Enhanced (photo)catalytic activity of Wells-Dawson (H6P2W18O62) in comparison to Keggin (H3PW12O40) heteropolyacids for 2-propanol dehydration in gas-solid regime

Article abstract of DOI:10.1016/j.apcata.2016.10.002

Catalytic and photocatalytic 2-propanol dehydration to propene at atmospheric pressure and a temperature range of 60–120?°C were carried out in gas-solid regime by using bare and supported Keggin H₃PW₁₂O₄₀ (PW₁₂) and Wells-Dawson H₆P₂W₁₈O₆₂ (P₂W₁₈) heteropolyacids (HPAs). Binary materials were prepared by impregnation of the HPAs on commercial SiO₂ and TiO₂. The Wells-Dawson was in any case more active than the Keggin heteropolyacid and the differences were enhanced when the supported samples were used. In particular, Wells-Dawson HPA supported on TiO₂ and under irradiation showed the highest activity. The HPA species played the key role both in the catalytic and photo-assisted reactions. The acidity of the cluster accounts for the catalytic role, whereas both the acidity and the redox properties of the HPA species were responsible for the increase of the reaction rate in the photo-assisted catalytic reaction. The estimated apparent activation energy resulted always lower for the photocatalytic process than for the catalytic one.

Full text of DOI:10.1016/j.apcata.2016.10.002

Applied Catalysis A: General 528 (2016) 113–122
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Enhanced (photo)catalytic activity of Wells-Dawson (H6P2W18O62) in
comparison to Keggin (H3PW12O40) heteropolyacids for 2-propanol
dehydration in gas-solid regime
Giuseppe Marcìa,∗, Elisa I. García-Lópeza, Francesca Rita Pomillaa,b, Leonarda F. Liottac,
Leonardo Palmisanoa
a “Schiavello-Grillone” Photocatalysis Group, Dipartimento di Energia, Ingegneria dell’informazione e modelli Matematici (DEIM), Università di Palermo,
Viale delle Scienze, 90128 Palermo, Italy
b “Dipartimento di Ingegneria per l’Ambiente e il Territorio e Ingegneria Chimica”, Università della Calabria via Pietro Bucci 87036 Arcavacata di Rende,
Cosenza, Italy
c Istituto per Lo Studio dei Materiali Nanostrutturati (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:
Received 14 May 2016
Received in revised form 9 September 2016
Accepted 3 October 2016
Available online 4 October 2016
Catalytic and photocatalytic 2-propanol dehydration to propene at atmospheric pressure and a tem-
perature range of 60–120 ◦C were carried out in gas-solid regime by using bare and supported Keggin
H3PW12O40 (PW12) and Wells-Dawson H6P2W18O62 (P2W18) heteropolyacids (HPAs). Binary materials
were prepared by impregnation of the HPAs on commercial SiO2 and TiO2. The Wells-Dawson was in any
case more active than the Keggin heteropolyacid and the differences were enhanced when the supported
samples were used. In particular, Wells-Dawson HPA supported on TiO2 and under irradiation showed
the highest activity. The HPA species played the key role both in the catalytic and photo-assisted reac-
tions. The acidity of the cluster accounts for the catalytic role, whereas both the acidity and the redox
properties of the HPA species were responsible for the increase of the reaction rate in the photo-assisted
catalytic reaction. The estimated apparent activation energy resulted always lower for the photocatalytic
process than for the catalytic one.
Keywords:
Keggin
Wells-Dawson
Heteropolyacid
Polyoxometalate
2-Propanol dehydration
Photocatalysis
© 2016 Elsevier B.V. All rights reserved.
]
3−, is composed of a globe-like cluster whose
1. Introduction
polyanion, [PW12O40
diameter is ca. 1.0 nm [5,6]. Wells-Dawson heteropolytungstate,
[P2W18O62
6−, may be considered derived from two Keggin units
Metal oxides can be subdivided into classical solid oxides and
nanosized transition metal oxygen clusters, also called polyox-
ometalates (POMs) [1]. The most explored POM materials are the
heteropolyacids (HPAs) which are strong inorganic acids with well-
defined molecular structures and it is convenient to classify them as
Keggin and Wells-Dawson HPAs, by taking into account the geom-
etry of the heteropolyanion. HPAs are widely used in homo and
heterogeneous catalysis [2] and they have been also studied as
photocatalysts both in homogeneous [3] and heterogeneous sys-
tems [4]. The use of HPAs as catalysts, in both homogeneous and
heterogeneous processes, has attracted great interest in organic
synthesis because of their stronger Brönsted acidity and redox
properties. These materials are also economically and environmen-
tally interesting because inexpensive and non-toxic [5]. Keggin
]
by removing a [W3O9] block from each of them and by link-
ing these two units by means of oxygen atoms forming a nearly
ellipsoidal anion cluster [7]. HPAs show hierarchical structures
that result of paramount importance to understand their role as
catalysts. The primary structure corresponds to the metal oxide
cluster heteropolyanion itself; the secondary structure to the three-
dimensional arrangement including counter cations and water
molecules. Misono et al. have demonstrated that a solid het-
eropoly compound can act as catalyst in a gas phase reaction in
three different ways, i.e. as (i) surface-type catalyst, (ii) pseudo-
liquid bulk-type and (iii) bulk-type catalyst [5,6]. The surface-type
catalysis is the ordinary heterogeneous catalysis that takes place bi-
dimensionally on the solid surface, in contrast with the bulk-type
catalysis. When the diffusion of reactant molecules in the lattice
rather than pores of the solid is faster than the reaction, a pseudo-
liquid phase is formed in the bulk where the catalytic reaction can
proceed. The catalytic behaviour of HPAs is strongly influenced by
∗
Corresponding author.
E-mail address: giuseppe.marci@unipa.it (G. Marcì).
http://dx.doi.org/10.1016/j.apcata.2016.10.002
0926-860X/© 2016 Elsevier B.V. All rights reserved.

114
G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
Fig. 1. XRD patterns of the photocatalysts: (A) PW12 bare or supported on SiO2 and TiO2. and (B) P2W18 bare or supported on SiO2 and TiO2. (* Rutile; +Anatase).
Fig. 2. SEM pictures of bare SiO2, bare P2W18 and P2W18 and PW12 supported on SiO2.

