Scalability of microwave-assisted organic synthesis. From single-mode to multimode parallel batch reactors

Article abstract of DOI:10.1021/op034075+

The direct scalability of microwave-assisted organic synthesis (MAOS) in a prototype laboratory-scale multimode microwave batch reactor is investigated. Several different organic reactions have been scaled-up typically from 1 mmol to 100 mmol scale. The transformations include multicomponent chemistries (Biginelli dihydropyrimidine and Kindler thioamide synthesis), transition metal-catalyzed carbon - carbon cross-coupling protocols (Heck and Negishi reactions), solid-phase organic synthesis, and Diels - Alder cycloaddition reactions using gaseous reagents in prepressurized reaction vessels. A range of different solvents (high and low microwave absorbing), Pd catalysts (homogeneous and heterogeneous), and varying reaction times and temperatures have been explored in these investigations. In all cases, it was possible to achieve similar isolated product yields on going from a small scale (ca. 5 mL processing volume) to a larger scale (max 500 mL volume) without changing the previously optimized reaction conditions (direct scalability). The prototype, benchtop multimode microwave reactor used in the present study allows parallel processing in either quartz or PTFE-TFM vessels with maximum operating limits of 300°C and 80 bar of pressure. The system features magnetic stirring in all vessels, complete on-line monitoring of temperature, pressure and microwave power, and the ability to maintain inert or reactive gas atmosphere.

Full text of DOI:10.1021/op034075+

Organic Process Research & Development 2003, 7, 707−716
Scalability of Microwave-Assisted Organic Synthesis. From Single-Mode to
Multimode Parallel Batch Reactors
Alexander Stadler, Behrooz H. Yousefi, Doris Dallinger, Peter Walla, Erik Van der Eycken,† Nadya Kaval,† and
C. Oliver Kappe*
Institute of Chemistry, Karl-Franzens-UniVersity Graz, Heinrichstrasse 28, A-8010 Graz, Austria
Abstract:
late 1990s, the number of publications related to MAOS has
increased dramatically to a point where it might be assumed
that, in a few years, most chemists will probably use
microwave energy to heat and drive chemical reactions. In
fact, it appears that rapid microwave protocols can be
developed for most chemical transformations requiring heat.
A large number of review articles and several books provide
extensive coverage of the subject.4-6 Apart from traditional
organic synthesis, more recent applications of microwaves
in this area include combinatorial chemistry (solid- and
fluorous-phase synthesis, parallel processing, use of polymer-
supported reagents),7 biochemical processes such as high-
speed polymerase chain reaction (PCR)8 and rapid enzyme-
mediated protein mapping,9 electrochemistry,10 photochemis-
try,11 and simultaneous ultrasound/microwave processing.12
Processing techniques employed in microwave chemistry
involve solventless (dry-media) procedures where the re-
agents are adsorbed onto either a microwave transparent (i.e.,
silica, alumina or clay) or strongly absorbing (i.e., graphite)
inorganic support, which additionally can be doped with a
catalyst or reagent.13 Alternatively, microwave-assisted syn-
thesis can be carried out using phase-transfer catalysis
(PTC),14 under heterogeneous catalytic gas-phase conditions15
or, most commonly, in solvents. Using the latter technique,
standard organic solvents, high-temperature (near critical)
The direct scalability of microwave-assisted organic synthesis
(MAOS) in a prototype laboratory-scale multimode microwave
batch reactor is investigated. Several different organic reactions
have been scaled-up typically from 1 mmol to 100 mmol scale.
The transformations include multicomponent chemistries (Bigi-
nelli dihydropyrimidine and Kindler thioamide synthesis),
transition metal-catalyzed carbon-carbon cross-coupling pro-
tocols (Heck and Negishi reactions), solid-phase organic syn-
thesis, and Diels-Alder cycloaddition reactions using gaseous
reagents in prepressurized reaction vessels. A range of different
solvents (high and low microwave absorbing), Pd catalysts
(homogeneous and heterogeneous), and varying reaction times
and temperatures have been explored in these investigations.
In all cases, it was possible to achieve similar isolated product
yields on going from a small scale (ca. 5 mL processing volume)
to a larger scale (max 500 mL volume) without changing the
previously optimized reaction conditions (direct scalability). The
prototype, benchtop multimode microwave reactor used in the
present study allows parallel processing in either quartz or
PTFE-TFM vessels with maximum operating limits of 300 °C
and 80 bar of pressure. The system features magnetic stirring
in all vessels, complete on-line monitoring of temperature,
pressure and microwave power, and the ability to maintain inert
or reactive gas atmosphere.
(5) (a) Hayes, B. L. MicrowaVe Synthesis: Chemistry at the Speed of Light;
CEM Publishing: Matthews, NC, 2002. (b) MicrowaVes in Organic
Synthesis; Loupy, A., Ed.; Wiley-VCH: Weinheim, 2002. (c) MicrowaVe-
Assisted Organic Synthesis; Lidstro¨m, P., Tierney, J. P., Eds.; Blackwell
Publishing: Oxford, 2004.
Introduction
High-speed microwave synthesis has attracted a consider-
able amount of attention in recent years.1 Since the first
reports on the use of microwave heating to accelerate organic
chemical transformations by the groups of Gedye and
Giguerre/Majetich in 1986,2,3 more than 2000 articles have
been published in the area of microwave-assisted organic
synthesis (MAOS).4 The initial slow uptake of the technology
has been attributed to its lack of controllability and reproduc-
ibility, coupled to a general lack of understanding of the
basics of microwave dielectric heating. However, since the
(6) (a) Caddick, S. Tetrahedron 1995, 51, 10403-10432. (b) Strauss, C. R.;
Trainor, R. W. Aust. J. Chem. 1995, 48, 1665-1692. (c) Lidstro¨m, P.;
Tierney, J.; Wathey, B.; Westman, J. Tetrahedron 2001, 57, 9225-9283.
(7) (a) Lew, A.; Krutzik, P. O.; Hart, M. E.; Chamberlin, A. R. J. Comb. Chem.
2002, 4, 95-105. (b) Kappe, C. O. Curr. Opin. Chem. Biol. 2002, 6, 314-
320. (c) Lidstro¨m, P.; Westman, J.; Lewis, A. Comb. Chem. High
Throughput Screening 2002, 5, 441-458. (d) Blackwell, H. E. Org. Biomol.
Chem. 2003, 1, 1251-1255.
(8) Ferme´r, C.; Nilsson, P.; Larhed, M. Eur. J. Pharm. Sci. 2003, 18, 129-
132.
(9) Pramanik, B. N.; Mirza, U. A.; Ing, Y. H.; Liu, Y.-H.; Bartner, P. L.; Weber,
P. C.; Bose, A. K. Protein Sci. 2002, 11, 2676-2687.
(10) Tsai, Y.-C.; Coles, B. A.; Compton, R. G.; Marken, F. J. Am. Chem. Soc.
2002, 124, 9784-9788.
* To whom correspondence should be addressed. Phone: +43-316-3805352.
Fax: +43-316-3809840. E-mail: oliver.kappe@uni-graz.at.
† Present address: Laboratory for Organic Synthesis, Department of Chem-
istry, University of Leuven, Celestijnenlaan 200F, B-3001 Leuven, Belgium.
(1) Adam, D. Nature 2003, 421, 571-572.
(2) Gedye, R.; Smith, F.; Westaway, K.; Ali, H.; Baldisera, L.; Laberge, L.;
Rousell, J. Tetrahedron Lett. 1986, 27, 279-282.
(11) For a review on microwave-assisted photochemistry, see ref 5b, chapter
14.
(12) (a) Peng, Y.; Song, G. Green Chem. 2001, 302-304. (b) Peng, Y.; Song,
G. Green Chem. 2002, 349-351.
(13) (a) Varma, R. S. Green Chem. 1999, 43-55. (b) Loupy, A.; Petit, A.;
Hamelin, J.; Texier-Boullet, F.; Jacquault, P.; Mathe´, D. Synthesis 1998,
1213-1234.
