Glycerol valorization by etherification to polyglycerols by using metal oxides derived from MgFe hydrotalcites

Article abstract of DOI:10.1016/j.apcata.2013.10.051

This work investigates the use of MgFe mixed oxides, derived from layered double hydroxides (LDH) with Mg/Fe molar ratio ranging from 1 to 4, as base catalysts for the etherification of glycerol. LDH precursors and catalysts were characterized by XRD, XPS, CO₂-TPD, NH₃-TPD, N₂ adsorption and DTA-TG analysis. The MgFe mixed oxides exhibit excellent textural properties, with specific surface areas close to 200 m² g^-1 and average pore diameters in the mesoporous range. This family of catalysts has shown to be active in the formation of polyglycerols from glycerol without solvent, at 220 C, in a batch reactor. The highest conversion (41%) is found for the MgFeO4 catalyst prepared with a Mg/Fe molar ratio of 4, whereas full selectivity to diglycerols is only reached for the MgFeO1 catalyst. Only diglycerols (DGs) and triglycerols (TGs) have been detected after 24 h of reaction.

Full text of DOI:10.1016/j.apcata.2013.10.051

Applied Catalysis A: General 470 (2014) 199–207
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Glycerol valorization by etherification to polyglycerols by using metal
oxides derived from MgFe hydrotalcites
∗
P. Guerrero-Urbaneja, C. García-Sancho, R. Moreno-Tost , J. Mérida-Robles,
J. Santamaría-González, A. Jiménez-López, P. Maireles-Torres
Departamento de Química Inorgánica, Mineralogía y Cristalografía (Unidad Asociada al ICP-CSIC), Facultad de Ciencias, Universidad de Málaga,
Campus de Teatinos s/n., 29071 Málaga, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 July 2013
Received in revised form 24 October 2013
Accepted 25 October 2013
Available online 5 November 2013
This work investigates the use of MgFe mixed oxides, derived from layered double hydroxides (LDH)
with Mg/Fe molar ratio ranging from 1 to 4, as base catalysts for the etherification of glycerol. LDH
precursors and catalysts were characterized by XRD, XPS, CO2-TPD, NH3-TPD, N2 adsorption and DTA–TG
analysis. The MgFe mixed oxides exhibit excellent textural properties, with specific surface areas close
2
−1
to 200 m g and average pore diameters in the mesoporous range. This family of catalysts has shown
◦
to be active in the formation of polyglycerols from glycerol without solvent, at 220 C, in a batch reactor.
Keywords:
Layered double hydroxide
Iron
Glycerol
Etherification
Polyglycerol
The highest conversion (41%) is found for the MgFeO4 catalyst prepared with a Mg/Fe molar ratio of 4,
whereas full selectivity to diglycerols is only reached for the MgFeO1 catalyst. Only diglycerols (DGs) and
triglycerols (TGs) have been detected after 24 h of reaction.
© 2013 Elsevier B.V. All rights reserved.
1
. Introduction
Biodiesel and bioethanol are the two most important biofuels
higher polyglycerols. This etherification process is carried out in
the presence of basic homogeneous catalysts, mainly alkaline car-
bonates or hydroxides. Thus, Ayoub et al. [7] have demonstrated
that LiOH exhibits an excellent catalytic performance, reaching full
glycerol conversion, with a selectivity of 33% towards diglycerol,
after 6 h of reaction. Clacens et al. [8–10] have shown, that concern-
ing the selectivity, the best results were obtained over mesoporous
solids modified by exchanged or impregnation with Cs. On the other
hand, Ruppert et al. [11] studied the use of alkaline earth metal
successfully supplied in the transportation sector. In Europe, the
production of biodiesel is increased from 4.1 MTm in 2006 up to
9
.6 MTm in 2010 [1]. Biodiesel is a mixture of long-chain alkyl
esters produced by transesterification of vegetable oils with short-
chain alcohols, such as methanol or ethanol. Glycerol is generated
as by-product in this transesterification process (roughly 10 wt%),
giving rise to a stock of glycerol which requires to be efficiently val-
orized in order to obtain a more sustainable biodiesel production.
In this sense, glycerol has been turned into an interesting starting
raw material for other chemical products. In recent years, several
reviews [2–4], dealing with the potential uses of glycerol as starting
compound for many industrial applications, have been published.
An interesting valorization alternative is the glycerol oligomerisa-
tion [5,6], because these oligomers are already used in cosmetics
and, as additives, in nutrition or lubricants. The condensation of
two glycerol molecules yields diglycerol, the simplest polyglycerol,
which can be linear, branched or cyclic depending on if the con-
densation takes place between primary or secondary hydroxyls, or
an intramolecular condensation is involved [5]. On the other hand,
the condensation of glycerol can progress yielding tri-, tetra- and
◦
oxides at lower temperature (220 C) and compared the catalytic
results with homogeneous sodium carbonate. Thus, they found
that glycerol conversion enhanced with increasing catalyst basic-
ity: MgO < CaO < SrO < BaO. The best selectivity values for (di- + tri-)
glycerol (>90% at 60% conversion) were obtained over CaO, SrO,
and BaO, with no substantial acrolein formation. However, alkaline
earth oxides suffer from instability toward ambient water and car-
bon dioxide, giving rise to less basic species, thus requiring very
high activation temperatures. Gholami et al. [12], evaluating the
activity of Ca:Al:La composites, concluded that the most suitable
La:Ca molar ratio was 1:2.7, reaching 91% glycerol conversion and
a selectivity towards diglycerol of 53%. Recently, we reported the
catalytic performance of MgAl mixed oxides derived from the cor-
responding hydrotalcite in the etherification of glycerol [13], being
the most active catalyst the mixed oxide derived from the hydrotal-
cite prepared by the coprecipitation method using Na CO /NaOH
2
3
∗ Corresponding author. Tel.: +34 952132021; fax: +34 952131870.
E-mail address: rmtost@uma.es (R. Moreno-Tost).
as precipitant agent. This catalyst yielded a maximum conversion
of 51% and a diglycerol selectivity of 85%. Razealy-Anuar et al.
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.10.051

2
00
P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
[
14] described the activity of MgAl hydrotalcite synthesized by the
2.2. Catalyst characterization
combustion method. This catalyst reached the highest glycerol con-
version in 16 h (77.7% glycerol conversion) and when the glucose
was used as fuel and calcined at 650 C.
Among basic solids, mixed metal oxides derived from layered
double hydroxides (LDHs) seem to be very promising for practi-
cal applications due to the easiness of synthesis and handling [15].
Elemental analysis was performed on a PERKIN–ELMER 2400
CHN with a LECO VTF900 pyrolysis oven. Mg, Fe and Na contents
have been determined by ICP-AES by using a Perkin Elmer (model
ELAN DRC-e) spectrometer. Powder XRD patterns were obtained
by using a Siemens D5000 automated diffractometer, over a 2ꢀ
range with Bragg–Brentano geometry using the Cu K␣ radiation
and a graphite monochromator. X-ray photoelectron spectroscopy
(XPS) studies were performed with a Physical Electronics PHI
5700 spectrometer equipped with a hemispherical electron ana-
lyzer (model 80-365B) and a Mg K␣ (1253.6 eV) X-ray source.
◦
2
+
3+
These solids are represented by the general formula [M
M
x
1−x
n
2+
3+
(
OH) ]A
·mH O [16], where M and M are di– and trivalent
2
x/n
2
2+ 2+ 2+ 2+ 2+
metal cations, including cations such as Mg , Fe , Ni , Cu , Co
,
2+
2+
2+
3+
3+
3+
3+
n−
Mn , Zn or Cd and Al , Cr , Ga or Fe , respectively. A is
an interlayer anion and x is the [M3+]/([M2+]+[M ]) molar fraction.
3+
2
+
◦
Their structure consists of brucite-type layers where M cations
are partially replaced with M3 cations, inducing a net positive
charge, which is compensated by anions situated in the inter-
High-resolution spectra were recorded at 45 take-off-angle by a
+
concentric hemispherical analyzer operating in the constant pass
energy mode at 29.35 eV, using a 720 mm diameter analysis area.
Charge referencing was done against adventitious carbon (C 1s at
284.8 eV). The pressure in the analysis chamber was kept lower
layer region (e.g., CO32−, SO4 , NO3 , C1 , OH ) together with
water molecules. An important feature of this family of solids is
that their basicity, interlayer distance and crystallites morphol-
ogy can be tailored by adjusting the nature of the cations, the
metal ratio in the brucite-like layers and the interlayer anionic
species [16]. Both the basic properties of hydrotalcite-like materials
and the existence of well-dispersed mixed oxides upon ther-
mal decomposition are key aspects for successful application in
catalysis.
2−
−
−
−
−
6
than 5 × 10 Pa. PHI ACCESS ESCA-V6.0 F software package was
used for data acquisition and analysis. A Shirley-type background
was subtracted from the signals. Recorded spectra were always
fitted using Gauss–Lorentz curves in order to determine more accu-
rately the binding energy of the different element core levels.
◦
N2 adsorption–desorption isotherms at −196 C were obtained
using an ASAP 2020 model of gas adsorption analyzer from
Micromeritics, Inc. Prior to N2 adsorption, the sample were evac-
Hydrotalcites, wherein Fe3+ ion partially substitutes Mg2+, are
among the most studied, as catalyst precursors. MgFe mixed oxides
generated after thermal treatment, usually at 450 C, have demon-
◦
−1
uated at 450 C (heating rate 10 K min ) for 18 h. Pore size
distribution and pore volume were calculated with the BJH method.
Thermogravimetric and differential thermal analyses (TG–DTA)
were performed on a Pyris–Diamond PerkinElmer apparatus. The
◦
strated to be excellent sorbent for phosphate [17] or perchlorate
ions [18]. Moreover, these solids show acid–base properties [19],
participating in dehydrogenation reactions over basic sites and
dehydration reactions over acidic ones [18,20]. Thus, these mixed
oxides have found applications in dehydrogenation of ethylben-
zene to styrene [21,22], alkylation of m-cresol with methanol [23]
and Friedel–Crafts alkylations [24].
In the present paper, the catalytic activity of MgFe mixed
oxides derived from hydrotalcite precursors in the etherifica-
tion of glycerol to diglycerol is reported. It has been studied
the influence of the Mg/Fe molar ratio, the catalytic kinetic, the
pretreatment of the catalysts and reusability of those catalysts,
proving that these catalysts are highly active and selective in such
reaction.
◦
temperature was varied from room temperature up to 1000 C, at a
heating rate of 10 K min- with a flux of nitrogen of 100 mL min
−1
−1
using a mass around 15 mg.
The basicity was studied by temperature-programmed desorp-
tion of CO . Samples (100 mg) were pretreated under a helium
2
◦
−1
◦
−1
stream at 450 C for 1 h (2 K min , 100 mL min ). Then, tem-
perature was decreased until 100 C, and a flow of pure CO
2
−
1
(50 mL min ) was subsequently introduced into the reactor dur-
◦
ing 1 h. The CO -TPD was carried out between 100 and 800 C under
2
−
1
−1
a helium flow (10 K min , 30 mL min ), and evolved CO2 was
analyzed by an on-line gas chromatograph (Shimadzu GC-14A) pro-
vided with a TCD, after passing by an ice-NaCl trap to eliminate any
trace of water. Temperature-programmed desorption of ammonia
(
NH -TPD) was carried out to evaluate the total acidity of the cat-
3
2
. Experimental
alysts. After cleaning the materials with helium and adsorption of
◦
ammonia at 100 C, the NH -TPD was performed between 100 and
3
◦
−1
2.1. Catalyst synthesis
550 C with a heating rate of 10 K min by using a helium flow and
maintained at 550 C for 15 min. The evolved ammonia was ana-
◦
The method of coprecipitation at constant pH [25] has been
lyzed by on-line gas chromatography (Shimadzu GC-14A) provided
with a TCD detector.
chosen to synthesize a series of Mg/Fe hydrotalcites. The Mg/Fe
molar ratio was varied between 1 and 4, and the concentration
of Mg + Fe was fixed at 1 M. The pH was maintained at a value of
2.3. Catalytic reaction
1
0 by using a precipitant solution of NaOH/Na CO with a con-
2 3
−
2−
centration of OH and CO3 of 2 and 0.125 M, respectively. Both
solutions were dropped wisely in a flask containing 100 mL of dis-
tilled water, under vigorous stirring. The synthesis was carried out
at room temperature and the precursor solution was aged during
The catalytic activity was evaluated in the etherification of
◦
glycerol (Aldrich) at 220 C in a three-necked glass batch reac-
tor without solvent, equipped with a water-cooled condenser
coupled with a Dean-Stark system to remove the formed water,
thermometer and vigorous stirring. The atmosphere of the reac-
2
4 h. The resulting solids were filtered, deeply rinsed with distilled
◦
−1
water and dried at 65 C. The solids were labelled as MgFex, where x
tor was kept inert by means of a N2 flow of 15 mL min . Before
stands for the Mg/Fe molar ratio. Prior the catalytic test, the hydro-
talcites were activated in a tubular furnace at 450 C during 15 h,
the reaction, catalysts were activated at different temperatures
for 15 h (heating rate, 2 K min ), under a helium flow. The reac-
◦
−1
under a helium flow, to transform them in the corresponding MgFe
mixed oxides (MgFeOx). Pure Mg(OH)2 was synthesized by using
this same methodology. Na CO (Merck) was also used, as received,
tion was stopped at 24 h and catalysts were separated by filtration
on a porous plate. Then, catalysts were washed deeply with
hot water and ethanol. The reaction products were analyzed by
means of gas chromatography. An aliquot (ca 80 mg) of the reac-
tion mixture was dissolved in dried pyridine (Aldrich) and then
N,O–bis(trimethylsilyl)trifluoroacetamide was added. This solution
2
3
for comparison, since it is the industrial catalyst. MgO catalyst was
◦
prepared from thermal treatment of Mg(OH)2 at 450 C under a
helium flow.

P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
201
Table 1
Chemical analysis and crystallographic parameters of as-synthesized MgFe hydrotalcites.
Sample
Mga (wt%)
Fea (wt%)
Mg/Fea (molar ratio)
Cb (wt%)
Nb (wt%)
c (Å)
a (Å)
d(003) (Å)
Crystallite size (c) (nm)
Crystallite size (a) (nm)
MgFe1
MgFe2
MgFe3
MgFe4
10.4
14.7
19.5
20.9
22.6
20.1
15.7
13.7
1.1
1.7
2.9
3.6
1.7
2.0
1.8
1.7
0.0
0.2
0.7
0.9
23.2
23.3
23.4
24.1
3.1
3.1
3.1
3.1
7.7
7.8
7.8
7.9
12.5
11.9
10.6
12.8
29.5
25.4
19.9
24.7
a
Calculated from AA analysis.
Calculated from CNH analysis.
b
Mg/Fe molar ratio, with sharp and intense diffraction peaks due to
basal planes, and broad and asymmetric peaks ascribed to non-
◦
◦
basal planes (Fig. 1). Thus, the sharp peaks at 2ꢀ ≈ 11 , 23 and
◦
3
4 correspond to the (0 0 3), (0 0 6) and (0 0 9) planes, respectively,
indicating a well-formed crystalline layered structure with rhom-
bohedral symmetry. No other crystalline phases, such as Fe(OH)3
MgFe4
or Mg(OH) , were detected in the XRD patterns. On the other
2
hand, the crystallinity increases with the Mg content, as can be
inferred from the intensification of the diffraction peak correspond-
ing to the (0 0 l) planes. The basal spacing (d003 reflection) values
have confirmed the presence of carbonate anions in the inter-
layer region [26,27]. The MgFe4 hydrotalcite shows a higher basal
spacing among the synthesized samples, and higher than those
reported for carbonate anions as compensating anions in the inter-
layer region [28]. This fact could be ascribed to the concomitant
presence of nitrate anions, as could point out the highest nitrogen
content found by CHN analysis (Table 1), because it has been pre-
viously reported that the basal spacing is higher in the presence of
nitrate anions [17].
MgFe3
MgFe2
MgFe1
5
10 15 20 25 30 35 40 45 50 55 60 65 70
θ (º)
2
The corresponding lattice parameters and d003 basal spacing are
presented in Table 1, together with the average crystallite sizes,
obtained from the full width a half maximum (FWHM) of diffrac-
tion peaks corresponding to (0 0 3) and (1 1 0) planes, by using the
Scherrer equation. The unit cell parameter, a, is the average dis-
tance between two metal ions within the layers, and it has been
calculated from the d-spacing of the (110) reflection (a = 2d110).
The c parameter is three times the distance between adjacent
hydroxide and it has been also calculated from the (0 0 3) reflection
Fig. 1. XRD patterns of as-synthesized MgFex hydrotalcites.
◦
was aged in a stove at 60 C during 1 h and analyzed in a gas
chromatograph (Shimadzu GC model 14A) equipped with FID and
a capillary silica fused TBR-14 column (Tecknochroma).The only
detected products were unreacted glycerol, di- and tri-glycerol. The
selectivities to the different products were calculated as the weight
ratio of the respective product to the sum of products formed. The
glycerol conversion was calculated as the ratio of the detected glyc-
erol to the sum of the glycerol and products formed.
(
c = 3d003). The a parameter barely changes with the decrease of the
3+
2+
Mg/Fe molar ratio, since the ionic radius of Fe and Mg cations
are very similar: 0.69 and 0.65 A˚ , respectively [23,29]. In return,
3
. Results and discussion
the c parameter shows a significant reduction as the Mg/Fe molar
ratio decreases, which is consistent with the increase in Coulombic
attractive forces between the negatively charged interlayer anions
and the positively charged brucite-like layers with the increas-
ing of Fe3+ cations. Therefore, the layers are packed closer with
lower Mg/Fe molar ratio [30,31]. The crystallite sizes are longer in
3.1. Characterization of hydrotalcite precursors and catalysts
The XRD patterns of MgFex hydrotalcites exhibit the diffraction
pattern characteristic of the hydrotalcite structure, irrespective of
MgFeO1
MgFeO1
MgFeO2
MgFeO2
MgFeO3
MgFeO3
MgFeO4
MgFeO4
20
25 30 35 40 45 50 55 60 65 70 75 80
θ (º)
20
25
30
35
40
45
θ (º)
50
55
60
65
70
2
2
◦
Fig. 2. (a) XRD patterns of MgFeOx mixed oxides derived from MgFex hydrotalcites calcined at 450 C and (b) XRD patterns of MgFeOx mixed oxides derived from MgFex
◦
hydrotalcites calcined at at 800 C.

2
02
P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
100
0.4
MgFeO4
9
9
8
8
7
7
6
6
5
5
5
0
5
0
5
0
5
0
5
0
0
.2
.0
0
MgFeO3
MgFeO2
-0.2
-0.4
-0.6
-0.8
-1.0
MgFeO1
800
-1.2
-1.4
100
200
300
400
500
600
700
800
100
200
300
400
500
600
700
Temperature (ºC)
Temperature (ºC)
Fig. 3. DTA–TG curves of the as-synthesized MgFe3 hydrotalcite.
MgFeO4
a direction, which it is expected from the well-known platelet-like
morphology of hydrotalcite crystallites [32].
MgFeO3
MgFeO2
The as-synthesized samples were calcined under a helium flow
◦
at 450 C in order to transform them into the respective MgFeOx
mixed oxides (Fig. 2a). This thermal treatment leads to the struc-
tural collapse with remarkable structural changes. All of them
exhibit XRD patterns quite similar to that of MgO phase (periclase),
MgFeO1
◦
◦
but with broad peaks (2ꢀ ≈ 43 , 63 ) indicating poor crystallinity.
On the other hand, the sample with Mg/Fe molar ratio of 1 shows
◦
◦
◦
other broad peaks at ca. 2ꢀ ≈ 30 , 35 and 57 , which can be ascribed
MgO
700 800
to the spinel MgFe O phase [23,24], although the presence of
2
4
Fe O cannot be ruled out since these two phases show a simi-
0
100
200
300
400
500
600
900
3
4
lar XRD patterns [23], and thermal treatment in inert atmosphere
Temperature (ºC)
3+
could result in a partial reduction of Fe . When the temperature
◦
of thermal treatment is raised to 800 C, all the samples exhibit the
Fig. 4. (a) TPD–NH3 curves of the MgFeOx catalysts and (b) TPD–CO2 curves of the
MgFeOx catalysts.
diffraction peaks of the spinel phase, irrespectively of the Fe content
(
Fig. 2b).
Thermogravimetric (TG) and Differential Thermal Analysis
DTA) curves are very similar, and Fig. 3 displays, as an example,
consisting of aggregates of particles forming slit-shaped pores
with non uniform size or shape [39].
(
those corresponding to the MgFe3 hydrotalcite. These results are
in good agreement with those from the literature [18–21,24,25].
The differences are only observed in the intensity of the three steps
of weight loss and the corresponding endothermic effects. The first
stage accounts for the removal of the water loosely bounded in
the interlayer region, with no significant differences among the
studied hydrotalcites, either the peak temperatures or weight loss.
The second weight loss is associated to dehydroxylation of the
brucite layers and decomposition of the interlayer anions with the
concomitant structural collapse, as evidenced the XRD patterns,
leading to the formation of metal oxides. The third one corresponds
to ca. 5 wt% of weight loss. By means of evolved gaseous analy-
sis by mass spectroscopy (Appendices A and B, Fig. S1), neither
mass 44 nor 18 were detected, therefore the thermal processes
involved in the decarbonation and dehydroxylation have finished
The pore size distributions of the MgFeOx samples are relatively
broad, extending from 2 to 70 nm with a maximum at around 15 nm
(Supplementary Information, Fig. S2). They present a bimodal pore
size distribution. Thus, the calcined hydrotalcites show pores in
the range of 2–3.5 nm, which formation is favoured by the evolved
gaseous products generated during the thermal decomposition of
the interlayer anions. Such mechanism of decomposition is known
as “cratering” [40] and it is responsible of the intraporosity of par-
ticles. On the other hand, a second pore size bands ranges from 4
to 70 nm, and these pores are formed as result of the collapse of
the plate like particles, provoking the interparticle porosity char-
acteristic of most clay materials [41]. The textural parameters of
the MgFeOx samples were observed to decrease with an increase
in Fe loading. The highest pore volume found in the MgFeO3 and
MgFeO4 samples could be ascribed to their highest porosity in the
range of 4–70 nm.
◦
before 500 C. In return, a peak corresponding to the mass 32 was
detected pointing out a release of oxygen and partial reduction of
the Fe(III) in the mixed oxides could take place [33].
The acid-base properties of catalysts play an important role in
their catalytic performance in the etherification of glycerol. In fact,
the acid sites may conduct the selectivity of the reaction towards
acrolein and the basic ones are involved in the etherification of glyc-
erol, and therefore the production of polyglycerols. The acid sites
The textural properties have been determined from nitrogen
◦
adsorption–desorption isotherms at −196 C of the MgFeOx, after
◦
calcination at 400 C, and the main textural parameters are shown
in Table 2. The shapes of all isotherms are type IV according to the
IUPAC classification (Supplementary Information, Fig. S2), typical
of mesoporous materials [34]. The analysis of the hysteresis loops
has been used to describe the pore structure of porous solids
have been evaluated by NH -TPD analysis (Table 2 and Fig. 4a).
3
The incorporation of Fe3+ cations in the brucite structure modified
drastically the acid properties of the resulting mixed oxide. Actu-
ally, the MgO sample does not show detectable acid sites, whereas
the incorporation of Fe3+ cations produces a broad heterogeneous
distribution of acid sites. The concentration of acid sites increases
[
17,19,25,35–38]. The hysteresis loops of calcined hydrotalcites
resemble to type H3 hysteresis, which is usually found in materials

P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
203
Table 2
Textural parameters of MgFeOx metal oxides.
2
3
Acidity
␮mol NH3/g)
bBasicity
Sample
SBET (m /g)
Vp (cm /g)
dp (av) (nm)
(
(␮mol NH3/m2)
(␮mol CO2/g)
(␮mol CO2/m2)
MgFeO1
MgFeO2
MgFeO3
MgFeO4
112
206
256
221
324
0.214
0.483
0.607
0.665
0.827
7.1
8.8
8.5
10.8
5.6
623
728
653
381
–
5.6
3.5
2.6
1.7
–
265
217
167
184
930
2.4 (0.42)
1.1 (0.31)
0.7 (0.27)
0.8 (0.47)
2.9 (−)
a
MgO
a
The acidity of MgO sample was not detected by means of TPD–NH3.
In brackets the ratio of specific basicity/acidity determined by means of CO2–TPD and NH3–TPD.
b
with decreasing of Mg/Fe molar ratio, reaching a maximum value
for the catalyst MgFeO2. The lower acidity of the MgFeO1 could be
ascribed to its lower SBET, as previously reported by Tu et al. [42] in
the study of the acidity of MgFe mixed oxides by microcalorime-
try of NH3 adsorption. These authors found the highest adsorption
heat and ammonia coverage for a sample with Mg/Fe molar ratio of
regions and it has been calculated the surface concentration of these
elements. The XPS results collected in Table 3.
Regarding the C1s signal (Supplementary Information, Fig. S3),
the MgFe hydrotalcites exhibit two bands: adventitious carbon
at 284.7–284.9 eV and carbonates at 287.8–289.0 eV. The mixed
oxides MgFeOx still show the presence of the carbonates even
◦
1
and the lowest to the MgO sample. Pavel et al. [43] reported that
after the thermal treatment at 450 C, which could be due to the
MgFeAlO mixed oxide showed that the total number of acid sites
existence of remaining strong bonded carbonates, although the
was increased due to the presence of a segregated spinel or Fe O3
contamination by adsorbed CO during the sample handling cannot
2
2
phase. However, León et al. [20] have reported a MgFe mixed oxide
with Mg/Fe molar ratio close to 3 with negligible ammonia adsorp-
tion. The authors ascribed this low acidity value to the impossibility
of Fe3+ ions to occupy a tetrahedral site because of its larger ionic
diameter.
be ruled out.
In the O 1s region (Supplementary Information, Fig. S3), the as-
synthesized hydrotalcites display an asymmetric band that can be
deconvoluted into two bands at 529.5 and 531.2 eV, which could be
ascribed to the presence of O2 and species like OH or carbonates
groups, respectively. The thermal treatment of the hydrotalcites to
generate the MgFe mixed oxides provokes an increase in the inten-
sity of the band at 529.5 eV and a decrease in the intensity of the
−
The CO -TPD profiles of calcined hydrotalcites (Fig. 4b) clearly
2
3+
reveal that the incorporation of Fe provokes a drastic change in
the CO -TPD profile, as well as in the concentration of basic sites
2
◦
(
Table 2). MgO only shows an intense peak centred around 275 C,
whereas the calcined hydrotalcites exhibit two well resolved peaks,
Table 3
◦
one at low temperature with the maximum at 200 C and another
Binding Energies of C 1s, O 1s, Mg 2p y Fe 2p of the as-synthesized hydrotalcites,
MgFeOx catalysts before and after catalytic test.
◦
one at 700 C. Based on these results, the high temperature peak
could be ascribed to strong basic sites generated by the presence of
iron ions in the MgO structure, whereas the low temperature peak
could be attributed to the periclase structure itself. In mixed oxides,
*Sample
C 1s (eV)
O 1s (eV)
Mg 2p (eV)
49.4
Fe 2p (eV)
710.3
MgFe1
285.4 (83.3)
529.6 (14.3)
531.1 (85.7)
529.7 (17.2)
531.4 (82.8)
529.5 (6.7)
531.2 (93.3)
529.5 (5.7)
531.2 (94.3)
288.5 (16.7)
the partial substitution of Mg2 by Fe generates an excess of pos-
itive charge which is compensated by cationic vacancies [44]. The
neighboured oxygen anions to these structural defects are unsat-
urated and provide the strongest basic sites. On the basis of CO2
adsorption coupled to FTIR analysis, it has been reported in the lit-
erature [45–48] that CO2 adsorption on surface hydroxyl groups
+
3+
MgFe2
MgFe3
MgFe4
284.8 (67.6)
289.0 (32.4)
284.9 (57.2)
48.9
49.5
49.5
711.2
711.6
711.1
2
88.8 (42.8)
284.9 (70.0)
88.8 (30.0)
2
(
low strength basic sites) generates bicarbonate ions, whereas car-
**Sample
MgFeO1
C 1s (eV)
O 1s (eV)
Mg 2p (eV)
48.9
Fe 2p (eV)
710.6
2−
bonate species are formed on O
ions (strong basic sites) [20].
284.9 (69.9)
289.7 (30.1)
529.6 (57.0)
531.3 (36.7)
532.7 (6.0)
529.4 (45.5)
531.1 (45.4)
Therefore, the peak at low temperature could be ascribed to bicar-
bonates species desorption whereas the peak at high temperature
could be derived from carbonates species desorption. Moreover,
MgFeO2
MgFeO3
MgFeO4
284.9 (69.1)
48.9
48.9
48.8
710.5
710.6
710.4
288.9 (30.9)
◦
the asymmetry of the peak centred at 700 C would point out the
532.4 (9.0)
formation of different types of carbonates, i.e. unidentate (strong
basic site), bidentate, chelating and bridging carbonates (medium
strength basic site) [20].
Regarding the concentration of basic sites, they decrease for
higher Mg/Fe molar ratio, but in all cases they are lower than MgO.
This fact is related to the increase of unsaturated oxygen anions
285.0 (69.0)
288.9 (31.0)
529.4 (48.2)
531.2 (42.0)
532.5 (9.8)
284.9 (63.9)
529.4 (48.5)
531.2 (42.1)
289.0 (36.1)
532.5 (9.4)
***Sample
MgFeO1
C 1s (eV)
284.8
O 1s (eV)
Mg 2p (eV)
48.8
Fe 2p (eV)
709.7
[
23], which are Lewis base sites, as the Fe loading is increased [42].
In summary, these mixed oxides show features related to both
530.2 (87.9)
532.1 (12.1)
530.3 (85.6)
532.1 (14.4)
530.5 (89.4)
531.8 (10.6)
530.2 (66.7)
531.8 (33.1)
acid and base sites, which surface concentration increases with the
iron loading. Besides, the density of surface acid and basic sites,
namely the concentration per square meter, follows the same trend,
i.e. the MgFeO1 shows the highest density, although the surface
basic/acid sites ratio does not vary irrespectively of the Mg/Fe molar
ratio.
MgFeO2
MgFeO3
MgFeO4
284.8 (93.8)
49.0
48.9
48.8
709.4
709.3
708.2
2
87.2 (6.2)
284.8 (86.5)
87.9 (13.5)
2
284.0 (39.1)
285.2 (60.9)
The surface characteristics of the as-synthesized hydrotalcites
and the mixed oxides have been studied by means of XPS. It has
been evaluated the signals in the C 1s, O 1s, Mg 2p and Fe 2p
*
As-synthesized hydrotalcites.
** Calcined hydrotalcites at 450 ◦C.
*
**
XPS results after catalytic test.

2
04
P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
8
7
7
6
6
5
5
4
4
3
0000
5000
0000
5000
0000
5000
0000
5000
0000
5000
6
5000
0000
5000
0000
5000
40000
5000
0000
6
5
5
4
3
3
7
40735730725720715710705 700695
7
40735 730725720 715710705 700695
B.E. (eV)
B.E. (eV)
5
5000
0000
5000
0000
5000
0000
5000
0000
5
4
4
3
3
2
2
7
40735730725720715710705700695
B.E. (eV)
MgFe1
MgFe2
MgFe3
MgFe4
Fig. 5. Fe2p core level of the as-synthesized hidrotalcites, MgFeOx catalysts before and after reaction (from left to right and top to bottom).
band at 531.2 eV. This fact remarks the dehydroxilation and decar-
bonation of the as-synthesized hydrotalcites, as ATD-TG analysis
showed, and it points out the increase of the strong basic centres
binding energy of the Fe 2p3/2 peak and the presence of a detectable
satellite peak at 8 eV from the Fe 2p3/2 peak (Fig. 5) [51–54].
on the surface of mixed oxides expressed in the presence of O2
−
3.2. Catalytic results
species. Moreover, the increase in the Fe loading is associated to an
increase in the band ascribed to the O2− ions. The XPS results cor-
roborate that the sample with the lower Mg/Fe molar ratio showed
the highest CO2 adsorption at the highest temperatures measured
Currently, much attention is being paid to the etherification
of glycerol to polyglycerol, under heterogeneous conditions, since
this process takes advantages, such as easiness of catalyst separa-
tion from the reaction medium and catalyst reutilization. However,
compared to homogeneous catalysts, the main drawbacks of solid
catalysts are related to lower catalytic activity, the requirement
of more drastic reaction conditions and the possible leaching in
the reaction media. In this etherification process, one of the main
catalytic goals is to address the selectivity towards di-glycerol and
tri-glycerol, since these oligomers are starting materials for the food
and cosmetic industries. In this present work, the etherification of
glycerol to polyglycerols using heterogeneous catalysis was carried
by CO -TPD.
2
Finally, in the Fe 2p region, the as-synthesized hydrotalcites and
the mixed oxides show a broad Fe 2p3/2 peak, which broadening is
attributed to multiplet splitting and shake-up phenomena [49]. The
presence or absence of shake up satellite near of the photoelecton
peak provides information about the oxidation state of iron [50].
If the iron is at the oxidation state (+II) the satellite is a shoulder
in the photoelectron peak, whereas in the case of oxidation state
(
+III) the satellite peak is a discernible peak at about 8 eV from the
◦
photoelectron peak and, finally if oxidation state of the iron is (0)
the satellite peak is absent. The photoelectron Fe 2p3/2 peak of the
as-synthesized hydrotalcites is ranged between 709.9 and 712.0 eV.
These binding energies together the shape of spectra are indicative
of the presence of Fe(III), as reported for compounds like FeOOH and
out in a batch reactor, without solvent, at 220 C under a continu-
ous nitrogen flow, during 24 h under atmospheric pressure. Under
these experimental conditions, di-glycerols and tri-glycerols were
only detected.
Fig. 6a shows the catalytic results obtained over MgFeOx cata-
◦
lysts, 2 wt% at 220 C, together with those of commercial Na CO ,
2 3
Fe(OH) [50]. When the hydrotalcites are converted in mixed oxides
3
◦
by the thermal treatment at 450 C, the Fe 2p spectra display the
MgO and Fe(OH) for comparison. Na CO was the most active cat-
3 2 3
photolectronic features of iron at oxidation state (+III), that is, the
alyst under these experimental conditions, reaching 100% glycerol

P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
205
100
8
0
0
0
0
Conversion
DG Selectivity
TG Selectivity
6
4
2
0
MgFeO1
MgFeO2
MgFeO3
MgFeO4
MgO
Na2CO3
Catalysts
4
4
3
3
2
2
1
1
5
MgFeO1
MgFeO2
MgFeO3
MgFeO4
0
5
0
5
0
5
0
5
0
0
2
4
6
8
10
12
14
16
18
20
22
24
26
Time (h)
Fig. 6. Catalytic performance MgFeOx catalysts: (a) glycerol conversion and diglycerol and triglycerol selectivity; (b) glycerol conversion versus reaction time (reaction
◦
conditions: glycerol = 15 g, catalyst weight = 300 mg; temperature = 220 C, reaction time: 24 h).
Table 4
conversion after 24 h of reaction as previous results have shown
Mg/Fe atomic ratio determined by means of XPS analysis and Mg/Fe molar ratio
determined by ICP-AES analysis.
[
9]. It is noteworthy that its catalytic activity was higher than that
obtained with a MgO prepared by activation of the corresponding
nitrate at 700 C [11]. The MgFeOx mixed oxides show a gradual
◦
aMg/Fe
Sample
increase of the catalytic activity with the Mg/Fe molar ratio, being
the MgFeO4 catalyst the most active. Thus, glycerol conversion
varies between 21.1% and 41.1%, after 24 h time on stream. None
of the MgFeOx catalysts reached full conversion of glycerol under
these experimental conditions. The catalytic behaviour seems to be
related to both the acid-basic properties and structural stability of
catalysts (as inferred from the superficial Mg/Fe atomic ratio). The
MgFeO4 catalyst shows the higher specific basicity/acidity ratio
*MgFeOx
bMg/Fe (%)
MgFex
MgFeOx
MgFe1
MgFe2
MgFe3
MgFe4
0.886
1.602
3.218
3.632
0.899
1.348
2.372
3.125
0.474
0.692
1.965
3.190
1.0
1.7
2.9
3.6
a
b
*
Measure from XPS analysis.
Measure from ICP–AES.
Data after catalytic tests.
(
see Table 2), whereas the surface Mg/Fe atomic ratio (Table 4)
after the catalytic reaction is hardly affected. The rest of catalysts

2
06
P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
Group 2
eluted after the cyclic ones. Based on reaction probabilities [57],
Linear Isomers
the ␣,␣-diglycerol will be formed more readily than the branched
isomers, and therefore the highest peak can be ascribed to that iso-
mer. Finally, the Group 3 accounts for the triglycerol isomers whose
intensity is lower than that of linear isomers
MgFeO1
MgFeO2
MgFeO3
MgFeO4
The etherification of glycerol is a shape-selective reaction in
which the pore framework plays an important role to lead the selec-
tivity to a particular product, in this case to diglycerol. The textural
analysis of the catalysts has proven that these catalysts possess pore
diameters large enough to facilitate the diffusion of glycerol and the
reaction products. Indeed, the sizes of glycerol isomers [57] range
between 0.667 and 0.753 nm for ␣␣-diglycerol and ␤␤-diglycerol,
respectively. Therefore, the reaction may occur into the pores of the
mixed oxides and the existence of large pore diameters in all cases
is responsible of the similar selectivity values observed in all cases.
Although, from a qualitative point of view and bearing in mind that
the statistical distribution of the ␣,␣:␣,␤:␤,␤ dyglycerol isomers
ratios might be 57:29:14 [57], the actual ratio is quite shifted to
the formation of the (␣,␣) isomer. This means that these catalysts
favour the synthesis of those diglycerol isomers.
Group 3
TriGlycerols
Group 1
Cyclic Isomers
18
20
22
24
26
28
30
32
34
Time (min)
It has been previously reported that the more active catalysts
shows basic features and acidic ones address the selectivity towards
the acrolein formation [5,11]. Weckhuysen et al. [11] also showed
that the Lewis acid sites of the alkaline-earth metals played a role
in the activation of glycerol molecules. In this mechanism, a basic
centre attacks a hydroxyl group of a glycerol molecule, extracting
a proton, whereas an unsaturated metal site on the catalyst sur-
face is able to activate a hydroxyl group of other glycerol molecule,
which is then attacked by the nucleophilic oxygen of the former
glycerol molecule yielding the diglycerol molecule. This acid–base
bifunctional mechanism has been postulated for other catalytic
reactions, in which mixed oxides derived from hydrotalcites have
been used [23,25,43,58]. This mechanism could be involved in pres-
ence of MgFeOx catalysts where acid and basic sites are present, as
demonstrated by NH -TPD and CO -TPD.
The influence of reaction time in the catalytic performance
(Fig. 6b) shows that the catalytic performance of MgFeO4 catalyst
is the highest in the whole range of time. It is noteworthy that the
catalysts require an activation period before showing any catalytic
activity. The selectivity to diglycerol is 100%, at least within the first
eight hours, and then triglycerol becomes measurable.