G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
115
914
1091 962
PW12
P2W18
982 890 810
780
PW12/TiO2 (a)
1080
P2W18/SiO2
PW12/SiO2
PW12/TiO2 (b)
P2W18/TiO2 (a)
1105
P2W18/TiO2
PW12/TiO2
P2W18/TiO2 (b)
600
400
(A)
(C)
(B)
1200
1000
800
600
400 1200
1000
800
600
400 1200
1000
800
Wavenumber [cm-1]
Wavenumber [cm-1]
Wavenumber [cm-1]
Fig. 3. FTIR spectra of (A) P2W18 bare and supported on SiO2 and TiO2; (B) PW12 bare and supported on SiO2 and TiO2 and (C) PW12 and P2W18 supported on TiO2 after the
catalytic (a) and photocatalytic experiments (b), carried out in both cases at 120 ◦C.
998
639
513
399
972
197
P2W18/TiO2
444
P2W18
920
1200
1100
1000
900
800
700
600
500
400
300
200
100
Raman Shift [cm-1]
Fig. 4. Raman spectra of the bare P2W18 and P2W18/TiO2.
the modifications that may occur in the secondary structure [8]
where the Brönsted acidity is located and which would be altered
by the number of water molecules involved. The structural modifi-
cations that may occur both in the secondary and tertiary structures
of the HPAs [2,6–8] influence their hydration state determining the
number and strength of the surface acid sites along with the acces-
sibility of the molecules. The acid sites in the secondary structure
described for the catalytic behaviour of the Keggin HPA [10]. Among
the HPAs, those having Keggin-type structure are currently used in
industrial catalytic processes [11], while the Dawson-type HPAs
have been used for the selective catalytic oxidative dehydrogena-
tion of isobutene to 2-methyl-propene [12] or for MTBE production
[13].
HPAs are used as photocatalysts and their ground electronic
state can absorb light producing a charge-transfer excited state
(HPA*). This photoexited species can easily become HPA−, the so-
called “heteropolyblue species”, by means of electron transfer from
another species [3]. Heteropolyblue species (HPAn−) are relatively
stable and are readily re-oxidized. The presence of the HPAs sup-
ported on different semiconductor oxides has shown a beneficial
role in photocatalytic reactions due to the ability of the activated
HPA* species to be reduced by the photoproduced electrons of the
+
are H5O2 species.
Baronetti et al. reported the importance of the pseudo-liquid
phase formation in Wells-Dawson acid during methyl ter-butyl
ether (MTBE) synthesis and methanol dehydration to dimethyl
ether in gas phase at 100 ◦C [9]. The authors observed that the loss of
catalytic activity of Wells-Dawson acid with the increase of the pre-
treatment temperature was due to the loss of water molecules in
the secondary structure. The important role of water has been also

116
G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
P2W18/SiO2 (a)
PW12/SiO2 (b)
P2W18 (c)
P2W18/TiO2 (a)
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
600
500
400
300
200
100
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
600
500
400
300
200
100
495 °C
(c)
PW12/TiO2 (b)
495 °C
(c)
(a)
P2W18 (c)
PW12 (d)
(A)
(B)
PW12 (d)
600 °C
600 °C
405 °C
(a)
(a)
300 °C
(d)
(b)
(b)
300 °C
(d)
0
20
40
60
80
0
20
40
Time [min]
60
80
Time [min]
600
425 °C
TiO2 (a)
SiO2 (b)
1200
480 °C
(a)
500
400
300
200
100
(C)
1000
800
600
400
200
0
295 °C
(b)
(b)
(a)
0
20
40
Time [min]
60
80
Fig. 5. NH3-TPD profiles versus time and temperature for SiO2 supported HPAs (A), for TiO2 supported HPAs (B), and for bare SiO2 and TiO2 (C). Bare P2W18 and PW12, were
inserted for comparison in (A) and (B).
conduction band of the UV-activated semiconductor [14–18]. The
redox potentials of Wells-Dawson and of Keggin HPA both in the
ground and even better in the excited state are suitable to allow
3.0
electron transfer from the conduction band of TiO2.
2.5
In this paper the (photo)catalytic 2-propanol dehydration to
propene in gas-solid regime has been carried out by using two types
of HPAs, i.e. H3PW12O40 or H6P2W18O62 both bare and deposited
on SiO2 or TiO2. Some bulk and surface properties of the solids
have been studied and both their catalytic and photocatalytic activ-
2.0
1.5
ity investigated. The research aimed to understand the role of
the HPA on the catalytic process and how the modified physico-
chemical properties of the supported HPAs give rise to different
(photo)catalytic activities with respect to the bare ones.
1.0
0.5
2. Experimental
0.0
2.1. Preparation of the binary heteropolyacid/support samples
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2-propanol concentration [mM]
Two sets of catalysts have been prepared by using two types of
heteropolyacid clusters (HPAs). The Keggin, H3PW12O40, (Aldrich
99.7%) was labelled as PW12. The Wells-Dawson heteropolyacid,
H6P2W18O62, labelled as P2W18, was synthetized in the laboratory.
It was obtained through the synthesis of K6P2W18O62 salt: H3PO4
acid (2 ml) was added to a solution of NaWO4 (0.01 mol) in water
(30 ml) and the solution was refluxed for 8 h. The salt was precipi-
tated by adding KCl (1 g) and purified by recrystallization (t = 5 ◦C).
The product was filtered, washed and then vacuum-dried for 8 h
[9,19,20]. The salt was transformed in phosphotungstic Wells-
Dawson acid, H6P2W18O62, according to the “etherate method” [9].
Fig. 6. Reaction rate of propene formation per gram of HPA versus 2-propanol con-
centration for runs carried out by using bare PW12 ) and bare P2W18 ) as
(
(
catalysts (dotted line) or as photocatalysts (full line). Flow rate of the feeding gas
equal to 100 mL min−1, temperature 80 ◦C and 0.5 g of catalyst.
