Influence of activated carbon surface chemistry on the activity of Au/AC catalysts in glycerol oxidation

Article abstract of DOI:10.1016/j.jcat.2011.04.008

The main goal of this work is the study of the relationship between the surface chemical characteristics of activated carbon and the performance of the respective gold-supported catalysts in the oxidation of glycerol. For that purpose, a set of modified activated carbons with different levels of oxygenated functional groups on the surface, but with no major differences in their textural parameters, was prepared. A strong effect of the activated carbon surface chemistry on the catalytic activity was observed. Gold particles with similar average sizes resulted in different performances, being the surface oxygenated acid groups particularly prejudicial for the catalytic activity. Basic oxygen-free supports characterized by a high density of free π-electrons lead to more active catalysts; the observation was tentatively explained on the basis of a recent proposed mechanism by considering the capability to promote electron mobility. However, the presence of oxygenated groups on the support does not influence significantly the selectivities.

Full text of DOI:10.1016/j.jcat.2011.04.008

Journal of Catalysis 281 (2011) 119–127
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Influence of activated carbon surface chemistry on the activity of Au/AC catalysts
in glycerol oxidation
a
a
b
b
a,
⇑
Elodie G. Rodrigues , Manuel F.R. Pereira , Xiaowei Chen , Juan J. Delgado , José J.M. Órfão
a
Laboratório de Catálise e Materiais (LCM), Laboratório Associado LSRE/LCM, Departamento de Engenharia Química, Faculdade de Engenharia,
Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal
b
Departamento de Ciência de los Materiales e Ingeniería Metalúrgica y Química Inorgánica, Facultad de Ciencias, Universidade de Cadiz,
Campus Rio San Pedro, 11510 Puerto Real, Cadiz, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The main goal of this work is the study of the relationship between the surface chemical characteristics of
activated carbon and the performance of the respective gold-supported catalysts in the oxidation of glyc-
erol. For that purpose, a set of modified activated carbons with different levels of oxygenated functional
groups on the surface, but with no major differences in their textural parameters, was prepared. A strong
effect of the activated carbon surface chemistry on the catalytic activity was observed. Gold particles with
similar average sizes resulted in different performances, being the surface oxygenated acid groups partic-
ularly prejudicial for the catalytic activity. Basic oxygen-free supports characterized by a high density of
Received 2 February 2011
Revised 29 March 2011
Accepted 13 April 2011
Available online 17 May 2011
Keywords:
Glycerol
Oxidation
Gold
Activated carbon
Surface chemistry
Mechanism
free p-electrons lead to more active catalysts; the observation was tentatively explained on the basis of a
recent proposed mechanism by considering the capability to promote electron mobility. However, the
presence of oxygenated groups on the support does not influence significantly the selectivities.
Ó 2011 Elsevier Inc. All rights reserved.
1
. Introduction
Over the last decades, biodiesel has emerged as a viable clean
Recently, there have been many studies dealing with the oxida-
tion of glycerol to chemical intermediates using supported gold
nanoparticles as catalysts [5–7]. One of the advantages of using
this metal is related with its strong resistance to oxygen poisoning,
compared with platinum-based catalysts, allowing for the use of
high oxygen partial pressures [8]. However, under alkaline condi-
tions, catalyst performance and selectivity seem to mainly depend
on the nanoparticles size, which in turn can be controlled by the
preparation method [9,10]. Generally, small nanoparticles (5 nm)
were found to be more active than larger particles (20 nm) [11].
Apparently, the gold–sol immobilization method is the most
appropriate technique in order to obtain these well-dispersed
nanoparticles on carbons [9,12]. In addition to the particles size,
the activity of gold catalysts seems to be affected by the nature
of support. Nano-sized gold particles supported on different carbon
materials (e.g., carbon black, activated carbon, and graphite) and
fuel. During its production via transterification of oils from plants,
about 100 kg of glycerol is produced for every ton of biodiesel [1].
During the last years, this market has been continuously increas-
ing, and it is expected that this trend remains [1]. The rapid expan-
sion of the global production of biodiesel has created a major
surplus of glycerol, leading to a decline in glycerin pricing. Con-
sumption of this extra glycerol is a necessary requisite for the com-
mercial viability of biodiesel production. Liquid phase catalytic
oxidation is a promising route to the valorization of glycerol by
its conversion into useful compounds, provided that catalysts used
are sufficiently active and selective for the formation of com-
pounds such as glyceric acid (GLYCEA) and/or dihydroxyacetone
(
DIHA), potentially useful as chemical intermediates in the fine
chemicals industry, particularly in pharmaceuticals [1,2]. However,
the extensive functionalization of the glycerol molecule, with
similar reactive hydroxyl groups, renders its selective oxidation
particularly difficult [3,4].
2 2 3
oxides (TiO , MgO, and Al O ) are active for the oxidation of
glycerol but show very different performances, being carbon-
supported gold catalysts more active than most oxides-supported
catalysts [5,8,13]. However, even when different carbon materials
are compared, textural and chemical properties could largely dif-
fer, which make the role of the support complex to understand.
Moreover, in spite of its well-known influence on the catalyst
activity, the role of the support in the mechanism of glycerol oxi-
dation is still not clear.
⇑
Corresponding author. Fax: +351 225 081 449.
E-mail addresses: elodie.rodrigues@fe.up.pt (E.G. Rodrigues), fpereira@fe.up.pt
(
M.F.R. Pereira), xiaowei.chen@uca.es (X. Chen), juanjose.delgado@uca.es (J.J.
Delgado), jjmo@fe.up.pt (J.J.M. Órfão).
0
021-9517/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2011.04.008

1
20
E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
In this study, a NORIT ROX 0.8 activated carbon was modified by
2.1.2.2. Preparation of catalysts. Gold on carbon catalysts were pre-
pared by the sol immobilization method with NaBH as reducing
agent [18]. Briefly, HAuCl O (35.1 mg) was dissolved in
ꢀ3H
690 mL of O, and polyvinyl alcohol was added (1.6 mL,
0.2 wt%) under stirring. The gold colloid was subsequently reduced
by the addition of NaBH (4 mL, 0.1 M) under vigorous magnetic
stirring. The resulting ruby-red sol was formed immediately. With-
in a few minutes of sol generation, the colloid was immobilized by
adding the support under vigorous stirring. The amount of support
was calculated to reach a final nominal metal loading of 1 wt%.
