Asymmetric Autoamplification in the Oxidative Kinetic Resolution of Secondary Benzylic Alcohols Catalyzed by Manganese Complexes

Article abstract of DOI:10.1002/cctc.201700438

Herein, chiral Mn–aminopyridine complexes have been shown to catalyze the oxidation of alkylarenes to enantiomerically enriched 1-arylalkanols with hydrogen peroxide. The observed enantiomeric excess values result from the direct enantioselective benzylic C?H hydroxylation, accompanied by stereoconvergent oxidative kinetic resolution of the resulting alcohol. Testing several (S,S)-bipyrrolidine derived Mn complexes has revealed a novel catalyst (6) that exhibits the best kinetic resolution in the series (k_rel up to 8.8), along with sufficient reactivity and efficiency (>1000 catalytic turnovers). The mechanistic study of the Mn-mediated alcohol oxidation witnesses electrophilic active species (ρ=?1.2), with rate-limiting H abstraction (k_H/k_D=2.2), followed by oxygen rebound and dehydration of the resulting gem-diol to form the ketone. Intriguingly, while for the resolution of the relatively bulky 1,2-diphenylethanol, k_rel is virtually constant throughout the reaction, for less bulky alcohols, k_rel increases with increasing conversion, in line with the rising optical purity of the 1-arylalkanol. The latter participates in the oxidation as an auxiliary ligand, assisting the chiral recognition. This effect is related to the previously described asymmetric autocatalysis and asymmetric autoinduction, but is not identical with either of those, with the differences being discussed. To unambiguously identify this effect, the term asymmetric autoamplification (chiral autoamplification) is proposed.

Full text of DOI:10.1002/cctc.201700438

Accepted Article
Title: Asymmetric Autoamplification in the Oxidative Kinetic Resolution
of Secondary Benzylic Alcohols Catalyzed by Manganese
Complexes
Authors: Evgenii Talsi, Denis Samsonenko, and Konstantin Petrovich
Bryliakov
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201700438
Link to VoR: http://dx.doi.org/10.1002/cctc.201700438
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201700438
Asymmetric Autoamplification in the Oxidative Kinetic Resolution of Secondary
Benzylic Alcohols Catalyzed by Manganese Complexes
[
a,b]
[a,c]
[a,b]
Evgenii P. Talsi,
Denis G. Samsonenko, and Konstantin P. Bryliakov
*
[
[
[
a]
b]
c]
Novosibirsk State University, Pirogova 2, Novosibirsk 630090, Russian Federation
Boreskov Institute of Catalysis, Pr. Lavrentieva 5, Novosibirsk 630090, Russian Federation
Nikolaev Institute of Inorganic Chemistry, Pr. Lavrentieva 3, Novosibirsk 630090, Russian
Federation
E-mail: bryliako@catalysis.ru
Abstract
Herewith, chiral Mn-aminopyridine complexes have been shown to catalyze the oxidation of
alkylarenes to enantiomerically enriched 1-arylalkanols with hydrogen peroxide. The observed
enantiomeric excess values result from the direct enantioselective benzylic C-H hydroxylation,
accompanied by stereoconvergent oxidative kinetic resolution of the resulting alcohol. Testing
several (S,S)-bipyrrolidine derived Mn complexes has revealed a novel catalyst (6) exhibiting the
best kinetic resolution in the series (k up to 8.8), along with sufficient reactivity and efficiency
rel
(> 1000 TN). The mechanistic study of the Mn mediated alcohol oxidation witnesses electrophilic
active species (ρ = -1.2), with rate-limiting H abstraction (k /k = 2.2), followed by oxygen-
H
D
rebound and dehydration of the resulting gem-diol to form the ketone. Intriguingly, while for the
resolution of the relatively bulky 1,2-diphenylethanol, krel is virtually constant throughout the
reaction, for less bulky alcohols, the k increases with increasing conversion, in line with the rising
rel
optical purity of the 1-arylalkanol. The latter participates in the oxidation as an auxiliary ligand,
assisting the chiral recognition. This effect is related to the previously described asymmetric
autocatalysis and asymmetric autoinduction, but is not identical with either of those, which
different is being discussed. To unambiguously identify this effect, the term asymmetric
autoamplification (chiral autoamplification) is proposed.
1
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
Introduction
Direct catalytic oxygenation of unactivated C-H groups has recently attracted great attention: this
approach is foreseen to provide novel efficient methodologies for the introduction of oxygen atom
[
1]
into complex organic molecules at late steps of multistep syntheses. One of the major trends has
been the design of bioinspired transition metal (mostly Fe based) catalysts, capable of conducting
[1k]
the C-H oxygenation with environmentally benign oxidants, i.e. H O and O .
In the last few
2
2
2
years, we have shown that Mn aminopyridine complexes with C -symmetric bipyrrolidine-derived
2
[
2a]
ligands efficiently oxidize unactivated C-H groups in alkanes with H O , and studied the C-H
2
2
[
2b,c]
hydroxylation mechanism.
000 TN), many times higher than for the structurally similar Fe complexes.
The next logical step could be the design of catalyst systems for the asymmetric C-H
The Mn based catalysts have exhibited very good efficiencies (up to
[2]
1
oxygenation of methylenic groups, aiming at achieving direct enantioselective access to optically
pure secondary alcohols from prochiral alkanes. To date, there have been very few reported
examples of catalyzed enantioselective C-H oxidations, affording the corresponding non-racemic
secondary alcohols. Regarding “green” processes, Simonneaux reported a water-soluble Mn-
porphyrin-catalyzed oxygenation of benzylic positions of several arylalkanes with H O , with
2
2
[
3a-c]
moderate enantioselectivities (32 to 57 % ee, depending on the alkane).
The catalytic protocol
required rather high (2.5 mol. %) catalyst loading and a 5-fold excess (relative to alkane) of H O ,
2
2
thus leaving space for further developments. Very recently, Bietti and Costas with co-workers have
published an intriguing Mn-catalyzed enantioselective ketonization of monosubstituted
cyclohexanes with H O , in which the asymmetric center did not coincide with the reaction center
2
2
[
3d]
(the 2° C atom which underwent catalytic oxidation).
