PdHx Entrapped in a Covalent Triazine Framework Modulates Selectivity in Glycerol Oxidation

Article abstract of DOI:10.1002/cctc.201500055

Pd nanoparticles within a nitrogen-containing covalent triazine framework (CTF) material are investigated to understand if the highly tunable CTF chemistry mediates the catalytic properties of the Pd nanoparticles. Surprisingly, our results demonstrate that the CTF stabilizes the formation of 2.6 nm PdH_x particles within the pores. These confined PdH_x particles are very active for the liquid-phase oxidation of glycerol and promote C-C cleavage, probably connected with the enhanced in situ formation of H₂O₂. During recycling tests, the confined particles are transformed progressively to very stable Pd⁰ particles. This stability has been attributed mainly to a confinement effect as nanoparticles trapped outside the pores lose activity rapidly. These results indicate that there is a potential to tune CTF chemistry to modify the chemistry of the catalytic metals significantly.

Full text of DOI:10.1002/cctc.201500055

DOI: 10.1002/cctc.201500055
Full Papers
PdHx Entrapped in a Covalent Triazine Framework
Modulates Selectivity in Glycerol Oxidation
Carine E. Chan-Thaw,[a] Alberto Villa,[a] Di Wang,[b] Vladimiro Dal Santo,[c]
Alessio Orbelli Biroli,[c] Gabriel M. Veith,[d] Arne Thomas,[e] and Laura Prati*[a]
Pd nanoparticles within a nitrogen-containing covalent triazine
framework (CTF) material are investigated to understand if the
highly tunable CTF chemistry mediates the catalytic properties
of the Pd nanoparticles. Surprisingly, our results demonstrate
that the CTF stabilizes the formation of 2.6 nm PdHx particles
within the pores. These confined PdHx particles are very active
for the liquid-phase oxidation of glycerol and promote CÀC
cleavage, probably connected with the enhanced in situ forma-
tion of H2O2. During recycling tests, the confined particles are
transformed progressively to very stable Pd0 particles. This sta-
bility has been attributed mainly to a confinement effect as
nanoparticles trapped outside the pores lose activity rapidly.
These results indicate that there is a potential to tune CTF
chemistry to modify the chemistry of the catalytic metals sig-
nificantly.
Introduction
In this work, we demonstrate that a covalent triazine frame-
work (CTF) is able to stabilize the in situ formation of PdHx par-
ticles, which mediate the activity and selectivity of this catalyst
directly for the liquid-phase oxidation of glycerol. Moreover,
we validated the role of nascent H2O2 towards CÀC cleavage
experimentally, which provides a new way to tune the chemi-
cal reactivity of these catalysts.
functionalities in the network. As a result of their fully covalent
structure, they possess an increased thermal and chemical sta-
bility. CTFs have been used as the support for metal nanoparti-
cles (NPs) in catalysis.[2] Indeed, we reported previously that
the N heteroatoms within a CTF stabilize PdNPs protected by
polyvinylalcohol (PVA; PdPVA/CTF) during alcohol oxidation
better than activated carbon (PdPVA/AC).[2b–d] Indeed, during the
liquid-phase oxidation of glycerol, PdPVA/AC deactivates imme-
diately in the first run; whereas the same NPs supported on
CTF (PdPVA/CTF) exhibited stable catalytic activity over three
cycles. This result is important because it is particularly chal-
lenging to design and prepare selective and robust catalysts
for liquid-phase reactions that avoid common problems of par-
ticle migration, coarsening, leaching, or modification of the
catalyst support structure.[3] To further increase the stability of
Pd/CTF, we investigated the impact of a preparation technique,
namely impregnation (IMP), that would lead to the confine-
ment of the particles of Pd within the pores of the CTF (PdIMP).
This confinement should result in a new chemical “dial” to con-
trol material properties and tune reactivity.