After 3–4 days, the colorless solution was filtered, the catalyst
washed thoroughly with distilled water until the filtrate was free
chemical and thermal treatments in order to obtain supports with
different surface chemical properties, but with no major differ-
ences in their textural parameters. The sol immobilization method
was used for preparing gold catalysts supported on the different
activated carbon samples. Then, the performance of the prepared
catalysts was investigated and the possible role of the support sur-
face chemistry in the reaction mechanism was discussed.
4
4
2
H
2
4
2
. Experimental
2
2
.1. Materials and methods
of chloride (AgNO
treatment under N
any possible organic waste, the catalysts were activated by reduc-
tion in a stream of H flow for 3 h at 350 °C.
3
test) and dried at 110 °C for 24 h. After heat
2
flow for 3 h at 350 °C, in order to remove
.1.1. Materials
The gold precursor, HAuCl
4
.3H
2
O (99.99% ACS, Au 49.5% min),
2
was supplied by Alfa Aesar. NaBH
4
of purity >95% from Riedel-de-
Haël, NaOH (>97%) and polyvinylalcohol (PVA) (purity >99%) from
Aldrich were used. The commercial activated carbon NORIT ROX
2.2. Characterization
0
.8 was selected as starting material for the preparation of
The supports were characterized by N
a NOVA Quantachrome Instruments apparatus.
2
adsorption at ꢁ196 °C in
supports.
The qualitative and quantitative determination of oxygenated
surface functional groups was performed by temperature pro-
grammed desorption – mass spectrometry (TPD-MS) [15,16]. CO
and CO2 TPD spectra were obtained with a fully automated AMI-
200 equipment (Altamira Instruments). The samples (150 mg)
were placed in a U-shaped quartz tube inside an electrical furnace
2
.1.2. Preparation procedures
The NORIT ROX 0.8 activated carbon has a small amount of sur-
face oxygenated groups. In order to evaluate how catalyst activity
correlates with the concentration of those groups, the original sup-
port was modified by chemical and thermal treatments in order to
obtain supports with different surface chemical properties. Gold
was supported on these modified activated carbon materials by
the sol immobilization method.
ꢁ1
and subjected to a 5 °C min linear temperature increase up to
3
ꢁ1
1100 °C under helium flow (25 Ncm min ). A quadrupole mass
spectrometer (Dymaxion 200, Ametek) was used to monitor CO
2 2
and CO signals. For quantification of the CO and CO released, cal-
ibration of these gases was carried out at the end of each analysis.
In some cases, the determination of the pH at the point of zero
charge (pHpzc) was also carried out.
2
.1.2.1. Preparation of modified activated carbons. A NORIT ROX 0.8
activated carbon (pellets of 0.8 mm diameter and 5 mm length)
was used as starting material for this study (sample AC ). This sam-
0
The average metal particles size was determined by high-reso-
ple was submitted to different chemical and thermal treatments in
order to obtain materials with different surface chemistries, while
maintaining the original textural properties as far as possible.
lution transmission electron microscopy (HRTEM) using
a
JEOL2010F instrument, with 0.19 nm spatial resolution at Scherzer
defocus conditions. High Angel Annular Dark Field – Scanning
Transmission Electron Microscopy (HAADF-STEM) images were
obtained with the same microscope. The average diameter was cal-
culated by the measurement of at least 200 particles.
The noble metal loading of the samples was determined by
inductively coupled plasma (ICP) analysis using an Iris Intrepid
Spectrometer (Thermo Elemental). An Ethos 1600 microwave lab-
station (Milestone) was used to prepare the acid solution of all
the samples for ICP analysis.
2
.1.2.1.1. Oxidation treatment. Initially, HNO
into a round-bottom flask. A Soxhlet extraction apparatus with
activated carbon AC was connected to the flask and to a con-
3
6 M was introduced
0
denser. The acid solution was heated to boiling temperature, and
the reflux was stopped after 3 h. The oxidized material was subse-
quently washed with distilled water until neutral pH, and then
dried at 110 °C for 24 h (sample AC ). The purpose of this liquid-
1
phase oxidation is to increase the amount of oxygen-containing
surface groups on the support.
Details of these characterization methods are described else-
where [19].
2
1
.1.2.1.2. Thermal treatments. Sample AC was used as the starting
material for the thermal treatments since it is important that the
starting material presents a large amount of surface groups in or-
der to produce activated carbons with a highly basic character
2.3. Catalysts evaluation
[
14]. Support AC
1
was treated under nitrogen flow for 30 min at
The standard oxidation experiments were carried out with oxy-
gen under 3 bar, at 60 °C. NaOH solution (NaOH/glycerol molar ra-
tio = 2) and catalyst (700 mg) were added to a 0.3 M aqueous
solution of glycerol (V = 150 mL) under stirring at 1000 rpm. Addi-
tionally, a limited number of runs were carried out at different oxy-
gen pressures (6, 10 bar) or temperatures (40, 80 °C). After heating
under nitrogen to the selected temperature, the reaction was initi-
ated by switching from inert gas to oxygen.
different temperatures (400, 600, or 900 °C) followed by a treat-
ment under dry air flow at room temperature for 1 h. Therefore,
the following samples were prepared: AC1tt400, AC1tt600, AC1tt900
The same procedure was repeated but under a flow of hydrogen
at 900 °C (sample AC1tt900-H2).
These thermal treatments selectively remove the oxygen-con-
taining surface groups, previously introduced in the treatment
.
with HNO
under air is intended to stabilize the surface chemistry of the sam-
ples. Besides, thermal treatments under H flow are expected to
generate higher basicity on the surface of the carbon than thermal
treatments under N flow [17].
3
, by thermal degradation [15,16]. The final treatment
The reaction was monitored by taking samples (0.5 mL) for
analysis at regular time intervals. The quantitative analysis of mix-
tures was carried out by high-performance liquid chromatography
(HPLC). The chromatograph (Elite LaChrom HITACHI) was
equipped with a refractive index and an ultraviolet (210 nm)
detector. Reactant and products were separated on an ion
exclusion column (Alltech OA 1000). Products were identified by
2
2
After these treatments, all samples were ground to a particle
diameter between 0.1 and 0.3 mm before use.

