Asymmetric Diels-Alder cycloadditions of d-erythrose 1,3-butadienes to achiral t-butyl 2H-azirine 3-carboxylate

Article abstract of DOI:10.1016/j.tetasy.2013.05.015

Two d-erythrose 1,3-butadienes were reacted with electrophilic achiral t-butyl 2H-azirine 3-carboxylate giving cycloadducts with good yields and moderate selectivity. The isomers could be separated to give the major (R)-isomers at C-2 in approximately 50%

Full text of DOI:10.1016/j.tetasy.2013.05.015

Tetrahedron: Asymmetry 24 (2013) 1063–1068
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Asymmetric Diels–Alder cycloadditions of D-erythrose 1,3-butadienes
to achiral t-butyl 2H-azirine 3-carboxylate
⇑
Vera C. M. Duarte, Hélio Faustino, Maria J. Alves , António Gil Fortes, Nuno Micaelo
Departamento de Química, Universidade do Minho, Campus de Gualtar, 4710-057 Braga, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 April 2013
Accepted 21 May 2013
Two D-erythrose 1,3-butadienes were reacted with electrophilic achiral t-butyl 2H-azirine 3-carboxylate
giving cycloadducts with good yields and moderate selectivity. The isomers could be separated to give
the major (R)-isomers at C-2 in approximately 50% yield in both cases. Alternatively LACASA-DA meth-
odology was applied to one of the reactions leading to homochiral (R)- and (S)-products by changing
the chiral nature of an extra chiral BINOL inductor used.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Ph
OH
OH
CO2Me
O
O
O
CO2t-Bu
D-Erythrose is an important chiral synthon used in many syn-
HO
N
2
6
theses.1–5 However, the use of its 1,3-butadiene derivatives 1 is
not common. The lack of interest in these dienes, namely as coun-
terparts in Diels–Alder cycloadditions (DA), is probably due to the
poor facial selectivity associated with DA thermal processes,
although exceptions are known.6,7 Nevertheless, a series of inter-
esting compounds could be envisaged by functional group trans-
formations (FGT) of these DA cycloadducts mainly in the field of
sugars/aza-sugars. Therefore we attempted to better understand
the facial selectivity of dienes 1 with dienophiles in order to im-
prove the selectivity. At first, 5-membered ring dienophiles (malei-
mides and 4-phenyl/methyl-1,2,4-triazoline-3,5-dione) were
examined. The major or exclusive isomer obtained [(S) at C-2]
was formed by the attack of the dienophile at the si face of the
OH
(R)
H
AcHN
OH
OH
4a
3
Figure 1. Structural comparison of neuraminic acid 3 and (2R)-cycloadduct 4a.
the same ratio. Primary cycloadducts obtained by the reaction of
diene 1b lost the t-butyldimethylsilyl group attached to the ery-
throse moiety according to 1H and 13C NMR spectra, thus giving
the free alcohols 4b and 5b (Scheme 1).
The regiochemistry of the cycloaddition results from the attack
of the nucleophilic terminal of the diene to the electrophilic center
in the azirine.8 Such an approach can develop a hydrogen bond
interaction between the nitrogen azirine ring atom of the dieno-
phile and the hydroxyl group at the diene, favoring an attack at
the si face (bottom representation in Fig. 2) over the re face (top
diene.7 Having in mind that the DA cycloadduct of
D-erythrose-
1,3-butadienes to 2H-azirines enables the synthesis of analogue
precursors of neuraminic acid 3, especially if the C-2 (R)-configura-
tion isomers could be formed in reasonable yields, we tested this
C@N dienophile to see if an approach by the re face would be pre-
ferred. Figure 1 represents the close relationship of neuraminic
acid with the (R)-cycloadduct 4a. Functionalization of the C@C
bond would give an epimeric analogue of compound 3 at C-6.
