Continuous Process for Preparing the Difluoromethylating Reagent [(DMPU)2Zn(CF2H)2] and Improved Synthesis of the ICHF2Precursor

Article abstract of DOI:10.1021/acs.oprd.0c00089

Complex [(DMPU)2Zn(CF2H)2] 1 is a crystalline organometallic reagent with utility in various catalytic difluoromethylation reactions under mild conditions. Unfortunately, this reagent is not commercially available and the procedure for the preparation of this air- and moisture-sensitive zinc complex has only been reported on a small scale. Herein, we report the development of a continuous process for the preparation of reagent 1 on >100 g scale by using a continuous stirred-tank reactor. The key to success is the design of a continuous reactive crystallization process in which the precipitation of the zinc complex 1 is driven by its low solubility in the reaction solvent. The improved synthesis of the iododifluoromethane precursor 2 from readily available bromodifluoroacetic acid further enables the synthesis of complex 1 on a larger scale.

Full text of DOI:10.1021/acs.oprd.0c00089

pubs.acs.org/OPRD
Article
Continuous Process for Preparing the Difluoromethylating Reagent
[(DMPU)2Zn(CF2H)2] and Improved Synthesis of the ICHF2 Precursor
Sebastien Monfette,* Yuan-Qing Fang,* Matthew M. Bio, Adam R. Brown, Ian T. Crouch,
Jean-Nicolas Desrosiers, Shengquan Duan, Joel M. Hawkins, Cheryl M. Hayward, Nikita Peperni,
and Joseph P. Rainville
Cite This: https://dx.doi.org/10.1021/acs.oprd.0c00089
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Complex [(DMPU)2Zn(CF2H)2] 1 is a crystalline organometallic reagent with utility in various catalytic
difluoromethylation reactions under mild conditions. Unfortunately, this reagent is not commercially available and the procedure for
the preparation of this air- and moisture-sensitive zinc complex has only been reported on a small scale. Herein, we report the
development of a continuous process for the preparation of reagent 1 on >100 g scale by using a continuous stirred-tank reactor. The
key to success is the design of a continuous reactive crystallization process in which the precipitation of the zinc complex 1 is driven
by its low solubility in the reaction solvent. The improved synthesis of the iododifluoromethane precursor 2 from readily available
bromodifluoroacetic acid further enables the synthesis of complex 1 on a larger scale.
KEYWORDS: difluoromethylation, catalysis, continuous flow, cross-coupling, zinc
INTRODUCTION
■
Fluorinated compounds are prominent in pharmaceutical
development due to the favorable pharmacological properties
they may confer compared to nonfluorinated analogs.1 Among
them, the difluoromethyl group has become popular2 for its
ability to serve as an isostere to the hydroxyl and thiol groups
while also affecting hydrophobicity and metabolic stability.3
The tactical importance of the difluoromethyl group in
medicinal chemistry has necessitated the development of
methods for its installation through traditional cross-coupling4
or via radical-based coupling methods.5 The entry of
difluoromethylated compounds into clinical trials requires the
development of large-scale, robust, versatile, and selective
difluoromethylation reactions or reagents, which are ideally
prepared from inexpensive feedstocks under mild conditions.
Figure 1. Application of zinc complex 1 in difluoromethylation
Among the reagents capable of mediating the difluoromethy-
lation of aryl electrophiles,6 the crystalline zinc complex
reactions using Pd, Ni, and Cu.
[(DMPU)2Zn(CF2H)2] (1) stands out for its reactivity under
mild conditions and for its activation by a variety of catalysts
nitrogen atmosphere implying thermal instability, vide infra.
including palladium, nickel, and copper complexes (Figure 1).7
Although reagent 1 is stable at a temperature of −35°C,9 we
During the development of a clinical-stage active pharma-
ceutical ingredient (API), we identified complex 1 as an
thought it would be prudent to generate and use it on demand,
rather than accumulate it in a large quantity for scale-up.
Herein, we report the development of a continuous flow
extremely effective reagent for the selective and high-yielding
difluoromethylation of an advanced intermediate.8 Given that
process allowing for the synthesis of complex 1 in >100 g scale
the synthesis of complex 1 had only been reported on a small
over 8 h.
lab scale, we set out to develop a process that would supply the
amount of reagent needed for the development of our API
Received: March 5, 2020
synthesis by using this difluoromethylating reagent. Zinc
complex 1 is known to be air and moisture sensitive7d and it
liberates the highly flammable difluoromethane gas (CH2F2)
when not properly inerted. During the process of development,
we also found that reagent 1 undergoes decomposition in the
solid state at room temperature, even when stored under a
https://dx.doi.org/10.1021/acs.oprd.0c00089
© XXXX American Chemical Society
Org. Process Res. Dev. XXXX, XXX, XXX−XXX
A

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 1. Reported Synthesis of Zinc Complex 1
We also describe a novel synthesis of iododifluoromethane
2, a key reagent employed in the synthesis of complex 1
(Scheme 1), using cheap and readily available bromodifluoro-
acetic acid. As complex 1 is a rare example of a nucleophilic
reagent capable of the difluoromethylation of various electro-
philes under mild reaction conditions, the development of a
scalable route to this reagent is expected to increase its use in
organic synthesis. Moreover, iodide 2 has found extensive use
in synthesis with application in the cross-coupling with aryl
Grignards,10 Simmons−Smith fluorocyclopropanation,11 Pd-
catalyzed Suzuki−Miyaura12 and Negishi13 couplings, and
heteroatom difluoromethylation via electrophilic alkylation.14
We expect the improved synthesis of intermediate 2 to also
have an impact beyond the work described here.
