Highly stereoselective epoxidation with H2O2 catalyzed by electron-rich aminopyridine manganese catalysts

Article abstract of DOI:10.1021/ol403018x

Fast, efficient, and highly stereoselective epoxidation with H ₂O₂ is reached by manganese coordination complexes with e-rich aminopyridine tetradentate ligands. It is shown that the electronic properties of these catalysts vary systematically with the stereoselectivity of the O-atom transfer event and exert fine control over the activation of hydrogen peroxide, reducing the amount of carboxylic acid co-catalyst necessary for efficient operation.

Full text of DOI:10.1021/ol403018x

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
Highly Stereoselective Epoxidation with
H O Catalyzed by Electron-Rich
Aminopyridine Manganese Catalysts
000–000
2
2
Olaf Cuss ꢀo , Isaac Garcia-Bosch, David Font, Xavi Ribas, Julio Lloret-Fillol, and
Miquel Costas*
Institut de Quimica Computacional i Cat aꢁ lisi (IQCC), and Departament de Quı´mica,
Universitat de Girona, Campus de Montilivi, E-17071 Girona, Catalonia, Spain
miquel.costas@udg.edu
Received October 21, 2013
ABSTRACT
2 2
Fast, efficient, and highly stereoselective epoxidation with H O is reached by manganese coordination complexes with e-rich aminopyridine
tetradentate ligands. It is shown that the electronic properties of these catalysts vary systematically with the stereoselectivity of the O-atom
transfer event and exert fine control over the activation of hydrogen peroxide, reducing the amount of carboxylic acid co-catalyst necessary for
efficient operation.
Asymmetric epoxidation is an important reaction in
organic synthesis because of the versatility of epoxides
as useful intermediates in the preparation of a diverse
array of functional molecules. Current challenges for
this reaction lay in finding environmentally more sustain-
able methods; for example, oxidations with catalysts based
on first-row, non-toxic earth-abundant metals and oxidants
exhibiting good atom economy that generate minimum
waste are highly desirable. In this regard, selected manganese
coordination complexes with aminopyridine-based li-
gands have emerged as efficient oxidation catalysts. Seminal
studies by Stack and co-workers described [Mn(CF SO ) -
3
3 2
0
1
,2
0
(R,R)-MCP)], MCP = N,N -bis(2-pyridylmethyl)-N,N -
dimethyl-trans-1,2-diaminocyclohexane as a very active
epoxidation catalyst employing peracetic acid (AcOOH)
3
as oxidant. Further elaboration of the manganese com-
plex by Sun et al. led to the development of a catalyst
that showed good enantioselectivity in the epoxidation of
4
chalcones. Hydrogen peroxide constitutes a particularly
(
1) (a) Sundermeier, U.; D o€ bler, C.; Beller, M. Modern Oxidation
interesting oxidant, but its use in manganese-catalyzed
reactions is challenging because of competing hydrogen
peroxide disproportionation reactions. Garcia-Bosch et al.
showed that hydrogen peroxide could be used as terminal
Methods. Wiley-VCH: Weinheim, 2004. (b) Lane, B. S.; Burgess, K. Chem.
Rev. 2003, 103, 2457. (c) Saisaha, P.; de Boer, J.; Browne, W. Chem. Soc.
Rev. 2013, 42, 2059.
(
2) (a) Katsuki, T. In Catalytic Asymmetric Synthesis, 2nd ed.; Ojima,
I., Ed.; Wiley-VCH: New York, 2000; p 287. (b) De Faveri, G.; Ilyashenko,
G.; Watkinson, M. Chem. Soc. Rev. 2011, 40, 1722. (c) Chatterjee, D.
Coord. Chem. Rev. 2008, 252, 176. (d) Wong, O. A.; Shi, Y. Chem. Rev.
2
008, 108, 3958. (e) Diez, D.; Nunez, M. G.; Anton, A. B.; Garcia, P.;
(3) Murphy, A.; Dubois, G.; Stack, T. D. P. J. Am. Chem. Soc. 2003,
125, 5250–5251.
(4) Wu, M.; Wang, B.; Wang, S.; Xia, C.; Sun, W. Org. Lett. 2009, 11,
Moro, R. F.; Garrido, N. M.; Marcos, I. S.; Basabe, P.; Urones, J. G.
Curr. Org. Synth. 2008, 5, 186. (f) Darwish, M.; Wills, M. Catal. Sci.
Technol. 2012, 2, 243. (g) Gopalaiah, K. Chem. Rev. 2013, 113, 3248.
3622.
