Diastereo-controlled DielsAlder cycloadditions of erythrose benzylidene-acetal 1,3-butadienes by 4-substituted-1,2,4-triazoline-3,5-dione: Evidence for the stereoelectronic effects on the dienes

Article abstract of DOI:10.1016/j.tetasy.2010.05.001

Erythrose benzylidene-acetal 1,3-butadienes are studied as partners in DielsAlder cycloadditions. A high diastereofacial improvement is found in cases where both the alcohol function is protected and a π π interaction between the diene and dienophile is possible. Several competing factors have been studied independently in order to explain its influence on the selectivity of the cycloadditions.

Full text of DOI:10.1016/j.tetasy.2010.05.001

Tetrahedron: Asymmetry 21 (2010) 1817–1820
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Diastereo-controlled Diels–Alder cycloadditions of erythrose benzylidene-acetal
1,3-butadienes by 4-substituted-1,2,4-triazoline-3,5-dione: Evidence for the
stereoelectronic effects on the dienes
*
Maria J. Alves , Vera C. M. Duarte, Hélio Faustino, António Gil Fortes
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:
Erythrose benzylidene-acetal 1,3-butadienes are studied as partners in Diels–Alder cycloadditions. A high
Received 23 November 2009
Revised 30 April 2010
Accepted 4 May 2010
Available online 30 June 2010
diastereofacial improvement is found in cases where both the alcohol function is protected and a
p–p
interaction between the diene and dienophile is possible. Several competing factors have been studied
independently in order to explain its influence on the selectivity of the cycloadditions.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
group with MeI in the presence of NaH. Compound 1d was also
obtained from aldehyde 2, by the intermediacy of ketone 37
Glucosyl-type dienes are excellent partners in Diels–Alder cyc-
loadditions, showing high diastereoselectivity to a range of cyclic
or acyclic carba-1 and aza-2,3dienophiles. As part of a project aimed
at the synthesis of C-aza-disaccharides we looked at the usefulness
of 1,3-butadienes bearing an erythrose benzylidene-acetal moiety
(R = Me) that was isolated and then converted into 1d by addition
of TBSOTf (2.2 equiv) in the presence of TEA (2.2 equiv) in quanti-
tative yield (Scheme 1). The literature preparation of 1a requires a
non-environmentally friendly synthesis by an oxymercuration–
demercuration method.4,8 Diene 1c had been obtained previously
by methylation from 1a.4 The other dienes 1b and 1d are, to the
best of our knowledge, new compounds and were fully character-
as a chiral inducer in 4p + 2p cycloadditions. There was a single
earlier investigation in this area: dienes of type 1 were combined
with 2-methoxycarbonyl-p-benzoquinones to build up the AB ring
skeleton of Forskolin with high diastereocontrol.4 Having decided
to combine dienes 1 with our target aza-dienophiles, 4-phenyl-
1,2,4-triazoline-3,5-dione (PTAD) and 4-methyl-1,2,4-triazoline-
3,5-dione (MTAD), it became apparent that the cycloaddition
selectivities would not be totally controlled in all cases. Based on
X-ray structure determinations of the cycloadducts we gained an
understanding of the facial selectivity of dienes 1.
0
0
ized. The typical trans coupling constant, J1 ,2 , was verified in every
case: diene 1a showed J = 14.8 Hz; diene 1b J = 15.3 Hz and so
diene 1c J = 14.6 Hz and diene 1d J = 15.4 Hz.
3. Cycloadditions of dienes 1a–d to PTAD and MTAD
PTAD reacted at ꢀ78 °C with the four dienes in dichlorometh-
ane. Products consisted of mixtures of diastereomers 4 and 5 when
the alcohol function was unprotected, or diastereomerically pure
compounds 4 when the alcohol function was protected (Scheme 2).
MTDA (4-methyl-3H-1,2,4-triazole-3,5(4H)-dione) was used in
reactions with dienes 1a and 1c keeping the reaction conditions
observed before with PTAD. Products consisted of mixtures of dia-
stereomers 6a and 7a in 2:1 isomeric ratio, when the alcohol func-
tion was unprotected, or in 4.5:1 ratio when the alcohol function
was protected with a methyl group giving 6a/7c (Scheme 2). The
closer ratio of isomers in reaction of diene 1a with MTAD com-
pared with the reaction with PTAD can be accounted for by the
2. Synthesis of 1,3-butadienes 1a–d
1,3-Butadienes 1a–c were easily prepared from the aldehyde 2,
which was available in a two-step procedure from
Butadiene 1a was obtained from 2 by a double Wittig reaction, in
one-pot procedure, by adding successively Ph3PCHCHO
D
-glucose.5,6
a
(1.1 equiv) under anhydrous PTSA catalysis in dry THF followed
t
by Ph3PCH3Br and BuLi. The enal intermediate 3 (R = H) was iso-
lated as a single isomer, its configuration around C-10/C-20 was as-
signed as trans, according to H-10–H-20 coupling constant
p–p interaction between the diene 1a with PTAD in the transition
0
0
(J1 ,2 = 15.8 Hz). Butadiene 1b was analogously prepared from 3,
by choosing the Bu3P@CHCO2Et reagent in the second step. Com-
pound 1c was obtained from 1a by methylation of the hydroxyl
state. The better ratio of isomers in the reaction of diene 1c with
MTAD (4.5:1) compared with reaction of 1a with MTAD shows
the importance of the stereoelectronic effect of the alcohol protec-
tion group. Both reactions of dienes 1a and 1c with MTAD demon-
strated that the charge transfer between phenyl groups is
important in Diels–Alder reactions of erythrose dienes with PTAD,
* Corresponding author. Tel.: +351253604376.
E-mail address: mja@quimica.uminho.pt (M.J. Alves).
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.05.001

