Development of scalable syntheses of selective PI3K inhibitors

Article abstract of DOI:10.1021/op100286g

On the basis of a more practical and scalable route to an iodothiophene, an efficient and reliable synthesis has been developed for three selective PI3K inhibitors. From this advanced intermediate, the three title compounds were each prepared in five additional steps. Key learnings also include: high throughput experimentation (HTE) screening toward a more robust Suzuki coupling, a more efficient triazole synthesis, and an acid/base cleanup developed to purify the final compounds. The final enabled synthesis required no column chromatography.

Full text of DOI:10.1021/op100286g

ARTICLE
pubs.acs.org/OPRD
Development of Scalable Syntheses of Selective PI3K inhibitors
Qinhua Huang,* Paul F. Richardson, Neal W. Sach, Jinjiang Zhu, Kevin K.-C. Liu, and Graham L. Smith
Pfizer Global R&D at San Diego, 10646 Science Center Drive, San Diego, California 92121, United States
Daniel M. Bowles
Pfizer Global R&D, Groton Laboratories, MS 8156-064 Eastern Point Road, Groton, Connecticut 06340, United States
ABSTRACT: On the basis of a more practical and scalable route to an iodothiophene, an efficient and reliable synthesis has been
developed for three selective PI3K inhibitors. From this advanced intermediate, the three title compounds were each prepared in five
additional steps. Key learnings also include: high throughput experimentation (HTE) screening toward a more robust Suzuki
coupling, a more efficient triazole synthesis, and an acid/base cleanup developed to purify the final compounds. The final enabled
synthesis required no column chromatography.
’ INTRODUCTION
Inhibition of the PI3K receptor may result in an effective
therapeutic intervention for the treatment of some types of
cancer.1 Our early research has identified compounds 1, 2, and
3 as highly promising candidates in our selective PI3K program
(Figure 1).
The initial route for analogue generation in this PI3K triazole
series is outlined in Scheme 1.2 Unfortunately, there are several
Figure 1. Potent selective PI3K inhibitors 1, 2, and 3.
disadvantages associated with this route: (1) the aryl moiety, the
point of diversity between the three compounds, is introduced in
the first step, and as such they could not share a common
Scheme 1. Original route to triazole analogues for PI3K
project
intermediate; (2) while 2,4-dichloroacetophenone is a commer-
cially available starting material,3 we were required to synthesize
2-chloro-4-methoxyacetophenone and 4-cyano-2-fluoroaceto-
phenone;4 and (3) the original medicinal chemistry route is
hampered by several steps which are not amenable for a large-
scale synthesis. In addition, the initial halogenation step suffers
from rapid bis-bromination, and the modified Gewald reaction
(Step 4, with a messy three-component mixture)5 is not a conve-
nient step to the formation of tetra-substituted thiophenes.
To carry out further in vivo studies on these three compounds,
gram amounts of each candidate were required. Therefore, a
facile route was highly desired not only for serving as a more
general and efficient route to access analogues of this series but
also for the purpose of scale-up. In this work we describe a more
efficient and scalable route to compounds 1-3. While the syn-
theses for compounds 1 and 2 were improved significantly, the
modified preparation of compound 3 contains improvements
that would allow further scaling into a pilot-plant setting with
minor changes.
Our retrosynthetic plan is outlined in Scheme 2. We envi-
sioned that compounds 1-3 could be synthesized from amides
due to the steric bulkiness and the poor intrinsic reactivity of
iodothiophenes. We found that intermediate 6 can be con-
structed rapidly over five steps from commercially available
malononitrile. Alternatively, amide 7 could be utilized as a more
4a, 4b, and 4c, respectively, followed by triazole formation.
Amides 4a-4c could be derived from esters 5a-5c via amino-
lysis. Suzuki coupling of common intermediate 6 with the
corresponding aryl boronic acids would generate esters 5a-5c.
On the basis of our previous experience in this series, we expected
that this Suzuki reaction could potentially be a problematic step
Received: October 25, 2010
Published: March 11, 2011
r
2011 American Chemical Society
556
dx.doi.org/10.1021/op100286g Org. Process Res. Dev. 2011, 15, 556–564
|

Organic Process Research & Development
ARTICLE
Scheme 2. Retrosynthetic analysis of compounds 1, 2 and 3
Scheme 3. Second-generation synthesis of iodothiophene 6
advanced common intermediate (route 2 in Scheme 2). The
Suzuki couplings of both ester 6 and amide 7 have been
examined, and the results are discussed herein.
converted to sulfoxide 11 using Oxone in aqueous MeOH.8
However, subsequent oxidation to the desired sulfone 12 was
very sluggish even when a large excess of Oxone was employed.
Fortunately, both sulfoxide 11 and sulfone 12 react smoothly
with morpholine to afford the desired product. Therefore, a
mixture of sulfoxide 11 and sulfone 12 (∼1:1 ratio) was treated
with morpholine in THF to afford the desired thiophene core 6,
which was precipitated out of the THF solution with water in
good purity (>98%) and yield (72% over two steps).
Despite the use of these hazardous reagents, the second-
generation synthesis afforded a convenient avenue to 6, so we
set out to further optimize the route to make it more process
friendly (Scheme 4). Replacement of NaH with K2CO3 elimi-
nated the primary risk associated with the step 1 condensation
reaction,9 and although the reaction still used MeI and CS2, these
reagents were charged in only near-stoichiometric quantities,
reducing the risk of exposure and flammability. Replacement of
THF with DMSO afforded a higher recovery of cleaner com-
pound 8, and we also observed that the yield of the Sandmeyer
’ RESULTS AND DISCUSSION
In an attempt to develop a more direct second-generation
synthesis, we found that iodothiophene 6 could be synthesized in
five steps from commercially available malononitrile (Scheme 3).3
Although the modified synthesis was relatively efficient, the
process suffered from multiple nonscalable steps, relying on
hazardous reagents such as NaH, CS2, and MeI, as well as
Sandmeyer chemistry. Treatment of malononitrile with NaH
followed by CS2 and MeI generated intermediate 8 in quantita-
tive yield, which was isolated by water dilution. Condensation of
compound 8 and ethyl thioacetate in the presence of Et3N
afforded aminothiophene 9 in 69% yield.6 Subsequently, the
Sandmeyer reaction of compound 9 using CH2I2 in CH3CN
gave iodothiophene 10 in 50% yield.7 Thiophene 10 was quickly
557
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
Scheme 4. Modified second-generation synthesis of 6
reaction improved from 50% to 78% upon scale-up. Calorimetric
examination allowed us to define the limits of the safe operation
of these steps sufficiently to safely undertake a 1 kg preparation10
of iodothiophene 6.
was investigated (Scheme 6). Under various conditions exam-
ined (once again a high throughput screen of approximately 100
conditions was carried out), the coupling reaction of amide 7
always resulted in both low conversion with significant amounts
of the des-iodo side product. We suspected that the poor solu-
bility of amide 7 was the culprit for the unsatisfactory Suzuki
coupling reactions. The desired coupling product from amide 7
also has poor solubility, and this caused difficulties with purifica-
tion. Therefore, the ester 6, rather than the amide 7, was used for
the Suzuki couplings to access 1-3.
