**3. Results and discussion**

170 Herbicides – Properties, Synthesis and Control of Weeds

\* previously identified using TiO2 Degussa P25

Table 2. MS/MS fragmentation data of CLP photodegradation intermediates (Part II).

\*\* wide spectral band *M*MI – monoisotopic weight

#### **3.1 Effects of the type of TiO2**

The photocatalytic activity of TiO2 Wackherr was compared to that of the most often used Degussa P25 under UV and visible irradiation. As can be seen from Figure 1, practically no degradation was observed under the visible light irradiation, either in the presence or absence of TiO2. The lack of CLP disappearance in the presence of TiO2 under these conditions also allows the exclusion of a significant adsorption of CLP on the catalyst surface during the course of the irradiation. In contrast, significant CLP removal could be observed under UV, and the process involving TiO2 Wackherr was slightly faster compared to that observed in the presence of Degussa P25. This insignificant acceleration of the degradation of CLP in the presence of TiO2 Wackherr is noteworthy, considering that this TiO2 specimen has much larger particles (average radii in solution are 3–4 times larger) than Degussa P25 and a surface area that is almost six times lower (Vione et al., 2005).

The direct photolysis of CLP was also checked under the adopted irradiation conditions, in the absence of a catalyst (Figure 1). It appears that CLP can be degraded by direct photolysis in the near UV region, but at a significantly lower rate compared to the photocatalytic process.

Under the relevant experimental conditions, the reaction followed a pseudo-first order kinetics. On the basis of the kinetic curves ln*c* (substrate concentration) vs. *t*, the values of the pseudo-first order rate constant *k′* were calculated. The degradation rate of CLP was calculated for all the investigated as the product *k′ c*0, where *c*0 is the initial concentration of CLP.

Comparative Assessment of the Photocatalytic Efficiency

**3.2 Effect of catalyst loading** 

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 173

Fig. 2. Comparison of the photocatalytic removal parent compound and mineralisation of CLP. The inset shows the degradation rate (*R*) calculated over 240 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 = 2.0 mg mL–1, *t* = 40 oC, at pH ~ 3.5.

Due to the inherent nature of heterogeneous photocatalytic systems, there is always an optimum catalyst concentration at which the removal rate is at its maximum. In this study, the optimum was determined by changing the concentration of TiO2 Wackherr over the loading range from 0.2 to 2.0 mg mL–1, as shown in Figure 3. As can be seen from the figure inset, the increase in TiO2 loading up to 1.0 mg mL–1 was accompanied by an increase in the degradation rate, but a further increase caused an opposite effect. Theoretically, the increase in the catalyst loading above an optimum value has no effect on the photodegradation rate since all the light available is already utilized. However, higher loading of TiO2 led to the aggregation of its particles and thus to a decrease in the contact surface between the reactant and photocatalyst particles, which caused a decrease in the number of active sites, resulting in a lower rate of photodegradation. Also, when TiO2 is overdosed, the intensity of the incident UV light is attenuated because of the decreased light penetration and increased scattering, which attenuates the positive effect coming from the dosage increment, and therefore the overall performance decreases (Wong & Chu, 2003). Optimum catalyst concentration is a complex function of a number of parameters including catalyst agglomeration, the suspension opacity, light scattering, mixing, reactor type, and the pollutant type (Toor et al., 2006; Mendez-Arriaga et al., 2008); hence it is not constant for all photocatalytic systems. Indeed, optimum catalyst concentrations have been reported to vary between as low as 0.1 mg mL–1 to as high as around 10 mg mL–1 (Mendez-Arriaga et al., 2008; Alhakimi et al., 2008; Chu et al., 2009a). An optimum catalyst concentration of around 1 mg mL–1 has been generally reported in many studies (Chen & Ray, 1998; Lu et al., 1999; Mendez-Arriaga et al., 2008; Rajeswari & Kanmani, 2009; Tizaoui et al., 2011). However, by comparing these results with our previously finding that the optimal catalyst loading of

Fig. 1. Kinetics of the photolytic and photocatalytic degradation of CLP (1.0 mM). When present, the TiO2 loading was 2.0 mg mL–1. Operation conditions: *t* = 40 oC at pH ~ 3.5.

