of the chiral centres in the minor diastereoisomer of aziridine
0 (X-ray) and the known absolute configuration of the chiral
centre in the quinazolinone 2-position [derived from (S)-lactic
acid], the stereostructures of the major diastereoisomers of 12,
131.6 (d), 131.4 (d), 131.1 (d), 130.7 (d), 130.3 (d), 130.0 (d),
2
129.5 (d), 129.1 (d), 128.9 (d), 128.8 (d), 124.3 (s), 69.8 (d), 68.6
(d), 55.9 (d), 52.3 (d), 51.3 (d), 46.2 (d), 33.1 (d), 28.6 (q),
24.3 (q), 21.4 (s), 20.8 (s), 1.5 (q), 0.4 (q) and 0.0 (q); minor
diastereoisomer (observable signals) 56.1 (d), 52.3 (d) and 50.8
(d).
1
8 and 20 can be deduced and are as illustrated in Schemes 4
and 5.
Further work to establish the generality of the N–(Q*) azirid-
ine to azirine to aziridine interconversion is in progress together
with an examination of the origin of the unexpectedly high
diastereoselectivity in the aziridination of vinylsilanes 5 and 17
Aziridination of â-triethylsilylstyrene 17 with Q*NHOAc 11
A solution of Q*NHOAc 11 in dichloromethane (10 cm ) was
prepared from 3-aminoquinazolinone 10 (1.00 g, 3.1 mmol)
3
1
3
with Q*NHOAc 11.
and LTA (1.53 g, 3.1 mmol) as described above and reacted
1
6
with a solution of β-triethylsilylstyrene 17 (0.82 g, 3.8 mmol)
containing HMDS (0.76 g, 4.7 mmol) in dichloromethane
Experimental
3
(2 cm ). After the work-up described above, chromatography
1
Unless otherwise indicated, H NMR spectra were run at 25 ЊC
over silica and elution with light petroleum–ethyl acetate–
triethylamine (89:9:2) gave (2S,3S)-1-{2-[(1S)-1-tert-
butyldimethylsilyloxyethyl]-4-oxo-3,4-dihydroquinazolin-3-yl}-2-
and 250 MHz in CDCl solution with SiMe as internal stand-
3
4
1
3
ard and C spectra at 75 MHz in the same solvent. IR Spectra
were run as solutions in dichloromethane. Optical rotations
were measured using a Perkin-Elmer 341 Polarimeter and are
phenyl-3-triethylsilylaziridine 18 (R 0.49) as an oil (0.88 g, 40%)
F
(Found: C, 67.65; H, 8.6; N, 7.75. C H N O Si requires C,
3
0
45
3
2
2
Ϫ1
2
Ϫ1
Ϫ1
recorded in units of 10 deg cm g . For other instrumen-
tation and general experimental details see ref. 14.