CTFs are produced by the ionothermal trimerization of car-
bonitrile groups (p-dicyanobenzene) in molten ZnCl2, which
acts as both catalyst and solvent at temperatures of 400–
6008C.[1] At this latter temperature, a dynamic reorganization
of the porous polymer networks takes place. Therefore, CTFs
exhibit very high surface areas and high amounts of nitrogen
[a] Dr. C. E. Chan-Thaw, Dr. A. Villa, Prof. L. Prati
Department of Chemistry
Università degli Studi di Milano
via Golgi 19, 20133 Milano (Italy)
E-mail: laura.prati@unimi.it
[b] Dr. D. Wang
Institute of Nanotechnology and Karlsruhe Nano Micro Facility
Karlsruhe institute of technology (KIT)
Hermann-von-Helmholtz-Platz 1, 76344
Eggenstein-Leopoldshafen (Germany)
Results and Discussion
[c] Dr. V. D. Santo, Dr. A. Orbelli Biroli
CNR-Istituto di Scienze e Tecnologie Molecolari
Via C. Golgi 19, 20133 Milano (Italy)
The TEM overview image of PdIMP/CTF shows that the PdNPs
were ꢀ2.6 nm in size (Table 1, Figure 1b), in close agreement
with the mean pore size of CTF determined by nitrogen ad-
sorption (2.7 nm).[4] Conversely, the use of the sol-immobiliza-
[d] Dr. G. M. Veith
Materials Science and Technology Division
Oak Ridge National Laboratory
Oak Ridge, TN 37831 (USA)
[e] Prof. A. Thomas
Department of Chemistry, Functional Materials
Technische Universität Berlin
Table 1. Mean PdNPs dimensions determined by TEM.
Hardenbergstr. 40, 10623 Berlin (Germany)
Statistical median [nm]
Standard deviation s
This publication is part of a Special Issue on Palladium Catalysis. To view
the complete issue, visit:
http://onlinelibrary.wiley.com/doi/10.1002/cctc.v7.14/issuetoc
PdPVA/CTF
PdIMP/CTF
3.1
2.6
0.7
0.6
ChemCatChem 2015, 7, 2149 – 2154
2149
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 2. Pd content and oxidation state determined by XPS.
Catalyst
Pd content Elemental analysis [%] Pd0 Pdd+
[at%]
C
N
O
[%] [%]
PdPVA/CTF
3.0
0.2
0.2
63.8
83.1
83.8
73.2
7.0
10.2
9.7
26.2
6.5
6.3
65
0
0
35
100
100
40
II
Pd /CTF
IMP
PdIMP/CTF
PdIMP/CTF after reaction 0.3
6.4
20.1
60
Further analysis of the Pd XPS data revealed interesting
trends in the Pd oxidation states. Indeed, exposure to air was
expected to produce partially oxidized Pd species, which hap-
pens in the case of the PdPVA material in which 35% of the Pd
was oxidized (binding energy=338.1 eV; Table 2).
Based on binding energies of known oxides, this Pd is con-
sistent with Pd4+. Nevertheless, care should be taken in the in-
terpretation of this data because of the difficulty to identify ox-
idation states of Pd with aromatic ligands,[5,6] such as the CTF,
formally because of electron-transfer effects with the ligands
or because of initial/final state effects that influence the inter-
pretation of XPS spectra of NPs.[7]
For this reason, we will describe the Pd as Pdd+ (Table 2).
The remaining Pd was consistent with metallic Pd particles
(binding energy=335.8 eV; Figure 2top). Surprisingly, we
found only oxidized Pd in the PdIMP/CTF sample (Fig-
ure 2bottom) as in the sample not yet reduced with NaBH4
II
IMP
(Pd /CTF) (Figure 2middle; binding energy=388.1 eV). If re-
oxidation occurred, we would expect the XPS data to be simi-
lar to that of the PdPVA samples because of the similar particle
sizes, that is, a mixture of reduced and oxidized Pd. Therefore,
we explored the possible diffusion limitation of the NaBH4 re-
ductant by the pore structure, which effectively shields the
Pdd+, and analyzed the Pdd+IMP/CTF after reduction with H2.
Again, there was no indication of reduced Pd. As there should
be no diffusion limitation of H2 during the reduction step, it is
likely that the CTF network stabilizes the oxidized Pd. The pres-
ence of oxidized Pd in PdHx would account for the XPS binding
energy shifts observed and explain the insensitivity to H2 re-
duction. It is likely that the stabilization of oxidized PdHx spe-
cies is enabled by the CTF.
Figure 1. TEM images of a) PdPVA/CTF and b) PdIMP/CTF.
