Electrochemical growth of porous titanium dioxide in a glycerol-based electrolyte at different temperatures

Electrochimica Acta 144 (2014) 127–135

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Electrochemical growth of porous titanium dioxide in a glycerol-based electrolyte at different temperatures Joanna Kapusta-Kołodziej a,1 , Olena Tynkevych b , Anna Pawlik a , Magdalena Jarosz a , Justyna Mech c , Grzegorz D. Sulka a,∗,1 a

Department of Physical Chemistry and Electrochemistry, Jagiellonian University in Krakow, Ingardena 3, 30060 Krakow, Poland Department of Inorganic Chemistry Solid State and Nanomaterials, Yuriy Fedkovych Chernivtsi National University, Lesya Ukrainka Str. 25, 58012 Chernivtsi, Ukraine c Faculty of Non-Ferrous Metals, AGH University of Science and Technology, Al. A. Mickiewicza 30, 30059 Krakow, Poland b

a r t i c l e

i n f o

Article history: Received 13 June 2014 Received in revised form 7 August 2014 Accepted 7 August 2014 Available online 27 August 2014 Keywords: Anodization Porous Titania Nanopores Nanotubes Glycerol

a b s t r a c t Anodic TiO2 layers on Ti were prepared via a three-step anodization in glycerol containing NH4 F (0.38 wt%) and H2 O (1.79 wt%) at the constant anodizing potential of 40 V and temperatures ranging from 10 to 40 ◦ C. It was confirmed that TiO2 growth is thermally activated and transport of oxygen species across the titanium dioxide layer at the pore bottoms is a rate-limiting step of anodization. The morphological characterizations of received anodic layers were performed for all studied temperatures. The structural features of anodic TiO2 such as pore diameter, cell diameter, porosity, pore density and pore circularity were investigated for various anodizing temperatures. It was found that the oxide thickness, pore diameter, cell diameter and porosity of formed TiO2 layers increase gradually with increasing temperature. The pore density and their arrangement analyses showed that the nanoporous anodic titania layer with the lowest density of pores and best pore arrangement was synthesized at 20 ◦ C. In addition, photoelectrochemical properties of formed TiO2 layers were investigated as well. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Nowadays, nanotechnology is considered as a key strategy to improve conventional and develop new technologies by exploring novel nanomaterials and nanoscale processes. Over the last years, the anodic formation of nanostructured TiO2 with ordered, straight and parallel tubes/channels, and a narrow distribution of their diameter has received significant scientific and technological attention. The exploitation of unique functional properties of nanostructured TiO2 holds outstanding promise for the achievement of significant breakthroughs in various areas of applications ranging from photovoltaics and photocatalysis to sensors and deposition of templates for secondary nanomaterials [1–3]. The increasing research interest in these fields has demonstrated that nanostructured TiO2 will play in the future an important role in the environmental and energy sectors. These include protection of

∗ Corresponding author. Department of Physical Chemistry & Electrochemistry, Jagiellonian University, Ingardena 3, 30060 Krakow, Poland; Tel.: +48 12 663 22 6; fax: +48 12 634 05 15. E-mail address: [email protected] (G.D. Sulka). 1 ISE member. http://dx.doi.org/10.1016/j.electacta.2014.08.055 0013-4686/© 2014 Elsevier Ltd. All rights reserved.

the environment and development of renewable and clean energy technologies [1,4]. Indeed, nanoporous/nanotubular anodic TiO2 has been successfully used as an efficient photocatalyst for water splitting [5,6], hydrogen generation [7,8], degradation of pollutants [9,10], and electrocatalytic oxidation of methanol [11,12]. Moreover, other potential technological applications of anodic TiO2 include dyesensitized solar cells [13–15], energy storage systems [15–17], humidity sensors [18], gas sensors [19,20], biosensors [21,22], and biocompatible bone grafting materials [23,24]. Similarly to porous anodic alumina, anodic TiO2 has been recently utilized as a porous template for synthesis of metal nanowires and nanoparticles [25,26], as well as conducting polymer nanowires and nanopore arrays [27,28]. The effectiveness of above mentioned applications depends evidently on the morphology of TiO2 layers. It is therefore not surprising that over the past years, many researchers have focused their studies on the fabrication of nanostructured titania [3]. In 1999, Zwilling et al. [29] first reported fabrication of nanotubular TiO2 through anodization of titanium in a mixture of chromic acid and hydrofluoric acid. Since then, researchers continuously studied and developed this approach to form porous/tubular TiO2 . The first-generation of anodic TiO2 nanotube arrays grown in acidic

