Scale-up of organic reactions in a pharmaceutical kilo-lab using a commercial microwave reactor

Article abstract of DOI:10.1021/op900269y

A range of pharmaceutically relevant reactions were investigated for scale-up in a kilo-lab environment using a commercial batch microwave reactor. Typical scale-up issues are discussed, taking into account the specific limitations of microwave heating in large-scale experiments. Examples of scale-up from 15 mL to 1 L are presented and demonstrate that the synthesis of compounds on greater than 100 g scale is feasible in one batch. Aided by this new technology reaction times have been significantly reduced and the productivity of our scale-up laboratory has been enhanced. Production rates of several hundred grams per day were achieved using microwave technology. The article concludes with a brief discussion of advantages and disadvantages of this type of batch microwave reactor.

Full text of DOI:10.1021/op900269y

Organic Process Research & Development 2010, 14, 650–656
Scale-Up of Organic Reactions in a Pharmaceutical Kilo-Lab Using a Commercial
Microwave Reactor
Hansjoerg Lehmann* and Luigi LaVecchia
Preparation Laboratories, Global DiscoVery Chemistry, NoVartis Institutes for BioMedical Research, Basel, Switzerland
Abstract:
industry where they are used for the high throughput synthesis
of chemical libraries.4,5
A range of pharmaceutically relevant reactions were investigated
for scale-up in a kilo-lab environment using a commercial batch
microwave reactor. Typical scale-up issues are discussed, taking
into account the specific limitations of microwave heating in large-
scale experiments. Examples of scale-up from 15 mL to 1 L are
presented and demonstrate that the synthesis of compounds on
greater than 100 g scale is feasible in one batch. Aided by this
new technology reaction times have been significantly reduced and
the productivity of our scale-up laboratory has been enhanced.
Production rates of several hundred grams per day were achieved
using microwave technology. The article concludes with a brief
discussion of advantages and disadvantages of this type of batch
microwave reactor.
Today, the most common application of microwave irradia-
tion is in the synthesis of new molecules on small scale for
medicinal chemistry purposes. However, microwave technology
is increasingly required in early scale-up phases, specifically
for the synthesis of intermediates and active compounds in larger
quantities. The question whether microwave technology could
be used for the scale-up of organic reactions from gram to
kilogram scale has become one of the major topics currently
discussed among industrial process development chemists.6-8
At present time the big challenge in this area is to establish a
reliable and safe process setup, where typical scale-up issues,
for example the limited penetration depth, a reliable temperature
control and a suitable reactor design, are carefully considered.8,9
Small-scale organic synthesis is traditionally conducted in
batches, and the majority of commercial microwave reactors
also operate in this way, utilizing vessels generally below a 100
mL volume. The most restricting factor for scale-up is the
limited penetration depth of the microwave field, which is
dependent on the dielectric properties of the microwave
absorbent reaction mixture, and is usually in a range of a few
centimetres. The maximum size of a batch reactor, that can be
simultaneously heated by standard magnetrons, is therefore
approximately 2-3 L. Despite the fact that some commercial
microwave batch reactors provide a relatively large power
output of more than 1000 W, chemists have not yet been able
to achieve the same steep heating and cooling profiles for litre
volumes as those observed for small-scale experiments. In order
to overcome these problems, stop-flow3c and continuous-flow
reactors3d have been developed by a number of providers.
Although being of advantage in terms of safety and productivity,
the major drawback of a continuous-flow reactor is the well-
reported fact that they are, in most cases, incompatible with
heterogeneous mixtures.8–10
Introduction
In the past few years, microwave heating has become a
widely accepted technique for organic synthesis, used both in
academic and industrial laboratories.1 The major driving force
for the widespread application of this innovative enabling
technology, is that many microwave reactions experience a
remarkable rate acceleration when compared to those with
conventional heating. Such reactions are frequently coupled with
increased yields and improved purity profiles.2 For the last 10
years, dedicated microwave reactors for small-scale synthesis
in research laboratories have been commercially available
(typically in the range of few hundred milligrams to some
grams) and offer a number of advantages.3 Reaction mixtures
can be rapidly heated under pressure in sealed vessels to
temperatures above 200 °C, thus shifting reaction times from
hours to minutes. Modern microwave reactors additionally allow
controlled parallel or automated sequential processing of a large
number of experiments and lead to significantly higher pro-
ductivity. The use of automated microwave reactors also enables
the rapid optimization of reaction conditions, and more recently,
such devices have found their way into the pharmaceutical
(4) (a) Kappe, C. O.; Kremser, J.; Stadler, A. J. Comb. Chem. 2007, 9,
285–291. (b) Kappe, C. O.; Dallinger, D. Nat. ReV. 2006, 5, 51–63.
(5) Wolkenberg, S. E.; Shipe, W. D.; Lindsley, C. W. Drug DiscoVery
Today 2005, 2, 155–161.