High-resolution transmission electron microscopy (HRTEM)
was performed to gain an insight into the structure and the
composition of the PdIMP/CTF sample (Figure 3). HRTEM studies
on over 100 Pd particles of the PdIMP/CTF catalyst revealed that
most of the Pd particles have a face-centered cubic (fcc) struc-
ture with a unit cell parameter averaged at (4.01Æ0.057) Š,
which is considerably larger than that of pure metallic Pd
(3.89 Š). It is known that deceptive lattice spacings could
appear because of the complication of NPs that tilt randomly
away from low-index-zone axis, particle shape effect, multiply
twinned structures, etc., which could lead to a large deviation
from the real lattice spacing and bending of the lattice planes
in HRTEM. To avoid any possible misinterpretation of the ex-
panded Pd lattice measured here, we selected the particles
that exhibited a high contrast of lattices without lattice bend-
tion technique (PdPVA/CTF) led to a PdNP size of 3.1 nm, which
depends on the preformed metallic sol (Table 1, Figure 1a).
Pd X-ray photoelectron spectroscopy (XPS) revealed a 93%
reduction in Pd intensity of the PdIMP sample compared to that
of the PdPVA material despite the identical 1 wt% metal load-
ings (XPS signal=3.0 at% Pd for PdPVA/CTF; 0.2 at% Pd for
PdIMP/CTF; Table 2). The high dispersion and narrow distribu-
tion of the Pd particle size with the lack of large Pd particles
evidenced by microscopy (Figure 1), indicate that the reduc-
tion in Pd XPS intensity is not caused by an enlargement of
the Pd particle size but, most probably, because of a decrease
in the concentration of Pd in the top few nanometers of the
external surface. This would confirm the confinement of Pd
particles within the CTF structure.
ChemCatChem 2015, 7, 2149 – 2154
www.chemcatchem.org
2150
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ing or abrupt shift carefully. The measured lattice spacings
were derived by averaging at least seven consecutive lattice
planes for each particle. According to the HRTEM simulation of
the particles at different orientations, the variation of the Pd
(111) lattice spacing should not exceed 1.2%, which is much
smaller than the present case of ꢀ3% expansion. Furthermore,
more than 100 particles were measured for (111) or (200) lat-
tice spacings and their random orientation should statistically
minimize the deviation caused by crystal tilting. The measured
lattice spacing value does not match any Pd oxide phase, but
is identical to the lattice of PdHx,[8] which suggests that H was
introduced into the Pd lattices during the reduction treatment
by NaBH4. The presence of PdHx (x>0–0.7) species have been
demonstrated after NaBH4 treatment in the case of PdNPs em-
bedded in Zr-modified mesoporous SiO2 films.[9] Furthermore,
in situ heating of the PdIMP/CTF catalyst to 4008C by using
a Gatan heating holder in TEM led to a decrease of the lattice
parameter to 3.94 Š, which was caused by the desorption of H.
Similar results have been reported upon hydrogenation and
dehydrogenation of a Pd film in environmental TEM.[10]
Temperature-programmed (palladium) hydride decomposi-
tion (TPHD) experiments further confirm the presence of palla-
dium hydrides stable under ambient conditions as appreciable
hydrogen desorption occurs at low and high temperatures
with onsets at 40 and 3258C, respectively. Although the first
hydrogen desorption peak occurs at PdHx decomposition tem-
peratures, typical of b-PdH in supported PdNPs,[11] the second
one occurs at an exceptionally high temperature. However, as
a peak that parallels the onset and shape of the high tempera-
ture m/z=2 peak is evident in the m/z=27 signal (HCN), a par-
tial decomposition of the CTF can occur. Thus, the hydrogen
evolution can be attributed to Pd-catalyzed dehydrogenation.
The activity and selectivity of Pd/CTF catalysts were tested
for the oxidation of glycerol. The reaction profiles obtained
with PdIMP/CTF and PdPVA/CTF are compared in Figure 4. PdIMP
/
CTF, which contains only Pdd+, showed a surprisingly good cat-
alytic activity, almost as good as that of PdPVA/CTF. The slight
differences in activity cannot be attributed to a different parti-
cle size as TEM revealed a comparable size distribution in both
cases.
II
Figure 2. XPS data collected for PdPVA (top), Pd
(middle), and reduced
IMP
PdIMP (bottom). The low PdIMP signal is because Pd is incorporated within the
CTF structure rather than confined on the outer surface of the CTF (PdPVA).
Figure 3. HRTEM image of PdIMP/CTF with enlarged (111) lattice spacing,
which corresponds to a lattice parameter of 4.01 Š.