1818
M. J. Alves et al. / Tetrahedron: Asymmetry 21 (2010) 1817–1820
Ph
O
O
OTBS
OTBS
vi)
1d, quant.
Ph
Ph
Ph
R=Me
Ph
i) or ii)
iii)
O
O
O
O
O
O
iv)
O
O
O
R
R=H
OH
OH
O
OH
OMe
1c, quant.
3
2
1a, 37 %
Ph
O
R=H
v)
O
OH
CO2Et
1b, 70 %
Scheme 1.
Ph
Ph
O
Ph
Ph
O
O
O
O
N
N
O
O
O
O
O
N
N
Ph N
O
2'
(S)
2'
(S)
H
H
H
Ph
N Me
O
4
4
5
H
R1
R1
R1
R2
5
R1
R2
+
+
3'
3'
H
H
(R)
N
H
OR3
O
(R)
H
OR3
O
O
O
OR3
O
N
OR3
O
N
N
R2
N
R2
R1
R2
N
N
N
N
N
N
N
OR3
O
O
O
O
Ph
Me
Ph
Me
4
5
6
7
isomers
d.r.
5:1
yield
isomers
d.r.
yield
1a, R1=R2=R3=H
1b, R1=H, R2=CO2Et, R3=H
1c, R1=R2=H, R3=Me
4a: 5a
4b: 5b
4c
(66 % + 16 %)a)
1.6:1 (27 % + 17 %)a)
6a: 7a
6c: 7c
2:1
4.5:1
(41 % + 32 %)a)
(53 %)b)c)
1d, R1=OTBS, R2=H, R3=TBS
----
----
67 %
61 %
a) both isomers were isolated;
b) total yield of both isomers after chromatography 73%
4d
c) isolated isomer 6c (53%)
a) both isomers were isolated
Scheme 2.
N
N
O
Ph
O
N
N
Ph
O
O
N
N
Ph
H
O
O
Ph
H
N
O
O
Ph
O
O
O
O
N
N
O
O
N
N
MeO
H
OMe
MeO
OMe
O
N
1c
O
1c
Not observed
4c
attack on the si face
favored approach
Ph
attack on the re face
disfavored approach
Observed
Figure 1.
since the isomeric ratio is substantially diminished in reactions
performed with MTAD. The results are summarized in Scheme 2.
To further rationalize the facial discrimination of the cycloaddi-
tion process we devised an experiment to test a possible hydrogen-
bonding coordination between the hydroxy group of the diene and
any of the polar atoms/groups in the dienophile. Such intermolec-
ular hydrogen-bonding has been described to influence the out-
come of the selectivity, both in Diels–Alder9,10 and in dipolar
cycloadditions.11 The solvation of the hydroxy group by DMF is
expected to interrupt the possible intermolecular hydrogen-bond-
ing in the approaching reagents. Diene 1a was reacted with PTAD
at ꢀ78 °C in dry DMF to determine any ratio difference of isomers
4a:5a. The same 5:1 ratio of isomers 4a:5a was obtained as previ-
ously described in dichloromethane. Clearly, no hydrogen-bonding
with the naked hydroxy group in compound 1a is responsible for
favoring the attack on the re face in cases c and d (see Fig. 1).
According to ChemBio 3D software a 180° dihedral angle is to be
expected for H-10–C10–C-4–H-4 moiety of the molecule, both in
(R)- and (S)-configurations. 1H NMR spectra, however, show in
some cases an anti disposition of H-10–H-4 and in other cases the
syn conformation is preferred. Comparison of 1H NMR coupling
constants between H-1 and H-4 for adducts showed values
between 9.3 and 9.8 Hz in both 4b, 5b and 4d (related to the anti
disposition) and in both 4a, 5a and in 4c (related to the syn dispo-
sition). The lack of consistency between the examples observed
clearly means a very narrow energy gap between the anti and
syn conformers. X-ray analysis, therefore, becomes the only tool
to solve the absolute configuration at C-10.
Fortunately, compound 4c could be crystallized and its struc-
ture determined by X-ray crystallography; C-10 was found to have
the (S)-configuration. Also 4a, the major component of the mixture
4a and 5a, was crystallized and the stereochemistry was found to

