Scale-Up of a Continuous Extraction Process for Driving an Equilibrium-Limited Reaction to Completion

Article abstract of DOI:10.1021/acs.oprd.8b00260

We report a strategy for implementing a scalable continuous extraction to drive an equilibrium-limited reaction to completion. Our approach is simple, requiring only standard equipment and glassware, and can be designed to meet the needs of the process (e.g., reduced cycle time or reduced solvent use) with minimal development time. We investigated the process parameters, such as the flow rate of fresh n-heptane (extracting solvent) into the system, at a 0.5-5 g scale and reduced the overall time and total amount of n-heptane needed to drive the reaction to completion. The improved conditions for the extraction were demonstrated at the 50 g scale, leading to a similar yield of the desired product and total reaction time as at the 5 g scale, verifying that our reactions and extraction conditions were scalable. In particular, our fit-for-purpose approach is customizable to a number of systems and can be implemented in a short period of time with a limited amount of material.

Full text of DOI:10.1021/acs.oprd.8b00260

Subscriber access provided by University of Sunderland
Communication
Scale-Up of a Continuous Extraction Process for
Driving an Equilibrium-Limited Reaction to Completion
Michael Tudesco, Eric G. Moschetta, and Eric A. Voight
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00260 • Publication Date (Web): 16 Oct 2018
Downloaded from http://pubs.acs.org on October 17, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Scale-Up of a Continuous Extraction Process for Driving an Equilibrium-Limited
Reaction to Completion
Michael T. Tudesco,1*† Eric G. Moschetta,1*† and Eric A. Voight2
1Process Research and Development, AbbVie, Inc., 1401 Sheridan Road, North Chicago,
IL 60064-1802, USA.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2Discovery Chemistry and Technology, AbbVie, Inc., 1 N Waukegan Road, North
Chicago, IL, 60064-1802, USA.
*Corresponding authors: michael.tudesco@abbvie.com, eric.moschetta@abbvie.com
†Both authors contributed equally to this work
For Table of Contents Only
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Abstract
We report a strategy for implementing a scalable continuous extraction to drive an
equilibrium-limited reaction to completion. Our approach is simple, requiring only
standard equipment and glassware, and can be designed to meet the needs of the process
(e.g. reduced cycle time or reduced solvent use) with minimal development time. We
investigated the process parameters, such as the flow rate of fresh n-heptane (extracting
solvent) into the system, at a 0.5-5 g scale and reduced the overall time and total amount
of n-heptane needed to drive the reaction to completion. The improved conditions for the
extraction were demonstrated at the 50 g scale, leading to similar yield of the desired
product and total reaction time at the 5 g scale, verifying that our reactions and extraction
conditions were scalable. In particular, our fit-for-purpose approach is customizable to a
number of systems and can be implemented in a short period of time with a limited
amount of material.
Keywords: continuous extraction, equilibrium reaction, deprotection, scale-up
Introduction
Liquid-liquid extraction is a simple, effective, and mild method for purifying
reaction mixtures in organic chemistry. Mixing two immiscible liquid solvents causes the
solutes to partition themselves between the two phases based on each solute’s
1
thermodynamic preference for residing in the two phases. This thermodynamic
partitioning can be exploited to drive mass transfer of either the desired products or the
2
ACS Paragon Plus Environment

Page 3 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
2
undesired impurities into the extracting phase. The two liquid phases are separated and
the phase containing the desired products is carried forward in the process. Extraction
occurs frequently at mild temperatures, which is important when thermally or chemically
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
unstable species are present and must be removed from the reaction mixture quickly.3
One of the most attractive features of extractions is that they are usually readily
4
transferrable from laboratory to manufacturing scale. Furthermore, extractions may be
executed in several stages in batch processes or operated as continuous flow systems,
5
demonstrating the flexibility of extraction as a unit operation. In flow, common
arrangements of continuous extraction equipment include mixer-settlers, packed columns,
and centrifugal extractors. Mixer-settlers are particularly useful for extractions that
require only a few equilibrium stages, provide flexible and simple operation, and are
readily scalable. Extraction technology is versatile and the desired unit operation is a
choice of cost, convenience, and ease of scale, among other factors.
In the pharmaceutical industry, the vast majority of liquid-liquid extractions are
carried out in batches, leaving continuous separation technology underdeveloped. In
particular, there are situations where batch extraction is inefficient and cumbersome to
scale compared to a continuous flow extraction process. For example, when a large
number of batch extractions are required to purify the reaction mixture, the process incurs
significant costs in labor and solvent use. In such cases, continuous flow extraction is
highly appealing from a number of perspectives. For extractions that operate far from
thermodynamic equilibrium, the flow rate of fresh extracting solvent can be tuned to
increase the rate of mass transfer (when quickly removing an unstable material is the
main objective at the expense of increased use of solvent) or to reduce the amount of
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
fresh solvent needed to perform the extraction (at the expense of increased operating
time). Additionally, implementing a continuous process reduces the labor requirement
inherent to a large number of manufacturing-scale batch extractions and can reduce the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
total operating time compared to a batch extraction. Continuous extractions can be
integrated with flow reactors and other continuous flow unit operations to intensify and
streamline the overall process and potentially reduce the footprint needed to execute the
process at larger scales.4d, 4e, 7
We recently encountered a situation in which implementing a continuous
extraction provided an opportunity for process intensification. During optimization of the
route to nucleoside 1 (see Figure 1), the nominal procedure for the
deprotection/lactonization reaction of cyclohexylidene ester 2 to lactone diol 4 (Step 4)
was identified as being difficult to scale up.
Figure 1. Synthesis of nucleoside 1.a
Me
OH
F
O
Me
N
O
P
H O
OH
O
HO
HO
O
F
EtO
O
N
p-TSA
p-TSA
O
O
OEt
O
O
NH
HO
OEt
O
O
Br
H O
n-PrCN
Br
O
2
OH F Br
MeCN
Ph
HO
Br
F
2
3
4
1
Although appearing to be quite simple, the initial deprotection step in the reaction proved
challenging due to the deprotected ethyl ester triol (3) and cyclohexanone being in
equilibrium with the cyclohexylidene ester starting material (2, Figure 2).
Figure 2. Equilibrium between the deprotected ethyl ester triol (3) and cyclohexylidene
ester (2) starting material.
a
The syntheses of compounds 1 and 2 will be disclosed in a future publication.
4
ACS Paragon Plus Environment

Page 5 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
3
OH
O
HO
HO
O
OH
OH
F
O
F
Br
10 mol% pTSA
PrCN
80 °C
O
O
O
+
O
Br
O
MeCN/H2O (2:1)
HO
F
O
50 °C
Br
2
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In order to drive the equilibrium towards desired deprotected ethyl ester triol (3), the
nominal procedure required distillation with an extensive amount of solvent to drive off
cyclohexanone. While this approach was effective on decagram scale, the distillations
proved to be difficult to perform and the results were thus expected to be challenging to
reproduce on kilogram scale. In addition, it was noted during the optimization of the
nominal procedure that most of the water used in the reaction must be removed prior to
lactonization to prevent hydrolysis of the deprotected ethyl ester triol (3) to the
corresponding acid. This acid would cyclize during the lactonization to an undesired six-
membered lactone that did not reject well in the crystallization. Removing the water
required azeotroping with butyronitrile several times, which would also be problematic at
kilogram scale. With multi-kilogram deliveries of nucleoside 1 planned to begin in six
weeks, the project team began exploring alternative methods to make the process more
amenable to running on kilogram scale.
Results and Discussion
The investigation into alternative processes began with attempting to find the best
conditions for the deprotection, since that part of the reaction was the most problematic to
scale. After exploring several different conditions for the deprotection, an approach
using an acidic resin (Dowex 50WX8) appeared most promising (see Figure 3).
Unfortunately, the reaction appeared to stall at ~50% conversion despite screening
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
various solvents, temperatures, resins, and loadings. Since the initial equilibrium
conversion could not be further improved, efforts were made to find ways other than
distillation to increase the conversion after reaching an initial equilibrium.
Figure 3. Deprotection using Dowex 50WX8 resin.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OH
O
OH
O
OH
O
OH
O
MeO OMe
Dowex 50WX8
MeOH
O
O
HO
HO
HO
HO
O
+
O
O
+
O
+
F
Br
F
Br
F
Br
F
Br
O
O
2
6
3
5
7
8
One such approach, inspired by the protocol used by Evans et. al., was to filter
off the acidic resin used for the deprotection, extract the MeOH solution with n-heptane
three times, and then re-subject the MeOH solution to the resin. After performing five
cycles of filtering, extracting, and then adding the resin back to the solution, the reaction
reached 100% conversion to a mixture of the ethyl ester triol (3) and methyl ester triol (5,
see Table 1, Entry 1). The reaction was therefore scaled-up from 100 mg to 5 g, which
required 14 cycles (42 total extractions) to reach 99% conversion (see Table 1, Entry 2).
The significant increase in the number of extractions needed when increasing the scale of
the reaction was alarming in the sense that the reaction would be run on kilogram
quantities of material and the cause of this dramatic increase was unclear.
Table 1. Extracting with n-heptane to push the reaction to complete conversion.
OH
O
OH
O
O
Dowex 50WX8 (10 wt%)
MeOH (10 vol.)
HO
HO
O
OR
F
Br
F
Br
O
30 °C
3
4
: R = Et
: R= Me
2
Entry Scale Inital Conversion Extractions Final Conversion
1
2
100 mg
5.00 g
49% (22.5 h)
47% (17.5 h)
15
42
100%
99%
To further explore this issue, an investigation into the partition coefficients of
cyclohexanone dimethyl ketal (7, generated instead of cyclohexanone when MeOH is the
solvent for the reaction) into other organic solvents was conducted. As shown in Table 2,
6
ACS Paragon Plus Environment

Page 7 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the partition coefficient of cyclohexanone dimethyl ketal only slightly favors extraction
from MeOH in all solvent systems explored (the small number of systems explored is due
to only n-hexane, n-heptane, and cyclohexane being immiscible with MeOH).9 These
partition coefficient values help to explain why so many extractions are needed to drive
the reaction to completion, as each extraction only removes small amounts of
cyclohexanone dimethyl ketal. This discovery led to consideration of a continuous
extraction approach, as it would allow for constant removal of cyclohexanone dimethyl
ketal.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Partition coefficient of cyclohexanone dimethyl ketal (7) in various solvent
systems.
[
푎푛푎푙푦푡푒 푖푛 푠표푙푣푒푛푡 퐴]
푝푎푟푡푖푡푖표푛 푐표푒푓푓푖푐푖푒푛푡 =
[푎푛푎푙푦푡푒 푖푛 푠표푙푣푒푛푡 퐵]
Solvent A Solvent B
Analyte
Partition Coefficient Ratio Solvent A:Solvent B for Phase Cut
n-hexane
n-heptane
MeOH
MeOH
cyclohexanone dimethyl ketal
cyclohexanone dimethyl ketal
cyclohexanone dimethyl ketal
1.24
1.36
1.51
2:3 (slow phase cut)
1:1 (fast phase cut)
2:1 (slow phase cut)
cyclohexane MeOH
A continuous extractor was constructed from simple and readily available lab
equipment, as shown in Figure 4, to obtain proof of concept. A pump was used to dose
n-heptane into the reaction vial, which contained the cyclohexylidene ester starting
material (2), resin, and MeOH at 30 °C (the mixer). A second pump removed solution
from the reaction vial and into a pressure-equalizing addition funnel (the settler). This
addition funnel was pre-filled with enough n-heptane to reach the sidearm. As the
reaction solution was pumped into the addition funnel, n-heptane would flow down the
sidearm and collect in a waste flask. Finally, a third pump transferred MeOH solution
from the bottom of the addition funnel and back into the reaction flask. Therefore, as
fresh n-heptane was being pumped into the reaction vial, the biphasic mixture in the vial
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was being pumped out into the addition funnel where the MeOH droplets (containing
compounds 2, 3, 5, and 6) would flow through the n-heptane layer (thus continuing the
extraction of cyclohexanone dimethyl ketal that started in the mixer) and settle to the
bottom before being pumped back into the reaction vial.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Continuous extractor design.
Reaction solution to extractor
Fresh n-heptane
MeOH
solution
recycle
n-heptane layer
mixer
settler
MeOH layer
n-heptane waste
With a suitable design for the continuous extractor in hand, initial experiments
were run to test feasibility. Two experiments were performed at 500 mg scale to evaluate
the most efficient way to use the continuous extractor. In one experiment, a pre-
equilibration period was utilized, where the continuous extractor was turned off for the
first 18 h, to allow the reaction to reach its initial equilibrium conversion. In a second
experiment, the continuous extractor was turned on from the beginning of the reaction.
As shown in Figure 5, both reactions proceeded noticeably faster than the 5 g batch
reaction. The difference in reaction time between these two reactions was most likely
due to using a longer than necessary pre-equilibration time, as the reaction likely stalled
8
ACS Paragon Plus Environment

Page 9 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
long before the continuous extractor was started. In light of these promising results, it
was hypothesized that the reaction could be even faster if a soluble acid was used in place
of the acidic resin. Once the MeOH phase exits the mixer, it is not exposed to the acidic
resin and the deprotection reaction stops. Therefore, any time the MeOH solution resides
in the tubing or the addition funnel, no reaction occurs, which unnecessarily increases the
time required to reach complete conversion. This change would potentially reduce
reaction time, which would in turn lower the amount of n-heptane used to reach 100%
conversion.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Reaction conversion: batch conditions vs. continuous extractor.
100
90
80
70
60
50
40
30
20
10
0
Batch Reaction with resin (extractions at each time point), 5 g
Continuous Extractions with resin, pre-equilibration
Continuous Extractions with resin, no pre-equilibration
0
20
40
60
80
Time (h)
To test this hypothesis, the reaction was set up as before (no pre-equilibration
period was used since it did not seem to positively affect the rate of the reaction with the
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
acidic resin), using p-TSA as the acid instead of the acidic resin. As shown in Figure 6,
the reaction reached 100% conversion in less than a day and was significantly faster than
the corresponding reaction with the acidic resin. With these results in hand, the reaction
was scaled up to 5 g (see Table 3, Entry 1 for results). In this reaction, the reactor
initially contained 25 mL (5 volumes) n-heptane, assuming that this would allow for
quicker removal of the cyclohexanone dimethyl ketal (7) from the reaction solution.
While the potency adjusted yield (PAY) of the lactone diol (4) was as expected (63%, see
Experimental Section for procedure to synthesize lactone diol 4 following completion of
the deprotection reaction), the final amount of n-heptane used (175 volumes) was
excessive for a commercial process. The process was thus improved to reduce the
amount of n-heptane being used.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Reaction conversion: acidic resin vs. p-TSA with the continuous extractor.
100
9
0
0
8
70
60
50
40
30
20
10
0
Continuous Extractions with resin, no pre-equilibration
Continuous Extractions with p-TSA
Continuous Extractions with p-TSA, 5 g
0
5
10
15
20
25
30
Time (h)
10
ACS Paragon Plus Environment

Page 11 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Table 3. Summary of results from scale-up/improvement of the continuous extractor
system.
Entry
Scale
PAY Lactone Diol (4) HPLC Purity/Potency Volumes n-Heptane Used
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
5 g
5 g
63%
98.9 peak area% purity
98.8 wt/wt% potent
97.9 wt/wt% potent
175
60.5
51.4
69.3%
63.8%
50 g
The reaction was modified, such that a 3 h pre-equilibration period was
incorporated (continuous extractor turned off), but the n-heptane pump was used to dose
5
volumes of n-heptane over the course of the pre-equilibration period. After the
equilibration period was over, the pumps to and from the extractor were turned on while
maintaining the same flow rate of fresh n-heptane over the remainder of the extraction.
These conditions led to a similar PAY of lactone diol 4 (69.3%); however, significantly
less n-heptane (60.5 volumes) was used (Table 3, Entry 2). The reaction did take longer
to reach full conversion (Figure 7), but since the reaction mixture appeared to be stable
over time and cycle time was not considered the driving cost factor for the scale-up, it
was deemed less of a concern than the n-heptane consumption. Finally, the reaction was
scaled up to 50 g, using the same conditions, which led to similar PAY for lactone diol 4
(Table 3, Entry 3). Gratifyingly, being able to use higher flow rates on larger scale led to
a similar overall reaction time (see Figure 7) as on the 5 g scale and even less n-heptane
being used (51.4 volumes).
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 7. Plot comparing conversion of various scale-up conditions for the continuous
extractor.
100
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9
0
0
8
70
60
50
40
30
20
10
0
Continuous Extractions with p-TSA, 5 g without pre-equilibration
Continuous Extractions with p-TSA, 5 g with pre-equilibration
Continuous Extractions with p-TSA, 50 g with pre-equilibration
0
10
20
Time (h)
30
40
50
Conclusion
We have developed a simple, fit-for-purpose scalable continuous flow extraction
to drive an equilibrium-limited reaction to completion, addressing an important project
challenge. We demonstrated proof-of-concept at lab scale with soluble and heterogeneous
acid catalysts and demonstrated ways to reduce the total amount of solvent used over the
course of the extraction. The established extraction protocol was successfully
demonstrated at 50 g scale. The approach described here is flexible and suitable for a
number of processes, but is especially appropriate for situations where starting material
and development time are limited, and the reaction must be driven to completion. This
method is customizable to fit the appropriate balance between cycle time and solvent
12
ACS Paragon Plus Environment

Page 13 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
usage and cost. The fact that this continuous extractor can be set up using readily
available glassware and equipment ensures that this approach can be readily adapted to
other processes that would benefit from continuous extraction.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental Section
General Methods. All commercially available materials and solvents were used directly
1
13
without further purification. H NMR (600 MHz) and C NMR (150 MHz) spectra were
recorded with a Bruker spectrometer. See Supporting Information for HPLC and GC
conditions used during experiments.
Representative Experimental Procedure for Deprotection/Lactonization Reaction
(4). A 2 L 4-neck round bottom flask equipped with a thermocouple and an overhead
stirrer was charged with cyclohexylidene ester starting material (2, 50.0 g, 1.00 equiv), p-
toluenesulfonic acid monohydrate (2.59 g, 0.100 equiv), and MeOH (500 mL, 10
volumes) and set up with a continuous extractor as shown in Figure 4 (see Supporting
Information for pump settings and types). A 25 mL pressure-equalizing addition funnel
was used as the settler in Figure 4 to separate the MeOH and n-heptane phases. It was
filled with 25 mL of n-heptane (enough to reach the sidearm) so that when solution was
pumped into it, the n-heptane overflow dripped through the sidearm and into a waste
flask. The solution was heated to 30 °C and stirred with the fresh n-heptane pump turned
on (See Supporting Information for pump flow rate) and the reaction solution to extractor
and MeOH solution recycle pumps turned off. After 3 h, the reaction solution to
extractor and MeOH solution recycle pumps were turned on (see Supporting Information
for pump flow rates). The reaction was stirred at 30 °C with the continuous extractor
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
running until <1 PA% total of the cyclohexylidene ester starting material (2) and the
methyl ester cyclohexylidene starting material (6) were observed via HPLC analysis.
The continuous extractor setup then was flushed with MeOH. The MeOH flush from the
pumps and tubing, the solution remaining in the addition funnel, and the solution in the 2
L 4-neck round bottom flask were combined and transferred to a separatory funnel. The
layers were separated and the MeOH layer (bottom layer) was washed twice with 250 mL
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(5 volumes) n-heptane. The MeOH solution was then partially concentrated under
reduced pressure at 35 °C. To the partially concentrated MeOH solution, 250 mL (5
volumes) n-PrCN was added and then the solution was concentrated to an oil under
reduced pressure at 35 °C. Butyronitrile (250 mL, 5 volumes) was added to the oil and
the solution again was concentrated to an oil under reduced pressure at 35 °C.
Butyronitrile (50 mL, 10 volumes) was then added to the oil a final time. The n-PrCN
solution was heated to 80 °C while sparging with N until <5 PA% total of the ethyl ester
2
triol (3) and the methyl ester triol (5) remained via HPLC analysis. The solution was
cooled to room temperature overnight. After cooling, the solution was partially
concentrated under reduced pressure at 35 °C to 80.17 g. The solution was then passed
through a 100 g (200 wt % relative to cyclohexylidene starting material 2) silica plug,
eluting with n-PrCN and collecting ~250 mL fractions. The fractions containing lactone
diol 4 (as determined by HPLC analysis) were collected and partially concentrated under
reduced pressure at 35 °C to 172.33 g. Chlorobenzene (250 mL, 5 volumes) was added
to the n-PrCN solution, which was then partially concentrated under reduced pressure at
3
5 °C to 138.2 g. Chlorobenzene (250 mL, 5 volumes) was added again and the solution
was partially concentrated under reduced pressure at 35 °C to 234.96 g, at which point a
14
ACS Paragon Plus Environment

Page 15 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
brown oil was observed crashing out of solution. The mixture was seeded with
crystalline lactone diol 4 (270 mg) from a previous reaction and then partially
concentrated under reduced pressure at 35 °C until a brownish orange solid was observed
crashing out of solution. The slurry was then stirred at room temperature as 150 mL (3
volumes) of dichloromethane was added slowly, causing more solid to crash out of
solution. The resulting slurry was stirred at room temperature for 2 h and then cooled to
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
°C and stirred for an additional 45 min. The slurry was then filtered while still cold and
the solids were washed twice with 50 mL (1 volume) of dichloromethane. The off-white
solids were collected and dried in a vacuum oven at 50 °C for 26 h to yield 20.31 g
(63.8% potency adjusted yield, 97.9 wt/wt% potent via HPLC analysis) of lactone diol 4
as an off-white solid. The solid contained 1.1 mol% p-TSA but was not purified further,
as it was determined that p-TSA had no impact on downstream chemistry. 1H NMR
((CD ) SO, 600 MHz) δ 6.92 (s, 1H), 4.56 (dd, J = 11.3, 5.0 Hz, 1H), 4.44 (tdd, J = 5.3,
3
2
H
.9, 1.5 Hz, 1H), 3.78 (dd, J = 12.7, 3.9 Hz, 1H), 3.68 (dd, J = 12.7, 5.4 Hz, 1H). 13C
NMR ((CD ) SO, 150 MHz) δ 166.45, 166.28, 96.38, 94.62, 85.43, 73.52, 73.42, 58.97.
3
3
2
C
Partition Coefficient Studies. The partition coefficient studies were carried out by
adding approximately 100 mg of 1,1-dimethoxycyclohexane (7) to a vial followed by
MeOH and either n-hexane, n-heptane, or cyclohexane. The vials were then shaken
vigorously and the layers were allowed to settle. The vials were allowed to sit over the
weekend at room temperature and then an aliquot from each layer was analyzed via GC
to determine the concentration of the 1,1-dimethoxycyclohexane in each solvent.
Supporting Information
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The Supporting Information contains experimental setup diagrams and pictures, pump
types and settings used, and detailed analytical data including methods and spectra.
Author Information
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Corresponding Authors
*E-mail: michael.tudesco@abbvie.com, eric.moschetta@abbvie.com
Author Contributions
†Both authors contributed equally to this work
Notes
The authors declare no competing financial interest. This study was sponsored by
AbbVie Inc. AbbVie contributed to the design, study conduct, financial support and
interpretation of data, writing, reviewing, and approving the publication. Michael
Tudesco, Eric Moschetta and Eric Voight are employees of AbbVie Inc.
Acknowledgments
The authors thank Nate Ide, Al Kruger, Moiz Diwan, Anuj Verma, Kaid Harper, Travis
Dunn, John Hartung, and Westin Morrill for helpful discussions and suggestions.
References
1
. (a) Selekman, J. A.; Tran, K.; Xu, Z.; Dummeldinger, M.; Kiau, S.; Nolfo, J.; Janey, J.,
High-Throughput Extractions: A New Paradigm for Workup Optimization in
Pharmaceutical Process Development. Org. Process Res. Dev. 2016, 20 (10), 1728-1737;
(b) Ashworth, I. W.; Meadows, R. E., A General Liquid-Liquid Partitioning Equation and
Its Consequences: Learning from the pH Dependent Extraction of a Pharmaceutical
Intermediate. J. Org. Chem. 2018, 83 (7), 4270-4274.
2
. (a) Hsieh, D.; Marchut, A. J.; Wei, C.; Zheng, B.; Wang, S. S. Y.; Kiang, S., Model-
Based Solvent Selection during Conceptual Process Design of a New Drug
Manufacturing Process. Org. Process Res. Dev. 2009, 13 (4), 690-697; (b) Abolhasani,
M.; Coley, C. W.; Jensen, K. F., Multiphase Oscillatory Flow Strategy for in Situ
Measurement and Screening of Partition Coefficients. Anal. Chem. 2015, 87 (21), 11130-
11136.
16
ACS Paragon Plus Environment

Page 17 of 18
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
3
. Boucher, M. M.; Furigay, M. H.; Quach, P. K.; Brindle, C. S., Liquid–Liquid
Extraction Protocol for the Removal of Aldehydes and Highly Reactive Ketones from
Mixtures. Org. Process Res. Dev. 2017, 21 (9), 1394-1403.
4
H. J., Continuous Chiral Separation of Amino Acid Derivatives by Enantioselective
Liquid−Liquid Extraction in Centrifugal Contactor Separators. Org. Process Res. Dev.
2
G.; Heeres, H. J., Scalable Enantioseparation of Amino Acid Derivatives Using
Continuous Liquid−Liquid Extraction in a Cascade of Centrifugal Contactor Separators.
Org. Process Res. Dev. 2009, 13 (5), 911-914; (c) Wang, Y.; Tang, K.; Zhang, P.; Zhou,
J.; Huang, Y.; Wen, P.; Sun, G., Continuous Separation of Pantoprazole Enantiomers by
Biphasic Recognition Chiral Extraction in Centrifugal Contactor Separators. Org.
Process Res. Dev. 2014, 19 (9), 1082-1087; (d) Falss, S.; Tomaiuolo, G.; Perazzo, A.;
Hodgson, P.; Yaseneva, P.; Zakrzewski, J.; Guido, S.; Lapkin, A.; Woodward, R.;
Meadows, R. E., A Continuous Process for Buchwald-Hartwig Amination at Micro-,
Lab-, and Mesoscale Using a Novel Reactor Concept. Org. Process Res. Dev. 2016, 20
. (a) Schuur, B.; Floure, J.; Hallett, A. J.; Winkelman, J. G. M.; deVries, J. G.; Heeres,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
008, 12 (5), 950-955; (b) Schuur, B.; Hallett, A. J.; Winkelman, J. G. M.; de Vries, J.
(2), 558-567; (e) Johnson, M. D.; May, S. A.; Calvin, J. R.; Remacle, J.; Stout, J. R.;
Diseroad, W. D.; Zaborenko, N.; Haeberle, B. D.; Sun, W.-M.; Miller, M. T.; Brennan, J.,
Development and Scale-Up of a Continuous, High-Pressure, Asymmetric Hydrogenation
Reaction, Workup, and Isolation. Org. Process Res. Dev. 2012, 16 (5), 1017-1038; (f)
Tsukanov, S. V.; Johnson, M. D.; May, S. A.; Kolis, S. P.; Yates, M. H.; Johnston, J. N.,
Continuous Platform To Generate Nitroalkanes On-Demand (in Situ) Using Peracetic
Acid-Mediated Oxidation in a PFA Pipes-in-Series Reactor. Org. Process Res. Dev.
2
018, 22 (8), 971-977; (g) Teoh, S. K.; Toh, Q. Y.; Noorulameen, M. S.; Sharratt, P. N.,
Sustainability Improvements through Catalyst Recycling in a Liquid–Liquid Batch and
Continuous Phase Transfer Catalyzed Process. Org. Process Res. Dev. 2017, 21 (4), 520-
5
5
Jensen, K. F., Design of Multistage Counter-Current Liquid-Liquid Extraction for Small-
Scale Applications. Ind. Eng. Chem. Res. 2017, 56 (14), 4095-4103.
6
Stage, Continuous High-Efficiency Extraction Device (HEED) for Flow Synthesis. Org.
Process Res. Dev. 2016, 20 (10), 1738-1743.
7
Flow Approach to 3,3,3-Trifluoromethylpropenes: Bringing Together Grignard Addition,
Peterson Elimination, Inline Extraction, and Solvent Switching. Org. Process Res. Dev.
30.
. Weeranoppanant, N.; Adamo, A.; Saparbaiuly, G.; Rose, E.; Fleury, C.; Schenkel, B.;
. Day, C.; Saldarriaga, A.; Tilley, M.; Hunter, H.; Organ, M. G.; Wilson, D. J., A Single-
. (a) Hamlin, T. A.; Lazarus, G. M. L.; Kelly, C. B.; Leadbeater, N. E., A Continuous-
2
014, 18 (10), 1253-1258; (b) Peer, M.; Weeranoppanant, N.; Adamo, A.; Zhang, Y.;
Jensen, K. F., Biphasic Catalytic Hydrogen Peroxide Oxidation of Alcohols in Flow:
Scale-up and Extraction. Org. Process Res. Dev. 2016, 20 (9), 1677-1685; (c) Chartoire,
A.; Claver, C.; Corpet, M.; Krinsky, J.; Mayen, J.; Nelson, D.; Nolan, S. P.; Penafiel, I.;
Woodward, R.; Meadows, R. E., Recyclable NHC Catalyst for the Development of a
Generalized Approach to Continuous Buchwald-Hartwig Reaction and Workup. Org.
Process Res. Dev. 2016, 20 (2), 551-557; (d) Glotz, G.; Lebl, R.; Dallinger, D.; Kappe, C.
O., Integration of Bromine and Cyanogen Bromide Generators for the Continuous-Flow
Synthesis of Cyclic Guanidines. Angew. Chem. Int. Ed. 2017, 56 (44), 13786-13789; (e)
Susanti, S.; Meinds, T. G.; Pinxterhuis, E. B.; Schuur, B.; de Vries, J. G.; Feringa, B. L.;
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Winkelman, J. G. M.; Yue, J.; Heeres, H. J., Proof of concept for continuous
enantioselective liquid–liquid extraction in capillary microreactors using 1-octanol as a
sustainable solvent. Green Chem. 2017, 19 (18), 4334-4343; (f) 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 (6343), 1144-1150; (g) Adamo,
A.; Beingessner, R. L.; Behnam, M.; Chen, J.; Jamison, T. F.; Jensen, K. F.; Monbaliu,
J.-C. M.; Myerson, A. S.; Revalor, E. M.; Snead, D. R.; Stelzer, T.; Weeranoppanant, N.;
Wong, S. Y.; Zhang, P., On-demand continuous-flow production of pharmaceuticals in a
compact, reconfigurable system. Science 2016, 352 (6281), 61-67; (h) Heider, P. L.;
Born, S. C.; Basak, S.; Benyahia, B.; Lakerveld, R.; Zhang, H.; Hogan, R.; Buchbinder,
L.; Wolfe, A.; Mascia, S.; Evans, J. M. B.; Jamison, T. F.; Jensen, K. F., Development of
a Multi-Step Synthesis and Workup Sequence for an Integrated, Continuous
Manufacturing Process of a Pharmaceutical. Org. Process Res. Dev. 2014, 18 (3), 402-
409.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8. Evans, D. A.; Burch, J. D., Asymmetric Synthesis of the Chlorocyclopropane-
Containing Callipeltoside A Side Chain. Org. Lett. 2001, 3 (4), 503-505.
9
. http://www.scharlab.com/tabla-reactivos-mezclabilidad.php Solvent Miscibility Table.
http://www.scharlab.com/tabla-reactivos-mezclabilidad.php.
For Table of Contents Only
18
ACS Paragon Plus Environment