A scalable route to an unusual 3,3-dimethyl-2,3-dihydrobenzofuran ring system present in an HCV drug candidate

Article abstract of DOI:10.1021/op4003467

A scalable synthesis of a key intermediate used for the preparation of an HCV inhibitor containing an unusual dimethyldihydrobenzofuran ring is described. A key element for the successful completion of the synthesis was the correct ordering of a sequence of bromination, chlorination, and methylation to provide optimized selectivity and improved yield. A tin hydride-mediated ring closure was replaced with a more environmentally benign sulfuric acid-catalyzed Friedel-Crafts reaction. The overall yield for the preparation of the key intermediate was increased from less than 5 to 40%.

Full text of DOI:10.1021/op4003467

Technical Note
pubs.acs.org/OPRD
A Scalable Route to an Unusual 3,3-Dimethyl-2,3-dihydrobenzofuran
Ring System Present in an HCV Drug Candidate
Ping Wang and Andrew J. Briggs*
Process Research and Synthesis, Hoffmann-La Roche, Inc., 340 Kingsland Street, Nutley, New Jersey 07110, United States
S
* Supporting Information
ABSTRACT: A scalable synthesis of a key intermediate used for the preparation of an HCV inhibitor containing an unusual
dimethyldihydrobenzofuran ring is described. A key element for the successful completion of the synthesis was the correct
ordering of a sequence of bromination, chlorination, and methylation to provide optimized selectivity and improved yield. A tin
hydride-mediated ring closure was replaced with a more environmentally benign sulfuric acid-catalyzed Friedel−Crafts reaction.
The overall yield for the preparation of the key intermediate was increased from less than 5 to 40%.
seen with either 1-bromo-2,4-difluoro-5-nitrobenzene (13) or
1-bromo-2,4-dimethoxy-5-nitrobenzene (14). Even elimination
of the electron-withdrawing nitro group, as in 15, failed to
provide any of the desired alkylation product (Scheme 2).
However, an approach based on Friedel−Crafts alkylation of
the less sterically encumbered, more electron-rich, and readily
available 4,6-dichlororesorcinol (16) was successfully developed
(Scheme 3).
INTRODUCTION
■
Compound 12, a potent inhibitor of hepatitis C virus (HCV)
NS5B polymerase, was a potential development candidate for
the treatment of HCV.1 The conversion of aniline 11, the final
intermediate, to 12 had previously been developed and
demonstrated to proceed in yields of up to 94% on a g-scale
by our Discovery Chemistry colleagues (Scheme 1).2 However,
the existing synthesis of 11 was extremely problematic above a
milligram scale.
A mixture of 16 and methallyl chloride (5.4 equiv) were
treated with sulfuric acid at 40−45 °C. This reaction was
exothermic and, with no applied cooling, a slow addition of
sulfuric acid was required to control the temperature.
Additional methallyl chloride, added in two equal portions,
followed by stirring overnight was needed for complete
reaction. Likely some competing polymerization of the
methallyl chloride under the strongly acidic conditions was
occurring.6 After the addition of water and cyclohexane, the
organic phase was treated with 4 N potassium hydroxide and
methanol to effect cyclization. Extraction with cyclohexane
removed some polymeric material, and acidification followed by
extraction provided crude 18 containing ∼7% of 19. Removal
of this impurity would be simplified if it could be converted to a
more polar derivative. Unfortunately, preliminary attempts to
derivatize the hindered alkyl chloride present in 19 were
unsuccessful. Consequently, the dichloride was purified via a
short plug of silica gel to provide 99% pure7 18 in ∼80% yield.
This separation was considered very manageable on a
multihundred gram scale, although a solution for pilot-plant
scale would need to take advantage of the subsequent discovery
that 18 was a solid. An alternative strategy for purification
would be to investigate crystallization of salts of 18 and 19.
Dechlorination of 18 using hydrogen, 10% palladium on
carbon, and sodium acetate (2.5 equiv) in methanol then gave
4. At atmospheric pressure, a loading of 22 w/w% of palladium
catalyst was required to ensure complete reaction (Table 1).
