Scalable Continuous Synthesis of Organozinc Reagents and Their Immediate Subsequent Coupling Reactions

Article abstract of DOI:10.1021/acs.oprd.0c00399

The continuous synthesis of organozinc reagents and their immediately following subsequent also continuous consumption in catalyzed and noncatalyzed coupling reactions were investigated. In the first step, a bed of Zn turnings at variable liquid throughputs and concentrations of organic halide solutions was used, and the formed Zn organometallics were analyzed for quality control. They were then directly pumped into a second step, namely, Reformatsky, Saytzeff, and Negishi coupling reactions. In the organozinc halides' formation, a novel process window was employed by using a large molar excess of Zn turnings and investigating mechanical as well as chemical Zn activation. Subsequent couplings of the freshly prepared Zn organometallics were done using examples of a Reformatsky, Saytzeff, and Negishi coupling reaction. For the Zn organometallics' formation, a laboratory-scale reactor setup previously built for Grignard reagent formation was evaluated including a Zn replenishing unit; the same reactor was also used in the metal-catalyzed subsequent step (Negishi coupling). The main objective of this work was to establish the scalable continuous formation of Zn organometallic reagents enabling fast and safe process optimization, analyze the reagents for their purity, and then immediately consume them in various follow-up steps, always only leaving a very small amount of reactive and sensitive organometallic reagent in the setup. It was found that full conversion of the employed halides could be achieved within a single passage through the reactor with organozinc yields of 82-92%, as well as being able to successfully perform subsequent non- and metal-catalyzed coupling steps with yields of up to 92%. A pilot-scale setup allowing a liquid throughput of up to 3-5 L/h has also been built and is ready to be tested with the synthesis as established here.

Full text of DOI:10.1021/acs.oprd.0c00399

pubs.acs.org/OPRD
Article
Scalable Continuous Synthesis of Organozinc Reagents and Their
Immediate Subsequent Coupling Reactions
̈
̈
Gabriele Menges-Flanagan,* Eva Deitmann, Lars Gossl, Christian Hofmann, and Patrick Lob
Cite This: https://dx.doi.org/10.1021/acs.oprd.0c00399
Read Online
Metrics & More
Article Recommendations
ACCESS
ABSTRACT: The continuous synthesis of organozinc reagents and their immediately following subsequent also continuous
consumption in catalyzed and noncatalyzed coupling reactions were investigated. In the first step, a bed of Zn turnings at variable
liquid throughputs and concentrations of organic halide solutions was used, and the formed Zn organometallics were analyzed for
quality control. They were then directly pumped into a second step, namely, Reformatsky, Saytzeff, and Negishi coupling reactions.
In the organozinc halides’ formation, a novel process window was employed by using a large molar excess of Zn turnings and
investigating mechanical as well as chemical Zn activation. Subsequent couplings of the freshly prepared Zn organometallics were
done using examples of a Reformatsky, Saytzeff, and Negishi coupling reaction. For the Zn organometallics’ formation, a laboratory-
scale reactor setup previously built for Grignard reagent formation was evaluated including a Zn replenishing unit; the same reactor
was also used in the metal-catalyzed subsequent step (Negishi coupling). The main objective of this work was to establish the
scalable continuous formation of Zn organometallic reagents enabling fast and safe process optimization, analyze the reagents for
their purity, and then immediately consume them in various follow-up steps, always only leaving a very small amount of reactive and
sensitive organometallic reagent in the setup. It was found that full conversion of the employed halides could be achieved within a
single passage through the reactor with organozinc yields of 82−92%, as well as being able to successfully perform subsequent non-
and metal-catalyzed coupling steps with yields of up to 92%. A pilot-scale setup allowing a liquid throughput of up to 3−5 L/h has
also been built and is ready to be tested with the synthesis as established here.
