Process development for sodelglitazar: A ppar panagonist

Article abstract of DOI:10.1021/op8002294

Three efficient syntheses of sodelglitazar (1) have been developed. In particular, the third synthesis avoids the use of zinc and eliminates the resulting heavy metal waste stream as well as the potential genotoxic methanesulfonate in the two earlier syntheses. This process produces sodelglitazar in 74% overall yield from readily available thiophenol (8) and thiazole alcohol (3).

Full text of DOI:10.1021/op8002294

Organic Process Research & Development 2009, 13, 297–302
Process Development for Sodelglitazar: A PPAR Panagonist
Andrew D. Brown, Roman D. Davis, Russ N. Fitzgerald, Bobby N. Glover, Kim A. Harvey, Lynda A. Jones, Bing Liu,*
Daniel E. Patterson, and Matthew J. Sharp
Chemical DeVelopment, GlaxoSmithKline, 5 Moore DriVe, Research Triangle Park, North Carolina 27709, U.S.A.
Scheme 1. Retrosynthetic analysis of sodelglitazar
Abstract:
Three efficient syntheses of sodelglitazar (1) have been developed.
In particular, the third synthesis avoids the use of zinc and
eliminates the resulting heavy metal waste stream as well as the
potential genotoxic methanesulfonate in the two earlier syntheses.
This process produces sodelglitazar in 74% overall yield from
readily available thiophenol (8) and thiazole alcohol (3).
Introduction
Sodelglitazar (1) is a panagonist active towards all three
peroxisome proliferator-activated receptors (PPAR).1 Sodelgli-
tazar (1) had been in the phase II clinical development for the
treatment of type 2 diabetes and metabolic syndrome. To supply
drug substance for the development studies, a robust and
efficient synthesis of sodelglitazar (1) was needed. Herein, we
detail our efforts on the process development for sodelglitazar
(1).
Retrosynthetic analysis of sodelglitazar (1) suggested that
disconnections of ether and thioether are the most accessible
(Scheme 1). Our first approach to sodelglitazar (1), a slight
modification of the medicinal chemistry routes,1a was to make
the ether linkage first followed by thioether formation through
coupling of thiophenol 2 with thiazole alcohol 3 (pathway a).
Our second approach to sodelglitazar was to construct the ether
linkage at the last stage of synthesis from phenol 4 (pathway
b). The phenol 4 was prepared from substitution of thiazole
alcohol methanesulfonate 5 by lithium thiolate 6 that was
obtained from aryl lithium 7 and elemental sulfur.2 Our third
approach to sodelglitazar was a slight variation of the second
synthesis. The phenol 4 was prepared from the direct coupling
of thiazole alcohol 3 with thiophenol 8 (pathway c).
Scheme 2. Preparation of thiazole alcohol 3
Synthesis of Starting Intermediate (3). For all syntheses
under consideration, the thiazole alcohol 3 was a key intermedi-
ate. This was efficiently prepared from readily available
benzonitrile 9 (Scheme 2). Reaction of benzonitrile 9 with
sodium hydrogen sulfide in methanol at 50 °C gave thioamide
10 in 32% yield along with 19% of methoxy substituted
thioamide 11. Column chromatography was required for
purification as recrystallization failed to provide the clean
thioamide 10. To eliminate the substitution reaction, we screened
different thionation agents such as NaHS in different solvents,
P2S5, thioacetamide, and thioacetic acid,3 and found that
thioacetic acid as a thionation agent worked best to convert
nitrile into thioamide without the undesired substitution reaction.
Thus, treatment of benzonitrile 9 with thioacetic acid and boron
trifluoride diethyl etherate followed by ammonia gave thioamide
10 in greater than 95% conversion. After aqueous workup and
azeotropic distillation to remove water, the crude thioamide 10
* Author for correspondence. E-mail: bing.b.liu@gsk.com.
(1) (a) Cadilla, R.; Gosmini, R. L. M.; Lambert, M. H., III; Sierra, M. L.
PCT WO 2002062774, 2002. (b) Evans, J. L.; Lin, J. J.; Goldfine, I. D.
Curr. Diabetes ReV. 2005, 1, 299–307.
(2) (a) Ingold, K.; Liu, B. PCT WO 2003074504, 2003. (b) Ko, J.; Ham,
J.; Yang, I.; Chin, J.; Nam, S.-J.; Kang, H. Tetrahedron Lett. 2006, 47,
7101–7106.
