Effects of Calcining Temperature on Photocatalytic Activity of Fe-DopedSulfated Titania

Ying Yang* and Congxue Tian

College of Biological and Chemical Engineering, Panzhihua University, Panzhihua, China

Received 30 December 2011, accepted 1 April 2012, DOI: 10.1111/j.1751-1097.2012.01157.x

ABSTRACT

Using industrial titanyl sulfate as a raw material, Fe-doped

sulfated titania (FST) photocatalysts were prepared by using the

one-step thermal hydrolysis method and characterized using

XRD, SEM, TGA–DSC, FTIR, UV–Vis DRS and N2 adsorp-

tion–desorption techniques. The effects of calcining temperature

on the structure of the titania were investigated. The photocat-

alytic activity of the FST was evaluated using the photodegra-

dation of methylene blue and photooxidation of phenol in

aqueous solutions under UV and visible light irradiation,

respectively. The results evinced that Ti4+ is substituted by

Fe3+ in titania lattice and forms impurity level within the band

gap of titania, which consequently induces the visible light

absorption and visible-light-driven photocatalytic activity. The

synergistic effects of Fe-doping and sulfation are beneficial to the

efficient separation of the photogenerated carriers and also

improve the quantum efficiency of photocatalysis. In addition,

Bronsted acidity arisen from the strong inductive effect of sulfate

is also conducive to enhancing the photocatalytic performance of

FST. However, when the calcining temperature is higher than

800�C, sulfur species and surface hydroxyl groups decompose

and desorb from FST and the specific surface area decreases

sharply. Moreover, severe sintering and rutile phase formation

occur simultaneously. All these are detrimental to photocatalytic

activity of FST.

INTRODUCTION

With industrialization and population growth, the environ-

mental contamination caused by organic pollutants is becom-ing a preponderant problem all over the world, especially indeveloping countries. Textile dyes and other industrial

dyestuffs constitute one of the largest groups of organicpollutants. These dyes exhibit great biotoxicity and possiblemutagenic and carcinogenic effects but at very low concentra-

tion, and most of them cannot be treated efficiently and ⁄ oreconomically using conventional biological and physicalmethods such as adsorption, ultrafiltration and coagulation.

Fortunately, photocatalysis technology, which can operate atambient temperature and atmospheric pressure and is totallyclean, safe and efficient, and thus environmental-friendly, may

provide a promising green solution to address these urgentissues (1–5).

Titanium dioxide is one of the most stable, cheap, nontoxic,easily available and environmentally harmonious photocatal-

ysis materials known today that has attracted considerableattention in the field of photocatalysis and environmentalpollution remediation over the past several decades. Nonethe-

less, the relatively large intrinsic band gap of titania (3.2 eV forthe anatase), which can be activated only by UV light(k < 400 nm), has hampered its widespread application as

well as the utilization of solar energy, because only 3–5% ofthe solar spectrum has wavelengths shorter than 400 nm. Onthe other hand, the photogenerated electrons and holes can

easily recombine, which consequently induce lower quantumefficiency. Both reasons mentioned above have largelyrestricted their industrial applications. How to extend thespectral response of wide-band-gap semiconductor such as

titania from UV to the visible range and improve thephotocatalysis quantum efficiency of titania in recent yearshas become a hot topic in the photocatalysis field (5–8). It has

been reported that iron doping can effectively expand thephotoresponse range of TiO2 to a visible region and increasephotocatalytic activity. Iron doping can also greatly reduce the

formation energy of oxygen vacancy, which favors thedissociation adsorption H2O and the formation of surfacehydroxyl group (9–14). In addition, the competition amongmigration, capture and recombination of the photogenerated

carrier determines the photocatalytic quantum efficiency ofTiO2. As a result, the effective separation of photoinducedelectrons and holes can be achieved through different mech-

anisms and approaches, such as iron doping and sulfatedmodification (15–20). However, iron-doped and ⁄ or sulfatedmodification TiO2 photocatalysts are generally prepared using

more expensive raw materials or complicated procedures,which is not practical for industrial-scale production.

In this article, using low-cost industrial titanyl sulfate

solution as a raw material, which contains abundant iron andsulfate, Fe-doped sulfated titania (FST) photocatalysts wereprepared and characterized using X-ray diffraction (XRD),scanning electron microscopy (SEM), thermogravimetry anal-

ysis–differential scanning calorimeter (TGA–DSC), Fouriertransform-infrared spectroscopy (FTIR), UV–Vis diffusereflectance spectra (UV–Vis DRS) and N2 adsorption–desorp-

tion techniques. The effects of calcining temperature on thestructure and morphology of FST photocatalyst were inves-tigated. The photocatalytic activity of FST photocatalyst was

*Corresponding author email: yydggpzh@163.com (Ying Yang)� 2012 Wiley Periodicals, Inc.Photochemistry and Photobiology� 2012 The American Society of Photobiology 0031-8655/12

Photochemistry and Photobiology, 2012, 88: 816–823

816

evaluated through photodegradation of methylene blue (MB)and photooxidation of phenol in aqueous solution under UVand visible light irradiation, respectively.