KEYWORDS: continuous processing, flow chemistry, organometallics, scale-up, organozinc compound
can be employed as reactive intermediates in complex
multistep synthetic processes. Among others, they are utilized
in non-metal-catalyzed Reformatsky reactions (condensation
reaction between a carbonyl compound and an α-halide
carboxylic acid ester),9,10 Saytzeff reactions (reaction between
an active organozinc halide and either an aldehyde or ketone to
form alcohols),11 and Blaise reactions (formation of β-
ketoesters via the reaction of Zn with an α-bromoester and a
nitrile).12,13 The most prominent metal-catalyzed coupling
reaction is the Negishi coupling between a zinc organometallic
species and an organic halide under Pd or Ni catalysis, which
earned its inventor the 2010 Nobel prize for chemistry, almost
100 years after Victor Grignard was awarded his for the
equivalent Mg organometallic species.3,14 Their unique
combination of above-mentioned functional group tolerance,
moderate reactivity, and formation of nontoxic side products as
compared to other organometallic reagents truly renders them
ideal candidates for a multitude of C−C bond formation
INTRODUCTION
■
Organozinc reagents were among the initial organometallic
species to be discovered and used in C−C bond formation
reactions. The first formation of a zinc reagent was performed
by E. Frankland in 1848, reacting zinc metal with ethyl iodide.1
They enjoyed prominence and were used for C−C bond
formation in the second half of the 19th century until the
discovery of Grignard reagents by Victor Grignard,2 and their
subsequent coupling reactions in the actual Grignard reaction
eclipsed their use. However, through their combination of a
lower reactivity compared to Grignard reagents and therefore a
higher tolerance toward functional groups (esters, ketones,
cyanides, halides, primary and secondary amines, etc.), within
reaction partners they constitute ideal reagents for multistep
synthesis particularly in the synthesis of active pharmaceutical
ingredients (APIs).3 The organozinc reagents as described in
this paper of the general formula RZnX (R constituting an
organic moiety and X denoting a halide, most often Cl or Br)
can be formed either via direct insertion of metallic zinc into
an organo-halide (RX) bond or via transmetalation.4−6 Newer
developments in the field of synthesizing Zn organometallic
reagents have been concerned with the improvement of their
handling and particularly their air and moisture sensitivity,
where the reagents can be stabilized as solid-form Zn pivalates
which can then more easily be handled and used in follow-up
coupling reactions.7,8 As mentioned above, organozinc halides
Special Issue: Celebrating Women in Process Chem-
istry
Received: September 6, 2020
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Organic Process Research & Development
pubs.acs.org/OPRD
Article
reactions and constitutes some crucial advantages over the use
of other reactive intermediates.
first step reactor via 3D laser melting allows cost-efficient
reactor fabrication while enabling access to heat exchanger
structures not amenable via conventional manufacturing, which
plays a crucial role in accomplishing effective heat exchange
structures for scale-up.29 Figure 1 details the reactor
In recent years, a number of studies have been published on
a small laboratory scale synthesizing organozinc species
continuously followed by their consumption in a second
reaction step.7,15−21 These studies prove how valuable
organozinc species are, particularly as intermediates for
manufacturing in the pharmaceutical industry; however, their
wider-spread utilization is hampered by their instability and
sensitivity to oxygen and water. The results presented here
demonstrate how a scalable continuous processing approach
can lead to “on demand” production of organozinc species
where these in situ generated reactive intermediates are
immediately consumed in a second subsequent, continuous,
and scalable processing step. The general interest in the topic
of combining flow chemistry with the generation and use of
organometallic reactive intermediates can also be seen in
several recent reviews on the subject.22−24
It is noteworthy here that, among the currently most used
synthetic pathways in API synthesis, no new routes have been
established within the last 20 years, and also within the 1980s
and 1990s, only two new synthetic pathways (Suzuki−Miyaura
and Buchwald−Hartwig) were established.25 In an analysis by
Roughley et al.,26 it was found that roughly 10 different
reaction types make up about two-thirds of the reactions used
in API syntheses today. Since there may be more and more
APIs synthesized but not necessarily more new drugs being
approved, and a bias is apparent toward certain classes of
reagents, there is a clear need to make new reaction pathways
accessible that possess a high functional group tolerance and
therewith eliminate the need for protecting groups, which is
applicable to the Zn organometallic species investigated here.25
The chemistry explored ideally combines well-established
chemistry (e.g., Pd-catalyzed C−C cross couplings) with the
wider accessibility of Zn organometallic reagents through
continuous processing. Particularly, the Suzuki−Miyaura
coupling is an impressive example of how the uptake of
novel chemistry (here for the synthesis of biphenyls) can lead
to a steady increase in using such synthetic fragments from 2%
in the 1980s to 12% today.25 This paper describes an
expansion of the playing field for process chemists in the
pharmaceutical industry through providing scalability of the
continuous processing of organozinc species to industrially
relevant throughputs, with technology ready to be adopted in
industry. More recently, a review on “the importance of
synthetic chemistry in the pharmaceutical industry” found that,
despite the advent of new pathways such as, e.g., biocatalysis,
olefin metathesis, and cross-couplings, there remains a need for
innovation in synthetic chemistry.27
Figure 1. Laboratory-scale reactor for the synthesis of organometallic
reactive intermediates via solids processing. Used with permission
from ref 28. Copyright 2020 American Chemical Society.
constructed and patented30 initially for Grignard reagent
formation and used here for organozinc reagent formation. It
allows for two different temperature zones to be applied within
a synthesis since the reactor is manufactured from two equal
halves laser welded together. Furthermore, it contains a manual
Zn replenishing unit that can be attached to the main reactor
body.
Titration was used as an immediate tool to determine RZnX
concentration following the procedure described by Huck15
with a modification for the analysis of allyl zinc bromide using
a back-titration as performed by Knochel31 and specifically
adapted by Deitmann.32 For a more detailed analysis of the
remaining starting material and possible side product content,
the samples were quenched with 1-butanol and analyzed via
gas chromatography (GC). Initially, activation of the Zn was
attempted solely through mechanical action through a jogging
motor. However, it was found that full halide conversion could
not be achieved using only mechanical activation. The start of
Zn organometallic reagent formation could be observed but
only occurred in the upper half of the reactor. Therefore, the
reactor volume was not fully used for the formation of the
organozinc species. This might be overcome by investigating
the use of a more sophisticated mechanical activation directly
within the reactor or the employment of larger Zn granules. In
the following, chemical activation has been used for the
employed Zn granules in the reactor, namely, following the
procedure by Huck et al.19 using a mixture of chlorotrime-
thylsilane and 1-bromo-2-chloroethane.
RESULTS AND DISCUSSION
■
For the synthesis of RZnX reagents, a laboratory-scale reactor
as designed and manufactured via 3D laser melting by
Fraunhofer IMM for Grignard reagent formation28 was used.
For the case of filling with Mg turnings, a behavior close to
plug-flow was found (Bodenstein numbers of up to 93 at flow
rates between 1 and 4 mL/min), and a similar behavior is
expected here; however, it has not been investigated in detail
for Zn granules. The same reactor was used in the subsequent
palladium-catalyzed Negishi coupling, while the noncatalyzed
Reformatsky and Saytzeff reactions were performed with a
PTFE coil immersed in a thermostat bath to provide sufficient
residence time for the synthesis in question. Manufacturing the
All substances investigated in the organozinc reagent
preparation were used at a concentration of 0.5 mol/L in
THF since reference substances at this concentration are
commercially available. Residence times in the reactor ranged
from about 3.0 to 13.0 min and were dependent on the halides’
reactivity and therewith time needed to reach full conversion
within a single passage through the reactor. To establish
optimal reaction conditions and ensure reproducibility,
B
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
multiple runs for each new RZnX reagent were performed.
Investigated variables were reaction temperature (through the
setting of the thermostat bath (TT) employed for heating/
cooling the reactor) and halide flow rate with the goal, again,
to achieve full halide conversion within a single passage of
halide solution through the reactor. In other words, the goal
was to achieve a maximum possible flow rate while
simultaneously using minimal energy for excessive heating/
cooling of the reactor. The organozinc formation reactions
could be investigated with the reactor as given above over a
period of up to 4 h, and the Zn replenishing unit could be
tested for each organozinc reagent in a two-step process
verifying its general applicability when running experiments
over a longer period of time. Figure 2 shows the temperature
profiles along the reactor for the synthesis of a 0.5 M benzyl
zinc bromide solution in THF.
benzyl zinc bromide, the suppression of coupling between
starting material and product is noteworthy; the coupling
product lay below the detection limit of the GC method used
and is a notable improvement to the batch synthesis where up
to 100% coupling product under nonoptimized conditions is
observed.33 The reactions were performed over a time period
of up to 4 h. In some cases, a small amount of white precipitate
could be observed after the reaction within the reactor, most
likely due to side product formation through reaction of
organozinc species with traces of water. However, this did not
impede experiments being run here and did not result in a
clogging of the reactor. In the future, longer runs and great care
to ensure the absence of water are needed to investigate the
true potential of using this form of continuous processing over
the course of a longer period of time. This will also clarify
further the influence of the Zn bed changing over time, with
particles shrinking in size and being partly eluded from the
reactor.
Each of the above synthesized RZnX compounds was used
in a specific coupling reaction (Reformatsky, Saytzeff, and
Negishi coupling) using either a Teflon tubing coil (inner
volume about 48 mL) as residence time loop immersed in a
thermostat bath for the noncatalyzed reactions or a catalyst-
filled stainless steel reactor identical in construction to the one
used in the organozinc reagent formation. Figure 3 shows the
Figure 2. Temperature profile along the reactor. Flow rate, 4 mL/min
(filling: 8 mL/min); 0.5 M benzyl Br in THF; thermostat bath setting
(TT), 40 °C; full conversion of benzyl Br reached; yields, 90−92%.
It was found that the temperature profiles along the reactor
are more regular compared to previously investigated Grignard
reagents formed from Mg turnings in the same reactor setup
and do not show hotspot formation, which can mainly be
attributed to the Zn organometallics’ lower reactivity. Table 1
summarizes the zinc organometallic reagent synthesis inves-
tigated detailing halide concentrations employed, maximum
flow rate possible for full conversion, temperature setting of the
thermostat bath, and yields of organozinc reagents achieved
(determined by manual titration).
Figure 3. Laboratory setup for the synthesis and Negishi coupling of
benzyl zinc bromide and 1-bromo-4-ethylbenzene.
It was found that all reagents could be formed in good to
very good yields, e.g., comparable to achievements reached by
Huck et al.15 with very limited coupling of starting material and
product (as investigated via GC analysis of quenched samples).
This is particularly noteworthy for the benzyl and allyl zinc
bromides since these tend to show a tendency to large amounts
of coupling. In the experiments, the reactor was freshly filled
with about 105 g of Zn granules each time which, using 0.5
molar starting solutions, corresponded to a molar excess of the
metal compound of about 250 and an accessible free volume of
about 13 mL within the reactor. Particularly for the case of
setup as used for the 2-step synthesis of benzyl zinc bromide
formation followed by the Negishi coupling of the formed
organozinc halide with 1-bromo-4-ethylbenzene. It must be
noted here that a flow rate of 4 mL/min could be used in the
synthesis of benzyl zinc bromide alone. However, in the
coupled reaction steps, the flow rate had to be reduced (1 mL/
min), so a sufficient residence time for the second step over the
catalyst bed could be achieved.
Table 1. Reaction Conditions and Yields Obtained for RZnX Reagents
concentration starting material
[mol/L]
flow rate
[mL/min]
residence time
[min]
temperature TT
concentration (achieved in flow)
yield
[%]
Zn reagent
[°C]
[mol/L]
Et Zn Br
acetate
0.5
filling, 8; run, 2
6.5
40
0.42−0.44
84−88
benzyl Zn Br
allyl Zn Br
0.5
0.5
filling, 8; run, 4
filling, 4; run, 1
3.25
13
40
10
0.45−0.46
0.41−0.42
90−92
82−84
C
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 4 shows the temperature profiles obtained in the
coupling of benzyl zinc bromide with 1-bromo-4-ethylbenzene,
In order to achieve a scale-up of the two-step synthesis as
described above, a pilot-scale reactor setup has been designed
and constructed, and initial tests have been performed for the
Grignard reagent formation. Figure 5 shows a pilot-scale
Figure 4. Temperature profile along the reactor for first step of benzyl
zinc bromide formation and Negishi coupling with 1-bromo-4-
ethylbenzene (T1−T4, temperatures along the reactor in first step;
T5, temperature observed directly after mixing via a T-piece; Kat T1−
Kat T4, temperatures along the 2nd step metal catalyst filled reactor).
showing the initial establishment of steady-state conditions for
the first step (organozinc reagent formation) followed by the
second metal-catalyzed Negishi coupling step.
Table 2 summarizes the results obtained in the three two-
step syntheses performed. It is noteworthy here that in all cases
an excess of organozinc reagent was used as following, e.g., the
example by Huck et al.15 Furthermore, it was found that, due
to the organozinc species’ nature, a peristaltic pump rather
than a syringe pump had to be used for dosing of the reactive
intermediate.
It was found that the best results could be achieved for the
Negishi coupling employing benzyl zinc bromide as well as for
the Saytzeff reaction using allyl zinc bromide with yields
around 90%, comparable to or even exceeding yields reported
in the literature.16,34 The Reformatsky synthesis reacting ethyl
zinc bromoacetate with 3-methoxyacetophenone posed more
problems, and the excellent yields achieved by Berton et al. on
a smaller-scale continuous process19 could not be reproduced
in the setup described here. This is associated with an
incomplete conversion of the carbonyl compound (3-
methoxyacetophenone); a maximum of 43% was converted
during the experiments even when the residence time was
tripled by using an additional stainless-steel coil as residence
time loop. The low yield observed for this particular synthesis
needs to be further investigated and is potentially due to gas
formation through local hotspots within the reactor. However,
the general applicability of the two-step synthesis forming a
reactive intermediate in situ and converting it in a follow-up
step could be shown.
Figure 5. Pilot-scale reactor established for the synthesis of
organometallic species to be tested for the scale-up of organozinc
species investigated here.
reactor used when piloting the formation of a Grignard reagent
at a throughput of 2.7 L/h. This setup is ready to also be tested
for the synthesis of organozinc reactive intermediates and can
be extended to include a second step with pilot-scale
throughputs.
Table 2. Reaction Conditions and Yields Obtained for RZnX Reagent Formation and Subsequent Couplings
total flow
yield,
2nd
carbonyl
carb.
rate
residence
time 2nd
conversion of
Zn
compound 2nd compound
[mL/min],
1:1 ratio
temp. [°C] starting materials
step
[%]
reagent
step
[mol/L]
0.15
reaction type
product
step [min] 1st/2nd step [%] 1st/2nd step
Et Zn Br 3-methoxy-
acetophenone
Reformatsky ethyl-3-hydroxy-3-(3-
methoxyphenyl)
4
12
40/60
100/33−47
38
butanoate
p-ethyldiphenylmethane
a
benzyl
Zn Br
allyl Zn
Br
1-bromo-4-
ethylbenzene
cyclohexanone
0.3
Negishi
2
2
6
40/60
10/18
100/100
100/100
92
88
0.35
Saytzeff
1-allyl cyclohexanol
24
a
Negishi coupling performed in a reactor similar in construction to the one for the first step; catalyst used was SiliaCat DPP-Pd with a 0.3 mmol/g
catalyst loading.
D
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
bromide (Acros Organics, 99%), 0.5 M allyl zinc bromide in
THF (Alpha Aesar), benzyl bromide (Thermo Fischer Kandel,
99%, stabilized), 0.5 M benzyl zinc bromide in THF (Sigma-
Aldrich), 1-bromo-2-chloroethane (Merck, ≥98%), 1-butanol
(Sigma-Aldrich, 99.7%), chlorotrimethylsilane (Sigma-Aldrich,
≥98%), cyclohexanone (Alfa Aesar, ≥99%), ethylbromoacetate
(Merck, >98%), iodine (Fluka, >99.5%), 3-methoxyacetophe-
none, (Acros Organics, 98%), SiliaCat DPP-Pd heterogeneous
catalyst (Silicycle Inc.), THF (T H Geyer, 99.9%, water-free),
and Zn turnings (Sigma-Aldrich, −30mesh).
Postnova syringe pumps (PN 1610 syringe dosing system)
were used for the first step of the RZnX formation and to
pump the carbonyl compound in the second step. For dosing
the freshly prepared organozinc reagent, a peristaltic pump
Masterflex L/S using PTFE tubing from Cole Palmer was used.
The reactors were tempered using thermostatic baths from
EXPERIMENTAL SECTION
■
Typical Flow Procedure for Zn Organometallics
Synthesis. The reactor was charged with about 105 g of
zinc turnings while the jogging motor at the bottom of the
reactor was already started (6 V). Argon was used for
inertization, and four thermocouples were placed along the
reactor’s Zn bed for temperature monitoring. The activation of
the zinc turnings was carried out by pumping a solution of 0.6
mol/L chlorotrimethylsilane and 0.24 mol/L 1-bromo-2-
chloroethane in dry THF with a flow rate of 2 mL/min and
a thermostat bath temperature of 40 °C through the zinc bed
for about 10 min. After activation was done, the reactor was
tempered using a thermostat bath temperature suitable for the
desired RZnX synthesis and flushed with the halide starting
material in dry THF at an elevated flow rate for up to 10 min,
resulting in a temperature increase as a hint of the initiation of
the exothermic reaction. The flow rate was then reduced to
keep the temperature below the boiling point of the solvent,
and the product solution was collected. Samples of the product
solution were taken for gas chromatography analysis and
titration to determine the organozinc halide concentration.
Specific reaction conditions for the investigated organozinc
halide reagents are shown in Table 1.
Typical Flow Procedure for Zn Organometallics
Coupling Reactions. The organozinc halide, which was
produced in the previous reaction step, was pumped from its
collection vessel through an in-line filter with ceramic fibers
and to the subsequent synthesis step by means of a peristaltic
pump. In a T-piece, the organozinc halide was contacted with
the electrophilic substrate (in dry THF), and the reaction
mixture was further fed into the reactor (PTFE coil or
stainless-steel reactor as in the first step). Temperature control
was maintained using a thermostat bath, and the outlet
solution leaving the reactor was collected in an ammonium
chloride solution cooled with ice. Samples of the outlet
solution were analyzed by means of gas chromatography.
Specific reaction conditions are shown in Table 2. In the case
of p-ethyldiphenylmethane synthesis, the reactor contained 5 g
of the catalyst SiliaCat DPP-Pd mixed with glass beads (used
for volume reduction).
Processing of Collected Samples. After quenching with
ammonium chloride solution in the previous synthesis step,
precipitated salts were filtered off, and the organic phase was
extracted in the separating funnel by means of a suitable
solvent (1-allylcyclohexanol: diethyl ether, p-ethyl diphenyl-
methane and ethyl 3-hydroxy-3-(3-methoxyphenyl)-butanoate:
ethyl acetate). The organic phase was dried using sodium
sulfate; the sodium sulfate was filtered off, and the organic
phase was concentrated on the rotary evaporator. The
concentrated product was again transferred to a separating
funnel with solvent. A white precipitate was formed that could
be transferred to the aqueous phase by adding some distilled
water. Acidification of the aqueous phase by a few drops of
10% hydrochloric acid dissolved the precipitate so that the
aqueous and organic phases could be separated in the
separating funnel. The ether phase was again concentrated
on the rotary evaporator to obtain the crude product (yield
given as determined by GC analysis).16,19,34
̈
Julabo (F31-C) and Peter Huber Kaltemaschinenbau (CC-
405, CC-3). The temperatures were measured using miniature
thermocouples (type K from RS Components), recorded by a
data logger (Type Expert Key 200L, from Delphin
Technology), and for evaluation transferred to a notebook.
Titration of the Zn reagent was done following the method of
Berton et al.19 and an adapted back-titration as described by
the Knochel group.30 Analysis for the RZnX compounds was
done via quenching the samples with 1-butanol and analyzing
via calibrated GC measurements, after workup crude products
from the investigated second steps were also analyzed via GC.
A Varian GC 3900 system with Varian 8400 GC-autosampler
was used for GC analysis.
CONCLUSIONS
■
In conclusion, a continuously operating two-step laboratory
setup for organozinc reagent formation and immediate
consumption in subsequent catalyzed and noncatalyzed
coupling reactions was established and tested. It was found
that mechanical activation within the reactor alone did not
create sufficient Zn surface activation and only yielded
incomplete halide conversion, so chemical Zn activation had
to be employed. However, for each organozinc species, full
halide conversion in one single reactor passage could be
achieved in good to very good yields with very little coupling
product formation observed via quenching and GC analysis of
samples. The reactor was operated at a large Zn excess (about
250-fold), and consumed Zn could be refilled via a metal
replenishing unit attached to the reactor. Reformatsky, Saytzeff,
and Negishi coupling reactions were performed using in situ
generated Zn organometallic species. For the examples of the
Saytzeff reaction and Negishi coupling, good yields could be
obtained. In the case of the Reformatsky synthesis, full
conversion of the employed carbonyl compound could not be
achieved, and gas formation was observed during the reaction.
This synthesis needs to be re-evaluated since a similar process
by Berton et al. on a smaller throughput scale resulted in
excellent yields.19 A scaled-up modular setup to pilot-scale
throughputs (scale-up factor of 15) employing the same
principle of using a large molar excess of the Zn within the
reactor and also including a truly continuous Zn replenishing
unit was designed and manufactured and is ready to be tested
for the formation of organozinc species.
MATERIALS AND METHODS
■
All solvents and reagents were purchased from commercial
suppliers and were used without further purification: allyl
E
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
14, 325−344. (d) Baba, A.; Yasuda, M.; Nishimoto, Y. Zinc Enolates:
The Reformatsky and Blaise Reactions. In Comprehensive Oragnic
Synthesis II, Vol. 2; Elsevier, 2014.
(10) Hirose, T.; Kodama, K. In Comprehensive Organic Synthesis II
Vol. 1; Elsevier, 2014.
AUTHOR INFORMATION
Corresponding Author
Gabriele Menges-Flanagan − Fraunhofer IMM, 55129
Mainz, Germany; orcid.org/0000-0002-8769-7247;
Phone: +49 6131 990425; Email: gabriele.menges-
flanagan@imm.fraunhofer.de; Fax: +49 6131 990205
■
̈
(11) Saytzeff, A. Aus dem technischen Laboratorium der Universitat
Kasan. Z. f. Chemie 13: Neue Folge 6 1870, 104−105.
́
́
́
́
(12) Blaise, E. E. Nouvelles reactions des derives organo-metalliques.
Compt. Rend. 1901, 132, 478−480.
(13) Baba, A.; Yasuda, M.; Nishimioto, Y. In Comprehensive Organic
Synthesis II Vol. 2; Elsevier, 2014.
Authors
Eva Deitmann − Fraunhofer IMM, 55129 Mainz, Germany;
Fachhochschule Münster, 48565 Steinfurt, Germany
(14) (aa) Negishi, E.; King, A. O.; Okukado, O. Selective carbon-
carbon bond formation via transition metal catalysis. 3. A highly
selective synthesis of unsymmetrical biaryls and diarylmethanes by the
nickel- or palladium-catalyzed reaction of aryl- and benzylzinc
derivatives with aryl halides. J. Org. Chem. 1977, 42, 1821−1823.
(b) Negishi, E.; Valente, L. F.; Kobayashi, M. J. Palladium-catalyzed
cross-coupling reaction of homoallylic or homopropargylic organo-
zincs with alkenyl halides as a new selective route to 1,5-dienes and
1,5-enynes. J. Am. Chem. Soc. 1980, 102, 3298−3299. (c) Haas, D.;
Hammann, J. N.; Greiner, R.; Knochel, P. Recent Developments in
Negishi Cross-Coupling Reactions. ACS Catal. 2016, 6, 1540−1552.
(d) Jana, R.; Pathak, T. P.; Sigman, M. S. Advances in Transition
Metal (Pd,Ni,Fe)-Catalyzed Cross-Coupling Reactions Using Alkyl-
organometallics as Reaction Partners. Chem. Rev. 2011, 111, 1417−
̈
Lars Gossl − Fraunhofer IMM, 55129 Mainz, Germany;
Hochschule Darmstadt, 64295 Darmstadt, Germany
Christian Hofmann − Fraunhofer IMM, 55129 Mainz,
Germany
̈
Patrick Lob − Fraunhofer IMM, 55129 Mainz, Germany
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.0c00399
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors would like to acknowledge the Fraunhofer IMM
for internal funding.
■
́
1492. (e) Palao, E.; Lopez, E.; Torres-Moya, I.; de la Hoz, A.; Díaz-
́
́
Ortiz, A.; Alcazar, J. Formation of quaternary carbons through cobalt-
catalyzed C(sp3)−C(sp3) Negishi cross-coupling. Chem. Commun.
2020, 56, 8210−8213. (f) Weng, Y.; Zhang, C.; Tang, Z.; Shrestha,
M.; Huang, W.; Qu, J.; Chen, Y. Nickel-catalyzed allylic carbonylative
coupling of alkyl zinc reagents with tert-butyl isocyanide. Nat.
Commun. 2020, 11, 392. (g) Acemoglu, M.; Boenziger, M.; Krell, C.
M.; Marterer, W. Experiences with Negishi Couplings on Technical Scale
in Early Development in Transition Metal-Catalyzed Couplings in Process
Chemistry: Case Studies from the Pharmaceutical Industry; Wiley, 2013.
REFERENCES
■
(1) Frankland, E. Ueber die Isolierung der organischen Radikale.
Liebigs Ann. Chem. 1849, 71, 171−213.
(2) (a) Grignard, V. Sur quelques nouvelles combinaisons
́
̀
̀
̀
organometalliques du magnesium et leur application a des syntheses
́
d’alcools et d’hydrocarbures. In CR Hebd. Seances Acad. Sci., Ser. C.
130; 1900; pp 1322−1324. (Nobel Prize for Chemistry, 1912).
(b) Silverman, G. S.; Rakita, P. E. Handbook of Grignard Reagents;
CRC Press: Boca Raton, FL, 1996. (c) Seyferth, D. The Grignard
Reagents. Organometallics 2009, 28, 1598−1605. (d) Garst, J. F.;
Soriaga, M. P. Grignard reagent formation. Coord. Chem. Rev. 2004,
248, 623−652.
́
(15) Huck, L.; Berton, M.; de la Hoz, A.; Díaz-Ortiz, A.; Alcazar, J.
Reformatsky and Blaise reactions in flow as a tool for drug discovery.
One pot diversity oriented synthesis of valuable intermediates and
heterocycles. Green Chem. 2017, 19, 1420−1424.
̌
́
(16) Alonso, N.; Miller, L. Z.; Munoz, J. de M.; Alcazar, J.;
McQuade, D. T. Continuous Synthesis of Organozinc Halides
Coupled to Negishi Reactions. Adv. Synth. Catal. 2014, 356, 3737−
3741.
(3) (a) Knochel, P.; Singer, R. D. Preparation and reactions of
polyfunctional organozinc reagents in organic synthesis. Chem. Rev.
1993, 93, 2117−2118. (b) Knochel, P.; Leuser, H.; Gong, L.-Z.;
Perrone, S.; Kneisel, F. F. In Organometallics; Wiley, 2005.
(c) Knochel, P.; Leuser, H.; Gong, L.-Z.; Perrone, S.; Kneisel, F. F.
In The chemistry of organozinc compounds; Wiley, 2006. (d) Lutter, F.
H.; Grokenberger, L.; Spieß, P.; Hammann, J. M.; Karaghiosoff, K.;
Knochel, P. Cobalt-Catalyzed Cross-Coupling of Functionalized
Alkylzinc Reagents with (Hetero)Aryl Halides. Angew. Chem., Int.
Ed. 2020, 59, 5546−5550.
́
(17) Abdiaj, I.; Horn, C. R.; Alcazar, J. Scalability of Visible-Light-
Induced Nickel Negishi Reactions: A Combination of Flow
Photochemistry, Use of Solid Reagents, and In-Line NMR
Monitoring. J. Org. Chem. 2019, 84 (8), 4748−4753.
̌
(18) Egle, B.; Munoz, J. de M.; Alonso, N.; De Borggraeve, W. M.;
́
de la Hoz, A.; Díaz-Ortiz, A.; Alcazar, J. First Example of Alkyl−Aryl
Negishi Cross-Coupling in Flow: Mild, Efficient and Clean
Introduction of Functionalized Alkyl Groups. J. Flow Chem. 2015, 4
(1), 22−25.
(4) Baba, A.; Yasuda, M.; Nishimioto, Y. In Comprehensive Organic
Synthesis II Vol. 2; Elsevier, 2014.
́
(19) Berton, M.; Huck, L.; Alcazar, J. On-demand synthesis of
(5) Krasovskiy, A.; Malakhov, V.; Gavryushin, A.; Knochel, P.
Efficient synthesis of functionalized organozinc compounds by the
direct insertion of zinc into organic iodides and bromides. Angew.
Chem., Int. Ed. 2006, 45, 6040−6044.
(6) Knochel, P. In Main-Group Metal Organometallics in Organic
Synthesis; Elsevier, 1995.
organozinc halides under continuous flow conditions. Nat. Protoc.
2018, 13 (1), 324−334.
(20) Lau, S.-H.; Bourne, S. L.; Martin, B.; Schenkel, B.; Penn, G.;
Ley, S. V. Synthesis of a Precursor to Sacubitril Using Enabling
Technologies. Org. Lett. 2015, 17, 5436−5439.
(21) Hafner, A.; Ley, S. V. Generation of Reactive Ketenes under
Flow Conditions through Zinc-Mediated Dehalogenation. Synlett
2015, 26, 1470−1474.
(7) Benischke, A. D.; Ellwart, M.; Becker, M. R.; Knochel, P.
Polyfunctional Zinc and Magnesium Organometallics for Organic
Synthesis: Some Perspectives. Synthesis 2016, 48, 1101−1107.
(8) Li, J.; Knochel, P. Cobalt-Catalyzed Cross-Couplings between
Alkenyl Acetates and Aryl or Alkenyl Zinc Pivalates. Angew. Chem.,
Int. Ed. 2018, 57, 11436−11440.
(22) Degennaro, L.; Carlucci, C.; De Angelis, S.; Luisi, R. Flow
Technology for Organometallic-Mediated Synthesis. J. Flow Chem.
2016, 6, 136−166.
(23) Zhao, T.; Micouin, L.; Piccardi, R. Synthesis of Organometallic
Compounds in Flow. Helv. Chim. Acta 2019, 102, No. e1900172.
(24) Colella, M.; Nagaki, A.; Luisi, R. Flow Technology for the
Genesis and Use of (Highly) Reactive Organometallic Reagents.
Chem. - Eur. J. 2020, 26, 19−32.
(9) (a) Reformatsky, S. Neue Synthese zweiatomiger einbasischer
̈
Sauren aus den Ketonen. Ber. Dtsch. Chem. Ges. 1887, 20 (1), 1210−
1211. (b) Rathke, M. W. The Reformatsky Reaction. In Organic
Reactions; Wiley, 2004. (c) Pellissier, H. Recent developments in the
asymmetric Reformatsky-type reaction. Beilstein J. Org. Chem. 2018,
F
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
pubs.acs.org/OPRD
Article
̈
(25) Brown, D. G.; Bostrom, J. Analysis of Past and Present
Synthetic Methodologies on Medicinal Chemistry: Where Have All
the New Reactions Gone? J. Med. Chem. 2016, 59, 4443−4458.
(26) Roughley, S. D.; Jordan, A. M. The Medicinal Chemist’s
Toolbox: An Analysis of Reactions Used in the Pursuit of Drug
Candidates. J. Med. Chem. 2011, 54, 3451−3479.
(27) Campos, K. R.; Coleman, P. J.; Alvarez, J. C.; Dreher, S. D.;
Garbaccio, R. M.; Terrett, N. K.; Tillyer, R. D.; Truppo, M. D.;
Parmee, E. R. The importance of synthetic chemistry in the
pharmaceutical industry. Science 2019, 363 (6424), eaat0805.
̈
(28) Menges-Flanagan, G.; Deitmann, E.; Gossl, L.; Hofmann, C.;
̈
Lob, P. Scalable continuous synthesis of Grignard reagents from in
situ activated magnesium metal. Org. Process Res. Dev. 2020, 24, 315−
321.
̈
(29) Lob, P. New Microreactor Designs for Practical Applications
Realized by Additive Manufacturing. In Flow Chemistry: Integrated
Approaches for Practical Applications; Green Chemistry Series No. 62;
Luis, S. V., Garcia-Verdugo, E., Eds.; Royal Society of Chemistry:
Cambridge, 2020; Chapter 12, pp 388−415.
(30) Hofmann, C.; Menges-Flanagan, G. Kontinuierliches Verfahren
zur Herstellung von Grignard-Addukten und Vorrichtung zu dessen
̈
Durchfuhrung. DE 10 2016 206 211 A1, 13; April, 2016.
(31) (a) Piazza, C. New methods for the synthesis of organozinc and
organocopper reagents. Ph.D. dissertation, Ludwig-Maximilians-
̈
̈
Universitat, Munchen, 2002. (b) Dunet, G. Preparation and reactions
of allylic zinc reagents and transition metal catalyzed cross coupling
̈
̈
reactions. Ph.D. dissertation, Ludwig-Maximilians-Universitat, Mun-
chen, 2007.
(32) Deitmann, E. Evaluierung eines kontinuierlichen Verfahrens zur
Synthese und Umsetzung von Organozinkhalogeniden. Master thesis,
̈
FH Munster Fraunhofer IMM, Mainz, 2019.
(33) Knochel, P.; Rozema, M. J.; Tucker, C. E.; Retherford, C.;
Furlong, M.; AchyuthaRao, S. The chemistry of polyfunctional
organozinc and copper reagents. Pure Appl. Chem. 1992, 64, 361−
369.
(34) Ranu, B. C.; Majee, A.; Das, A. R. Facile and efficient synthesis
of homoallylic alcohols using allyl bromide and commercial zinc dust.
Tetrahedron Lett. 1995, 36, 4885−4888.
G
https://dx.doi.org/10.1021/acs.oprd.0c00399
Org. Process Res. Dev. XXXX, XXX, XXX−XXX