Ag2O/Ag3VO4/Ag4V2O7 heterogeneous photocatalyst prepared by a facile hydrothermal synthesis with enhanced photocatalytic performance under visible light irradiation

Article abstract of DOI:10.1016/j.materresbull.2015.08.028

A novel Ag₂O/Ag₃VO₄/Ag₄V₂O₇ photocatalyst was synthesized by adjusting the molar ratio of silver-vanadium (Ag-V) in a facile hydrothermal method to obtain multi-phase Ag₂O/Ag₃VO₄/Ag₄V₂O₇ photocatalyst. The photocatalytic activity of the prepared samples was quantified by the degradation of Rhodamine B (RhB) model organic pollutant under visible light irradiation. Compared to pure Ag₃VO₄, Ag₄V₂O₇ and P25 TiO₂, respectively, the as-synthesized multi-phase Ag₂O/Ag₃VO₄/Ag₄V₂O₇ powders gave rise to a significantly higher photocatalytic activity, achieving up to 99% degradation of RhB in 2 h under visible light. This enhanced photocatalytic performance was attributed to the effect of the multi-phase Ag₂O/Ag₃VO₄/Ag₄V₂O₇ photocatalyst and the surface plasmon resonance (SPR) of the incorporated metallic silver (Ag⁰) nanoparticles (NPs) generated during the photocatalysis, as evidenced by post-use characterization, resulting in improved visible light absorption and electron-hole (e^--h⁺) separation. A mechanism was proposed for the photocatalytic degradation of RhB on the surface of Ag₂O/Ag₃VO₄/Ag₄V₂O₇.

Full text of DOI:10.1016/j.materresbull.2015.08.028

Accepted Manuscript
Title: Ag O/Ag VO /Ag V O heterogeneous photocatalyst
2
3
4
4
2
7
prepared by a facile hydrothermal synthesis with enhanced
photocatalytic performance under visible light irradiation
Author: Rong Ran Joanne Gamage McEvoy Zisheng Zhang
PII:
S0025-5408(15)30094-5
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.materresbull.2015.08.028
MRB 8381
To appear in:
MRB
Received date:
Revised date:
Accepted date:
13-5-2015
11-8-2015
22-8-2015
Please cite this article as: Rong Ran, Joanne Gamage McEvoy,
Zisheng Zhang, Ag2O/Ag3VO4/Ag4V2O7 heterogeneous photocatalyst
prepared by a facile hydrothermal synthesis with enhanced photocatalytic
performance under visible light irradiation, Materials Research Bulletin
http://dx.doi.org/10.1016/j.materresbull.2015.08.028
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.

Ag O/Ag VO /Ag V O heterogeneous photocatalyst prepared by a facile
2
3
4
4
2
7
hydrothermal synthesis with enhanced photocatalytic performance under
visible light irradiation
a, b
a, b
a, b,
Rong Ran , Joanne Gamage McEvoy , Zisheng Zhang
*
a
Department of Chemical and Biological Engineering, University of Ottawa,
Ottawa, Ontario, K1N 6N5, Canada
b
Centre for Catalysis Research and Innovation, University of Ottawa, Ottawa,
Ontario, K1N 6N5, Canada
*Corresponding author. Tel.: +1 613 562 5800 x 6110; fax: +1 613 562 5172;
Email: zzhang@uottawa.ca
Graphical abstract
Highlights



The photocatalyst was hydrothermally prepared by adjusting the ratio of Ag
to V.
Multi-phase
Ag O/Ag VO /Ag V O
obtained
exhibited
multi-
2
3
4
4
2
7
morphological features.
The photocatalyst exhibited strong visible light driven photoactivity towards
RhB.
1

Abstract
A novel Ag O/Ag VO /Ag V O photocatalyst was synthesized by adjusting the
2
3
4
4
2
7
molar ratio of silver to vanadium (Ag to V) in a facile hydrothermal method to
obtain multi-phase Ag O/Ag VO /Ag V O photocatalyst. The photocatalytic
2
3
4
4
2
7
activity of the prepared samples was quantified by the degradation of Rhodamine
B (RhB) model organic pollutant under visible light irradiation. Compared to pure
Ag VO , Ag V O and P25 TiO , respectively, the as-synthesized multi-phase
3
4
4
2
7
2
Ag O/Ag VO /Ag V O powders gave rise to a significantly higher photocatalytic
2
3
4
4
2
7
activity, achieving up to 99% degradation of RhB in 2 h under visible light. This
enhanced photocatalytic performance was attributed to the effect of the multi-
phase Ag O/Ag VO /Ag V O photocatalyst and the surface plasmon resonance
2
3
4
4
2
7
0
(
SPR) of the incorporated metallic silver (Ag ) nanoparticles (NPs) generated
during the photocatalysis, as evidenced by post-use characterization, resulting in
-
+
improved visible light absorption and electron-hole (e -h ) separation. A
mechanism was proposed for the photocatalytic degradation of RhB on the surface
of Ag O/Ag VO /Ag V O .
2
3
4
4
2
7
Keywords:
Semiconductors; Solvothermal; Chemical synthesis; Catalytic
properties; Optical properties
2

