From the grafting of NHC-based Pd(II) complexes onto TiO2 to the in situ generation of Mott-Schottky heterojunctions: The boosting effect in the Suzuki-Miyaura reaction. Do the evolved Pd NPs act as reservoirs?

Article abstract of DOI:10.1016/j.jcat.2021.04.016

The assumption that the real active species involved in the Suzuki-Miyaura reaction are homogeneous, heterogeneous or both is often proposed. However a lack of characterization of the true catalytic entities and their monitoring makes assumptions somewhat elusive. Here, with the aim of getting new insights into the formation of active species in the Suzuki-Miyaura reaction, a family of palladium(II) complexes bearing bis(NHC) ligands was synthesized for immobilization at the surface of TiO₂. The studies reveal that once the complexes are anchored onto TiO₂, the mechanism governing the catalytic reaction is different from that observed for the non-anchored complexes. All complexes evolved to Pd NPs at the surface of TiO₂ under reaction conditions and released Pd species in the liquid phase. Also, this reactivity was boosted by the in situ generation of Mott-Schottky heterojunctions, opening new routes towards the design of heterogenized catalysts for their further implementation in reverse-flow reactors.

Full text of DOI:10.1016/j.jcat.2021.04.016

Journal of Catalysis 398 (2021) 133–147
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
From the grafting of NHC-based Pd(II) complexes onto TiO2 to the in situ
generation of Mott-Schottky heterojunctions: The boosting effect in the
Suzuki-Miyaura reaction. Do the evolved Pd NPs act as reservoirs?
a
a,
Jonathan De Tovar a,b, , Franck Rataboul , Laurent Djakovitch
⇑
⇑
a Université de Lyon, Université Claude Bernard Lyon 1, CNRS, IRCELYON, F-69626 Villeurbanne, France
b Université Grenoble Alpes, CNRS, CEA, IRIG, Laboratoire de Chimie et Biologie des Métaux, 17 rue des Martyrs, Grenoble 38000, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The assumption that the real active species involved in the Suzuki-Miyaura reaction are homogeneous,
heterogeneous or both is often proposed. However a lack of characterization of the true catalytic entities
and their monitoring makes assumptions somewhat elusive. Here, with the aim of getting new insights
into the formation of active species in the Suzuki-Miyaura reaction, a family of palladium(II) complexes
bearing bis(NHC) ligands was synthesized for immobilization at the surface of TiO2. The studies reveal
that once the complexes are anchored onto TiO2, the mechanism governing the catalytic reaction is dif-
ferent from that observed for the non-anchored complexes. All complexes evolved to Pd NPs at the sur-
face of TiO2 under reaction conditions and released Pd species in the liquid phase. Also, this reactivity was
boosted by the in situ generation of Mott-Schottky heterojunctions, opening new routes towards the
design of heterogenized catalysts for their further implementation in reverse-flow reactors.
Ó 2021 Elsevier Inc. All rights reserved.
Received 13 January 2021
Revised 23 March 2021
Accepted 12 April 2021
Available online 21 April 2021
Keywords:
Suzuki-Miyaura
N-heterocyclic carbene
Palladium complexes
Palladium nanoparticles
Heterogeneous homogeneous catalysis
Titania
1. Introduction
coupling. This corresponds to a further step towards the design
of active and recyclable catalytic systems. For that we introduced
Palladium-based catalysts play an important role in C-C cross-
coupling reactions [1]. In particular, the impact of the Suzuki-
Miyaura reaction has been immense since it has resulted to be
one of the most efficient methodologies for the formation of biaryl
derivatives [2].
a function at the ligand backbone allowing the grafting onto a sup-
port. It has been demonstrated for reported systems that the sup-
port clearly influences the catalytic performance [9,10,11,12,13]. In
this view, TiO2, exhibiting good thermal, mechanical and chemical
stabilities can be regarded as a promising candidate. Therefore we
particularly followed here the evolution of TiO2-heterogenized pal-
ladium complexes of the type PdBr2(L \ L), in which L \ L is a
chelating bis(benzimidazol-2-ylidene) ligand. We aimed at identi-
fying their catalytic performances and the nature of the active spe-
cies generated from such complexes. For that we evaluated the
effect of the space-length (\) of the ligand, the reaction tempera-
ture, and the recyclability of the new catalytic system.
Despite the fact that many palladium-based complexes exhibit-
ing high performances have been developed, the recovery and
reuse of the catalytic system is still problematic. To tackle this
issue the heterogenization of the complexes by immobilization
onto a support is a possibility [3,4,5,6,7]. However, the catalytic
activity of such systems is usually lower than that of the homoge-
neous ones. Hence, the design of more active and stable heteroge-
nized catalysts remains a priority for C-C cross-coupling reactions.
Recently, we reported the excellent performances of novel N-
heterocyclic carbene (NHC)-Pd catalysts (C1a–c, Scheme 1) for
the Suzuki-Miyaura reaction and particularly demonstrated that
the activity was strongly affected by the complex architecture
[8]. In the present work we describe a comprehensive study of
the heterogenization of these systems for use in Suzuki-Miyaura
2. Results and discussion
2.1. Synthesis and characterization of ligands and complexes
The synthetic pathway to form the hybrid materials is pre-
sented in Scheme 1. It is based on the linkage of two benzimidazole
1 moieties giving bisbenzimidazolealkanes 2a–c [14] followed by
an alkylation with diethyl (3-bromopropyl)phosphonate leading
to ligands 3a-c. These were fully characterized by 1H, 13C and 31P
NMR spectroscopies (1D and 2D spectra, see Experimental section,
⇑
Corresponding authors.
E-mail addresses: Jonathan.DeTovarVillanueva@cea.fr (J. De Tovar), Laurent.
Djakovitch@ircelyon.univ-lyon1.fr (L. Djakovitch).
https://doi.org/10.1016/j.jcat.2021.04.016
0021-9517/Ó 2021 Elsevier Inc. All rights reserved.

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Scheme 1. (a) Synthetic pathway for 3a-c ligands, mC1a-c complexes and mC1a-c-TiO2 hybrid catalysts. (b) C1a-c complexes prepared in our previous study (reference [8]).
Figures S1–S3). Here note that 1H NMR indicates a C2 symmetry in
Table 1
Pd (wt%) and grafting (%) for the mC1a-c-TiO2.
solution for the three ligands in which not only the two alkylated
benzimidazolium moieties are equivalent but also bridge protons.
All ligands have also been characterized by HR-ESI-MS (Figures S4–
S6) and IR spectroscopies (Figures S7–S9).
Hybrid Catalyst
Pd (wt%)
Grafting (%)
mC1a-TiO2
mC1b-TiO2
mC1c-TiO2
0.23
0.21
0.19
82
75
68
Next, coordination to Pd(OAc)2 gave corresponding palladium
(II) molecular complexes mC1a-c obtained as air- and moisture-
stable solids. 1H NMR spectra show that the downfield signals
(~9–11 ppm) of acidic protons for the different bisbenzimidazolium
salts are no longer present (Figures S10–S12), confirming the
deprotonation of the benzimidazolium moiety upon formation of
biscarbene complexes [15]. Additionally, all methylene protons of
the bridges become diastereotopic due to ligand coordination
and retention of the bend-boat conformation in these complexes
[16]. The observed broadening of these resonances has been attrib-
uted to the fluxionality of the seven- or eight-membered chelate
rings [17]. Interestingly, the methylene protons of propyl chains
became also diastereotopic because of the HꢀꢀꢀBr interaction with
bromido ligands, these methylene protons being shifted downfield
(4–5 ppm). The formation of the mC1a-c mononuclear complexes
was also supported by their high-resolution ESI mass spectra at m/
z = 892.9868, 907.0019 and 921.0180 for the molecular cations
[M + Na]+ (mC1a, mC1b and mC1c, respectively, Figures S13–
S16), fitting with their simulated spectra. Finally, the complexes
have been complementary characterized by IR spectroscopy (Fig-
ures S16–S18).
amount of Pd(II) complexes did not increase the coverage. Also, the
Pd/P ratios present in the hybrid materials match with their non-
grafted counterparts (see Experimental section) indicating that
the integrity of the complexes remained upon grafting. The anchor-
ing through the phosphonate moiety of mC1a-c complexes was
attested by ATR-IR spectroscopy (Fig. 1 and S19-S20). As shown
in Fig. 1 the mC1a complex presents two absorption bands at
1222 cmꢁ1 and 1014 cmꢁ1 corresponding to the free P@O and
PAOAC units, respectively [22]. No free P@O and PAOAC bands
are visible for mC1a-TiO2 while appeared a band at 1037 cmꢁ1
attributed to –PAOATi bond [22,23].
2.2. Anchoring of the complexes on TiO2
The next step consisted in grafting the mC1a-c Pd(II) complexes
at the surface of TiO2 (P-25, see Experimental section) to form the
hybrid materials (mC1a-c-TiO2) (Scheme 1).
Phosphonate anchoring groups are well known to efficiently
interact with metal oxide surfaces [18,19,20]. For instance, the
cleavage of P-OEt bonds on TiO2 has already been reported in the
case of diethyl phenylphosphonate [21]. Here the grafting was per-
formed following Guerrero et al. [22] procedure. A TiO2 suspension
in dichloromethane containing the palladium(II) complex was stir-
red at 45 °C for 24 h. Then, dispersion was filtered off, washed and
dried at 120 °C under vacuum. ICP-OES analyses indicated that the
grafting was achieved with 68 to 82% yield (Table 1) relative to the
maximum expected 0.28 wt% Pd. Note that increasing the starting
Fig. 1. Overlay of ATR-IR spectra of TiO2 (black), mC1a complex (red) and mC1a-
TiO2 hybrid material (blue). The transmittance of each sample has been shifted
along the y-axis for comparison purposes. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
134

