Noble metal free Fe and Cr dual-doped nanocrystalline titania (Ti1?x?yMx+yO2) for high selective photocatalytic conversion of benzene to phenol at ambient temperature

Article abstract of DOI:10.1016/j.apcata.2018.07.035

Metal (Fe and Cr)-incorporated mesoporous nanocrystalline titania (Ti_1?x?yM_x+yO₂, or TiMxO₂) was synthesized using a simple solution combustion method. Structural characterization, microscopy, and spectroscopy t

Full text of DOI:10.1016/j.apcata.2018.07.035

AppliedCatalysisA,General565(2018)1–12
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Noble metal free Fe and Cr dual-doped nanocrystalline titania
(Ti1−x−yMx+yO2) for high selective photocatalytic conversion of benzene to
phenol at ambient temperature
Perumal Devaraji, Wan-Kuen Jo⁎
Department of Environmental Engineering, Kyungpook National University, Daegu 702-701, South Korea
A R T I C L E I N F O
A B S T R A C T
Keywords:
Metal (Fe and Cr)-incorporated mesoporous nanocrystalline titania (Ti1−x−yMx+yO2, or TiMxO2) was synthe-
sized using a simple solution combustion method. Structural characterization, microscopy, and spectroscopy
techniques confirmed the incorporation of Fe and Cr into the TiO2 lattice. Photocatalytic benzene oxidation to
phenol was investigated under ambient conditions, and under solar stimulator (AM1.5 or one sun condition) and
ultraviolet (UV) irradiation. Compared to pristine TiO2, TiO2 doped with 2 at% of Fe (Ti0.98Fe0.02O2) exhibited a
higher catalytic activity for benzene-to-phenol oxidation, highlighting the importance of the introduction of the
Fe dopant. Introduction of Fe and Cr into the TiO2 (Ti0.98Fe0.01Cr0.01O2) lattice further enhanced the oxidation of
benzene to phenol under UV light and gave a phenol yield that was ∼2 times higher than that given by
Ti0.98Fe0.02O2. The high photocatalytic activity of Ti1−x−yMx+yO2 and good photocurrent behaviour were
mainly ascribed to the generation of an electron trapping level composed of Fe3+ and Cr3+ in the TiO2 con-
duction band, disordered mesoporosity, high mobility, and a short diffusion length for charge carriers.
Disordered mesoporous nanocrystalline
Solution combustion synthesis
Functional application
Electron-trapping level
1. Introduction
irradiation in aqueous media [8]. Tanarungsm et al. conducted liquid-
phase photocatalytic hydroxylation of benzene to phenol with FeVCu/
Photocatalytic semiconductor materials have attracted attention as
a solution for current energy and environmental problems [1,2]. Among
the various wide-band-gap semiconductor photocatalysts, TiO2 has
been studied extensively because it demonstrates extraordinary che-
mical stability against photocorrosion and because holes with high
oxidizing power are generated in the valence band upon light absorp-
tion [3,4]. TiO2 is a semiconductor material with a wide band gap of
3.2 eV and is activated only under ultraviolet (UV) light. Phenol is one
of the most important organic chemicals in industry. Currently, phenol
is synthesized via the three-step cumene process at 250 °C and a pres-
sure of 30 bar [5,6]. In this process, the explosive cumene hydroper-
oxide is formed as an intermediate and acetone is formed as a major
side product. Therefore, an alternative green and one-step synthetic
route for phenol would be ideal to replace the cumene process [7].
Although several heterogeneous photocatalysts for the hydroxylation of
benzene to phenol have been reported to date, there have been no
significant breakthroughs [7–26]. Park et al. demonstrated that the
addition of polyoxometalates to the TiO2 suspension increases the
phenol yield [7]. In 2005, Shiraishi et al. reported the photocatalytic
conversion of benzene to phenol over mesoporous TiO2 under UV light
TiO2 as the photocatalyst under UV light and demonstrated that the tri-
metallic photocatalyst could improve phenol yield (9.7%) and se-
lectivity (52%) [11]. Recently, the one-step oxidation of benzene to
phenol on layered double hydroxides was reviewed [12].
In early 1981, Fujihira et al. examined the photocatalytic oxidation
of toluene on TiO2 in the presence of hydrogen peroxide with UV light
irradiation by a 500 W Xe lamp [13]. Palmisano et al. have demon-
strated the photocatalytic selective conversion of 4-methoxybenzyl al-
cohol to 4-methoxybenzaldehyde over TiO2 in an aqueous medium
under UV light [14]. Masami et al. demonstrated that doping TiO2 with
metal ions, such as Ir, Cr, Ru, or Rh, led to the generation of an electron
acceptor level in the conduction band (CB), which trapped the excited
electrons and increased the number of holes available in the valence
band (VB). Addition of grafted Fe3+ ions enhanced the electron transfer
process in TiO2, which in turn improved the photocatalytic activity
[15]. Han et al. have conducted the photocatalytic oxidation of benzene
to phenol using [RuII(Me2phen)3]2+ as a photocatalyst in the presence
of O2 and H2O [16]. Min et al. reported that doped Fe3+ along with
surface-grafted Fe3+ on TiO2 increased the availability of holes in the
VB, which enhanced the photocatalytic oxidation of organic compounds
⁎
Corresponding author.
E-mail address: wkjo@knu.ac.kr (W.-K. Jo).
https://doi.org/10.1016/j.apcata.2018.07.035
Received 7 May 2018; Received in revised form 13 July 2018; Accepted 25 July 2018
Availableonline27July2018
0926-860X/©2018PublishedbyElsevierB.V.

