CrystEngComm

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

PAPER View Article OnlineView Journal | View Issue

CrystEngCommThis journal is © The Royal Society of Chemistry 2014

a Environment Functional Materials Division, Shenyang National Laboratory for

Materials Science, Institute of Metal Research, Chinese Academy of Sciences,

Shenyang, 110016, PR China. E-mail: qili@imr.ac.cn; Fax: +86 24 23971215;

Tel: +86 24 83978028bDepartment of Materials Science and Engineering, University of Illinois at

Urbana-Champaign, Urbana, IL 61801, USA

† Electronic supplementary information (ESI) available: SEM images of thesample synthesized with only DI H2O as the solvent, SEM images of the samplescalcined at 350 °C and 500 °C, and the estimation of the CB and VB edges ofWO3 and WO3·H2O. See DOI: 10.1039/c4ce00857j

Cite this: CrystEngComm, 2014, 16,

7493

Received 23rd April 2014,Accepted 6th June 2014

DOI: 10.1039/c4ce00857j

www.rsc.org/crystengcomm

Template-free solvothermal synthesis ofWO3/WO3·H2O hollow spheres and theirenhanced photocatalytic activity fromthe mixture phase effect†

Yang Liu,a Qi Li,*a Shian Gaoa and Jian Ku Shangab

Well-defined WO3·H2O hollow spheres composed of nanoflakes were successfully synthesized by a

template-free solvothermal process with i-PrOH–H2O mixture solvent. In this process, the tungsten pre-

cursor was firstly hydrolyzed to form solid spheres composed of both WO3·2H2O and WO3·H2O phases.

Then, these solid spheres underwent an Ostwald ripening process and the WO3·2H2O phase was

dehydrated to WO3·H2O at the same time to form hollow spheres with a pure WO3·H2O phase. With

appropriate calcination temperature, hollow spheres with WO3 (major) and WO3·H2O (minor) mixture

phases were created. These hollow spheres with WO3 and WO3·H2O mixture phases demonstrated a

largely enhanced photocatalytic activity for RhB degradation under visible light irradiation compared with

either pure WO3·H2O hollow spheres or pure WO3 hollow spheres which could be attributed to the

matched band structure between the WO3 and WO3·H2O phases. Thus, an effective charge carrier

separation can occur between the WO3 and WO3·H2O phases of these hollow spheres under visible

light irradiation, contributing to the observed largely enhanced photocatalytic performance.

Introduction

Inorganic materials with hollow spherical structures haveattracted much research attention due to their specific voidinterior and consequently unique physical/chemical proper-ties for various applications, such as active-material encapsu-lation, drug delivery, micro-reactors, filters, piezoelectrictransducers, low dielectric constant substrates and so on.1–5

Many research efforts have been devoted to the developmentof synthesis approaches for hollow spherical materials,including both template and template-free processes.1–5 Inrecent years, a template-free process based on Ostwald ripen-ing has been proven to be an effective approach for the crea-tion of hollow spherical inorganic materials without thepotential contamination with templates, and a series ofhollow spherical inorganic materials has been successfullysynthesized through this process, such as TiO2, Cu2O, Co3O4,ZnS, MnO2, SnO2, Fe3O4, VO2, BaZrO3, and so on.6–13

As n-type semiconductors, tungsten oxide (WO3) and tung-sten oxide hydrates (WO3·nH2O) have attracted considerableattention for their applications in gas sensors, electrochromicdevices, optical recording devices, and as visible light-responsive photocatalysts.14 WO3·nH2O is usually synthesizedthrough the liquid-phase synthesis routes, and WO3 is synthe-sized by annealing the WO3·nH2O to remove crystal water. Tocreate WO3 hollow spheres, WO3·nH2O hollow spheres weresynthesized by both the template process with various sacrifi-cial templates15–18 and the template-free process via Ostwaldripening.19–21 However, at present, the reported approaches tosynthesize WO3·nH2O hollow spheres using the template-freeprocess are still quite limited in the literature, although previ-ous reports19–21 have demonstrated that various tungsten pre-cursors and specific synthesis conditions could produceproducts with different morphologies and crystal structures.Hollow spheres as photocatalysts are beneficial to the photo-catalytic performance because themultiple reflections (scatter-ing) from their specific hollow structure can result in betterlight harvesting and the subsequent generation of more photo-electron–hole pairs.22–26 Thus, the synthesis of WO3·nH2Ohollow spheres via the template-free Ostwald ripening mecha-nism should be further investigated. Recently, we developeda template-free process which successfully synthesizedwell-dispersed hollow TiO2 spheres via Ostwald ripening, inwhich the use of a PrOH–H2O mixture solvent was critical inthe creation of TiO2 hollow spheres with uniform size and

, 2014, 16, 7493–7501 | 7493

CrystEngCommPaper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

good dispersity.22 It would be of interest to examine if thePrOH–H2O mixture solvent could be applicable to the synthe-sis of well-defined WO3·nH2O hollow spheres.

