Suzuki-Miyaura coupling reactions using novel metal oxide supported ionic palladium catalysts

Article abstract of DOI:10.1016/j.molcata.2016.10.033

Palladium substituted CeO₂, SnO₂ and their mixed oxides have been synthesized in quantitative yields out of non-toxic and inexpensive precursors using a simple and rapid single step solution combustion method. The resulting oxides, especially the mixed oxides Ce_0.79Sn_0.20Pd_0.01O_2-δ, Sn_0.79Ce_0.20Pd_0.01O_2-δ and Sn_0.99Pd_0.01O_2-δ proved to be highly active (TOF > 12,000 h^?1) for Suzuki-Miyaura cross-couplings of phenylboronic acid with various bromoarenes. The reactions were carried out in ambient air at moderate temperatures using environmentally friendly aqueous ethanol solutions as reaction solvents. Minimal amounts of palladium in the product solution ( < 0.14 mg/L), the reaction kinetics as well as catalyst poisoning tests support the thesis that the reaction proceeds via dissolved palladium species in a homogeneous reaction mechanism. Nevertheless, the synthesized catalysts could be reused for at least five times with only minor changes in activity and no changes in the crystal structure, indicating the high potential of the investigated catalysts as quasi-heterogeneous C[sbnd]C coupling catalysts.

Full text of DOI:10.1016/j.molcata.2016.10.033

Accepted Manuscript
Title: Suzuki-Miyaura Coupling Reactions Using Novel Metal
Oxide Supported Ionic Palladium Catalysts
Author: G.J. Lichtenegger M. Maier M. Hackl J.G. Khinast
W. Go¨ ssler T. Griesser V.S.Phani Kumar H. Gruber-Woelfler
Parag A. Deshpande
PII:
S1381-1169(16)30456-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.10.033
MOLCAA 10095
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
18-4-2016
3-10-2016
31-10-2016
Please cite this article as: G.J.Lichtenegger, M.Maier, M.Hackl, J.G.Khinast,
W.Go¨ ssler, T.Griesser, V.S.Phani Kumar, H.Gruber-Woelfler, Parag A.Deshpande,
Suzuki-Miyaura Coupling Reactions Using Novel Metal Oxide Supported
Ionic Palladium Catalysts, Journal of Molecular Catalysis A: Chemical
http://dx.doi.org/10.1016/j.molcata.2016.10.033
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Highlights

Five novel cerium-tin-palladium-oxides (Ce0.99-xSnxPd0.01O2-δ (x = 0 – 0.99)) have been prepared out
of non-toxic and inexpensive precursors using a simple and rapid single step solution combustion
method

The catalysts proved to be highly active (TOF > 12,000h-1) for Suzuki-Miyaura cross-couplings of
phenylboronic acid with various bromoarenes


Only minimal amounts of palladium could be detected in the product solution (<0.14 mg/L)
The catalysts could be reused for at least five times with only minor changes in activity and with no
changes concerning the crystal structure

Suzuki-Miyaura Coupling Reactions Using Novel Metal Oxide Supported Ionic
Palladium Catalysts
G. J. Lichtenegger,a M. Maier,a M. Hackl,a J. G. Khinast,a W. Gössler,b T. Griesser,c V. S. Phani
Kumar,d H. Gruber-Woelfler,a,* Parag A. Deshpanded,*
a Graz University of Technology, Institute of Process and Particle Engineering, NAWI Graz, Graz,
Austria
b University of Graz, Institute of Chemistry – Analytical Chemistry, Graz, Austria
c Chair of Chemistry of Polymeric Materials & Christian Doppler Laboratory for Functional and
Polymer Based Ink-Jet Inks, University of Leoben, Leoben, Austria
d Quantum and Molecular Engineering Laboratory, Department of Chemical Engineering,
Indian Institute of Technology Kharagpur, Kharagpur 721302, India
* Corresponding authors: woelfler@tugraz.at (HGW), parag@che.iitkgp.ernet.in (PAD)

Abstract
Palladium substituted CeO2, SnO2 and their mixed oxides have been synthesized in quantitative yields
out of non-toxic and inexpensive precursors using a simple and rapid single step solution combustion
method. The resulting oxides, especially the mixed oxides Ce0.79Sn0.20Pd0.01O2-δ, Sn0.79Ce0.20Pd0.01O2-δ
and Sn0.99Pd0.01O2-δ proved to be highly active (TOF > 12,000 h-1) for Suzuki-Miyaura cross-couplings
of phenylboronic acid with various bromoarenes. The reactions were carried out in ambient air at
moderate temperatures using environmentally friendly aqueous ethanol solutions as reaction solvents.
Minimal amounts of palladium in the product solution (<0.14 mg/L), the reaction kinetics as well as
catalyst poisoning tests support the thesis that the reaction proceeds via dissolved palladium species in
a homogeneous reaction mechanism. Nevertheless, the synthesized catalysts could be reused for at
least five times with only minor changes in activity and no changes in the crystal structure, indicating
the high potential of the investigated catalysts as quasi-heterogeneous C-C coupling catalysts.
Keywords: Ionic catalysts; cross coupling; solid solution; palladium; recyclability; heterogeneous
catalysts.

