Synthesis of filibuvir. Part I. Diastereoselective preparation of a β-hydroxy alkynyl oxazolidinone and conversion to a 6,6- disubstituted 2H-pyranone

Article abstract of DOI:10.1021/op4002356

This is the first in a series of three papers describing the identification and development of a commercial synthesis of filibuvir (1). This contribution describes development of an Evans aldol reaction to control the tertiary alcohol stereocenter, a challenging variant of that strategy in that both reacting partners were nonstandard (acetate enolate and ketone electrophile). A sequence consisting of Sonogashira coupling, acylation and hydrogenation delivered acetate 24, and Dieckmann cyclization provided β-keto lactone 2.

Full text of DOI:10.1021/op4002356

Article
pubs.acs.org/OPRD
Synthesis of Filibuvir. Part I. Diastereoselective Preparation of a
β‑Hydroxy Alkynyl Oxazolidinone and Conversion to a 6,6-
Disubstituted 2H‑Pyranone
Robert A. Singer,*,† John A. Ragan,*,† Paul Bowles,† Esmort Chisowa,† Brian G. Conway,† Eric M. Cordi,†
Kyle R. Leeman,‡ Leo J. Letendre,† Janice E. Sieser,† Gregory W. Sluggett,‡ Corey L. Stanchina,†
and Holly Strohmeyer‡
‡
†Chemical Research and Development and Analytical Research and Development, Pfizer Worldwide Research and Development,
Eastern Point Road, Groton, 06340 Connecticut
Jon Blunt and Stuart Taylor§
Chemical Research and Development, Pfizer Process Development Facility Ramsgate Road, Sandwich, Kent, U.K.
Ciaran Byrne, Denis Lynch, Sandra Mullane, Maria M. O’Sullivan, and Marcella Whelan
Pfizer Global Supply, Ringaskiddy, Cork County, Ireland
ABSTRACT: This is the first in a series of three papers describing the identification and development of a commercial synthesis
of filibuvir (1). This contribution describes development of an Evans aldol reaction to control the tertiary alcohol stereocenter, a
challenging variant of that strategy in that both reacting partners were nonstandard (acetate enolate and ketone electrophile). A
sequence consisting of Sonogashira coupling, acylation and hydrogenation delivered acetate 24, and Dieckmann cyclization
provided β-keto lactone 2.
toxicological programs, we required a more efficient synthesis
of filibuvir than that originally developed.3
INTRODUCTION
■
Hepatitis C virus (HCV) is the leading cause of liver
transplantation in North America and infects over 170 million
people worldwide.1 There is no vaccine currently available for
HCV. The standard treatment, a combination of pegylated-
interferon α with ribavirin, is effective in less than 50% of
genotype 1 patients and is associated with several significant
side effects.2 Development of small-molecule HCV therapies is
a promising approach to improving this clinical situation.
Filibuvir 1 (Figure 1) was discovered in Pfizer’s La Jolla
Laboratories.3 As an RNA polymerase inhibitor, filibuvir
Our retrosynthesis of filibuvir is shown in Figure 2. This first
of three contributions describes the diastereoselective synthesis
of chiral propargylic alcohol 5 and its conversion to β-keto
lactone 2. Part II describes development of a more efficient,
second-generation route for the conversion of oxazolidinone 5
to β-keto lactone 2. Part III describes the development of the
final reductive coupling of β-keto lactone 2 and aldehyde 3 to
form filibuvir 1, as well as a rigorous impurity control strategy.
DISCUSSION
■
A fundamental challenge in the preparation of 1 was control of
chirality. In the initial medicinal chemistry synthesis this was
addressed through the use of a racemic synthesis followed by a
chiral HPLC separation of enantiomers.3 Scott and colleagues
in the legacy Pfizer - La Jolla Process Group developed an
efficient chemical resolution of β-hydroxy acid 12 (Scheme 1),
which was based on a similar strategy used in the synthesis of 9,
an earlier HCV candidate, and proceeded through resolution of
hydroxy acid 8 with amine 7.7 A similar resolution strategy was
adopted for filibuvir through resolution with (2R,3R)-2-amino-
3-hydroxy-3-(4-nitrophenyl)propanol 11. This strategy offered
significant benefits relative to a racemic synthesis by avoiding
the processing of the undesired enantiomer beyond the
Figure 1. Structure of filibuvir (1).
represents a complementary mechanism of action to telaprevir
(Incivek)4 and boceprevir (Victrelis),5 two small-molecule
HCV protease inhibitors approved by the FDA in 2011.6 In the
course of preparing supplies for ongoing clinical and
Received: August 27, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 2. Retrosynthesis of filibuvir.
Scheme 1. Chiral resolution strategy for the synthesis of filibuvir (1)
Scheme 2. Three routes to cyclopentyl ethynyl ketone (13)
resolution point. Additionally, a recycle method was developed
wherein the undesired enantiomer was converted to ketone 10
through a retro-aldol reaction. Nevertheless, any resolution
strategy suffers from loss of at least half of the material (absent
B
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
the recycle, which in practice is difficult to incorporate into a
commercial route), and the cost of 10 was unlikely to be
sufficiently low to tolerate this. Additionally, there were
toxicology and cost concerns with the resolving agent 11. We
felt that identification of an enantioselective synthesis would
overcome these shortcomings.
place of methyl), we found that the lithium enolate of
oxazolidinone 2119 reacted with ketone 13 with 93:7
diastereoselectivity, suggesting that the increased steric demand
of the cyclopentyl relative to the methyl group was beneficial.
During isolation of the crystalline aldol adduct 5 a significant
enhancement of the diastereoselectivity was observed (>200:1,
minor diastereomer not observed by HPLC). It was found that
self-condensation could be minimized by an inverse addition of
the acyloxazolidinone to LHMDS.
In the present work, we sought an enantioselective
preparation of a similar β-hydroxy acid or suitable precursor.
The prospect of using a Sonogashira coupling to install the 2,6-
diethylpyridine moiety made ethynyl cyclopentyl ketone 13 an
appealing starting point for a diastereoselective aldol approach.
Additionally, the small steric properties of the ethynyl
substituent (relative to other alkyl groups) offered the prospect
of effective differentiation from the α-branched cyclopentyl
moiety.