For semiconductor photocatalysts, the electron–hole pairrecombination problem could cause the loss of a large por-tion of photogenerated electron–hole pairs, thus reducingtheir photocatalytic efficiency. Besides noble metal deposi-tion and semiconductor coupling,27,28 the formation ofphotocatalysts with mixture phases has also been reported tobe beneficial to the separation of electron–hole pairs if thesemixture phases have matched band structures.29–38 For exam-ple, enhanced photocatalytic performance has been demon-strated in TiO2 photocatalysts with mixture phases, includinganatase/rutile,29–32 anatase/brookite,33,34 anatase/TiO2(B),

35

and rutile/brookite.36 Recently, In2O3/In2O3·3H2O37 and

WO3/WO3·H2O38 mixture phase photocatalysts have also been

reported. However, no report is available on the creation ofWO3-based hollow spheres with mixture phases, although thesynthesis of WO3 hollow spheres by the dehydration ofWO3·nH2O hollow spheres could offer the possibility of creat-ing hollow spheres with mixture phases of pure WO3 andWO3 with certain amounts of crystal water.

Herein, well-defined WO3·H2O hollow spheres composedof nanoflakes were successfully synthesized by a solvo-thermal process with i-PrOH–H2O mixture solvent. Thehollowing process was investigated, the results of whichdemonstrated clearly that their formation followed theOstwald ripening mechanism. Hollow spheres with WO3

(major) and WO3·H2O (minor) mixture phases created bycalcination with appropriate temperature demonstrated alargely enhanced photocatalytic activity for RhB degradationunder visible light irradiation compared with either pureWO3·H2O hollow spheres or pure WO3 hollow spheres. TheWO3·H2O and WO3 phases had matched band structures.Thus, an effective charge carrier separation can occurbetween the WO3 phase and the WO3·H2O phase under visi-ble light irradiation, contributing to their largely enhancedphotocatalytic performance.

ExperimentalChemicals and materials

Sodium tungstate dehydrate (Na2WO4·2H2O, 99.5%), oxalicacid (H2C2O4·2H2O, 99.8%), hydrochloric acid (HCl, 36%), andisopropyl alcohol (i-PrOH, 99.7%) were purchased fromSinopharm Chemical Reagent Co., Ltd. (Shanghai, P.R. China).Deionized (DI) water was prepared using a lab ultra-pure waterpurifier. Rhodamine B (RhB, AR) was purchased from ShenyangNo. 3 Chemical Reagent Factory (Shenyang, P.R. China).

Template-free synthesis of WO3·H2O hollow spheres andtheir dehydration

WO3·H2O hollow spheres were synthesized through asolvothermal process. Under vigorous stirring, 0.5 mmolNa2WO4·2H2O and 1 mmol H2C2O4·2H2O were added to30 mL of DI water. After they were dissolved completely,

7494 | CrystEngComm, 2014, 16, 7493–7501

36 mL of i-PrOH and 2 mL of HCl were added to this aque-ous solution in sequence to change the solvent from purewater to an i-PrOH–H2O mixture. The solution was thentransferred to a Teflon-lined autoclave with a capacity of90 mL which was then sealed, kept in an oven at 80 °C for12 h, and then air cooled to room temperature. To obtainsamples at various solvothermal reaction times, the auto-clave was kept at 80 °C for a desired time and then water-cooled to better preserve the morphology and structure ofthese samples. The obtained precipitate was centrifuged,washed with water and ethanol several times, and thendried at 70 °C to obtain WO3·H2O hollow spheres. WO3·H2Ohollow spheres were further calcinated at the desired tem-peratures (200 °C, 350 °C, and 500 °C) for 2 h in a mufflefurnace with a heating rate of 2 °C min−1.

Characterization of hollow spheres

SEM images were obtained using a Leo-Supra-55 scanningelectron microscope (Carl Zeiss, Oberkochen, Germany). TEMimages and SAED patterns were taken using a JEOL-2100transmission electron microscope (JEOL, Tokyo, Japan). XRDanalysis was conducted using a D/MAX-2004 X-ray powderdiffractometer (Rigaku, Tokyo, Japan) with Ni-filtered Cu Kαradiation (0.15418 nm) at 56 kV and 182 mA. TG-DSC curveswere measured using a STA 449 C Jupiter simultaneous ther-mal analyzer (Netzsch, Selb, Germany) from room temperatureto 550 °C with a heating rate of 10 °C min−1. The Brunauer–Emmett–Teller (BET) data were measured by nitrogen adsorp-tion–desorption isotherms using Autosorb-1 surface area andpore size analyzers (Quantachrome, Boynton Beach, USA). Thetotal pore volume and the average pore size were calculatedusing the Barrett–Joyner–Halenda (BJH) model. The opticalabsorbance was determined from the diffuse reflectance spec-tra (DRS) measurements with a UV-2550 UV-vis spectrophoto-meter (Shimadzu, Kyoto, Japan) using BaSO4 as the reference.