A comparison of the mineralisation capacities of TiO2 Wackherr and Degussa P25, presented in Figure 2, shows that the mineralisation efficiency in the presence of TiO2 Wackherr is significantly higher — 90% of CLP was mineralised during 240 min, whereas in the case of Degussa P25 only about 60%. Also, the ratio of the removal rate of the parent compound and total mineralisation in the case of TiO2 Wackherr is about 1.20, whereas in the case of P25 this ratio is significantly higher, amounting to even 2.80. If these results are compared with those of the photocatalytic degradation of pyridine pesticides such as picloram and triclopyr (Abramović et al., 2011) it can be seen that the ratios of the mineralisation rate to the rate of the parent compound degradation are different and that they depend on the type of the substituent in the pyridine ring. Namely, in the photocatalytic degradation of triclopyr, TiO2 Degussa P25 showed a higher efficiency both in the process of mineralisation and degradation of the parent compound. In the case of picloram, TiO2 Wackherr showed higher photocatalytic efficiency than Degussa P25. However, the ratio of the rates of removal of the parent compound and total mineralisation in the presence of TiO2 Wackherr is much higher in the case of picloram than of CLP, and it amounts to about 9, whereas in the presence of Degussa P25 this ratio is about 3, which is similar to the value obtained also for CLP.

Fig. 2. Comparison of the photocatalytic removal parent compound and mineralisation of CLP. The inset shows the degradation rate (*R*) calculated over 240 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 = 2.0 mg mL–1, *t* = 40 oC, at pH ~ 3.5.

#### **3.2 Effect of catalyst loading**

172 Herbicides – Properties, Synthesis and Control of Weeds

Fig. 1. Kinetics of the photolytic and photocatalytic degradation of CLP (1.0 mM). When present, the TiO2 loading was 2.0 mg mL–1. Operation conditions: *t* = 40 oC at pH ~ 3.5.

for CLP.

A comparison of the mineralisation capacities of TiO2 Wackherr and Degussa P25, presented in Figure 2, shows that the mineralisation efficiency in the presence of TiO2 Wackherr is significantly higher — 90% of CLP was mineralised during 240 min, whereas in the case of Degussa P25 only about 60%. Also, the ratio of the removal rate of the parent compound and total mineralisation in the case of TiO2 Wackherr is about 1.20, whereas in the case of P25 this ratio is significantly higher, amounting to even 2.80. If these results are compared with those of the photocatalytic degradation of pyridine pesticides such as picloram and triclopyr (Abramović et al., 2011) it can be seen that the ratios of the mineralisation rate to the rate of the parent compound degradation are different and that they depend on the type of the substituent in the pyridine ring. Namely, in the photocatalytic degradation of triclopyr, TiO2 Degussa P25 showed a higher efficiency both in the process of mineralisation and degradation of the parent compound. In the case of picloram, TiO2 Wackherr showed higher photocatalytic efficiency than Degussa P25. However, the ratio of the rates of removal of the parent compound and total mineralisation in the presence of TiO2 Wackherr is much higher in the case of picloram than of CLP, and it amounts to about 9, whereas in the presence of Degussa P25 this ratio is about 3, which is similar to the value obtained also Due to the inherent nature of heterogeneous photocatalytic systems, there is always an optimum catalyst concentration at which the removal rate is at its maximum. In this study, the optimum was determined by changing the concentration of TiO2 Wackherr over the loading range from 0.2 to 2.0 mg mL–1, as shown in Figure 3. As can be seen from the figure inset, the increase in TiO2 loading up to 1.0 mg mL–1 was accompanied by an increase in the degradation rate, but a further increase caused an opposite effect. Theoretically, the increase in the catalyst loading above an optimum value has no effect on the photodegradation rate since all the light available is already utilized. However, higher loading of TiO2 led to the aggregation of its particles and thus to a decrease in the contact surface between the reactant and photocatalyst particles, which caused a decrease in the number of active sites, resulting in a lower rate of photodegradation. Also, when TiO2 is overdosed, the intensity of the incident UV light is attenuated because of the decreased light penetration and increased scattering, which attenuates the positive effect coming from the dosage increment, and therefore the overall performance decreases (Wong & Chu, 2003). Optimum catalyst concentration is a complex function of a number of parameters including catalyst agglomeration, the suspension opacity, light scattering, mixing, reactor type, and the pollutant type (Toor et al., 2006; Mendez-Arriaga et al., 2008); hence it is not constant for all photocatalytic systems. Indeed, optimum catalyst concentrations have been reported to vary between as low as 0.1 mg mL–1 to as high as around 10 mg mL–1 (Mendez-Arriaga et al., 2008; Alhakimi et al., 2008; Chu et al., 2009a). An optimum catalyst concentration of around 1 mg mL–1 has been generally reported in many studies (Chen & Ray, 1998; Lu et al., 1999; Mendez-Arriaga et al., 2008; Rajeswari & Kanmani, 2009; Tizaoui et al., 2011). However, by comparing these results with our previously finding that the optimal catalyst loading of

Comparative Assessment of the Photocatalytic Efficiency

3.5.