Figure 4. Reaction profiles for glycerol oxidation (0.3m, NaOH/glycerol ra-
tio=4 mol/mol, glycerol/metal: 1000 mol/mol, 3 atm O2, and 508C)
ChemCatChem 2015, 7, 2149 – 2154
www.chemcatchem.org
2151
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The lower activity of PdIMP/CTF
than that of PdPVA/CTF could be
attributed to the lower Pd expo-
sure of the PdIMP sample, which
is close to that reported for simi-
lar reactions.[12,13] However, as
noted previously, PdIMP/CTF ap-
peared to contain significant
concentrations of oxidized Pd. If
the catalysts remained as Pdd+
,
there should be a significant de-
crease in activity as PdII is known
Scheme 1. Accepted mechanism of metal-catalyzed glycerol oxidation.[16b]
to become active as an oxida-
tion catalyst with oxygen under
conditions harsher than those
used in this study.[14]
hanced H2O2 production, the desorption of the primary oxida-
tion product (C3 products) could be hindered because the re-
action sites are located within the CTF structure, which would
thus favor subsequent oxidation (i.e., CÀC bond cleavage).[18]
The evolution of the PdHx species can be evaluated by recy-
cling the catalyst, which also constitutes an important practical
aspect. We compared the activity of PdIMP/CTF and PdPVA/CTF in
a nine-run recycling test (Figure 5). The recycling tests were
performed by simply filtering the catalyst and reusing it in the
To identify the real nature and role of the Pdd+ in the catalyt-
ic activity, PdIMP/CTF was tested before the reduction step,
II
IMP
II
namely, Pd /CTF. The unreduced Pd /CTF showed a fairly
IMP
good activity and reached 35% conversion after 4 h (Figure 4).
These findings possibly suggested an in situ reduction of Pd
inside the CTF framework. Indeed, XPS analysis of the used
PdIMP/CTF confirmed the presence of ꢀ60% Pd0 (Table 2),
which was likely produced from the reaction with glycerol or
its reaction products.
In terms of selectivity, the PdPVA/CTF catalyst showed
a higher selectivity to glyceric acid (81%) than PdIMP/CTF (62%;
Table 3). The higher production of C2 products in the case of
Table 3. Selectivity of PdPVA/CTF and PdIMP/CTF for glycerol oxidation.[a]
Catalyst
Selectivity at 90% conversion [%]
Glyceric acid
Glycolic acid
Oxalic acid
Tartronic acid
PdPVA/CTF
PdIMP/CTF
81
62
4
12
3
6
12
18
[a] Reaction conditions: 0.3m, NaOH/glycerol ratio=4 mol/mol, glycerol/
metal: 1000 mol/mol, 3 atm O2, 508C.
Figure 5. Conversion (4 h) of 1 wt% PdPVA/CTF and 1 wt% PdIMP/CTF upon re-
cycling.
PdIMP/CTF compared to that of PdPVA/CTF cannot be justified in
terms of the Pd particle size, which is similar (3.1 and 2.6 nm
for PdPVA/CTF and PdIMP/CTF, respectively).
Therefore, the hydride particles in the PdIMP/CTF are likely to
be responsible for the different selectivity of PdIMP/CTF with re-
spect to PdPVA/CTF. If we consider the accepted reaction mech-
anism for glycerol oxidation (Scheme 1) in which a metal hy-
dride is the intermediate formed, we could depict two differ-
ent reaction pathways for PdHx species: 1) the PdHx could be
transformed in situ by reaction with glycerol to form Pd0 and
an oxidation product or 2) the PdHx could reduce O2 directly to
generate H2O2 in higher concentrations than that generated
during the first reaction step (Scheme 1).
successive run without any further purification steps.[2b] PdIMP
/
CTF showed very little deactivation (less than 5%) in nine sub-
sequent runs (Figure 5) according to the physical/chemical sta-
bilization expected to be provided by the embedding of the
PdNPs inside the support framework. The small decrease of ac-
tivity observed from the fifth run is attributed to the partial
blocking of the active sites by adsorbed species. Indeed, XPS
revealed a higher oxygen content in the used catalyst com-
pared to the fresh one consistent with organic species trapped
on the surface (Table 2).
