An efficient synthesis of a dual PPAR α/γ agonist and the formation of a sterically congested α-aryloxyisobutyric acid via a Bargellini reaction

Article abstract of DOI:10.1021/jo051027+

A practical synthesis of benzisoxazole 1 and its conversion to α-aryloxyisobutyric acid 2 using 1,1,1-trichloro-2-methyl-2-propanol (chloretone) was developed. Benzisoxazole 1 was formed in high yields by the action of either methanesulfonyl chloride/base

Full text of DOI:10.1021/jo051027+

SCHEME 1. Alkylations of Phenol 2a
An Efficient Synthesis of a Dual PPAR r/γ
Agonist and the Formation of a Sterically
Congested r-Aryloxyisobutyric Acid via a
Bargellini Reaction
Raymond J. Cvetovich,*,† John Y. L. Chung,*,†
Michael H. Kress,† Joseph S. Amato,† Louis Matty,†
M. David Weingarten,†,§ Fuh-Rong Tsay,† Zhen Li,† and
George Zhou‡
Department of Process Research and Department of
Analytical Research, Merck Research Laboratories,
P.O. Box 2000, Rahway, New Jersey 07065
raymond_cvetovich@merck.com
Received May 23, 2005
a Reagents and conditions: (i) methyl 2-bromo-2-methylpropi-
onate (<50%); (ii) methyl 2-bromopropionate (90%); (iii) LDA, MeI,
THF (90%); (iv) chloretone; (v) aq NaOH.
from acetone, chloroform, and alkali metal hydroxide.5
Dual PPAR R/γ agonist 2, potentially useful for the
treatment of type II diabetes and dyslipidemia, is such
an R-aryloxyisobutyric acid possessing a highly sterically
congested ether linkage due to 2,6-di-n-propyl substitu-
tion flanking the phenolic oxygen.
A practical synthesis of benzisoxazole 1 and its conversion
to R-aryloxyisobutyric acid 2 using 1,1,1-trichloro-2-methyl-
2-propanol (chloretone) was developed. Benzisoxazole 1 was
formed in high yields by the action of either methanesulfonyl
chloride/base upon intermediate oxime 8 or with thionyl
chloride/base, which initially forms cyclic sulfite 10. A highly
reactive, short-lived intermediate derived from chloretone
was detected by ReacIR and its half-life determined to be
∼5 min. Reaction conditions for the Bargellini reaction were
developed that resulted in a 95% yield of 2 from the reaction
of highly hindered phenol 1 with chloretone hemihydrate
and powdered NaOH in acetone. Thus highly hindered
R-aryloxyisobutyric acids can be made in a single step in
high yield.
Attempted conversion of benzisoxazole 1 to methyl
R-aryloxyisobutyrate 3 by alkylation with methyl 2-bromo-
2-methylpropionate resulted in less than 50% conversion
of 1 (DMF, Cs2CO3, 60 °C) even after 7 days of heating
(see Scheme 1, Path A, 1-3-2).6a A high-yielding double
alkylation-hydrolysis sequence of benzisoxazole 1 with
methyl 2-bromopropionate6b/Cs2CO3 to form 4, followed
by alkylation with MeI/LDA and subsequent hydrolysis
(Path B, 1-4-3-2), was delineated for the preparation
of initial large amounts of R-phenoxyisobutyric acid 2.
But we were attracted by the potential efficiency of the
chloretone reaction with phenols7 (Path C, 1-2). A survey
of the literature revealed that the reported examples,
which employed hindered phenols, gave 15-60% yields
of R-aryloxyisobutyric acids.8 Our initial application of
the Bargellini chloretone reaction to phenol 1 using
chloroform/acetone/aqueous potassium hydroxide gave a
50% yield of R-phenoxyisobutyric acid 2. In this paper
we discuss the preparation of benzisoxazole 1 and the
development of a high-yielding reaction with chloretone
to prepare R-phenoxyisobutyric acid 2.
R-Aryloxyisobutyric acids are a motif present in a
variety of pharmaceutically active agents, including
hypolipidemics,1 antisicklings,2 plant-growth substances,3
and PPAR R/γ (peroxisome proliferator-activated recep-
tor) agonists.4 They can be prepared by condensing the
appropriate phenoxide with bromoisobutyryl acetates5 or
via reactions using trichloromethyl dimethylcarbinol
(chloretone), a commercially available reagent formed
The conversion of dipropylresorcinol 5 to oxime 8 was
accomplished via a 3-step synthesis as shown in Scheme
2. Treatment of dipropylresorcinol 59 with trifluoroacetic
* To whom correspondence should be addressed (mail stop: RY800-
B375).