Esters 5a and 5b were converted to acids 13a and 13b via
treatment with aqueous LiOH in water, and were readily preci-
pitated out of solution after pH adjustment. However, the
conversion of 5c to 13c did not work well under these conditions
(<40% yield). After undergoing reaction screening, it was
determined that saponification using aqueous NaOH in a mixture
of EtOH and 2-MeTHF gave smooth conversion without decom-
position of the sensitive aryl-fluoride group, but additional
decomposition was observed upon pH adjustment with HCl.
Inverse addition of the basic product solution into a mixture of
aqueous HCl and NaCl solved this problem, allowing isolation of
13c in 84% yield on 84 g scale (see Scheme 7).
The subsequent aminolysis reaction was carried out by first
activating the acid with 2-chloro-4,6-dimethoxy-[1,3,5]triazine
(CDMT) followed by treatment with gaseous ammonia to afford
amides 4a-4c.16 The primary amides were purified by tritura-
tion with EtOAc and saturated aqueous NaHCO3 in the case of
4a and 4b, and using a MeOH reslurry in the case of 4c, each to
afford product with over 95% purity. Compounds 4a-4c were
then treated with N,N-dimethylformamide dimethyl acetal
(DMF-DMA) to afford 14a-14c, which were purified via a
MeOH reslurry.
With key intermediate 6 in hand, we screened for the optimal
Suzuki coupling conditions,11 but unfortunately these reactions
typically generated significant amounts of des-iodo material and
suffered from consistently low yields. Furthermore, the desired
coupling products had poor solubility in organic solvents and
displayed polarity similar to that of the unreacted starting materials,
resulting in a problematic purification for a large-scale synthesis.
Therefore, a high-throughput experimentation methodology
(HTE) was initiated to identify optimal Suzuki conditions for
this particular substrate-reactant combination.12 A variety of
catalysts [Pd(PtBu3)2, Pd(PPh3)4, Pd(dppf)Cl2], bases (CsF,
DIPEA, K2CO3, K3PO4), and solvents (toluene, dioxane, EtOH,
water) were screened on 10 mg scale in varying stoichiometric
ratios at 80 °C for 16 h. The results were processed and analyzed
by Spotfire software and are illustrated in Scheme 5. In general,
Pd(PtBu3)2 was found to be superior to both Pd(PPh3)4 and
Pd(dppf)Cl2, whilst bases CsF and DIPEA gave better coupling
results than K2CO3 and K3PO4. Of the 200 screened reactions,
three potential generic condition sets were identified for each set
of coupling partners (Scheme 5). Unfortunately, a gram-scale
test using Hunig’s base resulted in much lower conversion and
more des-iodo side products than in the corresponding screening
reaction on 10 mg scale, possibly due to relatively easier Pd
aggregation and precipitation in larger-scale reactions. Even-
tually, we found the following conditions to be effective for all
substrates: 1.2 equiv boronic acid, 5 mol % Pd(PtBu3)2, 1.2 equiv
CsF, dioxane/water/toluene, 85 °C.13 Remarkably, for both 2,4-
dichlorophenylboronic acid and 2-chloro-4-methoxylphenyl-
boronic acid, Suzuki coupling afforded over 98% conversion.
On small scale, residual palladium was removed using Si-thiol
resin14 followed by aqueous extraction to obtain coupling products
5a and 5b as yellow solids with good purity. In the case of 5c,
extraction into toluene followed by cooling and crystallization
resulted in isolation of the desired product in 90% yield in >98%
purity on 250 g scale without the need for Si-thiol treatment.15
In addition, the Suzuki coupling of amide 7 with 2,4-dichlor-
ophenylboronic acid and 2-chloro-4-methoxyphenylboronic acid
When a solution of 14a-14c in 5:1 EtOH/HOAc was treated
in one portion with 10 equiv NH2NH2 H2O, on small scale
3
(∼500 mg) at 60 °C, desired triazoles 1-3 were obtained in near
quantitative yields (see Scheme 7). However, upon scaling up
(5.0 g), the same process resulted in 50-70% of the desired
triazole compounds along with 20-30% of the corresponding
hydrolyzed amides (4a-4c). We hypothesize that, under these
conditions, NH2NH2 was rapidly protonated to form an acetate
salt, generating an exotherm17 that might have led to decomposition
558
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
Scheme 5. Key Suzuki coupling to form esters 5a-5ca
a Screening conditions: 6 (10 mg, 0.025 mmol, 1.0 equiv), aryl boronic acid (0.030 mmol, 1.2 equiv), Pd catalyst (0.1 M in toluene, 25 μL, 0.1 equiv),
base (1 M aqueous solution, 75 uL, 3.0 equiv), solvent (300 μL), reaction concentration 0.06 M, 80 °C, overnight.
Scheme 6. Attempted Suzuki coupling to form amide 7
of 14a-14c in aqueous EtOH. The problem was resolved by
reversing the order of reagent addition and by carefully monitor-
ing the heat of reaction during slow addition of the aqueous
NH2NH2 solution. In the modified procedure, NH2NH2 was
added dropwise into a solution of 14a, 14b, or 14c in EtOH/
AcOH whilst carefully maintaining the internal temperature at -
10-0 °C. Heat generation was addition controlled, and as the
NH2NH2 reacted quickly during the slow addition, formation of
the corresponding amide impurity was virtually eliminated.
Incorporation of this modified procedure enabled the synthesis
of 3 in a more scalable manner, affording an 86% yield of high-
purity 3 using only 5 equiv of NH2NH2.18,19
Early on we found it necessary to purify compounds 1-3 via
silica gel chromatography. However, we later discovered that
triazole products 1-3 could be purified by extraction into
aqueous caustic, allowing removal of soluble organic impurities
by extraction with EtOAc. The pH of the resulting aqueous
solution was then adjusted to pH 4-5 with HOAc, and the
desired products were extracted into the organic layer. Solvent
removal afforded triazoles 1 and 2 with greater than 98% purity.
559
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
Scheme 7. Optimized synthesis of 1, 2, and 3
Compound 3 was isolated in a similar manner, although the basic
product solution was reacidified to pH 5 with 1 N HCl20 and
filtered to afford 3 directly in 86% yield on 50-g scale.21,22
product was triturated with MeOH (100 mL) to afford 77 g (100%)
of 8 as a white solid with spectral properties identical to those
previously reported.9a
Modified Laboratory Procedure. A suspension of K2CO3
(1.15 kg, 8.33 mol) in DMSO (5 L) at 20 °C was treated with
malononitrile (0.5 kg, 7.57 mol) in one portion. The reactor was
cooled to 0 °C, and CS2 (0.50 L, 8.33 mol) was added over
30 min at 18-22 °C. The resulting yellow suspension was stirred
at 20 °C for 2 h and then cooled to 0 °C over 20 min. MeI (0.77 L,
15.1 mol) was added over 30 min with pot temperature <5 °C.
