Continuous-flow microliter microwave irradiation in the synthesis of isoxazole derivatives: An optimization procedure

Article abstract of DOI:10.1055/s-0031-1290944

An efficient method was developed for the synthesis of 3,4,5-trisubstituted and 3,5-disubstituted isoxazoles by using continuous-flow microwave-heated microreactors. A study on the separate effects of the temperature, continuous-flow regime, and microwave irradiation showed that the continuous-flow regime had important effects for less reactive diketones, where microwave heating enhanced the reaction, permitting the formation of 5-methyl-3-phenylisoxazole, which was not formed in the absence of microwaves. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0031-1290944

SPECIAL TOPIC
▌2527
sCpeocianl ttoipnicuous-Flow Microliter Microwave Irradiation in the Synthesis of
Isoxazole Derivatives: An Optimization Procedure
Continuous-Flow Microliter Microwave-Irradiation Synthesis of Isoxazoles
Antonio M. Rodriguez,a Alberto Juan,b M. Victoria Gómez,*b Andres Moreno,a Antonio de la Hoz*a
a
Departamento de Química Orgánica, Facultad de Químicas, Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain
b
Instituto Regional de Investigación Científica Aplicada (IRICA), Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain
Fax 34(9022)04130; E-mail: Antonio.Hoz@uclm.es
Received: 01.03.2012; Accepted after revision: 15.03.2012
few centimeters) of microwaves at the established fre-
quency of 2.45 GHz, would be impossible in large-scale
batch reactors.5
Abstract: An efficient method was developed for the synthesis of
3,4,5-trisubstituted and 3,5-disubstituted isoxazoles by using con-
tinuous-flow microwave-heated microreactors. A study on the sep-
arate effects of the temperature, continuous-flow regime, and
microwave irradiation showed that the continuous-flow regime had
important effects for less reactive diketones, where microwave
heating enhanced the reaction, permitting the formation of 5-meth-
yl-3-phenylisoxazole, which was not formed in the absence of mi-
crowaves.
The optimization of chemical processes to reduce costs
and to eliminate unnecessary labor in purification of prod-
ucts and removal of byproducts is a key aspect of research
that is especially important when dealing with microwave
chemistry, where the analysis and optimization of the re-
action process can take longer to perform than the reaction
itself. With a focus on the process of optimization of a mi-
crowave-assisted chemical reaction, we have recently re-
ported on the design and fabrication of a microliter-
microwave reactor coupled with a custom-made nanoli-
ter-scale NMR spectrometer.6 This permits optimization
to be performed on a very small reaction volume and in a
short time, thereby reducing the quantity of waste pro-
duced, improving the safety of the process, and minimiz-
ing the consumption of reagents, solvents, and energy.
Previous optimization methods have involved, for exam-
ple, on-chip methodology7 or on-line HPLC analyses,
where various algorithms can be used to automate the op-
timization process.8
Key words: heterocycles, cyclizations, condensation, microwaves,
continuous-flow reactors, microreactors
Microreaction technology, which has recently been used
in both academia and industry as a tool to develop a wide
variety of synthetic organic transformations, has several
advantages over commonly used batch methods.1 Micro-
reactors are miniaturized reaction systems, fabricated by
techniques from microtechnology and precision engineer-
ing, that are used to study chemical reactions on a mi-
croscale.2 Microreactors consists of three-dimensional
structures (microchannels) with diameters typically less
than 1000 μm. The physical properties of microreactors,
such as their short diffusion distances, small internal vol-
umes, and high surface-to-volume ratios, permit improve-
ments in the yields and selectivities of many reactions,
mainly as a result of improved mass and heat transfer,
more efficient mixing, and regular flow profiles. Over the
past 15 years, microreactors have entered the field of flow
chemistry, providing additional advantages such as im-
proved safety, better reproducibility, ease of automation,
and precise control of reaction conditions.3
Here in is described an efficient method for the synthesis
of 3,5-substituted and 3,4,5-substituted 1,2-isoxazole de-
rivatives, together with an exhaustive optimization of the
reaction conditions. The method involved the use of con-
tinuous-flow microwave irradiation of a microreactor.
Complete conversions were possible within a few minutes
without using excessively high temperatures. In addition,
a modification of the setup afforded a good rate of produc-
tion of compound of interests by scaling up the process.
Because the isoxazole motif is present in numerous natu-
ral products of medicinal value,9 there is interest in the de-
velopment of highly regioselective syntheses of isoxazole
derivatives from simple starting materials. Efficient syn-
theses of 3-substituted and 3,5-substituted isoxazoles
have recently been reported that use α,β-unsaturated alde-
hydes or ketones.10 Haswell and co-workers described the
use of electroosmotic flow to induce flow of reagent solu-
tions within microchannels for the preparation of pyrazole
and isoxazole derivatives.11
The dielectric heating induced by microwave irradiation
can result reduced reaction times, high product yields, and
improved selectivity for a wide variety of chemical trans-
formations.4 Microwave irradiation has been recently ap-
plied to microreactor technology as an alternative mode of
heating, thereby combining the unique heating mecha-
nism of microwave irradiation with the typical advantages
of continuous-flow microreactors.5 Furthermore, the tech-
nique of continuous-flow microwave-assisted organic
synthesis permits scaling up of microwave processes in a
manner that, because of the low depth of penetration (a
Preliminary data for this reaction show that three factors
affect the conversion rate of the product: the temperature,
the use of microwave heating, and the flow regime. Com-
plete conversion in the preparation of 3,5-dimethylisoxa-
zole (4) is achieved by using a combination of microwave
SYNTHESIS 2012, 44, 2527–2530
Advanced online publication: 20.04.2012
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0031-1290944; Art ID: SS-2012-C0216-ST
© Georg Thieme Verlag Stuttgart · New York

2528
A. M. Rodriguez et al.
SPECIAL TOPIC
heating and a continuous flow regime at 80 °C. However, tion vessel and a microwave reactor (Discover, CEM
we were interested in determining the influence of each Corp.) or a heated sand bath for microwave or convention-
parameter separately to check whether the effect of micro- al heating, respectively.
wave irradiation is important in attaining full conversion,
In all cases, the percentage conversion was determined by
because continuous-flow reactors appear to be perfectly
direct analysis of samples of the effluent from the capil-
suited to mimicking the rapid heating achievable in micro-
lary (for continuous-flow conditions) or of samples from
wave chemistry.12 We therefore examined the individual
the reaction vessel (for batch-mode reactions) by 1H NMR
effects of microwave irradiation in comparison with con-
spectroscopy in an offline mode.
ventional heating, of operating at room temperature as op-
Various commercially available 1,3-diketones were treat-
posed to a higher temperature (80 °C), and of adopting a
ed with hydroxylamine hydrochloride in dimethyl sulfox-
continuous-flow regime as opposed to batch mode. We
ide to give the corresponding 1,2-isoxazoles (Scheme 1).
carried out a step-by-step analysis to test all possible com-
The optimization process was performed on three model
binations of these conditions by varying one factor at a
reactions in an attempt to identify any common behavior.
time to determine its influence while keeping other vari-
ables constant. This allowed us to evaluate the indepen-
R1
R2
O
O
dent effects of the temperature, the flow regime, and the
heating process on the conversion rate and permitted the
identification of the most effective combination of contin-
uous-flow and microwave irradiation conditions.5
O
NH2OH⋅HCl
R1
R3
N
DMSO
R2
R3
4: R1 = Me; R2 = H; R3 = Me
5: R1 = Me; R2 = Me; R3 = Me
6a: R1 = Ph; R2 = H; R3 = Me
6b: R1 = Me; R2 = H; R3 = Ph
1–3
Scheme 1 General reaction scheme for the preparation of three 1,2-
isoxazoles
Table 1 summarizes the conversion rates obtained for
isoxazole derivatives 4, 5, 6a, and 6b, for each combina-
tion of experimental conditions, as described above.
Three variables were tested: the temperature, the reaction
regime (batch or flow), and heating mode (microwave or
conventional). We therefore performed eight (23) experi-
ments to cover each of the possible combinations (Table
1, entries 1–8).
