Isomerization-free sulfonylation and its application in the synthesis of PHA-565272A

Article abstract of DOI:10.1021/op050208a

FeCl₃ catalyzed an isomerization-free Friedel-Crafts sulfonylation between 1-naphthalenesulfonyl chloride and halobenzenes. The coupled halide was then displaced using 35% hydrazine in DMSO to provide the Fischer indole precursor. Pure 5-chloro-2-pentanone was the key for a successful Grandberg modification of Fischer indole synthesis that effectively constructed both the indole core and side chain of the target molecule. The development of these methods enabled a rapid preparation of kilogram quantities of PHA-565272A.

Full text of DOI:10.1021/op050208a

Organic Process Research & Development 2006, 10, 334−338
Isomerization-Free Sulfonylation and Its Application in the Synthesis of
PHA-565272A
Thomas J. Fleck,§ Jiong Jack Chen,† Cuong V. Lu,* and Kari J. Hanson
Chemical Process Research and DeVelopment, Pfizer, Kalamazoo, Michigan 49001, U.S.A.
Abstract:
a significant amount of 1,2-isomerization of sulfonyl group
on the naphthalene ring (Table 1). The 1,2-isomerization of
naphthalene ketone is well-known and well studied.5,6
Surprisingly, the 1,2-isomerization of naphthalene sulfone
was reported only once in a case of bis-naphthalene sulfone.7
Therefore, a number of conditions and Lewis acids were
screened. Representative results are shown in Table 1.
Most Lewis acids efficiently mediated para-sulfonylation
of halobenzenes. Interestingly, FeCl3 produced a minimal
amount of the naphthalene 2-isomer (12a-c) (Table 1,
Entries 1, 3, 5 and 6), while GaCl3 caused significant
naphthalene isomerization (Table 1, Entries, 2, 4, 7 and 8).
Other Lewis acids also caused isomerization with poor yields
(Table 1, Entries 9, 10 and 12) or produced messy reaction
mixtures (Table 1, Entries 11, 13 and 14). The dramatic
difference between FeCl3 and other Lewis acids led us to a
further study of the isomerization.8
Entries 5-8 in Table 1 shows that the isomerization
proceeds more extensively at higher temperatures. This
suggested that the isomerization was a thermodynamic
process. We tested this hypothesis via sulfonylating fluo-
robenzene with both 1- and 2-naphthalenesulfonyl chloride
using 1 eq. of FeCl3 at several temperatures (Table 2). Entries
1-3 of Table 2 confirmed the observation that high tem-
perature caused more isomerization (Table 1, Entries 5-8).
Entries 4-6 of Table 2 further demonstrated that compound
12a is the thermodynamically more stable product. This result
is similar to that of naphthaleneketones.5
FeCl3 catalyzed an isomerization-free Friedel-Crafts sulfonyl-
ation between 1-naphthalenesulfonyl chloride and halobenzenes.
The coupled halide was then displaced using 35% hydrazine
in DMSO to provide the Fischer indole precursor. Pure
5-chloro-2-pentanone was the key for a successful Grandberg
modification of Fischer indole synthesis that effectively con-
structed both the indole core and side chain of the target
molecule. The development of these methods enabled a rapid
preparation of kilogram quantities of PHA-565272A.
Introduction
PHA-565272A (8) was a CNS drug candidate that
required kilogram quantity for preclinical studies. The initial
synthesis1 of compound 8 had several potential scale-up
problems (Scheme 1). The first three steps appeared straight-
forward and could be run on a large scale. The use of nitrite
in step 4 would require extensive safety evaluation. The use
of SnCl2 could make it very difficult to remove tin byprod-
ucts from the reactors. Although the Grandberg modification
of Fischer indole synthesis2 was very efficient in constructing
both the indole core and its side chain, poor yields suggested
that significant development work would be needed. Fur-
thermore, the cost and availability of 1-naphthalenethiol 1
presented a long-term cost of goods and sourcing issue.
Results and Discussions
The steps that led to the Fischer precursor 6 in the initial
route (Scheme 1) were quite lengthy. The limited availability
of 1-naphthalenethiol (1) further made this route unpractical.
