Electrochemical activation of Au nanoparticles for the selective partial oxidation of methanol

Article abstract of DOI:10.1016/j.jcat.2014.06.022

An electrochemical catalyst based on Au nanoparticles dispersed in an yttria-stabilized zirconia (YSZ) matrix was prepared by reactive cosputtering of zirconium-yttrium and Au targets on a K-βAl₂O₃ solid electrolyte. The Au/YSZ catalyst film deposited on K-βAl₂O ₃ was found to be active in the partial oxidation of methanol with a high selectivity toward methyl formate. This configuration allowed to electrochemically promote a highly dispersed Au catalyst by K⁺ ions and allowed to decrease the amount of metal used in the solid electrolyte cell. The back-spillover of K⁺ ions sharply increased the H₂ and HCOOCH₃ production rates up to more than 9 and 5 times, respectively. A number of experiments confirmed that the observed electro-promotional effect did not depend on the rate of K⁺ supply nor on the operation mode (galvanostatic or potentiostatic) and only depended on the promoter coverage (total applied charge). The stability of the Au-YSZ catalyst film under the explored conditions was also verified.

Full text of DOI:10.1016/j.jcat.2014.06.022

Journal of Catalysis 317 (2014) 293–302
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Electrochemical activation of Au nanoparticles for the selective partial
oxidation of methanol
J. González-Cobos a, D. Horwat b,c, J. Ghanbaja b,c, J.L. Valverde a, A. de Lucas-Consuegra a,
⇑
a Departamento de Ingeniería Química, Facultad de Ciencias y Tecnologías Químicas, Universidad de Castilla-La Mancha, Avenida Camilo José Cela 12, 13005 Ciudad Real, Spain
b Université de Lorraine, Institut Jean Lamour, UMR7198, F-54011 Nancy, France
c CNRS, Institut Jean Lamour, UMR7198, F-54011 Nancy, France
a r t i c l e i n f o
a b s t r a c t
Article history:
An electrochemical catalyst based on Au nanoparticles dispersed in an yttria-stabilized zirconia (YSZ)
matrix was prepared by reactive cosputtering of zirconium–yttrium and Au targets on a K-bAl2O3 solid
electrolyte. The Au/YSZ catalyst film deposited on K-bAl2O3 was found to be active in the partial oxidation
of methanol with a high selectivity toward methyl formate. This configuration allowed to electrochem-
ically promote a highly dispersed Au catalyst by K+ ions and allowed to decrease the amount of metal
used in the solid electrolyte cell. The back-spillover of K+ ions sharply increased the H2 and HCOOCH3
production rates up to more than 9 and 5 times, respectively. A number of experiments confirmed that
the observed electro-promotional effect did not depend on the rate of K+ supply nor on the operation
mode (galvanostatic or potentiostatic) and only depended on the promoter coverage (total applied
charge). The stability of the Au–YSZ catalyst film under the explored conditions was also verified.
Ó 2014 Elsevier Inc. All rights reserved.
Received 15 May 2014
Revised 17 June 2014
Accepted 18 June 2014
Keywords:
Electrochemical promotion
NEMCA effect
Permanent EPOC
Potassium
Selective oxidation
Methanol
Sputtering
Au–YSZ catalyst
Gold nanoparticles
1. Introduction
prepared by leaching Ag from an Au–Ag alloy [6], gold nanoparti-
cles supported on highly porous carbon materials [7] and, espe-
Methanol is one of the most important feedstock in the chemi-
cal industry, being involved in the production of formaldehyde,
MTBE, acetic acid, and methyl methacrylate, among others. It is
also a very promising liquid energy carrier that can be used as fuel
in direct methanol fuel cells (DMFC) [1,2] or as a hydrogen source
through the steam reforming (SRM) and partial oxidation (POM)
reactions. The main advantages of methanol are its high H/C ratio,
its moderate reaction temperatures (150–350 °C), and its easy
transport, storage, and handling. Moreover, despite being mainly
produced from natural gas, it can be also obtained from a wide
variety of biomasses [1]. The most extensively catalysts used in
SRM and POM are based on copper, zinc, or noble metals such as
platinum or palladium [1]. Until a few decades ago, gold has been
generally considered much less catalytically active due to its diffi-
culty to both adsorb and dissociate molecular oxygen [3] and
remain in a high dispersion state because of its low melting point
as compared to Pt or Pd [4,5]. However, several studies have shown
excellent catalytic properties of unsupported nanoporous gold
cially, on certain metal oxides (e.g., TiO2, CeO2, Fe2O3, ZnO, or
ZrO2) [4,5]. In fact, numerous applications of gold-based catalysts
in the field of environmental catalysis have been recently reviewed
[4]. Specifically, gold catalysts supported on oxide supports have
demonstrated a very high activity in CO oxidation [5,8,9] even at
temperatures below 0 °C [5], but also in other catalytic reactions
such as the oxidation of methanol [9,10–14] and many other
organic compounds [15]. In the case of the methanol partial oxida-
tion processes [6,9,10–14], it should be noted the complexity of the
reaction pathways that may take place. Effectively, a large number
of side reactions and intermediate species are generally involved,
making possible to obtain a wide variety of compounds. Along with
the main partial oxidation products, H2 and CO2, not only CO and
H2O can be coproduced, but also CH4 [10–13], CH3OCH3 (dimethyl
ether) [14], H2CO (formaldehyde) [13,14], (CH3O)2CH2 (dim-
ethoxymethane) [14], and HCOOCH3 (methyl formate) [11–14].
In this sense, Au-based catalysts are acquiring a great prominence
against the Pt and Pd, due to their capability to reach a very high
selectivity in the methanol partial oxidation toward products of
well-known interest in chemical manufacturing, such as formalde-
hyde [9] and methyl formate [6]. The influence of the gold particle
⇑
Corresponding author. Fax: +34 926295437.
E-mail address: Antonio.Lconsuegra@uclm.es (A. de Lucas-Consuegra).
http://dx.doi.org/10.1016/j.jcat.2014.06.022
0021-9517/Ó 2014 Elsevier Inc. All rights reserved.

294
J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
size and the oxide support on the catalytic performance in the
methanol oxidation has been widely studied [9,10–14] as well as
the benefits from adding a second active metallic component
[12,13].
which corresponds to a gold content of 7.5 at.% or a Au nanoparti-
cles loading of 40
l
g Au/cm2 (2.03 ꢁ 10ꢀ7 mol Au/cm2). More
details on the experimental procedure and principle of segregation
of metallic gold into the YSZ matrix during the formation of the
film can be found in previous studies [25]. For characterization
purposes, the same layer was prepared on glass substrates.