128

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

aqueous solutions had a length not exceeding 500 nm, because of a high dissolution rate of oxide in HF-based electrolytes [30]. The second generation of nanotubes were grown in neutral fluoride-based aqueous solutions, where the dissolution of TiO2 was considerably reduced, and thus, the nanotube length increased to approximately 7 ␮m [31,32]. Afterwards, the most crucial improvement to the method was to use fluoride-containing polar organic electrolytes such as formamide [33], dimethylsulfoxide (DMSO) [34,35] or ethylene glycol [36,37]. It was found that the water presence in electrolytes used for anodization of titanium play an appreciable role in the dissolution of formed oxide. Consequently, water as a main constituent of anodizing electrolytes was replaced by organic liquids with low water contents, and accordingly, the thirdgeneration TiO2 nanotube arrays with lengths of up to 1000 ␮m and a smooth surface morphology were achieved in ethylene glycol containing NH4 F and H2 O [38]. It is well known, that the most common electrolyte used today for electrochemical oxidation of Ti is ethylene glycol (ethane-1,2diol) containing fluoride ions. Glycerol (propan-1,2,3-triol) as an anodizing electrolyte is much less often used for the synthesis of nanotubular/nanoporous TiO2 compared to ethylene glycol. So far, research efforts were focused mainly on controlling the morphology of grown TiO2 nanotube/nanopore arrays and various parameters of electrochemical oxidation in glycerol-based electrolytes [39–56]. These include the studies on Ti anodization at various anodization voltages [44–53], electrolyte compositions and different values of pH [42,44,47,48,54,55], different hydrodynamic conditions [55], and anodizing times [44,46,48,53]. As it was shown for other anodizing electrolytes, temperature affects considerably the morphology and geometrical parameters of anodic TiO2 . However, the temperature influence on ATO layers formed in glycerol-based electrolytes has attracted attention only a few researcher groups. The earliest work on the effect of electrolyte temperature on the diameter and length of formed nanotubes has been reported by Macak et al. [40]. They reported self-organized nanotubular ATO layers with high aspect ratios up to 150 formed at the temperatures of 0, 20 and 40 ◦ C. The temperature of 20 ◦ C was recommended as the best one for the nanotube growth. Moreover, it was found that for the glycerol electrolyte containing 0.5 wt% NH4 F, the current efficiency of nanotube formation is close to 100%, which is significantly higher than that reported for water-based solutions. Then, Wang and Lin performed anodizations of titanium in glycerol containing 0.25 wt% NH4 F at room temperature and in an ice bath [56]. It was shown that anodizing temperature markedly affects nanotube dimensions, while the applied voltage does not influence the diameter of nanotubes. Next, Lee at al. [51] reported the fabrication of the self-ordered nanochannel TiO2 structure by anodization in a hot glycerol electrolyte (180 ◦ C) containing 10 wt% K2 HPO4 . Furthermore, the effect of bath temperature on the dimension of fabricated TiO2 nanotube arrays was studied by Wang and Chen [57]. They anodized Ti at 30 V in a mixture of glycerol and water (40 wt%) containing 0.5 wt% NH4 F at temperatures ranging between 10 and 80 ◦ C. The titania nanotubes were observed at the bath temperatures between 40 and 80 ◦ C, while below 40 ◦ C the formation of nanotubes was not detected. They claimed also that the inner diameter of nanotubes increased from 170 to 210 nm as bath temperature increased from 40 to 50 ◦ C. Above 50 ◦ C, comparable nanotube diameters were observed. On the other hand, the average length of nanotubes decreased from 2.10 to 0.94 ␮m with increasing bath temperature, while above 60 ◦ C the formed nanotubes had similar lengths. The listed above works give only limited information about the role of electrolyte temperature in the fabrication of nanoporous/nanotubular anodic TiO2 and many questions and issues are still open to discussion. In this context, the main purpose of this work was to evaluate how the electrolyte temperature

affects the morphology and growth of ATO layers in glycerol containing 0.38 wt% NH4 F and 1.79 wt% H2 O. To the best of our knowledge, the influence of the electrolyte temperature on growth of ATO layers was not studied in glycerol solutions containing less than 2 wt% of water. In addition, photoelectrochemical properties of formed TiO2 films were studied as well. 2. Experimental The titanium foil of 0.25 mm thickness and 99.5% purity from Alfa Aesar was used. Prior to anodizations, Ti substrates (1.5 × 2.5 cm) with a selected working area of 0.8 cm2 were degreased in acetone and ethanol, then dried in the air. The specimens were electrochemically polished at 20 ◦ C in a mixture of acetic acid (99.5 wt%), sulfuric acid (98 wt%), and hydrofluoric acid (40 wt%) (60:15:25 in volume) at a constant current density of 1.4 A cm−2 for 1 min, followed by chemical polishing carried out in a mixture of HF (40 wt%) and nitric acid (65 wt%) (1:3 in volume) for 10 s until mirror finish was observed. The anodization was performed in a two-electrode configuration with the Ti foil as the anode and a platinum grid as the cathode. The electrochemical oxidation process was carried out in a home-made Teflon electrochemical cell (volume of 100 cm3 ), where the anodized samples laid horizontally on the metallic cooled plate. The anodic TiO2 films were prepared via a three-step anodization in a glycerol solution containing NH4 F (0.38 wt%) and H2 O (1.79 wt%) at the constant anodizing potential of 40 V and temperatures ranging from 10 to 40 ◦ C. The duration of the first and second anodizing steps was 3 h. After the first and second anodizations, the grown oxide layers were removed. Then, immediately, the samples were re-anodized typically for 1 h. The resulting thicknesses of the oxide layer were 175, 602, 1434, 2615 nm for the temperatures of 10, 20, 30 and 40 ◦ C, respectively. In case when the fixed 1-␮m thick of ATO layers were synthesized (Fig. S1, supplementary content), the anodizing durations were selected to be 344, 100, 42 and 23 min for the corresponding temperatures 10, 20, 30 and 40 ◦ C. After anodizations, the as-prepared samples were rinsed with water and dried in the air at room temperature. The structural and morphological characterizations of anodic TiO2 were performed for all studied temperatures, and include an extensive study of the ATO morphology by field emission scanning electron microscopy (FE-SEM, Hitachi S-4700) and evaluation of structural features of ATO layers (cell diameter, pore density and regularity of pore arrangement) by using software [58,59]. Before photoelectrochemical measurements, the anodic titania films were calcined (in the air) at 400 ◦ C for 2 h using a muffle furnace (Czylok, FCF5-SHM). The photoelectrochemical characterization was performed using a photoelectric spectrometer (Instytut Fotonowy) equipped with the 150 W xenon lamp and coupled with the EG&G 273A (Princeton Applied Research) potentiostat. The photoelectrochemical experiments were carried out in a Teflon cell with a quartz window (35 × 35 mm) using a three-electrode configuration with a platinum foil as the counter electrode, a HaberLuggin capillary with a saturated calomel electrode (SCE) as the reference electrode and nanoporous TiO2 as the working electrode. The photocurrent vs. time curves were recorded at 0.5 V vs. SCE in a 0.1 M KNO3 aqueous solution under pulsed UV illumination in 5 and 10 s for a light and dark cycles, respectively. A wavelength step of 10 nm was applied in the range of 300–400 nm. 3. Results and discussion Fig. 1 shows the surface morphology, bottom and cross-sectional views of anodic TiO2 layers obtained after 1 h of the third anodization performed in a glycerol solution containing NH4 F (0.38 wt%)