**3.4 Effect of temperature** 

thermal activation (Topalov et al., 2004).

**3.5 Effect of the initial pH** 

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 175

Fig. 4. Effect of the initial CLP concentration on the kinetics of photodegradation. The inset shows the effect of initial CLP concentration on the CLP degradation rate (*R*) calculated for 60 min of irradiation. Operation conditions: TiO2 Wackherr = 2.0 mg mL–1, *t* = 40 oC at pH ~

The photocatalytic degradation of CLP was studied in a temperatures range from 25 to 40 oC (Figure 5) and the rate constant *k′* was determined from the pseudo-first order plots. In general, the rate constant is expected to increase at higher temperatures, but it appeared that CLP more easily degraded at lower temperatures in the TiO2 Wackherr suspension. Thus, the decrease in the rate constant observed in this temperature range may be attributed to the physisorption between the TiO2 surface and the CLP molecules (Ishiki et al., 2005). Namely, the temperature of fastest CLP removal was, surprisingly, 25 oC, and hence all the further measurements were carried out at this temperature. The energy of activation, *E*a, was calculated from the Arrhenius plot of ln*k* versus 1/*T* (K–1), and it amounted to 37.9 kJ mol–1. Obviously, this value is somewhat higher than that obtained for the photocatalytic degradation of CLP in the presence TiO2 Degussa P25 (Šojić et al., 2009), but it is acceptable since for TiO2 photocatalyst, irradiation is the primary source of the electron-hole pair generation at ambient temperature, as the band gap energy is too high to be overcome by

The effect of pH is very important in the heterogeneous photocatalytic removal of organic molecules since it influences both the surface charge of TiO2 and the ionic form of the reactant, influencing thus the electrostatic interactions between the reactant species and the catalyst surface. Moreover, the pH influences the sizes of TiO2 aggregates, interaction of the

TiO2 Degussa P25 was 4.0 mg mL–1 (Šojić et al., 2009), it can be concluded that the effect of catalyst loading on the efficiency of the photocatalytic removal of CLP is influenced by the type of TiO2 used. If we compare the efficiencies of the two catalysts at their optimal loadings (i.e. 1.0 mg mL–1 for TiO2 Wackherr and 4.0 mg mL–1 for Degussa P25), it comes out that TiO2 Wackherr, although being present in a lower amount, is more efficient in the removal of CLP.

Fig. 3. Effect of the TiO2 Wackherr catalyst loading on the kinetics of CLP photodegradation. The inset shows the effect of TiO2 Wackherr loading on the degradation rate (*R*) determined after 120 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, *t* = 40 oC at pH ~ 3.5.

#### **3.3 Effect of the initial CLP concentration**

Many studies have shown that the initial pollutant concentration has a significant effect on the rate of its photocatalytic removal. In this work, the effect of the initial concentration of CLP on the photodegradation rate was studied under UV light using TiO2 Wackherr in the loading range from 0.25 to 1.0 mM (Figure 4). As can been seen from the inset of Figure 4, the degradation rate decreased with increase in the CLP concentration above to 0.5 mM. Such behaviour may be explained by the fact that at an increased concentration of CLP more of its molecules can be adsorbed on the photocatalyst surface, needing thus a larger catalyst area for their degradation. However, as the intensity of light, irradiation time and amount of catalyst are constant, the relative amounts of O2 and OH radicals on the catalyst surface do not increase (Atiqur Rahman & Muneer, 2005; Qamar et al., 2006).

An alternative explanation for the effect of the substrate concentration is the competition for reactive species between the substrate and the transformation intermediates, the concentration of which increase with increasing substrate concentration, or the poisoning of the photocatalyst surface by the intermediates themselves (Abramović et al., 2011).

