Nonthermal microwave effects revisited: On the importance of internal temperature monitoring and agitation in microwave chemistry

Article abstract of DOI:10.1021/jo7022697

(Graph Presented) The concept of nonthermal microwave effects has received considerable attention in recent years and is the subject of intense debate in the scientific community. Nonthermal microwave effects have been postulated to result from a direct stabilizing interaction of the electric field with specific (polar) molecules in the reaction medium that is not related to a macroscopic temperature effect. In order to probe the existence of nonthermal microwave effects, four synthetic transformations (Diels-Alder cycloaddition, alkylation of triphenylphosphine and 1,2,4-triazole, direct amide bond formation) were reevaluated under both microwave dielectric heating and conventional thermal heating. In all four cases, previous studies have claimed the existence of nonthermal microwave effects in these reactions. Experimentally, significant differences in conversion and/or product distribution comparing the conventionally and microwave-heated experiments performed at the same measured reaction temperature were found. The current reevaluation of these reactions was performed in a dedicated reactor setup that allowed accurate internal reaction temperature measurements using a multiple fiber-optic probe system. Using this technology, the importance of efficient stirring and internal temperature measurement in microwave-heated reactions was made evident. Inefficient agitation leads to temperature gradients within the reaction mixture due to field inhomogeneities in the microwave cavity. Using external infrared temperature sensors in some cases results in significant inaccuracies in the temperature measurement. Applying the fiber-optic probe temperature monitoring device, a critical reevaluation of all four reactions has provided no evidence for the existence of nonthermal microwave effects. Ensuring efficient agitation of the reaction mixture via magnetic stirring, no significant differences in terms of conversion and selectivity between experiments performed under microwave or - oil bath conditions at the same internally measured reaction temperatures were experienced. The observed effects were purely thermal and not related to the microwave field.

Full text of DOI:10.1021/jo7022697

Nonthermal Microwave Effects Revisited: On the Importance of
Internal Temperature Monitoring and Agitation in Microwave
Chemistry
M. Antonia Herrero, Jennifer M. Kremsner, and C. Oliver Kappe*
Christian Doppler Laboratory for MicrowaVe Chemistry (CDLMC) and Institute of Chemistry,
Karl-Franzens-UniVersity Graz, Heinrichstrasse 28, A-8010 Graz, Austria
oliVer.kappe@uni-graz.at
ReceiVed October 19, 2007
The concept of nonthermal microwave effects has received considerable attention in recent years and is
the subject of intense debate in the scientific community. Nonthermal microwave effects have been
postulated to result from a direct stabilizing interaction of the electric field with specific (polar) molecules
in the reaction medium that is not related to a macroscopic temperature effect. In order to probe the
existence of nonthermal microwave effects, four synthetic transformations (Diels-Alder cycloaddition,
alkylation of triphenylphosphine and 1,2,4-triazole, direct amide bond formation) were reevaluated under
both microwave dielectric heating and conventional thermal heating. In all four cases, previous studies
have claimed the existence of nonthermal microwave effects in these reactions. Experimentally, significant
differences in conversion and/or product distribution comparing the conventionally and microwave-heated
experiments performed at the same measured reaction temperature were found. The current reevaluation
of these reactions was performed in a dedicated reactor setup that allowed accurate internal reaction
temperature measurements using a multiple fiber-optic probe system. Using this technology, the importance
of efficient stirring and internal temperature measurement in microwave-heated reactions was made evident.
Inefficient agitation leads to temperature gradients within the reaction mixture due to field inhomogeneities
in the microwave cavity. Using external infrared temperature sensors in some cases results in significant
inaccuracies in the temperature measurement. Applying the fiber-optic probe temperature monitoring
device, a critical reevaluation of all four reactions has provided no evidence for the existence of nonthermal
microwave effects. Ensuring efficient agitation of the reaction mixture via magnetic stirring, no significant
differences in terms of conversion and selectivity between experiments performed under microwave or
oil bath conditions at the same internally measured reaction temperatures were experienced. The observed
effects were purely thermal and not related to the microwave field.
1
Introduction
the groups of Gedye and Giguere/Majetich in 1986, more than
3
500 articles have been published in this fast-moving and
Since the first published reports on the use of microwave
irradiation to accelerate organic chemical transformations by
(1) (a) Gedye, R.; Smith, F.; Westaway, K.; Ali, H.; Baldisera, L.;
Laberge, L.; Rousell, R. Tetrahedron Lett. 1986, 27, 279. (b) Giguere, R.
J.; Bray, T. L.; Duncan, S. M.; Majetich, G. Tetrahedron Lett. 1986, 27,
4945.
*
To whom the correspondence should be addressed. Phone: +43-316-380-
5
352. Fax: +43-316-380-9840.
10.1021/jo7022697 CCC: $40.75 © 2008 American Chemical Society
3
6
J. Org. Chem. 2008, 73, 36-47
Published on Web 12/07/2007

exciting field, today generally referred to as microwave-assisted
organic synthesis (MAOS).2,3 In many of the published ex-
amples, microwave heating has been shown to dramatically
reduce reaction times, increase product yields, and enhance
product purities by reducing unwanted side reactions compared
to conventional heating methods. The advantages of this
enabling technology have more recently also been exploited in
at the same measured reaction temperature.12 Today, it is
generally agreed upon that in many cases the observed enhance-
ments in microwave-heated reactions are in fact the result of
purely thermal/kinetic effects, in other words, are a consequence
of the high reaction temperatures that can rapidly be attained
when irradiating polar materials/reaction mixtures under closed
1
1,12
vessel conditions in a microwave field.
Similarly, the
4
the context of multistep total synthesis and medicinal chemistry/
existence of so-called “specific microwave effects” which cannot
be duplicated by conventional heating and result from the
5
drug discovery and have additionally penetrated fields such as
6
7
8
polymer synthesis, material sciences, nanotechnology, and
uniqueness of the microwave dielectric heating phenomenon is
9
11-13
biochemical processes. The use of microwave irradiation in
largely undisputed.
In this category fall, for example, (i)
chemistry has thus become such a popular technique in the
scientific community that it might be assumed that, in a few
years, most chemists will probably use microwave energy to
heat chemical reactions on a laboratory scale.10
the superheating effect of solvents at atmospheric pressure, (ii)
the selective heating of, e.g., strongly microwave absorbing
heterogeneous catalysts or reagents in a less polar reaction
medium, and (iii) the elimination of wall effects caused by
1
3
inverted temperature gradients.
In contrast, the subject of “nonthermal microwave effects”
Regardless of the relatively large body of published work in
1
-10
this area,
the exact reasons why microwave irradiation is
1
1
(
also referred to as athermal effects) is highly controversial
able to enhance chemical processes are still unknown. Since
the early days of microwave synthesis, the observed rate-
accelerations and sometimes altered product distributions com-
pared to conventionally heated experiments have led to specu-
lation on the existence of so-called “specific-” or “nonthermal”
12,14
and has led to heated debates in the scientific community.
Essentially, nonthermal effects have been postulated to result
from a proposed direct interaction of the electric field with
specific molecules in the reaction medium that is not related to
a macroscopic temperature effect. It has been argued, for
example, that the presence of an electric field leads to orientation
effects of dipolar molecules or intermediates and hence changes
the pre-exponential factor A or the activation energy (entropy
term) in the Arrhenius equation for certain types of reactions.12
Furthermore, a similar effect has been proposed for polar
reaction mechanisms, where the polarity is increased going from
the ground state to the transition state, resulting in an enhance-
ment of reactivity by lowering of the activation energy.12
Significant nonthermal microwave effects have been suggested
1
1,12
microwave effects.
Such effects have been claimed when
the outcome of a synthesis performed under microwave condi-
tions was different from the conventionally heated counterpart
(
2) (a) MicrowaVes in Organic Synthesis; Loupy, A., Ed.; Wiley-VCH:
Weinheim, Germany, 2002. (b) Hayes, B. L. MicrowaVe Synthesis:
Chemistry at the Speed of Light; CEM Publishing: Matthews, NC, 2002.
(c) MicrowaVe-Assisted Organic Synthesis; Lidstr o¨ m, P., Tierney, J. P., Eds.;
Blackwell Publishing: Oxford, U.K., 2005. (d) Kappe, C. O.; Stadler, A.
MicrowaVes in Organic and Medicinal Chemistry; Wiley-VCH: Weinheim,
Germany, 2005. (e) MicrowaVes in Organic Synthesis, 2nd ed.; Loupy, A.,
Ed.; Wiley-VCH: Weinheim, Germany, 2006. (f) MicrowaVe Methods in
Organic Synthesis; Larhed, M., Olofsson, K., Eds.; Springer: Berlin,
Germany, 2006. (g) MicrowaVe-Assisted Synthesis of Heterocycles; Van
der Eycken, E., Kappe, C. O., Eds.; Springer, Berlin, Germany, 2006.
12
for a wide variety of synthetic transformations.
It should be obvious from a scientific standpoint that the
question of nonthermal microwave effects needs to be addressed
in a serious manner, given the rapid increase in the use of
microwave technology in chemical sciences, in particular
organic synthesis. There is an urgent need to provide a scientific
rationalization for the observed effects and to investigate the
general influence of the electric field (and therefore of the
microwave power) on chemical transformations. This is even
more important if one considers engineering and safety aspects
once this technology moves from the small-scale laboratory
(
3) Recent reviews: (a) Kappe, C. O. Angew. Chem., Int. Ed. 2004, 43,
6
250 and references cited therein. (b) Hayes, B. L. Aldrichim. Acta 2004,
3
7, 66. (c) Roberts, B. A.; Strauss, C. R. Acc. Chem. Res. 2005, 38, 653.
(4) (a) Baxendale, I. R.; Ley, S. V.; Nessi, M.; Piutti, C. Tetrahedron
2
002, 58, 6285. (b) Artman, D. D., III; Grubbs, A. W.; Williams, R. M. J.
Am. Chem. Soc. 2007, 129, 6336. (c) Appukkuttan, P.; Van der Eycken, E.
In MicrowaVe Methods in Organic Synthesis; Larhed, M., Olofsson, K.,
Eds.; Springer: Berlin, Germany, 2006; Chapter 1, pp 1-47.
(
5) (a) Larhed, M.; Hallberg, A. Drug DiscoVery Today 2001, 6, 406.
(
b) Wathey, B.; Tierney, J.; Lidstr o¨ m, P.; Westman, J. Drug DiscoVery
Today 2002, 7, 373. (c) Al-Obeidi, F.; Austin, R. E.; Okonya, J. F.; Bond,
D. R. S. Mini-ReV. Med. Chem. 2003, 3, 449. (d) Shipe, W. D.; Wolkenberg,
S. E.; Lindsley, C. W. Drug DiscoVery Today: Technol. 2005, 2, 155. (e)
Kappe, C. O.; Dallinger, D. Nature ReV. Drug DiscoV. 2006, 5, 51.
10
work to pilot or production scale instrumentation.
(11) For a more detailed definition and examples for thermal, specific,
and nonthermal microwave effects, see: Kappe, C. O.; Stadler, A.
MicrowaVes in Organic and Medicinal Chemistry; Wiley-VCH, Weinheim,
2005; Chapter 2, pp 9-28. See also refs 3a and 12.
(12) For leading reviews in the field, see: (a) Perreux, L.; Loupy, A.
Tetrahedron 2001, 57, 9199. (b) Perreux, L.; Loupy, A. In MicrowaVes in
Organic Synthesis; Loupy, A., Ed.; Wiley-VCH: Weinheim, Germany,
2002; Chapter 3, pp 61-114. (c) Perreux, L.; Loupy, A. In MicrowaVes in
Organic Synthesis, 2nd ed.; Loupy, A., Ed.; Wiley-VCH: Weinheim,
Germany, 2006; Chapter 4, pp 134-218. (d) De La Hoz, A.; Diaz-Ortiz,
A.; Moreno, A. Chem. Soc. ReV. 2005, 34, 164. (e) De La Hoz, A.; Diaz-
Ortiz, A.; Moreno, A. In MicrowaVes in Organic Synthesis, 2nd ed.; Loupy,
A., Ed.; Wiley-VCH: Weinheim, Germany, 2006; Chapter 5, pp 219-
277.
(13) It should be emphasized that specific microwave effects are
essentially still the result of a thermal phenomenon (that is, a change in
temperature compared to heating by standard convection methods), although
it may be difficult to experimentally determine the exact reaction temperature
in these cases, for example, on the surface of a strongly microwave-
absorbing catalyst that is selectively superheated by microwave irradiation.
For examples, see refs 11 and 12.
(6) (a) Bogdal, D.; Penczek, P.; Pielichowski, J.; Prociak, A. AdV. Polym.
Sci. 2003, 163, 193. (b) Wiesbrock, F.; Hoogenboom, R.; Schubert, U. S.
Macromol. Rapid Commun. 2004, 25, 1739. (c) Hoogenboom, R.; Schubert,
U. S. Macromol. Rapid Commun. 2007, 28, 368. (d) Bogdal, D.; Prociak,
A. MicrowaVe-Enhanced Polymer Chemistry and Technology; Blackwell
Publishing: Oxford, U.K., 2007.
(
7) (a) Barlow, S.; Marder, S. R. AdV. Funct. Mater. 2003, 13, 517. (b)
Zhu, Y.-J.; Wang, W. W.; Qi, R.-J.; Hu, X.-L. Angew. Chem., Int. Ed. 2004,
4
1
3, 1410. (c) Perelaer, J.; de Gans, B.-J.; Schubert, U. S. AdV. Mater. 2006,
8, 2101. (d) Jhung, S. H.; Jin, T.; Hwang, Y. K.; Chang, J.-S. Chem. Eur.
J. 2007, 13, 4410.
8) Tsuji, M.; Hashimoto, M.; Nishizawa, Y.; Kubokawa, M.; Tsuji, T.
Chem. Eur. J. 2005, 11, 440.
9) (a) Collins, J. M.; Leadbeater, N. E. Org. Biomol. Chem. 2007, 5,
141. (b) Lill, J. R.; Ingle, E. S.; Liu, P. S.; Pham, V.; Sandoval, W. N.
Mass Spectrom. ReV. 2007, 26, 657.
10) For reviews on large scale microwave synthesis, see: (a) Kremsner,
(
(
1
(
J. M.; Stadler, A.; Kappe, C. O. Top. Curr. Chem. 2006, 266, 233. (b)
Glasnov, T. N.; Kappe, C. O. Macromol. Rapid Commun. 2007, 28, 395.
(
c) Ondruschka, B.; Bonrath, W.; Stuerga, D. In MicrowaVes in Organic
Synthesis, 2nd ed.; Loupy, A., Ed.; Wiley-VCH: Weinheim, Germany, 2006;
(14) (a) Kuhnert, N. Angew. Chem., Int. Ed. 2002, 41, 1863. (b) Strauss,
C. R. Angew. Chem., Int. Ed. 2002, 41, 3589.
Chapter 2, p 62.
J. Org. Chem, Vol. 73, No. 1, 2008 37

Herrero et al.
In continuation of our general interest in microwave effects,15-17
we herein describe a detailed reevaluation of nonthermal
microwave effects and the influence of the microwave field on
chemical reactions. For this purpose, we have selected four
representative chemical transformations that were previously
reported to exhibit significant nonthermal microwave effects
type of fiber-optic device used and on the protective shielding
employed with the probe. For the standard probe/shielding used
in conjunction with the CEM Discover, we have experienced
response times up to 13 s before the correct temperature was
displayed by the sensor (Figure S1 in the Supporting Informa-
tion). In some instances, for example, for very fast reactions20
or for polar reaction media that are very rapidly heated by
microwave irradiation (see below), this delay time may lead to
an incorrect determination of the actual reaction temperature
in the initial phase of the experiment. For the current studies,
we have therefore employed an independent fiber-optic probe
that minimizes the response time to less than 7 s when using
an appropriate shielding device. In exceptional cases, this sensor
can also be used without protective shielding providing a correct
temperature measurement within less than 1 second (see Figure
S1 in the Supporting Information).
(see below). In all four cases, carefully conducted control
experiments using conventional heating at the same reaction
temperature as monitored during the microwave experiments
had previously shown large differences not only in yield/
conversion but also in product selectivity. For all examples, the
differing results between conventional and microwave heating
were rationalized by nonthermal microwave effects and in each
case a direct intervention of the electric field in the reaction
pathway was suggested. In order to reach a general conclusion
on the existence of nonthermal microwave effects, we have
reinvestigated these high-profile cases using a specialized fiber-
optic temperature probe that allows simultaneous temperature
detection at different positions of the reaction mixture.
Even more important, our previous experience in this field
has revealed that the location of the fiber-optic sensor in the
microwave-heated reaction vessel is of great importance. For
example, we have found that differences of 15-40 °C in the
measured temperature can be experienced heating microwave
absorbing solvents in standard microwave process vials utilizing
a single-mode microwave reactor, depending if the temperature
Results and Discussion
Microwave versus Conventional Heating and Tempera-
ture Measurement. In order to accurately compare the results
obtained by direct microwave heating with the outcome of a
conventionally heated reaction, we have recently described a
reactor system that allows us to perform both types of
transformations in the same reaction Vessel and to monitor the
internal reaction temperature in both experiments directly with
a fiber-optic probe device.17 Monitoring reaction temperatures
in microwave-assisted reactions by conventional infrared sensors
on the outside vessel wall is not an acceptable technique if an
accurate temperature profile for comparison studies needs to
21
probe is placed close to the bottom or in the middle of the vial.
In order to monitor these effects more accurately we have now
devised a multiple fiber-optic probe measurement device that
can be attached to a standard 10 mL microwave vial allowing
simultaneous temperature measurement at up to three different
positions inside the reaction vial (Figure 1b).
Using this system, surprisingly large temperature gradients
on heating pure organic solvents inside a common single-mode
microwave reactor with constant microwave power were
revealed. For example, using a sample of 5 mL of NMP as
solvent, the temperature difference between the bottom and the
middle of the vessel was 30 °C (Figure 1a). These internal
temperature gradients could also be detected on the outside
surface of the vessel by an IR camera after removing the vessel
from the microwave cavity (Figure S2, Supporting Information).
Gratifyingly, in the case of NMP these gradients could be
minimized (deviation <6 °C) by efficient magnetic stirring of
the reaction mixture (Figure S4, Supporting Information). While
stirring using the built-in magnetic stirring system is easily
possible for a pure solvent, this may not be the case for a
heterogeneous or viscous reaction mixture or in case where the
reaction is performed under solvent-free or dry-media conditions.
In order to verify this point, a similar experiment was performed
by heating a sample of Montmorillonite K10 Clay (a commonly
used inorganic solid support material for dry-media microwave
1
6-18
be obtained.
Our original system utilized a CEM Discover
single-mode microwave reactor equipped with a fiber-optic
probe provided by the instrument manufacturer for directly
monitoring the internal reaction temperature in a 10 mL sealed
1
7
reaction vessel. This setup can be either immersed into the
cavity of the microwave reactor or into a preheated and
temperature-equilibrated oil or metal bath placed on a magnetic
stirrer/hotplate. In both cases, the software of the microwave
instrument is recording the internal temperature. This system
has the advantage that the same reaction vessel and the same
method of temperature measurement is used. In this way, all
parameters apart from the mode of heating are identical, and
therefore, a fair comparison between microwave heating and
thermal heating can generally be made.1
9
For the high accuracy required for the current study, we have
made two significant changes to the measurement system
described above. During our recent work employing fiber-optic
temperature probes,16,17 we noticed that the response time of
the fiber-optic sensor can be quite long depending both on the
22
chemistry) using the same experimental setup. In this case,
the temperature gradients could not be eliminated by magnetic
(
20) For very rapid microwave-assisted reactions in the order of seconds,
(
15) (a) Stadler, A.; Kappe, C. O. J. Chem. Soc., Perkin Trans. 2 2000,
see: (a) Enquist, P.-A.; Nilsson, P.; Larhed, M. Org. Lett. 2003, 5, 4875.
(b) Leadbeater, N. E.; Smith, R. J. Org. Lett. 2006, 8, 4589. (c) Leadbeater,
N. E.; Smith, R. J.; Barnard, T. M. Org. Biomol. Chem. 2007, 5, 822. (d)
Leadbeater, N. E.; Smith, R. J. Org. Biomol. Chem. 2007, 5, 2770.
(21) For a theoretical prediction of this effect and modeling of the field
distribution in a single-mode cavity, see: (a) Berlan, J. Rad. Phys. Chem.
1995, 45, 581. (b) Saillard, R.; Poux, M.; Berlan, J.; Audhuy-Peaudecerf,
M. Tetrahedron 1995, 51, 4033. For a previous observation of this
phenomenon using a prototype single-mode cavity, see: (c) Kaiser, N.-F.
K. Ph.D. thesis, Uppsala University, 2001, p 37.
(22) (a) Loupy, A.; Petit, A.; Hamelin, J.; Texier-Boullet, F.; Jacquault,
P.; Math e´ , D. Synthesis 1998, 1213. (b) Varma, R. S. Green Chem. 1999,
43. (c) Kidawi, M. Pure Appl. Chem. 2001, 73, 147. (d) Varma, R. S. Pure
Appl. Chem. 2001, 73, 193. (e) Varma, R. S. Tetrahedron 2002, 58, 1235.
1
363. (b) Stadler, A.; Kappe, C. O. Eur. J. Org. Chem. 2001, 919. (c)
Strohmeier, G. A.; Kappe, C. O. J. Comb. Chem. 2002, 4, 154. (d) Garbacia,
S.; Desai, B.; Lavastre, O.; Kappe, C. O. J. Org. Chem. 2003, 68, 9136.
(
(
16) Kremsner, J. M.; Kappe, C. O. J. Org. Chem. 2006, 71, 4651.
17) Hosseini, M.; Stiasni, N.; Barbieri, V., Kappe, C. O. J. Org. Chem.
2
007, 72, 1417.
18) (a) N u¨ chter, M.; Ondruschka, B.; Bonrath, W.; Gum, A. Green
(
Chem. 2004, 6, 128. (b) N u¨ chter, M.; Ondruschka, B.; Weiâ, D.; Bonrath,
W.; Gum, A. Chem. Eng. Technol. 2005, 28, 871. (c) Leadbeater, N. E.;
Pillsbury, S. J.; Shanahan, E.; Williams, V. A. Tetrahedron 2005, 61, 3565.
(19) For related publications using this technique, see: (a) Hostyn, S.;
Maes, B. U. W.; Van Baelen, G.; Gulevskaya, A.; Meyers, C.; Smits, K.
Tetrahedron 2006, 62, 4676. (b) Vanier, G. S. Synlett 2007, 131.
38 J. Org. Chem., Vol. 73, No. 1, 2008