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Additionally, the 31P HPDEC solid-state NMR spectra of the
hybrid materials (Fig. 2 and S21-S22) were similar regardless of
the complex. The major resonance at 26.1 ppm is ascribed to tri-
dentate phosphonate sites –P(OTi)3. The presence of an additional
resonance near 31.5 ppm points to another bonding mode, which is
likely to be bidentate phosphonate –P(OTi)2(OEt) sites. 13C CP-MAS
NMR spectra show a good match with the free mC1a-c complexes
(Figures S23–S25).
TEM analyses of the hybrid nanomaterials revealed the pres-
ence of TiO2 nanoobjects with a mean size of 21 nm (Figure S26).
Therefore the initial size and morphology of TiO2 were maintained
after anchoring the mC1a-c without the decomposition of the com-
plexes. The mC1b-TiO2 and mC1c-TiO2 hybrid nanomaterials were
characterized accordingly, with very similar results to mC1a-TiO2
(Figures S19–S22 and S24–S26).
contrary to C1a-c catalysts, the performances observed for the
hybrid materials led to higher iTOFs for larger bridges (Table 1).
Therefore, once the complexes are grafted, the space-length (\)
affects the transformation through an opposite effect. From here,
it can be inferred a different catalytic mechanism when heteroge-
nizing the mC1a-c catalysts (vide infra, sections 2.4–2.5).
Interestingly, the new mC1b-TiO2 and mC1c-TiO2 materials
gave moderate to good yield when decreasing the temperature to
25 °C whereas C1a-c complexes showed almost no activity
(Fig. 4). This confirms a higher efficiency of the heterogeneous
catalysts.
The formation of Pd NPs from free C1a-c complexes was already
observed in our previous work under the same catalytic conditions
[8]. Surprisingly, it appears here that even when C1a-c complexes
are immobilized, Pd NPs were formed during the reaction as
revealed by TEM images. Note that they are present not only at
the surface of TiO2 but also on the holey carbon of the TEM grid
(Figures S27–S28 and Table S1). These particles exhibited a
monodisperse population of spherical NPs with a narrow size-
distribution. Interestingly, all NPs display a mean size lower than
those evolved directly from the free C1a-c complexes [8]. Addition-
ally, smaller NPs were observed after catalysis at higher tempera-
tures. Thus, increasing the temperature greatly enhanced the
complex decomposition followed by cluster formation as well as
favoring the leaching of highly active Pd species from grafted Pd
complexes or in situ formed Pd NPs.
To sum-up this section, easily obtained Pd(II) complexes bear-
ing phosphonate moieties have been prepared and successfully
grafted onto TiO2 surface giving new hybrid nanomaterials.
2.3. Catalytic performance studies
The different prepared hybrid nanomaterials have been evalu-
ated as catalysts for the model reaction depicted in Scheme 2,
and compared with the free C1a-c palladium(II) molecular com-
plexes (Scheme 1). We aimed at performing the reactions under
mild conditions and short times using a low catalyst loading.
p-Bromoacetophenone and phenylboronic acid were selected as
substrates in order to discriminate cross-coupling from homocou-
plings. Neither the homocoupling of phenylboronic acid nor that of
2.4. Leaching studies
Having demonstrated the superior performance in the presence
of the mC1c-TiO2 catalyst for the Suzuki-Miyaura cross-coupling
reaction, we addressed the leaching of the Pd-species in solution
during the catalytic reaction, suspected by the formation of Pd
NPs. Leaching is often observed when supported catalysts are used
in such cross-coupling reactions [24,25,26] determined following
several methodologies [24,27]. Among them the hot-filtration
method provides reliable information. The Pd leaching was exam-
ined for the coupling reaction at 25 °C using the hot-filtration
method: after 15 min of a standard catalytic run (corresponding
to ca. 35% conversion, Fig. 5), the reaction mixture was filtered to
remove mC1c-TiO2 solid leaving a clear filtrate (see Experimental
section). The latter was then used as reaction medium and ana-
lyzed over time. Here, the activity was completely stopped after fil-
tration (Fig. 5). This suggests that (1) no complex degrafting
occurred during this period, or (2) leached species if any were
not active at 25 °C, or (3) the catalytic performance could be due
to some Pd NPs formed from catalyst decomposition in the first
15 min. No Pd was found in the filtrate by ICP-OES analysis but
could be present at concentration less than 0.01 ppm, that is the
detection limit. With these results, to discard between the different
hypotheses, we decided to increase the reaction temperature to
60 °C when using the filtrate mixture. Interestingly, at this temper-
ature the reaction was reinitiated giving a complete conversion
after 20 min (Fig. 5). This result indicates clearly that Pd species
released from the immobilized palladium system or from evolved
Pd NPs are present in solution. These Pd-species play a significant
role in the catalyst activity at 60 °C, through homogeneous path-
ways. Thus, these Pd species, despite present at a very low concen-
tration, could be the true catalytic entities and would then be
redeposited on the functionalized TiO2 or merged as Pd NPs at
reaction completion.
p-bromoacetophenone
nor
dehalogenation
of
p-
bromoacetophenone were observed. Thus, since the formation of
p-acetylbiphenyl was exclusive, conversions and yields were con-
sidered as equal values. Once conditions optimized, phenylboronic
acid (PhB(OH)2/Pd = 1200), p-bromoacetophenone (substrate/Pd =
1000), and sodium methoxide (base/Pd = 1500) were employed in
absolute ethanol as solvent to assure a better solubility of the for-
mer. Final conversions, turnover numbers (TONs), turnover fre-
quencies (TOFs), initial activities (iTOFs) with induction periods
are reported in Table
determination).
In order to compare the catalytic performance of grafted and
non-grafted complexes, reactions were performed at two different
temperatures of 60 °C and 25 °C using the same Pd loading (Figs. 3
and 4).
2
(see Experimental section for
As previously observed for the free C1a-c complexes,[8] deple-
tion was noted in the first 10 min (Fig. 3a) while complete conver-
sions were achieved for all mC1a-c-TiO2 (Fig. 3b). Surprisingly and
26.1
31.5
*
*
150
100
50
0
-50
-100
2.5. Recycling studies
(ppm)
The recycling of the catalyst was examined for the coupling at
60 °C in the first 10 min. This procedure, described elsewhere,
Fig. 2. 31P HPDEC NMR spectrum of mC1a-TiO2 hybrid material. * Spinning bands.
135

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Scheme 2. Suzuki-Miyaura reaction studied in this work.
Table 2
iTOF, conversion, TON and TOF values for the Suzuki-Miyaura reaction of p-bromocetophenone with phenylboronic acid.
Entry
Catalyst
p-bromoacetophenone
60 °C
25 °C
iTOF ((mol product ∙ mol Pd-1 ∙ minꢁ1
)
Conv.
TON
TOF (hꢁ1
)
iTOF ((mol product ∙ mol Pd-1 ∙ minꢁ1
)
Conv.
TON
TOF (hꢁ1
)
1
2
3
4
5
6
C1a
C1b
C1c
mC1a-TiO2
mC1b-TiO2
mC1c-TiO2
611
294
90
169
253
1914
100
65
31
100
100
100
1000
648
309
1000
1000
1000
12,000
388
927
3000
6000
20,000
2
1
0.4
7
33
58
6
2
4
10
53
83
60
19
41
97
530
826
60
19
41
97
530
826
8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol phenylboronic acid and 1.2 mmol MeONa in 16 mL absolute EtOH. Conversion after 1 h reaction.
Fig. 3. Conversion versus the time for the coupling reaction of p-bromoacetophenone with phenylboronic acid using (a) C1a-c and (b) mC1a-c-TiO2 catalysts at 60 °C.
Fig. 4. Conversion versus the time for the coupling reaction of p-bromoacetophenone with phenylboronic acid using (a) C1a-c and (b) mC1a-c-TiO2 catalysts at 25 °C.
136

