The effective direct resolution procedure for the chiral drug bevantolol hydrochloride

Article abstract of DOI:10.1016/j.tetasy.2016.03.012

The solubility of the chiral drug bevantolol hydrochloride 1 in water and the azeotropic mixture ethanol-water has been investigated. It was found that rac-1 meets the requirements of Meyerhoffer's rule, so it was possible to reduce the ternary diagram, describing the solubility of 1, to a pseudo binary form, which facilitates the analysis of crystallization processes caused by temperature changes. On this basis, the effective and robust resolution procedure of racemic bevantolol hydrochloride by a preferential crystallization approach has been realized successfully.

Full text of DOI:10.1016/j.tetasy.2016.03.012

Tetrahedron: Asymmetry xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
The effective direct resolution procedure for the chiral drug
bevantolol hydrochloride
⇑
Zemfira A. Bredikhina, Olga A. Antonovich, Dmitry V. Zakharychev, Alexander A. Bredikhin
A.E. Arbuzov Institute of Organic and Physical Chemistry of Kazan Scientific Center of Russian Academy of Sciences, Arbuzov St., 8, Kazan 420088, Russian Federation
a r t i c l e i n f o
a b s t r a c t
Article history:
The solubility of the chiral drug bevantolol hydrochloride 1 in water and the azeotropic mixture ethanol–
water has been investigated. It was found that rac-1 meets the requirements of Meyerhoffer’s rule, so it
was possible to reduce the ternary diagram, describing the solubility of 1, to a pseudo binary form, which
facilitates the analysis of crystallization processes caused by temperature changes. On this basis, the
effective and robust resolution procedure of racemic bevantolol hydrochloride by a preferential crystal-
lization approach has been realized successfully.
Received 11 March 2016
Accepted 23 March 2016
Available online xxxx
Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction
Currently, direct methods for racemate resolution are deeply
understood from a fundamental and engineering point of view.
The scale of‘ fundamental problems associated with the phe-
nomenon of spontaneous resolution of enantiomers during crystal-
lization is recognized by the scientific community as ‘one of the
true challenges for science in the 21st century’.1 No less remark-
able is the practical aspect of this phenomenon. Spontaneous res-
olution makes it possible to obtain pure enantiomers, the
importance of which is increasing day by day, by direct methods,
which do not require the use of enantiopure auxiliary substances
and/or materials.2,3
The oldest of the direct methods is the preferential crystalliza-
tion. Described in 1866,4 it started to become more common in
chemical practice in the second half of the twentieth century5,2
and currently has a solid theoretical background and a variety of
embodiments.6 Another modification of the direct resolution, rela-
tively recently taken its shape, represents the so-called attrition-
enhanced deracemization,7 aka ‘Viedma ripening’.8 This approach
is based on the different rates of the dissolution of crystals of
different sizes (Ostwald ripening9), and can be applied to the der-
acemization of only those substrates that combine the ability for
spontaneous resolution with the ability to take part in
(spontaneous or induced) racemization in solution. The obvious
limitations of this approach are compensated, firstly, through the
ability to transform all of the available racemate into a single
enantiomer. Secondly, this modification allows the resolution to
proceed as a continuous process, as opposed to batch-crystalliza-
tion for the majority of preferential crystallization variants.
We believe that their penetration into routine practice would be
facilitated through the widening of the essential non-racemic com-
pounds prone to spontaneous resolution in racemic form. It
appears that the registered chiral active pharmaceutical ingredi-
ents could be the objects of choice in this regard.
The literature has many examples of how direct methods are
used for the deracemization of substances, which are used as syn-
thetic precursors for single enantiomeric active pharmaceutical
ingredients. For example, the resolution of racemic threo-1-(4-
nitrophenyl)-2-aminopropane-1,3-diol, the (R,R)-enantiomer of
which is used in the production of chloramphenicol.10,11 Racemic
modafinic acid, a key intermediate in the synthesis of (R)-modafinil,
was resolved in a preparative scale using a preferential crystalliza-
tion approach.12 The direct resolution of rac-2-(benzylide-
neamino)-2-(2-chlorophenyl)acetamide underlies the synthesis of
(S)-clopidogrel.13,14 Among recent publications we refer to our
synthesis of mexiletine enantiomers based on the preferential
crystallization of rac-(2,6-dimethylphenoxy)propane-1,2-diol.15
Another frequently used option that takes advantage of the
enantioselective crystallization in the synthesis of nonracemic
active pharmaceutical ingredients, is the conversion of a racemic
target product, incapable of spontaneous resolution, in a simple
derivative, which crystallizes in the form of normal conglomerate.
By this strategy, among the hundreds of fenfluramine and norfenflu-
ramine racemic salts, which have been studied, eight salts crystal-
lized in the form of a conglomerate were found, and six of those
were resolved into enantiomers via preferential crystallization.16
Pure enantiomers of atenolol were obtained via preferential
crystallization in the form of salts with p-toluic acid.17 Racemic
⇑
Corresponding author. Tel.: +7 843 2727393; fax: +7 843 2731872.