This work had been initiated with the aim to examine the catalytic behavior of the
[
2,4]
bioinspired chiral Mn-aminopyridine complexes
in the stereoselective hydroxylation of prochiral
methylenic sites with H O . However, in the course of catalyzed benzylic oxidation of
2
2
ethylbenzenes, the asymmetric induction has been found to arise from both enantioselective C-H
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
hydroxylation and subsequent stereoconvergent kinetic resolution of the resulting alcohols,
affording enantiomerically enriched alcohols with up to 61.5 % ee (for the oxidation of
ethylbenzene). The kinetic resolution step has been studied in detail; the mechanism of the
stereoselective alcohol oxidation and ketone formation has been examined, and an important kinetic
feature of the oxidative alcohol resolution – asymmetric autoamplification – has been observed and
rationalized. A new catalyst, bearing 1,1,1-trifluoroethoxy substituents at the ligand pyridine rings
has been designed, combining the highest C-H oxidation enantioselectivity and the highest k in
rel
the family (up to 8.8), and exhibiting good efficiencies (> 1000 TN).
Results and discussion
To examine the catalytic properties of Mn-aminopyridine complexes in the asymmetric oxidation of
[4e]
prochiral CH sites, we have screened a series of complexes 1-5 (Scheme 1), using ethylbenzene
2
as the substrate (Table 1). For Mn-aminopyridine catalyzed epoxidations and C-H oxidations, it is
standard to add carboxylic acids as co-catalytic additives (most often it is AcOH), enhancing the
[
2,3d,4,5]
catalytic activity, as well as chemo- and stereoselectivity.
In this work, the use of AcOH
(AA) as additive has resulted in the formation of alcohol with small optical purity (~0 to 32 % ee,
entries 1-5 of Table 1). Interestingly, the new catalyst 6 (Figure 1) has demonstrated the highest
alcohol yield in the series, along with the highest optical purity (34 % ee, entry 6 of Table 1).
Complex 6 has been obtained in single-crystalline form; the asymmetric unit of the structure
of 6 contains four neutral cis-[Mn(dpf)(CF SO ) ] complexes (Figures 1 and S1, SI). Mn(II) atoms
3
3 2
are in octahedral coordination environment built with for N atoms of organic ligand and two O
−
atoms of CF SO anions. Mn–N distances are in the range 2.212(10)–2.299(10) Å (average
3
3
2.244(25) Å). Mn–O distances are in the range 2.134(10)–2.170(8) Å (average 2.150(13) Å).
For further screening of different carboxylic acids, catalysts 6 and 3 were selected, as
affording the alcohol with the highest ees (Table 1, entries 3 and 6). In the case of 3, the
replacement of AA with EHA did not improve the optical purity of the alcohol (entry 7 of Table 1),
but deteriorated the alcohol yield. In contrast, the use of EHA instead of AA with catalyst 6 led to
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
the same alcohol yield and the A/K ratio, and the ee increased to 41 % (entry 8 of Table 1).
Screening other carboxylic acids, either linear, or branched (Table 1, entries 8-11), resulted in either
lower alcohol yields, or in lower enantiomeric purities, or both.
Y
X
X
N
Mn
N
N
Mn
N
N
Mn
N
OTf
OTf
OTf
OTf
N
N
N
N
OTf
OTf
N
N
X
X
Y
1
2
X= H
X= Ph
4
X= Me, Y= OMe
3
5 X= Me, Y= NMe2
6
X= H, Y= OCH CF3
2
O
O
O
O
O
OH
AA
OH
OH
OH
OH
EHA
DMBA
BA
EBA
Scheme 1. Structures of Mn(II) aminopyridine complexes (top) and carboxylic acid additives
bottom) used in this study. AA – acetic acid, BA – butyric acid, EBA – 2-ethylbutyric acid, EHA –
-ethylhexanoic acid, DMBA – 2,2-dimethylbutyric acid.
(
2
Figure 1. Molecular structure of complex 6. Thermal ellipsoids drawn at 50 % probability level.
Three (out of four) complexes of the asymmetric unit, hydrogen atoms, and rotational disorder of
triflates have been omitted for clarity. Full asymmetric unit of 6 is displayed in Figure S2 (SI).
CCDC 1455425 (6) contains the supplementary crystallographic data for this paper. These data can
be obtained free of charge from the Cambridge Crystallographic Data Centre via
ww.ccdc.cam.ac.uk/data_request/cif and from the authors.
4
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
Several p-substituted ethylbenzenes have been oxidized over catalyst 6 in the presence of
EHA (Table 1, entries 9-11). Good substrate conversion and alcohol yield was observed for the
oxidation of p-ethyltoluene (Table 1, entry 9), and the ee was somewhat higher than in the oxidation
of ethylbenzene under the same conditions (Table 1, entry 8).
[
a]
Table 1. Oxidation of p-substituted ethylbenzenes with H O in the presence of complexes 1-6.
2
2
HO
O
Mn cat.
CH3CN
+
+ byproducts
(additive)
H2O2
X
X
X
[c]
No
X
Catalyst Additive
Substrate
Yield of alcohol /
A / K
ee of alcohol
[b]
[d]
conversion ketone / other (µmol)
(%)
(
µmol)
1
2
3
4
5
6
7
8
9
H
1
2
3
4
5
6
3
6
6
6
6
6
AA
AA
17
10 / 5 / 2
7 / 3 / 2
2.0
2.3
1.5
1.0
0.4
0.8
2.2
0.8
1.6
2.6
0.6
0.4
14
6
H
12
23
27
41
37
13
30
44
14
18
8.5
H
AA
12 / 8 / 3
12 / 12 / 3
11 / 24 / 6
15 / 19 / 3
9 / 4 / −
32
~0
10
34
32
41
50
32
46
62
H
AA
H
AA
H
AA
H
EHA
EHA
EHA
EHA
EHA
−
H
13 / 17 / −
25 / 16 / 3
4 / 1.5 / 9
7 / 11 / −
2.5 / 6 / −
CH3
OCH3
Br
H
1
1
1
0
1
2
[
a]At 0 °C; [substrate]:[H O ]:[additive]:[Mn] = 200 µmol : 70 µmol : x µmol : 0.1 µmol (x = 600
2
2
µmol for AA and BA or 100 µmol for other acids), oxidant added via syringe pump over 30 min
and the mixture stirred for an additional 3 h, and analyzed by chiral HPLC. [b] Other products were:
esters; traces of (1-hydroperoxyethyl)benzene (entry 5). [c] Alcohol/ketone ratio. [d] Determined by
chiral HPLC; (S)-absolute configuration.
