Chiral N-acyloxazolidines: Synthesis, structure, and mechanistic insights

Article abstract of DOI:10.1021/jo702149m

(Chemical Equation Presented) A series of chiral imines derived from 1-amino-1-deoxyalditols such as D-glucamine, a rather unexplored raw material from the chiral pool, have been serendipitiously transformed into a novel family of N-acetyl-1,3-oxazolidines by means of an unexpected acetylation. The structure of these substances is supported by spectroscopic and crystallographic data. The acetylates also trigger a complex dynamic transformation, in which an initially configured trans oxazolidine converts into a more stable cis-configured derivative. Both isomers can also exist as rotational conformers (E,Z) as a consequence of the restricted rotation around the N-acetyl bond. The barriers to rotation have been determined by variable-temperature experiments. Overall, this transformation most likely involves the intermediacy of a chiral iminium ion, which has been documented in the synthesis of nitrogen heterocycles, thus explaining the experimental facts.

Full text of DOI:10.1021/jo702149m

Chiral N-Acyloxazolidines: Synthesis, Structure, and Mechanistic
Insights†
Mart´ın AÄ valos,‡ Reyes Babiano,‡ Pedro Cintas,‡ Jose´ L. Jime´nez,‡ Mark E. Light,§
Juan C. Palacios,‡ and Esther M. S. Pe´rez*,‡
Departamento de Qu´ımica Orga´nica e Inorga´nica, QUOREX Research Group,
UniVersidad de Extremadura, E-06071 Badajoz, Spain, and Department of Chemistry,
UniVersity of Southampton, Southampton SO17 1BJ, United Kingdom
espero@unex.es
ReceiVed October 3, 2007
A series of chiral imines derived from 1-amino-1-deoxyalditols such as D-glucamine, a rather unexplored
raw material from the chiral pool, have been serendipitiously transformed into a novel family of N-acetyl-
1,3-oxazolidines by means of an unexpected acetylation. The structure of these substances is supported
by spectroscopic and crystallographic data. The acetylates also trigger a complex dynamic transformation,
in which an initially configured trans oxazolidine converts into a more stable cis-configured derivative.
Both isomers can also exist as rotational conformers (E,Z) as a consequence of the restricted rotation
around the N-acetyl bond. The barriers to rotation have been determined by variable-temperature
experiments. Overall, this transformation most likely involves the intermediacy of a chiral iminium ion,
which has been documented in the synthesis of nitrogen heterocycles, thus explaining the experimental
facts.
Introduction and Background
established,2 that of aminopolyols remains quite underexploited.
Early preparations of 1-amino-1-deoxypolyols include catalytic
hydrogenation in the presence of Pt, Pd, or Raney Ni of the
corresponding glycosylamines3 or hydrazones,4 hydrogenation
of 1-deoxy-1-nitropolyols over Ag2O,5 as well as chemical3 or
electrolytic6 reduction of monosaccharide oximes.
Recently we turned our attention to the preparation of some
protected derivatives of the readily available 1-amino-1-deoxy-
D-glucitol (1), commonly referred to as D-glucamine.7 In
following well-known protocols, its condensation with aromatic
It should be largely unnecessary to emphasize the role played
by carbohydrates in contemporary organic synthesis, a conve-
nient source of both chiral synthons and densely functionalized
raw materials for varied purposes. As starting materials for the
preparation of enantiomerically pure compounds, carbohydrates
often provide advantages of availability and low cost. However,
to be truly practical, carbohydrate-based synthetic methodology
must also be efficient and in particular should avoid lengthy
protection/deprotection protocols.1 It is in this context that the
search for inexpensive and available chirons from sugars, and
methods to convert them into building blocks constitutes our
current challenge and this study concentrates on the use of
aminopolyols to this end.
(2) (a) Horton, D. In The Amino Sugars; Jeanloz, R. W., Ed.; Academic
Press: New York, 1969; Vol. IA, Chapter 1, pp 3-211. (b) Horton, D.;
Wander, J. D. In The Carbohydrates; Pigman, W., Horton, D., Wander, J.
D., Eds.; Academic Press: New York, 1980; Vol. IB, Chapter 16, pp 643-
760.
(3) (a) Kagan, F.; Rebenstorf, M. A.; Heinzelman, R. V. J. Am. Chem.
Soc. 1957, 79, 3541. (b) Neuberg, C.; Marx, F. Biochem. Z. 1907, 3, 539.
(4) Wolfrom, M. L.; Shafizadeh, F.; Wehrmu¨ller, J. O.; Armstrong, R.
K. J. Org. Chem. 1958, 23, 571.
(5) Angus, H. J. F.; Richtmyer, N. K. Carbohydr. Res. 1967, 4, 7.
(6) Ryan, G.; Utley, H. P.; Jones, H. F. Tetrahedron Lett. 1988, 29, 3699.
(7) Avalos, M.; Babiano, R.; Cintas, P.; Hursthouse, M. B.; Jime´nez, J.
L.; Light, M. E.; Palacios, J. C.; Pe´rez, E. M. S. Eur. J. Org. Chem. 2006,
657.
It is somewhat surprising that while the chemistry of reducing
aminosugars, particularly 2-amino-2-deoxyaldoses, is well
† Dedicated to Professor Miguel Yus on the occasion of his 60th birthday.
‡ Universidad de Extremadura.
§ University of Southampton.
(1) (a) Bols, M. Carbohydrate Building Blocks; John Wiley & Sons:
New York, 1995. (b) PreparatiVe Carbohydrate Chemistry; Hanessian, S.,
Ed.; Marcel Dekker: New York, 1997.
10.1021/jo702149m CCC: $40.75 © 2008 American Chemical Society
Published on Web 12/20/2007
J. Org. Chem. 2008, 73, 661-672
661

