Preparation and characterization of a new bis-layered supported ionic liquid catalyst (SILCA) with an unprecedented activity in the Heck reaction

Article abstract of DOI:10.1016/j.jcat.2019.01.029

A new bis-layered supported ionic liquid catalyst (SILCA) loaded with palladium was designed and successfully applied for the Heck reaction of iodobenzene and methyl acrylate. The silica modified catalyst consisting of the first ionic liquid layer – covalently anchored imidazolium bromide – on which the second layer, made of pyridine-carboxylic acid balanced with tetramethylguanidinium cation was attached, resulted in a catalyst with high activity. High turnover frequencies of 22,000 h^?1 were achieved in reactions with a low palladium loading as 0.009 mol %. Lower TOFs, indicating on palladium dimerization was detected when higher amounts were used. The TMG cation had a purpose to recapture and stabilize the Pd nanoparticles thus followed a release and catch mechanism. In order to get a full understanding of the catalyst structure and behaviour, the catalyst was characterized by means of nitrogen physisorption, thermal gravimetric analysis, infrared spectroscopy, scanning electron and transmission electron microscopes, solid-state NMR, X-ray photoelectron spectroscopy and inductively coupled plasma spectroscopy. The catalyst preserved good activity in five cycles.

Full text of DOI:10.1016/j.jcat.2019.01.029

Journal of Catalysis 371 (2019) 35–46
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Preparation and characterization of a new bis-layered supported ionic
liquid catalyst (SILCA) with an unprecedented activity in the Heck
reaction
a
Nemanja Vucetic a, , Pasi Virtanen , Ayat Nuri a,b, Ida Mattsson c, Atte Aho a, Jyri-Pekka Mikkola a,d
,
⇑
Tapio Salmi a
a Åbo Akademi University, Johan Gadolin Process Chemistry Centre, Laboratory of Industrial Chemistry and Reaction Engineering, Biskopsgatan 8, FI-20500 Turku, Finland
b University of Mohaghegh Ardabili, Faculty of Science, Department of Applied Chemistry, Ardabil 56199-11367, Iran
c Åbo Akademi University, Johan Gadolin Process Chemistry Centre, Laboratory of Organic Chemistry, Biskopsgatan 8, FI-20500 Turku, Finland
d Umeå University, Chemical-Biological Centre, Department of Chemistry, Technical Chemistry, SE-90187 Umeå, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
A new bis-layered supported ionic liquid catalyst (SILCA) loaded with palladium was designed and suc-
cessfully applied for the Heck reaction of iodobenzene and methyl acrylate. The silica modified catalyst
consisting of the first ionic liquid layer – covalently anchored imidazolium bromide – on which the sec-
ond layer, made of pyridine-carboxylic acid balanced with tetramethylguanidinium cation was attached,
resulted in a catalyst with high activity. High turnover frequencies of 22,000 hÀ1 were achieved in reac-
tions with a low palladium loading as 0.009 mol %. Lower TOFs, indicating on palladium dimerization was
detected when higher amounts were used. The TMG cation had a purpose to recapture and stabilize the
Pd nanoparticles thus followed a release and catch mechanism. In order to get a full understanding of the
catalyst structure and behaviour, the catalyst was characterized by means of nitrogen physisorption,
thermal gravimetric analysis, infrared spectroscopy, scanning electron and transmission electron micro-
scopes, solid-state NMR, X-ray photoelectron spectroscopy and inductively coupled plasma spectroscopy.
The catalyst preserved good activity in five cycles.
Received 28 October 2018
Revised 21 January 2019
Accepted 21 January 2019
Keywords:
Heck reaction
Supported ionic liquid catalyst
Palladium
Turn-over frequency
Catalyst characterization
Ó 2019 Elsevier Inc. All rights reserved.
1. Introduction
As well as many other organic reactions, the Heck reaction faces
a challenge to abound volatile organic solvents and transfer the
Growing demands for different kinds of molecules in fine-
chemical industry and drug production cannot be any more
satisfied from natural sources, and the question is should it be,
considered that today technologies can offer alternative routes to
different chemicals. From economical aspect, extraction and
isolation of molecules that preserve certain functionalities is cum-
bersome, much cheaper is just to produce them, favorably in–situ.
The Heck reaction catalyzed by the palladium and conducted in the
presence of base presents a significant tool in synthetic chemistry.
It was firstly introduced by Mizoroki [1] and described by Heck [2]
almost half a century ago. The Heck reaction is CAC coupling reac-
tion between various olefins and aryl, vinyl or allyl halides, triflates
or acetates to yield the molecule with preserved olefin double
bond. It is usually conducted in polar solvents at elevated temper-
atures, in slight excess of base and alkene in order to compensate
their losses due to the evaporation [2,3].
reaction into more ecologically acceptable and controllable envi-
ronment. Ionic liquids (ILs) have emerged and with the time
evolved into something that can replace bulk solvents and provide
many other advantages in processes involved with reactions, sep-
aration, electrochemistry etc. In the very beginning [4–8] ionic liq-
uids were ignored and for the first time acknowledged a century
ago [9], at which time they were described as salts, therefore
purely consistent of ions that are in liquid state at temperatures
below 100 °C and with no significant vapor pressure. From that
moment on, they slowly started to draw attention of the scientific
society and got a huge swing 30 years ago when few research
groups started to study and promote them [10,11,20,12–19]. Ionic
liquids are often characterized to be polar, water immiscible, ther-
mally stable, non-volatile, non-flammable and task-specific com-
pounds with wide liquidus range. However, in line with recent
studies these too generalized properties which were mainly
derived for most popular ILs based on imidazolium cation had to
be updated [21]. Features as such cannot be prescribed to complete
order of these organic compounds, especially when one becomes
⇑
Corresponding author.
E-mail address: nvucetic@abo.fi (N. Vucetic).
https://doi.org/10.1016/j.jcat.2019.01.029
0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

