734 cm-1. H NMR (400 MHz, CDCl3 ) δ = 1.32 (d, 3H, J = 6.2 Hz),
1.43 (s, 3H), 1.54 (s, 9H), 3.43 (dd, 1H, J = 2.9 Hz; 6.4 Hz), 3.96 (s, 2H),
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
J = 1.5 Hz) and C-5 epimer 9.90 (d, J = 1.6 Hz). 13C (100 MHz, CDCl3) δ
= 16.3, 18.7, 38.6, 52.8, 66.0, 72.4, 73.2, 172.0, 199.8 ppm.
1H, J = 14.5 Hz), 4.12 (m, 1H), 4.27 (m, 3H), 4.64 (dd, 1H, J = 6.2 Hz,
11.5 Hz), 7.23 – 7.36 (m, 5H, H-arom), 9.83 (d, 1H, J = 1.8 Hz) ppm.
Characteristic signals for diastereoisomers: endo 9.78 (d, J = 0.9 Hz) and
C-5 epimer 9.85 (d, J = 2.6 Hz). 13C NMR (100 MHz, CDCl3) δ = 14.2,
15.4, 20.7, 54.9, 62.1, 62.8, 64.2, 71.3, 73.0, 127.2, 128.2, 128.3, 137.4,
170.6, 170.9, 198.1 ppm. HRMS (CIID) m/z: calcd for C18H24NO6 [M+H]+:
350.1604, found 350.1619.
(3R,4S,5R)-Ethyl 2-benzyl-5-ethyl-4-formyl-3-methylisoxazolidine-3-
carboxylate 3aB
From Table 5, entry 1. Purification by column chromatography
(cyclohexane/ EtOAc = 19:1) gave brown oil (23 mg, 31%), Rf = 0. 52
(cyclohexane/ EtOAc = 8:2). Diastereomeric ratio: exo:endo: 96:4. [α]20
:
D
– 0.92 (c 0.152, CHCl3). IR (NaCl) ν = 2925, 1727, 1605, 1455, 1376,
1241, 1140, 1029, 731, 697 cm-1. 1H NMR (400 MHz, CDCl3) δ = 0.88 (t,
3H, J = 7.5 Hz), 1.34 (t, 3H, J = 7.1 Hz), 1.46 (s, 3H), 1.62 (m, 1H), 1.76
(m, 1H), 3.54 (dd, 1H, J = 2.9 Hz, 6.2 Hz), 3.94 (s, 2H), 4.27 (m, 3H),
7.25-7.38 (m, 5H, H-arom), 9.83 (d, 1H, J = 2.9 Hz) ppm. Characteristic
signals for diastereoisomers: endo 9.76 (d, J = 1.5 Hz) and C-5 epimer
9.87 (d, J = 2.8 Hz). 13C NMR (100 MHz, CDCl3) δ = 10.1, 14.2, 16.1,
26.5, 55.0, 61.8, 65.0, 71.9, 77.3, 127.1, 128.2, 128.3, 137.8, 171.3,
200.0 ppm. HRMS (ESI) m/z: calcd for C17H23NNaO4 [M+Na]+: 328.1519,
found 328.1520.
(3R,4S,5R)-Ethyl
isoxazolidine-3-carboxylate 3aF
5-(2-acetoxyethyl)-2-benzyl-4-formyl-3-methyl
From Table 5, entry 12. Purification by column chromatography
(cyclohexane/ EtOAc = 95:5) gave yellow oil (20 mg, 10%), Rf = 0.44
(cyclohexane/ EtOAc = 8:2). Diastereomeric ratio: exo:endo: 89:11. 1H
NMR (400 MHz, CDCl3) δ = 1.35 (t, 3H, J = 7.0 Hz), 1.46 (s, 3H), 1.98 (s,
3H), 1.92 and 2.05 (2m, 2H), 3.61 (dd, 1H, J = 2.6 Hz, 6.1 Hz), 3.93 (s,
2H), 3.98-4.13 (m, 2H), 4.10-4.32 (m, 2H), 4.55 (dt, 1H, J = 5.8 Hz, 7.3
Hz), 7.20-7.45 (m, 5H, H-arom), 9.81 (d, 1H, J
= 2.6 Hz) ppm.