1
0.1021/ol403018x r XXXX American Chemical Society

oxidant with this type of catalysts in the presence of acetic
5
acid (14 equiv). Subsequently, several manganese cata-
Table 1. Epoxidation of cis-β-Methylstyrene (1) Using Various
Manganese Complexes
lysts containing aminopyridine tetradentate ligands have
been described to efficiently operate under these experi-
6
mental conditions. Good to excellent enantioselectivities
a
have been obtained in the epoxidation of chalcones, but
rational strategies for further improvement of these cata-
lysts allowing for extending their rather narrow substrate
scope, and hopefully without requiring extensive trial and
error testing, need to be developed.
entry
catalyst
AcOH (x equiv)
conv (yield, %)
ee (%)
H
1
2
3
4
5
6
7
8
1
14
61 (38)
57 (33)
44 ( 22)
100 (75)
100 (86)
80 (59)
100 (67)
100 (79)
23 (2)
43
40
43
82
76
69
63
73
55
90
92
Cl
1
CO2Et
14
1
14
Me2N
1
14
dMM
1
14
MeO
1
14
Me
1
Me2N
14
1
0.35
0.35
14
dMM
9
1
b
Me2N
1
0
1
100 (89)
100 (86)
b
Me2N
11
1
0.35
a
2 2
Unless stated, reaction conditions are (S,S)-catalyst(0.1mol%), H O
b
(
1.2 equiv) and AcOH (x equiv) in CH CN at ꢀ30 °C. 2-ethylhexanoic
3
acid (2-eha) was used instead of AcOH.
CO Et) was prepared, characterized (see the Supporting
2
Information for details), and studied as epoxidation cata-
lysts employing H O as oxidant. Catalysts [Mn(CF SO ) -
2
2
3
3 2
H H
(S,S)- PDP)], (S,S)- 1, previously prepared and studied
(
by Talsi and co-workers, serves as a comparative bench-
1
serves as an illustrative example of the series, showing a
Figure 1. Manganese complexes studied and an ORTEP dia-
1. For
6c,g
mark. Crystallographic characterization of (R,R)-
Me2N
Me2N
gram of the X-ray-determined structure of (R,R)-
clarity, H atoms and triflate groups, except for O atoms bound
to the metal, are omitted.
C -symmetric octahedral complex where the ligand adopts
2
7
a cis-R topological coordination geometry. Structural and
metrical parameters are very similar to those described for
To this end, in this paper we show that the electronic
properties of the manganese catalysts in the series of com-
3
H 6g
the Mn(mcp) catalyst, and 1 and will not be described
further (see the Supporting Information). Hydrogen per-
oxide (1.2 equiv) diluted in acetonitrile (1:1 v:v) wasusedas
oxidant and delivered by syringe pump over 30 min to an
acetonitrile solution ofthe catalyst(0.1 mol %), a substrate
x‑
H
plexes [Mn(CF SO ) ( PDP)] ( PDP = 2-[[2-(1-(pyridin-
3
3 2
2-ylmethyl)pyrrolidin-2-yl)pyrrolidin-1-yl]methyl]pyridine,
Figure 1) exert a profound and systematic impact in the
activation of H O to efficiently generate epoxidizing spe-
2
2
(
cis-β-methylstyrene, 1), and AcOH (14 equiv) at ꢀ30 °C.
cies and also in the stereoselectivity exhibited by these
species in epoxidation reactions. This analysis has led us
to the discovery of a very active Mn catalyst that requires
substoichiometric amounts of a carboxylic acid for efficient
H O (1.2ꢀ2 equiv) activation, providing epoxides in short
After a further 30 min of stirring, reactions were analyzed
by chiral GC. Poor epoxide yield and enantioselectivity
H
were obtained with 1 (Table 1, entry 1). Yield and
enantioselectivity turned out to respond in a sensitive
manner to the electronic properties of the pyridine rings
of the ligand. When electron-withdrawing groups were
introduced at the para-position of the pyridine (Table 1,
entries 2 and 3), yields remained moderate and enantios-
electivities were lower. However, we observed a very sub-
stantial improvement in both parameters in the presence of
2
2
reaction times (30ꢀ60 min) with improved stereoselectiv-
ities and substrate scope in comparison with any amino-
pyridineꢀmanganese system described so far.
The series of manganese complexes [Mn(CF SO ) ((S,S)-
3
3 2
x
X
PDP)] (S,S)- 1 (Figure 1) (X = Me N, MeO, Me, Cl and
2
(
5) Garcia-Bosch, I.; Ribas, X.; Costas, M. Adv. Synth. Catal. 2009,
51, 348.
