Dehydrogenation properties of ZnO and the impact of gold nanoparticles on the process

Article abstract of DOI:10.1016/j.apcata.2016.01.020

The article presents the results of catalytic and surface properties of pure zinc oxide synthesized by hydrothermal method and surface-doped with gold nanoparticles. As a test reaction, the catalytic transformation of n-butyl alcohol towards the dehydrogenation or bimolecular condensation of symmetric ketone or an ester was studied. The tested materials catalyse both consecutive reactions, wherein the transformation towards the ketone is dependent on the presence of surface oxygen vacancies, whose concentration depends on the temperature. In turn, the transformation to the ester occurs in the presence of gold nanoparticles deposited on the surface of zinc oxide. The difference in work function of electrons from these materials create a change in the electron concentration in the surface area and will shift the balance of the coupling reaction of hydrogen with lattice oxygen, which prefers the formation of aldehyde and ester. The results were compared with the catalytic properties of other previously studied oxide systems in this group of changes. This analysis enabled the development of the mechanism of transformation and explanation of the impact of gold on the kinetics of the process.

Full text of DOI:10.1016/j.apcata.2016.01.020

Applied Catalysis A: General 514 (2016) 135–145
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Dehydrogenation properties of ZnO and the impact of gold
nanoparticles on the process
Patrycja Suchorska-Woz´niaka, Olga Raca,∗, Roman Klimkiewiczb, Marta Fiedota,
Helena Teterycza
a Faculty of Microsystem Electronics and Photonics, Wrocław University of Technology, Janiszewskiego 11/17, 50-372 Wrocław, Poland
b Wlodzimierz Trzebiatowski Institute of Low Temperature and Structure Research of the Polish Academy of Sciences, Okolna 2, 50-422 Wroclaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
The article presents the results of catalytic and surface properties of pure zinc oxide synthesized by
hydrothermal method and surface-doped with gold nanoparticles. As a test reaction, the catalytic trans-
formation of n-butyl alcohol towards the dehydrogenation or bimolecular condensation of symmetric
ketone or an ester was studied. The tested materials catalyse both consecutive reactions, wherein the
transformation towards the ketone is dependent on the presence of surface oxygen vacancies, whose
concentration depends on the temperature. In turn, the transformation to the ester occurs in the pres-
ence of gold nanoparticles deposited on the surface of zinc oxide. The difference in work function of
electrons from these materials create a change in the electron concentration in the surface area and will
shift the balance of the coupling reaction of hydrogen with lattice oxygen, which prefers the formation
of aldehyde and ester. The results were compared with the catalytic properties of other previously stud-
ied oxide systems in this group of changes. This analysis enabled the development of the mechanism of
transformation and explanation of the impact of gold on the kinetics of the process.
Received 24 July 2015
Received in revised form
22 December 2015
Accepted 15 January 2016
Available online 19 January 2016
Keywords:
Catalysis
Dehydrogenation
Gold nanoparticles
Zinc oxide
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
genation and hydrogen exchange reactions. The breakthrough was
achieved by Haruta et al. who showed that gold particles scattered
Zinc oxide is widely used in various areas of science and tech-
nology because of its physico-chemical and biological properties.
It is an important material in electronics and electrical engineer-
ing [1–4]. Because of its stability, zinc oxide was used in the first
chemical resistive gas sensors as a gas-sensitive material [5]. It is
still used in sensors of various gases (H2, SF6, C4H10, C2H5OH) [6–8].
The temperature of gas sensors comprising zinc oxide as a sensor
material is relatively high (400–500 ◦C), but if the sensors are made
of nanometric particles, then the operating temperature decreases
by approx. 200 ◦C [6]. The selectivity of the ZnO-based sensors is
small but it can be improved by doping ZnO with various com-
pounds. Among the dopes that significantly improve the sensors
parameters, there are gold nanoparticles [9–11].
on transition metal oxides as carriers demonstrate significantly
higher catalytic activity in the oxidation reaction of hydrogen and
carbon monoxide [14,15], and by Nkosi et al. in the hydrochlorina-
tion reactions [16]. In recent years, the role of gold in catalysis has
gained great importance [17–20].
Oxide sensor materials and phenomena occurring during gas
detection are typical of heterogeneous catalysis. For this reason,
the parameters of chemical resistance sensors, similarly to hetero-
geneous catalysts, are determined by the type, surface condition
and microstructure of the material. Preparation and methods for
characterization of materials in both cases are almost the same.
Only the final stages of preparing these materials (as sensor or as
catalytic materials) are different [21]. For these reasons, the authors
started working on determining the catalytic properties of pure zinc
oxide and zinc oxide surface-doped with gold in the transformation
reaction of n-butyl alcohol. This material was prepared for chem-
ical sensors for detection of low concentrations of volatile sulfur
compounds.
The catalytic properties of gold were noted by Bond et al. as early
as in the 1970’s [12,13]. They demonstrated that gold, although
has inferior catalytic properties in comparison with metals such
as platinum or palladium, it can catalyse hydrogenation, dehydro-
The article presents not only the results of the catalytic proper-
ties of pure ZnO and ZnO doped with gold, but it also proposes
the mechanism of the reactions taking place on the surface of
∗
Corresponding author.
E-mail address: olga.rac@pwr.edu.pl (O. Rac).
http://dx.doi.org/10.1016/j.apcata.2016.01.020
0926-860X/© 2016 Elsevier B.V. All rights reserved.

136
P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
Fig. 1. Microstructure of (a) ZnO undoped after drying; ZnO containing 1% by weight Au: (b) before the reaction; (c) after the reaction at 460 ◦C; (d) SnO2 after the reaction.
such amounts that the final concentration of ions in the Au3+ solu-
tion was 1 mM, and the concentration of Triton® X-100 was 0.2%
by weight. The molar ratio of PEI to chloroauric acid was 8:1 and
the molar ratio of ascorbic acid to chloroauric acid was 1:1. Ascor-
bic acid was added to a solution containing the precursor of gold
ions and stabilizers. The reduction of gold ions with ascorbic acid
resulted in obtaining a colloidal solution of gold having a red colour.
Surface doping of zinc oxide with gold nanoparticles was per-
formed by the process of reducting Au3+ gold ions in a suspension
containing grains of zinc oxide (ZnO). For this purpose, a solution of
surfactant, chloroauric acid and polyethyleneimine was added into
the suspension of zinc oxide obtained by hydrothermal method and
mixed. The content of gold ions in the solution relative to the weight
of zinc oxide varied between 0% and 2% by weight. It was followed
by reduction of gold ions with ascorbic acid, adding it slowly, por-
tionwise. The resulting nanoparticles were deposited on the surface
of the grains of zinc oxide. The resulting suspensions of zinc oxide
powder doped with gold were centrifuged, rinsed several times
with water in the decantation process and dried with a lyophilizer.
