Characterization of tungsten–titanium oxide electrode for electrochromic applications

Characterization of tungsten–titanium oxide electrode for electrochromic applications

Applied Surface Science 255 (2009) 3848–3853 Contents lists available at ScienceDirect Applied Surface Science journal homepage: www.elsevier.com/lo...

1MB Sizes 0 Downloads 69 Views

Applied Surface Science 255 (2009) 3848–3853

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Characterization of tungsten–titanium oxide electrode for electrochromic applications Ko-Wei Weng a, Sheng Han b,*, Ya-Chi Chen c, Da-Yung Wang a a

Department of Materials Science and Engineering, Mingdao University, 369 Wen-Hua Road, Peetow, Changhua 523, Taiwan, ROC Department of Leisure and Recreation Management, National Taichung Institute of Technology, 129 San-min Road, Section 3, Taichung 404, Taiwan, ROC c Department of Materials Science and Engineering, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 402, Taiwan, ROC b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 17 June 2008 Received in revised form 18 October 2008 Accepted 20 October 2008 Available online 31 October 2008

Titanium oxide films are of critical importance for the electrochromic device technology. The substrate, a conductive glass being coated with indium tin oxide (ITO) thin films, was deposited tungsten and titanium oxide by pulsed co-sputtering deposition system. The film thickness increased with the ion beam power. However, the slope of the curve of thickness against power at an ion beam power of less than 300 W was greater than that at a power of 400 or 500 W. A high ion beam power resulted produced a crystalline structure, as revealed by X-ray diffraction (XRD). Moreover, increasing the ion beam power resulted in the high Li-ions transport. The electrochromic behavior was optimal at an ion beam power of 200 W. ß 2008 Elsevier B.V. All rights reserved.

Keywords: Titanium oxide Co-sputtering Eelectrochromic behavior

1. Introduction Tungsten oxide is an opto-electronic material and can be used as windows in solar cells, in electrochromic (EC) devices, in information displays and in color memory devices [1–5]. Numerous physical, chemical and solid-state methods for synthesizing tungsten oxide films have been described [1–13]. Sangaletti et al. [8], using W–Ti alloy targets and radio-frequency (RF) sputtering, fabricated an amorphous W–Ti–O layer and analyzed the film morphology and structure. Wang and Hu [12] prepared TiO2-doped WO3 films via the peroxo sol–gel route and discussed the effect of adding Ti. Yamamoto et al. [4] synthesized WOx films using an RF sputtering system and explored the effect of the oxygen flow rate. Karuppasamy and Subrahmanyam [14] investigated the effect of oxygen chamber pressure on the EC properties of titanium-doped tungsten oxide thin films using co-sputtering metallic titanium and tungsten. Sputtering is a widely used technique for preparing thin films, and is superior in terms of composition reproducibility and thinness on a large-area substrate. Co-sputtering can be performed with a range of values of relevant variables to find the conditions that optimize EC behavior. This study is a preliminary survey based on co-sputtering with varying ion beam power. The relationships between electrochro-

* Corresponding author. Tel.: +886 4 22195902; fax: +886 4 22195911. E-mail address: [email protected] (S. Han). 0169-4332/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2008.10.071

mic applications and the other variables, such as gas flow rate, bias voltage and deposition time, are investigated. The goal is to gain insights that can be exploited on an industrial scale to optimize fabrication conditions. 2. Experimental details Indium tin oxide (ITO)-coated glass, with a sheet resistance of 10 V cm and dimensions of 2 cm  2 cm was selected as a substrate. Two targets, tungsten and titanium, were co-sputtered using a pulsed sputtering deposition system. Table 1 presents the experimental deposition parameters. Specimen assemblies were deposited at ion beam powers of 100, 200, 300, 400 and 500 W, and denoted S1, S2, S3, S4 and S5, respectively, for convenience. X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) Table 1 Experimental condition of the deposition parameters. Parameter

Value

Sputter target Substrate Ar flow rate (sccm) O2 flow rate (sccm) Work pressure (Torr) Pulsed power (W) Pulsed frequency (kHz) Deposition time (min)

W and Ti ITO glass 150 50 5  102 100, 200, 300, 400, 500 100 15

K.-W. Weng et al. / Applied Surface Science 255 (2009) 3848–3853

3849

Fig. 1. XRD patterns revealing that the crystalline structure is formed gradually as the ion beam power increases.

