The comparison of the photocatalytic performance shown by TiO2 and TiO2/WO3 composites—a parametric and kinetic study

Article abstract of DOI:10.1166/jnn.2019.15792

Kinetic and mechanistic related approaches for mostly titania were intensively studied in the literature. However, combined modelling and kinetic studies are few. Therefore, the present work focuses on modelling the dependence of the degradation kinetics of two model compounds (salicylic acid—SA and methyl orange—MO) on Evonik Aeroxide P25, hydrothermally prepared hierarchical TiO₂ and P25/WO₃ nanostars, obtained also by hydrothermal crystallization. The obtained individual semiconductors and the composites were characterized using XRD, DRS, SEM, while the photocatalytic degradation of the model pollutants were carried out varying the catalyst load, the initial pollutant concentration and incident light intensity. It was found that the degradation kinetics were independent from the hierarchical nature of the material, while significant dependencies of the degradation efficiency was found from the previously mentioned investigation parameters. All these parametric interdependences were successfully studied and a kinetic model was proposed for both bare TiO₂ and TiO₂/WO₃ composite systems.

Full text of DOI:10.1166/jnn.2019.15792

Article
Journal of
Nanoscience and Nanotechnology
Vol. 19, 356–365, 2019
Copyright © 2019 American Scientific Publishers
All rights reserved
www.aspbs.com/jnn
Printed in the United States of America
The Comparison of the Photocatalytic Performance
Shown by TiO2 and TiO2/WO3 Composites—
A Parametric and Kinetic Study
Endre-Zsolt Kedves1ꢀ2ꢀ†, István Székely1ꢀ2ꢀ†, Lucian Baia1ꢀ2, Monica Baia1ꢀ2,
Alexandra Csavdári3, and Zsolt Pap1ꢀ4ꢀ∗
1Faculty of Physics, Babes¸–Bolyai University, 400084 Cluj–Napoca, Romania
2Nanostructured Materials and Bio-Nano-Interfaces Center, Institute for Interdisciplinary Research on Bio-Nano-Sciences,
Babes¸–Bolyai University, 400271 Cluj-Napoca, Romania
3Faculty of Chemistry and Chemical Engineering, Babes¸–Bolyai University, 400028 Cluj-Napoca, Romania
4Institute of Environmental Science and Technology, University of Szeged, 6720, Szeged, Hungary
Kinetic and mechanistic related approaches for mostly titania were intensively studied in the lit-
erature. However, combined modelling and kinetic studies are few. Therefore, the present work
focuses on modelling the dependence of the degradation kinetics of two model compounds (salicylic
acid—SA and methyl orange—MO) on Evonik Aeroxide P25, hydrothermally prepared hierarchical
TiO2 and P25/WO3 nanostars, obtained also by hydrothermal crystallization. The obtained indi-
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
vidual semiconductors and the composites were characterized using XRD, DRS, SEM, while the
Copyright: American Scientific Publishers
photocatalytic degradation of the model pollutants were carried out varying the catalyst load, the
Delivered by Ingenta
initial pollutant concentration and incident light intensity. It was found that the degradation kinetics
were independent from the hierarchical nature of the material, while significant dependencies of the
degradation efficiency was found from the previously mentioned investigation parameters. All these
parametric interdependences were successfully studied and a kinetic model was proposed for both
bare TiO2 and TiO2/WO3 composite systems.
Keywords: Kinetic Modelling, Composite Materials, Salicylic Acid, Methyl Orange, Degradation.
1. INTRODUCTION
but more complex organic molecules.6 In several cases
the obtained degradation intermediate (issue valid in case
of many other advanced oxidation processes) could be
more toxic than the original molecule,7 therefore its
elimination/minimization from the degradation mechanism
is mandatory. Furthermore, not just the starting compound,
but also other facts such as the application conditions
should be considered. In classic photocatalytic experi-
ments, these are the light intensity, geometry of the reactor,
the concentration of the photocatalysts as well.8
Light intensity, is a more trivial parameter considered in
the case of photocatalytic reactions. No differences were
observed concerning the observations found in the litera-
ture, namely, that as the incident light intensity increases,
so does the degradation rate of the pollutant, which of
course at a specific light intensity value it can reach a
threshold, as recombination becomes a dominant factor.9–11
The applicability of TiO2 and in rare cases TiO2 based
composite photocatalysts achieved the everyday usage
level, being applied in mobile air and water cleaner
devices, self-cleaning paints, etc.1 This enlarged applica-
bility spectrum requires a careful study of the applicabil-
ity circumstances and post-usage impact. One of the main
concerns can be the degradation products of different pol-
lutants on the photocatalysts.
Important research was dedicated to uncovering the
mechanism of the charge generation,2 the charge
transfer3 and the degradation kinetics on TiO2 based
photocatalysts.4ꢀ5 Mostly, the degradation by-product is
not directly CO2, water and other simple compounds,
∗Author to whom correspondence should be addressed.
†These authors contributed equally to this work.
356
J. Nanosci. Nanotechnol. 2019, Vol. 19, No. 1
1533-4880/2019/19/356/010
doi:10.1166/jnn.2019.15792

Kedves et al.
The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
The catalyst loading is a more debated parameter of the
photocatalytic reactions, as for different model pollutants,
different optimal catalyst loading values were optimal.