Quite unexpectedly, the catalytic reaction without any additives has afforded (S)-1-
phenylethanol with significantly higher optical purity (62 % ee), albeit with low yield and
conversion (entry 12 of Table 1).
5
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
This catalytic behavior is quite intriguing, in particular the observation of the highest alcohol
ees in the absence of additives. The formation of significant amounts of ketones suggests that there
can be kinetic resolution of the initially formed alcohol, affecting the final optical purity of the
alcohols. To check this possibility, the oxidation of 1-phenylethanol in the presence of catalyst 6
has been examined (Table 2, entries 1-5). With additives of sterically demanding carboxylic acids
(EHA, EBA, DMBA), negligible (or even zero in the case of EHA) kinetic resolution has been
documented (Table 2, entries 1-3). This result is evidence that the asymmetric induction in the
oxidation of ethylbenzene in the presence of additive EHA (Table 1, entry 8) is generated
exclusively at the enantioselective oxidation step, without contribution of kinetic resolution.
On the contrary, in the presence of AA (Table 2, entry 4) or in the absence of additives
(Table 2, entry 5), kinetic resolution takes place, with selectivity factors being significantly higher
than 1. Catalysts 1-5 have also been tested without additives, all of them exhibiting lower k , and in
rel
most cases lower conversions than 6 (Table 2, entries 6-10). Oxidation of 1-phenylethanol has been
studied at different temperatures (Table 2, entries 5 and 11-15). At -10 °C, the stereodiscrimination
was most efficient, so further experiments were conducted at that temperature.
Several secondary benzylic alcohols, with different electronic properties (modulated by the
p-substituent X) and steric demand (governed by susbstituent Y) have been resolved (Table 2, cf.
entries 13 and 16-24) under the same conditions. In most cases, the selectivity factors have been
found inferior as compared with 1-phenylethanol, and the conversions were typically < 50 %. A
notable exception is p-ethyltoluene, which has been resolved with nearly the same conversion and
with a higher k = 8.0 (Table 2, entry 16).
rel
[
a]
Table 2. Kinetic resolution of p-substituted 1-phenylethanols in the presence of Mn complexes.
HO
Y
HO
Y
HO
Y
Y
O
Mn cat.
CH CN
+
+
3
+ byproducts
(additive)
H O
X
2
2
X
X
X
[b]
[c]
No
X
Y
Catalyst T (°C) Additive
Conversion (%)
ee (%)
krel
6
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
1
2
3
4
5
H
H
H
H
H
H
H
H
H
H
6
6
6
6
6
0
0
0
0
0
EHA
EBA
DMBA
AA
51
52
42
54
52
0
5.5
2
1.0
1.15
1.1
32
49
2.3
−
4.1
6
7
8
9
1
H
H
H
H
H
H
H
H
H
H
1
2
3
4
5
0
0
0
0
0
−
−
−
−
−
22
12
20
54
51
12
0
2.7
1.0
2.1
2.8
1.9
8
39
22
0
1
1
1
1
1
1
2
3
4
5
H
H
H
H
H
H
H
H
H
H
6
6
6
6
6
-30
-20
-10
+10
+20
−
−
−
−
−
48
50
53
53
58
31
40
61
46
43
2.7
3.3
6.2
3.7
2.8
16
17
18
19
20
21
22
CH3
OCH3
Cl
H
H
6
6
6
6
6
6
6
-10
-10
-10
-10
-10
-10
-10
−
−
−
−
−
−
−
52
39
34
44
59
44
41
65
27
22
29
30
36
34
7.6
3.2
3.0
2.9
2.0
3.7
3.9
H
F
H
CF3
H
H
CH3
H
CH Ph
2
2
3
4
6
6
-10
-10
−
−
40
44
24
17
2.6
1.8
2
[
a] [substrate]:[H O ]:[additive]:[Mn] = 100 µmol : 60 µmol : x µmol : 0.1 µmol (x = 600 µmol for
2
2
AA and BA or 100 µmol for other acids). For experiments at T ≥ 0 °C, oxidant was added via
syringe pump over 30 min and the mixture was stirred for additional 3 h. For experiments at T < 0
°
C, oxidant was added via syringe pump over 60 min and the mixture was stirred overnight. The
products were analyzed by chiral HPLC; trace amounts (< 0.2 %) of byproducts ((R)- and (S)-acetic
acid esters) were found in the case of 1-phenylethanol. [b] Determined by chiral HPLC. Entries 1-
21: (S)-absolute configuration. Entry 22: absolute configuration not assigned. [c] The
stereoselectivity factor was calculated as k = ln[(1-c)(1-ee)]/ln[(1-c)(1+ee)] (in which c is the
rel
conversion of secondary alcohol and ee is the enantiomeric excess of the secondary alcohol).
7
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
We notice that this is the first example of kinetic resolution of secondary alcohols in the
presence of bioinspired Mn aminopyridine catalysts. To test the synthetic power of the procedure,
two semipreparative-scale kinetic resolutions have been performed, using 0.1 and 0.06 mol. %
catalyst loadings, to afford (S)-arylethanols with 92 and 95 % ee (Scheme 2). The isolated yields
were not higher than 52.5 %, owing to the relatively small (i.e. < 10) k and inescapable product
rel
losses during the extraction and separation stages. At the moment, the selectivity factors
demonstrated by catalyst 6 are lower than for the most efficient to-date, palladium-sparteine based
catalysts systems; for the latter, the krel for p-substituted 1-phelylethanols in some cases exceed
[
6]
1
0. On the other hand, taking into account the drawbacks of the Pd-sparteine systems, i.e. high
catalyst loadings (5 mol. % Pd and up to 20 mol. % sparteine) and poor reactivity (reaction times of
[
6]
2
4 to 192 h at 80 °C and at 1 atm of O ), we believe that the simple, highly efficient and reactive
2
Mn systems hold a good practical promise; hopefully, with further ligand tuning, improved
stereoselection can be achieved.