The reaction was warmed to 20 °C and stirred for 12 h. The
resulting slurry was poured into ice water (12 L) to afford yellow
solids that were collected by filtration, washed with water (2 ꢀ
1 L) and MeOH (500 mL), and dried to afford 8 (1.28 kg,
7.52 mol, 99%) as a white solid with spectral properties identical
to those previously reported.9a
’ CONCLUSIONS
In conclusion, a general and scalable route has been success-
fully developed to synthesize PI3K inhibitors 1, 2, and 3 from
inexpensive commercially available starting materials. A few key
advantages and highlights associated with this route over the
original medicinal chemistry route include: (1) employment of
an advanced common intermediate strategy (i.e., 6) to enable
synthesis of multiple analogues on gram scale, (2) optimization
of the key Suzuki coupling step using a high throughput experi-
mentation approach, and (3) column chromatography was not
needed. A scalable synthesis for this series was demonstrated in
the synthesis of candidate 3.23
3-Amino-4-cyano-5-methylsulfanyl-thiophene-2-carboxylic
Acid Ethyl Ester (9). A suspension of 2-(bis-methylsulfanyl-methy-
lene)malononitrile (8) (2.58 kg, 15.1 mol) in EtOH (12 L) at 20 °C
was treated with ethyl 2-mercaptoacetate (1.66 L, 15.1 mol) in one
portion. The reactor was cooled with an ice bath, and Et3N (2.11 L,
15.1 mol) was added via addition funnel over 45 min with pot
temperature <20 °C. After being stirred at 20 °C for 12 h, the
resulting slurry was filtered to afford yellow solids that were washed
with EtOH (2 ꢀ 250 mL) and dried on the filter under vacuum for
12 h to afford 9 as a pale-yellow solid (2.47 kg, 10.1 mol, 67.4%
yield). 1H NMR (400 MHz, CDCl3) 1.34 (t, J = 7.05 Hz, 3 H) 2.65
(s, 3 H) 4.29 (q, J = 7.22 Hz, 2 H) 5.78 (br s, 2 H).
4-Cyano-3-iodo-5-methylsulfanyl-thiophene-2-carboxylic
Acid Ethyl Ester (10). To a stirred suspension of 3-amino-4-
cyano-5-methylsulfanyl-thiophene-2-carboxylic acid ethyl es-
ter (9) (2.47 kg, 10.2 mol) in CH3CN (10 L) was added
CH2I2 (1.25 L, 15.3 mol) in one portion. The reaction was
heated to 70 °C, and isoamyl nitrite (2.14 L, 15.3 mol) was
added via addition funnel at a rate to maintain the pot
temperature between 60 and 70 °C (50 mL was added rapidly
to initiate the reaction, and the remainder was added over
approx 120 min). After complete addition, the reaction
mixture was cooled to 20 °C and stirred for 12 h, then cooled
to 0 °C and stirred for 30 min. The resulting solids were
collected by filtration, washed with cold CH3CN (0 °C, 2 ꢀ
250 mL), and dried on the filter under vacuum for 12 h to
afford 10 as a yellow solid (2.8 kg, 7.93 mol, 78% yield). 1H
NMR (400 MHz, CDCl3) 1.34 (t, J = 7.18 Hz, 3 H) 2.64 (s,
3 H) 4.32 (q, J = 7.05 Hz, 2 H).
’ EXPERIMENTAL SECTION
General. HPLC assays were performed on Agilent 1100
chromatographs equipped with PDA detectors. Routine react-
ion monitoring was done using a Phenomenex Prodigy ODS3
4.6 mm ꢀ 50 mm column or an equivalent C-18 column and a
standard 8-min gradient program: 10:90 to 90:10 mixture of
acetonitrile-water containing 0.02% of TFA with constant flow
of 1 mL/min. LC/MS data were obtained on an Agilent 1100 LC
system with an Agilent 1100 LC/MS detector. NMR spectra
were recorded on a Bruker Avance DPX300 or a Bruker Avance
DRX400 NMR spectrometer. HRMS data were gathered on an
Agilent 1100 LC with MSD TOF (Agilent model G1969A) mass
spec detectors running with electrospray spray ionization source.
The purity of final APIs was determined by HPLC.
2-(Bis-methylsulfanyl-methylene)malononitrile (8). Original
Laboratory Procedure. To a suspension of NaH (60% on mineral oil,
40 g, 989 mmol, 2.20 equiv) in THF (500 mL) at 0 °C was added
malononitrile (30.0 g, 450 mmol, 1.0 equiv) at <5 °C. The resulting
mixture was stirred at 0 °C for 10 min, followed by addition of CS2
(30 mL, 495 mmol, 1.1 equiv) over 20 min. (Caution: significant
effervescence observed.) The resulting deep-orange solution was
allowed to warm to 25 °C for 4 h and was recooled to 0 °C. MeI
(57 mL, 899 mmol, 2.0 equiv) was added (again significant
effervescence) over 10 min, and the reaction mixture turned yellow.
The resulting mixture was stirred at 25 °C for 12 h and was then
poured into ice water (400 mL) to precipitate a yellow solid that was
filtered, washed with water, and dried on the filter for 1 h. The crude
560
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
4-Cyano-3-iodo-5-methanesulfinyl-thiophene-2-carboxylic
Acid Ethyl Ester (11) and 4-Cyano-3-iodo-5-methanesulfonyl-
thiophene-2-carboxylic Acid Ethyl Ester (12). A suspension of
4-cyano-3-iodo-5-methylsulfanyl-thiophene-2-carboxylic acid ethyl
ester (10) (900 g, 2.65 mol, 1.0 equiv) in MeOH (6 L) at 0 °C was
treated with an aqueous solution of oxone (3.58 kg, 5.84 mol, 2.20
equiv in 8 L water) via addition funnel over 1 h with pot temper-
ature <25 °C. The resulting mixture was stirred at 25 °C for 12 h
and diluted with water (4 L). The resulting solids were collected by
filtration, washed with water (2 ꢀ 1 L), and dried in vacuo at 40 °C
for 12 h to afford 950 g of a crude mixture of products (sulfoxide 11
and sulfone 12), that was carried into the next step without further
purification.
4-Cyano-3-(2,4-dichloro-phenyl)-5-morpholin-4-yl-thio-
phene-2-carboxylic Acid (13a). To a slurry of 4-cyano-3-(2,4-
dichloro-phenyl)-5-morpholin-4-yl-thiophene-2-carboxylic acid
ethyl ester (5a) (11.0 g, 8 mmol) in MeOH (120 mL) was
added a solution of LiOH (3.2 g, 134 mmol, 5 equiv) in water
(30 mL) at 25 °C. The resulting slurry was stirred at 25 °C for 3 h
to reach completion. The volatiles were removed under reduced
pressure. The aqueous solution solidified during concentration.