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
[4]. Fe-g-C3N4 (Fe-CN)/ titanium silicate zeolite (TS-1) hybrid materials
have also been utilized for the photocatalytic oxidation of benzene to
phenol under visible light irradiation. It was found that Fe doping in Fe-
CN/TS-1 improved the photocatalytic activity of these materials and
10% phenol yield was achieved [17]. Wang et al. conducted the direct
oxidation of benzene to phenol with ferrocene-modified carbon nitride
under outdoor sunlight irradiation and demonstrated that the use of
this photocatalyst improved the phenol yield to 14.4% as compared
with unmodified mesoporous graphitic carbon nitride (mpg-C3N4) and
ferrocene carboxyaldehyde [18]. Fe-doped g-C3N4 coated on SBA-15
catalyzed the oxidation of benzene to phenol at 60 °C under visible light
[19]. An FeCl3/mpg-C3N4 hybrid gave 38% benzene conversion and
97% phenol selectivity with light of λ ≥ 420 nm at 60 °C and only 4%
benzene conversion at 25 °C [20]. Dengke et al. reported that Fe-based
metal-organic frameworks could catalyze the selective hydroxylation of
benzene to phenol under visible light irradiation [21]. There are some
reports on the hydroxylation of benzene to phenol over Au/TiO2; the
excited electrons are stored in the Au nanoparticles and facilitate the
oxidation of benzene by TiO2 [22–25]. Ide et al. have reported that
benzene oxidation under visible light irradiation on immobilized gold
nanoparticles supported with layered titanate results in 62% yield and
96% phenol selectivity [22]. However, the above result was possible
only in the presence of a few hundred ppm (x) of benzene with the
addition of a large excess of seed phenol (≥10x). No phenol production
was observed in the absence of seed phenol [22]. The synthesis of
phenol via benzene oxidation over Au/TiO2 has also been reported; a
phenol yield of 13% with 89% selectivity was achieved after 24 h under
a CO2 atmosphere (230 kPa) and 1 sun irradiation conditions [23].
However, Au is an expensive noble metal and therefore, numerous ef-
forts have been undertaken to replace Au with low-cost materials
[10,26,27]. The above survey highlights the challenges involved in
designing a low-cost, effective, and efficient new photocatalyst that
gives high benzene conversions and phenol selectivities at ambient
temperature and pressure.
Doping with transition metal ions like Cr, Ir, Ru, and Rh leads to the
generation of a new acceptor level in the CB of TiO2, which acts as an
electron trapping center and facilitates the reduction of molecular
oxygen to H2O2 during the photocatalytic reaction. High photocatalytic
activity is observed for a grafted Fe3+ co-catalyst on Fe-doped TiO2
because Fe doping significantly enhances the electron-hole separation
and thus minimizes charge carrier recombination. This minimization of
charge carrier recombination is essential for redox reactions [15,17].
Soria et al. have demonstrated that bulk Fe3+ doping into TiO2 results
in better photocatalytic activity for NH3 reduction under UV light ir-
radiation as compared to surface-dispersed Fe3+ [28]. TiO2 doped with
Fe and Cr is useful for the oxidation of benzene and many other oxi-
dation reactions [29,30]. The oxidative degradation of phenol and
methylene blue using TiO2 doped with Fe, Cr, and other single metals
under visible and UV light (> 320 nm) irradiation have been examined
[30]. In our recently published study, we demonstrated that the in-
corporation of V in the TiO2 lattice generates a V5+ energy level below
the CB of TiO2, which helps to trap the excited electrons, and that Au
deposited over vanadium-doped TiO2 acts as an electron sink. The
Schottky junction between Au and TiO2, and V5+ synergistically in-
crease the availability of holes in the VB of TiO2, which can enhance the
photocatalytic conversion of benzene to phenol under UV light irra-
diation; 18% benzene conversion was achieved, along with a phenol
yield and selectivity of 15.9% and 88%, respectively [31]. For better
catalytic activity, it is important to prepare Fe and Cr co-doped TiO2
systems with mesoporosity. Ordered mesoporous materials have high
surface areas and long diffusion lengths. Consequently, the reactant and
product molecules have to diffuse a long distance to and from the active
sites. In these materials, there are a higher number of surface active
sites present on the porous channels. Hence, the short diffusion lengths
associated with disordered mesoporous materials are likely to be
helpful in minimizing the diffusion problem [32–35]. This is expected
to enhance the catalytic activity and selectivity for the target product.
In the present study, we synthesized anatase TiO2 doped with 2 at%
of Fe and Cr (TiMx = Ti0.98M0.02O2) by following a solution combustion
method (SCM), which exhibits advantages of lower costs, simple and
shorter synthesis time (within 15 min) compared to other methods. The
photocatalytic oxidation of benzene to phenol was carried out in a bi-
phasic system under UV/visible light under ambient conditions. After
12 h of UV irradiation in the presence of Ti0.98Fe0.01Cr0.01O2, 28%
benzene conversion and 25.2% and 90% phenol yield and selectivity,
respectively, were achieved. The present study is one of the major ef-
forts to synthesize rationally designed photocatalytic materials for the
application of organic conversion, water splitting and pollutant de-
gradation in our laboratories [31].
2. Experimental
2.1. Preparation of the photocatalysts
All chemicals used in this study were of analytical grade and used as
received. Ti[OCH(CH3)2]4, CrCl3·6H2O, and Fe(NO3)3·9H2O were used
as Ti, Cr, and Fe precursors, respectively, and urea was used as a fuel.
TiMx materials were synthesized using the solution combustion
synthesis method (SCM) [32]. The synthesis of TiO2 doped with 2 at%
M was carried out with a 1:1 M ratio of the metal ions (Ti + M
(Fe + Cr)) to urea, but with different Fe:Cr ratios of 100:0, 75:25,
50:50, 25:75, and 0:100. In a typical synthesis procedure, the required
amounts of the reactants were dissolved in water and introduced into a
250 mL beaker. The aqueous solution was stirred to get homogeneous
solution and then kept in a muffle furnace that was pre-heated at
400 °C. After the solution was introduced into the muffle furnace, the
water began to evaporate. This was followed by smoldering-type
combustion, and the resulting materials were designated as
Ti0.98Fe0.02O2,
Ti0.98Fe0.015Cr0.005O2,
Ti0.98Fe0.01Cr0.01O2,
Ti0.98Fe0.005Cr0.015O2, and Ti0.98Cr0.02O2, where the subscripts im-
mediately after Fe or Cr indicate the doping amount, in at%). Similarly,
TiO2 was synthesized without adding any Fe or Cr precursor.
2.2. Characterization
To examine the structural properties of the as-synthesized materials,
various characterization methods were used. Powder X-ray diffraction
(PXRD) data were recorded using a PANalytical X’Pert Pro dual goni-
ometer X-ray diffractometer with Cu-Kα (1.5418 Å) radiation and a Ni
filter, and the data were collected using a flat holder in the Bragg-
Brentano geometry (0.2°). High-resolution transmission electron mi-
croscopy (HRTEM), field emission scanning electron microscopy (FE-
SEM), N2 adsorption-desorption isotherm measurements, UV–vis ab-
sorption spectroscopy, photoluminescence (PL) spectroscopy, energy-
dispersive X-ray (EDX) spectroscopy, and X-ray photoelectron spectro-
scopy (XPS) were also conducted. An FE-SEM system (Hitachi SU8220)
equipped with an EDX attachment (Horiba X-MaxN with a silicon
DRIFT X-ray detector) was used to examine the surface morphologies
and chemical compositions. Prior to SEM analysis, samples were coated
with indium by ion sputtering (Hitachi, E-1030) for 60 s. HRTEM was
performed on all photocatalysts using an FEI Titan G2ChemiSTEM Cs
Probe microscope. The surface areas (SBET) and porosities of the ma-
terials were measured using a BET-BELSORP Mini II surface area ana-
lyzer; these measurements were carried out after degassing the mate-
rials at 300 °C. Raman analysis was carried out using a Renishaw (inVia
reflex) spectrometer combined with a microscope in reflectance mode
with a 632 nm excitation laser source and a spectral resolution of 0.5
cm−1. XPS was performed using an ULVAC-PHI Quantera SXMTM
scanning XPS microscope with an Al Kα radiation source or a Kratos
Axis HSi photoelectron spectrometer equipped with a charge neu-
tralizer and a monochromated Al Kα (1486.6 eV) radiation source.
Diffuse reflectance UV–vis studies were performed using a Shimadzu
2

