In situ assembled boronate ester assisted chiral carboxylic acid catalyzed asymmetric trans-aziridinations

Article abstract of DOI:10.1021/ja407764u

We developed herein a new chiral Bronsted acid catalyst which is composed of two independent organic molecules, a chiral diol, and 2-boronobenzoic acid. In situ formation of a boronate ester was utilized as a key process to generate an active catalyst. This boronate ester assisted chiral carboxylic acid catalyst was successfully applied to the trans-aziridination of N-Boc and N-benzyl imines with N-phenyldiazoacetamide. This is the first catalyst to achieve high enantioselectivities using N-benzyl imines.

Full text of DOI:10.1021/ja407764u

Communication
pubs.acs.org/JACS
In Situ Assembled Boronate Ester Assisted Chiral Carboxylic Acid
Catalyzed Asymmetric Trans-Aziridinations
Takuya Hashimoto, Alberto Osuna Galvez, and Keiji Maruoka*
́
Department of Chemistry, Graduate School of Science, Kyoto University, Sakyo, Kyoto 606-8502, Japan
S
* Supporting Information
making it easier to identify an optimal catalyst as is often the
case for chiral metal catalysts.
As an actual catalyst system to investigate, we set our focus
on the combination of a readily available chiral diol and 2-
boronobenzoic acid which are assembled in situ by the
formation of a boronate ester (Figure 2).12,13 This is expected
ABSTRACT: We developed herein a new chiral Brønsted
acid catalyst which is composed of two independent
organic molecules, a chiral diol, and 2-boronobenzoic acid.
In situ formation of a boronate ester was utilized as a key
process to generate an active catalyst. This boronate ester
assisted chiral carboxylic acid catalyst was successfully
applied to the trans-aziridination of N-Boc and N-benzyl
imines with N-phenyldiazoacetamide. This is the first
catalyst to achieve high enantioselectivities using N-benzyl
imines.
tronger chiral Brønsted acid catalysis has become a major
Figure 2. Concept of in situ assembled boronate ester assisted chiral
carboxylic acid catalyst.
S
research field in the past 10 years pioneered by Akiyama
and Terada’s invention of chiral phosphoric acid catalysis.1−3 As
is the case for all kinds of asymmetric catalysis, this research
area has grown in synergy with the evolution of new catalysts
having distinct acidities and chiral environments, such as N-
triflyl phosphoramides by Yamamoto,4 axially chiral dicarbox-
ylic acids by us,5 and imidodiphosphoric acids by List,6 among
others.7,8
These chiral Brønsted acids are in common single molecules,
consisting of a conjugate base as a chiral scaffold and a proton
as a catalytically active site. Since the proton cannot be
electronically and sterically modified, the acidity, thus the
reactivity, and the chirality of a catalyst are both dictated by the
basicity and the structure of its conjugate base (Figure 1a). Our
to form a boronate ester assisted chiral carboxylic acid catalyst
wherein the carboxylic acid coordinates to the Lewis acidic
boron atom. This coordination would help to rigidify the
structure of the catalyst and enhance the acidity of the
carboxylic acid.5 Actually, the enhanced acidity of preformed
achiral 2-Bpin benzoic acid was disclosed by Mattson during
the course of our study.14 Since a variety of chiral diols are
readily available, such a catalyst system could easily offer diverse
chiral environments. In addition, use of electronically modified
2-boronobenzoic acids, enabled by the attachment of functional
groups on the aromatic ring, is expected to directly change the
acidity of the catalyst.
We report herein the development of an in situ assembled
boronate ester assisted chiral carboxylic acid whose catalytic
performance was proven in the catalytic asymmetric trans-
aziridination of N-Boc imines and N-phenyldiazoacetamide.15,16
Furthermore, by taking advantage of the facile electronic tuning
of the catalyst, we applied this catalyst to the yet-unrealized
trans-aziridination of N-benzyl imines wherein the use of a
stronger acid catalyst became crucial for the successful
implementation. It is also worth emphasizing that we succeeded
in establishing a rare catalyst independent of a binaphthyl unit,
of which most stronger chiral Brønsted acids are composed.17
As a model reaction to evaluate this concept, we chose trans-
aziridination of N-Boc imines and N-phenyldiazoacetamide
developed in this laboratory (Table 1).15 The actual reactions
were conducted as follows. In a flask containing molecular
sieves were added 10 mol % of 2-boronobenzoic acid 1a and 12
Figure 1. Classification of chiral Brønsted acids.
idea is to split a chiral Brønsted acid into two organic
molecules, a chiral ligand which mainly controls the chiral
environment and an achiral Brønsted acid which governs the
acidity of the catalyst, and assemble these two in situ (Figure
1b).9−11 In such a system, each component can be tuned
independently to quickly generate a diverse array of chiral
Brønsted acid catalysts having different catalytic properties,
Received: July 31, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja407764u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
of 3,3′-disubstituted BINOLs such as L failed to give high
enantioselectivity (entry 10), presumably due to the poor
ability of less basic BINOLs to form a boronate ester complex,
resulting in the competitive reaction catalyzed by free 1a.