67.25; H, 8.45; N, 7.85%); νmax/cm 1680s and 1595m; δ (2:1
H
ratio of N-invertomers) major invertomer (observable signals)
8
.20 [d, J 7.5, 5-H(Q)], 6.80–7.60 (m, 8 H), 5.07 (q, J 7,
Improved procedure for preparation of 3-amino-2-[(1S)-1-tert-
CHCH ), 3.40 (d, J 7.9, CHPh), 2.90 (d, J 7.9, CHSi) and 1.40
3
butyldimethylsilyloxyethyl]-3,4-dihydroquinazolin-4-one 10
(d, J 7, CHCH ); minor invertomer (observable signals) 5.30 (q,
3
3
3
-Aminoquinazolinone 9 (5.67 g, 27.7 mmol), tert-butyl-
J 6, CHCH ), 3.88 (d, J 7.5, CHPh) and 2.15 (d, J 7.5, CHSi);
3
dimethylsilyl chloride (5.00 g, 33.2 mmol) and imidazole (4.70
signals from both invertomers at 1.00 (m), 0.85 (m) and 0.00
3
g, 69.1 mmol) were dissolved in DMF (11 cm ) and stirred at
(m) (total 30 H); δ (161.3 and 161.0) (s), (158.3 and 157.3) (s),
C
3
room temperature for 2 days. Water (30 cm ) was then added
146.3 (s), 138.5 (s), 132.7 (d), 134.0 (d), 129.2 (d), 128.9 (d),
128.4 (d), 128.0 (d), 127.1 (d), 126.7 (d), 126.4 (d), 122.0 (s),
(67.5 and 66.2) (d), (53.0 and 49.7) (d), 46.4 (d), 42.7 (d), 26.2
(q), 21.9 (d), 20.9 (d), (18.9 and 18.5) (s), 7.9 (q) and 2.7 (t);
δ (C D ) major diastereoisomer (3.7:1 ratio of N-invertomers)
and the aqueous layer was extracted with light petroleum
3
(
4 × 50 cm ). The combined organic extracts were washed with
3
3
brine (2 × 50 cm ), dried and reduced to ~20 cm by evapor-
ation under reduced pressure. Seeding with amino alcohol 9
and scratching the side of the flask removed a small amount of
this unchanged starting material, and evaporation of the separ-
ated light petroleum gave 3-aminoquinazolinone 10 as a colour-
H
6
6
major invertomer (assignable signals) 8.3 [d, J 7.5, 5-H(Q)],
6.70–7.60 (m, 8 H), 5.25 (q, J 6, CHCH ), 3.45 (d, J 7.9, CHPh),
3
3.10 (d, J 7.9, CHSi) and 1.40 (d, J 6, CHCH ); minor invert-
3
6
less oil (7.07 g, 80%) identical with that obtained previously.
omer (assignable signals) 5.60 (q, J 6.1, CHCH ), 4.30 (d, J 7,
3
CHPh) and 2.25 (d, J 7, CHSi); minor diastereoisomer (~1:1
ratio of invertomers) (assignable signals) 4.28 (d, J 7.5, CHPh),
3.58 (d, J 7.5, CHPh), 3.32 (d, J 7.5, CHSi) and 2.18 (d, J 7.5,
Aziridination of â-trimethylsilylstyrene 5 with Q*NHOAc 11
Dichloromethane (5 cm ) was cooled to Ϫ15 ЊC, lead tetra-
3
1
acetate (LTA) (0.76 g, 1.2 mmol) was added and the solution
stirred until the LTA dissolved. A solution of 3-aminoquin-
CHSi); the H NMR of the crude reaction mixture in C D
6
6
showed the ratio of major :minor diastereoisomers as ~13:1
from comparison of signal intensity for the aziridine ring
3
azolinone 10 (0.50 g, 1.6 mmol) in dichloromethane (2 cm ) was
ϩ
then added with stirring over 5 min and the mixture stirred at
Ϫ15 ЊC for a further 5 min. After cooling to Ϫ30 ЊC, the mix-
ture was filtered rapidly through a small column containing
Celite using a low positive pressure of nitrogen into a stirred
protons above; m/z 535 (M , 3.9%), 417 (42.6), 376 (100) and
247 (43.1).
Aziridination of â-triphenylsilylstyrene 19 with Q*NHOAc 11
1
5
3
solution of β-trimethylsilylstyrene 5 (0.33 g, 1.9 mmol) and
A solution of Q*NHOAc 11 in dichloromethane (25 cm ) was
3
3
HMDS (1.0 cm , 4.7 mmol) in dichloromethane (1 cm ) held at
Ϫ30 ЊC. The reaction mixture was allowed to warm to room
temperature over 1 h with stirring before addition of dichloro-
prepared from 3-aminoquinazolinone 10 (2.00 g, 6.3 mmol) and
LTA (3.05 g, 6.9 mmol) as described earlier (but without filtra-
1
6,17
tion through Celite) and β-triphenylsilylstyrene 19
(2.50 g,
3
3
methane (10 cm ). After washing the mixture with saturated
6.9 mmol) and HMDS (2.0 cm , 9.4 mmol) were added at
Ϫ30 ЊC. After reaction and work-up as described previously,
the crude product was chromatographed over silica eluting with
light petroleum–ethyl acetate (4:1) to give aziridine 20 (1.74 g,
41%) as a 2:1 ratio of diastereoisomers from comparison of
aqueous sodium hydrogen carbonate, the organic layer was
separated, dried and the solvent removed under reduced pres-
sure to give an oil (0.77 g).