This white slurry was diluted with water (150 mL) and adjusted
to pH ∼3 with 2 N HCl. The solids were collected by filtration,
washed with water and a minimal amount of MeOH, and dried
under vacuum at 50 °C for 7 h to provide 10.5 g (98%) of 13a as a
1
pale-yellow solid. H NMR (400 MHz, DMSO-d6) 3.51-3.69
4-Cyano-3-iodo-5-morpholin-4-yl-thiophene-2-carboxylic
Acid Ethyl Ester (6). A solution of 11 and 12 (approx. 2.65 mol,
crude from previous step) in THF (8 L) at 0 °C was treated with
morpholine (1.21 L, 13.8 mol) via addition funnel over 10 min
withthe pot temperature<10°C. The solutionwas stirred at25 °C
for 12 h. The resulting slurry was charged with water (6 L), cooled
to 5 °C, and stirred for 30 min. The solids were collected by
filtration, washed with water (2 ꢀ 500 mL), triturated with MeOH
(500 mL), and dried under vacuum at 30 °C for 2 h to afford 0.8 kg
(m, 4 H) 3.72-3.85 (m, 3 H) 7.39 (d, J = 8.31 Hz, 1 H) 7.50 (dd,
J = 8.31, 2.27 Hz, 1 H) 7.74 (d, J = 2.01 Hz, 1 H) 12.26-13.39
(m, 1 H).
4-Cyano-3-(2,4-dichloro-phenyl)-5-morpholin-4-yl-thio-
phene-2-carboxylic acid amide (4a). To a solution of 4-cyano-
3-(2,4-dichloro-phenyl)-5-morpholin-4-yl-thiophene-2-carboxylic
acid (13a) (15 g, 39 mmol, 1.0 equiv) in DMF (500 mL) was
added CDMT (7.71 g, 43 mmol, 1.1 equiv) and N-methyl mor-
pholine (NMM) (9.1 mL, 82.5 mmol, 2.0 equiv). The resulting
mixture was stirred at 25 °C for 2 h. Ammonia gas was then
bubbled through the suspension for 1 h, and the mixture was
allowed to stir at 25 °C for 2 h to reach completion. To the
reaction mixture was added 1.5 L water and solids precipitated
out. The solid was collected by filtration and triturated with a
mixture of EtOAc (50 mL) and saturated aq NaHCO3 (50 mL).
The resulting solids were filtered, washed with water and EtOAc,
and dried under vacuum to afford 13 g (87%) of 4a as a pale
yellow solid. LC-MS (APCI, Mþ þ 1) 382.2; 1H NMR
(400 MHz, DMSO-d6) 3.48-3.58 (m, 4 H) 3.69-3.85 (m, 4
H) 7.43 (d, J = 8.31 Hz, 1 H) 7.50-7.60 (m, 1 H) 7.78 (d, J =
2.01 Hz, 1 H) 9.31 (s, 2 H).
1
(77% over 2 steps) of 6 as a white solid. MP 186-187 °C; H
NMR (400 MHz, CDCl3) 1.37 (t, J = 7.18 Hz, 3 H) 3.55-3.70 (m,
4H) 3.82-3.94 (m, 4 H) 4.34 (q, J=7.22Hz, 2H);13CNMR(100
MHz, DMSO-d6) 14.1, 50.0, 61.1, 65.0, 95.7, 96.1, 113.2, 117.1,
159.5, 168.5; IR (neat, cm-1) 3055, 2984, 2211, 1709, 1494, 1265;
HRMS Calcd for C12H13IN2O3S: 391.9692. Found: 391.9680.
Screening Procedure for Key Step Suzuki Coupling. Com-
pound 6 and the corresponding boronic acid (1:1.2 molar ratio)
were suspended in THF. An appropriate amount of suspension
containing compound 6 (10 mg, 0.025 mmol, 1.0 equiv) and the
corresponding boronic acid (0.030 mmol, 1.2 equiv) was trans-
ferred into 0.8 mL-capacity vials charged with small magnetic stir
bars. The vials were placed in 96 vial plates, and all solvents were
removed under vacuum overnight to afford neat dry solids. To
the vials were added an appropriate solvent (300 μL), base (1 M
aqueous solution, 75 μL, 3.0 equiv), and catalyst (0.1 M in
toluene, 25 μL, 0.1 equiv). The resulting mixtures were stirred at
80 °C overnight. The screening plates were cooled to room
temperature, and DMF (300 μL) was added to each vial. After
being stirred at room temperature for 30 min, the mixtures were
sonicated and analyzed directly by LC/MS. The resulting LC/
MS data were processed and analyzed using Spotfire software.
4-Cyano-3-(2,4-dichloro-phenyl)-5-morpholin-4-yl-thio-
phene-2-carboxylic Acid Ethyl Ester (5a). A suspension of
4-cyano-3-iodo-5-morpholin-4-yl-thiophene-2-carboxylic acid ethyl
ester (6) (15 g, 38 mmol, 1.0 equiv), 2,4-dichlorophenylboronic
acid (9.12 g, 45.9 mmol, 1.2 equiv), CsF (17.6 g, 115 mmol,
3.0 equiv), Pd(PtBu3)2 (977 mg, 1.91 mmol, 0.05 equiv), dioxane
(240 mL), toluene (15 mL), and water (60 mL) was flushed with
N2 and warmed up to 85 °C. After being stirred at 85 °C for 12 h,
the reaction reached completion. To the reaction mixture was
added Si-thiol (∼10 g), and the mixture was stirred at 25 °C for
3 h to remove Pd catalyst. The resulting mixture was then diluted
with EtOAc and filtered through Celite. The filtrate was washed
with brine, dried over Na2SO4, filtered, and concentrated to
afford 11.0 g of 5a in 70% yield with over 85% purity as a yellow
solid. 1H NMR (400 MHz, DMSO-d6) 1.13 (t, J = 7.18 Hz, 3 H)
3.62-3.67 (m, 4 H) 3.85-3.91 (m, 4 H) 4.08-4.18 (m, 2 H)
7.21 (d, J = 8.06 Hz, 1H) 7.33 (dd, J = 8.31, 2.01 Hz, 1 H) 7.50 (d,
J = 2.01 Hz, 1 H).
4-Cyano-3-(2,4-dichloro-phenyl)-5-morpholin-4-yl-thio-
phene-2-carboxylic Acid [2-Methyl-prop-(E)-ylidene]amide
(14a). A brown slurry of 4-cyano-3-(2,4-dichloro-phenyl)-5-
morpholin-4-yl-thiophene-2-carboxylic acid amide (4a) (11.5 g,
0.1 mmol) in 80 mL of DMF-DMA was heated to reflux at
110 °C for 1.5 h to form a yellow slurry. During the reaction,
MeOH was generated. After cooling, the solvent was removed
under reduced pressure. The sticky residue was triturated with
MeOH and dried under vacuum to provide 12 g (91%) of 14a as
1
a white solid. LC-MS (APCI, Mþ þ 1) 437.2; H NMR
(400 MHz, DMSO-d6) 2.73 (s, 3 H) 3.07 (s, 3 H) 3.51-3.62
(m, 4 H) 3.73-3.81 (m, 4 H) 7.33 (d, J = 8.31 Hz, 1 H) 7.46 (dd,
J = 8.18, 2.14 Hz, 1 H) 7.68 (d, J = 2.01 Hz, 1 H) 8.32 (s, 1 H).