Characteristic signals for diastereoisomers: endo 9.77 (d, J = 0.9 Hz) and
C-5 epimer 9.85 (d, J = 2.6 Hz). 13C NMR (100 MHz, CDCl3) δ = 14.1,
15.7, 20.9, 20.9, 32.6, 55.2, 61.3, 61.7, 65.3, 73.3, 128.3, 137.4, 170.7,
171.0, 199.2 ppm. HRMS (ESI) m/z: calcd for C19H26NO6 [M+H]+:
364.1760, found 364.1755.
(3R,4S,5R)-Ethyl 2-benzyl-4-formyl-3-methyl-5-propylisoxazolidine-
3-carboxylate 3aC
From Table 5, entry 10. Purification by column chromatography
(cyclohexane/ EtOAc = 19:1) gave yellow oil (49 mg, 62%), Rf = 0.37
(cyclohexane/ EtOAc = 8:2). Diastereomeric ratio: exo:endo: 94:6 (crude),
rac-Ethyl
2-benzyl-4-formyl-3-methyl-5-phenylisoxazolidine-3-car-
> 99:1 (purified). . [α]20 = – 4.71 (c 0.104, CHCl3). IR (NaCl) ν = 2958,
boxylate 3aG
D
2932, 2872, 2733, 1726, 1605, 1455, 1377, 1231, 1141, 1020, 734, 697
cm-1. 1H NMR (400 MHz, CDCl3) δ = 0.88 (t, 3H, J = 7.3 Hz), 1.27 (m,
2H), 1.34 (t, 3H, J = 7.1 Hz), 1.46 (s, 3H), 1.56 (tdd, J = 4 Hz, 10.1 Hz,
12.7 Hz, 1H), 1.71 (m, 1H), 3.53 (dd, 1H, J = 2.9 Hz, 6.2 Hz), 3.90 (d, 1H,
J = 14.7 Hz), 3.95 (d, 1H, J = 14.5 Hz), 4.27 (m, 2H), 4.37 (m, 1H), 7.23-
7.38 (m, 5H, H-arom), 9.82 (d, 1H, J = 2.9 Hz) ppm. Characteristic
signals for diastereoisomers: endo 9.76 (d, J = 1.6 Hz) and C-5 epimer
9.87 (d, J = 2.6 Hz). 13C NMR (100 MHz, CDCl3) δ = 13.9, 14.2, 16.2,
19.2, 35.5, 55.0, 61.8, 65.3, 71.9, 75.9, 127.1, 128.2, 128.3, 137.8, 171.3,
200.1 ppm. HRMS (ESI) m/z: calcd for C18H25NNaO4 [M+Na]+: 342.1676,
found 342.1682. In parallel with this product, a by-product 9 was formed
(Oil, 28 mg, 58%). 1H NMR (400 MHz, CDCl3) δ = 1.00 (t, 3H, J = 7.3 Hz),
1.27 (t, 3H, J = 7.5 Hz), 1.64 (m, 2H), 2.67 (m, 2H), 2.72 (q, 2H, J = 7.5
Hz), 7.30 (d, 1H, J = 7.8 Hz), 7.64 (d, 1H, J = 7.8 Hz), 7.69 (s,1H), 9,95
(s, 1H).
Purification by column chromatography (EP/ EtOAc = 9:1) gave yellow oil
(45 mg, 26%), Rf = 0.61 (EP/ EtOAc = 8:2). Diastereomeric ratio:
exo:endo: 80:20. IR (NaCl) ν = 2982, 2935, 2736, 1725, 1604, 1496,
1455, 1374, 1243, 1126, 1027, 912, 735, 699 cm-1. 1H NMR (400 MHz,
CDCl3) δ = 1.32 (t, 3H, J = 7.1 Hz), 1.54 (s, 3H), 3.94 (dd, 1H, J = 1.4 Hz,
7.1 Hz), 4.0 (d, 1H, J = 14.3 Hz), 4.11(d, 1H, J = 14.3 Hz), 4.28 (q, 2H, J
= 7.1 Hz), 5.44 (d, 1H, J = 7.1 Hz), 7.20-7.45 (m, 10H, H-arom), 9.92 (d,
1H, J = 1.4 Hz) ppm. 13C NMR (100 MHz, CDCl3) δ = 14.2, 16.1, 55.2,
61.9, 68.3, 71.8, 76.9, 126.3, 127.2, 127.8, 128.2, 128.4, 128.5, 137.5,
137.6, 171.3, 198.9 ppm. HRMS (ESI) m/z : calcd for C21H24NO4 [M+H]+:
354.1700, found 354.1715.