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
In order to avoid the increased viscosity of the reaction media,
the recycling tests of mC1c-TiO2 were also performed by centrifu-
gation of the reaction crude for recovery of the solid after each run
(method B, see Experimental section). As shown in Fig. 7 and
Table 4, lower iTOFs were obtained when reusing the separated
catalyst and thus longer times were required to achieve complete
conversions.
Interestingly, the filtrate collected after the 1st run (method B)
was tested for possible catalytic activity. As shown in Fig. 7, it
reveals, as already observed for leaching studies, the presence of
species able to catalyze the reaction, at 60 °C, but rapidly deactivat-
ing as shown from the strong inflexion observed after few minutes,
leading to slower overall transformation. In parallel, a complete
conversion was achieved by using the filtered catalyst recovered
after a 1st run (method B). Comparing to the recycling method A
used previously when all species, i.e. dissolved Pd-species and
Pd-species immobilized on TiO2, are present at the same time, it
can be inferred that the catalytic performance of mC1c-TiO2 comes
from both, homogeneous and heterogeneous Pd-species. This can
be observed in the 2nd run when recycling the mC1c-TiO2 material
by method A (Table 3, 2nd run, Fig. 6). An equilibrium between the
two natures of Pd-species being more probably responsible for the
higher behavior observed when performing recycling by the
method A. It is noteworthy to mention that the reduced efficiency
of the isolated catalyst resulted from each run by method B could
also be due to a deactivation upon separation from the catalytic
reaction. However, as demonstrated in Figs. 6 and 7, the Pd species
in the liquid phase play a pivotal role in the catalytic activity com-
pared to the generated Pd species on TiO2.
Fig. 5. Activity of mC1c-TiO2 catalyst after hot-filtration at ca. 30% conversion
versus standard catalytic run.
[28,29] was performed as follows: at the completion of a 1st run
using fresh mC1c-TiO2, new reagents were added to the reaction
mixture and the conversion was set to 0 at the corresponding ‘‘ini-
tial time” for the next run of the catalyst (method A, see Experi-
mental section). Fig. 6 shows that an activation occurs after the
1st run of the catalyst as indicated by determination of the initial
rate iTOF1 = 1914 minꢁ1 for the 1st run and greater iTOF2 = 5532-
minꢁ1 for the 2nd run (Table 3). On a one hand, only small differ-
ence was observed in terms of conversions which was
completely achieved within 3 min for the 1st run whereas the 2nd
run needed 10 min. To confirm the results the procedure was
repeated twice. Under these conditions, comparable reaction rates
iTOF3 and iTOF4 to that for 2nd run were observed (Table 3) indicat-
ing that no ‘‘re-activation” of the catalyst was required. On another
hand, longer reaction time to achieve quantitative conversions
over successive recyclings can be attributed to the visually
observed change of the viscosity of the reaction mixture (increased
ionic strength), preventing efficient diffusion of the substrates to
the active sites.
2.6. Fate of the catalyst and mechanism considerations
To further study the fate of mC1c-TiO2 under turnover condi-
tions, the crude reaction mixture upon recycling (method B) was
analyzed by TEM, EDX, XPS, UV–Vis, solid-state 31P NMR and ICP-
OES.
Fig. 8 shows the evolution of the 31P HPDEC NMR spectrum and
indicates that the bidentate phosphonate moieties at ca. 31.5 ppm
were partially hydrolyzed to yield a net tripodal anchoring at ca.
25 ppm. In parallel, TEM images reveal the presence of Pd NPs of
4.7 0.6 nm both at the surface of TiO2 and along the grid (Fig. 8
and Table 5). After the 2nd cycle, Pd NPs were only present at the
surface of TiO2 showing two families of 4.3 1.0 and 17.9 2.3 n
m. After the 3rd cycle, a single family of Pd NPs of 4.3 0.9 nm
was observed. This indicates that, in fact, the evolved Pd NPs at
the surface of TiO2 act as reservoirs of active Pd species.
Additionally, the Pd and P content for the hybrid catalyst deter-
mined after each catalytic run (Table 5) indicates that the P/Pd
ratio remains stable upon recycling, supporting efficient Pd rede-
position on the support at reaction completion. As no P was
detected in solution we exclude detachment of the ligand from
the support, despite a slight increase of the Pd/P ratio in used cat-
alyst. This explains the differentiated performance between the
free C1a-c and mC1a-c-TiO2 catalysts since the nature of the Pd
NPs evolved by the latter is completely different than those
evolved by the former, as already suggested by their size
histograms.
To gain further insights into the effects of the state surface ele-
mental composition, the XPS patterns for bare TiO2, mC1c-TiO2 and
mC1c-TiO2 (after 3rd run reused by method B) were recorded as
shown in Figures S30-S33. Figure S31 shows the XPS signature of
the Pd 3d doublet (3d3/2 and 3d5/2). According to the observed
binding energies, mC1c-TiO2 (Figure S31, red) exhibits the Pd 3d
signal at 343.38 eV (Pd 3d3/2) and 338.0 eV (Pd 3d5/2), entirely cor-
responding to Pd+2 whereas mC1c-TiO2 after 3rd run reused by
method B (Figure S31, blue) exhibits at 340.6 eV (Pd 3d3/2) and
Fig. 6. Recyclability (method A) of the mC1c-TiO2 catalyst for the C-C coupling of p-
bromoacetophenone and phenylboronic acid. Reaction conditions: 8 ꢂ 10ꢁ4 mmol
Pd, 0.8 mmol p-bromoacetophenone,
MeONa, 16 mL absolute EtOH, 60 °C.
1 mmol phenylboronic acid, 1.2 mmol
137

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Table 3
Catalyst activities, conversion, TON and TOF values for the Suzuki-Miyaura reaction of p-bromoacetophenone and phenylboronic acid when recycling mC1c-TiO2 (method A).
Entry
iTOF (mol product ∙ mol Pdꢁ1 ∙ minꢁ1
)
Conv.
TON
TOF (hꢁ1
)
1st run
2nd run
3rd run
4th run
1914
5532
6235
6137
100
100
100
100
1000
1000
1000
1000
20,000
6000
3000
1500
Reaction conditions: 8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol phenylboronic acid and 1.2 mmol MeONa in 16 mL absolute EtOH. Conversion after 1 h
reaction.
100
attributed to steric hindrance due to the bulky ethyl phosphonate
moieties.
Second, the role of the TiO2 support was also studied. While no
reactivity was observed with TiO2 alone, TiO2 in the presence of
80
C1c (Fig. 9, pink, inversed triangle) gave a lower performance than
60
40
20
0
that of single C1c. Interestingly, Pd NPs evolved under these condi-
tions were not deposited at the surface of TiO2 (Figure S29). Thus,
the formation of Pd NPs at the surface of TiO2 could be ascribed to
their stabilization by the grafted ligands (Scheme 3). To confirm
this hypothesis, the 3a ligand (see Scheme 1) was grafted at the
surface of TiO2, yielding the 3a-TiO2 material (Figures S35-S36,
sand Experimental section), which was introduced in the standard
catalytic test in the co-presence of the C1c complex (see Experi-
mental section). As shown in Fig. 10a, no depletion of the C1c cat-
alyst was observed in the presence of the 3a-TiO2 material,
achieving a complete conversion after 40 min. Additionally, the
TEM image of 3a-TiO2 after performing the reaction (Fig. 10b)
reveals the formation of Pd NPs of 1.4 0.3 nm at the surface of
TiO2. This confirms the above hypothesis that ligand must be
grafted onto TiO2 to form Pd NPs at the TiO2 surface. Also, ICP-
OES analysis for the isolated hybrid catalyst provided a 0.14% Pd
content. Additionally, the resulting Pd-enriched hybrid catalyst
was further reused exhibiting a lower catalytic performance with
ca. 15% conversion as shown in Figure S37. This phenomenon is
attributed to a screen effect by the Pd NPs covering TiO2.
1st run (method B)
2nd run (method B)
3rd run (method B)
filtrate after 1st run (method B)
0
10
20
30
40
50
60
Time (min)
Fig. 7. Recyclability (method B) of the mC1c-TiO2 catalyst for the C-C coupling of p-
bromoacetophenone and phenylboronic acid. Reaction conditions: 8 ꢂ 10ꢁ4 mmol
Pd, 0.8 mmol p-bromoacetophenone,
MeONa, 16 mL absolute EtOH, 60 °C.
1 mmol phenylboronic acid, 1.2 mmol
335.2 eV (Pd 3d5/2), corresponding to that of bulk metallic Pd. This
confirms that no decomposition of the complex occurs upon graft-
ing and the previously observed nanoparticles are of Pd(0). Addi-
tionally, the phosphorous XPS spectra with signals are attributed
to the phosphonate moieties (Figure S32) and indicate no change
before and after the catalytic test. Interestingly, nitrogen XPS spec-
tra showed that part of the benzimidazole-2-ylidene moieties were
oxidized after catalysis (Figure S33).
The optical response of bare TiO2, mC1c-TiO2 and mC1c-TiO2
(after 3rd run reused by method B) was recorded showing a strong
absorption at less than 400 nm in all cases, which is attributed to
the intrinsic absorption of TiO2 (Figure S34). However, mC1c-
TiO2 (after 3rd run reused by method B) exhibited a slightly
increase in absorbance in the visible and near infrared region com-
pared to that of TiO2.
In order to unravel the enhanced catalytic activity for mC1a-c-
TiO2, a new series of experiments was performed (Fig. 9 and
Table 6).
First, the non-grafted mC1c complex (containing the phospho-
nate groups) was tested (Fig. 9, red, round) exhibiting a similar per-
formance to that of C1c (Fig. 9, black, squares). However, two times
lower iTOF was observed for this catalyst (Table 6, entry 2). This is
On the other hand, when the free mC1c complex was tested in
the presence of TiO2 (Fig. 9, green, rhombus), a medium overall
performance with an iTOF of 43 minꢁ1 was observed probably
due to (partial) in situ grafting of the complex.
Third, to elucidate the nature of the active species two separate
Hg poisoning tests were performed. A large excess of Hg(0) (Hg:Pd
ꢃ 2000:1 mol/mol) was added to the reaction mixture at the begin-
ning for mC1c-TiO2 at 60 °C (Fig. 9, violet, right arrow). Then, the
iTOF dramatically decreased to 68 minꢁ1 (Table 6, entry 7). In
the other experiment, when Hg(0) was added after 15 min of reac-
tion at 25 °C, the conversion ceased completely (Fig. 11). These
data also suggest that the grafted mC1c complex acts in fact as
pre-catalyst and its evolved Pd NPs are the active species or reser-
voirs of them.
Fourth, since the TiO2 P-25 support possesses semiconductor
properties, we evaluated the possibility of photocatalytic assis-
tance. For that, three different experiments were conducted. On
the one hand, the mC1c-TiO2 performance was studied in the dark
(Fig. 9, purple, hexagon) revealing a lower iTOF (Table 6, entry 8)
but a complete conversion in the first 20 min. Interestingly, when
Table 4
Catalyst activities, conversion, TON and TOF values for the Suzuki-Miyaura reaction of p-bromoacetophenone and phenylboronic acid when recycling mC1c-TiO2 (method B).
Entry
iTOF (mol product ∙ mol Pdꢁ1 ∙ minꢁ1
)
Conv.
TON
TOF (hꢁ1
)
1st run
2nd run
3rd run
1914
108
102
100
100
100
1000
1000
1000
20,000
1500
1500
Reaction conditions: 8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol phenylboronic acid and 1.2 mmol MeONa in 16 mL absolute EtOH. Conversion after 1 h
reaction.
138

