Materials Science in Semiconductor Processing 91 (2019) 47–57

Contents lists available at ScienceDirect

Materials Science in Semiconductor Processing

journal homepage: www.elsevier.com/locate/mssp

T
Dry-co-grinding of doped TiO2 with nitrogen, silicon or selenium for
enhanced photocatalytic activity under UV/visible and visible light
irradiation for environmental applications
Laila Galeanoa,c, , Sergio Valenciab,c, Gloria Restrepoc, Juan M. Marínc
⁎

a
Grupo de Investigaciones y Mediciones Ambientales (GEMA), Universidad de Medellín, Carrera 87 No 30-65, Medellín, Colombia
b
Grupo de investigación Integra, Tecnológico de Antioquia, Calle 78B No 72A-220, Medellín, Colombia
c
Grupo Procesos Fisicoquímicos Aplicados, Departamento de Ingeniería Química, Facultad de Ingeniería, Sede de Investigación Universit:aria, Universidad de Antioquia
UdeA, Calle 70 No. 52-21, Medellín, Colombia

A R TICL E INFO A BSTR A CT

Keywords: TiO2 doped with nitrogen (N), silicon (Si), or selenium (Se) (N-TiO2, Si-TiO2, and Se-TiO2) were obtained by the
Titanium dioxide integrated sol-gel and solvothermal method with short time of crystallization and low temperature. The UV/
Sol-gel visible and visible light absorption and photocatalytic activity of these doped TiO2 materials were improved by a
Solvothermal dry-co-grinding process with a short grinding time and low rotational speed (30 min at 200 rpm) to obtain N-
Doping TiO2
TiO2/Si-TiO2 and N-TiO2/Se-TiO2 catalysts. The materials were characterized by XRD, Raman, BET surface area
Co-grinding process
and porosity, XRF, SEM, TEM, FTIR-ATR, and UV/vis-DRS analyses. The photocatalytic activity of these mate-
rials was evaluated by the degradation of phenol under UV/visible and visible light irradiation. The integrated
sol-gel and solvothermal methods with short time of crystallization (2 h) and low temperature (225 °C), and the
dry-co-grinding process during 30 min at 200 rpm led to materials (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) with
higher specific surface area, a reduction in the band gap value, and an enhancement of the absorption in the
visible light spectrum. Moreover, N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 exhibited higher photocatalytic activities
for degradation of phenol under UV/visible and visible light irradiation than those obtained with the doped
TiO2, synthesized TiO2 or TiO2 P25.

1. Introduction where most of the solar energy is not used [16,42].

Several strategies have been proposed to improve the absorption of
Heterogeneous photocatalysis with titanium dioxide (TiO2) has visible light of TiO2 and its photoactivity, such as doping TiO2 with
become an attractive choice for environmental applications in the de- non-metal and transition metal elements [27,42,8] and physical mix-
gradation and mineralization of many pollutants. Some of these pol- ture with other semiconductors [21]. In addition to mechanical treat-
lutants are very difficult to remove by conventional methods, such as ments (grinding process) have been developed [20,21,35].
natural organic matter, emerging pollutants, dyes, endocrine disrupting The doping of TiO2 has focused various dopants [43], among them:
compounds, etc. [17,29,54]. TiO2 is the semiconductor most commonly nitrogen (N) [47], halogen atoms [27], silicon (Si) [59], selenium (Se)
studied and used due to its properties, such as strong oxidant, abundant [28], etc. Non-metal element N seems to be the most attractive dopant
material and chemically and biologically stable compound. Moreover, for TiO2. This is due to its small energy of ionization, stability, and ionic
TiO2 has low toxicity and low cost [64]. radius (which is approximate to the oxygen radius), thus making the
Heterogeneous photocatalysis with TiO2 continues to suffer from substitution of oxygen from the network more likely [22,58,6]. The
several limitations that restrict its application and commercialization in doping with Si generated a Si-O-Ti bond that could accelerate the
the environmental field. The main technical barriers are the inefficient transfer of the photogenerated electrons, significantly improving the
exploitation of visible light and the post-recovery of TiO2 particles after photocurrent density [13,48]. Additionally, Si improved the photo-
treatment [10,12]. TiO2 can only be excited by UV light irradiation catalytic activity of TiO2 (Si doped TiO2) when a suitable amount of Si
shorter than 380 nm, due to its wide band gap (3.2 eV) [36]. This band was used [32,59]. On the other hand, the doping with Se moderately
gap limits its applications to the UV part of the sunlight spectrum, extended the absorpiton of TiO2 to visible region. In addition, Se can act

⁎
Corresponding author at: Grupo de Investigaciones y Mediciones Ambientales (GEMA), Universidad de Medellín, Carrera 87 No 30-65, Medellín, Colombia.
E-mail address: [email protected] (L. Galeano).

https://doi.org/10.1016/j.mssp.2018.10.032

1369-8001/ © 2018 Elsevier Ltd. All rights reserved.

Received 22 May 2018; Received in revised form 19 October 2018; Accepted 29 October 2018
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

