Application of [3+2]-cycloaddition in the synthesis of valdecoxib

Article abstract of DOI:10.1080/00397911.2010.528572

A large scale synthesis of valdecoxib 1 is described. Our work features potential application of [3+2]-dipolar cycloaddition involving enamine and in situ-generated nitrile oxide derivatives. Dr. Reddy's Laboratories Ltd.

Full text of DOI:10.1080/00397911.2010.528572

This article was downloaded by: [Universite Laval]
On: 04 March 2015, At: 00:42
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Synthetic Communications: An
International Journal for Rapid
Communication of Synthetic Organic
Chemistry
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lsyc20
Application of [3 + 2]-Cycloaddition in
the Synthesis of Valdecoxib
Anumula Raghupathi Reddy a b , Gilla Goverdhan a , Aalla Sampath a
, Khagga Mukkanti b , Padi Pratap Reddy a & Rakeshwar Bandichhor a
a Research and Development, Dr. Reddy's Laboratories Ltd. ,
Qutubullapur , India
b Institute of Science and Technology, J. N. T. University ,
Hyderabad , India
Accepted author version posted online: 24 Aug 2011.Published
online: 01 Nov 2011.
To cite this article: Anumula Raghupathi Reddy , Gilla Goverdhan , Aalla Sampath , Khagga
Mukkanti , Padi Pratap Reddy & Rakeshwar Bandichhor (2012) Application of [3 + 2]-Cycloaddition
in the Synthesis of Valdecoxib, Synthetic Communications: An International Journal for Rapid
Communication of Synthetic Organic Chemistry, 42:5, 639-649, DOI: 10.1080/00397911.2010.528572
To link to this article: http://dx.doi.org/10.1080/00397911.2010.528572
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Synthetic Communications1, 42: 639–649, 2012
Copyright # Dr. Reddy’s Laboratories Ltd.
ISSN: 0039-7911 print=1532-2432 online
DOI: 10.1080/00397911.2010.528572
APPLICATION OF [3 þ 2]-CYCLOADDITION IN THE
SYNTHESIS OF VALDECOXIB
Anumula Raghupathi Reddy,1,2 Gilla Goverdhan,1
Aalla Sampath,1 Khagga Mukkanti,2 Padi Pratap Reddy,1 and
Rakeshwar Bandichhor1
1Research and Development, Dr. Reddy’s Laboratories Ltd.,
Qutubullapur, India
2Institute of Science and Technology, J. N. T. University, Hyderabad, India
GRAPHICAL ABSTRACT
Abstract A large scale synthesis of valdecoxib 1 is described. Our work features potential
application of [3 þ 2]-dipolar cycloaddition involving enamine and in situ–generated nitrile
oxide derivatives.
Keywords Chlorination; cycloaddition; enamine; NSAID; sulfonation
INTRODUCTION
Valdecoxib 1 (Fig. 1) is a nonsteroidal anti-inflammatory drug (NSAID),
which acts as an inhibitor of cyclooxygenase-2 (COX-2), although in practice it is
used for the treatment of rheumatoid arthritis, osteoarthritis, and dysmenor-
rhea.[1–3] Several synthetic procedures for valdecoxib 1 are known. Apart from utiliz-
ing dipolar cycloaddition, most of the synthetic methods involve either industrially
nonviable reagents such as butyl lithium,[4,5] lithium diisopropylamide,[6] diethyl
chlorophosphate, and hexamethylphospharamide[7] or commercially unavailable
materials,[8] and a few of them suffer from lack of regioselectivity.[7,9]
Received May 16, 2010.
Dr. Reddy’s Communication No. IPM-00206.
Address correspondence to Padi Pratap Reddy and Rakeshwar Bandichhor, Research and
Development, Innovation Plaza, Dr. Reddy’s Laboratories Ltd., Survey Nos. 42, 45, 46, and 54,
Bachupally, Qutubullapur, R. R. District 500 072, A. P., India. E-mail: reddyppou@yahoo.co.in;
rakeswarb@drreddys.com
639

640
A. R. REDDY ET AL.
Figure 1. Structure of valdecoxib 1.
Figure 2. Example of in situ–generated nitrile oxide and its cycloaddition with olefin. (Figure is provided
in color online.)
