Effects of Different Copper Loadings on the Photocatalytic Activity of TiO2-SiO2 Prepared at a Low Temperature for the Oxidation of Organic Pollutants in Water

Article abstract of DOI:10.1002/cctc.201800249

The objective of this research is to examine how Cu modification can improve the photocatalytic activity of TiO₂-SiO₂, to explain the correlation between the Cu concentration and the chemical state of Cu cations in the TiO₂

Full text of DOI:10.1002/cctc.201800249

Accepted Article
Title: Effects of different Cu loadings on photocatalytic activity of TiO2-
SiO2 prepared at low temperature oxidation of organic pollutants
in water
Authors: Tihana Cizmar, Urska Lavrencic Stangar, Mattia Fanetti, and
Iztok Arcon
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201800249
Link to VoR: http://dx.doi.org/10.1002/cctc.201800249
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Effects of different Cu loadings on photocatalytic activity of TiO2-
SiO prepared at low temperature for oxidation of organic
2
pollutants in water
[
a]
[a,b]
[a]
[a,c]
Tihana Čižmar , Urška Lavrenčič Štangar , Mattia Fanetti and Iztok Arčon*
[
1-3]
Abstract: The objective of this research is to examine how copper
modification can improve the photocatalytic activity of TiO -SiO , and
to explain the correlation between Cu concentration and chemical
state of Cu cations in the TiO -SiO matrix, and the photocatalytic
activity under the UV/solar irradiation. The Cu-modified TiO -SiO
cleaning glass, etc. . Photocatalytic activity only appears
under UV light irradiation since TiO cannot absorb visible part of
sunlight due to its wide band gap (3.0 - 3.2 eV). Addition of SiO
to TiO improves mechanical strength and thermal stability of the
2
2
2
2
2
2
2
2
2
photocatalyst, increases the surface area and pore volume and,
in some cases, inhibits the charge carrier surface recombination
and the crystallization of anatase phase, and therewith,
photocatalysts were prepared by a low temperature sol-gel method
based on organic copper, silicon and titanium precursors with varied
Cu concentrations (from 0.05 to 3 mol%). The sol-gels were dried at
[
4]
[5]
increases photocatalytic activity of pure TiO
2
.
Dong et al.
1
50 °C to obtain the photocatalysts in the powder form. The
studied the photocatalytic degradation of Rhodamine-B dye
under UV-light by nano-ordered two-dimensional hexagonal
photocatalytic activity was determined by a fluorescence-based
method of terephthalic acid decomposition. Up to three times
mesoporous TiO
synthesized by organic-solvent evaporation method, and found
that TiO -SiO composite materials have better photocatalytic
activity than pure TiO (Degussa P25) because of the synergetic
roles of coupled adsorption and photocatalytic oxidation.
2 2
-SiO composite materials that have been
2 2
increase in photocatalytic activity is obtained when TiO -SiO matrix
is modified with Cu in a narrow concentration range from 0.05 to 0.1
mol%. At higher Cu loadings the photocatalytic activity of Cu
modified photocatalyst is smaller than in the unmodified reference
2
2
2
[
6]
TiO
TiO
2
-SiO
-SiO
2
photocatalyst. XRD analysis shows that all Cu modified
composites with different Cu concentrations have the
Anderson and Bard studied the photocatalytic degradation of
Rhodamine-6G (R-6G) under UV-light and described the effect
2
2
same crystalline structure as unmodified TiO
addition of Cu does not change the relative ratio between anatase
and brookite phase or unit cell parameters of the two TiO crystalline
2
-SiO
2
composites. The
of incorporation of SiO
photocatalyst prepared by a sol-gel method from organometallic
precursors. TiO -SiO catalysts were heated to 200 °C for 12 h.
They demonstrated that a ratio TiO /SiO of 30/70 produces a
catalyst about 3 times more active than Degussa P25 TiO
Higher photocatalytic activity relates to the preferential
adsorption of R-6G on SiO which increases the surface
concentration of R-6G at or near the TiO sites promoting more
efficient oxidation by photogenerated species. They suggest that
the photogenerated intermediate oxidants on the TiO sites
(hydroxyl radical or perhydroxyl radical) must diffuse and react
with R-6G on the SiO sites.
2 2
on the behaviour of a TiO -based
2
2
2
structures. The Cu K-edge XANES and EXAFS analysis is used to
determine valence state and local structure of Cu cations in Cu-
2
2
2
.
modified TiO
2
-SiO
2
photocatalyst. The results elucidate the
mechanism responsible for the improved photocatalytic activity. In
the samples with low Cu content, which exhibit largest activity, Cu-
O-Ti connections are formed, suggesting that the activity
enhancement is due to Cu(II) cations attachment on the surface of
2
2
2
the photocatalytically active TiO
may act as free electron traps, reducing the intensity of
recombination between electrons and holes at the TiO
2
nanoparticles, so Cu(II) cations
2
2
To improve the photocatalytic activity under solar light
irradiation, and reduce high rate of recombination between
photocatalyst’s surface. At higher Cu loadings no additional Cu-O-Ti
connections are formed, instead only Cu-O-Cu connections are
established, indicating the formation of amorphous or nanocrystalline
2
photogenerated electrons and holes in TiO , metal doping was
[
7]
introduced. Improving photocatalytic activity can be achieved if
more photo-generated electrons and holes move to the surface
of the semiconductor particles before they recombine in bulk.
2
Cu oxide, which hinders the photocatalytic activity of TiO .
Doping of TiO
2
with transition metals can introduce electron
decrease in electron/hole
The photocatalytic activity of TiO and
2
photocatalysts is improved if Pt, Pd, Au or Ag are
or if Pt, Pd, Au or Ag are deposited on
but these metals are rare and expensive. Copper is a
capture centres resulting in
a
Introduction
[8,9]
recombination centres.
TiO -SiO
used as dopants
2
2
Titanium dioxide is one of the most popular photoactive
materials and it has been widely used for water purification,
degradation of air pollutants, removal of residual pesticides, self-
[10-13]
[14]
2
TiO
good candidate to replace expensive metals as dopants.
Improved photocatalytic activity was found for Cu-modification
under different reducing calcination atmospheres at low Cu
concentrations (0.05, 0.1, 0.2, 0.5 and 1 mol% with respect to
[
[
a]
b]
University of Nova Gorica
Vipavska cesta 13, Nova Gorica, 5000 (Slovenia)
University of Ljubljana, Faculty of Chemistry and Chemical
Technology,
[15]
Ti). The enhancement was attributed to Cu-ion participating in
lengthening of electron hole pair recombination time, and due to
creation of oxygen vacancy which helps to reduce the band
Večna pot 113, Ljubljana, 1000 (Slovenia)
Jožef Stefan Institute,
[
15]
[16]
[c]
gap. Wu et al. showed that Cu nanoparticles prepared by
Jamova 39, Ljubljana, 1000 (Slovenia)
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
wet impregnation, followed by low-temperature H
2
reduction,
relative to pure TiO
2
. In their opinion, one possible mechanism
could improve the catalytic activity of TiO
catalytic activity of Cu−TiO was about 10 times higher than that
of TiO at the optimum loading amount 1.5 mol% Cu. M. Tahir
and B. Tahir prepared Cu-modified TiO
modified sol-gel method and showed that CO production was
higher than that of pure TiO by more than 99 times when Cu
concentration is 3.5 mol%. They also co-doped TiO with Cu and
In and showed that performance of 1.0% Cu and 3.5% In co-
doped TiO catalyst for CO production is 113 times higher than
un-doped TiO The enhanced conversion efficiency was
attributed to improved charge separation in TiO , where Cu and
In species served as electron traps which suppressed electron–
2
under UV light. The
for the improved photocatalytic activity is that the
surface/interfacial Cu(II) sites function as unique adsorption sites
2
2
for methylene blue and facilitate the subsequent photocatalytic
[
17]
[31]
2
nanocatalysts using
degradation. Xin et al.
2
prepared Cu-TiO photocatalyst with
different Cu dopant content (0.02, 0.04, 0.06, 0.08, 0.10, 0.15,
0.20, 0.40, 0.60, 0.80, 1 and 3 mol%) by sol–gel method and
showed that 0.06 mol% of Cu increased photocatalytic activity of
2
2
Cu-doped TiO
ascribed to appropriate amount of oxygen vacancies and doped
Cu(II) ions on the surface layer of TiO , which can effectively
capture the photoinduced electrons. Hussein et al.
2
. The increase of photocatalytic activity is
2
2
.
2
[
32]
2
used
simulations to confirm experimental results about the reduction
of Cu(II) to Cu(I) and how it may serve as a further mechanism
[
18]
[19]
hole recombination.
