RSC Advances

PAPER

Band gap engine

aNational Laboratory of Mineral Materi

Technology, China University of Geoscienc

hhw@cugb.edu.cn; zyh@cugb.edu.cn; Tel: +bDepartment of Physics and Materials Scienc

Avenue, Kowloon, Hong Kong, China

† Electronic supplementary informationspectra of Bi2WO6, CB and VB of Mg1�xCanalysis of Mg0.7Cu0.3WO4/Bi2WO6

degradation curves of Mg1�xCuxWO4 sam

Cite this: RSC Adv., 2014, 4, 41219

Received 13th June 2014Accepted 14th August 2014

DOI: 10.1039/c4ra05708b

www.rsc.org/advances

This journal is © The Royal Society of C

ering design for construction ofenergy-levels well-matched semiconductorheterojunction with enhanced visible-light-drivenphotocatalytic activity†

Hongwei Huang,*a Shuobo Wang,a Yihe Zhang*a and Paul K. Chub

Energy-levels well-matched Mg1�xCuxWO4 (0.1 < x < 0.5)/Bi2WO6 heterojunctions with Type II staggered

conduction bands and valence bands have been successfully constructed by band gap engineering

based on solid-solution design and synthesized by a facile one-step hydrothermal method. X-ray

diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM),

transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM) and

UV-vis diffuse reflectance spectra (DRS) were utilized to characterize the crystal structures,

morphologies and optical properties of the as-prepared products. The as-designed Mg0.7Cu0.3WO4/

Bi2WO6 heterojunctions consisting of nanocube and nanoplate structures exhibit much higher visible-

light-driven (VLD) photocatalytic activity than the two individual components for the degradation of RhB

and photocurrent generation. The photoluminescence (PL) spectra, photoelectrochemical measurement,

active-species trapping and quantification experiments all indicated that the fabrication of energy-levels

well-matched overlapping band structures can greatly facilitate the separation and easy transfer of

photogenerated electrons and holes, thus resulting in remarkably enhanced photocatalytic activity. This

work provides a novel strategy for semiconductor heterojunction construction and energy band

structure regulation.

1. Introduction

The semiconductor photocatalysis technique is receiving greatattention due to its potential applications in the treatment oforganic contamination for environmental remediation.1–3

However, the application of traditional photocatalysts, e.g. TiO2

and ZnO, was limited by their inability to absorb visible light,though they may exhibit excellent photocatalytic performanceunder UV irradiation.4–6 To cope with these problems, greatefforts have been made in the development of new photo-catalysts with visible-light response and improvement of thephotocatalytic activity.7,8 Among these strategies, the fabricationof heterojunction photocatalysts by the coupling of two semi-conductors with appropriate energy band levels has been asignicant approach that can effectively separate and transfer

als, School of Materials Science and

es, Beijing, 100083, PR China. E-mail:

86-10-82332247

e, City University of Hong Kong, Tat Chee

(ESI) available: Diffuse reectanceuxWO4, SEM image and ICP elementalheterojunction and photocatalyticples. See DOI: 10.1039/c4ra05708b

hemistry 2014

the photogenerated electron–hole pairs, thus resulting in highphotocatalytic activity.9–11 Nevertheless, construction of a het-erojunction system demands that the energy band levels of thetwo semiconductors must be overlapping and well matched.Therefore, the most crucial problem in constructing a hetero-junction is to seek and couple photocatalysts with matchedconduction band (CB) and valence band (VB).

Recently, bismuth tungstate (Bi2WO6), as the most studiedexample of the Aurivillius oxide family with perovskite-likestructure, has been found to exhibit excellent photocatalyticperformance under visible light, which can be ascribed to itslayered structure and moderate band gap (2.6 � 2.8 eV).12,13 Inorder to further improve the photocatalytic activity of Bi2WO6,various modications were adopted, including element doping(such as F, B, Fe, Mo, etc.)14–17 and especially coupling withheterogeneous semiconductors, for instance, TiO2/Bi2WO6,18

ZnWO4/Bi2WO6,19 WO3/Bi2WO6,20 BiOI/Bi2WO6,21 C3N4/Bi2WO6

(ref. 22) and BiIO4/Bi2WO6.23 Here, we describe a novel designstrategy to develop a visible-light-driven (VLD) heterojunctionsystem with Bi2WO6 utilizing the tungstate solid solutionMg1�xCuxWO4 due to the following considerations: (1) thenarrow-band-gap semiconductor CuWO4 ensures the absorp-tion of visible light.24 (2) Compared with the high electronega-tivity of the Zn (4.45 eV) and Cd atoms (4.33 eV), the much lower

RSC Adv., 2014, 4, 41219–41227 | 41219

RSC Advances Paper

electronegativity of the Mg atom (3.75 eV) than the Cu atom(4.48 eV) could adjust the CB and VB positions and mightconstruct overlapping band structures in the heterojunction. (3)As belonging to the tungstate family, they can be developed via afacile one-step hydrothermal method.