This process implies the addition of ether and concentrated HCl
(37%) to an aqueous solution of the ␣/␤ mixture of the K6P2W18O62
salt. H6P2W18O62 formed a compound with ether, which can be
separated by extraction. The ether solution, containing the HPA,

G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
117
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
5.0
4.0
3.0
2.0
1.0
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2-propanol concentration [mM]
2-propanol concentration [mM]
Fig. 7. Reaction rate of propene formation per gram of HPA versus 2-propanol con-
centration for runs carried out by using PW12/SiO2 ) and P2W18/SiO2 ) as
Fig. 8. Reaction rate of propene formation per gram of HPA versus 2-propanol con-
centration for runs carried out by using PW12/TiO2 ) and P2W18/TiO2 ) as
(
(
(
(
catalysts (dotted line) or as photocatalysts (full line). Flow rate of the feeding gas
catalysts (dotted line) or as photocatalysts (full line). Flow rate of the feeding gas
equal to 100 mL min−1, temperature 80 ◦C and 0.5 g of catalyst.
equal to 100 mL min−1, temperature 80 ◦C and 0.5 g of catalyst.
was placed in a vacuum-desiccator until crystallization of the HPA
as a polycrystalline solid.
equipped with a thermal conductivity detector (TCD), a quadrupole
mass (QM) spectrometer (Ther-mostar, Balzers) and an ultraviolet
gas analyzer (ABB, Limas 11). The sample amount of 0.3 g was pre-
treated in He flow at 100 ◦C for 30 min. Then, after cooling down to
room temperature, ammonia adsorption was performed by admit-
ting a flow of 5% NH3/He stream (30 ml min−1) for 1 h. In order
to remove all the physically adsorbed ammonia, the sample was
purged by flowing 100 ml min−1 He at 100 ◦C for 1 h. Then, after
cooling down to room temperature, ammonia desorption started
by flowing He (30 ml min−1) and heating up to 600 ◦C (rate of
10 ◦C min−1), holding time at 600 ◦C for 30 min. Ammonia concen-
tration profiles were recorded with the ultraviolet gas analyzer.
TCD and QM data were used to qualitatively confirm the trends.
Both PW12 and P2W18 have been supported by impregnation on
commercial SiO2 (fumed, Aldrich) or TiO2 (Evonik P25). SiO2 or TiO2
were added to an aqueous solution of PW12 or P2W18. The resulting
suspension was stirred, dried and annealed at 50 ◦C overnight. The
aqueous HPA suspensions contained a sufficient amount of PW12
or P2W18 to theoretically form ca. one monolayer onto the oxide
surface, as detailed in the “Results and discussion” section. The
resulting powders were labelled as PW12/SiO2 and P2W18/SiO2 or
PW12/TiO2 and P2W18/TiO2.
2.2. Characterization of the binary heteropolyacid/support
samples
Bulk and surface characterizations were carried out in order
to define some physicochemical properties of the powders. The
crystalline structure of the samples was determined at room tem-
perature by powder X-ray diffraction analysis (PXRD) carried out
by using a Panalytical Empyrean, equipped with CuKa radiation
and PixCel1D (tm) detector. Scanning electron microscopy (SEM)
was performed by using a FEI Quanta 200 ESEM microscope, oper-
ating at 20 kV on specimens upon which a thin layer of gold had
been evaporated. An electron microprobe used in an energy disper-
sive mode (EDX) was employed to obtain information on the actual
metal content present in the samples in order to evaluate the over-
all dispersion of the HPA on the support. Specific surface area and
porosity were determined in accordance with the BET method from
the nitrogen adsorption-desorption isotherm using a Micromerit-
ics ASAP 2020. The structure of the HPA and the preservation of
the cluster structure after the deposition and the (photo)reactivity
experiments were studied by vibrational spectroscopy. FTIR spec-
tra of the samples in KBr (Aldrich) pellets were obtained by using a
FTIR-8400 Shimadzu spectrometer with 4 cm−1 resolution and 256
scans. Raman measurements were performed on pure powdered
samples. The spectra were recorded by a Reinshaw in-via Raman
equipped with an integrated microscope and with a charge-coupled
device (CCD) camera. A He/Ne laser operating at 632.8 nm was used
as the exciting source.
2.3. Catalytic and catalytic photo-assisted experiments
The reactivity set-up consisted of a cylindrical continuous Pyrex
photoreactor horizontally positioned (diameter: 10 mm, length:
100 mm). The reactivity experiments were carried out with 0.5 g of
solid powder placed as a thin layer inside the (photo)catalytic reac-
tor. A porous glass septum allowed to homogeneously distribute
the gaseous inlet mixture. It consisted of a nitrogen flow containing
2-propanol with a concentration ranging between 0.1 and 3.0 mM.
A mass flow controller allowed the N2 feeding, whereas 2-propanol
was added in the N2 stream by means of a home assembled infu-
sion pump. The flow rate of the gaseous stream for both the catalytic
and photocatalytic runs was 100 ml min−1. All the runs have been
carried out at atmospheric pressure. The reactor and the pipes of
the whole set-up, including the reactor, were heated by an electric
resistance. K-type thermocouples allowed to monitor the temper-
ature in the system. The temperature inside the (photo)-reactor
was modified in the range 60–120 ◦C both for catalytic and cat-
alytic photo-assisted experiments. For the photocatalytic runs the
reactor was illuminated from the top by two UV LED IRIS 40 lamps
with an irradiation peak centred at 365 nm. The irradiance reach-
ing the reactor, measured in the range 300–400 nm by a UVX Digital
radiometer, resulted 0.50 W. The temperature of the photoreactor
did not increase due to the irradiation by LEDs. A pretreatment of
the catalysts was carried out under N2 at 60 ◦C for 0.5 h. The runs
lasted ca. 8 or 24 h and samples of the reacting fluid were analyzed
by a GC-2010 Shimadzu gas chromatograph equipped with a Phe-
The acidity of the (photo)catalysts was determined by
temperature-programmed desorption of ammonia (NH3-TPD)
experiments by using a Micromeritics Autochem 2950 apparatus

118
G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
nomenex Zebron Wax-plus column (30 m × 0.32 ␮m × 0.53 ␮m)
and a flame ionization detector, using He as the carrier gas.
Preliminary runs indicated that the presence of the het-
eropolyacid is essential for the conversion of 2-propanol in the
(photo)catalytic reactions. No activity was observed in the pres-
ence of bare SiO2 or TiO2. The occurrence of the reaction needed
t ≥ 60 ◦C both under dark and irradiation conditions.