Momeni et al.
synthesized highly
[
30]
aligned TiO nanotube films co-sensitized with Cu and W
2
through which the lifetime of surface holes may be extended.
Arana et al.
[
33]
nanoparticles via chemical bath deposition method. Optimal
performance was obtained with photocatalysts containing 0.04
mol% Cu and 0.06 mol% W. The enhanced photocatalytic
activity was associated with the extended absorption in the
visible light region and effective separation of photo-generated
carriers by the co-sensitization of Cu and W nanoparticles.
2 2
prepared Cu-TiO by TiO impregnation with
CuSO aqueous solutions (incipient wetness impregnation
4
method) at 25 °C with Cu concentration 0.6 mol%. Catalysts
were calcined at 500 °C for 5 h. They suggested that the
presence of Cu(II) atoms on the TiO surface generate the
2
catalysts active centres and increase the photocatalytic activity
because photo-generated oxidising species do not migrate far
away from active centres where they are formed. Miyauchi et
[
20]
Zhang et al. synthesized Cu-doped titanium oxide films by dc
magnetron sputtering method using Ti and Cu mixed target. The
[
34]
[35]
Cu-doped TiO
2
films had different photocatalytic behaviour in
al.
interfacial charge transfer i.e., electrons in the valence band
(VB) of TiO are directly transferred to Cu(II) and form Cu(I), as
well as holes in the VB of TiO . The holes produced in the VB of
TiO efficiently oxidize organic contaminants. In contrast, the
Cu(I) produced by electron transfer is likely to reduce adsorbed
through multielectron reduction, and thus electrons are
consumed.
Jin et al.
molecular scale CuO clusters on the surfaces of TiO
(Degussa P25) in highly dispersed state
and Irie et al.
demonstrated that visible light initiates
accordance with the amount of doped copper. The optimum Cu
doping concentration was 1.45 mol%, which showed the most
effective photocatalytic activity. They suggested mechanism
responsible for photocatalytic activity enhancement: the doped
copper cation can provide a shallow trap for photo-generated
electron and hole so as to inhibit the recombination and extend
the lifetime of charge carrier. This optimum concentration can be
explained by the balance of an increase in trapping sites leading
to efficient trapping, and fewer trapped carriers leading to longer
2
2
2
O
2
[
36]
used Cu(acac)
2
as a precursor to form
particles
by
2
[
7]
lifetimes for interfacial charge transfer. In addition, the metal
ion doping of TiO has been used to induce the photoactivity in
the visible light region.
Improved photocatalytic activity of TiO
case of deposition of Cu O nanoclusters onto TiO
an impregnation technique at Cu concentrations from 0.001 to
a
2
chemisorption−calcination cycle technique and showed that the
photocatalytic activity strongly depends on Cu loading. They
explained that the effective mixing between the surface CuO
[
14,21-25]
2
was found also in
surfaces by
x
2
cluster levels and the O2p states from the surface of TiO
a surface d sub-band that overlaps with the TiO VB. The visible-
light absorption of CuO-TiO can mainly be attributed to the
interfacial electron transfer from the sub-band to the
conduction band (CB) of TiO
2
yields
2
[
26]
0
.02 mol%. Under UV/Vis light irradiation, the electrons in the
valence band of TiO are promoted to Cu(II) species in Cu
clusters, which results in the transformation of Cu(II) into Cu(I),
2
2
x
O
d
2
.
[
26-28]
[29]
[31]
while holes remain in the valence band.
that Cu-modified TiO -SiO nanocomposites (with the molar ratio
of Cu/Ti 0.01:1) prepared by a sol-gel method, calcined at
00 °C for 3h, exhibited higher photocatalytic activity than
commercial TiO (Degussa P25) for the degradation of
Li et al. showed
Xin et al. reported that at heavy Cu doping concentration,
p-type Cu O can cover the surface of TiO , which leads to
decrease in the photocatalytic activity of photocatalyst. Li et
2
2
2
2
[
30]
5
al.
explained that at higher Cu concentration (>0.5 mol%)
led to decreased photocatalytic
2
coupling between CuO and TiO
2
Rhodamine-B dye under both UV and visible light irradiation.
2
efficiency due to the charge transfer from TiO to CuO and
They explained that addition of Cu enhanced the water
subsequent charge recombination.
-
[37]
absorption on TiO
2
surface, more OH was formed and
In our previous research , we have developed a new low-
consequentially photocatalytic activity was increased. Cu
species with divalent copper cations states, dispersed on the
temperature sol-gel synthesis of Cu-modified TiO
photocatalysts, based on organic precursors. Colloidal solution
of TiO -SiO nanoparticles was prepared from titanium
tetraisopropoxide, tetraethyl ortosilicate, and copper
acetylacetonate as a source of Ti, Si and Cu, respectively. We
examined how copper modifications at two copper
concentrations (0.1 mol% and 3 mol%) processed at different
temperatures (drying at 150 °C and additional calcining at
2 2
-SiO
surface of TiO
2
nanoparticles, can effectively inhibit
2
2
[
30]
recombination of photoinduced charge carriers. Li et al.
and
Xin et al. concluded that metal to metal charge transfer in Ti-
O-Cu complexes at TiO surfaces enhance photoactivity through
charge carrier separation. Li et al.
photoactive CuO-TiO nanocomposites with different Cu
concentration (0.1, 0.6 and 5 mol% Cu) via a deposition
precipitation method. CuO-TiO nanocomposite with a very low
copper loading (0.1 mol%) demonstrated improved photoactivity
[
31]
2
[
30]
synthesized series of
2
2 2
500 °C) affect the photocatalytic activity of TiO -SiO under the
UV/solar irradiation. The enhancement of photocatalytic activity
was obtained only in the photocatalyst modified with low Cu
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
concentration (0.1 mol%), processed at low temperature.
further proportionally decreases with the increase of Cu
concentration.
2 2
Calcination of Cu-modified TiO –SiO photocatalysts at 500 °C
induces significant structural changes and reduces its
photocatalytic activity.
Structural properties
In this study we investigate the relation between Cu
loading and photocatalytic activity of Cu-modified TiO
2
-SiO
2
The BET surface areas of unmodified and Cu-modified TiO
2
-
photocatalyst, synthesized by low-temperature sol-gel synthesis,
in order to optimize synthesis conditions and obtain the
photocatalyst with highest photocatalytic activity under the
UV/solar irradiation. The Cu concentrations are varied in the
range from 0.05 to 3 mol%, at eight selected values: 0.05, 0.08,
SiO samples as determined by nitrogen adsorption and
desorption analysis are given in Table 1. All Cu-modified
samples exhibit same or very similar specific surface area as
2
unmodified TiO
value is found for the two Cu modified TiO
2
-SiO
2
nanocomposite. About 10% increased
-SiO
2
2
0
.1, 0.15, 0.3, 0.5, 1 and 3 mol%. The photocatalytic activity was
nanocomposites, TS_0.08Cu and TS_0.1Cu, which are the most
photocatalytically active. However, the increased photocatalytic
activity in these two samples compared to the unmodified one
cannot be attributed only to this small increase in their specific
tested by oxidative degradation of terephthalic acid (TPA) in
aqueous solution. Increased photocatalytic activity was found
only at low Cu loadings in a narrow range between 0.05 mol% to
0
.1 mol%. To clarify mechanisms responsible for the
2 2
surface area. Average pore size in Cu modified TiO -SiO
photocatalytic activity enhancement at low Cu concentrations,
and its decrease at high Cu loadings, different analytical and
spectroscopic techniques were used. Crystal structures present
nanocomposite is the same (within error-bars) as in the
unmodified one, at all Cu loadings.
The XRD patterns of unmodified and Cu-modified TiO
2
-
in the unmodified and Cu-modified TiO
were investigated by X-ray diffraction. Specific surface area and
average pore size was evaluated from N sorption. Local
structure and valence state of Cu cations in TiO -SiO
photocatalyst was determined by X-ray absorption spectroscopy
2
-SiO
2
photocatalysts
SiO samples are shown in Figure 2. Anatase is the major
2
crystal phase in all samples, with a minor amount of brookite
phase. The broad peak that appears at 2θ around 22° can be
attributed to glass-like amorphous silicate nanoparticles.
2
2
2
(
(
XAS) methods: extended X-ray absorption fine structure
EXAFS) and X-ray absorption near edge structure (XANES).
Correlations between the Cu cation local structure in TiO
matrix and photocatalytic activity of the Cu-modified TiO
nanocomposites at different Cu loadings elucidate competitive
mechanisms responsible for improvement or reduction of
photocatalytic activity of the material.
