Ligand-free, recyclable palladium-functionalized magnetite nanoparticles as a catalyst in the Suzuki-, Sonogashira, and Stille reaction

Article abstract of DOI:10.1016/j.jorganchem.2019.121005

A magnetically reusable ligand-free Fe₃O₄ palladium functionalized catalyst system was successfully prepared without the use of reducing agents, but by making use of the reduction potential of magnetite. The stabilizer was variated depending on the investigated reaction, whereby poly(ethylene glycol) (PEG) stabilized nanoparticles were used for the Suzuki reaction, as it requires protic conditions, while oleic acid stabilized nanoparticles were used for the Sonogashira and Stille reaction. It was found that it was possible to perform the Suzuki reaction and the Sonogashira reaction resulting in good to excellent conversions under air. Despite the good results for the Suzuki and the Sonogashira reaction it was not possible to perform the Stille reaction using this easily synthesized catalyst system due to the poisoning of the reusable catalyst by the tin-compound. Furthermore, the reusable catalyst system was recycled and reused for five times, resulting in a separable, straightforward and less time-consuming catalyst system.

Full text of DOI:10.1016/j.jorganchem.2019.121005

Journal Pre-proof
Ligand-free, recyclable palladium-functionalized magnetite nanoparticles as a catalyst
in the Suzuki-, Sonogashira, and Stille reaction
Amaury De Cattelle, Arne Billen, Galahad O'Rourke, Ward Brullot, Thierry Verbiest,
Guy Koeckelberghs
PII:
S0022-328X(19)30448-6
DOI:
https://doi.org/10.1016/j.jorganchem.2019.121005
JOM 121005
Reference:
To appear in: Journal of Organometallic Chemistry
Received Date: 10 September 2019
Revised Date: 22 October 2019
Accepted Date: 25 October 2019
Please cite this article as: A. De Cattelle, A. Billen, G. O'Rourke, W. Brullot, T. Verbiest, G.
Koeckelberghs, Ligand-free, recyclable palladium-functionalized magnetite nanoparticles as a catalyst
in the Suzuki-, Sonogashira, and Stille reaction, Journal of Organometallic Chemistry (2019), doi: https://
doi.org/10.1016/j.jorganchem.2019.121005.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
© 2019 Published by Elsevier B.V.

Ligand-free, recyclable palladium-functionalized
magnetite nanoparticles as a catalyst in the Suzuki-
, Sonogashira, and Stille reaction
Amaury De Cattellea, Arne Billenb, Galahad O’Rourkea, Ward Brullotb, Thierry Verbiestb,
Guy Koeckelberghsa*
aLaboratory for Polymer Synthesis, Department of Chemistry, KU Leuven, Celestijnenlaan
200F, B-3001 Heverlee, BELGIUM
bLaboratory for Molecular Electronics and Photonics, Department of Chemistry, KU Leuven,
Celestijnenlaan 200F, B-3001 Heverlee, BELGIUM
*
To
whom
correspondence
should
be
addressed.
E-mail
address:
guy.koeckelberghs@kuleuven.be (Guy Koeckelberghs)
KEYWORDS: Immobilized catalyst, magnetical separation, Suzuki reaction, Sonogashira
reaction, Stille reaction
Abbreviations:
PEG: Poly(Ethylene Glycol)
ICP-AES: Inductively Coupled Plasma Atomic Emission Spectroscopy

ABSTRACT
A magnetically reusable ligand-free Fe3O4 palladium functionalized catalyst system was
successfully prepared without the use of reducing agents, but by making use of the reduction
potential of magnetite. The stabilizer was variated depending on the investigated reaction,
whereby poly(ethylene glycol) (PEG) stabilized nanoparticles were used for the Suzuki
reaction, as it requires protic conditions, while oleic acid stabilized nanoparticles were used
for the Sonogashira and Stille reaction. It was found that it was possible to perform the
Suzuki reaction and the Sonogashira reaction resulting in good to excellent conversions under
air. Despite the good results for the Suzuki and the Sonogashira reaction it was not possible
to perform the Stille reaction using this easily synthesized catalyst system due to the
poisoning of the reusable catalyst by the tin-compound. Furthermore, the reusable catalyst
system was recycled and reused for five times, resulting in a separable, straightforward and
less time-consuming catalyst system.
1 Introduction
Catalysts are an indispensable component for the majority of chemical processes, which
makes them a key factor in the chemical research industry. More specifically, palladium-
catalyzed cross-coupling reactions have attracted much interest over the last 30 years,
because they unlocked valuable organic coupling pathways, not solely in academia but also
for industrial applications.[1–3] Academically, the importance was recognized by awarding
the 2010 Chemistry Nobel Prize to Richard F. Heck, Eiichi Negishi and Akira Suzuki for the
development of "palladium-catalyzed coupling reactions in organic synthesis".[4,5] The
opportunity for a tolerable functionalization of aryl halides with boronic acids, acetylenes or
tin compounds is the practical advantage for the use of Suzuki-, Sonogashira-, Stille- and
many other cross-coupling reactions. In the past, these reactions were typically performed
under homogeneous conditions using palladium and (phosphine) ligands as a catalytic
system.[6–9] Nowadays, more innovative catalyst systems are used.[10–13] However, the
separation and recovery of the catalyst represent major concerns for sustainable development.
Generally, homogeneous palladium catalysts are popular for their high selectivity, activity,
and resistance towards poisons, while heterogeneous catalysis offers the advantage of
simplified product separation and recyclability of the catalyst.[14–17] Therefore, the
immobilization of active homogeneous catalysts on solid supports has become a popular
strategy to combine the advantages of both homo-and heterogeneous catalysis. Efforts have
2

already been made to develop heterogeneous Pd catalysts for cross-coupling reactions[18–20]
by the immobilization of Pd on various supports, such as silica nanoparticles,[21–24]
graphene,[25,26] graphene oxide,[27] cellulose,[28,29] chitosan,[30–35] Agar,[36] bio-
composite,[37–39] starch,[35,40] carbon nanotubes,[41] dendrimers,[42–45] polymers,[46–
50] metal oxides,[51–53] graphene supported metal nanoparticles[54] and metal oxide
nanoparticles[55–61], which can not only be used using conventional thermal heating but
also using microwave assisted heating. These mentioned nanostructures can be separated into
two groups based on their method of separation. Nanoparticles can be separated by
precipitation using a poor solvent (e.g. silica nanoparticles, polymer nanoparticles) or they
can be separated by a magnet (e.g. magnetite nanoparticles) employing the
superparamagnetic property of the nanoparticle. Due to the simplicity of the last method and
ease of recovery, magnetite nanoparticles will be used during this study. The control of
nanoparticle dimensions by modern synthetic techniques has enabled the production of
catalysts with well-defined shapes, sizes, and compositions.[62] Because nanomaterials have
well-defined dimensions, they can be applied in surface chemistry. For example, spherical
nanoparticles with a size of 3 nm consist of 50% out of surface atoms, meaning a substantial
proportion of the atoms are available for reaction.[63] Facile modification of the size of the
particles alters the surface area ratio and consequently influences the activity. A dispersed
nanoparticle can be considered as a cluster of atoms, that resembles the classical
heterogeneous catalysts. Decreasing the particle size is beneficial because relatively more
atoms become available for interaction at the surface. Therefore, the efficiency of the
heterogeneous catalysts approximates the efficiency of homogeneous catalysts, which are
distinct chemicals consisting of single molecules. Previous research already described the use
of such heterogeneous catalysts which can be implemented in the Suzuki, Sonogashira, Stille
or other cross-coupling reactions. Nonetheless, many of these reported processes with
heterogeneous catalysts require prolonged reaction times, strict reaction conditions and time-
consuming ligand synthesis.[64–68] These ligands can be present as an attached ligand onto
the nanoparticle or as a ligand complexated to the Pd before the functionalization of the
nanoparticle with Pd. In both cases it requires a longer and, consequently, more costly
synthesis of the heterogeneous catalyst. Next to the ligand functionalized magnetite
nanoparticles, core-shell nanoparticles consisting out of a magnetite core whereupon a SiO2
shell or Ag shell is attached has also been described in literature [69–72], which also results
in a more times consuming catalyst preparation, just like the magnetically recyclable zeolite
catalyst systems[73,74]. Nevertheless, it remains important to make use of stabilizing groups
3