by trapping sites for the photogenerated electrons [28,44]. TiO2) titanium tetraisopropoxide (C12H28O4Ti, 98%, Merck) (TIOT), 2-
There are several methods to incorporate ions into TiO2, such as propanol (C3H8O, 99.8%, Merck), hydrochloric acid (HCl, 37%, Merck),
ammonolysis, sol-gel, hydrothermal and solvothermal reactions, sput- tetraethyl orthosilicate (TEOS) (SiC8H20O4, 99%, Merck), urea
tering, and ion implantation [6,67]. Among them, the sol-gel method is (CH4N2O,99%, Carlo Erba), selenium dioxide (sublimed) for synthesis
the most used [1]. The sol-gel method allows the formation of nanos- (SeO2, 99.9%, Merck) and Milli-Q water were used. phenol (C6H6O,
tructured material with controlled porosity and shape [53]. Never- 99%, Panreac) was used to evaluate the photocatalytic activity of the
theless, for an effective doping through sol-gel method, longer gelifi- prepared materials. Titanium dioxide (TiO2) P25 (Evonik) was used as a
cation time and a high-temperature of calcination are required [3]. A commercial photocatalyst.
high-temperature calcination leads to a lost in the surface area [45]. To
overcome these limitations, the sol-gel method has been combined with 2.2. Synthesis of TiO2 and doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2) by
the solvothermal technique for the synthesis of bare and doped TiO2 the integrated sol-gel and solvothermal method
[37,55,57]. This constitutes an integrated method that allows obtaining
a photocatalyst in shorter time and lower energy consumption com- TiO2 synthesis was performed using 8.2 mL of TIOT and 1.9 mL of
pared to the sol-gel method. water. They were added slowly to 45.8 mL of 2-propanol (molar ratio:
The physical mixture of TiO2 with other narrow band gap semi- water/TIOT = 3.5 and 2-propanol/TIOT = 20) stirring for 2 h at room
conductors is another way to extend the absorption spectrum of light temperature. For the crystallization reaction, the obtained gel was
and enhance the photocatalytic activity of TiO2 [21]. However, Hu acidified with 0.5 mL of HCl (3 mol/L) and placed in a Teflon-lined
et al. [21] found that photocatalytic activity of TiO2 under illumination stainless-steel reactor (autoclave) at 225 °C for 2 h. After that, the so-
with UV light was increased by physical mixture of sulfur and TiO2, obtained precipitates were filtered using a 0.45 µm filter and washed
although this physical mixture did not lead to an improvement in with distilled water. Finally, the obtained materials were dried at
photocatalytic activity under visible light. 100 °C for 2 h and ground.
On the other hand, the photocatalytic activity of TiO2 was increased Doped-TiO2 synthesis (N-TiO2, Si-TiO2, and Se-TiO2) was performed
by a mechanical treatment such as grinding process at high rotation using 50 mg of urea (0.5% N), 0.18 mL of TEOS (0.5% Si), or 33 mg of
speeds (mechanochemical process) [25,62,65,67]. The mechanical selenium dioxide (0.5% Se) as precursor in every case. Each precursor
forces during the grinding process lead to changes in the structural was added to 45.8 mL of 2-propanol under stirring. After that, 8.2 mL of
characteristics of the catalyst, such as the morphology of the more TIOT and 1.9 mL of water were added slowly to the mix of 2-propanol
amorphous surface, and the increased specific surface area. In addition, and precursor every 30 min for 2 h at room temperature (hydrolysis
mechanical forces could activate a portion of the electrons within the reaction). The crystallization and drying processes were similar to those
crystal lattices with the emergence of a highly reactive crystal-face followed for the TiO2 synthesis.
[35]. On the other hand, the grinding process is associated with the
decrease in the recombination of photo-injection electrons [34]. 2.3. Dry-co-grinding of doped TiO2
The photocatalytic activity of TiO2 under visible light irradiation was
improved by grinding using different conditions, such as atmospheres The mixing and grinding of N-TiO2 with Si-TiO2 (N-TiO2/Si-TiO2),
(grinding in gaseous NH3), solvents (grinding in ethanol, hexane, water, and N-TiO2 with Se-TiO2 (N-TiO2/Se-TiO2) was obtained by a grinding/
etc.) or temperatures [24,25]. Another way of increasing photocatalytic mixing process (co-grinding process) for 30 min at 200 rpm with a
activity of TiO2 is the grinding of TiO2 with other semiconductors (co- sample mass/ball mass ratio = 1/20. The co-grinding process was
grinding process), such as the hybrid organic or inorganic compound that performed dry at room temperature (dry-co-grinding process) using a
uses appropriate materials to incorporate in TiO2 matrix [33]. Fritsch Planetary Mill PULVERISETTE 5/2 with two grinding bowls and
All these grinding processes were developed using high rotational grinding balls (diameter = 12 mm) made of stainless steel.
speeds and/or high grinding times, such as 780 rpm and 0.5–20 h [20],
700 rpm and 0.5–5 h [24], 700 rpm and 2 h [25], 200 rpm and 0–24 h 2.4. Characterization of materials and photocatalytic degradation analysis
[35], as well as 400 rpm and 10–60 h [2]. This implies a high-energy
consumption and process costs. On the other hand, the processes that X-ray diffraction (XRD) analyses were performed using a Siemens
integrate the strategies mentioned above with low energy consumption diffractometer model D-501 with radiation Kα of Cu (ʎ = 1.5418) and
to improve the properties and effectiveness of the TiO2 and respect the 2θ from 10° to 60° with a step width of 0.05° per second. Raman ana-
environment were not found in literature. lyses were performed in Horiba JobinYvon, model Labram HR 800
In this study, TiO2 doped with N, Si, and Se (N-TiO2, Si-TiO2, and Microraman. Specific surface areas of materials were determined by the
Se-TiO2) was synthesized by the integrated sol-gel and solvothermal Brunauer-Emmet-Teller (BET) method using N2 phyisisorption at 77 K,
method with short time of crystallization (2 h) and low temperature with a Micromeritics Gemini V 2380 instrument. All samples were de-
(225 °C) to obtain high purity anatase and small particles with a large gassed at 150 °C for 1 h in a Micromeritics Vac Preo 06 Kbars and
surface area. The doped TiO2 N-TiO2 was mixed and dry-ground with purged with nitrogen for 2 h at 150 °C, prior to the surface area mea-
Si-TiO2 or Se-TiO2 in a ball-milling process for 30 min at 200 rpm (dry- surements.
co-grinding process), reducing the time and energy demand. The dry-co- X-ray fluorescence (XRF) analysis was performed on an XRF spec-
grinding process allowed to obtain N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 trometer with wavelength-dispersive Axios, Panalytical, with rhodium
catalysts with high UV/visible and visible light absorption and higher anode as radiation source. Scanning electron microscopy (SEM) ana-
photocatalytic activities. TiO2, doped TiO2, and dry-co-grinding mate- lysis was obtained on a Hitachi S-4800 microscope equipped with an
rials were characterized by XRD, Raman, specific surface area and energy-dispersive X-ray spectroscopy (EDS) system. Transmission
porosity, XRF, SEM, TEM, FTIR-ATR, and UV/vis-DRS analyses. The electron microscopy (TEM) analysis was performed in a transmission
materials were evaluated in the photocatalytic degradation of phenol electron microscope (Tecnai G2 F20) using a FP 5360/22 HAADF de-
under UV/visible and visible light irradiation. tector for 120/200 kV.
Fourier transform spectroscopy with attenuated total reflectance
2. Materials and methods (FTIR-ATR) analyses were performed in a Prestige-21/FTIR-8400S
spectrophotometer, Shimadzu, with attenuated total reflectance (ATR)
2.1. Materials device. The spectrum was taken in the range of 4000–500 cm−1. UV/
visible diffuse reflectance spectra (UV/vis-DRS) were performed on a
For the synthesis of TiO2 and doped TiO2 (N-TiO2, Si-TiO2, and Se- UV/visible Spectrophotometer Evolution 600, Thermo, which includes