36
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
aware of the tremendous number of ILs that can be designed. Even
though each group of ILs has to be separately defined, there are few
features which unify all of them: they consist of ions only, their
melting temperature is lower than the decomposition temperature
and they can be ‘task-specified’ to meet specific requirements. Fur-
thermore they can be utilized in catalyst design.
ride were purchased from Alfa Aesar. N-methyl-2-pyrrolidone,
imidazole, 6-bromo-nicotinic acid, 1,1,3,3-tetramethylguanidine,
sulfuric acid, sodium hydroxide, sodium bicarbonate and dimethyl
ether were supplied by Sigma Aldrich. Other chemicals were
obtained from common commercial sources. All the chemicals
were used as obtained, without further purification.
The first Heck reaction in ionic liquids was conducted in
tetraalkylammoinum and phosphonium salts without addition of
phosphonium ligands [22]. Compared to usual organic solvents
for this reaction such as DMF, DMA, NMP or acetonitrile, the ionic
liquid [Bu4N]Br was found to be a good medium, giving competi-
tive results [23,24]. Commonly used imidazolium [C4mim]PF6
and n-hexylpyridinium [C6Pyr]PF6 ionic liquids also present an
attractive alternative [25]. Based on these findings during the next
decade new compounds like tetramethylguanidine acetate and
hexafluorophosphate ([HTMG]OAc, [HTMG]PF6) [26], pyridyl-
functionalized imidazolium-based [27] and ethanolamine-
functionalized [28–30] ionic liquids were synthesized and applied
in different Heck reactions with various palladium catalysts and
alkaline compounds. It was concluded that ionic liquids are bene-
ficial for the product isolation and with ability to act as a base and
ligand in Heck chemistry.
Parallel to ILs development, new ligands for Pd were investigated
giving C, N, C pincer bis-carbene complexes [31] and pyrimidine-
functionalized N-heterocyclic carbenes [32,33] as alternative
replacement for toxic and expensive phosphine-containing ligands.
In chemical engineering the ubiquitous interest is to heteroge-
nize catalytically assisted reactions and diminish the use of expen-
sive materials whether it is metal, ligand or solvent. In conjunction
with this concept, ionic liquids and metals can be deposited on car-
rying materials, forming a supported ionic liquid catalyst or SILCA.
Within SILCA the amount of ionic liquid is kept on such a low level
that the mass transfer limitations due to viscosity are avoided, but
still sufficiently enough to preserve their bulky characteristics. In
this way the supported ionic liquid that interacts with the active
metal, is the phase where reaction takes place, while the bulk sol-
vent is used to deliver and overtake reactants/products.
2.2. General Heck reaction procedure
All the experiments were performed in flat bottom glass vial
tubes (8 ml volume) with Teflon cup screw and magnetic stirrer.
The vials were heated up in oil bath, submerged in just above the
level of the reaction mixture leaving the upper part of the vial
not heated. Temperature of the oil was monitored and reported
as a temperature of the reaction with assumption that the temper-
ature differences were minimal due to small volumes and good
stirring. In a typical experiment, the determined amount of SILCA
and 1 ml of solvent was introduced directly into vials, followed
by the addition of 1 mmol of aryl halide, base (1.5 mmol) and olefin
(1.5 mmol). The starting point of the reaction was set as the point
when the temperature reached the preset value and the reaction
was let to run until completion or it was quenched at a certain time
by submerging the vial into the ice bath.
The progress of the reaction was monitored with TLC and ana-
lyzed by GC. All the experiments were repeated at least two times,
and the average values are reported. The experimental error was
2%.
2.3. Products extraction and catalyst recovery
The reaction slurry was first diluted with DMF and separated
from the catalyst with a centrifuge leaving organic layer rich with
products. To remove all formed salts and products deposited on the
catalyst we continued with washing the catalyst with DMF and
centrifuging. At the end all the DMF was combined to give 4 vol%
diluted aliquot.
To recover the catalyst it was flushed with diethyl ether and
dried in air at 100 °C for 30 min. This catalyst was reused in next
cycle without removal from the vial or stored at ambient condi-
tions for characterization.
Soon after introducing this inspiring concept [34–37] it was
employed for Heck reaction of iodbenzene and acrylate in
[C4mim]PF6 layer immobilized on silica with n-decane [38] and
H2O [39] as a solvents. The interesting aspect was the simplicity
of the impregnation method for a reactions which proceed in a
liquid phase. The ability of imidazolium ion to act as an
N-heterocyclic carbene ligand was then used to affix Pd for the
support. In this case, silylated methylimidazoium chloride was
covalently anchored on the SiO2 [40] and SBA-16 [41] surface. On
the other hand, if it is preferred first to functionalize silica, metal
can be introduced just by dissolving it with IL and re-layering
[42] or again with help of covalently anchored IL [43].
2.4. Catalyst synthesis
Bis-layered SILCA consisted of 3-imidazolbromide nicotinic
tetramethylguanidinium ionic liquid on SiO2 loaded with palla-
dium (SiO2-Im/Br-Nic.Ac/TMG-PdCl2) was synthesized according
to Scheme 1.
2.4.1. Silica functionalization
In this article we report synthesis of a new type of SILCA with
bis- ionic liquid layer grown up one on another. The layer contains
N-heterocyclic carbene palladium complex in its primary ionic liq-
uid layer upgraded by pyridine linker with acidic head which is
balanced with tetramethylguanidinium cation in a secondary
layer. The catalyst gave rise to an impressive activity in the Heck
reaction and one of the best turnover frequencies reported in liter-
ature till now. The separation and recyclability was simple without
any significant loss of the activity after five cycles.
In a first step calcined SiO2 was functionalized with a procedure
reported in literature [44]. In a round bottom flask, under nitrogen
atmosphere, (3-chloropropyl)trimethoxysilane was reacted with
support in refluxing toluene for a 24 h. A modifying agent was
introduced dropwise in a ratio of 5 mmol per 1 g of support. The
obtained material (SiO2-Cl) was washed with ethanol in a Soxhlet
apparatus, vacuum dried at 80 °C for 8 h and dried at 100 °C for 3 h.
This 3-chloropropylated silica was further immobilized with
imidazole. In
a round bottom flask well grinded imidazole
(20 mmol, 1.3616 g) was first dissolved in 30 ml of anhydrous
toluene at 90 °C. 2 g of modified support was added to a stirred
solution and let to react for 24 h under reflux conditions and N2
atmosphere. The material was then separated by filtration in a
Soxhlet nimble, flushed with hot toluene and washed with ethanol
for 24 h in the Soxhlet apparatus to yield white solid (SiO2-Im). The
procedure is presented as reaction 1) and 2), respectively,
Scheme 1.
2. Experimental
2.1. Chemicals
Iodobenzene, methyl acrylate, triethanolamine, dimethylforma-
maide, (3-chloropropyl)trimethoxysilane and palladium(II) chlo-

N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
37
Scheme 1. Synthesis of SiO2-Im/Br-Nic.Ac/TMG-PdCl2.
2.4.2. Synthesis of (3-imidazoliume propylated nicotinic acid) bromide
supported SiO2
6-Bromo-nicotinic acid was anchored to functionalized SiO2 via
multistep route presented in Scheme 1 (reaction 3) and Scheme 2.
CDCl3, 25 °C): d = 8.97 (dd, JH2,H4 = 2.4 Hz, JH2,H5 < 1 Hz, 1H, pyH2),
8.13 (dd, JH4,H5 = 8.3 Hz, 1H, pyH4), 7.59 (dd, 1H, pyH5), 3.96 (s,
3H, AOCH3). 13C NMR (125.8 MHz, CDCl3, 25 °C): d = 165.1 (ACO2-
CH3), 151.4 (pyC2), 146.9 (pyC6), 139.2 (pyC4), 128.1 (pyC5), 125.3
(pyC3), 52.6 (AOCH3) (for more details see Supplementary
material).
2.4.2.1. Fischer esterification of 6-bromo-nicotinic acid (Scheme 2,
3a). This synthesis is slightly modified literature procedure [45]. In
the glass batch reactor with dean-stark apparatus and magnetic
stirrer 6-bromo-nicotinic acid (3 mmol, 0.606 g) was dissolved in
10 ml of methanol. The mixture was cooled in ice bath and added
dropwise 1.5 ml of concentrated H2SO4. The suspension was then
heated to refluxing conditions and under vigorous stirring let to
react overnight. After cooling down, 5 ml of H2O was added and
then saturated NaHCO3 solution was added to neutralize the acid,
The mixture was transferred to a separation funnel and the product
was extracted with dichloromethane (3 * 30 ml) as a bottom layer.
Dichloromethane was evacuated in rotary evaporator at room tem-
perature to obtain 6-bromo-nicotinic acid methyl ester as a white
solid (0.43 g, 66%) with m.p. 65–67 °C. 1H NMR (500.20 MHz,
2.4.2.2. Functionalization of 3-imidazolepropylated SiO2 with 6-
bromo-nicotinic acid methyl ester (Scheme 2, 3b). The reaction of
6-bromo-nicotinic acid methyl ester with imidazole groups was
done in refluxing DMF. The reaction itself was inspired by the reac-
tion of bulky imidazole and chloro-sibling compound performed by
others [46], while the workup and extraction was simplified to cor-
respond the heterogeneity of new system [47,48]. In the glass
batch reactor with dean-stark apparatus and magnetic stirrer 6-
bromo-nicotinic acid methyl ester (3 mmol, 0.6481 g) was dis-
solved in 30 ml of DMF followed with addition 2 g of SiO2-Im.
The mixture was heated to refluxing conditions and the reaction
kept for 24 h under N2 atmosphere and vigorous stirring. The solid
Scheme 2. (3a) esterification of 6-bromo-nicotinic acid, (3b) synthesis of 3-imidazolbromide nicotinic methyl ester functionalized SiO2, (3c) hydrolysis of 3-imidazolbromide
nicotinic methyl ester functionalized SiO2.