attached to the nanoparticle. This will prevent the nanoparticle to agglomerate as it will
improve the reactivity of the immobilized catalyst system.[75] The deposition of Pd onto the
magnetite surface using a reducing agent like NaBH4 or hydrazine, resulting in a ligand-free
immobilized Pd catalyst attached with or without stabilizing groups has been already
reported.[76–79] The use of such reducing agents leads to a certain toxicity and dangerous
reaction circumstances, which is deemed unnecessary. Therefore, our attempt is to make a
catalyst system that provides the advantages of the presence of a stabilizer without the
disadvantage of using a reducing agent. This can be achieved using the reducing nature and
coordinating ability of the Fe3O4 support. Although already reported, no use was made of any
stabilizing groups.[80] Our contribution is to combine both the advantage of the presence of a
stabilized group and the absence of a reducing agent during synthesis. Through the use of
poly(ethylene glycol) (PEG) or oleic acid as stabilizer for the nanoparticle together with a
straightforwardly available Pd source like Pd(OAc)2 and the reducing nature of the Fe3O4
support, it is possible to synthesize a less time consuming, cheap, non-toxic and reusable
catalyst for the implementation in cross-coupling reactions. Despite the fact that such easily
synthesized Pd functionalized magnetite nanoparticle was already used for the performance
of a magnetically induced Suzuki reaction, improvement is still possible.[81] Due to the
limitation of the heat generated by the magnetic field, the Suzuki reaction requires a
prolonged reaction time of 4 hours with average conversions, and the Sonogashira reaction
does not lead to any conversion to the end-product. By using conventional thermal heating, it
is possible to adapt the temperature to obtain high conversions with a short reaction time,
using an as low as possible amount of valuable catalyst.
In this study straightforwardly synthesized Pd-functionalized magnetite nanoparticles are
made starting from commercially available products in order to reduce the production time
and increase the safety of the heterogeneous catalyst synthesis. The novel catalytic system is
investigated for the Suzuki-, Sonogashira- and Stille-reactions. In consideration of the fact
that these reactions require different solvent conditions (protic versus aprotic), several types
of commercially available stabilizers are used to optimize the efficiency of the reactions. By
making use of the superparamagnetic property of the Fe3O4 support it is possible to easily
remove the heterogeneous catalyst system by the use of a magnet. Furthermore, the magnetite
will act as a relative cheap support for the Pd catalyst. This in comparison to Pd
nanoparticles, where only the surface of the nanoparticle can be used to catalyze the reaction,
which means that the Pd present in the center can be seen as a waste and is an unnecessary
4

cost. The use of a core-shell nanoparticle existing of a magnetite core surrounded by a golden
shell will also be examined because this can combine the advantages of the use of Au in
cross-coupling reactions[82] with the reusability of the catalyst system. Not only will the
reactions be performed using an easily reusable catalyst, they will also be executed under air
atmosphere using relative short reaction times, in order to facilitate these reactions.
2 Materials and methods
2.1 Experimental information
All reagents were purchased from TCI, Sigma-Aldrich, J&K Scientific and Acros Organics
1
and used without further purification. H-NMR spectra were recorded on a Bruker Avance
300 MHz spectrometer. The ultrasonication bath used in both functionalization of the
magnetite nanoparticle and the core-shell synthesis was a Bransonic Model 5510 sonicator
with a capacity of 10 L. The UV chamber used in the siloxane functionalization is equipped
with 3 LEDs (365 nm) with an output power of 200 mW. For the TEM images, an 80 kV
Zeiss EM-900 electron microscope was used in combination with 300 mesh Formvar coated
copper grids from Ted Pella. For the inductively coupled plasma – atomic emission
spectroscopy (ICP-AES) measurements, a Varian 720ES was used with a A Varian patented
vistachip CCD detector.
2.2 Pd functionalization of nanoparticles
Oleic acid stabilized Fe3O4 nanoparticles, PEG-silane stabilized Fe3O4 nanoparticles or gold-
coated Fe3O4 nanoparticles in 1,4-dioxane were brought into a 100 mL flask. Its weight in
palladium acetate was added accordingly, and the reaction mixture was stirred for four days
after it was brought under argon atmosphere. The mixture was transferred to a tarred vial and
the nanoparticles were isolated by a magnetic field and subsequently washed three times with
1,4-dioxane. The nanoparticles were dried under reduced pressure and the yield is determined
gravimetrically. Thereafter dioxane is added to obtain a stock dispersion of nanoparticles. For
more detailed information: see supplementary material.
2.3 General procedure for the Suzuki-Miyaura couplings
A two-necked flask was charged with, the aryl halide (0.180 mmol), the boronic acid-
compound (0.270 mmol), K2CO3 (0.360 mmol, 52.5 mg) and 1,3,5-trimethoxybenzene (0.180
mmol, 30.3 mg). A dispersion of Fe3O4/PEG/Pd nanoparticles in EtOH (0.500 mL, 0.100 mg
/mL) was added to the two-necked flask, together with 0.500 mL H2O. The reaction mixture
was stirred for 1 h at 60 °C and the nanoparticles were isolated from the reaction mixture
with a magnet. The conversion, with respect to the desired end-product, was determined by
5

1H-NMR of the crude reaction mixture, using 1,3,5-trimethoxybenzene as an internal
standard. For more detailed information: see supplementary material.
2.4 General procedure for the Sonogashira couplings
A two-necked flask was charged with, the aryl halide (0.180 mmol), 1,3,5-trimethoxybenzene
(0.180 mmol, 30.3 mg) and KOAc (0.270 mmol, 26.5 mg). A dispersion of Fe3O4/oleic
acid/Pd nanoparticles was made in d6-DMSO (1.00 mL, 2.00 mg /mL) and was added to the
two-necked flask. The acetylene (1.80 mmol) was added and the mixture was stirred for 3 h
under reflux at 100 °C. After stirring, the nanoparticles were isolated from the mixture with a
1
magnet. The conversion, with respect to the desired end-product, was determined by H-
NMR of the crude reaction mixture, using 1,3,5-trimethoxybenzene as an internal standard.
For more detailed information: see supplementary material.
3 Results and discussion
In this study, a reusable catalyst system is created for different types of cross-coupling
reactions. First, the Suzuki cross-coupling reaction is investigated, which is a typical example
of a reaction that requires protic conditions.[83,84] Secondly, water-free reactions are
performed and investigated by using the Sonogashira and Stille cross-coupling reaction. To
produce relatively simple reusable catalysts, two types of nanoparticles are synthesized, one
with a higher affinity for aprotic solvents, and one with a higher affinity for protic solvents.
The affinity of the nanoparticle for the solvent can easily be changed by the use of different
stabilizers. Nanoparticles stabilized with a poly(ethylene glycol) chain have a higher affinity
for protic solvents compared to the oleic acid stabilized nanoparticles. The reason for using
two types of nanoparticles instead of using one is due to the fact that the development of an
easy-to-apply catalyst system was kept in mind. The synthesis of the oleic acid stabilized
nanoparticles can be defined as a one-pot synthesis and is straightforward.[85] To obtain the
PEG-silane stabilized nanoparticles, on the other hand, two synthesis steps are required
consisting of, first, the synthesis of the nanoparticles with oleic acid as stabilizer and second,
an exchange of this oleic acid stabilizer by PEG.[86] Therefore, PEG stabilized nanoparticles
are harder to obtain and will only be used if necessary. Core-shell nanoparticle existing of a
magnetite core surrounded by a golden shell will also be used to examine the influence of
gold. Therefore, three catalyst systems will be used in this study: Pd functionalized magnetite
nanoparticles stabilized with PEG (Fe3O4/PEG/Pd), Pd functionalized magnetite
nanoparticles stabilized with oleic acid (Fe3O4/oleic acid/Pd) and, Pd functionalized core-
shell nanoparticles stabilized with PEG (Fe3O4/Au/PEG/Pd).
6

3.1 Catalyst preparation
The different types of magnetite nanoparticles are prepared using known procedures. The
magnetite nanoparticles stabilized with oleic acid are performed using the method of Chen et
al.[85], while the water-soluble PEG stabilized magnetite nanoparticles are synthesized
according to Bloemen et al.[86] The core-shell nanoparticles are synthesized and optimized
by the procedure of Billen et al.[87] The functionalization of all nanoparticles with Pd is done
by stirring Pd(OAc)2 together with the nanoparticles in 1,4-dioxane for 4 days under argon
atmosphere (Scheme 1). Uniquely, the functionalization is performed without the use of any
reducing reagents or surface modification, but occurs due to the reducing nature and
coordinating ability of the Fe3O4 support. Note that, in principle, it is possible to deposit any
metal with a reduction potential more positive than that of Fe2+ (e.g., Co, Ni, Cu, Rh, Ru, Pd,
Ag, Pt, and Au). This method was already reported in literature.[80] The amount of Pd
present on the different types of catalyst systems was determined using inductively coupled
plasma – atomic emission spectroscopy (ICP-AES) (Table 1). It should be mentioned that the
amount of Pd on a PEG stabilized nanoparticle is higher compared to the oleic acid stabilized
nanoparticle. This can be explained by the difference in stabilizer, where the palladium
originating from Pd2+, will be easier inserted inside the PEG stabilizer, due to the higher
polarity of the PEG stabilizer compared to the oleic acid stabilizer, and by this making easier
contact with the magnetite nanoparticle. The reason for the lower Pd content present on the
Fe3O4/Au/PEG nanoparticle compared with the Fe3O4/PEG nanoparticle is twofold. Firstly,
the size of the core-shell nanoparticles is larger compared with the magnetite nanoparticles,
resulting in a lower surface to mass ratio, which results in a lower amount of Pd to mass ratio.
Secondly, due to the coverage of the magnetite nanoparticle with the golden-shell, the
magnetite is less available.
Single column
Scheme 1: Functionalization of the different types of nanoparticles with Pd.
Table 1: ICP-AES results of Pd functionalized nanoparticles
Amount of Pd
(mg Pd/mg NPs)
Amount of Pd
Nanoparticles
(10-4 mmol Pd/mg NPs)
Fe3O4/PEG/Pd
Fe3O4/oleic acid/Pd
Fe3O4/Au/PEG/Pd
0.086
0.014
0.051
8.1
1.3
4.8
7

