Ti-salalen mediated asymmetric epoxidation of olefins with H2O2: Effect of ligand on the catalytic performance, and insight into the oxidation mechanism

Article abstract of DOI:10.1016/j.molcata.2016.05.019

The highly enantioselective epoxidation of several conjugated olefins with H₂O₂ in the presence of five chiral titanium(IV) salalen (dihydrosalen) complexes has been studied, with focus on the effects of substituents (electronic and

Full text of DOI:10.1016/j.molcata.2016.05.019

Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Ti-salalen mediated asymmetric epoxidation of olefins with H2O2:
Effect of ligand on the catalytic performance, and insight into the
oxidation mechanism
Evgenii P. Talsia,b, Tatyana V. Rybalovaa,c, Konstantin P. Bryliakova,b,∗
a Novosibirsk State University, Ul. Pirogova 2, Novosibirsk 630090, Russia
b Boreskov Institute of Catalysis, Pr. Lavrentieva 5, Novosibirsk 630090, Russia
c Vorozhtsov Novosibirsk Institute of Organic Chemistry, Pr. Lavrentieva 9, Novosibirsk 630090, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 April 2016
Received in revised form 13 May 2016
Accepted 16 May 2016
Available online 17 May 2016
The highly enantioselective epoxidation of several conjugated olefins with H2O2 in the presence of five
chiral titanium(IV) salalen (dihydrosalen) complexes has been studied, with focus on the effects of sub-
stituents (electronic and steric) on the epoxide yield and enantioselectivity. The introduction of electron
acceptors at the ligand’s remote positions enhances the catalytic reactivity, without drastic effect on
the stereoselectivity and ultimate catalytic efficiency. In turn, the replacement of Ph substituents of the
ligand core with o-MeO-Ph and o-Et-Ph substituents leads to the enhancement in enantioselectivity (up
to 99.8% ee), conversion (up to 100%) and efficiency (up to 125 TN). Kinetic and Hammett data provide
evidence in favour of a stepwise epoxidation mechanism on Ti-salalen catalysts. As compared to Ti-salan
protagonists, Ti-salalen catalysts exhibit much higher activity and efficiency, and in most cases compa-
rable enantioselectivity. Kinetic peculiarities of epoxidations on Ti-salalen and Ti-salan complexes are
Keywords:
Asymmetric oxidation
Hydrogen peroxide
Mechanism
Salalen
considered.
Titanium
© 2016 Published by Elsevier B.V.
1. Introduction
pure cis-1,2-diaminocyclohexane as the chirality source [19,20].
Other groups also contributed a series of works aimed at designing
In the last few years, catalytic asymmetric epoxidation of
unfunctionalized olefins has been attracting undiminishing inter-
est, owing to the need in novel environmentally friendly catalyst
systems based on non-toxic metal complexes (or even metal-free
catalysts) and green oxidants (e.g. H2O2 and O2) [1–6]. Within
transition-metal based catalysts, complexes of titanium are of
great interest due to non-toxicity of this metal (and products
of its hydrolysis), and proven ability to activate H2O2 molecule,
transferring one of its oxygen atoms to olefinic moieties in a
highly asymmetric fashion [7]. Katsuki and co-workers pioneered
the studies of chiral salan (tetrahydrosalen) [8–12] and salalen
(dihydrosalen) [13–16] complexes of titanium(IV) as catalysts for
the highly enantioselective epoxidation of conjugated and termi-
nal olefins. Berkessel and co-workers conducted comparative and
mechanistic studies of Ti-salalen catalyzed epoxidations [17,18]
and developed a library of salalen ligands, derived from optically
and testing related chiral titanium(IV) catalysts [21–25].
Both salan and salalen type complexes exhibited high to
excellent enantioselectivities for the epoxidation of conjugated
cis-olefins (up to >99% ee); furthermore, titanium salalen based cat-
alysts appeared active enough to epoxidize electron-poor terminal
olefins [13,20], suggesting a higher intrinsic reactivity of Ti-salalen
complexes as compared to their Ti-salan analogs, which may be an
important practical advantage. In spite of a plethora of available
catalytic data, systematic analyses of effects of substituents on the
catalytic reactivity and enantioselectivity of Ti-salalen complexes
have been hitherto lacking. In this work, we have undertaken such
a study, focusing on the electronic enfluence of remote electron-
donating and withdrawing substituents, as well as the effect of
introduction of bulky substituents into the ligand structure.