1. Introduction
C-C coupling reactions are an important class of reactions prevalent in synthesis protocols
adopted for making active pharmaceutical ingredients, natural products and bioinorganic materials [1]
on a lab scale, as well as in industrial settings. The importance of these reactions was further
demonstrated by the 2010 Nobel Prize in Chemistry to three eminent chemists (Richard F. Heck, Akira
Suzuki and Ei-ichi Negishi) on Pd-catalyzed cross couplings in organic synthesis [2]. Palladium has
been extensively used for catalyzing cross-coupling reactions. The catalytic systems are generally
organometallic complexes. Several groups have focused their attention towards the development of
ligands which effectively catalyze the reaction [3]. If the reaction is carried out with a homogeneous
catalyst, complete recovery of the catalyst often becomes a challenge. The catalyst impurities in the
product can be harmful when the products are intended as pharmaceuticals owing to the poisonous
nature of the metal, as well as of the organic ligands. Therefore, we have focused on ligand free,
heterogeneous catalysts for the Suzuki-Miyaura reaction, as these reaction systems have several
advantages [1,3]..
The Suzuki-Miyaura reaction in its native form involves the coupling of haloarenes with
phenylboronic acids with the major product being the corresponding biphenyls. Functionalized
biphenyls are common intermediates of pharmaceutical products. For example, biphenyl tetrazoles
show antimicrobial activity [4] and derivatives of biphenyl show anti-tumor effects [5]. Due to the
abundance of this compound in a large number of pharmaceutical products, biphenyl can be
considered to be a building block of a large variety of pharmaceutical products. Miyaura and Suzuki
[6] reported first the use of organoboron compounds for C-C bond formation with the help of Pd(Ph3)4.
A large number of organometallic Pd complexes has been reported since for catalyzing the reaction
and the protocol has successfully been applied for large-scale synthesis of pharmaceuticals [3]. Apart
from palladium, Suzuki-Miyaura reactions have been also catalyzed by several other transition metal
complexes. For example, an iron-MCPA [7] complex and Fe halides have been reported as catalysts
for the reaction [8]. Additionally, Cu has also been used for catalyzing the reaction, e.g., Cu

nanoclusters were found to be active for catalyzing the reaction [9]. Furthermore, bimetallic catalysts
consisting of Cu along with Pd, Pt and Ru also proved activity. Combined Cu/Pd catalysts showed the
highest activity among all combinations [9]. Additionally, Ni complexes have also been used for the
synthesis of biaryls using Suzuki-Miyaura reactions [10].
We report here palladium substituted CeO2, SnO2 and their mixed oxides as catalysts for Suzuki-
Miyaura reactions. Palladium-containing metal oxides and mixed metal oxides have already been used
successfully for Suzuki-Miyaura cross-coupling reactions. Catalysts of this kind, synthesized by sol-
gel methods (e.g., La1Fe0.95Pd0.05Ox) [11] or wet impregnation techniques (Pd/Al2O3, Pd/TiO2,
Pd/CeO2) showed extraordinary high reactivity, good recyclability and minimal metal leaching
[12,13].
Pd-doped CeO2 catalysts developed by Hegde and co-workers were reported as efficient and non-
deactivating catalysts for gas-phase oxidation reactions [14]. The solution combustion technique offers
a relatively simple, inexpensive and rapid method for the synthesis of noble metal substituted ceria and
ceria composites. Furthermore, solution combustion method permits homogeneous mixing of the
catalysts precursors on the molecular level (in an aqueous solutions) and ensure high product
crystallinity and purity due to the high combustion temperature (>900 °C) [15]. These characteristic
features allow the production of oxide-based catalysts with high specific surface and precise phase
composition [16]. Preliminary studies on C-C coupling reaction catalyzed by Pd-doped ceria-based
catalysts were published by Sanjay Kumar et al. [17].
In this study, we have carried out a thorough investigation of Pd-ion substituted CeO2, SnO2 and their
mixed oxides as catalysts for Suzuki-Miyaura reactions. In particular, a comparison of the activity of
the different catalysts depending on their Ce and Sn amount as well as their tolerance to different
starting materials were studied. Furthermore, the leaching behaviour and the recyclability of the
catalysts were investigated. All in all, the study demonstrates the high potential of the investigated
catalysts as heterogeneous C-C coupling catalysts.
2. Experimental details

2.1 Synthesis
General principles and methods for the solution combustion synthesis of nanocrystalline
materials have been described and reviewed elsewhere [18,19]. The synthesis of Ce0.99-xSnxPd0.01O2-δ
was done analogously to the method reported by Baidya et al. [20] using glycine instead of L-tartaric
acid as fuel. Glycine as the fuel has been used for combustion synthesis by several investigators and
has been reported to impart good nanocrystallinity in the product [14,16,20]. A set of five
Ce0.99-xSnxPd0.01O2-δ mixed oxides (with x = 0, 0.2, 0.495, 0.79 and 0.99) was prepared. The
appropriate proportions of catalyst precursors and fuel were calculated using the method described by
González-Cortés et al. [21] considering the oxidizing/reducing valencies (Φe) of the redox mixture and
assuming an ideal and complete combustion without secondary reactions. In this calculation, oxygen
was considered as an oxidizer with the valance -2, whereas carbon (+4), hydrogen (+1) and metals
(redox valence corresponding to metal valence) were considered as reducing elements. Nitrogen with
the redox valence 0 was not considered. Oxidizing/reducing valencies of the reacting species were
ammonium cerium (IV) nitrate =-24, tin (IV) nitrate =-20, palladium (II) nitrate = -10 and glycine = +9
to give the required amount of fuel for the synthesis of 1 mole Ce0.99-xSnxPd0.01O2-δ.
For the preparation of 3 g catalyst, appropriate amounts of ammonium cerium(IV) nitrate
((NH4)2Ce(NO3)6, Sigma Aldrich), tin(II) oxalate (SnC2O4, Sigma Aldrich), palladium(II) chloride
(PdCl2, Aldrich) and glycine (C2H5NO2, Sigma Aldrich) were dispersed in 3 ml of water in a
borosilicate dish with 600 ml capacity. 2.5 molar equivalent (related to tin (II) oxalate) of nitric acid
(HNO3, 65 %, Merck) were added and the mixture was slightly heated until a viscous, clear solution
was obtained. The redox mixture was treated in a muffle furnace at 350 °C. The mixture ignited within
5 minutes, initiating a self-propagating combustion reaction, which yielded a voluminous, porous
solid. This solid was ground with mortar and pestle and was heated to 350 °C for another 5 hours.
After that procedure, the obtained powders were directly usable as catalyst for the Suzuki-Miyaura
cross-coupling reactions.