Ph
N
O
O
H
CO2t-Bu
re
(R)
approach
H
N
Ph
O
OH
O
CO2t-Bu
2. Results and discussion
Ph
O
Two
the literature6,7 and combined with t-butyl 2H-azirine 3-carboxyl-
ate, obtained in situ from t-butyl
-azido acrylate;9 the reaction
D-erythrose butadienes 1a,b were prepared according to
si
approach
H
O
O
H
a
(S)
N
occurs at 60 °C. Two products were formed in each reaction in
H
CO2t-Bu
N
OH
CO2t-Bu
⇑
Corresponding author. Tel.: +351 253 604386; fax: +351 253 604 382.
E-mail address: mja@quimica.uminho.pt (M.J. Alves).
Figure 2. The endo facial approaches of 2H-azirine 2 to diene 1.
0957-4166/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetasy.2013.05.015

1064
V. C. M. Duarte et al. / Tetrahedron: Asymmetry 24 (2013) 1063–1068
Ph
O
O
OR2
R1
Ph
2'
Ph
1'
O
1a,R1=H, R2=H
R1
R1
O
O
O
H
H
CO2t-Bu
N3
N
1b,R1=OTBS, R2=TBS
toluene
R
1N
S
+
CO2t-Bu
H
H
N
reflux, 90 min.
60 ºC, 3h
6
OH
OH
CO2t-Bu
CO2t-Bu
2
4a, R1= H, 47%
5a, R1= H, 27%a)
, R1= OTBS, 46 %
5b, R1= OTBS, 32%
4b
a) contaminated with 4a
Scheme 1. Thermal cycloaddition of 2H-azirine 2 to diene 1.
representation depicted in Fig. 2). However the stereochemistry of
the products is incompatible with such reagent interactions: an
approximate (R):(S) ratio of 2.5:1 was obtained in the thermal
reaction of diene 1a to azirine 2 (Scheme 1).
der Waals’ distance were explored by using aromatic rings with a
wide range of substituents in the triptycene scaffold.10 Many
examples of such a positive interaction have also been reported be-
tween the oxygen atom of carbonyls and the aromatic groups in
proteins, within an average of 3.5 Å distance.11 In the si-approach,
the lone pair of the nitrogen is too far from the phenyl group due to
the different directions of attack. The (R)-configuration product
would also be favored from a kinetic profile point of view.
On the other hand, the conformational free energy difference
between the (R)- and (S)-diastereomers using computational free
energy perturbation methods was determined to be ꢀ8.9 kcal/
mol, meaning that the (S)-product is thermodynamically preferred
relative to the (R)-product. The isomeric ratio of products 4a/5a
probably reflects the major importance of kinetics over thermody-
namics in this cycloaddition.
The identification of the products (oils) was based on conforma-
tional analysis using molecular dynamics simulations. The two
lowest free energy conformers of the (R)-diastereomer (structures
A and B), and the (S)-diastereomer (structures C and D) were ob-
tained. In every case, the hydroxyl group establishes an intramo-
lecular hydrogen bond within O-1 in the dioxanyl moiety (Fig. 3).
This obviously predicts the same O–H–O interaction trend to occur
in the diene reagent, leaving no room for a possible hydrogen bond
with the nitrogen atom of the approaching azirine that would favor
the C-2 (S)-configuration product predicted by the approach out-
lined in Figure 2.
Figure 4 shows an alternative interaction of the approaching
azirine to the re/si faces of the diene. The re approach shows a
proximity between the nitrogen lone pair of the azirine and the
The identification of isomers 4a/5a was made by measuring the
dihedral angle between the two six-membered ring moieties in the
more stable conformers A, B [(R)-isomer], and C, D [(S)-isomer].
(Fig. 3) Both A and B display small dihedral angles: A 68°, B 47°;
p
current of the aromatic ring. Attractive interactions at the van
Figure 3. Lowest free energy conformations of (R)-4a/(S)-5a-diastereomers. (R)-Diastereomer: conformations A and B; (S)-diastereomer: conformations C and D. Structures
are rendered in ball-and-stick mode with carbon in green, oxygen in red, nitrogen in blue, and hydrogen in white. Intra-molecular interactions and distances are highlighted
with a dashed yellow line.

