Asymmetric transfer hydrogenation of imines catalyzed by a Noyori-type Ru(II) complex - A parametric study

Article abstract of DOI:10.1016/j.tetasy.2013.01.010

We present, to the best of our knowledge, the first parametric study of the asymmetric transfer hydrogenation of imines catalyzed by a Noyori-type catalytic complex based on ruthenium. A model imine for this study was 1-methyl-3,4-dihydroisoquinoline, and a well-known complex RuCl(η ⁶-p-cymene)((1S,2S)-N-p-toluenesulfonyl-1,2-diphenylethylenediamine) was chosen as the model catalyst. The reactions were performed in the presence of a formic acid-triethylamine mixture as the source of hydrogen. The parameters examined include general parameters, for example, concentration, temperature, and substrate-to-catalyst molar ratio, as well as parameters specific to this particular reaction, such as the amount of the hydrogenation mixture used, the ratio of its components, or the inhibitive effect of carbon dioxide. During this study, several unexpected parameters worth further investigation have emerged.

Full text of DOI:10.1016/j.tetasy.2013.01.010

Tetrahedron: Asymmetry 24 (2013) 233–239
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Asymmetric transfer hydrogenation of imines catalyzed by a Noyori-type
Ru(II) complex—a parametric study
a
a
a
a
a
a
a
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
Jan Pechácek , Jirí Václavík , Jan Prech , Petr Šot , Jakub Janušcák , Beáta Vilhanová , Jirí Vavrík ,
b
a,
⇑
ˇ
Marek Kuzma , Petr Kacer
a Department of Organic Technology, Institute of Chemical Technology, Technická 5, 166 28 Prague 6, Czech Republic
b
ˇ
Laboratory of Molecular Structure Characterization, Institute of Microbiology, v.v.i., Academy of Sciences of the Czech Republic, Vídenská 1083, 142 20 Prague 4, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 December 2012
Accepted 16 January 2013
We present, to the best of our knowledge, the first parametric study of the asymmetric transfer hydro-
genation of imines catalyzed by a Noyori-type catalytic complex based on ruthenium. A model imine
for this study was 1-methyl-3,4-dihydroisoquinoline, and a well-known complex RuCl(g
6-p-cyme-
ne)((1S,2S)-N-p-toluenesulfonyl-1,2-diphenylethylenediamine) was chosen as the model catalyst.
The reactions were performed in the presence of a formic acid–triethylamine mixture as the source of
hydrogen.
The parameters examined include general parameters, for example, concentration, temperature, and
substrate-to-catalyst molar ratio, as well as parameters specific to this particular reaction, such as the
amount of the hydrogenation mixture used, the ratio of its components, or the inhibitive effect of carbon
dioxide. During this study, several unexpected parameters worth further investigation have emerged.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
For imines, this mechanism does not plausibly explain some of
the observed phenomena, in particular the need for the imine to be
The increasing demand for enantiomerically pure compounds
arising mainly from the pharmaceutical industry provokes new
enantioselective synthetic methods to be sought, particularly
when the synthesis of a racemic mixture followed by chiral
separation is not economically feasible. The asymmetric transfer
hydrogenation (ATH) of C@N and C@O double bonds presents an
alternative to classical reduction procedures using gaseous
hydrogen allows the formation of enantiomerically enriched
alcohols and amines.1
Ruthenium-based complexes comprising of chiral diamine
ligands can effectively catalyze this reaction and achieve good
efficiency and excellent enantioselectivity. These Noyori-type
complexes work for both ketones and imines.2 For ketones, the
hydrogen donor can be either a mixture of formic acid and an or-
ganic base, or a mixture of a strong inorganic base and a primary
or a secondary alcohol.3 In the case of imines, on the other hand,
only a mixture of formic acid and a base is effective.4
either protonated or at least activated by a Lewis acid to undergo
the reduction,7 and also the formation of the opposite configura-
tion of the amine product. A mechanistic study of the ATH of imi-
nes was conducted by Martins et al. and a different, ‘ionic’,
mechanism was suggested.8 In this case, a protonated imine would
undergo the reduction, while only one hydrogen atom would be
transferred. The transition state is therefore acyclic.
