Rigid backbone 1,8-anthracene-linked bis-oxazolines (AnBOXes): Design, synthesis, application and characteristics in catalytic asymmetric aziridination

Article abstract of DOI:10.1016/j.tetasy.2005.09.025

A series of rigid backbone 1,8-anthracene-linked bis-oxazolines (AnBOXes) have been designed, synthesized, and evaluated in the catalytic asymmetric aziridination with [N-(p-toluenesulfonyl)imino]phenyliodinane (PhINTs) as a nitrene source. The results in

Full text of DOI:10.1016/j.tetasy.2005.09.025

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 3718–3734
Rigid backbone 1,8-anthracene-linked bis-oxazolines
(AnBOXes): design, synthesis, application and characteristics
in catalytic asymmetric aziridination
Linge Ma, Peng Jiao, Qihan Zhang and Jiaxi Xu*
Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education,
College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, PR China
Received 1 September 2005; accepted 27 September 2005
Available online 11 November 2005
Abstract—A series of rigid backbone 1,8-anthracene-linked bis-oxazolines (AnBOXes) have been designed, synthesized, and evalu-
ated in the catalytic asymmetric aziridination with [N-(p-toluenesulfonyl)imino]phenyliodinane (PhI@NTs) as a nitrene source. The
results indicate that highly enantioselective aziridination of chalcones catalyzed by an AnBOX and CuOTf complex with up to
>99% ee and the opposite enantioselectivity, compared with the ligands of Evans et al., can be achieved. The enantioselectivity
is substituent dependent with respect to chalcones. Chalcones with electron-donating substituents show higher enantioselectivities
due to the stronger Lewis basicity of the oxygen of their carbonyl groups than those with electron-withdrawing substituents. The
results also indicate that the coordination between the oxygen of the carbonyl group in chalcones and the ether group in alkenes
with the copper in the catalyst is essential for high enantioselectivity, while the p–p stacking interaction between two reactants plays
an importantly additional role for high enantioselectivity in asymmetric aziridination. An excellent backbone-controlled stereo-
selectivity was observed for the AnBOX ligands in asymmetric aziridination, as this will provide very important information for
designing novel ligands.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
furan,6 dibenzothiophene,7 dibenzopyrrole,8 bicyclic
compounds,9 dibenzo[a,c]cycloheptadiene,10 etc. have
been reported. The effects of the structure and chelate
size of bis-oxazoline ligands, the two oxazolines
of which were linked with a single bond, methylene,
ethylene, 1,2-phenylene, and 4,5-dioxolane, in the
asymmetric copper-catalyzed cyclopropanation and
aziridination of olefins,2,11 and in the asymmetric
Diels–Alder reaction12 were investigated previously.
The results indicated that fine tuning of the ligand struc-
ture could improve enantioselectivity. It is well known
that a well-ordered chiral environment at the catalytic
center will play an important role in controlling the
enantioselectivity in an asymmetric catalytic reaction.
The size of the chelate in the reactive metal complex
of bis-oxazolines is another important feature of the cat-
alyst since it will control the orientation of the substitu-
ents on the two oxazolines around the metal ion and the
distance of the substituents to the metal ions. This
implies that the chelate size of bis-oxazolines can tune
the chiral environment at the catalytic center and then
affect the enantioselectivity of asymmetric catalytic reac-
tions. To keep the designed chiral environment at the
The C2-symmetric chiral bis-oxazolines have emerged as
one class of important and efficient ligands in an increas-
ing number of asymmetric transformations over the last
decade.1 The bis-oxazolines have been found to be excel-
lent chiral ligands for numerous asymmetric reactions,
including cyclopropanation of olefins, aziridination of
olefins and imines, (hetero)-Dields–Alder reactions,
Henry reactions, allylic substitution and oxidation,
transfer hydrogenation, hydrosilylation and cyanosilyla-
tion of aldehydes and ketones, addition of dialkylzinc to
aldehydes and ketones, addition of alkyllithiums to imi-
nes, addition of silylketene acetal to aldehyde, free rad-
ical initiated addition of allyltributylstannane, etc.1
Over the last decade, numerous chiral bis-oxazolines
with different backbones, such as methylene,2 dimethyl-
methylene,3 ethylene,2 benzene,2,4 pyridine,5 dibenzo-
*
Corresponding author. Fax: +86 10 62751708; e-mail: jxxu@pku.
edu.cn
0957-4166/$ - see front matter ꢀ 2005Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.09.025

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3719
O
O
O
O
N
N
N
X
O
O
N
N
Ph
Ph
O
N
N
O
N
N
R
R
BOX
1
PyBOX
R
R
R
R
X = O DBFBOX
X = S DBTBOX
X = NH DBPBOX
AnBOX
2
Scheme 1. Different rigid backbone bis-oxazoline ligands.
catalytic center, a series of rigid backbone-linked bis-
oxazolines were designed and synthesized. Compared
with some conformationally rigid backbone-containing
bis-oxazoline ligands, in which the two oxazoline rings
were linked by benzene,2,4 pyridine,5 dibenzofuran,6
dibenzothiophene,7 dibenzopyrrole,8 etc. our design for
these ligands focuses on the backbone possessing some
different structural features from previously reported
bis-oxazoline ligands.2–11 Our ligand (AnBOX), the
two oxazoline rings of which are attached on the 1,8-
positions of a rigid anthracene ring, possesses only
bidentate, instead of tridentate in the previously reported
rigid bis-oxazoline ligands with pyridine,5 dibenzo-
furan,6 dibenzothiophene,7 or dibenzopyrrole8 as the
backbones, and their two substituents on the oxazoline
rings are closer than previously reported ones.5–8 When
they coordinate with a metal during the catalysis, their
stereogenic centers (substituents) are close to the metal
ion. Thus, they should have a different chiral environ-
ment near the reaction center (Scheme 1).
to prepare the dihydroxy diamide directly by aminolysis
of amino alcohols.19 However, we discarded this method
due to the very low yield. In order to synthesize our
ligands (AnBOXes) 2, we first had to prepare 1,8-
anthracenedicarboxylic acid firstly. When we attempted
to prepare 1,8-anthracenedicarboxylic diacid via sapon-
ification of its dimethyl ester, we found that the corre-
sponding disodium salt is insoluble in a variety of
solvents. Finally, we chose 1,8-dichloroanthraquinone
as starting material to prepare 1,8-anthracenedicarboxy-
lic diacid according to the method reported by Rogers
and Averill.20
Commercially available 1,8-dichloroanthraquinone was
converted to 1,8-anthraquinonedicarboxylic diacid via
cuprous cyanide substitution and hydrolysis in high
yields.20 1,8-Anthraquinonedicarboxylic diacid was
reduced to 1,8-anthracenedicarboxylic acid by zinc,
followed by the activation with SOCl2 in PhCl to give
1,8-anthracenedicarbonyl dichloride,21 which was trea-
ted with L-amino alcohols to afford the corresponding
dihydroxy diamides 3 in high yields. The diamides 3
were activated with SOCl2 in THF to yield dichlorodi-
amides, which were treated with K2CO3 in a mixture of
CH3CN–H2O as a co-solvent to produce 1,8-bisoxazo-
linylanthracenes 2 (AnBOXes) in satisfactory yields.22
In most cases, the intermediates 3 were used without fur-
ther purification due to their poor solubility. The final
ligands were conveniently purified by silica gel chroma-
tography (Scheme 2).
There have been three major approaches for catalytic
asymmetric aziridinations developed during the past
decade,13 including nitrene transfer to olefins3,14 and
carbene transfer to imines15 under the catalysis of chiral
ligands and transition metals,16 and carbene transfer to
imines via chiral sulfonium ylides.17 Among them, the
nitrene transfer to olefins is a powerful strategy for the
synthesis of enantiomerically enriched aziridines with a
variety of functional groups.3,14 In transition metal-
chiral ligand catalyzed asymmetric aziridinations, the
C2-symmetric chiral bis-oxazolines have emerged as
one class of the most efficient ligands. We have recently
reported preliminary results on the asymmetric aziridin-
ation of chalcones catalyzed by our bis-oxazoline ligand
(AnBOX)–copper complex.18 Herein, we present full
details on the preparation of anthracene-linked bis-
oxazolines, along with their stereoselectivity and
characteristics in asymmetric catalytic aziridinations of
olefins.
Because 1,8-anthraquinonedicarboxylic diacid is a key
intermediate in the preparation of 1,8-anthracenedicarb-
oxylic diacid, a tridentate bis-oxazoline ligand, 1,8-bis-
oxazolinylanthraquinone 4, with anthraquinone as a
backbone was also synthesized from 1,8-anthracenedi-
carboxylic diacid and enantiomerically pure L-phenyl-
alaninol using a similar method (Scheme 2).
2.2. Optimizing reaction conditions for the asymmetric
aziridination of chalcone
2. Results and discussion
First, the copper-catalyzed asymmetric aziridination of
chalcone with [N-(p-toluenesulfonyl)imino]phenyliodin-
ane (PhI@NTs) as a nitrene source in the presence of
AnBOXes was investigated in detail in order to deter-
mine the efficiency of the new ligand system and to opti-
mize reaction conditions. Different temperatures,
solvents, and copper salts were evaluated with chalcone
and ligand 2a. The results are summarized in Table 1. It
was found that excellent enantioselectivity and good
2.1. Synthesis of 1,8-bisoxazolinylanthracenes
Although several synthetic methods are available for the
preparation of bis-oxazolines, the most efficient one is
using diacids as starting materials via dihydroxy
diamides as intermediates. Dimethyl 1,8-anthracenedi-
carboxylate is commercially available and could be used

3720
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
O
O
O
1. SOCl2, THF, reflux
2. K2CO3, CH3CN-H2O
O
CO2Me
CO2Me
reflux
68%
HN
O
O
NH2
Bn
O
N
N
O
OH HO
Bn
Bn Bn
5
4
1. SOCl2, 99%
2.
L
-BnCHNH2CH2OH, 50%
X
HO
O
O
OH
Cl
O
Cl
O
O
CO2H
CO2H
Zn, 20% NH4OH
49%
1. CuCN, 93%
2. 70% H2SO4, 79%
O
1. SOCl2, 73%
1. SOCl2, THF, reflux
2. K2CO3, CH3CN-H2O
reflux
2. L-RCHNH2CH2OH
THF, 0oC to RT
O
N
N
O
HN
O
O
NH
OH HO
R
R
R
R
3a: R = CHMe2, 83%
3b: R = CHMeEt, 94%
3c: R = CMe3
3d: R = Bn, 98%
3e: R = Ph, 98%
2a: R = CHMe2, 80%
2b: R = CHMeEt, 54%
2c: R = CMe3, overall yield 56%
2d: R = Bn, 52%
2e: R = Ph, 72%
Scheme 2. Synthesis of 1,8-bisoxazolinylanthracenes 2 and 1,8-bisoxazolinylanthraquinone 4.
Table 1. Optimizing the reaction conditions by the aziridination of chalcone 6a
Ts
O
O
N
5 mol% Cu salt
6 mol% Ligand
PhI=NTs
+
solvent
6a
7a or ent-7a
Entry
Ligand
Solvent
Copper
Temp (ꢁC)
Yielda (%)
eeb (%)
Configurationc
1
2
3
4
5
6
7
8
2a
2a
2a
2a
2a
2a
2a
2a
2a
2b
2c
2d
2e
4
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH3CN
C6H6
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CuOTf
CuOTf
CuOTf
CuOTf
CuOTf
CuOTf
Cu(acac)2
Cu(CH3CN)4PF6
Cu(OTf)2
CuOTf
ꢀ17 warm to 23
46
64
80
80
80
18
33
81
52
79
2510
76
73
32
38
55
86
96
79
88
58
28
62
65
26
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
7a (2S,3R)
ent-7a (2R,3S)
0 warm to 23
24
30
24
24
24
24
24
24
24
24
24
24
24
9
10
11
12
13
14
15
CuOTf
CuOTf
CuOTf
CuOTf
61
32
23
86
1
CuOTf
a Isolated yield after flash silica gel chromatographic separation.
b Determined by HPLC analysis using chiralcel OD column.
c The absolute configuration was determined by comparing the determined specific rotations with that reported.24
yields were obtained in CH2Cl2 at 24 ꢁC (room temper-
ature) (Table 1, entry 3). Although Evans et al. reported
the highest enantioselectivity was obtained in benzene,3
we found that dichloromethane is the best solvent for
our asymmetric reaction. We also conducted the asym-
metric reaction at different temperatures and found that
almost no desired reaction occurred at ꢀ78 to 0 ꢁC pos-
sibly due to poor reactivity of chalcone with nitrene
TsN:. The reaction proceeded efficiently at room tem-
perature (around 24 ꢁC). The highest enantioselectivity
was achieved by portion wise addition of PhI@NTs.
For copper salts, CuOTf is soluble in many organic