That is why in some cases even extremely high 4 g · L−1
(case of reactive orange 412), and extremely low 0.3 g·L−1
(case of reactive red13) values were found as optimal ones.
The concentration of the pollutant is likewise a trivial, but
nevertheless very important, as photocatalytic reactions are
considered as showing mostly pseudo-first order kinetics.14
99.5% purity (Sinteza, Oradea). Ammonium metatungstate
hydrate (AMT) ((NH4ꢁ6H2W12O40 · xH2O, Sigma Aldrich,
99.99%) and hydrochloric acid (HCl, Merck, 37%, 12 M),
were used for the synthesis of WO3 micro-crystallites. In
the case of tungsten trioxide, the used reference catalyst
was commercial tungsten(VI) oxide (WO3, Sigma Aldrich,
99.9%). For the determination of the photocatalytic activ-
ity aqueous solution of methyl orange (C14H14N3NaO3S,
Sigma Aldrich) was used.
The most investigated material in the currently men-
15–18
tioned terms is TiO2
in suspension or in some cases
2.1. Synthesis of TiO2
fixed on a substrate (e.g., as films).19 In some studies
ZnO,20 while the newer generations of photocatalysts or
composite photocatalyst related kinetic studies are rare,
therefore represents a black spot in the knowledge of
photocatalysis.
When the targeted molecule is considered it was found
that several model pollutants were investigated in the terms
of their degradation kinetics, such as phenol,21ꢀ22 phe-
nol derivates,15ꢀ23 and sometimes more complex molecules
as well, such as bisphenol.24 However, photocatalytic
investigations are mostly carried out using specific dyes.
There were attempts to elucidate the kinetics of their
degradation,5ꢀ25 but all these studies considered titania as
the applied photocatalyst. Furthermore, few of them give
details about possible intermediate formation mechanism
and kinetics.
The synthesis method of this material was already
described in our previous publication,27 however a shorter
overview was provided here. In the first instance, SDS
(5.3 g) was dissolved in distilled water and mixed with
7 g of thiourea. Simultaneously the titanium dioxide
precursor TiCl4 was added drop-wise in 11 mL HClcc
(37%). After complete solubilization of the capping agents,
the mixture containing the precursor was added drop-
wise to the aqueous solution under vigorous stirring.
During the addition of the precursor a light gelation
occurred. The molar ratio applied during the synthesis was:
TiCl4:SDS:thiourea:HCl:H2O = 1:2:10:11:300. The solu-
tion was immediately transferred into a sealed, Teflon-
ꢀ
coated autoclave, and was further heated at 180 C for
24 h. The obtained white precipitate was washed several
times with distilled water until no H2S was noticed and the
organic impurities were completely removed. The product
was dried in an oven at 70 C for 12 h under airflow. The
sample will be denoted as TiO2-HT.
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
As it can be seen above there are some blank knowledge
regions of this research field, the most important one being
Copyright: American Scientific Publishers
ꢀ
Delivered by Ingenta
the scarce research data available concerning composite
type of photocatalysts. Therefore a two stage research was
proposed in the current work, which considers TiO2/WO3
composite materials and salicylic acid (a compound well-
known kinetically26) or methyl orange as model pollutants
(also a well-known compound in photocatalytic activity
testing).
To do a detailed reaction mechanism related study,
which is assumed to be feasible, a large amount of experi-
mental data is needed. To minimize somewhat the amount
of experimental work, without deteriorating the validity of
the mechanism, different modelling approaches, which are
based on kinetic and mechanistic investigations, can be
applied.
2.2. Synthesis of WO3 Nanostars
The synthesis method of this material was already
described in our previous publication,28 however a shorter
overview was provided here. 0.768 g AMT and 0.525 mL
37% HCl was dissolved in 12.5 mL of distilled water.
The solution was stirred for 15 min., then hydrothermally
ꢀ
treated at 180 C for 4 h and a green colloidal suspension
was obtained. The obtained products were collected by
centrifugation at 1600 rpm for 15 min., and were washed
with distilled water. After centrifugation, the products were
ꢀ
dried for 6 h at 70 C, and then they were annealed at
ꢀ
500 C for 30 min. and a yellowish powder was obtained.
The sample will be identified later as WO3-AMT; the
AMT abbreviation comes from the molecular formula of
ammonium metatungstate hydrate.
2. EXPERIMENTAL SECTION
The materials applied in the experiments were used as
received, without further purification commercially avail-
able TiO2 (Evonik Aeroxide P25, 89% anatase, 11%
rutile) as reference catalyst, titanium tetrachloride (TiCl4
99.9% Aldrich) as precursor, thiourea (H2NCSNH2, 98%,
Aldrich), sodium-dodecyl sulfate (SDS, CH3(CH2ꢁ11ꢁOSO3
99%, Aldrich), hydrochloric acid HCl 37% (Sigma-
Aldrich, Germany), for the degradation tests and for
the synthesis we used deionized water. The organic
compound used for photodegradation was salicylic acid,
2.3. Preparation of the WO3/TiO2 Nanocomposites
The chosen WO3 and TiO2 was used for the preparation
of the nanocomposites. In each case 24 wt.% WO3 and
76 wt.% TiO2 (Evonik Aeroxide P25) was the ratio of
the components.28ꢀ29 The nanocomposites were prepared
via mechanical mixing in an agate mortar for 3 × 5 min.
and the samples were denoted as follows: WO3 name +
TiO2 name. Evonik Aeroxide TiO2 will be referred as P25
J. Nanosci. Nanotechnol. 19, 356–365, 2019
357

The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
Kedves et al.
later, while the commercial WO3, WO3-COM. The com-
mercial tungsten trioxide was used as reference due to its
property that it wasn’t synthesized via hydrothermal treat-
ment and it does not contain shape tailored WO3 nano-
and microcrystals.
and WO3. Therefore, the characterization of the used mate-
rials was mandatory.