The origin of the enantioselectivity is another important aspect
of the mechanism. For ketones, an explanation was proposed by
Yamakawa et al. A weak interaction between the CH group of the
Ru-g
6-coordinated aromatic ligand and the aromatic part of the
substrate was considered to stabilize the favorable transition state
leading to the preferred product. The disfavored transition state
would lack such a stabilizing interaction and would be less
preferred.9
A theoretical study concerning the ATH of imines was recently
carried out by Václavík et al., focusing on this CH–p interaction.
The mechanism of the reduction of ketones was proposed by
Noyori5 and supported by means of a theoretical study.6 In this
case, both hydrogen atoms are transferred simultaneously and
the transition state has the form of a six-membered cycle.
The reaction coordinates for several different interactions were cal-
culated and their stabilization effects were enumerated, support-
ing the hypothesis on the stabilizing effect of this interaction10
(Fig. 1).
In contrast to the efforts made to explain the mechanism of the
ATH, very few studies describing the influence of the reaction con-
ditions on the course of the reaction presently exist. Kinetic studies
of ATH of acetophenone in water on Ru-based complexes have
⇑
Corresponding author. Tel.: +420 220 444 158; fax: +420 220 444 340.
ˇ
E-mail address: petr.kacer@vscht.cz (P. Kacer).
0957-4166/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetasy.2013.01.010

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
234
H
N
Ru
O
-OOCH
H
H
O
N
S
N
H
O
N S
H+
Ru
O
Ph
Ph
H2N
Ph
N
1b
PhMe
Ph
2a
Figure 1. A transition state of hydrogen transfer during the ATH of 3,4-dihydro-
isoquinolines as investigated by Václavík et al. The NH+Á Á ÁO–S–O hydrogen bond
NH+
and CH–p interaction are depicted.
CO2
2b
been conducted both with common complexes11 and their ‘teth-
ered’ modifications.12
To the best of our knowledge, the only kinetic study of the ATH
of imines published so far was conducted on a Rh-based catalytic
complex analogous to Noyori’s catalyst.13
HCOO-
NH
*
3
The aim of this study was to provide an insight into the param-
eters influencing the asymmetric transfer hydrogenation of imines
catalyzed by Noyori’s Ru-based catalytic complexes. A well-known
Noyori-type complex RuCl(g
6-p-cymene)((1S,2S)-N-p-toluenesul-
+
fonyl-1,2-diphenylethylenediamine) 1a was chosen as the model
catalyst and a simple prochiral imine 1-methyl-3,4-dihydroiso-
quinoline 2a was chosen as the model substrate. The hydrogena-
tions were performed in acetonitrile in the presence of a formic
acid–triethylamine mixture.
Ru
O
H
H
O
N
N
S
The first step of the reaction is an in situ transformation of inac-
tive chloride 1a into the active hydride species 1b. Haack et al. de-
scribed the base-induced elimination of hydrogen chloride under
strongly basic conditions in alcohol (as in the ATH of ketones).2
The exact mechanism of this reaction under the aforementioned
conditions remains, as yet, unclear (Scheme 1).
Ph
Ph
1c
Scheme 2. Suggested pathway of 1b–catalyzed ATH of 2a affording 3.
In the case of imines, the protonation of substrate 2a occurs in
the acidic reaction mixture. Iminium salt 2b can then enter the
transition state in which the hydrogen transfer takes place. The
transition state is stabilized by a relatively strong ion-dipole inter-
action between the NH+ group of the protonated substrate and the
sulfonyl group of 1b.10 After the hydrogen transfer, when the posi-
tive charge on the NH group has disappeared, this interaction
weakens and 1-methyl-1,2,3,4-tetrahydroisoquinoline 3 can leave
the active site. Species 1b is regenerated from the coordinatively
unsaturated Ru complex 1c and a formate ion, giving carbon diox-
ide as a by-product (Scheme 2).
The composition of the reaction mixture denominates most
parameters necessary to describe the entire catalytic system. They
include the concentration of the reaction mixture, temperature,
substrate-to-catalyst (S/C) molar ratio, amount of hydrogenation
mixture, and the ratio of the components of the mixture. The re-
sults indicate that the course of the reaction is strongly dependent
on these parameters and that their careful choice can considerably
improve the activity and, to a certain extent, selectivity of the reac-
tion as well.
2. Results and discussion
The main goal of the work presented herein was to devise a set
of parameters describing as many aspects of the system as possi-
ble. Subsequent variations of these parameters (preferably one at
a time—a goal not always achievable) were then employed in order
to determine their influence on the course of the reaction.