1. Introduction
Strategies for the development of ternary metal oxides have been investigated in
recent years as an alternate route for the fabrication of highly efficient
photocatalysts, which can be used for photocatalytic degradation and water
splitting applications [1-4]. Numerous ternary metal oxide photocatalysts have
been reported in literature, such as Ag VO [5], Ag V O [6], -AgVO [7],
3
4
4
2
7
3
AgNbO [8], AgMO (M = Al, Ga, In) [9, 10], Ag CO [11] , and Ag PO [12] and
3
2
2
3
3
4
have been found to exhibit enhanced photocatalytic performance under visible
light irradiation (VLI), compared to the traditional TiO photocatalyst [13].
2
Despite these research efforts, single phase ternary metal oxide photocatalysts
tend to suffer from poor visible light absorption due to the number of grain
boundaries present [1, 14]. These defects usually serve as recombination centers,
-
+
and impede the effective transfer of the photogenerated electrons (e ) and holes (h )
to adsorbed organic pollutants by trapping the separated charge species [1, 14, 15].
In addition, large specific surface area is a prerequisite for the effective
photocatalytic degradation of organic pollutants, as it contributes to increasing the
transfer rate of photogenerated charges, improving carrier separation, and
increasing surface reactivity during adsorption-desorption equilibration [16-20].
3

An approach to improving photocatalytic performance under VLI has been in
the development and preparation of heterogeneous photocatalysts based upon
silver species. For example, Xue et al [21] reported that novel Ag O/N-doped
2
helical carbon nanotubes (Ag O/N–HCNTs) were successfully synthesized via a
2
simple co-precipitation of N–HCNTs and Ag O, respectively, which gave rise to
2
the high photocatalytic performance towards degradation of methylene blue (MB)
and showed good stability in multiple degradation cycles under visible light
irradiation. Similarly, Wang et al. [22] investigated Ag O/Ag PO heterostructured
2
3
4
photocatalyst prepared by ion-exchange synthesis of Ag PO and subsequent
3
4
precipitation. The heterostructures prepared exhibited a much higher degradation
of methyl orange (MO) and phenol than the pure Ag PO , and possessed good
3
4
stability in the aqueous photosystem under VLI. Highly efficient p-n junction
Ag O-decorated flower-like ZnO photocatalyst exhibited high photoactivity
2
towards MO under ultraviolet (UV) irradiation by transferring photogenerated
carriers and suppressing the e-h recombination rate [23]. Zhou et al. [24] reported
that the photocatalytic activity of Ag O/TiO nanobelts’ heterostructure was
2
2
attributed to the Ag O NPs scattered uniformly on the surface of TiO , giving rise
2
2
to the higher photocatalytic decomposition of MO under both UV and VLI.
However, the conventional syntheses employed to prepare these heterostructure-
containing visible-light-induced photocatalysts are multi-step, time-consuming
4

procedures. Therefore, in this work, we propose a facile approach to obtain multi-
phase Ag O/Ag VO /Ag V O photocatalyst with high photocatalytic performance
2
3
4
4
2
7
under VLI, fabricated via a hydrothermal process and changing the ratio of
elements employed in the synthesis procedure. The prepared photocatalysts were
then studied for the degradation of RhB organic dye. The novel heterogeneous
Ag O/Ag VO /Ag V O photocatalyst could improve the utilization of light to
2
3
4
4
2
7
enhance the degradation of organic pollutants via the synergistic effects among
these three photosensitive phases. To the best of our knowledge, this approach to
the preparation of multi-phase Ag O/ Ag VO /Ag V O has not yet been reported.
2
3
4
4
2
7
2. Experimental
2.1. Synthesis
Multi-phase Ag O/Ag VO /Ag V O photocatalyst was synthesized via
2
3
4
4
2
7
hydrothermal method using potassium metavanadate (KVO ) and silver nitrate as
3
starting materials. KVO was prepared by calcination, where 0.38 g K CO (Fisher
3
2
3
Scientific, Certified ACS) and 0.5 g V O (ACROS Organics, 99.6%) were
2
5
dissolved in 35 mL deionized water (DW) under vigorous magnetic stirring. The
resulting red solution was poured into an evaporation dish, and dried overnight at
50 °C in an oven (Sheldon Manufacturing, Inc. Model No: 1350 GM). The dry
sample was then ground in an agate mortar and annealed in air at 457 °C (730 K)
5

for 5 h, yielding a pink KVO powder. To prepare the multi-phase
3
Ag O/Ag VO /Ag V O powders, the molar ratio of silver to vanadium (Ag to V)
2
3
4
4
2
7
used was 6:1.5. 15 mL of as-prepared KVO solution was heated in a water bath at
3
6
0 °C for 3 min, and 60 mL of 0.1 M AgNO (Fisher Scientific, Certified ACS)
3
was then added dropwise and allowed to react under continuous stirring for 30 min
to obtain an orange-yellow slurry. The pH of the resulting slurry was then adjusted
by adding 1M KOH, until a final pH of 7 was reached, as measured by a pH meter
(
Fisher Scientific, accumet basic, AB 15 pH meter). The slurry was then
transferred into 45 mL Teflon-lined stainless steel autoclave reactors, and the
reaction was allowed to proceed at 140 °C for 8 h. The resulting solid was
collected by filtration, washed five times with DW, and dried in an oven at 110 °C
for 6 h. The temperatures and times selected for the hydrothermal process were
based upon the study by Huang et al. [25].
For comparison, pure Ag VO was also prepared via the described method, but
3
4
applying a molar ratio of Ag to V of 3: 1. Single phase Ag V O (Ag to V of 3: 1)
4
2
7
was also similarly prepared, but adding Cetyltrimethyl Ammonium Bromide
CTAB, C H BrN) (ACROS Chemical, 99 + %) as a structural agent at a molar
(
1
9
49
ratio of 0.05:1 (CTAB: silver), in accordance with the report of Huang et al. [25,
6].
2
6