MATERIALS AND METHODS

Catalyst preparation. Titanyl sulfate solution used in this workwas obtained from an industrial source, and the primary composi-tions are as follows: TiO2 of 175 g L)1, F = m(effective H2SO4) ⁄m(TiO2) = 1.77, m(Fe) ⁄m(TiO2) = 0.35. FST photocatalysts wereprepared through one-step thermal hydrolysis method using industrialtitanyl sulfate solution as the raw material. The more detailedprocedure can be found in Yang et al. (21,22). In a typical synthesis,150 mL TiOSO4 solution and 34.5 mL preadding water were pre-heated to 96 ± 1�C, respectively. The heated TiOSO4 solution wasdropped into the preadding water under stirring and reflux at afeeding speed of 8.55 mL min)1. After feeding off, the mixturesolution was heated to boiling point (called the first boiling point) at aheating rate of 0.82�C min)1. Heating and stirring were stoppedimmediately when the mixture reached a gray color (called gray point)and then were turned on again after 30 min, then the mixture washeated to the boiling point again (called the second boiling point).Maintaining slight boiling 2.5 h after the second boiling point, thehydrolysis process is then finished. The slurry was promptly cooled toroom temperature, then filtered and washed. The as-preparedmetatitanic acid was dried at 80�C for 24 h, then calcined at differenttemperatures for 1.5 h in static air, and FST photocatalysts were thenobtained, denoted as FST(t), where t is the calcining temperature(�C).

Catalyst characterization. The structures of the FST samples wereconfirmed by both XRD (X¢Pertpro MPD with Cu Ka radiation) andFTIR (Thermo Electron Corporation, NICOLET 380, using KBrflakes). The average crystallite sizes were calculated from the (101)plane of titania according to Scherrer’s equation, and the weightfraction of rutile phase (WR) was calculated as (23):

WR ¼IR

0:884IA þ IRð1Þ

where IA and IR represent the integrated intensity of the anatase (101)main peak and the rutile (110) main peak, respectively. The surfacemorphology and the particle size were observed through the SEM(JSM-5900LV). The TGA–DSC was performed using a synchronouscomprehensive thermal analyzer (NETZSCH STA449C, Germany),under a N2 flow of 20 mL min)1 with a heating rate of 10�C min)1

from room temperature to 900�C. The UV–visible light absorptionspectra of the samples were recorded on a Hitachi U-4100 spectro-photometer equipped with integrator sphere, using BaSO4 as a refer-ence. Nitrogen adsorption–desorption isotherms were recorded on aMicromeritics ASAP2020M instrument. The S and Fe contents weredetermined by the chemical analysis method, and the Fe contents inthe surface layer were measured by X-ray photoelectron spectroscopytechnique.

Photocatalytic activity. Both photodegradation of MB and photo-oxidation of phenol were carried out at ambient temperature underUV and visible light irradiation, respectively. The UV light source is acommercial 8 W black-light tube with a spectral peak at around365 nm. The visible light source is produced by two 100 W tungsten-halogen lamps with any irradiation below 400 nm removed by UVcutoff filter (UBG-420; Accute Optical Technology Co. Ltd., China).The photocatalytic reactors used for UV and visible light irradiationare represented in Fig. 1a,b, respectively. Typical procedures ofphotocatalytic reactions are as follows: 200 mL MB solution(6 mg L)1) or phenol solution (6 mg L)1) and 0.2 g photocatalystwere fed into the reactor, and then the air was bubbled to the reactor at4 L min)1 rate from the bottom of reactor through a gas distributor.After 30 min of premixing, the irradiation was carried out and thereaction was completed after irradiating for 60 min. Solution samples(10 mL) were taken out from the reactor before irradiation andreaction finishing, and then centrifuged to separate the supernate andthe photocatalysts. The photocatalytic activity of FST samples forphotodegradation of MB was calculated by the difference between theinitial and the final optical absorbances of MB supernate measured at

666 nm wavelength using a spectrophotometer. However, the photo-catalytic activity of FST samples for photooxidation of phenol wasdetermined by the difference between the initial and the finalconcentrations of phenol supernate measured following the NationalEnvironmental Protection Standard of China (HJ 502-2009).

RESULTS AND DISCUSSION

XRD and SEM analysis

Figure 2 shows the XRD patterns of the FST samples calcined

at different temperatures for 1.5 h. It can be seen fromFig. 2a–f that all the samples calcined below 800�C areexclusively in the anatase phase (JPCDS 21-1272). At the

same time, all diffraction peaks assigned to anatase phasebecome sharper and stronger as the calcining temperatureincreases from 300 to 800�C, indicating the enhancement of

crystallization of titania. When the calcination temperaturefurther increases to 900�C, an obvious peak at 2h = 27.4�corresponding to the (110) plane diffraction of rutile (JPCDS21-1276) appears (as shown in Fig. 2g). In addition, according

to the calculation of Eq. (1), the weight fraction of rutileamounts to 93% for FST(900), suggesting that most of theanatase phase have been turned to rutile phase. Therefore, it is

reasonable to deduce that the phase transformation fromanatase to rutile occurs at ca 850�C. Although many factorscan affect the onset of phase transition temperature of titania

from anatase to rutile, it is commonly believed that the pure

(a) (b)Figure 1. Schematic diagram of photocatalytic reactor for UV light (a)and for visible light (b). 1. Air pipe; 2. UV light source; 3. Aluminumfoil; 4. Quartz glass reactor; 5. Gas distributor; 6. UV cutoff filter;7. Visible light source; 8. Glass reactor.

