Bimetallic nano-Pd/PdO/Cu system as a highly effective catalyst for the Sonogashira reaction

Article abstract of DOI:10.1016/j.jcat.2014.02.008

A copper-supported nanopalladium catalyst obtained by an innovative method of nanoparticle transfer from the intermediate carrier SiO₂ to the target Cu carrier was a highly efficient and selective catalyst, giving as much as quantitative conver

Full text of DOI:10.1016/j.jcat.2014.02.008

Journal of Catalysis 313 (2014) 1–8
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Bimetallic nano-Pd/PdO/Cu system as a highly effective
catalyst for the Sonogashira reaction
a
a,c
a,
⇑
b
a
Mateusz Korzec , Piotr Bartczak , Anna Niemczyk , Jacek Szade , Maciej Kapkowski ,
a
b
d
a,
⇑
Paulina Zenderowska , Katarzyna Balin , Józef Lel a˛ tko , Jaroslaw Polanski
a
Institute of Chemistry, University of Silesia, Szkolna 9, 40-006 Katowice, Poland
Institute of Physics, University of Silesia, Uniwersytecka 4, 40-007 Katowice, Poland
NANO-CHEM-TECH Ltd, Ligocka 90A/14, 40-568 Katowice, Poland
Institute of Materials Science, University of Silesia, 75 Pułku Piechoty 1A, 41-500 Chorzów, Poland
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
A copper-supported nanopalladium catalyst obtained by an innovative method of nanoparticle transfer
from the intermediate carrier SiO to the target Cu carrier was a highly efficient and selective catalyst,
giving as much as quantitative conversion and yield in a series of Sonogashira reactions. Comparisons
with other Pd systems, especially Pd/SiO , indicated substantial advantages for this novel catalytic sys-
Received 4 January 2014
Revised 9 February 2014
Accepted 14 February 2014
2
2
tem. X-ray photoelectron spectroscopy (XPS) studies of the prepared Pd/Cu catalyst revealed the devel-
opment of a PdO species in the Pd/Cu system and illustrated for the first time the importance of the
resulting Pd/PdO/Cu system for efficient Sonogashira coupling.
Keywords:
Pd nanoparticles
Sonogashira reaction
Bimetallic Pd/Cu
Heterogeneous catalysis
Sonogashira coupling
Palladium oxide
Ó 2014 Elsevier Inc. All rights reserved.
1
. Introduction
Catalytic C–C bond forming reactions provide attractive syn-
reuse [8]. Moreover, numerous investigations have indicated that
0
in coupling reactions, the catalytic active state is a Pd species
and where Pd(II) is used, in situ reduction to Pd0 occurs during
the reaction [9–17].
thetic routes to a variety of important chemicals and pharmaceuti-
cals or their building blocks. Palladium is one of the most
frequently transition metal used for catalytic Heck, Sonogashira,
Suzuki or Stille cross-couplings [1–4]. Although there has been tre-
mendous progress in recent years in this area, a number of prob-
lems remain to be solved [5]. More efficient catalysts are still
sought to allow us to run the reactions in higher yields under
milder conditions [6]. Easier catalyst separation, reuse, and lower
contamination of the final products are other important issues to
be addressed. Additionally, commercial availability at low cost is
required for industrial applications guided by economic efficiency.
The Pd-catalyzed Sonogashira C–C cross-couplings can be per-
formed under both heterogeneous and homogenous conditions
This rationalizes the application of heterogeneous catalysts con-
sisting of an active catalytic Pd species supported on solid materi-
als such as silica, polymers or carbon [18–21]. These types of
catalysts usually are easier for separation and reuse; however, they
often exhibit lower activity compared with many of their homoge-
neous counterparts [8]. Moreover, heterogeneous palladium-cata-
lyzed cross-couplings follow a mechanism in which palladium is
leached from the support in the oxidative addition step [22].
Although palladium re-precipitates later during the reductive
elimination, the activity of the catalyst significantly suffers from
this process [22]. Such a mechanism, termed quasi-heterogenous
catalysis, severely complicates the design of efficient fully hetero-
geneous nano-Pd systems for the demanding cross-couplings,
especially the Sonogashira reaction. One of the solutions is the
use of water as a solvent in which the mechanism was believed
to take the form of a fully heterogeneous process [23]. In another
[
7,8]. In general, homogeneous Pd catalysts are complexes of Pd(II)
salts such as chloride or acetate with such organic ligands as phos-
phines and phosphites. They display high activity, but are rela-
tively difficult to separate from reaction mixtures, especially for
2
+
approach, Choi et al. applied Pd exchanged mesoporous sodalite
and NaA zeolite for catalyzing various C–C couplings and observed
that reactions were catalyzed by a Pd0 species, which was gener-
ated in situ under the reaction conditions [9]. This species seemed
⇑
Corresponding authors. Fax: +48 32 2599978.
E-mail addresses: anna.niemczyk@us.edu.pl (A. Niemczyk), polanski@us.edu.pl
(
J. Polanski).
http://dx.doi.org/10.1016/j.jcat.2014.02.008
021-9517/Ó 2014 Elsevier Inc. All rights reserved.
0

2
M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
to be immediately oxidized back to the initial Pd(II) state when the
reaction was carried out in the presence of O , preventing the for-
with methanol (1000 mL) and centrifugation was repeated. The
scanning electron microscopy (SEM) images of the resultant silica
were obtained on a PHILIPS XL 30 microscope (Fig. S1).