Table 1. Small-Scale Experimentation for the Synthesis of
Iodide 2 Starting from Chloro- or Bromodifluoroacetic
a
Acid
Scale
CuI
%
%
b
Entry (mmol) M+I− equiv
Solvent
Conv. % 2 BrCF2H
c
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
K
K
K
K
K
K
K
K
1
1
0
0
0
0
0
0
0
0
DMF
DMF
DMF
n-PrCN
MEK
DMAc
n-BuOH
Sulfolane
Sulfolane
Sulfolane
40
100
100
54
0
5
0
3
15
1
0
20
0
28
13
9
c
c
c
c
c
c
d
16
24
16
30
0
55
65
55
40
100
30
RESULTS AND DISCUSSION
■
Improved Synthesis of Iododifluoromethane ICHF2
(2). The reported procedure for the synthesis of zinc reagent 1
is shown in Scheme 1.15 Although both diethyl zinc and
DMPU are readily available, commercial suppliers of iodide 2
are scarce and the cost of this reagent is high.16,17
100
100
100
1
660
Na
Na
10
a
1 mmol carboxylic acid, 2 mmol iodide salt, 1 mL of solvent, a sealed
To produce the quantities of iodide 2 needed for the large-
scale synthesis of zinc complex 1, we extensively explored
literature preparation methods.18 Reported methods for the
synthesis of this simple molecule are multistep processes
generating a large amount of triphenylphosphine oxide waste19
or requiring the use of stoichiometric silver20 or mercury14b
salts and were not investigated due to their liabilities in the
context of large-scale synthesis. While iodide 2 was successfully
prepared from 2-fluorosulfonyldifluoroacetic acid
(FSO2CF2CO2H), this route was not considered cost effective
due to the cost of the acid ($1 g−1) and the modest yield of the
reaction (91 g or 30% yield of 2 using 280 g of
FSO2CF2CO2H). An appealing synthesis of 2 reported by
Chen and co-workers proceeds by decarboxylation of
chlorodifluoroacetic acid using CuI and KI in DMF.21
Unfortunately, in our hands, this reaction only gave trace
amounts of the desired product after multiple attempts (Table
1, entry 1) and was abandoned. We reasoned that the more
reactive bromodifluoroacetic acid might be a more effective
starting material for this reaction. However, when identical
reaction conditions were employed with bromodifluoroacetic
acid as the starting material, the desired product was obtained
in only a 5% yield as determined by 19F-NMR spectroscopy
(entry 2). Omitting CuI increased the yield of 2 to 16% with a
nearly identical proportion of bromodifluoromethane also
present (entry 3). Solvents were tested with the aim of
obtaining greater selectivity for the desired product by
preferentially solvating iodide salts over bromide salts (entries
4−8). While propionitrile, methyl ethyl ketone, and N,N-
dimethyl acetamide all reduced the amount of bromodifluoro-
methane formed relative to 2, the yield of the desired iodide 2
remained lower than desired. Gratifyingly, the use of sulfolane
vial. After 14 h, yields and conversion were determined by 19F-NMR
b
spectroscopy. Equivalents relative to bromodifluoroacetic acid.
c
Multiple unidentified species observed by 19F-NMR spectroscopy.
d
Average of three trials.
as the solvent increased the yield of iodide 2 to 55% (entry 8).
The use of NaI in place of KI further increased the yield of 2 to
65% (entry 9) with only 13% of bromodifluoromethane being
formed. These conditions were carried out on a 115 g scale of
bromodifluoroacetic and gave a 55% yield of iodide 2 (entry
10). The desired iodide 2 was isolated as a hexane solution
after extraction and distillation. Bromodifluoromethane was
also detected in the distilled mixture in a ∼6:1 ratio of 2/
BrCHF2. The presence of bromodifluoromethane was accept-
able as it does not react in the subsequent step.
The improved procedure for the synthesis of iodide 2 is
shown in Scheme 2. The reaction was carried out in a three-
neck round-bottom flask with overhead stirring and an
attached dry ice cold trap to condense the volatile product.
Extraction using water and hexanes followed by washing with
sodium thiosulfate and drying with sodium sulfate provides a
0.68 M solution of 2 in hexanes with <60 ppm H2O as
determined by the Karl-Fisher analysis. We chose to prepare
and handle iodide 2 as a solution rather than a neat liquid
given its low boiling point of 20 °C. Hexane was chosen as a
solvent given the low solubility of zinc complex 1 in this
solvent, which helps drive the reaction in favor of the
bisdifluoromethyl zinc complex, vide infra. We note that
heptane can be substituted for hexanes.
Batch Synthesis of [(DMPU)2Zn(CF2H)2] (1). With an
improved procedure for the synthesis of iodide 2, we next
B
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 2. Improved Synthesis of 2 Starting from Bromodifluoroacetic Acid
sought to develop a scalable procedure for the synthesis of zinc
complex 1. Both Vicic7d and Mikami7a7c reported the reaction
of diethyl zinc with iodide 2 and DMPU using different solvent
systems.22 In the conditions reported by Vicic,7d selective
formation of the desired bisdifluoromethyl complex 1 is
obtained, likely owing to precipitation from the pentane/
hexane solvent mixture, which drives the reaction forward.
