Catalysis Today 285 (2017) 114–124

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Structural and surface properties of heterogeneous catalysts: Nature

of the oxide carrier and supported particle size effects
Anna Maria Venezia ∗ , Valeria La Parola, Leonarda Francesca Liotta
Institute of Nanostructured Materials (ISMN-CNR), Via Ugo La Malfa 153, 90146 Palermo, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Heterogeneous catalysts, investigated by several techniques, such as XRD, XPS, TEM, TPR/TPO, able to
Received 14 September 2016 ascertain their structural, chemical, electronic and morphology properties are reviewed with respect
Received in revised form 2 November 2016 to selected results obtained by the group during approximately the last twenty five years. The field of
Accepted 6 November 2016
application covers different catalytic processes such as alkene hydrogenation over noble metal cata-
Available online 22 November 2016
lysts, hydrodesulfurization (HDS) and environmental catalysis. In particular the following subjects are
described: the peculiarity of a support like pumice, in relation to the hydrogenation activity of the sup-
Keywords:
ported noble metals; supported gold and supported gold containing alloy catalysts for different types
Diene hydrogenation
HDS catalysts
of reactions; the importance of the CoMoS precursors and the structure of bimetallic PdAu and PtAu for
Au/CeO2 hydrotreating processes; the effect of a sulfatable oxide, such as TiO2 , on the CH4 oxidation activity of
CO oxidation supported Pd catalysts; different preparation methods influencing significantly the catalytic behavior of
Pd-Au catalysts supported nickel catalysts for methane conversion processes. The reviewing of the different case studies
CH4 oxidation illustrates the importance of the interaction between support and active metals ultimately determining
the surface distribution of the active sites and their final catalytic activity.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction photoelectron and Auger spectroscopy, using laboratory or Syn-

Catalysis and in particular heterogeneous catalysis, a technology
devoted to the enhancement of chemical process with increased
conversion and selectivity, relies very much on the structural
design and the surface properties of the material used as catalyst. As
commonly described in fundamental chemistry books, a catalytic
process involves a sequence of elementary steps, including, adsorp-
tion, surface diffusion, chemical rearrangement of the adsorbed
reaction intermediate and desorption of the products [1,2]. Each
of these steps is affected by the nature of the catalyst, typically but
not always, formed by an inert support and by the active species
present as metallic or as oxide state usually deposited at the surface
of the supports. A detailed structural and surface characterization
is therefore mandatory in order to understand the behavior of a
catalytic material and to be able to enhance its activity and its
selectivity in a particular reaction [3].
The advancement of our scientific career in the field of catalysis
has strongly relied on the knowledge of surface science and mate-
rial characterization tools [4–9]. Indeed the application of X-ray
∗ Corresponding author.
E-mail address: annamaria.venezia@ismn.cnr.it (A.M. Venezia).
chrotron facilities allowed to understand surface modification after
catalytic reactions or during adsorption of molecules [10].
When considering the case of supported metal catalyst, the
surface distribution of the active metal plays the main role [11].
Therefore, it is crucial to exploit synthetic routes allowing to obtain
catalysts with the highest exposed surface. In other words, it is
very important to achieve active site phases highly dispersed, with
the highest surface to volume ratio and with the least amount of
metal, saving also on the total cost of a catalyst [12]. Particles in the
nano size range are suitable for heterogeneous catalysis, however
it is important to keep in mind that the metallic properties and the
consequent catalytic benefits, depending on the particular reaction,
may be lost below a certain size [13,14]. As we are aware of, most
recently, advances in syntheses and characterization techniques
had allowed to demonstrate the role of even sub-nanometers parti-
cles in the enhancement of the catalytic oxidation activity [15,16].
Obviously, the design of a good catalyst takes into account the
nature of the support, either “inert” e.g. carbon, SiO2 , and to a
less extent Al2 O3 , or “active” such as the reducible oxides, e.g.
CeO2 , TiO2 [17,18]. In the first case the inert support acts as just
a dispersing carrier, whereas in the second case, the interaction
with the anchored active phase, modulated by different synthetic
approaches, affects the catalytic behavior [19]. Beside the chem-

http://dx.doi.org/10.1016/j.cattod.2016.11.004
0920-5861/© 2016 Elsevier B.V. All rights reserved.
 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
istry, the texture and the morphology of the supports are also
important. It is generally acknowledged that a high surface area
support is positive for the catalyst activity, allowing for a better
metal dispersion. However the use of a low surface area support,
when suitable syntheses succeed in making catalysts with high
metal dispersion, can improve the supported metal signal of a sur-
face technique like XPS, therefore allowing a better understanding
of the experimental results [4].
A way to enhance the catalytic behavior of a metal active species
is to add a second metal producing the so called “alloying” effect
[14,20]. Beside the main catalyst components, small amount of ele-
ments may be naturally present in the support or may be purposely
added, acting as structural or as chemical promoter [21–23].
Since none of the above effects works independently of each
other, for a better understanding of their role, we had tried to sep-
arate them by appropriate design of the catalysts. Obviously the
nature of the probing reaction must govern the selection of the
catalysts. The main achievements of our research activity carried
out during the last 25 years are summarized in the present review
article. Due to space limitations we will leave out some recent
work related to biomass valorization [24–27]. Different study cases
are grouped under two important types of test reactions, the cat-
alytic hydrogenation and the catalytic oxidation. Hydrotreating
reaction, hydrodesulfurization (HDS) and hydrodearomatization
(HDA), although involving hydrogenation steps are treated sep-
arately because of the different target and the peculiarity of the
active sites. For each typology of reaction the importance of the
support nature, with and without modifiers, the structure of the
active site, as single or multi-metal, will be demonstrated.
2. Experimental
2.1. Support and catalyst preparation
Different types of oxides, either commercially available either
laboratory made were used as catalyst support. Sol-gel, co-
precipitation or hydrothermal syntheses were employed when
a particular control of the morphology, chemical and electronic
properties was needed. The anchoring of the active metal was per-
formed through a variety of procedures, depending on the final
objective and application. Just to cite a few examples, in order
to obtain a high dispersion of palladium particle over a low sur-
face area support, an organometallic procedure, derived from the
Yermakov’s method, with the diallyl Pd(␩3 -C3 H5 )2 precursor was
adopted [28,29]; gold catalysts required a carefully pH controlled
deposition precipitation method [30]; solvated metal atom depo-
sition (SMAD) procedure was modified to attain high dispersed
gold particles [30,31]; CoMo catalysts were prepared by classical
impregnation and also by ligand (NTA) assisted procedures [32]; the
preparation of bimetallic particles with different degree of alloy-
ing and different core-shell structures involved also a preferential
interaction with appropriate functional groups previously attached
to the supports [33]; furthermore, microwave assisted procedures
were used in the latest investigations [34].
