Stereoselective intramolecular cycloadditions of homochiral N-alkenoyl aryl azides

Article abstract of DOI:10.1016/S0957-4166(01)00190-2

Starting from the commercially available (S)-1-phenylethylamine and L-alanine benzylester, we synthesised the homochiral N-alkenoyl aryl azides 2a-2d. The intramolecular cycloaddition of unsubstituted 2a and 2b gave enantiopure 3,3a-dihydro-1,2,3-triazolo

Full text of DOI:10.1016/S0957-4166(01)00190-2

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 1201–1206
Stereoselective intramolecular cycloadditions of homochiral
N-alkenoyl aryl azides
Gianluigi Broggini,b Luisa Garanti,a,* Giorgio Moltenia and Tullio Pilatic
aDipartimento di Chimica Organica e Industriale, Universita` di Milano, via Golgi 19, 20133 Milan, Italy
bDipartimento di Scienze Chimiche, Fisiche e Matematiche, Universita` dell’Insubria, via Lucini 3, 22100 Como, Italy
cConsiglio Nazionale delle Ricerche, Centro per lo Studio delle Relazioni tra Struttura e Reattivita` Chimica, via Golgi 19,
20133 Milan, Italy
Received 23 March 2001; accepted 25 April 2001
Abstract—Starting from the commercially available (S)-1-phenylethylamine and
L-alanine benzylester, we synthesised the
homochiral N-alkenoyl aryl azides 2a–2d. The intramolecular cycloaddition of unsubstituted 2a and 2b gave enantiopure
3,3a-dihydro-1,2,3-triazolo[1,5-a][1,4]benzodiazepine-4(6H)-ones 3a, 3b, 4a and 4b, while phenyl-substituted 2c and 2d gave
enantiopure 1,1a-dihydro-2H-azirino[2,1-c][1,4]benzodiazepine-4(6H)-ones 5c, 5d, 6c and 6d. © 2001 Elsevier Science Ltd. All
rights reserved.
1. Introduction
construction of valuable synthetic targets. Among
them, pyrrolizidine and indolizidine alkaloids,2 amaryl-
lidaceae alkaloids3 and biotin4 are of particular note.
However, all of these synthetic approaches rely on the
spontaneous degradation of the first-formed 4,5-dihy-
dro-1,2,3-triazole ring, which has very low stability
under the cycloaddition conditions. We present here the
first stereoselective synthesis of enantiopure 3,3a-dihy-
dro-1,2,3-triazolo[1,5-a][1,4]benzodiazepine-4(6H)-ones
3a, 3b, 4a and 4b and 1,1a-dihydro-2H-azirino[2,1-
Stereoselective 1,3-dipolar cycloaddition methodology1
represents one of the most fruitful fields within contem-
porary organic chemistry. In recent years, the cycload-
dition approach has allowed many successful syntheses
of enantiomerically pure molecules, which would be
difficult to obtain by different routes. There are many
examples in which the stereoselective intramolecular
cycloaddition of chiral azides plays a key role in the
Figure 1.
* Corresponding author.
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00190-2

1202
G. Broggini et al. / Tetrahedron: Asymmetry 12 (2001) 1201–1206
by us. Diazotisation of the latter followed by treatment
with sodium azide gave the desired N-alkenoyl aryl
azides 2 with good yields (Fig. 1). Unsubstituted 2a and
2b were not fully characterised because of their lability
even at room temperature. In fact, the latter partially
underwent spontaneous intramolecular cycloaddition to
the diastereoisomeric 1,2,3-triazolo[1,5-a][1,4]benzo-
diazepinones 3a, 4a and 3b, 4b, respectively, during the
reaction workup. The complete conversion of 2a and 2b
to the cycloadducts was achieved by stirring a 0.02 M
ethereal solution of the crude azides at room tempera-
ture (Scheme 1). By contrast, phenyl-substituted azides
2c and 2d, being far more stable than 2a and 2b, were
obtained as crystalline solids and could be fully charac-
terised. Compounds 2c and 2d, which were proven to
be stable under the above reaction conditions, were
further reacted in refluxing toluene in the presence of
1% triethylamine giving the diastereoisomeric aziridino-
[2,1-c][1,4]benzodiazepinones 5c–6c and 5d–6d, respec-
tively (Scheme 2). Products and their isolated yields, as
well as reaction times and chromatography eluents, are
collected in Table 1. The overall yields of the
intramolecular cycloaddition reactions were consis-
tently good. Furthermore, simple silica gel chromato-
graphy of the crude products allowed the clean separa-
Scheme 1.