4-(2,4-Dichlorophenyl)-2-morpholin-4-yl-5-(2H-[1,2,4]triazol-
3-yl)thiophene-3-carbonitrile (1). To a slurry of 4-cyano-
3-(2,4-dichlorophenyl)-5-morpholin-4-yl-thiophene-2-carboxylic
acid [2-methyl-prop-(E)-ylidene]-amide (14a) (25 g, 57 mmol,
1.0 equiv) in glacial HOAc (40 mL) and EtOH (200 mL) was
added dropwise NH2NH2 H2O (18.9 mL, 572 mmol, 10 equiv)
3
whilst maintaining the inner reaction mixture temperature at -
10-0 °C. Upon the completion of addition, the mixture was
warmed to 60 °C and allowed to stir at this temperature for 1.5 h
to reach completion. The resulting yellow slurry was cooled to
25 °C, diluted with water (1 L), and extracted with EtOAc (1 L).
The separated organic layer was treated with 1.0 M NaOH (1 L)
and the desired triazole product was taken into aqueous solution.
The aqueous solution was separated and further washed with
EtOAc (2 ꢀ 200 mL) to remove all other organic impurities. The
561
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
aqueous layer was then acidified by HOAc to pH = 4-5 to
generate a slurry. The slurry was extracted with EtOAc (3 ꢀ
400 mL). The organic layers were combined, dried over Na2SO4,
filtered, and concentrated to afford 18 g (78%) of 1 as a pale
yellow solid. After being dried under vacuum overnight, 1H NMR
showed that the product was an acetate salt (0.8 equiv by NMR
2 h. To the reaction mixture was added 1.5 L water, and lots of
solids were precipitated out. The solids were collected by filtration,
triturated with a mixture of EOAc (50 mL) and saturated aq
NaHCO3 (50 mL), filtered, washed with water and EtOAc, and
dried to afford 12.5 g (84%) of 4b as a white solid. LC-MS (APCI,
Mþ þ 1) 378.0; 1H NMR (400 MHz, DMSO-d6) 3.47-3.59 (m,
4 H) 3.72-3.83 (m, 4 H) 3.85 (s, 3 H) 7.06 (dd, J = 8.56, 2.52 Hz,
1 H) 7.22 (d, J = 2.52 Hz, 1H) 7.36 (d, J = 8.56 Hz, 1 H).
3-(2-Chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-
yl-thiophene-2-carboxylic Acid [2-methyl-prop-(E)-ylide-
ne]amide (14b). A brown slurry of 3-(2-chloro-4-methoxy-
phenyl)-4-cyano-5-morpholin-4-yl-thiophene-2-carboxylic acid
amide (4b) (12.5 g, 33.1 mmol) in 80 mL of DMF-DMA was
heated to reflux at 110 °C for 1.5 h to form a yellow slurry.
MeOH was generated as the reaction proceeded. After cooling,
the solvent was removed under reduced pressure. The sticky
residue was triturated with MeOH and dried under vacuum to
afford 14.0 g (95%) of 14b as a white solid. LC-MS (APCI, Mþ
þ 1) 433.0; 1H NMR (400 MHz, DMSO-d6) 2.75 (s, 3 H) 3.07
(s, 3 H) 3.51-3.60 (m, 4 H) 3.74-3.80 (m, 4 H) 3.81 (s, 3 H)
6.94 (dd, J = 8.56, 2.52 Hz,1 H) 7.08 (d, J = 2.52 Hz, 1 H) 7.19 (d,
J = 8.31 Hz, 1 H) 8.31 (s, 1 H).
1
integration). LC-MS (APCI, Mþ þ 1) 406.0; H NMR
(400 MHz, DMSO-d6) of its HOAc salt: 1.91 (s, 3H from
acetate salt), 3.49-3.58 (m, 4 H) 3.75-3.84 (m, 4 H) 7.37-7.44
(m, 1 H) 7.46-7.51 (m, 1 H) 7.72 (d, J = 2.01 Hz, 1 H) 8.42 (s,
1 H), triazole N-H protons were missing due to deuterium
exchange; 13C NMR (100 MHz, DMSO-d6) 50.2, 65.2, 88.8,
115.9, 127.2, 128.8, 132.6, 133.0, 133.8, 134.2, 134.5, 144.0,
155.5, 165.1, one triazole carbon missing due to overlap; IR
(neat, cm-1) 3054, 2900, 2207, 1597, 1508; HRMS Calcd for
C17H13Cl2N5OS: 405.0219. Found: 405.0218.
3-(2-Chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-
yl-thiophene-2-carboxylic Acid Ethyl Ester (5b). A mixture
of 4-cyano-3-iodo-5-morpholin-4-yl-thiophene-2-carboxylic acid
ethyl ester (6) (18 g, 46 mmol, 1.0 equiv), 2-chloro-4-methoxy
phenylboronic acid (10.3 g, 55.1 mmol, 1.2 equiv), Pd(PtBu3)2
(1.17 g, 2.3 mmol, 0.05 equiv), CsF (21.1 g, 138 mmol, 3.0 equiv),
dioxane (200 mL), toluene (12 mL) and water (50 mL) was
flushed with N2 and stirred at 85 °C. After being stirred at 85 °C
for 12 h, the reaction was judged complete by LC-MS. To the
reaction mixture was added Si-thiol (15 g), and the mixture was
stirred at 25 °C for 2 h to remove the Pd catalyst. The resulting
mixture was then diluted with EtOAc and filtered through Celite.
