Metalloporphyrin-mediated asymmetric nitrogen-atom transfer to hydrocarbons: aziridination of alkenes and amidation of saturated C-H bonds catalyzed by chiral ruthenium and manganese porphyrins.

Article abstract of DOI:10.1002/1521-3765(20020402)8:7<1563::AID-CHEM1563>3.0.CO;2-V

Chiral metalloporphyrins [Mn(Por*)(OH)(MeOH)] (1) and [Ru(Por*)(CO)(EtOH)] (2) catalyze asymmetric aziridination of aromatic alkenes and asymmetric amidation of benzylic hydrocarbons to give moderate enantiomeric excesses. The mass balance in these nitrog

Full text of DOI:10.1002/1521-3765(20020402)8:7<1563::AID-CHEM1563>3.0.CO;2-V

FULL PAPER
Metalloporphyrin-Mediated Asymmetric Nitrogen-Atom Transfer to
À
Hydrocarbons: Aziridination of Alkenes and Amidation of Saturated C H
Bonds Catalyzed by Chiral Ruthenium and Manganese Porphyrins
Jiang-Lin Liang, Jie-Sheng Huang, Xiao-Qi Yu, Nianyong Zhu, and Chi-Ming Che*[a]
Abstract: Chiral
metalloporphyrins
tetralin, and 2-ethylnaphthalene with
™PhI(OAc)2 ‡ NH2SO2Me∫, affording
the N-substituted methanesulfonamides
in up to 56% ee with substrate conver-
sions of up to 34% and amide selectiv-
ities of up to 92%. Extension of the
porphyrin aziridine adduct, [Ru(Por*)-
(CO)(TsAz)] (4, TsAz ˆ N-tosyl-2-
(4-chlorophenyl)aziridine), have been
isolated from the reaction of 2 with
[Mn(Por*)(OH)(MeOH)] (1) and [Ru-
(Por*)(CO)(EtOH)] (2) catalyze asym-
metric aziridination of aromatic alkenes
and asymmetric amidation of benzylic
hydrocarbons to give moderate enantio-
meric excesses. The mass balance in
these nitrogen-atom-transfer processes
ˆ
PhI NTs and N-tosyl-2-(4-chlorophen-
yl)aziridine, respectively. The imidoru-
thenium porphyrin 3 could be an active
species in the aziridination or amidation
catalyzed by complex 2 described above.
The second-order rate constants for the
reactions of 3 with styrenes, 2-vinyl-
naphthalene, indene, ethylbenzenes, and
2-ethylnaphthalene range from 3.7
42.5 Â 10À3 dm3 molÀ1 sÀ1. An X-ray
structure determination of complex 4
reveals an O- rather than N-coordina-
tion of the aziridine axial ligand. The
fact that the N-tosylaziridine in 4 does
not adopt an N-coordination mode dis-
favors a concerted pathway in the aziri-
dination by a tosylimido ruthenium
porphyrin active species.
ˆ
™complex 1 ‡ PhI NTs∫ or ™complex
1 ‡ PhI(OAc)2 ‡ NH2R (R ˆ Ts, Ns)∫
amidation protocol to a steroid resulted
in diastereoselective amidation of cho-
ˆ
has been examined. With PhI NTs as
the nitrogen source, the aziridination of
styrenes, trans-stilbene, 2-vinylnaphtha-
lene, indene, and 2,2-dimethylchromene
catalyzed by complex 1 or 2 resulted in
up to 99% substrate conversions and up
to 94% aziridine selectivities, whereas
the amidation of ethylbenzenes, indan,
tetralin, 1-, and 2-ethylnaphthalene cat-
alyzed by complex 2 led to substrate
conversions of up to 32% and amide
selectivities of up to 91%. Complex 1 or
2 can also catalyze the asymmetric
amidation of 4-methoxyethylbenzene,
À
lesteryl acetate at the allylic C H bonds
at C-7 with substrate conversions of up
to 49% and amide selectivities of up to
90% (a:b ratio: up to 4.2:1). An aziridi-
nation- and amidation-active chiral bis-
(tosylimido)ruthenium(vi)
porphyrin,
[Ru(Por*)(NTs)2] (3), and a ruthenium
Keywords: amidation ¥ asymmetric
catalysis ¥ aziridination ¥ manganese
¥ porphyrinoids ¥ ruthenium
active species, respectively (Scheme 1). Despite the current
absence of practical metalloporphyrin catalysts, metallopor-
phyrin-catalyzed hydrocarbon-functionalization reactions re-
main attractive not only because of their unique relationship
to heme-containing enzymes but also because of their
unusually high selectivity and catalyst turnover numbers
observed in many cases. To extend the utility of metal-
loporphyrin catalysts to asymmetric synthesis, a wide variety
of chiral metalloporphyrins have been developed[1b] and their
catalytic behavior toward asymmetric hydrocarbon-function-
alization reactions by oxygen- and carbon-atom-transfer
Introduction
Metalloporphyrin-mediated hydrocarbon functionalization is
of particular importance in biomimetic investigation and is
potentially useful in organic synthesis. The epoxidation,[1]
aziridination,[2] and cyclopropanation[3] of alkenes and the
hydroxylation[4] and amidation[2e, 5] of saturated C H bonds
À
are among the most common hydrocarbon-functionalization
reactions mediated by a metalloporphyrin, which are widely
believed to proceed by oxygen-, nitrogen-, and carbon-atom
transfer from oxo-, imido-, and carbene-metalloporphyrin
j, 3b,d,j l,n,p, 4c,g]
processes has been extensively investigated.[1a
However, little is known about the catalytic properties of a
chiral metalloporphyrin for asymmetric hydrocarbon-func-
tionalization reactions by nitrogen-atom-transfer processes.
Recently, we initiated exploration of chiral metallopor-
phyrins as catalysts for asymmetric nitrogen-atom-transfer
reactions and preliminarily reported the aziridination of
[a] Prof. C.-M. Che, J.-L. Liang, Dr. J.-S. Huang, Dr. X.-Q. Yu, Dr. N. Zhu
Department of Chemistry and Open Laboratory of Chemical Biology
of the Institute of Molecular Technology for Drug Discovery and
Synthesis, The University of Hong Kong, Pokfulam Road, Hong Kong
(China)
Fax : (‡852)2857-1586
E-mail: cmche@hku.hk
Chem. Eur. J. 2002, 8, No. 7
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1563 $ 20.00+.50/0
1563

C.-M. Che et al.
FULL PAPER
exceedingly rare examples of asymmetric amidation with
iminoiodanes catalyzed by a metal complex. The only
previous example is the rhodium-catalyzed asymmetric ami-
ˆ
dation of hydrocarbons with PhI NNs reported by M¸ller and
co-workers,[10] which afforded amides in up to 31% ee.[11]
Herein, we report the first examples of 1) asymmetric
aziridination of alkenes catalyzed by a chiral ruthenium
À
porphyrin, 2) asymmetric amidation of saturated C H bonds
with ™PhI(OAc)2 ‡ NH2R∫ catalyzed by a metal complex,
ˆ
and 3) amidation of a cholesteryl ester with either PhI NTs or
™PhI(OAc)2 ‡ NH2R∫ catalyzed by a metalloporphyrin,
together with a mass balance study on the 1- or 2-mediated
aziridination and amidation reactions. To help understand the
mechanism in the ruthenium-catalyzed asymmetric nitrogen-
atom-transfer reactions, a chiral bis(tosylimido)ruthenium(vi)
porphyrin, [Ru(Por*)(NTs)2] (3), and a ruthenium porphyrin
aziridine adduct, [Ru(Por*)(CO)(TsAz)] (4, TsAz ˆ N-tosyl-
2-(4-chlorophenyl)aziridine), were isolated; the former is the
only example of an isolated chiral metalloporphyrin bearing
an imido axial ligand and the latter contains an O-bound N-
tosylaziridine. Kinetic studies were carried out to measure the
rate constants for the stoichiometric reactions of 3 with a
series of hydrocarbons; the results were compared with those
obtained for bis(tosylimido)ruthenium complexes of simple
porphyrins[2e] to examine the effect of porphyrin ligands on
the ruthenium-porphyrin-mediated nitrogen-atom-transfer
reactions.
Scheme 1. Common metalloporphyrin-mediated hydrocarbon-functional-
ization reactions by oxygen-, nitrogen-, and carbon-atom transfer from
putative metal-oxo, -imido, and -carbene active species (X ˆ O, NSO2R,
and CRR', respectively).