AÄ valos et al.
SCHEME 1
aldehydes led to the corresponding Schiff bases (2), which serve
as a suitable protecting group of the amino function (Scheme
1). Such substrates could also be employed as stereodifferen-
tiating elements in asymmetric syntheses, although again this
potentiality has scarcely been employed. Thus, Barton et al. did
recognize the usefulness of acyclic imines derived from D-
glucosamine in the Staudinger synthesis of optically active
â-lactams.8,9 Kunz and his associates have successfully em-
ployed imines derived from O-protected glycosylamines in
numerous asymmetric reactions,10 such as Strecker,11 Ugi,12
Mannich,13 tandem Mannich-Michael,14 hetero-Diels-Alder,15
organometallic additions,16 or Staudinger.17 Likewise, glyco-
sylimines have been used in the stereoselective synthesis of
R-amino acids.18
Further interest on Schiff bases as ligands, not related to
carbohydrates nevertheless, is illustrated by metallosalen com-
plexes (especially Mn- and Cr-salen complexes) having a
structure of N,N-ethylenebis(salicylideneaminato) that catalyze
the epoxidation of a wide range of olefins.19
We did initially envisage that Schiff bases 2 could conve-
niently be O-protected by acylation yielding 3 and then
converted into 4. The latter is a useful intermediate that can be
employed as starting material en route to sugar isocyanates and
their ureas.7 Quite unexpectedly, the conventional O-acetylation
of imines 2 did not lead to compounds 3, but to anomalous
substances, namely polyol-based oxazolidines, which constitute
the subject of the present work. Compounds 4 could actually
be prepared by an alternative strategy that involves N-protection
as an enamine derivative (5) followed by O-acylation (6) and
further N-deprotection under mild conditions. It should be noted
that enantiomerically pure 1,3-oxazolidines (as well as their
N-acyl derivatives and other oxazolidinones), which can be
generated from 1,3-amino alcohols and functionalized aldehydes
(or aldehyde equivalents), have been explored as chiral auxil-
iaries in asymmetric synthesis.20 Our research might therefore
add novel chirons suitable for synthetic design.
Results and Discusion
Schiff Bases Derived from D-Glucamine. Formation of
imines was initially conducted by dissolving D-glucamine in
water and adding, under stirring, the corresponding aldehyde,
either solvent-free with liquid aldehydes or dissolved in
methanol. However, improved yields (∼90%) could be obtained
when a mixture of D-glucamine and aldehyde was refluxed in
benzene with azeotropic removal of water. Insoluble materials
can be obtained within short reaction times. Thus, Schiff bases
7-15 were prepared by condensation of 1 with substituted
benzaldehydes and cinnamaldehyde (Table 1).
(8) Barton, D. H. R.; Gateau-Olesker, A.; Anaya-Mateos, J.; Cleophax,
J.; Ge´ro, S. D.; Chiaroni, A.; Riche, C. J. Chem. Soc., Perkin Trans. I 1990,
3211.
(9) Anaya, J.; Barton, D. H. R.; Ge´ro, S. D.; Grande, M.; Hernando, J.
I. M.; Laso, N. M. Tetrahedron: Asymmetry 1995, 6, 609.
(10) Kunz, H. In SelectiVities in Lewis Acid Promoted Reactions;
Schinzer, D., Ed.; Kluwer Academic: Dordrecht, The Netherlands, 1989;
pp 189-202.
(11) Kunz, H.; Sager, W. Angew. Chem., Int. Ed. Engl. 1987, 26, 557.
(12) Kunz, H.; Pfrengle, W. J. Am. Chem. Soc. 1988, 110, 651.
(13) Kunz, H.; Pfrengle, W. Tetrahedron 1988, 44, 5487.
(14) Kunz, H.; Schanzenbach, D. Angew. Chem., Int. Ed. Engl. 1989,
28, 1068.
(15) Kunz, H.; Pfrengle, W. Angew. Chem., Int. Ed. Engl. 1989, 28, 1067.
(16) Pfrengle, W.; Kunz, H. J. Org. Chem. 1989, 54, 4261.
(17) Laschat, S.; Kunz, H. J. Org. Chem. 1991, 56, 5883.
(18) (a) Georg, G. I.; Akgu¨n, E.; Mashava, P. M.; Milstead, M.; Ping,
H.; Wu, Z.; Velde, D. V. Tetrahedron Lett. 1992, 33, 2111. (b) Ross, G.
F.; Herdwetck, E.; Ugi, I. Tetrahedron 2002, 58, 6127.
(19) Katsuki, T. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; Wiley-
VCH: New York, 2000; Chapter 6B, pp 295-314.
The structures of these substances have been confirmed by
analytical and spectroscopic data. The most prominent IR
(20) (a) Seyden-Penne, J. Chiral Auxiliaries and Ligands in Asymmetric
Synthesis; John Wiley & Sons: New York, 1995; Chapter 1.3. (b) Steif,
F.; Wibbeling, B.; Meyer, O.; Hoppe, D. Synthesis 2000, 743-753. (c)
Vega-Pe´rez, J. M.; Vega, M.; Blanco, E.; Iglesias-Guerra, F. Tetrahedron:
Asymmetry 2001, 12, 3189-3203. (d) Shen, M.; Li, C. J. Org. Chem. 2004,
69, 7906-7909. (e) Spino, C.; Tremblay, M.-C.; Gobdout, C. Org. Lett.
2004, 6, 2801-2804. (f) Elzner, S.; Maas, S.; Engel, S.; Kunz, H. Synthesis
2004, 2153-2164. (g) Gessier, F.; Schaeffer, L.; Kimmerlin, T.; Flo¨gel,
O.; Seebach, D. HelV. Chim. Acta 2005, 88, 2235-2249. (h) Hein, J. E.;
Geary, L. M.; Jaworski, A. A.; Hultin, P. G. J. Org. Chem. 2005, 70, 9940-
9946. (i) Coumbarides, G. S.; Dingjan, M.; Eames, J.; Flinn, A.; Motevalli,
M.; Northen, J.; Yohannes, Y. Synlett 2006, 101-105. (j) Gnas, Y.; Glorius,
F. Synthesis 2006, 1899-1930. (k) Bruschi, M.; Orlandi, M.; Rindone, B.;
Rummakko, P.; Zoia, L. J. Phys. Org. Chem. 2006, 19, 592-596. (l)
Chavda, S.; Coumbarides, G. S.; Dingjan, M.; Eames, J.; Flinn, A.; Northen,
J. Chirality 2007, 19, 313-320.
662 J. Org. Chem., Vol. 73, No. 2, 2008