V. C. M. Duarte et al. / Tetrahedron: Asymmetry 24 (2013) 1063–1068
1065
CO2t-Bu
due to the lack of force field parameters for silicon, the t-butyldi-
methylsilyl group was replaced with t-butyl. The same dihedral an-
gle/coupling constant trend was obtained for compounds 4b/5b.
The (R)-isomer shows 99% of its population to be in the E (89%)
and F conformers (10%). The dihedral angles H–C2–C4–H in the
two conformers are both small: 53° E and 61° F, which is consis-
tent with the small coupling constant of H-2 to H-40 (2.2 Hz) in
the 1H NMR spectrum. (Fig. 5) The (S)-isomer showed 94% of its
population to be in the G conformer (74%) and H conformer
(20%). The dihedral angles of these conformers are both large an-
gles: 167° G and 162° H, which is consistent with the large cou-
pling constant of H-2 to H-40 (10.1 Hz). (Fig. 5).
N
H
O
O
(R)
O
O
OH
N
OH
CO2t-Bu
attack on the re-face
favored approach
N
With the aim of enhancing the formation of compound 4a ver-
sus 5a, we thought that a coordination reaction might induce
kinetics to completely control the process. Previous experience
with a bimetallic complex of Mg(II) and Zn(II) having (R)-/(S)-BI-
NOL as a chiral ligand was applied in a Diels–Alder cycloaddition
based on a Lewis acid-catalyzed reaction of a ‘self-assembled’ com-
plex (LACASA-DA). This method involves with the existence of a
free hydroxyl group in the diene and a carbonyl in the dienophile
to coordinate all reagents and led to very good facial selectivities.
2,4-Pentadienol was combined with nitroso dienophiles,12,13 and
methyl acrylate14,15 in the presence of tartaric acid or BINOL with
excellent selectivities. We have previously reported on some con-
trol in reactions of diene 1a with maleimides.16
Combining diene 1a with azirine 2 at low temperatures gave
homochiral compounds 4a or 5a depending on the stereochemistry
of the BINOL (Scheme 2). Compound 5a was obtained in 47% yield
in the presence of (R)-BINOL, and 4a in 28% yield in the presence of
(S)-BINOL. The excess reacting diene remained untouched accord-
ing to 1H NMR spectra, even after the addition of fresh portions of
azirine or prolonged reaction times.
CO2t-Bu
H
O
O
(S)
O
O
OH
N
OH
CO2t-Bu
attack on the si-face
Figure 4. The two possible endo approaches of 2H-azirine 2 to diene 1a.
these values are in accordance with the small coupling constant
shown in the 1H NMR spectrum (J ꢁ2 Hz) for the (R)-diastereomer.
Considering that conformers A and B make up 98% of the total pop-
ulation in chloroform an obvious relationship of the (R)-isomer
with the small coupling constant of H-2 to H-40 occurs. In contrast,
C and D show larger dihedral angles: C 174°, D 176°, and so a large
coupling constant of H-2 to H-40 is predicted. In fact, the J is ca
10 Hz in the other (S)-isomer. The C and D population corresponds
to 91% of the total conformation species solubilized in chloroform
with 63% and 28%, respectively. An equivalent computational con-
formational analysis of compounds 4b/5b was also made, however
Figure 5. Lowest free energy conformations of the (R)-4b/(S)-5b diastereomers. (R)-Diastereomer: conformations E and F; (S)-diastereomer: conformations G and H.
Structures as rendered in ball-and-stick mode with carbon in green, oxygen in red, nitrogen in blue, and hydrogen in white. Intra-molecular interactions and distances are
highlighted with a dashed yellow line.

1066
V. C. M. Duarte et al. / Tetrahedron: Asymmetry 24 (2013) 1063–1068
a solution of diene 1a (0.041 g; 0.177 mmol) in a 1:1 mixture
of toluene/DCM (2 mL) was added. The mixture was main-
tained at 60 °C for 3 h. The solvent was evaporated until
dryness and the crude subjected to flash chromatography in
a 1:4 mixture of petroleum ether/ethyl ether. Two isomeric
oily products were obtained. The (S)-isomer 5a, (0.018 g;
0.048 mmol; 27%) was contaminated with the (R)-isomer 4a,
and (R)-4a was obtained pure (0.031 g; 0.083 mmol; 47%).