3.2 Suzuki cross-coupling reactions
The Suzuki-Miyaura cross-coupling reaction is a main example of a Pd cross-coupling
reaction in the presence of water. The reaction of 4-bromobenzoic acid with phenylboronic
acid is used for initial studies as a model reaction and is carried out in the presence of
aqueous K2CO3.[88–92] The solvent chosen for this reaction was an EtOH/H2O (1:1)
mixture, which according to Zhou et al. results in optimal conditions.[80] The presence of
water in the reaction mixture is the reason for the use of PEG-stabilized Pd-functionalized
magnetite nanoparticles (Fe3O4/PEG/Pd). When oleic acid stabilized nanoparticles are used in
this reaction flocculation of the NPs occurs in the reaction mixture, but by the use of PEG as
stabilizer a one-phase reaction mixture is obtained. As the Suzuki reaction is largely affected
by the amount of catalyst and temperature, we set out to optimize the reaction conditions
using the model reaction (Figure 1).
Temperature: RT
Temperature: 60 °C
Double column
Figure 1: Effect of the temperature and the amount of Fe3O4/PEG/Pd on the Suzuki cross-coupling. 4-bromobenzoic acid
(0.18 mmol), phenylboronic acid (0.27 mmol), K2CO3 (0.36 mmol), Fe3O4/PEG/Pd, EtOH/H2O (1:1) (1.0 mL), under air.
1
1
Reaction was monitored by H-NMR. H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as an internal
standard.
As evident from Figure 1, the use of a temperature of 60 °C with a catalyst amount of 0.05
mg of Fe3O4/PEG/Pd resulted in full conversion after 1h. At room temperature, the yield of
the target product decreases drastically when decreasing the catalyst amount, whereas a
decrease in the catalyst amount at 60 °C resulted in only a small influence on the reaction
yield. In summary, the best result is obtained at 60 °C during 1h using 0.05 mg of
Fe3O4/PEG/Pd. It should be mentioned that full conversion was also obtained at room
temperature after 2h with a catalyst loading of 0.3 mg of Fe3O4/PEG/Pd, resulting in an
8

energy-efficient Suzuki reaction with short reaction times, and a low catalyst loading of only
1.3 10-3 mol% relative to the aryl halide.
To explore the scope and limitations of this protocol, a range of aryl halides and arylboronic
acids were investigated (Table 2). It is observed that aryl iodides and aryl bromides couple
readily with phenylboronic acid to give significant results (Table 2), but that no reaction
occurs for aryl chlorides (Table 2; entry 5 and 6). Interestingly, almost no difference in yield
is noticeable between aryl iodides and aryl bromides. Generally speaking, a favorable effect
of electron-withdrawing substituents on the halide is normally observed in palladium-
catalyzed reactions. It is observed that aryl bromides containing electron-withdrawing groups
couple readily with phenylboronic acid to give excellent yields (Table 2; entry 10 and 12). By
introducing more steric hindrance on the aryl bromide, the yield decreases (Table 2; entry 3,
7 and 8). This was also noticed for the use of an electron-donating methoxy- or amine group
(Table 2; entry 9 and 11). Very interesting results are obtained when a comparison is made
between entry 4 and 9 (Table 2). Although all two halides have an electron-donating group,
entry 4 leads to a higher conversion compared to entry 9. This leads to conflicting results,
because due to the presence of K2CO3 the phenol-species is deprotonated to a phenoxide,
resulting in a higher electron donation and thus a deactivation compared to the 4-
bromoanisole. The reason for this opposite result is twofold. Firstly, this phenoxide has a
high affinity for magnetite nanoparticles. Due to the interaction of the phenoxide with the
nanoparticle, the substituent becomes a less electron-donating group compared to a “pure”
alkoxide-species. Second and most importantly, due to the high affinity of the phenoxide for
the nanoparticles, the binding brings the substrate in closer proximity to the catalytic system
resulting in a higher yield compared to 4-bromoanisole for which no deprotonation can occur.
The electronic nature of substituents of arylboronic acid was also investigated. The results
show that arylboronic acids bearing either electron-donating groups or electron-withdrawing
groups can provide the corresponding product (Table 2; entry 13 and 14). The use of an
electron-donating substituent results in a higher yield compared to an electron-withdrawing
group on the aryl boronic acid species.
9

Table 2: The scope of Suzuki cross-coupling reactions with different substituents.
Entry
Reactant 1
Reactant 2
Product
Conversionb
1
2
3
4
5
6
73 %
70 %
76 %
67 %
0 %
0 %
7
14 %
Br
Br
8
9
8 %
20 %
100 %
10
11
12
25 %
100 %
13
14
55 %
33 %
aReaction conditions: reactant 1 (0.18 mmol), reactant 2 (0.27 mmol),
K2CO3 (0.36 mmol), Fe3O4/PEG/Pd (0.05 mg), EtOH/H2O (1:1)
(1.0 mL), temp: 60 °C, 1h, under air.
b1H-NMR conversions using 1,3,5-trimethoxybenzene as an internal standard.
Recovery and recycling of supported catalysts are very important issues from both the
practical and environmental points of view. Therefore, catalyst recycling was performed at
the optimized reaction conditions, to evaluate the reusability of the Fe3O4/PEG/Pd in the
Suzuki reaction. After each cycle, when the reaction is finished, an external magnet is applied
on the reaction vessel. Whereafter the separation of the catalyst is achieved and the resulting
supernatant can be decanted, which minimizes the loss of catalyst during separation. The
nanoparticle is then subsequently washed with 4 mL of 1,4-dioxane, 4 mL of water, 2 times 4
mL of ethanol and is dried under reduced pressure. The catalyst system was reused without
further purification. As shown in Figure 2 the catalyst can be reused at least 5 times, although
with a loss in activity. At the 5th reaction, a conversion of only 18% was obtained. It should
10

be mentioned that the loss in nanoparticles after each recovery was taken into account. This
means that the amount of starting products, base and solvents are adapted to the amount of
nanoparticles present after the recuperation. Due to this measure, the loss in activity can only
be the cause of a change of the Fe3O4/PEG/Pd nanoparticle. By making use of ICP-AES the
Pd-amount present in the crude end-product can be measured. As depicted in Figure 2, there
is a large loss in activity of the catalyst after the first cycle and is accompanied by a loss in Pd
of 1.2%. After the 2nd, 3rd, and 4th, more moderate losses in conversions are observed and are
accompanied by a respectively loss of 0.4%, 0.3% and 3.4% of Pd. It should be mentioned
that the reaction is performed with only 0.22 10-3mol% Pd. Therefore, it is understandable
that even the leaching of a small quantity of Pd can cause a reduction in conversion.
Transition electron microscopy (TEM) images of the Fe3O4/PEG/Pd nanoparticles before and
after the reactions were recorded, in order to investigate the change in morphology of these
nanoparticles as a result of the reaction (Figure 3). According to Figure 3, there is almost no
difference in morphology between the nanoparticles before and after the reaction. Therefore
there can be concluded that the decrease in conversion is probably due to the leaching of the
Pd.
Single column
Figure 2: The reusability of the Fe3O4/PEG/Pd NPs for the Suzuki reaction using 4-bromobenzoic acid and phenylboronic
acids as substrates using 0.05 mg of nanoparticles/mL solvent (EtOH/H2O 1:1) with adapting the amount of substrates to the
amount of nanoparticles recycled after each cycle.
Figure 3:TEM images of the Fe3O4/PEG/Pd nanoparticle catalysts before (left) and after (right) the Suzuki reaction.
11

