A convenient method for the synthesis of roflumilast

Article abstract of DOI:10.1007/s11164-012-0742-3

A convenient synthetic method has been developed for the synthesis of roflumilast from 4-difluoromethoxy-3-hydroxybenzaldehyde and bromomethyl cyclopropane via O-alkylation, oxidation and N-acylation. With sodium hydroxide as alkali in the last step, the total yield of roflumilast can be up to 68.3 % and the purity of the target product reached 99.2 %. It was obvious that sodium hydroxide showed more economic advantage for scale-up production than sodium hydride or potassium tert-butoxide.

Full text of DOI:10.1007/s11164-012-0742-3

Res Chem Intermed (2013) 39:2107–2113
DOI 10.1007/s11164-012-0742-3
A convenient method for the synthesis of roflumilast
•
Yan Lin Peijun Huang Shuai Liu Lifeng Sima
•
•
•
•
Ligong Chen Donghua Wang
Received: 17 May 2012 / Accepted: 22 July 2012 / Published online: 7 August 2012
Ó Springer Science+Business Media B.V. 2012
Abstract A convenient synthetic method has been developed for the synthesis
of roflumilast from 4-difluoromethoxy-3-hydroxybenzaldehyde and bromomethyl
cyclopropane via O-alkylation, oxidation and N-acylation. With sodium hydroxide as
alkali in the last step, the total yield of roflumilast can be up to 68.3 % and the purity of
the target product reached 99.2 %. It was obvious that sodium hydroxide showed more
economic advantage for scale-up production than sodium hydride or potassium tert-
butoxide.
Keywords Acylation Á Chronic obstructive pulmonary diseases Á Roflumilast Á
Sodium hydroxide
Introduction
Roflumilast is one of the second generation PDE-4 inhibitors and is the first drug to
be approved for the treatment of chronic obstructive pulmonary disease (COPD) in
Europe in the last 10 years. Now, it is widely used for the treatment of COPD,
asthma, pulmonary hypertension and acute respiratory disease syndrome [1–3]. A
synthetic route of roflumilast and its pharmacological actions were reported by
Y. Lin Á S. Liu Á L. Sima Á L. Chen (&)
School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072,
People’s Republic of China
e-mail: lgchen@tju.edu.cn
P. Huang
Tianjin Bohai Chemical Industry Group Co. Ltd, Tianjin 300040,
People’s Republic of China
D. Wang
School of Pharmaceutical Science and Technology, Tianjin University, Tianjin 300072
People’s Republic of China
123

2108
Y. Lin et al.
Sorbera et al. [4], the synthetic route consisting of O-alkylation, oxidation and
N-acylation reaction, but there was just a scheme and no experimental details were
mentioned in that paper. Obviously, the N-acylation reaction is the key step in the
synthesis because 3,5-dichloropyridine-4-amine is not a good nucleophile and it is
difficult to carry out the nucleophilic substitution reaction under normal reaction
conditions. Sodium hydride was first reported to be used to activate the amine [5–7].
But the mineral oil in it is usually troublesome for the purification of the final
product, and the hydrogen generated would cause security problems. Kohl et al. [8]
intensively studied the synthesis of roflumilast, particularly the effects of alkalis
employed in the last step, in which potassium tert-butoxide yielded the best results.
In addition, alkalis in that paper only included lithium tert-butoixde, sodium tert-
butoxide and potassium tert-butoxide, thus developing a new method employing
sodium hydroxide in this reaction has good prospect.
Based on the above investigations, with consideration of the convenience of sodium
hydroxide, in this paper, we have synthesized roflumilast from 4-difluoromethoxy-3-
hydroxybenzaldehyde and bromomethyl cyclopropane, and have employed sodium
hydroxide in the last step (Scheme 1). By optimizing the reaction conditions, the total
yield of roflumilast can be up to 68.3 %.
Results and discussion
As described in Scheme 1, roflumilast was synthesized via O-alkylation, oxidation,
chlorination of the carboxylic acid and N-acylation reaction. The O-alkylation
Scheme 1 Synthesis of roflumilast
123

