Development of one-pot synthesis of new antiarthritic drug candidate S-2474 with high E-selectivity

Article abstract of DOI:10.1021/op800008w

A one-pot synthesis of S-2474 was developed to overcome the problems of a large number of steps, low stereoselectivity, low yield, a large amount of waste, and severe reaction conditions. Aldol-type condensation of 3,5-di-terf-butyl-4-hydroxybenzaldehyde and iV-ethyl-γsultam was carried out with LDA and then quenched with water. Dehydration proceeded under basic conditions, providing S-2474 directly as a single isomer on the benzylidene double bond. The reaction mechanism appears to involve a quinone methide intermediate. Environmental assessment of the development of this compound is also discussed in this paper.

Full text of DOI:10.1021/op800008w

Organic Process Research & Development 2008, 12, 442–446
Development of One-Pot Synthesis of New Antiarthritic Drug Candidate S-2474 with
High E-Selectivity
Katsuo Oda,† Takemasa Hida,*,† Teruo Sakata,‡ Masahiko Nagai,‡ Yoshihide Sugata,‡ Toshiaki Masui,† and Hideo Nogusa†
Chemical DeVelopment Department, CMC DeVelopment Laboratories, Shionogi & Co., Ltd., 1-3, Kuise Terajima 2-chome,
Amagasaki, Hyogo 660-0813, Japan, and Shionogi Research Laboratories, Shionogi & Co., Ltd., 12-4, Sagisu 5-chome,
Fukushima-ku, Osaka 553-0002, Japan
Abstract:
the Z-isomer 8 of S-2474 as a byproduct (E/Z ) 84/16). The
overall yield was only 19.8%. Another synthetic route was
A one-pot synthesis of S-2474 was developed to overcome the
problems of a large number of steps, low stereoselectivity, low
yield, a large amount of waste, and severe reaction conditions.
Aldol-type condensation of 3,5-di-tert-butyl-4-hydroxybenzalde-
hyde and N-ethyl-γ-sultam was carried out with LDA and then
quenched with water. Dehydration proceeded under basic condi-
tions, providing S-2474 directly as a single isomer on the ben-
zylidene double bond. The reaction mechanism appears to involve
a quinone methide intermediate. Environmental assessment of the
development of this compound is also discussed in this paper.
Scheme 1. Original synthesis (first-generation route) of
S-2474 (1)
Introduction
(E)-(5)-(3,5-Di-tert-butyl-4-hydroxybenzylidene)-2-ethyl-1,2-
isothiazolidine-1,1-dioxide (S-2474, 1), which was discovered
at Shionogi Research Laboratories, shows potent inhibitory
effects on both cyclooxygenase-2 (COX-2) and 5-lipoxygenase
(5-LO) and is anticipated to be promising as an antiarthritic
drug.1 Study of its structure-activity relationships in in vivo
assays shows that the tert-butyl group on benzene, N-alkylated
γ-sultam and the E-isomer are essential for its high bioavail-
ability.1 The E-isomer is particularly indispensable for oral
bioavailability because the Z-isomer is not detected in plasma
after oral administration.1
developed by Inagaki et al. using a quinone methide derivative
as an equivalent of the protected p-hydroxybenzaldeyde deriva-
tive (Scheme 2).2 It required four reactions and the overall yield
rose to 71.2%. Although the overall yield was improved and
the generation of 8 could be controlled, this route had a problem
with respect to scale-up manufacturing. Quinone methide
derivative 10 underwent sublimation, and its preparation
required a very high temperature reaction in vacuo (160 °C,
The original synthetic route (Scheme 1) required five
reactions, including protection and deprotection of the formyl
group to introduce the MOM group to the hydroxy group of
3,5-di-tert-butyl-4-hydroxybenzaldehyde 2, because methoxym-
ethylation of 2 gave the quinone enolate 3.2 This route required
chromatographic purification because dehydration of 7 produced
Scheme 2. Second-generation route of S-2474 (1)
* To whom correspondence should be addressed. Telephone: +81-6-6401-
8198. Fax: +81-6-6401-1371. E-mail: takemasa.hida@shionogi.co.jp.
† Chemical Development Department, CMC Development Laboratories.
‡ Shionogi Research Laboratories.