Moreover, since the Au–YSZ nanocomposite was electrically non-
conductive from its preparation, a silver mesh (Fuel Cell Materials,
AG-M40-100) was used as the current collector over the catalyst
surface in order to carry out the EPOC experiments. The three elec-
trodes (working, counter and reference) were connected to an
Autolab PGSTAT320-N potentiostat–galvanostat (Metrohm Auto-
lab) using gold wires (Alfa Aesar, 99.95% purity). The results
reported here were obtained in both operation modes: galvano-
static and potentiostatic, i.e. by setting the electrical current (from
An alternative way to in situ enhance the catalytic activity and
selectivity of metal particles is the well-known phenomenon of
electrochemical activation of catalysis also called electrochemical
promotion of catalysis (EPOC) or NEMCA effect (non-faradaic elec-
trochemical modification of catalytic activity) [16]. This is based on
the use of electrochemistry to activate a heterogeneous catalyst by
means of the controlled migration of promoter ions, via an applied
electric current or potential, from an electro-active catalyst sup-
port (solid electrolyte) to the active catalyst component generally
in form of a continuous thin film [16]. This phenomenon has been
applied on a wide variety of catalysts and in a large number of
important industrial and environmental reactions as has been
recently reviewed [17]. In particular, the electrochemical promo-
tion has been applied to enhance the activity and selectivity of Pt
catalysts in the methanol partial oxidation (POM) by electrochem-
ically supplying K+ [18,19] and O2ꢀ ions [20]. However, to the best
of our knowledge, this phenomenon has not been widely applied to
catalytically activate gold nanoparticles. Few works can be found
in the literature related to the application of the EPOC phenome-
non by using Au catalysts: in solid oxide fuel cell configuration
for the oxidation of CO [21] and CH4 [22] with O2ꢀ promoter ions,
in the liquid-phase SO2 oxidation with molten V2O5–K2S2O7 cata-
lyst [23] and in the preferential oxidation of CO in a low-tempera-
ture proton exchange membrane (PEM) fuel cell reactor promoted
by H+ ions and O2 crossover through the membrane [24].
In this work, we report for the first time the electrochemical acti-
vation of Au nanoparticles dispersed in yttria-stabilized zirconia
(YSZ) with electropositive alkali ions by using a ceramic K-bAl2O3
solid electrolyte in the catalytic partial oxidation of methanol
(POM). A novel solid electrolyte cell configuration based on a Au–
YSZ nanocrystalline thin film was prepared by sputter deposition
and tested under EPOC conditions by controlling and optimizing
the K+ promoter supply. The proposed configuration allowed to
both electrochemically activate Au nanoparticles highly dispersed
in an YSZ matrix and decrease the amount of metal used in the solid
electrolyte cell. The prepared Au–YSZ/K-bAl2O3/Au solid electrolyte
cell was characterized and tested under POM reaction conditions,
checking its reproducibility and stability.
0 to ꢀ20 A) and potential (from +2 to ꢀ2 V), respectively,
l
according to the procedure generally used in conventional three-
electrode electrochemical cells [16].
2.2. Electrochemical catalyst characterization
X-ray diffraction (XRD) was performed on the Au–YSZ nano-
composite layer before catalytic tests with a AXS Brücker D8
advance diffractometer using
a Cu anode (Cu Ka radiation,
k = 1.54 Å). The thermal stability of the nanocomposite layer was
verified by the same technique after a long catalytic test of 12 h
at 320 °C under POM reaction conditions. Transmission electron
microscopy (TEM) investigation and high-resolution TEM (HRTEM)
imaging of the tested sample after all the catalytic experiments
were carried out with a JEOL ARM 200F TEM/STEM – Cold FEG
(point resolution: 0.19 nm) fitted with a GIF Quatum ER. Cyclic vol-
tammetry measurements were also obtained with the potentio-
stat–galvanostat in order to investigate the polarization of the
solid electrolyte cell under POM reaction conditions. Before the
cyclic voltammetry experiment, the sample was kept at +2 V for
30 min in order to define a reference state (Au–YSZ film clean of
K+ promoter ions) as will be discussed later. Then, the potential
was cyclically varied between +2 and ꢀ2 V at a scan rate of
80 mV/s.
2.3. Catalytic activity measurements
The activity measurements were carried out in an experimental
setup used in previous works with a single chamber solid electro-
lyte cell reactor configuration [18,19]. The reaction gases (Praxair,
Inc.) were certified standards of air and Ar (99.999% purity), this
latter used as the carrier gas. The gas flow rates were controlled
by a set of calibrated mass flowmeters (Bronkhorst EL-FLOW),
while methanol was fed by sparging Ar through a thermostated
saturator (at 40 °C) containing liquid CH3OH (Panreac, 99.8% pur-
ity). All lines placed downstream from the saturator were heated
above 100 °C to prevent condensation. The catalytic experiments
were carried out at atmospheric pressure with an overall gas flow
rate of 6 NL hꢀ1, temperatures up to 300 °C and a feed composition
of CH3OH/O2 = 5.9%/0.43% (Ar balance). Reactant and product gases
were on line analyzed by using a double channel gas chromato-
graph (Bruker 450-GC) equipped with Hayesep Q-Molsieve 13X
consecutive columns and a CP-Wax 52 CB column, along with ther-
mal conductivity and flame ionization detectors, respectively. The
detected reaction products were as follows: H2, CO2, H2O,
HCOOCH3 (methyl formate), and traces of dimethoxymethane
and other unidentified compound, which could be dimethyl ether.
Without considering these latter compounds, the error in the car-
bon atom balance did not exceed 1%, which indicated no consistent
loss of material and no significant coking of the catalyst electrode.
Blank experiments were also performed under the same reaction
2. Experimental
2.1. Electrochemical catalyst preparation
The electrochemical catalyst consisted of a thin continuous
gold/yttria-stabilized zirconia (Au–YSZ) film (geometric area of
3.14 cm2) composed by fine gold nanoparticles dispersed in a YSZ
matrix, which have been deposited by magnetron sputtering on a
20-mm-diameter, 1-mm-thick K-bAl2O3 (Ionotec) disk. Firstly,
inert Au counter (C) and reference (R) electrodes were deposited
on one of the sides of the solid electrolyte by applying thin coatings
of gold paste (Gwent Electronic Materials C1991025D2) followed
by calcination at 800 °C for 2 h (heating ramp of 5 °C/min). Then,
the Au–YSZ nanocomposite catalyst film, which also behaved as a
working electrode (W), was deposited by reactive cosputtering of
Zr–Y and Au targets in an argon-oxygen gas mixture ensuring the
complete oxidation of the Zr and Y atoms in a nanocrystalline
nearly amorphous YSZ matrix. The deposition time was set such
that the resulting Au–YSZ film thickness was around 170 nm, as
measured by STEM imaging (not shown). Oxygen gas flow rate
and current applied to the targets were fixed such that the volume
fraction of gold particles into the composite layer was 11 vol.%,

J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
295
conditions without any catalyst and only with the Ag current col-
lector in order to determine its negligible catalytic contribution.
assisted diffusion. This way, the size of the particles is slightly
over-estimated, being the measurements in accordance with the
XRD data. This fact would demonstrate the good stability of the
gold nanoparticles after exposure to reaction conditions. Larger
particles, with diameters ranging between 15 and 50 nm, but
mostly located at the substrate/film interface, were also observed
(Fig. 2c). The presence of larger size gold crystallites after catalytic
tests was also observed in previous studies of methanol oxidation
over supported gold catalysts due to aggregation of smaller Au
particles [9,11,14]. Atom scale resolution HRTEM and Fourier filter
transform (FFT) of the images (not shown) confirmed the presence
of particles consisting of face centered cubic (FCC) metallic assem-
bly of gold atoms. For instance, the {111} atomic arrangement of
FCC gold was observed for particles in Fig. 2d.