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3721
solvents and is often employed in asymmetric aziridin-
ation.3 However, CuOTf is sensitive to air and moisture,
and relatively expensive. It was reported that bis-oxazo-
lines catalysts derived from Cu(I) and Cu(II) resulted in
the same level of asymmetric induction.23 We examined
So far, it can be seen from Table 1 that the AnBOX 2a–
CuOTf is the most efficient catalyst for our model reac-
tion, and the reaction performed at 24 ꢁC in CH2Cl2
gives the best results in terms of the enantioselectivity
and reactivity.
other
copper
sources,
such
as
Cu(acac)2,
Cu(CH3CN)4PF6, and Cu(OTf)2 (Table 1 , entries 7–9)
in the hope of finding a relatively inexpensive copper
salt. The results indicated that CuOTf is indeed the best
one and Cu(CH3CN)4PF6 ranked second with a compa-
rable yield to that of CuOTf but lower enantio-
selectivity.
In comparison with the asymmetric aziridination cata-
lyzed by bis-oxazoline (BOX) 1 of Evans et al. (Table
1, entry 15),3 it is very interesting to note that our ligand
AnBOXes 2 show a different stereoselectivity, whereby
chalcone is aziridinated in an opposite configuration
(2S,3R) despite AnBOXes being derived from the L-amino
alcohols as used in the BOX ligands of Evans et al.3
After optimizing reaction conditions, the bidentate
AnBOXes 2b–e (Table 1, entries 10–13) and tridentate
ligand 4 (Table 1, entry 14) were examined under the
same conditions. They showed lower enantioselectivities
than AnBOX 2a. Particularly, ligand 2c, with the more
sterically bulky group t-Bu on AnBOX (Table 1, entry
11), led to a further decrease in reactivity (25% yield)
and enantioselectivity (10% ee). It was suggested that
sterically hindered ligand might not provide suitable chi-
ral environment for our aziridination as reported by
Andersson et al. because it does not leave sufficient
space for the copper to simultaneously coordinate with
both nitrogen atoms of the oxazoline rings.2
2.3. Asymmetric aziridination of chalcones under the
optimized conditions
Aziridination of chalcones with PhI@NTs as the nitrene
precursor was conducted with the AnBOX 2a–CuOTf as
the catalyst under the optimal conditions. The results
are summarized in Table 2. Obviously, the ee values
are dependent on the substituents on both phenyl
groups in the chalcones. The chalcones with electron-
donating substituents, which do not contain hetero-
atoms with coordinated unshared electron pairs, for
example, O, S, show higher enantioselectivities (87% to
Table 2. Aziridination of chalcones 6 catalyzed by ligand 2a under the optimized conditions
Ts
N
O
O
5 mol% CuOTf
6 mol% Ligand 2a
R1
R2
R2
R1
PhI=NTs
+
CH2Cl2
6
7
Entry
Product
R1
R2
Yielda (%)
eeb (%)
1
2
3
4
5
6
7a
7b
7c
7d
7e
7f
H
H
H
H
H
H
H
80
86
3587
72
70
58
96
98
p-Me
p-Ph
p-F
p-Cl
p-Br
62
76
52
7
8
9
7g
7h
7i
p-MeO
o-MeO
p-MeS
p-Me
H
H
H
H
H
H
H
H
61
21
NR
59
92
72
74
64
61
74
68
66
51
91
76
8571
69
37
27
—
>99
>99
54
58
50
ND
62
43
39
68
10
11
12
13
14
15
16
17
18
19
20
21
22
23
7j
p-Me
p-Me
p-F
p-Cl
p-Br
p-NO2
p-MeO
p-Cl
p-Me
p-Cl
H
7k
7l
7m
7n
7o
7p
7q
7r
7s
7t
H
p-Cl
p-Cl
p-Me
o-Cl
m-Cl
m-F
p-CF3
79
84
7u
7v
7w
H
H
H
67
a Isolated yield after flash silica gel chromatographic separation.
b Ee value was determined by HPLC analysis using chiralpak AS column, chiralcel OD, OD-H columns. For 7a, the absolute configuration was
determined by comparing the determined specific rotation with that reported.24 For 7b–w except for 7i, the absolute configurations were tentatively
assigned according to the reaction mechanism and their relative retention times on chiral columns.

3722
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
>99% ee values) (Table 2, entries 2, 3, 10, and 11) than
those with electron-withdrawing substituents (Table 2,
entries 4–6, 12–14, and 17–23) in the asymmetric aziridi-
nation. To the best of our knowledge, the AnBOX 2a–
CuOTf is the best catalyst with the highest enantioselec-
tivity in the asymmetric aziridination of chalcones. It
was noticeable that chalcones with p-MeO, an elec-
tron-donating substituent (Table 2, entries 7, 8, and
16), showed decreased enantioselectivities (37%, 27%,
and 62% ee, respectively). There are two possible rea-
sons that cause the decreased enantioselectivity. The first
is that the oxygen in the MeO group reacts with
TsN@IPh to form an O–N ylide, which aziridinates
the C@C double bond to give rise to racemic aziridin-
ation product similar to halo substituted styrenes in
asymmetric cyclopropanation, in which a halo-carbon
ylide is formed and then reacts with the C@C double
bond to yield the racemic cyclopropanation product.25
The second is that the coordination of the oxygen in
MeO to the copper ion in the catalyst. This is a compet-
itive coordination with that between the copper and the
carbonyl group of the chalcones. To determine the rea-
son, 2-methoxychalcone was azirinated under same
reaction conditions and 27% ee value was obtained
(Table 2, entry 8). This indicates that no O–N ylide
formed in the reaction. If it was formed, complete race-
mic aziridination product would be obtained due to an
intramolecular aziridination via the O–N ylide. To ver-
ify the coordination, 4-methylthiochalcone with a strong
coordinating sulfur atom was prepared and tested. No
expected aziridine product was detected in the reaction
mixture and the starting materials were recovered quan-
titatively (Table 2, entry 9). It is obvious that both MeO
and MeS substituents, including coordinating hetero-
atoms oxygen and sulfur, are capable of coordinating
with the copper ion. Therefore, coordination seems to
be responsible for the decreased enantioselectivity and
reactivity.
donating groups afforded higher enantioselectivities
than those bearing electron-withdrawing groups. From
Hammett plots of Log([2S,3R]/[2R,3S]) versus Ham-
mett constants rp,26 a straight line for 4-substituted
chalcones (Table 2, entries 1–6) was obtained with
R2 = 0.88, slope = ꢀ3.65( Fig. 1). The same tendency
with more obvious influence, R2 = 0.94, slope = ꢀ5.41
(Fig. 2), was also observed for 40-substituted chalcones
(Table 2, entries 1 and 11–14). 40-Substituent showed
more obvious influence than 4-substituent because 40-
substituent is closer to the carbonyl group than 4-
substituent.
3
2.5
y = -3.65x + 1.47
S
R2 = 0.88
2
R,
R
1.5
1
S,
0.5
0
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Hammet constant (Sigma)
Figure 1. Hammett plot of the asymmetric aziridination of 4-substi-
tuted chalcones (Log([2S,3R]/[2R,3S]) of the aziridination products
versus Hammett constant rp, Me, ꢀ0.14; H, 0; Ph, 0.05; F, 0.15; Cl,
0.24; Br, 0.26).
3
y = -5.41x + 1.73
R2 = 0.94
2.5
S
2
R,
The oxygen of the MeO-group in the MeO-substituted
chalcones competed for coordination with the copper
in the catalyst, which decreased the amount of the car-
bonyl group-coordinated complex, so that their enantio-
selectivities decreased because the MeO-coordinated
complex could not undergo an asymmetric catalytic
aziridination due to unsuitable positioning of the C@C
bond of the chalcones in the ether-coordinated complex.
The MeS-substituted chalcone did not react with
PhI@NTs and was recovered quantitatively since the
coordination of the sulfur atom with the copper is so
strong that the catalytic amount of copper was caught
completely by the MeS-groups and removing the copper
from the reaction system. Thus, the nitrene cannot be
generated due to the absence of free non-coordinated
copper, and the aziridination of the MeS-substituted
chalcone did not occur.
R
1.5
1
S,
0.5
0
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Hammet constant (Sigma)
Figure 2. Hammett plot of the asymmetric aziridination of 40-
substituted chalcones (Log([2S,3R]/[2R,3S]) of the aziridination prod-
ucts versus Hammett constant rp, Me, ꢀ0.14; H, 0; F, 0.15; Cl, 0.24;
Br, 0.26).
The electronic effect-dependent enantioselectivity was
rationalized by the electron-donating groups in chal-
cones increasing the Lewis basicity of the oxygen of
the carbonyl group in the chalcones with electron-
donating groups so that they increase their coordination
with the copper. The stronger coordination will prefer-
entially allow the reaction to proceed via the asymmetric
catalytic path catalyzed by the chiral catalyst. Thus,
it produces higher enantioselectivity. A similar pheno-
menon was also observed in the asymmetric borane
2.4. Influence of the electronic effect on the enantioselec-
tivity in the asymmetric aziridination of chalcones
An obvious influence of the electronic effect of sub-
strates on the enantioselectivity was observed in our
asymmetric aziridination. The substrates with electron-

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
Table 3. Scope and limitation of AnBOX ligand 2a in the asymmetric aziridination
3723
Entry
Substrate
Product
Yield (%)a
ee (%)
44b
Configuratione
(2S,3R)24
Ts
N
O
O
1
61
Ph
Ph
OPh
O
Ph
O
O
8
2
3
4
NR
NR
NR
Ph
NMe2
H
Ph
Ph
Ph
N
Ts
N
5
92
7c
(R)14a
Ph
Ph
Ph
Ph
9
Ts
N
13d
(S,S)28
Ph
Ph
6
7
71
63
10
Ts
N
5d
(1R,2S)14a
11
a Isolated yields after the column chromatography; NR, no reaction occurred.
b Determined by 1H-NMR chiral shift reagent.
c Determined by HPLC analysis using a chiralcel OJ column.
d Determined by HPLC analysis using a chiralcel OD-H column.
e The absolute configurations were assigned according to the sign of the specific rotation as reported in Refs. 24,14a,28 and was further verified via
detosylation by the treatment with sodium naphthalene29 to afford (S,S)-2-benzyl-3-phenylaziridine according to the reported sign of the specific
rotation.30
reduction of ketones catalyzed by chiral boroxazolidine.27
The substituent electronic effect-dependent enantioselec-
tivity also indicated that the coordination step is an
enantioselectivity-determining step in the catalytic cycle.
of a carbonyl group in chalcones and cinnamate, which
led to the increased enantioselectivities due to the pres-
ence of the coordination between the oxygen in oxy-
gen-containing substrates and the copper in the
catalyst. The coordination was also previously proposed
via calculation.31
2.5. Scope and limitation of AnBOX ligand 2a in the
asymmetric aziridination
To verify the coordination, 1,3-diphenyl-1-propene 12
and cinnamyl phenyl ether 13, carbonyl group-free
structural analogues of chalcone and phenyl cinnamate
were designed, synthesized, and asymmetrically aziridi-
nated under the same reaction conditions (Scheme 3).
The results indicated that the reaction of 1,3-diphenyl-
1-propene 12 afforded aziridine product 14 with only
6% ee and 41% yield, along with allylic insertion
After successful asymmetric aziridination of chalcones,
we attempted to apply CuOTf–AnBOX 2a complex to
other substrates under the optimized conditions (Table
3). The reaction of phenyl trans-cinnamate (Table 3,
entry 1), a similar structure to chalcone, afforded the cor-
responding aziridine with 44% ee and 61% yield. It was
noteworthy that the aziridination of phenyl trans-cinna-
mate was also in (2S,3R) configuration, opposite to the
one obtained by Evans et al.3b However, no aziridin-
ation occurred for N,N-dimethyl trans-cinnamamide,
trans-cinnamaldehyde, and trans-N-cinnamylideneani-
line (trans-PhCH@CHCH@NPh), even though they
have similar structural features with chalcones. We also
tested unfunctionalized olefins, such as styrene, trans-
stilbene, and 1,2-dihyronaphthalene (Table 3, entries
5–7). For all of them, low enantioselectivities and mod-
erate to good yields were obtained.
product,
(E,R)-N-(1,3-diphenyl-2-propenylidene)-(4-
methyl)benzenesulfonamide 15 in 5% ee and 19%
yield.32 For cinnamyl phenyl ether 13, the reaction
proceeded in a very similar way. The corresponding
aziridine 16 with an obviously higher enantioselectiv-
ity (40% ee) was obtained in 24% yield. Meanwhile,
a
byproduct
(E)-N-(3-phenyl-2-propenylidene)-(4-
methyl)benzenesulfonamide 1733 was also obtained in
8% yield. It was assumed that the imine was formed
via an allylic insertion intermediate 18, which is unstable
and readily underwent a b-elimination to furnish 17.
Actually, phenol was also detected in the reaction mix-
ture, which confirmed our assumption. Obviously, the
enantioselectivities seriously decreased due to the
absence of the carbonyl group in these two substrates.
Herein, our results provide experimental evidence to
2.6. The coordination effect of oxyatoms in substrates
with copper in the catalyst
It should be noted that the structural difference of chal-
cones and cinnamate from simple olefins is the presence

3724
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
Ts
N
NHTs
Ph
2a, CuOTf
+
+
Ph
PhI=NTs
Ph
Ph
Ph
24 oC, CH2Cl2
Ph
12
15
14
yield 19%
5% ee
yield 41%
6% ee
Ts
N
NHTs
OPh
2a, CuOTf
+
Ph
OPh +
PhI=NTs
OPh
Ph
Ph
24 oC, CH2Cl2
Ph
13
18
16
yield 24%
40% ee
-PhOH
NTs
17
yield 8%
Scheme 3. Asymmetric aziridination of 1,3-diphenylpropene and cinnamyl phenyl ether, carbonyl group-free structural analogues of chalcone and
phenyl cinnamate.
support the coordination of the carbonyl group of chal-
cones with the copper in the catalyst in the copper-cata-
lyzed aziridination. This also suggests that the oxygen of
the ether group seems to play a coordination role with
the copper because olefin 13 shows higher enantioselec-
tivity than olefin 12 due to the existence of an ether
group in olefin 13.
cones and 1,3-diphenylpropene 12. Thus, it is very
important to know the configuration of aziridine 16.
In order to identify the configuration of aziridine 16 via
chemical correlation, tert-butyldimethylsilyl cinnamyl
ether 19 was prepared from cinnamyl alcohol with
tert-butyldimethylsilyl chloride.34,35 It was asymmetri-
cally aziridinated with TsN@IPh under AnBOX
2a–CuOTf complex catalysis to afford 2-tert-butyl-
The difference in enantioselectivity between chalcone
and phenyl cinnamate was rationalized to oxygen coor-
dination of substrates with the copper in the catalyst.
Because phenyl cinnamate could furnish two different
coordinations due to the existence of the carbonyl and
ether oxygens, which could result in two different orien-
tations of C@C bond of phenyl cinnamate in the transi-
tion state and thus opposite configurational aziridines
could be generated. Another possible reason is that
because the oxygen of the carbonyl group in phenyl cin-
namate is attached with a more electron-withdrawing
group PhO, instead of Ph group in chalcone, it shows
weaker Lewis bacisity than that in chalcone so that phe-
nyl cinnamate shows lower enantioselectivity than chal-
cone in the asymmetric aziridination. On the basis of the
configuration of aziridine 16 and the influence of the
electronic effect on the enantioselectivity, we consider
that the influence of the electronic effect is the major
reason for the decreasing enantioselectivity because
cinnamyl phenyl ether 13 shows the same enantiofacial
selectivity with chalcones and 1,3-diphenylpropene 12.
If it adopts a different coordinating orientation, it would
show an opposite enantiofacial selectivity between chal-
dimethylsiloxylmethyl-3-phenyl-1-tosyl-aziridine
20
with 27% ee, which was assigned the (2S,3R)-configura-
tion with the negative specific rotation according to the
reported sign of specific rotation.33 Aziridine 20 was
detrialkylsilylated with TABF to yield (2S,3R)-2-hydro-
xymethyl-3-phenyl-1-tosyl-aziridine 21, which was
tosylated with tosyl chloride to afford (2S,3R)-3-phe-
nyl-1-tosyl-aziridin-2-ylmethyl tosylate 22. Tosylate 22
was then substituted with sodium phenolate36 to give
rise to an aziridine with the configuration of 2S,3R
and a negative specific rotation in low yield due to the
existence of the aziridine ring-opening reaction. Both
the sign of the specific rotation and HPLC retention
time on a chiral column indicated that the obtained
aziridine has the same configuration with aziridine 16.
Thus, the configuration of aziridine 16 is 2S,3R. Mean-
while, in the substitution reaction, another aziridine 23
and a toluenesulfonamide 24 were also obtained in
yields of 30% and 34%, respectively (Scheme 4). Their
proposed formation processes are shown in Scheme 5.
Tosylate 22 could undergo an SN2 substitution with
sodium phenolate to afford aziridine 16, which could
PhI=NTs
2a, CuOTf
Ts
N
Ts
TBAF
N
Ph
OTBDMS
OTBDMS
OH
24 oC, CH2Cl2
27%
65%
Ph
Ph
19
20
21
OPh
OPh
Ts
N
Ts
PhOH, NaH
TsCl
63%
N
+
+ Ph
OPh
NHTs
24 34%
Ph
OTs
OPh
NTs
23 30%
Ph
Ph
22
16 24%
Scheme 4. Determination of the absolute configuration of asymmetric aziridination product of cinnamyl phenyl ether.

