Guanidine-catalyzed enantioselective desymmetrization of meso-aziridines

Article abstract of DOI:10.1039/c0cc05840h

An amino-indanol derived chiral guanidine was developed as an efficient Bronsted base catalyst for the desymmetrization of meso-aziridines with both thiols and carbamodithioic acids as nucleophiles, which provided 1,2-difunctionalized ring-opened products

Full text of DOI:10.1039/c0cc05840h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 3897–3899
www.rsc.org/chemcomm
COMMUNICATION
Guanidine-catalyzed enantioselective desymmetrization
of meso-aziridinesw
Yan Zhang,a Choon Wee Kee,a Richmond Lee,b Xiao Fu,a Julian Ying-Teck Soh,a
Esther Mun Foong Loh,a Kuo-Wei Huang*b and Choon-Hong Tan*a
Received 30th December 2010, Accepted 2nd February 2011
DOI: 10.1039/c0cc05840h
An amino-indanol derived chiral guanidine was developed as an
efficient Brønsted base catalyst for the desymmetrization of
meso-aziridines with both thiols and carbamodithioic acids as
nucleophiles, which provided 1,2-difunctionalized ring-opened
products in high yields and enantioselectivities.
Preliminary studies showed that the ring opening of
aziridine 2a with thiophenol can be catalyzed with 10 mol%
of guanidines 1a–d (Table 1, entries 1–4). The trans-product 4a
was isolated in 33% ee in the presence of 1b as a catalyst when
the reaction was conducted in ether at À20 1C. Different
N-protected aziridines were investigated with catalyst 1b
(entries 5 and 6). The enantioselectivity was slightly improved
when a 3,5-dinitrobenzoyl group was used and amongst
arylthiols that were tested, 2,6-dichlorobenzenethiol provided
the desired product in 84% ee (entries 7–9). This result was
further improved by decreasing the catalyst loading. When
2 mol% of catalyst was used in the desymmetrization, product
4f was obtained in 92% ee with 92% yield (entry 11). A
loading of 1 mol% catalyst gave the same enantioselectivity,
but at the expense of longer reaction time.
Aziridines and their ring-opened products are important
intermediates in synthetic organic chemistry.1 Asymmetric
desymmetrization of meso-aziridines has attracted much atten-
tion as it can generate enantiomerically pure 1,2-difunctionalized
products in a single step. Although some highly enantio-
selective desymmetrizations of meso-aziridines catalyzed by
chiral metal complexes have been developed,2 there are fewer
organocatalytic examples. Chiral phosphoric acids were reported
to be effective catalysts for the asymmetric ring opening of
meso-aziridines with TMSN3,3 TMS–SPh4 and thiols.5 Brønsted
base strategies using Cinchona alkaloid derivatives6,7 and prolinols8
have been recently exploited but with less success.
Density functional theory (DFT) calculations10 were performed
to elucidate the origin of enantioselectivity for the reaction
between aziridine 2c and thiol (Ar = 2,6-Cl2C6H3). Two
plausible mechanisms can be envisaged (Fig. 2a).
Our group has demonstrated the use of chiral bicyclic
guanidines as Brønsted base catalysts in several enantioselective
reactions.9 Recently, we are trying to design guanidines that
can be more easily accessible.9c In this communication, we
reported an amino-indanol derived guanidine 1b (Fig. 1),
which can be synthesized in two simple steps (see ESIw).
This catalyst was shown to be an excellent catalyst for the
desymmetrization of meso-aziridines with sulfur nucleophiles.
By comparing the lowest energy conformation located for
the relevant transition state (TS) structures, the absolute
Table 1 Desymmetrization of meso-aziridines 2a–c catalyzed by
guanidines 1a–d
Entry Catalyst 2 Ar
x mol% 4 Yielda (%) eeb (%)
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1b
1b
1b
1b
1b
1b
1b
1b
2a Ph
2a Ph
2a Ph
2a Ph
2b Ph
2c Ph
2c 2,6-Me2C6H3 10
2c 4-BrC6H4 10
2c 2,6-Cl2C6H3 10
10
10
10
10
10
10
4a 94
4a 94
4a 92
4a 89
4b 78
4c 95
4d 92
4e 92
4f 93
4f 93
4f 92
4f 92
4
33
2
Fig. 1 Amino-indanol derived chiral guanidines.
