Pd0.09Ce0.91O2-Δ: A sustainable ionic solid-solution precatalyst for heterogeneous, ligand free Heck coupling reactions

Article abstract of DOI:10.1016/j.mcat.2017.09.022

A quick and easy method for the preparation of Pd²⁺ metal ion substituted in ceria, Pd_0.09Ce_0.91O_2-δ solid solution oxide, is described. The Pd_0.09Ce_0.91O_2-δ solid solution oxide was fully characterized by XRD, ICP-OES, BET, XPS, SEM, EDX, TEM, TGA and Raman spectroscopy. All characterization techniques strongly suggested that Pd²⁺ was successfully incorporated into the lattice structure of ceria. The effect of the reaction conditions on the catalytic properties of the Pd_0.09Ce_0.91O_2-δ solid solution catalyst initially was studied in detail with the model Heck reaction of iodobenzene and methylacrylate to obtain optimum reaction conditions. The Pd_0.09Ce_0.91O_2-δ solid solution catalyst then afforded substituted alkenes in good to excellent yields under these optimum reaction conditions. Steric and electronic effects were also studied, and were found to influence the catalytic activity. Characterization of the used catalyst suggests that Pd²⁺ in Pd_0.09Ce_0.91O_2-δ is reduced in situ to Pd⁰ when employed in the Heck cross-coupling reactions. The catalyst was easily recovered by centrifuge and reused three times without significant loss of catalytic efficiency.

Full text of DOI:10.1016/j.mcat.2017.09.022

Molecular Catalysis 443 (2017) 60–68
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Research Paper
Pd0.09Ce0.91
O
2-␦: A sustainable ionic solid-solution precatalyst for
heterogeneous, ligand free Heck coupling reactions
Philani P. Mpungose, Neo I. Sehloko, Venkata D.B.C. Dasireddy,
Narayanappa Mahadevaiah, Glenn E. Maguire, Holger B. Friedrich∗
School of Chemistry and Physics, University of KwaZulu-Natal, Private Bag X54001, Durban 4001, South Africa
a r t i c l e i n f o
a b s t r a c t
A quick and easy method for the preparation of Pd2+ metal ion substituted in ceria, Pd0.09Ce0.91O2-␦ solid
solution oxide, is described. The Pd0.09Ce0.91O2-␦ solid solution oxide was fully characterized by XRD, ICP-
OES, BET, XPS, SEM, EDX, TEM, TGA and Raman spectroscopy. All characterization techniques strongly
suggested that Pd2+ was successfully incorporated into the lattice structure of ceria. The effect of the
reaction conditions on the catalytic properties of the Pd0.09Ce0.91O2-␦ solid solution catalyst initially was
studied in detail with the model Heck reaction of iodobenzene and methylacrylate to obtain optimum
Article history:
Received 2 May 2017
Received in revised form
18 September 2017
Accepted 20 September 2017
Keywords:
reaction conditions. The Pd0.09Ce0.91O2-␦ solid solution catalyst then afforded substituted alkenes in good
Pd2+ substituted ceria
Solution combustion
Nanoparticles
to excellent yields under these optimum reaction conditions. Steric and electronic effects were also
studied, and were found to influence the catalytic activity. Characterization of the used catalyst suggests
that Pd2+ in Pd0.09Ce0.91 2-␦ is reduced in situ to Pd0 when employed in the Heck cross-coupling reactions.
O
Heck coupling reaction
The catalyst was easily recovered by centrifuge and reused three times without significant loss of catalytic
efficiency.
© 2017 Elsevier B.V. All rights reserved.
1. Introduction
oxygen sensitive (easily oxidized) [17]. Furthermore, the efficient
separation and subsequent recycling of homogenous palladium
Palladium catalysed Heck coupling reactions are a well-known
and a well-established technique in organic chemistry and the
pharmaceutical industry [1]. The Heck coupling reaction is now
considered to be one of the most useful and practical tools in
modern organic synthesis [2]. It involves the construction of a
carbon-carbon (C C) bond through a chemical reaction between
an unsaturated halide and an alkene in the presence of a base and
catalysts and ligand remains a challenge and an aspect of economic
relevance [19,20].
As a result, the use of ligand-free heterogeneous palladium cat-
alyst is often desirable because it permits easy handling, simple
recovery and recyclability of the palladium catalyst. The increasing
number of reports indicates that there is a growing interest in find-
ing heterogeneous variants of Heck coupling reactions [6,7,20–26].
Most classes of heterogeneous catalysts described in literature
involve immobilization of homogeneous Pd containing catalysts
onto insoluble polymeric supports or solid inorganic supports. Fre-
quently used supports are activated carbon [7], molecular sieves
[27], metal oxides [8–10,28,29] and polymers [30,31].
Heterogeneous catalysis with the noble metal ion substituted in
solid-state materials, such as in simple and complex oxides, is not
well understood [32–34]. However, some authors of the present
work have published some papers addressing the synthesis and cat-
alytic behavior of binary oxide-based compounds substituted with
noble metal ions to make ionic catalysts [35–37]. Hedge and co-
workers have also published numerous papers addressing solution
combustion synthesis and application of noble metal ionic cata-
lysts [38–50]. All reports have shown that substitution of Pd2+ ions
within the CeO2 lattice allows for complete palladium dispersion,
a palladium catalyst to form substituted alkenes [3,4]. These C
C
bonds are of great importance in organic chemistry and are essen-
tial for all life on earth [5–10]. Several reviews have been published
[11–16].
The most common catalytic systems used for Heck coupling
reactions involve homogenous palladium-phosphine complexes,
such as PdCl2(PPh3)2 or Pd(PPh3)4, in the presence of a base
[11–18]. Many ligands and catalysts have also been reported, but
they are largely not available commercially and do not find exten-
sive use [19,20]. This is mostly because the phosphine ligands are
∗
Corresponding author.
E-mail addresses: friedric@ukzn.ac.za, holgerbfriedrich@gmail.com
(H.B. Friedrich).
http://dx.doi.org/10.1016/j.mcat.2017.09.022
2468-8231/© 2017 Elsevier B.V. All rights reserved.