ˆ
aromatic alkenes with iminoiodane PhI NTs in the presence
of chiral manganese porphyrin [Mn(Por*)(OH)(MeOH)] (1)
(which afforded N-tosylaziridines in up to 76% yield with up
to 68% ee)[6] and the PhI NTs amidation of benzylic hydro-
ˆ
carbons in the presence of 1 or chiral ruthenium porphyrin
[Ru(Por*)(CO)(EtOH)] (2) (which afforded N-substituted
toluenesulfonamides in up to 54% ee).[7] These, together with
Results
Synthesis and characterization of chiral ruthenium porphyrins
ˆ
3 and 4: Treatment of 2 with 4 equiv of PhI NTs in dichloro-
methane rapidly produced the bis(tosylimido)ruthenium(vi)
porphyrin 3 (reaction (1) in Scheme 2), which was isolated in
64% yield after chromatography on alumina. Note that in our
preliminary work[7] complex 3 was generated in situ and not
1
isolated in pure form. The main H NMR, UV/Vis, and IR
spectral features of 3 resemble those of bis(tosylimido)ruthe-
nium(vi) complexes of other meso-tetraarylporphyrins.[2e]
CO
CO
p-Cl-C6H4CH CH2
NTs
*
*
5
Ru
Ru
*
*
*
*
(2)
*
*
EtOH
O
S
O
ˆ
subsequent workby Marchon and co-workers on the PhI NR
2
N
(R ˆ Ts or SO2C6H4OMe) aziridination of styrene catalyzed
by manganese or iron tetramethylchiroporphyrins (which
afforded N-tosylaziridines in up to 34% yield with up to
57% ee),[8] are so far the only reports in the literature
concerning asymmetric nitrogen-atom-transfer reactions
mediated by a metalloporphyrin. Both 1 and 2 consist of a
sterically demanding D4-symmetric chiral porphyrin ligand
5,10,15,20-tetrakis-{(1S,4R,5R,8S)-1,2,3,4,5,6,7,8-octahydro-
1,4:5,8-dimethanoanthracene-9-yl}porphyrin [H2(Por*)] first
synthesized by Halterman and Jan.[9] The amidations of
benzylic hydrocarbons[7] catalyzed by complex 1 or 2 are
Cl
PhI=NTs
(1)
4
NTs
*
*
Ru
*
*
=
*
Por*
*
*
*
NTs
3
Scheme 2. Synthesis of chiral ruthenium porphyrins 3 and 4.
1564
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 0947-6539/02/0807-1564 $ 20.00+.50/0 Chem. Eur. J. 2002, 8, No. 7

Metalloporphyrin-Mediated Asymmetric N-Atom
1563 1572
Table 1. Crystal data and structure refinement for complex 4.
The ruthenium porphyrin 4 was obtained in 85% yield by
treatment of 2 with excess N-tosyl-2-(4-chlorophenyl)aziri-
dine (5) in chloroform (reaction (2) in Scheme 2). The O-
rather than N-coordination mode in 4 was established by
X-ray crystallography (see below). To our knowledge, this is
the first metalloporphyrin bearing an aziridine axial ligand.[12]
As expected, the key spectral features of 4 (including the
Soret and b bands in the UV/Vis spectrum, the n(CO) band in
the IR spectrum, and the proton resonances of the porphyrin
formula
MR
l [ä]
T [K]
crystal system
space group
a [ä]
b [ä]
c [ä]
a [8]
b [8]
C100H90N5ClO3RuS
1578.35
0.71073
301
triclinic
P1
10.801(2)
14.220(3)
15.294(3)
78.90(3)
77.39(3)
84.85(3)
2246.7(8)
1
1
ligand in the H NMR spectrum) are similar to those of 2.
Somewhat surprising is that the 1H NMR signals of the
coordinated 5 in 4 in CD2Cl2 are very similar to those of free 5
(Dd ꢀ 0.03 ppm) in the same deuterated solvent. This suggests
a liberation of 5 from 4 in the solution.
g [8]
V [ä3]
Z
1calcd [MgmÀ3
]
1.167
m(MoKa) [mmÀ1
F(000)
]
0.278
826
X-ray crystal structure of ruthenium porphyrin aziridine
adduct 4: Figure 1 depicts the structure of 4 with the key
atom-numbering scheme. The crystal data and structure
index ranges
À 12 ꢀ h ꢀ 11, À17 ꢀ k ꢀ 16, À18 ꢀ l ꢀ 18
reflections collected
independent reflections
refinement method
parameters
13110
11070
full-matrix least-squares on F 2
985
goodness-of-fit
1.059
final R indices
absolute structure parameter
largest diff. peak:hole [e äÀ3
R1 ˆ 0.076, wR2 ˆ 0.202
0.05(4)
]
1.31/ À 0.59
Table 2. Selected bond lengths [ä] and bond angles [8] for complex 4.
bond lengths
À
À
Ru(1) C(85)
1.76(1)
2.008(9)
2.07(1)
1.18(1)
1.50(1)
1.75(1)
1.46(2)
Ru(1) O(2)
2.256(8)
2.074(9)
2.045(9)
1.435(8)
1.59(1)
1.63(2)
1.57(2)
À
À
Ru(1) N(1)
Ru(1) N(2)
À
À
Ru(1) N(3)
Ru(1) N(4)
À
À
C(85) O(1)
S(1) O(2)
À
À
S(1) O(3)
S(1) N(5)
À
À
S(1) C(86)
C(93) C(94)
À
À
N(5) C(93)
N(5) C(94)
bond angles
Ru(1)-C(85)-O(1)
N(1)-Ru(1)-N(2)
N(3)-Ru(1)-N(4)
N(1)-Ru(1)-N(3)
C(85)-Ru(1)-N(1)
C(85)-Ru(1)-N(3)
O(2)-Ru(1)-N(1)
O(2)-Ru(1)-N(3)
O(2)-S(1)-C(86)
O(2)-S(1)-O(3)
179(1)
Ru(1)-O(2)-S(1)
N(2)-Ru(1)-N(3)
N(1)-Ru(1)-N(4)
N(2)-Ru(1)-N(4)
C(85)-Ru(1)-N(2)
C(85)-Ru(1)-N(4)
O(2)-Ru(1)-N(2)
O(2)-Ru(1)-N(4)
O(2)-S(1)-N(5)
148.0(6)
88.9(3)
90.2(4)
173.9(4)
92.5(5)
93.6(5)
86.8(3)
87.1(3)
108.2(6)
120.4(6)
113.7(9)
61.0(9)
90.2(4)
90.2(3)
175.2(4)
92.9(5)
91.9(4)
84.2(3)
91.0(3)
112.8(6)
118.3(9)
64.7(9)
54.2(8)
S(1)-N(5)-C(93)
C(93)-N(5)-C(94)
N(5)-C(94)-C(93)
S(1)-N(5)-C(94)
N(5)-C(93)-C(94)
As shown in Figure 1, the aziridine axial ligand 5 in 4 is
coordinated to ruthenium by an oxygen atom (O(2)) of the
tosyl group rather than by the nitrogen atom of the aziridine
Figure 1. X-ray crystal structure of [Ru(Por*)(CO)(TsAz)] (4) with the
À
key atom-numbering scheme viewed along the a) N(3) N(1) axis and
À
b) O(2) Ru(1) axis. Hydrogen atoms are not shown and the atoms of the
aziridine axial ligand in b) are filled with grey colors.
À
ring; the Ru O(2) bond length of 2.256(8) ä is comparable to
the Ru O(EtOH) bond length of 2.241(1) ä in 2.[3k] Both the
À
p-chlorophenyl and p-tolyl groups of 5 lie almost halfway
between adjacent norbornane moieties of the porphyrin
ligand (see Figure 1b). Compared to structurally character-
ized free aziridines,[13, 14] the coordinated 5 in 4 has an
aziridine ring with some unusual metrical parameters (for
refinements for 4 are compiled in Table 1; selected bond
lengths and angles for the complex are given in Table 2. This
ruthenium porphyrin, like complex 2,[3k] has a distorted
À
octahedral coordination geometry and has an axial Ru CO
À
moiety featuring an Ru C(CO) bond length of ꢁ1.76 ä and
À
À À
example the C(93) C(94) bond of 1.63(2) ä is rather long).
Ru C O angle of ꢁ1798. The porphyrin plane shows a slight
This may result from a disorder of the atoms in the three-
membered aziridine ring.
saddle distortion and the ruthenium atom is slightly out of the
mean porphyrin plane toward the CO axial group.
Chem. Eur. J. 2002, 8, No. 7
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1565 $ 20.00+.50/0
1565

C.-M. Che et al.
FULL PAPER
ˆ
Table 3. Catalytic asymmetric PhI NTs aziridination of aromatic alkenes by [Mn(Por*)-
(OH)(MeOH)] (1) or [Ru(Por*)(CO)(EtOH)] (2) and stoichiometric asymmetric aziridi-
nation of aromatic alkenes by [Ru(Por*)(NTs)2] (3).
Asymmetric aziridination of aromatic alkenes:
Mass balance in the aziridination catalyzed by
complex 1 or 2: We examined the reactions of a
X
H
PhCH
R
11
CHR
ˆ
series of aromatic alkenes with PhI NTs in the
6
7
8
9
4-Me
3-Cl
4-Cl
presence of ꢁ1.3 mol% complex 1 or 2. The
reactions were generally performed in dichloro-
methane at 408C for 2 h by employing excess
Me (cis)
12 Me (trans)
13 Ph (trans)
O
X
14
15
16
26
10 2-Br
TsN
NTs
NTs
NTs
17
ˆ
ˆ
X
H
PhI NTs (catalyst:substrate:(PhI NTs) molar
ratio ˆ 1:75:150) to facilitate assessment of the
mass balance in the catalytic processes. Table 3
shows the results obtained for the aziridination
of styrene (6), 4-methylstyrene (7), 3-chlorostyr-
ene (8), 4-chlorostyrene (9), 2-bromostyrene
(10), cis-b-methylstyene (11), trans-b-methyl-
styene (12), trans-stilbene (13), 2-vinylnaphtha-
lene (14), indene (15), and 2,2-dimethylchro-
mene (16). Under the conditions described
above, aziridination of these alkene substrates
afforded the products 17 19, 5, 20 26 (see
Table 3) in 1 57% ee with substrate conver-
sions of 41 99% (entries 1 17). The observed
aziridine selectivities (that is, the chemoselectiv-
ities defined as the yields of aziridines based on
the substrates consumed) range from 72 94%.
