Catalytic asymmetric ring openings of Meso and terminal aziridines with halides mediated by chiral 1,2,3-triazolium silicates

Article abstract of DOI:10.1021/ja3028668

Catalytic asymmetric chloride and bromide ring openings of meso aziridines with trimethylsilyl halides have been developed using modular chiral 1,2,3-triazolium chlorides as catalysts. Control experiments suggest the reaction pathway involving hypervalent

Full text of DOI:10.1021/ja3028668

Communication
pubs.acs.org/JACS
Catalytic Asymmetric Ring Openings of Meso and Terminal
Aziridines with Halides Mediated by Chiral 1,2,3-Triazolium Silicates
Kohsuke Ohmatsu, Yuta Hamajima, and Takashi Ooi*
Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Nagoya, 464-8603, Japan
S
* Supporting Information
hard, basic anions such as fluoride or phenoxide ions, there are
two possible mechanistic scenarios (Scheme 1). One is the ion-
ABSTRACT: Catalytic asymmetric chloride and bromide
ring openings of meso aziridines with trimethylsilyl halides
have been developed using modular chiral 1,2,3-triazolium
chlorides as catalysts. Control experiments suggest the
reaction pathway involving hypervalent silicate ions as
reactive intermediates. The application of this system to
the efficient kinetic resolution of terminal aziridines is also
reported.
Scheme 1. Two Possible Reaction Pathways for Chloride
Ring Opening of Aziridines Catalyzed by Onium Salts (Q·X)
xtra-coordinated silicon species, called hypervalent sili-
E
cates, play important roles as reagents and intermediates in
synthetic organic chemistry.1 Owing to their enhanced
reactivity, significant progress has been made in the develop-
ment of catalytic asymmetric reactions that proceed through
the intermediacy of penta- or hexacoordinated silicates.2 One of
the attractive features of the silicate-based methodologies is the
allowance of asymmetric transformations with otherwise
difficult-to-use nucleophiles, such as halide ions,3,4 under mild
conditions, making it feasible to rapidly construct optically
active halogen-containing organic molecules.5 For instance, the
combined use of chlorosilanes and chiral Lewis base catalysts
offers the fruitful opportunity of employing chloride ions as
nucleophiles for asymmetric bond-forming processes.6,7 In fact,
since the pioneering work by Denmark and co-workers,6a
several elegant studies have been reported on the design of
chiral catalysts for chlorosilicate-mediated stereoselective
reactions. However, despite the potential synthetic utility of
halosilicates, their successful applications have been limited to
the asymmetric ring openings of meso epoxides using the
highly reactive silicon tetrachloride (SiCl4). Herein, we describe
the highly enantioselective chloride and bromide ring openings
of aziridines based on the use of trimethylsilyl halides as a
stoichiometric halide source in combination with chiral 1,2,3-
triazolium chlorides as catalysts.8 This new catalytic system is
effective not only for the desymmetrization of meso
aziridines4,9 but also for the kinetic resolution of terminal
aziridines,10,11 including the hitherto unknown kinetic reso-
lution of differently 2,2-disubstituted aziridines. In addition, this
study shows the importance of chiral triazolium halosilicates as
a reactive intermediate, thereby providing mechanistic insight
into ion-pair catalysis with silicon-based nucleophiles.