With the optimized reaction conditions in hand, we then
examined the substrate scope of this catalytic asymmetric trans-
aziridination (Table 2). Use of aromatic imines bearing 3- or 4-
Table 1. Optimization of the Catalyst
a
Table 2. N-Boc Imine Scope
c
d
entry
R1
% yield
% ee
c
d
entry
diol
°C
% yield
% ee
1
2
3
4
5
6
7
8
9
Ph
64 (6a)
44 (6b)
42 (6c)
56 (6d)
54 (6e)
42 (6f)
68 (6g)
77 (6h)
54 (6i)
92
89
90
92
93
90
90
91
89
1
2
2
0
44
50
56
52
51
50
40
63
64
45
0
4-tolyl
3-tolyl
3
0
44
3
4
0
−61
−60
−70
−70
47
4-FC6H4
4
5a
5b
5c
5d
5e
5e
L
0
3-ClC6H4
4-MeO2CC6H4
3-MeOC6H4
4-PivOC6H4
2-Np
5
0
6
0
7
0
8
0
89
9
−20
0
92
a
Performed with N-Boc imine (0.10 mmol), N-phenyldiazoacetamide
(0.12 mmol), 5e (0.012 mmol), 1a (0.010 mmol), and MS4A (100
mg) in CH2Cl2 (1.0 mL). Determined by H NMR analysis of the
10
14
a
Performed with N-Boc imine (0.10 mmol), N-phenyldiazoacetamide
(0.12 mmol), chiral diol (0.012 mmol), 1a (0.010 mmol), and MS4A
(100 mg) in CH2Cl2 (1.0 mL). Determined by H NMR analysis of
the crude material. Isolated yield. Determined by chiral HPLC
analysis.
b
1
c
d
b
1
crude material. Isolated yield. Determined by chiral HPLC analysis.
c
d
substituents was tolerated to give the corresponding trans-
aziridines 6b−6h with modest yields and high enantioselectiv-
ities (entries 2−8). Whereas a 3-methoxybenzaldehyde derived
imine could be utilized to give the aziridine 6g in good yield
with high enantioselectivity (entry 7), the use of a 4-
methoxybenzaldehyde derived imine gave a complex mixture
(data not shown). To circumvent this issue, a 4-pivaloxy
substituted imine was used as a less electron-rich alternative
(entry 8). A 2-naphthaldehyde derived imine could be utilized
as well (entry 9). 2-Substituted aromatic imines and aliphatic
imines were not applicable, showing limitations of this
methodology.
With the success of catalytic asymmetric trans-selective
aziridination of N-Boc imines using the in situ assembled
boronate ester assisted chiral carboxylic acid, we turned our
attention to the use of N-benzyl imines. Compared with N-Boc
imines, N-benzyl imines are less electrophilic and less reactive
in nature while their facile synthesis and cleavage of the benzyl
group offer synthetic advantages. Although Wulff’s study using
a chiral boroxinate Brønsted acid catalyst revealed that N-
diarylmethyl imines can be used in catalytic asymmetric trans-
aziridination with N-phenyldiazoacetamide,16 simple N-benzyl
imines have not yet been used.
An initial attempt using the above-mentioned reaction
conditions gave N-benzyl trans-aziridine 7a in 51% yield with
94% ee (Scheme 1). While the enantioselectivity was already
high enough, we encountered difficulty in achieving a higher
conversion and yield in this reaction.18 In such a case, it is
desirable that the acidity of the catalyst be tuned to improve the
catalytic activity without affecting its chiral environment. This
could be easily realized in our catalyst system by replacing 2-
boronobenzoic acid with an electronically modified one.
Actually, by using the combination of the same diol 5e and
mol % of a chiral diol in CH2Cl2, and the mixture was stirred
for 30 min at rt to form the complex. Molecular sieves were
added to scavenge water generated in this event. After the
addition of the imine, the solution was cooled to the reaction
temperature, and N-phenyldiazoacetamide was then added to
start the reaction.