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
UV-2600 spectrophotometer with BaSO4 as the standard reference
material. PL measurements were performed using a Shimadzu RF-6000
with a tunable excitation wavelength. All PL measurements were car-
ried out with 325 nm excitation pulses.
2.3. Electrochemical measurements
Electrochemical measurements were conducted within a conven-
tional three-electrode cell system, using platinum foil as the counter
electrode and Ag/AgCl as the reference electrode. The working elec-
trode was prepared by dispersing 5 mg of the photocatalyst in 100 μL of
diluted Nafion (1 wt%) in ethanol. The above mixture was ground well
and the obtained slurry was spread uniformly over the conductive side
of a 0.25 cm2 indium tin oxide (ITO) glass plate and dried in air.
Photocurrent (PEC) measurements were carried out in a Na2SO4 (0.2 M)
electrolyte using a 300 W xenon arc lamp with an AM 1.5 filter [33].
2.4. Photocatalytic activity
Fig. 1. PXRD patterns of different materials, namely, (a) TiO2, (b)
Ti0.98Fe0.02O2, (c) Ti0.98Fe0.015Cr0.005O2, (d) Ti0.98Fe0.01Cr0.01O2, and (e)
Ti0.98Fe0.005Cr0.015O2.
Photocatalytic benzene oxidation reactions were conducted in a
quartz round bottom flask for different time intervals with a 450 W
mercury lamp (λ = 200–400 nm) using a flow-type double-jacketed
quartz reactor. Several experiments were conducted under 1 sun illu-
mination conditions using an AM 1.5 light source. In a typical experi-
ment, 30 mg of the photocatalyst was dispersed in a quartz round
bottom flask, and CH3CN (2 mL), benzene (1 mL), and 25% H2O2 (2 mL)
were added to afford the starting reaction mixture. CH3CN was used as
a solvent to dissolve H2O2 and benzene in the aqueous solution. The
temperature of the double-jacketed reactor and quartz round bottom
flask was maintained at 25 °C by allowing water to circulate. The re-
action mixture was first stirred for 30 min in the dark, and then irra-
diated for 3–18 h. After irradiation, the organic and aqueous layers
were separated using a separating funnel. The products of benzene
oxidation were analyzed using a gas chromatography (GC) system
(Agilent 4890) equipped with a flame ionization detector (FID) and an
HP-5% phenyl methyl siloxane capillary column. Analysis of both the
organic and aqueous layers confirmed the presence of phenol. It should
be noted that any other liquid products in the chromatogram were also
included in the quantitative estimation. Gas chromatography/mass
spectrometry (GCMS) analysis was performed to detect the inter-
mediates or products (GC model: Perkin Elmer Clarus 680 with DB-5
column; 60 m × 0.32 mm and film 1.0 μm; MS model: Perkin Elmer
Clarus –SQ 8 T). NIST library was referred to analyze phenol and its
fragmentation patterns. Additionally, H2O2 conversion was determined
by a calorimetric titration method based on the formation of a yellow
colored complex TiIV-H2O2, using a UV/Vis spectrophotometer at
410 nm [18].
the doping level did not change the crystal structure of TiO2 [34]. The
absence of any oxides of Fe or Cr in Ti1−x−yMx+yO2 illustrates the
advantage of the SCM approach for incorporating Fe and Cr into the
TiO2 lattice [35]. The observed peak broadening confirmed the pre-
sence of nanosize crystalline particles. The XRD pattern did not show
any peak shift after metal doping, most likely due to the low content of
the dopant introduced into the TiO2 lattice. The average crystallite size
was estimated using the Scherrer equation to be approximately 19 nm,
which was also confirmed by SEM and HRTEM analysis.
3.2. Surface area analysis
N2 adsorption-desorption isotherms were obtained in order to in-
vestigate the textural properties and mesoporous nature of TiO2,
Ti0.98Fe0.02O2,
Ti0.98Fe0.015Cr0.005O2,
Ti0.98Fe0.01Cr0.01O2,
and
Ti0.98Fe0.005Cr0.015O2. Fig. 2a shows the N2 adsorption-desorption iso-
therms measured at 77 K; the Brunauer-Emmett-Teller (BET) method
was used for surface area analysis, and the Barrett-Joyner-Halenda
(BJH) method was used to determine the pore size distribution, which is
shown in Fig. 2b. A type IV adsorption-desorption isotherm with an H2
hysteresis loop was observed for TiO2 and all TiMx substrates, which
confirmed the presence of mesopores [36,37]. The measured BET sur-
face areas (SBET), pore sizes, and pore volumes of all materials are
shown in Table 1. Ti0.98Fe0.02O2 has mesopores with a higher surface
area (132 m2/g) than those of TiO2 (97 m2/g). Increasing the amount of
Cr increased the surface area from 154 m2/g in the case of
Ti0.98Fe0.015Cr0.005O2 to 161 m2/g for Ti0.98Fe0.01Cr0.01O2; however, the
surface area decreased to 154 m2/g for Ti0.98Fe0.005Cr0.015O2. Sig-
nificant changes in the pore size and pore volume were observed for
Ti0.98Fe0.005Cr0.015O2. The high surface area of Ti0.98Fe0.01Cr0.01O2 of
161 m2/g confirmed the presence of mesopores with more active sites
on the surface; this is a very critical factor for enhancing photocatalytic
activity. Introduction of higher amounts of Cr further decreased the
surface area and the pore volume, which indicated that the Cr ions had
diffused into the bulk and blocked the pores of the material. These
observations were fully supported by microscopy results. The observed
2.5. Control tests
Three different control experiments were performed for studying
the photocatalytic oxidation of benzene under conditions of 1) H216O +
H216O2
Ar + benzene + CH3CN,
+
air + benzene + CH3CN, 2) H218O
+
H216O2
+
+
and
3)
H216O
+
H216O2
Ar + benzene + CH3CN. The final products were analyzed by GCMS.
3. Results and discussion
3.1. PXRD studies
average pore size distribution was approximately 5.3
average pore volume was 0.16
size distribution profiles, which indicate the presence of mesopores in
all of the materials. In the case of Ti0.98Fe0.005Cr0.015O2 (Fig. 2b), a
significant decrease in the pore size indicates the introduction of more
Cr ions into the pores.
0.1 nm and the
The structural aspects of the prepared photocatalysts were in-
vestigated using PXRD. Fig. 1 shows that the PXRD patterns of all of the
prepared Ti1−x−yMx+yO2 catalysts matched with that of anatase phase
TiO2 (JCPDS 21-1272). No additional peaks that corresponded to Fe or
Cr oxides were observed, clearly indicating that all of the
Ti1−x−yMx+yO2 materials retained the anatase phase of TiO2 and that
0.02 cm3/g. Fig. 2b shows the pore
3

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Fig. 2. (a–e) N2 adsorption-desorption isotherm and (f) pore size distribution profile of TiO2, Ti0.98Fe0.02O2, Ti0.98Fe0.015Cr0.005O2, Ti0.98Fe0.01Cr0.01O2, and
Ti0.98Fe0.005Cr0.015O2 materials respectively.
3.3. HRTEM and SEM
Table 1
Physicochemical properties of TiO2, Ti0.98Fe0.02O2, Ti0.98Fe0.015Cr0.005O2,
Ti0.98Fe0.01Cr0.01O2, and Ti0.98Fe0.005Cr0.015O2.
The textural properties and morphologies of the materials were
investigated by HRTEM and SEM. The HRTEM images of Ti0.98Fe0.02O2
and Ti0.98Fe0.01Cr0.01O2 are shown in Fig. 3. It is evident that
Ti0.98Fe0.02O2 and Ti0.98Fe0.01Cr0.01O2 (and TiO2, Fig. S1 in the Sup-
porting information (SI)) are well-interconnected spherical particles
with disordered mesoporous structures (Fig. 3a and b). Ti0.98Fe0.02O2
and Ti0.98Fe0.01Cr0.01O2 show similar textural characteristics as those of
TiO2 [34]. Both Ti0.98Fe0.02O2 and Ti0.98Fe0.01Cr0.01O2 have particle
sizes of 8–19 nm; larger nanoparticles were also observed by SEM (Fig.
S1 in the SI). The d-spacing of 0.35 nm corresponds to the (101) plane
of TiO2 [38]. These results are consistent with the PXRD and N2 ad-
sorption-desorption isotherm results. Disordered mesoporosity is
usually associated with short diffusion lengths and arises from the inter
Catalyst
BET surface
area (m2/g)
Pore size
(nm)
Pore volume
(cc/g)
TiO2
Ti0.98Fe0.02O2
Ti0.98Fe0.015Cr0.005O2
Ti0.098Fe0.01Cr0.01O2
Ti0.98Fe0.005Cr0.015O2
97
6.0
5.5
5.5
5.5
4.3
0.14
0.17
0.18
0.19
0.16
132
154
161
154
4

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Fig. 3. HRTEM images of a) Ti0.98Fe0.02O2, b) and c) Ti0.98Fe0.01Cr0.01O2, and d–h) represent the respective elemental mapping images of Ti0.98Fe0.01Cr0.01O2 sample.
Color coding for different elements is shown on the images (Ti, red; Fe, blue; Cr, green; and O, cyan) (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article).
growth of nanoparticles, which can also result in the formation of ag-
gregates with additional framework void space. These mesopores are
only a few nanometers in depth, unlike conventional ordered meso-
porous materials [31–39]. The HRTEM image (Fig. 3b) shows that the
particles are well connected with each other, ensuring that they are
electrically interconnected, which leads to increased mobility of the
excited charge carriers to the surface of the photocatalyst where the
reaction takes place. XPS results confirmed the presence of Fe3+ and
Cr3+. To verify the homogeneous distribution of Cr and Fe in the TiO2
matrix, HRTEM-EDX was conducted on Ti0.98Fe0.01Cr0.01O2 (Fig. 3c).
All of the elements can be seen clearly (Fig. 3d–h), and it was found that
Fe3+ and Cr3+ were highly dispersed (Fig. 3e and f, respectively). SEM-
EDX also well supported with the above results (Fig. S2 in the SI). These
observations confirmed the doping and uniform distribution of the
metal ions on the surface and in the pores of the parent TiO2 matrix.
This feature of Ti0.98Fe0.01Cr0.01O2 is essential for separating the charge
carriers by trapping and, hence, enhancing their mobility towards the
photocatalyst surface from the bulk. It is important to note that both
Fe3+ and Cr3+ can act as electron trapping centers and synergistically
enhance the hole utilization, and hence reduce charge recombination.
Fig. 4. UV–vis spectra of (a) TiO2, (b) Ti0.98Fe0.02O2, (c) Ti0.98Fe0.015Cr0.005O2,
(d) Ti0.98Fe0.01Cr0.01O2, and (e) Ti0.98Fe0.005Cr0.015O2.
5