Isomer (R)-4a: ½a D20
¼ ꢀ14:9 (c 1.5, CH2Cl2); mmax (neat) 3368,
ꢂ
2977, 1723, 1158 cmꢀ1; dH (400 MHz, CDCl3) 7.42–7.45 (m, 2H,
Ho, Ph), 7.33–7.35 (m, 3H, Hm and Hp, Ph), 5.78 (ddt, J 10.4, 5.4,
3.4 Hz, 1H, H-4), 5.46–5.54 (m, 1H, H-3), 5.44 (s, 1H, H-20), 4.38-
4.50 (m, 1H, H-50), 4.33 (dd, J 10.6, 5.3 Hz, 1H, H-60), 4.20 (d, J
2.0 Hz, 1H, H-2), 3.77 (dd, J 9.3, 3.1 Hz, 1H, H-40), 3.67 (t, J
10.4 Hz, 1H, H-60), 2.65 (t, J 4.0 Hz, 2H, H-5), 2.36 (s, 1H, H-7),
2.06 (s, 1H, H-7), 1.48 (s, 9H, 3ꢃCH3) ppm; dC (100 MHz, CDCl3)
172.0 (C@O), 138.1 (Cq, Ph), 128.8 (Cp, Ph), 128.1 (Cm, Ph), 126.1
(Co, Ph), 123.6 (C-4), 123.5 (C-3), 101.4 (C-20), 85.0 (C-40), 81.6
(Cq, t-Bu), 81.5 (C-6), 71.1 (C-60), 61.0 (C-50), 53.0 (C-2), 36.8 (C-
6), 30.9 (C-7), 28.0 (CH3), 27.9 (CH3), 27.8 (CH3), 22.4 (C-5) ppm;
HRMS (ESI): calculated for C21H28NO5: 374.1960; found: 374.1962.
Scheme 2. The LACASA-Diels–Alder cycloaddition between diene 1a and 2H-
azirine 2 tethered in a bimetallic complex of Mg(II) and Zn(II) with (R)-/(S)-BINOL as
a chiral ligand.
3. Conclusion
D
-Erythrose 1,3-butadienes were combined with t-butyl 2H-azi-
rine 3-carboxylate. Thermal reactions occur at 60 °C with moderate
facial selectivity, favoring the (R)-enantiomer at C-2. The approach
of the reagents has been inversed relative to other cases in the liter-
ature. The two (R)-products were separated to give the major isomer
in 46% and 47% yield in cases a and b respectively. The moderate
selectivity was explained by the antagonism of thermodynamics
and kinetics of the cycloadditions. An LACASA-DA methodology
using Zn/Mg bimetallic complexes and BINOL as an extra chiral
inductor gave pure (R)- and (S)-products. A significant improvement
in the yield was achieved in the case of the (S)-product but not in the
(R)-product. The relative configuration of the new stereocenter in
the products was obtained by combining the results of the confor-
mational analysis with the 1H NMR coupling constants.
Isomer (S) 5a: ½a D20
¼ ꢀ11:2 (c 1.05, CHCl3); mmax (neat) 3366,
ꢂ
2977, 1724, 1155 cmꢀ1; dH (400 MHz, CDCl3) 7.47–7.50 (m, 2H,
Ho, Ph), 7.35–7.38 (m, 3H, Hm+Hp, Ph), 5.75–5.84 (m, 2H, H-
3+H-4), 5.51 (s, 1H, H-20), 4.38 (dd, J 10.8, 5.2 Hz, 1H, H-60), 4.03
(ddd, J 10.1, 8.7, 5.2 Hz, 1H, H-50), 3.96 (d, J 10.0 Hz, 1H, H-2),
3.69 (t, J 10.8 Hz, 1H, H-60), 3.46 (dd, J 9.6, 8.4 Hz, 1H, H-40), 2.67
(dd, J 4.4, 1.6 Hz, 2H, H-5), 2.22 (s, 1H, H-7), 2.00 (s, 1H, H-7),
1.47 (s, 9H, 3ꢃCH3) ppm; dC (100 MHz, CDCl3) 170.6 (C@O),
137.6 (Cq, Ph), 128.9 (Cp, Ph), 128.2 (Cm, Ph), 126.0 (Co, Ph),
123.8 (C-3 or C-4), 120.8 (C-3 or C-4), 101.0 (C-20), 81.6 (Cq, t-
Bu), 80.4 (C-40), 70.5 (C-60), 68.1 (C-50), 59.9 (C-2), 39.4 (C-6),
27.9 (3ꢃCH3; C-7), 22.3 (C-5) ppm. HRMS (ESI): calculated for
4. Experimental
4.1. General
C
21H28NO5: 374.1952; found: 374.1962.