E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
121
comparison with standard samples. Glycerol (>99.5%) was pur-
chased from Fluka, glyceric acid (GLYCEA), dihydroxyacetone
(
a) 0.6
AC1tt600
(
DIHA) and the by-products glycolic acid (GLYCOA), glyceralde-
0.5
hyde (GLYCER), tartronic acid (TARTA), and oxalic acid (OXALA)
were obtained from Sigma–Aldrich. Further details can be found
0
0
0
.4
.3
.2
AC1tt400
elsewhere [19]. The selectivities (S
at time t were calculated as:
i
) into the different products i
AC0
C
i
S
i
¼
m
i
ð1Þ
ꢀ C
0
ꢀ X
0.1
AC1
AC1tt900
AC1tt900-H2
where C
concentration of glycerol (mol L ), X is the glycerol conversion,
and corresponds to the moles of i produced per mol of glycerol
consumed, according to the stoichiometry.
i
is the concentration of product i (mol Lꢁ1), C
0
is the initial
0
.0
ꢁ1
0
200
400
600
800
1000
m
i
Temperature (ºC)
(
b)
AC1
0
0
.3
.2
3
. Results and discussion
3.1. Characterization of supports and catalysts
The main goal of the present work is to study the relationship
AC1tt400
between the surface chemical characteristics of activated carbons
and the catalytic performance of gold supported on these materials
for glycerol oxidation. For that purpose, a set of modified activated
carbons with different levels of acidity/basicity was prepared. Both
liquid-phase oxidations and thermal treatments have been re-
ported not to significantly change the textural properties of acti-
vated carbons [15,16].
0.1
0.0
AC0
AC1tt900
AC1tt600
AC1tt900-H2
0
200
400
600
800
1000
Temperature (ºC)
2
Fig. 1. TPD spectra of the different activated carbons: (a) CO; (b) CO .
3.1.1. Textural properties
The textural characterization of the different supports and of
temperature ranges. Accordingly, it is possible to identify and esti-
mate the amount of the oxygenated groups on a given material by
TPD experiments. The nature of the groups can be assessed by the
decomposition temperature and the type of gas released, and their
respective amounts determined by the areas of the component
peaks, obtained by deconvolution techniques [15,16]. It was re-
ported that CO results from the decomposition of carboxylic acids
2
at low temperatures (150–450 °C) or from lactones at high temper-
atures (600–800 °C); carboxylic anhydrides originate both CO and
the Au/AC
changes in the textural properties of the materials were observed.
The slight decrease in AC surface area may be due to the abundant
presence of oxygen-containing groups introduced on the surface of
the activated carbon by the treatment with HNO , which possibly
block the entry of N inside the small pores, as reported by Yin
0
catalyst is presented in Table 1. As expected, no drastic
1
3
2
et al. [20]. After treatment at 900 °C, the mentioned functional
groups were removed by thermal decomposition and the area of
the corresponding support slightly increased comparatively to
the original carbon [21].
2
CO (400–650 °C); groups such as phenols (600–800 °C), and car-
bonyls and quinones (750–1000 °C) originate CO [15,16]. Conse-
quently, an increase in the amount of surface oxygen is
0
The BET surface area of the Au/AC catalyst decreased only
slightly compared with the unloaded carbon. Therefore, it was as-
sumed that the textural properties of modified supports and sup-
ported gold catalysts are not significantly different from those of
the original activated carbon.
evidenced by the increase in CO and CO
Fig. 1a and b shows, respectively, the CO and CO
the commercial activated carbon as received (AC ) and the modi-
fied samples. The deconvolution of these CO and CO spectra was
2
released.
2
TPD spectra of
0
2
carried out in order to determine the amount of each surface
group. A multiple Gaussian function was used for fitting each spec-
trum. The numerical calculations were based on a non-linear rou-
tine, which minimized the sum of squared deviations, using the
Levenberg–Marquardt method to perform the iterations and some
assumptions according to [15,16]. Table 2 shows the results
obtained.
3
.1.2. Surface chemistry characterization
TPD is a thermal analysis method that is becoming popular for
the characterization of the oxygen-containing groups on the sur-
face of carbon materials. In this technique, the total amount of
the various oxygenated surface groups (carboxylic acids, carboxylic
anhydrides, lactones, phenols, and carbonyls or quinones) on car-
bon materials can be determined, since these groups are decom-
Analyzing the results, it is clear that the sample AC
oxidized with HNO , has the highest amount of oxygen-containing
surface groups. The liquid-phase oxidation significantly increased
the amount of CO evolved at low temperatures (carboxylic acid
groups) and CO at high temperatures (phenol and carbonyl/qui-
none groups). As a result, the oxidized sample AC has a low pHpzc
see Table 1) due to the introduction of the mentioned oxygen-con-
1
, which was
posed upon heating by releasing CO and/or CO
2
at different
3
Table 1
2
Textural properties and pHpzc of activated carbons and the Au/AC
0
catalyst.
2
2
3
1
Sample
S
BET (m /g)
S
meso (m /g)
V
micro (cm /g)
pHpzc
(
AC
AC
AC1tt400
AC1tt600
AC1tt900
AC1tt900-H2
0
1
1025
965
952
1042
1106
1062
958
180
156
200
150
184
193
195
0.361
0.369
0.350
0.391
0.399
0.359
0.343
7.9
3.2
–
taining functional groups with acidic properties, namely carboxylic
acids (see Table 2), originating a support with acidic and hydro-
philic characteristics [22]. Thermal treatments selectively remove
the oxygen-containing surface groups, originating materials with
progressively lower oxygen contents (see Fig. 1 and Table 2). It
can also be observed in Table 2 that the amount of carboxylic acids
–
8.2
9.0
–
Au/AC
0

1
22
E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
Table 2
Amounts of the different oxygenated surface groups present on activated carbon supports (obtained by deconvolution of CO
oxygen released.
2
and CO spectra) and total amounts of CO, CO
2
and
)
ꢁ
Sample
Carboxylic acids
Carboxylic anhydrides
Lactones
Phenols (
l
mol g 1
)
Carbonyl /quinones
CO
CO
mol g
2
O
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
(lmol g
)
(l
mol g
)
(l
mol g
)
(l
mol g
)
(l
mol g
)
(
l
)
(lmol g
a
AC
AC
AC1tt400
AC1tt600
AC1tt900
AC1tt900-H2
0
134
608
48
44
84
0
37
18
82
75
101
12
0
191
788
895
658
71
500
776
189
929
356
172
127
0
1154
4272
2922
2109
526
a
1
239
233
27
31
0
1209
1082
1080
170
2414
2210
1765
272
96
26
70
96
a
The total amount of CO is slightly higher than the sum of the groups that decompose releasing CO. The justification can be found elsewhere [16].
and anhydrides considerably decreases for thermal treatments
above 400 °C; on the other hand, phenols and carbonyl/quinones
only decrease significantly after heat treatment above 600 °C. As
can be seen in the TPD spectra of samples AC1tt900 and AC1tt900-
1
can be seen in the TPD spectra (Fig. 1). The sample AC is the un-
ique support that has a significant amount of oxygenated groups
able to decompose at temperatures below 350 °C (Fig. 1). There-
fore, during the heat treatment and reducing steps used in the
H2, the CO
2
and CO releasing groups have almost completely been
1
preparation of the Au/AC catalyst, some groups are thermally
removed; these samples have basic properties [17]. According to
Leon y Leon et al. [23], this basic character is mainly due to the
electron rich oxygen-free sites located on the carbon basal planes.