c][1,4]benzodiazepine-4(6H)-ones 5c, 5d, 6c and 6d
from the intramolecular cycloaddition of homochiral
N-alkenoyl aryl azides 2. The inexpensive, commer-
cially available (S)-1-phenylethylamine and
benzylester were used as the starting chiral units.
L-alanine
2. Results and discussion
N-[(S)-1-Phenylethyl]-2-aminobenzylamines 1a and 1b5
and the benzyl N-[(2-aminophenyl) methylene]-L-alan-
inates 1c and 1d6 were synthesised as recently reported
Scheme 2.
Table 1. Intramolecular cycloadditions of N-alkenoyl aryl azides 2
Compd
Time (h)
Products and yieldsc
Eluentd
3
4
5
6
2a
2b
2c
2d
0.75a
0.5a
7.5b
6.5b
61
57
–
28
18
–
–
–
58
47
–
–
33
31
Et2O–LP (2:1)
Et2O
Et2O
–
–
AcOEt–hexane (1:1)
a In dry Et2O, rt.
b In refluxing toluene.
c Isolation yields.
d LP=light petroleum, bp 40–60°C.

G. Broggini et al. / Tetrahedron: Asymmetry 12 (2001) 1201–1206
1203
of the minor diastereoisomer 6c, thus suggesting the
(R)-configuration at C(3), while the lack of the shield
for the major diastereoisomer 5c may be indicative of
(S)-configuration. Furthermore, the coupling constants
found for the protons of the aziridine ring of 5c and 6c
(3.2–3.5 Hz) agree with those reported for similar nitro-
gen bridgehead aziridines8 and indicate a trans-arrange-
ment of these protons, thus accounting for the depicted
(1aS,2R)-configuration of 5c and the (1aR,2R)-configu-
ration of 6c. The X-ray diffraction analysis of minor
diastereoisomer 4a (Fig. 2) fully supported the above
assignments, and consequently the (S)-absolute
configuration can be assigned to the major cycloadduct
3a. Unfortunately, the 1H NMR spectra of cycload-
ducts 3b and 4b were very similar to each other and
lacked any shielding effect due to the
L-alanine ben-
zylester pendant. The (S)-absolute configuration to the
triazoline C(5) of the major diastereoisomer 3b was
elucidated by X-ray diffraction analysis, but the abso-
lute configuration of the aziridine C(3) of 5d and 6d
could only be tentatively assigned. In this respect, it can
be assumed that the transition states leading to the
major triazolines 3b and 3d are configurationally simi-
lar, thus accounting for the (S)-configuration of the
newly formed stereocentres in both 3b and 5d.
As can be seen in Table 1, intramolecular cycloaddi-
tions of azides 2 show stereoselectivities ranging from
fair to good; the ratios 3/4 and 5/6 encompass the range
from 60:40 (Table 1, 2d) to 73:27 (Table 1, 2b). This
result is noteworthy considering the large distance
between the pre-existing and the newly-formed
stereocentres.
