Microwave-enhanced reactions under open and closed vessel conditions. A case study

Article abstract of DOI:10.1016/S0040-4020(02)00270-3

Important differences have been observed when performing microwave-enhanced organic transformations, using open and closed vessel conditions. For the hydrolysis of benzamide with sulfuric acid in sealed vessels, where no volatile products are formed, no appreciable difference in reaction performance as a function of the filling volume of the vials or the reaction scale were detected. However, in the cyclocondensation of tetrahydroquinoline with substituted malonic esters, where 2equiv. of ethanol are formed, the outcome of the reaction under sealed vessel conditions is critically dependent on the scale of the reaction.

Full text of DOI:10.1016/S0040-4020(02)00270-3

TETRAHEDRON
Pergamon
Tetrahedron 58 (2002) 3177±3183
Microwave-enhanced reactions under open and closed vessel
conditions. A case study
Alexander Stadler, Sonja Pichler, Gabriela Horeis and C. Oliver Kappep
Institute of Organic Chemistry, Karl-Franzens-University Graz, Heinrichstrasse 28, A-8010 Graz, Austria
Received 20 December 2001; revised 12 February 2002; accepted 28 February 2002
AbstractÐImportant differences have been observed when performing microwave-enhanced organic transformations, using open and
closed vessel conditions. For the hydrolysis of benzamide with sulfuric acid in sealed vessels, where no volatile products are formed, no
appreciable difference in reaction performance as a function of the ®lling volume of the vials or the reaction scale were detected. However, in
the cyclocondensation of tetrahydroquinoline with substituted malonic esters, where 2 equiv. of ethanol are formed, the outcome of the
reaction under sealed vessel conditions is critically dependent on the scale of the reaction. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
general, regardless of the speci®c microwave reactor
employed, microwave-enhanced processes can be carried
out either under sealed vessel or under open vessel
(atmospheric pressure) conditions. In this article, we deal
with the speci®c advantages and disadvantages resulting
fromthe use of either of the two methods. We also report
on scale-up characteristics and on the use of so-called multi-
mode versus monomode microwave instruments.
High-speed microwave-enhanced chemistry has attracted a
considerable amount of attention in recent years. Since the
®rst report on the use of microwave heating to accelerate
organic chemical transformations by Gedye and co-workers
in 1986,1,2 more than 1000 articles have been published in
the area of microwave-enhanced organic synthesis.3 In fact,
it is becoming evident that rapid microwave protocols can
be developed for a large number of chemical transforma-
tions requiring heat. The main bene®ts of performing reac-
tions under microwave irradiation conditions are the
signi®cant 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 irradiation, a generally
accepted rationalization remains elusive.4 Regardless of
the origin/existence of a special microwave effect, micro-
wave-enhanced chemistry can be extremely ef®cient and is
applicable to a broad range of practical synthesis.3±8
Although many of the early pioneering experiments in
microwave-enhanced organic synthesis have been carried
out in domestic unmodi®ed microwave ovens, the current
trend clearly is to use dedicated instruments for chemical
synthesis, in particular for processes involving organic
solvents.3 Most of today's commercially available micro-
wave reactors feature built-in magnetic stirrers, direct
temperature control of the reaction mixture with the aid of
®ber-optic probes, shielded thermocouples or IR sensors,
and software that enables on-line temperature/pressure
control by regulation of microwave power output.3 In
2. Results and discussion
2.1. Hydrolysis of benzamide
As a suitable model reaction for studying some of the effects
outlined above we have chosen the acid hydrolysis of benza-
mide (1) to benzoic acid (2) (Scheme 1). In fact, this trans-
formation was the ®rst microwave-enhanced organic
reaction published in the literature.1,2 In 1986, Gedye and
co-workers reported on this hydrolysis under microwave
conditions, employing sealed Te¯on vessels in an unmodi-
®ed domestic microwave oven.1 A six-fold rate enhance-
ment, i.e. a reduction of the hydrolysis time from 1 h to
10 min, was observed for this particular process as
compared to classical re¯ux conditions. Due to the nature
of microwave dielectric heating,9 accurate temperature
measurements using conventional means of temperature
determination during the irradiation process were not
Keywords: microwave-enhanced chemistry; hydrolysis; cyclocondensa-
tions; 4-hydroxy-2(1H)-quinolones; non-thermal microwave effects.
Corresponding author. Tel.: 143-316-3805352; fax: 143-316-3809840;
p
Scheme 1.
e-mail: oliver.kappe@uni-graz.at
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00270-3

3178
A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
Figure 1. Kinetics of the hydrolysis of benzamide (1) at ca. 1008C in 20% (v/v) (a, left) and 5% (v/v) sulfuric acid (b, right). Hydrolysis rates were determined
by HPLC analysis (see Section 4).
possible at that time. Therefore, the reason for the observed
rate enhancement was not fully understood and these results
(as many others that followed later) contributed to specula-
tion on the existence of so-called non-thermal or speci®c
microwave effects.4 In order to have accurate data for
comparisons at hand, we ®rst investigated the hydrolysis
of benzamide under classical thermal re¯ux conditions
(,1008C). Experiments were run both with the originally
speci®ed1,2 20% (v/v) sulfuric acid, in addition to 5% (v/v)
sulfuric acid. The data so obtained are presented in Fig. 1.