We also performed the aziridination of styrene
catalyzed by complex 1 under a very low catalyst
NTs
Ph
R
R
18 4-Me
19 3-Cl
O
21 Me (cis)
22 Me (trans)
23 Ph (trans)
5
4-Cl
X
24
25
20 2-Br
Entry
Substrate
Product
Catalyst
Conversion [%]
Yield [%][a]
ee [%][b]
catalytic aziridination[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
6
6
7
7
8
8
9
9
10
10
11
12
13
14
14
15
16
17
17
18
18
19
19
5
1
2
1
2
1
2
1
2
1
2
99
71
99
80
99
45
99
61
97
41
76
88
81
99
91
99
99
78
84
76
82
91
91
93
83
75
82
78[d]
81
72
94
78
73
73
47
21
44
21
40
23
50
14
57
29
7[e]
11
1
5
20
20
21 ‡ 22
22
23
24
24
25
26
1
2
1
2
2
1
56
14
11
5
ˆ
loading of 0.05 mol% (1:PhI NTs:styrene mo-
stoichiometric aziridination by 3[f]
lar ratio ˆ 1:2000:5000), which led to formation
18
19
20
21
22
23
24
6
7
8
17
18
19
5
20
24
25
78
84
90
83
61
91
72
27
29
29
28
48
43
25
of aziridine 17 in 58% yield (based on the
ˆ
amount of starting PhI NTs) and 42% ee with
1160 catalyst turnovers.
9
10
14
15
Stoichiometric aziridination by complex 3: The
reactions of 3 with styrenes 6 10, naphthalene
derivative 14, and indene (15) were investigated.
These reactions were carried out by employing
excess alkene substrates to mitigate the auto-
degradation of 3 in solution. The results ob-
tained for the reactions performed in dichloro-
methane at 408C for 2 h are listed in entries 18
24 in Table 3.
[a] Yield of isolated product based on the amount of the substrate consumed (catalytic
aziridination) or on the amount of complex used (stoichiometric aziridination).
[b] Determined by HPLC with a chiral whelk-O1 column. [c] Reaction conditions: CH2Cl2,
3
ˆ
408C, 2 h; catalyst:substrate:(PhI NTs) molar ratio ˆ 1:75:150. [d] Ratio of 21:22 ˆ 9:91.
[e] For the trans-aziridine 22. [f] Reaction conditions: CH2Cl2, 408C, 2 h.
ˆ
Table 4. Asymmetric PhI NTs amidation of benzylic hydrocarbons catalyzed by
[Ru(Por*)(CO)(EtOH)] (2).[a]
À
Asymmetric amidation of saturated C H bonds:
Mass balance in the amidation catalyzed by
X
H
MeO
X
27
28
ˆ
complex 2 with PhI NTs: A series of benzylic
29
30
31
37
32
38
hydrocarbons, including ethylbenzene (27),
4-methoxyethylbenzene (28), indan (29), tetralin
(30), 2-ethylnaphthalene (31), and 1-ethylnaph-
NHTs
NHTs
NHTs
NHTs
NHTs
X
H
MeO
X
33
34
ˆ
thalene (32), were treated with excess PhI NTs
35
36
in the presence of ꢁ1.3 mol% complex 2. When
Entry
Substrate
Product
Conversion [%]
Yield [%][b]
ee [%][c]
the reactions were conducted in dichlorome-
ˆ
1
2
3
4
5
6
27
28
29
30
31
32
33
34
35
36
37
38
14
29
32
22
23
32
84
82
91
85
84
78
42
22
3
28
47
25
thane at 408C for 2 h at a 2:substrate:PhI NTs
molar ratio of 1:75:150, the N-tosylamides 33
38 as shown in Table 4 were formed in 3
47% ee with substrate conversions of 14 32%.
The amide selectivities in these reactions (that is,
the chemoselectivities defined as the yields of
amides based on the substrates consumed) range
from 78 to 91%.
ˆ
[a] Reaction conditions: CH2Cl2, 408C, 2 h; catalyst:substrate:(PhI NTs) molar ratio ˆ
1:75:150. [b] Yield of isolated product based on the amount of the substrate consumed.
[c] Determined by HPLC with a chiral AD column.
1566
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1566 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 7

Metalloporphyrin-Mediated Asymmetric N-Atom
1563 1572
Amidation catalyzed by complex 1 or 2 with ™PhI(OAc)2 ‡
NH2SO2Me∫: Reaction of 4-methoxyethylbenzene (28), tet-
ralin (30), and 2-ethylnaphthalene (31) with the commercially
available reagents PhI(OAc)2 and NH2SO2Me in dichloro-
methane containing 1 mol% complex 1 or 2 at 408C for 2 h
produced the N-substituted methanesulfonamide 39 41 in
40 56% ee with amide selectivities of 84 92% (Table 5).
with an 82% chemoselectivity (defined as the yield of 43
based on the 42 consumed), as shown in entry 1 of Table 6.
The amidation is a-selective,[15] with an a:b ratio of 4.2:1. To
examine the effect of porphyrin ligand on the diastereoselec-
ˆ
tivity in the amidation reactions, we treated 42 with PhI NTs
in the presence of other manganese porphyrins 44 47
depicted in Table 6 under the conditions identical to those
employed for complex 1. The a:b ratios observed for 44 47
range from 1.7:1 to 3.2:1 (entries 2 5 in Table 6), all lower
than that observed for complex 1 (entry 1). Lastly, we
investigated the amidation of 42 catalyzed by complex 1
with commercially available reagents PhI(OAc)2 and NH2R
(R ˆ Ts, Ns). These catalytic reactions were conducted in
dichloromethane (entries 6 and 7 in Table 6) and in
acetonitrile (entry 8 in Table 6) at a 1:42:PhI(OAc)2:NH2R
molar ratio of 1:100:125:150. The amides 43 and 48 (see
Table 6) were again formed as mixtures of a and b isomers
with a:b ratios of 3.6:1 (43), 1.8:1 (48, in dichloromethane),
and 1.2:1 (48, in acetonitrile) (entries 6 8 in Table 6). The
a:b ratio of 3.6:1 and the amide selectivity of 81% obtained
Table 5. Asymmetric amidation of benzylic hydrocarbons with ™PhI(OAc)2 ‡
NH2SO2Me∫ catalyzed by [Mn(Por*)(OH)(MeOH)] (1) or [Ru(Por*)(CO)-
(EtOH)] (2).[a]
NHSO2Me
NHSO2Me
NHSO2Me
MeO
39
40
41
Entry Substrate Product Catalyst Conversion [%] Yield [%][b] ee [%][c]
1
2
3
4
28
30
31
31
39
40
41
41
1
1
1
2
26
22
21
34
84
84
92
90
50
46
56
40
for the ™PhI(OAc)2
comparable to those obtained for the corresponding
‡
NH2Ts∫ amidation of 42 are
[a] Reaction conditions: CH2Cl2, 408C, 2 h; catalyst:substrate: PhI(OAc)2:
NH2R molar ratio ˆ 1:100:125:150. [b] Yield of isolated product based on the
amount of the substrate consumed. [c] Determined by HPLC with a chiral OD
column.
ˆ
PhI NTs amidation although the former amidation system
gave rise to a lower substrate conversion (compare entries 1
and 6 in Table 6).
Amidation of a cholesteryl derivative catalyzed by complex 1
ˆ
Kinetic studies of the reactions between complex 3 and
hydrocarbons: We determined the second-order rate con-
stants (k2) for the reactions of 3 with a variety of aromatic
alkenes including styrene and para-substituted styrenes
with PhI NTs or ™PhI(OAc)2 ‡ NH2R (R ˆ Ts, Ns)∫: The
catalytic amidation of a cholesteryl derivative was explored by
employing complex 1 as the catalyst. We first studied the
reaction of cholesteryl acetate (42) with PhI NTs in dichloro-
ˆ
À
ˆ
4-X C6H4CH CH2 (X ˆ H: 6, Me: 7, F: 49, Cl: 9, CF3: 50),
methane at 408C in the presence of ꢁ1.3 mol% complex 1,
2-bromostyrene (10), 2-vinylnaphthalene (14), and indene
which afforded the amide 43 as a mixture of a and b isomers
ˆ
Table 6. Amidation of cholesteryl acetate (42) with PhI NTs or ™PhI(OAc)2 ‡ NH2R (R ˆ Ts, Ns)∫ catalyzed by [Mn(Por*)(OH)(MeOH)] (1) and other
manganese porphyrins 44 47.[a]
Entry
Nitrogen source
Product
Catalyst
Conversion [%]
Yield [%][b]
Ratio of a:b[c]
ˆ
1
2
3
4
5
6
7
8[d]
PhI NTs
a-43 ‡ b-43
a-43 ‡ b-43
a-43 ‡ b-43
a-43 ‡ b-43
a-43 ‡ b-43
a-43 ‡ b-43
a-48 ‡ b-48
a-48 ‡ b-48
1
44
45
46
47
1
49
43
35
29
24
27
12
21
82
74
92
90
88
81
87
90
4.2:1
1.7:1
1.9:1
1.8:1
3.2:1
3.6:1
1.8:1
1.2:1
ˆ
PhI NTs
ˆ
PhI NTs
ˆ
PhI NTs
ˆ
PhI NTs
PhI(OAc)2 ‡ NH2Ts
PhI(OAc)2 ‡ NH2Ns
PhI(OAc)2 ‡ NH2Ns
1
1
ˆ
[a] Reaction conditions: CH2Cl2, 408C, 2 h; catalyst:substrate:(PhI NTs) molar ratio ˆ 1:75:150; catalyst:substrate:PhI(OAc)2:NH2R molar ratio ˆ
1
1:100:125:150. [b] Yield of isolated product based on the amount of 42 consumed. [c] Determined by H NMR spectroscopy. [d] In MeCN.