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Fig. 8. Evolution of the 31P HPDEC NMR for the mC1c-TiO2 catalyst upon recycling and TEM images with their Pd NPs size histograms for each run (method B).
Table 5
recorded for Ph-B(OH)2 (Fig. 12, black line), Ph-B(OH)2 after 1 h
reacting with 1.25 equivalents of MeONa in EtOH (Fig. 12, red line)
and Ph-B(OH)2 acid after 1 h reacting with 1.25 equivalents of
MeONa in the presence of 1 equivalent of TiO2 in EtOH (Fig. 12,
blue line). The Ph-B(OH)2 + OH– ꢀ Ph-B(OH)-3 equilibrium, which
presents the 55:45 ratio, was shifted to the right after the introduc-
tion of TiO2 revealing a 36:64 ratio. Thus, these three experiments
indicate that the photogenerated h+ in TiO2 contribute to the Ph-B
(OH)2-TiO2 surface interaction accelerating the trans-metalation
step [31]. On the other hand, when mC1c-TiO2 was tested in the
presence of an electron scavenger such as TEMPO (Fig. 9, olive,
pentagon), a quasi shift relative to mC1c-TiO2 was observed
prompting to hypothesize that the photogenerated electrons accel-
erate the oxidative addition step.
Evolution of the mean sizes and their standard deviations for Pd nanoparticles when
recycling (method B) the mC1c-TiO2 catalyst using p-bromoacetophenone as
substrate at 60 °C.
Catalyst
Mean size standard deviation (nm)
Pd (wt%)
P (wt%)
1st run
2nd run
3rd run
4.7 0.6
4.3 1.0; 17.9 2.3
4.3 0.9
0.12
0.12
0.10
0.04
0.04
0.04
Reaction conditions: 8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol
phenylboronic acid and 1.2 mmol MeONa in 16 mL absolute EtOH. Data after 1 h
reaction.
mC1c-TiO2 was tested as usual but in the presence of a hole scav-
enger (sodium oxalate) [30] (Fig. 9, brown, stars and Table 6, entry
9), the same reactivity was observed as that in the dark. Since Ph-B
(OH)2 can react in basic medium to form negative Ph-B(OH)-3 spe-
cies, the holes h+ can assist in cleaving the C-B bond to produce
biaryl-Pd species. Moreover, 11B solid-state NMR experiments were
Since the work function of palladium is located between the
HOMO and LUMO of TiO2
at the surface of TiO2 results in the in situ generation of a rectifying
metal-semiconductor contact, Mott-Schottky heterojunction
(UPd > UTiO2), the formation of Pd NPs
a
Fig. 9. Catalytic performance for the coupling of p-bromoacetophenone and phenylboronic acid by using (a) C1c, black, squares, (b) mC1c, red, round, (c) TiO2, blue, triangle,
(d) C1c+TiO2, pink, inversed triangle, (e) mC1c+TiO2, green, rhombus, (f) mC1c-TiO2, navy, left arrow, (g) mC1c-TiO2+Hg, violet, right arrow, (h) mC1c-TiO2 in the dark, purple,
hexagon, (i) mC1c-TiO2+sodium oxalate, brown, stars, (j) mC1c-TiO2+TEMPO, olive, pentagon. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
139

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Table 6
Catalyst activities, conversion, TON and TOF values for the Suzuki-Miyaura reaction of p-bromoacetophenone and phenylboronic acid.
Entry
Catalyst
iTOF (mol product mol Pdꢁ1 minꢁ1
)
Induction time (min)
Conv.
TON
TOF (hꢁ1
)
18
2
3
4
5
6
7
8
9
C1c
mC1c
TiO2
C1c+TiO2
mC1c+TiO2
mC1c-TiO2
mC1c-TiO2+Hg
mC1c-TiO2 in the dark
mC1c-TiO2+oxalate
mC1c-TiO2+TEMPO
90
43
–
41
786
1914
68
498
421
352
–
–
–
–
–
–
–
–
–
2
31
27
–
22
70
100
35
100
100
100
309
265
–
222
702
1000
345
1000
1000
1000
927
530
–
444
1404
20,000
690
3000
3000
4000
10
Reaction conditions: 8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol phenylboronic acid and 1.2 mmol MeONa in 16 mL absolute EtOH. Conversion after 1 h
reaction.
Scheme 3. Catalyst evolution of grafted Pd-bisNHC complexes at the surface of TiO2.
Fig. 10. (a) Comparison of the catalytic activities for the C1c catalyst in the absence and in the presence of the 3a-TiO2 material for the C-C coupling of p-bromoacetophenone
and phenylboronic acid. Reaction conditions: 8 ꢂ 10ꢁ4 mmol Pd, 0.8 mmol p-bromoacetophenone, 1 mmol phenylboronic acid, 1.2 mmol MeONa, 16 mL EtOH, 60 °C. (b) TEM
micrograph of the 3a-TiO2 material with (c) its size histogram after the catalytic reaction.
[32]. as shown in Scheme 4. Thus, a charge transfer at the interface
occurs resulting in a positively charged region in TiO2 and nega-
tively charged Pd NPs due to the Schottky effect. The energetic e-
located at the Pd NPs and h+ located at the TiO2 surface can con-
tribute to activate the p-bromoacetophenone and phenylboronic
acid, respectively. Consequently, the Suzuki-Miyaura reaction is
accelerated. However, this heterojunction can only be formed
thanks to the proximity of the Pd atoms by the grafted complexes,
140

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
(4q-4 t). Heteroaromatics bromides were also evaluated giving in
general moderate to quantitative yields for pyridyl, quinolines
and isoquinolines (4u-4y) but 8% and 0% yield for 4v and 4x,
respectively. However, the catalyst could not activate (benz)imida-
zolium bromides (4z-4ab), sulfur-containing heteroaromatics bro-
mides (4ac-4ae), oxygen-containing heteroaromatics bromides
(4af-4ag) and indole bromide (4ah-4al) but 4ak with a moderate
product yield in these conditions without substrate optimization.
3. Conclusions
We have prepared and characterized a new series of NHC-based
Pd(II) complexes, which have been immobilized at the surface of
TiO2. These new hybrid materials were used as heterogeneous cat-
alysts for the Suzuki-Miyaura reaction with the aim of providing
new insights into the nature of active species in such catalysts.
First, all transformations gave quantitative yields for the cou-
pling of p-bromoacetophenone and phenylboronic acid at 60 °C,
and moderate to excellent yields at 25 °C. Moreover, the TOF values
reached up to 20000 hꢁ1 at 0.1% catalyst loading under environ-
mentally friendly mild conditions.
Fig. 11. Comparison of the catalytic activities for mC1c-TiO2 catalyst in the absence
(standard run) and after the addition of Hg (Hg poisoning) for the coupling of p-
bromoacetophenone and phenylboronic acid at 25 °C.
Then, we have demonstrated that the heterogenized complexes
not only exhibit better performances than for the corresponding
homogeneous counterparts but also imply a different mechanism.
Following the catalytic and recyclability studies, mercury poison-
ing tests, TEM and 31P HPDEC NMR analyses we propose that the
enhanced catalytic activities can be ascribed to the leaching of
highly-active Pd species from the evolved Pd NPs at the surface
of TiO2. Interestingly for the mC1c-TiO2 catalyst, we evidenced that
the TiO2 nanostructure not only serves as support but also as pho-
toinduced boosting tool due to the in situ generation of Mott-
Schottky heterojunctions. Thus, the photogenerated electrons in
TiO2 transferred to the Pd NPs facilitate the cleavage of the C-Br
in aryl halides, while the photogenerated holes assist the cleavage
of the C-B bond in the phenylboronic acid. A possible mechanism of
the chemical transformation of the hybrid catalysts, the enhancing
effect by TiO2 and the leaching of Pd species is therefore proposed
here pointing that the evolved Pd NPs act as reservoirs of highly-
active Pd species.
Finally, after revealing the applicability of mC1c-TiO2 upon
recycling, a scope of substrates exhibited moderate to excellent
yields for (de)activated aryl bromides as well as for sterically hin-
dered and heteroaromatic derivatives.
Fig. 12. 11B MAS NMR spectrum for the phenylboronic acid (black line), phenyl-
boronic acid after 1 h reaction with 1.25 equivalents of MeONa in EtOH (red line)
and phenylboronic acid after 1 h reaction with 1.25 equivalents of MeONa in the
presence of 1 equivalent of TiO2 in EtOH (blue line). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.)
4. Experimental
4.1. Materials and methods
which could be attributed to both the contribution to the Pd(II) !
Pd(0) reduction by the photogenerated electrons in TiO2 and the
stabilization by the grafted ligands.
On the basis of the obtained results, we propose a mechanism
(Scheme 4) in which the phenomena above mentioned take place.
Evonik/Degussa AEROXIDEÒ TiO2 P-25 powder formed of
nanoparticles of 21 nm average size was sourced from Nippon
Aerosil (Yokkaichi, Japan). The rest of reagents were acquired from
Sigma-Aldrich and Alfa Aesar.
All analyses were performed at IRCELYON except notified.
GC analyses were performed on a Shimadzu GC-2030 gas chro-
matograph equipped with a FID detector, a HP-5 column (cross-
linked 5% phenylmethylsiloxane, 30 m ꢂ 0.25 mm ꢂ 0.25 mm) with
helium as carrier gas. Yields were determined by GC based on the
relative area of GC signals referred to an internal standard (naph-
thalene) calibrated to the corresponding pure compounds.
1H NMR, 13C NMR, 31P{1H} NMR, HSQC and HMBC spectra were
recorded on an AVANCE III 400 Bruker spectrometer at a tempera-
ture of 25 °C. All chemical shifts were given in ppm.
2.7. Scope of the Suzuki-Miyaura reaction
Following these mechanistic studies, a reaction scope for the
Suzuki-Miyaura reaction was evaluated for different aryl bromides
by using mC1c-TiO2 to form more functionalized products
(Fig. 13). As usually observed for such coupling, aryl bromides with
electron-withdrawing substituents were particularly reactive giv-
ing high yields (4a-4 g). Aryl bromides bearing electron-donating
substituents resulted in moderate yields (4 h-4o). However, 4p
could not be obtained due to its reactivity under basic condition
towards its enolate form, which can evolve to a quinone derivative
[33]. Also, bulky aromatics provided mainly quantitative yields
31P solid-state NMR spectra for hybrid materials were recorded
on an AVANCE III 500WB Bruker spectrometer with the high-
power decoupling (HPDEC) technique. 13C solid-state NMR spectra
141