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3725
Ts
N
OPh
OPh
H2O
-OH
SN2
Ph
OPh
NHTs
24
Ph
OPh
OPh
Ts
N
NaOPh
Ph
NTs
-
NaOPh
16
OTs
25
Ph
22
OPh
N-
SN2'
NaOPh
NaOPh
OPh
Ph
OTs
Ph
Ts
NTs
26
23
Scheme 5. Substitution reaction of (2S,3R)-3-phenyl-1-tosyl-aziridin-2-ylmethyl tosylate 22 with sodium phenolate.
Table 4. Aziridination of a,b-unsaturated ketones 27–30 catalyzed by
ligand 2a under the optimized conditions
be further attack at the benzylic position of the aziridine
ring by sodium phenolate to yield intermediate 25, fol-
lowed by work-up to afford toluenesulfonamide 24. Tos-
ylate 22 could also undergo an SN20 displacement via the
attack at the benzylic position of the aziridine ring by
sodium phenolate to yield an intermediate 26, followed
by an intramolecular SN2 substitution to produce aziri-
dine 23. Aziridine 23 could be further attacked at the
less sterically bulky carbon atom of the aziridine ring
by sodium phenolate to yield the intermediate 25, fol-
lowed by work-up to afford toluenesulfonamide 24.
Both benzylic and less sterically bulky carbon atoms in
an N-tosylated aziridine are active positions, which are
preferentially attacked by a nucleophile as we summa-
rized previously.37
Ts
O
R1
O
2a, CuOTf
R1
R2
N
+
PhI=NTs
R3
R2
R3
24 oC, CH2Cl2
27-30
31-34
Entry
Product
R1
R2
R3
Yielda (%)
eeb (%)
1
2
3
4
31
32
33
34
H
H
Me
H
Ph
Me
Ph
Me
Ph
650
42
20
Tracec
5
Me
Me
t-Bu
9
6
NDd
a Isolated yield after flash silica gel chromatographic separation.
b Ee value was determined by HPLC analysis using chiralcel OD and
OD-H columns. For 31, the absolute configuration was determined
by comparing the determined rotation to that reported.24 For 32 and
33, the absolute configurations were tentatively assumed according to
the reaction mechanism.
2.7. p–p Stacking interaction between aryl groups of
nitrene and substrates
c 93% starting materials were recovered.
d Not determined.
It is well known that the effects and benefits of the p–p
stacking interaction between reactants or reactants and
catalysts can improve the stereoselectivity in asymmetric
synthesis, especially in the asymmetric induction
achieved with chiral auxiliaries and asymmetric catalysis
catalyzed by chiral catalysts.38 The p–p stacking interac-
tion can lock one conformation of a reactant in a given
transition state assembly, which can preferentially yield
one of the product enantiomers.
Obviously, displacing a phenyl group with methyl group
in different positions of chalcone diminished the enantio-
selectivities in comparison with chalcone (96% ee). The
results indicate that the two phenyl groups in chalcone
show unequal effects on the enantioselectivity. The phe-
nyl group attached to the C@C double bond demon-
strates more efficient p–p stacking interaction with the
aryl group of the nitrene than that of the phenyl group
attached to the carbonyl group with the anthracene ring
in the ligand.
The p–p stacking interaction between the aryl groups of
substrates and the tosyl group of a nitrene was also
observed and presumed in our asymmetric catalytic sys-
tem. In the aziridination of chalcones, an AnBOX
ligand, chalcone, and CuOTf were allowed to stir for
one hour to furnish a pre-transition state. The subse-
quent portionwise addition of the nitrene precursor
PhI@NTs made the pre-transition state to transfer to a
transition state, in which nitrene TsN: was generated
via copper catalysis in the system and coordinated with
the copper. It was assumed that the p–p stacking inter-
action between the aryl group attached to the C@C
bond of chalcone and that of nitrene in the transition
state made the transition state relatively more stable
and favored to producing (2S,3R)-aziridination prod-
ucts of chalcones.
In order to further confirm the p–p stacking interaction,
we also attempted to prepare alkanesulfonyl nitrene pre-
cursors MeSO2N@IPh and BuSO2N@IPh and hoped to
test the asymmetric aziridination of chalcone with them.
Unfortunately, we did not obtain pure MeSO2N@IPh
and BuSO2N@IPh due to their instability although
intramolecular aziridination of olefins with alkyl nitrene
was achieved successfully.39
2.8. Explanation of the different stereoselectivities of
AnBOX 2 and BOX 1 and the proposed transition states
of aziridination of chalcones
To confirm our assumption, four substrates, 4-phenyl-3-
buten-2-one 27, 1-phenyl-2-buten-1-one 28, 4-methyl-3-
penten-2-one 29, and 5,5-dimethyl-3-hexen-2-one 30
were examined. The results are summarized in Table 4.
In comparison with the asymmetric aziridination cata-
lyzed by bis-oxazoline BOX 1 of Evans et al. (Table 1,
entry 15), it is very interesting to note that our ligand
AnBOX 2a showed completely different stereoselectivity,

3726
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
O
N
Cu
R
O
O
N
O
S
N
N
O
O
O
R
R
N
Cu
S
N
O
O
R
O
The nitrene attacks from the bottom
The nitrene attacks from the
(2R,3S)-aziridine produced
(2S,3R)-aziridine produced
Figure 3. Proposed transition states in the asymmetric aziridination of chalcone catalyzed by BOX 1 and AnBOX 2.
whereby chalcones were aziridinated in an opposite
configuration (2S,3R) with 96% ee although AnBOX
2a was derived from the L-amino alcohol as used in
the BOX 1 of Evans et al.3b To further confirm the
difference, additional chalcones 6k and 6l were subjected
to the asymmetric aziridination under the catalysis of
BOX 1 to afford ent-7k and ent-7l in yields 24% and
10% with 85% and 86% ee, respectively. Both of their
HPLC analysis and the sign of their specific rotations
indicate that they were asymmetrically aziridinated in
opposite enantiofacial selectivity in both cases compared
with those under the catalysis of AnBOX 2a. The differ-
ent orientations for the different enantioselectivities
could be due to the fact that AnBOX 2a has a more
crowded environment between two oxazoline rings than
BOX 1 of Evans et al. Copper has to locate above or
below the anthrancene ring after it coordinates with
AnBOX. Actually, the two locations are the same case
because of C2-symmetry of the AnBOX ligand. The
oxazoline rings rotate a certain angles and exist in a
non-planar mode to the anthrancene ring, in which the
nitrene attacks from the top of chalcone to produce
(2S,3R)-product. However, in the BOX ligands of Evans
et al., where the copper locates between two oxazoline
rings, in which the nitrene attacks from the bottom of
chalcone to generate (2S,3R)-product. Proposed transi-
tion states for the asymmetric aziridination of chalcones
for the reactions catalyzed by BOX 1 and AnBOX 2a
are shown, respectively, in Figure 3.
ing interaction between the aryl group of the nitrene and
aryl group attached to the C@C double bond in chal-
cones are indispensable for high enantioselectivity in
the asymmetric aziridination. It is worth noting that
our ligands 2 and 4 showed different enantiofacial selec-
tivity in the asymmetric aziridination of chalcones and
cinnamate, and better enantioselectivities in the asym-
metric aziridination of chalcones, as compared with
bis-oxazoline ligands of Evans et al.3 On the basis of
our results, the transition states of the aziridination
reaction catalyzed by BOX and AnBOX–Cu complexes
were proposed. It is very interesting that a backbone-
controlled stereoselectivity was observed for the
AnBOX ligands in the asymmetric aziridination. This
provides very important information for designing
novel ligands. This also indicated that the enantiofacial
selectivity of a ligand could be changed via tuning its
backbone, which linked the coordinating group(s).
Although opposite enantiofacial selectivity could be
achieved via displacing the stereogenic center(s) of the
ligand by using enantiopure starting materials of oppo-
site configuration in its synthesis, changing enantio-
facial selectivity via the tuning backbone method is a
very useful and alternative method in the case, in enan-
tiopure starting materials of opposite configuration are
not available.
4. Experimental
4.1. General
3. Conclusion
Melting points were measured on a Yanaco MP-500
melting point apparatus and are uncorrected. IR spectra
were recorded on a Bruker Vector 22 FT-IR spectropho-
tometer. 1H NMR and 13C NMR spectra were recorded
on a Varian Mercury 200 (200 MHz), Varian Mercury
Plus 300 (300 MHz), or Brucker 400 AMX (400 MHz)
spectrometer in CDCl3 or DMSO-d6 solution with
TMS as an internal standard and chemical shifts are
reported in parts per million. Mass spectra were obtained
on a VG-ZAB-HS spectrometer or a Bruker ESQUIREꢁ
LCTM ESI ion trap spectrometer. CH analyses were
performed on an Elementar Vario EL analyzer. Optical
rotations were measured on a Perkin–Elmer Model
341LC polarimeter with a thermally jacketed 10 cm cell
(concentration c given as g/100 mL). HPLC analyses
were performed on an HP1100 HPLC equipment with
chiralcel chiral columns (4.6 · 250 mm) using a mixture
In summary, a class of new bidentate bis-oxazolines
(AnBOXes) 2 with anthracene as the backbone and a
novel tridentate bis-oxazoline ligand 4 with anthraqui-
none as the backbone have been designed, synthesized,
and applied in the asymmetric aziridination of olefins
with PhI@NTs as a nitrene precursor. The results indi-
cated that AnBOX 2a with isopropyl substituent on
the oxazoline ring is superior to the other enantiopure
ligands 2 and 4 in the asymmetric aziridination, and
the enantioselectivity is substituent dependent with
respect to the chalcones. Generally, the chalcones with
electron-donating substituents show higher enantio-
selectivities than those with electron-withdrawing substi-
tuents. In addition, the results also indicated that the
coordination of the oxyatom of the carbonyl group in
chalcones with copper in the catalyst and the p–p stack-

