Asymmetric meso-aziridine ring-opening reactions using a chiral zirconium catalyst

Article abstract of DOI:10.1039/b914271c

A chiral zirconium catalyst prepared from Zr(O^tBu)₄ and a chiral tridentate BINOL was found to be effective for asymmetric meso-aziridine ring-opening reactions with aniline derivatives. The N-benzhydryl group on the product obtained

Full text of DOI:10.1039/b914271c

View Online / Journal Homepage / Table of Contents for this issue
This article was published as part of the
2009 ‘Catalysis in Organic
Synthesis’ web theme issue
Showcasing high quality research in organic chemistry
Please see our website
(http://www.rsc.org/chemcomm/organicwebtheme2009)
to access the other papers in this issue.

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Asymmetric meso-aziridine ring-opening reactions using a chiral
zirconium catalystwz
Kazutaka Seki, Rongmin Yu, Yumi Yamazaki, Yasuhiro Yamashita and Shu¯ Kobayashi*
Received (in Cambridge, UK) 15th July 2009, Accepted 18th August 2009
First published as an Advance Article on the web 7th September 2009
DOI: 10.1039/b914271c
A chiral zirconium catalyst prepared from Zr(OtBu)4 and a
chiral tridentate BINOL was found to be effective for
asymmetric meso-aziridine ring-opening reactions with aniline
derivatives. The N-benzhydryl group on the product obtained
was easily cleaved to give the corresponding amine in high yield
under reductive conditions.
Regioselective ring-opening reactions of meso-aziridines with
amines are an efficient methodology for providing chiral
1,2-diamine derivatives, which are useful building blocks for
the synthesis of natural products, as well as chiral organo-
catalysts and chiral ligands for enantioselective reactions.1
While asymmetric meso-epoxide ring-opening reactions have
been relatively well investigated,2 the corresponding reactions
with aziridines have been less progressed.3 Recently, some
successful examples using highly reactive nucleophiles such as
TMSCN, MeMgBr and TMSN3 have been reported.4
On the other hand, less reactive amine nucleophiles such as
aniline derivatives have not been well studied. Our group has
already disclosed that asymmetric ring-opening reactions of
meso-aziridines with aniline derivatives proceeded smoothly
using a chiral niobium catalyst, but that enantioselectivities
were not satisfactory (up to 84% ee).5 We also found that a
chiral titanium catalyst promoted the ring-opening reactions
of meso-aziridines with high enantioselectivities.6 In both
cases, o-methoxyphenyl group was employed as a N-protecting
group, however successful removal of this group from the
product in oxidative conditions was sometimes difficult. On
the other hand, in the course of our investigations to develop
more efficient catalysts for the activation of aziridines, we
focused on possibilities of Zr and Hf.7 Herein we report
asymmetric ring-opening reactions of meso-N-benzhydryl
aziridines with anilines catalyzed by a Zr-BINOLate complex
(Scheme 1).
Scheme 1 Asymmetric meso-aziridine ring-opening reaction of aniline.
facilitate catalytic Lewis acid activations. However, unexpectedly,
the reactions did not proceed at all in each case. Next, we
turned our attention to aziridines bearing N-benzyl and related
groups. While the aziridine bearing an N-benzyl group showed
poor reactivity, the aziridine bearing an N-benzhydryl group
gave the desired ring-opening product in good yield.
We further surveyed the catalysts and finally found that
zirconium-BINOLate complexes were promising, and a high
yield and moderate enantioselectivity were obtained when a
catalyst prepared from Zr(OiPr)4 and (R)-tridentate BINOL 1
was employed (Table 1, entry 1).
Table 1 Asymmetric meso-aziridine ring-opening reaction of anilinea
Yield ee
Entry M(OR0)4
Lb Additives/mol% Temp./1C (%)
(%)
1
2
3
4
5
6
7
8
Zr(OiPr)4
Zr(OtBu)4
Zr(OnPr)4
Zr(OnPr)4
Zr(OtBu)4
Zr(OtBu)4
Zr(OtBu)4
Zr(OtBu)4
Zr(OtBu)4
Zr(OtBu)4
Hf(OtBu)4
Hf(OtBu)4
Hf(OtBu)4
1
1
1
1
1
1
2
3
1
1
1
1
1
—
—
—
—
rt
rt
rt
0
0
0
70
54
73
76
83
82
58
69
50
29
69
73
73
75
29
9
First, we conducted several ring-opening reactions using
cyclohexene oxide-derived aziridines bearing N-tosyl and
N-benzoyl groups in the presence of a chiral Nb or Ti catalyst.