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
This aspect is very important in the context of oxidation reactions. The
porosity of the material was confirmed by the results of the adsorption
isotherm studies and was in good agreement with the HRTEM images.
(Fig. 5) were measured using an excitation wavelength of 325 nm. The
PL experiments revealed a decrease in the TiO2 and emission intensity
of the metal-doped TiO2 as compared to that of undoped TiO2 and,
hence, a reduction in charge recombination. TiMx showed character-
istic emission features at around 395, 405, 434, 446, and 464 nm. TiO2
showed high-intensity emission peaks at 395 and 464 nm, which were
attributed to the band gap transition and to the band edge free exciton
emission, respectively [46]. The broad emission feature that starts at
441 nm may have resulted from the charge-transfer transition from
Ti3+ to oxygen anions in a [TiO6]8− complex that was present in the
material along the (101) planes. All of the emission features were sig-
nificantly smaller when Fe was introduced into the TiO2 lattice and
even smaller after co-doping with Fe and Cr. While Ti0.98Fe0.02O2
showed emission features with the lowest intensity, undoped TiO2
showed emission features with the highest intensity, indicating that the
excited electrons are trapped by Fe3+ in TiO2 under light illumination,
which enhances the charge carrier separation and minimizes the elec-
tron-hole recombination [47]. The intensities of all emission features
decreased exponentially with increasing Cr doping. However, the
emission spectra of Ti0.98Fe0.005Cr0.015O2 and Ti0.98Fe0.01Cr0.01O2 were
similar, likely due to a small difference in the amounts of Fe and Cr
between the two photocatalysts. This observation is consistent with the
results of the UV–vis absorption studies. The above discussion shows
that excited electrons are synergistically trapped by both Fe and Cr
[46–49], which increases the availability of holes for oxidation on the
titania surface and leads to an effective charge separation. This also
suggests that there are electronic interactions between Fe3+, Cr3+, and
TiO2. On the basis of PL studies, it can be inferred that the materials
prepared using the SCM show substantially lower charge recombination
owing to selective electron trapping by Fe3+ and Cr3+ in TiO2, which
enhances the availability of holes for oxidation.
3.4. UV–vis absorption studies
UV-Vis absorption spectra were recorded to investigate the optical
properties, extent of the effect of doping, and nature of the doped Fe
and Cr species within the materials. Fig. 4 shows these UV–vis ab-
sorption spectra; the insets show the colors associated with TiO2 and all
of the TiMx materials. TiO2 shows absorption up to 380 nm in the UV
region with a band gap of 3.2 eV, which is typical for titania. A red shift
in the absorption edge was observed for Ti0.98Fe0.02O2, which showed
absorption at 400–600 nm in the visible light region, indicating the
presence of Fe3+ in the TiO2 lattice and the generation of a new Fe3+
3d energy level in the TiO2 CB, narrowing the band gap [40–43]. A
pronounced narrow peak was observed at 480 nm for Ti0.98Fe0.02O2,
which corresponded to the d-d electronic transitions of iron atoms [44].
With increasing Cr content, the visible light absorption by the materials
increased and the band gap decreased; the band gap of
Ti0.98Fe0.015Cr0.005O2 was 2.7 eV, which was lower than that of
Ti0.98Fe0.02O2 (2.9 eV). The highest visible light absorption was ob-
served for Ti0.98Fe0.01Cr0.01O2, which has a band gap of 2.6 eV, and no
significant changes were observed in the absorption for
Ti0.98Fe0.005Cr0.015O2. Two broad absorption peaks were observed for
Ti0.98Fe0.01Cr0.01O2, namely, an absorption peak with an onset at
∼600 nm and a broad absorption ranging from 630 to 800 nm, which
were attributed to the charge transfer from Cr3+ to Ti4+ and the d-d
4
4
transitions of A2g to T1g in Cr3+ ions, respectively [15,16,45]. This
suggested that the oxidation state of both Fe and Cr in the TiMx ma-
terials was +3. This conclusion was consistent with the XPS results. The
inset in Fig. 4 shows the color changes in the samples from white (a) to
pale yellow (b), then light brown (c) to brown (d), and finally dark
brown (e). These color changes in the sample reflect the changes in the
optical properties of the materials resulting from doping with Fe and Cr,
which reduce the optical band gap of TiO2 and enhance the visible light
absorption. These experimental observations suggest that there is a
strong interaction between Fe, Cr, and TiO2, and that the new energy
level formed below the CB because of metal doping assists the separa-
tion of charge carriers [46–49].
3.6. Raman analysis
Fig. 6 shows the Raman spectra of TiO2 and the TiMx photo-
catalysts. All photocatalysts showed the Raman-active fundamental
modes [50,51] of anatase-phase titania. After Fe and Cr doping, the
3.5. Photoluminescence studies
The PL emission spectra of TiO2 and all of the TiMx materials
Fig. 5. Photoluminescence studies (PL) of (a) TiO2, (b) Ti0.98Fe0.02O2, (c)
Fig. 6. Raman spectra of (a) TiO2, (b) Ti0.98Fe0.02O2, (c) Ti0.98Fe0.015Cr0.005O2,
Ti0.98Fe0.015Cr0.005O2 (d) Ti0.98Fe0.01Cr0.01O2, and (e) Ti0.98Fe0.005Cr0.015O2.
(d) Ti0.98Fe0.01Cr0.01O2, and (e) Ti0.98Fe0.005Cr0.015O2.
6

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Fig. 7. XPS spectra (a) Ti 2p (along with XPS of TiO2), (b) Fe 2p, and (c) Cr 2p for of fresh Ti0.98Fe0.01Cr0.01O2.
intensities of all Raman-active modes decreased and a small peak shift
from 145 to 147 cm−1 was observed (Fig. 6 inset). The decrease in the
peak intensity for all TiMx materials indicated that the incorporation of
Fe and/or Cr into the TiO2 lattice led to symmetry breaking in TiO2
[32,50,51]. The shift in the peak at 145 cm−1 confirmed the doping of
Fe and Cr into the TiO2 lattice. In addition, peak broadening confirmed
the presence of nanosize particles, and these results were well sup-
ported by XRD. There were no characteristic peaks corresponding to Fe
and Cr oxides in the Raman spectra, which implied the absence of Fe
and Cr oxides in the TiMx materials [52].
It was found that doping TiO2 with Fe and Cr did not affect the
binding energy. For more details on the band gap reduction and for a
deeper understanding of benzene oxidation, the valence band XPS
spectra of TiO2 and Ti0.98Fe0.01Cr0.01O2 were recorded (Fig. 8a). The
density of state (DOS) was determined with the help of UV–vis ab-
sorption spectroscopy (Fig. 8b) and the valence band XPS results. TiO2
and Ti0.98Fe0.01Cr0.01O2 showed valence band maximum (VBM) en-
ergies of ∼2.62 and ∼2.55 eV, and optical band gap energies of 3.2 and
2.6 eV, respectively. Therefore, the conduction band minima (CBM) of
TiO2 and Ti0.98Fe0.01Cr0.01O2 were calculated to be approximately
−0.58 and −0.05 eV, respectively. Based on the above results, the
absorption onset of Ti0.98Fe0.01Cr0.01O2 was located at ∼2.55 eV and
the maximum energy associated with the band tail was ∼1.90 eV.
Therefore, the substantial narrowing of the band gap of
Ti0.98Fe0.01Cr0.01O2 was attributed to slight VB tailing [56]. Combining
the experimental results with optical band gap calculations from the
absorption measurements, it was suggested that doping with Fe and Cr
reduces the band gap of Ti0.98Fe0.01Cr0.01O2. Furthermore, it can be
concluded that the VBM and CBM also decrease the band gap of
Ti0.98Fe0.01Cr0.01O2.
3.7. X-ray photoelectron spectroscopy
XPS analysis was conducted in order to investigate the chemical
compositions and electronic states of TiO2 and Ti0.98Fe0.01Cr0.01O2.
Fig. 7 shows the XPS spectra for the Ti 2p, Fe 2p, and Cr 2p core levels
of TiO2 and Ti0.98Fe0.01Cr0.01O2. The peaks at the binding energy of
458.6
0.1 eV (2p3/2) for TiO2 and Ti0.98Fe0.01Cr0.01O2 (Fig. 7a)
confirmed the presence of Ti4+ in all of the materials. Fe and Cr co-
doping in the titania lattice resulted in the appearance of a core-level Fe
2p3/2 and Fe 2p1/2 peaks at 711.3 and 724.1 eV, respectively (Fig. 7b),
and a Cr 2p3/2 peak at 577 eV and low intensity peak at ∼589 eV
(Fig. 7c). These results indicate that the oxidation states of both Fe
[53,54] and Cr [55] are predominantly +3 in Ti0.98Fe0.01Cr0.01O2. The
surface atom concentrations of Fe and Cr were 0.48% and 0.6%, re-
spectively. The O 1 s core-level peaks for TiO2 and Ti0.98Fe0.01Cr0.01O2
3.8. Photocatalytic benzene oxidation
The photocatalytic performances of all TiMx photocatalysts under
UV and visible light irradiation for the oxidation of benzene to phenol
for different time durations were evaluated and the results are shown in
Table 2. In the absence of light illumination, Ti0.98Fe0.01Cr0.01O2
appeared at 529.9
0.1 eV (Fig. S3 in the SI).
Fig. 8. a) Valence band XPS spectra of TiO2 and Ti0.98Fe0.01Cr0.01O2 and b) schematic diagram of density of state of TiO2 and Ti0.98Fe0.01Cr0.01O2.
7

