Skip to Main Content
Skip Nav Destination

This chapter discusses thoroughly the outcomes of the TiO2 photocatalytic degradation of organic contaminants of emerging concern, including manmade (insecticides, organochlorinated compounds, and antibiotics) and naturally occurring compounds (cyanotoxins and taste and odor compounds). Specifically, information is provided on the degradation of various organic contaminants in actual water samples, their corresponding reaction kinetics, the individual effects of water quality parameters (including pH, natural organic matter, and alkalinity) and nano-interfacial adsorption phenomena. Emphasis is given to the mechanisms of photocatalytic degradation of organic contaminants based on their structural differences and the corresponding transformation products formed.

Ever since Fujishima and Honda (1972) successfully applied titanium dioxide (TiO2) for the photoelectrochemical splitting of water, TiO2 has been widely utilized as a remediation technology for the removal of pollutants from the air, drinking water, and wastewater.1  TiO2 is considered an attractive alternative to other applied technologies because of its superior photo-activity, low toxicity, chemical and biological inertness, low cost, and corrosion resistance.2,3  TiO2 exists in three different polymorphic forms: anatase, rutile and brookite with band gaps of 3.2, 3.0, and ∼3.2 eV, respectively, activated in the UV range. Anatase and rutile are the most common polymorphs, with the anatase phase possessing higher photocatalytic activity than rutile and brookite.4  Higher reactivity is associated with the number and type of reactive oxygen or free radicals species (ROS) formed that react and transform the targeted pollutants first into smaller molecular weight products and ultimately mineralize them into simpler and non-toxic products H2O, CO2, and mineral acids. ROS such as hydroxyl radical (HO˙), superoxide anion radical (O2˙), hydroperoxyl radical (HO2˙), and singlet oxygen (1O2) are mainly produced during UV TiO2 photocatalysis.5–7  Hydroxyl radicals, generated on the surface of the catalyst following oxidation of water from the positive holes of TiO2, are non-selective oxidizing species with strong oxidation potential (+2.80 V) that rapidly react with most organic compounds with rate constants of the order of 106–1010M−1 s−1.8–10  While several ROS are formed during UV TiO2 photocatalysis, the effectiveness of the degradation process is generally dependent on the production and subsequent reactions of HO˙ at or very near the TiO2 surface.11,12  The surface adsorption and reactivity of the target compound with HO˙ are key factors in the UV promoted TiO2 photocatalytic degradation of organic compounds. Several heterogeneous kinetic models have been used to determine apparent kinetic parameters and effectively model the observed degradation processes.5  Detailed product studies as well as structure–reactivity relationships provide convincing evidence that generated HO˙ adsorbed at the surface of TiO2 has similar reactivity to HO˙ in a homogenous solution.13  Radiolysis methods are commonly used to study the HO˙ induced oxidative degradation of problematic pollutants and toxins. Radiolytic and UV TiO2 photocatalytic degradation of organic compounds can follow similar degradation pathways.14,15  Under photocatalytic conditions the adsorption of a target compound at a higher surface concentration than in the bulk solution can enhance the rate of HO˙ induced degradation since both the oxidant and target are localized at the surface. For the purpose of this chapter the differences in reactivities of free and adsorbed HO˙ will not be discussed.

Hydroxyl radicals generally react with organic compounds via three pathways: electrophilic addition (eqn (1.1)), hydrogen atom abstraction (eqn (1.2)) and electron transfer (eqn (1.3)). The hydroxyl radical additions to carbon–carbon double bonds and electron-rich aromatic systems leading to hydroxylated adducts are generally faster than H-abstraction and electron transfer pathways. Electron transfer with hydroxyl radical is limited to very electron rich systems.8  The electron transfer pathway generates a radical cation which in aqueous media is hydrolyzed, often generating the same hydroxylated adducts observed by direct addition of HO˙. The H-abstraction pathways result in oxidation of a sp3 hybridized carbon. The hydroxyl radical induced oxidation of primary (1°), secondary (2°), and tertiary (3°) aliphatic carbons is initiated by hydrogen atom abstraction. The hydroxyl radical mediated oxidation of a 1° alcohol via the H-atom abstraction pathways are given in eqn (1.4). The presence of a heteroatom attached to the carbon significantly enhances the rate of hydrogen atom abstraction and leads to a stabilized carbon-centered radical. Such stabilized carbon-centered radicals can add a molecular oxygen and subsequently eliminate a hydroperoxyl radical to yield the corresponding oxidation product.

Equation 1.1
Equation 1.2
Equation 1.3
Equation 1.4

UV/TiO2 photocatalytic and radiolytic oxidations of benzene and related aromatic compounds have been extensively studied.16,17  A primary product of UV/TiO2 photocatalytic degradation of benzene is phenol, which results from the reaction of HO˙ addition to the benzene ring. The TiO2 photocatalysis of benzene involves HO˙ addition to form hydroxylated and polyhydroxylated adducts, which upon further oxidation can be converted into quinone type structures. Subsequent oxidation can also lead to ring open products, affording low molecular weight carboxylic acids and ultimately CO2. 13C isotope labeling was used to establish different pathways involved in the formation of reaction products, muconaldehyde and phenol17,18  (Figure 1.1). Muconaldehyde can be readily oxidized to the corresponding carboxylic acids, with extensive oxidation leading to CO2 as final product. The reaction pathways leading to muconaldehyde account for 60–70% and pathways via phenol account for 30–40% of benzene consumption.17 

Figure 1.1

Two pathways for TiO2 photocatalysis of benzene. (Reproduced with permission from ref. 17. Copyright 2011 Elsevier.)

Figure 1.1

Two pathways for TiO2 photocatalysis of benzene. (Reproduced with permission from ref. 17. Copyright 2011 Elsevier.)

Close modal

Halogenated organic compounds are among the most problematic water pollutants. The presence of chlorine, fluorine, bromine, and iodine atoms can dramatically affect the toxicity and treatability of organic compounds. Halo-organics have been extensively used as herbicides, pesticides, and in several personal care products.10,19,20  Because of the negative health environmental consequences associated with many halogenated organic compounds extensive studies on TiO2 photocatalytic degradation of halogenated compounds have been reported.19,20  As an example, the degradation mechanisms of 1,2-dichlorobenzene are initiated with HO˙ addition to the 3-position of 1,2-dichlorobenzene to form 2,3-dichlorophenol (1A) or substitution of chloro by HO˙ to generate 2-chlorophenol (1B) (Figure 1.2).19  Intermediates produced along pathway A are 2,3-dichlorophenol (1A), 2,3-dihydroxychlorobenzene (2A) and 1,2,3-trihydrobenzene (3A) and further oxidation to CO2. Analogous substitution in pathway B can form 1,2-dihydroxybenzene (4B) and o-benzoquinone (5B).

Figure 1.2

Proposed pathways for TiO2 photocatalysis of 1,2-dichlorobenzene. (Reproduced with permission from ref. 19. Copyright 2002 Elsevier.)

Figure 1.2

Proposed pathways for TiO2 photocatalysis of 1,2-dichlorobenzene. (Reproduced with permission from ref. 19. Copyright 2002 Elsevier.)

Close modal

Phosphorus, sulfur and nitrogen containing organic substrates are also a target of TiO2 photocatalytic degradation due to their wide usage for insect controls in agricultural crops, pharmaceutical, and personal care products.2,21  These heteroatom containing compounds have widespread presence in groundwater and sediment. While there are numerous successful cases of remediation of such compounds, we choose to highlight examples of UV/TiO2 photocatalytic degradation of select phosphorus, sulfur, and nitrogen containing compounds. The TiO2 photocatalysis of organophosphorus insecticides (dimethyl 2,2-dichlorovinyl phosphate and dimethyl 2,2,2-trichloro-1-hydroxyethyl phosphate) has been investigated and the final degradation products are mineralized to Cl, PO43−, H+, and CO2.22  The photocatalytic degradation of sulfur and nitrogen containing molinate under simulated solar irradiation has been studied using a suspension of TiO2 in aqueous media.21  Different reaction pathways are involved in the degradation of molinate by the positive holes and HO˙. Ring fragmentation at the amine group is initiated via hole-mediated oxidation to form an immonium radical cation which undergoes hydrolysis to yield amine and carboxyl derivatives.23  An H-abstraction pathway leads to the alkyl radical, which undergoes addition of molecular oxygen with the subsequent elimination of peroxyl radical to yield keto-molinate. The holes can also mediate oxidation through the sulfur atom by an electron transfer process leading to sulfoxide derivatives and de-alkylated products.21  Extensive treatment can lead to complete mineralization with the formation of NH3, CO2, SO42−, HX, and H2O.

In the following sections, the degradation kinetics of TiO2 photocatalysis with various pollutants will be extensively discussed along with the effect of water quality parameters such as pH, natural organic matter (NOM), and alkalinity. Finally, a thorough discussion of the transformation products formed that lead to the degradation pathways of select compounds is given.

Photocatalysis has become one of the most effective approaches to degrade highly toxic naturally produced compounds, the cyanotoxins, that cannot be easily removed through conventional treatment processes. A lot of effort has been made to investigate the TiO2 photocatalytic degradation of various cyanotoxins, such as microcystins (MCs) and cylindrospermopsin (CYN), under UV or solar irradiation. Shephard et al. reported the high degradation efficiency of microcystin-LR (MC-LR) and MC-RR using immobilized TiO2 on a fiberglass sheet under UVC irradiation with the observed first rate constants of 0.255 ± 0.017 and 0.199 ± 0.016 min−1, respectively.24  The photocatalytic degradation of [d-Leu]-MC-LR using Degussa P25 TiO2 film under simulated solar light was investigated by Vilela et al.25  They reported a treatment time of 150 min could decrease the toxin concentration from 10 to 1 µg L−1 and remove 90% initial total organic carbon (TOC) under the experimental condition. Su et al. investigated the high degradation efficiency of MC-LR with a much higher concentration (5 mg L−1) using TiO2 nanotubes as the photocatalyst under the irradiation of UV and natural solar light.26  This study confirmed a better performance of TiO2 nanotubes than TiO2 nanofilm. Senogles et al. examined the effectiveness of two commercial TiO2 nanoparticles under UV light for the removal of CYN and found Degussa P-25 was more efficient than Hombikat UV-100.27  In addition, a polymorphic TiO2 nanoparticle investigated by Zhang et al.,28  containing anatase, brookite and rutile phases, showed good performance for CYN photocatalytic removal.