The stability of 3a, 3b, 4a and 4b is somewhat surpris-
ing in the light of the known thermal lability of D2-
1,2,3-triazolines,9 whose decomposition leads to
aziridines or imine derivatives. However, intramolecu-
lar cycloaddition of 2a and 2b occurs at room tempera-
ture, thus allowing the isolation of pure 3a, 3b, 4a and
4b from the diastereomeric mixtures in good overall
yields. Similar tricyclic structures 3c and 3d, which are
involved in the thermal decomposition of phenyl-substi-
tuted azides 2c and 2d, undergo loss of nitrogen to give
Figure 2. ORTEP plots of 3b (structure A, see Section 3.5)
and 4a (structure B, see Section 3.5) with the crystallographic
numbering scheme. Ellipsoids at 50% probability level. H
atoms not to scale.
tion of the diastereoisomeric products 3, 4, 5 and 6 in
pure form. Basic reaction media and eluents were
needed because of the known lability of compounds
such as 5 or 6 towards acidic species.7 Structural assign-
ment of the above products relied upon analytical and
spectral data. In the case of minor diastereoisomer 4a,
1
the H NMR spectra show resonance of the C(7) and
C(8) aromatic hydrogens at l=6.03 and 6.51, respec-
tively. These upfield shifted values can only be justified
by invoking a shielding effect from the phenyl ring of
the (S)-1-phenylethyl pendant. This shielding effect,
which is absent in the case of major diastereoisomer 3a,
can only be operative if the triazoline C-(5) has (R)-
absolute configuration, in close analogy with the struc-
turally related pyrazolo[1,5-a][1,4]benzodiazepine-4-
(6H)-ones, as recently proposed by us.5 On the basis of
the same considerations one can attribute the absolute
configuration of the aziridinic C-(3) of 5c and 6c. Here
again, a strong shielding effect causes upfielded reso-
nances (C(7)H, l=5.60 and C(8)H, l=6.40) in the case
Scheme 3.

1204
G. Broggini et al. / Tetrahedron: Asymmetry 12 (2001) 1201–1206
Table 2. Thermal behaviour of 1,2,3-triazolo[1,5-a][1,4]benzodiazepine-4(6H)-ones 3 and 4
Compd
Time (h)a
Products and yieldsb
Eluentc
5a
5b
6a
6b
7a
7b
3a
3b
4a
4b
1
1
6
8
43
–
–
–
48
–
–
–
30
–
–
–
–
53
44
44
–
–
–
42
43
AcOEt–CH2Cl2 (1:1)
AcOEt–LP (1:1)
Et2O
–
–
–
Et2O
a In refluxing toluene.
b Isolation yields.
c LP=light petroleum, bp 40–60°C.
the aziridines 5c, 5d, 6c and 6d. The lack of imine
derivatives when starting from azides 2c and 2d may be
attributed to the stabilizing effect of the phenyl group
on the adjacent electron-deficient carbon atom. This
picture is substantiated by thermal decomposition of
the triazolinic cycloadducts 3a, 3b, 4a and 4b, which
was carried out in refluxing toluene and gave a mixture
of enantiopure aziridines 5a and 6a and imines 7a and
7b (Table 2 and Scheme 3). It should be added that the
absence of geminal coupling in the case of 5a and 6a
parallels those observed for similar nitrogen bridgehead
aziridines.7,10
vigorous stirring and ice-cooling. After 1 h, the organic
layer was separated, washed with 5% aqueous sodium
hydrogen carbonate (30 mL), then with water (50 mL),
and dried over sodium sulfate. Evaporation of the
solvent gave crude 2a–2d as brown oily residues.
Compounds 2a and 2b were used without further purifi-
cation. 2a (2.94 g, 96%) IR (neat): 2120, 1660 (cm−1); 2b
(3.57 g, 98%) IR (neat): 2130, 1650 (cm−1).
In the remaining cases, the residue was chro-
matographed on a silica gel column with hexane–
AcOEt (2:1) as eluent giving 2c and 2d.
In conclusion, the course of the intramolecular cycload-
ditions of azides 2a–2d is strongly influenced by both
the reaction conditions and R1. Moreover, direct
intramolecular azide cycloaddition allowed the first suc-
cessful synthesis of enantiopure compounds 3–7 of
potential pharmacological interest.
