Res Technol. 2019. https:doi.org10.1016j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246 No. of Pages 11 [610642]

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
www.jmrt.com.brAvailable online at www.sciencedirect.com
Original Article
HZSM-5 zeolite supported boron-doped TiO 2for
photocatalytic degradation of ofloxacin
Haolun Li, Wenjie Zhang∗, Yuxuan Liu
School of Environmental and Chemical Engineering, Shenyang Ligong University, Shenyang 110159, China
a r t i c l e i n f o
Article history:
Received
25 November 2019
Accepted 28 December 2019
Available
online xxx
Keywords:PhotocatalyticHZSM-5TiO
2
OfloxacinDegradationa b s t r a c t
Boron-doped TiO 2was supported on HZSM-5 zeolite in a sol-gel route to prepare B-
TiO 2(x%)/HZSM-5 composites. The composites are composed of anatase TiO 2and HZSM-5
zeolite. The crystallite size of anatase TiO 2decreases after loading B-TiO 2on HZSM-5 zeolite.
The SEM and TEM images of the composite show the dispersion of the B-TiO 2crystals on
the external surface of HZSM-5. The bandgap energies of B-TiO 2(x%)/HZSM-5 composites
are approximately 3.2 eV. The N2adsorption-desorption isotherms indicate that B-TiO 2is
a mesoporous material and HZSM-5 zeolite is a microporous material. The mesoporous B-
TiO 2layer is loaded on the external surface of the zeolite and does not enter into the inner
micropores. The XPS results indicate the unchanged chemical environments of both tita-
nium and oxygen in the supported B-TiO 2. The B-TiO 2(x%)/HZSM-5 composites have higher
photocatalytic activity than the unsupported B-TiO 2. The optimum B-TiO 2loading content
is 20% in the composite, and 55.3% of the original ofloxacin molecules were degraded on the
sample after 30 min of irradiation. Photocatalytic oxidation not only causes breaking up of
the ofloxacin molecules, but also leads to decomposition of the major functional groups of
the ofloxacin molecules.
© 2020 The Authors. Published by Elsevier B.V. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/ ).
1. Introduction
Antibiotics in the wastewater of pharmaceutical plants are
among the most concerned hazardous pollutants that have
to be removed from the wastewater before discharging. These
harmful substances can hardly be removed through the tradi-
tional biochemical techniques. Deep oxidation methods such
as photocatalysis are used to handle most kinds of haz-
ardous organic substances [1–3] . Ofloxacin in the wastewater
can also be degraded during photocatalytic oxidation process
∗Corresponding author .
E-mail: [anonimizat] (W. Zhang).[4–8] . Photocatalytic oxidation efficiency depends mostly on
the activity of photocatalytic materials [9]. During more than
fifty years of investigation in this field, TiO 2-based material
is always the research focus and is the most applied material
[10–13] . On the other hand, pure TiO 2is not the optimal choice
in wastewater treatment at most conditions, and modification
is necessary to improve the performance of TiO 2[14,15] . The
essential purpose of modification is to enhance the activity
of pure TiO 2, and another purpose is to make it applicable to
complex water treating conditions.
A mostly adopted modification method is the doping of
nonmetal ions in TiO 2matrix, either in the voids of TiO 2crys-
tal skeleton or by substitution [16–18] . Simsek et al. studied
boron doped TiO 2catalysts for degradation of endocrine dis-
rupting compounds and pharmaceuticals under visible light
https://doi.org/10.1016/j.jmrt.2019.12.086
2238-7854/© 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/ ).

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
2 j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
irradiation [19]. We reported the sol-gel preparation of hollow
spherical x%B-TiO 2photocatalyst to study the effect of boron
content on RBR X-3B decoloration [20], and the effects of boron
content and calcination temperature on properties of B-TiO 2
photocatalyst prepared by a solvothermal method [21]. Pow-
der material is not suitable for most wastewater treatment
conditions since the separation of fine powders and water is
difficult. To solve this problem, photocatalytic material can
be used in the supported form, and the performance of the
supported material depends on the properties of the support
[22–24] .
ZSM-5 zeolite, with highly ordered micropores, surface
acidity,
and ion-exchange capacities, is one of the most widely
applied inorganic materials as catalyst support, adsorbent,
and molecular-sized space for various chemical or photo-
chemical reactions. In this work, we loaded boron-doped TiO 2
on HZSM-5 zeolite in a sol-gel route. The B-TiO 2(x%)/HZSM-
5 composites were characterized by X-ray diffraction (XRD),
scanning electron microscopy (SEM), transmission electron
microscopy (TEM), UV-Visible diffuse reflectance spectrome-
try, N2adsorption-desorption technique, X-ray photoelectron
spectroscopy and fluorescence spectrophotometry. Photocat-
alytic degradation of ofloxacin on the B-TiO 2(x%)/HZSM-5
composites was evaluated to show the effects of supporting
on photocatalytic activity.
2. Experimental
2.1. Synthesis of B-TiO 2(x%)/HZSM-5 composites
Boron-doped TiO 2was loaded on the surface of HZSM-5 zeo-
lite particles by sol-gel method. Ultrapure water (0.9 mL) was
dissolved in 4 mL anhydrous ethanol to prepare precursor A.