4.2.1.2. 2-LACASA-DA methodology
Starting with D-erythrose 1,3-butadienes 1a,b were obtained
from (2R,4R,5R)-5-hydroxy-2-phenyl-1,3-dioxane-4-carbaldehyde
according to the literature.6,7 2H-Azirine was prepared in situ from
Solution A. A solution of diene 1a (0.100 g; 0.43 mmol) in dry tol-
uene (2.2 mL) was added to a solution of Me2Zn (1.2 M) in toluene
(359 lL; 0.43 mmol) at 0 °C, and stirred for 5 min.
a
-azido t-butyl acrylate.9 All other reagents were purchased and
used without further purification. Solvents employed in reactions
were dried: CH2Cl2 was freshly distilled under CaH2, and toluene
was submitted to simple distillation to remove the head fraction.
The petroleum ether 40–60 °C used in flash chromatography was
previously distilled, while all other solvents were used as pur-
chased. Glassware was dried prior to use. Compounds were puri-
fied by dry flash chromatography, using silica 60 <0.063 mm as
the stationary phase and water pump vacuum. TLC plates (Silica
Gel 60 F254, Macherey–Nagel) were visualized either at a UV lamp
or in I2. 1H NMR and 13C NMR were run on a Varian Unity Plus 300,
or Brucker Avance III 400 or Bruker BioSpin GmbH spectrometers.
Infrared spectra were recorded on a Bomem MB 104 or on a Per-
kin–Elmer spectrophotometer. Samples were run as nujol mulls
and oils as thin films. MS spectra were recorded on a VG Autospec
M. spectrometer.
Solution B. A solution of (S)-BINOL (0.123 g; 0.43 mmol) in dry
toluene (2.2 mL) was added to a solution of MeMgBr (1.4 M) in tol-
uene/THF (307 lL; 0.43 mmol) at 0 °C, and stirred for 5 min.
Solution A was added to solution B, diluted with dry toluene
(3.6 mL), and stirred for 5 min. This mixture was cooled to ꢀ78 °C
and
a solution of t-butyl 2H-azirine-3-carboxylate (0.061 g;
0.43 mmol) in dry toluene (3 mL) was then added. The reaction
was stirred at ꢀ20 °C for 24 h. A new portion of t-butyl 2H-azirine-
3-carboxylate was then added and the reaction was stirred at rt
for 5 days. The reaction was quenched with NaHCO3 aq satd sol.
(1 mL), filtered through a pad of CeliteÒ, and the CeliteÒ was washed
with EtOAc (4 ꢃ 10 mL). The filtrates were combined and concen-
trated under reduced pressure to give an orange oil corresponding
to a 2:1 mixture of diene starting material and the expected product.
The crude was submitted to ‘dry-flash’ chromatography using a mix-
ture of PE (40–60)-Et2O. The (S)-BINOL was recovered from PE–Et2O
1:1 (0.069 g; 56%) and the product was eluted with PE/Et2O 2:3 to
give the (R)-isomer as an oil (0.044 g; 28%).