2c (3.29 g, 86%), mp 95–97°C, as a pale yellow solid
(from di-iso-propyl ether); [h]2D5=−19.6 (CHCl3, c=
1
0.20); IR (Nujol): 2125, 1640 (cm−1); H NMR l: 1.50
(3H, d, J=8.1), 4.30 (1H, d, J=19.0), 4.47 (1H, d,
J=19.0), 6.27 (1H, q, J=8.1), 6.55 (1H, d, J=15.3),
6.90–7.50 (14H, m), 7.80 (1H, d, J=15.3); MS m/z 382
(M+) (44%). Anal. calcd for C24H22N4O: C, 75.37; H,
5.80; N, 14.65. Found: C, 75.46; H, 5.86; N, 14.58%.
3. Experimental
Melting points were determined with a Bu¨chi apparatus
in open tubes and are uncorrected. IR spectra were
recorded with a Perkin–Elmer 1725 X spectrophotome-
ter. Mass spectra were determined with a VG-70EQ
2d (3.92 g, 89%), mp 75–76°C, as a pale yellow solid
(from di-iso-propyl ether); [h]2D5=+7.1 (CHCl3, c=
1
1.82); IR (Nujol): 2125, 1740, 1640 (cm−1); H NMR l:
1.48 (3H, d, J=7.8), 4.27 (1H, d, J=19.0), 4.55 (1H, d,
J=19.0), 4.73 (1H, d, J=17.4), 4.90 (1H, d, J=17.4),
5.16 (1H, q, J=7.8), 6.60 (1H, d, J=15.7), 6.90–7.50
(14H, m), 7.75 (1H, d, J=15.7); MS m/z 440 (M+)
(31%). Anal. calcd for C26H24N4O2: C, 70.89; H, 5.49;
N, 12.72. Found: C, 70.95; H, 5.46; N, 12.80%.
1
apparatus. H NMR spectra were taken with a Bruker
AC 300 or a Bruker AMX 300 instrument (in CDCl3
solutions at room temperature unless otherwise stated).
Chemical shifts are given as ppm from tetra-
methylsilane and J values are given in Hz. Optical
rotations, [h]2D5, were recorded on a Perkin–Elmer
Model 241 polarimeter at the sodium D line.
3.2. 3,3a-Dihydro-1,2,3-triazolo[1,5-a][1,4]benzo-
diazepine-4(6H)-ones 3 and 4
Because of the large overlapping of the signals, 1H
NMR spectra of compounds 7a and 7b were taken in
CDCl3 solutions at 60°C.
A solution of 2a and 2b (9.0 mmol) in dry diethyl ether
(450 mL) was stirred at room temperature for the time
indicated in Table 1. Evaporation of the solvent gave a
residue that was chromatographed on a silica gel
column with the eluent given in Table 1. The major
diastereoisomers 3 were eluted first, followed by minor
diastereoisomers 4.
3.1. General procedure for the synthesis of N-alkenoyl
aryl azides 2
A solution of 1 (10.0 mmol) in aqueous hydrochloric
acid (6 M, 8.0 mL) and acetic acid (4.0 mL) was treated
with sodium nitrite (1.04 g, 15.0 mmol) under stirring
and cooling at 0°C. After 30 min, the mixture was
treated with cold diethyl ether (25 mL) and sodium
azide (3.22 g, 0.05 mol) was added portionwise under
3a (1.68 g, 61%), mp 126–128°C, as a colourless solid
(from di-iso-propyl ether); [h]2D5=+73.6 (CHCl3, c=
0.24); IR (Nujol): 1660 (cm−1); 1H NMR l: 1.42 (3H, d,
J=7.9), 3.78 (1H, d, J=16.7), 4.55 (1H, dd, J=18.1,
14.0), 4.94 (1H, d, J=16.7), 5.15 (1H, dd, J=14.0,

G. Broggini et al. / Tetrahedron: Asymmetry 12 (2001) 1201–1206
1205
10.6), 5.40 (1H, dd, J=18.1, 10.6), 6.05 (1H, q, J=7.9),
6.90–7.90 (9H, m); MS m/z 306 (M+) (75%). Anal. calcd
for C18H18N4O: C, 70.57; H, 5.92; N, 18.29. Found: C,
70.65; H, 5.99; N, 18.19%.