6) (a) Wu,M.;Miao,C.-X.;Wang,S.;Hu,X.;Xia,C.;K u€ hn, F. E.; Sun,
electron-donating substituents (Table 1, entries 4ꢀ7). Up to
3
Me2N
82% ee was attained with 1, the most e-rich catalyst of
the series. The reaction was stereospecific, providing only
(
W. Adv. Synth. Cat. 2011, 353, 3014. (b) Gomez, L.; Garcia-Bosch, I.;
Company, A.; Sala, X.; Fontrodona, X.; Ribas, X.; Costas, M. Dalton Trans.
007, 0, 5539. (c) Ottenbacher, R. V.; Bryliakov, K. P.; Talsi, E. P. Adv.
Synth. Cat. 2011, 353, 885. (d) Garcia-Bosch, I.; G oꢀ mez, L.; Polo, A.; Ribas,
X.; Costas, M. Adv. Synth. Cat. 2012, 354, 65. (e) Wang, B.; Miao, C.; Wang,
S.; Xia, C.; Sun, W. Chem.;Eur. J. 2012, 18, 6750. (f) Dai, W.; Li, J.; Li, G.;
Yang, H.; Wang, L.; Gao, S. Org. Lett. 2013, 15, 4138. (g) Lyakin, O. Y.;
Ottenbacher, R. V.; Bryliakov, K. P.; Talsi, E. P. ACS Catal. 2012, 2, 1196.
epoxides where stereoconfiguration was retained. Good
dMM
performance was also obtained with catalyst 1. There-
fore, we concluded that chemo- and stereoselectivity were
2
(7) Knof, U.; von Zelewsky, A. Angew. Chem., Int. Ed. 1999, 38, 302.
Org. Lett., Vol. XX, No. XX, XXXX
B

systematic and substantially improved as the ligand was
Me2N
present system provides good yields and enantioselectivity
in the epoxidation of various families of olefinic substrates.
In general, more e-deficient olefins such as enones required
higher catalyst, H O , and carboxylic acid loadings.
dMM
more e-rich. For this reason, catalysts
were chosen for further investigations.
1 and
1
2
2
*
Table 2. Substrate Scope on the Asymmetric Epoxidation
Figure 2. Epoxidation of 1 as a function of AcOH loading for
Me2N
dMM
H
catalysts (S,S)-
ditions: catalyst (0.1 mol %), H
0.35ꢀ14 equiv) in CH CN at ꢀ30 °C.
1, (S,S)-
1, and (S,S)- 1. Reaction con-
2
O
2
(1.2 equiv), and AcOH
(
3
Carboxylic acids have been previously recognized as a
key additive in manganese catalyzed oxidations in order to
divert H O consumption from nonproductive peroxide
2
2
8
disproportionation reactions toward substrate oxidation.
Very interestingly, the amount of acetic acid necessary for
X
efficient operation in the 1 series appeared to be system-
*
Me2N
Unless stated, reaction conditions are (S,S)-
(2 equiv) and 2-eha (1 equiv) in CH
1 (0.5 mol %),
atically dependent in the nature of the manganese catalyst,
and more specifically in the e-releasing character of the
a
H
catalyst, H
2
O
2
3
CN at ꢀ30 °C. 0.2 mol % of
b
(2 equiv), 2-eha (0.5 equiv). GC conversion and (yield).
O
2 2
Me2N
c
d
ligand. Catalyst
1 retained full epoxidation activity
2 2
0.1 mol % of catalyst, H O (1.5 equiv), and 2-eha (0.35 equiv). Temp
ꢀ
10 °C. ee’s and configuration determined by chiral GC and HPLC (see
the Supporting Information for details).
with only 0.35 equiv of AcOH, although enantioselectivity
was somewhat eroded (Table 1, entry 8). A further de-
crease to 0.1 equiv of AcOH provided moderate epoxide
dMM
yield (51%). In contrast, for
1 the epoxide yield was
Styrenes are recognized as particularly difficult substrates
for asymmetric epoxidation and are epoxidized with a
remarkable 68ꢀ69% ee (entries 1 and 2). However, intro-
dution of a methyl group in the R-olefin site results in poor
stereoselection (entry 3). Excellent enantioselectivity was
achieved for a substituted chromene (97% ee, entry 4) and
also for cis-cinnamic derivatives (90ꢀ98% ee, entries 5ꢀ7).
Epoxidation of aromatic enones proceeded with good to
excellent yields, including not only the commonly studied
trans-chalcone (9) (96% ee, entry 8) but also more challen-
ging substrates such as cis-chalcone (8) (90% ee, entry 7),
trans-ethyl cinnamate (10) (92% ee, entry 9), an alkyl styryl
ketone (11) (96% ee, entry 10), and also an amide derivative
reduced substantially as the acetic acid loading was re-
duced from 14 to 0.35 equiv, and a more drastic effect was
H
observed for 1 (see Figure 2, and entry 9 in Table 1).