* To whom correspondence should be addressed. Telephone: +41 61 69
61607. Fax: +41 61 69 68757. E-mail: hansjoerg.lehmann@novartis.com.
(1) For reviews about microwave-assisted organic synthesis (MAOS) see:
(a) Loupy, A., Ed. MicrowaVes in Organic Synthesis, 2nd ed.; Wiley-
VCH: Weinheim, 2006. (b) Kappe, C. O.; Stadler, A.; MicrowaVes in
Organic and Medicinal Chemistry; Wiley-VCH: Weinheim, 2005. (c)
Hayes, B. L. MicrowaVe Synthesis: Chemistry at the Speed of Light:
CEM Publishing: Matthews, NC, 2002. (d) Lidstro¨m, P.; Tierney, J. P.
MicrowaVe Assisted Organic Synthesis; Blackwell: Oxford, 2005.
(2) Kappe, C. O. Angew. Chem., Int. Ed. 2004, 43, 6250.
(6) Moseley, J. D.; Lenden, P.; Lockwood, M.; Ruda, K.; Sherlock, J.;
Thomson, A. D.; Gilday, J. P. Org. Process Res. DeV 2008, 12, 30–
40.
(7) Lehmann, H. In New AVenues to Efficient Chemical Synthesis;
Seeberger, P. H., Blume, T., Eds.; Springer-Verlag: Berlin, 2007.
(8) Wolkenberg, S. E.; Shipe, W. D.; Lindsley, C. W.; Guare, J. P.;
Pawluczyk, J. M. Curr. Opin. Drug DiscoVery DeV. 2005, 8, 701–
708.
(9) Kappe, C. O.; Kremser, J.; Stadler, A. Top. Curr. Chem. 2006, 234–
278.
(3) (a) http://www.anton-paar.com. (b) http://www.biotage.com. (c) http://
www.cem.com. (d) http://www.milestonesrl.com.
(10) Wilson, N.; Sarko, C.; Roth, G. Org. Process Res. DeV. 2004, 8, 535–
538.
650
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development
10.1021/op900269y  2010 American Chemical Society
Published on Web 04/01/2010

A further, more general, obstacle that discourages chemists
from using these instruments for chemical synthesis is that, to
date, reactions in medicinal chemistry are normally performed
in a batch mode, and transferring them into a continuous process
requires additional time and optimization effort. It is therefore
not surprising that the number of publications which describe
the use of continuous flow reactors for synthesis on a 100 g
scale or more is still very small. Wilson et al. used a
noncommercial glass coiled flow cell which was introduced into
the microwave cavity of a Emrys Optimizer to investigate
nucleophilic aromatic substitution reactions, a Suzuki coupling,
and an esterification on a 10 g scale.10 They also reported that
crystallization of the product and the formation of solid particles
clogged the lines and frits, thus significantly reducing the utility
of such a tool. Leadbeater et al. had a similar experience when
they scaled up a range of organic reactions to hundreds of grams
using a commercially available continuous flow microwave
reactor.11 They concluded that scale-up from small sealed tubes
to continuous processing is feasible, as long as reaction mixtures
are homogeneous and do not contain solid particles.
A comparison of various commercially available microwave
reactors with focus on scale-up issues was recently published
by Moseley et al.6 In their study, all three concepts (batch-mode,
stop-flow, and continuous-flow) were tested by thorough
comparison of a rearrangement reaction. They showed that the
devices tested can be used for converting litres of reaction
mixtures to yield products in a range from 50 to 200 g.
However, a further investigation of a stop-flow microwave
reactor revealed that the same limitation of clogging also exists
for a stop-flow concept and that such a system works best when
only homogeneous solutions are converted in multiple batches.12
long reaction times (e.g. >8 h). In such a case, decreasing the
reaction time from hours to minutes often leads to an improved
work-flow and to higher productivity.
The second class of reactions is composed of those requiring
temperatures above 150 °C which normally have to be
performed in high-boiling solvents such as DMSO, DMF, or
NMP. In some cases, it is very laborious and time-consuming
to remove these high-boiling solvents during the workup in
order to isolate the product. Replacing these “work-up unfavor-
able” solvents by solvents which are “work-up friendly” can
help to avoid these kinds of problems as demonstrated by an
earlier work.13
Reactions which require more extreme conditions in terms
of temperature, or which do not result in high enough yield or
purity under conventional conditions, can be categorized in a
third class. As a consequence, chemists are not able to utilize
these reactions when using standard lab equipment. The
application of microwave technology expands the chemical
space which can be explored in the search for new interesting
compounds. The conversion of phenols into aromatic thiols by
the Newman-Kwart rearrangement,14 which in some cases
requires temperatures above 250 °C, is an interesting example
which falls into this category. A second example is the valuable
ortho-Claisen rearrangement, which normally requires temper-
atures above 200 °C.15 In many cases, the use of microwave
equipment instead of an oil bath for ortho-Claisen rearrangement
is more convenient and offers significant advantages.16-18
A more recently explored example of chemical scope
expansion are microwave-assisted organic reactions which are
performed in water, under the so-called “near-critical” conditions
reached in temperatures between 270 and 300 °C. Several useful
transformations, such as the hydrolysis of esters and amides,
Diels-Alder cycloadditions, pinacol rearrangements and the
Fischer indole synthesis were successfully performed under
these special near-critical conditions.19
Microwave Heating and Scale-Up of Batch Reactions.