5d (1.74 g, 47%) as a colourless oil; [h]2D5=−47.7
(CHCl3, c=0.40); IR (Nujol): 1740, 1650 (cm−1); 1H
NMR l: 1.39 (3H, d, J=7.6), 3.21 (1H, d, J=3.5), 3.35
(1H, d, J=3.5), 3.87 (1H, d, J=15.5), 5.12 (1H, d,
J=15.5), 5.17 (1H, d, J=12.8), 5.22 (1H, d, J=12.8),
5.43 (1H, q, J=7.6), 6.80–7.50 (14H, m); MS m/z 412
(M+) (63%).
4a (0.77 g, 28%), mp 106–107°C, as a colourless solid
(from di-iso-propyl ether); [h]2D5=−330.9 (CHCl3, c=
0.28); IR (Nujol): 1660 (cm−1); 1H NMR l: 1.55 (3H, d,
J=7.8), 3.75 (1H, d, J=17.0), 4.55 (1H, dd, J=18.5,
12.5), 5.06 (1H, d, J=17.0), 5.15 (1H, dd, J=12.5, 9.2),
5.38 (1H, dd, J=18.5, 9.2), 5.95 (1H, q, J=7.8), 6.03
(1H, dd, J=7.2, 1.6), 6.51 (1H, dt, J=7.3, 1.8), 7.20–
7.72 (7H, m); MS m/z 306 (M+) (59%). Anal. calcd for
C18H18N4O: C, 70.57; H, 5.92; N, 18.29. Found: C,
70.60; H, 5.94; N, 18.34%.
6d (1.15 g, 31%) as a pale yellow oil; [h]2D5=+83.7
(CHCl3, c=0.27); IR (neat): 1740, 1650 (cm−1); 1H
NMR l: 1.49 (3H, d, J=7.9), 3.18 (1H, d, J=3.2), 3.31
(1H, d, J=3.2), 3.92 (1H, d, J=15.5), 4.60 (1H, d,
J=12.6), 5.02 (1H, d, J=12.6), 5.12 (1H, d, J=15.5),
5.26 (1H, q, J=7.9), 6.60–7.50 (14H, m); MS m/z 412
(M+) (34%).
3b (1.87 g, 57%), mp 94–96°C, as a colourless solid
(from di-iso-propyl ether); [h]2D5=+106.9 (CHCl3, c=
3.4. Thermal behaviour of 1,2,3-triazoles 3a, 3b, 4a
and 4b
1
0.28); IR (Nujol): 1740, 1670 (cm−1); H NMR l: 1.42
(3H, d, J=8.0), 4.00 (1H, d, J=17.0), 4.52 (1H, dd,
J=18.0, 12.0), 5.16 (2H, s), 5.20–5.30 (4H, m), 6.90–
7.80 (9H, m); MS m/z 364 (M+) (36%). Anal. calcd for
C20H20N4O3: C, 65.92; H, 5.53; N, 15.37. Found: C,
65.88; H, 5.60; N, 15.44%.
A solution of 3a,b or 4a,b (5.0 mmol) and triethylamine
(2.20 g, 21.8 mmol) in dry toluene (250 mL) was
refluxed for the time indicated in Table 2. Evaporation
of the solvent gave a residue that was chromatographed
on a silica gel column with the eluent given in Table 2.
Imines 7a or 7b were eluted first, followed by aziridines
5a and 5b or 6a and 6b, respectively.
4b (0.59 g, 18%), mp 88–90°C, as a colourless solid
(from di-iso-propyl ether); [h]2D5=−122.0 (CHCl3, c=
1
0.21); IR (Nujol): 1740, 1660 (cm−1); H NMR l: 1.45
7a as a colourless oil; [h]2D5=+120.0 (CHCl3, c=0.20);
IR (neat): 1630 (cm−1); 1H NMR l: 1.45 (3H, d,
J=7.7), 2.60 (3H, s), 3.70 (2H, s), 5.85 (1H, q, J=7.7),
6.80–7.40 (9H, m); MS m/z 278 (M+) (74%).