2.2. Characterization
Powder X-ray diffraction (XRD) measurements were carried out in Bragg–
Brentano geometry on a Rigaku Ultima IV apparatus with Cu Kα1 (λ = 0.15418
nm) radiation, operating at 40 kV and 44 mA, and with a scanning range of 2θ
from 15° to 70°. Transmission electron microscopy (TEM) images were obtained
using an FEI (formerly Phillips) Tecnai G2 F20 field emission transmission
electron microscope at an acceleration voltage of 200 keV. Scanning electron
microscopy (SEM) images were obtained with a Tescan VegaII XMU electron
microscope operated at 20 kV, with Au/Pd alloy coated samples (coated with an
Anatech Hummer VII sputter coater). SEM-EDS was performed using an energy
dispersive X-ray detector. X-ray photoelectron spectroscopy (XPS) was conducted
on a Kratos Analytical Axis Ultra DLD instrument with mono-chromated Al X-
rays at 140 W. The powder UV-Vis diffuse reflectance spectra (DRS) were
recorded on a Thermo Evolution 300 UV/Vis spectrophotometer equipped with a
Praying Mantis diffuse reflectance accessory, and the spectra were collected at a
-1
scan rate of 240 nm min . UV/Vis spectra of RhB samples were obtained via a
Biochrom Ultrospec 60 UV/Vis spectrophotometer.
7

2.3. Photocatalytic activity
Photocatalytic performance was quantified by the decomposition of RhB
(
Sigma-Aldrich) as a model organic pollutant under VLI. A slurry reactor was
placed in a reflective housing to prevent outside light from entering the reactor,
and inside light from exiting the system. A 300 W ELH tungsten halide bulb
(
Ushio) was used as a light source with a 410 nm cut-off filter (Kenko Zeta, λ >
10 nm, transmittance > 90%) to provide visible light irradiation. The light source
4
was placed at a distance of 15 cm from the top of the slurry. The corresponding
irradiation was measured using a quantum meter (Biospherical QSL-2100; 400 nm
−
3
-2 -1
<
λ < 700 nm), and was found to be approximately 10.9 × 10 Einstein m s .
Cooling was provided by an external cooling jacket, and the temperature of the
reaction was controlled to 20 °C ± 2. Before illumination, 0.15 g of photocatalyst
-1
dispersed into 150 mL of 15 mg L RhB solution was allowed to reach adsorption-
desorption equilibrium under continuous magnetic stirring at 280 rpm for 30 min
in the dark. Irradiation was then provided for 2 h. Samples were withdrawn at 15
3
min time intervals and separated by centrifugation at 10  10 rpm for 3 min in an
accuSpin Micro 17 (Fisher Scientific) microcentrifuge to remove the suspended
catalyst. The supernatant fluid was then analyzed by monitoring the peak
absorbance (λ = 554 nm for RhB) with a Genysys 10-UV spectrophotometer
(
Geneq Inc.). A standard curve for RhB was prepared and the concentration was
8

determined by the measured absorbance and the Beer-Lambert Law. In the
recyclability studies, the used, unwashed photocatalysts were separated by
centrifugation after each run, the degraded RhB supernatant removed, and fresh 15
-1
mg L RhB solution added. The roles of various reactive species were studied by
employing reactive species scavengers (including tert-butyl alcohol (TBA) (Sigma-
Aldrich, ACS), benzoquinone (BQ) (reagent-grade ≥ 98%, Sigma Aldrich) and
ammonium oxalate (AO) (ACS, Sigma-Aldrich)), using the same procedures and
methodology described above.
3. Results and discussion
3.1. XRD analysis
The XRD pattern of the fresh, multi-phase photocatalyst obtained via
hydrothermal synthesis is shown in Fig. 1. The pattern observed was found to
exhibit features of the α-Ag VO (JCPDS Card No. 01-077-5276), Ag V O
3
4
4
2
7
(
JCPDS Card No.01-077-0097), and silver oxide phases (JCPDS Card No. 00-041-
1
3
104), respectively. The major reflections for the α-Ag VO phase occurred at
3
4
2.42° (3 1 2) and 30.96° (1 1 2), while the major peaks of the Ag V O structure
4
2
7
were observed at 32.93° (0 4 0) and 31.93° (2 2 4), and the reflection of the silver
oxide peaks was identified at 32.79° (1 1 1) and 38.07° (2 0 0) with low intensity,
suggesting that a low amount of Ag O was present in the fresh sample.
2
9

3.2. TEM analysis
As shown in Fig. 2a and 2b, the fresh multi-phase Ag O/Ag VO /Ag V O
2
3
4
4
2
7
photocatalyst was observed to range from 0.5 μm to 4 μm, and was decorated with
Ag O NPs clusters ranging from 50 nm to 180 nm; similar results were reported by
2
Yang et al .[27].
As seen in Fig. 2c, the high-resolution TEM (HRTEM) of the multi-phase
Ag O/Ag VO /Ag V O photocatalyst exhibited lattice spacings of 0.27 nm and
2
3
4
4
2
7
0
.24 nm, which corresponded to d-spacings of Ag O (1 1 1) and (2 0 0) planes of
2
silver oxide. Also, the selected area electron diffraction (SAED) patterns shown in
the inset of Fig. 2c further confirmed the crystal structure of Ag O present in the
2
multi-phase crystalline structure of Ag O/Ag VO /Ag V O photocatalyst,
2
3
4
4
2
7
demonstrating that the as-prepared silver species sample was polycrystalline
structure in nature. The rings were ascribed to diffraction from the (1 1 1) and (2 0
0
) reflections of Ag O phase (JCPDS Card No. 00-041-1104), based on the
2
calculated d-spacings of 2.73 Å and 2.36 Å, respectively.
.3. SEM and EDS analyses
The morphologies
Ag O/Ag VO /Ag V O photocatalyst were explored by SEM. The corresponding
3
present
in
the
synthesized
multi-phase
2
3
4
4
2
7
EDS spectra were also studied to qualitatively investigate the components present
1
0