Thus, a Friedel-Crafts sulfonylation was proposed to react
halobenzenes (9) and naphthalene-1-sulfonyl chloride (10)
to produce a precursor that may lead to hydrazine 6 (Scheme
2). The bulky naphthalene ring is likely to preclude the
formation of the undesired meta isomer of halobenzenes. It
was decided to simultaneously test various Lewis acids with
fluoro-, chloro- and bromobenzene (9a-c) (Scheme 3). If a
direct displacement of halides was not successful, chloro-
and bromobenzene (9b&c), especially bromide, would permit
a Pd-mediated hydrazone formation developed by Buchwald
and Hartwig.3,4 Immediately, we were surprised to discover
Faster conversion and higher yields were achieved using
fluorobenzene for this sulfonylation than chlorobenzene and
bromobenzene, because fluorobenzene is the most active
species for Friedel-Crafts reaction. Furthermore, the lower
boiling point of fluorobenzene as solvent facilitated a
convenient workup and isolation. The amount of fluoroben-
zene could be reduced to 2-2.5 eq. with no loss in yield,
though the reaction time was slightly longer. The reaction
could be performed in a combination of fluorobenzene and
CH2Cl2; however, the yield is slightly lower, and the reaction
requires a longer time. This reaction was readily scaled to
(3) Wagaw, S.; Yang, B. H.; Buchwald, S. L. J. Am. Chem. Soc. 1999, 121,
10251-10263.
(4) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2090-2093.
(5) Dowdy, D.; Gore, P. H.; Waters, D. N. J. Chem. Soc., Perkin Trans. 2
1991, 1149-1159.
(6) Pivsa-Art, S.; Okuro, K.; Miura, M.; Murata, S.; Nomura, M. J. Chem. Soc.,
Perkin Trans. 1 1994, 1703-1707.
(7) Yoshii, Y.; Ito, A.; Manabe, O. Nippon Kagaku Kaishi 1973, 2395-2397.
(8) Chen, J. J.; Fleck, T. J.; Lu, C. V.; Hanson, K. J. 224th ACS National
Meeting, August 18-22, Boston, MA; American Chemical Society:
Washington, DC, 2002.
* To whom correspondence should be addressed. E-mail: cuong.v.lu@
pfizer.com. Telephone: 860-686-1826. Fax: 860-441-5779.
† Current address: GlaxoSmithKline, 709 Swedeland Rd., King of Prussia,
PA 19406.
§ Deceased.
(1) Fu, J.-M. In PCT Intl WO 03011284, 2003.
(2) Hugel, H. M.; Kennaway, D. J. Org. Prep. Proced. Int. 1995, 27, 1-31.
334
•
Vol. 10, No. 2, 2006 / Organic Process Research & Development
10.1021/op050208a CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/07/2006

Scheme 1. Initial synthesisa
a Reagents and conditions: a) K2CO3, CH3CN, rt, 66%; b) oxone, H2O/MeOH/THF, 77%; c) H2, 5% Rh/C, MeOH, 92%; d) i. NaNO2/HCl, ii. SnCl2, HCl; e)
MeOH/H2O, reflux, 6% crystal, 16% column.
Table 1. Sulfonylationa with various Lewis acidsb
reaction
temp
yield of
yield of
reaction
temp
yield of
yield of
1-isomer 2-isomer
1-isomer 2-isomer
entry Ar-X Lewis acid
(°C)
(%)c
(%)
entry Ar-X
Lewis acid
GaCl3
TfOH
TfOH
cat. Sn(OTf)2
cat. Hf(OTf)4
BiCl3
(°C)
(%)c
(%)
1
2
3
4
5
6
7
9a
9a
9b
9b
9c
9c
9c
FeCl3
GaCl3
FeCl3
GaCl3
FeCl3
FeCl3
GaCl3
60
60
60
60
60
80
60
98
73
92
82
93
89
60
1
26
1
17
0
2
15
8
9c
9c
9c
9c
9c
9c
9c
80
80
130
150
150
150
60
36
68
65
57
52
35
30
30
5
12
0
28
0
0
9
10
11
12
13
14
d
d
ZrCl4
a The reaction was run on 1 mmol scale using halobenzenes (9a-c) as solvent and 1 equiv of Lewis acid unless stated otherwise. The reaction was monitored using
GC. Endpoint was determined based on the consumption of compound 10, or was called complete after 24 h. b AlCl3 caused an emulsion upon sampling and was
messy. ZnCl2 and BF3‚OEt2 did not induce sulfonylation. c GC yields. d Catalyst loading was 10 mol %.