FIGURE 1. (a) Temperature profiles for a sample of 5 mL of NMP contained in a 10 mL quartz vessel equipped with three internal fiber-optic
sensors. The sample was irradiated with constant 50 W magnetron output power without stirring for 40 min (CEM Discover). Shown are the
profiles for the three internal fiber-optic probes and for the external IR sensor located at the bottom of the instrument (see also Figure S3, Supporting
Information). (b) Multiple fiber-optic temperature probe assembly (see the Experimental Section for more details).
stirring since the stirring proved to be inefficient (Figure S5,
Supporting Information).23 It should also be noted that in all
cases the calibrated IR sensor shows a significant deviation even
from the fiber-optic probe located in close proximity near the
bottom of the vessel (see Figure 1b). This once again highlights
the problem of using external IR sensors for temperature
model reactions in order to reevaluate the existence of non-
thermal microwave effects in organic synthesis. According to
1
2
Perreux and Loupy, nonthermal microwave effects are most
likely to be observed for reactions that follow polar reaction
pathways via polar transition states. It has been suggested that
if the polarity of a system is increased from the ground state to
the transition state, acceleration can be observed by an increase
in the material-wave interactions during the course of the
reaction (leading to a stabilization of the polar transition state
by the electric field). The most frequently encountered examples
involve reactions between neutral molecules where dipoles are
developed in the transition state and/or in the product.12 Second,
to observe nonthermal microwave effects transformations should
possess high energies of activation with late product-like
transition states along the reaction coordinate in agreement with
1
6-18
measurements in microwave-heated vessels.
The fast
responding multiple fiber-optic probe system was also utilized
to investigate the effect of simultaneous external cooling by
compressed air on the internal temperatures.24 Also in this case,
the significant temperature gradients between the different
positions in the vial persisted (Figure S6, Supporting Informa-
tion). In contrast, a control experiment immersing the fiber-
optic probe device shown in Figure 1b (5 mL of NMP) into a
preheated oil bath confirmed that no gradients are observed on
such as small scale using conventional heating by conduction/
convection phenomena, regardless if the mixture was stirred or
not (Figure S7, Supporting Information). These temperature
monitoring studies demonstrate that microwave heating in high
field density single-mode cavities is in fact not as homogeneous
1
2
the Hammond postulate.
Another important factor in connection with nonthermal
microwave effects is the proper choice of the solvent. Micro-
wave heating generally relies on the ability of the reaction
mixture/solvent to efficiently absorb microwave energy, taking
advantage of “microwave dielectric heating” phenomena such
as dipolar polarization or ionic conduction mechanisms.26 The
ability of a specific material or solvent to convert microwave
energy into heat at a given frequency and temperature is
determined by the so-called loss tangent (tan δ), expressed as
the quotient, tan δ ) ꢀ′′/ꢀ′, where ꢀ′′ is the dielectric loss,
indicative of the efficiency with which electromagnetic radiation
is converted into heat, and ꢀ′ is the dielectric constant, describing
the ability of molecules to be polarized by the electric field.26
A reaction medium with a high tan δ at the standard operating
frequency of a microwave synthesis reactor (2.45 GHz) is
required for good absorption and, consequently, for efficient
3
as often portrayed and that extreme care must be taken in
determining the proper reaction temperature in these experi-
ments, especially in those cases where adequate mixing cannot
be assured.25
Selection of Chemistry Examples. Having a reliable tem-
perature monitoring system for microwave-assisted reactions in
place, we next set out to carefully define a set of appropriate
(23) For temperature gradients in the heating of solids under microwave
conditions in multimode cavities, see: (a) Stuerga, D.; Gaillard, P.
Tetrahedron 1996, 52, 5505. (b) Bogdal, D.; Bednarz, S.; Lukasiewicz, M.
Tetrahedron 2006, 62, 9440.
(24) Related to the issue of nonthermal microwave effects is the concept
that simultaneous external cooling of the reaction mixture (or maintaining
subambient reaction temperatures) while heating by microwaves can in some
cases lead to an enhancement of the overall process. Here, the reaction
vessel is cooled from the outside by compressed air or with the aid of a
cooling fluid while being irradiated by microwaves. This allows a higher
level of microwave power to be directly administered to the reaction mixture
thereby potentially enhancing nonthermal microwave effects that rely on
the electric field strength. At the same time, overheating will be prevented
by continuously removing heat. For recent examples using this technique,
see ref 17 and references cited therein.
27
heating. In the case of a high absorbing solvent, most of the
(25) For a recent qualitative description of problems resulting from
agitation issues in microwave chemistry involving single-mode cavities,
see: Moseley, J. D.; Lenden, P.; Thomson, A. D.; Gilday, J. P. Tetrahedron
Lett. 2007, 48, 6084.
(26) (a) Gabriel, C.; Gabriel, S.; Grant, E. H.; Halstead, B. S.; Mingos,
D. M. P. Chem. Soc. ReV. 1998, 27, 213. (b) Mingos, D. M. P.; Baghurst,
D. R. Chem. Soc. ReV. 1991, 20, 1. See also refs 2 and 3.
J. Org. Chem, Vol. 73, No. 1, 2008 39

Herrero et al.
microwave energy will in fact be absorbed by the solvent and
converted into heat, thereby masking any potential nonthermal
microwave effects as a result of material-wave interactions.
Therefore, the use of nonpolar, low microwave-absorbing
solvents or the use of solvent-free conditions has been sug-
gested.12
TABLE 1. Comparison of Microwave Heating and Conventional
Heating for the Diels-Alder Reaction of Pyrone 1 and
Phenylacetylene (2)a
In order to select the most suitable and meaningful examples
for our study, we have introduced a number of additional criteria.
All chosen published chemistry examples have been conducted
in dedicated single-mode microwave reactors utilizing open
vessel procedures using temperature control by either built-in
IR sensors or fiber-optic probes. The performed control experi-
ments using conventional heating at the same temperature range
led to a large difference between the microwave and the
conventionally heated run. We have deliberately chosen ex-
amples where the reported differences were large in order to
minimize the chance of experimental error. In addition, we have
selected examples where not only the conversion was different
in the microwave run, but in some cases also the product
distribution. Because of the inherent difficulties experienced in
the temperature measurement using solids under microwave
irradiation conditions (see above), dry-media reactions were
excluded from our studies. In contrast, we haveswhere
appropriatesselected cases where the reaction mixtures were
as homogeneous as possible, using either low-absorbing solvents
or solvent-free conditions. This would ensure that the mixtures
could be stirred, therefore avoiding potential problems with
temperature gradients as described above. Based on all of these
criteria, four representative examples for investigating nonther-
mal microwave effects were chosen. These are discussed in
detail in the following sections.
Tb (°C)
time (min)
30
heating methodc
conv (%)
d
entry
1
140
oil bath
MW
oil bath
MW
oil bath
MW
oil bath
MW
10
14
20
23
28
31
41
43
54
57
2
3
4
5
140
140
140
140
60
90
150
240
oil bath
MW
a
Pyrone 1 (6.5 mmol) and phenylacetylene (2, 9.1 mmol, 1.4 equiv)
were reacted neat in a 10 mL microwave process vial (Figure 1b). For further
details, see the Experimental Section. b Internal reaction temperature
measured by fiber-optic sensor. The heating profiles are reproduced in the
Supporting Information (Figures S8 and S9). c For details of the experi-
mental setup for microwave and oil bath heating, refer to the Experimental
Section. d Conversion based on 1H NMR analysis of the crude reaction
mixture.
accurately monitor internal reaction temperatures. The same
reaction conditions as reported previously were chosen with the
exception that the reaction temperature was adjusted to ca.
140 °C instead of 150 °C, since the boiling point of phenyl-
acetylene is 142-144 °C and in our hands experiments at 150 °C
internal temperature were difficult to execute. The cycloadditions
were performed on a 6.5 mmol scale in a 10 mL Pyrex
microwave vial. Since the reaction mixture (ca 2.0 mL) in this
particular case is a low viscosity liquid, magnetic stirring was
very efficient and therefore only one fiber-optic probe needed
to be employed. This allowed us to run the device under open
vessel conditions since the formed carbon dioxide vented
through the two unused fiber-optic probe inlet ports. For
performing the microwave-irradiated experiment the device was
introduced into the cavity of the Discover microwave reactor
and irradiated at 140 °C for 4 h using the calibrated built-in IR
temperature sensor as the lead sensor (temperature control, 80
W maximum magnetron output power). Simultaneously, the true
internal reaction temperature was monitored using the fiber-
optic probe. In this case, a very good agreement between the
external IR and internal fiber-optic temperature sensor was
obtained (139.5 vs 141.8 °C average temperature, see Figure
S8, Supporting Information). In order to obtain a kinetic profile
Diels-Alder Cycloaddition of 5-Methoxycarbonyl-2-py-
rone and Phenylacetylene. In a 2004 publication, the microwave-
assisted cycloaddition of pyrone 1 with phenylacetylene (2) was
2
8
reported. The resulting aromatized product 4 was obtained as
a single regioisomer by release of carbon dioxide from the
primary cycloadduct 3 (Table 1). The experiment was conducted
in an open vessel using the acetylene dienophile in slight excess
(1.4 equiv) without added solvent using a dedicated single-mode
microwave reactor at 150 °C (calibrated external IR sensor,
mechanical stirring). After 3 h of microwave irradiation at
150 °C (temperature control mode), a 64% conversion to product
4
was obtained. Importantly, the exact same experiment when
carried out in a preheated oil bath at the same temperature only
showed 19% conversion.28 This significant apparent “microwave
effect” was rationalized by the authors taking into account
theoretical calculations predicting a strong asynchronicity in the
reaction mechanism with a large enhancement in the dipole
2
8
1
moments from the ground states to the transition state.
of the reaction, samples for H NMR and HPLC analysis were
Stabilization of polar transition states by the microwave field
has been argued to reduce the required energy of activation and
therefore to accelerate these transformations.12
We have repeated the cycloaddition described in Table 1
using the fiber-optic probe setup shown in Figure 1b to
withdrawn after 30, 60, 90, 150, and 240 min through the unused
inlet ports.
For performing the conventionally heated control experiment,
the reaction and sample withdrawal regime was repeated, but
this time the assembly (Figure 1b) was immersed into a
preheated oil bath (140 °C). All other parameters such as
reaction scale, stirring speed, and the reaction vessel were kept
the same. As shown in Figure S9 (Supporting Information), the
internally measured reaction temperatures in the oil-bath and
the microwave experiment were nearly identical (141.0 vs
141.8 °C). Based on the previously published data for this
reaction (see above), we were surprised to see that in our hands
the results in terms of conversion wereswithin experimental
errorsnearly identical (Table 1, entries 1-5). No evidence for
(
27) Solvents used for microwave synthesis can be classified as high
(
(
tan δ > 0.5; for example, ethanol, DMSO, methanol, formic acid), medium
tan δ 0.1-0.5; for example, acetic acid, 1,2-dichlorobenzene, NMP, DMF,
water), and low microwave absorbing (tan δ < 0.1; for example, chloroform,
ethyl acetate, THF, dichloromethane, toluene, o-xylene, hexane). Other
common solvents without a permanent dipole moment such as carbon
tetrachloride, benzene, p-xylene and dioxane can be considered as micro-
wave transparent.
(
28) Loupy, A.; Maurel, F.; Sabati e´ -Gogov a´ , A. Tetrahedron 2004, 60,
1
683.
40 J. Org. Chem., Vol. 73, No. 1, 2008

TABLE 2. Kinetic Analysis of the Alkylation of
Triphenylphosphine with Benzyl Chloride under Solvent-Free
Conditions Using Conventional Heatinga
reactors under open vessel conditions (precalibrated IR and fiber-
optic temperature measurement). For example, performing the
solvent-free reaction of benzyl chloride (5) and triphenylphos-
phine (6) in a preheated oil bath at 100 °C for 10 min, the
reaction outcome was very different to the microwave irradiation
experiment using the same temperature/time conditions (24
3
0
versus 78%, respectively). This experimental difference in
favor of the microwave run was rationalized by the occurrence
of a nonthermal microwave effect involving the stabilization
entry
Tb (°C)
105
105
105
time (min)
yield (%)
c
1
2
3
4
5
6
7
10
20
45
90
10
10
10
25
59
78
98
87
91
97
30,31
of the polar transition state by the electric field.
During our reevaluation of this transformation, we first
concentrated on the solvent-free reaction conditions. A kinetic
analysis at 105 °C internal reaction temperature using a
preheated oil bath (105 °C) and the experimental setup displayed
in Figure 1b is shown in Table 2 (entries 1-4). In our hands,
the isolated 25% yield of phosphonium salt 7 after 10 min nicely
105
d
d
d
120 (156 °C)
150 (210 °C)
170 (253 °C)
a
Reactions were performed on a 10 mmol scale in the reaction vessel
b
shown in Figure 1b immersed into a preheated oil bath. Temperature of
the preheated oil bath. Isolated yields of pure product (see the Experimental
Section for details). Maximum internal temperatures monitored by fiber-
c
30
agrees with the result previously reported (24%). Full conver-
d
sion at 105 °C providing a nearly quantitative isolated product
yield under thermal conditions required 90 min (Table 2,
entry 4).
optic sensors (see the main text and the Supporting Information for a detailed
discussion).
It has to be noted that when immersing the vessel containing
the neat mixture of the two starting materials (Figure 1b, ca. 5
mL volume) into the preheated oil bath the reaction mixture
becomes completely homogeneous after a few seconds using
magnetic stirring. After approximately 1 min, the insoluble
phosphonium salt product starts to precipitate from the reaction
mixture which ultimately leads to complete solidification of the
reaction mixture inside the vessel and failure of the magnetic
stirring system (Figure S10, Supporting Information). While at
ca. 105 °C bath temperature this does not create a problem, we
have observed that at higher bath temperatures (>120 °C)
thermal runaways will occur leading to significantly higher
internal reaction temperatures than those of the bath fluid (Table
any nonthermal microwave effects could therefore be found in
this reaction. The slightly higher conversions seen in the
microwave runs (2-4%) may be attributed to the different
heating profiles experienced in the very early phase of the
microwave heated experiment (Figure S9, Supporting Informa-
tion).
Additional experiments were aimed at achieving full conver-
sion in this cycloaddition process. For this purpose, the reaction
was performed under sealed vessel microwave conditions at
higher temperatures using a Discover Labmate microwave
reactor including both standard IR temperature and online
pressure monitoring. Gratifyingly, a 96% conversion based on
1
H NMR measurements was obtained after 3 h at 200 °C,
2
and Figure S11, Supporting Information). This is most likely
resulting in a 71% isolated product yield of 4 after flash
chromatography. It is interesting to note that the pressure build-
up in the reaction vessel (ca 7.5 bar) is not preventing the
cycloaddition.29 Again, an identical experiment using oil bath
heating provided a more or less identical result (98% conversion,
a result of the strongly exothermic character of this alkylation
reaction and/or of the heat of crystallization resulting from
product precipitation as demonstrated by Differential Scanning
Calorimetry (DSC) measurements (Figure S12, Supporting
Information). For example, using a 150 °C external oil bath
temperature produces an internal peak temperature of 210 °C.
Under those conditions high isolated product yields can be
obtained within a 10 min reaction time frame (Table 2).
When performing the same solvent-free reaction under
microwave irradiation conditions the thermal runaway became
a serious issue. Since during the course of the alkylation process,
the reaction medium becomes strongly microwave absorbing
75% isolated product yield).
Nucleophilic Substitution of Benzyl Chloride with Triph-
enylphosphine. As a second example, we have chosen the
nucleophilic substitution of benzyl chloride (5) with triph-
enylphosphine (6) leading to phosphonium salt 7 (Table 2). In
contrast to the first example described above, this transformation
involves not only a polar transition state but also leads to a
polar ionic product.30 Moreover, the product-like transition state
26
due to the polar character of the initially still dissolved
3
0
is positioned late along the reaction coordinate. Therefore,
according to the current theory on microwave effects outlined
above, this transformation involving the reaction of two neutral
reagents to give a polar product would be ideally suited for
observing a nonthermal microwave effect by dipole-dipole
phosphonium salt 7, rapid temperature increases are experienced
in the first few minutes of the experiment. This is particularly
true when comparatively high initial microwave power levels
are used. However, even at very low magnetron output power
(for example 10 W) thermal runaways were experienced (Figure
1
2
electrostatic interaction/stabilization of the transition state.
S13, Supporting Information). As shown in Figure 2, the use
of 50 W maximum microwave power leads to significant
thermal runaways. Using a maximum set temperature of 100 °C
on the CEM Discover instrument (mimicking the previously
Indeed, in a 2004 publication, significant differences between
performing the reaction shown in Table 2 in an oil-bath and
under microwave conditions were observed.30 Alkylations have
been conducted either solvent-free or using xylene as a
microwave-transparent solvent employing dedicated single-mode
30
published conditions), the fiber-optic sensors revealed that the
actual internal reaction temperature was significantly higher than
the presumed reaction temperature recorded by the IR sensor
on the surface of the reaction vessel. In addition, the three
(29) For examples of microwave-assisted transformations that require
open vessel processing, see: Razzaq, T.; Kappe, C. O. Tetrahedron Lett.
2
007, 48, 2513 and references cited therein.
30) Cvengros, J.; Toma, S.; Marque, S.; Loupy, A. Can. J. Chem. 2004,
2, 1365.
(
(31) For related examples, see: (a) Kiddle, J. J. Synth. Commun. 2001,
31, 3377. (b) Kiddle, J. J. Tetrahedron Lett. 2000, 41, 1339.
8
J. Org. Chem, Vol. 73, No. 1, 2008 41

Herrero et al.
FIGURE 2. Temperature and microwave power profiles for the microwave-assisted solvent-free alkylation of triphenylphosphine with benzyl
chloride (Table 2) (CEM Discover single-mode reactor, reaction vessel; see Figure 1b, 50 W maximum magnetron output power, temperature
control using the built-in IR sensor). Shown are the profiles for the three internal fiber-optic temperature probes (FO), for the external IR sensor
located at the bottom of the instrument (IR), and for the magnetron output power (Power). While the IR sensor controlling the magnetron power
continuously measures 100 °C corresponding to the correct set temperature, the internal fiber-optic probes reveal the inhomogeneous temperature
distribution within the reaction vessel with a maximum recorded temperature of 165 °C.
internal probes (Figure 1b) displayed considerable differences
in the recorded temperature profiles and those experiments
furthermore proved difficult to reproduce. This is mainly a result
of the inherent temperature gradients experienced in the
microwave vessel/cavity (see above) and a consequence of the
inability to efficiently stir the reaction mixture once product
precipitation takes place (Figure S10, Supporting Information).
It is important to point out that without multiple internal
temperature monitoring, these severe problems connected to
temperature measurement in this particular reaction would not
be evident. As can be seen in Figure 2, the standard external
IR sensor displays the proper “reaction temperature” of 100 °C
which clearly does not represent the true reaction temperature
inside the microwave vessel. Not surprisingly, isolated product
yields from these microwave experiments were far higher than
one would expect from the nominal reaction temperature of
TABLE 3. Comparison of Microwave Heating and Conventional
Heating for the Reaction of Triphenylphosphine with Benzyl
Chloride (Table 2) in p-Xylenea
entry
1
Tb (°C)
time (min)
5
heating method
oil bath
yield (%)
c
140
5
4
MW
2
3
140
140
15
30
oil bath
MW
oil bath
11
8
17
14
14
18
43
37
53
56
73
77
MW
MW (sim cooling)d
MW (SiC)e
oil bath
MW
4
5
6
140
200
200
90
10
30
oil bath
MW
oil bath
MW
1
00 °C recorded by the IR sensor. For the experiment shown
a Triphenylphosphine (2.0 mmol) and benzyl chloride (2.0 mmol) in 2
mL of p-xylene were reacted in a 10 mL microwave process vial (Figure
in Figure 2 (maximum internal temperature 165 °C, ca. 12 min
reaction time) a 85% isolated yield of phosphonium salt 7 was
1
b). For further details, see the Experimental Section. b Internal reaction
temperature measured by two fiber-optic sensors. Some representative
heating profiles are reproduced in the Supporting Information (Figure S14).
3
0
obtained, matching the 78% yield previously obtained.
The results described above already indicate that also in this
case little evidence for the existence of a nonthermal microwave
effect could be found. Since, however, the solvent-free experi-
ments proved difficult to reproduce, we decided to additionally
re-evaluate the data previously obtained using p-xylene as
solvent.30 As both starting materials are completely soluble in
the microwave transparent solvent p-xylene, the reaction mixture
can be stirred efficiently, even after the insoluble phosphonium
salt 7 has precipitated. Using p-xylene as solvent for the
alkylation described in Table 2, the previous study has found
that while the reaction at 140 °C for 30 min in an oil bath
resulted in only 11% isolated yield of the phosphonium salt,
the microwave experiment using the same conditions led to a
c
Isolated yields of pure product (see the Experimental Section for details).
Applying simultaneous cooling using compressed air (ref 24). In the
d
e
presence of a SiC passive heating element (ref 16).
reaction vessel (Figure S14, Supporting Information). In addition
to the standard temperature controlled microwave runs two
additional experiments were performed exposing the reaction
mixture to microwave heating at 140 °C for 30 min: (i) By
applying the concept of simultaneous cooling of the reaction
24
mixture by compressed air, significantly more microwave
power was delivered to the reaction mixture as opposed to the
non-cooled experiment (292 versus 105 W) (Table 3, entry 3,
see also Figure S14b and S14a, Supporting Information). (ii)
Alternatively, the alkylation was also performed in the presence
of a strongly microwave absorbing silicon carbide (SiC)
3
3% product yield.30 The results of our control experiments
between microwave and oil bath heating employing again the
multiple fiber-optic probe device are shown in Table 3.
In contrast to the previously published data,30 we find that
for experiments carried out at 140 °C isolated product yields
are rather similar between microwave and conventionally heated
experiments (1-6% difference in yield). Since here efficient
stirring was possible, the IR and fiber-optic probe measurements
nicely matched and no temperature gradients developed in the
16
material. Since the SiC “passive heating element” is rapidly
heated by microwaves, less overall microwave power is used
to heat the reaction mixture (51 versus 105 W) (Table 3, entry
32
3
, see also Figure S14c and S14a, Supporting Information).
(32) Kremsner, J. M.; Stadler, A.; Kappe, C. O. J. Comb. Chem. 2007,
9, 285.
42 J. Org. Chem., Vol. 73, No. 1, 2008

SCHEME 1. Alkylation of 1,2,4-Triazole with
,2′,4′-Trichloroacetophenone
TABLE 4. Comparison of Microwave Heating and Conventional
Heating for the Reaction of 1,2,4-Triazole with
2
2
,2′,4′-Trichloroacetophenone (Scheme 1) in p-Xylenea
entry
Tb (°C)
time (min)
20
heating method
N1/N4/N1,4 (%)
c
1
2
3
4
5
6
140
oil bath
MW
oil bath
MW
oil bath
MW
oil bath
MW
oil bath
MW
oil bath
MW
42/33/25
43/31/26
38/32/30
41/32/28
56/20/24
57/20/23
97/1/2
140
170
200
200
200
60
20
20
40
60
If the electric field would have any influence on the reaction
pathway for this transformation (nonthermal microwave effect),
such dramatic differences in the used microwave power would
undoubtedly result in a noticeable change in reaction yields.
Since this is apparently not the case (Table 3, entry 3) we
conclude that no nonthermal microwave effects are present in
these reactions. As expected, higher reaction temperatures and
prolonged reaction times led to increased product yields, with
again no major differences between oil bath and microwave
heating being seen (Table 3, entries 5 and 6).
98/1/1
99/0/1
99/0/1
>99/0/0
>99/0/0
a
1
,2,4-Triazole (0.9 mmol) and 2,2′,4′-trichloroacetophenone (1.0 mmol)
in 0.5 mL of p-xylene were reacted in a 10 mL microwave process vial
(Figure 1b). For further details, see the Experimental Section. Reaction
b
temperature measured by IR sensor (microwave) and digital thermometer
1
(
oil bath fluid). c Product distributions based on H NMR analysis (Figure
S17, Supporting Information).
Alkylation of Triazole with 2,2′,4′-Trichloroacetophenone.
The phenacylation of 1,2,4-triazole (8) with 2,2′,4′-trichloro-
acetophenone (9) was selected as an example for our studies
since in this case not only different conversions between oil
bath and microwave heating have been experienced, but also
the regioselectivity in the alkylation process was found to be
strikingly influenced by the mode of heating and the solvent
will be susceptible to differential heating by microwave irradia-
tion.
3
5
In order to ensure efficient agitation and reproducibility, we
therefore ultimately used a total reaction volume of less than 1
mL inside the reaction vessel (0.5 mL of solvent). Preliminary
experiments using 1 mL of solvent demonstrated that here
adequate stirring inside the reaction mixture was possible in
most cases as indicated by the comparatively good agreement
between the measured external IR and internal fiber-optic
temperature probes (Figure S16a, Supporting Information). To
be confident about efficient agitation inside the closed micro-
wave reactor, the total reaction volume was reduced to 0.5 mL
for all subsequent experiments. Because of the low filling
volume no internal fiber-optic probe temperature measurement
in combination with magnetic stirring was possible, and the
determination of reaction temperature here relied exclusively
on IR sensor measurements. Using these experimental conditions
we could not detect any significant differences between
microwave and conventionally heated experiments when running
the alkylation reaction at 140 °C for 20 min. In contrast to the
employed (Scheme 1).3
3,34
Using polar solvents like pentanol
or DMF, no differences in the regioselectivity between con-
33
ventional and microwave heating at 140 °C were experienced.
Employing solvent-free conditions or nonpolar xylene as solvent,
however, dramatic differences between the oil bath and the
microwave heated alkylations were noted. For example, the
alkylation at 140 °C for 20 min in xylene as solvent produced
a mixture of the N1- (32%) (11), N4- (28%) (10), and N1,4-
bisalkylated triazoles (40%) (12) applying conventional heating
Scheme 1).33 In contrast, employing single-mode microwave
(
irradiation at the same measured temperature of 140 °C yielded
exclusively the N1 isomer.33 Nearly the same situation was
found using solvent-free conditions.33
Using our internal fiber-optic temperature monitoring tech-
nology, we initially set-out to reproduce the triazole alkylation
involving microwave transparent p-xylene as solvent. Perform-
ing oil bath experiments that allowed us to visually follow the
progress of the transformation in the reaction vessel shown in
Figure 1b immediately made the inability to efficiently agitate
the reaction mixture evident. While 2,2′,4′-trichloroacetophenone
3
3
previously published results (see above) in both cases nearly
identical mixtures of N1-, N4-, and N1,4-bisalkylated 1,2,4-
triazole products 10-12 (see Scheme 1) were obtained for both
heating modes (Table 4, entry 1). The composition of the crude
1
reaction mixture was readily determined by H NMR analysis
(Figure S17, Supporting Information). Column chromatography
(
1
mp 52-55 °C) is soluble in p-xylene at room temperature,
,2,4-triazole (mp 120-121 °C) remains insoluble in the solvent
using CHCl3/MeOH as eluent in combination with extraction
techniques allowed the isolation and full characterization of the
pure reaction products.
system even in the molten state. This leads to a biphasic mixture
that cannot be efficiently stirred using magnetic stirring in
combination with the standard 10 mL reaction vessel/stir bar
and filling volumes of >1 mL (Figure S15, Supporting
Information). Inefficient stirring under microwave conditions
may result in temperature gradients inside the reaction vessel
It has to be noted that the product distribution from our
experiments performed at 140 °C (20 min) is in comparatively
good agreement with the previously published oil bath data at
3
3
the same temperature (see above). In order to rationalize the
dramatic differences between the published and our microwave
runs several additional experiments were performed. As it is
evident from the data presented in Table 4, extending the
(
see above) and therefore to inaccurate reaction temperature
measurements. In the present case, the situation is further
aggravated since the biphasic 1,2,4-triazole/p-xylene mixture
(
33) Loupy, A.; Perreux, L.; Liagre, M.; Burle, K.; Moneuse, M. Pure
Appl. Chem. 2001, 73, 161.
34) For related examples, see: (a) P e´ rez, E. R.; Loupy, A.; Liagre, M.;
(35) By microwave irradiating a biphasic system consisting of immiscible
solvents/reagents with vastly different loss tangents, differential heating of
the two phases will occur. Depending on where and how the “reaction
temperature” is measured, different values will be obtained. A case in point
are biphasic mixtures of strongly microwave abosorbing ionic liquids and
nearly microwave transparent organic solvents. See ref 16 for more details.
(
de Guzzi Plepis, A. M.; Cordeiro, P. J. Tetrahedron 2003, 59, 865. (b)
Abenhaim, D.; Diez-Barra, E.; de la Hoz, A.; Loupy, A.; S a´ nchez-Migall o´ n,
A. Heterocycles 1994, 38, 793.
J. Org. Chem, Vol. 73, No. 1, 2008 43