Table 3: Comparison between the catalytic performance of Fe3O4/PEG/Pd and different catalysts in the Suzuki cross-
coupling reactions.
Pd content Yielda
Catalyst
Solvent
T (°C)
Time
Ref
(mol %)
0.015
0.01
0.02
(%)
63
54
47
51
80
93
67
85
75
62
72
75
72
Chitosan-cellulose/Pd(II)
OCMCS-SBL/Pd
Chitosan-Ulva/Pd(II)
OCMCS-3aPd
Pd NPs@Fe3O4/CS-AG
Fe3O4@SiO2-Pd(0)
Pd NPs@APC
/
/
/
50 (MW) 5 min
50 (MW) 4 min
50 (MW) 4 min
100
50 (MW) 5 min
85 26 min
50 (MW) 5 min
50 (MW) 6 min
60 (US) 30 min
50 (MW) 7 min
50 (MW) 5 min
50 (MW) 6 min
50 (MW) 6 min
[31]
[32]
[33]
[34]
[59]
[58]
[13]
[36]
[37]
[28]
[12]
[35]
[38]
Toluene
48 h
0.04
/
1.5 10-5
0.03
H2O/EtOH
/
/
/
/
/
/
/
1.8 10-6
3.0 10-6
1.0 10-5
2.0 10-3
0.004
AG-Pd
Pd NPs@CMC/AG
CL-Gly-Pd
Chitosan/cellulose-Pd NP
Pd@chitosan/starch
ST Shf Pd(II)
Fe3O4-
sporopollenin/Pd(II)
CS-NNSB-Pd(II)
OCS-NS-Pd(II)
CL-Sc-Pd(II)
Fe3O4/Ethyl-CN/Pd
ST-Sc-Pd(II)
γ-Fe2O3–Pd
Chitosan-pyridil-Pd(II)
Si-P-Pd
Fe@FexOy/Pd
Fe3O4/PEG/Pd
aIsolated yield of the pure product
bConversion
0.005
2.0 10-3
/
50 (MW) 5 min
1.0 10-3
70
[57]
/
/
/
50 (MW) 5 min
50 (MW) 4 min
50 (MW) 6 min
25
50 (MW) 5 min
110 1 h
50 (MW) 5 min
6.0 10-3
5.0 10-3
5.0 10-3
0.2
38
66
64
98
68
90
69
91
84
73b
[11]
[39]
[29]
[93]
[40]
[55]
[30]
[94]
[80]
H2O/EtOH
/
H2O/DMF
/
H2O/MeOH
H2O/EtOH
H2O/EtOH
30 min
4.0 10-3
0.1 10-4
5.0 10-3
0.5
25
25
60
4 h
2 h
1 h
0.5
2.2 10-4
This work
MW = Microwave
US = Ultrasonication
A comparison is made to other nanocatalysts reported in literature (Table 3). As seen in Table
3, the palladium functionalized magnetite nanoparticle show decent conversions compared to
already existing systems. When the Fe3O4/PEG/Pd catalyst system is compared with other
heterogeneous catalysts performed using thermal heating, there can be concluded that by
using this catalyst system the reactions can be performed using a catalyst loading of over
2000 times lower compared to already existing systems with a longer reaction time. When
this system is compared with a maghemite supported Pd catalyst, there can be concluded that
the catalyst system obtained in this work was performed at much lower reaction temperatures.
However, when this thermal induced reaction is compared with microwave induced reaction,
12

using recyclable catalyst systems there can be observed that for the last one shorter reaction
times are necessary, leading to similar conversions as the Fe3O4/PEG/Pd.
3.3 Sonogashira cross-coupling reactions
A second cross-coupling reaction investigated with the palladium-functionalized Fe3O4
catalyst is the Sonogashira reaction. This palladium catalyzed reaction is commonly
performed in the presence of a copper(I) salt and an amine base, under an inert
atmosphere.[95,96] When a terminal alkyne functionality is coupled to a substrate with a
carbon-halide bond, copper has a dual role.[97] First, it interacts with the triple bond of the
alkyne, lowering the Highest Occupied Molecular Orbital (HOMO), thereby increasing the
acidity of the acidic proton of the alkyne. The second role of the copper is that it can bond to
the deprotonated alkyne, generating a copper reagent that mediates the transmetalation step of
the Sonogashira reaction. While a copper co-catalyst is added to the reaction to increase
reactivity, the presence of copper can result in the formation of alkyne dimers.[98,99] This
leads to the Glaser coupling reaction, which is an undesired formation of homocoupling
products of acetylene derivatives upon oxidation. As a result, when running a Sonogashira
reaction with a copper co-catalyst, it is necessary to run the reaction in an inert atmosphere to
avoid the unwanted dimerization. Copper-free variations to the Sonogashira reaction have
been developed to avoid the formation of the homocoupling products and trace residues of
Cu+. A successful method of a recyclable catalyst for Cu-free Sonogashira reactions employs
palladium nanoparticles on a pectin support, resulting in a recyclable catalyst using
precipitation.[100] In the report by Khazaei and co-workers, complete conversion of starting
material is obtained with short reaction times in DMF and using KOAc as a base, without the
use of copper salts. In our approach, DMF is replaced by d6-DMSO for two reasons. Firstly, it
1
is easier to monitor the reaction using H-NMR, and secondly, the use of DMSO as the
solvent has a positive influence on the yield of the reaction.[101] Recyclable catalysts for the
use in Sonogashira reactions have already been examined, resulting in a catalyst that can only
be recycled using precipitation[100] or using a magnetite support which leads to reaction
times up to 72 h.[102] Therefore, our attempt is to develop an easily recoverable catalyst
using a magnetite support with short reaction times. For the investigation of the Sonogashira
reaction an Fe3O4/oleic acid/Pd nanoparticles is used. The reason for using a different
stabilizer compared to the Suzuki reaction is because of the simplicity of the nanoparticle
synthesis. The synthesis of the Fe3O4/oleic acid nanoparticle requires just one step, while the
synthesis of the Fe3O4/PEG nanoparticle requires two steps, and is in this way less
13

straightforward because it will not result in any advantage for the Sonogashira reaction.
Because side reactions tend to occur in the Sonogashira reaction with the use of copper in air
and because, in our approach, copper is absent, the Sonogashira reaction is attempted in air.
Single column
Scheme 2: Model reaction used for investigation of the Sonogashira reaction.
In a first step, the reaction is investigated and further optimized using the optimal conditions
reported by Khazaei and co-workers.[100] Therefore, an equimolar reaction is chosen using
1-(tert-butyl)-4-iodobenzene and phenylacetylene in air (Scheme 2). The reaction progress is
1
monitored by H-NMR of the crude reaction mixture. The conversion, with respect to the
end-product, is illustrated in Figure 4. It is observed that the conversion of the aryl iodide is
48% after 30 minutes, but the coupling reaction halts. With 1H-NMR is observed that the CH
peak of the phenylacetylene has completely disappeared after this period. The disappearance
of the phenylacetylene without complete conversion means the acetylene undergoes a side
reaction. The peak of the phenylacetylene protons shifts to 7.62 ppm, and this corresponds to
a homo coupled phenylacetylene.
Single column
Figure 4: Conversion in Sonogashira reaction with the substrates in equimolar amounts. Phenylacetylene (0.18 mmol), 1-
(tert-butyl)-4-iodobenzene (0.18 mmol), KOAc (0.27 mmol), Fe3O4/oleic acid/Pd (2.0 mg), d6-DMSO (1.0 mL), temp: 100
1
1
°C, under air. Reaction was monitored by H-NMR. H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as
an internal standard.
Therefore, the cause of the side-reaction is further investigated by heating phenylacetylene in
the absence of the aryl iodide. This shows if the catalyst enables the side reaction and to what
extent. In the first reaction identical conditions as the Sonogashira reaction are used without
the aryl iodide (Scheme 3). The second reaction is similar to the first, but no KOAc is added.
In the third reaction, unfunctionalized Fe3O4 nanoparticles are employed, these are of the
14

same batch as the palladium functionalized catalyst but consist solely of magnetite. The
fourth reaction is done without the addition of nanoparticles in the solvent and thus is
performed in d6-DMSO with the addition of KOAc (Scheme 3).
Single column
Scheme 3: Investigation of the side-reaction of phenylacetylene with A) Fe3O4/oleic acid/Pd with base, B) Fe3O4/oleic
acid/Pd without base, C) Fe3O4/oleic acid with base, D) d6-DMSO and base.
Single column
Figure 5: The conversions of the reaction with phenylacetylene in function of time. A) Phenylacetylene (0.18 mmol), KOAc
(0.27 mmol), Fe3O4/oleic acid/Pd (2.0 mg), d6-DMSO (1.0 mL), temp: 100 °C, under air. B) Phenylacetylene (0.18 mmol),
Fe3O4/oleic acid/Pd (2.0 mg), d6-DMSO (1.0 mL), temp: 100 °C, under air. C) Phenylacetylene (0.18 mmol), KOAc (0.27
mmol), Fe3O4/oleic acid (2.0 mg), d6-DMSO (1.0 mL), temp: 100 °C, under air. D) Phenylacetylene (0.18 mmol), KOAc
1
1
(0.27 mmol), d6-DMSO (1.0 mL), temp: 100 °C, under air. Reaction was monitored by H-NMR. H-NMR conversions are
calibrated using 1,3,5-trimethoxybenzene as an internal standard.
As depicted in Figure 5, the side reaction is catalyzed only by palladium, because no reaction
occurs with unfunctionalized Fe3O4. The conversion of the reaction with the palladium
catalyst and KOAc is 44% after 4h, the reaction occurs to a smaller extent in the absence of a
base. Hence, the homocoupling of phenylacetylene is an important side reaction to consider
in the thermal Sonogashira reaction in an air atmosphere without a Cu+I source. The
phenylacetylene reacts with itself, preventing high conversions to be obtained in the cross-
15

coupling reaction with the aryl halide. Because of the occurrence of the side reaction of
phenylacetylene, a large excess (10 equivalents) of the phenylacetylene is used in next
reactions. This results in high conversions. However, a reduction of the catalyst loading
would be beneficial in the Sonogashira reaction. Therefore, the catalyst loading in the
reaction is investigated by doing the reaction with various amounts of catalyst (Figure 6). As
expected, the conversion is the least for a catalyst loading of 0.1 mg. Little difference is
noticeable between the use of 2 mg, 1 mg, 0.5 mg and 0.3 mg of nanoparticles, but full
conversion is obtained using an amount of 2 mg of nanoparticles.
Single column
Figure 6: Effect of the amount of Fe3O4/oleic acid/Pd on the Sonogashira cross-coupling. 1-(tert-butyl)-4-iodobenzene (0.18
mmol), phenylacetylene (1.8 mmol), KOAc (0.27 mmol), Fe3O4/oleic acid/Pd, d6-DMSO (1.0 mL), temp: 100 °C, under air.
1
1
Reaction was monitored by H-NMR. H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as an internal
standard.
Further optimization was possible by investigating the temperature dependence of the
reaction progress. For this, the reaction is done with the optimized catalyst loading at various
temperatures. The temperatures investigated are 30 °C, 55 °C, 80 °C, 100 °C, and 120 °C
(Figure 7). The Fe3O4/oleic acid/Pd nanoparticles are isolated after each reaction and washed
to prepare them for a second reaction at 100 °C (Figure 8). This second reaction is done to
determine the catalyst degradation at each temperature in the first catalytic reaction. A
comparison of the conversions of starting material in the second reaction gives a measure for
the deterioration of the catalyst and determines which temperature is optimal for the reaction.
16