2
mation of unreactive Pd0 particles. More recently, Rossy and
coworkers attempted to design a system in which the aggregation
of the small Pd nanoparticles into larger crystallites having a lower
surface area and significantly lower catalytic activity was pre-
vented by the stabilization of carbon-supported Pd through the
formation a Pd/Au alloy [22].
2.2. Preparation of nanosized metallic particles dispersed on the
amorphous SiO2
Deionized water (100 mL) was added to the colloidal silica (ob-
tained as above) and the mixture was placed in an ultrasound bath
and sonicated for 20 min. Then, an aqueous saturated solution con-
taining nanometal precursor (Table 1) was added dropwise into a
suspension of colloidal silica and mixed in an ultrasound bath for
20 min. The mixture was dried at 60–90 °C for 12 h, ground, and
sieved. Finally, the reduction was conducted in an oven (FCF 7
SMW system) under hydrogen at 500 °C for 1 h affording nano-
Theoretical studies proved that in fact Cu can strongly modify
the Pd valence state by injecting charge into the sp subband [24].
Thus, electronic modification occurs, in that the Pd–Cu alloys affect
catalytic activity of the system. Significant enhancement of the cat-
alytic activity has been observed, for example, for bimetallic cop-
2
per palladium system in a CO + NO + O reaction [25]. The Pd–Cu
alloys facilitated NO dissociation with respect to Pd by varying
the nature of the NO interaction with Pd surface centers, which
change from almost purely covalent (in the case of the pure metal)
to a mixture with an ionic component. This ionic component in-
creases with the copper content of the alloy and progressively
weakens the N–O bond [24]. Recently Pd–Cu random alloy nano-
particles appeared to efficiently catalyze oxygen reduction. An in-
crease in catalytic activity was observed in this system up to a
peak at 1:1 Pd/Cu ratio [26].
sized metallic particles dispersed on the SiO intermediate carrier.
2
The content of nanoparticles on silica for each type of the compos-
ite was determined using atomic absorption spectrophotometry
(AAS), Varian model AA 1275 series atomic absorption spectropho-
tometer (Table 1). The transmission electron microscopy (TEM)
images of the resultant composites were obtained on a JEOL 2000
FX microscope (Fig. S2).
In this study, we present an innovative method for the prepara-
tion of bimetallic catalysts. To our best knowledge, this method has
not been reported before and it forms an attractive prospect for the
construction of a broad spectrum of complex nanocatalysts. We
applied this facile approach to formulate nanosized palladium
particles distributed on electrolytic copper. Our investigations were
aimed at the evaluation of the properties of such a bimetallic sys-
tem and its application as the catalyst in the Sonogashira reactions.
The role of the copper in the analyzed bimetallic system can be two-
fold – copper can be regarded as both the support for the palladium
nanoparticles and as the co-catalyst. The Sonogashira reaction with
iodobenzene or para-bromobezaldehyde proceeded smoothly and
2.3. Preparation of nanosized palladium particles dispersed on the
target carrier
Nanosized palladium particles dispersed on the intermediate
carrier, i.e., 1% Pd/SiO2 or 5% Pd/SiO2 (0.55 g) and target carrier,
i.e., Cu, CuO, Ni, Ag or Ag O (0.5 g) was added to deionized water
2
(100 mL), placed in an ultrasound bath and stirred for 10 min.
Then, sodium hydroxide solution (40 mL 40% w/w) was added
and the suspension was stirred for 4 h at the temperature 80 °C.
Next, the suspension was cooled to room temperature and left
for 2 h until the suspended solids sedimented. After decantation,
the sediment was washed five times with deionized water. The
resulting preparation of the catalyst was dried at room tempera-
ture to constant mass. The catalyst surface was probed by X-ray
photoelectron spectroscopy (XPS) using PHI 5700 Spectrometer
2
the comparison of the new catalyst with pure nano-Pd on SiO indi-
cated a markedly higher activity of the Pd–Cu system. Our studies of
the preparation of the Pd on the electrolytic Cu support revealed an
unexpected result: a tiny amount of Pd oxide is formed on the cat-
alyst surface. Taking into account the importance of the presence of
aerobic oxygen for the high activity of the Pd-catalyzed Sonogashira
coupling [9], we believe that the coexistence of Pd oxide is essential
for the observed enhancement in the activity of the bimetallic Pd/
Cu system. Although Pd oxide has been previously detected in the
with a AlKa radiation source.
Alternatively, the samples were suspended in ethanol and soni-
cated for 15 min in ultrasound bath and the resulted materials were
deposited on carbon adhesive tape for the preparation of the samples
for TEM analyses. The transmission electron microscopy (TEM)
images of the resultant composites were obtained on a JEOL 2000
FX or high resolution (HRTEM) JEM 3010 microscopes both equipped
with EDS detector. The scanning electron microscopy (SEM) JSM 6480
was used for morphology investigation of composite powders.
highly reactive Pd/Au system [22] or c-alumina supported Pd/Cu
that catalyzed liquid phase hydrogenation [27], this has never been
related to the activity of these systems.
2.4. General procedure for the catalytic cross-coupling
2
. Experimental
In a tightly sealed tube (septa system), aryl halides (5.4 mmol),
the catalyst, PPh and CuI (compare Table 2 and Table 3 for the
3
Bimetallic Pd catalysts were prepared using a novel facile ap-
proach involving the transfer of nanoparticles from the intermedi-
ate carrier, i.e., SiO , to the target carrier. The general method
includes several steps as described in the following subsections.
