Diels-Alder cycloadditions by microwave-assisted, continuous flow organic synthesis (MACOS): The role of metal films in the flow tube

Article abstract of DOI:10.1039/b715709f

Thin Pd films have been deposited on the inside of capillary-sized tubes through which compounds undergoing Diels-Alder reactions have been flowed while being heated with microwave irradiation; dramatic rate accelerations are observed in the presence of the film, which has been shown to play both a heating and catalytic role. The Royal Society of Chemistry.

Full text of DOI:10.1039/b715709f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Diels–Alder cycloadditions by microwave-assisted, continuous flow
organic synthesis (MACOS): the role of metal films in the flow tubew
Gjergji Shore and Michael G. Organ*
Received (in Cambridge, UK) 11th October 2007, Accepted 30th November 2007
First published as an Advance Article on the web 17th December 2007
DOI: 10.1039/b715709f
Thin Pd films have been deposited on the inside of capillary-sized
tubes through which compounds undergoing Diels–Alder reac-
tions have been flowed while being heated with microwave
irradiation; dramatic rate accelerations are observed in the
presence of the film, which has been shown to play both a heating
and catalytic role.
sensor that can provide information only on the portion of the
surface that fills the aperture of the sensor, which means that
only bulk measurements can be made. Thus, it could be that
the majority of the catalysis is happening in these isolated,
hotter regions in the film. This could account for why conver-
sion is far greater for microwaved couplings employing the Pd
film at 200 1C when compared with the identical reactions
performed using an oil bath set at the same temperature, i.e.,
without microwave irradiation no such hot spots are possible
in the Pd film.
Microwave-assisted organic synthesis (MAOS) has had a
significant impact on organic and medicinal chemistry by
shortening reaction times dramatically, even though a defini-
tive rationale for how this rate increase manifests itself has yet
to be elucidated.1 More recently, microwave technology has
also been applied to flowed2 reactions;3 to gain the full
advantage of working in flow, reactions should ideally proceed
very rapidly and complete during the time between the starting
materials entering and departing the reaction tube. Microwave
heating by itself has gone a long way toward driving a wide
variety of reactions in a flowed format to high levels of
completion during the brief time that the reactants spend in
the microwave chamber of the synthesizer.3
To further understand, and perhaps differentiate the roles of
heating from catalysis with metal films under microwave
irradiation conditions, we have begun to look at different
transformations that are not necessarily metal-catalyzed, as
are the cross-coupling experiments (vide supra). The Diels–
Alder reaction of diene 1 with dienophile 2 is reported to
proceed very slowly under batch heating conditions without
catalysis (80% yield after 24 h refluxing in toluene).5 We
examined this reaction, and related difficult cycloadditions,
in flow using microwave irradiation and the results are shown
in Table 1.
We have demonstrated recently that thin-metal films on the
inner wall of the flow tube exert a further rate acceleration on
flowed microwave reactions.4 In cross-coupling processes, the
film itself has been demonstrated to be capable of catalyzing
Suzuki–Miyaura and Heck reactions, i.e., Pd thin films pro-
moted these transformations without any additional catalyst
being added to the reactant stream(s) that enter the flow tube.
Metal films have been shown to be potent absorbers of
microwave irradiation leading to very high temperatures.4
Indeed, we have found that metal films that are too thick
(e.g., a Pd film of 410 mm) can absorb so much irradiation,
and convert it to heat via current, that the glass tube support-
ing the film can melt. Without microwave irradiation, the
above-described cross-coupling reactions did not proceed,
indicating that there is not only a catalytic effect, but also a
pivotal heating effect supplied by the film. It has been postu-
lated that irregularities (e.g., angles, steps) in metal films
accentuate the absorption of microwave irradiation; this may
lead to the creation of significant ‘hot-spots’ in the film and
therefore these areas are much hotter than the bulk of the
film.4 In microwave synthesizers, the reaction temperature at
the surface of the reaction vessel is measured by an infrared
Compounds 1 and 2 in DMSO were flowed together
through a clear glass capillary (Table 1, entry 1) while being
irradiated with sufficient power to match the temperature
achieved using the reported batch oil bath results
(B110 1C).5 Given the long reaction time reported in batch,
and the short residence time in flow (any one slug of reacting
solution spends B120 s in the irradiation zone of the micro-
wave), it is notable that even 12% conversion was attained; a
control experiment with no irradiation led to no detectable
conversion (Table 1, entry 9). To probe the role of temperature
on its own, the same clear tube was run through an oil bath set
at 205 1C (Table 1, entry 6) and a sizable increase in conver-
sion to 50% was observed. To see if the Pd film had a stand-
alone heating effect, we covered the outside of a thin-walled
capillary with a metal coating and performed the reaction with
microwave irradiation (Table 1, entry 3); the IR sensor in the
irradiation chamber recorded a temperature of 205 1C and the
same level of conversion (50%) was attained. With such thin
glass walls and small reaction volumes, heat transfer is gen-
erally accepted to be immediate and complete, thus the
temperature of the reaction in the capillary would also be
expected to be 205 1C, although it could not be measured
directly.6
Department of Chemistry, York University, 4700 Keele Street,
Toronto, Ontario, Canada M3J 1P3. E-mail: organ@yorku.ca
w Electronic supplementary information (ESI) available: Experimental
procedures and compound data. See DOI: 10.1039/b715709f
To probe catalytic roles of the Pd thin film, several experi-
ments were carried out. A Pd thin film was laid down on the
inner surface of the capillary and the MACOS reaction
ꢀc
This journal is The Royal Society of Chemistry 2008
838 | Chem. Commun., 2008, 838–840

View Article Online
Table 1 Diels–Alder cycloaddition reactions by MACOS employing Pd-coated and non-Pd-coated capillaries
Entry
Capillary (1200 mm ID)
Diene (1.0 eq.)
Dienophile (X eq.)
Pressure/atma
Temp./1C
Product (% conversion)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Clearc
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
2 (0.67)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
4 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
6 (1.5)
1
5
1
1
5
1
1
1
1
1
5
1
1
5
1
1
1
1
1
5
1
1
5
1
1
1
1
115
110
205
205
180
7 (12)
7 (39)
7 (50)
7 (93) [80]f
7 (99)
7 (50)
7 (75)
7 (16)
7 (0)
8 (10)
8 (70)
8 (58)
8 (86) [70]f
8 (96)
8 (46)
8 (64)
8 (5)
Clearc
Pd-coated outsided
Pd-coated insidee
Pd-coated insidee
Clearc
205 (oil bath)
205 (oil bath)
Pd-coated insidee
Pd-coated insidee
Clearc
23
23
115
110
205
205
Clearc
Clearc
Pd-coated outsided
Pd-coated insidee
Pd-coated insidee
Clearc
180
205 (oil bath)
205 (oil bath)
23
Pd-coated insidee
Pd-coated insidee
Clearc
23
115
110
205
8 (0)
Clearc
9 (10)
9 (50)
9 (47)
9 (90) [74]f
9 (87)
9 (54)
9 (72)
9 (1)
Clearc
Pd-coated outsided
Pd-coated insidee
Pd-coated insidee
Clearc
205
180
205 (oil bath)
205 (oil bath)
23
Pd-coated insidee
Pd-coated insidee
Clearc
23
9 (0)
a
For 1 atm reactions, the only pressure in the system originated from the pressure of the syringe pump that pushed the solution into and through
the capillary, otherwise it was an open system. For 5 atm reactions, the outlet end of the capillary was pushed through a septum into a sealed vial
b
that was charged with B75 psi of air pressure. Percent conversion was determined by taking samples of the effluent from the capillary and
c
analyzing it directly by 1H NMR spectroscopy. All capillaries used for this study were 12 cm long (1180 mm ID); approximately 2.5 cm of the tube
resides in the actual irradiation zone of the microwave. Pd colloids were deposited on the outer surface of the capillary; the inner surface of the
d
e
capillary had no metal coating on it and the reaction mixture was flowed through it. Pd colloids were deposited on the inner surface of the
f
capillary; the reaction mixture was flowed through the capillary where it was in direct contact with the metal film. % Yield following silica gel
chromatography.
repeated (Table 1, entry 4). Remarkably, conversion was now
essentially quantitative, which meant that a reaction that once
proceeded to only 80% conversion under 24 h of constant
reflux in toluene solvent under batch conditions could now
reach completion in less than 2 minutes with MACOS. In
order to dissect out the role of the microwave irradiation, the
reaction was performed at the same temperature (i.e., 205 1C),
but in an oil bath (Table 1, entry 7). Conversion dropped
sharply to 75%, suggesting that microwave irradiation is
necessary and simple bulk heating of the film, as one would
achieve in the oil bath, does not entirely suffice to promote the
highest possible conversion. Interestingly, when the same
transformation was simply flowed with no microwave irradia-
tion through the same Pd-coated capillary, 16% conversion
was achieved (Table 1, entry 8); with no film the conversion
was 0% (entry 9). It would appear as though the metal film is
playing a catalytic role in addition to its involvement in
heating.
The pressure within the flow system was also investigated.
With no imposed backpressure at the end of the flow system,
the only pressure generated during the MACOS process comes
from the heat of the reacting solution, the syringe pumps, and
the walls of the tubing, thus essentially it is functioning under
standard atmospheric pressure (i.e., 1 atm or 15 psi). Under
these conditions, the solvent cannot exceed its boiling point,
thus no overpressure is possible; when this temperature is
reached, the solvent changes state rapidly to a gas and the plug
of reactant in the irradiation zone is promptly expelled from
the capillary. With 5 atm (75 psi) of induced backpressure in
the MACOS set-up, no cavitation of the liquid occurs and the
reaction flows as a continuous stream. In all cases where
backpressure is imparted on the system (e.g., Table 1, entries
2 vs. 1 and 5 vs. 4), superior conversion was obtained. All
results obtained employing reactant pairs 3 with 4 and 5 with 6
were essentially identical to those obtained with 1 and 2 when
similar experiments to those described above were repeated.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 838–840 | 839

View Article Online
retained/coordinated material being returned to the infusing
stream. After flowing for long enough, the concentration of
maleic anhydride equalled the infusing concentration, i.e.,
0.2 mmol mLꢁ1. Thus, it would be reasonable to invoke a
catalytic role for the Pd(0) film that involves coordination of
the dienophile to the metal surface.
In summary, we have demonstrated that very long reaction
times for batch reactions using conventional heating can be
reduced to much shorter time frames using MACOS. Further,
reactions that transform so quickly are typically cleaner as the
completed product does not sit in a high temperature environ-
ment for prolonged periods while the rest of the starting
materials react. A Pd thin film on the flow tube walls drama-
tically shortens the reaction time; this film has been shown to
play a significant role both in effective heating and in catalyz-
ing these transformations.
Fig. 1 Changes in the concentration of 1-bromooctane and maleic
anhydride in a solution that was flowed through a capillary lined with
a Pd thin film.z
Based on the above experimentation, it is clear that the Pd
film has both a heating and a catalytic role in these Diels–
Alder cycloadditions; just how the film catalyzes the transfor-
mation is not clear. EDX analysis of the Pd film indicates that
it is 94% Pd, 5.5% C, and 0.3% O by weight, therefore it is
presumed that the vast majority of the film is Pd(0). To probe
this, we flowed hydrogen gas down the inside of a Pd-coated
capillary; significant sparking ensued, confirming that a sig-
nificant portion of the film’s surface is indeed metallic Pd. At
first glance, it would seem reasonable to try and invoke a
Lewis acid-type role for the film that would be analogous to
the well-established role of conventional electron-deficient
metals in homogeneous catalysis, such as complexes of boron
or aluminium. However, given the electron-rich nature of a
metallic film, such as Pd(0) in this case, it might seem counter-
intuitive to invoke such a role for the film. That said, reduced
metals are definitely known to coordinate suitable substrates,
such as is the case with olefins and Pd/C under catalytic
hydrogenation conditions. To probe such a coordinative role
of the Pd(0) film in these MACOS reactions we prepared a
solution comprised of an equimolar amount of maleic anhy-
dride and 1-bromooctane (0.2 M in each) and simply flowed
this solution through a metal-coated capillary at rt (with no
microwave irradiation) and collected the effluent one drop at a
time and analyzed each drop for its composition (Fig. 1).
Whereas the control 1-bromooctane, which presumably could
not coordinate to the film, came straight through the capillary,
the maleic anhydride displayed affinity behavior. In other
words, the film sequestered maleic anhydride out of the
solution that was being constantly infused at a fixed concen-
tration. This is seen when one looks at the amount of maleic
anhydride in drops 1 through 5. When the available sites on
the film become saturated, 2 came straight through and the
slight over-concentration seen in drops 5–10 is caused by the
Notes and references
z A stock solution containing equal molar portions of 1-bromooctane
and maleic anhydride (0.2 mmol mLꢁ1 each) in DMSO was flowed
through a Pd-coated capillary and each droplet of effluent separately
collected. The relative concentration of the two solutes was determined
by proton NMR spectroscopy.
1 (a) P. J. Tierney and P. Lidstrom, Microwave Assisted Organic
Synthesis, Blackwell Publishing, Oxford, 2005; (b) C. O. Kappe,
Angew. Chem., Int. Ed., 2004, 43, 6250.
2 For reviews on miniaturized flow reactors, see: (a) K. Jahnisch, V.
¨
406; (b) H. Pennemann, P. Watts, S. J. Haswell, V. Hessel and H.
¨
Hessel, H. Lowe and M. Baerns, Angew. Chem., Int. Ed., 2004, 43,
Lowe, Org. Process Res. Dev., 2004, 8, 422.
¨
3 (a) S. Bremner and M. G. Organ, J. Comb. Chem., 2007, 9, 14; (b)
E. Comer and M. G. Organ, J. Am. Chem. Soc., 2005, 127, 8160;
(c) M. G. Organ and E. Comer, Chem.–Eur. J., 2005, 11, 7223; (d)
M. C. Bagley, R. L. Jenkins, M. C. Lubinu, C. Mason and R.
Wood, J. Org. Chem., 2005, 70, 7003; (e) S. Saaby, I. R. Baxendale
and S. V. Ley, Org. Biomol. Chem., 2005, 3, 3365; (f) P. He, S. J.
Haswell and P. D. I. Fletcher, Appl. Catal., A, 2004, 274, 111; (g) T.
Cablewski, A. F. Faux and C. R. Strauss, J. Org. Chem., 1994, 59,
3408.
4 G. Shore, S. Morin and M. G. Organ, Angew. Chem., Int. Ed.,
2006, 45, 2761.
5 S. B. Kalindijan, I. M. Buck, J. Cushnir, D. Dunstone, M. Hudson,
C. Low, I. McDonald, M. Pether, K. Steel and M. Tozer, J. Med.
Chem., 1995, 38, 4294.
6 We recognize that the IR sensor measurement given by the Biotage
Initiator Synthesizert is not likely to deliver an accurate measure
of the temperature inside the capillary. However, we have not yet
been successful in obtaining more accurate measurements, for
example employing a miniature thermocouple, because such de-
vices also couple with microwave irradiation leading to current and
heating itself; this would lead to erroneous results. What we can
say is that the temperature in or at the surface of the metal film is
likely higher than 200 1C, as read by the IR sensor, but less than
600 1C, which is the melting temperature of borosilicate glass.
ꢀc
This journal is The Royal Society of Chemistry 2008
840 | Chem. Commun., 2008, 838–840