38
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
product was separated and repeatedly washed with dichloro-
methane and ethanol to remove non-reacted ester followed with
drying at 70 °C for 3 h.
13C and 29Si NMR spectra were obtained with a Bruker AVANCE-
III HD 400 MHz spectrometer. The powdered sample was span at
5 kHz spin rate in a Bruker 1H broadband double-resonance
4 mm CP MAS probe. For 13C, the proton 90° high-power pulse
was 2.9 ms and contact time 2 ms. The recovery delay time was
set to 2 s and 10,000 scans were accumulated. For 29Si, the proton
90° high-power pulse was 2.5 ms and contact time 2 ms.
The binding energies of all compounds and oxidation state of
palladium in ionic liquid was investigated with X-ray photoelec-
tron spectroscopy (XPS). A Perkin-Elmer PHI 5400 spectrometer
2.4.2.3. Hydrolysis of 3-imidazolbromide nicotinic methyl ester func-
tionalized SiO2 (Scheme 2, 3c). All the material synthetized in the
previous step was dissolved in 30 ml of methanol. Followed by
dropwise addition of 2.5 ml of 1 M NaOH. The suspension was
heated to refluxing conditions and under vigorous stirring reacted
overnight. After cooling, the slurry was acidified with 1 M HCl until
pH = 2–3. The solid product was separated with a centrifuge and
repeatedly washed with H2O and ethanol to remove the non-
reacted ester. The final solid was then dried at 70 °C for 3 h and
then at 100 °C prior to the use in the next step. The obtained sup-
ported ionic liquid layer of 3-imidazolbromide nicotinic acid on
SiO2 was a white solid (SiO2-Im/Br-Nic.Ac).
with monochromatized Mg K
a 945 was used; the X-ray source
operated at 14 kV and 200 W. The pass energy of the analyzer
was 35.5 eV and the energy step 0.1 eV. Samples were outgassed
overnight in high vacuum (8 * 10À9 mbar) before scanning. The
peak fitting was performed with the program XPS Peak 4.1. The
background was corrected with the Shirley function. The metal
content at the catalyst was determined by inductively coupled
plasma optical emission spectroscopy (ICP-OES) with Optima
4300 DV optical atomic emission spectrometer. The recovery delay
time was set to 90 s and 1000 scans were accumulated.
Progress of the Heck reaction was monitored with GC system
HP 6890 Series with HP-5, 5% Phenyl methyl siloxane capillary col-
umn (30.0 m * 320 mm, 0.25 mm) equipped with FID. The injector
temperature was 280 °C while gas flow was 9.5 ml minÀ1. Column
was heated from 40 °C to 250 °C with 10 °C minÀ1 heating rate and
at the other end detector was at the constant temperature of
300 °C. Hexadecane was used as internal standard.
2.4.3. Synthesis of tetramethylguaindinum ionic liquid layer on SiO2-
Im/Br-Nic.Ac
Reaction 4 presented in Scheme 1 is supposed to simulate the
common route in the ionic liquid synthesis where a simple neutral-
ization takes place and yields salt. For the heterogeneous reasons
three equivalents of base were used instead of the equimolar ratio.
To the 30 ml of ethanol 1,1,3,3-tetramethylguanidine (10 mmol,
1.5118 g) and 2 g SiO2-Im/Br-Nic.Ac (all obtained from previous
step) was slowly added. The reactor was then sealed and repeat-
edly washed with diethyl ether to remove non-reacted TMG. The
final solid was dried at 70 °C for 3 h and then at 100 °C for 3 h. Sup-
ported ionic liquid layer of 3-imidazolbromide nicotinic tetram-
ethylguanidinium on SiO2 (SiO2-Im/Br-Nic.Ac/TMG) was obtained.
3. Results and discussion
2.4.4. Palladium impregnation
FT-IR was used as a first tool for catalyst study during the
preparation process and obtained spectra are presented in Fig. 1.
The most evident peaks are the ones at 470, 810, 1090 and
1215 cmÀ1 denoted to SiO2, as well as those at 1635 and 3450 cor-
responding to Si-OH and H-OH stretching of adsorbed water. After
the first reaction where (3-chloropropyl)trimethoxysilane was
anchored on the SiO2 surface, peaks at 1635 and 3450 disappeared
which indicates that the silanol groups were occupied with modi-
fying agent. In following step, chlorine was exchanged with imida-
zole ring and this reaction broth changes in signal at 1670 cmÀ1
which is assigned to C@C stretching. With nicotinic acid peak at
1575 cmÀ1 belonging to C@N vibration of pyridine ring confirmed
new ionic liquid layer formed, together with a very broad acidic
CO-H peak that dominated at around 3400 cmÀ1. The disappear-
ance of this peak was obvious when TMG was introduced and pro-
tonated (light blue line in Fig. 1, a). However, this peak was
recovered after the PdCl2 introduction implying on hydrogen
migration between the acidic head of the IL layer and TMG
(Fig. 1, a).
Spectrum of the catalyst recovered after Heck reaction did not
show difference compared to the fresh one indicating that layer
is firmly bonded and stable in reaction conditions (see Fig. 1, b).
Elemental analysis of fresh and recycled catalysts gave informa-
tion of catalyst content. As presented in Fig. 2. Carbon, nitrogen,
oxygen and bromine were present as a part of the formed IL layer,
and they retained even after reaction. Palladium was loaded and
detected as a PdCl2 salt; nevertheless, these chlorine anions were
exchanged with iodine in the first Heck cycle. The elemental per-
centages were calculated without the carbon content due to limi-
tation introduced with the analytic practice. This disallows a
quantitative study of the catalyst, but still based on the results
listed in Table 1 it is indicated that the IL layer is stable during
the reaction because nitrogen, oxygen and bromine amount did
not change significantly. Traces of fluorine appeared as an
impurity.
In glass batch reactor 0.2 mmol of PdCl2 (0.03547 g) was dis-
solved in 150 ml of methanol. Followed with addition of 2 g SiO2-
Im/Br-Nic.Ac/TMG (all obtained from previous step). Reactor was
sealed and slurry stirred 8 h at room temperature. After reaction
completion solid product was separated with centrifuge and
repeatedly washed with ethanol and diethyl ether to remove phys-
ical absorbed PdCl2. Final solid was dried at 70 °C for 3 h and at
100 °C for 3 h. Pale yellow SILCA powder SiO2-Im/Br-Nic.Ac/TMG-
PdCl2 was obtained (Scheme 1, reaction 5).
2.5. Catalyst characterization
The thermogravimetric analysis (TGA) of the catalyst was done
on CAHN D-200 instrument in the temperature range of 20–700 °C
with a heating rate of 10 °C minÀ1 under Ar atmosphere. The speci-
fic surface area and pore volume measurements were done by
means of nitrogen physisorption with Carlo-Erba instrument,
1990 Sorptometer. The BET equation and Dollimore-Heal method
were used for calculations. Prior to the analysis, samples were out-
gassed for 3 h at 150 °C. The surface modification of the support
was monitored with infrared spectroscopy (FT-IR) on ATI Mattson
FTIR. Samples were pressed into pellets and dried at 100 °C. Ele-
mental analysis was performed with energy-dispersive X-ray spec-
troscopy (EDX), while the morphology and palladium deposition
were followed with scanning electron microscopy (SEM). Coupled
EDX-SEM analyses were done on a Zeiss Leo Gemini 1530 with a
Thermo-NORAN vantage X-ray detector. Imaging of the catalyst
was performed with transmission electron microscopy (TEM) on
JEM 1400 plus (120 kV, 0.38 nm) equipped with OSIS Quemesa
11 Mpix bottom mounted digital camera. The solution state 1H
and 13C NMR spectra were recorded at 298 K with a Bruker
Avance-III HD 500 MHz spectrometer equipped with a Bruker
SmartProbeTM. As the solvent, deuterated chloroform with 0.03%
tetramethylsilane (TMS) as internal standard was used. CP MAS

