Improved process for the preparation of montelukast: Development of an efficient synthesis, identification of critical impurities and degradants

Article abstract of DOI:10.1021/op900311z

An improved and scalable process for the production of montelukast (Singulair, drug for asthma) based on a new and advantageous method of carrying out the key substitution reaction has been developed. The present procedure is distinguished from the previous solutions in the use of linear or cyclic polyethers, which ensures higher selectivity of the key step. The improved process for the preparation of montelukast is able to minimize a content of impurities and allows the effective production of montelukast and its scale-up.

Full text of DOI:10.1021/op900311z

Organic Process Research & Development 2010, 14, 425–431
Improved Process for the Preparation of Montelukast: Development of an Efficient
Synthesis, Identification of Critical Impurities and Degradants
Alesˇ Halama, Josef Jirman,* Olga Bousˇkova´, Petr Gibala, and Kamal Jarrah
ZentiVa k.s., Department of Chemical Synthesis, U kabeloVny 130, Prague 102 01, Czech Republic
Abstract:
between carbon and sulfur atoms (thiolation step). In principle,
there are two basic methods of carrying out the key step, which
are described in Scheme 1. The first method is indicated as
method A,1,5–13 the second one as method B.14,15 Both the basic
methods can be further extended with a number of variants that
are mainly based on alternating the order of the reaction steps.
Only a few synthetic procedures were carried out in a different
way than is described in Scheme 1.16-18 The first and conven-
tional process for production of montelukast exploits com-
mercially available alcohol 2 and carboxylic acid 3 as starting
materials.6 This method is characterized by the use of a
methanesulfonyl derivate of 2 and the dilithium salt of 3 in the
key synthetic step. This solution also comprises a method of
purification of crude montelukast via its salt with dicyclohexy-
lamine and also a method of obtaining the amorphous sodium
salt of montelukast 1.
There are a number of functional groups in the montelukast
structure that impair the chemical stability of this substance,
see Figure 1. Montelukast 1 is known to be prone to several
types of degradation.19-26 It is mainly the case of the following
chemical transformation: oxidation of the mercapto group to
the sulfoxide,20–23 photoisomerisation at the location of the
double bond from geometry (E) to (Z),23,24 dehydration at the
tert-alcohol group, producing the corresponding olefin25 and
dehalogenation reaction.26 The chemical impurities are usually
removed by means of crystallization in the phase of montelukast
salts with amines6,9,25 or in the phase of montelukast acid.27
An improved and scalable process for the production of mon-
telukast (Singulair, drug for asthma) based on a new and
advantageous method of carrying out the key substitution reaction
has been developed. The present procedure is distinguished from
the previous solutions in the use of linear or cyclic polyethers,
which ensures higher selectivity of the key step. The improved
process for the preparation of montelukast is able to minimize a
content of impurities and allows the effective production of
montelukast and its scale-up.
Introduction
Montelukast 1 is a well-known drug indicated for the
prophylaxis and chronic treatment of asthma.1-4 It acts as a
selective antagonist of the leukotriene D4 receptor which leads
to the reduction of bronchoconstriction and results in less
inflammation. Montelukast is administered orally once daily
which yields a benefit in comparison with a majority of drugs
for pulmonary disorders such as asthma.1–4 The sodium salt of
montelukast described with formula 1, see Figure 1, is used
for asthma therapy.
Several synthetic methods were reported for the preparation
of montelukast 1.1,5-15 From the point of view of chemical
synthesis the crucial step is formation of the chemical bond
* Author for correspondence. E-mail: jirman.josef@seznam.cz.
(1) Labelle, M.; Belley, M.; Gareau, Y.; Gauthier, J. Y.; Guay, D.; Gordon,
R.; Grossman, S. G.; Jones, T. R.; Leblanc, Y. Bioorg. Med. Chem.
Lett. 1995, 5 (3), 283.
(16) Kidemet, D.; Kroselj, V.; Osolnik, R. WO 2008049922 A2 (Krka,
2008).
(17) Perez, A. J. A.; Huguet, C. J.; Dalmases, B. P. US 2008194825 A1
(Inke S.A., 2008).
(2) Schoors, D. F.; De Smet, M.; Reiss, T.; Margolskee, D.; Cheng, H.;
Larson, P.; Amin, R.; Somers, G. Br. J. Clin. Pharmacol. 1995, 40,
277.
(3) Markham, A.; Faulds, D. Drugs 1998, 56 (2), 251.
(4) Noonan, M. J.; Chervinsky, P.; Brandon, M.; Zhang, J.; Kundu, S.;
Mcburney, J.; Reiss, T. F. Eur. Respir. J. 1998, (11), 1232.