Tributyl borate (0.045 mL) and tetrabutyl titanate (2 mL) were
dissolved in 8 mL anhydrous ethanol, and hydrochloric acid
(0.1 mL) was added in the solution to prepare precursor B. Pre-
cursor A and HZSM-5 zeolite were mixed with precursor B,
and the mixture was stirred for 30 min to obtain a gel. The gel
was dehydrated at 80◦C for 10 h, and the solid was calcined at
450◦C for 3 h. The products are denoted as B-TiO 2(x%)/HZSM-
5, where x% is the loading content of B-TiO 2in the composites.
The molar ratio of n(B)/n(Ti) is 0.03:1 in the boron-doped TiO 2,
representing the optimal boron doping ratio in TiO 2[20].
2.2. Characterization methods
The XRD pattern of the material was taken on a D8 Advance
X-ray diffractometer (Cu K/H9251, /H9261 = 1.5416 Å). The SEM image of
surface morphology was obtained by a QUANTA 250 scanning
electron microscope. The TEM image of the material was taken
on a FEI Tecnai G2 20 transmittance electron microscope.
The UV–vis diffuse reflectance spectrum was obtained by a
LAMBDA 35 UV–vis spectrometer. N2adsorption-desorption
was conducted on an ASAP 2460 surface area and pore size
analyzer. XPS analysis was performed on an ESCALAB 250Xi
X-ray photoelectron spectroscopy (Al K/H9251). The steady-state PL
spectra of gadolinium titanate were recorded by a LS55 fluo-
rescence spectrophotometer.
Fig. 1 – Molecular structure of ofloxacin.
2.3. Photocatalytic degradation of ofloxacin
Fig. 1 illustrates the molecular structure of ofloxacin. Ofloxacin
solution (50 mL) and B-TiO 2(x%)/HZSM-5 composite were
mixed in a 100 mL quartz beaker. The ofloxacin concentra-
tion in the solution was 40 mg/L. The amount of B-TiO 2in
the solution was 20 mg. Adsorption of ofloxacin on the mate-
rial was measured after adsorption-desorption equilibrium. A
20 W ultraviolet lamp was turned on to initiate photocatalytic
degradation. The irradiation intensity was 2300 /H9262W/cm2at
253.7 nm, which was measured on the surface of the ofloxacin
solution. Subsequently, the solid material was removed from
the solution by centrifugation and filtration. Ofloxacin concen-
tration was determined on an Agilent 1260 HPLC (C18 column,
150 × 4.6 mm, 5 /H9262m; 1% phosphoric acid: acetonitrile = 4:1). The
UV–vis spectrum of the ofloxacin solution was recorded by a
LAMBDA 35 UV–vis spectrometer.
3. Results and discussion
3.1. Characterization results
Fig. 2 shows the XRD patterns of HZSM-5, B-TiO 2and B-
TiO 2(x%)/HZSM-5 composites. The diffraction pattern of the
unsupported B-TiO 2sample agrees with JCPDS 01-562. TiO 2
is in the solely anatase crystalline structure, and there is
no diffraction pattern of rutile phase TiO 2. Boron-containing
materials cannot be identified in the diffraction pattern of
B-TiO 2either. The B-TiO 2(x%)/HZSM-5 composites contain dif-
ferent contents of B-TiO 2and HZSM-5, and therefore, the
diffraction patterns of the B-TiO 2(x%)/HZSM-5 composites are
the combination of the diffraction patterns of both B-TiO 2
and HZSM-5. Meanwhile, diffraction intensity of each compo-
nent depends on the weight percentage of the substance. The
diffraction pattern of HZSM-5 zeolite comes from tetragonal
Si-O and Al-O structures in the zeolite. The diffraction peak
of the preferred (101) plane of anatase TiO 2can be identified
in the XRD patterns of the B-TiO 2(x%)/HZSM-5 comosite. The
loading of B-TiO 2on HZSM-5 does not cause phase transfor-
mation from anatase to rutile.
Table 1 gives lattice parameters of anatase TiO 2in B-
TiO 2(x%)/HZSM-5 composites with respect to B-TiO 2loading
content. The preferred (101) plane was used to calculate the
crystallite size of anatase TiO 2in the composite. Accord-
ing to the Scherrer equation, the crystallite size is inversely
proportional to the line broadening at half the maximum

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx 3
Table 1 – Lattice parameters of anatase TiO 2in B-TiO 2(x%)/HZSM-5 with respect to B-TiO 2loading content.
Samples a(=b) (nm) c (nm) V (nm3) Crystallite size (nm)
B-TiO 2(10%)/HZSM-5 0.37713 0.94900 0.13589 7.4
B-TiO 2(20%)/HZSM-5 0.37797 0.95196 0.13600 8.4
B-TiO 2(50%)/HZSM-5 0.37840 0.95279 0.13651 8.9
B-TiO 2 0.37951 0.95703 0.13783 10.6
Fig. 2 – XRD patterns of HZSM-5, B-TiO 2and
B-TiO 2(x%)/HZSM-5 composites as a factor of B-TiO 2loading
content.
intensity (FWHM). The crystallite size of anatase TiO 2in
the unloaded B-TiO 2powders is 10.6 nm. The preparation
conditions of boron-doped B-TiO 2powders, i.e. precursor com-
position, boron content, and calcination conditions, were
optimized to obtain high photocatalytic activity, as reported
in our previous work [20,21] . The as-prepared B-TiO 2powders
have high photocatalytic activity on azo dye degradation.
One of the effects of loading B-TiO 2on the surface of
HZSM-5 zeolite is the reduced crystallite size of anatase TiO 2.