OH
6
OH
O
OH
(0.1 mol. %)
workup
+
H O (0.9 mmol)
2
2
CH CN
3
5
mmol
92 % ee
krel = 7.1
(
S)-1-phenylethanol
92 % ee
6
9 % conversion
690 TN)
1
54 mg (1.13 mmol, 45 %)
(
OH
6 (0.06 mol. %)
H O (0.7 mmol)
OH
OH
O
workup
+
2
2
CH CN
3
5
mmol
95 % ee
krel = 8.8
(S)-1-(p-tolyl)ethanol
9
5 % ee
6
8 % conversion
1
79 mg (1.31 mmol, 52.5 %)
(
1133 TN)
Scheme 2. 5-mmol scale preparative kinetic resolutions of 1-arylethanols (for details see SI).
Previously, we have studied the mechanism of C-H oxidation in the presence of Mn-
aminopyridine complexes of the type 1-6, concluding that the reaction proceeded via rate-limiting
V
hydrogen-atom abstraction by the electrophilic, presumably [LMn =O] active species, followed by
[
2b]
oxygen rebound. It is intriguing to establish whether the mechanism of kinetic resolution embody
8
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
the same key step. To address this issue, competitive oxidations of p-substituted 1-phenylethanols
on catalyst 6 have been conducted. This task is not trivial since in the absence of additives, the Mn
catalyst exhibits different reactivities toward the (R)- and (S)- enantiomers of the substrate.
Therefore, for correct kinetic analysis, EHA has been used as additive, to suppress the
stereodiscrimination of catalyst 6 (Table 2, entry 5 vs. entry 1) and equalize the reactivities of the
active oxygen transferring species toward the (R)- and (S)- alcohols. The obtained log (k /k ) has
X
H
shown good linear correlation vs. σ (Figure 2), reflecting electrophilic active species (ρ = -1.18).
p
Such ρ value well fits within the typical range reported for the oxidation of benzylic alcohols by
[
7]
high-valent transition metal complexes (-0.8 to -2.0).
The observed Hammett correlation is somewhat different from that for the oxidation of p-
+
[2b]
substituted cumenes, for which log (k /k ) correlated with the polar Brown parameter σ .
In
X
H
p
+
principle, the absence of correlation with σ for the alcohol oxidations may argue against hydride-
p
proton transfer (HPT) mechanism, which assumes the formation of protonated ketone intermediate
[
8]
at the rate-limiting step. The hypothesis of hydrogen atom transfer (HAT) mechanism seems more
feasible, assuming that the presence of oxygen atom nearby the reaction centre efficiently assists in
compensation of the partial positive charge at the rate-limiting (presumably C-H abstraction) step.
p-МеО
0,4
0,2
0,0
p-Ме
p-Н
p-F
-0,2
-0,4
-0,6
-0,8
p-Cl
p-CF
3
y = -1.18x - 0.015
R = 0.990
2
-0,4
-0,2
0,0
0,2
0,4
0,6
σp
Figure 2. Hammett plot for the oxidation of p-substituted 1-phenylethanols with H O in the
2
2
presence of 6. Conditions: 0 °C, [substrate 1]:[substrate 2]:[H O ]:[EHA]:[Mn] = 0.100
2
2
mmol : 0.100 mmol : 0.135 mmol : 0.100 mmol : 0.2 µmol, oxidant was added via syringe pump
over 1 h, the mixture was stirred for an additional 3 h and analyzed by NMR and chiral HPLC.
9
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
To check whether C-H abstraction is indeed the rate-limiting step, primary kinetic isotope
effect was measured. To this end, two alternative approaches have been exploited. The first one
(method A) is based on the reaction sequence shown in the top of Scheme 3. This approach has
R
R
S
H
S
D
afforded a couple of KIE values, k /k , and k /k (Table 3; see Supporting Information for
H
D
details). The latter values are very close, 2.3 and 2.2, respectively (Table 3, entry 1). The alternative
approach (method B: Scheme 3, bottom) is based on the suppression of stereodiscrimination by
using EHA as additive; the resulting k /k of 2.2 (Table 3, entry 2) very well correspond to the KIE
H
D
values evaluated by method A. With catalyst 4, a similar k /k of 2.2 has been revealed (entry 3 of
H
D
Table 3). Such values are 1.5 times smaller than for the oxidation of C-H bond of cumene, and
[
2b]
almost two times smaller than for the oxidation of cyclohexane over Mn-aminopyridine catalysts.
On the other hand, the k /k values of 2.2 fall within the range (> 1 to ca. 12-15) typical for
H
D
benzylic alcohol oxidations, proceeding via rate-determining cleavage of the benzylic C-H
[
7b,7d,7e,8a,9a,b]
bond,
but are much smaller than the great k /k (36 to > 50) reported for alcohol
H D
[
8b,c]
[9d]
oxidations with oxoruthenium(IV)
and oxoiron(IV) complexes;
the latter high values
[9b]
required a separate discussion.
Scheme 3. The formal reaction schemes used for KIE measurements according to methods A and
B. For details see Supporting Information.
1
0
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
[a]
Table 3. KIE values for the kinetic resolution of 1-phelylethanol with H O .
2
2
No method complex
additive
−
KIE
R
R
S
H
S
D
1
2
3
A
B
B
6
6
4
k /k = 2.3; k /k = 2.2
H
D
EHA (0.2 mmol)
EHA (0.2 mmol)
k /k = 2.2
H
D
k /k = 2.2
H
D
[
a] Conditions: -10 °C, oxidant added via syringe pump over 1 h, the mixture stirred for an
additional 3 h and analyzed by NMR and chiral HPLC. For details see Supporting Information.
