Io

La

b

a

ARR1AA

KTSPA

1

Falra(tewbataceatpb

If

0d

Applied Surface Science 257 (2011) 4951–4955

Contents lists available at ScienceDirect

Applied Surface Science

journa l homepage: www.e lsev ier .com/ locate /apsusc

nfluence of ammonia on the morphologies and enhanced photocatalytic activityf TiO2 micro/nanospheres

i Cong-Jua,b,∗, Xu Guo-Ronga

College of Material Science and Engineering, Beijing Institute of Fashion Technology, Beijing 100029, PR ChinaBeijing Key Laboratory of Clothing Materials R&D and Assessment, Beijing 100029, PR China

r t i c l e i n f o

rticle history:eceived 28 October 2010eceived in revised form3 December 2010

a b s t r a c t

TiO2 micro/nanospheres were synthesized by a combination process contains hydrolysis of titaniumtetra-n-butyl in mixed solution of anhydrous ethanol/ammonia and the subsequent calcination under550 ◦C for 7 h. The pH values of the mixed solution were tuned to be 10.4, 11.0 and 11.6, respectively, byadding different amounts of ammonia. Scanning electron microscope (SEM) and transmission electron

ccepted 3 January 2011vailable online 5 January 2011

eywords:itanium dioxidepheres

microscope (TEM) were used to characterize the morphologies and the crystallinity. X-ray diffraction(XRD) patterns indicated that pH value of the precursors has an important effect on the crystal phasecomposition. UV–vis diffuse reflectance spectrum was applied to characterize the optical properties ofsamples. Degradation of methylene blue under the irradiation of 300 W Hg lamp confirmed the enhancedphotocatalytic activity of TiO2 micro/nanospheres. In addition, the formation mechanism was proposed.

hotocatalyticmmonia

. Introduction

TiO2 has attracted enormous research as the catalyst sinceujishima and Honda discovered the phenomenon of photocat-lytic splitting of water on a TiO2 electrode under ultraviolet (UV)ight in 1972 [1]. Till now, the mechanism of the photocatalyticeactions has been generally confirmed. When TiO2 was irradi-ted by the light with energy equal to or higher than its bandgap3.2 ev for anatase and 3.0 ev for rutile), the absorption of pho-ons leads to a charge separation due to the promotion of thelectrons (e−) from the valence band to the conduction band,ith the simultaneous generation of a hole (h+) in the valence

and. The photogenerated electrons and holes combine with thedsorbed chemical groups to form some radicals with strong oxida-ive activity, such as hydroxyl radicals (•OH), superoxides(O2

•−)nd hydroperoxyl radicals(HO2

•−) [2]. So most organic compoundsan be oxidized completely in the photocatalytic process [3]. How-ver, the recombination of photogenerated electron–hole pairs

nd the limited light source have blocked the development ofhe photocatalysts. Therefore, to prevent the recombination ofhotogenerated electron–hole pairs was the main effort that haseen done in recent decades. Metal-semiconductor system [4],

∗ Corresponding author at: College of Material Science and Engineering, Beijingnstitute of Fashion Technology, Beijing 100029, PR China. Tel.: +86 010 6428 8192;ax: +86 010 6428 8192.

E-mail address: congjuli@gmail.com (C.-J. Li).

169-4332/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.apsusc.2011.01.002

© 2011 Elsevier B.V. All rights reserved.

composite semiconductor [5] and transition metal doping [6]have been used to inhibit the recombination by increasing thecharge separation and therefore the efficiency of the photocatalyticprocess.

Besides the modifications mentioned above, another methodto improve the photocatalytic activity was to enlarge the surfacearea. It has been well known that the photodegradation is a sur-face reaction. So the properties of the surface have crucial effect onthe degradation efficiency. Large surface areas could facilitate thereaction/interaction between the photocatalyst and the interactingmolecules, thus increasing the degradation efficiency. Surface areahas a crucial correlation with surface morphologies. In addition, thesurface morphologies could influence the electron–hole recombi-nation. So many photocatalysts with unique morphologies, such asspherical morphologies, have been investigated [7,8].