2
Table 1
Preparation of the composition: Pd/SiO
2
.a
Pd/SiO
2
metal contentc (wt.%)
b
Precursor PdCl
2
2
.1. Preparation of the amorphous SiO
2
as the intermediate carrier
was prepared using
1
0
0
.72
.74
.65
5.0/4.70
2.0/2.09
1.9/1.84
1.0/0.94
0.1/0.1
The intermediate carrier amorphous SiO
2
the sol–gel technique [28]. In a typical procedure, a solution of
methanol (1050 mL) and aqueous ammonia (25 wt.%, 370 mL)
was stirred for 15 min, followed by addition of tetraethyl orthosil-
icate (75.09 mL). The reaction mixture was vigorously stirred for
h at room temperature affording colloidal silica suspension (with
theoretical mass of silica 20.2 g). The resultant colloidal silica sus-
pension was centrifuged. Then, the colloidal silica was washed
0.34
0.035
a
Obtained as described in the paragraph 2.2, acc. to the
method modified from [31].
b
2
Precursor used for a portion of silica (20.2 g) obtained as
described in paragraph 2.1.
c
Designed/actual metal content as found by the AAS method.

M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
3
Table 2
The catalytic performance of different palladium catalysts in the model Sonogashira reaction. Compare Table S1 (Supplementary data) for additional catalyst types.
Catalysta
Yield (%)
b
Temperature (°C)
Time (h)
Pd (mol%)
CuI (mol%)
3
PPh (mol%)
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
1% Pd/CuO
1% Pd/Cu
1% Pd/Cu
1% Pd/Cu
1% Pd/Cu
1% Pd/Cu
1% Pd/Cu
1% Pd/Cu
5% Pd/Cu
1% Pd/Cu
0
0
0
100
100
25
100
80
80
80
80
80
120
90
120
120
120
80
5
5
4
5
1.5
1.5
1.5
1.5
2.5
6
0.035
0.035
0.035
0.035
0.035
0.035
0.035
0.041
0.174
0.035
0.174
0.070
0.035
0.004
0.037
0
0
0
0
0
0
0
0
0.24
0
1.29
0.24
0.38
0.56
1.18
1.44
1.27
1.27
0.71
5.51
0.34
0.24
1.23
50
27
40
55
61
100
100
28
15
4
0
0
1
1
1
1
1
1
5% Pd/SiO
2% Pd/SiO
1% Pd/SiO
2
2
2
5
6
6
6
0.19
0.26
0.24
0.21
1.52
0.1% Pd /SiO
PdCl (PPh
2
3
89
2
3
)
2
4
a
Obtained as described in the paragraph 2.3 or modified acc. to the paragraph 2.3 for the systems other than Pd/Cu. Cu and CuO means – electrolytic and oxidized Cu
supports, respectively.
b
For entries 9, 10: yield of separated product without further purification: mp. 62–63 °C, (lit. mp. 61–63 °C [35]; 66 °C [36]); for other entries the yield calculated on the
basis of 1H NMR (400 MHz, DMSO): d 10.02 (s, 1H), 7.90 (d, J = 8.2 Hz, 2H), 7.67 (d, J = 8.1 Hz, 2H), 0.26 (s, 9H).
exact mixture compositions) were suspended in dry triethylamine
10 mL). The mixture was placed in an ultrasound bath and soni-
cated for 5 min. Then, the acetylene compound (5.6 mmol) was
added and the mixture was stirred as indicated in Table 2 and Ta-
ble 3. The mixture was cooled to room temperature, and the cata-
lyst was centrifuged, filtered, and washed with deionized water
size distribution of nanoparticles and their tendency to aggregate
or to form clumps [29,30]. To minimize these problems, herein
we have demonstrated an innovative and facile approach which
was implemented for the formation of a bimetallic Pd catalyst.
The concept of this newly developed method is based on the nano-
particle transfer from the intermediate carrier to the target carrier.
Amorphous silica synthesized by sol–gel [28] technique was used
as the intermediate carrier. SEM observations (Fig. S1 Supplemen-
tary materials) indicated that silica obtained by the sol–gel method
exhibited a regular spherical shape, a controlled size distribution,
and a uniform porous surface. A final catalyst was synthesized in
the transfer method, where nanosized palladium particles dis-
(
(
2 Â 10 mL) and ethyl acetate (2 Â 10 mL). The filtrate was poured
into 15 mL of dilute hydrochloric acid solution and the aqueous
layer was extracted three times with dichloromethane
(
3 Â 15 mL). The collected organic layer was washed with deion-
ized water (15 mL) and then dried over magnesium sulfate, fil-
tered, and concentrated under reduced pressure, affording the
1
product. The conversion and selectivity were determined by
H
persed on the intermediate carrier, i.e., 1% Pd/SiO
and target carrier, e.g., Cu, were suspended in deionized water,
placed in an ultrasound bath and stirred. Then, SiO was digested.
2 2
or 5% Pd/SiO
NMR spectroscopy. The spectra was recorded on a Bruker Avance
400 MHz or 500 MHz) using deuterated solvents: CDCl and
DMSO with TMS as internal standard.
(
3
2
We assumed that the appropriate digesting solvent should fulfill
the following requirements: it should digest only the intermediate
carrier, i.e., silica, and it must be inert both for the target carrier
and metallic nanoparticles. In our method, an alkali solution, i.e.,
2.5. Testing catalyst leaching
4
0% aqueous NaOH, was suitable as a digesting medium. The lack
The reaction performed as described in the previous paragraph
was stopped after 1.5 h centrifuged and the resulted supernatant
of silica (a tiny amount of ca. 1% can still be detected) in a final cat-
alyst has been proved by XPS and XRD analyses. Efficient sonifica-
tion appeared to be of crucial importance in this procedure. In
Table 1 we specify the composition of the individual synthetic sys-
tems used in the transfer method.
was filtered. Filtrate was reacted for the additional 3.5 h and ana-
lyzed in a standard mode using 1H NMR. The additional analysis
was performed using the ICP-MS (NexION 300 D ICP-MS) to deter-
mine a concentration of Pd.