The amounts of HPAs needed to form the theoretical coverage
of the support has been calculated by taking into account the size
of the HPAs primary structures and the specific surface areas of the
supports (these values are reported in Table 1). The diameter of the
Keggin is equal to ca. 10 Å and by considering a roundish shape,
each cluster results ca. 78.5 Å2 [5]. According to Sambeth et al. [24]
the parameters of the primary Wells-Dawson HPA structure can be
assumed as a rectangular prism sized (21.5 × 15.5 × 12.1) Å; conse-
quently the area occupied by each cluster resulted ca. 94 and 167 Å2
by considering the upright and reclined species, respectively.
FTIR experiments were useful to evaluate the presence of the
Wells-Dawson structure which was a home prepared HPA, but also
to confirm the retention of both Keggin and Wells-Dawson cluster
structures after the dispersion of the HPA onto the oxide surface.
The FTIR spectrum of the P2W18, shown in Fig. 3(A), presents the
3. Results and discussion
3.1. Bulk and textural photocatalysts characterization
Diffractograms of the Keggin based materials along with the
bare oxides are reported in Fig. 1(A). Bare PW12 presents a crys-
talline structure characterized by several diffraction peaks. Anatase
and rutile polymorphs can be identified in the commercial TiO2 P25
diffractogram, whereas SiO2 resulted amorphous. The main peaks
attributable to the Keggin HPA along with those related to anatase
and rutile are present in the PW12/TiO2 composite. The diffrac-
tion of the PW12/SiO2 material shows the main peaks attributed to
the HPA. Diffractogramms related to the Wells-Dawson HPA along
with its composites are reported in Fig. 1(B). The shape of Dawson-
type heteropolyanion is not spherical as the Keggin-type one and
its secondary structure is less ordered resulting in a complicated
diffactogram, as reported before [9,21]. This can be deduced from
XRD spectra, which presented less symmetry. This phenomenon
has been previously observed also for the Keggin structure when
the secondary structure was altered after an annealing process [22].
The HPAs peaks cannot be observed neither in P2W18/TiO2 nor
in P2W18/SiO2, due to their good dispersion on the surface of the
support, as confirmed by SEM.
SEM microphotographs of PW12 and P2W18 supported materi-
als show the characteristic morphology of the support used. Fig. 2
reports pictures of bare SiO2, bare P2W18 and both PW12 and P2W18
supported on SiO2. It can be noticed how the morphology of the
supported materials is very similar to that of the support. On the
contrary, PW12 (picture not reported for the sake of brevity) and
P2W18 lost their morphology when impregnated on the support
because they were well dispersed on SiO2. The same behaviour was
observed for the PW12 and P2W18 supported on TiO2 (pictures not
reported).
As far as the EDX measurements are concerned, they allowed
to obtain information on the actual HPA content only for the TiO2
based samples because the overlapping of Si and W signal in SiO2
based catalyst did not allow to properly analyse the percentage
of Si and W. Anyhow, the measured HPA/TiO2 ratios for both the
impregnated samples were very close to the nominal ones show-
ing small oscillations between various determinations in different
areas of the samples, indicating a good dispersion of HPA. Table 1
reports all of the samples used in the current research along with
some physico-chemical parameters, the mass percentage of the
HPA present in each binary material and the theoretical HPA cov-
erage.
The specific surface areas (SSA) of the oxides impregnated with
the HPAs decreased with respect to the bare oxides, in agree-
ment with previous studies [23] and this was more evident for the
Wells-Dawson cluster, as shown in Table 1. This decrease was more
significant in the case of HPAs supported on SiO2. The block of some
pores of the supports, due to the presence of the HPA clusters, could
account for this result. The observed increase of the diameter of
the pores could be explained by considering the formation of new
pores when HPA was supported. Consequently, the presence of HPA
blocks the smaller pores of the support but induces the formation
of new larger pores.
following strong vibration bands: 1091 cm−1 attributed to the ␯
as
frequency of the PO4 tetrahedron, 962 cm−1 to the ␯ of the ter-
as
minal W O bonds and two bands at 914 and 780 cm−1, assigned
to the ␯ vibration of the W
O W bridges, in agreement to the
as
Wells-Dawson HPA spectrum reported in the literature [25]. The
dispersion of P2W18 on SiO2 and TiO2 did not shift the vibrational
modes of the Wells-Dawson HPA, indicating the preservation of the
cluster structure.
Both Raman and FTIR spectra confirm the Wells-Dawson struc-
ture of the home prepared P2W18. The FTIR spectra confirmed the
retention of both Keggin and Wells-Dawson structures after the
dispersion of the HPA onto the support.
The spectrum of the bare PW12 showed in Fig. 3(B), presents
four characteristic bands [26,27] at 1080, 982, 890, 810 cm−1 corre-
sponding to ␯as(P-Oa), ␯as(W-Od), ␯as(W-Ob-W) and ␯as(W-Oc-W),
respectively. For spectra of the supported PW12 samples, reported
in Fig. 3(B), the Keggin anion skeletal vibrations were observed
at the same frequencies than for the bare PW12, indicating the
retention of the Keggin structure in the binary material. In the
PW12/SiO2 spectra, the band at 1105 is attributed to SiO2. Concern-
ing the PW12/TiO2 sample, the Ti-O bonds in the TiO2 P25 Evonik
gave rise to an absorption spectrum in the range 900 to 700 cm−1
dominating part of the spectrum.
,
Remarkably, all of the supported materials showed no impor-
tant differences in the FTIR spectrum before and after the catalytic
or the photocatalytic experiments, as reported in Fig. 3(C). The spec-
tra of the catalytically or photocatalytically used PW12 and P2W18
supported on TiO2, showed neither shifts nor differences in rela-
tive intensity of the bands or additional bands with respect to the
freshly prepared samples. With this respect, it has been reported
that the modification of the secondary structure of the HPAs, for
instance by a partial loss of water, increases the disorder in the sec-
ondary structure, hence inducing the bands at 911 and 778 cm−1 to
become broader and with lower intensities [25]. The primary struc-
ture remains stable, although the distance between the primary
units can change. This behaviour was not observed, as both HPAs
maintained their secondary structure after catalytic and photocat-
alytic reactions even after 120 ◦C. This finding can be explained by
taking into account that the dehydration reaction produced water
molecules that maintained an extent of hydration sufficient for the
Brönsted sites to be still active.