exchange reaction pathway, where the interaction of the silyl
compound with Lewis basic anions (X) leads to the exclusion
of Me3SiX, generating requisite yet reactive ion pairs. The other
mechanism involves the formation of onium silicate as a key
intermediate.1,13 Although the latter, silicate-mediated pathway
is conceivable, reactions of this type are often believed to
proceed through the ion-exchange pathway.14 If this ion-
exchange mechanism is operative in the chloride ring opening
of aziridines, onium chloride itself should be reactive enough to
effect C−N bond cleavage by the chloride ion transfer. To
examine this possibility, we attempted the reaction of N-p-tert-
butylbenzenesulfonyl aziridine 2a with L-phenylalanine-derived,
chiral 1,2,3-triazolium chloride 1a·Cl (see the scheme for Table
1).8 As shown in Scheme 2, treatment of 2a with 1 equiv of
1a·Cl in toluene at −40 °C for 12 h did not result in the
formation of a detectable amount of the desired ring-opening
product 3a. The exposure of 2a to a stoichiometric quantity of
trimethylsilyl chloride (Me3SiCl) led to an essentially
quantitative recovery of the starting material. In contrast, the
ring-opening reaction of 2a was promoted by the combined use
of a catalytic amount of 1a·Cl and Me3SiCl (1.2 equiv), giving
3a in 35% isolated yield even with an appreciable level of
enantioselectivity (65% ee). These results indicate that the
hypervalent chlorosilicate acts as a reactive intermediate in this
desymmetrization reaction. The intervention of the chlorosili-
cate was also supported by the 29Si NMR spectroscopic analysis
of a 1:1 mixture of 1a·Cl and Me3SiCl in toluene-d8; this
showed a single peak at δ = 6.6 ppm, which is considerably
shifted upfield from the original signal of Me3SiCl (29.5
Ring opening of N-protected aziridines with trimethylsilyl
chloride can be promoted by tetrabutylammonium fluoride
(TBAF), affording trans-β-chloroamine derivatives in nearly
quantitative yields.12 In the reactions of silicon-based
nucleophiles catalyzed by onium salts possessing relatively
Received: March 24, 2012
Published: May 14, 2012
© 2012 American Chemical Society
8794
dx.doi.org/10.1021/ja3028668 | J. Am. Chem. Soc. 2012, 134, 8794−8797

Journal of the American Chemical Society
Communication
modification of the structure of chiral triazolium ion 1 by
taking advantage of its modularity (Table 1).17 Initial screening
of the amide substituent (Ar1) revealed that the introduction of
a strongly electron-withdrawing p-nitrophenyl group (1b)
delivered a notable enhancement in enantioselectivity (entry
2). The property of the benzylic appendage at the triazolium
N(3) (Ar2) was also important for stereocontrol, and the p-
trifluoromethylphenyl group was optimal (1c, entry 3). This set
of aromatic substituents would impart an increased anion-
binding ability to the triazolium ion 1 rendering its ion pairing
with the silicate ion tight, which might be beneficial in terms of
attaining a higher asymmetric induction. On the basis of this
assumption, the electronic attribute of 1 was further tuned by
installing a p-chlorophenyl group as Ar3 (1d), leading to a slight
increase in the enantiomeric excess (entry 4). Furthermore, the
incorporation of a trifluoromethyl group into the para position
of the C(4) phenyl ring (1e) brought critical improvement in
the selectivity (entry 5). We then turned our attention to the
effect of the substituent attached to the stereogenic carbon of
the amino acid origin (R1) on the reaction profile. Interestingly,
the appropriate choice of parent α-amino acid appeared to be
crucial for eliciting the full potential of the catalyst, and 1g·Cl
prepared from L-cyclohexylalanine exerted excellent stereo-
controlling ability (entries 6 and 7). Throughout these fruitful
investigations, however, the conversion of aziridine 2a into the
product 3a remains insufficient; this seems to be associated
with the difficulty in facile regeneration of the active triazolium
chlorosilicate, probably because of the relatively weak Lewis
basicity of the p-tert-butylbenzenesulfonamide ion formed after
the ring opening. Therefore, we considered the use of alcoholic
additives to possibly facilitate this process by the generation of
alkoxide ions through a proton transfer. Fortunately, among the
several candidates examined, trimethylsilanol (Me3SiOH) was
identified to be suitable for increasing the catalyst turnover
without loss of enantioselectivity (entry 10).18 Eventually, the
addition of 1 equiv of Me3SiOH allowed the reaction to
proceed smoothly in the presence of 1g·Cl as a catalyst,
affording 3a in 85% yield with 95% ee (entry 11). It should be
added that the attempted reaction with triethylsilyl chloride
(Et3SiCl) instead of Me3SiCl proceeded slowly to give 3a in
moderate yield with slightly lower enantioselectivity (entry 12).