(1) Inagaki, M.; Tsuri, T.; Jyoyama, H.; Ono, T.; Yamada, K.; Kobayashi,
M.; Hori, Y.; Arimura, A.; Yasui, K.; Ohno, K.; Kakudo, S.; Koizumi,
K.; Suzuki, R.; Kato, M.; Kawai, S.; Matsumoto, S. J. Med. Chem.
2000, 43, 2040.
(2) Inagaki, M.; Haga, N.; Kobayashi, M.; Ohta, N.; Kamata, S.; Tsuri, T.
J. Org. Chem. 2002, 67, 125.
442
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development
10.1021/op800008w CCC: $40.75
2008 American Chemical Society
Published on Web 04/30/2008

Table 1. Screening of base on condensation of 2 and 6a
entry
base
product 12 (%)b
1
2
3
4
5
NaHMDS
KHMDS
LiHMDS
n-BuLi
46
59
75
80
90
LDA
a All reactions were carried out in THF at -60 °C. b Determined by HPLC.
threo-12a and erythro-12b mixture.
Table 2. Effect of temperature and solvent on condensation
of 2 and 6a
temp
(°C)
ratiob of
12a/12b
12a+12b
entry
solvent
(%)c
1
2
3
THF
THF
DMI
-60
25
25
6/1
1/1
1/1
90
Figure 1. Proposed mechanism of generation of threo- and
erythro-isomers on aldol-type reaction.
74d
97
a All reactions were carried out with 2.2 mol equiv of LDA. b Determined by H
NMR. c Determined by HPLC. d Slaggy lithium salt of 2 and 12 separated out in
the course of reaction, and the aldol reaction did not proceed completely.
20-100 mmHg). In sum, these two routes are not suitable for
industrial production. We thus aimed at the development of a
new synthetic route which could be used for pilot-scale
manufacturing.
Results and Discussion
Figure 2. Anti-elimination of threo-isomer 13a using K2CO3.
Scheme 3. Telescoped process of S-2474 (1)
Development of a New Synthetic Route of S-2474. First,
screening of the base of aldol-type condensation of 2 and
N-ethyl-γ-sultam 6 was investigated (Table 1). The counterca-
tion of the base was screened by using hexamethyldisilazane
(HMDS), which showed that lithium was better than sodium
and potassium (entries 1-3). Lithium diisopropylamide (LDA)
gave a better yield than LiHMDS and n-butyl lithium and was
thus chosen as an aldol reaction reagent (entry 5). The reaction
using LDA in THF required a very low temperature (Table 2,
entry 1) because lithium phenolate easily aggregates3 and
separates out at room temperature. The reaction did not proceed
to completion (entry 2). However, we found that lithium
phenolate dissolved in 1,3-dimethyl-2-imidazolidinone (DMI)
at room temperature, and the reaction could proceed smoothly
(entry 3). The ratio of threo-isomer 12a and erythro-isomer 12b
was dependent on the reaction temperature (Table 2). As the
temperature decreased, the ratio of threo-isomer increased.
These results show that steric hindrance of the sulfonyl group
and di-tert-butyl phenoxide seems to affect the stereoselectivity;
the reaction is controlled kinetically (Figure 1). The lithium salt
of 12 was quenched under acidic condition. The hydroxy group
of 12 was chlorinated and then eliminated with base. On using
potassium carbonate (K2CO3), Z-isomer 8 was generated at 20%
as a byproduct. Anti-elimination of threo-isomer 12a seemed
to proceed in the presence of a base such as K2CO3 (Figure 2).
Dechlorination with NaHCO3 did not give 8 even though the
reaction mixture included threo-isomer 12a, which provided 1
in good yield with high E-selectivity. These three reactions can
be carried out by a telescoped process (82.5%, Scheme 3),
without isolation of the intermediates.
Development of a One-Pot Process for Making S-2474.
Careful monitoring of the reaction showed that a very small
amount of 1 was generated when the lithium salt of 12 was
quenched. The yield of 1 seemed to be affected by the
exothermic heat of quenching. We estimated that 1 was obtained
directly without chlorination and dechlorination. Next, the
reaction conditions for dehydration of 12 were examined. The
dehydration preferred the basic condition over the acidic as
shown in Table 3. Addition of water and heating to 50-90 °C
(3) Jackman, L. M.; Chen, X. J. Am. Chem. Soc. 1992, 114, 403.