Determination of the enantioselectivity of the cycloadducts
Protocol A: In a NMR tube with a magnetic stirring and in the presence
of molecular sieves 4 Å, aldehyde (1 equiv) and (S)-methyl-benzylamine
(1 equiv) was dissolved in the indicated solvent. After stirring for 24 h,
The mixture was stirred for 24 h at room temperature, concentrated in
vacuum afforded the expected product 7.
rac-Ethyl
2-benzyl-4-formyl-5-heptyl-3-methylisoxazolidine-3-car-
boxylate 3aD
Purification by column chromatography (cyclohexane/ EtOAc = 98:2)
gave yellow oil (56 mg, 17%), Rf = 0.50 (cyclohexane/ EtOAc = 8:2). IR
(NaCl) ν = 2925, 2854, 1723, 1136, 795 cm-1. 1H NMR (400 MHz, CDCl3)
δ = 0.86 (t, 3H, J = 7.0 Hz), 1.20-1.30 (m, 10H), 1.34 (t, 3H, J = 7.0 Hz),
1.46 (s, 3H), 1.58 and 1.72 (2m, 2H), 3.53 (dd, 1H, J = 2.9 Hz, 6.1 Hz),
3.91 and 3.95 (2m, 2H, J = 14.4 Hz), 4.20-4.32 (m, 2H), 4.35 (dt, 1H, J =
6.4 Hz, 7.0 Hz), 7.20-7.40 (m, 5H, H-arom), 9.82 (d, 1H, J = 2.0 Hz) ppm.
Characteristic signals for diastereoisomers: endo 9.75 (d, J = 1.5 Hz) and
C-5 epimer 9.87 (d, J = 2.6 Hz). 13C NMR (100 MHz, CDCl3) δ = 14.1,
14.2, 22.6, 25.9, 29.0, 29.4, 31.7, 55.0, 61.8, 65.4, 71.9, 76.2, 127.1,
128.2, 128.3, 137.8, 171.3, 200.1 ppm. HRMS (ESI) m/z: calcd for
C22H34NO4 [M+H]+: 376.2488, found 376.2482.
Protocol B: In a 10 mL vial, to a solution of diastereomeric mixture of the
aldehyde (1 equiv) in THF were added NaBH4 (2 equiv). The mixture was
stirred at 0 °C for 1 h. The reaction mixture was quenched by (2 ml) HCl
(2M) and was stirred for 5 min at room temperature. Extraction of the
aqueous layer with CH2Cl2 (5 mL × 3), followed by drying the combined
organic layers with MgSO4, filtration, and concentration in vacuum
afforded the crude alcohol. Purification by column chromatography gave
the desired product 8.
(3R,4R,5R)-Methyl
2-benzyl-3,5-dimethyl-4-((E)-((S)-1-phenylethyl
imino) methyl)isoxazolidine-3-carboxylate 7bA
Protocol A: the product obtained is colorless oil. 1H NMR (400 MHz,
C6D6) δ = 1.23 (d, 3H, J = 6.1 Hz), 1.27 (s, 3H), 1.39 (d, 3H, J = 6 .6 Hz),
3.30 (s, 3H), 3.66 (dd, 1H, J = 5.9 Hz, 6.7 Hz), 3.95 (d, 1H, J = 14.4 Hz),
3.99 (d, 1H, J = 14.4 Hz), 4.11 (q, 1H, J = 6.6 Hz), 4.29 (m, 1H), 7.06 –
7.26 (m, 10H, H-arom), 7.53 (d,1H, J = 5.9 Hz). HRMS: (Cl+NH3CH4)
m/z: calcd for C23H29N2O3 [M+H]+: 381.2178, found 381.2180. The
enantiomeric excess was determined by NMR 1H, based on H5 (4.29
ppm) major and (4.38 ppm) minor with 93% ee and dr: 97:3.