[3 þ 2]-Cycloaddition of nitrile oxide with olefinic compounds is of synthetic
importance because the derivatives of isoxazoline are versatile intermediates for
the synthesis of bifunctional compounds of pharmaceutical relevance.[10] Tradition-
ally, the synthesis of nitrile oxides involves the oxidative dehydrogenation of
aldoximes using oxidants such as lead tetraacetate,[11] alkali hypochlorite,[12]
N-chlorosuccinamide (NCS) in dimethylformamide followed by base treatment,[13]
chloramine-T (CAT),[14] iodobenzene dichloride,[15] 1-chlorobenzotriazole,[16] or
mercuric acetate[17] as well as the reaction of nitro compounds with an aryl isocya-
nate as presented in Fig. 2.[18]
Based on the mechanistic pathway of cycloaddition as shown in Fig. 1 and the
reports on the application of enamine as an olefin equivalent in such reactions,[19,20]
we designed a novel synthetic route that allows us to access valdecoxib 1 efficiently.
Herein, we report the synthesis of valdecoxib 1 that involves [3 þ 2]-dipolar
cycloaddition, aromatization, and chlorosulfonation followed by sulfamidation.
RESULTS AND DISCUSSION
Required enamine 3 (olefin equivalent) was synthesized by employing phenyl-
acetone 2 with pyrrolidine in cyclohexane. The Safety of the process was ascertained
by conducting calorimetric experiments. Thereafter, enamine 3 was reacted with ben-
zonitrile oxide that was generated in situ from chlorobenzaldoxime[20] to provide dia-
rylisoxazole derivative 5. Safety concerns about in situ–generated nitrile oxide or its
precursor were addressed by conducting reaction calorimetric experiments following
the literature protocol.[21] The desired chlorobenzaldoxime was prepared by adopt-
ing the reported procedure.[13] Chlorosulfonation on 5 using chlorosulfonic acid in
dichloromethane followed by treatment with aqueous ammonia furnished the
desired valdecoxib 1 as shown in Scheme 1.
During the synthesis of 5, we encountered two intermediates, 3 and 4. While
intermediate 4 found to be quite stable (to make process robust, it was not isolated),
intermediate 3 was in equilibrium with the corresponding tautomer 7. As a result,

SYNTHESIS OF VALDECOXIB
641
Scheme 1. Novel approach to the synthesis of valdecoxib 1.
20% [by gas chromatography (GC)] of 7 was found to take part in cycloaddition to
afford cycloadduct 8 as an impurity [by high-performance liquid chromatography
(HPLC)] (Fig. 3) along with the desired product 4.
After acid-mediated aromatization of the mixture of 4 and 8, we were able to
isolate the desired isoxazole derivative 5 with 60% yield. Aromatized side product 9
(Fig. 4) originated from 8 and was separated by crystallization in isopropanol.
Chlorosulfonation of 5 afforded the corresponding chlorosulfonated inter-
mediate 6 in 60% yield. We observed poor yield (60%) of 6, and to find mass balance
we isolated at least two side products 10 (10% by HPLC) and 11 (15% by HPLC).
Additionally, we were also able to isolate chlorosulfonated side product of 9 as an
impurity 12. Removal of these two side products by crystallization using cyclohexane
led to the poor yield of 6. Because of poor stability of chlorosulfonated impurities,
without characterization, we converted them to corresponding sulfamidated moieties
(13, 14, and 15) (Fig. 5).
Figure 3. Existence of tautomer 7 and corresponding cycloadduct 8.

642
A. R. REDDY ET AL.
Figure 4. Aromatization and removal of 9.
In the final step, aqueous ammonia was used. There are a few more impurities
(16 and 17) formed as a result of the reaction of 6 with water and 1 respectively as
shown in Fig. 6.
After addressing impurity-related deficiencies and intensive process develop-
ment, we realized that the developed novel synthetic route can be carried out on a
multi-kilogram scale in the concurrent pilot plant scale.
CONCLUSION
We have developed an enamine directed [3 þ 2]-dipolar cycloaddition
process using commercially available inexpensive raw materials for the synthesis of
Figure 5. Structures of impurities 10–15.

SYNTHESIS OF VALDECOXIB
643
Figure 6. In-process impurities 16 and 17. (Figure is provided in color online.)
valdecoxib 1. This process is more advantageous than the reported processes and was
successfully implemented on a multi-kilogram scale in the plant.