The results with several other substrates are shown in Table
1. Entries 1−3 show that nonterminal alkynes suffered a modest
Table 1. Effect of alkyne and oxazolidinone substituents on
aldol selectivity
Ketone 138 was prepared via three routes (Scheme 2). The
first route developed proceeded via palladium- and copper-
catalyzed coupling of trimethylsilylacetylene with acid chloride
17 to form silylacetylene 18, followed by desilylation with
Hunig′s base in MeOH/toluene. Although this route was scaled
̈
successfully (>100 kg), an alternative Weinreb−Grignard route
offered the advantages of a shorter sequence and no
desilylation, which can be problematic due to the base lability
of the alkynyl ketone. Thus, Weinreb amide 16 was coupled
with commercial ethynyl magnesium chloride in THF to
directly provide 13. Amide 16 was prepared from acyl
imidazole 15, which was prepared either from commercially
available cyclopentanecarboxylic acid or from the less expensive
diethyl malonate via the sequence of bis-alkylation, hydrolysis,
isolation of the solid dicarboxylic acid 14,9 and CDI-mediated
decarboxylation.10
entry
R
R′
diastereoselectivity
comments
1
2
3
4
CH2Ph
Ph
Me3Si
Me3Si
2,6-Et2Pyr
H
83:17
87:13
92:8
modest selectivity
modest selectivity
product is an oil
CH2Ph
CH2Ph
93:7
mp 95 °C,
>200:1 d.r. after
crystallization
5
4,5-di-Ph
H
96:4
mp 155 °C, but
auxiliary more
expensive
than entry 4
6
Ph
H
96:4
product is an oil
Development of the acyloxazolidinone aldol addition to
alkynyl ketone 13 is shown in Scheme 3.11,12 Two features of
reduction in selectivity. Although the 2,6-diethylpyridyl-alkyne
(entry 3) gave good selectivity, the noncrystalline nature of the
aldol product made this intermediate less attractive. For the
terminal alkyne (entries 4−6), some improvements in
selectivity were realized with other oxazolidinone substitution
patterns, but these were viewed as less attractive than the
phenylalanine-derived auxiliary (entry 4) due to lack of
crystallinity (entry 6) or cost of the auxiliary (entry 5).
Scheme 3. Aldol reaction of acyloxazolidinone 21 with
cyclopentyl ethynyl ketone 13
A stereochemical model for the reaction is shown in Figure 3.
The model involves a Zimmerman−Traxler-type transition
Figure 3. Stereochemical model for Evans aldol with cyclopentyl
ethynyl ketone 3.
this aldol reaction limit the number of precedents: (a) the
electrophile is a ketone rather than an aldehyde, and (b) the
enolate is an acetate rather than a propionate. Several literature
precedents were identified, both with a phenylalanine-derived
oxazolidinone13−16 and other chiral auxiliaries.17,18 These
examples suggested that moderate to good selectivity could
be anticipated (58:42 to 88:12 dr). A particularly compelling
state20 in which lithium coordination with the enolate oxygen
and oxazolidinone carbonyl oxygen controls the absolute facial
selectivity, and the relative facial selectivity is dictated by
placement of the smaller ethynyl group in the more sterically
demanding pseudoaxial position (this position would be
occupied by a hydrogen atom with aldehyde electrophiles).
This explains the unique nature of alkynyl ketones to provide
reasonable degrees of selectivity in aldol reactions relative to
that of other dialkyl ketones.
example was reported by Furstner subsequent to the initiation
̈
of our studies, in which the same acyloxazolidinone was added
to ethynyl methyl ketone 19 to give alcohol 20 with 88:12
diastereoselectivity.15 With our related substrate (cyclopentyl in
With an efficient preparation of Evans aldol product 5 in
hand, we turned to the introduction of the pyridine and
C
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 4. Telescoped Sonogashira−acylation−hydrogenation sequence to acyloxazolidinone 24
conversion of the β-hydroxy oxazolidinone to the requisite 6,6-
disubstituted-2H-pyrone functionality (Scheme 4). Sonogashira
coupling of 4-bromo-2,6-diethylpyridine tosylate 421,22 and
alkyne 5 proceeded smoothly in the presence of (Ph3P)2PdCl2,
CuI, and Et3N in toluene. Formation of the oxidative alkyne
dimer 25 (Figure 4) was problematic in certain cases,
of acetic anhydride and methanesulfonic acid in toluene to
provide acetate 23. These conditions were identified after initial
screening showed Cu(OTf)2 to be an effective catalyst for
acylation of this hindered, tertiary alcohol.23 Subsequent
experiments demonstrated that the copper triflate could be
replaced by a strong acid such as methanesulfonic acid,
suggesting that the role of the copper triflate was to provide
catalytic quantities of trifluoromethanesulfonic acid. The
resulting acetate was prone to elimination in the presence of
base to form a mixture of enynes 27, and thus was carried
directly into the hydrogenation without purification beyond
that described below. Alkyne hydrogenation in toluene−
isopropanol over 10% Pd/C generated acetate 24 and was
isolated as the tosylate salt, which was found to be a stable,
easily filtered, crystalline solid.
While this telescoped, three-step sequence was effective for
the preparation of up to 150 kg batches of acyloxazolidinone
24·TsOH, a significant issue was poisoning of the hydro-
genation catalyst, requiring the use of ≥25 wt % catalyst to
achieve complete alkyne reduction in a single cycle. We
reasoned that this was due to residual copper and phosphine
impurities from the initial Sonogashira coupling poisoning the
hydrogenation catalyst. Decreased catalyst loadings could be
achieved following treatment of propargylic acetate 23 (as a
toluene solution) with aqueous ethylenediamine tetraacetic acid
disodium salt (EDTA·Na2) or, more effectively, ethylenedi-
amine (EDA). We approached the use of EDA with some
trepidation, as control experiments indicated that prolonged
exposure times and higher temperatures led to significant
elimination in the presence of EDA. However, further studies
evaluating addition times and agitation rates indicated that,
when added as a dilute toluene solution (10 wt %/wt) and with
careful control of pH, the EDA had minimal impact on the level
of elimination impurities formed.