48
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

the integrating sphere accessory for diffuse reflectance mode. UV/ TiO2, Si-TiO2, and Se-TiO2) and dry-co-grinding samples (N-TiO2/Si-
visible absorbance of the photocatalytic degradation of phenol was TiO2 and N-TiO2/Se-TiO2). All samples synthesized display the mono-
estimated with a spectrophotometer (Thermo Evolution 600) at 270 nm phasic anatase (JCPDS 01–089–4921). Some of the major peaks for the
corresponding to the wavelength of maximum absorbance. All analyses anatase phase were observed at 2θ = 25.28°, 36.94°, 48.05°, 55.06°,
were acquired in triplicate. and 62.69°. In addition, peaks associated with rutile phase were not
observed (JCPDS 01–086–0147). Therefore, the integrated sol-gel and
solvothermal method and the dry-co-grinding process produced a
2.5. Photolysis, adsorption and photocatalysis experiments
monophasic anatase in bare, doped and ground TiO2 materials. Ad-
ditionally, there were not characteristic peaks associated with oxides or
The experiments with TiO2 samples were evaluated in a 450 mL
other compounds of N, Si or Se. These dopants should have been em-
batch borosilicate glass reactor equipped with an aluminum housing
bedded into the anatase-crystalline phase of TiO2 [28,50,59].
light reflector on top, a magnetic stirring, and an aeration system that
For doped samples (N-TiO2, Si-TiO2 and Se-TiO2) the XRD patterns
pumps air through the suspension to maintain a saturated concentration
showed peaks with higher intensities and the narrowest widths than
of dissolved oxygen. For irradiation under UV/visible light, a sunlight
those obtained for TiO2, where Si-TiO2 presented the anatase peak with
simulation lamp (Osram, Ultra-Vitalux, 300 W) with a nominal voltage
the highest intensity and the narrowest width among doped TiO2 ma-
of 230 V was used. The sunlight simulation lamp was placed on top of
terials. The XRD patterns of the dry-co-grinding samples (N-TiO2/Si-
the reactor. For irradiation under visible light, a filter of polyester film,
TiO2 and N-TiO2/Se-TiO2) showed the characteristic anatase peaks with
0.04 mm thickness, 400 nm cut-off wavelength (Edmund Optics) was
higher intensities and narrower widths than those obtained for doped
added. This filter absorbs UV radiation and allows less than 10%
TiO2, where N-TiO2/Si-TiO2 presented the narrowest width and the
transmission below 390 nm. The intensity of the incident light was
highest intensity of the anatase peaks. As indicated below, these results
around 150 W/m2 for a wavelength higher than 400 nm. The radio-
are associated with the increase in particle size.
metric measurements were performed with an UV-A radiometer, Solar
Brookite or rutile phase were not formed by the dry-co-grinding
Light Co (Model PMA 2210) with an UV-A analogue measuring detector
process for 30 min at 200 rpm. This result is contrary to Hidalgo et al.
PMA 2110 for the range of 320–400 nm, and a PMA 21300 for the range
[20] and Kang et al. [24] who found the anatase-phase transformation
of 400–700 nm. For the experiments, the batch reactor was loaded with
to rutile with the grinding process of bare TiO2. Moreover, they found
230 mL of phenol solution (50 mg/L).
that the intensity of the anatase peak was decreasing as the grinding
For the photolysis experiments, the effect of UV/visible or visible
time increased. However, these grinding processes were performed with
light irradiation in the absence of TiO2 was performed. In the suspen-
rotation speeds higher than 700 rpm and during longer grinding times.
sion experiments, for each experiment 0.5 g/L of each photocatalyst
On the other hand, Kang et al. [25] and Ali [2] found that the grinding
was loaded in the glass reactor and 230 mL of phenol (50 mg/L), and
process with a high-energy ball mill and the use of a solvent did not
the stirrer system was turned on. For the adsorption experiments, the
change the phase constitutions and the crystallinity of N-doped TiO2
TiO2 suspension was kept in the dark to reach the equilibrium phenol
compared to TiO2 synthesized. This behavior was also obtained in this
adsorption on the photocatalysts. For the photocatalysis experiments,
study through a low energy grinding and without the use of solvents
the lamp was switched on after of the equilibrium adsorption. 5 samples
(dry milling).
of 3 mL of the reaction volume were taken at different times of pho-
Fig. 1b shows the Raman patterns for synthesized TiO2, doped TiO2
tocatalytic degradation (30, 45, 60, 90, and 120 min). The photo-
(N-TiO2, Si-TiO2, and Se-TiO2) and dry-co-grinding samples (N-TiO2/Si-
catalyst was immediately removed from the reaction medium and fil-
TiO2 and N-TiO2/Se-TiO2). The Raman modes in 144 cm−1 (Eg),
tered using 0.45 µm Millipore Millex 25 membranes filters before
197 cm−1 (Eg), 399 cm−1 (B1g), 513 cm−1 (A1g), 519 cm−1 (B1g), and
analysis. In this study, all photocatalytic experiments were performed
639 cm−1 (Eg) are attributed to the anatase phase [11]. Moreover, the
in triplicate.
Raman modes in 141 cm−1 (B1g), 444 cm−1, 608 cm−1 which are at-
tributed to the main Eg and A1g rutile vibration modes were not ob-
3. Results and discussion served [5]. These results corroborate the DRX patterns.
The main Eg Raman mode at 144 cm−1 shows a decrease in the
3.1. XRD and Raman analysis intensity of doped and dry-co-grinding samples. The decrease in the
intensity of the main Eg Raman mode with the increase of the amount of
Fig. 1a shows the XRD patterns for synthesized TiO2, doped TiO2 (N-

Fig. 1. XRD (a) and Raman spectra (b) of synthesized TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-
TiO2), during 30 min at 200 rpm.

49
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

Table 1
Physical properties of TiO2 P25, TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-
TiO2), and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2),
during 30 min at 200 rpm.
Catalyst SBET Crystal size Pore volume Pore size Eg (eV)
(m2/g) (nm) (cm3/g) (nm)

TiO2 P25 50 20.4 – – 3.23

TiO2 152 7.3 0.24 5.11 3.21
N-TiO2 137 6.2 0.30 8.27 3.18
Se-TiO2 136 5.1 0.27 8.09 3.11
Si-TiO2 133 8.3 0.28 8.56 3.11
N-TiO2/Si- 128 9.3 0.17 5.48 2.94
TiO2
N-TiO2/Se- 128 8.6 0.18 5.64 2.92
TiO2

Se of doped TiO2 material has been reported [28]. On the other hand,
the location of the main Eg Raman mode in doped and dry-co-grinding
samples remains almost unchanged in comparison with TiO2. This in-
dicates that there were no significant distortions of the crystal lattice.
The crystal sizes were calculated by the Scherrer formula according
to the full width at half maximum of the diffraction pattern [31],
Table 1. The integrated sol-gel and solvothermal method led to mate-
rials (TiO2, N-TiO2, Si-TiO2, and Se-TiO2) with a smaller crystallite size,
in the range of nanometers in comparison with TiO2 P25 Evonik. On the
other hand, N-TiO2 and Se-TiO2 crystallite size decreased in comparison
with the obtained for the TiO2. This is in accordance with literature
reports where there is a decrease in the crystallite size of the materials
modified with N or Se with the increase of the amount of these elements
[28,9]. On the other hand, the dry-co-grinding process led to an in-
crease of the crystallite size in both materials (N-TiO2/Si-TiO2 and N-
TiO2/Se-TiO2). In addition, the crystallite size is more defined in N-
Fig. 2. Pore size distributions (a), and N2 adsorption/desorption isotherms (b)
TiO2/Si-TiO2 and N-TiO2/Se-TiO2 materials compared to TiO2 or
of TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), and dry-co-grinding sam-
doped-TiO2. The increase in the crystalline size of N-TiO2/Si-TiO2 and
ples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) during 30 min at 200 rpm.
N-TiO2/Se-TiO2 may be due to the agglomeration of particles after the
dry-co-grinding process.
Kang et al. [26] found that crystallite size gradually decreased with grinding process for 30 min at 200 rpm for N-TiO2/Si-TiO2 and N-TiO2/
an increase in the grinding time. Moreover, these researchers found that Se-TiO2. This could be due to the tendency of the particles to agglom-
heating and the addition of ethanol resulted in the prevention of the erate with the dry-co-grinding processor of the mesopores, Fig. 2(a).
agglomeration of TiO2 particles. However, after a prolonged grinding Saito et al. [46] stated that in the initial phase of grinding, the specific
there was an increase in the size of the crystallite materials. Ali [2] also area increases with an increase in grinding time until reaching a critical
reported on crystal growth and attributed it to the coalescence process point where the surface area decreases. This is due to the agglomeration
to form larger particles to eliminate the excess energy supplied. of fine particles. Moreover, the decrease in the specific area could be
also due to the densification of the secondary particles [20], which
produced a loss of the largest diameter pores in N-TiO2/Si-TiO2 and N-
3.2. Specific surface area analysis and porosity properties
TiO2/Se-TiO2.
Fig. 2(a) shows the pore size distributions of the bare-TiO2, N-TiO2,
Table 1 shows the surface area, crystal size, pore volume, and pore
Si-TiO2, Se-TiO2, N-TiO2/Si-TiO2, and N-TiO2/Se-TiO2. The pore size
size of the doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), and the dry-co-
distributions were calculated by the BJH (Barret-Joyner-Halanda)
grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2). The surface
method. The maximum pore volume of doped-TiO2 was shifted to larger
area was greater in all the catalysts in comparison with the TiO2 P25
pore diameters in comparison with bare-TiO2. The doped materials
Evonik. However, doping of TiO2 with N, Si, or Se (N-TiO2, Si-TiO2, and
showed a higher content of mesopores of main size 7–9 nm. After the
Se-TiO2) led to a reduction in the surface area compared to the obtained
dry-co-grinding process during 30 min at 200 rpm, the maximum of the
for bare synthesized TiO2.
pore volume of the N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 suffered a shift
Doping of TiO2 can lead to an increase or decrease in the surface
to lower pore diameters in comparison with doped TiO2. This is due to
area of bare TiO2, depending of the methods of synthesis, doping,
the collapse of the largest pores and a slight reduction in the surface
content of dopant, and solvents used. Yin et al. [61] reported that the
area. N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 showed mesopores mainly
specific area of N-doped TiO2 changed greatly according to the pH and
between 3 nm and 4 nm. Therefore, the dry-co-grinding process during
the type of solvents used in the solvothermal treatment. For Si, Jin et al.
30 min at 200 rpm contributed to the transformation of larger pores to
[23] found that the surface area of Si-TiO2 could be increased or de-
lower pores. Additionally, this transformation was obtained by using
creased by increasing the concentration of Si. Xiao et al. [59] found that
less energy than that reported in literature.
the doping TiO2 with Si increased the surface area of bare TiO2 by in-
Fig. 2(b) shows the N2 adsorption/desorption isotherms of TiO2,
creasing the Si concentration. However, Estruga et al. [15] found that
doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), and dry-co-grinding sam-
doping of TiO2 with Si reduced the surface area with respect to bare-
ples during 30 min at 200 rpm (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2).
TiO2. For Se, Khan and Berk [28] found that the surface area of Se-TiO2
The TiO2, N-TiO2, Si-TiO2, and Se-TiO2 exhibited the typical IUPAC
can be increased or decreased depending on the concentration of Se.
type IV with a type H2 hysteresis loop, which suggests the presence of
A decrease in the specific area was also observed after the dry-co-