2. Experimental
2.1. Materials and methods
∗
Corresponding author at: Boreskov Institute of Catalysis, Pr. Lavrentieva 5,
Novosibirsk 630090, Russia.
H2O2 was used as a commercial analytical grade 30% aqueous
solution. Silica gel 60 (0.063–0.200 mm) for column chromatogra-
E-mail address: bryliako@catalysis.ru (K.P. Bryliakov).
http://dx.doi.org/10.1016/j.molcata.2016.05.019
1381-1169/© 2016 Published by Elsevier B.V.

132
E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
H
H
N
O
N
O
N
O
N
O
Ti
Ti
X
O
X
O
Ph
Ph
Y
2
Y
2
1 X= H
2 X= Br
4
5
Y= Et
Y= OMe
H
H
H
N
N
O
N
N
O
Ti
Ti
X
O
X
X
O
X
O
O O1/2
Et
Ph
Ph
Ph
Ph
2
2
1'
3'
X= H
X= OMe
3 X= OMe
Scheme 1. Titanium(IV) salalen complexes 1-5 considered in this work, and titanium(IV) salan catalysts 1ꢀ and 3ꢀ [22].
phy was purchased from Panreac. All other chemicals were Aldrich,
AlfaAesar, or Acros commercial reagents.
0.15 mmol of H2O2 was introduced at once and the reaction mix-
ture was stirred at 298 K for 24 h. An additional 0.1 mmol of H2O2
was added, and the mixture was allowed to stir for additional 48 h
at the same temperature. After the standard workup (see above),
the reaction outcome was analyzed by 1H NMR; the epoxide ee
was measured by 1H NMR with c.s.r. Eu(hfc)3. The optical purity of
chalcone epoxide was measured by chiral HPLC (SI).
1H and 13C NMR spectra were measured on a Bruker Avance
400 at 400.13 and 100.613 MHz, or on a Bruker DPX-250 at 250.13
and 62.903 MHz, respectively. Chemical shifts were referenced to
internal standard tetramethylsilane, with positive values in the
low-field direction.
Enantiomeric excess values and absolute configurations of
epoxides were measured on a Shimadzu LC-20 HPLC chromato-
graph equipped with Daicel chiral columns (250 × 4.6 mm). The
experimental uncertainty in ee measurements did not exceed
1.0% for ees falling in the range 74–90%, 0.5% for ees falling
in the range 90–98%, and 0.25% for ees higher than 98%. The
chiral HPLC separation details for epoxides and N-Boc protected
1,2-aminoalcohols can be found in the Supporting information and
references therein.
CCDC 1439962 (4) and 1439963 (5) contain 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.
3. Results and discussion
Complexes 1-5 (Scheme 1) have been studied as five model cata-
lysts of enantioselective epoxidation of olefins. Complexes 1-3 were
recently prepared and X-ray characterized, and reported to catalyze
the highly enantioselective sulfoxidation of pyridylmethylth-
iobenzimidazoles to optically pure proton pump inhibitors
(S)-omeprazole and (R)-lansoprazole with hydrogen peroxide [26].
In the solid state, complexes 1-3 have dimeric structures, with
either two bridging oxygen atoms (complexes 1, 2), or with one
bridging oxygen (complex 3) [26,27].
Complexes 4 and 5 have been first synthesized, isolated and
X-ray characterized in this work. The chiral ligands for 4 and 5
were prepared starting from commercial precursors (SI). X-ray
quality crystals of 4 and 5 were obtained by crystallization from
Et2O/CH2Cl2/pentane or from CHCl3/hexane (SI).
2.2. Typical asymmetric epoxidation procedure
A conjugated olefin (0.10 mmol) was added to the Ti-salan or
Ti-salalen catalyst (5 ꢀmol or 0.8 ꢀmol, respectively) dissolved in
CH2Cl2 (0.4 mL), and 0.05 mL of a saturated aqueous NaCl was
added. The mixture was thermostatted at the desired temperature
(290 K), and 30% aqueous hydrogen peroxide (0.20 mmol of H2O2)
was then introduced in one portion. Stirring (400 rpm) was contin-
ued at that temperature for (typically) 24 h. The reaction mixture
was diluted with aqueous NaCl (1 mL), the organic phase was sepa-
rated, and the aqueous phase was extracted with CH2Cl2 (3 × 1 mL).