2
The SEM images of the Pd/SiO precursor indicted the spheroi-
dal shape of the silica particles ranged from ca. 500 nm to 900 nm.
The average content of palladium was 4.2 wt.% as confirmed by
EDS method. HRTEM images of Pd/SiO2 (Fig. 1) indicated only min-
or areas between silica grains where crystalline Pd was formed. In-
stead, the amorphous Pd is present on the surface in the layer of ca.
2 nm (Fig. 1b). The diffraction analyses from these regions proved
the amorphous character of the Pd located here (Fig. 1c). Moreover,
an analysis of TEM image of Pd/SiO (Fig. S2) revealed that the ob-
2
tained average size of the Pd nanoparticles was below 10 nm. This
indicted also that the silica carrier facilitated an even distribution
3
. Results and discussion
3.1. The catalyst preparation and structure
Over the past few years, a number of techniques have been
developed for the production of nanosized metallic particles and
their distribution on different carriers [29,30]. The methods re-
cently used, which are based on ‘‘the bottom-up’’ and ‘‘the top-
down’’ techniques, still suffer some disadvantages, including broad

4
M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
Table 3
The Sonogashira reactions in the absence of CuI salt catalyzed by 5% Pd/Cu under air.
R1
X
R
Pd (mol%)
PPh
(mol%)
Time (h)
Temperature (°C)
Yield (%)
a
3
1
2
3
4
5
6
–H
–H
–NH
–CHO
–CHO
–H
Br
I
I
Br
Br
I
–Si(CH
–Si(CH
–Si(CH
3
3
3
)
)
)
3
3
3
0.183
0.191
0.183
0.174
0.191
0.174
1.34
1.27
1.27
1.27
1.27
1.20
5
2
6
2.5
5
120
80
100
100
100
80
63b
c
100
d
2
30
e
–CH
–Ph
–Ph
2
(OH)
93
f
100
100
g
2
a
b
c
d
e
f
Calculated on the basis of 1H NMR.
1
), d: 7.48–7.50 (m, 2H), 7.30–7.34 (m, 3H), 0.28 (s, 9H); lit 1H NMR [37].
H NMR (500 MHz, CDCl
H NMR (400 MHz, CDCl
H NMR (500 MHz, CDCl
H NMR (500 MHz, CDCl
H NMR (500 MHz, CDCl
3
3
3
3
3
1
1
1
1
1
1
), d: 7.47–7.51 (m, 2H), 7.30–7.33 (m, 3H), 0.28 (s, 9H); lit H NMR [37].
), d: 7.27–7.31 (m, 2H), 6.58–6.61 (m, 2H), 3.76 (s,2H), 0.25 (s,9H); lit 1H NMR [38].
), d: 10.01 (s, 1H), 7.84 (d, J = 8.2 Hz, 2H), 7.58 (d, J = 8.2 Hz, 2H), 4.54 (s, 2H), 2.48 (br s, 1H); lit 1H NMR [39].
), d: 10.04 (s, 1H), 7.89 (d, J = 8.1 Hz, 2H), 7.70 (d, J = 8.2 Hz, 2H), 7.53–7.63 (m, 2H), 7.33–7.43 (m, 3H); lit 1H NMR [40].
g
1
H NMR (500 MHz, DMSO), d: 7.55–7.58 (m, 4H), 7.40–7.46 (m, 6H); lit H NMR [41].
Fig. 1. TEM image of 5% Pd/SiO
c).
2
catalyst (a,b); and selected area diffraction pattern
Fig. 2. SEM image of electrolytic copper (a) and nanopalladium supported on
electrolytic copper – 5% Pd/Cu catalyst (b).
(
of the metallic precursors, thus resulting in a low polidispersity of
the nanostructures obtained. Since agglomeration is one of the ma-
jor problems in the synthesis of stable nanoparticles, we took
In Fig. 2 we presented SEM images of a Cu grain before assem-
bling nano-Pd on its surface and a final bimetallic Pd/Cu agglomer-
ate. The maximal size of the Cu grains was 63
material has a dendrite like morphology in which a size of the indi-
vidual dendrites did not exceed 5 m, and a clear change in the
lm; however; this
2
advantage of the stability of the obtained Pd/SiO for the synthesis
of the more complex systems.
l

M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
5
Fig. 4. X-ray diffraction patterns measured by Empyrean diffractometer with Cu
1,2 radiation for the sample of 5% Pd/SiO (a) and 5% Pd/Cu (b).
Ka
2
Fig. 3. TEM image of 5% Pd/Cu catalyst in bright field (a); dark field TEM image of
% Pd/Cu catalyst registered in the reflexes dhkl = 0.226 nm (b) and selected area
5
diffraction pattern (c).
surface morphology appeared as a result of a Pd transfer to the Cu
surface which is now ornamented with small nanoparticle size pat-
terns (Fig. 2b).
