O3/UV/TiO2 was used to effectively decompose humic acids (HAs) in drinking water. A series of Al-doped, Fe-doped, Zn-doped and co-doped TiO2 nanotubes were successfully synthesized by a hydrothermal method. According to the characterization of ion-doped TiO2 nanotubes, Al3+ and Fe3+ ions occupied substitutional positions in the crystal lattice, and Zn2+ ions were partially dispersed in the bulk of the TiO2 nanotubes, with ZnO formed on the surface of the TiO2. The calcination temperature and doping concentration could affect the anatase phase weight fractions, average crystallite sizes, Brunauer–Emmett–Teller (BET) surface area, bandgap energy, and photocatalytic activity of the catalysts. For single ion-doped TiO2 nanotubes, the best photocatalytic activities were achieved when the ion-doping amount was 1.0%, and the optimal calcination temperatures for Al-, Fe- or Zn-doped TiO2 were 600°C, 550°C and 550°C, respectively. The highest HA removal efficiency of 80.1% was achieved in the presence of 550°C calcined, 1.0% co-doped TiO2 nanotubes with an Fe:Zn ratio of 1:1 (atomic percent), with second-order rate constant of 0.0394 L/(min·mg). The addition of a third ion had little effect on the photocatalytic activity. The pollutants in filter influent from No. 9 waterworks in Beijing could be efficiently removed, with UV254 and total organic carbon (TOC) removal efficiencies of 57.2% and 44.7%, respectively.

INTRODUCTION

Humic acids (HAs) are notorious because they turn water brown, form complexes with metals and organic pollutants, and react with active chlorine during water sterilization in plants, resulting in the formation of trihalomethanes (Wei et al. 2011).

Several techniques have been developed to treat HAs. O3 can reduce color, UV254 and total organic carbon (TOC), and decrease high apparent molecular weight fractions. UV photocatalytic oxidation, which is characterized by the generation of highly reactive •OH, has emerged as a highly efficient method for decomposing pollutants. Therefore, O3/UV/TiO2 was chosen to generate •OH in the present study.

TiO2 was the most promising catalyst for eliminating contaminants because of its low cost, long-term stability and nontoxicity (Hao & Zhang 2009). However, the relatively wide bandgap of TiO2 limits the photocatalytic efficiency because of the high recombination rate of photogenerated e and h+ (Pang & Abdullah 2012a). Therefore, ion-doped TiO2 nanotubes have been studied (Yuan et al. 2014). Al, Fe or Zn ions could enhance the catalytic activity of TiO2 (Yuan et al. 2013). However, researchers have seldom considered Al, Fe and Zn co-doped TiO2 nanotubes for UV photocatalytic ozonation.

In this study, a UV/TiO2 photocatalytic ozonation method was used. The photocatalytic activities of Al-, Fe-, Zn- and co-doped TiO2 nanotubes with different doping concentrations and at different calcination temperatures were investigated. In order to obtain the removal efficiency of natural organic matter (NOM), the filter influent from No. 9 waterworks in Beijing was used, and the UV254 and TOC of the water were determined.

METHODS

Materials

P-25 TiO2, which has an SBET of 50 m2/g, an average crystallite size of 21 nm, and a crystal structure of 80% anatase and 20% rutile, was provided by the Degussa Corporation (Richfield Park, New Jersey). HAs (content ≥99%, formula weight 600–1000), AlCl3, Fe(NO3)3·9H2O and ZnCl2 were purchased from Beijing Chemical Reagent Company.

The major photocatalytic reactor was a cylindrical glass column (ϕ60 mm × 750 mm). The light source in the reaction column was a 37 W low-pressure mercury vapor lamp (ϕ22 mm × 793 mm) with a wavelength of 254 nm and intensity of 110 μW/cm2. O3-containing gas which was produced by an O3 generator (maximum output of 3 g/h, Beijing Tonglin Scitech Co., Ltd, China) was inlet into the reactor through a porous glass core aeration plate (pore size 4–7 μm) situated at the bottom of the reactor.