P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
61
which, as a result, increases the activity of the ceria-based catalysts
[38,51].
However, the 2 mol% level of Pd substitution usually inves-
tigated for these materials is at the detection limit of phase
identification by conventional powder X-ray diffraction [51]. Thus,
in most cases it is unclear whether the Pd is incorporated into the
lattice structure of CeO2 or is deposited on the surface of CeO2
as PdO [51]. In this regard, we thought it would be worthwhile
to prepare, and evaluate the phase composition and physico-
chemical properties of, a higher Pd-substituted PdxCe1-xO2-␦ (9
mol% Pd) solid solution oxide by solution combustion. This ionic
Pd0.09Ce0.91O2-␦ solid solution oxide was then tested on Heck cross
coupling reactions. Using an air, moisture and thermally stable
heterogeneous Pd0.09Ce0.91O2-␦ solid solution catalyst should offer
simplicity of workup, recyclability and minimization of metallic
waste. It should also offer higher turnover numbers since com-
plete dispersion is achieved in similar Pd2+ ion substituted CeO2
materials [38].
Fig. 1. Rietveld refined XRD pattern of Pd0.09Ce0.91O2-␦ and its STEM-EDX image
(insert).
2. Experimental
2.1. Catalyst synthesis
with N2 as probe gas. About 0.4 g of each powder sample was
degased overnight at 200 ◦C using a Micromeritics FlowPep 060
instrument prior to analysis.
SEM images and EDX data were obtained with a Jeol JSM-6100
scanning microscope using a Bruker signal processing unit detector.
The analysis was performed at random points along the surface of
the catalyst. The samples were first mounted on aluminium stubs
using double-sided carbon tape; they were then coated with gold
using a Polaron E5100 coating unit.
For TEM analysis, the samples were viewed on Joel JEM-
1010 Electron Microscope. For high resolution TEM (HR-TEM) and
scanning electron microscopy (STEM) analysis, the samples were
viewed on Joel JEM-2100 Electron Microscope and the images cap-
tured were analysed using iTEM software. The powder samples
were ultrasonically dispersed in ethanol and supported on a perfo-
rated carbon film mounted on a copper grid prior to analysis.
TGA analysis was conducted with a TA SDT Q600 instrument
under nitrogen flowing at 50 mL/min and at a temperature ramp
rate of 10 ◦C/min from room temperature up to 1000◦ C with ca.
10 mg of sample.
Cerium ammonium nitrate [(NH3)Ce(NO3)6, 99.9%], palladium
chloride [PdCl2, anhydrous, 60% Pd basis], urea [CH4N2O, 99.9%]
were obtained from Sigma-Aldrich and were used without further
purification.
2.1.1. Solution combustion synthesis of the Pd0.09Ce0.91O2-ı oxide
The following materials were synthesized using an adaptation
of a reported procedure [36]. For the synthesis of Pd0.09Ce0.91O2-␦
(6.7 wt% Pd), 2 mmol of PdCl2 were dissolved in 10 mL of HCl (32%)
in a borosilicate dish, and once all the PdCl2 had dissolved, 40 mL of
deionized water, 18 mmol of (NH3)Ce(NO3)6, and 72 mmol CH4N2O
(urea) were added. The solution was stirred at 100 ◦C to evapo-
rate water and form a gel. The formed gel was introduced into a
muffle furnace pre-heated to 120 ◦C, the furnace temperature was
then increased gradually to 600 ◦C over 30 min, and maintained at
that temperature for 5 h. A dark brown solid product was obtained.
For the synthesis of blank CeO2, the above procedure was followed
without the addition of PdCl2.
2.2. Catalyst characterization
2.3. General procedure for Heck cross-coupling reactions
A Bruker D8 Advance diffractometer, equipped with an XRK900
in situ cell and a Cu K␣ source (␭ = 1.5406 Å) was used to record the
powder X-ray diffraction patterns of the samples. The structures
were refined by the Rietveld method using the Full Prof Suite-
2000 program. The average crystallite size (D) and lattice strain
A dry two-necked pear-shaped flask, equipped with a condenser
containing a stirrer bar and 3 mL of DMF was charged with the
aryl halide (2 mmol), olefin (1.5 eq.), triethylamine (1.5 eq.) and
catalyst Pd0.09Ce0.91O2-␦ (0.3 mol% Pd). The reaction mixture was
heated and stirred at 130 ◦C. A GC/MS (PerkinElmer Inc.–Clarus
560 S GC/MS; PONA column: 50 m × 0.15 mm) was used for mon-
itoring the progress of the reaction and for product identification.
After the reaction had gone to completion, the reaction mixture was
cooled to room temperature and filtered. The filtrate was extracted
with ethyl acetate, hexane and brine (1:3:3). The organic layer was
dried with sodium sulphate and the solvent was then evaporated
under reduced pressure. The residue was finally purified by flash
chromatography on silica gel using ethyl acetate as the eluent.
(␧) of CeO2 and Pd0.09Ce0.91O2-␦ were estimated from the modified
Rietveld method and Williamson-Hall (W-H) plots (1).
Kꢀ
D
␤
hklcos␪ =
+ 4␧sin␪
(1)
X-ray photoelectron spectra (XPS) were recorded with a Thermo
Scientific Multilab 2000 spectrometer equipped with the Al K␣
radiation source (1486.6 eV). All the binding energies were refer-
enced to the C(1s) peak (284.5 eV).
ICP-OES was performed using a Perkin Elmer Optical Emission
Spectrometer Optima 5300 DV. The standards (1000 ppm Ce and
Pd) were purchased from Fluka.
3. Results and discussion
Raman spectroscopy was carried out using an Advanced 532
series spectrometer (NIR Spectrometer) utilizing Nuspec software,
equipped with a visible laser of 514 nm.
3.1. Physicochemical properties investigation
The microstructural parameters of the Pd0.09Ce0.91O2-␦ sample
were determined using Rietveld refinement; the experimental, cal-
culated and difference curves were plotted and are shown in Fig. 1.