Scheme 2. Sulfonylation and isomerization
was readily scaled to one-kilogram scale to provide 6 in a
92% yield (93% HPLC area purity). Attempts to purify
compound 6 by crystallization resulted in fair yields and no
apparent improvement in purity by HPLC. Since compound
6 could not be readily upgraded and conditions were found
to effect high-yield formation of PHA-565272A (8) (vide
infra), further crystallization studies were not pursued. Up
to this point, compound 6, the hydrazine penultimate, was
prepared in only two steps from readily available starting
materials (Scheme 3).
Although the Grandberg modification initially produced
8 in poor yields, the efficiency in constructing the indole
ring and its side chain was incomparable. The Grandberg
modification offered better opportunity for scale-up purpose.
Due to the poor regioselectivity in the Grandberg modifica-
tion, the purity of commercial 5-chloro-2-pentanone (7) was
immediately called into question. The black liquid appeared
to be very acidic. 1H NMR spectrum of compound 7 showed
a large impurity with a peak at 7 ppm. After aqueous K2CO3
treatment, CH2Cl2 extraction, and solvent removal, this
one kilogram. The product 11a was isolated from 2-propanol
as a white crystalline solid in a 93% yield.
The hydrazine displacements of compounds 11a-c pro-
ceeded smoothly. Aqueous hydrazine (35%) was chosen for
this displacement because of safety and availability. An
organic solvent was needed for this reaction since compounds
11a-c were not soluble in 35% aqueous hydrazine. A
number of solvents were screened, but only a combination
of DMSO/35% aqueous hydrazine gave a complete and clean
conversion. Not surprisingly, the fluoro-substituted 11a was
converted at 60 °C in a day, while the bromo analogue 11c
required 3 days of heating at 100 °C. Product isolation
involved a simple addition of water and a filtration to provide
compound 6 as a white solid in 88% yield. This reaction
1
unidentified impurity was removed. Based on the H NMR
integration and our isolated yield of compound 7, 80-90%
of the contents of the commercial material was this impurity!
Due to a lack of commercial pure compound 7, it was
prepared in-house starting from 2-acetylbutyrolactone (Scheme
4).9 Unlike the “crude” commercial material, the distilled
Vol. 10, No. 2, 2006 / Organic Process Research & Development
•
335

Scheme 3. Scale-up routea
a Reagents and conditions: a) FeCl3, 93%; b) 35% hydrazine/DMSO, 92%; c) EtOH/H2O, reflux, then crystallization, 74%.
a
Table 2. Sulfonylation of fluorobenzene (9a) using FeCl3
the 1.5%-2.4% level when the reactions were called com-
plete. After distillation, cooling, and product filtration, this
impurity level would generally drop to 1.2%-1.5%. This
impurity was later confirmed by LC-MS not to be the
hydrazone intermediate, but rather an impurity with a
chlorine on the naphthalene. This impurity could derive from
two possible sources. It could potentially arise from the
1-naphthalenesulfonyl chloride 10 or it could derive from
chlorination during Step 1 FeCl3-induced sulfonylation. A
recent paper showed that aryl chlorination might occur in
sulfonylation reactions in the presence of AlCl3.11 This
suggested that a close monitoring of the Friedel-Crafts
sulfonylation may be needed to reduce this impurity.
Several different conditions to recrystallize PHA-565272A
(8) were probed. Acetone and EtOAc did not efficiently
dissolve compound 8 at reflux. Recrystallization would not
occur from neat n-BuOH; however, when H2O was added,
the solids dissolved. Improved yields were achieved with
no loss in HPLC purity when partial distillation of the
n-BuOH:H2O solution occurred prior to cooling and filtration.