E-mail address: baa@iopc.ru (A.A. Bredikhin).
http://dx.doi.org/10.1016/j.tetasy.2016.03.012
0957-4166/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

2
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
propranolol hydrochloride crystallizes to form a normal racemic
compound,18 whereas its hydrofluoride becomes a conglomerate
and can be resolved into its enantiomers via resolution by entrain-
ment.19 Medetomidine is used in medical practice as a hydrochlo-
ride, but becomes a conglomerate in the form of oxalic acid salt.20
The registered racemic forms of omeprazole (free acid and its mag-
nesium salt) are not prone to spontaneous resolution. At the same
time the omeprazole potassium salt solvates with two molecules of
ethanol and can be separated into individual enantiomers by direct
methods in a batch crystallization mode,21 and in a continuous
mode.22
The most spectacular from the standpoint of demonstrating the
feasibility of the spontaneous resolution are the direct resolution of
chiral racemic active pharmaceutical ingredients. Apparently,
methyldopa should be considered as the first example of this kind,
because the pure (S)-enantiomer of it has been produced industri-
ally via the direct resolution of the racemic product.11 Spontaneous
resolution is a property exhibited by albuterol sulfate23 and cal-
cium pantothenate.24 In our previous publications we have shown
that the ability for spontaneous resolution is inherent in active
pharmaceutical ingredients guaifenesin,25,26 mephenesin27 and
methocarbamol.28
Potentially, this short list can be expanded by including the
active pharmaceutical ingredient bevantolol, the hydrochloride of
which is used in medical practice, and is crystallized in the form
of a normal conglomerate.29,30 Herein we examine the possibility
of using a direct resolution approach for to obtain the pure enan-
tiomers of bevantolol.
Bevantolol, chemical name 1-[[2-(3,4-dimethoxyphenyl)ethyl]
amino]-3-(3-methylphenoxy)-2-propanol, is a popular cardiovas-
cular remedy. It is used in the treatment in the form of hydrochlo-
ride 1 (see Scheme 1) under the trade names Ranestol, Sentiloc,
Vantol.31
2. Results and discussion
For the synthesis of racemic bevantolol hydrochloride, the reac-
tion of 3,4-dimethoxyphenylethylamine with racemic 1-chloro-3-
meta-tolyloxy-2-propanol or 2-meta-tolyloxymethyloxirane rac-2
has been used.38,30 Racemic oxirane rac-2 was obtained from race-
mic epichlorohydrin rac-3 and meta-cresol. Enantiomeric (R)-2 and
(S)-2 were prepared analogously from (S)-3 and (R)-3, respectively.
For the synthesis of enantiomeric bevantolol samples, needed as
seeds in the initial resolution step, we reacted 3,4-
dimethoxyphenylethylamine with enantiopure oxiranes (S)-2 or
(R)-2. The general sequence of the reaction is illustrated in
Scheme 2 with the (S)-enantiomer of bevantolol.
CH3
CH3
O
H
CH3
OH
OH
NH
i
iii
ii
O
H
(S)-1
O
H
+
Cl
O
(R)-3
(S)-2
O
CH3
O
CH3
Scheme 2. Reagents and conditions: (i) 2 M aq NaOH, 60 °C, 2 h; (ii) 3,4-
dimethoxyphenylethylamine, EtOH, rt; (iii) gaseous HCl, EtOAc.
As noted above, the tendency for the crystallization of racemic 1
as a racemic conglomerate has already been proved.30 However,
this very feature is a necessary but not sufficient condition for
the successful implementation of the preparative methods of direct
resolution. It is known that about half of conglomerate formative
compounds demonstrate poor efficiency in the stereoselective
crystallization processes.39–41 In addition, for efficient separation
some enantiomeric pre-enrichment of the initial mixture is often
required.25 Currently, there are no tests except for an experiment
to assess the ability of a substance to undergo preparative sponta-
neous resolution.
OH
H
H3C
O
N
O
O
CH3
CH3
.HCl
Bevantolol hydrochloride 1
Scheme 1. Chemical structure of bevantolol hydrochloride 1.
Bevantolol is a b1-adrenoceptor antagonist that has been shown
to be as effective as other beta blockers for the treatment of angina
pectoris and hypertension. Experiments confirm both agonist and
The choice of solvent is of great importance in the development
of methods of direct resolution. It should not have a too high or not
too low dissolving power with respect to the resolvable substrate.
antagonist effects on a-receptors, in addition to antagonist activity
at b1-receptors.32,33 For the family of b-adrenergic blockers with
general formula ArOCH2CH(OH)CH2NHAlk, it has been shown that
the (S)-enantiomers are eutomer components of the racemic drug,
whereas the (R)-enantiomers (distomers) usually display other
(often undesirable) activities.34 Racemic bevantolol as well as its
enantiomers also possesses different biological activity. Thus (S)-
(À)-bevantolol is a stereoselective b1L-AR2S antagonist; this sub-
stance, unlike its (+)-enantiomer, attenuates the increase in heart
rate caused by ( )-CGP 12177,35 whereas rac-1 only tended to
decrease the positive chronotropic effect of ( )-CGP 12177 and is
not considered to be an antagonist of b1L-adrenoceptors.36
The main advantages of homochiral (single enantiomer, enan-
tiopure) substances include reduced dosage and reducing the side
effects associated with off-target biological activity of the distomer
(undesired enantiomer). The latter property makes it possible to
create new single enantiomer drugs with different indications on
the basis of a racemate. Due to a general trend towards replacing
the racemic drugs with their enantiopure analogues,37 it is useful
to have a convenient approach to the individual enantiomers of
bevantolol.