Fig. 2. XPS spectrum of the specimen assemblies.

were used to elucidate microstructural changes. The morphology and cross-sectional microstructures of the specimen assemblies were examined by field emission scanning electron microscopy (FESEM). The surface topography of the specimen assemblies was

examined by atomic force microscopy (AFM), which measures the root mean square roughness (Rrms) in a square area of 3 mm. The same vertical scale was chosen for all the specimens for comparison.

Fig. 3. SEM images of the microstructure at different implanted powers, indicating that the films consisted of nanograins; the morphology became gradually coarser from S1 to S5.

3850

K.-W. Weng et al. / Applied Surface Science 255 (2009) 3848–3853

Fig. 4. Cross-sectional SEM micrographs at different ion beam powers.

EC experiments were performed in a three-electrode cell. The specimen assemblies were cycled in a non-aqueous electrolyte (0.1 M LiClO4 in propylene carbonate, PC). A highly pure graphite, a saturated calomel electrode (SCE) and specimen assemblies were used as the counter electrode, the reference electrode and the working electrode, respectively. Transmittance spectra were obtained at constant potential (1.0 V). The transmittances (DT, %) of the specimen was plotted between bleached and colored states (l = 550 and 800 nm). 3. Results and discussion Fig. 1 shows the XRD patterns of the specimen assemblies. Except for S5 (power 500 W), the films (S1–S4 assembly) have amorphous characteristics. The peaks of this spectrum (S2–S5 assembly) became more intense and narrow than those of the corresponding film deposited at 100 W. At a power of 500 W, phases (1 1 0), (1 0 2), (1 1 2), (2 0 2), (1 0 4), (2 2 2) are present in

Fig. 5. Variation of thickness with ion beam powers; a steady state exists at 300 W.

K.-W. Weng et al. / Applied Surface Science 255 (2009) 3848–3853

WO3 (1 0 4) for Ti2O3 and (1 1 1) for Ti0.54W0.48O2. Increasing the ion beam power requires a higher ion beam energy, given other fixed process parameters. Such the high energy may be responsible for the existence of a crystalline structure. Fig. 2 shows the XPS survey spectra of the specimen assemblies. The figure presents tungsten, titanium and oxygen peaks, and reveals a trace amount

3851

of surface physisorbed carbon with no other surface impurity on the surface of the film. The surface morphology characterization by SEM indicates that the specimen assemblies consisted of nanograins in Fig. 3. Its microstructure is fine and dense. The surface grains are quite homogeneous and uniform in size and distribution, because the

Fig. 6. Variation of AFM surface morphology with ion beam power.

K.-W. Weng et al. / Applied Surface Science 255 (2009) 3848–3853

3852

Table 2 Comparisons for specimen assemblies determined by roughness measurement. Specimen

S1

S2

S3

S4

S5

Pulsed power (W) Ra (nm)

100 1.31

200 2.73

300 3.22

400 3.92

500 3.92

energetic ion bombardment during the fabrication process remixes the surface atoms and clusters. Rice-shape pellets were observed gradually from S1 to S5, and are responsible for the coarse morphology. Fig. 4 shows the cross-sectional SEM micrographs of the specimen assembly films prepared at varying power. A columnar structure was found. Fig. 5 plots the film thickness versus ion beam power. When the ion beam power is low, fewer energetic atoms are sputtered from the targets. Therefore, when the deposition rate is low, the film is thin. The thickness of the film increases with the ion beam power. However, the slope of the curve at an ion beam power of less than 300 W is greater than that of the curve obtained at powers of 400 and 500 W, and indicates a steady state. The thickness of the film reaches a maximum of around 1800 nm at 500 W (Fig. 5). Re-sputtering causes the input kinetic energy and momentum to be so high that the deposited material is again vaporized by sputtering, reducing the growth velocity and the final thickness of the film [15,16]. The steady state of growth exists when the ion impingement rate equals the sputtering rate. Fig. 6 shows AFM images of 3 mm  3 mm scan areas for all specimens. The top-view image presents surface morphology with a roughness of between 1.3 and 3.9 nm, as listed in Table 2. Higher ion beam power yields a coarser morphology is. This finding is consistent with the plane-view SEM image. The films exhibited no apparent cracking. Fig. 7 plots the transmittances (DT, %) of the specimen assembly films between bleached and colored states curves (l = 550 and 800 nm). The maximum DT (%) for 550 nm is 55% and for 800 nm is 40%. The value of DT (%) is maximum in both cases at 200 W. It decreases when the ion beam power is either too low or too high. WO3 film in a bleach state (transparent) can be switched reversibly into a colored state (dark blue) by the insertion of both ions and electrons to form tungsten bronze (MxWO3), according to the reaction: WO3 þ xMþ þ xe $ Mx WO3 ðTransparentÞ

(1)

ðDark blueÞ

where M+ is H+, Li+, Na+ or K+ ions. The above equation describes a reduction reaction, which results in coloration, and an oxidation reaction, which results in transparency. The reaction rate depends on the concentration of ions, oxygen deficiency, temperature and other variables. In EC experiments, the transport of Li-ions is a major factor that affects EC behavior. The mobility of Li in the films depends strongly on the microstructure, including the density of the film, its degree of porosity and other factors [17,18]. The films, as revealed by SEM consisted of nanograins and the coarseness of the morphology increased with the ion beam power. The coarse morphology suits the transport of Li ions and improves EC behavior [19]. When Eq. (1) is reversible, EC behavior is improved. An irreversible crystalline phase is not appropriate for EC applications [20]. High ion beam power produced a crystalline structure, as revealed by XRD, which fact may explain the poor EC behavior. The transport of Li ions and the reversibility of Eq. (1) also were responsible for good EC behavior. Increasing the ion beam power, as discussed above, promoted the transport of Li ions and made the reaction irreversible. Varying ion beam power to improve EC behavior involves a trade-off with reducing EC behavior. In this study, EC behavior is favorable when the device is operated at the optimal ion beam power of 200 W. 4. Conclusions The substrate, a conductive glass being coated ITO thin films, was deposited by pulsed co-sputtering deposition system. The above analytical results support the following conclusions: (1) The peaks of this spectrum became more intense and narrow as the ion beam power increased. (2) The film surface grains are quite homogeneous and uniform in size and distribution. A higher ion beam power yields a coarser morphology. (3) The thickness of the film increases with the power of the ion beam. The slope of the curve at an ion beam power of less than 300 W is greater than that of the curve obtained at a power of 400 or 500 W. (4) The maximum DT (%) at 550 nm is 55% and at 800 nm is 40%. In both cases, DT (%) is maximal at 200 W. DT (%) decreases when the ion beam power is either too low or too high. Acknowledgements The authors would like to thank the National Science Council of the Republic of China, Taiwan, for financially supporting this research under Contract No. NSC 96-2745-E-451-004-URD. References

Fig. 7. Transmittance of specimen assembly films (DT, %) between bleached and colored states curves.

[1] X.G. Wang, Y.S. Jang, N.H. Yang, L. Yuan, S.J. Pang, XPS and XRD study of the electrochromic mechanism of WOx films, Surf. Coat. Technol. 99 (1998) 82. [2] E. Comini, G. Sberveglieri, V. Guidi, Ti–W–O sputtered thin film as n- or p-type gas sensors, Sens. Actuators B 70 (2000) 108. [3] J.L. He, M.C. Chiu, Effect of oxygen on the electrochromism of RF reactive magnetron sputter deposited tungsten oxide, Surf. Coat. Technol. 127 (2000) 43. [4] A. Yamamoto, Y. Abe, M. Kawamura, K. Sasaki, Effects of oxygen gettering and target mode change in the formation process of reactively RF sputtered WOx thin films, Vacuum 66 (2002) 269. [5] T.-S. Yang, Z.-R. Lin, M.-S. Wong, Structures and electrochromic properties of tungsten oxide films prepared by magnetron sputtering, Appl. Surf. Sci. 252 (2005) 2029. [6] C. Bechinger, M.S. Burdis, J.-G. Zhang, Comparison between electrochromic and photochromic coloration efficiency of tungsten oxide thin films, Solid State Commun. 101 (1997) 753. [7] N. Ozer, C.M. Lampert, Electrochromic performance of sol–gel deposited WO3– V2O5 films, Thin Solid Films 349 (1999) 205.

K.-W. Weng et al. / Applied Surface Science 255 (2009) 3848–3853 [8] L. Sangaletti, L.E. Depero, G. Sberveglieri, B. Allieri, E. Bontempi, S. Groppelli, Growth on WO3 crystals from W–Ti–O thin films, J. Cryst. Growth 198/199 (1999) 1240. [9] N. o¨zer, S. Sabuncu, J. Cronin, Electrochromic properties of sol–gel deposited Tidoped vanadium oxide film, Thin Solid Films 338 (1999) 201. [10] L. Su, L. Zhang, J. Fang, M. Xu, Z. Lu, Electrochromic and photoelectrochemical behavior of electrodeposited tungsten trioxide films, Sol. Energy Mater. Sol. Cells 58 (1999) 133. [11] A. Rougier, A. Blyr, Electrochromic properties of vanadium tungsten oxide thin films grown by pulsed laser deposition, Electrochim. Acta 46 (2001) 1945. [12] Z. Wang, X. Hu, Electrochromic properties of TiO2-doped WO3 film spin-coated from Ti-stabilized peroxotungstic acid, Electrochim. Acta 46 (2001) 1951. [13] L. Meda, R.C. Breitkopf, T.E. Haas, R.U. Kirss, Investigation of electrochromic properties of nanocrystalline tungsten oxide thin film, Thin Solid Films 402 (2002) 126. [14] A. Karuppasamy, A. Subrahmanyam, Studies on electrochromic smart windows based on titanium doped WO3 thin films, Thin Solid Films 516 (2007) 175.

3853

[15] W. Ensinger, On the mechanism of crystal growth orientation of ion beam assisted deposited thin films, Nucl. Instr. Meth. B 106 (1995) 142. [16] W. Ensinger, Low energy ion assist during deposition—an effective tool for controlling thin film microstructure, Nucl. Instr. Meth. B 127/128 (1997) 796. [17] G.J. Fang, K.-L. Yao, Z.L. Liu, Fabrication and electrochromic properties of double layer WO3(V)/V2O5(Ti) thin films prepared by pulsed laser ablation technique, Thin Solid Films 394 (2001) 64. [18] K.A. Gesheva, A. Cziraki, T. Ivanova, A. Szekeres, Crystallization of chemically vapor deposited molybdenum and mixed tungsten/molybdenum oxide films for electrochromic application, Thin Solid Films 515 (2007) 4609. [19] K.-W. Weng, S. Han, Y.-C. Chen, H.-C. Chuang, Effects of MPII-implanted titanium on the electrochromic properties of tungsten trioxide, Nucl. Instr. Meth. B 266 (2008) 1069. [20] I. Porqueras, E. Bertran, Efficiency of Li doping on electrochromic WO3 thin films, Thin Solid Films 377–378 (2000) 129.