4.2. Reaction of
3-carboxylate 2
L-erythrose diene 1a 1b to t-butyl 2H-azirine
4.2.1. Synthesis of cycloaducts 4a and 5a
4.2.1.1. Thermal method. The
-azido acrylate9 (0.151 g;
a
0.893 mmol) was dissolved in toluene (16 mL) and refluxed
under a nitrogen atmosphere for 90 min. The heating source
was then removed and when the temperature reached 60 °C,
4.2.2. Synthesis of cycloadducts 4b and 5b
4.2.2.1. Thermal method. The t-butyl
a-azido acrylate (0.340 g;
2.01 mmol) was dissolved in toluene (30 mL) and refluxed under

V. C. M. Duarte et al. / Tetrahedron: Asymmetry 24 (2013) 1063–1068
1067
a nitrogen atmosphere for 90 min. The heating source was re-
moved and when the reaction mixture temperature reached
60 °C, a solution of diene 1b (0.184 g; 0.386 mmol) in DCM
(4 mL) was added. The mixture was maintained at 60 °C for 5 h.
A second portion of azirine was prepared (0.180 g; 1.06 mmol) in
dry toluene (18 mL) and added to the reaction mixture and stirred
further at 60 °C for 1.5 h. The solvent was evaporated until dryness,
after which flash chromatography was carried out using petroleum
ether/ethyl ether 30% as the eluent. Two isomeric compounds were
obtained in 78% overall yield, and separated as oils: (S)-isomer 5b
(0.063 g; 0.125 mmol; 32%) and (R)-isomer 4b (0.090 g;
0.179 mmol; 46%).
(40–60)–Et2O. The (R)-BINOL was recovered from PE–Et2O 3:1
(0.076 g; 62%) and the product was eluted with PE–Et2O 1:1 to give
the (S)-isomer as an oil (0.076 g; 47%).
5. Materials and methods
Molecular dynamic (MD) simulations were performed with the
GROMACS package version 4.5.4 using an AMBER03/GAFF force
field. The geometry of each azirine enantiomer was optimized with
a quantum mechanical method at the Hartree–Fock level with the
6-31G(d) basis set using GAMESS. This optimized geometry and
the quantum mechanical electrostatic potential were used to cal-
culate the partial atomic charges using the RESP fitting method.
The azirine molecules were solvated with 500 chloroform solvent
molecules in a cubic box with dimension of 4.4 ꢃ 4.4 ꢃ 4.4 nm.
The equations of motion were numerically integrated using a time-
step of 2 fs. The nonbonded interactions were treated with a cut off
of 0.9 nm and updated every 5 steps. Long-range electrostatic
interactions were treated with particle-mesh-ewald (PME) using
a fourier spacing of 1.2 nm and a PME order of 4. The system
was simulated in an isothermal and isochoric ensemble using the
Berendsen thermostat at 300 K and relaxation time of 0.1 ps. The
pressure coupling was also accomplished with the Berendsen bar-
ostat with a relaxation time of 0.5 ps and isothermal compressibil-
ity of 4.5 ꢃ 10ꢀ5 bar. All bonds were constrained using the LINCS
algorithm. An initial energy minimization consisting of 5000 steps
was performed using the steepest descent algorithm. We per-
formed one long MD simulation of 400 ns for each system. Confor-
mations were recorded every 1 ps.
Isomer (R)-4b-viscous oil. ½a D20
¼ ꢀ40:5 (c 1.05, CHCl3); mmax
ꢂ
3746, 3682, 3093, 3067, 3037, 3005, 1727, 1678, 1164, 1135,
1090, 1030, 841 cmꢀ1 1H NMR (800 MHz, CDCl3) d 7.47–7.42 (m,
;
2H, CH, Ph), 7.38–7.31 (m, 3H, CH, Ph), 5.43 (s, 1H, H-20), 4.58 (br
t, J 2.2 Hz, 1H, H-3), 4.41 (td, J 9.8, 5.4 Hz, 1H, H-50), 4.33–4.37
(m, 2H, H-60+H-2), 3.70–3.66 (m, 2H, H-40+H-60), 2.74 (d,
J
17.7 Hz, 1H, H-5), 2.64 (br s, 1H, OH), 2.54 (d, J 17.8 Hz, 1H, H-5),
2.30 (s, 1H, H-7), 2.06 (s, 1H, H-7), 1.50 (s, 9H, OCCH3), 0.92 (s,
9H, SiCCH3), 0.12 (s, 3H, SiCH3), 0.10 (s, 3H, SiCH3) ppm; 13C NMR
(100 MHz, CDCl3) d 171.6 (C@O), 147.3 (C-4), 138.1 (Cq, Ph),
128.7 (CH, Ph), 128.1 (CH, Ph), 126.0 (CH, Ph), 101.3 (C-20), 99.2
(C-3), 84.9 (C-40), 81.7 (Cq), 71.1 (C-60), 61.3 (C-50), 54.0 (C-2),
38.4 (Cq, t-Bu), 30.9 (C-7), 28.0 (C-CH3), 27.7 (C-5), 25.6
(SiC(CH3)3), 18.0 (SiC(CH3)3), ꢀ4.4 (SiCH3), ꢀ4.6 (SiCH3) ppm;
HRMS (FAB): calculated for
504.2786.