Chem. Eur. J. 2002, 8, No. 7
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1567 $ 20.00+.50/0
1567

C.-M. Che et al.
FULL PAPER
ˆ
(15), and benzylic hydrocarbons including ethylbenzene and
with PhI NTs as the nitrogen source. The reactions, except
the amidation of 27, were all conducted using excess substrates
(ꢂ20-fold excess for aziridination and fivefold excess for
amidation), which hampered the determination of substrate
conversions and aziridine or amide selectivities in the
catalysis. The present workshows that, with a twofold excess
À
para-substituted ethylbenzenes 4-X C6H4CH2CH3 (X ˆ H:
27, MeO: 28, Me: 51, F: 52, Cl: 53, Br: 54) and 2-ethyl-
naphthalene (31). The k2 values obtained in dichloromethane
containing pyrazole (2% w/w) at 298 K are compiled in
Table 7. We previously reported the k2 values for the reactions
of [Ru(tpp)(NTs)2] (tpp ˆ meso-tetraphenylporphyrinato di-
anion) with some of the foregoing hydrocarbons under the
same conditions;[2e] those k2 values are also shown in Table 7
for comparison.
ˆ
PhI NTs, 1-catalyzed aziridinations of 6 10, 14, and 16, and
2-catalyzed aziridination of 16 resulted in substrate conver-
sions close to 100% (see Table 3). The substrate conversions
in 2-catalyzed amidations of 27 32 are significantly lower
(compare Table 3 and Table 4). For the hydrocarbons 6 16
and 27 32, the aziridine or amide selectivities are good-to-
excellent, with up to 57% ee observed for the aziridinations
and up to 47% ee for the amidations.[17] The 1160 catalyst
turnovers obtained for 1-catalyzed aziridination of styrene is
fairly high, which, to our knowledge, is the highest turnover
number ever reported for a metal-complex-catalyzed asym-
metric nitrogen-atom-transfer reaction.
Table 7. Second-order rate constants (k2) for nitrogen-atom transfer from
[Ru(Por*)(NTs)2] (3) to hydrocarbons in dichloromethane containing pyrazole
(2% w/w) at 298 K. The k2 values for nitrogen-atom transfer from
[Ru(tpp)(NTs)2] to the same hydrocarbons (from ref. [2e]) are also included
for comparison.
Entry Hydrocarbon
103k2 [dm3 molÀ1 sÀ1
[Ru(Por*)(NTs)2] (3) [Ru(tpp)(NTs)2]
]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
styrene (6)
13.7 Æ 0.5
35.6 Æ 1.4
18.7 Æ 0.5
19.9 Æ 0.3
9.0 Æ 0.1
16.0 Æ 0.8
11 Æ 1
4-methylstyrene (7)
4-fluorostyrene (49)
4-chlorostyrene (9)
The present workalso reveals that complex 1 can catalyze
ˆ
the PhI NTs amidation of steroid 42 in a diastereoselective
8.8 Æ 0.1
3.2 Æ 0.2
manner (a:b ˆ 4.2:1).[18] Manganese complexes 44 47 bear-
4-trifluoromethylstyrene (50) 8.1 Æ 0.3
ˆ
ing nonchiral porphyrin ligands can catalyze the PhI NTs
2-bromostyrene (10)
2-vinylnaphthalene (14)
indene (15)
9.9 Æ 0.5
42.5 Æ 0.5
40.3 Æ 0.6
3.92 Æ 0.06
amidation of 42 as well, but all afforded a lower diastereo-
selectivity. These manganese-porphyrin-catalyzed amidation
reactions are highly regio- and chemoselective: only the
amidation of the allylic C H bonds at C-7 (see Table 6) was
observed and the amide 43 was formed with up to 92%
chemoselectivity.
It is remarkable that the ™PhI(OAc)2 ‡ NH2R∫ amidation
protocol we developed previously by using nonchiral ruthe-
nium catalysts[5f, 19] is also effective for the asymmetric
amidations.[20] The amidation of 4-methoxyethylbenzene
(28), tetralin (30), and 2-ethylnaphthalene (31) with
™PhI(OAc)2 ‡ NH2SO2Me∫ catalyzed by complex 1 or 2
afforded the corresponding N-substituted methanesulfona-
mides 39 41 in 40 56% ee (Table 5), with substrate con-
versions of 21 34% comparable to those given in Table 4 for
ethylbenzene (27)
3.60 Æ 0.07
16.5 Æ 0.3
6.8 Æ 0.1
4-methoxyethylbenzene (28) 18.9 Æ 0.7
À
4-methylethylbenzene (51)
4-fluoroethylbenzene (52)
4-chloroethylbenzene (53)
4-bromoethylbenzene (54)
2-ethylnaphthalene (31)
6.1 Æ 0.1
5.9 Æ 0.4
3.7 Æ 0.1
6.6 Æ 0.6
21.0 Æ 0.5
4.3 Æ 0.1
5.0 Æ 0.1
Discussion
Metalloporphyrin-mediated nitrogen-atom transfer to hydro-
carbons: Metal-mediated nitrogen-atom transfer to hydro-
carbons constitutes an appealing route to aziridines and
amides/amines. The aziridination of alkenes and amidation of
ˆ
the PhI NTs amidations. This is so far the only example of a
À
saturated C H bonds with iminoiodanes catalyzed by a
metal-complex-catalyzed asymmetric nitrogen-atom transfer
directly with ™PhI(OAc)2 ‡ NH2R∫ as the nitrogen source,[21]
which realized an enantioselective formation of N-substituted
methanesulfonamides from direct amidation of saturated
metalloporphyrin were first reported by the groups of
Mansuy[2b] and Breslow,[5a] respectively, in the early 1980s.
Since then, a number of iminoiodane aziridinations/amida-
tions of hydrocarbons catalyzed by iron, manganese, and
À
C H bonds catalyzed by a metal complex.
e]
ruthenium porphyrins have been studied.[2c, 5b Surprisingly,
the catalyst turnover numbers in these reactions are usually
less than 50, not significantly higher than those observed for
the nitrogen-atom transfer catalyzed by non-porphyrin metal
Mechanistic aspects of 2-catalyzed nitrogen-atom transfer to
hydrocarbons: Despite a wide proposal of metal imido
species as the active intermediates in metal-complex-cata-
lyzed nitrogen-atom-transfer reactions with iminoiodanes,
isolation or direct observation of such intermediates remains a
serious challenge in most cases. For example, no imidomanga-
nese porphyrins have been detected in the reactions of
iminoiodanes with either complex 1 or any other manganese
porphyrins. The successful isolation of aziridination- and
amidation-active tosylimido ruthenium porphyrin [Ru-
(Por*)(NTs)2] (3) provides a useful insight into the mecha-
nism of asymmetric nitrogen-atom-transfer reactions cata-
lyzed by complex 2.
complexes.[16] Recently, we reported the PhI NTs aziridina-
ˆ
tion/amidation of hydrocarbons catalyzed by an electron-
deficient manganese porphyrin, [Mn(tpfpp)Cl] (tpfpp ˆ me-
so-tetra(pentafluorophenyl)porphyrinato dianion), which fea-
tures up to 2600 catalyst turnovers.[5f]
The chiral manganese and ruthenium porphyrins 1 and 2
are the first metalloporphyrins that have been found to be
able to catalyze asymmetric nitrogen-atom transfer to hydro-
carbon substrates. In our preliminary works on aziridinations
catalyzed by complex 1,[6] and amidations catalyzed by
complex 1 or 2,[7] only the aziridination of substrates 6 10
and the amidation of substrates 27 32 were reported, both
From the results in Table 3, it is evident that the alkene
aziridination by 3 afforded the aziridines in the yields and ee
1568
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1568 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 7

Metalloporphyrin-Mediated Asymmetric N-Atom
1563 1572
values roughly comparable to the aziridine selectivities and ee
values observed for aziridinations catalyzed by complex 2
give
a
good linearity,[24] unlike the case of complex
[Ru(tpp)(NTs)2].[2e]
ˆ
with PhI NTs. A similar phenomenon was also noted for
The above differences between the nitrogen-atom-transfer
reactions of 3 and [Ru(tpp)(NTs)2] should stem from the
different electronic or steric properties of the porphyrin
ligands. Fusing two electron-donating, bulky norbornane
moieties to each of the four meso-phenyl groups of tpp to
form Por* would render the macrocycle more electron-rich
and sterically more demanding. This would make the N-
tosylimido groups in 3 less electrophilic than those of
[Ru(tpp)(NTs)2] and would provoke a larger steric interaction
between 3 and hydrocarbons, both are expected to make 3
react more slowly with styrenes than [Ru(tpp)(NTs)2] (note
the electrophilic nature of the attackof styrenes by 3 and
[Ru(tpp)(NTs)2] as reflected by the negative 1mb values
described above). In this context, the larger aziridination
rates observed for 3 than for [Ru(tpp)(NTs)2] are striking,
which might be associated with the different steric interaction
between the tosylimido groups and the porphyrin ligands in 3
and [Ru(tpp)(NTs)2].