Single column
Figure 7: Effect of the temperature on the Sonogashira cross-coupling. 1-(tert-butyl)-4-iodobenzene (0.18 mmol),
phenylacetylene (1.8 mmol), KOAc (0.27 mmol), Fe3O4/oleic acid/Pd (2.0 mg), d6-DMSO (1.0 mL), under air. Reaction was
monitored by 1H-NMR. 1H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as an internal standard.
As depicted in Figure 7, the reaction goes to completion the quickest at 120 °C. At this
temperature, the conversion is already 92% after 30 minutes and 100% after 1h. However,
when the catalyst is recovered and reused at a temperature of 100 °C, it is observed that the
catalyst has lost activity (Figure 8). When the first reaction is applied at a temperature of 100
°C or lower, the catalyst does not show a loss in activity in the second reaction applied at 100
°C (Figure 8). This in combination with the fact when the first reaction applied at 100 °C,
results in higher conversions compared to lower temperatures, an optimum is obtained at 100
°C. Therefore, further reactions are done with a reaction time of 3 h and a temperature of 100
°C.
Single column
Figure 8: Conversion at 100 °C after 4 h with recovered nanoparticles of the reactions at various temperatures.
To explore the scope and limitations of the optimized reaction conditions for the copper-free
Sonogashira reaction in air, a range of aryl halides and phenylacetylenes were investigated
(Table 4). In general, all of the coupling products are obtained in good to excellent yields
17

using the optimal conditions. When a comparison is made between the Sonogashira and the
Suzuki cross-coupling reaction, there can be observed that the Sonogashira reaction demands
harsher conditions. The Sonogashira reaction can only be performed at high temperatures and
no conversion is obtained at 30 °C (Figure 7), while for the Suzuki reaction full conversion
was observed at room temperature with approximately the same Pd-loading. It is also
observed that a six times larger Pd-loading is necessary to reach full conversion for the
Sonogashira reaction alongside a reaction time, which is four times longer, at a higher
temperature. It should also be mentioned that the Sonosgashira reaction is only possible for
iodinated compounds, while the Suzuki reaction can also be performed using brominated
compounds. Nevertheless, almost all end-products are obtained with a conversion of 100 %
using the optimized conditions in the Sonogashira reaction, while this is not observed for the
Suzuki reaction. The only exception is the reaction performed with 4-iodoaniline, which only
has a conversion of 62 % in the Sonogashira reaction, due to its electron-donating character
(Table 4).[103] Furthermore, the sterically hindered effect of substituent-bearing aryl iodides
was also investigated. The results show that sterically hindered aryl iodides are also highly
reactive using the optimized reaction conditions (Table 4). It should be mentioned that in
comparison with the Suzuki reaction, no influence of the phenoxide was observed. This
results in an equal conversion of entry 3 and 10. The electronic nature of substituents of
phenylacetylene is also investigated. To decrease the acidity of the acetylene proton,
substrates with an electron-donating substituent are used. The results show that the
phenylacetylenes couple readily with 1-(tert-butyl)-4-iodobenzene, resulting in full
conversion (Table 4).
Table 4: The scope of Sonogashira cross-coupling reactions with different substituents.
Entry
Reactant 1
Reactant 2
Product
Conversionb
1
2
3
4
5
6
100 %
100 %
100 %
0 %
0 %
0 %
18

7
8
0 %
100 %
9
100 %
100 %
100 %
10
11
12
13
14
62 %
100 %
100 %
100 %
15
aReaction conditions: reactant 1 (0.18 mmol), reactant 2 (1.8 mmol),
KOAc (0.27 mmol), Fe3O4/oleic acid/Pd (2.0 mg), d6-DMSO (1.0 mL),
temp: 100 °C, 4h, under air.
b1H-NMR conversions using 1,3,5-trimethoxybenzene as an internal standard.
The recyclability of the supported palladium catalyst was also surveyed. Catalyst recycling
was performed, at the optimized reaction conditions, to evaluate the reusability of the
Fe3O4/oleic acid/Pd in the Sonogashira reaction (Figure 9). Therefore, after each cycle, an
external magnet is applied to the reaction vessel separating the catalyst from the reaction
mixture and the resulting supernatant can be decanted, which minimizes the loss of catalyst
during separation. Despite the fact that another stabilizer is attached on the nanoparticle
compared to the nanoparticle used for the Suzuki-Miyaura reaction, the magnetic separation
occurs equally fast and is done in a few seconds. The catalyst system is then subsequently
washed 2 times with 2 mL of DMSO and 2 times with 2 mL of 1,4-dioxane. After being
dried, the catalyst system could be reused directly without further purification. As shown in
Figure 9 there is some loss of catalyst activity in the first reaction. In this first reaction, the
conversion towards the end-product is 98%, which then declines to 94% in the second
reaction with the catalyst. In the third reaction, the conversion decreases to 77%, a further
drop in conversion is noticed in the fourth and fifth reaction with the recovered nanoparticles.
It should be mentioned that the loss in nanoparticles after each recovery was again taken into
account, similar as for the Suzuki reaction. TEM images of the Fe3O4/oleic acid/Pd
nanoparticles before and after the reactions were recorded, in order to investigate the change
in morphology of these nanoparticles as a result of the reaction (Figure 10). According to
Figure 10, almost no difference in morphology was observed between the nanoparticles
19

before and after the reaction. Therefore, there can be concluded that the decrease in
conversion is not due to a change in morphology of the nanoparticles, but most probably due
to the leaching of the palladium, which is investigated and confirmed by ICP-AES
(Supplementary materials Table S2). When a comparison is made between the recyclability
of the Suzuki cross-coupling reaction using Fe3O4/PEG/Pd nanoparticles and of the
Sonogashira reaction using Fe3O4/oleic acid/Pd nanoparticles, a different trend is observed.
While the conversion of the Suzuki reaction shows a fast decrease for the first recovery, the
recovery performed for the Sonogashira reaction shows only a significant decrease in
conversion after the 3rd cycle. It should also be mentioned that the leaching of Pd is 12.6 %,
4.7 %, 10.8 %, 0.5 and 0.7 % for the 1st, 2nd, 3rd, 4th, and 5th cycle respectively. When these
values are compared to the Suzuki reaction, there can be observed that there is a much lower
Pd-leaching for the Suzuki reaction. These dissimilar results can be explained by taken the
total amount of Pd present in the reaction into account. While, for example, the total amount
of Pd present in the second performed Sonogashira reaction exists out of 0.025 mg/mL, the
total amount of Pd present in the second performed Suzuki reaction is only 0.0043 mg/mL.
There should also be mentioned that a decrease of 50 % in nanoparticle amount, resulting in a
decrease of 50 % of Pd, results in a decrease of only 10 % of conversion (Figure 6). Another
possible explanation for the dissimilar results of the recuperation experiments between the
Suzuki and the Sonogashira reaction is the possibility that the oleic acid stabilized
nanoparticles have a higher stability in dry conditions compared to the PEG stabilized
nanoparticles. Bringing the nanoparticles to the dry state is inevitable, so the correct amount
of nanoparticles can be determined for the next cycle. Considering the results, it can be
concluded that the catalyst can be employed up to three times without significant loss of
activity.
Single column
20