20 30 40 50 60 70

rutile(g)

(f)

(e)

(d)

(c)

(b)Inte

nsity

(a. u

.)

2 Theta (degree)

(a)

anatase

Figure 2. XRD patterns of FST samples calcined at different temper-atures: (a) 300�C; (b) 400�C; (c) 500�C; (d) 600�C; (e) 700�C; (f) 800�C;(g) 900�C.

Photochemistry and Photobiology, 2012, 88 817

anatase phase titania changes into the rutile phase at 550–650�C (24–26). So the FST samples show superior thermalstability (up to ca 850�C), which may be attributed to theretarding effects of Fe-doping and sulfation on the phase

transformation of titania from anatase to rutile (12,18,27). It isinteresting to note that no significant characteristic peak ofiron-containing compounds, such as Fe2O3 or FexTiOy, can be

found in the XRD patterns of all FST samples calcined attemperatures from 300 to 900�C. One possible reason respon-sible for this is that all the iron ions, existing in trivalent ionic

state in FST samples, may be inserted into the structure oftitania, and occupy some of the titanium lattice sites, formingan iron–titanium oxide solid solution (22).

For the FST samples calcined below 800�C, the latticeparameters (a, c) and interplanar spacing (d101 and d200) werecalculated from (101) and (200) planes of anatase titania usingthe following two equations:

Bragg0s law: dðhklÞ ¼k

2 sin hð2Þ

1

d2ðhklÞ¼ h2 þ k2

a2þ l2

c2ð3Þ

where d(hkl) is the distance between crystal planes of (hkl), k isthe used X-ray wavelength, h is the diffraction angle of crystal

plane (hkl) and a and c are lattice parameters. The averagecrystallite sizes of anatase FST samples (L101) were calculatedaccording to Scherrer’s equation. All the results together with

the standard data of JCPDS 21-1272 are given in Table 1. Ascan be seen from Table 1, there is an obvious increase in thelattice c parameter and interplanar spacing of the anatase FSTsamples with the calcination temperature increases, whereas

the a parameter remains almost constant during calcination.So this evolution produces a progressive increasing on thetetragonality (c ⁄ a) as the temperature increases (28). However,

the lattice parameters (a, c) and interplanar spacing (d101, d200)both are smaller than that of the standard data reported in theJCPDS 21-1272. These results further indicate that almost all

of Fe3+ enter titania lattice and substitute Ti4+ through thecalcining process. However, the ionic radius of Fe3+ (0.64 A)is smaller than that of Ti4+ (0.68 A). Both the small sizesubstitution and oxygen vacancy formation contribute to the

decrease in the lattice parameters and interplanar spacing ofthe anatase FST samples. Meanwhile, the lattice substitutionof Fe3+ for Ti4+ causes a lattice distortion and accumulates

strain energy and then hampers the A–R transformation.However, the full width at half maximum (b101) and the

average crystallite sizes (L101) steadily decrease and mono-tonically increase, respectively, as the calcination temperatureincreases, which are shown in Fig. 2 and Table 1.

Figure 3 shows the SEM images of FST samples calcined at

500, 700 and 900�C, respectively. As can be seen from Fig. 3a,the anatase FST(500) sample consists of well-defined sphere-shaped structures with the diameter of about 1 lm. As the

calcining temperature increases, the FST(700) sample exhibitsa slight agglomeration between particles (Fig. 3b). The SEMimage (Fig. 3c) shows that the FST(900) sample is composed

of irregular particles due to the sintering of the sample,resulting in less specific surface area and pore volume (asshown in Table 2). Similar results have been reported by Lv

et al. (29). Based on the morphology information derived fromFig. 3, the thermal stability of FST samples against the phasetransformation from anatase to rutile is estimated to be higherthan 800�C, which is in agreement with the results of XRD

analysis (Fig. 2 and Table 1).

FTIR analysis

The FTIR spectra of FST samples calcined at different

temperatures are depicted in Fig. 4. A broad band between3500 and 3000 cm)1, centered at around 3420 cm)1, can befound for all FST samples, indicating the presence of hydroxyl

groups on the surface of FST samples. As the calciningtemperature increases, the band gets flatter with less intensity,showing the removal of hydroxyl groups from the surface ofFST samples. The peak at 1637 cm)1 should be assigned to the

H–O–H bending vibration of molecularly (physically)adsorbed water and its intensity was lowered with increasingthe calcination temperature. The peaks at 998 and 1050 cm)1

are associated with the symmetric and asymmetric S–Ostretching vibrations, respectively. Meanwhile, the peak cor-responding to 1139 cm)1 appears due to symmetric stretching

of S=O vibration (30,31). However, no obvious absorptionpeak in the range of 900–990 cm)1 was observed in all FSTsamples, suggesting that most sulfates are incorporated into

the titania bulk, i.e. Ti–O–Ti network and covalently boundedto titania in chelating or bridged bidentate mode (22,32,33). Inthis case, the oxygen atoms are in electron-deficient state andthe surface acidity, especially Bronsted site acid, increase

significantly. The former factor can effectively prevent therecombination of photoinduced electron–hole pairs and thenenhance the photocatalytic quantum efficiencies. However, the

latter one can remarkably improve the photocatalytic activityof FST samples (34). In addition, when the calcinationtemperature is higher than 800�C, all the peaks related to

Table 1. XRD analysis results of FST samples calcined at different temperatures.

Samples d(101) (nm) d(200) (nm) a (nm) c (nm) c ⁄ a b101 L(101) (nm)

JCPDS 21-1272 0.3520 0.1892 0.3785 0.9514 2.5134 — —FST(300) 0.3428 0.1869 0.3738 0.8593 2.2988 0.0114 12.34FST(400) 0.3448 0.1874 0.3746 0.8755 2.3327 0.0103 13.59FST(500) 0.3450 0.1877 0.3747 0.8790 2.3448 0.0069 20.23FST(600) 0.3451 0.1878 0.3749 0.8869 2.3678 0.0046 30.54FST(700) 0.3453 0.1880 0.3753 0.8895 2.3738 0.0032 43.63FST(800) 0.3483 0.1889 0.3757 0.9134 2.4245 0.0018 77.06

818 Ying Yang and Congxue Tian

sulfur-containing species have faded away, which is consistentwith the results of XRD and SEM analysis and is furtherconfirmed by TGA–DSC analysis.

TGA–DSC analysis

TGA and DSC curves for H2TiO3 used as precursors for FSTsamples are shown in Fig. 5. It is noteworthy that the TGAcurve shows a total mass loss of ca 15.77 wt% in the range

from room temperature to 500�C, which corresponds to theobvious endothermic peak in DSC curve. This mass lossshould be ascribed to the removal of physically adsorbed water

and ⁄ or surface hydroxyl group, which occurs before the startof the loss of the sulfate. When the temperature increases from500 to 800�C, the other mass loss of ca 6.07 wt% can be found

in the TGA curve, which may arise from sulfate decompositionand then removal (Table 2). This resulted in a sulfate-freesurface at 800�C, which is in good accordance with the resultsof FTIR analysis. Hereafter, calcination will lead to a fast

sintering and the phase transformation of titania from anataseto rutile, corresponding to a small endothermic peak, centered

at about 834�C in the DSC curve, which is in line with theresults of XRD and SEM analysis.

UV–Vis DRS analysis

Figure 6 depicts the UV–visible light absorption spectra of the

pure anatase titania, FST(500) and FST(600) samples. Allspectra have a shoulder centered at ca 360 nm, suggesting anincrease in the absorption at wavelength shorter than 390 nm.

This spectral behavior can be assigned to the intrinsicabsorption of titania for the electrons’ excitation from valenceband to conduction band of titania. Moreover, the diffuse

reflectance spectra of FST samples show a redshift andincreased absorbance in the visible range compared with thatof pure anatase titania. This can be attributed to the charge-

transfer transition between the iron ion d electrons and thetitania conduction or valence band (35,36). The other reasonaccounting for the redshift in FST samples is the d–d transitionof Fe3+ (2T2g fi 2A2g,

2T1g) or charge-transfer transition

between iron ions (Fe3+ + Fe3+ fi Fe2+ + Fe4+) (21).As shown in Fig. 6, the absorption thresholds (kg) are 390,

Figure 3. SEM photographs of (a) FST(500), (b) FST(700) and (c) FST(900) samples.

Table 2. N2-adsorption results, S and Fe contents of FST samples calcined at different temperatures.

Samples SBET (m2 g)1) Pore volume (mL g)1) Pore diameter (nm) S content (wt%) Fe content (wt%)

FST(300) 229.7 0.3293 5.233 3.35 0.211FST(400) 206.8 0.4256 5.969 3.02 0.223FST(500) 180.7 0.4776 12.21 1.86 0.249FST(600) 128.5 0.5793 24.84 0.98 0.259FST(700) 44.1 0.2665 28.17 0.44 0.265FST(800) 12.6 0.1656 20.15 0.07 0.267FST(900) 3.81 0.0199 20.05 0.00 0.267

Photochemistry and Photobiology, 2012, 88 819

449 and 469 nm, and corresponding band gap energies (Eg)calculated using the formula (Eg [eV] = 1240 ⁄ kg [nm]) are3.18, 2.76 and 2.64 eV for pure anatase titania, FST(600) andFST(500), respectively (37,38). These results confirm that

Fe-doping can effectively lower the apparent band gap energyof titania, in contrast to the pure anatase titania. It has beenreported that metal doping could introduce additional impu-rity level into the band gap of titania. The electronic

transitions from the valence band to impurity level or fromimpurity level to conduction band can effectively reduce theapparent band gap and redshift of the band edge adsorption

threshold (39). Thus, the utility range of light of FST samplesis extended, which may remarkably improve the photocatalyticactivity of FST samples under visible light irradiation.