Use of tert-butyldimethylsilyl chloride (t-BuMe2SiCl) resulted
in no detectable formation of 3a (entry 13). The observed
dependence of the reactivity and selectivity on the structure of
silyl substituents was consistent with the intermediacy of the
silicate in the chloride ring-opening event.
Table 1. Screening of the Structure of Triazolium Ions and
Reaction Conditions for Enantioselective Chloride Ring
Opening of Meso Aziridine 2a
a
b
c
entry
1
additive
% yield
% ee
1
2
3
4
5
6
7
8
9
10
1a
1b
1c
1d
1e
1f
none
35
24
26
42
35
49
31
44
41
49
85
55
NA
65
81
84
87
92
93
95
89
94
95
95
90
NA
none
none
none
none
none
1g
1g
1g
1g
1g
1g
1g
none
MeOH
PhOH
Me3SiOH
Me3SiOH
Me3SiOH
Me3SiOH
d
11
e
12
f
13
a
Reactions were carried out with 0.10 mmol of 2a, 0.12 mmol of
b
Me3SiCl, and 5 mol % of 1·Cl in 1 mL of toluene. Isolated yield.
c
d
Determined by chiral HPLC analysis. Reaction was performed with
e
1 equiv of Me3SiOH. With 1.2 equiv of Et3SiCl instead of Me3SiCl.
f
With 1.2 equiv of t-BuMe2SiCl instead of Me3SiCl. NA, not available.
Scheme 2. Control Experiments for Chloride Ring Opening
of Meso Aziridine 2a
The optimized conditions thus established were used for
reactions with a variety of meso aziridines (Table 2). A near-
quantitative yield was obtained in the chlorination of 2a simply
by extending the reaction time (entry 1). A cyclohexene-
derived substrate 2b was also employable for this desymmet-
rization, in which the chiral triazolium salt 1f·Cl was found to
be more effective than 1g·Cl (entries 2 and 3). Other bicyclic
and simple aziridines were well accommodated, and high
stereoselectivities were uniformly observed (entries 4−6).19 It
is noteworthy that the present system was successfully applied
to the bromide ring opening of the aziridines using
trimethylsilyl bromide, providing the corresponding β-bromo-
amine derivatives in high chemical yields with excellent
enantiomeric excesses (entries 7 and 8).20
ppm).15 When tetrabutylammonium chloride was mixed with
an equimolar amount of Me3SiCl, a single peak was also
detected at δ = 6.7 ppm, thereby suggesting the generation of a
silicate species through the interaction of chloride ions with
Me3SiCl.16
Based on this preliminary information, we next focused on
improving the reaction efficiency and stereoselectivity of the
asymmetric chloride ring opening reaction of the meso
aziridine. For this purpose, we undertook a systematic
The potential utility of the chiral 1,2,3-triazolium chloride
1·Cl-catalyzed, asymmetric halide ring-opening protocol was
further demonstrated by its successful application to the kinetic
8795
dx.doi.org/10.1021/ja3028668 | J. Am. Chem. Soc. 2012, 134, 8794−8797

Journal of the American Chemical Society
Communication
which exhibited impressive levels of selectivity factors, enabling
the preparation of enantiomerically pure 6, otherwise not
readily accessible by conventional asymmetric methodologies
(Scheme 4).23 When 6a was exposed to reaction conditions
Table 2. Enantioselective Chloride and Bromide Ring
Openings of Meso Aziridines
a
Scheme 4. Kinetic Resolution of 2,2-Disubstituted Aziridines
6
similar to those of the kinetic resolution of 4, a regioisomeric
mixture of the chlorinated products was isolated in 48%
combined yield (7a/8a = 80:20) and the enantiomeric excess of
the major β-chloro-tert-amine derivative 7a was 90% ee.24
Importantly, the starting 6a was recovered in 46% yield in an
essentially enantiopure form. While the regioselectivity was
dependent on the substrate structure, an almost complete
discrimination of the two enantiomers of 6 was consistently
realized.
a
Unless otherwise noted, reactions were carried out with 0.10 mmol of
In conclusion, we developed highly enantioselective chloride
and bromide ring openings of meso aziridines with
trimethylsilyl halides catalyzed by chiral 1,2,3-triazolium
chlorides under mild conditions. The potential utility of the
asymmetric halide ring-opening strategy has also been
demonstrated by its application to the kinetic resolution of
racemic, terminal aziridines. These catalytic asymmetric
nucleophilic halogenations based on the use of trimethylsilyl
halides rely on the ability of the appropriately modified, chiral
1,2,3-triazolium chloride to generate the requisite halosilicates
and extend precise stereocontrol over the halide ion transfer.