Fig. 3 shows the steady-state variation in the production rates
(hydrogen, carbon dioxide, and methyl formate) at different reac-
tion temperatures obtained in the partial oxidation of methanol
with the Au–YSZ catalyst film as well as the blank results (feed
composition of CH3OH/O2 = 5.9%/0.43% and Ar balance). In the case
of the electrocatalytic system (current collector + Au–YSZ catalyst),
the catalytic activity was measured at each temperature under
open circuit conditions, i.e. without applying electric current, after
the imposition of a catalyst potential (VWR) of +2 V for 30 min. In
agreement with previous studies with K-bAl2O3-based electrolytes
[18,19,27,28], the purpose of this previous positive polarization
was to remove from the Au–YSZ catalyst film the K+ ions that ther-
mally migrated from the solid electrolyte. This thereby defined a
reference state (unpromoted state). Both the reactor blank and
the silver mesh showed a negligible activity under the studied
reaction conditions as compared to that of the electrochemical cat-
alyst. On the other hand, it was confirmed that the Au–YSZ catalyst
film presented certain catalytic activity in the methanol partial oxi-
dation under unpromoted conditions. While large Au particles and
bulk gold are generally considered to be inert, catalysts based on
gold nanoparticles supported on different metal oxides have been
reported by several authors to be catalytically active in this reac-
tion [9,10–14]. In fact, the smaller the Au particle size, the higher
the methanol partial oxidation rate as well as the H2 selectivity
vs. CO and CH4 [11–13]. In particular, gold nanoparticles dispersed
on YSZ have already been tested in the CO oxidation reaction [8]
but, to the best of our knowledge, this is the first time that the cat-
alytic activity of an Au–YSZ composite film in the methanol partial
oxidation is studied. The H2 and CO2 production rates observed in
Fig. 3 were expected from the main POM reaction (Eq. (1)).
3. Results and discussion
3.1. Characterization of the electrochemical catalyst and blank
experiments
Fig. 1 shows the X-ray diffractograms for the fresh and the aged,
after 24 h at 320 °C under reaction conditions, nanocomposite
layer deposited on the glass substrate. Both diffractograms exhibit
the characteristic diffraction peaks corresponding to nanocrystal-
line gold at 2h = 38.2°, 44.4°, and 64.5°. Nevertheless, the diffracted
intensity was slightly increased after exposure to reaction condi-
tions. The average size of the nanoparticles was estimated from
the width of the (111) diffraction peak using the Scherrer’s for-
mula [26]. An average particle size of 3.3 nm in the fresh state
was measured, which is in line with the STEM analysis results.
The average particle size slightly rose up to 4.3 nm after exposure
to reaction conditions, which indicates a minor diffusion of the
gold atoms and a good thermal stability of the nanocomposite
layer. No diffraction peaks corresponding to gold oxides (AuO or
Au2O3) were detected. No diffraction lines of YSZ were identified
in the X-ray diffractograms due to the ultrafine grained, nearly
amorphous, nature of the YSZ matrix, which was confirmed by
HRTEM (not shown here). The XRD signal of YSZ could only be
detected after long time annealing at 500 °C, which is far above
the temperatures experienced by the samples in this study.
Analyses of the films deposited on the K-bAl2O3 led to similar con-
clusions, therefore indicating a limited influence of the substrate
on the thermal stability of the nanocomposite layers.
Further details of the aged gold particles after exposure to reac-
tions conditions were obtained using TEM and HRTEM. Fig. 2
shows bright field images of selected zones in this sample.
Fig. 2a and b shows extended and local views of the typical distri-
bution of gold nanoparticles in the core of the coatings and at the
surface exposed to the reaction mixture, respectively. The YSZ
matrix appears in clear, while dark zones correspond to individual
Au nanoparticles dispersed in it. The particles were round shaped
with a typical size distribution ranging between 2 and 10 nm.
Nevertheless, it must be pointed out that diffusion occurred upon
exposure to the electron beam during experiments, which trig-
gered a slight increase in the particle diameter by thermally
CH3OH þ 0:5O2 ! 2H2 þ CO2
ð1Þ
However, the most abundant product was methyl formate
(HCOOCH3) with a selectivity of 78% (with respect to carbon),
calculated
as
(100 ꢁ FHCOOCH3,out ꢁ 2)/(FCO2;out þ 2xFHCOOCH ;out),
3
where FHCOOCH3,out and FCO2,out are the molar flow rates of methyl
formate and carbon dioxide, respectively, at the outlet of the reac-
tor. There are many possible intermediate and side reactions
involved in the methanol oxidation as very recently summarized
[14]. Methanol dissociation resulting in methoxy species and
hydroxyl group is generally considered the first step of the
sequence. Then, formaldehyde, formic acid, and carbon dioxide
may be produced, in that order, from successive oxidation reactions,
but also dimethyl ether, dimethoxymethane, methyl formate, and
carbon monoxide may be obtained from the dehydrogenation of
the former compounds. Hence, the methyl formate could be formed
from the following consecutive steps [14]:
After 320°C
As deposited
30
40
50
60
70
80
CH3OH þ 0:5O2 ! H2CO þ H2O
H2CO þ 0:5O2 ! HCOOH
HCOOH þ CH3OH ! HCOOCH3 þ H2O
ð2Þ
ð3Þ
ð4Þ
Fig. 1. XRD diffractograms of the Au–YSZ nanocomposite layer as deposited and
after catalytic test under methanol partial oxidation conditions (CH3OH/O2 = 5.9%/
0.43%, Ar balance) at 320 °C.

296
J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
Fig. 2. Bright field micrographs of Au–YSZ nanocomposite film after catalytic tests under methanol partial oxidation conditions (CH3OH/O2 = 5.9%/0.43%, Ar balance) at
320 °C: typical microstructure in the core through the film thickness and at the film surface (a and b) and at the substrate/film interface (c). HRTEM micrograph of a selected
region in the same sample shows the atom scale arrangement of the nanoparticles (d).
can be associated with the formation of surface compounds
derived from the interaction among the different reactants
adsorbed on the catalyst-working electrode and the alkali ions
coming from the solid electrolyte [18,19,27,28,30,31]. No electro-
catalytic processes have been considered on the gold counter elec-
trode since it was constituted by large particles of negligible
adsorption activity [18,19,27,28,30,31]. In the present case, due
to the migration of K+ ions to the Au–YSZ film and the presence
of O2, CO2 and H2O in the reaction atmosphere, K-derived surface
compounds such as potassium oxides/superoxides [27,32] or car-
bonates/bicarbonates [18,27,33] may be formed. Then, the peaks
observed during the anodic scan could be linked to the decompo-
sition of these promoter-derived species and the migration of the
K+ ions back to the solid electrolyte.
Under lack of surface oxygen, HCOOCH3 production on supported
gold catalysts has also been postulated as a direct dehydrogenation
of methanol (Eq. (5)) [11,12].