An important characteristic of the solvent is the gap
D
T = TH À TN,
where the separating TH, the temperature at which the original
undersaturated solution becomes saturated, and TN, temperature
at which spontaneous crystallization becomes noticeable without
external seeds.42 Finally, the solvent of choice should be environ-
mentally friendly, accessible and inexpensive as possible. It is
known that compound 1 is very soluble in methanol, chloroform
and DMSO, but in the light of these requirements, none of these
solvents look attractive. It is also known that compound 1 is sol-
uble in water, but the quantitative data available differ by orders
of magnitude and cannot be regarded as reliable. For this reason,
we have carried out preliminary measurements of the solubility
of the racemic bevantolol hydrochloride in several available sol-
vents. The data are presented in Figure 1.
Figure 1 shows that ethyl acetate is not suitable practically for
crystallization, since even at 73 °C the concentration of the satu-
rated solution is only Csat = 1.4 mg/ml. Using isopropanol
(Csat = 6.4 mg/ml; 45 °C) or acetonitrile (Csat = 8.3 mg/ml; 41 °C)
requires working with dilute solutions. Ethanol (Csat = 60.0 mg/
ml; 43 °C) represents a suitable solvent for the resolution of
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
3
of rac-1 in a cycle of ‘dissolution-crystallization’ is characterized
by a magenta color. Concentration values for the racemate are
given in terms of the individual stereoisomer (i.e., the real values
of the concentration of the racemate are halved).
Analysis of the charts suggests that for both studied systems,
the experimental points, which correspond to dissolution process,
both for enantiopure and for racemate samples fall well on a single
curve (solid lines). This means that these systems well meet the
requirements of Meyerhoffer’s rule (the racemate solubility equals
twice the enantiomer solubility),43 i.e., their behavior is consistent
with the behavior of a normal conglomerate and not complicated
by concentration and solvation effects.
The experimental points that correspond to the crystallization
(contour circles in olive and magenta for enantiopure and racemic
samples, respectively) show a tendency to decrease the interval
DT
in passing from nonracemic to racemic samples. However, taking
into account the stochastic nature of the lower boundary of the
metastable zone, a considerable variance of the data set appears
natural, and the bottom boundaries of metastable supersaturated
solutions are satisfactorily described by the averaged dashed
curves. All of this gives us hope that when describing the dissolu-
tion and crystallization processes in the studied systems, the
mutual influence of bevantolol enantiomers in solution can be
neglected. In turn, this allows the use of the pseudo binary phase
diagram ‘enantiomer–solvent’ to describe the phase behavior of
the systems under consideration and when choosing a stereoselec-
tive crystallization conditions.
Figures 3 and 4 show the pseudo binary phase diagrams ‘bevan-
tolol hydrochloride–ethanol’ and ‘bevantolol hydrochloride–wa-
ter’. Figures (a) represent in diagram form their general
theoretical view. Figures (b) show the experimentally investigated
areas; these fragments are similar to Figure 2a and b and retain the
notation adopted there.
Figures 3c and 4c are of particular interest to the task solution of
rac-1 direct resolution into individual enantiomers. They represent
a detailed view of the region of the phase diagram near the temper-
ature interval 20–40 °C, i.e., near the ambient temperature. Using
moderate temperatures during the resolution simplifies a system
thermostating and supersaturated solution sampling for process
control, as well as allows minimizing the problems associated with
solvent evaporation.
The graphs indicate that in this area both the absolute value of
the solubility of compound 1 and the metastable zone width in
both solvents are satisfactory for the stereoselective crystallization
of bevantolol enantiomers. Preliminary experiments demonstrated
that when using ethanol, the crystallization rate decreased rapidly
with decreasing solution supersaturation, so the process was
Figure 1. Solubility of rac-1 in various solvents; filled circles correspond to TH
values and contour circles correspond to TN values.
rac-1, however the value of
is approximately 10–11 °C. The solubility of rac-1 in water is some-
what lesser (Csat = 47.7 mg/ml; 44.5 °C), but the value T is 20 °C.
D
T = TH À TN is sufficiently small and it
D
Both of these solvents have been tested by us in the process of res-
olution of rac-1 by entrainment. Since absolute ethanol is hygro-
scopic, we had to deal with an azeotrope mixture ethanol–water;
the measured solubility characteristics for this solvent consisted
of Csat = 60.7 mg/ml at 35 °C;
D
T ꢀ 10 °C.
In deciding on the stereoselective crystallization process
parameters, it is desirable to know the solubility phase diagram
for the system ‘solvent–enantiomers’, built with due regard for
the metastability zone of the supersaturated solution of the crys-
talline phase in the solvent.6,42 It should be noted that the con-
struction of a total phase diagram of the ternary system (two
enantiomers and a solvent) requires a large amount of experimen-
tal data. Furthermore, a ternary solubility diagram is usually repre-
sented as an isothermal cut, and it is inconvenient to represent the
crystallization process caused by changing (lowering in our case)
the temperature. As shown below, in some cases the information
contained in the ternary phase diagram can be adequately repre-
sented in the binary diagram, which takes account of the
temperature.