We believe that the present study adds a new dimension to the
development of silicate-based asymmetric methodologies.
2, 0.12 mmol of trimethylsilyl halide, 0.10 mmol of Me3SiOH, and 5
b
mol % of 1·Cl in 1 mL of toluene at −40 °C. Isolated yield.
c
Determined by chiral HPLC analysis. Absolute configuration of the
product 3b was determined after conversion to the known compound,
and the stereochemistries of other examples were assumed by analogy.
d
For details, see the Supporting Information. With 10 mol % of 1f·Cl.
e
Reaction was performed at −78 °C.
resolution of racemic, terminal aziridines 4 (Scheme 3). For
instance, the reaction of the vinylcyclohexane-derived aziridine
Scheme 3. Kinetic Resolution of Terminal Aziridines 4
ASSOCIATED CONTENT
* Supporting Information
■
S
Representative experimental procedures, spectral and analytical
data for all new compounds, and crystallographic data for 7a
(CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
tooi@apchem.nagoya-u.ac.jp
■
4a with Me3SiCl (0.55 equiv) in the presence of 1f·Cl (5 mol
%) and Me3SiOH (0.55 equiv) in toluene at −40 °C gave rise
to the ring-opening product 5a exclusively in 51% yield with
80% ee, and 4a was recovered in 44% yield with 94% ee
(selectivity factor: s = 23).21 Even higher selectivities were
observed with terminal aziridines bearing sterically demanding
tert-butyl and trimethylsilyl substituents (4b and 4c).22
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Funding Program for Next
Generation World-Leading Researchers, the Grant-in-Aid for
Young Scientists (B) from JSPS, the Global COE Program in
Chemistry of Nagoya University, and Program for Leading
Another distinct feature of this approach is an unprecedented
kinetic resolution of differently 2,2-disubstituted aziridines 6,
8796
dx.doi.org/10.1021/ja3028668 | J. Am. Chem. Soc. 2012, 134, 8794−8797

Journal of the American Chemical Society
Communication
(10) For palladium-catalyzed asymmetric alcoholysis of terminal
aziridines, see: Leung, W.-H.; Mak, W.-L.; Chan, E. Y. Y.; Lam, T. C.
H.; Lee, W.-S.; Kwong, H.-L.; Yeung, L.-L. Synlett 2002, 1688.
(11) For regiodivergent kinetic resolution of terminal aziridines with
trimethylsilyl azide, see: Wu, B.; Parquette, J. R.; RajanBabu, T. V.
Science 2009, 326, 1662.
(12) Wu, J.; Hou, X.-L.; Dai, L.-X. J. Org. Chem. 2000, 65, 1344.
(13) Gawronski, J.; Wascinska, N.; Gajewy, J. Chem. Rev. 2008, 108,
5227.
Graduate Schools “Integrative Graduate Education and
Research Program in Green Natural Sciences” in Nagoya
University.
REFERENCES
■
(1) Chuit, C.; Corriu, R. J. P.; Reye, C.; Young, J. C. Chem. Rev. 1993,
93, 1371.
(2) (a) Rendler, S.; Oestreich, M. Synthesis 2005, 1727. (b) Orito, Y.;
Nakajima, M. Synthesis 2006, 1391. (c) Benaglia, M.; Guizzetti, S.;
Pignataro, L. Coord. Chem. Rev. 2008, 252, 492.
(14) For selected examples, see: (a) Michida, M.; Mukaiyama, T.
Chem. Lett. 2008, 37, 890. (b) Michida, M.; Mukaiyama, T. Chem.
Asian J. 2008, 3, 1592.