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Table 2
photocatalysts under 12 h UV irradiation. Ti0.98Fe0.01Cr0.01O2 photo-
catalyst showed the highest photocatalytic activity, indicating that an
optimal level of Fe3+ and Cr3+ doping (1 at% each) is necessary. The
starting reaction solution (colorless) underwent a color change to
brown toward the end of the reaction (Fig. S4 in the SI), indicating that
this reaction took place photocatalytically. However, the phenol se-
lectivity decreased from 90% to 80% after 18 h of irradiation because of
the secondary reactions.
Control experiments for benzene oxidation either without a photo-
catalyst or under dark conditions were also carried out (Table S1 in the
SI). No changes in the starting reactant molecules were observed in the
reactions carried out without any catalyst in the dark. UV irradiation
for 18 h in the absence of any photocatalyst led to some photochemical
oxidation of benzene. The formation of side products indicated the
possibility of a free radical mechanism, which operated exclusively
under these conditions because of photochemical conversion and
homogeneous cleavage of hydrogen peroxide.
Photocatalytic benzene oxidation to phenol using M (= Fe and/or Cr) doped
TiO2 photocatalysts.*
Photocatalyst
t (h)
Benzene
conv. (%)
Phenol
Yield (%)
Selectivity (%)
Ti0.98Fe0.02O2
Ti0.98Fe0.02O2
Ti0.98Fe0.015Cr0.005O2
Ti0.98Fe0.015Cr0.005O2
Ti0.98Fe0.01Cr0.01O2
Ti0.98Fe0.01Cr0.01O2
Ti0.98Fe0.01Cr0.01O2
Ti0.98Fe0.005Cr0.015O2
Ti0.98Cr0.02O2
6
12
6
12
6
12
12
12
12
13
18
0.5
0.3
12.0
14.0
12.2
18.2
14.1
25.2
24.0
22.9
10.7
0.5
92
77
92
91
94
90
89
88
67
0.5
0.3
0.4
0.6
0.4
0.5
0.6
0.6
0.4
0.3
0.4
0.6
0.4
0.5
0.6
0.6
0.4
13.3
20
15
0.4
0.6
0.4
0.5
28
a
26.8
26
16
0.6
0.6
0.4
* Reaction conditions: catalyst 30 mg; benzene 1 ml, CH3CN 2 ml; H2O2
(25%) 2 ml; light source used 450 W mercury lamp (λ = 200–400 nm).
a
Dissolved oxygen was removed by argon bubbling for 30 min.
Other control experiments were performed for the photocatalytic
benzene oxidation under conditions of H216O + H216O2 + Argon (Ar)
+ benzene + CH3CN, H216O + H216O2+ Air + benzene + CH3CN,
and H218O + H216O2+ Ar + benzene + CH3CN) to examine the origin
of oxygen in the final product phenol. The yields obtained under later
two experimental conditions were similar each other (∼25%)
(Table 2), indicating that the contribution of oxygen to the formation of
reactive hydroxyl radicals was not significant in our photocatalytic
system. A possible explanation is that trapping of photo-excited elec-
tron by Cr3+ and Fe3+ dopant levels is more predominant than that by
oxygen molecule. Ti0.98Fe0.01Cr0.01O2 showed the highest activity under
UV light, which underscores the importance of energetic holes on the
titania surface for benzene oxidation. The low activity of
Ti0.98Fe0.01Cr0.01O2 under 1 sun illumination conditions (AM 1.5) sug-
gested that the photocatalytic activity for benzene oxidation depended
on the VB position (discussed in the proposed mechanism section) of
TiO2 [43]. The pristine TiO2 sample showed a lower activity than that
of the TiMx materials, underscoring the important role of metal doping
(Fe and Cr) in effective charge separation. Photocurrent measurements
for Ti0.98Fe0.02O2 and Ti0.98Fe0.01Cr0.01O2 under simulated solar light
with a cut-off filter indicated that light only having λ ≥ 400 nm passed
through (Fig. 11). The photocurrent generated by Ti0.98Fe0.01Cr0.01O2
was about four times higher than that of Ti0.98Fe0.02O2 under visible
light irradiation (λ ≥ 400 nm), indicating a greater visible light ab-
sorption by the former. The higher photocurrent of Ti0.98Fe0.01Cr0.01O2
was ascribed to more holes and electrons generated by this sample. The
effective utilization of holes would lead to higher photooxidation of
benzene to phenol by Ti0.98Fe0.01Cr0.01O2. There were no significant
changes in the photocatalytic activities of Ti0.98Fe0.01Cr0.01O2 after
three repeated reactions, which illustrated the stability of the catalyst in
continuous operating mode (Fig. S5 in the SI). In addition, there was no
significant variation in the photocurrent measurements of
Ti0.98Fe0.01Cr0.01O2 (Fig. 11). The XRD and XPS analyses showed no
changes in the corresponding patterns between the fresh and used
samples, confirming the high stability of the photocatalyst after three
consecutive reactions (Figs. S6 and S7 in the SI, respectively).
showed very low benzene conversion (∼0.5%) with trace amounts of
phenol after 24 h (Table S1 in the SI). The negligible conversion can be
rationalized by considering that the conventional Fenton oxidation re-
action is homogeneous, whereas the reaction in the current study is
heterogeneous. Moreover, strong acids, such as HClO4, CF3COOH, and
H2SO4, are necessary for the Fenton oxidation reaction [17–19] while
the photocatalytic benzene oxidation reaction was performed in the
absence of any acid. Therefore, the benzene oxidation reaction would
occur primarily via a photocatalytic process with assistance of H2O2,
without a significant contribution of Fenton reaction to the conversion
of benzene.
Using Ti0.98Fe0.02O2, 13% benzene conversion with 92% selectivity
was attained after 6 h of irradiation (Table 2). The conversion of ben-
zene increased to 18% when the irradiation time was increased to 12 h.
However, the reaction selectivity toward the phenol product decreased
to 77% with prolonged reaction time, primarily due to further oxidation
of the generated phenol to side products such as hydroquinone. In ad-
dition, a qualitative analysis of the gaseous products was carried out
using a GC system equipped with a thermal conductivity detector. It
was found that small amounts of hydrogen and CO2 were produced in
the reaction, likely due to the reformation of organic compounds via
secondary reactions [31,57–61]. Some of the generated charge carriers
were consumed in these secondaryreactions, affecting the yield and
selectivity for the primary product (phenol). However, we believe that
this occurs mainly to maintain the charge neutrality of the system,
which prevents the electron-hole recombination in the photocatalyst. In
addition, the conversion of benzene (16%) achieved using
Ti0.98Cr0.02O2 after 12 h irradiation was close to that attained using
Ti0.98Fe0.02O2 (18%) after 12 h irradiation.
The introduction of different amounts of Cr and Fe into the TiO2
lattice led to higher benzene conversions, yields, and selectivities than
those obtained with Fe-doped TiO2 (Ti0.98Fe0.02O2) after 12 h of UV
irradiation. The results are shown in Table 2. Co-doping with Fe and Cr
resulted further increasing of benzene conversions and phenol yields for
Ti0.98Fe0.015Cr0.005O2. The results obtained from different irradiation
time are shown in Fig. 9a, while the phenol yields obtained are shown
in Fig. 9b. The benzene conversion and phenol yield were significantly
increased from 3 to 12 h under light irradiation, which highlighted the
importance of Fe3+ and Cr3+ dopants in the trapping of excited elec-
trons. Ti0.98Fe0.005Cr0.015O2 displayed low benzene conversion (26%)
and phenol yield (22.9%) compared with those obtained using
Ti0.98Fe0.01Cr0.01O2 photocatalyst (Table 2). The selectivity of phenol
was decreased to 88%. The Ti0.98Fe0.01Cr0.01O2 sample exhibited a
As shown in Table 3, the benzene conversion (25%), phenol yield
(22.5%), and selectivity (90%) obtained using the Ti0.98Fe0.01Cr0.01O2
photocatalyst were higher than or comparable with those obtained
using other photocatalysts with or without containing a noble metal
reported in earlier studies [8,17–19,21,23,31,58,62–65,57]. For ex-
ample, MOFs based MIL-100(Fe) nanosphere photocatalyst displayed a
benzene conversion of 34.4% with a phenol selectivity of 98% under
visible light irradiation [62]. A mesoporous titania (mTiO2) showed a
phenol yield of 19% with a selectivity of 83% under UV-light irradia-
tion [8]. Fe-g-C3N4/SBA-15 photocatalyst exhibited a benzene conver-
benzene conversion of 28% (turnover frequency (TOF) = 38.7 h−1
;
Table S2 in the SI) with the phenol yield and selectivity of 25.2% and
90%, respectively.
sion of 11.9% with
a phenol selectivity of 83% [19]. An Au/
Ti0.98V0.02O2 photocatalyst provided a benzene conversion of 18% with
a phenol yield of 15.9% was achieved under UV-light irradiation [31].
Fig. 10 shows the phenol yields obtained using different
8