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Scheme 4. Catalyst evolution and Mott-Schottky heterojunction photocatalytic contribution showing the view of the rectifying metal-semiconductor (TiO2) contact. Legend:
Ev, Valence band (HOMO); Ec, Conduction band (LUMO); Ef, work function.
for hybrid materials were recorded on an AVANCE III 500WB Bru-
ker spectrometer with the cross-polarization magic angle spinning
(CP-MAS) technique. 11B solid-state NMR spectra for hybrid mate-
rials were recorded on an AVANCE III 500WB Bruker spectrometer
with the magic angle spinning (MAS) technique.
The palladium content was determined on an ICP-OES ACTIVA
Jobin Yvon apparatus from solutions obtained by treatment of
the catalysts with sulfuric acid and aqua regia in a Teflon reactor
at 400–450 °C.
Electrospray ionization Mass Spectrometry (ESI-MS) experi-
ments were performed on a QTOF Impact II Bruker instrument
equipped with an UHPLC U3000 Dionex system by the Centre Com-
mun de Spectrométrie de Masse at the University of Lyon 1.
Transmission electron microscopy (TEM) analyses were carried
out on a JEOL 2010 microscope with an instrumental magnification
of 50000x to 100000x and an acceleration voltage of 200 kV. The
point-to-point resolution of the microscope was 0.19 nm and the
resolution between the lines was 0.14 nm. The size distributions
were determined via manual analysis of enlarged images by mea-
suring ca. 200 particles on a given grid to obtain a statistical size
distribution and a mean diameter.
4.2. Synthesis and characterization of the ligands
4.2.1. 1,10-methylenebisbenzimidazole (2a)
A
DMSO (10 mL) solution of benzimidazole (2929 mg,
24.8 mmol, 2 eq) and potassium hydroxide (5563 mg, 99.2 mmol,
8 eq) was stirred at room temperature for 1 h. Then, diiodo-
methane (1 mL, 12.4 mmol, 1 eq) was added at once and the reac-
tion was stirred overnight. After the addition of water (150 mL),
the product was extracted with dichloromethane (3 ꢂ 10 mL),
dried over magnesium sulfate and concentrated. Diethyl ether
(50 mL) was added to precipitate
a white powder. Yield:
2470 mg, 80%. The spectroscopic data matched that reported in
the literature.9
4.2.2. 1,10-(dimethylene)bisbenzimidazole (2b)
A
DMSO (10 mL) solution of benzimidazole (1000 mg,
8.46 mmol, 2 eq) and potassium hydroxide (475 mg, 8.46 mmol,
2 eq) was stirred at room temperature for 1 h. Then, dibro-
moethane (0.37 mL, 4.23 mmol, 1 eq) was added at once and the
reaction was stirred overnight. After the addition of water
(150 mL), the product was extracted with dichloromethane
(3 ꢂ 10 mL), dried over magnesium sulfate and concentrated.
Diethyl ether (50 mL) was added to precipitate a white powder.
Yield: 154 mg, 14%. The spectroscopic data matched that reported
in the literature [9].
Infrared (IR) spectra were recorded with a Thermo Scientific
Nicolet iS5 spectrometer in the 4000–525 cmꢁ1 range in attenu-
ated total reflectance (ATR) mode.
X-ray photoelectron spectroscopy (XPS) analyses were carried
out with a Versa Probe II spectrometer (ULVAC-PHI) equipped with
4.2.3. 1,10-(trimethylene)bisbenzimidazole (2c)
a monochromated Al Ka source (hm = 1486.6 eV) at CEA Grenoble.
The core level peaks were recorded with constant pass energy of
23.3 eV. The XPS spectra were fitted with CasaXPS 2.3 software
using Shirley background. Binding energies are referenced with
respect to the adventitious carbon (C 1 s BE = 284.6 eV).
UV–Vis absorption spectra were carried out with a LAMBDA 365
UV/VIS Spectrophotometer from Pelkin Elmer (Walmar, MA, USA)
in solid-state at room temperature.
A
DMSO (10 mL) solution of benzimidazole (1000 mg,
8.46 mmol, 2 eq) and potassium hydroxide (475 mg, 8.46 mmol,
2 eq) was stirred at room temperature for 1 h. Then, diiodopropane
(0.43 mL, 4.23 mmol, 1 eq) was added at once and the reaction was
stirred overnight. After the addition of water (150 mL), the product
was extracted with dichloromethane (3 ꢂ 10 mL), dried over mag-
nesium sulfate and concentrated. Diethyl ether (50 mL) was added
142

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
Fig. 13. Reaction scope for the coupling of aryl bromides with phenylboronic acid in the presence of the mC1c-TiO2 catalyst.
143

