Novel use of N-benzoyl-N, O-acetals as N-acylimine equivalents in asymmetric heterocycloaddition: An extended enantioselective pathway to β-benzamido aldehydes

Article abstract of DOI:10.1021/jo0203138

For the first time, easily available N-(α-methoxyalkyl)amides were successfully used as synthetic equivalents of N-acylimines in an asymmetric heterocycloaddition process. The facial-controlled formation of 6-alkoxydihydrooxazines was thus achieved by SnC

Full text of DOI:10.1021/jo0203138

Novel Use of N-Ben zoyl-N,O-a ceta ls a s N-Acylim in e Equ iva len ts in
Asym m etr ic Heter ocycloa d d ition : An Exten d ed En a n tioselective
P a th w a y to â-Ben za m id o Ald eh yd es
Patricia Gizecki, Robert Dhal, Ce´line Poulard, Pascal Gosselin, and Gilles Dujardin*
Laboratoire de Synthe`se Organique UMR CNRS 6011, Universite´ du Maine,
F-72085 Le Mans Cedex 9, France
dujardin@univ-lemans.fr
Received May 2, 2002
For the first time, easily available N-(R-methoxyalkyl)amides were successfully used as synthetic
equivalents of N-acylimines in an asymmetric heterocycloaddition process. The facial-controlled
formation of 6-alkoxydihydrooxazines was thus achieved by SnCl4-promoted heterocycloaddition
of (R)-O-vinyl pantolactone. By simple acidic hydrolysis of the crude heteroadducts, new â-aryl-
and â-alkyl-substituted benzamido aldehydes were thus obtained in good overall yields with high
enantioselectivities.
SCHEME 1 a
In tr od u ction
Enantioselective access to N-protected â-aminoalde-
hydes remains an important tool for several reasons.
First, such compounds are key intermediates in natural
product chemistry (aminocyclitols,1 piperidinic alkaloids,2
aspartic semialdehyde, and derivatives3). In the field of
â-peptide chemistry, N-protected â-aminoaldehydes may
also give access to reduced peptides4 and “carba”pep-
tides.5 Beyond the homologating pathways from R-amino
acid derivatives,4-6 there are few general methods for
preparing enantiopure N-protected â-aminoaldehydes:
Michael addition of a chiral lithium amide to an R-un-
saturated Weinreb amide,7 or addition of an ester enolate
to a chiral sulfinimine.2a The two methods require the
careful reduction of an N-methoxyamide or an ester in
the final step.
a
Reagents and conditions: (i) 5 mol % Yb(fod)3, cyclohexane,
As a first example of a novel entry in this field, we
recently disclosed8 a direct route to (R)-3-benzamido-3-
phenylpropanal (1a ), produced by hydrolysis of the
stable9 6-alkoxy-5,6-dihydro-4H-1,3-oxazine 2a , asym-
reflux, 72 h (Procedure A); (ii) chromatography over SiO2; (iii)
recrystallization (Et2O); (iv) 1 equiv of SnCl4, CH2Cl2, -78 °C, 0.5
h (Procedure B); (v) Amberlyst-15H+, acetone, H2O, rt, 15 h.
metrically obtained from the heterocycloaddition between
(R)-O-vinyl pantolactone (3)10 and (E)-N-benzoyl benzal-
dimine (4a ) (Scheme 1). This key cycloaddition gave high
and divergent diastereoselectivities when done in the
presence of a catalytic amount of Yb(fod)3 (Procedure A)
or of a stoichiometric amount of SnCl4 (Procedure B).
The two procedures, when completed by diastereomeric
purification and recrystallization of the main isomer
(endo-â for A, exo-â for B) and followed by acidic hydroly-
sis, gave efficient access to enantiopure (R)-(-)-3-benza-
mido-3-phenylpropanal (1a ). Further experiments by our
group have been directed at exploring the scope of these
(1) Suami, T.; Tadano, K.; Horiuchi, S. Bull. Chem. Soc. J pn. 1975,
48, 2895.
(2) (a) Davis, F. A.; Szewczyk, S. M. Tetrahedron Lett. 1998, 39,
5951. (b) Davies, S. B.; McKervey, M. A. Tetrahedron Lett. 1999, 40,
1229. (c) J efford, C. W.; Wang, J . B. Tetrahedron Lett. 1993, 34, 2911.
(3) Coulter, C. V.; Gerrard, J . A.; Kraunsoe, J . A. E.; Moore, D. J .;
Pratt, A. J . Tetrahedron 1996, 52, 7127.
(4) Matsuura, F.; Hamada, Y.; Shioiri, T. Tetrahedron 1994, 50,
9457.
(5) (a) Rodriguez, M.; Aumelas, A.; Martinez, J . Tetrahedron Lett.
1990, 31, 5153. (b) Rodriguez, M.; Heitz, A.; Martinez, J . Tetrahedron
Lett. 1990, 31, 7319. (c) Llinares, M.; Devin, C.; Azay, J .; Berge´, G.;
Fehrentz, J . A.; Martinez, J . Eur. J . Med. Chem. 1997, 32, 767.
(6) Toujas, J .-L.; J ost, E.; Vaultier, M. Bull. Soc. Chim. Fr. 1997,
713.
(7) Davies, S. G.; McCarthy, T. D. Synlett 1995, 700.
(8) Gizecki, P.; Dhal, R.; Toupet, L.; Dujardin, G. Org. Lett. 2000,
2, 585.
(9) This study exemplified that the nature of the vinyl ether
substituent can significantly modulate the stability of the desired
heteroadduct.
(10) (a) Dujardin, G.; Rossignol, S.; Brown, E. Synthesis 1998, 798.
(b) For the preliminary study (Scheme 1) and the present work, the
two-step method used for the preparation of (R)-(+)-O-vinyl-pantolac-
tone 3 is an adaptation (see Experimental Section) of the previously
reported procedure.10a
10.1021/jo0203138 CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/03/2003
4338
J . Org. Chem. 2003, 68, 4338-4344

N-Benzoyl-N,O-acetals as N-Acylimine Equivalents
SCHEME 2 a
Extending the scope of the method to various 3-aryl
and 3-alkyl analogues of 3-benzamido-3-phenylpropanal
1a required the preparation of various N-benzoyl aryl-
or alkylaldimines via the corresponding N,O-acetals 5b-f
(Scheme 2, Table 1). The N,O-acetals 5b-f were prepared
by Katritzky’s method12 in satisfactory yields. In the case
of 5f (R ) t-Bu), the convenient preparation of the
intermediate benzotriazole adduct 6f by azeotropic re-
moval of water required the catalytic use of TsOH (5 mol
%). Contrary to our expectations, the thermal dehydro-
methoxylation of N,O-acetals 5b-f gave heterogeneous
results. Although highly successful for some arylated
substrates (5b, R ) p-F-Ph; 5c, R ) p-MeO-Ph), the
above-mentionned conditions (toluene, reflux) proved
sluggish for 5d (R ) p-O2N-Ph) and totally ineffective
for alkyl substrates 5e,f. For 5e (R ) Bn), complete
isomerization of 4e into the corresponding enamide 815
was observed. For 5f (R ) t-Bu), no reaction was
observed.
As previously observed for N-acyl imine 4a , the
Yb(fod)3-catalyzed heterocycloaddition of (R)-O-vinyl
pantolactone (3) with N-acylimines 4b and 4c led with
high yields to the selective formation of the endo-â and
exo-R diastereoisomers (Scheme 3, Table 2). Assignment
of the stereostructures for 2b and 2c thus obtained was
based on consistent comparisons between NMR data of
each isomer (Table 3)16 and those of 2a -endo-â and 2a -
exo-R, for which the configuration was previously estab-
lished by crystallographic analysis and chemical corre-
lation (Table 3).8 As expected, the stability of the
dihydroxazines was satisfactory again, as exemplified by
the isolation of the major diastereomer 2b-endo-â after
chromatography or recrystallization in 64% and 51%
yield, respectively. However, relative to the case of 2a ,
the overall diastereoselectivity endo-â/exo-R significantly
decreased (Table 2, entries 4 and 6 vs 1).
a
Reagents and conditions: (i) benzamide, benzotriazole, tolu-
ene, ∆, -H2O; (ii) MeONa, MeOH, rt; (iii) toluene, ∆.