After removal of the catalyst by filtration, evaporation of the
The biggest issue in the Discovery Chemistry synthesis of 11
was the lack of regioselectivity in the nitration step, which
resulted in a 1:1 mixture of 7 and isomer 8 from isomerically
pure 6. Isomer 8 presumably arises from ipso-substitution of 6.3
This mixture was chromatographically difficult to separate, and
the isolated yield of 7 was only ∼30%. The lack of selectivity in
the methallylation step, which gave approximately a 1:2:1
mixture of 2, 3, and dialkylated resorcinol, also resulted in a low
yield of the desired product (40%). As a consequence, the
overall yield of 12 from 1 was less than 5%. From a scale-up
standpoint, the tributyltin hydride ring closure, the limited
commercial availability of tribromoresorcinol, the microwave
Suzuki reaction, and the four chromatographies were also
undesirable.
RESULTS AND DISCUSSION
■
With the need for larger quantities of 12, a new route was
required as it seemed unlikely that the nitration reaction issues
could be overcome. Our Discovery Chemistry colleagues had
found that only the copper nitrate/acetic anhydride conditions
provided any 7, and exploration of a number of the reaction
parameters had also failed to improve the selectivity.
Consequently, several approaches to construct the 3,3-
dimethyl-2,3-dihydrobenzofuran ring system were explored.
Other than the tin hydride cyclization route, the two main
alternative approaches described in the literature were a
palladium-catalyzed ring closure4 or a Friedel−Crafts alkylation
with subsequent ring closure under basic conditions.5 For the
palladium route, preparation of an appropriately substituted
substrate that could be elaborated into 10 was the main
challenge. In the Friedel−Crafts approach, no alkylation was
Received: December 6, 2013
Published: April 1, 2014
© 2014 American Chemical Society
656
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661

Organic Process Research & Development
Technical Note
Scheme 1. Discovery chemistry route
Scheme 2. Initial Friedel-Crafts alkylation substrates
Scheme 3. 3,3-Dimethyl-2,3-dihydro-4-hydroxybenzofuran
methanol was followed by aqueous work-up with MTBE to
provide 4 as an oil in quantitative yield.
Differential functionalization of the ortho- and para-positions
of 4 was explored next.
Table 1. Dechlorination of 18
Bromination of 4 was investigated in acetonitrile and
dichloromethane using N-bromosuccinimide (NBS; 1.0
equiv). The halogenated solvent was more selective and, at
0−5 °C, provided an 89:9:2 mixture of ortho-, para-, and di-
brominated products (Scheme 4, Table 2). The different
partition coefficients of the potassium salts of the components
of the reaction mixture permitted isolation of the desired
product 5. After partitioning the mixture in cyclohexane/
methanol/potassium hydroxide, 5 was isolated in 88% yield and
94% purity; the major impurities were 20 (4%) and 21 (1.6%).
Unfortunately, as with the nitration of 6, chlorination of 5
with N-chlorosuccinimide (NCS; 1.0 equiv) in dichloro-
methane resulted in significant scrambling (Scheme 4).
Chlorination of 6, prepared by methylation of 5, gave a similar
result.
a
b
entry
scale (g)
10% Pd/C (w/w %)
4 (%)
1
2
0.2
1.8
10
22
70
100
a
b
50% wet. HPLC area % of 4
Since the bromination−chlorination sequence was unsuc-
cessful, chlorination−bromination of 4 was then examined
(Scheme 5). Fortunately, treatment of 4 with NCS (1.0 equiv)
in dichloromethane at room temperature was highly
regioselective, and gave a 97:3 mixture of ortho- and para-
isomers 24 and 25. Chlorination in acetonitrile was less
selective and provided a 70:16:14 ratio of 24, 25, and 18. This
was advantageously used to confirm the selectivity of the
657
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661

Organic Process Research & Development
Technical Note
Scheme 4. Sequential bromination−chlorination of 4
C−Br and C−Cl bond has been measured for bromo- and
chloro-benzene.15
Table 2. Bromination of 4
a
entry
solvent
4 (%)
5 (%)
20 (%)
21 (%)
Elaboration of the substituted anisole 27 to the aniline 11,
the key intermediate in the Discovery Chemistry route, was
achieved with sequential Buchwald−Hartwig amination fol-
lowed by a Suzuki−Miyaura reaction (Scheme 6). Bromo-
chloranisole 27 was coupled with tert-butyl carbamate by
treatment with sodium tert-butoxide, Pd2(dba)3 and t-
BuXPhos.16 A satisfactory reaction was obtained using 5 mol
% palladium in toluene for 10 min at 80 °C (Table 3), which
minimized the loss of the tert-butyl group.17 Extractive work-up
and crystallization provided BOC protected aniline 29 in 77%
yield.
