A series of magnesium vanadates (MgV 2O6, Mg2V 2O7, and Mg3V 2O8) were synthesized to investigate the effect of cation concentration on photocatalytic performance. The samples were characterized by X-ray diffraction, field emission-scanning electron microscopy, UV-visible diffuse reflectance spectroscopy, and fluorescence spectroscopy. The photocatalytic O2 evolution experiments under visible light irradiation showed Mg2V 2O7 exhibits the best performance, while Mg3V 2O8 has the lowest activity. The density functional theory calculations indicated that the lowest unoccupied states of Mg3V 2O8 are the mostly localized by the cation layers. The fluorescence spectra and fluorescence decay curves gave evident performances of excited states of magnesium vanadates and pointed out MgV 2O6 has a very short excited electron lift-time. Mg2V 2O7 performs high photocatalytic activity because of its high electron mobility and long electron life-time.

Solar light derived photocatalysis is of great importance due to its potential in solving the current energy crisis and environmental problem by providing unlimited, renewable, and environmental friendly energy source through splitting water into H2 and O2.1–7 A typical photocatalytic water splitting process generally involves three sections: (i) the electrons are excited from the valence band (VB) to the conduction band (CB) and leave holes in VB; (ii) the photo-generated electrons and holes migrate to the sample surface; and (iii) the electrons and holes reacted with water to generate H2 and O2, respectively.7,8 Generally, to study the relationship between excited states and photocatalytic activity is helpful to understand the process of photogeneration and migration of excited carriers. Up to now, the relevant studies have been performed on the TiO2, CdS, PbS, and C3N4 through fluorescence measurements and fluorescence decay analyses.9–13 

Vanadates were typical semiconductors which can perform visible light response photocatalytic reactions.14–17 Among various vanadates, alkaline earth metal vanadates are popular semiconductors. In an early report, Mg3V 2O8 has been demonstrated to be active for water oxidation under visible light irradiation.18 In our previous study, SrV 2O6 and Ag2SrV 4O14 were developed to oxidize water into O2.19 Besides photocatalysis, alkaline earth metal vanadates are also well studied as luminescence materials.20 In both of photocatalysis and luminescence field, the excitation is important because the migration rate and life-time of electrons in excited states could greatly affect the photocatalytic and fluorescent efficiency. The researches of fluorescent performance could be helpful to develop new vanadates materials with high photocatalytic activities.

In this work, we chose a series of magnesium vanadates (MgV 2O6, Mg2V 2O7, and Mg3V 2O8) as target materials to study the cation effect on photocatalytic water oxidation. MgV 2O6, Mg2V 2O7, and Mg3V 2O8 were prepared by a solid state reaction and then photocatalytic O2 evolution experiments, density functional theory (DFT) calculation, and fluorescent measurement were carried out to understand the effect of excited states of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 on photocatalytic performance.

The MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples were synthesized via a solid state reaction. MgO and V 2O5 were used as the starting reagents. First, the stoichiometrically mixed powder reagents of MgO and V 2O5 were grinded in the presence of 95% ethanol solution. Next, the mixture was pressed into a pellet and then calcined at 600 °C for 12 h. Second, the obtained pellet was crushed, grinded in dry state, pressed into a pellet, and calcined at 800 °C for 24 h. Then, the second step was repeated once. Finally, the obtained sample was grinded into powder state.

The crystal structures were determined with an X-ray diffractometer (X’Pert Powder, PANalytical B.V., Netherlands) with Cu-Kα radiation. The diffuse reflection spectra were measured with an integrating sphere equipped UV-visible recording spectrophotometer (UV-2600, Shimadzu Co., Japan) using BaSO4 as reference and the optical absorptions were converted from the reflection spectra according to Kubelka-Munk equation. The specific surface areas were determined with a surface-area analyzer (BEL Sorp-II mini, BEL Japan Co., Japan) by the Brunauer–Emmett–Teller (BET) method. Scanning electron microscopy (SEM) images were recorded with a field emission scanning electron microscopy (JSM-6701F, JEOL Co., Japan) operated at 15 kV. The fluorescence spectra and fluorescence decay curves were measured by a spectrofluorometer (Fluorolog-3, Horiba Jobin Yvon).

The O2 evolution experiments were carried out in a gas-closed circulation system. The catalyst powder (0.3 g) was dispersed by using a magnetic stirrer in NaIO4 aqueous solution (270 ml of distilled water + 5 m mol of NaIO4) in Pyrex cell with a side window. The light source was a 300 W of Xe arc lamp with/without an L42 cut-off filter (λ > 400 nm with L-42 cut-off filter and λ > 300 nm without L-42 cut-off filter). The O2 evolution was measured with an on-line gas chromatograph (GC-8A, Shimadzu) with a thermal conductivity detector (TCD) according to the standard curve.

All calculations were performed with the Vienna ab initio Simulation Package (VASP) based on the DFT.21,22 The Projector-augmented wave (PAW) was used for the electron-ion interactions. To obtain the reliable band gap, the HSE06 hybrid-functional with a mixing parameter of a = 0.25 was employed to evaluate the exchange-correlation energy.23 The number of k points and the cut-off energy were increased until the calculated total energy converged within an error of 1 × 10−5 eV/atom. Therefore, the cut-off energy of 500 eV was set. The energy convergence tolerance was set to below 5 × 10−6 eV/atom. The lattice vectors and atomic coordinates were relaxed until the Hellmann-Feynman force on each atom is reduced to less than 0.01 eV/Å.

After synthesizing, the as-synthesized MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples were first determined by the crystallographic structures to confirm the formation of target materials (as shown in Figure 1(a)). The measured XRD patterns of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 have good agreement with the standard diffraction patterns (JCPDS-071-1651 for MgV 2O6; JCPDS-070-7769 for Mg2V 2O7; and JCPDS-073-0207 for Mg3V 2O8), showing that all the samples were well crystallized in the structure of target materials and there is no obvious impurity existed. Then, the SEM was further used to observe the morphologies of the as-prepared samples and the SEM images of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 are shown as Figures 1(b)1(d). As observed in the SEM images, all the three magnesium vanadate samples are constituted by particles with the un-uniformed morphology, and the particle sizes generally range from 500 to 2000 nm. The surface areas were determined to be 1.4, 1.9, and 2.2 m2 g−1 by BET method for MgV 2O6, Mg2V 2O7, and Mg3V 2O8, respectively.

FIG. 1.

(a) p-XRD patterns of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples compared with the standard XRD patterns. SEM images of the as-prepared (b) MgV 2O6, (c) Mg2V 2O7, and (d) Mg3V 2O8 (the scale bar equals to 5 μm).

FIG. 1.

(a) p-XRD patterns of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples compared with the standard XRD patterns. SEM images of the as-prepared (b) MgV 2O6, (c) Mg2V 2O7, and (d) Mg3V 2O8 (the scale bar equals to 5 μm).

Close modal

UV-visible absorption spectra of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 powder samples, which were obtained from the Kubelka-Munk conversion from reflection spectra, are shown in Figure 2(a). With different Mg:V ratios, the samples exhibit different absorption performances. MgV 2O6, which contains more vanadium than the others, could absorb more visible light than Mg2V 2O7 and Mg3V 2O8, while the sample that has the lowest vanadium content (Mg3V 2O8) has the worst visible light absorption ability. However, all the absorption edges of three magnesium vanadate samples appear in the visible light region, suggesting it is possible to perform photocatalytic reactions over MgV 2O6, Mg2V 2O7, and Mg3V 2O8 under visible light irradiation. The optical band gaps Eg of the magnesium vanadate samples were determined according to the Tauc equation,

( α h ν ) 1 / 2 = A ( h ν E g )

in which α, ν, A, and Eg are absorption coefficient, light frequency, proportionality constant, and optical band gap, respectively.24 From Figure 2(b), the optical band gaps for MgV 2O6, Mg2V 2O7, and Mg3V 2O8 were determined to be 2.2, 2.6, and 3.1 eV, respectively. With the increasing of Mg:V ratios, the optical band gaps of magnesium vanadate continuously increase.

FIG. 2.

(a) UV-visible absorption spectra of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples. (b) The corresponding Tauc plots of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples.

