Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild conditions

Article abstract of DOI:10.1016/j.cattod.2016.05.002

This work aims to further the understanding of gold-based catalytic oxidation of n-octanol in liquid phase. Modification of catalysts with metal oxides additives (Fe or Mg) was used as a tool for transforming and stabilizing gold species. Structural, electronic and catalytic properties of gold catalysts were systematically investigated by means of DRS, H₂, CO FTIR, S_BET, EDS and SEM, HRTEM, SR-XRD, XANES, XPS and liquid phase n-octanol oxidation. Addition of modifiers affects Au electronic properties, but not the structural ones. Characterization results allow excluding Au³⁺ ions as candidates for active sites in n-octanol oxidation. In Au/Mg/TiO₂, gold exhibited more reduced states while in Au/Fe/TiO₂ gold was more oxidized; Au/TiO₂ for intermediate oxidized states was found. The proper balance of oxidation states in the gold surface of Au/Mg/TiO₂ can be responsible for its higher activity compared with Au/Fe/TiO₂ and Au/TiO₂ towards n-octanol oxidation. Finally our approach shed light on the nature of active sites for n-octanol oxidation on gold and furthers the development of green base-free catalytic oxidation of alcohols.

Full text of DOI:10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation
under mild conditions
Y. Kotolevicha, E. Kolobovab, G. Mamontovc, E. Khramovd, J.E. Cabrera Ortegae,
H. Tiznadoa, M.H. Faríasa, N. Bogdanchikovaa, Ya. Zubavichusd, J.D. Mota-Moralesa,f
,
V. Cortés Corberáng, R. Zanellah, A. Pestryakovb,∗
a Centro de Nanociencias y Nanotecnología, Universidad Nacional autónoma de México, Ensenada, 22860, Mexico
b Tomsk Polytechnic University, Tomsk, 634050 Russia
c Tomsk State University, Tomsk, 634050 Russia
d National Research Centerꢀꢀ Kurchatov Institute”, Moscow, 123182 Russia
e Universidad Autónoma de Baja California, Ensenada, 22860, Mexico
f
CONACYT Research Fellow at Centro de Nanociencias y Nanotecnología, UNAM, Ensenada, 22860, Mexico
g Institute of Catalysis and Petroleum Chemistry (ICP), CSIC, 28049 Madrid, Spain
h Centro de Ciencias Aplicadas y Desarrollo Tecnológico, Universidad Nacional autónoma de México, Cd. Mx, 04510, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
This work aims to further the understanding of gold-based catalytic oxidation of n-octanol in liquid phase.
Modification of catalysts with metal oxides additives (Fe or Mg) was used as a tool for transforming and
stabilizing gold species. Structural, electronic and catalytic properties of gold catalysts were systemati-
cally investigated by means of DRS, H2, CO FTIR, SBET, EDS and SEM, HRTEM, SR-XRD, XANES, XPS and liquid
phase n-octanol oxidation. Addition of modifiers affects Au electronic properties, but not the structural
ones. Characterization results allow excluding Au3+ ions as candidates for active sites in n-octanol oxi-
dation. In Au/Mg/TiO2, gold exhibited more reduced states while in Au/Fe/TiO2 gold was more oxidized;
Au/TiO2 for intermediate oxidized states was found. The proper balance of oxidation states in the gold
surface of Au/Mg/TiO2 can be responsible for its higher activity compared with Au/Fe/TiO2 and Au/TiO2
towards n-octanol oxidation. Finally our approach shed light on the nature of active sites for n-octanol
oxidation on gold and furthers the development of green base-free catalytic oxidation of alcohols.
© 2016 Elsevier B.V. All rights reserved.
Received 2 January 2016
Received in revised form 4 May 2016
Accepted 4 May 2016
Available online xxx
Keywords:
n-Octanol oxidation
Gold catalysts
Modification with Mg and Fe oxides
Gold electronic state
1. Introduction
Biomass offers not only a sustainable alternative to petrochem-
ical feedstock, but the industrial application of feedstock from
To address the upcoming petroleum shortage, two main
approaches have been developed to ensure fuel supply for energy
purposes. For instance, hydrate gas and shale gas exploitation have
become increasingly popular in this matter; while on the other
hand, by shifting to renewable sources, it is expected that the use
of petroleum undergoes a drastic reduction. Meanwhile, process-
ing and production of number of commodities in our daily life that
are based on petroleum-derived feedstock might also face paucity,
thus the search for alternative feedstock useful for fine-chemical
industry is a continuing and challenging task.
renewable resources could also be convenient in valorizing waste
from a variety of economic sectors such as food, agriculture and
paper industries. Biomass accounts for more than half of the renew-
able energy potential [1]; but also huge amounts of new raw
materials for chemicals production, including a vast source of alco-
hols due to fermentation, among others process. In this regard,
selective oxidation of fatty alcohols to the corresponding alde-
hydes, acids and ester, holds potential to open the path for the
synthesis of products with high added value for fine-chemicals.
From the sustainability viewpoint, the use of biomass fits well
with the principles of green chemistry, which include atom econ-
omy, chemicals with low toxicity, avoidance of volatile organic
solvents and derivatization, use of renewable resources, production
of biodegradable materials and use of catalysis.
∗
Corresponding author at: Tomsk Polytechnic University, Lenin ave. 30, Tomsk,
634050, Russia.