Boronate 9 was next prepared by first treating 6-bromo-2-
naphthylamine (31) with methanesulfonyl chloride and
pyridine. For solubility reasons, tetrahydrofuran was the best
solvent for this reaction, as compared to dichloromethane or
toluene.18 Methanesulfonamide 32 was isolated in >95% yield
after drowning into water, which was preferred to adding water.
The bromide was then converted to 9 by treatment with
pinacol diborane, potassium acetate, XPhos, and palladium
acetate in dioxane.19 After heating to reflux for 3 h, the reaction
was complete. Concentration to half volume, dilution with
heptane, and filtration provided 9 in ∼80% yield.
The Suzuki−Miyaura reaction of 29 with boronate 9 using
tripotassium phosphate, palladium acetate, and tricyclohexyl-
phosphine in aqueous DMF proceeded at 80 °C to give 30.
These conditions had been used successfully in several related
programs, although in this instance the reaction was more
sensitive to air and thorough degassing prior to heating was
critical. An excess of boronate 9 (1.3 equiv) was also required
since it undergoes slow protodeboronation to 33. Reducing the
level of potassium phosphate to 1.2 equiv slowed the reaction
1
2
CH2Cl2
MeCN
0.4
27
89
18
9
1.6
32
21
a
Normalized HPLC area % of 4, 5, 20, and 21
dichloromethane chlorination since for the minor isomer an
NOE was observed between the phenolic hydroxyl group and
the ortho hydrogen.8 Addition of NBS gave the differentially
substituted phenol 23.9 However, none of the methylation
conditions investigated subsequently provided 27; LC/MS and
1H NMR were consistent with formation of the dione 28.11
Methylation, followed by bromination, was therefore ex-
plored as our next strategy. Since it precluded telescoping the
halogenation reactions,12 this approach was initially considered
less attractive. Fortunately, it provided not only a successful
route to 27 but also an improvement in purity due to the faster
methylation of 24 over 25.13 During the work-up, the salt of
unreacted 25 remained in the aqueous DMF phase, while
product 26 was extracted into the heptane layer. After
concentration, chloroanisole 26 was obtained as an oil in
99% HPLC purity. Bromination with NBS (1 equiv) in
acetonitrile proceeded uneventfully to give 99% pure 27 as an
oil in 90% yield from 4. Thus, an efficient process to produce
the densely substituted key intermediate 27 was achieved by
the correct ordering of the chlorination, methylation, and
bromination steps.
Counterintuitively, bromination of chloride 24 resulted in no
halogen scrambling, in contrast to chlorination of bromide 5.14
One possible explanation could be the higher bond strength of
C−Cl versus C−Br; a difference of ∼13 kcal/mol between the
Scheme 5. 7-Bromo-5-chloro-4-methoxy-3,3-dimethyl-2,3-dihydrobenzofuran
658
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661

Organic Process Research & Development
Technical Note
Scheme 6. Buchwald−Hartwig and Suzuki−Miyaura sequence
to further development, leading to a process suitable for the
multikilogram scale.