A convenient method for the synthesis of roflumilast
2109
reaction has been reported to be achieved in N,N-dimethylformamide [9], followed by
the treatment of the reaction mixture with water and extraction of the aqueous phase by
organic solvents, which contributed to effluent burden in the scale-up industrializa-
tion. Thus, in this paper, a number of solvents were examined for this reaction and the
results are listed in Table 1. It can be seen that polar aprotic solvents would be helpful
to the reaction. The O-alkylation reaction in methanol proceeded quite slowly and
gave low yield, while higher yield was obtained in ethanol but in 40 h. Only in acetone
did the alkylation proceed efficiently to afford the desired compound in satisfactory
yield. In order to enhance the reaction time, KI was employed as catalyst and this led to
an excellent result. Compound 2 as yellow oil was obtained in the yield of 94.4 %,
which was directly used in the next oxidation step without further purification. The
oxidation was achieved straightforwardly by sulfamic acid and sodium chlorite in
acetonitrile to give the important intermediate, 3-(cyclopropylmethoxy)-4-(difluoro-
methoxy) benzoic acid in the total yield of 76.5 %.
Then, we focused on the investigation of the last step. As seen in Scheme 1,
roflumilast was achieved by the N-acylation of compound 5 with compound 4.
Because the nucleophilic ability of 3,5-dichloropyridine-4-amine was low due to the
electronic and steric effects, the activation of the amine is required by alkali, such as
potassium tert-butoxide and sodium hydride [7, 10]. In view of accessibility,
convenience and the low cost, we employed sodium hydroxide in this reaction. It
can be seen that the amino anion of 3,5-dichloropyridine-4-amine can be obtained in
the reverse reaction equilibrium (Scheme 2). And sodium hydroxide showed such
low basicity that it was more conductive to the forward equilibrium, which was
unfavorable for the generation of the amino anion. Thus, by continuously removing
the water, the equilibrium was very favorable for the reverse reaction balance and
the amino anion could be generated efficiently. The acyl chloride (compound 4) can
be easily obtained by the chlorination of compound 3 with SOCl2. Then it
(compound 4) was added to the suspending mixture to yield the roflumilast product.
So dehydration was of great importance in this reaction, and this could be the main
reason for the low yields in the preliminary experiments.
Then, the reactions were performed in the presence of different molar ratios of
compounds 3, 5 and sodium hydroxide to find the effects on the N-acylation
reaction. The results are shown in Table 2. The results showed that the yields of
roflumilast were all over 80 % under these conditions. And when the molar ratio of
compounds 3, 5 and NaOH was 1:2:4, the reaction gave the best yield of 89.2 %
while there was more than 0.5 % of compound 8 in the purified product after
repeated crystallization. But compound 8 was less in the final product when the
reaction molar ratio was 1:2.5:2.25, which contributed to a higher quality of
Table 1 Effects of solvents on
the O-alkylation reactiona
Entry
Solvents
Reaction time (h)
Yield %
1
2
3
4
Methanol
Ethanol
[16
40
45.7
88.2
91.7
94.4
Acetone
18
a
The yields of compound 2
were detected by HPLC
Acetone/KI
12
123

2110
Y. Lin et al.
Scheme 2 Formation of the sodium salt of 4-amino-3,5-dichloropyridine
Table 2 Effects of molar ratio
Entry Molar ratio of compounds Yield of 7 % Content of 8 %
3, 5 and NaOH
on the N-acylation reactiona
5
6
7
8
1.0:2.0:4.0
1.0:2.0:2.4
1.0:2.0:1.8
1.0:2.5:2.25
89.2
85.1
81.7
87.6
1.2
0.8
0.6
0.5
a
The reaction was refluxed in
toluene for 6 h. The yields and
contents were detected by HPLC
Scheme 3 Structure of
compound 8
Scheme 4 The supposed mechanism of the generation of compound 8
product. In addition, the mass of this compound 8 was measured at 353.9 by a liquid
chromatography–mass spectrometry instrument [8]. The structure of it can be seen
in Scheme 3, and the supposed mechanism of its formation in Scheme 4. In this
figure, the nucleophilic substitution of roflumilast yielded the phenoxy anion,
followed by the generation of compound 8. It can be seen that excess sodium
hydroxide contributed to the formation of compound 8. So the amounts of
compound 8 can be decreased by increasing the molar ratio of 3,5-dichloropyridine-
4-amine to sodium hydroxide, and this was of great help for the purification of
roflumilast. Thus, the optimum molar ratio of compounds 3, 5 and sodium
hydroxide was 1:2.5:2.25.
123