Focusing on the latter, catalysts comprising supported gold
nanoparticles (Au NPs) are among the most investigated systems
E-mail addresses: pestryakov2005@yandex.ru, pestryakov@tpu.ru
(A. Pestryakov).
http://dx.doi.org/10.1016/j.cattod.2016.05.002
0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
2
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
for the catalytic aerobic oxidation of alcohols in liquid phase, which
circumvent the use of bases [2]. Catalytic oxidation of fatty alcohols
is a complex reaction that involves several intermediates. In fact,
current data available in the literature fail identifying a clear gen-
eral trend and forecasts that permit to direct the catalyst selectivity
or that allows an overall optimization of the reaction conditions.
Although Au NP-based catalyst have proved to be efficient in cat-
alyzing oxidation of alcohols, a number of parameters such as gold
content, Au NPs size, electronic state of gold, influence of additives,
redox pretreatment, etc., have not been investigated in detail.
It is worth mentioning that oxidation of n-octanol is being used
as a convenient model for comparative studies between catalysts
activity towards oxidation of alcohols, specifically for primary alco-
hols of long chain (so called fatty alcohols). As oxidation of n-octanol
is more difficult than most alcohols of industrial importance, it is
expected that catalysts active for n-octanol oxidation will be also
effective in similar processes involving biomass transformation.
Normally, base-mediated oxidation of alcohols in liquid phase
requires neutralizing the carboxylates formed during oxidation
with strong acids, which generates large amounts of inorganic salts
as waste. This justifies the need for base-free catalytic oxidation
of alcohols in order to obtain aldehydes, acid or esters in a more
sustainable way. Villa et al. first reported a base-free oxidation of
n-octanol by Au NPs with the aid of O2, which were supported
on nanometer-sized NiO [3]; while Ishida et al. [4] reported an
extensive and specific study concerning a base-free oxidation of
n-octanol (in water) by Au NPs under O2 atmosphere at moderate
pressure. By screening the Au NPs supported on a variety of metal
oxides (Al2O3, TiO2, MnO2, Fe2O3, Co3O4, NiO, ZnO, ZrO2 and CeO2),
it was evidenced that their activity strongly depends upon both
the supportı´s nature and the solvent used, while selectivity varied
broadly. All these results were obtained under moderate pressure
of oxygen (5 bar), but recent studies have shown that, by using hep-
tane as solvent, n-octanol can be oxidized even with oxygen flowing
at normal pressure and in the absence of bases [5].
Catalytic oxidation of n-octanol under mild conditions, i.e. base-
free, at low temperature and under atmospheric pressure, has been
the focus of recent investigations conducted by our group [6,7]. We
have demonstrated that a catalyst consisting in Au NPs supported
on alumina CeO2-modified (Au/CeO2/Al2O3) is able to promote the
oxidation of n-octanol rendering a mixture of equal parts of the
corresponding aldehyde and ester, n-octanal and octyl octanoate
respectively [5]. On the other hand, selectivity towards aldehyde,
acid or ester formation can be tuned in by addition of metal oxides
as modifiers, even at low conversions [6]. In the present study TiO2
was chosen as support since Au/TiO2 catalysts are among the most
active gold catalysts in oxidation reactions [8–16]. Modification of
Au/TiO2 catalyst was carried out through introduction of electron-
donor (Mg) and electron-acceptor (Fe) oxides. Finally, this work
aims to shed light on the nature of gold active sites origin of the
catalytic activity of nanometric gold, which is an unexplored topic
for base-free catalytic oxidation of n-octanol.
HAuCl4 × 3H2O (Aldrich) was used as gold precursor. Au/TiO2 and
Au/M/TiO2 catalysts (nominal loading 4 wt.% Au) were prepared by
deposition-precipitation with urea in the absence of light, following
the previously reported procedure [17–19].
2.2. Samples characterization
Catalyst samples, either as-prepared or pretreated in hydrogen
at 300 ◦C for 1 h, were studied by diffuse reflectance UV–vis spec-
troscopy (DRS) with a CARY 300 SCAN (Varian) spectrophotometer.
Optical spectra of Au/TiO2 or Au/M/TiO2 samples presented in this
work were obtained by subtracting the spectra of pure supports
from those of catalyst samples.
H2-TPR measurements of as-prepared samples were performed
in a fixed-bed quartz reactor with an AutoChem 2950 analyzer,
¨
Micromeritics. Temperature-programmed experiments were per-
¨
formed by heating at a rate of 10 ◦C min−1 from 25 up to 900 ◦C
under the reducing feed (10 vol.% of H2/Ar, 20 cm3 min−1). Hydro-
gen consumption was measured by the thermal conductivity
detector.
Fourier transformed infrared spectra (FTIR) of CO adsorbed on
the catalysts were recorded by using a Bruker Tensor 27 FTIR spec-
trometer in transmittance mode with 4 cm−1 resolution. In situ
experiments were carried out in a quartz cell with NaCl windows
capable of working at temperatures from −100 to 300 ◦C and pres-
sures from 10−2 to 760 Torr. The sample powder was pressed into
disks of 13 mm diameter and weight ∼20 mg. The sample was pre-
treated in H2 or O2 (100 Torr) at 300 ◦C for 1 h and then cool down
for room temperature. After that, H2 or O2 was evacuated and CO
adsorption (Matheson Research grade, P0 = 30 Torr) were carried
out. CO spectra presented in the work were obtained by subtracting
the CO gas phase spectrum.
Due to conditions of equipment exploitation, prior any other
characterization the samples were pretreated in hydrogen at 300 ◦C
for 1 h.
Textural properties of samples were determined from nitro-
gen adsorption-desorption isotherms (–196 ◦C) recorded with a
Micromeritics TriStar 3000 apparatus. Prior to experiments, sam-
ples were degassed at 300 ◦C in vacuum for 5 h. The N2 adsorbed
volume was normalized to a standard temperature and pressure.