Table 3. Buchwald−Hartwig Amination
scale
temp. t-BuXPhos Pd2(dba)3
27
(%)
29
(%)
a
entry (g)
solvent
(°C)
(equiv)
(equiv)
EXPERIMENTAL SECTION
1
2
3
0.1
0.1
1.0
toluene
xylenes
toluene
90
100
80
0.04
0.04
0.07
0.020
0.020
15
30
10
85
70
■
General Methods. Reactions were monitored by reverse
phase HPLC (Zorbax Eclipse XDB C8 column, 4.6 mm × 50
mm, 1.8 μm particle size, 0.5 mL/min flow, detection at 220
nm, 5−100% acetonitrile/water containing 0.1% TFA over 10
min). HRMS was performed using an Orbitrap MSD (flow
injection, 0.25 mL/min, detection at 250 nm, 80% acetonitrile/
water containing 0.1% formic acid). LC/MS was performed
using reverse phase HPLC with either a triple quadrupole
detector (Zorbax Eclipse XDB C8 column, 3.5 mm × 50 mm,
1.8 μm particle size, 1.0 mL/min flow, detection at 220 nm with
either a 5−100% or 30−100% MeCN/water containing 0.1%
formic acid gradient over 10 min) or a single quadrupole
detector (Zorbax Eclipse XDB C8 column, 3.5 mm × 50 mm,
1.8 μm particle size, 1.0 mL/min flow, detection at 220 nm with
either a 5−100% or 30−100% MeCN/water containing 0.1%
TFA gradient over 10 min).
b
0.025
b
90
a
HPLC area % normalized to 27 and 29. Isolated yield after
crystallization was 77%.
and resulted in high levels of the protodeboronated side
product, as well as dechlorinated 29. In anhydrous Me-THF,
protodeboronation was particularly troublesome, giving a
mixture containing 90% of 33 and only 20% of product 30.
Work-up of the aqueous DMF reaction with ethyl acetate was
followed with filtration of the organic phase through a plug of
silica gel to remove some color. As some loss of the BOC group
occurred during the coupling reaction, the deprotection step
was conducted directly on the ethyl acetate solution of 30 using
HCl at 50 °C. After cooling, the slurry was filtered to give the
aniline hydrochloride in 96% purity. The freebase was prepared
by dropwise addition of aqueous sodium bicarbonate (1.4
equiv) to a methanol solution of the aniline hydrochloride.
After partial concentration of the slurry, filtration and a water
wash provided pure aniline 11 in 80% yield from 29.
5,7-Dichloro-3,3-dimethyl-2,3-dihydrobenzofuran-4-ol
(18). To a mixture of 4,6-dichlorobenzene-1,3-diol (10 g, 55.9
mmol) and methallyl chloride (30 mL, 304 mmol) was added
H2SO4 (5.48 g, 2.98 mL, 55.9 mmol) over 30 min at 40−45 °C.
After 2 h, methallyl chloride (3 mL, 30.4 mmol) was added,
followed 1 h later by additional methallyl chloride (3 mL, 30.4
mmol). The reaction mixture was stirred overnight, and diluted
with cold water (50 mL) and cyclohexane (40 mL). The
organic layer was separated and extracted with a mixture of 4 N
KOH (30 mL) and MeOH (30 mL), which completed the
cyclization of 17 to 18. The aqueous phase was washed with
cyclohexane (40 mL) and acidified with conc. HCl at 5 °C.
Extraction with cyclohexane (100 mL, then 50 mL) was
followed by washing the combined cyclohexane extracts with
water (50 mL). Concentration gave an oil, which was purified
by silica gel chromatography eluting with 0−10% EtOAc/
hexanes to give 18 (10 g, 77% yield), which partially solidified
CONCLUSION
■
Aniline 11 was prepared in 40% overall yield from cheap,
commercially available dichlororesorcinol. The new route
eliminated the major problem associated with the difficult
chromatographic separation of 7 and 8. Although halogen
scrambling initially occurred, by first incorporating the chloro,
rather than the bromo substituent, this effect was completely
suppressed. The tributyltin hydride ring closure, originally used
to construct the dimethyldihydrobenzofuran ring, was replaced
with a more environmentally friendly sulfuric acid-promoted
Friedel−Crafts alkylation. This synthetic sequence was not
scaled beyond 1−10 g for most steps due to a change in the
overall program direction. The remaining operations of this
process do include some scale-challenged steps, such as the
chromatography to remove 19, but these should be amenable
1
upon standing: H NMR (300 MHz, DMSO-d6) 9.49 (br s,
1H), 7.24 (s, 1H), 4.29 (s, 2H), 1.38 (s, 6H); HRMS (ESI) m/
z [M − H]− calcd for C10H9O2Cl2 230.99741; Found
230.99878.