Given the rate-determining abstraction of the C-H hydrogen, two major alternative scenarios
leading to the corresponding acetophenone may be considered, namely (A) second H abstraction
(from e.g. OH group) or (B) oxygen rebound, followed by dehydration of the resulting unstable
1
8
intermediate gem-diol (Scheme 4, top). Unfortunately, O labeling experiments are of little utility
for discrimination between these alternatives since ketones readily exchange with water (present in
[2b]
abundance) under our experimental conditions. Instead, it looks promising to replace hydroxyl
hydrogen with alkyl substituent, which would preclude the second H abstraction pathway A and
leave only pathway B available: the intermediate hemiketal is unstable in protic media toward
hydrolysis with subsequent dehydration, eventually affording the ketone (Scheme 4, bottom).
Two different ethers of 1-(p-methoxyphenyl)ethanol have been oxidized in the presence of
complex 6: both afforded the corresponding acetophenone (Table 4), thus confirming the feasibility
of the H abstraction / oxygen rebound pathway (B). This result itself, however, could not be
considered as irrefutable evidence against the co-existence of second H abstraction pathway (A). At
nd
the same time, the 2 H could also be abstracted from the CH group rather than from the OH
3
group (see pathway (C) in Scheme 4, top and bottom), given the 3-fold prevalence of the potentially
[
10]
reactive C-H groups and rather close bond dissociation energies of the C-H and O-H bonds.
However, (1-alcoxyethenyl)anisole (pathway (C) in Scheme 4, bottom) has not been the reaction
nd
product (Table 4), thus excluding 2 H abstraction from the CH group and leaving pathway B as
3
the only route for the ketone formation in the oxidation of 1-arylalkanol ethers. Altogether, these
1
1
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
data give evidence in support of the viability of the hydrogen abstraction / oxygen-rebound / gem-
diol dehydration mechanism of acetophenone formation.
Scheme 4. Proposed mechanism of alcohols (top) and related ethers (bottom) oxidation in the
presence of Mn-aminopyridine catalysts.
[
a]
Table 4. Oxidation of ethers of 1-phelylethanols with H O in the presence of complex 6.
2
2
RO
O
Mn cat.
CH CN
3
(additive)
H O
X
2
2
X
No
1
X
R
additive
−
conversion (%) yield (%)
OMe
iPr
6
5
[
b]
b]
2
3
OMe
OMe
nBu
nBu
−
11
19
10
17
[
EHA (0.1 mmol)
[
a] Conditions: 0 °C, [substrate]:[H O ]:[Mn] = 0.100 mmol : 0.55 mmol : 0.1 µmol, oxidant added
2
2
via syringe pump over 0.5 h, the mixture stirred for an additional 3 h and analyzed by NMR and
chiral HPLC. [b] Two sequential additions of H O (2×0.55 mmol) were performed.
2
2
Unexpectedly, when conducting the kinetic resolutions of 1-phenylethanol, we have
discovered that the observed krel is not constant during the reaction but increases in line with
1
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
increasing conversion (Figure 3A, black line). The krel for the resolution of p-methyl-1-phenyl
ethanol has varied in the same manner (Figure 3A, magenta line). Apparently, such behavior
reflects some alteration of the reaction conditions in the course of the reaction. This alteration could
not be due to the increase of concentration of H O in the mixture, since externally added water
2
(0.56 mmol) caused no effect on krel and the variation of krel with conversion. Moreover, the
resolution of a bulkier 1,2-diphenylethanol proceeded without significant variation of k , which
rel
remained in the range 3.9 to 4.1 for conversions between 40 and 70 % (Figure 3A, blue line).
However, there is another parameter which changes monotonously in the course of oxidation
–
the enantiomeric purity of the chiral alcohol (Figure 3B, black lines, and Figure S2, SI). If the ee
of the alcohol were connected with the resolution effectiveness, one could expect that the use of
non-racemic 1-phenyl ethanol would change the observed dependence of k on the conversion. In
rel
full agreement with this prediction, this dependence has visibly changed when (S)-1-phenylethanol
(15.4 % ee) has been taken instead of its racemic counterpart (Figure 3B): for the case of non-
racemic starting material, the observed k were always higher at similar conversions.
rel
A
9
8
7
6
5
4
3
B
1
0
100
80
60
40
20
0
8
6
4
2
0
0
10
20
30
40
50
60
70
80
0
10 20 30 40 50 60 70 80
conversion, %
conversion, %
Figure 3. Variation of the krel vs. conversion for 1-phenylethanol (black), 1-(p-
methylphenyl)ethanol (magenta), 1,2-diphenylethanol (blue) – (A). Variation of the k and ee vs.
rel
conversion for the resolution of racemic 1-phenylethanol (black) and non-racemic (15.4 % ee, (S)-
configuration) 1-phenylethanol (red) – (B). Conditions: -10 °C, catalyst 6 (0.1 mol %), portionwise
addition of oxidant, followed by HPLC analysis after each addition.
1
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
The mechanism of the influence of the enantiomeric composition of the substrate on the
kinetic resolution is very intriguing. Taking into account the observed effect of added carboxylic
acids on the stereoselection (Table 2, entries 1-5), which apparently occurs through coordination of
[
4b,e,g]
the acid to the catalytically active sites,
it would be logical to expect that in the absence of
acid, the substrate itself may coordinate to the active sites via the hydroxy moieties. Such situation
should lead to the generation of a pair of diastereomeric active sites, on default exhibiting different
reactivities and stereoselectivities toward the oxidation of 1-arylethanols (Scheme 5). The absence
of dependence of k for the oxidation of 1,2-diphenylethanol on the conversion may reflect the fact
rel
that coordination of the latter to the Mn center is substantially restricted due to steric congestion.