Preparation of TiO2 nanotubes

One gram of P-25 TiO2 was added into 16 mL of 10 mol/L NaOH solution in a Teflon vessel. The mixture was stirred for 2 h at 20°C to form a suspension, heated at 110°C for 24 h, and cooled to room temperature in air. The precipitates were washed with distilled water to pH 7 and subsequently dispersed in 0.1 mol/L HCl solution for 30 min. After washing to pH 7 and drying at 60°C, the titanic acid nanotubes were obtained. TiO2 nanotubes were prepared by calcining at 400–600°C for 2 h.

Al-, Fe-, Zn- and co-doped TiO2 nanotubes were prepared following the aforementioned procedure in the presence of AlCl3, Fe(NO3)3·9H2O or ZnCl2 with doping levels of 0.1%–5.0%. All dopant concentrations mentioned in this work are the nominal atomic ones.

Characterization of TiO2 nanotubes

X-ray diffraction (XRD) patterns were collected in a Rigaku Dmax-RB diffractometer (Tokyo, Japan). Transmission electron microscopy (TEM) images were acquired with a HITACHI HT-7700 electron microscope (Tokyo, Japan). The SBET of the samples were determined on a Quadrasorb SI-MP apparatus (Quantachrome Instrument, USA). Diffuse reflectance spectroscopy (DRS) was performed using a HITACHI U-3010 UV-vis scanning spectrophotometer (Tokyo, Japan).

Photocatalytic activity tests

Catalysts of 0.1 g/L were added into the water. The aqueous slurries (1.5 L) were stirred and bubbled with O3 (0.7 mg/min) for 30 min prior to UV irradiation. When the UV lamp was turned on, HA solution was poured into the slurries immediately, and the initial HA concentration was 10 mg/L. The reaction was stopped at 5 min intervals by adding 0.5 mL of 0.001 mol/L Na2S2O3 solution into a 10 mL sample. The suspension was filtered through 0.45-μm-pore-size microporous membranes to separate the photocatalyst particles. HA concentrations were determined using a UV-vis spectrophotometer (Hach DR5000, USA) operated at 254 nm. All the tests were repeated three times to minimize the errors.

In order to obtain the removal efficiency of NOM, the filter influent from No. 9 waterworks in Beijing (TOC 3.8 mg/L, UV254 0.078/cm, permanganate index 1.62 mg/L, turbidity 0.2 NTU, and pH value 7.5) was used, UV254 was determined using a UV-vis spectrophotometer operated at 254 nm, and TOC measurements were performed by the spectrophotometer using low-range TOC ampoules (Hach Chemical, USA).

RESULTS AND DISCUSSION

Characterization of photocatalysts

Morphology

TEM morphologies of 550°C calcined 1.0% Al-Fe-Zn (1:1:1) co-doped TiO2 nanotubes (Figure 1(a)) indicate that the TiO2 nanotubes exhibited hollow, open-ended structures with a diameter of 6–8 nm, and the tube wall thickness was approximately 1 nm. The Ti, Al, Fe, Zn and O elements are involved in the energy dispersive X-ray (EDX) spectra of ion-doped TiO2.
Figure 1

Catalytic properties of 550°C calcined 1.0% Al-Fe-Zn co-doped TiO2 nanotubes: (a) TEM images; (b) XRD patterns; (c) UV-vis DRS and Eg.

Figure 1

Catalytic properties of 550°C calcined 1.0% Al-Fe-Zn co-doped TiO2 nanotubes: (a) TEM images; (b) XRD patterns; (c) UV-vis DRS and Eg.