The XRD patterns of the used photocatalysts are shown
in Figure 1. The reference photocatalyst was Evonik
Aeroxide P25 (TiO2-P25, Danatase = 25 nm, Drutile = 50 nm),
which consisted from 89 wt.% of anatase and 11 wt.%
of rutile. TiO2 (TiO2-HT, Danatase = 14ꢄ2 nm obtained in
Section 2.1 contained just anatase. Both WO3s used (WO3-
2.4. Methods and Instrumentation
Crystalline phases and particle size were determined
by X-ray diffractometry (XRD, Shimadzu 6000) using
Cu-Kꢂ radiation (ꢃ = 1.5406 Å) equipped with a graphite
monochromator. The crystallites average size was calcu-
lated using the Scherrer equation.30
A JASCO-V650 spectrophotometer with an integration
sphere (ILV-724) was used to measure the DRS spectra
of the samples (250–800 nm). To obtain the band-gap
energy values, the well-known Kubelka-Munk approach
was applied.31
The microstructure of the samples was analyzed by the
field-emission scanning electron microscopy (SEM), (FEI
Quanta 3D FEG) operating at 20 kV. Samples for SEM
measurements were attached to a carbon adhesive pad
which was fixed to an aluminum stub.
COM, DWO -COM = 25 nm and WO3-AMT, DWO –AMT
=
3
3
60 nm) were crystalline and contained only the monoclinic
crystal phase of WO3.
The optical properties of the semiconductors were eval-
uated using DRS (Fig. 2). As it was expected, both TiO2s
band-gap energy values were in the UV range (3.12 eV
TiO2-P25 and 3.21 eV or TiO2-HT). TiO2-P25 showed a
small visible light absorption shoulder, due to its small
rutile content. Monoclinic WO3 showed an enhanced vis-
ible light absorption in case of both samples (2.61 eV
WO3-COM and 2.25 eV or WO3-AMT). However, a major
difference was observable in their DRS spectrum. The
WO3-COM sample showed a usual absorption edge in the
visible light as generally expected in many cases when
WO3 was measured. However, when WO3-AMT was ana-
lyzed, an inflection point was detected at 475 nm point-
ing out possible hierarchical structures, which may cause
this interesting behavior. The composites’ DRS spectra
was also registered, which showed the individual pecu-
liarities of the bare oxides’ spectra, while the measured
band-gap values were between the two oxides’ band
The photocatalytic tests were performed under UV irra-
diation in a photoreactor (continuous air flow, continuous
stirring. 6 × 6 W UV black-light lamps, ꢃmax = 365 nm,
thermostated at 25 ꢀC). The suspension containing the pho-
tocatalyst and the pollutant (initial concentrations applied
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
of salicylic acid (SA) and methyl orange (MO) were C0ꢀSA
Copyright: American Scientific Publishers
C0ꢀMO, the concentration of the applied photocatalysts were
Delivered by Ingenta
Ccat = 0ꢄ5, 1.0, 1.5 and 2.0 g · L−1, total volume of the
suspension Vsusp = 170 mL) was continuously purged with
air, assuring a constant dissolved oxygen concentration
during the whole experiment. Prior to the degradation
experiments the used suspension was kept in the dark for
10 min. to establish the adsorption/desorption equilibrium.
The kinetic modelling of the reaction system was carried
using Matlab R2016a.
WO3 - COM
The concentration of SA was followed by UV-Vis spec-
trophotometry with an Analytic Jena Specord 250+ spec-
trometer, where the samples’ spectra were registered in the
250–350 nm interval, while in the case of MO this range
was from 250–800 nm. Furthermore, SA was followed by
High Pressure Liquid Chromatography (HPLC), (Agilent
1100 series), using an ODS column (length 25 cm, an ODS
column is filled with a packing of octadecyldimethylsilyl
groups). An acceptable detection of the SA and its inter-
mediates was achieved within 10 minutes using a water:
methanol eluent (90:10) pumped at a rate of 0.7 cm3 ·
min−1, using three different detection wavelengths 210,
250 and 298 nm.
WO3 - AMT
TiO2 - P25
TiO2 - HT
3. RESULTS AND DISCUSSION
20
22
24
26
28
30
32
34
36
38
40
3.1. Structural and Morphological Characterization
As already mentioned, the present paper will focus on
several reaction kinetic issues directly related to TiO2
2θ (degree)
Figure 1. The XRD patterns of the used nanomaterials.
358
J. Nanosci. Nanotechnol. 19, 356–365, 2019

Kedves et al.