This phenomenon is quite similar to pure TiO 2that was sup-
ported on HZSM-5 zeolite [11,13] . Anatase TiO 2crystal growth
is restrained after loading on the surface of HZSM-5 parti-
cles [8,24] . TiO 2crystals tend to distribute on the surface of
HZSM-5 particles, especially for the B-TiO 2(x%)/HZSM-5 com-
posites with small B-TiO 2content. The doping of boron in
anatase TiO 2usually causes crystal refinement effect, while
nano-sized TiO 2is considered to have good photocatalytic
activity. Although the original purpose of supported photocat-
alyst is to solve the problem of solid-wastewater separation,
it is interesting that the supported photocatalyst has higher
photocatalytic activity than the unsupported powders.
Fig. 3 shows the SEM images of B-TiO 2(x%)/HZSM-5 com-
posites, unsupported B-TiO 2, and HZSM-5 zeolite. The well
crystallized HZSM-5 particles have regular octahedral shape
in the size of 1 /H9262m × 3 /H9262m × 5 /H9262m, and such HZSM-5 parti-
cles can be observed in Fig. 3(e). Meanwhile, there are also
some small HZSM-5 fragments among the large crystallized
HZSM-5 particles. The shape of the unsupported B-TiO 2par-ticles is irregular, and the particle size can be as large as
several micrometers, as shown in Fig. 3(d). The B-TiO 2sam-
ple was ground before characterization, so that there are many
small particles in the sample. Calcination of the material after
sol-gel transformation has significant effect on TiO 2particles
aggregation.
Fig. 3(a–c) give the surface morphologies of the
supported B-TiO 2(x%)/HZSM-5 composites with B-TiO 2load-
ing contents of 10%, 20% and 50%. Ti-OH is formed in the
sol
during controlled hydrolysis of tetrabutyl titanate in the
precursor. Dehydration of Ti-OH leads to formation of Ti-O-
Ti structure in the gel after sol-gel transformation, resulting
in combination of the HZSM-5 particles. The small particles
are also combined with the large ones, so that separation
of
water and the B-TiO 2(x%)/HZSM-5 particles is much easier
than using fine B-TiO 2powders.
Fig. 4 shows the TEM images of B-TiO 2and B-
TiO 2(20%)/HZSM-5 in two magnifications. The crystallite
size of B-TiO 2in Fig. 4(a1,a2) is around 10 nm, which is in
accordance
with the value calculated from the XRD pattern.
The B-TiO 2crystals may interact with others to form large
aggregates, and large B-TiO 2particles can be produced after
further agglomeration of the small B-TiO 2aggregates. The
supporting of B-TiO 2crystals on the surface of HZSM-5
particle can be clearly shown in Fig. 4(b1,b2). The content
of B-TiO 2is 20% in the sample B-TiO 2(20%)/HZSM-5, and
there is a thin layer of B-TiO 2crystals loading on HZSM-5
particles. Fig. 4(b1,b2) reveal the well dispersion of B-TiO 2
crystals on the external surface of HZSM-5, and in this case,
the aggregation of B-TiO 2crystals is reduced.
As shown in Fig. 4(b1,b2), the thickness of B-TiO 2layer is
approximately 50 nm. UV photons can penerate into TiO 2for
a few hundred nanometers. Based on this situation, the B-
TiO 2layer supported on HZSM-5 particles can all take part in
photon-absorption and the subsequent degradation process.
The particle size of the unsupported B-TiO 2particles can be as
large as several micrometers, and most part of the bulk mate-
rial cannot be applied to photocatalytic degradation process.
On the other hand, the interaction between B-TiO 2and HZSM-
5 zeolite cannot be ignored. Cozens et al. reported the role of
NaY zeolite in mediation of electron and hole migration over
vacant in the zeolite, leading to long-lived charge separated
species within the zeolite cavities [25]. Photogenerated elec-
trons in the B-TiO 2layer can be transferred to the vacant in
the zeolite, since the thin B-TiO 2layer is closely attached to
HZSM-5 zeolite.
Fig. 5(a) shows UV–vis diffuse reflectance spectra of B-
TiO 2(x%)/HZSM-5 composites with different B-TiO 2loading
content. HZSM-5 zeolite is an insulator and can only absorb
a small percent of photons at short wavelength. The absorp-
tion edges of B-TiO 2and B-TiO 2(x%)/HZSM-5 composites are
in the ultraviolet region. These materials do not have absorp-
tion in the visible region and cannot respond to visible light

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
4 j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 3 – SEM images of B-TiO 2(x%)/HZSM-5, B-TiO 2and HZSM-5. (a) 10%, (b) 20%, (c) 50%, (d) B-TiO 2, (e) HZSM-5.
irradiation. The absorption edge of the B-TiO 2(x%)/HZSM-5
composite moves to short wavelength region and shows a
blueshift with reduced B-TiO 2content. HZSM-5 zeolite can-
not absorb visible light and the majority of UV light. However,
it does not mean that HZSM-5 zeolite is transparent to
the light. Instead, the irradiation may be reflected by the
surface of HZSM-5 zeolite. The supported B-TiO 2layer has
significant influences on light absorption characters of theB-TiO 2(x%)/HZSM-5 composites. UV light at the wavelength
below the absorption edge can be thoroughly absorbed by the
B-TiO 2(x%)/HZSM-5 composite even if B-TiO 2loading content
is as small as 10%. As stated before, crystallite size of TiO 2
is reduced after supporting on the surface of HZSM-5 zeolite.