One of the first applications of a multimode parallel batch
reactor for scale-up of microwave reactions from 1 to 100 mmol
scale was published by Kappe et al. some years ago.20 They
found that for a range of reactions it was possible to achieve
similar yields both 5 and 500 mL scale. However, for the
experiments they reported, a prototype reactor with 8 reaction
vessels was used, thus limiting the reaction volume to 500 mL.
At that time, they did not have access to of the commercially
available machines with a total volume >1 L which would be
available for use today.
Results and Discussion
Specific Requirements of a Scale-Up Laboratory. Syn-
thesis work in a scale-up laboratory is traditionally performed
in a batch-wise manner using conventional thermal heating
techniques under atmospheric pressure. The limiting factor for
the reaction temperature under such conditions is the boiling
point of the solvent. In contrast to this, microwave heating in a
pressurized vessel allows access to reaction temperatures far
above the boiling point of the solvent. The superheating, thus
occurring, leads to remarkably shortened reaction times as well
as higher yields and reduced byproduct formation.13 When
choosing a reaction solvent, one can consider not only boiling
point but also more favorable workup and purification proper-
ties, thus avoiding time-consuming workup procedures. Solvent
choice is particularly important in scale-up and can lead to
shorter processing times for a chemical synthesis and, therefore,
to a higher productivity.
We typically begin our scale-up process with a trial on small
scale (5-20 mL) in a Biotage Initiator microwave reactor,
(14) (a) Moseley, J. D.; Sankey, R. F.; Tang, O. N.; Gilday, J. P.
Tetrahederon 2006, 62, 4685–4689. (b) Moseley, J. D.; Lenden, P.
Tetrahedron 2007, 63, 4120–4125.
For chemists working in a scale-up laboratory, the use of
microwave technology offers particular advantages in three main
classes of chemical reactions. The first class consists of reactions
which require moderately high temperatures (e.g. >80 °C) and
(15) Martin Castro, A. M. Chem. ReV. 2004, 104, 2939.
(16) Leadbeater, N. E.; Bowman, M. D.; Schmink, J. R.; McGowan, C. M.;
Kormos, C. M. Org. Process Res. DeV. 2008, 12, 1078–1088.
(17) Kappe, C. O.; Kremser, J. J. Org. Chem. 2006, 71, 4651–4658.
(18) Kappe, C. O.; Kremser, J.; Razzaq, T. J. Org. Chem. 2008, 73, 6321–
6329.
(11) Leadbeater, N. E.; Bowman, M. D.; Holocomb, J.; Kormos, C. M.;
Williams, V. A. Org. Process Res. DeV. 2008, 12, 41–57.
(12) Moseley, J. D.; Woodman, E. K. Org. Process Res. DeV. 2008, 12,
967–981.
(19) Kappe, C. O.; Dallinger, D. Chem. ReV. 2007, 107, 2563–2591.
(20) Kappe, C. O.; Stadler, A.; Yousefi, B. H.; Dallinger, D.; Walla, P.;
Van der Eycken, P.; Kaval, N. Org. Process Res. DeV. 2003, 7, 707–
716.
(13) Lehmann, H.; La Vecchia, L. J. Assoc. Lab. Autom. 2005, 10, 412–
417.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
651

Table 1. Comparison between ethanol heated to 150 °C in
temperature rise is not a linear function of the irradiation time.
Above 100 °C, acetone and other solvents such as ethanol and
methanol become more transparent to microwave irradiation,
and less energy is converted into heat.21 In total, 18.5 min were
required to heat this volume to 180 °C. Even more time is
required to remove the energy after the heating process is
stopped. As shown in Figure 1, about 37 min were required to
cool a volume of 1.12 L of acetone to 50 °C. In summary, the
time which is required to perform a reaction on a 1 L scale is
mainly defined by the heating and cooling processes which are,
especially for short reaction times, the dominating parts of the
reaction process, an observation which is significantly different
from the result of Kappe et al. who reported that it was possible
to reach reaction temperatures of 180 °C in 2-3 min on a close
to 500-mL scale.20
Examples for Optimization and Scale-Up. In the course
of our daily business in the kilo-lab, we are frequently faced
with the preparation of a number of intermediates or drug
candidates on a larger scale in which at least one step of the
original synthesis was performed in a microwave reactor. In
order to increase the productivity in this early phase, it is of
great importance to reduce the time which is required to produce
these compounds on a larger scale. As previously mentioned,
reactions requiring many hours or even days at elevated
temperature are particularly interesting targets for improvement
of the work-flow by the use of microwave technology.
The cyclization of a γ-ketoester with hydrazine hydrate in
Scheme 1 was part of a reaction sequence submitted to our
kilo-lab for scale-up to 100 g. In the literature this reaction is
described with a yield of approximately 60%.22 Under conven-
tional conditions (refluxing in ethanol for 4 h) we obtained
product 2 in 53% yield. By performing the reaction in a Synthos
3000 microwave reactor at 140 °C for 20 min, we were able to
improve the yield to 64% on a 20 g scale. Further scale-up to
a batch size of 900 mL (130 g in 16 vials filled each with ∼60
mL) under the same conditions (140 °C/20 min) requiring no
additional optimization, gave 73 g of product in a yield of 63%.