An unfortunate side effect of this recrystallization procedure
was the formation of greater color in the crystals. Prior to
recrystallization the crystals were tan. However, after re-
crystallization there was a definite brownish tint. To remove
the color, the solution of compound 8 in 20% H2O:n-BuOH
at 95 °C was treated with Ecosorb 793 carbon and then
filtered hot over Celite 545. After product clarification,
distillation, and filtration, crystals of light tan were achieved.
The Grandberg modification and the subsequent recrystal-
lization were performed on kilo-scale, which provided ∼1
kg of PHA-565272A (8) with an overall 74% yield in 98.8%
HPLC area purity along with the 6.9 min (HPLC peak)
impurity at <0.8%.
arylsulfony chloride
used
temp
(°C)
yield of
11a (%)
yield of
12a (%)
entry
1
2
3
4
5
6
10
10
10
20
40
60
20
40
60
98.5
97.6
92
0
0
0
0.25
2
6b
2-Naph-SO2Cl
2-Naph-SO2Cl
2-Naph-SO2Cl
98.9
98.9
98.9
a Reaction was run in fluorobenzene with 1 equiv of FeCl3. b This is a result
of a run >2 h.
Scheme 4. Preparation of 7a
a Reagents and conditions: a) 37% HCl, H2O, NaCl, reflux; b) hexane
extraction.
compound 7 prepared in-house was clear in color. With purer
supplies of 5-chloro-2-pentanone (7) now available, the
Grandberg modification was revisited. A recent review of
Fischer indole synthesis noted an increased use of thermal
conditions to effect this transformation, especially with
heteroaromatic hydrazines that were difficult to cyclize under
acidic conditions due to protonation of the heteroatom.10
Several thermal conditions were therefore probed. Using only
1.2 eq. of this pure 5-chloro-2-pentanone 7 prepared in-house,
reactions were performed in 9:1 MeOH:H2O at 60 °C, 9:1
EtOH:H2O at 80 °C, and 9:1 ethylene glycol:H2O at 100
°C. For each reaction, compound 6 was converted almost
immediately to an intermediate, probably the hydrazone.
After 3 days, the reaction using MeOH/H2O showed only
55.7% formation of PHA-565272A (8) with 26.1% of the
intermediate still present. The reaction using EtOH/H2O was
much quicker with only 2.4% unreacted intermediate after
48 h. The reaction using ethylene glycol was not as clean,
generating 12% of an unknown impurity and other impurities.
Importantly, the yields from this quick screen all appeared
to be >50%.
Conclusion
An isomerization-free sulfonylation was developed for
1-naphthalenesulfonyl chloride. This development allowed
a much shorter route for the preparation of PHA-565272A.
We discovered that the Grandberg modification of Fischer
indole synthesis required pure 5-chloro-2-pentanone to
achieve good yields. The new route was rapidly scaled to
one kilogram of PHA-565272A uneventfully.
This reaction was then scaled to 20 g in EtOH/H2O. When
the reaction was near completion, product solids formed in
the reaction. A distillation to a lower volume resulted in an
89% yield of the desired product. The 1H NMR and elemental
analysis for the product isolated confirmed the formation of
the mono-HCl salt.
Experimental Section
General Procedures. All reagents were commercially
obtained and used as received unless otherwise noted. All
nonaqueous reactions were performed in dry glassware under
an atmosphere of dry nitrogen. The drying of solids was
A material at 6.9 min (HPLC peak) with a similar
retention time as the hydrazone intermediate was present at
(9) Medina, F.; Manjarrez, A. Tetrahedron 1964, 20, 1807-10.
(10) Hughes, D. L. Org. Prep. Proced. Int. 1993, 25, 607-32.
(11) Hyatt, J. A.; White, A. W. Synthesis 1984, 214-7.
336
•
Vol. 10, No. 2, 2006 / Organic Process Research & Development

performed in a vacuum oven. NMR spectra were measured
was heated to 101 °C with distillate sent into the 4 L
separatory funnel. 2-acetylbutyrolactone (1000 g, 7.8 mol)
was slowly added to the 12-L flask over 8 h. Distillation
was continued for 2 h more while increasing the temperature
to 110-115 °C. Hexane (1840 mL) was added to the 4-L
separatory funnel, stirred, and the lower aqueous layer was
separated. H2O (800 mL) was added to the 4-L separatory
funnel, and 10 mL of the NaHCO3 solution was added to
adjust the pH to 7.2. The lower aqueous phase was separated,
and the hexane layer was washed again with H2O (800 mL).
