The effects of geometrical characteristics such as surface area (SA) and porosity of TiO2 nanotube arrays (TNAs) on its photocatalytic activity were investigated by applying variable voltages and reaction times for the anodization of Ti substrates. While larger SA of nanotubes was observed under higher applied potential, the porosity of TNAs decreased by increasing anodizing voltage. Under applied potential of 80 V, the SA of TNAs increased from 0.164 to 0.471 m2/g as anodization time increased from 1 to 5 hours, respectively. However, no significant effect on the porosity of TNAs was observed. On the other hand, both SA and porosity of TNAs, synthesized at 60 V, increased by augmenting the anodization time from 1 to 3 hours. But further increasing of anodization time to 5 hours resulted in a decreased SA of TNAs with no effect on their porosity. Accordingly, the TNAs with SA of 0.368 m2/g and porosity of 47% showed the highest photocatalytic activity for degradation of 4-chlorobenzoic acid (4CBA). Finally, the degradation of refractory model compounds such as carbamazepine and bisphenol-A was tested and more than 50% of both compounds could be degraded under UV-A irradiation (λmax = 365 nm).

INTRODUCTION

Trace organic compounds, including pharmaceuticals and personal care products (PPCPs) and endocrine disrupting compounds (EDCs) in effluents from wastewater treatment plants (WWTPs), have been posing significant environmental problems globally (Castiglioni et al. 2005). To reduce their discharge into water bodies, various advanced oxidation processes (AOPs) have been applied (Malato et al. 2009). In particular, heterogeneous photocatalysis employing TiO2 nanoparticles has demonstrated promising performance in degrading a wide range of organic contaminants at low cost. While TiO2 nanoparticles in aqueous suspensions are desirable for providing high surface areas (SA) in photocatalytic reactions, post-separation or recovery of the nanoparticles after use remains a major challenge (Liao et al. 2011). One approach to minimize the post-separation processes is to stabilize the catalyst particles by immobilization on suitable support materials. In the light of this, TiO2 nanotube arrays (TNAs) offering a larger SA to volume ratio compared to particulate films have been successfully developed (Roy et al. 2011).

The key parameters for enhancing the photocatalytic behaviour of TNAs are improvements in the active SA of TNAs and mass transport of reactants within the nanotubes. In general, nanotubes with larger SA are expected to produce relatively more reactive oxygen species (ROS) than nanotubes with smaller SA (Ku 2010). On the other hand, the porosity of TNAs not only affects the hole diffusion path (Liang et al. 2012) but also controls the diffusion of the reactant molecules (especially oxygen molecules) inside the nanotubes. Although TNAs have recently received significant attention in environmental applications, there are very limited reports on the effects of their geometrical structures (e.g. SA and porosity) on their photocatalytic activity (Wu et al. 2010).

In this study, we report the effects of anodization conditions on geometrical properties of TNAs and the subsequent influence on their photocatalytic activities via photocatalytic degradation of model refractory organic compounds such as 4-chlorobenzoic acid (4CBA), carbamazepine (CBZ) and bisphenol-A (BPA) under ultraviolet (UV) A irradiation.

MATERIALS AND METHODS

Fabrication of TiO2 nanotube arrays

TNAs were synthesized by anodic oxidation of Ti foil (2.5 cm × 1.5 cm) (Ti-Shop, UK) in a fluorinated solution of ethylene glycol containing 0.5 wt% NaF and 5 wt% water (Yun et al. 2011). Prior to anodization, the Ti foils were decontaminated by immersing in acetone and ethanol for 15 minutes, rinsing with ultrapure water and followed by vacuum drying. Anodization was carried out in a one compartment cell with two electrodes, comprised of platinum (Pt) sheet (cathode) and Ti foil (anode). A programmable direct current (DC) power supply interfaced with a computer was used to provide electrical potential of 60–80 V on both electrodes over 1–5 hours. Then, the resultant TNA samples were subjected to calcination at 450 °C for 3 hours with a ramp rate of 5 °C/min.

Characterization of TiO2 nanotube arrays

The surface morphology and crystallinity of TNAs were characterized using a scanning electron microscope (SEM) (Hitachi-S4500) and X-ray diffractometer (XRD) (Siemens D5000), respectively. Quantitative measurement of the structural dimensions (i.e. diameter and length) of TNAs was done using image analysis software (Image-Pro version 4.5, Media Cybernetics, Inc., Bethesda, MD, USA).