Figure 1 Illustration of the experimental set-up used for continuous-
flow microwave-assisted organic synthesis. The reagents were loaded
into the capillary system which was located inside the microwave
cavity with the aid of a graphite-impregnated Teflon (WeflonTM) bar.
Initially, the reactions conditions were optimized to pro-
vide the shortest possible reaction time (for lower costs)
by using the most appropriate temperature range and de-
sign of the microreactor/microwave system to facilitate
the translation of these reactions into high-throughput
chemistry for extension to the synthesis of a small library
of pyrazole and isoxazole derivatives.13 In addition, short
residence times were chosen to simplify studies on the in-
fluence of the other parameters. We therefore fabricated a
custom-built flow-cell that could be used to adapt a con-
ventional microwave oven (Discover, CEM Corp.) for
continuous-flow processing. A 70-cm-long fused-silica
capillary with external and internal diameters of 349 and
100 μm, respectively, was wound in and out of the micro-
wave cavity and wrapped around a bar of graphite-im-
pregnated Teflon (WeflonTM) containing the reaction
mixture, thereby defining a reaction volume of 3 μL that
could be exposed to microwave irradiation (Figure 1). For
conventionally heated continuous-flow reactions, a capil-
lary of the same length and diameter as that used for the
microwave reactions was introduced into a sand bath once
the desired temperature had become constant. Flow in the
system was controlled by a high-force pump and a gas-
chromatography syringe (SGE Analytical Science). An
appropriate flow-rate provided an optimum reaction time
of only 3–6 minutes for complete conversion. Batch-mode
reactions were carried out, either by using a 10-mL reac-
To determine which factors were the most relevant (tem-
perature, microwave heating, or flow regime), we had to
perform a detailed comparison between pairs of entries
that differed only in terms of one factor (for example, en-
try 1 versus entry 2), thereby eliminating the influence of
other factors that could have led to confusion in the anal-
ysis of data. For instance, the effect of temperature was
determined by comparing entries 1 and 2, in which both
reactions were performed in a batchwise manner in the ab-
sence of microwave irradiation. Thus, comparison of en-
tries 1 and 3 showed the effect of the flow regime, and
comparison of entries 1 and 5 showed the effect of micro-
wave heating. Table 2 lists the effects that were identified
by comparing the entries as discussed above. This analy-
sis showed that the reaction to form compound 4 shows
similar tendencies to the reaction to form compound 5,
whereas the reaction to form compound 6 showed a differ-
ent behavior. The temperature and flow regime had the
greatest effects on the formation of 4 and 5, whereas mi-
crowave heating and the flow regime were the most sig-
nificant factors in the formation of compound 6. In
addition, for all the isoxazole derivatives, continuous-
flow had three times the effect of microwave heating (Ta-
ble 2, entries 2 and 3), supporting recent studies in the lit-
erature.12 On the other hand, microwave heating produced
Synthesis 2012, 44, 2527–2530
© Georg Thieme Verlag Stuttgart · New York

SPECIAL TOPIC
Continuous-Flow Microliter Microwave-Irradiation Synthesis of Isoxazoles
2529
high conversions to isoxazoles 6a and 6b, the only case in
which the effect of microwave heating was significant
(entry 3). In fact, regioisomer 6b was obtained only in the
presence of microwave irradiation (Table 1, entries 1–5,
6, and 8) at particular temperatures; the nature of the flow
regime did not affect the formation of isomer 6b (Table 1,
entries 1 and 3 and entries 6 and 8).
Table 2 Conversion Data Ratios Extracted from Table 1, Illustrating
the Corresponding Effects
Entry Effects
Isoxazole Isoxazole Isoxazoles
4
5
6a/6b
1
2
3
temperature
6.5
4
5
3
1
6
33
11
flow regime
Table 1 Summary of the Conversions Obtained for the Three Isoxa-
zole Derivatives for Each Possible Combination of Experimental
Conditions
microwave heating
1
Resonance Instruments, Inc.) was used. This reactor has a
small microwave cavity separated from the magnetron by
a two-meter long coaxial cable, facilitating its use for con-
tinuous-flow studies.
Entry Temp Reaction Heating Conv. to Conv. to Conv. to
modea
(°C)
regime
4 (%)
5 (%)
6a/6b (%)
1
2
3
4
5
6
7
8
25
80
25
80
25
80
25
80
batch
batch
flow
flow
batch
batch
flow
flow
CH
6
39
23
43
6
7
35
19
43
6
1/0
6/0
Scale-up reactions were carried out for the synthesis of
3,4,5-trimethylisoxazole (5). The reaction was performed
under exactly the same conditions as the previous opti-
mized conditions and reached complete conversion, as ex-
pected. Thus, with a residence time of six minutes, a
temperature of 120 °C and a flow rate of 58 μL/min, we
obtained a total of 1.15 g/h of isoxazole 5.
CH
CH
33/0
23/0
11/0
57/13
37/0
76/10
CH
MW
MW
MW
MW
86
67
96
75
52
67
Pentane-2,4-dione (1), 3-methylpentane-2,4-dione (2), 1-phenylbu-
tane-1,3-dione (3), and NH2OH·HCl were commercially available
and were used without further purification. 1H and 13C NMR spectra
were recorded at 298 K in DMSO-d6 on a Varian Inova 500-MHz
spectrometer at 499.7 or 125.67 MHz, respectively.
a CH: conventional heating; MW: microwave heating.
Continuous-Flow Microwave Reactions; General Procedure
Microwave irradiations were performed in a microwave reactor
(Discover, CEM Corp.) in which the usual reaction vessel was re-
placed with a customized flow cell. The temperature was monitored
during the reaction by using the standard IR pyrometer in the micro-
wave reactor. The appropriate solutions of the 1,3-diketone and
NH2OH·HCl in DMSO were loaded into the fused-silica capillary
that defined the reaction volume by using a gas-chromatography sy-
ringe (SGE Analytical Science). The 70 cm long fused-silica capil-
lary with an outside diameter of 349 μm and an internal diameter of
100 μm was wrapped around a WeflonTM bar to facilitate heating of
the small sample volume. The capillary was wound in and out of the
microwave cavity to define a reaction volume of 3 μL. A flow rate
of the high-force pump (PHD 2000–71-2000, Harvard Apparatus
Inc.) that controlled the low flow rate in the system was adjusted to
1 μL/min, corresponding to a residence time of 3 min. To reach full
conversions for compounds 5 and 6, the flow rate was reduced to
give a longer residence time (6 min). When the reaction temperature
was 25 °C under microwave irradiation, the reactor operated at a
power of 1–7 W. The percentage conversion rate was determined by
analyzing samples of the effluent from the capillary directly by 1H
NMR spectroscopy in the offline mode.
The effect of the flow regime was more marked in the for-
mation of compounds 6, especially 6a (Table 2, entry 2
and Table 1, entries 1 and 3). The same trend was ob-
served with the use of microwave heating. Presumably,
the less-reactive conjugated ketone 1-phenylbutane-1,3-
dione (3) requires an additional input of energy to gener-
ate the transitions state for the reaction. This energy is best
supplied by the volumetric heating induced by micro-
waves or by the efficient heat transfer induced in continu-
ous-flow processing.14 In fact, complete conversion rates
for compounds 5 and 6 were only obtained by increasing
the temperature to 120 °C and the residence time to six
minutes under continuous-flow microwave-heating con-
ditions, thereby demonstrating the lower reactivity of ke-
tones 2 and 3, which, theoretically, have a lower density
of positive charge on the carbonyl carbon because of the
presence of the extra methyl or phenyl groups in sub-
strates 2 and 3, respectively. We aim to perform theoreti-
cal calculations to rationalize the observed differences
and to produce a general rule for predicting the enhance-
ment effects of microwave irradiation on microreactors.12
Continuous-Flow Conventionally Heated Reactions; General
Procedure
The appropriate solutions of the 1,3-diketone and NH2OH·HCl in
DMSO were loaded into a fused-silica capillary that defined the re-
action volume by using a gas-chromatography syringe (SGE Ana-
lytical Science). The 38 cm long fused silica capillary with an
outside diameter of 349 μm and an internal diameter of 100 μm was
introduced into a sand bath to heat up the small sample volume. The
heated capillary length defined a reaction volume of 3 μL. The flow
rate of the high-force pump (PHD 2000 – 71-2000, Harvard Appa-
ratus Inc.) was adjusted to 1 μL/min, corresponding to a residence
time of 3 min. The percentage conversion rate was determined by
Having established the optimal conditions for flow chem-
istry, we scaled the reaction up to a microliter scale by us-
ing a custom-built glass flow reactor with an internal
volume of 350 μL; this was more than one hundred times
the volume of the previous continuous-flow reactor used
in the optimization study. We selected continuous-flow
microwave-assisted experimental conditions because they
gave the highest conversion rates in the optimization ex-
periments. A commercial microwave reactor (Model 52;
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 2527–2530