Brunauer−Emmett−Teller (BET) surface area measurements
were determined using a MicroMetrics TriStar 3000 porosimeter

62
P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
Fig. 2. Core level XPS of (A) Ce (3d) and (B) Pd (3d) in Pd0.09Ce0.91O2-␦ catalyst.
35000
30000
5000
0000
5000
10000
5000
0
A
B
100
50
0
-50
-100
-5000
200
300
400
500
600
700
200
300
400
500
600
700
(cm-1)
Fig. 3. Raman spectra of (a) CeO2 and (b) Pd0.09Ce0.91O2-␦.
Table 1
Structural parameters of the investigated materials obtained from the XRD profiles.
Rietveld refinement
Williamson-Hall method
TEM
Lattice parameter a (Å)
Lattice strain (×10−3
)
Crystallite size (nm)
Average particle size (nm)
CeO2
Pd0.09Ce0.91O2-␦
5.411(9)
5.424(9)
0.9
3
19
17
22
19
The observed peaks in the X-ray patterns of Pd0.09Ce0.91O2-␦ cor-
respond only to the ceria phase with the fluorite structure (JCPDS
34-0394). The STEM-EDX (Fig. 1, insert) confirmed the presence of
Pd and also shows that palladium is homogeneously dispersed in
the ceria lattice.
CeO2 sample are enclosed in the electronic supplementary infor-
mation (ESI).
The Williamson-Hall plot was used to estimate the average
crystallite size and lattice strain of the samples (Table 1). The
Pd0.09Ce0.91O2-␦ sample showed about three times more lattice
strain than pure ceria prepared by the same route (Table 1). The
increased lattice strain in the Pd0.09Ce0.91O2-␦ sample indicates
that lattice distortion occurs upon introduction of Pd2+ ions in the
CeO2 lattice [37,51]. The average crystallite size decreases slightly
with the incorporation of the Pd2+ ions into the CeO2 lattice. The
average crystallite size of the blank CeO2 is 19 nm, while that of
Pd0.09Ce0.91O2-␦ decreased to 17 nm.
An assessment of recent reports on PdxCe1-xO2-␦ solid solutions
suggests that there is no conventional opinion to explain why the
lattice parameter increases with the increasing concentration of
Pd2+ ions [51]. We are of a view that the lattice parameters increase
due to the formation of larger Ce3+ (1.23 Å) ions and the creation
of more oxygen vacancies upon introduction of Pd2+ ions into the
The absence of the Pd0 and/or PdO phase in the X-ray pattern of
the Pd0.09Ce0.91O2-␦ sample suggests that the Pd2+ ions were suc-
cessfully incorporated into the ceria lattice. The incorporation of the
Pd2+ into the ceria lattice came with the expected physicochemi-
cal changes to ceria. The peak for the [111] plane shifts to lower 2
theta values upon the introduction of Pd2+ into the ceria lattice. The
2 theta value of the [111] plane shifts from 28.86◦ in pure CeO2 to
28.83◦ in the Pd0.09Ce0.91O2-␦ sample. This shift is associated with
increases in lattice parameters that cause the ceria lattice to be
strained [51]. Table 1 shows that the lattice parameter a increases
from 5.411(9) Å in CeO2 to 5.424(9) Å in Pd0.09Ce0.91
O2-␦. The XRD,
XPS, ICP-OES, BET, SEM-EDS, TEM, TGA and Raman analyses for the

P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
63
ceria lattice. The increase in the lattice parameters then causes uni-
form lattice strain and a decrease in crystallite size. The formation
of Ce3+ ions, that are larger than Ce4+ (0.97 Å) ions, was confirmed
by XPS results (Fig. 2). The oxidation states of Pd and Ce in the
Pd0.09Ce0.91O2-␦ catalyst were determined by XPS and core level
XPS of Pd (3d) of Pd0.09Ce0.91O2-␦ as shown in Fig. 2. The Pd (3d)
doublet lines were analysed to obtain the electronic state of palla-
dium and the binding energy for Pd (3d5/2) was found to be equal
to 337.5 eV, which is typical for Pd2+ in the catalyst [37,51,52]. This
main state of palladium was attributed to individual palladium ions
localized in the ceria lattice of the Pd0.09Ce0.91O2-␦ solid solution
oxide [51,53]. The elevated binding energy of the satellite peak at
342.5 eV further suggests that the palladium ions are localized in
the ceria lattice.
Fig. 4. TGA and DTA analysis of the investigated materials; (black) CeO2 and (red)
Pd0.09Ce0.91O2-␦. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
Hence, the analysis of the Pd (3d) doublet lines strongly suggests
that the environment of Pd2+ ions in the Pd0.09Ce0.91
O2-␦ solid solu-
tion oxide is different from the oxygen environment in PdO, since
the Pd (3d5/2) line for PdO oxide is in the range of 336.7–337.0 eV
[51,54]. The Ce4+ (3d5/2) peak obtained at 881.5 eV is a characteris-
tic of Ce4+ in CeO2. Thus, partial filling of the valley between 881 eV
and its satellite at 886 eV shows that Ce is in a mixed valence state
(+4 and +3) since the Ce3+ (3d) in Ce2O3, is characterized by Ce3+
(3d5/2) at 885.1 eV [52]. The ratios of oxidation states present in
the catalyst were calculated as Ce4+:Ce3+ of 4:1. Core level XPS of
Ce(3d) in CeO2 is shown in the Electronic supplementary informa-
tion (ESI Figure SI2). The Ce4+(3d3/2) peak obtained at 882.5 eV is
characteristic of Ce4+ in CeO2.
the CeO2 and Pd0.09Ce0.91O2-␦ samples are shown in Fig. 4. The
two samples show an initial weight loss is the temperature range
of 30–150 ◦C. Powdered CeO2 is known to be slightly hydroscopic
and it also absorbs small amount of carbon dioxide from the atmo-
sphere. Thus, for the ceria samples, the weight loss observed below
150 ◦C can be attributed to both water and carbon dioxide loss. Fur-
thermore, no weight loss was observed from 150 to 1000 ◦C in the
ceria sample. For the Pd0.09Ce0.91O2-␦ sample, a second weight loss
was observed at 760 ◦C (DTA) due to the thermal dissociation of
the Pd-O bond to form metallic palladium [34,55]. Furthermore,
the smooth nature of the weight loss change suggests that there
are no intermediates formed before complete reduction to metallic
palladium, implying that the reduction to the metallic state occurs
in one step.