It was also reported that heat treatments under hydrogen flow at
high temperatures are more effective in removing oxygen than un-
der inert atmosphere, once they originate stable basic surfaces by
forming stable C–H bonds, whereas the activated carbon heat trea-
ted under nitrogen has a surface with reactive sites capable of
readsorbing oxygen, leading to the formation of some of the previ-
ously removed groups, as it can be observed in Table 2 [17]. This is
in agreement with our TPD data, since sample AC1tt900 shows a
decomposed and the gold particles that are anchored to them
may have a low stability and could be prone to sintering. This
1
can possibly explain why the Au/AC catalyst has larger metallic
particles than Au/AC , Au/AC1tt400, Au/AC1tt600, Au/AC1tt900, and
0
Au/AC1tt900-H2, because on this latter group of samples, contrarily
to the former, there was no oxygenated groups that decompose
at low temperatures during the preparation step. The supports
AC1tt400, AC1tt600, AC1tt900, and AC1tt900-H2 were already heat treated
at temperatures above 350 °C before gold impregnation; therefore,
as it can be seen in TPD spectra (Fig. 1), they do not release CO and/
2
or CO below 350 °C.
higher content of CO
2
releasing groups than sample AC1tt900-H2
Table 3 summarizes the metal loadings of the catalysts. The
gold content obtained by ICP analysis is slightly lower than the
nominal. However, chemical and heat treatments of activated car-
bon did not influence substantially the gold content of catalysts
and all the samples present a similar value of approximately 0.6%.
(
see Fig. 1 and Table 2). As expected, sample AC1tt900-H2 has the
highest pHpzc value (see Table 1) and corresponds to the most basic
material.
3.1.3. Microscopy and ICP analyses
HAADF-STEM images were collected in order to get information
3.2. Kinetic experiments
about metal particle size distributions (Fig. 2). The average crystal-
lite sizes of the studied catalysts are presented in Table 3. For oxi-
dation catalysis by gold, it has been shown that the average
particle size influences both activities and selectivities [9,24].
Although small gold particle sizes (<5 nm) have a higher activity,
catalysts with gold crystallites as large as 42 nm have already
shown a relevant activity in glycerol oxidation [24]. Carrettin
et al. [25] reported that active catalytic sites have a broad size dis-
tribution between 15 and 30 nm diameter for gold nanoparticles
supported on carbon, the 15 nm size being predominant.
Nevertheless, in the present work, gold catalysts prepared by
the sol immobilization method have particles with sizes lower
than 15 nm, and therefore, these catalysts are expected to be
highly active. The same impregnation procedure was used in order
to obtain gold catalysts with similar particle sizes, independently
of the support used. This was effectively achieved for Au/AC0, Au/
AC1tt400, Au/AC1tt600, Au/AC1tt900, and Au/AC1tt900-H2, which have
fairly low and similar average gold particle sizes, suggesting that
the difference in the performance of these catalysts will be mainly
due to the different chemical surface properties of the supports.
3.2.1. Influence of pressure and temperature
The influence of oxygen pressure and temperature on the reac-
0
tion was examined for the Au/AC catalyst. Experiments were car-
ried out with 150 mL of glycerol (0.3 M), NaOH/glycerol = 2 mol/
mol, and 700 mg of catalyst. The oxygen pressure was varied be-
tween 3 and 10 bar and the temperature between 40 and 80 °C.
Under the tested conditions, it was found that the reaction rate
increases with the oxygen pressure as depicted in Fig. 3a). For in-
stance, when this value is 10 bar, total glycerol conversion is
achieved after about 1.5 h. Moreover, selectivity to DIHA slightly
increases with the increase in oxygen pressure, while selectivity
to unwanted GLYCOA decreases (Table 4).
The temperature increase also leads to an increase in the cata-
lytic activity as seen in Fig. 3b), but appears to be very unfavorable
for the selectivity to the products of interest, especially for glyceric
acid, which seems to oxidize very easily to glycolic acid at high
temperatures (Table 4). This result was described before by Porta
and Prati [9] under similar conditions.
For comparative purposes, selectivities to the different products
are presented at constant reaction time (2 h) and also at the same
glycerol conversion (90%) in Table 4. The selectivities obtained are
practically independent of time and glycerol conversion, varying
only slightly during the reaction for all the conditions and catalysts
tested in this work. As an illustrative example, the evolution of
DIHA and GLYCEA selectivities as a function of glycerol conversion
The Au/AC
due to the large amount of oxygen-containing groups on support
AC , which could influence the metal dispersion. On one hand,
1
sample has a higher average particle size. This can be
1
the surface oxygenated groups are considered to act as anchoring
sites that interact with metallic precursors, improving the metal
dispersion [26]. However, some of these functional groups present
on activated carbon are easily decomposed upon heating even at
low temperatures. This occurs with some carboxylic acid groups,
which already decompose at temperatures below 350 °C, as it
0
in the presence of Au/AC at different oxygen pressures and reac-
tion temperatures can be seen in Fig. 3c and d, respectively. In
some cases, the sum of the calculated selectivities is lower than

E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
123
(
a)10
5
4
3
2
1
0
(
b)
8
6
4
2
0
Particle Size (nm)
Particle Size (nm)
(
c)20
(d)10
8
6
4
2
0
1
5
0
5
0
1
Particle Size (nm)
Particle Size (nm)
18
(
e)10
(f)1
6
4
2
0
8
6
4
2
0
8
6
4
2
0
1
1
1
Particle Size (nm)
Particle Size (nm)
0 1
Fig. 2. Particle size distributions and HAADF-STEM images of the Au catalysts supported on (a) AC , (b) AC , (c) AC1tt400, (d) AC1tt600, (e) AC1tt900, and (f) AC1tt900-H2.