N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
39
A moderate increase in the nitrogen amount after the reaction
can also indicate on the salt formation (Table 1). Thus, for catalyst
recycling repeated washing with a polar solvent is necessary.
The successful propylation of the SiO2 surface was confirmed
with 29Si NMR and results are presented in Fig. 4. The peak at
À57.9 ppm (T3 band) confirmed that most of the alkane chains
were anchored to the surface via two SiAOASi bonds. Additionally
some of the chains were connected to the surface trough one and
three bonds (peaks at À50.7 and À67.1 ppm). However, the large
peak at À101.3 ppm pointed out that many of the silanol groups
(Q3 units) stayed unchanged and this was expected because the
IL loading was 8% (estimated with TGA). Similar results have been
reported in literature [50].
To have a final proof of the IL layer formations we performed
solid state 13C NMR. The carbon atoms originally belonging to
the 3-chloroprpylated chain have peaks at 7.7, 23.0 and in
49.7 ppm region (see Fig. 5). The imidazole carbon coordinated
with N atoms showed up at 136.2 and two other carbons are over-
lapping in 122.0 ppm region [41]. The carbon atoms belonging to
the nicotinic acid were detected downfield with the most polarized
carbon belonging to the acid group appeared at 171.4 ppm due to
the deshielding in the presence of the tetramethylganidinium
cation in vicinity of oxygen atom. Those carbons bonded to nitro-
gen are at 151.8 and a stronger peak at 143.8. The peak at
137.6 ppm was assigned to the carbon opposite to the nitrogen
atom in pyridine ring. As showing Fig. 4, carbon no. 11 is expected
to give week signal that is overlapping with signal of imidazole car-
bons (136.2 ppm region) which would be in agreement with theo-
retical predictions and earlier studies [51]. The carbon atoms
belonging to TMG cation are detected at 160.9 ppm and
37.5 ppm, later one belonging to the methyl groups. In addition
to these two, at 157.4 ppm one more peak was detected after the
TMG immobilization. We assume that this one belongs to the frac-
tion of TMG that eventually forms ionic liquid with bromine as
anion. Which indicate that TMG is not strictly attached to end
chain of grafted IL layer but that there is a free flow in between
the two different anions appearing in site.
Before going deeper in the catalyst characterization, the pre-
pared catalytic material was screened under different conditions
in the benchmark reaction of iodobenzene and methylacrylate to
give methyl cinnamate as the main product. In the first study the
solvent, base and temperature were optimized. When reaction
was performed without solvent (Table 3, entry 1), only a small
amount of the product was detected. Different solvents showed
that the reaction proceeds much better in polar aprotic solvents
(entries 2–8) compare to protic ones (entries 9, 12, 13) and their
mixtures (entries 10 and 11), while nonpolar solvents almost hin-
dered reaction (entries 14, 15, 16). To distinguish between the
DMA and NMP as best solvents, the reaction was conducted with
a reused catalyst for additional three cycles, and in the third cycle,
NMP showed a much better performance giving 70% yield of cinna-
mate. In the following stage, we studied the influence of different
organics (entries 18–27) and inorganic bases (entries 28–32) and
their mixture (entry 33). As expected the use of a base was neces-
sary (entry 17). Tertiary amines gave the best results out of which
triethanolamine outperformed the other ones (entry 21) resulting
in complete conversion in the second Heck cycle and, therefore,
became an obvious choice. To inspect the temperature influence
the reaction was tested at room temperature and at 50, 80 and
120 °C, respectively (entries 34–37). At temperatures less than
100 °C, the reaction could not proceed. On the other hand, the reac-
tion was completed in 5 min at 120 °C, so in further experiments
this was fixed as the reaction temperature.
Fig. 1. FT-IR spectra of catalysts on different preparation stages (a) and after one
reaction cycle (b).
The decomposition temperature of solid samples were analyzed
with TGA and illustrated in Fig. 3. The initial weight loss in the
range from room temperature to 220 °C was attributed to physi-
cally absorbed water in the catalyst pores and also some other
volatiles (for the spent catalyst). In case of fresh and used catalyst
this was 2.5% and 6.0% of the catalyst weight while in case of pris-
tine silica this was about 1.5%. The organic moieties in the IL layer
start to decompose at the 265 °C and continue to do so until 510 °C.
At 450 °C the decrease of weight was assigned to the decomposi-
tion of the methoxy side groups [49]. A similar thermal behavior
of modified silica has been reported elsewhere [50]. From TGA
thermogram the loading of IL can be estimated to be 8 wt% which
corresponds to 0.1621 mmol gÀ1 grafted amount.
The surface area measurements are presented in Table 2. With
the modification surface area decreased from 414.65 m2 gÀ1 to
333.66 m2 gÀ1 which is still a relatively high value. The modifica-
tion also affected the pore volume and diameter that decreased
from 0.58 cm3 gÀ1 (6.60 nm) to 0.33 cm3 gÀ1 (5.46 nm) indicating
that the organic groups were successfully anchored inside the
pores of the pristine material and the mesoporosity was preserved.
After the first Heck reaction, the recycled catalyst did not show any
significant decrease in the surface area; however, the pore volume
was drastically diminished. In our opinion, this indicates on cata-
lyst poisoning due to the formation of salts from the Heck cycle.
Under the optimized reaction conditions, we studied the
amount of catalyst and the reaction versatility as summarized in
Table 4. In the first set of experiments (entries 1–5), the reaction

40
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
Fig. 2. EDX analysis of fresh (a) and recycled catalysts (b).
Table 1
Elemental analysis of fresh and used catalysts.
N
O
F
Si
Cl
Br
Pd
I
Sample
atom %
SiO2-Im/Br-Nic.Ac/TMG-PdCl2
SiO2-Im/Br-Nic.Ac/TMG-PdCl2 – spent
13.86
15.44
68.76
65.76
0.00
1.00
16.85
17.44
0.27
0.00
0.09
0.09
0.17
0.07
0.00
0.21
was initially catalyzed with Pd loading as low as 0.00009 mmol
(0.009 mol % related to iodobenzene) and further with a higher
Pd content up to 0.0018 mmol (0.18 mol %). Catalyst appeared to
be highly active for iodobenzene with different alkenes, while a
decline of activity was observed for bromo- and chlorobenzene
due to their thermodynamic restrictions (entries 6–13). Turnover
frequency for our benchmark reaction was calculated to be about
22,000 hÀ1 for low loading and around 16,000 hÀ1 for higher
Table 2
BET surface area and pore characteristics of support and catalyst.
Sample
SBET
Vp
dp
a
b
(m gÀ1
)
(cm3 gÀ1
)
(nm)c
SiO2
414.65
333.66
329.82
0.58
0.33
0.16
6.60
5.46
4.92
SiO2-Im/Br-Nic.Ac/TMG-PdCl2
SiO2-Im/Br-Nic.Ac/TMG-PdCl2 – spent
a
b
c
Specific surface area.
Pore specific volume.
Average pore diameter.
Fig. 3. TGA of fresh SiO2-Im/Br-Nic.Ac/TMG-PdCl2 and spent catalysts.