(3) Giguere, R. J.; Bray, T. L.; Duncan, S. M.; Majetich, G. Tetrahedron Lett.
1986, 27, 4945-4958.
(4) For online resources and a literature database on microwave-assisted organic
synthesis, see: www.maos.net.
(14) Deshayes, S.; Liagre, M.; Loupy, A.; Luche, J.-L.; Petit, A. Tetrahedron
1999, 55, 10851-10870.
(15) Will, H.; Scholz, P.; Ondruschka, B. Chem.-Ing.-Tech. 2002, 74, 1057-
1067.
10.1021/op034075+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/09/2003
Vol. 7, No. 5, 2003 / Organic Process Research & Development
•
707

water,16 or ionic liquids17 have been employed as reaction
media in open18 or closed vessel systems.19
products on a kilogram scale.25 With some of the physical
limitations of microwave heating technology (magnetron
power, penetration depth) under consideration,5,26 two dif-
ferent approaches for microwave synthesis on a larger scale
(>100 mL volume) have emerged. While some groups have
employed larger batch-type multimode21 or monomode
reactors,27 others have used continuous flow techniques
(multi- and monomode)28 to overcome the inherent problems
associated with MAOS scale-up. In general, one should note
that published examples of MAOS scale-up experiments are
rare, in particular those involving complex organic reactions.
An important issue for the process chemist is the potential
of direct scalability of microwave reactions, allowing rapid
translation of previously optimized small-scale conditions
to a larger scale. Keeping these issues in mind, we herewith
report our findings on the microwave scalability of a range
of organic transformations typically from a less than 1 g to
100 g scale employing a dedicated multimode batch reactor.
Regardless of the specific chemistry or processing tech-
nique, the main benefits of performing reactions under
microwave conditions are the significant rate enhancements
and the higher product yields that can frequently be observed.
While different hypotheses have been proposed to account
for the observed rate enhancements under microwave ir-
radiation, a generally accepted rationalization remains elu-
sive.20 Regardless of the origin/existence of a special
microwave effect, microwave-enhanced chemistry can be
extremely efficient and is applicable to a broad range of
practical synthesis.
Although most of the early pioneering experiments in
microwave-enhanced organic synthesis have been carried out
in unmodified domestic microwave ovens, the current trend
clearly is to use dedicated instruments for chemical synthesis,
in particular for processes involving organic solvents. Begin-
ning in the late 1980s, dedicated multimode21 and mono-
mode22 microwave reactors for organic synthesis were
designed and later became commercially available. Most
recently, the focus in the published literature from both
academic and industrial laboratories has shifted toward the
use of small commercially available monomode (also called
single-mode) microwave applicators, that typically allow the
safe processing of up to 10 mL of reaction volume in a single
reaction vessel with a pressure limit of ca. 20 bar.23,24 These
reactors feature built-in magnetic stirrers, direct temperature
control of the reaction mixture aided by fiber-optic probes
or IR sensors, and software that enables on-line temperature/
pressure control by regulation of microwave power output.
Such applicators have been proven suitable and reliable for
the safe and controlled microwave synthesis of gram-scale
quantities of materials using organic solvents.
Results
General Considerations. One of the main limitations of
microwave scale-up technology is the restricted penetration
depth of microwave irradiation into absorbing materials, that
is, solvents or reaction mixtures. At the typical operating
frequency of most microwave reactors of 2.45 GHz, the
penetration depth is generally in the order of a few
centimeters, depending on the dielectric properties of the
medium.5,26 This means that the microwave power density
inside a large batch reactor (>1 L of volume) may only be
a small fraction of the density on the surface. Therefore,
solvents or reagents in the center of the reaction vessel are
heated by convection and not by direct “in core” microwave
dielectric heating.29 This physical limitation is one of the
main reasons for the development of continuous flow
reactors,28 where the reaction mixture is passed through a
relatively small microwave heated flow cell, avoiding
penetration depth problems. On the other hand, continuous
flow reactors with pumping systems may not be appropriate
for processing solids, highly viscous liquids, or heterogeneous
reaction mixtures.
While the above-mentioned techniques are very successful
for small-scale organic synthesis, in particular for the rapid
optimization of reaction conditions and in the context of the
drug discovery process, there is a clear need to develop
larger-scale MAOS techniques, which can ultimately provide
(25) For general references on microwave scale-up, see: (a) Bose, A. K.; Manhas,
M. S.; Ganguly, S. N.; Sharma, A.; Rao, K. V.; Nageshwar, K. V. 223th
National Meeting of the American Chemical Society, Orlando, FL, 2002;
American Chemical Society: Washington, DC, 2002; IEC 036. (b) Nu¨chter,
M.; Mu¨ller, U.; Ondruschka, B.; Tied, A.; Lautenschla¨ger, W. In Einsatz
innoVatiVer Energietra¨ger in der Verfahrenstechnik; Bathen, D., Schmidt-
Traub, H., Eds.; Shaker Verlag: Aachen, 2001; p 85. (c) Bykov, Yu.;
Rybakov, K. I.; Semenov, V. E. J. Phys. D.: Appl. Phys. 2001, 34, R55-
R75.
(16) (a) An, J.; Bagnell, L.; Cablewski, T.; Strauss, C. R.; Trainor, R. W. J.
Org. Chem. 1997, 62, 2505-2511. (b) Bagnell, L.; Cablewski, T.; Strauss,
C. R.; Trainor, R. W. J. Org. Chem. 1996, 61, 7355-7359.
(17) (a) Leadbeater, N. E.; Torenius, H. M.; Tye, H. Tetrahedron 2003, 59, 2253-
2258. (b) Berthold, H.; Schotten, T.; Ho¨nig, H. Synthesis 2002, 1607-
1610. (c) Vallin, K. S. A.; Emilsson, P.; Larhed, M.; Hallberg, A. J. Org.
Chem. 2002, 67, 6243-6246.
(18) Bose, A. K.; Manhas, M. S.; Ganguly, S. N.; Sharma, A. H.; Banik, B. K.
Synthesis 2002, 1578-1591.
(19) For a discussion of open versus closed microwave technology, see: Stadler,
A.; Pichler, S.; Horeis, G.; Kappe, C. O. Tetrahedron 2002, 58, 3177-
3183.
(20) For a discussion of nonthermal or specific microwave effects, see: (a)
Perreux, L.; Loupy, A. Tetrahedron 2001, 57, 9225-9283. (b) Kuhnert, N.
Angew. Chem., Int. Ed. 2002, 41, 1863-1866. (c) Strauss, C. R. Angew.
Chem., Int. Ed. 2002, 41, 3589-3590.
(21) Raner, K. D.; Strauss, C. R.; Trainor, R. W.; Thorn, J. S. J. Org. Chem.
1995, 60, 2456-2460. See also, www.milestonesci.com for a commercial
system.
(22) Commarmont, R.; Didenot, R.; Gardais, J. F. (Prolabo). French Patent 84/
03496, 1986. See also ref 13b.
(23) www.personalchemistry.com.
(26) MicrowaVe-Enhanced Chemistry; Kingston, H. M., Haswell, S. J., Eds.;
American Chemical Society: Washington, DC, 1997.
(27) (a) Perio, B.; Dozias, M.-J.; Hamelin, J. Org. Process Res. DeV. 1998, 2,
428-430. (b) Cle´ophax, J.; Liagre, M.; Loupy, A.; Petit, A. Org. Process
Res. DeV. 2000, 4, 498-504.
(28) (a) Cablewski, T.; Faux, A. F.; Strauss, C. R. J. Org. Chem. 1994, 59, 3408-
3412. (b) Kazba, K.; Chapados, B. R.; Gestwicki, J. E.; McGrath, J. L. J.
Org. Chem. 2000, 65, 1210-1214. (c) Khadlikar, B. M.; Madyar, V. R.
Org. Process Res. DeV. 2001, 5, 452-451. (d) Esveld, E.; Chemat, F.; van
Haveren, J. Chem. Eng. Technol. 2000, 23, 279-283 and 429-435. (e)
Shieh, W.-C.; Dell, S.; Repicˇ, O. Tetrahedron Lett. 2002, 43, 5607-5609.