2.2. Catalyst characterization
The catalysts, before and after the catalytic tests were char-
acterized by several structural and surface techniques. The most
commonly used ones were XRD, SAXS, XPS, TPR/TPO, TGA/DTA,
SEM,TEM, FT-IR. Temperature programmed desorption of ammo-
nia (NH3 -TPD) was used for the determination of the oxide support
acidity.
115
Table 1
Metal loadings of different pumice supported Pd catalysts, (wt.-%), surface metal
coverage (f) and diameters, d (Å) of particles as derived by SAXS.
Samplea Pd (wt%) f db
W0 0.05 0.005 18
W1 0.20 0.01 22
W2 0.11 0.006 29
W3 0.39 0.02 35
W4 0.61 0.03 35
W5 0.86 0.04 40
W6 0.37 0.01 50
W7 1.05 0.02 81
From Venezia et al. [13] with permission.
2.3. Catalytic activity
Different reactor set ups which are described in the published
papers were used. Dienes hydrogenations were performed in liquid
phase batch reactors, whereas gas phase reactions of CO, methane
and VOC oxidation were carried out in plug −flow reactors.
3. Results and discussion
3.1. Hydrogenation
3.1.1. Monometallic catalysts
Pumice, a natural stone of volcanic origin found in the nearby
island of Lipari, was used as catalyst support. Several studies related
to the catalytic effects of the supported metal particle sizes and
the alkali promotion have been published [13,35–40]. After a mild
acid treatment, the material, an amorphous aluminosilicate, had
a reproducible final composition of 85% of SiO2 , 7% of Al2 O3 with
minor amounts of Na2 O and K2 O [36]. Being characterized by a
low surface area (∼5 m2 /g) it was crucial to adopt metal sup-
porting procedures able to maximize the metal dispersion. In this
sense very good results were achieved using a procedure leading
to an allyl intermediates with Pd(␩3 -C3 H5 )2 precursor [29]. The
obtained series of pumice supported catalysts (w series) with dif-
ferent amounts of noble metal are listed in Table 1 along with the
surface metal coverage (f) calculated as f = x/(1-x) pd dS00 , where
x was the palladium weight fraction in the final catalyst, pd the
density of palladium, d the palladium particle diameter derived
by SAXS measurements and S0 the support specific surface area
[13]. These catalysts were tested in the selective hydrogenation
of 1,3-cyclooctadiene (1,3-COD) to mono-ene in liquid phase [29].
Surprisingly, the catalysts behaved quite differently from classical
catalysts of Pd over pure alumina or silica [12]. In those cases a dras-
tic decrease of the activity in terms of TOF was observed for metal
dispersion higher than 20% (particle size smaller than 60 Å). Such
behavior was explained with the rapid loss of metallic character of
the palladium crystallites associated to a positive chemical shift of
the Pd 3d photoelectron peaks in the X-ray photoelectron spectra
(XPS). On the contrary, the pumice supported catalysts exhibited a
constant turnover frequency (TOF) for metal particles larger than
30–40 Å and progressively smaller values as the dimensions of the
particles decreased below 30 Å [29]. At the same time a negative
Pd 3d chemical shift was observed [40]. A lowering of the bind-
ing energy with the increase of the metal dispersion suggested the
presence of another effect, opposite to the loss of metallic character.
XPS of the valence band and X-ray excited O KLL Auger transi-
tion revealed an interaction between pumice and metal, consistent
with an electron transfer from the support to the metal [39,40].
Through the combination of core electron binding energy shifts
and Auger transition shifts it was possible to calculate the Auger
parameters, which, being independent on the charging and energy
reference, yielded more reliable information on the charge distri-
116 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 1. Palladium Auger parameters and Turnover frequencies (TOF) relative to the selective hydrogen reduction of the 1,3-COD to mono-ene versus palladium particle size.
From Venezia et al. [13] with permission.
bution over the supported metal [41]. As shown in Fig. 1, a direct
correlation existed between the Auger parameters and the TOF’s for
the 1,3-COD hydrogenation both plotted versus palladium particle
size. Therefore, the decrease of the activity with high metal dis-
persion was related to the inhibited activation of the electron rich
diene when the electron density on the supported metal became
too high. By comparison with alkali free silica or alumina catalysts,
the presence of sodium and potassium in the structure of pumice
was considered responsible for such behavior. Such findings were
reproduced with synthetic pumice containing various amount of an
alkali metal, prepared by sol-gel procedure [22,23,42]. It was sug-
gested that the alkali ions would increase the electron density at
the metal particle through the formation of an electron donor com-
posite metal-oxygen- Na. The mechanism was essentially the same
either when the alkali ions were added as surface dopants or when
they were inside the support structure. However, when sodium
ions were deposited at the surface the decrease of the ensemble
size due to the partial coverage of the palladium active sites reduced
the positive effect of the increased electron density on palladium
and therefore the activity of the catalyst [22]. The synthetic mate-
rials with similar structural and chemical properties as the natural
pumice were characterized by larger surface area. To this respect
it was confirmed that the procedure of the metal deposition did
not have much relevance when dealing with the large surface area
supports, differently from the low surface area support like pumice,
for which only the allyl procedure allowed to obtain very small
supported metal particles [22].
The electronic effects related to the use of pumice as support
for nickel were investigated in the hydrogenation of CO as function
of different catalyst calcination temperature [37,38]. A combined
TPR, XRD and XPS characterization ascertained that the presence
of alkali ions in the support structure and their temperature driven
surface segregation inhibited both the activity and the C2+ selectiv-
ity, probably through a dilution ensemble effect.