In this study, the above considerations have been success-fully justied. We have prepared Mg0.7Cu0.3WO4/Bi2WO6

composite photocatalysts consisting of Mg0.7Cu0.3WO4 nano-cubes decorated with Bi2WO6 nanoplates by a one-step hydro-thermal method. Under visible-light irradiation, theMg0.7Cu0.3WO4/Bi2WO6 composite exhibited much higherphotocatalytic activity than those of the two individual compo-nents, which was veried by the photodegradation of rhoda-mine B (RhB) and photoelectrochemical measurements. Theenhancement of the VLD photocatalytic activity was ascribed tothe high separation efficiency for photogenerated electron–holepairs at the intimate interface of the heterojunctions.

2. Experimental section2.1 Materials and synthesis procedure

The raw materials Bi(NO3)3$5H2O, Mg(NO3)2$6H2O,Cu(NO3)2$6H2O and Na2WO4$2H2O, as well as other reagentsincluding ethylenediaminetetraacetic acid disodium salt(EDTA-2Na), isopropanol (IPA), benzoquinone (BQ) and nitrobluetetrazolium (NBT) from commercial sources, were all AR gradeand used as received without further purication. Mg1�xCuxWO4

(x ¼ 0, 0.1, 0.2, 0.3, 0.4 and 1) and Mg0.7Cu0.3WO4/Bi2WO6

samples were synthesized by a hydrothermal method.In a typical synthesis of Mg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3,

0.4 and 1), total amounts of 0.001 mol Mg(NO3)2$6H2O andCu(NO3)2$6H2O (molar numbers 0.001 : 0, 0.0009 : 0.0001,0.0008 : 0.0002, 0.0007 : 0.0003, 0.0006 : 0.0004 and 0 : 0.001,respectively) were dissolved in 30 ml deionized water, and thebeaker was placed in an ultrasonic bath for 10 min to dissolveraw materials. Meanwhile, 0.001 mol Na2WO4$2H2O was dis-solved in 30 ml deionized water to obtain a clear solution. Thenthe solution was mixed and then stirred for 3 h at roomtemperature. The resulting suspension was subsequentlytransferred into a 100 ml Teon-lined stainless steel autoclaveand heated at 180 �C for 24 h. Aer cooling, the products werecollected by ltration, washed repeatedly with deionized water,and then dried at 60 �C for 12 h.

In a typical synthesis of a Mg0.7Cu0.3WO4/Bi2WO6 samplewith a molar ratio 1 : 4, 0.008 mol Bi(NO3)3$5H2O, 0.0007 molMg(NO3)2$6H2O and 0.0003 mol Cu(NO3)2$6H2O were added to30 ml deionized water, and the beaker was placed in an ultra-sonic bath for 10 min to dissolve raw materials. Meanwhile,0.005 mol Na2WO4$2H2O was dissolved in 30 ml deionizedwater to obtain a clear solution. Then the solution was mixed,and then stirred for 3 h. The resulting suspension was subse-quently transferred into a 100 ml Teon-lined stainless steelautoclave and heated at 180 �C for 24 h. Aer cooling, theproducts were collected by ltration, washed repeatedly withdeionized water, and then dried at 60 �C for 12 h. According tothis method, different molar ratios of Mg0.7Cu0.3WO4/Bi2WO6

samples of 1 : 2, 1 : 6 and 1 : 10 were prepared, respectively.

41220 | RSC Adv., 2014, 4, 41219–41227

Pure Bi2WO6 samples were also synthesized under the sameconditions as references.

2.2 Catalyst characterization

Powder X-ray diffraction (XRD) was performed on an X/max-rAAdvance diffractometer with Cu Ka radiation. X-ray photoelec-tron spectroscopy (XPS) with Al Ka X-ray (hv ¼ 1486.6 eV) irra-diation operating at 150 W was employed to investigate thechemical composition and surface properties of the samples.An S-4800 scanning electron microscope (SEM) was used toobserve the general morphology and microstructure of theproducts. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) analyseswere performed using a JEM-2100F electron microscope (JEOL,Japan). A PerkinElmer Lambda 35 UV-vis spectrometer wasutilized to record the UV-vis diffuse reectance spectra (DRS).The photoluminescence (PL) spectra were measured on a JOBINYVON FluoroMax-3 uorescence spectrophotometer with a 150W xenon lamp as the excitation lamp.

2.3 Photocatalytic activity experiment

The photocatalytic activities of Mg0.7Cu0.3WO4, Bi2WO6 andMg0.7Cu0.3WO4/Bi2WO6 heterojunctions were evaluated bydecomposition of RhB under visible-light irradiation with a1000W xenon lamp (l > 420 nm). A total of 50 mg of as-preparedphotocatalyst was dispersed in an aqueous solution of RhB (50ml, 0.01 mM). First, the dye solution and photocatalyst werestrongly magnetically stirred in the dark for 1 h to obtainadsorption–desorption equilibrium. Under irradiation, about 2ml of the suspension was taken at given time intervals and thenwas separated by centrifugation. The concentration of RhB wasanalyzed by recording the absorbance at the characteristic bandof 553 nm using a Cary 5000 UV-vis spectrophotometer.

2.4 Photoelectrochemical measurements

Photoelectrochemical measurements were carried out in athree-electrode system with a 0.1 M Na2SO4 electrolyte solution.Saturated calomel electrodes (SCE) and platinum wire wereused as the reference electrodes and counter electrode,respectively. The working electrodes were Bi2WO6 and 1 : 4Mg0.7Cu0.3WO4/Bi2WO6 lm electrodes. The photo-electrochemical measurements were performed on an electro-chemical system (CHI-660B, China) with light intensity 1 mWcm�2. The photocurrent (PC) generation and electrochemicalimpedance spectra (EIS) of the photocatalysts with visible lighton and off were measured at 0.0 V. A 5 mV sinusoidal ACperturbation was applied to the electrode over the frequencyrange of 0.05–105 Hz.

2.5 Active-species trapping and cO2� quantication

experiments

To detect the active species generated in the photocatalyticreaction, various scavengers, including 1 mM BQ (a quencher ofcO2

�), 1.0 mM IPA (a quencher of cOH) and 1 mM EDTA-2Na (aquencher of h+) were added.25,26 The method was similar to the

This journal is © The Royal Society of Chemistry 2014

Paper RSC Advances

former photocatalytic activity experiment. NBT (0.025 mM,exhibiting an absorption maximum at 259 nm) was utilized todetermine the amount of cO2

� generated from Mg0.7Cu0.3WO4/Bi2WO6 heterojunctions. The production of cO2

� was quanti-tatively analyzed via recording the concentration of NBT with aUV-vis spectrophotometer. This method was also similar to theformer photocatalytic activity and active-species trappingexperiments with NBT replacing the RhB.

3. Results and discussion3.1 Band gap engineering design of Mg1�xCuxWO4 solid-solution and Mg0.7Cu0.3WO4/Bi2WO6 heterojunctions

The ultraviolet-visible diffuse reectance spectra ofMg1�xCuxWO4

(x ¼ 0, 0.1, 0.2, 0.3, 0.4 and 1) samples are displayed in Fig. 1a. Insemiconductors, the square of the absorption coefficient is linearwith energy for direct optical transitions in the absorption edgeregion; whereas the square root of the absorption coefficient islinear with energy for indirect transitions.27 The absorption edgesof MgWO4 and CuWO4 are caused by direct and indirect transi-tions, respectively.28 Since the visible-light absorption of Bi2WO6

was caused by band gap transition,29 the band gap of Bi2WO6 wasestimated from the plot of absorption versus energy (Fig. S1†). Asthe energy levels of MgWO4 and CuWO4 do not match that ofBi2WO6 (Fig. 1c), they could not form effective heterojunctionsfavoring the separation and transfer of charge carriers. In order toconstruct energy-levelmatched band structures, the solid solutionof Mg1�xCuxWO4 was synthesized to adjust the energy bandstructure, especially the position of the conduction band (CB) andvalence band (VB). When the amount of Mg is larger than that ofCu (x# 0.4) in the solid solution of Mg1�xCuxWO4, it possesses adirect band gap.

Fig. 1 (a) UV-vis diffuse reflectance spectra and (b) band gap energiesof Mg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3, 0.4 and 1) samples. (c)Comparison of energy levels of Mg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3, 0.4and 1) and Bi2WO6.

This journal is © The Royal Society of Chemistry 2014

The band positions of semiconductors can be predictedusing the electronegativity concept, and the CB and VB poten-tials of the semiconductor at the point of zero charge can becalculated by the following equation:30

EVB ¼ X � E e + 0.5Eg (1)

ECB ¼ EVB � Eg (2)

where X is the absolute electronegativity of the semiconductor,which is dened as the geometric average of the absolute elec-tronegativities of the constituent atoms, Ee is the energy of freeelectrons on the hydrogen scale (z4.5 eV), and Eg is the bandgap.2,30 The band gaps of Mg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3 and0.4) are determined from the data plots of absorption2 versusenergy in the absorption edge region (Fig. 1b), and the CB andVB were also estimated (Table S1†). Fig. 1c presents the energybands of Mg1�xCuxWO4 and Bi2WO6; it can be seen that band-matched overlapping band structures between Mg1�xCuxWO4

and Bi2WO6 have been fabricated when x is between 0.1 and 0.4.Fig. 2 displays the UV-vis DRS spectra of Mg0.7Cu0.3WO4,

Bi2WO6, and Mg0.7Cu0.3WO4/Bi2WO6 heterostructures. Theabsorption edge of pure Bi2WO6 is located at about 450 nm inthe visible region while that of Mg0.7Cu0.3WO4 is at approxi-mately 550 nm. In contrast, all the Mg0.7Cu0.3WO4/Bi2WO6

heterostructures exhibit red shis on their absorption edgescompared with pure Bi2WO6, and their edges range from 450 to550 nm.