2
-SiO
-SiO
2
TS_150
2
2
2.5
TS_0.05Cu
TS_0.08Cu
TS_0.1Cu
TS_0.15Cu
TS_0.3Cu
TS_0.5Cu
TS_1Cu
2.0
1.5
1.0
0.5
TS_3Cu
Results and Discussion
Photocatalytic activity
The photocatalytic activity of the samples was measured by the
fluorescence-based method of terephthalic acid (TPA)
decomposition in aqueous solution, taking into account that 2-
hydroxyterephthalic acid (HTPA) is formed as an intermediate
0.0
10
20
30
40
50
time / min
product. The formation rate constant of HTPA (k
compare photocatalytic activity of the
1
) is used to
Cu-modified
photocatalysts with an unmodified one. Modelling of the
experimental data of HTPA concentrations as function of UV/Vis
illumination time (Figure 1) was performed according to a
simplified kinetics analysis for the initial degradation curve,
where oxidation of TPA to HTPA under UV/solar irradiation can
Figure 1. HTPA concentrations as function of UV/Vis illumination time for
unmodified and Cu-modified TiO -SiO powder catalysts with different Cu
concentrations (from 0.05 to 3 mol%). (Dots – experiment, solid line – best fit
model ([HTPA] = k · t))
2
2
1
[
37,38]
be described by zero-order kinetics
fit values of the rate constants k , obtained for unmodified and
Cu-modified TiO -SiO catalysts with different Cu loadings and
relative photocatalytic activities of Cu-modified compared to
unmodified TiO -SiO photocatalyst are presented in Table 1.
The results suggest that there is up to three times increase
in photocatalytic activity when TiO -SiO matrix is modified with
(
[HTPA] = k
1
· t)
).
Best
1
2
2
2
2
2
2
low Cu concentrations in a narrow range from 0.05 to 0.1 mol%.
In this range photocatalytic activity increases with Cu
concentration reaching maximum activity in the interval between
0
.08 and 0.1 mol% Cu. With increase of Cu concentrations
above 0.1 mol%, photocatalytic activity first quickly drops to the
values lower than in the unmodified TiO -SiO photocatalyst and
2
2
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Table 1. Specific surface area, average pore size, and the rate constants
of hydroxyterephthalic acid (HTPA) formation k
modified TiO -SiO samples. Relative photoactivity of the photocatalysts,
i.e. k constant of the photocatalyst compared to the k constant of the
reference unmodified photocatalyst is given in the second column.
Uncertainty of the last digit is given in parentheses.
1
of unmodified and Cu-
2
2
2
500
000
1
1
2
Sample
k
1
Relative
photoactivity
Specific
Average
pore size
[nm]
TS_3Cu
TS_1Cu
-
6
[
10
M
surface area
-
1
2 -1
min ]
[m g ]
1500
TS_0.5Cu
TS
0.0212(5)
0.048(2)
1
258(8)
3.8(3)
3.5(3)
TS_0.3Cu
TS_0.15Cu
TS_0.1Cu
1
000
TS_0.05C
u
2.3(1)
262(8)
286(8)
TS_0.08Cu
TS_0.05Cu
TS
5
00
TS_0.08C
u
0.060(2)
2.8(1)
3.4(3)
0
TS_0.1Cu
0.056(9)
2.6(2)
281(8)
255(8)
3.5(3)
3.6(3)
10
20
30
40
50
60
70
80
TS_0.15C
u
0.0124(8)
0.58(4)
2
/ degree
TS_0.3Cu
TS_0.5Cu
TS_1Cu
0.0087(9)
0.0042(6)
0.0015(5)
0.0003(2)
0.41(4)
0.20(3)
0.07(2)
0.014(1)
262(8)
263(8)
259(8)
240(8)
3.7(3)
3.6(3)
3.8(3)
4.1(3)
Figure 2. XRD diffractograms of TiO
samples
2
-SiO
2
(TS) and Cu-modified TiO
2
−SiO
2
TS_3Cu
2 2
Table 2. Relative amount of Anatase crystalline phase in TiO -SiO and Cu-
modified TiO
2 2
-SiO nanocomposites, ist crystallite lattice parameters, unit cell
volume, and mean crystallite size.
Sample
Anatase
%]
Lattice parameter [Å]
Cell volume
Cr. size
[nm]
No other crystalline species were detected. For the unmodified
TiO -SiO sample, the final quantitative analysis of crystalline
3
[
[Å ]
2
2
a=b
c
phases resulted in the following crystalline phase composition:
anatase - 89 %, brookite - 11 %, with the crystallite sizes equal
TS
89
86
83
82
88
88
85
84
86
3.798(3)
3.801(3)
3.797(3)
3.802(3)
3.798(2)
3.796(3)
3.801(2)
3.796(4)
3.799(3)
9.327(8)
9.324(9)
9.332(8)
9.329(8)
9.331(8)
9.332(8)
9.328(7)
9.318(8)
9.321(8)
135.6(3)
134.7(3)
134.5(3)
134.8(3)
134.6(2)
134.5(3)
134.8(2)
134.3(4)
134.5(3)
3.7(9)
3.3(9)
3.6(9)
3.8(9)
3.5(9)
3.3(9)
3.3(9)
3.7(9)
3.4(9)
to 3.7 nm and 10.4 nm, respectively. Cu-modified TiO
samples with different Cu concentrations were found to have the
same crystalline structure as unmodified TiO -SiO composites.
2 2
-SiO
TS_0.05Cu
TS_0.08Cu
TS_0.1Cu
TS_0.15Cu
TS_0.3Cu
TS_0.5Cu
TS_1Cu
2
2
The addition of Cu does not change the relative ratio between
anatase and brookite phase. Diffraction peaks of the (101)
anatase and (121) brookite planes were used to evaluate the
lattice parameters as well as the crystallite size. The obtained
values are summarized in Table 2 (and Table S-1; see
supporting information). There are no changes in the anatase
(
A) and brookite (B) crystal unit cell parameters and crystallite
sizes in the Cu-modified samples, as compared to unmodified
TiO -SiO composite.
Morphology of Cu-modified and unmodified TiO
photocatalysts is examined with TEM and HRTEM imaging (see
supporting information, Figures S-1 and S-2). TiO and SiO are
recognized as separate phases in the material. The SiO phase
is composed of quasi-spherical particles with the size in the
range of about 10 nm to 20 nm, while TiO forms small
2
2
-SiO
2
TS_3Cu
2
2
2
Cu K-edge XANES
2
Cu K-edge XANES analysis was used to determine the average
Cu valence state and local symmetry of Cu cations in the
nanocomposite samples, where the valence is known to directly
affect the shape and position of the edge features. The analysis
of XANES spectra was performed with the IFEFFIT program
package ATHENA.
absorption spectra is obtained by the standard procedure by
removing the extrapolated best-fit linear function determined in
the pre-edge region (-150 eV - 30 eV), and by conventional
normalization extrapolating the post-edge spline background
determined in the range from 150 to 600 eV in order to set the
Cu K-edge jump to 1. The Cu K-edge XANES spectra of the Cu
2
crystallites with size in the range 2 nm to 6 nm. The two phases
are identified with EDX elemental mapping (see supporting
information, Figure S-3). HRTEM images show that SiO
2
nano-
particles are amorphous while TiO has crystalline structure, in
2
[39]
The relative K-shell contribution in the
agreement with XRD data. The same morphology is observed in
Cu modified and unmodified samples. The addition of Cu does
not change the morphology of TiO
phase in the Cu-modified TiO -SiO samples was not detected in
TEM images, indicating that Cu is dispersed on TiO -SiO matrix
2 2
-SiO catalyst. The CuO
2
2
2
2
in amorphous form or in the form of small nano-clusters, with a
size below detection limit.
2 2
modified TiO -SiO samples and reference copper compounds
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
with different Cu cation valence state (Cu(0), Cu(I) and Cu(II))
and Cu coordination are shown on Figure 3.
Different local environments of the Cu cation result in different K-
edge profiles and pre-edge lines in the XANES spectra. The
energy position of the absorption edge and the pre-edge
features are correlated with the valence state of the absorbing
atom in the sample, and can be used to determine the valence
1
0
9
8
7
6
5
4
3
2
1
0
0
Cu
Cu(0)-MOR
[
40]
state of Cu cations.
The energy of the Cu K-edge and pre-
TS_0.08Cu_v
edge features in Cu(II) compounds is shifted for about 4 eV to
[
40,41]
Cu O
higher energies, compared to Cu(I) compounds.
reference crystalline Cu where monovalent Cu(I) is
coordinated to four oxygen atoms, a well-defined absorption
In the
2
2
O
Cu(I)-MOR
CuO
[
41,42]
feature from the 1s-4p transition occurs at 8982 eV.