Quantum size effect declares the close relationship between
crystallite size in nano-scale and blueshift of absorption edge
of a semiconductor.

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx 5
Fig. 4 – TEM images of B-TiO 2and B-TiO 2(20%)/HZSM-5. (a1,a2) B-TiO 2, (b1,b2) B-TiO 2(20%)/HZSM-5.
Fig. 5(b) shows the hv-(ahv)2plots of HZSM-5, B-TiO 2and
B-TiO 2(x%)/HZSM-5 composites. Tauc-plot method was used
to calculate bandgap energy of semiconductor [26]. Here, ˛, h,
/ETB are the absorption coefficient, Planck constant, and light fre-
quency, respectively. According to Einstein relationship, the
energy of a photon is equal to hv. The hv-(ahv)2plot is appli-
cable to direct-gap material. The onset of absorption is the
intersection of the extending linear regime of the plot and
the abscissa axis. The bandgap energies of B-TiO 2(10%)/HZSM-
5, B-TiO 2(20%)/HZSM-5, B-TiO 2(50%)/HZSM-5 and B-TiO 2are
3.27, 3.25, 3.24 and 3.17 eV, respectively. The bandgap energy
of the boron-doped B-TiO 2is close to 3.2 eV of normal anatase
TiO 2. Meanwhile, the bandgap energy of the B-TiO 2(x%)/HZSM-
5 composite is slightly larger than the unsupported B-TiO 2.
Photon absorption character of the B-TiO 2(x%)/HZSM-5 com-
posite mostly depends on the properties of the supported
B-TiO 2and can hardly be affected by the insulated HZSM-5
zeolite.
N2adsorption-desorption isotherms of HZSM-5 zeolite, B-
TiO 2and B-TiO 2(x%)/HZSM-5 composites are illustrated in
Fig. 6(a). The adsorption-desorption isotherm of the unsup-
ported B-TiO 2is classified as type IV isotherm, which reveals
mesoporous character of the material. The type IV adsorption-
desorption isotherm has an apparent hysteresis loop, and theisotherm of B-TiO 2has a type H1 hysteresis loop. The hystere-
sis loop locates in N2relative pressure between 0.5 and 0.8.
When N2relative pressure is higher than 0.8, the increment
of the adsorbed N2quantity is quite slow. All the above-
mentioned characters of the adsorption-desorption isotherm
demonstrate the mesoporous character of the unsupported
B-TiO 2. The slow increase of adsorbed N2quantity at high
N2relative pressure shows that the material does not have
notable micropore or macropore, and the type H1 hysteresis
loop in the isotherm of B-TiO 2reveals a narrow pore size range
of the mesopores in the material.
HZSM-5 zeolite has micropores in the pore size around
0.6 nm. The adsorption-desorption isotherm of HZSM-5 zeo-
lite is the type I isotherm. The surface of the micropores
in HZSM-5 zeolite can be occupied at low N2relative pres-
sure, and the adsorbed quantity of N2is maintained in a
plateau during subsequent increase of N2relative pressure.
The adsorbed N2number has an apparent increase at the sat-
uration pressure of N2, due to capillary condensation of N2
molecules in the pores of HZSM-5 zeolite. At the same time,
there is also a type H4 hysteresis loop in the isotherm of HZSM-
5 zeolite. This type of hysteresis loop represents zeolite or
active carbon that has slits in narrow size.
From the above-mentioned observations, we can know
that B-TiO 2is a mesoporous material and HZSM-5 zeolite

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
6 j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 5 – (a) UV–vis diffuse reflectance spectra and (b)
hv-(ahv)2plots of B-TiO 2(x%)/HZSM-5 as a factor of B-TiO 2
loading content.
is a microporous material. The N2adsorption-desorption
isotherms of B-TiO 2(x%)/HZSM-5 composites present both
the characters of mesopores and micropores. The adsorbed
N2number decreases with rising B-TiO 2loading content in
the composites when N2relative pressure is below 0.8. N2
molecules are adsorbed on the surface of the material in
monolayer or multilayer. The difference in adsorbed N2num-
ber is attributed to the available surface area of the material,
Surface area is larger in the composite containing smaller con-
tent of B-TiO 2. Capillary condensation of N2molecules in the
pores occurs at saturation pressure of N2. At the saturation
pressure, the adsorbed N2number is almost the same for all
the materials. It is interesting to see that the pore volume in
B-TiO 2is almost as same as HZSM-5 zeolite, although pore size
is different for them.
Fig. 6(b) shows mesopore size distributions in B-
TiO 2(x%)/HZSM-5, B-TiO 2and HZSM-5 zeolite. There is
almost no mesopore in HZSM-5 zeolite, and pore size in
B-TiO 2is within a narrow range between 3 nm and 8 nm.
The B-TiO 2layer contributes nearly all the mesopores in the
B-TiO 2(x%)/HZSM-5 composite. Fig. 6(c) compares microp-
ore size distributions in B-TiO 2(20%)/HZSM-5 and HZSM-5
zeolite. Since the weight percentage of B-TiO 2is only 20%
in the B-TiO 2(20%)/HZSM-5 composite, micropore volume inB-TiO 2(20%)/HZSM-5 is slightly smaller than HZSM-5 zeolite.
The micropores in HZSM-5 zeolite are not blocked after
loading a thin B-TiO 2layer on the surface of HZSM-5 zeolite.