50
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

Table 2 loop that has been associated with the existence of aggregates of par-
XRF and EDX/TEM analyses of doped TiO2 (N-TiO2, Si-TiO2 and Se-TiO2), and ticles with plate-like forms and a more pronounced uptake. This is
dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2), during 30 min perhaps due to the filling or collapse of larger micropores [51,59].
at 200 rpm.
Catalyst XRF EDX/TEM 3.3. XRF analysis
% Si % Se %N % Si % Se
Table 2 shows XRF analyses of doped TiO2 (Si-TiO2 and Se-TiO2),
N-TiO2 – – 8.22 – – and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2). The
Si-TiO2 0.46 – – 1.21 – chemical composition of these doped-TiO2 materials are in agreement
Se-TiO2 – 0.54 – – 0.19
with the nominal concentration of 0.5% of Si and Se in the doped-TiO2
N-TiO2/Si-TiO2 0.44 – 8.00 0.24 –
N-TiO2/Se-TiO2 – 0.51 7.97 – 0.02 synthesis. This result indicates that the integrated sol-gel and sol-
vothermal method allowed incorporating a large amount of the added
dopants (Si and Se). Additionally, the dry-co-grinding process during
mesopores with greater complexity, partially defined or blocked. 30 min at 200 rpm kept the amount of these elements unchanged and
Moreover, The TiO2, N-TiO2, Si-TiO2, and Se-TiO2 contain a small did not contribute to the leaching of the dopant.
number of micropores [15]. The pore size distributions confirm the
mesoporosity of these TiO2 samples. In addition, N-TiO2/Si-TiO2 and N- 3.4. Morphological analysis (SEM and TEM)
TiO2/Se-TiO2 obtained by the dry-co-grinding process during 30 min at
200 rpm exhibited the Brunauer IV type, but with a type H3 hysteresis Fig. 3 shows the SEM micrographs of synthesized TiO2, doped TiO2

Fig. 3. SEM micrographs of TiO2 (a), doped TiO2 (N-TiO2 (b), Si-TiO2 (c), and Se-TiO2 (d)), and dry-co-grinding samples (N-TiO2-/Si-TiO2 (e) and N-TiO2/Se-TiO2
(f)), during 30 min at 200 rpm.

51
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

Fig. 4. TEM micrographs of doped TiO2 (N-TiO2 (a), Si-TiO2 (b), Se-TiO2 (c)), and dry-co-grinding samples (N-TiO2-/Si-TiO2 (d) and N-TiO2/Se-TiO2 (e)), during
30 min at 200 rpm.

(N-TiO2, Si-TiO2, and Se-TiO2), and dry-co-grinding samples (N-TiO2/ were present on the surface of the doped TiO2 and dry-co-grinding
Si-TiO2 and N-TiO2/Se-TiO2). Synthesized TiO2 shows a rougher sur- samples.
face. Si-TiO2 and Se-TiO2 powders show larger agglomerates than N-
TiO2. On the other hand, the dry-co-grinding samples (N-TiO2/Si-TiO2
3.5. FTIR-ATR analysis
and N-TiO2/Se-TiO2) show a more uniform spherical morphology.
However, after the dry-co-grinding process, N-TiO2/Si-TiO2 and N-
Fig. 5 shows the FTIR-ATR spectra of TiO2, doped TiO2 (N-TiO2, Si-
TiO2/Se-TiO2 consisted of larger agglomerates that also have associated
TiO2, and Se-TiO2), and dry-co-grinding samples (N-TiO2/Si-TiO2 and
fine particles.
N-TiO2/Se-TiO2). The spectra exhibit the characteristic bands for TiO2.
Fig. 4 shows TEM micrographs of doped TiO2 (N-TiO2, Si-TiO2, and
The broad band of approximately 3300 cm−1 is attributed to the
Se-TiO2) and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-
stretching vibrations of -OH groups due to surface hydroxylation
TiO2). The TEM of the doped materials (N-TiO2, Si-TiO2 and Se-TiO2)
[28,60]. The presence of OH in all samples favors the generation of
show that the particle sizes tend to be very uniform (Figs. 4a, 4b, and
oxidizing species such as •OH [33,60]. The peak at 1630 cm−1 is at-
4c). Moreover, the particle sizes are in agreement with those estimated
tributed to the molecular water (HOH) absorbed on the TiO2 surface
by the DRX analysis. The particle size of the dry-co-grinding samples (N-
[4,41].
TiO2/Si-TiO2 and N-TiO2/Se-TiO2) ranges from the sizes of the in-
The small peaks at 1055 or 1079, and 1355 or 1441 cm−1 in the
dividual doped particles to values close to 20 nm. This shows that the
doped TiO2 samples (N-TiO2, Si-TiO2, and Se-TiO2) and the dry-co-
dry-co-grinding process led to agglomeration of the individual particles,
grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) may be due to
although some particles remained disaggregated (Figs. 4d and 4e). The
the Ti-O-N, Ti-O-Si, and Ti-O-Se bands [52,60,66,7], respectively. The
materials that showed a greater aggregation were those with Si.
peak band at 930 or 952 cm−1 is attributed to the stretching vibration
The EDX/TEM analysis (Table 2) shows the doping elements (N, Si,
of Ti-O-Si or Si-O-Si bonds in Ti- that would imply a possible sub-
and Se) percentages on the surface of the materials. The amount of N
stitution of Si for Ti [30,60,66,7]. In addition, this implies that not only
remained in both the doped TiO2 (N-TiO2) and dry-co-grinding samples
Ti-O-Ti and Si-O-Si but also Ti-O-Si bands were formed during the
(N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2). Moreover, Si and Se were also
process [23].
detected in both the doped TiO2 (Si-TiO2, Se-TiO2) and the dry-co-
The bond at 550 cm−1 is attributed to the Ti-O stretching vibrations
grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2). However, the
of the crystalline TiO2 phase. The band located at about
percentages of these doping elements by EDX/TEM analysis were the
1200–1300 cm−1in the spectra of the Se-TiO2 and N-TiO2/Se-TiO2 is
average of some measurements in a specific surface area of the samples.
attributed to the vibration of the stretching Se [28]. The N-TiO2/Si-TiO2
Therefore, these percentages only confirm that the doping elements
and N-TiO2/Se-TiO2 materials presented bands at 3650–3850 cm−1