† Department of Process Research.
‡ Department of Analytical Research.
§ Current address: AtheroGenics, Inc., 8995 Westside Parkway,
Alpharetta, GA 30004.
(6) (a) Davis, R. D.; Fitzgerald, R. N.; Guo, J. Synthesis 2004, 1959.
(b) Synerholm, M. E.; Zimmerman, P. W. Contrib. Boyce Thompson
Inst. 1945, 14, 91.
(1) Hogberg, T.; Bondesson, G.; Misiorny, A.; Stjernstrom, N. E. Acta
Pharm. Suec. 1976, 13, 427.
(7) (a) Gilman, H.; Wilder, G. R. J. Am. Chem. Soc. 1955, 77, 6644.
(b) Galimberti, P.; Defranceshi, A. Gazz. Chim. Ital. 1947, 77, 431. (c)
Weizmann, C.; Sulzbacher, M.; Bergmann, E. J. Am. Chem. Soc. 1948,
70, 1153. (d) Clewley, R. G.; Fischer, A.; Henderson, G. N. Can. J.
Chem. 1989, 67, 1472. (e) Lai, J. T. Tetrahedron Lett. 2001, 42, 557.
(8) Adams, A. D.; Hu, Z.; von Langen, D.; Dadiz, A.; Elbrecht, A.;
MacNaul, K. L.; Berger, J. P.; Zhou, G.; Doebber, T. W.; Meurer, R.;
Forrest, M. J.; Moller, D. E.; Jones, A. B. Bioorg. Med. Chem. Lett.
2003, 13, 3185.
(2) Abraham, D. J.; Mehanna, A. S.; Williams, F. S.; Cragoe, E. J.,
Jr.; Woltersdorf, O. W., Jr. J. Med. Chem. 1989, 32, 2460.
(3) Kuramochi, H.; Yoshimura, T.; Miyazawa, T.; Makoto, K.;
Takematsu, T.; Yoneyama, K. J. Agric. Food Chem. 1996, 44, 1906.
(4) Liu, K. G.; Lambert, M. H.; Leesnitzer, L. M.; Oliver, W., Jr.;
Ott, R. J.; Plunket, K. D.; Stuart, L. W.; Brown, P. J.; Willson, T. M.;
Sternbach, D. D. Bioorg. Med. Chem. Lett. 2001, 11, 2959.
(5) (a) Bargellini, G. Gazz. Chim. Ital. 1906, 36, 329. (b) Jozicz, Z.
Chem. Zentralbl. 1897, 68, I, 1013. (c) Jozicz, Z. Zh. Russ. Fiz. Khim.
Ova. 1897, 29, 97.
(9) Ceglia, S. S.; Kress, M. H.; McNamara, J. M. Tetrahedron Lett.
2005, 46,1731.
10.1021/jo051027+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/20/2005
8560
J. Org. Chem. 2005, 70, 8560-8563

SCHEME 2. Synthesis of Oxime 8
SCHEME 3. Synthesis of Benzisoxazole 1
anhydride in the presence of a strong acid catalyst,10 such
as trifluoromethanesulfonic acid, produced trifluoroac-
etophenone 6 in 92% yield, along with 3-5% of O-
acetylated byproduct 7. In the absence of the strong acid,
O-acetylation competed with C-acetylation and produced
as much as 20% of trifluoroacetate 7. Variable ratios of
6:7 were sometimes observed with use of catalytic
amounts of trifluoromethanesulfonic acid when different
samples of 5 were used. The erratic nature of this
reaction was ameliorated when additional triflic acid was
used in the reaction. The nature of this erratic behavior
was traced to the presence of small quantities of sodium
phenoxides in 5 due to incomplete protonation during its
isolation following its formation in the presence of
NaOH.9 These difficulties were addressed by obtaining
titrations of the extraction solutions of 5 and incorporat-
ing additional washes with aqueous acid. Conversion of
trifluoroacetophenone 6 to oxime 8 was achieved by using
aqueous hydroxylamine in methanol. Rate acceleration
was observed when bases were included, particularly
primary amines. N-Butylamine was chosen as the opti-
mal base, and this base formed a crystalline salt of the
phenol-oxime which precipitated from the reaction mix-
ture upon the addition of water in 90% yield. The single-
crystal X-ray structure was solved for the n-butylamine-
oxime salt and it was determined that the salt exists as
the (E)-stereoisomer in the solid state.