(3H, d, J=8.0), 4.00 (1H, d, J=17.0), 4.49 (1H, dd,
J=18.2, 12.0), 4.53 (1H, d, J=11.8), 4.93 (1H, dd,
J=11.8), 5.12–5.30 (4H, m), 6.90–7.80 (9H, m); MS
m/z 364 (M+) (44%). Anal. calcd for C20H20N4O3: C,
65.92; H, 5.53; N, 15.37. Found: C, 65.99; H, 5.50; N,
15.31%.
5a (0.60 g, 43%) as a colourless oil; [h]2D5=+96.2
1
(CHCl3, c=0.28); IR (neat): 1650 (cm−1); H NMR l:
1.40 (3H, d, J=7.1), 2.13 (1H, d, J=3.7), 2.70 (1H, d,
J=5.9), 3.30 (1H, dd, J=5.9, 3.7), 3.60 (1H, d, J=
14.7), 4.75 (1H, d, J=14.7), 6.00 (1H, q, J=7.1),
7.10–7.40 (9H, m); MS m/z 278 (M+) (29%).
3.3. 1,1a-Dihydro-2-phenyl-2H-azirino[2,1-c][1,4]benzo-
diazepine-4(6H)-ones 5c, 5d, 6c and 6d
A solution of 2c, 2d (9.0 mmol) and triethylamine (3.90
g, 38.6 mmol) in dry toluene (450 mL) was refluxed for
the time indicated in Table 1. Evaporation of the
solvent gave a residue that was chromatographed on a
silica gel column with the eluent given in Table 1.
Major diastereoisomers 5c and 5d were eluted first,
followed by minor diastereoisomers 6c and 6d.
6a (0.42 g, 30%) as a colourless oil; [h]2D5=−256.4
1
(CHCl3, c=0.36); IR (neat): 1650 (cm−1); H NMR l:
1.59 (3H, d, J=7.6), 2.10 (1H, d, J=4.0), 2.70 (1H, d,
J=5.8), 3.27 (1H, dd, J=5.8, 4.0), 3.60 (1H, d, J=
15.0), 4.85 (1H, d, J=15.0), 5.60 (1H, dd, J=7.6, 1.1),
6.03 (1H, q, J=7.6), 6.40 (1H, dt, J=7.5, 1.2), 6.90–
7.30 (7H, m); MS m/z 278 (M+) (36%).
5c (1.85 g, 58%), mp 68–70°C, as a colourless solid
(from di-iso-propyl ether); [h]2D5=−79.2 (CHCl3, c=
0.26); IR (Nujol): 1650 (cm−1); 1H NMR l: 1.45 (3H, d,
J=7.9), 3.32 (1H, d, J=3.1), 3.39 (1H, d, J=3.1), 3.64
(1H, d, J=14.7), 4.78 (1H, d, J=14.7), 6.02 (1H, q,
J=7.9), 6.90–7.40 (14H, m); MS m/z 354 (M+) (49%).
Anal. calcd for C24H22N2O: C, 81.33; H, 6.26; N, 7.90.
Found: C, 81.40; H, 6.32; N, 7.97%.
7b as a colourless oil; [h]2D5=+204.8 (CHCl3, c=0.27);
1
IR (neat): 1740, 1650 (cm−1); H NMR l: 1.42 (3H, d,
J=7.9), 2.55 (3H, s), 4.12 (2H, s), 5.18 (2H, s), 5.29
(1H, q, J=7.9), 7.10–7.40 (9H, m); MS m/z 336 (M+)
(89%).
5b (0.81 g, 48%) as a colourless oil; [h]2D5=+67.6
(CHCl3, c=0.19); IR (neat): 1740, 1650 (cm−1); 1H
NMR l: 1.38 (3H, d, J=7.6), 2.05 (1H, d, J=3.7), 2.68
(1H, d, J=5.5), 3.28 (1H, dd, J=5.5, 3.7), 3.83 (1H, d,
J=15.8), 5.10–5.23 (3H, m), 5.37 (1H, q, J=7.6), 6.80–
7.40 (9H, m); MS m/z 336 (M+) (48%).