Fig. 4. Effect of the initial CLP concentration on the kinetics of photodegradation. The inset shows the effect of initial CLP concentration on the CLP degradation rate (*R*) calculated for 60 min of irradiation. Operation conditions: TiO2 Wackherr = 2.0 mg mL–1, *t* = 40 oC at pH ~ 3.5.

#### **3.4 Effect of temperature**

174 Herbicides – Properties, Synthesis and Control of Weeds

TiO2 Degussa P25 was 4.0 mg mL–1 (Šojić et al., 2009), it can be concluded that the effect of catalyst loading on the efficiency of the photocatalytic removal of CLP is influenced by the type of TiO2 used. If we compare the efficiencies of the two catalysts at their optimal loadings (i.e. 1.0 mg mL–1 for TiO2 Wackherr and 4.0 mg mL–1 for Degussa P25), it comes out that TiO2 Wackherr, although being present in a lower amount, is more efficient in the

Fig. 3. Effect of the TiO2 Wackherr catalyst loading on the kinetics of CLP photodegradation. The inset shows the effect of TiO2 Wackherr loading on the degradation rate (*R*) determined after 120 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, *t* = 40 oC at pH ~ 3.5.

Many studies have shown that the initial pollutant concentration has a significant effect on the rate of its photocatalytic removal. In this work, the effect of the initial concentration of CLP on the photodegradation rate was studied under UV light using TiO2 Wackherr in the loading range from 0.25 to 1.0 mM (Figure 4). As can been seen from the inset of Figure 4, the degradation rate decreased with increase in the CLP concentration above to 0.5 mM. Such behaviour may be explained by the fact that at an increased concentration of CLP more of its molecules can be adsorbed on the photocatalyst surface, needing thus a larger catalyst area for their degradation. However, as the intensity of light, irradiation time and amount of

and

An alternative explanation for the effect of the substrate concentration is the competition for reactive species between the substrate and the transformation intermediates, the concentration of which increase with increasing substrate concentration, or the poisoning of

the photocatalyst surface by the intermediates themselves (Abramović et al., 2011).

OH radicals on the catalyst surface do

**3.3 Effect of the initial CLP concentration** 

catalyst are constant, the relative amounts of O2

not increase (Atiqur Rahman & Muneer, 2005; Qamar et al., 2006).

removal of CLP.

The photocatalytic degradation of CLP was studied in a temperatures range from 25 to 40 oC (Figure 5) and the rate constant *k′* was determined from the pseudo-first order plots. In general, the rate constant is expected to increase at higher temperatures, but it appeared that CLP more easily degraded at lower temperatures in the TiO2 Wackherr suspension. Thus, the decrease in the rate constant observed in this temperature range may be attributed to the physisorption between the TiO2 surface and the CLP molecules (Ishiki et al., 2005). Namely, the temperature of fastest CLP removal was, surprisingly, 25 oC, and hence all the further measurements were carried out at this temperature. The energy of activation, *E*a, was calculated from the Arrhenius plot of ln*k* versus 1/*T* (K–1), and it amounted to 37.9 kJ mol–1. Obviously, this value is somewhat higher than that obtained for the photocatalytic degradation of CLP in the presence TiO2 Degussa P25 (Šojić et al., 2009), but it is acceptable since for TiO2 photocatalyst, irradiation is the primary source of the electron-hole pair generation at ambient temperature, as the band gap energy is too high to be overcome by thermal activation (Topalov et al., 2004).

#### **3.5 Effect of the initial pH**

The effect of pH is very important in the heterogeneous photocatalytic removal of organic molecules since it influences both the surface charge of TiO2 and the ionic form of the reactant, influencing thus the electrostatic interactions between the reactant species and the catalyst surface. Moreover, the pH influences the sizes of TiO2 aggregates, interaction of the

Comparative Assessment of the Photocatalytic Efficiency

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 177

Fig. 6. Effect of the pH on the kinetics of CLP photocatalytic degradation. The inset shows the effect of pH on the degradation rate (*R*) calculated for 120 min of irradiation. Operation