2
OCH CF3
OCH CF
2
3
HO
OCH CF
2
3
N
Mn
N
N
N
H O
2
OTf
OTf
2
N
N
O
V
N
N
O
V
N
N
Mn
O
N
Mn
HO
HO
O
N
2
OCH CF3
OCH CF
2
3
OCH CF
2
3
HO
HO
kR
rel
kS
rel
HO
O
+
unreacted
S)-1-phenylethanol
(
Scheme 5. Proposed formation of two diastereomeric catalytically active sites upon the
coordination of the enantiomers of 1-phenylethanol to the tentative active species. The enhancement
of stereoselection with the increase of the ee of (S)-1-phenylethanol (Figure 3B) suggests
S
rel
R
k
> k rel.
To verify this hypothesis, it would be logical to add an external ligand to the reaction
mixture, which would occupy the coordination site of the Mn catalyst; if coordination of 1-
1
4
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
arylethanol enantiomers to Mn is suppressed efficiently enough, the observed amplification of krel
with raising conversion should disappear or at least substantially reduce.
First of all, acetic acid was added (Figure 4, black curve). One can see that in this case,
variation of k was negligible, the latter remaining in the range 2.5-2.7 at conversions between 30-
rel
8
0 %. Curiously, there was a visible maximum k at ca. 48 % conversion.
rel
Next, we attempted the addition of chiral external ligands (Boc protected (L)- and (D)-
proline), to probe the chiral environment effects of the stereoselectivity (Figure 4). The observed
stereoselectivities were lower than for the same catalyst without additives (cf. Figure 3B). The
system with 6/Boc-(D)-proline exhibiting higher krel of 4.3 up to 4.7 (Figure 4, blue curve),
indicating “chiral match”. Again the variation of k was very small over the reaction course (<
rel
1
0 %, cf. the 1.5-fold variation (in particular monotonous increase) of k in the systems without
rel
additives, Figure 3). Like with AcOH, the k passed through a maximum at ca. 50 % conversion.
rel
For the system 6/Boc-(L)-proline, the observed k was lower (“chiral mismatch”), and gradually
rel
increased from 1.9 to 2.2 upon the conversion increase from 35 to 81 % (Figure 4, red curve).
Overall, the experiments on the addition of external coordinating ligands have provided evidence in
support of the proposed mechanism of k amplification via interaction with the enantiomers of 1-
rel
arylethanol.
5,0
4,5
4,0
3,5
3,0
2,5
2,0
1,5
1,0
6
/Boc-(D)-proline
6
/AcOH
6
/Boc-(L)-proline
20
30
40
50
60
70
80
90
conversion, %
Figure 4. Variation of the krel vs. conversion for the resolution of 1-phenylethanol by catalyst
systems 6/AcOH (black), 6/Boc-(L)-proline (red), 6/Boc-(D)-proline (blue). For details see
Supporting Information.
1
5
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
The origin of the observed maxima of the k (Figure 4, black and red curves) is intriguing.
rel
Plausible possibility is that the shielding of the catalyst from 1-arylethanol is not perfect, there
being a competition between the external ligand and 1-arylethanol for the coordination site of
manganese. In effect, with increasing (S)-arylethanol/(R)-arylethanol ratio over the reaction course,
V
the share of the (most stereoselective) active site [L*Mn =O·(S)-arylethanol)] grows up, thus
ensuring gradual increase of the k . However, the overall concentration of 1-arylethanol (both (R)-
rel
and (S)-) gradually decreases, so that at some point it becomes insufficient to compete with the
added external ligand for the coordination site of Mn. In effect, the k goes down, working for its
rel
“natural” value (the value which would be expected for the system 6/external ligand, provided that
1-arylethanol could not coordinate to the Mn center).
The described phenomenon represents a very special case of chiral environment
[
4g,11]
amplification,
1-arylethanol acting as both the substrate and the chiral environment, affecting
the stereoselectivity. The 1-arylethanol (S)-/(R)- ratio progressively increases in the course of
oxidation, thus increasing the observed k . We notice that the less reactive (S)-enantiomer is also
rel
the desired product of the kinetic resolution reaction, so its ascending effect on the stereoselectivity
with increasing conversion can be formally regarded as having autococatalytic-type character. To
the best of our knowledge, such phenomenon has not been described asymmetric catalysis. For its
unambiguous identification, we propose the term asymmetric autoamplification (chiral
autoamplification).
[
12a,b,d,e]
We notice that the reported phenomenon is related to asymmetric autocatalysis
and
[
12c-d]
asymmetric autoinduction,
but not identical to them. Scheme 6 illustrates the three concepts.
Basically, the difference between asymmetric autocatalysis and asymmetric autoamplification is
that in the latter case, the chiral product is not itself catalytically active. At the same time, it can
affect the stereochemical course of the reaction, like in the case of asymmetric autoinduction, via
complexation with the chiral catalyst. However, the mechanisms of amplification of chirality over
1
6
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
the reaction course in the latter two situations are different. In the case of asymmetric autoinduction,
the product is accumulated with time in the reaction mixture, thus increasing the concentration of
the [catalyst*·product*] complex, which exhibits higher stereoselectivity than [catalyst*] itself. In
the case of asymmetric autoamplification, the chiral product is not accumulated in the course of the
reaction but is originally present in the reaction as a component of the mixture of enantiomeric
reagents. Therefore, the enhancement of the catalyst’s stereoselectivity (here characterized by
stereodiscrimination, or k ) with the reaction course is in fact caused by the progressive decrease
rel
of concentration of the second component of the reagents mixture, which results in increase of the
share of the more stereoselective complex between the chiral catalyst and the enantiomeric product
[
C*-(S)-A*] compared with the share of the less stereoselective intermediate [C*-(R)-A*] (Scheme
).
5
P*
asymmetric autocatalysis:
A
A
+
+
B
B
P*
P*
product P* = catalyst
asymmetric autoinduction:
[C*-P*]
1. product P* is not catalytically active
2. product affects the stereochemical course of the reaction
asymmetric autoamplification:
1. product is not catalytically active
(
R)-A*
+
[C*-(S)-A*]
+
B
(S)-A*
2. product affects the stereochemical course of the reaction
(S)-A*
3. product {(S)-A*} is component of the mixture of reagents
Scheme 6. The difference between asymmetric autocatalysis, asymmetric autoinduction, and
asymmetric autoamplification. Asterisks indicate chiral reactants/catalysts/products.