(3) For examples of asymmetric ring openings of epoxides with
stoichiometric or catalytic amounts of chiral Lewis acid halides, see:
(a) Joshi, N. N.; Srebnik, M.; Brown, H. C. J. Am. Chem. Soc. 1988,
110, 6246. (b) Naruse, Y.; Esaki, T.; Yamamoto, H. Tetrahedron 1988,
44, 4747. (c) Nugent, W. A. J. Am. Chem. Soc. 1998, 120, 7139.
(d) Bruns, S.; Haufe, G. Tetrahedron: Asymmetry 1999, 10, 1563. For
recent reports on asymmetric nucleophilic fluorinations, see:
(e) Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132, 3268.
(f) Katcher, M. H.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132, 17402.
For asymmetric dichlorination of cyclopropanecarbaldehyde, see:
(g) Sparr, C.; Gilmour, R. Angew. Chem., Int. Ed. 2011, 50, 8391.
(4) For a single example of asymmetric chloride ring opening of
meso aziridines, in which hydrogen chloride was employed as a
chloride source, see: Mita, T.; Jacobsen, E. N. Synlett 2009, 1680.
(5) (a) Ibrahim, H.; Togni, A. Chem. Commun. 2004, 1147. (b) Ueda,
M.; Kano, T.; Maruoka, K. Org. Biomol. Chem. 2009, 7, 2005.
(c) Shibatomi, K. Synthesis 2010, 2679.
(6) (a) Denmark, S. E.; Barsanti, P. A.; Wong, K.-T.; Stavenger, R. A.
J. Org. Chem. 1998, 63, 2428. (b) Tao, B.; Lo, M. M.-C.; Fu, G. C. J.
Am. Chem. Soc. 2001, 123, 353. (c) Nakajima, M.; Saito, M.; Uemura,
M.; Hashimoto, S. Tetrahedron Lett. 2002, 43, 8827. (d) Paek, S. H.;
Shim, S. C.; Cho, C. S.; Kim, T.-J. Synlett 2003, 849. (e) Tokuoka, E.;
Kotani, S.; Matsunaga, H.; Ishizuka, T.; Hashimoto, S.; Nakajima, M.
Tetrahedron: Asymmetry 2005, 16, 2391. (f) Takenaka, N.;
Sarangthem, R. S.; Captain, B. Angew. Chem., Int. Ed. 2008, 47,
9708. (g) Chelucci, G.; Baldino, S.; Pinna, G. A.; Benaglia, M.; Buffa,
L.; Guizzetti, S. Tetrahedron 2008, 64, 7574. (h) Pu, X.; Qi, X.; Ready,
J. M. J. Am. Chem. Soc. 2009, 131, 10364. (i) Malkov, A. V.; Gordon,
(15) For details, see the Supporting Information.
(16) (a) Damrauer, R.; Burggraf, L. W.; Davis, L. P.; Gordon, M. S. J.
Am. Chem. Soc. 1988, 110, 6601. (b) Hao, C.; Kaspar, J. D.; Check, C.
E.; Lobring, K. C.; Gilbert, T. M.; Sunderlin, L. S. J. Phys. Chem. A
2005, 109, 2026.
(17) The reactions with aziridine bearing p-toluenesulfonyl group
(Ts) exhibited a similar level of reaction efficiency and slightly lower
enantioselectivity. For details, see Table S1 in the Supporting
Information.
(18) To interrogate the possibility of concomitant generation of
hydrogen chloride (HCl) and its participation as a reactive promoter,
the reaction of 2a was conducted with 10 mol% of a proton scavenger,
2,6-di-tert-butylpyridine, under otherwise identical conditions with
those of entry 7 in Table 1; this revealed that the differences in
chemical yield and ee were marginal and thus HCl was irrelevant to
the present catalytic system, even if it was generated.
(19) The present system seems to be ineffective for aziridines having
aromatic substituents. For instance, the reaction of cis-stilbene-derived
aziridine proceeded sluggishly even at 0 °C, resulting in the formation
of the desired product in 14% yield with 48% ee after 24 h.
(20) The chlorinated product was not obtained even in the reaction
with 30 mol% of 1f·Cl.