Conversions of 2′-hydroxy acetophenone oximes to
benzisoxazoles have been accomplished with a variety of
reagents: acetic anhydride,11 KOH,12 or trichloroacetyl
isocyanate.13 However, competitive formation of benzox-
azoles via Beckmann rearrangement followed by ring
closure (e.g., 11) has been reported using oxime acetates
with base or heat,11 excess KOH or p-toluenesulfonyl
chloride,12 phosphoryl chloride,14 zeolite catalysts,15 and
BiCl3 with microwave radiation.16 High-yielding prepara-
tion of the benzisoxazole ring of 1 was achieved from
either O-mesylate 9 or, uniquely, through cyclic sulfite
10 (see Scheme 3). Treatment of oxime 8 with methane-
sulfonyl chloride/diisopropylethylamine in isopropyl ac-
etate produced, initially, mesylate 9, which cyclized to
benzisoxazole 1 (92% yield). Alternatively, treatment of
SCHEME 4. Benisoxazole Formation from Cyclic
Sulfite 10
oxime 8 with thionyl chloride/diisopropylethylamine
produced, initially, cyclic sulfite 10, which also formed
benzisoxazole 1 (94% yield).
Although the mechanism of the cyclic sulfite 10 rear-
rangement has not been fully delineated, simple kinetic
experiments showed that the rate of formation of ben-
zisoxazole 2 was second order in the concentration of 10
and first order in the concentration of DIPEA, hinting
at the involvement of one phenoxide in initially cleaving
the cyclic sulfite of a second molecule to produce a species
such as 12 (see Scheme 3), which could then cyclize to
13. Sulfite ester 14 was isolated as a byproduct and
identified from reaction mixtures, and the addition of
dilute aq acid to reaction mixtures resulted in the
evolution of SO2.
The strength of the base used in the sulfite conversion
exerted an influence on the rate of product formation,
whereby the use of a weak base, such as imidazole, led
to formation of cyclic sufite 10, with very slow conversion
to the benzisoxazole. The order of addition of reagents
for the methanesulfonyl chloride mediated cyclization
was unimportant; however, in the cyclic sulfite conver-
sion, successful reactions required the addition of base
to a mixture of the oxime and thionyl chloride. An
alternate Beckmann rearrangement/ring closure path-
way of mesylate 9 or sulfite 10 to benzoxazole 11 (see
Scheme 3) was observed in both cases. This aryl rear-
rangement was limited with mesylate 9 to ∼1%, but
occurred in sulfite 10 in as high as 8%.
(10) (a) Casini, G.; Gualtieri, F.; Stein, M. L. J. Heterocycl. Chem.
1969, 6, 279. (b) Elkasaby, M. A.; Salem, M. A. I. Indian J. Chem.
1980, 19B, 571.
(11) Crabbe, P.; Villarino, A.; Muchowski, J. M. J. Chem. Soc.,
Perkins Trans. 1 1973, 2220.
(12) Stokker, G. J. Org. Chem. 1983, 48, 2613.
(13) Fujita, S.; Koyama, K.; Inagaki, Y. Synthesis 1982, 1, 68.
(14) Bhawal, B. M.; Mayabhate, S. P.; Lithite, A. P.; Deshmukh, A.
R. A. S. Synth. Commun. 1995, 25, 3315.
Conversion of benzisoxazole 1 to R-aryloxyisobutyric
acid 2 was successfully accomplished by a double alky-
lation/hydrolysis sequence, as illustrated above in Scheme
(15) Thakur, A. S.; Boruah, A.; Prajapati, D.; Sandhu, J. S. Synth.
Commun. 2000, 30, 2105.
(16) Butler, R. N.; O’Donoghue, D. A. J. Chem. Res. (S) 1983, 18.
J. Org. Chem, Vol. 70, No. 21, 2005 8561

FIGURE 2. Byproducts produced in the Bargellini chloretone
alkylation of benzisoxazole 1.
FIGURE 1. Profiles of the reaction obtained by spectral
decomposition.