N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
41
Fig. 4. 29Si CP MAS NMR spectra of propylated surface of silica.
Fig. 5. 13C CP MAS NMR spectra of SiO2-Im/Br-Nic.Ac/TMG.
catalyst loading. In our opinion the reason for deviation from first
order kinetic in this case can be due to formation of Pd dimers that
slowly agglomerate as palladium black. On the other hand, when
small amount of palladium is used, all the Pd is utilized in the reac-
tion cycle and the activity is preserved. This deviation from 1st
order kinetics was also hypothesized in other studies [52] where

42
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
Table 3
Reaction condition optimization for Heck reaction of iodobenzene and methylacrylate.
Entrya
Solvent
Base
T (°C)
Time (min)
Yield (%)b
1
2
3
4
5
6
7
8
/
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
r.t.
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
30
30
30
5
5
92
52
DMF
DMSO
DMA
Acetonitrile
NMP
THF
100 (98, 87, 12)c
11
100 (100, 87, 70)c
4
2
3
29
5
32
2
2
Et3N
H2O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
DMF + H2O (4:1)
DMF + H2O (1:4)
(EtOH)3N
EtOH
Toluene
Heptane
Decane
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
1
2
1
/
Et3N
Pr3N
Bu3N
(EtOH)3N
Pyridine
Imidazole
TMG
Bu-TMG
DBU
100 (29)d
100 (91)d
100 (88)d
100 (1 0 0)d
0
8
1
1
0
0
30
83
2
DBN
K2CO3
Na2CO3
NaHCO3
NaOAc
NaOH
Et3N + Na2CO3 (1:1)
(EtOH)3N
(EtOH)3N
(EtOH)3N
(EtOH)3N
NMP
NMP
NMP
NMP
NMP
NMP
NMP
NMP
5
30
100
0e
0e
0e
100
50
80
120
a
b
c
Reaction conditions: iodobenzene 1.0 mmol, methylacrylate 1.5 mmol, base 1.5 mmol, Pd 0.09 mol % (detected by ICP-OES), solvent 1 ml.
Isolated yield of main product based on iodobenzene, detected with GC.
2nd, 3rd and 4th reaction cycle (reaction time 30 min, temperature 120 °C).
2nd reaction cycle (reaction time 10 min, temperature 120 °C).
d
e
Monitoring with TLC.
Table 4
Catalyst amount optimization and reaction versatility tests.
d
Entrya
Aryl halide
Alkene
Pd (mmol)b
Product (mmol)c
time (min)
TOF (hÀ1
)
1
2
3
4
5
6
7
8
Iodobenzene
Iodobenzene
Iodobenzene
Iodobenzene
Iodobenzene
Iodobenzene
Iodobenzene
Bromobenzene
Bromobenzene
Bromobenzene
Chlorobenzene
Chlorobenzene
Chlorobenzene
Me. Acrylate
Me. Acrylate
Me. Acrylate
Me. Acrylate
Me. Acrylate
Bu. Acrylate
Styrene
Me. Acrylate
Bu. Acrylate
Styrene
0.00009
0.00018
0.00045
0.0009
0.0018
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
1.0000
1.0000
0.9371
0.9762
0.9479
0.916
0.71
0.02
0.02
0.02
0
30
15
8
4
2
60
60
60
60
60
60
60
60
22,222
22,222
15,618
16,270
15,798
1018
789
22
22
22
0
0
9
10
11
12
13
Me. Acrylate
Bu. Acrylate
Styrene
0
0
0
a
b
c
Reaction conditions: arylhalide 1.0 mmol, alkene 1.5 mmol, base 1.5 mmol, solvent 1 ml, temperature 120 °C.
Pd content detected by ICP-OES.
Isolated yield of main product based on aril halide, detected with GC.
Calculated as mmol of product per mmol of Pd in unit of time.
d

N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
43
Table 5
TOF numbers for similar catalytic systems found in literature for reaction of iodobenzene and methyl acrylate.
c
Entry
Catalyst
Pd (mmol %)b
T (°C)
TOF (hÀ1
)
a
1
2
3
4
5
6
7
8
9
SiO2-Im/Br-Nic.Ac/TMG-PdCl2
SBA-TMG-Pd [53]
SBA-15-SIL-Pd(OAc)2 [42]
0.009
0.01
1.0
0.04
0.5
120
140
120
140
100
90
140
120
130
22,222
8400
29
2500
8
Pd/SBA-15 nanocomposite [54]
SiO2-BisILs[PdCl24À] [55]
SiO2 -Im. Cross Linked-Pd [56]
Silica-Supported NHC-Pd/IL [40]
PDVB-IL-Pd [57]
0.1
46
380
0.05
0.02
0.02
1212
Pd–g-SILLP [58]
9375d
a
b
c
This work.
Amount related to iodobenzene.
Calculated as mmol of product per mmol of Pd in unit of time.
TOF calculated at 50% of conversion.
d
the catalyst in close presence of halide activates Pd and releases
it from the surface, where it is immediately recaptured after the
reaction. Therefore we would also propose that a release and catch
mechanism takes a place on our catalyst.
To visualize the state of the catalyst, SEM and TEM imaging was
recorded on fresh and recycled catalysts in one Heck cycle. From
the SEM scans displayed in Fig. 6 it is clear that the palladium par-
ticles were uniformly dispersed on the silica surface before the
reaction while after being exposed to reaction conditions these
particles were agglomerated. Looking deeper into the catalyst mor-
phology by studying TEM images (Fig. 7) it was found that before
reaction, palladium particles were in the range of 12–18 nm but
after the first reaction cycle this changed to the 32–52 nm range.
The activity of reused catalyst was still extremely high. This ren-
ders us believe that during the course of the reaction, active Pd par-
ticles are released from the surface of the catalyst and when the
reaction is finished, they are redeposited on the surface. Recap-
tured palladium assemble dimerized particles in grape-like forma-
tions. Since the activity of this palladium is still extremely high, we
consider that the dimerized form represent a reservoir of the active
species.
The release and catch mechanism for the Heck reaction was
already reported [56] and it comes from the ability of the moiety
to retake Pd. In our case, the nitrogen containing and mainly the
TMG molecule has a function to harvest leached Pd in the form
of nanoparticles. To confirm this hypothesis, we prepared a cata-
lyst in same manner as described above but without the TMG
appendix. This catalyst was equally efficient in the first reaction
cycle, however, when reused this catalyst completely lost its activ-
ity. The SEM analysis showed that Pd agglomerated into inactive
nanoparticles with a particle size exceeding 100 nm (Fig. 7c). This
indicates that positively charged nitrogen atoms in the imida-
zolium and pyridinum rings are able to keep the palladium metal
in smaller clusters [58]. Nevertheless, after Pd is dimerized into
more negative charges only tetramethylguanidinium cations are
capable to keep Pd anions tight enough. This ability of TMG to
behave as a ligand and stabilize Pd during the Heck reaction was
demonstrated in bulk ionic liquids [59] and in the heterogenized
form [53].
The ICP-OES tests showed that a fresh catalyst had a Pd loading
of 0.96 wt% or 0.090 mmol gcÀa1t which means that 90% of the palla-
dium that was used for the synthesis ended up immobilized in the
catalyst, however we believe that not all of it was chemically
bonded. After the reaction, recycled catalyst showed a decrease
in Pd amount to 0.77 wt% (0.085 mmol gÀca1t), and this leached pal-
ladium was detected in the solvent. We consider that the palla-
dium leached during the reaction was the one that was
physically absorbed on the catalyst surface as well as one that
did not found the way back to the to the anchoring places on the
Fig. 6. SEM images of fresh (a) and spent catalyst (b).
it was found out that the reaction rate depends on the catalyst
amount and is between half and first order, depending on the sub-
strate ratios. Beside the fact that the best activity was achieved
with 0.009–0.018 mol % of Pd (1–2 mg of catalytic material), we
decided to conduct our study with higher loadings of palladium
(0.09 mol % Pd, 10 mg of catalytic material) in order to decrease
the influence of experimental errors.
The catalyst activities are presented as the turnover frequen-
cies. When compared with similar catalytic systems found in liter-
ature and gathered in Table 5, our catalyst showed the best
performance. In our opinion the reason for such a high activity is
the stability of palladium species in the ionic liquid layer, where
they are firmly bonded via NHC, CO and TMG ligands. Exposing