Vol. 12, No. 3, 2008 / Organic Process Research & Development
•
443

Table 3. Optimization of quenching and dehydration of 12a
quenching
reagent
temp
(°C)
product 1
entry
(%)b
1
2
3
4
none
HCl
water
water
60
60
50
90
0c
0d
93
99
a All reactions were carried out using the aldol type condensation reaction
mixture. b Determined by HPLC. c Retro aldol reaction proceeded. 2 and 6 were
obtained. d No reaction. 12 was recovered.
gave 1 (entries 3 and 4). The pH value in this reaction was
over 13. To our surprise, Z-isomer 8 was not generated in spite
of such a strong basic condition. This means that E2 elimination
of threo-isomer 12a did not occur. The reaction details are
discussed in the next section.
This one-pot synthesis provided 1 in 84.1% isolated yield
with high quality (>99.9% purity, Scheme 4). This process did
not require extraction of the product, and crystals of 1 were
obtained directly on pH adjustment of the reaction mixture to
4.5.
Figure 3. Proposed dehydration mechanism.
threo-isomer 12a and erythro-isomer 12b, would give the same
intermediate. Our hypothesis was supported by molecular
mechanical (MM) calculation for dihedral angle rotation. The
dihedral angle energy of the quinone methide intermediate
corresponding to Z-isomer 8 was higher than that of
the E-isomer (Figure 4). This means that the conformation of
the quinone methide intermediate is controlled by steric
hindrance of the sulfonyl and the aryl groups. From this finding,
we propose a dehydration mechanism via a quinone methide
intermediate corresponding to the E-isomer (Figure 3). If the
mechanism were normal E2 elimination, the Z-isomer 8 would
be produced by anti-elimination of the threo-isomer 12a.
Environmental Assessment of Synthetic Route. This one-
pot process does not require chromatographic purification
because dehydration proceeds without generation of the Z-
isomer. Since it requires only two reactions without any
protection or introduction of a leaving group, the process has a
shorter operation time and is environmentally friendly. Also,
this one-pot process does not require any severe reaction
conditions. From the viewpoint of environmental assessment,
the preparation of 1 via four different routes is compared in
Table 4. With the one-pot process, atom economy3 and reaction
mass efficiency (RME)4 increase dramatically.
Scheme 4. One-pot process of S-2474 (1)
Reaction Mechanism of Dehydration with E-Selectivity.
The dehydration mechanism details were examined. 1 and 8
did not undergo interconversion even if heated at 100 °C. Thus,
the E-selectivity did not seem to be controlled by the thermo-
dynamic factor of the products. We focused on the unique
character of the phenol derivative which easily forms the
quinone methide.2 Dehydration of 12 proceeded even using 1
mol equiv of NaHCO3 with E-selectivity. On the other hand,
the p-methoxy derivative 14 did not undergo dehydration via
the aldol adduct 15 (Scheme 5). These results mean that the
dehydration mechanism is not normal E2 elimination. We
speculated that this dehydration would also proceed via the
quinone methide form as shown in Figure 3. A similar reaction
has been known to exist; for example, elimination of the acetoxy
group at the benzylic position of the phenol derivative gives
the p-quinone methide derivative.5 The quinone methide
intermediate quickly equilibrates with phenol compounds.6 In
addition, this quinone methide-type elimination seems to affect
the high E-selectivity. We hypothesized that phenoxide anions,
Conclusion
A one-pot synthesis of S-2474 for pilot-scale manufacturing
was developed. This route consisting of only two reactions gave
an overall yield of 84.1% with excellent E-selectivity. The
excellent stereoselective dehydration to form E-olefin was
clarified by examination and MM calculation, which relies on
Scheme 5
(4) Trost, B. M. Science 1991, 254, 1471 Atom economy is defined by the
following expression: [FW (g mol-1) product]/[FW of all reactants used
in reaction] ×100.
(5) Curzons, A. D.; Constable, D. J. C.; Mortimer, D. N.; Cunningham,
V. L. Green Chemistry 2001, 3, 1 Reaction mass efficiency is defined
by the following expression:[mass of isolated product (kg)] /[total mass
of reactants used in reaction (kg)] ×100.
(6) Takao, K.; Sasaki, T.; Kozaki, T.; Yanagisawa, Y.; Tadao, K.;
Kawashima, A.; Shinonaga, H. Org. Lett. 2001, 3, 4291.
(7) Hunter, S. E.; Savage, P. E. J. Org. Chem. Soc. 2004, 69, 4724.