2.2 Characterization
X-Ray diffraction
The synthesized catalysts were characterized using X-ray diffraction (XRD) with Cukα
radiation to analyze the crystal structure. The samples were scanned in a range of 2Θ of 20-100°, a
range in which characteristic peaks of CeO2 and SnO2 appear.
XPS spectroscopy
XPS analysis was performed with a monochromatic Thermo Fisher K-Alpha spectrometer
equipped with an Al X-ray source (1486.6 eV) operating at a base pressure in the range of 10-8 to 10-10
mbar. High resolution scans were acquired at a pass energy of 50 eV and a step size (resolution) of
0.1 eV. Survey scans were acquired with a pass energy of 200 eV and a step size of 1.0 eV. The
instrument work function was calibrated to give a binding energy (BE) of 83.96 eV for the Au 4f7/2
line for metallic gold. The peaks were fitted using a Gaussian/Lorenzian mixed function employing
Shirley background correction (Software Thermo Avantage v5.906). All analyses were performed at
room temperature.
Specific surface area
Specific surface areas were determined using a Tristar II 3020 (Micromeritics, Norcross,
Georgia) with nitrogen as analytical gas. Catalyst samples were degassed under vacuum at ambient
temperature for 24 h. The volume of nitrogen adsorption was recorded over a relative pressure range
between 0.01 and 0.99. 8 points in the relative pressure range of 0.05 to 0.2 were used for the
calculation of the surface area according to the Brunauer-Emmet-Teller (BET) theory.
Particle size distribution
The particle size measurements were conducted with a HELOS/KR Laser diffraction sensor
equipped with an OASISDRY/L dry dispersion system by Sympatec GmbH (injector: 4 mm,
dispersing pressure: 2.00 bar, optical concentration 1%-20%, measuring range: combined R2+R5 (0.45
µm - 875 µm), evaluation mode: Fraunhofer approximation, software: WINDOX 5.6.0.0). Each
analysis was conducted as a 3-fold determination.

2.3 Catalytic activity
The catalytic activities of the prepared catalysts were tested for the Suzuki-Miyuara cross-
coupling reaction of phenylboronic acid with various bromoarenes. In a typical experiment, 0.70 mmol
of the aryl bromide, 1.5 molar equivalents of phenylboronic acid and 1.5 molar equivalents of base,
respectively, and 1.3 mmol of anisole (internal standard) were dissolved in 20 ml of solvent. 5 mg of
the catalyst (corresponding to approximately 0.05 mol % palladium) were added. The reaction was
carried out under stirring at 75 °C. To monitor the reaction progress, samples from the reaction
solution (100 µL) were taken after 15, 30, 60, 90 and 120 minutes. HPLC sample solutions were
prepared by dissolving 100 µL of the reaction solution in 1 mL of the initial mobile phase. HPLC
analysis was done using an Agilent 1100 series HPLC system (Agilent, Waldbronn, Germany),
equipped with an online degasser, quaternary pump, autosampler, thermostatted column compartment
and UV-visible diode array detector. An Agilent Poroshell 120 EC-C18 reversed-phase column
(50x4.6 mm; 2.7 µm) was used as stationary phase. Analysis was carried out under gradient elution
conditions using a mobile phase consisting of methanol and aqueous phosphoric acid (water:
phosphoric acid = 300:1 v/v). Column temperature was set to 25 °C. UV-detection was performed at
wavelengths of 237 and 270 nm over a run time of 15 minutes.
2.4 Recycling tests
The cross-coupling reaction of phenylboronic acid with 4-bromotoluene using potassium
carbonate as base was chosen as model reaction to investigate the recyclability of the different
catalysts. The reactions were carried out in 20 ml reaction mixture using the same concentrations of 4-
bromotoluene, phenylboronic acid, potassium carbonate and internal standard (anisole) as in the
reactivity tests. Five subsequent reactions with each catalyst were carried out. After a reaction time of
120 minutes, the catalyst was filtered off using a filter crucible and was washed with ethanol and water
to remove organic and inorganic impurities. The catalyst was dried in a muffle furnace at 350 °C over
night. The filtrate was evaporated under reduced pressure. The solid residue was analyzed for its
palladium content by ICP-MS. Measurements were performed on an Agilent 7500ce ICP-MS after

microwave assisted acidic digestion of the samples. Due to a partial loss of catalyst throughout the
recovery procedure, the first experiment was carried out with an initial catalyst loading of 50 mg
(corresponding to approximately 0.5 mol % palladium, 10 times more than in the reactivity tests).
2.4. Heterogeneity studies
The reaction of 4-bromotoluene with phenylboronic acid (in EtOH/H2O = 7/3 v/v) was used for
the heterogeneity studies using the same substrate and base concentrations as in the reactivity tests.
Based on the results of the reactivity tests, the amounts of the catalysts were chosen so that the
obtained conversions of 4-bromotoluene was in the range between 25 and 80 percent at the moment of
filtration or catalyst poisoning.
2.4.1 Hot filtration test
Hot filtration experiments were done for each catalyst. The experiments were started in the
same manner as the reactivity tests. After a certain reaction time, which ensured conversions of 4-
bromotoluene between 25 and 80 % under the chosen reaction conditions, the catalyst was separated
from the reaction mixture by filtration through a hot (75 °C) filter crucible. The remaining reaction
solution was stirred and heated until a total reaction time of 120 minutes was reached. The reaction
progress before and after the filtration was monitored via HPLC-analysis. The used amounts of
catalysts and the times of filtration are given in Table 1.
Table 1: Catalyst masses and times of filtration for the hot filtration tests using
CexSn1-xPd0.01O2-δ (x = 0 – 0.99) as catalysts.
Catalyst mass
(mg)
Reaction time till
filtration (min)
Ce0.99Pd0.01O2-δ
25
2
20
15
20
10
5
Ce0.79Sn0.20Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ
5
2
2
2.4.2 Catalyst poisoning
Catalyst poisoning tests were started in the same manner as the hot filtration tests. After a
certain reaction time, 100 molecular equivalents (compared to the total molar amount of palladium in