Photocatalytic degradation of model organic compounds

4CBA (99%), BPA (99%), and CBZ (98%) were purchased from Sigma-Aldrich and used as model organic compounds to assess the photocatalytic activity of fabricated TNAs. 4CBA and aromatic compounds of similar chemical structure have been widely used in fungicides and pharmaceuticals exhibiting bactericide properties at high concentrations. These are resistant to biological treatments and could not be efficiently removed by conventional biological wastewater treatment processes (Dionysiou et al. 2004).

All experiments were performed in glass dishes (60 mm outer diameter × 35 mm height) in a UV chamber (Electron Microscopy Science, Hatfield, PA, USA) (Figure 1). Two 15 W fluorescent UV bulbs (Philips TLD 15W/08) with a maximum spectral power distribution (1600 μW/cm2) at 365 nm were used as the light source.

Figure 1

Schematic of the experimental set-up for photocatalytic degradation experiments.

Figure 1

Schematic of the experimental set-up for photocatalytic degradation experiments.

Different TNA samples were immersed in 30 ml aqueous suspensions of 4CBA (25 μM), BPA (2 μM) or CBZ (2 μM). The solution was agitated on a magnetic stirrer at 400 rpm in the absence of light for 1 hour for attaining the adsorption–desorption equilibrium and was subsequently exposed to UV-A irradiation for 2 hours.

One millilitre of the sample solution was collected every 20 minutes for further analysis. The concentrations of model compounds were quantified using a high performance liquid chromatography (HPLC) system equipped with a SPD-M20 photodiode array detector selected at λ = 210 nm, a ZORBAX Eclipse XBD-C18 analytical column (5 μm, 2.1 × 150 mm, Agilent) for 4CBA and a C-18 silica column (5 μm, 25 cm × 4.6 mm, Supelco Park) for the selected PPCP and EDC. The eluent consisted of a binary mixture of 0.1% phosphoric acid aqueous solution and acetonitrile (70:30, v/v) with a flow rate of 1 ml/min and an injection volume of 30 μl. All experiments were performed in triplicate. Student's t-test was used to assess the significance of the results employing a 95% confidence interval. Statistical significance was evaluated using one-way analysis of variance (ANOVA).

RESULTS AND DISCUSSION

Characteristics of the TNAs

The formation of TiO2 nanotubes on Ti substrates occurs via simultaneous reactions of electrochemical etching (i.e. field-enhanced dissolution of Ti4+ ions and TiO2 formation) and chemical dissolution by F ions (Mor et al. 2006). Therefore, the morphology and structure of TNAs are strongly influenced by the interaction between these two reactions. Figure 2 shows the top surface and cross-sectional SEM images of TNAs that are fabricated at two applied potentials using different anodization times. The corresponding size parameters derived from the SEM analysis are summarized in Table 1.

Table 1

Structural dimensions of TNA samples (L: nanotube length; Di: inner diameter; Do: outer diameter; W: wall thickness)

Sample ID L (μm) Di (nm) Do (nm) W (nm) 
TNA-1 (60 V-1 h) 2.0 77 ± 5 135 ± 7 29 ± 4 
TNA-2 (60 V-3 h) 7.6 101 ± 5 147 ± 7 23 ± 2 
TNA-3 (60 V-5 h) 7.7 119 ± 3 171 ± 6 26 ± 3 
TNA-4 (80 V-1 h) 4.2 ± 0.2 80 ± 4 148 ± 6 34 ± 2 
TNA-5 (80 V-3 h) 14.5 ± 0.5 110 ± 9 192 ± 10 41 ± 3 
TNA-6 (80 V-5 h) 16.3 ± 0.1 122 ± 9 212 ± 13 45 ± 4 
Sample ID L (μm) Di (nm) Do (nm) W (nm) 
TNA-1 (60 V-1 h) 2.0 77 ± 5 135 ± 7 29 ± 4 
TNA-2 (60 V-3 h) 7.6 101 ± 5 147 ± 7 23 ± 2 
TNA-3 (60 V-5 h) 7.7 119 ± 3 171 ± 6 26 ± 3 
TNA-4 (80 V-1 h) 4.2 ± 0.2 80 ± 4 148 ± 6 34 ± 2 
TNA-5 (80 V-3 h) 14.5 ± 0.5 110 ± 9 192 ± 10 41 ± 3 
TNA-6 (80 V-5 h) 16.3 ± 0.1 122 ± 9 212 ± 13 45 ± 4 
Figure 2

Surface morphology of TNA samples anodized in ethylene glycol (0.5 wt% NaF, 5 wt% H2O) at 60 V for (a) 1 hour, (b) 3 hours and (c) 5 hours, and 80 V for (d) 1 hour, (e) 3 hours and (f) 5 hours.

