Recyclable hybrid silica-based catalysts derived from Pd-NHC complexes for suzuki, heck and sonogashira reactions

Article abstract of DOI:10.1002/ejoc.201200205

Two silylated Pd-NHC complexes were immobilized on hybrid silicas by sol-gel cocondensation with tetraethyl orthosilicate (TEOS) and performed well as recyclable catalysts towards the Heck, Suzuki, and Sonogashira coupling reactions. Remarkable conversion

Full text of DOI:10.1002/ejoc.201200205

FULL PAPER
DOI: 10.1002/ejoc.201200205
Recyclable Hybrid Silica-Based Catalysts Derived from Pd–NHC Complexes
for Suzuki, Heck and Sonogashira Reactions
Guadalupe Borja,[a] Amàlia Monge-Marcet,[a,b] Roser Pleixats,*[a] Teodor Parella,[c]
Xavier Cattoën,[b] and Michel Wong Chi Man[b]
Keywords: Supported catalysts / Heterogeneous catalysis / Green chemistry / Palladium / Carbenes / Sol-gel processes /
Cross-coupling / C–C coupling
Two silylated Pd–NHC complexes were immobilized on hy-
brid silicas by sol-gel cocondensation with tetraethyl orthosil-
icate (TEOS) and performed well as recyclable catalysts
towards the Heck, Suzuki, and Sonogashira coupling reac-
tions. Remarkable conversion and recyclability were
achieved in the Suzuki reaction with a challenging aryl
chloride. No side products and no undesired homocoupling
were observed in Suzuki or Heck reactions, which facilitates
the final purification step for the cross-coupling products.
High turnover numbers and turnover frequencies were found
for copper- and phosphane-free Sonogashira reaction be-
tween p-bromoacetophenone and phenylacetylene.
Introduction
Palladium complexes incorporating Arduengo-type N-
heterocyclic carbenes (NHC)[8] appear to be efficient sys-
tems with which to achieve these goals. In some cases they
are formed in situ from the corresponding imidazolium salt
and a common palladium source;[9] however, this approach
has the drawback of requiring an excess of expensive li-
gands, which then need be removed from the reaction mix-
ture. Alternatively, well-defined monoligated PdII–NHC
precatalysts[10] have been synthesized that showed higher
levels of activity. However, they sometimes require
multistep syntheses and/or rigorously anhydrous and inert
conditions even if the intermediate carbene ligand is not
isolated. Recently, these drawbacks have been overcome to
some extent with the development of processes for the facile
preparation of air and moisture stable Pd–NHC precata-
lysts, without the need to generate the free carbene, using
readily available starting materials on a large scale. Nolan’s
group has reported the synthesis of [Pd(acac)Cl(NHC)]
complexes from [Pd(acac)2] and the corresponding imid-
azolium chloride, which have proven to be highly efficient
in the Buchwald–Hartwig amination and in α-ketone aryl-
ation with aryl bromides and chlorides.[11] Organ and col-
laborators have synthesized [PdCl2(3-ClPyr)(NHC)] com-
plexes from PdCl2, 3-chloropyridine, and the corresponding
imidazolium salt by direct C–H insertion. Furthermore,
they have demonstrated that such complexes perform well
when aryl bromides and chlorides are employed in Suzuki–
Miyaura and Negishi cross-coupling reactions, and in Pd-
catalyzed amination. The 3-chloropyridine plays the role of
a throw-away ligand (PEPPSI-type complexes: pyridine-en-
hanced precatalyst preparation, stabilization and initia-
tion).[12] To the best of our knowledge, to date, no Heck or
Sonogashira reactions have been described that use these
monoligated Pd–NHC complexes.
Palladium-catalyzed carbon–carbon bond forming reac-
tions such as the Suzuki–Miyaura,[1] Heck,[2] and Sonoga-
shira[3] coupling reactions are important synthetic transfor-
mations that are widely employed for the preparation of a
great variety of complex organic molecules and materials
such as natural products, fine chemicals, drugs, agrochem-
icals, and polymers.[4] However, these Pd-catalyzed reac-
tions suffer from the high cost of this metal. Furthermore,
Pd traces are undesired in the final pharmaceutically active
compounds. Therefore, the efficient recovery and recycling
of the catalyst remains a scientific challenge of economic
and environmental relevance. For these reasons, the immo-
bilization of homogeneous palladium catalytic systems is
the subject of intense research.[2d,5] Other challenges facing
this field are the design of catalysts that are more robust
and efficient with higher turnover numbers (TON) and
turnover frequencies (TOF),[6] as well as expanding the se-
arch for catalytic systems that are capable of activating aryl
chlorides,[7] which are substrates that are significantly less
reactive but also less expensive than the corresponding aryl
bromides or iodides.
[a] Department of Chemistry, Universitat Autònoma de Barcelona,
08193 Cerdanyola del Vallès, Barcelona, Spain
Fax: +34-93-5812477
E-mail: roser.pleixats@uab.cat
[b] Institut Charles Gerhardt Montpellier (UMR 5253 CNRS-
UM2-ENSCM-UM1), Architectures Moléculaires et Matériaux
Nanostructurés, Ecole Nationale Supérieure de Chimie de
Montpellier,
8 rue de l’école normale, 34296 Montpellier cédex 5, France
[c] Servei de Ressonància Magnètica Nuclear, Universitat
Autònoma de Barcelona,
08193 Cerdanyola del Vallès, Barcelona, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200205.
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3625

R. Pleixats et al.
FULL PAPER
Some of us have previously described several silica-sup- organic hybrid silica materials derived from monosilylated
ported phosphane-free palladium catalytic systems either [Pd(acac)Cl(NHC)] and [PdCl2(3-ClPyr)(NHC)] complexes
by applying sol-gel processes or grafting methods. Whereas by applying a sol-gel process. In this case, the Pd–NHC
macrocyclic triolefinic Pd0 complexes covalently anchored complex is generated at the organosilane precursor step,
to the silica matrix were active and recyclable catalysts for which is then subjected to the sol-gel hydrolysis step to form
Suzuki coupling only with aryl iodides,[13] hybrid silica ma- the immobilized Pd–NHC catalysts. Their activity and reus-
terials covalently Si–C bonded to di(2-pyridyl)methylamine- ability were then inspected in the Suzuki, Heck, and Sono-
palladium dichloride complex were found to be efficient re- gashira reactions.
cyclable catalysts for the Heck and Sonogashira reactions
with aryl bromides and for Suzuki–Miyaura couplings with
aryl bromides and chlorides.[14] A significant number of
publications have appeared in recent years on palladium
catalysts based on imidazolium-derived organosilicas for
C–C bond forming reactions; however, a successful recycla-
ble catalytic system that is able to operate with less expens-
Results and Discussion
Monomer Synthesis and Preparation of Supported
Catalysts
Syntheses of the monosilylated imidazolium salt 2, the
ive aryl chlorides is still a challenge. Indeed, only few stud- monocarbenic Pd–NHC complexes 3 and 4, and the hybrid
ies exist on Suzuki coupling that describe efficient and reus- silica materials M1 and M2 derived thereof, are summa-
able supported catalysts for this kind of substrate.[15]
rized in Scheme 1. Chloride 2 was obtained by heating a
Quite recently, we examined the immobilization of Pd– mixture of 1-mesitylimidazole 1[17] and (3-chloropropyl)tri-
NHC catalysts by using a sol-gel process. In this case, the ethoxysilane under argon at 90 °C for five days under sol-
hydrolysis–condensation, in the presence of a surfactant, of vent-free conditions.[18] Palladium complex 3 was prepared
a bis-silylated precursor derived from a dihydroimidazolium by treatment of [Pd(acac)2] with a slight excess of the mono-
salt was performed in the first step. Catalytic systems com- silylated salt 2 in anhydrous dioxane at 100 °C under argon
posed of the resulting hybrid silica materials and palladium for 20 h.[11a] Pure 3 was obtained in 83% isolated yield. On
acetate were tested in C–C coupling reactions.[16] Although the other hand, monosilylated salt 2 was reacted under ar-
some activity was found for the Suzuki cross-coupling with gon with palladium(II) chloride and excess 3-chloropyr-
an activated aryl chloride and with moderate success in re- idine at 80 °C overnight in the presence of potassium car-
cycling, the conversions remained quite low even under bonate,[12a] to afford complex 4 in 72% isolated yield.
harsh conditions.
Complex 3 was characterized by NMR and HRMS
To date, such silica-immobilized Pd–NHC catalysts have analyses. A complete chemical shift assignment was made
mostly been obtained by post-complexation of the pre- on the basis of two-dimensional NMR experiments (1H-
formed hybrid imidazolium or dihydroimidazolium derived 1H COSY, 1H-1H NOESY, 1H-13C HSQC, 1H-13C HMBC).
silica.[15] We describe herein the preparation of organic–in- Three different methyl and two aromatic signals belonging
Scheme 1. Synthesis of hybrid silica materials M1 and M2.