TABLE 1. P r ep a r a tion of N-Acylim in es 4
yield
(%)a
yield
(%)a
yield
(%)b
R
6
5
4
Ph
6a c
6b
59
60
63
80
65
73
5a d
5b
94
98
89
98
86
85
4a
4b
4c
4d
4e
4f
>85
>90
>95
36
p-F-C6H4
p-CH3O-C6H4
p-O2N-C6H4
Bn
6cc
6d c
6ed
6fc
5cd
5d e
5ed
5ff
0g
t-Bu
0h
a
b
d
Isolated yield. Crude product. c Reference 12a. Reference
12b. e Reference 13. f Reference 14. Isomerization of 4e into the
g
h
corresponding enamide 8. No reaction.
heterocycloadditions. In the present report, we describe
how the lack of availability of unstabilized N-acylimines
can be overcome by the use of stable N,O-acetals 5
(Scheme 2) acting as synthetic equivalents of N-acylimines
4 in the Lewis acid conditions used. We describe here
the enantioselective syntheses of various â-aryl and alkyl
â-benzamido aldehydes with ee ranging from 90 to 97%
using a straightforward sequence that avoids any inter-
mediate purification.
An interesting feature was that reaction with 3 also
proceeded when N-acylimines 4a -c were produced in
situ from N,O-acetals 5a -c under the thermal conditions
used (refluxing cyclohexane or toluene). 1H NMR showed
evidence of the transient N-acylimine formation.17 With
Yb(fod)3 as the catalyst, endo-â selectivity increased
(Table 2, entry 3 vs 2) as previously observed when
starting from preformed N-acylimine,8 but to a lesser
extent.18 Thus, considering our final goal, diastereomeric
distribution of the cycloadducts produced under these in
situ conditions remained unsatisfactory because of the
low overall â/R selectivity (e80/20, entries 3, 5, 7, and
8).
Resu lts a n d Discu ssion
Our preliminary heterocycloaddition study showed that
N-acylimine 4a , when prepared by previous procedures,11
led to some nonreproducible results. To overcome this
obstacle, we developed an alternative method for prepar-
ing 4a , based on the thermal dehydromethoxylation of
the readily available N,O-acetal 5a 12 (Scheme 2, Table
1, R ) Ph). Under the mild conditions used (toluene, 110
°C), N-acylimine 4a was cleanly obtained, without any
formation of side products such as bis-amide 7,13 and was
successfully used as heterodiene toward vinyl ether 3.
From this stereochemical point of view, results that
are more valuable were obtained with the use of stoichio-
metric amounts of SnCl4 at low temperature. Conversion
of N-acylimines 4b and 4c into the dihydrooxazines 2b
(14) Lokensgard, J . P.; Fischer, J . W.; Bartz, W. J . Org. Chem, 1985,
50, 5609.
(15) Enamide functionality of 8 was characterized in 1H NMR by
its vinylic proton at 6.27 ppm (d, J ) 14.8 Hz, E geometry).
(16) As a typical example of the correlations that can be established
between 1H diastereotopic signals of compounds 2a -c, the chemical
(11) Kupfer, R.; Meier, S.; Wurthwein, E. U. Synthesis 1984, 688.
(12) (a) Katritzky, A. R.; Pernak, J .; Fan, W.-Q.; Saczewski, F. J .
Org. Chem. 1991, 56, 4439. (b) Katritzky, A. R.; Fan, W.-Q.; Black,
M.; Pernak, J . J . Org. Chem. 1992, 57, 547. (c) Katritzky, A. R.; Pernak,
J .; Fan, W.-Q. Synthesis 1991, 868.
(13) Breuer, S. W.; Bernath, T.; Ben-Ishai, D. Tetrahedron, 1967,
23, 2869.
shifts of the N,O-acetalic protons H-6 are in the following range:
endo-â > δ exo-R > δ exo-â.
δ
(17) This observation is consistent with the fact that no Yb(fod)3-
catalyzed reaction occurred between 3 and the N,O-acetal 5f from
which the corresponding N-acylimine 4f could not have been obtained.
(18) A possible rationale would be the poisonous effect of the in situ
formed methanol on the lanthanide salt.
J . Org. Chem, Vol. 68, No. 11, 2003 4339

Gizecki et al.
SCHEME 3 a
a
Reagents and conditions: (i) Procedure A: 5 mol % Yb(fod)3, cyclohexane or toluene, reflux, 72 h; (ii) Procedure B: 1 equiv of SnCl4,
CH2Cl2, -78 °C, 0.5 h.
TABLE 2. Heter ocycloa d d ition of 3 w ith P r efor m ed or in Situ -F or m ed N-Acylim in es 4a -c
starting
material
(eq/3)
diastereomeric distribution of 2
cond.a
(time)
conv.
adduct
2/3 c
endo-â
exo-â
endo-R
exo-R
isolated yield of 2 (%)j
74 (pure 2a -endo-â)
1
2
3
4
5
6
7
8
9
4a (1)
5a (2)
5a (2)
A (3 d)
C (3 d)
A (3 d)
A (3 d)
A (3 d)e
A (8 d)f
A (8 d)e
A (1 d)e
B (0.5 h)
B (0.5 h)
B (1 h)a
2a
2a
2a
2b
2b
2c
2c
2d
2a
2b
2c
100
65
100
100
>95
72
90
38
72
86
67
64
63
57
13
9
0
6
8
0
11
2
10
3
87
91
86
0
1
2
0
0
0
0
0
0
0
5
10
55
18
14
22
34
27
40
0
4b (1.2)
5b (2)
4c (2)
5c (1.5)
5d (2)
4a (1.5)
4b (1.5)
4c (2)
64 (pure 2b-endo-â)d
78
48
100g
100h
100I
64 (pure 2a -exo-â)
10
11
0
8
1
a
Procedure A: 5% mol of Yb(fod)3, cyclohexane, reflux. Procedure B: 1 equiv mol of SnCl4, CH2Cl2, -78°C. Procedure C: no catalyst,
cyclohexane, reflux. -90 °C. c By 1H NMR. 51% after crystallization. e Solvent: toluene. f Solvent: toluene/cyclohexane, 1:3. 2% of
aldehyde 1a . 1% of aldehyde 1b. 3% of aldehyde 1c. After chromatography on SiO2.
b
d
g
h
i
j
TABLE 3. Selected 1H NMR Sp ectr oscop ic Da ta for Dih yd r ooxa zin es 2a -f
adduct
H-6 endo-â (ax)
H-6 endo-R (ax)
H-6 exo-R (eq)
H-6 exo-â (eq)
2
R
δ ppm (3J 6-5ax Hz)
δ ppm (3J 6-5ax Hz)
δ ppm (3J 6-5ax Hz)
δ ppm (3J 6-5ax Hz)
2a
2b
2c
2d
2e
2f
Ph
5.99 (8.9)
6.01 (8.4)
5.98 (8.9)
6.05 (7.9)
5.76 (7.9)
5.74 (9.8)
5.67 (9.4)
5.68
5.67
5.89 (3.0)
5.94 (3.0)
5.87 (3.0)
5.86
5.53 (3.0)
5.53 (3.0)
5.52 (3.0)
5.60 (2.5)
5.45 (3.0)
5.56 (3.0)
p-F-C6H4
p-CH3O-C6H4
p-O2N-C6H4
Ph-CH2
5.76
5.80 (3.0)
t-Bu
and 2c was complete in less than 1 h, and as for 2a , the
two â isomers were produced in both cases with a high
selectivity (Table 2, entries 9-11). Nevertheless, exten-
sion of the method was severely restricted by the un-
availability of the requisite N-acylimines 4. This prompted
us to examine how N,O-acetals 5 or even their precursors
6 could give access to the target heteroadducts 2 under
nonthermal conditions. For that purpose, given its well-
known ability to promote N-acyliminium chemistry to-
gether with the positive results obtained with neutral
acylimines (vide infra), SnCl4 appeared as the appropri-
ate Lewis acid. At -78 °C, the SnCl4-promoted reaction
between the benzotriazolyl adduct 6a 19 (in excess) and
vinyl ether 3 gave only low conversion of the latter
(<10%) into 1a . By contrast, under the same conditions
(-78 °C), an efficient reaction occurred between 3 and
the N,O-acetal 5a . Dihydrooxazine 2a was produced as
the main product together with small amounts of alde-
hyde 1a (Table 4, entry 1). Furthermore, the stereochem-
ical outcome of this reaction showed an amazing simi-
larity with those that prevailed when starting from 3 and
4a (Table 2, entry 9): the two â isomers were obtained
again almost exclusively in an 84/16 exo/ endo ratio. This
unexpected result, very positive in terms of synthetic
potential, poses some questions about the real nature of
the reactive intermediate produced by the N,O-acetal in
the presence of SnCl4.20 At -10 °C, under stoichiometric
conditions, N,O-acetal 5a gave rise to a nearly quantita-
tive reaction with vinyl ether 3 (Table 4, entry 2). This
new procedure also proved fruitful for new substrates.