The combined organic extract was dried with CaSO4, diluted with
0.6 mL of CCl4, and CH2Cl2 was carefully removed on a rotary evap-
orator at room temperature. The remaining solution was diluted
with 0.1 mL of CDCl3 and analyzed by 1H NMR to reveal the ratio
of olefin, epoxide, and byproduct(s), and by chiral HPLC to measure
the ee of the epoxide.
The asymmetric units of
4 and 5 contain complexes
[Ti2O2(C36H38N2O2)2] and [Ti2O2(C34H34N2O4)2], respectively. In
both complexes, two Ti(IV) cations have distorted octahedral coor-
dination environment formed by two N and two O atoms of an
organic ligand and O atoms of two oxo-ligands. The ethyl-groups of
4 are disordered over two orientations in approximate ratio 87:13
and 76:24. Hydrogen atoms of organic ligands were placed geomet-
rically and refined using a riding model except for the (R)-amine
hydrogen of the 5 which was refined independently. In the asym-
metric unit of 4, the solvent molecule of pentane was also located
and isotropically refined because of disordering. Additionally, there
are two CHCl3 solvent molecules in the asymmetric unit of 5.
As compared to the previously reported complexes 1-3 [26], 4
and 5 bear additional steric crowd at the aryl rings; their molecu-
lar structures are shown in Fig. 1. Curiously, while single-crystalline
complex 4 was homochiral (ꢁ,R,R,SN-ꢁ,R,R,SN)-4, complex 5 exhib-
For the epoxidation of 1-decene (0.1 mmol), the volumes of
CH2Cl2 and aqueous NaCl were 0.1 mL and 0.025 mL, respectively;

E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
133
Fig. 1. Molecular structures of complexes (ꢁ,R,R,SN-ꢁ,R,R,SN)-4 and (ꢁ,R,R,SN-ꢂ,R,R,RN)-5. Thermal ellipsoids are drawn at 30% probability level. Hydrogen atoms and solvent
molecules are omitted for clarity. C grey, Ti white, O red, N blue. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version
of this article.)
ited pseudoheterochiral configuration (ꢁ,R,R,SN-ꢂ,R,R,RN)-5, the
Ti atoms in 5 having different chiralities (Fig. 1). Such sort of iso-
merism was reported previously for structurally related Ti-salan
complexes, where three homo- and heterochiral stereoisomers
were isolated for the same Ti-salan species [28]. In catalytic
epoxidations, however, all those isomers demonstrated equal
enantioselectivities, pointing to the formation of the same active
species, apparently, monomeric titanium-peroxo intermediates
[7,28].
Complex 4 exhibits hydrogen bonding between the N H hydro-
gen of the ligand at one Ti atom and O at the other Ti atom (the
H· · ·O distances 2.54, 2.52 Å, SI), and C H· · ·␲ interactions between
Scheme 2. Proposed mechanism of titanium catalyzed olefin epoxidations. S is sol-
vent molecule or vacancy.
the salalen CH2 bridge of the ligand at one Ti atom and the phenyl
ring of the salalen ligand at the other Ti atom (distances between H
to the centers of rings 2.84 and 2.87 Å, SI). Contrariwise, in 5, only
the (R)-amine H forms such hydrogen bond (H· · ·O distance 2.18(7)
catalysts 4 and 5, higher product yields (65–76%) were achieved for
the epoxidation of styrene, the least electron-rich substrate of the
series of conjugated olefins.
Å).
The catalytic properties of complexes 1-5 have been tested in
the oxidation of five conjugated olefins (Table 1). Like for previ-
ously studied titanium-salan based catalysts [22], the introduction
of remote electron-donating or withdrawing substituents (com-
plexes 1, 2, 3) does not dramatically affect the enantioselectivity.
This observation is in agreement with the two-stage epoxida-
tion mechanism (previously suggested for titanium-salan catalysts
[22]), the oxygen transfer stage occurring in an intramolecular
For comparison, catalytic data for the titanium(IV) salan com-
plex 1ꢀ (used at 5 mol% loading) are given for each substrate. One
can see that the salan catalyst demonstrates a slightly higher enan-
tioselectivity than the corresponding salalen complex 1 for the
epoxidation of styrene and cis-␤-methylstyrene (cf. entries 1 vs.