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Fig. 9. Time-dependent photocatalytic benzene oxidation results for (a) benzene conversion and phenol selectivity and (b) phenol yield with Ti0.98Fe0.01Cr0.01O2 after
UV irradiation.
Leaf structured ZnO nano photocatalyst showed a benzene conversion
of 5.2% with a phenol selectivity of 92% under UV-light irradiation
[57]. And also, current photocatalyst showed a higher photocatalytic
activity than those obtained using Fe5V2.5Cu2.5/TiO2 [63], Fe/activated
carbon [64], Fe/SBA-16 catalyst [65], Fe-CN/TS-1 [17], Fc-MCN1.0-5
[18], and Au/TiO2 [23]. The superior or comparable photocatalytic
efficiency of Ti0.98Fe0.01Cr0.01O2 was primarily attributed to the sy-
nergistic effect of Fe and Cr co-doping for trapping charge carriers. The
superior photocatalytic efficiency of Ti0.98Fe0.01Cr0.01O2 to the other
photocatalysts was primarily attributed to the synergistic effect of Fe
and Cr co-doping for trapping charge carriers. Consequently, the pho-
tocatalyst developed in the current study can be used to achieve cost-
effective and highly efficient phenol production. Although TiO2 doped
with Fe and Cr showed visible light absorption, low benzene conversion
was observed even after irradiation for 24 h with visible light (AM1.5).
This indicated that a small number of holes were generated as com-
pared to those generated upon UV light irradiation. Further funda-
mental studies are required to improve the performance of the photo-
catalyst under 1 sun illumination condition. Longer irradiation times
with visible light increase the photocatalytic activity linearly.
Fig. 10. Phenol yields observed with (a) Ti0.98Cr0.02O2, (b) Ti0.98Fe0.02O2, (c)
However, the phenol selectivity decreased from 90% to 80% after
Ti0.98Fe0.015Cr0.005O2, (d) Ti0.98Fe0.01Cr0.01O2, and (e) Ti0.98Fe0.005Cr0.015O2 for
18 h of irradiation; this is mainly due to the over-oxidation of generated
12 h UV irradiation.
phenol to other side products such as p-benzoquinone (p-BQ), hydro-
quinone (HyQ), H2, and/or CO2 [58–61]. For example, Likozar et al.
observed p-BQ as a side product during the photocatalytic oxidation of
benzene to phenol over carbon nanotube (CNT)-supported Cu and TiO2
photocatalytic system under UV-light irradiation [58]. They also car-
ried out the UV light-assisted photocatalytic degradation of phenol on
TiO2 photocatalyst and observed that BQ and HyQ were as main in-
termediates, which were mineralized to CO2 after prolonged reaction
time [59]. Bui et al. examined the photocatalytic degradation of ben-
zene over TiO2 under UV-light irradiation and reported a small amount
of BQ, HyQ, and CO2 generated during the reaction [60]. Also, a phenol
degradation over TiO2-nanotube under UV-light irradiation led to the
generation of benzoquinone as a side product [61]. The GCMS results
obtained in our study show that p-benzoquinone was generated as a
major organic side product during the conversion reaction of benzene
to phenol (Fig. S8 in the SI).
3.9. Proposed mechanism
Fig. 11. Photocurrent measurements of (a) Ti0.98Fe0.02O2 and (b)
Ti0.98Fe0.01Cr0.01O2 with a solar simulated light source, λ ≥ 400 nm.
A possible mechanism for the oxidation of benzene to phenol was
proposed (Fig. 12) on the basis of the photocatalytic experiments car-
ried out under different conditions, characterization studies, GCMS
9