Figure 2 shows the temperature dependence of the solubility of
individual bevantolol hydrochloride stereoisomers in ethanol and
water. Olive solid circles denote the experimental results in terms
of dissolution, and olive contour circles denote the starting points
of crystallization for enantiopure samples. Similarly, the behavior
Figure 2. Temperature dependence of the solubility (solid lines) and the reverse process of crystallization (dotted lines) of rac-1 (magenta) and (S)-1 (olive) in rectified
ethanol (a) and in water (b).
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

4
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
Figure 3. (a) General view of the solubility phase diagram for the system ‘ethanol–bevantolol hydrochloride’; the dotted rectangle designated area which is enlarged in chart
b. (b) Experimentally studied region of the phase diagram; designations are the same as those adopted in Figure 2. The dotted box denotes the region, represented enlarged in
figure c. (c) Trajectory of the cyclic process of preferential crystallization of the individual enantiomers on the pseudo binary phase diagram (explanations are in the text).
Figure 4. Solubility phase diagram for the system ‘water–bevantolol hydrochloride’. Designations are the same as in Figure 3.
preferably carried out near the bottom temperature boundary of
the metastable zone. In aqueous solution, the crystallization pro-
ceeds efficiently to almost complete exhaustion of supersaturation
of the crystallizing component. At the same time, the bottom
boundary of the metastable zone for this system shows a high
degree of stochasticity, indicating that the process should be con-
ducted near the upper boundary of the metastable zone where
the system has much higher stability. The following process condi-
tions were chosen based on these considerations: crystallization
temperature—no more than 28 °C for ethanol and no more than
36 °C for water; the initial concentration of the racemate was
ꢁ60 mg/ml in ethanol and ꢁ40 mg/ml in water.
The crystallization process starts from a point in the KS zone,
which corresponds to the solution of the racemic composition at
a temperature above the boundary of the saturated solution
(solid black line in the phase diagram). Cooling this solution to a
temperature just below this boundary (ꢁ27 °C) with seeding with
the (S)-enantiomer gives a precipitate which consists substantially
of this enantiomer. At that its concentration in solution decreases
(heavy blue line KSLS). At the same time, the second enantiomer
remains in a supersaturated solution, and its concentration does
not significantly change (thin red line KRLR). After finishing this
step, the precipitated (S)-enantiomer is separated from solution.
The solution is then heated above the saturation temperature,
and the replenishing portion of the racemate is introduced, result-
ing in an increase in the concentration of each enantiomer. This
stage of the cycle in the phase diagram is designated by dashed
lines LSMS and LRMR. Upon cooling to supersaturation followed
by seeding with the (R)-enantiomer, this namely isomer
precipitates (heavy red line MRNR). The concentration of the
(S)-enantiomer remains fairly constant (thin red line MSNS). After
heating and adding an appropriate amount of the racemate, the
cycle is closed, and the system returns to virtually the initial com-
position (dotted lines NSKS and NRKR).
In Figures 3c and 4c, the concentrations of each component
before and after a regular resolution step (nodal points) are desig-
nated by triangles in red for (R)-1, and by inverted blue triangles
for (S)-1. Similar points for different cycles are grouped and
enclosed by ovals of an appropriate color; the ovals are designated
by capital letters. Red and blue arrows (trajectories) schematically
represent the change in concentration of corresponding enan-
tiomer during the realization of resolution by entrainment. Let us
discuss the details of the process by the example of the direct res-
olution of compound 1 in ethanol, Figure 3c.
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
5
Table 1
Resolution by entrainment of rac-bevantolol hydrochloride in ethanol (96%) (100 ml, 10 mg of crystal seeds on every run, crystallization temperature 27.5 0.5 °C)
Run
Added amount of rac-1, g
Operation amount
of enantiomers, g
Resolution time, min
(R)-1 and (S)-1 obtained
eea, % YEb
Yield, g
DRc, %
(R)-1
(S)-1
g
%
1
2
3
4
5
6
6.070
0.740
1.160
1.150
1.040
1.040
3.035
3.386
2.837
3.366
2.877
3.378
3.035
2.684
3.233
2.704
3.193
2.692
120
130
140
140
150
160
(S) 0.750
(R) 1.170
(S) 1.160
(R) 1.050
(S) 1.050
(R) 1.010
94.9
94.7
92.1
94.2
96.4
92.9
0.702
1.098
1.059
0.980
1.002
0.928
23.1
32.4
32.8
29.1
31.4
27.5
54.6
85.5
82.4
76.3
78.0
72.2
a
b
c
ee: enantiomeric excess (HPLC).
YE: Yield enantiomer; YE(g) = [Yield(g) Â ee (%)]/100 À 0.010; YE (%) = [YE(g) Â 100]/operation amount of (R)- or (S)-1(g).
DR: degree of resolution; DR (%) = YE(g) Á 100/[0.5 Á (6.07–3.5)]; 3.5—solubility rac-1 in 100 ml EtOH (96%) at 27.5 °C.
Table 1 shows the quantitative results of the stereoselective
at which the process was interrupted and the precipitate was fil-
tered off. The missing amount of racemic 1 was then added to
the mother liquor and the process was repeated (see Section 4).
In total, for three cycles (6 runs), from 11.20 g of rac-1, 2.763 g
(24.7%) of (S)-1 and 3.006 g (26.8%) of (R)-1 were obtained. Fig-
ure 4c shows the possible trajectories of the resolution by entrain-
ment process at the pseudo binary phase diagram of rac-1 in water.
Table 2 shows the quantitative results of this experiment. In total,
0.625 g (20.3%) of (S)-enantiomer and 0.599 g (19.4%) of (R)-enan-
tiomer was obtained from 3.076 g from rac-1 in three full cycles. It
should be noted that although the yield of the pure enantiomers is
somewhat higher in ethanol, in view of ecological requirements,
the separation in water as a solvent is preferred.
crystallization of bevantolol hydrochloride in ethanol under the
conditions described above; technical details of the process are
given in the Section 4.
From Table
1 it is clear that the process of bevantolol
hydrochloride separation is carried out with good enantiomeric
purity of the crystalline precipitate (ee 92–96%) and high yields
of pure enantiomers (23% in the first stage, then yield increases
substantially). The degree of separation at the first step exceeds
50% (54.6% in fact), thereby creating a high enantiomeric enrich-
ment of the mother liquor after the first crystallization cycle. At
other stages, the degree of separation increases to 72–85% and
the enantiomeric excess increases to ꢁ32%.
In Figure 5, the bevantolol hydrochloride separation process (3
cycles, 6 runs), is illustrated by the values of optical rotation of the
mother liquor. The red dots indicate the values of optical rotation
3. Conclusions
In investigating the solubility of the active chiral pharmaceuti-
cal ingredient bevantolol hydrochloride 1 in various solvents, it can
be seen that when rac-1 is dissolved in ethanol or water, it meets
the requirements of Meyerhoffer’s rule. This property makes it pos-
sible to reduce the ternary phase diagram, which describes the sol-
ubility of chiral compound 1, to pseudo binary phase diagram,
which facilitates a qualitative and quantitative analysis of crystal-
lization processes caused by temperature changes. The process of
the resolution of rac-1 is highly reproducible, does not require
enantiomeric pre-enrichment of the racemate in the first cycle,
and allows us to obtain crystalline precipitates with high enan-
tiomeric purity in each run. Furthermore this method can be
‘green’ by using water as the solvent.
4. Experimental
4.1. General
Figure 5. Mother liquor optical rotation vs time of preferential crystallization of
bevantolol hydrochloride (3 cycles, 6 runs). Red dots—the value of optical rotation,
on reaching which the process was interrupted; time spent on technical operations
associated with the interruption and resumption of stereoselective crystallization
process was not reflected in the chart.
The NMR spectra were recorded on a Bruker Avance-500
spectrometer (500.13 MHz for 1H) in CDCl3 with the signals of
Table 2
Resolution by entrainment of rac-bevantolol hydrochloride in water (50 ml, 10 mg of crystal seeds on every run, crystallization temperature 35.5 0.5 °C)
Run
Added amount of rac-1, g
Operation amount
of enantiomers, g
Resolution time, min
(R)-1 and (S)-1 obtained
eea, % YEb
Yield, g
DRc, %
(R)-1
(S)-1
g
%
1
2
3
4
5
6
2.300
0.222
0.242
0.266
0.245
0.269
1.150
1.048
1.155
1.033
1.143
1.023
1.150
1.252
1.145
1.267
1.157
1.267
60
80
90
80
90
80
(R) 0.232
(S) 0.252
(R) 0.276
(S) 0.255
(R) 0.279
(S) 0.417
93.6
90.1
92.6
91.4
89.7
70.4
0.207
0.217
0.246
0.223
0.241
0.284
18.0
17.3
21.3
17.6
21.1
22.4
69
72
82
74
80
95
a
b
c
ee: enantiomeric excess (HPLC).
YE: Yield enantiomer; YE(g) = [Yield(g) Â ee (%)]/100 À 0.010; YE(%) = [YE(g) Â 100]/Operation amount of (R)- or (S)-1(g).
DR: degree of resolution; DR (%) = YE(g) Á 100/[0.5 Á (2.3–1.7)]; 1.7—solubility rac-1 in 50 ml water at 35.5 °C.
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

6
Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
the solvent as the internal standard. Optical rotations were mea-
sured on a Perkin–Elmer model 341 polarimeter (concentration c
is given as g/100 ml). Melting points for general purposes were
determined using a Boëtius apparatus and are uncorrected. Optical
rotations were measured on a Perkin–Elmer model 341 polarime-
ter (concentration c is given as g/100 ml). The enantiomeric excess
(ee) of the substances and intermediate samples used was checked
by HPLC; HPLC analyses were performed on a Shimadzu LC-20AD
system controller.
CH2N+CH2CH2), 3.84 (s, 3H, OCH3), 3.85 (s, 3H, OCH3), 3.97 (dd,
J = 9.7, 5.9 Hz; 1H, 1OCH2), 4.06 (dd, J = 9.7, 4.3 Hz; 1H, 1OCH2),
4.69 (m, 1H, OCH), 5.26 (broad s, OH), 6.63–6.67 (m, 2H, Ar),
6.76–6.81 (m, 4H, Ar), 7.12 (t, J = 7.8 Hz, 1H, Ar), 8.90 (broad s,
N+H), 9.80 (broad s, 1H, N+H).