(3) Gauthier, J. Y.; Lebel, H. Phosphorus Sulfur 1994, 95 and 96, 325–
326.
10.1021/op8002294 CCC: $40.75
Published on Web 12/04/2008
2009 American Chemical Society
Vol. 13, No. 2, 2009 / Organic Process Research & Development
•
297

Scheme 3. First synthesis of 1
Table 1. Sulfenation of organolithium 7
run
equiv of 15:n-BuLi:S
ratio of 17:18a
1
2
3
4
1:1:0.9
1:1:0.9 (inverse addition)
1:1:1.2
0.4:99.6
3:97
reacted with 2-chloroacetoacetate in toluene at 100 °C to give
thiazole ester 12 in 70% yield in one pot. Ester 12 was reduced
with LiAlH4 to give thiazole alcohol 3 in 91% yield.
23:77
1:1:1.2 followed by 0.2 equiv of n-BuLi 17 not detected
First Synthesis. In our first synthesis of 1, we required
thiophenol 2. This was prepared through alkylation of 2-me-
thylphenol with 2-bromo-2-methylpropanoate in the presence
of t-BuOK in THF followed by sulfonation with chlorosulfonic
acid (Scheme 3). Subsequent reaction with thionyl chloride gave
sulfonyl chloride 13 in 74% yield, which was reduced using
zinc and TMSCl in ethyl acetate with acetic acid and water to
provide thiophenol 2. ZnCl2 generated in situ from zinc and
TMSCl facilitated the coupling of thiophenol 2 with thiazole
alcohol 3 followed by saponification to give 1 in 78% yield in
one pot on the multigram scale.4 The scalability of this process
was of concern due to the use of zinc, which tended to settle
on the bottom of the reactor, causing agitation issues on scale.
In addition, generation of a heavy metal waste stream was not
desirable. The one-pot process also lacked an isolation step prior
to the formation of 1, considered important for control of the
final drug substance quality. To address these issues, we
designed the second approach to sodelglitazar (1) (Scheme 1).
Second Synthesis. Our second synthesis of 1 started with
the readily available 4-bromo-2-methylphenol (14). Silylation
of 14 with TBSCl and DMAP in toluene gave silyl ether 15 in
quantitative yield. The crude bromide 15 in tert-butyl methyl
ether (TBME) at -20 °C was treated with n-BuLi to generate
aryllithium 7 that further reacted with elemental sulfur to form
the lithium arylthiolate intermediate 6. The lithium bromine
exchange reaction of bromide 14 was unsuccessful without
protection of the phenol as its tert-butyldimethylsilyl ether. In
one scale-up run of the coupling, the sulfenation of aryllithium
7 was incomplete as treatment of an aliquot of the reaction
mixture with MeI generated a significant amount of disulfide
17. Mechanistically, the disulfide 16 is the intermediate en route
to the sulfide 6 in the sulfenation reaction. The high residual
disulfide observed was a result of a probable process error
caused by a sulfur overcharge.
a The ratio was determined by HPLC.
Scheme 4. Second synthesis of 1
the sulfenation reaction, the order of addition has no major
impact (Table 1). The normal addition of sulfur to the aryl
lithium mixture (run 1) left an insignificant amount of disulfide.
The inverse addition, adding aryl lithium to a slurry of elemental
sulfur in TBME, led to a slight increase in the amount of
disulfide. However, the sulfenation process was highly depend-
ent upon the stoichiometry of organolithium 7 and elemental
sulfur in the reaction mixture. Excess sulfur resulted in
incomplete conversion of disulfide 16 to sulfide 6, generating
23% of disulfide 17 upon derivatization with MeI (run 3).
Gratifyingly, the addition of additional n-BuLi brought about
complete conversion of disulfide 16 to sulfide 6 (run 4).
Once the sulfenation reaction was complete as determined
by MeI derivatization, the lithium arylthiolate intermediate 6
was treated with thiazole alcohol methanesulfonate 5 prepared
from thiazole alcohol 3 and methanesulfonic anhydride to give
sulfide 19 in 74% yield (Scheme 4). Desilylation and alkylation
were carried out in one pot. Introduction of the isobutyric acid
moiety was first achieved with ethyl 2-bromo-2-methylpro-
To better understand the sulfenation reaction, we studied the
effects of order of addition and the stoichiometry of organo-
lithium 7 to sulfur. As measured by the amount of disulfide in
(4) Martin, M. T.; Thomas, A. M.; York, D. G. Tetrahedron Lett. 2002,
43, 2145–2147.