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

129

Fig. 1. SEM top-views (A-D), bottom-views (E-H) and corresponding cross-sectional views (I-L) of ATO layers formed after 1 h of the third step of anodization in the glycerol solution containing NH4 F (0.38 wt%) and H2 O (1.79 wt%) at the constant potential of 40 V and temperatures ranging from 10 to 40 ◦ C.

and H2 O (1.79 wt%) at the constant anodizing potential of 40 V and temperatures ranging from 10 to 40 ◦ C. As we can see the morphology of ATO depends on the electrolyte temperature. The regular nanopore arrays are formed at the temperatures below 30 ◦ C (Figs. 1A and 1B). For higher anodizing temperatures (Figs. 1C and 1D) a complex morphology of ATO with sub-pores is observed as a result of enhanced precipitation of hydrous titanium dioxide on the ATO surface [61]. The higher rate of the oxide growth the more enhanced hydrolysis of Ti4+ ions and precipitation of hydrous titanium dioxide are observed. The similar differences in the morphology of ATO layers were observed for the 1-␮m thick samples (Fig. S1, supplementary content). Furthermore, with increasing electrolyte temperature, the rate of oxide growth increases and the ATO layers become thicker. (Figs. 1I–1L) The temperature of anodizing affects not only the thickness of oxide layer,

but also the pore arrangement and the size of cells. This is especially visible at the pore bottoms (Figs. 1E–1H) where a random distribution of relatively small cells is observed at the temperature of 10 ◦ C. It suggests that the rearrangement of cells is not yet completed. Above 10 ◦ C, the cell diameter and cell arrangement observed at the pore bottoms are approximately the same for all studied electrolyte temperatures (Figs. 1F–1H). The current-time relations recorded during the three step anodization of Ti samples at 40 V and temperatures ranging from 10 to 40 ◦ C are shown in Fig. 2. The current-time curves reveals typical stages of anodic oxide growth, and the three main stages can be identified: (i) an exponential drop due to the formation of a compact and passive TiO2 layer, (ii) a rapid increase due to subsequently transformation of the compact oxide layer into the porous one, and (iii) a slow decrease with time after reaching

130

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

Fig. 2. Current density vs. time plots recorded for the three-step anodization at 40 V and temperatures ranging from 10 to 40 ◦ C. The anodizations were carried out in the glycerol solution containing NH4 F (0.38 wt%) and H2 O (1.79 wt%).

the maximum due to a steady-state formation of porous/tubular oxide [3,60,61]. For a given anodizing temperature, the current density increases in the order: first < second < third anodizing step. Moreover, the time at which the minimum and maximum current densities occur decreases with the increasing number of anodizing step. Consequently, for successive anodizing steps the steady-state conditions for the formation of pores are achieved within a shorter time period and recorded current densities are higher. It indicates that the pore nucleation proceeds much easier on the pre-textured surface. The current density increases also with increasing temperature for a given anodizing step. As a result of that thicker oxide layers are formed at higher temperatures. At 40 ◦ C, the plots are slightly different from those observed for the lower temperatures and a relatively rapid decrease with anodizing time is observed especially for the second and third anodizing steps (Fig. 2D). This is a typical behavior observed for the high rate formation of anodic TiO2 [60]. The recorded plots decrease with time due to mass transfer through the formed oxide layers and, consequently, increased diffusion path length to the oxide/metal interface. Fig. 3A illustrates the average current density measured during the third anodizing step at 40 V for various temperatures. The corresponding growth rates (Rp ) calculated from the cross sectional FE-SEM views of ATO layers are presented in Fig. 3B. The similar exponential trend with increasing temperature is observed for both plots. The rate of oxide growth was found to be about 2.62