Herrero et al.
SCHEME 2. Amidation of Methacrylic Acid with
reaction time for the 140 °C run to 60 min led to no significant
changes in the product composition (Table 4, entry 2). In
contrast, performing the alkylation process at higher reaction
temperatures clearly favored the formation of the N1-alkylated
triazole 11, both in microwave and conventionally heated
experiments (compare entries 1, 3, and 4). Choosing a reaction
temperature of 200 °C (entries 4-6), very high selectivities for
the N1-alkylated triazole 11 were obtained. After 60 min at
(R)-1-Phenylethylamine
2
00 °C only the N1-alkylated triazole 11 was observed in the
crude reaction mixture (entry 6) with an overall conversion of
ca. 90%.
In this context, it is important to note that the crude mixture
of triazoles 10-12 isolated from an experiment performed at
1
40 °C (Table 4, entries 1 and 2) can be cleanly converted into
absorbing N1,4-bisalkylated triazolium chloride intermediates
of type 12. These salts may behave similar to imidazolium-
type ionic liquids and will very rapidly be heated when exposed
the N1 isomer 11 by heating to 200 °C for 20 min (p-xylene,
microwaves or oil bath). These results are in good agreement
with previously published work on the temperature dependency
of 1,2,4-triazole alkylations.3 It has been shown earlier for a
closely related system (Scheme 1, Ar ) 4-chlorophenyl), that
while at lower temperatures a mixture of all three triazole
alkylation products is obtained, a higher reaction temperature
leads to the almost exclusive formation of N1-alkylated triazole
40
to microwave irradiation in the liquid or dissolved state.
6,37
Applying microwave irradiation during the alkylation of 1,2,4-
triazole raised the internal reaction temperature to 250 °C within
a few seconds (Figure S20, Supporting Information). Under
these conditions, mostly the N1-alkylated triazole 11 was
formed, but reactions were difficult to control and to reproduce.
Heating a mixture of 1,2,4-triazole and 2,2′,4′-trichloroacetophe-
none in an oil bath to a similar temperature range (for example,
36
11. The thermodynamically more stable N1 isomer is probably
formed by bimolecular rearrangement from the N4 isomer 10
involving the quaternary triazolium salt 12 as intermediate.
36-38
2
20 °C for 25 min) also provided product mixtures that mainly
Based on the outcome of our experiments, we therefore have
no evidence that the alkylation of 1,2,4-triazole shown in
Scheme 1 involves a nonthermal microwave effect as previously
stated. In our hands, control experiments between microwave
and oil bath heating at the same reaction temperature led to
comparable results. We believe that the dramatic changes in
regioselectivity previously observed in this reaction were the
result of inaccurate temperature measurements in the microwave
heated experiments, mainly as a consequence of inadequate
agitation/differential heating of the biphasic 1,2,4-triazole/xylene
reaction mixture.
consisted of the N1-alkylated triazole 11 (>85%).
Direct Amidation of Methacrylic Acid with (R)-1-Phenyl-
ethylamine. As a final example for our investigations on non-
thermal microwave effects, we have chosen the direct amidation
of a carboxylic acid with an amine. Transformations of this type
involve two neutral reagents and a neutral product but proceed
via highly polar intermediates (ammonium carboxylate salts)
and product-like transition states. According to the theory of
nonthermal microwave effects outlined above, these amidation
processes would be very well suited to exhibit nonthermal
microwave effects as a result of a direct stabilizing interaction
3
3
33
In addition to the results using p-xylene as solvent, we have
also re-evaluated the solvent-free alkylation of 1,2,4-triazole (8)
with 2,2′,4′-trichloroacetophenone (9). Heating a mixture of the
two components without solvent in an oil bath at 140 °C bath
temperature for 20 min (Figure S18, Supporting Information)
led to a product composition N1/N4/N1,4 ) 30:23:47, in the
same range as the previously published values (N1/N4/N1,4 )
12
of the electromagnetic field with the polar transition states.
Indeed, there have been several reports in the literature which
demonstrated that by using microwave irradiation under solvent-
free conditions, direct amide bond formation between a car-
boxylic acid and an amine can be achieved very efficiently
without the use of coupling reagents or other chemical activation
41-43
methods.
3
6:27:27).33 As in the alkylation of triphenylphosphine described
The particular example selected for this work involves the
above, the triazole alkylation shown in Scheme 1 also proved
to be exothermic and internal temperature monitoring revealed
reaction temperatures as high as 200 °C (Figure S19, Supporting
Information) during the initial phases of the 140 °C oil bath
experiment. Attempts to reproduce the solvent free triazole
alkylation under controlled microwave irradiation conditions
proved to be difficult.39 In all cases, significant thermal runaways
were experienced after only a few seconds of microwave
irradiation (Figure S20, Supporting Information). We believe
that those are related to the formation of strongly microwave
amidation of methacrylic acid (13) with (R)-1-phenylethylamine
(14) (Scheme 2). This example was chosen since a recent study
has shown that different product compositions are obtained when
comparing microwave heating and conventional heating in a
42
similar temperature region. Thus, microwave heating of an
equimolar mixture of the two starting materials in a dedicated
single mode microwave reactor at a monitored temperature of
1
80 °C (calibrated IR sensor) produced a 90% isolated yield of
(40) For recent reviews, see: (a) Leadbeater, N. E.; Torenius, H. M.;
Tye, H. Comb. Chem. High Throughput Screen. 2004, 7, 511. (b)
Habermann, J.; Ponzi, S.; Ley, S. V. Mini-ReV. Org. Chem. 2005, 2, 125.
(41) (a) Gelens, E.; Smeets, L.; Sliedregt, L. A. J. M.; Van Steen, B. J.;
Kruse, C. G.; Leurs, R.; Orru, R. V. A. Tetrahedron Lett. 2005, 46, 3751.
(b) Perreux, L.; Loupy, A.; Volatron, F. Tetrahedron 2002, 58, 2155. (c)
Massicot, F.; Plantier-Royon, R.; Portella, C.; Saleur, D.; Sudha, A. V.
Synthesis 2001, 2441. (d) Vasquez-Tato, M. P. Synlett 1993, 506.
(42) Ianelli, M.; Alupei, V.; Ritter, H. Tetrahedron 2005, 61, 1509.
(43) (a) Goretzki, C.; Krlej, A.; Steffens, C.; Ritter, H. Macromol. Rapid
Commun. 2004, 25, 513. (b) Ianelli, M.; Ritter, H. Macromol. Chem. Phys.
2005, 206, 349.
(
36) Smith, K.; Small, A.; Hutchings, M. G. Chem. Lett. 1990, 347.
(37) For a mechanistic discussion, see: Gautum, O. R.; Carlsen, P. H.
J. Eur. J. Org. Chem. 2000, 3745-3748.
38) Ab initio calculations at the B3LYP/6-31G* level confirmed that
the N1 isomer (11, Ar ) phenyl) is more stable than the N4 isomer (10, Ar
phenyl) by ca. 8 kcal/mol.
39) Difficulties reproducing this reaction in a solvent-free regime
(
)
(
under microwave conditions have been recently reported. For more
information, see: Lebouvier, N.; Giraud, F.; Corbin, T.; Na, Y.-M.; Le
Baut, G.; Marchand, P.; Le Borgne, M. Tetrahedron Lett. 2006, 47, 6479.
44 J. Org. Chem., Vol. 73, No. 1, 2008

the anticipated methacrylamide 17 after 15 min.42 This experi-
TABLE 5. Comparison of Microwave Heating and Conventional
Heating for the Reaction of Methacrylic Acid (13) with
(
ment was conducted using the concept of simultaneous cool-
R)-1-Phenylethylamine (14) (Scheme 2)a
2
4
ing. The actual internal reaction temperature was suggested
to be ca. 200 °C using a calibration method. In contrast, the
conversion to amide 17 in the experiment applying conventional
heating at 200 °C was only 12% after the same period of time
Tb
(°C)
timec
(min)
yieldd
(%)
entry
heating method
e,f
1
2
3
4
5
257
262
263
205
198
MW (sim cooling, power control)
MW (temp control)
15
15
15
15
15
71
78
75
25
19
f
(
15 min) and would not significantly increase by extending the
f
metal bath (258-268 °C bath temp)
42
reaction time to 30 min.
f
g
MW (temp control)
f
g
A detailed kinetic analysis conducted by the authors has
demonstrated that the first (kinetic) reaction product which is
produced from the initially formed salt 15 in this transformation
is propionic acid 18 which is the result of a (reversible) Michael
addition of the amine to methacrylic acid. After ca. 4 min using
microwave heating (ca 200 °C) the concentration of the Michael
adduct peaked at 50% and subsequently decreased dramatically
at the expense of the thermodynamically more stable methacry-
lamide 17 which was formed in 93% (GC-MS conversion) after
oil bath (200 °C bath temp)
a
Methacrylic acid (7.0 mmol) and (R)-1-phenylethylamine (7.0 mmol)
were reacted without solvent in a 10 mL microwave process vial (Figure
1b). For further details, see the Experimental Section. b Average reaction
c
temperature measured by one internal fiber-optic probe. Total reaction/
d
irradiation time. Isolated product yield of methacrylamide 17 after column
chromatography. e Applying 6 bar of compressed nitrogen (25 °C). Heating
profiles and GC-MS chromatograms are reproduced in the Supporting
Information (Figures S21-S25). g The major product at ca. 200 °C based
on GC-MS analysis is amide 16 (see Figure S25, Supporting Information).
f
1
5 min.42 Interestingly, when the same analysis was performed
for the conventionally heated experiment applying a similar
temperature profile44 the concentration of the initially formed
kinetic product (Michael adduct 18) would again decrease after
higher than the ca. 200 °C shown by the standard IR sensor. In
fact, the fiber-optic probe revealed reaction temperatures that
were as much as 60 °C above the recorded external IR
temperatures (average temperature 257 °C, see Figure S21,
Supporting Information). Additional experimentation has dem-
onstrated that the optimum temperature for the direct amidation
process 13 + 14 f 17 appears to be around 260 °C. Higher
temperatures (>270 °C) or longer reaction times led to
significant product decomposition and byproduct formation,
whereas lower temperatures favored the formation of amide 16
4
3
min, but now the major product of the reaction (>80% after
0 min) was amide 16 which is either formed by Michael
addition of amine 14 to methacrylamide 17 or by amidation of
propionic acid 18 with the amine (Scheme 2). The desired
methacrylamide 17 was formed in <20% yield using conven-
tional heating conditions, while amide byproduct 16 was
detected in only very small amounts (3.4%) in the microwave
4
2
experiment.
(see below). In addition to the “cooled” microwave experiment
The authors have suggested that these results cannot be
attributed exclusively to the very efficient heating experienced
using microwave irradiation, and have rationalized the high
selectivity for the formation of methacrylamide 17 from the acid
and amine components by arguments relating to the relative
described above (Table 5, entry 1) we have therefore carried
out experiments at the “correct” optimum reaction temperature
of ca. 260 °C using (i) microwave heating without simultaneous
cooling and (ii) applying conventional heating in a metal bath.
As seen from the data presented in Table 5 (entries 1-3), in all
three cases the isolated product yields for methacrylamide 17
were rather similar. Moreover, the GC traces for the crude
reaction mixtures were more or less identical showing the same
impurity profiles (Figure S22, Supporting Information). The
main impurity in all cases was amide 16 (Scheme 2). Based on
these results, we therefore have no evidence that this direct
amidation reaction involves a nonthermal microwave effect as
hardness of the transition states involved.4
2,45
It was proposed
that, when competitive reactions are involved, the mechanism
occurring via the hardest, most polar transition state (here 13
+
14 f 17) should be the favored one under microwave
irradiation conditions.
12
A reinvestigation of this solvent-free process using internal
fiber-optic temperature monitoring technology again demon-
strated the importance of an accurate reaction temperature
determination in microwave heated reactions. In order to ensure
appropriate agitation, reactions were typically run on a 7.0 mmol
scale under sealed vessel conditions using one single fiber-optic
temperature sensor. Under these conditions, the homogeneous
reaction mixture (ca. 1.5 mL) could be efficiently stirred after
the initially formed salt 15 had melted around 130 °C. In our
experiments we have tried to as close as possible mimic the
conditions previously reported in the literature.42 After extensive
optimization, we were able to achieve a 71% isolated yield of
the desired methacrylamide 17 after 15 min of microwave
irradiation at a ca. 200 °C IR temperature using simultaneous
cooling via a compressed nitrogen flow (Table 5, entry 1).
It became immediately apparent that by using such an
experimental protocol the internal temperature was considerably
4
2
previously assumed. The amount of microwave power used
in the experiment - modulated by simultaneous cooling - proved
to be irrelevant (94 W versus 173 W with simultaneous cooling).
Again, inadequate temperature measurements in the microwave
heated experiment, mainly as a consequence of using the
simultaneous cooling method in conjunction with IR temperature
1
8
monitoring, probably led to a misinterpretation of results in
4
2
the past.
In addition, we have also studied this reaction at a temperature
of 200 °C (15 min) in order to compare our results to the
4
2
previously published data. Indeed, at this substantially lower
reaction temperature the main reaction product is amide 16, with
only small amounts of the desired methacrylamide 17 being
formed (Table 5). Not surprisingly, there is no substantial
difference in yield regardless if the reaction has been processed
by microwave heating or conventional heating (Figure S25,
Supporting Information). Apparently, the amidation of meth-
acrylic acid (13) with (R)-1-phenyl-ethylamine (14) (Scheme
(44) In order to mimic the rapid heating profiles observed under
microwave conditions, the conventionally heated experiments were con-
ducted using a differential calorimetric scanning (DSC) apparatus.
(45) For further discussion on this reaction, see also the follwowing
2
) is rather sensitive to the reaction temperature. Here, small
references: (a) Koopmans, C.; Ianelli, M.; Kerep, P.; Klink, M.; Schmitz,
S.; Sinwell, S.; Ritter, H. Tetrahedron 2006, 62, 4709. (b) Reference 12c,
pp 162-164. (c) Reference 12d, pp 254-256.
changes in the reaction temperature (200 °C versus 250 °C)
will lead to a substantially different product distribution.
J. Org. Chem, Vol. 73, No. 1, 2008 45