The aqueous phases were combined and back extracted with
hexane (500 mL). The hexane extractions were combined,
washed with H2O (400 mL), and distilled atmospherically
to remove hexane. Vacuum distillation afforded a forerun
of 100 mL followed by 655 g (70%) of 5-chloro-2-pentanone
7 of 99.9% GC area purity.
1
on a Bruker AM-400 operated at 400 and 100 MHz for H
and 13C, respectively, with data reported as follows: chemical
shift (ppm), multiplicity (s, singlet; d, doublet; t, triplet; q,
quartet; m, multiplet), integration and coupling constant (J,
1
Hz). H - 13C Multiplicities are reported on the basis of
DEPT data. Where 13C -19F coupling occurs, the coupling
constant (JC-F) is reported. UV, IR and HRMS and elemental
analysis were conducted within Pfizer analytical support
groups. High-Pressure Liquid Chromatography (HPLC) were
performed on an Agilant 1100 Series HPLC System com-
posed of a G1316A Column Compartment, G1315A Diode
Array Detector, G1313A Automated Liquid Sampler, G1311A
Quaternary Gradient Pump and a G1322A Vacuum Degasser.
The conditions are as following: Solvent A; 2 L H2O, 10
mL TEA, 5 mL AcOH, Solvent B; CH3CN, Tinit ) 0 min,
35% A; Tfin ) 10 min, 35% A, Zorbax RX-C8 Column, 4.6
mm × 25 mm, PN880967, UV ) 254 nm, Flow rate ) 1.0
mL/min, Col T ) 40 °C. Gas Chromatography (GC) was
performed on a system composed of a Hewlett-Packard 6890
Series GC System and Integrator with a Hewlett-Packard
7683 Series Injector. Conditions are: Inject 50µL of the
resulting solution onto a 15 m capillary DB-1 column using
the following run parameters: Tinj ) 200 °C, Tinit ) 50 °C,
tinit ) 5 min, Tfin ) 300 °C, tfin ) 10 min, Rate ) 20 °C
/min.
2-Methyl-5-(1-naphthalenylsulfonyl)-1H-indole-3-eth-
anamine monohydrochloride (8). To a 22-L flask, com-
pound 6 (1090 g, 3.7 mol), 5-chloro-2-pentanone 7 (500 mL,
4.4 mol), EtOH (12 L) and H2O (1.15 L) were added. The
slurry was heated to reflux at which point all the solids
dissolved. The solution was filtered through a 0.6 micron
filter sending the filtrate into a second 22-L flask (GMP
requirement to remove possible contaminated particles). The
solution was stirred at reflux (78 °C) for 4 days at which
point the reaction was deemed complete with 1.93% of an
intermediate/impurity at 6.9 min (HPLC). Distillation was
performed to a volume of 5-6 L. The resulting mixture was
cooled to -35 °C to -27 °C, stirred 100 min, and filtered
onto a 9 L coarse fritted filter. The solids were washed with
cold EtOH (1 L) and then dried at 60 °C for 24 h to give
1296 g (88%) of PHA-565272A (8) with 95.7% HPLC area
purity in beige color. This material (500 g) was added to a
22-L flask, followed with n-BuOH (12 L), and H2O (3 L).
The slurry was heated to 95 °C at which point all the solids
dissolved. Ecosorb 793 (a type of carbon, 250 g) was added
and the slurry was stirred for 1 h. The slurry was poured
onto a 9 L coarse fritted filter containing Celite 545 (500
g). The filtrate was filtered through a 0.6 micron ultipore
filter sending the filtrate into a second 22-L flask. Distillation
was performed to a volume of 11-12 L. The resulting filtrate
was cooled to -29 °C, stirred 3 h, and filtered onto a 9-L
coarse fritted filter. The solids were rinsed with cold n-BuOH
(1 L) and dried at 60 °C for 3 days to give 418 g (84%) of
PHA-565272A (8) with 98.8% HPLC area purity. mp. 363-
1-[4-(1-naphthalenylsulfonyl)phenyl]hydrazine (6). To
a 22-L flask, compound 11a (1170 g, 4.1 mol), 35% aqueous
hydrazine (743 mL, 8.2 mol), and DMSO (5.1 L) were added.