At first, we examined two commercial C2-symmetric chiral
diols, (R,R)-diethyl tartrate 2 and (S,S)-hydrobenzoin 3, to
determine the viability of our concept. Whereas the use of 2
gave the trans-aziridine 6a in a racemic form (entry 1), use of 3
was found to give the product with apparent asymmetric
induction (entry 2). It should be noted that 2-boronobenzoic
acid 1a itself was able to promote the reaction despite its low
solubility in CH2Cl2, while chiral diols were not sufficiently
acidic to facilitate the reaction. In a further attempt we carried
out reactions using (S)-1,1-diphenyl-1,2-propanediol 4 and (S)-
1,1,2-triphenylethanediol 5a (entries 3 and 4). Contrary to our
initial assumption that these C1-symmetric diols would not
exceed C2-symmetric diols in terms of the enantioselectivity,
both diols 4 and 5a furnished 6a with promising
enantioselectivities. Encouraged by this result, we then
examined several (S)-1,1-diaryl-2-phenylethanediols 5b−5e,
which were prepared in one step from commercial (S)-
mandelic acid methyl ester. Whereas the 4- and 3,5-substitution
of the aryl group had no significant influence (entries 5 and 6),
use of 2-tolyl substituted diol 5d resulted in the inversion of the
sense of the enantioselectivity (entry 7). Finally, 2-biphenyl
substituted diol 5e was found to be optimal with which 6a was
obtained in 63% yield with 89% ee (entry 8). By lowering the
reaction temperature to −20 °C, the enantioselectivity could be
further improved to 92% ee (entry 9). It is noteworthy that use
B
dx.doi.org/10.1021/ja407764u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 1. Optimization of the Catalyst for the Trans-
Aziridination Using N-Benzyl Imine
2-boronobenzoic acid 1b bearing an electron-withdrawing 4-
trifluoromethyl group, the conversion reached 95% and a
considerable increase of the yield was observed while retaining
the high enantioselectivity.
With the quickly optimized catalyst in hand, we investigated
the substrate scope using a variety of N-Bn imines. Compared
with the trans-aziridination using N-Boc imines, the yields and
enantioselectivities were found to be generally higher (Table 3,
a
Table 3. N-Bn Imine Scope
Figure 3. NMR study of the catalyst.
1
soluble in CDCl3 and hard to detect by H NMR, the catalyst
solution apparently showed the signals corresponding to its
aromatic regions (open squares). These observations clearly
supported the formation of a boronate ester between the diol
and boronic acid moieties. In further study, we found that the
catalyst complex was actually isolable as a bench stable solid
(see Supporting Information for details), whose 1H NMR
shown in Figure 3c matched that of the in situ formed
catalyst.21 ESI-MS measurement of this isolated complex
provided an intense peak at m/z 571.21 (M−H)−, which
corresponded to the expected boronate ester.
c
d
entry
R2
% yield
% ee
1
Ph
80 (7a)
80 (7b)
67 (7c)
74 (7d)
70 (7e)
77 (7f)
61 (7g)
87 (7h)
53 (7i)
53 (7j)
78 (7k)
55 (7l)
48 (7m)
97
95
96
95
98
97
98
98
91
97
94
84
74
2
4-tolyl
3
3-tolyl
4
2-tolyl
5
4-BrC6H4
3-ClC6H4
2-ClC6H4
3-MeOC6H4
2-MeOC6H4
4-PivOC6H4
2-Np
6
7
8
9
As the optimal chiral diol 5e has C1 symmetry, its
complexation with 2-boronobenzoic acid potentially gives a
diastereomeric pair (S,S)-8 and (S,R)-8 arising from the
additional chirality at the spiro boronate moiety (Figure 4a).
10
11
12
13
(CH3)2CH
Cy
a
Performed with N-Bn imine (0.10 mmol), N-phenyldiazoacetamide
(0.12 mmol), 5e (0.012 mmol), 1b (0.010 mmol), and MS4A (100
b
1
mg) in CH2Cl2 (1.0 mL). Determined by H NMR analysis of the
crude material. Isolated yield. Determined by chiral HPLC analysis.
c
d
entries 2−11). The reaction was not sensitive to the position of
the functionality, as a variety of 2-substituted benzaldehyde-
derived imines could all participate in the reaction (entries 4, 7,
and 9). This reaction was also applied to branched aliphatic
imines in modest yields with a slight decrease of the
enantioselectivity (entries 12 and 13).19 Linear aliphatic
aldehydes were not suitable substrates, and the corresponding
aziridines were obtained in low yields (data not shown).