EXPERIMENTAL
The 1H NMR spectra were measured in CDCl3, dimethylsulfoxide (DMSO-d6),
mixture of CDCl3 and DMSO-d6, and mixture of CDCl3 and CD3CN using 200-
MHz and 400-MHz Fourier transform (FT)NMR spectrometers. The chemical shifts
are reported in d ppm relative to tetramethylsilane (TMS). The FT-IR spectra were
recorded in the solid state as KBr dispersion using a FT-IR spectrophotometer.
The mass spectrum (70 eV) was recorded on a liquid chromatography–mass specto-
metry (LC-MS) spectrometer. The solvents and reagents were used without further
purification.
Synthesis of Compound 3
A solution of phenylacetone (45.2 kg, 336.9 mol, 1.0 equiv), pyrrolidine
(42.5 kg, 597.6 mol, 1.8 equiv), and cyclohexane (370 L) was heated to azeotropic
reflux (70–80 ꢀC) for 8 h. Thereafter, cyclohexane and excess pyrrolidine were
removed from the reaction mixture under reduced pressure at below 50 ꢀC to provide
the title compound 3 [the residue was directly used in the next step without further
isolation (63.1 kg; crude yield was considered to be 100%), 89% (GC)].
Synthesis of Compound 4
A solution of chlorobenzaldoxime (61.6 kg, 395.9 mol, 0.8 equiv) and dichlor-
omethane (95 L) was added to a stirring solution of compound 3 (63.0 kg, 469.5 mol,
1.0 equiv), triethylamine (53 kg, 523.8 mol, 1.1 equiv), and dichloromethane (575.0 L)
at 5–10 ꢀC. The temperature of reaction mixture was raised to 25–35 ꢀC, and it was
stirred for 2–3 h. Subsequently, water (360 L) was added to the reaction mixture, and
it was stirred for 30 min. The organic layer was separated and concentrated at
50–55 ꢀC under atmospheric pressure to provide the title compound 4 [residue was

644
A. R. REDDY ET AL.
directly used in next step without further isolation (121.3 kg; crude yield was con-
sidered to be 100%), 70% (GC)].
Synthesis of Compound 5
Concentrated hydrochloric acid (226 L, 2197.2 mol, 5.6 equiv) was added to a
stirring mixture of water (480 L) and compound 4 (121.3 kg, 395.9 mol, 1.0 equiv) at
40–50 ꢀC. The resulting reaction mixture was heated, and the distillate was collected
until the temperature reached to 98–102 ꢀC. The mixture was stirred for 1.5–2.5 h at
98–102 deg;C. Thereafter, the reaction mixture was cooled to 25–35 ꢀC, and the pro-
duct was extracted with dichloromethane (2 ꢁ 220 L). The combined organic layers
were washed with water (145 L) and concentrated under atmospheric pressure at
45–50 ꢀC followed by reduced pressure at 45–50 ꢀC. Isopropyl alcohol (35 L) was
added to the residue and distilled off under vacuum at 45–50 ꢀC. Subsequently, an
additional quantity of isopropyl alcohol (75 L) was added and maintained for
15–30 min at 45–50 ꢀC. The resulting reaction mixture was cooled to 0–5 ꢀC and stir-
red for 45–60 min. The precipitated solid was filtered and washed with chilled iso-
propyl alcohol (3.15 L). The wet compound was charged into isopropyl alcohol
(75 L), heated to 55–60 ꢀC until the clear solution was obtained, cooled to 0–5 ꢀC,
and stirred for 50–60 min. The precipitated solid was filtered, washed with chilled
isopropyl alcohol (3.2 L), and dried under vacuum at 40–45 ꢀC to obtain 24 kg
(60%) of isoxazole derivative 5 with 99.4% purity by HPLC. Mass: m=z 236
(Mþ þ H); IR (KBr, cmꢂ1): 3051 (aromatic CꢂH stretching), 2928 (aliphatic CꢂH
=
stretching), 1619 and 1598 (aromatic C C stretching), 1416 and 1240 (aliphatic
CꢂH bending), 779 and 768 (aromatic CꢂH bending); 1H NMR (200 MHz, CDCl3þ
CD3CN): 2.4 (s, 3H), 7.0–7.5 (m, 10H).