2CH3OH $ HCOOCH3 þ 2H2
ð5Þ
In any case, the weak adsorption of the methanol partial oxidation
products on Au, as compared to other noble metals, makes the gold-
based catalysts suitable for the low-temperature selective oxidation
of methanol to oxygenated compounds other than CO2, such as
HCOOCH3 [6]. Moreover, the methyl formate is of interest in the
chemical industry as a solvent and a precursor in the synthesis of
formic acid, formamide, and dimethylformamide, among other
organic compounds [29].
Prior to the catalytic activity measurements under electrochem-
ical promotion conditions, the Au–YSZ/K-bAl2O3/Au electrochemi-
cal catalyst was in situ characterized by cyclic voltammetry
under the same POM reaction conditions at 280 °C, by recording
the current (I) variation with the applied potential (VWR) between
+2 and ꢀ2 V (Fig. 4). Previously, the electrochemical catalyst was
kept at +2 V for 30 min in order to define a Au–YSZ film free of
potassium ions (reference state). Firstly, it can be observed that
the cyclic voltammetry was reversible and almost completely
reproducible from the second cycle at the explored potential range.
It has been widely established that the cathodic peaks observed in
the cyclic voltammetries carried out with alkali ionic conductors
Hence two conclusions can be clearly obtained from this
voltammogram:
– The applied potential of +2 V was enough to electrochemically
decompose the promoter surface compounds, and hence, this
potential can be defined as a reference un-promoted state for
the further EPOC experiments.
– The reproducibility of the voltammogram after the 2nd cycle
revealed a good stability of the Au/YSZ catalyst during the
polarizations under reaction conditions. The reproducibility of
the Au catalytic activity will be also confirmed in the following
experiments.

J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
297
0.6
0.5
0.4
0.3
0.2
0.1
0
12
5
4
3
2
1
3
VWR = 2 V
I = -5 μA
I = 0 A VWR = 2 V
(a)
(b)
(c)
Promoter supply / molK+ x 10-7
0
0.5
1
1.5
2
(a)
Reactor blank
2
9
6
3
Currentcollector
Currentcollector +
Au-YSZ catalyst
1
0
2
1.5
1
Reactor blank
0
2
0
4
3
2
1
0
-1
6
4
2
(b)
Currentcollector
Currentcollector +
Au-YSZ catalyst
12
6
0.5
0
3
1
0
Reactor blank
2.5
2
Currentcollector
0
0
Currentcollector +
Au-YSZ catalyst
1.5
1
0
25
50
75
100 125 150 175
Time / min
0.5
Fig. 5. Variation of H2 (a), CO2 and HCOOCH3 (b) production rates as well as the
corresponding enhancement ratios ( i) and catalyst potential (VWR) (a) vs. time
upon different potentiostatic and galvanostatic transitions. Methanol partial
q
0
240
260
280
300
320
oxidation conditions: CH3OH/O2 = 5.9%/0.43%, Ar balance, 280 °C. Electrochemical
Temperature / ºC
promotion under the imposition of I = ꢀ5
lA.
Fig. 3. Steady-state H2 (a), CO2 (b) and methyl formate (HCOOCH3) (c) production
rates at different reaction temperatures during the blank experiments and the
catalytic tests using the Au–YSZ composite film (un-promoted state). Methanol
partial oxidation conditions: CH3OH/O2 = 5.9%/0.43%, Ar balance.
in both the hydrogen, carbon dioxide, and methyl formate produc-
tion rates and the catalyst potential (VWR) to the following step
changes: a positive potential (VWR = +2 V) for 32 min, a negative
current (I = ꢀ5
lA) for 65 min, open circuit (I = 0) for 37 min, and
again a positive potential (VWR = +2 V) for 37 min. In order to quan-
0.8
0.4
0
tify the magnitude of the electropromotional effect, the rate
enhancement ratio of each compound (qi) is also shown in the
figure, calculated by the following equation:
ri
ri;0
qi
¼
ð6Þ
-0.4
-0.8
where ri and ri,0 are the promoted (VWR < +2 V) and unpromoted
(VWR = +2 V) catalytic production rates, respectively, of the
corresponding compound. Moreover, the inserted abscissa along
the negative polarization indicates the accumulated promoter sup-
ply, which varies linearly vs. time in the case of galvanostatic exper-
iments and is calculated from the following general equation:
1st cycle
-1.2
2nd cycle
-1.6
-2
3rd cycle
Z
t
jIjdt
Promoter supply=mol Kþ
¼
ð7Þ
nF
0
-2.5
-1.5
-0.5
0.5
1.5
2.5
where n is the potassium ion charge in this case (+1), F is the Fara-
day constant (96484.6 C), and t is the polarization time.
VWR / V
Fig. 4. Current variation (I) vs. the applied potential (VWR) of the electrochemical
catalyst during the cyclic voltammetry between +2 and ꢀ2 V (scan rate = 80 mV/s)
under methanol partial oxidation conditions (CH3OH/O2 = 5.9%/0.43%, Ar balance)
at 280 °C.
The behavior of the electrochemical catalyst has been evaluated
upon the different polarization regimes. Firstly, the application of a
positive potential VWR = +2 V ensured the removal of K+ ions previ-
ously migrated to the catalyst surface by thermal migration or as a
consequence of the foregoing experiments [18,19,27,28] and
resulted in a Au catalytic activity practically coincident with that
observed at the same temperature under the open circuit condi-
3.2. Electrochemical promotion experiments
In order to study the phenomenon of electrochemical promo-
tion of catalysis (EPOC) on the Au–YSZ catalyst film, several closed
and open circuit transitions were carried out under POM reaction
conditions (CH3OH/O2 = 5.9%/0.43% and Ar balance)) at 280 °C.
The unpromoted catalyst was found to be inactive below this
temperature as shown in Fig. 3. Fig. 5 shows the response vs. time
tions in Fig. 3. Then, at t = 32 min, a negative current I = ꢀ5
lA
(current density, i = 1.6
l
A/cm2) was applied between the
Au–YSZ catalyst-working electrode and the Au counter electrode,
i.e. potassium promoter ions were electrochemically pumped from
the solid electrolyte to the Au–YSZ catalyst surface at a rate
I/F = 5.18 ꢁ 10ꢀ11 mol K sꢀ1 according to the Faraday equation.

298
J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
effect also sustains the stability of the Au nanoparticles supported on
the YSZ under EPOC conditions. Hence, it is clear that the use of the
YSZ support allowed to disperse the metal nanoparticles, thus avoid-
ing the metal sintering effect, very typical in this kind of catalytic
system.