(f) Hayes, B. L.; Collins, M. J.; Collins, J. M., Jr. Abstracts of Papers, 225rd
ACS National Meeting, New Orleans, LA, United States, March 23-27,
2003, ORGN-053.
(29) For an in-depth discussion of microwave dielectric heating mechanisms,
see refs 5 and 26. See also: Gabriel, C.; Gabriel, S.; Grant, E. H.; Halstead,
B. S. J.; Mingos, D. M. P. Chem. Soc. ReV. 1998, 27, 213-223.
(24) www.cemsynthesis.com.
708
•
Vol. 7, No. 5, 2003 / Organic Process Research & Development

Scheme 1
reaction times involving reflux temperatures, are the moder-
ate yields frequently observed when using more complex
building blocks. We have recently described a high yielding
and rapid microwave-assisted protocol that allows the
synthesis of gram quantities of DHPMs utilizing controlled
single-mode microwave irradiation.31
As the first model reaction for our scale-up experiments,
we selected the standard Biginelli cyclocondensation,30 where
in a one-pot process equimolar amounts of benzaldehyde,
ethyl acetoacetate, and urea react under Lewis acid (FeCl3)
catalysis to the corresponding dihydropyrimidine 1a (Scheme
1). Utilizing single-mode microwave irradiation, the reaction
can be carried out on a 4.0 mmol scale in AcOH/EtOH 3:1
at 120 °C within 10 min, compared to 3-4 h using
conventional thermal heating, providing DHPM 1a in 88%
isolated yield and high purity (>98%) (Figure 1).31
Adapting this chemistry to scale-up in the multimode
reactor, we filled four vessels of the eight-position rotor with
80 mmol of reagents each (0.32 mol in total), solvents, and
the catalyst in identical concentrations as in the small-scale
experiment. Microwave heating was performed under the
same conditions as in the small-scale run (120 °C for 10
min) providing the desired DHPM 1a in a slightly lower,
but still comparable overall isolated yield (73%). Importantly,
the product yields in the individual four vessels of the eight-
position rotor were virtually identical (70-74%), demon-
strating the good homogeneity of the microwave field in the
cavity.32 In contrast to the small-scale experiments a heating
ramp was programmed for the large-scale run (3 min to 120
°C) to have better control and reproducibility over the heating
profile.
To demonstrate the full range of scalability possible in
the eight-position rotor, another Biginelli condensation was
performed using different building blocks employing two,
four, or eight vessels of the rotor leading to a maximum scale
of 8 × 80 mmol in one single run (0.64 mol, ca. 400 mL of
total reaction volume). As a model reaction, we chose the
synthesis of DHPM 1b (Scheme 1), an important intermedi-
ate in the preparation of the orally active antihypertensive
agent SQ 32926.33 For this particular Biginelli condensation,
a 20 min reaction time at 120 °C using HCl as a catalyst in
EtOH provided the best yield (52%) that could be obtained
in the small-scale optimization experiments (data not shown).
Similar to the first example, the results of the small-scale
run performed in the monomode cavity were comparable to
the scale-up runs in the multimode instrument using the same
set of conditions (in addition to a 3 min linear heating ramp
to 120 °C). There was no significant difference between the
isolated total yields of experiments carried out in two, four,
or eight vessels (43%, 46%, and 48%, respectively of
combined overall yields; for details, see the Experimental
Section). The magnetron power of the reactor (1400 W) is
sufficient to allow the linear heating from room temperature
Having these issues in mind, we here report on MAOS
scale-up experiments involving up to 500 mL of reaction
volume in a dedicated multimode cavity, employing an eight-
position rotor system. This parallel batch processing tech-
nique not only circumvents the issue of penetration depth
but also allows for genuine parallel synthesis, that is,
performing different chemical reactions simultaneously in
one irradiation experiment. The reaction vessels (ca. 60 mL
filling volume each) used in the current studies are made of
either PTFE-TFM or quartz glass and are dedicated for
reactions at high pressures and temperatures (80 bar, 300
°C). The 1400 W multimode microwave reactor is equipped
with magnetic stirring, on-line temperature and pressure
control, and various safety features. The technical details of
this prototype reactor are presented in the Experimental
Section.
Case Studies. To test the general suitability of this reactor
platform to perform larger-scale microwave-assisted reac-
tions, and specifically to investigate the potential for direct
scalability of previously optimized small-scale microwave
protocols, a variety of organic reactions have been covered
in these proof-of-principles studies. These include multi-
component chemistries (Biginelli and Kindler reactions),
transition metal-catalyzed carbon-carbon cross-coupling
protocols (Heck and Negishi reactions), solid-phase organic
synthesis, and Diels-Alder cycloaddition reactions using
gaseous reagents in prepressurized reaction vessels. A range
of solvents (strongly and poorly microwave absorbing),
catalysts (homogeneous and heterogeneous), and varying
reaction times and temperatures have been studied to
rationalize general conclusive trends from these experiments.
A. Biginelli Dihydropyrimidine Synthesis. Multicom-
ponent reactions (MCRs) in general are of increasing
importance in organic and medicinal chemistry. In times
where a premium is put on speed, diversity, and efficiency
in the drug discovery process, MCR strategies offer signifi-
cant advantages over conventional linear-type syntheses. The
Biginelli protocol (Scheme 1) is particularly attractive, since
the resulting dihydropyrimidine (DHPM) scaffold displays
a wide range of biological activities which has led to the
development of a number of lead compounds based on that
structural core.30 In general, the standard procedure for the
Biginelli condensation involves one-pot condensation of the
three building blocks in a solvent such as ethanol using a
strongly acidic catalyst, that is, hydrochloric acid.30 One
major drawback of this procedure, apart from the long
(31) Stadler, A.; Kappe, C. O. J. Comb. Chem. 2001, 3, 624-630.
(32) For similar experiments with multivessel rotor systems in multimode
microwave cavities, see: Nu¨chter, M.; Ondruschka, B.; Tied, A.; Lauten-
schla¨ger, W.; Borowski, K. J. Am. Genomic/Proteomic Technol. 2001, 1,
34-39.
(33) Schnell, B.; Krenn, W.; Faber, K.; Kappe, C. O. J. Chem. Soc., Perkin Trans.
1 2000, 4382-4389 and references therein.
(30) For reviews on the Biginelli reaction, see: (a) Kappe, C. O. Tetrahedron
1993, 49, 6937-6963. (b) Kappe, C. O. Acc. Chem. Res. 2000, 33, 879-
888. (c) Kappe, C. O. Eur. J. Med. Chem. 2000, 35, 1043-1052.
Vol. 7, No. 5, 2003 / Organic Process Research & Development
•
709

Figure 1. Temperature and power profiles for a typical Biginelli condensation (1b, Scheme 1) in AcOH/EtOH (3:1) under sealed
quartz vessel/microwave irradiation conditions. Linear heating ramp to 120 °C (3 min), temperature control using the feedback
from the reference vessel temperature measurement (constant 120 °C, 20 min), and forced air cooling (20 min). The reaction was
performed in eight quartz vessels, each containing 40 mL of reaction volume. Shown is the temperature measurement in one reference
vessel via internal gas balloon thermometer (T), the surface temperatures monitoring of the eight individual vessels by IR
thermography (IR 1-8), and the magnetron power (P, 0-1400 W). Note that the power of the magnetron suffices to follow a linear
heating ramp from 20 to 120 °C in 3 min. After the maximum temperature has been reached, <200 W power is used to keep the
reaction temperature at 120 °C. Also note that the individual IR vessel surface temperatures deviate by less than 10 °C. For clarity,
the pressure graph is not shown.
to 120 °C within 3 min (ramp) of all eight vessels, even
when filled with ca. 50 mL each (total of 400 mL) of reaction
mixture.