298
•
Vol. 13, No. 2, 2009 / Organic Process Research & Development

Scheme 5. Third synthesis of 1
panoate. However, a significant amount of methacrylate formed,
and its polymerization presented a scale-up issue from a quality
perspective since polymer coating of reactors was observed.
The alternative method to make R-phenoxyisobutyric acid was
the Bargellini reaction that involves the condensation of phenol
with chloroform and acetone in the presence of sodium
hydroxide.5 Thus, the Bargellini reaction using 1,1,1-trichloro-
2-methyl-2-propanol, the condensation product of CHCl3 and
acetone, introduced the isobutyric acid segment, producing
sodelglitazar 1 in 65% yield. The use of the Bargellini reaction
avoided the formation of methacrylate and therefore the
polymerization issue. The second synthesis of sodelglitazar was
efficient and had been scaled up in 50-L reactors. However,
the process still had several issues that made it less than ideal
as a manufacturing route. The route involved the use of
potentially genotoxic methanesulfonate 5 which needed to be
controlled at low levels in the drug substance. In addition, the
sulfenation of organolithium 7 with elemental sulfur required
careful control to minimize the residual disulfide 16. Finally,
the Bargellini reaction was highly exothermic and involved the
use of polyhalogenated compounds that are a safety concern.
For these reasons, an alternative synthesis to address these issues
was desired.
Commercial Synthesis. Fortunately, during our work on
sodelglitazar, thiophenol 8 became available as a potential
starting material, and thus we opted to develop the third process
for sodelglitazar from 8 (Scheme 5). Selective coupling of
thiophenol 8 with thiazole alcohol 3 was achieved successfully
using strong acids such as trifluoroacetic acid in acetonitrile.
Strong inorganic acids such as sulfuric acid and sulfonic acids
such as methanesulfonic acid were less selective, leading to the
formation of ether 20 from the homocoupling of thiazole alcohol
3, and ether 21 from the further reaction of phenol 4 with
thiazole alcohol 3 (Figure 1). During the coupling of thiazole
alcohol 3 and thiophenol 8 in the mixture of trifluoroacetic acid
and acetonitrile, a transient intermediate determined by LC/
MS to be thiazole alcohol trifluoroacetate formed first and then
Figure 1. Structures of process impurities.
reacted with thiophenol 8 to give phenol 4. When the reaction
was complete, the product phenol 4 was precipitated from the
reaction mixture by addition of water.
Alkylation of phenol 4 was carried out under heterogeneous
conditions in the presence of sodium hydroxide using 2-bromo-
2-methylpropanoic acid instead of the Bargellini reaction
described in the second process (Scheme 4). The alkylation
using 2-bromo-2-methylpropanoic acid is less exothermic than
the Bargellini reaction.6 We next studied the process in more
detail to get the process ready for manufacturing scale. Screens
of different carbonate bases, hydroxides, and organic bases
confirmed solid sodium hydroxide as the optimal base for the
alkylation. No alkylation occurred with carbonate bases or
organic bases such as DBU as decomposition of 2-bromo-2-
methylpropanoic acid appeared to be the major pathway.
Soluble bases such as t-BuOLi and Me3SiOK failed to bring
the alkylation to completion. Attempts to run the alkylation in
the presence of 50% NaOH under phase transfer conditions
proved fruitless. 2-Butanone (MEK) was the preferred solvent
for the alkylation reaction over toluene and ether solvents such
as THF, DME, and 1,4-dioxane. Alcoholic solvents were
unsuitable for the alkylation as the fluorine in the phenol was
susceptible for substitution by alkoxide. Adequate agitation was
essential for the heterogeneous reaction to go to completion as
solid NaOH agglomerated as the reaction progressed. Different
forms of solid NaOH such as pellets, flakes, and beads had no
impact on the alkylation outcome once adequate agitation was
applied. After the reaction was complete, a water quench of
the reaction mixture and a 2-h hold at 50 °C destroyed the
excess 2-bromo-2-methylpropanoic acid to less than 10 ppm
that was the control specification for 2-bromo-2-methylpro-
panoic acid in the final drug substance. Water wash of the MEK
suspension removed salts and excess hydroxide. The sodium
salt of the product remained in the organic layer during the
(5) (a) Bargellini, G. Gazz. Chim. Ital. 1906, 36, 329–339. For a recent
example, see: (b) Bose, A. K.; Manhas, M. S.; Ganguly, S. N.; Pednekar,
S.; Mandadi, A. Tetrahedron Lett. 2005, 46, 3011–3013. (c) Caution!