(5) Belley, M. L.; Leger, S.; Roy, P.; Xiang, Y. B.; Labelle, M.; Guay,
D. EP 0480717 B1 (Merck Frost, 1992).
(18) Coppi, L.; Bartra, S. M.; Gasanz, G. Y.; Monsalvatje, L. M.; Talavera,
E. P. WO 2007051828 A2 (Esteve Qu´ımica, 2007).
(19) Radhakrishna, T.; Narasaraju, A.; Ramakrishna, M.; Satyanarayana,
A. J. Pharm. Biomed. Anal. 2003, 31, 359.
(20) Dufresne, C.; Gallant, M.; Gareau, Y.; Ruel, R.; Trimble, L.; Labelle,
M. J. Org. Chem. 1996, 61, 8518.
(6) Bhupathy, M.; McNamara, J. M.; Sidler, D. R.; Volante, R. P.; Bergan,
J. J. EP 0737186 B1 (Merck & Co, 1995).
(7) Sundaram, V.; Rajan, S. T.; Bulusu, V. V. N. C. S.; Srivastav, A.;
Kasturi, R. K.; Aavula S. K. US 20050234241 A1 (Dr. Reddy’s Lab.,
2005).
(8) Shapiro, E.; Yahalomi, R.; Niddam-Hildesheim, V.; Sterimbaum, G.;
Chen, K. WO 2005105751 A1 (TEVA Pharm. , 2005).
(9) Reguri, B. R.; Bollikonda, S.; Chandra, S. B. V. V.; Kasturi, R. K.;
Aavula, S. K. US 20050107612 A1 (Reddys Lab., 2005).
(10) Hung, J. T.; Wei, C. P. DE 102007061630 A1 (Formosa Lab, 2009).
(11) Agarwal, R. M.; Bhirud, S. B.; Pillai, B. G.; Khude, G. D. WO
2009138993 A1 (Macleods Pharm. , 2009).
(21) Nelson, E. D.; Harmon, P. A.; Szymanik, R. C.; Teresk, M. G.; Li,
L.; Seburg, R. A.; Reed, R. A. J. Pharm. Sci. 2006, 95, 1527.
(22) Balani, S. K.; Xu, X.; Pratha, V.; Koss, M. A.; Amin, R. D.; Dufresne,
C.; Miller, R. R.; Arison, B. H.; Doss, G. A.; Chiba, M. Drug Metab.
Dispos. 1997, 25 (11), 1282.
(23) Al Omari, M. M.; Zoubi, R. M.; Hasan, E. I.; Khader, T. Z.; Badwan,
A. A. J. Pharm. Biomed. Anal. 2007, 45, 465.
(24) Smith, G. A.; Rawls, C. M.; Kunka, R. L. Pharm. Res. 2004, 21 (9),
1539.
(25) Sterimbaum, G.; Shapiro, E.; Chen, K. WO 2007005965 A1 (TEVA
Pharm., 2007).
(12) Pathi, S. L.; Kankan, R. N.; Rao, D. R.; Phull, M. S. WO 2008032099
A2 (Cipla Ltd., 2008).
(13) Halama, A.; Jirman, J. WO 2008083635 A1 (Zentiva, 2008).
(14) Overeem, A.; Gieling, R. G.; Zhu, J.; Thijs, L. WO 2005105749 A2
(Synthon BV, 2005).
(15) Overeem, A.; Gieling, R. G.; Zhu, J.; Thijs, L. WO 2005105750 A1
(Synthon BV, 2005).
(26) Saravanan, M.; Kumari, K. S.; Reddy, P. P.; Naidu, M. N.; Babu,
J. M.; Srivastava, A. K.; Kumar, T. L.; Sekhar, B. V. V. N. C.;
Satyanarayana, B. J. Pharm. Biomed. Anal. 2008, 48, 708.
(27) (a) Suri, S.; Singh, J. WO 2004108679 A1 (Morephen Lab., 2004).
(b) Overeem, A.; van Den Heuvel, D. J. M. WO 2005040123 (Synthton
BV, 2005). (c) Alnabari, M.; Sery, Y. WO 2005074893 A1 (Chemagis,
2005).
10.1021/op900311z  2010 American Chemical Society
Published on Web 02/11/2010
Vol. 14, No. 2, 2010 / Organic Process Research & Development
•
425

Figure 1. Structure of montelukast sodium, 1; arrows indicates critical parts of the structure for chemical degradations; compounds
2 and 3 represent common intermediates.
Scheme 1. Two general methods of the key step for preparation of montelukast 1
The complicated synthesis and low stability of montelukast
1 present many problems for its large-scale production. We
developed a new process that obviates many previous problems
and limitations. Furthermore, we report critical impurities and
a procedure for achieving the pharmaceutical quality. The
improved process, whose details are presented in this paper, is
suitable for large-scale production of the active pharmaceutical
ingredient (API) with many advantages.
compound 4 was achieved using alcohol 2 by treating with
methanesulfonyl chloride under basic catalysis. Another starting
material of the present process is [1-(mercaptomethyl)cyclo-
propyl]acetic acid 3, which is converted to the corresponding
salt by the action of a base (2 equiv). Various substances can
be used as the base, e.g. organometallic compounds such as
butyl lithium.6 Suitable bases to ensure acceptable solubility
are alkoxides of alkali metals, e.g. tert-butoxides and tert-
amylates. The in situ obtained salt of acid 3 is the active form
of the reagent. In the literature was described a process for
preparation of alkali metal salts of 3 by saponification of
corresponding esters without isolation and characterization of
these salts.28 Several of our attempts to isolate the salts 3 with
alkali metals were carried out, but they failed due to very low
stability of these salts to oxygen. The properties of these salts
differ; mainly their solubility depends on the kind of metal ion.