659
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661

Organic Process Research & Development
Technical Note
3,3-Dimethyl-2,3-dihydrobenzofuran-4-ol (4). A mixture of
18 (1.8 g, 7.72 mmol) and NaOAc (1.58 g, 19.3 mmol) in
MeOH (20 mL) was degassed with nitrogen. Then, 10% Pd/C
(0.4 g, 50% wet) was added, and the mixture was stirred under
an atmosphere of hydrogen for 20 h. The catalyst was removed
by filtration and washed with methanol, and the combined
filtrate/wash was concentrated. The residue was partitioned
between water and MTBE. The organic layer was separated,
washed with water, and concentrated to give 4 (1.3 g, 103%
yield) as an oil: 1H NMR (300 MHz, DMSO-d6) 9.35 (s, 1H),
6.86 (apparent t, J = 7.91 Hz, 1H), 6.28 (dd, J = 8.10, 0.75 Hz,
1H), 6.19 (dd, J = 7.91, 0.75 Hz, 1H), 4.11 (s, 2H), 1.34 (s,
6H); HRMS (ESI) m/z [M − H]− calcd for C10H11O2
163.07536; Found 163.07664.
5-Chloro-3,3-dimethyl-2,3-dihydrobenzofuran-4-ol (24).
To a solution of 4 (0.9 g, 5.48 mmol) in CH2Cl2 (14 mL) at
rt was added NCS (732 mg, 5.48 mmol) in portions over 10
min. After stirring overnight, the mixture was concentrated, and
the residue was partitioned between water (5 mL) and heptane
(10 mL). The organic layer was washed with water (2 × 5 mL)
and concentrated to give 24 (1.03 g, 95% yield) as a semisolid:
1H NMR (300 MHz, DMSO-d6) 9.18 (s, 1H), 7.05 (d, J = 8.48
Hz, 1H), 6.29 (d, J = 8.48 Hz, 1H), 4.18 (s, 2H), 1.37 (s, 6H);
HRMS (ESI) m/z [M − H]− calcd for C10H10O2Cl 197.03638;
Found 197.03772.
5-Chloro-4-methoxy-3,3-dimethyl-2,3-dihydrobenzofuran
(26). To a slurry of 24 (1 g, 5.03 mmol) and K2CO3 (904 mg,
6.54 mmol) in DMF (6 mL) at rt was added CH3I (346 μL,
5.54 mmol) dropwise. After 25 min, the mixture was
partitioned between heptane (10 mL) and water (6 mL).
The organic layer was separated and washed with water (2 × 6
mL) and then concentrated to give 26 (1.0 g, 93% yield) as an
oil: 1H NMR (300 MHz, DMSO-d6) 7.19 (d, J = 8.55 Hz, 1H),
6.59 (d, J = 8.55 Hz, 1H), 4.21 (s, 2H), 3.85 (s, 3H), 1.38 (s,
6H); HRMS (ESI) m/z [M + H]+ calcd for C11H14O2Cl
213.06768; Found 213.06708.
N-(6-Bromonaphthalen-2-yl)methanesulfonamide (32).18
To a solution of 6-bromonaphthalen-2-amine (10 g, 45.0
mmol) in THF (30 mL) was added pyridine (3.82 mL, 47.3
mmol) and, after cooling to 5 °C, methanesulfonyl chloride
(3.68 mL, 47.3 mmol). The mixture was allowed to warm to rt
overnight, then was poured into water (50 mL). After 2 h, the
resulting solids were collected by filtration, washed with water
1
(25 mL), and dried to give 32 (13.5 g, 100% yield): H NMR
(300 MHz, DMSO-d6) 10.08 (s, 1H), 8.15 (d, J = 1.70 Hz,
1H), 7.89 (d, J = 8.85 Hz, 1H), 7.83 (d, J = 8.85 Hz, 1H), 7.70
(d, J = 1.88 Hz, 1H), 7.60 (dd, J = 8.85, 2.07 Hz, 1H), 7.42 (dd,
J = 8.85, 2.26 Hz, 1H), 3.08 (s, 3H); HRMS (ESI) m/z [M −
H]− calcd for C11H9O2NBrS 297.95319; Found 297.95441.