2 and 7 vs. 8), while for cyclic conjugated olefins the enantioselec-
tivities are virtually identical.
manner, and the enantioselection being virtually unaffected by the
presence of EDG and EWG, depending only on the substrate geom-
be their higher efficiency (in terms of turnover numbers per-
etry (Scheme 2).
The major advantage of the salalen type catalysts appears to
In turn, the introduction of o-substituted aryls instead of 3,3ꢀ-
Ph2 groups [29] in most cases improved the catalytic efficiency
(which was up to 125 TN for indene) and the optical yield
of the epoxides. This gain in enantioselectivity was higher for
styrene and cis-␤-methylstyrene and much smaller for cyclic con-
jugated olefins, which trend is similar to that documented for the
epoxidations catalyzed by titanium-salan complexes [22]. For the
epoxidations of indene, dihydronaphthalene and 2,2-dimethyl-2H-
chromene-6-carbonitrile, the ees obtained with the best catalysts
4 and 5 were good to excellent (96.5–99.8% ee). Interestingly, with
formed): while the epoxidation on titanium-salan based catalysts
required 5 mol% catalyst loading, for titanium-salalens, the use of
only 0.8 mol% loading ensured comparable epoxide yield and, in
most cases, comparable enantioselectivity.
The stepwise mechanism depicted in Scheme 2 assumes that
the presence of electron-acceptors should make the titanium cen-
ter more electrophilic and thus facilitate the rate-limiting olefin
coordination step, thus accelerating the overall reaction [22]. Coor-
dination of olefins to the metal centers is extremely widespread
in transition metal catalysis. It is an essential, in most cases rate-
limiting, step in coordination-insertion olefin polymerization by

134
E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
Table 1
Epoxidation of conjugated olefins with H2O2 in the presence of Ti-salalen catalysts 1–5.a
no
substrate
Ti cat.
conversion/selectivity [%]b
ee [%] (config.)c
1
2
3
4
5
6
7
8
1ꢀ
1e
2e
3e
4
60d
81.5 (S)
78.5 (S)
79.0 (S)
75.5 (S)
85.5 (S)
81.0 (S)
50.0/99.5
34.0/100
25.0/100
76.5/99.5
65.0/100
73d
70.0/100
36.5/100
64.5/95.5
56.0/100
85.0/100
92d
100/90.5
100/89.5
99.0/95.0
98.0/93.5
100/96.0
94d
95.0/98.0
95.5/98.5
95.5/97.5
95.5/98.0
95.5/98.5
76d
5
1ꢀ
1
74.5 (2R,3S)
66.0 (2R,3S)
66.0 (2R,3S)
69.0 (2R,3S)
74.5 (2R,3S)
82.0 (2R,3S)
95.0 (1S,2R)
95.5 (1S,2R)
95.5 (1S,2R)
96.0 (1S,2R)
96.5 (1S,2R)
96.5 (1S,2R)
96.0 (1S,2R)
96.5 (1S,2R)
96.5 (1S,2R)
96.5 (1S,2R)
97.5 (1S,2R)
97.5 (1S,2R)
99.5 (3S,4S)
99.4 (3S,4S)
99.3 (3S,4S)
98.5 (3S,4S)
99.7 (3S,4S)
99.8 (3S,4S)
68.5g
9
2
3
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
5
1ꢀ
1
2
3
4
5
1ꢀ
1
2
3
4
5
1ꢀ
1
88.5/97.5
66.5/99.0
78.0/99.5
80.0/99.0
92.0/99.0
6.0/100
7.0/100
21.0/100
11.5/100
2
3
4
5
2
3
4
5
69.0g
66.0g
f
70.0g
The bold values signifies the numbers of complexes.
a
Reaction conditions as shown in the sketch unless otherwise noted. Oxidant added in one portion.
b
c
Determined by 1H NMR analysis.
Determined by chiral HPLC.
d
Reported yields from Ref. 22; conditions: At 290 K; [H2O2]/[substrate]/[catalyst] = 150 ꢀmol:100 ꢀmol:5 ꢀmol, oxidant was added in one portion and the mixture was
stirred for 18 h.
e
2 + 1 equiv. H2O2, mixture stirred for 48 h (see SI for details).
f
Conditions reported in Section 2.
g
Determined by 1H NMR with chiral shift reagent Eu(hfc)3.
group IV metal complexes (Ti, Zr, Hf) as well as by classical Ziegler-
Natta titanium-based heterogeneous catalysts [30–32]. We note
that Ti-salan and Ti-salalen complexes are well-known catalysts of
olefin polymerization[33–42], which obviously implies their ability
to coordinate olefin molecules.