As in situ H2O2 production is expected to promote CÀC
After a single run, we also proved the absence of any cata-
lytically active soluble specie by filtering the hot solution and
letting the reaction proceed. No further conversion was ob-
served in either case.
cleavage,[15–17] this higher H2O2 local production would lead to
a higher selectivity toward CÀC cleavage products on PdIMP
/
CTF than that on PdPVA/CTF. Notably, in addition to the en-
ChemCatChem 2015, 7, 2149 – 2154
www.chemcatchem.org
2152
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Furthermore, inductively coupled plasma (ICP) analysis
showed that only 1% of the initial Pd amount was lost after
nine runs. Conversely, under similar conditions, PdPVA/CTF
showed a constant deactivation from the fourth run with
a total Pd leaching of 6% (measured by ICP). The selectivity
trend confirmed the transient presence of PdHx that decreased
during the reaction as highlighted by XPS. Indeed, the selectiv-
ity shown by PdIMP/CTF changed to become similar to that of
PdPVA/CTF on recycling (Table 4).
Sol immobilization
Pd sol: Na2PdCl4·2H2O (0.043 mmol) and freshly prepared PVA solu-
tion (2 wt%, 880 mL) were added to water (130 mL). After 3 min,
NaBH4 solution (0.1m, 860 mL) was added to the yellow-brown so-
lution under vigorous magnetic stirring. The brown Pd0 sol was
formed immediately. A UV/Vis spectrum of the Pd sol was recorded
to ensure the complete reduction of PdII. Within few minutes of its
generation, the suspension was acidified at pH 2 with sulfuric acid,
and the support was added under vigorous stirring. The catalyst
was collected by filtration and washed several times with distilled
water. The samples were dried at 808C for 2 h. The amount of sup-
port was calculated to obtain a final metal loading of 1 wt%.
Table 4. Selectivity to glyceric acid (Sga) during the recycling of PdIMP/CTF
(4 h reaction).
Runs
1
2
3
4
5
6
7
8
9
Sga [%]
62
56
64
58
70
73
84
79
80
Impregnation
Na2PdCl4·2H2O (0.235 mmol) diluted in water (20 mL) was added to
the catalyst (0.5 g) and stirred vigorously for 6 h. NaBH4 solution
(Pd/NaBH4 1:16) was added to the solution, and it was stirred for
6 h. The catalyst was collected by filtration and washed several
times to ensure the removal of the material that arose from the re-
duction treatment. The samples were dried at 808C for 2 h. The
amount of support was calculated to obtain a final metal loading
of 1 wt%. The obtained catalyst was labeled PdIMP/CTF.
Conclusions
Herein, we demonstrated that the NaBH4 reduction of PdII spe-
cies inside a covalent triazine framework generates partly
stable hydride species that have a significant effect on the se-
lectivity in the liquid-phase oxidation of glycerol, which pro-
vides an indirect confirmation of the mechanism of glycerol
oxidation in the presence of O2 catalyzed by metal. From an
application point of view, the physical confinement of Pd
nanoparticles within the covalent triazine framework showed
a beneficial effect on the durability of the catalyst.
Catalyst characterization
Morphology and microstructures of the catalysts were character-
ized by TEM. The powder samples of the catalysts were dispersed
ultrasonically in ethanol and mounted onto copper grids covered
with holey carbon film. A Philips CM200 LaB6 electron microscope
operated at 200 kV and equipped with a Gatan CCD camera was
used for TEM observation. A FEI Titan 80–300 aberration-corrected
electron microscope operated at 300 kV was used for HRTEM imag-
ing.
Experimental Section
Materials
1,4-Dicyanobenzene (DCB) was purchased from Alfa Aesar and was
used as received. Zinc chloride (Alfa Aesar, anhydrous, 98%) was
stored in a glovebox and used as received. Na2PdCl4·2H2O from Al-
drich (99.99% purity), NaBH4 of purity>96% from Fluka, polyviny-
lalcohol (PVA; Mw=13000–23000; 87–89% hydrolyzed) from Al-
drich were used. Gaseous oxygen from SIAD was 99.99% pure.
ICP analysis was performed by using a Jobin Yvon JV24 to verify
the quantitative metal loading on the support. The final total Pd
loading was 1 wt%.
XPS data were collected by using a PHI 3056 spectrometer with an
Al anode source operated at 15 kV and an applied power of 350 W.
Samples were pressed manually between two pieces of In foil; the
piece of In foil with the sample on it was then mounted on the
sample holder with a piece of carbon tape (Nisshin E.M. Co. LTD).