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
Table 3
Comparison of photocatalytic activity of Ti0.98Fe0.01Cr0.01O2 with other reported photocatalysts (or catalyst) for benzene oxidation to phenol reaction.
S.No.
Photocatalyst
(or catalyst)
Condition
Benzene conversion
(%)
Phenol Yield
(%)
Phenol Selectivity (%)
Reference
1.
mTiO2
UV-light
23
19
83
[8]
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Fe-CN/TS-1
Fc-MCN1.0-5
Visible light
Visible light
Visible light
Visible light
Visible light
UV-light
UV-light
Visible light
303 K
–
–
11.9
30.6
–
10
14.4
–
29.4
13
15.9
4.8
–
7.2
17.5
11.7
25.2
18.4
26.4
20.7
96
89
88
92
98
73
[17]
[18]
[19]
[21]
[23]
[31]
[58]
[62]
[63]
[64]
[65]
This work
Fe-g-C3N4/SBA-15
MIL-100 (Fe) bulk
Au/TiO2
Au/Ti0.98V0.02O2
ZnO
MIL-100 (Fe) nanosphere
Fe5V2.5Cu2.5/TiO2
Fe/activated carbon
Fe/SBA-16
18
5.2
34.4
9.8
19.6
12.1
28
303 K
65 °C
UV-light
89.3
96.4
90
Ti0.98Fe0.01Cr0.01O2
analysis, isotopic measurements, valence band XPS results, and pre-
vious literature. In the proposed mechanism, phenol was thought to be
generated via two reaction paths, namely path-A and path-B. TiO2
generates electron and hole charge carriers upon UV light absorption.
The photogenerated electrons are excited from the VB to the CB via the
electron trapping level resulting from doping with Fe3+ and Cr3+
[15–18]. In path-A, the trapped electrons in the dopant level reduce
Fe3+ to Fe2+, which then reacts with H2O2 and a proton (from H2O) to
produce a hydroxyl radical and Fe3+. The formed hydroxyl radicals
attack the aromatic benzene ring to form hydroxycyclohexadienyl ra-
dicals [15–18,66–68]. The photogenerated positive holes in the VB or
Fe3+ oxidize the hydroxycyclohexadienyl radicals to phenol via a de-
protonation process to restart the photocatalytic cyclic reaction
[60,66–69]. The control tests suggested that trapping of photo-excited
electron by Cr3+ and Fe3+ dopant levels is more predominant than that
by oxygen molecule. In path-B, the generated holes in the VB of TiO2
react with benzene to produce benzene radical ions. These radical ions
react with hydroxyl radicals to form phenol, likely via the deprotona-
tion of an unstable intermediate [31,49]. In the absence of peroxide and
under dark condition, no benzene oxidation occurred, confirming that
both of these conditions are crucial for benzene oxidation reactions.
Phenol incorporated with 16O (m/z 94) was observed in the GCMS
spectra, when 16O water was used for photocatalytic benzene oxidation
(Fig. S9 in the SI), indicating that the hydroxyl radical was originated
from water only. However, both 16O (m/z 94) and 18O (m/z 96)-in-
corporated phenols were observed, when 18O- water was used for the
photocatalytic benzene oxidation (Fig. S10 in the SI), indicating that
the hydroxyl radical was originated from both water and H2O2.
Upon irradiation, the formation of gaseous products such as CO2
and H2 was observed, likely due to the photocatalytic reforming of
benzene, acetonitrile, and/or phenol in possible secondary reactions
through the consumption of holes in the VB of titania. Based on the
carbon balance concept, the phenol selectivity was 90%, while the rest
10% was assigned to unknown byproducts. PL and optical studies re-
vealed the formation of an electron trapping level in the CB of TiO2 due
to Fe3+ and Cr3+ doping. The PL emission features were observed to be
weaker after the incorporation of Fe, indicateing the entrapment of
excited electrons. This phenomenon enhances the charge carrier se-
paration and lowers the incidence of their recombination. Thus, more
holes are available in the VB of TiO2 for benzene oxidation. Co-doping
with Fe and Cr led to further reduction of emission intensity, indicating
that Fe and Cr synergistically capture the excited electrons in the
trapping level. The photogenerated holes are then effectively utilized
for benzene oxidation, highlighting the importance and role of Fe and
Cr as trapping centers. Meanwhile, surface defects or non-radiative
decay may reduce PL intensity, but they are not likely to be a major
contributor for the reduced PL intensity in our photocatalytic system.
Since the VBM of the Ti0.98Fe0.01Cr0.01O2 catalyst is located at
2.55 eV, the DOS from the valence band XPS spectra (Fig. 8b) and the
photocatalytic benzene oxidation reaction appear to demonstrate that
benzene oxidation occurs on the surfaces of TiO2 and that Fe and Cr
assist in the trapping of excited electrons and generation of hydroxyl
radicals. This underscores the necessity of wide-band-gap semi-
conductor materials such as TiO2 and ZnO [57] with valence band
Fig. 12. Possible mechanism of photocatalytic benzene oxidation to phenol.
10