The mixture of chloroform and acetone in the presence of sodium
hydroxide has been reported as explosive and an exothermic hazard.
(6) Davis, R. D.; Fitzgerald, R. N.; Guo, J. Synthesis 2004, 1959–1962.
Vol. 13, No. 2, 2009 / Organic Process Research & Development
•
299

water wash. Acidification of the organic layer followed by
solvent exchange to heptane gave 1 in 75% yield.
at 40-50 °C, and water (200 mL) was added. Isooctane (200
mL) was added, and the layers were allowed to separate. The
organic phase was extracted with 1 N NaOH (200 mL) to
remove excess 2-methylphenol. The organic phase was dried
over Na2SO4 and concentrated at 40-50 °C to one-quarter
volume.
There was about 0.1% of methoxy impurity 22 detected in
the isolated drug substance (Figure 1). From a quality perspec-
tive, we needed to understand the origin of this impurity as no
methanol source was employed in the synthesis. The starting
phenol was confirmed to be free from any methoxy containing
contaminants. Analysis of the solvent MEK showed there were
43 ppm of methanol and 90 ppm of methyl propanoate in MEK.
It was interesting to note that treatment of MEK with solid
NaOH at 50 °C for 2 h increased the amount of methanol to
56 ppm and the amount of methyl propanoate to 1300 ppm. It
should be noted that the amount of methyl propanoate did not
increase significantly with extended reaction time in the open
air. Methyl propanoate could be hydrolyzed under the alkylation
conditions to methanol that caused the formation of methoxy-
substituted impurity 22 observed in the isolated product.
Alkylation in acetone-1,3-13C2 (C-13 methanol not present to
any extent) formed 13C-labeled methoxy impurity 23. This
suggested that methanol was derived from the hydrolysis of
methyl ester formed from 2-butanone through a Baeyer-Villiger
oxidation.7
The third synthesis of sodelglitazar (1) was a two-step
synthesis that eliminated potential genotoxic intermediate sul-
fonate 5 and the potentially nonrobust sulfenation reaction in
the second synthesis. This synthetic route was selected as the
manufacturing route for sodelglitazar.
In summary, we have developed three efficient syn-
theses of sodelglitazar. The third synthesis avoids the use
of zinc and eliminates the resulting heavy metal waste
stream as well as the potential genotoxic methanesulfonate
in the two earlier syntheses. The process produces
sodelglitazar in 74% overall yield from readily available
thiophenol 8 and thiazole alcohol 3.
The 2-methyl-2-(o-tolyloxy)propionic acid ethyl ester in
isooctane (from above) was dissolved in dichloromethane (330
mL), stirred, and cooled to -5 °C. Neat chlorosulfonic acid
(32.3 g, 0.28 mol) was added dropwise over 15-20 min,
controlling the temperature below 0 °C. The mixture was stirred
at -5 °C for 60 min. N,N-Dimethylformamide (57.9 g, 0.79
mol) was added over 15 min, while controlling the temperature
below 10 °C. The solution was cooled to -5 °C, and thionyl
chloride (35.3 g, 0.30 mol) was added over 5 min, with no
temperature rise. The reaction was allowed to warm to ambient
and to stir at ambient for 1.5 h. The mixture was cooled to 0
°C, and H2O (330 mL) was added. The layers were separated,
and the organic (lower) phase was washed with 0.1 N HCl (2
× 330 mL). The organic phase was dried over Na2SO4. Solvent
was exchanged at 40-50 °C to isooctane until distillation had
ceased. Upon cooling to ambient, a solid precipitated. The solid
was collected by filtration, washed with isooctane (25 mL), and
dried under vacuum at 25 °C to constant weight to give 60.0 g
of 2-(4-chlorosulfonyl-2-methylphenoxy)-2-methylpropionic acid
ethyl ester (13) as an amber solid in 74% yield over two steps.