N-[6-(4,4,5,5-Tetramethyl[1,3,2]dioxaborolan-2-yl)-
naphthalen-2-yl]methanesulfonamide (9). A mixture of 32
(5 g, 16.7 mmol), K3PO4 (9.81 g, 99.9 mmol), bis(pinacolato)-
diboron (5.08 g, 20.0 mmol), 2-dicyclohexylphosphino-2′,4′,6′-
tri-iso-propylbiphenyl (238 mg, 500 μmol), and Pd(OAc)2
(74.8 mg, 333 μmol) in degassed 1,4-dioxane (30 mL) was
heated in a 95 °C oil bath under a nitrogen atmosphere for 2.5
h. After cooling to rt, the slurry was diluted with 1,4-dioxane
(70 mL), then filtered through a Celite pad. After washing with
1,4-dioxane (2 × 40 mL), the combined filtrate/washes were
evaporated to dryness (7.4 g) then diluted with 1,4-dioxane (20
mL). After warming to dissolve the solids, heptane (30 mL)
was added slowly. The addition was halted at the cloud point to
initiate crystallization. After cooling slowly to 5 °C, the product
was collected by filtration, washed with cold 2:1 heptane/
1
dioxane (8 mL) and dried to give 9 (4.83 g, 84% yield): H
NMR (300 MHz, DMSO-d6) 10.07 (s, 1H), 8.24 (s, 1H), 7.98
(d, J = 8.85 Hz, 1H), 7.80 (d, J = 8.10 Hz, 1H), 7.60−7.74 (m,
2H), 7.38 (dd, J = 8.76, 2.17 Hz, 1H), 3.08 (s, 3H), 1.33 (s,
12H); HRMS (ESI) m/z [M + H]+ calcd for C17H23O4NBS
348.14354; Found 348.14279.
N-(6-(7-Amino-4-methoxy-3,3-dimethyl-2,3-dihydroben-
zofuran-5-yl)naphthalen-2-yl)methanesulfonamide hydro-
chloride (11-HCl). A mixture of 29 (660 mg, 2.01 mmol), 9
(909 mg, 2.62 mmol), K3PO4 (855 mg, 4.03 mmol), Pd(OAc)2
(18.1 mg, 80.5 μmol), and Cy3P (45.2 mg, 161 μmol) was
degassed by evacuating and filling three times with nitrogen.
Degassed DMF/water (5:1 v/v, 6.6 mL) was then added, and
the mixture was heated to 80 °C. After stirring for 6 h, the
reaction mixture was diluted with EtOAc (5 mL) and water (6
mL) and then was filtered through a Celite pad. After the pad
was washed with EtOAc (5 mL), the aqueous layer was
separated, and the organic layer was washed with water.
Filtration of the organic layer through a silica gel plug eluting
with 50% EtOAc/heptane and concentration then gave 30
(1.03 g): LC/MS m/z [M + H]+ 513. The residue was
dissolved in EtOAc (10 mL) and treated with 4 N HCl in
dioxane (2.5 mL, 10.0 mmol). The mixture was stirred at rt for
2 h, heated to 45 °C for 4 h, and then stirred at rt overnight.
After heating for a further 3 h at 50 °C, the reaction mixture
was cooled to rt, and the resulting solids were collected by
filtration, washed with EtOAc, and dried by suction to give 11-
7-Bromo-5-chloro-4-methoxy-3,3-dimethyl-2,3-dihydro-
benzofuran (27). To a solution of 26 (1 g, 4.7 mmol) in
MeCN (10 mL) was added NBS (837 mg, 4.7 mmol). After 2 h
at rt, the reaction was quenched with water and extracted with
heptane (10 mL). The organic layer was separated and washed
with water, then concentrated. Filtration through a silica gel
plug eluting with 10% EtOAc/hexane removed some color.