XRD analysis

Figure 1(b) shows that the crystallite phase containing Al, Fe and Zn metal oxides was not observed in the XRD patterns. The ionic radii of Al3+ and Fe3+ were smaller than that of Ti4+. Therefore, the dopant ions could be highly dispersed in the TiO2 nanotubes (Pang & Abdullah 2012a). Zn ions were partially dispersed in the bulk of the TiO2 nanotubes. ZnO might be formed on the surface of TiO2 because the radius of the Zn2+ ion is 74 pm, a little larger than that of Ti4+. However, these oxides were not observed in the XRD patterns because their metal sites were expected to be below the visibility limit of X-ray analysis (Ravichandran et al. 2009).

The anatase phase exhibited a higher photocatalytic activity than the other phases (Pang & Abdullah 2012b). The titanate/amorphous phase transformed into the anatase phase as the calcination temperature increased. When the calcination continuously increases, the rutile phase is formed. Only the anatase phase was found in most of the TiO2 nanotubes, except for the 600°C calcined TiO2, which exhibited both the anatase and rutile phases. Meanwhile, the diffraction peaks of the anatase phase became weaker after ion-doping.

The Bragg angles decreased when ions were added because the crystallite could not be considered as an ideal crystal, suggesting that the Al3+, Fe3+ and Zn2+ ions could replace Ti4+ in the TiO2 nanotube lattice (Deng et al. 2009) and disrupt the arrangement of the TiO2 basic octahedral units, hence distorting the overall crystallite structure (Bokhimi et al. 2001). As a result, broadening diffraction peaks and decreasing peak intensities were observed.

Brunauer–Emmett–Teller (BET) surface area analysis

The SBET values of the TiO2 nanotubes (shown in Table 1) were significantly higher than that of P-25 TiO2 because of the inner and outer surfaces of the layered tubular structure (Pang & Abdullah 2012b). The SBET of the TiO2 nanotubes decreased as the calcination temperature increased, which is attributed to nanotube aggregation (Zhang et al. 2012). The SBET values of ion-doped TiO2 nanotubes were smaller than those of un-doped samples, and the SBET decreased as the concentration of the dopant increased (Pang & Abdullah 2012a), which was because of partial pore blockages and framework defects.

Table 1

Some catalytic properties and photocatalytic activities of Al-, Fe-, Zn- and co-doped TiO2 nanotubes

Properties
BET surface areas (m2/g)
Indirect energy bandgap (eV)
HAs removal efficiency (%)
Calcination temperature (°C)500550600500550600500550600
Un-doped  232 138 108 3.26 3.24 3.23 38.7 34.1 28.4 
Al-doped 0.1% 175 133 101 3.24 3.23 3.22 40.8 29.8 59.8 
0.5% 172 135 98 3.24 3.22 3.21 41.8 35.5 64.9 
1.0% 156 134 91 3.23 3.22 3.20 52.5 60.4 76.7 
5.0% 82 70 53 3.17 3.08 3.05 26.5 13.7 14.1 
Fe-doped 0.1% 173 136 114 3.16 3.16 3.17 42.5 50.3 32.8 
0.5% 147 123 105 3.12 3.12 3.12 57.8 63.7 46.4 
1.0% 143 120 87 3.07 3.06 3.05 72.1 77.4 37.5 
5.0% 132 112 81 2.85 2.75 2.72 43.0 28.8 33.2 
Zn-doped 0.1% 170 135 111 3.24 3.22 3.21 46.1 59.6 41.7 
0.5% 167 132 112 3.23 3.21 3.19 52.9 63.6 49.4 
1.0% 153 134 110 3.21 3.19 3.18 52.8 73.2 50.9 
5.0% 116 97 72 3.17 3.12 3.08 25.2 32.1 21.7 
Al:Fe:Zn (1.0% co-doped) 1:1:0  153   3.10   67.6  
1:0:1  172   3.15   65.4  
0:1:1  157   3.09   80.1  
1:1:1  175   3.13   71.8  
Properties
BET surface areas (m2/g)
Indirect energy bandgap (eV)
HAs removal efficiency (%)
Calcination temperature (°C)500550600500550600500550600
Un-doped  232 138 108 3.26 3.24 3.23 38.7 34.1 28.4 
Al-doped 0.1% 175 133 101 3.24 3.23 3.22 40.8 29.8 59.8 
0.5% 172 135 98 3.24 3.22 3.21 41.8 35.5 64.9 
1.0% 156 134 91 3.23 3.22 3.20 52.5 60.4 76.7 
5.0% 82 70 53 3.17 3.08 3.05 26.5 13.7 14.1 
Fe-doped 0.1% 173 136 114 3.16 3.16 3.17 42.5 50.3 32.8 
0.5% 147 123 105 3.12 3.12 3.12 57.8 63.7 46.4 
1.0% 143 120 87 3.07 3.06 3.05 72.1 77.4 37.5 
5.0% 132 112 81 2.85 2.75 2.72 43.0 28.8 33.2 
Zn-doped 0.1% 170 135 111 3.24 3.22 3.21 46.1 59.6 41.7 
0.5% 167 132 112 3.23 3.21 3.19 52.9 63.6 49.4 
1.0% 153 134 110 3.21 3.19 3.18 52.8 73.2 50.9 
5.0% 116 97 72 3.17 3.12 3.08 25.2 32.1 21.7 
Al:Fe:Zn (1.0% co-doped) 1:1:0  153   3.10   67.6  
1:0:1  172   3.15   65.4  
0:1:1  157   3.09   80.1  
1:1:1  175   3.13   71.8  