52
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

3.6. XPS analysis

Fig. 6 shows the XPS spectra of TiO2, N-TiO2/Si-TiO2 and N-TiO2/

Se-TiO2. The binding energies for Ti (2p3/2) and Ti (2p1/2) correspond
to Ti4+ for all materials (Fig. 6a). Slight shifts to higher values of
binding energies with respect to the binding energies of TiO2 are ob-
served. It has been reported that these shifts are due to the greater
electronegativity of the doping elements acting on the Ti. [23]. Fig. 6b
shows the N 1s XPS spectra of N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2. For
both materials, very weak signals are observed although they are more
defined in the N-TiO2/Se-TiO2 XPS spectrum. The main observable peak
is slightly defined at 400 eV, which has been reported to correspond to
the N-O/N-N binding energy [25].
The XPS analysis of the material N-TiO2/Si-TiO2 in the Si 2p zone
(Fig. 6c) shows a peak at 101.1 eV, which has been associated with the
Fig. 5. FTIR-ATR spectra of TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), Si4+ that has a binding energy at 103.3 eV [59]. This peak shift has
and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2), during been reported as a possible consequence of an oxygen loss in SiO2 [23].
30 min at 200 rpm.
Se 3p XPS analysis in N-TiO2/Se-TiO2 material (Fig. 6d) indicates that
this element was possibly incorporated in the state Se4+ according to
that have been associated with the hydroxylation of the TiO2 surface the main peak (61.9 eV). Moreover, a deconvolution of this peak shows
[19], which were more defined in comparison to the bands of the doped a peak attached at 65.6 eV, which is associated with the binding energy
TiO2, TiO2 and TiO2 P25 Evonik. of the Ti 3 s [28,44].

Fig. 6. XPS spectra of Ti 2p of TiO2 and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) (a), N 1s of dry-co-grinding samples (N-TiO2/Si-TiO2 and N-
TiO2/Se-TiO2) (b), Si 2p of N-TiO2/Si-TiO2 material (c) and Se 3p of N-TiO2/Se-TiO2 material (d).

53
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

could imply that the dry-co-grinding process contributes to a better

incorporation of the doping elements (N, Si or Se) in the TiO2. The
absorption in the visible light spectrum trend was N-TiO2/Se-TiO2 > N-
TiO2/Si-TiO2 > Se-TiO2 > N-TiO2 > Si-TiO2 > TiO2 > TiO2 P25.
For the study of the optical properties of the doped TiO2 and dry-co-
grinding samples the band-gap was determined. The band-gap was
calculated by means of the optic absorption spectrum, and the remis-
sion function of Kubelka-Munk was used [56]. The indirect type tran-
sition analysis led to band-gap values of TiO2, doped TiO2, and dry-co-
grinding samples (Table 1).
The band-gap of TiO2 synthetized by the integrated sol-gel and
solvothermal method was similar to the TiO2 P25 Evonik. The doping of
TiO2 with N led to a slight reduction in the band gap from 3.21 eV to
3.18 eV in comparison with TiO2. Bergamonti et al. [6] synthesized and
doped TiO2 using TIOT and urea by the sol-gel method, and they found
a slight reduction in the band gap from 3.25 to 3.10 eV for TiO2 P25 and
N-TiO2, respectively; while a greater reduction in the band gap was
Fig. 7. UV/vis-DRS spectra of TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-TiO2), found using TiOSO4 and NH4OH, from 3.25 to 3.02 eV. Therefore, Ti-
and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2), during precursors, N-sources, and synthesis methods affect the reduction of N-
30 min at 200 rpm. TiO2 band gap. On the other hand, Si and Se led to a reduction in the
band gap of Si-TiO2 and Se-TiO2 in comparison with TiO2 (from 3.21 to
3.7. UV/vis-DRS analysis 3.11 eV). This is in agreement with Gurkan et al. [18] who obtained a
band gap of 2.76 eV for Se-TiO2 with 0.5 wt%. Moreover, they found a
Fig. 7 shows the UV/vis-DRS spectra of TiO2, doped TiO2 (N-TiO2, correlation between the band gap reduction of Se-TiO2 and the in-
Si-TiO2, and Se-TiO2), and dry-co-grinding samples (N-TiO2/Si-TiO2 creased concentration of Se.
and N-TiO2/Se-TiO2). The absorption of the visible light spectrum was The dry-co-grinding process during 30 min at 200 rpm led to a high
higher in the synthesized TiO2 than in the TiO2 P25. The integrated sol- reduction in band gap from 3.21 eV for TiO2 to 2.94 and 2.92 eV for N-
gel and solvothermal method led to enhance the absorption of syn- TiO2/Si-TiO2 and N-TiO2/Se-TiO2, respectively. Therefore, the dry-co-
thesized TiO2 in the visible light spectrum. On the other hand, all doped grinding process changed the optical properties of these materials (N-
TiO2 samples (N-TiO2, Si-TiO2, and Se-TiO2) and dry-co-grinding sam- TiO2/Si-TiO2 and N-TiO2/Se-TiO2). This was due to the fact that
ples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) had slightly lower absorp- grinding generates accumulations of defects such as oxygen vacancies
tion in the UV light region (200–350 nm) than TiO2 P25. in the structure of doped TiO2 [20,40]. And also because a better in-
Doping TiO2 with N from urea (CH4N2O), Si from tetraethyl or- corporation of the doping elements (N, Si or Se) in the TiO2 [24,62,67]
thosilicate (SiC8H20O4) or Se from selenium dioxide (SeO2) (N-TiO2, Si- increased the visible light absorption.
TiO2, and Se-TiO2) enhanced its absorption in visible light spectrum.
The light absorption for N-TiO2 is in agreement with the result of Du 3.8. Photolysis experiments of phenol
et al. [14], who found a high wavelength region absorption for TiO2
after nitrification with ammonia (NH3) (doping TiO2 with N from am- The photolysis experiment in the absence of TiO2, doped TiO2 (N-
monia). This was attributed to the shallow energy formation between TiO2, Si-TiO2, and Se-TiO2) or dry-co-grinding samples (N-TiO2/Si-TiO2
the conduction and valence bands. Xiao et al. [59] reported a red shift and N-TiO2/Se-TiO2), with 230 mL of phenol solution (50 mg/L) was
in the light absorption in TiO2 doping with Si, which was attributed to a performed using the photocatalytic reactor under UV/visible or visible
lower energy requirement for the electron movement from the valence light, during 180 min of irradiation. There was no change in the con-
band to the conduction band. Khan and Berk [28] reported that loca- centration of phenol in the photolysis experiment (data not shown). The
lized gap states (Se 4s) may induce visible light absorption in Se-doped experiment was performed in triplicate. The repeatability was below
TiO2 samples, while Gurkan et al. [18] reported that an increase in light 5% of the relative standard deviation.
absorption in Se-doped TiO2 generated a greater number of e-/h+ pairs
that led to surface reactions that increased the amount of hydroxyl 3.9. Adsorption experiments in the dark of phenol onto TiO2, doped TiO2,
radicals. and dry-co-grinding samples
Dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) en-
hanced the absorption in the visible light spectrum with respect to Adsorption experiments in the dark with 230 mL of phenol solution
doped TiO2, where N-TiO2/Se-TiO2 showed the highest visible light (50 mg/L), and 0.5 g/L of the photocatalysts in suspension showed that
absorption in the region from 400 to 600 nm. The reason for enhanced equilibrium of phenol on TiO2, doped TiO2 (N-TiO2, Si-TiO2, and Se-
adsorption of N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 in visible light TiO2), or dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-
spectrum could be probably attributed to a defect level in the TiO2 TiO2) was reached in less than 5 min (data not shown). An adsorption
formed during the dry-co-grinding process [49]. The dry-co-grinding time of 30 min was chosen to ensure saturation for all subsequent ex-
process increases the local temperature and the pressure at the collision periments. The amount of phenol adsorbed onto TiO2 in the dark was
sites, generating the formation of dislocations and microcracks [39,40]. less than 1% for all examples. The experiment was performed in tri-
Moreover, the increase in visible absorption may be due to a possible plicate. The repeatability was below 5% of the relative standard de-
generation of oxygen vacancies in the TiO2 [20], which are attributed viation.
to a release of oxygen from the titania lattice when some of the Ti-O
bonds in the surface layer are broken by the grinding process [40]. 3.10. Photocatalytic degradation experiments on phenol with TiO2, doped
On the other hand, Kang et al. [24] found that the absorption in the TiO2, and dry-co-grinding samples in suspension
visible range of TiO2 can be increased by increasing the grinding per-
iods. This is due to the mechanical process that contributes to the in- Fig. 8a shows the photocatalytic degradation under visible light
crease in the amount of doping of TiO2 [25,62,67]. Therefore, this irradiation, with 230 mL of phenol (50 mg/L), and with 0.5 g/L of the
photocatalyst (TiO2, N-TiO2, Si-TiO2, Se-TiO2, N-TiO2/Si-TiO2, and N-