Figure 2

Surface morphology of TNA samples anodized in ethylene glycol (0.5 wt% NaF, 5 wt% H2O) at 60 V for (a) 1 hour, (b) 3 hours and (c) 5 hours, and 80 V for (d) 1 hour, (e) 3 hours and (f) 5 hours.

The nanotube length and both the inner and outer diameter of nanotubes were found to increase by the increment of applied potential. The effect of applied potential on the diameter of the nanotubes can be related to the number of pits formed at the early stage of the anodization process (Lockman et al. 2010). Under higher applied potentials, more pits are formed in the oxide layer (due to the higher electric field-enhanced dissolution rates) which will be etched to form larger pores.

Similarly, the effect of applied potential on the length of nanotubes can be ascribed to the field-enhanced dissolution rate at the barrier layer. The length of TiO2 nanotubes is determined by a combined effect of the inward movement of the oxide barrier layer at the pore bottom (pore deepening rate) and the chemical dissolution of the formed TiO2 nanotubes (Mor et al. 2006; Lockman et al. 2010). Under a higher applied potential, the electric field dissolution at the barrier layer is accelerated, which results in a higher deepening rate of the pore and thus the formation of longer nanotubes.

As presented in Table 1, the length of nanotubes was proportional to the anodization time. As reported in the literature, longer anodization time has led to the formation of longer TiO2 nanotubes due to the continuous oxidation of Ti foil to form oxide layer (Yun et al. 2011). However, after the electrochemical etching rate equilibrates with the chemical dissolution rate, the length of the nanotube becomes independent of anodization time (Mor et al. 2006).

Contrary to the study by Xu et al. (2011) that claimed the diameter of TiO2 nanotubes is time independent, it was observed that both inner and outer diameter of TiO2 nanotubes increased by increasing anodization time (Table 1). It has been accepted that nanotube diameter is mainly proportional to the anodization voltage for fixed water content; however, increase in the inner and outer diameter by augmenting anodization time is mainly due to continuous chemical dissolution of the oxide layer (Yun et al. 2011).

The XRD patterns of fabricated TNAs confirmed that the as-anodized TNA sample before calcination (Figure 3, bottom) contains amorphous TiO2 but the calcinated sample is composed of anatase phase of TiO2 (2θ = 25 °) (Figure 3, top).

Figure 3

XRD patterns of TNAs before (bottom) and after calcination (top).

Figure 3

XRD patterns of TNAs before (bottom) and after calcination (top).

Geometrical characteristics of the TNAs

Recently, a simple geometrical model has been developed to study the effect of anodization conditions on geometrical properties of TNAs such as SA and porosity (Zhu et al. 2006; Kontos et al. 2010; Sulka et al. 2010). Assuming an ideal regular network of identical nanotubes, the real surface area of the TNAs can be estimated by the sum of the cylindrical (inner and outer) and flat top SA of nanotubes. Since TNAs fabricated by anodization process consist of closely packed nanotubes, the outer cylindrical SA of nanotubes has a negligible effect on its photocatalytic reactivity due to the poor light harvesting; therefore, the actual active SA of TNAs was considered to be the inner cylindrical SA plus the flat top side surface area of nanotubes. Therefore, the nanotube density (dNT, i.e. the total number of nanotubes occupying a unit area of 1 cm2) and total SA of TNAs can be calculated according to the following equations: 
formula
1
 
formula
2
where L is the average length of nanotubes (μm), Di is the average inner diameter of nanotubes (nm), and Do is the average outer diameter of nanotubes (nm).
Regardless of small inter-pore voids in closely packed nanotube arrays, the porosity of TNAs can also be defined as the ratio of the total volume of pores to the total volume of TNAs. Considering a unit area of 1 cm2, the porosity (P) of TNAs can be estimated from the following expression: 
formula
3
where Di and Do are the average inner diameter (nm) and outer diameter (nm) of nanotubes, respectively.

Using the dimensions of TNAs derived from the SEM images and Equations (2) and (3), the SA and porosity of TNA samples that were fabricated at different voltages were calculated and illustrated as a function of anodization time in Figures 4 and 5, respectively.