The filtrate was washed with brine, dried over Na2SO4, filtered,
and concentrated to afford 19 g (100% yield) 5b as a pale yellow
solid with over 85% purity, which was used in the subsequent
hydrolysis step without further purification. LC-MS (APCI, Mþ
4-(2-Chloro-4-methoxy-phenyl)-2-morpholin-4-yl-5-(2H-
[1,2,4]triazol-3-yl)-thiophene-3-carbonitrile (2). To a slurry
of 3-(2-chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-yl-
thiophene-2-carboxylic acid [2-methyl-prop-(E)-ylidene]-amide
(14b) (14 g, 32 mmol, 1.0 equiv) in HOAc (20 mL) and EtOH
(100 mL) was added dropwise NH2NH2 H2O (10.7 mL, 323
3
mmol, 10 equiv) whilst maintaining the reaction mixture tem-
perature at -10-0 °C with stirring. Upon the completion of
addition, the reaction was slowly warmed to 60 °C and allowed
to stir at this temperature for 1.5 h to reach completion. The
resulting yellow slurry was cooled to 25 °C, diluted with water
(500 mL), and extracted with EtOAc (500 mL). The separated
organic layer was treated with 1.0 M NaOH (500 mL) and the
desired triazole product was taken into aqueous solution. The
aqueous solution was separated and further washed with EtOAc
(2 ꢀ 200 mL) to remove all other organic impurities. The
aqueous layer was then acidified by HOAc to pH = 4-5 to
generate a slurry. The slurry was extracted with EtOAc (3 ꢀ
200 mL). The organic layers were combined, dried over Na2SO4,
filtered, and concentrated to afford 8.0 g (55%) of 2 as a white
1
þ 1) 407.0; H NMR (400 MHz, DMSO-d6) 0.99-1.05 (m,
3 H) 3.58-3.65 (m, 4 H) 3.75-3.81 (m, 4 H) 3.83 (s, 3 H)
3.98-4.08 (m, 2 H) 6.98 (dd, J = 8.56, 2.52 Hz, 1 H) 7.14 (d, J =
2.52 Hz, 1 H) 7.25 (d, J = 8.56 Hz, 1 H).
3-(2-Chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-
yl-thiophene-2-carboxylic Acid (13b). To a suspension of
3-(2-chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-yl-thio-
phene-2-carboxylic acid ethyl ester (5b) (19.0 g, 47 mmol,
1.0 equiv) in MeOH (200 mL) and THF (50 mL) was added a
solution of LiOH (5.59 g, 233 mmol, 5 equiv) in water (50 mL)
at 25 °C. The resulting clear-orange solution was stirred at 25 °C
for 12 h to reach completion. The volatiles were removed under
reduced pressure. The remaining aqueous solution was solidified
during concentration. This slurry was diluted with water and
acidified to pH ≈ 3 with 2 N HCl. The resulting solids were
filtered, washed with water and a minimal amount of MeOH, and
dried under vacuum at 50 °C for 7 h to afford 16 g (90%) of 13b
as a pale-yellow solid. LC-MS (APCI, Mþ þ 1) 379.0; 1H NMR
(400 MHz, DMSO-d6) 3.54-3.63 (m, 4 H) 3.74-3.81 (m, 4 H)
3.83 (s, 3 H) 6.97 (dd, J = 8.56, 2.52 Hz, 1 H) 7.12 (d, J = 2.52 Hz,
1H) 7.25 (d, J = 8.56 Hz, 1 H).
3-(2-Chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-
yl-thiophene-2-carboxylic Acid Amide (4b). To a solution of
3-(2-chloro-4-methoxy-phenyl)-4-cyano-5-morpholin-4-yl-thio-
phene-2-carboxylic acid (13b) (15.0 g, 40 mmol, 1.0 equiv)
in DMF (500 mL) was added CDMT (8.51 g, 47.5 mmol,
1.2 equiv) and NMM (9.2 mL, 83.4 mmol, 2.0 equiv). The
resulting mixture was stirred at 25 °C for 1 h to form a transparent
solution. Ammonia gas was bubbled through the mixture for
30 min, and the reaction was then allowed to stir at 25 °C for
1
solid. After being dried under vacuum overnight, the H NMR
showed that the product was an acetate salt (0.75 equiv). LC-
1
MS (APCI, Mþ þ 1) 402.0; H NMR (400 MHz, DMSO-d6)
1.91 (s, 3 H from HOAc salts) 3.49-3.56 (m, 4 H) 3.76-3.83
(m, 4 H) 3.84 (s, 3 H) 6.96 (dd, J = 8.69, 2.64 Hz, 1 H) 7.11 (d,
J = 2.52 Hz, 1H) 7.27 (d, J = 8.56 Hz, 1 H) 8.41 (s, 1 H), triazole
N-H protons were missing due to deuterium exchange; 13C
NMR (100 MHz, DMSO-d6) 50.2, 55.3, 65.2, 88.9, 113.0, 114.3,
116.1, 125.3, 132.3, 133.7, 135.7, 144.0, 155.5, 160.8, 164.9, one
triazole carbon missing due to overlap; IR (neat, cm-1) 3239,
2902, 2205, 1606, 1503; HRMS Calcd for C18H16ClN5O2S:
401.0708. Found: 401.0713.
Ethyl 4-cyano-3-(4-cyano-2-fluorophenyl)-5-morpholi-
nothiophene-2-carboxylate (5c). A slurry of ethyl 4-cyano-3-
iodo-5-morpholinothiophene-2-carboxylate (6) (250.0 g, 637 mmol)
and 4-cyano-2-fluorophenylboronic acid (126.2 g, 765 mmol,
1.2 equiv) in a mixture of toluene (188 mL) and 1,4-dioxane
(3 L) was charged a with premade solution of CsF (290.5 g,
1.91 mol, 3 equiv) in water (750 mL) in a jacketed reactor. The
resulting slurry was purged with N2 for 20 min and charged with
562
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
1
Pd(PtBu3)2 (5.0 g, 9.56 mmol, 1.5 mol %) in one portion. The
reaction mixture was heated to reflux for 2 h until TLC analysis
(3:1 heptane/EtOAc) indicated full conversion of iodide (0.3 Rf)
to desired product (0.2 Rf). Note: The jacket temperature was set to
357.0; H NMR (400 MHz, DMSO-d6) 3.35-3.43 (m, 4 H)
3.55-3.63 (m, 4 H) 7.29 (dd, J = 8.40, 2.49 Hz, 1 H) 7.52 (d, J =
2.47 Hz, 1H) 7.79 (d, J = 8.39 Hz, 1 H).
(Z)-4-Cyano-3-(4-cyano-2-fluorophenyl)-N-((dimethylamino)
methylene)-5-morpholinothiophene-2-carboxamide (14c).
A slurry of 4-cyano-3-(4-cyano-2-fluorophenyl)-5-morpholinothio-
phene-2-carboxamide (4c) (105.0 g, 295 mmol) in DMF-DMA
(1.2 L) was heated to reflux for 1.5 h in a jacketed reactor. The
reaction mixture was cooled to 40 °C, and the resulting solids were
collected by filtration, rinsed with MeOH (3 ꢀ 150 mL), and dried
under vacuum at 50 °C for 24 h to afford 125.0 g (100%) of 14c as a
o
65 C and was adjusted as necessary to maintain the internal pot
temperature between 55 and 65 oC during the remaining extractive
workup steps to maintain solubility. The reaction mixture was
charged with toluene (1.0 L), stirred for 10 min, and separated.
The lower aqueous layer was extracted with toluene (0.5 L) and
discarded. The combined organic layer was extracted with
aqueous NaCl (10 wt %, 2 ꢀ 1.5 L) followed by distillation
under vacuum at 55-70 °C under a Dean-Stark apparatus until
no more water was observed to collect. The resulting solution
was distilled down to a final volume of 1.5 L and was cooled to
18 °C at a rate of 0.5 °C/min. The resulting slurry was filtered,
rinsed with toluene (3 ꢀ 100 mL) followed by n-heptane (5 ꢀ
250 mL), and dried under vacuum at 50 °C for 12 h to afford
220 g (90%) of 5c as a granular white solid.24 LC-MS (Mþ þ 1)
386.0; 1H NMR (400 MHz, CDCl3) 1.05 (t, J = 7.22 Hz, 3 H)
3.55-3.65 (m, 4 H) 3.70-3.80 (m, 4 H) 4.05-4.13 (m, 2 H)
7.62 (dd, J = 8.22, 1.99 Hz, 1 H) 7.81 (d, J = 7.98 Hz, 1H) 8.05 (d,
J = 2.01 Hz, 1 H).