Raman spectroscopy was employed to comparatively analyze
the CeO2 and Pd0.09Ce0.91O2-␦ samples and the results are shown in
Fig. 3. The peak at 465–480 cm−1 corresponds to the F2g vibrational
mode of oxygen ions in the [CeO8] cubic subunit of the CeO2 struc-
ture [37,51]. In blank CeO2 this peak appears at 476 cm−1, while
for Pd0.09Ce0.91O2-␦ it shifts to a lower wavenumber of 466 cm−1
(Fig. 3). The bands at ca. 550 and 650 cm−1, which are weak for
blank CeO2 and more intense for the Pd0.09Ce0.91O2-␦ sample, are
assigned to the defect-induced vibrational mode (D-mode) [37,51].
The elemental composition of the samples was analysed using
EDS. The EDS spectrum of CeO2 shows the presence of three ele-
ments; cerium, oxygen and chlorine (from HCl), while the EDS
spectrum of the Pd0.09Ce0.91
ladium in addition to the other three elements (Fig. 5). The SEM
images of CeO2 and Pd0.09Ce0.91 2-␦ samples have very similar sur-
O2-␦ sample shows the presence of pal-
The absence of a PdO peak (≈660 cm−1) in the Pd0.09Ce0.91
O2-␦ sam-
ple further suggests that the Pd2+ ions are fully dispersed in the
lattice of CeO2 and not deposited as PdO on its surface (ESI Figure
SI5).
O
face morphology (Fig. 5). In both samples, the particles are present
in the form of different sized lumps or flakes with round-shaped
structures. These types of structures are generated due to the nature
of the solution combustion synthesis method, where gases escape
The Thermogravimetric analysis (TGA) (Fig. 4, solid lines) and
differential thermal analysis (DTA) curves (Fig. 4, dotted lines) of
Fig. 5. SEM/EDX of Pd0.09Ce0.91O2-␦.

64
P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
Fig. 6. (A) TEM (B) HR-TEM images of Pd0.09Ce0.91O2-␦.
Table 2
Table 3
Optimisation of reaction conditions for Heck coupling reactions (the bold values are
the chosen values).
Investigation of electronic effects and functional group tolerances of functionalised
olefins.
Reaction parameters
Reaction temperature
Solvent
25, 80 and 130 ◦
C
Anisole, benzonitrile, DMF, DMSO, toluene,
water and xylene
Base
K2CO3 and (CH3CH3)3N
1, 1.5 and 2
0.05, 0.1, 0.3, 0.5 and 1 mol%
Entry
Alkene
Reaction
time (h)
Isolated
yield
(trans) (%)
TON
Olefin equivalence
Catalyst loading (based
on Pd)
1
2
3
1
90
95
75
300
317
250
giving rise to porosity. The BET surface area of the Pd0.09Ce0.91O2-␦
sample was found to be 19 m2/g.
Fig. 6A shows the TEM image of the Pd0.09Ce0.91O2-␦ sample with
average particle size 19 nm. These particles are spherical in shape.
No particles associated with PdO or Pd nanoparticles were observed
1
24
in the Pd0.09Ce0.91O2-␦ TEM images. Qiao et al. and other researchers
4
5
6
1
1
2
96
79
78
320
263
260
have reported Pd particles as cubic and rectangular shaped particles
[56–59]. Hence, the absence of these cubic and rectangular shaped
particles in the Pd0.09Ce0.91O2-␦ sample further suggests that Pd2+
ions are substituted into the lattice structure of CeO2, and are not
supported on the surface of CeO2 as PdO.
The HR-TEM image of the solution combustion synthesized
Pd0.09Ce0.91O2-␦ sample show lattice fringes of 0.30 nm (Fig. 6B).
There were no lattice fringes associated with PdO or Pd nanopar-
ticles observed [60]. Again, the absence of PdO lattice fringes and
the presence of peaks due to Pd in the EDS spectrum (Fig. 5) of
7
8
9
2
89
90
92
297
300
307
24
24
the Pd0.09Ce0.91O
2-␦ sample further indicates that the Pd2+ ions are
completely substituted into the CeO2 lattice.
3.2. Catalyst activity investigation
10
1
90
300
To establish efficient Heck coupling conditions for the
Pd0.09Ce0.91O2-␦ catalyst, methylacrylate and iodobenzene were
used as model coupling partners. With this catalyst, neither cop-
per salts nor activating ligands were necessary. In fact, addition
of triphenylphosphine inhibited the reactions (ESI Table SI5). The
effect of each reaction parameter was then investigated and the
optimum reaction conditions are highlighted in Table 2. Excel-
lent yields were only obtained with polar aprotic solvents (DMF,
benzonitrile and DMSO) and DMF was chosen as the solvent in fur-
ther studies since it allowed easier product isolation. The reactions
were more rapid at 130 ◦C and 1.5 olefin equivalence was found
to be the optimum olefin loading. Potassium carbonate and tri-
ethylamine gave comparable yields, however, triethylamine was
chosen as the base in further studies because it simplifies catalyst
recovery. K2CO3 requires successive washes with water to remove
it from the recovered catalyst. It was also found that for the model
coupling partners, complete iodobenzene conversion was achieved
in two hours with catalyst loading as low as 0.05 mol% (based on
Pd). However, a 0.3 mol% loading was chosen for further studies to
allow shorter reaction times for less reactive coupling partners. Fur-
ther catalyst testing was carried out under the reaction conditions
highlighted in bold in Table 2.
