Catalytic Asymmetric Synthesis of N-Chiral Amine Oxides

Article abstract of DOI:10.1002/anie.201606354

Direct asymmetric synthesis of N-chiral amine oxides was accomplished (up to 91:9 e.r.) by means of a bimetallic titanium catalyst. A hydroxy group situated at the γ-position of the N stereocenter enables the desired N-oxidation through dynamic kinetic resolution of the trivalent amine substrates. The method was further extended to the kinetic resolution of racemic γ-amino alcohols with a preexisting stereocenter, giving an important class of enantioenriched (up to 99.9:0.1 e.r.) building blocks that are otherwise difficult to synthesize.

Full text of DOI:10.1002/anie.201606354

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201606354
German Edition: DOI: 10.1002/ange.201606354
Asymmetric Catalysis
Catalytic Asymmetric Synthesis of N-Chiral Amine Oxides
Sukalyan Bhadra* and Hisashi Yamamoto*
Abstract: Direct asymmetric synthesis of N-chiral amine
oxides was accomplished (up to 91:9 e.r.) by means of
a bimetallic titanium catalyst. A hydroxy group situated at
the g-position of the N stereocenter enables the desired N-
oxidation through dynamic kinetic resolution of the trivalent
amine substrates. The method was further extended to the
kinetic resolution of racemic g-amino alcohols with a preexist-
ing stereocenter, giving an important class of enantioenriched
(up to 99.9:0.1 e.r.) building blocks that are otherwise difficult
to synthesize.
We envisaged that this approach could be used to position
a hydroxy group at an appropriate distance from the N center
to confer enantioselectivity on the N-oxidation. We herein
report direct access to N-chiral amine oxides starting from the
corresponding unsymmetrical tertiary amino alcohols
through bimetallic catalysis (Figure 1).
Optically active chiral sulfoxides, phosphine oxides, and
amine oxides are frequently found in biologically relevant
compounds. In addition, these molecules have been used both
as chiral ligands in metal-catalyzed reactions and as organo-
catalysts.[1] The development of catalytic methods leading to
these molecules are of fundamental importance. Whereas, the
asymmetric syntheses of optically active sulfoxides and
phosphine oxides are well established, the synthesis of
amine oxides has not been satisfactorily developed. The
potential pitfall of the direct synthesis of N-chiral amine
oxides originates from the fact that the starting tertiary amine
enantiomers are always in equilibrium as a result of a rapid
pyramidal inversion on the stereogenic nitrogen atom of the
substrates. A literature survey demonstrates that N-chiral
N,N-dialkylarylamine oxides are primarily accessed through
various resolution methods.[2–4] Thus the direct asymmetric
oxidation of tertiary amines to the corresponding N-chiral
amine oxides remained unsolved except by using an enzy-
matic process that has rather low enantioselectivity and
efficiency.[5,6] A powerful catalytic asymmetric route to N-
chiral amine oxides is reported herein.
Figure 1. Asymmetric N-oxidation.
We started our investigation for a suitable asymmetric N-
oxidation method with various amino alcohol substrate
classes (for details see Table 1 and Tables S1,S2 in the
Supporting Information). The desired N-oxidation of the b-
amino alcohol gave the corresponding N-oxide 2 as a racemic
mixture in good yield when using a 2:1 complex of Ti and
N,O-ligand 1 in the presence of 70% aqueous tert-butyl
hydroperoxide solution (TBHP) at room temperature.[8]
Gratifyingly, we observed that the title reaction for the g-
amino alcohol provided the N-oxide 3a in reasonably good
yield and enantioselectivity (69%, 70:30 e.r.; entry 2). The
stereoselectivity was further improved when the reaction was
carried out at À208C. A 2:1 metal-to-ligand ratio was crucial;
the use of a 1:1 metal-to-ligand ratio had a deleterious effect.
The presence of the g-hydroxy group was the key for
successful N-oxidation, since the reaction does not proceed
in its absence, thus confirming that the hydroxy group acts as
an anchor for the Ti catalyst (entry 3). However, as expected,
the N-oxidation was not effective for other substrate classes,
including d-amino alcohols, hydroxymethylaniline, and ami-
nophenols (see Table S1).