Chiral N-Acyloxazolidines
consistent with a strong intramolecular hydrogen bonding.
Overall, the above results agree with literature data indicating
that most adducts (>90% according to our database searching)
generated from amines and 2-hydroxybenzaldehyde adopt imine
structures in the solid state,23-25 whereas other crystal data
support enamine structures,23,26-29 and all have invariably iminic
structures in solution.29 On the other hand, there are a few
adducts solved by X-ray crystallography, arising from amines
and 3-hydroxy-2-naphthaldehydes, but they correspond again
to imine skeleta.30,31
TABLE 1. Isolated Yields (%) for Compounds 7-20 by Methods
A and B
compd
method Aa
method Bb
7
8
9
92
55
95
63
87
73
91
26
86
95
58
90
89
89
99
93
98
90
99
99
91
95
96
98
94
c
10
11
12
13
14
15
16
17
18
19
20
The enamino structure associated with compounds 19 and
20 can easily be inferred from spectroscopic data; a strong IR
absorption at ∼1635 cm-1 points to the stretching vibration of
the enamine carbonyl group, while weaker absorptions are to
be expected for the corresponding imines. The methine proton
coupled to the NH proton appeared as a doublet signal at ∼9
ppm. The 13C NMR spectrum equally confirms the enamine
structure on showing the carbonyl group at ∼180 ppm. As noted
above, the NH proton, shifted downfield at δNH >13 ppm,
suggests the participation of this group within a strong intramo-
lecular hydrogen bond.
98
c
a In water. b In benzene. c Not determined.
In this context, it is convenient to note that most structures
elucidated by X-ray diffraction analyses involving adducts
between amines and 2-hydroxy-1-naphthaldehyde correspond
to enamines,32-42 while compounds adopting iminic arrange-
absorption appeared at ∼1640 cm-1 and can be attributed to
the imino group.21 1H NMR spectra showed the imino proton
resonating at ∼8.0-8.2 ppm, either as singlet for compounds
7-14 or doublet for 15 due to the coupling of that imino proton
with those of the olefinic group. Five OH groups were observed
in the range 4.2-5.0 ppm whereas all the protons of the sugar
moiety appeared more highfield (∼3.2-4.0 ppm).
(23) Paredes-Garc´ıa, V.; Venegas-Yazigi, D.; Lough, A. J.; Latorre, R.
Acta Crystallogr. 2000, C56, 283.
(24) Dominiak, P. M.; Grech, E.; Barr, G.; Teat, S.; Mallinson, P.;
Wozniak, K. Chem. Eur. J. 2004, 9, 963.
(25) Ozeryanskii, V. A.; Pozharskii, A. F.; Schiff, W.; Kamie´nski, B.;
Sawka-Dobrowolska, W.; Sobczyk, L.; Grech, E. Eur. J. Org. Chem. 2006,
782.
(26) (a) Cungen, Z.; Peizi, Z.; Dan, W.; Kaibei, Y. J. Chem. Res. 2000,
402. (b) Odabasoglu, M.; Albayrak, C¸ .; Bu¨yu¨kgu¨ngo¨r, O.; Lo¨nnecke, P.
Acta Crystallogr. 2003, C59, 616.
(27) Albayrack, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r, O.; Lo¨nnecke, P.
Acta Crystallogr. 2004, C60, 318.
(28) Ensali, C. C.; Albayrak, C¸ .; Odabasoglu, M.; Endo¨nmez, A. Acta
Crystallogr. 2003, C59, 601.
(29) Kia, R.; Esmazilbeig, A.; Harkema, S. Acta Crystallogr. 2004, A60,
267.
(30) Ferna´ndez-G., J. M.; Rosales, M. J.; Toscano, R. A.; Tapia, R. G.
Acta Crystallogr. 1986, C42, 1313.
(31) Lin, J.; Cui, G.-H.; Li, J. R.; Xu, S.-S. Acta Crystallogr. 2005, E61,
627.
(32) Pavlovic, G.; Sosa, J. M. Acta Crystallogr. 2000, C56, 1117.
(33) Yu¨ce, S.; Oze¨k, A.; Albayrak, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r,
O. Acta Crystallogr. 2004, E60, 1217.
(34) Oze¨k, A.; Yu¨ce, S.; Albayrak, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r,
O. Acta Crystallogr. 2004, E60, 826.
(35) Odabasoglu, M.; Albayrak, C¸ .; Bu¨yu¨kgu¨ngo¨r, O. Acta Crystallogr.
2004, E60, 142.
(36) Dilovic, I.; Matkovic-Calogovic, D.; Popovic, Z.; Roje, V. Acta
Crystallogr. 2005, C61, 351.
13C NMR spectra revealed the existence of the iminic carbon
at ∼162 ppm. The terminal methylene carbons of D-glucamine
showed almost coincidental chemical shifts at ∼64 ppm while
the remaining sugar carbons resonated around ∼72 ppm, a fact
usually typical of acyclic polyhydroxyalkyl chains. It is
noteworthy that the C-1 carbon, i.e., the position linked to the
imino function, appeared unusually downfield, whereas D-
glucamine derivatives show the same C-1 carbon at ∼43 ppm.22
When D-glucamine was treated with 2-hydroxybenzaldehyde,
2,4-dihydroxybenzaldehyde, and 2-hydroxy-3-naphthaldehyde,
the corresponding imines 16-18 could also be isolated. In stark
contrast, the condensation of the parent aminoalditol with
2-hydroxy-1-naphthaldehyde and 1-hydroxy-2-naphthaldehyde
yielded the enamine structures 19 and 20, respectively.
Spectroscopic data for compounds 16-18 are similar to those
found for 7-14, thereby supporting imine skeleta: a typical
IR absorption at ∼1640 cm-1 (νCdN) as well as proton and
carbon signals at ∼8.5 and ∼167 ppm, respectively, which can
be attributed to the imino function. The o-hydroxyl groups at
the aromatic rings are highly deshielded (δOH >13 ppm), a fact
(37) Oze¨k, A.; Yu¨ce, S.; Albayrak, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r,
O. Acta Crystallogr. 2004, E60, 1162.
(21) Nakanishi, K.; Solomon, P. Infrared Absorption Spectroscopy, 2nd
ed.; Holden-Day: San Francisco, CA, 1997; p 33.
(22) Bock, K.; Pedersen, Ch. AdV. Carbohydr. Chem. Biochem. 1983,
41, 27.
(38) Oze¨k, A.; Yu¨ce, S.; Albayrak, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r,
O. Acta Crystallogr. 2004, E60, 828.
(39) Acevedo-Aranz, E.; Ferna´ndez-G., J. M.; Rosales-Hoz, M. J.;
Toscano, R. A. Acta Crystallogr. 1992, C48, 115.
J. Org. Chem, Vol. 73, No. 2, 2008 663

