TiO2 films prepared by ultrasonic spray pyrolysis

3 downloads 0 Views 150KB Size Report
The TiO2 films were annealed subsequently in air for 4 h at. 600, 800 and 1000uC ... These data are interpreted in terms of phase transformation temperature,.
TiO2 films prepared by ultrasonic spray pyrolysis A. Nakaruk1, P. J. Reece2, D. Ragazzon1 and C. C. Sorrell1 Fully dense anatase and rutile (TiO2) films were coated on (0001) a-quartz wafers using ultrasonic spray pyrolysis in air at 400uC for 2 h. The TiO2 films were annealed subsequently in air for 4 h at 600, 800 and 1000uC respectively. The films were characterised using X-ray diffraction (mineralogy), ultraviolet visible spectrophotometry (optical properties) and photoluminescence (structural defects). These data are interpreted in terms of phase transformation temperature, optical mean free path, refractive index and effect of heat treatment on the concentration of oxygen vacancies and/or silicon contamination. Keywords: Titanium dioxide, Ultrasonic spray pyrolysis, Optical properties, Photoluminescence

Introduction

Methodology

Titanium dioxide (TiO2) has been studied widely owing to its variable and controllable electronic, magnetic, optical and electrochemical properties. As a wide energy band gap semiconductor,1,2 the applications for TiO2 span a large range, including self-cleaning materials,3 photodecomposition of water,4 purification of environmental pollutants5 as well as photovoltaics and photocatalysis.6–8 Recently, considerable efforts have been made in the synthesis of TiO2 thin films due to their excellent gas sensitivities.9 Most of these applications involve studies of thin and thick films that have been produced by many techniques, such as sputtering,10 pulsed laser deposition,11 sol–gel,1,12 gel oxidation,13 anodic oxidation,14 electrophoretic deposition15 and spray pyrolysis.16 Over the past two decades, considerable efforts have been made to develop metal oxide semiconductor materials with properties suitable for optoelectronic devices, including photoluminescence (PL), electroluminescence and nonlinear optics. Such optical properties can be used to elucidate useful information of the structures of semiconductors. It is generally known that, in metal oxides, oxygen vacancies are the most common defect and these can act as radiative centres in luminescence processes.17–21 However, although many efforts have been invested in the preparation and properties of TiO2 films,10–16 it appears that there is a limited number of reports on the PL of TiO2 films.22–26 The present work reports some data on a set of related mineralogical and optical properties, the latter being transparency, optical band gap and PL, which do not appear to have been reported simultaneously.

Ultrasonic spray pyrolysis was used to prepare fully dense TiO2 films; the details of this technique and the materials used are described in more detail elsewhere.16 In brief, a continuously produced aerosol was entrained in air flow to a (0001) a-quartz substrate, heated by a hot plate. The deposition temperature was 400¡2uC and the deposition time was 2 h. This condition was found to produce as deposited and annealed films of y1 mm thickness16 (measured growth rate5y0?67 mm h21).27 Characterisation involved the following techniques in situ: (i) X-ray diffraction (XRD): the mineralogies of the TiO2 films were examined by XRD using a Philips X’pert multipurpose X-ray diffraction system. (ii) ultraviolet visible (UV-VIS) spectrophotometry: the transmission spectra in the wavelength range 300–800 nm were obtained using a dual beam UV-VIS spectrophotometer (Cary 100 Bio) These spectra allowed the determination of the transparency and the indirect optical band gap (iii) photoluminescence: the PL was determined using a SPEX 270M spectrometer (thermoelectrically cooled charge coupled device). An excitation wavelength of 337 nm, generated by an N2 laser with a frequency of 10 Hz and power of 1?43 mJ per pulse, was selected since this is higher than the indirect optical band gaps of anatase and rutile observed in the present work. The sample was cooled to 20 K using liquid He.

1

School of Materials Science and Engineering, University of New South Wales, Sydney, NSW 2052, Australia 2 School of Physics, University of New South Wales, Sydney, NSW 2052, Australia Corresponding authors, email [email protected] and [email protected]

ß 2010 Institute of Materials, Minerals and Mining Published by Maney on behalf of the Institute Received 9 April 2009; accepted 24 June 2009 DOI 10.1179/026708309X12468927349299

Results and discussion Figure 1 shows the XRD patterns of films as deposited at 400uC and annealed at 600, 800 and 1000uC. The as deposited film showed very weak peaks for anatase, with a significant amount of amorphous contents being implied by the small size of the peaks and the absence of other peaks. In contrast, the better developed anatase

Materials Science and Technology

2010

VOL

26

NO

4

469

Nakaruk et al.

