CHEMCATCHEM
FULL PAPERS
though with longer reaction times. In addition, 2-naphthyl bro-
mide produced the corresponding product in high yield upon
reacting with ethyl acrylate under the same reaction conditions
(Table 3, entry 12). Notably, the successful application of aryl
chlorides in the Heck reaction is limited.[9c,19] As an example, re-
cently Nacci et al. reported the efficiency of palladium nano-
particles in the Heck reaction of aryl chlorides in tetraalkylam-
monium ionic liquids.[19g] The present catalyst system also dem-
onstrated excellent activity in the coupling of aryl chlorides
bearing electron-withdrawing groups, such as 4-chlorobenzal-
dehyde (Table 3, entries 13–15), 3-chlorobenzaldehyde (Table 3,
entry 16) and 4-chloronitrobenzene (Table 3, entry 17), with
various acrylates and gave high yields of the corresponding
Heck adducts under similar reaction conditions. These results
showed high efficiency and reactivity of the designed catalyst
compared to other reported catalytic systems used in this
field.[14] Because the recoverability and recycling of the catalyst
are important issues from both practical and economical view-
points, we next investigated the reusability of the Pd@PMO-IL-I
catalyst in the reaction of bromobenzene with methyl acrylate
under the optimised reaction conditions. The result showed
that the catalyst can be recovered and reused in nine reaction
cycles, although longer reaction times were needed after the
fifth run to ensure complete conversion (Figure 4). All recycling
runs furnished the corresponding Heck product in more than
99% purity by using GC analysis.
tion spectroscopy (AAS) performed for the filtrate demonstrat-
ed that the amount of leached palladium is less than 1 ppm.
However, the negative hot filtration test and the near absence
of leaching (as demonstrated by AAS) can sometimes result in
erroneous conclusion regarding the heterogeneous nature of
the catalyst[23e,f] because the amount of leached palladium spe-
cies can be lower than the detection limit of AAS (parts per bil-
lion level); however, the leached palladium species still demon-
strate extremely high catalytic activity.[24] In many instances the
leached palladium species are responsible for the observed
catalysis. In addition, these solubilised palladium species rede-
posit on the support after catalysis. Therefore, it is often neces-
sary to provide enough compelling evidences to verify wheth-
er a solid catalyst operates through a heterogeneous path-
way.[24] To verify whether the present catalyst system operates
through a heterogeneous pathway or whether it merely gener-
ates more active soluble palladium species, a series of control
experiments have been established. First, in a separate cou-
pling reaction of bromobenzene with ethyl acrylate under our
optimal reaction conditions, a large excess of Hg0 (Hg/Pd=
400:1) was added under vigorous stirring. The kinetic profile of
this reaction was then compared with that of the reaction in
the absence of Hg0. Then, in another separate experiment, an
excess amount of poly(4-vinylpyridine) (N/Pd=400:1) was
used in the same Heck coupling reaction under otherwise simi-
lar reaction conditions. As can be clearly seen from the reac-
tion profile (Figure S2), in both the poisoning experiments
a gradual (but not remarkable) decrease in catalytic activity
was observed upon the addition of poisons. These results
clearly indicate that the Pd@PMO-IL-I catalyst operates through
at least a partial homogeneous pathway.
The mechanism of the palladium-catalysed Heck coupling
reaction has been studied by several researchers.[8c,20] The re-
sults showed that in some cases the catalysts operated in a het-
erogeneous mode[4,9a,21] but in others the support acted as a re-
servoir for the soluble palladium species.[8e,22] To demonstrate
the behaviour of our catalyst during the reaction process,
a hot filtration test was performed in the Heck coupling of bro-
mobenzene with ethyl acrylate.[22a,23] After 45 min, the reaction
was stopped and filtered while it was hot. Then, the solid-free
filtrate was reacted under normal reaction conditions (1408C,
K2CO3). After 6 h, an additional conversion of only 5% was ob-
served in the coupling reaction. Furthermore, atomic adsorp-
The nitrogen sorption experiment and TEM micrograph of
the recovered catalyst were investigated to shed further light
on the catalyst evolution during catalysis and recycling stages.
The nitrogen adsorption–desorption isotherm of the recovered
palladium catalyst (RPd@PMO-IL-I) showed a type IV isotherm
with a sharp H1 hysteresis loop, which is observed typically in
the mesoporous materials with a regular rodlike structure (Fig-
ure S3). The BET surface area, pore volume and average pore
size of the RPd@PMO-IL-I catalyst decreased to 345 m2 gÀ1,
0.63 cm3 gÀ1 and 9.2 nm, respectively, which confirms the gen-
eration of palladium nanoparticles inside the mesopores of the
PMO-IL-I material.[25] Furthermore, the TEM image of the recov-
ered catalyst after five reaction cycles showed the high stability
of the mesochannels under the applied conditions, which is in
good agreement with the data obtained from the nitrogen
sorption experiments (Figure S4). This micrograph also showed
that the palladium nanoparticles are well dispersed inside the
mesochannels without any detectable aggregation and large
particle formation. In another study, the recyclability of an
ionic liquid-free Pd@SBA-15[26] catalyst with the same loading
of palladium in the Heck coupling reaction of bromobenzene
with ethyl acrylate was investigated. By using the same reac-
tion time as used in the case of the Pd@PMO-IL-I catalyst
(Table 3, entry 3), the product yields decreased rapidly from 98
to 48 to 17% upon three successive reaction cycles. This result
confirmed the higher efficiency of PMO-IL-I compared to SBA-
Figure 4. Reusability of the Pd@PMO-IL-I catalyst in the Heck coupling of
bromobenzene with ethyl acrylate over 10 runs.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 8
&
6
&
ÞÞ
These are not the final page numbers!