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3727
of hexane/isopropanol as an eluent. THF and benzene
were heated under reflux over sodium benzophenone
ketyl and distilled prior to use. Dichloromethane was
heated under reflux over calcium hydride and distilled
prior to use.
MS (EI) m/z (rel. intensity, %): 464 (M+, 1.1), 446
(M+ꢀH2O, 7.0), 428 (20), 416 (17), 389 (6.0), 371 (20),
344 (100), 330 (33), 303 (22), 287 (7.9), 272 (67), 244
(31), 230 (14), 216 (21), 202 (21), 190 (6.7), 176 (16);
Anal. Calcd for C28H36N2O4: C, 72.39; H, 7.81; N,
6.03. Found: C, 72.19; H, 8.02; N, 5.85.
4.2. Synthesis of chiral ligand AnBOXes 2 and 4
4.2.1.3. (S,S)-N,N0-Bis[1-(hydroxymethyl)-2,2-dimeth-
ylpropyl]-anthracene-1,8-dicarboxamide 3c. Yellow pow-
der was used directly in the following step without
purification due to very poor solubility.
Anthracene-1,8-dicarboxylic acid was prepared as previ-
ously reported20 and anthracene-1,8-dicarbonyl dichlo-
ride and anthraquinone-1,8-dicarbonyl dichloride were
synthesized according to the literature procedure.21
4.2.1.4. (S,S)-N,N0-Bis[1-(hydroxymethyl)-2-phenyl-
ethyl]-anthracene-1,8-dicarboxamide 3d. Yellow pow-
4.2.1. General procedure for the synthesis of (S,S)-N,N0-
bis[1-(hydroxymethyl)alkyl]-anthracene-1,8-dicarboxa-
mide 3. A 250 mL flask fitted with a magnetic stirring
bar was charged with a solution of anthracene-1,8-dicar-
bonyl dichloride (0.87 g, 2.9 mmol) in 50 mL of dry
THF. The solution was cooled in an ice bath, and a solu-
tion of L-amino alcohol (8.7 mmol) in 30 mL of dry
THF was added dropwise. The resulting mixture was
allowed to warm to room temperature and was stirred
overnight. After filtration, the solution was concentrated
to give a solid residue, which was recrystallized from
ethanol to afford yellow crystals of the product.
der yield 98%, mp 242–243 ꢁC; Rf = 0.38 (ethyl
24
acetate, silica gel plate); ½aŠ ¼ ꢀ21:2 (c 0.25, CH3OH);
D
1
IR (KBr) (cmꢀ1): 3306 (O–H), 1639 (C@O); H NMR
(200 MHz, DMSO-d6): 2.85(dd, J = 8.2, 14.0 Hz, 2H
in 2CH2Ph), 3.03 (dd, J = 6.1, 14.0 Hz, 2H in 2CH2Ph),
3.43–3.65(m, 4H, 2CH O), 4.25–4.29 (m, 2H, 2CHN),
2
4.86 (t, J = 5.6 Hz, 2H, 2OH), 7.20–7.34 (m, 10H,
ArH), 7.41–7.57 (m, 4H, ArH), 8.15 (d, J = 8.4 Hz,
2H, ArH), 8.42 (d, J = 8.2 Hz, 2H, 2NH), 8.66 (s, 1H,
ArH), 9.27 (s, 1H, ArH); 13C NMR (75.5 MHz,
DMSO-d6): 36.5, 53.1, 62.7, 123.0, 124.8, 125.1, 126.0,
126.9, 128.2, 129.3, 129.9, 131.0, 135.4, 139.4, 168.3;
MS (EI) m/z (rel. intensity, %): 532 (M+, 2.6), 502
(M+ꢀCH2O, 9.1), 484 (12), 441 (M+ꢀPhCH2, 21), 423
(21), 405(23), 382 (35), 378 (26), 364 (22), 337 (10),
307 (8.8), 290 (14), 272 (100), 248 (50), 233 (12), 216
(16), 204 (20), 190 (5.4), 176 (32), 152 (3.0); Anal. Calcd
for C34H32N2O4: C, 76.67; H, 6.06; N, 5.26. Found: C,
76.40; H, 6.02; N, 4.96.
4.2.1.1. (S,S)-N,N0-Bis[1-(hydroxymethyl)-2-methyl-
propyl]-anthracene-1,8-dicarboxamide 3a. After filtra-
tion, yellow solid crude product was obtained directly,
which was recrystallized from ethanol to afford 1.04 g
as yellow needle crystals in 83% yield, mp 252–254 ꢁC;
24
Rf = 0.47 (ethyl acetate, silica gel plate); ½aŠ ¼ ꢀ65:9
D
(c 0.41, CH3OH); IR (KBr) (cmꢀ1): 3314 (O–H), 1635
(C@O); 1H NMR (400 MHz, DMSO-d6): d 0.95(d,
J = 6.8 Hz, 6H, 2CH3), 1.01 (d, J = 6.8 Hz, 6H,
2CH3), 2.00 (q, J = 6.8 Hz, 2CH), 3.55–3.58 (m, 4H,
2CH2O), 3.93–3.97 (m, 2H, 2CHN), 4.62 (t,
J = 5.6 Hz, 2H, 2OH), 7.53–7.61 (m, 4H, 2NH and
2ArH), 8.15–8.19 (m, 2H, ArH), 8.66 (s, 1H, ArH),
9.18 (s, 1H, ArH); 13C NMR (100 MHz, DMSO-d6): d
18.3, 19.8, 28.4, 56.2, 61.5, 123.1, 124.8, 125.0, 126.8,
128.2, 129.6, 131.0, 136.0, 168.5; MS (EI) m/z (rel. inten-
sity, %): 436 (M+, 9), 418 (M+ꢀH2O, 6), 406 (38), 388
(17), 357 (4), 348 (5), 334 (33), 316 (27), 307 (23), 289
(35), 272 (35), 248 (61), 233 (13), 221 (16), 204 (36),
190 (5), 176 (33); Anal. Calcd for C26H32N2O4: C,
71.53; H, 7.39; N, 6.42. Found: C, 71.72; H, 7.28; N,
6.12.
4.2.1.5. (S,S)-N,N0-Bis[1-(hydroxymethyl)-phenylmeth-
yl]-anthracene-1,8-dicarboxamide 3e. Yellow powder,
yield 98%; mp 307–309 ꢁC; Rf = 0.59 (ethyl acetate, silica
22
gel plate); ½aŠ ¼ þ31:8 (c 0.42, DMF); IR (KBr)
D
1
(cmꢀ1): 3306 (O–H), 1639 (C@O); H NMR (200 MHz,
DMSO-d6): 3.68 (dd, J = 7.0, 11.4 Hz, 2H in 2CH2O),
3.80 (dd, J = 7.0, 11.4 Hz, 2H in 2CH2O), 4.95(t, J =
5.8 Hz, 2H, 2OH), 5.20 (ddd, J = 7.0, 7.0, 8.2 Hz, 2H,
2CHN), 7.22–7.74 (m, 14H, ArH), 8.21 (d, J = 8.6 Hz,
2H, ArH), 8.70 (s, 1H, ArH), 8.92 (d, J = 8.2 Hz, 2H,
2NH), 9.41 (s, 1H, ArH); 13C NMR (75.5 MHz,
DMSO-d6): 55.8, 64.7, 123.1, 124.9, 126.0, 126.9, 127.2,
128.2, 130.3, 131.1, 134.9, 137.2, 141.3, 168.2; MS (EI)
m/z (rel. intensity, %): 504 (M+, 2.6), 502 (9.1), 484
(12), 441 (M+ꢀPhCH2, 21), 423 (21), 405(23), 382
(35), 378 (26), 364 (22), 337 (10), 307 (8.8), 290 (14),
272 (100), 248 (50), 233 (12), 216 (16), 204 (20), 190
(5.4), 176 (32), 152 (3.0); MS (ESI) m/z: 505 (MH+);
Anal. Calcd for C32H28N2O4: C, 76.17; H, 5.59; N,
5.55. Found: C, 75.89; H, 5.61; N, 5.67.
4.2.1.2. (S,S)-N,N0-Bis[1-(hydroxymethyl)-2-methyl-
butyl]-anthracene-1,8-dicarboxamide 3b. Yellow pow-
der, yield 94%, mp 209–210 ꢁC; Rf = 0.52 (ethyl
24
acetate, silica gel plate); ½aŠ ¼ ꢀ64:2 (c 0.53, CH3OH);
D
1
IR (KBr) (cmꢀ1): 3291 (O–H), 1635(C @O); H NMR
(200 MHz, DMSO-d6): 0.82–0.98 (m, 12H, 4CH3),
1.14–1.29 (m, 2H in 2CH2CH3), 1.56 (m, 2H in
2CH2CH3), 1.76 (m, 2H, 2CHCH3), 3.58–3.61 (m, 4H,
2CH2O), 3.98 (m, 2H, 2CHN), 4.75(s, br, 2H, 2OH),
7.50–7.63 (m, 4H, ArH), 8.16 (d, J = 8.0 Hz, 2H,
ArH), 8.31 (d, J = 9.0 Hz, 2H, 2NH), 8.66 (s, 1H,
ArH), 9.20 (s, 1H, ArH); 13C NMR (75.5 MHz,
DMSO-d6): 22.2, 23.5, 24.5, 24.6, 49.4, 64.0, 123.1,
124.8, 125.2, 127.0, 128.2, 129.8, 131.1, 135.7, 168.3;
4.2.2. General procedure for the synthesis of (S,S)-1,8-
bis[4-alkyl/aryloxazolin-2-yl]anthracene (AnBOX) 2.
A
250 mL flask fitted with a magnetic stirring bar was
charged with a solution of above dihydroxy diamide 3
(2.39 mmol) in 80 mL of dry THF, 12.1 mL (166 mmol)
of SOCl2 was added and the resulting mixture was
heated to reflux for 4 h to give clear yellow solution.
After removal of solution and excess SOCl2, in vacuo,

3728
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
the residue was dissolved in 120 mL of ethyl acetate,
then washed with 48 mL of 2 mol/L K2CO3, water,
and brine, respectively. The organic layer was dried over
Na2SO4, filtered, and concentrated in vacuo to yield yel-
low solid, which was used in further reaction without
purification. The yellow solid was dissolved in 112 mL
of acetonitrile, then 24.2 g (175mmol) K 2CO3 and
9.7 mL of water were added. After refluxing for 8 h,
the resulting solution was cooled to room temperature
and concentrated, then 640 mL of ethyl acetate and
160 mL of water were added. The organic extract was
dried over Na2SO4, filtered, and concentrated to give
crude product, which was purified by column chroma-
tography (silica gel 50 g, ethyl acetate/petroleum 1:4,
v/v) to afford yellow crystals of the product.
NMR (50 MHz, CDCl3): d 26.1, 34.1, 68.4, 76.7,
124.5, 124.9, 126.2, 127.6, 129.5, 129.5, 131.6, 131.7,
164.0; MS (EI) m/z (rel. intensity, %): 428 (M+, 25),
371 (M+ꢀC4H9, 80), 344 (51), 313 (3.2), 287 (4.3), 271
(100), 244 (14), 216 (18), 202 (11), 190 (4.1), 176 (4.7).
Anal. Calcd for C28H32N2O2: C, 78.47; H, 7.53; N,
6.54. Found: C, 78.42; H, 7.83; N, 6.23.
4.2.2.4.
(S,S)-1,8-Bis(4-benzyloxazolin-2-yl)anthra-
cene 2d. Yellow crystals, yield 52%; mp 166–168 ꢁC;
Rf = 0.26 (ethyl acetate/petroleum ether 1:2, v/v, sil-
20
ica gel plate); ½aŠ ¼ þ50:7 (c 0.92, CHCl3); IR (KBr)
(cmꢀ1): 1638 (CD@N); 1H NMR (300 MHz, CDCl3):
2.91 (dd, J = 9.0, 13.8 Hz, 2H in 2CH2Ph), 3.45(dd,
J = 4.8, 13.8 Hz, 2H in 2CH2Ph), 4.23 (dd, J = 7.8,
8.1 Hz, 2H in 2CH2O), 4.43 (dd, J = 8.1, 9.3 Hz, 2H
in 2CH2O), 4.75(dddd, J = 4.8, 7.8, 9.0, 9.3 Hz, 2H,
2CHN), 7.24–7.53 (m, 12H, ArH), 8.1–8.15 (m, 4H,
ArH), 8.50 (s, 1H, ArH), 10.92 (s, 1H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 42.0, 68.6, 71.1, 124.5,
125.1, 125.5, 126.5, 127.6, 128.6, 129.3, 129.5, 131.5,
132.0, 138.0, 164.1; MS (EI) m/z (rel. intensity, %):
4.2.2.1. (S,S)-1,8-Bis[4-(1-methylethyl)oxazolin-2-yl]-
anthracene 2a. Yellow crystals, yield 80%, mp 122–
124 ꢁC; Rf = 0.35(ethyl acetate/petroleum ether 1:2, v/v,
20
silica gel plate); ½aŠ ¼ þ64:7 (c 0.97, CHCl3);
D
IR (KBr) (cmꢀ1): 1645(C @N); 1H NMR (300 MHz,
CDCl3): 1.03 (d, J = 6.9 Hz, 6H, 2CH3), 1.15(d,
J = 6.9 Hz, 6H, 2CH3), 2.00–2.06 (m, 2H, 2CH(CH3)2),
4.23–4.33 (m, 4H, 2CH2O), 4.51–4.56 (m, 2H, 2CHN),
7.48 (dd, J = 8.4, 6.9 Hz, 2H, ArH), 8.08–8.13 (m, 4H,
ArH), 8.47 (s, 1H, ArH), 10.64 (s, 1H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 18.1, 19.4, 32.8, 69.5,
73.2, 124.5, 124.9, 125.9, 127.5, 129.4, 129.5, 131.5,
131.7, 163.8; MS (EI) m/z (rel. intensity, %): 400 (M+,
22), 357 (M+ꢀC3H7, 19), 330 (100), 287 (7.9), 271
(72), 244 (24), 228 (6.0), 216 (12), 202 (12), 190 (4.1),
176 (4.3); Anal. Calcd for C26H28N2O2: C, 77.97; H,
7.05; N, 6.99. Found: C, 77.99; H, 7.26; N, 6.76.
+
496 (M+, 28), 405(M ꢀPhCH2, 65), 378 (82), 313
(3.0), 287 (20), 271 (100), 244 (22), 228 (48), 216 (21),
202 (16), 189 (4.6), 176 (4.6), 117 (8.8), 105(13), 91
(PhCHþ, 40); Anal. Calcd for C34H28N2O2: C, 82.23;
2
H, 5.68; N, 5.64. Found: C, 82.47; H, 5.67; N, 5.62.
4.2.2.5.
(S,S)-1,8-Bis(4-phenyloxazolin-2-yl)anthra-
cene 2e. Yellow crystals, yield 72%, mp 147–148 ꢁC;
Rf = 0.43 (ethyl acetate/petroleum ether 1:2, v/v, silica
23
gel plate); ½aŠ ¼ þ113 (c 0.56, CHCl3); IR (KBr)
D
1
m (cmꢀ1): 1638 (C@N); H NMR (300 MHz, CDCl3):
4.09 (dd, J = 8.4, 8.4 Hz, 2H, 2CHN), 4.62 (dd,
J = 8.4, 10.2 Hz, 2H in CH2O), 5.37 (dd, J = 10.2,
8.4 Hz, 2H in CH2O), 7.21–7.32 (m, 10H, ArH), 7.53
(dd, J = 7.2, 8.1 Hz, 2H, ArH), 8.15–8.23 (m, 4H,
ArH), 8.53 (s, 1H, ArH), 11.08 (s, 1H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 70.5, 74.1, 124.6, 125.3,
126.9, 127.5, 127.6, 128.6, 129.6, 129.9, 131.5, 132.2,
136.3, 142.5, 165.0. MS (EI) m/z (rel. intensity, %):
4.2.2.2. (S,S)-1,8-Bis[4-(1-methylpropyl)oxazolin-2-yl]-
anthracene 2b. Yellow crystals, yield 54%, mp 95–
97 ꢁC; Rf = 0.29 (ethyl acetate/petroleum ether 1:4,
20
v/v, silica gel plate); ½aŠ ¼ þ17:9 (c 0.80, CHCl3); IR
(KBr) (cmꢀ1): 1642 (DC@N); 1H NMR (400 MHz,
CDCl3): 1.03–1.06 (m, 12H, 4CH3), 1.54–1.59 (m, 2H
in 2CH2CH3), 1.88–1.98 (m, 4H, 2CHCH3 and 2H in
2CH2CH3), 4.14 (dd, J = 7.8, 8.0 Hz, 2H in 2CH2O),
4.53–4.60 (m, 2H in 2CHN), 4.65 (dd, J = 8.0, 9.2 Hz,
2H in 2CH2O), 7.49 (dd, J = 7.2, 8.4 Hz, 2H, ArH),
8.10–8.14 (m, 4H, ArH), 8.47 (s, 1H, ArH), 10.70 (s,
1H, ArH); 13C NMR (100.6 MHz, CDCl3): 22.6, 23.1,
25.6, 45.7, 65.7, 72.6, 124.5, 124.9, 125.6, 127.5, 129.4,
129.5, 131.4, 131.9, 163.8; MS (EI) m/z (rel. intensity,
%): 428 (M+, 14), 371 (M+ꢀC4H9, 6.6), 359 (5.5), 344
(100), 317 (3.7), 301 (6.6), 271 (21), 244 (31), 230 (9.6),
202 (9.8), 176 (4.1); Anal. Calcd for C28H32N2O2: C,
78.47; H, 7.53; N, 6.54. Found: C, 78.65; H, 7.52; N,
6.26.
+
468 (M+, 28), 405(M ꢀPhCH2, 65), 378 (82), 313
(3.0), 287 (20), 271 (100), 244 (22), 228 (48), 216 (21),
202 (16), 189 (4.6), 176 (4.6), 117 (8.8), 105(13), 91
(PhCHþ, 40); Anal. Calcd for C32H24N2O2: C, 82.03;
2
H, 5.16; N, 5.98. Found: C, 82.25; H, 5.19; N, 5.71.
4.2.3. Synthesis of (S,S)-N,N0-bis[1-(hydroxymethyl)-
benzyl]anthraquinone-1,8-dicarboxamide 5. To a solution
of anthraquinone-1,8-dicarbonyl dichloride (0.71 g,
2.1 mmol) in 20 mL of dry THF was added dropwise a
solution of L-phenylalaninol (0.91 g, 6.0 mmol) in
20 mL of dry THF at 0 ꢁC in an ice bath. The resulting
mixture was allowed to warm to room temperature and
was stirred overnight. After filtration, the solution was
concentrated to give a solid residue, which was purified
by column chromatography (silica gel, acetone/petro-
leum ether 3:2, v/v) to afford yellow crystals, yield
4.2.2.3. (S,S)-1,8-Bis[4-(1,1-dimethylethyl)oxazolin-2-
yl]anthracene 2c. Yellow crystals, yield 56%; mp 142–
144 ꢁC; Rf = 0.22 (ethyl acetate/petroleum ether 1:8,
20
v/v, silica gel plate); ½aŠ ¼ ꢀ10:9 (c 0.96, CHCl3); IR
(KBr) (cmꢀ1): 1641 (DC@N); 1H NMR (200 MHz,
CDCl3): d 1.07 (s, 18H, 6CH3), 4.23–4.38 (m, 4H,
2CH2), 4.49 (dd, 2H, J = 7.8, 9.4 Hz, 2CHN), 7.48
(dd, J = 7.2, 8.6 Hz, 2H, ArH), 8.06–8.13 (m, 4H,
50%, mp 275–276 ꢁC; Rf = 0.50 (acetone/petroleum
24
ether 3:2, v/v, silica gel plate); ½aŠ ¼ ꢀ111:5 (c 0.85,
D
CH3OH); IR (KBr) m (cmꢀ1): 3367 (O–H), 1676
1
(C9,10@O), 1626 (C@O); H NMR (200 MHz, DMSO-
ArH), 8.45(s, 1H, ArH), 10.45(s, 1H, ArH);
13C
d6): 2.78 (dd, J = 8.0, 13.6 Hz, 2H in 2CH2Ph), 2.95