The HRTEM analyses of the bimetallic Pd/Cu samples indicated
a presence of nano-Pd on the Cu surface (Fig. 3a and b). Electron
diffraction pattern from these regions indicated a view typical for
fine crystalline structures (Fig. 3c). The dhkl values can be attributed
to the Pd dhkl, while the single reflexes obtained for dhkllarger than
0
.226 nm comes from CuO. The observed crystallites do not exceed
a size of 5 nm. In some areas, we identified also a small amount of
silica. If identified, this was a background for Pd nanoparticles.
XRD analyses of the initial Pd/SiO2 system (Fig. 4) indicated
weak reflexes of Pd on
a background of amorphous silica
(Fig. 4a), while in a final Pd/Cu bimetallic catalyst the Pd reflexes
were accompanied by the signals of the Cu oxide phases (Fig. 4b).
We used XPS techniques to study the precise composition of
bimetallic catalysts. In particular, the analysis of XPS spectra for
5
% Pd/Cu provided us with useful information about the chemical
state of the catalyst components. The analysis of the Pd 3d doublet
Fig. 5) revealed two chemical states of Pd. Lower binding energy
Fig. 5. XPS analysis of the 5% Pd/Cu catalyst-Pd 3d doublet.
(
doublet was assigned to metallic Pd whereas higher binding en-
ergy one could be related to PdO or to the partial oxidation state
vacuum (UHV). UHV and possibly X-ray exposition resulted in oxy-
gen leakage and subsequently in the reduction of Pd(II) to Pd . In
0
PdO
x
. Such states have been mentioned in the literature [27] where
turn, the analysis of the O 1s line (Fig. 6) revealed two chemical
states, i.e., lower binding energy could be assigned to CuO, whereas
higher binding energy could be assigned to PdO and adsorbants
containing oxygen. Decreasing intensity of the second state corre-
sponded to reduction of PdO in UHV. The shape of the Cu 2p mul-
tiplet (Fig. 6) matched very well with the spectrum reported for
the position of the photoemission line close to 337 (eV) was similar
to that observed in our studies. Another argument for the presence
of PdO on the surface of the metallic nanoparticles was a gradual
x
reduction of the intensity attributed to the oxidized state with re-
spect to the metallic state during XPS measurements in ultra-high

6
M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
We beganour studieswith theexaminationof the catalyticutility
of the different bimetallic catalysts (Table 2) for the cross-coupling
of p-bromobenzaldehyde with trimethylsilylacetylene c as a model
Songashira reaction. The catalysts varied in their metal loadings on
the carrier, e.g., 1% Pd/carrier, 5% Pd/carrier, the palladium concen-
tration in the reaction system, and the type of the carrier metal or
its oxide conjugated with palladium in a bimetallic system (Table 2).
The carrier metals displayed a twofold role as both the palladium
support for the heterogenous catalysis and the co-catalyst to pro-
mote the transmetalation step in the Sonogashira coupling. The
majority of the potential co-catalysts tested, i.e., the metals Ag, Au
2
and Ni or the metallic oxides Ag O or CuO, did not indicate any capa-
2
bility of activating sp–sp coupling (Entry 1 of Table 2 and Table S1 in
Supplementary materials for the detailed list). Among the examined
bimetallic composites, only Pd/Cu compositions were found to be
active catalysts in the model reaction. Therefore, we focused further
investigation on only this bimetallic system.
Fig. 6. XPS analysis of the 5% Pd/Cu catalyst-O 1s signal.
For a given palladium loading on the carrier and the catalyst
concentration in the reaction system, the catalytic performance
of the Pd/Cu system was affected by the reaction conditions, espe-
cially the concentration of phosphine ligands and temperature
(Table 2). Although we did not perform a precise optimization of
the reaction conditions, we observed a general trend in the catalyst
performance. Based on control experiments (Table 2, entries 2 and
3
), we assumed that the coupling reaction did not occur in the ab-
sence of the phosphine ligands and the Pd/Cu system did not dis-
play catalytic activity at room temperature. A support of the Pd
system with the bulky and electron-rich phosphine ligands en-
hanced the formation of low-coordinate and highly active palla-
dium complexes. We noted that the increase in the phosphine
concentration improved the reactivity in the catalytic system (Ta-
ble 2, entries 5–7).
A comparison with pure Pd accompanied by CuI (Table 2, en-
tries 11–14) verified a significantly higher activity of the bimetallic
Pd–Cu system. Generally, the Pd–Cu composite exhibited catalytic
activity fairly comparable to that of the homogenous catalyst
PdCl (PPh ) (Table 2, entry 15) [6,33]. However, the maximal yield
2 3 2
for the Pd/Cu catalyzed systems achieved as much as 100% yield.
The catalytic activity for this system was also ca. 3–30 times higher
Fig. 7. XPS analysis of the 5% Pd/Cu catalyst-Cu 2p multiplet.
CuO [27]. In this case, no effects of UHV and X-rays were observed
in the spectra. Thus, the analysis of XPS spectra revealed the
formation of Pd–O and Cu–O bonds at the surface of the palla-
dium–copper bimetallic catalyst. Notably, the presence of the
expected Pd–Cu alloy was not confirmed by our analyses, in partic-
ular by the XPS analysis. The absence of signals attributed to the
Pd–Cu alloy formation in the Cu 2p spectra could be possibly due
to the low palladium content in the 5% Pd/Cu catalyst, insufficient
for the detection by XPS. It was also possible that Pd–Cu alloy par-
ticles could be completely covered by a thick surface layer of
metallic oxides, thus impeding the Pd–Cu alloy detection by XPS.