3626
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3625–3635

Recyclable Catalysts for Suzuki, Heck and Sonogashira Reactions
to the mesityl ring were observed in the 1H NMR spectrum predominates. Moreover, the complex was in equilibrium
of complex 3. On the other hand, the methylene group at- with free 3-chloropyridine ligand (see the Supporting Infor-
tached to the second nitrogen atom of the NHC ligand ap- mation, Figures S3 and S4) and a third, unknown, palla-
peared as two diastereotopic multiplets centered at δ = 4.4 dium species with higher molecular size was also present.
1
and 5.0 ppm. The H-1H NOESY spectrum of 3 presented Although the nature of the minor palladium species re-
strong chemical exchange cross-peaks for all these signals sulting from decoordination of the pyridine ligand from
(the two ortho methyl groups, the two aromatic protons and complex 4 could not be determined (variable-temperature
the diastereotopic methylene group), indicating restricted 1H NMR studies did not provide useful information), a di-
rotation around the N–Mes and N–CH2 bonds (see the meric complex with two bridging chloro ligands could be
Supporting Information, Figures S1 and S2).
envisaged. This should be in accordance with DOSY experi-
The PEPPSI-type complex 4 also deserves some com- ments (see the Supporting Information, Figure S4) in which
ments. To ensure complete removal of the excess of 3-chlo- we observed two species of similar molecular weight (corre-
ropyridine, vacuum distillation of this solvent was per- sponding to the two isomeric forms for 4), a species of
formed followed by thorough washing of the residue with higher molecular weight (the dimer), and a species of lower
anhydrous pentane and careful drying of the insoluble re- molecular weight (the 3-chloropyridine).
maining solid under vacuum. Although the results obtained
Palladium-containing hybrid silica materials M1 and M2
from elemental analysis showed that 4 was pure, the corre- were obtained by standard cogelification of the correspond-
sponding 1H NMR spectrum was complex and showed ing precursors with tetraethyl orthosilicate (TEOS) (40:1
more peaks than expected. Three different points must be molar ratio of TEOS to 3 or 4) in dimethylformamide, using
1
commented. First, in the H NMR spectrum of 4 (see the a stoichiometric amount of water with respect to the ethoxy
Supporting Information, Figure S3) two sets of signals in groups, and ammonium fluoride as catalyst (1 mol-% with
equilibrium were observed for 3-chloropyridine subunits, respect to silicon) (Scheme 1). Standard nucleophilic fluo-
one of which corresponded to the free ligand and the sec- ride catalysis was considered appropriate for the sol-gel co-
ond due to the coordinated ligand. This was confirmed by gelification because it avoids the acidic or basic medium
a
1H-1H NOESY experiment (see the Supporting Infor- required for alternative acid or basic catalysis, which may
mation, Figure S4) in which chemical exchange cross-sig- be detrimental to the coordinated complex. The materials
nals were observed. For instance, the signals at δ = 8.59 and were characterized by solid-state 29Si NMR spectroscopic
8.85 ppm would correspond to the same H-2 proton of the analysis, N2-sorption measurements, and elemental analy-
3-chloropyridine ring but in a different chemical environ- sis; the amount of palladium was also determined by induc-
ment. Secondly, two different triplets at δ = 4.74 and tively coupled plasma (ICP) analysis (Table 1). The presence
4.80 ppm, with a relative proportion of 1:0.25, were also of the organic ligand in the hybrid materials was confirmed
observed, corresponding to the protons of the methylene by solid state 29Si NMR analysis, which showed two sets of
group attached to the nitrogen atom of the NHC ligand for chemical shifts: T units at around δ = –55 to –75 ppm, re-
two different palladium complex species. Analogous duplic- sulting from the hydrolysis–condensation of organosilanes
ity of signals was also observed for the triethoxysilyl group 3 or 4, and Q units, ranging from δ = –90 to –120 ppm,
(superimposed methylene and methyl signals of the ethoxy formed from TEOS, as exemplified by the 29Si NMR solid
group centered at δ = 3.86 and 1.25 ppm, respectively). The state spectra of M1 (Figure 1). The low intensities of the T
rotation around the N–Mes bond was not restricted at units with respect to the Q units are due to the high dilution
room temperature in this case, and both ortho methyl of the organometallic moiety within the silica matrix. The
groups appeared as a singlet at δ = 2.2 ppm, whereas the organosilicas present type IV nitrogen sorption iso-
para methyl group presented a signal at δ = 2.3 ppm. As a therms,[19] corresponding to mesoporous materials (Fig-
third point, the spectrum showed additional broad signals ure 2). In both cases, the contribution of micropores is neg-
that were probably due to other palladium species derived ligible, most of the pores being in the mesoporous range.
from the decoordination of 3-chloropyridine. Although dif- Noticeable differences in terms of porosity were observed
ferent spectra were recorded at lower temperature (273 and depending on the precursor used, with the Brunauer–Em-
250 K), no further information could be obtained for this mett–Teller (BET) surface areas varying from 730 to
unknown species due to the fact that the signals remained 810 m2 g–1 (Table 1). Interestingly, the coordinating ligand
broad.
(acac or 3-chloropyridine) induces very important textural
From this study we concluded that palladium complex 4 differences to the resulting materials M1 and M2. A narrow
exists as a mixture of two isomeric structures, one of which pore size distribution is observed for M2, whereas a bi-
Table 1. Analytical and textural data of hybrid silica materials M1 and M2.
M
29Si CP MAS NMR
SBET
Pore
diameter [Å]
Pore
% Pd
mmol
Pd/g
T3
Q2
Q3
Q4
[m2g–1]
volume[a] [cm3g–1]
M1
M2
–67.6
–67.0
–92.8
–92.0
–102.0
–101.8
–110.9
–111.5
810
730
42; 70[b]
100–120
0.91
1.26
2.77
2.48
0.260
0.233
[a] Adsorbed volume at p/p0 = 0.99. [b] Bimodal distribution.
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3627

R. Pleixats et al.
FULL PAPER
modal distribution is found for M1. Significant differences theoretical value of 1:41 (with complete condensation).
in pore size were also observed between the materials Therefore, we can consider that most of the palladium has
(Table 1); indeed medium mesopores (42–70 Å) were ob- been incorporated within the material.
tained for M1, and large mesopores (100–120 Å) for M2
with a similar variation of the pore volume (from 0.91 to
1.26 cm3 g–1).
Assay of Supported Catalysts M1 and M2 in Suzuki, Heck,
and Sonogashira Reactions
The supported catalysts were tested as catalytic systems
(0.2 mol-% Pd) in the Suzuki cross-coupling reactions of
phenylboronic acid (5; 1.5 equiv.) with p-bromoacetophen-
one (6), p-bromoanisole (7), and p-chloroacetophenone (8)
to give the corresponding biaryls 9 and 10 (Scheme 2).
Scheme 2. Suzuki cross-coupling reactions catalyzed by M1 and
M2.
Figure 1. 29Si solid-state CP-MAS NMR spectrum of M1.
The two materials M1 and M2 gave efficient and fast
reactions with the activated aryl bromide 6, giving almost
quantitative isolated yields of 9 in 30 min for the first cycle
under the conditions indicated in Table 2 (K2CO3, DMF/
H2O (95:5), 110 °C). Furthermore, the catalysts could be
reused for five consecutive runs, although slightly longer
reaction times were required to achieve full conversions
upon recycling.
Table 2. Activity and recyclability of M1 and M2 for the reaction
of phenylboronic acid (5) with p-bromoacetophenone (6) to give 9
(see Scheme 2).[a]
Cycle
M1
t [h]
M2
t [h]
9 [%][b]
9 [%][b]
1
2
3
4
5
0.5
0.5
1
1.5
2
100
96
98
98
100
0.5
1
3
7
7
96
94
77
79
86
[a] Reactions performed in a 10 mL sealed tube under magnetic
stirring using 6 (1 mmol), M (0.2 mol-%), 5 (1.5 equiv.), K2CO3
(2 equiv.), DMF/H2O, 95:5 (2 mL), 110 °C. Full conversions at the
indicated times except for third, fourth and fifth cycle for M2, for
which about 90% conversion was observed. [b] Isolated yield.
When similar conditions were applied to the Suzuki reac-
tion between 5 and the deactivated aryl bromide 7
(Scheme 2) good performances were obtained for the first
cycle, but the activity decreased considerably in the second
and third runs (conditions A, Table 3, entries 1–3). It is
worth noting that the reaction progressed rapidly in the first
Figure 2. N2 sorption isotherm of M1 and M2 and plot of the pore
size distribution.