Figure 4

Influence of anodization time on SA of TNA samples fabricated at different applied potentials. (The weight (g) of the sample is referred to the total weight of TNA and Ti foil.) Error bars show standard deviation.

Figure 4

Influence of anodization time on SA of TNA samples fabricated at different applied potentials. (The weight (g) of the sample is referred to the total weight of TNA and Ti foil.) Error bars show standard deviation.

Figure 5

Influence of anodization time on the porosity of TNA samples fabricated at different voltages. Error bars show standard deviation.

Figure 5

Influence of anodization time on the porosity of TNA samples fabricated at different voltages. Error bars show standard deviation.

As shown in Figure 4, an increase in applied potential from 60 to 80 V resulted in larger SA of TNAs. According to Equation (2), the SA of TNAs is mainly affected by two parameters: the nanotube length and the nanotube density. Since dNT is inversely proportional to the square value of the outer diameter, a decrease in nanotube density is expected with an increase of either applied potential or anodization time leading to a smaller SA (Figure 6).

Figure 6

Influence of anodization time and applied potential on density of TNAs.

Figure 6

Influence of anodization time and applied potential on density of TNAs.

On the other hand, augmenting applied potential will result in significant increasing of nanotube length (Table 1) and therefore a larger SA of a single nanotube. Thus, the final effect of anodizing voltage on the total SA of TNAs can be predicted by the competition of these two phenomena. While the density of nanotubes for samples anodized at 60 V was greater than that of those anodized at 80 V (Figure 6), a two-fold increase of nanotube length at applied potential of 80 V dominated over the negative effect of reduced dNT leading to larger SA of TNAs.

Similarly, the effect of anodization time on the total surface area of TNAs can be explained by its influence on nanotube length and density. As anodization time increased from 1 to 3 hours, the length of nanotubes, synthesized under applied potential of 60 V, notably increased from 2 to 7.6 μm, whereas the nanotube density slightly decreased by 16%. Therefore, a significant increase of SA with respect to the anodization time was observed. However, as anodization time increased to 5 hours, the length of nanotubes became independent of anodization time (Table 1) and hence the reduction of nanotube density resulted in a decreased SA.

The data presented in Figure 5 indicate that the porosity of TNAs decreased with increasing applied potential from 60 to 80 V. According to Equation (3), the porosity of TNAs is proportional to the square value of inner diameter of the nanotubes and the nanotube density (1/Do2). Despite a slight increase in inner diameter of nanotubes under applied potential of 80 V, the nanotube density dropped dramatically (Figure 6), which resulted in less porosity of TNAs than that of ones fabricated at 60 V. Interestingly, the porosity of TNAs synthesized at 80 V was independent of anodization time with an average value of 32% which implies that the positive effect of pore enlargement was neutralized by the negative effect of reduction in nanotube density.

Nonetheless, the porosity of samples fabricated at 60 V firstly increased from 32 to 47% by increasing the anodization time from 1 to 3 hours, respectively and then became independent of anodization time. This different behaviour observed at 60 V can be possibly due to the higher growth rates of inner diameter for the first 3 hours of anodization.

Photocatalytic degradation of refractory model compounds by TNAs

The photocatalytic reactivity of TNA samples was evaluated by photocatalytic degradation of 4CBA in aqueous solutions under UV irradiation. Herein, it was hypothesized that the decay of 4CBA by TNAs could be fitted into the Langmuir-Hinshelwood (L-H) model (Wenhua et al. 2000) according to the following equation: 
formula
4
where r, C, kr, K and t denote the reaction rate for photocatalytic degradation of 4CBA (μM/min), concentration of 4CBA (μM), specific reaction rate constant (μM/min), equilibrium adsorption constant (/μM), and time (min), respectively.
If KC ≪ 1, as in our case, Equation (4) can be simplified to: 
formula
5
Integration of the above equation results in the following expression: 
formula
6
where kapp is the apparent reaction rate constant/min.

The logarithmic plots of normalized concentration of 4CBA vs UV irradiation time for different TNA samples are given in Figure 7.

Figure 7

Kinetics of 4CBA photocatalytic degradation by TNA samples ([4CBA] = 25 μM, pH = 5, temperature = 25 °C).

Figure 7

Kinetics of 4CBA photocatalytic degradation by TNA samples ([4CBA] = 25 μM, pH = 5, temperature = 25 °C).