44
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
Fig. 7. TEM images and particle size distribution for fresh catalyst (a) and recycled catalyst (b). Recycled catalyst synthetized without TMG moiety (c).
Fig. 8. XPS spectrum of Pd 3d of fresh SiO2-Im/Br-Nic.Ac/TMG-PdCl2 (a) and spent catalyst (b).
catalyst surface because of possible contamination with salts
formed in the Heck cycle. The catalyst prepared without TMG
showed similar results with initial Pd loaÀd1ing of 0.88 wt%
(0.083 mmol gÀca1t) and 0.77 wt% (0.072 mmol gcat) after the reac-
tion. This implies that the role of TMG was merely to hinder forma-
tion of huge agglomerates and not just to retake released metal.
The physical state of palladium in fresh and spent catalysts was
determined by X-ray photoelectron spectroscopy. As displayed in
Fig. 8 both Pd(II) and metallic Pd(0) were detected in the catalyst.
If the ratio of ionic to metallic palladium was compared before ver-
sus after reaction (Fig. 8(a) and (b)) it is evident that after the reac-
tion most of Pd(II) was reduced to Pd(0). This was ascribed to the
abundance of nitrogen in ionic liquid layer.
To investigate the catalyst recyclability we repeated the reac-
tion of iodbenzene and methylacrylate in presence of 0.09 mol %
of Pd and (EtOH)3N base in NMP, like presented in Table 4, entry

N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
45
Table 6
Catalyst amount optimization and reaction versatility tests.
Entrya
Cycle
Yield (%)b
1
2
3
4
5
6c
1
2
3
4
5
6
100
100
100
100
100
100
a
Reaction conditions: iodobenzene 1.0 mmol, methylacrylate 1.5 mmol, triethanolamine 1.5 mmol, NMP 1 ml.
Isolated yield of main product based on iodobenzene, detected with GC.
Reaction time 7 min.
b
c
4. After each cycle the catalyst was washed with DMF to extract all
the products, followed with diethyl ether washing and drying.
After drying, the catalyst was reused in the reaction (Table 6). A
high activity was maintained in five cycles before the activity
started to decrease and prolonged reaction times were needed to
reach comparable conversion.
Conflicts of interest
There are no conflicts to declare.
Acknowledgments
Although a general opinion is that the agglomeration is respon-
sible for the decrease in the activity of the palladium supported
catalysts, we assume that this is not the primary reason, because
in that case the catalyst might be deactivated after the first cycle
as agglomeration is initiated. We consider that the decrease of
the activity after five cycles indicates on catalyst leaching as well
as poisoning of some kind. To gain more information we performed
the studies at lower conversion by suppressing the reaction before
full conversion; however, this led to a complete catalyst deactiva-
tion in the subsequent cycle, once more confirming the release and
catch mechanism. As long as iodobenzene is present in the slurry,
palladium will be in its homogenous form and, therefore, not
recovered by physical removal of the solid. Nevertheless, the cata-
lyst deactivation mechanism is speculative at this point and would
require more detailed analysis which will be conducted in upcom-
ing studies.
This work is part of the activities of the Johan Gadolin Process
Chemistry Centre an Åbo Akademi University center of excellence
and financed by Academy of Finland – the Academy Professor
grants 319002 (T. Salmi) and 320115 (N.Vucetic). Additionally
the Graduate School in Chemical Engineering, Finland is gratefully
acknowledged for the financial support. Further, in Sweden, the
BIO4ENERGY programme, the Kempe Foundations and the Wallen-
berg Wood Science Center are acknowledged.
Appendix A. Supplementary material
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.01.029.
References
[1] T. Mizoroki, K. Mori, A. Ozaki, Arylation of olefin with aryl iodide catalyzed by
palladium, Bull. Chem. Soc. Jpn. 44 (1971) 581, https://doi.org/10.1246/
bcsj.44.581.
4. Conclusions
[2] R.F. Heck, J.P. Nolley Jr, Palladium-catalyzed vinylic hydrogen substitution
reactions with aryl, benzyl, and styryl halides, J. Org. Chem. 37 (1972) 2320–
2322.
[3] R.F. Heck, Palladium catalyzed reactions of organic halides with olefins, Acc.
Chem. Res. 12 (1979) 146–151.
[4] W. Ramsay, On picoline and its derivatives, Philos. Mag. (1878).
[5] P. Murrill, Halides and perhalides of the picolines, J. Am. Chem. Soc. 21 (1899)
828–854, https://doi.org/10.1021/ja02060a002.
[6] H.O. Jones, A study of isomerism and optical activity of quinquevalent nitrogen
compounds, J. Chem. Soc. Trans. 83 (1903) 1400–1421, https://doi.org/
10.1039/CT9038301400.
[7] H.O. Jones, Optically active nitrogen compounds. d- and l-
phenylbenzylmethylethylammonium salts, J. Chem. Soc. Trans. 85 (1904)
223–234, https://doi.org/10.1039/CT9048500223.
[8] A.W. Harvey, A note on phenyldimethylallylammonium compounds, J. Chem.
Soc. Trans. 85 (1904) 412–414, https://doi.org/10.1039/CT9048500412.
[9] P. Walden, Molecular weights and electrical conductivity of several fused salts,
Bull. l’Academie Imp. Des Sci. St.-Petersbg. (1914) 405–422.
[10] J.S. Wilkes, M.J. Zaworotko, Air and water stable 1-ethyl-3-methylimidazolium
based ionic liquids, J. Chem. Soc. Chem. Commun. (1992) 965–967, https://doi.
org/10.1039/c39920000965.
[11] J.D. Holbrey, K.R. Seddon, Ionic liquids, Clean Prod. Process. 1 (1999) 223–236,
https://doi.org/10.1007/s100980050036.
[12] J.S. Wilkes, J.A. Levisky, R.A. Wilson, C.L. Hussey, Dialkylimidazolium
chloroaluminate melts: a new class of room-temperature ionic liquids for
electrochemistry, spectroscopy and synthesis, Inorg. Chem. 21 (1982) 1263–
1264, https://doi.org/10.1021/ic00133a078.
In the current study we disclosed a new catalytic system for the
Heck reaction. Supported ionic liquid catalyst showed by far the
best activity reported in literature for the reaction of iodobenzene
and methyl acrylate, achieving TOFs of 22,000 hÀ1. The high activ-
ity was denoted to the multiple anchoring points in the ionic liquid
layer that can stabilize active palladium via NHC, CO and TMG
ligands. The mostly influential were the TMG ligands which were
able to stabilize the Pd nanoparticles and limit the agglomeration
to the 52 nm particle size or less. These particles are considered
to be the reservoirs of the active Pd that catalyzes reaction via a
release and catch mechanism. Extremely small amounts of palla-
dium (0.009 mol %) were able to catalyze the reaction in a short
time; on the other hand when this amount was increased the cat-
alyst started to loss the activity probably due to the dimerization of
Pd. Considering the Pd amount used in this study, the metal depo-
sition was efficient and the utilization reached 90%. For the upcom-
ing studies the results will bring significant savings in the
production costs. The described catalyst synthesis procedure gave
SiO2 functionalized with nicotinic acid as an intermediate which
can lead to new applications. The work provides solid ground for
further investigations with bis-layered catalysts with different
metals, building blocks and reactions.
[13] K.R. Seddon, Ionic liquids for clean technology, J. Chem. Technol. Biotechnol. 68
(1997) 351–356, https://doi.org/10.1002/(SICI)1097-4660(199704)68:4<351::
AID-JCTB613>3.0.CO;2-4.

46
N. Vucetic et al. / Journal of Catalysis 371 (2019) 35–46
[14] R. Hagiwara, Y. Ito, Room temperature ionic liquids of alkylimidazolium
cations and fluoroanions, J. Fluor. Chem. 105 (2000) 221–227, https://doi.org/
10.1016/S0022-1139(99)00267-5.
[15] P. Wasserscheid, W. Keim, Ionic liquids—new ‘‘Solutions” for transition metal
catalysis, Angew. Chem. 39 (2000) 3772–3789, https://doi.org/10.1002/1521-
3773(20001103)39:21<3772::AID-ANIE3772>3.0.CO;2-5.
[16] T. Welton, Room-temperature ionic liquids. solvents for synthesis and
catalysis, Chem. Rev. (Washington, D. C.). 99 (1999) 2071–2083, https://doi.
org/10.1021/cr980032t.
Prep. Heterog. Catal. 162 (2006) 87–94, https://doi.org/10.1016/S0167-2991
(06)80894-3.
[38] H. Hagiwara, Y. Sugawara, K. Isobe, T. Hoshi, T. Suzuki, Immobilization of Pd
(OAc)2 in ionic liquid on silica: application to sustainable Mizoroki-Heck
reaction, Org. Lett. (2004), https://doi.org/10.1021/ol049343i.
[39] H. Hagiwara, Y. Sugawara, T. Hoshi, T. Suzuki, Sustainable Mizoroki-Heck
reaction in water: remarkably high activity of Pd(OAc)2 immobilized on
reversed phase silica gel with the aid of an ionic liquid, Chem. Commun.
(2005), https://doi.org/10.1039/b502528a.
[17] C.L. Hussey, Room temperature haloaluminate ionic liquids. Novel solvents for
transition metal solution chemistry, Pure Appl. Chem. 60 (1988) 1763–1772,
https://doi.org/10.1351/pac198860121763.
[18] C.L. Hussey, Room-temperature molten salt systems, Adv. Molten Salt Chem. 5
(1983) 185–230.
[40] B. Karimi, D. Enders, New N-heterocyclic carbene palladium complex/ionic
liquid matrix immobilized on silica: application as recoverable catalyst for the
Heck reaction, Org. Lett. (2006), https://doi.org/10.1021/ol060129z.
[41] H. Yang, X. Han, G. Li, Y. Wang, N-Heterocyclic carbene palladium complex
supported on ionic liquid-modified SBA-16: an efficient and highly recyclable
catalyst for the Suzuki and Heck reactions, Green Chem. (2009), https://doi.
org/10.1039/b904136b.
[19] Y. Chauvin, H. Olivier-Bourbigou, Nonaqueous ionic liquids as reaction
solvents, Chemtech 25 (1995) 26–30.
[20] J. Dupont, R.F. de Souza, P.A.Z. Suarez, Ionic liquid (molten salt) phase
organometallic catalysis, Chem. Rev. (Washington, DC, United States) 102
(2002) 3667–3691, https://doi.org/10.1021/cr010338r.
[42] J.Y. Jung, A. Taher, H.J. Kim, W.S. Ahn, M.J. Jin, Heck reaction catalyzed by
mesoporous SBA-15-supported ionic liquid-Pd(OAc)2, Synlett (2009), https://
doi.org/10.1055/s-0028-1087488.
[21] R. Feng, D. Zhao, Y. Guo, Revisiting characteristics of ionic liquids: a review for
further application development, J. Environ. Prot. (Irvine, Calif). 01 (2010) 95–
104, https://doi.org/10.4236/jep.2010.12012.
[22] D.E. Kaufmann, M. Nouroozian, H. Henze, Molten salts as an efficient medium
for palladium-catalyzed C-C coupling reactions, Synlett (1996) 1091–1092,
https://doi.org/10.1055/s-1996-5658.
[23] W.A. Herrmann, V.P.W. Böhm, Heck reaction catalyzed by phospha-
palladacycles in non-aqueous ionic liquids, J. Organomet. Chem. 572 (1999)
141–145, https://doi.org/10.1016/S0022-328X(98)00941-3.
[24] V.P.W. Böhm, W.A. Herrmann, Coordination chemistry and mechanisms of
metal-catalyzed C-C coupling reactions; Part 12 nonaqueous ionic liquids:
superior reaction media for the catalytic Heck-vinylation of chloroarenes,
Chem. – A Eur. J. 6 (2000) 1017–1025, https://doi.org/10.1002/(SICI)1521-
3765(20000317)6:6<1017::AID-CHEM1017>3.0.CO;2-8.
[25] A.J. Carmichael, M.J. Earle, J.D. Holbrey, P.B. McCormac, K.R. Seddon, The Heck
reaction in ionic liquids: a multiphasic catalyst system, Org. Lett. 1 (1999)
997–1000, https://doi.org/10.1021/ol9907771.
[26] S. Li, Y. Lin, H. Xie, S. Zhang, J. Xu, Brønsted guanidine acid-base ionic liquids:
novel reaction media for the palladium-catalyzed heck reaction, Org. Lett. 8
(2006) 391–394, https://doi.org/10.1021/ol052543p.
[43] X. Shi, X. Han, W. Ma, J. Fan, J. Wei, A PdCl2-ionic liquid brush assembly: an
efficient and reusable catalyst for Mizoroki-Heck reaction in neat water, Appl.
Organomet. Chem. (2012), https://doi.org/10.1002/aoc.1857.
[44] J. Yang, T. Zeng, D. Cai, L. Li, W. Tang, R. Hong, T. Qiu, Supported ionic liquids as
green catalyst for 2-butanol synthesis from transesterification of sec-butyl
acetate, Asia-Pacific J. Chem. Eng. 11 (2016) 901–909, https://doi.org/10.1002/
apj.2024.
[45] B.V.L. Potter, J. Dowden, A. Galione, WO2005054198A2, 2005.
[46] J. Liao, Y. Hong, Y. Wang, T.W. Von Geldern, K.E. Zhang, WO2011094890A1,
2011.
[47] M. Li, P.J. Pham, C.U. Pittman, T. Li, SBA-15-supported ionic liquid compounds
containing silver salts: novel mesoporous
p-complexing sorbents for
separating polyunsaturated fatty acid methyl esters, Micropor. Mesopor.
Mater. 117 (2009) 436–443, https://doi.org/10.1016/j.micromeso.2008.07.017.
[48] S.K. Gupta, D. Ghorai, J. Choudhury, A new type of palladium-pincer complexes
generated
via
hydrolytic
ring-opening
of
imidazole-2-ylidenes,
Organometallics 33 (2014) 3215–3218, https://doi.org/10.1021/om500362x.