Many synthesis routes have been applied to fabricate TiO2spherical materials [9,10]. The most common used one wastemplate method [11]. It was reported that spherical morpholo-gies could be obtained by adding ammonia into agglomeratesof titanium hydroxide after the hydrolysis of titanium tetra-n-butoxide in anhydrous ethanol [12]. In the current experiment,titanium tetra-n-butoxide was hydrolyzed directly in mixedsolution of anhydrous ethanol and ammonia. The subsequent

calcination results in the formation of TiO2 micro/nanospheres.The mechanism was proposed. Degradation of methylene blueunder irradiation of 300 W Hg lamp indicated that TiO2micro/nanospherical structures showed enhanced photocatalyticactivity.

4 face Science 257 (2011) 4951–4955

2

2

jagw

dtu2t

2

ete(dasr

2

ectdb

F(

952 C.-J. Li, G.-R. Xu / Applied Sur

. Experimental

.1. Materials and preparation of TiO2 micro/nanospheres

Titanium tetra-n-butoxide (Ti(OBu)4) was purchased from Bei-ing Xingjin Fine Chemicals. LTD. Anhydrous ethanol (CH3CH2OH)nd ammonia solution (25%) were purchased from Beijing Tong-uang Fine Chemicals. LTD. All the materials were used as receivedithout any purification.

1 ml Ti(OBu)4 was dropped into 20 ml mixed solution of anhy-rous ethanol and ammonia. The pH values of the solution wereuned by adding different amounts of ammonia. Wait a few min-tes for the hydrolyzation and put the precipitation under 60 ◦C for4 h, finally anneal the precipitations under 550 ◦C for 7 h to obtainhe samples.

.2. Characterization and measurements

The morphologies of samples were examined by scanninglectron microscopy (SEM: JSM-6360LV, JAPAN). The deep inves-igation of the obtained samples was examined by transmissionlectron microscopy (TEM: Tecnai G2 20S-TWIN). X-ray diffractionXRD) patterns were collected on a Rigaku-D/max 40,000 V X-rayiffractometer equipped with Cu K� radiation (� = 0.15418 nm) atstep width of 5◦/min. The UV–vis absorption spectrum were mea-ured on a TU-1901 UV–vis spectrophotometer in the wavelengthange of 190–800 nm.

.3. Photocatalytic experiments

The photodegradation of methylene blue in water was used to

valuate the photoactivity of the samples as prepared, which isonsidered as one of the most standard method for the evalua-ion of photo-oxidation activity. In our experiment, a consistentosage of 150 mg samples were suspended in 400 ml methylenelue (MB) solutions with the initial concentration of 20 mg/L. The

ig. 2. Low and high magnified SEM images of TiO2 micro/nano materials after the calcinad) pH = 11.6.

Fig. 1. XRD spectra of samples: (a) without adding ammonia, tuning pH to (b) 10.4,(c) 11.0 and (d) 11.6 by adding ammonia during hydrolysis of Titanium tetra-n-butylin mixed anhydrous ethanol/ammonia solution.

photodegradation experiment was carried out in the photochemi-cal reactor (BL-GHX-II, Shanghai Bilon Instrument Co. LTD). Beforethe irradiation, the solutions were agitated for 3 h to establish theadsorption/degradation equilibrium. The concentration of methy-

lene blue in solutions was monitored and analyzed by measuringthe absorbance at 663 nm wavelength using TU-1901 UV–vis spec-trometer (Beijing Purkinje General Instrument Co. LTD) at givenirradiation time intervals.

tion of the precursors: (a) without adding ammonia, (b) pH = 10.4, (c) pH = 11.0 and

C.-J. Li, G.-R. Xu / Applied Surface Science 257 (2011) 4951–4955 4953

/nano

3

3

doa5(45cbfifwaorwhp

Fig. 3. Low and high TEM magnification images of TiO2 micro

. Results and discussions

.1. XRD patterns of samples

Fig. 1 shows the XRD patterns of samples under different con-itions. In Fig. 1a, the sample was obtained by the hydrolysisf titanium tetra-n-butyl in anhydrous ethanol without addingmmonia. The typical peaks with 2� values of 25.50◦, 37.49◦, 48.26◦,4.08◦ and 55.26◦ were in accordance with the (1 0 1), (0 0 4), (2 0 0),1 0 5) and (2 1 1) crystal planes for anatase TiO2 [PDF card no 4-77]. The typical peaks with 2� values of 27.60◦, 36.26◦, 41.42◦ and4.50◦ were in accordance with the (1 1 0), (1 0 1), (1 1 1) and (2 1 1)rystal planes for rutile TiO2 [PDF card no 21-1276]. In Fig. 1b–d,efore the hydrolysis of titanium tetra-n-butyl, pH values wererst tuned to be 10.4, 11.0 and 11.6, respectively, by adding dif-

erent amounts of ammonia. As shown in Fig. 1b, when pH valueas 10.4, rutile crystalline still exists. When pH values were 11.0

nd 11.6, however, pure anatase TiO2 without any rutile phase was

btained, as shown in Fig. 1c and d. As indicated in many othereports [13], the phase composition of TiO2 was in close correlationith the annealing temperature and time during the preparation byydrothermal method. In our experiment, however, annealing tem-erature and annealing time were all the same, the only variable

spheres with precursor pH value of: (a, b) 11.0 and (c, d) 11.6.

factor was the pH values. So it can be concluded that the additionof ammonia to the precursors play a key role in the formation ofTiO2 with different crystallinities. In other literatures [14], it wasreported that the growth rate of the different crystal planes of TiO2nanoparticles can be tuned under different pH values by addingamines. So it was suspected that the addition of ammonia play thesame role during the formation of TiO2 crystallinity.

3.2. TiO2 micro/nanospheres morphologies characterization

Fig. 2 shows the morphologies of different samples. Seen fromFig. 2a, when Ti(OBu)4 was hydrolyzed in anhydrous ethanolwithout adding ammonia, the subsequent calcination of the precip-itations resulted in the formation of irregular TiO2 powders despiteoccurrence of some spherical structures. As for other samples,before the hydrolyzation, ammonia was used to tune the mixedsolution pH values. When Ti(OBu)4 was hydrolyzed in the mixedsolution with pH value of 10.4, the samples after the calcination

showed sphere-like morphology (Fig. 2b). The spheres were irreg-ular in shape with diameters varying from hundreds of nanometersto a few micrometers. As for the process in which Ti(OBu)4 washydrolyzed in the mixed solution with the pH value of 11.0, sam-ples obtained in same way showed comparatively well-distributed

4954 C.-J. Li, G.-R. Xu / Applied Surface Science 257 (2011) 4951–4955

ation

ThtmeW1fsw

aciToipactcca

3

tihsitup

3.4. Photocatalytic activity

The UV–vis diffuse reflectance spectrum for the synthesizedTiO2 samples under different pH values is presented in Fig. 5. For

Fig. 4. Schematic illustrations of form

iO2 micro/nanospherical morphologies (Fig. 2c). The TiO2 spheresad very smooth surface with diameters of hundreds of nanometerso approximate 2 �m. In addition, it should be noticed that the for-

ation of regular TiO2 spheres inhibited the agglomeration to somextent, which could be observed in low magnification SEM image.hen Ti(OBu)4 was hydrolyzed in mixed solution with pH value of

1.6 (Fig. 2d), the calcination of the precipitations resulted in theormation of TiO2 nanomaterials with spherical structures. And thepheres here had a smaller diameter, however, the agglomerationas more severe.

Transmission electron microscopy (TEM) is used to differenti-te the crystalline and amorphous material structures. Structuralharacterization of as-prepared TiO2 micro/nanospheres is shownn Fig. 3a–d. Fig. 3a and b shows the low and high magnificationEM images of TiO2 microspheres obtained with precursor pH valuef 11.0. Fig. 3c and d shows the low and high magnification TEMmages of TiO2 nanospherical materials obtained with precursorH value of 11.6. Compared with Fig. 3a, in Fig. 3c, more severegglomeration could be observed when pH value was 11.6, whichan also be confirmed by SEM images in Fig. 2. However, better crys-allinity could be confirmed by the distinct crystal planes in Fig. 3d,ompared with that in Fig. 3b. In Fig. 3d, the distance between twoonsecutive planes is the same and the atomic planes are uniformlyrranged in parallel, which indicates good crystallinity.

.3. The proposed mechanism

A probable mechanism was proposed to explain the forma-ion of TiO2 micro/nanospheres. When Ti(OBu)4 was hydrolyzednto Ti(OH)4, ammonia molecules could complex with Ti(OH)4 byydrogen bond. When Ti(OBu)4 was first hydrolyzed in butanol

olution and then ammonia was added, Sugimoto and Kojima [12]ndicated that ammonia could play three key roles: accelerator ofhe precipitation of the hydrolysis products; inhibitor of the coag-lation of the particles; promote for production of highly sphericalarticles. In our experiment, Ti(OBu)4 were added into the mixed

process of TiO2 micro/nanospheres.

solution of ammonia and anhydrous ethanol solution, thereforeduring the hydrolysis of Ti(OBu)4 process, the complexation ofammonia molecules to hydrolysis products of Ti(OH)4 by hydro-gen bond occurred in the meantime. So the competition betweenhydrolysis of Ti(OBu)4 and the complexation of ammonia moleculesto Ti(OH)4 could be concluded, which can be seen in Fig. 4. On theother hand, Ti(OH)4 could also be agglomerated into spherical par-ticles with many “−OH” on surface. So ammonia molecules couldbe attached onto the surface of the Ti(OH)4 spherical particles byhydrogen bond, inhibiting the agglomeration of the spherical par-ticles. Subsequent calcination results into the formation of TiO2micro/nanospheres. It can be summarized that more ammoniamolecules contributes to the stronger complexation, results intothe formation of spheres with smaller diameter.

Fig. 5. UV–vis diffuse reflectance spectrum of TiO2 samples obtained under differentconditions.

C.-J. Li, G.-R. Xu / Applied Surface Sc

Fo

cdIaitiiseAp4n

pasaatiCpd6(arS

[

[[

ig. 6. Methylene blue concentration changes as a function of time at the presencef as-synthesized TiO2 micro/nanospheres.

omparison, the spectrum of TiO2 particles obtained by the proce-ure without adding ammonia during hydrolysis was also plotted.

t could be apparently observed that in the UV region (190–400 nm),s the pH value of the precursors increased, the absorption abil-ty of the samples became stronger. In addition, a blue shift inhe band gap absorption edge for the samples as the pH valuencreased is observed, which could be observed more clearly in thenset in Fig. 5. The shift may be ascribed to the smaller particlesize due to the quantum size effect [15]. The UV–vis absorptiondge of all the samples was clearly shown in the inset in Fig. 5.s is displayed, the band gap absorption edges for the samplesossess the sequence of TiO2 (pH = 11.6, 405 nm) < TiO2 (pH = 11.0,06 nm) <TiO2 (pH = 10.4, 416 nm) <TiO2 (without adding ammo-ia, 424 nm).

Methylene blue (MB) was adopted as a representative organicollutant to evaluate the photocatalytic activity performance of thes-synthesized photocatalysts. The photocatalytic activities of allamples were shown in Fig. 6. The degradation efficiency is defineds C/C0, where C0 was the initial concentration after the equilibriumdsorption and C was the reaction concentration of MB, respec-ively. As seen in Fig. 6, there is no appreciable degradation of MBn the absence of TiO2 under the irradiation of UV lamp. However,/C0 decreased gradually as the exposure time was extended in theresence of different TiO2 samples. The degradation efficiency forifferent TiO2 samples in 90 min was, respectively to be 6% (blank),

0% (without ammonia), 70% (pH = 10.6), 93% (pH = 11.0) and 94%pH = 11.6), in accordance with the absorption ability in UV regions shown in Fig. 5. The stronger absorption ability in UV regionesulted in the higher photocatalytic activity. Combined with theEM images in Fig. 2c and d, the formation of micro/nanospherical

[

[[[

ience 257 (2011) 4951–4955 4955

structures resulted in the enhanced photocatalytic activity. It hasbeen reported that TiO2 composed of anatase and rutile crystallinityhas a higher photocatalytic activity due to the electron transfer inthe interfaces [16]. In our experiment, however, the conclusion wasdifferent, which can be seen from the XRD patterns (Fig. 1) anddegradation efficiency figure (Fig. 6). So it was believed that theenhanced photocatalytic activity may be ascribed to the large sur-face area due to the decreased particles size and the well dispersion.

4. Conclusions

In summary, Ti(OBu)4 was hydrolyzed in the mixed solu-tion of anhydrous ethanol and ammonia. Ammonia was usedto tune the pH values of the solutions, respectively, to be 10.4,11.0 and 11.6. After the calcination of the precipitations, TiO2micro/nanospheres were obtained. Under different pH values,TiO2 micro/nanospheres showed different anatase/rutile compo-sitions. Degradation of methylene blue indicated that the TiO2micro/nanospheres showed the enhanced photocatalytic activityas the precursor pH value increased.

Acknowledgements

This study was partly supported by the Beijing Natural ScienceFoundation, PHR (IHLB), the 863 Project (grant no. 2007 AA021906)and the 973 Project (grant no. 2010CB 933501).

References

[1] A. Fujishima, K. Honda, Nature 238 (1972) 37–38.[2] U.G. Akpan, B.H. Hameed, J. Hazard. Mater. 170 (2009) 51–59.[3] T. Kawai, T. Sakata, Nature 286 (1980) 474–476.[4] Y. Shiraishi, Y. Sugano, S. Tanaka, T. Hirai, Angew. Chem. Int. Ed. 49 (2010)

1656–1660.[5] D.M. King, X.H. Liang, C.S. Carney, L.F. Hakim, P. Li, A.W. Weimer, Adv. Funct.

Mater. 18 (2008) 607–615.[6] H.Y. Chuang, D.H. Chen, Nanotechnology 20 (2009) 105704–105713.[7] G.S. Wu, J.P. Wang, D.F. Thomas, A.C. Chen, Langmuir 24 (2008) 3503–3509.[8] Y.J. Hwang, A. Boukai, P.D. Yang, Nano. Lett. 9 (2009) 410–415.[9] S. Zhang, C.Y. Liu, Y. Liu, Z. Zhang, G. Li, J. Am. Ceram. Soc. 91 (2008) 2067–2070.10] T.H. Eum, S.H. Kim, W.J. Jeong, S.J. Jeon, S.H. Kim, S.M. Yang, Chem. Mater. 21

(2009) 201–203.11] H. Strohm, P. Lobmann, J. Mater. Chem. 14 (2004) 2667–2673.12] T. Sugimoto, T. Kojima, J. Phys. Chem. C 112 (2008) 18760–18771.

13] A.K. Alves, F.A. Berutti, F.J. Clemens, T. Graule, C.P. Bergmann, Mater. Res. Bull.

44 (2009) 312–317.14] X.B. Chen, S.S. Mao, Chem. Rev. 107 (2007) 2891–2959.15] C.H. Wang, C.L. Shao, Y.H. Liu, X.H. Li, Inorg. Chem. 48 (2009) 1105–1113.16] Z.F. Zheng, H.W. Liu, J.P. Ye, J.C. Zhao, E.R. Waclawik, H.Y. Zhu, J. Mol. Catal. A:

Chem. 316 (2010) 75–82.