Then, oxides were heated at 500 ◦C for 30 min in order to remove
organic residues (PEI, Triton® X-100).
UV–vis spectroscopic measurements of the colloidal solutions
of gold were performed using the Optizen Aplha Hybrid Mecasys
spectrophotometer. Absorption spectra were recorded at room
temperature in the wavelength range of 200–700 nm. Quartz
cuvettes were used.
The size of the distribution of gold nanoparticles was deter-
mined by dynamic light scattering (DLS). Tests were performed
using a Nicomp 380 ZLS (Particle Sizing Systems, USA) device
with a laser having a 532 nm wavelength and a 50 mW power.
The frequency of photon counting by autocorrelator was set at
approximately 200 kHz, and a single measurement time was 3 min.
In the measurements, a 40 × 10 × 10 mm polymethylmethacrylate
(PMMA) cuvette was used.
that catalyst. Dehydrogenation properties of zinc oxide have long
been known [22,23], but publications relate to secondary alco-
hols [24–27] carboxylic acids [28] or dehydrogenation of primary
structures [29] without consecutive reactions, i.e. bimolecular con-
densation towards the symmetric ketone [30] or ester [31,32].
Dehydration features [33] are also observed.
2. Experimental
The particles of zinc oxide were obtained by hydrothermal
method. For this purpose, two aqueous solutions were prepared:
1 M zinc nitrate Zn(NO3)2 and 1 M solution of hexamethylenete-
tramine (HMT). These solutions were mixed in the 1:1 proportion
and diluted with water to the final concentration of 100 mM.
The reaction mixture containing both reactants was filtered and
was heated at 90 ◦C for 6 h. Simultaneously, a series of Reactions
(1)–(7) [34,35] occurred in the volume of the solution, resulting in
microstructures of zinc oxide were formed.
Zn NO
→ Zn2+ + 2NO−3
(1)
(2)
(3)
(
)
2
3
C6H12N4 + 6H2O → 6CH2O + 4NH3
NH3 + H2O → NH+ + OH−
4
2+
Zn2+ + 4NH3 → Zn NH
(4)
(5)
(6)
(7)
(
)
3
4
2−
Zn2+ + 4OH− → Zn OH
(
)
4
Zn NH 2+ + 2 OH− → ZnO + 4NH3 + H2O
(
)
4
3
2−
Zn OH
(
→ ZnO + H2O + 2OH−
)
4
The resulting ZnO suspension was filtered, and repeatedly
rinsed with deionized water having a conductivity of 0.5 ␮S/cm.
The gold nanoparticles were synthesized by chemical reduction
of gold ions in the aqueous solution in the presence of surfactant
and polymer as stabilizers. For this purpose, solutions of: 0.08 M
gold chloride (HAuCl4), 0.06 M ascorbic acid, low molecular weight
polyethyleneimine (PEI) having an average molecular weight of
10 kDa at a concentration of 43 g/dm3, and 0.08 M Triton® X-100
were prepared. The prepared solutions were mixed together in
The phase composition of the catalysts was determined on the
basis of the diffraction patterns obtained using an X-ray Philips
Materials Research Diffractometer (MRD) with monochromatic
micro-focusing source of CuK␣ radiation. Typical powder materials
were used, i.e. ꢀ/2ꢀ scan type.

P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
137
The microstructure of zinc oxide particles was examined by the
Ziess EVO LS15 scanning electron microscope: acceleration voltage
of 20 kV, sample current of 100 pA, equipped with an adapter for
qualitative and quantitative analysis by the non-model method.
The substructure of gold nanoparticles was observed by the
EM900 (Zeiss, Germany) transmission electron microscope (TEM).
The magnification used was equal to 81000× . The samples were
prepared by depositing drops of an aqueous dispersion of nanopar-
ticles onto 300 copper mesh, coated with a layer of carbon and
Formvar resin. The prepared samples were dried at ambient atmo-
sphere for 45 min, and then subjected to a microscopic observation.
The surface chemistry of pure and Au-doped ZnO catalysts
including their contaminations was characterized by XPS method
using the SPECS XPS spectrometer working under ultrahigh
vacuum conditions at 10−7 Pa maintained by the ion-sorption ultra-
high vacuum pump (IP-300 VARIAN model). It consists of vacuum
chamber equipped, among other, with the sample XYZ manipulator
(PREVAC SM-100 model), an X-ray source (Al-K␣ 1486.6 eV; XR-50
SPECS model) and a concentric hemispherical analyzer (PHOIBOS-
100 SPECS model). The XPS survey spectra, XPS spectral windows,
and XPS spectral lines of basic elements were recorded, and the reg-
istered binding-energy (BE) scale was calibrated using the Au4f7/2
peak at 84.0 eV. The analysis of XPS data was performed using the
CASA XPS Software.
All the powders of ZnO pure and Au doped were pelleted at
80 MPa pressure, crushed and sifted until a homogeneous granula-
tion of the catalysts was obtained (0.6–1.2 mm).
The catalytic activity of the new materials was studied in a fixed
bed standard flow system at the atmospheric pressure and in tem-
perature 400 ◦C and 460 ◦C with 1 h−1 load. The vertical quartz
reactor with 3 cm3 catalyst bed was used. The studies on catalytic
properties of zinc oxide compositions in the reactions of primary
alcohols conversion were carried out with the use of n-butanol. No
carrier gas was utilised. The substrate was fed in a continuous man-
ner from the top of the reactor with the use of a micropump with
the LHSV = 3.0 cm3/cm3 h. The products of the reaction were ana-
lysed with the use of a HP MSD 59 mass spectroscope and HP 6890
gas chromatograph.
3. Results
The microstructure of zinc oxide before/after doping with gold
and before/after the catalytic transformation of n-butanol was
examined using the SEM. The comparative analysis of SEM images
demonstrated that during the heating process, there is a change
in the microstructure of zinc oxide. Pure and dried zinc oxide is
composed of numerous grains having a spherical shape and a size
within the range of from 0.1 ␮m to 2 ␮m (Fig. 1a). There are also
visible larger structures having a hexagonal shape, and agglom-
erates of different sizes. After heating, the microstructure of pure
and surface-doped zinc oxide slightly changed (Fig. 1b). The grains
maintained an irregular shape, but also larger grains and agglomer-
ates can be found. In contrast, the microstructure of these materials
has not changed during the catalytic reaction of transforming n-
butyl alcohol at 460 ◦C (Fig. 1c). Under the reaction conditions, the
catalysts based on zinc oxide did not undergo a reduction, in con-
trast to the previously tested catalysts based on tin dioxide (Fig. 1d)
[36].