2.1. Concentration of the reaction mixture
The principal parameter influencing every chemical reaction is
the concentration of the reaction mixture (or the partial pressure
of the reactants in the case of reactions in the gaseous phase).
Two experiments were performed in round-bottom flasks with
two different S/C molar ratios. Other important molar ratios were:
FA/TEA = 2.5, HM/S = 8.8, and the temperature was 30 °C (Table 1
and Fig. 2).
The reaction rates related to the amount of the catalyst were
slightly lower in the case of S/C = 200 and this difference increased
HCOOH, Et3N
CH3CN
H
N
Ru
Cl Ru
N
O
O
H
H
O
H
H
O
N
S
N
S
Ph
Ph
Ph
Ph
1b
1a
Scheme 1. Activation of catalyst 1a.

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
235
because more apparent differences in the reaction rates were
expected. Other important molar ratios were: FA/TEA = 2.5,
HM/S = 8.8, temperature 30 °C (Table 2 and Fig. 3).
Table 2
Initial reaction rate and selectivity with various S/C ratios
S/Ca Initial reaction rateb (mmol/min mmolcat
)
Conversionc (%) eec (%)
75
100
130
170
200
225
4.17
3.99
3.83
3.44
3.34
3.24
99.7
99.2
94.9
89.6
76.5
68.9
82
84
84
85
85
86
a
The S/C ratios are molar.
Initial reaction rates were calculated from the linear part of each conversion
Figure 2. Dependence of the initial reaction rate on the concentration of the
b
reaction mixture for each of the S/C ratios.
curve.
c
Conversion and ee are given after 50 min.
Table 1
Reaction rate and selectivity achieved with different concentrations of the reaction
mixtures
Concentrationa (%)
Initial reaction rateb
Conversionc (%)
eec (%)
(mmol/min mmolcat
)
S/C = 100
45
35
25
15
10
7
5.60
5.42
4.87
3.99
3.57
3.26
99.8
99.9
99.8
99.2
97.6
95.8
86
86
86
85
84
83
S/C = 200
45
35
25
15
4.95
4.63
4.49
3.74
3.38
3.06
98.4
99.4
95.6
85.3
80.4
76.1
86
86
86
85
85
86
10
7
Figure 3. Dependence of the initial reaction rate on the S/C ratio.
a
Concentration is defined by the following formula: concentration = (mass of
It was confirmed that varying the S/C ratio leads to different
reaction rates even when these are relative to the amount of the
catalyst. The difference supports the hypothesis of the ‘inactive
fraction’ of the catalyst (3.1).
2a + mass of 1a + mass of formic acid + mass of triethylamine)/mass of solvent.
b
Initial reaction rates were calculated from the linear part of each conversion
curve.
c
Conversion and ee are given after 50 min.
2.3. Temperature
with increasing concentration. A possible cause of this behavior is
that some constant amount of the catalyst is not active during the
reaction (not regenerated yet, blocked by the base or by the prod-
uct), and for the S/C = 200 reactions, a larger fraction of the total
amount of the catalyst is blocked than for the S/C = 100 reactions,
which results in a lower reaction rate related to the total amount of
the catalyst.
The influence of concentration was found to be significant. One
can obtain up to 80% increase in the reaction rate by using a more
concentrated reaction mixture. We can also see that with each of
the S/C ratios, the rate increase per unit of concentration increase
diminishes with increasing concentration. A possible explanation
is that at higher concentrations, the reaction rate is no longer lim-
ited by the frequency of effective collisions between hydride 1b
and the protonated molecules of substrate 2b, but by the total
amount of 1b present in the mixture.
Temperature is another basic parameter influencing chemical
reactions. For this reaction, the measurement of its effect on both
the reaction rate and enantioselectivity was conducted in the range
of 10–50 °C. The concentration was 7% in all cases and other molar
ratios were also constant: S/C = 100, FA/TEA = 2.5; HM/S = 8.8
(Table 3 and Fig. 4).
An increase in the reaction rate with increasing temperature
was expected and is rather common.