Figure 4. Process-related impurities.
particularly without careful exclusion of oxygen or when CuI
was added prior to the palladium catalyst. By means of
thorough nitrogen purging and premixing of the Pd(II)
precatalyst with substrate and triethylamine, dimer formation
was kept at or below 2 area % (based on HPLC response
factors, 2 area % represents ∼6−8 wt % dimer). Diol 26 was
also observed under certain conditions. This impurity arose
from trace water in the Sonogashira coupling, which led to
endocyclic hydrolysis of the oxazolidinone. Its formation was
accelerated at higher reaction temperatures (50−65 °C) and
extended reaction times. By keeping the reaction temperature
<50 °C and quenching shortly after complete conversion was
achieved, the hydrolysis levels were minimized (<2%). The
resulting propargylic alcohol 22 was acylated through the action
Scheme 5. Acylation and Dieckmann cyclization to form β-keto lactone 2
D
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 6. Elimination pathways in the Dieckmann cyclization
Treatment with activated carbon was also found to offer
some benefits. Thus, the optimized workup consisted of
treatment of the toluene solution of alkyne 23 with EDA and
toluene followed by aqueous EDTA, maintaining a pH of 5−6,
followed by filtration through a series of CUNO cartridges
containing Darco KBB activated carbon. Even with these
precautions, catalyst loadings were maintained at 25 wt % to
facilitate acceptable conversions (≤2.0% olefin) in reasonable
reaction times.
possibility, and intramolecular proton transfer from the desired
C3′ enolate to C2 is also feasible.
We found that higher temperatures and slower addition of
LHMDS to the substrate solution increased the levels of
elimination byproducts. On laboratory scale, yields of 70−75%
could be realized through the rapid addition of LHMDS, which
minimized the elimination pathway (the cyclization itself is very
fast, and kinetic studies showed the rate-limiting step to be
enolization). This rapid addition strategy was not viable on
large scale, however, due to the reduced heat transfer capacity
of larger reaction vessels. We also examined an inverse addition
wherein a substrate solution was added to a pre-chilled solution
of LHMDS. This offered no yield improvements, however, as
the product now suffered elimination to form increased levels
of enones 29. Ultimately, we found that changing the
oxazolidinone to a less acidic carbonyl group (e.g., an ester)
led to the most dramatic improvements (see Part 2, the
subsequent paper in this journal).
Yield variations between different batches of LHMDS were
also observed. A more thorough discussion of this appears in
the following paper, which describes a similar Dieckmann
cyclization of a modified substrate (ester replacing oxazolidi-
none). For both cyclizations, we observed that LHMDS
prepared from n-BuLi and hexamethyldisilazane offered the
best yields, whereas some batches of LHMDS prepared from
lithium metal, 2-methylbutadiene, and hexamethyldisilazane
suffered ∼10% lower product yields. Reactivity differences for
LDA prepared from n-BuLi and i-Pr2NH vs lithium metal, α-
methylstyrene, and i-Pr2NH24 have also been previously
observed.25,26 Lower-yielding LHMDS batches could be
remediated by crystallization of the LHMDS at low temper-
ature and reformulation in fresh solvent. Control experiments
demonstrated that the residual 2-methyl-2-butene (formed in
the reduction of 2-methylbutadiene by lithium metal) was not
responsible for the yield differences. The reason for these
interbatch differences remains elusive, and further discussion
and data are presented in the accompanying contribution, in
which similar differences were observed with an ester substrate.
Two strategies were developed to address the variable yields
observed for this cyclization. The near-term solution was to
develop addition protocols via appropriate engineering and
equipment train modifications to minimize the elimination
pathways. The most successful of these was a dual-stream
addition in which simultaneous addition of cyclization substrate
24 and LHMDS (both as toluene solutions) to a prechilled
Conversion of 24·TsOH to pyrone 2 was achieved by first
performing a salt break in toluene and Et3N, followed by a
Dieckmann-type cyclization by treatment with lithium
hexamethyldisilazide (LiN(TMS)2, LHMDS) in toluene at
−20 °C. Following an aqueous citric acid workup, β-keto
lactone 2 was isolated as its crystalline hydrate (Scheme 5).
Identification of the hydrate of β-keto lactone 2 as a crystalline
solid was a breakthrough for the project, as the anhydrous
material was prone to decomposition via self-condensation and
elimination pathways. Direct crystallization of the hydrate from
the reaction mixture after neutralization with citric acid
provided a robust and scaleable isolation method that had
been lacking from earlier approaches with this intermediate.
There were two issues with this cyclization that complicated
our efforts to scale up the reaction from laboratory to pilot-
plant scale (e.g., 50 g to 50 kg). The first was impurity
formation by elimination of either starting material or product.
The second was yield variations observed with different sources
of LHMDS. These two issues will be discussed in sequence.
The two elimination pathways are shown in Scheme 6. The
first involves elimination of acetic acid from the cyclization
substrate 24, to form a mixture of acrylamides 28. An E1CB
mechanism via the lithium enolate is shown, but this could also
proceed through an E2 elimination in the presence of a weaker
base such as the lithiated oxazolidinone generated during the
cyclization. The second pathway is a ring-opening elimination
of the β-keto lactone product to generate a β-keto acid, which
undergoes decarboxylation to generate a mixture of enone
isomers 29. The top elimination pathway predominates in
“normal addition” mode, i.e. addition of LHMDS to the
substrate. We estimate the pKa values of the relevant protons at
∼25 for C3′ and ∼23 for C2. The use of a strong, hindered base
such as LHMDS and the low reaction temperature (−20 °C)
favor kinetic deprotonation at the more sterically accessible
position (C3′). Nonetheless, direct enolization at C2 remains a
E
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
reactor at −15 °C led to cyclization with reduced levels of
elimination byproducts. The goal of this simultaneous addition
was to minimize exposure of both starting oxazolidinone and
product lactone to strong base, which led to the eliminations
described in Scheme 6. Using this mode of addition increased
the yield from 40 to 55% on scale to 55−65%. The longer-term
solution was to suppress the elimination pathway by changing
the nature of the leaving group on the electrophilic carbonyl
(e.g., replace the oxazolidinone with an alternative function-
ality). The results of this approach will be discussed in the
following article (Synthesis of Filibuvir, Part 2).
instrument (Waters Corp., Milford, MA) and an Acquity BEH
C8 column (100 mm × 2.1 mm, 1.7 μm particle size). Mobile
phase A was 10 mM ammonium acetate in water, and mobile
phase solution B was acetonitrile. Gradient elution was
performed as shown in Table 4 with a run time of 16 min
and a re-equilibration time of approximately 3 min. The flow
rate was 0.3 mL/min, and the column temperature was 30 °C.