Consistent with this observation, the use of DMF as the
solvent in the Mikami procedure leads to a mixture of mono
and bisdifluoromethyl zinc complex. This mixture of 1 and
monodifluoromethyl [(DMPU)2Zn(CF2H)I] leads to a color-
less gelled liquid after concentration, which was expected to
cause problems in the isolation and was therefore deemed less
attractive. Additionally, mechanistic studies have revealed that
the bisdifluoromethyl zinc complex 1 undergoes faster
transmetallation than the monodifluoromethyl analog7a giving
additional impetus to isolate the clean, bisdifluoromethyl zinc
complex 1 to maximize reactivity. For these reasons, we
developed the synthesis of complex 1 in hydrocarbon solvent
as reported by Vicic.7d
We initially investigated the synthesis of 1 in small batches
to gather data that would aid in the design of a process. As
stated above, we chose to employ volatile iodide 2 as a solution
in hexanes for convenience as this can be handled at ambient
temperature and pressure while also promoting the precip-
itation of complex 1 as it is formed during the reaction.
Reaction progress was determined by monitoring the
disappearance of iododifluoromethane 2 as well as the
appearance of the co-product ethyl iodide by quantitative
high-performance liquid chromatography (HPLC) analysis.
Our initial experiment showed that the reaction reaches 60−
70% of the theoretical yield of ethyl iodide. We also noticed
the precipitation of an amorphous solid that was sticking to the
walls of the reactor and that could not be filtered. We
suspected that this gelling might be caused by adventitious
water in the reaction, most likely from DMPU. Indeed, the
Karl-Fisher analysis of the DMPU employed in the first
synthesis showed that it contained 2700 ppm of water (Table
2, entry 1). Decreasing the concentration of water from 2700
to 465 ppm (entry 2) by treatment with molecular sieves
improved the yield substantially (57%) but was still much
lower than desired. We hypothesized that this low yield could
be due to our initial batch of difluoroiodomethane 2 in hexanes
being a brown color, presumably by the presence of
contaminating iodine.23 Switching to a colorless batch of 2,
which was obtained by introduction of a sodium thiosulfate
wash during workup, resulted in another significant improve-
ment in the yield to 85−90% and yielded white crystalline zinc
complex 1 (entry 3). Although the literature synthesis of 1 is
carried out at a temperature of −20 °C, internal monitoring of
the reaction showed no significant exotherm. Therefore, we
carried out the reaction at a constant temperature between 22
and 28 °C and did not observe any negative effects on the
reaction yield (entry 4).
Table 2. Batch Optimization for the Synthesis of Zinc
Complex 1
Reaction
temperature
DMPU water
content (ppm)
Isolated yield
of 1 (%)
Entry
Color of 2
1
2
3
4
−5 to 23 °C
−20 to 25 °C
−15 to 23 °C
22 to 28 °C
2700
465
447
175
brown
brown
colorless
colorless
NA
57
87
86
Continuous Flow Synthesis of Zinc Complex 1. As
stated above, stability studies on the zinc complex 1 that was
obtained from the batch process showed some challenges.9
These studies showed that although complex 1 is stable under
an inert atmosphere at a temperature of −35 °C for at least 75
days, as judged by 19F-NMR spectroscopy, decomposition to
difluoromethane along with unidentified zinc species occurs
over the course of 1 month at 0 °C and in only 2 days at room
temperature under inert atmosphere. The thermal stability of
complex 1 at room temperature was expected to cause issues in
shipment and storage of this reagent. We reasoned that we
could address both challenges by designing a continuous
process to generate reagent 1 on demand. Such a process
would involve a continuous stirred-tank reactor (CSTR) for
the formation of complex 1 that would be pumped onto a pair
of dissolve-off filter as was elegantly demonstrated by Cole and
co-workers at Eli Lilly.24 Such dissolve-off filter would consist
of a pair of agitated single plate filters that would alternatively
collect, wash, and dry product 1 coming out of the CSTR
under an inert atmosphere. Reagent 1 could then be directly
used in downstream processing via a contiguous continuous
process after dissolution from the filter.
A single 250 mL reactor was used for our CSTR experiment
with a theoretical yield of 17.8 g/h (Figure 2). The DMPU
ligand was delivered by a syringe pump using disposable
polyethylene syringes. Diethyl zinc was delivered as a 1 M
solution in hexanes using a 25 mL glass syringe using two
syringe pumps in tandem. A distilled 0.39 M solution of iodide
2 in hexanes was pumped through a Tacmina diaphragm pump
and Quantim mass flow meter controlled by a proportional-
integral-derivative (PID) controller. Using this system, a steady
inflow of the three streams was maintained throughout the
7.35 h run with refilling of the diethyl zinc syringes at 36 min
intervals. The CSTR average reaction volume of 210 mL gave
an average residence time of 47 min, which was adequate time
for product precipitation based on batch experiments.
Overhead stirring with a vertical impeller provided good
C
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 2. Diagram of the CSTR setup for the continuous synthesis of [(DMPU)2Zn(CF2H)2] 1.
dispersion of solids. Pumping the product slurry out of the
reactor had presented difficulties in smaller-scale CSTRs.