Scheme 2
B. Kindler Reaction. Thioamides are essential building
blocks for the preparation of a number of biologically
relevant heterocyclic scaffolds (i.e., utilizing the Hantzsch
thiazole synthesis). While many different methods to prepare
thioamides have been reported in the literature, the three-
component coupling of an aldehyde, elemental sulfur, and
an amine (Kindler reaction) has so far received comparatively
little attention.34 This is despite the fact that this one-pot
process allows for an easy introduction of diversity into the
thioamide backbone by simple variation of the aldehyde and
amine components in the condensation step. Since a large
number of aldehydes and primary/secondary amines are
commercially available, a diverse set of synthetically useful
thioamide products can potentially be prepared in one step
using this method. While conventional Kindler condensations
require long reaction times and high temperatures, sealed
vessel microwave synthesis allows the rapid preparation of
thioamide building blocks in a comparatively short time
frame. Recently, we have reported a rapid microwave Kindler
protocol for the preparation of thioamide libraries on a 4.0
mmol scale in single-mode cavities, using 1-methyl-2-
pyrrolidone (NMP) as solvent employing microwave heating
at 110-180 °C for 2-20 min.35 As a model reaction for the
present scale-up studies, we have chosen the condensation
of elemental sulfur with 3-chlorobenzaldehyde and 4-fluo-
rophenethylamine to form the corresponding thioamide 2
(Scheme 2), an interesting building block for, for example,
heterocycle synthesis.34 The previously optimized 4 mmol
microwave conditions (140 °C, 10 min)35 were directly
applied to a large-scale run (3 min ramp to 140 °C), using
four vessels filled with 40 mmol of reagents each. As in the
previous cases, the isolated yield of 90% for the large-scale
run nicely compared with the 95% yield of thioamide 2
obtained in the previously published small-scale experiment.
C. Heck Reactions. The Heck reaction, the palladium-
catalyzed arylation and vinylation of olefins, was discovered
about 30 years ago and has since become an important
carbon-carbon coupling reaction in organic synthesis, due
to its high chemoselectivity and mild reaction conditions.36
The Heck coupling reaction requires a catalyst and usually
also a phosphine ligand (when a homogeneous catalyst is
used), to stabilize the catalyst-substrate complex. The
presence of an appropriate base is also necessary, and the
reaction is usually performed in a dipolar, aprotic organic
solvent, for example, acetonitrile (MeCN) or 1-methyl-2-
pyrrolidone (NMP). Under conventional heating, tempera-
tures between 60 and 120 °C for several hours are routinely
used with standard catalysts. Recently, it has been demon-
strated that Heck couplings can be dramatically enhanced
(34) For reviews on the Kindler reaction and related processes (e.g., the
Willgerodt reaction), see the following: (a) Carmack, M.; Spielman, M. A.
Org. React. 1946, 3, 83-107. (b) Brown, E. V. Synthesis 1975, 358-375.
For a review on the use of thioamides in heterocyclic synthesis, see:
Jagodzinski, T. S. Chem. ReV. 2003, 103, 197-227.
(35) Zbruyev, O. I.; Stiasni, N.; Kappe, C. O. J. Comb. Chem. 2003, 5, 145-
148.
(36) Handbok of Organopalladium Chemistry for Organic Synthesis; Negishi,
E., Ed.; Wiley-VCH: Weinheim, 2002; Vol. 1, p 1133.
710
•
Vol. 7, No. 5, 2003 / Organic Process Research & Development

Scheme 3
Scheme 4
vessels each with the two different arylbromide building
blocks (Scheme 3).
by microwave heating, reducing reaction times from hours
to minutes.37 For the present scale-up study, the Heck
coupling of aryl bromides with acrylic acid leading to
cinnamic acids has been investigated (Scheme 3). A variety
of related aryl bromide/acrylic acid coupling procedures have
been reported in the literature, in most cases using Pd(OAc)2/
tri(o-tolyl)phosphine as the catalytic system under conven-
tional heating conditions.38 Optimization of the reaction
conditions under small-scale (2 mmol) single-mode condi-
tions led to a protocol that employed acetonitrile as solvent,
1 mol % Pd(OAc)2/tri(o-tolyl)phosphine as catalyst/ligand
system, triethylamine (TEA) as base, and a 180 °C reaction
temperature for 15 min. For the cyano bromobenzene
substrate, 82% yield of the corresponding cinnamic acid 3a
was isolated, and for the fluoro/cyano substrate, the yield of
3b was considerably lower (55%), under otherwise identical
reaction conditions. Several recent reports in the literature
have dealt with the use of heterogeneous palladium on carbon
(Pd/C) as an alternative, simple, and inexpensive catalyst
system for Heck coupling reactions.39 In this context, we
have discovered that the homogeneous Pd(OAc)2/tri(o-tolyl)-
phosphine catalyst system can be replaced for 5% Pd/C40
(<0.1 mol % concentration of Pd catalyst) without the need
to change any of the other reaction parameters. Yields for
the Heck reaction providing cinnamic acid 3a were very
similar using either homogeneous or heterogeneous Pd
catalysis (see Experimental Section for details).
For scale-up, the homogeneous Pd coupling conditions
were employed for experiments on a 4 × 20 mmol scale in
the multimode batch reactor (ca. 30 mL reaction volume per
vessel). An overall yield of 79% of cinnamic acid 3a was
obtained. In a different run, cinnamic acid 3b was synthesized
from the corresponding aryl bromide in 56% yield. Both
yields closely agree with the yields obtained in the small-
scale microwave experiments (see above), demonstrating
once more the direct scalability of these reactions. Impor-
tantly, it was also possible to prepare both cinnamic acids
simultaneously in one single microwave irradiation experi-
ment by using all eight vessels in the rotor, charging four
Similarly, the Pd/C mediated coupling experiment for the
synthesis of 3a was performed without any problems on a 4
× 20 mmol scale providing an identical isolated yield (79%)
of cinnamic acid as compared to the homogeneous run.
D. Negishi Reactions. Another important carbon-carbon
bond forming reaction is the Negishi reaction, involving the
transition metal-catalyzed coupling of an arylhalide with an
arylzinc reagent.41 Despite the importance of this general
carbon-carbon bond forming reaction, reports on microwave-
induced Negishi couplings are limited.42,43 For the present
scale-up work, we have selected an example that was recently
reported on a 1.0 mmol scale in the literature with a single-
mode microwave reactor.42 The optimized reaction conditions
involved heating to 160 °C for 1 min in THF in the presence
of PdCl2(PPh3)2 as catalyst providing an isolated yield of
90% of the desired biaryl product 4 (Scheme 4). This
example was specifically selected to address the issue of very
short reaction times. In our hands, the previously published
small-scale method was reproducible (84% yield) despite the
fact that a different method for the preparation of the arylzinc
reagent was used (see Experimental Section for details).
The large-scale synthesis was carried out in two vessels
on a 2 × 20 mmol scale under an argon atmosphere in the
multimode batch reactor. The reaction was heated to 160
°C, employing a heating ramp of 2 min, and kept at 160 °C
for an additional minute, otherwise using identical reaction
conditions as in the published small-scale experiment.
Purification of the combined reaction mixtures by flash
chromatography provided biaryl 4 in 77% isolated yield.
E. Solid-Phase Synthesis. Solid-phase organic synthesis
(SPOS) shows several advantages compared with classical
protocols in solution. To accelerate reactions and to drive
them to completion, a large excess of reagents can be used,
as these can easily be removed by filtration and subsequent
washing of the solid support. In addition, SPOS can easily
be automated using appropriate robotics. Furthermore, SPOS
can be applied to the powerful split-and-mix strategy, which
turned out to be an important tool for combinatorial
chemistry.44 However, solid-phase organic synthesis exhibits
several shortcomings due to the nature of the heterogeneous
reaction conditions. Nonlinear kinetic behavior, slow reac-
tions, solvation problems, and degradation of the polymer
support due to the long reaction times are some of the
(37) Larhed, M.; Moberg, C.; Hallberg, A. Acc. Chem. Res. 2002, 35, 717-
727.
(38) (a) Rahim, E. H.; Kamounah, F. S.; Frederiksen, J.; Christensen, J. B. Nano
Lett. 2001, 1, 499-501. (b) Pratt, J.; Jimonet, P.; Bohme, G. A.; Boireau,
A.; Damour, D.; Debono, M. W.; Genevois-Borella, A.; Randle, J. C. R.;
Ribeill, Y.; Stutzmann, J.-M.; Vuilhorgne, M.; Mignani, S. Bioorg. Med.