3.2 Characterization of Mg0.7Cu0.3WO4/Bi2WO6

heterojunctions

As presented in Fig. 3a, orthorhombic phase Bi2WO6 with spacegroup Pca2 possesses an Aurivillius layered structure, which iscomposed of [Bi2O2]

2+ layers together with WO6 octahedrabetween them. Shown in Fig. 3b is the schematic crystal struc-ture of monoclinic phase Mg1�xCuxWO4. It is composed ofMg1�xCuxO6 and WO6 octahedra. Judging from the crystal

Fig. 2 UV-vis diffuse reflectance spectra of Mg0.7Cu0.3WO4/Bi2WO6

with different molar ratios.

RSC Adv., 2014, 4, 41219–41227 | 41221

Fig. 3 Crystal structure of (a) orthorhombic phase Bi2WO6 and (b) monoclinic phase Mg0.7Cu0.3WO4.

RSC Advances Paper

structure, Mg1�xCuxWO4 may display a cube-like surfacemorphology. X-ray powder diffraction (XRD) was utilized toverify the Mg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3, 0.4 and 1) solidsolution as shown in Fig. 4a. It is obvious that MgWO4,Mg0.9Cu0.1WO4, Mg0.8Cu0.2WO4, Mg0.7Cu0.3WO4, Mg0.6Cu0.4WO4

Fig. 4 XRD patterns of (a) Mg1�xCuxWO4 (x¼ 0, 0.1, 0.2, 0.3, 0.4 and 1)and (b) Mg0.7Cu0.3WO4/Bi2WO6 with molar ratios 0 : 1, 1 : 10, 1 : 6,1 : 4, 1 : 2 and 1 : 0.

41222 | RSC Adv., 2014, 4, 41219–41227

and CuWO4 with monoclinic structures were successfullysynthesized. Moreover, it can also be seen that the diffractionpeaks of the Mg1�xCuxWO4 solid solution shi from the 2qposition of MgWO4 to that of CuWO4 with the increase in Cucontent, further indicating the successful preparation of theMg1�xCuxWO4 (x ¼ 0, 0.1, 0.2, 0.3, 0.4 and 1) series.

Fig. 4b depicts the XRD patterns of the as-preparedMg0.7Cu0.3WO4, Bi2WO6 and Mg0.7Cu0.3WO4/Bi2WO6 hetero-junctions. From the image, it can be seen that the pure Bi2WO6

sample exhibits high purity and crystallinity. The diffractionpeaks can be identied as the orthorhombic phase of Bi2WO6

(PDF#39-0256). Due to the high content and intensity of theBi2WO6 diffraction peaks, the Mg0.7Cu0.3WO4/Bi2WO6 hetero-junctions with molar ratio <1 : 4 display similar diffractionpeaks to the pure Bi2WO6.22 In the Mg0.7Cu0.3WO4/Bi2WO6

composites with molar ratios 1 : 4 and 1 : 2, the characteristicdiffraction peaks of Mg0.7Cu0.3WO4 were successfully detected,conrming the coexistence of both Mg0.7Cu0.3WO4 and Bi2WO6.

The X-ray photoelectron spectroscopy (XPS) analysis wasemployed to investigate the chemical composition and surfacechemical states of the Mg0.7Cu0.3WO4/Bi2WO6 heterostructures.The typical survey XPS spectrum indicates that Bi, W, O, Mg andCu elements could be detected for the 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 heterostructure (Fig. 5a). The XPS peak for C 1s (285 eV)is ascribed to adventitious hydrocarbon from the XPS instru-ment. Fig. 5b–f show the high-resolution XPS spectra. As shownin Fig. 5b, the binding energies ofW 4f7/2 andW 4f5/2 are locatedat 37.6 and 35.5 eV, respectively.31 Two strong peaks at 159.4 and164.7 eV with a difference (delta) in binding energies of 5.35 eVshown in Fig. 5c are attributed to Bi 4f7/2 and Bi 4f5/2, respec-tively, which are characteristic of Bi3+ in Bi2WO6.32 The O 1speaks for the Mg0.7Cu0.3WO4/Bi2WO6 heterostructure (Fig. 5d)can be deconvoluted into three peaks at 530.1 eV, 531.2 eV and532.5 eV, which correspond to the lattice oxygen, –OH hydroxylgroups and chemisorbed water, respectively.16 As for the high-resolution Mg and Cu XPS spectra (Fig. 5e and f, respectively),two peaks at 1303.0 eV and 935.1 eV are associated with Mg 1s(ref. 33) and Cu 2p3/2,34 respectively. The XPS results furtherconrmed the coexistence of Mg0.7Cu0.3WO4 and Bi2WO6 in theMg0.7Cu0.3WO4/Bi2WO6 heterostructures.

To further conrm the composition of the heterojunctionsquantitatively, ICP elemental analyses were performed on the

This journal is © The Royal Society of Chemistry 2014

Fig. 5 XPS spectra of the 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 heterostructures: (a) survey, (b) W 4f, (c) Bi 4f, (d) O 1s, (e) Mg 1s and (f) Cu 2p.

Paper RSC Advances

Mg0.7Cu0.3WO4/Bi2WO6 products. The results (Table S2†)showed that the molar ratios of Mg : Cu : Bi : W are in accor-dance with the expected values.

The surface morphology of Mg0.7Cu0.3WO4/Bi2WO6 withmolar ratios 1 : 0, 1 : 2, 1 : 4, 1 : 6, 1 : 10 and 0 : 1 has beenobserved by scanning electron microscopy (SEM) as shown inFig. 6. It can be seen that Mg0.7Cu0.3WO4 (Fig. 6a) and Bi2WO6

(Fig. 6f) possess nanocube and nanoplate structures, respec-tively. With the increase in the Bi2WO6/Mg0.7Cu0.3WO4 molarratio, the content of the nanoplates also increased (Fig. 6a–d).Fig. 6c and S2† showed that the as-prepared products ofMg0.7Cu0.3WO4/Bi2WO6 with molar ratio 1 : 4 consist of irreg-ular nanocubes and nanoplates, and the crystal dimension ofthe nanocubes was estimated to be from several tens of nano-meters to one micron, and the thicknesses of the nanosheets ofBi2WO6 are in the range of 40–50 nm (Fig. S2†). TheMg0.7Cu0.3WO4/Bi2WO6 heterojunction was further characterizedby TEM. The low-magnication TEM images in Fig. 7a and bconrmed the nanocube and nanoplate morphologies of the

This journal is © The Royal Society of Chemistry 2014

sample. The high-resolution transmission electron microscopy(HRTEM) image (Fig. 7c) and fast Fourier transform (FFT)images (Fig. 7d and f) conrmed the single crystal natures ofMg0.7Cu0.3WO4 and Bi2WO6. The lattice-resolved HRTEMimages indicate that the spacings of the lattice are 0.181 and0.310 nm (as seen in Fig. 7e and g, respectively), which isconsistent with the spacings of the corresponding (022) and(131) planes of monoclinic Mg0.7Cu0.3WO4 and orthorhombicBi2WO6, respectively. The interface between Bi2WO6 andMg0.7Cu0.3WO4 crystals can also be observed from the image.

3.3 Photocatalytic performance

On the basis of the above results, improved photocatalyticactivity would be obtained when Mg1�xCuxWO4 (x¼ 0.1, 0.2, 0.3and 0.4) and Bi2WO6 were combined into heterojunctions. SinceMg0.7Cu0.3WO4 exhibits the highest photocatalytic activity(Fig. S3†) among the solid-solution samples, the photo-degradation of RhB has been investigated to evaluate the pho-tocatalytic activity of as-prepared Mg0.7Cu0.3WO4/Bi2WO6

RSC Adv., 2014, 4, 41219–41227 | 41223

Fig. 6 SEM images of as-prepared products with Mg0.7Cu0.3WO4/Bi2WO6 molar ratios: (a) 1 : 0, (b) 1 : 2, (c) 1 : 4, (d) 1 : 6, (e) 1 : 10 and (f)0 : 1.

Fig. 7 (a and b) TEM, (c) HRTEM images, (d and f) FFT (fast Fouriertransition) patterns and (e and g) inverse FFT patterns of the latticefringe of Mg0.7Cu0.3WO4/Bi2WO6 heterostructures.

Fig. 8 (a) Photocatalytic degradation curves of RhB under irradiationof visible light for Mg0.7Cu0.3WO4/Bi2WO6 with different molar ratios.(b) Temporal absorption spectral patterns of RhB during the photo-degradation process over 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6. (c) Apparentrate constants for the photodegradation of RhB over Mg0.7Cu0.3WO4/Bi2WO6 with different molar ratios.

RSC Advances Paper

heterojunctions under visible-light irradiation. The degradationdegree of RhB was examined by determining the change in itscharacteristic absorption peak at 554 nm. As displayed in

41224 | RSC Adv., 2014, 4, 41219–41227

Fig. 8a, the pure Mg0.7Cu0.3WO4 and Bi2WO6 show relativelypoor activity. When Mg0.7Cu0.3WO4 and Bi2WO6 were combinedto construct Mg0.7Cu0.3WO4/Bi2WO6 heterostructures, it wasfound that the photocatalytic activity of Mg0.7Cu0.3WO4/Bi2WO6

heterojunctions with a molar ratio higher than 1 : 10 is

This journal is © The Royal Society of Chemistry 2014

Paper RSC Advances

signicantly improved compared with pure Mg0.7Cu0.3WO4 andBi2WO6. When the theoretical molar ratio of Mg0.7Cu0.3WO4 toBi2WO6 was 1 : 4, the highest photocatalytic activity wasobtained, resulting in a degradation efficiency of RhB of 97.2%aer 2 h irradiation. As shown in Fig. 8b, the characteristicabsorption peak of RhB at 554 nm decreases with increasingtime, which is consistent with the degradation curve.

Besides, in order to quantitatively understand the reactionkinetics of the photocatalytic degradation process of RhB, akinetic study was performed by employing the pseudo-rst-order model:35,36

ln(C0/C) ¼ kappt (3)

where kapp is the apparent pseudo-rst-order rate constant(h�1), C0 is the initial RhB concentration (mg L�1), and C is theRhB concentration in aqueous solution at time t (mg L�1).Fig. 8c shows the RhB photodegradation apparent rateconstants for different Mg0.7Cu0.3WO4/Bi2WO6 molar ratios.The experimental data present the corresponding kapp values ascalculated to be 0.10 h�1, 0.68 h�1, 1.62 h�1, 0.81 h�1, 0.236 h�1

and 0.30 h�1 for Mg0.7Cu0.3WO4/Bi2WO6 with molar ratios 1 : 0,1 : 2, 1 : 4, 1 : 6, 1 : 10 and 0 : 1, respectively. The results showthat 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 demonstrates the highest kappvalue (1.62 h�1), which is almost 30 and 5.4 times higher thanthose of pure Mg0.7Cu0.3WO4 and Bi2WO6, respectively, indi-cating that Mg0.7Cu0.3WO4/Bi2WO6 is an excellent compositephotocatalyst under visible light.

3.4 Investigation of photocatalytic mechanism

Photoluminescence (PL) emission spectra can serve as a usefulapproach to investigate the separation and transfer efficiency ofphotogenerated charge carriers in semiconductors, since PLemissions may result from the recombination of free carriers.37

Generally, a decrease in recombination rate leads to a lower PLintensity, thus higher photocatalytic activity. Fig. 9 presents thePL spectra of the Mg0.7Cu0.3WO4/Bi2WO6 composites with

Fig. 9 PL spectra of Mg0.7Cu0.3WO4/Bi2WO6 with molar ratios 1 : 2,1 : 4, 1 : 6 and 1 : 10.

This journal is © The Royal Society of Chemistry 2014

molar ratios 1 : 2, 1 : 4, 1 : 6 and 1 : 10 at room temperature. Itcan be seen that 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 displays thelowest emission peaks and thus possesses the highest photo-catalytic activity, which is in good agreement with the resultsfrom the photodegradation experiment.

Fig. 10a shows the photocurrent of Bi2WO6 and 1 : 4Mg0.7Cu0.3WO4/Bi2WO6 samples generated in an electrolyteunder visible light, which may indirectly correlate with thegeneration and transfer of the photoinduced charge carriers inthe photocatalytic process.38 As shown in the gure, theobserved photocurrent generation is quite reversible and showsgood reproducibility, indicating that the electrode is stable.Compared to pure Bi2WO6, 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6

exhibits an obviously enhanced photocurrent response, whichis about 4 times that of pristine Bi2WO6. This remarkablyenhanced photocurrent response further conrmed the moreefficient separation and transfer of photoinduced electron–holepairs occurring at the interface of the Mg0.7Cu0.3WO4/Bi2WO6

heterojunction.

Fig. 10 (a) Comparison of transient photocurrent response and (b) EISNyquist plots of Bi2WO6 and 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 with lighton/off cycles under visible-light irradiation (l > 420 nm, [Na2SO4] ¼0.1 M).

RSC Adv., 2014, 4, 41219–41227 | 41225

Scheme 1 Schematic diagrams of Mg0.7Cu0.3WO4/Bi2WO6 hetero-structures under visible-light irradiation.

RSC Advances Paper

As the separation and transfer processes of charges in theelectrode–electrolyte interface region are supposed to be indi-cated by the electrochemical impedance spectra (EIS) Nyquistplots,39 EIS technology can be used to investigate the photo-catalytic performance. Fig. 10b shows Nyquist plots of Bi2WO6

and 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 with and without visible-lightirradiation. It can be seen that the arc radius of 1 : 4Mg0.7Cu0.3WO4/Bi2WO6 is smaller than that of Bi2WO6, whichindicates that the heterojunction possesses a stronger ability inthe separation and transfer of photogenerated electron–hole pairs.

To detect the active species during the photodegradation ofRhB over Mg0.7Cu0.3WO4/Bi2WO6 heterojunctions, a trappingexperiment was carried out by adding various scavengers to thephotodegradation system. We utilized ethylenediaminetetra-acetic acid disodium salt (EDTA-2Na), isopropanol (IPA) andbenzoquinone (BQ) as hole (h+), hydroxyl radical (cOH) andsuperoxide radical (cO2

�) scavengers, respectively.25,26 It can beseen that the photodecomposition of RhB was almost unaf-fected by adding IPA. In contrast, the degradation efficiency forRhB was inhibited by about 100% and 80% with the addition of

Fig. 11 (a) Photodegradation of RhB over 1 : 4 Mg0.7Cu0.3WO4/Bi2WO6 in the presence of different scavengers. (b) Absorption spectraof NBT with 4 h visible-light irradiation (l > 420 nm).

41226 | RSC Adv., 2014, 4, 41219–41227

EDTA-2Na and BQ, respectively (Fig. 11a). Thus it can besupposed that photogenerated holes (h+) and cO2

� are the mainactive species of Mg0.7Cu0.3WO4/Bi2WO6 for RhB degradationunder visible-light irradiation.

To further understand the change in active species over theMg0.7Cu0.3WO4/Bi2WO6 heterojunction, detailed cO2

� quanti-cation experiments have been carried out. Fig. 11b shows theabsorption spectra centered at 259 nm of NBT under visible-light irradiation (l > 420 nm) for 4 h during the photocatalyticreaction. It is obvious that the absorption peak graduallydecreased with increasing irradiation time, which conrms thatcO2

� plays an important role in the photocatalytic reaction overMg0.7Cu0.3WO4/Bi2WO6 heterojunctions. It further demon-strates that in the Mg0.7Cu0.3WO4/Bi2WO6 heterojunction(Scheme 1), the photogenerated electrons on Mg0.7Cu0.3WO4

could easily transfer to Bi2WO6, leading more photogeneratedelectrons to react with O2 to produce cO2

� and take part in thedecomposition of RhB. Meanwhile, the holes (h+) migrate fromBi2WO6 to the VB of Mg0.7Cu0.3WO4, making charge separationmore efficient and reducing the recombination probability,which is in good agreement with the photocatalytic activity.

4. Conclusions

In summary, energy-levels well-matched Mg0.7Cu0.3WO4/Bi2WO6 heterojunctions have been successfully constructedthrough semiconductor band gap engineering based on solid-solution design and synthesized by a facile hydrothermalmethod. The as-designed Mg0.7Cu0.3WO4/Bi2WO6 hetero-junctions consisting of nanocube and nanoplate structuresexhibit highly improved visible-light-driven photocatalyticperformance compared to the pure component samples. Thiswas also conrmed by the photoelectrochemical measure-ments. The optimum photocatalytic activity of the 1 : 4Mg0.7Cu0.3WO4/Bi2WO6 sample for the degradation of RhB wasalmost 30 and 5.4 times higher than those of pristineMg0.7Cu0.3WO4 and Bi2WO6, respectively. Active-species

This journal is © The Royal Society of Chemistry 2014

Paper RSC Advances

trapping and quantication measurements indicated thatsuperoxide radicals (cO2

�) and photogenerated holes (h+) playa crucial role in the photodegradation of RhB overMg0.7Cu0.3WO4/Bi2WO6 heterojunctions. The fabrication ofwell-matched overlapping band structures can result in efficientphotogenerated charge transfer between Mg0.7Cu0.3WO4 andBi2WO6, enhancing the VLD photocatalytic reactivity.

Acknowledgements

This work was supported by the Fundamental Research Fundsfor the Central Universities (2652013052), and the NationalNatural Science Foundation of China under Grants 50590402,and 91022036, and the National Basic Research Project of China(2010CB630701, and 2011CB922204).

Notes and references

1 M. W. Kanan and D. G. Nocera, Science, 2008, 321, 1072.2 A. Kubacka, M. Fernandez-Garcıa and G. Colon, Chem. Rev.,2012, 112, 1555.

3 H. Tong, S. X. Ouyang, Y. P. Bi, N. Umezawa, M. Oshikiri andJ. H. Ye, Adv. Mater., 2012, 24, 229.

4 A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev., 1995, 95,735.

5 X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891.6 F. Z. Liu, Y. H. Leung, A. B. Djurisic, A. M. Ching Ng andW. K. Chan, J. Phys. Chem. C, 2013, 117, 12218.

7 Q. Li, H. Meng, P. Zhou, Y. Q. Zheng, J. Wang, J. G. Yu andJ. R. Gong, ACS Catal., 2013, 3, 882.

8 H. W. Huang, Y. He, Z. S. Lin, L. Kang and Y. H. Zhang, J.Phys. Chem. C, 2013, 117, 22986.

9 J. Cao, B. Xu, B. Luo, H. Lin and S. Chen, Catal. Commun.,2011, 13, 63.

10 L. Q. Ye, J. Y. Liu, Z. Jiang, T. Y. Peng and L. Zan, Appl. Catal.,B, 2013, 142, 1.

11 T. H. Yu, W. Y. Cheng, K. J. Chao and S. Y. Lu, Nanoscale,2013, 5, 7356.

12 H. Cheng, B. Huang, Y. Dai, X. Qin, X. Zhang, Z. Wang andM. Jiang, J. Solid State Chem., 2009, 182, 2274.

13 M. Ge, Y. Li, L. Liu, Z. Zhou and W. Chen, J. Phys. Chem. C,2011, 115, 5220.

14 R. Shi, G. L. Huang, J. Lin and Y. F. Zhu, J. Phys. Chem. C,2009, 113, 19633.

15 Y. Fu, C. Chang, P. Chen, X. L. Chu and L. Y. Zhu, J. Hazard.Mater., 2013, 254, 185.

16 S. Guo, X. F. Li, H. Q. Wang, F. Dong and Z. B. Wu, J. ColloidInterface Sci., 2012, 369, 373.

This journal is © The Royal Society of Chemistry 2014

17 Y. Tian, L. D. Zhang and J. X. Zhang, J. Alloys Compd., 2012,537, 24.

18 M. Shang, W. Z. Wang, L. Zhang, S. M. Sun, L. Wang andL. Zhou, J. Phys. Chem. C, 2009, 113, 14727.

19 D. Q. He, L. L. Wang, D. D. Xu, J. L. Zhai, D. J. Wang andT. F. Xie, ACS Appl. Mater. Interfaces, 2011, 3, 3167.

20 M. S. Gui, W. D. Zhang, Y. Q. Chang and Y. X. Yu, Chem. Eng.J., 2012, 197, 283.

21 H. Q. Li, Y. M. Cui and W. S. Hong, Appl. Surf. Sci., 2013, 264,581.

22 Y. L. Tian, B. B. Chang, J. L. Lu, J. Fu, F. N. Xi and X. P. Dong,ACS Appl. Mater. Interfaces, 2013, 5, 7079.

23 H. W. Huang, S. B. Wang, Na. Tian and Y. H. Zhang, RSCAdv., 2014, 4, 5561.

24 J. E. Yourey, K. J. Pyper, J. B. Kurtz and B. M. Bartlett, J. Phys.Chem. C, 2013, 117, 8708.

25 L. Ye, J. Liu, C. Gong, L. Tian, T. Peng and L. Zan, ACS Catal.,2012, 2, 1677.

26 X. J. Bai, L. Wang and Y. F. Zhu, ACS Catal., 2012, 2, 2769.27 X. Zhang, Z. H. Ai, F. L. Jia and L. Z. Zhang, J. Phys. Chem. C,

2008, 112, 747.28 J. E. Yourey, J. B. Kurtz and B. M. Bartlett, Inorg. Chem., 2012,

51, 10394.29 L. W. Zhang, Y. Man and Y. F. Zhu, ACS Catal., 2011, 1, 841.30 Y. Ohko, K. Hashimoto and A. Fujishima, J. Phys. Chem. A,

1997, 101, 8057.31 Y. G. Su, B. L. Zhu, K. Guan, S. S. Gao, L. Lv, C. F. Du,

L. M. Peng, L. C. Hou and X. J. Wang, J. Phys. Chem. C,2012, 116, 18508.

32 H. Liu, W. R. Cao, Y. Su, Y. Wang and X. H. Wang, Appl.Catal., B, 2012, 111, 271.

33 J. S. Corneille, J. W. He and D. W. Goodman, Surf. Sci., 1994,306, 269.

34 J. Morales, L. Sanchez, F. Martın, J. R. Ramos-Barrado andM. Sanchez, Thin Solid Films, 2005, 474, 133.

35 C. H. Wu, H. W. Chang and J. M. Chern, J. Hazard. Mater.,2006, 137, 336.

36 Y. He, Y. H. Zhang, H. W. Huang, N. Tian and Y. Luo, Inorg.Chem. Commun., 2014, 40, 55.

37 X. J. Wang, Q. Wang, F. T. Li, W. Y. Yang, Y. Zhao, Y. J. Haoand S. J. Liu, Chem. Eng. J., 2013, 234, 361.

38 H. Kim, P. Borse, W. Choi and J. Lee, Angew. Chem., Int. Ed.,2005, 44, 4585.

39 Z. Hosseini, N. Taghavinia, N. Shari, M. Chavoshi andM. Rahman, J. Phys. Chem. C, 2008, 112, 18686.

RSC Adv., 2014, 4, 41219–41227 | 41227