With regard to productivity, this procedure allows the production
of 500-600 g per 8 h working day.
two different-scale microwave machines
Biotage Initiator
Synthos 3000
reaction volume
20 mL
300 W
15 W/mL
3 min
16 × 60 mL ) 960 mL
max. power
1400 W
ratio power/volume
time to heat to 150 °C
time to cool to 50 °C
1.46 W/mL
10 min
30 min
5 min
which provides a so-called focused monomode microwave field
and a power of 300 W. For synthesis on a larger scale, a Synthos
3000 Multiwave reactor is used. The Synthos machine is a
multimode reactor capable of providing a maximum irradiation
power of 1400 W and a maximum reaction volume of
approximately 1.2 L, distributed between 16 vials. When we
look at the heating process in more detail, it is obvious that
rapid heating and, even more importantly, rapid cooling strongly
depend on the ratio of volume to surface area of the reaction
mixture. Of course, this ratio changes significantly when a
reaction is scaled from a 20-mL scale to a 1000-mL scale. As
a consequence, one should expect a longer heating period for
the bigger batch reaction compared to a reaction performed on
smaller scale. Although the volume of 1 L is distributed between
16 100-mL vials in a Synthos 3000 reactor, the same step
heating rates as on small scale are rarely observed. A compari-
son of relevant parameters for a typical scale-up experiment is
shown in Table 1.
The Biotage Initiator provides a maximum power of 300
W, and when a 20-mL vial is used to perform a reaction, the
ratio of available power per volume of reaction mixture is 15
W/mL. Heating a 20-mL sample of ethanol to 150 °C in the
Biotage Initiator required 3 min. The reactor required about 5
min to cool the sample back down to 50 °C.
In a second experiment using the Synthos 3000 Multiwave
microwave, 960 mL (16 vials each filled with 60 mL) of ethanol
were heated to 150 °C. The two magnetrons of the Synthos
3000 provide a maximum power of 1400 W. Correlated with
the volume of 960 mL, a much lower ratio of power per volume
of 1.46 W/mL is obtained. Nevertheless, only 10 min were
required to heat the 48-fold amount of ethanol to 150 °C, which
corresponds to an increase in the heating period by a factor of
3.3. To cool this volume down again to 50 °C, 30 min were
required, which is 6 times longer compared to that for the
Biotage Initiator.
Scheme 1. Microwave-promoted cyclization of a γ-ketoester
with hydrazine hydrate
When we look in more depth into the heating process (Figure
1) it becomes obvious that microwave heating, especially for
batches of 1 L volume, has its own limitations. A simple test,
where we heated 1.12 L of acetone in 16 vials to 180 °C using
a constant power irradiation of 1400 W, revealed that the
A well-known standard procedure in organic chemistry to
form pyrazoles is the reaction of ꢀ-ketonitriles with hydrazines.
The reaction shown in Scheme 2 is part of the synthesis of a
p38 MAP kinase Inhibitor, which is described in literature with
(21) The ability of a solvent to absorb microwave energy is related to three
main dielectric parameters: tangent delta, dielectric constant, and
dielectric loss. Since these parameters are dependent on the temper-
ature, the ability to absorb microwave energy changes with the
temperature, and a lot of common solvents show a decrease in
microwave susceptibility with temperature rise.
Figure 1. Temperature profile when 1.12 L (16 × 70 mL) of
acetone is heated to 180 °C.
(22) Schenker, E.; Salzmann, R. Arzneim.-Forsch. 1979, 29, 1835–1843.
652
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development

Scheme 2. Microwave-promoted aminopyrazole synthesis
Scheme 3. Microwave-promoted nucleophilic substitution of
a chloropyrimidine
6-bromoquinazin-4-one 7 (Scheme 4) as an intermediate in the
synthesis of biologically active compounds.25-27 A common
method for the formation of the 3H-quinazolin-4-one ring is
based on the Niementowski reaction which involves the
condensation and cyclization of anthranilic acid derivatives with
formamide.28 Such a procedure usually requires high temper-
atures (150-180 °C) and is coupled with lengthy and tedious
workup conditions.29 On bench scale, microwave-assisted
reactions gave yields varying from 55% to 75%. Similar
procedures using microwave technology are described in
literature, however, with examples on a milligram scale.26,29 We
tested these conditions (150 °C, 20 min) on a 5 g scale using
our microwave equipment but found that the reaction did not
go to completion. Analysis of the reaction mixture by HPLC
revealed about 55% conversion to product with about 45% of
remaining starting material. On performing a scale-up experi-
ment we found that the reaction proceeded in a fast and very
clean manner at 180 °C with a reaction time of 30 min.