Chromatography over silica, previously washed with light
petroleum–ethyl acetate (4:1) containing 2% triethylamine, and
elution with light petroleum–ethyl acetate (4:1) gave (2S,3S)-
1
the aziridine ring proton signals in the H NMR spectrum
(see below). Trituration with light petroleum gave the minor
diastereoisomer (2R,3R)-1-{2-[(1S)-1-tert-butyldimethylsilyl-
oxyethyl]-4-oxo-3,4-dihydroquinazolin-3-yl}-2-phenyl-3-tri-
phenylsilylaziridine 20 as a colourless solid, mp 151–152 ЊC
1
-{2-[(1S)-1-tert-butyldimethylsilyloxyethyl]-4-oxo-3,4-dihydro-
quinazolin-3-yl}-2-phenyl-3-trimethylsilylaziridine 12 (R 0.30)
F
ϩ
(
4
0.27 g, 35%) (Found: M , 493.2580. C H N O Si requires M,
93.2580); νmax/cm 1920m, 1680s and 1600m; δ (1.6:1 ratio
2
7
39
3
2
2
Ϫ1
ϩ
(from ethanol) (Found: M 679.3050. C H N O Si requires
H
42 45
3
2
2
Ϫ1
of N-invertomers); major invertomer (observable signals) 8.15
d, J 7.5, 5-H(Q)], 5.03 (q, J 6.0, CHOSi), 3.35 (d, J 7.5, CHPh)
M, 679.3050); νmax/cm 1680s and 1600s; δH major diastereo-
isomer (1.5:1 ratio of N-invertomers) major invertomer 8.15 [d,
J 7.5, 5-H(Q)], 7.80 (m, 4 H), 7.40 (m, 15 H), 7.00 (m, 3 H), 5.3
[
and 2.9 (d, J 7.5, CHSiMe ); minor invertomer (observable
3
signals) 8.05 [d, J 7.9, 5-H(Q)], 5.25 (q, J 6.0, CHOSi), 3.70
(q, J 6.2, CHCH ), 4.20 (d, J ~7.5, CHPh), 2.90 (d, J 7.5, CHS-
3
(d, J 7.2, CHPh) and 2.10 (d, J 7.2, CHSi); signals for both
iPh ), 1.05 (d, J 6.2, CHCH ), 0.70 [s, SiC(CH ) ] and Ϫ0.05 [s,
3
3
3 3
invertomers at 6.9–7.6 (m, 8 H), 1.4 (m), 0.7–1.85 (m) and
Ϫ0.1 to Ϫ0.2 (m), (total 27 H); δC major diastereoisomer (2
N-invertomers) 163.4 (s), 163.3 (s), 160.3 (s), 159.3 (s), 148.8 (s),
Si(CH ) ]; minor N-invertomer (observable peaks) 5.05 (d, J
3 2
6.0, CHCH ), 4.00 (d, J 7.3, CHPh) and 3.40 (d, J 7.3, CHS-
3
iPh ); minor diastereoisomer (1:1 ratio of N-invertomers), sig-
3
1
48.6 (s), 140.7 (s), 136.3 (d), 136.0 (d), 134.8 (d), 132.0 (d),
nals for both invertomers at 8.45 [d, J 7.5, 5-H(Q)], 7.95 (m, 4
J. Chem. Soc., Perkin Trans. 1, 1997
899