Thus, dihydrooxazines 2d -f bearing respectively a p-
nitrophenyl, benzyl, or tert-butyl group at C-4 were
(19) Katritzky’s group recently reported the ZnBr2-mediated R-
amidoalkylation of ethyl vinyl ether with the use of 6-type N-(1-
benzotriazol-1-alkyl)amides: Katritzky, A. R.; Fan, W. Q.; Silina, A.
J . Org. Chem. 1999, 64, 7622.
(20) As suggested with N-acylimine 4a in our previous paper,8 the
endo/exo stereodivergency observed with N,O-acetal 5a -c between Yb-
(fod)3-catalyzed and SnCl4-promoted reactions most probably indicates
two different mechanisms. The first role of SnCl4 would be to promote
the formation of the electrophilic reactive heterodiene from the N,O-
acetal 5a -c. As the present time, we have little evidence about (i) the
real nature of this intermediate (neutral N-acylimine or more likely
N-acyliminium salt 21) and (ii) the subsequent putative influence of the
stannic salt (SnCl3OMe or SnCl4) on the expected stepwise mechanism
that would involve this intermediate.
4340 J . Org. Chem., Vol. 68, No. 11, 2003

N-Benzoyl-N,O-acetals as N-Acylimine Equivalents
TABLE 4. Sn Cl4-P r om oted Rea ction of 3 a n d N,O-Aceta ls 5
diastereomeric distribution of 2c
5:3a
ratio
temp
(°C)
time
(min)
adduct
2:1
ratio
5
2
conv./5b,c
endo-â
exo-â
endo-R
exo-R
1
2
3
4
5
6
7
8
9
5a
5a
5b
5c
5d
5d
5e
5e
5f
1:1
1:1
1:1
1:1
1:1
1:2
1:2
1:2
1:1.5
-78
-10
-10
-10
-10
-10
-78
-10
-10
15
15
15
15
15
15 d
30
30
30
2a
2a
2b
2c
2d
2d
67
100
80
82
36
63
0
97:3
99:1
99:1
99:1
91:9
82:18d
16
18
13
14
20
20
84
0
0
0
0
3
3
0
80.5
85.5
83
75
77
1.5
1.5
3
2
0
2e
2f
71
69
100:0
100:0
22
24
77
76
0
0
1
0
a
b
d
1 equiv of SnCl4, CH2Cl2 , -78°C (Procedure B). Into 2/1 mixture. c Estimated by 1H NMR. No improvement with 30 min reaction.
conveniently obtained, but not isolated,22 from the cor-
responding N,O-acetals 5d -f (Table 4), provided that the
conditions fit with the reactivity of the substrate. Indeed,
compared to the other aryl-substituted N,O-acetals 5a -c
under the same conditions, N,O-acetal 5d (R ) p-NO2-
C6H4) proved less reactive (-10 °C, 1:1 ratio, entry 5).23
In this case, more convenient conditions were found with
an excess of vinyl ether 3 (entry 6). 5e and 5f proved
reactive only at -10 °C in an extended time (entry 7),
and thus again an excess of vinyl ether 3 was necessary
because of its sensitivity under these harsher conditions
(entries 8 and 9). However, homogeneous high levels of
â-stereoselectivity were observed in both aryl and alkyl
series (â: R ratio g97:3).
SCHEME 4
For all heteroadducts 2b-f obtained with SnCl4, as-
signment of the exo-â topicity to the main isomer was
based on a consistent H NMR correlation with 2a -exo-
TABLE 5. En a n tioselective Syn th esis of
â-Am id oa ld eh yd es 1a -f (w ith ou t p u r ifica tion of
in ter m ed ia te a d d u cts 2a -f)
1
â8 (Table 3). In each case, the exo-â proton H-6 is
significantly high-field displaced (e5.60 ppm) relative to
the H-6 protons of other configurations. The exo character
is evidenced by the weak value of the coupling constant
(3.0 Hz or less) with the vicinal axial proton.
preparation
method
for 2
alde- overall
hyde
1
starting
material
yielda eec
R
(%)
(%)
1
2
3
4
5
6
7
8
Ph
p-F-C6H4
Ph
p-F-C6H4
p-CH3O-C6H4 5c
p-O2N-C6H4
Bn
t-Bu
4a
4b
5a
5b
Table 2, entry 9 (R)-1a
Table 2, entry 10 (R)-1b
Table 4, entry 2 (R)-1a
Table 4, entry 3 (R)-1b
Table 4, entry 4 (R)-1c
Table 4, entry 6 (R)-1d
Table 4, entry 8 (S)-1e
Table 4, entry 9 (R)-1f
63b
80b
63
62
62
45
70
43
97
97
96
96
93
90
96
97
Since all the SnCl4-promoted heterocycloadditions of
3 we tested (using either 4 or 5 as the co-reactant) led to
a high overall â-control (g97:3), acidic hydrolysis of the
dihydrooxazines thus obtained was performed without
any prior purification (Scheme 4, Table 5). Treatment of
the crude heteroadducts 2a -f with Amberlyst 15-H+ in
aqueous acetone at room temperature gave the desired
â-benzamido aldehydes 1a -f in 43-80% overall yields,
with an efficient simultaneous recovery of the chiral
5d
5e
5f
a
b
Per 5 unless otherwise noted. Per 3. c Determined by 1H
NMR with 25-100 mol % of (+)-Eu(tfc)3.
1
auxiliary. H NMR measurements with use of (+)-Eu-
Supporting Information) of the corresponding â-benzyl-
â-benzamidoacetal 9 deriving from (+)-1,4-bis-O-(4-chlo-
robenzyl)-D-threitol25 (Scheme 5).
(tfc)3 as the chiral shift reagent gave homogeneous ee
values ranging from 90 to 97%.24
Finally, assignment of the absolute configuration or
each â-benzamido aldehyde 1 was unambiguously de-
duced from the exo-â character of its main cyclic precur-
sor. To ascertain the configuration of the â-benzyl-â-
benzamido aldehyde 1e, additional proof was introduced
by way of the X-ray crystallographic structure (see
Con clu sion
In summary, the present study demonstrates the
synthetic usefulness of SnCl4 as a Lewis acid promotor
for the asymmetric synthesis of type-2 dihydrooxazines.
First, through the novel use of N,O-acetals 5,26 this
method can be applied to a wide range of starting
compounds, easily obtained from the corresponding al-
dehyde via Katritzky’s procedure. In addition, much
benefit can be gained from the unique stereochemical
outcome that prevails when 3 is used under these Lewis
acid conditions, to obtain the corresponding â-benzamido
(21) Meester, W. J . N.; Rutjes, F. P. J . T.; Hermkens, P. H. H.;
Hiemstra, H. Tetrahedron Lett. 1999, 40, 1601.