Plausible reason for the higher activity of Ti-salalen catalysts
(as compared to the salan ones) is the lower electron-donor char-
acter of the salalen ligands, owing to the presence of conjugated
parameters ꢃp. The slope ꢄ was found to be −0.52, which is very
close to the slope for the epoxidation of substituted styrenes in
the presence of salan complex 1ꢀ [22]. At the same time, the enan-
tioselectivity is not visibly affected by the substituents variation,
remaining in the range 85. . .87% (with the only exception of p-F-
styrene for which the ee was only 83%). This behavior is identical to
that documented for Ti-salan epoxidation catalysts [22], and fully
consistent with the stepwise mechanism depicted in Scheme 2,
which implies that the epoxidation rate and epoxidation selectivity
are determined by two separate reaction steps.
It would be logical to expect that catalyst 2, bearing Br sub-
stituents, should be more active as compared to unsusbtituted
1 and OMe-substituted 3. However, our catalytic data (Table 1)
witnessed that for Br-substituted catalyst 2, the epoxide yields
(achieved within 24 h) were in most cases lower than for H- or
OMe- containing catalysts 1 and 3. To disclose the reasons of this
apparent discrepancy, we have monitored the kinetics of epoxida-
tion of cis-␤-methylstyrene in the presence of Ti-salalen catalysts
N
C moiety. The decrease of electron-donor character of the lig-
and, when passing from Ti-salan to Ti-salalen complexes, should
make the latter more electrophilic and facilitate olefin coordination
to the titanium centers, thus speeding up the overall process. Con-
trariwise, the decrease of electron-donor character of the substrates
would be expected to attenuate the rate of substrate coordination
to titanium, thus slowing down the overall process. To illustrate
the influence of electronic factors on the relative epoxidation rates,
five substituted styrenes were taken to competitive epoxidations
with catalyst 4 (Fig. 2). It is seen that the relative epoxidation rates
demonstrate good linear correlation with the Hammett substituent

E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
135
increased to adopt a value which further remained virtually con-
stant during the reaction. At the same time, while for salan systems
there was a steady-state regime, resulting in linear ln([S]/[S]0)
vs. time dependence [22], for the salalen catalyst 3 the epoxi-
dation kinetics was nonsteady, such that no linear dependence
could be observed and no pseudo-first-order rate constants could
be derived. At the same time, the slope of the kinetic curve, pro-
portional to the rate of decrement of olefin concentration, may be
taken as a measure of relative activities of the active species for dif-
ferent catalyst systems. One can see that the kinetic curve for the
system 2/H2O2 (after the induction period) is more steep than for
3/H2O2 (Fig. 3B); however, the former systems deactivates faster.
In effect, the system 3/H2O2, which retains its activity for longer,
eventually (after 24 h) ensures higher conversion (Fig. 3B). For com-
parison, titanium-salan catalyst 3ꢀ showed a much lower activity,
and deactivated much faster, to end up at only 34% conversion
(corresponding to 21 TN) within 24 h.
The nature of this deactivation is most likely complex. On the
one hand, adding another 2 equiv. of H2O2 to the reaction mixtures
in Fig. 3B after 24 h did not affect the olefin/epoxide ratio for the
system 3ꢀ/H2O2 within the next 24 h. On the other hand, a similar
manipulation with the system 2/H2O2 caused an increase of the
olefin conversion from 55% to 65% within the next 24 h. Therefore,
we believe that the reaction deceleration may be due to (1) titanium
catalyst degradation and (2) catalytic decomposition of H2O2 which
competes with the target epoxidation reaction. The contribution of
the latter channel exhibits itself most explicitly in epoxidations of
the least electron-rich conjugated olefin, styrene (Table 1, entries
2–5), for which the documented yields were in most cases (except
for catalysts 4 and 5) achievable within only 48 h, with the addition
of another 1 equiv. of H2O2 after 24 h.