54
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

Fig. 8. Photocatalytic degradation activity of phenol with visible light irradiation (a) and UV/visible light irradiation (b) using TiO2, doped TiO2 (N-TiO2, Si-TiO2,
and Se-TiO2) and dry-co-grinding samples (N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2) in suspension.

TiO2/Se-TiO2). Besides 0.5 g/L of TiO2 P25 as a reference. The experi- The photocatalyst activity trend under UV/visible light irradiation
ments were performed in triplicate. The repeatability was below 5% of was N-TiO2/Si-TiO2 > N-TiO2/Se-TiO2 > TiO2 > TiO2 P25. TiO2
the relative standard deviation. sample showed a high phenol photocatalytic degradation as the ob-
The photocatalyst activity trend under visible light irradiation was N- tained with TiO2 P25 under UV/visible light irradiation. In addition, N-
TiO2/Si-TiO2 ≈ N-TiO2/Se-TiO2 > N-TiO2 > Se-TiO2 > Si-TiO2 > TiO2 TiO2/Si-TiO2 and N-TiO2/Se-TiO2 showed a higher phenol degradation
≈ TiO2 P25. Si and Se dopants into TiO2 led to a slight increase in the in comparison with TiO2. On the other hand, the photocatalytic activity
photocatalytic degradation of phenol under visible light irradiation in of physical mixture of N-TiO2 with Si-TiO2 and N-TiO2 with Se-TiO2
comparison with synthesized TiO2 or TiO2 P25, with an amount of Si or Se (without the dry-co-grinding process) did not lead to an improvement in
doped TiO2 of 0.5%. On the other hand, N-TiO2 showed a greater pho- the photocatalytic activity under UV/visible light irradiation. These
tocatalytic activity in comparison to Si-TiO2 or Se-TiO2, although the ab- results were consistent with those obtained under visible radiation.
sorption of visible light by N-TiO2 was lower than that absorbed by Se- Therefore, the improving of photocatalytic activity of N-TiO2/Si-TiO2
TiO2 but higher than the absorption achieved by Si-TiO2. This shows that and N-TiO2/Se-TiO2 can be associated with the indicated properties of
this property does not always lead to higher photocatalytic activity. These these materials after dry-co-grinding process.
results are in agreement with Nagaveni et al. [38] who has reported that
doping with some metal ions into TiO2 was not beneficial, even though all 4. Conclusions
the doped TiO2 absorbed in the visible region.
The dry-co-grinding process during 30 min at 200 rpm contributed The dry grinding/mixing process, during 30 min at 200 rpm (dry-co-
to obtain materials with higher photocatalytic activity compared with grinding process), over doped TiO2 photocatalyts materials that were
TiO2, doped TiO2 or TiO2 P25 (Fig. 8a). This is perhaps due to the synthesized by the integrated sol-gel and solvothermal method (N-TiO2,
smallest band gap, which is in agreement with the UV/visible-DRS Si-TiO2 and Se-TiO2), led to obtain materials (N-TiO2/Si-TiO2 and N-
spectra (Fig. 7), where N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 had shown TiO2/Se-TiO2) with high specific area, anatase phase, and a reduction in
the best visible light absorption in the region from 400 to 600 nm. On the band gap values. Besides an improvement in absorption in the
the other hand, some researchers have shown that particle size, surface visible light spectrum, and with a higher UV/visible and visible light
area, and phase composition strongly influence the photocatalytic ac- response. On the other hand, N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 ma-
tivity. These properties improve absorbability of hydroxyl/water on terials exhibited higher photocatalytic activity for degradation of
surface [26,63]. phenol under UV/visible and visible light irradiation than TiO2 P25,
The properties of N-TiO2/Si-TiO2 and N-TiO2/Se-TiO2 materials synthesized TiO2 or doped TiO2 with N, Si, and Se. Therefore, the in-
such as high visible light absorption, low band gap values, high surface tegrated sol-gel and solvothermal method (at short crystallization time
hydroxylation, the co-doping, possibly the amount of defect levels, (2 h) and low temperature (225 °C) and the dry-co-grinding process
oxygen vacancies, and oxygen-rich surface formed during the dry-co- during 30 min at 200 rpm (a process with low energy consumption)
grinding process contributed to a high photocatalytic activity. On the contributed to modify the optical and surface properties of TiO2 pho-
other hand, the photocatalytic activity of physical mixture of N-TiO2 tocatalysts with a higher visible light response.
with Si-TiO2 and N-TiO2 with Se-TiO2 (without the dry-co-grinding
process) did not lead to an improvement in the photocatalytic activity Acknowledgements
under visible light irradiation.
Fig. 8b shows the photocatalytic degradation under UV/visible light The authors thank Universidad de Antioquia and COLCIENCIAS for
irradiation, during 120 min of irradiation, with 230 mL of phenol their financial support.
(50 mg/L) and 0.5 g/L of TiO2, N-TiO2/Si-TiO2, and N-TiO2/Se-TiO2 in
suspension with TiO2 P25 as a reference. The experiments were per- References
formed in triplicate. The repeatability was below 5% of the relative
standard deviation. [1] U. Akpan, B.H. Hameed, The advances in sol-gel method of doped-TiO2

55
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

photocatalyst, Appl. Catal. A: Gen. 375 (2017) 1–11. [33] W. Mekprasart, N. Vittayakorn, W. Pecharapa, Ball-milled CuPc/TiO2 hybrid na-
[2] M. Ali, Transformation and powder characteristics of TiO2 during high-energy nocomposite and its photocatalytic degradation of aqueous rhodamine B, Mater.
milling, Ceram. Process. Res. 15 (2014) 290–293. Res. Bull. 47 (2012) 3114–3119.
[3] N. Aman, N.N. Das, T. Mishra, Effect of N-doping on visible light activity of TiO2- [34] W. Mekprasart, W. Jarernboon, W. Pecharapa, nanocomposites prepared by low-
SiO2 mixed oxide photocatalyst, J. Environ. Chem. Eng. 4 (2016) 191–196. energy ball Milling for dye-sensitized solar cell application, Mater. Sci. Eng. B 172
[4] B. Appavu, S. Thiripuranthagan, Visible active N, S codoped TiO2/graphite pho- (2010) 231–236.
tocatalyst for the degradation of hazardous dyes, J. Photochem. Photobiol. A: [35] J. Miao, R. Zhang, L. Zhang, Photocatalytic degradations of three dyes with dif-
Chem. 340 (2017) 146–156. ferent chemical structures using ball-milled TiO2, Mater. Res. Bull. 97 (2018)
[5] M.N. Asiah, M.H. Mamat, Z. Khusaimi, S. Abdullah, M. Rusop, Z. Qurashi, 109–114.
Structural and optical properties of hydrothermally synthesized mesoporous Si/ [36] A. Mills, C. O´Rourke, K. Moore, Powder semiconductor photocatalysis in aqueous
TiO2 nanowire composites, Microelectron. Eng. 136 (2015) 31–35. solution: an overview of kinetics-based reaction mechanisms, J. Photochem.
[6] L. Bergamonti, G. Predieri, Y. Paz, L. Fornasini, P.P. Lottici, F. Bondioli, Enhanced Photobiol. A: Chem. 310 (2015) 66–105.
self-cleaning properties of N-doped TiO2 coating for cultural heritage, Microchem. [37] J. Mosquera-Pretelt, M. Mejía, Juan M. Marín, Synthesis and characterization of
J. 133 (2017) 1–12. photoactive S-TiO2 from TiOSO4 precursor using an integrated sol-gel and sol-
[7] D. Bui, S. Kang, X. Li, J. Mu, Effect of Si doping on the photocatalytic activity and vothermal method at low temperatures, J. Adv. Oxid. Technol. 21 (2018), https://
photoelectrochemical property of TiO2 nanoparticles, Catal. Commun. 13 (2011) doi.org/10.26802/jaots.2017.0008.
14–17. [38] K. Nagaveni, M.S. Hagde, G. Madras, Structure and photocatalytic Activity of Ti1-
[8] X. Cheng, X. Yu, Z. Xing, J. Wan, Enhanced photocatalytic activity of nitrogen xMxO2 ± δ (M = W, V, Ce, Zr, Fe, and Cu) synthesized by solution combustion
doped TiO2 anatase nano-particles under simulated sunlight irradiation, Energy method, J. Phys. Chem. B 108 (2004) 20204–20212.
Procedia 16 (2012) 598–605. [39] S. Palmas, A.M. Polcaro, J. Rodriguez, A. Da, Effect of the mechanical activation on
[9] B. Chi, L. Zhao, T. Jin, One-step template-free route for synthesis of mesoporous N- the photoelectrochemical properties of anatase powders, Int. J. Hydrog. Energy 34
doped titania spheres, J. Phys. Chem. C 11 (2007) 6189–6193. (2009) 9662–9670.
[10] M. Chong, B. Jin, C. Chow, C. Saint, Recent development in photocatalytic water [40] X. Pan, X. Ma, Phase transformations in nanocrystalline TiO2 milled in different
treatment technology: a review, Water Res. 44 (2010) 2997–3027. milling atmospheres, J. Solid State Chem. 177 (2004) 4098–4103.
[11] H. Choi, Y. Jung, S. Kim, Size effects in the raman spectra of TiO2 nanoparticles, [41] E. Park, B. Jeong, M. Jeong, Y. Kim, Synergetic effects on hydrophilic surface
Vib. Spectrosc. 37 (2005) 33–38. modification and N-doping for visible light response on photocatalytic activity of
[12] H. Dong, G. Zeng, L. Tang, C. Fan, C. Zhang, X. He, Y. He, An overview on lim- TiO2, Curr. Appl. Phys. 14 (2014) 300–305.
itations of TiO2-based particles for photocatalytic degradation of organic pollutants [42] S. Phattalung, S. Limpijumnong, J. Yu, Passivated co-doping approach to band gap
and the corresponding countermeasures, Water Res. 79 (2015) 128–146. narrowing of titanium dioxide with enhanced photocatalytic activity, Appl. Catal.
[13] J. Du, X. Li, K. Li, X. Gu, W. Qi, K. Zhang, High hydrophilic Si-doped nanowires by B. Environ. 200 (2017) 1–9.
chemical vapor deposition, J. Alloy. Compd. 687 (2016) 893–897. [43] N. Rahimi, R. Pax, E. Gray, Review of functional titanium oxide. I: TiO2 and its
[14] J. Du, G. Zhao, Y. Shi, H. Yang, Y. Li, G. Zhu, Y. Mao, R. Sa, W. Wang, A facile modifications, Prog. Solid State Chem. 4 (2016) 86–105.
method for synthesis of N-doped TiO2 nanooctahedra, nanoparticles, and nano- [44] E. Rockafellow, J. Haywood, T. Witte, R. Houk, W. Jenks, Selenium modified TiO2
spheres and enhanced photocatalytic activity, Appl. Surf. Sci. 273 (2013) 278–286. and its impact on photocatalysis, Langmuir 26 (2010) 19052–19059.
[15] M. Estruga, C. Domingo, X. Doménech, J. Ayllón, Zirconium-doped and silicon- [45] S. Sahni, S. Reddy, B. Murty, Influence of process parameters on the synthesis of
doped TiO2 photocatalysts synthesis from ionic-liquid-like precursors, J. Colloid nano-titania by sol-gel route, Mater. Sci. Eng. A 452–453 (2007) 758–762.
Interface Sci. 344 (2010) 327–333. [46] F. Saito, M. Baron, J. Dodds, Morphology control in size reduction processes, in:
[16] W. Fang, M. Xing, J. Zhang, Modifications on reduced titanium dioxide photo- Y. Waseda, A. Muramatsu (Eds.), Morphology Control of Materials and
catalysts: a review, J. Photochem. Photobiol. C: Photochem. Rev. 32 (2017) 21–29. Nanoparticles, Springer Series in Materials Science, Berlin, Heidelberg, 2004, pp.
[17] M. Gmurek, M. Olak-Kucharczyk, S. Ledakowicz, Photochemical decomposition of 1–24.
endocrine disrupting compounds - a review, Chem. Eng. J. 310 (2017) 437–456. [47] T. Sano, N. Mera, Y. Kanai, C. Nishimoto, S. Tsutsui, T. Hirakawa, N. Negishi, Origin
[18] Y. Gurkan, E. Kasapbasi, Z. Cinar, Enhanced solar photocatalytic activity of TiO2 by of visible-light activity of N-doped TiO2 photocatalyst: behaviors of N and S atoms
selenium(IV) ion-doping: characterization and DFT modeling of surface, Chem. Eng. in a wet N-dopping process, Appl. Catal. B: Environ. 128 (2012) 77–87.
J. 214 (2013) 33–44. [48] W. Shi, Q. Chen, Y. Xu, D. Wu, C.F. Huo, Investigation of the silicon concentration
[19] J.M. Hernández, L.A. Cortez, R. García, E. San Martín, P. García, E. Brent, effect on Si-doped anatase TiO2 by first-principles calculation, J. Solid Chem. 184
G. Cárdenas, L.A. García, Synthesis of solid acid catalysts base on TiO2-SiO42- and (2011) 1983–1988.
Pt/TiO2-SO42- applied in n-Hexane isomerization, O. J. Metal. 3 (2013) 34–44. [49] C. Shifu, C. Lei, G. Shen, C. Gengyu, The preparation of coupled SnO2/TiO2 pho-
[20] M.C. Hidalgo, G. Colón, J. Navio, Modification of physicochemical properties of tocatalyst by ball milling, Mater. Chem. Phys. 98 (2006) 116–120.
commercial TiO2 samples by soft mechanical activation, J. Photochem. Photobiol. [50] P. Si, H. Wang, Z. Li, J. Liu, J. Lee, C. Choi, Large scale synthesis of nitrogen doped
A: Chem. 148 (2002) 341–348. TiO2 nanoparticles by reactive plasma, Mater. Lett. 161 (2012) 161–163.
[21] C. Hu, C. Lian, S. Zheng, S. Duo, R. Zhang, Q. Hu, S. Zhang, X. Li, Y. Sun, Grinding [51] M. Thommes, H. Kaneko, A. Neimark, J. Olivier, F. Rodríguez-Reinoso, J. Rouquero,
combined melt-diffusion synthesis of sulfur/P25 heterostructure for enhanced K. Sing, Physisorption of gases with special reference to the evaluation of Surface
photocatalytic activity under visible light, J. Mol. Catal. A: Chem. 407 (2015) area and pore size distribution (IUPAC Technical report), Pure Appl. Chem. (2015),
182–188. https://doi.org/10.1515/pac-2014-1117.
[22] T. Jia, F. Fu, D. Yu, J. Cao, G. Sun, Facile synthesis and characterization of N-doped [52] V. Tran, T. Truong, T. Phan, T. Nguyen, T. Huynh, A. Agresti, S. Pescetelli, T. Le,
TiO2/C nanocomposites with enhanced visible–light photocatalytic performance, A. Di Carlo, Application of nitrogen-doped TiO2 nano-tubes in dye sensitized solar
Appl. Surf. Sci. 438 (2018) 438–447. cells, Appl. Surf. Sci. 399 (2017) 515–522.
[23] R. Jin, Z. Wu, Y. Liu, B. Jiang, H. Wong, Photocatalytic reduction of NO with NH3 [53] Q. Truong, L. Dien, D. Vo, T. Le, Controlled synthesis of titanium using water-so-
using Si-doped TiO2 prepared by hydrothermal method, J. Hazard. Mater. 161 luble titanium complexes: a review, J. Solid State Chem. 251 (2017) 143–163.
(2009) 42–48. [54] S. Valencia, J. Marin, G. Restrepo, F.H. Frimmel, Evaluation of natural organic
[24] I. Kang, Q. Zhang, J. Kano, S. Yin, T. Sato, F. Saito, Synthesis of nitrogen doped TiO2 matter changes from Lake Hohloh by three-dimensional excitation/emission matrix
by grinding in gaseous NH3, J. Photochem. Photobiol. A: Chem. 189 (2007) fluorescence spectroscopy during TiO2/UV process, Water Res. 51 (2014) 124–133.
232–238. [55] S. Valencia, X. Vargas, L. Rios, G. Restrepo, J. Marín, Sol-gel and low-temperature
[25] I. Kang, Q. Zhang, S. Yin, T. Sato, F. Saito, Novel method for preparing of high solvothermal synthesis of photoactive nanotitanium dioxide, J. Photochem.
visible active N-doped TiO2 photcatalyst with its grinding in solvent, Appl. Catal. B: Photobiol. A: Chem. 251 (2013) 175–181.
Environ. 84 (2008) 570–576. [56] S. Valencia, J. Marín, G. Restrepo, Study of the band gap of synthesized titanium
[26] I. Kang, Q. Zhang, S. Yin, T. Sato, F. Saito, Preparation of a visible sensitive carbon dioxide nanoparticles using the sol-gel method and a hydrothermal treatment, Open
doped TiO2 photo-catalyst by grinding TiO2 with ethanol and heating treatment, Mater. Sci. J. 4 (2010) 9–14.
Appl. Catal. B: Environ. 80 (2008) 81–87. [57] X. Vargas, E. Tauchertb, J.M. Marín, G. Restrepo, R. Dillert, D. Bahnemann, Fe-
[27] M. Khaki, M. Shafeeyan, A. Raman, W. Daud, Application of doped photocatalysts doped titanium dioxide synthesized: photocatalytic activity and mineralization
for organic pollutants degradation – a review, J. Environ. Manag. 198 (2017) study for azo dye, J. Photochem. Photobiol. A: Chem. 243 (2012) 17–22.
78–94. [58] J. Wen, X. Li, W. Liu, Y. Fang, J. Xie, Y. Xu, Photocatalysis fundamentals and surface
[28] H. Khan, D. Berk, Selenium modified oxalate chelated titania: characterization, modification of TiO2 nanomaterials, Chin. J. Catal. 36 (2015) 2019–2070.
mechanistic and photocatalytic studies, Appl. Catal. A: Gen. 505 (2015) 285–301. [59] J. Xiao, Z. Pan, B. Zhang, G. Liu, H. Zhang, X. Song, G. Hu, C. Xiao, Z. Wei, Y. Zheng,
[29] R. Khataee, M.B. Kasiri, Photocatalytic degradation of organic dyes in presence of The research of photocatalytic activity on Si doped TiO2 nanotubes, Mater. Lett. 188
nanostructured titanium dioxide: influence of the chemical structured of dyes, J. (2017) 66–68.
Mol. Catal. A: Chem. 328 (2010) 8–26. [60] X. Yan, J. He, D. Evans, X. Duan, Y. Zhu, Preparation, characterization and pho-
[30] R.B. Khomane, B.D. Kulkarni, A. Paraskar, S.R. Sainkar, Synthesis, characterization tocatalytic activity of Si-doped and rare earth-doped TiO2 from mesoporous pre-
and catalytic performance of titanium silicate prepared in micellar media, Mater. cursors, Appl. Catal. B: Environ. 55 (2005) 243–252.
Chem. Phys. 76 (2002) 99–103. [61] S. Yin, Y. Aita, M. Komatsu, T. Sato, Visible-light-induced photocatalytic activity of
[31] Z. Lu, X. Jiang, B. Zhou, X. Wu, L. Lu, Study of effect annealing temperature on the TiO2-xNy prepared by solvothermal process in urea-alcohol system, J. Eur. Ceram.
structure, morphology and photocatalytic activity of Si doped TiO2 thin films de- Soc. 26 (2006) 2735–2742.
posited by electron beam evaporation, Appl. Surf. Sci. 257 (2011) 10715–10720. [62] S. Yin, Q. Zhang, F. Saito, T. Saito, Synthesis of titanium dioxide-based, visible-light
[32] B. Mazinani, A. Masron, A. Beitollani, R. Luque, Photocatalytic activity, surface area induced photocatalysts by mechanochemical doping, in: M. Sopicka-Lizer (Ed.),
and phase modifications of mesoporous SiO2-TiO2 prepared by a one-step hydro- High-Energy Ball Milling Mechanochemical Processing of Nanopowders, CRC Press
thermal procedure, Ceram. Int. 40 (2014) 1125–11532. L.L.C., Boca Raton, Florida, 2010, pp. 304–330.

56
L. Galeano et al. Materials Science in Semiconductor Processing 91 (2019) 47–57

[63] J. Yu, H. Yu, B. Cheng, M. Zhou, X. Zho, Enhanced photocatalytic of TiO2 powder 1161–1163.
(P-25) by hydrothermal treatment, J. Mol. Catal. A: Chem. 253 (2006) 112–118. [66] Z. Zheng, X. Wang, J. Liu, J. Xiao, Z. Hu, Si doping influence on the catalytic
[64] H. Zangeneh, A. Zinatizadeh, M. Habibi, M. Akia, M. Isa, Photocatalytic oxidation of performance of Pt/TiO2 mesoporous film catalyst for low-temperature methanol
organic dyes and pollutants in wastewater using different modified titanium diox- combustion, Appl. Surf. Sci. 309 (2014) 144–152.
ides: a comparative review, J. Ind. Eng. Chem. 25 (2015) 1–36. [67] Z. Zhihuan, F. Jimin, C. Honghong, A. Yusuke, Y. Shu, Recent progress on mixed-
[65] Q. Zhang, J. Wang, S. Yin, T. Sato, F. Saito, Synthesis of visible-light active TiO2-xSx anion type visible-light induced photocatalysts, Sci. China Technol. Sci. 60 (2017)
photocatalyst by means of mechanochemical doping, J. Am. Ceram. Soc. 87 (2004) 1447–1457.

57