C27H42NO6Si: 504.2781; found:
Isomer (S)-5b: viscous oil. ½a D20
ꢂ
¼ ꢀ4:0 (c 1.8, CHCl3);
mmax (neat)
3566, 3554, 3092, 3068, 3036, 1727 cmꢀ1
;
1H NMR (400 MHz,
From these long MD simulations the multidimensional free en-
ergy landscape of each azirine enantiomer was calculated. The
multidimensional free energy landscape was calculated using a
principal component analysis (PCA) approach based on the struc-
tural dissimilarity of all pairs of conformations recorded. The pro-
tocol for the PCA calculations and identification of the lowest free
energy conformations used herein was the same as described by
Sara et al.17
CDCl3) d 7.50–7.45 (m, 2H, C-H, Ph), 7.40–7.33 (m, 3H, C-H, Ph),
5.52 (s, 1H, H-20), 4.91–4.84 (m, 1H, H-3), 4.38 (dd, J 10.8, 5.2 Hz,
1H, H-60), 4.07 (d, J 9.7 Hz, 1H, H-50), 4.00 (ddd, J 10.1, 8.7, 5.2 Hz,
1H, H-2), 3.68 (t, J 10.8, 1H, H-60), 3.42 (dd, J 9.5, 8.8 Hz, 1H, H-
40), 2.77 (dtd, J 17.6, 2.5, 1.1 Hz, 1H, H-5), 2.56 (dd, J 17.6, 0.7 Hz,
1H, H-5), 2.21 (s, 1H, H-7), 1.99 (s, 1H, H-7), 1.47 (s, 9H, OC(CH3)3),
0.93 (s, 9H, SiC(CH3)3), 0.16 (s, 3H, SiCH3), 0.15 (s, 3H, SiCH3) ppm;
13C NMR (100 MHz, CDCl3) d 170.4 (C@O), 148.1 (C-4), 137.6 (Cq,
Ph), 128.8 (CH, Ph), 128.2 (CH, Ph), 125.8 (CH, Ph), 100.7 (C-20),
95.8 (C-3), 81.8 (Cq), 81.2 (C-40), 70.4 (C-60), 68.0 (C-2), 60.4 (C-
50), 40.9 (Cq, t-Bu), 28.2 (C-7), 28.1 (C-5), 27.9 (C-CH3), 25.6
(SiC(CH3)3), 18.0 (SiC(CH3)3), ꢀ4.4 (SiCH3), ꢀ4.5 (SiCH3) ppm;
HRMS (FAB): calculated for C27H42NO6Si: 504.2781; found:
504.2779.
Acknowledgments
We thank FCT for project funding PTDC/QUI/67407/2006, for
financial support to the NMR portuguese network (PTNMR, Bruker
Avance III 400-Univ. Minho); FCT and FEDER (European Fund for
Regional Development)-COMPETE-QREN-EU for financial support
to the Research Centre, CQ/UM [PEst-C/QUI/UI0686/2011
(FCOMP-01-0124-FEDER-022716)]. N.M. acknowledges the con-
tract research program ‘Compromisso com a Ciência’ reference:
C2008-UMINHO-CQ-03 and access to the Minho University GRIUM
cluster. V.C.M.D. also thanks for PhD grant (SFRH/BD/61290/2009).