Photocatalytic degradation of Rhodamine B under visiblelight irradiation

The photocatalytic performance of both uncalcined and cal-cined samples was evaluated by examining their photocata-lytic degradation of RhB under visible light irradiation.50 mg of the photocatalyst was added to 50 mL of RhB solu-tion (20 mg L−1) in a 250 mL glass reactor with a quartz topcover. The solution was stirred for 30 min in the dark andthen exposed to visible light irradiation. The light source wasa 300 W Xe arc lamp (PLS-SXE300, Beijing Trusttech Co. Ltd)with a UV cutoff filter (λ > 400 nm). The light intensity was~100 mW cm−2. 2 mL of the solution was sampled at eachtime and the concentration of RhB was determined using aUV-vis spectrophotometer (UV-2550, Shimadzu, Kyoto, Japan).

Results and discussionThe structure and morphology of WO3·H2O hollow spheres

Fig. 1 shows the XRD pattern of the samples synthesized bya solvothermal process at 80 °C for 12 h in an i-PrOH–H2O

This journal is © The Royal Society of Chemistry 2014

Fig. 1 XRD pattern of the as-prepared WO3·H2O hollow spheres.

CrystEngComm Paper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

mixture solvent. All of the diffraction peaks could be indexedas the orthorhombic structure of the WO3·H2O phase (JCPDSno. 84-0886, space group: Pmnb(62), RIR: 12.81), suggestingthat the synthesis process we developed could produce asingle-phase WO3·H2O product. Fig. 2 shows the SEMimages, TEM images and corresponding SAED patterns ofthese WO3·H2O hollow spheres. It could be observed fromthe lower magnification SEM image in Fig. 2a that well-defined WO3·H2O hollow spheres were successfully createdusing our approach. Their sizes are highly uniform with anouter diameter of approximately 2 μm. Fig. 2b shows thehigher magnification SEM image of a single hollow sphereenlarged from the center area in Fig. 2a. From its broken

This journal is © The Royal Society of Chemistry 2014

Fig. 2 SEM images of the as-prepared WO3·H2O hollow spheres at (a) a loa single WO3·H2O hollow sphere. (d) The corresponding SAED pattern of tnanoflake. (f) The corresponding SAED pattern of the single nanoflake in (e)

shell, the hollow interiors of these spheres could be clearlyidentified. The shell thickness is ca. 200 nm, and the shell iscomposed of nanoflakes having a side length ranging fromtens to hundreds of nanometers and a thickness ranging fromseveral to tens of nanometers. Furthermore, nanoflakes of asmaller size are located at the inner area (marked as A) of theshell, while nanoflakes of a larger size are located at the outerarea (as B marked) of the shell. This observation indicates atypical Ostwald ripening process, in which smaller crystallitesdissolve and are transferred to the outer part for the growthof larger crystallites located at the outer part.6 Fig. 2c shows theTEM image of a single WO3·H2O hollow sphere; its hollownature could be easily identified from the contrast difference.It could also be observed that these nanoflakes were growncrosswise and large interstices exist between them, resultingin good permeability of the shells. The permeable nature ofthe shells could allow easy access of water, other molecules,and part of the incident photons to the shells and the interiorvoids. Thus, photocatalytic reactions could happen either atthe outer/inner surfaces or the shells of these hollow spheres.Fig. 2d shows the corresponding SAED pattern which indi-cates that these hollow spheres have a polycrystalline ortho-rhombic WO3·H2O phase. Fig. 2e shows the TEM image of asingle nanoflake, which has a nearly square shape, approxi-mately 280 × 200 nm2 in size. Its corresponding SAED pattern(Fig. 2f) suggests that these nanoflakes are single-crystalline.The diffraction spots could be indexed as (002) and (200) withthe electron beam parallel to the [020] direction. The SAEDpatterns are in accordance with the XRD analysis result.

CrystEngComm, 2014, 16, 7493–7501 | 7495

wer magnification and (b) a higher magnification. (c) The TEM image ofhe single WO3·H2O hollow sphere in (c). (e) The TEM image of a single.

Fig. 4 XRD patterns of the time-dependent samples synthesized at80 °C for 30 min, 90 min, 180 min, and 360 min.

CrystEngCommPaper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

The formation process and mechanism of WO3·H2O hollowspheres

To investigate the formation process and mechanism of thesehollow WO3·H2O spheres, time-dependent experiments wereconducted at 80 °C for different reaction times. Watercooling was adopted to stop the reaction immediately at thedesired time to better preserve the morphology and structureof these samples. Fig. 3a–d show the TEM images of thetime-dependent samples obtained at reaction times of 90,120, 150, and 180 min, respectively. Fig. 3 demonstrates thatthe hollow spheres evolved from the hollowing of solidspheres. Small voids firstly formed at the center of the solidspheres, gradually enlarging to finally form hollow interiors.Fig. 4 shows the XRD patterns of these time-dependent sam-ples. Different from the pure WO3·H2O phase of the sampleobtained after 12 h of reaction (Fig. 1), these time-dependentsamples (reaction time up to 6 h) were all composed of boththe WO3·H2O phase (JCPDS no. 84-0886) and the WO3·2H2Ophase (JCPDS no. 18-1420), wherein the WO3·H2O phase wasthe major one. As the reaction time increased from 30 min to360 min, the diffraction peak strength of the samplesincreased, indicating the enhanced crystallinity. Meanwhile,for each sample, the relative diffraction peak strength of theWO3·2H2O phase became weaker and weaker compared withthat of the WO3·H2O phase, indicating a phase transitionfrom WO3·2H2O to WO3·H2O with increasing reaction time.After 6 h of reaction, only a very small fraction of WO3·2H2Oremained. After 12 h of reaction, the sample had a pureWO3·H2O phase.

Thus, the synthesis of WO3·H2O hollow spheres followedthe Ostwald ripening mechanism. In this process, H2C2O4

and WO42− firstly formed the complex ion [WO2(C2O4)2]

2−

following eqn (1):

WO42− + 2H2C2O4 → [WO2(C2O4)2]

2− + 2H2O (1)

It is rather stable and wouldn't react with HCl at room tem-perature. This step is critical to the creation of the hollowspherical structure. When no H2C2O4 is present, precipitateswill form immediately when HCl is added to the sodium tung-state solution; the final product after solvothermal treatment iscomposed of irregularly shapedmicrometer-sized crystals.

7496 | CrystEngComm, 2014, 16, 7493–7501

Fig. 3 TEM images of the time-dependent samples synthesized at 80 °C fo

At elevated temperature under solvothermal conditions,[WO2(C2O4)2]

2− reacts with HCl to form precipitates of bothWO3·2H2O and WO3·H2O at a much slower hydrolysis ratefollowing eqn (2) and (3), respectively:

[WO2(C2O4)2]2− + 2H+ + 3H2O → WO3·2H2O + 2H2C2O4 (2)

[WO2(C2O4)2]2− + 2H+ + 2H2O → WO3·H2O + 2H2C2O4 (3)

These primary crystallites first aggregate into solid spheresas demonstrated in Fig. 3a. Then, these solid spheresundergo Ostwald ripening to finally form the hollow spheri-cal structure. In addition, the phase transition fromWO3·2H2O to WO3·H2O occur following eqn (4):

WO3·2H2O → WO3·H2O + H2O (4)

Thus, the amount of WO3·2H2O gradually decreased asthe reaction time increased, finally disappearing after 12 h ofreaction due to this dehydration reaction.

In this approach, another critical factor in the formationof well-defined hollow WO3·H2O spheres is the use ofi-PrOH–H2O mixture solvent. When only water was used asthe solvent, the obtained product was composed of bothindividual nanoflakes and hollow spherical aggregates of

This journal is © The Royal Society of Chemistry 2014

r (a) 90 min, (b) 120 min, (c) 150 min, and (d) 180 min.

CrystEngComm Paper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

nanoflakes with different sizes and shapes (see Fig. S1 inthe ESI†). As we found for n-PrOH in the creation ofwell-dispersed hollow TiO2 spheres in our previous work,22

the introduction of i-PrOH to H2O could reduce thehydrolysis rate of the tungsten precursor, which is beneficialto the aggregation of primary crystallites into well-definedsolid spheres.

The creation of hollow spheres with WO3·H2O and WO3

mixture phases

Thermal analysis was conducted on the WO3·H2O hollowsphere sample by measuring its TG and DSC curves as dem-onstrated in Fig. 5a. The sample was heated from roomtemperature to 550 °C at a heating rate of 10 °C min−1 andthen cooled to room temperature. From room temperature to~170 °C, the TG curve showed a slight decrease of ca.0.7% and a broad weak endothermic peak appeared on thecorresponding DSC curve. This part of mass loss mainlyresulted from the evaporation of surface-adsorbed water.From ~170 °C to ~280 °C, the TG curve had a largedecrease of ca. 6.7%, and a large endothermic peakappeared on the corresponding DSC curve with the peakposition at ~218 °C. From ~280 °C to ~500 °C, the TG curvehad a slight decrease of ca. 0.3%, and the DSC curveshowed a continuous endothermic process. Thus, the sum

This journal is © The Royal Society of Chemistry 2014

Fig. 5 (a) TG-DSC curves of the as-prepared WO3·H2O hollow spheres. ((c) The SEM image of the sample calcined at 200 °C.

of these two parts of mass loss of ca. 7% resulted from thephase transition from WO3·H2O to WO3. The theoreticalmass loss in the process is 7.2%, very close to the experi-mental data. Thermal analysis suggested that the phasetransition from WO3·H2O to WO3 began at ~170 °C andmostly finished at ~280 °C. With further temperatureincrease, the residual crystal water was removed and thecrystallinity of WO3 enhanced.