Although it is reported that formamide may decompose above
170 °C to form ammonia, the highest pressure measured inside
the vials was 2.3 bar. Since the boiling point of formamide is
210 °C, this pressure reflects the formation of water during the
reaction or a slow decomposition of formamide and the
formation of ammonia.30 After executing a simple workup
procedure (the reaction mixture was poured into water, the
precipitated product was filtered and dried), we isolated
compound 7 in 89% yield and a purity of >99% on a 200 g
scale in one single batch.
a yield of 60-85% after refluxing in toluene for 24 h.23 In the
course of our investigations we replaced toluene with methanol
and obtained a yield of 86% after 16 h at reflux. Under
microwave conditions (130 °C, 20 min) the reaction of
p-tolylhydrazine and pivaloylacetonitrile was scaled up by a
factor of 4 (entry 1, Table 2) to give pyrazole 4a in 85% yield
(entry 2, Table 2). Furthermore, the required processing time
(from the start of the reaction until the end of the cooling down
period) was reduced significantly from 16 h to 55 min, allowing
the production of multihundred grams per day. Similar results
were obtained for the cyclization of 3,4-dimethyl phenylhydra-
zine with pivaloylacetonitrile to give pyrazole 4b (entries 3 and
4, Table 2). Yields for this reaction were almost the same for
both microwave and conventional heating, but due to the much
shorter processing time under microwave irradiation, we were
able to obtain nearly 200 g of product in less than one hour.
By preparing 4 batches during an 8 h working day, it is possible
to produce 800 g of this compound in high quality. When we
tried to convert the deactivated 4-cyanophenylhydrazine into
pyrazole 4c under conventional conditions, we found that the
reaction was still not complete after 16 h of reflux in methanol.
However by using our microwave equipment, we were able to
push the reaction to completion. Scaled-up to a 100 g scale,
we isolated the pyrazole 4c in a yield of 77% after 25 min
reaction time at 140 °C (entry 6, Table 2, and Scheme 2).
These examples clearly show that a shift of the reaction
temperature from 16 h to 15 min by application microwave
conditions under pressure allows significant enhancement of
the productivity. To produce some hundred grams of a product
under conventional heating requires at least 16 h, independent
of the scale. However, the use of a batch microwave with 1 L
reaction volume enables a skilled person to run several batches
sequentially per day and to produce several hundred gram within
an 8 h working day.
Compared to other published procedures in which the
synthesis of compound 7 is performed in two steps using various
reagents (e.g., ammonia, EDC ·HCl, HOBt), solvents (e.g.,
DMF), and a relatively complicated multistep workup,25 our
scale-up approach fulfills several principles of green chemistry
and can be regarded as environmentally benign.31 Our reaction
protocol is performed under solvent-free conditions; for workup,
only water is used, and the amount of chemical waste is
minimized. The reaction is efficient with regard to atom
economy; the only byproduct of the reaction is water. The use
Substitution of halogens on heteroaromatic rings is a
common way to introduce new functionalities. Diamino pyri-
midine 6 in Scheme 3 was required on a 100 g scale as an
intermediate for the synthesis of purines and pteridines. In the
literature, this exchange was performed on a 5 g scale using
ammonia in ethanol in a sealed tube under pressure for 6 h at
125-130 °C with a yield of 76%.24 After a couple of trials on
small scale using our Synthos 3000 equipment, we found
suitable microwave conditions of 170 °C for 180 min and
obtained the desired product on a 120 g scale in 83% yield and
high purity.
(25) Hayashi, H.; Tobe, M.; Isobe, Y.; Tomizawa, H.; Nagasaki, T.;
Takahashi, H.; Fukazawa, T. Bioorg. Med. Chem. 2003, 11, 383–391.
(26) Besson, T.; Alexandre, F.; Berecibar, A.; Wrigglesworth, R. Tetra-
hedron 2003, 59, 1413–1419.
(27) Liu, L.; Yuan, H.; Liu, H.; Chen, S.; Wu, Y. Bioorg. Med. Chem.
Lett. 2007, 17, 6373–6377.
(28) von Niementowski, S. J. Prakt. Chem. 1895, 51, 564–572.
(29) Besson, T.; Alexandre, F.; Berecibar, A. Tetrahedron Lett. 2002, 43,
3911–3913.
(30) Besson, T.; Nouira, I.; Kostakis, I. K.; Dubouilh, C.; Chosson, E.;
Iannelli, M. Tetrahedron Lett. 2008, 49, 7033–7036.
In the course of our work as a support function for medicinal
chemistry, we were asked to produce a larger quantity of
(31) (a) Varma, R. S.; Polshettiwar, V. Chem. Soc. ReV. 2008, 37, 1546–
1557. (b) Strauss, C. R. Aust. J. Chem. 2009, 62, 3–15. (c) Andrews,
I.; Cui, J.; DaSilva, J.; Dudin, L.; Dunn, P.; Hayler, J.; Hinkley, B.;
Hughes, D.; Kaptein, B.; Kolis, S.; Lorenz, K.; Mathew, S.; Rammeloo,
T.; Wang, L.; Wells, A.; White, T.; Xie, C.; Zhang, F. Green Chemistry
Highlights. Org. Process Res. DeV. 2009, 13, 397–408, and literature
cited therein.
(23) Regan, J.; Breitfelder, S.; Cirillo, P. J. Med. Chem. 2002, 45, 2994–
3008.