In these aziridines their basicities are lower, which might
EtOH (100)
nC5H11OH (100)
nC5H11OH (100) rt
nC5H11OH (100) rt
9c
10c
11
12
13d
nC5H11OH (50)
nC5H11OH (50)
—
0
quant 73
ꢀ20
54
96
86
83
77
20
62
71
rt
nC5H11OH (100) rt
Department of Chemistry, School of Science and Graduate School of
Pharmaceutical Sciences, The University of Tokyo, The HFRE
Division, ERATO, Japan Science Technology Agency (JST),
HongoBunkyo-ku, Tokyo 113-0033, Japan.
EtOH (50)
0
a
The reaction was performed using aziridine 4a and aniline 5a
(1.2 equiv.) in toluene (0.1 M) at 0 1C for 48 h in the presence of a
chiral catalyst prepared from M(OR0)4 (20 mol%), (R)-ligand (22 mol%)
and an additive unless otherwise noted. The absolute configuration of
E-mail: shu_kobayashi@chem.s.u-tokyo.ac.jp
w This article is part of a ChemComm ‘Catalysis in Organic Synthesis’
web-theme issue showcasing high quality research in organic
chemistry. Please see our website (http://www.rsc.org/chemcomm/
organicwebtheme2009) to access the other papers in this issue.
z Electronic supplementary information (ESI) available: Experimental
details and characterization data. See DOI: 10.1039/b914271c
b
c
the product has not been determined yet. (R)-Ligand. Catalyst
d
loading was 10 mol%, and concentration was 0.3 M. Catalyst
loading was 10 mol%, and concentration was 0.5 M.
ꢁc
This journal is The Royal Society of Chemistry 2009
5722 | Chem. Commun., 2009, 5722–5724

We then optimized the reaction conditions using the
aziridine 4a derived from cyclohexene oxide and aniline (5a)
as models. It was found that the alkoxide part of the Zr
catalyst influenced the reactivity and selectivity significantly.
While a complex of zirconium tert-butoxide and chiral ligand
1 showed poor enantioselectivity, the complex prepared from
zirconium primary alkoxide gave a higher selectivity (entries 2
and 3). Furthermore, the length of the alkoxide moiety was
investigated, and finally the zirconium complex prepared from
n-pentanol via exchange of the remaining tert-butoxy part was
found to be the best (entries 4–6). We also employed tridentate
BINOL 2 and tetradentate BINOL 3; however, the selectivities
were not satisfactory (entries 7 and 8). While the reaction
at lower temperature showed higher selectivities, and good
enantioselectivities were obtained (73 and 77% ee), the yields
were also moderate to high, even in the presence of 10 mol%
of the catalyst (entries 9 and 10). On the other hand,
Hf(OtBu)4 was investigated instead of Zr(OtBu)4, and a similar
tendency was observed as a result, except that the Hf catalyst
was less stable and showed a higher selectivity when using
ethanol as an alcohol additive (entries 11–13).
cyano, nitro and ester groups gave high enantioselectivities
(entries 6–8). On the other hand, 3-methoxyaniline showed
good selectivity in spite of its electron-donating nature
(entry 9). However, selectivities slightly decreased when
methylanilines were used, and 2-substituted aniline did not
give a good result (entries 10–12). The trifluoromethyl group, a
strong electron-withdrawing group, was also a promising
substituent, and high enantioselectivities were obtained
(entries 13 and 14). Notably, 92% ee was obtained when
3,5-bis(trifluoromethyl)aniline was used (entry 15).