and 0.16 ␮m. h−1 for the anodizations performed at 40 and 10 ◦ C, respectively. These values are at least several times lower than those obtained in an ethylene glycol solution. For instance, the rate of oxide growth in the similar glycol-based electrolyte was about 5.1 ␮m. h−1 at 10 ◦ C [60]. In a very viscous electrolyte, such as glycerol, the recorded current density was a few orders of magnitude lower than that in ethylene glycol. As a consequence, a very slow rate of oxide growth is observed in the glycerol-based electrolyte. When we consider the anodic titanium dioxide formation as a pseudo-zero order reaction, the oxide growth rate (Rp ) is proportional to the rate constant (k) of the reaction and the Arrhenius equation can be used for the temperature (T) dependence of the rate of anodic reaction. Rp =

dh ∼k = A exp dt

 −Ea  RT

(1)

where h is the oxide thickness, t is anodizing time, A is the preexponential factor, Ea is the activation energy, and R is the universal gas constant. On the other hand, the thickness of anodic oxide is proportional to the passing current density and, therefore, we can expressed the growth rate by the relationship Rp =

MTiO2 ITO2− (1 − p)z F dTiO2

(2)

where MTiO2 is the molar mass of TiO2 , I is the measured current density, which corresponds primarily to the ionic migration of O2- ,

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

131

Fig. 3. Average current density (A), growth rate of the oxide layer (B), Arrhenius-type plot of anodization average current density I versus 1000/T (C) and Arrhenius-type plot of thickness growth rate Rp versus 1000/T (D) for samples anodized for 1 h in the glycerol solution containing 0. 38 wt.% NH4 F and 1.79 wt. % H2 O at 40 V and temperatures ranging from 10 to 40 ◦ C.

Fig. 4. Temperature influence on the average cell size (A) and pore diameter (B) of ATO layers formed after 1 h of the third step of anodization at 40 V in the glycerol solution containing 0. 38 wt% NH4 F and 1.79 wt % H2 O. Insets: the data for the 1-␮m thick ATO layers.

132

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

Fig. 5. Pore density (A) and regularity ratio (B) of ATO layers obtained after 1 h of the third step of anodization in glycerol solution containing 0. 38 wt% NH4 F and 1.79 wt% H2 O at 40 V and temperatures ranging from 10 to 40 ◦ C. Insets: the data for the 1-␮m thick ATO layers.

TO2− is the transport number of O2- , p is the porosity of oxide layer, z is the number of electrons involved in the reaction, F is the Faraday constant, and dTiO2 the density of TiO2 . Therefore, the Arrhenius relationship should be also observed between the current density and anodizing temperature. Both current density and nanotube growth rate fit well with the Arrhenius equation (Figs. 3C and 3D). It indicates that the anodic process of nanotube growth is thermally activated and activation energy, calculated from the slopes of the linear relations of ln I vs. 1/T and ln Rp vs. 1/T, are 151.6 kJ mol−1 (1.57 eV) and 153.2 kJ mol−1 (1.59 eV), respectively. These values suggest rather a reaction rate-limited process. The effect of temperature on the structural features of anodic TiO2 formed at 40 V in different temperatures is shown in Fig. 4 for samples anodized for 1 h and for the samples with the fixed oxide thickness. Since, the very similar temperature relations were observed for the 1-␮m thick samples to those obtained for ATOs after 1 h of anodization (overlapping of points), these results are presented as insets in Fig. 4. We observed that the average cell size increases from 123 to 154 nm when the electrolyte temperature increases from 10 to 20 ◦ C. Between the temperatures of 20 and 40 ◦ C, a very slight increase or a constant value of the cell diameter are observed. This fact is in good agreement with the bottom views of ATO layers presented in Fig. 1. Similarly, the linear relationship between the temperature and pore diameter can be seen. The average pore diameter increases from 21 to 33 nm when the bath temperature increases from 10 to 40 ◦ C, respectively. For the obtained nanoporous ATO structures the analyses of pore shape were performed (Fig. S2, supplementary content). The circularity of 100% indicates a perfect circle shape of pores, while the value gradually decreasing to 0% indicates an increasingly elongated polygon. Figure S2A shows the average pore circularities calculated for various temperatures. As can be seen, pore circularity decreases from about 83% to 73% for the samples anodized for 1 h and from about 87% to 80% for the 1-␮m thick samples when the anodizing temperature increases. It means that the best pore circularity of ATO is obtained at 10 ◦ C. For all studied temperatures, porosity of ATO, defined as a ratio of the surface area occupied by pores to the whole surface area of the sample, was calculated directly from the FE-SEM images (Fig. S2B, supplementary content). The average porosity increases slowly at the temperature

Fig. 6. Photocurrent dependences on the electrolyte temperature of fabrication of TiO2 photoelectrodes, anodized for 1 h (A) and with the fixed ATO thickness (B), and wavelength of incident light recorded in 0.1 M KNO3 at 0.5 V vs. SCE. The ATO films were calcined in the air at 400 ◦ C for 2 h.