Synthesis of Compound 6
A solution of compound 5 (25 Kg, 106.2 mol, 1.0 equiv) in dichloromethane
(50 L) was added to a stirring solution of chlorosulfonic acid (98 Kg, 841.0 mol,
8.0 equiv) and dichloromethane (75 L) at 0–10 ꢀC. The reaction mixture was heated
to reflux, stirred for 9–11 h, cooled to 25–35 ꢀC, and quenched into chilled water
(175 L) below 10 ꢀC. Thereafter, the temperature of the reaction mixture was raised
to 25–35 ꢀC, and the organic layer was separated. The aqueous layer was extracted
with dichloromethane (2 ꢁ 62.5 L). The combined organic layers were washed with
water (3 ꢁ 190 L) and concentrated under atmospheric pressure below 60 ꢀC. Cyclo-
hexane (250 L) was added and heated to reflux for 15–30 min. Water (75 L) was
added to the reaction mixture and again heated to reflux for 15–30 min. The organic
layer was separated, cooled to 25–35 ꢀC, and stirred for 45 min. The precipitated solid
was filtered and washed with cyclohexane (22 L). Recrystallization employing cyclo-
hexane was repeated twice and dried under vacuum at 50–55 ꢀC to obtain 21.3 kg
(60%) of compound 6 with 98.7% purity by HPLC. Mass: m=z 334 (Mþ þ H); IR
(KBr, cmꢂ1): 3091 and 3064 (aromatic CꢂH stretching), 1626 (C N stretching),
=
=
1591 and 1492 (aromatic C C stretching), 1464 and 1396 (aliphatic CꢂH bending),
= =
1383 and 1191 (O S O asymmetric and symmetric stretching), 782 and 755
1
(aromatic CꢂH bending); H NMR (200 MHz, CDCl3 þ CD3CN): 2.53 (s, 3H),

SYNTHESIS OF VALDECOXIB
645
7.2–7.6 (m, 5H), 7.45 (d, J 8.6 Hz, 2H), 8.03 (d, J 8.6 Hz, 2H); 13C NMR (200 MHz,
DMSO-d6): 11.3, 114.9, 126.0, 128.1, 128.7, 128.8, 129.0, 129.7, 130.0, 147.2, 160.6,
167.1.
Synthesis of Valdecoxib 1
Charcoal (1 kg) was added to a stirring solution of compound 6 (20 kg,
59.9 mol, 1.0 equiv) in dichloromethane (120 L) and stirred for 30–45 min. There-
after, the charcoal was filtered and washed with dichloromethane (40 L). To the
combined filtrate was added 15% aqueous ammonia (90 L, 794.1 mol, 13.2 equiv)
at 20–30 ꢀC for 1–1.5 h. Subsequently, dichloromethane was distilled off from the
reaction mixture below 45 ꢀC under atmospheric pressure, cooled to 5–10 ꢀC, and
stirred for 30–45 min. The precipitated solid was filtered and washed with water
(20 L). The wet compound was charged into water (100 L) and stirred for 45–
60 min and thereafter the solid was filtered and washed with water (20 L). The wet
compound was dried under vacuum at 80–85 ꢀC to afford the 15 kg (80%) of title
compound 1 with 99.92% purity by HPLC. Mass: m=z 315 (Mþ þ H); IR (KBr,
cmꢂ1): 3378 and 3251 (NꢂH stretching), 2927 (aliphatic CꢂH stretching), 1622
=
=
(C N stretching), 1595 and 1564 (aromatic C C stretching), 1466 and 1392
= =
(aliphatic CꢂH bending), 1334 and 1151 (O S O asymmetric and symmetric
1
stretching), 785 (aromatic CꢂH bending); H NMR (400 MHz, DMSO-d6): 2.47
(s, 3H), 7.3–7.5 (m, 5H), 7.41 (d, J 8.2 Hz, 2H), 7.84 (d, J 8.2 Hz, 2H); 13C NMR
(400 MHz, DMSO-d6): 11.3, 114.2, 126.1, 128.2, 128.4, 128.8, 129.7, 130.0, 133.3,
143.3, 160.7, 167.6; Anal. calcd. for C16H14N2O3S: C, 61.13; H, 4.49; N, 8.92; S,