The SBET values of co-doped catalysts were higher than those of single ion-doped catalysts. When a second metal was doped into the crystal lattice, the catalyst surface structure changed because of the conservation of a large number of micropores (Thirupathi & Smirniotis 2011). However, the SBET of Al-Fe-Zn (1:1:1) co-doped TiO2 was not much higher than those of the co-doped TiO2 catalysts which were doped by only two dopants.

A larger SBET value indicates more pollutant could be adsorbed on the surface of the catalyst, and therefore affects the photocatalytic activities.

UV-vis DRS analysis

The Eg of the catalysts were calculated based on the UV-vis DRS analysis by the method given in the literature (Yuan et al. 2013), and the relationships between the Eg and the type and concentration of the doped ions are shown in Figure 1(c) and Table 1.

The Eg of TiO2 calcined at higher temperatures were wider than those calcined at lower temperatures. This tendency results from the narrower Eg of rutile TiO2 (3.0 eV) compared with that of anatase TiO2 (3.2 eV).

The Eg of undoped TiO2 nanotubes ranged from 3.23 to 3.26 eV, corresponding to the absorption of 384–381 nm. The reflectance spectra of ion-doped TiO2 nanotubes (Al-doped 383–407 nm, Fe-doped 391–456 nm, and Zn-doped 383–403 nm) slightly shifted toward longer wavelengths compared with those of un-doped specimens, and the Eg of ion-doped TiO2 nanotubes were visibly narrower. These results revealed that the dopant elements were successfully incorporated into the TiO2 nanotube lattice and changed its crystal and electronic structures (Yu et al. 2009). The Eg decreased as the concentration of dopant increased because of a charge-transfer transition between the new dopant level near the valence band and the conduction band of the TiO2 nanotubes (Pang & Abdullah 2012a). The reduction in the Eg of the TiO2 nanotubes allowed the excitation of the catalyst under a lower irradiation power and thus enhanced its photocatalytic activity (Pang & Abdullah 2013).

Moreover, the Eg of 1.0% co-doped TiO2 calcined at 550°C was between those of single ion-doped TiO2. The Eg of Al-Zn co-doped TiO2 was the widest because the Eg of Al-doped and Zn-doped TiO2 was wider than that of the Fe-doped TiO2.