The increase in the accumulated potassium supply was found to
enhance the overall methanol conversion (not shown) and increase
the reaction rate of hydrogen (Fig. 5a), carbon dioxide, and methyl
formate (Fig. 5b). Selectivities toward H2 and HCOOCH3 were in
turn enhanced vs. those toward CO2 and H2O. At the same time,
the applied negative current caused a decrease in the catalyst
potential measured between the working and reference electrodes
(Fig. 5a), as typically observed in studies with alkali solid electro-
lytes [28,30,34]. The trend of the catalytic activity instead depends
on the NEMCA behavior of the system. According to the theory of
electrochemical promotion [16], the back-spillover of potassium
ions onto the Au–YSZ catalyst film under applied negative currents
led to the decrease in the catalyst potential and work function. It
strengthened the Au chemical bond with the electron acceptor
adsorbates (oxygen) and weakened that with the electron donors
(methanol) [20]. In fact, Broqvist et al. [35] already reported the
interest of the alkali compounds as ‘‘oxygen attractors’’ for signifi-
cantly improving the activity of CO oxidation on an Au-based cat-
alyst. It should be noted that the methanol conversion obtained in
this work did not exceed the 10% probably due to the low surface
area of the active catalyst film and reactants bypass. However, very
interestingly, the tested Au nanoparticles dispersed in a YSZ matrix
were electrochemically promoted by K+ ions leading to an
electrophilic NEMCA behavior under the explored POM reaction
conditions. Other studies have also successfully promoted metal
catalysts dispersed on YSZ both by alkali electropromotion [36]
and by alkali conventional promotion [37]. In the present work,
maximum production rates of 3.7 ꢁ 10ꢀ8 mol H2 sꢀ1, 4.3 ꢁ 10ꢀ8
mol CO2 sꢀ1, and 1.5 ꢁ 10ꢀ7 mol HCOOCH3 sꢀ1 were achieved at
the end of the negative polarization, which corresponded to
Fig. 6a and b shows the results obtained in two additional
experiments under the same POM reaction conditions, which con-
sisted of the same potentiostatic and galvanostatic transitions than
the previous one (Fig. 5), but applying higher currents during the
negative polarization: I = ꢀ10
lA (i = 3.2 l
A/cm2) in Fig. 6a-1 and
a-2, and ꢀ20 A (i = 6.4
l
l
A/cm2) in Fig. 6b-1 and b-2. The applica-
tion of a potential of +2 V at the beginning and the end of each
experiment defined again a Au–YSZ free of K-derived species
(unpromoted state). In this way, all the reaction rates remained
almost at the same values upon every positive polarization, con-
firming the complete reversibility of the promotional effect. Then,
as also observed in the first experiment during the application of a
negative current, the migration of potassium ions to the catalyst
surface at a rates, I/F = 1.04 ꢁ 10ꢀ10 mol K sꢀ1 (Fig. 6a-1 and a-2)
and 2.07 ꢁ 10ꢀ10 mol K sꢀ1 (Fig. 6b-1 and b-2) sharply enhanced
the gold catalytic activity. However, it can be found under the
polarizations at ꢀ10
l
A and, more clearly, at ꢀ20
lA that all the
production rates decreased from a certain accumulated amount
of supplied promoter, which was not achieved upon the imposition
of ꢀ5
lA. This poisoning effect could be attributed to the excessive
formation of alkali-derived surface compounds and the concomi-
tant blocking of active sites, as widely observed with both Na-
[37–39] and K-based [18,19,32] promoted systems. Moreover, the
catalyst potential measured after 65 min of negative polarization
decreased from ꢀ0.55 V (Fig. 5a) to ꢀ0.60 V (Fig. 6a-1) and
ꢀ0.74 V (Fig. 6b-1), due to an increase in the maximum potassium
coverage achieved in each experiment. A permanent EPOC was also
observed in these experiments upon the subsequent current inter-
ruption, although it was inevitably affected by the mentioned poi-
soning effect. For instance, in the case of the hydrogen production,
enhancement ratios (qi) of 8.9, 2.5 and 5.3, respectively, with
respect to the production rates obtained under unpromoted condi-
tions. Such an improvement in the catalytic activity was caused by
an accumulated promoter supply of 2.02 ꢁ 10ꢀ7 mol K+ (Eq. (7)).
þ
Hence, an optimum potassium coverage, hK , of 0.32 can be
permanent rate enhancement ratios (cH ) of 8.7, 6.4 and 3.2 can be
estimated in this experiment by considering the total number of
deposited Au sites, N = 6.38 ꢁ 10ꢀ7 mol Au, according to the
following equation:
2
observed at the end of the open circuit imposition in Fig. 5a,
Fig. 6a-1 and b-1, respectively. This parameter has been calculated
using the following equation:
R
t
0
jIjdt
nF
ri;per
ri;0
þ
hK
¼
ð8Þ
ci
¼
ð9Þ
N
It is also interesting to note that upon current interruption (open cir-
cuit conditions, at t = 97 min) a permanent EPOC phenomena was
observed probably due to the remaining of the promoter phases
on the catalyst surface under open circuit conditions. Thus, all the
product reaction rates were kept at their electro-promoted values.
In addition, a very slight increase in the catalyst potential close to
the open circuit potential value typically observed in this work
(V0WR ꢂ ꢀ0.4 V) was detected. This kind of irreversible promotional
effect (Permanent EPOC or Permanent NEMCA) has been reported
elsewhere for alkali-electropromoted systems [27,28,34]. It has been
mainly attributed to the high stability under reaction conditions of
certain promoter species formed on the catalyst surface, like potas-
sium oxides or superoxides [27,28]. It is very important to highlight
the practical application of the observed irreversible NEMCA behav-
where ri,per is the permanent catalytic production rate of the i com-
pound following the promoted state. Nevertheless, the optimum
amount of supplied K+ ions which maximized the catalytic activity
was found to be the same in each experiment (around 3 ꢁ 10ꢀ7
mol K+, which corresponded to a potassium coverage, hK ;opt, of
þ
about 0.5). In addition, the maximum values of the reaction rates
obtained were practically the same. Specifically, the H2, CO2 and
HCOOCH3 production rates were in all cases enhanced around 9.2,
2.6 and 5.5 times, respectively, under optimum electrochemically
promoted conditions. Then, it is interesting to note that the applied
current had no influence on the behavior and magnitude of the
electropromotional effect, but only on the speed of the promotion
process. Finally, the stability of the Au/YSZ catalyst film along the
reproducible behavior in the different experiments was also
demonstrated.
ior: not only a very low current (a few lA) is enough to enhance the
catalytic performance of the gold nanoparticles by several times, but
also a first polarization step may be already sufficient to maintain
this catalytic modification for a certain time, with the consequent
electricity saving. Finally, the imposition of a potential of +2 V again
at t = 134 min restored the initial production rates values within a
few minutes, leading to a completely reversible electropromotional
effect. It was due to the decomposition of the promoter phases and
the electrochemical pumping of the K+ ions back to the solid electro-
lyte in good agreement with the cyclic voltammetry. This reversible
The magnitude of the electrochemical promotion effect is also
typically described by the Faradaic efficiency (K) defined by:
D
r
K
¼
ð10Þ
I
nF
where
Dr is the promotion-induced change in the catalytic rate and
n is the number of electrons involved in the corresponding electro-
catalytic reaction. Under the studied POM reaction conditions at

J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
299
5
4
3
2
1
3
4
3
2
1
3
12
9
VWR = 2 V
I = -10 μA
I = 0 A VWR = 2 V
VWR = 2 V
I = -20 μA
I = 0 A VWR = 2 V
10
8
Promoter supply / molK+ x 10-7
Promoter supply / molK+ x 10-7
0
1
2
3
4
0
2
4
6
8
(a-1)
(b-1)
2
2
6
4
2
1
1
6
3
0
0
6
4
0
2
0
2
0
4
-1
0
3
2
1
0
-1
6
4
2
0
(a-2)
(b-2)
3
2
1
0
12
6
12
6
1
0
1
0
2
0
0
0
0
25
50
75 100 125 150 175
0
25
50
75 100 125 150 175
Time / min
Time / min
Fig. 6. Variation of H2, CO2 and HCOOCH3 production rates as well as the corresponding rate enhancement ratios (
potentiostatic and galvanostatic transitions. Methanol partial oxidation conditions: CH3OH/O2 = 5.9%/0.43%, Ar balance, 280 °C. Electrochemical promotion under the
imposition of I = ꢀ10 A (a-1 and a-2), and ꢀ20 A (b-1 and b-2).