The specific areas of the samples were calculated by applying the
BET method to the nitrogen adsorption data within the P/P0 range
0.05–0.25.
A JEOL-5300 scanning electronic microscope (SEM) was utilized
for a general sample morphology observation. Gold contents were
measured by energy dispersive spectroscopy (EDS) in the same
system equipped with a Kevex Superdry detector.
High resolution transmission electronic microscopy (HRTEM)
studies were carried out using a JEM 2100F microscope operat-
ing with a 200 kV accelerating voltage. The samples were ground
into a fine powder and dispersed ultrasonically in hexane at room
temperature. Then, a drop of the suspension was put on a lacey
carbon-coated Cu grid. At least ten representative images were
taken for each sample. Particle size distribution was obtained by
counting ca. 100 particles for each sample.
2. Experimental
2.1. Catalysts preparation
X-ray powder diffraction was conducted by the step-scanning
procedure (step size 0.02◦; 0.5 s) with
a Philips XPert PRO
Titania Degussa P25 (45 m2 g−1, nonporous, 70% anatase and
30% rutile, purity >99.5%) was used as starting support. Before
use, TiO2 was dried in air at 100 ◦C for at least 24 h. Modifica-
tion of titania with molar ratio Ti/M = 40 (M = Mg, Fe) was made
by impregnation (2.5 cm3/g) of initial TiO2 with aqueous solutions
of modifier precursors Fe(NO3)3 × 9H2O or Mg(NO3)2 from Aldrich.
Then, impregnation products were dried at room temperature for
48 h and at 110 ◦C for 4 h, and calcined at 550 ◦C for 4 h. Commercial
diffractometer, using Ni-filtered CuK␣ (␭ = 0.15406 nm) radiation.
Assignment of crystalline phases was based on the ICDD- 2013
powder diffraction database. Synchrotron radiation X-ray diffrac-
tion (SR-XRD) experiments were carried out at the Structural
Materials Science beamline of the Kurchatov Synchrotron Radia-
tion Source as described in [20]. Diffraction patterns of powdered
materials were taken in transmission mode at ␭ = 0.68886 Å, using
an Imaging Plate 2D detector (exposure time 30 min). EXAFS exper-
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
3
b
a
20
15
10
5
90
Au/Mg/TiO2
60
Au/Mg/TiO2
30
0
90
0
20
Au/Fe/TiO2
15
10
5
60
Au/Fe/TiO2
30
0
0
20
90
as-prepared
H2-treated
n
15
10
5
aldehyde
60
30
0
acid
ester
Au/TiO2
Au/TiO2
0
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Time, (h)
Time, (h)
Fig. 1. Conversion of n-octanol on Au catalysts: the as-prepared and after pretreatment in H2 (a). Selectivity of catalysts pretreated in H2 (b).
H
H
B
C
H
C
OH
CH
O
C
H
H
R
OH
Me
Me
C
H2O
1/2 O2
1/2 O2
R
R
R
R
OH
R
O
OH
OH
H
O
H
OH
CH
A
C
C
R
Me
OH
R
O
O
CH2
1/2 O2
CH2
R
R
Scheme 1.
iments were carried out as described in [20]. The XANES spectra
were taken in transmission mode. Primary processing of EXAFS
spectra was done using IFEFFIT software package [21,22]. Extended
analysis was carried in-situ at room temperature. The Fourier
transformants were analyzed for k = 2.0–11.6 Å−1 with the weight
coefficient k3 using the phases and amplitudes of photoelectrons
scattering calculated in terms of FEFF8 software [23].
mated by calculating the integral of each peak after subtracting
a Shirley type background and fitting the experimental peak to
a combination of Lorentzian/Gaussian lines with a 30/70 propor-
tion, considering the spin-orbit 4f7/2 and 4f5/2 doublet with a 4:3
intensity ratio and the same width on all lines.
The samples were investigated by X-ray photoelectron spec-
2.3. Catalytic testing
troscopy (XPS) with
a SPECS GmbH custom made system
using a PHOIBOS 150 WAL hemispherical analyzer and a non-
monochromated X-Ray source. All the data were acquired using
Al K␣ X-rays (1486.6 eV, 200 W). A pass-energy of 50 eV, a step size
of 0.1 eV/step and a high-intensity lens mode were selected. The
diameter of the analyzed area was 3 mm. Charging shifts were refer-
enced against adventitious carbon (C 1 s at 284.5 eV). The pressure
in the analysis chamber was maintained lower than 1 × 10−8 mbar.
Catalysts were mounted on a sample holder and kept overnight
in high vacuum in the preparation chamber before they were
transferred to the analysis chamber of the spectrometer. Energy
regions were selected after a general survey and scanned with sev-
eral sweeps until a good signal-to-noise ratio was observed. The
accuracy of the binding energy (BE) values was 0.1 eV. Spectra
are presented without smoothing or background subtraction, with
intensity in counts-per-second (CPS). Peak intensities were esti-
Catalytic measurements in n-octanol oxidation were performed
with catalysts as-prepared or after being treated in pure hydrogen
flow at 300 ◦C for 1 h. Typically, supported gold catalyst (sub-
strate/metal (R) = 100 mol/mol) was added to 20 ml of n-octanol
substrate (0.1 M) in n-heptane as solvent, in four-necked round
bottom flask equipped with reflux condenser, oxygen feed, ther-
mometer and a septum cap. The reaction mixture was stirred in a
semi batch reactor operation under atmospheric pressure at 80 ◦C.