Peristaltic pumping using poly(tetrafluoroethylene) (PTFE)
tubing of 3/16″ ID proved unable to draw the dense solid out
of the reactor adequately, instead drawing out mostly the
supernatant. Switching to tubing of 1/8″ ID drew the slurry at
a higher velocity, allowing the solids to remain suspended in
the PTFE tube while avoiding clogging. Intermittent fast
pumping also helped the slurry transfer. The slurry was
collected on a 600 mL filter under an argon atmosphere, which
was periodically drained of solvent by applying vacuum. At the
end of the run, the dip tube was lowered and the remaining
suspension was transferred into the filter chamber. Three
washings of hexanes were added to the reactor and transferred
to the filter, which served to transfer the remaining solids and
wash the solids in the filter. Residual hexanes were removed
with an argon flow through the filter until a constant mass was
obtained. Shut valves allowed the sealed filter to be transferred
to a glovebag for collecting the product under an argon
atmosphere. Using this setup, we were able to obtain 120 g of
zinc complex 1 as a white crystalline solid. The potency of the
material measured at 91 wt % as determined by quantitative
19F-NMR spectroscopy in dry deuterobenzene using trifluor-
otoluene as an internal standard.
complex 1 is expected to promote the use of this reagent for
organic synthesis.
EXPERIMENTAL SECTION
■
Unless otherwise noted, reagents and solvents were used as
received from commercial suppliers. Proton nuclear magnetic
resonance spectra were obtained on a Bruker AV-300
spectrometer at 300 MHz using tetramethylsilane as an
internal standard for proton spectra. Unless otherwise noted,
yields are uncorrected for purity or potency.
Synthesis of Bromodifluoroacetic Acid. In a 5 L three-
neck round-bottom flask open to air and equipped with
overhead stirring, 98.4 g (2.46 mol) of NaOH pellets and 2.25
L of MeOH were combined and stirred until dissolved. The
solution was cooled in an ice bath and a thermocouple was
fitted. Ethyl bromodifluoroacetate (Oakwood Chemicals
#002480) (500 g, 2.46 mol) was added in 100 g portions,
keeping the temperature below 20 °C. After the addition was
complete, the ice bath was removed and the solution was
stirred at 150 rpm for 12 h. The homogeneous solution was
concentrated by rotary evaporation in a 1 L flask to produce a
white solid. The white solid was dissolved in 1 L H2O. To this
was added 380 g (2.46 mol) of 98% H2SO4/H2O (1:1 v/v)
over 10 min with magnetic stirring in an ice bath. After the
addition, a pH of ≤0 was measured using pH strips. This
solution was extracted with methyl ethyl ketone (4 × 250 mL)
in a separatory funnel. The combined organic phase was dried
over Na2SO4 and concentrated by rotary evaporation with a 50
°C bath. The oil was dried again with Na2SO4 and filtered
using methyl ethyl ketone to wash through before concentrat-
ing again to give 378.5 g of oil (88% of theoretical mass). H2O
(2.74 wt %) was detected by the Karl-Fisher apparatus. This
was taken on to the next step.
CONCLUSIONS
■
The foregoing describes the synthesis of the difluoromethyla-
tion reagent 1 as well as a novel synthesis of the
iododifluoromethane precursor 2. Decarboxylation of bromo-
difluoroacetic acid and trapping with sodium iodide in
sulfolane allows for a cheap and efficient synthesis of 2 in a
good yield. Synthesis of the zinc complex 1 was accomplished
in batch and continuous flow. The key to its synthesis was the
use of a nonpolar solvent allowing for the precipitation of
product 1 as it is formed thus driving the reaction forward. The
development of a scalable process for the synthesis of zinc
Synthesis of Iododifluoromethane 2. In a 2 L, three-
neck round-bottom flask with overhead stirring, a heating
mantle, nitrogen inlet and thermocouple, NaI (196 g, 1.31
mol) and sulfolane (680 g) were heated to 90 °C, producing a
D
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
brown solution. A cold trap with acetone/dry ice was attached
to the flask to prevent the loss of the product through
evaporation. Bromodifluoroacetic acid (115 g, 0.66 mol) was
combined with 50 g of sulfolane and this solution was added to
the stirred 90 °C reactor over 30 min. The mixture was stirred
under a nitrogen atmosphere at this temperature and
iododifluoromethane concentration was monitored by quanti-
tative HPLC with observation at 254 nm. Alternatively,
conversion can be monitored by 19F-NMR spectroscopy.
Upon completion (∼40 h), the contents of the flask and cold
trap were transferred into a 5 L tank with overhead stirring
after letting the reaction flask cool to room temperature.
Hexanes (200 mL) was added, followed by water until the
aqueous layer volume was 3.5 L. The combined hexane
extracts were dried with MgSO4 and filtered, giving 362 g of a
17.9 wt % solution of ICHF2, a 55% yield. This solution was
distilled at atmospheric pressure with a dry ice/acetone cold
trap attached to the receiving flask. The receiving flask and
cold trap contents were combined to give 338 g of a 16 wt %
solution in hexanes; a 46% yield of ICHF2. Bromodifluoro-
methane was also detected in the distilled mixture in a 6.3:1
ratio of ICHF2/BrCHF2 by 19F-NMR integration. The distilled
solution was stored at 4°C in the dark and was used as is.