We also examined the reactions with other meso-aziridines
using 3,5-bis(trifluoromethyl)aniline as a nucleophile. Other
aziridines containing 6-membered ring systems gave high
yields and high enantioselectivities (entries 16 and 17). The
aziridine containing a 5-membered ring system worked well to
afford the desired diamine in high selectivity (entry 18). The
monocyclic aziridines derived from cis-butene oxide and
cis-hexene oxide also reacted with the aniline derivatives
in moderate to high yields with high enantioselectivities
(entries 19 and 20).
For cleavage of the benzhydryl group from the product,
Pd/C-H2 or Et3SiH/TFA reduction8 was found to be effective,
and the desired monoamine-free diamine was obtained in high
yield. In this reaction, recoverable and reusable polymer-
incarcerated palladium catalyst (PI-Pd)9 also worked well,
and an excellent yield was obtained (Scheme 2), which was
successfully employed in hydrogenation and Suzuki coupling
reactions.
Next, we investigated the substrate scope of the aziridine
ring opening reaction, and the results are summarized in
Table 2. It was found that aniline derivatives bearing
electron-withdrawing groups such as halogen atoms on the
benzene ring showed high yields and high enantioselectivities,
regardless of their less nucleophilic nature. When 3-halogen
and 3,4-dihalogen substituted aniline derivatives were used,
high enantioselectivities of over 80% ee were obtained
(entries 2–5). Other aniline derivatives containing electron-
withdrawing groups were also examined, and anilines bearing
Finally, we investigated amplification of the optical yield to
get information of structure of the Zr catalyst (Fig. 1). First,
we conducted the ring-opening reaction of 4a with 5a using
(R)-1 with low optical purity (op) that was prepared by mixing
Table 2 Substrate scopea
Entry
4
Ar
Yield (%)
ee (%)
1b
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4b
4c
4d
4e
4f
Ph (5a)
3-ClC6H4 (5b)
3-BrC6H4 (5c)
3-IC6H4 (5d)
3,4-Cl2C6H3 (5e)
3-NCC6H4 (5f)
3-O2NC6H4 (5g)
3-(EtOCO)C6H4 (5h)
3-MeOC6H4 (5i)
4-MeC6H4 (5j)
quant
80
80
75
90
86
55
94
92
86
72
54
77
73
81
96
93
51
77
56
73
86
83
83
85
80
84
80
80
73
76
65
83
86
92
93
92
86
93
80
2b
3b
4b
5
6
7
8
Scheme 2 Cleavage of the N-protecting group.
9
10
11
12
13b
14
15c
16de
17de
18cd
19d
20f
3-MeC6H4 (5k)
2-MeC6H4 (5l)
4-CF3C6H4 (5m)
3-CF3C6H4 (5n)
3,5-(CF3)2C6H3 (5o)
3,5-(CF3)2C6H3 (5o)
3,5-(CF3)2C6H3 (5o)
3,5-(CF3)2C6H3 (5o)
3,5-(CF3)2C6H3 (5o)
3-ClC6H4 (5b)
a
The reaction was performed using aziridine 4 and aniline derivative 5
in toluene (0.3 M) at 0 1C for 48 h in the presence of the chiral Zr
catalyst prepared from Zr(OtBu)4 (10 mol%), (R)-1 (11 mol%) and
n-pentanol (50 mol%) unless otherwise noted. The absolute
configurations of the products have not been determined yet.
b
c
Reaction runs for 24 h. 3.0 equiv. of aniline derivatives were used.
d
e
Reaction time was 68 h. Reaction runs at ꢀ10 1C. Concentration
f
was 0.5 M. 20 mol% catalyst was used. Concentration was 1.0 M.
Fig. 1 Non-linear effect on the product ee.
Chem. Commun., 2009, 5722–5724 | 5723
ꢁc
This journal is The Royal Society of Chemistry 2009

optical purity of the BINOL and the ee of the product. Further
investigation of structure of the active species is now in
progress.
This work was partially supported by a Grant-in-Aid for
Science Research from the Japan Society for the Promotion
of Science (JSPS) and Global COE Program (Chemistry
Innovation through Cooperation of Science and Engineering),
The University of Tokyo, MEXT, Japan.
Fig. 2 Assumed structures of the chiral zirconium complex.
