Synthesis of layered perovskite Ag,F-Bi2MoO6/rGO: A surface plasmon resonance and oxygen vacancy promoted nanocomposite as a visible-light photocatalyst

Article abstract of DOI:10.1016/j.jphotochem.2019.04.046

Heterojunction z-scheme based Ag,F co-doped Bi₂MoO₆/reduced graphene oxide (Ag,F@BMO/rGO) photocatalysts were synthesized via a facile solvothermal method. The present work describes the improved photocatalytic activity of BMO/rGO nanocomposite by co-doping of F^? and Ag⁺ ions, to remove RhB from aqueous solution. The XRD, N₂ adsorption, SEM, TEM, EDS, UV–Vis DRS, FT-IR, Raman, and PL measurements were employed to characterize the crystallographic, morphological, and optical properties. X-ray diffraction analysis suggests that crystal growth of all the as-prepared nanoparticles with different F^? and Ag⁺ contents has occurred in Aurivillius phase and the crystal structure did not affected by doping. The insertion of Ag⁺ and F^? into Bi₂MoO₆ led to a red-shift in the absorption edge of nanocomposite and decrease the band gap energy from 2.78 eV to 2.6 eV, due to the synergetic effects of Surface Plasmon Resonance and surface oxygen vacancy induced by Ag⁺ and F^?, respectively. These beneficial properties are explored toward the photodegradation of RhB under visible-light source, resulting in better yields at lesser exposure time. The photocatalytic activity was significantly influenced by rGO in the nanocomposite, which was 2 times higher than that of pure Bi₂MoO₆, by effective separation of the charge carriers. The separation behaviors of photogenerated electron-hole were also systematically investigated by the PL. Based on the radical trapping experiments, photogenerated holes and O₂ ^?? were the main active species in RhB photodegradation and the detailed decolorization pathway has been suggested, using liquid chromatography/mass spectrometry (LC/MS) technique. In addition, the Ag,F@BMO/rGO nanocomposite does not display dramatic reduction of catalytic performance after four recycles, reveals its great prospect and promising application for water purification.

Full text of DOI:10.1016/j.jphotochem.2019.04.046

Accepted Manuscript
Title: Synthesis of Layered Perovskite Ag,F-Bi MoO /rGO: A
2
6
Surface Plasmon Resonance and Oxygen Vacancy Promoted
Nanocomposite as a Visible-Light Photocatalyst
Authors: Zeynab Khazaee, Ali Reza Mahjoub, Amir Hossein
Cheshme Khavar, Varsha Srivastava, Mika Sillanpaa
PII:
S1010-6030(19)30368-5
DOI:
Reference:
https://doi.org/10.1016/j.jphotochem.2019.04.046
JPC 11835
To appear in:
Journal of Photochemistry and Photobiology A: Chemistry
Received date:
Revised date:
Accepted date:
1 March 2019
25 April 2019
29 April 2019
Please cite this article as: Khazaee Z, Mahjoub AR, Cheshme Khavar AH, Srivastava
V, Sillanpaa M, Synthesis of Layered Perovskite Ag,F-Bi MoO /rGO: A Surface
2
6
Plasmon Resonance and Oxygen Vacancy Promoted Nanocomposite as a Visible-Light
Photocatalyst, Journal of Photochemistry and amp; Photobiology, A: Chemistry (2019),
https://doi.org/10.1016/j.jphotochem.2019.04.046
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Synthesis of Layered Perovskite Ag,F-
Bi MoO /rGO: A Surface Plasmon Resonance and
2
6
Oxygen Vacancy Promoted Nanocomposite as a
Visible-Light Photocatalyst
Zeynab Khazaee, Ali Reza Mahjoub*, Amir Hossein Cheshme Khavar, Varsha Srivastava, Mika
Sillanpaa
Department of Chemistry, Faculty of Basic Sciences, Tarbiat Modares University, Tehran, Iran.
Department of Green Chemistry, School of Engineering Science, Lappeenranta-Lahti University
of Technology, Sammonkatu 12, FI-50130 Mikkeli, Finland.
*
Corresponding Author
E-mail address: mahjouba@modares.ac.ir (Ali Reza Mahjoub)
Present Addresses
(Ali Reza Mahjoub) Department of Chemistry, Faculty of Basic Sciences, Tarbiat Modares
University, Tehran, Iran.
1

Graphical Abstract
Highligh
Heterojunction Ag,F co-doped Bi
2
MoO
6
/reduced graphene oxide photocatalysts were synthesized
led to decrease the band gap energy from 2.78 eV to
via a solvothermal-calcination method.
+
−
The insertion of Ag and F into Bi
.6 eV.
2
MoO
6
2
RhB removal by the novel photocatalyst exceeded 99% within 120ꢀmin.
The novel photocatalyst remained stable up to 4 consecutive runs.
2

ABSTRACT: Heterojunction z-scheme based Ag,F co-doped Bi
2
MoO
6
/reduced graphene oxide
(Ag,F@BMO/rGO) photocatalysts were synthesized via a facile solvothermal method. The present
work describes the improved photocatalytic activity of BMO/rGO nanocomposite by co-doping of
+
−
F and Ag ions, to remove RhB from aqueous solution. The XRD, N
2
adsorption, SEM, TEM,
EDS, UV–Vis DRS, FT-IR, Raman, and PL measurements were employed to characterize the
crystallographic, morphological, and optical properties. X-ray diffraction analysis suggests that
+
−
crystal growth of all the as-prepared nanoparticles with different F and Ag contents has occurred
+
in Aurivillius phase and the crystal structure did not affected by doping. The insertion of Ag and
−
F into Bi
2
MoO led to a red-shift in the absorption edge of nanocomposite and decrease the band
6
gap energy from 2.78 eV to 2.6 eV, due to the synergetic effects of Surface Plasmon Resonance
+
−
and surface oxygen vacancy induced by Ag and F , respectively. These beneficial properties are
explored toward the photodegradation of RhB under visible-light source, resulting in better yields
at lesser exposure time. The photocatalytic activity was significantly influenced by rGO in the
nanocomposite, which was 2 times higher than that of pure Bi
2
MoO , by effective separation of
6
the charge carriers. The separation behaviors of photogenerated electron-hole were also
systematically investigated by the PL. Based on the radical trapping experiments, photogenerated
•
−
holes and O
decolorization pathway has been suggested, using liquid chromatography/mass spectrometry
LC/MS) technique. In addition, the Ag,F@BMO/rGO nanocomposite does not display dramatic
2
were the main active species in RhB photodegradation and the detailed
(
reduction of catalytic performance after four recycles, reveals its great prospect and promising
application for water purification.
KEYWORDS: Photocatalytic degradation; Bi
2
MoO ; F-doping; Ag-doping; Organic pollutants
6
3