The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
100
90
80
70
60
50
40
30
20
3.2. The Photocatalytic Activity of the
Obtained Nanomaterials
3.2.1. The Photocatalytic Performance of
TiO2-HT and TiO2-P25
The photoactivity of these materials was tested on a model
pollutant, which degrades by direct hole oxidation and
ꢁOH attack.32 Also, it is known that SA forms surface
complexes with TiO2, which sometimes can confer a yel-
TiO2-HT
TiO2-P25
WO3-AMT
WO3-COM
TiO2-P25/WO3-COM
TiO2-P25/WO3-AMT
low color to the catalyst, while permitting the well-known
photosensitization mechanism.33 Therefore, it was interest-
ing to investigate whether secondary shape controlling can
intervene in the degradation mechanism of this model pol-
lutant, because it is known that this simple morphological
10
250
350
450
550
650
750
feature may have a serious impact on the general efficiency
of titania photocatalysts.27 The first approach was to use
different initial concentrations of SA, considering the exis-
tence of the well-known chemisorption of this compound.
As it can be seen in Figure 4 different removal effi-
ciency values were obtained, as expected. Additionally, the
degradation curves’ shape was quite different suggesting
different degradation mechanisms or different rate deter-
mining steps. Similar results can be found in the literature,
such as the case 4-nitrophenol.15 In that work it was found
that, the degradation reaction mostly occurs on the sur-
face, however other works, such the one published by Ollis
and coworkers,34 show that a smaller fraction of the pho-
Wavelength (nm)
Figure 2. DRS spectra of the used semiconductors.
gap energy (3.02 eV–TiO2-P25/WO3-AMT, 2.94 eV–TiO2-
P25/WO3-COM).
The morphology of the obtained photocatalysts was
both hierarchical as it is shown in Figure 3. The
secondary crystals were in the micrometric range for
WO3-AMT (4 ꢅm, homogenous size distribution) and
few hundreds of nanometers for TiO2-HT (750–900 nm).
The individual primary crystallite size values were cal-
culated as it was shown in the beginning of this
tocatalytic reactions occurs in the liquid phase, despite the
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
section. This structural organization may lead to enhanced
charge separation mechanism as already discussed in
our previous publication27 and may result an interesting
kinetic behavior during the degradation of different model
pollutants.
Copyright: American Scientific Publishers
strong adsorption of the model compound. This observa-
Delivered by Ingenta
tion may be important later. The variation of the photocat-
alyst concentration showed the expected results, namely,
that there is an optimum value which was ≈1 g · L−1
.
Figure 3. SEM micrographs of the obtained samples (TiO2-HT and WO3-AMT).
J. Nanosci. Nanotechnol. 19, 356–365, 2019
359

The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
Kedves et al.
0.5
0.4
0.3
0.2
0.1
0.0
0.5
0.4
0.3
0.2
0.1
0.0
SA 0.5 mM
SA 0.5 mM
0.5
0.4
0.3
0.2
0.1
0.0
0.50 mM
0.35 mM
0.25 mM
0.15 mM
36W
24W
18 W
12 W
1.25 g/L
1 g/L
0.75 g/L
0.5 g/L
TiO2
UV
intensity
0
20 40 60 80 100 120
0
20 40 60 80 100 120
Irradiation time (min)
Irradiation time (min)
0.35
0.30
0.25
0.20
0.15
0.35
0.30
0.25
0.20
0.15
SA 0.35 mM
SA 0.35 mM
0
20 40 60 80 100 120
Irradiation time (min)
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
Copyright: American Scientific Publishers
0D.10elivered by Ingenta
0.10
36 W
24 W
18 W
12 W
1.25 g/L
1 g/L
0.75 g/L
0.5 g/L
TiO2
0.05
0.05
0.00
UV
intensity
0.00
0
20 40 60 80 100 120
0
20 40 60 80 100 120
Irradiation time (min)
Irradiation time (min)
Figure 4. Degradation curves of SA on TiO2-HT, when different degradation parameters were varied (the initial concentration of SA, the applied
light intensity and suspension concentration).
Lower catalyst concentration may favor efficient illumina-
tion, but the amount of catalyst is not sufficient to achieve
high degradation rates. Suspension concentration values
higher than 1 g · L−1 are sufficient to achieve high degra-
dation yield, but caused a serious inhibition factor in illu-
mination efficiency.
As, the light intensity was varied, first it was expected
that the shape of the degradation curves will be the same,
while the slope of these curves will be in direct relation-
ship with the applied light intensity. While the second
hypothesis was partially confirmed (Fig. 4), the first one
not, pointing out an interesting phenomenon. The shape of
the degradation curves may change when the intermediate
profiles change during the degradation process. Therefore,
the following strategy was applied: the degradation curves
obtained from the spectrophotometric measurements were
fitted considering a kinetic model. If the proposed model
does not fit the chosen degradation curves, then sup-
plementary HPLC measurements should be carried out.
Indeed, as expected the model was showed several errors
when spectrophotometric data was the only data taken
in consideration. Therefore, it was mandatory to consider
hydroxylated intermediates as previously suggested.34
That is why the 3 intermediates were considered (3,5-
dihydroxybenzoic acid, another dihydroxybenzoic acid-
not identified and catechol) and the kinetic equations
detailed subsequently were applied, in accordance with the
Langmuir-Hinshelwood kinetics.
It is generally accepted that the Langmuir-Hinshelwood
equation describes the kinetics of a photocatalytic oxi-
dation system.35ꢀ36 The reaction rate Rt of the pollutant
St at any time t during the irradiation will be given by Rt
360
J. Nanosci. Nanotechnol. 19, 356–365, 2019

Kedves et al.