The corresponding decrease of the active agent concentration
reduces selectivity of the substitution reaction, see Figure 2.
Low conversion of 4 to montelukast 1 in the course of the key
step relates to both competitive reactions of starting material 4
and consecutive degradations of montelukast 1, see Scheme 2.
Results and Discussion
A procedure of synthesis as well as isolation and purification
of montelukast 1 has a significant influence on the economy of
process. Our need for large quantities of montelukast led us to
develop a scalable synthesis for this compound in pharmaceuti-
cal quality. Hence, we tried the earlier synthesis, and we
distinguished three serious troubles: low selectivity of the key
synthetic step, a suitable form of montelukast for efficient
isolation and specific impurities formation.
Key Synthetic Step and Its Selectivity. The present process
for synthesis of montelukast 1 relates to a new method of
carrying out the substitution reaction that represents the key
step in the process. This reaction is described in the Scheme 2.
The compound 4 containing the methanesulfonic group as the
good leaving group was used as the first starting material. The
(28) Chawla, H. P. S.; Chowdhary, A. S.; Patel, A. M.; Patel, M. P. WO
2007116240 A1 (Glade Organics, 2007).
426
•
Vol. 14, No. 2, 2010 / Organic Process Research & Development

Scheme 2. Present process for the key step (inside the ellipse); unwanted reactions lead to impurities (out of the ellipse)
Unlike the previous solutions, the present process uses the
linear or cyclic polyethers as agents to increase reaction
selectivity. These compounds play the role of phase transfer
catalyst, which can solvate metal ions and thus increase the
solubility and reactivity of the nucleophilic reagent. The
increased reactivity of the nucleophilic reagent (dialkali metal
salt of 3) resulted in higher selectivity of the process, and
unwanted competitive reactions were suppressed. As suitable
linear polyethers various polyethylene glycols (PEG) can be
used. CROWN ethers with different cycle sizes can serve as
suitable cyclic polyethers. When the key step was carried out
without addition of any polyethers, compound 5 was found as
a main product (typical final content of reaction mixture by
HPLC: 53.9% of 5, 20.6% of 1, 8.1% of 6, 4.6% of 2, other
minor impurities making up the rest of 100%).
The set up of rough conditions for the key step procedure
was made by means of a series of preliminary experiments.
Their results and experimental conditions are described in Table
1. The experiment based on the use sodium tert-butoxide and
PEG-600 was chosen as a ground for the scale-up procedure.
For instance, the final montelukast content in the reaction
mixture is usually about only 20% when the mixture of toluene
and tetrahydrofuran is used as a solvent and sodium tert-
Table 1. Conversions of 4 and contents of montelukast 1
after 24 h from mixing of the reaction components in
different modifications of the key synthetic stepa
component
increasing
selectivity
(1 equiv)
conversion
of starting content of
material montelukast
base
(2 equiv)
4 (%)
1 (%)
tert-BuOLi
tert-BuOLi
tert-BuONa
s
97.9
97.3
82.3
96.1
94.6
93.2
81.4
94.8
96.2
89.5
95.0
92.7
60.8
59.6
21.4
75.9
73.3
81.0
68.2
65.6
72.7
12.0
9.6
12-CROWN-4
s
tert-BuONa 18-CROWN-6
tert-BuONa 15-CROWN-5
tert-BuONa PEG-600
tert-BuONa PEG-1500
tert-BuOK
tert-BuOK
tert-BuOK
tert-AmylOK
dibenzo-18-CROWN-6
PEG-600
s
s
tert-AmylOK 18-CROWN-6
51.5
a Twenty milliliters of toluene, 0.28 g of acid 3, a base (2 equiv to acid) and a
polyether (1 equiv) were mixed under argon atmosphere and cooled to ∼-10 °C.
The solution of 4 (1 g) in 5 mL of THF was then added to obtain the slurry, and
the mixture was gradually allowed to warm from -10 °C up to room temperature
in 1 h and stirred at room temperature for several hours. Conversion of 4 and
montelukast content was checked by HPLC (isocratic mode).
Figure 2. Time dependence of the contents of montelukast in
the reaction mixture of substitution reactions carried out with
various tert-butoxides of alkali metals and without addition of
agents increasing selectivity.
Vol. 14, No. 2, 2010 / Organic Process Research & Development
•
427

advantages without the risk of excessive losses of the product
that would otherwise remain stuck on the production equip-
ment.29 Pure amine salts were used for direct transformation to
the pharmaceutically useful amorphous sodium salt of 1. The
setup of conditions for the isolation and purification procedure
was made by means of a series of preliminary experiments with
many alkylamines, see Table 2. Better results were observed
for primary amines in comparison with secondary as well as
tertiary amines. The results for n-propylamine and isopropy-
lamine obtained at the stage of preliminary experiments were
almost the same. However, repeated experiments with isopro-
pylamine on real reaction mixtures show slightly better repro-
ducibility, yields and product quality (isolated yields from
reaction mixture were 70-75%, and purity was around 94%
by HPLC) in comparison with n-propylamine (isolated yields
65-70%, purity around 93% by HPLC). The experiment based
on the use of isopropylamine was chosen as a ground for the
scale-up procedure.