444
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development

Procedure for a Telescoped Process of S-2474 Synthesis
(Scheme 3). Into a solution of 3,5-di-tert-butyl-4-hydroxyben-
zaldehyde 2 (15.0 g, 64.0 mmol) and N-ethyl-γ-sultam 6 (10.0
g, 73.4 mmol) in DMI (120 mL) was added dropwise 25% LDA
(60.3 g, 140.8 mmol) at 25 °C. This was stirred for 30 min,
and then water (144 mL) was added dropwise at 25 °C. After
toluene (150 mL) and 62% H2SO4 (27.44 g) had been added,
the organic layer was separated and washed with water (75 mL,
twice). The organic layer was concentrated to 154 g under
reduced pressure. Into the residue were added DMI (7.5 mL)
and triethylamine (11.7 g, 115.6 mmol), and then thionyl
chloride (8.38 g, 70.5 mmol) was slowly added dropwise at 0
°C. Into this reaction mixture, 8.2% NaHCO3 solution (196.1
g, 192.0 mmol) was slowly added dropwise, followed by heating
to room temperature. After monitoring of the end-point of the
elimination, the organic layer was separated, washed with water
(75 mL, twice), and concentrated to 50 g under reduced
pressure. The residue was cooled to -10 °C, and white crystals
were obtained by filtration. They were recrystallized from
2-propanol. By this procedure, 19.3 g (52.8 mmol) of S-2474
1 was obtained (isolated yield: 82.5%). The purity was over
99.9% as determined by HPLC; mp 135-137 °C.1 Z-isomer 8
was not detected.
Procedure for a One-Pot Process to S-2474 (Scheme 4).
Into a solution of 3,5-di-tert-butyl-4-hydroxybenzaldehyde 2
(18.0 g, 76.8 mmol) and N-ethyl-γ-sultam 6 (12.0 g, 80.4 mmol)
in DMI (144 mL) was added dropwise 25% LDA (77.5 g, 180.9
mmol) at 25 °C. This was stirred for 30 min; then water (144
mL) was added, and the mixture was heated to 90 °C for an
hour. After monitoring of the end-point of the reaction, the
mixture was cooled, then water (108 mL) and 2-propanol (145
mL) were added. Adjustment of the pH to 4.5 with 20%
hydrochloric acid gave a slurry and white crystals by filtration.
Crystallization from 2-propanol gave 23.6 g (64.6 mmol) of
S-2474 1 (isolated yield: 84.1%). The purity was over 99.9%
as determined by HPLC; mp 135-137 °C.1 Z-isomer 81 was
not detected.
Figure 4. Dihedral energy plot of the quinone intermediate.
This was obtained using the dihedral energy plot of MOE
2006.06, and the structure of the quinone methide intermediate
was minimized using MMFF94x. The dihedral angle energy of
the quinone methide intermediate corresponding to the E-
isomer (-90°) was about 12 kcal/mol based on the energy of
the most stable conformation (dihedral angle -9°), and that of
the Z-isomer (+90°) was about 47 kcal/mol.
Table 4. Comparison of enviromental assessment for routes
of S-2474 (1)
reaction
atom
overall
route
number economy (%) RME (%) yield (%)
first generation
Scheme 1
second generation
Scheme 2
telescoped process
Scheme 3
one-pot process
Scheme 4
5
4
3
2
31.8
48.9
46.2
69.3
3.1
21.2
20.2
48.0
19.8
71.2
82.5
84.1
(5R*,1′R*)-5-[3,5-Di-tert-butyl-4-hydroxyphenyl)hydroxy-
methyl]-2-ethyl-1,2-isothiazolidine-1,1-dioxide 12a (threo)
and (5S*,1′R*)-5-[3,5-Di-tert-butyl-4-hydroxyphenyl)hy-
droxymethyl]-2-ethyl-1,2-isothiazolidine-1,1-dioxide 12b (eryth-
ro). LDA was prepared with diisopropylamine (118 mL) in THF
(282 mL) and 1.7 mol/L of n-butyl lithium/hexane solution (500
mL). In a 2-L flask, a solution of 2 (94 g), 6 (71.8 g) and THF
(1410 mL) was cooled to -60 °C and LDA was added dropwise
for 40 min. After the dropwise addition, the reaction mixture
was heated to 15 °C followed by stirring for 4.5 h. The mixture
was poured into 1 N HCl (3760 mL) and ethyl acetate (3760
mL). The organic layer was separated, washed with water (1880
mL × 2), and evaporated. The residue (226 g) was dissolved
in ethyl acetate (1630 mL) and methanol (1130 mL), then
methanol (2000 mL) including Girard’s reagent T (146 g) was
added. The mixture was left standing for 40 h at room
temperature. Next, ethyl acetate (1500 mL) and water (1000
mL) were added, and the mixture was stirred. The organic layer
was separated and washed with water (1500 mL × 4).