3.1.2. Bimetallic catalysts
The combination of the electronic properties of pumice with
those of a bimetallic system was investigated in a series of pumice
supported palladium catalysts promoted by copper and tested in
the liquid phase hydrogenation of phenylacetylene and in the gas
phase isomerization-hydrogenation of but-1-ene [43,44]. Addition
of small amount of a second metal may improve the catalytic behav-
ior of palladium catalysts in the hydrogenation of diene to alkene
on the bases of geometric or ligand effects [20]. The geometric
effect consists of a modification of the “ensemble size” intended
as the number of surface atoms needed for the organic substrate
activation. The “ligand effect” is an electronic effect associated
with a change in the electronic properties of a central metal atom
brought about by varying its near neighbor atoms. For the Pd-Cu
system, the importance of the chemical state of copper in control-
ling the catalytic behavior of palladium in two different reactions
was demonstrated. After a mild catalyst reduction at room tem-
perature, copper, which in accord with the Cu 2p3/2 XP spectra
shown in Fig. 2 was present as CuO, caused a depletion of the
electron density at the palladium atoms [43]. Assuming a pref-
erential localization of the CuO on the support, with consequent
invariance of the palladium active site distribution, the electronic
density change, in accord with the “structure insensitivity” of the
reaction, did not affect the conversion rate, whereas it increased
the selectivity to styrene as shown in Fig. 3. The preferential forma-
tion of the monoalkene was ascribable to a structure sensitivity of
the subsequent hydrogenation of styrene to ethylbenzene inhibited
by the electron deficiency of the palladium particles. After reduc-
tion at 623 K, the catalytic active sites changed with formation of
metallic copper and Cu-Pd alloyed particles. Due to the ensemble
size variation with dilution of the palladium sites by copper, the
conversion rate decreased. Different behavior was observed in the
hydrogenation reaction of the but-1-ene. In this case the presence
of CuO, by decreasing the electron density over palladium, caused
a decrease of the hydrogenation rate and an increase of the isomer-
ization selectivity to but-2-ene reflecting the structure sensitivity
of the reaction [43]. In contrast, after the catalyst reduction, in the
presence of metallic copper, the re-establishment of the electronic
density restored the typical hydrogenation rate of the monometal-
lic palladium catalyst. However, the concurrent decrease of the
ensemble size, needed for multiply adsorbed alkylidene species,
favored the selectivity towards isomerization requiring, on the con-
trary, only one metal centre according to the intramolecular shift
mechanism [44,45].
3.2. Hydrotreating
3.2.1. CoMo catalysts
There is an increasing concern about the environmental impact
of fossil fuels with respect to their sulfur content and consequent
emission of SO2 in the atmosphere. This worry, along with an
increased exploitation of heavier feedstocks such as shale oil and
sand tar, thrusted in the last 15–20 year a strong scientific and
industrial interest toward development of more active and sta-
ble hydrodesulfurization (HDS), catalysts [46]. The most commonly
used HDS catalysts consisted of ␥- Al2 O3 supported MoS2 promoted
by cobalt. The ascertained structure of such catalysts was a layered
structure of Co-Mo-S formed by MoS2 with Co atoms located at the
MoS2 edges. Single slab or multiple slabs with different degree of
interaction with the support would influence the catalytic activity
[47]. Generally the stronger was the interaction with the support,
 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 2. Cu 2p3/2 photoelectron peak of the 0.05Cu–1.5Pd/pumice catalyst after the
sequential treatment. (a) H2 at 298 K; (b) H2 at 623 K; (c) air treatment at 623 K; (d)
H2 at 623 K. From Venezia et al. [43] with permission.
as in the single slab model (CoMoS type I), the less active was the
catalyst, opposite to the case of the multiple slabs (CoMoS type
II). The occurrence of one or the either slab models depended on
the carrier. Series of CoMo catalysts, with and without sodium ions
added as promoters, were supported on SiO2 and on amorphous
117
Fig. 3. Styrene selectivity versus phenylacetylene conversion of the copper-
promoted 1.5 Pd/pumice-supported samples prepared by organometallic route and
reduced at 298 K. The reaction wasperformed in liquid phase in Tetrahydrofuran
(THF) at 298 K under atmospheric pressure of H2 . From Venezia et al. [43] with
permission.
Fig. 4. Variation of the thiophene HDS rate constant, k, versus wt% Na for the ASA
and the SiO2 -supported catalysts. Samples prepared from both cobalt precursors,
CoCl2 and Co(NO3 )2 are included (reaction conditions: 5.2vol% thiophene in H2 at
25.7 ml min−1 , T = 603 K). From Venezia et al. [48] with permission.
alumino-silicate (ASA) by classical impregnation method starting
from (NH4 )6 Mo7 O24 ·4H2 O and Co(NO3 )2 ·6H2 O, also in the pres-
ence of the complexing agent, nitrilotriacetic acid (NTA) [32]. The
catalysts were then tested in the HDS of thiophene [48–51]. As
seen in Fig. 4, the HDS rate constant as a function of the wt% Na,
for the ASA supported catalysts, regardless the cobalt precursor,
followed a volcano type curve with a maximum of activity in corre-
spondence of 3 wt% of sodium in the support [48]. On the contrary
on the SiO2 support, a drastic decrease of the rate constant with
the increase of the sodium content was observed. By relating the
catalytic results with the structural characterization, the different
behavior was attributed to the acidic nature of the ASA support as
compared to the SiO2 oxide. Indeed, Na+ , by suppressing the ASA
support acidity would favor the Co-Mo interaction as compared
to separate support- Mo(Co) interaction, leading to the ␤-CoMoO4
phase. The formation of such phase during the calcination step was
beneficial since from a structural point of view it was considered
the precursor to the active CoMoS type II phase after the sulfidation
treatment [48]. However above a certain sodium content mixed
sodium molybdate species would preferentially form with conse-
quent decrease of the activity. On the contrary, a negative effect of
sodium was observed on the silica supported catalysts where the
formation of the ␤-CoMoO4 precursor phase did not need the pres-
ence of sodium. On the contrary, sodium had the deleterious effect
of replacing the active precursor ␤-CoMoO4 phase with sodium
118 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 5. Variation of the thiophene HDS rate constant with the support formulation
sx for the sodium free and 1 wt% Na doped series (reaction conditions: 5.2 vol% thio-
phene in H2 at 25.7 ml min−1 , T = 603 K). From La Parola et al. [51] with permission.
molybdate phases, producing also a drastic decrease of the surface
area.
On the bases of the above results, a combination of the two
effects, composition of the support in terms of Al/Si atomic ratio
and presence of sodium, was investigated. To this aim, a series
of supports labeled as s with different Al/Si atomic ratio ranging
from 0 to ∝, was prepared by sol-gel-technique and characterized
by FT-IR and XPS spectroscopies. The same supports were impreg-
nated with a constant amount of sodium [51]. The FT-IR of adsorbed
pyridine allowed to identify Lewis acid sites of similar strength
independently on the Al/Si atomic ratio, and Brønsted acid sites
which formed essentially in the samples with a higher percent-
age of silica [50]. The good molecular mixing between the two
oxides was confirmed by the gradual variation of the structure of
the valence spectra [50]. As seen in Fig. 5 a maximum of the thio-
phene HDS activity was obtained in correspondence of the support
s2 (Al/Si ∼ 2), thereafter the activity decreased. The support depen-
dence was clearly modulated by the presence of 1 wt% Na. The right
balance between overall Brønsted/Lewis acidity and the catalyst
reducibility as determined by TPR, shown in Fig. 6, were the main
reasons for the catalytic behavior. Sodium, hardly influencing the
reducibility of the catalysts, by neutralizing the Brønsted acidity,
responsible for weaker metal-support interaction, and driving the
metals towards the more interacting Lewis site, depending on the
Al/Si ratio, would determine the type of CoMoO4 precursor and
therefore the type of CoMoS active phase [51]. The studies demon-
strated the possibility of obtaining good performing CoMo HDS
catalysts by choosing the appropriate support acidity and playing
with the amount of sodium.