The main advantage of using heterogeneous versus homoge-
neous catalysts is the possibility of reusing. For that, the active
phase must not be leached to the reaction medium. Thus, the
organic phase was analyzed by ICP-AES to detect any trace of iron
or magnesium. The absence of metals in the reaction medium, even
with those catalysts showing a decrease of Mg/Fe atomic ratio, con-
firms the stability of the solid catalysts. Therefore, the Mg/Fe atomic
ratio depletion observed after the catalytic test is not accounted for
the leaching of Mg or Fe but it is explained by the segregation of
iron species.
Fig. 7. Chromatograms of the reaction mixture after 24 h as function of the MgFeOx
catalysts.
evidence a significant reduction of the exposed iron, being nearly
5
0% for the MgFeO and MgFeO catalysts (Table 4). It is noteworthy
2 1
to point out that all catalysts present a similar carbon deposition
degree on their surfaces, so this decrease in the surface Mg/Fe
atomic ratio is not associated to carbon deposition. Thus, this fact
could point out a partial segregation of iron species from the MgFeO
framework during the catalytic process. The extraction of the Fe3
from the periclase structure could be related to the partial reduc-
+
3
+
2+
tion of Fe to Fe , as it is observed by XPS analysis, since the Fe 2p
core level of the mixed oxides (Fig. 5) reveals a shift to lower bind-
ing energies, roughly 1 eV. Furthermore, the satellite peak appears
as a shoulder of the Fe 2p3/2 photoelectronic peak, thus corrobo-
3
2
rating the partial reduction of Fe3 to Fe . Therefore, the existence
of extra-framework iron could be related to the lower catalytic
performance of catalysts with Mg/Fe molar ratio lower than 4.
Regarding the selectivity (Fig. 6a), all these catalysts display
selectivity towards diglycerol greater than 90%, which declines as
glycerol conversion is increased. This selectivity results are sim-
ilar to those reported by Richter et al. [55] and Clacens et al. [9]
for similar levels of glycerol conversion. It is well known that
oligomerisation of glycerol can lead to dimers, trimers and higher
oligomers (Supplementary Information, Scheme S1). Moreover,
once a dimer is formed, cyclic products can be produced as result of
intramolecular ring closure reactions [56] and therefore the compo-
sition of a polyglycerol mixture is extremely complex. Taking into
account the chromatograms of standard reference compounds, the
chromatogram peaks of reaction mixture after the silylation pro-
cess have been grouped in three classes, which have been labelled
as Group 1, 2 and 3 (Fig. 7). The Group 1 included those species
eluted between 20 and 22 min, the Group 2 included those peaks
between 22 and 23 min and Group 3, those between 30 and 31 min.
The Group 1 (Supplementary Information Scheme S1) consisted
of small peaks associated to the existence of cyclic diglycerol iso-
mers. These cyclic isomers only have two hydroxyl groups to react
and therefore they are trimethyl-disilyl derivatives and as conse-
quence their molecular weight are lower and they eluted at lower
retention times. The peaks included in the Group 2 are the most
intense, and hence these isomers are more abundant. These are
linear isomers (Supplementary Information Scheme S1), such as
+
2+
4. Conclusions
Magnesium iron hydrotalcites with variable Mg/Fe molar ratio,
ranging from 1 to 4, have been prepared as precursors of MgFe
mixed oxides, which have demonstrated to be active in the selective
etherification of glycerol to yield mainly diglycerol. The catalytic
behaviour seems to be influenced by the stability of iron in the
structure of MgO oxide. This is achieved with the Mg/Fe molar
ratio of 4. The rest of catalysts show a noticeable decrease of Mg/Fe
atomic ratio, as determined by XPS, after the catalytic test, indi-
cating a partial segregation of iron species over the MgO surface.
These catalysts display a catalytic selectivity to diglycerol (<90%)
without significant differences between the tested catalysts. More-
over, the catalysts do not show any leaching of the Fe or Mg in the
␣
,␣-diglycerol, ␣,␤-diglycerol and ␤,␤-diglycerol [56]. The linear
diglycerol isomers have four hydroxyl groups and therefore their
molecular weight is higher than such cyclic isomers and then they

P. Guerrero-Urbaneja et al. / Applied Catalysis A: General 470 (2014) 199–207
207
reaction mixture after the catalytic test, which is important when
considering their reuse in several catalytic cycles.
[23] V. Crocellà, G. Cerrato, G. Magnacca, C. Morterra, F. Cavani, S. Cocchi, S. Passeri,
D. Scagliarini, C. Flego, C. Perego, J. Catal. 270 (2010) 125–135.
[
24] N. Tahir, Z. Abdelssadek, D. Halliche, S. Saadi, R. Chebout, O. Cherifi, K. Bachari,
Surf. Interface Anal. 40 (2008) 254–258.
Acknowledgements
[25] M. León, E. Díaz, S. Bennici, A. Vega, S. Ordó n˜ ez, A. Auroux, Ind. Eng. Chem. Res.
9 (2010) 3663–3670.
4
[
[
26] S.K. Sharma, P.A. Parikh, R.V. Jasra, Appl. Catal., A 386 (2010) 34–42.
27] C.M. Jinesh, C.A. Antonyraj, S. Kannan, Appl. Clay Sci. 48 (2010) 243–249.
The authors are grateful to financial support from the Span-
ish Ministry of Science and Innovation (ENE2009-12743-C04-03
project), Junta de Andalucía (P09-FQM-5070) and FEDER funds.
RMT would like to thank this Ministry of Science and Innovation
for the financial support under the Program Ramón y Cajal (RYC-
[28] M.A. Ulibarri, M.J. Hernández, J. Cornejo, Thermochim. Acta 113 (1987)
9–86.
[
7
29] O.P. Ferreira, O.L. Alves, D.X. Gouveia, A.G. Souza Filho, J.A.C. de Paiva, J.M. Filho,
J. Solid State Chem. 177 (2004) 3058–3060.
[30] J.S. Valente, H. Pfeiffer, E. Lima, J. Prince, J. Flores, J. Catal. 279 (2011)
196–204.
[
[
[33] R.L. Frost, K.L. Erickson, J. Therm. Anal. Calorim. 76 (2004) 217–225.
[
2
008-03387).
31] M. Vucelic, W. Jones, G.D. Moggridge, Clays Clay Miner. 45 (1997) 803–813.
32] J. Pérez-Ramírez, S. Abelló, N.M. van der Pers, Chem. Eur. J. 13 (2007) 870–879.
Appendix A. Supplementary data
34] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquérol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
35] S.K. Sharma, P.K. Kushwaha, V.K. Srivastava, S.D. Bhatt, R.V. Jasra, Ind. Eng.
Chem. Res. 46 (2007) 4856–4860.
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.2013.
[
1
0.051.
[36] E. Angelescu, O.D. Pavel, R. Bîrjega, M. Florea, R. Zavoianu, Appl. Catal., A 341
2008) 50–57.
(
[
37] P. Kustrowski, L. Chmielarz, E. Bozek, M. Sawalha, F. Roessner, Mater. Res. Bull.
39 (2004) 263–281.
38] S. Kannan, J. Mater. Sci. 39 (2004) 6591–6600.
References
[
[
[
1] <http://www.ebb-eu.org/stats.php/>.
2] M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi, C. Della Pina, Angew. Chem. Int.
Ed. 46 (2007) 4434–4440.
[39] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids,
Academic Press, Waltham, MA, 1999.
[40] W.T. Reichle, S.Y. Kang, D.S. Everhardt, J. Catal. 101 (1986) 352–359.
[41] J.A. Gursky, S.D. Blough, C. Luna, C. Gomez, A.N. Luevano, E.A. Gardner, J. Am.
Chem. Soc. 128 (2006) 8376–8380.
[3] B. Katryniok, S. Paul, V. Bellière-Baca, P. Rey, F. Dumeignil, Green Chem. 12
(
2010) 2079–2080.
[
[
4] C.H. Zhou, J.N. Beltramini, Y.-X. Fan, G.Q. Lu, Chem. Soc. Rev. 37 (2008) 527–549.
5] M.V. Sivaiah, S. Robles-Manuel, S. Valange, J. Barrault, Catal. Today 198 (2012)
[42] M. Tu, J. Shen, Y. Chen, J. Solid State Chem. 128 (1997) 73–79.
[43] O.D. Pavel, D. Tichit, I.-C. Marcu, Appl. Clay Sci. 61 (2012) 52–58.
[44] F. Cavani, L. Maselli, D. Scagliarini, C. Flego, C. Perego, Stud. Surf. Sci. Catal. 155
(2005) 167–177.
[45] H.A. Prescott, Z.-J. Li, E. Kemnitz, A. Trunschke, J. Deutsch, H. Lieske, A. Auroux,
J. Catal. 234 (2005) 119–130.
3
05–313.
[
[
6] A. Martin, M. Richter, Eur. J. Lipid Sci. Technol. 113 (2011) 100–117.
7] M. Ayoub, M.S. Khayoon, A.Z. Abdullah, Bioresour. Technol. 112 (2012)
3
08–312.
[
[
8] J.-M. Clacens, Y. Pouilloux, J. Barrault, Stud. Surf. Sci. Catal. 143 (2002) 687–695.
9] J.-M. Clacens, Y. Pouilloux, J. Barrault, Appl. Catal., A 227 (2002) 181–190.
[46] H. Du, C.T. Williams, A.D. Ebner, J.A. Ritter, Chem. Mater. 22 (2010) 3519–3520.
[47] J. Shen, M. Tu, C. Hu, J. Solid State Chem. 137 (1998) 295–301.
[48] F. Prinetto, G. Ghiotti, R. Durand, D. Tichit, J. Phys. Chem. B 104 (2000)
11117–11126.
[49] A.P. Grosvenor, B.A. Kobe, M.C. Biesinger, N.S. McIntyre, Surf. Interface Anal. 36
(2004) 1564–1570.
[50] M. Descostes, F. Mercier, N. Thromat, C. Beaucaire, M. Gautier-Soyer, Appl. Surf.
Sci. 165 (2000) 288–302.
[51] M. Hadnadjev, T. Vulic, R. Marinkovic-Neducin, Y. Suchorski, H. Weiss, Appl.
Surf. Sci. 254 (2008) 4297–4300.
[
[
10] J. Barrault, J.-M. Clacens, Y. Pouilloux, Top. Catal. 27 (2004) 137–142.
11] A.M. Ruppert, B.M. Weckhuysen, J.D. Meeldijk, B.W.M. Kuipers, B.H. Erné, Chem.
Eur. J. 14 (2008) 2016–2020.
[
[
12] Z. Gholami, A.Z. Abdullah, K.-T. Lee, J. Taiwan Inst. Chem. Eng. 44 (2013)
1
17–122.
13] C. García-Sancho, R. Moreno-Tost, J.M. Mérida-Robles, J. Santamaría-González,
P. Maireles-Torres, Catal. Today 167 (2011) 84–90.
[
[
[
[
14] M.R. Anuar, A.Z. Abdullah, M.R. Othman, Catal. Commun. 32 (2011) 67–70.
15] D. Tichit, F. Fajula, Stud. Surf. Sci. Catal. 125 (1999) 329.
16] F. Cavani, F. Triffirò, A. Vaccari, Catal. Today 11 (1991) 173–301.
17] K.S. Triantafyllidis, E.N. Peleka, V.G. Komvokis, P.P. Mavros, J. Colloid Interface
Sci. 342 (2010) 427–436.
[52] R.J. Balasamy, A. Khurshid, A.A.S. Al-Ali, L.A. Atanda, K. Sagata, M. Asamoto, H.
Yahiro, K. Nomura, T. Sano, K. Takehira, S.S. Al-Khattaf, Appl. Catal., A 390 (2010)
225–234.
[53] T.L. Barr, J. Phys. Chem. 82 (1978) 1801–1810.
[
[
18] Y. Yang, N. Gao, Y. Deng, S. Zhou, Appl. Clay Sci. 65–66 (2012) 80–86.
19] J.S. Valente, J. Hernández-Cortez, M.S. Cantu, G. Ferrat, E. López-Salinas, Catal.
Today 150 (2010) 340–345.
[54] A.C. Heredia, M.I. Oliva, C.I. Zandalazini, U.A. Agú, G.A. Eimer, S.G. Casuscelli, E.R.
Herrero, C.F. Pérez, M.E. Crivello, Ind. Eng. Chem. Res. 50 (2011) 6695–6700.
[55] M. Richter, Y.K. Krisnandi, R. Eckelt, A. Martin, Catal. Commun. 9 (2008)
2112–2120.
[56] B. De Meulenaer, B. Vanhoutte, A. Huyghebaert, Chromatographia 51 (2000)
44–52.
[57] A. Martin, M.P. Checinski, M. Richter, Catal. Commun. 25 (2012) 130–135.
[58] D. Tichit, M.H. Lhouty, A. Guida, B.H. Chiche, F. Figueras, A. Auroux, D. Bartalini,
E. Garrone, J. Catal. 151 (1995) 50–59.
[
[
[
20] M. León, E. Díaz, A. Vega, S. Ordó n˜ ez, A. Auroux, Appl. Catal., B 102 (2011)
5
90–599.
21] P. Kustrowski, A. Rafalska-Lasocha, D. Majda, D. Tomaszewska, R. Dziembaj,
Solid State Ionics 141–142 (2001) 237–242.
22] Y. Ohishi, T. Kawabata, T. Shishido, K. Takaki, Q. Zhang, Y. Wang, K. Nomura, K.
Takehira, Appl. Catal., A 288 (2005) 220–231.