Concentration of the filtrate gave 27 (1.3 g, 95% yield) as a
1
light-brown oil: H NMR (300 MHz, DMSO-d6) δ 7.50 (s,
1H), 4.32 (s, 2H), 3.86 (s, 3H), 1.39 (s, 6H).
tert-Butyl 5-chloro-4-methoxy-3,3-dimethyl-2,3-dihydro-
benzofuran-7-ylcarbamate (29). A mixture of 27 (1 g, 3.43
mmol), tert-butyl carbamate (482 mg, 4.12 mmol) and t-
BuONa (396 mg, 4.12 mmol) in toluene (10 mL) was degassed
by evacuating and filling three times with nitrogen. Pd2(dba)3
(78.5 mg, 85.7 μmol) and 2-di-tert-butylphosphino-2′,4′,6′-tri-
iso-propylbiphenyl (102 mg, 240 μmol) were added, and the
mixture was heated to 80 °C over 10 min and stirred for 10
min. The reaction mixture was diluted with EtOAc and water,
then filtered through a Celite pad. The organic layer was
separated, washed with water, and concentrated. Filtration
through a silica gel plug eluting with 6% EtOAc/hexane gave an
oil, which was crystallized from heptane to give 29 (870 mg,
1
HCl (724 mg, 80% yield) as an off-white solid: H NMR (300
MHz, DMSO-d6) 10.45−9.45 (br, 2H), 10.06 (s, 1H), 7.86−
8.04 (m, 3H), 7.72 (d, J = 1.70 Hz, 1H), 7.62 (dd, J = 8.57, 1.70
Hz, 1H), 7.42 (dd, J = 8.85, 2.07 Hz, 1H), 7.27 (s, 1H), 4.40 (s,
2H), 4.20−3.35 (br, 1H), 3.30 (s, 3H), 3.08 (s, 3H), 1.47 (s,
6H); HRMS (ESI) m/z [M + H]+ calcd for C22H25O4N2S
413.15295; Found 413.15234.
1
77% yield) as colorless crystals: H NMR (300 MHz, DMSO-
d6) major rotamer 8.54 (s, 1H), 7.35 (br s, 1H), 4.23 (s, 2H),
3.82 (s, 3H), 1.43 (s, 9H), 1.37 (s, 6H); HRMS (ESI) m/z [M
+ H]+ calcd for C16H23O4NCl 328.13101; Found 328.13028.
660
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661

Organic Process Research & Development
Technical Note
1601. (b) Kowalczyk, B. A.; Roberts, P. N.; McEwen, G. K.; Robinson,
J. Org. Process Res. Dev. 1997, 1, 355.
N-(6-(7-Amino-4-methoxy-3,3-dimethyl-2,3-dihydroben-
zofuran-5-yl)naphthalen-2-yl)methane-sulfonamide (11).
Salt 11-HCl (724 mg, 1.61 mmol) was dissolved in MeOH
(10 mL), and saturated aqueous NaHCO3 (2.0 mL) was added
dropwise, during which time the free base crystallized out of
solution. MeOH was removed partially under vacuum, and the
solid was then collected by filtration, washed with water, and
dried to give 11 (661 mg, 100% yield) as an off-white solid: 1H
NMR (300 MHz, DMSO-d6) 9.96 (s, 1H), 7.76−8.04 (m, 3H),
7.58−7.76 (m, 2H), 7.38 (dd, J = 8.76, 2.17 Hz, 1H), 6.59 (s,
1H), 4.52 (br s, 2H), 4.21 (s, 2H), 3.21 (s, 3H), 3.06 (s, 3H),
1.42 (s, 6H); HRMS (ESI) m/z [M + H]+ calcd for
C22H25O4N2S 413.15295; Found 413.15247.20
(11) Tashiro, M.; Fukata, G. J. Org. Chem. 1977, 42, 428.
(12) Although this was not investigated, all the available evidence
indicated that halogenation in a polar solvent like DMF would be less
selective. Likewise methylation in dichloromethane was considered
unlikely.