Table 3
Average crystallite size and metal content of the studied catalysts.
It should be noticed that the comparison between the results
obtained in this work and those reported in literature is particu-
larly difficult since the reaction conditions changed from author
to author. Moreover, the results are often presented as conversion
obtained after a certain reaction time. For example, under condi-
tions identical to the present research, Carretin et al. [27] obtained
Loading (wt%)a
Catalyst
d
M
(nm)
Au/AC
Au/AC
Au/AC1tt400
Au/AC1tt600
Au/AC1tt900
Au/AC1tt900-H2
0
1
0.59
0.62
0.60
0.56
0.57
0.66
0.62
0.56
6.8
12.5
4.4
5.8
6.4
5.0
8.3
5
6% glycerol conversion after 3 h of reaction at 60 °C and 3 bar
using 1 wt% Au/graphite. In the present work, total conversion
was achieved at this same reaction time with Au/AC . Additionally,
the turnover frequencies (TOFs) calculated after 1 h of reaction for
the Au/AC catalyst tested at different temperatures, and oxygen
Au/AC
Au/AC
0
(after 2 run)
(after 4 run)
0
0
11.4
a
Obtained by ICP analysis after combustion of activated carbon.
0
pressures are shown in Table 5 and compared with some typical
results from the literature [28–30]. It can be concluded that the
activities obtained in this work are similar or even higher than
those reported by different groups.
1
(see Table 4), since some oxalic acid was also detected but always
with low concentrations.

1
24
E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
3
.2.2. Investigation of the support effects in the glycerol oxidation
Fig. 4 shows the evolution of glycerol conversion in the pres-
(
(
(
a)1.0
0
0
0
0
0
.8
.6
.4
.2
.0
ence of the gold catalysts prepared on different activated carbons.
The surface chemical functional groups of activated carbon can be
controlled by chemical and heat treatments. Since no drastic
changes were observed in the textural properties of the supports
3
6
1
bar
bar
0 bar
(
cf. Table 1), in gold average particle sizes and in metal loadings
of the catalysts (cf. Table 3), the disparities in performances ob-
served in Fig. 4 must be related almost exclusively to the differ-
ences in support chemical properties. In fact, whether the
activity was only dependent on the gold particles size, the follow-
ing order of performances would be expected: Au/AC1tt400 > Au/
0
50
100
150
200
250
300
Time [min]
0 1
AC1tt900-H2 > Au/AC1tt600 > Au/AC1tt900 > Au/AC > Au/AC (see Table
b)1.0
3). However, sample Au/AC1tt400, which has the smallest gold par-
ticles size (average diameter = 4.4 nm), only shows a weak effi-
ciency in glycerol oxidation, mainly when compared with Au/
AC1tt900-H2, which has a similar average particle size (5.0 nm), but
0
0
0
0
0
.8
.6
.4
.2
.0
4
6
0 ºC
0 ºC
0
also with Au/AC (6.8 nm) and Au/AC1tt900 (6.4 nm). These cata-
lysts, all with similar distributions of gold particle sizes, clearly re-
sulted in different performances depending on the amount of
oxygenated groups on the carbon surface. In general, as the oxy-
genated groups of the support are removed, the catalytic perfor-
mance gradually increases.
80 ºC
0
50
100
150
200
250
300
0
We also tested the Au/AC catalyst in successive experiments.
Each run was carried out under the standard conditions described
previously. The catalyst was recovered by filtering off the solution
of the previous run after 5 h of reaction and reused. Due to some
losses during the filtration procedure, a limited amount of fresh
catalyst (<5% of the total weight) was also added to the reaction
medium. It can be observed in Fig. 5 that the first two runs corre-
spond to similar results, but in the third and mainly in the fourth
the catalyst performance significantly decreased (conversion after
Time [min]
c)0
.6
3
bar
6 bar
0 bar
0
0
0
.4
.2
.0
GLYCEA
DIHA
1
2
h is about 40% lower in run 4). ICP analyses (see Table 3) show
that the gold content remains unchanged after the four runs; then,
metal leaching is not the cause of the observed deactivation. How-
ever, a growing size of Au particles was observed by microscopy
analysis of the recovered catalyst after successive experiments
0
.0
0.2
0.4
0.6
0.8
1.0
(
see Table 3), which probably explains the poor performance. On
Conversion
the other hand, this may affect positively the selectivity to GLYCEA,
which slightly increased from 0.62 on the first run to 0.65 on the
fourth, whereas the selectivity to DIHA remains practically un-
changed (Fig. 5). The effect of the particle size on the GLYCEA selec-
tivity is in agreement with the results of Porta and Prati [9].
Although these authors [9] prepared gold catalysts by a similar
PVA-protected sol preparation, they used them directly as obtained
(d)
0
0
0
0
.6
.4
.2
.0
GLYCEA
DIHA
40 ºC
6
0 ºC
0 ºC
after the synthesis (i.e., without any previous heat treatments in N
and H , as described in Section 2.1.2.2). In order to study the influ-
ence of the mentioned activation in the catalyst performance, we
tested the Au/AC catalyst as obtained and heat treated at 350 °C.
2
8
2
0
A better performance was achieved with the catalyst prepared by
the standard procedure (Section 2.1.2.2), as it can be seen in Table
0
.0
0.2
0.4
0.6
0.8
1.0
5
(TOF reduces by more than 50% in the catalyst prepared without
heat treatments). Moreover, successive runs carried out under the
standard experimental conditions with the Au/AC catalyst, with or
Conversion
0
0
Fig. 3. Glycerol conversion over Au/AC varying (a) the oxygen pressure or (b) the
without heat treatments at 350 °C (data not shown), allow to con-
clude that deactivation of the catalyst used without heat treat-
ments is much more significant than that of the sample heat
treated at 350 °C. For example, the glycerol conversion after 2 h
in the third run is about 25% lower in the presence of the catalyst
prepared by the standard procedure (Fig. 5), while the sample used
without PVA removal leads to a relative conversion decrease of
53%. Moreover, it was observed by microscopy analysis of the
recovered catalyst after successive experiments that the average
temperature, and evolution of DIHA and GLYCEA selectivities as a function of
glycerol conversion at different (c) oxygen pressures and (d) reaction temperatures.