10.19. Found: C, 60.89; H, 4.35; N, 8.98; S, 10.21.
Synthesis of Impurity 9
A solution of 3 (100 g, 0.745 mol, 1.0 equiv) and pyrrolidine (88 g, 1.239 mol,
1.7 equiv) in cyclohexane (740 mL) was heated for azeotropic reflux until water
collection stoped. The resulting reaction mass was concentrated under reduced press-
ure below 70 ꢀC. Dichloromethane (1000 mL) and triethylamine (116 g, 1.146 mol)
were charged at 25–35 ꢀC. To the resulting reaction mixture was added a solution
of chlorobenzaldoxime (125 g, 0.803 mol) in dichloromethane (200 mL) and stirred
at 25–35 ꢀC for 3 h. Subsequently, the reaction mixture was quenched with water
(750 mL), and layers were separated. The product was extracted with dichloro-
methane (200 mL) from the aqueous layer. The combined organic layers were
concentrated under reduced pressure below 55 ꢀC. Thereafter, concentrated hydro-
chloric acid (350 mL) and water (700 mL) were charged to the reaction mixture,
which was heated to 85–90 ꢀC under distillation mode, and the dichloromethane
was collected. The resulting reaction mixture was refluxed for 3 h at 100 ꢀC and
cooled to 25–35 ꢀC. Thereafter, dichloromethane (450 mL) was added to the reaction
mixture at 25–35 ꢀC, and layers were separated. The product was extracted with
dichloromethane (200 mL) from the aqueous phase. The combined organic layers
were concentrated under reduced pressure below 50 ꢀC. After cooling the reaction
mixture at 25–35 ꢀC, isopropyl alcohol (100 mL) was charged, and the solid (com-
pound 5)was filtered. The obtained mother liquor was concentrated below 60 ꢀC

646
A. R. REDDY ET AL.
under reduced pressure, charged with acetone (200 mL) at 25–35 ꢀC, and stirred for
1 h. The solid was filtered and dried at 40–45 ꢀC to obtain 26 g (15%) of compound 9
with 98.5% purity (HPLC). MS: m=z 236 (Mþ þ H); IR (KBr, cmꢂ1): 2919 (aliphatic
1
=
CꢂH stretching), 1598 and 1578 (aromatic C C stretching); H NMR (200 MHz,
CDCl3): 4.1 (s, 2H), 6.3 (s, 1H), 7.2–7.8 (m, 10H).
Synthesis of Impurity 13
A solution of 5 (100 g, 0.425 mol, 1.0 equiv) in dichloromethane (100 mL) was
added to a solution of chlorosulfonic acid (396.2 g, 3.400 mol, 8.0 equiv) in dichlor-
omethane (400 mL) at 0–10 ꢀC. The reaction mixture was refluxed for 7–8 h. The
resulting reaction mixture was quenched with water (1000 mL) below 10 ꢀC. The
organic layers were separated, and the product was extracted from the aqueous layer
with dichloromethane (2 ꢁ 500 mL). The combined organic layers were concentrated
below 45 ꢀC under reduced pressure. Thereafter, cyclohexane (1000 mL) was added
to the residue at 25–35 ꢀC and stirred for complete solid isolation. The precipitated
solid (compound 6) was filtered, and the filtrate was concentrated at below 60 ꢀC
under reduced pressure to obtain 10 (not isolated) along with some quantity of 6
(not quantified). Aqueous ammonia (15%, ꢃ650 mL 5.735 mol, 13.2 equiv) was
added to a solution of 10 (crude) and 6 in dichloromethane (1000 mL) at
25–35 ꢀC. The resulting reaction mixture was stirred for 1 h at 25–35 ꢀC. The layers
were separated, and the product was extracted from the aqueous layer with dichlor-
omethane (2 ꢁ 200 mL). The combined layers were concentrated under reduced
pressure below 45 ꢀC, and dichloromethane (100 mL) was added at 25–35 ꢀC and
thereafter cooled to 10–15 ꢀC. After stirring for 1 h, the precipitated solid (1) was fil-
tered. The filtrate was concentrated, and the crystallization was repeated twice using
dichloromethane to remove 1 completely. The filtrate free from 1 was taken and
concentrated below 45 ꢀC under reduced pressure. Ethyl acetate was added to the
resulting residue, heated to reflux for 30 min followed by cooling to 0–5 ꢀC, and stir-
red for 1 h. The precipitated solid was filtered and dried at 80–85 ꢀC to provide 13.5 g
(10%) of compound 13 with 96.0% purity (HPLC); MS: m=z 315 (Mþ þ H); IR (KBr,
cmꢂ1): 3337 and 3247 (NꢂH stretching), 1334 and 1165 (O S O asymmetric and
= =
1
symmetric stretching); H NMR (200 MHz, CD3CN þ CDCl3): 2.48 (s, 3H), 7.37
(m, 5H), 7.43 (d, J 7.8 Hz, 1H), 7.52 (t, J8.0, 7.8 Hz, 1H), 7.74 (s, 1H), 7.87 (d,
J7.8 Hz, 1H); 13C NMR (400 MHz, DMSO-d6): 11.2, 114.9, 125.1, 126.0, 126.5,
128.0, 128.3, 129.0, 129.2, 130.0, 147.5, 148.8, 160.6, 167.0.