A practical problem arising in the use of TiO2 as a photocatalyst is the undesired e

electron acceptors to the reaction mixture. They may have several different effects such as (1) to increase the number of trapped electrons and, consequently, avoid recombination, (2) to generate more radicals and other oxidising species, (3) to increase the oxidation rate of intermediate compounds, and (4) to avoid problems caused by low oxygen concentration. In highly toxic wastewater, where the degradation of organic pollutants is the major concern, the addition of electron acceptors to enhance the degradation rate may often be justified (Singh et al., 2007). The rates of photocatalytic degradation and mineralisation of CLP in the presence of various electron acceptors such as KBrO3, H2O2, and (NH4)2S2O8 in addition to

As can be seen, the mentioned electron acceptors showed different effects. Namely, only the addition of KBrO3 enhanced the rate of photocatalytic degradation of the parent compound (by a factor of 1.4), indicating that this compound is a more effective electron acceptor compared with other oxidants employed in this study. A possible explanation might be the change in the reaction mechanism of the photocatalytic degradation, since the reduction of

by themselves can act as oxidising agents (Singh et al., 2007). However,

by electrons does not lead directly to the formation of

the mineralisation rate is slightly lower (by a factor of 1.1).

formation of other reactive radicals or oxidising reagents e.g. BrO2

h+ pair

OH, but rather to the

and HOBr.

h+ pair recombination is to add other (irreversible)

conditions: *c* (CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC.

**3.6 Effect of electron acceptors** 

recombination. One strategy to inhibit e

the molecular oxygen are shown in Figure 7.

BrO3

Furthermore, BrO3

Fig. 5. Arrhenius plot of ln*k* versus 1/*T* for the photocatalytic degradation of CLP for the first 120 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1 at pH ~3.5.

solvent molecules with the catalyst and the type of radicals or intermediates formed during the photocatalytic reaction (Muneer & Bahnemann, 2002; Šojić at al., 2009; Tizaoui et al., 2011). All of these factors have a significant effect on the adsorption of solutes on TiO2 surfaces and, as a result, on the observed photodegradation rates. Because the real effluent stuff of pesticide can be discharged at a different pH, the pH effect on the photocatalytic rate degradation of CLP was studied in the pH range from 2.4 to 9.8 (Figure 6). The point of zero charge (pHpzc) of anatase is 5.8 (Karunakaran & Dhanalakshmi, 2009). Thus, the TiO2 surface will be positively charged ( TiOH2 ) in acidic media (pH pHpzc) and negatively charged (TiO ) in alkaline media (pH pHpzc). On the other hand, the p*K*a values for CLP are 1.4±0.1 and 4.4±0.1 (Corredor et al., 2006), so that at pH < 1.4 the herbicide is mainly present in its protonated form, and at pH > 4.4 in the anionic form. In the pH interval from 2.4 to 3.5, one can expect a great increase in the photodegradation rate, arising as a consequence of the dissociation of the carboxylic group and deprotonation of the pyridine nitrogen (to a significantly smaller extent). In this way, favourable electrical forces are generated that are manifested as the attraction between the positively charged surface of the catalyst and CLP anion. As can be seen in Figure 6 (inset), in the pH interval from 3.5 to 4.8, a distinct decrease of the photodegradation rate is observed, arising probably as a consequence of the decrease in the number of positive sites on the catalyst surface. A further increase in the pH up to 9.8 caused a decrease in the photodegradation rate, which was probably a consequence of the influence of several factors. Namely, at pH > pHpzc, the TiO2 surface is negatively charged, causing the repulsion of the CLP anion. Besides, unfavourable electrical forces are generated, i.e., the repulsion between the negatively charged surface of the catalyst and OH .

Fig. 6. Effect of the pH on the kinetics of CLP photocatalytic degradation. The inset shows the effect of pH on the degradation rate (*R*) calculated for 120 min of irradiation. Operation conditions: *c* (CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC.

#### **3.6 Effect of electron acceptors**

176 Herbicides – Properties, Synthesis and Control of Weeds

Fig. 5. Arrhenius plot of ln*k* versus 1/*T* for the photocatalytic degradation of CLP for the first 120 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0

solvent molecules with the catalyst and the type of radicals or intermediates formed during the photocatalytic reaction (Muneer & Bahnemann, 2002; Šojić at al., 2009; Tizaoui et al., 2011). All of these factors have a significant effect on the adsorption of solutes on TiO2 surfaces and, as a result, on the observed photodegradation rates. Because the real effluent stuff of pesticide can be discharged at a different pH, the pH effect on the photocatalytic rate degradation of CLP was studied in the pH range from 2.4 to 9.8 (Figure 6). The point of zero charge (pHpzc) of anatase is 5.8 (Karunakaran & Dhanalakshmi, 2009). Thus, the TiO2 surface