the stoichiometric amidation by 3[7] and the PhI NTs
ˆ
amidation catalyzed by 2 (Table 4). These observations,
together with the facile formation of 3 from 2 and PhI NTs,
suggest the involvement of 3 as an active intermediate in the
asymmetric aziridination or amidation processes catalyzed by
complex 2.
ˆ
We previously isolated several aziridination- and amida-
tion-active, nonchiral, sterically-unencumbered bis(tosylimi-
do)ruthenium(vi) porphyrins and studied the mechanisms of
the reactions between [Ru(tpp)(NTs)2] and styrenes or
benzylic hydrocarbons in considerable detail.[2e] The results
in that workreveal that the aziridination of styrenes by
[Ru(tpp)(NTs)2] most probably occurs by the rate-limiting
formation of a benzylic carboradical intermediate rather than
in a concerted manner, whereas the amidation of ethyl-
benzenes most probably proceeds by hydrogen-atom abstrac-
tion by [Ru(tpp)(NTs)2] to form benzylic radicals followed by
scavenging of the radicals by the ruthenium porphyrin.
Similar mechanisms may remain valid for the reactions
between the chiral imido complex 3 and styrenes or benzylic
hydrocarbons considering that both 3 and [Ru(tpp)(NTs)2]
are bis(tosylimido)ruthenium(vi) meso-tetraarylporphyrins.
Indeed, the aziridination by 3 also appears to occur by a
nonconcerted pathway: the aziridination of cis-b-methylstyr-
ene 11 catalyzed by complex 2 afforded a mixture of cis and
trans aziridines 21 and 22 (entry 11 in Table 3), like the
aziridination of the same alkene by [Ru(tpp)(NTs)2]. More-
over, the substituent effects in 3- and [Ru(tpp)(NTs)2]-
mediated aziridinations/amidations are similar. As shown in
Table 7, for both of the imido complexes, the aziridination of
para-substituted styrenes is basically promoted by electron-
donating substituents but retarded by electron-withdrawing
substituents, whereas the amidation of para-substituted ethyl-
benzenes is basically promoted by either electron-donating or
-withdrawing substituents. Further, the larger k2 values
observed for 2-vinyl- and 2-ethylnaphthalene than for styr-
enes and ethylbenzenes, respectively (see Table 7), are con-
sistent with the intermediacy of benzylic radicals in the
reactions in view of a wider conjugation in the radicals with a
naphthyl ring than with a phenyl ring.
The tosylimido groups in 3 and [Ru(tpp)(NTs)2] are
expected to have essentially linear Ru-N-Ts moieties, a
geometry previously observed for the same imido groups in
the X-ray crystal structures of bis(tosylimido)osmium(vi)
porphyrins.[3q, 25] Suppose that the TsN Ru NTs moieties of
3 and [Ru(tpp)(NTs)2] have the same metrical parameters as
ˆ
ˆ
ˆ
ˆ
the TsN Os NTs moieties of bis(tosylimido)osmium(vi)
porphyrins,[26] modeling studies reveal that the rotation of
À
the Ts groups in 3 about the N S bonds is hampered by the
steric interaction between the Ts groups and the norbornane
moieties of the Por* ligand (see Figure 2a). However, the Ts
À
groups in [Ru(tpp)(NTs)2] can freely rotate about the N S
bonds (see Figure 2b). The rapid rotation of Ts groups about
À
the S N bond in [Ru(tpp)(NTs)2] could make it harder for
styrenes to approach the tosylimido N atoms (compare
Figure 2c and Figure 2d) and thus slow down the nitrogen-
atom-transfer reactions. This may rationalize the faster
reactions of 3 with styrenes. Alternatively, if complex 3 favors
a structure that has a smaller steric interaction between the Ts
groups and the norbornane moieties (relative to the structure
ˆ
shown in Figure 2a), the Ru NTs bonds in the complex
should be longer than those in [Ru(tpp)(NTs)2]. This would
ˆ
make 3 have weaker Ru NTs bonds, which may also account
However, we noted the following significant differences
between the nitrogen-atom-transfer reactions mediated by 3
and [Ru(tpp)(NTs)2]. First, the ratio of cis- versus trans-
aziridine (21:22) in the aziridination of 11 catalyzed by 2
(9:91) is substantially lower than that mediated by
[Ru(tpp)(NTs)2] (34:66).[2e] Second, the aziridination of
styrenes mediated by 3 is considerably faster than that by
[Ru(tpp)(NTs)2] (as reflected by the k2 values compiled in
for its faster reactions with styrenes.[27]
Concerning the modes by which terminal aromatic alkenes
ˆ
ArCH CH2 (such as styrenes and 2-vinylnaphthalene) ap-
proach the tosylimido N atoms of 3 or [Ru(tpp)(NTs)2],
several possibilities as shown in Scheme 3 are to be consid-
À À
ered. Evidently, the linear Ru N Ts arrangements would
prevent the alkenes from approaching the tosylimido N atoms
in a ™top-on∫ manner (Scheme 3a). ™Side-on∫ approaches as
shown in Scheme 3b and Scheme 3c are possible, which
minimize the interaction of the Ar groups with the porphyrin
ligands and allow a concerted nitrogen-atom transfer via
transition states 55 and 56 (see the inset to Scheme 3). The
transition state 56 may be more stable than 55 since the
former has a larger separation between the Ar and Ts groups.
However, even in 56, the steric interaction between the Ar
and Ts groups could still be significant due to their mutually
.
Table 7). Third, the log kR versus (smb, sJJ ) plot[22] for 3-
mediated aziridination of 6 and para-substituted styrenes 7, 9,
.
.
49, and 50 features 1mb of À0.58 and 1JJ of 1.05 with j1mb
1
JJ
j
/
.
of 0.55, in contrast to the 1mb of À1.05 and 1JJ of 0.52 with
.
j1mb
1
JJ
j
of 2.02 observed for [Ru(tpp)(NTs)2]-mediated
/
aziridination of the same alkenes.[2e] Fourth, for the 3-
mediated amidation of ethylbenzenes 27, 28, and 51 53,
plotting log kR against the radical parameter TE[23] did not
Chem. Eur. J. 2002, 8, No. 7
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1569 $ 20.00+.50/0
1569

C.-M. Che et al.
FULL PAPER
nitrogen-atom transfer via benzylic radical intermediates 57
or 58 to mitigate the steric interaction between the Ar and Ts
groups. In our previous kinetic studies on the aziridinations by
[Ru(tpp)(NTs)2],[2e] a concerted nitrogen-atom-transfer path-
way is also disfavored. The formation of the O-bound
aziridine adduct 4, rather than an N-bound aziridine adduct
similar to transition state 55 or 56,[28] from reaction of 2 with 5
provides additional evidence in disfavor of the concerted
pathway in the ruthenium porphyrin-mediated aziridinations.
Conclusion
We have shown that the nitrogen-atom-transfer reactions
ˆ
between PhI NTs and a variety of hydrocarbons catalyzed by
chiral metalloporphyrins [Mn(Por*)(OH)(MeOH)] (1) and
[Ru(Por*)(CO)(EtOH)] (2) feature good-to-excellent aziri-
dine selectivities for aromatic alkenes and good-to-excellent
amide selectivities for benzylic hydrocarbons. Reaction of
ˆ
cholesteryl acetate with PhI NTs in the presence of catalyst 1
afforded the corresponding amide in high regio- and chemo-
selectivity and moderate diastereoselectivity. The isolation of
the aziridination- and amidation-active chiral imido rutheni-
um porphyrin [Ru(Por*)(NTs)2] (3) and the ruthenium
porphyrin aziridine adduct [Ru(Por*)(CO)(TsAz)] (4,
TsAz ˆ N-tosyl-2-(4-chlorophenyl)aziridine) with an O- rath-
er than N-bound aziridine ligand provides useful insight into
the mechanism of asymmetric nitrogen-atom-transfer reac-
tions catalyzed by complex 2. Kinetic studies of the reactions
between 3 and aromatic hydrocarbons disclose a significant
effect of porphyrin ligand on the
Figure 2. Side-view model structures of a) [Ru(Por*)(NTs)2] (3) and
b) [Ru(tpp)(NTs)2]. The top-views of these model structures are depicted
in c) and d), respectively. Hydrogen atoms are not shown and the atoms of
the Ts groups are filled with grey colors. The modeling was based on the
X-ray crystal structures of complex
4
reported here and complex
[Os(tpp)(NTs)2] reported in reference [25].
nitrogen-atom-transfer reactions of
bis(tosylimido)ruthenium(vi) por-
a)
Ar
Ar
Ar
Ar
Ar
Ts
N
Ts
N
Ts
Ts
N
d)
Ts
N
b)
c)
e)
phyrins. Finally, the present work
first demonstrated the feasibility of
using ™PhI(OAc)2 ‡ NH2R∫ as the
nitrogen source in metal-catalyzed
asymmetric nitrogen-atom-transfer
reactions, and by employing this
amidation protocol, N-substituted
methanesulfonamides became ac-
cessible in up to 56% ee from
N
Ru
Ru
Ru
Ru
Ru
Ar
Ar
Ar
Ar
Ts
N
Ts
N
Ts
N
Ts
À
direct amidation of saturated C H
N
bonds catalyzed by complex 1 or 2.