Synthesis of Impurity 14
A solution of 9 (20 g, 0.085 mol, 1.0 equiv) in dichloromethane (50 mL) was
added to a solution of chlorosulfonic acid (79.2 g, 0.680 mol, 8.0 equiv) in dichloro-
methane (50 mL) at 0–5 ꢀC. The reaction mixture was stirred at 40 ꢀC for 3–5 h and
quenched with water (140 mL) below 10 ꢀC. The layers were separated, and the pro-
duct was extracted with dichloromethane (2 ꢁ 50 mL) from the aqueous layer. There-
after, aqueous ammonia solution (15%, 130 mL 1.147 mol, 13.5 equiv) was added to
the organic layer at 25–35 ꢀC and stirred for 1–2 h. The precipitated solid was filtered
and washed with water (40 mL) and dichloromethane (20 mL) successively. The wet

SYNTHESIS OF VALDECOXIB
647
solid was dried at 80–85 ꢀC to afford 16 g (60%) of compound 14 with 99.5%
(HPLC); MS: m=z 315 (Mþ þ H); IR (KBr, cmꢂ1): 3363 and 3271 (NꢂH stretching),
1
1342 and 1158 (O S O asymmetric and symmetric stretching); H NMR (200 MHz,
= =
CDCl3): d 4.27 (s, 2H), 6.33 (s, 1H), 7.20–7.80 (m, 5H), 7.52 (d, J ¼ 8.4 Hz, 2H), 8.02
(d, J ¼ 8.4 Hz, 2H); 13C NMR (400 MHz, DMSO-d6): 31.8, 100.4, 126.0, 126.5,
128.5, 129.0, 129.2, 130.1, 140.3, 142.7, 161.9, 171.7.
Synthesis of Impurity 15
A solution of 5 (20 g, 0.085 mol, 1.0 equiv) in dichloromethane (40 mL) was
added to a stirring solution of chlorosulfonic acid (347 g, 2.978 mol, 35.0 equiv)
and dichloromethane (400 mL) at 0–10 ꢀC. The reaction mixture was refluxed for
7–8 h. The resulting reaction mixture was quenched with water (900 mL) below
10 ꢀC. The layers were separated, and the product was extracted from the aqueous
layer with dichloromethane (2 ꢁ 100 mL). Aqueous ammonia (15%, 260 mL,
2.294 mol, 27.0 equiv) was added to the combined organic layer at 25–35 ꢀC and stir-
red for 1 h. The precipitated solid was filtered and dried at 80–85 ꢀC to provide 23.5 g
(70%) of compound 15. Purity by HPLC 97.0%; MS: m=z 394 (Mþ þ H); IR (KBr,
cmꢂ1): 3322 and 3241 (NꢂH stretching), 1346 and 1161 (O S O asymmetric and
= =
1
symmetric stretching); H NMR (400 MHz, DMSO-d6): 2.45 (s, 3H), 7.36 (d, J
8.4 Hz, 1H), 7.45 (d, J 8.4 Hz, 2H), 7.58 (t, J 8.0, 7.6 Hz, 1H), 7.86 (d, J 8.4 Hz,
2H), 7.90 (d, J 8.0 Hz, 1H), 8.05 (s, 1H); 13C NMR (200 MHz, DMSO-d6): 11.4,
114.2, 125.1, 126.1, 126.8, 129.0, 129.4, 129.9, 131.2, 132.6, 143.4, 144.8, 159.6, 168.1.