(3R,4R,5S)-Ethyl 5-acetoxymethyl-2-benzyl-4-formyl-3-methylisoxa-
zolidine-3-carboxylate 3aE
From Table 5, entry 11. Purification by column chromatography
(cyclohexane/ EtOAc = 85:15) gave yellow oil (53 mg, 61%), Rf = 0.06
(cyclohexane/ EtOAc = 19:1). Diastereomeric ratio: exo:endo: 85:15. IR
(NaCl) ν = 2983, 2736, 1735, 1455, 1372, 1238, 1136, 1043, 735 cm-1. 1H
NMR (400 MHz, CDCl3) δ = 1.34 (t, 3H, J = 7.1Hz), 1.47 (s, 3H), 2.01 (s,
3H), 3.70 (dd, 1H, J = 1.5 Hz, 6.2 Hz), 3.92 (d, 1H, J = 14.5 Hz), 4.03 (d,
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
(3R,4R,5R)-tert-Butyl
2-benzyl-4-(hydroxymethyl)-3,5-dimethyl
(3R,4R,5R)-Ethyl
2-benzyl-5-ethyl-4-(hydroxymethyl)-3-methyl
isoxazolidine-3-carboxylate 8cA
isoxazolidine-3-carboxylate 8aB
Protocol B: 3cA (0.078 mmol, exo:endo = 98:2). Purification by column
chromatography (cyclohexane/ EtOAc = 8:2) gave white powder (20 mg,
87%), Rf = 0. 5 (cyclohexane/ EtOAc = 6:4). IR (NaCl) ν = 3443, 2975,
2929, 1719, 1455, 1369, 1138, 734 cm-1. 1H NMR (400 MHz, CDCl3 ) δ =
1.33 (d, 3H, J = 6.1 Hz), 1.40 (s, 3H), 1.52 (s, 9H), 2.45 (br s,1H), 2.79
(dt, 1H, J = 6.3 Hz, 7.2 Hz), 3.69 (m, 1H), 3.82 (m, 2H), 3.92 (d, 1H, J =
Protocol B: 3aB (0.028 mmol, exo:endo = 95:5). Purification by column
chromatography (cyclohexane/ EtOAc = 8:2) gave colorless oil (5 mg,
63%), Rf = 0.22 (cyclohexane/ EtOAc = 7:3). IR (NaCl) ν = 3433, 2926,
1726, 1455, 1260, 1023, 803, 698 cm-1. 1H NMR (400 MHz, CDCl3) δ =
0.91 (t, 3H, J = 7.5 Hz), 1.31 (t, 3H, J = 7.1 Hz), 1.47 (s, 3H), 1.62 (m,
1H), 1.73 (m, 1H), 2.34 (br s, 1H), 2.89 (ddd, 1H, J = 5.3 Hz, 6.2 Hz, 7.6
Hz), 3.62 (dd, 1H, J = 6.0 Hz, 13.1 Hz), 3.77 (m, 2H), 3.92 (d, 1H, J =
14.6 Hz), 4.03 (d, 1H, J = 14.6 Hz), 4.23 (m, 2H), 7.21 – 7.38 (m, 5H, H-
arom) ppm. 13C NMR (100 MHz, CDCl3) δ = 10.5, 14.1, 14.2, 27.5, 55.1,
55.5, 61.5, 61.8, 71.5, 80.5, 126.9, 128.2, 128.3, 138.5, 173.2 ppm.
14.7 Hz), 4.05 (d, 1H, J = 14.7 Hz), 7.21 -7.38 (m, 5H, H-arom) ppm. 13
C
NMR (100 MHz, CDCl3) δ = 14.6, 19.8, 28.1, 55.1, 56.8, 61.6, 71.9, 75.6,
82.4, 126.9, 128.1, 128.2, 138.7, 172.4 ppm. HRMS: (ESI) m/z: calcd for
C18H28NNaO4 [M+Na]+: 344.1832, found 344.1847. Enantioselectivity
was determined by chiral HPLC analysis, CHIRALPAK AS-H, isooctane/i-
PrOH = 98:2, 1 mL/min, 254 nm; tR (min): 13.4 (minor) and 20.9 (major)
with 96% ee for exo and dr: 98:2.
HRMS: (ESI) m/z
:
calcd for C17H26NO4 [M+H]+: 308.1856,
found 308.1864. Enantioselectivity was determined by chiral HPLC
analysis, CHIRALPAK AS-H, isooctane/i-PrOH = 98:2, 1 mL/min, 211
nm; tR (min): 22.2 (minor) and 27.4 (major) with 56% ee for exo and dr:
78:22.