As far as the Raman vibrational spectra are concerned, the
Wells-Dawson heteropolyanion exhibits multiple WdO vibrations
between 950 and 1005 cm−1, reflecting a distribution of distortions
among the WO6 units in the framework. Those species possess a
strong Raman signal at 998 cm−1 corresponding to the ␯ of the
s
terminal W O bonds, along with other signals of lower intensity
at 972, 920 and 853 cm−1, attributed to ␯ of the terminal W-
as
O bonds, ␯ of bridging W-O-W and ␯ of O-W-O bonds of the
as
as
extended polytungstate framework surrounding the central PO4
group [27]. The Raman spectrum of the home prepared Wells-

G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
119
4
2
4
PW12
P2W18
2
0
0
-2
-4
-6
-2
-4
-6
-8
-8
0.0025
0.0027
0.0029
0.0031
0.0025
0.0027
1/T [K-1]
0.0029
0.0031
1/T [K-1]
Fig. 9. Plots of “ln r” versus “1/T” used to calculate “Ea” and “ln A”. Experiments carried out by using the bare HPAs as catalysts (ꢀ) or photocatalysts (᭿). The units of “r” are
mmol h−1 g−1
.
HPA
6
4
8
6
P2W18/SiO2
PW12/SiO2
2
4
0
2
-2
-4
-6
0
-2
-4
-8
-6
0.0025
0.0027
0.0029
0.0031
0.0027
0.0029
0.0031
0.0025
1/T [K-1]
1/T [K-1]
6
5
4
5
4
PW12/TiO2
P2W18/TiO2
3
3
2
2
1
1
0
0
-1
-2
-3
-1
-2
-3
0.0025
0.0027
0.0029
0.0031 0.0025
0.0027
0.0029
0.0031
1/T [K-1]
1/T [K-1]
Fig. 10. Plots of “ln r” versus “1/T” used to calculate “Ea” and “ln A”. Experiments carried out by using the SiO2 and TiO2 supported HPAs as catalysts (ꢀ) or photocatalysts (᭿).
The units of “r” are mmol h−1 g−1
.
HPA
Table 1
BET specific surface area (SSA), pore size, mass percentage of HPA on the binary material, theoretical HPA coverage on the support.
Sample
SSA [m2 g−1
]
Pore size [nm]
Mass percentage of HPA [%]
Theoretical HPA coverage
SiO2
TiO2
PW12
P2W18
319
56
15
5
9.5
22.5
–
–
–
–
–
–
–
–
–
–
PW12/SiO2
PW12/TiO2
P2W18/SiO2
P2W18/TiO2
53
50
38
46
25.2
28.8
28.5
24.9
70
26
70
26
1.1
0.9
0.8–1.4a
0.7–1.3a
a
Values calculated by considering two different positions of the species on the surface of the support: upright or reclined.

120
G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
Table 2
3.2. Catalytic and photocatalytic reactivity
Apparent activation Energy (Ea) and logarithm of the pre-exponential factor (ln A)
for catalytic and photocatalytic 2-propanol dehydration.
The presence of the heteropolyacid is essential for the dehydra-
tion of 2-propanol both in catalytic and photocatalytic reactions;
in fact, no activity was observed in the presence of both bare TiO2
and SiO2 samples. Moreover, the occurrence of the reaction needed
t ≥ 60 ◦C both under irradiation and dark conditions. Preliminary
runs indicated that it was necessary to feed the (photo)reactor with
a flow rate of 100 mL min−1 in order to work under kinetic regime
conditions.
Ea [kJ/mol]
Catalytic
ln A
Catalyst
Photocatalytic
Catalytic
Photocatalytic
PW12
186
151
101
175
161
92
157
144
96
152
139
86
56
51
35
56
52
32
50
50
34
51
47
31
PW12/SiO2
PW12/TiO2
P2W18
P2W18/SiO2
P2W18/TiO2
Catalytic and photocatalyticexperimentswere carried out at dif-
ferent initial 2-propanol concentrations and by using both the bare
and supported HPAs. Propene and di-isopropyl ether were the main
reaction products observed both in the absence and in the presence
of UV light, but the extent of ether formation was generally much
lower with respect to that of propene (ca. one order of magnitude
less, under irradiation). This finding indicates that the selectivity
of the 2-propanol dehydration reaction versus propene formation
ranged between 85 and 93% in the absence or in the presence of
light, respectively. These values were approximately the same for
all of the catalysts used. The rate of propene formation per gram
of HPA versus 2-propanol concentration in the feeding stream is
reported in Fig. 6 for runs carried out by using bare PW12 or P2W18
as the (photo)catalysts.
In all of the experiments, the rate of propene formation firstly
increased and then decreased by increasing 2-propanol concen-
tration. Consequently, a maximum substrate concentration exists
above which a dramatic decrease of reactivity occurs. HPAs present
the uncommon ability to absorb polar substrates in their bulk giv-
ing rise to catalytic reactions in the “pseudo-liquid” phase [5,6].
Therefore the solubility of 2-propanol in the water molecules of
the secondary structure of the HPA should be invoked to explain
their reactivity. This phenomenon has been observed for the cat-
alytic formation of tertiary ethers in gas phase by addition of polar
alcohols (methanol or ethanol) to isobutene for experiments car-
ried out with H4SiW12O40 and H6P2W18O62 [32] or in the presence
of PW12 and P2W18 [8,13]. The same phenomenon was observed by
using bare and supported H3PW12O40 in the catalytic and photo-
catalytic 2-propanol dehydration [33]. In Fig. 6 it can be observed
that the reaction rate decrease started at lower 2-propanol con-
centration for the catalytic experiments than for the photocatalytic
ones. Moreover, Wells-Dawson HPA material was more active than
the Keggin one.
According to the literature, the catalytic dehydration of 2-
propanol occurs by means of an acid-base mechanism (elimination
E1) involving the dioxonium ions placed between the HPA anions
[33,34]. Consequently, the amounts of acid sites present in the cata-
lysts should account for their activity. The NH3-TPD experiments in
accord with the reactivity results, indicate that P2W18 shows higher
amount of acid sites with respect to PW12. Here it is important to
underline that, although it is reported the higher catalytic activity
of the Wells-Dawson HPA material with respect to the Keggin one,
this finding is considered surprising because the adsorption heat
of NH3 (evaluated by microcalorimetry) indicates the presence of
stronger acid sites in Keggin HPA [6,30,32].