Fig. 2. Hammett plot of Log(kX/kH) (blue) and enantiomeric excess (red) vs ꢃp for
the epoxidation of p-substututed styrenes by the 4/H2O2 system. (For interpretation
of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
A
The higher catalytic activity of the Ti-salalen complexes
(compared to Ti-salan analogs) was previously exploited for
the epoxidation of non-conjugated (especially terminal) olefins
[13,14,19,20]. We have conducted the epoxidation of a non-
conjugated olefin − 1-decene − (entries 31–34 of Table 1). The
enantioselectivities shown by catalysts 2-5 did not vary signifi-
cantly, remaining in the range 66–70%. Under the conditions used,
the yields were much lower than for conjugated olefins; with cata-
lyst 4, the highest epoxide yield of 21% was achieved. In spite of the
low yield, we note that the efficiency of 4 (26 TN) was comparable
to those demonstrated by the Katsuki’s “second-generation” Ti-
salalen catalyst in 1-octene epoxidation (23–42 TN) [13,14]. For the
epoxidation of a typical electron-deficient olefin, trans-chalcone in
the presence of catalyst 2 (under the standard conditions of Table 1),
only a 2% conversion within 24 h was achieved, with the epoxide
having 18% ee (2S,3R).
B
To exemplify the practical utility of Ti-salalen catalyzed epox-
idations, we have attempted the synthesis of optically pure
1,2-aminoalcohols starting from indene, 1,2-dihydronaphthalene
and 2,2-dimethyl-2H-chromene-6-carbonitrile, proceeding via
ring cleavage of the intermediate epoxide with ammonia
(Scheme 3). The nucleophilic attack of NH3 occurred without dete-
rioration of the enantiomeric excess. Boc-protection provides a
practical tool for purification: by recrystallization of the N-Boc-
protected aminoalcohols, the epoxide HPLC purity and the ee can be
further increased up to > 99%. After purification, the Boc protecting
group can be removed under acidic conditions,
Fig. 3. (A) Kinetics of epoxidation of cis-␤-methylstyrene (CH2Cl2, 290 K, 1.6 mol%
of catalyst, [H2O2]/[olefin] = 2.0) by catalyst system 3/H2O2: substrate (S) concen-
tration black, epoxide concentration blue, enantiomeric excess brown. (B) Kinetics
of epoxidation of cis-␤-methylstyrene (CH2Cl2, 290 K) by catalyst systems 3/H2O2
(black), 2/H2O2 (brown) and 3ꢀ/H2O2 (blue). Dashed lines demonstrate the highest
slopes of kinetic curves for catalyst systems 3/H2O2 (black) and 2/H2O2 (brown).
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
to release the desired optically pure aminoalcohol [41]. This
approach to optically pure 1,2-aminoalcohols from chiral epoxides
may be a technically easier alternative to that relying on azidolysis
of the epoxide, followed by H2 reduction of the azidoalcohol (over
Pd/C), with subsequent Boc-protection, purification by recrystal-
lization, and deprotection [42].
2, 3 and (for comparison) Ti-salan catalyst 3ꢀ. The resulting kinetic
curves are presented in Fig. 3.
Like for titanium-salan based catalysts [22], there was some ini-
tial induction period of ca. 100–120 min (Fig. 3A), which apparently
reflects the formation and accumulation of the true catalytically
active sites. Also like for titanium-salan based catalysts, during
the induction period, the epoxidation enantioselectivity (Fig. 3A)

136
E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
1) NH3 (aq.)
3 (0.4 mol. %)
73 % crude yield
OH
O
96.5 % ee
H2O2 (2.0 equiv.)
290 K, 24 h
2) Boc2O
O
after recrystallization:
56 % overall yield
HN
tBuO
>99.5 % conversion
98.5 % selectivity
95.5 % ee
99.7 % HPLC purity
99.6 %
ee
6
3
(0.55 mol. %)
1) NH3 (aq.)
76 % crude yield
96 % ee
H2O2 (2.0 equiv.)
290 K, 24 h
2) Boc2O
O
OH
O
after recrystallization:
50 % overall yield
99 % HPLC purity
99.7 % ee
HN
99 % conversion
93.5 % selectivity
96.5 % ee
tBuO
7
O
68.5 % crude yield
O
1) NH3 (aq.)
2) Boc2O
99.7 % ee
O
5
(0.4 mol. %)
NC
after recrystallization:
55 % overall yield
99.8 % HPLC purity
100 % ee
NC
OH
O
H2O2 (2.0 equiv.)