pale-tan solid. LC-MS (Mþ þ 1) 412.0; H NMR (400 MHz,
1
DMSO-d6) 2.75 (s, 3 H) 3.09 (s, 3 H) 3.55-3.61 (m, 4 H) 3.70-
3.79 (m, 4 H) 7.55 (dd, J = 8.51, 2.49 Hz,1 H) 7.71 (d, J = 2.49 Hz,
1 H) 7.91 (d, J = 8.20 Hz, 1 H) 8.25 (s, 1 H).
4-(4-Cyano-2-fluorophenyl)-2-morpholino-5-(1H-1,2,4-
triazol-5-yl)thiophene-3-carbonitrile (3). A slurry of (Z)-4-
cyano-3-(4-cyano-2-fluorophenyl)-N-((dimethylamino)methylene)-
5-morpholinothiophene-2-carboxamide (14c) (50.0 g, 122 mmol,
1.0 equiv) in EtOH (600 mL) and HOAc (120 mL) was charged
with NH2NH2 (65 wt % in water, 30.0 g, 0.6 mol, 5.0 equiv) via
addition funnel over 10 min with pot temperature <0 °C. The
resulting solution was then warmed to maintain the pot tem-
perature at 60 °C for 1 h, cooled to 18 °C, charged with water
(1.25 L) via addition funnel over 5 min, and stirred for 1 h. The
product solids were partitioned between isopropyl acetate (1 L)
and aqueous sodium hydroxide (3 N, 1 L), and the organic layer
was discarded. The product-containing aqueous layer was ex-
tracted with isopropyl acetate (2 ꢀ 200 mL), concentrated
slightly on the rotary evaporator at 50 °C and 30 Torr to remove
any residual organic solvents, and was treated with aqueous 1 N
HCl to pH 5. The resulting solids were collected by vacuum
filtration, rinsed on the filter with water (5 ꢀ 200 mL) followed
by n-heptane (5 ꢀ 200 mL), and then dried under vacuum at
50 °C for 12 h to afford 40.0 g (86%) of 3 as a pale-yellow
granular solid. LC-MS (Mþ þ 1) 381.0; 1H NMR (400 MHz,
DMSO-d6) 3.54-3.59 (m, 4 H) 3.71-3.80 (m, 4 H) 7.60-7.65
(m, 1 H) 7.75-7.79 (m, 1 H) 7.92-7.99 m, 1 H) 8.43 (s, 1 H),
triazole N-H protons missing (deuterium exchange); 13C NMR
(100 MHz, DMSO-d6) 50.9, 64.8, 89.2, 113.1, 116.5, 118.0,
119.9, 120.1, 128.3, 128.8, 130.5, 133.9, 158.1, 160.9, 166.3, one
triazole carbon missing due to overlap; one triazole carbon
missing due to overlap; IR (neat, cm-1) 3230, 2910, 2230,
2208, 1618, 1499; HRMS Calcd for C18H13FN6OS: 380.0856.
Found: 380.0861.
4-Cyano-3-(4-cyano-2-fluorophenyl)-5-morpholinothio-
phene-2-carboxylic Acid (13c). A slurry of ethyl 4-cyano-3-(4-
cyano-2-fluorophenyl)-5-morpholinothiophene-2-carboxylate
(5c) (80.0 g, 208 mmol) in EtOH (800 mL) and 2-MeTHF
(320 mL) was charged with NaOH (50 wt %, 34.9 g, 436 mmol,
2.1 equiv) in one portion in a jacketed reactor. After being stirred
at 55 °C for 0.5 h, TLC analysis (1:1 heptane/EtOAc) indicated
full conversion of starting ester (0.4 Rf) to carboxylate (origin).
The reaction mixture was charged with EtOAc (1.5 L) and
transferred to a carboy. The reactor was charged with aqueous
1 N HCl (1.3 L) and saturated NaCl solution (500 mL), and the
mixture was adjusted to 50 °C. The original reaction mixture was
transferred back to the reactor over 10 min, stirred for 15 min,
and the layers were separated. The organic layer was extracted
with aqueous NaCl (10 wt %, 2 ꢀ 500 mL) and concentrated by
rotary evaporation to ∼250 mL total volume. The slurry was
charged with n-heptane (150 mL) and stirred for 1 h at 18 °C.
The resulting solids were collected by filtration, washed with n-
heptane (2 ꢀ 100 mL), and dried on the filter under vacuum for
12 h to afford 62.4 g (84%) of 13c as a pale-yellow solid. LC-MS
1
(Mþ þ 1) 358.0; H NMR (400 MHz, DMSO-d6) 3.55-3.64
(m, 4 H) 3.70-3.80 (m, 4 H) 7.63 (dd, J = 8.21, 2.49 Hz, 1 H)
7.80 (d, J = 2.42 Hz, 1H) 8.01 (d, J = 8.21 Hz, 1 H).
4-Cyano-3-(4-cyano-2-fluorophenyl)-5-morpholinothio-
phene-2-carboxamide (4c). A slurry of 4-cyano-3-(4-cyano-2-
fluorophenyl)-5-morpholinothiophene-2-carboxylic acid (13c)
(62.0 g, 173 mmol) in DMF (620 mL) was charged with CDMT
(33.5 g, 191 mmol, 1.1 equiv) in one portion at 18 °C. NMM
(35.1 g, 347 mmol, 2 equiv) was added via addition funnel over 5
min. The resulting slurry was stirred between 18 and 25 °C for
1 h. Ammonia gas was then bubbled through the suspension for
30 min, allowing the pot temperature to rise to 35 °C. Once the
exotherm subsided, the ammonia source was removed, and the
mixture was allowed to cool to 18 °C with stirring over 1 h. The
reactor was charged with water (1 L) via addition funnel over 10
min. The resulting solids were collected by filtration, rinsed with
water (3 ꢀ 150 mL), and dried on the filter for 30 min under
vacuum. The resulting wet cake was reslurried in MeOH
(100 mL), filtered, and dried under vacuum at 50 °C for 24 h
to afford 51.5 g (83%) of 4c as a white solid. LC-MS (Mþ þ 1)
’ AUTHOR INFORMATION
Corresponding Author
E-mail: Qinhua.Huang@pfizer.com. Telephone: 858-622-3165.
’ ACKNOWLEDGMENT
We thank Dr. Kevin Bunker, Michael Collins, Hayden Thomas,
and Jared VanHaitsma for helpful discussions related to this work,
and Jeffrey Raggon for outsourcing support. We are grateful to
Xiukun Qin and Heping Wu from Chemizon Ltd. for preparing
iodothiophene 6 on scale.
’ REFERENCES
(1) Nuss, J. M.; Tsuhako, A. L.; Anand, N. K. Annu. Rep. Med. Chem.
2009, 44, 339–356.
563
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564

Organic Process Research & Development
ARTICLE
(2) For examples of recently reported potent compounds containing
2-aminothiophene cores, see: (a) Renou, C. C. WO/2009/094224 A1,
2009. (b) Lethu, S.; Ginisty, M.; Bosc, D.; Dubois, J. J. Med. Chem. 2009,
52, 6205–6208. (c) Castano Mansanet, A. M.; Dominguez-Manzanares,
E.; Escribano, A. M.; Fernandez, M. C.; Hornback, W. J.; Jimenez-
Aguado, A. M.; Tromiczak, E. G.; Wu, Z.; Zarrinmayeh, H.; Zimmer-
man, D. M. WO/2005/070916 A1, 2005. (d) Cardin, D. P.; Gaulin, J. L.;
Greenspan, P. D.; Vyskocil, S.; Xu, T.; Renou, C. C. WO/2009/154741
A1, 2009.