Chem. Lett. 2000, 10, 2749-2754. (c) Mitra, J.; Mitra, A. K. Indian J.
Chem. 1992, 31B, 613-616. (d) Kogon, A. A.; Bochkarev, D. E.; Baskunov,
B. P.; Cheprakov, A. V. Liebigs Ann. Chem. 1992, 879-881. (e) Bumagin,
N. A.; Bykov, V. V.; Beletskaya, I. P. IzV. Akad. Nauk SSSR, Ser. Khim.
1990, 2672. (f) Alabaster, R. J.; Cottrell, I. F.; Hands, D.; Humphrey, G.
R.; Kennedy, D. J.; Wright, S. H. B. Synthesis 1989, 598-603.
(39) Heidenreich, R. G.; Ko¨hler, K.; Krauter, J. G. E.; Pietsch, J. Synlett 2002,
1118-1122 and references therein.
(41) (a) Negishi, E.; King, A. O.; Okukado, N. J. Org. Chem. 1977, 42, 1821-
1823. (b) Negishi, E. In Handbook of Organopalladium Chemistry for
Organic Synthesis; Negishi, E., Ed.; John Wiley & Sons: New York, 2002;
Vol. 1, p 229.
(42) O¨ hlberg, L.; Westman, J. Synlett 2001, 1893-1896.
(43) (a) Hayes, B. L. Abstracts of Papers, 223rd ACS National Meeting, Orlando,
FL, United States, April 7-11, 2002, ORGN-036. (b) See also ref 5a, pp
166-169.
(40) The Pd/C catalyst used in the experiments described in this publication is
characterized by a high palladium dispersion (36%), a low reduction degree,
and a high water content (E 105 CA/W 5% Pd, product of Degussa AG;
see ref 39 for details.)
(44) Nicolaou, K. C., Hanko, R., Hartwig, W., Eds.; Handbook of Combinatorial
Chemistry: Wiley-VCH: Weinheim, 2002.
Vol. 7, No. 5, 2003 / Organic Process Research & Development
•
711

Scheme 5
Scheme 6
that it is possible to perform such cycloadditions under
microwave conditions using 1,2-dichlorobenzene (DCB) as
solvent, saturated with ethylene. For 2-(1H)-pyrazinone 6,
for example, the ethylene addition was carried out in 140
min at 190 °C on a 0.2 mmol scale to provide 89% of the
final bis-lactam product 8.50 Interestingly, it was not possible
to further increase the reaction rate by raising the tempera-
ture. At temperatures above 200 °C, an equilibrium between
the cycloaddition 6 f 7 and the competing retro-Diels-
Alder fragmentation process was observed.50 To perform
microwave-assisted protocols utilizing gaseous reagents as
described herein, it would clearly be desirable to have a setup
where a microwave compatible reaction vessel could be
pressurized with a gaseous reagent before irradiating with
microwaves. For small-scale single-mode reactors,23,24 such
equipment is not available at the present time.51 The
multimode instrument described in the present study allows
the prepressurization of one or more individual reaction
vessels in the eight-position rotor.
For the experiment involving prepressurized ethylene
conditions, a solution of 1.0 mmol of 2-(1H)-pyrazinone 6
(Scheme 6) in 25 mL of DCB in a single quartz vessel was
first flushed with ethylene at atmospheric conditions and then
pressurized to ca. 7 atm with ethylene (see Experimental
Section for details). After irradiation at 190 °C for 30 min,
the solvent was removed under reduced pressure and the
crude cycloadduct 7 was hydrolyzed by treatment with dilute
NaOH at 70 °C for 30 min (microwave irradiation) to furnish
bis-lactam 8 in 77% isolated yield. Clearly, here the increased
pressure is driving the reaction along the desired pathway,
shifting the equilibrium to the cycloadduct side.
problems typically experienced in SPOS.45 Performing solid-
phase organic synthesis under microwave irradiation condi-
tions has become an increasingly popular technology.7 Many
examples of this high-speed technique have been reported
in the literature for small-scale reactions, in most instances
using cross-linked polystyrene resins as solid supports.
As a model reaction for solid-phase synthesis scale-up
under microwave conditions, we have chosen a recently
published protocol, describing the amination of chloro-
methylated polystyrene resin (Merrifield resin) with methyl-
amine at 150 °C within 5 min (Scheme 5).46 Interestingly,
this transformation was carried out in water as a solvent,
which is a highly unusual solvent for SPOS because of the
inferior swelling properties of most cross-linked resins in
water (in particular polystyrene resins).45
In contrast to the other scale-up work described in this
report, here dedicated PTFE-TFM reaction vessels have been
used instead of quartz vessels, to allow a more convenient
handling of the polymer beads. The amination of Merrifield
resin with commercially available aqueous methylamine
(40% w/w) according to Scheme 5 was carried out in four
vessels containing 5.0 g of resin each (0.84 mmol/g loading).
Microwave heating for 5 min at 150 °C provided the same
degree of amination (loading 0.74 mmol/g of amine func-
tionality as determined by combustion analysis), comparable
with the small-scale experiment.
F. Diels-Alder Cycloadditions with Ethylene. The
readily accessible and broadly functionalized 2-azadiene
system of the 2-(1H)-pyrazinone scaffold (e.g., 6, Scheme
6) has been shown to offer unique opportunities for inter-
and intramolecular cycloaddition reactions with electron-rich
and electron-poor dienophiles.47 A particularly interesting
cycloaddition route in the 2-(1H)-pyrazinone series involves
Diels-Alder cycloaddition reaction of the heterocyclic
pyrazinone heterodienes with ethylene, leading to bicyclic
cycloadducts of type 8, which can be elaborated further to
conformationally restricted dipeptide analogues, useful as
â-turn mimetics.48 Under conventional conditions, these
cycloaddition reactions have to be carried out in an autoclave
applying 25 atm of ethylene pressure before the setup is
heated to 110 °C for 12 h.49 We have recently demonstrated
Discussion
An overall analysis of the scale-up results presented in
this study demonstrates the general direct scalability of
microwave-assisted transformations optimized under small-
scale single-mode conditions to a larger scale, using multi-
mode batch reactors. By keeping the previously optimized
reaction parameters (time, temperature) constant, very similar
results in terms of reaction conversions and isolated product
yields/purities have been obtained. One critical point in
scaling up the volume of microwave-heated transformations
is the reproducibility of heating and cooling profiles. Since
the power density in single-mode microwave reactors is much
higher than those in multimode cavities,5a it is possible to
obtain very rapid heating (microwave flash heating) of
comparatively small volumes (1-5 mL) in these small-scale
reactors (with temperature increases of 10 °C s-1 not being
uncommon). In multimode cavities, it is generally not
possible to duplicate those heating profiles for larger volumes
(500 mL).52 With the current 1400 W multimode reactor
(49) Loosen, K. P.; Tutonda, M. G.; Khorasani, M. F.; Compernolle, F.;
Hoornaert, G. J. Tetrahedron 1991, 47, 9259-9268.
(50) Van der Eycken, E.; Appukkuttan, P.; De Borggraeve, W.; Dehaen, W.;
Dallinger, D.; Kappe, C. O. J. Org. Chem. 2002, 67, 7904-7909.
(51) For a very recent application of a small-scale prepressurized reaction (<2.5
bar) in a single-mode microwave reactor, see: Milijanic´, O. Sˇ.; Vollhardt,
K. P. C.; Whitener, G. D. Synlett 2003, 29-34.
(45) Do¨rwald, F. Z. Organic Synthesis on Solid Phase; Wiley-VCH: Weinheim,
2002.
(46) Westman, J.; Lundin, R. Synthesis 2003, 1025-1030.
(47) (a) Vekemans, J.; Pollers-Wiee¨rs, C.; Hoornaert, G. J. Heterocyclic Chem.
1983, 919-923. (b) Hoornaert, G. Bull. Soc. Chim. Belg. 1994, 103, 583-
589.
