Cathodic deposition of TiO2 thin films with supercritical CO2 emulsified electrolyte

(1)

Short communication

Cathodic deposition of TiO

2

thin

ﬁlms with supercritical CO

2

emulsi

ﬁed electrolyte

Tso-Fu Mark Chang

a,

⁎

, Wei-Hao Lin

b

, Yung-Jung Hsu

b

, Chun-Yi Chen

c

, Tatsuo Sato

a

, Masato Sone

a

Precision and Intelligence Laboratory, Tokyo Institute of Technology, 4259-R2-35 Nagatsuta-cho, Midori-ku, Yokohama 226-8503, Japan b_{Department of Materials Science and Engineering, National Chiao Tung University, 1001 University Road, Hsinchu 30010, Taiwan, ROC} c

Department of Electrochemistry, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8503, Japan

a b s t r a c t

a r t i c l e i n f o

Article history: Received 3 April 2013

Received in revised form 24 April 2013 Accepted 24 April 2013

Available online 2 May 2013 Keywords: TiO2 Cathodic deposition Supercritical CO2 Morphology Grain size

TiO2thinﬁlms were fabricated by cathodic deposition with a supercritical CO2(sc-CO2) emulsiﬁed TiCl3+ NaNO3

electrolyte. Morphology and average grain size were evaluated by SEM, TEM and XRD. SEM micrographs showed that the TiO2ﬁlms fabricated by TiCl3+ NaNO3electrolyte were porous, and theﬁlms were composed of particles

and aggregates of the particles. Also, size of the particles increased when the sc-CO2emulsiﬁed electrolyte was

used. Average grain size of the TiO2ﬁlms was calculated using Scherrer equation. The average grain size was

found to increase when the sc-CO2emulsiﬁed electrolyte was used. In addition, both particle sizes observed

from SEM and average grain size calculated using Scherrer equation were found to increase when pressure was increased from atmospheric pressure to 20 MPa.

1. Introduction

TiO2has been extensively investigated for applications in various

ﬁelds of technology, such as solar cell[1], photocatalysis[2], hydrogen generation[3], and gas sensing[4]. TiO2thinﬁlms can be fabricated

by chemical vapor deposition[5], sol–gel method[6], spray pyrolysis [7], and hydrothermal reaction [8]. These techniques usually involve prolonged reaction time which causes poor morphological control of the ﬁlms. Among the different synthetic routes, cathodic deposition offers a low-cost yet effective process for fabrication of TiO2thinﬁlms[9–11].

Supercritical CO2 (sc-CO2) is one of the solutions to overcome

problems encountered when applying electrochemistry in miniaturi-zation of electronic devices[12]. However, CO2is non-polar. Metal

salt solubility and electrical conductivity are both extremely low in CO2, which limit application of sc-CO2in electrochemical reactions.

Therefore, a surfactant is used to form an emulsiﬁed electrolyte composed of the aqueous electrolyte and sc-CO2in order to conduct

electrochemical reactions[13]. Surface roughness and grain size of the electroplated Ni[14]and Cu[15]ﬁlms are both found to decrease with application of sc-CO2emulsiﬁed electrolyte. On the other hand,

Cufilled into a confined space such as nano-via by electroplating with sc-CO2emulsified electrolyte was reported to be single crystal[16].

This study would be theﬁrst report on application of sc-CO2

emulsi-ﬁed electrolyte in cathodic deposition of oxide ﬁlms, such as TiO2.

Applications of TiO2thinﬁlms are highly dependent on morphology

and crystal structure of theﬁlms[17]. The sc-CO2emulsiﬁed electrolyte

is expected to affect the morphology and crystal structure of the TiO2

films deposited significantly. Therefore, this study is focused on study-ing effects of sc-CO2emulsified electrolyte on morphology and crystal

structure of the TiO2ﬁlms deposited.

2. Experimental

Details of the high-pressure experimental apparatus used can be found in a previous study[13]. CO2with a minimum purity of 99.9%

was used. The TiO2electrolyte was composed of 0.47 M NaCl, 25 mM

TiCl3, and 75 mM NaNO3. A non-ionic surfactant, polyoxyethylene

lauryl ether (C12H25(OCH2CH2)15OH) was used to form the emulsion.

Ni plates with dimensions of 1.0 × 2.0 cm2_{were used as the working}

electrode at cathode. Pt plates with dimensions of 1.0 × 2.0 cm2_were

used as the counter electrode at anode.