6c (1.05 g, 33%) as a pale yellow oil; [h]2D5=+46.7
1
(CHCl3, c=0.15); IR (neat): 1640 (cm−1); H NMR l:
1.61 (3H, d, J=7.9), 3.31 (1H, d, J=3.3), 3.37 (1H, d,
J=3.3), 3.64 (1H, d, J=14.9), 4.89 (1H, d, J=14.9),
5.67 (1H, dd, J=7.4, 1.8), 6.07 (1H, q, J=7.9), 6.42
(1H, dt, J=7.8, 1.5), 6.90–7.60 (12H, m); MS m/z 354
(M+) (59%).
6b (0.89 g, 53%) as a colourless oil; [h]2D5=−83.8
(CHCl3, c=0.28); IR (neat): 1740, 1650 (cm−1); 1H
NMR l: 1.45 (3H, d, J=7.5), 2.03 (1H, d, J=3.7), 2.68

1206
G. Broggini et al. / Tetrahedron: Asymmetry 12 (2001) 1201–1206
(1H, d, J=5.7), 3.24 (1H, dd, J=5.7, 3.7), 3.88 (1H, d,
J=16.0), 4.57 (1H, d, J=12.7), 4.9 (1H, d, J=12.7),
5.08 (1H, d, J=16.0), 5.23 (1H, q, J=7.5), 6.80–7.25
(9H, m); MS m/z 336 (M+) (51%).
phenylethylamine. The two independent molecules
(with the same numbering scheme but signed as A and
B, respectively) show similar geometrical parameters.
The main differences are on the torsion angles around
the C(16)ꢀC(18) bond (see Fig. 1): for example the
torsion N(7)ꢀC(16)ꢀC(18)ꢀC(23) is −127.6(3)° for
molecule A and −115.8(3)° for molecule B.
3.5. Crystal data for compounds 3b and 4a
3b: C20H20N4O3, Fw=364.40, triclinic, space group P1,
,
,
,
a=8.5433(8) A, b=11.1139(11) A, c=11.1251(12) A,
h=72.371(7)°, i=69.721(6)°, k=89.073(7)°, V=
Acknowledgements
3
−3
,
939.6(2) A , Z=2, Dx=1.288 Mg m , v(Mo Ka)=
0.089 mm−1; crystal dimensions 0.60×0.36×0.16 mm3,
,
u=0.71073 A (Mo Ka radiation, graphite monochro-
We are grateful to CNR and MURST for financial
support.
mator, Bruker P4 diffractometer). Data collection at
291 K, ꢀ−2q scan mode, 4<2q<55°, h 0“11, k −13“
14, l −13“14; 4989 collected reflections, 4252 unique
[3122 with Io>2|(Io)], merging R=0.0179. The struc-
ture was solved by SIR-9211 and refined by SHELXL-
References
9712 by full-matrix least-squares based on Fo , with
1. (a) Gothelf, K. V.; Jørgensen, K. A. J. Chem. Soc., Chem.
Commun. 2000, 1449; (b) Gothelf, K. V.; Jørgensen, K.
A. Chem. Rev. 1998, 98, 863; (c) Cinquini, M.; Cozzi, F.
Methoden der Organische Chemie (Houben-Weyl), E 21c,
Stereoselective Synthesis; G. Thieme: Stuttgart, 1995; pp.
2953–2987; (d) Curran, D. P. In Advances in Cycloaddi-
tion; Curran, D. P., Ed.; JAI Press: London, 1988; Vol. 1,
p. 129.
2. Pearson, W. H.; Bergmeier, S. C.; Degan, S.; Lin, K.-C.;
Poon, Y.-F.; Schkeryantz, J. M.; Williams, J. P. J. Org.
Chem. 1990, 55, 5719.
3. Pearson, W. H.; Schkeryantz, J. M. Tetrahedron 1996, 52,
3107.
