- Email: [email protected]
Anodic titanium oxide photonic crystals prepared by novel cyclic anodizing with voltage versus charge modulation
Electrochemistry Communications 91 (2018) 5–9
Contents lists available at ScienceDirect
Electrochemistry Communications journal homepage: www.elsevier.com/locate/elecom
Anodic titanium oxide photonic crystals prepared by novel cyclic anodizing with voltage versus charge modulation
T
⁎
N.A. Sapoletovaa, S.E. Kushnira,b, , K.S. Napolskiia,b a b
Department of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia Department of Materials Science, Lomonosov Moscow State University, 119991 Moscow, Russia
A R T I C LE I N FO
A B S T R A C T
Keywords: Anodic titania Anodizing Photonic crystals Porous structure Film
Photonic crystals based on titania are of great practical importance in modern photonics owing to a variety of possibilities for their applications in optoelectronics, sensorics, and solar photovoltaics. However, the reproducible, scalable, and low-cost method of the preparation of titania photonic crystals with desired optical characteristics is still absent. Here the novel anodizing regime with voltage versus electric charge modulation, U (Q), is suggested for the precise morphology control of anodic porous films of valve metals oxides. The potential of the suggested approach is demonstrated on the preparation of high-quality one-dimensional titania photonic crystals. A new type of anodic titania nanotubes possessing periodic change of inner diameter and constant outer diameter is prepared using sine-wave modulation of applied voltage with electric charge. The possibility to tune the position of photonic band gap within the whole visible spectrum is shown.
1. Introduction Photonic crystals made of titania attract great attention owing to perspectives for practical applications in optoelectronics, sensorics, and solar photovoltaics. They are caused by the synergy of anomalous dispersion in photonic crystals and of unique properties of titania, such as a high refractive index (nTiO2 = 2.6 at λ = 600 nm [1]), high chemical stability, low toxicity, semiconductor conductivity, and electrochromism. One of the promising methods of the preparation of titania with controlled nanomorphology is titanium anodizing in pore-forming electrolytes [2–4]. In the case of valve metal anodizing under periodically changing conditions one can obtain porous oxide film with periodic modulation of porosity across the thickness [5–8]. This kind of anodic oxide films can be considered as one-dimensional photonic crystals because the effective refractive index changes periodically along the surface normal. The first results on anodizing of titanium at periodically alternating voltage were reported in 2008 [9]. It has been shown that obtained under square-wave voltage modulation anodic titanium oxide (ATO) films, consisted of bamboo-like titania nanotubes, show higher efficiency of solar energy conversion in dye-sensibilized solar cells in comparison with titania nanotubes with smooth walls [10]. It is worth noting that in the case of periodic variation of voltage with time, the accurate periodicity of porosity modulation in ATO cannot be achieved.
⁎
The layer thickness formed during one cycle of anodic oxidation decreases gradually from cycle to cycle due to the decrease in current with the increase in porous film thickness [11]. In 2011, ATO photonic crystals with well-defined photonic band gaps were obtained under periodically alternating current conditions [12]. The resulting structure of porous titania contained tubes with periodically tapering and expanding of outer diameter and was therefore called “concave”. Later, the alternating-current anodizing of titanium has been successfully applied for the preparation of anodic titania films with extended periodical morphology used as one of the components of dye-sensitized solar cells [11,13–17]. Several complex timedependencies of periodic voltage pulses have been proposed for the preparation of porous multilayer films [18,19] and structures similar to the “concave” type ATO [20]. The positions of the photonic band gaps depend on the composition of the medium inside the pores of ATO, that makes possible to use ATO photonic crystals as refractive index sensors [20,21]. Moreover, the optical response of anodic titania photonic crystals can be reproducibly switched on/off by an anodic/cathodic polarization [18]. This effect can be used to create smart color windows or electronic displays. To the best of our knowledge, all anodizing regimes, which have been previously applied to prepare one-dimensional ATO photonic crystals, used either time-periodic voltage or current profiles. However, as was previously noted, titanium anodizing with periodic U(t) profile leads to different layer thickness in the upper and the lower parts of the
Corresponding author at: Department of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia. E-mail address: [email protected] (S.E. Kushnir).
https://doi.org/10.1016/j.elecom.2018.04.018 Received 22 March 2018; Received in revised form 6 April 2018; Accepted 23 April 2018 Available online 24 April 2018 1388-2481/ © 2018 Elsevier B.V. All rights reserved.
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
film. On the other hand, in the case of periodic I(t) profile, anodizing voltage increases from cycle-to-cycle that also break the optical periodicity. Here, we propose a new anodizing regime with voltage versus electric charge modulation U(Q). In situ measuring of charge allows us to control the thickness of the layers with high precision, whereas control of voltage guarantees the identity of voltage-dependent parameters of the ATO structure for each layer from the top to the bottom. The ability to design the morphology of anodic oxide films using the suggested approach has been demonstrated on the example of the synthesis of one-dimensional photonic crystals.
profile. 3) The steps described in item #2 were repeated until the charge spent during anodizing became higher than the maximum value of charge in U(Q) profile. Finally, DC power supply was switched off. Ti was anodized at the voltages in the range from 40 to 60 V using sine wave U(Q) profiles:
2πQ π U = 50 + 10⋅sin ⎛ − ⎞, 2⎠ ⎝ P
(1)
where P is the electric charge spent during one cycle of anodizing. Four samples with charge density per cycle of 0.177, 0.213, 0.256, and 0.295 C·cm−2 were obtained (denoted as S1, S2, S3, and S4, respectively). The total charge density spent during anodizing of each sample S1–S4 was 17.7 C·cm−2. Sample S5 with charge density per cycle of 0.426 C·cm−2 was prepared similar to the sample S2 with the same sine-wave increase and decrease in voltage in U(Q) profile, but with the additional constant voltage anodizing at 60 V for 0.213 C·cm−2 in each cycle (Fig. 1a). The anodized area for all samples was 1.13 cm2. After the end of anodizing the porous oxide films were washed in ethanol and then dried in air. The morphology characterization of ATO films was performed by scanning electron microscopy (SEM) using LEO Supra 50 VP instrument. Perkin Elmer Lambda 950 spectrophotometer was used to record reflectance spectra at incident angle of 8°. The size of the light spot was 3 × 3 mm2.