XRD tests have shown that the crystallographic structure of
tested catalysts is the same as the crystallographic structure of
standard zinc oxide. The grains of zinc oxide crystallized in the
wurtzite-type structure with hexagonal unit cell, space group of
P63mc and lattice parameters a = 3.2501 Å and c = 5.2071 Å. The
pure ZnO was a partial amorphous but after the process the material
becomes crystalline (Fig. 2a). The zinc oxide doped with nanopar-
Fig. 2. X-ray diffraction pattern of catalysis materials: (a) zinc oxide before and after
reaction; (b) zinc oxide doped 1% Au before and after reaction; (c) zinc oxide doped
1% Au after reaction and AuZn alloy.
ticles is more crystalline before reaction, because it is heated after
the doping process (at 500 ◦C for 30 min) in order to remove organic
residues (PEI, Triton® X-100). On diffraction patterns, no peaks
characteristic for gold were observed, when its content is less than
1 wt.%. This is due to a very small average molecular size and large
dispersion over the surface of the ZnO grains (Fig. 2b). The differ-
ence which exists between these two materials is the formation
of alloy from interstitial zinc atoms with nanoparticles of gold
(Fig. 2c).

138
P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
Table 1
Strains and size of ZnO crystallites.
Sample
ε
D [nm]
␯
ZnO before
ZnO after
ZnO + 1% Au before
ZnO + 1% Au after
0.00453
0.00426
0.00178
0.00275
34.74
158.88
112.49
571.35
Fig. 3. The W–H analysis of ZnO + 1% Au before reaction.
By using of the Williamson–Hall (W–H) method (1) the average
crystallite size and microstrains in ZnO were determined [37,38]:
kꢂ
ˇhkl cos ꢁ =
+ 4ε sin ꢁ
(1)
(
)
( )
Dꢃ
where:
ˇ—half-width of the (FWHM) peak [rad],
ϕ—the angle of the highlight for the given band interference
[rad],
Fig. 4. UV–vis spectrum of gold nanoparticles.
k—the Scherrer constant (the 0.9 value),
ꢂ—the length of the X-ray beam [nm],
in samples doped with gold nanoparticles. Such doped material
has much larger size than the pure ZnO. It should be noted that
the increase of the dimension of crystallites is caused by addi-
tional annealing of samples after the doping process. Similarly to
the samples without dopant a several fold increase in size of crys-
tallites is observed. The increase of strains after the reaction with
n-butyl alcohol could be associated with the formation of Zn–Au
alloy (Table 1).
In the UV–vis range, on the light absorption spectra of colloidal
solutions of gold, there was observed a characteristic narrow peak
at a radiation wavelength of 530 nm, which according to the Mie
theory corresponds to nanoparticles having a diameter of not more
than 10 nm (Fig. 4). These values are consistent with the values
typical for the spherical forms of gold nanoparticles [39].
In order to confirm the results of the tests of the UV–vis radiation
absorption, the average size of gold nanoparticles was determined
by the dynamic light scattering (DLS) method. Studies have shown
D —the average crystallite size [nm],
␯
ε—strain.
This equation represents Uniform Deformation Model (UDM)
which takes into account uniform strain in all crystallographic
directions, thus assuming the isotropic nature of analysed crys-
tal. To calculate crystallite size and strain curve ␤cos␸ = f (4sin␸)
was plotted for preferred diffraction peaks (1 0 0), (0 0 2), (1 0 1),
(1 0 2), (1 1 0), (1 0 3). The W–H analysis of zinc oxide doped with
gold before reaction was demonstrated (Fig. 3).
Results clearly showed that crystallite size of the ZnO before
the reaction was found to be 35 nm and that after reaction was
159 nm. This is due to the fact that analysed catalytic reaction takes
place at an elevated temperature at which occurs recrystallization
of ZnO. Moreover, after the reaction internal strains in the mate-
rial are slightly reduced. This is because the reorganization of the
structure at higher temperatures. Different situation is observed
Fig. 5. Distribution of the size of gold nanoparticles performed by the DLS method.

P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
139
At 400 ◦C, the conversion rate in the presence of pure zinc oxide
was only approximately 27%, and along with the increasing gold
dopant, it increased to 60%. So, the presence of gold doubles the
conversion. Since increasing the temperature by 60 ◦C resulted in a
threefold increase in the conversion in the case of pure zinc oxide,
the gold dopant catalytic effect manifested itself more evidently in
respect of selectivity.
Also, large differences exist in the efficiency and selectivity
(Figs. 10 and 11). At 400 ◦C, in the presence of pure zinc oxide,
butyraldehyde is almost exclusively formed. When the ZnO catalyst
is doped with gold, the total dehydrogenation efficiency is much
higher, because the products contain little more aldehyde and a
large amount of ester—the subsequent product of dehydrogena-
tion (Tishchenko reaction) [31,32]. Ester efficiency values vary to a
small degree along with the increasing dopant concentration above
0.5% by weight. When the gold content is 2%, ketone appears in the
reaction products, which is also a subsequent product of dehydro-
genation and bimolecular condensation (Fig. 10a). In contrast, at a
temperature of 460 ◦C, the ketone is formed at all the tested cat-
alysts, even in the undoped zinc oxide. In the doped catalysts, its
efficiency is comparable to the efficiency of the ester (Fig. 10b).
In this case, the total efficiency of the dehydrogenation products
(aldehyde, ketone, ester) exceeds 90%.
The selectivity of the reaction in the presence of pure ZnO is
very high compared to butyraldehyde. When ZnO is doped with Au,
butyl butyrate is the product resulting from the additional expense
of aldehyde at both temperatures. This reduces the selectivity to
aldehyde by approximately 10% and 25%, respectively, at lower and
higher temperatures. At 400 ◦C the selectivity of the butyl butyrate
reaches the highest value when the dopant concentration is 0.5% by
weight. (Fig. 11a). The ketone at the lower temperature is formed
on the most active catalyst with a gold content of more than 1%
by weight. At higher temperatures, the selectivity to the ketone is
higher after the doping with gold, but it depends on the amount
of the gold dopant when its content is more than 0.25% by weight.
(Fig. 11b).
Fig. 6. The picture in atomic number contrast of ZnO powder with 2% wag. Au: (a)
before reaction; (b) after reaction.
that the average particle size of the gold is approximately 4.0 nm
for the vast majority of particles (93.5%) (Fig. 5).
On the basis of the results of the SEM and EDX tests, it was found
that the surface of the ZnO grains there are gold particles having
very different sizes, when their content is high (Figs. 6a and 7a ). No
presence of these particles was determined during the observation
of powder containing only 0.25% by weight Au, which is caused by a
large development of zinc oxide surface and highly uniform disper-
sion of gold nanoparticles (Fig. 7b). The distribution and size of the
gold particles have not changed during the catalytic transformation
of alcohol (Fig. 6b).