Table 3
Reaction rates and selectivity for the reactions performed at different temperatures
Temperature (°C)
Initial reaction ratea
Conversionb (%)
eeb (%)
(mmol/min mmolcat
)
10
20
30
40
50
0.37
1.54
3.14
5.27
7.64
18.2
71.4
95.5
95.9
96.8
89
87
85
84
82
2.2. S/C Ratio
The differences of the reaction rates related to the mass of the
catalyst with different S/C ratios were examined further, as the
S/C ratio is a parameter that generally affects the course of catalytic
reactions. For this experiment, a concentration of 15% was chosen
a
Initial reaction rates were calculated from the linear part of each conversion
curve.
b
Conversion and ee are given after 50 min.

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
236
Figure 4. Dependence of the initial reaction rate on temperature.
Figure 5. Dependence of the initial reaction rate on the amount of the hydroge-
nation mixture used.
The decrease in enantioselectivity can be explained by Yama-
kawa’s theory on the cause of the selectivity. Yamakawa stated
(and supported this statement by ab initio molecular modeling) that
two different transition states exist.9 One would lead to the pre-
ferred configuration of the product; the other would lead to the
other configuration. For catalyst 1b and substrate 3, this has been
further elaborated by Václavík et al., who calculated the actual en-
ergy difference between these transition states.10 This difference
would make the first transition state favorable and its formation
would be advantageous in comparison with the other, disfavorable
transition state.14 Obviously, molecules with higher energy (as in
the higher temperature) would overcome thisbarrier to a greater ex-
tent than molecules with lower energy and the non-preferred prod-
uct would become more abundant, as we can see from Table 3.
activity. The reaction would be then limited by the amount of the
active catalytic species. Some other experimental results supported
this hypothesis (vide infra), although no direct observation of the
protonated ligand was made.
The other hypothesis states that in a large excess of the hydro-
genation mixture, the protonated triethylamine is in a large excess
over the protonated substrate. The triethylammonium cation is be-
lieved to be able to create a hydrogen bond with the sulfonyl group
of the ligand, by means of which it sterically hinders the active site
of the catalytic complex and prevents the substrate from reaching
the active site. The reaction would then be limited by the amount
of free active sites.16 A theoretical calculation of such structure was
performed and the stabilization caused by this bond was calculated
to be
mol (Fig. 6).
D
E(DFT//MP2) = À89.4 kJ/mol and
D
G(DFT//DFT) = À31.0 kJ/
2.4. Amount of the hydrogenation mixture
The mixture of formic acid and triethylamine provides hydrogen
for the reduction. The most commonly used ratio of these two is 5/2
and this mixture exhibits azeotropism,15 therefore it is commonly
referred to as the azeotropic mixture. Variation of its amount was
expectedto have a major influenceonthe course of thereaction. Sev-
eral hydrogenation mixture/substrate ratios were tested. Concen-
tration of the reaction mixture was 7% in all cases, as well as
S/C = 100, FA/TEA = 2.5, and temperature 30 °C (Table 4 and Fig. 5).
Table 4
Reaction rate and selectivity with different amounts of hydrogenation mixture used
HM/Sa
Initial reaction rateb
Conversionc (%)
eec (%)
(mmol/min mmolcat
)
4.2
8.8
14.0
28.0
3.74
3.26
2.31
1.17
86.7
95.8
92.4
57.7
83
84
84
85
a
The hydrogenation mixture/substrate ratios are molar: (amount of formic
acid + amount of triethylamine)/amount of substrate.
b
Initial reaction rates were calculated from the linear part of each conversion
curve.
c
Conversion and ee are given after 50 min.
Contrary to our original expectations, an increase in the amount
of the azeotropic mixture leads to a decrease in the initial reaction
rate.
Figure 6. An associate of 1b and the triethylammonium ion. The hydrogen atoms of
the catalyst are omitted for clarity.
We have come up with two working hypotheses to explain such
behavior. One of them suggests that under strongly acidic condi-
tions (if the word ‘acidity’ can be used in an aprotic solvent), the
chiral diamine ligand becomes protonated and subsequently deco-
ordinates from the Ru atom, which would lead to a loss of catalytic
2.5. Ratio of formic acid and triethylamine
Variation of the ratio between formic acid and triethylamine,
the two components of the hydrogenation mixture, could provide
an insight into several subtle aspects of the reaction mechanism.