1H NMR (300 MHz, CDCl3): δ 7.85 (d, 1 H, J ) 2.4 Hz),
7.78 (dd, 1 H, J ) 8.9, 2.4 Hz), 6.70 (d, 1 H, J ) 8.9 Hz), 4.26
(q, 2 H, J ) 7.1 Hz), 2.34 (s, 3 H), 1.73 (s, 6 H), 1.25 (t, 3 H,
J ) 7.1 Hz). HPLC purity: 96.3% AUC.
2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-5-thia-
zolecarboxylic Acid Ethyl Ester (12). To a solution of
thioacetic acid (320 mL, 4.48 mol) and boron trifluoride diethyl
etherate (550 mL, 4.34 mol) in toluene (1.5 L), was added
slowly a solution of 2-fluoro-4-trifluoromethylbenzonitrile (9)
(303 g, 1.6 mol) in toluene (300 mL) over 60 min at 20 °C.
After the addition was complete, the mixture was allowed to
stir for 3 h. The mixture was cooled to 5 °C, and water (300
mL) was added over 30 min to quench boron trifluoride. Once
the quenching was complete, water (900 mL) was added to
dilute the reaction mixture. The aqueous layer was separated.
To the organic layer was added 10% ammonia solution (1.2 L)
over 30 min while maintaining the temperature below 5 °C.
The mixture was heated to 20 °C where it was maintained for
an additional 30 min. The aqueous layer (basic) was separated.
The organic layer was washed with water (2 × 1.2 L) and
concentrated under reduced pressure to ∼900 mL. To the
mixture was added ethyl 2-chloroacetoacetate (241 mL, 1.74
mol), and the mixture was heated at 100 °C until the reaction
was complete (∼16 h). The reaction mixture was cooled to 50
°C, and toluene was removed under reduced pressure. Ethanol
(1.5 L) was added followed by adding water (0.6 vol). The
mixture was allowed to cool to ambient over 1 h and maintained
at ambient for 1 h. The solid was collected by filtration, washed
with cold 90% ethanol (300 mL), and dried under vacuum at
40 °C to a constant weight to give 370 g of thiazole ester 12 as
Experimental Section
1H NMR spectra were obtained with 300 or 400 MHz
instruments. All reagents and solvents were purchased from
commercial sources and were used without further purification.
HPLC purity was determined on a Hewlett Packard series 1100
system using Agilent Eclipse XDB C18 columns (150 mm ×
4.6 mm, 3.5 µm), and a mixture of water and acetonitrile as
mobile phase (gradient at a flow rate of 1.0 mL/min and UV
detector at 220 nm). Mass spectral data were collected on an
Agilent quadrupole ion trap mass spectrometer using electron-
spray ionization (ESI) in positive ion mode.
2-(4-Chlorosulfonyl-2-methylphenoxy)-2-methylpropion-
ic Acid Ethyl Ester (13). A solution of 1 M t-BuOK (325 mL,
0.325 mol) was stirred and cooled to between -10 and 0 °C.
2-Cresol (35.1 g, 0.325 mol) in THF (70 mL) was added
dropwise over 10 min, keeping the temperature below 0 °C.
The main exotherm occurred during the first half of addition.
The mixture was then cooled to -5 °C, and ethyl 2-bromo-2-
methylpropanoate (50.7 g, 0.28 mol) was added at once. The
mixture was warmed to 55-65 °C. During the warm-up, a dark
pink/magenta color formed. The mixture was stirred at 62-65
°C for 50 min. The mixture was concentrated to half-volume
1
a white solid in 70% yield. H NMR (300 MHz, CDCl3): δ
(7) Fermin, M. C.; Bruno, J. W. Tetrahedron Lett. 1993, 34, 7545–7548.
8.49 (t, 1 H, J ) 7.7 Hz), 7.54 (d, 1 H, J ) 8.5 Hz), 7.50 (d,
300
•
Vol. 13, No. 2, 2009 / Organic Process Research & Development

1 H, J ) 12.1 Hz), 4.39 (q, 2 H, J ) 7.1 Hz), 2.82 (s, 3 H),
1.43 (q, 3 H, J ) 7.1 Hz). HPLC purity: 96.9% AUC. MS:
MH+ ) 334.