The resulting slurry was heated to 58 °C (The contents went
into solution at 55 °C). After 24 h, the reaction was deemed
complete with 2.3 area% starting material (GC) remaining.
The solution was cooled to 16 °C. Water (28 L) was added
to the 40 L wash tank followed by the reaction mixture. The
reaction flask was rinsed with H2O (5 L) and added to the
wash tank. The resulting slurry was filtered onto the 9-L
coarse fritted filter. The wash tank and filter were rinsed
with H2O (10 L). The solids were dried at 60 °C for 5 days
to yield 1120 g (92%) of compound 6 with 93.2% HPLC
area purity. The yield took into account the presence of 6.6%
1
H2O. mp. 129.2-131.4 °C. H NMR (400 MHz, DMSO-
d6) δ 8.62 (d, 1H, J ) 8.7 Hz), 8.31 (d, 1H, J ) 7.1 Hz),
8.22 (d, 1H, J ) 8.2 Hz), 8.05 (d, 1H, J ) 8.1 Hz), 7.76 (s,
1H), 7.71-7.58 (m, 5H), 6.80 (d, 2H, J ) 8.6 Hz), 4.22 (s,
2H). 13C NMR (100 MHz, DMSO- d6) δ 155.88 (s), 137.51
(s) 134.30 (d), 133.8 (s), 129.12 (d), 128.41 (d), 128.05 (d),
127.30 (s), 126.82 (d), 125.35 (s), 124.80 (d), 124.04 (d),
110.03 (d). IR (diffuse reflectance) 3367, 1940 (w), 1902
(w), 1591 (s), 1288 (s), 1151 (s), 1145 (s), 1132 (s), 1082,
834, 822, 802, 768 (s), 718, 698, cm-1; HRMS (FAB) calcd
for C16H14N2O2S +H 299.0854, found 299.0847. Anal. Calcd
for C16H14N2O2S: C, 64.41; H, 4.73; N, 9.39; S, 10.74.
Found: C, 63.46; H, 4.92; N, 8.85; S, 10.52.
5-Chloro-2-pentanone (7). NaHCO3 (200 g) was dis-
solved in H2O (2 L) and set aside. To the 4 L separatory
funnel was added H2O (65 mL) and hexane (65 mL). To a
12-L flask, NaCl (230 g) and H2O (1215 mL) were added.
37% HCl (1178 mL, 14.3 mol) was slowly added at 20 °C
to 50 °C over 15 min from a 2 L addition funnel. The solution
1
368 °C. H NMR (400 MHz, DMSO- d6) δ 11.65 (s, 1H),
8.73 (d, 1H, J ) 8.7 Hz), 8.41 (d, 1H, J ) 7.6 Hz), 8.36 (s,
1H), 8.24 (d, 1H, J ) 8.2 Hz), 8.3-8.1 (br s, 3H), 8.03 (d,
1H, J ) 8.1 Hz), 7.72 (t, 1H, J ) 7.6 Hz), 7.69 (t, 1H, J )
8.0 Hz), 7.58 (t, 1H, J ) 7.3 Hz), 7.47 (d, 1H, J ) 8.7 Hz),
7.39 (d, 1H, J ) 8.1 Hz), 3.07 (t, 2H, 7.6 Hz), 2.92 (t, 2H,
J ) 7.6 Hz), 2.36 (s, 3H). 13C NMR (100 MHz, DMSO- d6)
δ 137.28 (s), 137.16 (s), 136.50 (s), 134.62 (d), 133.78 (s),
130.47 (s), 129.13 (d), 128.75 (d), 128.21 (d), 127.28 (s),
126.88 (d), 124.89 (d), 124.12 (d), 119.29 (d), 117.93 (d),
111.41 (d), 107.34 (s), 39.49 (t), 21.66 (t), 11.25 (q). IR
(mull) 3050, 3012 (b), 1988 (w), 1951 (w), 1938 (w), 1506,
1300 (s), 1155 (s), 1136 (s), 1131 (s), 1104, 805, 768 (s),
638, 614, cm-1; UV λmax 232 (60200, 95% Ethanol); HRMS
Vol. 10, No. 2, 2006 / Organic Process Research & Development
•
337

(FAB) calcd for C21H20N2O2S +H 365.1324, found 365.1315;
Anal. Calcd for C21H21N2O2SCl: C, 62.91; H, 5.28; N, 6.99;
Cl, 8.84; S, 8.00. Found: C, 62.57; H, 5.32; N, 6.93; Cl,
8.82; S, 7.92.