Herrero et al.
Accurate online temperature monitoring during microwave
experiments is therefore essential.
to be critically reexamined and that a considerable amount of
research work will be required before a definitive answer about
the existence or nonexistence of these effects can be given.46
Concluding Remarks
Experimental Section
In summary, we have performed a critical reinvestigation on
the existence of nonthermal microwave effects for four carefully
selected organic reactions. In all four cases, previous studies
have found significant differences in conversion and/or product
distribution comparing conventionally heated and microwave-
heated experiments performed at the same measured reaction
temperature.
Of critical importance for our reevaluation was the introduc-
tion of a multiple fiber-optic probe system as accurate temper-
ature measurement device in both the microwave and the
conventionally heated reactors. Using this technology, we have
demonstrated that efficient stirring/agitation of microwave-
heated reactions is essential. In contrast to an oil bath experi-
ment, even completely homogeneous solutions need to be stirred
when using single-mode microwave reactors. If efficient stirring/
agitation cannot be ensured, temperature gradients may develop
as a consequence of inherent field inhomogeneities inside a
single-mode microwave cavity. The formation of temperature
gradients is therefore a particular problem in case of, for
example, solvent-free or dry media reactions and for very
viscous or biphasic reaction systems where standard magnetic
stirring is not effective.
Applying the fiber-optic temperature monitoring device, a
critical reevaluation of the four model reactions discussed herein
has provided no evidence for the existence of non-thermal
microwave effects (see the Results and Discussion). Working
at a reaction scale that allowed for an efficient agitation of the
reaction mixture via magnetic stirring, no significant differences
in terms of conversion and selectivity between experiments
performed under microwave conditions and runs conducted in
an oil or metal bath at the same internally measured reaction
temperature were experienced. The observed effects were purely
thermal and not related to the microwave field. Modulating the
amount of microwave power delivered to the reaction mixture
Microwave Irradiation Experiments. All microwave irradiation
experiments described herein were performed using a single-mode
Discover Labmate System from CEM Corp. using standard Pyrex
or quartz vessels (capacity 10 mL). Experiments were performed
in temperature-control mode where the temperature was controlled
using the built-in calibrated IR sensor. In most experiments, the
internal reaction temperature was additionally monitored by a
multiple fiber-optic probe sensor. The assembly shown in Figure
1b consists of a standard 10 mL CEM process vial (quartz or Pyrex)
which is sealed with a special PTFE cap fitted with three quartz
tubes (inner diameter ) 1 mm, outside diameter ) 2 mm) for
inserting three individual fiber-optic probes (GaAs principle,
Opsens, accuracy (1.5 °C). The quartz tubes are flexible in height
and can be fixed by tightening the three screws on top of the cap.
When the main screw (made of PEEK) is tightened, the assembly
is sealed and can be operated at temperatures up to 250 °C and a
pressure of 20 bar (pressure is not monitored). The whole setup
can be introduced in the open-vessel applicator of the CEM
Discover using a specially designed PEEK support or immersed
into a preheated oil/metal bath.
Diels-Alder Cycloaddition of 5-Methoxycarbonyl-2-pyrone
and Phenylacetylene (Table 1). For the kinetic studies, a 10 mL
Pyrex reaction vessel equipped with a magnetic stir bar (Figure
1b) was filled with 1.0 g (6.5 mmol) of 5-methoxycarbonyl-2-
pyrone (1) and 1.0 mL (464 mg, 9.1 mmol, 1.4 equiv) of freshly
distilled phenylacetylene (2). The vessel was sealed (the PTFE cap
was equipped with only one quartz immersion tube allowing sample
withdrawal with a syringe through the two open fiber optic inlets)
and either immersed in a preheated oil bath or irradiated in a
microwave reactor for 240 min (Table 1; Figures S8, S9, Supporting
Information). For preparative experiments, a sealed 10 mL Pyrex
reaction vessel was filled with 499 mg (3.24 mmol) of 5-meth-
oxycarbonyl-2-pyrone (1) and 0.500 mL (464 mg, 4.54 mmol, 1.4
equiv) of freshly distilled phenylacetylene (2) and heated for 3 h
at 200 °C either with microwave irradiation (IR temperature
measurement) or in a preaheated oil bath to reach >96% conversion
1
(
H NMR). The cycloadduct (4) could be isolated as a light brown
(for example by applying simultaneous cooling) proved to be
oil (71 and 75% yield, respectively) after column chromatography
ineffective. We believe that the previously observed differences
between experiments performed under conventional and mi-
crowave heating conditions for these four transformations are
in fact the result of inaccurate temperature measurements using
external IR sensors. As we have demonstrated, the use of IR
sensors is in many cases not appropriate for microwave heated
reactions and can easily lead to a misinterpretation of results
since the true reaction temperatures during microwave irradiation
are not known, in particular for those cases where stirring is
problematic and/or where strongly microwave-absorbing polar
intermediates or products are formed.
1
using CH
2
Cl
2 3
as an eluent: H NMR (CDCl ) δ 3.65 (s, 3H), 7.33-
7.45 (m, 7H), 7.55 (dt, J ) 7.5 Hz, J ) 1.4 Hz, 1H), 7.85 (dd, J
47
)
7.7 Hz, J ) 1.0 Hz, 1H).
Nucleophilic Substitution of Benzyl Chloride with Triph-
enylphosphine in p-Xylene (Tables 2 and 3). A 10 mL Pyrex
reaction vessel equipped with a stir bar (Figure 1b) was filled with
a solution of 524 mg (2.0 mmol) of triphenylphosphine (6) and
278 µL (253 mg, 2.0 mmol) of benzyl chloride (5) in 2 mL of
p-xylene. For solvent-free experiments, 2.62 g (10 mmol) of
triphenylphosphine (6) and 1.39 mL (1.27 g, 10 mmol) of benzyl
chloride (5) were mixed. The reaction mixture was either exposed
to microwave irradiation or immersed into a preheated oil bath at
different temperatures and times as shown in Tables 2 and 3 (for
representative heating profiles, see Figures S11, S13, S14, Sup-
porting Information). After the mixture was cooled to room
temperature, the precipitated salt was filtered, washed with cold
toluene, and kept in the drying oven at 50 °C overnight to produce
In this context, it has to be noted that it is probably no
coincidence that nonthermal microwave effects have in many
cases been claimed for processes involving solvent-free/dry
media reactions and/or for transformations involving polar
reaction intermediates or products (which will strongly absorb
12
microwave energy). Based on the results presented herein, we
suspect that in most of the published cases, the observed
differences between microwave and conventional heating can
in fact be rationalized by inaccurate temperature measurements
often using external IR temperature probes, rather than being
the consequence of a genuine nonthermal effect. We therefore
believe that the concept of nonthermal microwave effects has
pure phosphonium salt 7 (for yields, refer to Tables 2 and 3): mp
32-334 °C (lit.30 mp 333-337 °C); H NMR (CDCl
1
) δ 5.43 (d,
3
3
(46) For a recent study on microwave effects, see: Dressen, M. H. C.
L.; van de Kruijs, B. H. P.; Meuldijk, J.; Vekemans, J. A. J. M.; Hulshof,
L. A. Org. Process Res. DeV. 2007, 11, 865.
(47) Kondolff, I.; Doucet, H.; Santelli, M. Organometallics 2006, 25,
5219.
46 J. Org. Chem., Vol. 73, No. 1, 2008

J ) 14.5 Hz, 2H), 7.07-7.08 (m, 4H), 7.57-7.62 (m, 6H), 7.68-
bar impossible. Therefore, the starting materials were additionally
mixed with a spatula before sealing the reaction vessel as shown
in Figure 1b. The salt was subjected to microwave irradiation (140-
300 W maximum magentron ouput power, 200 °C set IR temper-
ature), and simultaneous cooling (6 bar of nitrogen) was started
immediately after the target IR temperature (200 °C) was reached
(Table 5, entry 1, Figure S21, Supporting Information). After 15
min of total irradiation time, the reaction mixture was allowed to
cool to room temperature. Subsequent silica gel column chroma-
tography using dichloromethane/methanol (20:1) as eluent resulted
7
.75 (m, 10H).30
General Procedure for the Alkylation of Triazole with
Trichloroacetophenone (Table 4). A 10 mL Pyrex vessel equipped
with a stir bar (Figure 1b) was charged with a solution of 223 mg
(1.0 mmol) of 2,2′,4′-trichloroacetophenone (9) in 0.5 mL of
p-xylene and 62 mg (0.9 mmol) of 1,2,4-triazole (8). The reaction
mixture was either exposed to microwave irradiation or immersed
into a preheated oil bath at different temperatures and times shown
in Table 4 (for representative heating profiles, see Figures S16,
S19, and S20, Supporting Information). After being cooled to room
temperature, the mixture was homogenized with methanol. Sub-
sequently, the organic solvents were removed under vacuum and
in 932 mg (71%) of the final methacrylamide product 17: mp 88-
90 °C (lit.42 mp 91-92 °C); H NMR (CDCl
1
) δ 1.54 (d, J ) 6.9
3
Hz, 3H), 1.98 (s, 3H), 5.19 (quint, J ) 7.2 Hz, 1H), 5.34 (s, 1H),
5.69 (s, 1H), 6.01 (br s, 1H), 7.26-7.39 (m, 5H); MS (positive
1
42
the conversions determined using H NMR spectroscopy (Figure
S17, Supporting Information). Isolation via tituration with dioxane
APCI) m/z 190 (M + 1, 100). Experiments at different temperatures
and using conventional heating (Table 5, entries 2-5) were
performed in a similar way. The salt 15 was either subjected to
microwave irradiation without simultaneous cooling or heated in a
and subsequent silica gel flash chromatography using CHCl
3
/MeOH
as eluent provided the pure alkylation products as colorless solids.
1
1
(
(
+
2
1
6
0: mp 175-177 °C; H NMR (DMSO-d ) δ 6.03 (s, 2H), 7.71
dd, J ) 8.4 Hz, J ) 1.9 Hz, 1H), 7.86 (d, J ) 1.9 Hz, 1H), 8.09
preheated bath for 15 min. Data for amide 16: mp 137-142 °C;
1
d, J ) 8.4 Hz, 1H), 9.41 (s, 2H); MS (positive APCI) m/z 256 (M
3
H NMR (CDCl ) δ 1.19 (d, J ) 7.1 Hz, 3H), 1.50 (d, J ) 6.9 Hz,
1
1, 100). 11: mp 116-118 °C; H NMR (DMSO-d
6
) δ 5.84 (s,
3H), 1.79 (d, J ) 6.8 Hz, 3H), 1.91 (br s, 2H), 2.82-2.87 (m, 2H),
3.40-3.49 (m, 1H), 4.21-4.25 (m, 1H), 5.01 (quint, J ) 7.3, 1H),
7.22-7.58 (m, 10H); MS (positive APCI) m/z 311 (M + 1, 100).
H), 7.64 (dd, J ) 8.4 Hz, J ) 2.0 Hz, 1H), 7.81 (d, J ) 1.9 Hz,
H), 7.95 (d, J ) 8.4 Hz, 1H), 8.02 (s, 1H), 8.53 (s, 1H);39 MS
1
(
positive APCI) m/z 256 (M + 1, 100). 12: mp 226-228 °C; H
NMR (DMSO-d
8
6
) δ 6.23 (s, 2H), 6.40 (s, 2H), 7.71-7.76 (dt, J )
Acknowledgment. This work was supported by a grant from
the Christian Doppler Society (CDG). M.A.H. thanks “Junta
de Comunidades de Castilla la Mancha” for a scholarship. We
also thank Mr. Hannes Zach (Anton Paar GmbH) for provisions
and help with fiber-optic probes and IR imaging as well as Mr.
Syaed Tahir Alin and Dr. Walter M. F. Fabian for performing
ab initio calculations.
.1 and 1.9 Hz, 2H), 7.89 (dd, J ) 5.0 and 1.9 Hz, 2H), 8.13 (dd,
J ) 8.4 and 1.5 Hz, 2H), 9.35 (s, 1H), 10.28 (s, 1H); MS (positive
APCI) m/z 443 (M + 1, 100).
General Procedure for the Amidation of Methacrylic Acid
with (R)-1-Phenylethylamine (Scheme 2). To a 10 mL Pyrex
reaction vessel equipped with a stir bar and charged with 900 µL
(846 mg, 7.0 mmol) of (R)-1-phenylethylamine (14) was added
dropwise 595 µL (604 mg, 7.0 mmol) of freshly distilled meth-
Supporting Information Available: Description of general
experimental procedures, images, and heating profiles for reactions.
This material is available free of charge via the Internet at
http://pubs.acs.org.
acrylic acid (13) under ice cooling. In an exothermic reaction, the
1
solid ammonium salt 15 (mp 130-134 °C; H NMR (CDCl
3
) δ
1
.51 (d, J ) 6.8 Hz, 3H), 1.79 (s, 1H), 4.21 (q, J ) 6.8 Hz, 1H),
.29 (s, 1H), 5.49 (br s, 3H), 5.77 (s, 1H), 7.25-7.38 (m, 5H)) is
5
formed spontaneously, which makes further mixing with the stir
JO7022697
J. Org. Chem, Vol. 73, No. 1, 2008 47