AÄ valos et al.
FIGURE 1. 1H NMR spectrum in CDCl3 for compound 25.
ments in the solid state43,44 behave contrarily as enamines in
solution.44
produce the corresponding per-O-acetyl derivatives under the
same reaction conditions.7
In a further attempt to obtain simplified models, at least from
a spectroscopic viewpoint, the imine derivative 22 has been
obtained starting from R,R,R-tris(hydroxymethyl)methylamine
(21). By virtue of the symmetry of the latter, compounds such
as 22 show spectra that can easily be interpreted: the stretching
vibration of the CdN bond (νCdN ≈ 1640 cm-1), singlet signals
at ∼8.5 (δCHdN) and 14.5 ppm (δOH,phenol), and a carbon
resonance at ∼165 ppm all support an iminic structure in
solution. Conversely, crystallographic data point to an enamine
derivative in the solid state as evidenced by recent studies.26 In
contrast, the adduct generated from 21 and 2-hydroxy-1-
naphthaldehyde shows an enamine structure (23) as inferred
from its carbonyl resonance at ∼180 ppm and those of the NH
proton (∼14 ppm) and the CdCH-N group (∼8.9 ppm), both
as doublets. The large downfield shifts for the phenolic OH
group and the NH of the enamine moiety in 22 and 23,
respectively, suggest again the existence of strong intramolecular
hydrogen bonds.
A preliminary inspection of their IR spectra shows a strong
absorption at ∼1750 cm-1, resulting from the acetate groups,
and a medium intensity band at ∼1652 cm-1 characteristic of
the stretching vibration of the amide carbonyl. However, NMR
data provide the most significant data supporting the formation
of a ring system (Figure 1). At first glance the observed
1
multiplicity of signals in both the H and 13C NMR spectra,
having similar chemical shifts and multiplicities, suggested the
existence of two isomers for 24-27 and 32 and four isomers
in the remaining cases. In the latter compounds 28-31 could
be isolated by recrystallization, as an isomer couple each, in a
similar way to that of 24-27 and 32. If these compounds had
an imine structure, this functional group would be expected to
give rise to NMR signals at ∼8.3 and ∼161 ppm, corresponding
to the iminic proton and carbon, respectively. In contrast,
resonances at ∼6.0 and ∼90 ppm were found. Such shifts agree
with those of saturated groups and their magnitude is consistent
with the typical values found for the C-2 position of an
oxazolidine ring (90 ppm)45 and its proton (∼5.30-6.00
ppm).45-47
Synthesis of Chiral Oxazolidines. When imines 7-13, 15,
and 16 were treated with acetic anhydride in pyridine, the
resulting acetyl derivatives show spectroscopic data that rule
out the expected structure of per-O-acetylimines (3) (with overall
yields ranging from 40% to 86%). Such data are rather consistent
with N-acyloxazolidines (24-32). This behavior is clearly
distinct from that of D-glucamine-based enamines, which
(40) Elerman, Y.; Kabak, M.; Elmali, A.; Svoboda, I. Acta Crystallogr.
1998, E54, 128.
(41) Oze¨k, A.; Yu¨ce, S.; Albayrak, C¸ .; Odabasoglu, M.; Bu¨yu¨kgu¨ngo¨r,
O. Acta Crystallogr. 2004, E60, 356.
(42) Kaitner, B.; Pavlovic, G. Acta Crystallogr. 1996, C52, 2573.
(43) Popovic, Z.; Roje, V.; Pavlovic, G.; Matkovic-Calogovic, D.; Giester,
G. J. Mol. Struct. 2001, 597, 39.
(44) Pavolic, G.; Sosa, J. M.; Vikic-Topic, D.; Leban, I. Acta Crystallogr.
2002, E58, 317.
(45) Neuvonen, K.; Fu¨lo¨p, F.; Neuvonen, H.; Koch, A.; Kleinpeter, E.;
Pihlaja, K. J. Org. Chem. 2001, 66, 4132.
(46) (a) Fu¨lo¨p, F.; Pihlaja, K. Tetrahedron 1993, 49, 6701. (b) Fu¨lo¨p,
F.; Pihlaja, K.; Neuvonen, K.; Berna´th, G.; Argay, G.; Ka´lma´n, A. J. Org.
Chem. 1993, 58, 1967.
664 J. Org. Chem., Vol. 73, No. 2, 2008

Chiral N-Acyloxazolidines
SCHEME 2
to obtain the unprotected oxazolidines 37 having a greater atom-
economy orientation.52
Rotational Equilibria of N-Acetyloxazolidines. As men-
tioned before on the basis of NMR data, compounds 24-32
exist in solution as isomeric couples that could not be separated.
After repeated crystallizations, these mixtures remain in the same
ratio, a fact pointing to rotational isomers. This surmise could
be assessed by means of variable-temperature experiments. Thus,
the broad NMR signals observed at room temperature become
increasingly sharp and well-differentiated resonances by de-
creasing the temperature. On heating, the signals become
increasingly broad again until they merge to form simplified
spectra showing only one signal set (Figure 3). The existence
of such isomers should be reasonably linked to the restricted
rotation around the amide bond. Assuming that this function
should adopt a coplanar arrangement with the oxazolidine ring,
thereby facilitating the electronic delocalization between the lone
pair on the nitrogen atom and the carbonyl group, this system
could in principle have the possibility to exist in two orientations
as depicted in Scheme 3: either placing the carbonyl oxygen
close to the C-4 position of the heterocycle (E configuration)
or pointing the former group to the C-2 position of the ring (Z
configuration).
FIGURE 2. ORTEP diagram for N-acyloxazolidine 33.
A similar behavior was found in imines derived from R,R,R-
tris(hydroxymethyl)methylamine. Thus, acetylation of 22 led
to compound 33, an N-acyloxazolidine derivative whose struc-
ture could be unambiguously elucidated by single-crystal X-ray
diffraction (Figure 2).48 The solid-state structure also reveals
an E configuration around the amide bond.
Since compound 33 show spectroscopic data similar to those
of 24-32, the latter substances should equally be oxazolidine
derivatives. Characteristic signals are the amide band in their
IR spectra at 1661 cm-1 and resonances at ∼87 (C-2) and ∼6.3
ppm as mentioned above. The methylene protons of the
acetoxymethyl groups are diastereotopic and give rise to AB-
type splitting patterns centered at 4.74 and 4.57 ppm. The
heterocyclic methylene gives an AB splitting too having
chemical shifts markedly different for its protons (∆δ ∼0.25
ppm). It is noteworthy that the acetamido resonance occurs at
unusually high field around ∼1.84 ppm. If compound 33 had
in solution a structure identical with that of the solid state, the
methyl group of the acetamido function would be placed next
to the aromatic ring and would undergo an appreciable shielding.
It is also interesting to point out that N-acyloxazolidines
derived from 21 had been previously synthesized using a
lengthier route (Scheme 2).49 Condensation of 21 with 2 equiv
of aldehyde in refluxing benzene with concomitant water
elimination leads to 1-aza-3,7-dioxabicyclo[3.3.0]octano (34),50,51
whose acetylation with acetyl chloride and further hydrolysis
gives rise to an oxazolidine hydrochloride (36). Neutralization
with aqueous solutions of sodium hydroxide or sodium hydrogen
carbonate also causes migration of the acetyl group to yield
N-acetyloxazolidines (37).51
SCHEME 3
The protons of the C-2 and C-4 positions will undergo the
influence of the carbonyl group in a different way for each
rotamer. This assumption correlates well with the chemical shift
ranges observed experimentally. Thus, the geminal C-4 protons
exhibit chemical shift differences of about 1 ppm for one of
the rotamers and approximately 0.5 ppm for the other. The C-2
protons show typical shift differences of 0.3 ppm for both
rotamers.
A similar behavior has been previously reported for glyco-
sides derived from 5-acetamido-5-deoxypyranoses (38),53 6-acy-
lamino-2,6-anhydro-2-deoxyheptitols (39),54 and glycosides or
nucleosides from 4-acetamido-4-deoxyfuranoses (40 and 41).53,55
The present synthesis of N-acetyloxazolidines such as 33,
which can further be easily deacetylated, envisages a strategy
(47) Maireanu, C.; Darabantu, M.; Ple´, G.; Berghian, C.; Condamine,
E.; Ramondenc, Y.; Silaghi- Dumitrescu, I.; Mager, S. Tetrahedron 2002,
58, 2681.
(48) Crystal data for compound 33 have been deposited with the
Cambridge Crystallographic Data Centre (CCDC-644232) and can be
obtained, upon request, from the Director, Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, U.K.
(49) For a revision in the synthesis of oxazolidines see: Bergmann, E.
D. Chem. ReV. 1953, 53, 309.
(50) Senkus, M. J. Am. Chem. Soc. 1945, 67, 1515.
(52) (a) Trost, B. M. Science 1991, 254, 1471. (b) Trost, B. M. Angew.
Chem., Int. Ed. Engl. 1995, 34, 259.
(51) Pierce, J. S.; Lunsford, C. D.; Raiford, R. W., Jr.; Rush, J. L.; Riley,
D. W. J. Am. Chem. Soc. 1951, 73, 2595.
(53) Szarek, W. A.; Wolfe, S.; Jones, J. K. N. Tetrahedron Lett. 1964,
2743.
J. Org. Chem, Vol. 73, No. 2, 2008 665