(24) Bendich, A.; Russel, P.; Fox, J. J. Am. Chem. Soc. 1954, 76, 6073–
6077.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
653

Table 2. Comparison of aminopyrazole 4a-c formation in methanol
entry
product
scale
reaction temp., time
yielda (%)
isol. product (g)
1
2
3
4
5
6
4a: R1 ) Me, R2 ) H
4a: R1 ) Me, R2 ) H
4b: R1 ) R2 ) Me
20 g, 200 mL
82 g, 800 mL
8.6 g, 100 mL
128 g, 900 mL
2 g, 15 mL
reflux, 16 h
86
28.8
117
12.8
192
n.d.
MW, 130 °C, 20 min
85
reflux, 16 h
91.5
92
4b: R1 ) R2 ) Me
MW, 130 °C, 15 min
reflux, 16 h
4c: R1 ) CN, R2 ) H
4c: R1 ) CN, R2 ) H
n.d.b
77
102 g, 800 mL
MW, 40 °C, 25 min
128.5
a Isolated yield. b Only 70% conversion to product detected by HPLC.
Scheme 4. Microwave-assisted synthesis of
6-bromoquinazin-4-one 7
reactions which are otherwise difficult to perform under
conventional conditions. Especially when considering the
synthesis of compounds on a 100-g scale, the use of a
microwave reactor supports the development of environmentally
benign processes and green chemistry.
At the present time, none of the three strategies that have
evolved for scale-up (batch, stop-flow, and continuous flow)
offers solutions to all the scale-up-related problems that chemists
face.6,9 For a microwave batch reactor, the penetration depth is
still a limiting factor, and the reaction volume of commercially
available microwave reactors does not exceed the 1-2 L range.
With regard to a stop-flow or continuous flow regime, the
handling of suspensions and precipitating products which may
block the lines and valves, is still a severe limitation that hinders
the broad use of these tools for the scale-up of organic reactions.
Nevertheless, we have demonstrated that a microwave batch
reactor, for example the Synthos 3000, is a useful tool for the
scale-up of organic reactions and suitable for the delivery of
several hundred grams per day to meet medicinal chemistry
needs. We were also able to show that our productivity and
output can be significantly increased by using microwave
technology. With regard to true process development and scale-
up towards multikilogram scale we conclude, along with others
in this field, that there is presently no commercially available
microwave solution that fulfils all the requirements without any
drawbacks.6
of microwave heating and the shortened reaction time lead to
a better energy efficiency. Due to the cleanliness of such a
reaction and to the high yield, no additional purification is
required, thus avoiding the formation of additional waste.
Summary
The scale-up of a range of organic reactions using a
commercially available microwave reactor was investigated,
with special focus on handling, workflow, and typical scale-up
issues. The Synthos 3000 batch mode reactor provides a
relatively large reaction volume (maximum 16 vials which can
be filled with up to 70 mL each), allows high temperature and
pressure (240 °C/40 bar), and proved to be very robust. The
magnetic stirring system allows the handling of suspensions;
however, there is way of controlling whether stirring is efficient
in each vial. In some cases we experienced variability in
conversion in parallel vials because the magnetic stirrer bar was
blocked during the reaction or stirring did not work efficiently
enough. Most of the time the reaction conditions applied on a
15-mL scale in a Biotage Initiator unit were easily transferred
to our Synthos 3000 microwave reactor and led to comparable
results. Only in a few cases was additional optimization
required, and this was mostly limited to moderate adjustments
in reaction temperature or reaction time.
In general, the results from small-scale trials were reproduced
on a larger scale without any significant optimization effort.
Nevertheless, filling and removal of all 16 of the vials is
laborious, and the cooling period even for large reaction
volumes is comparatively long. A further disadvantage with
regard to the production of larger quantities is the fact that it is
not possible to run several batches sequentially in an automated
manner.
Experimental Section
General Experimental. All reagents were obtained com-
mercially and used as received unless otherwise noted. TLC
was performed on Merck silica gel plates 60 F-254, Art. Nr.
5729 with detection by UV (254 nm).
Reverse phase HPLC analyses were performed on an
Agilent-1100 machine using a Macherey-Nagel CC 70/4
Nucleosil 100-13 C18 HD column, acetonitrile and water (both
containing 0.05% TFA), a column temperature of 35 °C, with
a flow rate of 1.0 mL/min, and measuring at 216 nm. The
standard gradient used was 20-100% MeCN over 6 min, 100%
MeCN for 3 min followed by 100-20% MeCN over 0.5 min.
NMR spectroscopy was performed using a 400 MHz Varian
In many scale-up experiments, we experienced the use of
microwave technology leading to a significant decrease in the
reaction time and in some cases to less byproduct formation
and higher yield. Microwave technology allows us to optimize
a reaction with special focus on workup and purification,
independent of the required reaction temperature and the boiling
point of the solvent. Furthermore, the possibility to reach
temperatures greater than 200 °C significantly expands the
chemical scope of organic chemistry and allows us to run
1
instrument, AS 400 Oxford; H shifts were referenced to
DMSO-d6 at 2.49 ppm and CDCl3 at 7.25 ppm with tetra-
methylsilane as internal standard for 1H NMR. Melting points
were measured using a Bu¨chi, B-545 instrument.