M. J. Alves et al. / Tetrahedron: Asymmetry 21 (2010) 1817–1820
1819
Figure 2. ORTEP view of the molecular structure of the compound 4a.
Figure 4. ORTEP view of the molecular structure of the compound 6a.
charge transfer can happen between the two phenyl groups of re-
agents on the left side of Figure 1.
Of course the observed structure 4 obtained by X-ray proves
neither an endo or an exo approach; the easy inversion of the nitro-
gen bridge atom configuration makes such a distinction not possi-
ble. The endo approach of reagents is suggested on the basis of the
stereochemistry of the adducts in the cases of 2-methoxycarbonyl-
p-benzoquinones with dienes of type 1 previously reported,4 and
by the fact that endo approaches had been suggested for PTAD to
other glucosyl-bond dienes.1,2 X-ray structures of compounds 4a
and 4c are presented in Figures 2 and 3. X-ray structure of com-
pound 6a is represented in Figure 4.
Assignments of isomers 4 and 5 were based on 1H NMR spectro-
scopic analysis. Chemical shifts and coupling constants of the
hydrogens on the tetrahydropyridazine ring of these compounds
were based on H-10 and H-40 that show up in the aliphatic region;
H-10 shows at a lower field (dH = 5.0–5.2 ppm) due to the combined
effects of the nitrogen atom and the vinyl moiety. The two H-40 in
compounds 4a, c are ddd signals with a large geminal coupling
(J ꢁ 16.5 Hz) and two small J vicinal and long range-couplings con-
stants (J ꢁ 5.0 Hz and J ꢁ 2.0 Hz). The vinyl protons H-20 and H-30
appeared as multiplets at the typical vinylic region dH = 6.0–
6.3 ppm. The main differences between isomers 4 and 5 are the
higher dH ca. 0.1 ppm for H-10 in isomer 5 compared to 4. Com-
pounds 4b and 5b were differentiated only on the basis of the
chemical shift difference between the two H-10. Also compound
from case d, which was not possible to crystallize display H-10 con-
sistent with structure 4 (Scheme 1). Compound 6a was identified
by X-ray, with an H-10 chemical shift about 0.04 ppm higher than
that of 7a. Also the same kind of H-10 difference was obtained for
adducts 6c and 7c. 13C NMR spectra are consistent with the as-
signed structures for compounds 4–7.
Figure 3. ORTEP view of the molecular structure of the compound 4c.
be the same. Curiously, the X-ray of compound 4c exhibits the
opposite conformation (anti) to that observed in CDCl3 solution
(syn). Compound 6a showed to be (S)-configuration with an anti
disposition of H10–H4.
Most predictably, the bulky substituents on the six-membered
acetal ring would prefer to occupy equatorial positions. Two
ground state conformers of diene 1c are represented in Figure 1.
These two conformers direct PTAD with a different angle of attack.
In the upper conformer PTAD attacks the rear face (si) leading to
the observed (S)-configuration at C-10. In the lower conformer
PTAD accesses the rear face (re) of the diene leading to the not
observed (R)-configuration. Figure 1 shows how a protecting group
on the alcohol function at C-5 is able to cause a stereoblocking
effect on the front faces of either conformers and how electronic
4. Conclusion
An analysis of the stereofactors influencing the selectivity of the
Diels–Alder cycloaddition between PTAD and dienes 1a–d was
investigated by performing a designed set of experiments. It was
demonstrated how different factors control the stereoelectronic
effect. We expect that this knowledge could be further applied to
other dienophiles aiding to the synthesis of interesting molecules.