3.2.2. Noble metal catalysts
In spite of improvement in their chemical and structural design,
CoMo catalysts were not suitable in the desulfurization of diesel
fraction arising from heavy crudes with higher sulfur and aromatic
contents. In order to overcome the thermodynamic constraints of
the high temperature saturation of the polyaromatic molecules and
to efficiently activate the C-S bond of the heavy aromatic molecules,
noble metals such as Pd or Pt were suggested as low temperature
hydrotreatment catalysts [52]. By the same concepts developed for
hydrogenation reactions and hydrodechlorination, the activity of
the palladium catalyst supported over amorphous aluminosilicate
was modified by gold added as a structural/chemical promoter [53].
Good mixing of palladium and gold was achieved in a series of
Aux Pdy catalyst supported on ASA by the slow deposition- precip-
itation method with urea. As seen in Fig. 7 the thiophene HDS rate
constant, normalized by the palladium content, increased going
from pure palladium to the bimetallic Au50 Pd50 , reaching almost a
plateau, and the pure gold catalyst being completely inert. Accord-
ing to the analyses of the X-ray diffraction patterns, the high gold
loaded samples contained true alloyed particles with an estimated
composition of Au80 Pd20 . Such gold enriched composition was
ascribed to the method of synthesis where preferential nucleation
of the gold hydroxides particles on the support hydroxyls took
place. The beneficial effects of gold were of geometrical and struc-
tural nature; the geometrical effect was related to the decrease of
the active Pd ensemble by simply diluting the palladium atoms
in the Aux Pdy solid solution, in agreement with a structure sen-
sitivity of the HDS reaction. The structural effect, consisted in the
capture of palladium within the alloy lattice, avoiding, as con-
firmed by XRD analyses, formation of the less active Pd4 S species.
Similar type of bimetallic catalysts but with nanosized particles
were prepared by an alcohol reduction method, adding during the
synthesis poly(N-vinyl-2-pyrrolidone) (PVP) as protective agent
against metal sintering [54]. The obtained Aux Pdy catalysts were
tested in the hydrodesulfurization of thiophene and benzothio-
phene (DBT). The study, carried out in collaboration with the group
of prof. Josè Luis Fierro in Madrid has confirmed the beneficial
effect of the gold-palladium alloys in the hydrodesulfurization of
both substrate molecules at different experimental conditions. The
alloy, in accord with XRD and IR spectra of chemisorbed CO, was
proved to be more resistant to sulfur poisoning and to optimize
the ensemble size for the thiophene activation. Surprisingly, in the
case of the DBT desulfurization performed at high hydrogen pres-
sure, (P = 30 bar), the conversion obtained with monometallic gold
catalyst was superior to the conversion over the palladium catalyst
and substantially increased in the bimetallic catalysts. Depending
on the Au/Pd atomic ratios, different selectivity to biphenyl (BP) and
to cyclohexylbenzene (CHB) were obtained, reflecting the incidence
of the two possible HDS pathways, the intermediate hydrogena-
tion and the direct sulfur extrusion. Bimetallic AuPt and AuPd and
gold promoted PdPt catalysts were investigated in the hydrogena-
tion of aromatics also in the presence of sulfur [55–58]. The lack of
any sulfur compound on the spent catalysts confirmed the positive
effect of gold against sulfur poisoning [55]. Aiming to control the
composition, structure and sizes of supported bimetallic particles,
an innovative synthetic procedure, involving mercaptopropyltri-
ethoxysilane (MPTES) as stabilizing agent and as carrier for the
particle deposition over silica, was adopted [33,59–61]. The thiol
functionalization procedure was used for preparing PdAu and PtAu
bimetallic catalysts containing 1 wt% of each metal. In the first case,
it was possible to demonstrate how a fine tuning of the molar
ratio between the −SH groups and the total amount of the two
metals allowed to modify the structure of the bimetallic PdAu cat-
alyst. By using the deposition-precipitation method, over the −SH
functionalized supports, the interaction of each metals with the
hydroxyl and with the sylanol groups would indeed determine the
final bimetallic particle composition with a remarkable effect on
the thiophene HDS activity, as shown in Fig. 8 [60]. For comparison
reason catalysts prepared by wet impregnation are also included. In
the case of the PtAu catalyst, the synthesis procedure involved aging
of the metal precursors solution with the MPTES reagent [61]. As
confirmed by XRD and FTIR analyses combined with the XPS spec-
tra given in Fig. 9, a structural change from the less active Pt@AuPt
core-shell like structure to the more active Au@AuPt core-shell like
structure took place by simply extending the time the two metal
ions spent in solution with the MPTES ligand before adding the
support [61].
Mesoporous silica MCM-41, due to its sigh surface area, well
defined pore structure and good thermal stability was used as sup-
port for Pd catalysts in HDS of thiophene [62]. To increase the
support acidity without affecting the thermal stability and the
 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124 119

Fig. 6. TPR profiles of catalysts without sodium on different supports. From La Parola et al. [51] with permission.

structural order of the hexagonal framework, post synthesis alu-
mination of the mesoporous silica was performed by simple Al3+
ions impregnation. As shown in Fig. 10, very small amount of Al3+
ions produced a large increase of the conversion. Above 0.5 wt%
of doping ions lower initial conversions and quicker deactivation
were obtained. A bi-functional mechanism as sketched in Fig. 11
was invoked. The acid sites, created on the supports by post alu-
mination, interacted with the electron rich thiophene molecules.
Being close to the activated hydrogen over Pd particles, hydrogena-
tion by the spillover hydrogen, followed by the C S bond breaking
would occur. However the efficiency of this process is nullified
when, above a certain amount of acid sites, formation of coke would
prevail.