(3R,4R,5R)-Ethyl 2-benzyl-3-ethyl-5-methyl-4-((E)-((S)-1-phenylethyl
imino) methyl) isoxazolidine-3-carboxylate 7dA
According the protocol A, but this reaction was stirred in chloroform on a
silver foil. 1H NMR (400 MHz, CDCl3) δ = 0.93 (t, 3H, J = 7.4 Hz), 1.29 (d,
3H, J = 6.2 Hz), 1.35 (t, 3H, J = 7.1 Hz), 1.51 (d, 3H, J = 6.6 Hz), 1.65 (m,
1H), 1.94 (m, 1H), 3.65 (dd, 1H, J = 5.5 Hz, 7.6 Hz), 3.76 (d, 1H, J = 15.0
Hz), 4.08 (m, 2H), 4.28 (m, 2H), 4.39 (q, 1H, J = 6.6 Hz), 7.21 – 7.37 (m,
10H, H-arom), 7.87 (d, 1H, J = 7.6 Hz) ppm. HRMS: (ESI) m/z : calcd
for C25H33N2O3 [M+H]+: 409.2486, found 409.2469.The enantiomeric
excess was determined by 1H NMR, based on H12 (0.93 ppm) major and
(0.63 ppm) minor with 92% ee and dr: 96:4.
(3R,4R,5R)-Ethyl
isoxazolidine-3-carboxylate 8aC
2-benzyl-4-(hydroxymethyl)-3-methyl-5-propyl
Protocol B: 3aC (0.093 mmol, exo:endo = 97:3) Purification by column
chromatography (cyclohexane/ EtOAc = 8:2) gave yellow oil (23 mg,
77%), Rf = 0.27 (cyclohexane/ EtOAc = 7:3). IR (NaCl) ν = 3448, 2957,
2931, 2872, 1726, 1455, 1143, 1024, 698 cm-1. 1H NMR (400 MHz,
CDCl3) δ = 0.88 (t, 3H, J = 7.3 Hz), 1.31 (t, 3H, J = 7.1 Hz), 1.34 (m, 2H),
1.45 (s, 3H), 1.55 (m, 1H), 1.72 (m, 1H), 2.39 (br s, 1H), 2.87 (ddd, 1H, J
= 5.4 Hz, 6.3 Hz, 7.5 Hz), 3.72 (m, 3H), 3.92 (d, 1H, J = 14.6 Hz), 4.01 (d,
1H, J = 14.7 Hz), 4.22 (m, 2H), 7.21 – 7.37 (m, 5H, H-arom) ppm. 13C
NMR (100 MHz, CDCl3) δ = 14.1, 14.2, 19.5, 36.7, 55.1, 55.8, 61.5, 61.7,
71.4, 79.1, 126.9, 128.1, 128.3, 138.4, 173.2 ppm. HRMS: (ESI) m/z:
(3R,4R,5R)-Methyl
2-benzyl-4-(hydroxymethyl)-5-methyl-3-propyl
isoxazolidine-3-carboxylate 8eA
Protocol B: 3eA (0,024 mmol, exo:endo = 100:0). Purification by column
chromatography (cyclohexane/ EtOAc = 9:1) gave colorless oil (6 mg,
74%), Rf = 0. 27 (cyclohexane/ EtOAc = 7:3). IR (NaCl) ν = 3447, 2962,
2876, 1732, 1454, 1214, 1082, 735 cm-1. 1H NMR (400 MHz, CDCl3) δ =
0.97 (t, 3H, J = 7.3 Hz), 1.32 (d, 3H, J = 6.3 Hz), 1.35 (m, 2H), 1.71 (m,
1H), 1.83 (m,1H), 2.66 (br s, 1H), 2.94 (dd, 1H, J = 4.7 Hz, 9.9 Hz), 3.80
(s, 4H), 3.86 (m, 2H), 4.15 (m, 2H), 7.23-7.34 (m, 5H, H-arom) ppm. 13C
NMR (100 MHz, CDCl3) δ = 14.6, 18.7, 18.8, 32.2, 51.8, 53.3, 55.0, 62.1,
76.0, 76.1, 127.0, 128.1, 128.3, 138.5, 172.5 ppm. HRMS: (ESI) m/z :
calcd
for C18H28NO4
[M+H]+:
322.2013,
found
322.2018.
Enantioselectivity was determined by chiral HPLC analysis, CHIRALPAK
AS-H, isooctane/i-PrOH = 98:2, 1 mL/min, 211 nm; tR (min): 18.7 (minor)
and 21.5 (major) with 88% ee for exo obtained by Catalyst II with dr =
94:6; tR (min): 18.8 (minor) 21.5 (major) with 54% ee for exo obtained by
Catalyst I with dr: 78:22.