Reaction conditions: 150 mL of glycerol 0.3 M, catalyst amount = 700 mg, NaOH/
glycerol = 2 mol/mol.
It is important to mention that about 85% total selectivity to
products of commercial interest was achieved at 40 °C and 3 bar
or 60 °C and 10 bar. Then, it may be concluded that gold prepared
by the sol immobilization method is highly effective, and the selec-
tivity can be slightly altered by changing the reaction conditions.
Furthermore, gold allows high levels of oxygen in the reaction
medium without suffering any significant deactivation.
0
size of gold particles in the sample Au/AC heat treated at 350 °C
increases from 6.8 nm (fresh catalyst) to 11.4 nm after four runs
(see Table 3), while the average crystallite size of the sample used

E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
125
Table 4
GLYCEA, GLYCOA, DIHA, and TARTA selectivities (SGLYCEA, SGLYCOA,
reaction with the Au/AC catalyst.
S
DIHA, STARTA) at different oxygen pressures and temperatures for 90% glycerol conversion (XGLY) and after 2 h of
0
Catalyst
t = 120 min
XGLY = 90%
X
GLY
S
GLYCEA
S
GLYCOA
S
DIHA
S
TARTA
S
GLYCEA
S
GLYCOA
S
DIHA
S
TARTA
Au/AC
0
3
6
1
bara
0.72
1
1
0.62
0.60
0.63
0.15
0.10
0.10
0.18
0.19
0.22
0.05
0.05
0.05
0.59
0.61
0.64
0.14
0.11
0.10
0.17
0.19
0.21
0.05
0.05
0.05
a
bar
0 bara
b
4
6
8
0 °Cb
0.59
0.72
1
0.62
0.62
0.50
0.11
0.15
0.26
0.22
0.18
0.19
0.05
0.05
0.05
0.63
0.59
0.47
0.10
0.14
0.25
0.21
0.17
0.17
0.06
0.05
0.05
0 °Cb
0 °C
a
Reaction conditions: 60 °C, 150 mL of glycerol 0.3 M, NaOH/glycerol = 2 mol/mol, catalyst amount = 700 mg.
Reaction conditions: pO2 = 3 bar, 150 mL of glycerol 0.3 M, NaOH/glycerol = 2 mol/mol, catalyst amount = 700 mg.
b
Table 5
1
0
.0
.8
Comparison between turnover frequencies (TOFs) obtained after 1 h of reaction for
the Au/AC catalyst tested at different temperatures and oxygen pressures and some
0
typical results from the literature.
Run 1
Run 2
Run 3
Run 4
Reaction conditions
TOF (hꢁ1)
Ref.
T (°C)
pO2 (bar)
NaOH/GLY (mol/mol)
0.6
4
6
6
6
6
8
6
5
5
7
0
0
0
0
0
0
0
0
0
0
3
3
3
6
10
3
10
3
3
2
2
2
2
2
2
2
4
4
4
3930
4050
1800
6550
10,080
7070
8640
900
This work
This work
This work
This work
This work
This work
[28]
[29]
[30]
[30]
GLYCEA
DIHA
0
.7
0
.4
0.6
0.5
a
0.4
0.3
0.2
0.2
0.0
0.1
.0
0
1
2
3
4
Number of runs
830
960
0
50
100
150
200
250
300
3
Time [min]
a
0
Au/AC catalyst used without prior heat treatments at 350 °C.
Fig. 5. Successive experiments with the Au/AC
and DIHA at t = 2 h in each cycle are also presented inside. Reaction conditions:
0 °C, pO2 = 3 bar, 150 mL of glycerol 0.3 M, catalyst amount = 700 mg, NaOH/
glycerol = 2 mol/mol.
0
catalyst. Selectivities to GLYCEA
6
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
Au/AC0
Au/AC1
Au/AC1tt400
Au/AC1tt600
Au/AC1tt900
AC
efficiency of the Au/AC
ticle size, cannot be explained only by the gold dispersion.
The Au/AC catalyst has a high content of oxygenated functional
groups and only a residual performance (see Table 2 and Fig. 4).
When carboxylic acid groups were removed from the surface (sam-
ple Au/AC1tt400), no significant enhancement was observed. How-
ever, after removing carboxylic acid and carboxylic anhydride
groups (sample Au/AC1tt600), the catalytic performance improves.
Finally, catalysts with a low content of oxygenated surface groups,
0
catalyst shows a fairly high activity (Fig. 5). Therefore, the poor
1
catalyst, which has a similar average par-
Au/AC1tt900-H
2
1
0
such as Au/AC , Au/AC1tt900, and Au/AC1tt900-H2, show a similar high
0
50
100
150
200
250
activity. Therefore, Fig. 4 reveals a strong influence of the surface
chemical properties of the support on the catalytic performance
in glycerol oxidation, indicating that surface oxygenated groups,
mainly those with acid nature, are prejudicial.
As the temperature used in the preparation of supports is in-
creased, oxygenated groups are decomposed sequentially, releas-
Time [min]
Fig. 4. Influence of the modified activated carbon supports on the performance of
gold catalysts. Reaction conditions: 60 °C, pO2 = 3 bar, 150 mL of glycerol 0.3 M,
catalyst amount = 700 mg, NaOH/glycerol = 2 mol/mol.
2
ing CO and/or CO , and therefore, the resulting supports have a
in the reaction without heat treatments increases from 4.6 nm
gradual lower amount of oxygen, as it can be seen in Table 2. As
a result, the oxygen content of the surface seems to be a factor of
large influence in the reaction, since a trend can be observed in
Fig. 6. Additionally, TOFs obtained after 2 h of reaction for the dif-
ferent catalysts are also presented in Fig. 6 as a function of the oxy-
gen content. It seems that the catalysts tested can be assorted in
three distinct groups:
(
fresh catalyst) to 25 nm after only three runs. Therefore, it seems
that the standard procedure for the preparation of catalysts leads
to stronger metal/support interactions, slowing down the sinter-
ization rate of Au particles. As the catalyst heat treated at 350 °C
presents higher performance and stability, the procedure adopted
seems to be the most adequate to prepare the gold catalysts.
0
The successive experiments carried out with the Au/AC cata-
lyst heat treated at 350 °C confirm the influence of the gold disper-
sion on the catalytic performance, but also allow concluding that
even with a metal average size of 11.4 nm (see Table 3) the Au/
j
1
Au/AC and Au/AC1tt400: catalysts with a high amount of oxy-
gen-containing functional groups (mainly acid groups) and
low activity.