2. Materials and methods Before anodizing, the titanium foils (99.6% purity, 0.15 mm thick) were electrochemically polished in the mixture of 99.5 wt% acetic acid and 65 wt% perchloric acid with a volume ratio of 9:1. Electropolishing was conducted at temperature below 25 °C during 4 min under squarewave applied voltage: 40 V for 10 s and 60 V for 10 s. Ti was anodized in an ethylene glycol (EG) electrolyte containing 0.3 wt% NH4F, 0.66 wt% CH3COONa (NaAc), and 2 wt% H2O at 30 °C using programmable DC power supply. The electrolyte was agitated at the rate of 480 RPM using overhead stirrer. The distance between Ti foil and Ti counter electrode was 2 cm in all experiments. To create periodic modulation of porosity across thickness of anodic titania film, a specific voltage profile was set as a staircase function of electric charge spent during anodizing using lab-made software. U(Q) anodizing profile was realized using the following algorithm:
3. Results and discussion
1) At the initial stage (Q = 0 C), the anodizing voltage was set to 40 V. 2) After a short time interval Δt (ca. 30 ms), the increment of charge ΔQ was calculated by numerical integration of measured current. New value of charge Q(t + Δt) is the sum of the charge value Q(t) at previous step and the increment ΔQ. Then the voltage was set to the new value U(Q(t + Δt)), which corresponded to the value of charge spent during anodizing in accordance with required voltage – charge
Electrochemical responses recorded during anodizing of titanium under cyclic voltage versus charge modulation are shown in Fig. 1. It can be clearly seen that the voltage applied to the electrochemical cell coincides well with the desired sine-wave U(Q) anodizing profile (Fig. 1a). The periodic variation of anodizing voltage between 40 and 60 V leads to the cyclic change of current density (Fig. 1b). Current density, j, hits the maximum and minimum values at the edges of the
Fig. 1. Electrochemical responses recorded during anodizing of titanium. (a) Applied voltage versus charge modulation U(Q) of the samples S1 (blue solid), S2 (cyan dash), S3 (green dot), S4 (red dash dot), and S5 (black solid). (b) Dependences of current density on voltage with different cycle number for the sample S2: 3 (black solid), 25 (red dash), 50 (green dot), 75 (blue dash dot). Inverse of current density (j−1) versus charge density (q) is plotted in panels (c) and (d): data for the cases of potentiostatic anodizing at 40 V (orange solid) and 60 V (violet dot). Data for the samples S1 (blue solid), S4 (red dash dot) and linear approximation of j−1 maxima for the sample S4 in the case of q ≥ 0.5 C·cm−2 (green dash line) are shown in panel (c); data for the samples S2 (cyan dash) and S5 (solid black) are given in panel (d). Thickness of the oxide films is calculated as a ratio of q to the experimental value of specific charge density. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 6
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
Fig. 2. SEM images of cross-section of the samples (a) S2, (b) S3, (c) S5, and (d) S4. Dark and light grey arrows in panels (a) and (c) show positions of the layers with higher and lower porosity, respectively. The inset in Fig. 2b shows the lower central part of Fig 2b with higher magnification. SEM images of top (e) and bottom (f) surfaces of the sample S2. (g) Dependence of the thickness of ATO film on the number of anodizing cycles and corresponding cross-sectional SEM image of the sample S5. (h) Scheme of the structure of the anodic titania photonic crystal. (j) Reflectance spectra of the samples S1 (blue solid), S2 (cyan dash), S3 (green dot), and S4 (red dash dot). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
that although the cell boundary oxide is weaker than titania nanotubes walls, it surely exists. Thus, the structure of ATO obtained under considered anodizing conditions is similar to the structure of anodic alumina formed under hard anodizing conditions is sulfuric acid electrolytes [25,26]. Although applied voltage changes more than by factor of √2 during each anodizing cycle, nanotubes grow from the top (Fig. 2e) to the bottom (Fig. 2f) of ATO film without branching (Fig. 2a-d, g) due to constant oscillation of anodizing voltage. Contrary, if the applied voltage reduces by more than a factor of √2 and then keeps constant, the branching of channels is observed [27–29]. Periodic alternation of light and dark layers, corresponding to the regions of ATO with lower and higher porosity, respectively, can be seen in the cross-sectional SEM images of ATO films (Fig. 2a–d). It is worth noting that the porosity modulation is caused by the variation of
voltage cycling interval, demonstrating hysteresis loop on j(U) curve. The width of the hysteresis loop and the values of current densities decrease with the number of cycle. The decrease in j with ATO film thickness was observed earlier [11,12] and was explained by increasing of the diffusion path in the pores for ionic species. Morphology and thickness of the ATO films were characterized using SEM (Fig. 2). The average film thickness of all samples S1–S5 is equal to 17.4 ± 0.9 μm. Thus, thickness-to-charge density ratio is 980 ± 50 nm·cm2·C−1. This value is comparable to the highest thickness-to-charge density ratios (600–960 nm·cm2·C−1) observed in the case of ATO formation in regular NH4F–H2O–EG electrolyte [12,22–24]. It can be clearly seen in cross-sectional images that most of the fracture propagates through the cell boundaries. There are also regions, where cleavage occurred through the pores. The latter proves 7
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
the inner diameter of nanotubes, whereas their outer diameter remains constant across the film thickness (see scheme of ATO porous structure in Fig. 2h). Moreover, the inner diameter of nanotubes changes sinusoidally, similar to the sine-wave profile of the applied voltage U(Q). To the best of our knowledge, this kind of ATO morphology has never been observed earlier and is substantially different from the smooth, corrugated, bamboo-type or concave tubes [12]. To find out the correlation between anodizing voltage and porosity of ATO, in the case of sample S5 additional charge density of 0.213 C·cm−2 was passed at constant voltage of 60 V in each cycle of U (Q) profile, in comparison with the sample S2 (Fig. 1a). As a result, dark layers have become thicker (see dark arrows in Fig. 2c), which proves that the dark layers with larger inner diameter of nanotubes and higher porosity are formed during anodizing at high voltage. According to SEM, the period lengths (175, 210, 265, and 290 nm for the samples S1, S2, S3, and S4, respectively) are constant across the whole ATO films thickness. Fig. 2g demonstrates linear dependence of distance from the top surface of ATO to light layers formed at 40 V on the number of anodizing cycle. It shows that the charge density (q) is the key factor of the control of the ATO thickness. Knowledge of specific charge density, required for the growth of ATO with unit thickness, allows us to plot the dependence of current on film thickness and to obtain new insight into