(3R,4R,5S)-Ethyl
isoxazolidine-3-carboxylate 8aE
5-acetoxy-2-benzyl-4-(hydroxymethyl)-3-methyl
calcd for
C17H25NNaO4 [M+Na]+: 330.1676, found 330.1684.
Protocol B: 3aE (0.048 mmol, exo:endo = 85:15). Purification by column
chromatography (cyclohexane/ EtOAc = 7:3) gave colorless oil (13 mg,
73%), Rf = 0.25 (cyclohexane/ EtOAc = 5:5). IR (NaCl) ν = 3520, 2935,
1734, 1239, 1039, 698 cm-1. 1H NMR (400 MHz, CDCl3) δ = 1.31 (t, 3H, J
= 7.1 Hz), 1.42 (s, 3H), 2.03 (s, 3H), 2.28 (br s, 1H), 2.94 (m, 1H), 3.75
(m, 1H), 3.86 (m, 1H), 3.92 (d, 1H,J = 14.6 Hz), 3.96 (m, 1H), 4.02 (d, 1H,
J = 14.6 Hz), 4.23 (m, 4H), 7.22 – 7.37 (m, 5H, H-arom) ppm. 13C NMR
(100 MHz, CDCl3) δ = 13.4, 14.1, 20.8, 53.4, 54.9, 61.5, 61.7, 65.7, 70.9,
77.6, 127.0, 128.1, 128.2, 137.9, 171.1, 171.3 ppm. HRMS: (ESI) m/z :
calcd for C18H25NNaO6 [M+Na]+: 374.1574, found 374.1580.
Enantioselectivity was determined by chiral HPLC analysis, Lux-5µ-
Cellulose II, isooctane/i-PrOH = 90:10, 1 mL/min, 211 nm; tR (min): 23.9
(minor) and 30.3 (major) with 82% ee for exo obtained by Catalyst II with
dr: 91:9; tR (min): 24 (minor) and 30 (major) with 76% ee for exo obtained
by Catalyst I with dr: 88:12.
Enantioselectivity was determined by chiral HPLC analysis, CHIRALCEL
OD, isooctane/i-PrOH = 98:2, 1 mL/min, 254 nm; tR (min): 15.9 (major)
and 18.6 (minor) with 90% ee for exo and dr: 95:5.
(3R,4R,5R)-Ethyl
2,3,5-trimethyl-4-((E)-((S)-1-phenylethylimino)
methyl) isoxazolidine-3-carboxylate 7fA
According the protocol A, the product obtained is colorless oil. 1H NMR
(400 MHz, C6D6) δ = 0.91 (t, 3H), 1.23 (d, 3H), 1.37 (m, 6H), 2.64 (s, 3H),
3.69 (m, 1H), 3.92 (m, 2H), 4.12 (m,1H), 4.44 (m, 1H), 7.08 -7.58 (m, 6H,
H-arom, 1H). HRMS: (ESI) m/z: calcd for C18H26N2NaO3 [M+Na]+:
341.1836, found 341.1847. The enantiomeric excess was determined by
1H NMR, based on H5 (4.44 ppm) major and (4.53 ppm) minor with 92%
ee and dr: 96:4.
(3R,4R,5R)-Methyl
2,3,5-trimethyl-4-((E)-((S)-phenylethylimino)
methyl) isoxazolidine-3-carboxylate 7gA
(3R,4R,5S)-Ethyl
5-(2-acetoxyethyl)-2-benzyl-4-(hydroxymethyl)-3-
According the protocol A, the product obtained is colorless oil. 1H NMR
(400 MHz, CDCl3) δ = 1.27 (s, 3H), 1.34 (d, 3H, J = 5.8 Hz), 1.49 (d, 3H,
J = 6.7 Hz), 2.62 (s, 3H), 3.46 (dd, 1H, J =6.4 Hz, J = 6.4 Hz), 3.76 (s,
3H), 4.32 (m, 1H), 4.37 (q, 1H, J = 6.7 Hz), 7.2 -7,4 (m, 5H, H-arom),
7.76 (d, 1H, J = 61 Hz). The enantiomeric excess was determined by 1H
NMR, based on MeO (3.76 ppm) major and (3.74 ppm) minor with 92%
ee and dr: 96:4.