Conclusions
It has been shown that bioinspired chiral Mn-aminopyridine complexes catalyze the oxidation of
alkylarenes with H O to the corresponding non-racemic 1-arylalkanols and further to ketones, with
2
2
the hydroxylation enantioselectivity depending on the nature of the externally added carboxylic
1
7
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
acid. 2-Ethylhexanoic acid (EHA) ensures the highest enantioselection in the series, affording
substituted 1-phenylethanols with up to 50 % ee.
It has been revealed that the enantioselective hydroxylation of alkylarenes is accompanied
by the stereoconvergent oxidative kinetic resolution of the resulting benzylic alcohols. The
stereoselectivity of the kinetic resolution also depends on the added carboxylic acid, EHA being the
least efficient additive of the series, leading to complete loss of stereoselection in the oxidation of 1-
phenylethanol. Moreover, the highest k were achieved without added acid at all, approaching the
rel
value of 8.8 for the resolution of 1-(p-methylphenyl)ethanol.
A novel bipyrrolidine derived Mn complex 6 has been synthesized and X-ray characterized,
which combines the highest efficiency in the series and the highest stereoselectivity in both the
enantioselective oxidation of alkylarenes and kinetic resolution of 1-arylalkanols.
The mechanistic study reveals an electrophilic oxidant (Hammett ρ = -1.2), the observed
primary KIE (k /k = 2.2) for the oxidation of 1-d-phenylethanol-1 witnessing rate-limiting C-H
H
D
abstraction step. The data on the oxidation of ethers of 1-aryl ethanols on catalyst 6 provide
evidence in favor of the viability of the C-H abstraction / oxygen-rebound / gem-diol dehydration
mechanism of acetophenone formation.
For the relatively small alcohols (1-phenylethanol and p-methylphenyl-1-ethanol), the
observed krel substantially increases with increasing (S)-/(R)- substrate ratio, reflecting the
participation of the benzylic alcohol in the kinetic resolution. Apparently, this participation is
realized via coordination of the enantiomers of the alcohol to the Mn center, resulting in the
formation of two diastereomeric active sites, possessing different stereoselectivities. As the reaction
proceeds, the (S)-/(R)- ratio monotonously increases, leading to increasing prevalence of one of the
two diastereomeric active sites and hence to monotonous increase of the observed krel with
increasing conversion. To unambiguously distinguish this dynamic substrate-product effect on the
stereoselectivity, we propose the term asymmetric autoamplification (chiral autoamplification).
Further studies, aimed at designing more stereoselective Mn catalysts, the search for substrates
1
8
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
ensuring higher autoamplification impact on the oxidative resolution, and extended elaboration of
the concept of asymmetric autoamplification are in progress.
Acknowledgments
This work was conducted within the framework of budget project 0303-2016-0005 for the Boreskov
Institute of Catalysis. Financial support from the Russian Foundation for Basic Research (project
16-29-10666) is gratefully acknowledged.
Keywords
Asymmetric autoamplification; hydrogen peroxide; kinetic resolution; manganese; oxidation
References
[
1] a) A. E. Shilov, G. B. Shul’pin, Chem. Rev. 1997, 97, 2879-2932; b) A. E. Shilov, A. A.
Shteinman, Acc. Chem. Res. 1999, 32, 763-771; (c) M. Christmann, Angew. Chem. Int. Ed. 2008,
7, 2740-2742; d) T. Newhouse, P. S. Baran, Angew Chem. Int. Ed. 2011, 50, 3362-3374’ e) C.
4
Che, V. Kar-Yan Lo, C. Y. Zhou, J. S. Huang, Chem. Soc. Rev. 2011, 40, 1950-1975; f) L.
McMurray, F. O'Hara, M. J. Gaunt, Chem. Soc. Rev. 2011, 40, 1885-1898; g) J. Yamaguchi, A. D.
Yamaguchi, K. Itami, Angew. Chem. Int. Ed. 2012, 51, 8960-9009; h) T. B. Gunnoe, Alkane C–H
Activation by Single-Site Metal Catalysis (Ed.: P. J. Pérez), Springer, Dordrecht, 2012, pp. 1-15; i)
E. P. Talsi, K. P. Bryliakov, Coord. Chem. Rev. 2012, 256, 1418-1434; j) E. Roduner, W. Kaim, B.
Sarkar, V. B. Urlacher, J. Pleiss, R. Gläser, W. D. Einicke, G. A. Sprenger, U. Beifuß, E. Klemm,
C. Liebner, H. Hieronymus, S. F. Hsu, B. Plietker, S. Laschat, ChemCatChem 2013, 5, 82-112; k)
K. P. Bryliakov, Environmentally Sustainable Catalytic asymmetric Oxidations, CRC Press, Boca
Raton, 2014, pp. 109-119; l) Y. Qiu, S. Gao, Nat. Prod. Rep. 2016, 33, 562-581; m) G. B. Shul’pin,
Catalysts 2016, 6, 50.
1
9
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
[
2] a) R. V. Ottenbacher, D. G. Samsonenko, E. P. Talsi, K. P. Bryliakov, Org. Lett. 2012, 14,
310-4313; b) R. V. Ottenbacher, E. P. Talsi, K. P. Bryliakov, ACS Catalysis 2015, 5, 39-44; c) E.
P. Talsi, R. V. Ottenbacher, K. P. Bryliakov, J. Organomet. Chem. 2015, 793, 102-107.
3] a) H. Srour, P. Le Maux, G. Simonneaux, Inorg. Chem. 2012, 51, 5850-5856 ; b) N. Amiri, P.
4
[
Le Maux, H. Srour, H. Nasri, G. Simonneaux, Tetrahedron 2014, 70, 8836-8842 ; c) R. V.
Ottenbacher, E. P. Talsi, K. P. Bryliakov, Molecules 2016, 21, 1454. (d) Milan, M.; Bietti, M.;
Costas, M. ACS Cent. Sci. 2017, DOI: 10.1021/acscentsci.6b00368.