Photocatalytic activities

The photocatalytic activities of TiO2 nanotubes were determined from the UV photocatalytic ozonation of HAs in the presence of the catalysts (Figure 2 and Table 1). The HA concentration decreased rapidly in the first 10 min, and then tended to be stable because the concentrations of O3 and HAs reduced significantly as the time went on, and the HA degradation during the first 10 min was consistent with second-order reaction.
Figure 2

Degradation of 10 mg/L HAs by the O3/UV/co-doped TiO2 process: (a) P-25 TiO2, 550°C calcined un-doped and 1.0% ion-doped TiO2; (b) 550°C calcined Al-Fe-Zn co-doped TiO2 (1.0% total).

Figure 2

Degradation of 10 mg/L HAs by the O3/UV/co-doped TiO2 process: (a) P-25 TiO2, 550°C calcined un-doped and 1.0% ion-doped TiO2; (b) 550°C calcined Al-Fe-Zn co-doped TiO2 (1.0% total).

A smaller SBET value indicates a smaller amount of adsorbed pollutant on the catalyst surface that can affect the photocatalytic activity. The reduction in the Eg of ion-doped TiO2 enhanced the photocatalytic activity. The HA removal efficiency in the presence of P-25 TiO2 (39.2%) was lower than that in the presence of ion-doped TiO2 nanotubes because the SBET values of the TiO2 nanotubes were larger, and the Eg were narrower.

Single ion-doped TiO2

The best photocatalytic activities were achieved when the ion-doping amount was 1.0%. The optimal calcination temperatures for Al-, Fe- or Zn-doped TiO2 were 600°C, 550°C and 550°C, corresponding to HA removal efficiencies of 76.7%, 77.4% and 73.2%, respectively. Another test showed that only approximately 1%–2% of HAs was adsorbed by the catalysts, indicating that the differences between the HA removal efficiencies were owing to the photocatalytic activities of the diverse catalysts. The SBET and Eg decreased as the calcination temperature and dopant concentration increased (Table 1). The photocatalytic activities are attributed to the SBET, Eg and crystalline phases of the catalysts (Zhang et al. 2000) which are dependent on the calcination temperature and the dopant concentration.

There are different reasons for the improvement of the photocatalytic activities of TiO2 after Al-, Fe- or Zn-doping. The ionic radii of Al3+ (53.5 pm) and Fe3+ (55 pm) are slightly smaller than that of Ti4+ (60.5 pm). After two doping ions replace two Ti4+ ions, one O2 hole appears. This phenomenon alters the crystal shape and promotes the generation of e-h+ pairs, which results in increased photoactivity (Yuan et al. 2013, 2014).

The ionic radius of Zn2+ is slightly larger than that of Ti4+. ZnO might be formed on the surface of TiO2. In the TiO2/ZnO composite, the e transfer occurs from the conduction band of light-activated ZnO to the conduction band of light-activated TiO2 and, conversely, h+ transfer can take place from the valence band of TiO2 to the valence band of ZnO (Sukharev & Kershaw 1996). This efficient charge separation increases the photocatalytic activity of the TiO2/ZnO composite. In addition, as the Zn2+ ions can be partially doped into the TiO2 lattices, the photogenerated e can be effectively scavenged by Zn2+ ions (Zang et al. 1995).

Co-doped TiO2

The 550°C calcined 1.0% Fe-Zn co-doped TiO2 nanotubes (Fe:Zn = 1:1) showed the highest catalytic activity (80.1% HA removal), and the rate constant was 0.394 L/(min·mg). The addition of a third metal ion had little effect on HA removal. SBET values increased when two or more metal ions were used as dopants (Table 1). The rutile phase was not found in the co-doped TiO2 (Figure 1(b)), which subsequently promoted the HA removal. The doped Fe3+ dissolved in TiO2 changed the color of Fe-doped and co-doped TiO2 nanotubes from white to orange, and the Eg of Fe-doped and co-doped samples decreased compared with that of un-doped TiO2. The Eg of 550°C calcined TiO2 containing 1:1 of Al:Fe and Fe:Zn were narrower than that of Al-Zn co-doped TiO2. The Eg of Fe-Zn co-doped TiO2 was not considerably wider than that of Fe-doped TiO2. Therefore, the highest photocatalytic activity that was achieved depended on all of the abovementioned reasons.