To extend the light response of TiO2 into the visible region, many studies focused on the modification of TiO2 by doping with metals (Fe, Co, Ag, or Ni)29  or non-metals (N, F, S, or C).30–33  This causes narrowing of the bandgap of TiO2 or introduces mid-gap energy states. The study of Choi et al.34  reported the efficient removal of MC-LR using a self-synthesized nitrogen-doped TiO2 (N–TiO2) under visible light. In addition, the N–TiO2 was found to be 3–4 times more efficient than control TiO2 for the degradation of MC-LR under UVA irradiation. Another study compared the removal efficiency of MC-LR, -LA, -YR, -RR, and CYN using N and F doped TiO2 films under visible or UV-vis irradiation.35  In a mixture of MCs, the reaction rate of an individual toxin under UV-vis irradiation was shown in this study as follow: MC-LA ≥ MC-LR ≥ MC-YR > MC-RR, which was attributed to the extent of adsorption of an individual toxin.

Oxytetracycline (OTC) belongs to tetracyclines, one of the most common antibiotics group for disease treatment and prevention, and growth promotion in livestock worldwide. Thus, it is one of the most frequently detected antibiotic chemicals in water bodies and sediments around the world, and it was also found to be potentially toxic for the ecosystem and humans.36–39  The occurrence of OTC in the environment is due to the discharge of pharmaceutical, agriculture, and livestock waste. In the past few years, several studies have demonstrated the presence of OTC in surface water at concentrations ranging from a few µg L−1 to mg L−1.40,41  The detoxification of water by removal of OTC has become a pressing environmental problem because of the high toxicity of OTC and the inability of conventional treatment processes to remove it due to its stable naphthol ring structure and the high toxicity to the microorganisms used in biological treatment. Herein, the most important studies related to the application of TiO2 based photocatalysts for the treatment of OTC will be presented.

OTC is degradable by both hydrolysis and photolysis. Xuan et al.42  reported the effects of pH and temperature on the decomposition of OTC. Solution pH and temperature were found to have important effects on OTC hydrolysis and photolysis. The hydrolysis was much faster in acidic and alkaline solutions than in neutral. With the enhancement of temperature from 4 ± 0.8 to 60 ± 1 °C, the half-life of OTC reduced from 1.2 × 102 d to 0.15 d. In addition, the photolytic degradation of OTC was found to be fast with a degradation rate constant of 3.61 ± 0.06 per day at pH 5.85. The degradation rate is highly pH dependent and drastically increased in alkaline solutions. OTC has four species at different pH ranges according to the three ionization equilibrium constants with pKa values of 3.57, 7.49, and 9.88. The four protonation states can be represented by H3OTC+, H2OTC, HOTC, and OTC2−. The adsorption spectra of OTC at different pH values exhibited a redshift to visible light, paralleled to the change of the protonation states of OTC. Zhao et al.43  revealed a self-photosensitization pathway with evidence of singlet oxygen generation for HOTC and OTC2− during photolysis under solar light irradiation. Meanwhile, the three-dimensional fluorescence spectra of OTC at different pH values also demonstrate that only HOTC and OTC2− exhibit significant emission spectra, but not H3OTC+ and H2OTC. Therefore, the change of the internal electrostatic force of OTC molecule as a function of solution pH values and species was proposed as a significant factor influencing the energy states and observed reaction pathways of OTC species under light irradiation. Nevertheless, only 13.5% reduction of TOC occurred despite a rapid photolytic degradation of 90% of OTC at 20 mg L−1 after 240 min of irradiation by a 500 W medium mercury lamp (light intensity 5.25 × 10−4 W cm−2 at 365 nm).44  Moreover, the photolytic degradation byproducts (maintained the naphthol ring) have been proved to be even more toxic than the parent when tested with luminescent bacterium with the inhibition rate increasing from 21% to 47% on P. phosphoreum.44 

The photocatalytic removal of OTC with TiO2 was also investigated under UV, visible, and solar light irradiation. Pereira et al.45  reported the effects of TiO2 load, solution pH, and inorganic ions on the degradation, mineralization, and detoxification of OTC using simulated solar light and natural solar radiation. The solution pH was critical in decomposing OTC. The highest photocatalytic activity of TiO2 was observed at pH 4.4 and 0.5 g L−1 TiO2 for the removal of both OTC (100% after 40 min of irradiation; 7.5 kJ L−1 of UV dose) and TOC (>90% after 180 min of irradiation; 38.3 kJ L−1 of UV dose), while 100% OTC and ∼80% TOC were removed by 1.8 and 11.3 kJ L−1, respectively, by natural solar energy. In addition, the BOD5 (biological oxygen demand)/COD (chemical oxygen demand) ratio rose from 0 to nearly 0.5, showing a remarkable improvement in biodegradability, while the inhibition percentage of bioluminescence of Vibrio fischeri decreased significantly from 35% to 7%. The remaining degradation by-products were mainly low-molecular-weight carboxylate anions without any antibacterial activity. The presence of phosphates hindered the removal of OTC, whereas the presence of chlorides, sulfates, nitrates, ammonium, and bicarbonates did not alter the degradation kinetics.46  Han et al.47  synthesized Ag-decorated, monodisperse TiO2 aggregates for the degradation of OTC, and found enhanced photocatalytic activity with a low leaching of Ag in the solution. Zhao et al.48  also reported the photocatalytic degradation, mineralization, and detoxification of OTC by 5A and 13X zeolite particles loaded with nano-TiO2 under UV irradiation. The best loading rate of composite photocatalysts was 15 wt% TiO2/5A and 10 wt% TiO2/13X for OTC degradation. The composite photocatalysts showed a better anti-interference capability of radical scavengers, and humic acid, at a low concentration than unsupported TiO2 during the removal of OTC in aqueous solutions.49  Meanwhile, the 10 wt% TiO2/13X kept the inhibition percentage of bioluminescence of Vibrio qinghaiensis sp.-Q67 below 35% during the process due to its excellent adsorption capability of OTC and its degradation byproducts.48  Moreover, the degradation of OTC was investigated with nitrogen and fluorine doped TiO2 film at different solution pH under visible and solar light in the presence of a series of scavengers. Five pathways, including direct photolytic degradation, UV/vis light-induced photocatalytic oxidation and reduction, and visible light-induced self-photosensitized oxidation and reduction, were proposed and verified as the mechanism for OTC photochemical degradation with TiO2.43 

The application of TiO2 photocatalysis for the efficient removal of contaminants of emerging concern (CEC) is dependent on many different parameters associated with both water quality and the physical chemical parameters of the contaminants. The main characteristics of the influent (i.e., pH, alkalinity, dissolved oxygen (DO), and the presence of NOM), the physical chemical characteristics of the contaminant and the treatment goal will determine the final design of photocatalytic water treatment systems.50–53  In this section we review the main publications that investigated the effects of water quality on the efficiency of TiO2 photocatalysis to remove an array of contaminants.

Pelaez et al.50  reported the effect of water quality parameters on the degradation of the cyanotoxin MC-LR when treated with visible light-active nitrogen and fluorine co-doped TiO2. Four water parameters (solution pH, alkalinity, NOM, and DO) were chosen to be studied. The solution pH significantly affected the decomposition of MC-LR since it is directly associated with the surface charge of TiO2 and the contaminant (in this case MC-LR). The optimal pH range to achieve the highest efficiency for MC-LR decomposition was 3–4 since TiO2 is positively charged30  and MC-LR is negatively charged at those pH values,54  which allows the toxin to better interact with the surface of the catalyst and better utilize the surface formed radicals. The effect of pH was also observed with other CEC such as carbamazepine (CBZ) when treated with different types of photocatalysts51,52  such as nitrogen doped TiO2, conventional TiO2 or zinc oxide (ZnO). For the degradation of CBZ, the highest degradation was achieved at basic conditions around pH values of 8–9 using nitrogen doped TiO2.52  At basic conditions, available hydroxyl ions on the surface of the catalyst increased to form hydroxyl radicals (HO˙) due to changes in the catalyst’s surface ionization states.55,56  Consequently, increased formation of HO˙ at these conditions enhanced the photocatalysis for the decomposition of CBZ. However, when conventional TiO2 and ZnO were employed to decompose CBZ, the highest degradation was observed at pH 6.4 when compared to pH 3 and pH 11.51  Therefore, the effect of pH on the degradation of CEC was found to be dependent on both the targeted CEC and type of photocatalysts used.