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3729
(dd, J = 6.4, 13.6, 2H in 2CH2Ph), 3.45–3.63 (m, 4H,
2CH2O), 4.11–4.14 (m, 2H, 2CHN), 7.17–7.27 (m, 10H,
ArH), 7.46 (d, J = 7.6 Hz, 2H, ArH), 7.88 (dd, J = 7.6,
7.6 Hz, 2H, ArH), 8.21–8.27 (m, 4H, 2H in ArH and
2NH); 13C NMR (75.5 MHz, DMSO-d6): 36.2, 52.8,
62.0, 126.0, 127.2, 128.2, 129.3, 131.3, 132.7, 133.8,
139.0, 139.4, 168.2, 182.05, 182.11; MS (EI) m/z (rel.
intensity, %): 544 (M+ꢀH2O, 3.8), 531 (7.0), 513 (6.3),
471 (M+ꢀPhCH2, 39), 453 (40), 435 (22), 413 (54), 395
(25), 320 (18), 302 (70), 279 (77), 263 (57), 248 (10), 234
(100), 206 (13), 178 (13), 150 (28). Anal. Calcd for
C34H30N2O6: C, 72.58; H, 5.37; N, 4.98. Found: C,
72.66; H, 5.38; N, 4.74.
the reaction mixture was kept stirring for another 3 h.
The aziridine product was obtained after flash silica
gel chromatography with a mixture of petroleum ether
(60–90 ꢁC) and ethyl acetate (6:1, v/v) as an eluent.
For analytic data of known chiral aziridines 7a, 7b, 7e,
7i, 7j, 7o, 7p, 7q, and 7r, see the electronic supporting
information in Ref. 18; ent-7a, ent-7k, and ent-7l, see
the supporting information in Ref. 40.
4.3.1. (2S,3R)-2-Benzoyl-3-(4-phenylphenyl)-1-(4-toluene-
sulfonyl)aziridine 7c. Colorless crystals, yield 35%;
mp 152.5–154.0 ꢁC; Rf = 0.29 (ethyl acetate/petroleum
ether 1:5, v/v, silica gel plate); 87% ee determined by
HPLC with Chiralcel OD-H column with hexane/2-pro-
4.2.4. Synthesis of (S,S)-1,8-bis[(4-phenylmethyl)oxazo-
lin-2-yl]anthraquinone 4. To a solution of above dihy-
droxy diamide 5 (0.60 g, 1.1 mmol) in 30 mL of dry
THF was added 5.0 mL (68.6 mmol) of SOCl2. The
resulting mixture was heated to reflux for 4 h to give
clear yellow solution. After removal of solution and
excess SOCl2 in vacuo, the residue was dissolved in
50 mL of ethyl acetate, then washed with 20 mL of
2 mol/L K2CO3, water, and brine, respectively. The
organic layer was dried over Na2SO4, filtered, and con-
centrated in vacuo to yield yellow solid, which was
used in further reaction without purification. The yellow
solid was dissolved in 45mL of acetonitrile, 10.5g
(76.1 mmol) of K2CO3, and 4.0 mL of water were
added. After refluxing for 8 h, the resulting solution
was cooled to room temperature and concentrated, then
200 mL of ethyl acetate and 70 mL of water were added.
The organic extract was dried over Na2SO4, filtered, and
concentrated to give crude product, which was purified
by column chromatography (silica gel, acetone/petro-
leum ether 1:1, v/v) to afford yellow crystals, yield
panol (70:30, v/v) as an eluent at a flow rate of 0.6 mL/
20
min (smajor = 22.4 min; sminor = 37.5min); ½aŠ ¼ þ1:3
D
(c 0.64, CHCl3, 87% ee); IR (KBr) m (cmꢀ1): 1689
(s), 1332 (s), 1161 (s); 1H NMR (300 MHz, CDCl3):
2.40 (s, 3H, CH3), 4.34 (d, J = 4.2 Hz, 1H, CH), 4.55
(d, J = 4.2 Hz, 1H, CH), 7.24 (d, J = 5.7 Hz, 2H,
ArH), 7.36–7.63 (m, 12H, ArH), 7.74 (d, J = 5.7 Hz,
2H, ArH), 8.05–8.09 (m, 2H, ArH); 13C NMR
(75.5 MHz, CDCl3): d 21.6, 47.4, 50.2, 127.1, 127.3,
127.6, 127.7, 128.0, 128.80, 128.84, 128.9, 129.5, 131.8,
134.1, 135.9, 140.3, 141.8, 144.4, 190.3; MS (EI) m/z
(rel. intensity, %): 453 (M+, 11), 298 (M+ꢀTs, 11), 105
(PhCO+, 100), 77 (Ph+, 41). Anal. Calcd for
C28H23NO3S: C, 74.15; H, 5.11; N, 3.09. Found: C,
74.14; H, 5.08; N, 2.91.
4.3.2. (2S,3R)-2-Benzoyl-3-(4-fluorophenyl)-1-(4-toluene-
sulfonyl)aziridine 7d. Colorless crystals, yield 72%; mp
130.5–131.5 ꢁC; Rf = 0.19 (ethyl acetate/petroleum ether
1:5, v/v, silica gel plate); 62% ee determined by HPLC
with Chiralcel OD-H column with hexane/2-propanol
68%; mp 185–186 ꢁC; Rf = 0.57 (acetone/petro-
(90:10, v/v) as an eluent at a flow rate of 0.8 mL/min
20
20
leum ether 1:1, v/v, silica gel plate); ½aŠ ¼ ꢀ121 (c
(smajor = 23.7 min;
s
minor = 32.3 min);
½aŠ ¼ þ1:7
D
D
0.51, acetone); IR (KBr) m (cmꢀ1): 1684 (C@N), 1670
(C9,10@O); 1H NMR (300 MHz, CDCl3): 2.93 (dd,
J = 8.4, 13.6 Hz, 2H in 2CH2Ph), 3.31 (dd, J = 5.7,
13.6 Hz, 2H in 2CH2Ph), 4.23 (dd, J = 7.2, 8.2 Hz, 2H
in 2CH2O), 4.42 (dd, J = 8.2, 9.0 Hz, 2H in 2CH2O),
4.62–4.70 (m, 2H, 2CHN), 7.21–7.33 (m, 10H, ArH),
7.77–7.89 (m, 4H, ArH), 8.41 (dd, J = 1.5, 7.5 Hz, 2H,
ArH); 13C NMR (75.5 MHz, CDCl3): 44.4, 68.2, 73.1,
126.5, 128.6, 129.2, 129.8, 133.2, 133.4, 136.1, 138.0,
164.7, 181.6, 181.8; MS (EI) m/z (rel. intensity, %):
526 (M+, 9.7), 512 (8.9), 435 (M+ꢀPhCH2, 89), 381
(3.7), 378 (4.2), 343 (4.2), 301 (100), 287 (6.7), 260
(7.6), 246 (18), 235 (3.3), 218 (5.3), 190 (11), 176 (6.5).
Anal. Calcd for C34H26N2O4: C, 77.55; H, 4.98; N,
5.32. Found: C, 77.78; H, 4.90; N, 5.11.
(c 0.77, CHCl3, 62% ee); IR (KBr) m (cmꢀ1): 1689
(s), 1332 (s), 1162 (s); 1H NMR (300 MHz, CDCl3):
2.40 (s, 3H, CH3), 4.28 (d, J = 4.2 Hz, 1H, CH), 4.49
(d, J = 4.2 Hz, 1H, CH), 7.00–7.05(m, 2H, ArH),
7.22–7.35(m, 4H, ArH), 7.46–7.15 (m, 2H, ArH),
7.60–7.72 (m, 3H, ArH), 8.02–8.06 (m, 2H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 21.6, 46.8, 45.0, 115.6 (d,
2JF–C–C = 27.9 Hz), 127.7, 128.6, 128.8, 128.9, 129.4,
1
129.5, 129.5, (d, JF–C = 288.8 Hz), 134.1, 135.8, 136.4,
144.5, 190.2; MS (EI) m/z (rel. intensity, %): 395
+
(M+, 2), 240 (M+ꢀTs, 78), 105(PhCO
, 100),
77 (Ph+, 79). Anal. Calcd for C22H18FNO3S: C, 66.82;
H, 4.59; N, 3.54. Found: C, 66.67; H, 4.67; N, 3.29.
4.3.3. (2S,3R)-2-Benzoyl-3-(4-bromophenyl)-1-(4-toluene-
sulfonyl)aziridine 7f. Colorless crystals, yield 58%;
mp 150–152 ꢁC; Rf = 0.23 (ethyl acetate/petroleum ether
1:6, v/v, silica gel plate); 52% ee determined by HPLC
with chiral OD column with hexane/2-propanol (90:10,
4.3. General procedure for the asymmetric aziridination of
chalcones or olefins
A three-necked flask (25mL) was charged with chalcone
v/v) as an eluent at a flow rate of 0.8 mL/min
20
6 or olefin (1.50 mmol), AnBOX 2 (0.06 mmol), and
CuOTfÆ1/2PhH (13 mg, 0.05mmol) under nitrogen
atmosphere. Dichloromethane (8 mL) was added by
syringe and the resulting mixture was stirred for 1 h at
24 ꢁC. PhI@NTs (373 mg, 1.00 mmol) was added por-
tionwise to the mixture over 2 h. After the addition,
(sminor = 32.25min;
s
major = 48.16 min); ½aŠ ¼ þ3:8
(c 1.02, CHCl3, 52% ee); IR (KBr) m (cmꢀ1): D1686 (s),
1
1332 (s), 1161 (s); H NMR (300 MHz, CDCl3): 2.40
(s, 3H, CH3), 4.24 (d, J = 4.2 Hz, 1H, CH), 4.48 (d,
J = 4.2 Hz, 1H, CH), 7.0–7.27 (m, 4H, ArH), 7.45–
7.52 (m, 4H, ArH), 7.61–7.73 (m, 3H, ArH), 8.02–8.06