FIG. 2.

(a) UV-visible absorption spectra of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples. (b) The corresponding Tauc plots of the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples.

Close modal

The O2 evolutions from aqueous NaIO4 solution (5 m mol of NaIO4 in 270 ml of H2O) over MgV 2O6, Mg2V 2O7, and Mg3V 2O8 (0.3 g) under the irradiation of visible light (λ > 400 nm) are presented in Figure 3(a). Under the irradiation of visible light, all the samples exhibited photo-activities for O2 evolution in the presence of NaIO4 as a sacrificial reagent. Furthermore, O2 was generated almost linearly over all the samples during 8 h. As plotted in Figure 3(a), the O2 evolution rates are significantly different: Mg2V2O7 > MgV2O6 > Mg3V2O8, and the Mg2V 2O7 had much higher activity than the other samples with the O2 evolution rate up to 110 μmol ⋅ h−1. Because of its wider band gap, the O2 evolution rate over Mg3V 2O8 increased by 4 times when the light source was changed from visible light (λ > 400 nm) to UV-visible light (λ > 300 nm), while the activities of MgV 2O6 and Mg2V 2O7 under UV-visible light were only about 3 times of them under visible light (as shown in Figure 3(b)). The photocatalytic O2 evolution experiments indicate that the photocatalytic activity of Mg2V 2O7 is much higher than MgV 2O6 and Mg3V 2O8 in water oxidation. After O2 evolution experiments, there is no obvious changes in crystal structure which was observed from the p-XRD patterns.25 

FIG. 3.

Photocatalytic O2 evolutions from the aqueous NaIO4 solution over the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples under the irradiation of (a) visible light (λ > 400 nm) and (b) UV-visible light (λ > 300 nm).

FIG. 3.

Photocatalytic O2 evolutions from the aqueous NaIO4 solution over the as-prepared MgV 2O6, Mg2V 2O7, and Mg3V 2O8 samples under the irradiation of (a) visible light (λ > 400 nm) and (b) UV-visible light (λ > 300 nm).

Close modal

To understand the mechanism of different photocatalytic performances appeared over MgV 2O6, Mg2V 2O7, and Mg3V 2O8, DFT calculations were carried out. As shown in Figure 4(a), the theoretical band gaps of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 are somewhat overestimated, but the relative magnitude of the band gap MgV2O6 < Mg2V2O7 < Mg3V2O8 is consistent with our experimental results. The density of states (DOS) indicates all the magnesium vanadates have a similar character at the valence band maximum (VBM), which is mainly composited of O p states. The major difference in the electronic structures is found at the conduction band minimum (CBM), which is a hybrid band of V d and O p states; the width of the conduction band is narrower in Mg3V 2O8 than in MgV 2O6. This is related to the difference of the cation content in three vanadates, that is, the lower the cation content the vanadate is, the more V-O-V bonding network thrives, forming a wider band. Figures 4(b)4(d) show the difference in partial charge densities corresponding to the lowest unoccupied states of MgV 2O6, Mg2V 2O7, and Mg3V 2O8. As expected from the density of states, MgV 2O6 has the highest contribution from O p state as well as V d state in the CBM reflecting the formation of V-O-V network. By contrast, in Mg3V 2O8, the V-O-V network is disrupted by the magnesium layer, leading to the localized V d state at CBM. This indicates that the mobility of photo-generated electrons is very low and they could hardly migrate to the sample surface, agreeing with the previous study that it has an internal luminescence quantum efficiency up to about 6%.20 When photocatalytic experiment is carried out, the electrons are excited from O p orbitals to V d orbitals and the holes are left in O p orbitals under visible light irradiation followed by carrier migrations to the sample surface where photocatalytic reactions occur. From the above analysis over the electronic structures, it is likely that the electron mobility of MgV 2O6 and Mg2V 2O7 is higher than that of Mg3V 2O8. However, the experimental result could not support this simple conclusion that MgV 2O6 is not an efficient water oxidation photocatalyst as demonstrated below.

FIG. 4.

(a) The calculated density of states of MgV 2O6, Mg2V 2O7, and Mg3V 2O8. The partial charge densities corresponding to the lowest unoccupied states of (b) MgV 2O6, (c) Mg2V 2O7, and (d) Mg3V 2O8 (the isosurface is at 0.004 electrons/Å3).

FIG. 4.

(a) The calculated density of states of MgV 2O6, Mg2V 2O7, and Mg3V 2O8. The partial charge densities corresponding to the lowest unoccupied states of (b) MgV 2O6, (c) Mg2V 2O7, and (d) Mg3V 2O8 (the isosurface is at 0.004 electrons/Å3).

Close modal

For the further study of the properties of excited states to find out why Mg2V 2O7 exhibits much higher photocatalytic O2 evolution activity than the others, fluorescence spectra were recorded (as shown in Figure 5(a)). Under the irradiation of UV light (λex = 330 nm), MgV 2O6, Mg2V 2O7, and Mg3V 2O8 showed significantly different fluorescent characterizations. MgV 2O6 had quite weak light emission with the peak wavelength of about 430 nm; Mg2V 2O7 showed slight light emission with the peak wavelength of about 555 nm; and Mg3V 2O8 exhibited the strongest light emission with the peak wavelength of about 545 nm. The fluorescent intensities could also be easily observed when the sample was irradiated by UV light which are shown as the inset of Figure 5(a). As the fluorescence is induced by the recombination of photo-generated electron-hole pairs, the low activity of Mg3V 2O8 could be attributed to its high electron-hole recombination rate. Nevertheless, MgV 2O6, which presented lowest fluorescent intensity, has lower photocatalytic O2 evolution rate than Mg2V 2O7. To make a clear understanding of this phenomenon, fluorescence decay analyses were performed (as shown in Figure 5(b)). In the fluorescence decay curves, MgV 2O6 showed very short life-time of excited electrons, which means the photo-excited electrons have too short lift-time to migrate to the sample surface, especially when its particle size is very large. Mg2V 2O7 exhibited a slight longer life-time of excited electrons than Mg3V 2O8, suggesting the photo-excited electrons could more easily migrate to the surface.

FIG. 5.

(a) The fluorescence spectra of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 (λex = 330 nm) and the inset photos show the sample images under room light and UV light irradiation. (b) The fluorescence decay curves of MgV 2O6, Mg2V 2O7, and Mg3V 2O8.

FIG. 5.

(a) The fluorescence spectra of MgV 2O6, Mg2V 2O7, and Mg3V 2O8 (λex = 330 nm) and the inset photos show the sample images under room light and UV light irradiation. (b) The fluorescence decay curves of MgV 2O6, Mg2V 2O7, and Mg3V 2O8.

Close modal

In conclusion, MgV 2O6, Mg2V 2O7, and Mg3V 2O8 were synthesized to realize the visible light response water oxidation reaction. In the experiments, the photocatalytic performances of Mg2V 2O7 were much higher than those of MgV 2O6 and Mg3V 2O8. According to the DFT calculation, the low activity of Mg3V 2O8 is caused by its highly localized excited states, inducing a high electron-hole recombination rate and low electron mobility, which was confirmed by the fluorescence spectrum. The fluorescence decay analysis indicated that the life-time of photo-generated electrons in MgV 2O6 is too short to effectively migrate to the sample surface, resulting in a low photocatalytic O2 evolution activity. Therefore, this work is helpful for developing new vanadate for visible light response photocatalysis by studying its excited state property.

This work was partly supported by Precursory Research for Embryonic Science and Technology (PRESTO) and the Core Research for Evolutional Science and Technology (CREST) program, the Japan Science and Technology Agency (JST), and the National Basic Research Program of China (973 Program, 2014CB239301).