The XPS analysis was performed to determination of the sur-
face chemistry. On the overall spectra strong contribution of basic
elements of ZnO catalysts was observed from the core level Zn2p,
O1s, C1s, Zn3s, Zn3p and Zn3d lines for Zn, reference line Au4f7/2
at 85.0 eV for gold and from C1s lines at BE ∼286.0 eV for carbon
(Fig. 8a). Weak trace of contribution of Au at the surface of Au-doped
ZnO (Fig. 8b) is caused by the formation of alloy from interstitial zinc
atoms with gold nanoparticles (Fig. 2c). Before reaction, for both
catalysts pure and Au-doped ZnO, mainly on the surface O
C OH
bondings are observed. It is the residue after the synthesis of zinc
oxide, which was used hexamethylenetetramine. After reaction on
the both surfaces only C OH bondings were detected. They are the
remnants of the catalytic reaction products (Fig. 8c).
The catalytic properties of pure zinc oxide as well as zinc oxide
surface-modified by gold nanoparticles were tested in the reac-
tions of transformation of n-butyl alcohol. These reactions included
dehydrogenation of aldehyde (butyraldehyde) and bimolecular
condensation to ester (butyl butyrate), and ketone (heptan-4-one).
These tests were carried out in a flow reactor at 1 h−1 load at two
temperatures of 400 ◦C and 460 ◦C. These results were compared
with the results of other metal oxides used as catalysts in the same
reaction.
It is worth mentioning that high activity of ZnO surface-modified
gold nanoparticles was observed in oxidation reactions both with-
out and in the presence of UV radiation [40,41].
All the catalytic experiments were performed in the func-
tion of an increasing temperature. To check reproducibility, after
an interruption of the experiments, cooling down the catalytic
bed, the restarted system—after return to the starting param-
eters and then to the chosen temperature—enabled to obtain
a satisfactory repetitiveness of results. For ZnO containing 1%
by weight Au, at 460 ◦C, yields of butyraldehyde, butyl butyrate
and heptan-4-one were 42%, 24%, 24% in the first run and 43%,
The conversion rate of the transformation reaction of n-butyl
alcohol conducted in the presence of catalysts based on zinc oxide
depended on the temperature and the content of the dope (Fig. 9).
Fig. 7. The result of the EDX analysis: (a) the dependence of the intensity of the signal measurement on the energy of the electrons incident on the surface of zinc oxide
doped with 0.5% by weight gold nanoparticles; (b) the distribution of gold atoms over the surface of zinc oxide doped with gold nanoparticles.

140
P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
Fig. 8. The result of the XPS analysis: (a) overall spectra; (b) range spectra of core level for Au and Zn; (c) range spectra of core level for carbon C1s.
bon balance corresponds to the sum of unreacted butanol and
the cumulative yield of distinguished dehydrogenation products.
Therefore, at 400 ◦C and at 460 ◦C, the carbon balance ranges
between 94–98% and 92–96%, respectively.
4. Discussion
The presented reaction of transformation of n-butyl alcohol
occurs on the border of solid-gas phases. Its progress, and con-
sequently the efficiency and selectivity depend not only on the
structural characteristics, but also on the electrical properties of
that material [42].
In the present study, pure zinc oxide and zinc oxide surface-
doped with gold nanoparticles was used as a catalyst. Pure ZnO
demonstrated high efficiency and selectivity in relation to the
aldehyde at both temperatures (Figs. 10 and 11). In contrast, the
material surface-doped with gold also catalysed the subsequent
reaction to an ester with high selectivity and efficiency. At higher
temperatures, i.e. 460 ◦C, the significant proportion of the ketone
also made up a higher total selectivity and efficiency of dehydro-
genation products, similar to the pure zinc oxide and the doped
zinc oxide (Fig. 11).
Fig. 9. The conversion rate of the transformation reaction of the n-butyl alcohol in
the presence of the catalysts based on ZnO with different gold content. The reaction
was carried out in a fixed bed standard flow system at the atmospheric pressure and
in temperature 400 ◦C and 460 ◦C with 1 h−1 load.
The efficiency and selectivity of the reaction conducted in
relation to the pure zinc oxide differed from the efficiency and
selectivity of that process previously conducted on the catalysts
created on the basis of tin dioxide [43], and stabilized zirconium
dioxide (Fig. 12) [44]. The main difference in the results of the catal-
23%, 24% in the second run, respectively, at practically the same
conversions.
Since no carbonaceous deposits were accumulated on the
catalyst surface (Figs. 1c and 6b), it can be assumed that the car-

P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
141
Fig. 10. The dependence of efficiency of the formation of aldehyde, ester, ketone and other (not analysed) products as a function of the amount of the Au dopant.
Fig. 11. The dependence of the selectivity of the formation of aldehyde, ester, ketone and other (not analysed) products as a function of the amount of the dopant.
ysis on the pure zinc oxide relative to the comparative materials is
much higher selectivity to the aldehyde.
photocatalysis [47–49]. The formation of point defects with the
ZnO structure causes the change of the ZnO volume and surface
parameters [45,46].
These native point defects occurring in ZnO, which significantly
affect its properties recorded in the Kröger–Vink notation include:
These previously tested oxides are the materials having the
characteristics of the n-type semiconductors (SnO2, SnO2 + CeO2) or
solid electrolytes (CeO2, ZrO2 stabilized by MgO). At a temperature
of 400 ◦C, all these materials are characterized by both a higher effi-
ciency and selectivity to the ketone than ZnO-based catalysts. The
catalysts based on tin dioxide or stabilized by zirconium dioxide
possess strong Lewis acid sites [44]. The presence of these centers
is due to the existence of these materials in a number of surface
oxygen vacancies [44]. Their concentration changes as a function
of the partial gas pressure of oxygen in the ambient atmosphere
and temperature. In the case of stabilized zirconium dioxide, their
concentration depends only on the concentration of the stabilizing
dopant. Oxygen vacancies are also present in zinc oxide. However,
a unique feature of ZnO as the semiconductor photocatalyst is its
open structure and the presence of the series of other structural
defects in addition to oxygen vacancies [45,46].
Zinc oxide crystallizes in the wurtzite-type hexagonal structure,
wherein the zinc atoms occupy half of the tetrahedral interstice
created by oxygen atoms, and all octahedral interstice are empty.
It is due to the numerous intrinsic defects occurring in the zinc
oxide and the crystallization of this material in the form of a num-
ber of different microstructures, it is currently intensively studied
in terms of applications in optoelectronics, sensor technique and
ꢀꢀ
- zinc interstice (V ),
Zn
- oxygen interstice (V•O• ),
- zinc atoms in interstitial sites (Zn2i +),
- oxygen atoms in interstitial sites (O2i −),
- zinc atoms in place of oxygen (Zn•O••• ),
ꢀꢀꢀꢀ
- oxygen atoms in place of zinc (O ).