1
We then set out to elucidate the catalyst structure by H
NMR experiments in CDCl3.20 Shown in Figure 3a and 3b are
the charts of diol 5e and the catalyst solution prepared in situ
from a 1.2:1 mixture of 5e and 1a in the presence of molecular
sieves. The OH signals of 5e which appeared around 3 ppm
(filled circles) diminished in the catalyst solution, and the
doublet at 5.6 ppm (filled triangle) which corresponds to the α-
hydrogen of the alcohol moiety appeared as a singlet at 5.9 ppm
(open triangle). While 2-boronobenzoic acid itself is poorly
Figure 4. (a) Two diastereomers of the complex 8. (b) Ortep diagram
of the lithium salt of the 1:1 complex of 1a and 5a shown at the 50%
probability level. Hydrogen atoms and DMF are omitted for clarity.
(c) Two possible catalyst−substrate complexes.
C
dx.doi.org/10.1021/ja407764u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(2) Akiyama, T.; Itoh, J.; Yokota, K.; Fuchibe, K. Angew. Chem., Int.
Ed. 2004, 43, 1566.
From the fact that essentially a single isomer was detected in
the 1H NMR experiment even at low temperature,22 either one
of these two complexes would be operating in the reaction. To
gain insight into this, we worked extensively to obtain a crystal
of the catalyst suitable for X-ray analysis. After many attempts,
we finally found that the lithium salt of the complex derived
from 5a and 1a could be crystallized in the presence of a small
amount of DMF (Figure 4b). From this experiment, the
chirality at the boron center of the 5a·1a complex was
determined to be (S). Accordingly, (S,S)-8 was tentatively
assigned as the true catalyst, although we could not exclude the
possibility that the chirality at the boron may be inverted
completely depending on the nature of a chiral diol. A closer
look at the crystal structure indicated that it could be described
as a lithium borate salt (Figure 4b, right) in which the double
bond character of the carboxylate moiety resides on the exo C−
O bond (1.22(1) Å) rather than the endo C−O bond (1.33(0)
Å).23,24 Uniquely, this lithium borate salt forms a 1-D
coordination polymer in the crystalline state wherein the
lithium cation coordinated not only to the oxygen atom of the
carboxylic acid but also to the oxygen of the diol in addition to
two molecules of DMF (see SI for details). From this
observation, we assumed that the active catalyst complexed
with an imine can be depicted as a spiro borate having the
protonated imine either at the carbonyl oxygen (A) or at the
oxygen derived chiral diol (B) as shown in Figure 4c.25
Although it is not conclusive, complex B seems to be more
likely, as it bears the substrate closer to the chiral environment
of the diol.26,27
(3) Uraguchi, D.; Terada, M. J. Am. Chem. Soc. 2004, 126, 5356.
(4) Nakashima, D.; Yamamoto, H. J. Am. Chem. Soc. 2006, 128, 9626.
(5) Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc. 2007, 129, 10054.
̌
́
(6) Coric, I.; List, B. Nature 2012, 483, 315.
(7) Hatano, M.; Maki, T.; Moriyama, K.; Arinobe, M.; Ishihara, K. J.
Am. Chem. Soc. 2008, 130, 16858.
(8) Garci
́ ́ ́ ́
a-Garcia, P.; Lay, F.; Rabalakos, C.; Garcia-Garcia, P.; List,
B. Angew. Chem., Int. Ed. 2009, 48, 4363.
(9) For the combination of an achiral Brønsted acid and a chiral
Lewis base, see: Nugent, B. M.; Yoder, R. A.; Johnston, J. N. J. Am.
Chem. Soc. 2004, 126, 3418.
(10) For the combination of an achiral sulfonic acid and a chiral
thiourea, see: (a) Xu, H.; Zuend, S. J.; Woll, M. G.; Tao, Y.; Jacobsen,
E. N. Science 2010, 327, 986. (b) Lin, S.; Jacobsen, E. N. Nat. Chem.
2012, 4, 817.
(11) In chiral Lewis acid assisted Brønsted acid catalysis, the active
proton is provided from the phenolic proton of a chiral scaffold, not
from an achiral Brønsted acid. Yamamoto, H.; Futatsugi, K. Angew.
Chem., Int. Ed. 2005, 44, 1924.