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

133

Fig. 7. IPCE values for ATOs anodized for 1 h (A) and with the fixed oxide thickness (B) at 350 nm.

range from 10 to 30 ◦ C. As the temperature is further increased, the oxide porosity increases to about 10%. Fig. 5A presents the dependence between anodizing temperature and the average pore density, calculated as a total number of pores occupying a given surface area of the FE-SEM image, for samples anodized for 1 h and for the samples with the fixed oxide thickness. Our results show that the average pore densities on the top of ATO layers do not change considerably within the temperature range between 10 and 30 ◦ C, while at 40 ◦ C a significant increase in the pore density is observed for both sample types: anodized for 1 h and with the fixed ATO thickness. We also conducted a quantitative analysis of the pore arrangement regularity on the basis of FFT profiles of FE-SEM images as described previously [60,61]. A regularity ratio (R), defined as a ratio of the maximum intensity of the FFT peak to its width at half-maximum was calculated from the average FFT profile for various anodizing temperatures. Fig. 5B shows the average regularity ratios calculated from FFT profiles for various anodizing temperatures. Independently of the sets of analyzed samples (with the fixed anodizing time or with the fixed ATO thickness), the best pore arrangement is observed for nanoporous ATO structure formed at 20 ◦ C. Above and below this temperature, a progressive decrease in the regularity of pore arrangement in ATO films is observed. Fig. 6 shows the photocurrent dependence on the electrolyte temperature used for oxide growth and wavelength of incident light for different TiO2 nanopore arrays fabricated at the constant potential of 40 V in the glycerol based electrolyte at temperatures ranging from 10 to 40 ◦ C. The generated photocurrents were recorded for the samples anodized for 1 h (Fig. 6A) and with the fixed ATO thickness (Fig. 6B) at 0.5 V vs. SCE in a 0.1 M KNO3 aqueous solution under pulsed UV illumination in 5 s light/10 s dark cycles with a step of 10 nm in the range of 300–400 nm. As can be seen in Fig. 6A, a maximum photocurrent is observed for the TiO2 layer obtained at 20 ◦ C and at the wavelength of 350 nm. The photocurrent generated on samples synthesized at 20 ◦ C is about twice larger than that observed for samples fabricated at 10 ◦ C. The obtained results can be explain in terms of factors influencing the phoelectrochemical performance of TiO2 layers. It is widely recognized that the photocurrent generated on the ATO layer depends on both: the thickness of the ATO layer (length of nanochannels) [44,62,63] and its porosity (consequently pore diameter and wall thickness) [64]. When a thinner oxide layer is

illuminated smaller effective surface area is accessible for light (photon) absorption, and a decrease in photogenerated charge carriers (photocurrent) is observed. On the other hand, increasing the thickness of ATO layer above the certain value may result in a decline of photocurrent intensity [44]. In the thicker ATO films, some photogenerated electrons and holes might recombine again, when they diffuse/drift from the top of the nanoporous film to the Ti substrate. Therefore, relatively weak photoresponses were recorded for the short (175 nm at 10 ◦ C) and long (1434 nm at 30 ◦ C and 2615 nm at 40 ◦ C) TiO2 nanochannels. Porosity of the ATO layer is a second factor influencing the generated photocurrent. According to the Bruggeamn effective medium approximation, increasing porosity of the ATO layer decreases average dielectric constant, similarly as it is in anodic alumina [65]. Consequently, larger amounts of incident light are reflected at surfaces with lower porosities. In our case, the porosity increases with increasing temperature of the oxide formation (Fig.S2, supplementary content), but the generated charge carriers recombine more effectively along the long nanotubes by the time the metallic substrate is reached. To sum up, the observed photoelectrochemical performance of ATO formed after 1 h of anodization at 10 ◦ C is dominated by the limited length of nanochannels. The lower photocurrents generated on the photoanodes fabricated at 30 and 40 ◦ C result from large nanochannel lengths. In order to check if the nanochannel length is a predominant factor which affects the generated photocurrent, we performed an additional set of experiments in which the thickness of formed oxide layer was fixed at 1 ␮m for all studied temperatures (Fig. 6B). The obtained results showed that the recorded photocurrents were similar for all anodizing temperatures. The photoconversion efficiency was determined by incident photon-to-current efficiency (IPCE) calculations performed for 350 mn where the maxima in the generated photocurrents are observed. The IPCEs for both sets of samples: anodized for 1 h and with the fixed oxide thickness are shown in Fig. 7. As can be seen, the IPCE values varied with the oxide thickness, but the similar data are obtained for the samples with the fixed oxide thickness. 4. Conclusions In this work, highly ordered TiO2 nanopore arrays, varying in pore diameters, porosities, and pore densities were fabricated using

134

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135

a three-step anodization of Ti at 40 V in glycerol based electrolyte containing 0.38 wt.% NH4 F and 1.79 wt. % H2 O. It was found that ATO growth is thermally activated and the rate-determining step is transport of oxygen species across the titanium dioxide layer at the pore bottoms. We conclude also that the electrolyte temperature applied for anodizations has a significant influence on the surface morphology and characteristic parameters of porous anodic titanium dioxide. Generally, the oxide thickness, pore diameter and porosity of formed TiO2 layers increase gradually with increasing temperature. The cell diameter initially increases sharply up to 20 ◦ C and then very slightly up to 40 ◦ C with increasing temperature of anodization. The pore density on the top of ATO layer does not change considerably within the temperature range between 10 and 30 ◦ C. At 40 ◦ C the pore density value is twice higher than those observed at lower anodizing temperatures. The best regularity in pore arrangement is observed in the ATO layer obtained at 40 V and 20 ◦ C. Finally, we demonstrated that the maximum photocurrent at 350 nm can be obtained for the sample anodized for 1 h at 20 ◦ C. On the other hand, for the 1-␮m thick ATO samples similar photocurrents were observed for all anodizing temperatures. Acknowledgments This research was partially supported by the National Science Centre, Poland (Grant No. N N204 213340). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.electacta. 2014.08.055. References [1] C.A. Grimes, G.K. Mor in: TiO2 nanotube arrays: Synthesis, properties, and applications, Springer Dordrecht Heidelberg London New York (2009) 1-345. [2] A. Ghicov, P. Schmuki, Self-ordering electrochemistry: A review on growth and functionality of TiO2 , Chem. Commun. (2009) 2791–2808. [3] P. Roy, S. Berger, P. Schmuki, TiO2 nanotubes: Synthesis and applications, Angew. Chem. Int. Ed. 50 (2011) 2904–2939. [4] X. Chen, S.S. Mao, Titanium dioxide nanomaterials: Synthesis, properties, modifications, and applications, Chem. Rev 107 (2007) 2891–2959. [5] B. Chen, J. Hou, K. Lu, Formation mechanism of TiO2 nanotubes and their applications in photoelectrochemical water splitting and supercapacitors, Langmuir 29 (2013) 5911–5919. [6] C.W. Lai, S. Sreekantan, Incorporation of WO3 species into TiO2 nanotubes via wet impregnation and their water-splitting performance, Electrochim. Acta 87 (2013) 294–302. [7] W. Krengvirat, S. Sreekantan, A-F. Mohd Noor, G. Kawamura, H. Muto, A. Matsuda, Single-step growth of carbon and potassium-embedded TiO2 nanotube arrays for efficient photoelectrochemical hydrogen generation, Electrochim. Acta 89 (2013) 585–593. [8] Y. Li, Y. Xiang, S. Peng, X. Wang, L. Zhou, Modification of Zr-doped titania nanotube arrays by urea pyrolysis for enhanced visible-light photoelectrochemical H2 generation, Electrochim. Acta 87 (2013) 794–800. [9] R. Ojani, J-B. Raoof, A. Khanmohammadi, E. Zarei, Photocatalytic degradation of 3-nitrophenol at surface of Ti/TiO2 electrode, J. Solid State Electrochem. 17 (2013) 63–68. [10] Z. Zhang, Y. Yu, P. Wang, Top-porous/bottom-tubular TiO2 nanostructures decorated with Pd nanoparticles for efficient photoelectrocatalytic decomposition of synergistic pollutants, ACS Appl. Mater. Interfaces 4 (2012) 990–996. [11] Y-Y. Song, Z-D. Gao, P. Schmuki, Highly uniform Pt nanoparticle decoration on TiO2 nanotube arrays: A refreshable platform for methanol electrooxidation, Electrochem. Commun. 13 (2011) 290–293. [12] H. He, P. Xiao, M. Zhou, F. Liu, S. Yu, L. Qiao, Y. Zhang, PtNi alloy nanoparticles supported on carbon-doped TiO2 nanotube arrays for photo-assisted methanol oxidation, Electrochim. Acta 88 (2013) 782–789. [13] L-K. Tsui, G. Zangari, The influence of morphology of electrodeposited Cu2 O and Fe2 O3 on the conversion efficiency of TiO2 nanotube photoelectrochemical solar cells, Electrochim. Acta 100 (2013) 220–225. [14] M. Ye, D. Zheng, M. Lv, C. Chen, C. Lin, Z. Lin, Hierarchically structured nanotubes for highly efficient dye-sensitized solar cells, Adv. Mater. 25 (2013) 3039–3044. [15] W. Guo, X. Xue, S. Wang, C. Lin, Z.L. Wang, An integrated power pack of dyesensitized solar cell and Li battery based on double-sided TiO2 nanotube arrays, Nano Lett. 12 (2012) 2520–2523.