to cool to room temperature. The solid was collected by
filtration, washed with ethanol/water (2:1, 40 mL), and dried
under vacuum at 60 °C to a constant weight to give 26.9 g of
1 as a slightly yellow solid in 78% yield. 1H NMR (400 MHz,
DMSO-d6): δ 13.02 (br s, 1 H), 8.32 (t, 1 H, J ) 7.9 Hz), 7.86
(d, 1 H, J ) 10.4 Hz), 7.67 (d, 1 H, J ) 8.4 Hz), 7.20 (d, 1 H,
J ) 1.8 Hz), 7.10 (dd, 1 H, J ) 8.4, 2.2 Hz), 6.60 (d, 1 H, J )
8.4 Hz), 4.35 (s, 2 H), 2.20 (s, 3 H), 2.07 (s, 3 H), 1.46 (s, 6
H). HPLC purity: 95.6% AUC. MS: MH+ ) 500.
4-tert-Butyldimethylsilyloxy-3-methylbromobenzene (15).
To a solution of 4-bromo-2-methylphenol (149.6 g, 0.80 mol),
4-N,N-dimethylaminopyridine (102.6 g, 0.84 mol) and toluene
(450 mL) was added a solution of tert-butyldimethylsilyl
chloride (126.6 g, 0.84 mol) in toluene (280 mL), at such a
rate that the temperature was maintained between 20 and 25
°C (about 20 min). The resulting slurry was stirred at ambient
overnight, then quenched with water (115 mL). The organic
layer was washed with 1 N HCl (100 mL) followed by 2 N
NaOH (100 mL). The organic layer was concentrated under
vacuum to give 241 g of silyl ether 15 as a colorless oil in
quantitative yield. 1H NMR (300 MHz, DMSO-d6): δ ) 7.35
(s, 1 H), 7.24 (d, 1 H, J ) 8.4 Hz), 6.75 (d, 1 H, J ) 8.4 Hz),
2.30 (s, 3 H), 0.98 (s, 9 H), 0.20 (s, 6 H).
5-[[[4-[[(1,1-Dimethylethyl)dimethylsilyl]oxy]-3-methyl-
phenyl]thio]methyl]-2-[2-fluoro-4-(trifluoromethyl)phenyl]-
4-methylthiazole (19). A solution of silyl ether 15 (3.16 kg,
10.5 mol) and TBME (12.6 kg) was stirred and cooled to -20
°C. A solution of 2 M n-BuLi in cyclohexane (4.70 kg, 12.7
mol) was added over 15 min. The solution was allowed to warm
to 0 °C for 1 h. The clear pale yellow solution was cooled to
-10 °C, and sulfur (0.335 kg, 10.5 mol) was added in small
portions to maintain the temperature at -10 to -5 °C. The
solution containing lithium thiolate 6 was stirred at -10 °C for
1 h and held there for subsequent use.
Meanwhile, a separate reactor was charged with thiazole
alcohol 3 (2.65 kg, 9.1 mol), acetonitrile (7.6 kg), TBME (4.7
kg), and methanesulfonic anhydride (1.67 kg, 9.56 mol). The
slurry was stirred and cooled to 0 °C, and then treated dropwise
with Et3N (1.00 kg, 9.9 mol) while maintaining the mixture
below 5 °C. The mixture of mesylate 5 was held at 0 °C for
1 h and cooled to -20 °C.
2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-5-thia-
zolemethanol (3). A dry nitrogen purged vessel was charged
with THF (1.26 L) and 1 M LiAlH4 in THF (0.67 L, 0.670
mol), and the solution was cooled to -15 °C. A solution of
thiazole ester 12 (360 g, 1.08 mol) dissolved in THF (0.72 L)
was added over 1.5 h at a rate as to maintain the reaction
mixture temperature between -10 and -15 °C. The mixture
was allowed to stir at -15 °C for 0.5 h. The reaction was
consecutively quenched with a mixture of water and THF (1/
1, 54 mL) over 30 min, a 20% NaOH solution (20.2 mL) over
15 min, and water (93.6 mL) over 15 min. During the quenching
process, the reaction temperature was maintained at -10 and
-15 °C with vigorous stirring. The quenched mixture was
warmed to ambient over 30 min where it was maintained for
an additional 0.5 h The granular aluminum salts were filtered
and washed with THF (3 × 0.36 L). The combined filtrate was
concentrated under reduced pressure to ∼1.44 L. Water (2.88
L) was added to the mixture over 30 min while maintaining
the solution temperature at 50 °C. The resultant slurry was
cooled to 20 °C over 1 h, cooled further to 10 °C, and stirred
at 10 °C for 30 min. The solid was filtered, washed with heptane
(2 × 0.72 L), and dried under vacuum at 50-55 °C to a constant
weight to give 286.3 g of thiazole alcohol 3 as a slightly yellow
solid in 91%. 1H NMR (400 MHz, CDCl3): δ 8.38 (t, 1 H, J )
7.8 Hz), 7.49 (d, 1 H, J ) 8.5 Hz), 7.44 (d, 1 H, J ) 10.8 Hz),
4.88 (s, 2 H), 2.49 (s, 3 H). HPLC purity: 98.2% AUC. MS:
MH+ ) 272.