Trifluorotoluene was used as a reference standard set to −63.3
ppm. 19F-NMR (CDCl3): ICHF2 δ −66.5 ppm (d, J = 56.3
Hz) BrCHF2 δ −68.6 ppm (d, J = 60.2 Hz).
transferring the residual solids. This was done three times to
wash the solids in the filter. A small amount remained on the
sides of the flask. A heavy stream of argon was passed through
the solids in the filter until a constant mass was obtained. By
pre-tared weighing, 121 g of a white power was collected in the
filter after drying. In total, 121 g of white powder was collected
(92% yield unadjusted for purity). Quantitative 19F-NMR assay
in C6D6 was 91.0 wt %.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.oprd.0c00089.
NMR spectra; temperature monitoring during batch
synthesis of complex 1; delivery system of diethyl zinc;
experimental setup for the synthesis of 1 in flow; reagent
flow rates; data collection for the continuous run
producing 120 g of 1; stability of zinc complex 1;
attempts to prepare complex 1 via a new route (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
Sebastien Monfette − Process Chemistry, Chemical R&D,
Pfizer Worldwide R&D, Groton, Connecticut 06340, United
States; orcid.org/0000-0002-8394-4423;
Continuous Flow Synthesis of Complex 1. The reactor
system was rinsed with dry tetrahydrofuran (4 × 100 mL) and
then with hexanes (4 × 100 mL). Residual solvent was dried
off with an argon flow sweep. Seeds of complex 1 (1.00 g, 2.36
mmol) were weighed out under an argon atmosphere and
added to the 300 mL jacketed reactor serving as CSTR with
the recirculating chiller set at 23 °C. A stock solution of iodide
2 was pumped out of a 2 L Schott bottle with the IDEX inlet
cap under an argon headspace and through a Tacmina
diaphragm pump (10 mL/min max) and Quantim mass flow
meter controlled by PID. This was set to 2.58 g/min and was
recirculated into the bottle until the flow rate stabilized. The
solution of diethyl zinc was drawn into a 25 mL glass Hamilton
syringe connected to a tee with two shut valves. This was
connected to the CSTR inlet with 1/8″ OD PTFE tubing,
allowing repeated drawing of the diethyl zinc solution in
hexanes without disconnecting needles. Two syringe pumps
were used in tandem to allow uninterrupted delivery of the
diethyl zinc solution, see Figure S6. The stock solution of
iodide 2, diethyl zinc in hexanes, and neat DMPU started
pumping into the CSTR concurrently. When the fill volume
reached 200 mL, the slurry-transferring peristaltic pump
adapted with the PTFE tubing head (1/4″ OD, 1/8″ ID
PTFE tubing on the CSTR side) started in intermittent
pumping cycles (pumping for 20 s at 600 rpm followed by 5
min off time), giving an average CSTR volume of 210 mL. The
supernatant of the CSTR was sampled periodically for reaction
conversion assay based on EtI and ICHF2 (2) HPLC assay.
When the volume of the solvent in the filter reached about 500
mL, a 1 L Schott bottle was connected to the bottom of the
filter with 1/8″ OD PTFE tubing with the shut valve. Vacuum
was generated in this bottle with a rotavap pump and the valve
was opened to pull the solvent through the filter and into the
bottle. The pumping was stopped after 442 min of run time.
The material remaining in the CSTR was transferred to the
filter by lowering the dip tube and swirling the flask. Hexanes
(200 mL) were added to the CSTR and this aided in
Email: sebastien.monfette@pfizer.com
Yuan-Qing Fang − Snapdragon Chemistry Inc., Waltham,
Massachusetts 02451, United States; orcid.org/0000-0002-
8952-1018; Email: eric.fang@snapdragonchemistry.com
Authors
Matthew M. Bio − Snapdragon Chemistry Inc., Waltham,
Massachusetts 02451, United States
Adam R. Brown − Process Chemistry, Chemical R&D, Pfizer
Worldwide R&D, Groton, Connecticut 06340, United States
Ian T. Crouch − Snapdragon Chemistry Inc., Waltham,
Massachusetts 02451, United States
Jean-Nicolas Desrosiers − Process Chemistry, Chemical R&D,
Pfizer Worldwide R&D, Groton, Connecticut 06340, United
States; orcid.org/0000-0003-3859-2390
Shengquan Duan − Process Chemistry, Chemical R&D, Pfizer
Worldwide R&D, Groton, Connecticut 06340, United States;
orcid.org/0000-0002-8126-2592
Joel M. Hawkins − Process Chemistry, Chemical R&D, Pfizer
Worldwide R&D, Groton, Connecticut 06340, United States
Cheryl M. Hayward − Process Chemistry, Chemical R&D,
Pfizer Worldwide R&D, Groton, Connecticut 06340, United
States
Nikita Peperni − Process Chemistry, Chemical R&D, Pfizer
Worldwide R&D, Groton, Connecticut 06340, United States
Joseph P. Rainville − Process Chemistry, Chemical R&D, Pfizer
Worldwide R&D, Groton, Connecticut 06340, United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.0c00089
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors would like to thank Steven Guinness for useful
discussions. The authors would also like to thank Prof. David
E
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
mides: General Methods To Construct Fluorinated 3,3-Disubstituted
2-Oxindoles Using Fluoroalkylsulfonyl Chlorides. Org. Lett. 2014, 16,
4594−4597. (d) Tang, X.-J.; Zhang, Z.; Dolbier, W. R., Jr. Direct
Photoredox-Catalyzed Reductive Difluoromethylation of Electron-
Deficient Alkenes. Chem. - Eur. J. 2015, 21, 18961−18965. (e) He, Z.;
Tan, P.; Ni, C.; Hu, J. Fluoroalkylative Aryl Migration of Conjugated
N-Arylsulfonylated Amides Using Easily Accessible Sodium Di- and
Monofluoroalkanesulfinates. Org. Lett. 2015, 17, 1838−1841. (f) Li,
Z.; Cui, Z.; Liu, Z.-Q. Copper- and Iron-Catalyzed Decarboxylative
Tri- and Difluoromethylation of α,β-Unsaturated Carboxylic Acids
with CF3SO2Na and (CF2HSO2)2Zn via a Radical Process. Org. Lett.