In addition to the effect of pH on the photocatalytic degradation of CEC, the effect of alkalinity was studied. For the degradation of MC-LR by visible light-induced photocatalysis, three alkalinity values at 50, 100, and 150 mg L−1 of sodium carbonate were tested at pH 7.1 using a phosphate buffer.50  Increasing alkalinity significantly decreased the degradation of MC-LR because carbonates and bicarbonates scavenged ROS. Since ROS attack and decompose CEC in photocatalytic degradation processes, MC-LR degradation decreased as alkalinity increased. Avisar et al.52  also reported the effect of alkalinity on the degradation of CBZ under simulated solar light-induced photocatalysis. In their study, five alkalinity values at 25, 50, 100, 200, and 300 mg L−1 of calcium carbonate (CaCO3) were tested at pH 9.0–9.5. The degradation efficiency of CBZ was also found to be inversely proportional to the concentration of alkalinity due to the scavenging effect of carbonates and bicarbonates. Above 200 mg L−1 of CaCO3, the effect of alkalinity on the degradation of CBZ decreased compared to lower alkalinity. It was reported that the reaction of carbonate and bicarbonate anions with HO˙ led to the formation of carbonate radicals, which have a lower redox potential (1.59 V)57  compared to hydroxyl radicals (2.8 V).58  The reduced reactivity of these radicals resulted in a decrease in the efficiency of the photocatalytic process.

Additionally, the effect of DO on the removal of CEC was investigated because oxygen has the ability to trap electrons and form additional ROS such as the superoxide anion (O2˙), therefore enhancing photocatalytic processes and CEC removal.50,59  The enhancement of photocatalytic degradation of MC-LR in the presence of DO was also reported.50  Nitrogen or oxygen was supplied into each solution for 30 min in order to study the effect of DO on the degradation of MC-LR. The MC-LR degradation rate in the reactor purged with oxygen was higher than that of the nitrogen purged reactor. The degradation rate was also proportional to the concentration of DO. Since electron–hole pairs generated during photocatalysis recombined fast in low concentrations of oxygen, the formation of ROS reduced, and the photocatalytic degradation of CEC decreased.60 

Finally, the effect of NOM on CEC degradation was investigated since natural water contains various forms of NOM including humic and fulvic acids, which can either inhibit or enhance photocatalytic degradation of CEC.50,53,61,62  For the degradation of MC-LR by visible light-induced photocatalysis, the degradation rates were inversely proportional to the concentration of NOM at pH values equal to 3.0, 5.7, and 7.1. Other studies also reported the inhibition of NOM for the degradation of CEC due to their scavenging effects of ROS.61,62  Although enhanced photolysis of CEC under simulated solar irradiation has been reported due to photosensitization of NOM,63,64  the degradation of CEC was inhibited in the presence of photocatalysts such as TiO2 and ZnO. Additional information on the effect of NOM is discussed in Section 1.3.2.

The presence of co-existing organic chemicals is problematic in decomposing target chemicals of concern in water. Since hydroxyl radicals produced from TiO2 non-selectively attack organic chemicals, the decomposition rates of many organic chemicals seem very similar.55  The non-selective nature looks attractive because TiO2 photocatalysts do not need special modification when a target chemical to decompose changes. The photocatalytic process does not differentiate one chemical from another in decomposition. However, this also represents a huge problem when only target chemicals of concern should be decomposed in a mixture with non-target chemicals.

In particular, typical source water is contaminated with highly toxic organic substances at low concentrations (e.g., anthropogenic chemicals such as pharmaceuticals) and less toxic naturally occurring organic substances at high concentrations (e.g., NOM). A photocatalytic process is required to focus on the decomposition of the toxic target chemicals rather than splitting its reactivity to both toxic targets and non-toxic NOM. However, target chemicals have to compete with abundant NOM. Competing chemicals take catalytic sites and consume HO˙. Moreover, they foul the surface of TiO2 materials. As demonstrated in Figure 1.3(a), the presence of humic acid (as representative NOM) significantly retarded the decomposition of a target chemical, ibuprofen (IBP, a well-known analgesic pharmaceutical found in water resources) on conventional TiO2.65  IBP decomposition was significantly decreased to around 30% in the presence of NOM from 80% in the absence of NOM.

Figure 1.3

Ibuprofen (IBP) decomposition in water by (a) conventional TiO2 (control) and (b) porous TiO2 in the absence and presence of humic acid (NOM). The result clearly shows the impact of the presence of NOM and the significance of the surface modification of TiO2. (Reproduced with permission from ref. 65. Copyright 2013 Elsevier.)

Figure 1.3

Ibuprofen (IBP) decomposition in water by (a) conventional TiO2 (control) and (b) porous TiO2 in the absence and presence of humic acid (NOM). The result clearly shows the impact of the presence of NOM and the significance of the surface modification of TiO2. (Reproduced with permission from ref. 65. Copyright 2013 Elsevier.)

Close modal

Consequently, nonselective sorption followed by oxidation of co-existing chemicals at TiO2 surface should be prevented. Operational parameters such as reaction pH and temperature can change decomposition rates among organic chemicals while TiO2 materials with controlled properties (e.g., crystal phase) can improve selectivity.66  An extrinsic material exhibiting specific affinity for a target chemical can also be introduced onto TiO2. Coating TiO2 with molecules with hydrophobic moieties can enhance sorption and oxidation of hydrophobic chemicals.67  Specially designed organic or inorganic domains as molecular recognition sites are integrated onto TiO2 surface to preferentially attract targets which then diffuse to the photocatalytic sites for chemical oxidation.68  Cavities of target chemicals can be imprinted onto TiO2 surfaces.69  Such chemical modifications exhibit ultimate selectivity only to target chemicals, meaning that a chemical marker should be replaced upon change in a target chemical.

Considering the heterogeneous reaction, size-based selection onto well-defined porous materials has also been investigated. TiO2 particles are embedded into clay structure and thus molecules smaller than the distance between sheet silicate layers can reach the embedded TiO2 particles.70  Similarly, they can also be coated with a SiO2 shell with well-developed porous structure.71  However, the approaches discussed above to improve selectivity commonly require introduction of a secondary material, which generally decreases overall reactivity due to covered TiO2 surface, reduced UV penetration, and limited mass transport. A recent study delivered an important message that even a pure TiO2 material, when properly designed, can work for size exclusion (Figure 1.3b).65  TiO2 with well-controlled pores slightly smaller than NOM and larger than IBP showed improved reactivity towards IBP in the presence of NOM as compared to nonporous control TiO2. However, studies on preferential photocatalytic degradation in a real mixed stream have still been rarely reported.

Photocatalysis with TiO2 is considered to be an alternative to conventional methods for environmental detoxification. Its efficiency is based on the formation of ROS that are able to decompose a great variety of environmental pollutants of different origin (both manmade and naturally found compounds),72–75  functionality (such as pesticides, pharmaceuticals, chlorophenols, and azo dyes),76–79  and chemical structure (organic compounds with heteroatoms such as N, P, and S).80  Although photocatalysis in the presence of TiO2 under UV-A proceeds to mineralization of the organic substrates, the degradation pathways are quite complicated through the formation and decay of numerous intermediates.81  The elucidation of the reaction mechanism is a difficult but necessary task in TiO2 photocatalysis not only for understanding the process in detail but also for detoxification purposes due to toxicity and persistency of many intermediates.

Consequently, it is important to monitor together with the destruction of the parent compounds the formation and decay of the intermediates formed in order to control all transformation steps, to identify hazardous intermediates, and to clarify the reaction mechanism.82–84  On some occasions, toxicity studies are contacted on the treated mixture and/or the identified intermediates to prove water detoxification following treatment.81 

Structural identification of transformation products (TPs) requires advanced analytical methods that combine high separation efficiency with increased structural identification capabilities. Due to their high selectivity and sensitivity, gas and liquid chromatography-mass spectrometry techniques are frequently used methods of choice for the structural elucidation of TPs.85 

Determination of TPs based on mass spectrometry requires three working steps. Following the exposure of the contaminant of interest to the relevant treatment, the sample is analyzed using a chromatographic and mass spectrometric method optimized for the parent compound. Then the mass spectrometer scans a range of masses that account for potential additions on the contaminant or its fragments. The second step consists of careful examination of the TP mass spectrum and comparison with the mass spectrum of the parent compound (contaminant). The final third step consists of additional analysis by tandem mass spectrometry which insures proper structural identification of TPs. Following that, TPs can be organized in degradation routes and pathways.

This section gives an overview of the different intermediates and their mechanisms of formation upon TiO2 photocatalysis with UV light. For that reason selected compounds (cyanotoxins i.e. MC-LR, CYN and water taste and odor compounds, i.e. geosmin (GSM) and 2-methylisoboeneol (MIB)) will be discussed to cover different mechanisms of degradation based on the functional groups treated with the various ROS that TiO2 photocatalysis forms. This target group of compounds are emerging contaminants of biological origin, the photocatalytic degradation of which was recently studied in detail.6,86–88 

Light activation of titania results in the formation of an array of ROS including hydroxyl radicals HO˙ (main species), superoxide radical anion (O2˙), and perhydroxyl radical (HO2˙) and the conduction band electron (eCB).89,90  More importantly, the positive holes formed in the valence band following electron photoexcitation can oxidize any compound (organic and/or inorganic) found adsorbed on the surface of the catalyst, also contributing to the routes of oxidation.

Since it is not possible to go over all the previously mentioned studies in this section, select groups of contaminants that incorporate in their structure basic functional groups (such as aromatic bonds, simple carbon bonds) and heteroatoms (N) will be discussed as well as their TPs and mechanisms of formation with TiO2 photocatalysis. The selected group of contaminants chosen is characterized by great structural variability and concerns a family of naturally occurring toxins, the cyanotoxins. Cyanotoxins are formed from the harmful strains of cyanobacterial algal blooms and greatly affect the ecosystem and human health.81,91  They can also be grouped based on structure: cyclic peptides (microcystins, nodularins–hepatotoxins), alkaloids (saxitoxin–neurotoxins), and lipopolysaccharides (LPSs). Cylindrospermopsin though structurally grouped as an alkaloid is not a neurotoxin but rather affects the liver and kidneys. Herein, the intermediates of MC-LR, CYN, and two taste and odor compounds with TiO2 photocatalysis (HO˙ attack) will be discussed.

In general, the main mechanistic steps that organic compounds undergo following HO˙ attack are hydroxylation via substitution or addition with possible simultaneous isomerization, oxidation, and oxidative bond cleavage. Unsaturated bonds such as aromatic rings and carbon double bonds can undergo hydroxyl substitution of a hydrogen to form hydroxylated intermediates such as m/z 1011.5 (Table 1.1). The necessary mechanistic steps include the addition of a HO˙ to one of the double bonds and formation of a carbon-centered radical, which rapidly reacts with oxygen, to form a peroxy radical. The release of a perhydroxyl radical results eventually in the substitution of the hydrogen with a hydroxyl group.92  If another HO˙ reacts with the carbon centered radical, then hydroxyl addition occurs which changes the carbon hybridization from sp2 to sp3, and therefore the overall three-dimensional structure of the TP compared to the parent compound. As will be seen from the studied cases extensively analyzed below, further radical attack will cause the complete oxidation of carbon and the fragmentation of the treated compound.

Table 1.1

Structures of reaction intermediates of microcystin-LR with TiO2 based photocatalysts

No.Structurem/zCompoundTechnologyaReference
MC-LR  995.5 C49H74N10O12 b b 
1A  1029.5 C49H76N10O14 TiO2_p/UVA 15  
1B  TiO2_f/UVA 95 c 
1C  NTiO2_p/λ > 420 nm 34  
1D  NTiO2_p/UVA 15  
1E  NFTiO2_f/λ > 420 nm 103d 
 835.5 C37H58N10O12 TiO2_p/UVA 15  
TiO2_f/UVA 95  
NTiO2_p/UVA 34  
NFTiO2_f/λ > 420 nm 103  
 795.4 C34H54N10O12 TiO2_p/UVA 15  
TiO2_f/UVA 95  
NTiO2_p/UVA 34  
NFTiO2_f/λ > 420 nm 103  
 811.5 C34H54N10O13 TiO2_p/UVA 15  
 965 C46H74N10O11Na TiO2_p/UVA 15  
943 C46H74N10O11 
 999 C46H75N10O13Na TiO2_p/UVA 15  
977 C46H76N10O13 
7A  1011.5 C49H74N10O13 TiO2_p/UVA 15  
7B  
7C  
7D  
7E  
 1027.5 C49H74N10O14 TiO2_f/UVA 95  
NFTiO2_f/λ > 420 nm 103  
 1009.6 C49H72N10O13 TiO2_f/UVA 95  
10A  965.6 C48H72N10O11 TiO2_f/UVA 95  
11A  1063.5 C49H78N10O14 TiO2_f/UVA 95  
11B  
11C  
12A  783.4 C33H54N10O12 TiO2_f/UVA 95  
NTiO2_p/UVA 34  
12B  783.4 C34H58N10O11 TiO2_p/UVA 15  
TiO2_f/UVA 95  
13  1015.5 C48H74N10O14 TiO2_f/UVA 95  
14  743.4 C31H54N10O11 TiO2_p/UVA 15  
15  759.4 C31H54N10O12 TiO2_p/UVA 15  
16  617 C31H48N6O7 TiO2_p/UVA 15  
17  651 C31H50N6O9 TiO2_p/UVA 15  
18A  1045.5 C49H77N10O15 NFTiO2_p/λ > 420 nm 103  
18B  
18C  
18D  
19  1025.5 C49H72N10O14 NFTiO2_f/λ > 420 nm 103  
NOD  825.5 C41H60N8O10 — — 
20A  859 C41H62N8O12 TiO2_p/UVA 94  
20B  
21  665 C29H44N8O10 TiO2_p/UVA 94  
22  625 C26H40N8O10 TiO2_p/UVA 94  
23  695 C29H42N8O12 TiO2_p/UVA 94  
24  286 C11H19N5O4 TiO2_p/UVA 94  
25  175 C6H14N4O2 TiO2_p/UVA 94  
No.Structurem/zCompoundTechnologyaReference
MC-LR  995.5 C49H74N10O12 b b 
1A  1029.5 C49H76N10O14 TiO2_p/UVA 15  
1B  TiO2_f/UVA 95 c 
1C  NTiO2_p/λ > 420 nm 34  
1D  NTiO2_p/UVA 15  
1E  NFTiO2_f/λ > 420 nm 103d 
 835.5 C37H58N10O12 TiO2_p/UVA 15  
TiO2_f/UVA 95  
NTiO2_p/UVA 34  
NFTiO2_f/λ > 420 nm 103  
 795.4 C34H54N10O12 TiO2_p/UVA 15  
TiO2_f/UVA 95  
NTiO2_p/UVA 34  
NFTiO2_f/λ > 420 nm 103  
 811.5 C34H54N10O13 TiO2_p/UVA 15  
 965 C46H74N10O11Na TiO2_p/UVA 15  
943 C46H74N10O11 
 999 C46H75N10O13Na TiO2_p/UVA 15  
977 C46H76N10O13 
7A  1011.5 C49H74N10O13 TiO2_p/UVA 15  
7B  
7C  
7D  
7E  
 1027.5 C49H74N10O14 TiO2_f/UVA 95  
NFTiO2_f/λ > 420 nm 103  
 1009.6 C49H72N10O13 TiO2_f/UVA 95  
10A  965.6 C48H72N10O11 TiO2_f/UVA 95  
11A  1063.5 C49H78N10O14 TiO2_f/UVA 95  
11B  
11C  
12A  783.4 C33H54N10O12 TiO2_f/UVA 95  
NTiO2_p/UVA 34  
12B  783.4 C34H58N10O11 TiO2_p/UVA 15  
TiO2_f/UVA 95  
13  1015.5 C48H74N10O14 TiO2_f/UVA 95  
14  743.4 C31H54N10O11 TiO2_p/UVA 15  
15  759.4 C31H54N10O12 TiO2_p/UVA 15  
16  617 C31H48N6O7 TiO2_p/UVA 15  
17  651 C31H50N6O9 TiO2_p/UVA 15  
18A  1045.5 C49H77N10O15 NFTiO2_p/λ > 420 nm 103  
18B  
18C  
18D  
19  1025.5 C49H72N10O14 NFTiO2_f/λ > 420 nm 103  
NOD  825.5 C41H60N8O10 — — 
20A  859 C41H62N8O12 TiO2_p/UVA 94  
20B  
21  665 C29H44N8O10 TiO2_p/UVA 94  
22  625 C26H40N8O10 TiO2_p/UVA 94  
23  695 C29H42N8O12 TiO2_p/UVA 94  
24  286 C11H19N5O4 TiO2_p/UVA 94  
25  175 C6H14N4O2 TiO2_p/UVA 94  
a

TiO2_p = TiO2 particles. TiO2_f = TiO2 films.

b

— Not available or not referred at the publication. UVA stands for UV(300 < λ < 400 nm).

c

Displayed structures A–D were observed in Ref. 95.

d

Structure 1E was only reported by Ref. 103.

The first study on TiO2 photocatalytic transformations of MC-LR was conducted by Liu and coworkers.15,93  MC-LR (Co = 1000 mg L−1) degradation was performed with TiO2 nanoparticles in slurry systems (1.0% w/v) in the presence of hydrogen peroxide (42.8 mM H2O2) at acidic pH (pH 4.0). Based on these experimental conditions, ten TP were detected (651 ≤ m/z ≤ 1029) leading to three oxidation routes. The side of MC-LR that was affected initially was the chain of the Adda amino acid, which contains conjugated carbon double bonds that underwent simultaneous hydroxylation and isomerization (m/z = 1029.5 products, Table 1.1), and bond cleavage giving smaller molecular weight aldehydes (m/z = 795.5) and ketones (the m/z = 835.5). Further oxidation of m/z = 795.5, produced the carboxylated intermediate with m/z = 811.5. The next proposed pathway is similar to the first with the difference that the cyclic structure of MC-LR was initially cleaved between Mdha and Ala to produce a linear TP with [M + Na = 965]. After that, the C6–C7 double bond of Adda was hydroxylated (m/z = 977) and oxidized to the corresponding ketone (m/z = 783), aldehyde (m/z = 743), and carboxylic acid (m/z = 759). The last pathway involved a highly oxidized linear MC-LR where the Mdha–Ala–MeAsp moieties were removed (m/z = 617), followed by the dihydroxylation of the Adda chain (m/z = 651). In addition to the intermediate studies, the authors performed toxicity tests based on the inhibition of the protein phosphatases 1 (PP1) enzymes of the treated solution and on fractions of the treated solution collected following HPLC separation. Based on the inhibition data, though the fractions contained high concentrations of linearized TPs no significant toxicity against PP1 enzymes was observed.15,93 

The same research group also studied the pentapeptide nodularin (NOD) by utilizing TiO2 nanoparticles in a slurry (0.1% w/v).94  Following treatment eleven TP were identified with m/z in the range 175 ≤ m/z ≤ 859. The degradation pathways that NOD followed were equivalent to those of MC-LR15,93  since the same types of bonds underwent hydroxyl radical attack. Initially, dihydroxylated TPs of different stereochemistry were formed at the carbon diene bonds (Table 1.1, m/z = 859), followed by bond cleavage at C4–C5 and/or C6–C7 bonds of Adda to form the corresponding ketone (Table 1.1, m/z = 665) and aldehyde products (Table 1.1, m/z = 625). These by-products were then oxidized to the corresponding peroxidated products (Table 1.1, m/z = 695), followed by hydrolysis of the peptide bonds resulting in small amino acid fragments (Table 1.1, m/z = 286; m/z = 175). Toxicity studies with the PP1 enzyme showed again loss of toxicity due to the photocatalytic oxidation (PCO), as well as the lack of formation of toxic intermediates.

Another study investigated the degradation of MC-LR with TiO2/UVA, Antoniou et al. (2008),95  but utilized two different photocatalytic films with different thickness (mass of catalyst was 1.4 mg and 50.4 mg per thin and thick film, respectively) instead of nanoparticles. The degradation of MC-LR (Co = 20 mg L−1) occurred at neutral pH and 21 types of TPs were detected as m/z. Differences in the type and number of intermediates isolated between the two studies15,93  can be attributed to differences in the experimental conditions, frequency of sampling, addition of oxidants (which enhance degradation), and the pH of the treated solution (which affects the interaction between pollutant and catalyst). In their study, Antoniou and coworkers conducted their experiments in the absence of additional oxidants and at pH of milli-Q water which reduced interactions between MC-LR and the TiO2 and allowed the HO˙ to interact with more sites of the toxin.95 

Tandem mass spectrometry analysis led to the structural characterization of the formed TPs and revealed four sites of MC-LR where degradation was initiated: three of them are on the Adda amino acid (aromatic ring, methoxy group, and conjugated double bonds) and one is on the cyclic structure (Mdha amino acid). Overall, the most susceptible group for hydroxyl radical attack was the unsaturated bonds of the Adda amino acid (side chain of the toxin) possibly because of their location and vulnerability to oxidation by HO˙ (kOH ∼109−10 M−1 s−1).96  The first oxidation step was the hydroxylation of the aromatic ring through substitution of an aromatic hydrogen to form the m/z 1011.5 intermediates at o-, p- and at a lesser extend the m-substitution, followed by a second hydroxylation (m/z 1027.5, Table 1.1). Detection of m/z 1027.5 confirmed the first substitution, because hydroxyl groups increase the electron density of the aromatic ring and, thus, the kinetic rates of electrophilic reactions increase.97,98 

Hydroxyl radical attack not only causes hydrogen abstraction but also removal of small functional groups such as methyl and methoxy groups. Antoniou et. al. (2008)95  were the first to observe the demethoxylation of the Adda chain (m/z 965.6, DmADDA) through the formation of the formic acid ester (m/z 1009.6, Table 1.1). Another mechanist step observed was hydroxyl addition, especially at the diene carbon bonds. The m/z 1029.5 and m/z 1063.5 were formed following double and quadruple hydroxyl addition. It has been observed that when conjugated dienes undergo electrophilic reactions, mixtures of TPs are formed (additional to the stereoisomers) because of 1,2- (m/z = 1029.5, 1A and 1B) or 1,4-additions (m/z = 1029, 1C) (Table 1.1). A separate oxidation pathway initiated at the diene bonds resulted in the complete removal of the Adda chain.15,95  Once the enol-MC-LR (m/z 1011.5, 7D) was formed it was isomerized to the more stable tautomer of ketone-MC-LR (m/z = 1011.5, 7E). Following consequent oxidative bond cleavage steps, the ketone tautomer m/z = 1011.5 (7E) transformed into m/z 835.4, then into the aldehyde-derivative with m/z = 795.4, and eventually to a hydroxyl-derivative with m/z 783.4 (product 12A, Table 1.1).

The last site where Antoniou et. al. (2008)95  reported initiation of MC-LR’s oxidation was in the cyclic structure at the Mdha amino acid. The cyclic structure proved to be more resilient to ROS attack, possibly due to shielding from the functional groups of the other amino acids, as well as competition with the three other sites for ROS utilization. TPs formed following consecutive oxidation, including double hydroxylation of the double bond of the Mdha (m/z 1029.5, 1D), its oxidation to aldehyde (m/z 1011.5, 7D), and cleavage of the R2C–COR bond (m/z 1015.5), were reported. Finally, TPs where degradation occurred simultaneously at the Adda chain and cyclic structure, such as m/z 783.4 (product 12B, Table 1.1), were observed.

Following TPs’ identification, the authors performed toxicity studies measuring the activity of the PP1 enzymes in the treated solution,99  though it was already proven that some of the identified TPs did not possess toxic properties [(Z)-MC-LR and the demethoxylated-MC-LR (DmADDA)].95,100,101  The PP1 inhibition studies showed that as photocatalytic degradation progressed, the enzyme’s activity was recovered, and towards the end of the treatment it was completely restored. Treatment with the photocatalytic films impaired MC-LR’s structure so much that its toxic properties disappeared with limited carbon mineralization. This proves that water detoxification can be achieved without the complete mineralization of a compound, but instead through targeted oxidation.

Aligned with the current trend of PCO for utilizing sustainable light sources for catalyst activation such as sunlight, a few studies reported intermediates with light activated modified TiO2 photocatalysts. Sunlight contains only 5% of UV radiation; however, incorporation of heteroatoms (such as N and F) into the TiO2 structure allows visible light activation due to narrowing of the band gap energy.30,34,102  In this case, HO˙ radicals are not the primary oxidation species produced by the positive holes of the conduction band and the oxidation is speculated to arise from O2˙ and HO2˙ radicals. The latter react in a similar way to HO˙ and can also produce hydroxylated TPs. So far, two studies have identified the reaction intermediates of MC-LR under visible light activation of N–TiO2 nanoparticles34  and NF–TiO2 photocatalytic films103  respectively. When N–TiO2 nanoparticles where illuminated with visible light, only the m/z = 1029.5 intermediate was isolated in multiple peaks. When UVA radiation was used instead for N–TiO2 activation, besides m/z = 1029.5 the oxidative cleavage of the Adda chain through the formation of a ketone (m/z 835.5), an aldehyde (m/z 795.5), and a hydroxyl derivative (m/z 783.5, product 12A, Table 1.1) were observed as well. In Andersen’s study, seven different TPs were formed following 8 h of treatment with m/z = 795.4; 1011.5 (4 peaks); 1027.5; 1029.5; 1045.5; 835.4; and 1025.5 (Table 1.1). The latter one was a unique intermediate that was never before reported with PCO, and was formed after the hydroxyl substitution of the Adda amino acid double bond at C4 (enol ↔ ketone), followed the oxidation of the methoxy group to an aldehyde. The reported pathways of the remaining of TPs were the same as those reported by Antoniou et al. (2008).95 

Microcystins are cyanobacterial metabolites with relatively large and complex structures and, therefore, unveiling the corresponding TPs structures following free radical attack can be tedious work. TPs have also been identified for simpler secondary metabolites of cyanobacterial blooms such as taste and odor causing compounds GSM and MIB.104  Removing these metabolites from water is necessary for its general usage and consumption.

Two studies have worked on TPs formed during the photocatalytic degradation of GSM with Degussa P-25 nanoparticles.87,105  During the first study only a few intermediates were identified which involved oxidation of GSM via HO˙ and by the reactive holes on the surface of the catalyst. Results in this study pointed that GSM undergoes rapid ring opening giving aliphatic saturated and unsaturated compounds.105  In a more elaborate study by Fotiou et al.,87  many degradation products were detected. The majority of the TPs formed were oxygenated, suggesting that the primer degradation mechanism was due to HO˙ oxidation through electrophilic substitutions. Bond cleavages also occurred at multiple sites, producing cyclic ketones that with further bond cleavages gave open chain saturated and unsaturated products (Figure 1.4).87  Mechanisms concerning the formation of the main intermediate products are α-hydrogen abstraction from GSM tertiary carbon, hydroxylation from HO˙ attack and ketone formation with β-scission abstraction.

Figure 1.4

Intermediates products of geosmin formed using TiO2 under UV-A light. (Reproduced with permission from ref. 87. Copyright 2015 Elsevier.)

Figure 1.4

Intermediates products of geosmin formed using TiO2 under UV-A light. (Reproduced with permission from ref. 87. Copyright 2015 Elsevier.)

Close modal

In the same study, degradation of MIB using of TiO2 under UV-A light revealed the formation of several products, prior to their decay and final decomposition to CO2.87  Alcohol-, ketone- and diketone-derivatives of MIB, oxygen containing cyclic compounds and open chain aliphatic compounds, were the identified intermediates (Figure 1.5). d-Camphor was the primary intermediate formed during the photocatalytic degradation of MIB which was formed with a β-scission reaction mechanism on the methyl group of MIB, giving the ketone. Starting from d-camphor with hydrogen elimination followed by HO˙ addition and with further oxidation of a secondary alcohol, leading to the formation of a ketone, three diketone products could be formed. All other intermediates identified were mainly oxygen containing saturated and unsaturated cyclic compounds, through HO˙ oxidations, driven by electrophilic substitutions. Ring opening reactions on the MIB molecule result in the formation of five-membered ring compounds. Similar to previous studies by Hiskia et al.,106  during the later stages of photo-oxidation linear aliphatic compounds were identified.

Figure 1.5

Intermediate products of 2-methylisoboeneol (MIB) formed using TiO2 under UV-A light. (Reproduced with permission from ref. 87. Copyright 2015 Elsevier.)

Figure 1.5

Intermediate products of 2-methylisoboeneol (MIB) formed using TiO2 under UV-A light. (Reproduced with permission from ref. 87. Copyright 2015 Elsevier.)

Close modal

The degradation routes of compounds with heteroatoms in their structure (such as N, P, S) may differ from the ones with only carbon atoms because of differences in the kinetic rates but also the reaction mechanisms that need to take place for stable intermediates to form. Herein, the TPs of CYN and taste and odor compounds will be reported as characteristic examples of this group of compounds.

CYN is a uracil derivative with a tricyclic guanidine and a sulfate group. Only a few studies have dealt with the degradation mechanism of CYN using TiO2 photocatalysis,6,86  where it has been stated that the main reaction pathways were through hydroxyl radical attack, with hydrogen abstractions and addition to double bonds. Table 1.2 gives the structures of reaction intermediates identified upon TiO2 photocatalysis under UV light.

Table 1.2

Structures of reaction intermediates of cylindrospermopsin with TiO2 photocatalysis

No.Structurem/zCompoundReference
CYN  416 C15H21N5O7 
 414 C15H19N5O76,87  
2A  432 C15H21N5O86,87  
2B   6,87  
2C  C15H21N5O86  
 450 C15H23N5O987  
4A  448 C15H21N5O96,87  
4B  448 C15H21N5O96,87  
5A  464 C15H21N5O106  
5B  464 C15H23N5O106  
 434 C15H23N5O887  
7A  392 C13H21N5O76  
7B  393 C13H20N4O86  
 375 C14H22N4O687  
 375 C15H19N4O76  
10  373 C14H20N4O687  
11  350 C12H19N3O787  
12  350 C15H19N5O5 6  
13  347 C12H18N4O687  
14  338 C15H23N5O4+ 87  
15  334 C15H19N5O4 6  
16  322 C11H19N3O66  
17  320 C11H17N3O687  
18  316 C13H23N4O5+ 87  
19A  290 C10H15N3O56  
19B  292 C10H17N3O56  
19C  292 C9H13N3O66  
20  287 C13H24N3O4+ 87  
21  280 C8H13N3O687  
22  227 C11H20N3O2+ 87  
23  195 C10H16N3O+ 87  
No.Structurem/zCompoundReference
CYN  416 C15H21N5O7 
 414 C15H19N5O76,87  
2A  432 C15H21N5O86,87  
2B   6,87  
2C  C15H21N5O86  
 450 C15H23N5O987  
4A  448 C15H21N5O96,87  
4B  448 C15H21N5O96,87  
5A  464 C15H21N5O106  
5B  464 C15H23N5O106  
 434 C15H23N5O887  
7A  392 C13H21N5O76  
7B  393 C13H20N4O86  
 375 C14H22N4O687  
 375 C15H19N4O76  
10  373 C14H20N4O687  
11  350 C12H19N3O787  
12  350 C15H19N5O5 6  
13  347 C12H18N4O687  
14  338 C15H23N5O4+ 87  
15  334 C15H19N5O4 6  
16  322 C11H19N3O66  
17  320 C11H17N3O687  
18  316 C13H23N4O5+ 87  
19A  290 C10H15N3O56  
19B  292 C10H17N3O56  
19C  292 C9H13N3O66  
20  287 C13H24N3O4+ 87  
21  280 C8H13N3O687  
22  227 C11H20N3O2+ 87  
23  195 C10H16N3O+ 87  

As previously mentioned, HO˙ radicals primarily react through hydrogen abstraction and addition on the unsaturated carbon bond on the uracil group, resulting in the formation of product m/z = 432.6,86  Further hydroxylation results in the formation of product m/z = 450.1  Oxidation on the uracil group of product m/z = 432 can lead first to the formation of product m/z = 375, upon ring opening at the urea group moiety, and then by a further oxidation to product m/z = 350 (cylindrospermopsic acid).86  From m/z = 375, product m/z = 373 can also be formed. From there onwards, products with m/z = 347 and the acid-like product with m/z = 320 are formed by further oxidations. HO˙ radical attack on m/z = 347 can result in compound m/z = 280.86  A product with m/z = 392 was detected, which can likely derive from hydroxyl addition on product m/z = 432.6 

Elimination of the sulfate group can also occur under free radical attack. Product m/z = 334 resulted from hydroxylation on the tricyclic guanidine moiety followed by elimination of the sulfuric acid group. With further hydroxylation, product m/z = 350 was produced,6  from which further hydroxylation resulted to give product m/z = 366 and m/z 382.6 

Starting from product m/z = 450, TPs with m/z = 287 and m/z = 448 are formed. Oxidation of m/z = 450 on either of the secondary alcohols, through hydrogen abstraction, can lead to the formation of two carbonyl isomers with m/z = 448,86  while further hydroxylation leads to product m/z = 464.6  In addition, m/z = 287 is formed, through substitution of the sulfate group followed by oxidation on the uracil group. By a similar pathway, the carbonyl product with m/z = 414 corresponds to the oxidation of CYN by HO˙ radical attack on the secondary alcohol (bridging methane group), followed by substitution of the sulfate group with HO˙ addition to form product m/z = 338, from which a compound with m/z = 316 can be produced with oxidation on the uracil group.86 

Overall, it was found that the mechanism of CYN degradation proceeds through HO˙ attack on different sites of CYN: (a) hydroxylation on the uracil group, with uracil ring opening and further oxidation, (b) attack on the tricyclic alkaloid group, (c) attack on the tricyclic alkaloid group with ring opening, and (d) attack on the sulfate group, leading to various oxidation products (Table 1.2). After prolonged irradiation under UV-A light, complete mineralization of CYN to CO2, SO42−, NO3, and NH4+ takes place.86 

UV-based TiO2 photocatalysis is effective for the degradation of a wide variety of organic compounds with different functional groups. Complex organic compounds can be broken down into individual functional groups. The different functional groups have different reactivity and susceptibility to photocatalytic degradation. Given the large number of studies reported on the UV/TiO2 degradation of an extensive number of organic pollutants and toxins, it is possible to predict the relative reactivity of different reaction sites and reaction pathway partitioning in complex substrates. Fundamental mechanistic understanding is critical for the development and application of UV/TiO2 photocatalysis for real water treatment of problematic pollutant and mixed waste. Therefore, the influent water quality parameters must be optimized to achieve the highest efficiency of TiO2 photocatalysis. In addition, physical and chemical modification of TiO2 materials can improve selectivity towards target chemicals, thus achieving higher removal efficiencies than conventional with UV/TiO2. Future research will also focus on unveiling experimental conditions under which the photocatalytic processes applied in water treatment results in detoxified water by focusing on the contaminant’s sites that initiate its toxic behavior.

M. G. Antoniou is thankful to the Cyprus University of Technology for a start-up package (EX-90). H. Choi is grateful to the Texas Higher Education Coordinating Board for the Norman Hackerman Advanced Research Program fund (THECB13311). D. D. Dionysiou also acknowledges support from the University of Cincinnati through a UNESCO co-Chair Professor position on “Water Access and Sustainability”. C. Han was supported by the Postgraduate Research Program at the National Risk Management Research Laboratory administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Environmental Protection Agency. T. Fotiou, T. M. Triantis and A. Hiskia acknowledge financial support from the European Social Fund and Greek national funds through ARISTEIA operational research program “Cyanotoxins in Fresh Waters, Advances in Analysis, Occurrence and Treatment CYANOWATER” (Grant N.: 2455).

1

The U.S. Environmental Protection Agency, through its Office of Research and Development, funded and managed, or partially funded and collaborated in, the research described herein. It has been subjected to the Agency's administrative review and has been approved for external publication. Any opinions expressed in this chapter are those of the author(s) and do not necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred. Any mention of trade names or commercial products does not constitute endorsement or recommendation for use.

1.
Fujishima
A.
,
Honda
K.
,
Nature
,
1972
, vol.
238
pg.
37
2.
Fox
M. A.
,
Dulay
M. T.
,
Chem. Rev.
,
1993
, vol.
93
pg.
341
3.
Banerjee
S.
,
Pillai
S. C.
,
Falaras
P.
,
O’Shea
K. E.
,
Byrne
J. A.
,
Dionysiou
D. D.
,
J. Phys. Chem. Lett.
,
2014
, vol.
5
pg.
2543
4.
Pelaez
M.
,
Nolan
N. T.
,
Pillai
S. C.
,
Seery
M. K.
,
Falaras
P.
,
Kontos
A. G.
,
Dunlop
P. S. M.
,
Hamilton
J. W. J.
,
Byrne
J. A.
,
O’Shea
K.
,
Entezari
M. H.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2012
, vol.
125
pg.
331
5.
Zhao
C.
,
Pelaez
M.
,
Dionysiou
D. D.
,
Pillai
S. C.
,
Byrne
J. A.
,
O’Shea
K. E.
,
Catal. Today
,
2014
, vol.
224
pg.
70
6.
Zhang
G.
,
Wurtzler
E. M.
,
He
X.
,
Nadagouda
M. N.
,
O’Shea
K.
,
El-Sheikh
S. M.
,
Ismail
A. A.
,
Wendell
D.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2015
, vol.
163
pg.
591
7.
Zheng
S.
,
Jiang
W.
,
Cai
Y.
,
Dionysiou
D. D.
,
O’Shea
K. E.
,
Catal. Today
,
2014
, vol.
224
pg.
83
8.
Buxton
G. V.
,
Greenstock
C. L.
,
Helman
W. P.
,
Ross
A. B.
,
J. Phys. Chem. Ref. Data
,
1988
, vol.
17
pg.
513
9.
Andreozzi
R.
,
Caprio
V.
,
Insola
A.
,
Marotta
R.
,
Catal. Today
,
1999
, vol.
53
pg.
51
10.
Zhao
C.
,
Arroyo-Mora
L. E.
,
DeCaprio
A. P.
,
Sharma
V. K.
,
Dionysiou
D. D.
,
O’Shea
K. E.
,
Water Res.
,
2014
, vol.
67
pg.
144
11.
Turchi
C. S.
,
Ollis
D. F.
,
J. Catal.
,
1990
, vol.
122
pg.
178
12.
Lawless
D.
,
Serpone
N.
,
Meisel
D.
,
J. Phys. Chem.
,
1991
, vol.
95
pg.
5166
13.
O’Shea
K. E.
,
Cardona
C.
,
J. Org. Chem.
,
1994
, vol.
59
pg.
5005
14.
Song
W.
,
Xu
T.
,
Cooper
W. J.
,
Dionysiou
D. D.
,
de la Cruz
A. A.
,
O’Shea
K. E.
,
Environ. Sci. Technol.
,
2009
, vol.
43
pg.
1487
15.
Liu
I.
,
Lawton
L. A.
,
Robertson
P. K. J.
,
Environ. Sci. Technol.
,
2003
, vol.
37
pg.
3214
16.
d’Hennezel
O.
,
Pchat
P.
,
Ollis
D. F.
,
J. Photochem. Photobiol., A
,
1998
, vol.
118
pg.
197
17.
Bui
T. D.
,
Kimura
A.
,
Higashida
S.
,
Ikeda
S.
,
Matsumura
M.
,
Appl. Catal., B
,
2011
, vol.
107
pg.
119
18.
Li
X.
,
Jenks
W. S.
,
J. Am. Chem. Soc.
,
2000
, vol.
122
pg.
11864
19.
Lin
H. F.
,
Ravikrishna
R.
,
Valsaraj
K. T.
,
Sep. Purif. Technol.
,
2002
, vol.
28
pg.
87
20.
Hidaka
H.
,
Honjou
H.
,
Koike
T.
,
Mitsutsuka
Y.
,
Oyama
T.
,
Serpone
N.
,
J. Photochem. Photobiol., A
,
2008
, vol.
197
pg.
115
21.
Konstantinou
I. K.
,
Sakkas
V. A.
,
Albanis
T. A.
,
Appl. Catal., B
,
2001
, vol.
34
pg.
227
22.
Harada
K.
,
Hisanaga
T.
,
Tanaka
K.
,
Water Res.
,
1990
, vol.
24
pg.
1415
23.
Vidal
A.
,
Dinya
Z.
,
Mogyorodi
F.
,
Appl. Catal.,B
,
1999
, vol.
21
pg.
259
24.
Shephard
G. S.
,
Stockenstrom
S.
,
de Villiers
D.
,
Engelbrecht
W. J.
,
Wessels
G. F. S.
,
Water Res.
,
2002
, vol.
36
pg.
140
25.
Vilela
W. F. D.
,
Minillo
A.
,
Rocha
O.
,
Vieira
E. M.
,
Azevedo
E. B.
,
Sol. Energy
,
2012
, vol.
86
pg.
2746
26.
Su
Y. L.
,
Deng
Y. R.
,
Zhao
L.
,
Du
Y. X.
,
Chin. Sci. Bull.
,
2013
, vol.
58
pg.
1156
27.
Senogles
P. J.
,
Scott
J. A.
,
Shaw
G.
,
Stratton
H.
,
Water Res.
,
2001
, vol.
35
pg.
1245
28.
Zhang
G. S.
,
Nadagouda
M. N.
,
O’Shea
K.
,
El-Sheikh
S. M.
,
Ismail
A. A.
,
Likodimos
V.
,
Falaras
P.
,
Dionysiou
D. D.
,
Catal. Today
,
2014
, vol.
224
pg.
49
29.
Choi
J.
,
Park
H.
,
Hoffmann
M. R.
,
J. Phys. Chem. C
,
2010
, vol.
114
pg.
783
30.
Pelaez
M.
,
de la Cruz
A. A.
,
Stathatos
E.
,
Falaras
P.
,
Dionysiou
D. D.
,
Catal. Today
,
2009
, vol.
144
pg.
19
31.
Liu
G. L.
,
Han
C.
,
Pelaez
M.
,
Zhu
D. W.
,
Liao
S. J.
,
Likodimos
V.
,
Ioannidis
N.
,
Kontos
A. G.
,
Falaras
P.
,
Dunlop
P. S. M.
,
Byrne
J. A.
,
Dionysiou
D. D.
,
Nanotechnology
,
2012
, vol.
23
pg.
29
32.
Han
C.
,
Pelaez
M.
,
Likodimos
V.
,
Kontos
A. G.
,
Falaras
P.
,
O’Shea
K.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2011
, vol.
107
pg.
77
33.
Zhang
Y. C.
,
Yang
M.
,
Zhang
G.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2013
, vol.
142–143
pg.
249
34.
Choi
H.
,
Antoniou
M. G.
,
Pelaez
M.
,
De la Cruz
A. A.
,
Shoemaker
J. A.
,
Dionysiou
D. D.
,
Environ. Sci. Technol.
,
2007
, vol.
41
pg.
7530
35.
Pelaez
M.
,
Falaras
P.
,
Kontos
A. G.
,
de la Cruz
A. A.
,
O’Shea
K.
,
Dunlop
P. S. M.
,
Byrne
J. A.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2012
, vol.
121–122
pg.
30
36.
Kiszczak
L.
,
Tropilo
J.
,
Rocz. Panstw. Zakl. Hig.
,
1988
, vol.
39
pg.
314
37.
Milic
N.
,
Milanovic
M.
,
Letic
N. G.
,
Sekulic
M. T.
,
Radonic
J.
,
Mihajlovic
I.
,
Miloradov
M. V.
,
Int. J. Environ. Health Res.
,
2013
, vol.
23
pg.
296
38.
Wei
R.
,
Ge
F.
,
Huang
S.
,
Chen
M.
,
Wang
R.
,
Chemosphere
,
2011
, vol.
82
pg.
1408
39.
Brausch
J. M.
,
Rand
G. M.
,
Chemosphere
,
2011
, vol.
82
pg.
1518
40.
Hirsch
R.
,
Ternes
T.
,
Haberer
K.
,
Kratz
K. L.
,
Sci. Total Environ.
,
1999
, vol.
225
pg.
109
41.
Li
K.
,
Yediler
A.
,
Yang
M.
,
Schulte-Hostede
S.
,
Wong
M. H.
,
Chemosphere
,
2008
, vol.
72
pg.
473
42.
Xuan
R.
,
Arisi
L.
,
Wang
Q.
,
Yates
S. R.
,
Biswas
K. C.
,
J. Environ. Sci. Health, Part B
,
2010
, vol.
45
pg.
73
43.
Zhao
C.
,
Pelaez
M.
,
Duan
X.
,
Deng
H.
,
O’Shea
K.
,
Fatta-Kassinos
D.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2013
, vol.
134–135
pg.
83
44.
Jiao
S.
,
Zheng
S.
,
Yin
D.
,
Wang
L.
,
Chen
L.
,
J. Environ. Sci
,
2008
, vol.
20
pg.
806
45.
Pereira
J. H. O. S.
,
Vilar
V. J. P.
,
Borges
M. T.
,
González
O.
,
Esplugas
S.
,
Boaventura
R. A. R.
,
Sol. Energy
,
2011
, vol.
85
pg.
2732
46.
Pereira
J. H. O. S.
,
Reis
A. C.
,
Queirós
D.
,
Nunes
O. C.
,
Borges
M. T.
,
Vilar
V. J. P.
,
Boaventura
R. A. R.
,
Sci. Total Environ.
,
2013
, vol.
463–464
pg.
274
47.
Han
C.
,
Likodimos
V.
,
Khan
J. A.
,
Nadagouda
M. N.
,
Andersen
J.
,
Falaras
P.
,
Rosales-Lombardi
P.
,
Dionysiou
D. D.
,
Environ. Sci. Pollut. Res.
,
2014
, vol.
21
pg.
11781
48.
Zhao
C.
,
Deng
H.
,
Li
Y.
,
Liu
Z.
,
J. Hazard. Mater.
,
2010
, vol.
176
pg.
884
49.
Zhao
C.
,
Zhou
Y.
,
Ridder
D. J. D.
,
Zhai
J.
,
Wei
Y.
,
Deng
H.
,
Chem. Eng. J.
,
2014
, vol.
248
pg.
280
50.
Pelaez
M.
,
de la Cruz
A. A.
,
O’Shea
K.
,
Falaras
P.
,
Dionysiou
D. D.
,
Water Res.
,
2011
, vol.
45
pg.
3787
51.
Haroune
L.
,
Salaun
M.
,
Ménard
A.
,
Legault
C. Y.
,
Bellenger
J.-P.
,
Sci. Total Environ.
,
2014
, vol.
475
pg.
16
52.
Avisar
D.
,
Horovitz
I.
,
Lozzi
L.
,
Ruggieri
F.
,
Baker
M.
,
Abel
M.-L.
,
Mamane
H.
,
J. Hazard. Mater.
,
2013
, vol.
244–245
pg.
463
53.
Chong
M. N.