The NH3-TPD study reported in this work confirmed that PW12
presents stronger acid sites with respect to P2W18. However, the
average amount of acid sites in P2W18 (0.33 mmolNH3 g−1HPA) is
significantly higher than that in PW12 (0.13 mmolNH3 g−1HPA). Con-
sequently, the activity of the samples should be related not only to
the strength of the acid sites but also to their amount.
Dawson heteropolyacid P2W18 is shown in Fig. 4. The selected
sample P2W18/TiO2, shown for the sake of comparison, confirmed
the upshifted position of the vibrational modes, in accord with the
FTIR conclusions regarding the preservation of the cluster after the
dispersion onto the oxides surface. Moreover, Fig. 4 shows charac-
teristic Raman vibrational modes of the TiO2 anatase centered at
144, 197, 399, 513, and 639 cm−1, attributable to the Eg, Eg, B1g, A1g
and B2g [28]. A peak attributed to the rutile phase is also present and
located at 444 cm−1. Intense peaks at 142 and 144 cm−1, assigned to
rutile and anatase, respectively, are out of the range of the spectra
reported.
The acid properties of HPAs supported samples were evaluated
by NH3-TPD experiments carried out from room temperature up to
600 ◦C. Such high temperature was chosen in order to detect also
the contribution of strong acid sites that are known to desorb NH3
above 550 ◦C. According with the literature we were confident that
in the selected range of temperature both Keggin and Dawson acids
still keep their heteropolyoxoanion structure [9,10].
The technique provides information on the total acidity of the
catalysts without distinguishing between Brönsted acid sites (typi-
cal of HPAs and SiO2) and Lewis acid sites, characteristic of defective
TiO2 support. The amount of NH3 desorbed (ppm/g HPAs) gives
a quantitative evaluation of the number of active sites, while the
temperature of desorption is an indication of the strength of the
acid sites. In Fig. 5(A)–(C) the NH3-TPD profiles are displayed
for bare HPAs, for SiO2 and TiO2 supported HPAs and for bare
supports, respectively. No NH3 desorption occurs at temperature
below 200 ◦C, suggesting the absence of weak acid sites [23,29].
For bare and SiO2 supported P2W18 the main desorption occurs
with a broad and intense peak centred at 495 ◦C corresponding to
15,260 and to 12,780 ppm of NH3 desorbed per gram of HPA, respec-
tively. According with our previous results [29] NH3 desorption in
the range 200–550 ◦C has been attributed to medium strength acid
sites. A pronounced shoulder at around 300 ◦C was also detected for
P2W18/SiO2. By comparison with the profile of bare P2W18 and that
of bare SiO2 support it can be concluded that such low temperature
feature is due to the contribution of SiO2 medium acid sites. More-
over, a careful analysis of the TPD curves suggests for both samples,
P2W18 and P2W18/SiO2, the presence of some strong acid sites.
Unlike P2W18 samples, in the case of bare and SiO2 supported PW12
the main desorption peak of ammonia (around 10,000 ppm/g HPA)
was detected at 600 ◦C confirming that the acid sites in Keggin-type
heteropolyacids are stronger than that of Dawson-type HPAs [30].
By looking at the NH3-TPD curves of TiO2 supported HPAs, a broad
peak centered at around 450 ◦C was observed for both samples.
Along with such features, attributed to the Lewis acid sites typical
of TiO2 (see Fig. 5(C)) [31], the peaks corresponding to the Brönsted
acid sites of the HPAs were clearly detected at 495 (12,200 ppm of
NH3/g HPA) for P2W18/TiO2 and at 600 ◦C (∼6000 ppm NH3/g HPA)
for PW12/TiO2.
By irradiating the system, the reaction rate of propene formation
increased for both HPAs and the 2-propanol concentration corre-
sponding to the maximum reactivity shifted to a higher substrate
concentration (from 0.25 to 0.5 mM for PW12, and from 0.5 to 1 mM
for P2W18), as showed in Fig. 6. The increase of the reactivity under

G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
121
UV irradiation will be explained later along with the behaviour of
the supported samples.
trap an electron from 2-propanol giving rise to the heteropolyblue
species (see Introduction Section). Notably, all materials contain-
ing HPAs become strongly blue coloured under irradiation only in
the presence of 2-propanol. The further hypothesized evolution of
2-propanol radical species to propene is reported in [33].
The catalytic and photocatalytic propene formation rates per
gram of HPA in the presence of SiO2- and TiO2-supported HPAs
versus the concentration of 2-propanol in the feeding stream are
reported in Figs. 7 and 8, respectively. The overall catalytic and
photocatalytic behaviours of the supported PW12 and P2W18 were
similar to those observed for the bare HPAs, indicating that the
dehydration reaction occurred in the pseudo-liquid phase also for
the supported HPAs. Consequently, as reported in Figs. 7 and 8,
the rate of propene formation firstly increased (reaching a maxi-
mum) and then decreased by increasing 2-propanol concentration
when the supported samples were used as the (photo)catalysts. The
supported Wells-Dawson HPAs were more active than the analo-
gous Keggin HPAs for the catalytic process, although the activities of
the supported samples were very similar at the lowest 2-propanol
concentration. The different catalytic activity of PW12 and P2W18
supported samples could be related, as for the bare HPAs, to the
higher amounts of acid sites present on the P2W18 supported cat-
alysts. Propene formation rate increased under UV irradiation for
all of the solids and the supported P2W18 samples were always the
most active ones. Notably 2-propanol conversion always decreased
by increasing 2-propanol concentration in the feeding gas, being
averagely ca. 70% and 8% for the lowest and the highest concentra-
tions, respectively, for both types of supported catalysts.
The fact that TiO2 showed to be the best support in the photo-
catalytic reaction suggests the occurrence of a synergistic effect in
which the 2-propanol radical species formed as above reported are
reduced to propene by trapping photo-generated electrons in the
conduction band of TiO2 and contemporaneously the heteropoly-
blue species are re-oxidized to HPAs [33]. Moreover, it can not
be excluded that the photosensitized HPA possesses an oxidation
potential sufficient to abstract a photogenerated electron from the
conduction band of TiO2. Such electron transfer could (i) produce
a higher number of HPA− species and (ii) inhibit the fast electron-
hole recombination on TiO2 [16,18]. Nevertheless, these two effects
should result very useful when TiO2 is used as the photocatalyst (for
oxidation reaction) but not in this work in which the active phases
were the HPA species. The increase of the reactivity observed under
irradiation was sometimes very high, and similar performances
were achieved only when the temperature for catalytic reactions
was ca. 15–20 ◦C higher with respect to the photocatalytic ones.