2013, 15, 406−409. (g) O’Brien, A. G.; Maruyama, A.; Inokuma, Y.;
Fujita, M.; Baran, P. S.; Blackmond, D. G. Radical C-H
functionalization of heteroarenes under electrochemical control.
Angew. Chem., Int. Ed. 2014, 53, 11868−11871. (h) Rong, J.; Deng,
L.; Tan, P.; Ni, C.; Gu, Y.; Hu, J. Radical Fluoroalkylation of
Isocyanides with Fluorinated Sulfones by Visible-Light Photoredox
Catalysis. Angew. Chem., Int. Ed. 2016, 55, 2743−2747. (i) Su, Y.-M.;
Hou, Y.; Yin, F.; Xu, Y.-M.; Li, Y.; Zheng, X.; Wang, X.-S. Visible
Light-Mediated C-H Difluoromethylation of Electron-Rich Hetero-
arenes. Org. Lett. 2014, 16, 2958−2961.
A. Vicic for the gracious gift of zinc reagent 1 that initiated this
unit of work.
REFERENCES
■
(1) (a) Harsanyi, A.; Sandford, G. Organofluorine chemistry:
applications, sources and sustainability. Green Chem. 2015, 17, 2081−
2086. (b) Neumann, C. N.; Ritter, T. Late-Stage Fluorination: Fancy
Novelty or Useful Tool? Angew. Chem., Int. Ed. 2015, 54, 3216−3221.
(c) Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell,
N. A. Applications of Fluorine in Medicinal Chemistry. J. Med. Chem.
2015, 58, 8315−8359. (d) Shah, P.; Westwell, A. D. The role of
fluorine in medicinal chemistry. J. Enzyme Inhib. Med. Chem. 2007, 22,
527−540. (e) Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V.
Fluorine in medicinal chemistry. Chem. Soc. Rev. 2008, 37, 320−330.
(f) Caron, S. Where Does the Fluorine Come From? A Review on the
Challenges Associated with the Synthesis of Organofluorine
Compounds. Org. Process Res. Dev. 2020, 24, 470−480.
(2) (a) Meanwell, N. A. Synopsis of Some Recent Tactical
Application of Bioisosteres in Drug Design. J. Med. Chem. 2011, 54,
2529−2591. (b) Hagmann, W. K. The Many Roles for Fluorine in
Medicinal Chemistry. J. Med. Chem. 2008, 51, 4359−4369. (c) Muller,
K.; Faeh, C.; Diederich, F. Fluorine in Pharmaceuticals: Looking
Beyond Intuition. Science 2007, 317, 1881−1886. (d) Desrosiers, J.-
N.; Kelly, C. B.; Fandrick, D. R.; Nummy, L.; Campbell, S. J.; Wei, X.;
Sarvestani, M.; Lee, H.; Sienkiewicz, A.; Sanyal, S.; Zeng, X.;
Grinberg, N.; Ma, S.; Song, J. J.; Senanayake, C. H. A Scalable and
Regioselective Synthesis of 2-Difluoromethyl Pyridines from
Commodity Chemicals. Org. Lett. 2014, 16, 1724−1727.
(6) (a) Rong, J.; Ni, C.; Hu, J. Metal-Catalyzed Direct
Difluoromethylation Reactions. Asian J. Org. Chem. 2017, 6, 139−
152. (b) Yerien, D. E.; Barata-Vallejo, S.; Postigo, A. Difluoromethy-
lation Reactions of Organic Compounds. Chem. - Eur. J. 2017, 23,
14676−14701.
(7) (a) Serizawa, H.; Ishii, K.; Aikawa, K.; Mikami, K. Copper-
Catalyzed Difluoromethylation of Aryl Iodides with (Difluoromethyl)
zinc Reagent. Org. Lett. 2016, 18, 3686−3689. (b) Aikawa, K.;
Serizawa, H.; Ishii, K.; Mikami, K. Palladium-Catalyzed Negishi
Cross-Coupling Reaction of Aryl Halides with (Difluoromethyl)zinc
Reagent. Org. Lett. 2016, 18, 3690−3693. (c) Aikawa, K.; Ishii, K.;
Endo, Y.; Mikami, K. Copper-catalyzed allylic difluoromethylation of
allyl carbonates with (difluoromethyl)zinc reagent. J. Fluorine Chem.