Critical Impurities and Degradants. Montelukast 1 and
its intermediates show extremely low chemical stability which
has great influence on its industrial production.19–26 One of the
sources of chemical contamination of montelukast 1 is low
stability of the normally used starting material 4. In parallel to
the key thiolation reaction, compound 4 is subjected to undesired
reactions, namely cyclization and elimination that lead to
compounds 5 and 6,30 see Scheme 2. Impurities of this type
are not critical and can be easily removed by crystallization of
montelukast salts with amines, especially from nonpolar sol-
vents, e.g., toluene.
Figure 3. Dependencies of montelukast content on reaction time
for the key substitution carried out under various conditions;
sodium tert-butoxide was used as a base and the mixture of
toluene and tetrahydrofuran as a solvent in all cases
butoxide as a base. The addition of an agent to increase
selectivity, for example PEG-600 (1 equiv) or 18-CROWN-6
(1 equiv), to this reaction mixture in the beginning increases
the final montelukast content to 80%, and the final content of
competitive products is reduced. Figure 3 compares the
composition of reaction mixtures with or without the use of a
polyether and demonstrates the positive influence of the agent
added.
Isolation and Purification Process. The output of the key
synthetic step is a reaction mixture that contains a salt of
montelukast 1 with an alkali metal and several undesired
impurities dissolved together in organic solvents used. It was
impossible to isolate and purify sodium montelukast directly
due to its good solubility in numerous solvents from polar ones
(e.g., water, ethanol) to nonpolar ones (e.g., diethyl ether,
toluene). The product was isolated from the reaction mixture
by means of conversion of the crude substance to well-
crystallizing salts with amines, followed by recrystallization of
these salts with simultaneous removal of chemical impurities.
A very important aspect in the method of isolation of mon-
telukast ammonium salts is the use of acetonitrile as a cosolvent.
Acetonitrile specifically prevents the adhesion of separated
crystals to the crystallization vessel or to the agitator. This
finding can be applied in production scale with substantial
The impurities generated by degradation of the target
substance are structurally very similar to the source substance,
and therefore, it is very difficult to reduce their content in the
API by common methods (e.g., crystallization). It has been
found that the transformation of montelukast 1 to (Z)-isomer 8
was unexpectedly fast under sunlight and air. Already within
minutes the concentration of (Z)-isomer 8 may grow to a level
of percent units.31 Simultaneously, but more slowly, the
mercapto group was oxidized, producing (E)-montelukast
sulfoxide 7, see Scheme 2.20–24,31
A mixture of diastereoisomers 9 and 10 has been found as
a specific and so far unknown impurity in crude montelukast
sodium 1. These compounds are generated through the reaction
of alkali metal salts of 3 with primarily obtained montelukast
Table 2. Preparation and yields of salts of various amines with montelukasta
melting point
start of
crystallization (min)
amine
separated form
yield (%)
(°C)
n-propylamine
solid state
solid state
solid state
solid state
solid state
oil
95
94
92
85
92
s
s
s
82
s
52
97-100
97-100
98-101
93-96
98-102
s
10
10
45
120
5
isopropylamine
tert-butylamine
benzylamine
R-methylbenzylamine
2-methylaminoethanol
di-n-propylamine
di-isopropylamine
dicyclohexylamine
triethylamine
s
not separated
not separated
solid state
oil
s
s
s
s
120
s
180
118-121
s
152-154
di-isopropylethylamine
solid state
a Ten milliliters of toluene, 2.5 mL of acetonitrile and 1.05 equiv of the amine were added to 1 g of montelukast acid; 10 mL of n-heptane was gradually added to the
solution under stirring. In case of solid separation the montelukast salt was filtered, washed with 5 mL of heptanes and dried.
428
•
Vol. 14, No. 2, 2010 / Organic Process Research & Development

1. This unwanted reaction is probably allowed due to increased
reactivity of nucleophilic reagent. This also is only one negative
side effect of added polyether which was observed. Stereo-
chemical differentiation of compounds 9 and 10 was made after
assignment of their 1H and 13C NMR signals in NMR spectra
measured in deuteriochloroform solution. This stereochemical
differentiation comes from the premise that both impurities 9
and 10 come into existence from the montelukast which has
one chiral centre with configuration R, and this configuration
remains unchanged. Stereochemistry of the second chiral carbon
was determined on the basis of NOE interactions in NMR
spectra and thus distances of protons in the surrounding area
of both chiral centers of the molecule. The contents of
diastereoisomers 9 and 10, shortly referred to as montelukast
diastereoisomer I and diastereoisomer II, respectively, were
successfully removed by means of crystallization of salts of
montelukast with amines in polar solvents, e.g., isopropyl
alcohol. During the montelukast sodium process development,
four critical impurities were observed, namely (Z)-montelukast
8, (E)-montelukast sulfoxide 7, and diastereoisomers 9 and 10.
The preparation of critical montelukast impurities, including the
methods of their separation and isolation, are described in the
Experimental Section.
Figure 4. Contents of montelukast sulfoxide 7 (by HPLC) in
montelukast 1 during final step, depending on base used.
action of sodium methylate as a suitable source of sodium ions.
Previously selected sodium tert-butoxide has been changed due
to its mild oxidation effect, see Figure 4. The overall yield of
montelukast sodium 1 comprising five whole separate operations
was approximately 28%; the achieved chemical purity was
higher than 99.5% with the contents of individual impurities
below 0.1%. Analytical data obtained by means of XRPD
unambiguously characterize the amorphous features of mon-
telukast sodium 1.
Conclusions
The improved process represents a new and advantageous
method of carrying out the key substitution reaction that leads
to montelukast 1. In carrying out the key step in accordance
with the present process, the resulting composition of the
reaction mixture is conveniently controlled without the necessity
of long-term cooling or the use of protective groups. Isolation
and purification lead to the product which conforms to all the
regulatory requirements.
Scalable Process. The process for preparation of mon-
telukast sodium 1 used in the scale-up experiment consists of
five separate operations. In the first step alcohol 2 was converted
to the intermediate 4 by action of methanesulfonyl chloride in
acetonitrile medium and with the addition of DIPEA as a base.
Compound 4 of low stability was isolated in the paste form,
and it was used immediately in the next synthetic step. In the
second step was mixed acid 3 with sodium tert-butoxide and
PEG-600 in toluene. The obtained mixture was cooled below
-5 °C, and then a solution of 4 in THF was added dropwise.
Further, the reaction mixture was maintained under the inert
atmosphere and stirred for several hours. This solution usually
contained 80-85% of montelukast 1 according to HPLC
analyses. In the third step crude montelukast sodium was
converted to a solution of montelukast acid and further isolated
in the form of a crystalline salt of montelukast with isopropy-
lamine. In the fourth step the salt of montelukast with isopro-
pylamine was purified by crystallizations in isopropyl alcohol
and toluene. The final removal of the impurities occurs in this
step of the scale-up process. The target amorphous form of
montelukast sodium was obtained in the last step by direct
conversion of the montelukast isopropylammonium salt by
Experimental Section
Analytical Methods. The standards of critical impurities 7,
8, 9 and 10 were obtained by separations with the use of the
Waters autopurification system, and their structures were
checked by NMR and MS. The Waters autopurification system
with SQ detector, XBridge Prep OBC C18 column, 100 mm
× 19 mm, and mixture of two mobile phases was used. The
Waters autopurification system is a combination of various
instruments integrated in a specific configuration that enables
automated separation and isolation of particular substances from
a sample on the basis of a signal from UV and MS detectors.
1H and 13C NMR spectra were measured with the Bruker
Avance 500 spectrometer with the measuring frequency of
500.131 and 125.762 MHz, respectively. The spectra were
measured in DMSO-d6 or in CDCl3 solutions. 1H chemical shifts
were related to TMS (δ ) 0.00 ppm) and 13C chemical shifts
to DMSO-d6 (δ ) 39.6 ppm) or CDCl3 (δ ) 77.6 ppm). The
mass spectra were measured using a Sciex API 3000 Mass
Spectrometer (Sciex, Canada) with positive atmospheric pres-
sure ionization (TurboIonspray) or using LTQ Orbitrap Hybrid
Mass Spectrometer (Thermo Finnigan, U.S.A.) with direct
injection into APCI source in positive mode. X-ray diffraction
patterns (XRPD) of amorphous montelukast sodium were
measured with the X’PERT PRO MPD PAN analytical dif-
fractometer: radiation Cu KR (λ ) 1.54178 Å), graphite
(29) Halama, A.; Jirman, J., Petrickova, H. WO 2009006861 A2 (Zentiva,
2009).
(30) Egekeze, J. O.; Danielski, M. C.; Grinberg, N.; Smith, G. B.; Sidler,
D. R.; Perpall, H. J.; Bicker, G. R.; Tway, P. C. Anal. Chem. 1995,
67 (13), 2292.
(31) Halama, A.; Bouskova, O.; Gibala, P.; Jirman, J. WO 2009111998
A2 (Zentiva, 2009).
Vol. 14, No. 2, 2010 / Organic Process Research & Development
•
429

monochromator, 45 kV excitation voltage, 40 mA anode current,
measurement range 2-40° 2θ, increment 0.01 2θ. Melting
points of the salts of montelukast with amines obtained in the
solid state were measured on the Kofler block with the sample
heating speed of 10 °C (to 70 °C) and 4 °C (over 70 °C) per
minute. The measured values of melting points are summarized
in Table 2. HPLC (isocratic mode) chromatograms were
measured with the EliteLachrom device made by the Hitachi
Company. Stationary phase: RP-18e was used for the analyses;
column temperature was 20 °C. Mobile phase: Acetonitrile
(80%) and a 0.1 M aqueous solution of ammonium formate
adjusted to pH 3.6 with formic acid (20%) were used. The flow
rate of the mobile phase was 1.5 mL/min. Detection at the
wavelength of 234 nm was used. Methanol was used as the
solvent for preparation of samples; 10-20 µL of the solution
was used for the injection. The isocratic HPLC method was
used for checking the compositions of the reaction mixtures.
HPLC (gradient mode) chromatograms were measured with
the Alliance HPLC device with PDA detector. Stationary phase:
STAR RP-8e, 250 mm × 4 mm, 5 µm was used for the
analyses; column temperature was 15 °C. Mobile phase:
Acetonitrile (A) and 0.01 M aqueous solution of KH2PO4
adjusted to pH 2.2 with phosphoric acid (B) were used. Gradient
mode with the flow rate of mobile phase 0.8 mL/min was used.
Composition on the start was 60% of A and 40% of B, then
changed to 15% of A and 85% of B over 20 min; this
composition was held for 5 min, then changed to 60% of A
and 40% of B over 5 min, and this composition was held to
the end (overall time 35 min.). Detection at the wavelength of
234 nm was used. Methanol was used as the solvent for the
preparation of the samples; 10-20 µL of the solution was used
for the injection. The gradient HPLC method was used for
checking the quality of the target substance including its salts
with amines and of isolated standards of impurities. HPLC
(determination of (S)-enantiomer by HPLC) chromatograms
were measured with the Alliance HPLC device with PDA
detector. Stationary phase: Chiralpak IA (5 µm), size 0.25 m,
internal diameter 4.6 mm (manufactured by Daicel) was used
for the analyses, column temperature 10 °C. Mobile phase:
hexane/ethanol/1,4-dioxan/trifluoroacetic acid (77:3:20:0,1 v/v/
v) was used. The flow rate of the mobile phase was 1.0 mL/
min. Detection at the wavelength of 285 nm was used. Methanol
was used as the solvent for preparation of samples; 10 µL of
the solution was used for the injection. The isocratic elution
was used for checking the optical purity of target montelukast.
Typical retention times: montelukast: 9.3 min, (S)-montelukast:
12.9 min.
under an inert atmosphere. The cake was washed with cooled
acetonitrile (2 × 4 L, -15 °C). The yield of the crude product
in a paste form was 6.4 kg which corresponds to 3.33 kg of 4
recalculated to dry product (yield 71 %). Obtained crude product
was used immediately in the next step.
Preparation of Crude Montelukast Sodium 1 (Scale-Up
Process for the Key Synthetic Step). In 25.6 L of toluene,
[1-(mercaptomethyl)cyclopropyl]acetic acid 3 (950 g), sodium
tert-butoxide (1.25 kg) and PEG-600 (3.6 kg in 3.8 L of toluene)
were mixed together; the mixture was stirred under argon
atmosphere and cooled to ∼-10 °C. Then, a solution of 4 (3.33
kg recalculated to dry product) in 14 L of tetrahydrofuran was
added to the obtained slurry. The reaction mixture was gradually
stirred from -10 °C up to the room temperature for 2 h. It was
further stirred at room temperature (about 21 °C) for 5-6 h.
The reaction mixture was monitored by means of HPLC. At
the end of the monitoring, the reaction mixture contains 85.7%
of montelukast.
Isolation and Crystallization of Montelukast Salt with
Isopropylamine (Scale-Up Process). The reaction mixture
from the previous step was concentrated in vacuum; 10.5 L of
toluene was added to the residue and concentrated in vacuum
again. The residue was diluted with toluene to the volume of
27 L. This solution was washed twice with a 0.5 M solution of
tartaric acid (4 L) and once with 4 L of water, and the obtained
toluene solution was dried over sodium sulfate. Then, filtration
of the drying agent and addition of 3.4 L of acetonitrile, 970
mL of isopropylamine and 16.1 L of n-heptane followed. After
one hour of stirring another 8.5 L of n-heptane was added to
the suspension, and it was stirred for another hour. Then,
filtration was performed, and the cake was washed with 2 × 2
L of toluene and 2 × 2 L of n-heptane. An off-white powder,
2.8 kg, was obtained after drying at room temperature in
vacuum. The salt of montelukast with isopropylamine was
repeatedly recrystallized from isopropyl alcohol (twice) and
toluene (once). An off-white powder (1.81 kg, HPLC 99.7%)
was obtained after drying at 55 °C. The yield comprising both
the synthesis of the crude sodium salt of montelukast and
isolation and recrystallization of the salt with isopropylamine
was 45%.
1H NMR (DMSO-d6) δ (ppm) 0.23-0.47 (m, 4H, 2 × CH2
cyclopropyl), 1.08 (d, 6H, 2 × CH3 isopropyl), 1.44 (s, 6H, 2
× CH3), 2.10-2.30 (m, 4H, 2 × CH2), 2.51 (m, 1H, CH), 2.52
and 2.63 (m, 2H, CH2), 2.77 a 3.07 (2 × m, 2H, CH2), 3.06
(m, 1H, CH isopropyl), 4.01 (t, 1H, CH), 5.70 (bb, 4H, NH3+,
OH), 7.03-8.41 (m, 15H, CHdCH, and CH-arom.).
Preparation of the Amorphous Form of Montelukast
Sodium (Scale-Up Process). Toluene (19.4 L) was added to
the crystalline salt of montelukast with isopropylamine (1.3 kg).
The suspension was stirred for 20 min, then sodium methylate
(112.6 g) was added, and the suspension was further stirred at
the temperature of approximately 68-72 °C for 45 min. Then,
filtration was performed, and the clear, pale-yellow-colored
filtrate was injected into 43 L of intensively stirred n-heptane.
The obtained suspension was stirred for another hour and then
was filtered. An off-white amorphous powder (1.09 kg, yield
89%, HPLC 99.7%, amorphous feature was checked by X-ray
Preparation of 2-(3-(S)-(3-(2-(7-Chlorquinolinyl)ethe-
nyl)phenyl)-3-methanesulfonyloxypropyl)phenyl-2-pro-
panol 4 (Scale-Up Process). A mixture of 2-(2-(3(S)-(3-(7-
chloro-2-quinolinyl)ethenyl)phenyl)-3-hydroxypropyl)phenyl-
2-propanol 2 (4 kg), acetonitrile (21 L) and toluene (8,4 L) was
stirred under inert atmosphere and cooled to -15 °C. Then di-
isopropylethylamine (1.89 L in 0.5 L acetonitrile) was added
gradually. A solution of methanesulfonylchloride (0.79 L) in
toluene (1.1 L) was added in portions over 20 min, maintaining
the temperature at -13 to -15 °C. The reaction mixture was
stirred for 3.5 h at -15 °C, and the separated solid was filtered
430
•
Vol. 14, No. 2, 2010 / Organic Process Research & Development

powder diffraction) was obtained after vacuum drying under
the stream of nitrogen at the temperature of 55 °C.
washed with a solution of tartaric acid and water. The organic
phase was dried over sodium sulfate and concentrated to the
volume of 30 mL after filtration of the desiccant. To the
concentrated residue were added 3 mL of acetonitrile, 0.5 mL
of isopropylamine and gradually 30 mL of n-heptane. The
separated suspension of the salt of montelukast with isopropy-
lamine was filtered off, and the filtered mother liquor was
concentrated in vacuum. An oily product (0.4 g) was obtained
containing 70% of a mixture of montelukast diastereoisomers
I and II (according to HPLC). The standards of both the
diastereoisomers in the form of free acids were obtained in the
quantities of approximately 80 mg with the use of the Waters
autopurification system (see Analytical Methods).
Preparation of the Standard of [R-(E)]]-1-[[[1-[3-[2-(7-
Chloro-2-quinolinyl)ethenyl]phenyl]-3-[2-(1-hydroxy-1-me-
thylethyl)phenyl]propyl]sulfinyl]methyl]cyclopropane Acetic
Acid 7 (Montelukast Sulfoxide). Montelukast sodium (6.0 g)
was dissolved in 100 mL of methanol; subsequently, 30%
hydrogen peroxide (30 mL) was added. After 3 h of stirring,
methanol was evaporated in vacuum, and acetic acid (3 mL)
was added to the residue. Water (50 mL) was added to the
separated suspension, and the mixture was stirred for 30 min;
then, the suspension was filtered, the cake was washed with
water, with a toluene/heptane (1:1) mixture and finally with
n-heptane. After vacuum drying at 65 °C, 5.5 g of a yellow
powder was obtained with the chemical purity 93% (by HPLC).
A portion of the crude material was subjected to the autopu-
rification separation system (see Analytical Methods), while the
standard (∼250 mg) was obtained.
Montelukast Diastereoisomer II: 1H NMR (CDCl3) δ (ppm):
0.36-0.62 (m, 8H); 1.56 (s, 3H); 1.57 (s, 3H); 1.98 (m, 1H);
2.13 (m, 1H); 2.00 (m, 1H); 2.46 (m, 1H); 2.20 (d, 1H); 2.65
(d, 1H); 2.36 (d, 1H); 2.48 (d, 1H); 2.77 (m, 1H); 2.96 (m,
1H); 3.29 (dd, 1H); 3.68 (dd, 1H); 3.95 (t, 1H); 4.36 (dd, 1H);
7.09 (m, 1H); 7.10 (m, 1H); 7.16 (dd, 1H); 7.18 (dd, 1H); 7.27
(m, 1H); 7.28 (m, 1H); 7.30 (m, 1H); 7.35 (m, 1H); 7.36 (m,
1H); 7.40 (dd, 1H); 7.61 (d, 1H); 8.02 (d, 1H); 8.09 (d, 1H).
13C NMR (CDCl3) δ (ppm): 12.1; 12.3; 12.5; 12.8; 16.6; 17.0;
31.6; 31.7; 32.1; 38.8; 39.1; 40.0; 40.5; 45.7; 50.3; 51.0; 74.3;
123.3; 125.3; 125.4; 125.6; 125.9; 126.8; 127.1; 127.2; 127.4;
128.2; 129.1; 131.7; 135.9; 136.7; 140.4; 142.4; 142.7; 144.8;
147.3; 161.0; 175.9; 176.2. MS: 432.2591 (M + 1)+.