Due to limited photoelectron mean free path, the information
obtained from the XPS data is limited to about 3–4 nm from the
surface. (See Fig. 7).
than that for the combination of Pd/SiO
2
and CuI (Table 2, entries
9
–10 vs. entries 11–14). Moreover, in comparison with the Pd/
2
SiO system, all of the Pd/Cu catalysts exhibited much lower palla-
dium loadings (Table 2). Nonetheless, this amount was sufficient to
ensure catalyst reactivity. This is a vital advantage taking into con-
sideration other reported Sonogashira catalytic systems with Pd
loadings usually ranging from 1 mol% to 5 mol% [1,6,33]. Indeed,
a Pd loading as low as 0.035 mol% was sufficient to activate the
Sonogashira reaction in our catalytic system (Table 2, entries: 4–
7
, 10). We believe this value could further optimized in the later
generations of the catalyst.
Another important advantage of the new Pd/Cu catalyst was its
1
3.2. The Sonogashira C–C coupling
selectivity. H NMR spectra indicated a high purity of the products
obtained under the catalytic Pd/Cu system. The model reaction
with the highly effective Pd/Cu system could provide even quanti-
tative conversions and subsequently quantitative yields of the cou-
pling products (Table 2, entries 9–10). The formation of the alkyne
homo-coupling Glaser-type products [6,8,33] was not observed,
though the model Sonogashira reactions were performed under
air. In our Sonogashira protocol, no Cu(I) salt was used, thereby
avoiding undesired formation of homo-coupling products.
According to the original protocol, the Sonogashira coupling
generally occurs in the presence of a substantial amount of palla-
dium catalyst with copper (I) salt as the co-catalyst under an inert
atmosphere [32]. The main limitations of the Sonogashira reaction
were considered to arise from the high cost of Pd catalyst and its
low reusability, the residual contamination of a toxic Pd species
(
a major problem in the synthesis of pharmaceuticals), and the
use of cooper salt sensitive to oxygen which induced a side reac-
tion, i.e., oxidative homocoupling of the alkyne substrates (the so
called Glaser-type reaction) [6,8,33]. Our studies were focused on
the development of a more efficient, less costly, and more simply
formulated catalyst which would be active enough to perform
the Sonogasira reaction under mild conditions.
The new system appeared selective in the case of cross-coupling
with trimethylsilylacetylene (Table 2, entries 4–10 and Table 3, en-
tries 1–3) where the C(sp)–Si bond was not affected and the forma-
tion of a ‘‘sila’’-Sonogashira side-product via the Si–C bond
activation [34] was not observed. The reactions with trimethylsi-
lylacetylene could be regarded as a highly useful synthetic tool,

M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
7
as they allowed attachment of the protecting group, i.e., trimethyl-
silyl group [34] into the corresponding Sonogashira products. This
transformation, in turn, could enable further modifications of the
products.
dissolved in the reaction mixture was at the level of ca. 4% of the to-
tal initial Pd content. Moreover, the catalyst after being centrifuged
and filtrated additionally released a low amount of catalyst, ca. 1%
(total initial Pd), to ethyl acetate used as a washing solvent. Since
these results proved catalyst leaching typical for the Pd quasi-
heterogenous catalysis, we performed an additional experiment
in which the reaction mixture is filtered after partial conversion
of the reactants and the filtrate is monitored to check for further
reaction. Thus, we stopped the reaction (entry 4 in Table 3) after
1.5 h which appeared to provide 23% of conversion. After filtration
of the solid catalyst, the obtained filtrate was reacted for the addi-
tional 3.5 h yielding 67% of the expected coupling product (i.e.,
67–23 = 44%; without heterogenous catalyst), which compares to
93% of the original reaction after 2.5 h (entry 4 in Table 3). This
experiment indicates that catalyst leaching occurs during the inves-
tigated reaction, which is characteristic for the quasi-heterogenous
Pd catalysis. On the other hand, simple mechanical damage to the
catalyst can also be the important reason for the catalyst deactiva-
tion, since the amount of Pd determined in the reaction mixture
was relatively small (ca. 4%) and the measured XPS spectra (not
shown) of the catalyst worn in the reaction indicated that no Pd
was observed on the Cu surface after the third cycle. Thus, we also
checked the composition of the catalyst after the first use, perform-
ing the XPS test at the same conditions as for the fresh sample
(Fig. S3). The relative Pd/Cu ratio on the surface was considerably
decreased to about 0.7% while in the fresh sample it was about
12.5%. The analysis of the Pd most pronounced photoemission line
indicates the presence of two chemical states with the dominant
contribution from the oxidized state. At the same time the chemical
state of Cu is reduced. The position of the lines is characteristic for
In further studies, we investigated the performance of the Pd/Cu
system used to catalyze the reactions with different aryl halides and
various terminal acetylenes (Table 3). The substrates ubstituted
with both electron-donating groups and electron-withdrawing
groups were used to probe the Pd/Cu catalytic potential. Notably,
the studies proved that the characteristics of the aryl halides in
the tested catalytic system exerted a considerable influence on
the reaction rates and yields. The results were consistent with the
well-known trend that iodobenzenes (Table 3, entries 2 and 6)
exhibited higher Sonogashira reactivity [6,8] in comparison with
the reactivity of bromobenzene (Table 3, entry 1 vs. entry 2). In
general, electron-deficient aryl halides favor this reaction, while
electron-rich aryl halides are less prone to the process. Indeed,
the results proved that the aryl bromide substituted with an
electron-withdrawing group (R = CHO Table 2, entry 9 and Table 3,
entry 4 and 5) displayed higher reactivity, while electron-donating
2
groups, i.e., –NH (Table 3, entry 3), disfavored the reaction. In sum-
mary, the studies confirmed that terminal acetylenes bearing an
electron-donating group afforded excellent yields of the desired
products at 100 °C within 2–5 h.