[16] Y. Fan, N. Zhang, L. Zhang, H. Shao, J. Wang, J. Zhang, C. Cao, Co3 O4 -coated TiO2 nanotube composites synthesized through photo-deposition strategy with enhanced performance for lithium-ion batteries, Electrochim. Acta 94 (2013) 285–293. [17] N.A. Kyeremateng, C. Lebouin, P. Knauth, T. Djenizian, The electrochemical behaviour of TiO2 nanotubes with Co3 O4 or NiO submicron particles: Composite anode materials for Li-ion micro batteries, Electrochim. Acta 88 (2013) 814–820. [18] F. Liang, L-B. Luo, C-K. Tsang, L. Zheng, H. Cheng, Y.Y. Li, TiO2 nanotube-based field effect transistors and their application as humidity sensors, Mater. Res. Bull. 47 (2012) 54–58. [19] Y. Gönüllü, G.C.M. Rodríguez, B. Saruhan, M. Ürgen, Improvement of gas sensing performance of TiO2 towards NO2 by nano-tubular structuring, Sens. Actuat. B 169 (2012) 151–160. [20] Y. Kimura, S. Kimura, R. Kojima, M. Bitoh, M. Abe, M. Niwano, Micro-scaled hydrogen gas sensors with patterned anodic titanium oxide nanotube film, Sens. Actuat. B 177 (2013) 1156–1160. [21] S. Yu, X. Peng, G. Cao, M. Zhou, L. Qiao, J. Yao, H. He, Ni nanoparticles decorated titania nanotube arrays as efficient nonenzymatic glucose sensor, Electrochim. Acta 76 (2012) 512–517. [22] A.K.M. Kafi, G. Wu, P. Benvenuto, A. Chen, Highly sensitive amperometric H2 O2 biosensor based on hemoglobin modified TiO2 nanotubes, J. Electroanal. Chem. 662 (2011) 64–69. [23] N.T.C. Oliveira, J.F. Verdério, C. Bolfarini, Obtaining self-organized nanotubes on biomedical Ti–Mo alloys, Electrochem. Commun. 35 (2013) 139–141. [24] S. Wu, X. Liu, K.W.K. Yeung, H. Guo, P. Li, T. Hu, C.Y. Chung, P.K. Chu, Surface nano-architectures and their effects on the mechanical properties and corrosion behavior of Ti-based orthopedic implants, Surf. Coat. Technol. 233 (2013) 13–26. [25] V.M. Prida, M. Hernández-Vélez, K.R. Pirota, A. Menéndez, M. Vázquez, Synthesis and magnetic properties of Ni nanocylinders in self-aligned and randomly disordered grown titania nanotubes, Nanotechnology 16 (2005) 2696–2702. [26] J.M. Macak, B.G. Gong, M. Hueppe, P. Schmuki, Filling of TiO2 nanotubes by self-doping and electrodeposition, Adv. Mater. 19 (2007) 3027–3031. [27] D. Wang, Y. Liu, C. Wang, F. Zhou, W. Liu, Highly flexible coaxial nanohybrids made from porous TiO2 nanotubes, ACS Nano 3 (2009) 1249–1257. [28] D. Kowalski, A. Tighineanu, P. Schmuki, Polymer nanowires or nanopores? Site selective filling of titania nanotubes with polypyrrole, J. Mater. Chem. 21 (2011) 17909–17915. [29] V. Zwilling, M. Aucouturier, E. Darque-Ceretti, Anodic oxidation of titanium and TA6 V alloy in chromic media, An electrochemical approach, Electrochim. Acta 45 (1999) 921–929. [30] D. Gong, C.A. Grimes, O.K. Varghese, Titanium oxide nanotube arrays prepared by anodic oxidation, J. Mater. Res. 16 (2001) 331–334. [31] Q. Cai, M. Paulose, O.K. Varghese, C.A. Grimes, The effect of electrolyte composition on the fabrication of self-organized titanium oxide nanotube arrays by anodic oxidation, J. Mater. Res. 20 (2005) 230–236. [32] G.K. Mor, O.K. Varghese, M. Paulose, K. Shankar, C.A. Grimes, A review on highly ordered, vertically oriented TiO2 nanotube arrays: Fabrication, material properties, and solar energy applications, Solar Energy Mater. Solar Cells 90 (2006) 2011–2075. [33] M. Paulose, K. Shankar, S. Yoriya, H.E. Prakasam, O.K. Varghese, G.K. Mor, T.A. Latempa, A. Fitzgerald, C.A. Grimes, Anodic growth of highly ordered TiO2 nanotube arrays to 134 ␮m in length, J. Phys. Chem. B 110 (2006) 16179–16184. [34] C. Ruan, M. Paulose, O.K. Verghese, G.K. Mor, C.A. Grimes, Fabrication of highly ordered TiO2 nanotube arrays using an organic electrolyte, J. Phys. Chem. B 109 (2005) 15754–15759. [35] S. Yoriya, M. Paulose, O.K. Varghese, G.K. Mor, C.A. Grimes, Fabrication of vertically oriented TiO2 nanotube arrays using dimethyl sulfoxide electrolytes, J. Phys. Chem. C 111 (2007) 13770–13776. [36] C.A. Grimes, Synthesis and application of highly ordered arrays of TiO2 nanotubes, J. Mater. Chem. 17 (2007) 1451–1457. [37] H.E. Prakasam, K. Shankar, M. Paulose, O.K. Verghese, C.A. Grimes, A new benchmark for TiO2 nanotube array growth by anodization, J. Phys. Chem. C 111 (2007) 7235–7241. [38] M. Paulose, H.E. Prakasam, O.K. Varghese, L. Peng, K.C. Popat, G.K. Mor, T.A. Desai, C.A. Grimes, TiO2 nanotube arrays of 1000 ␮m length by anodization of titanium foil: phenol red diffusion, J. Phys. Chem. C 111 (2007) 14992– 14997. [39] J.M. Macak, H. Tsuchiya, L. Taveira, S. Aldabergerova, P. Schmuki, Smooth anodic TiO2 nanotubes, Angew. Chem. Int. 44 (2005) 7463–7465. [40] J.M. Macak, P. Schmuki, Anodic growth of self-organized anodic TiO2 nanotubes in viscous electrolytes, Electrochim. Acta 52 (2006) 1258–1264. [41] W. Chanmanee, A. Watcharenwong, R.C. Chenthamaraksha, P. Kajitvichyanukul, N.T. de Tacconi, K. Rajeshwar, Titania nanotubes from pulse anodization of titanium foils, Electrochem. Commun. 9 (2007) 145–2149. [42] Y. Yin, Z. Jin, F. Hou, X. Wang, Synthesis and morphology of TiO2 nanotube arrays by anodic oxidation using modified glycerol-based electrolytes, J. Am. Ceram. Soc. 90 (2007) 2384–2389. [43] J.M. Macak, S. Albu, D.H. Kim, I. Paramasivam, S. Aldabergerova, P. Schmuki, Multilayer TiO2 nanotube formation by two-step anodization, Electrochem. Solid-State Lett. 10 (2007) K28–K31. [44] Y. Lai, H. Zhuang, L. Sun, Z. Chen, C. Lin, Self-organized TiO2 nanotubes in mixed organic–inorganic electrolytes and their photoelectrochemical performance, Electrochim. Acta 54 (2009) 6536–6542.