Interestingly, some runs carried out for long time (ca. 24 h)
showed that the reactivity and the selectivity of the catalysts did
not change indicating a good stability of these samples.
A perusal of Figs. 6–8 indicates that the reaction rates of propene
formation in the presence of HPAs/SiO2 or HPAs/TiO2 samples were
higher than in the presence of bare HPAs, both for catalytic and
photocatalytic experiments.
The increase of the activity of the supported materials with
respect to the bare ones can be explained by considering the higher
SSA values of these samples accounting for a larger surface of con-
tact between the cluster and 2-propanol.
The reactivity study of the supported samples, as above
reported, was carried out by using catalysts prepared with the same
theoretical HPA coverage, i.e ca. 1 monolayer, both on TiO2 and
SiO2, because the aim of this work was to compare the reactivity of
two heteropolyacids both bare and supported with the same cover-
age in two different oxides. Anyway, for the sake of completeness
two additional samples were prepared in which HPAs were sup-
ported on SiO2 with a weight percentage of 26% (the same used
for the HPAs/TiO2 samples) and they were tested to compare also
the reactivity of supported samples with the same HPA percentage.
For these two samples, whose the theoretical coverage of PW12 or
P2W18 on SiO2 was ca. 0.2 monolayer, the (photo)catalytic tests
were carried out only with 1 mM 2-propanol in the feeding gas.
The results, not reported for the sake of brevity, indicated that the
(photo)reactivity figures were lower than those of the correspond-
ing HPAs/SiO2 samples with the highest percentage of HPAs when
the whole catalyst mass was considered, but they showed to be
higher per gram of heteropolyacid. Moreover, the (photo)reactivity
was very similar to that showed by the HPAs/TiO2 samples. This
finding indicates that a more detailed study on the HPA coverage
is needed in the future to establish what is the most performing
catalyst.
Notably, the reaction rate decrease vs. 2-propanol concentra-
tion in the feeding stream was more dramatic for the catalytic
than for the photocatalytic experiments. As shown in Figs. 6–8,
it is impossible to find a unique concentration of 2-propanol for
which the reactivity was maximum, however the 1 mM concentra-
tion resulted a good compromise in order to compare the reactivity
of the various samples. By studying the activity results reported in
Figs. 7 and 8 for runs carried out at 2-propanol concentration equal
to 1 mM, it can be concluded that: (i) the activity under dark con-
dition in the presence of PW12/SiO2 was higher than that observed
with the PW12/TiO2 sample; (ii) the activity under dark condition
by using P2W18/SiO2 was very similar than that obtained in the
presence of P2W18/TiO2 samples; (iii) the activity under irradiation
of the HPAs supported on TiO2 was always higher than that of the
corresponding HPAs/SiO2 samples. The NH3-TPD study indicates
that, among the supported samples, PW12/SiO2 and PW12/TiO2
were the less acidic ones (the total amount of ammonia des-
orbed was 0.24 and 0.36 mmolNH3 g−1HPA, respectively), whereas
P2W18/SiO2 and P2W18/TiO2 showed very similar acidity (0.44
and 0.46 mmolNH3 g−1HPA, respectively) that was higher than that
observed for the corresponding supported PW12 samples (see also
Fig. 5). These findings are in good agreement with the reactivity
showed under dark conditions, i.e. the catalytic experiments, indi-
cating that the most important factor controlling the reactivity is
the acidity of the catalyst. The UV irradiation clearly gave rise to an
increase of the performance of both HPAs, which was significantly
higher for the samples supported on TiO2. The beneficial role of UV
irradiation in the dehydration reaction of 2-propanol to propene
carried out in the presence of bare and hydrothermally prepared
supported PW12 was previously reported [33]. Briefly, during the
catalytic photo-assisted reactions, in addition to the previous men-
tioned acid-base mechanism, a further key role is played by both
HPAs (PW12 and P2W18). Indeed, HPA after photosensitization can
3.3. Study on the apparent activation energy of the catalytic and
catalytic photo-assisted reactions
The apparent activation energy of 2-propanol dehydration reac-
tion to propene has been estimated by applying the Arrhenius
equation by considering a zero order rate reaction. The experiments
at increasing temperature were carried out at concentration of 2-
propanol equal to 3 mM because, as showed in Figs. 7 and 8, at
this concentration the reaction rate of propene formation was in
a plateau by using all of the (photo)catalysts and 2-propanol con-
version was less than 10%. This finding indicates that 2-propanol
completely covered the catalysts surface for a concentration equal
to 3 mM giving rise to a paramount absorption in the pseudo-liquid
phase [5,6], and consequently the reaction rate became of pseudo-
zero order [34].
Therefore, the rate of propene formation can be written as:
r = k0
(1)

122
G. Marcì et al. / Applied Catalysis A: General 528 (2016) 113–122
in which “k0” is the pseudo-zero order rate constant of the reaction.
The apparent activation energies Ea and the pre-exponential
factors A were determined from the Arrhenius equation in the form:
Acknowledgements
The authors wish to thank Prof. Anna Micek-Ilnicka for fruitful
suggestions and MIUR (Rome) and Università degli Studi di Palermo
for financial support.
Ea
ln r = ln k0 = ln A −
(2)
RT
Eq. (2) has been applied for the runs carried out both in the
presence and in the absence of UV-light by using both the bare and
the supported HPA materials in the catalytic or the photocatalytic
reaction. Figs. 9 and 10 report the Arrhenius plots for all of the
runs. The values of Ea and ln A estimated by means of these plots
are reported in Table 2.
The values of Ea calculated in the presence of UV light (pho-
tocatalytic) were always lower than those estimated under dark
conditions (catalytic), justifying the higher reactivity observed in
the photocatalytic reactions, as shown in Table 2. It can be observed
that the value of Ea for P2W18/SiO2 under irradiation decreased
more significantly (13.6%) than for P2W18/TiO2 (6.5%), although the
photoreactivity for the last sample appeared to be always higher.
