Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of citral to unsaturated alcohols

Article abstract of DOI:10.1016/j.jcat.2016.09.026

A series of carbon xerogels-TiO₂ composites with different TiO₂ contents were prepared, exhaustively characterized and used as a Pt-support to develop selective hydrogenation catalysts. The carbon phase in the composite hinders the TiO₂ crystal growth and the transformation to rutile during thermal treatments. Textural, chemical and catalytic properties are determined by the TiO₂ content, with an optimum around 40 wt.% of TiO₂ content. The mesoporosity of the supports, the Pt-dispersion and Pt-support interactions are favoured in this sense. During the H₂-pretreatment, the Pt and TiO₂ phases were simultaneously reduced and the formation of oxygen vacancies leads to the mobility of Pt-species inside the TiO₂ structure, avoiding sintering in surface and strongly improving both catalytic activity and selectivity. The catalytic performance was discussed on the basis of the sample characteristics. Unsaturated alcohols were obtained as main reaction products in all cases, being the only product in the case of the optimized catalyst.

Full text of DOI:10.1016/j.jcat.2016.09.026

Journal of Catalysis xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Chemoselective Pt-catalysts supported on carbon-TiO2 composites for
the direct hydrogenation of citral to unsaturated alcohols
⇑
Esther Bailón-García, Francisco Carrasco-Marín, Agustín F. Pérez-Cadenas, Francisco J. Maldonado-Hódar
Department of Inorganic Chemistry, Faculty of Sciences, University of Granada, Avda. Fuentenueva s/n, 18071 Granada, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of carbon xerogels-TiO2 composites with different TiO2 contents were prepared, exhaustively
characterized and used as a Pt-support to develop selective hydrogenation catalysts. The carbon phase
in the composite hinders the TiO2 crystal growth and the transformation to rutile during thermal treat-
ments. Textural, chemical and catalytic properties are determined by the TiO2 content, with an optimum
around 40 wt.% of TiO2 content. The mesoporosity of the supports, the Pt-dispersion and Pt-support inter-
actions are favoured in this sense. During the H2-pretreatment, the Pt and TiO2 phases were simultane-
ously reduced and the formation of oxygen vacancies leads to the mobility of Pt-species inside the TiO2
structure, avoiding sintering in surface and strongly improving both catalytic activity and selectivity. The
catalytic performance was discussed on the basis of the sample characteristics. Unsaturated alcohols
were obtained as main reaction products in all cases, being the only product in the case of the optimized
catalyst.
Received 7 June 2016
Revised 22 August 2016
Accepted 19 September 2016
Available online xxxx
Keywords:
Carbon xerogels
Pt-dispersion
C/TiO2-composites
Oxygen vacancies
Metal-support interactions
Unsaturated alcohols
Selective hydrogenation
Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction
[11,12]. The catalytic performance is influenced by the active
phase, the support nature and their interactions. Probably, the
Nowadays, unsaturated alcohols (UA) are highly demanded by
organic synthesis, fine chemistry, perfumery or pharmaceutical
industries [1,2]. These compounds can be obtained from available
most important parameter is the selection of the active phase. This
factor has been studied by different authors [13–16], and Vannice
and Singh [16] show that the catalytic activity in citral hydrogena-
tion varies as Pd > Pt > Ir > Os > Ru > Rh > Ni > Co ꢀ Fe. Neverthe-
less, the selectivity to UA of these monometallic catalysts is high
only for Os-catalyst (at around 88%), moderate for Ru and Co
(55%) and null (0% UA) being selective to citronellal, for Pt, Ir, Rh,
Ni and Pd-catalysts.
This performance however, strongly depends on the support, as
revealed by different studies [17,18]. Therefore, by studying the
influence of the supports (graphite, KL zeolite and ZrO2) on Ru-
catalyst performance [17], it was pointed out that using graphite
as Ru-support, the selectivity to UA is favoured by electronic trans-
fers from graphite to the Ru-particles with the formation of
electron-rich metal species. In the case of Ru-supported on ZrO2,
this effect is associated with the formation of Ru0-Zrn+ species
and when supported on KL zeolite the hydrogenation of the car-
bonyl group and poisoning resistance are favoured by geometrical
factors (pore characteristics and localization of the Ru-particles).
Ru/KL is the most selective catalyst to UA (around 50%), but it is
poorly active.
a,b-unsaturated aldehydes. Among them, citral (3,7-dimethyl-
2,6-octadienal) is a cheap raw material which is one of the main
components of the lemongrass oil and it can also be produced at
a large scale trough petrochemical processes [3]. Valuable UA,
nerol and geraniol, can be obtained from citral. However, citral is
a multi-unsaturated compound containing both an isolated double
bond (C@C) and another C@C conjugated to a carbonyl (AC@O)
group [4,5]. Consequently, the citral hydrogenation presents a
complex reaction scheme [6] leading to a variety of products. The
main problem is that the hydrogenation of C@C bonds is thermo-
dynamically favoured over the C@O bonds; moreover, there is also
a series of side reactions such as acetalization, cyclization or crack-
ing [4,7,8] that reduces the UA yield.
To obtain the chemoselective C@O hydrogenation, specific cat-
alysts should be designed [8–10]. The use of monometallic hetero-
geneous catalysts, instead of forming UA, leads mainly to
unsaturated aldehyde (citronellal) or saturated alcohol (3,7
dimethyl octanol) due to the hydrogenation of the C@C bonds
In previous works [13,19] we also pointed out the influence of
the chemical and porous characteristics of both carbon and
inorganic supports used to develop Pt-catalysts for citral
⇑
Corresponding author. Fax: +34 958248526.
E-mail address: fjmaldon@ugr.es (F.J. Maldonado-Hódar).
http://dx.doi.org/10.1016/j.jcat.2016.09.026
0021-9517/Ó 2016 Elsevier Inc. All rights reserved.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

2
E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
hydrogenation. Using carbon supports, the performance of the
derivatives Pt-catalysts strongly depends on the micro/meso-
porous character of this support and the presence of impurities,
namely inorganic components. Mesoporosity and acid impurities
leads to a decrease in UA yields; therefore, the best results were
obtained with pure microporous carbon xerogels [19]. Similar con-
clusions were obtained when using inorganic supports: meso-
porosity favours secondary reactions inside the pores, while
Brönsted acidity strongly favours cracking reactions [13]. On such
a basis, Al2O3 always provides worse results than TiO2 as Pt-
supported catalysts for citral hydrogenation, in spite of their simi-
lar porous characteristics and pHpzc values.
wave heating under an argon atmosphere in periods of 1 min at
300 W until constant weight, using a Saivod MS-287W microwave
oven. The pyrolysis of organic xerogel-titanium oxide composites
to obtain the corresponding carbon-TiO2 composites, was carried
out at 900 °C in a tubular furnace using a N2 flow at 300 cm3/
min and a heating rate of 1 °C/min, in order to allow a soft remov-
ing of pyrolysis gases, and
a soaking time of 2 h at this
temperature.
The recipes were fitted to obtain six carbon-TiO2 composites
containing a different amount of titanium oxide (around 10, 20,
30, 40, 50 and 80 weight% in the carbonized materials). For that,
the proportion of organic and inorganic precursor was balanced
and a weight loss during carbonization of around 50% was assumed
according to previous experiences. The real TiO2 content of the
samples was finally determined by thermogravimetric analysis
(TGA) using a Mettler-Toledo TGA/DSC1 thermobalance and burn-
ing the organic fraction at 500 °C in flowing air until constant
weight. Carbon-TiO2 composites were refereed as CTiX (X corre-
sponding to the theoretical TiO2 content programmed for each
sample, e.g. CTi40 should contain around 40 wt.% of TiO2). Addi-
tionally, a pure carbon xerogel prepared in our laboratories
[13,19] and titanium dioxide (anatase, from Alfa Aesar) were used
as reference materials.
This behaviour also highly depends on the pretreatment condi-
tions of the catalyst, which determine the chemical and crystalline
transformation of the supports as well as the nature and dispersion
of the active phase. In spite of the fact that Pt-sintering in general is
favoured by H2-pretreatment regarding the He-pretreatment [20]
(which should lead to a stronger catalyst deactivation) using Pt/
TiO2 catalysts, its catalytic performance is improved after the H2-
pretreatment. This is because the H2 treatment also favours the
partial reduction of the TiO2 surface in such a way that the syner-
getic combination of Pt site/oxygen vacancy, strongly favours the
C@O hydrogenation to UA during citral hydrogenation [21].
These results show new opportunities in the preparation of
highly selective hydrogenation catalysts by optimizing the support
characteristics and the interactions with the corresponding active
phase. New Pt-supports can be developed by taking advantage of
the ability of TiO2 to produce specific active site for citral adsorp-
tion through the C@O bond, the developed porosity and fitted sur-
face chemistry (basicity) of pure carbon xerogels. The sol-gel
synthesis procedure guarantees the purity of these supports avoid-
ing interferences of the mineral matter present on classical acti-
vated carbons. The high surface area of carbon xerogels should
facilitate the distribution and reduction in the titanium dioxide
particles in the composite, as well as the Pt-dispersion favouring
the metal-support interactions (SMSI effect) in the derivative cata-
lysts. So that, in this work, TiO2-carbon xerogel composites are pre-
pared with different TiO2 percentages and used as supports to
develop specific and chemoselective Pt-catalysts for citral hydro-
genation reactions. The catalytic behaviour of these samples was
studied, relating activity and mainly selectivity to their chemical,
structural and porous characteristics.
2.2. Catalysts preparation
All supports were milled and sieved to a diameter smaller than
0.15 mm before impregnation. Platinum catalysts were prepared
by impregnation at 3 wt.% Pt-loading using an aqueous solution
containing the appropriate amount of [Pt(NH3)4]Cl2, dried over
night at 120 °C and finally pretreated in H2 flow at 400 °C (heating
rate of 5 °C/min) for 12 h. Catalysts were refereed indicating the
support, the Pt-content and the final mean Pt-particle size deter-
mined by H2-chemisorption. Therefore, as an example, CTi40Pt3-
9 indicates that Pt nanoparticles present a mean particle size of
9 nm when deposited with a 3 wt.% Pt-loading on the carbon-
TiO2 composite which contains 40% of titanium oxide.
2.3. Textural and chemical characterization
Textural characterization was carried out by N2 and CO2 adsorp-
tion at ꢁ196 °C and 0 °C, respectively, using a Quantachrome
Autosorb-1 equipment. The BET and Dubinin–Radushkevich equa-
2. Experimental
tions were applied to determine the apparent surface area (SBET)
and the micropore volume (W0), the mean micropore width (L0)
and the microporous surface (Smic), respectively. Furthermore,
the BJH method was used to calculate the mesopore volume of
the samples (Vmes). Pore size distributions were also obtained by
applying the BJH method. The total pore volume was considered
as the volume of N2 adsorbed at P/P0 = 0.95.
The titania phase was determined by a powder X-ray diffraction
(XRD) pattern, using a Bruker D8 Advance X-ray diffractometer
with Cu Ka radiation at a wavelength (k) of 1.541 Å. The 2h angles
2.1. Synthesis of TiO2-carbon xerogel composites
Carbon-TiO2 composites were prepared by the sol-gel synthesis
procedure using resorcinol-formaldehyde and titanium isopropox-
ide (IV) as carbon and TiO2 precursor, respectively. In a typical syn-
thesis procedure, Span 80 (S) was dissolved in 900 ml of n-heptane
and heated at 70 °C under reflux and stirring (450 rpm). After sta-
bilizing this temperature, another aqueous (W) solution containing
the corresponding amount of resorcinol (R) and formaldehyde (F)
was added drop-wise. The organic RF gel was formed immediately
and after that, the proper amount of titanium isopropoxide was
also added drop by drop to the suspension. The final molar ratio
of the mixture was R/F = 1/2, R/W = 1/14 and R/S = 4.5.
The formed hydrogel was aged at 70 °C for 24 h under stirring
after which the solid, displaying an intense orange colour, was
recovered by filtering and placed in acetone for 5 days (changing
acetone twice daily) to remove the rest of unreacted products
and to exchange solvents within the pores by acetone. This proce-
dure reduces the collapse of porosity during the subsequent drying
process [22]. Then, the gel was filtered again and dried by micro-
were scanned from 20° to 70° and the average crystallite sizes (D)
were estimated by the Debye-Scherer equation.
Pt dispersion (D) and mean particle size (d) were obtained by
H2-chemisorption and high-resolution transmission electron
microscopy (HRTEM). The H2-chemisorption isotherms were mea-
sured at 25 °C. The Pt dispersion (D) is obtained from the amount
of H2-chemisorbed assuming a stoichiometric ratio H2:Pt = 1:2
(dissociative chemisorption) and the average particle size was cal-
culated as dPt(H2) = 1.08/D (nm). The chemical characterization of
the catalysts was further analysed by X-ray photoelectron spec-
troscopy (XPS). The spectra were obtained on a Kratos Axis Ultra-
DLD X-ray photoelectron spectrometer equipped with a hemi-
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
3
spherical electron analyzer connected to a detector DLD (delay-line
detector).
2.4. Catalytic performance
The citral hydrogenation was carried out in 100 ml heptane
solution at a constant hydrogen pressure of 8.3 bar and 90 °C using
a Parr reactor model 5500. The experimental conditions such as
citral concentration, catalyst weight and stirring speed, were previ-
ously optimized in order to avoid mass transfer limitations (results
not showed) and fixed in 0.05 M, 500 mg and 1500 rpm, respec-
tively. A small volume of the sample (1 mL) was periodically with-
drawn and analysed by chromatography using a Bruker 430-GC
equipped with a FID detector and a Varian GC Capillary Column
CP7485 (25 m ꢂ 0.32 mm ꢂ 0.45
lm). Citral and all possible prod-
ucts were previously calibrated.
3. Results and discussion
3.1. Support characterization
The TiO2 content, present in each composite and determined by
TGA is shown in Table 1. The TiO2 percentages are slightly higher
than the theoretical ones, due to the weight lost during the car-
bonization which is not exactly 50%; however, the experimental
data are much closer to the expected ones. The pHpzc values point
out the different acid-base characteristics of each support. Because
pure carbon material is a basic support while pure TiO2 phase is
acid, in general the pHpzc values tend to decrease when increasing
the TiO2 content in the composite.
Powder XRD measurements of carbon-TiO2 composites were
performed in order to determine the nature and particle size of
TiO2 (Fig. 1). Below a TiO2 content of 30%: XRD patterns do not
show any defined diffraction peak corresponding to the inorganic
phase, the small shoulders observed are characteristic of the amor-
phous carbon component, indicating a very small nanoparticle size
or the amorphous character of the TiO2 phase. With a greater TiO2
content, crystallinity of the samples increases and wide peaks cor-
responding to the anatase polymorphic form, are observed in the
CTi40 XRD pattern. An incipient rutile phase is observed on
CTi50, being rutile however predominant only in the composite
CTi80. For pure TiO2 oxide (patterns not included) this transforma-
tion is fast at temperatures above 730 °C, but the phase transition
temperature has been shown to depend on impurity content, par-
ticle size and surface area [23]. The above results even after car-
bonization at 900 °C indicate the following: the low crystallinity
of TiO2 in the composites denotes that this phase is mainly well
dispersed on the carbon phase, which prevents the sintering of
TiO2 nanoparticles and stabilizes the anatase structure at high tem-
perature when TiO2 content is smaller than 50 wt.% (i.e. avoiding
the rutile formation).
20
30
40
50
60
70
80
2θ (degrees)
Fig. 1. Powder XRD patterns of samples: CTi10 (
D
), CTi20 (}), CTi30 (h), CTi40 (s),
CTi50 ( ) and CTi80 ( ). Anatasa (j), Rutile (N).
r
are collected in Table 2. It is well known that the CO2 adsorption
provides the information about the narrow microporosity, corre-
sponding to micropores with a diameter lower than 0.7 nm, while
the total microporosity is obtained from N2 isotherm only in the
absence of diffusion restrictions [24]. The results obtained from
CO2 and N2 adsorption show a great variation between the textural
properties of the composites. This fact is due to the different poros-
ity ranges also present in both pure phases. Therefore, the pure car-
bon phase is typically a microporous material that presents a high
surface area value. The narrow porosity is predominant, as indi-
cated by the fact that Smic ꢀ SBET, which points out the diffusional
restrictions of N2 molecules to the interior of the narrowest micro-
porosity. However, the TiO2 phase presents smaller surface area
values but a significantly higher mesoporosity (Vmes). Therefore,
the microporosity is progressively widening when increasing the
TiO2 content, the volume of narrow micropores (W0 (CO2)) pro-
gressively decreases and the mean micropore size (L0 (CO2))
remains practically constant at around 0.6 nm in all the composites
(as in the carbon support), although (L0 (N2)) progressively
increases.
The mesopore size distribution (PSD) for the composites,
obtained by the application of the BJH method to the correspond-
ing N2-adsorption isotherms, is shown in Fig. 2a. It is noteworthy
that samples with TiO2 content between 20 and 40% present a
main maximum for mesopores at around 16.8 nm. The composites
with smaller (CTi10) or higher (CTi80) TiO2-content present a PSD
with the corresponding maxima shifted to smaller pores. Increas-
ing the% TiO2, the total porosity (V0.95) of the composites is
favoured (Table 2) which is mainly induced by the development
of mesoporosity (Fig. 2a and b), but this strongly decreases at a
high TiO2 content (CTi80), probably also induced by the rutile
formation.
The pore structure of the composites was analysed by compar-
ing the results from CO2 and N2-adsorption experiments. Results
Table 1
Composition, TiO2 crystal size and pHpzc of supports.
Sample
TiO2 (wt.%)
Crystal size (DRX)
pHpzc
Carbon
CTi10
CTi20
CTi30
CTi40
CTi50
CTi80
TiO2
0
–
–
–
3.4
4.3
6.7
8.0
24.6
9.4
7.9
7.8
7.3
7.1
7.0
6.4
4.9
15
22
34
45
54
84
100
Therefore, the mesopore volume increases more or less linearly
with the TiO2 content (up to CTi50, Fig. 2b) while the micropore
volume (determined by N2 or CO2 adsorption) shows the opposite
tendency (Fig. 2c and d). In this sense, the micropore size (L0) also
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

4
E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
Table 2
Textural characteristics of supports.
Carbon
SBET (m2 gꢁ1
)
W0 (N2) (cm3 gꢁ1
)
W0 (CO2) (cm3 gꢁ1
)
L0 (N2) (nm)
L0 (CO2) (nm)
Vmes (cm3 gꢁ1
)
Smic (CO2) (m2 gꢁ1
)
V0.95 (cm3 gꢁ1
)
C
614
523
480
481
586
507
176
116
0.245
0.206
0.190
0.192
0.227
0.163
0.070
0.047
0.285
0.223
0.205
0.176
0.174
0.144
0.057
0.054
0.78
0.87
0.76
0.97
1.21
1.22
1.20
1.89
0.60
0.58
0.56
0.58
0.62
0.61
0.62
0.68
0.060
0.116
0.203
0.242
0.438
0.392
0.081
0.272
952
768
737
608
562
473
185
159
0.305
0.322
0.393
0.434
0.665
0.555
0.151
0.326
CTi10
CTi20
CTi30
CTi40
CTi50
CTi80
Ti
1.2
1.0
0.8
0.6
0.4
0.2
0.5
a
b
0.25
0.0
1
0
0
50
100
10
100
D (nm)
% TiO2
0.3
0.2
0.1
0.3
0.2
0.1
d
c
0
0
0
0
50
100
50
100
% TiO2
% TiO2
Fig. 2. (a) Pore size distribution of supports: CTi10 ( ), CTi20 ( ), CTi30 ( ), CTi40 ( ) and CTi80 ( ) and evolution of the corresponding pore volumes (b) Vmeso, (c) W0 (CO2)
and (d) W0 (N2) with TiO2 content.
progressively increases and consequently, with the micropore
widening, the commented diffusional restriction decreases and
therefore also the difference between the Smic and SBET values.
Pretreated catalysts were also characterized before reaction,
following the same procedure previously described for the sup-
ports. Textural properties and pHpzc are summarized in Table 3.
All the pretreated catalysts present a smaller pHpzc value than their
respective supports; nevertheless, these values also slowly
decrease when increasing the TiO2 content.
It is also noteworthy that the surface values of the catalysts are
significantly smaller than those of their corresponding supports.
These results are obviously a consequence of the partial blockage
of the support porosity by the Pt-nanoparticles, as denoted by
the decrease in both micropore (W0 (N2)) and mesopore (Vmes) vol-
umes. Nevertheless, these differences strongly decrease when
increasing the TiO2 content of the composite (Table 3), in such a
way that the N2-adsorption isotherms of supports and catalysts
are quite coincident for TiO2 contents greater than 20% (Fig. 3).
Table 3
Textural characteristics and pHpzc of pretreated catalysts.
Catalyst
SBET (m2 gꢁ1
)
W0 (N2) (cm3 gꢁ1
)
L0 (N2) (nm)
Vmes (cm3 gꢁ1
)
V0.95 (cm3 gꢁ1
)
pHpzc
CPt3
531
315
359
352
483
533
154
111
0.21
0.62
1.21
0.95
1.23
1.33
1.42
1.51
1.62
0.000
0.077
0.146
0.300
0.427
0.355
0.107
0.397
0.220
0.200
0.290
0.440
0.620
0.567
0.169
0.442
8.8
5.9
5.8
5.7
5.6
5.5
4.9
3.6
CTi10Pt3
CTi20Pt3
CTi30Pt3
CTi40Pt3
CTi50Pt3
CTi80Pt3
TiPt3
0.123
0.144
0.140
0.193
0.212
0.062
0.045
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
5
CTi10
CTi40
1.6
0.4
0.2
0.0
0.8
0.0
0.0
0.5
1.0
0.0
0.5
1.0
P/P0
P/P0
Fig. 3. Nitrogen adsorption-desorption isotherms.
CTiX,
CTiXPt3. Open symbols: desorption, closed symbols: Adsorption.
Clearly, microporosity is preferentially blocked by the metallic
nanoparticles: L0 (N2) increases while W0 (N2) decreases. These
results suggest that Pt-nanoparticles are mainly located on the lar-
gest micropores and mesopores of the composite, blocking the nar-
rowest micropores from the carbon phase. Therefore, when using
composites with a high TiO2 content (>40%), similar PSD is
obtained regarding their supports (Fig. 3), due to the low propor-
tion of micropores in these samples.
The Pt-dispersion of pretreated catalysts was analysed by H2-
chemisorption, XRD and HRTEM. Results are summarized in
Table 4. Nevertheless, the results should be carefully analysed
because different tendencies can be described. Previous studies
[25] showed that the mesopore volume of the support contributes
to a higher Pt-dispersion. However, in this case, according to the
H2-chemisorption data, dispersion does not improve by increasing
the mesopore volume of the supports. On the contrary, even a
slight but progressive increase is observed of the Pt-particle size
when increasing the percentage of TiO2.
However, XRD seems to indicate that smaller particles are
formed when increasing the TiO2 content. When comparing the
XRD patterns of pretreated catalysts at 400 °C (Fig. 4), a decrease
in the intensity of the main Pt [111] diffraction peak is observed,
increasing the % of titania in the composite from 10 to 30% above
which no diffraction peaks (corresponding to the Pt-phase) are
observed. The observed improved Pt-dispersion could be related,
as commented, with the higher mesopore volume of the supports.
Nevertheless, the Pt-diffraction peaks were not observed when
using the composite CTi80 or pure TiO2 supports, which have a
low mesopore volume. Therefore, the different textural properties
do not explain this effect.
expected, because increasing the temperature should favour sin-
tering as observed by H2-chemisorption. However, again,
decrease in the intensity of the Pt [111] diffraction peak is clearly
observed regarding the same sample pretreated at 400 °C (Fig. 5),
therefore leading to a smaller Pt-particle size value determined
by the Scherrer equation.
a
In order to clarify the controversy between XRD and chemisorp-
tion results, samples were also analysed by HRTEM (Fig. 6). This
technique confirms a higher Pt-dispersion (smaller Pt-particle size)
increasing the amount of TiO2 in the carbon composite materials
(Fig. 6a and b). Small Pt-nanoparticles of around 2 nm were
detected at high TiO2 loadings. Ti and Pt distributions were anal-
ysed by EDX and the corresponding mapping (Fig. 6c and d) shows
that Ti is presented along the complete composite surface while Pt
is forming nanoparticles, although these nanoparticles are also
homogeneously distributed. The size of these Pt-nanoparticles is
higher on the CTi10 than on CTi40 or CTi80 composites.
A previous work [21] pointed out the formation of 3-D and crys-
talline Pt-particles on Pt/TiO2 catalysts after He-pretreatment.
However, Pt-XRD peaks were not observed after H2-pretreatment
at the same temperature. This fact was justified because the reduc-
tive character of the H2-atmosphere simultaneously induces the
partial reduction of the TiO2 phase and the formed oxygen vacan-
cies favour the mobility of Pt-species into the TiO2 structure. This
Pt [111]
Equally, when increasing the pre-treatment temperature of
TiC20Pt3 catalysts to 450 °C, an increase in Pt XRD peaks was also
Table 4
Mean Pt-particle size determined for pretreated catalysts by H2 chemisorption (dH ),
2
XRD or HRTEM.
Catalyst
QH (lmol/g)
2
dH (nm)
dXRD (nm)
dHRTEM (nm)
2
CPt3-8
10.5
12.8
12.0
9.0
8.9
8.5
8.1
7.7
9.0
7.9
6.5
6.9
9.2
9.3
9.7
10.3
10.7
9.2
9.3
8.2
5.1
4.5
–
1.3
–
1.6
2.2
3.6
CTi10Pt3-7
CTi20Pt3-7
CTi30Pt3-9
CTi40Pt3-9
CTi50Pt3-10
CTi80Pt3-10
CTi20Pt3-11a
TiPt3-9
25.3
11.3
4.9
n.d.
n.d.
n.d.
5.6
20
30
40
50
60
70
80
2
Θ
n.d.
Fig. 4. Influence of the TiO2 content on the XRD-patterns of catalysts pretreated at
400 °C: CTi10Pt3-7 ( ), CTi20Pt3-7 ( ), CTi30Pt3-9 ( ), CTi40Pt3-9 ( ), CTi50Pt3-
10 ( ), CTi80Pt3-10 ( ), TiPt3-9 (j).
(n.d.) no XRD peaks detected.
a
After pretreatment at 450 °C.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

6
E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
fact avoids sintering on the support surface and the formation of 3-
Pt [111]
D crystallites that could be observed by XRD, but simultaneously
also decreases the Pt-concentration on the surface, therefore
decreasing the amount of H2-chemisorbed during the characteriza-
tion experiments.
A similar effect is observed in the present catalyst series. Using
composites with low TiO2-content as supports, 3D and crystalline
Pt-particles are observed by XRD or HRTEM. They are formed
mainly on the predominant carbon phase, as denoted by the textu-
ral and Pt-dispersion analysis. When increasing the TiO2 loading
above 20–30%, the formation of small Pt particles is favoured. In
addition, the vacancies concentration and the mobility of the Pt-
species increase with the TiO2 content and the pre-treatment tem-
perature, favouring the progressive migration of Pt species into the
TiO2 structure. This fact reduces the H2-chemisorption capacity of
the samples, leading to a false increase in the Pt-particle size values
determined by this technique, in spite of the fact that smaller par-
ticles are formed.
20
30
40
50
2Θ
60
70
80
The surface chemistry of pretreated catalysts was analysed by
XPS. Results are collected in Table 5 and Fig. 7. Three components
Fig. 5. Influence of the pretreatment temperature on the XRD-patterns of CTi20Pt3
catalyst. 400 ( ) 450 °C ( ).
Fig. 6. HRTEM images of (a) CTi10Pt3-7, (b) CTi80Pt3-10 and EDX mapping, (c) CTi10Pt3-7 and (d) CTi40Pt3-9.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
7
Table 5
Pt-surface concentration and distribution of Pt-species determined by XPS.
Catalyst
BE (eV) Pt0
Pt (0) (%)
BE (eV) Pt2+
Pt (II) (%)
PtXPS (wt.%)
OXPS (wt.%)
CXPS (wt.%)
TiXPS (wt.%)
CPt3-8
71.3
71.6
71.5
71.6
71.6
71.6
71.5
71.3
71.4
70
67
67
72
70
70
68
66
73
72.3
72.5
72.5
72.6
72.7
72.7
72.6
72.3
72.3
30
33
33
28
30
30
32
34
27
4.38
1.71
1.39
1.61
1.19
1.07
1.51
3.60
1.95
2.54
3.80
3.99
5.92
5.86
9.80
14.27
16.89
30.02
93.08
93.30
92.06
81.78
83.57
72.10
62.38
56.05
0.00
0.00
1.19
2.56
10.69
9.38
17.02
21.83
23.47
68.03
CTi10Pt3-7
CTi20Pt3-7
CTi20Pt3-11
CTi30Pt3-9
CTi40Pt3-9
CTi50Pt3-10
CTi80Pt3-10
TiPt3-9
Ti3+
Ti3+
TiPt3
TiPt3
CTi40Pt3
CTi30Pt3
CTi40Pt3
CTi30Pt3
CTi20Pt3
CTi20Pt3
CTi10Pt3
CTi10Pt3
536
533
530
527
467
463
459
455
BE (eV)
BE (eV)
Fig. 7. Survey and deconvolution of the XPS Ti2p (left) and O1s (right) spectral regions. Line: BE of Ti3+ or oxygen linked to Ti3+, dashed line for composites.
at binding energy BE = 459.2 eV, 457.9 eV and 456.5 eV are used for
the deconvolution of the Ti2p spectral region of catalyst prepared,
using pure TiO2 support (TiPt3) which according to the bibliogra-
phy [26] correspond to Ti4+, Ti3+ and Ti2+ species, respectively
(Fig. 7). However, only two peaks were needed to fit this Ti2p spec-
tral region for catalysts supported on composites samples. The first
one, at 459.3 eV corresponds to Ti4+ as previously described, but
the band at 458.6 eV, which should be assigned to Ti3+, appears
therefore at higher BE (at around 0.7 eV) than on pure titania.
These results point out that Ti-species distribution is quite dif-
ferent in the composites (Ti2+ is not detected) and that these spe-
cies are located in a significantly different environment (with
different BE). Moreover, while in the case of TiPt3-9 catalyst 70%
of Ti atoms remain unreduced (as Ti4+) after treatments, this per-
centage decreases until only 30% for CTi10Pt3-7 catalyst. This fact
can be related to the low Ti-content of this support and the reduc-
ing character of the carbon matrix, which favours the extent of the
Ti-reduction during the pretreatment. The percentage of reduced
Ti4+ quickly increases on higher TiO2 content in the composite
(60% for CTi50Pt3-10 and 66% for CTi80Pt3-10).
Four components were used to fit the O1s spectral region for all
samples. In the case of TiPt3-9 the component at 530.7 eV corre-
sponds to the ‘bulk’ oxygen atom in the stoichiometric TiO2, and
the peaks at 529.3 and 527.9 eV were assigned to oxygen linked
to reduced species (Ti3+ and Ti2+, respectively) [27]. The high BE
component, at 532.2 eV, could correspond to various oxygen-
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

8
E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
containing surface functional groups. Mainly hydroxyl (AOH)
groups were described [28]. When using composite materials as
supports, the oxygen distribution species is also influenced by
the oxygen content linked to the carbon phase, clearly differenti-
ated because they appeared at higher BE. The peaks at 532.4 and
533.7 eV are assigned to C@O and CAO of carbon phase, respec-
tively [29]. These peaks are predominant in the composites with
low TiO2 content but with increasing this parameter, the main
components are located at 530.2 and 531.6 eV associated in this
case with oxygen bonded to Ti3+ and Ti4+ species, respectively
(Fig. 7). Fig. 7 shows how the BE values of Ti3+ and oxygen linked
to Ti3+ in the composites are around 0.7–1.0 eV higher than those
detected for pure titania, which can be due to the ability of the car-
bon phase to accommodate electrons from the inorganic phase
leading to a poorest electronic environment.
The XPS analyses also show the surface concentration and the
chemical state of Pt-species (Table 5). A mixture of Pt0 and Pt2+
with about 70% Pt0, was detected in all samples, denoting an insuf-
ficient pretreatment or the partial reoxidation of the samples dur-
ing storage. Regarding the surface Pt concentration (PtXPS, wt.%), a
smooth decrease in the Pt concentration with the increase of TiO2
loading in the composite, is observed. Usually, this parameter
depends on the Pt-particle size (dispersion) and Pt-distribution
inside the pores, both effects causing a decrease in the amount of
surface Pt detected by XPS. Therefore, the Pt-content decrease
could be related to a bigger particle size, but both XRD and HRTEM
prove the contrary behaviour. Therefore, this fact could be due to
the presence of Pt-particles inside the mesopores. Nevertheless,
in spite of that the mesoporosity range should be easily reached
by H2 during chemisorption experiments, the H2-chemisorption
is quite low according to the Pt-content detected. Therefore, the
most plausible explanation to justify the PtXPS decrease is not sin-
tering or localization inside the pores, but results show again that
Pt-species move into the TiO2 structure through the oxygen vacan-
cies, decreasing the H2-adsorption and PtXPS % in spite of the fact
that small Pt-particles are actually formed on the surface.
3.2. Catalytic performance
Fig. 8a and b (in order or clarity) shows the evolution of the con-
version values obtained with the different catalysts as a function of
the reaction time. Clearly, a slower and less active catalyst is
obtained depositing the Pt on the pure carbon phase. Although
the Pt-activity is favoured on pure TiO2 regarding the pure carbon
phase (reaching total citral conversion after 6 h of reaction), faster
hydrogenation processes are always obtained using composites as
Pt-supports. With increasing the TiO2 content of carbon-TiO2 com-
posites up to CTi40Pt3 catalyst, the reaction rate becomes progres-
sively faster (Fig. 8a). However, catalysts prepared by using
composites with a higher TiO2 content, become progressively
slower (Fig. 8b) in such a way that finally composite CTi80
and pure TiO2 present a similar behaviour (CPt3 < CTi10Pt3 <
CTi20Pt3 < CTi30Pt3 < CTi40Pt3 > CTi50Pt3 > CTi80Pt3 ꢃ TiPt3).
When comparing the performance of the catalysts in terms of
selectivity to unsaturated alcohols (SUA), Fig. 8c and d, the initial
SUA values vary in a similar way with the conversion, i.e. they pro-
gressively increase from 0% to 40% TiO2 in the composites and then,
progressively decrease, in such a way that CTi80 and pure TiO2 pre-
sent again a similar behaviour. These results point out clearly a
100
80
100
80
60
60
40
40
20
20
b
a
0
0
0
120
240
360
0
120
240
360
t (min)
t (min)
100
80
60
40
20
0
100
80
60
40
20
0
d
c
0
20
40
60
80
100
0
20
40
60
80
100
C (%)
C (%)
Fig. 8. Comparison of the catalytic performance of sample CPt3 ( ), CTi10Pt3 ( ), CTi20Pt3 ( ), CTi30Pt3 ( ), CTi40Pt3 ( ), CTi50Pt3 ( ), CTi80Pt3 ( ) and TiPt3 ( ).
Evolution of conversion with the time of reaction (a, b) and variation of the selectivity to unsaturated alcohols with increasing conversion (c, d). Reaction conditions: citral in
heptane solution (0.05 M), 500 mg of catalyst, 1500 rpm, T = 90 °C and hydrogen pressure = 8.3 bar.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
9
synergetic effect between both carbon and TiO2 phases on the Pt-
activity and selectivity. Therefore, conversion and selectivity
increase with the CTi10 composite regarding pure carbon, but lar-
ger carbon contents are necessary in the composite to improve the
catalytic performance regarding pure TiO2. Clearly, the best cat-
alytic performance is obtained when the TiO2 proportion in the
carbon-TiO2 supports is around 40% (Fig. 9).
concentration in surface determined by XPS analysis (Table 5), is
greater for CPt3 and CTi80Pt3 samples, but however, these samples
are not the most active or selective catalysts.
The environment of Pt-particles and, consequently, the catalytic
performance are related to the TiO2 content of the support, deter-
mining the Pt-support interactions. After thermal treatment in H2-
flow, the partial reduction of the TiO2 phase with formation of oxy-
gen vacancies occurs, which hinders the Pt-sintering but favours
the diffusion of Pt-species into the TiO2 structure. This fact causes
an intimate contact between supports and the Pt-active phase,
inducing a strong SMSI effect. Therefore, in spite of the loss of
accessible Pt-sites, this generates highly active and selective Pt-
TiO2 sites [21]. These sites obviously cannot be formed when using
the pure carbon xerogel. Lower conversion values are attached in
spite of the catalysts presenting an optimal Pt-particle size
(Table 4). Using composites with low amount of TiO2 (i.e. 10% or
20%), large Pt-agglomerates are still obtained (observed by XRD),
which mainly would be localized on the main carbon phase. Nev-
ertheless, both conversion and selectivity increase because the
low Ti-content of these samples (the small TiO2-particle size and
the reducing character of the carbon matrix) permits a large reduc-
tion of the TiO2 phase (around 70% of Ti atoms). Increasing the
amount of TiO2 up to 40%, the number of this highly selective
and active sites increases, and therefore, the higher are the activity
and selectivity. However, at a high TiO2 content (i.e. 80%), the TiO2
reducibility decreases due to the larger crystals formed and the
rutile formation, decreasing therefore the interaction of C-TiO2.
The Pt-environment depends therefore on the support composition
(carbon/TiO2 ratio), TiO2 crystal structure and the degree of reduc-
tion (formation of vacancies). The electronic d+ charge associated
with the oxygen vacancies and/or the significant increase in the
BE of Pt-species when supported on composites with an intermedi-
ate TiO2 content, can favour the electrophilic adsorption of citral
molecules (through the C@O bond), inducing a chemoselective
character to the UA formation. Hydrogen is adsorbed and dissoci-
ated on the small Pt-nanoparticles of this samples providing a high
concentration of atomic hydrogen that therefore favour the C@O
hydrogenation to unsaturated alcohols.
This different catalytic behaviour is obviously a consequence of
the different characteristics of the corresponding Pt-active sites. It
is well known that the hydrogenation of
a-b unsaturated aldehy-
des is strongly dependent on Pt-dispersion [30–32], i.e. citral
hydrogenation is a structure sensitive reaction. In previous works
[33] the influence of Pt particle size in the selective hydrogenation
of citral has been studied. An increase in the Pt particle size up to
around 8 nm produces an increase in both activity and selectivity;
however, larger particles cause the contrary effect. On the basis of
these results, and considering only the Pt-particle size, a decrease
in activity and selectivity increasing the TiO2 amount should be
expected due to the increase in Pt-dispersion in that sense. How-
ever, both activity and selectivity increase up to around 40% of
TiO2 in the composite. Consequently, in this catalyst series, disper-
sion is not the main factor to take into account; the main factor is
the synergetic effects between both organic and inorganic phases
and the different Pt-support interactions. In fact, the Pt-
140
120
100
80
100
80
60
40
20
0
60
40
This behaviour is also corroborated when analysing the influ-
ence of the pretreatment temperature of the CTi20Pt3 sample on
the catalytic performance (Fig. 10). As previously commented,
the increasing temperature produces a Pt-particle size decrease
but the reduction of the TiO2-phase and the mobility of the Pt-
species are favoured. This fact leads to a significant increase in
activity and selectivity.
20
0
0
50
100
(%)
TiO2
Fig. 9. The influence of the support composition on the activity (conversion at
90 min of reaction) and selectivity (at 70% of isoconversion) of supported Pt-
catalysts.
In this sense it is also interesting to analyse the evolution of the
product distribution obtained with the different Pt-catalysts along
100
80
100
80
60
60
40
40
20
b
0
20
a
0
0
50
100
0
60
120
t (min)
C (%)
Fig. 10. Influence of the pretreatment temperature: 400 °C ( ) and 450 °C ( ) on the conversion (a) and selectivity (b) of CTi20Pt3 catalyst.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

10
E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
TiPt3-9
CPt3-8
60
40
20
0
60
40
20
0
0
25
50
75
100
0
25
50
75
C (%)
C (%)
CTi10Pt3-7
CTi40Pt3-9
60
60
40
20
0
40
20
0
0
25
50
75
100
0
25
50
75
100
C (%)
C (%)
Fig. 11. Evolution of the selectivity values to each product as a function of the conversion reached. Product distributions: Citronellal ( ), Nerol ( ), Geraniol ( ), Citronellol
), 3,7 DMO ( ). Reaction conditions: citral in heptane solution (0.05 M), 500 mg of catalyst, 1500 rpm, T = 90 °C and hydrogen pressure = 8.3 bar.
(
the reaction progress, i.e. as the conversion increases (Fig. 11). It is
noteworthy that at the beginning of the reaction, geraniol is the
main product obtained in all cases except for the CPt3 catalysts.
The higher reactivity of geranial, regarding neral in the citral
racemic mixture, was previously pointed out [21,33] as a conse-
quence of a smaller steric impediment. It should be noted that
when using pure carbon as a Pt-support, CPt3 initially produces
citronellal as a main product and confirms that in this case the
citral adsorption and hydrogenation is carried out basically
through the C@C bonds, while the presence of TiO2 strongly
favours the interaction with the C@O bond. The active sites pro-
ducing citronellal on the CPt3 catalyst are however quickly deacti-
vated, increasing the selectivity to UA. This behaviour is also
observed on the TiPt3 catalysts at short reaction times, where
the UA formation is favoured at the expense of the citronellal
decrease. At high conversion values however, the formation of
citronellol by hydrogenation of the previously formed UA, is
favoured due to the low citral concentration.
Using carbon xerogel-TiO2 composites, the product distribution
strongly changes regarding CPt3 even at a low TiO2 content. In the
case of CTi10Pt3, geraniol becomes the main reaction product from
the reaction beginning, decreasing the citronellal that is trans-
formed into citronellol along the reaction; however, the hydro-
genation of UA seems to be negligible except at high conversion
values. Therefore, in the case of the best catalyst prepared,
CTi40Pt3, UA (nerol + geraniol) are obtained selectively; only at
conversion values higher than 80% UA are hydrogenated to
citronellol.
4. Conclusions
Carbon xerogels-TiO2 composites with different content ratios
were successfully prepared. These materials were used as Pt-
supports to develop specific hydrogenation catalysts for the con-
version of citral into derivative unsaturated alcohols (nerol
+ geraniol). Reference catalysts using pure phases as supports were
also used.
All the textural, chemical, crystallographic and catalytic proper-
ties are determined by the TiO2 content in the composite and by
the synergetic effect between both phases. The carbon xerogel is
eminently a microporous support, but the composites are meso-
porous materials, increasing mesoporosity at the expense of micro-
porosity as increasing the TiO2 content up to 40 wt.%. The presence
of carbon in the composite prevents the TiO2 crystal growth, and
difficults the anatase to rutile transition. Only composites with a
TiO2 content of 80 wt.% present a significant rutile phase. After
impregnation and H2-pretreatment, the partial reduction in the
titania phase and certain pore blockage by the Pt-nanoparticles is
observed. These nanoparticles are smaller when increasing TiO2
content, and complementary analysis techniques are necessary in
order to obtain a correct interpretation of the results.
Both H2-chemisorption and Pt-content are determined by XPS
decrease (suggesting sintering) but the XRD peaks corresponding
to the Pt-phase decrease and even disappear by increasing the
TiO2 content or treatment temperature. The fact that Pt-
dispersion is favoured in this sense was also confirmed by HRTEM.