2530
A. M. Rodriguez et al.
SPECIAL TOPIC
analyzing samples of the effluent from the capillary directly by 1H
NMR spectroscopy in the offline mode.
Acknowledgment
We acknowledge financial support from the Spanish Ministry of
Science and Innovation (CTQ2011-22410) and JJCC Castilla-La
Mancha (Project PII2I09-0100-9633). A.M.R. thanks the Ministry
of Science and Innovation for his F.P.U. grant. M.V.G. thanks
Marie Curie Reintegration Grants as well as Albacete Science and
Technology Park for the INCRECYT research contract. Thanks are
also expressed to F. Jiménez for adjustments of the microwave in-
strument.
Microwave-Heated Batch Reactions; General Procedure
Microwave-assisted reactions were performed in a microwave reac-
tor (Discover, CEM Corp.). The temperature during the reaction
was monitored by using the standard IR pyrometer in the micro-
wave reactor. The appropriate solutions of the 1,3-diketone and
NH2OH·HCl in DMSO were mixed and stirred in a conventional
microwave reaction vessel. The mixture was irradiated for 3 min at
the appropriate temperature (25 or 80 °C). When the reaction tem-
perature was 25 °C under microwave irradiation, the reactor operat-
ed at a power of 1–7 W. The percentage conversion rate was
determined by analyzing samples of the effluent from the capillary
directly by 1H NMR spectroscopy in the offline mode.
References
(1) Wiles, C.; Watts, P. Chem. Commun. (Cambridge) 2011, 47,
6512.
Conventionally Heated Batch Reactions; General Procedure
The appropriate solutions of the 1,3-diketone and NH2OH·HCl in
DMSO were mixed and stirred in a small reaction vessel (10 mL)
and heated for 3 min to the appropriate temperature (25 or 80 °C) in
a sand bath. The percentage conversion rate was determined by an-
alyzing samples of the effluent from the capillary directly by H
NMR spectroscopy in the offline mode.
(2) Ehrfeld, W.; Hessel, V.; Löwe, H. Microreactors: New
Technology for Modern Chemistry;Wiley-VCH: Weinheim,
2000.
(3) Wiles, C.; Watts, P. Eur. J. Org. Chem. 2008, 1655.
(4) (a) Díaz-Ortiz, A.; de la Hoz, A.; Moreno, A. Curr. Org.
Chem. 2004, 8, 903. (b) Díaz-Ortiz, A.; de la Hoz, A.;
Moreno, A. Adv. Org. Synth. 2005, 1, 119.
1
(5) Shore, G.; Organ, M. G. Chem. Commun. (Cambridge)
2008, 838.
(6) Gomez, M. V.; Verputten, H.; Díaz-Ortiz, A.; Moreno, A.;
de la Hoz, A.; Velders, A. H. Chem. Commun. (Cambridge)
2010, 46, 4514.
(7) Belder, D.; Ludwig, M.; Wang, L.-W.; Reetz, M. T. Angew.
Chem. Int. Ed. 2006, 45, 2463; Angew. Chem. 2006, 118,
2523.
3,5-Dimethylisoxazole (4)15,16
A solution of pentane-2,4-dione (1; 3 mmol, 3 M) and NH2OH·HCl
(3.33 mmol, 3.3 M) in DMSO was exposed to the appropriate ex-
perimental conditions. The percentage conversions to the product
are listed in Table 1. The spectroscopic properties of the product
agreed with those reported in the literature.