In case
Cu(acac)2
of Cu O nanoparticles in the reference Cu(I)-containing MOR-
2
[
43]
type zeolite (Cu(I)-MOR) ), the 1s-4p transition occurs at 8983
eV. In case of Cu(II) compounds the 1s-4p transition appears as
pre-edge shoulder in the energy range from 8985 eV to 8989 eV.
The intensity of the shoulder increases from distorted octahedral
Cu(acac) _lq
2
Cu(OH)2
CuSO4
[
40,44]
to pure planar Cu coordination.
2
In Cu(acac) , Cu(II) is
Cu malathion
TS_3Cu
TS_1Cu
coordinated to four oxygen atoms at 1.92 Å, the 1s-4p transition
occurs at energy between 8985-8986 eV. In the crystalline CuO,
where Cu(II) cations are octahedrally coordinated to six oxygen
atoms in elongated octahedron (four oxygens at 1.95 Å and two
at 2.88 Å), a pre-edge 1s-4p shoulder is present at 8985 eV. In
case of Cu(II) octahedrally coordinated to six oxygen atoms
tetragonally distorted due to Jahn–Teller effect with four oxygens
at 1.9 Å and two at 2.3 Å, the 1s-4p shoulder appears in the
TS_0.5Cu
TS_0.3Cu
TS_0.15Cu
TS_0.1Cu
TS_0.08Cu
TS_0.05Cu
energy range between 8986 eV and 8989 eV (CuSO
4
, Cu
malathion, Cu(OH) and liquid Cu(acac) precursor).
2
2
Judging from the energy position of the Cu K-edge and
sharp pre-edge features in the XANES spectra of the catalyst
samples compared to the Cu reference compounds, the
oxidation state of copper is Cu(II) (Figure 3). The same
conclusion can be drawn also for Cu cations in the liquid
photocatalyst precursor.
8
960
8980
9000
9020
9040
E / eV
The Cu K-edge profiles of all eight samples (TS_0.05Cu,
TS_0.08Cu, TS_0.1Cu, TS_0.15Cu, TS_0.3Cu, TS_0.5Cu,
TS_1Cu, TS_3Cu) are practically identical to one another. This
qualitative observation is confirmed also by detailed quantitative
analysis of the XANES spectra by Principal Component Analysis
Figure 3. Normalized Cu K-edge XANES spectra of the Cu-modified TiO
SiO samples dried in air at 150 °C for 1 h and TS_0.08Cu_v sample
additionally dried in vacuum. Reference Cu(0), Cu(I) and Cu(II) copper
2
-
2
43
[ ]
compounds with different Cu cation coordination (Cu(0), Cu(0)-MOR , Cu
2
O,
precursor , Cu(OH)
, Cu malathion) are shown for comparison. The spectra are displaced
[
43
]
37
[ ]
Cu(I)-MOR , CuO, Cu(acac)
CuSO
2
and liquid Cu(acac)
2
2
,
[
39]
4
(PCA). The number of eigenvectors above noise level denotes
vertically for clarity. Vertical line shows the pre-edge 1s-4p shoulder. Vertical
dotted line is placed at the 1s-4p shoulder position (8988.5 eV) in the samples
to facilitate the comparison of the Cu K-edge edge shifts.
number of independent components with a physical meaning in
the set of the analysed spectra. The results of PCA suggest that
there is only one dominant principal component that describes
all eight spectra in the set. Minor differences between spectra
can be attributed to noise.
The shape of Cu K-edge in the XANES spectra of the air-dried
catalyst samples is similar to the shape of Cu K-edge of the
reference Cu compounds where Cu(II) is located at the centre of
Jahn-Teller distorted octahedron of six oxygen atoms. The linear
[
39]
combination fit (LCF) analysis
shows that the Cu K-edge
XANES spectra of the photocatalyst samples can be almost
completely described by a linear combination of two XANES
reference spectra: Cu malathion (87 %) and Cu hydroxide
(13 %) (Figure 4), where Cu(II) cations are coordinated with four
planar oxygens at short distance of about 1.95 Å and two axial
oxygens at about 2.3 Å, indicating that local symmetry and
coordination of Cu(II) cations in the catalysts is similar.
The energy position of Cu K-edge of the sample
additionally dried in vacuum (TS_0.08Cu_v) is shifted towards
lower energies, indicating the reduction of Cu(II) during drying in
vacuum. Results of LCF analysis (see supporting information,
Figure S-4) suggest that sample contains a mixture of three
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
valence states of copper: 66% of Cu(0), 29% of Cu(I) and 5% of
Cu(II).
5
4
3
2
1
0
TS_0.08Cu
1
1
1
1
0
0
0
0
0
.6
.4
.2
.0
.8
.6
.4
.2
.0
TS_3Cu
TS_1Cu
Cu malathion (87%)
TS_0.5Cu
TS_0.3Cu
Cu(OH) (13%)
2
8960
8980
9000
9020
9040
9060
TS_0.15Cu
TS_0.1Cu
E / eV
Figure 4. The Cu K-edge XANES spectrum measured on TS_0.08Cu sample.
Blue solid line – experiment; red dashed line – best-fit linear combination of
XANES profiles of Cu malathion (87%) and Cu(OH)
are shown below.
2
(13%). Fit components
TS_0.08Cu
TS_0.05Cu
Cu K-edge EXAFS
The Cu K-edge EXAFS analysis was used to determine the local
structure around Cu cations in the Cu-modified TiO -SiO
0
2
4
6
2
2
2
R / Å
catalysts. Fourier transforms (FT) of the k weighted EXAFS
spectra of the samples and references compound are shown in
Figure 5. The Fourier transforms spectra from the Cu modified
catalysts series (Figure 5) reveal the contributions of
consecutive shells of Cu neighbours of up to ~4 Å. Qualitative
comparisons of the FT spectra shows that average Cu
neighbourhood in the catalyst samples dried in air are similar but
not the same. Small differences clearly indicate structural
differences around Cu cations among the samples with different
Cu loadings.
2
Figure 5. Fourier transform magnitudes of the k -weighted Cu EXAFS spectra
of Cu-modified TiO -SiO calculated in the R range of 1–4 Å and in the k range
of 3–11 Å (blue solid line – experiment, red dashed line – EXAFS model).
2
2
-
1
In the first step, FEFF models were built for local coordination of
Cu(II) cations in mentioned Cu species and tested on the
corresponding reference Cu compounds where available (liquid
Cu(acac) precursor, crystalline Cu(acac) and crystalline CuO)
2 2
(see supporting information, Figure S-5, Table S-2).
The quantitative analysis of Cu K-edge EXAFS spectra
[
39]
were performed with the IFEFFIT program package. Structural
parameters were quantitatively resolved by comparing the
measured signal with the model EXAFS spectra, constructed
2
The Cu K-edge EXAFS spectra of the Cu-modified TiO -
[
45]
with the FEFF6 program code
in which the photoelectron
2
SiO catalyst samples are modelled with a combined FEFF
scattering paths are calculated ab initio from a tentative spatial
distribution of neighbour atoms. We can expect a mixture of
model, composed of neighbour atoms at distances characteristic
for the expected Cu(II) species that may be present in the
different Cu(II) species in the Cu-modified TiO
nanocomposites at different Cu loadings, after drying in air at
2
-SiO
2
samples. A combination of three FEFF models is used, one for
-
Cu(II) cations bound to acac groups as in crystalline Cu(acac)
2
-
1
50 °C. Part of Cu(II) cations may retain the bonding to acac
groups as in the liquid precursor or as in crystalline Cu(acac)
Cu(acac) complex may be chemisorbed on TiO particles via
coordination by surface Ti−OH groups without elimination of the
or liquid precursor, second for Cu(II) cations in crystalline CuO
nanoparticles, and third for Cu(II) cations bound to the surface of
crystalline TiO particles, taking into account the structural
2
information obtained from EXAFS analysis of the reference
Cu(II) compounds (see supporting information) and theoretical
2
.
2
2
-
[36]
acac ligand as Jin et al. indicated.
deposited on the TiO -SiO matrix as crystalline CuO
nanoparticles or amorphous Cu(II) oxides
Cu(II) cations may be
predictions for possible atomic structures of CuO nanoclusters
2
2
[
27,30,46]
[27,30,32,36]
or bound to the
2
bound to the surface of crystalline TiO particles.
[
32,36,37]
surface of crystalline TiO
2
particles.
The FEFF model included eight single scattering paths. Six
oxygen atoms were included in the first coordination shell,
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
arranged in a distorted octahedron (as suggested by XANES
analysis): four at 1.95 Å and two at 2.35 Å. Carbon atoms were
added in the second coordination shell at about 2.9 Å, to
describe the contribution of Cu-O-C connections which may be
the best fit parameters for all of the samples is given in the Table
3.