As a result, the B-TiO 2layer is reason ably loaded on the
external surface of the zeolite and does not enter into the
inner micropores. This phenomenon is also observed in the
TEM images of the materials. The precursor did not penetrate
into the micropores of HZSM-5 zeolite in sol-gel synthesizing
route.
Table 2 gives the BET surface area, average pore size, and
pore
volume of B-TiO 2, HZSM-5 and B-TiO 2(x%)/HZSM-5. The
values in the table prove the conclusions obtained from Fig. 6.
The surface area and pore volume of the B-TiO 2(x%)/HZSM-
5 are contributed by both B-TiO 2and HZSM-5. The average
pore size of B-TiO 2is 7.1 nm, which decreases with declining B-
TiO 2loading content. Total pore volume of the material varies
slightly with B-TiO 2loading content. HZSM-5 zeolite has many
micropores although pore volume of single micropore is much
smaller than mesopore in B-TiO 2. Surface area of photocata-
lyst is essential to absorbing photons in the initiating step of
photocatalytic degradation process. The porous HZSM-5 zeo-
lite with large surface area is a suitable support for B-TiO 2.
As stated before, B-TiO 2particles aggregation is prohibited
after dispersing a thin layer of B-TiO 2on the external sur-
face of HZSM-5 zeolite. The highly dispersed B-TiO 2crystals
will inevitably provide more surface area to absorb incoming
photons and to adsorb pollutant as well.
Fig. 7 shows the XPS Ti2p and O1s spectra of B-TiO 2and
B-TiO 2(20%)/HZSM-5 composite. The Ti2p electron binding
energy
peaks of B-TiO 2are as similar as B-TiO 2(20%)/HZSM-5,
except that the former peaks have higher intensity. The bind-
ing energies of Ti2p1/2and Ti2p3/2electrons are 464.7 eV and
459.1 eV, respectively. The distance between the two binding
energy peaks is 5.6 eV, representing the status of fully oxidized
Ti4+oxidation state [27]. The binding energies of Ti2p elec-
trons do not change after supporting on the surface of HZSM-5
zeolite.
Since oxygen exists in both B-TiO 2and HZSM-5 zeolite, the
XPS O1s spectra of B-TiO 2and B-TiO 2(20%)/HZSM-5 compos-
ite are more complex, as shown in Fig. 7(b). The O1s electrons
in the spectrum of B-TiO 2have three binding energy peaks
situated at 531.9, 530.5 and 530.1 eV. Oxygen and hydroxyl
group adsorbed on the surface of the materials have two bind-
ing energy peaks at 531.9 eV and 530.5 eV. The other binding
energy peak at 530.1 eV is related to Ti-O bonding in B-TiO 2
[11]. These binding energy peaks can also be found in the O1s
spectrum of B-TiO 2(20%)/HZSM-5 composite. Meanwhile, two
extra peaks at 532.7 eV and 533.3 eV in the spectrum can be
assigned to oxygen in Si O and Al O bonds in the HZSM-5
zeolite [11,28] . Based on the analyses of the XPS Ti2p and O1s
spectra, we can find that the chemical environments of both
titanium and oxygen in B-TiO 2are not affected by HZSM-5
zeolite. There is no reaction between HZSM-5 zeolite and the
supported B-TiO 2layer.
Photoluminescence spectrum of photocatalyst is used
to study photogenerated electron-hole recombination effi-
ciency. Fig. 8 shows photoluminescence spectra of the B-
TiO 2(x%)/HZSM-5 composites after being excited at 253.7 nm.
The B-TiO 2(20%)/HZSM-5 has the lowest photoluminescence

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx 7
Fig. 6 – (a) N2adsorption-desorption isotherms of B-TiO 2(x%)/HZSM-5; (b) Mesopore size distributions in
B-TiO 2(x%)/HZSM-5, B-TiO 2and HZSM-5 zeolite; (c) Micropore size distributions in B-TiO 2(20%)/HZSM-5 and HZSM-5 zeolite.
Table 2 – Specific surface area and porous properties of B-TiO 2, HZSM-5 and B-TiO 2(x%)/HZSM-5.
Samples BET surface area
(m2/g)Average pore size
(nm)Pore volume
(cm3/g)
HZSM-5 286.2 2.0 0.1445
B-TiO 2(10%)/HZSM-5 260.8 2.3 0.1506
B-TiO 2(20%)/HZSM-5 240.0 2.6 0.1518
B-TiO 2(50%)/HZSM-5 174.2 3.6 0.1548
B-TiO 2 85.9 7.1 0.1513
intensity, and the B-TiO 2has the highest photoluminescence
intensity. Low photoluminescence intensity, i.e. low electron-
hole recombination efficiency, means long lifetime of the
charge carriers.
3.2. Photocatalytic degradation of ofloxacin
Fig. 9(a) shows the adsorption and photocatalytic degradation
of ofloxacin on B-TiO 2(x%)/HZSM-5 composites with differ-
ent B-TiO 2loading content. Ofloxacin can be adsorbed on
the materials but the maximum removal efficiency is less
than 10%. Although HZSM-5 zeolite is a microporous mate-
rial with large surface area, ofloxacin molecule is large andenters into the internal micropores. On the other hand, the
external surface of HZSM-5 zeolite can only adsorb 3.2% of the
original ofloxacin molecules in the solution. The mesoporous
B-TiO 2can remove 6.4% of the original ofloxacin molecules via
adsorption. Photogenerated electrons and holes may combine
in a very short time if they are not used for reacting with pollu-
tant. Migration of the electrons and holes are difficult so that
photocatalytic reaction occurs mainly on the surface of pho-
tocatalyst. A readily reaction between the oxidative reagents
and the pollutant depends on fast migration of pollutant from
the solution to the surface of photocatalyst.