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
to precipitate a white powder. Yield: 494 mg, 42%. The spectro-
scopic data matched that reported in the literature.9
4.3. Synthesis and characterization of the complexes
4.3.1. Dibromido-(1,10-di(3-diethoxyphosphoryl)propyl-
4.2.4. 1,10-Di(3-diethoxyphosphoryl)propyl-3,30-
methylenebisbenzimidazolium dibromide (3a)
3,30methylenedibenzimidazolin-2,20-diylidene)palladium(II) (mC1a)
Palladium(II) acetate (102 mg, 0.45 mmol, 1 eq) and ligand 3a
(350 mg, 0.46 mmol, 1.01 eq) were dissolved in degassed DMSO
(3 mL) in a round-bottom flask. The reddish mixture was stirred
for 4 h at room temperature, 17 h at 40 °C and, finally, 2 h at
120 °C. After cooling to room temperature, dichloromethane
(3 mL) was added and the crude was filtered through a plug of
celite. Then, diethyl ether (50 mL) was added to the yellow solution
to precipitate the product as an orange powder. Yield: 268 mg, 69%.
1H NMR (DMSO d6, 25 °C): d1.18 (m, 12H, H12), 1.86 (m, 4H, H10),
2.10 (m, 4H, H9), 3.95 (m, 8H, H11), 4.60 (m, 2H, H8), 5.01 (m, 2H,
H8), 6.79 (d, J = 13.98 Hz, 1H, H13), 7.35 (d, J = 14.19 Hz, 2H, H13),
7.45 (m, 4H, H4 and H5), 7.80 (d, J = 7.99 Hz, 2H, H3), 8.24 (d,
J = 7.92 Hz, 2H, H6). 31P{1H} NMR (DMSO d6, 25 °C): d31.01. 13C
{1H} NMR (DMSO d6, 25 °C): d16.7 (d, J = 5.23 Hz, C12), 22.09 (d,
J = 140.28 Hz, C10), 23.1 (C9), 48.5 (C8), 58.0 (C13), 61.5 (t,
J = 7.25 Hz, C11), 111.8 (C6), 112.2 (C3), 124.5 (C5), 124.7 (C4),
A mixture of diethyl (3-bromopropyl)phosphonate (387 mL,
2 mmol, 2 eq) and 2a (250 mg, 1 mmol, 1 eq) was stirred at
80 °C for 24 h. After cooling to room temperature, the crude was
solved in dichloromethane (3 mL) and diethyl ether (25 mL) was
added to precipitate a white off hygroscopic solid. Yield: 760 mg,
99%. 1H NMR (CDCl3, 25 °C): d1.25 (t, J = 6.97 Hz, 12H, H12), 1.93
(m, 4H, H10), 2.38 (m, 4H, H9), 4.03 (m, 8H, H11), 4.77 (t,
J = 7.46 Hz, 4H, H8), 7.55 (t, J = 7.96 Hz, 2H, H3), 7.65 (t,
J = 7.83 Hz, 2H, H6), 7.86 (d, J = 8.40 Hz, 2H, H5), 8.16 (s, 2H,
H
13), 8.74 (d, J = 8.63 Hz, 2H, H4), 11.32 (s, 2H, H1). 31P NMR (CDCl3,
25 °C): d30.17. 13C NMR (CDCl3, 25 °C): d16.5 (d, J = 5.83 Hz, C12),
22.2 (d, J = 141.87 Hz, C10), 22.6 (d, J = 5.00 Hz, C9), 47.8 (d,
J = 12.56 Hz, C8), 56.5 (C13), 62.2 (d, J = 6.55 Hz, C11), 113.6 (C5),
114.8 (C4), 127.7 (C3), 128.4 (C6), 130.8 (C2), 131.2 (C7), 144.4
(C1). IR (ATR) cmꢁ1: 2978
m(C-H)ar, 2931 m(C-H), 1397 (d(C = C),
d(C = N))ar, 1215 (P = O)st, 1014 (P-O-C)st, 758 d(C-H)oop. HR-
ESI-MS (DMSO): calculated: m/z = 687.1899 ([M + Na]+); found:
m/z = 687.1895 ([M + Na]+), 303.1362 ([Mꢁ2Br]+2).
133.5 (C7), 133.72 (C2), 170.9 (C1). IR (ATR) cmꢁ1: 3093
2972 (C-H), 1451, 1397 (d(C = C), d(C = N))ar, 1222 (P = O)st,
m(C-H)ar,
m
1014 (P-O-C)st, 757 d(C-H)oop. HR-ESI-MS (DMSO): calculated:
m/z = 892.9866 ([M + Na]+); found: m/z = 892.9868 ([M + Na]+),
791.0794 ([MꢁBr]+).
4.2.5. 1,10-Di(3-diethoxyphosphoryl)propyl-3,30-(dimethylene)
bisbenzimidazolium dibromide (3b)
A mixture of diethyl (3-bromopropyl)phosphonate (147 mL,
0.76 mmol, 2 eq) and 2b (100 mg, 0.38 mmol, 1 eq) was stirred
at 80 °C for 4 h. After cooling to room temperature, the crude
was filtered and washed first with dichloromethane (5 mL) and
then with diethyl ether (10 mL). Yield: 244 mg, 89%. 1H NMR
(D2O, 25 °C): d1.35 (t, J = 7.18 Hz, 12H, H12), 1.98 (m, 4H, H10),
2.10 (m, 4H, H9), 4.17 (m, 8H, H11), 4.57 (t, J = 7.80 Hz, 4H, H8),
5.27 (s, 4H, H13), 7.36 (d, J = 8.50 Hz, 2H, H5), 7.57 (t, J = 7.95 Hz,
2H, H3), 7.72 (t, J = 7.70 Hz, 2H, H6), 7.93 (d, J = 8.43 Hz, 2H, H4),
9.44 (s, 2H, H1). 31P NMR (D2O, 25 °C): d33.3. 13C NMR (D2O,
25 °C): d15.6 (d, J = 5.9 Hz, C12), 21.0 (d, J = 141.6 Hz, C10), 22.0
(d, J = 4.55 Hz, C9), 46.5 (C13), 47.1 (d, J = 19.4 Hz, C8), 63.5 (d,
J = 6.67 Hz, C11), 111.6 (C5), 113.7 (C4), 127.9 (C3 and C6), 130.8
4.3.2. Dibromido-(1,10-di(3-diethoxyphosphoryl)propyl-
3,30ethylenedibenzimidazolin-2,20-diylidene)palladium(II) (mC1b)
Palladium(II) acetate (43 mg, 0.19 mmol, 1 eq) and ligand 3b
(150 mg, 0.19 mmol, 1.01 eq) were dissolved in degassed DMSO
(3 mL) in a round-bottom flask. The reddish mixture was stirred
for 4 h at room temperature, 17 h at 40 °C and, finally, 2 h at
120 °C. After cooling to room temperature, dichloromethane
(3 mL) was added and the crude was filtered through a plug of
celite. Then, diethyl ether (50 mL) was added to the yellow solution
to precipitate the product as pale yellow powder. Yield: 115 mg,
68%. 1H NMR (DMSO d6, 25 °C): d1.22 (m, 12H, H12), 1.94 (m, 4H,
H13), 2.10 (m, 2H, H3), 2.22 (m, 2H, H3), 4.00 (m, 8H, H11), 4.72
(m, 2H, H10), 4.82 (m, 2H, H10), 5.07 (m, 2H, H2), 5.65 (m, 2H,
H2), 7.39 (m, 4H, H5 and H8), 7.75 (m, 4H, H6 and H7). 31P{1H}
NMR (DMSO d6, 25 °C): d31.37. 13C{1H} NMR (DMSO d6, 25 °C):
d16.3 (d, J = 5.29 Hz, C12), 21.7 (d, J = 140.79 Hz, C13), 22.2 (d,
J = 3.86 Hz, C3), 43.9 (C2), 48.7 (C10), 61.1 (t, J = 5.93 Hz, C11),
111.3 (C6 and C7), 123.6 (C5), 123.8 (C8), 133.3 (C4), 133.9 (C9). IR
(C7), 131.1 (C2), 141.1 (C1). IR (ATR) cmꢁ1: 2986
m(C-H)ar, 2898 m
(C-H), 1431 (d(C = C), d(C = N))ar, 1222 (P = O)st, 1020 (P-O-C)st,
758 d(C-H)oop. HR-ESI-MS (DMSO): calculated: m/z = 701.2055
([M + Na]+); found: m/z = 701.2064 ([M + Na]+), 310.1442
([Mꢁ2Br]+2).
(ATR) cmꢁ1: 2978
m(C-H), 1464, 1404 (d(C = C), d(C = N))ar, 1229
(P = O)st, 1006 (P-O-C)st, 751 d(C-H)oop. HR-ESI-MS (DMSO): cal-
culated: m/z = 907.0023 ([M + Na]+); found: m/z = 907.0019
([M + Na]+), 805.0943 ([MꢁBr]+).
4.2.6. 1,10-Di(3-diethoxyphosphoryl)propyl-3,30-(trimethylene)
bisbenzimidazolium dibromide (3c)
A mixture of diethyl (3-bromopropyl)phosphonate (348 mL,
1.8 mmol, 2 eq) and 2c (250 mg, 0.9 mmol, 1 eq) was stirred at
80 °C for 4 h. After cooling to room temperature, the crude was
solved in dichloromethane (3 mL) and diethyl ether (25 mL) was
added to precipitate a white hygroscopic solid. Yield: 671 mg,
92%. 1H NMR (CDCl3, 25 °C): d1.28 (t, J = 6.92 Hz, 12H, H12), 1.88
(dt, J = 6.96 Hz, J = 18.57 Hz, 4H, H10), 2.36 (m, 4H, H9), 2.97 (q,
J = 6.98 Hz, 2H, H14), 4.06 (m, 8H, H11), 4.68 (t, J = 7.48 Hz, 4H,
H8), 5.13 (t, J = 7.49 Hz, 4H, H13), 7.58 (t, J = 7.99 Hz, 2H, H5),
7.65 (t, J = 7.48 Hz, 2H, H4), 7.78 (d, J = 8.31 Hz, 2H, H3), 8.53 (d,
J = 8.25 Hz, 2H, H6), 10.94 (s, 2H, H1). 31P NMR (CDCl3, 25 °C):
d29.8. 13C NMR (CDCl3, 25 °C): d16.5 (d, J = 5.86 Hz, C12), 22.2 (d,
J = 135.73 Hz, C10), 22.9 (d, J = 2.19 Hz, C9), 29.6 (C14), 44.5 (C13),
47.2 (d, J = 11.8 Hz, C8), 62.1 (d, J = 6.57 Hz, C11), 112.7 (C3),
115.2 (C6), 127.5 (C5), 127.8 (C4), 131.0 (C7), 131.6 (C2), 142.0
4.3.3. Dibromido-(1,10- di(3-diethoxyphosphoryl)propyl-
3,30propylenedibenzimidazolin-2,20-diylidene)palladium(II) (mC1c)
Palladium(II) acetate (69 mg, 0.31 mmol, 1 eq) and ligand 3c
(250 mg, 0.31 mmol, 1.01 eq) were dissolved in degassed DMSO
(3 mL) in a round-bottom flask. The reddish mixture was stirred
for 4 h at room temperature, 17 h at 40 °C and, finally, 2 h at
120 °C. After cooling to room temperature, dichloromethane
(3 mL) was added and the crude was filtered through a plug of
celite. Then, diethyl ether (50 mL) was added to the yellow solution
to precipitate the product as a pale green powder. Yield: 235 mg,
85%. 1H NMR (DMSO d6, 25 °C): d1.21 (dt,
J = 12.97 Hz,
J = 16.72 Hz, 12H, H12), 1.87 (m, 2H, H9 and 1H, H14), 2.06 (m,
4H, H10), 2.32 (m, 2H, H9), 2.55 (m, 1H, H14), 4.00 (m, 8H, H11),
4.55 (td, J = 3.99 Hz, J = 10.51 Hz, 2H, H8), 4.86 (dd, J = 5.52 Hz,
J = 14.50 Hz, 2H, H13), 5.05 (m, 2H, H8), 5.26 (dd, J = 11.87 Hz,
J = 14.53 Hz, 2H, H13), 7.27 (dd, J = 3.09 Hz, J = 6.02 Hz, 4H, H3
and H6), 7.64 (m, 4H, H4 and H5). 31P{1H} NMR (DMSO d6, 25 °C):
(C1). IR (ATR) cmꢁ1: 2978
m(C-H)ar, 2898 m(C-H), 1437 (d(C = C),
d(C = N))ar, 1229 (P = O)st, 1014 (P-O-C)st, 751 d(C-H)oop. HR-
ESI-MS (DMSO): calculated: m/z = 715.2212 ([M + Na]+); found:
m/z = 715.2220 ([M + Na]+), 317.1521 ([Mꢁ2Br]+2).
144