The injection volume was 2 μL, and UV detection was
performed at 220 nm. Solutions of reaction mixtures were
prepared for analysis by dilution of 0.5 mL reaction mixture
into 100 mL acetonitrile.
Gradient Program for UPLC Analysis (Table 4). The
approximate retention times are listed in Table 5.
CONCLUSION
■
A synthesis of oxazolidinone 5 has been developed via the
diastereoselective addition of an acyl oxazolidinone to ethynyl
cyclopentyl ketone. A remarkably high level of diastereose-
lectivity was realized, given the unusual nature of the reacting
partners (acetate enolate with ketone electrophile), and this
provided an efficient strategy for control of the single
stereocenter in filibuvir (1). Approximately 20 batches of
oxazolidinone 5 have been prepared on 100 kg scale via this
route. Conversion of oxazolidinone 5 to β-keto lactone 2 was
achieved via a sequence of Sonogashira coupling, acylation,
hydrogenation, and base-mediated (“Dieckmann-type”) cycliza-
tion. Although yield challenges were encountered in this
sequence, two robust isolation points were identified
(oxazolidinone 24·TsOH and β-keto lactone 2·H2O) that
provided effective control of impurities.
Table 4. Gradient program for UPLC analysis
time (min)
% mobile phase A
% mobile phase B
0
12
14
16
45
45
5
55
55
95
95
5
The UPLC method used for monitoring the conversion of 23
to 24 employed a Waters Acquity series instrument (Waters
Table 5. Approximate retention times
cmpd
retention time (min)
toluene
4
2.2
2.4
EXPERIMENTAL SECTION
■
5
2.9
Analytical Methods. The HPLC method used for analysis
of 5 employed an Agilent 1100 series instrument (Agilent
Technologies, Santa Clara, CA) and a Waters XBridge C8
column (150 mm × 3.0 mm, 3.5 μm particle size). Mobile
phase A was 10 mM ammonium acetate in water, and mobile
phase solution B was acetonitrile. Gradient elution was
performed as shown in Table 2 with a run time of 50 min
and a re-equilibration time of approximately 5 min. The flow
rate was 0.6 mL/min, and the column temperature was 30 °C.
The injection volume was 5 μL, and UV detection was
performed at 220 nm. Solutions of 5 were prepared at 2 mg/
mL in acetonitrile for analysis.
22
5.9
23
9.9
25
11.0
5.0
26
27
10.8
Corp., Milford, MA) and an Acquity BEH C18 column (100
mm × 2.1 mm, 1.7 μm particle size). Mobile phase A was 10
mM ammonium acetate in water, and mobile phase solution B
was acetonitrile. Gradient elution was performed as shown in
Table 6with a run time of 14 min and a re-equilibration time of
approximately 3 min. The flow rate was 0.3 mL/min, and the
column temperature was 30 °C. The injection volume was 2
μL, and UV detection was performed at 220 nm. Solutions of
reaction mixtures were prepared for analysis by dilution of 1
mL reaction mixture into 100 mL acetonitrile/water (50/50, v/
v). Under these conditions, 24 elutes at approximately 8 min.
Gradient Program for UPLC Analysis (Table 6). The HPLC
method used for monitoring the conversion of 24 to 2
employed an Agilent 1100 series instrument (Agilent
Technologies, Santa Clara, CA) and a Waters XBridge C8
column (150 mm × 3.0 mm, 3.5 μm particle size). Mobile
phase A was 10 mM ammonium acetate in water and mobile
phase solution B was acetonitrile. Gradient elution was
Gradient Program for HPLC Analysis (Table 2). The
approximate retention times are listed in Table 3.
Table 2. Gradient program for HPLC analysis
time (min)
% mobile phase A
% mobile phase B
0
20
40
50
60
60
5
40
40
95
95
5
Table 3. Approximate retention times
cmpd
retention time (min)
Table 6. Gradient program for UPLC analysis
5
18.7
17.7
5.6
Epi-5 (3-S-diastereomer)
time (min)
% mobile phase A
% mobile phase B
13
21
4.1
0
10
12
14
40
30
5
60
70
95
95
The UPLC method used for monitoring the conversion of 5
to 22 and 22 to 23 employed a Waters Acquity series
5
F
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
performed as shown in Table 7 with a run time of 20 min and a
re-equilibration time of approximately 3 min. The flow rate was
0.6 mL/min, and the column temperature was 30 °C. The
injection volume was 4 μL, and UV detection was performed at
220 nm. Solutions of solid 2 were prepared at 0.8 mg/mL in
acetonitrile/water (80/20, v/v) for analysis. Reaction mixtures
were quenched by addition of an equal volume of 1 M citric
acid prior to analysis. Approximately 0.5 mL of each layer of the
quenched reaction mixture was combined and diluted to 25 mL
with acetonitrile/water (80/20, v/v) for analysis.
Spectral data were in accord with those reported previously.8
Cyclopentyl Ethynyl Ketone (13). [Note: the initial
aqueous NaOH treatment in this procedure was found useful
for reducing residual copper levels from the preceding reaction
and improved the conversion in the desilylation]. A 3000-L
glass-lined reactor was charged with water (950 kg) and sodium
hydroxide (1.9 kg) and stirred until homogeneous. The
solution was cooled to 0 to 10 °C, and toluene (827 kg) and
the TMS-alkynyl ketone 18 (192 kg, 990 mol) were added. The
resulting two-phase solution was stirred for 90 min and then
allowed to settle, and the aqueous phase was removed. Water
(950 kg) was added, stirring was initiated for 30 min, and then
the phases were allowed to settle and separate. The resulting
organic phase was divided into two portions for processing.
Each portion was transferred to a 3000-L glass-lined reactor and
cooled to −5−0 °C. Diisopropylethylamine (0.64 kg, 5.0 mol)
and methanol (15 kg) were added, and stirring was continued
until <3% of the starting material remained by HPLC assay.
Glacial acetic acid (0.29 kg, 4.8 mol) was added followed by
calcium chloride (25 kg), and the slurry was stirred for 60 min.
The supernatant was assayed by Karl Fischer analysis, which
showed <0.15% water; GC analysis showed <0.1% methanol.