Synthesis of Impurity 16
A mixture of 6 (20 g, 0.060 mol, 1.0 equiv), water (65 mL), and THF(280 mL)
was heated to reflux and maintained for 24 h. The solvent was distilled off completely
from the reaction mixture below 80 ꢀC under reduced pressure. The reaction mixture
was cooled to 25–35 ꢀC, charged with toluene (200 mL), and stirred for 1 h. The solid
was filtered and dried at 50–55 ꢀC to obtain 18.5 g (98.4%) of compound 16. Purity
by HPLC 99.7%; MS: m=z 316 (Mþ þ H); IR (KBr, cmꢂ1): 3063 (OꢂH stretching),
1
1396 and 1176 (O S O asymmetric and symmetric stretching); H NMR (400 MHz,
= =
DMSO-d6): d 2.44 (s, 3H), 7.19 (d, J ¼ 8.0 Hz, 2H), 7.35–7.45 (m, 5H), 7.64 (d,
J ¼ 8.0 Hz, 2H); 13C NMR (400 MHz, DMSO-d6): 11.3, 114.9, 126.0, 128.1, 128.7,
128.8, 129.0, 129.6, 129.9, 147.4, 160.6, 167.1.
Synthesis of Impurity 17
A mixture of 1 (10 g, 0.032 mol, 1.0 equiv), 6 (16 g, 0.048 mol, 1.5 equiv), and
pyridine (50 mL) was heated to 110–115 ꢀC and stirred for 9 h. The reaction mixture
was concentrated under reduced pressure at 95 ꢀC. The resulted solid was recrystal-
lized from isopropyl alcohol (300 mL), and the wet compound was again recrystal-
lized from the mixture of water (100 mL) and acetone (100 mL). The wet solid was
dried at 80–85 ꢀC to afford 12.5 g (64%) of compound 17 with 98.0% purity (HPLC);
MS: m=z 612 (Mþ þ H); IR (KBr, cmꢂ1): 3339 (NꢂH, stretching), 1369 and
1165 (O S O asymmetric and symmetric stretching); 1H NMR (400 MHz,
= =

648
A. R. REDDY ET AL.
CDCl3 þDMSO-d6): d 2.5 (s, 6H), 7.32 (d, J ¼ 8.2 Hz, 4H), 7.36–7.45 (m, 10H), 7.95
(d, J ¼ 8.2 Hz, 4H); 13C NMR (400 MHz, DMSO-d6): 11.3, 114.5, 126.5, 128.1,
128.5, 128.8, 129.0, 129.7, 131.5, 145.5, 160.7, 167.3.
ACKNOWLEDGMENTS
The authors thank the management of Dr. Reddy’s Laboratories Ltd. for sup-
porting this work. Cooperation from colleagues in the Research and Development
Department and Analytical Research and Development Department is also
acknowledged.
REFERENCES
1. Vane, J. R. Inhibition of prostaglandinsynthesis as a mechanism of action for aspirin-like
drugs. Nature (New Biol.) 1971, 231, 232–235.
2. Lewis, P. J.; Dollery, C. T. Clinical pharmacology and potential of prostacyclin. Br. Med.
Bull. 1983, 39, 281–284.
3. Fitzgerald, G. A.; Patrono, C. The coxibs, selective inhibitors of cyclooxygenase-2. New
Engl. J. Med. 2001, 345, 433–442.
4. Talley, J. J.; Nagarajan, S.; Carter, J. S.; Weier, R. M.; Stealey, M. A.; Collins, P. W.;
Rogers, R. S.; Seibert, K. Substituted isoxazoles for the treatment of inflammation.
U.S. Patent 5,633,272, 1997.
5. Talley, J. J.; Brown, D. L.; Carter, J. S.; Graneto, M. J.; Koboldt, C. M.; Masferrer, J. L.;
Perkins, W. E.; Rogers, R. S.; Shaffer, A. F.; Zhang, Y. Y.; Zweifel, B. S.; Seibert, K.