HZSM-5 zeolite has no photocatalytic activity on ofloxacin
degradation under the same reaction conditions of B-TiO 2.

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
8 j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 7 – XPS spectra of B-TiO 2and B-TiO 2(20%)/HZSM-5. (a)
Ti2p, (b) O1s.
Fig. 8 – Photoluminescence spectra of B-TiO 2(x%)/HZSM-5
composites after excitation at 253.7 nm.
Although photocatalyst in supported form usually is not as
active as the powder form, this is not the case for HZSM-5
supported B-TiO 2. As shown in Fig. 9(a), the B-TiO 2(x%)/HZSM-
5 composites have higher photocatalytic activity than the
unsupported B-TiO 2. The optimum B-TiO 2loading content
is 20%, and the sample B-TiO 2(20%)/HZSM-5 has the maxi-mum efficiency on photocatalytic degradation of ofloxacin.
55.3% of the original ofloxacin molecules were degraded on
B-TiO 2(20%)/HZSM-5 after 30 min of irradiation. The mech-
anism for improved photocatalytic activity of B-TiO 2after
loading on HZSM-5 can be complex, and at present, it needs
to clarify what causes the optimum loading content. Disper-
sion of B-TiO 2as a layer on HZSM-5 zeolite is beneficial to both
absorption of irradiating photons and adsorption of ofloxacin
molecules from solution. On the other hand, since the same
amount of B-TiO 2was used to evaluate the activity of different
material, more HZSM-5 existed in the solution when the B-
TiO 2(x%)/HZSM-5 with low B-TiO 2content was used. Although
the
solution was stirred throughout the degradation process,
the particles in deep solution were not involved in the process.
Fig. 9(b) illustrates the effects of photocatalyst dosage on
ofloxacin removal efficiencies. B-TiO 2(20%)/HZSM-5 was used
as the photocatalyst since it has the strongest activity. Both
the adsorption efficiency and photocatalytic degradation effi-
ciency increase with rising photocatalyst dosage, especially
at low photocatalyst dosage. The adsorption efficiency has a
relative linear relationship with the amount of B-TiO 2. When
B-TiO 2dosage is more than 400 mg/L, the increase of photo-
catalytic degradation efficiency is very small and may slightly
decrease when B-TiO 2dosage is more than 600 mg/L. After
Fig. 9 – (a) Adsorption and photocatalytic degradation of
ofloxacin on B-TiO 2(x%)/HZSM-5 composites; (b) Effects of
B-TiO 2(20%)/HZSM-5 dosage on ofloxacin removal
efficiencies.

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx 9
Fig. 10 – (a) Photocatalytic degradation of ofloxacin with extended irradiation time; (b) Kinetic plots of photocatalytic
degradation of ofloxacin on B-TiO 2(x%)/HZSM-5 and B-TiO 2.
photocatalyst dosage reaches the optimal amount, not all
the
powders can absorb enough irradiation. In this work,
400 mg/L of B-TiO 2was applied in photocatalytic degradation
of ofloxacin.
Photocatalytic degradation of ofloxacin on the materials
was examined with extended irradiation time, as presented
in
Fig. 10(a). Ofloxacin can be degraded on the materials after
enough time. Fully degradation of ofloxacin in the solution
can be achieved in 90 min, using B-TiO 2(20%)/HZSM-5 as the
photocatalyst. The other B-TiO 2(x%)/HZSM-5 composites also
have higher activity than the unsupported B-TiO 2. Photocat-
alytic degradation of ofloxacin follows the first order reaction
law. Fig. 10(b) illustrates kinetic plots of photocatalytic degra-
dation
of ofloxacin on B-TiO 2(x%)/HZSM-5 and B-TiO 2. The first
order reaction rate constants are 0.0365, 0.0410, 0.0262 and
0.0248 min−1, using B-TiO 2(10%)/HZSM-5, B-TiO 2(20%)/HZSM-
5, B-TiO 2(50%)/HZSM-5 and B-TiO 2as the photocatalysts.
The B-TiO 2(20%)/HZSM-5 sample was reused to examine
the stability on ofloxacin degradation. After each photo-
catalytic degradation cycle, the solution was mixed with
ofloxacin stock solution to restore the initial ofloxacin concen-
tration. 98.0% of the total ofloxacin molecules are degraded
after 90 min of photocatalytic reaction. There is a slight
decrease in degradation efficiency for the next cycles, and the
degradation efficiency is 93.5% in the fifth cycle.
Based on the experimental results in this work, we would
try to clarify photocatalytic activity enhancing mechanism of
the supported B-TiO 2(x%)/HZSM-5 composite. Even though the
essential purpose of preparing supported photocatalyst is to
facilitate water-solid separation after wastewater treatment, it
is absolutely interesting to enhance the activity of the material
at the same time. It is difficult to find a promising suppor with
such functions. HZSM-5 zeolite is an interesting material that
has sufficient microporous structure. The functions of HZSM-
5 zeolite in improving the activity of B-TiO 2must be complex.