) in alkaline media (pH pHpzc). On the other hand, the p*K*a values for CLP are 1.4±0.1 and 4.4±0.1 (Corredor et al., 2006), so that at pH < 1.4 the herbicide is mainly present in its protonated form, and at pH > 4.4 in the anionic form. In the pH interval from 2.4 to 3.5, one can expect a great increase in the photodegradation rate, arising as a consequence of the dissociation of the carboxylic group and deprotonation of the pyridine nitrogen (to a significantly smaller extent). In this way, favourable electrical forces are generated that are manifested as the attraction between the positively charged surface of the catalyst and CLP anion. As can be seen in Figure 6 (inset), in the pH interval from 3.5 to 4.8, a distinct decrease of the photodegradation rate is observed, arising probably as a consequence of the decrease in the number of positive sites on the catalyst surface. A further increase in the pH up to 9.8 caused a decrease in the photodegradation rate, which was probably a consequence of the influence of several factors. Namely, at pH > pHpzc, the TiO2 surface is negatively charged, causing the repulsion of the CLP anion. Besides, unfavourable electrical forces are generated, i.e., the repulsion between the negatively charged surface of the

) in acidic media (pH pHpzc) and negatively charged

mg mL–1 at pH ~3.5.

(TiO

catalyst and OH

will be positively charged ( TiOH2

.

A practical problem arising in the use of TiO2 as a photocatalyst is the undesired e h+ pair recombination. One strategy to inhibit e h+ pair recombination is to add other (irreversible) electron acceptors to the reaction mixture. They may have several different effects such as (1) to increase the number of trapped electrons and, consequently, avoid recombination, (2) to generate more radicals and other oxidising species, (3) to increase the oxidation rate of intermediate compounds, and (4) to avoid problems caused by low oxygen concentration. In highly toxic wastewater, where the degradation of organic pollutants is the major concern, the addition of electron acceptors to enhance the degradation rate may often be justified (Singh et al., 2007). The rates of photocatalytic degradation and mineralisation of CLP in the presence of various electron acceptors such as KBrO3, H2O2, and (NH4)2S2O8 in addition to the molecular oxygen are shown in Figure 7.

As can be seen, the mentioned electron acceptors showed different effects. Namely, only the addition of KBrO3 enhanced the rate of photocatalytic degradation of the parent compound (by a factor of 1.4), indicating that this compound is a more effective electron acceptor compared with other oxidants employed in this study. A possible explanation might be the change in the reaction mechanism of the photocatalytic degradation, since the reduction of BrO3 by electrons does not lead directly to the formation of OH, but rather to the formation of other reactive radicals or oxidising reagents e.g. BrO2 and HOBr. Furthermore, BrO3 by themselves can act as oxidising agents (Singh et al., 2007). However, the mineralisation rate is slightly lower (by a factor of 1.1).

Comparative Assessment of the Photocatalytic Efficiency

degradation scheme (Figure 9).

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 179

them, 3,6-dichloro-pyridin-2-ol (compound **3**) and isomeric 3,6-dichloro hydroxypyridine-2 carboxylic acids (compounds **4** and **7**) were previously identified (Šojić et al., 2009) in the presence of TiO2 Degussa P25. Using the positive and negative ionization MS2 spectra, it was possible to identify the remaining compounds and propose a photocatalytic

Fig. 8. Kinetics of the appearance/disappearance of CLP and intermediates in the photocatalytic degradation of CLP monitored by LC-ESI-MS/MS. Operation conditions:

Compound **2**, eluting at 0.81 min, had *M*MI 223, two chlorine atoms (on the basis of the A+2 isotopic peak intensity, as well as two consecutive losses of HCl in the MS2 spectra: 178142 and 142106) and an odd number of nitrogen atoms (odd molecular weight), and was visible only in negative mode. In both the first-order and second-order MS spectra, the loss of CO2 was observed (222178), pointing out to the presence of carboxylic group. On the basis of the molecular weight (32 units higher than that of CLP) and spectral data, it was concluded that the compound is 3,6-dichloro-4,5-dihydroxypyridine-2-carboxylic acid.