the used catalysts) of the catalyst poison 3-mercaptopropyl-functionalized silica gel (MPSG, Sigma-
Aldrich, 200-400 mesh, extend of labeling: 1.2 mmol/g) were added to the reaction solution. The
reaction progress before and after poisoning was monitored via HPLC-analysis. The used amounts of
catalysts, MPSG and the reaction times till the start of the catalyst poisoning are given in Table 2.
Table 2: Employed catalyst masses, masses of the added catalyst poison MPSG (100 molar
equivalents) and times of poisoning for the catalyst poisoning tests using CexSn1-xPd0.01O2-δ (x = 0 –
0.99) as catalysts.
Catalyst mass
(mg)
MPSG mass
(mg)
Reaction time till
poisoning (min)
Ce0.99Pd0.01O2-δ
25
2
120
15
26
15
15
30
10
20
10
5
Ce0.79Sn0.20Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ
5
2
2
3. Results
3.1 Catalyst synthesis
The synthesis of the five catalysts via combustion of the prepared redox-mixtures yielded the
desired compounds in quantitative amounts (> 99 %) within 1 hour. Redox solutions with high
amounts of cerium (Ce ≥ 0.8) underwent a highly violent combustion reaction and produced a powdery
solid. With increasing amounts of tin (Sn > 0.2), the combustion reactions became more and more
gentle, yielding voluminous spongiform solids. After grinding with mortar and pestle and heating to
350 °C for another 5 hours, the solids could be directly used for characterization or reactivity tests.
3.2 Characterization
3.2.1 XRD
The XRD-profiles of the as-synthesized CexSn1-xPd0.01O2-δ (x = 0 – 0.99) compounds are shown
in Figure 1. The obtained lattice parameters are summarized in Table 3. The powder XRD pattern
showed broad peaks indicative of nanocrystallinity in the synthesized compounds. Single phase cubic
fluorite structures were observed for oxides with tin proportions up to 0.2. The measured cubic lattice
constants were in good accordance with the data obtained by Baidya et al. [20] for the cerium rich

oxides, Ce0.99Pd0.01O2-δ, Ce0.79Sn0.2Pd0.01O2-δ. Due to the smaller ionic radius of Sn4+ (0.81 Å)
compared to Ce4+, a decrease of the lattice constant from a = 5.412 Å (for x = 99) to a = 5.400 Å (for x
= 0.79) indicated a successful incorporation of tin ions at Ce4+ sites in the cubic lattice structure. The
XRD patterns in Figure 1 have been marked with filled triangles (▲) for prominent CeO2 peaks.
Ce0.79Sn0.2Pd0.01O2-δ shows a good correspondence for these peaks, as can be observed from Figure 1.
Contrary to the study of Baidya, a single phase cubic structure for the mixed oxides with equimolar
amounts of cerium and tin (Ce0.495Sn0.495Pd0.01O2-δ) could not be observed in our work. Instead, a more
amorphous cubic/tetragonal mixed phase structure was observed when high ratios of tin were used
(Ce0.495Sn0.495Pd0.01O2-δ and Ce0.20Sn0.79Pd0.01O2-δ). The binary oxide Sn0.99Pd0.01O2-δ showed a clearly
single phase tetragonal structure. The prominent tetragonal peaks of pure SnO2 have been marked by a
filled star () in Figure 1 and crystallization of Sn0.99Pd0.01O2 in tetragonal structure can be clearly
observed.
In order to confirm substitution of Pd in the lattice, XRD was recorded in a small range of 35-
50° as shown inFigure 2 . XRD-patterns clearly attributable to PdO or metallic Pd species could not be
observed (Figure 2). This confirmed the absence of PdO or Pd metal particles and indicated a Pd2+ ion
substitution in the metal oxide lattice.

Figure 1: X-Ray diffraction pattern of as-synthesized CexSn1-xPd0.01O2-δ (x = 0 – 0.99), ▲ - Cubic
fluorite and  - Tetragonal
Figure 2: XRD-pattern of CexSn1-xPd0.01O2-δ (x = 0 – 0.99). The arrow shows the peak position for Pd.

Table 3: Lattice constants of the as-synthesized mixed oxides
Lattice constants (Å)
Tetragonal
Catalyst
Phase
Cubic
a
a
b
cubic
cubic
5.412
5.400
5.385
-
-
Ce0.99Pd0.01O2-δ
-
-
Ce0.79Sn0.2Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
mixed
4.562
3.482
cubic/tetragonal
mixed
cubic/tetragonal
5.260
-
4.732
7.739
3.234
3.186
Ce0.20Sn0.79Pd0.01O2-δ
tetragonal
Sn0.99Pd0.01O2-δ
3.2.2 XPS
The predominantly cationic nature of Pd in the Ce/Sn-oxide lattice could also be confirmed by XPS
analysis. XPS-scans in the Pd 3d region for all synthesized catalysts are shown in Figure 3.

Figure 3: XPS spectra in the Pd 3d region for the novel Ce0.99-xSnxPd0.01O2-δ catalysts: Ce0.99Pd0.01O2-δ
(a), Ce0.79Sn0.2Pd0.01O2-δ (b), Ce0.495Sn0.495Pd0.01O2-δ (c), Ce0.20Sn0.79Pd0.01O2-δ (d), Sn0.99Pd0.01O2-δ (e).
XPS-signals at 337.4 – 338.0 eV for Pd3d5/2 and at 342.8 and 343.5 eV for Pd3d3/2, respectively, are
generally attributed to cationic Pd2+. [17,20,22]. Nevertheless, small amounts of metallic palladium
(binding energies around 335.3 and 340.8 eV) could also be detected on the surface. Observed binding
energies of Pd2+ and metallic Pd0 as well as the relative atomic amounts of both palladium species are
summarized in Table 4. As can be seen, the ratio between Pd2+ and Pd 0 is approximately 90/10 for cat
1 – 4. The highest amount for metallic Pd was observed for Sn0.99Pd0.01O2-δ.
Table 4: Ratio between Pd2+ and Pd0 at the catalyst surface and observes binding energies for Pd3d5/2
and Pd3d3/2.
Binding energies (eV)
Relative Amount (%)
Pd2+
Pd0
Pd2+
89.5
Pd0
Pd3d5/2
337.4
Pd3d3/2
342.8
Pd3d5/2
340.8
Pd3d3/2
335.4
Ce0.99Pd0.01O2-δ
10.5
Ce0.79Sn0.20Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ
88.9
90.4
89.2
77.3
11.1
9.6
337.7
337.9
338.0
337.7
343.1
343.3
343.5
343.1
340.7
340.7
340.9
340.9
335.4
335.2
335.4
335.4
10.8
22.7
3.2.2 Particle size distribution and specific surface area
The particle size distribution and specific surface area are important parameters for catalytic activity,
and were therefore included in our studies. The produced oxides appear as fine powders, consisting of
agglomerates of the nanocrystalline materials. The analysis of the particle size distribution of these