Notes and references
optically pure (R)-1 ligand and (S)-1 ligand in proper ratio
(method A).
1 For example of non-chiral aziridine ring-opening reaction with
aniline see: (a) S. Peruncheralathan, M. Henze and C. Schneider,
Synlett, 2007, 2289; (b) J. Wu, X. Sun and W. Sun, Org. Biomol.
Chem., 2006, 4, 4231; (c) J. Wu, X. Sun and Y. Li, Eur. J. Org.
Chem., 2005, 46, 4271; (d) U. K. Nadir and A. Singh, Tetrahedron
Lett., 2005, 46, 2083; (e) S. Chandrasekhar, S. J. Prakash and
T. Shyamsunder, Synth. Commun., 2004, 34, 3865; (f) J. S. Yadav,
B. V. S. Reddy and K. Premalatha, Adv. Synth. Catal., 2003, 345,
948; (g) J.-Y. Goujon, D. Gueyrard, P. Compani, O. R. Martin
and N. Asano, Tetrahedron: Asymmetry, 2003, 14, 1969; (h) I. D.
G. Watson and A. K. Yudin, J. Org. Chem., 2003, 68, 5160;
(i) N. R. Swamy and Y. Venkateswarlu, Synth. Commun., 2003,
33, 547; (j) R.-H. Fan and X.-L. Hou, J. Org. Chem., 2003, 68, 726;
(k) R. V. Anand, G. Pandey and V. K. Singh, Tetrahedron Lett.,
2002, 43, 3975; (l) J. S. Yadav, B. V. S. Reddy, B. Jyothirmai and
M. S. R. Murty, Synlett, 2002, 53; (m) G. Sekar and V. K. Singh,
J. Org. Chem., 1999, 64, 2537; (n) M. Meguro, N. Asao and
Y. Yamamoto, Tetrahedron Lett., 1994, 35, 7395.
2 Review for ring-opening reaction of epoxide: (a) C. Schneider,
Synthesis, 2006, 3919; (b) I. M. Pastor and M. Yus, Curr. Org.
Chem., 2005, 9, 1; (c) E. N. Jacobsen and M. H. Wu,
Comprehensive Asymmetric Catalyst, ed. E. N. Jacobsen,
A. Pfaltz and H. Yamamoto, Springer, Berlin, 1999, vol. 2, p. 649.
3 (a) C. Schneider, Angew. Chem., Int. Ed., 2009, 48, 2082;
(b) Aziridines and Epoxides in Organic Synthesis, ed.
A. K. Yudin, Wiley-VCH, Weiheim, 2006; (c) M. Pineschi, Eur.
J. Org. Chem., 2006, 4979; (d) X. E. Hu, Tetrahedron, 2004, 60,
2701; (e) W. McCoull and F. A. Davis, Synthesis, 2000, 1347.
4 Review for ring opening reaction of aziridine: (a) Z. Wang, X. Sun,
S. Ye, W. Wang, B. Wang and J. Wu, Tetrahedron: Asymmetry,
2008, 19, 964; (b) E. B. Rowland, G. B. Rowland, E. Rivera-Otero
and J. C. Antilla, J. Am. Chem. Soc., 2007, 129, 12084;
(c) I. Fujimori, T. Mita, K. Maki, M. Shiro, A. Sato,
S. Furusho, M. Kanai and M. Shibasaki, J. Am. Chem. Soc.,
2006, 128, 16438; (d) Y. Fukuta, T. Mita, N. Fukuda, M. Kanai
and M. Shibasaki, J. Am. Chem. Soc., 2006, 128, 6312; (e) Z. Li,
M. Fermandez and E. N. Jacobsen, Org. Lett., 1999, 1, 1611;
In this case, higher ees were nevertheless observed the less
optically pure ligands were used. This significant positive
non-linear effect (NLE) on the product ee could be explained
based on an assumption that some catalytically inactive
species containing an equal amount of R and S ligands formed
in the reaction system. Next, we conducted the reaction using
‘‘mixed catalyst’’ prepared by mixing two optically pure
zirconium catalysts ((R)-catalyst and (S)-catalyst) that were
already prepared (method B). In method B, we performed two
types of experiments; the difference between the experiments
was only the ‘‘catalyst mixing time’’. One reaction was
conducted using the mixed catalyst, which was prepared by
30 min mixing of (R)-catalyst and (S)-catalyst (method B-1)
and the other was conducted using the catalyst prepared by 3 h
mixing (method B-2). In each case, while a significant NLE on