(13) The reason for the faster methylation is unclear. The hydroxyl
group in 24 would appear more hindered than in 25. Also, the pKa of
25 is likely higher than in 24, based on an analogy to 4- and 2-
chlorophenol (9.41 and 8.56, respectively), leading to faster reaction
for 25 if both 24 and 25 are fully ionized. The pKa data were taken
from Dissociation Constants of Organic Acids and Bases. In CRC
Handbook of Chemistry and Physics, 94th ed. (Internet Version 2014)
[online]: Haynes, W. M., Ed.; CRC Press/Taylor and Francis: Boca
Raton, FL, 2013−2014; Chapter 5, pp 94−103. http://www.
hbcpnetbase.com.
ASSOCIATED CONTENT
■
S
(14) No halogen scrambling of 27 was detected.
* Supporting Information
(15) Blanksby, S. J.; Ellison, G. B. Acc. Chem. Res. 2003, 36, 255.
(16) (a) Bhagwanth, S.; Waterson, A. G.; Adjabeng, G. M.;
Hornberger, K. R. J. Org. Chem. 2009, 74, 4634. (b) Hartwig, J. F.;
Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roman, L. M.
J. Org. Chem. 1999, 64, 5575.
(17) Loss of the tert-butyl group from 30 was observed at the higher
temperatures (Table 3, entry 2); presumably, decomposition of the
tert-butyl carbamate was also occurring.
NOE spectra for the reaction products from the chlorination of
4. This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: andrew.briggs@roche.com
(18) Lopez, N. A.; Abelt, C. J. J. Photochem. Photobiol., A 2012, 238,
35.
(19) Yasuda, S.; Yorimitsu, H.; Oshima, K. Organometallics 2008, 27,
Notes
The authors declare no competing financial interest.
4025.
(20) 1H NMR identical to an authentic spectrum provided by
Discovery Chemistry.
ACKNOWLEDGMENTS
■
We thank G. Sasso for help with the identification of 19, T.
Burchfield for the NOE data on the reaction products from the
chlorination of 4, T. Tran for analytical support, and Dr. M.
Okabe and Dr. C. Exon for valuable input during the
preparation of the manuscript.
REFERENCES
■
(1) Schoenfeld, R. C.; Staben, L. R.; Talamas, F. X. Heterocyclic
antiviral compounds. U.S. Patent Application 2011/0123490, May 26,
2011, CA154:606427.
(2) Dominique, R. Hoffmann-La Roche, Inc., Nutley, NJ. Personal
communication.
(3) The ipso-substitution would likely result in the formation of the
corresponding dibromo- and dinitro- derivatives of methylated 4, but
these side products were not reported by our Discovery Chemistry
colleagues.
(4) Janusz, J. M.; Young, P. A.; Scherz, M. W.; Enzweiler, K.; Wu, L.
I.; Gan, L.; Pikul, S.; McDow-Dunham, K. L.; Johnson, C. R.;
Senanayake, C. B.; Kellstein, D. E.; Green, S. A.; Tulich, J. L.; Rosario-
Jansen, T.; Magrisso, I. J.; Wehmeyer, K. R.; Kuhlenbeck, D. L.;
Eichhold, T. H.; Dobson, R. L. M. J. Med. Chem. 1998, 41, 1124.
(5) Stanetty, P.; Koller, H.; Purstinger, G. Monatsh. Chem. 1990, 121,
̈
883.
(6) Polymerization of allyl chloride in strong acid has been noted:
Solvay North America, HSE and Responsible Care, Product Safety
S u m m a r i e s : h t t p : / / w w w . s o l v a y n o r t h a m e r i c a . c o m /
S i t e C o l l e c t i o n D o c u m e n t s / P D F / P S _
AllylChlorideAndEpichlorohydrinFinal.pdf.
(7) Purity refers to HPLC area %.
(8) See Supporting Information.
(9) Nitration of 24 or 26 was not investigated, so it is unknown if the
scrambling seen in 6 would be suppressed. Developing safe nitration
chemistry for large-scale use is more challenging due in part to the
limited number of suitable solvents.10 In addition, nitrated 26
produced a mutagenicity alert in an in silico Derek analysis.
(10) (a) Urben, P. G. Bretherick’s Handbook of Reactive Chemical
Hazards, 6th ed.; Butterworth-Heinemann: 1999; Vol 1, pp 1566−
661
dx.doi.org/10.1021/op4003467 | Org. Process Res. Dev. 2014, 18, 656−661