agglomerates revealed particles sizes of ~ 10 – 100 µm with a monomodal size distribution of the
cerium-rich catalysts. The catalysts with higher ratios of tin, which originated from more gentle
combustion reactions yielding spongiform solids, featured polymodal particle size distributions
(Figure 4).
Figure 4: Particle size distributions of as-synthesized CexSn1-xPd0.01O2-δ (x = 0 – 0.99).
Ce0.495Sn0.495Pd0.01O2-δ, which contained equimolar amounts of cerium and tin, had the highest mass-
median-diameter (x50). Minimum x50 were observed for the binary oxides Ce0.99Pd0.01O2-δ and
Sn0.99Pd0.01O2-δ. BET measurements showed specific surface areas at the same order of magnitude for
all catalysts with Ce0.99Pd0.01O2-δ having the lowest (26.9 m²/g) and Ce0.20Sn0.79Pd0.01O2-δ having the
highest surface (97.7 m²/g). The other catalysts had comparable specific surfaces of approximately 50
m²/g (Table 5).
Table 5: Characteristic mean diameters x10, x50, x90 and specific surface areas of as-synthesized
CexSn1-xPd0.01O2-δ (x = 0 – 0.99) catalysts
BET-Surface
x10 [µm]
x50 [µm]
x90 [µm]
[m²/g]
26.9
41.3
48.7
97.7
54.8
Ce0.99Pd0.01O2-δ
0.8
2.5
5.5
1.0
0.6
9.6
15.3
33.8
22.4
5.2
28.2
36.6
189.4
109.6
92.8
Ce0.79Sn0.2Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ

3.3 Reactivity
The catalytic activity of the as-synthesized catalysts was investigated for the Suzuki-Miyaura cross-
coupling reaction of phenylboronic acid with various bromoarenes containing electron donating
(methyl-, hydroxyl-), as well as electron withdrawing (acetyl-, trifluoromethyl-) groups in para
position and the electron withdrawing nitrile group in ortho position. The results of the reactivity tests
are summarized in Table 7. In all experiments, the conversions of the bromoarenes and the yields of
the products formed out of the bromoarenes were equal within the error limits of the HPLC
measurements. The absence of bromoarene-deriving side products (dehalogenation product Arʼ and
bromoarene homocoupling product Arʼ Arʼ) indicated that the bromoarene substrates were converted
to the desired biphenyl compound to the greatest extent in a highly selective manner. However, side
products originating from boronic acid homocoupling (biphenyl, ArAr) and boronic acid oxidation
(phenol, ArOH) could be found in quantifiable amounts, whereas the formation of the
protodeboronation product (benzene, Ar) could not be observed.
Table 7 shows the reaction progress expressed as relative yield of the desired Suzuki cross-coupling
product based on the concentration of the bromoarene (limiting component). The selectivity is defined
as the concentration of the desired product divided by the sum of the concentrations of all formed
organic products. All the reactions were carried out using 50 % molar excess of phenylboronic acid
and potassium carbonate (K2CO3, used as base) at 75 °C, either in pure ethanol or in an aqueous
ethanolic solutions (EtOH/H2O = 7/3 v/v). Due to the low solubility of potassium carbonate in ethanol,
reaction mixtures using pure ethanol appeared as suspensions, whereas the use of aqueous ethanol
gave clear and colorless reaction solutions.
In general, Ce0.79Sn0.2Pd0.01O2-δ, Ce0.20Sn0.79Pd0.01O2-δ and Sn0.99Pd0.01O2-δ (change with tin proportions
of 0.2, 0.79 and 0.99, respectively), showed extraordinarily high activities in all investigated Suzuki-
Miyaura cross coupling reactions (up to TOF of >12,000 h-1 for the coupling of 4-bromoacetophenone
with phenylboronic acid). The use of less than 0.05 mol% of active palladium led to a quantitative
conversion of all para-substituted bromoarenes in less than two hours with aqueous ethanol as reaction

solvent. Full conversion within 2 hours could also be reached using 2-bromobenzonitrile as aryl halide,
however, 4 times more catalyst had to be used to achieve comparable reaction rates.
Ce0.495Sn0.495Pd0.01O2-δ also showed catalytic activity in all reactions, however significantly lower
reaction rates were observed. The binary oxide Ce0.99Pd0.01O2-δ proved to be the least active catalyst in
all investigated reactions except for the very fast reaction of phenylboronic acid with 4-
bromoacetophenon. In this reaction, total conversion of the bromoarene substrate could be achieved
within 30 minutes even when the less reactive catalyst 1 and 3 were used (whereby both catalysts
showed comparable reaction rates). Suzuki reactions of 4-bromotoluene with phenylboronic acid over
pure CeO2 and pure SnO2 did not show any progress, indicating that Pd is the active species in the
investigated catalysts. All in all, at the current stage, our results do not show any direct context
between Ce or Sn loading and catalytic activity. However, there might be a relation between the
surface area, as well as the particle size and the catalytic activity: our results show that Ce0.99Pd0.01O2-δ
has the lowest surface area and the lowest catalytic activity. The large particle size of
Ce0.495Sn0.495Pd0.01O2-δ might be the reason for its low activity. But these are just preliminary
conclusions and more investigations are needed prove this theory.
In our studies we also investigated the influence of the solvents, bases and starting bromoarenes. As
can be seen in Table 7, in all experiments, the use of aqueous ethanolic solvents instead of pure
ethanol led to increased reaction rates, which can be attributed to the higher solubility of potassium
carbonate in water containing alcoholic solvents. p-bromobenzenes containing electron withdrawing
substituents showed higher reactivity than bromobenzenes containing electron donating groups.
Regardless of the reaction solvent, the reactivity follows the order: 4-bromoacetophenone ≥ 4-
bromobenzotrifluoride > 4-bromotoluene > 4-bromophenol.
The influence of different bases and solvent systems were investigated using the moderately fast
reaction of 4-bromotoluene with phenylboronic acid catalyzed by Ce0.495Sn0.495Pd0.01O2-δ. The results
of these tests are summarized in Table 6.