290 K, 40 h
NC
HN
O
94.5 % conversion
98.5 % selectivity
99.8 % ee
tBuO
8
Scheme 3. Synthesis of optically pure 1,2-aminoalcohols.
4. Conclusions
the Multi-Access Chemical Service Center SB RAS for providing the
equipment for the XRD experiments.
The effects (electronic and steric) of substituents on the product
yield and enantioselectivity in the course of epoxidation of conju-
gated olefins with H2O2 in the presence of five chiral titanium(IV)
salalen (dihydrosalen) complexes has been studied systematically.
The introduction of electron-acceptors at the ligand’s remote posi-
tions enhances the catalytic reactivity (without significant effect
on the stereoselectivity) but causes faster reaction deceleration.
The latter is apparently due to (1) irreversible catalyst degra-
dation as well as (2) Ti-promoted decomposition of hydrogen
peroxide. In turn, the replacement of Ph substituents of the lig-
and with o-MeO-Ph and o-Et-Ph substituents affords more efficient
and stereoselective catalysts. The methoxy-substituted catalyst 5
exhibits the highest enantioselectivities (up to 99.8% ee) and in most
cases the highest yields and efficiencies.
Hammett correlation (Ln of epoxidation rates of p-substituted
styrenes) provides evidence in favour of an electrophilic active
species (ꢀ = −0.52). At the same time, the epoxidation enantiose-
lectivity is virtually independent of the p-substituent. This picture
is consistent with a stepwise mechanism (previously proposed
for epoxidations on similar Ti-salan catalysts [22]), with separate
rate-limiting and enantioselectivity-determining steps. As com-
pared with Ti-salan protagonists, Ti-salalen catalysts demonstrate
higher activity and efficiency, and, depending on substrate, similar
or slightly lower enantioselectivity. Chiral epoxides of cyclic con-
jugated olefins, obtained with the aid of Ti-salalen catalysts, can
be further readily cleaved by ammonia to afford optically active
trans-1,2-aminoalcohols without deterioration of optical purity.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molcata.2016.05.
019.
References
[1] E.P. Talsi, K.P. Bryliakov, Coord. Chem. Rev. 256 (2012) 1418–1434.
[2] H. Srour, P. Le Maux, S. Chevance, G. Simonneaux, Coord. Chem. Rev. 257
(2013) 3030–3050.
[3] K.P. Bryliakov, Environmentally Sustainable Catalytic Asymmetric Oxidations,
CRC Press, Boca Raton, 2014, pp. 5–36.
[4] R.L. Davis, J. Stiller, T. Naiker, H. Juang, K.A. Jørgensen, Angew. Chem. Int. Ed.
53 (2014) 7406–7426.
[5] O. Cussó, X. Ribas, M. Costas, Chem. Commun. 51 (2015) 14285–14298.
[6] C. Wang, H. Yamamoto, Chem. Asian J. 10 (2015) 2056–2068.
[7] E.P. Talsi, A.A. Bryliakova, K.P. Bryliakov, Chem. Rec. 16 (2016) 924–939.
[8] Y. Sawada, K. Matsumoto, S. Kondo, H. Watanabe, T. Ozawa, K. Suzuki, B. Saito,
T. Katsuki, Angew. Chem. Int. Ed. 45 (2006) 3478–3480.
[9] K. Matsumoto, Y. Sawada, T. Katsuki, Synlett (2006) 3545–3547.
[10] K. Matsumoto, B. Saito, T. Katsuki, Chem. Commun. (2007) 3619–3627.
[11] Y. Shimada, S. Kondo, Y. Ohara, K. Matsumoto, T. Katsuki, Synlett (2007)
2445–2447.
[12] K. Matsumoto, T. Oguma, T. Katsuki, Angew. Chem. Int. Ed. 48 (2009)
7432–7435.
[13] K. Matsumoto, Y. Sawada, B. Saito, K. Sakai, T. Katsuki, Angew. Chem. Int. Ed.
44 (2005) 4935–4939.
[14] Y. Sawada, K. Matsumoto, T. Katsuki, Angew. Chem. Int. Ed. 46 (2007)
4559–4561.
[15] K. Matsumoto, T. Kubo, T. Katsuki, Chem. Eur. J. 15 (2009) 6573–6575.
[16] K. Matsumoto, C. Feng, S. Handa, T. Oguma, T. Katsuki, Tetrahedron 67 (2011)
6474–6478.