3.3. Oxidation
3.3.1. CO oxidation over Au catalysts
The remarkable results published by Haruta more than 25 years
ago [63] has driven a large academic and industrial interest in the
development of gold catalysts for low temperature oxidation reac-
tion, particularly relevant to environmental catalysis [64–70]. In
particular, the effect of the supports and of different preparation
120 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124

20

16
k (ml*gPd s )
-1 -1

12

0
0 20 40 60 80 100
wt% Au
Fig. 7. Variation in the thiophene HDS rate constant, obtained at 613 K, versus
weight percent of Au for the ASA supported Aux Pdy catalysts (reaction conditions: Fig. 10. Thiophene conversion curve as a function of time for three selected samples
5.2 vol% thiophene in H2 at 25.7 ml min−1 ; T = 613 K). From Venezia et al. [53] with (reaction conditions: 5.3 vol% thiophene in H2 ; WHSV = 7500 h−1 ; T = 613 K). From
permission. Venezia et al. [62] with permission.

procedures was extensively investigated by our group, using CO

Fig. 8. Thiophene HDS reaction rate as a function of molar ratio between SH and
metal, molSH /molmetal . The open symbols refer to catalysts prepared by wet impreg-
nation (reaction conditions: 5.2 vol% thiophene in H2 at 25.7 ml min−1 ; T = 613 K).
From La Parola et al. [60] with permission.
oxidation as a probe reaction [30,71–75]. Some results from the
first published study in this topic will be here mentioned. The study
regarded the effect of the preparation procedure of ceria supported
gold catalysts on the final catalyst structure and on its oxidation
activity [30]. Au/CeO2 catalysts with the same gold loading were
prepared by adopting three different synthetic procedures. One cat-
alyst was prepared by deposition-precipitation (DP) using HAuCl4
as gold precursor and Na2 CO3 as precipitating agent, then after
filtering and washing, it was calcined at 393 K. A second cata-
lyst was prepared by the solvated metal atom deposition (SMAD)
method [31]. Accordingly, metal atoms were simultaneously vapor-
ized under vacuum and rapidly deposited on the frozen solvent of
the walls, followed by warming up and condensing down to the
bottom of the reactor over the support [30]. A third catalyst was
prepared by co-precipitation (CP) of gold and cerium hydroxides
with Na2 CO3 at the constant pH = 10, followed by final calcina-
tion temperature of 673 K needed to remove the nitrate ions of
the precursor reagents. As shown in Fig. 12 the best performing
catalyst was the one prepared by DP followed by the SMAD sam-
ple. As proved by the slight but still significant contraction of the

Fig. 9. Pt 4f and Au 4f XPS high resolution region of bimetallic PtAu/SiO2 .

 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 11. Suggested bi-functional mechanism of HDS of thiophene over Pd catalyst
supported on siliceous MCM-41 modified by aluminum ions. From Venezia et al.
[62] with permission.
Fig. 12. CO conversion% as a function of temperature for the different catalysts (reac-
tion conditions: 1%CO +1% O2 in He; WHSV = 60 000 ml g−1 h−1 ). From Venezia et al.
[30] with permission.
fluorite lattice parameter estimated by Rietveld refinement of the
corresponding XRD pattern, the higher activity of the DP sample
was attributed to the presence of a mixed oxide Aux Ce1-x O2-␦ with
ionic gold. In the case of the CP and SMAD Au/CeO2 samples, the
significant catalytic activity in the 400–600 K temperature range
was related to the presence of metallic gold, following an inverse
relationship with the Au crystallite sizes. By comparing the activity
of different gold catalysts supported on different carrier, such as the
reducible CeO2 , TiO2, MnO2 and inert oxides, like SiO2, a major dis-
tinction could be ascertained [71,72]. Indeed, the presence of non-
metallic Au species with an increased oxygen vacancy mobility was
the main requisite for the achievement of the highest CO conversion
rate at the lowest temperature in the reducible oxides. On the other
121
Fig. 13. Methane conversion as a function of reaction temperature (reaction condi-
tions: 0.3 vol% CH4 + 2.4 vol%O2 in He; WHSV = 60 000 mlg−1 h−1 ): curve Ia (1st run
without SO2 ); curve Ib (1st run with SO2 ); curve II (2nd run without SO2 after curve
Ib). From Venezia et al. [77] with permission.
hand, the presence of small metallic gold particles were responsi-
ble for the high activity exhibited by inert carrier supported gold
catalysts.
3.3.2. CH4 total oxidation over Pd catalysts
The research in improved catalytic systems for the total com-
bustion of methane, has been lately receiving a strong scientific
and industrial interest due to a steady increase of the number of
natural gas fueled engines (NGVs). Indeed, in spite of the lower
amount of NOx, particulates and CO2 emissions, associated with
the NGVs vehicles, the unconverted methane, a stronger green-
house effect gas than CO2 , is still a major environmental issue to be
solved [76]. We have focused our research on the development of
palladium based catalyst, which are the most efficient systems for
this type of reaction but suffer from drawbacks related to sintering
of the active phase and poisoning by water and sulfur-containing
compounds. In particular the effect of the support morphology
and of a second metal promotion was investigated [77–80]. The
research has produced important results related to the morphology
effect of the silica support on the total methane conversion without
and in the presence of SO2 [77]. As shown in Fig. 13 represent-
ing the methane conversion curves versus reaction temperature,
a remarkable improvement of the activity was registered for the
palladium catalysts supported on a mesoporous silica as compared
to an amorphous silica. Moreover the decreasing activity upon SO2
exposure was easily recovered and even improved in a following
run with SO2 - free reagent mixture. On the basis of XRD analyses,
the improved tolerance to the SO2 was attributed to a bimodal dis-
tribution of palladium particles with larger particles outside the
support pores and smaller particles inside the pores. Palladium
sintering, generally reported for palladium catalysts supported on
amorphous silica, did not occur in the present case because of the
mesoporous nature of the silica inhibiting the particle growth at
high temperature [77,78]. In this system the presence of gold did
not change substantially the behavior, except for a slight decrease
of activity likely due to a ligand effect reflected in an increase of
activation energy. However, by detailed XRD and XPS analyses,
the presence of gold allowed to ascertain a dynamic reduction-
oxidation process occurring at the palladium surface. As shown in
Fig. 14, the reducing action of SO2 over PdO, just speculated to occur
on the monometallic catalyst, was in fact confirmed by the reduced
palladium components of the Pd 3d XP spectrum, stabilized by gold.
The combination of a sulfating and non-sulfating support on the
Pd catalytic conversion of methane was also explored [78–80]. The
122 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 14. Pd 3d XP spectra of PdAu/HMS after the 3rd run (including the overnight
SO2 treatment). From Venezia et al. [77] with permission.
modification of SiO2 , induced by small amounts of TiO2 (5–10 wt%),
influenced positively the methane oxidation performance and the
SO2 tolerance of the supported PdO. The increased methane con-
version was attributed to the formation of Si-O-Ti linkages favoring
the intrinsic activity of the supported PdO. The tolerance to the
SO2 and water was attributed to the preferential adsorption of the
molecules to the oxygen vacancy sites of TiO2 , avoiding Pd poison-
ing. For the first time, the catalyst recovered the initial activity after
one cycle under lean–burn conditions without additional regener-
ating treatments [79,80].