1H NMR (DMSO-d6) δ (ppm): 0.34 (m, 1H); 0.48 (m, 1H);
0.63 (m, 1H); 1.42 (s, 3H); 2.22 (m, 1H); 2.26 (m, 1H); 2.44
(m, 1H); 2.46 (m, 1H); 2.61 (d, 1H, J ) 13.7 Hz); 2.67 (d, 1H,
J ) 13.7 Hz); 2.76 (m, 1H); 2.97 (m, 1H); 4.05 (dd, 1H, J )
11.0 and 4.1 Hz); 7.10 (m, 1H); 7.15 (m, 1H); 7.36 (d, 1H, J
) 7.7 Hz); 7.38 (d, 1H, J ) 7.6 Hz); 7.48 (t, 1H, J ) 7.7 Hz);
7.54 (d, 1H, J ) 16.4 Hz); 7.61 (dd, 1H, J ) 8.7 and 2.1 Hz);
7.73 (d, 1H, J ) 7.8 Hz); 7.76 (s, 1H); 7.91 (d, 1H, J ) 16.4
Hz); 7.96 (d, 1H, J ) 8.6 Hz); 8.03 (d, 1H, J ) 8.8 Hz); 8.04
(d, 1H, J ) 2 Hz); 8.45 (d, 2H, J ) 8.6 Hz). 13C NMR (DMSO-
d6) δ (ppm): 10.8; 12.3; 13.9; 30.9; 31.3; 31.6; 40.1; 56.9; 66.4;
71.6; 120.3; 125.3; 125.6; 126.4; 126.8; 127.0; 127.9; 128.3;
129.3; 129.4; 129.8; 131.0; 134.5; 135.2; 136.0; 136.4; 136.9;
139.6; 146.7; 147.5; 156.5; 172.6. MS: 602.2125 (M + 1)+.
Preparations of the Standards of 2-[(R)-1-[[[1-[3-[2-(7-
Chloro-2-quinolinyl)-(S)-1-({[1-(carboxymethyl)cyclopropyl]-
methyl}thio)ethyl]phenyl]-3-[2-(1-hydroxy-1-methylethyl)phe-
nyl]propyl]thio]methyl]cyclopropane]acetic Acid 9 (Diastereo-
isomer I) and 2-[(R)-1-[[[1-[3-[2-(7-Chloro-2-quinolinyl)-
(R)-1-({[1-(carboxymethyl)cyclopropyl]methyl}thio)ethyl]-
phenyl]-3-[2-(1-hydroxy-1-methylethyl)phenyl]propyl]thi-
o]methyl]cyclopropane]acetic Acid 10 (Diastereoisomer II).
A solution of 2.6 g of PEG-600 in 3 mL of toluene was added
under argon atmosphere to the mixture of acid 3 (0.7 g), toluene
(20 mL) and sodium tert-butoxide (0.85 g). Then, a solution of
montelukast sodium (2.72 g) in 15 mL of tetrahydrofuran was
added dropwise to the stirred mixture. The obtained mixture
was stirred at the room temperature and under argon atmosphere
for 30 days. Then, toluene (100 mL) was added, and 45 mL of
the liquid was removed by vacuum distillation. The residue was
Montelukast Diastereoisomer I: 1H NMR (CDCl3) δ (ppm):
0.38 (m, 2H); 0.41-0.48 (m, 4H); 0.56 (m, 2H); 1.60 (s, 3H);
1.61 (s, 3H); 2.00 (d, 1H); 2.40 (d, 1H); 2.16 (q, 2H); 2.33 (d,
1H); 2.43 (d, 1H); 2.35 (d, 1H); 2.55 (d, 1H); 2.37 (d, 1H);
2.49 (d, 1H); 2.97 (q, 2H); 3.37 (dd, 1H); 3.62 (dd, 1H); 3.93
(t, 1H); 4.29 (dd, 1H); 7.11 (m, 1H); 7.17 (m, 2H); 7.23 (m,
1H); 7.24 (m, 1H); 7.28 (m, 1H); 7.30 (m, 1H); 7.38 (d, 1H);
7.4 (dd, 1H); 7.70 (d, 1H); 8.08 (d, 1H); 8.11(d, 1H). 13C NMR
(CDCl3) δ (ppm): 12.2; 12.5; 12.6; 17.1; 17.7; 31.6; 31.8; 32.4;
38.1; 39.2; 39.3; 40.0; 41.2; 45.4; 50.7; 51.2; 123.2; 125.4;
125.5; 126.1; 126.2; 126.8; 127.3; 127.5; 128.2; 128.8; 129.4;
131.8; 135.9; 139.7; 140.7; 142.3; 142.6; 144.6; 147.2; 161.0;
175.7; 175.9. MS: 432.2586 (M + 1)+.
Acknowledgment
We thank the management of Zentiva k.s. for supporting
this work. Close cooperation with colleagues from the R&D
Department is appreciated.
Received for review December 3, 2009.
OP900311Z
Vol. 14, No. 2, 2010 / Organic Process Research & Development
•
431