The nitrogen content of the materials was higher than hour and then more slowly. After some experimentation, it
expected (see the experimental section), which could be due was found that under conditions B (KOtBu, iPrOH, 90 °C)
to residual solvent (DMF) remaining in the material. More- the reusability of the supported catalysts was noticeably im-
over, the elemental analyses revealed a Pd/Si ratio of 1:46 proved (Table 3, entries 4–8). Full conversions were not at-
for M1 and M2, which is only slightly different from the tained under these conditions for the indicated times and
3628
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3625–3635

Recyclable Catalysts for Suzuki, Heck and Sonogashira Reactions
the reaction was stopped when no further significant evolu-
tion was observed.
We then assessed the recyclability of M1 and M2 for this
challenging Suzuki coupling. The best conditions (condi-
tions C) found for M1 (Table 4, entry 4) provided an excel-
lent 92% isolated yield of biphenyl 9 in the first cycle, but
the activity decreased drastically for the second run (Ͻ 10%
yield). For M2 under conditions D (Table 4, entry 15), a
79% isolated yield of 9 was obtained in the first cycle after
24 h and the catalyst could be recycled up to three consecu-
tive runs (41 and 35% after 72 h). We then adopted these
conditions (conditions D) for catalyst M1 and, although the
conversion and yield were lower for the first run than under
conditions C, the reusability was improved (77, 53 and 29%
yields after 24, 48 and 48 h, respectively, for three consecu-
tive runs) (Table 4, entry 8).
Table 3. Activity and recyclability of M1 and M2 for the Suzuki
reaction of phenylboronic acid (5) with p-bromoanisole (7) to give
10 (see Scheme 2).
Entry Conditions[a] Cycle M1
t [h]
M2
10 [%][b] t [h] 10 [%][b]
1
2
3
4
5
6
7
8
A
A
A
B
B
B
B
B
1
2
3
1
2
3
4
5
21.5
24
24
2
5
5
82
93
28
74
65
68
67
65
21.5 90
24
24
2
84
26
75
65
72
60
67
5
5
5
24
5
5
Taking into account the isolated yields of 9 obtained un-
der the conditions detailed in entries 8 and 15 of Table 4,
the turnover numbers (TON) and turnover frequencies
(TOF) for this reaction were 385 and 16 h–1 (for M1), and
395 and 16 h–1 (for M2). These values are higher than those
obtained for this aryl chloride with the systems Pd(OAc)2/
silica-supported imidazolium salts[16] and similar to those
recently found for the same reaction with other silica-sup-
ported Pd–NHC catalysts.[15,20] When the amount of cata-
lyst was reduced from 0.2 to 0.01 mol-%, higher values were
obtained but, unfortunately, the progress of the reaction
was found to stop at low conversions (for M1: 8% at 12 h,
TON 800, TOF 66 h–1; for M2: 9% at 3 h, TON 900, TOF
300 h–1).
[a] Conditions A: reactions performed in a 60 mL sealed tube un-
der orbital stirring and an inert atmosphere, using 7 (2 mmol), M
(0.2 mol-%), 5 (1.5 equiv.), K2CO3 (2 equiv.), degassed DMF/H2O,
95:5 (4 mL), 100 °C. Conditions B: reactions performed in a 10 mL
sealed tube under magnetic stirring and inert atmosphere, using 7
(2 mmol), M (0.2 mol-%), 5 (1.2 equiv.), KOtBu (1.3 equiv.), de-
gassed iPrOH (2 mL), 90 °C. [b] Isolated yield.
We then turned to the more challenging aryl chloride 8
(Scheme 2), for which a screening of different reaction con-
ditions (base, solvent, temperature, catalyst loading, con-
centration) was performed using M1 and M2 as catalysts
(Table 4). For M1, the best result was found using 0.2 mol-
% catalyst and K2CO3 as base, in degassed THF/H2O
(80:20) under an inert atmosphere at 110 °C (95% GC yield
after 24 h, although the reaction was almost complete after
It is worth mentioning that all the Suzuki coupling reac-
5 h) (conditions C; Table 4, entry 4). In contrast, the use of tions with aryl bromides and chlorides gave very selective
M2 as catalyst under the same conditions afforded only reactions because neither the homocoupling of phenyl-
44% GC yield of 9 after 24 h (Table 4, entry 16) whereas boronic acid nor formation of other side products was ob-
88% GC yield was obtained in ethanol at 110 °C with served. A simple experimental protocol was followed that
KOtBu as base (conditions D; Table 4, entry 15).
allowed easy recovery of the catalyst and isolation of pure
Table 4. Suzuki reaction between phenylboronic acid (5) and p-chloroacetophenone (8) to give 9 with supported catalysts M1 and M2
(see Scheme 2).
Entry
M [mol-%]
[8] (m)
Base [equiv.]
Solvent
T [°C]
9 [%][a]
1
2
3
4
5
6
7
8
M1 (0.2)
M1 (1.0)
M1 (0.2)
M1 (0.2)
M1 (0.2)
M1 (1.0)
M1 (0.2)
M1 (0.2)
M2 (0.2)
M2 (1.0)
M2 (0.2)
M2 (0.2)
M2 (0.2)
M2 (1.0)
M2 (0.2)
M2 (0.2)
M2 (0.2)
M2 (0.2)
M2 (0.2)
M2 (0.2)
0.5
0.5
1.0
1.0
1.0
1.0
0.5
0.5
0.5
0.5
1.0
0.5
1.0
1.0
0.5
1.0
1.0
0.5
0.5
0.5
K2CO3 (2)
K2CO3 (2)
K2CO3 (2)
K2CO3 (2)
KOtBu (1.3)
KOtBu (1.3)
KOtBu (1.3)
KOtBu (1.3)
K2CO3 (2)
K2CO3 (2)
K2CO3 (2)
K2CO3 (2)
KOtBu (1.3)
KOtBu (1.3)
KOtBu (1.3)
K2CO3 (2)
K2CO3 (2)
KOH (2)
DMF/H2O 95:5
DMF/H2O 95:5
DMF/H2O 95:5
THF/H2O 80:20
iPrOH
iPrOH
iPrOH
110
110
110
110
110
110
110
110
110
110
110
130
110
110
110
110
130
110
110
110
69
81
63
95[b]
48
63
44
EtOH
86[c]
66
9
DMF/H2O 95:5
DMF/H2O 95:5
DMF/H2O 95:5
DMF/H2O 95:5
iPrOH
iPrOH
EtOH
THF/H2O 80:20
toluene/H2O 80:20
H2O
dioxane
dioxane
10
11
12
13
14
15
16
17
18
19
20
73
70
68
75
83
88[d]
44
13
28
64
74
Cs2CO3 (2)
KOH (2)
[a] GC yield (undecane standard) after 24 h of reaction performed in a 45 mL sealed multireactor tube under an inert atmosphere and
magnetic stirring in degassed solvents. [b] 92% isolated yield, but Ͻ10% yield in the second cycle after 72 h. [c] 77% isolated yield; 53
and 29% for the second and third cycles after 48 h. [d] 79% isolated yield; 41 and 35% for the second and third cycles after 72 h.
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3629

R. Pleixats et al.
FULL PAPER
biphenyls 9 and 10 without the need for chromatographic loading of 0.01 mol-% (for M1: 86% conversion at 6 h,
purification (see experimental section).
TON 8600, TOF 1433 h–1; for M2: 85% conversion at 4.5 h,
In all cases, the solid catalysts turned grey after the first TON 8500, TOF 1900 h–1). Similar values were reported by
cycle and darkened progressively to black after several runs, Karimi and Enders[22] in the reaction between p-bromo-
which is an indication of the in situ formation of palladium acetophenone and methyl acrylate with Pd–NHC systems
nanoparticles. This was confirmed by high-resolution trans- immobilized in silica. For this type of supported catalyst,
mission electron microscopy (HR-TEM), which revealed higher TOF could be achieved for this reaction by the use
particle diameters ranging from 2.3 to 4.5 nm depending of microwave activation.[23]
on catalyst and reaction (see the Supporting Information,
Table S1). Electron diffraction (ED) of the samples exhib-
ited the characteristic pattern of face-centered cube (fcc)
palladium(0) (see the Supporting Information, Figures S5,
S6, and S7).
The amounts of Pd in the final biphenyl products 9 and
10 were determined by ICP-MS analysis (see the Support-
ing Information, Table S1), and leaching values ranging
Scheme 3. Mizoroki–Heck reaction catalyzed by M1 and M2.
from 0.3 to 2.8% with respect to the initial palladium were
found, depending on the catalyst, substrate, and conditions.