J. Kapusta-Kołodziej et al. / Electrochimica Acta 144 (2014) 127–135 [45] S. Sreekantan, R. Hazan, Z. Lockman, Photoactivity of anatase-rutile TiO2 nanotubes formed by anodization method, Thin Solid Films 518 (2009) 16–21. [46] S. Sreekantan, K.A. Saharudin, Ch.L. Wei, Formation of TiO2 nanotubes via anodization and potential applications for photocatalysts, biomedical materials, and photoelectrochemical cell, IOP Conf. Ser.: Mater. Sci. Eng. 21 (2011) 1–18. [47] Y. Alivov, Z.Y. Fan, D. Johnstone, Titanium nanotubes grown by titanium anodization, J. Appl. Phys. 106 (2009) 034314. [48] C. Cao, G. Zhang, X. Song, Z. Sun, Morphology and microstructure of assynthesized anodic TiO2 nanotube arrays, Nanoscale Res. Lett. 6 (2011) 64. [49] Y.-J. Chang, J.-W. Lee, H.-P. Chen, L.-S. Liu, G.-J. Weng, Photocatalytic characteristics of TiO2 nanotubes with different microstructures prepared under different pulse anodizations, Thin Solid Films 519 (2011) 3334–3339. [50] J.-H. Kim, K. Zhu, J.Y. Kim, A.J. Frank, Tailoring oriented TiO2 nanotube morphology for improved Li storage kinetics, Electrochim. Acta 88 (2013) 123–128. [51] K. Lee, D. Kim, P. Schmuki, Highly self-ordered nanochannel TiO2 structure by anodization in a hot glycerol electrolyte, Chem. Commun. 47 (2011) 5789–5791. [52] H. Li, J. Wang, K. Huang, G. Sun, M. Zhou, In-situ preparation of multi-layer TiO2 nanotube array thin films by anodic oxidation method, Mater. Lett. 65 (2011) 1188–1190. [53] D. Regonini, A. Satka, A. Jaroenworaluck, D.W.E. Allsopp, C.R. Bowen, R. Stevens, Factors influencing surface morphology of anodized TiO2 nanotubes, Electrochim. Acta 74 (2012) 244–253. [54] K. Indira, S. Ningshen, U. Kamachi Mudali, N. Rejendran, Effect of anodization parameters on the structural morphology of titanium in fluoride containing electrolytes, Mater. Charact. 71 (2012) 58–65. [55] R. Sánchez-Tovar, K. Lee, J. GarcÍa-Antón, P. Schmuki, Formation of anodic TiO2 nanotube or nanosponge morphology determined by the electrolyte hydrodynamic conditions, Electrochem. Commun. 26 (2013) 1–4.

135

[56] J. Wang, Z. Lin, Anodic formation of ordered TiO2 nanotube arrays: Effects of electrolyte temperature and anodization potential, J. Phys. Chem. C 113 (2009) 4026–4030. [57] W.-Y. Wang, B.-R. Chen, Characterization and photocatalytic activity of TiO2 nanotube films prepared by anodization, Int. J. Photoenergy (2013) 1–12, 2013. [58] I. Horcas, R. Fernández, J.M. Gómez-Rodríguez, J. Colchero, J. Gómez-Herrero, A.M. Baro, WSXM: a software for scanning probe microscopy and a tool for nanotechnology, Rev. Sci. Instrum. 78 (2007) 013705. [59] Image J. National Institute of Mental Health. Bethesda, MD, USA, http://rsb.info.nih.gov/ij [60] G.D. Sulka, J. Kapusta-Kołodziej, A. Brzózka, M. Jaskuła, Fabrication of nanoporous TiO2 by electrochemical anodization, Electrochim. Acta 55 (2010) 4359–4367. [61] G.D. Sulka, J. Kapusta-Kołodziej, A. Brzózka, M. Jaskuła, Anodic growth of TiO2 nanopore arrays at various temperatures, Electrochim. Acta 104 (2013) 526–535. [62] R. Hahn, T. Stergiopoulus, J.M. Macak, D. Tsoukleris, A.G. Kontos, S.P. Albu, D. Kim, A. Ghicov, J. Kunze, P. Falaras, P. Schmuki, Efficient solar energy conversion using TiO2 nanotubes produced by rapid breakdown anodization–a comparison, Phys. Status Solidi RRL 1 (2007) 135–137. [63] M. Paulose, G.K. Mor, O.K. Varghese, K. Shankar, C.A. Grimes, Visible light photoelectrochemical and water-photoelectrolysis properties of titania nanotube arrays, J. Photochem. Photobiol A 178 (2006) 8–15. [64] S. Liang, J. He, Z. Sun, Q. Liu, Y. Jiang, H. Cheng, B. He, Z. Xie, S. Wei, Improving photoelectrochemical water splitting activity of TiO2 nanotube arrays by tuning geometrical parameters, J. Phys. Chem. C 116 (2012) 9049–9053. [65] G.D. Sulka, K. Hnida, Distributed Bragg reflector based on porous anodic alumina fabricated by pulse anodization, Nanotechnology 23 (2012) 075303.