Presumably, potassium and tetraalkylammonium car-
boxylate salts of R-aryloxyisobutyric acid 2 are either
more reactive than the corresponding sodium carboxylate
or more soluble. In using KOH, as much as 35% of the
oligomer 15 was isolated.
Neither LiOH nor DBU produced alkylation products.
t-BuONa in acetone or DME led to significant decomposi-
tion. Slow addition of both chloretone and NaOH to
benzisoxazole 1 produced less acetone aldol impurities
than other modes of addition. The formation of mesityl
oxide did not present a problem in terms of yield or
isolation of product, so the slow addition of the chloretone
solution to the mixture of components remained the
chosen mode of operation.
Solvents such as methyl ethyl ketone, isopropyl methyl
ketone, tert-butyl methyl ketone, diethyl ketone, and di-
isopropyl ketone gave less of the aldol byproducts, but
the alkylation proceeded at much slower rates and gave
lower yields of R-aryloxyisobutyric acid 2 relative to
reactions performed in acetone. Similar slow conversions
were also observed with MTBE, DME, and NMP. Reac-
tions in acetonitrile at 20 °C were exothermic, and
although an 85% conversion of starting material was
achieved the yield of desired product was only 45%.
Cooling the acetonitrile reaction to 10 °C slowed the
reaction rate, but an increase in polymeric impurities was
observed.
Benzisoxazole 1 was found to be stable (97% recovery)
for at least 4 days under the reaction conditions (solid
NaOH/acetone) in the absence of chloretone. A study on
the reaction temperature revealed that an accumulation
of reagents occurs below 30 °C, displaying a delayed
exotherm when chloretone was added below this tem-
perature. When chloretone was added to mixtures at >30
°C a mildly exothermic reaction was observed. Reactions
performed in these temperature ranges resulted in dif-
fering reaction yields (83% at 25-27 °C, 90% at 30-35
°C, and 83% at 40 °C). Lowering the reaction temperature
to 5-7 °C lowered the yield by another 10%.
1, and although excellent yields were achieved (91% over-
all) this required three separate steps for this sequence.
Since the conversion of phenols to R-aryloxyisobutyric
acid using chloroform, acetone, and sodium hydroxide
was first reported by Bargellini,5 others have reported
moderate success in this alkylation utilizing similar
conditions. A wide range of yields (20-90%) have been
reported with a variety of phenols and reaction conditions
with most reports in the 15-60% range.8 Our initial
attempts at converting benzisoxazole 1 to R-aryloxyisobu-
tyric acid 2 using chloretone-hemihydrate and aqueous
NaOH obtained modest yields (45-55%). Adding phase
transfer agent (n-Bu4NBr) offered little improvement.
FT-IR techniques were used to monitor reaction mix-
tures. When chloretone-hemihydrate was added to solid
NaOH/acetone, a short-lived species was observed with
a unique peak of 758 cm-1 that formed rapidly and
decayed within 5 min (see Figure 1). It is hypothesized
that this absorption is due to formation and degradation
of an intermediary dimethyl dichloro epoxide. Efforts to
observe this intermediate by NMR did not succeed.
In further optimization of the Bargellini chloretone
alkylation, when a large excess of solid NaOH was used
in conjunction with excess chloretone-hemihydrate added
in portions over a 5 h period, an 80% yield of R-aryloxy-
isobutyric acid 2 was achieved with 96% consumption of
benzisoxazole 1. The reaction is heterogeneous with both
the sodium salt of phenol 1 and acid 2 having limited
solubility in the reaction mixture. This improvement in
yield brought about by the slow addition of excess chlore-
tone was not observed in reactions using aqueous NaOH.
A number of byproducts (15-18) were isolated and
identified from reaction mixtures, as shown in Figure 2,
in addition to acetone self-condensation byproducts and
possible carbon monoxide formation.17 No significant
difference was observed in yield or byproduct distribution
by reversing the addition order of solid NaOH and
chloretone. The use of phase transfer agent (n-Bu4NBr)
with solid NaOH reactions gave a slower initial conver-
sion to product but did achieve an 88% conversion within
the same time period. With KOH (s), 50% aq NaOH/
phase transfer agents, or t-BuOK, R-aryloxyisobutyric
acid 1 reacted further to incorporate two or more isobu-
tyrates and produced 15 and 16 in significant amounts.