1.
M. R.
Hoffmann
,
S. T.
Martin
,
W. Y.
Choi
, and
D. W.
Bahnemann
,
Chem. Rev.
95
(
1
),
69
(
1995
).
2.
K.
Maeda
,
J. Photochem. Photobiol., C
12
(
4
),
237
(
2011
).
3.
A.
Kubacka
,
M.
Fernandez-Garcia
, and
G.
Colon
,
Chem. Rev.
112
(
3
),
1555
(
2012
).
4.
H.
Tong
,
S.
Ouyang
,
Y.
Bi
,
N.
Umezawa
,
M.
Oshikiri
, and
J.
Ye
,
Adv. Mater.
24
(
2
),
229
(
2012
).
5.
G.
Li
,
Z.
Yi
,
Y.
Bai
,
W.
Zhang
, and
H.
Zhang
,
Dalton Trans.
41
(
34
),
10194
(
2012
).
6.
J.
Shi
and
L.
Guo
,
Prog. Nat. Sci.: Mater. Int.
22
(
6
),
592
(
2012
).
7.
A.
Kudo
and
Y.
Miseki
,
Chem. Soc. Rev.
38
(
1
),
253
(
2009
).
8.
Z.
Yi
,
J.
Ye
,
N.
Kikugawa
,
T.
Kako
,
S.
Ouyang
,
H.
Stuart-Williams
,
H.
Yang
,
J.
Cao
,
W.
Luo
,
Z.
Li
,
Y.
Liu
, and
R. L.
Withers
,
Nat. Mater.
9
(
7
),
559
(
2010
).
9.
K.
Ishibashi
,
A.
Fujishima
,
T.
Watanabe
, and
K.
Hashimoto
,
Electrochem. Commun.
2
(
3
),
207
(
2000
).
10.
T.
Tachikawa
,
M.
Fujitsuka
, and
T.
Majima
,
J. Phys. Chem. C
111
(
14
),
5259
(
2007
).
11.
K.
Wu
,
Z.
Chen
,
H.
Lv
,
H.
Zhu
,
C. L.
Hill
, and
T.
Lian
,
J. Am. Chem. Soc.
136
(
21
),
7708
(
2014
).
12.
D.
Wang
,
J. K.
Baral
,
H.
Zhao
,
Vo-V.
Truong
,
B. A.
Gonfa
,
My A. E.
Khakani
,
R.
Izquierdo
, and
D.
Ma
,
Adv. Funct. Mater.
21
(
21
),
4010
(
2011
).
13.
P.
Niu
,
G.
Liu
, and
H. M.
Cheng
,
J. Phys. Chem. C
116
(
20
),
11013
(
2012
).
14.
A.
Kudo
,
K.
Omori
, and
H.
Kato
,
J. Am. Chem. Soc.
121
(
49
),
11459
(
1999
).
15.
J.
Ye
,
Z.
Zou
,
M.
Oshikiri
,
A.
Matsushita
,
M.
Shimoda
,
M.
Imai
, and
T.
Shishido
,
Chem. Phys. Lett.
356
(
3–4
),
221
(
2002
).
16.
R.
Konta
,
H.
Kato
,
H.
Kobayashi
, and
A.
Kudo
,
Phys. Chem. Chem. Phys.
5
(
14
),
3061
(
2003
).
17.
D.
Wang
,
Z.
Zou
, and
J.
Ye
,
Catal. Today
93–95
,
891
(
2004
).
18.
D.
Wang
,
J.
Tang
,
Z.
Zou
, and
J.
Ye
,
Chem. Mater.
17
(
20
),
5177
(
2005
).
19.
P.
Li
,
N.
Umezawa
,
H.
Abe
, and
J.
Ye
,
J. Mater. Chem. A
3
,
10720
(
2015
).
20.
T.
Nakajima
,
M.
Isobe
,
T.
Tsuchiya
,
Y.
Ueda
, and
T.
Manabe
,
J. Phys. Chem. C
114
(
11
),
5160
(
2010
).
21.
G.
Kresse
and
J.
Hafner
,
Phys. Rev. B
47
(
1
),
558
(
1993
).
22.
G.
Kresse
and
J.
Furthmüller
,
Phys. Rev. B
54
(
16
),
11169
(
1996
).
23.
J.
Heyd
,
G. E.
Scuseria
, and
M.
Ernzerhof
,
J. Chem. Phys.
118
(
18
),
8207
(
2003
).
24.
M. A.
Butler
,
J. Appl. Phys.
48
(
5
),
1914
(
1977
).
25.
See supplementary material at http://dx.doi.org/10.1063/1.4922833 for additional data.

Supplementary Material