Oxygen (30 ml/min) was bubbled through the suspension and reac-
tion was followed for 6 h. Small aliquots of the reacting mixture
were obtained during and at the end of the test, by using nylon
syringe filters (pore 0.45 ␮m), for monitoring the reaction progress.
Reactants and products were analyzed in a Varian 450 gas chro-
matograph, using a capillary DB wax column (15 m × 0.548 mm)
and He as the carrier gas.
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
4
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
Fig. 2. Microphotographs and corresponding Au particle size distribution of Au/TiO2, Au/Fe/TiO2 and Au/Mg/TiO2 (dm is arithmetic mean diameter, ds is diameter, calculated
as average by surface).
3. Results and discussion
(1–3 h), becoming slightly active afterwards probably due to partial
reduction of gold ions during the course of the reaction.
Au NPs supported on pure and modified TiO2 supports were
tested in n-octanol oxidation (see Fig. 1) under mild conditions,
e.g. atmospheric conditions at 80 ◦C. On as-prepared catalysts
Au/Mg/TiO2, Au/Fe/TiO2 and Au/TiO2 reaction started after 1, 2 and
3 h, respectively, reaching ∼5% after 6 h in all cases. It can be inferred
that Au3+ is not active because all as-prepared catalysts containing
predominantly Au3+ were not active during the first period of time
In contrast, after H2 treatment, modified catalysts (especially
with Mg) showed higher n-octanol conversion than unmodified
catalyst supported on pure titania. It is also evident that the type
of modifier used influenced octanol conversion. For all catalysts,
octanoic acid formation was practically negligible. With increasing
run time, minimal changes of selectivity were observed for Au/TiO2,
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
5
Table 1
Specific surface area and gold content of studied samples.
Samples
SBET, m2/g
Support
Au content. wt.%
Catalyst
Au/TiO2
Au/Fe/TiO2
Au/Mg/TiO2
55.5
45.5
42.6
45.5
44.2
43.7
4.5
4.2
4
Au/Mg/TiO2
while Au/Fe/TiO2 selectivity towards octanyl octanoate increased
at the expense of octanal formation
Ishida et al. [24] proposed that the first step in n-octanol oxi-
dation is its oxidation to octanal on the supported Au NPs, which
may then follow two reaction routes: a hydration to a geminal diol
(Scheme 1A) that can be further oxidized to octanoic acid, or an
acetalization to form the hemiacetal, that can be oxidized to octyl
octanate. The ester could be also formed by esterification of the acid
with the alcohol (Scheme 1, B).
Au/Fe/TiO2
Au/TiO2
For all catalysts, selectivity towards octanoic acid formation was
negligible. On the other hand, selectivity towards octanal decreased
at the expense of octanyl octanoate formation. Minimal changes
regarding selectivity were observed for Au/TiO2, while Au/Fe/TiO2
gives rise to a varied amount of product during the course of the
reaction.
10
15
20
25
30
35
2theta (deg)
Fig. 3. The SR-XRD patterns for Au catalysts. Diffraction lines corresponding to TiO2
anatase phase are marked with an “a” while those corresponding to rutile have an
“r”; Au stands for gold.
It is important to mention that octanal formation occurs only on
the gold surface. Formation of octanoic acid and octanyl octanoate
occurs through intermediates formed by acid-base catalyzed reac-
tions. Two stages are then drawn: the first one consisting in octanal
formation due to the catalytic oxidation of alcohol on gold, and the
second stage in which formation of acid and ester is due to acid-base
reactions occurring mainly on the support surface. In the second
stage of the reaction, water traces, Broensted and Lewis acid char-
acter of supports become more relevant. Consequently, the more
acid and ester are produced, the more contribution of n-octanol
transformation through acid-base pathway (without gold partici-
pation) is observed. No difference in ester formation was observed
during the first 2 h for Au/Fe/TiO2 and Au/Mg/TiO2, which could be
interpreted as due to similar acid/base character of both catalysts.
After 2 h of reaction, formation of ester reached a maximum on
Au/Fe/TiO2 indicating higher acid/base contribution; in contrast to
Au/TiO2, whose acid-base contribution was minimal. Hence, when
and in what extent acid/base stages are involved in the catalytic
process determine the selectivity towards acid or ester.
To get deeper insights into the catalytic oxidation of n-octanol by
gold, an extensive characterization of gold was carried out in this
work. As aldehyde, the only primary product, is produced exclu-
sively on the gold surface, characterization of gold properties is
the key step in understanding the catalytic activity during gold-
catalyzed oxidation of n-octanol. The influence of modifiers on
the structural and electronic properties of the catalysts through
physical-chemical studies, were also conducted.
Table 1 summarizes the catalysts’ specific surface area and Au
content. SBET of TiO2 decreased as a result of modification with both
Fe and Mg oxides. Further Au deposition did not alter the surface
area, excepting for unmodified TiO2, resting in similar final val-
ues for all catalysts (ca. 45 m2/g). Au content was ca. 4 wt.% for all
catalysts.
HRTEM results are shown on Fig. 2. Au particle size is found
within 1–10 nm interval in all catalyst. Au/TiO2 histogram appeared
monodispersed with an average diameter of 3.7 nm, while those of
Au/Mg/TiO2 and Au/Fe/TiO2 were similar, comprising larger aver-
age diameter (∼5.2 nm). The main difference between Au/Mg/TiO2
and Au/Fe/TiO2 consists in higher contribution of 1 nm particles in
the former. Hence, the modification with Mg oxide led to appear-
ance of 1 nm Au particles along with larger ones, suggesting also
the existence of subnanometer Au species. Several cases where
Table 2
Hydrogen consumption during H2-TPR of as-prepared samples.