The mixture was filtered, and the filtrates were combined to
give a red-brown liquid. HPLC assay indicated 89% chemical
purity. This solution of alkynyl ketone was used directly in the
aldol reaction, basing stoichiometry on the quantity of TMS-
acetylenic ketone 18 that went into the reaction.
Gradient Program for HPLC Analysis (Table 7). The
approximate retention times are listed in Table 8.
Table 7. Gradient program for HPLC analysis
time (min)
% mobile phase A
% mobile phase B
0
7
80
40
40
20
20
20
60
60
80
80
17
17.1
20
Table 8. Approximate retention times
cmpd
retention time (min)
2
5.7
16.9
17.6
24
28
Spectral data were in accord with those reported previously.8
(S)-4-Benzyl-3-((R)-3-cyclopentyl-3-hydroxy4-
pentynoyl)oxazolidin-2-one (5). Preparation of LHMDS:
To a 1500-L low-temperature reactor were charged toluene
(152 kg), THF (388 kg), and hexamethyldisilazane (87.1 kg,
540 mol). The mixture was cooled to below −5 °C, n-butyl
lithium (144 kg, 208 L, 2.51 mol/L, 522 mol) was added at the
rate of 30−50 kg/h at <0 °C. After addition, the mixture was
stirred for 2 h at <0 °C and stored until needed in the
procedure below.
29
11.6, 13.2
9.1
toluene
Cyclopentyl Trimethylsilylethynyl Ketone (18). Into a
3000-L glass-lined reactor was charged toluene (653 kg), and it
was then purged with nitrogen with stirring. (Ph3P)2PdCl2
(2.38 kg, 3.39 mol) and CuI (2.15 kg, 11.3 mol) were added in
single portions, followed by trimethylsilylacetylene (171 kg,
1,741 mol) and cyclopentanecarbonyl chloride (150 kg, 1,130
mol). The resulting slurry was cooled to 15 to 25 °C, and
triethylamine (229 kg, 2270 mol) was added over ∼4 h (50−70
kg/h) at 25 to 35 °C. Stirring was continued until GC analysis
indicated <0.5% remaining acid chloride (4 h). The reaction
was then cooled to 15−23 °C and quenched with an aqueous
citric acid solution (prepared in a separate vessel by addition of
citric acid (238 kg, 1240 mol) to 600 kg of water). Celite (75
kg) was added in one portion, and stirring continued for 30−60
min. The mixture was then filtered, rinsing with toluene (197
kg). The aqueous phase was separated, and the organic layer
was concentrated at 100 mbar and <60 °C until the w/w% of
toluene (relative to product) was <12%. The resulting solution
was diluted with n-heptane (100 kg) and vacuum distilled (15−
25 mbar and jacket temperature <60 °C) until distillation
ceased. The mixture was cooled to 15−25 °C, and activated
carbon (45 kg) and n-heptane (90 kg) were added as a slurry,
followed by an additional portion of n-heptane (195 kg).
Stirring was continued for 2 h, and the slurry was then filtered,
rinsing with n-heptane (153 kg). The filtrate was then filtered
through a pad prepared from silica gel (60 kg) and n-heptane
(45 kg), rinsing the pad with additional n-heptane (312 kg).
The filtrate was then concentrated (partial vacuum and jacket
temperature <60 °C) until the w/w% of n-heptane was <7%.
The resulting pale-yellow liquid contained 192 kg of product
(83%) by HPLC analysis, and showed a chemical purity (less
residual n-heptane) of 96%.
Acyloxazolidinone 21 (98 kg, 578 mol)19 and THF (349 kg)
were charged to a 2000-L glass-lined reactor and stirred until
dissolution was complete. The LHMDS solution prepared
above was charged to a 3000-L stainless steel reactor and
cooled to between −85 and −75 °C. The solution of
acyloxazolidinone 21 was then transferred rapidly to the
LHMDS solution, rinsing with 11 kg of THF. The resulting
solution was stirred at −80 to −70 °C for 2 h. A THF solution
of cyclopentyl ethynyl ketone 13 (450 mol) from the preceding
reaction was then added at a rate between 120 and 170 kg/h (3
h addition). Stirring was continued until HPLC analysis of an
aliquot showed <5% unreacted acyloxazolidinone (1 h). The
reaction mixture was quenched with a solution of citric acid
(prepared from 108 kg citric acid and 262 kg THF) and added
at a rate of 80−130 kg/h such that the internal temperature
remained <−60 °C. The mixture was warmed to between −10
and 0 °C and water (294 kg) was added. The layers were
separated, and the organic phase washed with 5% aq sodium
bicarbonate (393 kg). The organic phase was then concentrated
under partial vacuum at 30−45 °C to a volume of 200−380 L.
Isopropanol (287 kg) was then added, and distillation
continued to a volume of 180−250 L, at which point a toluene
content of <1.5% was determined by GC analysis (the IPA
displacement was repeated if this value was exceeded). A
mixture of n-heptane (333 kg) and IPA (130 kg) was added,
and the mixture was heated to 60−70 °C to achieve complete
G
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
dissolution. It was maintained at that temperature for 60 min
and then cooled to 20−25 °C at a rate of 8−10 °C/h. The
resulting slurry was stirred for 2 h and then filtered. After drying
under vacuum, the product 5 was obtained as an off-white solid,
mp 93 °C (107 kg, 313 mol, 70% yield, 95% HPLC purity).
Data for oxazolidinone 5 prepared in a laboratory pilot.
1H NMR (300 MHz, CDCl3): δ 7.37−7.23 (m, 5H), 4.76−4.71
(m, 1H), 4.44 (s, 1H), 4.23−4.18 (m, 2H), 3.79 (d, 1H, J =
18), 3.36−3.31 (m, 1H), 2.92 (d, 1H, J = 18), 2.43 (s, 1H),
2.20−2.17 (m, 1H), 1.82−1.57 (m, 8H).
Data for alkyne 23 obtained from a laboratory pilot,
isolated by crystallization from tert-butyl methyl ether: H
1
NMR (400 MHz, CDCl3): δ 7.32−7.25 (m, 3H), 7.23−7.21
(m, 2H), 7.01 (s, 2H), 4.73−4.68 (m, 1H), 4.15−4.06 (m, 2H),
4.01 (d, 1H, J = 16), 3.87 (d, 1H, J = 16), 3.34−3.30 (m, 1H),
2.89−2.82 (m, 1H), 2.79−2.67 (m, 5H), 2.09 (s, 3H), 1.83−
1.59 (m, 8H), 1.28−1.24 (m, 6H).