ABBREVIATIONS
BMO, Bi MoO ; rGO, Reduced Graphene Oxide.
2
6
1
. INTRODUCTION
Semiconductor photocatalysis as an advanced oxidation technology, has become one of the
most promising green methods for elimination all the dangerous pollutants, including azo dyes,
because of its high performance and mineralization efficiency for environmental purification [1].
Among the various photo activated materials, Bismuth based semiconductors such as Bi
BiVO [3], BiPO [4], Bi WO [5], and Bi MoO [6] exhibited great photodegradation efficiency
for removing azo dyes, because of their excellent visible light absorption and high chemical
stability. Among them, Bi MoO with the band gap of 2.7- 2.8 eV is an excellent choice in order
to photodegradion of organic pollutants [7]. This semiconductor is belonged to the Aurivillius
2
O
3
[2],
4
4
2
6
2
6
2
6
2
+
phases with a layered structure containing alternating (Bi
2
O
2
) layers placed between perovskite
) octahedral layers [8]. Its appropriate band gap energy, high inherent electrical and optical
properties, oxygen storage capacity and notable visible light harvesting, make it a promising
photocatalyst material. Nevertheless, the utilization of Bi MoO as photoctalyst under visible light
2
−
(
MoO
4
2
6
is restricted due to its low quantum yield, rapid recombination of photo-induced charge carriers,
slow carrier transportation and poor surface chemical states [9]. To overcome these problems,
different approaches such as design heterojunction structures, introducing crystal defects,
depositing noble metal nanoparticles and doping of metals or non-metals have been employed [6].
In the recent years, graphene oxide (GO) has been widely applied, in order to improve of
photocatalytic yield of various semiconductors [10]. GO as an inherent electron sink can play
4

critical role in the effective separation and transportation of the photogenerated electron-hole and
significantly reduce the charge carriers recombination in a heterogeneous photocatalytic oxidation
PCO system [11-13]. Su et al. [6] reported a ZnO/rGO composite and investigated the electron
transfer of accumulated electrons on the conduction band of ZnO to the rGO layer. Recently,
graphene co-modified Bi
graphene oxide presented strong adsorption for RhB and excellent charge carrier conductivity on
the Bi WO surface. Meanwhile, Vadivel et al [15] successfully synthesized Bi co-modified by
2
WO nanosheets were prepared by Sun et al. [14]. The deoxygenated
6
2
6
2
O
3
reduced graphene oxide to enhance photodegradation of MB. They showed that some of the
oxygen functional groups were reduced during the solvothermal process, improving the mobility
of electrons on the graphene-like layer.
Heteroatom doping has been proved to be an efficient method to introduce crystal defects into a
semiconductor. Crystal defects could change the electronic behavior of photocatalysts and improve
charge separation trough trapping the photogenerated electrons [16, 17]. Among the various
dopant agents, fluorine has been widely employed to improve the photocatalytic performance of
metal oxides [18-20]. Fluorine can be substituted with surface oxygen atoms and create surface
oxygen vacancies. The surface oxygen vacancies can act as trapping centers for photogenerated
electrons and reduce the charge recombination rate [19-21]. In addition, fluorine modification
increases the crystallinity and visible-light absorption [21, 22]. It was reported that F-doping can
improve generation of OH radicals, which is one of the most active species in photodegradation
process.[23] Recently, many fluorinated semiconductors such as F-doped TiO
F/Bi [25], F/BiPO [4], and F/Bi WO [26] has been reported. Based on above mentioned works,
surface fluorination can remarkably enhance the photodegradation efficiency, due to produce more
2
(F/TiO ) [24],
2
2
O
3
4
2
6
5

surface hydroxyl groups, modification of electronic structure and improve charge separation. Also,
Noble metals (Au, Ag etc.)-doping has been investigated as another efficient method for enhancing
the photocatalytic efficiency of various semiconductors, due to their localized surface plasmonic
resonance (LSPR) effect [27-29], which can highly improve visible light absorption and regenerate
of charge-carriers. This phenomenon is attributed to d–d transitions of noble metals [30], which
increases the molar extinction coefficients [31]. Previous reports revealed the enhanced visible
light absorption efficiency by Ag-doping due to the LSPR effect of silver [32]. Furthermore, Ag-
modified semiconductor can act as electron acceptor site to facilitate electron transfer and electron-
hole separation, resulting in reduce of recombination rate and enhance of quantum efficiency [33].
Some novel and efficient Ag-doped photocatalysts, e.g., Ag/TiO
36], and AgI/TiO [37], showed enhanced visible light photocatalytic degradation. Very recently,
Huynh et al. [38] synthesized a novel Ag, F-modified ZnO. They found that the synergic effects
2
[34], Ag/ZnO [35], Ag/Bi
2
WO
6
[
2
+
−
0
+
of Ag and F created silver species (Ag and Ag ) and zinc vacancies, and generated more free OH
radicals on the surface of ZnO, leading to a remarkable improvement in its photocatalytic activity.
As the mentioned above, the photocatalysts usually confront with various challenges including
unsuitable electronic structures [39], fast charge recombination [40], and low light absorption
capacity [41]. However, in many cases, using single dopant resolves one of these challenges, while
the other remains unanswered. To further achieve optimal photocatalytic activities, double or triple
doped matrixes [42] can be used as a prevalent method to modify the optical efficiency. The co-
doping of two elements into a photocatalyst such as F/Ca and La/F, co-doped-TiO
dramatic improvement on the efficient photodecomposition of contaminants [43, 44]. Recently,
Ce/F co-doped-Bi WO photocatalysts with synergistic effect for enhancing the photoctalytic
2
, can make
2
6
6

performance were reported by Huang et al. [43]. Meanwhile, Fu et al. [6] successfully synthesized
a novel La/F co-doped Bi MoO photocatalyst via one-step hydrothermal method which
significantly removed RhB (10 M) within 80 min under visible light irradiation. They found that
the content of Ce and La dopants in Bi WO and Bi MoO lattice could enhance photocatalytic
2
6
−4
2
6
2
6
performance, due to their remarkable influence on the separation of charge carriers.
As the mentioned above, doping a metal oxide with Nobel metals or fluorine could create several
benefits and significantly improve quantum yield. Therefore, when a Noble metal such as silver
and a heteroatom like fluorine combine to modify a metal oxide, the synergic effects of high visible
light absorption (LSPR properties), effective charge separation (silver species /defect
levels/vacancies) and production of more hydroxyl radicals, can significantly improve the
photocatalytic activity of the pure metal oxide under visible light illumination [45,46].
Accordingly, the unique properties of Ag and F as dopant along with inherent characteristics of
GO and Bi
2
MoO encouraged us to design a novel Ag,F co-doped BMO/rGO photocatalyst as an
6
efficient visible light activated catalyst for degradation and mineralization RhB dye in the PCO
process. The effects of fluorine and silver ions doping on the electronic structures and optical
properties of Bi
displayed a remarkable enhance in photodegradation efficiency of RhB compared with pure
Bi MoO . Also, the mechanism of photocatalytic degradation of RhB over the co-doped catalyst
was discussed according to the results from the radical capture experiment and LC/MS technique.
2
MoO
6
was investigated. Under visible light irradiation, the co-doping Ag,F-BMO
2
6
2
2
. EXPERIMENTAL SECTION
.1. Materials
7