Hot filtration tests performed with all catalytic materials
for the Suzuki cross-coupling reaction between phenylbor-
onic acid (5) and p-bromoacetophenone (6) suggest that a
homogeneous pathway plays a significant role in the Suzuki
coupling reactions, with homogeneous Pd species released
from the immobilized palladium systems being the true cat-
alysts; these species would then be redeposited on the func-
tionalized silica support.[21] When M1 and M2 were filtered
off from the hot reaction mixture under the conditions de-
tailed in Table 2 after 5 and 2 min, respectively (53 and 22%
GC conversion of 6) and the remaining filtrates were made
to react under the same conditions after addition of a new
equivalent of base, the GC conversions increased to 100 and
96%, respectively, after 30 min.
Table 5. Activity and recyclability of M1 and M2 for the Heck reac-
tion between p-bromoacetophenone (6) and n-butyl acrylate (11)
to give 12 (see Scheme 3).[a]
Cycle
M1
t [h]
M2
t [h]
12 [%][b]
12 [%][b]
1
2
3
4
5
3
2
3
3
3
98
99
98
96
97
3
2
3
3
3
98
99
99
98[c]
95[c]
[a] Reactions performed in a 10 mL sealed tube under magnetic
stirring, with 6 (2 mmol), 11 (1.5 equiv.), NBu3 (1.5 equiv.) in DMF
(4 mL). [b] Isolated yield. [c] Yields based on 1H NMR spectro-
scopic analysis. No side products observed, only starting substrate
and 12.
We then performed the Mizoroki–Heck reaction between
p-bromoacetophenone (6) and n-butyl acrylate (11) to give
n-butyl 4-acetyl-trans-cinnamate (12; Scheme 3). Classical
For both M1 and M2 a hot filtration test revealed the
conditions were adopted (NBu3 as base in DMF at 150 °C). homogeneous nature of the catalytic Heck process. Thus,
For M1 and M2, almost quantitative isolated yields of 12 the catalysts were filtered off after 3 min of reaction (31 and
were obtained in 3 h, and the systems were successfully re- 39% GC conversion), the filtrates were left under the same
cycled up to five runs (Table 5). No side products were conditions and the conversions increased until 100% after
formed for any of the catalysts and the cinnamate derivative 15 and 10 min, respectively. The content of Pd in the crude
12 was isolated in pure form without the need for chroma- mixture of the Heck reaction was found to be 12.2 and
tographic purification. Again, HR-TEM observations re- 5.6 ppm for M1 and M2, respectively (0.9 and 0.5% of
vealed the in situ formation of Pd0 nanoparticles, which leaching with respect to initial palladium).
were larger in size than those found previously in Suzuki
Finally, the immobilized Pd–NHC complexes were as-
couplings (12.6 nm for M1 and 19.5 nm for M2, see the sayed in the Sonogashira reaction between p-bromoaceto-
Supporting Information, Figure S8). When the reaction phenone (6) and phenylacetylene (13), to afford 1-(4-acet-
progress was monitored over time in the first and second ylphenyl)-2-phenylacetylene (14), under the conditions
cycle for both M1 and M2 (see the Supporting Information, summarized in Scheme 4 (tetrabutylammonium acetate as
Figures S9–S10 for M1 and Figures S11–S12 for M2) we base, dimethylformamide, 110 °C), which had previously
found a fast reaction that required much shorter times than provided successful results for other silica-supported sys-
those mentioned in Table 5. Thus, complete conversion was tems.[14b] In all cases, conversions were complete after one
attained after 15 min for the first and second runs in the hour for the first run, good isolated yields of disubstituted
case of M1 (TON 500, TOF 2000 h–1). For M2, it took only acetylene 14 being obtained after chromatographic purifica-
10 min to reach 100% conversion in the first cycle (TON tion of the crude mixture (for M1: TON 355; for M2: TON
500, TOF 3000 h–1), whereas in the second cycle some de- 365). The materials could be recycled up to five consecutive
crease of activity was found (93% conversion after 20 min, runs, although increased reaction times were needed for
TON 465, TOF 1410 h–1). The reaction was also success- complete conversion (4 h) from the second to the fifth cy-
fully performed under the same conditions with a catalyst cles (Table 6). It is worthwhile noting that the recovered
3630
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3625–3635

Recyclable Catalysts for Suzuki, Heck and Sonogashira Reactions
catalytic materials of the Sonogashira reactions did not
The reaction cycle for C–C coupling reactions with sup-
contain any nanoparticles, and no darkening was observed ported Pd catalysts may be complex and involve a consider-
after the successive runs.
able number of processes with Pd colloids on the support
or in the solution, molecular Pd0, and/or PdII species either
supported or released in the solution, with the nature of the
truly active catalytic species not being well-established in
most cases.[25] The dissolution and redeposition of palla-
dium has been proven by several authors.[21,25a] In our case,
hot filtration tests suggest that the reactions take place, at
least in part, through a homogeneous pathway. Whereas in
the Suzuki and Heck reactions the formation of Pd nano-
particles is observed upon recycling the materials, this is not
the case for the Sonogashira reaction. Both the inorganic
matrix and the NHC moieties of our hybrid silica materials
should have a contribution to the stabilization of nanopar-
ticles or other Pd species formed in the processes. However,
palladium nanoparticles have also been claimed to act as a
reservoir in C–C coupling reactions, from which a small
amount of leached palladium atoms or low molecular
weight catalytically active Pd species may be formed.[25]
Scheme 4. Sonogashira reaction catalyzed by M1 and M2.
Table 6. Activity and recyclability of M1 and M2 for the Sonoga-
shira reaction between p-bromoacetophenone (6) and phenylacetyl-
ene (13) to give 14 (see Scheme 4).[a]
Cycle
M1
t [h]
M2
14 [%][b]
t [h]
14 [%][b]
1
2
3
4
5
1
4
4
4
5
90 (71)
76
79
84
62
1
4
4
4
4
87 (73)
78
82
83
86
[a] Reactions performed in a 10 mL sealed tube under magnetic
stirring, with 6 (2 mmol), 13 (1.5 equiv.), NBu4OAc (1.45 equiv.) in
DMF (4 mL). [b] GC yield (undecane standard). Isolated yield af-
ter chromatography given in brackets.
Conclusions
Two hybrid silica gels (M1 and M2) have been prepared
by cocondensation of TEOS with silylated Pd–NHC com-
The content of Pd in the crude mixture of the Sonoga- plexes (3 and 4). These were obtained by reaction of a sil-
shira reaction was determined by ICP-MS and revealed that ylated mesityl-imidazolium derivative (2) with Pd(acac)2 or
leaching was very low (3.9 and 3.1 ppm for M1 and M2, PdCl2/3-chloropyridine, respectively. These two supported
respectively, corresponding to a loss of 0.3 and 0.2% with catalysts were evaluated for the Heck, Suzuki, and Sonoga-
respect to initial palladium).
shira coupling reactions. It is worth mentioning that, al-
Monitoring the progress of the reaction over time though the corresponding homogeneous complexes [Pd-
showed that the Sonogashira reaction was faster than ex- (acac)Cl(NHC)] and [PdCl2(3-ClPyr)(NHC)] have not been
pected. For the first run with catalyst M1 complete conver- described in Heck and Sonogashira reactions, both materi-
sion was achieved after 5 min. The catalyst had lost some als performed efficiently in all three catalytic reactions and,
activity in the second cycle, with conversion reaching a notably, very fast reactions and excellent recyclabilities (for
value of 90% after 10 min, and the reaction subsequently at least five runs) were obtained for aryl bromides. For the
progressed very slowly (see the Supporting Information, Suzuki reaction with the challenging aryl chloride 8, signifi-
Figures S13–S14).
cantly better conversions were achieved than with pre-
The reaction was even faster with M2 than with M1 viously reported[16] systems with imidazolium-immobilized
(100% GC conversion after 2 min for the first run; for the silicas, and the catalysts could be reused for up to three
second run the activity decreased and required 10 min for runs. Moreover, no homocoupling reaction was observed
the reaction to complete) (see the Supporting Information, and no other side products were obtained in the Suzuki or
Figure S15).