1-[(4-chlorophenyl)sulfonyl]naphthalene (11b): A )
11.3, B ) 50, C ) 40, D ) 5, 11.6 g (81%), 1H NMR (400
MHz, DMSO-d6) δ 8.48 (m, 2H), 8.33 (d, 1H, J ) 8.1 Hz),
8.08 (d, 1H, J ) 8.1 Hz), 7.98 (d, 2H, J ) 8.1 Hz), 7.77 (t,
1H, J ) 7.6 Hz), 7.60-7.69 (m, 4H). Anal. Calcd for
C16H11ClO2S: C, 63.47; H, 3.66; N, 0.00. Found: C, 63.32;
H, 3.63; N, 0.15.
1-[(4-bromophenyl)sulfonyl]naphthalene (11c): A )
11.3, B ) 50, C ) 50, D ) 3, 12.4 g (72%), 1H NMR (400
MHz, DMSO-d6) δ 8.47 (d, 2H, J ) 8.6 Hz), 8.33 (d, 1H,
J ) 8.2 Hz), 8.09 (d, 1H, J ) 7.5 Hz), 7.87-7.92 (m, 2H),
7.74-7.80 (m, 3H), 7.60-7.70 (m, 2H). Anal. Calcd for
C16H11BrO2S: C, 55.35; H, 3.19; N, 0.00. Found: C, 55.01;
H, 3.11; N, 0.16.
2-[(4-fluorophenyl)sulfonyl]naphthalene (12a): A )
11.3, B ) 30, C ) 40, D ) 1, 12.7 g (89%), 1H NMR (400
MHz, DMSO-d6) δ 8.71 (s, 1H), 8.21 (d, 1H, J ) 7.6 Hz),
8.09-8.14 (m, 3H), 8.03(d, 1H, J ) 8.2 Hz), 7.67-7.74 (m,
2H), 7.45 (t, 2H, J ) 9.1 Hz). 19F NMR (400 MHz, DMSO-
d6) δ 105.12. Anal. Calcd for C16H11FO2S: C, 67.12; H, 3.87;
N, 0.00. Found: C, 66.90; H, 3.91; N, 0.14.
2-[(4-chlorophenyl)sulfonyl]naphthalene (12b): A )
11.3, B ) 50, C ) 40, D ) 2, 12.9 g (90%), 1H NMR (400
MHz, DMSO-d6) δ 8.72 (s, 1H), 8.21 (d, 1H, J ) 7.6 Hz),
8.14 (d, 1H, J ) 8.6 Hz), 8.01-8.05 (m, 3H), 7.90 (d, 1H,
J ) 8.7 Hz), 7.68-7.75 (m, 4H). Anal. Calcd for C16H11-
ClO2S: C, 63.47; H, 3.66; N, 0.00. Found: C, 63.01; H,
3.61; N, 0.15.