1820
M. J. Alves et al. / Tetrahedron: Asymmetry 21 (2010) 1817–1820
3. Aspinall, I. H.; Cowley, P. M.; Mitchell, G.; Stoodley, R. J. J. Chem. Soc., Chem.
Acknowledgments
Commun. 1993, 1179–1180.
4. Mukhopadhyay, A.; Ali, S. M.; Husain, M.; Suryawanshi, S. N.; Bhakuni, D. S.
Tetrahedron Lett. 1989, 30, 1853–1856.
5. Baker, S. R.; Clissold, D. W.; McKillop, A. Tetrahedron Lett. 1988, 29, 991–994.
6. Zimmermann, P.; Schmidt, R. R. Liebigs Ann. Chem. 1988, 663–667.
7. Ruijter, E.; Schltingkemper, H.; Wessjohann, L. A. J. Org. Chem. 2005, 70, 2820–
2823.
We thank FCT for project funding PTDC/QUI/67407/2006; FCT
and FEDER for funding NMR spectrometer Bruker Avance II 400
as part of the National NMR Network. We thank Daniela Salgueiro
for having preparing some precursors and Dr Thomas Gilchrist for
reading the manuscript.
8. Mukhopadhyay, A.; Suryawanshi, S. N.; Bhakuni, D. S. Indian J. Chem., Sect B
1988, 27B, 1009–1011.
9. Trost, B. M.; Lee, D. C. J. Org. Chem. 1989, 54, 2271–2274.
10. Tripathy, R.; Carroll, P. J.; Thornton, E. R. J. Am. Chem. Soc. 1991, 113, 7630–
7640.
References
1. Larsen, D. S.; Stoodley, R. J. J. Chem. Soc., Perkin Trans. I 1989, 1841–1852.
2. Aspinall, I. H.; Cowley, P. M.; Stoodley, R. J. Tetrahedron Lett. 1994, 3397–3400.
11. Houk, K. N.; Moses, S. R.; Wu, Y.-D.; Rondan, N. G.; Jager, V.; Schohe, R.;
Fronczek, F. R. J. Am. Chem. Soc. 1984, 106, 3880–3882.