2017, 203, 122−129. (d) Xu, L.; Vicic, D. A. Direct Difluoromethy-
lation of Aryl Halides via Base Metal Catalysis at Room Temperature.
J. Am. Chem. Soc. 2016, 138, 2536−2539. (e) Pan, F.; Boursalian, G.
B.; Ritter, T. Palladium-Catalyzed Decarbonylative Difluoromethyla-
tion of Acid Chlorides at Room Temperature. Angew. Chem., Int. Ed.
2018, 57, 16871−16876.
(8) The development of the difluoromethylation route to the API
will be reported independently.
(9) See Supporting Information for details
(10) Motohashi, H.; Mikami, K. Nickel-Catalyzed Aromatic Cross-
Coupling Difluoromethylation of Grignard Reagents with Difluor-
oiodomethane. Org. Lett. 2018, 20, 5340−5343.
(11) (a) Beaulieu, L.-P. B.; Schneider, J. F.; Charette, A. B. Highly
Enantioselective Simmons-Smith Fluorocyclopropanation of Allylic
Alcohols via the Halogen Scrambling Strategy of Zinc Carbenoids. J.
Am. Chem. Soc. 2013, 135, 7819−7822. (b) Navuluri, C.; Charette, A.
B. Diastereoselective Fluorocyclopropanation of Chiral Allylic
Alcohols Using an α-Fluoroiodomethylzinc Carbenoid. Org. Lett.
2015, 17, 4288−4291. (c) Jubault, P.; Poisson, T.; Pannecoucke, X.
Fluorocyclopropanes from Difluoroiodomethane and Alkenes. Sci.
Synth., Knowl. Updates 2017, 2, No. 338.
(3) Erickson, J. A.; McLoughlin, J. I. Hydrogen Bond Donor
Properties of the Difluoromethyl Group. J. Org. Chem. 1995, 60,
1626−1631.
(4) (a) Prakash, G. K. S.; Ganesh, S. K.; Jones, J.-P.; Kulkarni, A.;
Masood, K.; Swabeck, J. K.; Olah, G. A. Copper-mediated
difluoromethylation of (hetero)aryl iodides and β-styryl Halides
with tributyl(difluoromethyl)stannane. Angew. Chem., Int. Ed. 2012,
51, 12090−12094. (b) Fujikawa, K.; Kobayashi, A.; Amii, H. An
efficient route to difluoromethylated pyridines. Synthesis 2012, 44,
3015−3018. (c) Ashwood, M. S.; Cottrell, I. F.; Cowden, C. J.;
Wallace, D. J.; Davies, A. J.; Lennedy, D. J.; Dolling, E. H. Copper-
mediated reaction of 2-halopyridines with ethyl Bromodifluoroace-
tate. Tetrahedron Lett. 2002, 43, 9271. (d) Zeng, X.; Wei, X.; Song, J.
J.; Sarvestani, M.; Fandrick, D. R.; Qu, B.; Rodriguez, S.; Sieber, J. D.;
Desrosiers, J.-N.; Yee, N. K.; Roschangar, F.; Senanayake, C. H.
Development of a Practical Sequence for Difluoromethylation of 2-
Bromopyridines via Copper-Mediated Reductive Coupling and
Decarboxylation. Asian J. Org. Chem. 2015, 4, 1262−1264.
(e) Fujikawa, K.; Fujioka, Y.; Kobayashi, A.; Amii, H. A new method
for aromatic difluoromethylation: copper-catalyzed cross-coupling
and decarboxylation sequence from aryl iodides. Org. Lett. 2011, 13,
5560−5563. (f) Fier, P. S.; Hartwig, J. F. Copper-Mediated
Difluoromethylation of Aryl and Vinyl Iodides. J. Am. Chem. Soc.
2012, 134, 5524−5527. (g) Jiang, X.-L.; Chen, Z.-H.; Xu, X.-H.;
Qing, F.-L. Copper-mediated difluoromethylation of electron-poor
aryl iodides at room temperature. Org. Chem. Front. 2014, 1, 774−
776. (h) Ge, S.; Chaladaj, W.; Hartwig, J. F. Pd-Catalyzed α-Arylation
of α,α-Difluoroketones with Aryl Bromides and Chlorides. A Route to
Difluoromethylarenes. J. Am. Chem. Soc. 2014, 136, 4149−4152.
(i) Chang, D.; Gu, Y.; Shen, Q. Pd-Catalyzed Difluoromethylation of
Vinyl Bromides, Triflates, Tosylates, and Nonaflates. Chem. Eur. J.
2015, 21, 6074−6078.
(12) Hori, K.; Motohashi, H.; Saito, D.; Mikami, K. Precatalyst
Effects on Pd-Catalyzed Cross-Coupling Difluoromethylation of Aryl
Boronic Acids. ACS Catal. 2019, 9, 417−421.
(13) Nitta, J.; Motohashi, H.; Aikawa, K.; Mikami, K. Palladium-
Catalyzed Negishi Cross-Coupling Reaction of Difluoroiodomethane
with Arylzinc Reagents. Asian J. Org. Chem. 2019, 8, 698−701.
(14) (a) Hine, J.; Ketley, A. D. Methylene derivatives as
intermediates in polar reactions. XX. Reaction of aqueous and
alcoholic base with chlorodifluoromethane and difluoroiodomethane.