13C NMR (100 MHz, CD3OD): δ 170.3, 168.6, 163.4, 154.1,
135.7, 132.2, 129.5, 128.7, 127.0, 121.5, 91.0, 84.4, 79.3, 66.3,
55.2, 48.5, 40.7, 37.2, 30.6, 28.2, 27.8, 25.8, 20.7, 13.5.
IR (thin film, KBr disc, cm−1): 2968, 2871, 1782, 1741, 1703,
1217.
13C NMR (75 MHz, CDCl3): δ 172.2, 153.2, 134.9, 129.4,
129.0, 127.4, 84.8, 72.3, 71.5, 66.2, 55.0, 49.6, 45.0, 37.8, 28.0,
27.3, 25.83, 25.80.
MS (CI): 517.3 (M − H)+
IR (thin film, KBr disc, cm−1): 3504, 3296, 2963, 2928, 1786,
1765, 1691, 1374, 1215.
HPLC purity (achiral): 97.4%; residual solvents 10.2%
(MTBE).
MS (EI): 324.5 (M − OH)+
(R)-5-((S)-4-Benzyl-2-oxooxazolidin-3-yl)-3-cyclopen-
tyl-5-(2,6-diethylpyridin-4-yl)-1-oxopentan-3-yl Acetate
Tosylate Salt (24·TsOH). The toluene solution of alkyne 23
prepared in the previous step (29.3 mmol, assumed 100% from
previous step, 56 kg of solution) was transferred to a 250-L
Hastelloy pressure reactor, rinsing with toluene (24.7 kg). It
was treated with activated carbon (2.3 kg, ∼15 wt %, Norit E
SUPRA USP grade carbon) at ambient temperature for at least
10 h. The carbon was removed by circulating through a
Gauthier-type filter until the filtrate was clear, rinsing with a
mixture of toluene (13.1 kg) and isopropanol (27.2 kg), and
then with a second portion of isopropanol (33.9 kg) into
holding drums. The Hastelloy reactor was then cleaned and
dried and then charged with the filtrate from the carbon
filtration. Ten percent of Pd/C (50 wt % in water, 3.0 kg on dry
basis, ∼20 wt %) was then charged, and the slurry was
hydrogenated at 3.5 barg (∼50 psig) and 30 °C, with
monitoring of hydrogen uptake. When hydrogen uptake has
leveled off and HPLC analysis showed adequate conversion
(specification ≤2% residual Z-olefin, observed 0.10%),27 the
system was purged with nitrogen, diluted with methanol (23
L), and filtered through a Gauthier-type filter to remove
catalyst, rinsing with toluene (26 kg) and methanol (23 L),
transferring to a 250-L, glass-lined steel reactor. The solution
was subjected to vacuum distillation (100 mbar, distillation
temperature ≤50 °C) to a final volume of ∼60 L (a total of
∼200 L distillate is collected). An additional portion of toluene
was added (118 kg), and vacuum distillation was resumed to a
final volume of ∼76 L, followed by a second portion of toluene
(89 kg) and another vacuum distillation to a volume of ∼100 L.
Analysis at this point showed adequate removal of water and
isopropanol (0.02% and 0.14%, respectively; specifications
NMT 0.3% water and 1.0% isopropanol). The solution was
diluted with THF (54 kg) and then treated with a solution of p-
toluenesulfonic acid monohydrate (5.6 kg, 29.3 mol) in THF
(26.8 kg), added over at least 60 min and rinsing with an
additional 26 kg of THF. The resulting slurry was stirred for at
least 30 min at 45 °C and then cooled to 20 °C and stirred for
6 h, then cooled to −10 °C and stirred for 6 h. The product was
collected by filtration, rinsing with two portions of THF (26 kg
each), and dried at 45 °C and 100 mbar on tray driers. The
product 24·TsOH was obtained as a white solid (11.95 kg, 17.2
mol, 58.9% yield).
HPLC purity: 99.0% (0.63% total HPLC impurities, 0.13%
water, 0.21% residue on ignition).
(R)-5-((S)-4-Benzyl-2-oxooxazolidin-3-yl)-3-cyclopen-
tyl-1-(2,6-diethylpyridin-4-yl)-5-oxopent-1-yn-3-yl ace-
tate (23). To a 250-L glass-lined, steel reactor was charged
toluene (32.0 kg), 4-bromo-2,6-diethylpyridine tosylate salt 4
(11.6 kg, 29.9 mol), and alkyne 5 (10.0 kg, 29.3 mol), and the
resulting slurry was stirred at ambient temperature for 10 min.
Triethylamine (8.90 kg, 87.9 mol, 3.0 equiv) was then added,
followed by a 5 L toluene line rinse. Copper(I) iodide (0.056
kg, 0.293 mol, 1.0 mol %) and PdCl2(Ph3P)2 (0.206 kg, 0.293
mol, 1.0 mol %) were added, and the vessel was purged with
nitrogen and warmed to 45 °C for 5 h. When HPLC analysis
indicated complete conversion (specification ≤2.0%, observed
0.3%), the slurry was cooled to −5 °C. Methanesulfonic acid
(15.5 kg, 161 mol, 5.5 equiv) was then added slowly so as to
maintain an internal temperature of 0−10 °C (the addition is
exothermic; on this scale 25 min was required to complete the
addition), followed by a line rinse with 5 L of toluene. After
addition the reaction was warmed to 15 °C and stirred for at
least 2 h. Acetic anhydride (6.0 kg, 59 mol, 2.0 equiv) was
charged (no significant exotherm), followed by a line rinse with
5 L toluene, and the reaction was warmed to 20 °C and stirred
for at least 12 h.