A convenient method for the synthesis of roflumilast
2111
Table 3 Effects of the reaction
temperature and time on the
N-acylation reactiona
Entry
Reaction
temperature (°C)
Reaction
time (h)
Yield of
7 %
Yield of
8 %
9
35
110
110
110
110
6
2
3
6
7
17.2
68.5
81.5
87.6
87.3
0.1
0.2
0.2
0.5
2.5
10
11
12
13
a
The molar ratio of compound
3, 5 and NaOH in the reactions
was 1:2.5:2.25. The yields of
were detected by HPLC
In order to further decrease the amount of compound 8, both the reaction
temperature and reaction time were studied. The results are listed in Table 3. We
can see that the N-acylation reaction and the generation of compound 8 competed in
this reaction system, while the low temperature slowed down the N-acylation
reaction process more significantly and this was disadvantageous for the roflumilast
product yield. However, the reaction conducted at reflux afforded better yield of the
roflumilast, which was 87.6 %. Furthermore, the results showed that the amounts of
both compound 8 and roflumilast increased with the extension of the reaction time.
So 3 h was the suitable reaction time.
Conclusions
In conclusion, we have developed a convenient synthetic method of roflumilast from
4-difluoromethoxy-3-hydroxybenzaldehyde and bromomethyl cyclopropane via
O-alkylation, oxidation and N-acylation reaction. By optimizing the reaction
conditions, roflumilast can be obtained in 68.3 % total yield and the purity of the
product reached up to 99.2 %. The results showed that, with sodium hydroxide as
the alkali in the N-acylation reaction, roflumilast can be obtained at the best yield of
89.2 %. Compared with other methods, the present method has several advantages
including convenience, low cost and easy work-up, which allows an easy scale-up.
Experimental
General information
Reagents and solvents were obtained from commercial suppliers. All reactions were
monitored by thin layer chromatography using commercial silica gel plates. The
purity was detected by HPLC on Agilent 1,100 series. Melting points were observed
on YRT-3 Melting Point Tester and were uncorrected. NMR spectra were recorded
on Varian Inova-500 and 400 MHz NMR spectrometer with TMS as an internal
reference. MS were recorded on a LCQ Advanced MAX mass spectrometer.
HR-MS was recorded on MicroOTOF-Q II.
123

2112
Y. Lin et al.
Synthesis of 3-(cyclopropylmethoxy)-4-(difluoromethoxy) benzaldehyde (2)
4-(Difluoromethoxy)-3-hydroxybenzaldehyde (compound 1, 53.16 mmol), K2CO3
(79.74 mmol), KI (10.63 mmol) in 250 mL acetone were added to a flask with
magnetical stirring under nitrogen atmosphere. After the mixture was refluxed for
0.5 h, bromomethyl cyclopropane (79.74 mmol) was added to the flask within
0.5 h. Then, the mixture was stirred and refluxed for 12 h. After cooled to room
temperature, the reaction mixture was filtered and concentrated. The obtained oil
was dissolved in ethyl acetate and washed with water for three times. Then, the
solvent was evaporated giving the compound 2 in yellow oil with the purity of
1
90.09 %. And this can be used for the next process without purification. H NMR
(CDCl3, 500 MHz) d: 9.93 (s, 1H, CHO), 7.47 (d, J = 2.5 Hz, 1H, CH of benzene),
7.45 (d, J = 1.5 Hz, 1H, CH of benzene), 7.32 (d, J = 8.0 Hz, 1H, CH of benzene),
6.76 (t, J = 74.5 Hz, 1H, CF2H), 3.95 (d, J = 7.0 Hz, 2H, OCH2), 1.35–1.29
(m, 1H), 0.70–0.66 (m, 2H), 0.40–0.37 (m, 2H); 13C NMR (CDCl3, 100 MHz)
d: 190.68 (C=O), 150.88, 145.20, 134.55, 124.65, 121.94 (5 C benzene), 118.27,
115.68, 113.08 (CHF2), 112.61, 73.98 (CH2O), 9.89, 3.13. HR-MS (ESI),
calcd C12H12F2O3: [M?Na]?, m/z: 265.0652, found: 265.0650.
Synthesis of 3-(cyclopropylmethoxy)-4-(difluoromethoxy) benzoic acid (3) [9, 11]
Compound 2 was dissolved in 100 mL acetonitrile and added to the solution of
sulfamic acid (74.42 mmol) in 50 mL water. Then, sodium chlorite (74.42 mmol) in
50 mL water was added slowly at 0–10 °C and the reaction mixture was stirred at
20 °C for 5 h. Then, the acetonitrile was evaporated and the residue was dissolved
in 5 wt%. NaOH solution, followed by washing with CH2Cl2 for three times. The
aqueous solution was acidized by 1 mol/L hydrochloric acid until pH 3. The
suspension was filtered and the solid was washed with water and dried to afford
compound 3 as white solid (10.50 g, with the purity of 99.82 %). Mp.120 °C
(lit. [12] 118–120 °C) 1H NMR (CDCl3, 500 MHz) d: 7.73 (d, J = 8.0 Hz, 1H, CH
of benzene), 7.68 (s, 1H, CH of benzene), 7.24 (d, J = 8.5 Hz, 1H, CH of benzene),
6.75 (t, J = 74.5 Hz, 1H, CF2H), 3.95 (d, J = 7.0 Hz, 2H, OCH2), 1.36–1.28
(m, 1H), 0.70–0.66 (m, 2H), 0.40–0.38 (m, 2H). (lit. [12].: 1H NMR (CDCl3,
300 MHz) d: 7.70 (d, J = 8.4 Hz, 1H), 7.65 (s, 1H), 7.22 (d, J = 8.1 Hz, 1H), 6.72
(t, J = 74.7 Hz, 1H), 3.94 (d, J = 6.6 Hz, 2H), 1.2–1.37 (m, 1H), 0.71–0.65 (m, 2H),
0.41–0.36 (m, 2H).)
Synthesis of 3-(cyclopropylmethoxy)-4-(difluoromethoxy) benzoyl chloride (4) [13]
Compound 3 (3.9 mmol) in 50 mL toluene was heated to reflux until dissolving and
the water was removed by Dean-Stark apparatus. The solution was then cool to
90 °C, SOCl2 (5.8 mmol) and several drops of DMF were added. The solution was
heated at reflux for 3.0 h. Then, it was cooled to room temperature and concentrated
to give compound 4 as yellow crystal.
123