AÄ valos et al.
FIGURE 3. 1H NMR spectra of compound 32 at the temperatures indicated (recorded in CDCl3).
TABLE 2. Dynamic NMR Data for the Interconversion of the Z,E
Rotamers
product
solvent
∆V (Hz) k (s-1
)
TC (K) ∆Gq a ∆Gq b
25
CDCl3
DMSO-d6
CDCl3
DMSO-d6
CDCl3
DMSO-d6
CDCl3
116.0
26.0
119.6
29.6
107.6
48.82
104.2
30.4
257.6
57.7
328
317
327
320
325
325
321
318
65.2
64.4
65.1
67.4
65.0
67.1
63.4
66.0
15.6
15.4
15.6
16.1
15.5
16.0
15.2
15.8
27
28
32
265.6
65.7
239.0
108.4
231.4
67.5
DMSO-d6
a In kJ mol-1 b In kcal mol-1 (1 cal ) 4.183 J).
.
For compound 33 only one rotamer could be observed in
solution, which should be correlated with its solid-state structure
(Figure 2). It seems obvious that in the indicated E configuration,
the oxygen atom of the amide group bisects the angle between
the two acetoxymethyl groups at C-4, and such an arrangement
is less sterically encumbered than its Z counterpart, for which
the bulkier methyl group would bisect the acetoxymethyl groups.
Estimation of Rotational Barriers. Temperature-dependent
NMR spectra enabled the measurement of the interconversion
barriers between rotational isomers, by determining experimen-
tally the coalescence temperature for some signals, in both
CDCl3 and DMSO-d6. It is well-known that the rate constant at
the coalescence point between exchanging systems in equilib-
rium with NMR signals showing no coupling and being of the
same intensity follows the equation:56
signal at the coalescence point TC; it also corresponds to the
difference in chemical shift observed for slow exchange.
Although this expression is not accurate when equilibria occur
between unequally populated states, errors are usually small and
it gives an approximate estimation of the barriers to rotation.
By combining the above expression with the Eyring equation,
according to which the exchange rate constant decreases
exponentially with the free molar activation enthalpy, and after
substitution of the fundamental constants, one gets the following
expression for the free molar activation enthalpy for the
exchange process:57
∆Gq (cal‚mol-1) ) 1.987TC[22.96 + ln(TC/∆V)]
Table 2 collects the different data obtained for compounds 25,
27, 28, and 32, which are quite similar in both solvents.
Geometric Isomerization of 1,3-Oxazolidines. It was in-
triguing to see that the rotational pair observed for 25 converted
slowly and gradually in CDCl3 into a novel rotamer couple
(Figure 4). This transformation should be associated with a
stereochemical change of the new chiral center at C-2, generated
during the imine cyclization that occurs by acetylation. Such a
k ) π21/2∆ν ) 2.221∆ν
where ∆ν represents the full width at half-maximum of the
(54) Saavedra, O. M.; Martin, O. R. J. Org. Chem. 1996, 61, 6987.
(55) Reist, E. J.; Gueffroy, D. E.; Blackford, R. W.; Goodman, L. J.
Org. Chem. 1966, 31, 4025.
(56) (a) Breitmaier, E. Structure Elucidation by NMR in Organic
Chemistry. A Practical Guide; John Wiley & Sons: New York, 1993; pp
63-64. (b) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic
Compounds; John Wiley & Sons: New York, 1994; p 503.
(57) Williams, D. H.; Fleming, I. Spectroscopic Methods in Organic
Chemistry, 4th ed.; McGraw-Hill: New York, 1984; p 103.
666 J. Org. Chem., Vol. 73, No. 2, 2008

Chiral N-Acyloxazolidines
FIGURE 4. 1H NMR spectrum in CDCl3 of compound 25 after equilibration.
FIGURE 5. Equilibration of Z,E rotamers derived from 45 with their Z,E partners of the new isomer in CDCl3 acidified with trifluoroacetic acid.
ring closure affords a kinetically favored mixture of rotamers,
which further evolves into the products of thermodynamic
control. This isomerization, however, was not detected in
DMSO-d6 at room temperature.
The crude mixture of synthetic 28-31 showed H and 13C
1
FIGURE 6. Oxazolidine ring conformations (R1) D-arabino-
NMR spectra similar to those obtained in the equilibration of
(CHOAc)3CH2OAc, R ) Ar or PhCHdCH-).
25 (Figure 4). On the basis of these results, it was thought that
this isomerization took place via acid catalysis promoted by
DCl traces in CDCl3. To verify this surmise, we studied the
fate of diverse oxazolidine solutions in CDCl3 acidified with
trifluoroacetic acid. It was shown that a slow equilibration of
Z,E rotamers derived from the starting oxazolidine with their
Z,E partners of the new isomer (Figure 5). The presence of the
aldehydo group reveals that such an isomerization is ac-
companied by a degradation of the heterocyclic ring.
Absolute Configuration at C-2 of Chiral Oxazolidines.
There are some antecedents dealing with the steric course of
imine cyclization to produce oxazolidines,58-60 although they
are not completely applicable to the present case. In a further
attempt to label the absolute stereochemistry at C-2, a series of
NOE experiments involving the heterocyclic protons at C-2,
C-4, and C-5 have been carried out.
J. Org. Chem, Vol. 73, No. 2, 2008 667