Nitrogen adsorption–desorption analysis and S and Fe content

Figures 7 and 8 provide the nitrogen adsorption–desorptionisotherms and the corresponding pore-size distribution curves

of FST samples calcined at different temperatures, respec-tively. All three samples in Fig. 7 display isotherms of Type-IVand hysteresis loops of Type H1 according to the IUPACclassification, which is the representative of mesoporous

structure. However, with the calcining temperature increase,

200 400 600 80075

80

85

90

95

100

TGADSC

Temperature

Mas

s fra

ctio

n%

-0.4

-0.2

0.0

0.2

0.4

dH.dt -1

W.g

- 1

Figure 5. TGA–DSC curves for H2TiO3 dried at 80�C for 24 h.

300 400 500 600 700 800

0.0

0.2

0.4

0.6

0.8

1.0

(c)

(b)

Abso

rban

ce

Wavelength (nm)

Pure Titania (a) FST(500) (b) FST(600) (c)

(a)

Figure 6. UV–visible light absorption spectra of pure titania,FST(500) and FST(600) samples.

0.0 0.2 0.4 0.6 0.8 1.00

40

80

120

160

200

Vol

ume

Ads

orbe

d m

l.g-1

Relative Pressure P/P0

FST(500) FST(600) FST(700)

Figure 7. N2 adsorption–desorption isotherms of FST samples cal-cined at different temperatures.

0 30 60 90 120

0.00

0.01

0.02

0.03

0.04

FST(500) FST(600) FST(700)

dV/d

D

ml.g

-1.n

m-1

Pore diameter (nm)

Figure 8. The BJH pore-size distribution curves of FST samplescalcined at different temperatures.

4000 3500 3000 2500 2000 1500 1000 500

(e)

(d)

(c)

(b)Tran

smitt

ance

a.

u.

Wavenumber cm-1

(a)

Figure 4. FTIR spectra of FST samples calcined at differenttemperatures: (a) 400�C; (b) 500�C; (c) 600�C; (d) 700�C; (e) 800�C.

820 Ying Yang and Congxue Tian

the hysteresis loop gradually shifts toward the high pressureside, indicative of an increase in pore size, as presented inTable 2. Narrow pore-size distribution of all samples shown inFig. 8 shows good uniformity of channel, which may be due to

the orientation, stability and support roles of sulfate formesoporous structure in the calcination process (22). Thespecific surface area (SBET), pore volume and pore diameter of

the FST samples calcined at temperatures from 300 to 900�Care presented in Table 2. It can be seen from the values thatthere is a monotonous decrease in specific surface area with an

increase in calcining temperature. The pore volume and thepore diameter both were enhanced as the calcination temper-ature increases from 300 to 600�C, and then decreased with the

calcining temperature further increasing from 600 to 900�C.The former may be attributed to the improvement of thecrystallization degree of FST samples and thus this enhancesthe density of the pore structure. In addition, calcining in this

range of temperature will induce the pore restructuring;namely, several small pores can be merged into a large one,which is another possible reason. As shown in Table 2, the S

content is almost constant for the samples calcined under500�C. However, when calcined at temperatures from 500 to900�C, sulfur-containing species will be decomposed and

desorbed from FST samples, resulting in the decrease of Scontent and the consequent severe sintering and the collapse ofthe pores. All of the specific surface area, pore volume andpore diameter of the FST samples calcined at a temperature

higher than 600�C decrease evidently. This agrees well with theresults of XRD, SEM and FTIR mentioned above. Inaddition, it can also be seen from Table 2 that the Fe contents

gradually increase with the rise in calcination temperaturefrom 300 to 800�C, which may be caused by the S speciesand ⁄ or H2O(OH) desorption from FST. While the total

average Fe contents determined by chemical analysis methodare 0.249 and 0.259 wt%, for the FST(500) and FST(600)samples, respectively, which are approximately equal to those

Fe contents in the surface layer determined by using X-rayphotoelectron spectroscopy technique (0.250 wt% forFST(500) and 0.261 wt% for FST(600)). This may furtherprove that almost all iron ions were inserted into the bulk of

titania and formed homogenous Fe-doping.

Photocatalytic activity

The photocatalytic activity of the FST samples calcined at

different temperatures was evaluated by the photodegradationof MB and photooxidation of phenol at ambient temperatureunder UV and visible light irradiation, respectively. As shown

in Fig. 9, it can be seen that the photocatalytic activity undervisible light irradiation shows a trend similar to that under UVlight irradiation for both photodegradation of MB andphotooxidation of phenol. The photocatalytic activity in-

creases with the increase of the calcining temperature, and thendecreases with an additional increase in calcination tempera-ture. The FST(500) and FST(600) samples have the maximal

photocatalytic activity under both UV and visible lightirradiation, for photodegradation of MB and photooxidationof phenol, respectively, which can be attributed to the larger

BET surface area and the improvement of crystallization ofFST(500) and FST(600) samples. For comparison, the photo-catalytic activity of pure anatase titania was also performed

under the same conditions and are 68.49% and 45.31% under

UV, and 1.38% and 0.86% under visible light, for thephotodegradation of MB and photooxidation of phenol,respectively. The results under visible light are due mainly to

the adsorption of substrate (MB or phenol) onto the surface oftitania. This may affirm that the FST(500) and FST(600)samples have higher visible light photocatalytic activity, asexpected from the UV–Vis DRS analysis. However, the pure

anatase titania does not show the similar property. Thephotocatalytic efficiency of modified titania appears to be acomplex function of parameters of titania, such as the

crystalline phase, crystallinity, particle size, specific surfacearea and surface chemical state. However, these parametersusually conflict with each other to affect the photocatalytic

activity of titania. For example, as calcination temperatureincreases, the crystallization degree of FST samples increasesevidently, which is favorable to improve the photocatalytic

activity of FST samples. But when the calcination furtherincreases from 500 or 600 to 800�C, the photocatalytic activityunder both UV and visible light steadily decreases. This can beexplained by the decomposition and desorption of sulfur-

containing species and ⁄ or surface hydroxyl group of FSTsamples, and ⁄ or the decrease of BET surface area (as shown inTable 2). The difference in calcining temperature of maximal

photocatalytic activity for the photodegradation of MB andthe photooxidation of phenol may be ascribed to the differencein molecular structure and chargeability of MB cation and

phenoxy anion. It is worth noting that the photocatalyticactivity of FST(900) sharply decreases due to severe sinteringof the sample and the formation of rutile phase (Figs 2 and 3).It is usually considered that the thermally induced phase

transformation from anatase to rutile of titania is initiated inthe core of particles, whereas the anatase phase in the surfaceregion can remain at a relatively higher temperature (24,25,40).

Compared with rutile phase, anatase phase can contain moreadsorbed water and hydroxyl groups on the surface of titania,which is beneficial to improving the photocatalytic activity of

FST samples. Although the presence of rutile may improve theabsorption properties of FST samples, the effective holeconcentration at the anatase surface is lowered. Therefore,

the formation of rutile at 900�C for FST sample is detrimentalto the photodegradation of MB and photooxidation of phenol,and does not show the so-called mixed-phase effects (41,42).

300 400 500 600 700 800 9000

20

40

60

80

100

0

20

40

60

80

100

Phot

odeg

rada

tion

of M

B

%

Methylene Blue under UV light Methylene Blue under visible light

Calcining temperature

Photooxidation of Phenol %

Phenol under UV light Phenol under visible light

Figure 9. Photocatalytic activity of FST samples calcined at differenttemperatures.

Photochemistry and Photobiology, 2012, 88 821

CONCLUSIONS

Using industrial titanyl sulfate as the raw material, FSTphotocatalysts with high photocatalytic activity under UV andvisible light irradiation were prepared by one-step thermal

hydrolysis method. The results indicated that Ti4+ is substi-tuted by Fe3+ in titania lattice and then forms impurity levelwithin the band gap of titania, which consequently induces the

visible light absorption and visible-light-driven photocatalyticactivity for FST samples. The synergistic effects of Fe-dopingand sulfation can effectively hinder the crystal growth andphase transformation from anatase to rutile of titania.

Therefore, the FST samples show superior thermal stability(up to 850�C). In addition, Bronsted acidity arisen from thestrong inductive effect of sulfate is also conductive to

enhancing the photocatalytic performance of FST samples.With the increase of the calcination temperature, the crystal-lization degree of FST samples increases evidently, which is

favorable to improve the photocatalytic activity of FSTsamples. But, as the calcining temperature further increases,sulfur species and surface hydroxyl groups will decompose and

desorb from FST samples and the specific surface area willsteadily decrease. Moreover, calcining at a temperature higherthan 800�C will give rise to severe sintering and the formationof rutile phase. All these are detrimental to photocatalytic

activities of FST photocatalysts. The optimal calcinationtemperatures for photodegradation of MB and photooxida-tion of phenol under both UV and visible light irradiation are

500 and 600�C, respectively.

Acknowledgements—This work was supported by the National Nat-

ural Science Foundation of China (50804025), the Applied and Basic

Research Program of Sichuan Province, China (2008JY0140), the

Youths Foundation of Sichuan Province, China (09ZQ026-067), the

Talents Innovation Project of Panzhihua City, China (2009TX-5(1)),

and the Promoting Industrialization Program of Panzhihua City,

China (2011CY-G-23).

REFERENCES1. Hoffmann, M. R., S. T. Martin, W. Choi and D. W. Bahnemann

(1995) Environmental applications of semiconductor photocatal-ysis. Chem. Rev. 95, 69–96.

2. Chen, X. and S. S. Mao (2007) Titanium dioxide nanomaterials:Synthesis, properties, modifications, and applications. Chem. Rev.107, 2891–2959.

3. Diebold, U. (2003) The surface science of titanium dioxide. Surf.Sci. Rep. 48, 53–229.

4. Fujishima, A., X. Zhang and D. A. Tryk (2008) TiO2 photoca-talysis and related surface phenomena. Surf. Sci. Rep. 63, 515–582.

5. Anpo, M. and P. V. Kamat (2010) Environmentally BenignPhotocatalysts: Applications of Titanium Oxide-based Materials.Springer, New York.

6. Linsebigler, A. L., G. Lu and J. T. Yates (1995) Photocatalysis onTiO2 surfaces: Principles, mechanisms, and selected results. Chem.Rev. 95, 735–758.

7. Anpo, M. and M. Takeuchi (2003) The design and development ofhighly reactive titanium oxide photocatalysts operating undervisible light irradiation. J. Catal. 216, 505–516.

8. Rehman, S., R. Ullah, A. M. Butt and N. D. Gohar (2009)Strategies of making TiO2 and ZnO visible light active. J. Hazard.Mater. 170, 560–569.

9. Liu, Z., Y. Wang, W. Chu, Z. Li and C. Ge (2010) Characteristicsof doped TiO2 photocatalysts for the degradation of methyleneblue waste water under visible light. J. Alloys Compd. 501, 54–59.

10. Tu, Y. F., S. Y. Huang, J. P. Sang and X. W. Zou (2010) Prep-aration of Fe-doped TiO2 nanotube arrays and their photocata-lytic activities under visible light. Mater. Res. Bull. 45, 224–229.

11. Roldan, A., M. Boronat, A. Corma and F. Illas (2010) Theoreticalconfirmation of the enhanced facility to increase oxygen vacancyconcentration in TiO2 by iron doping. J. Phys. Chem. C 114, 6511–6517.

12. Yu, J., Q. Xiang and M. Zhou (2009) Preparation, characteriza-tion and visible-light-driven photocatalytic activity of Fe-dopedtitania nanorods and first-principles study for electronic struc-tures. Appl. Catal. B 90, 595–602.

13. Wang, X. H., J. G. Li, H. Kamiyama, M. Katada, N. Ohashi,Y. Moriyoshi and T. Ishigaki (2005) Pyrogenic iron(III)-dopedTiO2 nanopowders synthesized in RF thermal plasma: Phaseformation, defect structure, band gap, and magnetic properties.J. Am. Chem. Soc. 127, 10982–10990.

14. Schaub, R., P. Thostrup, N. Lopez, E. Lægsgaard, I. Stensgaard,J. K. Nørskov and F. Besenbacher (2001) Oxygen vacancies asactive sites for water dissociation on rutile TiO2 (110). Phys. Rev.Lett. 87, 266104–266107.

15. Thompson, T. L. and J. T. Yates (2006) Surface science studies ofthe photoactivation of TiO2-new photochemical processes. Chem.Rev. 106, 4428–4453.

16. Liu, Y., J. Liu, Y. Lin, Y. Zhang and Y. Wei (2009) Simplefabrication and photocatalytic activity of S-doped TiO2 underlow power LED visible light irradiation. Ceram. Int. 35,3061–3065.

17. Antony Raj, K. J. and B. Viswanathan (2009) Single-step synthesisand structural study of mesoporous sulfated titania nanopowderby a controlled hydrolysis process. ACS Appl. Mater. Interfaces 1,2462–2469.

18. Parida, K. M., N. Sahu, N. R. Biswal, B. Naik and A. C. Pradhan(2008) Preparation, characterization, and photocatalytic activityof sulfate-modified titania for degradation of methyl orange undervisible light. J. Colloid Interface Sci. 318, 231–237.

19. Barraud, E., F. Bosc, D. Edwards, N. Keller and V. Keller (2005)Gas phase photocatalytic removal of toluene effluents on sulfatedtitania. J. Catal. 235, 318–326.

20. Srinivasan, S. S., J. Wade, E. K. Stefanakos and Y. Goswami(2006) Synergistic effects of sulfation and co-doping on the visiblelight photocatalysis of TiO2. J. Alloys Compd. 424, 322–326.

21. Yang, Y. and C. Tian (2010) Synergistic effects of sulfation andFe-doping on the photocatalysis of titania. Res. Chem. Intermed.36, 889–895.

22. Yang, Y., H. Zhong, C. Tian and Z. Jiang (2011) Single-steppreparation, characterization and photocatalytic mechanism ofmesoporous Fe-doped sulfated titania. Sur. Sci. 605, 1281–1286.

23. Zhang, H. and J. F. Banfield (2000) Understanding polymorphicphase transformation behavior during growth of nanocrystallineaggregates: Insights from TiO2. J. Phys. Chem. B 104, 3481–3487.

24. Zhang, J., M. Li, Z. Feng, J. Chen and C. Li (2006) UV Ramanspectroscopic study on TiO2. I. Phase transformation at the sur-face and in the bulk. J. Phys. Chem. B 110, 927–935.

25. Zhang, J., Q. Xu, M. Li, Z. Feng and C. Li (2009) UV Ramanspectroscopic study on TiO2. II. Effect of nanoparticle size on theouter ⁄ inner phase transformations. J. Phys. Chem. C 113,1698–1704.

26. Liu, L., J. Chan and T.-K. Sham (2010) Calcination-inducedphase transformation and accompanying optical luminescence ofTiO2 nanotubes: An X-ray absorption near-edge structures andX-ray excited optical luminescence study. J. Phys. Chem. C 114,21353–21359.

27. Wu, Y., J. Zhang, L. Xiao and F. Chen (2009) Preparation andcharacterization of TiO2 photocatalysts by Fe3+ doping togetherwith Au deposition for the degradation of organic pollutants.Appl. Catal. B 88, 525–532.

28. Colon, G., M. C. Hidalgo, G. Munuera, I. Ferino, M. G.Cutrufello and J. A. Navıo (2006) Structural and surface approachto the enhanced photocatalytic activity of sulfated TiO2 photo-catalyst. Appl. Catal. B 63, 45–59.

29. Lv, K., Q. Xiang and J. Yu (2011) Effect of calcination temperatureon morphology and photocatalytic activity of anatase TiO2 nano-sheets with exposed {0 0 1} facets. Appl. Catal. B 104, 275–281.

822 Ying Yang and Congxue Tian

30. Ropero-Vega, J. L., A. Aldana-Perez, R. Gomez and M. E. Nino-Gomez (2010) Sulfated titania [TiO2 ⁄ SO4

2)]: A very active solidacid catalyst for the esterification of free fatty acids with ethanol.Appl. Catal. A 379, 24–29.

31. Yamaguchi, T., T. Jin, T. Ishida and K. Tanabe (1987) Structuralidentification of acid sites of sulfur-promoted solid super acid andconstruction of its structure on silica support. Mater. Chem. Phys.17, 3–19.

32. Li, H., G. Li, J. Zhu and Y. Wan (2005) Preparation of an activeSO4

2) ⁄TiO2 photocatalyst for phenol degradation under super-critical conditions. Appl. Catal. A 226, 93–100.

33. Jin, T., M. Machida, T. Yamaguchi and K. Tanabe (1984)Infrared study of sulfur-containing iron oxide. Behavior of sulfurduring reduction and oxidation. Inorg. Chem. 23, 4396–4398.

34. Wang, Q., M. Zhang, C. Chen, W. Ma and J. Zhao (2010)Photocatalytic aerobic oxidation of alcohols on TiO2: Theacceleration effect of a Brønsted acid. Angew. Chem. Int. Ed. 49,7976–7979.

35. Cong, Y., J. Zhang, F. Chen, M. Anpo and D. He (2007) Prep-aration, photocatalytic activity, and mechanism of nano-TiO2

co-doped with nitrogen and iron (III). J. Phys. Chem. C 111,10618–10623.

36. Yalcln, Y., M. Klllc and Z. Clnar (2010) Fe+3-doped TiO2: Acombined experimental and computational approach to the eval-uation of visible light activity. Appl. Catal. B 99, 469–477.

37. Joshi, M. M., N. K. Labhsetwar, P. A. Mangruldar, S. N. Tijare,S. P. Kamble and S. S. Rayalu (2009) Visible light inducedphotoreduction of methyl orange by N-doped mesoporous titania.Appl. Catal. A 357, 26–33.

38. Khan, S. U. M., M. Al-Shahry and W. B. Ingler Jr. (2002)Efficient photochemical water splitting by a chemically modifiedn-TiO2. Science 297, 2243–2245.

39. Yang, Y., H. Zhong and C. Tian (2011) Photocatalytic mecha-nisms of modified titania under visible light. Res. Chem. Intermed.37, 91–102.

40. Carneiro, J. T., T. J. Savenije, J. A. Moulijn and G. Mul (2010)Toward a physically sound structure—Activity relationship ofTiO2-based photocatalysts. J. Phys. Chem. C 114, 327–332.

41. Ding, Z., G. Q. Lu and P. F. Greenfield (2000) Role of the crys-tallite phase of TiO2 in heterogeneous photocatalysis for phenoloxidation in water. J. Phys. Chem. B 104, 4815–4820.

42. Carneiro, J. T., T. J. Savenije, J. A. Moulijn and G. Mul (2011)How phase composition influences optoelectronic and photocat-alytic properties of TiO2. J. Phys. Chem. C 115, 2211–2217.

Photochemistry and Photobiology, 2012, 88 823