Equipment Used. Reactions on small scale were performed
in the commercial Biotage Initiator Sixty monomode microwave
unit with IR temperature monitoring and noninvasive pressure
transducers, using 20-mL quartz vials.
654
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development

Scale-up experiments were performed in a commercial
Synthos 3000 Multiwave microwave unit. We used the con-
figuration with 16 100-mL PTFE lined tubes in ceramic cases,
allowing a maximum load of ∼70 mL per tube. Temperature
and pressure were monitored continuously in a reference tube
by an internal gas balloon thermometer; 240 °C and 40 bar are
reachable. In addition, all vials were monitored by an external
IR sensor.
Experimental Procedures for the Reactions Performed
in this Study. Preparation of 3-Oxo-3,4,4a,5,7,8-hexahydro-
2H-pyrido[4,3-c]pyridazine-6-carboxylic Acid Ethyl Ester (2)
under ConVentional Conditions. To a solution of 3-ethoxycar-
bonylmethyl-4-oxo-piperidine-1-carboxylic acid ethyl ester 1
(20 g, 78 mmol) in ethanol (60 mL) were added hydrazine
hydrate (3.9 g, 78 mmol) and acetic acid (7.8 mL, 136 mmol)
at ambient temperature. The mixture then was heated to reflux
for 4 h. After cooling to ambient temperature, the reaction
mixture was concentrated in Vacuo, and the remaining solid
was triturated in ethanol (40 mL). The product was filtered and
dried in a vacuum oven at 60 °C to yield the title compound 2
as an off-white crystalline solid (9.4 g, 54%). HPLC (tR 4.33
min, 99%); mp 166-167 °C (lit.22 167-169 °C).
Preparation of 3-Aminopyrazoles 4a-c on Large Scale in
Anton Paar Synthos 3000. In a typical procedure, p-tolylhy-
drazine hydrochloride (82 g, 517 mmol) and pivaloylacetonitrile
(71.2 g, 569 mmol) were dissolved in methanol (650 mL).
Sixteen vials were filled with 50 mL each of this solution and
placed in the rotor. The reaction mixtures were heated with
magnetic stirring to 130 °C over 5 min with 1400 W available
power, held for 20 min at 130 °C, and then cooled by air
ventilation to 50 °C over 30 min. The contents of all tubes were
combined, and solvent was evaporated in Vacuo. The remaining
solid was partly dissolved in hot DCM (100 mL) and precipi-
tated again by addition of diisopropyl ether (100 mL). The
suspension was cooled to 0 °C, stirred for 20 min, and filtered.
After filtration and drying 5-tert-butyl-2-p-tolyl-2H-pyrazol-3-
ylamine 4a was isolated as an off-white solid (117 g, 85%).
1
HPLC (tR 3.74 min, 100%); mp 196 °C dec; H NMR (400
MHz, DMSO-d6) δ 1.29 (s, 9H), 2.38 (s, 3H), 5.67 (s, 1H),
7.36-7.50 (m, 4H).
Preparation of 5-tert-Butyl-2-(3,4-dimethyl-phenyl)-2H-pyra-
zol-3-ylamine (4b). 3,4-Dimethyl-phenylhydrazine hydrochlo-
ride (128 g) was dissolved in methanol (700 mL), and 16 tubes
were filled with 60 mL each. The reaction mixtures were heated
to 130 °C over 5 min and held for 15 min at 130 °C. After
cooling to 50 °C, the solvent was evaporated, and the residual
oil was precipitated by addition of diethyl ether (500 mL). The
crystalline solid was filtered and dried. Compound 4b was
isolated as a white solid (178 g, 86%); HPLC (tR 3.87 min,
100%); mp 220 °C. 1H NMR (400 MHz, DMSO-d6) δ 1.29 (s,
9H), 2.29 (s, 6H), 5.66 (s, 1H), 7.26-7.29 (m, 1H), 7.32-7.39
(m, 2H).
Preparation of 5-tert-Butyl-2-(4-cyanophenyl)-2H-pyrazol-
3-ylamine (4c). 4-Cyanophenylhydrazine hydrochloride (102 g)
was dissolved in methanol (700 mL), and 16 tubes were filled
with 50 mL each. The reaction mixtures were heated to 140
°C over 5 min and held for 25 min at 140 °C. After cooling to
50 °C, the solvent was evaporated and the residual oil solidified
upon standing. The crude material was suspended in DCM (400
mL), cooled to 0 °C, and filtered. The crystalline product was
dried, compound 4c was obtained as a white solid (128.5 g,
77%). HPLC (tR 3.97 min, 100%); mp 199-200 °C; 1H NMR
(400 MHz, DMSO-d6) δ 1.26 (s, 9H), 5.62 (s, 1H), 7.83 (d, J
) 8.6 Hz, 2H), 7.99 (d, J ) 8.6 Hz, 2H).