From the thermal analysis result, three temperatures of200, 350 and 500 °C were chosen for the calcination of theas-prepared WO3·H2O hollow spheres to create final prod-ucts with different phases. Fig. 5b shows the XRD patternsof these three samples calcined at different temperatures.The as-prepared WO3·H2O hollow spheres had a pureorthorhombic WO3·H2O phase. After calcination at 500 °Cfor 2 h, the sample completely transformed to the well-crystallized, pure WO3 phase (JCPDS no. 43-1035, spacegroup: P21/n(14), RIR: 5.04) with three distinct strong dif-fraction peaks of (002), (020), and (200). The sample cal-cined at 350 °C also only had a pure WO3 phase; howeverthe intensity of its diffraction peaks was weaker. In theXRD pattern of the sample calcined at 200 °C, however,two sets of diffraction peaks could be identified. Besidesthe major phase of WO3, the (111) and (020) diffractionpeaks of WO3·H2O could also be observed, suggesting thatit was composed of a mixture of WO3 and WO3·H2O. The

CrystEngComm, 2014, 16, 7493–7501 | 7497

b) XRD patterns of samples calcinated at 200 °C, 350 °C, and 500 °C.

CrystEngCommPaper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

WO3/WO3·H2O weight ratio (ω1/ω2) of the sample calcinedat 200 °C could be estimated by the reference intensitymethod using eqn (5):

II1

2

1

2

RIRRIR2

1

(5)

where I1 and I2 are the intensities of the WO3 (200) andWO3·H2O (111) peaks, respectively, measured from the XRDpattern. RIR1 and RIR2 are the reference intensities of theWO3 and WO3·H2O phases compared to the α-Al2O3 stan-dard, respectively, taken from the JCPDS cards. The bound-ary condition is that the sum of ω1 and ω2 is 1. From theXRD data, it could be roughly estimated that ω1/ω2 is~0.91/0.09. Thus, the chosen calcination temperature of200 °C based on the thermal analysis result created thefinal product with the WO3–WO3·H2O mixture phases.

Fig. 5c shows the SEM image of the sample calcined at200 °C; the SEM images of samples calcined at 350 °C and500 °C could be found in Fig. S2 in the ESI.† After calcina-tion, all samples kept their hollow sphere morphology, whichindicates that the mechanical strength of these hollowspheres was pretty good during the calcination to sustaintheir structure. Similar observations have been reported forWO3·nH2O hollow structures17,18,21 and TiO2 hollow struc-tures,22,23 in which the hollow structures were well-maintained during the calcination process.

Table 1 summarizes the BET specific surface area, porevolume, and pore size data for these samples. The samplecalcinated at 200 °C had BET specific surface area, pore vol-

7498 | CrystEngComm, 2014, 16, 7493–7501

Fig. 6 (a) Optical absorbance curves of the as-prepared WO3·H2O hollow splots of the as-prepared WO3·H2O spheres and the sample calcinated at 50

Table 1 BET data for the as-prepared WO3·H2O hollow spheres and sampl

Sample Specific surface area (m2 g−1)

WO3·H2O 22.0200 °C 22.2350 °C 17.6500 °C 9.3

ume, and pore size very close to those of the as-preparedWO3·H2O hollow spheres. When the calcination temperatureincreased to 350 °C and 500 °C, the obtained samples' BETspecific surface area decreased by ~20% and ~58%, respec-tively, and their pore volume and pore size also decreased.These results were in accordance with the XRD and DSC anal-ysis results. With a higher calcination temperature, the sam-ple's crystallinity increases, resulting in a lower BET specificsurface area and a smaller pore size/volume. Similar resultshave been reported for the calcination of WO3·H2O hollowspheres.18

Optical properties of hollow sphere samples

The optical absorbance of these samples was investigated byUV-vis diffuse reflectance spectroscopy (DRS) measurement,as shown in Fig. 6a. From the reflectance data, optical absor-bance could be approximated by the Kubelka–Munk functiongiven by eqn (6):

F R RR

( ) ( )

12

2

(6)

where R is the diffuse reflectance.39 The as-preparedWO3·H2Ohollow spheres had an absorption edge at ~525 nm, indicat-ing a good visible light absorption capability. After calcina-tion, the absorption stopping edges of all samples blue-shifted. Samples calcined at 350 °C and 500 °C had very closelight absorption curves with a stopping edge at ~455 nm,while the sample calcined at 200 °C had better visible lightabsorption with a stopping edge at ~510 nm. The samples

This journal is © The Royal Society of Chemistry 2014

pheres and samples calcinated at 200 °C, 350 °C, and 500 °C. (b) Tauc0 °C (pure WO3 phase) constructed from (a).

es calcinated at 200 °C, 350 °C, and 500 °C

Total pore volume (cm3 g−1) Average pore size (nm)

0.077 15.40.079 14.20.062 13.80.022 9.6

CrystEngComm Paper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

calcined at 350 °C and 500 °C were composed of a pure WO3

phase, therefore they had almost identical light absorptioncurves. The sample calcinated at 200 °C, however, was com-posed of a mixture of both WO3 (major) and WO3·H2O(minor) phases. Thus, its light absorption curve was betweenthose of WO3 and WO3·H2O and closer to that of WO3.

For semiconductors, the optical absorption near the bandedge follows eqn (7):39

αhν = A(hν − Eg)n (7)

where α, ν, A, and Eg are the absorption coefficient, light fre-quency, proportionality constant, and band gap, respectively,and n is equal to either 0.5 for allowed direct transition or 2for allowed indirect transition. The optical band gap of tung-sten oxide (hydrate) is between the oxygen 2p orbitals andthe tungsten 5d orbitals, which are indirect band transi-tions.40,41 Hence the value of n was determined to be 2.Fig. 6b shows the Tauc plots ((αhv)1/2 vs. hv) of theas-prepared WO3·H2O hollow spheres and the samplecalcinated at 500 °C (pure WO3 phase) constructed fromFig. 6a, demonstrating the linear Tauc region just above theoptical absorption edge to determine the semiconductorband gap. Extrapolation of this line to the photon energy axisyields the semiconductor band gap. Thus, the Eg of WO3·H2Owas determined to be ~2.35 eV, and the Eg of WO3 was deter-mined to be ~2.72 eV.

Photocatalytic degradation of RhB under visible lightirradiation by hollow sphere samples

The photocatalytic activities of these hollow sphere sampleswere demonstrated by their degradation effect on RhB undervisible light irradiation, as shown in Fig. 7. A control experi-ment with only visible light irradiation was also conducted,and no RhB concentration change was observed, suggestingthat visible light irradiation itself could not degrade RhB (not

This journal is © The Royal Society of Chemistry 2014

Fig. 7 Photocatalytic degradation curves of RhB under visible lightirradiation by the as-prepared WO3·H2O hollow spheres and the sam-ples calcinated at 200 °C, 350 °C, and 500 °C.

shown in Fig. 7). Before the photocatalytic degradation exper-iment, all samples were mixed with the RhB solution in thedark to reach the adsorption–desorption equilibrium. It wasfound that 0.5 h was enough for the adsorption–desorptionequilibrium to be established in these samples. For example,the adsorption kinetic curve of the sample calcinated at200 °C shown in Fig. 7 demonstrated that the residual RhBconcentration reached a plateau after 0.5 h of adsorption.Dark adsorption could remove ~18%, ~18%, ~13%, and ~8%RhB from the RhB solution for the as-prepared WO3·H2O hol-low spheres and samples calcinated at 200 °C, 350 °C, and500 °C, respectively. The difference observed may be attrib-uted to their BET specific surface area differences. Under visi-ble light irradiation, all samples demonstrated photocatalyticRhB degradation activities, which could be attributed to theirvisible light absorption as demonstrated in Fig. 6a. After 5 hof visible light irradiation, ~24%, ~60%, ~23%, and ~22%RhB in the RhB solution was further removed by photocata-lytic degradation using the as-prepared WO3·H2O hollowspheres and the samples calcinated at 200 °C, 350 °C, and500 °C, respectively. Thus, the sample calcinated at 200 °Cdemonstrated the best photocatalytic RhB degradationeffect under visible light irradiation among the four hollowsphere samples.

The slope of the RhB degradation curves in Fig. 7 repre-sents the RhB degradation rate at certain treatment times.The photocatalytic activity enhancement could be furtherdemonstrated quantitatively by the initial RhB degradationrates over different photocatalysts. The initial RhB degrada-tion rates were determined to be ~1.2, ~4.4, ~1.8, and~1.4 mg (g h)−1 for the as-prepared WO3·H2O hollow spheresand the samples calcinated at 200 °C, 350 °C, and 500 °C,respectively. The initial photocatalytic RhB degradation rateby the sample calcinated at 200 °C was ~3.7, ~2.4, and~3.1 times greater than those by the as-prepared WO3·H2Ohollow spheres and the samples calcinated at 350 °C and500 °C, respectively, clearly demonstrating its largely enhancedphotocatalytic activity under visible light irradiation.

Photocatalytic enhancement mechanism of theWO3/WO3·H2O sample

Although the visible light absorption of the as-preparedWO3·H2O hollow spheres was stronger than that of the sam-ple calcinated at 200 °C and their BET specific surface areaswere very close, its photocatalytic RhB degradation perfor-mance under visible light illumination was much worse thanthat of the sample calcinated at 200 °C. The pure WO3 hollowspheres (samples calcinated at 350 °C and 500 °C) had muchbetter crystallinity than the sample calcinated at 200 °C.However, their initial photocatalytic RhB degradation rateswere just ~41% and ~32% compared to that of the samplecalcinated at 200 °C, while their BET specific surface areaswere ~79% and ~42% compared to that of the samplecalcinated at 200 °C, respectively. This observation suggests

CrystEngComm, 2014, 16, 7493–7501 | 7499

CrystEngCommPaper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

that the surface area difference itself could not explain thedifferent photocatalytic activities.

It is well-known that photocatalysts may not demonstrategood photocatalytic activity even with high light absorptioncapacity because of the loss of charge carriers due to recom-bination. The effective separation of charge carriers is criticalto their photocatalytic performance. The largely enhancedphotocatalytic activity of the sample calcinated at 200 °Cunder visible light irradiation could be attributed to its spe-cific mixture nature of both WO3 and WO3·H2O phases. TheEg values of WO3·H2O and WO3 were determined from theTauc plots (Fig. 6b) to be ~2.35 eV and ~2.72 eV, respectively.The conduction band (CB) bottom values and valence band(VB) top values of the two semiconductors could be estimatedusing eqn (8) and (9):42

ECB ≈ Xcomp − Ee − 1/2Eg (8)

EVB = ECB + Eg (9)

where Ee is equal to 4.5 eV and Xcomp is the electronegativityof a compound. Calculation details could be found in theESI;† the CB and VB edges of WO3 are ~0.73 and ~3.45 eVwith respect to the NHE, respectively, and the CB and VBedges of WO3·H2O are ~1.21 and ~3.56 eV with respect to theNHE, respectively. Thus, an effective charge carrier separa-tion can occur between the WO3 and WO3·H2O phases, asschematically illustrated in Fig. 8. When the samplecalcinated at 200 °C was exposed to visible light irradiation,both WO3 and WO3·H2O phases could be excited, subse-quently generating electron–hole pairs. Because the CB ofWO3 is more negative than that of WO3·H2O, the photogene-rated electrons in the CB of WO3 could flow to the CB ofWO3·H2O, while the photogenerated holes in the VB ofWO3·H2O could flow to the VB of WO3 because the VB ofWO3·H2O is more positive than that of WO3. Thus, thephotogenerated electrons and holes could be separated effec-tively from each other and could further react with electronacceptors and electron donors, respectively, contributing tothe largely enhanced photocatalytic performance.

7500 | CrystEngComm, 2014, 16, 7493–7501

Fig. 8 Schematic illustration of the matched band structure and thecharge carrier separation in the WO3–WO3·H2O mixture phase sampleunder visible light irradiation.

Conclusions

In summary, well-defined WO3·H2O hollow spheres com-posed of nanoflakes were synthesized through a template-free solvothermal process via Ostwald ripening. Hollowspheres with WO3 (major) and WO3·H2O (minor) mixturephases were created by calcination with appropriate tempera-ture, and a matched band structure was created between thetwo phases. Thus, an effective charge carrier separation canoccur between the WO3 phase and the WO3·H2O phase undervisible light irradiation. These hollow spheres with WO3 andWO3·H2O mixture phases demonstrated a largely enhancedphotocatalytic activity for RhB degradation under visible lightirradiation compared with either pure WO3·H2O hollowspheres or pure WO3 hollow spheres.

Acknowledgements

The useful discussion with Dr. Wuzhu Sun is greatlyappreciated. This study was supported by the NationalNatural Science Foundation of China (Grant No. 51102246),the Knowledge Innovation Program of Institute of MetalResearch, Chinese Academy of Sciences (Grant No.Y0N5A111A1), the Youth Innovation Promotion Association,Chinese Academy of Sciences (Grant No. Y2N5711171), andthe Scientific Research Foundation for the Returned OverseasChinese Scholars, State Education Ministry, P. R. China.

Notes and references

1 J. K. Cochran, Curr. Opin. Solid State Mater. Sci., 1998, 3, 474.

2 X.W. Lou, L. A. Archer and Z. Yang, Adv.Mater., 2008, 20, 3987. 3 Q. Zhang, W. Wang, J. Goebl and Y. Yin, Nano Today, 2009,

4, 494.4 Y. Zhao and L. Jiang, Adv. Mater., 2009, 21, 3621.

5 J. Hu, M. Chen, X. Fang and L. Wu, Chem. Soc. Rev., 2011,

40, 5472.6 H. G. Yang and H. C. Zeng, J. Phys. Chem. B, 2004, 108,

3492.7 Y. Chang, J. J. Teo and H. C. Zeng, Langmuir, 2005, 21, 1074.

8 B. Liu and H. C. Zeng, Small, 2005, 1, 566. 9 B. Li, G. Rong, Y. Xie, L. Huang and C. Feng, Inorg. Chem.,

2006, 45, 6404.10 X. W. Lou, Y. Wang, C. Yuan, J. Y. Lee and L. A. Archer, Adv.

Mater., 2006, 18, 2325.11 B. Jia and L. Gao, J. Phys. Chem. C, 2008, 112, 666.

12 F. Y. Kong, M. Li, X. Y. Yao, J. M. Xu, A. D. Wang, Z. P. Liu

and G. H. Li, CrystEngComm, 2012, 14, 3858.13 Z. Dong, T. Ye, Y. Zhao, J. Yu, F. Wang, L. Zhang, X. Wang

and S. Guo, J. Mater. Chem., 2011, 21, 5978.14 H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell and

K. Kalantar-Zadeh, Adv. Funct. Mater., 2011, 21, 2175.15 X. L. Li, T. J. Lou, X. M. Sun and Y. D. Li, Inorg. Chem., 2004,

43, 5442.16 X. Zhao, T. L. Y. Cheung, X. Zhang, D. H. L. Ng and J. Yu,

J. Am. Ceram. Soc., 2006, 89, 2960.

This journal is © The Royal Society of Chemistry 2014

CrystEngComm Paper

Publ

ishe

d on

06

June

201

4. D

ownl

oade

d by

Yor

k U

nive

rsity

on

13/0

8/20

14 0

2:37

:37.

View Article Online

17 D. Chen and J. Ye, Adv. Funct. Mater., 2008, 18, 1922.

18 J. Yu, L. Qi, B. Cheng and X. Zhao, J. Hazard. Mater., 2008,

160, 621.19 S. Jeon and K. Yong, Chem. Commun., 2009, 7042.

20 Z. G. Zhao andM.Miyauchi, J. Phys. Chem. C, 2009, 113, 6539. 21 J. Huang, X. Xu, C. Gu, G. Fu, W. Wang and J. Liu, Mater.

Res. Bull., 2012, 47, 3224.22 Y. Liu, Q. Li, S. Gao and J. K. Shang, J. Am. Ceram. Soc.,

2013, 96, 1421.23 H. Li, Z. Bian, J. Zhu, D. Zhang, G. Li, Y. Huo, H. Li and

Y. Lu, J. Am. Chem. Soc., 2007, 129, 8406.24 Y. Li, T. Kunitake and S. Fujikawa, J. Phys. Chem. B, 2006,

110, 13000.25 Y. Kondo, H. Yoshikawa, K. Awaga, M. Murayama, T. Mori,

K. Sunada, S. Bandow and S. Iijima, Langmuir, 2008, 24, 547.26 H. J. Koo, Y. J. Kim, Y. H. Lee, W. I. Lee, K. Kim and

N. G. Park, Adv. Mater., 2008, 20, 195.27 H. Zhang, G. Chen and D. W. Bahnemann, J. Mater. Chem.,

2009, 19, 5089.28 S. G. Kumar and L. G. Devi, J. Phys. Chem. A, 2011, 115, 13211.

29 D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh and

M. C. Thurnauer, J. Phys. Chem. B, 2003, 107, 4545.30 S. Bakardjieva, J. Šubrt, V. Štengl, M. J. Dianez and

M. J. Sayagues, Appl. Catal., B, 2005, 58, 193.

This journal is © The Royal Society of Chemistry 2014

31 B. Sun and P. G. Smirniotis, Catal. Today, 2003, 88, 49.

32 T. van der Meulen, A. Mattson and L. Österlund, J. Catal.,

2007, 251, 131.33 S. Ardizzone, C. L. Bianchi, G. Cappelletti, S. Gialanella,

C. Pirola and V. Ragaini, J. Phys. Chem. C, 2007, 111,13222.

34 T. Ozawa, M. Iwasaki, H. Tada, T. Akita, K. Tanaka and

S. Ito, J. Colloid Interface Sci., 2005, 281, 510.

35 W. Li, C. Liu, Y. Zhou, Y. Bai, X. Feng, Z. Yang, L. Lu, X. Lu

and K. Y. Chan, J. Phys. Chem. C, 2008, 112, 20539.

36 H. Xu and L. Zhang, J. Phys. Chem. C, 2009, 113, 1785.

37 S. Chen, X. Yu, H. Zhang and W. Liu, J. Hazard. Mater.,

2010, 180, 735.38 J. Cao, B. Luo, H. Lin, B. Xu and S. Chen, Appl. Catal., B,

2012, 111–112, 288.39 J. Tauc, R. Grigorovici and A. Vancu, Phys. Status Solidi,

1966, 15, 627.40 C. G. Grandqvist, A. Azens, A. Hjelm, L. Kullman,

G. A. Niklasson, D. Rönnow, M. Strømme Mattsson,M. Veszelei and G. Vaivars, Sol. Energy, 1998, 63, 199.

41 S. H. Baeck, T. F. Jaramillo, C. Brändli and E. W. McFarland,

J. Comb. Chem., 2002, 4, 536.

42 P. Wang, B. Huang, X. Zhang, X. Qin, Y. Dai, H. Jin, J. Wei

and M. H. Whangbo, Chem. – Eur. J., 2008, 14, 10543.

CrystEngComm, 2014, 16, 7493–7501 | 7501