The corresponding linear dependence observed in Figure 7 implies that the degradation of 4CBA by all TNA samples follows pseudo first-order reaction kinetics, confirming the L-H photocatalytic mechanism. The apparent reaction rate constants and the removal efficiency of 4CBA by all TNA samples are summarized in Table 2.

Table 2

Apparent reaction rate constants and removal efficiency of 4CBA by various TNA samples under UV-A irradiation

Sample kapp (/min) (R2SA (m2/g) Porosity (%) Removal (%) 
TNA-1 0.0058 (0.99) 0.086 32 50 
TNA-2 0.0104 (0.99) 0.369 47 71 
TNA-3 0.0084 (0.99) 0.331 48 64 
TNA-4 0.0045 (0.99) 0.164 29 42 
TNA-5 0.0090 (0.99) 0.451 33 66 
TNA-6 0.0081 (0.99) 0.471 33 62 
Sample kapp (/min) (R2SA (m2/g) Porosity (%) Removal (%) 
TNA-1 0.0058 (0.99) 0.086 32 50 
TNA-2 0.0104 (0.99) 0.369 47 71 
TNA-3 0.0084 (0.99) 0.331 48 64 
TNA-4 0.0045 (0.99) 0.164 29 42 
TNA-5 0.0090 (0.99) 0.451 33 66 
TNA-6 0.0081 (0.99) 0.471 33 62 

The highest removal efficiency of 4CBA (>70%) in aqueous solution and the largest kapp (0.0104/min) were obtained for TNA-2 sample which showed a SA of 0.369 m2/g and porosity of 47%. It has been accepted that the photocatalytic reactivity of TNAs depends on their SA and porosity. Larger SA of TNAs allows more adsorption of aqueous reactants (specifically water molecules and hydroxyl ions) onto the outer and also inner surfaces of the nanotubes, whereas the higher porosity of TNA results in a faster diffusion of reactants (especially oxygen molecules) and improved light penetration during the photocatalytic reaction (Liang & Li 2009).

However, further increase in surface area of TNAs in TNA-6 sample compared to TNA-5 with the same porosity (33%) did not lead to an improved performance in terms of larger kapp and higher removal efficiency. This finding means that the photocatalytic reactivity of TNAs is not dependent upon its SA and porosity alone and therefore, there must be some other geometrical or structural factors that have a notable influence on the reactivity of TNAs.

As reported in the literature, the mechanism of ROS production by TNAs is different from suspended TiO2 nanoparticles (Liu et al. 2012). Under irradiation of UV light, photo-generated electrons and holes in the nanotube wall will diffuse to the nanotube surface in the radial and axial direction that is accompanied by bulk recombination. However, diffusion along the axial direction was found to be less effective for the nanotube wall is much thinner than its length. The transfer of electrons from TiO2 to O2 is also considered to be the rate-limiting step of photocatalysis (Mora Seró et al. 2005) particularly in the absence of electron trapping centres on the surface. Therefore, the thinner wall of nanotubes enhances the migration of photo-generated electrons and holes from the bulk to the surface of nanotube walls leading to increased production of ROS (Zhuang et al. 2007; Liu et al. 2012).

Based upon these considerations, it was found that samples fabricated under applied potential of 80 V exhibited the thickest walls among other samples with the largest thickness (45 nm) observed for TNA-6 (Table 1). Consequently, the unexpected decrease of photocatalytic reactivity of TNA-6 sample can be attributed to the higher wall thickness which severely affected the diffusion of photo-generated electrons and holes.

Photocatalytic reactivity of the TNA-2 sample has been investigated for removal of selected PPCP (CBZ) and EDC (BPA). The results of the photocatalytic removal of both compounds, as well as 4CBA, are illustrated in Figure 8.

Figure 8

Kinetics of photocatalytic degradation of the selected PPCP and EDC by the TNA-2 sample ([BPA or CZP] = 2 μM, pH = 5, temperature = 25 °C).

Figure 8

Kinetics of photocatalytic degradation of the selected PPCP and EDC by the TNA-2 sample ([BPA or CZP] = 2 μM, pH = 5, temperature = 25 °C).