the regime of ATO formation under cyclic anodizing conditions. After initial stage of pore nucleation and rearrangement of porous structure to the steady-state parameters (q < 0.5 C·cm−2), in the case of the samples obtained at constant anodizing voltage of 40 and 60 V, the dependence of j−1 increases linearly with q or with porous film thickness (l) (see orange solid and violet dot lines in Figs. 1c, d). This behaviour proves that diffusion of ionic species in the pores of ATO is the limiting stage under considered anodizing conditions [30,31]. The slope of j−1(l) is inversely proportional to the difference between concentrations of limiting species (ΔC) at pore base and pore mouth. The slope of j−1(l) curve decreases > 20 times from 13.6 to 0.66 A−1·cm2·μm−1 when anodizing voltage changes from 40 to 60 V. Such a big difference could not be caused solely by the change of porosity and could not be observed if the diffusion of reagents from bulk of electrolyte to the pore bases limits the anodizing current. Thus, under applied conditions the limiting stage is the diffusion of reaction products and the reason of the increase in slope of j−1(l) dependences with anodizing voltage is the concentration rise of [TiF6]2− (the product of barrier layer dissolution [32]) at the pore bases. Moreover, the rise of the current testifies that the rate of barrier layer dissolution also increases with the anodizing voltage. In the case of the stationary process, the simultaneous increase in the concentration of reaction product and the rate of the reaction could be only the result of the increase in the reaction constant. The latter is possibly caused by the rise of temperature at the pore bases. Although voltage modulates periodically during the formation of ATO photonic crystals, j−1(l) at 40 V (see maximа of j−1 curves for the samples S1 and S4 in Fig. 1c) is linear with the slope of 21 A−1·cm2·μm−1, which is greater by factor of ~1.5 than in the case of the potentiostatic anodizing of Ti at 40 V. Such a difference is probably caused by the accumulation of [TiF6]2− in the pore after anodizing at higher voltages (> 40 V) and, as a consequence, the decrease in concentration gradient at the pore base. Comparison of j−1(l) plots for the samples S2 and S5 (see Fig. 1d) shows that longer anodizing at higher voltage leads to the decrease in j at 40 V and the increase in the slope to 33 A−1·cm2·μm−1. This observation supports our assumption about the variation of the reaction constant of the barrier layer dissolution with the anodizing voltage. The diffraction of visible light on the layered structure of ATO films results in iridescence of oxide coatings. Well-defined maximum on the reflectance spectra of the samples S1–S4 (Fig. 2j) corresponds to the first stop band of one-dimensional ATO photonic crystals. The red shift of the positions of stop bands with the increase in charge spent for one anodizing cycle is clearly seen, as predicted by Bragg–Snell law [33]:
2 λ = 2d neff − sin2 θ ,
(2)
where λ is the wavelength of the first stop band, d is the period length, θ is the angle of incidence, neff is the effective refractive index of ATO film. Taking account of the lengths of structure periods measured by SEM, according to Eq. (2), the effective refractive index of ATO films S1–S4 lies in the range of 1.30–1.37. 4. Conclusions New anodizing regime with voltage versus electric charge modulation U(Q) was successfully applied for the preparation of one-dimensional photonic crystals based on anodic titania. In situ measuring of electric charge allows one to control the thickness of the formed layers with high precision, whereas control of voltage guarantees the identity of voltage-dependent parameters of the ATO structure for each layer from the top to the bottom. A new type of anodic titania structures, possessing periodic change of inner diameter and constant outer diameter of nanotubes, were obtained at 40–60 V at 30 °C in NaAc–NH4F–H2O–EG electrolyte with sine wave profile of the voltage versus electric charge. The thickness of anodic titania is proportional to the charge density with the coefficient of 980 ± 50 nm·cm2·C−1. The variation of electric charge spent during each cycle of anodizing allowed us to prepare anodic titania photonic crystals with different structure periods (175–290 nm) and, as a consequence, with tuned photonic stop band position from 460 to 780 nm. The proposed method is universal and could be applied for anodizing of other valve metals and their alloys. Acknowledgments This work is supported by the Russian Science Foundation under grant No. 17-73-10471. The experiments were carried out using the scientific equipment purchased by M.V. Lomonosov Moscow State University Program of Development. References [1] J.R. DeVore, Refractive indices of rutile and sphalerite, JOSA 41 (1951) 416–419, http://dx.doi.org/10.1364/JOSA.41.000416. [2] V. Zwilling, E. Darque-Ceretti, A. Boutry-Forveille, D. David, M.Y. Perrin, M. Aucouturier, Structure and physicochemistry of anodic oxide films on titanium and TA6V alloy, Surf. Interface Anal. 27 (1999) 629–637, http://dx.doi.org/10. 1002/(SICI)1096-9918(199907)27:7<629::AID-SIA551>3.0.CO;2-0. [3] J.M. Macak, H. Tsuchiya, L. Taveira, S. Aldabergerova, P. Schmuki, Smooth anodic TiO2 nanotubes, Angew. Chem. Int. Ed. 44 (2005) 7463–7465, http://dx.doi.org/ 10.1002/anie.200502781. [4] L.V. Taveira, J.M. Macak, K. Sirotna, L.F.P. Dick, P. Schmuki, Voltage oscillations and morphology during the galvanostatic formation of self-organized TiO2 nanotubes, J. Electrochem. Soc. 153 (2006) B137–B143, http://dx.doi.org/10.1149/1. 2172566. [5] B. Wang, G.T. Fei, M. Wang, M.G. Kong, L.D. Zhang, Preparation of photonic crystals made of air pores in anodic alumina, Nanotechnology 18 (2007) 365601, http://dx.doi.org/10.1088/0957-4484/18/36/365601. [6] D.-L. Guo, L.-X. Fan, F.-H. Wang, S.-Y. Huang, X.-W. Zou, Porous anodic aluminum oxide bragg stacks as chemical sensors, J. Phys. Chem. C 112 (2008) 17952–17956, http://dx.doi.org/10.1021/jp806926f. [7] A. Santos, J.H. Yoo, C.V. Rohatgi, T. Kumeria, Y. Wang, D. Losic, Realisation and advanced engineering of true optical rugate filters based on nanoporous anodic alumina by sinusoidal pulse anodisation, Nanoscale 8 (2016) 1360–1373, http://dx. doi.org/10.1039/C5NR05462A. [8] S.E. Kushnir, K.S. Napolskii, Thickness-dependent iridescence of one-dimensional photonic crystals based on anodic alumina, Mater. Des. 144 (2018) 140–150, http://dx.doi.org/10.1016/j.matdes.2018.02.012. [9] S.P. Albu, D. Kim, P. Schmuki, Growth of aligned TiO2 bamboo-type nanotubes and highly ordered nanolace, Angew. Chem. Int. Ed. 47 (2008) 1916–1919, http://dx. doi.org/10.1002/anie.200704144. [10] D. Kim, A. Ghicov, S.P. Albu, P. Schmuki, Bamboo-type TiO2 nanotubes: improved conversion efficiency in dye-sensitized solar cells, J. Am. Chem. Soc. 130 (2008) 16454–16455, http://dx.doi.org/10.1021/ja805201v. [11] Y.-L. Xie, Z.-X. Li, H. Xu, K.-F. Xie, Z.-G. Xu, H.-L. Zhang, Fabrication of TiO2 nanotubes with extended periodical morphology by alternating-current anodization, Electrochem. Commun. 17 (2012) 34–37, http://dx.doi.org/10.1016/j.elecom.