TiO 2 films prepared by ultrasonic spray pyrolysis

1 X-ray diffraction (XRD) data at room temperature for films as function of annealing temperature

peaks for the film annealed at 600uC showed a greater amount of recrystallisation. After annealing at 800uC, the presence of mixed phase anatase–rutile is consistent with having exceeded the time–temperature conditions for the anataseRrutile phase transformation at y600uC.28 While it is possible to retain the anatase phase up to y900uC,29 the annealing temperature of 1000uC was sufficient to convert all of the anatase to rutile. The light transmission spectra of the TiO2 films and the quartz substrate are shown in Fig. 2. It can be seen that the films as deposited at 400uC and annealed at 600uC, consisting of anatase, were more transparent than the films containing rutile, which were annealed at 800 and 1000uC. The anatase films showed relatively constant values of about 75–85% transmission over the visible range (400–800 nm) while the mixed phase anatase–utile (annealed at 800uC) and rutile (annealed at 1000uC) films showed variable light transmission spectra in the range 10–65% transmission. The possible reasons for decreasing light transmission are: (i) scattering (photon mean free path): it is generally known that if the optical mean free path is approximately smaller than or equal to the grain size, optical transmission will be decreased; conversely, if it is larger than the grain size, there is little effect.30 For example, the optical transmission of Ga doped ZnO thin films decreased when the mean grain size was greater than the mean free path.31 In the present case, the anatase and rutile grain sizes were about 50– 100 nm and 700 nm respectively.16 Hence, rutile and not anatase would be expected to decrease optical transmission (ii) reflection (refractive indices): the higher the refractive index, the higher the amount of total internal reflection.32 Since the refractive index of rutile (2?35) is higher than that of anatase (2?10),29 then it can be expected that the amount of light transmission through anatase should be greater than that through rutile (iii) scattering (grain boundary interfaces): the amount of light scattering depends on the particle size (number of grain boundaries), phases (number and types), and mutual orientations. This effect is unlikely to be relevant with the following aspects: first, since anatase is of a small grain size (about 50–100 nm) relative to rutile (y700 nm), the former would have more

N

470

Materials Science and Technology

2010

VOL

26

NO

4

2 Dual beam UV-VIS spectrophotometry transmission data for films as function of annealing temperature and for (0001) a-quartz substrate

grain boundaries and hence would be expected to cause more scattering. The greater transmission in the small grained films indicates that this is not the case second, since anatasezrutile introduce interfaces between dissimilar phases, enhanced light scattering at the grain boundaries relative to a single-phase material would be expected. The similarity of the data for the films annealed at 800uC (mixed phase) and 1000uC (single-phase) indicates that this is not the case. It can be added that there was no clear evidence of epitaxial alignment of anatase or rutile,16 which would tend to decrease the light scattering if present. In summary, it is probable that the relation between the mean free path and the grain size is the dominant experimental effect (while the grain size is variable, the refractive indices are not). That is, since the anatase does not affect the mean free path owing to its small grain size, then the number of its grain boundaries (the first point above) and the scattering of the interfaces between anatase and rutile (the second point above) are not important. However, since the conversion efficiency of photocatalytic TiO2 depends on the amount of light absorbed,6 then these observations have some importance in that they may provide some guidance concerning the preparation of microstructures of specific grain sizes that may improve light absorption. However, absorption of light depends on the location of the absorption edge and the corresponding band gap, and these are known to allow strong absorption only in the ultraviolet range in pure TiO2.10–12,18 Also, surface versus bulk issues and the importance of the properties of grain boundaries are critical considerations.33 That is, the lower transmissions of the curves in Fig. 2 (800 and 1000uC) do not result from absorption but from scattering and/or reflection. Thus, fabrication of high absorption TiO2 depends on: (i) modification of the semiconducting properties to lower the band gap (ii) control of the grain size to maximise the relation between the mean free path and the grain size (iii) control of the grain boundary:bulk ratio to optimise the relative effects of these two components of grains. From the light transmission data, the optical band gap can be obtained using the well known formula for

N

Nakaruk et al.

TiO 2 films prepared by ultrasonic spray pyrolysis

Table 1 Summary of PL data Annealing temperature

Emission peaks

Origin of peaks

Explanation

As deposited at 400uC

Small peak: y560 nm

y560 nm: oxygen vacancy with one electron.17

Annealed at 600uC

Large peaks: y475 nm y545 nm y665 nm

y475 nm: oxygen vacancy with two electrons.17 y545 nm: oxygen vacancy with one electron.17 y665 nm: surface defect or surface state energy level.18

Unknown defects in metastable amorphous materials potentially could quench most of the PL, resulting in very little emission. However, PL data for amorphous and crystalline SiO2 reveal that the former contained more oxygen vacancies than the latter,40 which would explain the small peak in TiO2. Increases in temperature (and time) during annealing in air can be expected to change the number of defects resulting from: (i) decreasing concentration of oxygen vacancies owing to move toward equilibration (ii) increasing Si contamination owing to diffusion from the substrate (iii) decreasing concentration of oxygen vacancies owing to the phase transformation from metastable anatase to stable rutile.28