J. Org. Chem. 1960, 25, 606−607. (b) Burg, A. B. New chemistry of
(difluoromethyl)phosphines. Inorg. Chem. 1985, 24, 3342−3347.
(15) An alternative synthesis approach consists of treating Zn(CF3)2
with NaBH4, see: Xu, Long. “Advances in Metal-Catalyzed
(5) (a) Fujiwara, Y.; Dixon, J. A.; Rodriguez, R. A.; Baxter, R. D.;
Dixon, D. D.; Collins, M. R.; Blackmond, D. G.; Baran, P. S.; New, A.
A New Reagent for Direct Difluoromethylation. J. Am. Chem. Soc.
2012, 134, 1494−1497. (b) Fujiwara, Y.; Dixon, J. A.; O’Hara, F.;
Funder, E. D.; Dixon, D. D.; Rodriguez, R. A.; Baxter, R. D.; Herle, B.;
Sach, N.; Collins, M. R.; Ishihara, Y.; Baran, P. S. Practical and innate
carbon-hydrogen functionalization of heterocycles. Nature 2012, 492,
95−99. (c) Tang, X.-J.; Thomoson, C. S.; Dolbier, W. R., Jr.
Photoredox-Catalyzed Tandem Radical Cyclization of N-Arylacryla-
F
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Difluoromethylation and Polydifluoromethylenation Reactions”.
Ph.D. Theses and Dissertations; Lehigh University: 2017.
(16) (a) Hu, J.; Ni, C. Using Difluoro(iodo)methane. Sci. Synth., C-1
́
́
Build. Blocks Org. Synth. 2013, 2, No. 431. (b) Levesque, E.
Encyclopedia of Reagents for Organic Synthesis Copyright 2; John Wiley
& Sons, Ltd., 2014. (c) Sperry, J. B.; Farr, R. M.; Levent, M.; Ghosh,
M.; Hoagland, S. M.; Varsolona, R. J.; Sutherland, K. A Robust
Process for an mGluR5 Negative Allosteric Modulator: Difluorome-
thylation and Sonogashira Coupling on Large Scale. Org. Process Res.
Dev. 2012, 16, 1854−1860.
(17) Potential alternatives to iododifluoromethane 2 were explored
for this reaction but only in the case of 2 did we observe reactivity.
See Supporting Information for details.
(18) (a) Weyerstahl, P.; Mathias, R.; Blume, G. Darstellung von 1-
fluor-1-jod-cyclopropanen. Tetrahedron Lett. 1973, 14, 611−612.
(b) Hine, J.; Butterworth, R.; Langford, P. B. Methylene derivatives as
intermediates in polar reactions. X. The hydrolysis and deuterium
exchange of dibromofluoromethane and fluorodiiodomethane. J. Am.
Chem. Soc. 1958, 80, 819−824. (c) Sharp, K. G. Silicon-substituted
derivatives of trifluoro(trifluoromethyl)silane. Chemistry of the
silicon-iodine bond in a polyhalo system. Inorg. Chem. 1975, 14,
1241−1244.
(19) Flynn, R. M.; Manning, R. G.; Kessler, R. M.; Burton, D. J.;
Hansen, S. W. The hydrolysis of bromodifluoromethyltriphenylphos-
phonium bromide. J. Fluorine Chem. 1981, 18, 525−531.
(20) Haszeldine, R. N. The reactions of metallic salts of acids with
halogens. III. Some reactions of salts of fluorohaloacetates and of
perfluoro acids. J. Chem. Soc. 1952, 4259−4268.
(21) Cao, P.; Duan, J. X.; Chen, Q. Y. Difluoroiodomethane:
practical synthesis and reaction with alkenes. J. Chem. Soc., Chem.
Commun. 1994, 737−738.
(22) We note that it is also possible to generate 1 using Zn dust as
the source of Zn. We chose not to explore this route given the
possible batch to batch variability expected with heterogeneous Zn
powder, potential irreproducible kinetics arising from mass transfer
and the potential additional complexity that would arise from the use
of Zn dust in continuous flow.
(23) This compound was reported to form iodine upon exposure to
light, see: Atherton, J. H.; Fields, R.; Haszeldine, R. N. Carbene
chemistry. Part II. Migration in fluoroalkylcarbenes. J. Chem. Soc. C
1971, 366. After distillation the solution was stored in the dark.
(24) Cole, K. P.; Groh, J. M.; Johnson, M. D.; Burcham, C. L.;
Campbell, B. M.; Diseroad, W. D.; Heller, M. R.; Howell, J. R.;
Kallman, N. J.; Koenig, T. M.; May, S. A.; Miller, R. D.; Mitchell, D.;
Myers, D. P.; Myers, S. S.; Phillips, J. L.; Polster, C. S.; White, T. D.;
Cashman, J.; Hurley, D.; Moylan, R.; Sheehan, P.; Spencer, R. D.;
Desmond, K.; Desmond, P.; Gowran, O. Kilogram-scale prexasertib
monolactate monohydrate synthesis under continuous-flow CGMP
conditions. Science 2017, 356, 1144−1150.
G
https://dx.doi.org/10.1021/acs.oprd.0c00089
Org. Process Res. Dev. XXXX, XXX, XXX−XXX