Sample
Au reduced, ␮mol/g (cat)
Au loading, ␮mol/g (cat)
Au/TiO2
Au/Mg/TiO2
Au/Fe/TiO2
194
161
200
228 46
184 37
213 43
most of Au is in the form of subnanometer species, and HRTEM
histograms (d > 1 nm) represent only a small part of the real dis-
tribution have been reported in the literature [25–27]. To answer
the question of whether modified catalysts in this work include
subnanometer species or not, several physicochemical methods of
characterization were applied.
SR-XRD patterns (Fig. 3) of the studied catalysts included not
only narrow anatase and rutile diffraction lines, but also a series of
broad diffraction lines corresponding to the Au metal phase with
size >2 nm, which is in agreement with HRTEM histograms. The
absence of well-defined lines corresponding to the modifiers evi-
dences their uniform distribution on the TiO2 surface.
By DRS, registration of subnanometer Au clusters (< 1 nm) can
be directly assessed (Fig. 4); they are characterized by absorption
peaks at <470 nm [24,28–30]. In our case, all samples showed strong
absorption below 420 nm due to the band gap excitation of sup-
ports, which impedes of Au clusters. However, comparison between
the spectra of as-prepared catalyst and supports, demonstrates
wide peaks with low intensive in the cluster range (320–470 nm).
This is indicative that part of gold precursor in as-prepared catalysts
is already reduced and exists in the form of Au clusters.
Absorption spectra of the catalysts after H2 pretreatment
(Fig. 4d) exhibit a peak with a maximum at 500–700 nm. This peak
is located at the wavelength area corresponding to Au surface plas-
mon, thus confirming AuNPs formation [31]. Au NPs plasmon peaks
of all catalysts are very similar (Fig. 1). This is also in agreement
with their corresponding HREM histograms. For catalysts after H2
pretreatment the information about Au subnanometer clusters was
not possible to assess.
Fig. 5 and Table 2 show the results of H2-TPR measurements
for as-prepared catalysts. For all catalysts H2-TPR profiles show an
acute peak with maximum at 145 ◦C that is typical for the reduc-
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
6
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
a
b
1.8
1.8
Au/Mg/TiO2
Au/TiO2
support
support
1.5
1.5
1.2
0.9
0.6
0.3
0.0
as-prepared
as prepared
1.2
0.9
0.6
0.3
0.0
200
400
600
800
1000
200
400
600
800
1000
Wavelength, (nm)
Wavelength, (nm)
c
d
1.8
1.8
Au/Fe/TiO2
H2-treated
support
Au/TiO2
1.5
1.2
0.9
0.6
0.3
0.0
1.5
1.2
0.9
0.6
0.3
0.0
as-prepared
576
543
Au/Mg/TiO2
Au/Fe/TiO2
200
400
600
800
1000
300
450
600
750
900
Wavelength, (nm)
Wavelength, (nm)
Fig. 4. Diffuse reflectance spectra of supports and as-prepared Au/TiO2 (a), Au/Mg/TiO2 (b) and Au/Fe/TiO2. Catalysts after reduction in H2 (d).
consumption at 100–250 ◦C corresponds to reduction from Au3+ to
Au/TiO
Au/Mg/TiO
Au/Fe/TiO
2
Au0 of the most part (85–94%) of the gold precursor (Table 2). Dif-
ferences in the H2 consumption value, different hypothesis can be
drawn. These differences can be associated with the precision limit
of H2-TPR method (20%); alternatively with the presence of minor
amounts of Au+ or Au0 in as-prepared catalysts, or partial reduction
of Au3+ into Au+. Results of DRS described above, point to the exis-
tence of – at least partially – reduced gold clusters in as-prepared
catalysts.
0.08
0.06
0.04
0.02
0.00
2
2
EXAFS method was applied in order to rule out the contribu-
tion of subnanometer gold species in the studied catalysts. Average
coordination number of metallic particles measured by EXAFS
correlates well with average particle size. Application of EXAFS
method can give the most direct answer on whether modified cat-
alysts of this work include substantial amount of subnanometer
species. From EXAFS data are presented in Fig. 6 it can be seen that
average coordination number for Au is 9.7–10.6 for all catalysts.
These values are close to the coordination number of Au atoms in
gold metallic foil. [12], this implies that only Au-Au distances are
observed and Au-O distances were not detected. This evidences that
the majority of Au volume in the studied catalysts is represented
by relatively large metallic particles Au coordination numbers in
Au samples decrease down to 2–4 when subnanometer species are
dominant [32]. Consequently, on the basis of H2-TPR and EXAFS
0
100
200
300
400
500
600
700
Temperature (oC)
Fig. 5. H2-TPR profiles of as-prepared Au catalysts.
tion of Au precursor. The shift of temperature at which reduction
of gold cations occurred was not significant among the studied
catalysts. Insignificant peaks corresponding to support reduction
were observed at higher temperatures (detailed profiles for sup-
ports are not shown). The main result of this study is that the H2
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
7
Au foil (CN=12)
8
6
4
2
0
Au/TiO2 (CN=10.6)
2103
Au/Mg/TiO2 (CN=9.8)
Au/Fe/TiO2 (CN=9.7)
0
1
2
3
4
5
6
R (A)
Fig. 6. Catalyst EXAFS spectra and coordination numbers (CN).
Au/Mg/TiO2
Au/Fe/TiO2
Au/TiO2
it can be concluded that Au particle size distributions, measured
by HRTEM (Fig. 2), are correct and that contribution of gold sub-
nanometer species is not significant in the studied catalysts. This
conclusion permits to use HRTEM histograms for the calculation
of average surface diameter for gold nanoparticle. These diameters
are 4.9, 6.1 and 6.4 for Au/TiO2, Au/Fe/TiO2 and Au/Mg/TiO2, respec-
tively (Fig. 2). Based on these values, time dependence of turnover
numbers were calculated (Fig. 7).