Zn
In ZnO, there are also combinations of defects, e.g. Schottky
pair (anion and cation interstice) and Frenkel pairs (cation gap and
interstitial cation). The energy needed for the creation of the defect
depends on the difference between the charge of the given defect
and the charge on the lattice in which the defect is located. In ZnO,
interstices and interstitial atoms have a charge of 2, and antisite
defect 4 [50].
According to some scientists, the dominant defects in the zinc
ꢀ
ꢁ
oxide are oxygen vacancies V•O• . In contrast, other authors believe
that zinc is the dominant defect in the interstitial position Zni
[51,52]. These defects result from the Frenkel and Schottky Reaction

142
P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
Fig. 12. The efficiency and selectivity of the condensation reaction of n-butanol in the presence of various oxides at: (a, b) 400 ◦C; (c, d) 460 ◦C.
ZnxZn + OxO ↔ Znx + Vꢀꢀ + OOx + V•O• (Schottky defect)
Zn
Zn
OxO ↔ O•O + eꢀ(ionization)
(12)
(13)
OO•
↔
+ V•O• + e (decompositon reaction)
1
2O2
The concentration of these major defects in ZnO depends on
the oxygen gas pressure, and can be described by the following
relationships (14)–(18) [53]:
1
ꢀ
ꢁ
O2 + V•• + 2eꢀ ↔ OO; VO ∝ pO−
(14)
(15)
(16)
(17)
(18)
1
2
••
x
2
O
2
1
2
ꢀ
ꢁ
ꢀꢀ
ꢀꢀ
O2 + Vꢀꢀ ↔ O ; V
∝ p−O
1
2
Zn
Zn
Zn
2
Fig. 13. Diagram of the energy levels resulting from the position arising from
the presence of occurrence of spotting defects in ZnO. To describe the dia-
gram Kröger–Vink notations was used: i = interstitial site, Zn = zinc, O = oxygen,
and V = vacancy. Superscript symbols indicate charges: a dot—positive charge, a
prime—negative charge and a symbol x—zero charge.
1
2
ꢀ
ꢁ
1
2
x
Zn0i
1
+
O2 ↔ Znx + OO; Zno ∝ p−O
Zn
i
2
1
ꢀ
ꢁ
O2 ↔ Oo; Oo ∝ pO2
i
i
2
2
(8)–(13) during the heat treatment of ZnO and cause the creation of
the number of donor and acceptor levels in the band gap (Fig. 13).
1
ꢀ
ꢁ
ꢀ
ꢁ
ꢀꢀ
ꢀꢀ
O2 + Vꢀꢀ ↔ O ; O
∝ pO2
×
V
Zn
1
2
ꢀꢀ
Zn
Zn
Zn
2
ZnxZn ↔ Zn2+ + VZꢀꢀ n(Frenkel defect)
(8)
ꢀ
ꢁ
ꢀ
ꢁ
where: V•O• and
V
is the concentration of non-ionized vacan-
ꢀꢀ
Zn
i
ꢀ
ꢁ ꢀ
ꢁ
ꢀ
ꢁ
ꢀꢀꢀꢀ
Zn
cies of oxygen and zinc, respectively. While Znoi , a Oio nd
are concentrations of the interstitial zinc, the interstitial oxygen
and another atom, i.e. oxygen in the zinc node. Eqs. (14)–(16) show
O
Zn2i + + eꢀ ↔ Zn+i (1.degree of ionization)
Zn+i + eꢀ ↔ Znoi (2.degree of ionization)
(9)
(10)

P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
143
ꢀ
ꢁ
ꢀ
that the concentration of oxygen vacancies V•O• , zinc vacancies
ꢀ
ꢁ
ꢀ
ꢁ
ꢀꢀ
V
and interstitial zinc Znoi increases as the oxygen pressure
Zn
decreases pO2 . In contrast, the other Eqs. (17) and (18) show that
ꢁ
that the concentration of the interstitial oxygen Ooi and oxygen
in zinc ion nodes increases along with the increasing oxygen pres-
sure pO2 . Accordingly, heating in the anaerobic atmosphere, i.e. in
the atmosphere, in which the reaction was conducted, causes an
ꢀ
ꢁ
increase in the concentration of both the oxygen vacancies VO••
ꢀ
ꢁ
and interstitial zinc Znoi and the atrophy of interstitial oxygen
ꢀ
ꢁ
Ooi and oxygen in the zinc ion nodes.
In the oxygen-containing atmosphere on the surface oxygen
vacancies, oxygen undergoes physical and chemical adsorption
(19).
O2(gas) ↔ O2(abs)
(19)
As the temperature increases, the oxygen molecules ionize and
dissociate (20)–(21).
Fig. 14. The distribution pattern of gold nanoparticles on the surface of the zinc
oxide grains.
O2(ads) + e− ↔ O−2
O−2 + e− ↔ 2O−
(20)
(21)
At 400 ◦C, butene is one of the reaction products (in particular
the reaction carried out in the presence of pure ZnO) (Fig. 10). This
compound can be produced in the dehydration Reaction (32).
In an atmosphere containing alcohol, the oxygen ions adsorbed
on the surface vacancies may react in two ways, by undergoing
dehydration (22) or dehydrogenation (23)–(24) [54,55].
•
O
RCH2OH + V•• + OOx → CH3CH2CHCH2 + 2 OH
(32)
ZnO;T
(
)
O
RCH2OH(gas) + O− ↔ RCHO + H2O + e−
RCH2OH(gas) ↔ RCH2O(surf) + H
(22)
(23)
(24)
Taken into account that, under the reaction conditions, there is
no modification in the microstructure and in the crystalline struc-
ture of zinc oxide, it should therefore be assumed that at this
temperature there is no Reaction (31), because then it would result
in the reduction of zinc oxide and an increase in the concentration
of surface oxygen vacancies that significantly affect the kinetics
of the tested process [44]. In addition, in the process of thermal
decomposition, in the zinc oxide the excess zinc ions pass into the
interstitial positions and migrate towards the surface. This migra-
tion of zinc ions also reduces the increase in the concentration of
oxygen vacancies in the material. This is consistent with the litera-
ture data, which show that the thermal decomposition of zinc oxide
takes place very slowly below 500 ◦C [58]. This cations migration is
not observed in tin dioxide and cerium dioxide.
RCH2O(surf ↔ RCHO + H
)
In our studies, the reaction is carried out under anaerobic con-
ditions at a temperature of 400 ◦C and 460 ◦C. The atmosphere
contained only n-butyl alcohol. Under these conditions, the par-
ticles of alcohol may also be adsorbed on the surface vacancies of
zinc oxide, which are the Lewis acid sites. These vacancies interact
with the oxygen of the hydroxyl group. In contrast, the (electron-
poor) carbon atom of the hydroxyl group becomes the center of the
nucleophilic attack of nucleophilic agents which in these materials
are surface (lattice) oxygen ions. This is consistent with the results
obtained by Bielan´ ski and Haber [56].