Preparation of 3-Oxo-3,4,4a,5,7,8-hexahydro-2H-pyrido[4,3-
c]pyridazine-6-carboxylic Acid Ethyl Ester (2) in Anton Paar
Synthos 3000. 3-Ethoxycarbonylmethyl-4-oxo-piperidine-1-car-
boxylic acid ethyl ester, 1 (130 g, 505 mmol), hydrazine hydrate
(24.6 mL, 505 mmol), and acetic acid (50.6 mL, 884 mmol)
were dissolved in ethanol (800 mL). This stock solution (∼900
mL) was charged to 16 PTFE tubes, each containing a magnetic
stirring bar, sealed in ceramic cases, and placed inside a 16-
position rotor. One tube was fitted with a gas balloon ther-
mometer; the temperature of the other vials was monitored by
an IR-sensor. The reaction mixtures were heated with magnetic
stirring to 140 °C over 12 min with 1400 W available power,
held for 20 min at 140 °C, and then cooled by air ventilation to
60 °C over 30 min. The contents of all tubes were combined,
and solvent was evaporated in Vacuo. The remaining solid was
recrystallized from ethanol (300 mL). After filtration and drying,
product 2 was isolated as a white crystalline solid (73 g, 64%).
1
HPLC (tR 4.32 min, 99%); mp 167-168 °C; H NMR (400
MHz, CD3OD) δ 1.27 (t, J ) 7 Hz, 3H), 2.50 (dt, J ) 17, 6
Hz, 4H), 2.83 (m, 1H), 3.57 (t, J ) 6 Hz, 2H), 3.65 (t, J ) 6
Hz, 2H), 4.14 (q, J ) 7 Hz, 2H).
Preparation of 5-tert-Butyl-2-p-tolyl-2H-pyrazol-3-ylamine,
4a, and 5-tert-Butyl-2-(3,4-dimethyl-phenyl)-2H-pyrazol-3-
ylamine (4b) on Small Scale under ConVentional Conditions.
In a typical procedure, p-tolylhydrazine hydrochloride (20 g,
126 mmol) and pivaloylacetonitrile (17.3 g, 139 mmol) were
mixed in methanol (200 mL) and heated to reflux. Conversion
to product was monitored by HPLC and found to be complete
after 16 h. Methanol was evaporated, and the remaining solid
was triturated in refluxing DCM (100 mL). After cooling to 0
°C, the precipitated product was filtered and dried to give
3-aminopyrazole, 4a, as colorless crystals (28.8 g, 86% as HCl
salt). HPLC (tR 3.74 min, 100%); mp 195 °C dec.
Preparation of 4,5-Diamino-6-chloropyrimidine (6). Portions
of 5-amino-4,6-dichloropyrimidine (7.5 g each, 732 mmol) were
charged to 16 PTFE tubes, and a solution of 10% ammonia in
ethanol (60 mL for each tube) was added. The tubes were
carefully sealed in ceramic cases, placed in the rotor, and heated
to 170 °C in the microwave cavity over 5 min and held for
180 min at 170 °C. After cooling to 50 °C, the content of all
tubes was combined, and the precipitated solid was filtered and
suspended again in water (750 mL) and stirred at ambient
temperature for 30 min. The precipitate was filtered, washed
with water, and dried in vacuo to give compound 6 as a yellow
solid (87.8 g, 83%); mp 248-250 °C dec (lit.24 249-250 °C).1H
NMR (400 MHz, DMSO-d6) δ 4.95 (s, 2H), 6.76 (s, 2H), 7.65
(s, 1H).
Compound 4b was obtained by a similar procedure and
isolated as an off-white solid (12.8 g, 91.5% yield). HPLC (tR
3.87 min, 100%); mp 219-220 °C.
Vol. 14, No. 3, 2010 / Organic Process Research & Development
•
655

Preparation of 6-Bromoquinazin-4(3H)-one (7). Portions of
2-amino-5-bromobenzoic acid (15 g each, 1.11 mol) were
charged to 16 PTFE tubes and suspended in formamide (35
mL for each tube). The tubes were carefully sealed in ceramic
cases, placed in the rotor and heated to 180 °C in the microwave
cavity over 5 min and held at 170 °C for 30 min. After cooling
to 70 °C, the content of all tubes was poured into water (2.4 L)
and stirred at ambient temperature for 30 min. The precipitated
solid was filtered and washed with water (500 mL) and dried
in vacuo to give compound 7 as an off-white solid (222.5 g,
89%); mp 273-274 °C (lit.26 262 °C). HPLC (tR 3.83 min,
Hz, 1H), 7.92 (dd, J ) 8.6 Hz, 2.35 Hz, 1H), 8.11 (s, 1H),
8.16 (d, J ) 2.35 Hz, 1H), 12.4 (br s, 1H).
Acknowledgment
We thank Cara Brocklehurst for her assistance to prepare
this manuscript and for fruitful discussions. We also thank
Thomas Ruppen, Dominik Wiss, and Martin Roggwiller for
technical assistance.
Received for review October 19, 2009.
OP900269Y
1
100%); H NMR (400 MHz, DMSO-d6) δ 7.59 (d, J ) 8.6
656
•
Vol. 14, No. 3, 2010 / Organic Process Research & Development