The lowest values of time dependence of TON were observed
for Au/TiO2, while the highest ones corresponded to Au/Mg/TiO2;
Au/Fe/TiO2 presented intermediate values (Fig. 7). Initial TON (after
the first 5 min) for Au/Fe/TiO2 (0.26) is slightly higher than that for
Au/TiO2 (0.21), and approximately two-fold higher for Au/Mg/TiO2
(0.41).
Method of CO adsorption at room temperature can be used to
identify gold species (e.g. Aun␦−, Au0, Aun␦+, Au+ [33]) was applied
to find a correlation between these adsorbed CO species with gold
catalytic activity in n-octanol oxidation. Fig. 8 the shows FTIR spec-
tra of CO adsorbed at room temperature on studied Au catalysts.
For all catalysts a peak at 2103 cm−1 was observed. This peak
ascribed to CO-Au0 on small particles, which corresponds to bands
2000
2050
2100
2150
2200
2250
Wavenumber, cm-1
Fig. 8. FTIR spectra of CO adsorbed at room temperature on studied Au catalysts.
within 2080–2120 cm−1 interval in the literature [25,26,33–37].
Peak intensity for Au/TiO2 and Au/Fe/TiO2 were similar, while that
for Au/Mg/TiO2 was ∼10 times higher. This order of peak intensity
of adsorbed CO coincided with that of their corresponding initial
TON (after 5 min reaction): Au/Mg/TiO2 > Au/Fe/TiO2–Au/TiO2.
TON for Au/TiO2 and Au/Fe/TiO2 were similarly low (0.21 and
0.26 mol/mAu2), and for Au/Mg/TiO2 TON it is two-fold higher
(0.41 mol/mAu2). This difference decreases with time (Fig. 7) and
after 6 h is only ∼36%, while after 5 min is ∼95%. Consequently, fol-
lowing that trend, it is plausible that this difference could be larger
5 min
1.5
1.2
0.9
0.6
0.3
0.0
Au/Mg/TiO2 (0.41)
Au/TiO2 (0.21)
Au/Fe/TiO2 (0.26)
0
1
2
3
4
5
6
Time, (h)
Fig. 7. Turnover number (TON) with time for n-octanol oxidation on gold catalysts. In parenthesis TON after 5 min of reaction are indicated.
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
8
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
Table 3
Binding energy of different gold electronic states and their relative atomic concentrations.
Catalysts
BE of different gold electronic states (eV), and their relative atomic concretions, % (indicated in parenthesis)
Au0
Au+
Au2Ox (1 < x < 3)
Au3+
␦−
Aun
≤83.6
∼84.0
∼85.0
∼85.3
∼87.0
Au/TiO2
Au/Mg/TiO2
Au/Fe/TiO2
83.3 (68%)
83.2 (73%)
–
84.1 (20%)
84.5 (20%)
84.1 (28%)
85.4 (12%)
–
85.0 (32%)
–
–
–
86.1 (7%)
86.0 (29%)
86.9 (11%)
Bold numbers refer to literature data [37–43].
expected because it is well known that Fe oxide possess strong
electron acceptor character and therefore can cause shifting in the
electron density of gold species.
Results of HRTEM, DRS, EXAFS and H2-TPR showed that struc-
tural properties of Au NPs in the three catalysts were very similar.
Only XPS method, characterizing surface states, revealed some
differences between the gold surface states. Hence, variation in cat-
alytic properties could be attributed to the difference in surface
states of gold particles.
As shown by catalytic tests with as-prepared catalysts (Fig. 1),
Au3+ is not active in n-octanol oxidation. Modification with Fe
oxide led to important contribution (11%) of inactive Au3+ state on
gold surface. According to XPS data, Au/TiO2 gold exhibited more
reduced states while gold was more oxidized in Au/Fe/TiO2; for
Au/Mg/TiO2 intermediate oxidized states were found. It is then pos-
sible that the proper balance of intermediate oxidation states in the
gold surface, displayed by Au/Mg/TiO2, is responsible for its higher
activity towards n-octanol oxidation. The other suggestion is that
Au/TiO2 is less active than Au/Mg/TiO2 probably the presence of
smaller Au NPs in Au/Mg/Ti plays a role in its higher activity. Also
we have to note that during the catalytic reaction small Au clus-
ters can obtain certain effective charge (␦+) as a result of oxygen
adsorption.
In summary, in the present work the characterization of gold
catalysts with a number of complex physicochemical methods
allowed studying Au NPs structural and electronic properties in
the frame of base-free catalytic oxidation of alcohols. This helped
to shed light on the nature of active sites for n-octanol oxidation in
gold.
Au/TiO2
83.3
84.1
85.4
2450
2100
1750
1400
2100
Au/Mg/TiO2
86.1
84.5
83.2
1800
1500
1200
3150
86.0
85.0
Au/Fe/TiO2
2800
2450
2100
86.9
84.1
80
82
84
86
88
90
92
94
Binding energy (eV)
Fig. 9. Au 4f photoelectron spectra for studied catalysts.
during very first stage of the reaction, i.e. before 5 min. FTIR spectra
of CO adsorbed at room temperature were measured after in-situ
H2 pretreatment. Strictly speaking, adequate comparison between
catalytic data should be done immediately after H2 pretreatment
(i.e. before 5 min of reaction). As the first measurement was made
after 5 min, TON difference can be already decreased as compared
to the initial one. Nevertheless, this qualitative correlation permits
suggesting that active sites for n-octanol oxidation are reduced Au0
states on small particles, capable of adsorbing CO at room temper-
ature.