Adventitious carbon was used to calibrate the binding energy
shifts of the sample (C1s=284.8 eV).[19] High-resolution data were
collected at a pass energy of 5.85 eV with 0.05 eV step sizes and
a minimum of 200 scans to improve the signal-to-noise ratio; low-
resolution survey scans were collected at a pass energy of 93.5 eV
with 0.5 eV step sizes and a minimum of 25 scans.
Catalyst preparation
The support was prepared according to Ref. [1] by modifying the
preparation as described by Chan-Thaw et al.[2b] The monomer DCB
and ZnCl2, in a molar ratio of 1:5, were transferred into a quartz
ampoule (3”12 cm) under an inert atmosphere. The ampoule was
then evacuated, sealed, and heated within 30 min to 4008C. After
20 h at 4008C, the ampoule was further heated within 1 h to
6008C and it was held for additional 20 h. The ampoule was then
cooled to RT and opened. The resulting reaction mixture was
washed several times with dilute HCl (0.1m) under stirring to
remove most of the ZnCl2. The resulting black powder was washed
with water until a neutral pH was obtained. After filtration, the ma-
terial was dried at 1508C and finely ground to give the CTF in
a yield of 98%. The residual Zn content was checked by ICP, which
resulted in a maximum amount of 0.3 wt%.
TPHD was performed by using a home-made reaction rig. We
placed 100 mg of sample in a Pyrex tubular reactor inserted in
a furnace. The sample was heated under He flow (10 mLminÀ1) at
a rate of 108CminÀ1. Decomposition products were detected by
using a quadrupole MS (VGQ, Thermo electron Corporation) con-
nected downstream. Hydrogen evolution was monitored by follow-
ing the m/z=2 signal, in addition the signals at m/z=12, 14, 16,
18, 26, 27, 28, 44, 45, and 77 were collected to reveal any eventual
decomposition of the CTF material.
ChemCatChem 2015, 7, 2149 – 2154
www.chemcatchem.org
2153
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[1] a) P. Kuhn, M. Antonietti, A. Thomas, Angew. Chem. Int. Ed. 2008, 47,
3450–3453; Angew. Chem. 2008, 120, 3499–3502; b) P. Kuhn, A. Forget,
J. Hartmann, A. Thomas, M. Antonietti, Adv. Mater. 2009, 21, 897–901;
c) A. Thomas, Angew. Chem. Int. Ed. 2010, 49, 8328–8344; Angew. Chem.
2010, 122, 8506–8523.
[2] a) R. Palkovits, M. Antonietti, P. Kuhn, A. Thomas, F. Schüth, Angew.
Chem. Int. Ed. 2009, 48, 6909–6912; Angew. Chem. 2009, 121, 7042–
7045; b) C. E. Chan-Thaw, A. Villa, P. Katekomol, D. Su, A. Thomas, L.
Prati, Nano Lett. 2010, 10, 537–541; c) C. E. Chan-Thaw, A. Villa, L. Prati,
A. Thomas, Chem. Eur. J. 2011, 17, 1052–1057; d) C. E. Chan-Thaw, A.
Villa, G. M. Veith, K. Kailasam, L. A. Adamczyk, R. R. Unocic, L. Prati, A.
Thomas, Chem. Asian J. 2012, 7, 387–393.
[3] T. Mallat, A. Baiker, Catal. Today 1994, 19, 247–283.
[4] P. Kuhn, A. Forget, D. S. Su, A. Thomas, M. Antonietti, J. Am. Chem. Soc.
2008, 130, 13333–13337.
[5] Z. M. Michalska, B. Ostaszewski, J. Zientarska, J. W. Sobczak, J. Mol.
Catal. A 1998, 129, 207–218.
[6] M. Heckenroth, E. Kluser, A. Neels, M. Albrecht, Angew. Chem. Int. Ed.
2007, 46, 6293–6296; Angew. Chem. 2007, 119, 6409–6412.
[7] W. P. Zhou, A. Lewera, R. Larsen, R. I. Masel, P. S. Bagus, A. Wieckowski, J.
Phys. Chem. B 2006, 110, 13393–13398.
Catalytic test
The reactions were performed in a thermostatted glass reactor
(30 mL) equipped with an electronically controlled magnetic stirrer
connected to a large reservoir (5000 mL) that contained oxygen at
3 atm. The oxygen uptake was followed by using a mass flow con-
troller connected to a PC through an A/D board, and we plotted
flow versus time. Glycerol solution (0.3m, NaOH/glycerol ratio=
4 mol/mol) and the catalyst (substrate/metal) 1000 mol/mol were
mixed in distilled water (total volume 10 mL). The reactor was pres-
surized at the desired pressure of O2 and thermostatted at 508C.
The reaction was initiated by stirring. Samples were removed peri-
odically and analyzed by HPLC using an Alltech OA-10308,
300 mm”7.8 mm column and UV and refractive index (RI) detec-
tion to analyze the mixture of the samples. H3PO4 0.1% solution
was used as the eluent. The identification of the possible products
was performed by comparison with the original samples.
Recycling tests
[8] W. Wunderlich, M. Tanemura in Advances in Solid State Physics, Vol. 43
(Ed.: B. Kramer), Springer, 2003, pp. 171–180.
[9] a) J. Saha, A. Dandapat, G. De, J. Mater. Chem. 2011, 21, 11482–11485;
b) M. Khanuja, B. R. Mehta, P. Agar, P. K. Kulriya, D. K. Avasthi, J. Appl.
Phys. 2009, 106, 093515–093522.
Each run was performed under the same conditions. The catalyst
was recycled in the subsequent run after filtration without any fur-
ther treatment using a fresh glycerol/NaOH solution.
[10] T. Yokosawa, T. Alan, G. Pandraud, B. Dam, H. Zandbergen, Ultramicro-
scopy 2012, 112, 47–52.
[11] M. Bonarowska, W. Raróg-Pilecka, Z. Karpinski, Catal. Today 2011, 169,
223–231.
[12] S. Biella, F. Porta, L. Prati, M. Rossi, Catal. Lett. 2003, 90, 23–29.
[13] A. Villa, A. Gaiassi, I. Rossetti, C. L. Bianchi, K. van Benthem, G. M. Veith,
L. Prati, J. Catal. 2010, 275, 108–116.
Hot-filtration test
After the first run, the hot slurry (508C) was filtered to remove the
catalyst, and the solution was stirred for an additional 1 h under O2
pressure (3 atm).
[14] a) B. P. Buffin, N. L. Belitz, S. L. Verbeke, J. Mol. Catal. A 2008, 284, 149–
ˇ
154; b) V. Jurcík, T. E. Schmid, Q. Dumont, A. M. Z. Slawin, C. S. J. Cazin,
Acknowledgements
Dalton Trans. 2012, 41, 12619–12623; c) C. Melero, O. N. Shishilov, E. l-
varez, P. Palma, J. Cµmpora, Dalton Trans. 2012, 41, 14087–14100.
[15] W. C. Ketchie, Y. Fang, M. S. Wong, M. Murayama, R. J. Davis, J. Catal.
2007, 250, 94–101.
[16] a) L. Prati, P. Spontoni, A. Gaiassi, Top. Catal. 2009, 52, 288–296; b) D.
Wang, A. Villa, D. Su, L. Prati, R. Schlçgl, ChemCatChem 2013, 5, 2717–
2723.
[17] M. Sankar, N. Dimitratos, D. W. Knight, A. F. Carley, R. Tiruvalam, C. J.
Kiely, D. Thomas, G. J. Hutchings, ChemSusChem 2009, 2, 1145–1151.
[18] A. Villa, C. E. Chan-Thaw, L. Prati, Appl. Catal. B 2010, 96, 541–547.
[19] J. F. Moulder, W. F. Stickle, P. E. Solbol, K. D. Bomben, Handbook of X-ray
Photoelectron Spectroscopy, PerkinElmer Corp., 1992.
We thank Phisan Katekomol for the preparation of the support.
Financial support by the UniCat cluster of excellence (Unifying
Concepts in Catalysis, Berlin) is gratefully acknowledged. A por-
tion of this research was sponsored by the U.S. Department of
Energy, Office of Basic Energy Sciences, Materials Sciences and
Engineering Division (GMV). TEM characterization was performed
at KIT and sponsored by Karlsruhe Nano Micro Facility (KNMF).
Keywords: cleavage reactions · heterogeneous catalysis ·
nitrogen heterocycles · oxidation · palladium
Received: January 21, 2015
Revised: March 2, 2015
Published online on June 26, 2015
ChemCatChem 2015, 7, 2149 – 2154
www.chemcatchem.org
2154
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim