Preparation of carbon aerogel supported platinum catalysts for the selective hydrogenation of cinnamaldehyde

Article abstract of DOI:10.1016/j.apcata.2012.03.016

The selective hydrogenation of cinnamaldehyde is investigated using platinum catalysts supported on carbon aerogels with different textural and chemical properties. Despite the large amount of oxygenated surface groups introduced after the oxidation step, the original porosity of the carbon aerogels is maintained. The oxidation treatments performed on the materials are found to strongly influence the surface chemistry which in turn affects the Pt dispersion, yielding larger metal particles after the chemical modifications and the H₂ pre-treatment. The presence of mesopores and the increase of the acidic character in the carbon aerogels lead to a higher catalytic activity and selectivity towards cinnamyl alcohol when compared with that obtained for the untreated materials. A thermal treatment at 973 K is found to favor the hydrogenation of the olefinic bond when using carbon aerogels, due to the remaining oxygenated surface groups, at variance with other previously reported carbon supports (xerogels and nanotubes).

Full text of DOI:10.1016/j.apcata.2012.03.016

Applied Catalysis A: General 425–426 (2012) 161–169
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Preparation of carbon aerogel supported platinum catalysts for the selective
hydrogenation of cinnamaldehyde
Bruno F. Machadoa,∗, Sergio Morales-Torresa,b,∗∗, Agustín F. Pérez-Cadenasb,
Francisco J. Maldonado-Hódarb, Francisco Carrasco-Marínb, Adrián M.T. Silvaa,
José L. Figueiredoa, Joaquim L. Fariaa
a LCM – Laboratory of Catalysis and Materials – Associate Laboratory LSRE/LCM, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal
b Departamento de Química Inorgánica, Facultad de Ciencias, Universidad de Granada, Campus 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:
The selective hydrogenation of cinnamaldehyde is investigated using platinum catalysts supported on
carbon aerogels with different textural and chemical properties. Despite the large amount of oxygenated
surface groups introduced after the oxidation step, the original porosity of the carbon aerogels is main-
tained. The oxidation treatments performed on the materials are found to strongly influence the surface
chemistry which in turn affects the Pt dispersion, yielding larger metal particles after the chemical mod-
ifications and the H2 pre-treatment. The presence of mesopores and the increase of the acidic character
in the carbon aerogels lead to a higher catalytic activity and selectivity towards cinnamyl alcohol when
compared with that obtained for the untreated materials. A thermal treatment at 973 K is found to favor
the hydrogenation of the olefinic bond when using carbon aerogels, due to the remaining oxygenated
surface groups, at variance with other previously reported carbon supports (xerogels and nanotubes).
© 2012 Elsevier B.V. All rights reserved.
Received 4 November 2011
Received in revised form 27 February 2012
Accepted 6 March 2012
Available online 15 March 2012
Keywords:
Carbon aerogel
Surface chemistry
Platinum
Selective hydrogenation
Cinnamaldehyde
1. Introduction
structure and consequently the porous texture of the organic aero-
gels. In this context, meso- or macroporous carbon aerogels can be
Following the first synthesis of carbon aerogels reported by
Pekala [1] other types of carbon gels have been reported based on
the use of different solvent removal steps in the synthetic route
[2]: (i) aerogels, when supercritical CO2 is used, (ii) xerogels, when
the removal takes place under ambient temperature and pressure
conditions, and (iii) cryogels, in the case of using a freeze-drying
method. Carbon aerogels usually have surface areas between 400
and 1000 m2 g−1 and are promising materials for applications such
as adsorbents, catalysts or capacitors [3–8]. Their potential is based
on their unique properties: purity, homogeneity and above all, con-
trollable porosity. Different precursors and methods have been
developed over the years to produce highly porous carbon aero-
gels. The porous texture of carbon aerogels, and consequently their
applications, depend strongly on several experimental conditions.
The most important is the polymerization step, since it defines the
developed by simply modifying the polymerization catalyst, while
maintaining constant the other synthesis parameters [9].
Because the support plays an important role in catalyst design,
in order to optimize the dichotomy between metal and support,
the use of a material whose properties can be finely tuned is highly
desirable. Furthermore, due to the importance of the oxygenated
surface groups in the design of efficient catalysts it is advantageous
to have the possibility of easily modifying the surface chemistry
of the support materials to incorporate those groups [10,11]. Thus,
polymer-based carbon materials are an important category of
materials suitable to produce noble metal supported catalysts
[5]. This type of catalyst is extremely useful in heterogeneously
catalyzed reactions, namely for fuel cell applications [12,13] and
selective hydrogenation of ␣,␤-unsaturated aldehydes (acrolein,
crotonaldehyde, cinnamaldehyde, etc.) [14–17]. The latter is a
key process for the production of important intermediates in the
preparation of fine chemicals for fragrance, pharmaceutical and
agrochemical industries [18,19]. Unfortunately, there are some
thermodynamic and kinetic constrictions that limit the selectivity
towards the unsaturated alcohol formation [20], commonly the
most valuable intermediate compound. In spite of these draw-
backs, the selectivity to unsaturated alcohols using heterogeneous
catalysts can be improved by careful design of the catalyst. Several
∗
Corresponding author. Tel.: +351 225 081 582; fax: +351 225 081 449.
Corresponding author at: Departamento de Química Inorgánica, Facultad de
∗∗
Ciencias, Universidad de Granada, Campus Fuentenueva s/n, 18071 Granada, Spain.
Tel.: +34 958 243 235; fax: +34 958 248 526.
E-mail addresses: bmachado@fe.up.pt (B.F. Machado), semoto@ugr.es
(S. Morales-Torres).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2012.03.016

162
B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
works in the literature indicate that an increase of the metal
particle size favors the selectivity towards the unsaturated alcohol,
either due to steric or electronic effect between the molecule and
metal surface or due to the distribution of crystal surfaces and the
presence of certain edges and corners [17,21]. Another possibility
to control the selectivity and activity of hydrogenation is the design
of the catalyst support, i.e., by changing its porous texture, chem-
ical properties or by making thermal treatments that modify these
characteristics, or influence the state and particle size of the metal
[15,22]. In our previous works [14,16], mesoporous carbon xerogels
demonstrated to be excellent supports for noble metals (Pt, Ru and
Ir) in the selective hydrogenation of cinnamaldehyde. In a similar
context, Job et al. [23] prepared carbon xerogels with different
porous textures: micro- and macroporous and micro- and meso-
porous, which were used as supports of Pt catalyst and tested in the
hydrogenation of benzene. They concluded that these supported
catalysts were more active than others on active carbon and the
most active catalyst was prepared on a carbon xerogel with micro-
and mesopores, which also presented the highest Pt dispersion.
The hydrogenation of ␣,␤-unsaturated aldehydes is a structure-
sensitive reaction, i.e., the activity and selectivity of the catalyst are
influenced by the metal particle size and orientation [17]. There-
fore, a comprehensive study on how the porous texture and surface
chemistry influences the activity and selectivity of carbon aero-
gel supported Pt catalysts in the liquid-phase hydrogenation of
cinnamaldehyde is described in this work. In addition, a high tem-
perature thermal activation of the materials was also carried out
in order to determine its influence on the catalytic performance of
the supported catalysts.
microscope. Textural characterization of the supports and the
catalysts was carried out using N2 physical adsorption at 77 K
(Quantachrome Autosorb-1). The BET equation [25] and t-method
[26] were used for the analysis of the corresponding adsorption
isotherms, from which the BET surface area (SBET), the external
porous surface area (SEXT) and the micropore volume (VMIC) were
obtained. The BJH method [27] was applied to the desorption
branch of the N2 adsorption isotherms to obtain the pore size dis-
tributions curves in the case of mesoporous samples, while for
macroporous ones the Hg porosimetry (Quantachrome Autoscan
60) was used.
The surface chemistry of the carbon aerogels was characterized
by temperature programmed desorption (TPD) and pHPZC (point of
zero charge) measurements, accordingly to a previously reported
procedure [28]. The pHPZC values correspond to the equilibrium
pH of 250 mg of carbon dispersed in 4 mL of H2O at 298 K. TPD
experiments were carried out by heating the samples to 1273 K in
He flow (60 cm3 min−1) at a heating rate of 50 K min−1. The amount
of evolved gases was recorded as a function of temperature using
a quadrupole mass spectrometer (Balzers, model Thermocube), as
described elsewhere [29]. The oxygen content was calculated from
the amounts of CO and CO2 released during the TPD experiments.
Pt dispersion (DPt) was obtained by H2 chemisorption measure-
ments performed at 298 K. In addition, assuming the formation of
spherical particles and a H:Pt = 1:1 stoichiometry it was possible
to calculate the Pt average particle size, using dPt = 1.08/DPt (nm).
Pt particle size was also determined by high-resolution transmis-
sion electron microscopy (TEM). TEM experiments were carried out
using a Phillips CM-20 microscope equipped with an EDAX micro-
analysis system.
2. Experimental
2.1. Support preparation
2.4. Catalytic hydrogenation
Carbon aerogels were obtained by carbonization (at 1173 K) of
the corresponding organic aerogels, which were synthesized by
polymerization of resorcinol with formaldehyde in aqueous solu-
tion using alkali carbonates (M2CO3; M = Li or Cs) as polymerization
catalysts, according to a methodology described elsewhere [9].
These samples will be referred as CA-Li and CA-Cs indicating the
alkaline metal used in their synthesis.
Samples of CA-Li were further oxidized (1 g carbon/10 mL of
solution) with concentrated hydrogen peroxide (9.8 M) and with
a saturated solution of ammonium peroxydisulfate in sulfuric acid
(1 M) for 48 h at ambient temperature [24]. After oxidation, the
samples were washed with distilled water and dried at 393 K in an
oven during 24 h. The samples oxidized with H2O2 and (NH4)2S2O8
will be referred as CA-LiH and CA-LiS, respectively.
The liquid-phase hydrogenation of cinnamaldehyde was per-
formed according to a procedure described elsewhere [30]. Briefly,
the reaction mixture containing heptane, 0.14 g cinnamaldehyde,
decane (as internal standard for gas chromatography) and 0.20 g
of catalyst was bubbled with nitrogen several times to remove
traces of dissolved oxygen before being purged with hydrogen.
The temperature was set at 363 K and the reactor pressurized with
hydrogen to the desired 10 bar immediately before starting the
reaction.
The results of the reaction runs were analyzed in terms of cin-
namaldehyde (CAL) conversion (Eq. (1)) and product selectivity (Eq.
(2)), measured at 50% conversion:
CCAL,0 − CCAL
XCAL
=
(1)
(2)
CCAL,0
2.2. Catalyst preparation
Ci
CCAL,0 − CCAL
The materials described in the previous section were used as
supports to prepare 1 wt.% Pt catalysts. Incipient wetness impreg-
nation was used to deposit Pt(NH3)4(NO3)2 over the different
carbon aerogels, being the metal precursor previously dissolved
in an aqueous solution. Prior to reaction, the resulting catalysts
(Pt/CA-Li, Pt/CA-LiH, Pt/CA-LiS and Pt/CA-Cs) were treated in N2
during 4 h and reduced in H2 during 3 h at 573 K. A post-reduction
treatment (PRT, 2 h, N2) was performed at 973 K to remove part of
the surface groups (Pt/CA-Li973, Pt/CA-LiH973, Pt/CA-LiS973 and
Pt/CA-Cs973).
Si =
where Ci represents the concentration of cinnamyl alcohol (COL),
hydrocinnamaldehyde (HCAL), and hydrocinnamyl alcohol (HCOL).
To compare the activity exhibited by the catalysts, the turn-over
frequency (TOF) was calculated based on CAL disappearance:
molCAL converted
time(mcatyPt/MPt)DPt
TOF =
(3)
2.3. Catalyst characterization
where mcat, yPt, and MPt represent the catalyst mass, Pt load and
molar mass, respectively. DPt was determined based on TEM anal-
ysis.
The surface morphology of the aerogels was studied by scanning
electron microscopy (SEM) using a LEO (Carl Zeiss) GEMINI-1530

B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
163
Table 1
Surface oxidation in terms of evolved CO and CO2 (␮mol g−1), oxygen content (OTPD, wt.%) and acidity pHPZC of the original and treated at 573 K and 973 K carbon aerogels.
Sample
CO
20 ␮mol g−1
CO2
20 ␮mol g−1
OTPD
CO/CO2
pHPZC
( 0.1)
(
)
(
)
(
0.1 wt.%)
CA-Li
420
419
314
1110
1102
735
2846
2818
1734
534
61
50
5
335
270
28
1621
1330
112
59
0.9
0.8
0.5
2.9
2.9
1.3
9.7
8.9
3.1
1.0
1.0
0.7
7
8
64
3
4
26
2
2
16
9
12
137
10.2
n.d.
n.d.
4.9
n.d.
n.d.
3.2
n.d.
n.d.
10.3
n.d.
n.d.
CA-Li573
CA-Li973
CA-LiH
CA-LiH573
CA-LiH973
CA-LiS
CA-LiS573
CA-LiS973
CA-Cs
CA-Cs573
CA-Cs973
532
410
44
3
n.d., not determined.
3. Results and discussion
CA-LiH573 and CA-LiH973, respectively), due to their higher oxy-
gen content (9.7 wt.% in comparison with 2.9 wt.% for CA-LiH) and
the presence of stronger acid surface groups. Therefore, a less acidic
surface is also obtained for carbon aerogel oxidized with H2O2 and
treated at 573 K or 973 K. The surface acidity of the materials was
also seriously affected by the oxidation procedure. The acid charac-
ter of the samples, determined by pHPZC measurements (Table 1),
was found to increase in the following order: CA-LiS (3.2) > CA-LiH
(4.9) > CA-Li ≈ CA-Cs (10.2). In addition, the surface oxygen content
3.1. Support functionalization
The surface chemistry of carbon materials is mainly determined
by the acidic and basic character of their surface, which can be mod-
ified by treating the material with different oxidizing agents [24,31]
and by submitting it to thermal treatments under inert atmosphere.
It is commonly accepted that the oxygenated surface groups that
contribute to the acidic character of a carbon material are carboxylic
acids, anhydrides, lactones and phenols [32]. Regarding the basic
groups some restraints arise concerning the strength and extent
of their contribution to the overall carbon basicity. Nevertheless,
groups such as chromene structures, quinone, pyrone-like groups
and ␲-electrons of the carbon basal planes have been suggested
[32].
determined by TPD was found to increase with decreasing pHPZC
,
i.e., CA-LiS (9.7 wt.%) > CA-LiH (2.9 wt.%) > CA-Li ≈ CA-Cs (1 wt.%).
The introduction of these oxygenated surface groups was
achieved without any significant changes of the initial textu-
ral properties of the carbon aerogels synthesized with Li2CO3: a
decrease of only ca. 4% was observed in both SBET and SEXT values
(Table 2), while the micropore volume and the pore size distri-
bution (dpore) remained practically unchanged. The difference in
porous texture between CA-Li and CA-Cs (meso- and macroporous
materials, respectively), is explained in terms of the polarizing
power of the alkaline cation during the polymerization of resorcinol
and formaldehyde [9].
The changes in the surface chemistry of the carbon aerogel
induced by oxidation treatments with H2O2 and (NH4)2S2O8 were
followed by TPD (Fig. 1 and Table 1), and resulted in the forma-
tion of different oxygenated surface groups. Carboxylic acids (CAc,
released as CO2 at 550 K), carbonyls/quinones situated in different
energetically sites (CQ, released as CO at 1163–1253 K), ethers (Eth,
released as CO at 1073 K), lactones (Lac, released as CO2 at 923 K),
carboxylic anhydrides (CAn, released as CO and as CO2 at 773 K)
and phenol groups (Ph, released as CO at 973 K) were introduced
with both treatments [33,34], although higher amounts of more
acidic groups (evolved as CO2) were obtained with the (NH4)2S2O8
treatment, i.e., the CO/CO2 ratio decreased as follows: CA-Li < CA-
LiH < CA-LiS. The amount of oxygenated surface groups evolved
as CO, increased from 420 ␮mol g−1 for CA-Li (non-oxidized) to
2846 ␮mol g−1 for CA-LiS (oxidized with (NH4)2S2O8), which corre-
sponds to an increase of ca. 85%; an increase of ca. 96% was observed
for the oxygenated surface groups evolved as CO2. In the case of CA-
LiH (oxidized with H2O2), an increase of ca. 62% and 82% for CO and
CO2, respectively, was observed. Similar results were reported with
identical oxidation procedures using an activated carbon [24,31],
and the enhanced acidity using (NH4)2S2O8 is attributed to the
creation of stronger carboxylic groups close to other groups, such
as carbonyls and phenols (electron-withdrawing groups), which
enhance their acidity due to a negative inductive effect [24].
The surface chemistry of the carbon aerogels was also stud-
ied by TPD after thermal treatment with N2 at 573 K and 973 K
(Table 1), which are the temperatures used in order to reduce the
Pt particles and to remove some oxygenated surface groups in the
supported Pt catalysts. For all materials, a decrease of these groups
was reached, being more effective for those evolved as CO2 (mainly
acid groups) during TPD experiments. CA-LiS treated at different
temperatures presented always a lower CO/CO2 ratio (2 and 16 for
CA-LiS573 and CA-LiS973, respectively) than CA-LiH (4 and 26 for
3.2. Catalyst characterization
In general, BET, external surface areas and micropore vol-
umes (Table 3) decreased for Pt catalysts in comparison to the
corresponding supports. This is thought to be related to a partial
blockage of porosity induced by the Pt particles deposition. An
increase of BET surface area was also observed for all Pt catalysts
treated at 973 K. In addition, the percentage of increase is different
and increases with the oxygen surface content of the support. BET
surface area increased 18 and 42% for Pt/CA-LiH973 and Pt/CA-
LiS973, respectively, while only a 10% increase was observed
for Pt/CA-Li973. This can be explained in terms of pore opening,
produced as a consequence of the release of surface groups during
the N2 treatment at 973 K and confirmed by the decrease on the
surface oxygen (OTPD, Table 1) of the treated carbon aerogels. In
Table 2
Textural characterization of the different carbon aerogels. BET (SBET, m2 g−1) and
external (SEXT, m2 g−1) surface areas, micropore volumes (VMIC, cm3 g−1) and pore
dimension (dpore, nm).
Sample SBET
SEXT
(
VMIC
(
dpore
(
5 m2 g−1
)
5 m2 g−1
)
0.01 cm3 g−1
)
(
0.5 nm)
CA-Li
902
863
861
758
284
275
269
94
0.25
0.24
0.24
0.27
22
23
CA-LiH
CA-LiS
CA-Cs
22
90.0a
a
Determined by Hg porosimetry.

164
B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
6
5
4
3
2
1
Eth
Ph
(a)
CQ
CAn
0
CA-LiS
CA-LiH
CA-Li
273
473
673
873
1073
1273
T (K)
6
5
4
3
2
1
(b)
CAc
CAn
Lac
0
CA-LiS
CA-LiH
CA-Li
273
473
673
873
1073
1273
T (K)
Fig. 1. TPD spectra for the carbon aerogel CA-Li before and after oxidation with hydrogen peroxide (CA-LiH) and ammonium peroxydisulfate (CA-LiS): (a) CO and (b) CO2
desorption profiles.
fact, increases in the average pore size (23–27 nm) and in the
external surface area were also observed for these Pt catalysts.
Platinum dispersions and particle sizes of the carbon aerogel
supported catalysts are given in Table 4. Excellent Pt dispersions
over the untreated aerogels (Pt/CA-Li and Pt/CA-Cs) were observed,
but somewhat lower values were detected when using the oxidized
supports (Pt/CA-LiH and Pt/CA-LiS).
Comparing the Pt particle size obtained by H2 chemisorption
with that obtained by TEM, it can be observed that there is a general
agreement between these two techniques.
There are two main factors that can affect the metal dispersion
over a support: (i) the available surface area and (ii) the interac-
tion between the surface and the precursor solution. The textural
Table 3
Textural characterization of the Pt catalysts. BET (SBET, m2 g−1) and external (SEXT
m2 g−1) surface areas, micropore volumes (VMIC, cm3 g−1) and pore dimension (dpore
nm).
,
,
Table 4
Pt dispersion (DPt, in %) and particle size (in nm) determined by H2 chemisorption
(dPt) and TEM analysis (dPt†).
Catalyst
SBET
5 m2 g−1
SEXT
5 m2 g−1
VMIC
0.01 cm3 g−1
dpore
†
(
)
(
)
(
)
(
0.5 nm)
Catalyst
DPt
dPt
dPt
(
0.1%)
(
0.1 nm)
(
0.1 nm)
Pt/CA-Li
Pt/CA-Li973
Pt/CA-LiH
Pt/CA-LiH973 1069
Pt/CA-LiS
Pt/CA-LiS973
Pt/CA-Cs
Pt/CA-Cs973
882
967
908
275
289
274
313
242
324
90
0.25
0.27
0.26
0.31
0.23
0.33
0.25
0.30
23
26
23
27
23
27
n.d.
n.d.
Pt/CA-Li
Pt/CA-Li973
Pt/CA-LiH
Pt/CA-LiH973
Pt/CA-LiS
Pt/CA-LiS973
Pt/CA-Cs
74.1
38.1
46.1
30.5
44.4
28.7
87.3
42.5
1.5
2.8
2.3
3.5
2.4
3.8
1.2
2.5
1.4
2.8
2.5
2.6
2.1
2.2
1.5
2.4
801
1141
720
832
104
Pt/CA-Cs973
n.d., not determined

B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
165
Fig. 2. TEM micrographs of Pt/CA-Li and Pt/CA-LiH catalyst: (a and b) before and (c and d) after a post-reduction treatment at 973 K, respectively.
properties of the supports were not significantly changed by the
oxidation treatments, as already shown, but the presence of oxy-
genated surface groups increased the hydrophilic character of the
surface. Taking into consideration that the surface is negatively
charged (pHPZC < 10.5, pH of precursor solution) and the cationic
nature of the precursor ([Pt(NH3)4]2+), an improved metal dis-
persion should be expected when using the oxidized supports, in
comparison with the catalysts prepared using the untreated sup-
ports. However this is not the case. The explanation could rely
on the oxygenated surface groups, which act as anchoring sites,
retaining the Pt-precursor molecules at the entrance of the pores
and preventing the diffusion of Pt through the porous structure.
Furthermore, during the reduction treatment, the less stable oxy-
gen surface complexes (carboxylic acids groups) may be removed,
promoting the mobility of the platinum particles, thus favoring
their agglomeration [35]. The different porous texture observed for
Pt/CA-Li and Pt/CA-Cs does not appear to have a marked influence
over the Pt particle size, since both Pt catalysts presented similar
particle sizes (1.5 and 1.2 nm for Pt/CA-Li and Pt/CA-Cs, respec-
tively). Therefore, the low external surface area of Pt/CA-Cs (with
respect to Pt/CA-Li) seems to be sufficient to obtain very small and
accessible Pt particles.
When a post-reduction treatment is carried out in N2 at 973 K,
the experimental conditions are severe enough to induce sintering
of the Pt particles [14]. This effect was particularly more intense in
the catalysts with higher dispersion. After the thermal treatment,
all materials (independently of their surface chemistry) show Pt
particles with sizes between 2.2 and 2.8 nm (values determined by
TEM analysis). In addition to sintering related with the tempera-
ture effect, particle sizes could also be influenced by the surface
chemistry of the support materials, as the Pt particles supported in
untreated aerogels increased ca. 2 times (Fig. 2a and c), while for
H2
O
H2
OH
-H2O
Cinnamaldehyde (CAL)
Cinnamyl alcohol (COL)
β
-Methylstyrene (MS)
H2
2H2
H2
H2
H2
O
H2
OH
-H2O
Hydrocinnamyl alcohol (HCOL)
Hydrocinnamaldehyde (HCAL)
1-Propylbenzene (PB)
Fig. 3. Reaction network for the selective hydrogenation of cinnamaldehyde with carbon aerogel supported catalysts.

166
B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
100
100
80
60
40
20
0
(c)
(a)
,
,
,
,
Conversion of CAL
Selectivity to COL
Selectivity to HCAL
Selectivity to HCOL
,
,
,
,
Conversion of CAL
Selectivity to COL
Selectivity to HCAL
Selectivity to HCOL
80
60
40
20
0
0
60 120 180 240 300 360 420 480 540 600
Time (min)
0
60 120 180 240 300 360 420 480 540 600
Time (min)
(d) 100
(b) 100
,
,
,
,
Conversion of CAL
Selectivity to COL
Selectivity to HCAL
Selectivity to HCOL
,
,
,
,
Conversion of CAL
Selectivity to COL
Selectivity to HCAL
Selectivity to HCOL
80
80
60
40
20
0
60
40
20
0
0
60 120 180 240 300 360 420 480 540 600
Time (min)
0
60 120 180 240 300 360 420 480 540 600
Time (min)
Fig. 4. Product distribution for the selective hydrogenation of cinnamaldehyde using (a) CA-Li, (b) CA-LiH, (c) CA-LiS and (d) CA-Cs. Full and open symbols: before and after
thermal treatment in N2 at 973 K, respectively (CCAL,0 ≈ 12 mM, T = 363 K and P = 10 bar).
the oxidized supports no sintering effect was observed (Fig. 2b and
d) once the less stable oxygenated surface groups were removed.
As previously reported for multi-walled carbon nanotubes [30],
surface groups can act as anchoring sites and increase the ther-
mal stability of the catalysts while providing a stronger interaction
between the metal and the surface, resulting in improved catalytic
properties at the metal–support interface. Oxidized carbon aero-
gels could, thus, present a similar behavior.
Besides the sintering effect, the treatment at 973 K is able to
purge most of the CO2-releasing groups, but only a part of the oxy-
genated surface groups evolving as CO are removed as previously
indicated. Hence, groups like ethers, carbonyls or quinones, which
are only released at higher temperatures due to their stability [36],
should remain to some extent on the surface.
The reaction kinetics at 363 K and 10 bar (total pressure) are rep-
resented in Fig. 4 in terms of conversion of CAL and selectivity to
COL, HCAL and HCOL with time of reaction. In Table 5 it is given the
turn-over frequency (TOF), the time required to reach 50% conver-
sion of CAL (t50%), as well as the selectivities towards COL (SCOL),
HCAL (SHCAL), HCOL (SHCOL) and other by-products (SOTHERS), at 50%
conversion of CAL.
There are two effects, generally attributed to the support, which
can have a significant influence on the hydrogenation reaction:
steric and electronic interactions [37]. These effects can be related
with the different modes of adsorption/repulsion of cinnamalde-
hyde on the metal and/or the support surface. Taking into account
the kind of porosity of the materials tested and the experimen-
tal conditions used in the hydrogenation of cinnamaldehyde, the
Table 5
Turn-over frequency (TOF), time required to reach 50% conversion of CAL (t50%),
and selectivities towards COL, HCAL, HCOL and other products (measured at 50%
conversion of CAL).
3.3. Selective hydrogenation
The network of reactions observed for the selective hydro-
genation of cinnamaldehyde in heptane is shown in Fig. 3. The
thermodynamically preferred route goes through the hydrogena-
tion of the C C bond yielding the saturated aldehyde (HCAL).
Selective hydrogenation of the C O bond yields the unsaturated
alcohol (COL). Both COL and HCAL can be further hydrogenated to
produce the fully saturated alcohol (HCOL). Side-products involv-
ing the loss of the hydroxyl group, like ␤-methylstyrene (MS) and
1-propylbenzene (PB) were also detected in different amounts,
indicating a strong adsorption of the substrate over the metal sites
and a possible poisoning effect.
Catalyst
TOF
(s−1
t50%
(min)
SCOL
(%)
SHCAL
(%)
SHCOL
(%)
SOTHERS
(%)
)
Pt/CA-Li
Pt/CA-Li973
Pt/CA-LiH
Pt/CA-LiH973
Pt/CA-LiS
Pt/CA-LiS973
Pt/CA-Cs
1.6
3.4
1.6
3.5
2.3
4.4
0.9
2.8
64.1
27.3
36.2
26.8
30.2
19.5
102
11
12
53
21
36
21
24
12
54
54
18
41
23
44
34
54
20
25
20
28
25
27
35
19
15
9
9
10
16
8
7
15
Pt/CA-Cs973
42.0
CCAL,0 ≈ 12 mM, T = 363 K and P = 10 bar.

B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
167
absence of internal mass transfer limitations during the reaction
before PRT
after PRT
50
40
30
20
10
0
could be expected. Since the support surface chemistry and the Pt
particle size were similar for both Pt/CA-Li and Pt/CA-Cs, the higher
activity (TOF) exhibited by Pt/CA-Li should be related to its textural
characteristics.
Another way of monitoring the activity of the catalysts is to mea-
sure the time required to reach 50% conversion of cinnamaldehyde
(t50%). If one compares these values against those obtained for the
TOF, a general correspondence can be observed: higher TOF values
produce lower t50%. In this context, a much lower t50% was observed
for Pt/CA-Li (64.1 min) when compared to Pt/CA-Cs (102 min). Thus,
Pt/CA-Li not only leads to a higher conversion of cinnamaldehyde
but also is able to do so in a shorter period of time, indicating that
the material porous texture could be playing an important role in
the activity of the Pt catalyst.
Pt/CA-Li
Pt/CA-LiH
Pt/CA-LiS
Because both catalysts presented a similar micropore volume,
the different activity observed could be related to the presence of
mesopores and, consequently, to the higher external surface area in
the case of Pt/CA-Li. Thus, the porous texture formed by micro- and
mesopores of Pt/CA-Li should increase the interactions between the
cinnamaldehyde molecules and the Pt particles located throughout
the mesopores.
Fig. 5. Effect of the oxidation and post-reduction treatment of CA-Li on the selec-
tivity towards cinnamyl alcohol (CCAL,0 ≈ 12 mM, T = 363 K and P = 10 bar).
for Pt/CA-LiH (53% at 50% conversion of cinnamaldehyde) and 3.3
for Pt/CA-LiS (36%), compared with the catalyst prepared using the
untreated support (11%).
Selectivity results also appear to be somewhat affected by
This result is in line with the work published by Coloma et al. [39]
on the gas-phase hydrogenation of crotonaldehyde over carbon
black supported Pt catalysts. They reported that the selectiv-
ity towards the desired unsaturated alcohol increased when the
support was previously oxidized with hydrogen peroxide. This
behavior, which was not related to the metal particle size, was
said to be associated to the decomposition of the oxygenated sur-
face groups upon the thermal treatments in hydrogen. In our work,
the oxidation treatments carried out on the materials produce an
increase in the catalytic activity and selectivity towards the hydro-
genation of the C O bond, which improves when H2O2 is used.
As previously reported for other carbon materials [14,30,40],
high-temperature treatments at 973 K in N2 do enhance both the
the porous texture, as the hydrogenation of both C O and C
C
bonds are favored for Pt/CA-Cs, while the formation of HCAL was
favored when Pt/CA-Li was used. In accordance with the adsorp-
tion/repulsion model, the Pt particle size affects the selectivity,
leading to the hydrogenation of C C and C O, or C O bonds, prefer-
ably when small or large Pt particles are used, respectively [22]. In
our case, both catalysts presented a similar Pt particle size around
1.5 nm and thus, the different selectivity detected should be due to
the different porous texture. Therefore, when the surface chem-
istry and the metal particle size of the Pt catalysts are similar,
the mesoporous catalyst provides a higher selectivity towards the
hydrogenation of only the C C bond, when compared to the macro-
porous one.
activity and the selectivity (towards hydrogenation of the C
O
Oxidation of the carbon aerogel was found to have a marked
effect over the activity and the selectivity towards COL (Fig. 5).
Activity values increased when Pt particles were deposited on oxi-
dized carbon aerogels of similar porous texture, where slightly
larger Pt particles were obtained. This correlation between Pt-
dispersion and catalytic activity was previously observed by Teddy
et al. [38]. Comparing the catalytic performance of Pt/CA-LiH and
Pt/CA-LiS, which have similar Pt particle sizes (2.3 and 2.4 nm,
respectively) it is clear that additional effects (namely the surface
chemistry changes) should be taken into account. Selectivity to
the hydrogenation of the C O bond increased by a factor of 4.8
bond) of the catalysts. This same result was also corroborated by
Guo et al. [41] and Vu et al. [42] regarding both activity and selectiv-
ity to cinnamyl alcohol, using Pt/CNT catalysts treated at different
calcination temperatures to eliminate oxygen from the CNT surface.
According to them, the enriched electron density surrounding the
metallic nanoparticles reduces the probability of C C bond coor-
dination to Pt active sites due to the electronic repulsion, thereby
increasing the selectivity towards cinnamyl alcohol.
In the present work, the same thermal treatment was found to
increase the TOF for all tested materials by a factor up to 3, but
shifted selectivity to the C C bond (Fig. 5), due to the amount
Fig. 6. Influence of the porous texture, surface chemistry and Pt particle size on the activity and selectivity to COL for the Pt catalysts not treated at 973 K (CCAL,0 ≈ 12 mM,
T = 363 K and P = 10 bar).