Recently we have introduced a new concept based on
bimetallic catalysis for various asymmetric oxidation process-
es, including epoxidation of homoallylic alcohols and 2-allylic
phenols, and sulfoxidation of g-hydroxypropyl sulfides.[7,8] In
the new approach, one of the two independent metal centers
inside the bimetallic catalyst scaffold holds the hydroxy
substrate in close proximity to the other metal center, which
enantioselectively transfers the oxidant to the reactive site.
[*] Dr. S. Bhadra, Prof. Dr. H. Yamamoto
Molecular Catalyst Research Center, Chubu University
1200, Matsumoto-cho, Kasugai, Aichi 487-8501 (Japan)
E-mail: bhadra@isc.chubu.ac.jp
hyamamoto@isc.chubu.ac.jp
Dr. S. Bhadra
With an effective reaction system in hand, we subse-
quently studied the scope of the asymmetric N-oxidation for
various g-amino alcohols (Scheme 1). A range of (N,N-
benzylalkyl)amino alcohols were converted into the corre-
sponding N-chiral oxide in satisfactory yields and with
moderate to good enantioselectivity (3a–3j). The presence
Inorganic Materials and Catalysis Division, CSIR-Central Salt and
Marine Chemicals Research Institute
G.B. Marg, Bhavnagar 364002, Gujarat (India)
E-mail: sukalyanbhadra@csmcri.org
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201606354.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Identification of the appropriate amino alcohol for the asym-
side chain instead of the benzyl group at the N center (3k).
Further, g-(N,N-isopropylphenyl)amino alcohol did not
undergo oxidation under the optimized conditions (3l).
We speculate that the asymmetric N-oxidation proceeds
via dynamic kinetic resolution at the N center of the chiral
racemic g-amino alcohol.[9] The enantioselectivity of the
product was induced by our new bimetallic catalysis
approach. Although the reaction mechanism is not clear at
this point, the stereodiscrimination should take place during
the N-oxidation stage at one of the titanium centers of the
bimetallic catalyst.
Encouraged by these promising results, we became
interested in applying this strategy to the practical synthesis
of optically active g-amino alcohols via kinetic resolution
(KR) of the racemates through formation of the correspond-
ing N-oxide. The synthesis of optically active g-amino
alcohols and amine oxides is of profound significance given
their abundance in bioactive natural products and pharma-
ceutical agents, and their increasing use as chiral ligands
(Figure 2).[10,11] While the racemic g-amino alcohols are
metric N-oxidation.[a]
Entry
1
N-Oxides
T
Yield [%], e.r. [%]
85, 1:1
RT
RT
67, 70:30
68, 75:25
45, 60:40
60, 67:33[b]
30, 58:42[c]
À208C
À788C
À208C
À208C
2
3
RT
0
[a] Reaction Conditions: 0.15 mmol amino alcohol (1 equiv), 2.1 equiv
70% aq TBHP, 10 mol% Ti(O-iPr)4, 5 mol% 1, in 4 mL CH2Cl2 at a given
temperature for 24 h. Yields of isolated product are shown, and the e.r.
values were determined by HPLC with a chiral stationary phase. [b] With
5 mol% Ti(O-iPr)4 and 2.5 mol% 1. [c] With 5 mol% Ti(O-iPr)4 and
5 mol% 1.
Figure 2. Some optically active g-amino alcohol derivatives.
simply accessible, only a handful of straightforward synthetic
routes to the optically active variants have been reported,
including transition-metal-catalyzed asymmetric reduction of
the corresponding amino ketones,[12] stereoselective addition
of dialkyl zinc to aldehydes,[13] and the recently reported
Buchwaldꢀs hydroamination reactions.[14] Nonetheless, vari-
ous classes of amino alcohols remain that cannot be synthe-
sized with excellent enantioselectivity.
We predicted that our N-oxide formation strategy could
be utilized for a kinetic resolution (KR) of g-amino alcohols
that contain a preexisting stereogenic center, a transformation
that has failed to be achieved through the Sharpless
method.[15] After a quick survey of various reaction param-
eters (see Table S3,S4), it turned out that the quantity of
TBHP is crucial to achieve excellent enantioselectivity. To our
delight we found that when the racemic secondary alcohol 5’a
was subjected to oxidation with 1.4 equiv of 70% aq. TBHP in
the presence of the optimized catalyst system, the optically
active g-amino alcohol 5a was obtained at room temperature
in virtually enantiopure form (> 99% ee). The KR may
proceed via the formation of a stereocongested transition
state driven by the crowded secondary hydroxy group of the
substrate. In this way, an array of optically active sec-g-amino
alcohols and the corresponding N-oxides were obtained in
good yields and with excellent enantioselectivity (Table 2A).