This apparent contradictory result can be explained by taking into
account that the percentage decrease under irradiation of the pre-
exponential factor A, which is related to physico-chemical factors as
adsorption/desorption of reagent and products, was also higher for
P2W18/SiO2 (9.6%) than for P2W18/TiO2 (3.1%). Notably, this factor
influences the kinetic constant in the opposite way to Ea as before
reported [34].
References
[1] M.T. Pope, A. Müller, Angew. Chem. Int. Ed. Engl. 30 (1991) 34–48.
[2] S.S. Wang, G.Y. Yang, Chem. Rev. 115 (2015) 4893–4962.
[3] E. Papaconstantinou, Chem. Soc. Rev. 18 (1989) 1–33.
[4] E.I. García-López, G. Marcì, L. Palmisano, Heterogeneous Photocatalysis: From
Fundamentals to Green Applications, in: J.C. Colmenares, Y.J. Xu (Eds.),
Springer-Verlag, Berlin Heidelberg, 2016, pp. 63–107.
[5] N. Mizuno, M. Misono, Chem. Rev. 98 (1998) 199–218.
[6] T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 41 (1996) 113–252.
[7] L.E. Briand, G. Baronetti, H.J. Thomas, Appl. Catal. A 256 (2003) 37–50.
[8] S. Shikata, M. Misono, Chem. Commun. (1998) 1293–1294.
[9] G. Baronetti, L. Briand, U. Sedran, H.J. Thomas, Appl. Catal. A 172 (1998)
265–272.
[10] A. Micek-Ilnicka, J. Mol. Catal. 308 (2009) 1–14.
[11] I.V. Kozhevnikov, Catalysis for Fine Chemical Synthesis. Vol. 2: Catalysis by
Polyoxometalates, John Wiley and Sons, Chichester, 2002.
[12] F. Cavani, C. Comuzzi, G. Dolcetti, E. Etienne, R. Finke, G. Selleri, F. Trifirò, A.
Trovarelli, J. Catal. 160 (1996) 317–321.
[13] S. Shikata, T. Okuhara, M. Misono, J. Mol. Catal. A 100 (1995) 49–59.
[14] G. Marcì, E.I. García-López, L. Palmisano, Eur. J. Inorg. Chem. (2014) 21–35.
[15] H. Park, W. Choi, J. Phys. Chem. B 107 (2003) 3885–3890.
[16] T. Tachikawa, M. Fujitsuka, T. Majima, J. Phys. Chem. C 111 (2007) 5259–5275.
[17] A. Hiskia, M. Athanasios, E. Papacostantinou, Chem. Soc. Rev. 30 (2001) 62–69.
[18] M. Yoon, J.A. Chang, Y. Kim, J.R. Choi, K. Kim, S.J. Lee, J. Phys. Chem. B 105
(2001) 2539–2545.
[19] D.K. Lyon, W.K. Miller, T. Novet, P.J. Domaille, E. Evitt, D.C. Johnson, R.G. Finke,
J. Am. Chem. Soc. 13 (1991) 7209–7221.
4. Conclusions
[20] C.R. Graham, R.G. Finke, Inorg. Chem. 47 (2008) 3679–3686.
[21] E. Arendt, S. Zehri, P. Eloy, E.M. Gaigneaux, Eur. J. Inorg. Chem. (2012)
2792–2801.
Catalytic and photocatalytic tests for 2-propanol dehydration to
propene were successfully carried out by using bare SiO2 or TiO2
supported Keggin and Wells-Dawson HPAs. The Wells-Dawson het-
eropolyacid presented a higher catalytic and photocatalytic activity
than the Keggin one particularly when supported. The UV irradia-
tion increased the activity of both HPAs. The heteropolyacid species
played a key role both for the catalytic and the photocatalytic reac-
tions, indicating that the acidity of the cluster accounts for the
catalytic role (P2W18 showed higher activity with respect to PW12
because of the presence of higher amount of acid sites), whereas
both the acidity of the cluster and the oxidant ability of excited
PW12 and P2W18 were responsible for the increase of the reaction
rate of the photocatalytic reaction. Moreover, the presence of TiO2,
used as a photoactive semiconductor support, showed a beneficial
effect to enhance the reactivity of the binary materials. Finally, the
apparent activation energy of 2-propanol catalytic and photocat-
alytic dehydration, determined in the range 60–120 ◦C, decreased
in the presence of light for all of the (photo)catalysts used.
[22] U.B. Miocˇ, R.Z. Dimitrijevic´, M. Davidovic´, Z.P. Nedic´, M.M. Mitrovic´, Ph.
Colomban, J. Mater. Sci. 29 (1994) 3705–3718.
[23] G. Marcì, E.I. García-López, M. Bellardita, F. Parisi, C. Colbeau-Justin, S. Sorgues,
L.F. Liotta, L. Palmisano, Phys. Chem. Chem. Phys. 15 (2013) 13329–13342.
[24] J.E. Sambeth, G. Romanelli, J.C. Autino, H.J. Thomas, G.T. Baronetti, Appl. Catal.
A 378 (2010) 114–118.
[25] G.T. Baronetti, H. Thomas, C.A. Querini, Appl. Catal. A 217 (2001) 131–141.
[26] C. Rocchiccioli-Deltcheff, M. Fournier, R. Franck, R. Thouvenot, Inorg. Chem.
22 (1983) 207–216.
[27] A. Bridgeman, Chem. Phys. 287 (2003) 55–69.
[28] L.M. Martínez-Tejada, A. Mun˜oz, M.A. Centeno, J. A: Odriozola, J. Raman,
Spectroscopy 47 (2016) 189–197.
[29] M.L. Testa, V. La Parola, F.L. Liotta, A.M. Venezia, J. Mol. Catal. A 367 (2013)
69–76.
[30] M.N. Timofeeva, Appl. Catal. A 256 (2003) 19–35.
[31] G. Martra, Appl. Catal. A 200 (2000) 275–285.
[32] A. Bielan´ ski, A. Micek-Ilnicka, Inorg. Chem. Acta 363 (2010) 4158–4162.
[33] E.I. García-López, G. Marcì, F.R. Pomilla, A. Micek-Ilnicka, A. Kirpsza, L.
Palmisano, Appl. Catal. B 189 (2016) 252–265.
[34] G.C. Bond, S.J. Frodsham, P. Jubb, E.F. Kozhevnikova, I.V. Kozhevnikov, J. Catal.
293 (2012) 158–164.