Bi(NO
3
)
3
.5H
2
O, Na
2
MoO
4
.2H
2
O, NaF, AgNO , Ethanolamine, and Ethylene glycol were
3
purchased from Merck Co. and used without further purifications. Double distilled water (DW)
was used in all experiments.
2
.2. Synthesis of Ag, F co-doped Bi
2
MoO
6
Pure BMO and silver and/or fluorine doped samples were synthesized via a solvothermal-
calcination method. For F and Ag doped Bi MoO , a typical synthesis procedure is as follows: 1
mmol of Na MoO .2H O, 2 mmol of Bi(NO .5H O, stoichiometric amounts of NaF (with the
molar ratio of F to Mo at 0.50, 0.75, 1 respectively) and AgNO (with the molar ratio of Ag to Mo
2
6
2
4
2
3
)
3
2
3
at 0.005, 0.01, 0.02, 0.05, 0.10 respectively), was dissolved in 70 mL of ethylene glycol. The pH
was adjusted to 7.0 using Ethanolamine. After stirring at room temperature for 30 min, the
resulting pale-yellow suspensions were transferred into a Teflon-lined stainless autoclave and
ᵒ
heated to 160 C for 12 h. After natural cooling, the obtained powder was collected, washed
repeatedly with deionized water and ethanol and dried at 80 °C for 10 h in a vacuum oven. Finally,
◦
◦
the powder was placed in a furnace and calcined at 400 C (1 C/min) for 2 h. The sample with x
mol.% of Ag and y mol.% of F was labeled as Ag @BMO. Also, the pure BMO was synthesized
based on the above procedure without addition of NaF or AgNO into precursor solution.
x
F
y
3
2
.3. Preparation of AgF@BMO/rGO nanocomposite
The graphene oxide was prepared by chemical oxidation and exfoliation of graphite flakes,
based on an improved Hummers’ method [45]. The AgF@BMO/rGO nanocomposite was
prepared by a one-pot solvothermal method. Typically, 0.0075 g of graphene oxide (0.5 wt.%) was
first added into a 4:1 mixture of EtOH/H O and ultrasonicated for 1 h. Then, 1.5 g AgF@BMO
2
8

was added to the GO suspension. The mixture was stirred (30 min) and sonicated (30 min) for four
times continuously and then, transferred into a Teflon-lined stainless autoclave and heated to 130
◦
C for 6 h. Finally, the autoclave was allowed to cool naturally to room temperature and the product
AgF@BMO/rGO was filtered out and several times washed with ethanol and deionized water and
◦
dried at 60 C for 6 h.
2
.4. Characterization
X-ray diffraction (XRD) analysis was performed on a STOE power diffraction system at 40 kV
and 40 mA equipped with a Cu/Kα(λ=1.54060 Å) radiation source. XRD spectra were recorded in
a 2θrange of 1-80 0. A Micromeretic/Gemini-2372 surface area analyzer was used to measure the
Brunauer–Emmett–Teller (BET) surface areas of the samples using multipoint estimation. Also,
the pore volumes were obtained based on volumes of adsorbed N
2
at p/p = 0.977; and the Barrett–
0
Joyner–Halenda (BJH) method was applied to calculate the pore size data. Scanning electron
microscopy (SEM, XL30 Philips) and a transmission electron microscope (TEM, Philips-CM30
instrument) images were recorded to study the morphologies of as-prepared photocatalysts. UV–
Vis diffuse reflectance spectra (DRS) were measured using a Shimadzu-UV-2550
spectrophotometer. Elemental analysis of catalyst samples obtained by using X-ray photoelectron
spectroscopy (XPS, ESCALAB 250) with 1486.6 eV Al-K x-ray source for further investigation
of the element composition. Total organic carbon (TOC) was studied with a Multi N/C 3100 TOC
analyzer. Fourier transform infrared spectroscopy (FTIR) were performed on a Shimadzu-8400S
spectrometer in the range of 400- 4000 cm-1 using KBr pellets. Raman spectra were recorded by
using an Almega Thermo Nicolet Dispersive Raman Spectrometer (Laser second harmonic @532
nm of an Nd:YLF laser). The UV/Vis absorption spectra were examined on a 4802 UV/Vis
9

spectrometer Unico in the wavelength range of 200-800 nm. Room temperature
photoluminescence spectra (PL) were detected with an RF6000 Shimadzu fluorescence
spectrophotometer. The zeta potentials of Ag0.5F75@BMO suspension were measured using a
Malvern Zetasizer Nano ZS particle analyzer. The RhB intermediates generated in the
photodegradation process were analyzed with a Shimadzu LCMS 2010 A and Eclipse Atlantis T3
C18 3µ, 2.1×100 mm column in the mixed eluent (A: Acetonitrile 0.1% Formic Acid, and B: H
.1% Formic Acid by volume).
2
O
0
2
.5. Photocatalytic experiments
Photocatalytic performance of pure BMO and doped samples was quantified in a photochemical
reactor. A 400W Mercury-vapor lamp was used as the visible-light source (λ > 420 nm). The
distance between lamp and photoreactor was 30 cm. In a typical photocatalytic experiment, a given
amount of photocatalyst was suspended in 50 mL RhB (10 mg/L) aqueous solution with a known
pH. Before irradiation, the mixed solution was kept in the dark for 30 min under magnetic stirring
to establish an adsorption–desorption equilibrium and obtain a good dispersion. After that, the
suspended solution was illuminated by visible light source under stirring (500 rpm). The system
was maintained at room temperature by circulating water. At certain intervals, 3 mL of solution
was taken and the photocatalyst was separated through centrifugation. The concentration of RhB
during the photodegradation was monitored by measuring the UV-Vis maximum absorption band
at 556 nm. The effect of Ag,F content and rGO ratio, was evaluated on the photocatalytic
performance of BMO. The kinetics of RhB photodegradation and mineralization was studied in
the PCO process by optimum catalyst (Ag,F@BMO/rGO). The mechanism of enhanced
photocatalytic activity was investigated through the scavenging tests by using three sacrificial
1
0

agents, p-benzoquinone, isopropanol, and ethylenediaminetetraacetic acid disodium salt. The
reusability of Ag,F@BMO/rGO was evaluated after 4 recycling runs reusing the optimum
photocatalyst in the oxidation process under the same condition.
3
. RESULTS AND DISCUSSION
In this research, three different series of photocatalyts based on bismuth molybdate were
prepared, by employing silver and fluorine as dopants, to improve the photocatalytic properties.
Bismuth molybdates are crystallized in three phases, α-Bi
46] In this work, γ-Bi MoO , only with an Aurivillius structure, was studied because of its
photocatalytic activity under visible-light irradiation. The Ag,F co-doped Bi MoO
Ag @BMO) samples were synthesized by one-pot solvothermal method and using EG as
2
Mo
3
O12, β-Bi
2
Mo
2
O
9
, and γ-Bi
2
MoO
6
[
2
6
2
6
(
F
x y
solvent. Also, EG could act as a coordination agent to grow 3D structures and reductant agent to
reduce the oxygen functional groups of GO under solvothermal condition. Then, we optimized the
dosage of each dopants and also the pH of solution to obtain the best condition for
photodegradation of organic pollutants. Rhodamine B (RhB) as one of the most commonly used
dyes which causes serious environmental and biological problems, was used for photodegradation
experiments. Optical properties were studied to show the modified photocatalytic activity of BMO
+
−
nanoparticles by doping of Ag and F . In addition, the structure of the as-synthesized
nanoparticles was investigated to understand the effect of mono and co-doping on the crystal
structure and morphology.
3
.1. Texture properties
1
1