The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
(Eq. (1)). Also, stable intermediates are generated.
and degradation occurs in the liquid phase, not permitting
a coverage of the surface by SA molecules.
dS
dt
Kads ·St ·k
At lower light intensity values the generated ꢁOH radical
•
Rt = −
=
ꢀ
(1)
n
1 +Kads ·St + j=1ꢆKꢆIꢁ ·Itꢁj
amount is surely lower; therefore, the degradation must
occur also by an alternate route, which could be direct
oxidation by photogenerated holes.
Ij is the jth stable reaction intermediate generated dur-
ing the degradation of pollutant to CO2; k—reaction
rate constant [mol · L−1 · min−1]; Rt—initial reaction rate
[mol·L−1 ·min−1].
•
If the above situation occurs, then the surface must be
covered with SA. This could be true, as the compounds
generated by the hydroxyl radical route are accumulating
in the liquid, pointing out the lack of “space” on the cata-
lysts surface.
For TiO2-HT catalyst, the ratio of kj /k
(forma-
−j
tion/consumption reaction rate constant) was obtained
from the experimental data, considering the ratio of the
maximum intermediate concentration and the concentra-
tion of SA at that time. The consumption reaction rate
•
This also means that in the case of pollutants with good
adsorption an investigation concerning the light intensity
may give hints about reaction kinetics and intermediates’
generation.
constant k was obtained from the experimental data. The
−j
formation reaction rate constant kj did not provide relevant
data graphically, while kj was determined from the ratio
The same observations were made when Evonik
Aeroxide P25 was used, so it seems that the observed
degradation kinetics for SA is not secondary morpholog-
ical dependent. Even if Evonik Aeroxide P25 was more
efficient, the kinetic trends and observations made were
the same and will be not listed separately. Therefore, in
the next step just P25 was used together with WO3. The
chosen pollutant will be this time MO, to eliminate sur-
face sensitization and increased adsorption/chemisorption
of SA and to demonstrate the general validity of the stud-
ied degradation parameters.
of kj /k . Our predictive kinetic model (Eqs. (2–5)) based
−j
on the L-H kinetics for SA and its intermediates can be
described with the equations showed below. Using these
equations, it was possible to calculate the concentration of
SA and Ij during the photocatalytic experiment.
dS
dt
−KS ·St ·k
=
1 +ꢆKs ·St +KI ·ꢆꢇI1ꢈt +ꢇI2ꢈt +ꢇI3ꢈtꢁꢁ
(2)
dI1
dt
KI ·ꢆk1 ·St −k ·ꢇI ꢈ ꢁ
1 +ꢆK ·S +K ·ꢆꢇI ꢈ +ꢇI ꢈ +ꢇI ꢈ ꢁꢁ
−1IP:12t12.115.51.207 On: Sat, 05 Jan 2019 11:41:54
=
3.2.2. The Photocatalytic Performance of TiO -P25 and
Copyright: American Scientific Publishers
2
s
t
I
1
t
2
t
3 t
Delivered by IngenWtaO3 Based Composites and Giving Feedback to
(3)
the Bare Photocatalysts’ Shown Kinetics
dI2
dt
KI ·ꢆk2 ·St −k−2 ·ꢇI2ꢈtꢁ
WO3 acts as a charge separator in the present case as
=
(4)
1 +ꢆKs ·St +KI ·ꢆꢇI1ꢈt +ꢇI2ꢈt +ꢇI3ꢈtꢁꢁ
KI ·ꢆk3 ·St −k−3 ·ꢇI3ꢈtꢁ
these materials (WO3-AMT and WO3-COM) do not show
any adsorption or photocatalytic activity towards MO.28
Also, when the UV-Vis spectrum of MO was recorded, no
dI3
=
dt
1 +ꢆKs ·St +KI ·ꢆꢇI1ꢈt +ꢇI2ꢈt +ꢇI3ꢈtꢁꢁ
absorption maximum shifts were observed, pointing out
the fact that the degradation lead to products, which are
(5)
where, KS and KI are adsorption constants, k1ꢀ2ꢀ3 for-
mation reaction rate constants and k consumption
most likely non-aromatic small organic molecules. There-
fore, it was proposed to compare the behavior of TiO2-P25,
TiO2-HT with the composite made from TiO2-P25 and
WO3-AMT. Thus, all the obtained results will be presented
considering the results from the previous section (Section
3.2.1). It should be noted that the MO degradation capac-
ity of TiO2-P25 and WO3-COM was also investigated, but
the results were not presented, as they showed the same
trends during the parameter investigation.
In the case of MO, the first step was likewise to test
the effect of the initial concentration of the dye (Fig. 7).
The trend shown by the applied composite revealed the
same behavior as in the case of SA, namely that with the
increase of the MO concentration the achieved conversion
decreased.
−1ꢀ−2ꢀ−3
reaction rate constants. Two very important aspects should
be emphasized here. First, if the spectrophotometric data
are corrected using the HPLC measurements the degrada-
tion curve will take a new form, as shown in Figure 5.
Therefore, the degradation rate constant of SA was inde-
pendent from the applied light intensity (when lnꢆC/C0ꢁ
vs. time was represented graphically), while the main rea-
son for it was the different formation and consumption rate
constant of the obtained degradation intermediates, as it is
shown in Figure 6. This also means that these light inten-
sity values are still sufficient to promote the degradation of
SA (same degradation efficiency), but changes in the light
intensity values. This could suggest a degradation pathway
switching:
When the catalyst load was investigated an interesting
phenomenon was observed, namely that at a smaller scale,
changes in the suspension concentration did not have any
significant effect in the degradation of SA (case of P25),
•
When, the applied light intensity is high the chosen
semiconductor generates higher amount of charge carrier
ꢁ
pairs, which can produce a higher amount of OH radicals
J. Nanosci. Nanotechnol. 19, 356–365, 2019
361

The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
Kedves et al.