AÄ valos et al.
FIGURE 7. 1H NMR spectrum of compound 31 (recorded in CDCl3)
SCHEME 4
Had the substituent at C-2 adopted a trans relationship relative
to the substituent at C-5 (polyhydroxyl chain), the oxazolidine
ring would exhibit either T1 or E1 conformations. In such
5
circumstances, the bulkier polyhydroxyl side chain and the H-2
proton adopt pseudoequatorial dispositions, while the substituent
at C-2 would be at a pseudoaxial arrangement. This does mean
that the H-2 proton lies far away with respect to the remaining
oxazolidine protons and, accordingly, no appreciable NOE
effects would be expected among them (Figure 6). A distinctive
behavior would take place if the substituents at C-2 and C-5
had adopted a cis relationship, which would lead to pseudoequa-
torial configurations for such groups. The steric effect decreases
and the H-2 proton lies closer to the H-5 proton and one of the
protons on the C-4 position; then significant NOE effects should
be observed.61
When such decoupling experiments are conducted on the
initial rotamers (kinetic control products), no NOEs are observed
for the H-2 signal and the rest of the heterocyclic protons, thus
suggesting the presence of a trans isomer. Assuming that the
absolute configuration at C-5 is known to be S (coming from
D-glucamine), the stereochemistry at C-2 should logically be R
(24-32).
to rotamers having a similar spatial environment for the
acetamido group, i.e., the E rotamers of 25 and 32. These
conjectures could also be confirmed by assessing the NMR
spectra of the rotamers from 31 (Figure 7). This styryloxazo-
lidine shows spectra analogous to those observed for 25 and
32, and with a similar rotamer distribution (Scheme 4). However,
there are no proton signals at chemical shifts less than 1.98 ppm.
Due to the CHdCH linkage, the N-acetyl group is located
outside of the shielding zone of the aromatic ring, and therefore
this methyl group will be expected to be more deshielded as
compared to the E rotamers of 25 and 32.
When both kinetically controlled rotamers (2R-oxazolidines)
of 25 isomerize to the corresponding thermodynamic products
(2S-oxazolidines), the major rotamer shows again a shielded
acetyl group (δMeCON 1.86 ppm), for which the E conformation
has been assigned (Figure 4).
Absolute Stereochemistry of Rotational Pairs. Further
analysis of NMR data provides further evidence for the
stereochemistry of these rotamers around the C-N bond of the
amide linkage. As noted above, compound 33 exists exclusively
as its E rotamer in solution with the methyl group of the
acetamide function appearing unusually upfield (δMeCON ∼1.84
ppm), and that shielding results from the spatial arrangement
of this methyl group with respect to the aromatic ring (Figure
2). Obviously this effect will be absent in the case of the Z
rotamer, in which the methyl group lies far from the aryl
Scheme 5 displays a plausible rationale consistent with
experimental data. Thus, acetylation of imines 7-17 does
initially produce their corresponding trans isomers (24-32),
having the 2R absolute configuration and existing as a mixture
of E and Z conformers in equilibrium, the former being
1
substituent. For compounds 25 and 32 the H NMR spectrum
of the kinetically controlled rotamers (2R isomer) shows a signal
for the acetyl group of one rotamer (δMeCON ∼1.72 ppm)
matching that of E-33. The latter signal should therefore belong
(58) For a revision on the cyclization of iminium ions in the synthesis
of chiral heterocycles see: Royer, J.; Bonin, M.; Micouin, L. Chem. ReV.
2004, 104, 2311.
(59) Just, G.; Potvin, P.; Uggowitzer, P. J. Org. Chem. 1983, 48, 2923.
(60) Khruscheva, N. S.; Loim, N. M.; Sokolov, V. I.; Makhaev, V. D.
J. Chem. Soc., Perkin Trans. I 1997, 2425.
(61) (a) Santiesteban, F.; Grimaldo, C.; Contreras, R.; Wrackmeyer, B.
J. Chem. Commun. 1983, 1486. (b) Hussain, A.; Wyatt, P. B. Tetrahedron
1993, 49, 2123.
668 J. Org. Chem., Vol. 73, No. 2, 2008

Chiral N-Acyloxazolidines
TABLE 3. Calculated Relatives Energies of Cis/Trans and Z/E
SCHEME 5
Configurations of Some 1,3-Oxazolidinesa,b
cis (5S)
trans (5R)
product
E
Z
E
Z
25
27
30
32
0.00
1.42
0.00
0.00
1.21
0.00
2.60
0.94
8.85
0.40
0.11
3.70
14.56
13.46
15.95
12.05
a At DFT level (B3LYP/6-31G*) in the vacuum. b In kcal mol-1
.
31G*),63 on compounds 25, 27, 30, 32, and their 5S isomers.
Table 3 shows the calculated energies for the trans isomers (2R
configured) and the corresponding cis isomers (2S configured).
The above results are consistent with our experimental
observations and stereochemical conclusions. The trans oxazo-
lidines are less stable and should therefore be the kinetic product,
the energy difference relative to its cis isomers, thereby
accounting for the conversion, being the thermodynamic product.
Formation of Oxazolidine Rings: Mechanistic Consider-
ations. At first sight, the most intuitive mechanistic hypothesis
explaining the formation of oxazolidines 24-32 should first
involve cyclization of imine derivatives followed by acetylation
as shown in Scheme 7. However, imines such as 25, 31, or 32
were stable in pyridine-d5 solutions for prolonged reaction times
and no oxazolidine formation could be observed by NMR
monitoring.
These results clearly suggest that formation of oxazolidines
from imines under acetylating conditions should be occurring
by a mechanism other than that of Scheme 7. An alternative
hypothesis, both plausible and consistent with the whole range
of experiments, is depicted in Scheme 8. Given the inherent
basicity of the iminic nitrogen, formation of the corresponding
acetyliminium ion 60 should be occurring quickly, which would
give rise to the heterocyclic derivative, a step presumably
prevalent. Under appropriate conditions, these compounds could
be partially converted into the more stable cis isomers of 2S
absolute configurations and exist again as a mixture of two
conformers (Scheme 5).
Synthesis of Polyhydroxyalkyl Oxazolidines. Deacetylation
of compounds 24-32 can easily be carried out by treatment
with saturated solutions of ammonia in methanol, which affords
the corresponding D-arabino-tetrahydroxybutyl oxazolidines
42-50 in good to excellent yields (68-100%). These substances
maintain unaffected their trans (2R) stereochemistry, because
deacetylation in basic medium has no effect on that chiral center.
Synthesis of N-acyloxazolidines by reaction of N-acylamino
alcohols with aldehydes under acid catalysis has been reported.
Therefore, we have attempted an alternative synthesis of the
above substances using N-acetyl-D-glucamine (51)62 as raw
material, easily generated by reaction of 1 with acetic anhydride
in methanol, which was further treated with the corresponding
aldehydes in refluxing benzene with azeotropic removal of water
(Scheme 6). Unfortunately, the entire range of aldehydes tested,
both in the presence and the absence of p-toluenesulfonic acid
as catalyst, failed to give N-acyloxazolidines (52), the precursor
being recovered unaffected.
The spectroscopic data of 42-50 clearly support the O-
unprotected structures, particularly a broad IR absorption of the
OH groups (νOH 3600-3100 cm-1) along with the amide band
(63) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 2004
User’s Reference; Gaussian, Inc.: Wallingford, CT, 2004. (b) Becke, A.
D. J. Chem. Phys. 1993, 98, 5648. (c) Lee, C.; Yang, W.; Parr, R. G. Phys.
ReV. B 1998, 37, 75. (d) Hehre, W. J.; Radom, L.; Scheleyer, P. v. R.;
Pople, J. A. Ab Initio Molecular Orbital Theory; Wiley: New York, 1986.
(νCO ∼1623 cm-1) of the oxazolidine moiety. Both 1H and 13
C
NMR spectra showed duplicated signals, with singlets at ∼6
and ∼85 ppm characteristic of the heterocyclic ring. The
acetamido groups appeared with distinctive chemical shifts, for
instance, 1.59, 167.2, and 22.2 ppm for the major rotamer of
50, while resonances at 2.01, 167.9, and 23.4 ppm were observed
for the minor one.
Stability of N-Acetyloxazolidines. To test the validity of
Scheme 5 as a working hypothesis representing the possible
equilibria of the N-acetyloxazolidine rotamers in solution, a
series of theoretical calculations have also been performed.
Estimations have been obtained at the DFT level (B3LYP/6-
(62) Whistler, R. C.; Pauzer, H. P.; Roberts, H. J. J. Org. Chem. 1961,
26, 1583.
J. Org. Chem, Vol. 73, No. 2, 2008 669