Figure 9: The reusability of the Fe3O4/oleic acid/Pd NPs for the Sonogashira reaction using 1-(tert-butyl)-4-iodobenzene
and phenylacetylene as substrates using 2.0 mg of nanoparticles/mL solvent (d6-DMSO) with adapting the amount of
substrates to the amount of nanoparticles recycled after each cycle.
Figure 10: TEM images of the Fe3O4/oleic acid/Pd nanoparticle catalysts before (left) and after (right) the Sonogashira
reaction.
3.4 Stille cross-coupling reactions
A third and last cross-coupling reaction investigated is the Stille reaction. The Stille coupling
is an organometallic reaction that has the advantage of being tolerant to a lot of functional
groups, without needing a base in the cross-coupling reaction but is notorious for its use of
toxic tin reagents, that are hard to remove after the reaction.[104,105] Because no prior work
was done with this catalyst in the Stille reaction, a first step in the investigation is finding
optimal solvent conditions (Table 5). The reaction of 1-(tert-butyl)-4-iodobenzene with
trimethyl(phenyl)stannane is used for initial studies as a model reaction. Because relatively
short reaction times are optimal for the investigation, a predetermined reaction time of 4 h is
used. A reaction temperature was chosen of 100 °C since this temperature is commonly used
for the use in Stille reactions.[6,106] Although all reactions have similar low conversions, of
the reactions in the different solvents, 1,4-dioxane has the highest conversion. Therefore, the
following reactions are done in 1,4-dioxane.
Table 5: The effect of the solvent on the Stille reaction.
Entry
Solvent
Toluene
Toluene/DMF (1:1)
DMF
Conversiona
1
2
3
4
6 %
6 %
8 %
1,4 Dioxane
10 %
a1H-NMR conversions using 1,3,5-trimethoxybenzene as an internal standard.
Because low conversions are obtained in the Stille reaction with 1-(tert-butyl)-4-iodobenzene
as the substrate in the investigation of the solvent conditions, the substrate is changed to
methyl-4-iodobenzoate. This aryl iodide contains an ester substituent in the para position,
which results in an activated carbon-iodide bond. This is due to the stronger electron-
withdrawing effect of an ester compared to the tertiary butyl group. Further, a longer reaction
time is applied in order to investigate the progress of the reaction in time (Figure 11), and to
21

understand the low yield obtained after a reaction time of 4 h at 100 °C. The conversion was
found to be approximately 15% higher compared with the less reactive 1-(tert-butyl)-4-
iodobenzene but further conversion stagnates after 2 h. This stop in conversion could indicate
that the catalyst degrades during the reaction or gets poisoned. To further investigate this
behavior the catalyst is recovered and reused in an identical reaction. The nanoparticles are
recovered with a magnet and washed 4 times with 2 mL 1,4-dioxane to remove all traces of
the reaction mixture. The recovered nanoparticles are then dried under vacuum to prepare
them for the next catalytic reaction. Adapting the quantity of the reactants to the mass of
recovered catalyst, the reaction is repeated with the same nanoparticles. In the second
reaction, there is no reaction occurring at all. The lack of any conversion of starting material
clearly indicates that the Fe3O4 supported catalyst is degraded or poisoned in the first
catalytic reaction. Therefore, the cause of the degradation is further investigated.
Single column
Figure 11: Conversion of the Stille reaction in function of time at “optimal” solvent conditions. Methyl 4-iodobenzoate (1.44
mmol), trimethyl(phenyl)stannane (1.73 mmol), Fe3O4/oleic acid/Pd (8.0 mg), 1,4-dioxane (8.0 mL), temp: 100 °C, under
argon atmosphere. 1H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as an internal standard.
Single column
Scheme 4: Experiments for investigation of catalyst poisoning with A) trimethyl(phenyl)stannane B) trimethyltin iodide.
22

Because the conversion halts completely around 25% and does not increase after 2 h, it is
speculated that the catalyst is poisoned either by trimethyl tin iodide, which is the by-product
formed during the reaction or by the trimethyl(phenyl)stannane (Scheme 4). The poisoning is
investigated by testing the two potential poisons. First, the Fe3O4/oleic acid/Pd nanoparticles
are stirred at 100 °C for a period of 1h in 1,4-dioxane with the addition of the potential poison
(Scheme 4). Afterwards, the nanoparticles, that were stirred with one of the poisons, are
recovered. To remove all traces of the tin compounds in the solvent, the nanoparticles are
washed 3 times with 1,4-dioxane. Thereafter, the recovered nanoparticles are dried and, with
adapting the scale of the solvent and reactants to the mass of the recovered nanoparticles, a
Stille reaction is initiated. The conversion of the two reactions with the recovered
nanoparticles is followed in time, and even after a reaction time of 5 h no conversion was
obtained. The nanoparticles stirred both with trimethyl tin iodide as
trimethyl(phenyl)stannane, have lost their catalytic activity. Therefore, the conclusion of the
experiment is that both the tin reagent as the tin by-product, that is formed in the Stille
reaction, are poisons for the catalyst. Because of this, it will be impossible to obtain optimal
conditions in the Stille reaction using the Fe3O4/oleic acid/Pd catalyst. Because both the
starting reagent and the product are a poison for the catalyst, the activity ceases before high
conversion of the starting material is obtained and the catalyst is not suitable for reuse.
3.5 The influence of Au on the catalytic activity
The use of gold as co-catalyst in cross-coupling reactions has already widely been described
in literature.[82] M.G. Speziali and co-workers are the only ones describing the influence of
gold on the catalytic properties of a reusable Pd catalyst for the use in Suzuki-reactions.[107]
Knowing the importance and positive effect of gold present as Au(III) or Au(I) together with
Pd in the Sonogashira reaction,[82,108–110] our goal is to combine this effect with the
advantage of a reusable catalyst. Therefore, a core-shell nanoparticle was made consisting of
a magnetite core surrounded by a golden shell, stabilized with PEG and functionalized with
Pd. The reason for the use of PEG as stabilizer is the result of the fact that the bonding of
oleic acid to the golden shell of the nanoparticle is much weaker compared to a thiolated
PEG-chain. Second, the synthesis of the Fe3O4/Au core-shell nanoparticle is performed using
ascorbic acid as stabilizer, therefore it is not possible to attach a group as for example oleic
acid, which has an approximately equal binding energy. It should also be mentioned that the
structure of the PEG-ligands protects the core-shell nanoparticles from aggregation through
steric hindrance. Due to the presence of the magnetite core, it is possible to reuse the catalyst
23

system in the Sonogashira reaction as described in the previous section (Figure 9). Because it
is aimed to investigate the influence and the possible improvement on the yield of the
reaction, reaction conditions are used resulting in a yield of 0 %. Therefore, the reaction of 1-
(tert-butyl)-4-iodobenzene with phenylacetylene at a temperature of 30 °C and a reaction
time of 3 h is used (Figure 7). The influence of gold on the catalytic activity is investigated
by testing the two catalytic systems in parallel for three times (Figure 12). The catalyst
systems used are the already described and investigated Fe3O4/oleic acid/Pd and, second, the
gold/magnetite core-shell nanoparticle functionalized with Pd (Fe3O4/Au/PEG/Pd). To
investigate the influence of the gold, the same amount of Pd is used, resulting in the use of 2
mg Fe3O4/oleic acid/Pd and 0.56 mg Fe3O4/Au/PEG/Pd.
Single column
Figure 12: Conversion obtained using Fe3O4/Pd compared with Fe3O4/Au/Pd using the model Sonogashira reaction. 1-(tert-
butyl)-4-iodobenzene (0.18 mmol), phenylacetylene (1.8 mmol), KOAc (0.27 mmol), Pd functionalized nanoparticles, d6-
1
DMSO (1.0 mL), temp: 30 °C, 4h, under air. H-NMR conversions are calibrated using 1,3,5-trimethoxybenzene as an
internal standard.
As depicted in Figure 12, the presence of gold has indeed a positive influence on the yield of
the reaction and activity of the catalyst. Due to this, it is not only possible to perform
Sonogashira reactions at a shorter reaction time, but also at much lower reaction temperature,
resulting in a more time- and energy-efficient reaction.
4 Conclusion
In summary, it is possible to synthesize a less time consuming recyclable Pd catalyst by the
use of commercially available stabilizers and Pd-sources, and without the use of toxic
reducing agents. It is found that ligand-free palladium functionalized magnetite nanoparticles
can be used in Suzuki cross-coupling reactions as well as in Sonogashira reactions under
ligand-free and aerobic conditions. By varying the stabilizer present on the magnetite
nanoparticle it is possible to make a nanoparticle that is suitable for the implementation in a
24