Heck processes. Hence, the corresponding coupling prod-
Due to the fast processes observed, we then performed ucts were easily obtained without any need for chromato-
the reaction with a much lower loading of catalyst graphic separation. For the copper- and phosphane-free
(0.01 mol-%). Under these conditions, the GC yields of al- Sonogashira reaction, we found TON and TOF values up
kyne 14 after 1.5 h were 54% for M1 (TON 5400, TOF to 5400 and 3600 h–1, respectively, which are much higher
3600 h–1) and 48% for M2 (TON 4800, TOF 3200 h–1). It than those reported for other silica-supported Pd–NHC
is worthwhile mentioning that silica-supported Pd–NHC systems previously described. Moreover, we have shown
systems have rarely been used as recyclable catalysts in the here that the materials act as a reservoir of the Pd species
Sonogashira reaction.[15] To the best of our knowledge, only that actually react in solution. In the Suzuki and Heck reac-
one report describes the coupling between iodobenzene and tions the formation of Pd nanoparticles is observed upon
phenylacetylene and, in this case, quite low yields and TOF recycling the materials, whereas this is not the case for the
values (0.08 h–1) were reported.[24] A hot filtration test for Sonogashira reaction. However, palladium nanoparticles
the second run with M1 and M2 also suggested a homogen- have also been claimed to be a reservoir in C–C coupling
eous pathway, despite the fact that, in these cases, no nano- reactions, from which a small amount of leached palladium
particles were formed.
atoms or low molecular weight catalytically active Pd spe-
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3631

R. Pleixats et al.
FULL PAPER
Ar-CH3 and NCH2CH2), 1.14 (t, 3J = 7.0 Hz, 9 H, OCH2CH3),
0.55–0.60 (m, 2 H, SiCH2) ppm. 13C NMR (90 MHz, CDCl3): δ =
141.1, 138.6, 134.1, 130.8, 129.8, 123.2, 122.8, 58.6, 52.0, 24.6, 21.1,
18.3, 17.6, 6.9 ppm.
cies may be formed. Furthermore, this work demonstrates
that the use of periodic mesoporous materials obtained with
surfactants is not a prerequisite and that a simple sol-gel
cocondensation with TEOS to generate nonuniform meso-
porosity is enough to achieve such performance. Similar
Synthesis of the Monosilylated Pd–NHC Complex 3: In a Schlenk
efficiency (recyclability, selectivity, yield, and cleanliness of tube under argon, 2 (0.533 g, 1.25 mmol) was suspended in anhy-
drous dioxane (18 mL). This suspension was transferred to a
Schlenk tube containing Pd(acac)2 (0.318 g, 1.04 mmol) and the
resulting mixture was stirred for 20 h at 100 °C under argon. The
solvent was removed under vacuum, the residue was extracted with
anhydrous Et2O and the resulting suspension filtered off. The fil-
trates were concentrated under vacuum and the final solid was fur-
ther washed with anhydrous pentane to afford 3 as a yellow solid
the reaction) was quite recently observed for the direct
asymmetric aldol reaction with supported hybrid silica-
based organocatalysts derived from prolinamide, despite the
low surface area obtained through the sol-gel process.[26]
1
(0.547 g, 83%). M.p. 132–134 °C. H NMR (500 MHz, CDCl3): δ
Experimental Section
3
= 7.13 (d, J = 1.5 Hz, 1 H, CH=CH), 7.01 (s, 1 H, Ar–H), 6.89
General Remarks: When required, experiments were carried out
with standard high vacuum and Schlenk techniques, and the sol-
vents were dried and distilled just before use following standard
3
(s, 1 H, Ar–H), 6.84 (d, J = 1.5 Hz, 1 H, CH=CH), 5.17 (s, 1 H,
CH acac), 5.07–5.01 (m, 1 H, CHHCH2CH2Si), 4.45–4.39 (m, 1 H,
3
CHHCH2CH2Si), 3.82 (q, J = 7.0 Hz, 6 H, OCH2CH3), 2.34 (s, 3
1
procedures. Solution H and 13C NMR spectra were recorded with
H, Ar-CH3), 2.24 (s, 3 H, Ar-CH3), 2.15 (m, 2 H, CH2CH2Si), 2.01
(s, 3 H, Ar-CH3), 1.91 (s, 3 H, CH3 acac), 1.75 (s, 3 H, CH3 acac),
1.22 (t, 3J = 7.0 Hz, 9 H, OCH2CH3), 0.80–0.72 (m, 2 H,
CH2Si) ppm. 13C NMR (100.6 MHz, CDCl3): δ = 187.4 (C=O),
183.5 (C=O), 151.9 (Pd-C), 139.2 (C), 136.7 (C), 135.9 (C), 135.0
(C), 129.7 (CH), 128.6 (CH), 123.4 (CH), 121.9 (CH), 99.8 (CH),
58.6 (CH2), 53.5 (CH2), 27.3 (CH3), 25.7 (CH3), 24.9 (CH2), 21.3
(CH3), 18.8 (CH3), 18.4 (CH3), 17.9 (CH3), 7.5 (CH2-Si) ppm. IR
Bruker DRX-250 MHz, DPX-360 MHz, AVANCE-III 400 MHz
or AVANCE 500 MHz spectrometers. The 29Si CP-MAS solid-state
NMR spectra were recorded with a Bruker AV-400-WB spectrome-
ter. All NMR experiments were performed at the Servei de
Ressonància Magnètica Nuclear of the Universitat Autònoma de
Barcelona. HRMS were recorded by the Servei dЈAnàlisi Química
at the Universitat Autònoma de Barcelona, with an ESI hybrid
quadrupole time of flight microTOFQ spectrometer from Bruker
Daltonics. IR data were obtained with a Bruker Tensor 27 spectro-
photometer with ATR Golden Gate and with a Perkin–Elmer (Sys-
tem 2000 FTIR) spectrophotometer. N2-sorption isotherms were
obtained with a Micromeritics ASAP2020 analyzer at the Institut
Charles Gerhardt Montpellier after degassing the materials at
55 °C for 30 h under vacuum. The surface areas and the pore-size
distributions were determined by using the Brunauer–Emmett–
Teller (BET) and the Barrett–Joyner–Halenda (BJH) methods,
respectively. Elemental analyses were performed at the Serveis Ci-
entificotècnics of the Universitat de Barcelona, at the Servei dЈAnà-
lisi of the Universitat Autònoma de Barcelona or at the Servei de
Microanàlisi of the CSIC in Barcelona. The content of palladium
was determined at the Serveis Cientificotècnics of the Universitat
de Barcelona by inductively coupled plasma (ICP) analysis. High-
resolution transmission electron microscopy (HR-TEM) observa-
tions were carried out at the Servei de Microscòpia of the Uni-
versitat Autònoma de Barcelona with a JEOL JEM 2011 (200 kV).
Chromatographic purifications were performed under N2 pressure
using 230–400 mesh silica gel (flash chromatography). Distilled and
deionized water (MilliQ) was used for the sol-gel process. GC
analyses were performed with an Agilent Technologies 7890A in-
strument equipped with an Agilent HP-5 capillary column. 1-Mes-
itylimidazole (1),[17b] was prepared according to previously de-
scribed procedures. Compounds 9,[27] 12,[28] and 14[29] are described
in the literature and their spectroscopic data matched those pre-
viously reported.
(ATR): ν = 2971, 2920 and 2885 (H–Csp3), 1579 and 1514 (C=O–
˜
Pd), 1101 and 1075 (Si–O–C) cm–1. IR (PEG): ν = 339, 301 and
˜
215 (Pd–Cl) cm–1. HRMS (ESI): calcd. for C26H41ClN2O5PdSi +
Na+ [M + Na+] 655.1403; found 655.1396.