(22) Contrariwise to dihydroxazines 2a -c, dihydroxazines 2d (R )
C6H4NO2) and 2e (R ) Bn) showed a weak stability over SiO2 and
column chromatography gave low yields of purified product (17% for
2e).
(23) This fact agrees with the total lack of reactivity we observed
under the same conditions when using imino ester 5g (R ) CO2Me) if
we assume that the reaction is disfavored when N,O-acetal is substi-
tuted by an electron-withdrawing group.
(24) The (R)-(+)-O-vinyl-pantolactone 3 used in this study was found
to have 98.5% ee (determined by chiral GC, see Experimental Section).
(25) Tamoto, K.; Sugimori, M.; Terashima, S. Tetrahedron 1984, 40,
4617.
J . Org. Chem, Vol. 68, No. 11, 2003 4341

Gizecki et al.
SCHEME 5 a
column chromatography (silica gel 20/1; toluene/AcOEt from
100:0 to 97:3).
N-[(4-F lu or op h en yl)(m eth oxy)m eth yl]ben za m id e (5b).
Colorless crystal (98%), mp 106-108 °C (CH2Cl2); IR (Nujol)
3271, 1643, 1525, 1223, 1092, 982, 845, 700 cm-1 1H NMR
;
(400 MHz, CDCl3) δ 7.81 (2H, d, J ) 6.9 Hz), 7.54 (1H, t, J )
7.4 Hz), 7.46 (4H, m), 7.06 (2H, t, J ) 8.6 Hz), 6.57 (1H, d, J
) 9.4 Hz), 6.36 (1H, d, J ) 9.4 Hz), 3.54 (3H, s); 13C NMR
1
(100 MHz, CDCl3) δ 167.1, 162.3 (d, J C-F ) 247.2 Hz), 134.8
4
3
(d, J C-F ) 3.0 Hz), 133.2, 131.8, 128.3, 127.4 (d, J C-F ) 8.4
2
Hz), 126.8, 115.1 (d, J C-F ) 22.1 Hz), 81.0, 55.8. Anal. Calcd
for C15H14FNO2: C, 69.49; H, 5.44; F, 7.13; N, 5.44. Found:
C, 69.51; H, 5.44; F, 7.16; N 5.41.
a
Reagents and conditions: (a) (+)-1,4-bis-O-(4-chlorobenzyl)-
D-threitol, p-TsOH‚H2O, C6H6, ∆.
Gen er a l P r oced u r e for th e P r ep a r a tion of N-Acyl-
im in es 4a -d . By adaptation of the Breuer’s method,13
a
solution of freshly chromatographed N,O-acetal 5 (1 mmol) in
anhydrous toluene (15 mL) under nitrogen was stirred at 110
°C for 24-60 h. After concentration in vacuo, N-acylimine 4
was obtained without further purification.
aldehydes 1, as yet not described to our knowledge, with
a valuable enantioselectivity and by a very simple
preparative pathway. Extension of this methodology to
other types of N-substituted N,O-acetals is under inves-
tigation in our laboratory.
N-[(E)-4-F lu or oben zylid en e]ben za m id e (4b). Yellowish
crystal (100%, purity g90%), mp 82-83.5 °C (toluene); IR
1
(Nujol) 1674, 1520, 1354, 1049, 852, 715 cm-1; H NMR (400
Exp er im en ta l Section
MHz, CDCl3) δ 8.76 (1H, s), 8.16 (2H, dd, J ) 8.4 and 1.5 Hz),
8.00 (2H, dd, J ) 8.9 and 5.4 Hz), 7.60 (1H, t, J ) 7.4 Hz),
Gen er a l Meth od s. Reagents were purchased from com-
mercial suppliers and used without purification. Commercial
(R)-(-)-pantolactone was recrystallized from xylene before use
(99.5% ee, determined by chiral GC). All solvents were dried
following standard procedures. Chromatography was per-
formed with 40-60 µm silica gel under medium pressure (1
bar). All melting points are uncorrected. Infrared spectra were
performed on an FT spectrophotometer. 1H and 13C NMR
spectra were recorded on 200- and 400- MHz spectrometers
in CDCl3 with TMS as reference. LCMS (EI or CI) were
7.49 (2H, t, J ) 7.6 Hz), 7.21 (2H, dd, J ) 8.9 and 8.4 Hz); 13
C
1
NMR (100 MHz, CDCl3) δ 180.7, 165.8 (d, J C-F ) 255.6 Hz),
3
4
163.2, 133.5, 133.2, 132.2 (d, J C-F ) 9.2 Hz), 130.9 (d, J C-F
) 3.0 Hz), 130.1, 128.5, 116.3 (d, 2J C-F ) 22.1 Hz). HRMS calcd
for C14H10FNO [M]•+ 227.0746, found 227.0743.
(R)-(+)-O-Vin yl P a n tola cton e (3). By adaptation of the
previous method,10a to a solution of the crude mixed acetal
(4.94 g, 24.4 mmol, quantitatively obtained from (R)-(-)-
pantolactone and ethyl vinyl ether10a) in anhydrous dichlo-
romethane (30 mL) under argon was added dropwise at 0 °C
triethylamine (1.2 equiv, 29.3 mmol, 4.1 mL) and then tri-
methylsilyl trifluoromethanesulfonate (1.1 equiv, 26.9 mmol,
4.9 mL). The reaction mixture was allowed to warm at room
temperature and was stirred for 4 h. The cooled solution was
treated at 0 °C with aqueous 1 M NaOH (7.4 mL) and then
immediately diluted with pentane (300 mL). The resulting
organic layer was washed (H2O), dried (MgSO4), and concen-
trated in vacuo at room temperature. Purification by column
chromatography (silica gel 15/1; cyclohexane/AcOEt from 95:5
to 80:20) afforded the vinyl ether 3 as a colorless oil (2.70 g,
70%). The enantiopurity of (R)-(+)-O-vinyl pantolactone 3 was
determined by chiral GC (column: Restek-â Dex Sm (25m ×
0.25 mm), inlet 200 °C, detector 200 °C, oven 120 °C for 1 min
then 2 °C/min until 180 °C): tR ) 8.4 min, (R)-(+)-3, tR ) 9.4
min, (S)-(-)-3, ee 98.5%.
P r ep a r a tion of Dih yd r ooxa zin es 2 w ith Yb(fod )3: P r o-
ced u r e A (w ith fin a l p u r ifica tion ). 6-[(2R)-3,3-Dim eth yl-
γ-b u t yr ola ct on e-2-yl]oxy-4-(4-flu or op h en yl)-2-p h en yl-
5,6-d ih yd r o-4H-1,3-oxa zin es (2b-en d o-â) (4R,6S) a n d (2b-
exo-r) (4S,6S). A solution of N-acylimine 4b (556 mg, 2.45
mmol), (R)-(+)-O-vinylpantolactone (3) (318 mg, 2.04 mmol),
and Yb(fod)3 (107 mg, 0.1 mmol) in dry cyclohexane (40 mL)
was refluxed under nitrogen for 3 d. After concentration in
vacuo, the crude product (endo-â/exo-R: 86/14) was chromato-
graphed (silica gel 50/1, toluene/AcOEt 99:1 to 95:5) to yield
in order of elution the pure dihydrooxazines 2b-exo-R (56 mg,
9%) and 2b-endo-â (381 mg, 64%).
performed on
a particle beam mass spectrometer. High-
resolution mass spectra were performed at the University of
Rennes I.
Gen er a l P r oced u r e for th e P r ep a r a tion of Ben zotr ia -
zole Der iva tives 6a -f.12 A solution of an aldehyde (0.05 mol),
benzotriazole (5.96 g, 0.05 mol), and benzamide (6.06 g, 0.05
mol) in anhydrous toluene (or benzene for 6f) (50 mL) was
heated to reflux for 48-72 h with azeotropic distillation of
water. After concentration in vacuo, the residue was treated
with ethyl ether (100 mL) and the resulting solid was
separated by filtration.