The corresponding linear dependency of logarithmic normalized concentration of selected compounds vs UV irradiation time indicated that the photocatalytic degradation of BPA and CBZ followed the pseudo first-order kinetic (L-H model). Under UV irradiation for 2 hours, more than 50% of both selected pollutants were degraded. The apparent reaction rate constants of 0.007/min (R2 = 0.9265) and 0.006/min (R2 = 0.9984) were obtained for CBZ and BPA, respectively. However, these observed values are 1.5 times smaller than the kapp obtained for 4CBA (0.0104/min), which may have resulted from the presence of additional aromatic rings in their molecular structure.

Another possible reason for the lower removal efficiency of BPA and CBZ compared to 4CBA might rely on their surface charge, which is strongly affected by the pH of the solution. At pH below pKa value, an organic compound exists as neutral specie while the organic compound surface is negatively charged above the pKa value (Ahmed et al. 2010). At pH 5, the surface of 4CBA (pKa = 3.98) is negatively charged while BPA (pKa = 9.9–11.3) and CBZ (pKa = 7–13.9) are present as neutral compounds (Rykowska 2006; Yamamoto et al. 2009). The point of zero charge (pzc) of TiO2 has been reported as pH 6.2 (Chen & Ray 1998). Therefore, electrostatic attraction forces between the positively charged TiO2 nanotubes and the negatively charged 4CBA molecules at this solution pH 5 resulted in enhanced adsorption, and hence led to the largest apparent reaction rate constant and 4CBA removal efficiency. Further investigations are recommended to elucidate the photocatalytic degradation kinetics of these compounds by TNAs.

CONCLUSIONS

The application of TNAs for photocatalytic treatment of water and wastewater relies on their geometrical characteristics. This study clearly showed that photocatalytic reactivity of TNAs was influenced by its SA and porosity. However, wall thickness of nanotubes also plays a very crucial role as it determines the diffusion path for photo-generated electrons and holes. The photo-activated TNA with SA of 0.368 m2/g and porosity of 47% could degrade 4CBA (25 μM) by 71% and selected PPCP and EDC (2 μM) by 50% under UV-A irradiation for 2 hours. Where existing technologies for the elimination of refractory organic compounds from WWTPs are limited, the TNA may hold promise for niche oxidative treatment systems for wastewater recycling. Nonetheless, the overall removal efficiency of those compounds by TNAs could be improved by modifying the surface properties of nanotubes in terms of larger SA, higher porosity, and thinner nanotube wall.

ACKNOWLEDGEMENTS

This study was supported by the Korea Ministry of Environment as The Eco-Innovation Project (Global Top Project, Project no. GT-SWS-11-01-002-0) and also by the National Research Foundation of Korea (NRF) grant funded by the Korea government (NRF-2013R1A2A1A09007252).