Synthesis of the Monosilylated Pd–NHC Complex 4: In a Schlenk
tube under argon, PdCl2 (0.266 g, 1.50 mmol),
2 (0.713 g,
1.67 mmol), and K2CO3 (99%; 1.038 g, 7.41 mmol) were suspended
in freshly distilled 3-chloropyridine (6.00 mL, 1.21 g/cm3,
64 mmol). The mixture was stirred overnight at 80 °C under argon,
then anhydrous CH2Cl2 (6 mL) was added and the solution was
placed on a short silica gel/Celite® column and eluted with anhy-
drous CH2Cl2 (3ϫ4.5 mL). The volatiles of the eluted fraction
were distilled off under vacuum (1 Torr, 40 °C) and the residue was
washed with anhydrous pentane, affording 4 (0.734 g, 72%) as a
1
4
pale-brown solid. H NMR (500 MHz, CDCl3): δ = 8.85 (d, J =
3
5
2.5 Hz, 1 H, Ar–H pyr), 8.75 (dd, J = 5.5 Hz, J = 1.0 Hz, 1 H,
Ar–H pyr), 7.68–7.66 (m, 1 H, Ar–H pyr), 7.21–7.15 (m, 1 H, Ar–
H pyr), 7.00–6.90 (m, 4 H, Ar–H Mes and NCH=CHN), 4.74 (t,
3J = 7.5 Hz, 2 H, CH2CH2CH2Si), 3.86 (q, 3J = 7.0 Hz, 6 H,
OCH2CH3), 2.35 (s, 3 H, Ar-CH3), 2.25 (s, 6 H, Ar-CH3), 1.25–
3
1.22 (m, 2 H, CH2CH2Si), 1.25 (t, J = 7.0 Hz, 9 H, OCH2CH3),
0.83–0.81 (m, 2 H, CH2Si) ppm. Signals corresponding to free 3-
chloropyridine were also observed due to a complexation–decom-
plexation equilibrium. 13C NMR (100.6 MHz, CDCl3): δ = 150.5,
149.5, 149.1, 147.7, 139.3, 137.9, 136.6, 136.5, 135.9, 135.1, 135.0,
132.4, 132.3, 129.35, 129.27, 129.2, 128.3, 124.6, 124.5, 124.1,
122.2, 121.8, 58.7 (CH2), 53.6 (CH2), 24.2 (CH2), 21.2 (CH3), 19.0
1-Mesityl-3-[3-(triethoxysilyl)propyl]-1H-imidazol-3-ium Chloride
(2): A mixture of 1-mesitylimidazole 1 (0.601 g, 3.23 mmol) and (3-
chloropropyl)triethoxysilane (0.90 g, 3.55 mmol) was stirred under
argon at 90 °C for 5 days. The volatiles were removed under vac-
uum and the residue was thoroughly washed with anhydrous pent-
ane to afford 2 as a solid (1.166 g, 85%). M.p. 134–135 °C (Lit.[18]
m.p. 135 °C). 1H NMR (360 MHz, CDCl3): δ = 10.54 (s, 1 H,
NCHN), 7.75 (s, 1 H, NCH=CHN), 7.20 (s, 1 H, NCH=CHN),
(2ϫCH ), 18.5 (CH ), 7.6 (CH -Si) ppm. IR (ATR): ν = 3092 (H–
˜
3
3
2
Csp2), 2971 and 2918 (H–Csp3), 1463, 1419, 1220, 1073, 825,
691 cm–1. IR (PEG): ν = 379, 348 and 323 (Pd–Cl) cm–1. MS (ESI):
˜
m/z (%) = 644.2 (100) [M – Cl]+. C26H38Cl3N3O3PdSi (681.45):
calcd. C 45.82, H 5.62, N 6.17, Cl 15.61; found C 45.52, H 5.06,
N 6.91, Cl 16.03.
Synthesis of M1: Complex 3 (0.378 g, 0.598 mmol) and TEOS
(98%; 5.06 g, 23.8 mmol) were dissolved in anhydrous DMF
(14.7 mL). To this solution was added a mixture of distilled and
3
3
6.91 (s, 2 H, Ar–H), 4.64 (t, J = 6.5 Hz, 2 H, NCH2), 3.74 (q, J
= 7.0 Hz, 6 H, OCH2CH3), 2.26 (s, 3 H, Ar-CH3), 1.99 (m, 8 H,
3632
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3625–3635

Recyclable Catalysts for Suzuki, Heck and Sonogashira Reactions
deionized water (1.49 mL, 82.5 mmol) and NH4F (1 m aqueous
solution, 0.260 mL, 14.4 mmol H2O, 0.260 mmol F) in anhydrous
DMF (9.8 mL). The resulting yellow solution (molar composition
3/TEOS/H2O/F = 1:40:162:0.4) was shaken for 1 min at room tem-
perature and then allowed to stand. After 1 h a gel formed that
was aged for 5 days at room temperature. The gel was crushed,
filtered, and washed with water (ϫ2), EtOH (ϫ3), and CH2Cl2
(ϫ3). The resulting solid was dried overnight at 60 °C under vac-
uum (1 Torr) affording M1 (1.81 g, 0.260 mmol Pd/g) as a pale-
yellow solid. 29Si CP-MAS NMR (79.5 MHz): δ = –58.6 (T2), –67.6
EtOAc (10 mL) was added to the reaction mixture, the suspension
was centrifuged and the supernatant solution was separated; this
process was repeated twice. The ethyl acetate from the combined
layers was removed under reduced pressure and, upon addition of
water to the remaining liquid phase, compound 10 precipitated,
which was filtered off and washed with water. Compound 10 was
dissolved again in EtOAc, the solution was dried with anhydrous
Na2SO4, and the solvent was removed under reduced pressure to
give 10 as a white solid. The catalyst that had been separated by
centrifugation was washed successively with EtOH (ϫ2), water
(T3), –92.8 (Q2), –102.0 (Q3), –110.9 (Q4) ppm. C20H26ClN2O2- (ϫ2), EtOH (ϫ2), and Et2O (ϫ2), then dried under vacuum
SiO1.5Pd·40SiO2 (2923.8) calcd. (considering complete condensa-
tion) C 8.22, H 0.89, N 0.96, Cl 1.21, Si 39.4, Pd 3.64; found C
11.44, H 2.26, N 1.88, Cl 1.09, Si 33.9, Pd 2.77. SBET 810 m2/g;
pore diameter: bimodal distribution centered at 42 and 70 Å; pore
volume: 0.91 cm3/g.
(2 Torr) and directly reused in the next cycle.
Procedure for the Suzuki Cross-Coupling between 5 and 7 To Give
10 (Conditions B, Table 3): In a sealed tube (10 mL) with a mag-
netic stir bar under argon, the supported catalyst (0.004 mmol), p-
bromoanisole (7; 0.255 mL, 1.49 g/cm3, 2.0 mmol), phenylboronic
acid (5; 0.301 g, 2.4 mmol) and KOtBu (0.300 g, 2.6 mmol) were
suspended in degassed 2-propanol (2 mL). The mixture was stirred
at 90 °C and the reaction was monitored by GC analysis. After the
time indicated in Table 3, EtOAc (5 mL) was added to the reaction
mixture, the suspension was centrifuged, and the supernatant solu-
tion separated; this process was repeated twice. The ethyl acetate
from the combined layers was removed under reduced pressure
and, upon addition of water to the remaining liquid phase, com-
pound 10 precipitated, which was filtered off and washed with
water. Compound 10 was dissolved again in EtOAc, the solution
was dried with anhydrous Na2SO4, and the solvent was removed
under reduced pressure to furnish 10 as a white solid. The catalyst
that had been separated by centrifugation was washed successively
with EtOH (ϫ2), water (ϫ2), EtOH (ϫ2) and Et2O (ϫ2), then
dried under vacuum (2 Torr) and directly reused in the next cycle.
Synthesis of M2: Complex 4 (0.738 g, 1.08 mmol) and TEOS 98%
(9.203 g, 43.3 mmol) were dissolved in anhydrous DMF (26.6 mL).
To this solution was added a mixture of distilled and deionized
water (2.74 mL, 152 mmol) and NH4F (1 m aqueous solution,
0.445 mL, 24.7 mmol H2O, 0.445 mmol F) in anhydrous DMF
(17.7 mL). The resulting yellow solution (molar composition 4/
TEOS/H2O/F = 1:40:163:0.4) was shaken for one minute at room
temperature and then allowed to stand. After 3 h a gel formed that
was aged for 5 days at room temperature. At this time, the gel was
crushed, filtered, and washed with water (ϫ2), EtOH (ϫ3), and
CH2Cl2 (ϫ3). The resulting solid was dried overnight at 60 °C un-
der vacuum (1 Torr) affording M2 (3.65 g, 0.233 mmol Pd/g) as a
pale-yellow solid. 29Si CP-MAS NMR (79.5 MHz): δ = –59.1 (T2),
–67.0 (T3), –92.0 (Q2), –101.8 (Q3), –111.5 (Q4) ppm.
C20H23Cl3N3O3PdSiO1.5·40SiO2 (3021.7) calcd. (considering com-
plete condensation) C 7.95, H 0.76, N 1.39, Si 38.11, Pd 3.52; found
C 11.65, H 2.29, N 1.97, Si 30.61, Pd 2.48. SBET 730 m2/g; pore
diameter: distribution centered at 110 Å; pore volume: 1.26 cm3/g.
Procedure for the Suzuki Cross-Coupling between 5 and 8 To Give 9
by Using K2CO3 as Base and THF/H2O (80:20) as Solvent
(Entries 4 and 16, Table 4): In a sealed multireactor tube (45 mL)
with a magnetic stir bar under argon, the supported catalyst
(0.004 mmol), p-chloroacetophenone (8; 2.0 mmol), phenylboronic
acid (5; 3.0 mmol), and K2CO3 (4.0 mmol) were suspended in a
degassed mixture of THF/H2O (80:20, 2 mL). The mixture was
stirred at 110 °C and the reaction was monitored by GC analysis.
After the time indicated in Table 4, EtOAc (10 mL) was added to
the reaction mixture, the suspension was centrifuged, and the su-
pernatant solution was separated; this process was repeated twice.
The ethyl acetate from the combined layers was removed under
reduced pressure and, upon addition of water to the remaining li-
quid phase, a crude product precipitated that was filtered off and
washed with water. The latter was dissolved in EtOAc, the solution
was dried with anhydrous Na2SO4, and the solvent was removed
under reduced pressure to give 9 as a white solid. The supported
catalyst that had been separated by centrifugation was washed suc-
cessively with EtOH (ϫ2), water (ϫ2), EtOH (ϫ2), and Et2O
(ϫ2), then dried under vacuum (2 Torr) and directly reused in the
next cycle.
Catalytic Tests
Procedure for the Suzuki Cross-Coupling between 5 and 6 To Give 9
(Table 2): In a sealed tube (10 mL) with a magnetic stir bar, the
supported catalyst (0.002 mmol), p-bromoacetophenone (6;
0.203 g, 1.00 mmol), phenylboronic acid (5; 0.187 g, 1.5 mmol), and
K2CO3 (0.279 g, 2.0 mmol) were suspended in a mixture of DMF
and H2O (95:5, 2 mL). The mixture was stirred at 110 °C and the
reaction was monitored by GC analysis. After the time indicated
in Table 2, EtOAc (5 mL) was added, the suspension was centri-
fuged, and the supernatant solution separated; this process was re-
peated twice. The ethyl acetate from the combined layers was re-
moved under reduced pressure and, upon addition of water to the
remaining liquid phase, a crude product precipitated that was fil-
tered off and washed with water. The latter was dissolved in EtOAc,
the solution was dried with anhydrous Na2SO4, and the solvent
was removed under reduced pressure to give 9 as a white solid.
The catalyst that had been separated by centrifugation was washed
successively with EtOH (ϫ2), water (ϫ2), EtOH (ϫ2), and Et2O
(ϫ2), then dried under vacuum (2 Torr) and directly reused in the
next cycle.
Procedure for the Suzuki Cross-Coupling between 5 and 8 To Give 9
by Using KOtBu as Base and EtOH as Solvent (Table 4, entry 15):
In a sealed multireactor tube (45 mL) with a magnetic stir bar
under argon, the supported catalyst (0.004 mmol), p-chloroaceto-
phenone (8; 2.0 mmol), phenylboronic acid (5; 3.0 mmol), and
KOtBu (2.7 mmol) were suspended in degassed EtOH (4 mL). The
mixture was stirred at 110 °C and the reaction was monitored by
GC analysis. After the time indicated in Table 4, EtOAc (10 mL)
was added to the reaction mixture, the suspension was centrifuged,
and the supernatant solution was separated; this process was re-
Procedure for the Suzuki Cross-Coupling between 5 and 7 To Give
10 (Conditions A, Table 3): In a sealed multireactor tube (60 mL)
with orbital stirring under argon, the supported catalyst
(0.004 mmol), p-bromoanisole (7; 0.255 mL, 1.49 g/cm3, 2.0 mmol),
phenylboronic acid (5; 0.373 g, 3.0 mmol), and K2CO3 (0.563 g,
4.0 mmol) were suspended in a degassed mixture of DMF/H2O
(95:5, 4 mL). The mixture was stirred at 100 °C and the reaction
was monitored by GC analysis. After the time indicated in Table 3,
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3633

R. Pleixats et al.
FULL PAPER
peated twice. The ethyl acetate from the combined layers was re-
moved under reduced pressure and, upon addition of water to the
remaining liquid phase, a crude product precipitated that was fil-
tered off and washed with water. The latter was dissolved back in
EtOAc, the solution was dried with anhydrous Na2SO4, and the
solvent was removed under reduced pressure to give 9 as a white
solid. The supported catalyst that had been separated by centrifu-
gation was washed successively with EtOH (ϫ2), water (ϫ2),
EtOH (ϫ2), and Et2O (ϫ2), then dried under vacuum (2 Torr) and
directly reused in the next cycle.
[1]
a) S. Kotha, K. Lahiri, D. Kashinath, Tetrahedron 2002, 58,
9633–9695; b) A. Suzuki, J. Organomet. Chem. 2002, 653, 83–
90; c) F. Bellina, A. Carpita, R. Rossi, Synthesis 2004, 2419–
2440; d) N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv.
Synth. Catal. 2006, 348, 609–679; e) F. Alonso, I. P. Beletskaya,
M. Yus, Tetrahedron 2008, 64, 3047–3101.
a) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100,
3009–3066; b) N. J. Whitcombe, K. K. Hii, S. E. Gibson, Tetra-
hedron 2001, 57, 7449–7476; c) F. Alonso, I. P. Beletskaya, M.
Yus, Tetrahedron 2005, 61, 11771–11835; d) V. Polshettiwar, A.
Molnár, Tetrahedron 2007, 63, 6949–6976.
a) K. Sonogashira, J. Organomet. Chem. 2002, 653, 46–49; b)
H. Doucet, J.-C. Hierso, Angew. Chem. 2007, 119, 850; Angew.
Chem. Int. Ed. 2007, 46, 834–871; c) R. Chinchilla, C. Nájera,
Chem. Rev. 2007, 107, 874–922; d) R. Chinchilla, C. Nájera,
Chem. Soc. Rev. 2011, 40, 5084–5121.
a) H. U. Blaser, A. Indolese, F. Naud, U. Nettekoven, A.
Schnyder, Adv. Synth. Catal. 2004, 346, 1583–1598; b) K. C.
Nicolaou, P. G. Bulger, D. Sarlah, Angew. Chem. 2005, 117,
4516; Angew. Chem. Int. Ed. 2005, 44, 4442–4489; c) C. Tor-
borg, M. Beller, Adv. Synth. Catal. 2009, 351, 3027–3043.
[2]
[3]
[4]
[5]
Procedure for the Heck Reaction between 6 and 11 To Give 12
(Table 5): In a sealed tube (10 mL) with a magnetic stir bar, the
supported catalyst (0.004 mmol), p-bromoacetophenone (6;
2.0 mmol), and n-butyl acrylate (11; 3.0 mmol) were suspended in
DMF (4 mL) and the mixture was heated at 150 °C. Tributylamine
(3.0 mmol) was added and the reaction mixture was maintained at
150 °C while stirring (progress of the reaction monitored by GC
analysis). After the time indicated in Table 5, DMF (3 mL) was
added to the reaction mixture, the suspension was centrifuged, and
the supernatant solution was separated; this process was repeated
once more and the solvent from the combined layers was removed
under vacuum. Water was added to the residue and compound 12
was extracted with EtOAc. The combined organic layers were dried
with anhydrous Na2SO4 and concentrated under reduced pressure
to give 12 as an oily product. The supported catalyst that had been
separated by centrifugation was washed successively with water,
EtOH, and Et2O, then dried under vacuum (2 Torr) and directly
reused in the next cycle.
a) A. Zamboulis, N. Moitra, J. J. E. Moreau, X. Cattoën, M.
Wong Chi Man, J. Mater. Chem. 2010, 20, 9322–9338; b) V.
Polshettiwar, C. Len, A. Fihri, Coord. Chem. Rev. 2009, 253,
2599–2626; c) A. Molnár, Chem. Rev. 2011, 111, 2251–2320.
[6]
[7]
V. Farina, Adv. Synth. Catal. 2004, 346, 1553–1582.
a) A. F. Littke, G. C. Fu, Angew. Chem. 2002, 114, 4350; An-
gew. Chem. Int. Ed. 2002, 41, 4176–4211; b) R. B. Bedford,
C. S. J. Cazin, D. Holder, Coord. Chem. Rev. 2004, 248, 2283–
2321.
[8]
a) E. A. B. Kantchev, C. J. O’Brien, M. G. Organ, Angew.
Chem. 2007, 119, 2824; Angew. Chem. Int. Ed. 2007, 46, 2768–
2813; b) N. Marion, S. P. Nolan, Acc. Chem. Res. 2008, 41,
1440–1449; c) S. Díez-González, N. Marion, S. P. Nolan,
Chem. Rev. 2009, 109, 3612–3676; d) G. C. Fortman, S. P. No-
lan, Chem. Soc. Rev. 2011, 40, 5151–5169.
Procedure for the Sonogashira Reaction between 6 and 13 To Give
14 (Table 6): In a sealed tube (10 mL) with a magnetic stir bar,
the supported catalyst (0.004 mmol), p-bromoacetophenone (6;
2.0 mmol) and Bu4NOAc (2.85 mmol) were suspended in DMF
(4 mL) and the mixture was heated at 110 °C. Phenylacetylene 13
(3.0 mmol) was added and the reaction mixture was maintained at
110 °C while stirring (progress of the reaction monitored by GC
analysis). After the time indicated in Table 6, DMF (3 mL) was
added to the reaction mixture, the suspension was centrifuged, and
the supernatant solution was separated; this process was repeated
once more and the solvent from the combined layers was removed
under vacuum. Water was added to the residue and compound 14
was extracted with EtOAc. The combined organic layers were dried
with anhydrous Na2SO4 and concentrated under reduced pressure.