A convenient method for the synthesis of roflumilast
2113
Synthesis of roflumilast (compound 7)
Sodium hydroxide (15.50 mmol) was added to the mixture of compound 5
(7.75 mmol) in 40 mL toluene. Water was removed by azeotrope distillation. After
2 h, the chloride compound 4 in dry toluene was added and the mixture was stirred
at reflux for 6 h. After the reaction mixture was cooled to room temperature, it was
washed successively by 1 mol/L hydrochloric acid, saturated sodium bicarbonate
solution and water. The organic phase was dried with magnesium sulfate and
concentrated to give crude product of roflumilast as yellow solid (1.05 g) in the
yield of 89.2 %. It was recrystallized in methanol/water = 90/10 (v/v) to yield the
1
product (99.26 %). Mp. 158 °C (lit. [10] 158 °C) H NMR (CDCl3, 500 MHz)
d: 8.56 (s, 2H, 2 CH of pyridine), 7.92 (s, 1H, NH), 7.59 (s, 1H, CH of benzene),
7.50–7.48 (dd, J1 = 2.0 Hz, J2 = 6.5 Hz, 1H, CH of benzene), 7.28 (d, J = 8.0
Hz, 1H, CH of benzene), 6.76 (t, J = 93.5 Hz, 1H, CF2H), 3.96 (d, J = 8.5 Hz, 2H,
OCH2), 1.36–1.30 (m, 1H), 0.71–0.66 (m, 2H), 0.40–0.37 (m, 2H); 13C NMR
(CDCl3, 125 MHz) d: 163.94 (C=O), 151.15, 148.57, 144.07 (3 C pyridine), 139.89,
131.03, 129.12, 122.54, 120.12, 114.42 (6 C benzene), 117.95, 115.87, 113.79
(CHF2), 74.43, 10.23, 3.54. HR-MS (ESI), calcd C17H14Cl2F2N2NaO3: [M?Na]?
m/z: 425.0242, found: 425.0208.
References
1. A.M. Vignola, Respir. Med. 98, 495–503 (2004)
2. M. Kobayashi, S. Kubo, M. Iwata et al., Int. Immunopharmacol. 11, 732–739 (2011)
3. R. Beume, A. Hatelmann, EP2366393 (2006)
4. L.A. Sorbera, P.A. Leeson, J. Castan˜er, Drugs Future 25, 1261–1264 (2000)
5. A. Hermann, F. Dieter, G. Beate, et al., WO9501338 (1994)
6. L.E. Williams, T.C. Wu, WO2004033430 (2004)
7. G.W. Smith, WO2005041864 (2005)
8. B. Kohl, B. Mueller, W. Palosch, WO2004080967 (2004)
9. C.D. Cook, H.R. Jones, H. Kabir, J.D. Lythgoe et al., Org. Process Res. Dev. 2, 157–168 (1998)
10. B. Kohl, B. Mueller, W. Palosch, WO20090171096 (2009)
11. E. Dalacanale, J. Org. Chem. 51, 567–569 (1986)
12. A. Thomas, K.P. Bhavar, V.S. Lingam, et al., WO200422536 (2004)
13. J.F. Xin, J.H. Ma, S.F. Zhang, Chem. Ind. Hebei 29, 16–20 (2006)
123