Synthesis of phenylboronic acids in continuous flow by means of a multijet oscillating disc reactor system operating at cryogenic temperatures

Article abstract of DOI:10.1021/op3000493

A multijet oscillating disk (MJOD) millireactor system suitable for operating at cryogenic temperatures has been developed, assembled, and investigated. This new reactor system (cryoMJOD) was realized with the purpose to prepare various phenylboronic acids in a continuous two (three)-step telescoped synthetic process at temperatures in the interval -50 to -75 °C. In this process, n-butyllithium was reacted with a phenylbromide to provide the corresponding phenyllithium derivative, whereupon a borate was added under the formation of target product phenylboronic acid in good selectivity and medium-to-good yield (50-75%). These results were competitive with results previously revealed in the literature. The residence times of the telescoped two-step process were considerably shorter compared to those of batch mode operations for the identical syntheses. The flow process was optimized by means of statistical experimental design and multivariate regression, upon which the process was utilized for the production of a series of phenylboronic acid derivatives, all in medium-to-good yield. One of the substrates, 4-methoxyphenylboronic acid, was submitted for throughput improvements, resulting in a process with capability to produce the product phenylboronic acid in a quantity of 2.0 kg × day^-1.

Full text of DOI:10.1021/op3000493

Article
pubs.acs.org/OPRD
Synthesis of Phenylboronic Acids in Continuous Flow by Means of a
Multijet Oscillating Disc Reactor System Operating at Cryogenic
Temperatures
Dagfinn Sleveland† and Hans-Rene Bjørsvik*,†,‡
́
†Department of Chemistry, University of Bergen, Alleg
́
aten 41, N-5007 Bergen, Norway
‡Fluens Synthesis, Thormøhlensgate 55, N-5008 Bergen, Norway
S
* Supporting Information
ABSTRACT: A multijet oscillating disk (MJOD) millireactor system suitable for operating at cryogenic temperatures has been
developed, assembled, and investigated. This new reactor system (cryoMJOD) was realized with the purpose to prepare various
phenylboronic acids in a continuous two (three)-step telescoped synthetic process at temperatures in the interval −50 to −75 °C.
In this process, n-butyllithium was reacted with a phenylbromide to provide the corresponding phenyllithium derivative,
whereupon a borate was added under the formation of target product phenylboronic acid in good selectivity and medium-to-
good yield (50−75%). These results were competitive with results previously revealed in the literature. The residence times of
the telescoped two-step process were considerably shorter compared to those of batch mode operations for the identical
syntheses. The flow process was optimized by means of statistical experimental design and multivariate regression, upon which
the process was utilized for the production of a series of phenylboronic acid derivatives, all in medium-to-good yield. One of the
substrates, 4-methoxyphenylboronic acid, was submitted for throughput improvements, resulting in a process with capability to
produce the product phenylboronic acid in a quantity of 2.0 kg × day−1.
standard solution-phase reaction conditions in a temperature
range of −10 to +150 °C, and multiphase (gas−solid−liquid)
conditions. A recently disclosed paper from our laboratory
demonstrates the MJOD flow reactor as a platform for
conducting gas−liquid phase, namely molecular oxygen as the
terminal oxidant in an organocatalyzed epoxidation of olefins,10
for catalytic processes including processes that require inert gas
(Ar, N2) conditions, and for reactions that involve slurries or
precipitates of solids during the course of the reaction.
Another important application area for flow chemistry lately
visited by three different research groups involves reactions and
processes that require cryogenic temperatures in batch. Werner
and collaborators11 have investigated the preparation of
phenylboronic acids in flow using micromixers combined
with tubular reactors. Ley and collaborators12 reported a study
involving an in-house developed microreactor system dedicated
for low-temperature reactions. By means of this system, a series
of pinacol boronate esters and 4-fluorophenylboronic acid were
prepared at cryogenic temperatures. Buchwald and collabo-
rators13 also disclosed results that involved flow synthesis of
phenylboronic acids at room temperature or above. A flow
process setup was used, which was integrated with one more
step that permitted production of various substituted biphenyls
via the Suzuki reaction. Herein we will reveal results from a
project that encompassed design and development of a
cryoMJOD flow reactor and results from the development of
a highly effective two-step telescoped process for the synthesis
INTRODUCTION
■
Continuous flow chemistry in terms of microreactor technology
has during the past two decades been implemented both in the
academic laboratory as well as within industrial R&D and
production of fine- and pharmaceutical chemicals.1 A number
of flow reactor suppliers have been established that offer a
variety of different flow reactor systems and equipment.2 The
major part of the suppliers provides microreactor technology
dedicated to flow chemistry in solution phase only. However,
other technologies have also been developed and adopted for
processing slurries using a flow chemistry platform. Recently,
Ley and collaborators3 investigated the potential for processing
slurries in a reaction that produced a nonsoluble salt during the
course of the reaction. Such operations are generally impossible
to conduct in a microreactor due to clogging of the micro-sized
flow channel. For this purpose, a “Coflore” agitating cell reactor
(ACR)4 that roughly operates according to the principles of a
continuously stirred tank reactor (CSTR)5 was utilized. In our
laboratory we have also demonstrated the use of the multijet
oscillating disk (MJOD) reactor system (Fluens Synthesis)6 for
heterogeneous reactions, namely by successfully conducting the
Hoffmann rearrangement reaction7 that involved a heteroge-
neous reaction system.8 Hulshof and collaborators9 have
demonstrated the use of the continuous flow microwave
reactor system “FlowSynth” by Milestone srl2c to conduct two
processes in heterogeneous phase, namely a biocatalyzed
esterification of (R,S)-1-phenylethanol with vinyl acetate and
esterification of (S)-pyroglutamic acid with n-decanol.
Special Issue: Continuous Processes 2012
The MJOD reactor technology previously reported from our
laboratory,8 was designed, manufactured, and developed as a
general reactor platform for conducting flow reactions under
Received: February 25, 2012
Published: April 16, 2012
© 2012 American Chemical Society
1121
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
of a reagent of paramount importance in organic synthesis,
namely phenylboronic acids.
The importance of this reagent was confirmed by the award
of the Nobel Prize14 in chemistry in 2010 to Akira Suzuki.
Suzuki and collaborators published two papers15 in 1979 on the
discovery of a new versatile and practical palladium-catalyzed
reaction, today known as the Suzuki reaction.16 This reaction
involves an organoboron compound as a coupling partner with
a vinyl or an aryl halide in a palladium-catalyzed cross coupling.
The reaction was shown to take place by base activation of the
organoboron reagents to produce the corresponding boronate
intermediates, which facilitate a transmetalation reaction that
implies an organic group transfer from boron to the palladium
catalyst whereby two organic moieties are coupled under
release from palladium. The reaction was also extended to
include coupling reactions with alkyl groups and arylboronic
acids, which can also take part in the palladium-catalyzed cross-
coupling reaction. More than 30 years of research, innovation,
and development have resulted in a copious body of total
syntheses that lead to unsymmetrical substituted biphenyls,
some of which have great benefit to society and life. Examples
of such products are the fungicide, Boscalid,17 and the
pharmaceutical chemical valsartan18 that is an AT1 angiogenesis
receptor that dilates blood vessels and reduces the blood
pressure. Furthermore, compounds bearing the boronic acid
moiety have also been investigated as an antineoplastic19
protease inhibitor,20 and a number of other medicinal
remedies.21−23
Figure 1. Three-dimensional drawing of the MJOD millireactor (left)
shows the four various sections, (1) the oscillator section, (2) the
output section, (3) the reactor body, and (4) the input section. A
fraction of the reactor body (3) and the input section (4) are shown in
enlarged view to the right in the figure. The reactor body consists of
the heat exchanger chamber that surrounds the reactor tube where the
multijet oscillating disk (MJOD) assembly is inserted. The MJOD unit
comprises the perforated discs (5) that are fixed on the oscillator
piston shaft with “the spacers” (6). This MJOD unit is connected to
the (1) oscillator unit that transfers an oscillating vertical (up−down)
movement of this “numerous-headed” piston.
by linear translation of the cam assembly to a predefined
position (i.e., the distance to the motor shaft). Frequencies in
the range of f = 1−10 Hz and amplitudes in the range of A =
0.5−15 mm can be achieved by adjusting the motor speed and
the cam assembly. Various types of feeding pumps can be used
to feed the substrate, reagents, catalysts, and solvents into the
MJOD reactor through the feeding channels of the input
section, which is located at the bottom of the vertically placed
reactor body. The substrate and various reagents are
continuously pumped into the reactor tube via the input
section, which implies that the reaction mixture is pushed
upwards in the reactor tube practically without backmixing.24
From the input section, the reaction mixture passes through the
jets of the MJOD discs from one reaction cavity upwards to the
next one. Together with the up−down movements of the
MJOD discs, the tightening of the cross section (that is the tiny
jets in one of the MJOD discs) creates vortices and thus an
extremely good mixing of the reacting components. The
postreaction mixture (that it is the mixture that comes out from
the output section) is collected in suitable batches for workup.
The temperature control for the MJOD reactor body was
performed by means of a thermostat and a circulator pump.
Previous studies in our laboratory have shown that the MJOD
reactor system possesses an outstanding mass and heat transfer
ability for the system, which in general results in a significantly
increased reaction rate.
Modification of the MJOD Flow Reactor for Utilization
at Cryogenic Temperatures. The parts of the reactor section
were manufactured identical to those of the general MJOD flow
reactor model. The MJOD unit was constructed by four-jet (djet
= 1.25 mm) discs; in total 133 pieces (pcs) were used, which
correspond to 132 interdisc cavities, each of a volume of Vcavity
≈ 0.65 mL. The four-jet disks that cover the entire cross section
of the tubular reactor (D = 10 mm) were manufactured with
Teflon for the general MJOD model and were replaced by discs
manufactured with Polyzene for the cryogenic model. Three
reagent supply (inlet) lines of the input section were used.
These were connected to precision piston- or syringe-feeding
pumps. One input line at the output section of the MJOD
The goal of the project disclosed herein was two-fold, namely
(1) to design and realize a flow reactor suitable for performing
continuous flow processes at cryogenic temperatures based on
the multijet oscillating disk (MJOD) reactor system that
previously was designed and developed in our laboratories8 and
(2) to utilize this new flow chemistry equipment as a reactor
platform for development and synthesis of a series of
phenylboronic acid derivatives.
METHODS AND RESULTS
■
The MJOD Flow Reactor System. Figure 1 displays a
general three-dimensional (3D) drawing of a MJOD reactor
that includes (1) the oscillator section, (2) the output section,
(3) the reactor body, and (4) the input section. In addition, a
number of auxiliary supporting units, such as feeding pumps, a
cryostat with a circulating pump, a precooling unit (heat
exchanger) placed prior to the input section, and reservoirs for
raw materials, solvents, and final product were utilized.
The right-hand side of Figure 1 shows a transparent, bottom-
up view of the input section, including a small fraction of the
reactor zone (reactor body). The MJOD assembly is located in
the center of the reactor tube. The outer shell of the reactor
body forms a ring-shaped room (cavity) that encapsulates the
whole length of the reactor tube through which the cooling
fluid (pure ethanol) was circulated. The whole length of the
cryoMJOD flow reactor body was encapsulated using a
polyethylene foam insulation sleeve with a wall thickness of
lwt = 13 mm.
Due to the advantageous contact surface ratio vs the net
reactor volume, an outstanding heat transfer capacity was
attained. A variable-frequency and variable-amplitude oscillator
was used for the vertical “piston movement” of the MJOD unit.
An electric motor connected to a cam mechanism was used to
power the up−down movement of the MJOD assembly. In
addition, the cam assembly provides control of the amplitude
1122
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
reactor was connected to a pressurized gas tank containing
argon. The argon flow was adjusted to a pressure slightly above
atmospheric pressure.
The cooling fluid, pure ethanol, was cooled and regulated by
a thermostat (temperatures were selected in the range of −50
to −75 °C, as measured in the cooling fluid) and circulated
through the heat exchanger. The cam mechanism and electrical
motor that constitute the oscillator was tuned to provide
oscillations with an amplitude of A = 5 mm with a frequency f
≈ 2 Hz.
Screening, Development and Suboptimization of the
Telescoped Lithiation and Boronation Process. The
syntheses of phenylboronic acid 1b and derivatives comprise
two distinct synthetic steps that are both conducted at
cryogenic temperature. Werner et al.11 have studied the
production of phenylboronic acid in flow using micromixer/
tubular reactors at room temperature and even higher.
Recently, Ley and collaborators12 and Buchwald and
collaborators13 demonstrated the synthesis of phenylboronic
acids in microreactor systems, although only under restricted
production capacity. Concurrently with the recently disclosed
studies,12,13 we were in progress with the design and
development aiming at approaching an MJOD flow reactor
system suitable for cryogenic temperatures. This flow reactor
system was designed in such a way that it was possible to
introduce a reagent at various positions in the flow reactor and
thus at various time intervals during the course of a reaction.
Such an MJOD reactor body enables performing two or more
different reactions in a telescoped manner, which is a
requirement for the process leading to phenylboronic acid
derivatives.
at the entrance (the input section) to the flow reactor. From
this point and throughout the reactor part L1 of the reactor
body, the substrate and the reagent were thoroughly mixed by
the MJOD unit. Then, at the beginning (at the lower point) of
the reactor part L2 of the reactor body, a third input line was
connected. By means of the pump P3, this tube supplied the
borate reagent from reservoir R3. The target product was
received and collected at the output section at the end of (at
the top) the reactor part L2. The two reservoirs R4 and R5
contained anhydrous hexane and anhydrous THF, respectively.
When the reagent reservoirs and substrate reservoirs were
emptied, the feeding from the each solvent reservoir was
connected to the respective feeding pumps.
Each of the two reactor parts, L1 and L2, corresponds to a
certain reaction residence time, which was a function of the
flow rate within the reactor body, and thus the pump rates.
Exploratory Flow Experiments. The initial experiments
using the process setup of Figure 2a were rather discouraging,
because only low yields and selectivity were achieved. Since we
were, at the outset, unfamiliar with operating the MJOD flow
reactor system under cryogenic temperatures and with the
different experimental parameter settings, the existing batch
records were copied too strictly. In practice, initially relatively
long residence times were applied when various reaction
temperatures were explored. Table 1 provides a small selection
of the results in this initial phase of the experimental work
carried out with a cryoMJOD flow reactor as shown in the flow
diagram in Figure 2a.
Based on earlier experience with the MJOD flow reactor8 and
the results of the new exploratory experiments (Table 1), the
reaction time was considerably reduced and reaction temper-
atures were used in accordance with previously published batch
records.
The Batch Record. The three-step synthesis leading to
phenylboronic acid 1a is performed at cryogenic temperatures
under inert (argon) atmosphere. The first step (a) of Scheme 1
Some few different substrates were also used in the screening
study. However, this approach did not appear to have any
significant influence on the outcome. For further exploration
and development of the flow process 1-bromo-4-methoxyben-
zene 2a was selected as substrate. We decided to postpone a
more thorough investigation of the influence of various
functional groups in the substrate and to focus on establishing
and optimizing a flow chemistry record.
Scheme 1
Optimization of the cryoMJOD Flow Process. A simple
statistical experimental design25 aimed at investigating the
influence of two experimental variables, x1 as the reactor
residence time [min] and x2 as the reaction temperature [°C].
The exploratory experiments, using substrate 2a, revealed that
both of these variables (x1, x2) possess a major influence on the
outcome of the reaction. Thus, a full factorial design including
two center-point experiments, 2k + 2 = 22 + 2 = 6 experiments
was generated and conducted with the cryoMJOD flow reactor.
The experimental design (1−6) with corresponding responses
is summarized in Table 2.
consists of a reaction between a bromobenzene a and n-
butyllithium, which provides the corresponding phenyllithium
derivative c. Then in the second, but telescoped step (b), a
boron reagent is introduced, which reacts with the phenyl-
lithium intermediate c that was produced during the course of
reaction step a to provide the target product phenylboronic
acid 1b after an acidic quench, step c, of Scheme 1.
The model matrix [1 x1 x2 x1 × x2] was multivariate
correlated to the response vector by using multiple linear
regression (MLR).26 The estimated model with numerical
values of the regression coefficients is provided in eq 1. Figure 3
shows the isocontour projection of the response surface
(contour map) using eq 1. The product statistics estimated
for the model provided in eq 1 were: R2 = 0.714 and RMSEP =
7.71.
Implementation of the Synthesis on the MJOD Flow
Reactor. A cryoMJOD flow reactor setup based on the
conventional addition sequence was assembled as shown in the
process flowchart of Figure 2a. In this process setup, the
substrate (in reservoir R1) was pumped with pump P1, and the
n-BuLi reagent (diluted in pentane and placed in reservoir R2)
was pumped with pump P2. The two separate tubes from P1
and P2, respectively, were connected to a precooler (C/HE2)
1123
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
Figure 2. Flow diagrams for two setups (a) and (b) for the multijet oscillating disk (MJOD) flow reactor systems. P: pump; R: reservoir, (R4’s
contain anhydrous hexane, R5’s contain anhydrous THF); O: oscillator [A = 0.5−5 mm; f = 0.5−3 Hz], C/HE1: cryostat/heat exchanger, T = −65
°C. C/HE2: precooler (T = −65 °C) for substrate and reagents, each line (tube length) of 500 mm with internal diameter of d = 1.5 mm. MJOD:
multijet oscillating disc reactor [K = 133 discs of d = 9.8 mm; each disc has four jets of d = 1.25 mm, with reactor tube lengths, L1 = 1530 mm, L2 =
960 mm, and L = L1 + L2 = 2490 mm with internal diameter of di = 10 mm.
1. The numbers printed in black are the actual results achieved
at the current experimental settings (entries 1−5 of Table 2).
By means of this contour map, the first step optimization was
attempted. With the goal to further optimize the yield, a
moderate extrapolation was performed. By using the settings x1
= +1.5 and x2 = −2.0 (in coded units), the contour map (model
eq 1) predicts a yield of ∼100%. However, when the actual
y = f(x1, x2)
= 59.66 − 3.75 × x1 − 6.75 × x2 − 12.75 × x1 × x2
(1)
The blue-colored lines and numbers displayed in the contour
map of Figure 3 exhibit the yield predicted by the model of eq
1124
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
a
Table 1. Introductory experiments
residence time [min]
product
entry
reactor platform
substrate
nsubstrate [mmol]
temp. [°C]
step 1
step 2
yield [%]
b
1
2
3
4
5
6
7
8
9
MJOD
MJOD
MJOD
Batch
3a
3a
3a
3a
3a
3a
2a
2a
2a
15.0
15.0
15.0
13.3
14.0
14.0
25.0
42.0
42.0
−50
−60
−70
−78
−70
−70
−78
−70
−65
30
15
45
22
3b
3b
3b
3b
3b
3b
2b
2b
2b
10
b
40
c
15
22
61
c
120
10
240
15
74
c
MJOD
MJOD
Batch
57
c
5.0
110
3.0
9.62
7.4
180
4.3
−
72
d
69
d
MJOD
MJOD
71
d,e
50
a
b
2a = 1-bromo-4-methoxybenzene, 2b = 4-methoxyboronic acid. 3a = 1-bromo-2-methoxybenzene, 3b = 2-methoxyphenylboronic acid. Isolated by
column chromatography. Isolated by crystallization and filtration from mother liquor. Isolated by acidic extraction. The reaction was performed in
one step of length L1 + L2
c
d
e
level outside the region investigated herein. Further attempt to
optimize the process was not performed.
Table 2. Full factorial design with center-point experiments
22 + 2 for investigation of the relation between the reaction
temperature and the residence time of substrate 2a
Improvement of the Throughput. A high reactor
throughput is of paramount importance in chemical produc-
tion. In the current case, a maximized throughput can be
obtained by (1) reducing the reactor residence time, (2)
increasing the concentration of the reactants, or (3) by a
combination of (1) and (2). For the process leading to
phenylboronic acid derivatives, the MJOD flow reactor
residence time was optimized as described in the previous
section and could not be shortened more, at least not for the
substrate 1-bromo-4-methoxybenzene 2a that was utilized
during the optimization study above.
Thus, we turned our efforts towards investigating the
performance of our flow process using various concentrations
of the substrate 2a. For this investigation again the cryoMJOD
flow reactor setup of Figure 2a was utilized.
A series of experiments involving substrate concentrations in
the range of 0.18−1.34 mmol × mL−1 was carried through.
Detailed reaction conditions with corresponding results for this
series of experiments are summarized in Table 3.
a
experimental design
response
b
entry
x1
x2
yield [%]
1
2
3
4
−1
+1
−1
+1
0
−1
−1
+1
+1
0
52
70
64
31
72
69
69
c
5
c
6
0
0
7
+1.5
−2
a
A quantity of 40 mmol of the substrate 2a was used in each of the
experiments. x1 the reactor residence time [min] [−1, 0, +1] = (3.00 +
4.30, 3.75 + 5.37, 4.50 + 6.45) [min]. x2 the reaction temperature [°C]
b
c
[−1, 0, +1] = (−70, −65, −60) [°C]. Percent isolated yield. The
experiment was performed two times, providing a mean yield of 71%.
At the highest concentration (entry 6, Table 3), the
cryoMJOD flow process did not operate properly due to a
high backpressure, which was caused by the high viscosity of
the reaction mixture. The experiments of entries 1−5 of Table
3 provided all good isolated yields (60−75%) of target product
4-methoxyphenylboronic acid 2b. Variations in the outcome
appeared to be due to workup difficulties.
Our effort to improve the throughput of the flow process was
successfully completed, starting out from a capacity of 0.35 kg
× day−1 and approaching a throughput of >2.0 kg × day−1. This
throughput capacity improvement corresponds to an improve-
ment of approximately 6 times the capacity of the initial
process.
Investigation of the Versatility of the Flow Process to
Phenylboronic Acid Derivatives. Next, the versatility of our
two-step telescoped flow synthesis process described in the
previous sections was investigated. Thus, a small selection of
substituted bromobenzenes (Table 4) that were submitted to
our flow process, was made. This selection was based on our
interest for the target products, but these substrates also
represent a span in substitution pattern from nonsubstituted at
entry 1 to the crowded ring of substrate in entry 6. For
comparison purposes, the corresponding batch mode process27
was also carried out for the same series of substrates. Both
series of experiments revealed comparable results, see Table 4.
Figure 3. Response surface plot (isocontour map) describing the
performance (yield) of the reaction as a function of the two variables,
x1, reaction time (reactor residence time), and x2, reaction temper-
ature, were varied.
experiment was conducted in the laboratory, the predicted yield
was not achieved; instead, yield was only 69% (printed in red in
the contour map).
On the basis of this result, one might preliminarily conclude
that the response surface ascends and reaches a maximum in a
yield at approximately 70% and then levels off. However, this
limited screening design included only two of the experimental
variables; thus, some of the other variables that were not
systematically investigated can be of significance and at a fixed
1125
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
Table 3. Conditions and results of reactions testing increased concentrations of substrate 2a
Csubstrate
a
[mmol × mL−1
]
residence time [min]
ratio THF:pentane
production capacity
isolated yield [%]
72
entry
nsubstrate [mmol]
step 1
step 2
step 1
step 2
step 1
step 2
[g × h−1
]
[kg × day−1
]
1
2
3
4
5
40
80
0.18
0.40
0.60
0.86
1.15
1.34
0.17
0.33
0.50
0.67
0.83
1.00
3.75
3.75
3.75
3.75
3.75
3.75
5.37
4.89
4.88
4.55
4.22
4.38
10.1
3.8
9.1
5.7
3.5
2.7
10.8
4.5
14.7
27.0
43.8
72.9
84.0
−
0.35
0.65
1.05
1.75
2.02
−
b
60
c
120
160
200
240
10.0
6.9
64
75
64
−
4.7
d
6
3.3
a
Production based on the isolated yield. The production capacity in the [kg × day−1] unit is calculated on the basis of the observed production
b
c
d
capacity [g × h−1]. Difficult workup The workup procedure was not scaled up properly The experiment failed due to a high viscosity of the
reaction mixture, causing a backpressure that exceeded the pressure of the feeding pump.
Table 4. Series of bromobenzenes (1a−6a) for the
preparation of the corresponding phenylboronic acids (1b−
6b) using the cryoMJOD flow reactor
nylboronic acid 4b,29a,c were reported in the range of 59−91%
and 34−74%, respectively.
The only experiment with a substantial difference between
the cryoMJOD flow reactor and the batch mode is reported in
entry 6 of Table 4. This substrate contains three methoxy
substituents and one methylgroup, which give a very electron-
rich and activated aromatic ring.
The isolated yield obtained for product 6b with the
cryoMJOD reactor (entry 6 in Table 4) is significantly higher
compared to that with the batch mode protocol, corresponding
with 50% relative improvement. This improvement can be due
to the much shorter residence time in the flow reactor. The
flow process we have developed requires only one-sixth of the
residence time as demanded by a typical batch mode protocol
in order to achieve a complete conversion of charged substrate.
An Alternative cryoMJOD Flow Process Record. Li and
collaborators30 have previously demonstrated that if the borate
reagent and the substrate were mixed before the n-BuLi reagent
was added, an increased yield could be achieved in some cases,
while the contrary was observed for some other substrates. For
example, when 1-bromo-4-methoxybenzene 2a was used as
substrate, a substantially reduced yield (26%)31 was achieved
compared to the standard addition method for batch reactions
(96%).31
In order to explore this approach, and thus potentially
simplify the flow reactor setup by changing from setup a to
setup b of Figure 2, our test experiment using the cryoMJOD
reactor resulted in a slightly lower yield (50% isolated yield)
compared to the outcome (72% isolated yield) when the
standard addition method was used (with the setup of Figure
2a). Compared to the batch protocol of Li and collaborators,30
we achieved approximately a 2-fold yield with our modified
flow process, although lower than that of the setup using the
telescoped fashion (Figure 2a). The process performance with
the setup of Figure 2b was in line with the batch record
disclosed by Li and collaborators, but with higher yield. The
conversion of the bromoderivative was in the range 70−75%.
The small variations in isolated yield between the flow and
batch protocols can be justified by product losses during the
experimental workup procedure that involved precipitation of
solid product followed by filtration.
EXPERIMENTAL SECTION
■
General Methods. NMR spectra were recorded on a
Bruker Spectrospin DMX 400 with field strength of 400 MHz
The isolated yields were according to literature,28 exemplify-
ing that our flow synthesis process operates reasonably well
with both electron-donating and -withdrawing substituents in
the substrates.
1
for H nuclei and 100 MHz for the 13C nuclei. Chemical shifts
were calibrated using the solvent peaks. High-resolution mass
spectra were recorded on a JEOL AccuTOF T100LC
spectrometer operated in DART mode.
Starting materials and reagents were purchased commercially
and used without further purification unless indicated
The literature reports a large variation in the isolated yields
of phenylboronic acid derivatives. For example, the isolated
yields for 4-methoxyphenylboronic acid 2b29 and 4-chlorophe-
1126
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
otherwise. Anhydrous THF and hexane were prepared by
distillation over sodium and benzophenone under nitrogen.
General Procedure for Flow Experiments (setup
Figure 2a). Prior to utilization, the reactor system was flushed
with argon. The pumps were connected before the reactor was
cooled to −65 °C. All reservoirs and the reactor were kept
under argon at all times via a manifold. PTFE-tubing (i.d. 1.6
mm) was utilized.
A dried Schlenk flask designated as reservoir 1 (R1) was
charged with 1-bromo-4-methoxybenzene 2a (5.06 mL, 40.0
mmol) and anhydrous THF (190 mL). A second dried Schlenk
flask designated as reservoir 2 (R2) was charged with 2.0 M n-
BuLi in pentane (24.0 mL, 48.0 mmol, 1.2 equiv). A syringe
designated as reservoir 3 (R3) was charged with trimethyl
borate (6.76 mL, 60.0 mmol, 1.5 equiv) and dry THF (13.5
mL).
The tubing leading to the reactor from the three reservoirs,
R1, R2, and R3, were filled with the solutions before the
reaction was started so that they would enter the reactor
immediately as soon as the pumps were started.
The oscillator was started, and the substrate (R1) was
pumped into the reactor at 11.0 mL × min−1 using pump (P3)
drive FMI QBG-MB-Q48532 and pump head Q1-SSY. n-BuLi
in pentane (R2) was pumped into the reactor at 1.34 mL ×
min−1 using pump drive FMI QG20-2-MB-Q48532 and pump
head Q1-CKC-W. The pump head of P2, Q1-CKC-W, was
protected from the outside atmosphere by pumping anhydrous
hexane (mixture of isomers) through the gland port with a
syringe pump (1 mL × min−1) and drained to the waste
container.
Trimethyl borate (R3) was pumped into the reactor at a rate
of 1.13 mL × min−1 by means of a syringe pump. However, the
pump P3 was started after 3.75 min, which corresponds to the
residence time of reaction step 1. At that point of time, the
pumps P1 and P2 had filled the reactor body up to the P3 inlet
point. The postreaction mixture started to leave the output
section after a total residence time of 9.12 min.
The first 60 mL of reaction mixture was thrown away before
120 mL of reaction mixture was collected into a round-bottom
flask containing saturated ammonium chloride solution (120
mL) under stirring. The reaction mixture leaving the reactor
was clear and colorless. The last 60 mL of reaction mixture was
also thrown away. After 17.7 min, the reservoirs R1 and R2
were empty, and the pumps were switched to pumping dry
THF. After 21.5 min, the reservoir R3 was empty.
The pumps and reactor were washed with anhydrous THF
followed by acetone before the reactor was emptied through
one of the bottom inlets, and the reactor was dried overnight
with a stream of N2. The pump head of P2 was disassembled,
washed with water, and dried in an oven at a temperature of
100 °C.
The volume of the collected fraction was measured before
the organic solvents were removed under reduced pressure.
The remaining aqueous phase was acidified with 3 M HCl. The
aqueous phase was then extracted with dichloromethane (3 ×
50 mL), and the combined organic phase was washed with
saturated NaCl solution (1 × 50 mL) and dried over anhydrous
Na2SO4. The solution was filtered and the solvent removed
using a rotary evaporator under reduced pressure, giving a
white semisolid contaminated with a yellow-colored oil.
The crude product was stirred with hexane (mixture of
powder. The yield (in the range 54−77%) was adjusted
according to the volume collected from the output section of
the MJOD reactor. All reactions proceeded with full conversion
of the bromobenzene.
Procedure for Flow Experiment (setup Figure 2b).
Prior to utilization, the reactor system was flushed with argon.
The pumps were connected before the reactor was cooled
down to −65 °C. All reservoirs and the reactor were kept under
argon at all times via a manifold. PTFE-tubing (i.d. 1.6 mm)
was utilized.
A dried Schlenk flask designated as reservoir 1 (R1) was
charged with 1-bromo-4-methoxybenzene 2a (5.31 mL, 42.0
mmol), trimethyl borate (7.09 mL, 63.0 mmol, 1.5 equiv) and
anhydrous THF (200 mL). A second dried Schlenk flask
designated as reservoir 2 (R2) was charged with 2.0 M n-BuLi
in pentane (25.2 mL, 50.4 mmol, 1.2 equiv). The tubing
leading to the reactor from the two reservoirs were filled with
the solutions before the reaction was started so that they would
enter the reactor immediately as the pumps were started. The
oscillator was started and the substrate and borate (R1) was
pumped into the reactor at 11.2 mL × min−1 using pump drive
FMI QBG-MB-Q485 and pump head Q1-SSY. n-BuLi (R2)
was pumped into the reactor at 1.32 mL × min−1 using pump
drive FMI QG20-2-MB-Q485 and pump head Q1-CKC-W.
The pump head of P2, Q1-CKC-W, was protected from the
outside atmosphere by pumping anhydrous hexane (mixture of
isomers) through the gland port with a syringe pump at 1 mL ×
min−1. This solution passed through the pump head and into a
waste container.
The first 60 mL of reaction mixture were thrown away before
120 mL of reaction mixture was collected into a round-bottom
flask containing saturated ammonium chloride (120 mL) under
stirring. The solution coming out of the reactor was clear and
colorless. The last 60 mL of reaction mixture was also thrown
away. After 19.2 min R1 and R2 were empty, and the pumps
were switched to pumping dry THF, giving a residence time of
3.75 min in reaction step one and 5.87 min in step two. The
pumps and reactor were washed with anhydrous THF followed
by acetone before the reactor was emptied of solvent through
one of the bottom inlets, and the reactor was dried overnight
with a stream of N2. The pump head of P2 was disassembled,
washed with water, and dried in an oven at a temperature of
100 °C.
The volume of the collected fraction (114 mL) was measured
before the organic solvents were removed under reduced
pressure. The remaining aqueous phase was acidified with 3 M
HCl. The aqueous phase was then extracted with dichloro-
methane (3 × 50 mL), and the combined organic phase was
washed with a saturated NaCl solution (1 × 50 mL) and dried
over anhydrous Na2SO4. The solution was filtered and the
solvent removed under reduced pressure, giving a white
semisolid contaminated with yellow oil. The solid was stirred
in hexane (mixture of isomers) (3 × 15 mL). The solvent was
then removed on a Buchner filter. The isolated solid product
̈
was air-dried to provide a white powder (1.51 g), which
corresponds to an isolated yield of 50% after adjusting
according to the volume collected from the reactor.
General Procedure for Batch Experiments. A dried
Schlenk flask was charged with 1-bromo-4-methoxybenzene 2a
(1.28 mL, 10.0 mmol) and anhydrous THF (48 mL) under
argon. It was cooled in a dry ice bath (−78 °C) for 5 min
before adding 2.0 M n-BuLi in pentane (6.0 mL, 12.0 mmol, 1.2
equiv) dropwise over 20 min using a syringe pump. After
isomers) (15 mL × 3). The solvent was removed on a Buchner
̈
filter. The filter cake was dried to provide a white-colored
1127
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
1
addition, it was left under stirring for 1 h before adding
trimethyl borate (1.69 mL, 15.0 mmol, 1.5 equiv) as a neat
liquid and stirred for 30 min. The mixture was allowed to warm
to room temperature and stirred for another 1.5 h. The mixture
was quenched in a saturated ammonium chloride solution (30
mL), and the organic solvents were removed under reduced
pressure. The remaining aqueous phase was acidified with 3 M
HCl. The aqueous phase was extracted with dichloromethane
(3 × 25 mL), and the combined organic phase was washed with
saturated NaCl solution (1 × 25 mL) and dried over anhydrous
Na2SO4. The solution was filtered and the solvent removed
under reduced pressure, giving a white semisolid contaminated
with a yellow oil. The crude product was stirred with hexane
(mixture of isomers) (15 mL × 3). The solvent was removed
g). H NMR (400 MHz, deut-DMSO): δ 7.82 (d, 2H, J = 8.0
Hz), 7.39 (d, 2H, J = 8.0 Hz).
2-Methylphenylboronic Acid (5b). [16419-60-6], C7H9BO2,
MW 135.96. 2-Bromo-1-methylbenzene 5a (4.63 mL, 40.0
mmol) was used in the flow experiment. The experiment was
conducted as described above in the general procedure for flow
experiments. A sample (117 mL) of the reaction mixture was
collected from the output section of the MJOD flow reactor.
This sample was worked up to provide a white powder (1.28
g), which corresponds to an isolated yield of 48%. The
corresponding batch experiment, utilizing 5a (1.16 mL, 10.0
1
mmol), provided an isolated yield of 52% (0.71 g). H NMR
(400 MHz, CDCl3): δ 8.20 (d, 1H, J = 7.4 Hz), 7.44 (t, 1H, J =
7.4 Hz), 7.27 (m, 2H), 2.80 (s, 3H).
on a Buchner filter. The filter cake was dried to provide a white-
2,3,4-Trimethoxy-6-methylphenylboronic Acid (6b).
[212573-50-7], C10H15BO5, MW 226.03. 2-Bromo-3,4,5-
trimethoxy-1-methylbenzene 6a (10.550 g, 40.0 mmol) was
used in the flow experiment. The experiment was conducted as
described above in the general procedure for flow experiments.
A sample (114 mL) of the reaction mixture was collected from
the output section of the MJOD flow reactor. This sample was
worked up to provide a white powder (2.30 g), which
corresponds to an isolated yield of 54%. The corresponding
batch experiment, utilizing 6a (2.638 g, 10.0 mmol), provided
an isolated yield of 36% (0.81 g). 1H NMR (400 MHz,
CDCl3): δ 6.54 (s, 1H), 5.85 (s, 1.74H), 3.92 (s, 3H), 3.86 (s,
3H), 3.81 (s, 3H), 2.50 (s, 3H). 13C NMR (100 MHz, CDCl3):
δ 160.0, 155.5, 142.8, 139.7, 111.6, 77.9, 62.5, 61.5, 56.5, 24.2.
HRMS (DART): m/z (MH+) calcd for C10H16BO5 227.10908,
found 227.10925.
Long Time Run Flow Experiment. Prior to the
experiment, the three main pumps (P1, P2, and P3) were
connected to the input sections of the MJOD reactor, and the
reactor was flushed with argon before it was cooled to a
temperature of −65 °C. All the reagent and substrate reservoirs,
the product collector tank, and the reactor tube were kept
under argon throughout the experiment through an argon
distribution manifold. The reservoir R1 (a dried borosilicate
flask [5 L]), was charged with 1-bromo-4-methoxybenzene 2a
(68.5 mL, 542 mmol) and anhydrous THF (2600 mL).
Reservoir R2 (a dried Schlenk flask [500 mL]) was charged
with 2.0 M n-BuLi in pentane (325 mL, 650 mmol, 1.2 equiv).
Reservoir R3 (a dried Schlenk flask [500 mL]) was charged
with trimethyl borate (91.6 mL, 813 mmol, 1.5 equiv) and
anhydrous THF (183 mL). Reservoir R4 (a dried Schlenk flask
[500 mL]) was charged with anhydrous hexane (400 mL,
mixture of isomers). Reservoir R5 (a dried Schlenk flask [500
mL]) was charged with dried THF (400 mL). The
interconnecting tubing from the various reservoirs to the
pumps and further to the input sections of the MJOD reactor
were filled with the various reagent and substrate solutions in
order to prepare for initiating the reaction, allowing entrance to
the reactor at the identical time. The substrate (R1) was
pumped into the reactor at a rate of 11.1 mL × min−1 using a
piston pump (P1).33 n-BuLi in pentane (R2) was pumped into
the reactor at a rate of 1.36 mL × min−1 using a piston pump
(P2). Trimethyl borate (R3) was pumped into the reactor at a
rate of 1.14 mL × min−1 with a piston pump (P3). Barrier
liquid (R4) was pumped through the pump head of P2 by
pump P4 at a rate of 1.6 mL × min−1. Barrier liquid (R5) was
pumped by P5 at a rate of 1.3 mL × min−1. The various
reservoirs and pumps were interconnected to the MJOD
reactor body input sections by means of PTFE tubing (1.6 mm
̈
colored powder.
Phenylboronic Acid (1b). [98-80-6], C6H7BO2, MW 121.93.
Bromobenzene 1a (4.26 mL, 40.0 mmol) was used in the flow
experiment. The experiment was conducted as described above
in the general procedure for flow experiments. A sample (108
mL) of the reaction mixture was collected from the output
section of the MJOD flow reactor. This sample was worked up
to provide a white powder (1.32 g), which corresponds to an
isolated yield of 60%. The corresponding batch experiment,
utilizing 1a (1.06 mL, 10.0 mmol), provided an isolated yield of
65% (0.75 g). 1H NMR (400 MHz, CDCl3): δ 8.24 (dd, 2H, J1
= 8.0 Hz, J2 = 1.4 Hz), 7.59 (tt, 1H, J1 = 7.4 Hz, J2 = 1.4 Hz),
7.50 (m, 2H).
4-Methoxyphenylboronic Acid (2b). [5720-07-0],
C7H9BO3, MW 151.96. 1-Bromo-4-methoxybenzene 2a (5.06
mL, 40.0 mmol) was used in the flow experiment. The
experiment was conducted as described above in the general
procedure for flow experiments. A sample (114 mL) of the
reaction mixture was collected from the output section of the
MJOD flow reactor. This sample was worked up to provide a
white powder (2.07 g), which corresponds to an isolated yield
of 72%. The corresponding batch experiment, utilizing 8 (1.26
mL, 10.0 mmol), provided an isolated yield of 69% (1.05 g). 1H
NMR (400 MHz, CDCl3): δ 8.15 (d, 2H, J = 8.1 Hz), 6.99 (d,
2H, J = 8.1 Hz), 3.87 (s, 3H).
2-Methoxyphenylboronic Acid (3b). [5720-06-9],
C7H9BO3, MW 151.96. 1-Bromo-2-methoxybenzene 3a (5.14
mL, 40.0 mmol) was used in the flow experiment. The
experiment was conducted as described above in the general
procedure for flow experiments. A sample (111 mL) of the
reaction mixture was collected from the output section of the
MJOD flow reactor. This sample was worked up to provide a
white powder (2.17 g), which corresponds to an isolated yield
of 77%. The corresponding batch experiment utilizing 3b (1.28
mL, 10.0 mmol) provided an isolated yield of 86% (1.30 g). 1H
NMR (400 MHz, CDCl3): δ 7.84 (d, 1H, J = 7.3 Hz), 7.44 (t,
1H, J = 7.8 Hz), 7.02 (t, 1H, J = 7.3 Hz), 6.90 (d, 1H, J = 8.3
Hz), 6.26 (s, 2H), 3.90 (s, 3H).
4-Chlorophenylboronic Acid (4b). [1679-18-1],
C6H6BClO2, MW 156.37. 1-Bromo-4-chlorobenzene 4a
(7.735 g, 40.0 mmol) was used in the flow experiment. The
experiment was conducted as described above in the general
procedure for flow experiments. A sample (120 mL) of the
reaction mixture was collected from the output section of the
MJOD flow reactor. This sample was worked up to provide
white, fluffy needles (2.07 g), which corresponds to an isolated
yield of 66%. The corresponding batch experiment, utilizing 11
(1.934 g, 10.0 mmol), provided an isolated yield of 59% (0.93
1128
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
i.d.) connected to the various reservoirs by means of rubber
taps/seals.
trimethyl borate (20.3 mL, 180 mmol, 1.5 equiv) and
anhydrous THF (20.3 mL). P1: 10.68 mL × min−1, P2: 1.80
mL × min−1, P3: 2.53 mL min−1. Retention time L1: 3.75 min,
L2: 4.88 min; 111 mL reaction mixture was collected giving the
product 2b as a white powder in 5.42 g, 64% isolated yield.
Experiment at 160 mmol scale: R1: 1-bromo-4-methox-
ybenzene 2a (20.3 mL, 160 mmol) and anhydrous THF (128
mL). R2: n-BuLi in pentane 10.0 M (19.2 mL, 192 mmol, 1.2
equiv) and hexane, mixture of isomers, (19.2 mL). R3:
trimethyl borate (27.0 mL, 240 mmol, 1.5 equiv) and
anhydrous THF (27.0 mL). P1: 9.91 mL × min−1, P2: 2.57
mL × min−1, P3: 3.62 mL × min−1. Retention time L1: 3.75
min, L2: 4.55 min; 105 mL reaction mixture was collected
giving the product 2b as a white powder in 8.02 g, 72% isolated
yield.
Experiment at 200 mmol scale: R1: 1-bromo-4-methox-
ybenzene 2a (25.3 mL, 200 mmol) and anhydrous THF (100
mL). R2: n-BuLi in pentane 10.0 M (24.0 mL, 240 mmol, 1.2
equiv) and hexane, mixture of isomers, (24.0 mL). R3:
trimethyl borate (33.8 mL, 300 mmol, 1.5 equiv) and
anhydrous THF (33.8 mL). P1: 9.02 mL × min−1, P2: 3.46
mL × min−1, P3: 4.87 mL × min−1. Retention time L1: 3.75
min, L2: 4.22 min; 105 mL reaction mixture was collected
giving the product 2b as a white powder in 8.52 g, 64% isolated
yield.
The pumps, P1, P2, P4, and the oscillator were started
simultaneously. The pumps P3 and P5 were started 3.75 min
later, which corresponds to the residence time of reaction step
1. At that point of time, the pumps P1 and P2 had filled the
reactor body up to the P3 inlet point. The postreaction mixture
started to leave the output section after a total residence time of
9.12 min.
The reactor was run for 4 h before the reservoirs were empty.
The postreaction mixture (∼3200 mL) was collected in four
equal amounts (800 mL each) in four flasks, each containing
saturated NH4Cl solution (320 mL) under stirring. The reactor
was washed with anhydrous THF (200 mL) for 15 min, and
this wash was collected in the last of the four collection flasks.
The pumps, P1, P2, P3, and the reactor were washed with
acetone before the reactor was emptied of solvent through one
of the bottom inlets, and the reactor was dried overnight with a
stream of N2. The pump heads of P2 and P3 were dismantled,
washed with water, and dried at 100 °C.
Workup. The organic solvents were removed under reduced
pressure. The combined aqueous solutions (∼1300 mL) were
acidified with 3 M HCl (120 mL). Because of the large volume
a continuous liquid−liquid extractor was utilized instead of a
separatory funnel. Dichloromethane (500 mL × 3) was refluxed
in the extractor for 2.5 h each time to extract the organic
compounds. The combined organic fractions (∼1600 mL) were
split into two parts and each washed with a saturated sodium
chloride solution (250 mL). The combined organic phases
were dried over Na2SO4 and filtered, and the solvent was
removed under reduced pressure using a rotary evaporator to
obtain a white solid contaminated with a yellow oil.
Experiment at 240 mmol scale: R1: 1-bromo-4-methox-
ybenzene 2a (30.4 mL, 240 mmol) and anhydrous THF (91
mL). R2: n-BuLi in pentane 10.0 M (28.8 mL, 288 mmol, 1.2
equiv) and hexane, mixture of isomers, (28.8 mL). R3:
trimethyl borate (40.5 mL, 360 mmol, 1.5 equiv) and
anhydrous THF (20.3 mL). P1: 8.47 mL × min−1, P2: 4.01
mL × min−1, P3: 4.24 mL × min−1. Retention time L1: 3.75
min, L2: 4.38 min. The experiment failed due to high
backpressure caused by the high viscosity of the reaction
mixture.
The resulting crude was stirred with hexane (mixture of
isomers) (50 mL × 3). The solvent was removed on a filter
nutch to provide a white-colored powder. The filtrate was
evaporated, giving another two crops with the same method to
1
achieve a total isolated yield of 55%. H NMR (400 MHz,
CONCLUSION
■
CDCl3): δ 8.15 (d, 2H, J = 8.1 Hz), 6.99 (d, 2H, J = 8.1 Hz),
3.87 (s, 3H).
We have designed and realized modifications of the multijet
oscillating disk (MJOD) millireactor system that allows the
reactor system also to be utilized as a flow reactor platform for
conducting reactions under cryogenic conditions using a
cryostat circulator as cooling source. Subsequently, a two-step
synthetic process was investigated at cryogenic temperatures
using the prototypes and resulting in a flow reactor model that
operated according to our predefined requirements.
Experimental Conditions for the Investigation of the
Throughput Capacity of the cryoMJOD Flow Reactor.
Experiment at 40 mmol scale: R1: 1-bromo-4-methoxyben-
zene 2a (5.06 mL, 40.0 mmol) and anhydrous THF (192 mL).
R2: n-BuLi in pentane 2.0 M (24.0 mL, 48.0 mmol, 1.2 equiv).
R3: trimethyl borate (6.76 mL, 60.0 mmol, 1.5 equiv) and
anhydrous THF (13.5 mL). P1: 11.13 mL × min−1, P2: 1.36
mL × min−1, and P3: 1.14 mL × min−1. Residence time in L1:
3.75 min and L2: 5.37 min. Vsample = 114 mL, reaction mixture
was collected at the output section of the MJOD reactor giving
the product 2b as a white powder in 2.07 g, 72% isolated yield.
Experiment at 80 mmol scale: R1: 1-bromo-4-methox-
ybenzene 2a (10.1 mL, 80.0 mmol) and anhydrous THF (144
mL). R2: n-BuLi in pentane 2.0 M (48.0 mL, 96.0 mmol, 1.2
equiv). R3: trimethyl borate (13.5 mL, 120 mmol, 1.5 equiv)
and anhydrous THF (27.0 mL). P1: 9.52 mL min−1, P2: 2.96
mL × min−1, and P3: 2.50 mL × min−1. Retention time L1:
3.75 min, L2: 4.89 min; 110 mL reaction mixture was collected
giving the product 2b as a white powder in 3.37 g, 60% isolated
yield.
This cryoMJOD flow reactor operates with net volumes at
multimilliliter scale rather than at the microliter scale, which is
the normal volume of commercially available flow reactor
systems. Moreover, the flow reactor setup involves a reactor
construction that allows accomplishment of two-step tele-
scoped reactions. This reactor setup was utilized for developing
a telescoped flow-process comprising two steps: (1) lithiation
of a phenylbromide and (2) reaction between the phenyl-
lithium derivate and trimethyl borate or other borates.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR spectra of all compounds. The H NMR and 13C
1
Experiment at 120 mmol scale: R1: 1-bromo-4-methox-
ybenzene 2a (15.2 mL, 120 mmol) and anhydrous THF (156
mL). R2: n-BuLi in pentane 10.0 M (14.4 mL, 144 mmol, 1.2
equiv) and hexanes, mixture of isomers, (14.4 mL). R3:
NMR spectra and HRMS data for 2,3,4-trimethoxy-6-
methylphenylboronic acid 6b. This material is available free
of charge via the Internet at http://pubs.acs.org.
1129
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130

Organic Process Research & Development
Article
97. (f) Stanforth, S. P. Tetrahedron 1998, 54, 263−303. (g) Suzuki, A.;
Brown, H. C. Suzuki Coupling. In Organic Syntheses via Boranes;
Aldrich Chemical Company: Milwaukee, 2003; Vol. 3, pp 1−314.
(h) Zapf, A. Coupling of Aryl and Alkyl Halides with Organoboron
Reagents (Suzuki Reaction). In Transition Metals for Organic Synthesis,
2nd ed.; Beller, M., Bolm, C., Eds.; Wiley−VCH: Weinheim, Germany,
2004; 1, pp 211−229. (i) Molander, G. A.; Dehmel, F. J. Am. Chem.
Soc. 2004, 126, 10313−10318. (j) Coleman, R. S.; Lu, X.; Modolo, I. J.
Am. Chem. Soc. 2007, 129, 3826−3827.
AUTHOR INFORMATION
Corresponding Author
* E-mail: hans.bjorsvik@kj.uib.no. Telephone: +47 55 58 34
52, and fax: +47 55 58 94 90.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(17) Glasnov, T. N.; Kappe, C. O. Adv. Synth. Catal. 2010, 352,
3089−3097.
Mr. Gianfranco Liguori and Mr. Steinar Vatne are acknowl-
edged for excellent technical assistance in connection with the
realization and manufacturing of the various parts that
constitute the dedicated MJOD flow reactor system for low-
temperature reactions. D.S. is grateful to the Department of
Chemistry at the University of Bergen for funding of the
research fellowship.
(18) Goossen, L. J.; Melzer, B. J. Org. Chem. 2007, 72, 7473−7476.
(19) Adams, J.; Elliot, P. J. Oncogene 2000, 19, 6687−6692.
(20) Groviak, M. P. Am. J. Ther. 2001, 8, 321−328.
(21) Westmark, P. R.; Smith, B. D. J. Am. Chem. Soc. 1994, 116,
9343−9344.
́
(22) Weston, G. S.; Blazquez, F.; Baquero, F; Schoichet, B. K. J. Med.
Chem. 1998, 41, 4577−4586.
(23) James, T. D.; Sandanayake, K.; Iguchi, R.; Shinkai, S. K. J. Med.
Chem. Soc. 1995, 117, 8982−8987.
REFERENCES
■
(1) (a) Watts, P.; Haswell, S. J. Chem. Soc. Rev. 2005, 34, 235−246.
(b) Brivio, M.; Verboom, W.; Reinhoudt, D. N. LabChip 2006, 6,
329−344. (c) Ahmed-Omer, B.; Brandt, J. C.; Wirth, T. Org. Biomol.
Chem. 2007, 5, 733−740. (d) Watts, P.; Wiles, C. Chem. Commun.
2007, 443−467. (e) Wirth, T., Ed. Microreactors in Organic Synthesis;
Wiley-VCH: New York, 2008; pp 1−297. (f) Yoshida, J.; Nagaki, A.;
Yamada, T. Chem.Eur. J. 2008, 14, 7450−7459. (g) Special Issue on
Flow Chemistry. ChemSusChem 2012, 5(2), 215−350.
(2) (a) Syrris Ltd. http://www.syrris.com/. (b) Chemtrix BV http://
www.chemtrix.com/ (c) Uniqsis Ltd. http://www.uniqsis.com/
(c) Milestone srl http://www.milestonesrl.com/.
(3) Browne, D. L.; Deadman, B. J.; Ashe, R.; Baxendale, I. R.; Ley, S.
V. Org. Process Res. Dev. 2011, 15, 693−697.
(4) AM Technology: http://www.amtechuk.com/.
(5) Peacock, D. G., Richardson, J. F., Eds. Chemical and Biochemical
Reactors and Process Control. In Chemical Engineering, 3rd. ed.;
Butterworth-Heinemann: Oxford, 1994; Vol. 3.
(24) In the MJOD flow reactor, backmixing is dependent on the fluid
flow rate and the oscillating frequency. With a very low fluid flow rate
from the feeding pumps (via the input section) and a high frequency
of the oscillator, backmixing will in some degree take place. In the case
of the MJOD reactor we say that backmixing occurs when there is
backflow from a reactor cavity (the volume between two discs) to a
cavity at a step closer (lower step) to the input section.
(25) Box, G. E. P.; Hunter, W. G.; Hunter, J. S. Statistics for
Experimenters. In An Introduction to Design, Data Analysis, and Model
Building; Wiley: New York, 1978.
(26) (a) Draper, N. R.; Smith, H. Applied Regression Analysis, 3rd ed.;
Wiley: New York, 1998. (b) Montgomery, D. C.; Peck, E. A.
Introduction to Linear Regression Analysis; Wiley: New York, 1982.
(27) The process conditions were identical, except for the reactor
residence time (3.75 min + 5.37 min) that was according to the
literature; see, e.g., ref 29a and the Experimental Section for more
details.
(28) Wenjie, L.; Nelson, D. P.; Jensen, M. S.; Hoerrner, R. S.; Cai,
D.; Larsen, R. D.; Reider, P. J. J. Org. Chem. 2002, 67, 5394−5397.
(29) (a) Fukuda, T.; Sudo, E.; Shimokawa, K.; Iwao, M. Tetrahedron
2008, 64, 328−338. (b) Wang, C.; Rakshit, S.; Glorius, F. J. Am. Chem.
Soc. 2010, 132, 14006−14008. (c) Lloud, D. G.; Hughes, R. B.;
Zisterer, D. M.; Williams, D. C.; Fattorusso, C.; Catalanotti, B.;
Campiani, G.; Meegan, M. J. J. Med. Chem. 2004, 47, 5612−5615.
(30) Li, W.; Nelson, D. P.; Jensen, M. S.; Hoerrner, R. S.; Cai, D.;
Larsen, R. D.; Reider, P. J. J. Org. Chem. 2002, 67, 5394−5397.
(31) The reported result was measured by HPLC as described by Li
and collaborators, reference 30.
(6) Fluens Synthesis http://www.FluensSynthesis.com/.
(7) (a) Hofmann, A. W. Ber. 1881, 14, 2725−2736. (b) Moriarty, R.
M. J. Org. Chem. 2005, 70, 2893−2903.
(8) Liguori, L.; Bjørsvik, H.-R. Org. Process Res. Dev. 2011, 15, 997−
1009.
(9) Dressen, M. H. C. L.; van de Kruijs, B. H. P.; Meuldijk, J.;
Vekemans, J. A. J. M.; Hulshof, L. A. Org. Process Res. Dev. 2010, 14,
351−361.
(10) Spaccini, R.; Liguori, L.; Punta, C.; Bjørsvik, H.-R.
ChemSusChem 2012, 5, 261−265.
(11) Hessel, V.; Hofmann, C.; Lowe, H.; Meudt, A.; Scherer, S.;
̈
Schonfeld, F.; Werner, B. Org. Process Res. Dev. 2004, 8, 511−523.
̈
(32) Information about FMI pumps can be found at http://www.
fmipump.com/.
(33) Pump P1: FMI QBG-MB-Q485 and pump head Q1-SSY.
Pumps P2 and P3: FMI QG20-2-MB-Q485 and pump head Q1-CKC-
W. Pump P4: FMI QG20-2-MB-Q485 and pump head Q1-SSY. Pump
P5: FMI QD-2-MB and pump head Q0-SSY. http://www.fmipump.
com/.
(12) Brown, D. L.; Baumann, M.; Harji, B. H.; Baxendale, I. R.; Ley,
S. V. Org. Lett. 2011, 13, 3312−3315.
(13) Shu, W.; Pellegatti, L.; Oberli, M. A.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2011, 50, 10665−10669.
(14) (a) Heck, R. F.; Negishi, E.-i.; Suzuki, A. Scientific Background of
the Nobel Prize in Chemistry 2010. Palladium-Catalyzed Cross Couplings
in Organic Synthesis; The Royal Swedish Academy of Sciences:
Stockholm, Sweden; October 6, 2010; http://www.nobelprize.org/
n o b e l _ p r i z e s / c h e m i s t r y / l a u r e a t e s / 2 0 1 0 / a d v a n c e d -
chemistryprize2010.pdf. (b) Suzuki, A. Angew. Chem., Int. Ed. 2011, 50,
6722−6737.
(15) (a) Miyaura, N.; Yamada, Y.; Suzuki, A. Tetrahedron Lett. 1979,
20, 3437−3440. (b) Miyaura, N.; Suzuki, A. J. Chem. Soc., Chem.
Commun. 1979, 866−867.
(16) (a) Tidwell, J. H.; Peat, A. J.; Buchwald, S. L. J. Org. Chem. 1994,
59, 7164−7168. (b) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95,
2457−2483. (c) Kawasaki, I.; Katsuma, H.; Nakayama, Y.; Yamashita,
M.; Ohta, S. Heterocycles 1998, 48, 1887−1901. (d) Kawaski, I.;
Yamashita, M.; Ohta, S. Chem. Pharm. Bull. 1996, 44, 1831−1839.
(e) Suzuki, A. In Metal-Catalyzed Cross-Coupling Reactions; Diederich,
F., Stang, P. J., Eds.; Wiley−VCH: Weinhein, Germany, 1998; pp 49−
1130
dx.doi.org/10.1021/op3000493 | Org. Process Res. Dev. 2012, 16, 1121−1130