Next, we studied the KR of primary g-amino alcohols with
Scheme 1. Scope of the asymmetric N-oxidation. Reaction Conditions:
0.15 mmol 3’ (1 equiv), 2.1 equiv 70% aq TBHP, 10 mol% Ti(O-iPr)4,
5 mol% 1, in 4 mL CH2Cl2 at À208C for 24 h. Yields of isolated
product are shown, and the e.r. values were determined by HPLC with
a chiral stationary phase. [a] At 08C. [b] With 3.5 equiv 70% aq. TBHP,
and for 72 h.
of a bulky substituent at the N stereocenter facilitated the
formation of more enantioenriched products (3a–3h).
Unfortunately, however, the enantioselectivity drops in the
presence of a less bulky substituent (3i–3j) or a simple alkyl
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 2: Kinetic resolution of g-amino alcohols.
g-amino alcohols (7a–7c) were obtained with excellent
enantiopurity, along with their N-oxides (8a–8c) in diaste-
reomerically pure form and with satisfactory enantiomeric
excess. The KR was further extended successfully to other
primary g-amino alcohols with different substituents in the g-
amino alcohol chain (7d–7 f). The N-oxides thus formed can
be transformed into the corresponding g-amino alcohols of
opposite configuration through various reduction methods.[17]
However the KR did not proceed for the g-amino tertiary
alcohols with a geminal dialkyl substituents.
In conclusion, we have developed a catalytic asymmetric
N-oxidation of unsymmetrical g-hydroxy tertiary amines that
proceeds via dynamic kinetic resolution, leading to N-chiral
amine oxides. A hydroxy group located at the g-position with
respect to the N stereocenter is key to the success of this
approach. The concept of hydroxy-directed N-oxidation was
further extended to the KR of racemic g-amino alcohols with
a preexisting stereogenic center. These oxidative transforma-
tions based on bimetallic catalysis approach provide a new
platform for “catalyst-controlled” chemical reactions, and
further investigation is currently ongoing in our laboratory.
Amino Alcohol 5’
5
6
C[c] s[d]
e.r.[a] [% Yield][b] e.r.[a] [% Yield][b]
5a
6a
51
62
56
213
99.5:0.5 [43]
97.7:2.3 [48]
5b
6b
80:20 [50]
17
54
98.9:1.1 [40]
5c
6c
90:10 [49]
99.9:0.1 [45]
5d
6d
97:3 [48]
50
139
97.4:2.6 [44]
Experimental Section
General procedure for the asymmetric N-oxidation of g-amino
alcohols: To the ligand 1 (5.2 mg, 0.0075 mmol, 5 mol%) in 3 mL
dicloromethane, Ti(O-iPr)4 (4.5 mL, 0.015 mmol, 10 mol%) was
added at room temperature with a microlitre syringe under nitrogen
atmosphere. The resulting yellow solution almost immediately turned
red and was stirred for additional 2 h. Next, 70% aq. tBuOOH
solution (31 mL, 0.315 mmol, 2.1 equiv) was added and stirred for an
additional 10 min before cooling to À208C. The g-amino alcohol
(0.15 mmol, 1 equiv) in 1 mL CH2Cl2 was then added dropwise with
a syringe. The reaction mixture was stirred at the same temperature
for the required time (TLC). It was then directly subjected to column
chromatography over silica gel using CH2Cl2 and MeOH as eluents to
obtain the corresponding N-oxide.
Amino Alcohol 7’
7
8
C[c] s[d]
e.r.[a] (% Yield)[b] e.r.[a] (% Yield)[b]
7a
8a
52 56
51 87
52 133
97.5:2.5 [45]
93.7:6.3 [52]
7b
8b
97.7:2.3 [44]
95.7:4.3 [48]
7c
8c
Acknowledgements
99.7:0.3 [39]
95.9:4.1 [48]
This work was supported by ACT-C, JST. We thank JSPS for
a postdoctoral fellowship to SB.