4-[5-Methyl-3-phenylisoxazol-4-yl]-bezenesulfonamide, valdecoxib: A potent and selective
inhibitor of COX-2. J. Med. Chem. 2000, 43, 775–777.
6. Nunno, L. D.; Vitale, P.; Scilimati, A.; Tacconelli, S.; Patrignani, P. Novel synthesis of
3,4-diarylisoxazole analogues of valdecoxib: Reversal cyclooxygenase-2 selectivity by sul-
fonamide group removal. J. Med. Chem. 2004, 47, 4881–4890.
7. Reddy, M. V. R.; Bell, S. C. Process for the preparation of substituted isoxazoles and
2-isoxazolines. U.S. Patent 6,809,204 B2, 2004.
8. Waldo, J. P.; Larock, R. C. The synthesis of highly substituted isoxazoles by electrophilic
cyclization: An efficient synthesis of valdecoxib. J. Org. Chem. 2007, 72, 9643–9647.
9. Kochetkov, N. K.; Sokolov, S. D. Advances in Heterocyclic Chemistry; Academic Press:
New York, 1963; Vol. 2, pp. 365–421.
10. (a) Lokanatha Rai, K. M. Hassner, A. Intermolecular 1,3-dipolar cycloaddition of nitrile
oxide with vinyl acetate and acrylonitrile. Ind. J. Chem. 1997, 36B, 242; (b) Grundmann,
C.; Grunanger, P. In The Nitrile Oxide; Springer Verlag: New York, 1971.
11. Just, G. Dahl, L. Lead tetraacetate oxidation of aldoximes. Tetrahedron 1968, 24,
5251–5269.
12. (a) Peake, C. J.; Strickland, J. H. A simple regioselective synthesis of hydroximinoyl chlor-
ides (nitrile oxide precursors). Synth. Commun. 1986, 16, 763–765; (b) Kozikowski, A. P.;
Cheng, X. M. Use of a thermally stable, optically active nitrile oxide in the synthesis of a
lyngbyatoxin A intermediate. Tetrahedron Lett. 1987, 28, 3189–3192; (c) Grundmann, C.;
Dean, J. M. Nitrile oxides, V: Stable aromatic nitrile oxides. J. Org. Chem. 1965, 30,
2809–2812.
13. Liu, K.; Shelton, B. R.; Howe, R. K. A particularly convenient preparation of benzohy-
droximinoyl chlorides (nitrile oxide precursors). J. Org. Chem. 1980, 45, 3916–3918.

SYNTHESIS OF VALDECOXIB
649
14. Hassner, A.; Lokanatha Rai, K. M. A new method for the generation of nitrile oxides and
its application to the synthesis of 2-isoxazolines. Synthesis 1989, 1, 57–59.
15. Radhakrishna, A. S.; Sivaprakash, K.; Singh, B. B. Iodobenzene dichloride—An efficient
reagent for preparation of nitrile oxides from aldoximes. Synth. Commun. 1991, 21,
1625–1629.
16. Kim, J. N.; Ryu, E. K. A convenient synthesis of nitrile oxides from aldoximes by
1-chlorobenzotriazole. Synth. Commun. 1990, 20, 1373–1377.
17. Rai, K. M. L.; Linganna, N.; Hassner, A.; Murthy, C. A. Mercuric acetate as synthetic
reagent for generation of highly reactive nitrile oxide intermediate. Org. Prep. Proc. Int.
1992, 24, 91–93.
18. Mukaiyama, T.; Hoshino, T. The reactions of primary nitroparaffins with isocyanates.
J. Am. Chem. Soc. 1960, 82, 5339–5342.
19. McMurry, J. E. A general synthesis of 4-isoxazolecarboxylic esters: Ethyl 3-ethyl-5-
methyl-4-isoxazolecarboxylate. Org. Synth. 1988, 6, 592–594.
20. Mosher, M. D.; Natale, N. R. J. The preparation of intercalating isoxazoles via a nitrile
oxide cycloaddition. Heterocycl. Chem. 1995, 32, 779–781.
21. Hansen, E. C.; Levent, M.; Connolly, T. Safe and scaleable oxidation of benzaldoximes to
benzohydroximinoyl chlorides. Org. Process Res. Dev. 2010, 14, 574–578.