4. Conclusions
Modification of gold catalysts with Fe and Mg oxides showed
significant influence on the catalytic properties (n-octanol conver-
sion and products selectivity) in n-octanol oxidation on Au/TiO2
under mild conditions. Additive with electron-donor properties
(e.g. MgO) provided the maximum catalytic activity (TON), while
electron-acceptor one (Fe oxide) resulted less effective.
The modifiers also affect Au electronic properties, but not the
structural ones. Characterization results allow excluding Au3+ ions
as candidates for active sites in n-octanol oxidation. In Au/Mg/TiO2,
gold exhibited more reduced states while in Au/Fe/TiO2 gold was
more oxidized; for Au/TiO2 intermediate oxidized states were
found. It is suggested that the proper balance of intermediate
oxidation states in the gold surface, displayed by Au/Mg/TiO2, is
responsible for its higher activity towards n-octanol oxidation.
Finally our approach furthers the development of green base-
free catalytic oxidation of alcohols.
XPS method was applied to investigate the electronic states of
gold surface. Fig. 9 and Table 3 show the catalysts Au 4f photo-
electron spectra and their analyses. Au 4f peak doublets showing
large width were subjected to a deconvolution process. The decon-
volution showed that there are three gold surface states in Au/TiO2
and Au/Mg/TiO2, while four different states were registered in
Au/Mg/TiO2. It can be noticed that BE shifts of 0.4 and 0.7 eV for gold
surface states Au0 and Au+, respectively, occurred in Au/Mg/TiO2
as compared to Au/TiO2. A much more significant difference was
found for Au/Fe/TiO2. It consists in that while reduced states
account for the majority of surface gold in Au/TiO2 and Au/Mg/TiO2,
oxidized surface states are dominant Au/Fe/TiO2. For Au/TiO2 and
Acknowledgements
␦−
Au/Mg/TiO2 gold reduced states (Aun and Au0) stand for about
This work is funded by CONACYT-Mexico under projects 79062,
83275 and 130407 and PAPIIT-UNAM projects 105416, IN114209,
IT200311 and IT200114 (Mexico); CSIC, project 201180E104
and MICINN, project ENE2009-14522-C05, (Spain); Government
90%, while for Au/Fe/TiO2 they account for only 28%. In Au/Fe/TiO2
∼40% of surface gold is in highly charged states (with charge higher
than +1). High contribution of these gold states in Au/Fe/TiO2 is
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002

G Model
CATTOD-10189; No. of Pages9
ARTICLE IN PRESS
Y. Kotolevich et al. / Catalysis Today xxx (2016) xxx–xxx
9
Program ”Science” of Tomsk Polytechnic University, grant No.
4.1187.2014/K. We thank Drs. O. Martynyuk, S. Martinez-Gonzalez,
E. Flores, A. Olivas Sarabia, and B. Acosta Ruelas, and G. Torres
Otan˜ez, Z. I. Bedolla Valdez, F. Ruiz Medina, I. Gradilla, J. Mendoza, E.
M. Aparicio Ceja, J. Peralta, David A. Domínguez, M. Sainz, R. Valdez
Castro, M. Martínez, F. Ramírez Hernández, M. López Cisneros, R.
Vélez for valuable technical assistance.
[20] A.A. Chernyshov, A.A. Veligzhanin, Y.V. Zubavichus, Nucl. Instr. Meth. Phys.
Res. A 603 (2009) 95.
[21] R.B. Athena, A. Hephaestus, J. Synchrotron Rad. 12 (2005) 537.
[22] M. Newville, J. Synchrotron Rad. 8 (2001) 322.
[23] A.L. Ankudinov, B. Ravel, J.J. Rehr, S.D. Conradson, Phys. Rev. B 58 (1998) 7565.
[24] A.A. Herzing, C.J. Kiely, A.F. Carley, P. Landon, G.J. Hutchings, Science 321
(2008) 1331.
[25] N. Bogdanchikova, A. Pestryakov, M.H. Farías, J.A. Díaz, M. Avalos, J. Navarrete,
Solid State Sci. 10 (2008) 908.
[26] A. Pestryakov, N. Bogdanchikova, A. Simakov, I. Tuzovskaya, F. Jentoft, M.
Farias, A. Díaz, Surf. Sci. 601 (2007) 3792.
[27] S. Chrétien, S.K. Buratto, H. Metiu, Curr. Opin. Solid State Mater. Sci. 11 (2007)
62.
References
[1] D. Gielen, F. Boshell, D. Saygin, Nat. Mater. 15 (2016) 117.
[2] V.C. Corberán, M.E. González-Pérez, S. Martínez-González, A. Gómez-Avilés,
Appl. Catal. A Gen. 474 (2013) 211.
[3] A. Villa, C.E. Chan-Thaw, G.M. Veith, K.L. More, D. Ferri, L. Prati, ChemCatChem
3 (2011) 1612.
[28] S. Peters, S. Peredkov, M. Neeb, W. Eberhardt, M. Al-Hada, Surf. Sci. 608 (2013)
129.
[29] T. Takei, I. Okuda, K.K. Bando, T. Akita, M. Haruta, Chem. Phys. Lett. 493 (2010)
207.
[30] J. Zhang, X. Suo, J. Zhang, B. Han, P. Li, Y. Xue, H. Shi, Mater. Lett. 162 (2016)
235.