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
SA 0.35 mM
SA 0.35 mM
SA 0.35 mM
36 W
18 W
12 W
SA-HPLC
SA-HPLC
Intermediate 1
Intermediate 2
Intermediate 3
SA-UV/Vis
Model SA
Model I1
Model I2
Model I3
Intermediate 1
Intermediate 2
Intermediate 3
SA-UV/Vis
Model SA
Model I1
Model I2
Model I3
SA-HPLC
Intermediate 2
Intermediate 3
SA-UV/Vis
Model SA
Model I2
Model I3
0
20 40 60
80 100 120
0
20 40 60 80 100 120
0
20
40
60
80 100 120
Irradiation time (min)
Irradiation time (min)
Irradiation time (min)
Figure 5. HPLC based degradation curves versus UV/Vis spectrophotometric based curves and the degradation intermediate evolution profiles, when
the impact of the UV-A light intensity was evaluated.
that is why the investigations were carried out on a larger
catalyst load interval when the composite’s photocatalytic
performance was carried out. It was found that at low cat-
is a positive component in the composite but, at
high density catalyst suspensions, it can seriously hin-
der the light transmission within the volume of the
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
alyst load values a very low photocatalytic conversion was
suspension.
Delivered by Ingenta
Copyright: American Scientific Publishers
achieved (Fig. 8). As TiO2 can be bound quite efficiently
When the light intensity was the variable parameter
(Fig. 9), the linear dependence of the conversion with the
light intensity was found only in the case of the composite
material, due to the facts mentioned in the previous section
of the paper. This means that the degradation of MO could
occur just in the solution, because:
to the surface of WO3 as it was previously shown,29 it can-
not be excluded that larger agglomerations were formed
diminishing thus the contact surface between solid and liq-
uid entities.
As the catalyst load was raised, the previously men-
tioned effect disappeared, while further increase resulted
a direct decay of the degradation efficiency that may
be attributed to the WO3 component, which normally
•
Insignificant adsorption on the catalyst surface after the
dark step of the photodegradation process.
Methyl orange concentration (µM)
0
50
97.6
100
150
98.3
84.6
200
250
300
100
0.050 Formation
Catechol
Cathecol2
DHBA I2
DHBAI22
DHBA I3
DHBAI32
100
80
60
40
20
0
Degradation
100
0.040
0.030
0.020
0.010
74.3
80
60
40
20
0
52.3
24.3
38.4
0
100
200
300
400
500
600
0.000
36
18
12
Salicylic acid concentration (µM)
Light intensity (W)
Figure 7. Comparison of the effect of the initial concentration of the
pollutant on TiO2-P25 and TiO2-P25/WO3-AMT composite photocata-
lysts’ degradation efficiency.
Figure 6. Formation and degradation rate constants for the three differ-
ent intermediates.
362
J. Nanosci. Nanotechnol. 19, 356–365, 2019

Kedves et al.
The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
Table I. The k values obtained from the applied kinetic model where
the catalyst load was modified.
Catalyst load (g·L–1
)
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
100
80
60
40
20
0
100
80
60
40
20
0
Catalyst load
k1
k2
k3
93.5
(g·L−1
)
(min−1
)
R2
(min−1
)
R2
(min−1
)
R2
90.5
89.4
84.6
93.5
0.5
1
1.5
2
0.0021
0.0096
0.0056
0.0041
0.992
0.992
0.991
0.969
0.3863
0.8683
0.5961
0.4362
0.993
0.999
0.959
0.997
0.0952
0.2353
0.1596
0.1374
0.999
0.999
0.987
0.994
58.3
49.7
32.0
TiO2
TiO2/WO3
modelling in heterogenous photocatalysis the Langmuir-
Hinshelwood mechanism is the most used one, especially
when adsorption–desorption equilibrium can be consid-
ered. According to Langmuir-Hinshelwood approach the
reaction rate can be calculated from the equation below:
0.25
0.75
1.25
1.75
2.25
Catalyst load (g·L–1
)
Figure 8. Comparison of the effect of the catalyst load on TiO2-P25 and
TiO2-P25/WO3-AMT composite photocatalysts’ degradation efficiency.
k1 ·k2 ·ꢇCꢈ
r =
(8)
ꢉꢇCꢈꢊ+1
Hence r is the reaction rate, k1 is the adsorption appar-
ent rate constant, k2 is the apparent rate constant of the
pollutant degradation, C is the pollutant concentration.
In Tables I–III the R2 values were attributed to the
respective k values, which were calculated from plotting
the concentration versus the time using linear fitting in
each case. The slopes of the linear fits will yield the k val-
ues. In case of k1, lnC/C0 (0, 10, 20 min.); in case of
•
No shifts of the MO peak occurred during the degra-
dation, which could mean direct oxidation to low molar
weight carboxylic acids.
•
Linear dependence of the degradation efficiency with
the light intensity.
Considering the above mentioned facts, a kinetic model
was further proposed.