The samples of enantiopure bevantolol hydrochloride, which
were used as seeds, were obtained according to the literature:30
(R)-1: mp 157–158 °C; [a]
20 = +19.3 (c 1.1, EtOH); 99.8% ee [chiral
D
HPLC analysis; Daicel Chiralcel OD-RH (0.46 Â 25 cm) column; col-
umn temperature 24 °C; eluent KPF6 aq (0.05 M pH 2.1): CH3-
CN = 60:40 v/v; flow rate: 0.7 ml minÀ1; UV detector 254 nm;
4.2. Compounds
tR = 8.8 min]. (S)-1: mp 157–158 °C; [
a
]
20 = À19.2 (c 1.1, EtOH);
D
Racemic epichlorohydrin rac-3, meta-cresol and 3,4-di-
methoxyphenylethylamine were commercially available. Enan-
tiopure epichlorohydrins (R)-3 and (S)-3 was prepared via the
Jacobsen kinetic hydrolytic resolution of rac-3. (R)-3: bp 113–
99.9% ee (chiral HPLC analysis; tR = 13.3 min). NMR spectra were
identical with those published earlier.30
4.3. Solubility measurement
114 °C; [a]
20 = –33.1 (c 4, MeOH), 96% ee [chiral HPLC analysis;
D
Chiralpack AD (0.46 Â 25 cm) column; column temperature
20 °C; eluent: hexane/isopropanol = 9:1; flow rate: 1.0 ml/min;
refractive index detector; tR = 7.5 min]. (S)-3: [
In order to determine the solubility, the polythermal laser mon-
itoring last crystal disappearance method was used. Precisely con-
trolled samples [Shimadzu AUW 120D analytic balance
(accuracy 0.01 mg)] of rac-1 and (S)-1 in 1 ml of solvent were
placed in sealed glass vessels, equipped with magnetic stir bar.
For more precise temperature control in the vicinity of the test
samples a vessel equipped with a controlling thermometer, a stir
bar and a solvent were placed. All the vessels were placed in a ther-
mostatic container. The suspension of the samples was agitated at
about 400 rpm, increasing the temperature gradually (0.5 °C/min).
The turbidity degree was registered for each individual reactor to
detect the so-called ‘clear point’. The temperature where the last
crystals disappeared was taken as saturation temperature. After
complete dissolution of the crystals in all the studied reactors
the coolant was slowly cooled down (ꢁ0.3 °C/min), detecting the
so-called ‘cloud points’ (the temperature where the first crystals
were reliably detected) in cooling cycles.
a]
20 = +33.3 (c 4,
D
MeOH), 96% ee (chiral HPLC analysis; tR = 9.7 min).
4.2.1. 1,2-Epoxy-3-(3-methylphenoxy)propane rac-2
To a solution of rac-3 (51.36 g, 0.555 mol) and m-cresol (20.09 g,
0.185 mol), aqueous 2 M NaOH solution (130 ml) was added drop-
wise at 60 °C, after which the reaction mixture was stirred for 2 h
at 60 °C. Then the reaction mixture was extracted with ether, and
the combined organic portions were dried (MgSO4) and concen-
trated. The residue was purified by distillation to yield rac-2
(21.43 g, 71%) as a colorless oil, bp 100–105 °C at 0.1 mm Hg. 1H
NMR d: 2.36 (s, 3H, CH3), 2.78 (dd, J = 5.0, 2.7 Hz, 1H, CH2), 2.92
(t, J = 4.5 Hz, 1H, CH2), 3.36–3.38 (m, 1H, CH), 3.99 (dd, J = 11.0,
5.6 Hz, 1H, OCH2), 4.22 (dd, J = 11.0, 3.3 Hz, 1H, OCH2), 6.75–6.78
(m, 2H, C2,4ArH), 6.82 (d, J = 7.5 Hz, 1H, C6ArH), 7.25 (t, J = 7.8 Hz,
1H, C5ArH).
4.4. Preferential crystallization
4.2.1.1. (S)-1,2-Epoxy-3-(3-methylphenoxy)propane (S)-2.
This
compound was obtained from (R)-epichlorohydrin (R)-3 (8.32 g,
0.09 mol) and m-cresol (3.89 g, 0.036 mol) as described for racemic
compound. The configuration of the product is inverted as against
the configuration of the starting epichlorohydrin; colorless oil
Experiments on the stereoselective crystallization of rac-1 were
performed in a three-necked flask equipped with a magnetic stir-
rer, water condenser and thermometer placed in a water bath at
a controlled temperature. During the resolution small portions of
the mother liquor were taken through a syringe equipped with
Teflon filter and were analyzed either by polarimetry (the case of
EtOH as resolution solvent; cell path length 100 mm; wavelength
365 nm; 20 °C), or by HPLC (the case of water).
(3.90 g, 66%), [a]D
20 = +13.1 (c 1.7, MeOH). 1H NMR d: 2.36 (s, 3H,
CH3), 2.78 (dd, J = 4.9, 2.7 Hz, 1H, CH2), 2.92 (dd, J = 4.9, 4.3 Hz,
1H, CH2), 3.35–3.37 (m, 1H, CH), 3.99 (dd, J = 11.0, 5.5 Hz, 1H,
OCH2), 4.21 (dd, J = 11.0, 3.3 Hz, 1H, OCH2), 6.74–6.78 (m, 2H,
C2,4ArH), 6.82 (dd, J = 7.4, 0.7 Hz, 1H, C6ArH), 7.14 (t, J = 7.8 Hz, 1H,
C5ArH).