A red shifting tendency was observed for the doped catalysts compared with the un-doped samples. Given that the reactor was not protected from light during the experiments, the UV and part of the visible light in the laboratory would have promoted the degradation of HAs in the presence of ion-doped TiO2.

Mechanism of co-doped TiO2 photocatalytic process

The Fe-Zn co-doped TiO2 nanotubes exhibited high photocatalytic activity. A valence band e was promoted to the conduction band and left h+ which reacts with H2O (Equations (1)–(3)) (Pang & Abdullah 2013): 
formula
1
 
formula
2
 
formula
3
Fe3+ in the ion-doped TiO2 acted as shallow trapping sites for e and h+ (Equations (4) and (5)) and formed Fe2+ and Fe4+. The unstable Fe2+ and Fe4+ reacted with H2O2, OH and O2 and subsequently transformed into Fe3+ (Equations (6)–(10)) (Wei et al. 2011). Similarly to Fe3+, the inclusion of Zn2+ ions as dopants in TiO2 could also generate •OH and O2• radicals (Equations (11)–(17)) (Rauf et al. 2011): 
formula
4
 
formula
5
 
formula
6
 
formula
7
 
formula
8
 
formula
9
 
formula
10
 
formula
11
 
formula
12
 
formula
13
 
formula
14
 
formula
15
 
formula
16
 
formula
17
The combination of TiO2 with the doped ions resulted in the generation of Ti3+ (Equations (18) and (19)). Zn3+ could also react with e and Fe3+ (Equation (20)): 
formula
18
 
formula
19
 
formula
20

Treatment of filter influent through O3/UV/co-doped TiO2 process

In order to obtain the removal efficiency of NOM, the filter influent from No. 9 waterworks in Beijing was used, and the UV254 and TOC of the water were determined. The UV254 was attributed to aromatic compounds and other compounds containing C=C and C=O bonds in the water, and the TOC removal efficiency was used to evaluate the actual degree of HA mineralization caused by the co-doped TiO2 nanotubes. Figure 3 shows the changes in UV254 and TOC removal efficiency for the filter influent that was photocatalyzed in the presence of 550°C calcined TiO2 with an ion-doping concentration of 1.0% (Fe:Zn = 1:1).
Figure 3

TOC and UV254 removal for filter influent in the presence of 550°C calcined 1.0% Fe-Zn co-doped TiO2 nanotubes: (a) concentrations versus irradiation time; (b) reaction kinetics.

Figure 3

TOC and UV254 removal for filter influent in the presence of 550°C calcined 1.0% Fe-Zn co-doped TiO2 nanotubes: (a) concentrations versus irradiation time; (b) reaction kinetics.

The UV254 removal efficiency increased obviously during the first 5 min, and then the removal rate slowed down in the period of 5–15 min. The UV254 became stable after 15 min because of the degradation of aromatic compounds into intermediate molecules during photocatalysis, and the UV254 removal efficiency was 57.2%. As the photocatalytic reaction proceeded, an increasing number of intermediates were produced and may have hindered the adsorption process, eventually leading to the termination of the degradation of the pollutants on the surface of the co-doped TiO2 (Jiang et al. 2012). Meanwhile, the TOC rapidly decreased during the first 15 min of the reaction, and then the removal rate slowed down in the period of 15–20 min. When the reaction time exceeded 20 min, the TOC remained constant. The TOC removal rate was 44.7%, indicating that 44.7% of the pollutants were mineralized into CO2 and H2O. The second-order rate constants of UV254 and TOC were 1.4426 cm/min and 0.0162 L/(min·mg), respectively.