To study the crystal structure, X-ray diffraction analysis of the pure BMO and doped BMO
samples was carried out. As shown in Fig. 1a, synthesized samples exhibited well crystalline nature
and no characteristic peaks related to other impurities could be observed in the XRD patterns. The
◦
◦
◦
◦
diffraction peaks at 2θof 28.3 , 32.6 , 47.2 and 56.2 well correspond to the crystal planes of (131),
200), (202) and (133) of orthorhombic Bi MoO phase (JCPDS No.72-1524), respectively [43].
Meanwhile, XRD patterns of the F75@BMO, Ag0.5@BMO and Ag0.5 75@BMO were quite similar
(
2
6
F
to those of pure BMO and the peaks remained at the same angles. Based on the Fig. 1b, (the
magnified (131) peak), no systematic peak position shift is observed for the F-doped sample,
indicating that the lattice and crystal phase of Bi
a peak position shift is observed for the Ag-doped sample, slightly toward a lower 2θ value. The
same shift is also observed for Ag0.5 75@BMO sample, indicating that the crystal planes of BMO
2
MoO
6
were not affected by inducing F. whereas,
F
were affected by the substitution of Ag and F dopants.
As observed, a slight decreasing in the peak intensity of F75@BMO and Ag0.5@BMO confirms
that a decrease in crystallinity was occurred, and nanoparticles with smaller size was obtained [47].
+
−
However, upon co-doping with Ag and F in Ag0.5
peaks increased, suggesting an increase in crystallinity. These observations are in agreement with
those reported by Lin et al. [25]. Meanwhile, XRD pattern of Ag0.5 75@BMO/rGO exhibited quite
F75@BMO sample, the intensity of diffraction
F
similar peak position and no specific peak can be observed for carbon, which may be due to low
amount of GO (0.5 wt.%) as well as low crystallinity of GO. Also, it can be suggest that the crystal
structure of Ag0.5F75@BMO is not affected by loading of graphene oxide.
1
2

According to the Bragg equation, d(hkl) = nλ/(2sinθ), where d(hkl), λ, and θ are the space between
crystal planes of (h k l), the X-ray wavelength, and the Bragg diffraction angle, respectively, the
lattice parameters of pure Bi
2
MoO were calculated as a = 5.506 Å, b = 16.226 Å, and c = 5.487
6
Å. The slight shift to a lower angle reflects the lattice expansion of the crystals, and well suggests
that the insertion of interstitial dopants could increases the lattice parameters. The similar lattice
−
+
parameters of un-doped and doped-samples further confirms that inserting of F and Ag ions have
−
2−
not a considerable influence on the crystal structure of BMO [26]. Since, F and O possesses the
−
2−
+
similar ionic radii (119 and 126 pm, respectively) F could replace with O in the Bi
2
O
2
layers or
+
substituted with the surface oxygen atoms [43]. In contrast, due to larger ionic radius of Ag (129
3
+
+
pm) compared with Bi (117 pm), Ag couldn’t integrate into the crystal structure, and it is only
+
3+
inserted to the surface of BMO by replacement of Ag ions with Bi ionic sites on the surface [48].
The specific surface area, pore volume, and average pore size of pure Bi
2
MoO and as-prepared
6
samples were detected by N adsorption–desorption analysis, and the results are summarized in
2
Table 1. For the pure BMO, F75@BMO, Ag0.5@BMO, and Ag0.5F75@BMO, the surface area slightly
2
decreased from 22.08 to 18.61, 16.78 and 17.78 m /g respectively. In addition, an increase in
average pore size can be observed. The results may be attributed to the presence of Ag and F on
the surface of Bi
2
MoO ; because the loading of Ag and F on the surface causes a block of some
6
pores and consequently, decreasing the surface area. As observed, Ag0.5
F
75@BMO/rGO
75@BMO sample (17.78
F75@BMO nanoparticles on the graphene sheets
2
−1
nanocomposite showed higher SBET (20.59 m .g ) than that of Ag0.5
F
2
−1
m .g ). The results exhibit that loading of Ag0.5
leads to an enhancement in the specific surface area due to the high inherent surface area of
graphene oxide, resulting in the improvement of adsorption capacity and photocatalytic activity.
1
3

Fig. 1c shows FTIR spectra of as-prepared GO, pure BMO samples and Ag,F@BMO/rGO
1
−
nanocomposite. The absorption peaks at 400-1000 cm were assigned to be the Bi-O stretching,
Mo-O stretching, and Mo-O-Mo bridging stretching modes. Typically, the band located at 416
1
−
cm is corresponded to the Bi-O vibration. As observed, the characteristic bands at 547 and 717
1
−
cm are attributed to the bending and asymmetric stretching of Mo-O involving vibrations of the
1
−
equatorial oxygen atoms in MoO
6
octahedrons. Also, the absorption bands at 771 and 809 cm are
related to the asymmetric and symmetric stretching modes of MoO
6
vibration of the apical oxygen
1
−
atoms, respectively [49]. The bands in the range of 1000-1700 cm can be attributed to impurities
1
−
of C=O and C-O stretching in CO
2
[50]. The broad band centered at 3200-3600 cm and 1630
1
−
cm are involved with hydroxyl group of water, adsorbed on the surfaces of the as-prepared
samples [51].
The FT-IR spectrum of synthesized graphene oxide showed the broad absorption peak at 3440
1
1
−
−
cm , due to the stretching vibration of hydroxyl (-OH) group. The peak at nearly 1095 cm , 1250
1
1
cm− and 1727 cm , corresponded to the stretching vibration bands of oxygen containing
−
1
1
−
−
functional groups including C-O, C-OH and C=O. The absorption peaks of 1650 cm , 2940 cm ,
1
−
and 1396 cm can be related to the C=C stretching vibration, the CH
2
stretching vibration and the
C-O-C deformation vibration, respectively [52]. Comparatively, some peaks related to the oxygen
functional groups (-OH, C=O, C-O-C, and C-OH) were removed and the intensities of other peaks
were weakened in the FT-IR spectrum of Ag0.5
F75@BMO/GO, indicating that graphene oxide has
been reduced to rGO during the solvothermal process in the presence of ethylene glycol. Hence, it
1
4