Example of the resolution by entrainment of racemic bevantolol
hydrochloride in ethanol without the initial enantiomeric enrich-
ment of a starting mixture: 6.07 g of (RS)-1 was dissolved in
100 ml of EtOH (96%) at 38–40 °C and was cooled with stirring to
28 °C. Then a portion of finely ground crystalline seeds of (S)-
bevantolol hydrochloride (10 mg, 0.17% by weight) was added.
The solution was stirred and incubated at a temperature of
27.5 0.5 °C for 120 min. The precipitate was filtered, collected
and dried to obtain 750 mg of (S)-bevantolol (ee 94.9%). To the fil-
trate, which remained after separation of the precipitate, 740 mg of
(RS)-bevantolol were added, and the system was heated to 38–
40 °C until full homogenization after which it was cooled to
28 °C, and seeded with finely ground crystals of (R)-bevantolol
hydrochloride (10 mg). The solution was stirred at a temperature
of 27.5 0.5 °C, and filtered after 130 min; 1170 mg of (R)-bevan-
tolol, ee 94.7%, were obtained. Likewise, the cycle was repeated
several times, each time adding the lacking amount of racemic
bevantolol hydrochloride. Details of subsequent runs No. 3–6 are
shown in Table 1. A single recrystallization of the combined crystal
crop of each of the enantiomers from acetonitrile enhances the
enantiomeric excess to 99%.
4.2.1.2. (R)-1,2-Epoxy-3-(2-methylyphenoxy)propane (R)-2.
This
compound was synthesized analogously from (S)-3; colorless oil;
[
a D
]
20 = À12.9 (c 1.0, MeOH).
4.2.2. rac-1-[[2-(3,4-Dimethoxyphenyl)ethyl]amino]-3-(3-meth-
ylphenoxy)-2-propanol hydrochloride, rac-Bevantolol
hydrochloride, rac-1
This compound was prepared by analogy with a previously
described procedure.38,30 rac-1,2-Epoxy-3-(3-methylphenoxy)-
propane rac-2 (21.0 g, 0.128 mol) and 23.2 g (0.128 mol) of 3,4-
dimethoxyphenethylamine in 50 ml of ethanol were stirred at
25–30 °C for 24 h. The reaction was monitored by TLC. After the
disappearance of the starting epoxide, the mixture was evaporated
to dryness and the residue was dissolved in EtOAc and gaseous HCl
was passed through the resulting solution to give 37.5 g (75%) of
crude rac-1ÁHCl. After recrystallization from CH3CN/EtOH (9:1)
was obtained 26.4 g (54%) of rac-1ÁHCl: white solid, mp
140–141 °C. 1H NMR d: 2.29 (s, 3H, CH3), 3.21–3.37 (m, 6H,
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012

Z. A. Bredikhina et al. / Tetrahedron: Asymmetry xxx (2016) xxx–xxx
7
18. Bredikhin, A. A.; Savel’ev, D. V.; Bredikhina, Z. A.; Gubaidullin, A. T.; Litvinov, I.
A. Rus. Chem. Bull. 2003, 52, 853–861.
19. Bredikhin, A. A.; Bredikhina, Z. A.; Gubaidullin, A. T.; Krivolapov, D. B.; Litvinov,
I. A. Mendeleev Commun. 2004, 14, 268–270.
Experiment on rac-1 separation in water was performed simi-
larly without enantiomeric enrichment. Experimental details are
given in Table 2.
20. Choobdari, E.; Fakhraian, H.; Peyrovi, M. H. Chirality 2014, 26, 183–188.
21. Coquerel, G.; Petit, M. N.; Tauvel, G. Pat. US 2009/0124811 A1, 2011.
22. Hein, J. E.; Cao, B. H.; Van der Meijden, M. W.; Leeman, M.; Kellogg, R. M. Org.
Process Res. Dev. 2013, 17, 946–950.
Acknowledgments
23. Palacios, S. M.; Palacio, M. A. Tetrahedron: Asymmetry 2007, 18, 1170–1175.
24. Synoradzki, L.; Hajimowicz, H.; Wisialski, J.; Mizerski, A.; Rowicki, T. Org.
Process Res. Dev. 2008, 12, 1238–1244.
25. Bredikhina, Z. A.; Novikova, V. G.; Zakharychev, D. V.; Bredikhin, A. A.
Tetrahedron: Asymmetry 2006, 17, 3015–3020.
26. Fayzullin, R. R.; Lorenz, H.; Bredikhina, Z. A.; Bredikhin, A. A.; Seidel-
Morgenstern, A. J. Pharm. Sci. 2014, 103, 3176–3182.
27. Bredikhin, A. A.; Bredikhina, Z. A.; Novikova, V. G.; Pashagin, A. V.;
Zakharychev, D. V.; Gubaidullin, A. T. Chirality 2008, 20, 1092–1103.
28. Bredikhin, A. A.; Bredikhina, Z. A.; Zakharychev, D. V.; Pashagin, A. V.
Tetrahedron: Asymmetry 2007, 18, 1239–1244.