1-[(4-fluorophenyl)sulfonyl]naphthalene (11a). To a
22-L flask, 1-naphthalenesulfonyl chloride 10 (1004 g, 4.4
mol) and fluorobenzene 9a (1061 g, 11 mol). The slurry was
heated to 45 °C. FeCl3 (856 g, 5.3 mol) was added to the
22-L flask over 30 min while maintaining the temperature
around 40 °C. After 70 min, GC analysis showed the reaction
was complete. The reaction mixture was cooled and quenched
with 4 L of 1M HCl. The mixture was extracted with 5 L of
CH2Cl2. The aqueous layer was back extracted with CH2Cl2
(2 × 2 L). The combined CH2Cl2 extractions were washed
with 1M HCl (4 L). The aqueous layer was back extracted
with CH2Cl2 (1 × 2 L). The combined organic layers were
distilled and swapped to 2-propanol (7.2 L). The slurry was
heated to reflux to dissolve the solids, cooled to -15 °C,
stirred 30 min, and filtered onto a 9-L coarse fritted filter.
The solids were washed with cold 2-propanol (2 L) and dried
at 60 °C for 24 h to give 1172 g (93%) of compound 11a
with 98.7% GC and 94.9% HPLC area quality. mp 127.8-
1
128.1 °C. H NMR (400 MHz, CDCl3) δ 8.61 (d, 1H, J )
8.1 Hz), 8.50 (d, 1H, J ) 6.6 Hz), 8.10 (d, 1H, J ) 7.6 Hz),
7.98 (br s, 2H), 7.90 (d, 1H, J ) 7.7 Hz), 7.63-7.54 (m,
3H), 7.13 (t, 2H, J ) 7.7 Hz). 13C NMR (100 MHz, CDCl3)
δ 136.88 (s, JC-F ) 223.8 Hz), 134.38 (s), 132.83 (s), 130.94
(s), 128.47 (s) 135.7 (d), 130.37 (d), 130.27 (d), 130.11 (d),
129.28 (d), 129.61 (d), 127.11 (d), 124.40 (d, JC-F ) 26.1
Hz), 116.55 (d, JC-F ) 22.9 Hz). IR (diffuse reflectance)
2447 (w), 2413 (w), 2295 (w), 2215 (w), 2041 (w), 1312
(s), 1227 (s), 1161 (s), 1136 (s), 1131 (s), 1083 (s), 847 (s),
800 (s), 775 (s), 690 (s), cm-1; HRMS (FAB) calcd for
C16H11FO2S +H 287.0542, found 287.0548; Anal. Calcd for
C16H11FO2S: C, 67.12; H, 3.87; S, 11.20; Found: C, 66.85;
H, 3.91; N, 0.20; S, 11.09.
2-[(4-bromophenyl)sulfonyl]naphthalene (12c): A )
11.3, B ) 50, C ) 40, D ) 3, 12.1 g (70%), 1H NMR (400
MHz, DMSO-d6) δ 8.71 (s, 1H), 8.21 (d, 1H, J ) 8.1 Hz),
8.13 (d, 1H, J ) 8.6 Hz), 8.03 (d, 1H, J ) 8.2 Hz), 7.93 (d,
2H, J ) 8.7 Hz), 7.89 (d, 1H, J ) 8.7 Hz), 7.82 (d, 2H, J )
8.6 Hz), 7.66-7.74 (m, 2H). Anal. Calcd for C16H11BrO2S:
C, 55.35; H, 3.19; Br, 23.01; N, 0.00. Found: C, 55.28; H,
3.23; N, 0.19.
Synthesis of compounds 11b, 11c, 12a, 12b, and 12c:
General procedure. To a solution of 1- naphthalenesulfonyl
chloride (10) or 2-naphthalenesulfonyl chloride (1 eq. A g)
in the corresponding halobenzene (B mL), FeCl3 (1.2 eq.)
was added in one portion at C °C. After D h, the dark solution
was quenched with 2 × B mL of CH2Cl2 and 1 × B mL of
1M HCl. The aqueous layer was extracted with 1 × B mL
of 1M HCl. The organic layers were concentrated to give a
yellow solid, which was then re-recrystallized in IPA to give
desired product.
Acknowledgment
This paper is dedicated to the memory of Thomas Fleck,
a passionate process chemist and a dear friend. We thank
Stephen Grode, Mark Mowery, and Dave Russell for
analytical support and Bruce Pearlman for the reference to
prepare 5-chloro-2-pentanone.
Received for review October 19, 2005.
OP050208A
338
•
Vol. 10, No. 2, 2006 / Organic Process Research & Development