2
34
44
67
59
84
89
92
92
a Department of Chemistry, 3 Science Drive 3,
National University of Singapore (NUS), 117543 Singapore,
Singapore. E-mail: chmtanch@nus.edu.sg; Fax: +65 6779 1691;
Tel: +65 6516 2845
9c
10
11
12d
2c 2,6-Cl2C6H3
2c 2,6-Cl2C6H3
2c 2,6-Cl2C6H3
5
2
1
b KAUST catalysis Center and Division of Chemicals and Life
Sciences and Engineering, King Abdullah University of Science and
Technology, Thuwal 23955-6900, Kingdom of Saudi Arabia.
E-mail: hkw@kaust.edu.sa
w Electronic supplementary information (ESI) available. CCDC
806368–806369. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c0cc05840h
a
b
Isolated yield. Determined by chiral HPLC analysis on a chiral phase.
c The absolute configuration of 4f was determined by X-ray analysis (see
the ESIw for details). d The reaction was complete after 40 hours.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3897–3899 3897

View Article Online
Table 2 Desymmetrization of N-3,5-dinitrobenzoyl aziridines 2c–i
catalyzed by guanidine 1b
Entry 2 (R = 3,5-dinitrobenzoyl)
4
x mol% Yielda (%) eeb (%)
1c
2
2c
2d
2e
4f
1
92
94
93
94
94
90
4g 1
4h 2
3
4
5
6
2f
4i
4j
5
5
91
90
95
90
Fig.
2
(a) Possible mechanisms for aziridine-opening reaction.
DE = ESS À ERR (kcal molÀ1), where E is the electronic energy at
M06-2X/6-311+G(2df,2p)//B3LYP/6-31G(d,p). (b) Lowest energy TS
structure for (R,R) product for mechanism 2 optimized at RB3LYP/
2g
6-31G(d,p). On the right is the ChemDraw representation of TSRR
.
2h
2i
4k 5
93
91
88
93
configuration of 4f, that was determined by single-crystal
diffraction, could be predicted correctly with both models.
The (R,R) TS structure is lower in energy relative to the (S,S)
one in both mechanisms: 1.2 kcal molÀ1 for mechanism 1 and
2.3 kcal molÀ1 for mechanism 2.
7
4l 2
a
b
Isolated yield. Determined by chiral HPLC analysis on a chiral
c
phase. The reaction was performed at À50 1C.
Both mechanisms were further tested in the prediction of the
trend in enantioselectivity observed in Table 1 by using different
thiols (Ar = Ph and Ar = 4-BrC6H4). Mechanism 1 predicted
an increase in enantioselectivity relative to Ar = 2,6-Cl2C6H3
ring-opened products with good ees and yields (Scheme 1).
Absolute configuration of 6a is determined by X-ray crystallo-
graphic analysis (see ESIw).
for both cases (DEAr=Ph = 2.4 kcal molÀ1; DEAr=4-BrC6H4
=
Dithiocarbamates with substitutions at the thiol chain are
compounds with potential biological activity.14a One of the
most convenient approaches to these compounds is the genera-
tion of the carbamodithioic acid from an amine and carbon
disulfide, followed by the ring opening of an epoxide.14 However,
no variant with aziridines was reported. To further extend the
scope of our methodology, we studied the desymmetrization of
N-3,5-dinitrobenzoyl meso-aziridines with in situ generated
carbamodithioic acid. This three component reaction was
investigated with amines, and bis(2-methoxybenzyl)-amine
was found to provide the best enantioselectivity. The results
were less ideal when we used carbamodithioic acid that was
prepared separately. The reactions of various aziridines with
different ring-size and acyclic aziridine were investigated.
Good enantioselectivities and yields were observed for bicyclic
aziridines containing six- and five-membered rings (Table 3,
entries 1–3). A lower ee and much lower yield were observed
when aziridine 2f with a seven-membered ring was used (entry 4).
With 20 mol% 1b as a catalyst, acyclic meso-aziridine 2h
provided the ring-opened product with high enantioselectivity
and yield (entry 5). Excellent optical purity of the products
could be easily obtained through a single recrystallization.
The enantiopure 1,2-difunctionalized products obtained can
be readily transformed into valuable compounds (Scheme 2).