N-[Ben zot r ia zol-1-yl-(4-flu or op h en yl)m et h yl]b en za -
m id e (6b). Colorless crystal (60,5%), mp 163-164 °C (MeOH);
IR (Nujol) 3305, 1637, 1514, 1232, 1155, 1090, 845, 787, 742,
1
690 cm-1; H NMR (400 MHz, DMSO-d6) δ 10.22 (1H, d, J )
8.4 Hz), 8.21 (1H, d, J ) 7.9 Hz), 8.09 (1H, d, J ) 8.4 Hz),
7.95 (3H, m), 7.58 (2H, m), 7.50 (4H, m), 7.43 (1H, t, J ) 7.4
Hz), 7,27 (2H, t, J ) 9.2 Hz). Anal. Calcd for C20H15FN4O: C,
69.36; H, 4.37; F, 5.49; N, 16.18. Found: C, 69.37; H, 4.43; F,
5.43; N, 16.21.
Gen er a l P r oced u r e for th e P r ep a r a tion of N,O-Aceta ls
5a -f.12b The benzotriazole derivative 6 (10 mmol) was added
to a solution of sodium methoxide (2 mmol) in methanol
(10 mL), and the mixture was stirred at room temperature
for one night. After dilution with water (20 mL), or adjust-
ment to pH 7 with HCl for products 5c and 5e, the precipi-
tate was separated by filtration. After solubilization of the
product in dichloromethane or ethyl acetate, the organic layer
was dried (MgSO4), concentrated in vacuo, and purified by
2b-endo-â: amorphous white solid;1H NMR (400 MHz,
CDCl3) δ 8.02 (2H, d, J ) 6.9 Hz), 7.48 (1H, t, J ) 7.1 Hz),
7.41 (2H, t, J ) 7.4 Hz), 7.39 (2H, dd, J ) 8.4 and 5.4 Hz),
7.05 (2H, ∼t, J ∼ 8.9 Hz), 6.01 (1H, dd, J ) 8.4 and 3.4 Hz),
4.79 (1H, dd, J ) 10.3 and 4.4 Hz), 4.45 (1H, s), 4.05 and 4.01
(2H, 2d, J ) 8.9 Hz), 2.62 (1H, ddd, J ) 13.3, 4.4, and 3.4 Hz),
1.87 (1H, ddd, J ) 13.3, 10.3, and 8.4 Hz), 1.27 (3H, s), 1.02
(3H, s); MS (EI) m/z (%) 393 [M]•+ (4.7), 270 [M - C6H9O2]+
(11), 123 (25.5), 105 [C7H5O]+ (100), 77 [C6H5]+ (44.4).
2b-exo-R: colorless crystal, mp 159-160 °C (toluene); IR
(26) To our knowledge, 5-type compounds have never been used for
the preparation of 2-type 6-alkoxy dihydrooxazines. By contrast, the
successful use of N,O-acetals as iminium precursors in the asymmetric
preparation of â-aminocarbonyl compounds is well documented, see:
(a) Ferraris, D.; Young, B.; Dudding, T.; Leckta, T. J . Am. Chem. Soc.
1998, 120, 4849. (b) Kise, N.; Ueda, N. Org. Lett. 1999, 1, 1803. (c)
Kise, N.; Ueda, N. J . Org. Chem. 1999, 64, 7511. (d) Pilli, R. A.; Alves,
C. F.; Bo¨ckelmann, M. A.; Mascarenhas, Y. P.; Nevy, J . G.; Vencato, I.
Tetrahedron Lett. 1999, 40, 2891. (e) Ferraris, D.; Young, B.; Dudding,
T.; Leckta, T. J . Am. Chem. Soc. 2002, 124, 67 and references therein.
1
(Nujol) 1784, 1655, 1508 cm-1; H NMR (400 MHz, CDCl3) δ
4342 J . Org. Chem., Vol. 68, No. 11, 2003

N-Benzoyl-N,O-acetals as N-Acylimine Equivalents
8.06 (2H, d, J ) 6.9 Hz), 7.48 (1H, t, J ) 7.4 Hz), 7.42 (4H,
m), 7.07 (2H, ∼t, J ) 8.6 Hz), 5.94 (1H, br s), 4.90 (1H, dd, J
) 12.3 and 4.9 Hz), 4.47 (1H, s), 4.08 and 4.01 (2H, 2d, J )
8.9 Hz), 2.50 (1H, ddd, J ) 13.8, 4.9, and 2.0 Hz), 1.87 (1H,
ddd, J ) 13.8, 12.3, and 3.0 Hz), 1.13 (3H, s), 1.09 (3H, s); MS
(EI) m/z (%) 383 [M]•+ - (1), 270 [M - C6H9O2]+ (13.4), 151
(8), 123 (19.3), 105 [C7H5O]+ (100), 77 [C6H5]+ (22.9). Anal.
Calcd for C22H22FNO4: C, 68.92; H, 5.78; F, 4.95; N 3.65.
Found: C, 68.47; H, 5.77; F, 5.08; N, 3.68.
6-[(2R)-3,3-Dim eth yl-γ-bu tyr olacton e-2-yl]oxy-4-(4-m eth -
oxyp h en yl)-2-p h en yl-5,6-d ih yd r o-4H -1,3-oxa zin es (2c-
en d o-â) (4R,6S) a n d (2c-exo-r) (4S,6S). A solution of N,O-
acetal 5c12b (333 mg, 1.23 mmol, 1.5 equiv), (R)-(+)-O-vinyl
pantolactone (3) (124 mg, 0.79 mmol), and Yb(fod)3 (42 mg,
0.04 mmol) in dry toluene (16 mL) was refluxed under nitrogen
for 8 d. After concentration in vacuo, the crude product was
chromatographed (silica gel 30 /1, toluene/AcOEt 100:0 to 95:
5) to yield in order of elution the pure dihydrooxazine 2c-exo-R
(85.5 mg, 27%) and a (84:16) mixture of the dihydrooxazines
2c-endo-â and 2c-exo-â (139.3 mg, 45%).
(R)-(+)-O-vinyl pantolactone 3 (1 equiv, 130 mg, 1 mmol) and
N,O-acetal 4b (1 equiv, 260 mg, 1 mmol) at -10 °C (15 min).
6-[(2R)-3,3-Dim eth yl-γ-bu tyr olacton e-2-yl]oxy-4-(4-m eth -
oxyp h en yl)-2-p h en yl-5,6-d ih yd r o-4H-1,3-oxa zin e (2c). Ob-
tained with procedure C as a foam (exo-â/endo-â/exo-R 83/14/
3) from N,O-acetal 5c13 (1 equiv, 200 mg, 0.735 mmol) and
(R)-(+)-O-vinyl pantolactone 3 (1 equiv) in dichloromethane
(11 mL) at -10 °C (15 min). Isomer (4R,6R) (2c-exo-â): 1H
NMR (400 MHz, CDCl3) δ 8.05 (2H, d, J ) 6.9 Hz), 7.51 (1H,
t, J ) 8.6 Hz), 7.40 (2H, t, J ) 7.9 Hz), 7.32 (2H, d, J ) 8.9
Hz), 6.91 (2H, d, J ) 8.9 Hz), 5.52 (1H, dd, J ) 3.4 and 3.0
Hz), 4.90 (1H, dd, J ) 10.8 and 4.9 Hz), 4.25 (1H, s), 4.03 and
3.92 (2H, 2d, J ) 8.9 Hz), 3.81 (3H, s), 2.33 (1H, ddd, J )
13.5, 4.9, and 3.4 Hz), 1.92 (1H, ddd, J ) 13.5, 10.8, and 3.0
Hz), 1.27 (3H, s), 1.17 (3H, s).