As we understand in the current work, the dispersion of a
thin B-TiO 2layer on the external surface of HZSM-5 zeolite
plays the key role. More surface area of B-TiO 2is available in
B-TiO 2(x%)/HZSM-5 composite. The shrinkage of anatase TiO 2
crystals after supporting might have a minor influence either,
since it is believed that small TiO 2crystals have good activity.
Fig. 11 shows the UV–vis spectra of ofloxacin solution
during photocatalytic degradation on B-TiO 2(x%)/HZSM-5 andB-TiO 2. Photocatalytic oxidation not only causes the breaking
up
of ofloxacin molecules, but also results in decomposition
of the major functional groups of the pollutant. As shown in
Fig. 1, the main functional groups in ofloxacin molecule are the
quinolone substituent, the benzoic ring, and the piperazinyl
and oxazinyl groups. Ofloxacin molecule does not absorb pho-
tons
in the wavelength above 400 nm. There are four major
absorption peaks in the ultraviolet region in the UV–vis spec-
trum of original ofloxacin solution. The strongest absorption
peak at 288 nm is used to determine the concentration of
ofloxacin solution using a Agilent 1260 HPLC. This absorption
is caused by the quinolone substituent of ofloxacin molecule.
The absorption peaks situated at 226 nm and 330 nm are
attributed
to the piperazinyl and oxazinyl groups in ofloxacin,
while the other absorption at 255 nm is due to /H9266–/H9266* excitation
in benzoic ring [6,24] .
These functional groups in ofloxacin molecule can be
decomposed during photocatalytic oxidation process. The
absorption intensities of these groups are reduced gradu-
ally with extended irradiation time. The fastest ofloxacin
degradation rate appears in the solution containing B-
TiO 2(20%)/HZSM-5. On the other hand, all these four
absorption peaks disappear almost at the same time, and
it is hard to find a priority for individual functional group.
The oxidative reagents formed on the materials seem to have
enough power to degrade the whole ofloxacin molecule.
A brief mechanism for ofloxacin degradation can be pro-
posed. Electrons and holes are generated after absorbing the
incoming photons by the B-TiO 2(x%)/HZSM-5 composite. Since
these charge carriers have very short lifetime, the electrons
have to migrate to the surface of the material as soon as pos-
sible. The electrons can react with O2to produce •O2−, and
the holes in the valence band react with H2O to produce •OH.
These oxidative species take part in the subsequent degrada-
tion of ofloxacin.
4. Conclusions
B-TiO 2(x%)/HZSM-5 composites were prepared by loading B-
TiO 2on the surface of HZSM-5. TiO 2was in anatase phase in
both B-TiO 2and the composites. Anatase TiO 2crystal growth
was restrained after loading on the surface of HZSM-5 parti-

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
10 j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 11 – UV–vis spectra of ofloxacin solution during photocatalytic degradation on B-TiO 2(x%)/HZSM-5 and B-TiO 2. (a) 10%,
(b) 20%, (c) 50%, (d) B-TiO 2.
cles. The absorption edge of the B-TiO 2(x%)/HZSM-5 composite
moves to short wavelength region and shows a blueshift
with reduced B-TiO 2content. The N2adsorption-desorption
isotherms of B-TiO 2(x%)/HZSM-5 composites present both
the characters of mesopores and micropores. The microp-
ores in HZSM-5 zeolite are not blocked after loading a thin
B-TiO 2layer. The binding energies of both Ti2p and O1s
electrons in B-TiO 2layer are not affected by the zeolite. The B-
TiO 2(x%)/HZSM-5 composites have stronger activity than the
unsupported B-TiO 2, and B-TiO 2(20%)/HZSM-5 has the max-
imum efficiency on photocatalytic degradation of ofloxacin.
The functional groups in ofloxacin molecule can be decom-
posed during the photocatalytic oxidation process.
Conflicts of interest
The authors declare no conflicts of interest.
Acknowledgments
This work was supported by Scientific Research Project of Edu-
cation Department of Liaoning Province (No. LG201913).r e f e r e n c e s
[1] Hoffmann MR, Martin ST, Choi W, Bahnemann W.
Environmental applications of semiconductor
photocatalysis. Chem Rev 1995;95:69–96.
[2] Fujishima A, Rao TN, Tryk DA, Photochem J. Titanium
dioxide photocatalysis. J Photochem Photobiol C Photochem
Rev 2000;1:1–21.
[3] Dumrongrojthanath P, Thongtem T, Phuruangrat A,
Thongtem S. Synthesis and characterization of hierarchical
multilayered flower-like assemblies of Ag doped Bi2WO6 and
their photocatalytic activities. Superlattices Microstruct
2013;64:196–203.
[4] Kaur A, Gupta G, Ibhadon AO, Salunke DB, Sinha ASK, Kansal
SK. A Facile synthesis of silver modified ZnO nanoplates for
efficient removal of ofloxacin drug in aqueous phase under
solar irradiation. J Environ Chem Eng 2018;6:3621–30.
[5] Gupta G, Umar A, Kaur A, Sood S, Dhir A, Kansal SK. Solar
light driven photocatalytic degradation of Ofloxacin based
on ultra-thin bismuth molybdenum oxide nanosheets. Mater
Res Bull 2018;99:359–66.
[6] Zhang WJ, Yang J, Li CG. Role of thermal treatment on sol-gel
preparation of porous cerium titanate: Characterization and
photocatalytic degradation of ofloxacin. Mater Sci Semicond
Process 2018;85:33–9.