3.3.3. CH4 partial oxidation and dry reforming for syngas
production
Methane, besides being used as vehicle fuel and for power gen-
eration, allows to obtain fuels and chemicals as alternative to the
petrol-chemistry derivatives. The direct synthesis of such products
from methane is, however, not practical because of the low yield.
It is therefore preferred to generate from methane the syn gas, a
mixture of H2 and CO, which through the Fisher Tropsch synthesis
would lead to various products. The main processes for syn gas pro-
duction are the reforming and to a less extent the catalytic partial
oxidation [81,82].
We have recently started working in the field of dry reform-
ing of methane (DRM) with CO2 [83,84]. Nickel is generally the
metal of choice for this type of process in virtue of its low cost,
availability and relatively high activity. In our studies we had tried
to overcome the two main drawbacks associated with the nickel
catalysts which are the easy metal sintering and the coke forma-
tion leading to a quick deactivation. The addition of small amount
of gold was explored in order to increase the nickel dispersion
and to maintain it for long time by inhibiting metal sintering. Ni
catalysts with different gold loading supported over commercial
spinel MgAl2 O4 were prepared by an unconventional sol method
developed in prof. Guczi’s laboratory [83]. Several characterization
techniques, TPO/TPR, TEM, XPS, HRTEM, before and after the DRM
test, confirmed highly dispersed and quite stable alloyed particles
as detected by XRD patterns and by the TEM images of Fig. 15. The
study proved a positive effect of gold on the DRM activity. More-
over, the dilution of the nickel sites by gold avoided the formation
of large carbon molecules generally requiring a large ensemble
of Ni sites. The combined effect of Mg and Ce doping of an alu-
mina support and the Ni promotion by small amount of Au and
Fig. 15. HRTEM image of Ni–3Au catalyst after a second TPR. From Horvath et. al.
[83] with permission.
Pt were recently investigated [84]. The best performance in terms
of methane conversion and catalyst stability was exhibited by the
CeO2 doped Ni/Al2 O3 catalyst and by the trimetallic NiAuPt sup-
ported on the undoped Al2 O3 . In the first case, modification of
alumina by ceria contributed to the increased catalyst reducibility
and to the easy carbon removal due to the peculiar redox prop-
erties of the CeO2 oxide in accord with H2 -TPR analyses. In the
second case the simultaneous presence of nickel and two noble
metals contributed to the enhancement of activity and to the resis-
tance towards carbon formation because of particle size growth
constraint during reaction.
The second important reaction to generate syn gas is the catalytic
partial oxidation (CPO) of methane. As in the dry reforming process
nickel is still the metal of choice for possible industrial application.
Contrary to the endothermic reforming reactions, highly energetic
and capital intensive, the CPO is mildly exothermic and is a bet-
ter cost and energy efficient process. As discussed before for the
dry reforming process, the main issues related to the nickel cata-
lyst also for this type of reaction was the deactivation caused by
nickel sintering and/or by coke formation. Series of Ni catalysts
were prepared using different Ni introduction method and differ-
ent support oxides. In particular the research was focused on CeO2 ,
La2 O3 and a combination of the two [34,85]. The selection of the two
oxides was dictated by their redox properties and their tendency
to chemically interact with the supported metals. Both properties
were indeed important to enhance the carbon removal and to main-
tain a stable metal dispersion [85]. Structural analyses indicated
different degree of interaction between nickel and support, modu-
lated by the preparation procedure. Same catalyst formulation had
completely different TPR patterns, as a function of the preparation
procedure, and produced quite different amount of deposited car-
bon. Very interesting results were obtained with series of Ni/CeO2
 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
Fig. 16. TPR profiles of the differently prepared nickel catalysts and corresponding
supports. From Pantaleo et al. [34] with permission.
Fig. 17. CH4 conversion after different time on stream for the differently prepared
nickel catalysts. From Pantaleo et al. [34] with permission.
catalysts prepared in different ways leading to different degree
of Ni-CeO2 interaction and different Ni and CeO2 particle sizes.
As seen in Fig. 16, the TPR profiles of the three catalysts were
very different. The sample labeled as Ni/CeO2 (HT), prepared in the
presence of cetyl trimetyl ammonium bromide (CTAB) as protect-
ing ligand, exhibited a TPR profile typical of separated NiO and
CeO2 . Quite different were the profiles for the catalysts prepared
by the microwave assisted wet impregnation and co-precipitation.
The deep interaction between NiO and CeO2 was reflected in the
peak maxima and in the increased hydrogen consumption [34].
As shown in Fig. 17, the differences in terms of methane con-
version were not so striking, except for the sample prepared by
microwave assisted wet-impregnation Ni/CeO2 (mw-WI). However
as confirmed by TGA profiles of the catalysts after the reaction, the
production of carbon was very different. In fact, the most active
catalyst, Ni/CeO2 (HT), produced more carbon. In conclusion it was
shown that the achievement of small NiO crystallite sizes not neces-
sarily interacting with the support were beneficial for the catalytic
activity. On the other hand, a stronger interaction between nickel
and CeO2 , by increasing the oxygen mobility of the cerium oxide
favored the carbon removal. In order to design an appropriate cata-
lyst for industrial CPO application, with high activity, high stability
123
and without carbon buildup, the right balance between the two
different factors should be taken into account.
4. Conclusion
As it has been partially confirmed in this compendium of our
research activity, catalysis is a chemistry technology which needs
a multidisciplinary approach. Generally, the multiplicity of active
sites present in a catalyst makes very difficult to relate the catalytic
behavior with the catalyst structure. Along with properly designed
catalyst syntheses, of paramount importance are also the mate-
rial characterization tools allowing to ascertain the fundamental
aspects behind the catalytic behavior of the catalysts. By its own
mission, catalysis should be seen as a continuing evolving technol-
ogy the entire society must rely on, to solve the many issues of
modern society related to environment preservation and energy
supply.
Acknowledgment
Financial support by CNR (Progetto Finalizzato “Chimica Fine
II”, MURST (40% and 60%) European Community for COST Action
D15 and D36 and CM0903, NATO Grant ESP.CLG. No. 984160 are
kindly acknowledged. Special thanks go to Prof. Giulio Deganello
who intitiated us to the catalysis world.
References
[1] G.A. Somorjai, Introduction to Surface Chemistry and Catalysis, John Wiley &
Sons, Inc, NY, USA, 1994.
[2] J.R.H. Ross, Heterogeneous Catalysis, Fundamental and Applications, Elsevier
B.V, Amsterdam, The Netherlands, 2012.
[3] M.S. Chen, D.W. Goodman, Catal. Today 111 (2006) 22–33.
[4] A.M. Venezia, Catal. Today 77 (2007) 359–370.
[5] A.M. Venezia, R.G. Cavell, J. Chem. Phys. 84 (1986) 4797–4806.
[6] A.M. Venezia, J. Vayrynen, R.G. Cavell, J. Chem Phys. 87 (1987) 5195–5201.
[7] A.M. Venezia, C.M. Loxton, Surf. Sci. 194 (1988) 136–148.
[8] A.M. Venezia, C.M. Loxton, J.A. Horton, Surf. Sci. 225 (1990) 195–205.
[9] R.T. Haasch, A.M. Venezia, C.M. Loxton, J. Mater. Res. 7 (1992) 1341–1349.
[10] A.M. Venezia, C.M. Loxton, R.F. Garrett, Surf. Sci. 275 (1992) 75–81.
[11] B.C. Gates, Chem. Rev 95 (1995) 511–522.
[12] M. Che, C.O. Bennett, Adv. Catal. 36 (1989) 55–172.
[13] A.M. Venezia, A. Rossi, D. Duca, A. Martorana, G. Deganello, Appl. Catal. A 125
(1995) 113–128.
[14] G.C. Bond, Chem. Soc. Rev. 20 (1991) 441.
[15] A.A. Herzing, C.J. Kiely, A.F. Carley, P. Landon, G. Hutchings, J. Sci. 321 (2008)
1331–1332.
[16] A. Corma, P. Concepción, M. Boronat, M.J. Sabater, J. Navas, M.J. Yacaman, E.
Larios, A. Posadas, M.A. López-Quintela, D. Buceta, E. Mendoza, G. Guilera, A.
Mayoral, Nat. Chem. 5 (2013) 775–781.
[17] M.M. Schubert, S. Hackenberg, A.C. van Veen, M. Muller, V. Plzak, J. Behn, J.
Catal. 197 (2001) 113–122.
[18] A.M. Venezia, L.F. Liotta, G. Pantaleo, V. la Parola, G. Deganello, A. Beck, Zs.
Koppany, K. Frey, D. Horvath, L. Guczi, Appl. Catal. 251 (2003) 359–368.
[19] L. Ilieva, T. Tabakova, G. Pantaleo, I. Ivanov, R. Zanella, D. Paneva, N. Velinov,
J.W. Sobczak, W. Lisowski, G. Avdeev, A.M. Venezia, Appl. Catal. A 467 (2013)
76–90.
[20] R. Burch, Acc. Chem. Res. 15 (1982) 24–31.
[21] A.C. Gluhoi, B.E. Nieuwenhuys, Catal. Today 122 (2007) 226–232.
[22] L.F. Liotta, A.M. Venezia, A. Martorana, A. Rossi, G. Deganello, J. Catal. 171
(1997) 169–176.
[23] L.F. Liotta, A.M. Venezia, A. Martorana, G. Deganello, J. Catal. 171 (1997)
177–183.
[24] M.L. Testa, V. la Parola, A.M. Venezia, Catal. Today 158 (2010), 198-113.
[25] M.L. Testa, V. La Parola, L.F. Liotta, A.M. Venezia, J. Mol. Cat. A 367 (2013)
69–76.
[26] M.L. Testa, V. la Parola, A.M. Venezia, Catal. Today 223 (2014) 115–121.
[27] M.L. Testa, L.C. Demailly, V. La Parola, A.M. Venezia, C. Pinel, Catal. Today 257
(2015) 291–296.
[28] Y.I. Yermakov, Catal. Rev. Sci. Eng. 13 (1976) 77.
[29] G. Deganello, D. Duca, L.F. Liotta, A. Martorana, A.M. Venezia, La Chimica e
l’Industria 80 (1998) 741–746.
[30] A.M. Venezia, G. Pantaleo, A. Longo, G. Di Carlo, M.P. Casaletto, F.L. Liotta, G.
Deganello, J. Phys. Chem. B 109 (2005) 2821–2827.
[31] K. Klabunde, Acc. Chem. Res. 8 (1975) 399.
[32] A.M. Venezia, V. La Parola, G. Deganello, D. Cauzzi, G. Leonardi, G. Predieri,
Appl. Catal. A 229 (2002) 261–271.
124 A.M. Venezia et al. / Catalysis Today 285 (2017) 114–124
[33] V. La Parola, A. Longo, A.M. Venezia, A. Spinella, E. Caponetti, Eur. J. Inorg.
Chem. (2010) 3628–3635.
[34] G. Pantaleo, V. La Parola, F. Deganello, R.K. Singha, R. Bal, A.M. Venezia, Appl.
Catal. B 189 (2016) 233–241.
[35] M.A. Floriano, A.M. Venezia, G. Deganello, E.C. Svensson, J.H. Root, J. Appl.
Cryst. 27 (1994) 271–277.
[36] A.M. Venezia, M.A. Floriano, G. Deganello, A. Rossi, Surf. Interface Anal. 18
(1992) 532–538.
[37] A.M. Venezia, R. Bertoncello, G. Deganello, Surf. Interface Anal. 23 (1995)
239–247.
[38] A.M. Venezia, A. Parmaliana, A. Mezzapica, G. Deganello, J. Catal. 172 (1997)
463–470.
[39] A.M. Venezia, D. Duca, M.A. Floriano, G. Deganello, A. Rossi, Surf. Interface
Anal. 18 (1992) 619–622.
[40] A.M. Venezia, D. Duca, M.A. Floriano, G. Deganello, A. Rossi, Surf. Interface
Anal. 19 (1992) 543–547.
[41] G. Moretti, J. Electron Spectrosc. Relat. Phenom. 95 (1998) 95–144.
[42] A.M. Venezia, A. Rossi, L.F. Liotta, A. Martorana, G. Deganello, Appl. Catal. A
147 (1996) 81–94.
[43] A.M. Venezia, L.F. Liotta, G. Deganello, Z. Schay, L. Guczi, J. Catal. 182 (1999)
449–455.
[44] L. Guczi, Z. Schay, Gy. Stefler, L.F. Liotta, G. Deganello, A.M. Venezia, J. Catal.
182 (1999) 456–462.
[45] J. Goetz, M.A. Volpe, R. Touroude, J. Catal. 164 (1996) 369–377.
[46] H. Topsoe, B.S. Clausen, F.E. Massoth, Hydrotreating Catalysis, in: J.R.
Anderson, M. Boudart (Eds.), Springer, Berlin, 1996.
[47] R.G. Leliveld, A.J. van Dillen, J.W. Geus, D.C. Koningsberger, J. Catal. 175 (1998)
108–116.
[48] A.M. Venezia, F. Raimondi, V. la Parola, G. Deganello, J. Catal. 194 (2000)
393–400.
[49] V. La Parola, G. Deganello, C.R. Tewell, A.M. Venezia, Appl. Catal. A 235 (2002)
171–180.
[50] V. La Parola, G. Deganello, S. Scirè, A.M. Venezia, J. Solid. State Chem. 174
(2003) 188–482.
[51] V. La Parola, G. Deganello, A.M. Venezia, Appl. Catal. A 260 (2004) 237–247.
[52] B.C. Cooper, B.B.L. Donnis, Appl. Catal. A 137 (1996) 203–223.
[53] A.M. Venezia, V. La Parola, V. Nicolì, G. Deganello, J. Catal. 212 (2002) 56–62.
[54] A.M. Venezia, V. La Parola, G. Deganello, B. Pawelec, J.L.G. Fierro, J. Catal. 215
(2003) 317–325.
[55] A.M. Venezia, V. La Parola, B. Pawelec, J.G. Fierro, Appl. Catal. A 264 (2004)
43–51.
[56] B. Pawelec, A.M. Venezia, V. La Parola, E. Cano-Serrano, J.M. Campos-Martin,
J.L.G. Fierro, Appl. Surf. Sci. 242 (2005) 380–391.
[57] B. Pawelec, A.M. Venezia, V. La Parola, S. Thomas, J.L.G. Fierro, Appl. Catal. A
283 (2005) 165–175.
[58] B. Pawelec, V. La Parola, S. Thomas, J.L.G. Fierro, J. Mol. Catal. A: Chem. 253
(2006) 30–43.
[59] V. La Parola, B. Dragoi, A. Umgureanu, E. Dumitriu, A.M. Venezia, Appl. Catal. A
386 (2010) 43–50.
[60] V. La Parola, M.L. Testa, A.M. Venezia, Appl. Catal. B 119–120 (2012) 248–255.
[61] V. La Parola, M. Kantcheva, M. Milanova, A.M. Venezia, J. Catal. 298 (2013)
170–173.
[62] A.M. Venezia, R. Murania, V. La Parola, B. Pawelec, J.L.G. Fierro, Appl. Catal. A
383 (2010) 211–216.
[63] M. Haruta, N. Yamada, T. Kobayashi, S. Lijima, J. Catal. 115 (1989) 301–309.
[64] Q. Fu, A. Weber, M. Flytzani-Stephanopoulos, Catal. Lett. 77 (2001) 87–95.
[65] S. Scirè, S. Minico, C. Crisafulli, C. Satriano, A. Pistone, Appl. Catal. B 40 (2003)
43–49.
[66] D. Cameron, R. Holliday, D. Thompson, J. Power Sources 118 (2003) 298–303.
[67] L. Ilieva, G. Pantaleo, I. Ivanov, A.M. Venezia, D. Andreeva, Appl. Catal. B:
Environ. 76 (2007) 107–114.
[68] M. Kantcheva, O. Samarskaya, L. Ilieva, G. Pantaleo, A.M. Venezia, D. Andreeva,
Appl. Cata. B 88 (2009) 113–126.
[69] L. Ilieva, P. Petrov Ivanov, G. Munteanu, M. Boutonnet, J.W. Sobczac, W.
Lisowski, Z. Kaszkur, P. Markov, A.M. Venezia, T. Tabakova, Mat. Chem. Phys.
157 (2015) 138–146.
[70] L. Ilieva, P. Petrova, G. Pantaleo, R. Zanella, L.F. Liotta, V. Georgiev, S.
Boghosian, Z. Kaszkur, J.W. Sobczak, W. Lisowski, A.M. Venezia, T. Tabakova,
Appl. Catal. B 188 (2016) 154–168.
[71] M.P. Casaletto, A. Longo, A. Martorana, A. Prestianni, A.M. Venezia, Surf.
Interface Anal. 38 (2006) 215–218.
[72] M.P. Casaletto, A. Longo, A.M. Venezia, A. Martorana, A. Prestianni, Appl. Catal.
A 302 (2006) 309–316.
[73] A. Longo, L.F. Liotta, G. Di Carlo, F. Giannici, A.M. Venezia, A. Martorana, Chem.
Mater. 22 (2010) 3952–3960.
[74] A. Longo, L.F. Liotta, G. Pantaleo, F. Giannici, A.M. Venezia, A. Martorana, J.
Phys. Chem. C 116 (2012) 2960–2966.
[75] L.F. Liotta, G. Pantaleo, F. Puleo, A.M. Venezia, Catal. Today 187 (2012) 10–19.
[76] J.K. Lambert, M.S. Kazi, R. Farrauto, J. Appl. Catal. B 14 (1997) 211–223.
[77] A.M. Venezia, R. Murania, G. Pantaleo, G. Deganello, J. Catal. 251 (2007)
94–102.
[78] A.M. Venezia, G. Di Carlo, L.F. Liotta, G. Pantaleo, M. Kantcheva, Appl. Catal. B
106 (2011) 529–539.
[79] A.M. Venezia, G. Di Carlo, G. Pantaleo, L.F. Liotta, G. Melaet, N. Kruse, Appl.
Catal. B 88 (2009) 430–437.
[80] G. Di Carlo, G. Melaet, N. Kruse, L.F. Liotta, G. Pantaleo, A.M. Venezia, Chem.
Commun. 46 (2010) 6317–6319.
[81] B.C. Enger, R. Lødeng, A. Holmen, Appl. Catal. A 346 (2008) 1–27.
[82] J.F. Munera, S. Irusta, L.M. Conaglia, E.A. Lombardo, D.V. Cesar, M. Schmal, J.
Catal. 245 (2007) 25–34.
[83] A. Horváth, L. Guczi, A. Kocsonya, G. Sáfrán, V. La Parola, L.F. Liotta, G.
Pantaleo, A.M. Venezia, Appl. Catal. A 468 (2013) 250–259.
[84] H. Wu, G. Pantaleo, V. La Parola, A.M. Venezia, X. Collard, C. Aprile, L.F. Liotta,
Appl. Catal. B 156–157 (2014) 350–361.
[85] G. Pantaleo, V. La Parola, F. Deganello, P. Calatozzo, R. Bal, A.M. Venezia, Appl.
Catal. B 164 (2015) 135–143.