3,4,5-Trimethylisoxazole (5)16
A mixture of 3-methylpentane-2,4-dione (2; 3 mmol, 3 M) and
NH2OH·HCl (3.33 mmol, 3.3 M) in DMSO was exposed to the cor-
responding experimental conditions. The percentage conversions to
the product are listed in Table 1. The spectroscopic properties of the
product agreed with those reported in the literature.
(8) Nelder, J. A.; Mead, R. Comput. J. 1965, 7, 308.
(9) Hewings, D. S.; Wang, M.; Philpott, M.; Fedorov, O.;
Uttarkour, S.; Filippakopoulos, P.; Picaud, S.; Vuppusetty,
C.; Marsden, B.; Knapp, S.; Conway, S. J.; Heightman, T. D.
J. Med. Chem. 2011, 54, 6761.
(10) Tang, S.; He, J.; Yongquan, S.; Liuer, H.; She, X. Org. Lett.
2009, 11, 3982.
(11) Wiles, C.; Watts, P.; Haswell, S. J. Org. Process Res. Dev.
2004, 8, 28.
(12) (a) Viviano, M.; Glasnov, T. N.; Reichart, B.; Tekautz, G.;
Kappe, C. O. Org. Process Res. Dev. 2011, 15, 858.
(b) Glasnov, T. N.; Kappe, C. O. J. Heterocycl. Chem. 2011,
48, 11. (c) Wiles, C.; Watts, P. Beilstein J. Org. Chem. 2011,
7, 1360.
3-Methyl-5-phenylisoxazole (6a) and 5-Methyl-3-phenylisoxa-
zole (6b)17
A mixture of 1-phenylbutane-1,3-dione (3; 3 mmol, 3M) and
NH2OH·HCl (3.33 mmol, 3.3 M) in DMSO was exposed to the cor-
responding experimental conditions. The percentage conversions to
the relevant products are listed in Table 1. The spectroscopic prop-
erties of the product agreed with those reported in the literature.
Scaled-Up Continuous-Flow Microwave-Heated Reactions;
General Procedure
(13) Rodriguez, A. M.; de la Hoz, A.; Velders, A. H.; Fratila, R.;
Gomez, M. V. manuscript in preparation.
The scaled-up reactions were performed in a commercial micro-
wave reactor (Model 521, Resonance Instruments Inc.) by using a
customized glass flow-cell with an internal volume of 350 μL. The
temperature during the reaction was monitored by using the stan-
dard IR pyrometer in the microwave reactor. The percentage con-
version rate was determined by analyzing samples of the effluent
from the capillary directly by 1H NMR spectroscopy in the offline
mode.
(14) (a) Rodríguez, A. M.; Cebrián, C.; Prieto, P.; García, J. I.; de
la Hoz, A.; Díaz-Ortiz, A. Chem.–Eur. J. 2012, in press;
DOI: 10.1002/chem.201103560. (b) Microreactors in
Organic Synthesis and Catalysis; Wirth, T., Ed.; Wiley-
VCH: Weinheim, 2008.
(15) Griesbeck, A. G.; Franke, M.; Neudörfl, J.; Kotaka, H.
Beilstein J. Org. Chem. 2011, 7, 127.
3,4,5-Trimethylisoxazole (5)16
(16) Kashima, C.; Yamamoto, Y.; Tsuda, Y.; Omote, Y. Bull.
Chem. Soc. Jpn. 1976, 49, 1047.
A mixture of 3-methylpentane-2,4-dione (2; 2.1 mmol, 3 M) and
NH2OH·HCl (3.3 M) in DMSO was heated to 120 °C for a residence
time of 6 min at a flow rate of 58 μL/min. The rate of production of
the product was 1.15 g/h.
(17) (a) Sechi, M.; Sannia, L.; Orecchioni, M.; Carta, F.;
Paglietti, G.; Neamati, N. J. Heterocycl. Chem. 2003, 40,
1097. (b) Zaitsev, A. B.; Schmidt, E. Y.; Mikhaleva, A. M.;
Afonin, A. V.; Ushakov, I. A. Chem. Heterocycl. Compd. (N.
Y., NY, U. S.) 2005, 41, 722.
Synthesis 2012, 44, 2527–2530
© Georg Thieme Verlag Stuttgart · New York