Table 6: Yields and selectivities after 120 minutes of reaction time observed by HPLC-measurements
for the Suzuki cross-coupling reaction of phenylboronic acid with 4-bromotoluenea
Entry
Solvent (v/v ratio)
EtOH/H2O = 7/3
EtOH/H2O = 7/3
EtOH/H2O = 7/3
EtOH/H2O = 7/3
EtOH/H2O = 7/3
EtOH/H2O = 7/3
EtOH/H2O = 7/3
MeOH/H2O = 7/3
i-PrOH/H2O = 7/3
n-PrOH/H2O = 7/3
Base
Na2CO3
K2CO3
Cs2CO3
NaOH
KOH
NaOAcb
KOAcb
K2CO3
K2CO3
K2CO3
Temperature [K]
Yield [%]
Selectivity [%]
1
2
353
353
353
353
353
353
353
343
353
353
32
92
81
88
88
<1
<1
96
>99
5
97.1
94.4
96.2
94.6
95.8
-
3
4
5
6
7
-
8
93.1
90.7
93.5
9
10
a
Reaction conditions: aryl bromide (0.7 mmol), phenylboronic acid (1.5. mol eq.), K2CO3 (1.5 mol eq.), 5 mg
Ce0.495Sn0.495Pd0.01O2-δ (corresponding to 0.05 mol% Pd), 20 ml of solvent.
using 4 mol eq. of base
b
Although considerable reactivity could be observed using cesium carbonate, potassium hydroxide and
sodium hydroxide as base, the often used potassium carbonate proved to be the optimal base when
EtOH/H2O 7/3 (v/v) was used as solvent. The minor activity of the catalyst using sodium carbonate as
base may be attributed to the low solubility of Na2CO3 in the applied solvent system. Only traces of
the product 4-methylbiphenyl could be detected when the weak bases sodium acetate and potassium
acetate were used.
A slight increase in catalytic activity with comparable selectivities could be achieved when 2-propanol
(at 75 °C) and methanol (at 65°C) were used as alcoholic solvent instead of ethanol. The low product
yield when 1-propanol was used can be attributed to the low capability of this system to dissolve
highly polar inorganic bases. Taking ecological, toxicological and safety aspects into account 2-
propanol/water mixtures may be considered as most practical solvent system for this reaction.
Selectivity
Using HPLC as analytical method, not only the disappearance of the arylbromide and the formation of
the biphenyl-product were monitored, also the concentrations of possible side products were

determined. Typically observed side products of Suzuki-Miyaura coupling reactions [23,24] are shown
in Figure 5.
Figure 5: Main product (Ar’Ar) and typically observed side products in Suzuki-Miyaura cross
coupling reactions.
Table 7: Yields and selectivities observed by HPLC-measurements for the Suzuki cross-coupling
reactions of phenylboronic acid with various bromoarenes, using CexSn1-xPd0.01O2-δ (x = 0 – 0.99) as
catalyst.a
Pure Ethanol
15 Minutes 120 Minutes
Catd Yield
Sel.
EtOH/H2O = 7/3
15 Minutes 120 Minutes
Product
Yield
<1
Sel.
-
Yield
<1b
9b
Sel.
-
98.4
97.3
97.3
96.7
Yield
6
>99
91
Sel.
98.5
91.4
94.4
91.9
92.3
<1
31
7
39
48
-
1
2
3
4
5
92.8
97.9
92.7
96.4
75
39
89
85
90.7
94.9
88.0
93.6
26
4-Methylbiphenyl
77b
82b
>99
98
<1
41
7
44
31
-
10
77
34
93
70
96.0
89.1
92.5
90.3
93.1
2b
-
45
>99
80
97
>99
95.3
84.6
93.7
91.0
91.9
1
2
3
4
5
92.2
97.1
94.1
97.0
50b
23b
30b
69
94.0
96.4
95.6
95.8
4-Phenylphenol
4-Acetylbiphenyl
7
-
-
-
85
>99
98
>99
>99
99.7
96.9
98.6
95.9
98.2
55b
>99b
38b
99.8
97.7
99.8
95.9
94.7
>99
>99
>99
>99
>99
88.4
87.1
93.3
85.9
90.0
1
2
3
4
5
96
53
98
96
99.6
99.7
98b
98b
<1
20
3
53
61
-
2
-
<1
95
29
94
>98
-
2
>99
99
>99
>98
-
1
2
3
4
5
99.0
-
98.2
97.6
88
53
98
97
97.5
98.3
92.1
92.5
97.8
98.8
95.3
92.2
87.3
92.7
81.6
80.3
4-
Trifluoromethylbiphenyl
2-Biphenylcarbonitrilec
1
<1
-
<1
-
<1
-
2
-

78
9
68
64
99.6
-
99.7
99.8
>99
20
>99
>99
89.5
99.6
83.1
83.0
38
6
74
66
98.8
-
99.0
98.9
>99
29
>99
>99
98.4
98.5
83.3
88.7
2
3
4
5
a
Reaction conditions: aryl bromide (0.7 mmol), phenylboronic acid (1.5. mol eq.), K2CO3 (1.5 mol eq.), 5 mg
CexSn1-xPd0.01O2-δ (corresponding to 0.05 mol% Pd), 20 ml of solvent, 75 °C.
After 10 minutes
Using 20 mg of catalyst (0.2 mol% Pd)
Cat 1: Ce0.99Pd0.01O2-δ, Cat 2: Ce0.79Sn0.2Pd0.01O2-δ, Cat 3: Ce0.495Sn0.495Pd0.01O2-δ, Cat 4: Ce0.20Sn0.79Pd0.01O2-δ, Cat 5:
b
c
d
Sn0.99Pd0.01O2-δ.
The obtained selectivities are closely related to the catalyst’s activity in our calculations. This can be
illustrated comparing the concentration profiles and selectivities during the cross-coupling reactions of
phenylboronic acid with 2-biphenylcarbonitrile using Ce0.79Sn0.2Pd0.01O2-δ and Ce0.20Sn0.79Pd0.01O2-δ,
respectively (Figure 6).
Figure 6: Concentration profiles for the reaction of 2-bromobenzonitrile and phenylboronic acid
catalyzed by (a) Ce0.79Sn0.2Pd0.01O2-δ and (b) Ce0.20Sn0.79Pd0.01O2-δ.
When full conversion of the starting halide is reached, the values for selectivity are equally high in
both reactions (98.4 % after 120 and 45 minutes for Ce0.79Sn0.2Pd0.01O2-δ and Ce0.20Sn0.79Pd0.01O2-δ
respectively). The side product biphenyl and phenol are formed predominantly at high conversion
levels, i.e., in the absence of the already consumed reaction partner 2-bromobenzonitrile. Hence,
highly selective reactions can be achieved by termination of the reaction just at the moment of full
conversion, even when a high excess of phenylboronic acid is used.
3.3 Recyclability and Pd-leaching