From the general kinetic point of view the removal of
a pollutant can be described with the following simple
k2 the concentration (30, 40, 50, 60 min.); in the case
0ꢄ75
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
of k3 the C
(60, 80, 100, 120 min.) values were plot-
equations:
Copyright: American Scientific Publishers
ted versus time. From the above statements, which were
Delivered by Ingenta
adjusted to our case, the following equation can be used:
ꢇCꢈ
dt
r = k ·ꢇMOꢈn
−
= r
(6)
(7)
ꢇMOꢈ ꢉk1 ·k2 ·k3 ·ꢇMOꢈnꢊ·ꢇMOꢈ
−
=
(9)
dt
ꢆ1 +k1ꢁ·ꢇMOꢈ
ln the equation above k1 is the adsorption apparent rate
constant, k2 is the apparent rate constant of the MO degra-
dation, k3 is the apparent rate constant of the degradation
where r is the reaction rate, k is the apparent rate con-
stant, C is the concentration of the pollutant an n is the
order of the reaction. When approaching mathematical
Table II. The k values obtained from the applied kinetic model where
the initial concentration was modified.
100
TiO2-P25
88.4
93.5
84.6
TiO2-P25/WO3-AMT
Initial concentration
(ꢅM)
k1
k2
k3
80
60
40
20
0
(min−1
)
R2
(min−1
)
R2
(min−1
)
R2
61.9
62.5
125
187.5
250
0.0162 0.984 0.6897 0.997 0.1836 0.991
0.0096 0.992 0.8683 0.999 0.2353 0.999
0.0013 0.999 0.5570 0.977 0.1780 0.998
0.0015 0.997 0.7294 0.999 0.0913 0.990
55.5
47.9
34.1
Table III. The k values obtained from the applied kinetic model where
the light intensity was modified.
14.8
UV
intensity
Light
k1
k2
k3
(min−1
intensity (W) (min−1
)
R2
(min−1
)
R2
)
R2
8
14
20
26
32
38
Light intensity (W)
12
18
24
36
0.0011
0.986
0.988
0.996
0.992
0.1487
0.3138
0.5510
0.8683
0.993
0.991
0.998
0.999
0.0426
0.0879
0.1506
0.2353
0.994
0.998
0.996
0.999
0.0031
0.0050
0.0096
Figure 9. Comparison of the effect of the light intensity on
TiO2-P25 and TiO2-P25/WO3-AMT composite photocatalysts’ degrada-
tion efficiency.
J. Nanosci. Nanotechnol. 19, 356–365, 2019
363

The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
Kedves et al.
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
Copyright: American Scientific Publishers
Delivered by Ingenta
Figure 10. The comparison of the model and the experimental data.
of the MO intermediates and n is the order of the reac-
tion, in our case, between 0 and 1. This equation is used
when k1 and k2 have first order kinetics (the k values were
summarized in Table I).
Considering the data collected we have concluded, that
in our case k1 and k2 show a pseudo-first order kinetics
followed by a fractional order k3. We have implemented
this in our mathematical model, hence we obtained the
equation below:
the MO degradation with WO3–TiO2 composite photocata-
lysts. The scripts for the simulations were written and val-
idated in Matlab 2016a, and ordinary differential equation
(ode45) was used to solve these equations and the chosen
model fitted acceptably the experimental data (Fig. 10).
4. CONCLUSIONS
The present work reemphasized the fact that a degrada-
tion process can be influenced severely, by just changing
some of the reaction parameters, including catalyst con-
centration, initial concentration of the pollutants and light
intensity. Not in all the cases the expected dependencies
were obtained. It was found that in the case of the catalyst
concentration and initial pollutant concentration the usual
observations were made, while in the case of the light
intensity the formed intermediates intervene in the degra-
dation product kinetics. In each of the cases a specialized
model was elaborated and applied which fitted the experi-
mental data accordingly. However, it should be noted that
by introducing the WO3 in the system resulted the same
ꢇMOꢈ ꢉꢆk1/2ꢄ303ꢁ·k2 ·k3 ·ꢇMOꢈnꢊ·ꢇMOꢈ
−
=
(10)
dt
1 +ꢆk1/2ꢄ303ꢁ·ꢇMOꢈ
Where [MO] is the dye concentration, k1 is the adsorp-
tion apparent rate constant, k2 is the apparent rate constant
of the MO degradation, 2.303 is a divisor to convert the nat-
ural logarithm to a base 10 log ꢇlog10ꢆxꢁ = lnꢆxꢁ/2ꢄ303ꢈ,
k3 is the apparent rate constant of the degradation of the
MO intermediates and n is a fractional order, between in
this case the attributed reaction order for the MO interme-
diates is 0.75. The equation above was used for modeling
364
J. Nanosci. Nanotechnol. 19, 356–365, 2019

Kedves et al.
The Comparison of the Photocatalytic Performance shown by TiO2 and TiO2/WO3 Composites
behavior but with different impact intensities of the cho-
sen parameters. Furthermore, the hierarchical nature of the
oxides’ crystals did not show any impact on the observed
kinetics.
18. H. Zhang, D. Liu, S. Ren, and H. Zhang, Res. Chem. Intermed.
43, 1529 (2016).
19. T. A. McMurray, J. A. Byrne, P. S. M. Dunlop, J. G. M. Winkelman,
B. R. Eggins, and E. T. McAdams, Appl. Catal., A 262, 105 (2004).
20. O. Bechambi, L. Jlaiel, W. Najjar, and S. Sayadi, Mat. Chem. Phys.
173, 95 (2016).