29. Neau, S. H.; Shinwari, M. K.; Hellmuth, E. W. Int. J. Pharm. 1993, 99, 303–310.
30. Bredikhina, Z. A.; Zakharychev, D. V.; Gubaidullin, A. T.; Bredikhin, A. A. J. Mol.
Struct. 2009, 936, 171–176.
31. The Merck Index; O’Neil, M. J., Ed., 14th ed.; Merck and Co., Inc.: Whitehouse
Station, NJ, USA, 2006; p 195. entry 1192.
The authors thank the Russian Fund of Basic Research and the
Government of the Republic of Tatarstan for financial support
(Grant No. 15-43-02238). We also wish to thank Dr. A.V. Pashagin
for valuable assistance with chiral HPLC analysis.
References
1. Perez-Garcia, L.; Amabilino, D. B. Chem. Soc. Rev. 2007, 36, 941–967.
2. Collet, A.; Brienne, M. J.; Jacques, J. Chem. Rev. 1980, 80, 215–230.
3. Lorenz, H.; Seidel-Morgenstern, A. Angew. Chem., Int. Ed. 2014, 53, 1218–1250.
4. Gernez, M. Compt. Rend. Acad. Sci. 1866, 63, 1192.
5. Secor, R. M. Chem. Rev. 1963, 63, 297–309.
6. Coquerel, G. Top. Curr. Chem. 2007, 269, 1–51.
7. Hein, J. E.; Cao, B. H.; Viedma, C.; Kellogg, R. M.; Blackmond, D. G. J. Am. Chem.
Soc. 2012, 134, 12629–12636.
8. Sogutoglu, L.-C.; Steendam, R. R. E.; Meekes, H.; Vlieg, E.; Rutjes, F. P. J. T. Chem.
Soc. Rev. 2015, 44, 6723–6732.
9. Noorduin, W. L.; Vlieg, E.; Kellogg, R. M.; Kaptein, B. Angew. Chem., Int. Ed. 2009,
48, 9600–9606.
10. Amiard, G. Experientia 1959, 15, 38–40.
32. Vaughan, W. E. M. J. Clin. Pharmacol. 1987, 27, 450–460.
33. Frishman, W. H.; Goldberg, R. J.; Benfield, P. Drugs 1988, 35, 1–21.
34. Mehvar, R.; Brocks, D. R. J. Pharm. Pharm. Sci. 2001, 4, 185–200.
_
35. Schlicker, E.; Pe˛dzin´ ska-Betiuk, A.; Kozłowska, H.; Szkaradek, N.; Zelaszczyk,
´
D.; Baranowska-Kuczko, M.; Kiec-Kononowicz, K.; Marona, H.; Malinowska, B.
Naunyn-Schmiedeberg’s Arch. Pharmacol. 2014, 387, 743–752.
11. Crosby, J. Tetrahedron 1991, 47, 4789–4846.
´
36. Malinowska, B.; Kiec-Kononowicz, K.; Flau, K.; Godlewski, G.; Kozłowska, H.;
12. Wermester, N.; Lambert, O.; Coquerel, G. CrystEngComm 2008, 10, 724–733.
13. Van der Meijden, M. W.; Leeman, M.; Gelens, E.; Noorduin, W. L.; Meekes, H.;
Van Enckevort, W. J. P.; Kaptein, B.; Vlieg, E.; Kellogg, R. M. Org. Process Res. Dev.
2009, 13, 1195–1198.
Kathmann, M.; Schlicker, E. Br. J. Pharmacol. 2003, 139, 1548–1554.
37. Murakami, H. Top. Curr. Chem. 2007, 269, 273–299.
38. Hoefle, M. L.; Hastings, S. G.; Meyer, R. F.; Corey, R. M.; Holmes, A.; Stratton, C.
D. J. Med. Chem. 1975, 18, 148–152.
14. Noorduin, W. L.; Van der Asdonk, P.; Bode, A. A. C.; Meekes, H.; Van Enckevort,
W. J. P.; Vlieg, E.; Kaptein, B.; Van der Meijden, M. W.; Kellogg, R. M.; Deroover,
G. Org. Process Res. Dev. 2010, 14, 908–911.
15. Bredikhina, Z. A.; Kurenkov, A. V.; Krivolapov, D. B.; Bredikhin, A. A.
Tetrahedron: Asymmetry 2015, 26, 577–583.
39. Dufour, F.; Perez, G.; Coquerel, G. Bull. Chem. Soc. Jpn. 2004, 77, 79–86.
40. Levilain, G.; Coquerel, G. CrystEngComm 2010, 12, 1983–1992.
41. Gonella, S.; Mahieux, J.; Sanselme, M.; Coquerel, G. Org. Process Res. Dev. 2011,
16, 286–293.
42. Coquerel, G. Chem. Soc. Rev. 2014, 43, 2286–2300.
16. Coquerel, G.; Bouaziz, R.; Brienne, M. J. Chem. Lett. 1988, 1081–1084.
17. Kitaori, K.; Takehira, Y.; Furukawa, Y.; Yoshimoto, H.; Otera, J. Chem. Pharm.
Bull. 1998, 46, 505–507.
43. Meyerhoffer, W. Ber. Dtsch. Chem. Ges. 1904, 37, 2604–2610.
Please cite this article in press as: Bredikhina, Z. A.; et al. Tetrahedron: Asymmetry (2016), http://dx.doi.org/10.1016/j.tetasy.2016.03.012