in the Ag O/Ag VO /Ag V O photocatalyst, and the results are shown in Figs. 3a
2
3
4
4
2
7
and 3b, respectively.
As shown in Fig. 3a, the fresh Ag O/Ag VO /Ag V O photocatalyst exhibited
2
3
4
4
2
7
plate-like morphology with size ranges from 0.5 μm to 4 μm, which agreed with
the results reported by Pan et al. [28], and were sparsely decorated with Ag O NPs,
2
agreed with the report by Zhou et al. [24] and the results observed from XRD and
TEM. According to the results shown in Fig. 3b, EDS data showed that only the
silver, vanadium and oxygen elements constituted the as-synthesized sample. The
results indicated that no other elements occurred as impurities in the prepared
Ag O/Ag VO /Ag V O . It should be noted that the Au peak occurred as a result of
2
3
4
4
2
7
the sputter coating, and carbon was attributed to the adsorbed carbon present in the
samples.
3.4. XPS analysis
To investigate the chemical and electronic states present in the prepared multi-
phase photocatalyst, XPS analyses of the fresh Ag O/Ag VO /Ag V O were
2
3
4
4
2
7
conducted, and the corresponding high resolution XPS spectral patterns are shown
in Figs. 4a, 4b and 4c, for Ag 3d, V 2p and O 1s states, respectively.
As shown in Fig. 4a, the spectra of the Ag species exhibited two symmetric
peaks located at binding energies of 367.5 eV and of 373.5 eV, with a separation of
1
1

6
eV, ascribed to the spin-orbit splitting characteristic of Ag 3d and Ag 3d ,
5
/2
3/2
+
respectively, and corresponding to Ag [29, 30]. The binding energy observed at
367.5 eV and 373.5 eV indicated that silver was probably in the state of oxides
[
31], agreeing with the results in shown in XRD and SEM. No peaks attributable to
0
Ag (located at 368.3 eV and 374.5 eV) [30, 32, 33] were observed, indicating that
0
no Ag NPs existed in the fresh sample.
The observed peak shown in Fig. 4b at a binding energy of 516.4 eV was
attributed to the V 2p3/2 orbital, and that at 523.8 eV to the V 2p1/2 orbital,
5
+
indicating that the vanadium was present in the form of V [7], which corresponds
well with the data expected from KVO [34].
3
The oxygen peak in Fig. 4c for the O 1s spectra was observed to be broad and
was deconvoluted into three separate peaks (dashed lines in Fig. 4c). The main
peak occuring at binding energy of 530.9 eV was thought to be dominant, while
the peaks located at binding energies of 532.1 eV and 534.5 eV were of lower
intensity, and indicated that the oxygen species was not purely ionic but also
contained metal-oxygen bonds with covalent characteristics [29]. The main peak at
530.9 eV was considered to be oxygen species in the presence of covalent mixing
2
-
states. As reported in literature, the oxide ions, such as O , would have given rise
to a single Gaussian peak around 532 eV [29]. However, the results obtained
indicated that the oxygen species present contained mainly covalent characteristics
1
2

in the presence of dominant metal-oxygen bonds rather than pure oxide ionic
states. This covalent characteristic with metal-oxygen bonds promoted a shift of
the broad oxygen peak observed, in which binding energies of 532.1 eV and 534.5
eV were attributed to the lattice oxygen in the multi-phase silver speices composite,
and the absorbed oxygen, respectively [35]. Therefore, the results suggested the
existence of covalent bonding including the Ag-O bonding [36] and between the
3
-
4-
vanadium and oxygen in the form of (VO ) and (V O ) species [29],
4
2
7
respectively, which were also confirmed with the results in SAED patterns and
EDS data present in the previous section.
3.5. Optical properties
In order to estimate the band gap energy of the fresh Ag O/Ag VO /Ag V O ,
2
3
4
4
2
7
UV-Vis diffuse reflectance spectra of the samples were investigated and the
corresponding patterns indicating intrinsic absorption in the visible light region are
shown in Fig. 5.
The band gap energies of the as-prepared fresh catalysts were estimated by the
DRS data and the following equation:
λ = 1240/Ebg.
(1)
1
3

Where λ is the maximum wavelength of absorption by photocatalysts (nm)
illustrated by the tangent lines in Fig. 5), and E is the estimated band gap energy
(
bg
of the photocatalysts (eV).
As shown in Fig. 5, the onset of visible light absorption by
Ag O/Ag VO /Ag V O occurred around 625 nm, resulting in a calculated band
2
3
4
4
2
7
gap energy of ~ 1.98 eV, which was smaller than that of both Ag VO (E = 2.1
3
4
bg
eV) and Ag V O (E = 2.5 eV), agreeing with the results from literature [6, 25,
4
2
7
bg
37]. This improved visible light absorption observed in the multi-phase
photocatalyst was thought to be due to the presence of Ag O (E = 1.29 eV) [38],
2
bg
where more visible light could be absorbed by these incorporated species
containing narrower band gap energy.
3
.6. Photocatalytic activity of Ag O/Ag VO /Ag V O
2
3
4
4
2
7
3
.6.1. Photodegradation of RhB
To quantify photocatalytic activity, RhB was used as a model organic pollutant
under VLI, and the results obtained are shown in Fig. 6.
As shown in Fig. 6, the photolysis control in the absence of photocatalyst was
found to contribute approximately 2.4 % self-degradation in 2 h under VLI. Due to
this relatively small contribution, the effect of photolysis was thought to be
negligible in the photocatalytic system. As seen in Fig. 6, the
1
4