168
B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
and kind remaining oxygenated surface groups after treatment,
since both catalysts (Pt/CA-LiH973 and Pt/CA-LiS973) presented
similar Pt particle size and porosity. Thus, selectivity towards the
hydrogenation of the olefinic bond in carbon aerogel supported
Pt catalysts appears to be enhanced by the presence of certain
amount of stable oxygenated groups (such as ether and quinone
groups), on the material surface. Upon oxidation treatment with
HNO3, carbon xerogels and nanotubes [14,30,40] were found to
possess a strong acidic surface, which decreased after a thermal
treatment at 973 K. Similarly to these materials, after an identi-
cal thermal treatment, carbon aerogels also lost part of its acidic
character, although not to the extent of the other reported car-
bon materials, since they conserved a higher amount of surface
groups. This fact could explain the selectivity towards olefinic
bond for the carbon aerogels. Hence, for carbon aerogels, an acidic
surface favors the interaction with the carbonyl group (i.e., car-
bon aerogels oxidized with H2O2 or (NH4)2S2O8) while a less
acidic surface (e.g., after PRT) favors the reduction of the olefinic
bond.
be carried out by designing carbon aerogels with tailored porous
texture and surface chemistry.
Acknowledgements
Financial support for this work was partly provided by LSRE/LCM
Associate Laboratory financing from FCT and FEDER under Pro-
gram COMPETE, project FCOMP-01-0124-FEDER-022706 (Ref.
FCT Pest-C/EQB/LA0020/2011), and from MEC-FEDER (CTM2010-
18889). BFM and SMT gratefully acknowledge Fundac¸ ão para
a Ciência e a Tecnologia the grants SFRH/BPD/70299/2010 and
SFRH/BPD/74239/2010, respectively. AMTS acknowledges the
financial support from POCI/N010/2006.
References
[1] R.W. Pekala, J. Mater. Sci. 24 (1989) 3221–3227.
[2] N. Job, A. Théry, R. Pirard, J. Marien, L. Kocon, J.-N. Rouzaud, F. Béguin, J.-P.
Pirard, Carbon 43 (2005) 2481–2494.
Another possible interpretation could be associated to the polar-
ity of the surface. Toebes et al. [43] related an increased activity and
selectivity towards the olefinic bond, in the non-polar surface of
Pt/CNF catalysts, upon surface group removal. They suggested that
the hydrogenation of the C C bond was assisted by adsorption of
the cinnamaldehyde aromatic ring on the non-polar CNF support
surface, thus increasing the selectivity towards hydrocinnamalde-
hyde.
[3] S.A. Al-Muhtaseb, J.A. Ritter, Adv. Mater. 15 (2003) 101–114.
[4] J. Wang, X. Yang, D. Wu, R. Fu, M.S. Dresselhaus, G. Dresselhaus, J. Power Sources
185 (2008) 589–594.
[5] C. Moreno-Castilla, F.J. Maldonado-Hodar, Carbon 43 (2005) 455–465.
[6] F.J. Maldonado-Hódar, C. Moreno-Castilla, F. Carrasco-Marín, A.F. Pérez-
Cadenas, J. Hazard. Mater. 148 (2007) 548–552.
[7] S. Morales-Torres, F.J. Maldonado-Hódar, A.F. Pérez-Cadenas, F. Carrasco-
Marín, J. Hazard. Mater. 183 (2010) 814–822.
[8] H.T. Gomes, B.F. Machado, A. Ribeiro, I. Moreira, M. Rosário, A.M.T. Silva, J.L.
Figueiredo, J.L. Faria, J. Hazard. Mater. 159 (2008) 420–426.
[9] S. Morales-Torres, F.J. Maldonado-Hodar, A.F. Perez-Cadenas, F. Carrasco-
Marin, Phys. Chem. Chem. Phys. 12 (2010) 10365–10372.
[10] P.V. Samant, F. Goncalves, M.M.A. Freitas, M.F.R. Pereira, J.L. Figueiredo, Carbon
42 (2004) 1321–1325.
To have an overall idea of catalyst performance both textural and
chemical characteristics had to be taken in account simultaneously
(Fig. 6).
[11] A.M.T. Silva, B.F. Machado, J.L. Figueiredo, J.L. Faria, Carbon 47 (2009)
1670–1679.
[12] N. Job, J. Marie, S. Lambert, S. Berthon-Fabry, P. Achard, Energy Convers. Manage.
49 (2008) 2461–2470.
[13] C. Arbizzani, S. Beninati, E. Manferrari, F. Soavi, M. Mastragostino, J. Power
Sources 172 (2007) 578–586.
[14] B.F. Machado, H.T. Gomes, P. Serp, P. Kalck, J.L. Figueiredo, J.L. Faria, Catal. Today
149 (2010) 358–364.
[15] P.V. Samant, M.F.R. Pereira, J.L. Figueiredo, Catal. Today 102–103 (2005)
183–188.
[16] N. Mahata, F. Gonc¸ alves, M.F.R. Pereira, J.L. Figueiredo, Appl. Catal. A 339 (2008)
159–168.
[17] J.C. Serrano-Ruiz, A. López-Cudero, J. Solla-Gullón, A. Sepúlveda-Escribano, A.
Aldaz, F. Rodríguez-Reinoso, J. Catal. 253 (2008) 159–166.
[18] B. Chen, U. Dingerdissen, J.G.E. Krauter, H.G.J. Lansink Rotgerink, K. Möbus, D.J.
Ostgard, P. Panster, T.H. Riermeier, S. Seebald, T. Tacke, H. Trauthwein, Appl.
Catal. A 280 (2005) 17–46.
Comparison of both CA supports (CA-Li and CA-Cs) allow to
estimate the influence of the porous texture, because both oxy-
gen contents (OTPD) and the particle sizes of their corresponding
Pt catalysts are similar. The effect of the surface chemistry on the
activity and selectivity can be discussed by comparing Pt/CA-LiH to
Pt/CA-LiS, both catalysts with similar porous textures and Pt parti-
cle sizes. As referred earlier, it is noteworthy that the activity can be
favored by the increase of the Pt particle size and oxygen content,
while the high SCOL obtained for Pt/CA-LiH seems be related with
some specific oxygen surface groups.
4. Conclusions
[19] H.-U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth.
Catal. 345 (2003) 103–151.
[20] C. Mohr, P. Claus, Sci. Prog. 84 (2001) 311–334.
Oxidation treatments with H2O2 and (NH4)2S2O8 allow the
porous structure to remain relatively unchanged while introducing
significant amounts of oxygenated groups.
The main acidic character of the fixed oxygenated groups leads
to a strong decrease of the pHPZC of the carbon support, which
strongly influence the Pt dispersion over the carbon aerogels,
decreasing in all cases after oxidation treatments.
When the surface chemistry and Pt particle size are similar,
mesoporous Pt catalysts exhibit a higher activity and selectivity
to the C C bond than macroporous ones in the hydrogenation of
cinnamaldehyde.
[21] E. Ramos-Fernández, J. Ramos-Fernández, M. Martínez-Escandell, A.
Sepúlveda-Escribano, F. Rodríguez-Reinoso, Catal. Lett. 133 (2009) 267–272.
[22] A.J. Plomp, H. Vuori, A.O.I. Krause, K.P. de Jong, J.H. Bitter, Appl. Catal. A 351
(2008) 9–15.
[23] N. Job, M.F.R. Pereira, S. Lambert, A. Cabiac, G. Delahay, J.-F. Colomer, J. Marien,
J.L. Figueiredo, J.-P. Pirard, J. Catal. 240 (2006) 160–171.
[24] C. Moreno-Castilla, M.A. Ferro-García, J.P. Joly, I. Bautista-Toledo, F. Carrasco-
Marín, J. Rivera-Utrilla, Langmuir 11 (1995) 4386–4392.
[25] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309–319.
[26] B.C. Lippens, J.H. de Boer, J. Catal. 4 (1965) 319–323.
[27] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373–380.
[28] A.F. Pérez-Cadenas, F.J. Maldonado-Hódar, C. Moreno-Castilla, Carbon 41
(2003) 473–478.
In the case of similar porous texture and Pt particle size, the
increased acidity of the Pt/CA-LiH and Pt/CA-LiS supports leads to a
higher activity and selectivity towards cinnamyl alcohol, especially
in the case of carbon aerogels oxidized with H2O2, which could be
related with a lower content of stronger acid surface groups.
At variance with other carbon supported Pt catalysts, a thermal
[29] M.A. Alvarez-Merino, F. Carrasco-Marin, J.L.G. Fierro, C. Moreno-Castilla, J. Catal.
192 (2000) 363–373.
[30] A. Solhy, B.F. Machado, J. Beausoleil, Y. Kihn, F. Gonc¸ alves, M.F.R. Pereira, J.J.M.
Órfão, J.L. Figueiredo, J.L. Faria, P. Serp, Carbon 46 (2008) 1194–1207.
[31] C. Moreno-Castilla, M.V. López-Ramón, F. Carrasco-Marín, Carbon 38 (2000)
1995–2001.
[32] M.A. Montes-Morán, D. Suárez, J.A. Menéndez, E. Fuente, Carbon 42 (2004)
1219–1225.
[33] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Orfão, Carbon 37 (1999)
1379–1389.
treatment at 973 K favors the activity and hydrogenation of the C
C
bond when using carbon aerogels as supports, due to the kind and
amount of remaining stable oxygenated groups.
Finally, the improvement of catalytic performance and selectiv-
ity towards the desired product using supported Pt catalysts can
[34] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Orfão, Ind. Eng. Chem. Res.
46 (2007) 4110–4115.
[35] M.C. Román-Martínez, D. Cazorla-Amorós, A. Linares-Solano, C.S.-M. De Lecea,
H. Yamashita, M. Anpo, Carbon 33 (1995) 3–13.

B.F. Machado et al. / Applied Catalysis A: General 425–426 (2012) 161–169
169
[36] G. Barco, A. Maranzana, G. Ghigo, M. Causa, G. Tonachini, J. Chem. Phys. 125
(2006) 194706–194712.
[40] B.F. Machado, H.T. Gomes, P. Serp, P. Kalck, J.L. Faria, ChemCatChem 2 (2010)
190–197.
[37] P. Gallezot, D. Richard, Catal. Rev. Sci. Eng. 40 (1998) 81–126.
[38] J. Teddy, A. Falqui, A. Corrias, D. Carta, P. Lecante, I. Gerber, P. Serp, J. Catal. 278
(2011) 59–70.
[41] Z. Guo, Y. Chen, L. Li, X. Wang, G.L. Haller, Y. Yang, J. Catal. 276 (2010) 314–326.
[42] H. Vu, F. Gonc¸ alves, R. Philippe, E. Lamouroux, M. Corrias, Y. Kihn, D. Plee, P.
Kalck, P. Serp, J. Catal. 240 (2006) 18–22.
[39] F. Coloma, J. Narciso-Romero, A. Sepulveda-Escribano, F. Rodriguez-Reinoso,
Carbon 36 (1998) 1011–1019.
[43] M.L. Toebes, Y.H. Zhang, J. Hajek, T.A. Nijhuis, J.H. Bitter, A.J. van Dillen, D.Y.
Murzin, D.C. Koningsberger, K.P. de Jong, J. Catal. 226 (2004) 215–225.