Table 8 and Figure 7 show the observed relative conversions of 4-bromotoluene after 30 and 120
minutes reaction time, as well as the recoverability of the used solid catalyst in mass percent (Table 8).
Detailed diagrams of the obtained conversion vs. time are shown in the supporting information. Using
0.5 mol% Pd, full conversion after 2 hours could be achieved in five subsequent reactions, when the
tin-containing catalysts (x = 0 – 0.99) were used, despite a significant loss of catalysts during the
whole recycling procedure (filtration and drying). Using Ce0.495Sn0.495Pd0.01O2-δ, full conversion of 4-
bromotoluene within two hours could not be reached in run 4 and 5. However, comparison of the
initial reaction rates and the whole reaction progress with the results of the reactivity test shows only
minor decrease of the mass specific catalyst activity. A similar recycling behavior could be observed
for Ce0.99Pd0.01O2-δ at a lower activity level.
Table 8: Results of the recycling experiments for the Suzuki-Miyaura reaction of 4-bromotoluene with
phenylboronic acid.
Run 1 Run 2 Run 3 Run 4 Run 5
30 min
120 min
13
78
84
100
100
88
100
100
81
100
100
81
100
100
81
50
91
70
100
100
76
94
100
68
100
100
66
100
100
68
25
81
55
99
100
68
89
100
58
99
100
50
16
69
43
99
100
68
81
98
48
100
100
45
100
100
40
8
49
33
98
100
60
76
97
38
98
100
32
96
100
32
Conversion [%]
Recoverability [%]
Conversion [%]
Recoverability [%]
Conversion [%]
Recoverability [%]
Conversion [%]
Recoverability [%]
Conversion [%]
Recoverability [%]
Ce0.99Pd0.01O2-δ
30 min
120 min
Ce0.79Sn0.2Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ
30 min
120 min
30 min
120 min
30 min
120 min
100
100
54

Figure 7: Observed conversions after 30 (a) and 120 minutes (b) for the Suzuki-coupling of 4-
bromotoluene with phenylboronic acid in 5 subsequent runs using 0.5 mol% Pd. Cat 1:
Ce0.99Pd0.01O2-δ, Cat 2: Ce0.79Sn0.2Pd0.01O2-δ, Cat 3: Ce0.495Sn0.495Pd0.01O2-δ, Cat 4: Ce0.20Sn0.79Pd0.01O2-δ,
Cat 5: Sn0.99Pd0.01O2-δ.
To investigate the leaching behaviour of the catalysts, the catalyst was filtered off after each reaction
cycle and washed with water and ethanol. The combined reaction and washing solutions were
evaporated under reduced pressure and the remaining solids were analyzed for their palladium content
by means of ICP-MS. When Ce0.79Sn0.2Pd0.01O2-δ and Ce0.20Sn0.79Pd0.01O2-δ were used as catalysts, the
palladium contents were determined after each run. Due to the good reproducibility of the results (0.08
± 0.01 mg/L and 0.15 ± 0.02 mg/L for Ce0.79Sn0.2Pd0.01O2-δ and Ce0.20Sn0.79Pd0.01O2-δ, respectively) the
Pd content in the reaction solutions was determined representatively for run 3 for all other catalysts.

Table 9 shows the obtained Pd-concentrations in the remaining solids as well as the corresponding
palladium concentration in solution, represantively for run 3.
Table 9: Pd-contents of the residual solids and corresponding Pd-concentrations in solution for the 3rd
reaction
Pd in remaining solids [mg/kg]
Pd in solution [mg/L] after 3rd run
after 3rd run
Catalyst
Ce0.99Pd0.01O2-δ
5.0
6.7
5.3
10.0
11.0
0.06
0.09
0.07
0.13
0.14
Ce0.79Sn0.2Pd0.01O2-δ
Ce0.495Sn0.495Pd0.01O2-δ
Ce0.20Sn0.79Pd0.01O2-δ
Sn0.99Pd0.01O2-δ
Figure 8 shows a comparison of the initial reaction rates (expressed as turnover frequencies after 15
minutes reaction time) and the amount of palladium in solution for the third recycling run (catalysts are
ordered according to increasing activity). The results demonstrate an apparent correlation between
reactivity and amount of leached palladium and support the thesis, that the coupling reaction is actually
catalyzed by small amounts of leached palladium via a homogeneous reaction mechanism. Amoroso
et al. [12] described a sigmoidal shape of the temporal conversion profile for Suzuki-couplings using
Pd/CeO2 catalysts and attributed this finding to an induction time period of the reaction due to the
formation of dissolved palladium species, acting as true catalytically active species. Sigmoidal
conversion profiles also could be observed in this study in moderately fast reactions, not only for
Ce0.99Pd0.01O2-δ but also for the tin containing catalysts. Figure 9 illustrates such sigmoidal profiles for
the reaction of 4-bromotoluene with phenylboronic acid catalyzed by Ce0.79Sn0.2Pd0.01O2-δ and
Ce0.495Sn0.495Pd0.01O2-δ, respectively.