7d
95:5 [48]
8d
60 12
63 16
80.3:19.7 [48]
7e
8e
79:21 [53]
Keywords: asymmetric catalysis · dynamic kinetic resolution ·
homogeneous catalysis · nitrogen oxides · titanium
99.1:0.9 [43]
7 f
8 f[e]
64:36 [38]
70
3
82:18 [57]
[a] The e.r. values were determined by HPLC with a chiral stationary
phase. [b] Yield of isolated product. [c] Conversion, C=eeS/(eeS +eeP).
[d] The selectivity factor s was calculated as s=ln[(1ÀC)(1ÀeeS)]/
ln[(1ÀC)(1+eeS)]. [e] For 24 h.
[1] For an overview see: a) Asymmetric Synthesis, Vol. 4 (Eds.: J. D.
Morrison, J. W. Scott), Academic Press, New York, 1983;
b) K. M. Pietrusiewicz, M. Zablocka, Chem. Rev. 1994, 94,
1375 – 1411; c) N. V. Dubrovina, A. Boerner, Angew. Chem. Int.
Ed. 2004, 43, 5883 – 5886; Angew. Chem. 2004, 116, 6007 – 6010;
d) H. B. Kagan in Organosulfur Chemistry in Asymmetric Syn-
thesis (Eds.: T. Toru, C. Bolm), Wiley-VCH, Weinheim, 2008,
pp. 1 – 29; e) M. Carmen CarreÇo, G. Hernꢁndez-Torres, M.
Ribagorda, A. Urbano, Chem. Commun. 2009, 6129 – 6144;
f) O. I. Kolodiazhnyi, Tetrahedron: Asymmetry 2012, 23, 1 – 46;
diverse substitution patterns, which represent a more chal-
lenging but important substrate class (Table 2B). The 2-
hydroxyalkyl-substituted piperidines and their derivatives
constitute a framework that is frequently found in the
structure of drug candidates.[16] We were pleased to find that
when using our new KR method, 2-piperidineethanol-derived
ˇ
g) S. Liao, I. Coric, Q. Wang, B. List, J. Am. Chem. Soc. 2012, 134,
10765 – 10768; h) G. Desimoni, G. Faita, P. Quadrelli, Chem. Rev.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
2014, 114, 6081 – 6129; i) B. M. Trost, M. Rao, Angew. Chem. Int.
Ed. 2015, 54, 5026 – 5043; Angew. Chem. 2015, 127, 5112 – 5130.
[2] a) J. Meisenheimer, Chem. Ber. 1908, 41, 3966 – 3976; b) J.
Meisenheimer, H. Glawe, H. Greeske, A. Schorning, E.
Vieweg, Justus Liebigs Ann. Chem. 1926, 449, 188 – 213; c) J.
Meisenheimer, Justus Liebigs Ann. Chem. 1922, 428, 252 – 285.
[3] F. Toda, K. Mori, Z. Stein, I. Goldberg, Tetrahedron Lett. 1989,
30, 1841 – 1844.
123, 353 – 354; e) M. Saito, M. Nakajima, S. Hashimoto, Chem.
Commun. 2000, 1851 – 1852; f) M. Nakajima, M. Saito, M.
Uemura, S. Hashimoto, Tetrahedron Lett. 2002, 43, 8827 – 8829;
g) T. Shimada, A. Kina, S. Ikeda, T. Hayashi, Org. Lett. 2002, 4,
ˇ
´
2799 – 2801; h) A. V. Malkov, P. Kocovsky, Eur. J. Org. Chem.
2007, 29 – 36; i) X. H. Liu, L. L. Lin, X. M. Feng, Acc. Chem. Res.
2011, 44, 574 – 587, and references therein.
[11] a) G. T. Rice, M. C. White, J. Am. Chem. Soc. 2009, 131, 11707 –
11711; b) H. Ren, W. D. Wulff, Org. Lett. 2013, 15, 242 – 245.
[12] a) D. Liu, W. Gao, C. Wang, X. Zhang, Angew. Chem. Int. Ed.
2005, 44, 1687 – 1689; Angew. Chem. 2005, 117, 1715 – 1717;
b) J. M. M. Verkade, P. J. L. M. Quaedflieg, G. K. M. Verzijl, L.