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
237
The following approach was used: the molar ratios of formic acid
2.6. Atmosphere
and triethylamine with respect to the substrate were taken from
the standard experiment described in the experimental section
(i.e., FA/S = 6.31, TEA/S = 2.52; entries 3 and 8). The ratio of one
of the components to the substrate was held constant, while the
ratio of the other one was varied and vice versa. The concentration
of the reaction mixture was held constant at all times (7%), S/C ra-
tio was 100 and temperature 30 °C. The HM/S ratio was inevitably
varied throughout this experiment (Table 5).
During the experiments, significant color differences between
the reaction mixtures were observed. Generally, the mixtures with
higher formic acid/triethylamine ratio were yellowish to clear,
whereas the more basic mixtures were orange, closely resembling
the color of the stock solution of the catalyst. This could indicate
some changes to the catalyst under acidic conditions.
ATH performed using a mixture of formic acid and triethyl-
amine as the source of hydrogen has one by-product: carbon diox-
ide. Since this compound is a gas under standard conditions, the
reaction is not strictly homogeneous and carbon dioxide can be as-
sumed to have an influence on the course of the reaction. Four
reactions employing the standard setting (S/C = 100, FA/TEA = 2.5,
HM/S = 8.8, concentration 7%, temperature 30 °C) were carried
out under different conditions: in an air atmosphere with magnetic
stirring, in an air atmosphere with sonication degassing, in an inert
argon atmosphere with stirring, and in CO2 under atmospheric
pressure (Table 6).
The results indicate that there is an optimal FA/TEA ratio for the
reaction, and that the ratio is close to the azeotropic ratio of 5:2.
With excessive formic acid, the reaction proceeded very slowly
or even not at all, probably because the catalyst was destroyed
by the formic acid. Protonation of the amino-group of the diamine
ligand followed by its decoordination from the Ru atom seems to
be a likely explanation of this phenomenon.17
With excess triethylamine, the reaction proceeded slowly as
well. One explanation suggests that triethylamine neutralizes all
of the formic acid prior to the reaction and then the substrate can-
not be protonated, which has been reported to be of vital impor-
tance.7 The aforementioned possibility of triethylamine’s binding
to the sulfonyl group through a hydrogen bond is another possible
explanation.
The difference between entries 4 and 7 can be explained by
these hypotheses because in the case of entry 7, the amount of
free formic acid with respect to the catalyst is roughly twice
the amount as in entry 4, and therefore the catalyst can be
decomposed to a greater extent. The amount of triethylamine
is also bigger in the case of entry 7, so more active sites can
be blocked by coordinated triethylamine. The major difference
between entries 2 and 9 can also be explained by these two
hypotheses.
Table 6
Reaction rate and selectivity under different gas treatment
Conditions
Initial reaction ratea Conversionb (%) eeb (%)
(mmol/min mmolcat
)
Air, magnetic stirring
3.41
3.40
3.42
1.08
96.2
96.7
99.6
71.5
83
83
84
83
Air, sonication degassing
1 atm Ar, magnetic stirring
1 atm CO2, magnetic stirring
a
Initial reaction rates were calculated from the linear part of each conversion
curve.
b
Conversion and ee are given after 50 min.
Carbon dioxide was confirmed to have a negative effect on the
reaction rate. This can be attributed to its reported ability to form
a complex with the catalyst, where it could prevent the active hy-
dride species from being formed and block the active site.18 These
negative effects, however, can be completely overcome by stirring
and no advanced techniques, such as sonication degassing or the
use of an inert atmosphere proved more advantageous.
3. Conclusion
To sum up, we present the first evaluation of the parametric
sensitivity of the asymmetric transfer hydrogenation of imines cat-
alyzed by a Noyori-type ruthenium catalytic complex 1a in the
presence of a mixture of formic acid and triethylamine as a hydro-
gen source. During this study, several mechanistic aspects worth
further investigation have emerged, most notably the low catalytic
activity in overly acidic conditions as well as the possible interac-
tion between 1a and the base present in the hydrogenation
mixture.
Low reaction rates in entries 1 and 10 can obviously be ex-
plained by the need of the substrate to be protonated. In these
mixtures, most formic acid was consumed for the protonation
of triethylamine, so the substrate was protonated to a small
extent.
Extremely low rates in entry 5 and the absence of any
product in entry 6 were probably caused by substantial dam-
age that the catalyst had sustained under the strongly acidic
conditions.