In the first stage of the reaction, the adsorption of alcohol
molecules takes place on the surface vacancies of the catalyst (Lewis
acid sites), and the gradual dehydrogenation process (25)–(26)
takes place. Hydrogen adsorbs on the electrophilic center (lat-
tice oxygen) according to the Reaction (27) [21,57]. The resulting
aldehyde is desorbed (28) or it reacts in the consecutive reaction
with the intermediate product to form an ester (29). Groups may
undergo the Reaction (30) to form hydrogen or may undergo the
Reaction (31) to form water and oxygen vacanices. If this reaction
takes place, zinc oxide will be reduced, similar to the catalysts based
on tin or cerium dioxide [36].
By analysing the catalytic properties of the gold particles
deposited on the transition metal oxides, it is commonly believed
that the dominant role is played by the oxygen molecules that
dissociate on the surface of the gold [59,60]. At elevated temper-
atures, the atomic oxygen undergoes recombination to O2. The
temperature of the recombination depends on the crystallographic
orientation of the gold surface. It occurs on the surface of the poly-
crystalline Au at 650 K (377 ◦C) [55], on the gold surface having the
Au (1 1 0) orientation it occurs at 590 K (317 ◦C) [61] and on the gold
surface with the Au (1 0 0) orientation it occurs at 470 K (200 ◦C)
[62]. In addition to this chemical aspect, the electronic aspect must
also be taken into account [42].
Au nanoparticles randomly distributed on the surface of the
ZnO particles cause the formation of the Schottky barriers between
the metallic Au islands and the semiconductor ZnO. The work
function of the electrons from zinc oxide—the n-type semicon-
ductor (ФZnO = 5.30 eV) is different than in from noble metals
(ФAu = 5.10 eV, ФPt = 5.65 eV). Therefore, the Schottky barrier is
formed at the interface of zinc oxide and gold. Since the Fermi lev-
els of Au and ZnO must be balanced, the electrons pass from gold
into ZnO [63]. Thus, the Au nanoparticles cause local changes in
the Fermi level of zinc oxide [64]. As a result of this process, the
Au islands become positively charged and the layer enriched with
electrons (Fig. 14) is formed in the vicinity of the islands on the
surface of the ZnO particles. For this reason, the conductance of the
•
RCH2OH + V•• → RCH2OO + H•
(25)
ZnO;T
O
RCH2O• → RCHOxO + H•
(26)
(27)
ZnO;T
O
H + OxO ꢀ (OH)•O + e−
••
ZnO;T
RCHOxO → RCHO ↑ +VO
(28)
(29)
(30)
(31)
RCH2O•O + RCHOxO
→
ZnO;T; AuRCOOCH2R ↑ +H•
•
ZnO;T
2 OH
(
→ 2OxO + H2
↑
→ H2O ↑ +OxO + V•O•
)
O
•
ZnO;T
2 OH
(
)
O

144
P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
ZnO grains modified with gold nanoparticles is greater than that of
the pure ZnO.
on ZnO is a consecutive reaction occurring at a higher temperature,
because along with the increasing temperature, the concentration
of the surface vacancies increases. However, an increase in their
concentration is less than the one on the catalysts based on tin
oxide, since in zinc oxide there is also the reaction of oxygen and
interstitial zinc diffusion towards the surface of ZnO.
The height of the Schottky barrier formed on the Au–ZnO border
affects not only the process of the electric charge transportation,
but also on the kinetics of the adsorption process in the area of
the junction. In the metal-semiconductor boundary area, the crys-
talline structure of metal may also change, which also affects the
adsorption process. In contrast, the height and width of the barrier
may change due to the adsorption or desorption of gas molecules.
In addition, gold is also known as the promoter of the catalytic dis-
sociation of oxygen molecules, which is also known as a spillover
phenomenon in the catalysis [65]. The resulting oxygen atoms
migrate to the sensor surface and form oxygen ions that capture free
electrons from the semiconductor. Since the reaction was carried
out in an anaerobic atmosphere, the spillover phenomenon was not
considered during the analysis of the occurring phenomena.
It is known that the ZnO activity towards the dehydrogenation
or dehydration of alcohols is also dependent on the morphology
[66]. In the present paper, zinc oxide was synthesized under the
same conditions, and according to the microscopic observations,
the structure of the catalysts was very similar. Moreover, the pres-
ence of gold nanoparticles on the surface of the grains is likely to
partially reduce their sintering due to the high energy value of the
phase boundary [67]. The microscopic observations confirmed the
literature findings, since there was no change in the microstruc-
ture of ZnO during the reaction carried out at 460 ◦C (Fig. 1c). It was
therefore assumed that the energy of the phase boundary between
gold and zinc oxide has a significant impact on the properties of the
tested catalysts.
By analyzing the impact of gold nanoparticles on the kinetics of
the reaction, it was observed that at the temperature of 400 ◦C in
the presence of pure ZnO, only the reaction of formation of the alde-
hyde (26) is preferred, whereas in the presence of the ZnO-based
catalysts doped with gold, the reaction of formation of the aldehyde
(26) and the consecutive reaction of formation of the ester (29) is
preferred. Both of these reactions are dehydrogenation reactions. A
thorough analysis of the proposed mechanism leads to the conclu-
sion that the presence of Au nanoparticles shifts the equilibrium of
the Reaction (27) to the left. Shifting the equilibrium is caused by
to the transition of electrons from gold to zinc oxide (Fig. 14). As
a result, there is also a shift in the equilibrium Reaction (26) also
to the left. As a consequence of these changes, the ester formation
Reaction (29) may occur.
5. Conclusions
In the present study, the kinetics of the transformation of the
n-butyl alcohol towards the dehydrogenation or bimolecular con-
densation in the presence of the catalysts based on zinc oxide
and zinc oxide surface-doped with gold nanoparticles was inves-
tigated. Zinc oxide was synthesized by the hydrothermal method.
The nanoparticles of gold were obtained by reduction of chloroau-
ric acid (HAuCl4) with ascorbic acid. As stabilizers in the reaction
of gold reduction, we have used low-molecular polyethyleneimine
(PEI) having an average molecular weight of 10 kDa and Triton®
X-100. The reduction of the gold ions was carried out in a solution
containing zinc oxide.
The obtained zinc oxide had the wurtzite-type crystalline
structure and its structure did not change after doping with
gold nanoparticles. The EDX studies demonstrated that the gold
nanoparticles are uniformly dispersed on the surface of zinc oxide.