3730
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
(m, 2H, ArH); 13C NMR (75.5 MHz, CDCl3): d 21.6,
46.5, 50.2, 123.0, 127.6, 128.8, 128.9, 129.1, 129.6,
131.8, 132.0, 134.2, 135.8, 136.3, 144.6, 190.0; MS (EI)
m/z (rel. intensity, %) 455 (M+, 15), 299 (M+ꢀTs, 12),
105(PhCO +, 100), 77 (Ph+, 31). Anal. Calcd for
C22H18BrNO3S: C, 57.90; H, 3.98; N, 3.07. Found: C,
57.89; H, 4.02; N, 2.92.
(d, J = 8.1 Hz, 2H, ArH), 8.08–8.13 (m, 2H, ArH); 13C
NMR (75.5 MHz, CDCl3): 21.6, 47.2, 50.2,
d
115.9, 116.2, 127.4, 127.7, 128.6, 128.9, 129.5, 131.7,
131.8, 132.9, 136.5, 144.5, 188.8; MS (EI) m/z (rel. inten-
sity, %): 395(M +, 19), 240 (M+ꢀTs, 100), 155 (Ts+, 21),
123 (4-FC6H4CO+, 100). Anal. Calcd for C22H18FNO3S:
C, 66.82; H, 4.59; N, 3.54. Found: C, 66.70; H, 4.61;
N, 3.55.
4.3.4. (2S,3R)-2-Benzoyl-3-(4-methoxyphenyl)-1-(4-tolu-
enesulfonyl)aziridine 7g. Yellow crystals, yield 61%;
mp 43–45 ꢁC; Rf = 0.21 (ethyl acetate/petroleum ether
1:4, v/v, silica gel plate); 37% ee determined by HPLC
with chiral OD-H column with hexane/2-propanol
4.3.7. (2S,3R)-2-(4-Chlorobenzoyl)-3-phenyl-1-(4-toluene-
sulfonyl)aziridine 7m. Colorless crystals, yield 74%;
mp 145–147 ꢁC; Rf = 0.32 (ethyl acetate/petroleum ether
1:5, v/v, silica gel plate); 58% ee determined by HPLC
with Chiral OD-H column with hexane/2-propanol
(90:10, v/v) as an eluent at a flow rate of 0.7 mL/min
20
(sminor = 27.44 min; smajor = 35.07 min); ½aŠ ¼ þ23:8
(90:10, v/v) as an eluent at a flow rate of 0.5mL/min
(c 0.98, CHCl3, 37% ee); IR (KBr) m (cmꢀ1)D: 1694 (s),
(smajor = 46.79 min; sminor = 51.26 min); ½aŠ ¼ þ10:6
20
D
1
1329 (s), 1162 (s); H NMR (300 MHz, CDCl3): 2.42
(c 1.01, CHCl3, 58% ee); IR (KBr) m (cmꢀ1): 1687 (s),
1
(s, 3H, CH3), 3.67 (s, 3H, CH3), 4.33 (d, J = 7.5Hz,
1H, CH), 4.38 (d, J = 7.5Hz, 1H, CH), 6.67–6.70 (m,
2H, ArH), 7.14–7.17 (m, 2H, ArH), 7.33–7.41 (m, 4H,
ArH), 7.50–7.53 (m, 1H, ArH), 7.84–7.87 (m, 2H,
ArH), 7.95–7.98 (m, 2H, ArH); 13C NMR (75.5 MHz,
CDCl3): d 21.7, 46.3, 48.3, 55.1, 113.7, 123.1, 128.0,
128.3, 128.5, 128.7, 129.8, 133.8, 134.4, 135.6, 145.0,
159.6, 189.1; MS (EI) m/z (rel. intensity, %): 407 (M+,
6), 251 (M+ꢀTsꢀH, 100), 236 (M+ꢀTsꢀHꢀCH3, 18),
105(PhCO +, 27), 77 (Ph+, 32). Anal. Calcd for
C23H21NO4S: C, 67.79; H, 5.19; N, 3.44. Found: C,
67.75; H, 5.11; N, 3.13.
1330 (s), 1161 (s); H NMR (300 MHz, CDCl3): 2.41
(s, 3H, CH3), 4.18 (d, J = 4.0 Hz, 1H, CH), 4.52 (d,
J = 4.4 Hz, 1H, CH), 7.24 (d, J = 8.2 Hz, 2H, CH),
7.33 (s, 5H, ArH), 7.45 (d, J = 8.8 Hz, 2H, ArH), 7.71
(d, J = 8.2 Hz, 2H, ArH), 8.00 (d, J = 8.4 Hz, 2H,
ArH); 13C NMR (75.5 MHz, CDCl3): d 21.6, 47.2,
50.3, 127.3, 127.7, 128.7, 128.9, 129.1, 129.5, 130.3,
132.8, 134.3, 136.3, 140.7, 144.6, 189.2; MS (EI) m/z
(rel. intensity, %): 411 (M+, 5), 255 (M+ꢀTsꢀH, 100),
139 (ClC6H4CO+, 71), 111 (C1C6Hþ, 35), 77 (Ph+,
4
24). Anal. Calcd for C22H18ClNO3S: C, 64.15; H, 4.40;
N, 3.40. Found: C, 64.10; H, 4.53; N, 3.12.
4.3.5. (2S,3R)-2-Benzoyl-3-(2-methoxyphenyl)-1-(4-tolu-
enesulfonyl)aziridine 7h. Colorless crystals, yield 21%;
mp 143–145 ꢁC; Rf = 0.24 (ethyl acetate/petroleum ether
1:5, v/v, silica gel plate); 27% ee determined by HPLC
with chiral OD-H column with hexane/2-propanol
4.3.8. (2S,3R)-2-(4-Bromobenzoyl)-3-phenyl-1-(4-toluene-
sulfonyl)aziridine 7n. Colorless needle crystals, yield
64%; mp 143–145 ꢁC; Rf = 0.32 (ethyl acetate/petroleum
ether 1:6, v/v, silica gel plate); 50% ee determined by
HPLC with chiral OD-H column with hexane/2-propa-
(90:10, v/v) as an eluent at a flow rate of 0.8 mL/min
nol (93:7, v/v) as an eluent at a flow rate of 0.8 mL/min
20
20
(smajor = 27.29 min; sminor = 32.05min); ½aŠ ¼ ꢀ13:6
(smajor = 36.2 min; sminor = 39.7 min); ½aŠ ¼ þ10:8 (c
D
D
(c 0.45, CHCl3, 27% ee); IR (KBr) m (cmꢀ1): 1693 (s),
1.00, CHCl3, 50% ee); IR (KBr) m (cmꢀ1): 1688 (s),
1
1
1328 (s), 1162 (s); H NMR (300 MHz, CDCl3): 2.41
1329 (s), 1160 (s); H NMR (300 MHz, CDCl3): 2.41
(s, 3H, CH3), 3.73 (s, 3H, CH3), 4.41 (d, J = 4.5Hz,
1H, CH), 4.64 (d, J = 4.5Hz, 1H, CH), 6.84–6.92 (m,
2H, ArH), 7.20–7.34 (m, 4H, ArH), 7.48–7.51 (m, 2H,
ArH), 7.61–7.64 (m, 1H, ArH), 7.72–7.74 (m,
2H, ArH), 8.09–8.13 (m, 2H, ArH); 13C NMR (75.5
MHz, CDCl3): d 21.6, 44.9, 48.2, 55.2, 110.3, 120.3,
120.7, 127.8, 128.0, 128.7, 128.9, 129.3, 130.1, 133.9,
136.0, 136.6, 144.1, 158.8, 190.9; MS (EI) m/z (rel. inten-
sity, %): 407 (M+, 28), 376 (M+ꢀOMe, 4.5), 252
(M+ꢀTs, 20), 155 (Ts+, 3.3), 105(PhCO +, 100), 77
(Ph+, 8.9). Anal. Calcd for C23H21NO4S: C, 67.79; H,
5.19; N, 3.44. Found: C, 67.87; H, 5.11; N, 3.13.
(s, 3H, CH3), 4.17 (d, J = 4.2 Hz, 1H, CH), 4.52 (d,
J = 4.2 Hz, 1H, CH), 7.22–7.33 (m, 7H, ArH), 7.62 (d,
J = 8.7 Hz, 2H, ArH), 7.70 (d, J = 8.4 Hz, 2H, ArH),
7.91 (d, J = 8.7 Hz, 2H, ArH),; 13C NMR (75.5 MHz,
CDCl3): d 21.6, 47.1, 50.2, 126.4, 127.3, 127.7, 128.6,
128.9, 129.5, 130.3, 132.1, 132.8, 134.6, 136.3, 144.5,
189.4; MS (EI) m/z (rel. intensity, %): 455 (M+, 22),
300 (M+ꢀTs, 7.8), 185(95), 183 (4-BrC 6H4CO+, 100),
155 (Ts+, 39). Anal. Calcd for C22H18BrNO3S: C,
57.90; H, 3.98; N, 3.07. Found: C, 57.85; H, 4.09; N,
3.06.
4.3.9. (2S,3R)-2-(4-Nitrobenzoyl)-3-phenyl-1-(4-toluene-
sulfonyl)aziridine 7o. Colorless needle crystals, yield
61%; mp 146–148 ꢁC; Rf = 0.28 (ethyl acetate/petroleum
4.3.6. (2S,3R)-2-(4-Fluorobenzoyl)-3-phenyl-1-(4-toluene-
sulfonyl)aziridine 7l. Colorless needle crystals, yield
72%; mp 123.5–125.5 ꢁC; Rf = 0.34 (ethyl acetate/petro-
leum ether 1:5, v/v, silica gel plate); 54% ee determined
by HPLC with chiral OD-H column with hexane/2-pro-
ether 1:6, v/v, silica gel plate); The ee value cannot be
25
determined due to very poor solubility; ½aŠ ¼ þ6:3 (c
D
0.73, CHCl3); IR (KBr) m (cmꢀ1): 1697 (s), 1335(s),
1159 (s); 1H NMR (300 MHz, CDCl3): 2.42 (s, 3H,
CH3), 4.17 (d, J = 4.2 Hz, 1H, CH), 4.58 (d,
J = 4.2 Hz, 1H, CH), 7.25–7.35 (m, 7H, ArH), 7.72 (d,
J = 8.4 Hz, 2H, ArH), 8.23 (d, J = 9.0 Hz, 2H, ArH),
8.34 (d, J = 9.0 Hz, 2H, ArH); 13C NMR (50 MHz,
CDCl3): d 21.6, 47.2, 50.5, 123.9, 127.2, 127.7, 128.8,
panol (94:6, v/v) as an eluent at a flow rate of 0.8 mL/
25
min (smajor = 35.06 min; sminor = 38.21 min); ½aŠ ¼ þ7:9
D
(c 0.68, CHCl3, 54% ee); IR (KBr) m (cmꢀ1): 1687 (s),
1
1333 (s), 1160 (s); H NMR (300 MHz, CDCl3): 2.41
(s, 3H, CH3), 4.20 (d, J = 4.2 Hz, 1H, CH), 4.52 (d,
J = 4.2 Hz, 1H, CH), 7.13–7.33 (m, 9H, ArH), 7.72

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3731
129.1, 129.7, 130.0, 132.6, 138.0, 140.2, 144.9, 189.5; MS
(EI) m/z (rel. intensity, %): 422 (M+, 20), 267 (M+ꢀTs,
14), 155 (Ts+, 17), 150 (100, 4-NO2C6H4CO+, 100)
HRMS (EI) m/z calcd for C22H18N2O5S (M+)
422.0936; Found 422.0924.
127.7, 128.8, 128.9, 129.1, 129.5, 131.6, 133.4, 134.8,
136.4, 144.5, 145.4, 189.5; MS (EI) m/z (rel. intensity,
+
%): 425(M
,
11), 269 (M+ꢀTsꢀH, 100), 179
(M+ꢀTsꢀCH3C6H4, 49), 119 (CH3C6H4CO+, 76), 91
(CH3Ph+, 84). Anal. Calcd for C23H20ClNO3S: C,
64.86; H, 4.73; N, 3.29. Found: C, 64.79; H, 4.76; N,
3.10.
4.3.10. (2S,3R)-2-(4-Methoxybenzoyl)-3-phenyl-1-(4-tol-
uenesulfonyl)aziridine 7p. Colorless needle crystals,
yield 74%; mp 123–125 ꢁC; Rf = 0.22 (ethyl acetate/
petroleum ether 1:4, v/v, silica gel plate); 62% ee deter-
mined by HPLC with chiral OD-H column with hex-
ane/2-propanol (90:10, v/v) as an eluent at a flow rate
4.3.13. (2S,3R)-2-Benzoyl-3-(4-trifluoromethylphenyl)-1-
(4-toluenesulfonyl)aziridine 7w. Colorless crystals, yield
69%; mp 156.0–158.0 ꢁC; Rf = 0.36 (ethyl acetate/petro-
leum ether 1:5, v/v, silica gel plate); 67% ee determined
by HPLC with Chiralcel OD-H column with hexane/2-
propanol (90:10, v/v) as an eluent at a flow rate of
of 0.8 mL/min (smajor = 48.75min; sminor = 62.22 min);
20
½aŠ ¼ þ3:4 (c 1.02, CHCl3, 62% ee); IR (KBr) m
D
(cmꢀ1): 1677 (s), 1332 (s), 1164 (s); 1H NMR
(300 MHz, CDCl3): 2.40 (s, 3H, CH3), 3.88 (s, 3H,
CH3), 4.26 (d, J = 4.2 Hz, 1H, CH), 4.50 (d,
J = 4.2 Hz, 1H, CH), 6.93–6.96 (m, 2H, ArH), 7.21–
7.30 (m, 7H, ArH), 7.71 (d, J = 8.4 Hz, 2H, ArH),
8.03–8.06 (m, 2H, ArH); 13C NMR (75.5 MHz, CDCl3):
d 21.6, 47.3, 50.1, 55.6, 114.0, 127.5, 127.7, 128.6, 128.8,
129.1, 129.5, 131.4, 133.0, 136.6, 144.3, 164.3, 188.4; MS
(EI) m/z (rel. intensity, %): 407 (M+, 12), 251
(M+ꢀTsꢀH, 100), 135(CH 3OC6H4CO+, 60), 107
0.8 mL/min
(smajor = 23.0 min;
sminor = 34.8 min);
20
½aŠ ¼ þ9:0 (c 0.55, CHCl3, 67% ee); IR (KBr) m
D
(cmꢀ1): 1694 (s), 1325(s), 1164 (s);
1H NMR
(300 MHz, CDCl3): 2.42 (s, 3H, CH3), 4.22 (d,
J = 4.2 Hz, 1H, CH), 4.60 (d, J = 4.2 Hz, 1H, CH),
7.24–7.26 (m, 2H, ArH), 7.45–7.53 (m, 4H, ArH),
7.59–7.65 (m, 3H, ArH), 7.71–7.74 (m, 2H, ArH),
8.03–8.07 (m, 2H, ArH); 13C NMR (75.5 MHz, CDCl3):
d 21.6, 46.0, 50.8, 125.6, 125.6, 127.7, 128.8, 129.0,
129.6, 134.2, 135.8, 136.2, 137.3, 144.7, 189.7; MS (EI)
m/z (rel. intensity, %): 445(M +, 1), 290 (M+ꢀTs, 56),
105(PhCO +, 100), 77 (Ph+, 62). Anal. Calcd for
C23H18F3NO3S: C, 62.01; H, 4.07; N, 3.14. Found: C,
62.05; H, 4.06; N, 2.94.
(CH3OC6Hþ, 10), 77 (Ph+, 42). Anal. Calcd for
4
C23H21NO4S: C, 67.79; H, 5.19; N, 3.44. Found: C,
67.76; H, 5.21; N, 3.50.
4.3.11. (2S,3R)-2-(4-Chlorobenzoyl)-3-(4-chlorophenyl)-
1-(4-toluenesulfonyl)aziridine 7q. Colorless crystals,
yield 68%; mp 161–163 ꢁC; Rf = 0.30 (ethyl acetate/
petroleum ether 1:6, v/v, silica gel plate); 43% ee deter-
mined by HPLC with Chiralcel OD column with hex-
ane/2-propanol (90:10, v/v) as an eluent at a flow rate
4.3.14. (R,R)-2-Benzyl-3-phenyl-1-(4-toluenesulfonyl)azi-
ridine 14. Colorless crystals, yield 41%; mp 113–
114 ꢁC; Rf = 0.45(ethyl acetate/petroleum ether 1:6,
v/v, silica gel plate); 6% ee determined by HPLC with
Daicel Chiralcel OJ column with hexane/2-propanol
of 0.8 mL/min (smajor = 35.46 min; sminor = 42.53 min);
(70:30, v/v) as an eluent at a flow rate of 0.6 mL/min
20
20
½aŠ ¼ þ9:25( c 1.01, CHCl3, 43% ee); IR (KBr) m
(smajor = 40.95min;
s
minor = 54.46 min); ½aŠ ¼ ꢀ2:45
D
D
(cmꢀ1): 1695(s), 1330 (s), 1162 (s);
1H NMR
(c 1.06, CHCl3, 6% ee), the configuration was assigned
tentatively on the basis of the configuration of aziridine
16; 1H NMR (300 MHz, CDCl3): 2.41 (s, 3H, CH3), 3.06
(ddd, J = 9.6, 4.2, 4.2 Hz, 1H, CHCH2), 3.41 (dd,
J = 14.4, 9.6 Hz, 1H, CH2), 3.61 (dd, J = 14.4, 4.2 Hz,
1H, CH2), 4.01 (d, J = 4.2 Hz, 1H, CH), 7.09–7.13 (m,
3H, ArH), 7.22–7.24 (m, 4H, ArH), 7.29–7.31 (m, 5H,
ArH), 7.84 (d, J = 8.4 Hz, 2H, ArH); 13C NMR
(75.5 MHz, CDCl3): d 21.6, 34.7, 49.1, 52.8, 126.5,
126.8, 127.3, 128.1, 128.5, 128.6, 128.7, 129.6, 135.0,
137.5, 137.7, 144.1; MS (EI) m/z (rel. intensity, %):
363 (M+, 0.72), 208 (M+ꢀTs, 100), 117
(M+ꢀTsꢀPhCH2, 36), 103 (M+ꢀTsꢀPhCH2CH–H,
15), 91 (PhCHþ2 , 100), 77 (Ph+, 15). Anal. Calcd for
C22H21NO2S: C, 72.70; H, 5.82; N, 3.85. Found: C,
72.71; H, 5.84; N, 3.65.
(300 MHz, CDCl3): 2.42 (s, 3H, CH3), 4.14 (d,
J = 4.2 Hz, 1H, CH), 4.49 (d, J = 4.2 Hz, 1H, CH),
7.24–7.32 (m, 6H, ArH), 7.44–7.47 (m, 2H, ArH),
7.69–7.71 (m, 2H, ArH), 7.97–8.00 (m, 2H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 21.6, 46.3, 50.3, 127.7,
128.7, 128.9, 129.2, 129.6, 130.3, 131.4, 134.1, 134.9,
136.1, 140.8, 144.8, 188.9; MS (EI) m/z (rel. intensity,
%): 445(M +, 22), 290 (M+ꢀTs, 59), 255 (M+ꢀTsꢀCl,
3), 139 (ClC6H4CO+, 100), 111 (C1C6Hþ, 42). Anal.
4
Calcd for C22H17Cl2NO3S: C, 59.20; H, 3.84; N, 3.14.
Found: C, 59.20; H, 3.90; N, 2.99.
4.3.12. (2S,3R)-2-(4-Methylbenzoyl)-3-(4-chlorophenyl)-
1-(4-toluenesulfonyl)aziridine 7r. Colorless crystals,
yield 66%; mp 163.5–165 ꢁC; Rf = 0.30 (ethyl acetate/
petroleum ether 1:6, v/v, silica gel plate); 39% ee deter-
mined by HPLC with chiral OD-H column with hex-
ane/2-propanol (90:10, v/v) as an eluent at a flow rate
4.3.15. (2S,3R)-2-Phenyl-3-phenoxylmethyl-1-(4-toluene-
sulfonyl)aziridine 16. Colorless crystals, yield 24%; mp
108–109 ꢁC; Rf = 0.27 (ethyl acetate/petroleum ether
1:10, v/v, silica gel plate); 40% ee determined by HPLC
with Chiralcel OD column with hexane/2-propanol
of 0.5mL/min ( smajor = 39.00 min; sminor = 56.87 min);
20
½aŠ ¼ þ3:6 (c 1.04, CHCl3 39% ee); IR (KBr) m
D
(cmꢀ1): 1683 (s), 1331 (s), 1162 (s); 1H NMR
(300 MHz, CDCl3): 2.41 (s, 3H, CH3), 2.44 (s, 3H,
CH3), 4.23 (d, J = 4.2 Hz, 1H, CH), 4.46 (d,
J = 4.0 Hz, 1H, CH), 721–7.29 (m, 8H, ArH), 7.71 (d,
J = 8.2 Hz, 2H, ArH), 7.94 (d, J = 8.2 Hz, 2H, ArH);
13C NMR (75.5 MHz, CDCl3): d 21.6, 21.8, 46.6, 50.2,
(90:10, v/v) as an eluent at a flow rate of 0.8 mL/min
20
(smajor = 20.4 min; sminor = 29.8 min); ½aŠ ¼ ꢀ10:8 (c
D
1.05, CHCl3, 40% ee); IR (KBr) m (cmꢀ1): 1598 (s),
1325(s), 1161 (s), 1089 (s); 1H NMR (300 MHz,
CDCl3): 2.41 (s, 3H, CH3), 3.29 (ddd, J = 5.8, 6.4,