Table 5
Reaction rate and selectivity with various FA/TEA ratios
Entry
FA/Sa
TEA/Sa
FA/TEAa
Initial reaction rateb (mmol/min mmolcat
)
Conversionc (%)
eec (%)
1
2
3
4
5
6.31
6.31
6.31
6.31
6.31
12.60
6.31
2.52
1.26
0.63
0.5
1
2.5
5
0.10
0.12
3.16
0.49
0.02
4.6
8.4
95.8
26.6
0.6
80
67
85
87
81
10
6
7
8
9
25.20
12.60
6.31
2.52
1.26
2.52
2.52
2.52
2.52
2.52
10
5
2.5
1
—
—
1.5
95.8
20.2
1.4
—
0.04
3.16
0.48
0.03
83
85
81
73
10
0.5
a
b
c
The ratios are molar.
Initial reaction rates were calculated from the linear part of each conversion curve.
Conversion and ee are given after 50 min.

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
238
solution of sodium hydroxide, allowed to cool down slightly
4. Experimental
4.1. General
and extracted with toluene (5 Â 60 ml) and diethyl ether
(4 Â 30 ml). The extracts were combined and dried over anhy-
drous sodium sulfate. The dried extract was evaporated on a ro-
tary evaporator, affording a reddish-brown oily substance which
was then distilled in vacuo. The fraction of 95–97 °C (4 Torr)
contained the desired 2a (slightly yellowish oil). Yield: 12.5 g
(64%), purity: 99% (GC). 1H NMR (400.00 MHz, CDCl3, 303.2 K):
d 2.374 (3H, t, J = 1.5 Hz, 1-CH3), 2.689 (2H, m, H-4), 3.652
(2H, tq, J = 7.5, 1.5 Hz, H-3), 7.163 (1H, m, H-5), 7.275 (1H, m,
H-7), 7.323 (1H, ddd, J = 7.4, 7.4, 1.4 Hz, H-6), 7.462 (1H, dd,
J = 7.6, 1.4 Hz, H-8). 13C NMR (100.58 MHz, CDCl3, 303.2 K): d
23.17 (1-CH3), 25.96 (C-4), 46.85 (C-3), 125.19 (C-8), 126.78
(C-7), 127.32 (C-5), 129.49 (C-8a), 130.45 (C-6), 137.32 (C-4a),
164.14 (C-1). Mass calcd: 145.09; measured: 145.16.
The following chemicals were used: 2-phenethylamine (Fluka,
99%), acetylchloride (Sigma–Aldrich, 98%), triethylamine (Sigma–
Aldrich, 98%), phosphorus(V) oxychloride (Fluka, 98%), phospho-
rus(V) oxide (Lachema, 99%), xylene (Penta, 99%), hydrochloric acid
(Lachner, 36.4%), sodium hydroxide (Lachner, 99.6%), sodium sul-
fate (Lachner, 99.9%), diethyl ether (Sigma–Aldrich, 99%), toluene
(Penta, 99%), acetonitrile (LC–MS grade, Sigma–Aldrich), RuCl(p-
cymene)-(1S,2S)-N-(p-toluenesulfonyl)-1,2-diphenylethylenediam-
ine (Sigma–Aldrich), formic acid (Fluka, 98%), sodium carbonate
(Penta, 99%), magnesium sulfate (Lachner, 98%), and (À)-men-
thyl-chloroformate (Sigma–Aldrich, 99%). The synthesis of 2a is
based on a published procedure.19
4.3. General protocol for reactions carried out in round-bottom
flasks
4.2. Preparation of 2a
Phenethylamine 4 (18.0 g, 149 mmol) was dissolved in dichlo-
romethane (375 ml) and triethylamine (25.0 ml, 180 mmol) was
added. The flask was cooled on an ice bath to 0 °C and acetylchlo-
ride (12.7 ml, 178 mmol) was added dropwise (Scheme 3). After
A solution of substrate 2a in acetonitrile (LC–MS grade) was
prepared, so that its concentration was 150 mg/ml. A solution of
catalyst 1a was prepared by dissolving 5.4 mg of 1a in 1 ml aceto-
nitrile. A round-bottom flask was equipped with a magnetic stirrer
and a septum with a needle, and pre-heated on a water bath. The
initial volume of acetonitrile was transferred into the flask, the
hydrogenation mixture (HM) consisting of formic acid (FA) and tri-
ethylamine (TEA) was added, followed by the solution of the cata-
lyst. Active catalytic species 1b was allowed to form by stirring the
mixture for 5 min. After that, the solution of 2a (S) was added.
Standard reaction conditions: 0.11 mmol 2a; 2a/1a (S/C) ratio
100; HM/S ratio 8.83; FA/TEA ratio 2.5, total reaction mixture vol-
NH
NH2
CH3COCl
O
Et3N, DCM
5
4
Scheme 3. N-Acetylation of 4 affording 5.
ume 1500 ll, concentration 7% (see Section 2.1), temperature
30 °C. All ratios are molar.
the addition, the reaction mixture was gradually heated to 40 °C
for 15 min. After cooling down, water (165 ml) was added with
intensive stirring for 15 min. The organic phase was separated, ex-
tracted with 5% aqueous solution of sodium carbonate (2 Â 45 ml),
then with 5% aqueous solution of hydrochloric acid (2 Â 45 ml),
again with 5% aqueous solution of sodium carbonate (1 Â 45 ml)
and finally with water (60 ml). The organic phase was dried over
anhydrous magnesium sulfate and evaporated, affording 22.3 g
(92%) of 2-phenethyl-acetamide 5. Mass calcd: 163.10, measured:
163.16.
Samples were taken in the following way: the calculated amount
of the reaction mixture containing approximately 2 mg total of 2b
and 3 was transferred into a vial containing saturated aqueous solu-
tion of sodium carbonate (1 ml). The mixture was shaken well and
extracted with diethyl ether (3 Â 1 ml). The extract was dried over
anhydrous magnesium sulfate and stripped in a stream of argon
to dryness. The residue was dissolved in acetonitrile (700
analyzed on GC for conversion. After that, triethylamine (20
l) were added to the sample,
l
l) and
ll)
and (À)-menthyl-chloroformate 6 (10
l
affording a pair of diastereomeric carbamates 7a, 7b. The mixture
Compound 5 (21.8 g, 134 mmol) was mixed with phospho-
rus(V) oxychloride (47.0 ml, 502 mmol) and phosphorus(V) oxide
(23.8 g, 168 mmol). The resulting mixture was refluxed in dry
xylene (400 ml) for 4 h (Scheme 4). The cooled mixture was
was analyzed on GC for enantioselectivity (Scheme 5).
4.4. GC analyses
Varian CP-3800 gas chromatograph equipped with an 1177
injector, a Varian VF-1 column (length: 60 m, inner diameter:
0.25 mm, film thickness 0.25 lm, stationary phase: poly(dimethyl-
NH
POCl3, P4O10
siloxane)), and an FID 11 flame-ionization detector was used. For
the determination of conversion, the setting was as follows: injec-
N
O
xylenerf., 160°C
5
2a
Scheme 4. Bischler–Napieralski cyclization of 5 affording 2a.
tion volume 1 ll, injector temperature 250 °C, split ratio 25, col-
umn flow 0.5 ml/min, detector temperature 250 °C. Retention
times: 3 26.81 min, 2a 27.06 min.
For enantioselectivity, the following setting was used: injection
volume 1 ll, injector temperature 250 °C, split ratio 25, column
slowly hydrolyzed with warm water, until the addition of more
water did not heat the reaction mixture. The aqueous phase was
separated, concentrated HCl (25 ml) was added and the mixture
was extracted with toluene (3 Â 50 ml). The combined organic
extracts were added to the separated organic phase, which was
then extracted with 3.6% (w/w) HCl (1 Â 110 ml). The extract
was added to the previously separated water phase and the or-
ganic phase was discarded. The aqueous solution was cooled in
an ice bath, alkalized by an addition of 400 ml concentrated
flow 0.5 ml/min, detector temperature 250 °C. Retention times:
carbamate 7a derived from (R)-3 43.88 min, carbamate 7b derived
from (S)-3 44.31 min.
4.5. Theoretical computations
In performing the calculations of the catalyst-base associates
(see Section 2.4), a DFT level of theory utilizing the novel restricted

ˇ
J. Pechácek et al. / Tetrahedron: Asymmetry 24 (2013) 233–239
239
Cl
O
O
6
N
+
N
O
O
NH
*
Et3N
O
O
3
7a
7b
Scheme 5. Formation of two diastereomeric carbamates during pre-column derivatization.
x
B97XD functional20 was used in combination with the Def2-SVP
References
basis set21 for all atoms and additional effective core potential for
the ruthenium atom.22 The IEFPCM model was used to simulate
solvation in acetonitrile. Frequency analyses were performed on
optimized structures to obtain thermodynamic data, and so were
the MP2-level calculations providing more accurate single-point
energies.