can modify the electrical conductivity of the nanocomposite and improve the mobility of electrons
on the graphene-like layer [53].
To further investigate the element substitution on BMO and the distortion of crystal structure in
Ag,F co-doped BMO nanostructures, Raman spectra of GO and prepared nanocomposites were
performed. Raman patterns of GO, BMO and Ag0.5F75@BMO/rGO samples are shown in Fig. 1d.
1
−
The band at 147 cm can be related to the contribution of Bi-O stretches and lattice modes. The
1
−
bands observed in the 180–500 cm region, originates from the bending modes of the MoO
6
octahedra which could be coupled with both stretching and bending modes of the bismuth-oxygen
1
−
polyhedral. The modes at 286 cm , are attributed to the E
u
symmetry bending modes. The Raman
1
1
1
−
−
−
modes at 798 cm (A1g mode) with two shoulders (718 and 844 cm ) and 844 cm (A2u mode)
correspond to the symmetric and asymmetric stretching vibrations of a MoO octahedron which
6
are involved with vibrations of the apical oxygen atoms [54]. All the as-prepared samples
1
−
demonstrated similar vibration modes between 100 and 900 cm , and no detectable Raman shifts
+
after doping of Ag and F can be found. Raman spectrum of Ag0.5
F
75@BMO indicates that Ag and
F− substitution did not affect the coordination sphere around Mo atoms (the MoO
octahedron),
6
+
and the dopants are only substituted onto Bi
2
O
2
layers on the surface.
1
−
The two characteristic peaks of graphene oxide are located at around 1358 cm (D band) and at
1
−
1
604 cm (G band) for the graphitized structure. The G band represents the scattering of the E2g
2
3
mode of sp carbon atoms, and D band is common for disordered sp C [55]. The D/G integral
2
intensity ratio (I
/I
D G
) correlates with the disorder degree and the average size of the sp domains.
The I /I was increased from 0.66 for GO to 0.99 for Ag0.5
D
G
F
75@BMO/GO sample. This is possibly
1
5

attributed to a decrease in the amount of oxygen functional groups in GO during the solvothermal
process, which is in agreement with FT-IR results [39]. Moreover, the G band of the
1
−
nanocomposite is slightly blue-shifted (1599 cm ), which further confirms the reduction of GO
under selected synthesis method [39].
3
.2. Morphologic structure
The morphology and microstructure of pure BMO and the as-synthesized Ag and F doped
nanostructured materials investigated by SEM and TEM. The size of synthesized samples was in
the range of 25-75 nm (Fig. 2). As shown in Fig. 2a, pure BMO was seen to possess nanoparticles
with no regular morphology and diameters of approximately 50 nm, with a smooth surface. When
the fluorine was doped into BMO lattice, the particle size and assemble morphologies were not
significantly different from the pure BMO, (Fig. 2b), but, the particles became slightly smaller.
For the Ag-doped sample, the final products had a spongy texture which was composed from very
small nanoparticles, as shown in Fig. 2c. It can be seen that the sample consists of Ag showed
slightly agglomerated morphology and the particles possess smooth surface, leads to a decrease in
surface area. However, with the coexistence of Ag and F in Ag0.5F75@BMO, the differences in the
particle size and morphology were not considerable compared with the pure BMO (Fig. 2d), but it
illustrated a slight increasing in particle size compared to those of individually F and Ag doped
samples, which is in good agreement with the results from XRD and BET. Fig. 2e illustrates the
exfoliated and crumpled sheets of layered graphene oxide, which is a typical morphology for this
material [56]. The layered structure of GO was remained after solvothermal process, while
1
6

nanoparticles were attached on the surface of rGO as shown in Fig. 2f
,
which could lead to an
increase in surface area of Ag0.5
F
75@BMO.
Fig. 2g and 2h illustrate TEM images of Ag0.5
F
75@BMO/rGO which confirms that the prepared
75@BMO particles with average
nanocomposite is composed of the high amounts of Ag0.5
F
diameters between 10 and 25 nm. TEM images showed the occupation of the most available
surface area of reduced graphene oxide by nanoparticles, while rGO serves as a matrix for the
Ag0.5
F
75@BMO nanoparticles. This layered structure provides a large amount of particles in
nanocomposite. To further identify the presence of F and Ag in Bi MoO , the chemical
composition and elements distribution of sample Ag0.5 75@BMO/rGO investigated by EDS-
2
6
F
mapping microanalysis as shown in Fig. 2 i-l, confirming that the sample contains Bi, Mo, F, and
Ag elements.
3
.3. Optical properties
UV-Vis absorption spectra of pure BMO, F75@BMO, Ag0.5@BMO, Ag0.5F75@BMO and
Ag0.5F75@BMO/rGO are presented in Fig. 3a. The band gap energies of the as-synthesized
1
/2
photocatalysts were obtained by plotting (αhν) against hν (the photonic energy), where α is
the optical absorption coefficient. It was found that Ag and F dopants have approximately
similar effect on the optical properties of BMO and a red shift in the band gap energy is shown
for both Ag and F-doped samples. The absorption edge of pure BMO was around 470 nm. The
Ag and F modified samples exhibited a red-shift in the absorption edge (480 nm and 500 nm,
respectively), which could be attributed to the band gap transition [10]. It can be observed that
1
7

silver and fluorine doping led to a slight decrease in the band gap energy from 2.78 eV (for
pure BMO) to 2.75 eV and 2.65 eV, respectively (Fig. 3b). In fluorinated BMO, the decrease
in the band gap energy could come from the hybridization of F 2p and O 2p states in the valence
band and creating a half-filled state in the band gap by F impurity as hole carrier [57]. Also,
+
when Ag was introduced, the absorption capacity in the visible light region slightly improved.
This improvement in the band gap energy is attributed to the strong surface plasmon resonance
effect of Ag on the surface of BMO [28]. As seen in Fig. 3c, UV-Vis absorption spectrum of
Ag
6
@BMO shows two new absorption band at 550 nm and 625 nm which can be assigned to
the presence of Ag and its LSPR effect [58].
When Ag0.5
F
75@BMO was combined with rGO, the ability of harvesting visible light energy was
enhanced. As seen, the absorption edge of as-prepared graphene oxide nanocomposite enhanced
to 570 nm, which reduce the band gap from 2.60 eV (for Ag0.5 75@BMO) to 2.50 eV (for
75@BMO/rGO).This enhancement could be due to the chemical bonding between
75@BMO and graphene oxide. Moreover, when the oxygen functional groups reduce under
F
Ag0.5
Ag0.5
F
F
solvothermal condition, more carbon atoms could bond with BMO and improve the light
absorption intensity [59]. Also, to realize the value of the valence and conduction band’s edge
potential (EVB and ECB) of pure BMO and Ag0.5
using the following formula:
F
75@BMO/rGO, the band positions were calculated
E
VB = X − E
e
+ 0.5E
g
(1)
(2)
ECB = EVB – E
g
1
8