3732
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
4.4 Hz, 1H, CHCH2), 4.01 (d, J = 4.4 Hz, 1H, CH), 4.53
(dd, J = 10.6, 6.4 Hz, 1H, CH2), 4.73 (dd, J = 10.6,
5.8 Hz, 1H, CH2), 6.89–7.02 (m, 3H, ArH), 7.20–7.29
(m, 9H, ArH), 7.84 (d, J = 8.2 Hz, 2H, ArH); 13C
NMR (75.5 MHz, CDCl3): d 21.6, 46.9, 49.6, 114.7,
116.0, 121.4, 126.8, 127.6, 128.4, 128.6, 129.5, 134.3,
136.7, 144.3, 158.0; MS (EI) m/z (rel. intensity, %):
379 (M+, 1), 286 (M+ꢀPhO, 3), 272 (M+ꢀPhOCH2,
2), 224 (M+ꢀTs, 44), 155 (Ts+, 12), 130
(M+ꢀTsꢀPhOꢀH, 52); Anal. Calcd for C22H21NO3S:
C, 69.63; H, 5.58; N, 3.69. Found: C, 69.60; H, 5.63;
N, 3.45.
CH2), 4.35(dd, J = 4.9, 11.3 Hz, 1H in CH2), 7.15–
7.28 (m, 7H, ArH), 7.82 (d, J = 8.2 Hz, 2H, ArH); 13C
NMR (50 MHz, CDCl3): d ꢀ5.3, 18.2, 21.6, 25.8, 47.5,
52.3, 60.7, 126.8, 127.4, 128.2, 128.4, 129.5, 134.7,
137.2, 144.0.
4.4.1. Preparation of (2S,3R)-2-hydroxymethyl-3-phenyl-
1-(4-toluenesulfonyl)aziridine 21. To a stirred solution
of 20 (0.147 g, 0.36 mmol) in 2.5mL of THF was added
tetrabutylammonium fluoride (0.36 mL, 0.36 mmol,
1 mol/L solution in THF) by syringe at ꢀ78 ꢁC, and
the resulting mixture was allowed to warm to 0 ꢁC then
stirred at this temperature for an additional 1 h. The
mixture was poured into ice-water and extracted with
Et2O. The extract was washed successively with 5% citric
acid, water, 5% NaHCO3, and water, and dried over
anhydrous Na2SO4. Concentration under reduced pres-
sure gave an oily residue, which was purified by flash
chromatography over silica gel with ethyl acetate/petro-
leum ether (1:4, v/v) to give colorless oil 69 mg of 21,
4.3.16. (2S,3R)-2-Benzoyl-3-methyl-1-(4-toluenesulfonyl)-
aziridine 32. Colorless viscous oil, yield 42%; Rf =
0.28 (ethyl acetate/petroleum ether 1:5, v/v, silica gel
plate); 9% ee determined by HPLC with Chiralcel
OD column with hexane/2-propanol (90:10, v/v) as an
eluent at a flow rate of 0.8 mL/min (smajor = 14.93 min;
20
D
s
minor = 18.30 min); ½aŠ ¼ ꢀ5:4 (c 0.21, CHCl3, 9% ee);
IR (KBr) (cmꢀ1): 1689 (s), 1325(s), 1162 (s); 1H NMR
(300 MHz, CDCl3): 1.83 (d, J = 6.0 Hz, 3H, CH3),
2.38 (s, 3H, CH3), 3.25(qd, J = 6.0 Hz, 5.2 Hz, 1H,
CHCH3), 4.15(d, J = 5.2 Hz, 1H, CHCH), 7.25–7.28
(m, 2H, ArH), 7.43–7.60 (m, 3H, ArH), 7.84–8.97 (m,
4H, ArH); 13C NMR (75.5 MHz, CDCl3): d 13.6, 21.5,
45.4, 46.9, 127.4, 128.5, 128.7, 129.5, 134.0, 135.5,
136.9, 144.4, 191.6; MS (EI) m/z (rel. intensity, %):
315(M +, 0.14), 160 (M+ꢀTs, 47), 105(PhCO +, 100),
77 (Ph+, 54). HRMS (EI) m/z calcd for C17H17NO3S
(M+) 315.0929; Found 315.0921.
yield 65%. Rf = 0.12 (ethyl acetate/petroleum ether 1:4,
20
v/v, silica gel plate); ½aŠ ¼ ꢀ12:9 (c 0.51, CHCl3,
D
15
27% ee), lit.34 ½aŠ ¼ þ48:4 (c 1.5, CHCl3, >98% ee)
D
1
for (2R,3S)-enantiomer; H NMR (200 MHz, CDCl3):
d 2.41 (s, 3H, CH3), 3.15(dd, J = 4.9, 9.8 Hz, 1H,
OH), 3.19 (ddd, J = 3.1, 4.4, 8.4 Hz, 1H, CH), 4.03 (d,
J = 4.4 Hz, 1H, CH), 4.19 (ddd, J = 4.9, 8.4, 13.3 Hz,
1H in CH2), 4.32 (ddd, J = 3.2, 9.8, 13.3 Hz, 1H in
CH2), 7.13–7.31 (m, 7H, ArH), 7.82 (d, J = 8.2 Hz,
2H, ArH); 13C NMR (50 MHz, CDCl3): d 21.6, 46.3,
54.7, 60.7, 126.4, 127.1, 128.4, 128.6, 129.7, 134.5,
137.0, 144.4.
4.3.17. 2-Acetyl-3,3-dimethyl-1-(4-toluenesulfonyl)aziri-
dine 33. Colorless viscous oil, yield 20%; Rf = 0.21
(ethyl acetate/petroleum ether 1:5, v/v, silica gel plate);
6% ee determined by HPLC with Chiralcel OD-H col-
umn with hexane/2-propanol (90:10, v/v) as an eluent
4.4.2. Preparation of (2S,3R)-3-phenyl-1-(4-toluenesul-
fonyl)aziridin-2-ylmethyl tosylate 22. To a stirred solu-
tion of 21 (69 mg, 0.23 mmol) in 4 mL of CH2Cl2 were
added 0.035mL (0.25mmol) of Et 3N and 46 mg
(0.24 mmol) of TsCl at 0 ꢁC. The mixture was allowed
to stir overnight with warming to room temperature,
then diluted with 5mL of water and extracted with
CH2Cl2. The combined organic layer was washed with
brine and dried over anhydrous Na2SO4. The usual
work-up was followed by flash chromatography over
silica gel with ethyl acetate/petroleum ether (1:5, v/v)
as an eluent to give colorless oil 65mg of 22, yield
63%. Rf = 0.23 (ethyl acetate/petroleum ether 1:5,
at a flow rate of 0.8 mL/min (sminor = 10.7 min; smajor
=
20
11.4 min); ½aŠ ¼ ꢀ1:8 (c 0.85, CHCl3, 6% ee); 1H NMR
D
(300 MHz, CDCl3): 1.30 (s, 3H, CH3), 1.82 (s, 3H,
CH3), 1.97 (s, 3H, CH3), 2.45(s, 3H, CH 3), 3.48 (s,
1H, CH), 7.33–7.36 (m, 2H, ArH), 7.86–7.88 (m, 2H,
ArH); 13C NMR (75.5 MHz, CDCl3): d 21.0, 21.6,
21.9, 28.7, 53.3, 55.2, 127.4, 129.7, 137.2, 144.4, 202.1.
Racemate was reported in the literature.41
20
1
4.4. Determination of the absolute configuration of (2S,3R)-
2-phenyl-3-phenoxylmethyl-1-(4-toluenesulfonyl)aziridine
16
v/v, silica gel plate); ½aŠ ¼ ꢀ2:3 (c 0.27, CHCl3); H
D
NMR (200 MHz, CDCl3): 2.41 (s, 3H, CH3, 27% ee),
2.45(s, 3H, CH ), 3.14 (ddd, J = 4.0, 5.8, 7.2 Hz, 1H,
3
CH), 3.89 (d, J = 4.0 Hz, CH), 4.60 (dd, J = 7.2,
11.4 Hz, 1H in CH2), 4.71 (dd, J = 5.8, 11.6 Hz, 1H in
CH2), 7.13–7.31 (m, 9H, ArH), 7.79 (d, J = 8.2 Hz,
2H, ArH), 7.81 (d, J = 8.6 Hz, 2H, ArH); 13C NMR
(50 MHz, CDCl3): d 21.59, 21.64, 47.3, 47.9, 66.7,
126.8, 127.6, 128.0, 128.6, 129.7, 130.0, 133.6, 136.4,
144.6, 145.2.
tert-Butyldimethylsilyl cinnamyl ether 19 was prepared
in the yield of 84% according to the literature method,34
and was aziridinated asymmetrically by using general
procedure as mentioned above to afford (2S,3R)-2-tert-
butyldimethylsiloxymethyl-3-phenyl-1-(4-toluenesulfo-
nyl)aziridine (20), colorless oil, yield 27%; Rf = 0.43
(ethyl acetate/petroleum ether 1:10, v/v, silica gel plate);
20
D
27% ee, ½aŠ ¼ ꢀ10:7 (c 0.37, CHCl3, 27% ee), lit.34
4.4.3. Substitution reaction of (2S,3R)-3-phenyl-1-(4-tol-
uenesulfonyl)aziridin-2-ylmethyl tosylate 22 and sodium
phenolate for the preparation of (2S,3R)-2-phenyl-3-
phenoxylmethyl-1-(4-toluenesulfonyl)aziridine 16. Seven
milligrams (0.175mmol) of 60% NaH in mineral oil
was added to 1.5mL of dry DMF at 0 ꢁC. After
20
D
½aŠ ¼ þ35:7 (c 1.29, CHCl3, >98% ee) for (2R,3S)-
1
enantiomer; H NMR (200 MHz, CDCl3): d 0.095(s,
6H, 2CH3), 0.90 (s, 9H, 3CH3), 2.39 (s, 3H, CH3),
3.10 (1H, ddd, J = 4.5, 6.8, 6.9 Hz, 1H, CH), 3.89 (d,
J = 4.4 Hz, CH), 4.14 (dd, J = 7.1, 11.3 Hz, 1H in

L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
3733
10 min, to the above white suspension 14 mg
(0.149 mmol) of PhOH in 1 mL of dry DMF was added
dropwise at the same temperature. The resulting
mixture was stirred for 30 min to give the clear solution.
Then 65mg (0.142 mmol) of 22 in 1 mL of dry DMF was
added dropwise to the solution. The mixture was allowed
to stir overnight with warming to room temperature, and
diluted with 10 mL of water. The mixture was extracted
with EtOAc and the organic layer was washed with
brine, dried over anhydrous Na2SO4. Concentration
under reduced pressure gave an oily residue, which was
separated by flash chromatography over silica gel with
ethyl acetate/petroleum ether (1:10, v/v) to give colorless
oil 13 mg of 16, yield 24%; colorless oil 16 mg of 23, yield
30%; white solid 23 mg of 24, yield 34%, respectively.
128.3, 128.8, 129.3, 129.4, 129.5, 136.6, 137.5, 143.3,
157.9; MS (EI) m/z (rel. intensity, %): 473 (M+, 0.67),
380 (M+ꢀPhO, 4.25), 290 (23), 209 (14), 183 (52), 172
(94), 155 (Ts+, 50), 130 (36), 119 (41), 91 (CH3Ph+,
100), 77 (Ph+, 49). HRMS (EI) m/z calcd for
C28H27NO4S (M+) 473.1661; Found 473.1660.
Acknowledgments
This work was supported in part by the National Natu-
ral Science Foundation of China (Project No. 20272002
and 20472005), Ministry of Education of China (EYTP).
References
4.4.3.1. (2S,3R)-2-Phenyl-3-phenoxymethyl-1-(4-tolu-
enesulfonyl)aziridine 16 (from22). Yield 24%; 26% ee
determined by HPLC with Chiralcel OD column with
hexane/2-propanol (90:10, v/v) as an eluent at a flow
1. For recent reviews on the application of chiral bis(oxazo-
line)–metal complexes in catalytic asymmetric reactions,
see: (a) Pfaltz, A. Acc. Chem. Res. 1993, 26, 339–345; (b)
Ghosh, A. K.; Mathivanan, P.; Cappiello, J. Tetrahedron:
Asymmetry 1998, 9, 1–45; (c) McManus, H. A.; Guiry, P.
J. Chem. Rev. 2004, 104, 4151–4202.
rate of 0.8 mL/min (smajor = 20.9 min; sminor = 30.3 min);
20
½aŠ ¼ ꢀ7:1 (c 1.00, CHCl3, 26% ee).
D
2. Bedekar, A. V.; Koroleva, E. B.; Andersson, P. G. J. Org.
Chem. 1997, 62, 2518–2526.
4.4.3.2. (S,S)2-(1-Phenoxy-phenylmethyl)-1-(4-tolu-
enesulfonyl)aziridine 23. Colorless crystals, yield 30%;
mp 92.5–94 ꢁC; Rf = 0.12 (ethyl acetate/petroleum ether
1:10, v/v, silica gel plate); 19% ee determined by HPLC
with Chiralcel OD-H column with hexane/2-propanol
3. Asymmetric aziridination catalyzed by bis-oxazoline cop-
per complexes, see: (a) Evans, D. A.; Woerpel, K. A.;
Hinman, M. M.; Faul, M. M. J. Am. Chem. Soc. 1991,
113, 726–728; (b) Evans, D. A.; Faul, M. M.; Bilodeau, M.
T.; Anderson, B. A.; Barnes, D. M. J. Am. Chem. Soc.
1993, 115, 5328–5329.
(90:10, v/v) as an eluent at a flow rate of 0.8 mL/min
20
D
(sminor = 12.05min;
s
major = 19.59 min); ½aŠ ¼ þ10:7
´
4. Hatimi, A. E.; Gomez, M.; Jansat, S.; Muller, G.; Font-
(c 0.45, CHCl3, 19% ee); IR (KBr) (cmꢀ1): 1598 (s),
1494 (s), 1327 (s), 1234 (s), 1163 (s); 1H NMR
(300 MHz, CDCl3): 2.40 (d, J = 4.2 Hz, 1H in CH2N),
2.43 (s, 3H, CH3), 2.81 (d, J = 6.9 Hz, 1H in CH2N),
3.14 (ddd, J = 4.2, 6.0, 6.9 Hz, 1H, CHN), 4.87 (d,
J = 6.0 Hz, 1H, CHO), 6.70–6.73 (m, 2H, ArH), 6.85–
6.90 (m, 1H, ArH), 7.12–7.19 (m, 9H, ArH), 7.61–7.64
(m, 2H, ArH); 13C NMR (75.5 MHz, CDCl3): d 21.6,
31.0, 43.8, 78.8, 116.0, 121.4, 126.3, 127.9, 128.0,
128.5, 129.3, 129.5, 134.4, 137.7, 144.4; MS (EI) m/z
(rel. intensity, %): 379 (M+, 0.25), 286 (M+ꢀPhO, 35),
166 (13), 155 (Ts+, 27), 130 (100), 91 (CH3Ph+, 78), 77
(Ph+, 22). HRMS (EI) m/z calcd for C22H21NO3S
(M+) 379.1242; Found 379.1249.
´
Bardıa, M.; Solans, X. J. Chem. Soc., Dalton Trans. 1998,
4229–4236.
5. For a recent review on the synthesis and application of
pyridine-2,6-bis(oxazolines), see: Desimoni, G.; Faita, G.;
Quadrelli, P. Chem. Rev. 2003, 103, 3119–3154.
6. Kanemasa, S.; Oderaotoshi, Y.; Yamamoto, H.; Tanaka,
J.; Wada, E.; Curran, D. P. J. Org. Chem. 1997, 62, 6454–
6455.
7. Voituriez, A.; Schulz, E. Tetrahedron: Asymmetry 2003,
14, 339–346.
8. (a) Suzuki, T.; Kinoshita, K.; Kawada, H.; Nakada, M.
Synlett 2003, 570–572; (b) Inoue, M.; Suzuki, T.; Nakada,
M. J. Am. Chem. Soc. 2003, 125, 1140–1141.
9. Takacs, J. M.; Quincy, D. A.; Shay, W.; Jones, B. E.;
Ross, C. R., II. Tetrahedron: Asymmetry 1997, 8, 3079–
3087.
10. Du, D. M.; Fu, B.; Hua, W. T. Tetrahedron 2003, 59,
1933–1938.
11. Jiao, P.; Xu, J. X.; Zhang, Q. H.; Choi, M. C. K.; Chan,
A. S. C. Tetrahedron: Asymmetry 2001, 12, 3081–3088.
12. Lipkowitz, K. B.; Pradhan, M. J. Org. Chem. 2003, 68,
4648–4656.
4.4.3.3. (S,S)-1,3-Diphenoxy-1-phenyl-N-(4-toluene-
sulfonyl)propan-2-amine 24. Colorless crystals, yield
34%; mp 162–164 ꢁC; Rf = 0.08 (ethyl acetate/petroleum
ether 1:10, v/v, silica gel plate); 23% ee determined by
HPLC with Chiralcel OD-H column with hexane/2-pro-
panol (90:10, v/v) as an eluent at a flow rate of 0.8 mL/
13. For recent reviews on the asymmetric aziridination, see:
(a) Jacobsen, E. N. In Comprehensive Asymmetric Catal-
ysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.;
Springer: Berlin, 1999; Vol. 2, p 607; (b) Muller, P.; Fruit,
C. Chem. Rev. 2003, 103, 2905–2920; (c) Li, A. H.; Dai, L.
X.; Aggarwcal, A. K. Chem. Rev. 1997, 97, 2341–2372.
14. Asymmetric aziridination catalyzed by diimine copper
complexes, see: (a) Li, Z.; Cosner, K. R.; Jacobsen, E. N.
J. Am. Chem. Soc. 1993, 115, 5326–5327; (b) Li, Z.; Quan,
R. W.; Jacobsen, E. N. J. Am. Chem. Soc. 1995, 117,
5889–5890; (c) Quan, R. W.; Li, Z.; Jacobsen, E. N. J. Am.
Chem. Soc. 1996, 118, 8156–8157; (d) Gillespie, K. M.;
Sanerd, C. J.; OꢀShaughnessy, P.; Westmoreland, I.;
20
min (sminor = 12.01 min; smajor = 14.83 min); ½aŠ ¼ ꢀ5:8
D
(c 0.45, CHCl3); IR (KBr) (cmꢀ1): 1599 (s), 1494 (s),
1330 (m), 1235(s), 1161 (s); 1H NMR (300 MHz,
CDCl3): 2.40 (s, 3H, CH3), 3.90 (dd, J = 5.1, 9.9 Hz,
1H in CH2O), 4.06 (dddd, J = 4.5, 5.1, 5.7, 9.0 Hz,
1H, CHN), 4.23 (dd, J = 5.7, 9.9 Hz, 1H in CH2O),
5.00 (d, J = 9.0 Hz, 1H, CHO), 5.45 (d, J = 4.5Hz,
1H, NH), 6.63–6.68 (m, 4H, ArH), 6.84–6.95(m, 2H,
ArH), 7.10–7.29 (m, 11H, ArH), 7.69 (d, J = 8.4 Hz,
2H, ArH); 13C NMR (75.5 MHz, CDCl3): d 21.5, 57.6,
65.5, 78.9, 114.4, 115.7, 121.2, 121.3, 126.5, 127.2,

3734
L. Ma et al. / Tetrahedron: Asymmetry 16 (2005) 3718–3734
Thickitt, P.; Scott, C. P. J. Org. Chem. 2002, 67, 3450– 27. Xu, J. X.; Wei, T. Z.; Zhang, Q. H. J. Org. Chem. 2004,
3458; (e) Nishimura, M.; Minakata, S.; Takahashi, T.;
Oderaotoshi, Y.; Komatsu, M. J. Org. Chem. 2002, 67,
2101–2110.
69, 6860–6866.
28. Saito, T.; Sakairi, M.; Akiba, D. Tetrahedron Lett. 2001,
42, 5451–5454.
15. (a) Hansen, K. B.; Finnney, N. S.; Jacobsen, E. N. Angew.
Chem., Int. Ed. Engl. 1995, 34, 676–678; (b) Juhl, K.;
Hazell, R. G.; Jorgensen, K. A. J. Chem. Soc., Perkin.
Trans. 1 1999, 2293–2297.
16. Brandt, P.; Sodergren, M. J.; Andersson, P. G.; Norrby,
P.-O. J. Am. Chem. Soc. 2000, 122, 8013–8020.
17. (a) Aggarwal, A. K.; Templeton, J. L. J. Org. Chem. 1996,
61, 8368–8369; (b) Wang, D. K.; Dai, L. X.; Hou, X. L.
Chem. Commun. 1997, 1231–1232.
29. Alonso, D. A.; Andersson, P. G. J. Org. Chem. 1998, 63,
9455–9461.
30. Aggarwal, V. K.; Thompson, A.; Jones, R. V. H.;
Standen, M. C. H. J. Org. Chem. 1996, 61, 8368–8369.
31. Gillespie, K. M.; Crust, E. J.; Deeth, R. J.; Scott, P. Chem.
Commun. 2001, 785–786.
32. Wang, Y.; Ding, K. L. J. Org. Chem. 2001, 66, 3238–3241.
33. Geory, G. I.; Harriman, G. C. B.; Peterson, S. A. J. Org.
Chem. 1995, 60, 7366–7368.
18. Xu, J. X.; Ma, L. G.; Jiao, P. Chem. Commun. 2004, 1616–
1617.
19. Imai, Y.; Zhang, W. B.; Kida, T.; Nakatsuji, Y.; Ikeda, I.
Tetrahedron: Asymmetry 1996, 7, 2453–2462.
20. Rogers, M. E.; Averill, A. I. J. Org. Chem. 1986, 51, 3308–
3314.
21. (a) Waldmann, H.; Stengl, R. Chem. Ber. 1950, 83, 167–
170; (b) Beckett, A. H.; Walker, J. Tetrahedron 1963, 19,
545–556.
22. Andrus, M. B.; Asgari, D. Tetrahedron 2000, 56, 5775–
5780.
23. Muller, P.; Fruit, C. Chem. Rev. 2003, 103, 2905.
24. Suga, H.; Kakehi, A.; Ibata, T.; Fudo, T.; Watanabe, Y.;
Kinoshita, Y. Bull. Chem. Soc. Jpn. 2003, 76, 189–199.
25. Muller, P.; Bernardinelli, G.; Allenbach, Y. F.; Ferri, M.;
Flack, H. D. Org. Lett. 2004, 6, 1725–1728.
26. Isaacs, N. Physical Organic Chemistry, 2nd ed.; Addison
Wesley Longman: Essex, England, 1995; p 152.
34. Fujii, N.; Nakai, K.; Habashita, H.; Hotta, Y.; Tamam-
ura, H.; Otaka, A.; Ibuka, T. Chem. Pharm. Bull. 1994, 42,
2241–2250.
35. Olah, G. A.; Gupta, B. G. B.; Narang, S. C.; Malhotra, R.
J. Org. Chem. 1979, 44, 4272–4275.
36. Klunder, J. M.; Onami, T.; Sharpless, K. B. J. Org. Chem.
1989, 54, 1295–1304.
37. For a recent review on ring-opening reactions of unsym-
metric aziridines with nucleophiles, see: Ma, L. G.; Xu, J.
X. Prog. Chem. (Huaxue Jinzhan) 2004, 16, 220–235.
38. For a recent review on the p stacking interaction in
asymmetric synthesis, see: Jones, G. B.; Chapman, B.
Synthesis 1995, 475–497.
39. Dauban, P.; Dodd, P. R. H. Org. Lett. 2000, 2, 2327–2329.
40. Ma, L. G.; Du, D. M.; Xu, J. X. J. Org. Chem. 2005, 70,
10155–10158.
41. Ali, S. L.; Nikalje, M. D.; Sudalai, A. Org. Lett. 1999, 1,
705–707.