(48) De Borggraeve, W. M.; Rombouts, F. J. R.; Van der Eycken, E. V.; Toppet,
S. M.; Hoornaert, G. J. Tetrahedron Lett. 2001, 42, 5693-5695.
(52) Nu¨chter, M.; Mu¨ller, U.; Ondruschka, B.; Tied, A.; Lautenschla¨ger, W.
Chem.-Ing.-Tech. 2002, 74, 910-920.
712
•
Vol. 7, No. 5, 2003 / Organic Process Research & Development

system, it was nevertheless possible to follow the pro-
grammed heating ramps (see above), that is, reaching desired
reaction temperatures of up to 180 °C within 2-3 min, even
for runs involving eight reaction vessels with close to 500
mL of reaction volume. Cooling in single-mode microwave
reactors is achieved by active gas-jet cooling of the reaction
vessel by compressed air.23,24 For larger reaction vessels as
used in the present study, this technique is significantly less
effective and the rapid cooling times of typically 2-3 min
achieved on a small scale could not be reproduced. Typically
20-30 min were necessary to bring down the reaction
temperature to ca. 40 °C.53 However, this had no effect on
the specific chemistries presented herein.
The transformations described herein were conducted in
a parallel fashion. Therefore, rather than exposing one large
reaction vessel to microwave irradiation, up to eight vessels
can be placed in a rotor system and irradiated simultaneously.
Apart from avoiding potential problems related to microwave
penetration depth (see above), this method allows one to
conduct genuine parallel synthesis, that is, to perform eight
different chemical synthesis at the same time. To ensure
similar temperatures in a parallel setup, however, the same
amount of the identical solvent has to be used in each reaction
vessel due to the dielectric properties involved. Since the
maximum suggested filling volume of each the reaction
vessels is ca. 50-60 mL, a total volume of ca. 500 mL can
be processed. In addition, in case of an unexpected vessel
failure, only the contents of one vessel are lost and the risk
of an explosion is significantly reduced.
raphy was performed using silica gel 60 (0.040-0.063 mm,
Merck). Melting points were obtained on a Gallenkamp
melting point apparatus, model MFB-595 in open capillary
tubes. 1H NMR spectra were recorded on a Bruker AMX360
or AMX500 instrument in CDCl3 or DMSO-d6, operating
at 360 or 500 MHz, respectively, using TMS as reference.
On-bead FTIR spectra were recorded on a Unicam Galaxy
Series FTIR 7000 (Mattson Instruments Inc.) using mashed
resin beads in KBr pellets. Conventional IR spectra were
taken on a Perkin-Elmer 298 spectrophotometer in KBr
pellets. Mass spectra were obtained on a Hewlett-Packard
LC/MSD 1100 series instrument in the atmospheric pressure
chemical ionization (negative or positive APCI) mode.
Microanalyses were performed on a Carlo Erba 1106
elemental analyzer.
Small-Scale Microwave Chemistry. All reactions de-
scribed herein were initially run and carefully optimized on
a typically 1.0-4.0 mmol scale (max reaction volume 5 mL)
using an Emrys Synthesizer single-mode microwave reactor
(Personal Chemistry AB). Inside the cavity, the ca. 10 mL
reaction vials can be exposed to 20 bar of pressure and 250
°C. Temperature is measured by infrared thermometry on
the outer surface of the process vial. After the irradiation
period, the reaction vessel is cooled rapidly (approximately
1-2 min) to ambient temperature by compressed air (gas
jet cooling). Further details of this instrument have been
reported elsewhere.31
Large-Scale Microwave Synthesis. The experiments on
a large scale described herein (500 mL reaction volume) have
been carried out in a prototype multimode batch reactor
(Anton Paar GmbH, Graz).54 The instrument is equipped with
two magnetrons, operating at a frequency of 2.45 GHz with
continuous microwave output power from 0 to 1400 W. The
cavity (dimensions: W × D × H, 45 × 42 × 35 cm3) is
equipped with an eight-vessel rotor, employing either 100
mL PTFE-TFM or 80 mL quartz glass vessels (max filling
volume ca. 50-60 mL), both types dedicated for reactions
at high pressure (60 or 80 bar controlled pressure, respec-
tively) and temperatures (260 and 300 °C, respectively). The
PTFE-TFM vessels are inserted into a ceramic vessel jacket,
which provides structural strength and dimensional stability.
The ceramics, respectively, the quartz vessels rest inside
protecting air cooling jackets made of PEEK. Both types
are capped with special seals and a protective PEEK cap.
The seals comprise a release valve, which can be manually
operated and allows the prepressurization of the vessel via
a connection port with a syringe adapter. The individual
vessels are placed in the corresponding rotor, fixed by
screwing down the upper rotor plate, and finally the rotor is
closed with a protection hood (Figure 2). Accurate temper-
ature measurement is achieved by inserting a gas balloon
thermometer in one reference vessel. Additionally, the surface
temperature of all vessels can be monitored by IR thermog-
raphy. Pressure is monitored by a load-cell type simultaneous
hydraulic pressure sensing system for all vessels, monitoring
Conclusion
The feasibility of direct scale-up from small single-mode
microwave reactors to larger multimode systems has been
demonstrated. A number of different organic transformations
involving different solvents and reaction conditions have
been investigated. In all cases, it was possible to achieve
similar yields of products on varying reaction scales from 1
to 4 mmol (5 mL processing volume) to a ca. 100-500 mmol
scale (max 500 mL volume) without changing the previously
optimized conditions (direct scalability). All reactions were
repeated several times showing very small deviations in yield.
The prototype laboratory-scale multimode microwave reactor
used in the present study allows parallel processing in either
quartz or PTFE-TFM vessels with maximum operating limits
of 300 °C and 80 bar of pressure (quartz vessel). The system
allows magnetic stirring in all vessels, full on-line monitoring
of temperature, microwave power and pressure and, there-
fore, seems ideally suited for conducting microwave-assisted
transformations on up to a 1000 mL scale.
Experimental Section
General Methods. All starting materials were purchased
from commercial sources or were synthesized as described
in the cited references. TLC analysis was performed on
Merck precoated 60 F254 plates. Flash column chromatog-
(54) The prototype microwave reactor used for synthesis in this study is based
on the Multiwave 3000 microwave digestion system (www.anton-paar.com).
For applications and further details, see: Kettisch, P. Int. Labmate 2003,
27, 39-41.
(53) Previoulsy described microwave batch reactors possess a coldfinger inside
the reaction vessel for rapid cooling after completion of irradiation (see
refs 21 for details).
Vol. 7, No. 5, 2003 / Organic Process Research & Development
•
713

position rotor. The reaction mixtures were heated to 120 °C
employing a 3 min linear heating ramp and then irradiated
for an additional 10 min at 120 °C. After cooling to 40 °C
by an air flow (20 min), individual vessels were removed
from the rotor and kept at 4 °C for ca. 1 h. The precipitates
were filtered by suction and washed with cold EtOH to
provide DHPM 1a in 70%, 72%, 71%, and 74% yield,
respectively. In every case, the purity determined by HPLC
1
and H NMR was >98%. The combined yield was 59.80 g
1
(72%). H NMR (DMSO-d6) δ 1.12 (t, J ) 7.5 Hz, 3H),
2.28 (s, 3H), 4.03 (q, J ) 7.5 Hz, 2H), 5.17 (d, J ) 3 Hz,
1H), 7.22-7.41 (m, 5H), 7.78 (br s, 1H), 9.22 (br s, 1H).
The spectroscopic and analytical properties of this material
were in agreement with an authentic sample.31
Isopropyl 4-(3-Nitrophenyl)-2-oxo-1,2,3,4-tetrahydro-
pyrimidine-5-carboxylate (1b). This DHPM derivative was
synthesized in a similar way as described above for 1a, in
two, four, and eight quartz reaction vessels. For the eight
vessel run, all eight vessels were filled with the appropriate
reagents as described below: 4.80 g (80 mmol) of urea and
32 mL of EtOH were placed in a quartz vessel and stirred
for approximately 10-15 min to enable the urea to dissolve.