where X is denoted as the absolute electronegativity of the semiconductor, E
of free electrons on the hydrogen scale (∼4.5 eV), and E refers to the band gap energy of the
semiconductor [60]. Therefore, the EVB values of pure BMO and Ag0.5 75@BMO/rGO can be
e
stands for the energy
g
F
determined as +3.19 and +3.05 eV, and the corresponding ECB values calculated as +0.41 and +0.55
eV.
3
.4. Composition and chemical state analysis
To investigate the elemental composition and the oxidation states of the as-prepared
75@BMO sample XPS spectra of Ag0.5 75@BMO was carried out and shown in Fig. 4. The
full XPS survey spectrum (0-1300 eV) (Fig 4a) clearly indicates that Ag0.5 75@BMO is composed
of Bi, Mo, O, F, Ag, and C elements, implying that the high purity of Ag0.5 75@BMO. Fig. 4b−e
Ag0.5
F
F
F
F
demonstrates the high-resolution XPS spectrum of Bi 4f, Mo 3d, O 1s, F 1s, and Ag 3d regions,
respectively. In Fig 4b, the XPS peaks of Bi 4f show the binding energies of Bi 4f7/2 and Bi 4f5/2
at 159.0 and 164.4 eV respectively, which are originated from the oxidation state of Bi3+ in
Bi
2
MoO
6
. In Fig. 4c, peaks at 232.3 and 235.3 eV are corresponding to Mo 3d3/2 and Mo 3d5/2
,
6
+
respectively, which assigned to the Mo in MoO
6
octahedron [61]. Fig. 4d shows two kinds of
oxygen species with the binding energy at 529.8 and 536.1 eV, corresponding to the lattice oxygen
and surface hydroxyl (or crystal water), respectively [62]. Fig. 4e demonstrated that there was an
overlap of two peaks of F 1s at the binding energy of 681.0 and 686.5 eV, which can be attributed
−
to F anions adsorbed on the surface of photocatalyst, and the substituted Fluorine atoms that
occupied oxygen sites in the crystal lattice [63]. In Fig. 4f, the Ag 3d3/2 and Ag 3d5/2 peaks at 368.0
+
1
and 374.0 eV suggest the presence of silver in Ag0.5
F
75@BMO in the state of Ag [64].The results
+
−
confirm that there are both Ag and F species in the nanocomposite system.
1
9

3
.5. PL properties
To further understand the recombination process of charge carriers in samples, the
photoluminescence (PL) properties of prepared samples were investigated. Photoluminescence
experiment of semiconductor photocatalyst reflects the trapping of charge carriers, migration, and
transfer of electron-hole pairs. For as-prepared photocatalysts, the room temperature PL spectra
are illustrated in Fig. 3d. When the excitation wavelength was 400 nm, the PL spectrum of pure
BMO demonstrated two peaks centered at 440 and 465 nm in the range of 400–550 nm. The blue
emission peak at 465 nm was related to the inherent luminescence of BMO, which originates from
the photo-induced charge transfer between the hybrid of Bi 6s and O 2p (VB) orbitals to the empty
4
d orbital of Mo (CB) in MoO octahedra [6]. The strongest blue emission peak at 440 nm ascribed
6
3
+
to the transition from 3P1 (6s6p) excited state to the 1S0 (6S2) ground state of Bi ions [65]. As
observed, with the doping of Ag and F atoms into BMO phase, the intensity of the PL spectra was
obviously decreased comparing to the pure BMO, revealing the significant inhibition of
recombination of photogenerated charge carriers by F and Ag doping. In F75@BMO, because of
2
−
−
stronger affinity of F ions for electrons compared with O , the electronic cloud around Bi-O
polyhedron could be deviated, which leads to weaker Bi-O bonds and replacing more F atoms [47].
Therefore, with illumination of visible light, more oxygen vacancies can be generated. In addition,
−
these oxygen vacancies induced by the F ions on the surface of BMO can hold the trapped
electrons, leading to improve the interfacial charge-transfer rates [47]. Meanwhile, for
Ag0.5@BMO, the lower PL intensity revealed that the doping of Ag can decrease the probability of
the recombination rate of photogenerated electron and hole by the LSPR effect, which can highly
2
0

inhibit the chance of the recombination of charge carriers. As shown, Ag0.5
displayed the lowest relative PL intensity, and based on the following formula:
F
75@BMO/rGO
PL ꢁꢂtꢃꢂꢄꢁty of ꢅꢆꢇꢃ BMO−PL ꢁꢂtꢃꢂꢄꢁty of Naꢂocomꢅoꢄꢁtꢃ
%
suppressing =
× 100%
(3)
PL ꢁꢂtꢃꢂꢄꢁty of ꢅꢆꢇꢃ BMO
implies the lowest recombination rate of electron-hole pair by suppressing the recombination to
7
7% compared to the pure BMO. Reduced graphene oxide sheets can act as an electron shuttle to
accept photogenerated electrons from the BMO conduction band through interfacial interactions.
Accordingly, the charge recombination rate could be remarkably diminished in the
Ag0.5F75@BMO/rGO nanocomposite as compared with that in the pure BMO.
3
.6. Evaluation of photocatalytic activity of the as-prepared catalysts:
The photocatalytic performances of the as-prepared samples were evaluated by studying the
photodegradation of the aqueous solution of RhB dye under visible light irradiation. Before
irradiation, the suspensions containing 25 mg of photocatalyst and 50 mL aqueous solution of RhB
-
1
(
10 mg.L ) were magnetically stirred under dark condition for 30 min to establish an
adsorption/desorption equilibrium. When the visible light source was turned on, at certain
intervals, 3 mL of solution were taken and the photocatalysts were separated through
centrifugation. The concentration of RhB during the photodegradation was monitored by
measuring the UV-Vis absorption through its maximum absorption band.
Fig. 5 shows the concentration profiles of RhB (C/Co) under photodegradation process. As
shown, in the dark condition, only less than 20% of RhB adsorbed onto the surface of
photocatalysts. The removal efficiency for the pure BMO was 45% after 120 min photocatalytic
2
1

reaction time (λ > 420 nm). F-doped BMO samples with different molar ratio of fluorine to Mo at
.50, 0.75, and 1 were examined under visible light irradiation. The photodegradation efficiency
0
for F50@BMO, F75@BMO, and F100@BMO are presented in Fig. 5a. As can be seen, the degree of
−
photodegradation varied with change of F content in the catalyst, and after 120 min illumination,
the removal efficiency was 79%, 84% and 59% for F50@BMO, F75@BMO, and F100@BMO
+
−
samples, respectively. To employ the synergistic effect of Ag and F for improving the
photocatalytic performance, three different molar ratio of Ag, 0.5, 1, and 2 mol.%, were added into
the optimum fluorinated BMO (F75@BMO). As viewed in Fig. 5b, the corresponding RhB removal
for Ag0.5
F
75@BMO, Ag
1
F
75@BMO, and Ag
2
F
75@BMO were 93%, 79%, and 75%, respectively. To
evaluate the effects of the composition of rGO sheets on photocatalytic activity, 0.5 wt% of GO
was added to the optimized Ag0.5 75@BMO sample. It can be observed that the photodegradation
efficiency has markedly improved and reached to 100%. Based on the variations in RhB
concentration profiles (changes in C /C ), it was found that all the Ag, F, and rGo modified BMO
samples exhibited much better photocatalytic performance than pure BMO. As shown Fig. 5b, the
ordering of photocatalytic activity was Ag0.5 75@BMO/rGO > Ag0.5F75@BMO > F75@BMO >
F
x
0
F
Ag0.5@BMO > BMO. In addition, the adsorption rate of rGO in the system obtained as 30%
percent, which refers to the strong adsorption capability of graphene oxide. As seen in Fig. 5d, the
main absorption peak of RhB at 556 nm is gradually shifted to 498 nm, indicating the destruction
of the conjugated structure and N-deethylation of RhB dye which is in good agreement with
2
2

quickly changing in the color of solution from bright red to colorless under visible-light irradiation
66].
[
To study the reaction kinetics of the photodegradation of RhB, a pseudo-first-order rate constant
can be ascribed based on the Langmuir-Hinshelwood (L-H) kinetics model [67] as shown in the
following equation:
Ln (C
0
/C
t
) = kapp
t
(4)
1
−
where kapp is the apparent pseudo-first-order rate constant (h ), C
0
and C are the initial
t
concentration of RhB (mol/L) and the instantaneous concentration of RhB after irradiation for
different time intervals. The apparent rate constants of as-prepared photocatalysts in
photodegradation process of RhB are exhibited in Fig. 5e. The corresponding kapp values are
3
2
2
2
2
1
−
−
−
−
−
−
4
.98×10 , 1.64×10 , 1.26×10 , 2.22×10 , and 4.32×10 min for pure BMO, F75@BMO,
Ag0.5@BMO, Ag0.5
F
75@BMO, and Ag0.5
F
75@BMO/rGO respectively. It implies that the
1
−
photodegradation rate constant of RhB over Ag0.5
F
75@BMO/rGO (0.02216 min ) is 8.7 times
1
−
higher than the pure Bi
2
MoO
6
(0.00498 min ) under visible light illumination. The results clearly
showed that the doping of BMO with Ag and F and deposition on the rGO sheets obviously
promoted its photocatalytic activity under visible light irradiations.
In F-doped BMO samples the photodegradation efficiency was gradually enhanced as the F
content was increased from 0 to 0.5, and 0.75 mol.%, and with further increase of fluorine molar
2
3

ratio to 1, the photocatalytic activity decreased, which indicates that fluorine doping is an effective
2
−
−
method to improve the photocatalytic performance of BMO. The substitution of O anion by F
could induce oxygen vacancies in BMO and also decrease the band gap energy by creating a half-
filled state in the band gap of BMO. Oxygen vacancies and adsorbed fluorine anions on the surface
−
of BMO can trap the photogenerated electrons, due to the high electron affinity of F and reduce
the charge recombination rate. Surface doped fluorine can enhance interfacial electron-transfer
rates, leading to improve the separation of charge carriers, and consequently, more active radicals
+
•
•
(
h , OH, O
can increase the amount of •OH. The possible route would be that the electrons on the surface of
BMO can be trapped by adsorbed H O and O to form •OH [68]. However, excessive quantities
2
−) are available for photodegradation of RhB [47]. As mentioned, surface F-doping
2
2
of F on the surface will decrease the photocatalytic activity of catalyst. F doping can increase the
surface acidity, which leads to detachment of pollutant from surface due to pH decreasing [69].
As observed in Fig. 5b, Ag0.5F75@BMO had excellent photodegradation efficiency among all the
co-doped samples and RhB was completely removed within 2 h, illustrating the remarkable effect
of co-doping of F and Ag. Silver as a noble metal can efficiently improve photocatalytic activity
by localized surface plasmonic resonance (LSPR) effect, which facilitates electron transfer and
electron-hole separation. Theoretically, LSPR is an optical phenomenon attributed to the
increasing of the local electromagnetic field in nanoparticles smaller than the wavelength of
irradiation. The interaction between light wave and surface electrons is because of the energy
levels of d–d transitions of Ag in the visible range of the spectrum [30]. Therefore, nanoparticles
undergoing LSPR can have high molar extinction coefficients for light absorption [31].
Furthermore, the surface Ag species can act as electron acceptor site and facilitate electron transfer
2
4

and electron-hole separation, which resulted in decreasing of recombination rate and enhancement
of quantum efficiency. However, when the Ag content was more than 2 mol.% (Fig. 5c), it was
observed that the photocatalytic performance was reduced. As the mentioned, Ag can act as a
trapping center for the photogenerated electrons, leading to an improvement in the charge
separation. However, in the high doping quantities of Ag, it may serve as a recombination center
for photogenerated electron and hole pair and therefore, decreases the photocatalytic activity [70].
The results from Fig. 4 demonstrated that with optimum molar ratio of dopants in Ag0.5F75@BMO,
approximately 2 times enhancement in removal (%) of RhB was obtained in compare to pure
BMO. This is attributed to the synergic effects of Ag in decreasing the recombination rate of
electrons and holes, and also successfully generating oxygen vacancies and increasing hydroxyl
groups by F atoms on the surface, which decrease the band gap energy and strongly enhance their
photocatalytic performance under visible light irradiation.
As mentioned above, Ag0.5
F
75@BMO/rGO showed the best photocatlytic activity in the all
75@BMO/rGO is related to the
prepared catalysts. The better photocatalytic performance of Ag0.5
F
several parameters. Based on the N
2
adsorption–desorption analysis reults, deposition of BMO on
2
the rGO sheets enhances its specific surface to the high specific surface area of GO (2600 m /g),
resulting in the higher adsorption capacity for RhB. Concurrently, GO with a two-dimensional π-
conjugated structure could serve as an electron mediator for photogenerated electrons in the
conductive band of BMO and thus it could significantly improve the separation of photoexcited
electron-hole pair [71, 72]. Besides, after the reduction of GO during the solvothermal process, the
mobility of photogenerated electrons on the graphene-like layers could be improved, because of
reducing some of the oxygen functional groups. Therefore, the recombination rate of electron-hole
2
5