Ru
Ru
Ru
Ru
57
55
56
58
Experimental Section
General: PhI(OAc)2 (Aldrich), NH2Ts
(Aldrich), NH2Ns (Acros), NH2SO2Me
(Aldrich), cholesteryl acetate (Acros),
and all the solvents (AR grade), except
dichloromethane, were used as received.
Scheme 3. a) Top-on and b) e) side-on approaches of terminal aromatic alkenes to the N atoms of the
tosylimido groups in 3 or [Ru(tpp)(NTs)2] (only one tosylimido axial group is shown in each case). The inset
shows the transition states 55 58 corresponding to the side-on approaches in b) e), respectively.
cis arrangement in the preformed aziridine. In fact, all the
Dichloromethane and aromatic hydrocarbons were purified as described
[29]
elsewhere.[2e] H2(Por*),[1g, 9] PhI NTs, cis-b-methylstyrene,[30] and com-
X-ray crystal structures of monocyclic C-substituted N-
tosylaziridines feature a trans rather than cis configuration[13g]
(note also the trans configuration of aziridine 5 in the
structure of 4). Therefore, the alkenes may preferentially
approach the tosylimido N atoms in a manner as shown in
Scheme 3d or Scheme 3e, which results in a nonconcerted
ˆ
plexes 1,[6] 2,[3k] 44 47[31] were prepared according to the literature
methods. 1H NMR spectra were measured on
a Bruker DPX300
spectrometer with CDCl3 as the solvent (the chemical shifts are relative
to tetramethylsilane). IR spectra were recorded on a Bio-Rad FT-IR
spectrometer (KBr pellet). UV/Vis spectra were obtained on a HP8453
diode array spectrophotometer. Mass spectra were measured on a Finnigan
1570
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1570 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 7

Metalloporphyrin-Mediated Asymmetric N-Atom
1563 1572
MAT95 mass spectrometer. HPLC measurements were carried out on a
HP1050 Series HPLC. Elemental analyses were performed by the Institute
of Chemistry, the Chinese Academy of Sciences.
a- and b-48: 1H NMR (CDCl3): d ˆ 8.37 (m, 2H), 8.05 (m, 2H), 5.63 (d, J ˆ
5.7 Hz, 1H), 5.23 (s, 1H), 4.55 (m, 1H), 3.83 (m, 1H), 3.64 (m, 1H), 2.04 (s,
‡
3H), 2.22 0.71 (m, 41H); MS: m/z: 628 [M] .
General procedure for the stoichiometric asymmetric aziridination/amida-
tion by complex 3: Complex 3 (0.05 mmol) was added to a solution of
substrate (2 mmol) in dichloromethane (4 mL) containing pyrazole (2%
w/w) in a 10-mL Schlenktube. The mixture was stirred for 2 h at 40 8C.
After removal of solvent, the amidation products were purified by column
chromatography on silica gel with n-hexane/ethyl acetate (6:1 v/v) as the
eluent.
Synthesis of bis(tosylimido)ruthenium(vi) porphyrin [Ru(Por*)(NTs)2] (3):
ˆ
A
mixture of [Ru(Por*)(CO)(EtOH)] (2, 0.05 mmol) and PhI NTs
(0.2 mmol) in dichloromethane (10 mL) was stirred at room temperature
under argon for 5 min. The mixture was then evaporated to dryness in
vacuo followed by chromatography on a short column of neutral alumina
with dichloromethane as the eluent. The first red band was collected and
the solvent was removed in vacuo, affording complex 3 as a darkpurple
powder in 64% yield. 1H NMR (300 MHz, CD2Cl2): d ˆ 8.77 (s, 8H), 7.38 (s,
4H), 6.19 (d, J ˆ 8.2 Hz, 4H), 4.90 (d, J ˆ 8.1 Hz, 4H), 3.57 (s, 8H), 2.72 (s,
8H), 1.98 (m, 8H), 1.87 (m, 14H), 1.36 (m, 24H), 1.11 (m, 8H); IR (KBr):
nÄ ˆ 1019 cmÀ1 (™oxidation-state marker∫ band); UV/Vis (CH2Cl2): lmax (log
e) ˆ 424 (5.26), 536 (4.14), 570 nm (3.76, sh); elemental analysis calcd (%)
for C98H90N6O4S2Ru ¥ 2CH2Cl2 (1750.87): C 68.60, H 5.41, N 4.80; found: C
69.04, H 5.40, N 4.29.
Kinetic studies: Kinetic measurements were performed on a HP8453
Diode Array spectrophotometer with an IBM-compatible PC and equip-
ped with a Lauda RM6 circulating water bath by using standard 1.0-cm
quartz cuvettes. The temperature of solutions during kinetic experiments
was maintained to within Æ0.28C. The rates of nitrogen-atom transfer from
3 to hydrocarbon substrates were measured by monitoring the decrease in
absorbance of 3 at 424 nm in dichloromethane containing pyrazole (2%
w/w) with a [substrate]:[3] concentration ratio of ꢂ100:1, a condition
identical to that employed in the rate measurements for the reactions
between [Ru(tpp)(NTs)2] and hydrocarbons.[2e] Pseudo-first-order rate
constants (kobs) were determined from the (Af À At) versus t plot (Af: the
final absorbance, At: the absorbance at time t) by nonlinear least-squares
fits of the data over four half-lives (t1/2) according to Equation (1), where Ai
is the initial absorbance.
Synthesis of [RuII(Por*)(CO)(TsAz)] (4): To a solution of [RuII(Por*)-
(CO)(EtOH)] (10 mg, 7.6 mmol) in chloroform (2 mL) was added free
aziridine 5 (9.4 mg, 30 mmol). The mixture was stirred for 1 h, and then
treated with excess hexane, leading to the formation of a darkpurple
precipitate. The precipitate was collected by filtration, washed with hexane,
and air-dried. Yield: 85%; 1H NMR (300 MHz, CD2Cl2): d ˆ 8.55 (dd, 8H),
7.37 (s, 4H), 3.56 (m, 8H), 2.72 (s, 8H), 1.97 (m, 8H), 1.89 (m, 8H), 1.39
0.86 (m, 32H), TsAz: 7.80 (d, J ˆ 8.3 Hz, 4H), 7.36 (d, 4H, partially
overlapped with d 7.37 signal), 7.28 (d, J ˆ 8.6 Hz, 4H), 7.16 (d, J ˆ 8.4 Hz,
4H), 3.67 (m, 1H), 2.91 (d, J ˆ 7.2 Hz, 1H), 2.35 (d, J ˆ 4.4 Hz, 1H), 2.43 (s,
3H); IR (KBr): nÄ ˆ 1935 cmÀ1 (CO); UV/Vis (CH2Cl2): lmax (log e) ˆ414
(5.02), 529 nm (4.05); elemental analysis calcd (%) for C100H90ClN5O3SRu ¥
(Af À At) ˆ (Af À Ai)exp(Àkobst)
(1)
Second-order rate constants, k2, were obtained from the plots of kobs versus
substrate concentration by linear least-squares fitting. For each substrate, at
least four data points were used to determine the k2 value, with the straight
line fit giving an R value of >0.99. The errors in the k2 values shown in
Table 7 are the errors of the straight-line fits.
5 4CHCl (1727.63): C 70.39, H 5.32, N 4.05; found: C 70.55, H 5.61, N 3.83.