Subsequently, 12.10 g (80 mmol) of 3-nitrobenzaldehyde,
11.50 g (80 mmol) of isopropyl acetoacetate, and 1.50 mL
of HCl concd were added. All eight quartz reaction vessels
were sealed and inserted into the eight-position rotor. The
reaction mixtures were heated to 120 °C employing a 3 min
linear heating ramp and then irradiated for an additional 20
min at 120 °C. After cooling to 40 °C by an air flow (20
min), individual vessels were removed from the rotor and
the contents combined, treated with 150 mL of water and
stirred for ca. 30 min at room temperature. Subsequent
cooling at 4 °C for 1 h and filtration provided 98.1 g (48%)
Figure 2. View of the eight-position rotor system. Visible are
the protecting air cooling jackets made of PEEK containing
the quartz or PTFE-TFM vessels and the pressure release
valves. The protection hood is shown in the back. This rotor
system is placed in the cavity of the reactor.
the highest pressure level and pressure increase. Both,
pressure and temperature data are transferred to the micro-
wave oven wireless by an infrared LED on the rotor. The
ovens built-in electronics allows reaction control in a
temperature versus time mode. Alternative rotor types for
the simultaneous treatment of 16 vessels are available but
have not been used for this work. After irradiation, the rotor
is cooled to approximately 40 °C within 20 min by venting
air through cooling gaps which are surrounding the reaction
vessels. The system is equipped with various safety features/
devices, including a solvent detection system, and shut-down
mechanisms in case of a too rapid temperature or pressure
increase in one of the reaction vessels. Each vessel is
equipped with a pressure release seal that is designed to vent
at 120 bar of pressure, avoiding rupture of the vessel.
CAUTION: Performing chemical synthesis with organic
solvents in sealed vessels under pressure constitutes a safety
hazard. Proper care must be taken in the event of any
unexpected vessel rupture or explosion (exothermic reactions,
thermal runaways, arcing phenomena in electromagnetic
fields). In our synthetic attempts with the reactor system
described above, such experimental hazards were not expe-
rienced.
Ethyl 4-Phenyl-2-oxo-1,2,3,4-tetrahydropyrimidine-5-
carboxylate (1a). Four 80 mL quartz reaction vessels
equipped with Teflon-coated stirring bars (2.5 cm) were
individually charged with 4.80 g (80 mmol) of urea and 2.40
g (8 mmol) of FeCl3‚6H2O. To each of these vessels was
added 32 mL of a 3:1 (v/v) mixture of AcOH and EtOH,
followed by stirring at room temperature for approximately
10-15 min to enable the starting materials to dissolve.
Subsequently, 8.49 g of benzaldehyde (80 mmol) and 10.40
g (80 mmol) of ethyl acetoacetate were added. All four quartz
reaction vessels were sealed and inserted into the eight-
1
of DHPM 1b in >95% purity (HPLC). H NMR (DMSO-
d6) δ 1.00, 1.18 (2d, J ) 6.0 Hz, 6H), 2.28 (s, 3H), 4.84 (m,
1H), 5.31 (d, J ) 3.0 Hz, 1H), 7.65-7.77 (m, 2H), 7.91 (br
s, 1H), 8.09-8.19 (m, 2H), 9.38 (br s, 1H). The spectroscopic
and analytical properties of this material were in agreement
with an authentic sample.31
2-Chloro-N-[2-(4-flurophenyl)ethyl]thiobenzamide (2).
The reaction was carried out in four 80 mL quartz vessels,
each charged with 1.44 g (45 mmol) of elemental sulfur (S8),
suspended in 40 mL of NMP. Subsequently, 6.96 g (50
mmol) of 4-fluorophenethylamine and 5.62 g (40 mmol) of
3-chlorobenzaldehyde were added (in that order). The vessels
were sealed, put into the eight-position rotor system, and
inserted into the cavity. The reaction mixtures were heated
to 140 °C employing a 3 min linear heating ramp and then
irradiated for an additional 10 min at 140 °C. After
irradiation, the vessels were cooled to 40 °C within 20 min.
The dark brown solutions were combined, poured onto
crushed ice and stirred for 20 min. The formed precipitate
was filtered and purified by redissolving the crude material
in acetone, followed by filtration and reprecipitation with
ice-water. This protocol afforded 42.2 g (90%) of thioamide
1
product 2 in excellent purity (>98%, HPLC). H NMR
(CDCl3) δ 3.08 (t, J ) 6.5 Hz, 2H), 4.07 (q, J ) 6.0 Hz,
2H), 7.03-7.53 (m, 8H), 7.65 (s, 1H). The spectroscopic
714
•
Vol. 7, No. 5, 2003 / Organic Process Research & Development

and analytical properties of this material were in agreement
with an authentic sample.35
(79%) in excellent purity (>98%). Carrying out the large-
scale reaction under heterogeneous conditions on the same
scale (four vessels) led to an identical overall product yield
of 79%.
3-Cyano-trans-cinnamic Acid (3a). 1. Small Scale. (A)
Homogeneous Catalysis. In a typical procedure for homo-
geneous catalysis, 1.0 mol % of Pd(II) acetate (4.5 mg, 0.02
mmol), 2.2 mol % of tri(o-tolyl)phosphine (13.4 mg, 0.044
mmol), and 2.0 mmol (275 µL) of TEA were dissolved in
3.0 mL of MeCN and stirred for approximately 15 min in
the microwave process vial. Then 2.0 mmol (364 mg) of
3-bromobenzonitrile was added, and the mixture stirred for
an additional 5 min at room temperature. Finally, 2.0 mmol
(137 µL) of acrylic acid was added, and the vial was sealed
and subsequently irradiated at 180 °C for 15 min. After
cooling, the precipitate was filtered off and treated with 5
mL of water for an initial purification (removal of the TEA
HBr). The mother liquor was evaporated, and the residue
also treated with water to provide additional product. The
combined solid crude products were dissolved in EtOH and
filtered to remove residual Pd(0). Evaporation led to a
colorless, crystalline product in 82% yield (280 mg) and
excellent purity (>98%, HPLC), mp ) 230-232 °C (lit.55
3-Cyano-4-fluoro-trans-cinnamic Acid (3b). 1. Small
Scale. The reaction under homogeneous conditions is carried
out as described above for 3a on a 2.0 mmol scale, employing
3-bromo-4-fluorobenzonitrile (2.0 mmol, 400 mg) as starting
material. Identical workup lead to 210 mg (55%) slightly
greenish product in high purity (>95%, HPLC), mp ) 203-
205 °C; 1H NMR (DMSO-d6) δ 6.64 and 7.58 (2d, J ) 16.0
Hz, 1H), 7.56 (m, 1H), 8.12 (d, J ) 7.5 Hz, 1H), 8.32 (d, J
) 7.5 Hz, 1H), 12.58 (s, OH). MS (neg. APCI): m/z 190
(M-1). 2. Scale-Up Experiment. The reaction was carried
out as mentioned above under homogeneous conditions
employing four 80 mL quartz vessels on a 4 × 20 mmol
scale each according to the general protocol described above.
Identical workup led to 8.6 g (56%) of product 3b in high
purity (>95%).
4-(2-Cyanophenyl)benzaldehyde (5). 1. Preparation of
2 M Stock Solution of ZnBr2 in THF. Anhydrous ZnBr2
(Fluka 96465) was fused at 15 mbar for ca. 15 min in a
round-bottomed flask utilizing a heating gun. This reagent
(10.0 g, 44.4 mmol) was dissolved in 22 mL of dry THF
(distilled from sodium benzophenone ketyl) by vigorous
shaking of the flask. 2. Preparation of 0.38 M Stock
Solution of 2-Bromozinc Benzonitrile. A two-necked
round-bottomed flask was filled with a solution of 9.10 g
(50 mmol) 2-bromobenzonitrile in 40 mL of dry THF under
argon atmosphere. The flask was equipped with a stirring
bar, a thermometer, and a septum. The flask was immersed
into a Dewar vessel cooled by an ethanol/liquid N2 slush.