REFERENCES

REFERENCES
Ahmed
S.
Rasul
M. G.
Martens
W. N.
Brown
R.
Hashib
M. A.
2010
Heterogeneous photocatalytic degradation of phenols in wastewater: a review on current status and developments
.
Desalination
261
(
1–2
),
3
18
.
Castiglioni
S.
Bagnati
R.
Fanelli
R.
Pomati
F.
Calamari
D.
Zuccato
E.
2005
Removal of pharmaceuticals in sewage treatment plants in Italy
.
Environmental Science & Technology
40
(
1
),
357
363
.
Chen
D.
Ray
A. K.
1998
Photodegradation kinetics of 4-nitrophenol in TiO2 suspension
.
Water Research
32
(
11
),
3223
3234
.
Dionysiou
D. D.
Suidan
M. T.
Baudin
I.
Laîné
J.-M.
2004
Effect of hydrogen peroxide on the destruction of organic contaminants-synergism and inhibition in a continuous-mode photocatalytic reactor
.
Applied Catalysis B: Environmental
50
(
4
),
259
269
.
Kontos
A. G.
Katsanaki
A.
Maggos
T.
Likodimos
V.
Ghicov
A.
Kim
D.
Kunze
J.
Vasilakos
C.
Schmuki
P.
Falaras
P.
2010
Photocatalytic degradation of gas pollutants on self-assembled titania nanotubes
.
Chemical Physics Letters
490
(
1–3
),
58
62
.
Ku
Y.
Fan
Z.
Chou
Y.
Wang
W.
2010
Characterization and induced photocurrent of TiO2 nanotube arrays fabricated by anodization
.
Journal of the Electrochemical Society
157
(
6
),
671
675
.
Liang
S.
He
J.
Sun
Z.
Liu
Q.
Jiang
Y.
Cheng
H.
He
B.
Xie
Z.
Wei
S.
2012
Improving photoelectrochemical water splitting activity of TiO2 nanotube arrays by tuning geometrical parameters
.
Journal of Physical Chemistry C
116
(
16
),
9049
9053
.
Liao
J.
Lin
S.
Zhang
L.
Pan
N.
Cao
X.
Li
J.
2011
Photocatalytic degradation of methyl orange using a TiO2/Ti mesh electrode with 3D nanotube arrays
.
ACS Applied Materials & Interfaces
4
(
1
),
171
177
.
Liu
B.
Nakata
K.
Liu
S.
Sakai
M.
Ochiai
T.
Murakami
T.
Takagi
K.
Fujishima
A.
2012
Theoretical kinetic analysis of heterogeneous photocatalysis by TiO2 nanotube arrays: the effects of nanotube geometry on photocatalytic activity
.
Journal of Physical Chemistry C
116
(
13
),
7471
7479
.
Lockman
Z.
Sreekantan
S.
Ismail
S.
Schmidt Mende
L.
MacManus Driscoll
J. L.
2010
Influence of anodisation voltage on the dimension of titania nanotubes
.
Journal of Alloys and Compounds
503
(
2
),
359
364
.
Malato
S.
Fernández-Ibáñez
P.
Maldonado
M. I.
Blanco
J.
Gernjak
W.
2009
Decontamination and disinfection of water by solar photocatalysis: recent overview and trends
.
Catalysis Today
147
(
1
),
1
59
.
Mor
G. K.
Varghese
O. K.
Paulose
M.
Shankar
K.
Grimes
C. A.
2006
A review on highly ordered, vertically oriented TiO2 nanotube arrays: fabrication, material properties, and solar energy applications
.
Solar Energy Materials and Solar Cells
90
(
14
),
2011
2075
.
Roy
P.
Berger
S.
Schmuki
P.
2011
Tio2 nanotubes: synthesis and applications
.
Angewandte Chemie International Edition
50
(
13
),
2904
2939
.
Rykowska
I.
Wasiak
W.
2006
Properties, threats, and methods of analysis of bisphenol A and its derivatives
.
Acta Chromatographica
19
,
7
27
.
Sulka
G. D.
Kapusta Kolodziej
J.
Brzózka
A.
Jaskuła
M.
2010
Fabrication of nanoporous TiO2 by electrochemical anodization
.
Electrochimica Acta
55
(
14
),
4359
4367
.
Wenhua
L.
Hong
L.
Sao'an
C.
Jianqing
Z.
Chunan
C.
2000
Kinetics of photocatalytic degradation of aniline in water over TiO2 supported on porous nickel
.
Journal of Photochemistry and Photobiology A: Chemistry
131
(
1–3
),
125
132
.
Wu
N.
Wang
J.
Tafen
D. N.
Wang
H.
Zheng
J.-G.
Lewis
J. P.
Liu
X.
Leonard
S. S.
Manivannan
A.
2010
Shape-enhanced photocatalytic activity of single-crystalline anatase TiO2 (101) nanobelts
.
Journal of the American Chemical Society
132
(
19
),
6679
6685
.
Xu
H.
Liu
Y.
Wang
H.
Zhao
W.
Huang
H.
Liang
C.
Ye
Q.
Li
M.
Deng
Y.
Shen
H.
2011
The fabrication of highly ordered TiO2 nanotube arrays and their application in dye-sensitized solar cells
.
Advanced Materials Research
217–218
,
1553
1558
.
Yamamoto
H.
Nakamura
Y.
Moriguchi
S.
Nakamura
Y.
Honda
Y.
Tamura
I.
Hirata
Y.
Hayashi
A.
Sekizawa
J.
2009
Persistence and partitioning of eight selected pharmaceuticals in the aquatic environment: laboratory photolysis, biodegradation, and sorption experiments
.
Water Research
43
(
2
),
351
362
.
Yun
J. H.
Ng
Y. H.
Ye
C.
Mozer
A. J.
Wallace
G. G.
Amal
R.
2011
Sodium fluoride-assisted modulation of anodized TiO2 nanotube for dye-sensitized solar cells application
.
ACS Applied Materials & Interfaces
3
(
5
),
1585
1593
.
Zhuang
H. F.
Lin
C. J.
Lai
Y. K.
Sun
L.
Li
J.
2007
Some critical structure factors of titanium oxide nanotube array in its photocatalytic activity
.
Environmental Science & Technology
41
(
13
),
4735
4740
.