Therefore, the data from Table 1 and Figure 2 show that
activation of H O to form epoxidizing species is uniquely
2
Me2N
2
facilitated by
1. Lyakin and co-workers have recently
reported that alkyl carboxylic acids can be used instead of
acetic acid, and in some specific cases this change results in
6
g
improvement of the enantioselectivity. Therefore, a
range of carboxylic acids were tested in the epoxidation
of 1 (see the Supporting Inforamtion, Table 1). When
Me2N
1
was used with low loadings of racemic 2-ethyl
hexanoic acid (2-eha), the enantioselectivity was improved
up to 90% ee, and this value was slightly improved at low
carboxylic acid loading (0.35 equiv) (entries 10 and 11,
respectively).
(
12) (97% ee, entry 11). However, as noticed for styrenes, the
epoxidation of the R-methyl-substituted chalcone (13) pro-
ceeded with moderate enantioselectivity (57% ee, entry 12),
albeit with high epoxide yield and stereospecificity. Finally,
the trisubstituted tetralone-derived (14) epoxide was achieved
with moderate yield and good enantioselectivity (84% ee,
A substrate scope study is presented in Table 2. Reaction
conditions involved low catalyst loadings (0.1ꢀ0.5 mol %),
H O in slight to moderate excess (1.2ꢀ2 equiv), and
2
2
entry 13). Overall, with the single exception of the epoxida-
Me2N
relatively low amounts of the carboxylic acid 2-eha
tion of substrate 4, catalyst
tioselectivities with regard to any of the Mn-based catalyst
1 provides improved enan-
(
0.35 to 1 equiv). With this set of conditions in hand, the
6
with N-based polydentate ligands described up to date and
6f
(
8) (a) Berkessel, A.; Sklorz, C. A. Tetrahedron Lett. 1999, 40, 7965.
b) Shul’pin, G. B.; Suss-Fink, G.; Smith, J. R. L. Tetrahedron 1999, 55,
345. (c) de Boer, J. W.; Browne, W. R.; Brinksma, J.; Alsters, P. L.;
broader or complementary substrate scope.
An important and interesting set of olefinic substrates is
Δ -unsaturated steroids. It is well established that some
(
5
5
Hage, R.; Feringa, B. L. Inorg. Chem. 2007, 46, 6353.
Org. Lett., Vol. XX, No. XX, XXXX
C

β-epoxide stereoids, such as withanolides, exhibit antitu-
9
moral activities, and for this reason, their β-selective
5
a
Table 3. Epoxidation of Δ -Unsaturated Steroids
epoxidation isaninteresting target. Thesenatural products
contain methyl groups in their β face that exert steric
protection, and their epoxidation with peracids usually
1
0
produces the R-epoxide as the major product. Reagents
1
0,11
for selective β-epoxidation include the Parish reagent,
1
2
structurally elaborated dioxiranes, and catalytic oxida-
tions with porphyrin ruthenium catalysts that use 2,6-
1
Cl pyNO as oxidant. An attractive alternative is to
3
b
2
entrysubstrate
catalyst
yield (%) (18 þ 19) (20) ratio 19/18
employ Mn-based catalysts, making use of H O as oxi-
2
2
c
Me2N
1
2
3
4
5
6
7
15a (S,S)-
1
ꢀ (ꢀ)
ꢀ
dant, but to the best of our knowledge, this approach has
Me2N
15a (S,S)-
15a (R,R)-
1
36 (23)
93:7
67:33
96:4
69:32
94:6
97:3
66/34
48/52
1
4
been described with modest selectivity. Considering the
Me2N
d
1
39 (20)
Me2N
dMM
highly selective nature of electron-rich catalysts
dMM
1 and
15a (S,S)-
15a (R,R)-
1
50 (5)
66 (4)
56 (8)
5
dMM
1
, their application to the oxidation of Δ -unsaturated
1
dMM
dMM
15b (S,S)-
15c (S,S)-
15b (S,S)-
15b (S,S)-
1
steroids was studied. Preliminary optimization experi-
ments disclosed that pivalic acid was a more convenient
coligand than 2-eha and that 15 equiv was necessary since
the use of 0.35 equiv resulted in no epoxide forma-
e
f
1
66 (5)
g
Me2N
8
9
1Fe
46 (4)
60 (2)
h
dMM
1Fe
a
tion (Table 3, entry 1). Using 15 equiv of pivalic acid
Me2N
Unless stated, reaction conditions are catalyst (0.25 mol %), H
(2 equiv), and pivalic acid (15 equiv) in ACOEt/CH CN (4:1) at room