Evaporation of the organic layer gave 145 g of residue of which
133 g was crystallized using ethyl acetate and hexane to yield
the unique character of the phenol derivative that can easily
form the quinone methide under basic conditions. This process
is environmentally friendly and practical for large-scale
manufacturing.
Experimental Section
NMR spectra were measured on Varian Unity Inova-600
and Varian Mercury 300 spectrometers. High-resolution mass
spectra were recorded on JEOL JMS-SX/SX102A. Di-tert-
butylbenzaldehyde was obtained from Tokyo Chemical Indus-
try. Lithium diisopropylamide (LDA) was obtained from
Chemetall Gmbh. N-Ethyl-γ-sultam was prepared by a proce-
dure described in ref 1. HPLC analysis was carried out using
Shimadzu 10A-VP. S-2474 1 and (Z)-S-2474 8 were identified
by melting point range and NMR spectra in comparison with
reference data.1 Its purity was determined by HPLC: column,
Capcell Pak C18 (5 µm) 4.5 mm × 150 mm (Shiseido); eluent,
water/methanol/acetonitrile ) 7:10:3; flow rate, 1.2 mL/min;
column oven temperature, 25 °C; wavenumber, 230 nm; tR of
S-2474, 18 min. tR of (Z)-S-2474, 16 min.
Vol. 12, No. 3, 2008 / Organic Process Research & Development
•
445

the threo-isomer 12a (63.39 g). The filtrate included the erythro-
isomer 12b. It was evaporated to a residue (66.39 g) from which
58 g was crystallized from hexane to give crude 12b (42.5 g).
The crude crystals (30 g) were purified by silica gel chroma-
tography (70-230 mesh, 240 g, dichloromethane as eluent).
Crystallization with diethylether gave 12b (23.35 g). Compound
12a: mp 160-162 °C. 1H NMR (500 MHz, (CDCl3 δ) 1.25 (t,
J ) 7.3 Hz, 3H), 1.43 (s, 18H), 1.84 (m, 1H), 1.94 (m, 1H),
3.03 (ddd, J ) 9.1, 8.2, 7.2 Hz, 1H), 3.08 (dq, J ) 13.4, 7.2
Hz, 1H), 3.18 (ddd, J ) 9.1, 8.0, 3.5 Hz, 1H), 3.23 (dq, J )
13.4, 7.2 Hz, 1H), 3.53 (m, 1H), 4.84 (d, J ) 9.7 Hz 1H), 5.27
(s, 1H), 7.16 (s, 2H). 13C NMR (150 MHz, (CDCl3 δ) 13.18,
23.09, 30.24, 34.41, 39.42, 43.92, 63.67, 74.64, 123.58, 129.99,
136.38, 154.22. Elemental analysis: Calcd for C20H33O4NS: C;
62.63, H; 8.67, N; 3.65, S; 8.36, Found: C; 62.58, H; 8.62, N;
3.66, S; 8.32. Compound 12b: mp 78-96 °C. 1H NMR (500
MHz, (CDCl3 δ) 1.25 (t, J ) 7.3 Hz, 3H), 1.44 (s, 18H), 2.08
(m, 1H), 2.60 (dq, J ) 13.0, 8.5 Hz, 1H), 3.07 (m, 1H), 3.08
(m, 1H), 3.21 (dq, J ) 13.3, 7.2 Hz, 1H), 3.28 (m, 1H), 3.31
(d, J ) 2.6 Hz, 1H), 3.38 (td, J ) 8.5, 2.1 Hz 1H), 5.21 (s,
1H), 5.40 (brs, 1H), 7.15 (s, 2H). 13C NMR (150 MHz, (CDCl3
δ) 13.18, 18.30, 30.28, 34.43, 39.46, 44.56, 63.21, 69.19,
122.41, 129.94, 136.14, 153.54. Elemental analysis: Calcd for
C20H33O4NS: C; 62.63, H; 8.67, N; 3.65, S; 8.36, Found. C;
62.19, H; 8.63, N; 3.62, S; 8.10.
Into the residue, acetone (15 mL) and 1 N HCl (5 mL) were
added for hydration of acetal. The reaction mixture was
extracted with ethyl acetate, and then the organic layer was
evaporated to give 3,5-di-tert-butyl-4-methoxybenzaldehyde 14
(3.25 g).
Aldol-type condensation of 14 and 6 and dehydration were
tried according to the “Procedure for a One-Pot Process to
S-2474”. However, no dehydration was detected with HPLC.
The reaction mixture was extracted with toluene, and evapora-
tion gave white crystals of 15 as a diastereomer mixture (the
1
threo-isomer was the major product). H NMR (600 MHz,
(CD3)2SO δ) 1.10 (t, J ) 7.2 Hz, 3H), 1.38 (s, 18H), 1.61 (m,
1H), 1.69 (m, 1H), 2.95 (m, 2H), 2.98 (m, 2H), 3.38 (minor)
and 3.43 (major) (ddd, J ) 11.9, 9.7, 8.4 Hz, 1H), 3.61 (s, 3H),
4.65 (major) and 4.80 (minor) (dd, J ) 9.7, 4.8 Hz, 1H), 5.65
(minor) and 5.72 (major) (d, J ) 4.8 Hz, 1H), 7.26 (major)
and 7.28 (minor) (s, 2H). 13C NMR (150 MHz, (CD3)2SO δ)
13.0 (major) and 13.1 (minor), 21.7 (minor) and 22.7 (major),
31.8, 35.3, 39.2, 43.4 (major) and 44.2 (minor), 63.2 (major)
and 62.9 (minor), 63.9 (minor) and 64.0 (major), 70.1 (minor)
and 72.3 (major), 124.7 (minor) and 125.0 (major), 135.8
(major) and 136.1 (minor), 142.2 (minor) 142.7 (major), 158.3.
HRMS (FAB+) Calcd for C21H35NO4S: ([M + Na]+), 420.2185:
Found. m/z 420.2183.
Examination of Reaction Mechanism Study (Scheme 5).
3,5-Di-tert-butyl-4-hydroxybenzaldehyde (7.0 g, 29.9 mmol),
ethylene glycol (2.8 g, 45.2 mmol), toluene (70 mL), and
p-toluene sulfonic acid monohydrate (113 mg, 0.59 mmol) were
stirred and heated. The water formed during the reaction was
removed azeotropically over the 7 h. The reaction mixture was
poured into saturated NaHCO3, and a product was extracted
with ethyl acetate. The organic layer was separated and washed
with water. This organic layer was dried and concentrated to
give 6.1 g of (3,5-di-tert-butyl-4-hydroxybenzaldehyde)-1,3-
dioxolane as white crystals. The crystals (5.0 g) were dissolved
with THF (5 mL) and DMF (5 mL), and this solution was
slowly added dropwise to a stirred suspension of NaH (60% in
mineral oil, 0.80 g, 20 mmol) in THF (5 mL) with ice-cooling
and stirred for 15 min at 0 °C. Iodomethane (10.2 g, 71.9 mmol)
was added to the reaction mixture at 0 °C, and stirring was
continued for 1 h. The reaction mixture was poured into
saturated NaHCO3 and extracted with ethyl acetate. The organic
layer was separated and concentrated under reduced pressure.
Acknowledgment
We thank Dr. Hideki Tsujisita, Dr. Yasuyuki Hiramatsu, and
Mr. Yusuke Sakou for helpful discussions on the molecular
mechanical calculation.
Supporting Information Available
1
HPLC chart of S-2474 1; copies of H NMR spectra of
S-2474 1; copies of H and 13C NMR spectra of (5R*,1′R*)-
1
5-[3,5-di-tert-butyl-4-hydroxyphenyl)hydroxymethyl]-2-ethyl-
1,2-isothiazolidine-1,1-dioxide 12a (threo), (5S*,1′R*)-5-[3,5-
di-tert-butyl-4-hydroxyphenyl)hydroxymethyl]-2-ethyl-1,2-
isothiazolidine-1,1-dioxide 12b (erythro), 5-[(3,5-di-tert-butyl-
4-methoxyphenyl)hydroxymethyl]-2-ethyl-1,2-isothiazolidine-
1,1-dioxide 15. This material is available free of charge via the
Internet at http://pubs.acs.org.
Received for review January 15, 2008.
OP800008W
446
•
Vol. 12, No. 3, 2008 / Organic Process Research & Development