Ag O/Ag VO /Ag V O sample exhibited the highest photocatalytic activity
2
3
4
4
2
7
compared to the other photocatalysts tested, exhibiting 99% RhB decomposition in
h, while the Ag VO , Ag V O and P25 TiO affected 78%, 83% and 59%
2
3
4
4
2
7
2
degradations in the same time period, respectively. The improved photocatalytic
performance of the Ag O/Ag VO /Ag V O compared to the other photocatalysts
2
3
4
4
2
7
was thought to be attributed to the effects of the multi-phase
Ag O/Ag VO /Ag V O photocatalyst.
2
3
4
4
2
7
3.6.2. RhB degradation mechanism
To further investigate the mechanism of RhB photocatalytic degradation in the
presence of Ag O/Ag VO /Ag V O , the changes to the UV-visible spectra of the
2
3
4
4
2
7
dye during the photocatalytic reaction were investigated, and the results are shown
in Fig. 7.
As seen in Fig. 7, the intensity of the characteristic absorption peak at 554 nm
reduced significantly along the course of reaction, with progressive shifts of the
absorption bands towards the blue region from 554 nm to 520 nm, in agreement
with the results reported in literature on the photocatalytic degradation of RhB over
TiO and Bi WO under visible light, respectively [39, 40]. Competitive reactions
2
2
6
between the cleavage of the RhB chromophore ring structure and N-de-ethylation
were thought to occur during the entire photocatalytic degradation process for
1
5

RhB. From Fig. 7, the decline at the maximum absorption wavelength of 554 nm
was indicative of the cleavage of the whole conjugated chromophore structure of
RhB, while the minor shifts from 554 nm to 520 nm were indicative of the
stepwise N-de-ethylation of RhB. Of these processes, the cleavage of the whole
conjugated chromophore structure for RhB was thought to be predominant during
the observed degradation process.
3
.7. Stability of Ag O/Ag VO /Ag V O
2
3
4
4
2
7
3.7.1. Analyses of crystal structural changes
The multi-phase material synthesized consisted of dual-phase Ag VO /Ag V O
3
4
4
2
7
decorated with Ag O NPs at the surface, and the instability of the photocatalyst
2
under the photocatalytic conditions was thought to induce the loss crystallinity of
0
+
dual-phase Ag VO /Ag V O and the generation of Ag NPs. Some Ag ions from
3
4
4
2
7
-
Ag V O were thought to first react with the adsorbed hydroxyl ions (OH ) to give
4
2
7
rise to Ag O, as shown in reaction (2):
2
+
-
2
Ag + 2 OH → Ag O + H O [41, 42],
(2)
2
2
Ag O were unstable with prolonged photocatalytic reaction under VLI, and
2
0
some of Ag O could be reduced to Ag NPs by receiving photogenerated electrons
2
[
38]:
1
6

+
-
0
Ag + e → Ag
(3)
0
With the increase of quantity of Ag NPs, Ag VO was thought to be converted
3
4
to pyrovanadate (Ag V O ) and some Ag O, under prolonged photocatalytic
4
2
7
2
reaction. This is shown by the following reaction:
VLI
2
Ag VO → Ag V O + Ag O
(4)
3
4
4
2
7
2
Accordingly, the loss of crystallinity of both Ag VO and Ag V O promoted the
3
4
4
2
7
0
generation of Ag O and the increase of Ag NPs, as reported by Belver et al. [17].
2
This process decreased the crystallinity of Ag VO during photocatalytic reaction
3
4
under VLI. Although the instability of Ag VO facilitated the transformation of
3
4
+
0
Ag into Ag , resulting in the photocorrosion of Ag VO in the absence of the
3
4
0
electron acceptors, Ag NPs with the SPR served as active photocatalytic sites for
accepting photogenerated electrons and transferring them to the adsorbed oxygen,
where more hydroxyl radicals were generated due to photogenerated holes that
reacted with absorbed H O, which acted to enhanced photocatalytic reaction under
2
VLI.
Structural changes to the multi-phase Ag O/Ag VO /Ag V O photocatalyst
2
3
4
4
2
7
occurred during the course of photocatalytic reaction, and were studied via a
number of characterization techniques and investigations. The post-use
characterization results are discussed in subsequent sections.
1
7

3.7.2. Post-use XRD analysis
In order to further investigate changes to the crystal structure of the multi-phase
Ag O/Ag VO /Ag V O photocatalyst upon exposure in the photoreactive system,
2
3
4
4
2
7
XRD analyses were performed on the recycled catalysts. The post-use patterns
obtained for various samples are shown in Fig. 8.
Compared to the fresh sample shown in Fig. 1, the XRD patterns obtained from
the material recovered post-use and after recycling evidenced that structural
changes occurred to Ag O/Ag VO /Ag V O in the photosystem. In the XRD
2
3
4
4
2
7
-1
pattern shown in Fig. 8a, after reaction with 15 mg L RhB for 2 h, an obvious loss
of crystallinity of the Ag V O structure was observed, as evidenced in the decline
4
2
7
of major peaks at 32.93° and 31.93°. This was thought to be due to the reported
+
instability of Ag V O under visible light irradiation [43]. The Ag ions in
4
2
7
-
Ag V O reacted with the adsorbed OH ions to produce Ag O. In addition, some
4
2
7
2
+
0
lattice Ag of Ag O was thought to be reduced to the Ag by receiving
2
photogenerated electrons during the photocatalytic reaction [38]. This was
0
evidenced by the Ag phase discovered and the major reflections which occurred at
0
3
8.10° and 44.28° for Ag in the pattern (JCPDS Card No. 01-071-4613), as shown
in Fig. 8b. Also, the produced Ag O phase was not stable in the presence of ethanol,
2
and could be dissolved in ethanol after photocatalytic reaction according to the
literature reported by Singh et al. [44] and Mehta et al. [45] due to the solubilty of
1
8

this phase in the solvent. As such, the used photocatalyst was rinsed with ethanol
after recovery from the degraded RhB supernatant, and then recovered from the
ethanol extract by centrifugation. By comparison of the rinsed photocatalyst and
the unwashed photocatalyst after use, respectively, crystallinity changes and the
characteristics of these post-use photocatalysts were observed. The loss of
crystallinity of Ag V O during photocatalysis was appreciable, as observed in Fig.
4
2
7
8a.
The XRD pattern shown in Fig. 8b represented that of the unwashed
-1
photocatalyst after a single use in the photosystem with 15 mg L RhB solution
under VLI for 2 h. Comparison of Fig. 8b with Fig. 8a illustrates an obvious
0
increase in the intensity of the diffraction peak for Ag in the unwashed sample at
3
8.10°; and was thought to be due to more Ag V O being photocorroded, so that
4
2
7
+
0
more Ag ions from Ag V O were reduced to Ag eventually with prolonged
4
2
7
reaction under VLI. Furthermore, the intensity of the diffraction peaks of Ag V O
4
2
7
at 31.93° and 32.93° were much higher than the characteristic peaks for Ag V O
4
2
7
shown in Fig. 8a. This was ascribed to the instability of Ag VO with prolonged
3
4
0
photocatalytic reaction in the presence of generated Ag NPs, resulting in the
conversion from Ag VO to pyrovanadate (Ag V O ), and a new generation of
3
4
4
2
7
Ag O [17].
2
1
9

The XRD patterns of Fig. 8b and Fig. 8c were used to further investigate these
structural changes in the multi-phase Ag O/Ag VO /Ag V O upon cyclic use. As
2
3
4
4
2
7
0
shown in Fig. 8c for the recycled photocatalyst used three times, more Ag NPs
were generated upon prolonged reaction, as shown in the intensity of diffraction
peaks at 38.10°. In addition, the loss of crystallinity of Ag VO was more severe
3
4
+
than the sample shown in Fig. 8b, suggesting that Ag ions in Ag VO were both
3
4
0
reduced to Ag NPs and gave rise to the increase of Ag V O phase, respectively.
4
2
7
3.7.3. Post-use SEM analysis
The morphologies of the synthesized multi-phase Ag O/Ag VO /Ag V O
2
3
4
4
2
7
photocatalysts after photocatalytic reactions were explored by SEM, in order to
investigate the microstructure and morphology changes that occurred, and the
images obtained are shown in Fig. 9.
No difference in morphologies was observed between Fig. 3a and Fig. 9a,
indicating that the photocatalytic reaction towards dye degradation did not strongly
impact the morphology of the catalyst. Comparison of Figs. 9a, 9b and 9c indicated
that multi-morphology of the synthesized photocatalysts emerged after use in the
0
photocatalytic reaction, where more Ag NPs were generated during the course of
VLI, as confirmed by the XRD results obtained. In addition, a new generation of
two-dimensional Ag O nanotubes (NTs) (width: 0.2 m-0.4 m; length: 4 m-40
2
2
0

m) was also observed in the unwashed samples (Figs. 9b and 9c), which was
mainly attributed to the loss of crystallinity of Ag VO and Ag V O under
3
4
4
2
7
prolonged photocatalytic conditions [17]. As seen in Figs. 9b and 9c, two-
dimensional NTs were observed and were thought to be Ag O, based upon the
2
results obtained by XRD. These Ag O NTs were not observed in the pure rinsed
2
sample shown in Fig. 9a, suggesting that the new generation of Ag O NTs resulted
2
from both Ag VO and Ag V O after photocatalytic performance.
3
4
4
2
7
0
As shown in Fig. 9c, more Ag NPs were obtained in the unwashed recycled
sample after three runs, while the amount of Ag O NTs decreased. This may
2
indicate that the loss crystallinity of both Ag VO and Ag V O was severe after
3
4
4
2
7
prolonged photocatalytic reaction, resulting in the generation of more Ag O NTs,
2
while Ag O was unstable in the reducing ambient RhB solution under VLI, and
2
0
0
gave rise to Ag NPs [44], increasing the amount of Ag NPs decorated on the
multi-phase photocatalyst, in good agreement with the results from XRD.
3.7.4. Post-use XPS analysis
To investigate the changes of chemical and electronic states present in the multi-
phase photocatalyst, XPS analyses of the post-use Ag O/Ag VO /Ag V O were
2
3
4
4
2
7
discussed, and the corresponding high resolution XPS spectral patterns are shown
in Figs. 10a, 10b and 10c, for Ag 3d, V 2p and O 1s states, respectively.
2
1

Comparing Ag 3d spectra in Fig. 4 and Fig. 10, the observed Ag 3d spectra of
post-use sample changed markedly. The devolution of the Ag 3d was performed,
and Ag 3d and Ag 3d spectra assigned to three sets of individual peaks
5
/2
3/2
centered at 367.0 eV & 373.0 eV, 367.9 eV & 373.6 eV, and 369.2 eV & 374.7 eV
dashed lines in Fig. 10a) were obtained. Of these six peaks, the peaks at 367.0 eV
(
&
373.0 eV and 367.9 eV & 373.6 eV were ascribed to the binding energies of Ag,
+
corresponding to Ag , and peaks at 369.2 eV & 374.7 eV were ascribed to the
0
binding energies of Ag, corresponding to Ag [30, 32, 46]. The chemical and
electronic states of silver were thought to be changed during photocatalytic
+
reactions under VLI, indicating that some Ag species were reduced into metallic
silver. V 2p and V 2p spectra were observed to be broad in Fig. 10b, and were
3
/2
1/2
deconvoluted into two pairs of individual peaks centered at 516.0 eV & 522.6 eV,
5
+
and 516.9 eV & 523.8 eV (dashed lines in Fig. 10b), corresponding to V , and
corresponding well with the data from KVO [34]. Photocatalytic reactions also
3
influenced the chemical and electronic states of oxygen shown in Fig.10. O 1s
spectra observed were changed noticeably, and deconvoluted into four separated
peaks (dashed lines in Fig. 10c), centered at 529.3 eV, 530.8 eV, 531.8 eV and
533.1 eV, respectively. Of these four peaks, peaks at 529.3eV, 530.8 eV and 531.8
eV were attributed to the lattice oxygen in the heterogeneous multi-phase
2
2

photocatalysts, while the observed peak of O1s at 533.1 eV was ascribed to the
absorbed oxygen according to the literature [35].
3.7.5. Post-use optical properties
As shown in Fig. 11, the UV-Vis diffuse reflectance spectra of post-use
Ag O/Ag VO /Ag V O samples also revealed intrinsic absorption in the visible
2
3
4
4
2
7
light region.
Compared to the fresh catalyst shown in Fig. 5, no apparent absorption peaks of
two spectra shown in Figs. 11a and 11b were observed at longer wavelengths,
0
indicating that Ag NPs were present in low quantities after a single use. However,
as shown in Fig. 11c, a noticeable broad peak at around 573 nm occurred in the
absorption spectrum for the recycled sample used three times, and was thought to
0
be due to the increased number of Ag NPs present in the catalyst [47]. The
0
broadening of the absorption peak was mainly due to the non-uniformity of the Ag
0
nanoparticle sizes and shapes [48]. These Ag NPs both acted to increase the range
-
+
of visible light absorption and to decrease the rate of recombination of e -h pairs
-
+
[
49] resulting in enhancing the efficiency of photogenerated e -h pairs capture by
the adsorbed organic pollutants.
2
3

3.8. Role of reactive species testing
To further investigate the role of the reactive species impacting the
photocatalytic degradation, quenching tests were applied to the degradation studies.
Photocatalytic degradation of organic pollutants is caused by reaction with reactive
+
-
-
species such as h , e , •OH and O • , which are generated on the surface of
2
photocatalysts after irradiation [50]. Quenching tests for photocatalytic degradation
of RhB by Ag O/Ag VO /Ag V O were performed by employing various radical
2
3
4
4
2
7
species scavengers, and the results are shown in Fig. 12.
As shown in Fig. 12, without scavengers, the photocatalytic degradation of RhB
was 99% within 2 h. 3 mL TBA was added to the photocatalytic reaction system as
the scavenger of •OH [6], resulting in a reduction in the degradation of RhB to
-
9
3% at the end of the 2 h reaction. 0.015 g BQ was used as a scavenger of O • [51,
2
5
2], limiting the capture of photogenerated electrons by molecular oxygen (O ) to
2
-
generate
O • , and causing the photodegradation of
RhB on
2
Ag O/Ag VO /Ag V O to reduce to 80%. These results indicated that the
2
3
4
4
2
7
photocatalytic reaction was less likely to be restricted by reducing the amount of
-
•
OH and O • . In contrast, by adding 0.15 g AO into the photocatalytic system [53,
2
54], the degradation was suppressed to 38% during the course of reaction, which
indicated that holes played a major role in the degradation mechanism for RhB.
2
4

Therefore, holes were thought to dominate the mechanism of photocatalytic
degradation of RhB under VLI for the prepared photocatalyst.
3
.9. Mechanism of photocatalytic activity
mechanism for RhB photodegradation in the presence of
Ag O/Ag VO /Ag V O under VLI was proposed, and the applicable reactions are
A
2
3
4
4
2
7
given as follows:
Ag O + VLI → Ag O (h + e )
+
-
(5a)
(5b)
(5c)
(6)
2
2
+
-
Ag VO + VLI → Ag VO (h + e )
3
4
3
4
+
-
Ag V O + VLI → Ag V O (h + e )
4
2
7
4
2
7
+
-
h + e → Heat
-
-
Ag V O (e ) + Ag O → Ag O
+ Ag V O
(7a)
(7b)
(7c)
(7d)
(8a)
(8b)
4
2
7
2
2
(injected e )
4
2
7
-
-
Ag VO (e ) + Ag O → Ag O
+ Ag VO4
3
4
2
2
(injected e )
3
+
+
Ag VO (h ) + Ag O → Ag O (h ) + Ag VO
3
4
2
2
3
4
+
+
Ag O (h ) + Ag V O → Ag V O (h ) + Ag O
2
4
2
7
4
2
7
2
+
-
Ag (in Ag O) + e → Ag
2
Ag + VLI → Ag (SPR)
2
5