[49] V. Antochshuk, M. Jaroniec, Functionalized mesoporous materials obtained via
interfacial reactions in self-assembled silica-surfactant systems, Chem. Mater.
12 (2000) 2496–2501, https://doi.org/10.1021/cm000268p.
[27] R. Wang, J.-C. Xiao, B. Twamley, J.M. Shreeve, Efficient Heck reactions catalyzed
[50] Q. Wang, G.A. Baker, S.N. Baker, L.A. Colón, Surface confined ionic liquid as a
stationary phase for HPLC, Analyst 131 (2006) 1000–1005, https://doi.org/
10.1039/B607337A.
[51] D. Brunel, A. Cauvel, F. Fajula, F. DiRenzo, MCM-41 type silicas as supports for
immobilized catalysts, Stud. Surf. Sci. Catal. 97 (1995) 173–180, https://doi.
org/10.1016/S0167-2991(06)81887-2.
[52] T. Rosner, J. Le Bars, A. Pfaltz, D.G. Blackmond, Kinetic studies of heck coupling
reactions using palladacycle catalysts: experimental and kinetic modeling of
the role of dimer species, J. Am. Chem. Soc. 123 (2001) 1848–1855, https://doi.
org/10.1021/ja003191e.
[53] X. Ma, Y. Zhou, J. Zhang, A. Zhu, T. Jiang, B. Han, Solvent-free Heck reaction
catalyzed by a recyclable Pd catalyst supported on SBA-15 via an ionic liquid,
Green Chem. 10 (2008) 59–66, https://doi.org/10.1039/b712627a.
[54] 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, https://doi.org/10.1016/j.molcata.2007.03.006.
[55] X. Shi, X. Han, W. Ma, J. Fan, J. Wei, A PdCl2-ionic liquid brush assembly: an
efficient and reusable catalyst for Mizoroki-Heck reaction in neat water, Appl.
Organomet. Chem. 26 (2012) 16–20, https://doi.org/10.1002/aoc.1857.
[56] C. Pavia, F. Giacalone, L.A. Bivona, A.M.P. Salvo, C. Petrucci, G. Strappaveccia, L.
Vaccaro, C. Aprile, M. Gruttadauria, Evidences of release and catch mechanism
in the Heck reaction catalyzed by palladium immobilized on highly cross-
linked-supported imidazolium salts, J. Mol. Catal. A Chem. 387 (2014) 57–62,
https://doi.org/10.1016/j.molcata.2014.02.025.
[57] G. Liu, M. Hou, J. Song, T. Jiang, H. Fan, Z. Zhang, B. Han, Immobilization of Pd
nanoparticles with functional ionic liquid grafted onto cross-linked polymer
for solvent-free Heck reaction, Green Chem. 12 (2010) 65–69, https://doi.org/
10.1039/b913182e.
by
a highly recyclable palladium(II) complex of a pyridyl-functionalized
imidazolium-based ionic liquid, Org. Biomol. Chem. 5 (2007) 671–678, https://
doi.org/10.1039/b616529j.
[28] L. Wang, H. Li, P. Li, Task-specific ionic liquid as base, ligand and reaction
medium for the palladium-catalyzed Heck reaction, Tetrahedron 65 (2009)
364–368, https://doi.org/10.1016/j.tet.2008.10.042.
[29] Z.D. Petrovic´, D. Simijonovišc´, V.P. Petrovic´, S. Markovic´, Diethanolamine and
N, N-diethylethanolamine ionic liquids as precatalyst-precursors and reaction
media in green Heck reaction protocol, J. Mol. Catal. A Chem. 327 (2010) 45–
50, https://doi.org/10.1016/j.molcata.2010.05.010.
[30] Z.D. Petrovic´, S. Markovic´, V.P. Petrovic´, D. Simijonovic´, Triethanolammonium
acetate as a multifunctional ionic liquid in the palladium-catalyzed green Heck
reaction, J. Mol. Model. (2012), https://doi.org/10.1007/s00894-011-1052-1.
[31] J.A. Loch, M. Albrecht, E. Peris, J.W. Faller, R.H. Crabtree, Palladium complexes
with tridentate pincer bis-carbene ligands as efficient catalysts for C - C
coupling, Organometallics (2002) 700–706, https://doi.org/10.1021/
cr050004s.
[32] D. Meyer, M.A. Taige, A. Zeller, K. Hohlfeld, S. Ahrens, T. Strassner, Palladium
complexes with pyrimidine-functionalized N-heterocyclic carbene ligands:
synthesis structure and catalytic activity, Organometallics 28 (2009) 2142–
2149, https://doi.org/10.1021/om8009238.
[33] F. Schroeter, J. Soellner, T. Strassner, Cross-coupling catalysis by an anionic
palladium complex, ACS Catal. 7 (2017) 3004–3009, https://doi.org/10.1021/
acscatal.6b03655.
[34] C.P. Mehnert, E.J. Mozeleski, R.A. Cook, Supported ionic liquid catalysis
investigated for hydrogenation reactions, Chem. Commun. (2002) 3010–
3011, https://doi.org/10.1039/b210214e.
[35] C.P. Mehnert, R.A. Cook, N.C. Dispenziere, M. Afeworki, Supported ionic liquid
catalysis–a new concept for homogeneous hydroformylation catalysis, J. Am.
Chem. Soc. 124 (2002) 12932–12933, https://doi.org/10.1021/ja0279242.
[36] P. Virtanen, H. Karhu, G. Toth, K. Kordas, J.P. Mikkola, Towards one-pot
synthesis of menthols from citral: modifying Supported Ionic Liquid Catalysts
(SILCAs) with Lewis and Brønsted acids, J. Catal. 263 (2009) 209–219, https://
doi.org/10.1016/j.jcat.2009.02.005.
[58] M.I. Burguete, E. García-Verdugo, I. Garcia-Villar, F. Gelat, P. Licence, S.V. Luis,
V. Sans, Pd catalysts immobilized onto gel-supported ionic liquid-like phases
(g-SILLPs): a remarkable effect of the nature of the support, J. Catal. 269 (2010)
150–160, https://doi.org/10.1016/j.jcat.2009.11.002.
[59] S. Li, Y. Lin, H. Xie, S. Zhang, J. Xu, Brønsted guanidine acid À base ionic liquids:
novel reaction media for the palladium-catalyzed heck reaction, Tetrahedron
2003–2006 (2006).
[37] P. Mäki-Arvela, J.-P. Mikkola, P. Virtanen, H. Karhu, T. Salmi, D.Yu. Murzin,
Supported ionic liquid catalyst (SILCA) in the hydrogenation of citral, Sci. Bases