A wide range of olefins and aryl halides were then tested to
investigate group tolerance, steric and electronic effects under opti-
mum reaction conditions using Pd0.09Ce0.91O2-␦ as a catalyst. It is

P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
65
Table 4
arylation of other styrene derivatives. The styrene derivative with a
methyl or methoxy group as a substituent took longer to go to com-
pletion (Table 3, Entries 8, 9). This further confirms that electron
donating substituents on olefins slow down the reaction. Hence,
the electronic effect controls the catalyst activity for both cyclic
and acyclic olefins.
Investigation of steric effects on the Heck coupling reactions.
Entry
Alkene
Reaction
time (h)
Isolated yield
(trans) (%)
TON
1
2
3
1
95
97
96
317
323
320
To investigate any possible steric effects, the reactivity of
methylacrylate, ethyl crotonate and isobutylmethacrylate with
iodobenzene was monitored (Table 4). It was found that steric
effects play a major role in the catalysis. Thus, the reaction with
methyacrylate went to completion in an hour, while it took 12
and 70 h for isobutylmethacrylate and ethyl crotonate to react to
completion, respectively. Isobutylmethacrylate (Table 4, Entry 2)
has a 1,1 disubstitution and ethyl crotonate (Table 4, Entry 3) is
1,2-disubstituted. Thus, isobutylmethacrylate has a more exposed
carbon-carbon double bond than ethyl crotonate, and hence it
reacts faster. In conclusion, the Heck cross coupling reactions favour
electron deficient and less hindered (less substituted) olefins.
The scope of reactivity of various aryl halides was then investi-
gated. It was found that aryl bromides and chlorides do not react
under our optimum conditions. Various aryl iodides were then
investigated to gain better understanding of the catalyst activity.
With aryl iodides, both the electronic effect and substituent posi-
tion effect were explored (Table 5). We found that most aryl iodides
gave comparative reaction times; the electronic effect did not
greatly influence the reactions (with an exception of iodonitroben-
zenes). It was also observed that the position of the substituent on
the aryl iodide has some influence in the reaction rate. Thus, Entries
3–4 show that the position of the methyl group influences the ary-
lation rate of iodotoluene in that 4-iodotoluene is more reactive
than 3-iodotoluene. The substituent position effect is more pro-
nounced in Entries 9–10. In this case, 4-nitroiodobenzene is far less
reactive than 3-nitroiodobenzene. It is unclear at this point how
the substituents on aryl iodides affect the reactivity of the catalyst,
however, the initial observation is that electron rich aryliodides
are coupled more efficiently to methyl acrylate than to electron
deficient aryliodides. This is discussed further in the reaction mech-
anism section.
12
70
Table 5
Investigating the effect of varying the aryl iodide on the Heck coupling reactions.
Entry
R
Reaction time (h)
Isolated yield (% trans)
TON
1
2
3
4
5
6
7
8
H
4-OH
1
2
1
3
2
4
2
2
30
7
2
95
75
98
96
95
92
90
94
71
75
93
317
250
327
320
317
307
300
313
237
250
310
4-CH3
3-CH3
4-NH2
3-NH2
4-COCH3
3-COCH3
4-NO2
3-NO2
3-CF3
9
10
11
reported that both steric and electronic factors can favour arylation
at the terminal position of the olefin or on the more exposed carbon
[3,61,62]. Two sets of experiments were carried out to learn more
about the catalyst activity. First, the olefins were varied and the
aryl halide was kept constant as iodobenzene (Tables 3 and 4). Sec-
ondly, the aryl halides were varied and the olefin was kept constant
as methylacrylate (Table 5).
Aliphatic and aromatic olefins with electron withdrawing and
donating group were arylated with iodobenzene and the results
are tabulated in Table 3. Entries 1–5 in Table 3 are of aliphatic
olefins with varying functional groups. It was observed that olefins
with electron withdrawing groups are easily arylated (Entries 1,
2, 4 and 5), with reaction times of less than an hour. However,
Entry 3, which shows an olefin with an electron-donating group,
has a longer reaction time. Entries 6–10 show the arylation of var-
ious styrene derivatives. The arylation of styrene (Entry 6) goes to
completion in 2 h, however, three products formed; predominantly
trans-stilbene (78%), and cis-stilbene (19%) and 1,1-diphenylethene
as minor products. These side products were not observed in the
The product yield (stilbene) of the present catalytic system was
compared to other heterogeneous, ligand-free palladium catalysed
Heck coupling reactions reported in literature (Table 6) [63–66].
The stilbene yield in this work was found to be comparable to those
obtained by Nasrollahzadeh et al., Sanjaykmar et al. and Monopoli
et al. [19,64,66]. However, the present catalyst system gives better
results than those reported by Perosa et al. and Karami et al. [63,65].
3.3. Catalyst recovery, recyclability and leaching investigations
3.3.1. Used catalyst characterisation
The EDS spectrum of the used catalyst shows that it still contains
palladium (Fig. 7). It further indicated that the surface concentra-
Table 6
Performance comparison of the present catalyst in Heck cross coupling reactions of iodobenzene and styrene against other related heterogeneous catalysts.
Entry
Catalyst
Reaction conditions
Time/h
Yield/%
[Ref.]
1
2
3
4
5
6
Pd/C 5% (0.05 mol% Pd)
Isooctane, H2O, Aliquant 336, 100 ◦
C
20
4
7
6
10
2
46
81
30
64
93
78
[63]
[64]
[65]
[66]
[19]
This paper
Pd-NPs/ZrO2 (0.3 mol% Pd)
Pd-complex/TiO2 (0.12 mol% Pd)
Ce0.98Pd0.02O2-␦ (1.52 mol% Pd)
TiO2@Pd (1 mol%)
H2O, TBAOH, 90 ◦
C
MeOH, Na2CO3, 65 ◦
C
DMF, K2CO3, 130 ◦
C
DMF, (CH3CH2)3N, 140 ◦
DMF, (CH3CH2)3N, 130 ◦
C
C
Pd0.09Ce0.91O2-␦ (0.3 mol% Pd)

66
P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
Fig. 7. The EDS spectrum and TEM image (insert) of the recovered catalyst.
Fig. 8. The XRD pattern and STEM image (insert) of the recovered catalyst.
tion of palladium in the used catalyst is not significantly different to
that of the fresh catalyst. The TEM image of the used catalyst shows
the presence of palladium nanoparticles on the surface of ceria, this
indicates that palladium ions are extruded from the lattice of ceria
during the reaction (Fig. 7, insert).
ysed C H bond activation [33]. In their system, the Ce1-xPdxO2-␦
material did not become active until it was significantly reduced
by ethylene and hydrogen flowing over the catalyst.
3.3.2. Catalyst leaching and recyclability tests
The XRD pattern of the recovered catalyst also shows that the
fluorite structure of the catalyst was intact; however, there is a
small peak at 40◦ 2-␪, indicating the presence of Pd0. This fur-
ther suggests that Pd2+ ions are extruded from the lattice of ceria
and sit on the surface as palladium nanoparticles (Fig. 8). In addi-
tion, the STEM images of the used catalyst also show Pd0 particles
deposited on ceria (Fig. 8, insert). This thus suggests that the single
A mixture of Pd0.09Ce0.91O2-␦, iodobenzene, methylacrylate and
triethylamine in DMF was stirred for an hour at 130 ◦C. After com-
pletion of the reaction, the catalyst was separated by centrifuge. (a)
The mother liquor was analysed by ICP-OES for palladium content.
(b) To the recovered catalyst, fresh substrates were added, namely
iodobenzene, methylacrylate, triethylamine, and stirred in DMF at
130 ◦C for an hour. After completion of the reaction, the catalyst
was separated by centrifuge, and parts (a) and (b) were repeated.
This process was repeated three times and the results are tabulated
in Table 7.
phase Pd2+0.09Ce0.91
O2-␦ solid solution is not the active species and
that catalysis occurs instead over the reduced two phase Pd0/CeO2.
Misch et al. reported similar observations for Ce1-xPdxO2-␦ catal-

P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
67
rich than Pd2+, it allows for oxidative addition of the aryl iodide,
which marks the beginning of the catalytic cycle (2).
Table 7
Leaching and recyclability testa.
Reaction time/h
% Conversion
Amount Pd leached
(ICP)/ppm
The olefin coordinates and forms a pi complex with palladium
(3). The catalytic results suggest that this olefin insertion is likely
the rate-determining step. The oxidative addition step cannot be
the rate-limiting step, since varying aryl iodides gave comparable
activity. Similar observations were reported by de Vries et al., who
suggested that oxidative addition only becomes a rate determining
step under ligand-free Heck coupling of aryl bromide [73]. Olefin
insertion is then followed by ␤-hydride elimination, and the disso-
ciation of the product moiety liberates the reaction product (4). The
active catalyst is then regenerated through reductive elimination
of hydrogen by the base triethylamine.
Cycle 1
Cycle 2
Cycle 3
1
1
1
100
98
98
0.35
0.35
0.25
a
Pd content in fresh catalyst = 390 ppm.
4. Conclusions
In conclusion, a more highly Pd substituted ceria catalyst (9
mol%), Pd0.09Ce0.91O2-␦ was prepared, characterized by various
techniques and shown to be catalytically active towards the Heck
coupling reaction between aryl iodides and alkenes. The physico-
chemical properties of this material indicate that the Pd2+ ions are
substituted in the lattice structure of ceria and not deposited on its
surface as PdO. However, used catalyst characterisation revealed
that the catalysis occurs over the reduced two phase Pd0/CeO2
and not on the as prepared monophasic Pd0.09Ce0.91O2-␦. Good
to excellent yields of substituted olefins were obtained with the
Pd0.09Ce0.91O2-␦ precatalyst when tested on ligand-free Heck cou-
pling reactions. The Heck reactions were also shown to be affected
by electronic and steric effects. Electron deficient olefins react more
rapidly compared to electron rich olefins. In addition, an olefin
with a more exposed carbon-carbon double bond also reacts more
rapidly. Finally, the active catalyst (Pd0/CeO2) can be easily recov-
ered and reused for at least three times without any significant
loss in efficiency and only a negligible amount of palladium was
detected in the product solution (0.35 ppm).
Scheme 1. Proposed Heck coupling reaction mechanism catalysed by the
Pd0.09Ce0.91O2-␦ precatalyst.
The catalyst was recycled successfully three times without any
significant loss in activity. However, the ICP results show that there
was a minimal metal contamination of the products; a total of about
0.25% of palladium leached out of the Pd0.09Ce0.91O2-␦ solid solu-
tion over the three cycles. When the filtrate from cycle one was
stirred with methylacrylate and 3-iodoacetophenone only 15% con-
version was achieved in two hours, as compared to 100% conversion
with the fresh catalyst. This implies that the reaction mechanism
of the filtrate is different to that of the fresh catalyst. Although
these results suggest that the Pd0.09Ce0.91O2-␦ catalysed Heck cou-
pling reactions are largely heterogeneous, reports in literature have
shown that supported palladium catalyst are mostly reservoirs of
active soluble zerovalent palladium species [2,67–70]. Thus there
is a possibility of a dissolution-recapture mechanism, with ceria
acting like a Pd0 scavenger. In any case, the Pd0.09Ce0.91O2-␦ pre-
catalyst has proven effective in minimisation of metallic waste and
allows for simplification of work-up.
Acknowledgements
We would like to express our gratitude to Mintek and the
Department of Science and Technology, South Africa (Advanced
Metals Initiative program) for financial support. The authors would
also like to thank Dr M Shozi, Dr S Mohamed, Dr S Singh and the
rest of the UKZN catalysis research group for helpful discussions.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.mcat.2017.09.
022.