methyl isoxazolidine-3-carboxylate 8aF
Protocol B: 3aF (0.16 mmol, exo:endo = 89:11). Purification by column
chromatography (cyclohexane/ EtOAc = 7:3) gave colorless oil (38 mg,
65%), Rf = 0.37 (cyclohexane/ EtOAc = 5:5). IR (NaCl) ν = 3520, 2935,
1734, 1239, 1039, 698 cm-1. 1H NMR (400 MHz, CDCl3) δ = 1.29 (t, 3H, J
= 7.3 Hz), 1.45 (s, 3H), 1.90 – 2.10 (m, 2H), 1.98 (s, 3H), 2.21 (br s, 1H),
2.92 (q, 1H, J = 6.1 Hz), 3.74 (m, 1H), 3.83 (m, 2H), 3.90 (d, 1H,J = 14.3
Hz), 3.98 (d, 1H, J = 14.3 Hz), 4.00 – 4.20 (m, 2H) 4.25 (m, 4H), 7.22 –
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
7.37 (m, 5H, H-arom) ppm. 13C NMR (50 MHz, C6D6) δ = 14.0 (2C), 20.3,
Acknowledgements
33.8, 55.4, 61.0, 61.5, 61.7, 71.2, 138.2, 171.1, 172.8 ppm. HRMS: (ESI)
m/z
:
calcd for C19H28NO6 [M+H]+: 366.1911, found 366.1904.
This work was supported by the ANR (ANR-11-BS07-005). The
authors are particularly grateful to F. Legros, P. Gangnery and
C. Jacquemmoz for technical and analytical assistance.
Enantioselectivity was determined by chiral HPLC analysis, CHIRALPAK
AS-H, hexane/i-PrOH = 95:5, 1 mL/min, 254 nm; tR (min): 19.9 (minor)
and 26.9 (major) with 85% ee for exo obtained by Catalyst II with dr:
89:11.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701307
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Kawther Ben Ayed, Mathieu Y. Laurent,
Arnaud Martel*, Khalid B. Selim, Souhir
Abid, Gilles Dujardin*
Page No. – Page No.
Enantioselective 1,3-DC reactions of
C-Carboxy Ketonitrones and Enals
with MacMillan catalyst: Evidence of
a non-concerted mechanism
Key Topic : Organocatalyzed Cycloaddition
Organocatalyzed 1,3-dipolar cycloaddition between ketonitrones and substituted enals are promoted
by MacMillan’s catalyst with high diastereo- and enantioselectivities. Reaction proceeds with a high
exo selectivity, however, a third diastereomer, not compatible with a concerted mechanism, was
observed. DFT calculations evidence that the cycloaddition proceeds in a non-concerted fashion by a
first oxa-Michael-type addition of the nitrone to the double bond followed by a cyclization. In addition
the diastereo- and enantioselectivities of the reaction were shown to be thermodynamically controlled
and modulated by the kinetics of iminium hydrolysis.
1 M. Berthet, T. Cheviet, G. Dujardin, I. Parrot, J. Martinez, Chem. Rev., 2016, 116, 15235-15283.
2 Reviews on dipolar cycloaddition of nitrones : (a) J. N. Martin, R. C. F. Jones, In Synthetic Applications of 1,3-Dipolar Cycloaddition Chemistry Toward
Heterocycles and Natural Products, (Eds.: A. Padwa, W. H. Pearson), Wiley and Sons: New York, 2002, Chapter 1, pp 1-82; (b) K. V. Gothelf, In Cycloaddition
Reactions in Organic Synthesis (Eds.: S. Kobayashi, K. A. Jørgensen), Wiley-VCH, Weinheim, 2002; Chapter 6, pp 211-248; (c) S. Kanemasa, In Cycloaddition
Reactions in Organic Synthesis (Eds.: S. Kobayashi, K. A. Jørgensen), Wiley-VCH, Weinheim, 2002; Chapter 7, pp 249-300; (d) M. Frederickson Tetrahedron
1997, 53, 403-425; (e) T. B. Nguyen, A. Martel, C. Gaulon, R. Dhal, G. Dujardin, Org. Prep. Proced. Int., 2010, 42, 387-431; (f) T. B. Nguyen, A. Martel, C.
Gaulon-Nourry, R. Dhal, G. Dujardin, Org. Prep. Proced. Int., 2012, 44, 1-81.