The residue was purified by flash chromatography on silica gel
(hexane/EtOAc, 20:1) to afford 14 as a pale-brown solid. The sup-
ported catalyst that had been separated by centrifugation was
washed successively with water, EtOH, and Et2O, then dried under
vacuum (2 Torr) and directly reused in the next cycle.
[9]
G. A. Grasa, M. S. Viciu, J. Huang, C. Zhang, M. L. Trudell,
S. P. Nolan, Organometallics 2002, 21, 2866–2873.
[10]
a) R. Jackstell, M. G. Andreu, A. Frisch, K. Selvakumar, A.
Zapf, H. Klein, A. Spannenberg, D. Rottger, O. Briel, R.
Karch, M. Beller, Angew. Chem. 2002, 114, 1028; Angew. Chem.
Int. Ed. 2002, 41, 986–989; b) D. R. Jensen, M. J. Schultz, J. A.
Mueller, M. S. Sigman, Angew. Chem. 2003, 115, 3940; Angew.
Chem. Int. Ed. 2003, 42, 3810–3813; c) M. S. Viciu, O. Navarro,
R. F. Germaneau, R. A. Kelly, W. Sommer, N. Marion, E. D.
Stevens, L. Cavallo, S. P. Nolan, Organometallics 2004, 23,
1629–1635; d) M. S. Viciu, E. D. Stevens, J. L. Petersen, S. P.
Nolan, Organometallics 2004, 23, 3752–3755; e) O. Diebolt, P.
Braunstein, S. P. Nolan, C. S. J. Cazin, Chem. Commun. 2008,
3190–3192.
a) N. Marion, E. C. Ecarnot, O. Navarro, D. Amoroso, A. Bell,
S. P. Nolan, J. Org. Chem. 2006, 71, 3816–3821; b) N. Marion,
P. De Fremont, I. M. Puijk, E. C. Ecarnot, D. Amoroso, A.
Bell, S. P. Nolan, Adv. Synth. Catal. 2007, 349, 2380–2384; c)
O. H. Winkelmann, A. Riekstins, S. P. Nolan, O. Navarro, Or-
ganometallics 2009, 28, 5809–5813; d) D. Amoroso, A. Bell,
O. N. Fernandez, S. P. Nolan, N. Marion, US Patent 7442800,
2008.
a) C. J. O’Brien, E. A. B. Kantchev, C. Valente, N. Hadei, G. A.
Chass, A. Lough, A. C. Hopkinson, M. G. Organ, Chem. Eur.
J. 2006, 12, 4743–4748; b) M. G. Organ, S. Avola, I. Dubovyk,
N. Hadei, E. A. B. Kantchev, C. J. O’Brien, C. Valente, Chem.
Eur. J. 2006, 12, 4749–4755; c) M. G. Organ, S. Calimsiz, M.
Sayah, K. H. Hoi, A. J. Lough, Angew. Chem. 2009, 121, 2419;
Angew. Chem. Int. Ed. 2009, 48, 2383–2387; d) K. H. Hoi, S.
Calimsiz, R. D. J. Froese, A. C. Hopkinson, M. G. Organ,
Chem. Eur. J. 2011, 17, 3086–3090; e) M. G. Organ, C. J.
O’Brien, E. A. B. Kantchev, CA Patent 2556850, 2007.
[11]
[12]
Supporting Information (see footnote on the first page of this arti-
cle): Figures S1–S15 and Table S1, as well as the characterization
data for 2, 3, 4, M1, and M2 (29Si CP-MAS NMR, IR, N2-sorption
measurements).
Acknowledgments
We acknowledge financial support from the Spanish Ministerio de
Ciencia e Innovación (MICINN) (project numbers CTQ2009-
07881 and CTQ2009-08328 and a grant to A. M.-M.), Consolider
Ingenio 2010 (project number CSD2007-00006), Generalitat de Ca-
talunya (project number SGR2009-01441 and a grant to G. B.) and
the French Centre National de la Recherche Scientifique (CNRS).
3634
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 3625–3635

Recyclable Catalysts for Suzuki, Heck and Sonogashira Reactions
[13]
[21] a) F. Y. Zhao, B. M. Bhanage, M. Shirai, M. Arai, Chem. Eur.
J. 2000, 6, 843–848; b) A. Biffis, M. Zecca, M. Basato, Eur. J.
Inorg. Chem. 2001, 1131–1133.
[22] B. Karimi, D. Enders, Org. Lett. 2006, 8, 1237–1240.
[23] V. Polshettiwar, R. S. Varma, Tetrahedron 2008, 64, 4637–4643.
[24] A. Corma, C. González-Arellano, M. Iglesias, S. Pérez-Ferr-
eras, F. Sánchez, Synlett 2007, 1771–1774.
[25] a) M. Weck, C. W. Jones, Inorg. Chem. 2007, 46, 1865–1875; b)
D. Astruc, Inorg. Chem. 2007, 46, 1884–1894; c) J. G. de Vries,
Dalton Trans. 2006, 421–429; d) J. Durand, E. Teuma, M.
Gómez, Eur. J. Inorg. Chem. 2008, 3577–3586; e) L. Djakov-
itch, K. Köhler, J. G. De Vries, in: Nanoparticles and Catalysis
(Ed.: D. Astruc), Wiley-VCH, Weinheim, 2008, chapter 10, pp.
303–348; f) V. P. Ananikov, I. P. Beletskaya, Organometallics
2012, 31, 1595–1604.
a) B. Blanco, A. Mehdi, M. Moreno-Mañas, R. Pleixats, C.
Reyé, Tetrahedron Lett. 2004, 45, 8789–8791; b) B. Blanco, M.
Brissart, M. Moreno-Mañas, R. Pleixats, A. Mehdi, C. Reyé,
S. Bouquillon, F. Henin, J. Muzart, Appl. Catal. A 2006, 297,
117–124; c) B. Blanco, M. Moreno-Mañas, R. Pleixats, A.
Mehdi, C. Reyé, J. Mol. Catal. A 2007, 269, 204–213.
a) M. Trilla, R. Pleixats, M. Wong Chi Man, C. Bied, J. J. E.
Moreau, Tetrahedron Lett. 2006, 47, 2399–2403; b) M. Trilla,
R. Pleixats, M. Wong Chi Man, C. Bied, J. J. E. Moreau, Adv.
Synth. Catal. 2008, 350, 577–590.
A. Monge-Marcet, R. Pleixats, X. Cattoën, M.
Wong Chi Man, Catal. Sci. Technol. 2011, 1, 1544–1563.
M. Trilla, G. Borja, R. Pleixats, M. Wong Chi Man, C. Bied,
J. J. E. Moreau, Adv. Synth. Catal. 2008, 350, 2566–2574.
[14]
[15]
[16]
[17] a) M. G. Gardiner, W. A. Herrmann, C. P. Reisinger, J.
Schwarz, M. Spiegler, J. Organomet. Chem. 1999, 572, 239–247;
b) A. J. Arduengo III, F. P. Gentry Jr, P. K. Taverkere, H. E.
Simmons III, US Patent 6177575, 2001.
[18] K. Koehler, K. Weigl, WO Patent 2005016940, 2005.
[19] K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A.
Pierotti, J. Rouquerol, T. Siemieniewska, Pure Appl. Chem.
1985, 57, 603–619.
[20] a) H. Yang, X. Han, G. Li, Z. Ma, Y. Hao, J. Phys. Chem. C
2010, 114, 22221–22229; b) H. Yang, G. Li, Z. Ma, J. Chao, Z.
Guo, J. Catal. 2010, 276, 123–133; c) H. Yang, Y. Wang, Y.
Qin, Y. Chong, Q. Yang, G. Li, L. Zhang, W. Li, Green Chem.
2011, 13, 1352–1361.
[26] a) A. Monge-Marcet, R. Pleixats, X. Cattoën, M.
Wong Chi Man, D. A. Alonso, D. Almasi, C. Nájera, New J.
Chem. 2011, 35, 2766–2772; b) A. Monge-Marcet, X. Cattoën,
D. A. Alonso, C. Nájera, M. Wong Chi Man, R. Pleixats, Green
Chem. 2012, DOI: 10.1039/c2gc35227c.
[27] G. A. Olah, M. Arvanaghi, Y. D. Vankar, J. Org. Chem. 1980,
45, 3531–3532.
[28] X. Cui, Z. Li, C.-Z. Tao, Y. Xu, J. Li, L. Liu, Q.-X. Guo, Org.
Lett. 2006, 8, 2467–2470.
[29] A. Arcadi, S. Cacchi, G. Fabrizi, F. Marinelli, P. Pace, Eur. J.
Org. Chem. 1999, 3305–3313.
Received: February 20, 2012
Published Online: May 23, 2012
Eur. J. Org. Chem. 2012, 3625–3635
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3635