AÄ valos et al.
SCHEME 6
SCHEME 7
SCHEME 8
SCHEME 9
SCHEME 10
favored by basic catalysis from solvent molecules. Formation
of 1,3-oxazolidines arises from a 5-endo-trig cyclization64 is
entropically favored with respect to the alternative 6-endo-trig
that leads to a 1,3-oxazine ring.
The stereochemical outcome can be rationalized assuming
that cyclization of the acetyliminium cation takes place through
the most stable conformation. Iminium ion cyclization has in
fact become a useful strategy in the synthesis of chiral
heterocycles, particularly alkaloid skeleta.58
The first chiral center of D-glucamine has the S configuration
and its hydroxyl substituent lies in the Re face of the acetyl-
iminium ion. If the subsequent cyclization took place through
such a face, the newly created chiral center would be R and the
relative stereochemistry between the polyhydroxyl chain at C-5
and the aromatic substituent at C-2 would be trans, thus
generating the kinetic product (Scheme 9). The starting con-
formations 65 and 66 would lead by attack to the Re or Si faces
to the corresponding transition structures, which would ulti-
mately evolve into the oxazolidine derivatives (Scheme 10). A
conformation such as 66 (and hence the corresponding transition
structure) should be less stabilized than 65, because the bulkier
acetyliminium group bisects the angle between the hydroxyl
substituent and the sugar chain, thus generating two strong
gauche interactions; conversely, only one gauche interaction
with the hydroxyl group occurs in 65. Further transformation
into the more stable cis isomer 67 (of 2S configuration) should
be occurring under acid catalysis, a fact that regenerates the
cation intermediate, which slowly and irreversible cyclizes via
the Si face (Scheme 11). The requirement for acid catalysis is
also supported by other experimental observations. Thus,
elimination of acid substances in CDCl3, by passing the solution
through alumina, inhibits the isomerization. Likewise, no
isomerization takes place when the process is conducted in
DMSO-d6, lacking acid properties.
Conclusions
This work demonstrates that condensation of aromatic alde-
hydes with 1-amino-1-deoxypolyols, exemplified here by D-
glucamine, produces either imine or enamine skeleta. The former
derivatives, under conventional O-acetylation, give rise to
(64) Alva, M. E.; Chokotho, N. C. J.; Jarvis, T. C.; Johnson, C. D.; Lewis,
C. C.; McDonnell, P. D. Tetrahedron 1985, 41, 5919.
670 J. Org. Chem., Vol. 73, No. 2, 2008