4.2.2.2. LACASA-DA methodology
Solution A. A solution of diene 1a (0.100 g; 0.43 mmol) in dry tol-
uene (2.2 mL) was added to a solution of Me2Zn (1.2 M) in toluene
(359 lL; 0.43 mmol) at 0 °C, and stirred for 5 min.
References
Solution B. A solution of (R)-BINOL (0.123 g; 0.43 mmol) in dry
toluene (2.2 mL) was added to a solution of MeMgBr (1.4 M) in tol-
uene/THF (307 lL; 0.43 mmol) at 0 °C, and stirred for 5 min.
Solution A was added to solution B, diluted with dry toluene
(3.6 mL), and stirred for 5 min. This mixture was cooled to
1. Ruiz, M.; Ruanova, T. M.; Blanco, O.; Nunez, F.; Pato, C.; Ojea, V. J. Org. Chem.
2008, 73, 2240–2255.
2. Ruiz, M.; Ojea, V.; Ruanova, T. M.; Quintela, J. M. Tetrahedron: Asymmetry 2002,
13, 795–799.
3. Pino-Gonzaléz, M. S.; Oña, N. Tetrahedron: Asymmetry 2008, 19, 721–729.
4. Hudlicky, T.; Luna, J. H.; Price, J. D.; Rulin, F. J. Org. Chem. 1990, 55, 4683–4687.
5. Fujita, T.; Nagasawa, H.; Uto, Y.; Hashimoto, T.; Asakawa, Y.; Hori, H. Org. Lett.
2004, 6, 827–830.
6. Mukhopadhyay, A.; Ali, S. M.; Husain, M.; Suryawanshi, S. N.; Bhakuni, D. S.
Tetrahedron Lett. 1989, 30, 1853–1856.
7. Alves, M. J.; Duarte, V. C. M.; Faustino, H.; Gil Fortes, A. Tetrahedron: Asymmetry
2010, 21, 1817–1820.
8. Alves, M. J.; Gilchrist, T. L. J. Chem. Soc., Perkin Trans. 1 1998, 299–303.
9. Alves, M. J.; Gilchrist, T. L. Tetrahedron Lett. 1998, 39, 7579–7582.
10. Gung, B. W.; Zou, Y.; Xu, Z.; Amicangelo, J. C.; Irwin, D. G.; Ma, S.; Zhou, H.-C. J.
Org. Chem. 2008, 73, 689–693.
ꢀ78 °C and
a solution of t-butyl 2H-azirine-3-carboxylate
(0.061 g; 0.43 mmol) in dry toluene (3 mL) was added. After mix-
ing the reagents, the temperature was allowed to rise gradually
to rt. The reaction was quenched with NaHCO3 aq. satd. sol.
(1 mL), filtered through a pad of CeliteÒ, and the CeliteÒ washed
with EtOAc (4 ꢃ 10 mL). The filtrates were combined and concen-
trated under reduced pressure to give an orange oil, consisting of
a 1:1 mixture of starting diene and product. The crude was
submitted to ‘dry-flash’ chromatography using a mixture of PE

1068
V. C. M. Duarte et al. / Tetrahedron: Asymmetry 24 (2013) 1063–1068
11. Jain, A.; Purohit, C. S.; Verma, S.; Sankararamakrishnan, R. J. Phys. Chem. B 2007,
111, 8680–8683.
12. Ding, X.; Ukaji, Y.; Fujinami, S.; Inomata, K. Chem. Lett. 2003, 32, 582–583.
13. Ukaji, Y.; Inomata, K. Synlett 2003, 1075–1087.
15. Ward, D. E.; Mohammad, S. A. Org. Lett. 2000, 3937–3940.
16. Salgueiro, D. A. L.; Duarte, V. C. M.; Sousa, C. E. A.; Alves, M. J.; Gil Fortes, A.
Synlett 2012, 1765–1768.
17. Campos, S. R. R.; Baptista, A. M. J. Phys. Chem. B 2009, 113, 15989–16001.
14. Ward, D. E.; Souweha, M. S. Org. Lett. 2005, 3533–3536.