q) and catalyst potential (VWR) vs. time upon different
l
l
280 °C, the increase in the oxygen catalytic consumption rate (not
shown) caused by the optimum promoter coverage was around
2.1 ꢁ 10ꢀ7 mol O sꢀ1, which was 4075 times larger than the rate
of K+ supply (I/F) to the catalyst film under the application of
-2500
-2000
-1500
-1000
-500
(a)
H2
CO2
ꢀ5
l
A (I/F = 5.18 ꢁ 10ꢀ11 mol K sꢀ1). The corresponding Faradaic
HCOOCH3
efficiency (
K
) would be ꢀ8150, i.e. by considering the equivalent
oxygen electrocatalytic reaction (n = 2 in Eq. (9)). However, in the
present study the source of promoter ions is not any O2ꢀ conductor
material such as YSZ, but K-bAl2O3. Thus, the Faradaic efficiency
must be considered with caution since the alkali ions cannot inter-
act with the gas reactants at the same extent [40]. Hence, the most
significant parameters from a catalytic point of view to characterize
the alkali-based promoted systems are the rate enhancement ratio
þ
0
25
20
15
10
5
(b)
H2
(q, Eq. (6)) and the promotion index (PIK ) [16], the latter being
CO2
defined according to the following equation:
HCOOCH3
D
r0
r
þ
PIK
¼
ð11Þ
þ
hK
þ
Fig. 7b depicts the promotion index ðPIK ) for the maximum produc-
tion rates of the different compounds (H2, CO2 and HCOOCH3)
observed in previous experiments (Figs. 5 and 6) under optimum
promoted conditions, and Fig. 7a shows the corresponding faradaic
efficiency values related to the rate of promoter supply (n = 1 in Eq.
0
0
-5
-10
I / A x 10-6
-15
-20
-25
Fig. 7. Influence of the applied negative current on the optimum faradaic efficiency
þ
(9)). This latter parameter has been denoted as KK and expressed in
relative to the rate of K+ supply (KK ;opt) (a) and the optimum promotion index
þ
units of mol product/mol K+. Namely, in both Fig. 7a and b, three
points are depicted which correspond with the maximum catalytic
rates observed in the three previous experiments (Figs. 5 and 6a and
b). Firstly, a clear EPOC effect can be observed in all the experiments
þ
(PIK ;opt) (b) of the different products. Data obtained from the experimental results
shown in Figs. 5 and 6.
þ
þ
(PIK > 0Þ with an electrophilic behavior (KK < ꢀ1), since the
the simultaneous production of up to more than 600 mol H2,
500 mol CO2 and 2300 mol HCOOCH3. Similar maximum production
rates were achieved in all the experiments despite increasing the
rate of promoter supply. Consequently, the faradaic efficiency val-
ues decreased as higher negative current was applied. On the other
hand, from Fig. 7b, it can be drawn that hydrogen showed the
highest promotion index values. In fact, the H2 production rate
presented the most pronounced increase with respect to the
application of negative currents and the consequent back-spillover
of K+ promoter ions increased the Au catalytic activity (I < 0,
Dr > 0),
as previously discussed. In Fig. 7a, methyl formate presented the
highest KK values, being the most abundant product obtained
under the studied reaction conditions. This parameter represents
the ratio between the mol of product and the mol of K+ electro-
chemically supplied. In this sense, under the application of
I = ꢀ5
þ
l
A, each K+ mol migrated to the Au–YSZ catalyst film allowed
unpromoted rate, as also reflected in the highest q values. Higher

300
J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
þ
optimum promotion indexes (PIK ;opt) were found for all the prod-
A positive polarization at VWR = +2 V before each different potential
allowed starting from a reference (unpromoted) state in all cases
[18,19,27,28]. The imposition of VWR = 0.5 V did not involve any
modification in the catalytic activity since it still remained above
the open circuit potential value. Then, the negative polarization
at VWR = ꢀ0.5 V caused a sharp increase of all the production rates
as a result of the back-spillover of the promoter ions from the
K-bAl2O3 electrolyte to the Au–YSZ catalyst-working electrode
ucts under the imposition of the lowest negative current,
I = ꢀ5
lA, because these data were obtained from the maximum
production rates observed at the end of this polarization step
(Fig. 5), which did not correspond with the actual optimum electro-
promoted conditions. With such a low rate of K+ supply in the first
experiment (5.18 ꢁ 10ꢀ11 mol K sꢀ1), the polarization time should
have been extended for another 35 min to achieve the optimum
potassium coverage (around 0.5). In this sense, the slight difference
(VWR 6 ꢀ0.5 V,
Dr > 0). As noted in previous galvanostatic experi-
þ;opt
between the depicted PIK
values obtained at I = ꢀ10
lA and
ments, the observed NEMCA effect could be explained in terms of
the work function variation and the consequent change in chemi-
sorptive bond strengths of the catalyst surface [16]. Very interest-
ingly, under this negative potential the steady-state H2, CO2 and
HCOOCH3 production rates were close to those obtained under
optimum electropromoted conditions in previous experiments,
ꢀ20
lA could be attributed to the non-complete coincidence in
the respective potassium coverages, which was probably due to
slight differences in the GC analysis time. However, fixed promotion
index values of around 16, 3 and 9 for hydrogen, carbon dioxide and
methyl formate, respectively, could be considered, regardless the
magnitude of the applied negative current, under optimum electro-
with enhancement ratios (qi) of around 8, 2 and 5, respectively.
þ;opt
chemically promoted conditions (hK
ꢂ 0:5). Hence, in contrast to
The associated amount of migrated promoter at the end of this
polarization was 3.58 ꢁ 10ꢀ7 mol K+, which was calculated from
the integration of the current vs. time curve (not shown here) via
þ
the KK parameter, it was demonstrated the independence of the
promotion index with respect to the rate of K+ ions supply, being
possible to induce the maximum improvement in the Au–YSZ cata-
lytic activity under given reaction conditions as long as a specific
potassium coverage is reached.
Other kind of electrochemical promotion experiments under
the same methanol partial oxidation conditions consisted of suc-
cessive potentiostatic transitions. Fig.
þ
Eq. (7). It corresponded to a potassium coverage ðhK Þ equal to
0.56, which was slightly higher than the optimum value found in
previous experiments (around 0.5). This fact could indicate some
possible blocking of Au active sites by an excess of K-derived sur-
face compounds [18,19,32] and could explain the mildly lower cat-
alytic activity obtained under these promoted conditions
compared to the maximum achieved in Fig. 6. This poisoning effect
was more clearly observed under the application of VWR = ꢀ1 V.