the product ee was observed, the degree of the NLE was a little
different; the NLE of B-2 was stronger than that of B-1. These
results indicated that the (R)- and (S)-Zr catalyst prepared
from each optically pure 1 could also form the inactive species
containing an equal amount of R and S ligands, easily. We
have already reported that the Zr catalyst prepared from
Zr(OtBu)4, ligand 1 and N-methylimidazole was efficient for
an asymmetric Mannich-type reaction, and elucidated that the
proposed structure of this catalyst was a monomeric form
according to DFT calculations, and NLE and NMR studies.10
Based on those experiments, we firstly assumed that the Zr
catalyst consisted of a 1 : 1 complex of Zr and ligand 1, and
that it mainly existed as a monomer. However, the NLE
experiments of method A, method B-1 and B-2 showed
different curves on the graphs, which meant that the inactive
RS dimer or oligomer complex gradually generated. Considering
those results, the current Zr complex could form a dimer or
oligomer structure in the solution state. The main difference
from the complex for the Mannich-type reaction is absence
of N-methylimidazole, a coordinative Lewis base, which
could control the whole aggregation structure. The assumed
structures of the Zr complexes are shown in Fig. 2, although, it
is still unclear whether the real active species as a catalyst is
monomeric, or a dimeric or oligomeric form of the complex.
In conclusion, we have found that the chiral Zr-tridentate
BINOL complex was successfully employed for meso-aziridine
ring-opening reactions with a variety of aniline derivatives in
good yields with high enantioselectivities. The N-benzhydryl
group on the product obtained was easily cleaved in high yield
under reductive conditions. Furthermore, it was revealed that
the zirconium complex could exist mainly in dimer or oligomer
form according to investigations of the relationship between
(f) P. Muller and P. Nury, Org. Lett., 1999, 1, 439.
¨
5 K. Arai, S. Lucarini, M. M. Salter, K. Ohta, Y. Yamashita and
S. Kobayashi, J. Am. Chem. Soc., 2007, 129, 8103.
6 (a) R. Yu, Y. Yamashita and S. Kobayashi, Adv. Synth. Catal.,
2009, 351, 147; recently, Schneider et al. also reported a chiral
Ti-catalyzed meso-aziridine ring opening reactions with aromatic
amines, see: (b) S. Peruncheralathan, H. Teller and C. Schneider,
Angew. Chem., Int. Ed., 2009, 48, 4849.
7 For examples of asymmetric Zr-catalyzed reactions, see:
(a) S. Kobayashi, M. Ueno, S. Saito, Y. Mizuki, H. Ishitani and
Y. Yamashita, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 5476;
(b) S. Yamasaki, M. Kanai and M. Shibasaki, Chem.–Eur. J., 2001,
7, 4066; (c) E.-I. Negishi, Pure Appl. Chem., 2001, 73, 239;
(d) S. Huo, J.-C. Shi and E.-I. Negishi, Angew. Chem., Int. Ed.,
2002, 41, 2141; (e) T. Okachi, N. Murai and M. Onaka, Org. Lett.,
2003, 5, 85.
8 T. W. Green and P. G. M. Wuts, in Protective groups in organic
synthesis, John Wiley & Sons, Inc., New York, 4th edn, 2007,
p. 824.
9 (a) K. Okamoto, R. Akiyama and S. Kobayashi, J. Org. Chem.,
2004, 69, 2871; (b) R. Akiyama and S. Kobayashi, J. Am. Chem.
Soc., 2003, 125, 3412.
10 Y. Ihori, Y. Yamashita, H. Ishitani and S. Kobayashi, J. Am.
Chem. Soc., 2005, 127, 15528.
ꢁc
This journal is The Royal Society of Chemistry 2009
5724 | Chem. Commun., 2009, 5722–5724