1. Blaser, H.-U.; Malan, C.; Pugin, B.; Spindler, F.; Steiner, H.; Studer, M.
ChemInform 2003, 34.
2. Haack, K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R. Angew. Chem., Int. Ed.
1997, 36, 285–288.
3. Ikariya, T.; Blacker, A. J. Acc. Chem. Res. 2007, 40, 1300–1308.
4. Uematsu, N.; Fujii, A.; Hashiguchi, S.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
1996, 118, 4916–4917.
5. Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001, 66, 7931–7944.
6. Yamakawa, M.; Ito, H.; Noyori, R. J. Am. Chem. Soc. 2000, 122, 1466–1478.
7. Åberg, J. B.; Samec, J. S. M.; Backvall, J.-E. Chem. Commun. 2006, 2771–2773.
8. Martins, J. E. D.; Clarkson, G. J.; Wills, M. Org. Lett. 2009, 11, 847–850.
9. Yamakawa, M.; Yamada, I.; Noyori, R. Angew. Chem., Int. Ed. 2001, 40, 2818–
2821.
4.6. NMR spectroscopy
NMR spectra were acquired using a Bruker Avance III 400 MHz
spectrometer (1H 400.00 MHz and 13C 100.58 MHz). 1H and 13C
NMR spectra were measured in chloroform, whose residual signals
(dH 7.265 ppm, dC 77.00 ppm) were used as internal standards for
the chemical shift scale. Standard software was used, as supplied
by the manufacturers (Bruker BioSpin GmbH, Rheinstetten,
Germany).
ˇ
ˇ
10. Václavík, J.; Kuzma, M.; Prech, J.; Kacer, P. Organometallics 2011, 30, 4822–
4829.
11. Wu, X.; Liu, J.; Di Tommaso, D.; Iggo, J. A.; Catlow, C. R. A.; Bacsa, J.; Xiao, J.
Chem. Eur. J. 2008, 14, 7699–7715.
12. Cheung, F. K.; Clarke, A. J.; Clarkson, G. J.; Fox, D. J.; Graham, M. A.; Lin, C.;
Criville, A. L.; Wills, M. Dalton Trans. 2010, 39, 1395–1402.
13. Blackmond, D. G.; Ropic, M.; Stefinovic, M. Org. Process Res. Dev. 2006, 10, 457–
463.
14. Balcells, D.; Maseras, F. New J. Chem. 2007, 31, 333–343.
15. Narita, K.; Sekiya, M. Chem. Pharm. Bull. 1977, 25, 135–140.
Acknowledgments
ˇ
ˇ
ˇ
´
ˇ
16. Kuzma, M.; Václavík, J.; Novák, P.; Prech, J.; Janušcák, J.; Cerveny, J.; Pechácek,
ˇ
J.; Šot, P.; Vilhanová, B.; Matoušek, V.; Goncharova, I. I.; Urbanová, M.; Kacer, P.
This work has been financially supported by the Grant Agency
of the Czech Republic (Grant GACR 104/09/1497 and P106/12/
1276), and by a Grant for long-term conceptual development of
the Institute of Microbiology RVO: 61388971.
The access to computing and storage facilities owned by parties
and projects contributing to the National Grid Infrastructure Meta-
Centrum, provided under the program ‘Projects of Large Infrastruc-
ture for Research, Development, and Innovations’ (LM2010005) is
appreciated.
Dalton Trans., 2013, http://dx.doi.org/10.1039/C3DT32733G.
17. Wu, X.; Li, X.; King, F.; Xiao, J. Angew. Chem., Int. Ed. 2005, 44, 3407–3411.
18. Koike, T.; Ikariya, T. Adv. Synth. Catal. 2004, 346, 37–41.
19. Sánchez-Sancho, F.; Mann, E.; Herradón, B. Synlett 2000, 509–513.
20. Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10, 6615–6620.
21. Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297–3305.
22. Andrae, D.; Häußermann, U.; Dolg, M.; Stoll, H.; Preuß, H. Theor. Chim. Acta
1990, 77, 123–141.