Figure 8: TOF (after 15 minutes) vs. Pd-concentration in solution (mg/L) for the third recycling
experiment.
Figure 9: Sigmoidal profile of the reaction progress for the coupling reaction of 4-bromotoluene with
phenylboronic acid using Ce0.79Sn0.2Pd0.01O2-δ and Ce0.495Sn0.495Pd0.01O2-δ.
The results of the recycling tests agree very well with the findings of other studies using Pd-doped
metal oxides as precatalysts, and substantiate the hypothesis, that the precatalysts act as palladium

reservoir, that slowly releases extremely reactive palladium species [11,12]. The successful recycling
of the catalysts might be attributed to a Pd release/capture mechanism, which was not only proposed
for Pd on metal oxides [11,12] but also for a wide variety of Pd catalysts on different solid supports
[25–28].
In a second recycling series, the catalysts were used for 3 Suzuki reactions in the same manner and
afterwards characterized by means of XRD (Figure 10). The XRD-patterns of the used catalysts reveal
intact fluorite structures of the cerium rich catalysts, Ce0.99Pd0.01O2-δ and Ce0.79Sn0.2Pd0.01O2-δ
respectively, a crystalline mixed phase structure of the mixed oxides Ce0.495Sn0.495Pd0.01O2-δ and
Ce0.20Sn0.79Pd0.01O2-δ and the remaining tetragonal structure of the binary oxide Sn0.99Pd0.01O2-δ. These
findings prove the high stability of the produced catalysts at the applied reaction conditions. Also in
the used catalysts, patterns attributable to Pd (111) could not be observed, which confirms the absence
of PdO or segregated Pd metal species.
Figure 10: X-Ray diffraction pattern of three times used CexSn1-xPd0.01O2-δ (x = 0 – 0.99)
3.4. Heterogeneity studies

In order to get more insight in the homo- or heterogeneous nature of the novel catalysts, we carried out
further tests, including the hot filtration test and catalyst poisoning.
3.4.1 Hot filtration test
An often applied test to determine the heterogeneity of a system is the so called hot filtration test. After
a certain reaction time, the solid catalyst is removed from the reaction solution by filtration through a
hot filter (filter frit or filter crucible). The subsequent reaction progress in the remaining reaction
solution without solid catalyst is monitored over time. In our study, this test was performed for all
produced catalysts employing the Suzuki cross-coupling reaction of 4-bromotoluene with
phenylboronic acid in ethanol/water = 7/3 v/v. Based on the results of the reactivity test, the amount of
catalyst and the moment of filtration were chosen in such a way that the obtained conversions of 4-
bromotoluene was in the range between 25 and 80 percent. The results of the hot filtration tests are
depicted in Figure 11.
Figure 11: Results of the hot filtration test employing the Suzuki-Miyaura cross-coupling reaction of
4-bromotoluene and phenylboronic acid
In all experiments, the removal of the catalyst by filtration led to a complete stop of the conversion of
4-bromotoluene to 4-methylbiphenyl, although all required reaction participants remained in solution

(which was confirmed by HPLC-analysis). This could lead to the assumption, that the reaction is
catalyzed by an insoluble catalyst in a heterogeneous manner and there is no catalytic activity of the
compounds in the solution after filtration. However, it is known that hot filtration tests are rather
problematic for the precise distinction between homogeneous and heterogeneous catalysis [29], since
catalytically active leached metal can redeposit on the catalyst very quickly during filtration and may
not be detected by a filtration test. Furthermore, massive interruptions in the reaction process (such as
filtration) can deactivate dissolved metal species. Thus, more reliable methods for the determination of
the homogeneous or heterogeneous nature of metal catalyzed reactions, including three-phase-tests
[30] and catalyst poisoning tests [31] have been developed.
3.4.1 Catalyst poisoning
According to Jones et al. [31], the catalyst poisoning test represents one of the most potential methods
to distinguish between homogeneous and heterogeneous catalytic systems. During this test, materials,
which are known to act as selective catalyst poisons for dissolved palladium, are added to the reaction
mixtures. Frequently used poisons for palladium catalyzed cross-coupling reactions include
Quadrapure TU (polymeric resin), poly(4-vinylpyridine) (PVPy) and 3-mercaptopropyl functionalized
silica gel (MPSG) [29]. The addition of such materials leads to the removal of soluble reactive metal
from the reaction solution and would lead to a suppression of a further reaction progress in a
homogeneously catalyzed reaction.
In this work, we used 3-mercaptopropyl functionalized silica gel (MPSG, 100 molecular equivalents of
SH-groups with respect to the employed amount of palladium) as selective catalyst poison for the
reaction of 4-bromotoluene with phenylboronic acid in ethanol/water = 7/3 v/v (at the same reaction
conditions as in the hot filtration test). The complete suppression of the reaction after addition of the
catalyst poison (Figure 12) strongly indicates a homogeneous reaction mechanism via dissolved
palladium species. Although the good recyclability of the catalysts and the results of the hot filtration
test seemingly suggest a heterogeneous reaction mechanism, the findings of the catalyst poisoning
experiments, the sigmoidal profile of the reaction progress and the low but detectable amount of

palladium in the filtered reaction solution strongly support the above mentioned thesis, that the
produced oxides act as precatalysts and release extremely reactive palladium species, which represent
the true catalytically active species. Further studies to get more inside in the reaction mechanism will
be carried out in the near future.
Figure 12: Catalyst poisoning test employing mercaptopropyl functionalized silica gel in the Suzuki-
Miyaura cross-coupling reaction of 4-bromotoluene and phenylboronic acid
4. Conclusions
Palladium substituted CeO2, SnO2 and their mixed oxides proved to be efficient catalysts for Suzuki-
Miyaura cross-coupling reactions. The synthesis of the oxides via a one-step solution combustion
method represents a further simplification of rapid and inexpensive preparation strategies, such as sol-
gel methods or wet impregnation techniques. In general, tin containing catalysts proved to be clearly
more catalytically active than previously reported binary cerium-palladium oxides. Especially, the
mixed oxides Ce0.79Sn0.2Pd0.01O2-δ, Ce0.20Sn0.79Pd0.01O2-δ and Sn0.99Pd0.01O2-δ showed extraordinarily