The deposition with only the TiO2electrolyte is denoted as

conven-tional deposition (CONV). TiO2ﬁlms were also prepared by CONV with

0.08 vol.% (CONV-8) and 0.16 vol.% (CONV-16) of the surfactant with respect to total volume of the reaction chamber to clarify the effects of the surfactant. The deposition with the sc-CO2emulsiﬁed electrolyte

is denoted as DSCE. For DSCE, 20 vol.% of CO2and 0.08 vol.% of the

surfactant with respect to total volume of the reaction chamber were used. The pressures used for DSCE were 10 MPa (DSCE-10) and 20 MPa (DSCE-20). Cathodic current density of 25 mA/cm2_{, temperature of}

40 °C, and deposition time of 10 min were used for all the deposition processes.

Electrochemistry Communications 33 (2013) 68–71

⁎ Corresponding author. Tel.: +81 45 924 5631.

http://dx.doi.org/10.1016/j.elecom.2013.04.025

Contents lists available atSciVerse ScienceDirect

Electrochemistry Communications

(2)

Morphology of the TiO2ﬁlms was examined by a scanning electron

microscope (FESEM, S-4300SE, Hitachi). Crystal structural analysis was conducted by transmission electron microscopy (TEM, JEM-2100, JEOL, operated at 200 keV) and X-ray diffraction (XRD, Ultima IV, Rigaku) at a glancing angle of 1.0°. The samples were annealed in air at 400 °C for 1 h before the SEM, TEM, and XRD analysis. Average grain size was calculated using Scherrer equation. At least three samples were prepared with each condition for calculation of the average grain size. 3. Results and discussion

Morphology of the TiO2 ﬁlm deposited by CONV (no surfactant

used) was porous as shown inFig. 1(a) and (b). Theﬁlm was composed of nano-scale pores and mostly micro-scale aggregates. The aggregates were composed of nano-scale particles. The results are similar to the results reported by Hu et al., where nano-scale pores were observed when the temperature was higher than 30 °C [9–11]. When the surfactant was added, increase in size of the pores was observed. Size of the aggregates and the particles did not change much as shown inFig. 1(c), where 0.08 vol.% of the surfactant was used. The pores were suggested to be mainly caused by evolution of gases as shown in reactions1 to 3 [10]:

2H2Oþ 2e−→2OH−þ H2 ð1Þ

2NO3−þ 6H2Oþ 10e−→12OH−þ N2 ð2Þ

NO3−þ 6H2Oþ 8e−→9OH−þ NH3: ð3Þ

Adsorption of gas bubbles on the surface of cathode was expected to be more signiﬁcant after addition of the surfactant, which were reported in electroplating of Ni and Cu with sc-CO2 emulsiﬁed

electrolyte[13,15]. This should be the main cause for increase of the pore size observed after addition of the surfactant.

When sc-CO2emulsiﬁed electrolyte was used, particles with size

in several hundreds of nm were observed as shown in Fig. 1(d), where the pressure was 10 MPa. Nano-scale particles could still be found surrounding the micro-scale particles for DSCE-10. For the size of the pores, distribution of the pore size became more uniform when comparing betweenFig. 1(c) and (d). The more uniform distri-bution of the pore size should be caused by more uniform adsorption and desorption of the H2·N2, and NH3from the surface of cathode

after emulsifying the electrolyte with sc-CO2. The dispersed phases

in the sc-CO2emulsiﬁed electrolyte are like micelles in oil-in-water

emulsion[13]. The dispersed phases would constantly have contact with the cathode to lead to more uniform adsorption and desorption of the H2·N2, and NH3. Morphology of the TiO2ﬁlm fabricated by

DSCE-20 was similar to the case of DSCE-10 as shown inFig. 1(e), where particles with size in several hundreds of nm were observed.

From XRD patterns shown in Fig. 2, the particles forming the aggregates and the ﬁlms observed inFig. 1were conﬁrmed to be

500 nm 2 µm 2 µm 2 µm

a)

d)

c)

b)

2 µm

e)

Fig. 1. SEM micrographs of the TiO2films fabricated by (a) conventional deposition (CONV), (b) higher magnification of CONV, (c) conventional deposition with 0.08 vol.% of the surfactant (CONV-8), (d) deposition with sc-CO2emulsified electrolyte at 10 MPa (DSCE-10), and (e) deposition with sc-CO2emulsified electrolyte at 20 MPa (DSCE-20).

69 T.-F.M. Chang et al. / Electrochemistry Communications 33 (2013) 68–71

(3)

anatase structure TiO2. Full width at half maximum of the strongest

diffraction peak corresponding to the facet (101) of anatase was used in Scherrer equation to calculate average grain size of the TiO2

films. For CONV, increase in grain size of the films was observed after addition of the surfactant, where the grain size increased from 8.84 ± 0.45 to 11.27 ± 0.55 nm for surfactant volume fraction from 0 to 0.16 vol.% as shown inFig. 3. Grain size of the TiO2films obtained

in this study was in the same range as reported in Hu et al.'s study [11]. Although the experimental conditions used in Hu et al.'s study are not exactly the same as the conditions used in this study, however, composition of the base electrolyte is the same between the two studies. When sc-CO2was introduced to the electrolyte containing 0.08 vol.%

of the surfactant, the grain size increased from 9.63 ± 0.33 nm for CONV-8 to 11.46 ± 0.77 nm for DSCE-10. The grain size further

increased to 14.84 ± 0.25 nm when DSCE-20 was used. Grain size of theﬁlm fabricated by DSCE-20 was conﬁrmed to be roughly 10 nm from TEM micrograph shown inFig. 3. The trend of increase in grain size observed here matched the increase in size of the particles observed in the SEM micrographs.

For cathodic deposition of TiO2, formation of the intermediate

products, TiO2+_{and oxy-hydroxyl-Ti species, is required, as shown}

in reactions4 and 5:

Ti3þþ NO3−→TiO2þþ NO2 ð4Þ

TiO2þþ 2OH−þ xH2O→TiO OHð Þ2∙xH2O: ð5Þ

Fig. 2. XRD patterns of the TiO2ﬁlms deposited under different conditions, anatase with ICDD# 01-070-6826, and Ni with ICDD# 00-004-0850.

Fig. 3. Effects of the surfactant volume fraction and pressure on average grain size of the TiO2ﬁlms. Left and lower axes correspond to the effects of the surfactant. Right and upper axes correspond to the effects of pressure. The embedded image is high resolution TEM of DSCE-20.

(4)

After formation of the oxy-hydroxyl-Ti species, TiO2could be formed

after annealing as shown in reaction6:

TiO OHð Þ2→TiO2þ H2O: ð6Þ

Source of OH−is believed to be the most important step in cathodic deposition of TiO2, shown in reaction5 [9–11,17]. Kholmanov et al.

reported that larger oxy-hydroxyl-Ti particles led to the formation of larger TiO2particles and grains[18], and Hu et al. reported higher

gen-eration rate of the OH−would lead to formation of larger TiO2grains

[11]. In DSCE, supply of reactants to the reaction site, such as NO3−to

the surface of cathode, and removal of products from the reaction site, such as H2, N2and NH3from the surface of cathode, would be enhanced

because the dispersed phases could improve transfer of materials into and out of the diffusion layer[15,19,20]. In addition, N2and NH3

produced in reactions2 and 3would be in supercritical and liquid state, respectively, where pressure higher than 10 MPa and tempera-ture of 40 °C were used. Critical point of N2is−147 °C and 3.3 MPa

[21]. Phase transition pressure of NH3at 40 °C is about 1.6 MPa[22].

On the other hand, N2and NH3produced under atmospheric pressure

and 40 °C would be both in gas state. Therefore, we believe the reaction rates of reactions2 and 3 would be enhanced in sc-CO2emulsiﬁed

electrolyte to give a higher generation rate of OH−and lead to the grain coarsening effect.

Results obtained in this study showed that morphology of the TiO2

ﬁlms could be affected by the surfactant used in this study and the sc-CO2 emulsiﬁed electrolyte. Furthermore, the surfactant and the

sc-CO2were found to have synergetic effect on the grain coarsening

effect. Effects including the interaction between the surfactant and the surface of cathode, and the enhancement in transfer of materials to and away from the reaction site after emulsifying the electrolyte with sc-CO2, N2in supercritical state and NH3in liquid state at high

pressure are all believed to contribute to the phenomena observed in this study. However, further study is still needed to clarify the main factor causing the phenomena.