After further optimization of the reaction conditions,
the best reaction yields were achieved by suspending
solid NaOH in acetone, adding benzisoxazole 1, adjusting
and maintaining the reaction temperature to 30 °C, and
adding a solution of excess chloretone-hemihydrate in
acetone over 5 h. All the steps in this synthesis were
demonstrated on a kilogram scale. The Bargellini reac-
tion produced consistent 92-95% isolated yields of highly
(17) Nair, A. D.; Lach, J. L. J. Am. Pharm. Assoc. 1959, 48, 390.
8562 J. Org. Chem., Vol. 70, No. 21, 2005

Anal. Calcd for C14H16F3NO2: C, 58.53; H, 5.61; N, 4.88.
Found: C, 58.53; H, 5.65; N, 4.88. LCMS (API-ES+) m/z 288.1
(M + 1, 100). Oxime mesylate 9 intermediate was isolated by
using only 1.05 equiv of DIPEA. 1H NMR (400.25 MHz, CDCl3)
δ 6.83 (s, 1H), 5.16 (br s, 2H), 3.25 (s, 3H), 2.56 (t, J ) 7.7 Hz,
2H), 2.50 (t, J ) 7.7 Hz, 2H), 1.66-1.50 (om, 4H), 0.98 (t, J )
7.3 Hz, 3H), 0.96 (t, J ) 7.3 Hz, 3H). 13C NMR (100.65 MHz,
CDCl3) δ 155.2, 154.2 (J ) 34.2 Hz), 150.8, 126.0, 120.7, 119.7
(J ) 277.9 Hz), 115.5, 104.7, 36.8, 31.5, 25.3, 22.6, 22.0, 13.9,
13.8.
5,7-Dipropyl-3-(trifluoromethyl)-6-hydroxy-1,2-benzisox-
azole (1) with Thionyl Chloride. To a solution of oxime 8
(15.50 g, 50.8 mmol) in ethyl acetate (50 mL) cooled to 5 °C was
added thionyl chloride (4.0 mL, 54.8 mmol). A solution of
diisopropylethylamine (22.0 mL, 126.3 mmol) in ethyl acetate
(25 mL) was added dropwise over 10-15 min, during which time
the reaction temperature rose to 25 °C and precipitate formed.
The mixture was stirred for 2 h at 25 °C then aq HCl (1 N, 30
mL) was added. The ethyl acetate layer was concentrated in
vacuo to an oil, which was then dissolved into acetic acid (65
mL). Dropwise addition of water (65 mL) to this solution over
1.5 h produced crystalline benzisoxazole, which was isolated by
filtration to give 13.8 g of product after washing and drying as
above. The intermediate cyclic sulfite can be prepared and
isolated by using MTBE as the solvent and imidazole as the base
to afford an oil. 1H NMR (400.25 MHz, CDCl3) δ 7.21 (s, 1H),
5.54 (br s, 1H), 2.94-2.64 (om, 4H), 1.80-1.53 (om, 4H), 1.00 (t,
J ) 7.3 Hz, 6H). 13C NMR (100.65 MHz, CDCl3) δ 165.8 (q, J )
32.9 Hz), 156.9, 147.4, 127.6, 127.1, 123.8, 119.7 (q, J ) 278.9
Hz), 112.4, 32.0, 26.5, 22.58, 22.55, 14.3, 14.0. LCMS (API-ES+)
m/z 352.0 (M + 1, 5), 288.1 (M - 63, 100). HRMS (ESI) calcd for
C14H17F3NO4S 352.0830 (M + 1), found 352.0829.
hindered R-aryloxyisobutyric acid 1 on a kilogram scale
in a single step and is the preferred procedure to the
3-step (double alkylation and hydrolysis) process.