(3) Starting material 2,4-dichloroacetophenone was available from
Aldrich at $31/25 g, and malononitrile was purchased from Aldrich at
$68/500 g.
(4) (a) Gust, R.; Burgemeister, T.; Mannschreck, A.; Sch€onenberger,
H. J. Med. Chem. 1990, 33, 2535–2544. (b) Lautens, M.; Schmid, G. A.;
Chau, A. J. Org. Chem. 2002, 67, 8043–8053.
conditions were slightly modified, in which 1.5 mol % of Pd(PtBu3)2 and
3.0 equiv of CsF were used (also see Experimental Section for detail).
(16) The direct amination of esters 5a and 5b to amides 4a and 4b
with NH3 in various solvents has been briefly investigated. The reactions
proceeded very slowly at elevated temperature using pressure vessels.
However, longer reaction time and higher reaction temperatures led to
decomposition of both the starting materials and products, while the
reaction was difficult to achieve good conversion.
(17) We observed a ∼ 25 °C exotherm on 5-g scale.
(18) While we were confident running this chemistry on 50 g scale
(based on crude calorimetry data) with close monitoring of reaction
internal temperature, it should be noted that additional process safety
analysis would be required before attempting to scale NH2NH2
chemistry further. It is likely that the number of equivalents of hydrazine
could be further reduced with additional studies.
(5) Sabnis, R. W.; Rangnekar, D. W.; Sonawane, N. D. J. Heterocycl.
Chem. 1999, 36, 333–345.
(19) We did not investigate quenching excess NH2NH2 on this
scale; unreacted NH2NH2 was purged in the aqueous sidestream.
(20) The acetate salt was avoided by using HCl with a direct-drop
product isolation.
(21) The largest scales that this reaction was run are 25 g, 14 g, and
50 g for compounds 1, 2 and 3, respectively. No further scale-up was
carried out because our PI3K project was held. The materials obtained
from this method are pure enough for animal studies.
(22) As with any transformation, attempts to scale this chemistry
further should be met with additional calorimetric scrutiny. DSC analysis
of steps 7-10 indicated that all transformations except step 9 were
carried out at more than 100 °C under any significant exothermic onset.
DSC onsets for reaction mixtures of interest: step 7 saponification,
230 °C (120 J/g); step 8 amide formation, 244 °C (524 J/g); step 9
condensation, 179 °C [122 J/g, DMA-DMF has an onset of 255 °C
(255 J/g)]; step 10 triazole formation, 169 °C [697 J/g, hydrazine
hydrate has an onset of 180 °C (2334 J/g)].
(6) (a) For
a review on the synthesis of tetrasubstituted
thiophenes, see: Sabnis, R. W.; Rangnekar, D. W.; Sonawane, N. D.
J. Heterocycl. Chem. 1999, 36, 333–345. (b) Sommen, G.; Comel, A.;
Kirsh, G. Synthesis 2003, 735–741. (c) Gewald, K.; Schinke, E.; B€ottcher,
H. Chem. Ber 1966, 99, 94–100.
(7) (a) For a review, see: Merkushev, E. B. Synthesis 1988, 923–937.
(b) While the synthesis of ester 10 was previously reported in ref 2c, our
chemistry was carried out in a relatively large scale, and procedures
described herein were modified to enable scale-up.
(8) Although the use of excess amounts of mCPBA in CH2Cl2
resulted in complete oxidation from sulfide 9 to sulfone 11, it was not
chosen for scale-up synthesis because the protocol using Oxone
provided a simpler workup and the mCPBA/CH2Cl2 procedure has
potential risk if oxygen is liberated in the flammable solvent during
reaction.
(9) For the replacement of NaH with other bases, such as NaOEt
and KOH, see: (a) Yu, S.-Y.; Cai, Y.-X. Synth. Commun. 2003,
33, 3989–3995. (b) Ryan, G.; Mettler, H. P.; Previdoli, F. EP508353
A1, 1992; (c) Zhao, W.; Wang, S.; Wang, W.; Li, Z. Xuaxue Shiji 2000,
22, 376.
(10) DSC and ARC analysis of reaction mixtures and raw materials,
as applicable, indicated that in all but one case the desired reactions
could be carried out in a dose-controlled manner at a temperature more
than 100 °C under any significant exothermic onset. Exothermic
additions were found to be dose-controlled; however, it should be noted
that further extensive calorimetric studies should be required before this
process could be considered for further scale-up in a plant setting. DSC
onsets for reaction mixtures of interest: step 1 alkylation, 224 °C (190 J/g);
step 2 condensation, 205 °C (142 J/g), step 3 Sandmeyer, 119 °C
(342 J/g); step 4 oxidation, 184 °C (213 J/g); step 5 morpholine
addition, 221 °C (148 J/g).
(23) While several intermediates leading to compounds 1 and 2 were
concentrated to dryness (see experimental description), workup and
purification of intermediates 5c, 13c, 4c, and 14c were demonstrated in
such a manner that we believe a successful plant campaign could be
realized with minor additional process development work, such as
solvent exchanges in place of concentrations to dryness.
(24) Determined by atomic absorption, the synthesized Suzuki
product 5c contained <50 ppm palladium.
(11) (a) Billingsley, K.; Buchwald, S. L. J. Am. Chem. Soc. 2007,
129, 3358–3368. (b) Martin, R.; Buchwald, S. L. Acc. Chem. Res. 2008,
41, 1461–1473. (c) Chemler, S. R.; Trauner, D.; Danishefsky, S. J.
Angew. Chem., Int. Ed. 2001, 40, 4544–4568.
(12) Higginson, P. D.; Sach, N. W. Org. Process Res. Dev. 2004,
8, 1009–1014.
(13) As the Pd catalysts (0.1 M in toluene) and bases (1.0 M aqueous
solution) were prepared and utilized as solutions, identical amounts of
water and toluene were incorporated for all the screening conditions.
The screening reactions were carried out under highly dilute conditions;
thus, the conditions were slightly modified on the concentration and
ratio of solvents for 1 g condition validation and scale-up synthesis.
(14) For a reference describing the efficiency of use Si-thiol resin to
remove palladium, see: Flahive, E. J.; Ewanicki, B. L.; Sach, N. W.;
O’Neill-Slawecki, S. A.; Stankovic, N. S.; Yu, S.; Guinness, S. M.; Dunn, J.
Org. Process Res. Dev. 2008, 12, 637–645.
(15) In the case of 5c, Suzuki coupling of compound 6 with 4-cyano-
2-fluorophenylboronic acid was carried out multiple times on 20- and
250-g scales, respectively. On 20-g scale, the optimal conditions
indicated in the Scheme 5 were used, and on 250-g scale, the coupling
564
dx.doi.org/10.1021/op100286g |Org. Process Res. Dev. 2011, 15, 556–564