Compound **5** eluted at 1.57 min. On the basis of the A+2 isotopic peak intensity and molecular weight, it could be concluded that it contains one chlorine atom and an odd number of nitrogen atoms. The only fragmentation observable in the NI MS1 and MS2 spectra was the loss of the carboxylic group as CO2 (172128). The monoisotopic weight of 173 mass units could be explained by the loss of one chlorine atom (which is in agreement with the isotopic profile) from the CLP molecule and introduction of one hydroxyl. Thus, the compound was identified as either 6-chloro-3-hydroxypyridine-2-carboxylic acid or 3-

Finally, compound **6** was characterized by the odd monoisotopic weight of 179 units (pointing out to the odd number of nitrogen atoms), presence of two chlorine atoms, and the absence of carboxylic group loss both in positive mode (no sequential loss of H2O and CO)

*c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH ~3.5.

chloro-6-hydroxypyridine-2-carboxylic acid.

However, the presence of H2O2 caused a decrease in both the rate of removal of CLP (by a factor of 1.7) and its mineralisation (by a factor of 1.3). Such a negative effect of H2O2 is probably a consequence of the fact that it can also act as an OH scavenger, generating much less reactive hydroperoxyl radicals ( HO2 ). The HO2 can further react with the remaining strong OH to form ineffective oxygen and water. Besides, at a higher dose, H2O2 might absorb and thus attenuate the incident UV light available for the photocatalysis process (Chu & Wong, 2004; Muruganandham & Swaminathan, 2006).

The presence of <sup>2</sup> 2 8 S O had an insignificant effect on the rate of degradation of CLP, but it decreased the rate of mineralisation more than KBrO3 and H2O2, which can be explained by an increase in the concentration of <sup>2</sup> <sup>4</sup> SO adsorbed on the TiO2 surface, reducing thus the catalytic activity. The excess of adsorbed <sup>2</sup> <sup>4</sup> SO also reacts with the photogenerated holes and with the OH (San et al., 2001; Muruganandham & Swaminathan, 2006).

Fig. 7. Comparison of the degradation rate (*R*) of CLP removal and mineralisation in the presence of different electron acceptors (3 mM) calculated for 60 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH ~3.5.

#### **3.7 Effect of OH scavenger**

In order to investigate whether the heterogeneous photocatalysis takes place via OH, ethanol was added to the reaction mixture. Namely, it is known that alcohols, e.g. ethanol, act as OH scavengers (Daneshvar et al., 2004). The results obtained (data not shown) indicate that the degradation rate was significantly slower (by about 100 times) compared to that observed in the absence of ethanol, which proves that the reaction of photocatalytic degradation proceeded via OH.

#### **3.8 Intermediates and the mechanism of photodegradation**

The LC-MS analysis of the irradiated CLP solutions indicated the formation of six intermediates (labelled 1–7, Table 2), whose kinetic curves are shown in Figure 8. Three of

However, the presence of H2O2 caused a decrease in both the rate of removal of CLP (by a factor of 1.7) and its mineralisation (by a factor of 1.3). Such a negative effect of H2O2 is

). The HO2

absorb and thus attenuate the incident UV light available for the photocatalysis process

decreased the rate of mineralisation more than KBrO3 and H2O2, which can be explained by

OH (San et al., 2001; Muruganandham & Swaminathan, 2006).

Fig. 7. Comparison of the degradation rate (*R*) of CLP removal and mineralisation in the presence of different electron acceptors (3 mM) calculated for 60 min of irradiation. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH ~3.5.

In order to investigate whether the heterogeneous photocatalysis takes place via

ethanol was added to the reaction mixture. Namely, it is known that alcohols, e.g. ethanol,

indicate that the degradation rate was significantly slower (by about 100 times) compared to that observed in the absence of ethanol, which proves that the reaction of photocatalytic

The LC-MS analysis of the irradiated CLP solutions indicated the formation of six intermediates (labelled 1–7, Table 2), whose kinetic curves are shown in Figure 8. Three of

OH scavengers (Daneshvar et al., 2004). The results obtained (data not shown)

OH to form ineffective oxygen and water. Besides, at a higher dose, H2O2 might

2 8 S O had an insignificant effect on the rate of degradation of CLP, but it

<sup>4</sup> SO adsorbed on the TiO2 surface, reducing thus the

<sup>4</sup> SO also reacts with the photogenerated holes

OH scavenger, generating much

OH,

can further react with the remaining

probably a consequence of the fact that it can also act as an

(Chu & Wong, 2004; Muruganandham & Swaminathan, 2006).

less reactive hydroperoxyl radicals ( HO2

an increase in the concentration of <sup>2</sup>

catalytic activity. The excess of adsorbed <sup>2</sup>

**OH scavenger** 

OH.