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
d31.37. 13C{1H} NMR (DMSO d6, 25 °C): d16.4 (dd, J = 4.39 Hz,
filled with water/ethyl acetate mixture (2:1.5, 3.5 mL) and with
naphthalene (0.09 mmol) as standard.
J
= 5.72 Hz, C12), 21.81 (d, J = 3.86 Hz, C9), 21.88 (d,
J = 139.59 Hz, C10), 29.15 (C14), 48.38 (d, J = 19.92 Hz, C8), 48.76
(C13), 61.16 (dd, J = 6.72 Hz, J = 12.21 Hz, C11), 110.86 (C5),
111.28 (C4), 123.30 (C6), 123.50 (C3), 132.88 (C7), 133.99 (C2),
The organic phase was analyzed by GC and GC–MS.
All products gave analytical data in agreement with those
reported in the literature: {p-acetylbiphenyl [92–91-1], 2-
methoxybiphenyl [86–26-0], 4-methoxybiphenyl [613–37-6], 4-
phenyltoluene [644–08-6], 2-chloro-1,1-biphenyl [2051–60-7], 1-
phenylnaphthalene [605–02-7], 1-phenylpyrene [5101–27-9], 4-
phenylisoquinoline [19571–30-3], 3-phenylquinoline [1666–96-
2], 2-phenylnaphthalene [612–94-2], 2-phenyl-9H-fluorene
173.49 (C1). IR (ATR) cmꢁ1: 2986
m(C-H), 1457, 1417 (d(C = C), d
(C = N))ar, 1229 (P = O)st, 1014 (P-O-C)st, 751 d(C-H)oop. HR-
ESI-MS (DMSO): calculated: m/z = 921.0180 ([M + Na]+); found:
m/z = 921.0165 ([M + Na]+), 819.1093 ([MꢁBr]+).
[28065–98-7],
hydroxybiphenyl
3-methoxybiphenyl
[90–43-7], 3,4-dimethoxy-1,1‘-biphenyl
[2113–56-6],
2-
4.4. Synthesis and characterization of the hybrid nanomaterials
[17423–55-1], 1,3-dimethoxy-5-phenylbenzene [64326–17-6], 4-
(trifluoromethyl)-1,1‘-biphenyl [398–36-7], 3-(trifluoromethyl)
biphenyl [366–04-1], 4-cyanobiphenyl [2920–38-9], biphenyl
[613–37-6], 3-fluoro-[1,1‘-biphenyl]-4-carbonitrile [503177–15-
9], 4-fluorobiphenyl [324–74-3], 2,3,4,5,6-pentafluorobiphenyl
[784–14-5], 2-([1,1‘-biphenyl]-3-yl)pyridine [458541–39-4], 5-
phenyl-1H-indole [66616–72-6], 7-phenyl-1H-indole [1863–21-
4]}. For TEM analyses, a drop of the crude mixture was deposited
on the holey carbon-covered copper.
In order to obtain the iTOF values, the evolved mmol of p-
acetylbiphenyl divided by 8 ꢂ 10-4 mmol of Pd vs time (in minutes)
was plotted. Thus, a polynomial equation was fitted for the region
0–3 min and the first derivative was calculated. Finally, the iTOF
values were obtained for ꢂ = 0 for the first derivative equation.
TON values were calculated dividing the evolved mmol of p-
acetylbiphenyl by 8 ꢂ 10-4 mmol of Pd. TOF values were calculated
dividing the TON values by the required time (in hours) to achieve
the maximum conversion for each catalyst.
mC1a-TiO2: A solution of mC1a (16.3 mg, 0.02 mmol, 0.002 eq)
in dichloromethane (4 mL) was added to an ACE pressure tube con-
taining TiO2 P-25 (740 mg, 9.4 mmol, 1 eq). The reaction mixture
was kept under vigorous stirring at 45 °C overnight. Then, the
orange powder was filtered off and washed with dichloromethane
(3 ꢂ 10 mL), methanol (3 ꢂ 10 mL) and diethyl ether (3 ꢂ 10 mL).
Finally, the product was dried under vacuum at 120 °C for 6 h.
Recovered: 709 mg. 31P HPDEC NMR (25 °C, ppm): d26.10, 31.55.
FT-IR (ATR, cmꢁ1): 1037 (P-O-Ti) st. ICP-OES (w/w%): Pd, 0.23; Ti,
56.12; P, 0.11.
mC1b-TiO2: A solution of mC1b (16.4 mg, 0.02 mmol, 0.002 eq)
in dichloromethane (4 mL) was added to an ACE pressure tube con-
taining TiO2 P-25 (740 mg, 9.4 mmol, 1 eq). The reaction mixture
was kept under vigorous stirring at 45 °C overnight. Then, the pow-
der was filtered off and washed with dichloromethane (3 ꢂ 10 mL),
methanol (3 ꢂ 10 mL) and diethyl ether (3 ꢂ 10 mL). Finally, the
product was dried under vacuum at 120 °C for 6 h. Recovered:
693 mg. 31P HPDEC NMR (25 °C, ppm): d26.02, 32.41. FT-IR (ATR,
cmꢁ1): 1052 (P-O-Ti) st. ICP-OES (w/w%): Pd, 0.21; Ti, 54.38; P,
0.11.
4.5.2. Procedures for recycling
Method A. After calculating the remaining volume, the appropri-
ate amount of a stock solution of p-bromoacetophenone, phenyl-
boronic acid and MeONa in EtOH (0.2 M, 0.24 M and 0.3 M,
respectively) was introduced (to reach final concentrations of
0.05 M, 0.06 M and 0.075 M, respectively) in a sealed Glass cat-
alytic reactor of the Radleys Carousel 12 Plus StationTM fitted with
a water-cooled aluminum reflux head and septa. The station was
preheated at 60 °C. The reaction was conducted under vigorous
stirring and followed by GC analysis until completion. Then, a
new aliquot of the stock solution was added keeping the same con-
centration of the reagents as in the previous runs.
Method B. After determining the Pd content in the hybrid cata-
lyst, in a two-neck 250 mL round-bottom flask fitted with a reflux
condenser and septa, the appropriate amounts of p-
bromoacetophenone, phenylboronic acid and MeONa and EtOH
were introduced to reach final concentrations of 0.05 M, 0.06 M
and 0.075 M, respectively. The reaction mixture was preheated at
60 °C. Then, the hybrid catalyst was added at once. The reaction
was conducted under vigorous stirring and followed by GC analysis
until completion. Then, after cooling to room temperature, the
reaction crude was centrifuged to isolate the hybrid catalyst as a
powder. Furthermore, the catalyst was washed with acetone and
dried under vacuum for further analyses and reuse by using the
same Pd loading.
mC1c-TiO2: A solution of mC1c (16.9 mg, 0.02 mmol, 0.002 eq)
in dichloromethane (4 mL) was added to an ACE pressure tube con-
taining TiO2 P-25 (740 mg, 9.4 mmol, 1 eq). The reaction mixture
was kept under vigorous stirring at 45 °C overnight. Then, the pow-
der was filtered off and washed with dichloromethane (3 ꢂ 10 mL),
methanol (3 ꢂ 10 mL) and diethyl ether (3 ꢂ 10 mL). Finally, the
product was dried under vacuum at 120 °C for 6 h. Recovered:
685 mg. 31P HPDEC NMR (25 °C, ppm): d25.24, 31.81. FT-IR (ATR,
cmꢁ1): 1052 (P-O-Ti) st. ICP-OES (w/w%): Pd, 0.19; Ti, 55.28; P,
0.10.
3a-TiO2: A solution of 3a (15 mg, 0.019 mmol, 0.002 eq) in
dichloromethane (4 mL) was added to an ACE pressure tube con-
taining TiO2 P-25 (600 mg, 7.6 mmol, 1 eq). The reaction mixture
was kept under vigorous stirring at 45 °C overnight. Then, the pow-
der was filtered off and washed with dichloromethane (3 ꢂ 10 mL),
methanol (3 ꢂ 10 mL) and diethyl ether (3 ꢂ 10 mL). Finally, the
product was dried under vacuum at 120 °C for 6 h. Recovered:
591 mg. 31P HPDEC NMR (25 °C, ppm): d25.9, 31.1. FT-IR (ATR,
cmꢁ1): 1052 (P-O-Ti) st. ICP-OES (w/w%): Ti, 56.28; P, 0.29.
4.5. Catalytic experiments
4.5.1. Catalytic studies for Suzuki-Miyaura reactions
In glass tubes of a Radleys Carousel 12 Plus StationTM fitted with
a water-cooled aluminum reflux head and septa, 16 mL of an abso-
lute EtOH solution containing p-halogenoacetophenone (0.05 M,
1 eq), phenylboronic acid (0.06 M, 1.2 eq) and MeONa (0.075 M,
1.5 eq) were added. The reaction mixtures were stirred and heated
at 25 °C or 60 °C under nitrogen atmosphere for 30 min. Then, the
palladium catalyst (8 ꢂ 10-4 mmol Pd, 5 ꢂ 10-5 M, 0.001 eq) was
added. The mixture was vigorously stirred and kept at 60 °C for
1 h under nitrogen. Then, an aliquot (ca. 0.5 mL) of the reaction
crude was taken at different reaction times and quenched in a vial
4.5.3. Catalytic performance study for C1c in the presence of 3a-TiO2
In a glass tube of a Radleys Carousel 12 Plus StationTM fitted
with a water-cooled aluminum reflux head and septa, 70 mL of
an absolute EtOH solution containing p-bromoacetophenone
(700 mg, 3.5 mmol, 0.05 M, 1 eq), phenylboronic acid (515 mg,
4.2 mmol, 0.06 M, 1.2 eq), MeONa (285 mg, 5.3 mmol, 0.075 M,
1.5 eq) and 3a-TiO2 (74.8 mg, 0.001 eq ligand-to-palladium) were
added. The reaction mixtures were stirred and heated at 60 °C or
25 °C under nitrogen atmosphere for 30 min. Then, the palladium
145