3.4. Proposed mechanism
References
There are two main proposed mechanisms for the palladium
catalysed Heck coupling reaction using homogeneous catalysts,
namely the Pd0/Pd2+ and Pd2+/Pd4+ catalytic cycles. For the het-
erogeneous Heck reaction discussed here, the mechanism is not
clear. However, the catalytic activity data and the physicochem-
ical properties of the fresh and used catalyst suggest that the
reaction follows a Pd0/Pd2+ catalytic cycle (Scheme 1). The XRD
diffractogram and HR-TEM images of the used catalyst suggest that
palladium extrudes from the ceria lattice and is deposited as palla-
dium nanoparticles on the surface of ceria. Hence, we proposed that
the first step in the reaction mechanism is ‘pre-activation’ which
allows Pd2+0.09Ce0.91O2-␦ to enter the catalytic cycle as Pd0/CeO2
(1). The Pd2+ ions in the fresh catalyst are reduced in situ by DMF
and/or triethylamine to Pd0 [14,71,72]. Since Pd0 is more electron
[1] J.C. Hierso, Angew. Chem. Int. Ed. 53 (2014) 920.
[2] A. Biffis, M. Zecca, M. Basato, J. Mol. Catal. A: Chem. 173 (2001) 249.
[3] R.F. Heck, J.P. Nolley, J. Org. Chem. 37 (1972) 2320.
[4] T. Mizoroki, K. Mori, A. Ozaki, Bull. Chem. Soc. Jpn. 44 (1971) 581.
[5] S. Sisodiya, L.R. Wallenberg, E. Lewin, O.F. Wendt, Appl. Catal. A: Gen. 503
(2015) 69.
[6] J. Xiao, Z. Lu, Y. Li, Ind. Eng. Chem. Res. 54 (2015) 790.
[7] X.U. Qijie, S.H.I. Wenzhong, Z.U.O. Chunshan, L.I.U. Dayong, Y.A.N.G. Chun,
Z.H.A.O. Xiang, L.I.U. Pengcheng, Chin. J. Appl. Chem. 32 (2015) 428.
[8] C. Battilocchio, B.N. Bhawal, R. Chorghade, B.J. Deadman, J.M. Hawkins, S.V.
Ley, Isr. J. Chem. 54 (2014) 371.
[9] M. Hosseini-Sarvari, Z. Razmi, M.M. Doroodmand, Appl. Catal. A: Gen. 475
(2014) 477.
[10] C. Pavia, F. Giacalone, L.A. Bivona, A.M.P. Salvo, C. Petrucci, G. Strappaveccia, L.
Vaccaro, C. Aprile, M. Gruttadauria, J. Mol. Catal. A: Chem. 387 (2014) 57.
[11] A. de Meijere, F.E. Meyer, Angew. Chem. Int. Ed. 33 (1995) 2379.
[12] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.

68
P.P. Mpungose et al. / Molecular Catalysis 443 (2017) 60–68
[13] D. Mc Cartney, P.J. Guiry, Chem. Soc. Rev. 40 (2011) 5122.
[14] L. Yin, J. Liebscher, Chem. Rev. 107 (2007) 133.
[15] K. Köhler, R.G. Heidenreich, J. Krauter, Chem. Eur. J. 8 (2002) 622.
[16] A. Kumbhar, Top. Curr. Chem. 375 (2016) 2.
[17] M. Islam, P. Mondal, K. Tuhina, A.S. Roy, J. Braz. Chem. Soc. 22 (2011) 319.
[18] H. Doucet, Angew. Chem. Int. Ed. 53 (2014) 629.
[19] M. Nasrollahzadeh, A. Azarian, A. Ehsani, M. Khalaj, J. Mol. Catal. A: Chem. 394
(2014) 205.
[20] A. Banazadeh, A. Pirisedigh, F. Aryanasab, H. Salimi, S. Shafiei-Haghighi, Inorg.
Chim. Acta. 429 (2015) 132.
[21] J.W. Brown, N.N. Jarenwattananon, T. Otto, J.L. Wang, S. Gloggler, L. Bouchard,
Catal. Commun. 65 (2015) 105.
[22] B. Movassagh, F.S. Parvis, M. Navidi, Appl. Organomet. Chem. 29 (2015) 40.
[23] A. Naghipour, A. Fakhri, Catal. Commun. 73 (2016) 39.
[24] S. Sobhani, Z.M. Falatooni, S. Asadi, M. Honarmand, Catal. Lett. 146 (2015) 255.
[25] S. Sobhani, Z. Zeraatkar, F. Zarifi, New J. Chem. 39 (2015) 7076.
[26] L. You, W. Zong, G. Xiong, F. Ding, S. Wang, B. Ren, I. Dragutan, V. Dragutan, Y.
Sun, Appl. Catal. A: Gen. 511 (2016) 1.
[27] J. Demel, M. Lamac, J. Cejka, P. Stepnicka, ChemSusChem 2 (2009) 442.
[28] Z. Li, J. Chen, W. Su, M. Hong, J. Mol. Catal. A: Chem. 328 (2010) 93.
[29] A.R. Hajipour, N.S. Tadayoni, Z. Khorsandi, Appl. Organomet. Chem. 30 (2016)
590.
[46] S. Roy, A. Marimuthu, M.S. Hegde, G. Madras, Catal. Commun. 9 (2008) 101.
[47] P. Singh, M.S. Hegde, Dalton Trans. 39 (2010) 10768.
[48] P. Singh, M.S. Hegde, Chem. Mater. 21 (2009) 3337.
[49] P. Singh, M.S. Hegde, J. Gopalakrishnan, Chem. Mater. 20 (2008) 7268.
[50] K.C. Patil, M.S. Hegde, T. Rattan, S.T. Aruna, Chemistry of Nanocrystalline
Oxide Materials: Combustion Synthesis, Properties and Applications; World
Scientific: 5 Toh Tuck Link, Singapore, 2008, pp. 9–41.
[51] R.V. Gulyaev, T.Y. Kardash, S.E. Malykhin, O.A. Stonkus, A.S. Ivanova, A.I.
Boronin, Phys. Chem. Chem. Phys. 16 (2014) 13523.