6-[(2R)-3,3-Dim eth yl-γ-bu tyr ola cton e-2-yl]oxy-4-(4-n i-
tr op h en yl)-2-p h en yl-5,6-d ih yd r o-4H-1,3-oxa zin e (2d ) Ob-
tained with procedure C as a foam (exo-â/endo-â/endo-R 77/
20/3) from N,O-acetal 5d 13 (1 equiv, 175 mg, 0.61 mmol) and
(R)-(+)-O-vinyl pantolactone 3 (2 equiv) in dichloromethane
(18 mL) at -10 °C (15 min). Isomer (4R,6R) (2d -exo-â): 1H
NMR (400 MHz, CDCl3) δ 8.24 (2H, d, J ) 8.6 Hz), 8.05 (2H,
d, J ) 7.4 Hz), 7.64 (2H, d, J ) 8.6 Hz), 7.47 (1H, t, J ) 7.4
Hz), 7.41 (2H, t, J ) 7.4 Hz), 5.60 (1H, t, J ) 2.5 Hz), 5.05
(1H, dd, J ) 11.8 and 4.9 Hz), 4.32 (1H, s), 4.04 and 3.96 (2H,
2d, J ) 8.9 Hz), 2.42 (1H, ddd, J ) 13.8, 4.9, and 2.5 Hz), 1.85
(1H, ddd, J ) 13.8, 11.8, and 2.5 Hz), 1.31 (3H, s), 1.19 (3H,
s). HRMS calcd for C23H22N2O6 [M]•+ 410.14779, found 410.1471.
Selected data for 2d -endo-â: 1H NMR (400 MHz, CDCl3) δ 6.05
(1H, dd, J ) 7.9 and 3.4 Hz, H-6), 4.92 (1H, dd, J ) 10.1 and
5.2 Hz), 4.44 (1H, s). Selected data for 2d -endo-R: 1H NMR
(400 MHz, CDCl3) δ 5.76 (1H, m).
2c-endo-â: yellow oil; 1H NMR (400 MHz, CDCl3) δ 8.02 (2H,
d, J ) 6.9 Hz), 7.46 (1H, t, J ) 7.1 Hz), 7.40 (2H, t, J ) 7.4
Hz), 7.34 (2H, d, J ) 8.6 Hz), 6.90 (2H, d, J ) 8.6 Hz), 5.98
(1H, dd, J ) 8.9 and 3.4 Hz), 4.76 (1H, dd, J ) 10.8 and 4.4
Hz), 4.45 (1H, s), 4.03 and 4.00 (2H, 2d, J ) 8.9 Hz), 3.81 (3H,
s), 2.61 (1H, ddd, J ) 13.3, 4.4, and 3.4 Hz), 1.85 (1H, ddd, J
) 13.3, 10.8, and 8.9 Hz), 1.27 (3H, s), 1.04 (3H, s); MS (EI)
m/z (%) 395 [M]•+ (2.7), 282 [M - C6H9O2]+ (8.3), 239 (10.8),
105 [C7H5O]+ (100), 77 [C6H5]+ (25.6).
2c-exo-R: yellow oil; 1H NMR (400 MHz, CDCl3) δ 8.04 (2H,
d, J ) 7.4 Hz), 7.46 (1H, t, J ) 7.4 Hz), 7.40 (2H, t, J ) 7.4
Hz), 7.35 (2H, d, J ) 8.4 Hz), 6.92 (2H, d, J ) 8.9 Hz), 5.87
(1H, ∼t, J ) 2.5 Hz), 4.85 (1H, dd, J ) 12.3 and 4.9 Hz), 4.40
(1H, s), 4.04 and 3.97 (2H, 2d, J ) 8.9 Hz), 3.82 (3H, s), 2.47
(1H, ddd, J ) 13.8, 4.9, and 2.0 Hz), 1.88 (1H, ddd, J ) 13.8,
12.3, and 3.0 Hz), 1.23 (3H, s), 1.19 (3H, s); SM (EI) m/z 395
[M]•+ (2.2), 282 [M - C6H9O2]+ (17.8), 105 [C7H5O]+ (100), 77
[C6H5]+ (24.8). HRMS calcd for C23H25NO5 [M]•+ 395.17327,
found 395.1718.
Gen er a l P r oced u r es for th e P r ep a r a tion of Dih yd r o-
oxa zin es 2 w ith Sn Cl4 w ith ou t P u r ifica tion . P r oced u r e
B (w ith N-a cylim in e 4). To a cooled solution at -78 °C under
nitrogen of (R)-(+)-O-vinyl pantolactone (3) (1 equiv) and
N-acylimine 4 (1.5-2 equiv) in dichloromethane (15 mL/mmol)
was rapidly added dropwise SnCl4 (1 equiv, 1 M in dichloro-
methane). After being stirred (30 to 60 min), the mixture was
quenched with saturated aqueous NaHCO3 solution (30 mL/
mmol). After return to room temperature and extraction with
dichloromethane (2 × 15 mL/mmol), the resulting organic layer
was dried (MgSO4) and concentrated in vacuo at room tem-
perature. The crude dihydrooxazine 2 is used for hydrolysis
without further purification (analytical samples were obtained
by chromatography).
P r oced u r e C (w ith N,O-a ceta l 5). To a cooled solution at
-10 °C under nitrogen of N,O-acetal 5 (1 equiv) and (R)-(+)-
O-vinyl pantolactone (3) (1.5-2 equiv) in dichloromethane
(15-30 mL/mmol) was rapidly added SnCl4 (1 equiv, 1 M in
dichloromethane). After being stirred (15 to 30 min), the
mixture was quenched with saturated aqueous NaHCO3
solution (30 mL/mmol) and treated as in protocol B.
4-Ben zyl-6-[(2R)-3,3-dim eth yl-γ-bu tyr olacton e-2-yl]oxy-
2-p h en yl-5,6-d ih yd r o-4H-1,3-oxa zin e (2e). Obtained with
procedure C as a foam (exo-â/endo-â/exo-R 77/22/1) from N,O-
acetal 5e12b (1 equiv, 401 mg, 1.57 mmol) and (R)-(+)-O-vinyl
pantolactone (3) (2 equiv) in dichloromethane (47 mL) at -10
°C (30 min). Isomer (4S,6R) (2e-exo-â): 1H NMR (400 MHz,
CDCl3) δ 7.97 (2H, d, J ) 6.4 Hz), 7.38 (3H, m), 7.35 to 7.20
(5H, m), 5.45 (1H, dd, J ) 3.0 and 2.5 Hz), 4.14 (1H, s), 4.02
(1H, s), 3.97 and 3.87 (2H, 2d, J ) 8.9 Hz), 3.19 (1H, dd, J )
13.3 and 5.9 Hz), 2.79 (1H, dd, J ) 13.3 and 8.4 Hz), 1.96 (1H,
ddd, J AB ) 13.8, 4.9, and 2.5 Hz), 1.62 (1H, ddd, J ) 13.8,
11.0, and 3.0 Hz), 1.19 (3H, s), 1.07 (3H, s). HRMS calcd for
C
16H18NO4 [M - CH2Ph]+ 288.1236, found 288.1246. Selected
data for 2e-endo-â: 1H NMR (400 MHz, CDCl3) δ 5.76 (1H,
dd, J ) 7.9 and 3.4 Hz), 4.40 (1H, s), 2.19 (1H, ddd, J ) 13.8,
4.9, and 3.4 Hz), 1.65 (1H, ddd, J ) 13.8, 8.9, and 7.9 Hz).
Selected data for 2e-exo-R: 1H NMR (400 MHz, CDCl3) δ 5.80
(1H, ∼t, J ) 2.5 Hz).
4-ter t-Bu tyl-6-[(2R)-3,3-d im eth yl-γ-bu tyr ola cton e-2-yl]-
oxy-2-p h en yl-5,6-d ih yd r o-4H-1,3-oxa zin e (2f). Obtained
with procedure C as a foam (exo-â/endo-â 76/24) from N,O-
acetal 5f14 (1 equiv, 94.5 mg, 0.43 mmol) and (R)-(+)-O-vinyl
pantolactone (3) (1.5 equiv) in dichloromethane (13 mL) at -10
°C (30 min). Isomer (4R,6R) (2f-exo-â): 1H NMR (400 MHz,
CDCl3) δ 7.97 (2H, d, J ) 6.4 Hz), 7.39 (3H, m), 5.56 (1H, dd,
J ) 3.0 and 2.0 Hz), 4.19 (1H, s), 4.00 to 3.90 (2H, m), 3.40
(1H, dd, J ) 12.5 and 4.9 Hz), 2.04 (1H, ddd, J ) 13.3, 4.9,
and 2.0 Hz), 1.65 (1H, ∼dt, J ) 13.3 and 3.0 Hz), 1.22 (3H, s),
1.13 (3H, s), 1.01 (9H, s). Selected data for 2f-endo-â: 1H NMR
(400 MHz, CDCl3) δ 5.74 (1H, dd, J ) 9.8 and 3.0 Hz), 4.45
(1H, s), 3.25 (1H, dd, J ) 12.3 and 4.4 Hz), 2.35 (1H, m).
6-[(2R )-3,3-Dim e t h yl-γ-b u t yr ola ct on e -2-yl]oxy-4-(4-
flu or op h en yl)-2-p h en yl-5,6-d ih yd r o-4H-1,3-oxa zin e (2b).
(i) Obtained with procedure B as a foam (exo-â/endo-â: 91/9)
from (R)-(+)-O-vinyl pantolactone 3 (1 eq, 210 mg, 1.34 mmol)
and N-acylimine 4b (1.5 equiv, 470 mg, 2.07 mmol) at -78 °C
(30 min). Isomer (4R,6R) (2b-exo-â): 1H NMR (400 MHz,
CDCl3) δ 8.06 (2H, d, J ) 6.9 Hz), 7.39 (5H, m), 7.05 (2H, t, J
∼ 8.8 Hz), 5.53 (1H, t, J ) 3.0 Hz), 4.91 (1H, dd, J ) 11.3 and
4.9 Hz), 4.27 (1H, s), 4.02 and 3.93 (2H, 2d, J ) 8.9 Hz), 2.35
(1H, ddd, J ) 13.8, 4.9, and 3.0 Hz), 1.87 (1H, ddd, J ) 13.8,
11.3, and 3.0 Hz), 1.28 (3H, s), 1.17 (3H, s). (ii) Obtained with
procedure C as a foam (exo-â/endo-â/exo-R 85.5/13/1.5) from
Gen er a l P r oced u r e for th e P r ep a r a tion of â-Am id o-
a ld eh yd es 1a -f: Hyd r olysis27 of Cr u d e Dih yd r ooxa zin es
2a -f. A solution of dihydrooxazine 2 (1 mmol, see Table 5) in
a mixture of acetone (8 mL) and water (2 mL) was stirred with
Amberlyst-15 H+ (320 mg) at room temperature under nitrogen
for one night. After filtration and concentration in vacuo, the
residue was dissolved in dichloromethane (20 mL), and the
organic layer was dried (MgSO4) and concentrated in vacuo.
The residual (R)-(-)-pantolactone was removed by sublimation
J . Org. Chem, Vol. 68, No. 11, 2003 4343

Gizecki et al.
with use of a bulb-to-bulb distillation apparatus (100 °C, 10-1
Torr, 1 h) and the â-amidoaldehyde 1 thus obtained was
chromatographed (silica gel 50/1; toluene/AcOEt 100:0 to
80:20).
MHz, CDCl3) δ 9.79 (1H, ∼t, J ∼ 1.0 Hz), 7.68 (2H, dt, J ) 6.9
and 1.5 Hz), 7.50 (1H, tt, J ) 7.4 and 1.5 Hz), 7.42 (2H, tt, J
) 7.4 and 1.5 Hz), 7.33 (2H, tt, J ) 7.4 and 1.5 Hz), 7.26 (1H,
m), 7.22 (2H, dt, J ) 6.9 and 1.5 Hz), 6.54 (1H, d, J ) 8.5 Hz),
4.76 (1H, m), 3.10 (1H, dd, J ) 13.3 and 6.9 Hz), 2.98 (1H, dd,
J ) 13.3 and 7.9 Hz), 2.79 (1H, ddd, J ) 17.3, 5.4, and 1.0
Hz), 2.74 (1H, ddd, J ) 17.7, 5.9, and 2.0 Hz); 13C NMR +
DEPT 135 (100 MHz, CDCl3) δ 201.4 (C), 167.0 (C), 137.3 (C),
134.2 (C), 131,6 (C), 129.2 (CH), 128.8 (CH), 128.6 (CH), 127.0
(CH), 126.9 (CH), 46.9 (CH), 46.7 (CH2), 40.1 (CH). HRMS
(3R)-3-Ben za m id o-3-(4-flu or op h en yl)p r op a n a l (R)-1b.
(i) Obtained from the crude dihydrooxazine 2b (Procedure B,
1.34 mmol) as an amorphous white solid (289 mg, 80%); IR
(Nujol) 3330 (NH), 2729 (CHO), 1716 (HCdO), 1637 (HNCd
O), 1525 (CdC Ar), 1228, 853 cm-1; 1H NMR (400 MHz, CDCl3)
δ 9.83 (1H, s), 7.77 (2H, d, J ) 6.9 Hz), 7.52 (1H, t, J ) 7.4
Hz), 7.44 (2H, t, J ) 7.4 Hz), 7.36 (2H, dd, J ) 8.4 Hz and
J H-F ) 4.9 Hz), 7.05 (2H, ∼t, J ∼ 8.6 Hz), 6.93 (1H, d, J ) 7.9
Hz), 5.69 (1H, m, J ) 5.9 Hz), 3.24 (1H, ddd, J ) 17.2, 5.9,
and 2.0 Hz), 3.08 (1H, ddd, J ) 17.2, 6.2, and 1.2 Hz); 13C NMR
+ DEPT 135 (100 MHz, CDCl3) δ 200.4 (C), 166.8 (C), 162.1
calcd for C17H17NO2 [M]•+ 267.12593, found 267.1259. [R]20
D
-12 (c 0.9, CHCl3). 1H NMR (400 MHz, CDCl3, (+)-Eu(tfc)3
25% mol, CHO): δ(S) 9.94 (98%), δ(R) 9.88 (2%); ee 96%.
(3R)-3-Ben za m id o-4,4-d im eth ylbu ta n a l (R)-1f. Obtained
from the crude dihydrooxazine 2f (Procedure C, 0.43 mmol)
as a colorless amorphous solid (42 mg, 43%); 1H NMR (400
MHz, CDCl3) δ 9.78 (1H, dd, J ) 1.0 and 4.4 Hz), 7.73 (2H, d,
J ) 7.4 Hz), 7.51 (1H, t, J ) 7.1 Hz), 7.44 (2H, t, J ) 7.6 Hz),
6.20 (1H, d, J ∼ 9.4 Hz), 4.61 (1H, ∼dt, J ∼ 10.3 and 3.9 Hz),
2.79 (1H, ddd, J ) 14.5, 3.9, and 1.0 Hz), 2.46 (1H, ddd, J )
1
4
(C, d, J C-F ) 246.4 Hz), 136.3 (C, d, J C-F ) 3.1 Hz), 133.8
(C), 131,8 (C), 128.6 (CH), 128.2 (CH, d, 3J C-F ) 8.4 Hz), 127.0
2
(CH), 115.7 (CH, d, J C-F ) 21.4 Hz), 48.7 (CH2), 48.4 (CH).
HRMS calcd for C16H14FNO2 [M]•+ 271.1008, found 271.1008.
[R]20 -2.85 (c 0.43, CHCl3). H NMR (400 MHz, CDCl3, (+)-
1
D
Eu(tfc)3 30% mol, CHO) δ(R) 9.97 (98.5%), δ(S) 9.93 (1.5%); ee
97%. (ii) Obtained from the crude dihydrooxazine 2b (Proce-
dure C, 1 mmol) as an amorphous white solid (169 mg, 62%).1H
NMR (400 MHz, CDCl3, (+)-Eu(tfc)3 55% mol, CHO) δ(R) 10.15
(98%), δ(S) 10.06 (2%); ee 96%.
14.5, 10.3, and 4.4 Hz), 1.03 (9H, s). HRMS calcd for C14H19
-
NO2 [M]•+ 233.14158, found 233.1412, [M - C2H3O]+ 190, [M
- t-Bu]+ 176. [R]20 +13 (c 1.3, CHCl3). 1H NMR (400 MHz,
D
CDCl3, (+)-Eu(tfc)3 100% mol, CHO): δ(R) 10.81 (98.5%), δ(S)
10.72 (1.5%); ee 97%.
(3S)-3-Ben za m id o-4-(m eth oxyp h en yl)p r op a n a l (S)-1c.
Obtained from the pure dihydrooxazine 2c-exo-R (Procedure
A, 0.2 mmol) as a yellowish oil (23 mg, 40%); IR (Nujol) 3321
(NH), 2834 (CHO), 1722 (HCdO), 1637 (HNCdO), 1178, 1032,
833 cm-1; 1H NMR (400 MHz, CDCl3) δ 9.79 (1H, s), 7.75 (2H,
d, J ) 6.9 Hz), 7.49 (1H, t, J ) 7.4 Hz), 7.40 (2H, t, J ) 7.4
Hz), 7.29 (2H, d, J ) 8.6 Hz), 6.95 (1H, br s), 6.87 (2H, d, J )
8.6 Hz), 5.65 (1H, ∼q, J ∼ 7 Hz), 3.78 (3H, s), 3.18 (1H, dd, J
) 16.7 and 6.9 Hz), 3.01 (1H, dd, J ) 16.7 and 5.9 Hz); 13C
NMR + DEPT 135 (100 MHz, CDCl3) δ 200.4 (C), 166.4 (C),
158.9 (C), 133.7 (C), 132.1 (C), 131,4 (CH), 128.3 (CH), 127.5
(CH), 126.7 (CH), 114.0 (CH), 55,0 (CH3), 48.6 (CH2), 48.3 (CH).
Deter m in a tion of th e Absolu te Con figu r a tion of Al-
d eh yd e 1e: P r ep a r a tion of th e Am id o Aceta l 9. A mixture
of 1e (0.250 g, 0.95 mmol), (+)-1,4-bis-O-(4-chlorobenzyl)-D-
threitol (mp 72-74 °C, [R]20 +6.0 ( 0.5 (c 3, CHCl3); 0.372 g,
D
1 mmol), and p-TsOH‚H2O (9.5 mg, 0.05 mmol) in C6H6 (13
mL) was heated at reflux for 13h in a Dean-Stark apparatus.
After being cooled, the mixture was diluted with CH2Cl2 (15
mL), then neutralized by adding powdered K2CO3 (0.8 g).
Filtration and concentration in vacuo, followed by purification
by column chromatography (silica gel 20/1; cyclohexane/AcOEt
from 9:1 to 8:2), afforded acetal 9 as a thick oil (0.42 g, 70%)
(Rf 0.32; cyclohexane/AcOEt 9:1), that slowly crystallized as
colorless needles, mp 92-94 °C (Et2O); 1H NMR (400 MHz,
CDCl3) 7.73 (2H, d, J ) 7.7 Hz), 7.46 (1H, t, J ) 7.5 Hz), 7.34
(2H, t, J ) 7.7 Hz), 7.22 (13H, m), 6.75 (1H, d, J ) 7.2 Hz),
5,22 (1H, d, J ) 4.5 Hz), 4.52 (1H, m), 4.50 (2H, s), 4.46 (2H,
2d, J ) 12.1 Hz), 4.05 and 3.96 (2H, 2dt, J ) 6.7 and 4.9 Hz),
3.56 (4H, m), 3.14 (1H, dd, J ) 13.4 and 4.8 Hz), 2.86 (1H, dd,
J ) 13.4 and 8.1 Hz), 1.98 (1H, dt, J ) 14.7 and 4.4 Hz), 1.88
(1H, ddd, J ) 14.7, 8.8, and 4.6 Hz); 13C NMR + DEPT 135
(100 MHz, CDCl3) δ 166.7 (C), 137.8 (C), 136.2 (C), 136.1 (C),
134.8 (C), 133.5 (C), 131.3 (C), 129.6 (CH), 129.0 (CH), 128.8
(CH), 128.6 (CH), 128.5 (CH), 128.4 (CH), 126,8 (CH), 126,5
(CH), 103.3 (CH), 77.9 (CH), 77.0 (CH), 72.7 (CH2), 70.3 (CH2),
48.1 (CH), 40.5 (CH2), 36.2 (CH2).
HRMS calcd for C17H17NO3 [M]•+ 283.1208, found 283.1213.
1
[R]20 -17 (c 0.965, CHCl3). H NMR (400 MHz, CDCl3, (+)-
D
Eu(tfc)3 25% mol, CHO) δ(S) 9.92; ee >98%.
(3R)-3-Ben za m id o-3-(4-m eth oxyp h en yl)p r op a n a l (R)-
1c. Obtained from the crude dihydrooxazine 2c (Procedure C,
1
0.735 mmol) as a yellowish oil (181 mg, 62%); H NMR (400
MHz, CDCl3) according to (S)-1c. [R]20 +15 (c 1.335, CHCl3).
D
1H NMR (400 MHz, CDCl3, (+)-Eu(tfc)3 65% mol, CHO): δ(R)
10.17 (96.5%), δ(S) 10.10 (3.5%); ee 93%.
(3R)-3-Ben za m id o-3-(4-n itr op h en yl)p r op a n a l (R)-1d .
Obtained from the crude dihydrooxazine 2d (Procedure C, 0.61
mmol) as a yellowish amorphous solid (82 mg, 45%); IR (Nujol)
3271 (NH), 2729 (CHO), 1720 (HCdO), 1637 (HNCdO), 1520
(NO2), 1346 (NO2) cm-1 1H NMR (400 MHz, CDCl3) δ 9.83
;
Ack n ow led gm en t. We thank Dr. Vincent Maison-
neuve (Laboratoire des Fluorures UMR CNRS 6010,
Universite´ du Maine, F-72085 Le Mans Cedex 9, France)
for performing the X-ray crystallography of compound
9.
(1H, s), 8.21 (2H, d, J ) 8.9 Hz), 7.79 (2H, d, J ) 6.9 Hz), 7.56
(2H, d, J ) 8.9 Hz), 7.53 (1H, m), 7.46 (1H, t, J ) 7.6 Hz),
7.17 (1H, m), 5.76 (1H, ∼q, J ∼ 6.1 Hz), 3.32 (1H, ddd, J )
17.7, 5.7, and 1.2 Hz), 3.18 (1H, dd, J ) 17.7 and 5.7 Hz).
HRMS calcd for C16H14N2O4 [M]•+ 283.12084, found 283.1213.
1
[R]20 -6 (c 1.13, CHCl3). H NMR (400 MHz, CDCl3, (+)-Eu-
D
(tfc)3 30% mol, CHO): δ(R) 10.06 (95%), δ(S) 9.97 (5%); ee 90%.
(3S)-3-Ben za m id o-4-p h en ylbu ta n a l (S)-1e.28 Obtained
from the crude dihydrooxazine 2e (Procedure C, 1.57 mmol)
Su p p or tin g In for m a tion Ava ila ble: Crystal structure
data for 9 and copies of 1H NMR spectra of all new compounds
lacking elemental analysis, including ee determination by H
NMR with use of chiral shift reagent in the case of 1b. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
1
as a colorless amorphous solid (318 mg, 74%); H NMR (400
(27) Coppola, G. M. Synthesis 1984, 1021.
(28) Greenlee, W. J .; Allibone, P. L.; Perlow, D. S.; Patchett, A. A.;
Ulm, E. H.; Vassil, T. C. J . Med. Chem. 1985, 28, 434.
J O0203138
4344 J . Org. Chem., Vol. 68, No. 11, 2003