In the experiments discussed above (Tables 2 and 3), triethyl-
amine was used both as a solvent and a base. In a series of exper-
iments, we preserved triethylamine as a base while attempting to
replace the large excess of this reagent with other polar solvents of
the aprotic type to investigate their effects on the reactivity of the
new system (Table 4). We observed quantitative conversions and
the yields of products of the reaction, indicating the compatibility
of such solvents as acetonitrile and ethyl acetate (Table 4, entries 2
and 3), while THF (Table 4, entry 1) significantly decreased the
yield.
In a series of additional experiments, we tested the catalytic sta-
bility of the 5% Pd/Cu system in the cross-coupling Sonogashira
reaction of iodobenzene with trimethylsilylacetylene. The catalyst
separation was facilitated via centrifuge-assistance. A facile recy-
cling method involved the catalyst filtration, washing, and drying.
In addition, a fixed amount of phosphine ligand was added for each
round of the cycle to enhance the stability of the palladium species
and to prevent catalyst deactivation. No noticeable reduction of
activity was observed for two consecutive cycles (100% conversion).
However, a fairly high deactivation of Pd catalyst was observed for
the third cycle (5% conversion). The reused catalyst was extensively
studied using XPS analyses, while the solvents were analyzed by
ICP-MS. ICP-MS indicated that a level of the total amount of Pd
2
metallic Cu although we cannot rule out the presence of Cu O.
Another feature of the XPS spectra after the first catalyst use is
the presence of high amount of carbon. Additional elements like
bromine and silicon were detected as well. These results indicated
that a substantial amount of Pd was removed from the Cu surface
during the first reaction cycle. The presence of C, Br, and Si indicates
the remains after the performed coupling reaction. Apparently, the
dominant co-occurrence of oxidized PdO species and reduced form
of Cu may indicate the oxidative-reductive influence into the Sono-
gashira reaction of Pd and Cu, respectively.
Thus, a presence of PdO within the new bimetallic catalyst can
be of the crucial importance for a possible explanation for the high
activity of the Pd/PdO/Cu catalytic system in the Sonogashira reac-
tion. Numerous recent investigations have indicated that in cou-
0
pling reactions, the catalytic active state is a Pd species even
2
+
0
where Pd is used, with in situ reduction to Pd taking place dur-
ing the reaction. Moreover, more recently, the importance of the
presence of oxygen in the form of aerobic conditions has been re-
ported to promote the efficiency of the Pd nano-system [9].
Table 4
Effect of different solvent on the Sonogashira reaction catalyzed by 1% Pd/Cu under
air.
According to this mechanism, O
2
in the reaction medium prevents
0
the formation of less reactive Pd nanoparticles by shifting the
2
+
equilibrium toward Pd . The equilibrium seems to be highly
2
+
shifted toward the Pd state as a result of the strong electrostatic
2
+
stabilization of Pd ions by the negatively charged zeolite surface
9]. We adapted this scheme to describe an analogous mechanism
[
explaining the reactivity of Pd/PdO/Cu system (Scheme 1 Supple-
mentary material). Apparently, PdO preformed on the catalyst pro-
2
+
vides a template stabilizing Pd . Thus, the coexistence of a Pd/PdO
species in our catalyst can be advantageous for the activity of the
Pd/Cu system in the Sonogashira reaction.
Solvent
PPh
3
(mol%)
Time (h)
Yielda,b (%)
1
2
3
THF
Ethyl acetate
Acetonitrile
1.604
1.708
1.195
1.5
2.5
2
7
100
100
4
. Conclusions
a
Calculated on the basis of 1H NMR.
TON/TOF (h ): entry 1 (157/105); entry 2 (2234/894); entry 3 (2482/1241).
In summary, a copper-supported nano-Pd catalyst obtained by
b
À1
an innovative method of nanoparticle transfer from the SiO
2

8
M. Korzec et al. / Journal of Catalysis 313 (2014) 1–8
[7] D. Astruc, Inorg. Chem. 46 (2007) 1884.
intermediate carrier to the target Cu carrier appeared to be a highly
efficient and selective catalyst in a series of Sonogashira reactions.
We observed up to quantitative conversions. Ultrasonication was
used to prevent nanoparticle aggregation into larger forms having
a lower surface area and subsequently a lower catalytic activity.
[
[
8] H. Doucet, J.C. Hierso, Angew. Chem. Int. Ed. 46 (2007) 834.
9] M. Choi, D.H. Lee, K. Na, B.W. Yu, R. Ryoo, Angew. Chem. Int. Ed. 48 (2009)
3673.
[10] L. Djakovitch, K. Koehler, J. Am. Chem. Soc. 123 (2001) 5990.
[
[
11] K. Shimizu, S. Koizumi, T. Hatamachi, H. Yoshida, S. Komaid, T. Kodama, Y.
Kitayama, J. Catal. 228 (2004) 141.
12] J.G. de Vries, Dalton Trans. 3 (2006) 421.
2
Comparisons with other nano-Pd systems, especially Pd/SiO , indi-
cated a substantial advantage of the novel catalytic system. X-ray
photoelectron spectroscopy (XPS) studies of the prepared Pd/Cu
catalyst revealed the development of a PdO species in the Pd/Cu
system and we have shown for the first time the importance of
the resulting Pd/PdO/Cu system for an efficient Sonogashira cou-
pling. Up to two cycles of full catalyst performance (100% conver-
sion) can be attained, and a fairly high deactivation in a third cycle
[13] M.T. Reetz, J.G. de Vries, Chem. Commun. 14 (2004) 1559.
[
[
[
14] M.T. Reetz, E. Westermann, Angew. Chem. 112 (2000) 170.
15] M.T. Reetz, E. Westermann, Angew. Chem. Int. Ed. 39 (2000) 165.
16] A.F. Schmidt, V.V. Smirnov, J. Mol. Catal. A: Chem. 203 (2003) 75.
[17] A.F. Schmidt, V.V. Smirnov, Top. Catal. 32 (2005) 71.
[
[
18] V. Polshettiwar, A. Molnar, Tetrahedron 63 (2007) 6949.
19] R. Bernini, S. Cacchi, G. Fabrizi, G. Forte, F. Petrucci, A. Prastaro, S. Niembro, A.
Shafir, A. Vallribera, Green Chem. 12 (2010) 150.
[20] B. Tamami, S. Ghasemi, J. Mol. Catal. A: Chem. 322 (2010) 98.
[
[
21] J. Cookson, Platinum Met. Rev. 56 (2012) 83.
22] C. Rossy, J. Majimel, E. Fouquet, C. Delacôte, M. Boujtita, C. Labrugère, M.
Tréguer-Delapierre, F.X. Felpin, Chem. Eur. J. 19 (2013) 14024.
(
5% conversion). The performed experiments controlled by ICP-MS
indicated catalyst leaching on a level of ca. 4%. However, XPS spec-
tra of the worn catalyst indicated no Pd on the Cu surface after the
third cycle which may results from the mechanical destruction.
This provides a basis for the design of future generations of cata-
lysts that are not hampered by the disadvantage of the previous
nano-Pd systems where Pd agglomeration has been observed.
[23] S. Mukhopadhyay, G. Rothenberg, A. Joshi, M. Baidossi, Y. Sasson, Adv. Synth.
Catal. 344 (2002) 348.
[
24] F. Illas, N. López, J.M. Ricart, A. Clotet, J.C. Conesa, M. Fernández-García, J. Phys.
Chem. B 102 (1998) 8017.
[
25] M. Fernández-García, A. Martínez-Arias, C. Belver, J.A. Anderson, J.C. Conesa, J.
Soria, J. Catal. 190 (2000) 387.
[
[
26] W. Tang, L. Zhang, G. Henkelman, J. Phys. Chem. Lett. 2 (2011) 1328.
27] J. Batista, A. Pintar, D. Mandrino, M. Jenko, V. Martin, Appl. Catal. A:Gen. 206
Acknowledgments
(
2001) 113.
[
[
28] H. Okudera, A. Hozumi, Thin Solid Films 434 (2003) 62.
29] H. Skaff, T. Emrick, in: V. Rotello (Ed.), Nanoparticles: Building Blocks for
Nanotechnology, Springer Science and Business Media Inc, New York, 2004, p.
This study was supported by the Grant ORGANOMET No: PBS2/
A5/40/2014 from the Polish National Research and Development
Center. PZ appreciates the support of the FORSZT and PB, MK,
and MK to the Doktoris fellowships.
3
2.
[30] J. Kao, K. Thorkelsson, P. Bai, B.J. Rancatore, T. Xu, Chem. Soc. Rev. 42 (2013)
654.
[
[
2
31] Y. Wu, Y. Zhang, L. Zhang, China Particuol. 2 (2004) 19.
32] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 50 (1975) 4467.
Appendix A. Supplementary material
[33] K. Sonogashira, J. Organomet. Chem. 653 (2002) 46.
[
[
34] Y. Nishihara, E. Inoue, D. Ogawa, Y. Okada, S. Noyori, K. Takagi, Tetrahedron
Lett. 50 (2009) 4643.
35] C.B. Aakeröy, A.S. Sinha, K.N. Epa, C.L. Spartz, J. Desper, Chem. Commun. 48
(2012) 11289.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2014.02.008.
[
[
[
36] S. Thorand, N. Krause, J. Org. Chem. 63 (1998) 8551.
37] X.F. Wu, H. Neumann, M. Beller, Chem. Commun. 47 (2011) 7959.
38] J.J. Dai, C. Fang, B. Xiao, J. Yi, J. Xu, Z.J. Liu, X. Lu, L. Liu, Y. Fu, J. Am. Chem. Soc
References
1
35 (2013) 8436–8439.
[
[
[
[
[
1] R. Chinchilla, C. Nájera, Chem. Soc. Rev. 40 (2011) 5084.
2] K.C. Nicolaou, P.G. Bulger, D. Sarlah, Angew. Chem. Int. Ed. 44 (2005) 4442.
3] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.
4] A. Suzuki, Angew. Chem. Int. Ed. 50 (2011) 6722.
5] C.C. Johansson Seechurn, M.O. Kitching, T.J. Colacot, V. Snieckus, Angew. Chem.
Int. Ed 51 (2012) 5062.
[39] E. Ko, J. Liu, L.M. Perez, G. Lu, A. Schaefer, K. Burbess, J. Am. Chem. Soc. 133
(2011) 462.
[40] J.M. Englert, J. Malig, V.A. Zamolo, A. Hirsch, N. Jux, Chem. Commun. 49 (2013)
4827.
[41] A. Komáromia, Z. Novák, Chem. Commun. 40 (2008) 4968.
[
6] R. Chinchilla, C. Nájera, Chem. Rev. 107 (2007) 874.