3 Reviews on asymmetric 1,3-DC reactions : (a) K. V. Gothelf, K. A. Jorgensen Chem. Rev. 1998, 98, 863-909; (b) H. Pellissier Tetrahedron 2007, 63, 3235-3285;
(c) C. Najera, J. M. Sansano, M. Yus J. Braz. Chem. Soc. 2010, 21, 377-412; (d) M. S. Singh, S. Chowdhury, S. Koley Tetrahedron 2016, 72, 1603-1644.
4 (a) K. B. Selim, A. Martel, M. Y. Laurent, J.Lhoste, S. Py, G. Dujardin J. Org. Chem. 2014, 79, 3414-3426; (b) X. Zhang, P. Cividino, J.-F. Poisson, P. Shpak-
Kraievskyi, M. Y. Laurent, A. Martel, G. Dujardin, S. Py, Org. Lett. 2014, 16, 1936–1939; (c) K. Ben Ayed, A. Beauchard, J.-F. Poisson, S. Py, M. Y. Laurent, A.
Martel, H. Ammar, S. Abid, G. Dujardin Eur. J. Org. Chem. 2014, 2924–2932; (d) P. Shpak-Kraievskyi, A. Mankou Makaya, A. Beauchard, A. Martel, M. Y.
Laurent, G. Dujardin Eur J. Org. Chem. 2015, 3923-3934.
5 W. S. Jen, J. J. M. Wiener., D. W. C. MacMillan, J. Am. Chem. Soc. 2000, 122, 9874–9875.
6 (a) Ł.Weseliński, E. Kalinowska, J. Jurczak Tetrahedron: Asymmetry 2012, 23, 264-270; (b) K. B. Selim, A. Beauchard, J. Lhoste, A. Martel, M. Y. Laurent, G.
Dujardin Tetrahedron: Asymmetry 2012, 23, 1670-1677.
7 For examples of other organocatalyzed 1,3-DC of nitrones with enals: (a) ref (5); (b) S. S. Chow, M. Nevalainen, C. A. Evans, C. W. Johannes Tetrahedron Lett.
2007, 48, 277-280; (c) M. Lemay, J. Trant W. W. Ogilvie Tetrahedron 2007, 63, 11644-11655; (d) S. Karlsson, H.-E. Högberg Tetrahedron: Asymmetry 2002, 13,
923-926; (e) S. Karlsson, H.-E. Högberg Eur. J. Org. Chem. 2003, 2782-2791; (f) Ł. Weseliński, E. Słyk, J. Jurczak Tetrahedron Lett. 2011, 52, 381-384; (g) T.
Otsuki, J. Kumagai, Y. Kohari, Y. Okuyama, E. Kwon, C. Seki, K. Uwai, Y. Mawatari, N. Kobayashi, T. Iwasa, M. Tokiwa, M. Takeshita, A. Maeda, A. Hashimoto,
K. Turuga, H. Nakano Eur. J. Org. Chem. 2015, 7292-7300.
8 T. B. Nguyen, A. Beauseigneur, A. Martel, R. Dhal, M. Laurent, G. Dujardin J. Org. Chem. 2010, 75, 611-620.
9 Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.
A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.
Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2013.
10 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215-241.
11 Although the corresponding transition state could not be found at the M06-2X/6-311+G(d,p) level, it could be obtained at Hartree-Fock level and the geometry
proved to be very close to the geometry of the intermediate 9gA, suggesting that the transition state must be very close on the energy profile and explaining the
difficulties encountered while attempting to locate it.
12 (a) C. Allemann, R. Gordillo, F. R. Clemente, P. H.-Y. Cheong, K. N. Houk, Acc. Chem. Res. 2004, 37, 558-569. (b) R. Gordillo, J. Carter, K. N. Houk, Adv.
Synth. Catal. 2004, 346, 1175-1185.
13 (a) P. James, F.-X. Felpin, Y. Landais, K. Schenk J. Org. Chem. 2005, 70, 7985-7995; (b) A. A. C. van Wijk, J. Lugtenburg Eur. J. Org. Chem. 2002, 4217-
4221.
14 I. A. Vasilenko, D. V. Shamsin, A. V. Tsytovich, A. N. Kan, S. G. Alekseeva, V. I. Shvets Russ. J. Bioorg. Chem. 2004, 30, 547-552.
15 (a) X. Song, X. Zhang, S. Zhang, H. Li, W. Wang Chem. Eur. J. 2012, 18, 9770-9774; (b) B. C. Hong, H. C. Tseng, S. H. Chen. Tetrahedron 2007, 63, 2840-
8250.
This article is protected by copyright. All rights reserved.