[
4] a) M. Wu, B. Wang, S. Wang, C. Xia, W. Sun, Org. Lett. 2009, 11, 3622-3625 ; b) O. Y.
Lyakin, R. V. Ottenbacher, K. P. Bryliakov, E. P. Talsi, ACS Catalysis 2012, 2, 1196-1202; c)
B. Wang, C. Miao, S. Wang, C. Xia, W. Sun, Chem. Eur. J. 2012, 18, 6750-6753; d) W. Dai, J. Li,
G. Li, H. Yang, L. Wang, S. Gao, Org. Lett. 2013, 15, 4138-4141; e) R. V. Ottenbacher, D. G.
Samsonenko, E. P. Talsi, K. P. Bryliakov, ACS Catalysis 2014, 4, 1599-1606; f) D. Shen, C. Miao,
S. Wang, C. Xia, W. Sun, Eur. J. Inorg. Chem. 2014, 5777-5782; g) R. V. Ottenbacher, D. G.
Samsonenko, E. P. Talsi, K. P. Bryliakov, ACS Catalysis, 2016, 6, 979-988.
[
5] a) D. Shen, C. Miao, D. Xu, C. Xia, W. Sun, Org. Lett. 2015, 17, 54–57 ; b) A. M. Adams, J. Du
Bois, H. A. Malik, Org. Lett. 2015, 17, 6066–6069 ; c) W. Dai, Y. Lv, L. Wang, S. Shang, B. Chen,
G. Li, S. Gao, Chem. Commun. 2015, 51, 11268-11271.
[
6] a) D. R. Jensen, J. S. Pugsley, M. S. Sigman, J. Am. Chem. Soc. 2001, 123, 7475-7476; b) E. M.
Ferreira, B. M. Stolz, J. Am. Chem. Soc. 2001, 123, 7725-7726.
7] a) K. K. Banerji, J. Chem. Soc. Perkin Trans II 1973, 435-437 ; b) C. M. Che, W. T. Tang, W.
[
O. Lee, K. Y. Wong, T. C. Lau, J. Chem. Soc. Dalton Trans 1992, 1551-1556; c) S. Campestrini, A.
Cagnina, J. Mol. Catal. A: Chemical 1999, 150, 77-86; d) M. M. Whittaker, J. W. Whittaker,
Biochemistry 2001, 40, 7140-7148; e) S. C. Sawant, J. Cho, K. B. Cho, S. H. Kim, M. S. Seo, Y. M.
Lee, M. Kubo, T. Ogura, S. Shaik, W. Nam, Angew. Chem. Int. Ed. 2010, 49, 8190-8194.
2
0
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
[
8] a) M. S. Thompson, T. J. Meyer, J. Am. Chem. Soc. 1982, 104, 4106-4115; b) L. Roecker, T. J.
Meyer, J. Am. Chem. Soc. 1987, 109, 746-754; c) A. Paul, J. F. Hull, M. R. Norris, Z. Chen, D. H.
Ess, J. J. Concepcion, T. J. Meyer, Inorg. Chem. 2011, 50, 1167–1169.
[
9] a) Y. Wang, J. L. DuBois, B. Hedman, K. O. Hodgson, T. D. P. Stack, Science, 1998, 279, 537-
5
40; b) J. Brinksma, M. T. Rispens, R. Hage, B. L. Feringa, Inorganica Chimica Acta 2002, 337,
5-82; c) M. E. Marmion, K. J. Takeuchi, J. Chem. Soc. Dalton Trans, 1988, 2385-2391; d) N. Y.
7
Oh, Y. Suh, M. J. Park, M. S. Seo, J. Kim, W. Nam, Angew. Chem. Int. Ed. 2005, 44, 4235–4239.
10] Cf. e.g. the close (and rather small) C-H and O-H bond energies in •CH CH -H and •CH O-H
[
2
2
2
-1
-1
radical species (35.7 kcal mol and 30.2 kcal mol , respectively): S. J. Blanksby, G. B. Ellison,
Acc. Chem. Res. 2003, 36, 255-263. On the other hand, the difference of C-H and O-H bond
dissociation energies can be estimated based on the difference of stabilities of the keto end enolic
forms of acetophenone, see Figure S3 of the Supporting Information.
[
11] a) T. Hashihayata, Y. Ito, T. Katsuki, Synlett 1996, 1079-1081; b) T. Hashihayata, Y. Ito, T.
Katsuki, Tetrahedron 1997, 53, 9541-9552; c) K. Miura, T. Katsuki, Synlett 1999, 783-785; d) J.
Balsells, P. J. Walsh, J. Am. Chem. Soc. 2000, 122, 1802-1803; e) J. W. Faller, A. R. Lavoie, J.
Parr, Chem. Rev. 2003, 103, 3345-3367.
[
12] a) K. Soai, S. Niwa, H. Hori, J. Chem. Soc. Chem. Commun. 1990, 982-983; b) C. Bolm, F.
Bienewald, A. Seger, Angew. Chem. Int. Ed. Engl. 1996, 35, 1657-1658; c) A. H. Alberts, H.
Wynberg, J. Am. Chem. Soc. 1989, 111, 7265-7266; d) K. Soai, T. Shibata, I. Sato, Acc. Chem. Res.
2000, 33, 382-390; e) K. Soai, I. Sato, Chirality 2002, 14, 548-554.
2
1
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201700438
Layout 2:
FULL PAPER
Evgenii P. Talsi, Denis G. Samsonenko,
Konstantin P. Bryliakov*
9
8
7
6
5
k
rel
Page No. – Page No.
Asymmetric Autoamplification in the
Oxidative Kinetic Resolution of
Secondary Benzylic Alcohols
0
10 20 30 40 50 60 70
conversion, %
Münchhausen-type catalysis: oxidative kinetic resolution of 1-arylethanols in the
presence of chiral Mn-aminopyridine catalysts has been shown to be affected by
coordination of the substrate to the catalytically active sites, which results in
monotonous increase of the krel with increasing conversion. To identify this effect,
the term asymmetric autoamplification is proposed.
Catalyzed by Manganese Complexes
2
2
This article is protected by copyright. All rights reserved.