2
weights w=1/[|2(Fo)2+(0.0410P)2], where P=(Fo +
2
2
2Fc )/3. H atoms of benzyl, phenyl and methyl groups
were calculated (riding model). The final consistency
indexes were R=0.0573 and Rw=0.0896 (0.0378 and
0.0821, respectively, for observed reflections), goodness-
of-fit=1.011. The final map ranges between −0.10 and
−3
,
0.11 e A . The absolute configuration was determined
on the basis of the known configuration of
L
-alanine
benzylester. The two independent molecules (with the
same numbering scheme but signed as A and B, respec-
tively) have similar geometrical parameters. The main
differences are on the conformation of the seven-mem-
bered
ring:
here
the
torsion
angles
4. Pearson, W. H.; Postich, M. J. J. Org. Chem. 1994, 59,
5662.
5. Broggini, G.; Casalone, G.; Garanti, L.; Molteni, G.;
Pilati, T.; Zecchi, G. Tetrahedron: Asymmetry 1999, 10,
4447.
C(8)ꢀC(13)ꢀN(14)ꢀC(4) and N(6)ꢀC(7)ꢀC(8)ꢀC(9) mea-
sure 10.3(4) and −129.7(2)°, respectively, in molecule A
(see Fig. 1) versus −1.3(4) and −121.2(3)° in molecule B.
4a: C18H18N4O, Fw=364.40, triclinic, space group P21,
6. Broggini, G.; Garanti, L.; Molteni, G.; Pilati, T. Synth.
Commun., in press.
7. Fusco, R.; Garanti, L.; Zecchi, G. J. Org. Chem. 1975,
40, 1906.
8. Batterham, T. J. NMR Spectra of Simple Heterocycles;
Wiley-Interscience: New York, 1973; p. 138.
9. (a) Bourgois, J.; Bourgois, M.; Texier, F. Bull. Soc. Chim.
Fr. 1978, 485; (b) Schultz, A. G. In Advances in Cycload-
ditions; Curran, D. P., Ed.; JAI Press: London, 1988;
Vol. 1, p. 53; (c) Gru¨nanger, P.; Vita-Finzi, P. Isoxazoles;
Wiley: New York, 1991; p. 625.
10. Woerner, F. P.; Reimlinger, H.; Mere´nyi, R. Chem. Ber.
1971, 104, 2786.
11. Altomare, A.; Cascarano, G.; Giacovazzo, G.;
Guagliardi, A.; Burla, M. C.; Polidori, G.; Camalli, G. J.
Appl. Cryst. 1994, 27, 435.
,
,
,
a=16.0395(16) A, b=5.9632(6) A, c=17.0305(19) A,
3
,
i=100.605(7)°, V=1601.1(3) A , Z=4, Dx=1.271 Mg
m−3, v(Mo Ka)=0.082 mm−1; crystal dimensions 0.54×
3
,
0.31×0.20 mm , u=0.71073 A (Mo Ka radiation,
graphite monochromator, Bruker P4 diffractometer).
Data collection at 291 K, ꢀ−2q scan mode, 4<2q<52°,
h 0“19, k 0“7, l −21“20; 8452 collected reflections,
3474 unique [2173 with Io>2|(Io)], merging R=0.0344.
The structure was solved by SIR-9211 and refined by
SHELXL-9712 by full-matrix least-squares based on
Fo , with weights w=1/[|2(Fo)2+(0.0396P)2], where P=
2
2
2
(Fo +2Fc )/3. H atoms of phenyl and methyl groups
were calculated (riding model). The final consistency
indexes were R=0.0618 and Rw=0.0768 (0.0321 and
0.0659, respectively, for observed reflections), goodness-
of-fit=0.851. The final map ranges between −0.09 and
12. Sheldrick, G. M. SHELX97 Program for the Refinement
of Crystal Structures; University of Go¨ttingen, Go¨ttin-
gen, 1997.
−3
,
0.11 e A . The absolute configuration was determined
on the basis of known configuration of (S)-1-
.