reaction making use of an aprotic solvent, like the Sonogashira reaction, as well as it is
possible to make them usable in a Suzuki reaction, which is a reaction using protic solvents.
By analyzing the scope of the Suzuki reaction, it can be concluded that a substituent with a
high affinity for the magnetite nanoparticle has a positive influence on the yield of the
reaction. This phenomenon was not observed for the Sonogashira reaction. It should be
noticed that although excellent yields are obtained in the Sonogashira reaction, harsh reaction
conditions are necessary in comparison with the Suzuki reaction, resulting in higher
temperatures and a higher catalyst loading. Although good yields were obtained in the
Suzuki- and Sonogashira-reaction it can be concluded that it is not possible to use the as
described Pd-functionalized nanoparticles in the Stille reaction. This was found to be due to
poisoning of the catalytic system by the tin-reagent as by the tin by-product. The supported
catalyst developed in the study also had the advantage of being recoverable with the simple
application of a magnetic field. Furthermore, the influence of Au as co-catalyst is described
and leads to a positive influence in the yield of the Sonogashira reactions, resulting in a
reusable catalyst with short reaction times and low reaction temperatures. All these
advantages make this an eco-friendly and economic catalyst for the use in Suzuki reactions
and Sonogashira reactions with added value in achieving green chemistry goals.
Acknowledgments
The authors are grateful to the Onderzoeksfonds KU Leuven/Research Fund KU Leuven
and the Special Research Fund (BOF) of the Flemish community as a C3 project for financial
support. AB acknowledges financial support from the Agency for Innovation and Technology
in Flanders (IWT). Thanks to Hendrik Reynders for the TEM measurements.
Notes and references
[1] V. L. Budarin, P. S. Shuttleworth, J. H. Clark, R. Luque, Industrial Applications of C-
C Coupling Reactions, Curr. Org. Synth. 7 (2010) 614–627.
doi:10.2174/157017910794328529.
[2] A. Dumrath, C. Lübbe, M. Beller, Palladium-Catalyzed Cross-Coupling Reactions -
Industrial Applications, in: Palladium-Catalyzed Coupling React., Wiley-VCH Verlag
GmbH & Co. KGaA, Weinheim, Germany, 2013: pp. 445–489.
doi:10.1002/9783527648283.ch12.
[3] M. Picquet, Organometallics as catalysts in the fine chemical industry, Platin. Met.
Rev. 57 (2013) 272–280. doi:10.1595/147106713X672320.
[4] C.C.C.J. Seechurn, M.O. Kitching, T.J. Colacot, V. Snieckus, Cross-Coupling
Palladium-Catalyzed Cross-Coupling: A Historical Contextual Perspective to the 2010
25

Nobel Prize, Angew. Chem. Int. Ed. Engl. 51 (2012) 5062–5085.
doi:10.1002/anie.201107017.
[5] N. Miyaura, A. Suzuki, Palladium-Catalyzed Cross-Coupling Reactions of
Organoboron Compounds, Chem. Rev. 95 (1995) 2457–2483.
doi:10.1021/cr00039a007.
[6] W.-J. Zhou, K.-H. Wang, J.-X. Wang, Pd(PPh3)4 -PEG 400 Catalyzed Protocol for the
Atom-Efficient Stille Cross-Coupling Reaction of Organotin with Aryl Bromides, J.
Org. Chem. 74 (2009) 5599–5602. doi:10.1021/jo9005206.
[7] L. Del Valle, J.K. Stille, L.S. Hegedus, Palladium-catalyzed coupling of allylic
acetates with aryl- and vinylstannanes, J. Org. Chem. 55 (1990) 3019–3023.
doi:10.1021/jo00297a014.
[8] G.A. Molander, S.L.J. Trice, S.M. Kennedy, Scope of the Two-Step, One-Pot
Palladium-Catalyzed Borylation/Suzuki Cross-Coupling Reaction Utilizing Bis-
Boronic Acid, J. Org. Chem. 77 (2012) 8678–8688. doi:10.1021/jo301642v.
[9] C. Sun, B. Potter, J.P. Morken, A Catalytic Enantiotopic-Group-Selective Suzuki
Reaction for the Construction of Chiral Organoboronates, J. Am. Chem. Soc. 136
(2014) 6534–6537. doi:10.1021/ja500029w.
[10] V. V Namboodiri, R.S. Varma, Microwave-accelerated Suzuki cross-coupling reaction
in polyethylene glycol (PEG), Green Chem. 3 (2001) 146–148. doi:10.1039/b102337n.
[11] T. Baran, A new chitosan Schiff base supported Pd(II) complex for microwave-
assisted synthesis of biaryls compounds, J. Mol. Struct. 1141 (2017) 535–541.
doi:10.1016/j.molstruc.2017.03.122.
[12] N. Yılmaz Baran, T. Baran, A. Menteş, Production of novel palladium nanocatalyst
stabilized with sustainable chitosan/cellulose composite and its catalytic performance
in Suzuki-Miyaura coupling reactions, Carbohydr. Polym. 181 (2018) 596–604.
doi:10.1016/j.carbpol.2017.11.107.
[13] T. Baran, Pd(0) nanocatalyst stabilized on a novel agar/pectin composite and its
catalytic activity in the synthesis of biphenyl compounds by Suzuki-Miyaura cross
coupling reaction and reduction of o-nitroaniline, Carbohydr. Polym. 195 (2018) 45–
52. doi:10.1016/j.carbpol.2018.04.064.
[14] M. Nasrollahzadeh, Advances in Magnetic Nanoparticles-Supported Palladium
Complexes for Coupling Reactions, Molecules. 23 (2018) 2532.
doi:10.3390/molecules23102532.
[15] M. Nasrollahzadeh, M. Atarod, M. Alizadeh, A. Hatamifard, S. MohammadSajadi,
Recent Advances in the Application of Heterogeneous Nanocatalysts for Sonogashira
Coupling Reactions, Curr. Org. Chem. 21 (2017) 708–749.
https://www.ingentaconnect.com/contentone/ben/coc/2017/00000021/00000008/art000
07 (accessed September 24, 2019).
[16] D. Astruc, F. Lu, J.R. Aranzaes, Nanoparticles as Recyclable Catalysts: The Frontier
between Homogeneous and Heterogeneous Catalysis, Angew. Chemie Int. Ed. 44
(2005) 7852–7872. doi:10.1002/anie.200500766.
[17] E. Farnetti, R. Di Monte, J. Kašpar, Homogeneous and Heterogeneous Catalysis, Life
Support Syst. 2 (1999) 10.
26

[18] H.A. Elazab, A.R. Siamaki, S. Moussa, B.F. Gupton, M.S. El-Shall, Highly efficient
and magnetically recyclable graphene-supported Pd/Fe3O4 nanoparticle catalysts for
Suzuki and Heck cross-coupling reactions, Appl. Catal. A Gen. 491 (2015) 58–69.
doi:10.1016/J.APCATA.2014.11.033.
[19] H. Veisi, P. Sarachegol, S. Hemmati, Palladium(II) anchored on polydopamine coated-
magnetic nanoparticles (Fe3O4@PDA@Pd(II)): A heterogeneous and core–shell
nanocatalyst in Buchwald–Hartwig C–N cross coupling reactions, Polyhedron. 156
(2018) 64–71. doi:10.1016/J.POLY.2018.09.019.
[20] H. Veisi, J. Gholami, H. Ueda, P. Mohammadi, M. Noroozi, Magnetically palladium
catalyst stabilized by diaminoglyoxime-functionalized magnetic Fe3O4 nanoparticles
as active and reusable catalyst for Suzuki coupling reactions, J. Mol. Catal. A Chem.
396 (2015) 216–223. doi:10.1016/J.MOLCATA.2014.10.012.
[21] H. Veisi, A. Amini Manesh, N. Eivazi, A.R. Faraji, Palladium nanoparticles supported
on 1,3-dicyclohexylguanidine functionalized mesoporous silica SBA-15 as highly
active and reusable catalyst for the Suzuki–Miyaura cross-coupling reaction, RSC
Adv. 5 (2015) 20098–20107. doi:10.1039/C4RA14668A.
[22] V. Polshettiwar, C. Len, A. Fihri, Silica-supported palladium: Sustainable catalysts for
cross-coupling reactions, Coord. Chem. Rev. 253 (2009) 2599–2626.
doi:10.1016/J.CCR.2009.06.001.
[23] Z. Zheng, H. Li, T. Liu, R. Cao, Monodisperse noble metal nanoparticles stabilized in
SBA-15: Synthesis, characterization and application in microwave-assisted Suzuki–
Miyaura coupling reaction, J. Catal. 270 (2010) 268–274.
doi:10.1016/J.JCAT.2010.01.004.
[24] P. Han, X. Wang, X. Qiu, X. Ji, L. Gao, One-step synthesis of palladium/SBA-15
nanocomposites and its catalytic application, J. Mol. Catal. A Chem. 272 (2007) 136–
141. doi:10.1016/J.MOLCATA.2007.03.006.
[25] S. Moussa, A.R. Siamaki, B.F. Gupton, M.S. El-Shall, Pd-Partially Reduced Graphene
Oxide Catalysts (Pd/PRGO): Laser Synthesis of Pd Nanoparticles Supported on PRGO
Nanosheets for Carbon–Carbon Cross Coupling Reactions, ACS Catal. 2 (2012) 145–
154. doi:10.1021/cs200497e.
[26] A.R. Siamaki, A.E.R.S. Khder, V. Abdelsayed, M.S. El-Shall, B.F. Gupton,
Microwave-assisted synthesis of palladium nanoparticles supported on graphene: A
highly active and recyclable catalyst for carbon–carbon cross-coupling reactions, J.
Catal. 279 (2011) 1–11. doi:10.1016/J.JCAT.2010.12.003.
[27] B.F. Mohazzab, B. Jaleh, Z. Issaabadi, M. Nasrollahzadeh, R.S. Varma, Stainless steel
mesh-GO/Pd NPs: catalytic applications of Suzuki-Miyaura and Stille coupling
reactions in eco-friendly media , Green Chem. 21 (2019). doi:10.1039/c9gc00889f.
[28] T. Baran, N. Yılmaz Baran, A. Menteş, Preparation, structural characterization, and
catalytic performance of Pd(II) and Pt(II) complexes derived from cellulose Schiff
base, J. Mol. Struct. 1160 (2018) 154–160. doi:10.1016/j.molstruc.2018.01.074.
[29] N. Yılmaz Baran, T. Baran, A. Menteş, Fabrication and application of cellulose Schiff
base supported Pd(II) catalyst for fast and simple synthesis of biaryls via Suzuki
coupling reaction, Appl. Catal. A Gen. 531 (2017) 36–44.
doi:10.1016/j.apcata.2016.12.005.
27