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
maxima of ∼2.5 eV for benzene oxidation. The electrons in the CB of
TiO2 can undergo three different processes under the reaction condi-
tions: (a) recombination, (b) consumption by H2O2 or (c) trapping by
Fe3+ and Cr3+. The increase in phenol yield from 14% (Ti0.98Fe0.02O2)
to 25.2% (Ti0.98Fe0.01C0.01O2) underscores the prevalence of processes b
and c, which help the reaction. A colorimetric titration test showed that
97% of the initially added H2O2 was used for the benzene oxidation
reaction (Eq. S1 in the SI).
The phenol production selectivity was obtained in the range of
67–92% for our catalysts, indicating the formation of side products. As
mentioned earlier, the GCMS results obtained in our study showed that
p-benzoquinone was a major organic side product during the conver-
sion reaction of benzene to phenol (Fig. S8 in the SI). In addition, hy-
droquinone, CO2, and H2 can be produced during the benzene-to-
phenol conversion reactions [31].
[6] H. Mimoun, L. Saussine, E. Daire, M. Postel, J. Fischer, R. Weiss, J. Am. Chem. Soc.
105 (1983) 3101–3110.
[7] H. Park, W. Choi, Catal. Today 101 (2005) 291–297.
[8] Y. Shiraishi, N. Saito, T. Hirai, J. Am. Chem. Soc. 127 (2005) 12820–12822.
[9] O. Tomita, B. Ohtani, R. Abe, Catal. Sci. Technol. 4 (2014) 3850–3860.
[10] O. Tomita, R. Abe, B. Ohtani, Chem. Lett. 40 (2011) 1405–1407.
[11] G. Tanarungsun, W. Kiatkittipong, S. Assabumrungrat, H. Yamada, T. Tagawa,
P. Praserthdam, J. Ind. Eng. Chem. 13 (2007) 870–877.
[12] A. Antonyraj, K. Churchil, Srinivasan, Catal. Surv. Asia 17 (2013) 47–70.
[13] M. Fujihira, Y. Satoh, T. Osa, Nature 293 (1981) 206–208.
[14] G. Palmisano, S. Yurdakal, V. Augugliaro, V. Loddo, L. Palmisano, J. Adv. Synth.
Catal. 349 (2007) 964–970.
[15] M. Nishikawa, W. Shiroishi, H. Honghao, H. Suizu, H. Nagai, N. Saito, J. Phys.
Chem. A 121 (2017) 5991–5997.
[16] J.W. Han, Y.M. Lee, W. Nam, S. Fukuzumi, Chem. Sci. 8 (2017) 7119–7125.
[17] X. Ye, Y. Cui, X. Qiu, X. Wang, Appl. Catal. B 152–153 (2014) 383–389.
[18] X. Ye, Y. Cui, X. Wang, ChemSusChem 7 (2014) 738–742.
[19] X. Chen, J. Zhang, X. Fu, M. Antonietti, X. Wang, J. Am. Chem. Soc. 131 (2009)
11658–11659.
[20] P. Zhang, Y. Gong, H. Li, Z. Chen, Y. Wang, RSC Adv. 3 (2013) 5121–5126.
[21] D. Wang, M. Wang, Z. Li, ACS Catal. 5 (2015) 6852–6857.
[22] Y. Ide, M. Matsuoka, M. Ogawa, J. Am. Chem. Soc. 132 (2010) 16762–16764.
[23] Y. Ide, N. Nakamura, H. Hattori, R. Ogino, M. Ogawa, M. Sadakane, T. Sano, Chem.
Commun. 47 (2011) 11531–11533.
4. Conclusion
The SCM approach was employed for the synthesis of disordered
mesoporous Ti0.98Fe0.02O2 and Ti0.98(Fe + Cr)0.02O2 nanocrystalline
photocatalysts. XPS, Raman analysis, EDX spectroscopy, and other
characterization studies confirmed the doping of Fe and Cr into the
TiO2 matrix. PL studies revealed an electron transfer mechanism be-
tween the metal dopants and titania. The mobility of the photoexcited
charge carriers was enhanced by a disordered mesoporous structure,
and the electrically interconnected and integrated Ti0.98(Fe + Cr)0.02O2
nanocrystallites. Moreover, doping with metal ions reduced the re-
combination of excited charge carriers and thus expedited their se-
paration and migration to the surface of the catalyst for enhanced re-
activity. Introduction of Fe3+ and Cr3+ increased the conversion of
benzene and the selectivity of phenol. Co-doping with Fe and Cr re-
sulted in visible light absorption because of the formation of a trapping
level in the band gap of TiO2, and co-doped TiO2 (Ti0.98Fe0.01Cr0.01O2)
generated four times as much current as Fe-doped TiO2 (Ti0.98Fe0.02O2).
Fe and Cr would act as electron trapping centers and assist in charge
separation. This would increase the diffusion and availability of holes
on the titania surface and enhance their utilization for oxidation, which
led to high phenol yields. Visible light absorption and efficiency of light
energy into current or chemical energy conversion highlight the po-
tential of Ti0.98(Fe + Cr)0.02O2 photocatalyst for the application of CeH
activation, optoelectronics and variety organic transformation reac-
tions.
[24] Z. Zheng, B. Huang, X. Qin, X. Zhang, Y. Dai, M.-H. Whangbo, J. Mater. Chem. 21
(2011) 9079–9087.
[25] Z. Zheng, B. Huang, X. Meng, J. Wang, S. Wang, Z. Lou, Z. Wang, X. Qin, X. Zhang,
Y. Dai, Chem. Commun. 49 (2013) 868–870.
[26] R. Su, L. Kesavan, M.M. Jensen, R. Tiruvalam, Q. He, N. Dimitratos, S. Wendt,
M. Glasius, C.J. Kiely, G.J. Hutchings, F. Besenbacher, Chem. Commun. 50 (2014)
12612.
[27] J. Yu, Y. Hai, B. Cheng, J. Phys. Chem. C 115 (2011) 4953–4958.
[28] J. Soria, J.C. Conesa, V. Augugliaro, L. Palmisano, M. Schiavello, A. Sclafani, J.
Phys. Chem. 95 (1991) 274–282.
[29] J. Kou, C. Lu, J. Wang, Y. Chen, Z. Xu, R.S. Varma, Chem. Rev. 117 (2017)
1445–1514.
[30] J. Choi, H. Park, M.R. Hoffmann, J. Phys. Chem. C 114 (2010) 783–792.
[31] P. Devaraji, N.K. Sathu, C.S. Gopinath, ACS Catal. 4 (2014) 2844–2853.
[32] K. Sivaranjani, A. Verma, C.S. Gopinath, Green Chem. 14 (2012) 461–471.
[33] W.K. Jo, S. Kumar, M.A. Isaacs, A.F. Lee, S. Karthikeyan, Appl. Catal. B 201 (2017)
159–168.
[34] K. Sivaranjani, C.S. Gopinath, J. Mater. Chem. 21 (2011) 2639–2647.
[35] A. Varma, A.S. Mukasyan, A.S. Rogachev, K.V. Manukyan, Chem. Rev. 116 (2016)
14493–14586.
[36] P.C.S. Bezerra, R.P. Cavalcante, A. Garcia, H. Wender, M.A.U. Martines,
G.A. Casagrande, J. Giménez, P. Marco, S.C. Oliveiraa, A.M. Jr, J. Braz. Chem. Soc.
28 (2017) 1788–1802.
[37] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol,
T. Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
[38] X. Chen, L. Liu, P.Y. Yu, S.S. Mao, Science 331 (2011) 746–750.
[39] K. Sivaranjani, S.R. Ambal, T. Das, K. Roy, S. Bhattacharyya, C.S. Gopinath,
ChemCatChem 6 (2014) 522–530.
[40] K.K. Patra, B.D. Bhuskute, C.S. Gopinath, Sci. Rep. 7 (2017) 6515.
[41] Z. Luo, J.C. Smith, W.H. Woodward, A.W. Jr, J. Phys. Chem. C 116 (2012)
7219–7226.
[42] M. Asilturk, J. Photochem. Photobiol. A 203 (2009) 64–71.
[43] W. Choi, A. Termin, M.R. Hoffmann, J. Phys. Chem. 98 (1994) 13669–13679.
[44] M.A.A. Mamun, Y. Kusumoto, M.S. Islam, J. Mater. Chem. 22 (2012) 5460–5469.
[45] X. Fan, X. Chen, S. Zhu, Z. Li, T. Yu, J. Ye, Z. Zou, J. Mol. Catal. A 284 (2008)
155–160.
Disclosure statement
The author declares that no conflicting actual or potential interests
exist.
[46] H. Dang, X. Dong, Y. Dong, Y. Zhang, S. Hampshire, Int. J. Hydrogen Energy 38
(2013) 2126–2135.
[47] X.Y. Li, P.L. Yue, C. Kutal, New J. Chem. 27 (2003) 1264–1269.
[48] T. Umebayashi, T. Yamaki, H. Itoh, K. Asai, J. Phys. Chem. Solid. 63 (2002)
1909–1920.
Acknowledgements
[49] A. Litke, E.J.M. Hensen, J.P. Hofmann, J. Phys. Chem. C 121 (2017) 10153–10162.
[50] K. Bhattacharya, S. Varma, A.K. Tripathi, S.R. Bharadwaj, A.K. Tyagi, J. Phys.
Chem. C 112 (2008) 19102–19112.
[51] R.S. Zeferino, M.B. Flores, U. Pal, J. Appl. Phys. 109 (2011) 014308.
[52] H. Ren, P. Koshy, F. Cao, C.C. Sorrell, Inorg. Chem. 16 (2016) 8071–8081.
[53] S. Mansingh, D.K. Padhi, K.M. Parida, Int. J. Hydrogen Energy 41 (2016)
14133–14146.
[54] D. Channei, B. Inceesungvorn, N. Wetchakun, S. Ukitnukun, A. Nattestad, J. Chen,
S. Phanichphan, Sci. Rep. 4 (2014) 5757.
[55] Y.H. Peng, G.F. Huang, W.Q. Huang, Adv. Power Technol. 23 (2012) 8–12.
[56] S.A. Ansari, M.H. Cho, Sci. Rep. 6 (2016) 25405.
This work was supported by the National Research Foundation of
Korea (NRF) grant funded by the Korea government (MSIP) (No.
2016R1A2B4009122 and No. 2017R1A4A1015628).
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi: https://doi.org/10.1016/j.apcata.2018.07.035.
[57] N.K. Sathu, P. Devaraji, C.S. Gopinath, J. Nanosci. Nanotechnol. 16 (2016)
9203–9208.
References
[58] V.D.B.C. Dasireddy, B. Likozar, J. Taiwan Inst. Chem. Eng. 82 (2018) 331–341.
[59] M. Krivec, A. Pohar, B. Likozar, G. Drazic, AIChE J. 61 (2015) 572–581.
[60] T.D. Bui, A. Kimura, S. Ikeda, M. Matsumura, J. Am. Chem. Soc. 132 (2010)
8453–8458.
[61] L. Suhadolnik, A. Pohar, B. Likozar, M. Ceh, Chem. Eng. J. 303 (2016) 292–301.
[62] B. Xu, Z. Chen, B. Han, C. Li, Catal. Commun. 98 (2017) 112–115.
[63] G. Tanarungsun, W. Kiatkittipong, P. Praserthdam, H. Yamada, T. Tagawa,
S. Assabumrungrat, J. Ind. Eng. Chem. 14 (2008) 596–601.
[1] K. Maeda, K. Domen, J. Phys. Chem. Lett. 18 (2010) 2655–2661.
[2] A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol. C 1 (2000) 1–21.
[3] A. Kudo, Y. Miseki, Chem. Soc. Rev. 38 (2009) 253–278.
[4] L. Min, Q. Xiaoqing, M. Mashiro, H. Kazuhito, J. Am. Chem. Soc. 135 (2013)
10064–10072.
[5] N. Herron, C.A. Tolman, J. Am. Chem. Soc. 109 (1987) 2837–2839.
11

P. Devaraji, W.-K. Jo
AppliedCatalysisA,General565(2018)1–12
[64] Y. Zhong, G. Li, L. Zhu, Y. Yan, G. Wu, C. Hu, J. Mol. Catal. A Chem. 272 (2007)
169–173.
[65] M. Jourshabani, A. Badiei, Z. Shariatinia, N. Lashgari, G.M. Ziarani, Ind. Eng. Chem.
Res. 55 (2016) 3900–3908.
[67] C. Walling, Acc. Chem. Res. 8 (1975) 125–131.
[68] J.R.L. Smith, R.O.C. Norman, J. Chem. Soc. 113 (1963) 2897–2905.
[69] S. George, S. Pokhrel, Z. Ji, B.L. Henderson, T. Xia, L. Li, J.I. Zink, A.E. Nel,
L. Madler, J. Am. Chem. Soc. 133 (2011) 11270–11278.
[66] A. Kunai, S. Hata, S. Ito, K. Sasaki, J. Am. Chem. Soc. 108 (1986) 6012–6016.
12