1
1
.4
.2
1
5
preserved in Cu(acac)
50 °C. An additional shell of Ti atoms around 3.0 Å is added to
the model to test Cu-O-Ti connections (Cu atoms bonded to
2
from liquid precursor after drying at
1
4.5
4
[
36,37,47]
TiO
2
surface, forming Cu-O-Ti bridges).
The Cu–Cu (R =
3
3
.5
3
.1 Å, 3.3 Å and 3.7 Å) and Cu–O (
R
= 3.6 Å) single scattering
paths are included to detect the presence of Cu and O
neighbours characteristic for second and third coordination shell
in amorphous CuO or in CuO (nano)oxides which can be
expected, if a part of Cu atoms are present in the form of
0
.8
2.5
2
0.6
0.4
[
27,36]
1.5
1
different nanostructured CuO species.
included three variable parameters for each shell of neighbours
in total 24 parameters): the coordination number ( ), the
distance to the neighbour atoms (∆ ), and the Debye-Waller
The EXAFS model
0
.2
(
N
0
0
300
250
200
.5
r
2
2
0
0
factors (σ ). The amplitude reduction factor (
S
) was fixed at the
3.5
3
value of 0.8 obtained in the model of reference compounds. A
shift of energy origin ∆E , common to all scattering paths, was
o
also varied.
The fits of individual spectra reveal similar structures
around Cu cations in all the samples. Cu EXAFS results show
that Cu atoms in all eight samples are coordinated to six oxygen
atoms in a distorted octahedron: four oxygen atoms in the
equatorial plane are at shorter Cu–O distance (1.94 Å –1.97 Å)
and two oxygen atoms on axial positions at longer distance
2
1
0
.5
2
1
1
5
0
50
00
0
.5
1
.5
0
(
2.35 Å). The second shell of neighbours consists of Ti and C
atoms at 2.85 Å and 2.9 Å, respectively and Cu neighbours
distributed from 3.0 Å to 3.9 Å. The presence of Cu-O-C
connections indicate that Cu coordination to two acac groups,
0
0.05
0.08
0.1
0.15
0.3
0.5
1
3
-
Samples with differentCu concentration
as found in in the liquid precursor, are partially preserved after
drying at 150 °C, however, a part of Cu cations is bound to TiO
2
Figure 7. Histogram of relative photocatalytic activity of unmodified TiO
and Cu-modified TiO -SiO samples dried at 150 °C, relative amount of Cu-O-
Ti bonds (Red line – experiment, dashed green line – model), Cu-O-Cu bonds
blue line), and specific surface area of a given catalysts.
2 2
-SiO
2
2
surface forming Cu-O-Ti connections and part of Cu cations are
present in the form of CuO nanoparticles as indicated by Cu-O-
Cu connections. In the fitting of the second coordination shell,
the contributions of silicon backscattering were ruled out, judging
from the unfavourable fit results (absence of Cu-O-Si
connections).
(
The most important structural difference between the Cu
modified TiO -SiO nanocomposites at different Cu loadings
2
2
To detect small structural differences between the samples
in the local Cu neighbourhood and to minimize uncertainties of
fitting parameters due to high correlations between them in the
fit of individual spectra, a parallel fit of the spectra (divided in the
groups of three) was performed, where some parameters for all
spectra were constrained to common values and some
parameters were fixed at the values obtained in the fit of
reference spectra. The coordination number of nearest oxygen
neighbours and their distances to Cu atom in the first
coordination shell is varied for each sample. Debye–Waller
factors of all paths except the first were fixed to common values
for the same type of atoms (Ti, Cu and O). For all spectra, Cu-Ti
distance was constrained to a common value. Number of axial O
atoms and their distance to Cu atom is fixed to the known values
based on the liquid Cu precursor model (two O at 2.35 Å).
Number of C atoms was varied at fixed distance of 2.8 Å. At
larger distances, fixed number of Cu and O atoms that belong to
CuO oxide nanoparticles found in crystalline CuO. There were in
total 42 independent points and 27 independent variables in the
detected by EXAFS analysis is the average coordination number
of Ti neighbours found in the second Cu coordination shell at
2
.85 Å. Average number of Cu-O-Ti connections changes with
Cu concentration (Figure 7). At low Cu loadings (from 0.05 to 0.1
mol% Cu), we found highest number of Cu-O-Ti connections (Cu
is adsorbed on TiO surface). Those samples are also most
2
photocatalytically active. At increased Cu loadings (from 0.1 to 3
mol%), we observed drop in the average number of Cu-O-Ti
connections. Average number of Cu neighbours in more distant
coordination shells is almost constant, independent of Cu
loading.
The presence of Cu-O-C connections clearly indicate that
part of Cu(II) cations remain in the form of Cu(acac) preserved
2
from liquid precursor after drying. Relative number of Cu
-
connections to organic acac group is also independent of Cu
loading. We can estimate that about 20% to 30% of Cu remains
-
bonded to acac groups after drying at 150 °C at all Cu
concentrations. The Cu(acac)
TiO via coordination by surface Ti−OH groups without
elimination of the acac ligands.
2
is chemisorbed on the surface of
2
parallel fit of three spectra within the fitting range in the
k
interval
range of 1 Å up to 4 Å. A very
good fit is obtained for all spectra (Figure 5). A complete list of
-
[36]
-
1
-1
from 3 Å to 11 Å and the
R
We can explain the observed decrease of average number
of Ti neighbours at higher Cu loadings (from 0.1 to 3 mol%), if
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
we assume that at Cu concentrations above 0.1 mol% absolute
holes migrate to the catalyst surface and participate in redox
reactions.
2
amount of Cu directly bonded to TiO surface remains the same
as at the concentration of 0.1mol%, independent of the Cu
loading, and the remaining Cu forms CuO nanoparticles or
[
30]
[31]
The experimental findings of Li et al.
and Xin et al.
showed that modification of TiO surface with Cu(II) cations may
2
-
remains attached to acac groups. The theoretical prediction of a
increase the photocatalytic activity at low Cu concentrations.
They explained that metal to metal charge transfer in Ti-O-Cu
decrease of an average number of Ti neighbours at higher Cu
loadings, based on above assumption, matches with the
observed measured trend (Figure 7).
complexes at TiO
2
surfaces enhance photoactivity through
charge carrier separation. Our results strongly support
[
30]
[31]
The fact that average number of Cu neighbours in more
distant coordination shells, attributed to Cu cations bounded in
CuO nanoparticles, is approximately constant and independent
of Cu loading, suggests that CuO nanoparticles do not grow to
form larger particles, but rather that at higher Cu loadings,
mentioned findings of Li et al. and Xin et al. . We also found
that that interfacial Ti-O-Cu linkages exist in a Cu modified TiO
SiO nanocomposite at low copper loadings and we can ascribe
the enhancement of photocatalytic activity to the metal to metal
charge transfer across Ti-O-Cu bridges at TiO surfaces.
2
–
2
2
smaller CuO nanoparticles are dispersed on the surface of TiO
SiO matrix.
To examine the stability of the nanocomposites after drying
at 150 °C, one Cu-modified photocatalyst (TS_0.08Cu) was
additionally dried in vacuum (TS_0.08Cu_v). The results (see
supplementary information) reveal that by exposing Cu-modified
2
-
At higher Cu loadings Cu forms CuO nanoparticles or
remains attached to acac groups and consequently, relative
amount of Cu-O-Ti bridges is reduced. Decrease of
photocatalytic activity at higher Cu concentrations can be
-
2
[
48]
ascribed to the shielding effect
that prevents the access of pollutants to the photoactive surface
of TiO and hinders its photocatalytic activity. The mechanism of
inhibition of TiO photocatalytic activity by CuO nanoclusters on
its surface is additionally supported by findings of G. Li et al.
who showed that CuO nanoparticles on TiO surface decrease
photocatalytic activity due to charge transfer from TiO to CuO
and subsequent charge recombination in CuO nanoparticles.
2
of CuO on the TiO surface
TiO
2
-SiO
2
photocatalyst nanocomposites to vacuum Cu(II)
2
cations in the sample are reduced to Cu(0) metal nanoparticles
with fcc crystal structure and Cu(I) nanooxide particles, while
Cu-O-C connections are preserved. As we already showed in
2
[
30]
,
2
[
37]
our previous work
additional calcination of the dried Cu-
2
photocatalysts at 500 °C also induces
2
modified TiO -SiO
2
significant structural changes of the nanocomposite: Cu-O-Ti
connections are lost, Cu partially incorporates into the SiO
2
matrix, and amorphous copper oxides are formed. In this way Conclusions
the photocatalytic activity of the material is lost, so Cu-modified
TiO
vacuum.
XRD results showed that Cu modification of TiO
matrix, did not change anatase and brookite crystal unit cell
parameters and crystallite size of the unmodified TiO -SiO
2 2
-SiO photocatalysts is unstable if calcined or exposed to
2 2
In this research we have shown that modification of TiO -SiO
photocatalysts with low amount of copper, synthesized by low
temperature sol-gel synthesis based on organic Si, Ti, and Cu
precursors leads to increased photocatalytic activity under
UV/Vis irradiation. Photocatalytic activity measurements showed
up to three times increase in photocatalytic activity of air dried
-SiO
2 2
2
2
composite, and the specific surface area of all Cu modified
photocatalysts is the same or very similar as that of unmodified
2 2
photocatalysts when TiO -SiO matrix is modified with Cu in a
TiO
that the addition of Cu does not change the morphology of TiO
SiO catalyst. From the observed difference in Cu local structure,
we can conclude that the key feature influencing the
photocatalytic activity of Cu modified TiO -SiO photocatalysts is
the relative number of direct connections of Cu(II) cations to the
surface of TiO nanoparticles, distinguishing high active from low
2 2
-SiO nanocomposite. Also TEM and HRTEM images show
narrow concentration range from 0.05 to 0.1 mol%. At higher Cu
concentrations dried samples exhibit smaller photocatalytic
2
-
2
activity compared to the unmodified reference TiO
photocatalyst. All samples exhibit predominantly nanocrystalline
anatase crystal structure of TiO with a smaller amount of
brookite phase, with the same crystal unit cell parameters and
same crystal sizes as in unmodified TiO -SiO photocatalyst,
independent of the relative amount of Cu in the TiO -SiO matrix.
SiO is in the form of glass-like amorphous silicate nanoparticles.
Cu-modified samples containing low amount of copper (from
.05 to 0.1 mol%) showed largest number of Cu-O-Ti
-SiO
2 2
2
2
2
2
2
2
active and inactive samples.
2
2
The observed dependence of photocatalyst activity at
different Cu loadings indicates a presence of two competitive
mechanisms responsible for improvement or reduction of
2
0
2 2
photocatalytic activity of the Cu modified TiO -SiO
connections, and consequentially increased photocatalytic
activity. Metal to metal charge transfer in Ti-O-Cu complexes at
TiO surfaces enhances photoactivity through charge carrier
2
separation. At higher Cu loadings (more than 0.1 mol% Cu),
average number of Cu-O-Ti connections decreases, major part
of Cu cations formed amorphous or nanocrystalline Cu(II) oxide
photocatalysts: one mechanism responsible for promoting
photocatalytic activity at low Cu concentrations (from 0.05 to 0.1
mol% Cu) and inhibition mechanism strongly expressed at high
Cu loadings (higher than 0.1 mol% Cu).
To understand the nature of the photocatalytic
improvement/inhibition, we need to elucidate the photo-reaction
2
nanoparticles on the surface of TiO and photocatalytic activity
mechanism of TiO
2
. The basic mechanism of photoinitiated
of the material is hindered.
oxidative transformations of organic pollutants on the TiO
surface is well-known. Water or the hydroxyl ion (OH ) is the
2
-
possible trap for the valence band hole, which leads to the
•
formation of the hydroxyl radical ( OH). Oxygen over the TiO
2
surface traps the covalent band electron and generates the
superoxide radical anion. After charge separation, electrons and
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Table 3. Parameters of the nearest coordination shells around Cu atoms in
2
2
Cu
neighbour
N
R [Å]
σ [Å ]
R-factor
2
Cu-modified TiO
2
-SiO
2
: N - the number of neighbour atoms, R - distance, σ -
Debye-Waller factors, and R-factor - the goodness of fit parameter. The best
fit is obtained with the amplitude reduction factor S = 0.8 eV.
0
2
TS_0.3Cu
1.956(6)
2.35
O
2
2
3.8(3)
2
0.004(1)
0.02
Cu
N
R [Å]
σ [Å ]
R-factor
neighbour
O
TS_0.05Cu
1.943(8)
2.35
C
1
0
1
.0(5)
.2(3)
.3(4)
2
2.8
0.008
0.009
0.009
0.009
0.01
O
O
4.7(5)
2
0.006(2)
0.02
Ti
0.0068
2.80(3)
3.15(4)
3.37(4)
3.8(1)
Cu
Cu
O
C
1.0(5)
0.8(5)
1.4(6)
2
2.8
0.008
0.009
0.009
0.009
0.01
Ti
2.80(3)
2.93(6)
3.12(5)
3.71(9)
3.94(7)
0.0063
2
Cu
Cu
O
Cu
1
3.9(1)
0.01
TS_0.5Cu
2
O
3
1
.9(5)
2
1.956(6)
2.35
0.004(2)
0.02
Cu
1
0.01
O
TS_0.08Cu
C
.0(5)
2.8
0.008
0.009
0.009
0.009
0.01
O
Ti
0.0068
0.0075
0.0075
4
.0(5)
2
1.974(8)
2.35
0.005(2)
0.02
0.3(4)
2.80(3)
3.2(1)
3.48(3)
3.6(1)
O
Cu
Cu
O
0.5(7)
C
1
1
1
.0(5)
.0(4)
.4(6)
2
2.8
0.008
0.009
0.009
0.009
0.01
2
2
1
Ti
0.0063
2.80(3)
3.03(2)
3.25(3)
3.8(1)
Cu
Cu
O
Cu
3.7(1)
0.01
TS_1Cu
O
2
3.8(3)
2
1.962(6)
2.35
0.003(1)
0.02
Cu
O
1
3.91(7)
0.01
TS_0.1Cu
C
1
.0(5)
2.8
0.008
0.009
0.009
0.009
0.01
O
3
.7(5)
2
1.974(8)
2.35
0.004(2)
0.02
Ti
<
0.1
.8(9)
2
2.85
O
Cu
Cu
O
0
3.3(1)
3.50(7)
3.6(1)
C
1
0
2
.0(5)
.6(5)
.3(9)
2
2.8
0.008
0.009
0.009
0.009
0.01
Ti
0.0063
2.80(3)
3.03(2)
3.28(4)
3.8(1)
2
Cu
Cu
O
Cu
1
3.79(5)
0.01
TS_3Cu
2
O
3
1
1
.6(3)
2
1.958(6)
2.35
0.004(1)
0.02
Cu
1
3.9(1)
0.01
O
TS_0.15Cu
C
.0(5)
2.8
0.008
0.009
0.009
0.009
0.01
O
4
.1(4)
2
1.956(6)
2.35
0.006(1)
0.02
Ti
<
0.1
.2(6)
2
2.85
O
Cu
Cu
O
3.12(3)
3.35(4)
3.50(9)
3.93(5)
C
1
0
2
.0(5)
.4(3)
.6(4)
2
2.8
0.008
0.009
0.009
0.009
0.01
Ti
0.0068
2.80(3)
3.03(2)
3.25(3)
3.85(7)
3.80(5)
2
Cu
Cu
O
Cu
1
0.01
2
Cu
1
0.01
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Experimental Section
photocatalyst (TS_0.08Cu) was additionally dried in vacuum at room
temperature (sample denoted as TS_0.08Cu_v) to examine the stability
2 2
of the Cu species, present in the Cu-modified TiO -SiO nanocomposite
photocatalysts after drying at 150 °C in air, if the photocatalyst is
additionaly dried in vacuum.
2 2
Synthesis of Cu-modified TiO -SiO photocatalysts
Characterization
Photocatalytic activity tests
2 2
The photocatalytic activity of the unmodified and Cu-modified TiO -SiO
catalysts in the powder form was determined using a fluorescence-based
method of terephthalic acid decomposition. The stock solution of TPA
[
50]
was prepared by mixing together TPA (130 mg) and aqueous solution
-
3
[
38]
of NaOH (1L, 2·10 M).
Working solution of TPA (100 mL, 83 mg/L)
was freshly made prior to photocatalytic experiments. 25 mL of TPA
working solution and 10 mg of photocatalyst were mixed in 25 mL of
double deionized water and stirred under sunlight irradiation. Samples (1
mL) of the aqueous solution were taken from the 100 mL glass beaker at
different UV irradiation times (0 min, 5 min, 10 min, 15 min, 20 min, 30
-
1
min, and 45 min) and centrifuged (1300 min ) for 3 min. A fixed volume
159 μL) of the solution was then sampled with an automatic pipette, and
(
transferred into microtiter plate wells (microtiter plate with 96 wells, flat
bottom, black) for fluorescence measurements. Photocatalytic tests were
carried out in solar simulator (Suntest XLS+, Atlas, USA) chamber with a
simulated solar irradiation source (Xenon lamp), using daylight filter
2
(
300–800 nm), at UV light flux of 750 W/m . During irradiation in the
presence of the photocatalyst, TPA is decomposed and highly
fluorescent 2-hydroxyterephthalic acid (HTPA) is formed as an
intermediate oxidation product. Fluorescence measurements were
performed using a microplate reader in the fluorescence mode (Infinite
F200 Microplate reader, Tecan, Switzerland). The wavelength of the
excitation light was 320 nm (filter bandwidth: 25 nm) and emission was
measured at 430 nm (filter bandwidth: 35 nm). The instrument was
operating in top mode with 25 reads per well, with 20 μs integration time.