Chiral N-Acyloxazolidines
SCHEME 11
(2 × 50 mL), a saturated solution of NaHCO3 (2 × 50 mL), and
distilled water (2 × 50 mL). The organic layer was dried (MgSO4)
and evaporated. If the resulting product was a solid this was
separated by filtration and washed with water.
unexpected N-acetyl-1,3-oxazolidine chirons, whose structures
are supported by spectroscopic data. This behavior can also be
extended to an imine derived from tris(hydroxymethyl)methy-
lamine and salicylaldehyde; the resulting oxazolidine structure
(33) has been elucidated unambiguously by X-ray diffraction
analysis. The N-acetyloxazolidines undergo dynamic equilibria
in solution where an initial mixture of E,Z rotamers (around
the N-acetyl bond) of a trans oxazolidine evolves into a more
stable cis isomer that equally exists as a mixture of rotational
isomers, in which the E rotamer is preferentially formed. By
means of theoretical calculations and NMR experiments, we
have now figured out how these transformations occur and a
mechanistic rationale satisfactorily accounts for all the experi-
mental observations. Such functionalized chiral oxazolidines
bearing a polyhydroxyl side chain constitute appropriate raw
materials for further use in asymmetric methodologies.
(2R,5S)-2-(4-Acetoxyphenyl)-3-acetyl-5-(1,2,3,4-tetra-O-acetyl-
D-arabino-tetritol-1-yl)oxazolidine (24E,Z): 42%; recrystallized
from ethanol had mp 149-151 °C; [R]24 -33.0 (c 0.5, chloro-
D
form); IR (KBr) νmax 1747 (CdO), 1652 (CdO, amide), 1225 (C-
O-C, ester), 1082, 1034 cm-1 (C-O); 1H NMR (400 MHz, CDCl3)
δ 7.53 (2H, d, J ) 7.6 Hz, H-arom), 7.34 (2H, d, J ) 7.6 Hz,
H-arom), 7.13 (2H, d, J ) 7.2 Hz, H-arom), 7.05 (2H, d, J ) 7.6
Hz, H-arom), 6.29 (1H, s, H-2E), 5.96 (1H, s, H-2Z), 5.42 (1H, m,
H-2′E), 5.39 (1H, m, H-2′Z), 5.39 (1H, m, H-1′E), 5.31 (1H, m,
H-1′Z), 5.13 (1H, m, H-3′E), 5.08 (1H, m, H-3′Z), 4.29-4.16 (4H,
m, H-4Z, H-4′Z, H-5Z, H-5E, H-4′′Z and H-4′′E), 3.90 (1H, dd, J4,4
) 9.2 Hz, J4,5 5.6 Hz, H-4E, oxaz), 3.31 (1H, t, J4,4 ) J4,5 ) 9.8
Hz, H-4E), 3.21 (1H, t, J4,4 ) J4,5 ) 8.0 Hz, H-4Z, oxaz), 2.30,
2.28, 2.08, 2.06, 2.03, 2.01 (10 × 3H, s, CH3 acetates); 13C NMR
(100 MHz, CDCl3) δ 170.5, 170.3, 170.1, 170.0, 169.8, 169.3 (Cd
O), 169.1, 168.1 (N-CdO), 151.5, 150.8, 136.5, 135.7, 128.1,
122.1, 121.2 (C-arom), 89.7 (C-2Z), 88.4 (C-2E), 76.6 (C-5E), 76.4
(C-5Z), 69.1, 69.0 (C-2′Z and C-2′E), 68.6, 68.4 (C-1′Z and C-1′E),
68.1, 68.0 (C-3′Z and C-3′E), 61.5 (C-4′Z and C-4′E), 47.4 (C-4E),
46.5 (C-4Z), 23.2 (CH3, Ac-N of the E isomer), 22.8 (CH3, Ac-N
of the Z isomer), 21.1, 20.8, 20.7, 20.6, 20.4 (CH3, acetates). Anal.
Calcd for C25H31NO12 (537.51): C, 55.86; H, 5.81; N, 2.61.
Found: C, 55.37; H, 5.84; N, 2.49.
Experimental Section
Compound 51 was synthesized as described.62
Condensation of D-Glucamine with Arylaldehydes. Method
A: To a solution of D-glucamine (10.0 g, 55.2 mmol) in water (70
mL) was added slowly a solution of the corresponding aldehyde
(55.0 mmol) in the minimal volume of methanol. The reaction
mixture was kept at room temperature under stirring until the
appearance of a solid within a few minutes. Precipitation was
continued at the refrigerator, then the solid was filtered and washed
successively with water, cold ethanol, and diethyl ether.
Method B: A mixture of D-glucamine (0.91 g, 5.0 mmol) and
the corresponding aldehyde (7.5 mmol) in benzene (15 mL) was
refluxed with azeotropic separation of water during 5 h. Then the
solid was filtered and washed with benzene.
Synthesis of Polyhydroxyalkyl Oxazolidines. To a solution of
the corresponding oxazolidine (0.98 mmol) in methanol (16 mL)
was added a saturated solution of ammonia in methanol (16 mL).
The transformation was monitored by thin layer chromatography
(benzene-methanol 9:1) and then the mixture was filtered off and
evaporated to dryness at a temperature below 30 °C. The title
compound is obtained as a solid.
1-Deoxy-1-(4-hydroxybenzylidene)amino-D-glucitol (7): method
A 92%, method B 99%; mp 227-228 °C; [R]23 +12.8 (c 0.5,
(2R,5S)-3-Acetyl-2-(4-hydroxyphenyl)-5-(D-arabino-tetritol-1-
yl)oxazolidine (42E,Z): 98%; recrystallized from ethanol had mp
233-234 °C; [R]25578 -37.0 (c 0.5, pyridine); IR (KBr) νmax 3500-
3100 (OH), 1612 (CdO), 1518 (aryl), 1234 (C-N), 1079, 1038
D
DMSO); IR (KBr) νmax 3400-2900 (OH), 1642 (CdN), 1608, 1587
(aryl), 1104, 1084, 1020 cm-1 (C-O); 1H NMR (400 MHz, DMSO-
d6) δ 9.70 (1H, bs, OH-arom), 8.16 (1H, s, CHdN), 7.55 (2H, d,
J ) 8.4 Hz, H-arom), 6.81 (2H, d, J ) 8.0 Hz, H-arom), 4.67-
4.31 (4H, bs, OH), 3.82 (1H, c, J1,2 ) J2,3 ) JC2,OH ) 4.8 Hz,
1
cm-1 (C-O); H NMR (400 MHz, CDCl3) δ 9.50 (2H, bs, OH-
arom), 7.30 (1H, bs, H-arom), 7.21 (1H, d, J ) 8.0 Hz, H-arom),
7.19 (1H, d, J ) 8.4 Hz, H-arom), 6.78 (1H, d, J ) 8.0 Hz, H-arom),
6.70 (1H, d, J ) 8.4 Hz, H-arom), 5.96 (1H, s, H-2E), 5.88 (1H, s,
H-2Z), 4.71 (2H, m, OH-1Z and OH-1E), 4.53 (2H, m, OH), 4.37
(2H, m, OH), 4.14 (1H, dt, J4A,5 ) 10.0 Hz, J5,1′ ) J4B,5 ) 6.4 Hz,
H-5Z), 4.06 (2H, m, J5,1′ ) J4B,5 ) 6.8 Hz, H-4E and H-5Z), 3.90
(1H, dd, J4,4 ) 9.6 Hz, J4,5 ) 5.6 Hz, H-4Z), 3.78 (2H, m, H-1′Z
and H-1′E), 3.59 (2H, m, H-4′Z and H-4′E), 3.49 (2H, m, H-3′Z and
H-3′E), 3.39 (2H, m, H-4′′Z and H-4′′E), 3.23 (2H, m, H-2′Z and
H-2′E), 3.13 (2H, t, J4,4 ) J4,5 ) 9.6 Hz, H-4E and H-4Z, oxaz),
2.00 (3H, s, CH3, Z), 1.58 (3H, s, CH3, E); 13C NMR (100 MHz,
CDCl3) δ 167.7, 167.5 (CdO), 158.5, 157.8, 131.0, 130.51, 129.3,
128.8, 115.7, 115.0 (C-arom), 89.8 (C-2E), 89.3 (C-2Z), 80.4 (C-
5Z), 79.8 (C-5E), 71.7, 71.6, 71.4, 71.3, 71.2, 70.9 (C-1′Z and C-1′E
and C-2′Z and C-2′E, C-3′Z and C-3′E), 63.8 (C-4Z and C-4E), 48.4
H-2), 3.68 (2H, m, H-1, H-3), 3.60 (1H, dd, J5,6 ) 2.4 Hz, J6,6′
)
8.8 Hz, H-6′), 3.51 (1H, m, H-1), 3.39 (1H, m, H-4, H-5, H-6′);
13C NMR (100 MHz, DMSO-d6) δ 161.7 (CdN), 160.2, 130.1,
127.9, 115.9 (C-arom), 73.0 (C-2), 72.4 (C-4), 71.9 (C-5), 70.3
(C-3), 64.0 (C-6), 63.6 (C-1). Anal. Calcd for C13H19NO6
(285.29): C, 54.73, H, 6.71, N, 4.91. Found: C, 54.58; H, 6.56,
N, 4.95.
Acetylation of D-Glucamine Schiff Bases: Synthesis of Chiral
Oxazolidines. To a solution of the corresponding 1-(arylmethylene)-
amino-1-deoxy-D-glucitol (5.0 mmol) in pyridine (6.7 mL) was
added acetic anhydride (6.5 mL). The reaction mixture was kept at
0 °C for 24 h, and then it was poured into ice-water. If the resulting
product was an oil this was extracted with chloroform (3 × 50
mL), and the organic layer was sequentially washed with 1 N HCl
J. Org. Chem, Vol. 73, No. 2, 2008 671

AÄ valos et al.
(C-4′E), 47.4 (C-4′Z), 23.7, 22.9 (4 CH3). Anal. Calcd for C15H21-
NO7 (327.33): C, 55.04, H, 6.47, N, 4.28. Found: C, 54.93, H,
6.39, N, 4.18.
Supporting Information Available: General methods, synthe-
ses, and structural characterization for compounds 8-20, 22, 23,
25-33, and 43-50, NMR spectra for all new compounds, variable-
temperature experiments for compounds 25, 27, 28, and 32,
Cartesian coordinates of compounds 25, 27, 28, and 32, and
crystallographic data for compound 33. This material is available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We thank the Ministry of Education and
Science and FEDER (Grant nos. CTQ2005-07676 and CTQ2007-
66641) for financial support. E. M. S. Pe´rez also thanks the
University of Extremadura for a research grant linked to the
NMR laboratory.
JO702149M
672 J. Org. Chem., Vol. 73, No. 2, 2008