A quench solution was prepared by addition of sodium
carbonate (7.8 kg, 73.3 mol) to water (80 L) in a 100-L mobile
head tank and stirred to dissolve. When HPLC analysis
indicated complete reaction (specification ≤2.0% alcohol,
observed 0.38%), the reaction mixture was cooled to 10 °C,
and the aqueous Na2CO3 quench solution was added at a rate
to minimize foaming from the CO2 off-gassing and to maintain
an internal temperature ≤15 °C (2 h 10 min were required for
this addition). The resulting solution was warmed to 40−50 °C
and stirred for 2 h. The solution was allowed to settle for 30
min, and the pH of the aqueous phase checked and adjusted
with additional sodium carbonate if needed (target pH range is
3.5−4.5, observed 4.02). The lower aqueous phase was
separated and discarded. The 100-L mobile head tank was
charged with a second portion of sodium carbonate (1.6 kg, 15
mol) and water (40 L) and stirred to dissolve. This solution
was then added in portions, with 10 min stirring intervals
between charges, to reach an aqueous pH of 5−8 (observed pH
8). The mixture was then warmed to 40 °C (this assists with
the phase separation), and the lower aqueous phase was
separated. After cooling to ambient temperature, the product-
containing toluene phase was stored for use in the subsequent
hydrogenation.
Data for oxazolidinone 24·TsOH prepared in a laboratory
pilot: mp 98−105 °C (initial, loss of THF solvate), 130−133
°C (mp of desolvated solid)
1H NMR (400 MHz, CD3OD): δ 7.70−7.65 (m, 4H), 7.35−
7.19 (m, 7H), 4.74 (m, 1H), 4.22−4.15 (m, 3H), 3.18 (m, 3H),
H
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
3.01 (q, 4H, J = 7.6), 2.92−2.88 (m, 2H), 2.65 (m, 1H), 2.52
(m, 1H), 2.36 (s, 3H), 2.24 (m, 1H), 1.95 (s, 3H), 1.90−1.55
(m, 8H), 1.40 (t, 6H, J = 7.6).
A solution of citric acid (43.7 kg, 308 mol) in water (180 L)
was prepared separately and added slowly to the aqueous
product solution until a pH of 6.5−7.5 was reached (target 7.0,
achieved 7.02, initial pH 12.4). The solution was stirred for 30
min at 20 °C, during which time the product began to
crystallize. Additional aqueous citric acid was added to reduce
the pH to 5.5−6.5 (target 6.0, achieved 5.61), and the slurry
was stirred for at least 30 min. The solids were then collected
on an agitated filter/drier, rinsing with water (120 L) and
methyl ethyl ketone (120 L), and dried at 35 °C and 100 mbar
until the specified water content was reached (specification
NMT 10.0%, final value 6.2%). The product 2·H2O was
obtained as a white solid (40.1 kg, 111 mol, 64% yield).
13C NMR (100 MHz, CD3OD): δ 172.0, 170.3, 164.8, 157.8,
154.6, 142.4, 140.5, 135.9, 129.5, 128.7, 128.6, 127.0, 125.8,
123.6, 85.8, 66.2, 55.3, 47.5, 38.4, 37.4, 35.6, 31.0, 26.52, 26.50,
25.42, 25.37, 20.7, 20.2, 12.4.
IR (thin film, KBr disc, cm−1): 2943, 2867, 1780, 1736, 1704,
1166.
MS (CI): 521.3 (M − H)+
HPLC purity (achiral): 99.2%; TsOH content 22.7% (theory
24.8%); K−F 0.18% water; residual solvents 5.1% (4.3% tol,
0.8% THF).
1
Data for 2·H2O obtained from a laboratory pilot: H
(R)-6-Cyclopentyl-6-(2-(2,6-diethylpyridin-4-yl)ethyl)-
dihydro-2H-pyran-2,4(3H)-dione Hydrate (2·H2O). Salt
Break To Generate Oxazolidinone 24. To a 1600-L, glass-
lined steel reactor were charged toluene (623 kg) and
oxazolidinone 24·TsOH (60.0 kg, 86.6 mol). In another 630-
L, glass-lined steel reactor were combined water (300 L) and
NaHCO3 (14.6 kg, 173 mol, 2.0 equiv), and stirring was
initiated until all visible solids were dissolved. The aq NaHCO3
solution was transferred to the tosylate salt slurry over ∼30
min, maintaining an internal temperature of 15−25 °C, and
stirring was continued for 60 min. Stirring was halted, the layers
were allowed to settle, and the lower aqueous phase was
removed. A vacuum distillation of the toluene layer was then
performed until the water content was NMT 0.05% by Karl
Fischer analysis (50 mbar, maximum pot temperature 41 °C,
final volume ∼500 L). The resulting solution was cooled to 20
°C and held for the cyclization.
Preparation of LHMDS Solution. A 630-L, glass-lined steel
reactor was charged with toluene (242 kg), and assayed for
water content. If >0.05% then an atmospheric distillation was
performed in 25 L portions until the water content was
≤0.05%. When the water content IPC was met, LHMDS (25
wt % in toluene, 162 kg, 243 mol, 2.80 equiv) was added, and
the contents were cooled to 0 °C with stirring.
NMR (400 MHz, CD3OD): (note that the protons on C3 of
the β-keto lactone are not visible due to deuterium exchange
with solvent): δ 7.08 (s, 2H), 2.77 (q, 4H, J = 7.6), 2.70−2.65
(m, 3H), 2.43−2.42 (m, 2H), 2.09−2.06 (m, 2H), 1.75−1.4
(m, 8H), 1.27 (t, 6H, J = 7.6).
13C NMR (100 MHz, CD3OD): (note that the C3 β-keto
lactone resonance is broadened due to deuterium exchange): δ
175.9, 170.3, 161.9, 155.7, 120.5, 89 (br), 83.8, 47.0, 37.8, 33.6,
29.6, 29.5, 26.8, 26.7, 25.5, 13.4.
IR (thin film, KBr disc, cm−1): 3294 (br), 2963, 2870, 1656,
1605.
MS (CI): 344.2 (M − H)+
HPLC purity (achiral): 99.5%; 5.1% water (theory for
monohydrate 5.0%); residual solvents 0.27% (MeTHF and
MeOH).
AUTHOR INFORMATION
■
Corresponding Authors
*robert.a.singer@pfizer.com
*john.a.ragan@pfizer.com
Present Address
§Canterbury Christ Church University, North Holmes Road,
Canterbury, Kent, U.K.
Cyclization (via Dual Stream Addition of LHMDS and
Oxazolidinone 24). A 2500-L, glass-lined steel reactor was
charged with toluene (474 kg) that had been precharged to the
two addition tanks and rinsed through the transfer lines (during
this transfer the addition valves were calibrated to deliver 5.0 L/
min at 0.4 barg). An atmospheric distillation was then
performed until the water content met specification (spec
≤0.05%, assayed value initially 0.1%, and <0.05% after
distillation of ∼50 L). With stirring, the toluene was cooled
to −17 °C. The solutions of oxazolidinone 24 and LHMDS
were then added simultaneously at a rate of 5.0 L/min over
∼100 min. The maximum internal temperature during this
addition was −9 °C. Each addition tank was rinsed with an
additional portion of toluene (13 kg), and the solution was
stirred for an additional 30 min at −17 °C.