Thus, complete hydrolysis of benzamide at 1008C under
atmospheric pressure conditions requires ca. 1 h in 20%
and 12 h in 5% sulfuric acid.
reached. The actual reaction time at 1808C is therefore
considerable shorter (Fig. 2). Active gas jet cooling allows
the mixture to be cooled very ef®ciently. In addition we also
carried out microwave-enhanced hydrolysis reactions
involving 5% sulfuric acid. Here, 1408C reaction tempera-
ture proved to be insuf®cient to hydrolyze benzamide within
a reasonable time (Table 1). However, increasing the
temperature to 1808C allowed complete hydrolysis of
benzamide within 7 min. Comparing the results of thermal
and microwave-enhanced hydrolysis rates, signi®cant rate
enhancements (10±100-fold) could indeed be achieved for
both the 20 and 5% sulfuric acid runs. Impressive as these
numbers may seem, they are not surprising at all if one
considers the signi®cantly higher con®rmed reaction
temperatures under which these processes took place, and
the fact that these temperatures are rapidly obtainable in
sealed vessels utilizing microwave technology.3 Performing
this reaction under open vessel microwave irradiation
conditions would not be expected to lead to any signi®cant
rate enhancement (due to the aqueous medium), apart from
what can be attributed to the well-known, but comparatively
small microwave superheating effect at atmospheric
pressure.11
We next turned our attention to hydrolysis reactions under
sealed vessel conditions using heating by microwave
irradiation. For these small-scale experiments a monomode
instrument with integrated robotics interface for automated
use (see Section 4) was employed.10 In contrast to the
original work by Gedye et al.,1,2 here both direct tempera-
ture and pressure measurements were possible. After a few
optimization runs using the same reaction mixture composi-
tion as described by Gedye et al.1,2 we discovered that
employing 1408C as reaction temperature complete con-
version to benzoic acid could be achieved within 7 min
(Table 1), corresponding nicely to the rate-enhancement
data given in the original publication.1,2 Making use of the
automated sequential processing capabilities of the micro-
wave reactor,10 this hydrolysis reaction could rapidly be
further optimized to arrive at conditions where complete
hydrolysis is possible within only 1±2 min (e.g. by increas-
ing the reaction temperature to 1808C) (Table 1). The
reaction time here is speci®ed as the total irradiation time,
thus including the rapid microwave ¯ash heating period
until the 1808C preselected reaction temperature was
The above experiments involving microwave-enhanced
hydrolysis in monomode instruments involved 2.0 mL of
solvent in glass process vials designed for a ®lling volume
of 2.0±5.0 mL. In order to cover a variable range of
different reaction volumes we have also carried out experi-
ments in the smaller of the two types of available process
vials, designed for reaction volumes of 0.5±2.0 mL (see
Table 1. Conversion rates (%) for the microwave-enhanced hydrolysis of
benzamide under different reaction conditions (sealed vessels)
Time (min)
20% H2SO4
5% H2SO4
1408C
1608C
1808C
1408C
1608C
1808C
1
2
5
7
10
25
71
95
.99
±
72
95
.99
±
95
.99
±
±
±
9
37
62
90
91
73
79
96
.99
±
31
51
64
83
Figure 2. Heating pro®le for the benzamide hydrolysis at 1808C in 20%
sulfuric acid under sealed vessel/microwave irradiation conditions. Micro-
wave ¯ash heating (300 W, 0±50 s), temperature control using the feedback
fromIR thermography (constant 180 8C, 50±120 s), and active cooling
(120±210 s).
±
.99
Hydrolysis rates were determined by HPLC analysis (see Section 4).

A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
3179
per¯uoroalkoxy Te¯on (PFA) reaction vessel. Again, the
use of this set-up allowed direct temperature and pressure
monitoring, as well as stirring of the reaction mixture. Using
the same concentrations (100 mg benzamide/mL sulfuric
acid) and reaction conditions successful hydrolysis runs
were carried out with 10, 25, and 50 mL volume, i.e. hydro-
lyzing up to 5 g of benzamide. Runs were carried out at
1408C (20% sulfuric acid) and 1808C (5% sulfuric acid),
leading to complete hydrolysis within 10 min. Although
the heating and cooling pro®les for the two reactors were
different, these experiments demonstrate that conditions can
be successfully taken fromone microwave platformto
another, going from0.5 to 50 mL in scale, and froma mono-
mode to a multimode system. In particular, the fact that
there was no active cooling incorporated with the multi-
mode reactor (heating pro®les not shown) made a more
quantitative comparison impossible.
Figure 3. Heating pro®les for the benzamide hydrolysis at 1408C in 20%
sulfuric acid under sealed vessel/microwave irradiation conditions; (a) for
0.5 mL reaction volume (small process vial); (b) for 5.0 mL reaction
volume (large process vial). For very small volumes (curve a) the tempera-
ture equilibration is not as smooth as for larger solvent volumes (curve b).
Section 4). Therefore a series of experiments involving
hydrolysis of 100 mg/mL of benzamide in 20% sulfuric
acid was carried out using 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, and
5.0 mL volume of sulfuric acid in the appropriate vials.
Irradiation of all these reaction mixtures using the optimized
conditions (1408C, 7 min) led to quantitative hydrolysis of
benzamide. The maximum pressure, depending on the type
of vial and the volume, was 5±8 bar. To get more accurate
information on the reproducibility and scale-up character-
istics of this process we have also run a set of experiments
involving only 2 min of irradiation time (1408C), i.e. for an
incomplete hydrolysis (see Table 1), in the solvent range
indicated above (0.5±5.0 mL). The hydrolysis data were
spread out ^10% around the 71% value found for the
2.0 mL run (cf. Table 1), the higher rates (70±80%) gener-
ally being obtained for the lower solvent volumes, and vice
versa. This is not surprising if one considers the difference
in the heating pro®les between, e.g. a 0.5 and 5.0 mL run
(Fig. 3), where a distinctly faster heating to the desired ®nal
temperature is observed for smaller volumes. Apart from
these comparatively small deviations no appreciable
difference in reaction performance depending on the ®lling
volume of the vials was detected. It should be pointed out
that the ®lling volume may be a major concern when
performing microwave chemistry with systems lacking
temperature or pressure control, or when working outside
the speci®ed volume ranges.
2.2. 4-Hydroxy-2-quinolinones by cyclocondensation
In order to investigate in detail the effect of open- versus
closed vessel conditions in microwave-heated chemical
transformations we next turned our attention to a second
model reaction, namely the formation of 4-hydroxyquino-
lin-2-(1H)-ones 5a,b fromanilines (i.e. 3) and malonic
esters (Scheme 2).13 Note that here, in contrast to the ®rst
model reaction, 2 equiv. of a volatile byproduct (ethanol)
are formed, that under normal (atmospheric pressure) condi-
tions are simply distilled off.13,14 At least one of the ethanol
molecules may be involved in an equilibrium type process
(6!7, Scheme 2), and preventing removal of ethanol from
the reaction mixture, i.e. by using a closed vessel, may drive
the reaction along undesired pathways.13 The desired
synthesis is conventionally dif®cult to achieve, mainly
because of the high temperatures required (250±3508C).
Such high temperature conditions are crucial here in order
to generate the key a-oxoketene intermediates 7 which
subsequently cyclize to the desired quinolinones 5.15,16
Therefore the rapid heating and high-temperature pro-
cessing capabilities offered by microwave technology
seemed to be ideally suited for this type of chemistry.
Furthermore, carrying out the reaction sequence in a closed
vessel would assure that the desired high-temperatures
Another issue in microwave-enhanced synthesis is the use
of dedicated multimode versus monomode cavities.3 In the
so-called multimode instruments (conceptually similar to a
domestic oven), the microwaves that enter the cavity are
being re¯ected by the walls and the load over the typically
large cavity. A mode stirrer ensures that the ®eld distribu-
tion is as homogeneous as possible. In the much smaller
mono- or single-mode cavities, only one mode is present
and the electromagnetic irradiation is focused directly
through an accurately designed wave-guide onto the reac-
tion vessel mounted in a ®xed distance from the radiation
source. We were therefore interested to investigate if the
relatively small-scale reaction conditions that were opti-
mized in the monomode instrument (see above) could be
successfully transformed to a multimode reactor, capable of
processing much larger volumes. In this context we have
carried out the hydrolysis of benzamide in an ETHOS Synth
Labstation (see Section 4),12 employing a 100 mL sealed
Scheme 2.

3180
A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
could be reached, regardless of the boiling points of solvents
or reagents employed.
reagents run (entries 8±12). The results obtained clearly
indicate that there is no dilution effect at play here. The
optimum yield (91% isolated yield in 10 min) was achieved
for a 1.0 mmol composition of reagents in 0.5 mL of DCB
(entry 12). Similar high yields were also obtained without
solvent, but here the isolated product was not in an
analytical state of purity (entry 13). It should be pointed
out that for the corresponding conventional, thermal
protocol, reaction conditions involving several hours of
heating in an oil bath at 220±3008C (without solvent)
were reported providing similar high yields (94%).14 We
also note that while this work was in progress, a micro-
wave-enhanced synthesis of 4-hydroxy-2-quinolinones
appeared in the literature, involving the solvent-less fusion
of anilines and malonic esters in open vessels employing a
dedicated multimode reactor.17 Interestingly, the authors
speci®cally point out that it was `essential to work in open
vessels', and that `in a closed vessel the reaction does not
proceed at all'.17 For comparison purposes we have there-
fore carried out the synthesis of 5a under open vessel,
solvent-free, microwave conditions using the monomode
reactor (entry 14, see Section 4). In agreement with the
published work17 we ®nd that this method provides a high
yield of product, albeit not in analytical purity since DCB
cannot be used as solvent (see above). It is important to note,
however, that hereÐin contrast to sealed vessel condi-
tionsÐone is not limited by the amount of material in the
vial, since there is no pressure build-up (entries 14 and 15).
Therefore, for larger scale synthesis the atmospheric
pressure conditions are evidently preferred. On the other
hand, one has to keep in mind that large-scale experiments
in open vessels will release signi®cant amounts of ¯am-
mable ethanol into the microwave cavity which presents a
severe safety hazard.
With these issues in mind we have carried out the following
sets of microwave experiments involving the reaction of
diethyl phenylmalonate (4a) with tetrahydroquinoline (3)
using 1,2-dichlorobenzene (DCB) as solvent. DCB is a
solvent ideally suited for this process as it is microwave
absorbing due to its dielectric properties9 and also dissolves
both starting materials, whereas the quinolinone products 5
are virtually insoluble at roomtemperature. This facilitates
product isolation and, e.g. provided 4-hydroxy-2-quinoli-
none 5a in analytical purity directly upon cooling of the
reaction mixture. All initial experiments were carried out
in the monomode reactor using sealed vessel conditions at
the maximum operable temperature of 2508C (Table 2). A
number of optimization runs were carried out involving
equimolar amounts of starting materials in varying concen-
trations at 2508C. In one series of experiments, the effect of
irradiation time on the yield was studied for runs involving
2.0 mmol of reagents and 2.0 mL of DCB. Not surprisingly,
the yield increased from54 to 78% on going from5 to
30 min of irradiation time (entries 1±4). In another set of
experiments (entries 5±7) the effect of starting material
concentration on the observed yields was investigated keep-
ing the irradiation time (10 min) and amount of solvent
(2.0 mL) constant. We were surprised to ®nd that there is
a signi®cant dependence of the yield on the amount of
reagents used in this process. Apparently, the lower is the
amount of starting materials, the higher the yield. At ®rst
glance these results may indicate a relationship between
dilution and yield. However, we attribute this unexpected
effect to the pressure built-up in the reaction vessel. Moni-
toring the pressure in the vial, it became evident that the
larger the amount of material in the vial the higher the
pressure that was formed in the sealed vessel, going from
3.6 bar (1.0 mmol) to 7.4 bar (4.0 mmol). Clearly, the pres-
sure build-up in the vial is a result of the ethanol eliminated
during the reaction.13 In an effort to increase the available
head-space in the sealed reaction vessels and to check for
possible dilution effects we have performed another set of
experiments, varying the amount of solvent for a 1.0 mmol
To further address the issue of open- versus closed reaction
vessels we have studied another example of the above
cyclocondensation process, involving diethyl ethylmalonate
(4b) as building block.13 In contrast to the corresponding
phenylmalonate, the ethylmalonate has a signi®cantly lower
boiling point (2088C for 4b18 versus 2858C for 4a;19 2458C
for 320) which makes the preparation of quinolinone 5b
under conventional conditions troublesome.13 Indeed, in a
control experiment using oil-bath heating (2508C bath
temperature) (Table 3, entry 1), only traces of the desired
Table 2. Optimization runs for the microwave-promoted synthesis of 4-
hydroxyquinolin-2-(1H)-one 5a under sealed and open vessel conditions
(2508C, monomode reactor)
Entry
(3)/(4a) (mmol)
DCBa (mL)
Time (min)
Yield (%)
Table 3. Comparison of thermal- versus microwave-heated synthesis of
quinolinone 5b under sealed and open vessel conditions (20 min reaction
time)
1
2
3
4
5
6
7
8
2.0
2.0
2.0
2.0
1.0
2.0
4.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
4.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
4.0
3.0
2.0
1.0
0.5
±
5
54
67
72
78
76
67
60
54
65
76
84
91
90
92
90
10
20
30
10
10
10
10
10
10
10
10
10
10
10
Entry
Method
(3)/(4b) (mmol)
DCB (mL)
Yield (%)
1
2
3
4
5
6
7
8
9
CONVa
CONVa
MWb
2.0
2.0
2.0
2.0
2.0
4.0
2.0
4.0
2.0
1.0
±
1.0
0.5
±
±
±
±
±
,5%
40
37
49
75
34
75
75
71
MWb
9
MWb
10
11
12
13
14b
15b
MWb
MWc
MWc
CONVd
±
±
a
Open vessel, pre-heated oil-bath (2508C).
Sealed vessel microwave irradiation (2508C).
Open vessel microwave irradiation (2508C).
Open vessel, pre-heated oil bath (2908C).
b
c
d
a
Solvent: 1,2-dichlorobenzene (DCB).
Open vessel in the monomode reactor (see Section 4).
b

A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
3181
product were obtained when the reaction was carried out in
1,2-dichlorobenzene (DCB) as solvent. This is not surpris-
ing, since here the boiling point of DCB (1808C) will
prevent the reaction mixture from reaching the required
high temperatures. When the same experiment was carried
out in a solvent-free manner, a moderate yield of 40% of 5b
was obtained (entry 2). It should be noted that the tempera-
ture of the reaction mixture reached only ca. 2258C at the
end of the 20 min period with concomitant removal of
volatile ethanol fromthe reaction vessel. In a further set
of experiments the cyclocondensation reaction was carried
out under sealed vessel conditions using the monomode
microwave reactor (entries 3±6). Similar to the results
described with the phenyl analog (Table 2), there was a
distinct correlation between the amount of solvent (entries
3±5) and starting materials (compare entries 5 and 6) used,
and the yield of product. Again, the lower is the total amount
of material in the vial (i.e. the larger the head-space and the
lower the pressure build-up) the higher the yield. For quino-
linone 5b the best result (75% isolated yield) was obtained
using solvent-free conditions at 2.0 mmol concentration of
starting materials (entry 5). Using microwave dielectric
heating in sealed vessels the desired reaction temperature
of 2508C was rapidly reached within 3 min, in contrast to
thermal heating utilizing an oil bath, preheated to 2508C,
where the maximum reachable temperature was 2258C.
Directly comparing the conventional and the microwave
solvent-free runs (entries 2 and 5, respectively), it seems
that the signi®cantly higher yield in the latter experiment
may be rationalized by the rapid heating possible under
microwave conditions (microwave-¯ash heating), and the
signi®cantly higher ®nal temperature that is obtained by
direct `in core' microwave heating. In order to address
this issue further, we have carried out additional experi-
ments in open vessels. Similar to the phenyl analog 5a
(Table 2) open vessel microwave runs at 2508C produced
exactly the same yield as closed vessel experiments at lower
concentrations (entries 7 and 8). Apparently, the lower boil-
ing point of malonate 4b does not play a major role here. In
a ®nal experiment we have tried to mimic the rapid heating
and high temperatures obtainable by microwave irradiation
by performing the cyclocondensation in a preheated oil bath
at 2908C. Under these conditions the heating pro®le of the
reaction was virtually identical to the microwave runs (open
or closed vessel conditions) leading to a temperature of ca.
2458C in the process vial within 3 min. Not surprisingly, a
yield of 71% of isolated pure product was obtained (entry
9). Therefore, by simulating the rapid heating possible
utilizing microwave heating by conventional methods, it is
indeed possible to obtain a similar high yield as in the
microwave-heated experiments! One has to be aware,
however, that the current transformations were carried out
on a relatively small scale (2.0±4.0 mmol). It may in fact
not be possible to reproduce the crucial rapid heating
pro®les experienced herein on a signi®cantly larger scale,
in particular not with conventional heating methods.
tions. In the ®rst example, the hydrolysis of benzamide (1)
to benzoic acid (2) in sulfuric acid (Scheme 1), no volatile
reaction products are formed and this transformation can be
performed successfully under a variety of sealed vessel
conditions. The outcome of the reaction is not dependent
on the ®lling volume in the vial as long as similar heating
pro®les can be achieved. Therefore this process can be
scaled-up from 0.5 to 50 mL volume without a signi®cant
change in process time, even switching microwave plat-
forms (monomode to multimode). In the second example,
the condensation of tetrahydroquinoline (3) with malonates
4, 2 equiv. of ethanol are formed during the cyclocondensa-
tion process (Scheme 2). This leads to a signi®cant pressure
build-up in the sealed reaction vessel, which has a marked
effect on the progression of the reaction (Tables 2 and 3),
and makes the scale-up of these processes under sealed
vessel conditions troublesome. However, these condensa-
tion reactions can also be carried out successfully under
open vessel conditions in a microwave heated, solvent-
free experiment. That way, the volatile reaction product is
removed rapidly from the reaction mixture and therefore
fromthe equilibrium(Scheme 2).
In summary, we have demonstrated that care must be taken
in deciding on the appropriate reaction conditions in micro-
wave-heated organic transformations. While sealed vessel
reactors provide the important convenience and advantage
that the reaction temperature is not limited by the boiling
point of the solvent or reagent, in some cases the closed
vessel conditions can also present a severe problemas
demonstrated herein.21 We believe that this is particularly
relevant for those processes where equilibria are involved,
and where volatile molecules can otherwise be removed
fromthe reaction vessel.
We would also like to point out that in the studies reported
herein we have not observed any non-thermal or speci®c
microwave effects.4 All rate enhancements described in
this work can be rationalized on the basis of the increased
reaction temperatures.
4. Experimental
1H NMR spectra were obtained on a Bruker AMX 360
instrument at 360 MHz. FTIR spectra were recorded on a
Mattson Instruments Unicam FTIR 7000 spectrophotometer
using the KBr pellet method. Micro-analyses were obtained
on a Fisons Mod. EA 1108 elemental analyzer. Benzamide
(1), tetrahydroquinoline (3), malonates 4, 1,2-dichloro-
benzene (DCB) and sulfuric acid were obtained from
Aldrich Chem. Co. and used without further puri®cation.
4.1. Microwave instrumentation
Monomode reactor: Smith Synthesizere (Personal
Chemistry AB).10 The systemoperates at a frequency of
2.45 GHz with continuous microwave irradiation power
from0±300 W. The reaction vials (Smith Process Vials e)
are glass-based ca. 10 mL tubes, sealed with Te¯on septa
and an aluminum crimp top. Vials for 0.5±2.0 and 2.0±
5.0 mL ®lling volume (total volume 8.3 and 11.0 mL,
respectively) are available, both equipped with appropriate
3. Conclusion
In the present article, we have described the differences
observed in carrying out microwave-enhanced organic
transformations under sealed vessel and open vessel condi-

3182
A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
stirring bars. The process vials are moved in an automated
fashion by a gripper incorporated to the platform. Inside the
cavity, applying the pressure sensor as additional seal, the
vials can be exposed to 20 bars of pressure and 2508C.
Temperature is measured by infrared thermometry on the
outer surface of the process vial. The microwave output
power is regulated by the software algorithm, so that the
preselected maximum temperature is maintained for the
desired reaction time. After the irradiation period the reac-
tion vessel is cooled down rapidly (approx. 60 s) to ambient
temperature by compressed air (gas jet cooling). For experi-
ments in open vessels, the Te¯on septa were omitted during
capping of the vials. Caution: note that this is not an
approved method of operation for the microwave instru-
ment. Care must be taken since volatile and/or ¯ammable
organic material may enter the microwave cavity.
with stirring for 90 min. After 1, 2, 5, 7, 10, 20, 30, 40, 50,
60 and 90 min, aliquots were taken from the reaction
mixture and the exact conversion rates were determined
by HPLC measurements as described in Section 4.2.
Complete conversion (.99%) at re¯ux temperature
(1008C) required 60 min. An identical experiment was
carried out with 5% (v/v) sulfuric acid involving heating
for 24 h. Conversion rates were determined for 10, 20, 30,
40, 50, 60, 90 min and 2, 3, 5, 8, 10, 12, 14, 16 and 24 h.
Complete conversion (.99%) required 12 h.
4.4. Microwave-enhanced hydrolysis of benzamide
(Table 1)
A. Monomode reactor. Mixtures of 200 mg (1.65 mmol) of
benzamide (1) and 2.0 mL 20% (v/v) sulfuric acid were
®lled in large Smith Process Vialse. After the vials were
sealed and placed in the Smith Synthesizer microwave
cavity they were irradiated sequentially using the conditions
given in Table 1. Conversion rates were determined by
HPLC measurements as described in Section 4.2. Identical
experiments were carried out, employing 2.0 mL of 5%
(v/v) sulfuric acid and for volumes from 0.5±2.0 mL
(small vial), and 2.0±5.0 mL (large vial) using the same
concentration of benzamide (100 mg/mL).
Multimode reactor: ETHOS Synth Labstation (Milestone
Inc.).11 The multimode microwave reactor has a twin
magnetron (2£800 W, 2.45 GHz) with
a maximum
delivered power of 1000 W in 10 W increments (pulsed
irradiation). A rotating microwave diffuser ensures homo-
geneous microwave distribution throughout the plasma
coated PTFE cavity (35 cm£35 cm£35 cm). For the experi-
ments carried out in sealed vessels a 100 mL PFA reaction
vessel contained in a single high-pressure HPR1000 rotor
block segment was employed. Built-in magnetic stirring
(Te¯on-coated stirring bar) was used in all operations.
During experiments, time, temperature, pressure, and
power was monitored/controlled with the `easyWAVE'
software package (Version 3.2). Temperature was moni-
tored with the aid of a shielded thermocouple (ATC-300)
inserted directly into the corresponding reaction container.
For experiments in sealed vessels a pressure sensor (APC-
55) was additionally employed.
B. Multimode reactor. Mixtures of 100 mg benzamide/mL
of sulfuric acid (20 and 5% (v/v), respectively) were placed
in the 100 mL sealed PFA Te¯on vessel and successful
hydrolysis runs were carried out with 10, 25, and 50 mL
volume. The 20% mixtures were irradiated at 1408C
(700 W maximum power), the 5% mixtures at 1808C
(800 W maximum power) for 10 min. Cooling after the
irradiation was performed by placing the vessel in an ice
bath. At ca. 808C the vessel was opened and the mixture was
allowed to stand in a refrigerator (48C) for several hours to
enable complete crystallization of product. Conversion rates
were determined by HPLC measurements as described in
Section 4.2.
4.2. Monitoring benzamide hydrolysis
As benzamide (1) and benzoic acid (2) are virtually
insoluble in 48C water and sulfuric acid, the reaction
products could be simply isolated by suction after standing
for several hours in a refrigerator and washed with cold
water. For all kinetic investigations described herein a
HPLC method was developed based on the analysis of
reprecipitated material after cooling (.95% recovery).
After drying, equivalent amounts of material were dissolved
in a minimum amount of acetonitrile and ®lled up to 10 mL
with water. That solution was injected into the HPLC (HP
Series 1050), Column LiChrospher 100 RP-18 (5 mm)
119 mm£3 mm, E. Merck, Darmstadt, Germany), using a
gradient of (acetonitrile/water 5:95)/(acetonitrile) 7:3
(0±2 min) and acetonitrile (2±10 min) as the mobile
solvent. The elution resulted in a retention time of 2.5 min
for benzamide and 3.7 min for benzoic acid. For reference
purposes some samples were also worked-up according to
the method described by Gedye et al.2 involving extraction
methods. In our hands both methods gave virtually identical
results.
4.5. Cyclocondensation of tetrahydroquinoline (3) and
malonic esters (4a,b) (Tables 2 and 3)
A. Monomode reactor. Equimolar amounts of tetrahydro-
quinoline (3) and the corresponding malonic esters 4a or
4b (1.0±4.0 mmol) were placed in the appropriate reaction
vial. 1,2-Dichlorobenzene as a solvent was added in
0±4.0 mL range. Individual vessels were sealed, placed in
the microwave cavity and irradiated at 2508C for 5±30 min.
After rapid gas jet cooling the crude precipitated products
were diluted with ca. 1.0 mL of hexanes and allowed to
stand in the refrigerator for approximately 1 h. The precipi-
tate was ®ltered by suction and dried in vacuo. In case of the
open vessel experiments, the reaction vials were placed in
the microwave cavity without sealing (alumina crimp top
without Te¯on septum).
B. Thermal heating (Table 3, entry 2). A mixture of 266 mg
(2.0 mmol) of tetrahydroquinoline (3) and 376 mg
(2.0 mmol) of ethyl diethylmalonate was placed in a small
Smith Process Viale. The ®ber optic probe was inserted and
the vessel was placed in a pre-heated oil bath (2508C bath
temperature). The temperature inside the vessel reached
4.3. Thermal hydrolysis of benzamide (Fig. 1)
Mixtures of 3.0 g (24.8 mmol) of benzamide (1) in 30 m L of
20% (v/v) sulfuric acid were heated under re¯ux (1008C)

A. Stadler et al. / Tetrahedron 58 (2002) 3177±3183
3183
Â
Jacquault, P.; Mathe, D. Synthesis 1998, 1213±1234.
(b) Varma, R. S. Green Chem. 1999, 43±55.
2258C within 90 s and the mixture was further heated at this
temperature for 20 min and subsequently cooled using a
water bath. The reaction mixture was diluted with 1.0 mL
of hexanes and allowed to stand in the refrigerator for
approximately 1 h. The precipitate was ®ltered by suction
and dried in vacuo. This procedure yielded pure 5b in 40%
yield.
6. (a) Caddick, S. Tetrahedron 1995, 51, 10403±10432.
(b) Bose, A. K.; Banik, B. K.; Lavlinskaia, N.; Jayaraman,
M.; Manhas, M. S. Chemtech 1997, 27, 18±24. (c) de la
Â
Hoz, A.; Dõaz-Ortis, A.; Moreno, A.; Langa, F. Eur. J. Org.
Chem. 2000, 3659±3673.
7. Larhed, M.; Hallberg, A. Drug Discov. Today 2001, 6, 406±
416.
4.5.1.
1-Hydroxy-2-phenyl-6,7-dihydro-5H-benzo[ij]-
quinolizin-3-one (5a). Mp 221±2228C (lit.14 2208C). H
NMR (DMSO-d6) d 2.00 (t, Jˆ7.0 Hz, 2H), 2.95 (t,
Jˆ7.0 Hz, 2H), 4.02 (t, Jˆ7.0 Hz, 2H), 7.13±7.87 (m,
8H), 9.96 (s, OH). IR (KBr): n 3200, 1625, 1605,
1565 cm21. Anal. calcd: C, 77.96%; H, 5.45%; N, 5.05%.
Found: C, 78.21%; H, 5.55%; N, 4.98%.
8. Elander, N.; Jones, J. R.; Lu, S.-Y.; Stone-Elander, S. Chem.
Soc. Rev. 2000, 239±250.
1
9. Gabriel, C.; Gabriel, S.; Grant, E. H.; Halstead, B. S. J.;
Mingos, D. M. P. Chem. Soc. Rev. 1998, 27, 213±223.
10. For a detailed description, see: Stadler, A.; Kappe, C. O.
J. Comb. Chem. 2001, 3, 624±630.
11. Chemat, F.; Esveld, E. Chem. Engng Technol. 2001, 24, 735±
741.
4.5.2. 1-Hydroxy-2-ethyl-6,7-dihydro-5H-benzo[ij]qui-
1
nolizin-3-one (5b). Mp 192±1938C (lit.22 294±2958C). H
12. For a detailed description, see: Kappe, C. O. Am. Lab. 2001,
33 (10), 13±19.
NMR (DMSO-d6) d 1.01 (t, Jˆ7.5 Hz, 3H), 1.97 (t, Jˆ6 Hz,
2H), 2.60 (q, Jˆ7.5 Hz, 2H), 2.91 (t, Jˆ6 Hz, 2H), 4.01 (t,
Jˆ6 Hz, 2H), 7.10 (t, Jˆ8.0 Hz, 1H), 7.29 (d, Jˆ8.0 Hz,
1H), 7.78 (d, Jˆ8.0 Hz, 1H), 9.94 (br s, OH). IR (KBr): n
3200, 1630, 1605, 1570 cm21. Anal. calcd: C, 73.34%; H,
6.59%; N, 6.11%. Found: C, 73.55%; H, 6.43%; N, 6.01%.
13. Stadlbauer, W.; Laschober, R.; Lutschounig, H.; Schindler,
G.; Kappe, T. Monatsh. Chem. 1992, 123, 617±636 and
references cited therein.
14. Stadlbauer, W.; Schmut, O.; Kappe, T. Monatsh. Chem. 1980,
111, 1005±1013.
15. Ziegler, E. Chimia 1970, 24, 62±64.
16. Ziegler, E.; Sterk, H. Monatsh. Chem. 1968, 99, 1958±1961.
17. Lange, J. H. M.; Verveer, P. C.; Osnabrug, S. J. M.; Visser,
G. M. Tetrahedron Lett. 2001, 42, 1367±1369.
18. Clark, J. H.; Miller, J. M. J. Chem. Soc., Perkin Trans. 1 1977,
1743±1745.
Acknowledgements
We wish to thank Personal Chemistry AB (Uppsala) for use
of the Smith Synthesizere. We also would like to thank Dr
W. Stadlbauer for helpful discussions and G. A. Strohmeier
for assisting with the HPLC measurements.
19. Kappe, T.; Kos, C. Synthesis 1989, 624±625.
20. Steele, W. V.; Chirico, R. D.; Hossenlopp, I. A.; Nguyen, A.;
Smith, N. K.; Gammon, B. E. J. Chem. Thermodyn. 1989, 21,
1121±1150.
References
21. In addition, the unique temperature and pressure relationships
involved in closed vessel microwave systems have also to be
considered. Detailed studies have shown that the internal pres-
sure in a microwave heated vessel is considerably lower than
in a conventionally heated autoclave due to the fact that the
vessel material is generally microwave transparent and the
vessel walls remain relatively cool during the heating process.
Richter, R. C.; Kingston, H. M. Anal. Chem. 2001, 73, 31A±
37A.
1. Gedye, R.; Smith, F.; Westaway, K.; Ali, H.; Baldisera, L.;
Laberge, L.; Rousell, J. Tetrahedron Lett. 1986, 27, 279±282.
2. For a full description of this work, see: Gedye, R. N.; Smith,
F. E.; Westaway, K. C. Can. J. Chem. 1987, 66, 17±26.
È
3. Lidstrom, P.; Tierney, J.; Wathey, B.; Westman, J.
Tetrahedron 2001, 57, 9222±9283.
4. (a) Perreux, L.; Loupy, A. Tetrahedron 2001, 57, 9199±9223.
Â
(b) Langa, F.; de la Cruz, P.; de la Hoz, A.; Dõaz-Ortiz, A.;
Â
Dõez-Barra, E. Contemp. Org. Synth. 1997, 373±386.
5. (a) Loupy, A.; Petit, A.; Hamelin, J.; Texier-Boullet, F.;
22. Kappe, T.; Aigner, R.; Roschger, P.; Schnell, B.; Stadlbauer,
W. Tetrahedron 1995, 51, 12923±12928.