The amplification factor for the photomultiplier tube was 56. For each
irradiation time, at least four parallel photocatalytic tests were done.
Measurements of HTPA concentrations were performed in the time
interval from 0 to 45 minutes.
2 2
Figure 8. Diagram of sol-gel synthesis of Cu-modified TiO -SiO
photocatalysts.
Low temperature sol-gel method was used to synthesize Cu-modified
TiO -SiO photocatalysts with different copper concentrations (0.05, 0.08,
.1, 0.15, 0.3, 0.5, 1 and 3 mol%). We have revised a formerly proposed
XRD
2
2
0
2 2
The crystal structures of unmodified and Cu-modified TiO -SiO samples
[ ]
49
37
[ ]
synthesis method , as described in detail in our previous publication.
The chemicals used for synthesis were titanium tetraisopropoxide (TTIP),
tetraethyl ortosilicate (TEOS) and colloidal SiO and copper
acetlyacetonate (Cu(acac) ) as Ti, Si and Cu sources (Figure 8). For the
TiO -SiO colloidal solution we prepared silica binder from tetraethyl
orthosilicate (TEOS), colloidal SiO Levasil 200/30% aqueous solution,
HCl to catalyse TEOS hydrolysis and after 1h of mixing 1-propanol was
added. The TiO sol was prepared by dissolving titanium
were investigated by X-ray diffraction (MiniFlex Benchtop 300/600, 150)
using Cu Kα irradiation from 10 to 80° at a scan rate of 2°/min.
Quantitative phase composition analysis was performed using Rietveld
refinement method by the High Score Plus software. The crystallite size
was determined from XRD pattern, using Scherrer formula:
2
,
2
2
2
2
d = (0.9λ/β cos θ)
(1)
2
tetraisopropoxide (TTIP) in absolute ethanol. In the first step, double
deionized water and perchloric acid were mixed separately. This solution
was then added to the TTIP solution drop-wise, under reflux and heating,
where exothermic reaction of uncontrolled hydrolysis and condensation
where d is the crystallite size in nm, λ is the wavelength of X-ray in Å
1.5418 Å), β is the full width of diffraction peaks at half maxima (FWHM)
in radians, and θ is the Bragg angle.
(
of TTIP took place, gaining white precipitate of hydrated amorphous TiO
2
.
After heating and refluxing for 48 h, a stable translucent TiO sol was
2
2
N -physisorption
obtained. The TiO
binder solution to TiO
2
-SiO
2
sol (denoted TS) was obtained by adding silica
sol. The sol was further diluted with double
2
The specific surface area of the unmodified and Cu-modified TiO -SiO
2
2
deionized water and organic solvents (1-propanol and 2-propoxyethanol).
Copper cations were added by direct incorporation during the sol-gel
photocatalyst samples was evaluated from N2 sorption isotherms
obtained at 77 K using a Tristar 3000 Micromeritics volumetric adsorption
[
51
]
synthesis. Cu(acac)
copper, was added into the TiO
and Cu-modified) were dried at 150 °C for 1 h, to obtain unmodified TiO
SiO and Cu-modified TiO -SiO catalysts with different copper
2
dissolved in 2-propoxyethanol, used as a source of
analyser. The BET specific surface area
was calculated from
2
-SiO sol. All TiO -SiO sols (unmodified
2
2
2
adsorption data in a relative pressure range from 0.05 to 0.25. The pore
size distributions (PSDs) were calculated from nitrogen adsorption data
2
-
[
52]
2
2
2
using the Barrett-Joyner-Halenda (BJH) method.
concentrations (0.05, 0.08, 0.1, 0.15, 0.3, 0.5, 1 and 3 mol%), denoted as
TS, TS_0.05Cu, TS_0.08Cu, TS_0.1Cu, TS_0.15Cu, TS_0.3Cu,
TS_0.5Cu, TS_1Cu, TS_3Cu, respectively. One Cu modified
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Giuliana Aquilanti and Luca Olivi, from XAFS beamline at
ELETTRA, and Edmund Welter from P65 at PETRA III, for the
support and expert advice on beamline operation. Thanks also
goes to Mojca Opresnik from the National Institute of Chemistry
TEM /STEM/EDX
Unmodified and Cu-modified TiO
2
-SiO
2
photocatalyst samples for
for N
2
-physisorption measurements.
TEM/STEM/EDX characterization have been prepared by immersing a
Lacey-carbon TEM grid in a water suspension and air-dried at room
temperature. The electron microscopy analysis has been performed by
means of a Field-Emission Transmission Electron Microscope (JEM-
Keywords:
titanium dioxide,
Cu-modified
2 2
TiO -SiO
photocatalysts, photocatalytic activity, Cu K-edge XANES,
EXAFS.
2
100F, JEOL) operated at 200 kV. The microscope is equipped with a
STEM unit and a detector for EDX mapping and spectroscopy (X-MAX80,
Oxford).
[
1]
2]
K. Guan, Surf. Coat. Technol. 2005, 191, 155–160.
M. R. Hoffmann, S. T. Martin, W. Choi, D. W. Bahnemann, 1995, Chem
Rev. 95, 69–96.
[
X-ray absorption spectroscopy
[
3]
R. W. Matthews, J. Phys. Chem. 1987, 91, 3328–3333.
X. Gao, I.E. Wachs, Catal. Today 1999, 51, 233-254.
W. Dong, C. Wee Lee, X. Lu, Y. Sun, W. Hua, G. Zhuang, S. Zhang, J.
Chen, H. Hou, D. Zhao, Appl. Catal. B, 2010, 95, 197–207.
C. Anderson, A. J. Bard, J. Phys. Chem. 1995, 99, 9882-9885.
W. Choi, A. Termin, M. R. Hoffmann, J. Phys. Chem. 1994, 98, 13669–
13679.
The local atomic structure and the chemical state of copper cations in
Cu-modified TiO -SiO photocatalysts were analysed by X-ray absorption
2 2
[4]
[5]
spectroscopy methods XANES and EXAFS. Cu K-edge absorption
spectra of the prepared photocatalyst samples were recorded at room
temperature in transmission and fluorescence detection mode at the
XAFS beamline of the ELETTRA synchrotron radiation facility in Trieste,
Italy, and at P65 beamline of PETRA III, DESY, Hamburg, Germany. The
[6]
[7]
[8]
[9]
M. Tahir, B. Tahir, N. A. S. Amin, A. Muhhamad, Energy Convers.
Manage. 2016, 119, 368–378.
Cu-modified TiO
2
-SiO
2
samples were prepared in the form of
homogenous pellets, pressed from micronized powder mixed with BN. In
all cases the total absorption thickness (µd) of the sample was about 2
above the Cu K-edge. A Si (111) double crystal monochromator was
used with energy resolution of 1 eV at Cu K-edge. The intensity of the
monochromatic X-ray beam was measured by three consecutive
ionization detectors (XAFS beamline: 30 cm long ionisation chambers
M. M. Momeni, M. Hakimian, A. Kazempour, Ceram. Int. 2015, 41,
13692–13701.
[10] B. Xin, L. Jing, Z. Ren, B. Wang, H. Fu, J. Phys. Chem. B 2005, 109,
2805–2809.
[11] M. M. Momeni, Y. Ghayeb, A. A. Mozafari, J. Mater. Sci. Mater.
Electron. 2016, 27, 11201–11210.
filled with gas mixtures: (first) 1250 mbar N
mbar Ar, 1000 mbar N 750 mbar He; (third) 1000 mbar N
00 mbar Ar; P65 beamline: 5 cm long ionisation chambers filled with:
first) 1000 mbar N , (second) 1000 mbar Ar, (third) 1000 mbar Kr). The
samples were mounted on the sample holder placed after the first
ionization detector. For fluorescence detection, the samples were rotated
2
, 750 mbar He; (second) 250
[12] M. M. Momeni, Y. Ghayeb, S. Gheibee, Ceram. Int. 2017, 43, 564–570.
[13] M. M. Momeni, Z. Nazari, Ceram. Int. 2016, 42, 8691–8697.
[14] S. Sakthivel, M. V. Shankar, M. Palanichamy, B. Arabindoo, D. W.
Bahnemann, V. Murugesan, Water Res. 2004, 38, 3001–3008.
[15] R. Chand, E. Obuchi, K. Katoh, H. N. Luitel, K. Nakano, Catal.
Commun. 2011, 13, 49–53.
2
2
, 300 mbar He,
7
(
2
4
5° with respect to the X-ray beam. The absorption spectra were
[16] N. Wu, M.S. Li, Int. J. Hydrogen Energy 2004, 29, 1601−1605.
[17] M. Tahir, B. Tahir, Appl. Surf. Sci. 2016, 377, 244–252.
[18] M. Tahir, N. S. Amin, Appl. Catal., A 2015, 493, 90–102.
[19] M. M. Momeni, Y. Ghayeb, F. Ezati, J. Colloid Interface Sci. 2018, 514,
70–82.
measured in the energy region from -150 eV to +1000 eV relative to the
Cu K-edge. In the XANES region equidistant energy steps of 0.25 eV
were used, while for the EXAFS region equidistant k steps of 0.03 Å
were adopted, with an integration time of 2s/step for transmission
measurements. For samples with low Cu concentration, fluorescence
detection mode was used, with SDD fluorescence detector at XAFS
beamline, and 7-pixel Ge detector at P65 beamline, to measure the
intensity of Cu K-alpha emission line. At both beamlines we used 100-
micron thick aluminium foil as a filter to reduce Ti K-alpha signal from the
sample. Three to eight repetitions of the scans with integration time of
-
1
[20] W. Zhang, Y. Li, S. Zhu, F. Wang, Catal. Today 2004, 93–95, 589–594.
[21] S. Klosek, D. Raftery, J. Phys. Chem. B 2001, 105, 2815–2819.
[22] K. Su, Z. Ai, L. Zhang, J. Phys. Chem. C 2012, 116, 17118–17123.
[23] R. Lopez, R. Gomez, M. Llanos, Catal. Today 2009, 148, 103–108.
[24] Z. Ambrus, N. Balazs, T. Alapi, G. Wittmann, P. Sipos, A. Dombi, K.
Mogyorosi, Appl. Catal. B 2008, 81, 27–37.
5
s/step, were superimposed to improve signal-to-noise ratio. The exact
[25] M. M. Momeni, Y. Ghayeb, J. Mater. Sci. Mater. Electron. 2016, 27,
1062–1069.
energy calibration was established with simultaneous absorption
measurement on a 5-micron thick Cu metal foil placed between the
second and the third ionization chamber. The first inflection point in the
spectrum of the copper K-edge in the Cu metal with fcc crystal structure
is at 8979.0 eV Absolute energy reproducibility of the measured spectra
was ±0.03 eV. The quantitative analysis of EXAFS spectra is performed
[26] X. Qiu, M. Miyauchi, K. Sunada, M. Minoshima, M. Liu,Y. Lu, D. Li, Y.
Shimodaira, Y. Hosogi, Y. Kuroda, K. Hashimoto, ACS Nano 2012, 6,
1609–1618.
[27] X. Q. Qiu, M. Miyauchi, H. G. Yu, H. Irie, K. Hashimoto, J. Am. Chem.
Soc. 2010, 132, 15259–15267.
[39]
with the IFEFFIT program package
program code for ab initio calculation of photoelectron scattering paths.
in combination with FEFF6
[28] H. Irie, S. Miura, K. Kamiya, K. Hashimoto, Chem. Phys. Lett. 2008,
457, 202–205.
[45]
[
[
[
29] J. Li, D. Zhen, G. Sui, C. Zhang, Q. Deng, L. Jia, Nanosci. Nanotechnol.
012, 12, 6265-6270.
30] G. Li, N. M. Dimitrijevic, L. Chen, T. Rajh, K. A. Gray, J. Phys. Chem. C
008, 112, 19040–19044.
2
Acknowledgements
2
31] B. Xin, P. Wang, D. Ding, J. Liu, Z. Ren, H. Fu, Appl. Surf. Sci. 2008,
254, 2569–2574.
This work was supported by the Ministry of Education, Science
and Sport of Slovenia, the Slovenian Research Agency
[32] M. Hussein, N. Assadi, D. A. H. Hanaor, Appl. Surf. Sci. 2016, 387,
82–689.
6
(
research programs P1-0112, and P1-0134). Access to
[
33] J. Arana, C. Fernandez Rodrıguez, O. Gonzalez Diaz, J.A. Herrera
Melian, J. Perez Pena, Catal. Today 2005, 101, 261–266.
34] M. Miyauchi, H. Irie, M. Liu, X. Qiu, H. Yu, K. Sunada, K. Hashimoto, J.
Phys. Chem. Lett. 2016, 7, 75−84.
synchrotron radiation facilities of ELETTRA, Trieste, Italy,
beamline XAFS, project 20140041, and P65 beamline of PETRA
III at DESY, Hamburg, Germany, a member of the Helmholtz
Association, project I-20160044 EC, is acknowledged. We thank
[
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
[
[
[
[
35] H. Irie, K. Kamiya, T. Shibanuma, S. Miura, D. A. Tryk, T. Yokoyama, K.
Hashimoto, J. Phys. Chem. C 2009, 113, 10761–10766.
36] Q. Jin, M. Fujishima, A. Iwaszuk, M. Nolan, H. Tada, J. Phys. Chem. C
2
013, 117, 23848−23857.
37] T. Čižmar, I. Arčon, U. Lavrenčič-Štangar, Catal. Today 2017, 287,
55–160.
38] U. Černigoj, M. Kete, U. Lavrenčič-Štangar, Catal. Today 2010, 151,
6-52.
1
4
[
39] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537–541.
40] F. W. Lytle, R. B. Greegor, A. J. Panson, Phys. Rev. B 1988, 37, 1550–
[
1562.
[
[
[
[
[
[
[
[
[
[
[
[
41] L. Kau, D. J. Spira-Solomon, J. E. Penner-Hahn, K. O. Hodgson, E. I.
Solomon, J. Am. Chem. Soc. 1987, 109, 6433-6442.
42] C. Maurizio, F. d’Acapito, M. Benfatto, S. Mobilio, E. Cattaruzza, F.
Gonella, Eur. Phys. J. B 2000, 14, 211–216.
43] E. M. C. Alayon, M. Nachtegaal, A. Bodi, M. Ranocchiari, J. A. van
Bokhoven, Phys.Chem.Chem.Phys. 2015, 17, 7681- 7693.
44] A. Manceau, A. Matynia, Geochim. Cosmochim. Acta 2010, 74, 2556–
2
580.
45] J. J. Rehr, R. C. Albers, S. I. Zabinsky, Phys. Rev. Lett. 1992, 69,
397–3400.
46] X. H. Xia, Y. Gao, Z. Wang, Z. J. Jia, J. Phys. Chem. Solids 2008, 69,
888–2893.
3
2
47] O. Pliekhov, I. Arčon, N. Novak Tušar, U. Lavrenčič Štangar,
ChemCatChem 2016, 8, 2109–2118.
48] N. Slamet, H. W. Purnama, E. Kosela, S. Gunlazuardi, J. Catal.
Commun. 2005, 6, 313-319.
49] A. Soklič, M. Tasbihi, M. Kete, U. Lavrenčič-Štangar, Catal. Today 2015,
252, 54-60.
50] R. Thiruvenkatachari, T. O. Kwon, J. C. Jun, S. Balaji, M.
Matherswaran, I. S. Moon, J. Hazard. Mater. 2007, 142, 308-314.
51] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60,
3
09–319.
52] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 73,
73–380.
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201800249
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Optimal Cu loading: Cu-modified
TiO -SiO is synthesized using low-
2 2
T. Čižmar, Prof. Dr. Urška Lavrenčič
Štangar, Prof. Dr. Iztok Arčon*
3
.0
.5
2.0
Cu-O-Ti connections
Relative photoactivity
2
temperature sol-gel method, with Cu
concentrations from 0.05 to 3 mol%.
Photocatalytic activity is increased
up to three times for Cu
concentrations from 0.05 to 0.1
mol%, due to metal to metal charge
transfer in Ti-O-Cu complexes at
1
.5
2.0
Page No. – Page No.
1.5
1.0
0.5
0.0
1.0
Effects of different Cu loadings on
photocatalytic activity of TiO -SiO
2 2
prepared at low temperature for
oxidation of organic pollutants in
water
0.5
0.0
TiO
CuO nanoparticles on the surface of
TiO hinder photocatalytic activity of
the material.
2
surfaces. At higher Cu loadings
0
0.05 0.08 0.1 0.15 0.3 0.5
1
3
Samples with different Cu concentration / mol%
2
This article is protected by copyright. All rights reserved.