[52] R.V. Gulyaev, A.I. Stadnichenko, E.M. Slavinskaya, A.S. Ivanova, S.V. Koscheev,
A.I. Boronin, Appl. Catal. A: Gen. 41 (2012) 439.
[53] K.R. Priolkar, P. Bera, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro, N.P.
Lalla, Chem. Mater. 14 (2002) 2120.
[54] S. Colussi, A. Gayen, M. Farnesi Camellone, M. Boaro, J. Llorca, S. Fabris, A.
Trovarelli, Angew. Chem. Int. Ed. 48 (2009) 8481.
[55] S. Hinokuma, H. Fujii, M. Okamoto, K. Ikeue, M. Machida, Chem. Mater. 22
(2010) 6183.
[56] Z.A. Qiao, Z. Wu, S. Dai, ChemSusChem 6 (2013) 1821.
[57] H. Zhang, M. Jin, Y. Xia, Angew. Chem. Int. Ed. 51 (2012) 7656.
[58] W. Niu, L. Zhang, G. Xu, Nanoscale 5 (2013) 3172.
[59] Y. Xia, Y. Xiong, B. Lim, S.E. Skrabalak, Angew. Chem. Int. Ed. 48 (2009) 60.
[60] C. Rossy, J. Majimel, M.T. Delapierre, E. Fouquet, F.X. Felpin, Appl. Catal. A:
Gen. 482 (2014) 157.
[30] N. Kann, Molecules 15 (2010) 6306.
[31] S.J. Tabatabaei Rezaei, A. Shamseddin, A. Ramazani, A. Mashhadi Malekzadeh,
P. Azimzadeh Asiabi, Appl. Organomet. Chem. (2017) e3707.
[32] J.A. Kurzman, J. Li, T.D. Schladt, C.R. Parra, X. Ouyang, R. Davis, J.T. Miller, S.L.
Scott, R. Seshadri, Inorg. Chem. 50 (2011) 8073.
[61] A. King, N. Yasuda, Organometallics in Process Chemistry, 6, Springer, Berlin
Heidelberg, 2004, pp. p205.
[62] E. Mieczyn´ ska, A. Gniewek, I. Pryjomska-Ray, A.M. Trzeciak, H. Grabowska,
M.A. Zawadzki, Appl. Catal. A: Gen. 393 (2011) 195.
[33] L.M. Misch, J.A. Kurzman, A.R. Derk, Y.I. Kim, R. Seshadri, H. Metiu, E.W.
McFarland, G.D. Stucky, Chem. Mater. 23 (2011) 5432.
[63] A. Perosa, P. Tundo, M. Selva, S. Zinovyev, A. Testa, Org. Biomol. Chem. 2
(2004) 2249.
[34] J.A. Kurzman, L.M. Misch, R. Seshadri, Dalton Trans. 42 (2013) 14653.
[35] M. Narayanappa, V.D.B.C. Dasireddy, H.B. Friedrich, Appl. Catal. A: Gen. 135
(2012) 447.
[36] T. Cwele, N. Mahadevaiah, S. Singh, H.B. Friedrich, Appl. Catal. B: Environ. 182
(2016) 1.
[64] A. Monopoli, A. Nacci, V. Calò, F. Ciminale, P. Cotugno, A. Mangone, L.C.
Giannossa, P. Azzone, N. Cioffi, Molecules 15 (2010) 4511.
[65] K. Karami, M.B. Shehni, N. Rahimi, Appl. Organomet. Chem. 27 (2013) 437.
[66] S.R. Sanjaykumar, B. Mukri, S. Patil, G. Madras, M.S.J. Hegde, Chem. Sci. 123
(2011) 47.
[37] T. Cwele, N. Mahadevaiah, S. Singh, H.B. Friedrich, A.K. Yadav, S.N. Jha, D.
Bhattacharyya, N.K. Sahoo, Catal. Sci. Tech. 6 (2016) 8104.
[38] M.S. Hegde, G. Madras, K.C. Patil, Acc. Chem. Res. 42 (2009) 704.
[39] T. Baidya, G. Dutta, M.S. Hegde, U.V. Waghmare, Dalton Trans. (2009) 455.
[40] A. Baidya, G. Deshpandey, M.S. Hegde, J. Phys. Chem. C 113 (2009) 4059.
[41] P. Bera, K.C. Patil, V. Jayaram, G.N. Subbanna, M.S. Hegde, J. Catal. 196 (2000)
293.
[67] M. Bhanage, M. Shirai, Chem. Eur. J. 6 (2000) 843.
[68] P. Albers, J. Pietsch, S.F. Parker, J. Mol. Catal. A: Chem. 173 (2001) 275.
[69] J.A. Gladysz, Recoverable and Recyclable Catalysts, John Wiley & Sons, Ltd,
Chichester UK, 2009, pp. 1–45.
[70] R.L. Augustine, S.T. O’Leary, J. Mol. Catal. A: Chem. 95 (1995) 277.
[71] A. Jutand, The Mizoroki-Heck Reaction, John Wiley & Sons, Ltd., Chichester,
UK, 2009, pp. 6–50.
[42] P. Bera, K.R. Priolkar, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro, N.P.
Lalla, Chem. Mater. 14 (2002) 3591.
[43] A. Gayen, K.R. Priolkar, P.R. Sarode, V. Jayaram, M.S. Hegde, G.N. Subbanna, S.
Emura, Chem. Mater. 16 (2004) 2317.
[44] A. Gupta, A. Kumar, U.V. Waghmare, M.S. Hegde, Chem. Mater. 21 (2009) 4880.
[45] S. Roy, M.S. Hegde, Catal. Commun. 9 (2008) 811.
[72] J.P. Collman, R.G. Finke, J.R. Norton, Principles and Applications of
Organotransition Metal Chemistry, 2nd ed., University Science Books, Mill
Valley, California, 1987, pp. 1145–1146.
[73] A.H.M. de Vries, J.M.C.A. Mulders, J.H.M. Mommers, H.J.W. Henderickx, J.G. de
Vries, Org. Lett. 5 (2003) 3285.