8
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
2012.01.021. [12] J. Lin, K. Liu, X. Chen, Synthesis of periodically structured titania nanotube films and their potential for photonic applications, Small 7 (2011) 1784–1789, http://dx. doi.org/10.1002/smll.201002098. [13] C.T. Yip, H. Huang, L. Zhou, K. Xie, Y. Wang, T. Feng, J. Li, W.Y. Tam, Direct and seamless coupling of TiO2 nanotube photonic crystal to dye-sensitized solar cell: a single-step approach, Adv. Mater. 23 (2011) 5624–5628, http://dx.doi.org/10. 1002/adma.201103591. [14] M. Guo, K. Xie, J. Lin, Z. Yong, C.T. Yip, L. Zhou, Y. Wang, H. Huang, Design and coupling of multifunctional TiO2 nanotube photonic crystal to nanocrystalline titania layer as semi-transparent photoanode for dye-sensitized solar cell, Energy Environ. Sci. 5 (2012) 9881–9888, http://dx.doi.org/10.1039/c2ee22854h. [15] M. Guo, K. Xie, X. Liu, Y. Wang, L. Zhou, H. Huang, A strategy to reduce the angular dependence of a dye-sensitized solar cell by coupling to a TiO2 nanotube photonic crystal, Nanoscale 6 (2014) 13060–13067, http://dx.doi.org/10.1039/ C4NR03712J. [16] K. Xie, M. Guo, X. Liu, H. Huang, Enhanced efficiencies in thin and semi-transparent dye-sensitized solar cells under low photon flux conditions using TiO2 nanotube photonic crystal, J. Power Sources 293 (2015) 170–177, http://dx.doi.org/10. 1016/j.jpowsour.2015.05.025. [17] M. Guo, K. Xie, Y. Wang, L. Zhou, H. Huang, Aperiodic TiO2 nanotube photonic crystal: full-visible-spectrum solar light harvesting in photovoltaic devices, Sci. Rep. 4 (2014) 6442, , http://dx.doi.org/10.1038/srep06442. [18] L. Zheng, H. Cheng, F. Liang, S. Shu, C.K. Tsang, H. Li, S.-T. Lee, Y.Y. Li, Porous TiO2 photonic band gap materials by anodization, J. Phys. Chem. C 116 (2012) 5509–5515, http://dx.doi.org/10.1021/jp212416c. [19] L. Zheng, Y. Dong, H. Bian, C. Lee, J. Lu, Y.Y. Li, Self-ordered nanotubular TiO2 multilayers for high-performance photocatalysts and supercapacitors, Electrochim. Acta 203 (2016) 257–264, http://dx.doi.org/10.1016/j.electacta.2016.04.049. [20] G. Wang, J. Wang, Y. An, C. Wang, Anodization fabrication of 3D TiO2 photonic crystals and their application for chemical sensors, Superlattice. Microst. 100 (2016) 290–295, http://dx.doi.org/10.1016/j.spmi.2016.09.027. [21] W.-T. Kim, W.-Y. Choi, Fabrication of TiO2 photonic crystal by anodic oxidation and their optical sensing properties, Sensors Actuators A Phys. 260 (2017) 178–184, http://dx.doi.org/10.1016/j.sna.2017.04.039. [22] S.P. Albu, N. Taccardi, I. Paramasivam, K.R. Hebert, P. Schmuki, Oxide growth
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
9
efficiencies and self-organization of TiO2 nanotubes, J. Electrochem. Soc. 159 (2012) H697–H703, http://dx.doi.org/10.1149/2.015208jes. Y. Zhang, H. Fan, X. Ding, Q. Yan, L. Wang, W. Ma, Simulation of anodizing currenttime curves and morphology evolution of TiO2 nanotubes anodized in electrolytes with different NH4F concentrations, Electrochim. Acta 176 (2015) 1083–1091, http://dx.doi.org/10.1016/j.electacta.2015.07.110. F.J.Q. Cortes, P.J. Arias-Monje, J. Phillips, H. Zea, Empirical kinetics for the growth of titania nanotube arrays by potentiostatic anodization in ethylene glycol, Mater. Des. 96 (2016) 80–89, http://dx.doi.org/10.1016/j.matdes.2016.02.006. D.J. Arrowsmith, A.W. Clifford, D.A. Moth, Fracture of anodic oxide formed on aluminium in sulphuric acid, J. Mater. Sci. Lett. 5 (1986) 921–922, http://dx.doi. org/10.1007/BF01729276. W. Lee, K. Schwirn, M. Steinhart, E. Pippel, R. Scholz, U. Gösele, Structural engineering of nanoporous anodic aluminium oxide by pulse anodization of aluminium, Nat. Nanotechnol. 3 (2008) 234, http://dx.doi.org/10.1038/nnano.2008.54. G. Meng, Y.J. Jung, A. Cao, R. Vajtai, P.M. Ajayan, Controlled fabrication of hierarchically branched nanopores, nanotubes, and nanowires, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 7074–7078, http://dx.doi.org/10.1073/pnas.0502098102. D.I. Petukhov, K.S. Napolskii, A.A. Eliseev, Permeability of anodic alumina membranes with branched channels, Nanotechnology 23 (2012) 335601, , http://dx.doi. org/10.1088/0957-4484/23/33/335601. B. Chen, K. Lu, Hierarchically branched titania nanotubes with tailored diameters and branch numbers, Langmuir 28 (2012) 2937–2943, http://dx.doi.org/10.1021/ la204154h. K. Yasuda, P. Schmuki, Control of morphology and composition of self-organized zirconium titanate nanotubes formed in (NH4)2SO4/NH4F electrolytes, Electrochim. Acta 52 (2007) 4053–4061, http://dx.doi.org/10.1016/j.electacta.2006.11.023. I.V. Roslyakov, E.O. Gordeeva, K.S. Napolskii, Role of electrode reaction kinetics in self-ordering of porous anodic alumina, Electrochim. Acta 241 (2017) 362–369, http://dx.doi.org/10.1016/j.electacta.2017.04.140. J.M. Macak, H. Hildebrand, U. Marten-Jahns, P. Schmuki, Mechanistic aspects and growth of large diameter self-organized TiO2 nanotubes, J. Electroanal. Chem. 621 (2008) 254–266, http://dx.doi.org/10.1016/j.jelechem.2008.01.005. G.A. Ozin, A. Arsenault, L. Cademartiri, Nanochemistry: a chemical approach to nanomaterials, (2008) 978-1-84755-895-4, p. 463 (see details on http://pubs.rsc. org/en/content/ebook/9781847558954 ).
Contents lists available at ScienceDirect
Electrochemistry Communications journal homepage: www.elsevier.com/locate/elecom
Anodic titanium oxide photonic crystals prepared by novel cyclic anodizing with voltage versus charge modulation
T
⁎
N.A. Sapoletovaa, S.E. Kushnira,b, , K.S. Napolskiia,b a b
Department of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia Department of Materials Science, Lomonosov Moscow State University, 119991 Moscow, Russia
A R T I C LE I N FO
A B S T R A C T
Keywords: Anodic titania Anodizing Photonic crystals Porous structure Film
Photonic crystals based on titania are of great practical importance in modern photonics owing to a variety of possibilities for their applications in optoelectronics, sensorics, and solar photovoltaics. However, the reproducible, scalable, and low-cost method of the preparation of titania photonic crystals with desired optical characteristics is still absent. Here the novel anodizing regime with voltage versus electric charge modulation, U (Q), is suggested for the precise morphology control of anodic porous films of valve metals oxides. The potential of the suggested approach is demonstrated on the preparation of high-quality one-dimensional titania photonic crystals. A new type of anodic titania nanotubes possessing periodic change of inner diameter and constant outer diameter is prepared using sine-wave modulation of applied voltage with electric charge. The possibility to tune the position of photonic band gap within the whole visible spectrum is shown.
1. Introduction Photonic crystals made of titania attract great attention owing to perspectives for practical applications in optoelectronics, sensorics, and solar photovoltaics. They are caused by the synergy of anomalous dispersion in photonic crystals and of unique properties of titania, such as a high refractive index (nTiO2 = 2.6 at λ = 600 nm [1]), high chemical stability, low toxicity, semiconductor conductivity, and electrochromism. One of the promising methods of the preparation of titania with controlled nanomorphology is titanium anodizing in pore-forming electrolytes [2–4]. In the case of valve metal anodizing under periodically changing conditions one can obtain porous oxide film with periodic modulation of porosity across the thickness [5–8]. This kind of anodic oxide films can be considered as one-dimensional photonic crystals because the effective refractive index changes periodically along the surface normal. The first results on anodizing of titanium at periodically alternating voltage were reported in 2008 [9]. It has been shown that obtained under square-wave voltage modulation anodic titanium oxide (ATO) films, consisted of bamboo-like titania nanotubes, show higher efficiency of solar energy conversion in dye-sensibilized solar cells in comparison with titania nanotubes with smooth walls [10]. It is worth noting that in the case of periodic variation of voltage with time, the accurate periodicity of porosity modulation in ATO cannot be achieved.
⁎
The layer thickness formed during one cycle of anodic oxidation decreases gradually from cycle to cycle due to the decrease in current with the increase in porous film thickness [11]. In 2011, ATO photonic crystals with well-defined photonic band gaps were obtained under periodically alternating current conditions [12]. The resulting structure of porous titania contained tubes with periodically tapering and expanding of outer diameter and was therefore called “concave”. Later, the alternating-current anodizing of titanium has been successfully applied for the preparation of anodic titania films with extended periodical morphology used as one of the components of dye-sensitized solar cells [11,13–17]. Several complex timedependencies of periodic voltage pulses have been proposed for the preparation of porous multilayer films [18,19] and structures similar to the “concave” type ATO [20]. The positions of the photonic band gaps depend on the composition of the medium inside the pores of ATO, that makes possible to use ATO photonic crystals as refractive index sensors [20,21]. Moreover, the optical response of anodic titania photonic crystals can be reproducibly switched on/off by an anodic/cathodic polarization [18]. This effect can be used to create smart color windows or electronic displays. To the best of our knowledge, all anodizing regimes, which have been previously applied to prepare one-dimensional ATO photonic crystals, used either time-periodic voltage or current profiles. However, as was previously noted, titanium anodizing with periodic U(t) profile leads to different layer thickness in the upper and the lower parts of the
Corresponding author at: Department of Chemistry, Lomonosov Moscow State University, 119991 Moscow, Russia. E-mail address: [email protected] (S.E. Kushnir).
https://doi.org/10.1016/j.elecom.2018.04.018 Received 22 March 2018; Received in revised form 6 April 2018; Accepted 23 April 2018 Available online 24 April 2018 1388-2481/ © 2018 Elsevier B.V. All rights reserved.
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
film. On the other hand, in the case of periodic I(t) profile, anodizing voltage increases from cycle-to-cycle that also break the optical periodicity. Here, we propose a new anodizing regime with voltage versus electric charge modulation U(Q). In situ measuring of charge allows us to control the thickness of the layers with high precision, whereas control of voltage guarantees the identity of voltage-dependent parameters of the ATO structure for each layer from the top to the bottom. The ability to design the morphology of anodic oxide films using the suggested approach has been demonstrated on the example of the synthesis of one-dimensional photonic crystals.
profile. 3) The steps described in item #2 were repeated until the charge spent during anodizing became higher than the maximum value of charge in U(Q) profile. Finally, DC power supply was switched off. Ti was anodized at the voltages in the range from 40 to 60 V using sine wave U(Q) profiles:
2πQ π U = 50 + 10⋅sin ⎛ − ⎞, 2⎠ ⎝ P
(1)
where P is the electric charge spent during one cycle of anodizing. Four samples with charge density per cycle of 0.177, 0.213, 0.256, and 0.295 C·cm−2 were obtained (denoted as S1, S2, S3, and S4, respectively). The total charge density spent during anodizing of each sample S1–S4 was 17.7 C·cm−2. Sample S5 with charge density per cycle of 0.426 C·cm−2 was prepared similar to the sample S2 with the same sine-wave increase and decrease in voltage in U(Q) profile, but with the additional constant voltage anodizing at 60 V for 0.213 C·cm−2 in each cycle (Fig. 1a). The anodized area for all samples was 1.13 cm2. After the end of anodizing the porous oxide films were washed in ethanol and then dried in air. The morphology characterization of ATO films was performed by scanning electron microscopy (SEM) using LEO Supra 50 VP instrument. Perkin Elmer Lambda 950 spectrophotometer was used to record reflectance spectra at incident angle of 8°. The size of the light spot was 3 × 3 mm2.
2. Materials and methods Before anodizing, the titanium foils (99.6% purity, 0.15 mm thick) were electrochemically polished in the mixture of 99.5 wt% acetic acid and 65 wt% perchloric acid with a volume ratio of 9:1. Electropolishing was conducted at temperature below 25 °C during 4 min under squarewave applied voltage: 40 V for 10 s and 60 V for 10 s. Ti was anodized in an ethylene glycol (EG) electrolyte containing 0.3 wt% NH4F, 0.66 wt% CH3COONa (NaAc), and 2 wt% H2O at 30 °C using programmable DC power supply. The electrolyte was agitated at the rate of 480 RPM using overhead stirrer. The distance between Ti foil and Ti counter electrode was 2 cm in all experiments. To create periodic modulation of porosity across thickness of anodic titania film, a specific voltage profile was set as a staircase function of electric charge spent during anodizing using lab-made software. U(Q) anodizing profile was realized using the following algorithm:
3. Results and discussion
1) At the initial stage (Q = 0 C), the anodizing voltage was set to 40 V. 2) After a short time interval Δt (ca. 30 ms), the increment of charge ΔQ was calculated by numerical integration of measured current. New value of charge Q(t + Δt) is the sum of the charge value Q(t) at previous step and the increment ΔQ. Then the voltage was set to the new value U(Q(t + Δt)), which corresponded to the value of charge spent during anodizing in accordance with required voltage – charge
Electrochemical responses recorded during anodizing of titanium under cyclic voltage versus charge modulation are shown in Fig. 1. It can be clearly seen that the voltage applied to the electrochemical cell coincides well with the desired sine-wave U(Q) anodizing profile (Fig. 1a). The periodic variation of anodizing voltage between 40 and 60 V leads to the cyclic change of current density (Fig. 1b). Current density, j, hits the maximum and minimum values at the edges of the
Fig. 1. Electrochemical responses recorded during anodizing of titanium. (a) Applied voltage versus charge modulation U(Q) of the samples S1 (blue solid), S2 (cyan dash), S3 (green dot), S4 (red dash dot), and S5 (black solid). (b) Dependences of current density on voltage with different cycle number for the sample S2: 3 (black solid), 25 (red dash), 50 (green dot), 75 (blue dash dot). Inverse of current density (j−1) versus charge density (q) is plotted in panels (c) and (d): data for the cases of potentiostatic anodizing at 40 V (orange solid) and 60 V (violet dot). Data for the samples S1 (blue solid), S4 (red dash dot) and linear approximation of j−1 maxima for the sample S4 in the case of q ≥ 0.5 C·cm−2 (green dash line) are shown in panel (c); data for the samples S2 (cyan dash) and S5 (solid black) are given in panel (d). Thickness of the oxide films is calculated as a ratio of q to the experimental value of specific charge density. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 6
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
Fig. 2. SEM images of cross-section of the samples (a) S2, (b) S3, (c) S5, and (d) S4. Dark and light grey arrows in panels (a) and (c) show positions of the layers with higher and lower porosity, respectively. The inset in Fig. 2b shows the lower central part of Fig 2b with higher magnification. SEM images of top (e) and bottom (f) surfaces of the sample S2. (g) Dependence of the thickness of ATO film on the number of anodizing cycles and corresponding cross-sectional SEM image of the sample S5. (h) Scheme of the structure of the anodic titania photonic crystal. (j) Reflectance spectra of the samples S1 (blue solid), S2 (cyan dash), S3 (green dot), and S4 (red dash dot). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
that although the cell boundary oxide is weaker than titania nanotubes walls, it surely exists. Thus, the structure of ATO obtained under considered anodizing conditions is similar to the structure of anodic alumina formed under hard anodizing conditions is sulfuric acid electrolytes [25,26]. Although applied voltage changes more than by factor of √2 during each anodizing cycle, nanotubes grow from the top (Fig. 2e) to the bottom (Fig. 2f) of ATO film without branching (Fig. 2a-d, g) due to constant oscillation of anodizing voltage. Contrary, if the applied voltage reduces by more than a factor of √2 and then keeps constant, the branching of channels is observed [27–29]. Periodic alternation of light and dark layers, corresponding to the regions of ATO with lower and higher porosity, respectively, can be seen in the cross-sectional SEM images of ATO films (Fig. 2a–d). It is worth noting that the porosity modulation is caused by the variation of
voltage cycling interval, demonstrating hysteresis loop on j(U) curve. The width of the hysteresis loop and the values of current densities decrease with the number of cycle. The decrease in j with ATO film thickness was observed earlier [11,12] and was explained by increasing of the diffusion path in the pores for ionic species. Morphology and thickness of the ATO films were characterized using SEM (Fig. 2). The average film thickness of all samples S1–S5 is equal to 17.4 ± 0.9 μm. Thus, thickness-to-charge density ratio is 980 ± 50 nm·cm2·C−1. This value is comparable to the highest thickness-to-charge density ratios (600–960 nm·cm2·C−1) observed in the case of ATO formation in regular NH4F–H2O–EG electrolyte [12,22–24]. It can be clearly seen in cross-sectional images that most of the fracture propagates through the cell boundaries. There are also regions, where cleavage occurred through the pores. The latter proves 7
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
the inner diameter of nanotubes, whereas their outer diameter remains constant across the film thickness (see scheme of ATO porous structure in Fig. 2h). Moreover, the inner diameter of nanotubes changes sinusoidally, similar to the sine-wave profile of the applied voltage U(Q). To the best of our knowledge, this kind of ATO morphology has never been observed earlier and is substantially different from the smooth, corrugated, bamboo-type or concave tubes [12]. To find out the correlation between anodizing voltage and porosity of ATO, in the case of sample S5 additional charge density of 0.213 C·cm−2 was passed at constant voltage of 60 V in each cycle of U (Q) profile, in comparison with the sample S2 (Fig. 1a). As a result, dark layers have become thicker (see dark arrows in Fig. 2c), which proves that the dark layers with larger inner diameter of nanotubes and higher porosity are formed during anodizing at high voltage. According to SEM, the period lengths (175, 210, 265, and 290 nm for the samples S1, S2, S3, and S4, respectively) are constant across the whole ATO films thickness. Fig. 2g demonstrates linear dependence of distance from the top surface of ATO to light layers formed at 40 V on the number of anodizing cycle. It shows that the charge density (q) is the key factor of the control of the ATO thickness. Knowledge of specific charge density, required for the growth of ATO with unit thickness, allows us to plot the dependence of current on film thickness and to obtain new insight into the regime of ATO formation under cyclic anodizing conditions. After initial stage of pore nucleation and rearrangement of porous structure to the steady-state parameters (q < 0.5 C·cm−2), in the case of the samples obtained at constant anodizing voltage of 40 and 60 V, the dependence of j−1 increases linearly with q or with porous film thickness (l) (see orange solid and violet dot lines in Figs. 1c, d). This behaviour proves that diffusion of ionic species in the pores of ATO is the limiting stage under considered anodizing conditions [30,31]. The slope of j−1(l) is inversely proportional to the difference between concentrations of limiting species (ΔC) at pore base and pore mouth. The slope of j−1(l) curve decreases > 20 times from 13.6 to 0.66 A−1·cm2·μm−1 when anodizing voltage changes from 40 to 60 V. Such a big difference could not be caused solely by the change of porosity and could not be observed if the diffusion of reagents from bulk of electrolyte to the pore bases limits the anodizing current. Thus, under applied conditions the limiting stage is the diffusion of reaction products and the reason of the increase in slope of j−1(l) dependences with anodizing voltage is the concentration rise of [TiF6]2− (the product of barrier layer dissolution [32]) at the pore bases. Moreover, the rise of the current testifies that the rate of barrier layer dissolution also increases with the anodizing voltage. In the case of the stationary process, the simultaneous increase in the concentration of reaction product and the rate of the reaction could be only the result of the increase in the reaction constant. The latter is possibly caused by the rise of temperature at the pore bases. Although voltage modulates periodically during the formation of ATO photonic crystals, j−1(l) at 40 V (see maximа of j−1 curves for the samples S1 and S4 in Fig. 1c) is linear with the slope of 21 A−1·cm2·μm−1, which is greater by factor of ~1.5 than in the case of the potentiostatic anodizing of Ti at 40 V. Such a difference is probably caused by the accumulation of [TiF6]2− in the pore after anodizing at higher voltages (> 40 V) and, as a consequence, the decrease in concentration gradient at the pore base. Comparison of j−1(l) plots for the samples S2 and S5 (see Fig. 1d) shows that longer anodizing at higher voltage leads to the decrease in j at 40 V and the increase in the slope to 33 A−1·cm2·μm−1. This observation supports our assumption about the variation of the reaction constant of the barrier layer dissolution with the anodizing voltage. The diffraction of visible light on the layered structure of ATO films results in iridescence of oxide coatings. Well-defined maximum on the reflectance spectra of the samples S1–S4 (Fig. 2j) corresponds to the first stop band of one-dimensional ATO photonic crystals. The red shift of the positions of stop bands with the increase in charge spent for one anodizing cycle is clearly seen, as predicted by Bragg–Snell law [33]:
2 λ = 2d neff − sin2 θ ,
(2)
where λ is the wavelength of the first stop band, d is the period length, θ is the angle of incidence, neff is the effective refractive index of ATO film. Taking account of the lengths of structure periods measured by SEM, according to Eq. (2), the effective refractive index of ATO films S1–S4 lies in the range of 1.30–1.37. 4. Conclusions New anodizing regime with voltage versus electric charge modulation U(Q) was successfully applied for the preparation of one-dimensional photonic crystals based on anodic titania. In situ measuring of electric charge allows one to control the thickness of the formed layers with high precision, whereas control of voltage guarantees the identity of voltage-dependent parameters of the ATO structure for each layer from the top to the bottom. A new type of anodic titania structures, possessing periodic change of inner diameter and constant outer diameter of nanotubes, were obtained at 40–60 V at 30 °C in NaAc–NH4F–H2O–EG electrolyte with sine wave profile of the voltage versus electric charge. The thickness of anodic titania is proportional to the charge density with the coefficient of 980 ± 50 nm·cm2·C−1. The variation of electric charge spent during each cycle of anodizing allowed us to prepare anodic titania photonic crystals with different structure periods (175–290 nm) and, as a consequence, with tuned photonic stop band position from 460 to 780 nm. The proposed method is universal and could be applied for anodizing of other valve metals and their alloys. Acknowledgments This work is supported by the Russian Science Foundation under grant No. 17-73-10471. The experiments were carried out using the scientific equipment purchased by M.V. Lomonosov Moscow State University Program of Development. References [1] J.R. DeVore, Refractive indices of rutile and sphalerite, JOSA 41 (1951) 416–419, http://dx.doi.org/10.1364/JOSA.41.000416. [2] V. Zwilling, E. Darque-Ceretti, A. Boutry-Forveille, D. David, M.Y. Perrin, M. Aucouturier, Structure and physicochemistry of anodic oxide films on titanium and TA6V alloy, Surf. Interface Anal. 27 (1999) 629–637, http://dx.doi.org/10. 1002/(SICI)1096-9918(199907)27:7<629::AID-SIA551>3.0.CO;2-0. [3] J.M. Macak, H. Tsuchiya, L. Taveira, S. Aldabergerova, P. Schmuki, Smooth anodic TiO2 nanotubes, Angew. Chem. Int. Ed. 44 (2005) 7463–7465, http://dx.doi.org/ 10.1002/anie.200502781. [4] L.V. Taveira, J.M. Macak, K. Sirotna, L.F.P. Dick, P. Schmuki, Voltage oscillations and morphology during the galvanostatic formation of self-organized TiO2 nanotubes, J. Electrochem. Soc. 153 (2006) B137–B143, http://dx.doi.org/10.1149/1. 2172566. [5] B. Wang, G.T. Fei, M. Wang, M.G. Kong, L.D. Zhang, Preparation of photonic crystals made of air pores in anodic alumina, Nanotechnology 18 (2007) 365601, http://dx.doi.org/10.1088/0957-4484/18/36/365601. [6] D.-L. Guo, L.-X. Fan, F.-H. Wang, S.-Y. Huang, X.-W. Zou, Porous anodic aluminum oxide bragg stacks as chemical sensors, J. Phys. Chem. C 112 (2008) 17952–17956, http://dx.doi.org/10.1021/jp806926f. [7] A. Santos, J.H. Yoo, C.V. Rohatgi, T. Kumeria, Y. Wang, D. Losic, Realisation and advanced engineering of true optical rugate filters based on nanoporous anodic alumina by sinusoidal pulse anodisation, Nanoscale 8 (2016) 1360–1373, http://dx. doi.org/10.1039/C5NR05462A. [8] S.E. Kushnir, K.S. Napolskii, Thickness-dependent iridescence of one-dimensional photonic crystals based on anodic alumina, Mater. Des. 144 (2018) 140–150, http://dx.doi.org/10.1016/j.matdes.2018.02.012. [9] S.P. Albu, D. Kim, P. Schmuki, Growth of aligned TiO2 bamboo-type nanotubes and highly ordered nanolace, Angew. Chem. Int. Ed. 47 (2008) 1916–1919, http://dx. doi.org/10.1002/anie.200704144. [10] D. Kim, A. Ghicov, S.P. Albu, P. Schmuki, Bamboo-type TiO2 nanotubes: improved conversion efficiency in dye-sensitized solar cells, J. Am. Chem. Soc. 130 (2008) 16454–16455, http://dx.doi.org/10.1021/ja805201v. [11] Y.-L. Xie, Z.-X. Li, H. Xu, K.-F. Xie, Z.-G. Xu, H.-L. Zhang, Fabrication of TiO2 nanotubes with extended periodical morphology by alternating-current anodization, Electrochem. Commun. 17 (2012) 34–37, http://dx.doi.org/10.1016/j.elecom.