(12) For a general review, see: Hall, D. G. Boronic Acids; Wiley-VCH:
Weinheim, 2005.
(13) For a recent review on the catalytic use of boronic acid
́
derivatives, see: Dimitrijevic, E.; Taylor, M. S. ACS Catal. 2013, 3, 945.
(14) (a) Auvil, T. J.; Mattson, A. E. Synthesis 2012, 44, 2173. See
also: (b) So, S. S.; Burkett, J. A.; Mattson, A. E. Org. Lett. 2011, 13,
716.
(15) (a) Hashimoto, T.; Uchiyama, N.; Maruoka, K. J. Am. Chem. Soc.
2008, 130, 14380. (b) Zeng, X.; Zeng, X.; Xu, Z.; Lu, M.; Zhong, G.
Org. Lett. 2009, 11, 3036.
(16) (a) Desai, A. A.; Wulff, W. D. J. Am. Chem. Soc. 2010, 132,
13100. (b) Huang, L.; Zhang, Y.; Staples, R. J.; Huang, R. H.; Wulff,
W. D. Chem.Eur. J. 2012, 18, 5302.
́
(17) For notable exceptions, see: (a) Coric, I.; Muller, S.; List, B. J.
̈
In summary, we succeeded in developing a new chiral
Brønsted acid catalyst composed of two independent organic
molecules, a chiral diol as a ligand and 2-boronobenzoic acid as
an achiral Brønsted acid. This binary system offers an
opportunity to tune the chiral environment and acidity of the
catalyst independently, while avoiding a tedious synthesis of a
variety of different catalysts. This concept was validated in
catalytic asymmetric trans-aziridinations of N-Boc and N-benzyl
imines with N-phenyldiazoacetamide. We have also succeeded
̌
Am. Chem. Soc. 2010, 132, 17370. (b) Rowland, G. B.; Zhang, H.;
Rowland, E. B.; Chennamadhavuni, S.; Wang, Y.; Antilla, J. C. J. Am.
Chem. Soc. 2005, 127, 15696.
(18) Modification of the other reaction conditions such as substrate
ratio, solvent, and catalyst loading did not improve the result. It seems
to be partially due to the deactivation of the catalyst by alkylation of
the carboxylic acid with the diazoacetamide.
(19) In these two cases, the corresponding triazolines were also
obtained in approximately 10% yield. Troyer, T. L.; Muchalski, H.;
Hong, K. B.; Johnston, J. N. Org. Lett. 2011, 13, 1790.
(20) For the 11B NMR of the catalyst, see SI.
(21) We subjected this preformed catalyst to the trans-aziridination
of benzaldehyde-derived N-Boc imine and N-phenyldiazoacetamide,
giving the trans-aziridine 6a in 66% yield with 92% ee. The result was
no different from that obtained by the in situ formed catalyst (see
Table 2, entry 1).
1
in shedding light on the nature of the catalyst using H NMR
and X-ray crystallography.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details and characterization data for new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) For the VT NMR of the catalyst, see SI.
(23) Hashimoto, T.; Kimura, H.; Nakatsu, H.; Maruoka, K. J. Org.
Chem. 2011, 76, 6030.
(24) Zhang, L.; Cheng, J.; Carry, B.; Hou, Z. J. Am. Chem. Soc. 2012,
AUTHOR INFORMATION
Corresponding Author
maruoka@kuchem.kyoto-u.ac.jp
■
134, 14314.
(25) Whereas the X-ray analysis strongly supports the formation of a
borate, the possibility of chiral Brønsted acid assisted boron Lewis acid
cannot be excluded. See refs 11 and 13.
Notes
The authors declare no competing financial interest.
(26) For a review on chiral anions, see: Lacour, J.; Moraleda, D.
Chem. Commun. 2009, 7073.
(27) Hu, G.; Gupta, A. K.; Huang, R. H.; Mukherjee, M.; Wulff, W.
D. J. Am. Chem. Soc. 2010, 132, 14669.
ACKNOWLEDGMENTS
■
This work was partially supported by a Grant-in-Aid for
Scientific Research from the MEXT (Japan). A.O.G. thanks the
Japanese Government (MEXT) Scholarship Program for the
fellowship.
REFERENCES
■
(1) For authoritative reviews, see: (a) Akiyama, T. Chem. Rev. 2007,
107, 5744. (b) Terada, M. Synthesis 2010, 1929.
D
dx.doi.org/10.1021/ja407764u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX