Ethanol dehydration on silica-aluminas: Active sites and ethylene/diethyl ether selectivities

Article abstract of DOI:10.1016/j.catcom.2015.05.009

Commercial silica-alumina catalysts prepared by different procedures have been characterized. Both present strong Lewis acidity together with Br?nsted sites able to protonate pyridine. No evidence of "zeolitic" bridging OH's but significant heterogeneity of terminal silanol groups, part of which are likely "pseudobridging", was found. Similar high activity in ethanol conversion but markedly different selectivities to ethylene and diethyl ether were found. They are less active than both zeolites and γ-Al₂O₃. Lewis sites with alumina-like acidobasic neighbor are more selective for ethylene production while Lewis sites with silica-like covalent neighbor are more selective for diethyl ether.

Full text of DOI:10.1016/j.catcom.2015.05.009

Ethanol dehydration on silica-aluminas: Active sites and ethylene / diethyl
ether selectivities
Thanh Khoa Phung, Guido Busca
PII:
DOI:
Reference:
S1566-7367(15)00198-3
doi: 10.1016/j.catcom.2015.05.009
CATCOM 4328
To appear in:
Catalysis Communications
Received date:
Revised date:
Accepted date:
3 April 2015
8 May 2015
9 May 2015
Please cite this article as: Thanh Khoa Phung, Guido Busca, Ethanol dehydration on
silica-aluminas: Active sites and ethylene / diethyl ether selectivities, Catalysis Commu-
nications (2015), doi: 10.1016/j.catcom.2015.05.009
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

ACCEPTED MANUSCRIPT
ETHANOL DEHYDRATION ON SILICA-ALUMINAS: ACTIVE SITES AND ETHYLENE /
DIETHYL ETHER SELECTIVITIES
a,b
a,
Thanh Khoa Phung , Guido Busca *
a
Dipartimento di Ingegneria Civile, Chimica e Ambientale, Università di Genova, P.le
J.F. Kennedy 1, I-16129, Genova, Italy
b
NTT Institute of Hi-Technology (NIH), Nguyen Tat Thanh University, 298-300A
Nguyen Tat Thanh Street, Ward 13, District 4, Ho Chi Minh City, Vietnam
*
Tel: (+39) 010 353 6024, Fax: (+39) 010 353 6028,
Email: Guido.Busca@unige.it (G. Busca)
Abstract
Commercial silica-alumina catalysts prepared by different procedures have been
characterized. Both present strong Lewis acidity together with Brønsted sites able to
protonate pyridine. No evidence of “zeolitic” bridging OH’s but significant heterogeneity of
terminal silanol groups, part of which are likely “pseudobridging”, was found. Similar high
activity in ethanol conversion but markedly different selectivities to ethylene and diethyl
ether were found. They are less active than both zeolites and -Al O . Lewis sites with
2
3
alumina-like acidobasic neighbour are more selective for ethylene production while Lewis
sites with silica-like covalent neighbour are more selective for diethyl ether.
Keywords: silica-alumina; ethanol; ethylene; diethyl ether; surface acidity; selectivity.
1
Introduction
Silica-aluminas are widely used as solid acid catalysts, catalyst supports and binders in
various refining and petrochemical industrial processes, including hydrotreating, mild
1

ACCEPTED MANUSCRIPT
cracking, isomerization, oligomerization and alkylation [1-4]. Silica-aluminas are actually
different materials depending on the preparation method, Si/Al ratio and pretreatments.
Amorphous silica-aluminas (ASA) are silica-rich materials prepared by co-precipitation or
co-gelling from mixed Si and Al precursors. Their most typical composition is with a
SiO /Al O molar ratio 10-12 corresponding to an alumina content of 12-15 wt.% [5,6]. In
2
2
3
recent years, a number of materials belonging to this system with relevant mesoporosity
have been prepared and developed at the industrial level [7]. Similar amorphous materials
can be prepared by grafting moderate amounts of Al compounds on high surface area
amorphous silica followed by calcination [8,9].
Another family of materials are prepared by grafting silica precursors on alumina or
boehmite. This is the case in particular of the Siralox family of Sasol (previously Condea),
whose preparation and characterization has been reported in some detail [10].
Silica-aluminas are usually reported possessing Brønsted acid sites together with Lewis
acid sites, and have been characterized since early times by different methods [11-15].
However, still the nature of Brønsted acid sites in silica-alumina is not fully established.
Several authors propose that Brønsted acidity of silica-alumina is due to small amounts of
“zeolitic” bridging Si-OH-Al groups [16-18]. Other authors instead do not find in these
-1
samples the spectroscopic features typical of zeolitic OH’s (IR bands at 3650-3500 cm ,
1
H NMR bands at 3.8-5.2 ppm) but only those typical of terminal silanols Si-OH (IR band
-1
1
at ca. 3745 cm and H peak at 1.7-1.8 ppm) [8,19-21], whose acidity may be enhanced
by nearest aluminum ions. In agreement, theoretical studies suggest the presence of
“pseudobridging” OH’s (i.e. terminal silanols prone to bridge over Al ions when they
interact with a basic molecule) as the active sites [22,23]. A very recent study by Caillot et
al. suggests that the nature of the active site may depend on the preparation procedure
[
24].
We recently investigated the activity of several acid catalysts in converting ethanol to
ethylene [25-27]. Although zeolites are more active than silica-alumina, the key question is
stability on time on stream, with reduced coking. Silica-aluminas have been used in the
past and found suffuiciently stable for this reaction [28]. Over all catalysts, at low
temperature and conversion, diethyl ether (DEE) is found as the main product, while, at
higher temperature and conversion, ethylene becomes the main product. The mechanisms
2

ACCEPTED MANUSCRIPT
of these reactions are also the object of studies and some disagreement [29-31]. In this
communication we report on our study of ethanol conversion on silica-alumina materials.
The obtained data will allow us to propose some conclusions on both ethanol dehydration
mechanism and on the surface acidity of silica-aluminas.
2
Experimental
The properties of the catalysts investigated and their notations are summarized in Table 1.
Catalytic experiments were performed at atmospheric pressure in a tubular flow reactor
feeding 7.9% v/v ethanol in nitrogen at 298 K (total flow rate of 80 cc/min) as described in
our previous studies [25-27].
IR spectra were recorded using Nicolet 380 FT-IR spectrometers, with KBr disks for
skeletal studies and pure powders pressed disks for adsorption studies, performed as
described elsewhere [25,27]. Results
XRD analysis of the samples under study show complete amorphicity of the SiO (S) and
2
the SA87 samples. XRD of A and SA5 show the patterns of a spinel-type alumina phases,
slightly different (-Al O for A and ,-Al O for SA5), while that of SA30 shows the same
2
3
2
3
features, weak and broad. The skeletal IR spectra of the catalysts are presented in Fig. 1.
Two very broad and poorly resolved bands in the medium IR region in the spectrum of A at
-1
ca. 550 and 830 cm are typical for -Al O [32]. The spectrum of the sample SA5 is
2
3
similar, but shows the evolution towards the spectrum of -Al O (sharp components on
2
3
-1
the main absorption) and a pronounced broad feature in the region1150-980 cm , due to
Si-O stretchings of silicate species. The spectrum of SA87 and SiO are typical of
2
-1
amorphous silica based materials (Si-O stretchings at ca. 1230 and 1090 cm , Si-O-Si in
-1
-1
plane deformation at 808 cm and Si-O-Si rocking at 459 cm [33]). In the case of silica
-1
gel the additional absorption at ca. 1000 cm is due to the Si-O stretching of silanol
groups, very abundant. The spectrum of SA30 combines those of amorphous silica
alumina and spinel type Al O , in agreement with the XRD pattern.
2
3
The IR spectra of investigated samples in the region of the OH stretching of the surface
hydroxyl groups, recorded after outgassing at 773 K, are presented in Fig. 2. The
-1
spectrum of A shows the main bands at 3770, 3725 and 3680 cm , typical of the surface
OHs of -Al O , as discussed previously [32,34]. SA5 shows weaker nearly the same
2
3
-1
bands with an additional sharp band at 3741 cm , assigned to OH stretching of a silanol
3

ACCEPTED MANUSCRIPT
group. The spectra of SA30 and SA87, and that of Si as well, give rise to a single band at
-1
3
744  2 cm with a tail at lower frequencies, attributed to terminal silanol groups [34]. On
SA30 and SA87 part of these groups may be “pseudobridging”, i.e. located near Al cations
and able to bridge after deprotonation [19,23]. As for comparison, in Fig. 2 the spectrum of
a typical zeolite, H-ferrierite in this case, is also reported. In this case, together with a band
due to silanol groups assumed to be located in the outer surface, the band of zeolitic
-1
bridging OH’s at 3595 cm is well evident [34]. It is clear that this band is absent in the
spectra of silica-alumina samples.
Fig. 3 shows the IR subtraction spectra of residual adsorbed pyridine on investigated
catalysts after outgassing at 373 K. Pyridine adsorbed on A gives rise to three 8a
-1
components observed at 1578, 1616 and 1624 cm , corresponding to a complex 19b
-1
band center at 1452 cm , revealing the existence of at least three different families of
Lewis acid sites, and no significant Brønsted acidity, as typical of γ-Al O . It seems likely
2
3
[
32] that the most acidic sites are predominantly located on corners and edges in the
nanocrystals of aluminas. The spectrum of SA5 is closely similar to γ-Al O , with possibly a
2
3
-1
smaller relative intensity of the 8a component at 1624 cm . This indicates that this
material does not show Brønsted acidity sufficiently strong to cause pyridine protonation. It
possesses strong Lewis acidity, and that part of the strongest Lewis sites might be
neutralized by interaction with silicate species.
In the case of both SA30 and SA87, the features of pyridine adsorbed on strong Lewis
-1
-1
sites are also evident (8a mode at 1621 cm , 19b mode at 1455 cm ) together with the
-1
-1
bands associated to pyridinium ions at 1639 cm , 8a, and 1547 cm , 19a. This indicates,
in agreement with literature data, that such silica-alumina samples have, together with
strong Lewis acidity, also Brønsted acidity sufficiently strong to cause pyridine protonation.
Pyridine adsorbs only weakly on S via H-bond and is fully desorbed at 373 K, thus
confirming that amorphous silica neither possess Brønsted acidity sufficiently strong to
cause pyridine protonation, nor Lewis acidity.
In Fig. 4 a full experiment of adsorption and desorption of pyridine on SA30,
showing the formation of hydrogen bonded and weakly adsorbed pyridine (H),
Lewis bonded pyridine (L) and pyridinium ion (Brønsted bonded, B,Fig. 4C). In Fig.
4
A and 4B, the analysis of the region of the surface OH groups provides evidence
-1
of some complexity of the main band centered at ca. 3744 cm , where three
4

ACCEPTED MANUSCRIPT
different components behaving differently can be distinguished at 3747, 3741 and
-1
3
726 cm , likely due to different kinds of terminal and/or “pseudobridging” hydroxyl
-1
groups. The overall band shifts upon pyridine adsorption down to near 3000 cm , in
parallel to the formation of the H species,and is recovered by outgassing at
increasing temperature when the species disappear. Components in the region
-1
3
650-3550 cm are found neither in the normal spectra nor in the un-subtracted
ones, thus excluding the presence of zeolite-like OH’s. This agrees with the
conclusion of Caillot et al. for the same sample [24]. The same is found for SA87,
as already reported years ago [35].
In the table 2 and table 3, as well as in Fig. 5, the data concerning the catalytic activity of
investigated catalysts in ethanol dehydration are summarized. The mass balance is
essentially always fulfilled in these conditions. The trend of ethanol conversion in the
experiments performed using the same catalyst weight (Fig. 5a) is A > SA87 > SA30 >
SA5 >> S. -Al O , sample A, whose behavior has been discussed previously [25] and that
2
3
is considered here as a reference catalyst, gives rise to the highest conversion at T  473
K. Pure SiO , sample S, has weak activity only at T  573 K. In the case of silica-
2
aluminas, the activity trend is closely related to their surface area, the higher the surface
area, the higher the catalytic activity. Looking at the product selectivities, all catalysts
produce mainly DEE at low temperature and low conversion and ethylene at high
temperature and high conversion (Table 2). Only SiO2 produces also significant
selectivities to acetaldehyde ( 10%) at  623 K.
In the case of the experiments performed with the same catalyst surface area (Fig. 5b), the
conversion on γ-Al O is again definitely more than on all silica-aluminas, where similar
2
3
conversions are found, the small differences being near the experimental error. However,
the conversion trend is systematically SA5 > SA30 > SA87 at 573 K. Thus the catalytic
activity of silica-aluminas seem to slightly decrease in parallel with increasing SiO /Al O
3
2
2
ratio in agreement with the data of Caillot el al. [24].
More important differences are found for product yield and selectivity. As evident in the
lower sections of Fig. 5, the yields (and selectivities too) to ethylene markedly increase at
similar conversion and temperature by increasing SiO /Al O ratio, the reverse occurring to
2
2
3
DEE yields and selectivities. Selectivities to other products are very small ( 0.3 %) except
for SA5 where, at high temperature and conversion, they are  4.5 %.
5

ACCEPTED MANUSCRIPT
It must, in any case, be confirmed that the catalytic activity of silica-aluminas is lower than
that of zeolites with comparable SiO -Al O compositions [27]. In particular, experiments
2
2
3
performed in the same conditions show that several zeolites give rise to detectable
conversion of ethanol (mainly to DEE) already at 393 K where all silica-aluminas are still
fully inactive.
3
Discussion
The data reported here confirm that silica-aluminas are active catalysts in converting
ethanol to DEE and ethylene. However, they are less active than both zeolites and -
alumina. Surface characterization confirms that silica-aluminas present strong Lewis
acidity and also sufficient Brønsted acidity to protonate in part pyridine. IR spectra confirm
that terminal silanol groups are the largely predominant surface hydroxyl groups on silica-
aluminas. A detailed analysis of the spectra reveal some heterogeneity of these OH
groups and do not produce any evidence of zeolitic hydroxyl groups [34]. It is consequently
confirmed that the Brønsted acidity of silica-aluminas is due to very acidic terminal
silanols.
The behavior, both from the point of view of the surface chemistry and from the point of
view of the catalytic activity, suggest that silica-rich fully amorphous materials like typical
co-precipitated or co-gelled cracking catalysts, and -Al O containing sample such as
2
3
Siralox 30/260 have close similarities, in agreement with Caillot et al. [24].
Concerning the ethanol dehydration reaction, the lower activity of silica-alumina with
respect to protonic zeolites, the similar behavior with respect to aluminas and the
conversion trend (more catalytic activity in absence of Brønsted acidity as for SA5)
suggest that Lewis acidity is predominant in the catalysis of silica-aluminas in the reaction.
The significant difference found in the DEE vs ethylene selectivities at closely similar
conversion and temperature conditions allows some mechanistic insight. In our previous
papers we emphasized the parallel behavior of all different catalysts: in all cases DEE is
found very predominant at low temperature and conversion, while ethylene is predominant
at high temperature and conversion. Surface ethoxy groups were found to be the active
surface intermediates for both reactions [31]
C H OH  C H + H O
(1)
(2)
2
5
2
4
2
2
C H OH  C H OC H + H O
2
5
2
5
2
5
2
6

ACCEPTED MANUSCRIPT
Properly designed experiments [31] showed that the behavior is due to at least two factors:
i) the two reactions are parallel and reaction (2) is characterized by lower activation energy
and higher reaction order than reaction (1); ii) however, there is also a third reaction, the
cracking of DEE to ethylene, showing that DEE can also act as intermediate in ethylene
production.
C H OC H  2 C H + H O
(3)
2
5
2
5
2
4
2
The experiments performed with different zeolites [27] suggested also that shape
selectivity effects may also play a role, e.g. favouring ethylene formation also at low
conversion when occurring in small pores: e.g. in the side pockets of mordenite and in
ferrierite, where the addition of two molecules can be sterically hindered.
Here we show that the nature of different surface sites may also play a role. The
increasing ethylene selectivity and decreasing DEE selectivity at the same temperature
and conversion by increasing the silica content may be interpreted suggesting that the
more active sites for DEE production may be different from the more active sites for
ethanol ethylene production.
Our data [31] suggest that the main way to ethylene is essentially the elimination of an
ethoxy group. Instead, the synthesis of DEE implies the reaction of ethoxy group with
some form of undissociated ethanol. It can be supposed that activation of the second
ethanol molecule on nearest acido-basic sites can be beneficial for DEE formation. This
can more easily occur on silica-aluminas with lower SiO /Al O ratio where the Lewis
2
2
3
active sites should be more frequently surrounded by extended alumina-like surface. This
is depicted in scheme 1. The activation of the second ethanol molecule could occur less
efficiently in the predominantly covalent surface of silica-rich catalysts. This would explain
why DEE selectivity at low conversion follows the trend (at similar low conversion): -Al O3
2
(
A)> SA5 > SA30 > SA87 > SiO (S).
2
4
Conclusions
The conclusions of this paper are the following:
1
2
. Amorphous silica-alumina and silica-alumina prepared by deposition of silicic acid
on boehmite precursor (Siralox) have similar surface structures and behavior.
. In all cases the materials present strong Lewis acidity together with Brønsted sites
able to protonate pyridine.
7

ACCEPTED MANUSCRIPT
3
. Both these materials do not show any evidence of “zeolitic” bridging OH’s, while
they do show significant heterogeneity of terminal silanol groups, likely
“pseudobridging”.
4
5
. These materials show high activity in ethanol conversion to ethylene and diethyl
ether.
. The diethyl ether/ethylene selectivity depends on the nature of the active site,
diethyl ether formation being more favoured for lower SiO /Al O ratios where Lewis
2
2
3
sites are more frequently surrounded by alumina-like surface.
6
. The diethyl ether synthesis from ethanol occurs by reaction of ethoxy groups with
an activated undissociated ethanol molecule.
Acknowledgements
TKP acknowledges funding by EMMA (Erasmus Mundus Mobility with Asia) in the
framework of the EU Erasmus Mundus Action 2.
8

ACCEPTED MANUSCRIPT
Table 1. The properties of investigated catalysts
Notation
XRD data
Commercial
name
Manufacturer
SiO wt% Al O wt%
Preparation
SBET
2
2
3
2
(m /g)
A
-Al O
Puralox SBa200
Sasol
0
100
From boehmite
produced via Al
alkoxides
190
2
3
SA5
,-Al O
Siralox 5/170
Siralox 30/260
Sasol
Sasol
5
95
70
From silicic-acid
impregnated boehmite
produced via Al
alkoxides
170
260
2
3
SA30
-Al O , amorphous
30
2
3
SA87
S
amorphous
amorphous
Cracking catalyst Strem Chemicals
Silica Gel SG127 Grace Davison
87
13
0
Co-gel
 330
 300
100
Gel
9

ACCEPTED MANUSCRIPT
Table 2. Ethanol conversion and product selectivities (S) upon ethanol dehydration
-1
(
experiments were performed with the same catalyst weight of 0.5 g, WHSV = 1.43 h )
Catalyst Temp. (K)
Conversion (%) SEthylene (%) SDEE (%) SOthers (%)
423
473
523
0.2
20.8
78.6
97.7
99.7
99.9
100.0
0.0
0.0
0.9
100.0
99.1
79.7
0.3
0.0
0.0
0.1
0.0
0.0
0.1
1.0
0.0
0.0
0.0
1.4
4.5
3.9
3.7
0.0
0.0
0.1
0.2
0.2
0.3
0.3
0.0
0.0
0.0
0.1
0.0
0.0
0.1
0.0
0.0
0.0
0.0
9.0
19.4
19.0
20.2
99.7
100.0
99.8
98.9
0.0
A
573
623
673
723
423
473
523
0.0
0.0
0.1
0.0
2.6
0.0
100.0
97.3
74.3
0.1
35.6
74.1
99.7
100.0
100.0
1.0
2.7
SA5
SA30
SA87
S
573
24.3
95.4
96.1
96.3
0.0
623
673
723
423
473
523
0.0
0.0
100.0
93.4
84.1
76.0
0.0
8.3
6.6
60.2
84.5
100.0
100.0
100.0
0.8
15.8
23.8
99.8
99.7
99.6
0.0
573
623
673
723
423
473
523
0.0
0.1
100.0
86.3
64.9
0.3
15.7
67.5
96.0
99.8
100.0
100.0
0.0
13.7
35.1
99.6
100.0
100.0
99.9
0.0
573
623
673
723
423
473
523
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.2
0.0
100.0
75.4
55.5
31.8
8.5
573
0.6
24.6
35.5
48.8
72.5
623
673
723
2.5
8.6
15.0
10

ACCEPTED MANUSCRIPT
Table 3. Ethanol conversion and product selectivities (S) upon ethanol dehydration
(experiments were performed with the same catalyst surface area)
Catalyst
Temp. (K)
Conversion (%) SEthylene (%) SDEE (%) SOthers (%)
423
473
523
0.0
0.0
0.0
99.1
91.4
5.5
0.0
0.0
0.0
0.4
1.1
0.6
0.8
0.0
0.0
0.0
0.3
1.0
3.5
3.1
0.0
0.0
0.2
0.1
0.3
0.4
0.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
9.5
0.9
61.5
91.5
99.2
100.0
99.7
0.0
8.8
A
-1
573
94.1
98.9
99.4
99.2
0.0
2
.97 h WHSV
623
673
723
423
473
523
0.1
0.0
0.0
0.0
1.3
0.0
100.0
97.8
81.1
0.6
31.0
80.2
98.3
100.0
100.0
0.2
2.2
SA5
-1
573
18.6
98.4
96.5
96.8
0.0
2
.7 h WHSV
623
673
723
423
473
523
0.0
0.1
100.0
92.0
85.7
58.0
0.1
2.8
8.0
32.6
74.8
100.0
100.0
100.0
0.3
14.2
41.8
99.6
99.6
99.3
0.0
SA30
-1
573
4
.0 h WHSV
623
673
723
423
473
523
0.0
0.0
100.0
87.6
69.5
32.5
0.4
3.7
12.4
30.5
67.5
99.6
100.0
100.0
26.0
73.8
99.0
100.0
100.0
SA87
-1
573
4
.95 h WHSV
623
673
723
0.0
0.0
11

ACCEPTED MANUSCRIPT
H C H C
H C CH
2 2
3
2
O
O
HO
(
H CCH ) O
3
2 2
H C H C
3
2
CH CH
2
3
HO
O
H
Scheme 1.
12

ACCEPTED MANUSCRIPT
Figure Captions
Fig. 1. FT-IR skeletal spectra (KBr pressed disks) of investigated catalysts.
Fig. 2. FT-IR spectra of investigated catalysts after activated at 773 K (OH region).
Fig. 3. FT-IR subtraction spectra of surface species arising from pyridine adsorbed on
investigated catalysts at 373 K.
Fig. 4. FT-IR spectra of the OH stretching region (A) and of adsorbed pyridine (C) on
SA30. (B) spectra in the OH stretching region after subtraction of the spectrum of the
activated sample. (a) sample activated at 500 °C in vacuum, (b) in contact with pyridine,
later under outgassing at 300 K, 5 min (c), 300 K, 30 min, 5 min (d), 323 K, 5 min (e), 373
K, 5 min (f), 423 K, 5 min (g), 473 K, 5 min (h), 523 K, 5 min (i), 573 K, 5 min (l), 623 K,
5
min (m), 673 K, 5 min (n), 723 K, 5 min (o), 773 K, 5 min (p).
Fig. 5. Conversion of ethanol over investigated catalysts as a function of temperature: (a)
-1
performed using the same catalyst weight of 0.5 g, WHSV = 1.43 h and (b) same catalyst
surface area; (c) ethylene yield and (d) diethyl ether yield as a function of ethanol
conversion in the case of experiments performed using same catalyst surface area.
13

ACCEPTED MANUSCRIPT
14

ACCEPTED MANUSCRIPT
15

ACCEPTED MANUSCRIPT
16

ACCEPTED MANUSCRIPT
17

ACCEPTED MANUSCRIPT
18

ACCEPTED MANUSCRIPT
Highlights





Differently prepared commercial silica-aluminas have similar surface sites
Brønsted sites are pseudobridging terminal silanols.
Ethanol conversion on slica-aluminas occurs on Lewis acid sites
Lewis sites with alumina-like acidobasic neighbor are more selective for ethylene
Lewis sites with silica-like covalent neighbor are more selective for diethyl ether
19

ACCEPTED MANUSCRIPT
References
[
[
[
[
[
[
[
[
1]
2]
3]
4]
5]
6]
7]
8]
T.J. Gray, J. Phys. Chem., 61 (1957) 1341–1343.
S. Peratello, M. Molinari, G. Bellussi, C.Perego, Catal. Today 52 (1999) 271- 277.
D. P. Serrano, J. Aguado , J. M. Escola , ACS Catal., 2 (2012) 1924–1941.
A. Finiels, F.Fajula, V. Hulea, Catal. Sci. Technol., 4 (2014) 2412–2426.
W. H. Shearon, W. R. Fullem, Ind. Eng. Chem., 51 (1959) 720–726.
A. de Klerk, Energy & Fuels 21 (2007) 625-632.
C. Perego, R. Millini, Chem. Soc. Rev., 42 (2013) 3956-3976.
M. Bevilacqua, T. Montanari, E. Finocchio and G. Busca, Catalysis Today, 116
(
2006) 132-142.
M. Caillot, A. Chaumonnot, M. Digne, J.A. Van Bokhoven, ChemCatChem 6 (2014)
32-841.
[
[
9]
8
10] W. Daniell, U. Schubert, R. Glöckler, A. Meyer, K. Noweck, H. Knözinger, Appl.
Catal. A Gen. 196 (2000) 247-260.
[
[
11] V.C.F. Holm, G.C. Bailey, A. Clark, J. Phys. Chem., 63 (1959) 129–133.
12] G.E. Maciel, J.F. Haw, I.S. Chuang, B.L. Hawkins, T.A. Early, D.R. McKay, L.
Petrakis, J. Am. Chem. Soc., 105 (1983) 5529–5535.
[
[
[
13] S. Sato, M. Toita, T. Sodesawa, F. Nozaki, Appl. Catal., 62 (1990) 73–84.
14] C. Pereira, R.J. Gorte, Appl. Catal. A Gen., 90 (1992) 145–157.
15] L. Heeribout, R. Vincent, P. Batamack, C. Dorémieux-Morin, J. Fraissard, Catal.
Lett., 53 (1998) 23-31.
[
[
[
16] K. Góra-Marek, J. Datka, Appl. Catal. A Gen., 302 (2006) 104-109.
17] O. Cairon, Phys. Chem. Chem. Phys., 12 (2010) 6333-6336.
18] E.J.M. Hensen, D.G. Poduval, V. Degirmenci, D.A.J.M. Ligthart, W. Chen, F. Maugé
M.S. Rigutto, J.A.R. van Veen, J. Phys. Chem. C 116 (2012) 21416−21429.
19] M. Trombetta, G. Busca, S. Rossini, V. Piccoli, U. Cornaro, A. Guercio, R..; Catani,
R.J. Willey, J. Catal. 179 (1998) 581-596.
[
[
[
20] Z. Luan, J.A. Fournier, Micropor. Mesopor. Mat. 79 (2005) 235-240.
21] G. Crépau, V. Montouillout, A. Vimont, L.Mariey, T. Cseri , F. Maugé, J. Phys.
Chem. B, 110 (2006) 15172-15185.
[
22] F. Leydier, C. Chizallet, A. Chaumonnot, M. Digne, E. Soyer, A.-A. Quoineaud, D.
Costa, P. Raybaud, J. Catal. 284 (2011) 215-229.
20

ACCEPTED MANUSCRIPT
[
[
23] C. Chizallet, P. Raybaud, Catal. Sci. Technol., 4 (2014) 2797–2813.
24] M. Caillot, A. Chaumonnot, M. Digne, J.A. Van Bokhoven, J. Catal. 316 (2014) 47-
5
6.
[
[
[
25] T.K. Phung, A. Lagazzo, M.Á. Rivero Crespo, V. Sanchez Escribano, G. Busca, J.
Catal. 311 (2014) 102-113.
26] T.K. Phung, L. Proietti Hernández, G. Busca, Appl. Catal. A Gen. 489 (2015) 180-
1
87.
27] T.K. Phung, L. Proietti Hernández, A. Lagazzo, G. Busca, Appl. Catal. A Gen. 493
2015) 77-89.
(
[
[
[
28] J. McKetta, Chemical Processing Handbook, 1993, Wiley, p. 786-789.
29] M.A. Christiansen, G. Mpourmpakis, D.G. Vlachos, ACS Catal. 3 (2013) 1965-1975.
30] J.F. DeWilde, H. Chiang, D. A. Hickman, C.R. Ho, and A. Bhan, ACS Catal. 3 (4)
(2013) 798–807.
[
[
[
31] T. K. Phung, G. Busca, Chem. Eng. J., 272 (2015) 92-101.
32] G. Busca, Advan. Catal., 57 (2014) 319-404.
33] G. Busca, in: S.D. Jackson and J.S.J. Hargreaves (Eds.), Metal Oxide Catalysis,
Wiley-VCH, Vol. 1, 2009, pp. 95.
[
[
34] K.H. Hadjiivanov, Advan. Catal., 57 (2014) 99-318.
35] G. Busca, Catal. Today 41 (1998) 191-206.
21