There was no reduction of zinc oxide in the reaction conducted at
the temperatures of 400 ◦C and 460 ◦C.
At a temperature of 400 ◦C in the presence of a pure zinc oxide,
aldehyde is the main product of n-butyl alcohol transformation. By
increasing the temperature to 460 ◦C, additionally the ketone was
also the product. In the presence of zinc oxide doped with gold
when the reaction is carried out at a 400 ◦C, then in addition to
the aldehyde, ester also was created while in higher temperature
aldehyde, ester and ketone were presented.
The catalytic properties of tested catalysts were compared with
other previously studied oxides. In comparison with them, zinc
oxide was characterized by a high selectivity of dehydrogenation to
the aldehyde, but a low selectivity in relation to the ketone, where
zinc oxide doped with gold had a high selectivity towards the ester.
Through the analysis of the literature and research results, the
authors were able to develop a mechanism for this reaction and
explain the impact of gold on the kinetics of the process. Due to
the difference in their work function, the electrons from the gold
particles pass to zinc oxide. This causes the shift to the left of the
equilibrium of joining hydrogen and lattice oxygen. Consequently,
in the presence of the ZnO-based catalysts doped with Au, the dehy-
drogenation and the formation of aldehyde and the consecutive
reaction of ester formation is preferred.
The differences in the catalytic activity of zinc oxide and tin
oxides have also been discovered by testing their sensory prop-
erties, in an atmosphere of 1-butanol and butanal [49]. When
examining the composition of the gas atmosphere during sensors
operation, the authors found that on the sensor surface of the zinc
oxide layer, the intensive dehydrogenation of butanol into butanal
occurs. In contrast, they had suggested that esters may be formed
by further oxidation of butanal to butanoic acid and the subse-
quent esterification with butanol to form the butyl ester of butanoic
acid. The authors of this article observed this behaviour when they
applied zinc oxide doped with gold nanoparticles as a catalyst.
According to the Frost theory the junction metal/oxide disturbs
the balance of ZnO defects [68]. Ghiotti et al. has been reported that
cooper in Cu/ZnO catalyst for methanol synthesis is able to activate
the CO molecule [69], but in case of Au/ZnO catalyst gold is not
able to activate the support [70]. According to the above-presented
theory the presence of gold nanoparticles do not disturb the bal-
ance of defects in ZnO but is a result of enriched layer formation
(Fig. 14) at Au/ZnO boundary. Therefore the oxygen adsorbed in
the region of enriched layer adjacent to the gold makes that nucle-
ophilic attack of surface oxygen atoms is facilitated. Also, FTIR tests
[70,71] shows that gold nanoparticles do not activate the oxidative
dehydrogenation but modify the electrical properties of the zinc
oxide.
At a temperature of 460 ◦C, in the ZnO-based catalysts, addi-
tional oxygen vacancies begin to form, which participate in the
reaction of the ketone [28] formation. The reaction temperature
(460 ◦C) is close to the temperature of thermal decomposition
(above 500 ◦C) of zinc oxide [58]; furthermore, the reaction is
carried out in an anaerobic atmosphere. In this atmosphere, accord-
ing to the Reaction (14), the concentration of oxygen vacancies
increases. In addition, in this reaction, electrons are produced that
limit the impact of the electric potential of gold nanoparticles in
the surface area (Fig. 8). Thus, the ketonization reaction that occurs
Acknowledgements
The project was funded by the National Science Centre based on
Decision Number DEC-2012/07/N/ST7/02304 and from Funds for
Science during 2014–2015 for Project No. 0067 W12.

P. Suchorska-Wo´zniak et al. / Applied Catalysis A: General 514 (2016) 135–145
145
References
[36] H., Teterycz, R., Klimkiewicz, B. Licznerski, Conference proceedings of 26th
International Spring Seminar on Electronics Technology, ISSE 2003,
Piscataway, NJ : IEEE (2003) 379–383.
[37] G.K. Williamson, W.H. Hall, Acta Metall. 1 (1953) 22–31.
[38] R. Yogamalar, R. Srinivasan, A. Vinu, K. Ariga, A. Chandra Bose, Solid State
Commun. 149 (2009) 1919–1923.
[39] P.K. Jain, K.S. Lee, I.H. El-Sayed, M.A. El-Sayed, J. Phys. Chem. B 110 (2006)
7238–7248.
[40] S.A.C. Carabineiro, B.F. Machado, R.R. Bacsa, P. Serp, G. Drazic, J.L. Faria, J.L.
Figueiredo, J. Catal. 273 (2010) 191–198.
[41] C.G. Silva, M.J. Sampaio, S.A.C. Carabineiro, J.W.L. Oliveira, D.L. Baptista, R.
Bacsa, B.F. Machado, P. Serp, J.L. Figueiredo, A.M.T. Silva, J.L. Faria, J. Catal. 316
(2014) 182–190.
[42] T. Wolkenstein, Adv. Catal. 12 (1960) 189–264.
[43] H. Teterycz, R. Klimkiewicz, M. Łaniecki, Appl. Catal. A 274 (2004) 49–60.
[44] H. Teterycz, R. Klimkiewicz, M. Łaniecki, Appl. Catal. A 249 (2003) 313–326.
[45] L. Schmidt-Mende, J.L. Mac Manus-Driscoll, Mater. Today 10 (2007) 40–48.
[46] M.D. McCluskey, S.J. Jokela, J. Appl. Phys. 106 071101 (2009) 1–13.
[47] Ü. Özgür, Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Dogˇan, V. Avrutin, S.J.
Cho, H.;¸ Morko, J. Appl. Phys. 98 (2005) 1–103.
[48] A. Kołodziejczak-Radzimska, T. Jesionowski, Materials 7 (2014) 2833–2881.
[49] B.P.J. de Lacy Costello, R.J. Ewen, P.R.H. Jones, N.M. Ratcliffe, R.K.M. Wat, Sens.
Actuators B 61 (1999) 199–207.
[50] J. Han, Y. Liu, N. Singhal, L. Wang, W. Gao, Biochem. Eng. J. 213 (2012)
150–162.
[51] K.I. Hagemark, J. Solid State Chem. 16 (1976) 293–299.
[52] M.H. Sukkar, H.L. Tuller, Adv. Ceram. 7 (1984) 71–90.
[53] B. Lin, Z. Fu, Y. Jia, Appl. Phys. Lett. 79 (2001) 943–945.
[54] M.A. Barteau, R.J. Madix, J. Chem. Phys. 74 (1981) 4144–4149.
[55] N.D.S. Canning, D.A. Outka, R.J. Madix, Surf. Sci. 141 (1984) 240–254.
[56] A. Bielanski, J. Haber, Catal. Rev. Sci. Eng. 19 (1979) 1–41.