[31] R.B. Greecor, F.W. Lytle, J. Catal. 63 (1980) 476.
[32] O. Martynyuk, Y. Kotolevich, R. Vélez, J. Efren Cabrera Ortega, H. Tiznado, T.
Zepeda Partida, J.D. Mota Morales, A. Pestryakov, N. Bogdanchikova,
Molecules 21 (2016) 432.
[33] A. Pestryakov, V. Lunin, A. Kharlanov, N. Bogdanchikova, I. Tuzovskaya, EPJD
24 (2003) 307.
[34] E. Smolentseva, N. Bogdanchikova, A. Simakov, A. Pestryakov, M. Avalos, M.H.
Farias, A. Tompos, V. Gurin, J. Nanosci. Nanotech. 7 (2007) 1882.
[35] A. Pestryakov, I. Tuzovskaya, E. Smolentseva, N. Bogdanchikova, F. Jentoft, A.
Knop-Gericke, Int. J. Mod. Phys. B 19 (2005) 2321.
[36] A.N. Pestryakov, V.V. Lunin, N. Bogdanchikova, O.N. Temkin, E. Smolentseva,
Fuel 110 (2013) 48.
[37] N. Bogdanchikova, A.N. Pestryakov, I. Tuzovskaya, T.A. Zepeda, M.H. Farias, H.
Tiznado, O.A. Martynyuk, Fuel 110 (2013) 40.
[38] N.M. Figueiredo, N.J.M. Carvalho, A. Cavaleiro, Appl. Surf. Sci. 257 (2011) 5793.
[39] S. Peters, S. Peredkov, M. Neeb, W. Eberhardt, M. Al-Hada, Surf. Sci. 608 (2013)
129.
[40] O. Martynyuk, Y. Kotolevich, R. Vélez, J.E. Cabrera Ortega, H. Tiznado, T.
Zepeda, J.D. Mota-Morales, A. Pestryakov, N. Bogdanchikova, Molecules 21
(2016) 432.
[41] G.M. Veith, A.R. Lupini, S.J. Pennycook, G.W. Ownby, N.J. Dudney, J. Catal. 231
(2005) 151.
[42] J. Llorca, M. Domínguez, C. Ledesma, R.J. Chimentão, F. Medina, J. Sueiras, I.
Angurell, M. Seco, O. Rossell, J. Catal. 258 (2008) 187.
[43] E. Kolobova, Y. Kotolevich, E. Pakrieva, G. Mamontov, M.H. Farías, N.
Bogdanchikova, V. Cortés Corberán, A. Pestryakov, Molecules 21 (2016) 486.
[4] T. Ishida, Y. Ogihara, H. Ohashi, T. Akita, T. Honma, H. Oji, M. Haruta,
ChemSusChem 5 (2012) 2243.
[5] A. Gomez-Aviles, S. Martinez-Gonzalez, S. Ivanova, M.I. Dominguez, J.A.
Odriozola, R. Fernandez-Lafuente, V. Cortes Corberan, 7th World Congress on
Oxidation Catalysis (7WCOC), St. Louis (USA), 2013, p. 93.
[6] V.C. Corberán, A. Gómez-Avilés, S. Martínez-González, S. Ivanovab, M.I.
Domínguez, M.E. González-Pérez, Catal. Today 238 (2014) 49.
[7] S. Martínez-González, A. Gómez-Avilés, O. Martynyuk, A. Pestryakov, N.
Bogdanchikova, V.C. Corberán, Catal. Today 227 (2014) 65.
[8] Y. Zhang, X. Cui, F. Shi, Y. Deng, Chem. Rev. 112 (2012) 2467.
[9] S.K. Klitgaard, A.T. Dela Riva, S. Helveg, R.M. Werchmeister, C.H. Christensen,
Catal. Lett. 126 (2008) 213.
[10] Y.F. Yang, P. Sangeetha, Y.W. Chen, Int. J. Hydrogen Energy 34 (2009) 8912.
[11] N. Hammer, I. Kvande, D. Chen, M. Rønning, Catal. Today 122 (2007) 365.
[12] F. Pinna, A. Olivo, V. Trevisan, F. Menegazzo, M. Signoretto, M. Manzoli, F.
Boccuzzi, Catal. Today 203 (2013) 196.
[13] J. Xu, J. Huang, Y.M. Liu, Y. Cao, Y.X. Li, K.N. Fan, Catal. Lett. 141 (2011) 198.
[14] S.E. Dapurkar, Z. Shervani, T. Yokoyama, Y. Ikushima, H. Kawanami, Catal. Lett.
130 (2009) 42.
[15] A.V. Kucherov, O.P. Tkachenko, O.A. Kirichenko, G.I. Kapustin, I.V. Mishin, K.V.
Klementiev, S. Ojala, L.M. Kustov, R. Keiski, Top. Catal. 52 (2009) 351.
[16] A. Zwijnenburg, M. Makkee, J.A. Moulijn, Appl. Catal. A Gen. 270 (2004) 49.
[17] S.A.C. Carabineiro, X. Chen, O. Martynyuk, N. Bogdanchikova, M. Avalos-Borja,
A. Pestryakov, P.B. Tavares, J.J.M. Orfao, M.F.R. Pereira, J.L. Figueiredo, Catal.
Today 244 (2015) 103.
[18] R. Zanella, L. Delannoy, C. Louis, Appl Catal. A 291 (2005) 62.
[19] R. Zanella, C. Louis, Catal. Today 107–08 (2005) 768.
Please cite this article in press as: Y. Kotolevich, et al., Au/TiO2 catalysts promoted with Fe and Mg for n-octanol oxidation under mild
conditions, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.05.002