However, all the promoter species formed during the negative
polarizations were decomposed upon the successive imposition
of VWR = +2 V, again proving, in good agreement with the Au–YSZ
characterization results, the reversibility and reproducibility of
the EPOC phenomenon, and the stability of the YSZ-supported Au
nanoparticles. Moreover, it should be noted that in this experiment
the electrochemical catalyst showed a similar qualitative and
quantitative behavior as that observed in previous figures under
applied negative currents.
8 shows the dynamic
response in the H2 (a), CO2 (b) and HCOOCH3 (c) production rates
to step changes in the applied catalyst potential (VWR) at 280 °C.
4
3
2
1
3
(a)
(b)
(c)
rH
VWR
2
2
1
0
Finally, the influence of the reaction temperature in the produc-
tion rate of different compounds was studied at the same feed
composition (CH3OH/O2 = 5.9%/0.43% and Ar balance) under both
unpromoted and promoted conditions. Fig. 9 shows the variation
of the steady-state production rates and the accumulated promoter
supply vs. the applied catalyst potential at 280 and 300 °C. The cor-
-1
-2
2
0
4
3
2
1
0
rCO
VWR
2
responding rate enhancement ratios (qopt) and promotion indexes
þ
(PIK ;opt) are also indicated for the optimum applied potential in
each case. Starting from the reference positive polarization,
VWR = +2 V (un-promoted state), the decrease in the applied poten-
tial led, according to the obtained negative currents (inset of Fig. 9d
for the application of VWR = ꢀ0.5 V), to the electrochemical pump-
ing of K+ ions from the solid electrolyte to the Au–YSZ catalyst film.
As typically observed on K-bAl2O3-based electrocatalytic systems
under potentiostatic transitions [19,28], the resultant negative cur-
rents were more pronounced at the beginning of the polarization
and decreased in a few minutes down to negligible residual values
1
0
-1
-2
2
rHCOOCH
VWR
3
of the order of ꢀ1
lA. All the production rates (rH ; rCO and
rHCOOCH ) were increased by the potassium promotional effect at
12
6
2 2
1
3
both studied temperatures, showing an electrophilic EPOC behav-
0
þ
ior (VWR 6 ꢀ0.5 V,
D
r > 0,
q
> 1 and PIK > 0). Moreover, in both
cases the maximum catalytic activity was found at the same opti-
-1
-2
mum applied potential, VWR,opt = ꢀ0.5 V. This optimum potential
corresponded with the migration of 3.58 ꢁ 10ꢀ7 and 4.74 ꢁ 10ꢀ7
-
mol K+ and potassium coverages of 0.56 and 0.74, at 280 and
300 °C, respectively. The higher the reaction temperature, the
higher the solid electrolyte ionic conductivity and, therefore, the
higher the amount of electrochemically transferred K+ ions at fixed
potential. It can be observed an increase in the catalytic activity of
the Au–YSZ catalyst film with the temperature under unpromoted
0
0
100
200
300
400
Time / min
Fig. 8. H2 (a), CO2 (b) and HCOOCH3 (c) production rates vs. time under different
applied potentials (VWR) between 2 and ꢀ1 V. Methanol partial oxidation condi-
tions: CH3OH/O2 = 5.9%/0.43%, Ar balance, 280 °C.

J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
301
3.5
3
16
+
ρH ,opt = 7.9 PIK ,opt = 12.3
+
,HCOOCH
ρHCOOCH ,opt = 5.1 PIK
,opt = 7.3
3
,H
2
2
3
(a)
(c)
12
8
2.5
2
+
ρH ,opt = 3.6 PIK ,opt = 3.5
,H
2
2
+
ρHCOOCH ,opt = 2.7 PIK
,opt = 2.3
,HCOOCH
3
3
280 ºC
300 ºC
280 ºC
300 ºC
1.5
1
4
0.5
0
0
2
0.2
-0.2
-0.6
(b)
(d)
-0.5 V
+
ρCO ,opt = 1.9 PIK
,opt = 1.2
,CO
2
2
4
+
ρCO ,opt = 2.3 PIK
,opt = 2.3
,CO
2
2
3.5
3
1.5
1
280 ºC
300 ºC
280 ºC
300 ºC
2.5
2
-1
0
0.5
0
20
40
60
280 ºC
300 ºC
1.5
1
Time / min
-1.5
-0.5
0.5
1.5
2.5
-1.5
-0.5
0.5
1.5
2.5
VWR / V
VWR / V
Fig. 9. Steady-state H2 (a), CO2 (b) and HCOOCH3 (c) production rates and accumulated promoter supply (d) vs. applied potential (VWR) as well as the optimum production
þ
rates enhancement ratios (
q
i,opt) and promotion indexes ðPIK ;i;optÞ at different reaction temperatures. Methanol partial oxidation conditions: CH3OH/O2 = 5.9%/0.43%, Ar
balance. Inset of (d) shows the current vs. time curve during the potentiostatic imposition of ꢀ0.5 V.
conditions (positive potentials) as previously observed with
supported Au catalyst [9,10–14]. However, under electrochemical
promotion (VWR = ꢀ0.5 and ꢀ1 V) the overall methanol conversion
decreased at 300 °C, very likely due to the poisoning effect of the
excessive promoter-derived species. In fact, the value of potassium
– The stability of the Au/YSZ catalyst film was also verified by dif-
ferent characterization techniques: cyclic voltammetry, TEM
and XRD, which demonstrates the interest of the catalyst film
preparation technique and the use of the YSZ support to dis-
perse metal nanoparticles, avoiding the metal sintering.
– The magnitude of the electrochemical promotion effect with
alkali solid electrolyte did not depend on the applied current,
i.e. the rate of promoter supply, but on the promoter coverage.
The H2, CO2 and HCOOCH3 production rates were enhanced,
regardless the operation mode (galvanostatic or potentiostatic),
more than 9, 2 and 5 times, respectively, at 280 °C upon an opti-
mum potassium coverage of around 0.5.
þ
coverage, hK ¼ 0:74, obtained at this temperature under the appli-
cation of ꢀ0.5 V was already far above the optimum value found in
þ
previous experiments performed at 280 °C (hK
ꢂ 0.5). Further-
;opt
more, in good agreement with several studies [6,11,12,14], an
increase in the temperature favored the formation of CO2 and hin-
dered the HCOOCH3 selectivity due to the further oxidation of the
latter. The activity and selectivity of other metallic catalysts have
been also in situ modified by pumping alkali ions, as reported in
previous EPOC studies on the methanol partial oxidation [18,19]
but also on other catalytic reactions, such as the epoxidation of
alkenes [32] and the selective catalytic reduction of nitrogen oxi-
des [38–40]. In the present work, gold nanoparticles dispersed in
a YSZ matrix were catalytically activated by K+ ions via EPOC for
the simultaneous production of hydrogen and methyl formate
from methanol oxidation. Hence, two products of great interest
could be simultaneously produced (and electro-promoted), which
could be easily separated for their further use in different
applications.
– A permanent EPOC effect was found upon negative current
interruption, showing the stability of the formed promoter spe-
cies under the studied reaction conditions, which may be of
great interest for practical applications.
Acknowledgment
We gratefully acknowledge Spanish MICINN project CTQ 2010-
16179/PQ for the financial support of this work.
References
4. Conclusions
[1] D.R. Palo, R.A. Dagle, J.D. Holladay, Chem. Rev. 107 (2007) 3992–4021.
[2] S. Sundarrajan, S.I. Allakhverdiev, S. Ramakrishna, Int. J. Hydrogen Energy 37
(2012) 8765–8786.
The following conclusions can be drawn from this study:
[3] X. Deng, B.K. Min, A. Guloy, C.M. Friend, J. Am. Chem. Soc. 127 (2005) 9267–
9270.
[4] G. Avgouropoulos, T. Tabakova, J.J. Spivey, Environmental Catalysis over Gold-
Based Materials, The Royal Society of Chemistry, Cambridge, 2013.
[5] M. Haruta, Gold Bull. 37 (2004) 27–36.
– An electrochemical catalyst based on Au nanoparticles dis-
persed in a YSZ matrix was prepared by reactive cosputtering
of Zr–Y and Au targets. This technique allowed depositing a cat-
alyst film with a very low particle size (around 3 nm) and a
metal loading of just 40 l
g Au/cm2.
[6] A. Wittstock, V. Zielasek, J. Biener, C.M. Friend, M. Bäumer, Science 327 (2010)
319–322.
– The obtained Au–YSZ catalyst was shown to be active in the
partial oxidation of methanol with a very high selectivity
toward methyl formate production, being promoted by electro-
chemical pumping potassium ions under negative polarization
at two different temperatures.
[7] S. Gil, M. Marchena, C.M. Fernández, L. Sánchez-Silva, A. Romero, J.L. Valverde,
Appl. Catal. A 450 (2013) 189–203.
[8] H.A. Dole, J.M. Kim, L. Lizarraga, P. Vernoux, E.A. Baranova, ECS Trans. 45 (2012)
265–274.
[9] A.N. Pestryakov, V.V. Lunin, N. Bogdanchikova, O.N. Temkin, E. Smolentseva,
Fuel 110 (2013) 48–53.

302
J. González-Cobos et al. / Journal of Catalysis 317 (2014) 293–302
[10] B.P.C. Hereijgers, T.M. Eggenhuisen, K.P. de Jong, H. Talsma, A.M.J. van der
Eerden, A.M. Beale, B.M. Weckhuysen, J. Phys. Chem. C 115 (2011) 15545–
15554.
[24] F.M. Sapountzi, M.N. Tsampas, C.G. Vayenas, Catal. Today 146 (2009) 319–325.
[25] R. Catrin, T. Gries, B. Raillard, F. Mücklich, S. Migot, D. Horwat, J. Mater. Res. 27
(2012) 879–885.
[11] F.-W. Chang, S.-C. Lai, L.S. Roselin, J. Mol. Catal. A 282 (2008) 129–135.
[12] W. Jianxin, L. Laitao, Catal. Lett. 126 (2008) 325–332.
[13] T.-C. Ou, F.-W. Chang, L.S. Roselin, J. Mol. Catal. A 293 (2008) 8–16.
[14] K. Kähler, M.C. Holz, M. Rohe, A.C. van Veen, M. Muhler, J. Catal. 299 (2013)
162–170.
[15] S. Scirè, L.F. Liotta, Appl. Catal. B 125 (2012) 222–246.
[16] C.G. Vayenas, S. Bebelis, C. Pliangos, S. Brosda, D. Tsiplakides, Electrochemical
Activation of Catalysis: Promotion, Electrochemical Promotion and Metal-
Support Interactions, Kluwer Academic Publishers/Plenum Press, New York,
2001.
[17] P. Vernoux, L. Lizarraga, M.N. Tsampas, F.M. Sapountzi, A. de Lucas-Consuegra,
J.L. Valverde, S. Souentie, C.G. Vayenas, D. Tsiplakides, S. Balomenou, E.A.
Baranova, Chem. Rev. 113 (2013) 8192–8260.
[18] A. de Lucas-Consuegra, J. González-Cobos, Y. García-Rodríguez, A. Mosquera,
J.L. Endrino, J.L. Valverde, J. Catal. 293 (2012) 149–157.
[19] A. de Lucas-Consuegra, J. González-Cobos, V. Carcelén, C. Magén, J.L. Endrino,
J.L. Valverde, J. Catal. 307 (2013) 18–26.
[26] H.P. Klug, L.E. Alexander, X-Ray Diffraction Procedures for Polycrystalline and
Amorphous Materials, Wiley, New York, 1974.
[27] A. de Lucas-Consuegra, F. Dorado, J.L. Valverde, R. Karoum, P. Vernoux, J. Catal.
251 (2007) 474–484.
[28] A. de Lucas-Consuegra, A. Princivalle, A. Caravaca, F. Dorado, C. Guizard, J.L.
Valverde, P. Vernoux, Appl. Catal. B 94 (2010) 281–287.
[29] J.S. Lee, J.C. Kim, Y.G. Kim, Appl. Catal. 57 (1990) 1–30.
[30] N. Kotsionopoulos, S. Bebelis, Top. Catal. 44 (2007) 379–389.
[31] N. Kotsionopoulos, S. Bebelis, J. Appl. Electrochem. 40 (2010) 1883–1891.
[32] A. Palermo, A. Husain, M.S. Tikhov, R.M. Lambert, J. Catal. 207 (2002) 331–340.
[33] A.J. Urquhart, J.M. Keel, F.J. Williams, R.M. Lambert, J. Phys. Chem. B 107 (2003)
10591–10597.
[34] C.G. Vayenas, S. Bebelis, M. Despotopoulou, J. Catal. 128 (1991) 415–435.
[35] P. Broqvist, L.M. Molina, H. Grönbeck, B. Hammer, J. Catal. 227 (2004) 217–226.
[36] A. de Lucas-Consuegra, A. Caravaca, P.J. Martínez, J.L. Endrino, F. Dorado, J.L.
Valverde, J. Catal. 274 (2010) 251–258.
[37] I.V. Yentekakis, R.M. Lambert, M.S. Tikhov, M. Konsolakis, V. Kiousis, J. Catal.
176 (1998) 82–92.
[38] F. Dorado, A. de Lucas-Consuegra, C. Jiménez, J.L. Valverde, Appl. Catal. A 321
(2007) 86–92.
[39] I.V. Yentekakis, A. Palermo, N.C. Filkin, M.S. Tikhov, R.M. Lambert, J. Phys.
Chem. B 101 (1997) 3759–3768.
[40] P. Vernoux, F. Gaillard, C. Lopez, E. Siebert, J. Catal. 217 (2003) 203–208.
[20] C.G. Vayenas, S. Neophytides, J. Catal. 127 (1991) 645–664.
[21] O.A. Mar’ina, V.A. Sobyanin, Catal. Lett. 13 (1992) 61–69.
[22] O.A. Mar’ina, V.A. Sobyanin, V.D. Belyaev, V.N. Parmon, Catal. Today 13 (1992)
567–570.
[23] I.M. Petrushina, V.A. Bandur, F. Cappeln, N.J. Bjerrum, J. Electrochem. Soc. 147
(2000) 3010–3013.