2-[4-[[[2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-
1,3-thiazol-5-yl]methyl]sulfanyl]-2-methylphenoxy]-2-meth-
ylpropanoic Acid (1). A stirred suspension of zinc (15.7 g,
0.24 mol) in ethyl acetate (220 mL) was heated at 40 °C, and
then acetic acid (7.87 mL, 0.137 mol) and water (2.47 mL, 0.137
mol) were added. To the resultant suspension was added
sulfonyl chloride 13 (22.0 g, 0.0687 mol) in small portions over
15 min while keeping the temperature below 60 °C. After the
addition was complete, the mixture was stirred for one hour.
HPLC showed a complete conversion to sulfinic acid. To the
mixture was added TMSCl (52.3 mL, 0.412 mol) over 1 h while
keeping the temperature below 55 °C. After the addition was
complete, the mixture was heated to gentle reflux (about 77
°C) for 3 h. Then thiazole alcohol 3 (20.0 g, 0.0687 mol) was
added in one portion. No obvious exotherm occurred. The
mixture was maintained at about 77 °C for 15 h or until the
reaction was complete. The mixture was cooled to ambient,
washed with water (2 × 100 mL), dilute NaCl solution (2 ×
100 mL), and concentrated under reduced pressure to ∼40 mL.
Ethanol (120 mL) was added, and the mixture was concentrated
to ∼40 mL. The solvent exchange process was repeated once.
To the mixture was added ethanol (120 mL), followed by water
(40 mL), and 50% sodium hydroxide (5.0 mL, 0.094 mol). The
cloudy mixture was heated to 60 °C for 2 h. A brown solution
was afforded. The mixture was cooled to 40 °C, and concen-
trated hydrochloric acid (15.0 mL, 0.175 mol) was added over
10 min. The resultant mixture was stirred at 60 °C for 15 min.
A slurry with yellow solid formed. The mixture was allowed
The TBME solution of lithium thiolate 6 was added to
the acetonitrile solution of mesylate 5 at such a rate that the
temperature was maintained between -10 and -5 °C. The
resulting mixture was held at -5 °C for 45 min and then
warmed to ambient. The mixture was quenched with water (20
kg). The organic layer was washed with water (14 kg). The
organic layer was concentrated to ∼9 L and then treated with
95% ethanol (20 kg) to effect crystallization at 15-20 °C. The
solid was filtered, washed with cold ethanol (6 kg), and dried
under vacuum at 55 °C to a constant weight to give 3.53 kg of
sulfide 19 as a yellow solid in 74% yield. 1H NMR (300 MHz,
CDCl3): δ 8.40 (t, 1 H, J ) 7.4 Hz), 7.51 (d, 1 H, J ) 8.1 Hz),
7.45 (d, 1 H, J ) 11.3 Hz), 7.19 (s, 1 H), 7.08 (d, 1 H, J ) 8.2
Hz), 6.68 (d, 1 H, J ) 8.2 Hz), 4.14 (s, 2 H), 2.26 (s, 3 H),
2.16 (s, 3 H), 1.02 (s, 9 H), 0.21 (s, 6 H). HPLC purity: 97.1%
AUC. MS: MH+ ) 528.