The reaction was quenched by the addition of water (360 L)
and stirred at 3 °C for 30 min. The lower aqueous phase
(containing product) was separated and transferred to a 1600-
L, glass-lined steel reactor. The organic phase was extracted
with an additional portion of water (60 L), and this aqueous
extraction was transferred to the same reactor.
The entire process is then repeated on the same scale as
described above. The total input of 24·TsOH is thus 120.0 kg
(173.2 mol).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Kris Jones for demonstrating the diastereoselective
aldol reaction on lab scale prior to scale up. We thank BASF
and Asymchem for scaling the production of oxazolidinone 5 to
>100 kg. We are grateful to the Pfizer Structure Elucidation
Group for assisting with structural characterization of process
impurities.
REFERENCES
■
(1) Sarbah, S. A.; Younossi, Z. M. J. Clin. Gastroenterol. 2000, 30, 125.
(2) Hayashi, N.; Takehara, T. J. Gastroenterol. 2006, 41, 17.
(3) Li, H.; Tatlock, J.; Linton, A.; Gonzalez, J.; Jewell, T.; Patel, L.;
Ludlum, S.; Drowns, M.; Rahavendran, S. V.; Skor, H.; Hunter, R.; Shi,
S. T.; Herlihy, K. J.; Parge, H.; Hickey, M.; Yu, X.; Chau, F.;
Nonomiya, J.; Lewis, C. J. Med. Chem. 2009, 52, 1255.
(4) Grillot, A.-L.; Farmer, L. J.; Rao, B. G.; Taylor, W. P.; Weisberg, I.
S.; Jacobson, I. M.; Perni, R. B.; Kwong, A. D. Discovery and
Development of Telaprevir. In Antiviral Drugs: From Basic Discovery
Through Clinical Trials; Kazmierski, W. M., Ed.; John Wiley & Sons,
Inc.: New York, 2011; p 209.
(5) Njoroge, F. G.; Chen, K. X.; Shih, N.-Y.; Piwinski, J. J. Acc. Chem.
Res. 2008, 41, 50.
Crystallization of Hydrate 2·H2O. The combined aqueous
layers from the two batches were diluted with ethanol (284 kg).
(6) Jensen, D. M. N. Engl. J. Med. 2011, 364, 1272.
I
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(7) Camp, D.; Matthews, C. F.; Neville, S. T.; Rouns, M.; Scott, R.
W.; Truong, Y. Org. Process Res. Dev. 2006, 10, 814.
(8) Karpf, M.; Dreiding, A. S. Helv. Chim. Acta 1979, 62, 852.
(9) Griffith, J. C.; Jones, K. M.; Picon, S.; Rawling, M. J.; Kariuki, B.
M.; Campbell, M.; Tomkinson, N. C. O. J. Am. Chem. Soc. 2010, 132,
14409.
(10) Lafrance, D.; Bowles, P.; Leeman, K.; Rafka, R. Org. Lett. 2011,
13, 2322.
(11) Evans, D. A.; Nelson, J. V.; Taber, T. R. Top. Stereochem. 1982,
13, 1.
(12) Cowden, C. J.; Paterson, I. Org. React. (New York) 1997, 51, 1.
(13) Zucca, C.; Bravo, P.; Malpezzi, L.; Volonterio, A.; Zanda, M. J.
Fluorine Chem. 2002, 114, 215.
(14) Liu, W.; Sheppeck, J. E., II; Colby, D. A.; Huang, H.-B.; Nairn,
A. C.; Chamberlin, A. R. Bioorg. Med. Chem. Lett. 2003, 13, 1597.
(15) Fuerstner, A.; Bouchez, L. C.; Morency, L.; Funel, J.-A.; Liepins,
V.; Poree, F.-H.; Gilmour, R.; Laurich, D.; Beaufils, F.; Tamiya, M.
Chem.Eur. J. 2009, 15, 3983.
(16) Theurer, M.; Fischer, P.; Baro, A.; Nguyen, G. S.; Kourist, R.;
Bornscheuer, U.; Laschat, S. Tetrahedron 2010, 66, 3814.
(17) Peters, R.; Althaus, M.; Diolez, C.; Rolland, A.; Manginot, E.;
Veyrat, M. J. Org. Chem. 2006, 71, 7583.
(18) Song, J. J.; Xu, J.; Tan, Z.; Reeves, J. T.; Grinberg, N.; Lee, H.;
Kuzmich, K.; Feng, X.; Yee, N. K.; Senanayake, C. H. Org. Process Res.
Dev. 2007, 11, 534.
(19) Ager, D. J.; Allen, D. R.; Schaad, D. R. Synthesis 1996, 1283.
(20) Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79,
1920.
(21) Matthews, C. F.; Scott, R. W.; Tucker, J. L.; (Pfizer Inc., USA).
Application: WOWO, 2007, p 79.
(22) Johnson, S.; Drowns, M.; Tatlock, J.; Linton, A.; Gonzalez, J.;
Hoffman, R.; Jewell, T.; Patel, L.; Blazel, J.; Tang, M.; Li, H. Synlett
2010, 796.
(23) Chandra, K. L.; Saravanan, P.; Singh, R. K.; Singh, V. K.
Tetrahedron 2002, 58, 1369.
(24) Reetz, M. T.; Maier, W. F. Liebigs Ann. Chem. 1980, 1471.
(25) Hoepker, A. C.; Gupta, L.; Ma, Y.; Faggin, M. F.; Collum, D. B.
J. Am. Chem. Soc. 2011, 133, 7135.
(26) Schreiber, S. L.; Ragan, J. A.; Standaert, R. F. Strategies Tactics
Org. Synth. 1991, 3, 417.
(27) The first IPC showed incomplete conversion, and after filtration
the batch was resubjected to the same conditions until the IPC was
met. This behavior was typical for this problematic hydrogenation and
was improved dramatically with the Ester Dieckmann approach
described in Part II.
J
dx.doi.org/10.1021/op4002356 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX