Enhanced photocatalytic performance of N-nitrosodimethylamine on TiO2 nanotube based on the role of singlet oxygen

Chemosphere 120 (2015) 521–526

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Enhanced photocatalytic performance of N-nitrosodimethylamine on TiO2 nanotube based on the role of singlet oxygen Xiaoyan Guo a,⇑, Qilin Li b, Man Zhang a, Mingce Long c, Lulu Kong a, Qixing Zhou a,⇑, Huaiqi Shao d, Wanli Hu a, Tingting Wei a a College of Environmental Science and Engineering, Tianjin Key Laboratory of Environmental Remediation and Pollution Control, Ministry of Education Key Laboratory of Pollution Processes and Environmental Criteria, Nankai University, Wei Jin Road 94, Tianjin 300071, China b Department of Civil & Environmental Engineering, George R. Brown School of Engineering, Rice University, 6100 Main Street, Houston, TX 77005, United States c School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China d College of Material Science and Chemical Engineering, Tianjin University of Science & Technology, Thirteenth Street 29, TEDA, Tianjin 300457, China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 The anatase TiO2 nanotube was

prepared from anatase TiO2 nanopowder.  Almost 100% NDMA degradation efficiency was obtained on anatase TiO2 nanotube.  Tubular morphology may be responsible for the high NDMA removal on TiO2 nanotube. 1  A pathway initiated by O2 may contribute much to enhance NDMA degradation.

a r t i c l e

i n f o

Article history: Received 17 March 2014 Received in revised form 24 August 2014 Accepted 1 September 2014

Handling Editor: Tamara S. Galloway Keywords: N-nitrosodimethylamine Photocatalysis TiO2 nanotube Reactive oxygen species

a b s t r a c t N-nitrosodimethylamine (NDMA) photocatalytic degradation performance and mechanism were investigated on the TiO2 nanotube prepared from anatase TiO2 nanopowder in terms of the production of reactive oxygen species including hydroxyl radical, singlet oxygen and superoxide radical. Significantly higher NDMA degradation efficiency was obtained on anatase TiO2 nanotube rather than anatase TiO2 nanopowder. The tubular morphology may be responsible for almost 100% NDMA removal on TiO2 nanotube, presumably due to its confinement effect leading to NDMA molecules within the nanotube being attacked by reactive oxygen species such as hydroxyl radical and singlet oxygen, and initiating reaction inside the nanotube. In particular, the ability of the nanotubular structure of TiO2 nanotube to promote a singlet oxygen oxidation pathway contributes much to the enhanced NDMA degradation efficiency and favors the formation of dimethylamine and NO 3 . Such function originating from nanotube morphology could bring new insights for the photocatalytic degradation of organic pollutants. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction

⇑ Corresponding authors. Tel.: +86 22 66229528; fax: +86 22 66229535 (X. Guo). Tel.: +86 22 66229522; fax: +86 22 66229562 (Q. Zhou). E-mail addresses: [email protected] (X. Guo), [email protected] (Q. Zhou). http://dx.doi.org/10.1016/j.chemosphere.2014.09.002 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.

N-nitrosodimethylamine (NDMA) is a growing health concern and is present in groundwater, recycled water and treated surface water (Gerecke and Sedlak, 2003; Mitch et al., 2003). Many technologies, such as air stripping, adsorption, reverse osmosis (Fleming et al., 1996; Fujioka et al., 2012, 2013a, 2013b; Kong

522

X. Guo et al. / Chemosphere 120 (2015) 521–526

et al., 2010a, 2010b; Kommineni et al., 2003; Mitch et al., 2003), nanofiltration (Fujioka et al., 2013a, 2013b), biodegradation (Tate and Alexander, 1976; Mallik and Tesfai, 1981; Yang et al., 2005), UV photolysis (Mitch et al., 2003), advanced oxidation (Yang et al., 2009), photocatalytic degradation (Lee et al., 2005a, 2005b) and electrochemical treatment (Chaplin et al., 2009, 2010), have been tried to remove NDMA from contaminated waters. Among those, photocatalytic degradation has recently attracted considerable attention due to such merits as photochemical stability, non-toxicity and low cost (Hoffmann et al., 1995). Efficient photocatalytic degradation of NDMA over titanium dioxide (TiO2, Degussa P25) and surface-modified TiO2, which are the mixture nanopowder of anatase and rutile, has been reported by Lee et al. (2005a, 2005b). However, previous researchers (Augustynski, 1993; Hoffmann et al., 1995; Linsebigler et al., 1995; Mills and Le Hunte, 1997; Lai et al., 2006) showed that TiO2 in the anatase form had a better catalytic performance than the rutile form, mixtures of anatase and rutile, or amorphous structures; and moreover, TiO2 nanotubes presented a significantly better photocatalytic activity in comparison with a Degussa P25 nanopowder (Macak et al., 2007). Therefore, it is of great significance for developing more efficient NDMA photocatalytic degradation techniques to explore the photocatalytic performance of NDMA over anatase TiO2 and TiO2 nanotube prepared from anatase TiO2. Although photocatalytic degradation of environmental pollutants by TiO2 has been given considerable attention, details of the photocatalytic mechanism at the TiO2 surface are still controversial (Hufschmidt et al., 2004; Fujishima et al., 2008). Some researchers thought that photocatalytic oxidation proceeded via direct electron transfer between substrate and positive holes, while others argued for a hydroxyl radical-mediated pathway. Lee et al. (2005a, 2005b) thought that the photocatalytic degradation of NDMA seemed to be initiated exclusively by hydroxyl radical on pure and modified TiO2. However, more recently, the role of reactive oxygen species (ROS) other than hydroxyl radicals (OH), such as singlet oxygen (1O2) and superoxide radicals (O 2 ), is stressed in the photooxidation reactions (Raja et al., 2005; Daimon and Nosaka, 2007; Daimon et al., 2008; Fujishima et al., 2008; Zhang et al., 2009). The high quantum yield of 1O2 in photocatalytic reactions was confirmed by Nosaka (Daimon and Nosaka, 2007; Daimon et al., 2008), but due to the short lifetime in aqueous solution, its contribution to photodegradation reactions is limited in most cases. There are few investigations on the reaction pathways and product distributions that are regarded to be initiated by such ROS as singlet oxygen (1O2). This paper reports on the photocatalytic performances of NDMA over anatase TiO2 and TiO2 nanotube prepared from anatase TiO2, and the production of the principal ROS including OH, 1O2 and O 2 in the corresponding suspension. Furthermore, we will also discuss the photocatalytic mechanism of NDMA on TiO2 nanotube in terms of roles of ROS.

2. Materials and methods 2.1. Chemical and materials All chemicals were used as received in this study. NDMA (99.5% purity) was purchased from Wako Pure Chemical, Japan. Methanol (HPLC grade), acetonitrile (HPLC grade) and methanesulfonic acid (MSA, analytical grade) were obtained from Tianjin Concord Co., China. Methylamine (CH3NH2, 40% water solution), dimethylamine ((CH3)2NH, 40% water solution), sodium nitrite (NaNO2), sodium nitrate (NaNO3), sodium carbonate (Na2CO3), sodium hydroxide (NaOH), hydrochloride acid (HCl) and perchloric acid (HClO4) were analytical grade reagents procured from Tianjin Kermel Chemical Reagent Co., China. p-Chlorobenzoic acid (pCBA, >98% purity) was

purchased from Newprobe Co., Beijing, China. 2,3-bis(2-methoxy4-nitro-5-sulfophenyl)-2 h-tetrazolium-5-carboxanilide sodium salt (XTT sodium salt) and furfuryl alcohol (FFA, >98.5% purity) were obtained from Nanjing Duly Biotech Co., Ltd, China. NaN3 (>99% purity) was purchased from Tianjin Fengchuan Chemical Reagent Science and Technology Co., Ltd., China. All solutions were prepared in ultrapure water (18 MX cm). The nano-sized anatase TiO2 (Anatase) powder (grain size: 14 nm and BET surface area: 138 m2 g1) (Hehai Nanometer Science and Technology Co., Ltd, China) was used as the raw material for preparing TiO2 nanotube. 2.2. Preparation and characterization of TiO2 nanotube TiO2 nanotube was synthesized by the alkaline hydrothermal method with Anatase powder as the precursor material. Anatase (1 g) was mixed with 50 ml of NaOH aqueous solution (10 M) followed by hydrothermal treatment at 150 °C in a Teflon-lined autoclave for 24 h. The treated powder was washed carefully with 0.1 M HCl aqueous solution until the pH value of the washing solution was 2. After filtration and drying at 80 °C, the solid was heated in a furnace at a rate of 2 °C min1 to a calcination temperature of 380 °C for 2 h. The TiO2 nanotube prepared by using Anatase, was denoted as Anatase-TiNT. The morphology characteristics and crystal phase structure of the TiO2 nanotube were characterized by transmission electron microscopy (TEM, JEM-2100F, JEOL, Japan) and X-ray diffraction (XRD, Rigaku D/MAX 2200PC, Japan), shown in Figs. 1 and 2, respectively. Its BET surface area, obtained by a Nova 100 (Quantachrome Instruments) with nitrogen as the adsorption gas, was 110 m2 g1. 2.3. Photocatalysis, products and ROS analyses Photocatalytic degradation of NDMA and production of ROS on Anatase and Anatase-TiNT were performed in a 400 ml photocatalytic reactor consisting of two screwed cylindrical parts sealed by a gasket and placed horizontally in a constant temperature bath. Irradiated directly by a 500 W xenon arc lamp, all suspensions were prepared at a catalyst concentration of 0.5 g L1 and continuously dispersed by magnetic stirrer, the initial pH of the suspension was adjusted to a desired value with 1 M HClO4 or NaOH, and then kept in dark for 30 min to evaluate the impact of adsorption. Sample aliquots of 10 mL were withdrawn from the illuminated reactor with a syringe at regular time intervals, filtered through a 0.45 lm filter to remove catalysts, and injected into a 10 mL glass vial. A set of triplicate experiments was carried out, and the average values and the standard deviations are presented. A blank test of NDMA photolysis without catalysts under the same illumination showed that after 240 min irradiation, about 32% of NDMA was converted at neutral pH mainly to dimethylamine (DMA) and NO 2 . This photolysis can be attributed to the weak absorption of NDMA centered at 332 nm (n ? p⁄ transition band) (Chow et al., 1972; Lee et al., 2005a, 2005b; Plumlee and Reinhard, 2007). Quantitative analysis of NDMA was performed by high-performance liquid chromatography (HPLC, Waters 1525, UV detector) equipped with an XTerra RP18 column (Waters, 250 mm  4.6 mm, 5 lm packing material). The mobile phase used for NDMA was an isocratic 80/20 water/methanol mixture with flow rate of 1 mL min1 and monitored at 228 nm. The main products of the  NDMA photocatalysis (methylamine (MA), DMA, NO 3 , and NO2 ) were analyzed by ion chromatograph (IC, Dionex, DX 120) with a conductivity detector (Kong et al., 2010a, 2010b). MSA (3 mM) or Na2CO3/NaHCO3 (1.8 mM/1.7 mM) at a flow rate of 1.2 mL min1 served as the eluent for cation or anion measurement, respectively. The concentration of ROS was monitored by using specific scavengers. The production of OH was monitored via the loss of pCBA,

X. Guo et al. / Chemosphere 120 (2015) 521–526

523

Fig. 1. TEM images of Anatase and Anatase-TiNT.

Fig. 2. XRD patterns of Anatase and Anatase-TiNT.

through which the pseudo-steady-state concentration of OH was also determined. XTT sodium salt and FFA was used to detect O 2 and 1O2, respectively. The concentrations of pCBA and FFA were measured by HPLC (Waters 1525, UV detector) equipped with an XTerra RP18 column (Waters, 250 mm  4.6 mm, 5 um packing material). The mobile phase for pCBA determination consisted of 86/14 (v/v) 40 mM phosphate buffer/acetonitrile. The flow rate was 1.0 mL min1 and determination was at 230 nm. A solvent mixture of 50/50 (v/v) water/methanol was used as the mobile phase for FFA detection. The presence of O 2 was determined by quantifying XTT sodium salt by a T6 new century UV–Vis spectrophotometer (Beijing Purkinje General Instrument Co., China) at its absorbance maximum (470 nm). A more complete description of the methods used to detect and quantify the ROS has been provided previously (Chen and Jafvert, 2010). NaN3 (250 mM), an effective quencher of singlet oxygen, was added to the photocatalytic reaction system to observe its effect on the photocatalytic efficiency and products distribution of NDMA over Anatase and Anatase-TiNT, respectively (Zhang et al., 2009). 3. Results and discussion 3.1. Characterization of TiO2 nanotube Figs. 1 and 2 show the TEM images and XRD patterns of the asprepared TiO2 nanotube and its precursor nanopowder, respectively. The Anatase has a particle diameter of about 20–25 nm, and the Anatase-TiNT was successfully obtained by the hydrothermal synthesis, showing the obvious tubular morphology and welldefined structure. The nanotubes in the sample have an average diameter of 10 nm, with a tube wall of about 2 nm and a tube length of several tens of nm. In comparison with the changes of

morphology, the crystal phase composition of the Anatase-TiNT was quite consistent with its precursor nanopowder in the form of anatase, with only a slightly decreased intensity. 3.2. Photocatalytic degradation of NDMA on TiO2 nanotube The time profiles of NDMA photocatalytic degradation on the Anatase-TiNT and its precursor nanopowder photocatalysts at solution pH of 7 are shown in Fig. 3. When the NDMA photocatalytic degradation performances between these two photocatalysts are compared, NDMA degradation on Anatase-TiNT was found to proceed at a faster rate and the product distribution was very different from that obtained on Anatase. It is noticed that DMA and  NO 3 were evolved as the major products and detectable NO2 was produced on the Anatase-TiNT, moreover, NDMA removal efficiency on Anatase-TiNT is also significantly higher than that of the Degussa P25 nanopowder (Lee et al., 2005a, 2005b), reaching almost 100% after 240 min irradiation. These results can be ascribed to the tubular morphology of Anatase-TiNT with anatase phase. This finding is well in line with literature reporting that TiO2 nanotubes show a significantly better photocatalytic activity for the decomposition of organic azo dyes in comparison with a Degussa P25 nanopowder (Macak et al., 2007). In addition, the total N balance presented in Fig. 3 was satisfactory in both cases, suggesting the NDMA photocatalytic degradation products were  mainly composed of MA, DMA, NO 2 and NO3 . 3.3. Mechanism of photocatalytic degradation of NDMA on TiO2 nanotube based on the roles of ROS A general process for photocatalysis initiates from the photoexcitation of TiO2 with an energy greater than its band gap, resulting

524

X. Guo et al. / Chemosphere 120 (2015) 521–526

Fig. 3. Time profiles of NDMA photocatalytic degradation on Anatase and Anatase-TiNT at solution pH of 7.

in the formation of active electron–hole pairs (e h+) (Eqs. (1) and (2)) (Dung et al., 1982; Rothenberger et al., 1985). The separated photogenerated electrons and holes can respectively react with molecular oxygen, water or surface absorbed groups to generate 1  ROS, including OH, O 2 and O2. The main process for OH and O generation is described in following equations (Eqs. (3) and 2 + 1 (4)). Moreover, O 2 can be oxidized by h to form O2 (Eq. (5)) (Daimon and Nosaka, 2007; Tachikawa et al., 2007; Daimon et al., 2008). In fact, the generation processes of 1O2 are more complicated than this. It can form by the combination of O 2 (Eq. (7)), by a reaction with H2O2 (Eq. (8)) or OH (Eq. (9)) (Atlante and Passarella, 1999; Fujishima et al., 2008), and via an energy transfer pathway as described in Eqs. (10) and (11) (Tachikawa et al., 2007). These ROS have been identified as the predominant reactive intermediates responsible for photocatalytic reactions (Raja et al., 2005; Daimon and Nosaka, 2007; Tachikawa et al., 2007; Daimon et al., 2008; Fujishima et al., 2008; Zhang et al., 2009)].

½TiO2  þ light ! ½TiO2 



ð1Þ

þ



½TiO2  ! ½TiO2  þ h þ e

ð2Þ

þ

H2 O þ h !  OH þ Hþ

ð3Þ

O2 þ e ! O 2

ð4Þ

þ

1 O 2 þ h ! O2

ð5Þ

O2 þ 2e þ 2Hþ ! H2 O2

ð6Þ

 þ 1 O 2 þ O2 þ 2H ! H2 O2 þ O2

ð7Þ

 1  O 2 þ H2 O2 ! O2 þ OH þ OH

ð8Þ

  1 O 2 þ OH ! O2 þ OH

ð9Þ

½TiO2  þ light ! ½TiO2 



ð10Þ



½TiO2  þ 3 O2 ! TiO2 þ 1 O2

ð11Þ 1 O 2 , O2

Fig. 4 shows the production of ROS including and OH on Anatase and Anatase-TiNT. According to the concentration data, the final production of OH, which could be the main ROS in the Anatase and Anatase-TiNT suspension, is more than one order of  magnitude higher than O 2 . The highest concentration of OH is understandable because the separation of photogenerated charges on Anatase and Anatase-TiNT is facile due to the anatase crystal phases (Augustynski, 1993; Linsebigler et al., 1995; Mills and Le

Fig. 4. Time profiles of ROS generated from Anatase and Anatase-TiNT (the solution pH is 7).

Hunte, 1997; Lai et al., 2006). At the same time, the O 2 could be quickly consumed according to the Eqs. (5)–(9), resulting in a lower concentration in the irradiated suspension. The higher concentration of 1O2 in the irradiated Anatase and Anatase-TiNT suspension shown in Fig. 4 is the evidence for such changes. In most cases, OH is the ROS dominating the organic degradation because of its relatively long lifetime and significant oxidation power. In a previous report (Lee et al., 2005a, 2005b), OH was thought to play the most important role for NDMA degradation, which attacked one of three positions on NDMA: the methyl group, the amine nitrogen and the nitrosyl nitrogen, obtaining the products distribu tion: MA, NO 3 , DMA, NO2 , etc. Hence, the high concentration of  OH and the products distribution shown in Fig. 3 may imply the important role that OH plays in the NDMA photocatalytic degradation on Anatase and Anatase-TiNT. However, it is interesting that the NDMA degradation efficiency on Anatase-TiNT is significantly higher than Anatase, despite the lower OH generation from Anatase-TiNT than that from Anatase during the photocatalysis process, even if they are very close at the end of reaction. This indicates that species other than OH may make a significant contribution to NDMA photocatalysis. Of the three kinds of ROS, the amount of 1O2 was only found to slightly increase from Anatase to the Anatase-TiNT, which may be due to a higher energy transfer process for Anatase-TiNT with tubular morphology (Tatsuma et al., 1999; Janczyk et al., 2006; Riss et al., 2007). The reason for this could be that there was no change in the crystal phase, but a decreased crystallinity leading to the lower production of OH and O 2 for Anatase-TiNT.

X. Guo et al. / Chemosphere 120 (2015) 521–526

525

Fig. 5. Degradation efficiency, products distribution and total N balance at 240th min of NDMA photocatalytic degradation on Anatase-TiNT and Anatase with and without 1 O2 quenching by NaN3.

H 3C H 3C

N

N

NDMA

O +

1

O2

H2 O

H 3C H 3C

NH2+ +

DMA

-

O

O

+ N

-

O-

contribution to NDMA photocatalytic degradation on Anatase with nanopowder morphology.

4. Conclusions

NO31

Scheme 1. Pathway for the photocatalytic reaction of NDMA initiated by O2.

There are only a few reports on the role of 1O2 in photocatalytic reactions (Nosaka et al., 2004; Raja et al., 2005; Daimon and Nosaka, 2007; Daimon et al., 2008), but the quantum yields of 1O2 in different TiO2 suspensions were measured by Nosaka and the significantly high values suggested that it may contribute to the oxidation of organic molecules. In comparison with the amounts of ROS produced and the photocatalytic degradation rates for NDMA between Anatase and Anatase-TiNT, it seems likely that 1O2 plays an important role for NDMA degradation on the nanotubular photocatalyst. The logical reason that makes this possible must be the confinement effect of nanotube morphology. One fact supporting this is that the size of NDMA molecule is about 0.45 nm (Zhu et al., 2001), less than the 5–6 nm diameters of nanotube. Therefore it is likely that NDMA molecules enter the nanotube, and be attacked by a ROS initiated reaction inside the nanotube. In such kind of confined situation, 1O2, which has a high quantum yield in photocatalytic process, can involve the degradation of NDMA at a significant rate. Considering DMA and NO 3 become to be the main products when NDMA photocatalytic degradation takes place on Anatase-TiNT photocatalyst, a pathway favoring the formation of DMA and NO 3 might be involved in the NDMA degradation. In order to confirm the possible role of 1O2 in NDMA degradation on Anatase-TiNT, the photocatalytic degradation experiment was carried out in the presence of excess NaN3 as a 1O2 scavenger (Zhang et al., 2009). Fig. 5 compares the photocatalytic degradation efficiency of NDMA and the concurrent product generation in the Anatase-TiNT and Anatase suspensions with or without NaN3, respectively. It is noticed that the addition of excess NaN3 as a 1O2 scavenger significantly inhibited the photocatalytic degradation of NDMA and the production of DMA and NO 3 on Anatase-TiNT, rather than on Anatase, which indicates that a pathway favoring the production of DMA and NO 3 is involved in the photocatalytic degradation of NDMA on AnataseTiNT, probably initiated by 1O2, as shown in Scheme 1. Moreover, we can see that the NDMA degration efficiency and products are only slightly different when NaN3 is added as a 1O2 scavenger to an Anatase mediated oxidation, which demonstrates that 1O2 has little

We believe that high NDMA photocatalytic degradation efficiency is achieved by Antase-TiNT, whose nanotubular structure promotes the 1O2 oxidation pathway, contributes much to the NDMA degradation and favors the formation of DMA and NO 3. Such function originating from nanotube morphology could bring new insight for the photocatalytic degradation of other organic pollutants. Moreover, selective removal of DMA would further enhance the NDMA removal efficiency; this is to be the subject of future work. However, it should be noted photocatalytic degradation of organics is too complicated to be described by any mechanism involving only one kind of ROS. Therefore, research on the roles of different kinds of ROS in photocatalytic degradation of organics has great significance for providing much more insight into the photocatalytic mechanisms. Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant Nos. 50808102 and 20907031). We gratefully acknowledge the valuable suggestion from Prof. Chen Wei of Nankai University. References Atlante, A., Passarella, S., 1999. Detection of reactive oxygen species in primary cultures of cerebellar granule cells. Brain Res. Protoc. 4, 266–270. Augustynski, J., 1993. Characteriyation of electrodeposited TiO2 films. Electrochim. Acta 38, 43–47. Chaplin, B.P., Schrader, G., Farrell, J., 2009. Electrochemical oxidation of Nnitrosodimethylamine with boron-doped diamond film electrodes. Environ. Sci. Technol. 43, 8302–8307. Chaplin, B.P., Schrader, G., Farrell, J., 2010. Electrochemical destruction of Nnitrosodimethylamine in reverse osmosis concentrates using boron-doped diamond film electrodes. Environ. Sci. Technol. 44, 4264–4269. Chen, C.Y., Jafvert, C.T., 2010. Photoreactivity of carboxylated single-walled carbon nanotubes in sunlight: reactive oxygen species production in water. Environ. Sci. Technol. 44, 6674–6679. Chow, Y.L., Lau, M.P., Perry, R.A., Tam, J.N.S., 1972. Photoreactions of nitroso compounds in solution. XX. Photoreduction, photoelimination, and photoaddition of nitrosamines. Can. J. Chem. 50, 1044–1050. Daimon, T., Nosaka, Y., 2007. Formation and behavior of singlet molecular oxygen in TiO2 photocatalysis studied by detection of near-infrared phosphorescence. J. Phys. Chem. C 111, 4420–4424.

526

X. Guo et al. / Chemosphere 120 (2015) 521–526

Daimon, T., Hirakawa, T., Kitazawa, M., Suetake, J., Nosaka, Y., 2008. Formation of singlet molecular oxygen associated with the formation of superoxide radicals in aqueous suspensions of TiO2 photocatalysts. Appl. Catal. A 340, 169–175. Dung, D., Ramsden, J., Graetzel, M., 1982. Dynamics of interfacial electron-transfer processes in colloidal semiconductor systems. Am. Chem. Soc. 104, 2977–2985. Fleming, E.C., Pennington, J.C., Wachob, B.G., Howe, R.A., Hill, D.O., 1996. Removal of N-nitrosodimethylamine from waters using physical–chemical techniques. J. Hazard. Mater. 51, 151–164. Fujioka, T., Khan, S.J., Poussade, Y.J., Drewes, E., Nghiem, L.D., 2012. N-nitrosamine removal by reverse osmosis for indirect potable water reuse – a critical review based on observations from laboratory, pilot and full scale studies. Sep. Purif. Technol. 98, 503–515. Fujioka, T., Khan, S.J., McDonald, J.A., Henderson, R.K., Poussade, Y., Drewes, J.E., Nghiem, L.D., 2013a. Effects of membrane fouling on N-nitrosamine rejection by nanofiltration and reverse osmosis membranes. J. Membr. Sci. 427, 311–319. Fujioka, T., Khan, S.J., McDonald, J.A., Roux, A., Poussade, Y., Drewes, J.E., Nghiem, L.D., 2013b. N-nitrosamine rejection by nanofiltration and reverse osmosis membranes: the importance of membrane characteristics. Desalination 316, 67–75. Fujishima, A., Zhang, X., Tryk, D.A., 2008. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 63, 515–582. Gerecke, A.C., Sedlak, D.L., 2003. Precursors of N-nitrosodimethylamine in natural waters. Environ. Sci. Technol. 37, 1331–1336. Hoffmann, M.R., Martin, S.T., Choi, W., Bahnemann, D.W., 1995. Environmental applications of semiconductor photocatalysis. Chem. Rev. 95, 69–96. Hufschmidt, D., Liu, L., Seizer, V., Bahnemann, D., 2004. Photocatalytic water treatment: fundamental knowledge required for its practical application. Water Sci. Technol. 49, 135–140. Janczyk, A., Krakowska, E., Stochel, G., Macyk, W., 2006. Singlet oxygen photogeneration at surface modified titanium dioxide. J. Am. Chem. Soc. 128, 15574–15575. Kommineni, S., Ela, W.P., Arnold, R.G., Huling, S.G., Hester, B.J., Betterton, E.A., 2003. NDMA treatment by sequential GAC adsorption and Fenton-driven destruction. Environ. Eng. Sci. 20, 361–373. Kong, L., Guo, X., Zhou, Q., Hu, W., Yun, H., Chen, C., Lu, J., 2010a. Impact of eluent concentration on resolution of low molecular weight aliphatic amine at low levels by ion chromatography. Chin. J. Anal. Chem. 38, 1191–1194. Kong, L., Guo, X., Zhou, Q., Li, Q., Hu, W., Lu, J., 2010b. Degradation methods of NDMA in surface and drinking water. Prog. Chem. 22, 734–739. Lai, Y., Sun, L., Chen, Y., Zhuang, H., Lin, C., Chin, J.W., 2006. Effects of the structure of TiO2 nanotube array on Ti substrate on its photocatalytic activity. J. Electrochem. Soc. 153, D123–D127. Lee, C., Choi, W., Kim, Y.G., Yoon, J., 2005a. UV photolytic mechanism of Nnitrosodimethylamine in water: dual pathways to methylamine versus dimethylamine. Environ. Sci. Technol. 39, 2101–2106. Lee, J., Choi, W., Yoon, J., 2005b. Photocatalytic degradation of Nnitrosodimethylamine: mechanism, product distribution, and TiO2 surface modification. Environ. Sci. Technol. 39, 6800–6807.

Linsebigler, A.L., Lu, G., Yates, J.T., 1995. Photocatalysis on TiO2 surfaces—principles, mechanisms, and selected results. Chem. Rev. 95, 735–758. Macak, J.M., Zlamal, M., Krysa, J., Schmuki, P., 2007. Self-organized TiO2 nanotube layers as highly efficient photocatalysts. Small 3, 300–304. Mallik, M.A.B., Tesfai, K., 1981. Transformation of nitrosamines in soil and in vitro by soil microorganisms. Bull. Environ. Contam. Toxicol. 27, 115–121. Mills, A., Le Hunte, S., 1997. An overview of semiconductor photocatalysis. J. Photochem. Photobiol., A 108, 1–35. Mitch, W.A., Sharp, J.O., Trussell, R.R., Valentine, R.L., Alvarez-Cohen, L., Sedlak, D.L., 2003. N-nitrosodimethylamine (NDMA) as a drinking water contaminant: a review. Environ. Eng. Sci. 20, 389–404. Nosaka, Y., Daimon, T., Nosaka, A.Y., Murakami, Y., 2004. Singlet oxygen formation in photocatalytic TiO2 aqueous suspension. Phys. Chem. Chem. Phys. 6, 2917– 2918. Plumlee, M.H., Reinhard, M., 2007. Photochemical attenuation of Nnitrosodimethylamine (NDMA) and other nitrosamines in surface water. Environ. Sci. Technol. 41, 6170–6176. Raja, P., Bozzi, A., Mansilla, H., Kiwi, J., 2005. Evidence for superoxide-radical anion, singlet oxygen and OH-radical intervention during the degradation of the lignin model compound (3-methoxy-4-hydroxyphenylmethylcarbinol). J. Photochem. Photobiol., A 169, 271–278. Riss, A., Berger, T., Grothe, H., Bernardi, J., Diwald, O., Knozinger, E., 2007. Chemical control of photoexcited states in titanate nanostructures. Nano Lett. 7, 433–438. Rothenberger, G., Moser, J., Gratzel, M., Serpone, N., Sharmaf, D.K., 1985. Charge carrier trapping and recombination dynamics in small semiconductor particles. J. Am. Chem. Soc. 107, 8054–8059. Tachikawa, T., Fujitsuka, M., Majima, T., 2007. Mechanistic insight into the TiO2 photocatalytic reactions: design of new photocatalysts. J. Phys. Chem. C 111, 5259–5275. Tate, R.L., Alexander, M., 1976. Resistance of nitrosamines to microbial attack. J. Environ. Qual. 5, 131–133. Tatsuma, T., Tachibana, S., Miwa, T., Tryk, D.A., Fujishima, A., 1999. Remote bleaching of methylene blue by UV-irradiated TiO2 in the gas phase. J. Phys. Chem. B 103, 8033–8035. Yang, W., Gan, J., Liu, W., Green, R., 2005. Degradation of N-nitrosodimethylamine in landscape soils. J. Environ. Qual. 34, 336–341. Yang, L., Chen, Z., Shen, J., Xu, Z., Liang, H., Tian, J., Ben, Y., Zhai, X., Shi, W., Li, G., 2009. Reinvestigation of the nitrosamine-formation mechanism during ozonation. Environ. Sci. Technol. 43, 5481–5487. Zhang, D., Qiu, R., Song, L., Eric, B., Mo, Y., Huang, X., 2009. Role of oxygen active species in the photocatalytic degradation of phenol using polymer sensitized TiO2 under visible light irradiation. J. Hazard. Mater. 163, 843–847. Zhu, J.H., Yan, D., Xai, J.R., Ma, L.L., Shen, B., 2001. Attempt to adsorb N-nitrosamines in solution by use of zeolites. Chemosphere 44, 949–956.