Experimental Section
1-(2,4-Dihydroxy-3,5-dipropylphenyl)-2,2,2-trifluoroet-
hanone (6). Toluene (58 mL) and bis-n-propyl resorcinol 59,10
(11.57 g, 57.4 mmol) were stirred under a N2 atmosphere and
triflic acid (25 µL) was added. The mixture was warmed to 35
°C and TFAA (10.53 mL) was added over 15 min while control-
ling the temperature between 30 and 40 °C. After stirring 1.5-
2.0 h, MeOH (10 mL):water (1 mL) was added slowly while
maintaining the temperature between 35 and 45 °C. The mixture
was stirred at 45 °C for 2 h (to hydrolyze the O-acylated
byproduct) then cooled to 22 °C and quenched slowly into cold 6
wt % aq NaHCO3 (200 mL). After stirring for 30 min, the layers
were separated and the organic layer was washed with 6 wt %
aq NaHCO3 (100 mL) then water (22 mL). The yield of ketone
in toluene solution was 92.4%. 1H NMR (400.25 MHz, CDCl3) δ
11.7 (s, 1H), 7.453 (d, J ) 2.2 Hz, 1H), 5.73 (s, 1H), 2.66 (q, J )
5.1 Hz, 2H), 2.56 (q, J ) 5.1 Hz, 2H), 1.70-1.55 (om, 4H), 1.02-
0.97 (om, 6H). 13C NMR (100.665 MHz, CDCl3) δ 182.4 (q, J )
34.5 Hz), 163.9, 161.0, 129.3 (q, J ) 3.8 Hz), 121.3, 116.9 (q, J
) 289.6 Hz), 115.7, 107.8, 31.6, 24.5, 22.5, 21.7, 14.0, 13.7. LCMS
(API-ES+) m/z 291.1 (M + 1, 100), 221.1 (80). Anal. Calcd for
C
14H17F3O3: C, 57.93; H, 5.90; F, 19.63. Found: C, 58.03; H,
6.06; F, 19.39.
(1E)-1-(2,4-Dihydroxy-3,5-dipropylphenyl)-2,2,2-trifluo-
roethanone Oxime (8). To a solution of trifluoroketone 6 (50
g, 172 mmol) in methanol (172 mL) under a N2 atmosphere was
added n-butylamine (25.2 g, 344 mmol). Hydroxylamine (50%
aq solution, 22.8 g solution, 344 mmol) was added and the
mixture was warmed to 60 °C, then stirred for 15-20 h. The
mixture was cooled to 35 °C, and water (30 mL) was added to
produce crystals. The crystallization was completed by adding
water (214 mL) then cooling to 20 °C. The product was filtered
and washed with 10% methanol/water and then water. The
solids were dried in vacuo to produce 47.3 g of oxime-amine salt
2-{[5,7-Dipropyl-3-(trifluoromethyl)-1,2-benzisoxazol-6-
yl]oxy}-2-methylpropanoic Acid (1). To a 15 °C solution of
benzisoxazole 2 (11.56 g, 99.4% pure, 40 mmol) in acetone (160
mL), stirring at 450 rpm, was added 20-40 mesh NaOH (32.7
g, 800 mmol). Upon mixing, the temperature rose to 25-30 °C.
The temperature was adjusted to 27-30 °C, then a solution of
chloretone hemihydrate (38.1 g, 200 mmol) in acetone (80 mL)
was added subsurface over 5 h. The reaction mixture was
maintained at 28-32 °C during this period, during which time
gaseous carbon monoxide17 is generated and sodium chloride
precipitates. The reaction mixture was aged at 28-30 °C for 1
h and at 21 °C for 12-18 h, during which time oligomer and
dimer impurities hydrolyze. The reaction mixture (assay yield
∼92%) was slowly added to 0 °C H2O (200 mL) while maintain-
ing the temperature at <10 °C. The pH of the mixture was
adjusted to pH 4.5-4.7 using concentrated HCl (15-16 mL),
during which time the mixture separated into two layers. The
mixture was warmed to 22 °C and extracted with a mixture of
heptane (180 mL) and MTBE (18 mL). The organic layer was
washed with H2O (5 × 200 mL), then diluted with MTBE (144
mL). A 1 wt % aq NaCl (335 mL) solution was added, which
was then treated with 5 N NaOH (∼8 mL) to adjust the pH to
11.2-11.5. The organic layer was removed and the aqueous layer
was mixed with heptane (200 mL) and the pH was adjusted to
4.1-4.5 with concentrated HCl (3.5 mL). The organic layer was
filtered, concentrated in vacuo, dissolved in acetic acid (54 mL),
and crystallized with the addition of water (54 mL) to give 13.2
g as an off-white solid. This material could also be crystallized
1
8 in 90% isolated yield. H NMR (400.25 MHz, CD3OD) δ 6.90
(s, 1H), 5.09 (br s, OH), 2.82 (t, J ) 7.6 Hz, 2H), 2.68 (m, 2H),
2.53 (t, J ) 7.5 Hz, 2H), 1.62-1.50 (om, 6H), 1.38 (m, 2H), 1.00-
0.91 (om, 9H). 13C NMR (100.65 MHz, CD3OD) δ 155.8, 154.9,
148.3 (J ) 27.9 Hz), 125.3, 123.1 (q, J ) 274.5 Hz), 120.0, 118.5,
109.8, 39.3, 32.0, 30.0, 25.4, 22.9, 22.1, 19.3, 13.2, 12.7, 12.5.
Oxime-amine Salt Break. Oxime 8, free of n-butylamine, was
obtained by extraction with IPAc and 1N HCl, followed by
solvent removal. 1H NMR (400.25 MHz, CDCl3) δ 9.8 (br s, 1H),
6.97 (s, 1H), 6.00 (br s, 1H), 5.08 (br s, 1H), 2.70 (t, J ) 7.7 Hz,
2H), 2.55 (t, J ) 7.6 Hz, 2H), 1.69-1.57 (om, 4H), 1.00 (om, 6H).
13C NMR (100.65 MHz, CDCl3) δ 155.4, 152.4, 149.0 (q, J ) 32.2
Hz), 126.0, 121.3, 120.6 (q, J ) 276.4 Hz), 117.7, 106.5, 31.8,
25.7, 22.7, 22.2, 14.1, 13.8. Anal. Calcd for C14H18F3NO3: C,
55.08; H, 5.94; N, 4.59. Found: C, 55.10; H, 5.94; N, 4.51. LCMS
(API-ES+) m/z 306.1 (M + 1, 16), 260 (100).
5,7-Dipropyl-3-(trifluoromethyl)-6-hydroxy-1,2-benzisox-
azole (1) with Methanesulfonyl Chloride. To a solution of
oxime 8 (39.6 g, 0.13 mol) in isopropyl acetate (0.50 M) and
diisopropylethylamine (0.156 mol, 20.1 g, 27.1 mL) under a N2
atmosphere was added methanesulfonyl chloride (16.36 g, 11.05
mL) dropwise over 30 min while maintaining the temperature
at -5 to 0 °C. Additional DIPEA (0.156 mol, 20.1 g, 27.1 mL)
was added, and the mixture was heated to 55 °C and stirred for
10 h. The reaction was cooled to 25 °C, 4 N HCl was added, and
the phases were separated. The organic phase was washed with
aq bicarbonate, then concentrated to an oil. The oil was dissolved
with acetic acid (concentration ∼140 g/L), and water was added
dropwise to crystallize to give 38.0 g (isolated yield: 92%) of
benzisoxazole 1. 1H NMR (400.25 MHz, CDCl3) δ 7.37 (s, 1H),
5.51 (br s, 1H), 2.92 (t, J ) 7.6 Hz, 2H), 2.71 (t, J ) 7.7 Hz, 2H),
1.81-1.66 (om, 4H), 1.04-0.99 (om, 6H). 13C NMR (100.65 MHz,
CDCl3) δ 163.7, 154.8, 149.6 (q, J ) 38.1 Hz), 128.9, 120.4 (q, J
) 271.5 Hz, 117.7, 110.2, 109.7, 32.7, 25.7, 22.7, 2.0, 13.9, 13.8.
1
from heptane. H NMR (400.25 MHz, CDCl3) δ 11.4 (br s, 1H),
7.43 (s, 1H), 2.92 (m, 2H), 2.69 (m, 2H), 1.85-1.64 (om, 4H),
1.59 (s, 6H), 1.03-0.97 (om, 6H). 13C NMR (100.65 MHz, CDCl3)
δ 179.8, 163.6, 154.5, 149.6 (q, J ) 38.2 Hz), 137.1, 121.0, 120.2
(q, J ) 271.5 Hz), 116.2, 113.6, 81.9, 33.9, 27.8, 25.2 (2C), 23.3,
22.3, 14.1, 13.9. Anal. Calcd for C18H22F3NO4: C, 57.90; H, 5.94;
N, 3.75. Found: C, 57.94; H, 5.90; N, 3.69. LCMS (API-ES+)
m/z 374.0 (M + 1, 15), 288.1 (100).
Supporting Information Available: 1H NMR and 13C
NMR spectra for all compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO051027+
J. Org. Chem, Vol. 70, No. 21, 2005 8563