For instance, the ring-opened product 4f (95% ee) could be easily
oxidized to a sulfoxide intermediate as a single diastereoisomer
2.6 kcal molÀ1
)
which contradicts experimental results.
Mechanism 2 is able to predict qualitatively the observed
decrease in enantioselectivity for both substituted thiols
(DEAr=Ph = 0 kcal molÀ1; DEAr=4-BrC6H4 = 0.7 kcal molÀ1).
The TS structure of mechanism 2 in Fig. 2b provides a basis
for rationalizing the enantioselectivity observed. Despite the
much shorter hydrogen bond in TSSS, it is less stable than
TSRR. This is ascribed to larger steric repulsion which arises
from bringing the aziridine and thiol closer to the catalyst. The
large increase in enantioselectivity when the phenyl group of
the thiol is substituted with chlorine at the 2- and 6-positions
can be attributed to an increase in steric repulsion between the
two chloro-substituents and neighbour atoms of the catalyst–
aziridine complex. The number of neighbouring atoms within
5 A to both chloro-substituents is 29 in TSSS (average: 4.1 A)
and 25 in TSRR (average: 4.3 A).11 As chlorine (1.75 A) has a
larger van der Waals radius relative to hydrogen (1.20 A),12
destabilization of TSSS relative to TSRR will be more signifi-
cant in Ar = 2,6-Cl2C6H3 than Ar = Ph and contributes to
the observed increase in ees.
The ring opening of aziridine-2,3-dicaboxylates is a direct
synthetic approach towards b-substituted aspartates.13 Catalyst
1b proved to be an efficient catalyst for the desymmetrization
of the N-tosyl meso-aziridine 5 with thiols. Both 2,6-dichloro-
benzenethiol and 2,6-dimethylbenzenethiol provided the
c
3898 Chem. Commun., 2011, 47, 3897–3899
This journal is The Royal Society of Chemistry 2011

View Article Online
NaOH and finally oxidized with performic acid to form the
zwitterionic chiral b-amino sulfonic acid.
In summary, we have developed an easy, efficient catalyst 1b
for highly enantioselective desymmetrization of meso-aziridines.
This is the first report on the use of carbamodithioic acid
as a nucleophile for asymmetric ring opening of aziridines.
Compounds that are useful in medicinal chemistry such as
chiral allylic amides and b-amino sulfonic acids could be
readily synthesized from the ring-opened products.
Scheme 1 Desymmetrization of cis-aziridine-2,3-dicaboxylate 5.
Table 3 Desymmetrization of N-3,5-dinitrobenzoyl aziridines 2c–h
with amine and CS2
Notes and references
1 For recent reviews, see: (a) D. Tanner, Angew. Chem., Int. Ed.
Engl., 1994, 33, 599; (b) A. K. Yudin, Aziridines and Epoxides in
Organic Synthesis, Wiley-VCH, Weinheim, 2006; (c) X. E. Hu,
Tetrahedron, 2004, 60, 2701.
2 For selected examples, see: (a) Z. Li, M. Fernandez and
E. N. Jacobsen, Org. Lett., 1999, 1, 1611; (b) T. Mita,
I. Fujimori, R. Wada, J. Wen, M. Kanai and M. Shibasaki,
J. Am. Chem. Soc., 2005, 127, 11252; (c) K. Arai, S. Lucarini,
M. M. Salter, K. Ohta, Y. Yamashita and S. Kobayashi, J. Am.
Chem. Soc., 2007, 129, 8103; (d) B. Wu, J. C. Gallucci,
J. R. Parquette and T. V. RajanBabu, Angew. Chem., Int. Ed.,
2009, 48, 1126; For recent review, see: (e) C. Schneider, Angew.
Chem., Int. Ed., 2009, 48, 2082.
3 E. B. Rowland, G. B. Rowland, E. Rivera-Otero and J. C. Antilla,
J. Am. Chem. Soc., 2007, 129, 12084.
4 G. Della Sala and A. Lattanzi, Org. Lett., 2009, 11, 3330.
5 S. E. Larson, J. C. Baso, G. Li and J. C. Antilla, Org. Lett., 2009,
11, 5186.
Entry 2 (R = 3,5-dinitrobenzoyl)
7
Yielda (%) eeb (%)
1c
2
2c
2d
2e
7a 98
7b 98
7c 80
89 (96)
85 (91)
84 (95)
3
4
2f
7d 67
7e 91
80 (90)
90 (98)
6 Z.-B. Luo, X.-L. Hou and L.-X. Dai, Tetrahedron: Asymmetry,
2007, 18, 443.
7 Z. Wang, X. Sun, S. Ye, W. Wang, B. Wang and J. Wu,
Tetrahedron: Asymmetry, 2008, 19, 964.
5d
2h
8 A. Lattanzi and G. D. Sala, Eur. J. Org. Chem., 2009, 1845.
9 For our recent work on bicyclic guanidine, see: (a) D. Leow and
C.-H. Tan, Chem. –Asian J., 2009, 4, 488; (b) Z. Jiang, Y. Pan,
Y. Zhao, T. Ma, R. Lee, Y. Yang, K.-W. Huang, M. W. Wang and
C.-H. Tan, Angew. Chem., Int. Ed., 2009, 48, 3627; (c) X. Fu,
W.-T. Loh, Y. Zhang, T. Chen, T. Ma, H. Liu, J. Wang and
C.-H. Tan, Angew. Chem., Int. Ed., 2009, 48, 7387; (d) H. Liu,
D. Leow, K.-W. Huang and C.-H. Tan, J. Am. Chem. Soc., 2009,
131, 7212; (e) S. Lin, D. Leow, K.-W. Huang and C.-H. Tan,
Chem. –Asian J., 2009, 4, 1741; (f) Y. Pan, Y. Zhao, T. Ma,
Y. Yang, H. Liu, Z. Jiang and C.-H. Tan, Chem. –Eur. J., 2010, 16,
779; (g) D. Leow and C.-H. Tan, Synlett, 2010, 1589; For recent
contributions from other groups on guanidine catalysts, (h) Z. Yu,
X. Liu, L. Zhou, L. Lin and X. Feng, Angew. Chem., Int. Ed., 2009,
48, 5195; (i) H. Ube, M. Shimada and M. Terada, Angew. Chem.,
Int. Ed., 2010, 49, 1858; (j) T. Misaki, G. Takimoto and
T. Sugimura, J. Am. Chem. Soc., 2010, 132, 6286; (k) S. Dong,
X. Liu, X. Chen, F. Mei, Y. Zhang, B. Gao, L. Lin and X. Feng,
J. Am. Chem. Soc., 2010, 132, 10650.
a
b
Isolated yield. Determined by chiral HPLC analysis on a chiral
c
phase; ees after recrystallization are reported in parentheses. The
d
reaction was performed at À50 1C. The reaction was carried out with
20 mol% of catalyst 1b.
10 M. J. Frisch, et al., Gaussian 09, revision A.02, Gaussian, Inc.,
Wallingford, CT, 2009.
11 Discovery Studio, version 2.5.5.9350, Accelrys, San Diego, CA, 2009.
12 A. Bondi, J. Phys. Chem., 1964, 68, 441.
Scheme 2 Preparation of allylic amide 9 and b-amino sulfonic acid 11.
13 L. Antolini, M. Bucciarelli, E. Caselli, P. Davolli, A. Forni,
I. Moretti, F. Prati and G. Torre, J. Org. Chem., 1997, 62, 8784.
14 (a) S. K. Kumar, L. Kumar, V. L. Sharma, A. Jain, R. K. Jain,
J. P. Maikhuri, M. Kumar, P. K. Shukla and G. Gupta, Eur. J.
Med. Chem., 2008, 43, 2247; (b) M. Karikomi, T. Yamazaki and
T. Toda, Chem. Lett., 1993, 1965; (c) A. Ziyaei-Haimjani and
M. R. Saidi, Can. J. Chem., 2006, 84, 1515; (d) B. C. Ranu, A. Saha
and S. Banerjee, Eur. J. Org. Chem., 2008, 519.
with high yield. Pyrolysis15 then afforded the chiral allylic
amide, which is difficult to synthesize via other means. In
addition, oxidative cleavage of the CQC bond allows access to
a-amino acid of high optical purity. The ring-opened product
7a (96% ee) was subjected to Boc protection, followed by
removal of the 3,5-dinitrobenzoyl group with 6 M aqueous
15 V. Kesavan, B.-D. Daniele and J.-P. Bgeue, Tetrahedron Lett.,
2000, 41, 2895.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 3897–3899 3899