7504
F.-Y. Tsai et al. / Tetrahedron Letters 45 (2004) 7503–7506
Reaction of aryl iodides with methyl acrylate in the pres-
(EtO)3Si
(CH2)3
HN
Si(OEt)3
(CH2)3
NH
ence of Et3N yielded the corresponding cinnamate in
ꢁ80% yields (entries 1, 4). Replacing Et3N with Bu3N,
the products could be isolated almost quantitatively
(entries 2–3, 5–6), presumably due to the higher boiling
point of such base. However, the reaction of aryl bro-
mides with butyl acrylate or styrene did not proceed at
100ꢁC. It took place smoothly at an elevated tempera-
ture with excellent turnover numbers for activated aryl
bromide (entries 7–11). In the case of p-bromoacetophen-
one, the turnover frequency (TOF) can reach upto
6.6 · 104 (entry 7). As for bases, trialkylamine appeared
to be a better choice. The use of sodium acetate or potas-
sium carbonate provided only trace of product (entries
12–13) presumably due to the poor solubility of these
salts in the organic solvents. Reaction of bromobenzene
with butyl acrylate afforded the coupling product in good
yield, but poor result with styrene (entries 14, 15). Unfor-
tunately, this catalyst performance appeared to be poor
with the deactivated aryl bromides (entry 16).
Br
Br
(i)
N
N
N
N
1
2
(EtO)3Si
(CH2)3
HN
Si(OEt)3
(CH2)3
O
Si
O
O
O
O
O
NH
Si
(iii)
(CH2)3
NH
(ii)
(CH2)3
HN
N
N
Pd
Cl
Cl
N
N
Pd
Cl
Cl
3
4
Scheme 1. Preparation of Pd grafting nano-sized MCM-41: (i)
H2N(CH2)3Si(OEt)3, Et3N, THF; (ii) (CH3CN)2PdCl2; (iii) (a)
MCM-41, toluene, D; (b) Me3SiCl.
To demonstrate the good transport of molecules inside
the nanosized MCM-41, the identical palladium com-
plex was modified onto the wall of fibrous MCM-41 sili-
cas (pore size ꢁ2.7nm) for comparison.12 For the
fibrous silica consisted of long channels (in 103 nm
dimension), one would expect the difficulty in material
transporting inside the channels. In addition, the same
palladium complex catalyzed the coupling reaction
under homogeneous conditions was studied for compar-
ison. Figure 2 shows the difference of yield versus time in
the coupling reaction of phenyl iodide with acrylate
among these systems. It appears that nanosized catalyst
4 behaves in the same order of activity as the homogene-
ous one, but much better than the catalyst prepared via
the support of the fibrous MCM-41 silica. This result
clearly demonstrates very little diffusion interference
along the nanosized MCM-41 channels for transporting
molecules (i.e., the reactant molecules are easily access-
ing to the catalytic sites). On the other hand, the reac-
tion proceeded slowly in the fibrous MCM-41 system,
reflecting the difficult transport of substrates.
metal complex decreases the pore size and surface area
somewhat, but with enough space left for the reactant
molecules. One notices that there is a very large jump
of adsorption volume near saturation pressure. This re-
flects the large contribution to adsorption by the inter-
particle texture porosity, which shows these MCM-41
particles are indeed in nanosized dimension. The TEM
images of as calcined and grafted material are shown
in Figure 1 for comparison, indicating the preserva-
tion of nanoparticle morphology upon grafting of
complexes.
For the catalytic activity test, coupling reactions of var-
ious aryl iodides and bromides with acrylates and styr-
ene were investigated. In a typical experiment, freshly
distilled N-methylpyrrolidinone (NMP) solvent was
added to a degassed flask containing a mixture of aryl
halide, olefin, and triethylamine in a ratio of 1:1.5:1, fol-
lowed by the addition of the catalyst. The resulting mix-
ture was heated in an oil bath at the specified
temperature for a certain period of time. The catalyst
was recovered by centrifugation and washed sub-
sequently with THF and water twice for each washing.
The liquid portion was added to water, extracted with
ethyl acetate, and concentrated to yield the desired
For the recycling study, reaction was conducted by
using 100mg (4.76 · 10ꢀ2 mol% Pd) of catalyst in the
coupling of phenyl iodide with methyl acrylate ([sub-
strate]/[Pd] = 2100). After the first run of the reaction,
the catalyst was recovered by simple centrifugation
and then dried under vacuum overnight. Under the
same conditions, a complete conversion of the substrate
into the desired product by using the recovered catalyst
was observed within 5h. Furthermore, the catalyst
remained the same activity after four reuse runs, for
example, close to 100% conversion for each use, without
any saturation phenomenon during the reaction. As for
the metal leaching, we checked the solution of the reac-
tion mixture after the catalysis and found less than ppb
level of palladium in the liquid portion of the reaction
mixture. This observation also explains that the recycled
catalysts remain the high activity after several uses.11
1
product, which was either characterized by H NMR
spectroscopy or separated by column chromatography.
All of the results are summarized in Table 1.
In summary, we have prepared an extremely efficient
and recyclable palladium anchored nano-sized MCM-
Figure 1. TEM images of (a) as calcined; (b) Pd complex grafted
MCM-41 silicas.