-
+
Ag (SPR) + Ag V O → Ag V O
+ Ag (SPR)
(8c)
(9a)
4
2
7
4
2
7 (injected e )
-
-
e + O → O •
2
2
+
+
h + H O → • OH + H
(9b)
2
+
+
4
4
h + 2 H O → 4 H + O
(9c)
(9d)
2
2
+
-
+
h + 2 OH → O + 2 H
2
-
+
O • + H → HO •
(10a)
(10b)
(11a)
(11b)
2
2
-
+
-
HO • + 2e + H → •OH + OH
2
*
RhB ads + VLI → RhB
ads
*
-
+
RhB
- e → RhB
+ •OH → intermediates → CO + H O
ads
ads
2
2
[
40, 50]
In accordance with reactions (5)-(11) and Fig. 13, the photocatalyst prepared
consisted of multi-phase Ag O/Ag VO /Ag V O , namely Ag VO /Ag V O
2
3
4
4
2
7
3
4
4
2
7
decorated with Ag O NPs. As seen from Fig. 13, photocatalytic reaction was
2
initiated by the absorption of photons from the visible light range with energy
higher than the band gap energies of for Ag O, Ag VO and Ag V O , causing the
2
3
4
4
2
7
-
+
photogenerated e -h pairs to occur in the photoexcited semiconductors, in
-
+
accordance with reactions (5a-5c). In addition, reaction (6) showed that these e -h
2
6

pairs could either separate and move freely on the surface of photocatalysts, or
recombine as energy emission under the internal electric field caused by the
-
+
heterosturcture, which occurred when the migration of e -h pairs was slow from
the bulk to reaction sites at surface [1].
After the photoexcitation on the heterogenerous photocatalyst, the
photogenerated electrons were promoted to the conduction band (CB) and the
holes were left at the valence band (VB) of the photocatalyst, respectively. These
electrons were transferred from the CB of Ag V O to the CB of Ag O firstly
4
2
7
2
(
reaction 7a), meanwhile, some photogenerated electrons were shifted from the CB
of Ag VO to the CB of Ag O as well, due to the internal electric field (reaction
3
4
2
7
b), because CB potential of Ag V O (-2.03 eV vs. NHE) [6] and CB potential
4
2
7
of Ag VO (-0.1 eV vs. NHE) [37] were more negative than that of Ag O (-0.04
3
4
2
eV vs. NHE) [38]. In addition, the CB of Ag V O was more negative than the
4
2
7
-
standard redox potential of O /O • (-0.33 eV vs. NHE). Therefore, electrons were
2
2
-
able to reduce the adsorbed O to O • species (reaction 9a). However, an excess of
2
2
electrons accumulated in the CB of Ag O rather than being captured by the
2
adsorbed O molecules due to the more positive CB potential of Ag O than that of
2
2
-
O /O • (-0.33 eV vs. NHE). The extra photogenerated electrons transferred to CB
2
2
+
0
0
of Ag O could reduce Ag to Ag (reaction 8a), Ag NPs exhibited SPR in the
2
-
+
visible region, exciting e -h pairs on their surface. During the reaction processes,
2
7

0
Ag NPs exhibited higher absorption intensities in the visible region (λ > 410 nm)
0
by absorbing photons, in which, one Ag nanoparticle could absorb one visible
light photon resulting in the effectively polarized state by plasmon resonance of
silver, causing effective separation of the SPR-excited electrons and holes [55]
0
(
reaction 8b). And these electrons were transferred from the SPR-excited Ag NPs
to CB of Ag V O [56] (reaction 8c), while, holes caused more •OH by reacted
4
2
7
with absorbed H O (reaction 9b). Therefore, more electrons transferred happened
2
from the CB of Ag V O to that of Ag O (reaction 7a) due to the plasmon
4
2
7
2
resonance of silver.
Holes also migrated from the VB of Ag VO to VB of Ag O then to VB of
3
4
2
Ag V O because of the internal electric field present, as shown in reactions (7c
4
2
7
and 7d), accompanied with more positive VB potential, Ag VO preferentially
3
4
-
-
oxidized OH to give •OH (•OH/OH , 1.55 eV vs. NHE) (reaction 9b), H O to give
2
-
-
O (O /H O, 1.23 eV vs. NHE) (reaction 9c), and OH to give O (O /OH , 0.4 eV
2
2
2
2
2
vs. NHE) (reaction 9d) [57]. Ag O was able to give rise to reactions 9c and 9d as
2
well, due to the internal electric field present. In addition, holes transferred to VB
-
of Ag V O were trapped by OH ions originating from hydrolysis of H O at
4
2
7
2
+
-
surface to yield O and protons (H ) (reaction 9d). Furthermore, the generated O •
2
2
radicals at CB of Ag V O combined with the generated protons to yield HO •
4
2
7
2
radicals (reaction 10a). Finally, additional •OH radicals were generated via HO2•
2
8

radicals reacting with both photoexcited electrons and protons (reaction 10b) [58].
During the whole photocatalytic degradation, photogenerated holes played a
significant role of generating the strong oxidant •OH radicals, resulting in the
direct oxidization of RhB (reactions 11a and 11b) [56] on the surface of
Ag O/Ag VO /Ag V O . Therefore, this high photocatalytic performance was
2
3
4
4
2
7
attributed to the positive effects between the photocatalysis of multi-phase
0
Ag O/Ag VO /Ag V O and SPR of Ag NPs [56].
2
3
4
4
2
7
4. Conclusions
Multi-phase Ag O/Ag VO /Ag V O photocatalyst was prepared through a facile
2
3
4
4
2
7
hydrothermal method. Characterization tests indicated that dual-phase Ag VO and
3
4
Ag V O were present in the system and were surface-decorated by Ag O NPs.
4
2
7
2
Although each single phase (Ag O, Ag V O and Ag VO ), suffered from
2
4
2
7
3
4
structural changes and loss of crystallinity in the photocatalytic system, the
0
existence of the produced Ag NPs induced positive effects of SPR on
photocatalysis, which promoted an overall high photocatalytic performance
towards the degradation of RhB under VLI. A new generation of Ag O nanotube-
2
0
like structures and the increase of Ag formed from the morphological changes that
occurred under the photocatalytic reaction were observed. A mechanism was
proposed for the photocatalytic degradation of RhB.
2
9