pair is remarkably reduced by insertion of GO into nanocomposite. However, the photocatalytic
activity of the composite deteriorated, when the GO content was increased above 0.5wt%.
Although, increase in the GO content might leads to adsorption of greater amount of RhB,
however, it could disadvantageously occupy the active sites of BMO, resulting in inhibition of
light photons arriving to BMO surface and decrease the photocatalytic performance [73, 74]
3
.6.1. The effect of pH on photocatalytic activity
The pH of an aquatic solution plays an important role in the photocatalytic degradation, by
controlling the surface charge and the rate of aggregations [75]. The surface charge of
photocatalyst is highly dependent on the solution pH and the adsorption of pollutants is directly
related to the surface charge of catalyst [76]. In order to study the influence of pH, the
photocatalytic degradation of RhB was investigated at acidic (pH = 4), natural (pH = 5.4), neutral
(pH = 7) and alkaline (pH = 9) solution pHs. The reaction condition was as follow: RhB
concentration= 10 mg/L,catalyst concentration = 0.5 g/L and reaction time = 80 min. The results
indicated (Fig. 6a) that the RhB photodegradation efficiency was maximum at neutral pH and it
dropped to 60% and 49% in the solution pH of 4 and 9.
For more clarification, the surface charge of photocatalyst was measured by the point of zero
charge (PZC) for Ag0.5F75@BMO. Fig. 6b shows zeta potentials of −3.37, −26.9, −31.8, and −32.0
mV for pH=4, 5.4, 7 and 9, respectively. As observed, the surface of catalyst was negative in all
pH and by increasing the pH, the surface of catalyst became more negatively up to neutral pH.
With considering the negative surface of Ag0.5F75@BMO nanoparticles, by increasing the pH value
2
6

more hydroxyl anions in solution will expose to radiation, leading to generate greater amount of
•
OH. Since, the surface of catalyst becomes more negative with increase of pH (up to pH=7),
through a strong electrostatic attraction RhB as a cationic dye could adsorb on the catalyst surface
efficiently, which is in good agreement with results reported by Azeez et al. [77]. In addition, at
+
low solution pH, H as dominant specie is in competition with the RhB as a cationic contaminant
which leads to decrease in the adsorption of RhB on the surface of catalyst.
3
.6.2. Mineralization studies:
The dye mineralization degree was measured by analyzing the total organic carbon content
(TOC) in the solutions catalyzed by the most active catalyst, Ag0.5
F75@BMO/GO nanocomposite,
at different irradiation times. The mineralization ratio of RhB was determined as follows:
TOC −TOC
ꢌ
ꢍ
%
ꢈinerꢉlizꢉꢊiꢋn =
× 100%
(5)
TOCꢌ
Where TOC
0
and TOC are the concentration of initial and residual total organic carbon
t
respectively. As seen in Fig. 6c, the mineralization of RhB reached to 64.58% at the reaction time
of 120 min while the removal efficiency was 100% at this time. The different degradation and
mineralization efficiencies infers that some of intermediates of RhB oxidation process might
remain in the solution, which can be removed by increasing the reaction time. It is believed that
the mineralization of RhB is due to the redox reaction of the highly reactive oxygen species (ROS)
generated on the surface of Ag0.5F75@BMO/rGO.
3
.6.3. Discussion on the enhanced degradation performance:
2
7

It was proposed that the reactive oxygen species (ROS) generated by Ag,F@BMO play a
•
crucial role in the photodegradation. Reactive oxygen species such as hydroxyl ( OH) and
•
superoxide ( O
2
−) are of significance for environmental chemistry. These species could be
basically produced in photocatalytic process, and are capable to contribute in photodegradation.
•
•
2−
The photogenerated electron can reduce O
2
to O and hole can oxidize hydroxyl groups to OH
+
−
−
on H
2
O and OH [78-80]. The rapid recombination of photoinduced e - h in bulk phase and on
the surface of photocatalyst, leads to the low charge transfer, and thus reducing the amount of
ROS in the system [81-83]. To reveal the reaction mechanism for photocatalytic performance of
the as-synthesized samples, the trapping experiments were carried out by introduction of various
sacrificial agents to determine the active species generated during the photodegradation process.
•
2−
Three different quenchers, p-benzoquinone (BZQ, as O radicals scavenger, 1 mM),
•
isopropanol (IPA, as the scavenger of OH radicals, 1 mM) and ethylenediaminetetraacetic acid
+
disodium salt (EDTA-2Na, as the scavenger of h , 1 mM) were added [84]. Fig. 6d shows that
the photocatalytic activity of Ag0.5F75@BMO was not affected by IPA, while with introducing of
BZQ the photodegradation efficiency of RhB was slightly decreased. In contrast, by the addition
of trace EDTA-2Na significantly suppressed the photocatalytic performance, suggesting that
•
2−
photogenerated holes are the main active species in RhB photodegradation process. Besides, O
radicals can play a relatively important role and participate in the oxidation reactions.
Herein, based on the above results, schematic interactions between the nanocomposite and RhB
molecule, and also a possible mechanism for the photocatalytic activity of z-scheme
Ag0.5
F
75@BMO/rGO is proposed and illustrated in Fig. 7. And Fig. 8. When Bi
2
MoO harvests a
6
visible-light photon with energy equal or higher than the band-gap, a valance band electron could
2
8

be excited and migrate to the conduction band, resulting in the generation a hole in the valance
−
band. The CB-electrons can transfer to dissolved O
2
molecules or OH and produce reactive
−
oxygen species (ROSs). Also the VB-holes could react with OH and create hydroxyl radicals or
•
directly oxidize the contaminants. Based on the above discussion, both O2− and photogenerated
holes in VB can participate in photodegradation process of RhB. Nevertheless, low quantum yield
and the rapid recombination of electron-hole pair in BMO restricts its utilization in
photodegradation of RhB. On the basis of the photocatalytic performance of Ag0.5F75@BMO/rGO,
we infer that the recombination rate is much slower, due to the synergistic effect of Ag and F
dopants as well as rGO, which was proved trough PL results (Fig. 3d). It is believed that a slight
red-shift in the absorption edge of Ag0.5F75@BMO is related to the fluorine doping, which
originated from new energy sub-levels consisted of the F 2p and O 2p orbitals on the top of the
−
valence band [25]. Oxygen vacancies which induced by F on the surface of photocatalyst, play a
crucial role in the photodegradation process and decrease the electron-hole recombination [47].
+
3+
Also, the Ag on the surface of BMO could substitute with Bi ions to induce many defects which
would capture free electrons in BMO and also improve the electron-hole separation efficiency due
to the surface plasmon resonance effect. In the other hand, GO as a powerful electron acceptor
could capture the photogenerated electrons and improve charge separation and electron transfer
efficiently. Therefore, the photoexcited electrons on CB can be captured by oxygen vacancies and
+
also, transfer to Ag ions and to the π-conjugated system of rGO. This process is faster than the
recombination of electron–hole pair between the VB and CB of BMO. Therefore, many of CB-
electrons could escape from the recombination process and are captured by absorbed oxygen
2
9