,
Jin
B.
,
Chow
C. W. K.
,
Saint
C.
,
Water Res.
,
2010
, vol.
44
pg.
2997
54.
Lawton
L. A.
,
Robertson
P. K. J.
,
Cornish
B. J. P. A.
,
Marr
I. L.
,
Jaspars
M.
,
J. Catal.
,
2003
, vol.
213
pg.
109
55.
Mathews
R. W.
,
Water Res.
,
1986
, vol.
20
pg.
569
56.
Shifu
C.
,
Gengyu
C.
,
Sol. Energy
,
2005
, vol.
79
pg.
1
57.
Shafirvich
V.
,
Dourandin
A.
,
Huang
W.
,
Geacintov
N. E.
,
J. Biol. Chem.
,
2001
, vol.
276
pg.
24621
58.
Zhao
D.
,
Liao
X.
,
Yan
X.
,
Huling
S. G.
,
Chai
T.
,
Tao
H.
,
J. Hazard. Mater.
,
2013
, vol.
254–255
pg.
228
59.
Assalin
M. R.
,
De Moraes
S. G.
,
Queiroz
S. C. N.
,
Ferracini
V. L.
,
Duran
N.
,
J. Environ. Sci. Health, Part B
,
2010
, vol.
45
pg.
89
60.
Cho
M.
,
Chung
H.
,
Choi
W.
,
Yoon
J.
,
Water Res.
,
2004
, vol.
38
pg.
1069
61.
Andreozzi
R.
,
Marotta
R.
,
Pinto
G.
,
Pollio
A.
,
Water Res.
,
2002
, vol.
36
pg.
2869
62.
Doll
T. E.
,
Frimmel
F. H.
,
Catal. Today
,
2005
, vol.
101
pg.
195
63.
Sakkas
V. A.
,
Lambropoulou
D. A.
,
Albanis
T. A.
,
Chemosphere
,
2002
, vol.
48
pg.
939
64.
Grbin
J. R.
,
Milori
D. M.
,
Simões
M. L.
,
da Silva
W. T.
,
Neto
L. M.
,
Chemosphere
,
2007
, vol.
66
pg.
1692
65.
Zakersalehi
A.
,
Nadagouda
M.
,
Choi
H.
,
Catal. Comm.
,
2013
, vol.
41
pg.
79
66.
Paz
Y.
,
C. R. Chim.
,
2006
, vol.
9
pg.
774
67.
Inumaru
K.
,
Murashima
M.
,
Kasahara
T.
,
Yamanaka
S.
,
Appl. Catal., B
,
2004
, vol.
52
pg.
275
68.
Paz
Y.
,
Solid State Phenom.
,
2010
, vol.
162
pg.
135
69.
Ramstörm
O.
,
Ansell
R. J.
,
Chirality
,
1998
, vol.
10
pg.
195
70.
Shimizu
K. I.
,
Kaneko
T.
,
Fujishima
T.
,
Kodama
T.
,
Yoshida
H.
,
Kitayama
Y.
,
Appl. Catal., A
,
2002
, vol.
225
pg.
185
71.
Kato
R.
,
Shimura
N.
,
Ogawa
M.
,
Chem. Lett.
,
2008
, vol.
37
pg.
76
72.
Hoffmann
M. R.
,
Martin
S. T.
,
Choi
W.
,
Bahnemann
D. W.
,
Chem. Rev.
,
1995
, vol.
95
pg.
69
73.
Malato
S.
,
Fernández-Ibáñez
P.
,
Maldonado
M. I.
,
Blanco
J.
,
Gernjak
W.
,
Catal. Today
,
2009
, vol.
147
pg.
1
74.
Triantis
T. M.
,
Fotiou
T.
,
Kaloudis
T.
,
Kontos
A. G.
,
Falaras
P.
,
Dionysiou
D. D.
,
Pelaez
M.
,
Hiskia
A.
,
J. Hazard. Mater.
,
2012
, vol.
211–212
pg.
196
75.
Antoniou
M. G.
,
Shoemaker
J. A.
,
De la Cruz
A. A.
,
Dionysiou
D. D.
,
Toxicon
,
2008
, vol.
51
pg.
1103
76.
Vinodgopal
K.
,
Bedja
I.
,
Hotchandani
S.
,
Kamat
P. V.
,
Langmuir
,
1994
, vol.
10
pg.
1767
77.
Pelizzetti
E.
,
Maurino
V.
,
Minero
C.
,
Carlin
V.
,
Tosato
M. L.
,
Pramauro
E.
,
Zerbinati
O.
,
Environ. Sci. Technol.
,
1990
, vol.
24
pg.
1559
78.
D’Oliveira
J. C.
,
Minero
C.
,
Pelizzetti
E.
,
Pichat
P.
,
J. Photochem. Photobiol., A
,
1993
, vol.
72
pg.
261
79.
Hapeshi
E.
,
Achilleos
A.
,
Vasquez
M. I.
,
Michael
C.
,
Xekoukoulotakis
N. P.
,
Mantzavinos
D.
,
Kassinos
D.
,
Water Res.
,
2010
, vol.
44
pg.
1737
80.
Calza
P.
,
Pelizzetti
E.
,
Minero
C.
,
J. Appl. Electrochem.
,
2005
, vol.
35
pg.
665
81.
A.
Hiskia
,
M. T.
Triantis
,
M. G.
Antoniou
,
A. A.
De la Cruz
,
K.
O’Shea
,
W.
Song
,
T.
Fotiou
,
T.
Kaloudis
,
X.
He
,
J.
Andersen
and
D. D.
Dionysiou
, in
Transformation Products of Emerging Contaminants in the Environment: Analysis, Processes, Occurrence, Effects and Risks
, ed. L. Nollet and D. Lambropoulou,
John Wiley & Sons, Inc.
,
2013
, pp. 687–720
82.
Kormali
P.
,
Dimoticali
D.
,
Tsipi
D.
,
Hiskia
A.
,
Papaconstantinou
E.
,
Appl. Catal., B
,
2004
, vol.
48
pg.
175
83.
Kouloumbos
V.
,
Tsipi
D.
,
Hiskia
A.
,
Nikolic
D.
,
van Breemen
R.
,
J. Am. Soc. Mass Spectrom.
,
2003
, vol.
14
pg.
803
84.
Calza
P.
,
Massolino
C.
,
Pelizzetti
E.
,
J. Photochem. Photobiol., A
,
2008
, vol.
200
pg.
356
85.
I. K.
Konstantinou
, in
Mass Spectrometry for Analysis of Pesticide Residues and their Metabolites
, ed. D. Tsipi, H. Botitsi and A. Economou,
John Wiley & Sons
,
2015
86.
Fotiou
T.
,
Triantis
T.
,
Kaloudis
T.
,
Hiskia
A.
,
Chemosphere
,
2015
, vol.
119
pg.
89
87.
Fotiou
T.
,
Triantis
T. M.
,
Kaloudis
T.
,
Papaconstantinou
E.
,
Hiskia
A.
,
J. Photochem. Photobiol., A
,
2014
, vol.
286
pg.
1
88.
Antoniou
M. G.
,
De la Cruz
A. A.
,
Dionysiou
D. D.
,
Environ. Sci. Technol.
,
2010
, vol.
44
pg.
7238
89.
Hoffmann
M. R.
,
Martin
S. T.
,
Choi
W.
,
Bahnemann
D. W.
,
Chem. Rev.
,
1995
, vol.
95
pg.
69
90.
Pichat
P.
,
Water Sci. Technol.
,
2007
, vol.
55
pg.
167
91.
Sharma
V. K.
,
Triantis
T. M.
,
Antoniou
M. G.
,
He
X.
,
Pelaez
M.
,
Han
C.
,
Song
W.
,
O’Shea
K. E.
,
De La Cruz
A. A.
,
Kaloudis
T.
,
Hiskia
A.
,
Dionysiou
D. D.
,
Sep. Purif. Technol.
,
2012
, vol.
91
pg.
3
92.
Cermenati
L.
,
Pichat
P.
,
Guillard
C.
,
Albini
A.
,
J. Phys. Chem. B
,
1997
, vol.
101
pg.
2650
93.
Liu
I.
,
Lawton
L. A.
,
Cornish
B.
,
Robertson
P. K. J.
,
J. Photochem. Photobiol., A
,
2002
, vol.
148
pg.
349
94.
Liu
I.
,
Lawton
L. A.
,
Bahnemann
D. W.
,
Robertson
P. K. J.
,
Appl. Catal., B
,
2005
, vol.
60
pg.
245
95.
Antoniou
M. G.
,
Shoemaker
J. A.
,
De la Cruz
A. A.
,
Dionysiou
D. D.
,
Environ. Sci Technol.
,
2008
, vol.
42
pg.
8877
96.
Buxton
G. V.
,
Greenstock
C. L.
,
Helman
W. P.
,
Ross
A. B.
,
J. Phys. Chem. Ref. Data
,
1988
, vol.
17
pg.
513
97.
Song
W.
,
De La Cruz
A. A.
,
Rein
K.
,
O’Shea
K. E.
,
Environ. Sci. Technol.
,
2006
, vol.
40
pg.
3941
98.
J.
McMurry
,
Organic Chemistry
,
Brooks Cole Pub Co
,
1995
99.
Antoniou
M. G.
,
Nicolaou
P. A.
,
Shoemaker
J. A.
,
De la Cruz
A. A.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2009
, vol.
91
pg.
165
100.
Tsuji
K.
,
Watanuki
T.
,
Kondo
F.
,
Watanabe
M. F.
,
Suzuki
S.
,
Nakazawa
H.
,
Suzuki
M.
,
Uchida
H.
,
Harada
K.
,
Toxicon
,
1995
, vol.
33
pg.
1619
101.
Takenaka
S.
,
Tanaka
Y.
,
Chemosphere
,
1995
, vol.
31
pg.
3635
102.
Pelaez
M.
,
Falaras
P.
,
Likodimos
V.
,
Kontos
A. G.
,
De la Cruz
A. A.
,
O’shea
K.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2010
, vol.
99
pg.
378
103.
Andersen
J.
,
Han
C.
,
O’Shea
K.
,
Dionysiou
D. D.
,
Appl. Catal., B
,
2014
, vol.
154–155
pg.
259
104.
Burlingame
G. A.
,
Dann
R. M.
,
Brock
G. L.
,
Am. Water Works Assoc.
,
1986
, vol.
78
pg.
56
105.
Bamuza Pemu
E. E.
,
Chirwa
E. M.
,
Chem. Eng. Trans.
,
2011
, vol.
24
pg.
91
106.
Hiskia
A.
,
Androulaki
E.
,
Mylonas
A.
,
Boyatzis
S.
,
Dimoticali
D.
,
Minero
C.
,
Pelizzetti
E.
,
Papaconstantinou
E.
,
Res. Chem. Intermed.
,
2000
, vol.
26
pg.
235
Close Modal

or Create an Account

Close Modal
Close Modal