Acknowledgments: The present work was financially
supported by the SH 7/2/20 project.
21. V. Auguliaro, E. Davi, L. Palmisano, M. Schiavello, and A. Sclafani,
Appl. Catal. 65, 101 (1990).
22. P. Górska, A. Zaleska, and J. Hupka, Sep. Purif. Technol. 68, 90
(2009).
References and Notes
1. A. Ibhadon and P. Fitzpatrick, Catalysts 3, 189 (2013).
2. P. Zhang, T. Wang, X. Chang, and J. Gong, Acc. Chem. Res. 49, 911
(2016).
23. J. C. Doliveira, C. Minero, E. Pelizzetti, and P. Pichat, J. Photochem.
Photobiol., A 72, 261 (1993).
24. N. Lu, Y. Lu, F. Liu, K. Zhao, X. Yuan, Y. Zhao, Y. Li, H. Qin, and
J. Zhu, Chemosphere 91, 1266 (2013).
25. F. L. Zhang, J. C. Zhao, T. Shen, H. Hidaka, E. Pelizzetti, and
N. Serpone, Appl. Catal., B 15, 147 (1998).
26. M. M. Ibrahim, S. A. Ahmed, K. S. Khairou, and M. Mokhtar, Appl.
Catal., A 475, 90 (2014).
3. K. Osako, K. Matsuzaki, H. Hosono, G. Yin, D. Atarashi, E. Sakai,
T. Susaki, and M. Miyauchi, APL Materials 3, 104409 (2015).
4. J. Yang, C. Chen, H. Ji, W. Ma, and J. Zhao, J. Phys. Chem. B
109, 21900 (2005).
27. K. Vajda, K. Saszet, E. Zs. Kedves, Zs. Kása, V. Danciu, L. Baia,
K. Magyari, K. Hernádi, G. Kovács, and Zs. Pap, Ceram. Int.
42, 3077 (2016).
28. I. Székely, G. Kovács, L. Baia, V. Danciu, and Zs. Pap, Materials
9, 258 (2016).
29. L. Baia, E. Orbán, S. Fodor, B. Hampel, E. Zs. Kedves, K. Saszet,
I. Székely, É. Karácsonyi, B. Réti, P. Berki, A. Vulpoi, K. Magyari,
A. Csavdári, Cs. Bolla, V. Cos¸oveanu, K. Hernádi, M. Baia,
A. Dombi, V. Danciu, G. Kovács, and Zs. Pap, Mater. Sci. Semicond.
Process 42, 66 (2016).
5. A. Ajmal, I. Majeed, R. N. Malik, H. Idriss, and M. A. Nadeem,
RSC Adv. 4, 37003 (2014).
6. E. Grabowska, J. Reszczynska, and A. Zaleska, Water Res. 46, 5453
(2012).
7. E. Illes, E. Szabo, E. Takacs, L. Wojnarovits, A. Dombi, and
K. Gajda-Schrantz, Sci. Total Environ. 472, 178 (2014).
8. S. Merabet, A. Bouzaza, and D. Wolbert, J. Hazard. Mater.
166, 1244 (2009).
9. C. Karunakaran and S. Senthilvelan, Catal. Comm. 6, 159 (2005).
10. M. Qamar, M. Muneer, and D. Bahnemann, J. Environ. Manage.
80, 99 (2006).
11. J. F. Montoya, J. A. Velásquez, and P. Salvador, Appl. Catal., B
88, 50 (2009).
12. M. Muruganandham and M. Swaminathan, Sol. Energy Mater. Sol.
Cells 81, 439 (2004).
13. S. Kaur and V. Singh, J. Hazard. Mater. 141, 230 (2007).
14. C. Minero, C. Aliberti, E. Pelizzetti, R. Terzian, and N. Serpone,
Langmuir 7, 928 (1991).
30. R. Jenkins and R. L. Snyder, Introduction to X-ray Diffractometry,
First edn., Wiley, New York (1996), p. 90.
31. P. Kubelka and F. Munk, Zeitschrift Für Technische Physik 12, 601
(1931).
32. A. Mills, C. E. Holland, R. H. Davies, and D. Worsley, J. Photoch.
IP: 212.115.51.207 On: Sat, 05 Jan 2019 11:41:54
Photobio. A 83, 257 (1994).
Copyright: American Scientific Publishers
33. M. A. Morosanova and E. I. Morosanova, Chem. Central J. 9, 64
Delivered by Ingenta
(2015).
34. C. Turchi and D. F. Ollis, J. Catal. 122, 192 (1990).
35. M. A. Behnajady, N. Modirshahla, and R. Hamzavi, J. Hazard.
Mater. 133, 226 (2006).
15. D. W. Chen and A. K. Ray, Water Res. 32, 3223 (1998).
16. R. Fretwell and P. Douglas, J. Photochem. Photobiol., A 143, 229
(2001).
36. A. N. Rao, B. Sivasankar, and V. Sadasivam, J. Hazard. Mater.
166, 1357 (2009).
17. G. Sivalingam, M. H. Priya, and G. Madras, Appl. Catal., B 51, 67
(2004).
Received: 31 October 2017. Accepted: 19 January 2018.
J. Nanosci. Nanotechnol. 19, 356–365, 2019
365