2 2
O
3
(
the epoxidation of 15 but with poor yield and formation
S,S)-
1 provided excellent β-diastereoselectivites in
b
c
d
e
temperature. Isolated yields. Pivalic acid (0.35 equiv). GC yield. H
NMR yield. Ratio β/R determined by GC, except for entry 7 which was
1
f
g
of 23% of allylic oxidation side product 20a (Table 3,
Me2N
determined by H NMR. Allylic 7-β-OH alcohol. Reaction catalyzed
Me2N h
by [Fe(CF SO ) ((S,S)-
(
3
dMM
3 2
3 3 2
PDP)]. Reaction catalyzed by [Fe(CF SO ) -
entry 2). Enantiomeric (R,R)-
1 isomer retained the
(S,S)-
PDP)],
low product yield and modest selectivity toward the β
dMM
isomer (entry 3). Fortunately, (S,S)-
1 proved to be a
the oxygen atom transfer event, providing a useful guiding
principle for rational design of a future generation of cata-
lysts. Catalysts described herein complement or compare
favorably in terms of enantioselectivities and substrate scope
against structurally related examples described so far. When
compared with the analogous iron complexes, obvious
more convenient catalyst providing improved yields, che-
moselectivity, and excellent diastereoselectivity (Table 3,
entry 4). On the other hand, as also observed for (R,
Me2N
dMM
R)-
1 (entry 3), the enantiomer (R,R)-
1 offered a
more modest diastereoselectivity (Table 3, entry 5). Under
5
optimized conditions Δ -unsaturated stereoids 16 and 17
15
similarities arise in the electronic effects, but in addition
the manganese catalysts require lower catalyst loadings, are
more tolerant to aromatic substrates, and show enhanced
were also epoxidized with moderate to good yields and
excellent β-diastereoselectivities (table 3, entries 6 and 7) in
short reaction times (1 h). Interestingly, oxidation of the
same steroidal substrates with the analogous iron
5
β-selectivities in the epoxidation of Δ -unsaturated steroids.
1
5
dMM
Me2N
complexes (S,S)-
1Fe and (S,S)-
1Fe afforded
Acknowledgment. Financial support from ERC-2009-
StG-239910, MINECO of Spain (CTQ2012-37420-C02-
an equimolar ratio of β/R epoxides and modest selectivity
toward the β epoxide, respectively.
01/BQU and CSD2010-00065), and the Catalan DIUE
In conclusion, the present work describes highly efficient
and stereoselective manganese catalysts for the epoxida-
tion of olefins with H O under mild conditions and short
(2009SGR637). J.Ll.-F. thanks MICINN for a RyC con-
tract. X.R. and M.C. acknowledge ICREA-Academia
awards. X.R. is grateful for financial support from IN-
NPLANTA Project No. IPN-2011-0059-PCT-42000-ACT1.
I.G.-B. acknowledges an IOF Marie Curie fellowship. We
acknowledge STRs from UdG for technical support.
2
2
reaction times. The work demonstrates that electronic
effects can be used to improve the activation of H O in
a productive manner, as well as the steroselectivity in
2
2
(
9) Ichikawa, H.; Takada, Y.; Shishodia, S.; Jayaprakasam, B.; Nair,
M. G.; Aggarwal, B. B. Mol Cancer Ther. 2006, 5, 1434.
Supporting Information Available. Experimental de-
tails for preparation of catalysts, epoxidation reactions,
(
(
(
(
(
10) Parish, E. J.; Li, H.; Li, S. Synth. Commun. 1995, 25, 927.
11) Baqi, Y.; Giroux, S.; Corey, E. J. Org. Lett. 2009, 11, 959.
12) Yang, D.; Jiao, G.-S. Chem.;Eur. J. 2000, 6, 3517.
13) Zhang, J.-L.; Che, C.-M. Chem.;Eur. J. 2005, 11, 3899.
14) Bruyneel, F.; Letondor, C.; Bast u€ rk, B.; Gualandi, A.; Pordea,
0
Me2N
and product characterization. X-ray data for (R,R )- 1
CIF). This material is available free of charge via the
(
Internet at http://pubs.acs.org.
A.; Stoeckli-Evans, H.; Neier, R. Adv. Synth. Catal. 2012, 354, 428.
15) Cuss oꢀ , O.; Garcia-Bosch, I.; Ribas, X.; Lloret-Fillol, J.; Costas,
(
M. J. Am. Chem. Soc. 2013, 135, 14871.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX