Optimizing the deprotection of the amine protecting p-methoxyphenyl group in an automated microreactor platform

Article abstract of DOI:10.1021/op900139u

Three factors (temperature, stoichiometry and reaction temperature) were investigated in continuous flow microreactors in an automated fashion for optimization of the removal of the p- methoxyphenyl (PMP) protecting group, thereby consuming only minute amounts of substrate (0.2 mg/sample). The optimal reaction conditions were also applied to a larger microreactor system, in which the corresponding free amine was obtained at a preparative scale.

Full text of DOI:10.1021/op900139u

Organic Process Research & Development 2009, 13, 1003–1006
Optimizing the Deprotection of the Amine Protecting p-Methoxyphenyl Group in an
Automated Microreactor Platform
Kaspar Koch, Bram J. A. van Weerdenburg, Jorge M. M. Verkade, Pieter J. Nieuwland, Floris P. J. T. Rutjes,* and
Jan C. M. van Hest*
Radboud UniVersity Nijmegen, Institute for Molecules and Materials, Heyendaalseweg 135,
NL-6525 AJ Nijmegen, The Netherlands
Scheme 1. General scheme for the deprotection of
p-methoxyphenyl-protected amines
Abstract:
Three factors (temperature, stoichiometry and reaction temper-
ature) were investigated in continuous flow microreactors in an
automated fashion for optimization of the removal of the p-
methoxyphenyl (PMP) protecting group, thereby consuming only
minute amounts of substrate (0.2 mg/sample). The optimal reaction
conditions were also applied to a larger microreactor system, in
which the corresponding free amine was obtained at a preparative
scale.
phenyl (PMP) group stood out as a crucial protecting group
for the amine function, giving rise to optimal enantio- and
diastereoselectivity. The inevitable removal of the PMP group,
however, appeared an important drawback for scale-up of this
methodology, since common deprotection methods require the
use of toxic and expensive reagents (e.g., ceric ammonium
nitrate (CAN), or PhI(OAc)2) and the use of column chroma-
tography. Recently, Verkade et al.15 developed a mild and
efficient method to remove the PMP-group, leading smoothly
to the corresponding amines in a one-pot procedure (see Scheme
1).16,17
The latter method involved the use of either periodic acid
or trichloroisocyanuric acid (TCCA) in the presence of one
equivalent of sulfuric acid and water. This causes oxidation of
the aromatic ring to give the corresponding quinone-derived
iminium ion, which is then hydrolyzed, resulting in overall
removal of the PMP group. Due to the cheap reagents and
favorable atom economy, these conditions are well-suited for
large-scale application in an industrial setting. These particular
conditions, however, have not been fully optimized yet.
We have previously shown that small microreactors with
internal volumes in the range of a few microliters can be
successfully applied to screen and optimize chemical reactions
using tiny amounts of reaction fluids.18,19 Although these
microreactors in combination with a robot and efficient analyti-
cal means are capable of screening many reaction parameters
in a short time frame, we chose to apply the principles of Design
of Experiment (DOE), more specifically D-optimal design,20
Introduction
In the recent past, the interest in using microreactors for
synthetic purposes has increased enormously.1-7 Traditionally,
the emphasis has been either on the production of chemicals in
microstructured flow reactors providing several benefits over
conventional batch reactors or on rather specialized and novel
reaction processes on a very small scale.8,9 Scaling up, or scaling
out using microreactor setup multiplication, has been a particular
subject of investigation.10 Surprisingly, only a few examples
exist on the application of microliter or nanoliter volume reactors
for screening purposes, in particular for reactions that are
commonly used in organic synthesis laboratories.11,12
With the advent of new catalytic strategies to produce
enantiopure products, asymmetric one-pot direct crossed-
Mannich reactions constitute an elegant entry into ꢀ-
aminoketones.13,14 In many of these reactions, the p-methoxy-
* Authors for correspondence. E-mail: f.rutjes@science.ru.nl; j.vanhest@
science.ru.nl.
(1) Ehrfeld, W.; Hessel, V.; Lo¨we, H.,Eds. Microreactors: New Technol-
ogy for Modern Chemistry; Wiley-VCH: Weinheim, 2000; p xi.
(2) Jahnisch, K.; Hessel, V.; Lowe, H.; Baerns, M. Angew. Chem., Int.
Ed. 2004, 43, 406.
(3) Watts, P.; Wiles, C. Chem. Commun. 2007, 443.
(4) Geyer, K.; Codee, J. D. C.; Seeberger, P. H. Chem.sEur. J. 2006,
12, 8434.
(5) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan, A. R.; McQuade,
D. T. Chem. ReV. 2007, 107, 2300.
(6) Usutani, H.; Tomida, Y.; Nagaki, A.; Okamoto, H.; Nokami, T.;
Yoshida, J. J. Am. Chem. Soc. 2007, 129, 3046.
(7) Wirth, T. E., Ed. Microreactors in Organic Synthesis and Catalysis;
Wiley-VCH: Weinheim, 2008.
(15) Verkade, J. M. M.; van Hemert, L. J. C.; Quaedflieg, P. J. L. M.;
Alsters, P. L.; van Delft, F. L.; Rutjes, F. P. J. T. Tetrahedron Lett.
2006, 47, 8109.
(16) For deprotection using laccases, see: Verkade, J. M. M.; van Hemert,
L. J. C.; Quaedflieg, P. J. L. M.; Schoemaker, H. E.; Schurmann, M.;
van Delft, F. L.; Rutjes, F. P. J. T. AdV. Synth. Catal. 2007, 349, 1332.
(17) For application in total synthesis, see: Verkade, J. M. M.; van der
Pijl, F.; Willems, M. A. J. H. P.; Quaedflieg, P. J. L. M.; van Delft,
F. L.; Rutjes, F. P. J. T. J. Org. Chem. 2009, 74, 3207.
(8) Thomsen, M. S.; Polt, P.; Nidetzky, B. Chem. Commun. 2007, 2527.
(9) Fortt, R.; Wootton, R. C. R.; de Mello, A. J. Org. Process Res. DeV.
2003, 7, 762.
(10) Watts, P.; Wiles, C.; Haswell, S. J.; Pombo-Villar, E.; Styring, P. Chem.
Commun. 2001, 990.
(11) Fernandez-Suarez, M.; Wong, S. Y. F.; Warrington, B. H. Lab Chip
2002, 2, 170.
(18) Trieu, H. C.; Slotkowski, J.; Klieber, R.; van Hest, J. C. M.; Rutjes,
F. P. J. T.; Koch, K.; Nieuwland, P. J.; Wiebe, P. Chip holder, fluidic
system and chip holder system. WO2008049447, 2008.
(19) Koch, K.; van den Berg, R. J. F.; Nieuwland, P. J.; Wijtmans, R.;
Schoemaker, H. E.; van Hest, J. C. M.; Rutjes, F. P. J. T. Biotechnol.
Bioeng. 2008, 99, 1028.
(12) Odedra, A.; Seeberger, P. H. Angew. Chem., Int. Ed. 2009, 48, 2699.
(13) Marques, M. M. B. Angew. Chem., Int. Ed. 2006, 45, 348.
(14) Verkade, J. M. M.; van Hemert, L. J. C.; Quaedflieg, P. J. L. M.;
Rutjes, F. P. J. T. Chem. Soc. ReV. 2008, 37, 29.
10.1021/op900139u CCC: $40.75  2009 American Chemical Society
Published on Web 08/27/2009
Vol. 13, No. 5, 2009 / Organic Process Research & Development
•
1003

Scheme 2. p-Methoxyphenyl (PMP)-protected model
substrate
Scheme 3. Schematic of the microreactor setup (internal
microreactor (M) reaction volume 7 µL)
in order to come up with a rationalized set of reaction conditions.
Traditional experimental designs (Full Factorial Designs, Frac-
tional Factorial Designs, and Response Surface Designs) are
appropriate for calibrating linear models in experimental settings
where factors are relatively unconstrained in the region of
interest. In some cases, however, models are necessarily
nonlinear. In other cases, certain treatments (combinations of
factor levels) may be expensive or impossible to measure.
D-optimal designs are model-specific designs that address these
limitations of traditional designs. With D-optimal design the
number of experiments can be reduced compared to those for
standard factorial or fraction factorial designs so that less time
and resources are needed.
The first goal was to optimize the removal of a PMP
protecting group in a continuous flow microreactor platform
using D-optimal design. A second goal was to apply the
optimized reaction conditions to a larger microreactor to scale
up the process to preparative amounts. We selected the protected
amine 1 (Scheme 2) as a representative example for PMP group
removal.
3 would be well-suited to study the flow process. Due to the
fact that sulfuric acid can deactivate the periodic acid over time,
we reasoned that sulfuric acid had to be combined with the
PMP-substrate instead of periodic acid. Since the deprotection
is efficiently stopped at higher pH values, we used aqueous
sodium hydroxide and sodium dithionite for quenching the
process.
During the continuous flow optimization experiment, the
reaction temperature was varied between 60 and 90 °C, the
reaction time between 0.5 and 4 min, and the stoichiometry
between 1 and 4. The design for screening the domain contained
51 samples and was transferred to the automated microreactor
platform,18 in which the individual 51 reactions (approximately
0.2 mg per sample) were automatically conducted in a total
sampling time of 5.6 h. The samples were analyzed by HPLC
resulting after polynomial fitting depicted in the contour plots
in Figure 2 (raw data available in Table S1, Supporting
Information). Note that the response is shown in a logarithmic
scale to better visualize the modeled response at high conversions.
The contour plots indeed show that the factors, reaction time
and temperature, have a considerable effect on the yield. On
the other hand, the effect of an increasing stoichiometry is less
pronounced. The reaction conditions for a conversion of >99%
combined with the shortest reaction time (highest flow rate)
were observed at a reaction time of 1.3 min, a stoichiometry of
3.2, and a reaction temperature of 90 °C.
The next step was to transfer the optimal conditions from
the 7 µL to a 950 µL internal volume microreactor in order to
conduct the same deprotection at a preparative scale. For this
purpose, a standard commercially available stainless steel
continuous flow reactor was selected and the optimal settings
from the screening experiments were applied. At a reaction
temperature of 90 °C, however, the solvents (water and
acetonitrile) started to boil inside the continuous flow reactor,
giving rise to an unstable outflow of the product. In order to
circumvent this issue, the temperature was lowered to 80 °C,
while the reaction time was increased to 4 min in order to
achieve 100% conversion (according to the screening experi-
ments displayed in Figure 2). The reaction fluids were continu-
ously pumped through the reactor for approximately 4 h. The
flow rates of the individual pumps were set to 176 µL/min
(approximately 10 mL/h), leading to a continuous deprotection
of PMP-protected amine 1 at a rate of 213 mg/h. The conversion
of the outflow was monitored at intervals and always appeared
>99%, confirming that the initially identified optimal depro-
tection conditions can also be successfully used in the larger
continuous flow reactor.
Results and Discussion
Initially, the deprotection was studied in small batch reactors
at various temperatures (60-90 °C) and monitored by taking
small samples from the reaction medium for subsequent HPLC
analysis, resulting in the graph depicted in Figure 1. This clearly
shows that elevating the reaction temperature leads to an
enhancement of the reaction rate. Thus, both factors (reaction
time and temperature) have a significant effect and have to be
invoked as factors in the reaction model. Additionally, the
stoichiometry (ratio PMP-substrate:periodic acid) was taken as
a factor in the DOE.
In order to be able to optimize the reaction in the microre-
actor platform, it also had to be translated into a continuous
flow process. We envisioned that the design shown in Scheme
Figure 1. Conversion of the PMP-substrate into its correspond-
ing free amine (in batch reactors).
1004
•
Vol. 13, No. 5, 2009 / Organic Process Research & Development

Figure 2. Contour plots of the optimized model of the PMP-group removal in the automated microreactor platform (polynomial
fit).
Conclusion
Three factors (temperature, stoichiometry and reaction time)
in. ID 1/16 in. OD, custom-filed needle tip) was used as outlet.
A sample robot (Gilson Aspec XL) with dedicated software
(Gilson 735 Sample software, Version 1.00) was used to
dispense all samples during reaction screening. The pumps,
robot, and temperature controller were automatically controlled
with a custom-designed software program (developed by
Fraunhofer IMS, Duisburg, Germany).
Microreactor. The microchannel structure was designed
using the software program CleWIN. The actual microreactor
was fabricated from borosilicate glass by Micronit Microfluidics
BV, Enschede, The Netherlands (HF etched). Chip dimension:
length 45 mm, width 15 mm, height 2.2 mm. Channel
dimension: width 120 µm, depth 60 µm, total length 132 cm.
Reactor reaction volume 7.02 µL.
Microreactor Experiment for Deprotection Using D-
Optimal Design. The first syringe was loaded with solution A
containing substrate 1 (200 mg, 0.880 mmol), 1 M sulfuric acid
(0.88 mL, 1 equiv) and p-nitrophenol (100 mg, 0.718 mmol,
internal standard) dissolved in MeCN/H2O (10 mL, 1:1). The
second syringe was loaded with solution B containing periodic
acid (201 mg, 0.880 mmol, 1 equiv) and benzoic acid (100
mg, 0.819 mmol, internal standard) dissolved in MeCN/H2O
(10 mL, 1:1). The third syringe was loaded with the quenching
solution (Q) containing sodium dithionite (391 mg, 2.24 mmol,
2.55 equiv) dissolved in MeCN/0.1 M NaOH (25 mL, 1:1).
Solution C was prepared by dissolving biphenol (30 mg, 0.161
mmol, internal standard) in MeCN/0.1 M NaOH (25 mL, 1:1).
Solution D was made by diluting solution C (3 mL) to 25 mL
with MeCN/0.1 M NaOH (1:1). Solutions A, B, and Q were
then connected to the microreactor system. Of each reaction
mixture, 20 µL was collected in 500 µL of solution D. Due to
the varying flow rates, sampling times differ for every experi-
ment. All reaction conditions were randomized except for the
temperature. The samples were analyzed with HPLC. The
retention times are 6.23, 8.20, 8.51, 9.59, and 10.97 min,
respectively, for the deprotected amine, benzoic acid, PNP, 2,2-
biphenol, and PMP substrate 1.
were investigated in continuous flow microreactors in an
automated fashion for optimization of the removal of the
p-methoxyphenyl (PMP) amine protecting group. As a result
of the small dimensions of the microreactor, only tiny amounts
of reagents and solvents were required in order to identify
optimal reaction conditions. The latter conditions were also
applied to a larger microreactor system to synthesize the free
amine product at a preparative scale. In conclusion, the PMP
group could be conveniently removed in 100% conversion
within two minutes in a continuous flow microreactor, both at
small and at preparative scale, which clearly underlines the
potential of flow chemistry in organic synthesis. Although we
efficiently used HPLC as the analysis method for determining
reaction conversion, online detection such as IR and UV will
be considered to decrease the time required for optimization
even further.
Experimental Section
HPLC Analysis. HPLC analysis was performed on a
Shimadzu VP LC10 equipped with a 250 mm × 4 mm × 4.6
mm Inertsil ODS-3 column, using a gradient program with
acetonitrile and 20 mM phosphate buffer (pH 3) with a flow
rate of 2 mL min-1. Detection took place at a wavelength of
254 nm.
Synthesis of PMP-Protected Amine 1. See procedure
described by Verkade et al.15
Microreactor Setup. All syringes (Harvard apparatus; high
pressure syringe, 2 mL) mounted on a syringe pump (New Era;
type NE-1000 or NE 500) were connected to FEP tubing (1.59
mm OD, 254 µm ID). At the end of each piece of tubing, a
special ‘flat bottom headless nut’ (Upchurch Scientific; type:
M 660) was mounted which pressed down onto a flat bottom
ferrule (Upchurch Scientific; type: M 650) to achieve a leak-
free fluid connection to the microreactor. The microreactor was
placed in a custom-designed chip holder18 with threaded holes
on the top side in which the nuts were screwed. For temperature
control, a custom-designed heater (peltier element) was used,
which could easily slide into the microreactor chipholder. A
stainless steel needle (UpChurch Scientific; type U 106 1/100
Deprotection in the Larger-Scale Continuous Flow Sys-
tem. A stainless steel reactor (internal volume 0.95 mL)
was used in combination with a commercial available
T-junction (Upchurch) acting as a mixer. The reactor was
submerged in an oil bath, and the two inlets of the
T-junction were connected to the syringes. Both syringes
(20) de Aguiar, P. F.; Bourguignon, B.; Khots, M. S.; Massart, D. L.; Phan-
Than-Luu, R. Chemom. Intell. Lab. Syst. 1995, 30, 199.
Vol. 13, No. 5, 2009 / Organic Process Research & Development
•
1005

(50 mL) were mounted on a syringe pump (New Era; type
NE-1000) and were connected with FEP tubing (1.59 mm
OD, 254 µm ID) to the T-junction. The following solutions
were prepared: Solution A: compound 1 (1.01 g, 4.46
mmol), p-nitrophenol (internal standard, 49.3 mg), and 1
M H2SO4 (4.4 mL) were diluted to 50 mL with acetoni-
trile/H2O (1:1). Solution B: periodic acid (1.00 g, 4.41
mmol) was diluted to 50 mL with acetonitrile/H2O (1:1).
Both solutions were degassed for 50 min with ultrasound.
The flow rates of the pumps were set to a total of 167
µL · min-1 corresponding to a total reaction time of 4 min.
After stabilizing the system for 15 min, the outflow was
collected for 255 min. During this time frame, the
conversion was regularly measured and shown to be >99%
according to HPLC analysis. Subsequently, the collection
vial was worked up according to the procedure by Verkade
et al.15 resulting in an isolated yield of 60% (329 mg) of
the corresponding free amine.
Acknowledgment
Dr. Ron Wehrens and Noud van Harskamp are kindly
acknowledged for contributions regarding the Design of Experi-
ment and the modeling and validation of the models.
Supporting Information Available
Raw data from the optimization experiment is shown in
Table S1. This material is available free of charge via the
Internet at http://pubs.acs.org.
Received for review June 4, 2009.
OP900139U
1006
•
Vol. 13, No. 5, 2009 / Organic Process Research & Development