The removal efficiency of UV254 was higher than that of TOC because of the generation of intermediates that could significantly reduce the UV254, but not the TOC. The pH of water before and after the reaction was also measured. The results show that the pH decreased from 7.5 to 5.6, indicating the formation of carboxylic acids, which are oxidation-resistant compounds (Yuan et al. 2013). In the period between 15 and 20 min, the UV254 removal stopped but TOC removal continued because of the mineralization of intermediates. The TOC removal efficiency was lower than that of UV254 because of the presence of intermediates.

CONCLUSIONS

A series of Al-doped, Fe-doped, Zn-doped and co-doped TiO2 nanotubes were successfully synthesized by a hydrothermal method. Results of various characterization techniques, including XRD and UV-vis DRS, show that Al3+ and Fe3+ ions occupied substitutional positions in the crystal lattice, and Zn2+ ions were partially dispersed in the bulk of the TiO2 nanotubes, while ZnO might be formed on the surface of the TiO2. As the calcination temperature and ion doping concentration increased, the SBET decreased, and the SBET of co-doped catalysts were higher than those of single ion-doped catalysts. The Eg of ion-doped TiO2 nanotubes were narrower than those of un-doped TiO2 nanotubes.

HA degradation by the O3/UV/ion-doped TiO2 process followed second-order kinetics. The catalytic activities of the catalysts were affected by the SBET, Eg and the crystalline phases of the catalyst. For single ion-doped TiO2 nanotubes, the best photocatalytic activities were achieved when the ion-doping amount was 1.0%, and the optimal calcination temperatures for Al-, Fe- or Zn-doped TiO2 were 600°C, 550°C and 550°C, respectively. The highest HA removal efficiency was obtained in the presence of 1.0% co-doped TiO2 nanotubes containing Fe:Zn of 1:1 and calcined at 550°C. The addition of a third ion had little effect on HA removal.

The filter influent from No. 9 waterworks in Beijing was used to obtain the removal efficiency of NOM. The pollutants in the water could be efficiently removed. About 44.7% of the pollutants were mineralized into CO2 and H2O. The TOC removal efficiency was lower than that of UV254 because of the presence of intermediates.

ACKNOWLEDGEMENT

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (51178043).

REFERENCES

REFERENCES
Bokhimi
X.
Morales
A.
Aguilar
M.
Toledo-Antonio
J. A.
Pedraza
F.
2001
Local order in titania polymorphs
.
Int. J. Hydrogen Energy
26
(
12
),
1279
1287
.
Deng
L.
Wang
S.
Liu
D.
Zhu
B.
Huang
W.
Wu
S.
Zhang
S.
2009
Synthesis, characterization of Fe-doped TiO2 nanotubes with high photocatalytic activity
.
Catal. Lett.
129
(
3–4
),
513
518
.
Ravichandran
L.
Selvam
K.
Krishnakumar
B.
Swaminathan
M.
2009
Photovalorisation of pentafluorobenzoic acid with platinum doped TiO2
.
J. Hazard. Mater.
167
(
1–3
),
763
769
.
Sukharev
V.
Kershaw
R.
1996
Concerning the role of oxygen in photocatalytic decomposition of salicylic acid in water
.
J. Photochem. Photobiol. A: Chem.
98
,
165
169
.
Wei
M. C.
Wang
K. S.
Hsiao
T. E.
Lin
I. C.
Wu
H. J.
Wu
Y. L.
Liu
P. H.
Chang
S. H.
2011
Effects of UV irradiation on humic acid removal by ozonation, Fenton and Fe0/air treatment: THMFP and biotoxicity evaluation
.
J. Hazard. Mater.
195
,
324
331
.
Zhang
G. W.
He
G. H.
Xue
W. L.
Xu
X. F.
Liu
D. N.
Xu
Y. H.
2012
Enhanced photocatalytic performance of titania nanotubes modified with sulfuric acid
.
J. Mole. Catal. A: Chem.
363–364
,
423
429
.