[57] G. Heiland, D. Kohl, Chem. Sens. Technol. 1 (1988) 15–38.
[58] M. Kobayashi, M. Flytzani-Stephanopoulos, Ind. Eng. Chem. Res. 41 (2002)
3115–3123.
[59] R. Meyer, C. Lemire, Sh K. Shaikhutdinov, H.J. Freund, Gold Bull. 37 (2004)
72–124.
[60] T.V.W. Janssens, B.S. Clausen, B. Hvolbæk, H. Falsig, C.H. Christensen, T.
Bligaard, J.K. Nørskov, Top. Catal. 44 (2007) 15–26.
[61] D.A. Outka, R.J. Madix, Surf. Sci. 179 (1987) 351–360.
[62] J.M. Gottfried, N. Elghobashi, S.L.M. Schroeder, K. Christmann, Surf. Sci. 523
(2003) 89–102.
[63] L. Sun, D. Zhao, Z. Song, C. Shan, Z. Zhang, B. Li, D. Shen, J. Colloid Interface Sci.
363 (2011) 175–181.
[64] F. Flores, R. Miranda, Adv. Mater. 6 (1994) 540–548.
[65] C. Li, L. Li, Z. Du, H. Yu, Y. Xiang, Y. Li, Y. Cai, T. Wang, Nanotechnology 19
(2008) 1–4.
[66] W.H. Hirschwald, Acc. Chem. Res. 18 (1985) 228–234.
[67] L.F. Gutiérrez, S. Hamoudi, K. Belkacemi, Catalysts 1 (2011) 97–154.
[68] J.C. Frost, Nature 334 (1988) 577–580.
[1] S. Mansouri, R. Bourguiga, F. Yakuphanoglu, Curr. Appl. Phys. 12 (2012)
1619–1623.
[2] K.D. Gunaratne, C. Berkdemir, C.L. Harmon, A.W. Castelman Jr., J. Phys. Chem.
A 116 (2012) 12429–12437.
[3] M. Purica, E. Budianu, E. Rusu, Thin Solid Films 383 (2001) 284–286.
[4] T. Aoki, Y. Hatanaka, D.C. Look, Appl. Phys. Lett. 76 (2000) 3257–3258.
[5] T. Seiyama, A. Kato, K. Fujiishi, M. Nagatani, Anal. Chem. 34 (1962) 1502–1507.
[6] S. Roy, S. Basu, Bull. Mater. Sci. 25 (2002) 513–515.
[7] K.V. Gurav, P.R. Deshmukh, C.D. Lokhande, Sens. Actuators B 151 (2011)
365–369.
[8] L. Wang, Y. Kang, X. Liu, S. Zhang, W. Huang, S. Wang, Sens. Actuators B 162
(2012) 237–243.
[9] N.L. Hung, H. Kim, S.K. Hong, D. Kim, Sens. Actuators B 151 (2010) 127–132.
[10] J. Guo, J. Zhang, M. Zhu, D. Ju, H. Xu, B. Cao, Sens. Actuators B 199 (2014)
339–345.
[11] P. Rai, Y.S. Kim, H.M. Song, M.K. Song, Y.T. Yu, Sens. Actuators B 165 (2012)
133–142.
[12] G.C. Bond, P.A. Sermon, G. Webb, D.A. Buchanan, P.B. Wells, J. Chem. Soc.
Chem. Commun. (1973) 444–445.
[13] P.A. Sermon, G.C. Bond, P.B. Wells, J. Chem. Soc. Faraday Trans. 75 (1979)
385–394.
[14] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 16 (1987) 405–408.
[15] M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 115 (1989) 301–309.
[16] B. Nkosi, N.J. Coville, G.J. Hutchings, Appl. Catal. 43 (1988) 33–39.
[17] G.C. Bond, D.T. Thompson, Cat. Rev. Sci. Eng. 41 (1999) 319–388.
[18] G.J. Hutchings, Catal. Today 100 (2005) 55–61.
[19] G.J. Hutchings, M. Haruta, Appl. Catal. A 291 (2005) 2–5.
[20] O. Martin, C. Mondelli, D. Curulla-Ferré, C. Drouilly, R. Hauert, J.
Pérez-Ramírez, ACS Catal. 5 (2015) 5607–5616.
[21] D. Kohl, J. Phys. D: Appl. Phys. 34 (2001) 125–149.
[22] W.A. Lazier, H. Adkins, J. Am. Chem. Soc. 47 (1925) 1719–1722.
[23] S.M. Hosseini-Sarvari, Curr. Org. Synth. 10 (2013) 697–723.
[24] S.M. Chaudhari, A.S. Waghulde, V. Samuel, M.L. Bari, V.R. Chumbhale, Res. J.
Chem. Sci. 3 (2013) 38–44.
[25] G. Djéga-Mariadassou, L. Davignon, A.R. Marques, J. Chem. Soc. Faraday Trans.
78 (1982) 2447–2454.
[26] H.R. Mardania, M. Forouzania, M. Ziaria, A. Malekzadehb, P. Biparvac, Chem.
Commun. 3 (2015) 199–207.
[27] L. Saad, M. Riad, J. Serb. Chem. Soc. 73 (2008) 997–1009.
[28] M. Glin´ ski, G. Zalewski, E. Burno, A. Jerzak, Appl. Catal. A 470 (2014) 278–284.
[29] V.K. Raizada, V.S. Tripathi, D. Lal, G.S. Singh, C.D. Dwivedi, A.K. Sen, J. Chem.
Technol. Biotechnol. 56 (1993) 265–270.
[30] M.J.L. Gines, E. Iglesia, J. Catal. 176 (1998) 155–172.
[31] W.L. Chen, H.C. Liu, Chin. Sci. Bull. 60 (2015) 1–12.
[32] B. Xu, J. Haubrich, C.G. Freyschlag, R.J. Madix, C.M. Friend, Chem. Sci. 1 (2010)
310–314.
[33] I. Fatimah, Proceeding of International Conference On Research,
Implementation And Education Of Mathematics And Sciences 2014,
Yogyakarta State University, 18–20 May 2014, 79–86.
[34] L. Vayssieres, Adv. Mater. 15 (2003) 464–466.
[35] Q.C. Li, V. Kumar, Y. Li, H.T. Zhang, T.J. Mark, R.P.H. Chang, Chem. Mater. 17
(2005) 1001–1006.
[69] G. Ghiotti, F. Boccuzzi, A. Chiorino, J. Chem. Soc. Chem. Commun. (1985)
1012–1014.
[70] M. Manzoli, A. Chiorino, F. Boccuzzi, Appl. Catal. B 57 (2004) 201–209.
[71] F. Boccuzzi, A. Chiorino, S. Tsubota, M. Haruta, Sens. Actuators B 24-25 (1995)
540–543.