After cooling the solution to -100 °C, 35 mL (56 mmol) of
1.6 M n-BuLi in hexanes was added carefully, keeping the
reaction temp < -78 °C. The addition took approximately
30 min, and afterwards, the mixture was kept at -78° for
an additional 3 min. The solution was cooled again to -100-
110 °C, and 30 mL of 2 M ZnBr2 in THF was added slowly,
keeping the temp < -78 °C. After addition (∼30 min), the
flask was allowed to warm to ambient temp within ca. 30
min under vigorous stirring. A sample was taken from the
vigorously stirred biphasic mixture, quenched with diluted
HCl, and extracted with ether. The etheral extract was
analyzed by HPLC (210 nm), showing 76% conversion to
the desired 2-bromozinc benzonitrile. 3. Negishi Coupling.
The microwave scale-up was conducted in two 80 mL quartz
vessels, equipped with stirring bars. A melted mixture of
3.30 g (17.8 mmol) of 4-bromobenzaldehyde and 250 mg
(0.36 mmol) of PdCl2(PPh3)2 was poured under argon into
each vessel. Subsequently, both vessels were temporarily
sealed with rubber septa and charged with 50 mL (corre-
sponding to 19 mmol of 2-bromozinc benzonitrile) of
organozinc reagent using a syringe. Both vessels were
warmed (ca 50 °C) to ensure dissolution of the substrate.
The vessels were then permanently sealed and placed into
the eight-position rotor. The reaction was heated to 160 °C,
employing a heating ramp of 2 min, and kept at 160 °C for
additional 1 min. The cooling period required ca. 20 min to
1
236-238 °C). H NMR (DMSO-d6) δ 6.70 and 7.61 (2d, J
) 16.0 Hz, 1H each), 7.61 (t, J ) 8.0 Hz, 1H), 7.85 (d, J )
7.5 Hz, 1H), 8.03 (d, J ) 7.5 Hz, 1H), 8.22 (s, 1H), 12.6 (br
s, OH); MS (pos. APCI): m/z 174 (M + 1). (B) Heteroge-
neous Catalysis. Utilizing heterogeneous catalysis, the
catalytic system of phosphine ligand and Pd(II)-acetate was
substituted by 5.0 mg of Pd/C (5% Pd, 56% water, Degussa
AG),40 removing the need for prestirring. The reaction was
carried out under identical conditions as described above (180
°C, 15 min), and identical workup led to the desired product
in comparable yield (270 mg, 79%) and purity (>98%). The
spectroscopic and analytical properties of this material were
identical to the product isolated from a homogeneous run.
2. Scale-Up Experiment. In a typical procedure for homo-
geneous catalysis, 1.0 mol % of Pd(II) acetate (0.2 mmol,
45 mg), 2.2 mol % tri(o-tolyl)phosphine (0.44 mmol, 134
mg), and 20 mmol (2.75 mL) TEA were dissolved in 25
mL of MeCN and stirred for approximately 15 min in an 80
mL quartz reactor vessel. Then, 20 mmol (3.64 g) of
3-bromobenzonitrile were added, and the mixture stirred for
an additional 5 min. Finally, 20 mmol (1.37 mL) of acrylic
acid was added, and the vessel was sealed and inserted into
the eight-position rotor system. The rotor was equipped with
four vessels prepared according to the above protocol and
then inserted into the cavity. The reaction mixtures were
heated to 180 °C employing a 3 min linear heating ramp
and then kept for an additional 15 min at 180 °C. After
cooling to 40 °C within a period of 20 min, the vessels were
opened and the contents from individual vessels combined.
The formed precipitate was filtered, and the filtrate was
evaporated. The combined crude products were treated with
50 mL of water, stirred for 20 min, and then filtered by
suction. To eliminate traces of Pd, the residue was dissolved
in hot ethanol (approximately 50 mL per g) and filtered.
Evaporation of the solvent left 10.8 g of off-white product
(55) Dhurjati, M. S. K.; Sharma, J. A. R. P.; Desiraju, G. R. J. Chem. Soc.,
Chem. Commun. 1991, 1702-1703.
Vol. 7, No. 5, 2003 / Organic Process Research & Development
•
715

reach approximately 40 °C. The contents of both vessels were
combined, and after evaporation of the solvent, the crude
mixture was purified by silica gel flash column chromatog-
raphy, employing EtOAc/hexane 1:2.5 as eluent to yield 5.70
g (77%) of product in high purity (>98% by 1H NMR). 1H
NMR (CDCl3): δ 7.52-7.84 (m, 6H), 8.03 (d, J ) 7.0 Hz,
2H), 10.12 (s, 1H); FT-IR (KBr): 2215 cm-1 (CN), 2840-
2950 cm-1 (CH); MS (pos. APCI): m/z 208 (M + 1). The
spectroscopic and analytical properties of this material were
in agreement with an authentic sample.42
Methylaminomethyl Polystyrene Resin. Washed and
dried Merrifield resin (5.0 g, 0.84 mmol/g loading, Nova-
Biochem No. 01-64-0104, washed with H2O, THF/H2O,
DCM, and MeOH (20 mL each) before use) was filled,
respectively, into four dedicated 100 mL Teflon vessels and
suspended with 50 mL of commercially available aqueous
MeNH2 (40% w/w). The vessels were capped, coated with
ceramics tubes and a PEEK sheath, and inserted into the rotor
system. The rotor was closed and placed inside the cavity
of the reactor. The reaction mixtures were heated at 150 °C
for 5 min (no heating ramp) and afterwards cooled to ca. 40
°C within 20 min. The combined resins were washed 3 times
with H2O, DCM, and MeOH (20 mL each) and dried at 70
°C to furnish 20.0 g of methylaminomethyl polystyrene resin.
The loading was calculated to 0.74 g/mol based on the N
content determined by elemental analysis of 1.03%. The
reaction progress was also monitored by on-bead FTIR
measurement observing the C-Cl stretch at 1270 cm-1,
which decreased due to substitution of the chlorine func-
tionality.
the rotor, together with three dummy vessels, each filled with
25 mL of solvent. After the vessels were fixed, the reaction
vessel containing the pyrazinone precursor was prepressur-
ized with 7 bar of ethylene from a lecture bottle via the gas
inlet system of the vessel. Then the rotor was placed inside
the cavity and irradiated, employing a 1.5 min heating ramp
to reach 190 °C and additional irradiation time at 190 °C
for 30 min. After cooling to 40 °C (ca 20 min), the solvent
was removed by distillation under reduced pressure. The
crude cycloadduct 7 was suspended in 10 mL of 1 M NaOH/
dioxane (1:2) and irradiated at 70 °C for 30 min. After
cooling, the mixture was neutralized with 1 M HCl and
evaporated. The crude bis-lactam product 8 was dissolved
in CHCl3 and extracted with water. The organic layer was
separated and evaporated, and the residue was purified by
silica gel flash chromatography (DCM/ethyl acetate 1:2) to
1
furnish 206 mg (77%) of adduct 8 as a colorless solid. H
NMR (CDCl3): 2.52-1.94 (m, 4H), 4.97 (m, 1H), 7.25-
7.54 (m, 5H). The spectroscopic and analytical properties
of this compound were in agreement with authentic mate-
rial.50
Acknowledgment
This work was supported by the Austrian Science Fund
(P15582). We thank Anton Paar Gmbh (Graz, Austria) and
Personal Chemistry AB (Uppsala, Sweden) for providing the
microwave equipment and for helpful discussions on chem-
istry and instrumentation issues. D.D. wishes to thank
Personal Chemistry for a fellowship. Parts of this study were
also supported by Boehringer Ingelheim Austria (Vienna,
Austria).
Pyrazinone/Ethylene Diels Alder Reaction. Pyrazinone
648 (227 mg, 1.0 mmol) was placed in a 80 mL quartz vessel
and dissolved in 25 mL of 1,2-dichlorobenzene. The vessel
was flushed several minutes with ethylene and inserted into
Received for review June 17, 2003.
OP034075+
716
•
Vol. 7, No. 5, 2003 / Organic Process Research & Development