Therefore, the formation of vacancies during pretreatment favours
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026

E. Bailón-García et al. / Journal of Catalysis xxx (2016) xxx–xxx
11
[10] P.F. Qu, J.G. Chen, Y.H. Song, Z.T. Liu, Z.W. Liu, Y. Li, J. Lu, J. Jiang, Catal.
Commun. 68 (2015) 105.
[11] E. Bailón-García, F.J. Maldonado-Hódar, A.F. Pérez-Cadenas, F. Carrasco-Marín,
Catalysts 3 (2013) 853.
the mobility of Pt-species inside the TiO2-structure, avoiding sin-
tering and improving the metal-support interactions, as demon-
strated according to the BE shifting of the XPS spectra.
[12] A. Giroir-Fendler, D. Richard, P. Gallezot, Stud. Surf. Sci. Catal. 41 (1988) 171.
[13] E. Bailón-García, F. Carrasco-Marín, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar,
Appl. Catal. A: Gen. 482 (2014) 318.
[14] E. Bailón-García, F. Carrasco-Marín, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar,
Appl. Catal. A: Gen. 512 (2016) 63.
[15] H.J. Jiang, H.B. Jiang, D.M. Zhu, X.L. Zheng, H.Y. Fu, H. Chen, R.X. Li, Appl. Catal.
A: Gen. 445–446 (2012) 351.
[16] A. Vannice, U.K. Singh, Abstr. Pap. Am. Chem. Soc. 223 (2002) U437.
[17] J. Alvarez-Rodríguez, I. Rodríguez-Ramos, A. Guerrero-Ruiz, E. Gallegos-Suarez,
A. Arcoya, Chem. Eng. J. 204–205 (2012) 169.
The combination of oxygen vacancies-Pt was suggested as
active and selective sites for the citral hydrogenation. The product
distribution significantly changes with the support composition.
The adsorption and hydrogenation of citral through the C@C bonds,
forming citronellal, are favoured in the beginning of the reaction
when using pure carbon as support. However, using composite
supports, UA were always the main products. The catalysts devel-
oped using a carbon composite containing around 40 wt.% of TiO2,
permit the specific hydrogenation of C@O bonds transforming
citral into UA.
[18] A.B. Dongil, L. Pastor-Pérez, J.L.G. Fierro, N. Escalona, A. Sepúlveda-Escribano,
Appl. Catal. A: Gen. 513 (2016) 89.
[19] E. Bailón-García, F. Carrasco-Marín, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar,
Catal. Commun. 58 (2014) 64.
[20] F.J. Maldonado-Hódar, C. Moreno-Castilla, A.F. Pérez-Cadenas, Appl. Catal. B:
Environ. 54 (2004) 217.
Acknowledgments
[21] E. Bailón-García, F. Carrasco-Marín, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar,
J. Catal. 327 (2015) 86.
[22] E. Gallegos-Suárez, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar, F. Carrasco-
Marín, Chem. Eng. J. 181–182 (2012) 851.
[23] D. Hanaor, C. Sorrell, J. Mater. Sci. 46 (2011) 855.
[24] D. Cazorla-Amorós, J. Alcañiz-Monge, A. Linares-Solano, Langmuir 12 (1996)
2820.
EBG acknowledges for a pre-doctoral fellowship to the MCINN
project CTM2010-18889. This research has been supported by
the Spanish projects P12-RNM-2892 and CTQ2013-44789-R, from
Andalucia’s government and MINECO, respectively.
[25] S. Morales-Torres, F.J. Maldonado-Hódar, A.F. Pérez-Cadenas, F. Carrasco-
Marín, J. Hazard. Mater. 183 (2010) 814.
References
[26] A.R. González-Elipe, A. Fernández, J.P. Espinós, G. Munuera, J. Catal. 131 (1991)
51.
[27] M.E. Nagassa, A.E. Daw, W.G. Rowe, A. Carley, D.W. Thomas, R. Moseley, Clin.
Oral Implan. Res. 19 (2008) 1317.
[28] A. Samokhvalov, E.C. Duin, S. Nair, B.J. Tatarchuk, Surf. Interface Anal. 42
(2010) 1476.
[29] J.M. Jiménez Mateos, J.L.G. Fierro, Surf. Interface Anal. 24 (1996) 223.
[30] F. Delbecq, P. Sautet, J. Catal. 152 (1995) 217.
[1] M. Eisenacher, S. Beschnitt, W. Hölderich, Catal. Commun. 26 (2012) 214.
[2] J. Groling, Ullmann’s Encyclopedia of Industrial Chemistry, Weinheim, 2003.
[3] H. Surburg, J. Panten, Common Fragrance and Flavor Materials: Preparation,
Properties and Uses, WILEY-VCH, Weinheim, Germany, 2006.
[4] P. Barbaro, F. Liguori, Heterogenized Homogeneous Catalysts for Fine
Chemicals Production, Springer, London, 2010.
[5] W. Li, G. Fan, L. Yang, F. Li, Catal. Sci. Technol. 6 (2016) 2337.
[6] A.R. Rautio, P. Mäki-Arvela, A. Aho, K. Eränen, K. Kordas, Catal. Today 241
(2015) 170.
[31] M.G. Prakash, R. Mahalakshmy, K.R. Krishnamurthy, B. Viswanathan, Catal. Sci.
Technol. 5 (2015) 3313.
[7] J. Aumo, S. Oksanen, J.P. Mikkola, T. Salmi, D.Y. Murzin, Ind. Eng. Chem. Res. 44
(2005) 5285.
[32] B.S. Takale, S. Wang, X. Zhang, X. Feng, X. Yu, T. Jin, M. Bao, Y. Yamamoto,
Chem. Commun. 50 (2014) 14401.
[8] T. Ekou, L. Ekou, A. Vicente, G. Lafaye, S. Pronier, C. Especel, P. Marecot, J. Mol.
Catal. A-Chem. 337 (2011) 82.
[33] E. Bailón-García, F. Carrasco-Marín, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar,
Catal. Commun. 82 (2016) 36.
[9] D. Spasyuk, C. Vicent, D.G. Gusev, J. Am. Chem. Soc. 137 (2015) 3743.
Please cite this article in press as: E. Bailón-García et al., Chemoselective Pt-catalysts supported on carbon-TiO2 composites for the direct hydrogenation of
citral to unsaturated alcohols, J. Catal. (2016), http://dx.doi.org/10.1016/j.jcat.2016.09.026