Vol. 13, No. 2, 2009 / Organic Process Research & Development
•
301

2-[4-[[[2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-
1,3-thiazol-5-yl]methyl]sulfanyl]-2-methylphenoxy]-2-meth-
ylpropanoic Acid (1). A slurry of silyl ether 19 (3.16 kg, 6.0
mol) and 20-40 mesh sodium hydroxide (2.4 kg, 6.0 mol) in
acetone (25 kg) was stirred at ∼33 °C for 1 h. A solution of
1,1,1-trichloro-2-methyl-propanol hydrate (2.0 kg, 11.4 mol) in
acetone (4 kg) was added over at least 60 min, while maintain-
ing the temperature at 36-38 °C. The reaction mixture was
cooled to ambient temperature over approximately 1 h. The
reaction volume was reduced under vacuum to 1/4 to 1/2 of
the total volume, at a temperature below 40 °C. TBME (22
kg) and 1 N hydrochloric acid (26 kg) were added, keeping
the temperature under 35 °C. After separation, the organic layer
was washed with water (2 × 20 kg) and concentrated to ∼16
L. The solution was heated to approximately 50 °C and heptane
(26 kg) was added. The mixture was concentrated at atmo-
spheric pressure until the head temperature reached 83-85 °C.
The solution was cooled slowly at <1 °C/min, during which
time crystallization began (typically 60 °C). The slurry was
further chilled to below 15 °C and stirred for at least 1 h. The
solid was collected by filtration, washed with heptane (10 kg),
and dried under vacuum at 45-55 °C to a constant weight to
give 1.95 kg of 1 as a slightly yellow solid in 65% yield. HPLC
purity: 97.3% AUC.
4-[[[2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-5-
thiazolyl]methyl]thio]-2-methylphenol (4). To a mixture of
thiazole alcohol 3 (48.6 kg, 166.8 mol), thiophenol 8 (26.3 kg,
187.6 mol), and acetonitrile (98.5 kg) was added trifluoroacetic
acid (185 kg). The mixture was stirred under nitrogen until the
exotherm was realized, ∼15 min. The mixture was heated to
∼65 °C over 45 min and maintained at that temperature for
6 h. To the mixture was added water (400 kg) while maintaining
the mixture above 40 °C. The mixture was held at 40 °C with
good stirring for one hour. The mixture was cooled to ambient
over 30 min. The solid was filtered, washed with a mixture of
acetonitrile (19.6 kg) and water (37.5 kg) followed by a heptane
(68 kg) wash, and dried under vacuum at 60 °C to a constant
weight to give 63.1 kg of phenol 4 as a slightly yellow solid in
89% yield. 1H NMR (400 MHz, CDCl3): δ 8.34 (t, 1 H, J )
7.7 Hz), 7.47 (d, 1 H, J ) 7.3 Hz), 7.43 (d, 1 H, J ) 9.9 Hz),
7.18 (s, 1 H), 7.03 (d, 1 H, J ) 8.2 Hz), 6.62 (d, 1 H, J ) 8.2
Hz), 4.10 (s, 2 H), 2.18 (s, 3 H), 2.17 (s, 3 H). HPLC purity:
97.6% AUC. MS: MH+ ) 414.
2-[4-[[[2-[2-Fluoro-4-(trifluoromethyl)phenyl]-4-methyl-
1,3-thiazol-5-yl]methyl]sulfanyl]-2-methylphenoxy]-2-meth-
ylpropanoic Acid (1). A mixture of 2-butanone (313.6 kg),
phenol 4 (56.0 kg, 135.4 mol), and 20-40 mesh NaOH (29.7
kg, 742.5 mol) was heated to 50 °C with good stirring over
approximately 30 min under nitrogen. The mixture was stirred
at 50 °C for 1 h. To the mixture was added a solution of
2-bromo-2-methylpropanoic acid (39.2 kg, 230.2 mol) in
2-butanone (44.8 kg) over approximately 1 h at 50 °C. After
the addition, the mixture was held at 50 °C for 2 h with stirring.
Water (224 kg) was added, and the mixture was held at 50 °C
for 1 h. The mixture was cooled to 20-25 °C. The layers were
separated, and the organic layer was washed with 1 N HCl (170
kg). The organic layer was filtered through a 30 µm in line
filter and concentrated to ∼112 L under reduced pressure.
Heptane (191.5 kg) was added, and the mixture was heated to
65-70 °C. The mixture was cooled to 10-15 °C over 2 h.
The solid was filtered, washed with heptane (57.4 kg), and dried
under vacuum at 50-55 °C to a constant weight to give 56.0
kg of 1 as a slightly yellow solid in 83% yield. HPLC purity:
98.5% AUC.
Received for review September 17, 2008.
OP8002294
302
•
Vol. 13, No. 2, 2009 / Organic Process Research & Development