4. Conclusions

This study is the_{ﬁrst study to report application of sc-CO}2emulsiﬁed

electrolyte in cathodic deposition of TiO2thinﬁlms. Pore size of the TiO2

ﬁlm fabricated by CONV was found to increase after addition of the surfactant. Size distribution of the pores became more uniform and size of the particles observed from SEM increased from several to several hundreds of nm when DSCE was applied. In addition, grain size of the TiO2 ﬁlms increased after addition of the surfactant and further

increased after emulsifying the electrolyte with sc-CO2. The results

obtained in this study showed that DSCE is a promising process to control morphology, particle size, and grain size of the TiO2thinﬁlms

deposited cathodically.

Acknowledgment

Funding Program for Next Generation World-leading Researchers (NEXT Program) GN037, Cabinet Ofﬁce (CAO), Japan is acknowledged. W.H. Lin and Y.J. Hsu acknowledge theﬁnancial support from the National Science Council of Taiwan (NSC-101-2213-M-009-018).

References

[1] B. O'Regan, M. Grätzel, Nature 353 (1991) 737.

[2] M.R. Elahifard, S. Rahimnejad, S. Haghighi, M.R. Gholami, Journal of the American Chemical Society 129 (2007) 9552.

[3] N. Lakshminarasimhan, A.D. Bokare, W. Choi, Journal of Physical Chemistry C 116 (2012) 17531.

[4] F. Liang, T.L. Kelly, L.B. Luo, H. Li, M.J. Sailor, Y.Y. Li, ACS Applied Materials & Interfaces 4 (2012) 4177.

[5] W.J. Lee, Y.M. Sung, Crystal Growth and Design 12 (2012) 5792.

[6] S. Lee, I.S. Cho, J.H. Lee, D.H. Kim, D.W. Kim, J.Y. Kim, H. Shin, J.K. Lee, H.S. Jung, N.G. Park, K. Kim, M.J. Ko, K.S. Hong, Chemistry of Materials 22 (2010) 1958. [7] E.L. Unger, E. Ripaud, P. Leriche, A. Cravino, J. Roncali, E.M.J. Johansson, A. Hagfeldt,

G. Boschloo, Journal of Physical Chemistry C 114 (2010) 11659. [8] A. Nakahira, T. Kubo, C. Numako, Inorganic Chemistry 49 (2010) 5845. [9] C.C. Hu, C.C. Huang, K.H. Chang, Electrochemistry Communications 11 (2009) 434. [10] C.C. Huang, H.C. Hsu, C.C. Hu, K.H. Chang, Y.F. Lee, Electrochimica Acta 55 (2010)

7028.

[11] C.C. Hu, H.C. Hsu, K.H. Chang, Journal of the Electrochemical Society 159 (2012) D418.

[12] J. Ke, W. Su, S.M. Howdle, M.W. George, D. Cook, M. Perdjon-Abel, P.N. Bartlett, W. Zhang, F. Cheng, W. Levason, G. Reid, J. Hyde, J. Wilson, D.C. Smith, K. Mallik, P. Sazio, Proceedings of the National Academy of Science 106 (2009) 14768.

[13] T.F.M. Chang, M. Sone, A. Shibata, C. Ishiyama, Y. Higo, Electrochimica Acta 55 (2010) 6469.

[14] S.T. Chung, H.C. Huang, S.J. Pan, W.T. Tsai, P.Y. Lee, C.H. Yang, M.B. Wu, Corrosion Science 50 (2008) 2614.

[15] T.F.M. Chang, T. Shimizu, C. Ishiyama, M. Sone, Thin Solid Films 529 (2013) 25. [16] N. Shinoda, T. Shimizu, T.F.M. Chang, A. Shibata, M. Sone, Thin Solid Films 529

(2013) 29.

[17] X. Chen, S.S. Mao, Chemistry Reviews 107 (2007) 2891.

[18] I.N. Kholmanov, E. Barborini, S. Vinati, P. Piseri, A. Podestà, C. Ducati, C. Lenardi, P. Milani, Nanotechnology 14 (2003) 1168.

[19] D. Roha, in: R.A. Mackay, J. Texter (Eds.), Electrochemistry in Colloids and Dispersions, VCH, New York, 1992, pp. 179–194.

[20] P.J. Sonneveld, W. Visscher, E. Barendrecht, Journal of Applied Electrochemistry 20 (1990) 563.

[21] P. Nowak, R. Kleinrahm, W. Wagner, The Journal of Chemical Thermodynamics 29 (1997) 1137.

[22] B.F. Dodge, H.N. Davis, Journal of the American Chemical Society 47 (1927) 610.

71 T.-F.M. Chang et al. / Electrochemistry Communications 33 (2013) 68–71