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
catalyst C1c (3.5 ꢂ 10ꢁ3 mmol Pd, 5 ꢂ 10ꢁ5 M, 0.001 eq) was added
from a stock solution. The mixture was vigorously stirred and kept
at 60 °C for 1 h under nitrogen. Then, an aliquot (ca. 0.5 mL) of the
reaction crude was taken at different reaction times and quenched
in a vial filled with H2O/EtOAc mixture (2:1.5, 3.5 mL) and with
naphthalene (0.09 mmol) as standard. The organic phase was ana-
lyzed by GC and GC-MS.
References
[1] S. Kotha, K. Lahiri, D. Kashinath, Recent applications of the Suzuki-Miyaura
cross-coupling reaction in organic synthesis, Tetrahedron 58 (2002) 9633–
9695.
[2] A. Ganesan, New tools for parallel automated chemistry, Drug Discovery Today
6 (5) (2001) 238–241.
[3] L. Yin, J. Liebscher, Carbon-carbon coupling reactions catalyzed by
heterogeneous palladium catalysts, Chem. Rev. 107 (1) (2007) 133–173.
[4] V. Polshettiwar, C. Len, A. Fihri, Silica-supported palladium: sustainable
catalysts for cross-coupling reactions, Coord. Chem. Rev. 253 (2009) 2599–
2626.
[5] M. Lamblin, L. Nassar-Hardy, J. Hierso, E. Fouquet, F. Felpin, Recyclable
heterogeneous palladium catalysts in pure water: sustainable developments in
Suzuki, Heck, Sonogashira and Tsuji-Trost reactions, Adv. Synth. Catal. 352
(2010) 33–79.
4.5.4. Procedure for the leaching studies by hot-filtration
In glass tubes of the Radleys Carousel 12 Plus StationTM fitted
with a water-cooled aluminum reflux head and septa, 16 mL of
an absolute EtOH solution containing p-bromoacetophenone
(160 mg, 0.8 mmol, 0.05 M, 1 eq), phenylboronic acid (118 mg,
0.96 mmol, 0.06 M, 1.2 eq) and MeONa (65 mg, 1.2 mmol,
0.075 M, 1.5 eq) were added. Next, the reaction mixtures were stir-
red and heated at 60 °C under nitrogen atmosphere for 30 min.
Then, the palladium catalyst (8 ꢂ 10ꢁ4 mmol Pd, 5 ꢂ 10ꢁ5 M,
0.001 eq) was added. The reaction was conducted under vigorous
stirring for 15 min of reaction. The supernatant solution was fil-
tered through a cannula with a microglass WhatmanÒ filter and
CeliteÒ (in order to remove all fine particles). The reaction was
monitored by GC over the total period and the results compared
to a standard catalytic reaction.
[6] S. Navalón, M. Álvaro, H. García, Polymer- and ionic liquid-containing
palladium: recoverable soluble cross-coupling catalysts, ChemCatChem
(2013) 3460–3480.
[7] K.V.S. Ranganath, S. Onitsuka, A.K. Kumar, J. Inanaga, Recent progress of N-
heterocyclic carbenes in heterogeneous catalysis, Catal Sci. Technol. 3 (2013)
2161–2181.
5
[8] J. De Tovar, F. Rataboul, L. Djakovitch, Insights into the Suzuki-Miyaura
reaction catalyzed by novel Pd
tetracarbene entities the key active species?, ChemCatChem 12 (2020)
5797–5808.
– carbene complexes. are palladium –
[9] A. Wolfson, O. Levy-Ontman, Recent developments in the immobilization of
palladium complexes on renewable polysaccharides for Suzuki-Miyaura
cross-coupling of halobenzenes and phenylboronic acids, Catalysts 10
(2020) 136.
[10] Topics in Current Chemistry, Immobilized Catalysts, A. Kirschning (Ed.), vol.
242, Springer-Verlag, Berlin Heidelberg, 2004.
[11] F. Amoroso, S. Colussi, A. Del Zotto, J. Llorca, A. Trovarelli, Room-temperature
Suzuki-Miyaura reaction catalyzed by Pd supported on rare earth oxides:
influence of the point of zero charge on the catalytic activity, Catal. Lett. 143
(2013) 547–554.
[12] A. Del Zotto, D. Zuccaccia, Metallic palladium, PdO, and palladium supported
on metal oxides for the Suzuki-Miyaura cross-coupling reaction: a unified
view of the process of formation of the catalytically active species in solution,
Catal. Sci. Technol. 7 (2017) 3934–3951.
[13] A. Del Zotto, S. Colussi, A. Trovarelli, Pd/REOs catalysts applied to the Suzuki-
Miyaura coupling. a comparison of their catalytic performance and reusability,
Inorganic Chim Acta 470 (2018) 275–283.
4.6. TEM analyses
For the hybrid mC1a-c-TiO2 materials, the as-prepared powder
was dispersed in EtOH followed by deposition of some drops under
the holey carbon-covered copper grid. For catalytic tests, after 1 h
of catalytic reaction, a drop of the crude mixture was deposited
under the holey carbon-covered copper grid for TEM analysis.
mC1x-TiO2 catalysts were dispersed in EtOH and a drop was
deposited under the holey carbon-covered copper grids.
[14] P.O. Asekunowo, R.A. Haque, M.R. Razali, S. Budagumpi, Benzimidazole-based
silver(I) -N-heterocyclic carbene complexes as anti-bacterials: synthesis,
crystal structures and nucleic acids interaction studies, Appl. Organometal.
Chem. 29 (2015) 126–137.
CRediT authorship contribution statement
[15] F. Gómez-Villarraga, J. De Tovar, M. Guerrero, P. Nolis, T. Parella, P. Lecante, N.
Romero, L. Escriche, R. Bofill, J. Ros, X. Sala, K. Philippot, J. García-Antón,
Dissimilar catalytic behavior of molecular or colloidal palladium systems with
a new NHC ligand, Dalton Trans. 46 (2017) 11768–11778.
[16] T. Scherg, S.K. Schneider, S.G. Frey, J. Schwarz, E. Herdtweck, W.A. Hermann,
Bridged imidazolium salts used as precursors for chelating carbene
complexes of palladium in the Mizoroki-Heck reaction, Synlett 18 (2006)
2894–2907.
[17] R. Jothibasu, H.V. Huynh, Syntheses and catalytic activities of Pd(II) dicarbene
and hetero-dicarbene complexes, J. Organomet. Chem. 696 (2011) 3369–3375.
[18] R. Boissezon, J. Muller, V. Beaugeard, S. Monge, J.-J. Robin,
Organophosphonates as anchoring agents onto metal oxide-based materials:
synthesis and applications, RSC Adv. 4 (67) (2014) 35690, https://doi.org/
10.1039/C4RA05414H.
[19] S.A. Paniagua, A.J. Giordano, O.L. Smith, S. Barlow, H. Li, N.R. Armstrong, J.E.
Pemberton, J.-L. Brédas, D. Ginger, S.R. Marder, Phosphonic acids for interfacial
engineering of transparent conductive oxides, Chem. Rev. 12 (2016) 7117–
7158.
Jonathan De Tovar: Conceptualization, Validation, Formal anal-
ysis, Investigation, Writing - Original Draft, Writing - Review and
Editing, Visualization. Franck Rataboul: Writing - Review and
Editing, Visualization, Supervision, Project administration. Laurent
Djakovitch: Conceptualization, Resources, Writing - Review and
Editing, Visualization, Supervision, Project administration, Funding
acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
[20] J. De Tovar, N. Romero, S.A. Denisov, R. Bofill, C. Gimbert-Suriñach, D.
Ciuculescu-Pradins, S. Drouet, A. Llobet, P. Lecante, V. Colliere, Z. Freixa, N.
McClenaghan, C. Amiens, J. García-Antón, K. Philippot, X. Sala, Light-driven
water oxidation using hybrid photosensitizer-decorated Co3O4 nanoparticles,
Mater. Today Energy 9 (2018) 506–515.
[21] M.B. Mitchell, V.N. Sheinker, E.A. Mintz, Adsorption and decomposition of
dimethyl methanephosphonate on metal oxides, J. Phys. Chem. B 101 (1997)
11192–11203.
[22] G. Guerrero, P.H. Mutin, A. Vioux, Anchoring of Phosphonate and Phosphinate
Coupling Molecules on Titania Particles, Chem. Mater. 13 (2001) 4367–4373.
[23] A. Bachinger, S. Ivanovici, G. Kickelbick, Formation of Janus TiO2 nanoparticles
by a pickering emulsion approach applying phosphonate coupling agents, J.
Nanosci. Nanotechnol. 11 (2011) 8599–8608.
[24] L. Djakovitch, K. Koehler, J.G. De Vries, Nanoparticles and Catalysis, The Role of
Palladium Nanoparticles as Catalyst for Carbon – Carbon Coupling Reactions,
D. Astruc (Ed.) Wiley-VCH, Weinheim, 2008, pp. 303–348.
Acknowledgements
Financial support from ANR research program is gratefully
acknowledged (HYPERCAT, contract N° ANR-18-CE07-0021-03).
We thank Dr V. Meille and Dr C. de Bellefon (CP2M, Lyon) for fruit-
ful discussions. We kindly acknowledge L. Burel (TEM, IRCELYON,
N. Bonnet and P. Mascunan (ICP-OES, IRCELYON), Dr A. Demes-
sence (UV-vis, IRCELYON), Dr D. Aldakov (XPS, SyMMES CEA
Grenoble) and the Centre Commun de Spectrométrie de Masse
(University Lyon 1) for analyses and discussions.
Appendix A. Supplementary material
[25] N.T.S. Phan, M. Van Der Sluys, C.W. Jones, On the nature of the active species in
palladium catalyzed Mizoroki-Heck and Suzuki-Miyaura couplings
homogeneous or heterogeneous catalysis, a critical review, Adv. Synth. Catal.
348 (2006) 609–679.
-
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2021.04.016.
146

J. De Tovar, F. Rataboul and L. Djakovitch
Journal of Catalysis 398 (2021) 133–147
[26] G. Borja, A. Monge-Marcet, R. Pleixats, T. Parella, X. Cattoën, M. Wong Chi Man,
Recyclable hybrid silica-based catalysts derived from Pd-NHC complexes for
Suzuki, heck and sonogashira reactions, Eur. J. Org. Chem. (2012) 3625–3635.
[27] J.M. Richardson, C.W. Jones, Strong evidence of solution-phase catalysis
associated with palladium leaching from immobilized thiols during Heck
and Suzuki coupling of aryl iodides, bromides, and chlorides, J. Catal. 251
(2007) 80–93.
[28] S. Chouzier, M. Gruber, L. Djakovitch, New hetero-bimetallic Pd-Cu catalysts
for the one-pot indole synthesis via the Sonogashira reaction, J. Mol. Catal. A:
Chem. 212 (2004) 43–52.
[30] J.A. Byrne, B.R. Eggins, Photoelectrochemistry of oxalate on particulate TiO2
electrodes, J. Electroanal. Chem. 457 (1998) 61–72.
[31] A. Eskandari, M. Jafarpour, A. Rezaeifard, M. Salimi, Supramolecular
photocatalyst of palladium (II) encapsulated within dendrimer on TiO2
nanoparticles for photo-induced Suzuki-Miyaura and Sonogashira cross-
coupling reactions, Appl. Organometal. Chem. 33 (2019) e5093.
[32] M. Sakar, R.M. Prakashi, T.-O. Do, Insights into the TiO2-based photocatalytic
systems and their mechanisms, Catalysts 9 (2019) 680.
[33] R. Santhosh Reddy, R. Mahamadali Shaikh, V. Rawat, P.U. Karabal, G. Dewkar,
G. Suryavanshi, A. Sudalai, A novel synthesis and characterization of titanium
superoxide and its application in organic oxidative processes catal, Suv. Asia
14 (2010) 21–32.
[29] L. Djakovitch, P. Rollet, Sonogashira cross-coupling reactions catalyzed by
copper-free palladium zeolites, Adv. Synth. Catal. 346 (2004) 1782–1792.
147