/
3
ˆ
General procedure for asymmetric aziridination/amidation with PhI NTs
catalyzed by complex 1 or 2: A solution of substrate (0.15 mmol) in dry
dichloromethane (4 mL) was added by syringe into a Schlenkflask
containing 1 or 2 (0.002 mmol) and molecular sieves (4 ä, 50 mg). The
mixture was stirred at room temperature for 10 min, then treated with
X-ray structure determination of complex 4: Single crystals of 4 were
obtained by slow diffusion of hexane into a solution of 4 in chloroform
containing free aziridine 5. A crystal of dimensions 0.5 Â 0.2 Â 0.15 mm
mounted in a glass capillary was used for data collection at 288C on a MAR
ˆ
PhI NTs (0.30 mmol), and maintained at 408C for 2 h. After the mixture
diffractometer with
a 300 mm image plate detector using graphite-
was cooled to room temperature, the molecular sieves were filtered off and
washed with dichloromethane. The filtrate and washings were combined
and then evaporated to dryness. The product was purified by column
chromatography on silica gel with n-hexane/ethyl acetate (6:1 v/v) as the
eluent.
monochromated MoKa radiation (l ˆ 0.71073 ä). The images were inter-
preted and the intensities integrated by using program DENZO. The
structure was solved by direct methods by employing the SHELXS-97
program. Many non-hydrogen atoms including the ruthenium atom were
located by direct methods. The positions of the other non-hydrogen atoms
were found after successful refinement (against F 2) by full-matrix least-
squares using SHELXL-97. In the final stage of the least-squares refine-
ment, all non-hydrogen atoms were refined anisotropically, whereas
hydrogen atoms were generated by SHELXL-97. The positions of hydrogen
atoms were calculated based on riding mode with thermal parameters
equal to 1.2 times that of the associated carbon atoms.
a-43: The spectral data of this compound are identical with those reported
in the literature.[15]
b-43: 1H NMR (CDCl3): d ˆ 7.73 (d, J ˆ 8.2 Hz, 2H), 7.30 (d, J ˆ 8.2 Hz,
2H), 4.76 (s, 1H), 4.51 (m, 1H), 4.04 (d, J ˆ 9.2 Hz, 1H), 3.65 (t, 1H), 2.44
‡
(s, 3H), 2.00 (s, 3H), 2.20 0.85 (m, 41H); MS: m/z: 597 [M] .
CCDC-172416 (4) contains the supplementary crystallographic data
(excluding structure factors) for the structure reported in this paper. These
data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: (‡44)1223-336033; or
deposit@ccdc.cam.ac.uk).
General procedure for asymmetric amidation with ™PhI(OAc)2 ‡ NH2R
(R ˆ Ts, Ns, SO2Me)∫ catalyzed by complex 1 or 2: To a well-stirred
suspension of molecular sieves (4 ä, 50 mg) in dry dichloromethane (4 mL)
containing 1 or 2 (0.002 mmol) at room temperature was added the
substrate (0.20 mmol) by means of
a syringe. After 10 min, NH2R
(0.30 mmol) and PhI(OAc)2 (0.25 mmol) were added and the mixture
was stirred at 408C for 2 h. The solution was then filtered and the products
were purified by column chromatography on silica gel with n-hexane/ethyl
acetate (6:1 v/v) as the eluent.
Acknowledgements
39: 1H NMR (CDCl3): d ˆ 7.26 (d, J ˆ 8.4 Hz, 2H), 6.90 (d, J ˆ 8.3 Hz, 2H),
4.62 (m, 1H), 4.55 (d, J ˆ 6.0 Hz, 1H), 3.81 (s, 3H), 2.62 (s, 3H), 1.52 (d, J ˆ
This workwas supported by The University of Hong Kong, the Hong Kong
Research Grants Council (HKU 7096/97P and HKU 7099/00P) the Hong
Kong University Foundation, and Area of Excellence Scheme (AoE/P-10/
01), University Grants Committee of the Hong Kong SAR, China. J.S.H. is
grateful to The University of Hong Kong for a Postdoctoral Fellowship.
‡
‡
6.9 Hz, 3H); MS: m/z: 229 [M] ; HRMS m/z [M] calcd for C10H15NO3S:
229.0773, found: 229.0769.
40: 1H NMR (CDCl3): d ˆ 7.42 (m, 1H), 7.20 (m, 2H), 7.09 (m, 1H), 4.65 (m,
1H), 4.51 (d, J ˆ 7.9 Hz, 1H), 3.07 (s, 3H), 2.80 (m, 2H), 2.10 (m, 2H), 1.89
‡
‡
(m, 2H); MS: m/z: 225 [M] ; HRMS m/z [M] calcd for C11H15NO2S:
[1] Reviews: a) B. Meunier, Chem. Rev. 1992, 92, 1411; b) J. P. Collman,
X. Zhang, V. J. Lee, E. S. Uffelman, J. I. Brauman, Science 1993, 261,
1404; selected recent examples: c) J. P. Collman, V. J. Lee, C. J.
Kellen-Yuen, X. Zhang, J. A. Ibers, J. I. Brauman, J. Am. Chem. Soc.
1995, 117, 692; d) Z. Gross, S. Ini, M. Kapon, S. Cohen, Tetrahedron
Lett. 1996, 37, 7325; e) A. Berkessel, M. Frauenkron, J. Chem. Soc.
225.0824, found: 225.0837.
41: M.p. 148 1498C; 1H NMR (CDCl3): d ˆ 7.83 (m, 4H), 7.49 (m, 3H),
4.85 (m, 1H), 4.75 (d, J ˆ 5.9 Hz, 1H), 2.62 (s, 3H), 1.63 (d, J ˆ 6.9 Hz, 3H);
‡
‡
MS: m/z: 249 [M] ; HRMS m/z [M] calcd for C13H15NO2S: 249.0824,
found: 249.0819.
Chem. Eur. J. 2002, 8, No. 7
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1571 $ 20.00+.50/0
1571

C.-M. Che et al.
FULL PAPER
Perkin Trans. 1 1997, 2265; f) Z. Gross, S. Ini, J. Org. Chem. 1997, 62,
5514; g) R. L. Halterman, S.-T. Jan, H. L. Nimmons, D. J. Standlee,
M. A. Khan, Tetrahedron 1997, 53, 11257; h) T.-S. Lai, R. Zhang, K.-K.
Cheung, H.-L. Kwong, C.-M. Che, Chem. Commun. 1998, 1583; i) T.-
S. Lai, H.-L. Kwong, R. Zhang, C.-M. Che, J. Chem. Soc. Dalton Trans.
1998, 3559; j) J. P. Collman, Z. Wang, A. Straumanis, M. Quelquejeu,
E. Rose, J. Am. Chem. Soc. 1999, 121, 460; k) X.-Q. Yu, J.-S. Huang,
W.-Y. Yu, C.-M. Che, J. Am. Chem. Soc. 2000, 122, 5337.
B. G. Siim, K. G. Robinson, W. A. Denny, P. J. Brothers, G. R. Clark,
Inorg. Chem. 1991, 30, 3750.
[13] a) L. M. Trefonas, R. Majeste, J. Heterocycl. Chem. 1965, 2, 80;
b) L. M. Trefonas, T. Sato, J. Heterocycl. Chem. 1966, 3, 404; c) H. M.
Zacharis, L. M. Trefonas, J. Heterocycl. Chem. 1968, 5, 343; d) H.
Zacharis, L. M. Trefonas, J. Heterocycl. Chem. 1970, 7, 755; e) H. M.
Zacharis, L. M. Trefonas, J. Heterocycl. Chem. 1970, 7, 1301; f) J. N.
Brown, R. L. R. Towns, L. M. Trefonas, J. Heterocycl. Chem. 1970, 7,
1321; g) W. Clegg, S. L. Heath, L. Horsburgh, R. F. W. Jackson, A.
Wood, Acta Crystallogr. C 1996, 52, 2779.
[2] a) J. T. Groves, T. Takahashi, J. Am. Chem. Soc. 1983, 105, 2073; b) D.
Mansuy, J.-P. Mahy, A. Dureault, G. Bedi, P. Battioni, J. Chem. Soc.
Chem. Commun. 1984, 1161; c) J.-P. Mahy, G. Bedi, P. Battioni, D.
Mansuy, J. Chem. Soc. Perkin Trans. 2 1988, 1517; d) S.-M. Au, S.-B.
Zhang, W.-H. Fung, W.-Y. Yu, C.-M. Che, K.-K. Cheung, Chem.
Commun. 1998, 2677; e) S.-M. Au, J.-S. Huang, W.-Y. Yu, W.-H. Fung,
C.-M. Che, J. Am. Chem. Soc. 1999, 121, 9120.
[3] a) H. J. Callot, C. Piechocki, Tetrahedron Lett. 1980, 21, 3489; b) S.
O'Malley, T. Kodadek, Tetrahedron Lett. 1991, 32, 2445; c) J. L.
Maxwell, K. C. Brown, D. W. Bartley, T. Kodadek, Science 1992, 256,
1544; d) J. L. Maxwell, S. O'Malley, K. C. Brown, T. Kodadek,
Organometallics 1992, 11, 645; e) S. O'Malley, T. Kodadek, Organo-
metallics 1992, 11, 2299; f) K. C. Brown, T. Kodadek, J. Am. Chem.
Soc. 1992, 114, 8336; g) D. W. Bartley, T. Kodadek, J. Am. Chem. Soc.
1993, 115, 1656; h) D. A. Smith, D. N. Reynolds, L. K. Woo, J. Am.
Chem. Soc. 1993, 115, 2511; i) J. R. Wolf, C. G. Hamaker, J.-P. Djukic,
T. Kodadek, L. K. Woo, J. Am. Chem. Soc. 1995, 117, 9194; j) E.
Galardon, P. Le Maux, G. Simonneaux, Chem. Commun. 1997, 927;
k) W.-C. Lo, C.-M. Che, K.-F. Cheng, T. C. W. Mak, Chem. Commun.
1997, 1205; l) M. Frauenkron, A. Berkessel, Tetrahedron Lett. 1997, 38,
[14] The three-membered aziridine rings of the structurally characterized
À
free aziridines reported in references [13a g] feature C C bond
À
lengths of 1.44 1.53 ä and C N bond lengths of 1.44 1.57 ä. In each
À
case, the lengths of the two C N bonds are comparable or almost
identical; the C C N and C N C angles of the ring are all close to
608.
À À
À À
[15] The spectral data of the predominant isomer of 43 are identical with
those of a-43 reported in the literature (D. H. R. Barton, R. S. Hay-
Motherwell, W. B. Motherwell, J. Chem. Soc. Perkin Trans. 1 1983,
445). The a and b configurations were further confirmed by NOESY
measurements.
[16] P. M¸ller, in Advances in Catalytic Processes, Vol. 2, Asymmetric
Catalysis (Ed.: M. P. Doyle), JAI Press, Greenwich, 1997, p. 113.
ˆ
[17] Note that the use of excess PhI NTs generally resulted in a decrease
in enantioseletivity in these aziridination or amidation reactions. We
employed such conditions in this workmainly to evaluate the mass
balance in the catalytic nitrogen-atom-transfer processes.
[18] Metal-complex-catalyzed amidation of steroids is important consid-
ering the noteworthy pharmacological activity of amino steroids (see
for example: P. H. D. Chenna, P. Dauban, A. Ghini, G. Burton, R. H.
Dodd, Tetrahedron Lett. 2000, 41, 7041). Prior to the present work,
Breslow and co-workers reported the benzylic amidation of equilenin
¬
7175; m) E. Galardon, S. Roue, P. Le Maux, G. Simonneaux,
Tetrahedron Lett. 1998, 39, 2333; n) Z. Gross, N. Galili, L. Simkhovich,
Tetrahedron Lett. 1999, 40, 1571; o) E. Galardon, P. Le Maux, G.
Simonneaux, Tetrahedron 2000, 56, 615; p) C.-M. Che, J.-S. Huang, F.-
W. Lee, Y. Li, T.-S. Lai, H.-L. Kwong, P.-F. Teng, W.-S. Lee, W.-C. Lo,
S.-M. Peng, Z.-Y. Zhou, J. Am. Chem. Soc. 2001, 123, 4119; q) Y. Li, J.-
S. Huang, Z.-Y. Zhou, C.-M. Che, J. Am. Chem. Soc. 2001,
123, 4843.
ˆ
acetate with PhI NTs catalyzed by [Mn(tpfpp)Cl] (see ref. [5e]).
[19] S.-M. Au, J.-S. Huang, C.-M. Che, W.-Y. Yu, J. Org. Chem. 2000, 65,
7858.
[20] PhI(OAc)2 and NH2R are the well-known precursors to iminoiodanes
ˆ
[4] Reviews: a) P. E. Ellis, Jr., J. E. Lyons, Coord. Chem. Rev. 1990, 105,
181; b) D. Mansuy, Coord. Chem. Rev. 1993, 125, 129; selected recent
examples: c) J. T. Groves, P. Viski, J. Am. Chem. Soc. 1989, 111, 8537;
d) H. Ohtake, T. Higuchi, M. Hirobe, J. Am. Chem. Soc. 1992, 114,
10660; e) A. Sorokin, A. Robert, B. Meunier, J. Am. Chem. Soc. 1993,
115, 7293, and references therein; f) M. W. Grinstaff, M. G. Hill, J. A.
Labinger, H. B. Gray, Science 1994, 264, 1311; g) R. Zhang, W.-Y. Yu,
T.-S. Lai, C.-M. Che, Chem. Commun. 1999, 1791; h) Z. Gross, S. Ini,
Org. Lett. 1999, 1, 2077; i) K. Wietzerbin, J. G. Muller, R. A. Jameton,
G. Pratviel, J. Bernadou, B. Meunier, C. J. Burrows, Inorg. Chem.
1999, 38, 4123; j) J. F. Bartoli, V. Mouries-Mansuy, K. Le Barch-
Ozette, M. Palacio, P. Battioni, D. Mansuy, Chem. Commun. 2000, 827;
k) W. Nam, M. H. Lim, S. K. Moon, C. Kim, J. Am. Chem. Soc. 2000,
122, 10805.
[5] a) R. Breslow, S. H. Gellman, J. Chem. Soc. Chem. Commun. 1982,
1400; b) R. Breslow, S. H. Gellman, J. Am. Chem. Soc. 1983, 105, 6728;
c) J. P. Mahy, G. Bedi, P. Battioni, D. Mansuy, Tetrahedron Lett. 1988,
29, 1927; d) J. P. Mahy, G. Bedi, P. Battioni, D. Mansuy, New J. Chem.
1989, 13, 651; e) J. Yang, R. Weinberg, R. Breslow, Chem. Commun.
2000, 531; f) X.-Q. Yu, J.-S. Huang, X.-G. Zhou, C.-M. Che, Org. Lett.
2000, 2, 2233.
[6] T.-S. Lai, H.-L. Kwong, C.-M. Che, S.-M. Peng, Chem. Commun. 1997,
2373.
[7] X.-G. Zhou, X.-Q. Yu, J.-S. Huang, C.-M. Che, Chem. Commun. 1999,
2377.
[8] J. P. Simonato, J. Pecaut, W. R. Scheidt, J. C. Marchon, Chem.
Commun. 1999, 989.
[9] R. L. Halterman, S.-T. Jan, J. Org. Chem. 1991, 56, 5253.
[10] I. N‰geli, C. Baud, G. Bernardinelli, Y. Jacquier, M. Moran, P. M¸ller,
Helv. Chim. Acta 1997, 80, 1087.
PhI NR. As we pointed out elsewhere (see references [5f, 19]), the
use of ™PhI(OAc)2 ‡ NH2R∫ not only bypasses the preparation of
ˆ
PhI NR, but also makes it feasible to prepare certain N-substituted
ˆ
amides that are inaccessible by the PhI NR protocol because the
required iminoiodanes are unstable or unknown.
[21] Dauban, Dodd, and their co-workers recently reported the asymmet-
ˆ
ˆ
ric aziridination of alkenes with ™PhI O ‡ NH2R∫ (the PhI O was
prepared from PhI(OAc)2). See: P. Dauban, L. Saniere, A. Tarrade,
R. H. Dodd, J. Am. Chem. Soc. 2001, 123, 7707.
[22] X.-K. Jiang, Acc. Chem. Res. 1997, 30, 283.
[23] Y. D. Wu, C.-L. Wong, K. W. K. Chan, G.-Z. Ji, X.-K. Jiang, J. Org.
Chem. 1996, 61, 746.
[24] In this case, a good linearity (R ˆ 0.99) was obtained for the log kR
.
.
versus (smb, sJJ ) plot (1mb ˆ À0.79, 1JJ ˆ 0.26).
[25] S.-M. Au, W.-H. Fung, J.-S. Huang, K.-K. Cheung, C.-M. Che, Inorg.
Chem. 1998, 37, 6564.
ˆ
ˆ
[26] This was based on the almost identical Ru O and Os O bond lengths
in the dioxoruthenium(vi) and dioxoosmium(vi) porphyrins [Ru-
ˆ
(Por*)(O)2] (Ru O: ꢁ1.74(1) ä, see ref. [1h]) and [Os(tpfpp)(O)2]
ˆ
(Os O: 1.741(2) ä, see ref. [3q]).
[27] The larger k2 values for the amidation of ethylbenzenes 27, 28, and 52
by 3 than by [Ru(tpp)(NTs)2] (see Table 7) may be rationalized in a
similar manner. However, it is unclear why the amidations of 51 and
53 by 3 are slower than by [Ru(tpp)(NTs)2].
[28] This is in contrast to the isolation of a styrene oxide adduct,
[Ru(tdcpp)(CO)(styrene oxide)] (tdcpp ˆ meso-tetrakis(2,6-dichloro-
phenyl)porphyrinato dianion), from reaction of [Ru(tdcpp)(CO)-
(MeOH)] with styrene oxide by Groves and co-workers. See: J. T.
Groves, Y. Han, D. V. Engen, J. Chem. Soc. Chem. Commun. 1990,
436.
[11] Recently, Katsuki and Kohmura reported the [Mn(salen)]-catalyzed
[29] Y. Yamada, T. Yamamoto, M. Okawara, Chem. Lett. 1975, 361.
[30] H. Lindlar, R. Dubuis, Org. Synth. Collect. Vol. 5 1973, 880.
[31] A. Robert, B. Loock, M. Momenteau, B. Meunier, Inorg. Chem. 1991,
30, 706.
asymmetric amidation of allylic and benzylic hydrocarbons with
ˆ
PhI NTs, which produced amides in up to 89% ee. See: Y. Kohmura,
T. Katsuki, Tetrahedron Lett. 2001, 42, 3339.
[12] Metal complexes bearing aziridine ligands are rare, very few of which
are characterized by X-ray structure determination. See: D. C. Ware,
Received: October 18, 2001 [F3623]
1572
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0807-1572 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 7