[7] Hapeshi E, Achilleos A, Vasquez MI, Michael C,
Xekoukoulotakis NP, Mantzavinos D, et al. Drugs degrading
photocatalytically: Kinetics and mechanisms of ofloxacin

Please cite this article in press as: Li H, et al. HZSM-5 zeolite supported boron-doped TiO 2for photocatalytic degradation of ofloxacin. J Mater
Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.12.086ARTICLE IN PRESSJMRTEC-1246; No. of Pages 11
j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx 11
and atenolol removal on titania suspensions. Water Res
2010;44:1737–46.
[8] Zhang WJ, Tao YJ, Li CG. Sol-gel synthesis of
Gd2Ti2O7/HZSM-5 composite photocatalyst for ofloxacin
degradation. J Photochem Photobio A: Chem 2018;364:787–93.
[9] Phuruangrat A, Maneechote A, Dumrongrojthanath P,
Ekthammathat N, Thongtem S, Thongtem T. Effect of pH on
visible-light-driven Bi2WO6 nanostructured catalyst
synthesized by hydrothermal method. Superlattices
Microstruct 2015;78:106–15.
[10] Bellardita M, Paola A, Megna B, Palmisano L. Absolute
crystallinity and photocatalytic activity of brookite TiO2
samples. Appl Catal B: Environ 2017;201:150–8.
[11] Zhang WJ, Bi FF, Yu Y, He HB. Phosphoric acid treating of
ZSM-5 zeolite for the enhanced photocatalytic activity of
TiO2/HZSM-5. J Mole Catal A: Chem 2013;372:6–12.
[12] Enesca A, Yamaguchi Y, Terashima C, Fujishima A, Nakata K,
Duta A. Enhanced UV–Vis photocatalytic performance of the
CuInS2/TiO2/SnO2 hetero-structure for air decontamination.
J Catal 2017;350:174–81.
[13] Ku Y, Shiu SJ, Wu HC. Decomposition of dimethyl phthalate
in aqueous solution by UV–LED/TiO2 process under periodic
illumination. J Photochem Photobio A: Chem
2017;332:299–305.
[14] Cai JB, Wu XQ, Li SX, Zheng FY. Controllable location of Au
nanoparticles as cocatalyst onto TiO2@CeO2 nanocomposite
hollow spheres for enhancing photocatalytic activity. Appl
Catal B: Environ 2017;201:12–21.
[15] Jo WK, Kumar S, Isaacs MA, Lee AF, Karthikeyan S. Cobalt
promoted TiO2/GO for the photocatalytic degradation of
oxytetracycline and Congo Red. Appl Catal B: Environ
2017;201:159–68.
[16] Zhang WJ, Liu YX, Pei XB, Chen XJ. Effects of indium doping
on properties of xIn-0.1%Gd-TiO2 photocatalyst synthesized
by sol-gel method. J Phys Chem Solids 2017;104:45–51.
[17] Du J, Li X, Li K, Gu X, Qi W, Zhang K. High hydrophilic
Si-doped TiO2 nanowires by chemical vapor deposition. J
Alloys Compd 2016;687:893–7.[18] Juma A, Acik IO, Oluwabi AT, Mere A, Mikli V, Danilson M,
et al. Zirconium doped TiO2 thin films deposited by chemical
spray pyrolysis. Appl Surf Sci 2016;387:539–45.
[19] Simsek EB. Solvothermal synthesized boron doped TiO2
catalysts: Photocatalytic degradation of endocrine disrupting
compounds and pharmaceuticals under visible light
irradiation. Appl Catal B: Environ 2017;200:309–22.
[20] Zhang W, Pei X, Yang B, He H. Effects of boron content and
calcination temperature on properties of B-TiO2
photocatalyst prepared by solvothermal method. J Adv Oxid
Technol 2014;17:66–72.
[21] Zhang W, Liu Y, Xin H. Sol-gel preparation of hollow spherical
x%B-TiO2 photocatalyst: the effect of boron content on RBR
X-3B decoloration. Curr Nanosci 2018;14:209–15.
[22] Cao B, Li G, Li HX. Hollow spherical RuO2@TiO2@Pt
bifunctional photocatalyst for coupled H2 production and
pollutant degradation. Appl Catal B: Environ 2016;194:42–9.
[23] Sun GH, Zhu CS, Zheng JT, Jiang B, Yin HC, Wang H, et al.
Preparation of spherical and dendritic CdS@TiO2 hollow
double-shelled nanoparticles for photocatalysis. Mater Lett
2016;166:113–5.
[24] Zhang WJ, Tao YJ, Li CG. Sol-gel synthesis and
characterization of Gd2Ti2O7/SiO2 photocatalyst for
ofloxacin decomposition. Mater Res Bull 2018;105:55–62.
[25] O’Neill MA, Cozens FL, Schepp NP. Photogeneration and
Migration of Electrons and Holes in Zeolite NaY. J Phys Chem
B 2001;105:12746–58.
[26] Butler MA. Photoelectrolysis and physical properties of the
semiconducting electrode WO2. J Appl Phys 1977;48:1914–20.
[27] Pouilleau J, Devilliers D, Groult H, Marcus P. Surface study of
a titanium-based ceramic electrode material by X-ray
photoelectron spectroscopy. J Mater Sci 1997;32:5645–51.
[28] Zhang GX, Song AK, Duan YW, Zheng SL. Enhanced
photocatalytic activity of TiO2/zeolite composite for
abatement of pollutants. Microporous Mesoporous Mater
2018;255:61–8.