1
26
E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
AC1tt900-H2
O-H bond cleavage can occur on the support basic sites. On the
5000
AC1tt900
0
0
0
0
0
.8
.6
.4
.2
.0
contrary, also in a study of benzyl alcohol oxidation, but using acti-
vated carbon supports, Zhu et al. [32] concluded that the activity of
Au/AC catalysts was improved by the presence of oxygen-contain-
ing species on the surface, which were proposed as responsible for
oxygen activation.
In Table 6, the selectivities obtained in the presence of the most
active catalysts are compared at constant reaction time and at the
same glycerol conversion (X = 40%). Contrary to what was observed
with the conversions, the presence of oxygenated groups on the
support did not influence significantly the selectivities. The pre-
pared catalysts lead to a high total selectivity to products of com-
mercial interest (GLYCEA + DIHA) of about 80%, and a very similar
distribution of the products was observed, independently of the
amount of oxygenated surface groups present.
AC
0
AC1tt900
AC
0
4000
3000
2000
1000
0
AC1tt900-H2
AC1tt600
AC
AC1tt400
1
0
1000
2000
3000
4000
-1
Oxygen content (µmol g
)
AC1tt600
AC1tt400
AC1
0
1000
2000
3000
4000
-
1
Oxygen content (µmol g )
Fig. 6. Influence of the oxygen content of activated carbons on the catalytic
performance. Insert: TOF after 2 h of reaction for the different catalysts vs. the
oxygen content.
3.2.3. Discussion on the influence of support surface chemistry
The role of the support in the catalytic reaction studied is still
under discussion. The mechanism for oxygen adsorption and acti-
vation on supported gold catalysts is not yet completely estab-
lished. For the discussion of our observations, it could be
assumed that oxygen adsorption proceeds directly on the metal
particles [33]. However, in this mechanism, the support would
have only a secondary role in the catalytic activity, limited to the
dispersion and stabilization of metal particles, and therefore could
not explain the results obtained. Additionally, this is not compati-
ble with the well-known low capacity of gold for adsorbing molec-
ular oxygen.
Nevertheless, recent experiments in glycerol oxidation cata-
lyzed by gold supported on carbon showed that oxygen atoms
incorporated into the reaction products were originated from
hydroxide ions, which play a decisive role during the reaction,
and not from molecular oxygen, which is necessary for the
j
j
Au/AC1tt600: catalyst with an intermediate amount of oxygen-
containing functional groups (lower concentration of acid
groups) and intermediate activity.
0
Au/AC , Au/AC1tt900 and Au/AC1tt900-H2: catalysts corresponding
to basic supports (only with a limited concentration of oxygen-
ated surface groups, particularly those of acid nature) and high
activity.
Summarizing, supports with basic characteristics lead to more
active gold catalysts.
Yang et al. [31] tested a series of metal carbonate-supported
gold nanoparticles for the liquid-phase oxidation of benzyl alcohol
and observed that the catalytic performance depends strongly on
the basicity of the support material. They suggest that the initial
Table 6
Influence of the support on the selectivities of GLYCEA, GLYCOA, DIHA, and TARTA for Au catalysts prepared on activated carbons with different surface chemistries.
Catalyst t = 120 min GLY = 40%
X
X
GLY
S
GLYCEA
S
GLYCOA
S
DIHA
S
TARTA
S
GLYCEA
S
GLYCOA
S
DIHA
S
TARTA
Au/AC
0
0.72
0.32
0.78
0.82
0.62
0.64
0.61
0.62
0.15
0.18
0.17
0.15
0.18
0.13
0.13
0.18
0.05
0.05
0.06
0.05
0.59
0.65
0.57
0.61
0.15
0.17
0.20
0.15
0.18
0.13
0.13
0.20
0.04
0.05
0.05
0.04
Au/AC1tt600
Au/AC1tt900
Au/AC1tt900-H2
OH
HO
OH
OH
OH
HO
OH
HO
OH
O
HO
HO
O
e-
e-
e-
e-
e-
HO
HO
e-
e-
Au
Au
Au
II
I
e-
Oxygen-free AC
Oxygen-free AC
Oxygen-free AC
V
O2
O2
OH-
OH-
e-
e-
e-
OH-
OH-
Au
Au
Oxygen-free AC
IV
III
e-
Oxygen-free AC
Fig. 7. Role of the basic supports on the mechanism of glycerol oxidation.

E.G. Rodrigues et al. / Journal of Catalysis 281 (2011) 119–127
127
regeneration of hydroxide ions [34]. In fact, very poor or no activity
Acknowledgements
is observed when using gold catalysts without a relatively high
concentration of added base [24]. Glycerol oxidation is aided by
high-pH conditions, since the hydroxide species adsorbed over
gold have the ability to facilitate the activation of both the C-H
and the O–H bonds of glycerol (also adsorbed on the catalyst sur-
face), leading to the addition of electrons to the metal surface [34]
This work was carried out with the support of Fundação para a
Ciência e a Tecnologia (FCT) under research fellowship BD/45280/
2008 (E.G. Rodrigues) and FCT/FEDER in the framework of Program
COMPETE (project PTDC/EQU-ERQ/101456/2008). It was also par-
tially supported by Acção Integrada Luso-Espanhola N° E28/11
(Portugal) and Acciones Integradas with reference number
AIB2010PT-00377 (Spain). X. Chen acknowledges Ramón y Cajal
contract from Ministry of Science and Innovation of Spain.
(
see Fig. 7; step I). In this way, the oxidation mechanism does not
involve the dissociation of molecular oxygen to atomic oxygen. In-
stead, activation of O occurs by the formation and dissociation of
2
peroxide and hydrogen peroxide intermediates on the metal sur-
face, according to the following mechanism [34]:
References
ꢂ
ꢂ
ꢂ
ꢂ
O þ H
2
O ! HOO þ HO
ð2Þ
ð3Þ
ð4Þ
2
[1] C.H.C. Zhou, J.N. Beltramini, Y.X. Fan, G.Q.M. Lu, Chem. Soc. Rev. 37 (2008) 527.
[
2] K. Golz-Berner, L. Zastrow, US Patent 10 570 031, 2007.
ꢂ
ꢂ
ꢂ
ꢂ
HOO þ H
2
O ! H
2
O þ HO
[3] P. McMorn, G. Roberts, G.J. Hutchings, Catal. Lett. 63 (1999) 193.
2
[
4] M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi, C. Della Pina, Angew. Chem. Int.
Ed. 46 (2007) 4434.
HO þ eꢁ $ HO þꢂ
where ꢂ represents a site on the metal surface. The last step allows
for the removal of electrons added to the surface during the adsorp-
tion of hydroxide ions and regenerates these species. In that study
ꢂ
ꢁ
[
[
5] S. Demirel, K. Lehnert, M. Lucas, P. Claus, Appl. Catal. B 70 (2007).
6] K. Musialska, E. Finocchio, I. Sobczak, G. Busca, R. Wojcieszak, E. Gaigneaux, M.
Ziolek, Appl. Catal. A 384 (2010) 70.
[7] A. Villa, A. Gaiassi, I. Rossetti, C.L. Bianchi, K. van Benthem, G.M. Veith, L. Prati,
J. Catal. 275 (2010) 108.
[
[
8] L. Prati, M. Rossi, J. Catal. 176 (1998) 552.
9] F. Porta, L. Prati, J. Catal. 224 (2004) 397.
[
34], the support was not considered to be involved in the mecha-
nism. However, the presence of delocalized -electrons in basic
p
[10] G.C. Bond, D.T. Thompson, Catal. Rev. – Sci. Eng. 41 (1999) 319.
[
11] W.C. Ketchie, Y.-L. Fang, M.S. Wong, M. Murayama, R.J. Davis, J. Catal. 250
2007) 94.
12] F. Porta, L. Prati, M. Rossi, S. Coluccia, G. Martra, Catal. Today 61 (2000) 165.
carbon supports leads to a high electronic mobility, whereas oxy-
gen-containing functional groups (electron-withdrawing groups)
decrease the electron density and the electrical conductivity
(
[
[13] S. Demirel, P. Kern, M. Lucas, P. Claus, Catal. Today 122 (2007) 292.
[
[
14] E. Papirer, S. Li, J.B. Donnet, Carbon 25 (1987) 243.
15] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Carbon 37 (1999)
[
35,36]. The mobility of electrons to the basic support could poten-
tially reduce the excess of negative charge on gold and enhance the
possibility of further hydroxide bonding (Fig. 7, step II), whereas the
transition of this electrons from the support to the metal enhance
the regeneration of the hydroxide ions (Eqs. (2)–(4) and Fig. 7, steps
III and IV). Therefore, the mobility of the electrons from and to the
gold surface could potentially promote both adsorption and regen-
eration of hydroxide ions and thus the catalytic performance.
1
379.
[16] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Ind. Eng. Chem. Res.
6 (2007) 4110.
[
[
4
17] J.A. Menendez, J. Phillips, B. Xia, L.R. Radovic, Langmuir 12 (1996) 4404.
18] Y. Onal, S. Schimpf, P. Claus, J. Catal. 223 (2004) 122.
[19] E.G. Rodrigues, S.A.C. Carabineiro, X. Chen, J.J. Delgado, J.L. Figueiredo, M.F.R.
Pereira, J.J.M. Órfão, Catal. Lett. 141 (2011) 420.
20] C.Y. Yin, M.K. Aroua, W.M.A.W. Daud, Sep. Purif. Technol. 52 (2007) 403.
21] V. Gómez-Serrano, M. Acedo-Ramos, A.J. López-Peinado, C. Valenzuela-
Calahorro, Thermochim. Acta 291 (1997) 109.
[
[
[
[
22] C. Moreno-Castilla, M.V. Lopez-Ramon, F. Carrasco-Marin, Carbon 38 (2000)
4
. Conclusions
1995.
23] C.A. Leon y Leon, J.M. Solar, V. Calemma, L.R. Radovic, Carbon 30 (1992) 797.
A strong effect of the activated carbon surface chemistry on the
[24] S. Demirel-Gülen, M. Lucas, P. Claus, Catal. Today 102–103 (2005) 166.
[
[
[
[
25] S. Carrettin, P. McMorn, P. Johnston, K. Griffin, C.J. Kiely, G.J. Hutchings, Phys.
Chem. Chem. Phys. 5 (2003) 1329.
26] Y.A. Ryndin, O.S. Alekseev, P.A. Simonov, V.A. Likholobov, J. Mol. Catal. 55
(1989) 109.
27] S. Carrettin, P. McMorn, P. Johnston, K. Griffin, C.J. Kiely, G.A. Attard, G.J.
Hutchings, Top. Catal. 27 (2004) 131.
28] W.C. Ketchie, M. Murayama, R.J. Davis, J. Catal. 250 (2007) 264.
activity of Au/AC catalysts for glycerol oxidation was observed.
Gold particles with similar average sizes resulted in different per-
formances, depending on the amount of oxygenated groups on the
surface of the support used. As a result, the oxygen content on the
surface is a factor of large influence in the reaction. Basic oxygen-
free supports, characterized by a high density of free
p
-electrons,
[29] D. Wang, A. Villa, F. Porta, D. Su, L. Prati, Chem. Commun. (2006) 1956.
[30] N. Dimitratos, J.A. Lopez-Sanchez, D. Lennon, F. Porta, L. Prati, A. Villa, Catal.
Lett. 108 (2006) 147.
lead to an enhancement of the gold catalyst activity. These charac-
teristics can easily be achieved by thermal treatments at high tem-
peratures, which remove the oxygen-containing surface groups.
The role of the activated carbon surface chemistry was explained
on the basis of a recent published mechanism, by considering the
capability of oxygen-free supports to promote electron mobility.
The prepared catalysts lead to a high total selectivity to products
of commercial interest (GLYCEA + DIHA) of about 80%.
[
31] J. Yang, Y.J. Guan, T. Verhoeven, R. van Santen, C. Li, E.J.M. Hensen, Green Chem.
1 (2009) 322.
[32] J.J. Zhu, S.A.C. Carabineiro, D. Shan, J.L. Faria, Y.J. Zhu, J.L. Figueiredo, J. Catal.
74 (2010) 207.
1
2
[
[
33] H.H. Kung, M.C. Kung, C.K. Costello, J. Catal. 216 (2003) 425.
34] B.N. Zope, D.D. Hibbitts, M. Neurock, R.J. Davis, Science 330 (2010) 74.
[35] D. Sebastián, I. Suelves, R. Moliner, M.J. Lázaro, Carbon 48 (2010) 4421.
36] Z. Hashisho, M.J. Rood, S. Barot, J. Bernhard, Carbon 47 (2009) 1814.
[