[30] T. Baran, I. Sargin, A. Menteş, M. Kaya, Exceptionally high turnover frequencies
recorded for a new chitosan-based palladium(II) catalyst, Appl. Catal. A Gen. 523
(2016) 12–20. doi:10.1016/j.apcata.2016.05.012.
[31] T. Baran, I. Sargin, M. Kaya, A. Menteş, Green heterogeneous Pd(II) catalyst
produced from chitosan-cellulose micro beads for green synthesis of biaryls,
Carbohydr. Polym. 152 (2016) 181–188. doi:10.1016/j.carbpol.2016.06.103.
[32] T. Baran, T. Inanan, A. Menteş, Synthesis, characterization, and catalytic activity in
Suzuki coupling and catalase-like reactions of new chitosan supported Pd catalyst,
Carbohydr. Polym. 145 (2016) 20–29. doi:10.1016/j.carbpol.2016.03.019.
[33] T. Baran, I. Sargin, M. Kaya, A. Menteş, An environmental catalyst derived from
biological waste materials for green synthesis of biaryls via Suzuki coupling reactions,
J. Mol. Catal. A Chem. 420 (2016) 216–221. doi:10.1016/j.molcata.2016.04.025.
[34] T. Baran, E. Açıksöz, A. Menteş, Carboxymethyl chitosan Schiff base supported
heterogeneous palladium(II) catalysts for Suzuki cross-coupling reaction, J. Mol.
Catal. A Chem. 407 (2015) 47–52. doi:10.1016/j.molcata.2015.06.008.
[35] T. Baran, N. Yılmaz Baran, A. Menteş, Sustainable chitosan/starch composite material
for stabilization of palladium nanoparticles: Synthesis, characterization and
investigation of catalytic behaviour of Pd@chitosan/starch nanocomposite in Suzuki-
Miyaura reaction, Appl. Organomet. Chem. 32 (2018) e4075. doi:10.1002/aoc.4075.
[36] T. Baran, N. Yılmaz Baran, A. Menteş, An easily recoverable and highly reproducible
agar-supported palladium catalyst for Suzuki-Miyaura coupling reactions and
reduction of o-nitroaniline, Int. J. Biol. Macromol. 115 (2018) 249–256.
doi:10.1016/j.ijbiomac.2018.04.057.
[37] T. Baran, Ultrasound-accelerated synthesis of biphenyl compounds using novel Pd(0)
nanoparticles immobilized on bio-composite, Ultrason. Sonochem. 45 (2018) 231–237.
doi:10.1016/j.ultsonch.2018.03.017.
[38] T. Baran, N. Yılmaz Baran, A. Menteş, A new air and moisture stable robust bio-
polymer based palladium catalyst for highly efficient synthesis of biaryl compounds,
Appl. Organomet. Chem. 32 (2018) e4076. doi:10.1002/aoc.4076.
[39] T. Baran, A. Menteş, Construction of new biopolymer (chitosan)-based pincer-type
Pd(II) complex and its catalytic application in Suzuki cross coupling reactions, J. Mol.
Struct. 1134 (2017) 591–598. doi:10.1016/j.molstruc.2017.01.005.
[40] T. Baran, Practical, economical, and eco-friendly starch-supported palladium catalyst
for Suzuki coupling reactions, J. Colloid Interface Sci. 496 (2017) 446–455.
doi:10.1016/j.jcis.2017.02.047.
[41] V.I. Sokolov, E.G. Rakov, N.A. Bumagin, M.G. Vinogradov, New Method to Prepare
Nanopalladium Clusters Immobilized on Carbon Nanotubes: A Very Efficient Catalyst
for Forming Carbon-Carbon Bonds and Hydrogenation, Fullerenes, Nanotub. Carbon
Nanostructures. 18 (2010) 558–563. doi:10.1080/1536383X.2010.488077.
[42] C. Deraedt, D. Wang, L. Salmon, L. Etienne, C. Labrugère, J. Ruiz, D. Astruc, Robust,
Efficient, and Recyclable Catalysts from the Impregnation of Preformed Dendrimers
Containing Palladium Nanoparticles on a Magnetic Support, ChemCatChem. 7 (2015)
303–308. doi:10.1002/cctc.201402775.
28

[43] C. Deraedt, L. Salmon, D. Astruc, “Click” Dendrimer-Stabilized Palladium
Nanoparticles as a Green Catalyst Down to Parts per Million for Efficient C-C Cross-
Coupling Reactions and Reduction of 4-Nitrophenol, Adv. Synth. Catal. 356 (2014)
2525–2538. doi:10.1002/adsc.201400153.
[44] Lei Wu, Bao-Lin Li, Yi-Yong Huang, Hai-Feng Zhou, Yan-Mei He, Qing-Hua Fan,
Phosphine Dendrimer-Stabilized Palladium Nanoparticles, a Highly Active and
Recyclable Catalyst for the Suzuki−Miyaura Reaction and Hydrogenation, Org. Lett. 8
(2006) 3605–3608. doi:10.1021/OL0614424.
[45] Karical R. Gopidas, James K. Whitesell, M.A. Fox, Synthesis, Characterization, and
Catalytic Applications of a Palladium-Nanoparticle-Cored Dendrimer, Nano Lett. 3
(2003) 1757–1760. doi:10.1021/NL0348490.
[46] Y. Zheng, P.D. Stevens, Y. Gao, Magnetic Nanoparticles as an Orthogonal Support of
Polymer Resins: Applications to Solid-Phase Suzuki Cross-Coupling Reactions, J.
Org. Chem. 71 (2006) 537–542. doi:10.1021/jo051861z.
[47] N. Yılmaz Baran, T. Baran, A. Menteş, M. Karakışla, M. Saçak, Highly effective and
recoverable Pd(II) catalyst immobilized on thermally stable Schiff base polymer
containing phenol group: Production, characterization and application in Suzuki
coupling reactions, J. Organomet. Chem. 866 (2018) 87–94.
doi:10.1016/J.JORGANCHEM.2018.04.022.
[48] S. Ogasawara, S. Kato, Palladium Nanoparticles Captured in Microporous Polymers:
A Tailor-Made Catalyst for Heterogeneous Carbon Cross-Coupling Reactions, J. Am.
Chem. Soc. 132 (2010) 4608–4613. doi:10.1021/ja9062053.
[49] I. Beletskaya, A. Kashin, I. Khotina, A. Khokhlov, Efficient and Recyclable Catalyst
of Palladium Nanoparticles Stabilized by Polymer Micelles Soluble in Water for
Suzuki-Miyaura Reaction, Ostwald Ripening Process with Palladium Nanoparticles,
Synlett. 10 (2008) 1547–1552. doi:10.1055/s-2008-1078430.
[50] R. Najman, J.K. Cho, A.F. Coffey, J.W. Davies, M. Bradley, Entangled palladium
nanoparticles in resin plugs, Chem. Commun. 47 (2007) 5031–5033.
doi:10.1039/b711978j.
[51] J.A. Bobb, A.A. Ibrahim, M.S. El-Shall, Laser Synthesis of Carbonaceous TiO 2 from
Metal–Organic Frameworks: Optimum Support for Pd Nanoparticles for C–C Cross-
Coupling Reactions, ACS Appl. Nano Mater. 1 (2018) 4852–4862.
doi:10.1021/acsanm.8b01045.
[52] A. Gniewek, J.J. Ziółkowski, A.M. Trzeciak, M. Zawadzki, H. Grabowska, J.
Wrzyszcz, Palladium nanoparticles supported on alumina-based oxides as
heterogeneous catalysts of the Suzuki–Miyaura reaction, J. Catal. 254 (2008) 121–130.
doi:10.1016/J.JCAT.2007.12.004.
[53] M. Lakshmi Kantam, S. Roy, M. Roy, B. Sreedhar, B.M. Choudary, Nanocrystalline
Magnesium Oxide-Stabilized Palladium(0): An Efficient and Reusable Catalyst for
Suzuki and Stille Cross-Coupling of Aryl Halides, Adv. Synth. Catal. 347 (2005)
2002–2008. doi:10.1002/adsc.200505198.
[54] M. Nasrollahzadeh, Z. Issaabadi, M.M. Tohidi, S. Mohammad Sajadi, Recent Progress
in Application of Graphene Supported Metal Nanoparticles in C−C and C−X Coupling
Reactions, Chem. Rec. 18 (2018) 165–229. doi:10.1002/tcr.201700022.
29