Annealed at 800uC

Medium peaks: y475 nm y545 nm y665 nm yNil

Annealed at 1000uC

yNil

indirect band gap semiconductors;34–37 these details are described elsewhere.16 The values obtained from these data were 3?48, 3?54, 3?38 and 3?26 eV for as deposited at 400uC and annealed at 600, 800 and 1000uC respectively. It is not clear if these values are meaningful in terms of absolute values and trend because anatase and rutile have a wide range of band gaps, these being 3?20–3?56 eV for anatase2,34–37 and 3?00–3?34 eV for rutile.2,36–39 An N2 laser with band pass filter at 337 nm (3?68 eV) was used for the excitation (Ex) wavelength and subsequent decay by the emission (Em) peaks, as shown in Fig. 3 (Table 1 interprets the data shown in Fig. 3). Since this excitation energy is higher than that of the measured band gaps, the electron is transferred directly from the valence band to the conduction band by band– band transition. This transition avoids the potential for the observation of data resulting from a mid gap transition, which may occur in mixed phase anatase– rutile and/or from the presence of impurities and defects in the single-phase anatase (400 and 600uC) and rutile (1000uC) films.

Conclusions X-ray diffraction The XRD data for fully dense TiO2 films of y1 mm thickness deposited by ultrasonic spray pyrolysis on

Completion of equilibration eliminated all of the oxygen vacancies.

(0001) a-quartz substrates effectively are as expected from the literature28,29 Amorphous

400{6000 C

? Anatase

8000 C

?

0

AnatasezRutile

1000 C

? Rutile

UV/VIS Spectrophotometry The optical transmission data derive largely from the relation between the mean free path and the grain size, where the anatase grain sizes (about 50–100 nm) were too small to affect the mean free path. However, the wavelengths used were largely less than or equal to the rutile grain size (y700 nm), which resulted in scattering in these films. This observation leads to the conclusion that optical light absorption in photocatalytic TiO2 requires consideration of modification of the absorption edge/band gap, control of the grain size to improve light trapping by scattering and control of the grain boundary:bulk ratio to optimise their relative effects.

Photoluminescence Emission peaks are likely to result from the presence of oxygen vacancies. However, quenching of the PL peaks probably results from unknown defects in the metastable amorphous TiO2 precursor and/or elimination during annealing of the oxygen vacancies owing to equilibration, transformation of metastable anatase to stable rutile and/or Si contamination from the substrate.

Acknowledgements The authors are grateful for the financial support of Austral Brick Co. Pty. Ltd, which has allowed this and other developmental work to be undertaken. The authors also wish to thank Dr Jani O’Rourke (UNSW School of Biotechnology and Biomolecular Sciences) for assistance in obtaining the UV-VIS spectrophotometry data.

References 3 Photoluminescence data at 20 K for films as function of annealing temperature

1. M. Gra¨tzel: J. Sol–Gel Sci. Technol., 2001, 22, (1–2), 7–13. 2. H. H. Huang, C. C. Huang, P. C. Huang, C. F. Yang and C. Y. Hsu: J. Nanosci. Nanotechnol., 2008, 8, (5), 2659–2664.

Materials Science and Technology

2010

VOL

26

NO

4

471

Nakaruk et al.

TiO 2 films prepared by ultrasonic spray pyrolysis

3. I. P. Parkin and R. G. Palgrave: J. Mater. Chem., 2005, 15, (17), 1689–1695. 4. S. Sato and J. M. White: J. Phys. Chem., 1981, 85, (5), 592–594. 5. H. Ichiura, T. Kitaoka, and H. Tanaka: Chemosphere, 2003, 50, (9), 79–83. 6. T. Bak, J. Nowotny, M. Rekas and C. C. Sorrell: Int. J. Hydrogen Energy, 2002, 27, (10), 991–1022. 7. T. Bak, J. Nowotny, M. Rekas and C. C. Sorrell: Int. J. Hydrogen Energy, 2002, 27, (1), 19–26. 8. J. Nowotny, C. C. Sorrell, L. R. Sheppard and T. Bak: Int. J. Hydrogen Energy, 2005, 30, (5), 521–544. 9. C. Garzella, E. Comini, E. Tempesti, C. Frigeri and G. Sberveglieri: Sens. Actuators B, 2000, 68B, (1–3), 189–196. 10. M. F. Hossain, S. Biswas, T. Takahashi, Y. Kubota and A. Fujishima: Thin Solid Films, 2008, 516, (20), 7149–7154. 11. Y. Suda, H. Kawasaki, T. Ueda and T. Ohshima: Thin Solid Films, 2004, 453–454, (1), 162–66. 12. M. Sasani Ghamsari and A. R. Bahramian: Mater. Lett., 2008, 62, (3), 361–364. 13. H. Z. Abdullah and C. C. Sorrell: Mater. Sci. Forum, 2007, 561– 565, 2167–2170. 14. H. Z. Abdullah and C. C. Sorrell: Mater. Sci. Forum, 2007, 561– 565, 2159–2162. 15. H. Z. Abdullah and C. C. Sorrell: Mater. Sci. Forum, 2007, 561– 565, 2163–2166. 16. A. Nakaruk, D. Ragazzon and C. C. Sorrell: ‘Anatase-rutile transformation through high-temperature annealing of titania films produced by ultrasonic spray pyrolysis’, Unpublished work. 17. J. M. Wu, H. C. Shih, W. T. Wu, Y. K. Tseng and I. C. Chen: J. Cryst. Growth, 2005, 281, (2–4), 384–390. 18. L. Q. Jing, X. J. Sun, B. F. Xin, B. Q. Wang, W. M. Cai and H. G. Fu: J. Solid State Chem., 2004, 177, (10), 3375–3382. 19. L. Forss and M. Schubnell: Appl. Phys. B, 1993, 56B, (6), 363–366. 20. Y. X. Zhang, G. H. Li, Y. X. Jin, Y. Zhang, J. Zhang and L. D. Zhang: Chem. Phys. Lett., 2002, 365, (3–4), 300–304 21. Y. K. Lai, L. Sun, C. Chen, C. G. Nie, J. Zuo and C. J. Lin: Appl. Surf. Sci., 2005, 252, (4), 1101–1106. 22. S. K. Deb: Solid State Commun., 1972, 11, (5), 713–715.

472

Materials Science and Technology

2010

VOL

26

NO

4

23. K. M. Krishna, Md. Mosaddeq-ur-Rahman, T. Miki, T. Soga, K. Igarashi, S. Tanemura and M. Umeno: Appl. Surf. Sci., 1997, 113–114, 149–154. 24. M. M. Rahman, K. M. Krishna, T. Soga, T. Jimbo and M. Umeno: J. Phys. Chem. Solids, 1999, 60, (2), 201–210. 25. J. Yu, J. Xiong, B. Cheng and S. Liu: Appl. Catal. B, 2005, 60B, (3– 4), 211–221. 26. A. P. Huang, Z. F. Di and P. K. Chu: Surf. Coat. Technol., 2007, 201, (9–11), 4897–4900. 27. D. Ragazzon, A. Nakaruk and C. C. Sorrell: ‘Deposition Rate of Anatase Films by Ultrasonic Spray Pyrolysis’, Unpublished work. 28. L. E. Depero, P. Bonzi, M. Zocchi, C. Casale and G. de Michele: J. Mater. Res., 1993, 8, (10), 2709–2715. 29. D. J. Kim, S. H. Hahn, S. H. Oh and E. J. Kim: Mater. Lett., 2002, 57, (2), 355–360. 30. W. D. Kingery, H. K. Bowen and D. R. Uhlmann: ‘Introduction to ceramics’, 2nd edn; 1976, New York, John Wiley and Sons. 31. K. Shimakawa and T. Itoch: Jpn J. Appl. Phys., 2007, 46, (24), L577–L579. 32. R. Webster and B. W. Anderson: ‘Gems: their sources, description and identification’, 4th edn; 1987, Sevenoaks, Butterworths. 33. L. R. Sheppard, J. Nowotny and T. Bak: Adv. Appl. Ceram., 2007, 106, (1/2), 49–56. 34. H. Tang, K. Prasad, R. Sanijine`s, P. E. Schmid and F. Le´vy: J. Appl. Phys., 1994, 75, (4), 2042–2047. 35. H. Takikawa, T. Matsui, T. Sakakibara, A. Bendavid and P. J. Martin: Thin Solid Films, 1999, 348, (1–2), 145–151. 36. D. Mardare, M. Tasca, M. Delibas and G. I. Rusu: Appl. Surf. Sci., 2000, 156, (1–4), 200–206. 37. D. J Won, C. H. Wang, H. K. Jang and D. J. Choi: Appl. Phys. A, 2001, 73A, (5), 595–600. 38. L. Miao, P. Jin, K. Kaneko, A. Terai, N. Nabatova-Gabain and S. Tanemura: Appl. Surf. Sci., 2003, 212–213, 255–263. 39. C. C. Ting, C. H. Chen and D. M. Liu: J. Appl. Phys., 2000, 88, (8), 4628–4633. 40. Z. Y. Lu, C. J. Nicklaw, D. M. Fleetwood, R. D. Schrimpf and S. T. Pantelides: Phys. Rev. Lett., 2002, 89, (28), 285505.