Lefort, F. L. van Delft, J. G. de Vries, F. P. J. T. Rutjes, Chem.
Commun. 2015, 51, 14462 – 14464, and references therein.
[13] C. Lutz, V. Lutz, P. Knochel, Tetrahedron 1998, 54, 6385 – 6402.
[14] S.-L. Shi, Z. L. Wong, S. L. Buchwald, Nature 2016, 532, 353 –
356.
[15] Although a Ti-based catalyst reported by Sharpless et al. enables
the KR of b-hydroxy amines, it does not allow a similar
transformation of g-hydroxy amines. For details, see: a) S.
Miyano, L. D.-L. Lu, S. M. Viti, K. B. Sharpless, J. Org. Chem.
1983, 48, 3608 – 3611; b) S. Miyano, L. D.-L. Lu, S. M. Viti, K. B.
Sharpless, J. Org. Chem. 1985, 50, 4350 – 4360.
´
[4] J. Drabowicz, B. Dudzinski, M. Mikołajczyk, S. Colonna, N.
Gaggero, Tetrahedron: Asymmetry 1997, 8, 2267 – 2270.
[5] F. A. Davis, J. R. H. Jenkins in Asymmetric Synthesis, Vol. 4
(Eds.: J. D. Morrison, J. W. Scott), Academic Press, New York,
1983.
[6] S. Colonna, N. Gaggero, J. Drabowicz, P. Lyzwa, M. Mikolajczyk,
Chem. Commun. 1999, 1787 – 1788.
[7] For reviews, see: a) Multimetallic Catalysts in Organic Synthesis
(Eds.: M. Shibasaki, Y. Yamamoto), Wiley-VCH, Weinheim,
2004; b) S. Matsunaga, M. Shibasaki, Bull. Chem. Soc. Jpn. 2008,
81, 60 – 75; c) S. Matsunaga, M. Shibasaki, Chem. Commun.
2014, 50, 1044 – 1057; d) Cooperative Catalysis: Designing Effi-
cient Catalysts for Synthesis (Ed.: R. Peters), Wiley-VCH,
Weinheim, 2015.
[8] S. Bhadra, M. Akakura, H. Yamamoto, J. Am. Chem. Soc. 2015,
137, 15612 – 15615.
[16] a) G. M. Strunzand, J. A. Finlay in The Alkaloids, Vol. 26 (Ed.:
A. Brossi), Academic Press, New York, 1985; b) M. G. P. Buffat,
Tetrahedron 2004, 60, 1701 – 1729.
[17] H. P. Kokatla, P. F. Thomson, S. Bae, V. R. Doddi, M. K.
Lakshman, J. Org. Chem. 2011, 76, 7842 – 7848.
[9] The barrier to pyramidal inversion at the N center in amines is 6–
7 kcalmolÀ1 at room temperature; For details, see Ref. [5].
[10] An overview on chiral amines: a) Chiral Amine Synthesis:
Methods, Developments and Applications (Ed.: T. C. Nugent),
Wiley-VCH, Weinheim, 2010; For applications of chiral N-oxide
ligands see: b) M. Nakajima, M. Saito, M. Shiro, S.-i. Hashimoto,
J. Am. Chem. Soc. 1998, 120, 6419 – 6420; c) M. Nakajima, Y.
Sasaki, H. Iwamoto, S.-i. Hashimoto, Tetrahedron Lett. 1998, 39,
87 – 88; d) B. Tao, M. M.-C. Lo, G. C. Fu, J. Am. Chem. Soc. 2001,
Received: June 30, 2016
Published online: && &&, &&&&
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Asymmetric Catalysis
S. Bhadra,*
H. Yamamoto*
&&&& — &&&&
Catalytic Asymmetric Synthesis of N-
Chiral Amine Oxides
One-sided: An asymmetric synthesis of
N-chiral amine oxides was developed that
proceeds via dynamic kinetic resolution
of unsymmetrical g-hydroxy tertiary
amines. A binuclear titanium complex
enables the desired hydroxy-directed N-
oxidation process. The concept was fur-
ther expanded to the kinetic resolution of
racemic g-amino alcohols with a preexist-
ing stereocenter.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!