(21) S = kfast/kslow = In[(1 − C/100)(1 − ee/100)]/In[(1 − C/100)
(1 + ee/100)] (C = conversion, ee = enantiomeric excess of recovered
4 or 6).
(22) The ring opening of styrene-derived aziridine showed an
opposite regioselectivity and diminished stereoselectivity. For details,
see the Supporting Information.
(23) (a) Morton, D.; Pearson, D.; Fielda, R. A.; Stockman, R. A.
Chem. Commun. 2006, 1833. (b) Colpaert, F.; Mangelinckx, S.;
Leemans, E.; Denolf, B.; De Kimpe, N. Org. Biomol. Chem. 2010, 8,
3251.
(24) The absolute configuration of 7a was determined to be R by X-
ray diffraction analysis. The minor regioisomers 8 were also obtained
in enantioenriched form (8a: 64% ee, 8b: 62% ee, 8c: 80% ee). These
products would arise from the stereo-invertive ring opening of (R)-
aziridines, and the stereochemistry of 8 shown in Scheme 4 is based on
this assumption. For stereo-invertive nucleophilic substitution at the
disubstituted carbon of aziridine, see: Forbeck, E. M.; Evans, C. D.;
Gilleran, J. A.; Li, P.; Joullie, M. M. J. Am. Chem. Soc. 2007, 129, 14463.
̂ ̂
́
M. R.; Stoncius, S.; Hussain, J.; Kocovsky, P. Org. Lett. 2009, 11, 5390.
(j) Denmark, S. E.; Barsanti, P. A.; Beutner, G. L.; Wilson, T. W. Adv.
Synth. Catal. 2007, 349, 567.
(7) For asymmetric ring opening of terminal aziridine using a
stoichiometric amount of chiral amine and trimethysilyl cyanide or
iodide, see: Minakata, S.; Murakami, Y.; Satake, M.; Hidaka, I.; Okada,
Y.; Komatsu, M. Org. Biomol. Chem. 2009, 7, 641.
(8) Ohmatsu, K.; Kiyokawa, M.; Ooi, T. J. Am. Chem. Soc. 2011, 133,
1307.
(9) For representative references of desymmetrization of aziridines
with azide, see: (a) Li, Z.; Fernandez, M.; Jacobsen, E. N. Org. Lett.
́
1999, 1, 1611. (b) Fukuta, Y.; Mita, T.; Fukuda, N.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2006, 128, 6312. (c) Rowland, E. B.;
Rowland, G. B.; Rivera-Otero, E.; Antilla, J. C. J. Am. Chem. Soc. 2007,
129, 12084. (d) Nakamura, S.; Hayashi, M.; Kamada, Y.; Sasaki, R.;
Hiramatsu, Y.; Shibata, N.; Toru, T. Tetrahedron Lett. 2010, 51, 3820.
With cyanide, see: (e) Mita, T.; Fujimori, I.; Wada, R.; Wen, J.; Kanai,
M.; Shibasaki, M. J. Am. Chem. Soc. 2005, 127, 11252. (f) Fujimori, I.;
Mita, T.; Maki, K.; Shiro, M.; Sato, A.; Furusho, S.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2006, 128, 16438. (g) Wu, B.; Gallucci,
J. C.; Parquette, J. R.; RajanBabu, T. V. Angew. Chem., Int. Ed. 2009,
48, 1126. With other nucleophiles, see: (h) Muller, P.; Nury, P. Org.
̈
Lett. 1999, 1, 439. (i) Arai, K.; Lucarini, S.; Salter, M. M.; Ohta, K.;
Yamashita, Y.; Kobayashi, S. J. Am. Chem. Soc. 2007, 129, 8103.
(j) Della Sala, G.; Lattanzi, A. Org. Lett. 2009, 11, 3330. (k) Larson, S.
E.; Baso, J. C.; Li, G.; Antilla, J. C. Org. Lett. 2009, 11, 5186. (l) Xu, Y.;
Lin, L.; Kanai, M.; Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc. 2011,
133, 5791.
8797
dx.doi.org/10.1021/ja3028668 | J. Am. Chem. Soc. 2012, 134, 8794−8797