8
Electrochemistry Communications 91 (2018) 5–9
N.A. Sapoletova et al.
2012.01.021. [12] J. Lin, K. Liu, X. Chen, Synthesis of periodically structured titania nanotube films and their potential for photonic applications, Small 7 (2011) 1784–1789, http://dx. doi.org/10.1002/smll.201002098. [13] C.T. Yip, H. Huang, L. Zhou, K. Xie, Y. Wang, T. Feng, J. Li, W.Y. Tam, Direct and seamless coupling of TiO2 nanotube photonic crystal to dye-sensitized solar cell: a single-step approach, Adv. Mater. 23 (2011) 5624–5628, http://dx.doi.org/10. 1002/adma.201103591. [14] M. Guo, K. Xie, J. Lin, Z. Yong, C.T. Yip, L. Zhou, Y. Wang, H. Huang, Design and coupling of multifunctional TiO2 nanotube photonic crystal to nanocrystalline titania layer as semi-transparent photoanode for dye-sensitized solar cell, Energy Environ. Sci. 5 (2012) 9881–9888, http://dx.doi.org/10.1039/c2ee22854h. [15] M. Guo, K. Xie, X. Liu, Y. Wang, L. Zhou, H. Huang, A strategy to reduce the angular dependence of a dye-sensitized solar cell by coupling to a TiO2 nanotube photonic crystal, Nanoscale 6 (2014) 13060–13067, http://dx.doi.org/10.1039/ C4NR03712J. [16] K. Xie, M. Guo, X. Liu, H. Huang, Enhanced efficiencies in thin and semi-transparent dye-sensitized solar cells under low photon flux conditions using TiO2 nanotube photonic crystal, J. Power Sources 293 (2015) 170–177, http://dx.doi.org/10. 1016/j.jpowsour.2015.05.025. [17] M. Guo, K. Xie, Y. Wang, L. Zhou, H. Huang, Aperiodic TiO2 nanotube photonic crystal: full-visible-spectrum solar light harvesting in photovoltaic devices, Sci. Rep. 4 (2014) 6442, , http://dx.doi.org/10.1038/srep06442. [18] L. Zheng, H. Cheng, F. Liang, S. Shu, C.K. Tsang, H. Li, S.-T. Lee, Y.Y. Li, Porous TiO2 photonic band gap materials by anodization, J. Phys. Chem. C 116 (2012) 5509–5515, http://dx.doi.org/10.1021/jp212416c. [19] L. Zheng, Y. Dong, H. Bian, C. Lee, J. Lu, Y.Y. Li, Self-ordered nanotubular TiO2 multilayers for high-performance photocatalysts and supercapacitors, Electrochim. Acta 203 (2016) 257–264, http://dx.doi.org/10.1016/j.electacta.2016.04.049. [20] G. Wang, J. Wang, Y. An, C. Wang, Anodization fabrication of 3D TiO2 photonic crystals and their application for chemical sensors, Superlattice. Microst. 100 (2016) 290–295, http://dx.doi.org/10.1016/j.spmi.2016.09.027. [21] W.-T. Kim, W.-Y. Choi, Fabrication of TiO2 photonic crystal by anodic oxidation and their optical sensing properties, Sensors Actuators A Phys. 260 (2017) 178–184, http://dx.doi.org/10.1016/j.sna.2017.04.039. [22] S.P. Albu, N. Taccardi, I. Paramasivam, K.R. Hebert, P. Schmuki, Oxide growth
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
9
efficiencies and self-organization of TiO2 nanotubes, J. Electrochem. Soc. 159 (2012) H697–H703, http://dx.doi.org/10.1149/2.015208jes. Y. Zhang, H. Fan, X. Ding, Q. Yan, L. Wang, W. Ma, Simulation of anodizing currenttime curves and morphology evolution of TiO2 nanotubes anodized in electrolytes with different NH4F concentrations, Electrochim. Acta 176 (2015) 1083–1091, http://dx.doi.org/10.1016/j.electacta.2015.07.110. F.J.Q. Cortes, P.J. Arias-Monje, J. Phillips, H. Zea, Empirical kinetics for the growth of titania nanotube arrays by potentiostatic anodization in ethylene glycol, Mater. Des. 96 (2016) 80–89, http://dx.doi.org/10.1016/j.matdes.2016.02.006. D.J. Arrowsmith, A.W. Clifford, D.A. Moth, Fracture of anodic oxide formed on aluminium in sulphuric acid, J. Mater. Sci. Lett. 5 (1986) 921–922, http://dx.doi. org/10.1007/BF01729276. W. Lee, K. Schwirn, M. Steinhart, E. Pippel, R. Scholz, U. Gösele, Structural engineering of nanoporous anodic aluminium oxide by pulse anodization of aluminium, Nat. Nanotechnol. 3 (2008) 234, http://dx.doi.org/10.1038/nnano.2008.54. G. Meng, Y.J. Jung, A. Cao, R. Vajtai, P.M. Ajayan, Controlled fabrication of hierarchically branched nanopores, nanotubes, and nanowires, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 7074–7078, http://dx.doi.org/10.1073/pnas.0502098102. D.I. Petukhov, K.S. Napolskii, A.A. Eliseev, Permeability of anodic alumina membranes with branched channels, Nanotechnology 23 (2012) 335601, , http://dx.doi. org/10.1088/0957-4484/23/33/335601. B. Chen, K. Lu, Hierarchically branched titania nanotubes with tailored diameters and branch numbers, Langmuir 28 (2012) 2937–2943, http://dx.doi.org/10.1021/ la204154h. K. Yasuda, P. Schmuki, Control of morphology and composition of self-organized zirconium titanate nanotubes formed in (NH4)2SO4/NH4F electrolytes, Electrochim. Acta 52 (2007) 4053–4061, http://dx.doi.org/10.1016/j.electacta.2006.11.023. I.V. Roslyakov, E.O. Gordeeva, K.S. Napolskii, Role of electrode reaction kinetics in self-ordering of porous anodic alumina, Electrochim. Acta 241 (2017) 362–369, http://dx.doi.org/10.1016/j.electacta.2017.04.140. J.M. Macak, H. Hildebrand, U. Marten-Jahns, P. Schmuki, Mechanistic aspects and growth of large diameter self-organized TiO2 nanotubes, J. Electroanal. Chem. 621 (2008) 254–266, http://dx.doi.org/10.1016/j.jelechem.2008.01.005. G.A. Ozin, A. Arsenault, L. Cademartiri, Nanochemistry: a chemical approach to nanomaterials, (2008) 978-1-84755-895-4, p. 463 (see details on http://pubs.rsc. org/en/content/ebook/9781847558954 ).