[17] A. Berkessel, M. Brandenburg, E. Leitterstorf, J. Frey, J. Lex, M. Schäfer, Adv.
Synth. Catal. 349 (2007) 2385–2391.
[18] A. Berkessel, M. Brandenburg, M. Schäfer, Adv. Synth. Catal. 350 (2008)
1287–1294.
Acknowledgments
The authors gratefully acknowledge the financial support from
the Russian Science Foundation (grant 14-13-00158), and thank

E.P. Talsi et al. / Journal of Molecular Catalysis A: Chemical 421 (2016) 131–137
137
[19] A. Berkessel, T. Günther, Q. Wang, J.M. Neudörfl, Angew. Chem. Int. Ed. 52
[30] P. Cossee, J. Catal. 3 (1964) 80–88.
(2013) 8467–8471.
[31] M. Bochmann, Dalton Trans. (1996) 255–270.
[32] H.G. Alt, A. Köppl, Chem. Rev. 100 (2000) 1205–1221.
[33] S. Segal, I. Goldberg, M. Kol, Organometallics 24 (2005) 200–202.
[34] S.C. Gagieva, V.A. Tuskaev, D.A. Kurmaev, N.A. Kolosov, A.O. Borissova, B.M.
Bulychev, J. Organomet. Chem. 793 (2015) 155–159.
[35] A. Yeori, S. Gendler, S. Groysman, I. Goldberg, M. Kol, Inorg. Chem. Commun. 7
(2004) 280–282.
[36] E.L. Whitelaw, M.D. Jones, M.F. Mahon, Inorg. Chem. 49 (2010) 7176–7178.
[37] K. Press, A. Cohen, I. Goldberg, V. Venditto, M. Mazzeo, M. Kol, Angew. Chem.
Int. Ed. 50 (2011) 3529–3532.
[20] Q. Wang, J.M. Neudörfl, A. Berkessel, Chem. Eur. J. 21 (2015) 247–254.
[21] D. Xiong, M. Wua, S. Wang, F. Li, C. Xia, W. Sun, Tetrahedron: Asymmetry 21
(2010) 374–378.
[22] E.P. Talsi, D.G. Samsonenko, K.P. Bryliakov, Chem. Eur. J. 20 (2014)
14329–14335.
[23] J.L. Jat, S.R. De, G. Kumar, A.M. Adebesin, S.K. Gandham, J.R. Falck, Org. Lett. 17
(2015) 1058–1061.
[24] D. Xiong, M. Wu, S. Wang, F. Li, C. Xia, W. Sun, Tetrahedron: Asymmetry 21
(2010) 374–378.
[25] D. Xiong, X. Hu, S. Wang, C.X. Miao, C. Xia, W. Sun, Eur. J. Org. Chem. (2011)
4289–4292.
[38] S.H. Kim, D.J. Kim, M.J. Go, Y.S. Ko, J. Lee, Y. Kim, Dalton Trans. 41 (2012)
11619–11626.
[26] E.P. Talsi, T.V. Rybalova, K.P. Bryliakov, ACS Catal. 5 (2015) 4673–4679.
[27] For complex 2, both crystalline forms (bis-(-oxo and mono-(-oxo) are known,
exhibiting identical catalytic properties (Ref. 26).
[28] S. Kondo, K. Saruhashi, K. Seki, K. Matsubara, K. Miyaji, T. Kubo, K. Matsumoto,
T. Katsuki, Angew. Chem. Int. Ed. 47 (2008) 10195–10198.
[29] We intentionally introduced o-substutuents of comparable size (OCH3 and
CH2CH3): previous experience with titanium-salan complexes witnessed that
bigger (Ph) or smaller (Me, Cl) substituents afforded less enantioselective
catalysts, (Ref. 22).
[39] K. Huang, S. Zhou, D. Zhang, X. Gao, Q.R. Wang, Y.J. Lin, J. Organomet. Chem.
741 (2013) 83–90.
[40] Y. Wang, Y.S. Qin, X.H. Wang, F.S. Wang, Catal. Sci. Technol. 4 (2014)
3964–3972.
[41] S. Rodrígues-Escrich, L. Solà, C. Jimeno, C. Rodrígues-Escrich, M. Pericàs, Adv.
Synth. Catal. 350 (2008) 2250–2260.
[42] R. Li, D.J. Jansen, A. Datta, Org. Biomol. Chem. 7 (2009) 1921–1930.