**3.8 Intermediates and the mechanism of photodegradation** 

strong

The presence of <sup>2</sup>

and with the

**3.7 Effect of** 

degradation proceeded via

act as

them, 3,6-dichloro-pyridin-2-ol (compound **3**) and isomeric 3,6-dichloro hydroxypyridine-2 carboxylic acids (compounds **4** and **7**) were previously identified (Šojić et al., 2009) in the presence of TiO2 Degussa P25. Using the positive and negative ionization MS2 spectra, it was possible to identify the remaining compounds and propose a photocatalytic degradation scheme (Figure 9).

Fig. 8. Kinetics of the appearance/disappearance of CLP and intermediates in the photocatalytic degradation of CLP monitored by LC-ESI-MS/MS. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH ~3.5.

Compound **2**, eluting at 0.81 min, had *M*MI 223, two chlorine atoms (on the basis of the A+2 isotopic peak intensity, as well as two consecutive losses of HCl in the MS2 spectra: 178142 and 142106) and an odd number of nitrogen atoms (odd molecular weight), and was visible only in negative mode. In both the first-order and second-order MS spectra, the loss of CO2 was observed (222178), pointing out to the presence of carboxylic group. On the basis of the molecular weight (32 units higher than that of CLP) and spectral data, it was concluded that the compound is 3,6-dichloro-4,5-dihydroxypyridine-2-carboxylic acid.

Compound **5** eluted at 1.57 min. On the basis of the A+2 isotopic peak intensity and molecular weight, it could be concluded that it contains one chlorine atom and an odd number of nitrogen atoms. The only fragmentation observable in the NI MS1 and MS2 spectra was the loss of the carboxylic group as CO2 (172128). The monoisotopic weight of 173 mass units could be explained by the loss of one chlorine atom (which is in agreement with the isotopic profile) from the CLP molecule and introduction of one hydroxyl. Thus, the compound was identified as either 6-chloro-3-hydroxypyridine-2-carboxylic acid or 3 chloro-6-hydroxypyridine-2-carboxylic acid.

Finally, compound **6** was characterized by the odd monoisotopic weight of 179 units (pointing out to the odd number of nitrogen atoms), presence of two chlorine atoms, and the absence of carboxylic group loss both in positive mode (no sequential loss of H2O and CO)

Comparative Assessment of the Photocatalytic Efficiency

longer dominated by the parent compound.

**H-4-II-E cell line**

**MRC-5 cell line**

a)

b)

(*n* = 8).

**80 100 120**

**Cell growth (%)**

**Cell growth (%)**

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 181

concentration of toxic degradation intermediates and to the toxicity of the mixture that is no

**0 min 60 min 120 min 240 min Clopyralid**

**a b b a b b a b a**

**0 min 60 min 120 min 240 min Clopyralid**

**b b b a b b a b a**

**Control 160 80 40 20 10 Dilution**

**Control 160 80 40 20 10 Dilution**

Fig. 10. Cell growth activity of the serial dilutions of CLP and its photocatalytic degradation intermediates obtained after different irradiation times in a) H-4-II-E and b) MRC-5 cell line. (One-way analysis of variance, compared to the control; **a**: p 0.05, **b**: p 0.01). Results are expressed as mean SD of two independent experiments, each performed in quadruplicate

On the other hand, in the H-4-II-E cell line, the reaction mixture obtained after 60 minutes of irradiation produced significant (p 0.01) stimulation of cell growth compared to control (Figure 10a). The effects that are not concentration-dependent may be explained by the well known concept of hormesis (low dose stimulation and high dose inhibition). Since hormetic effects have been reported in a highly diverse array of biological models, for numerous organs and endpoints and chemical/physical stressors, it is evident that no single mechanism can account for these phenomena. In pharmacology, such dose responses have been studied with the aid of synthetic agonists and antagonists of receptors which mediate hormetic biphasic effects (Calabrese & Baldwin, 2001). A single agonist with differential

and in negative mode (no loss of CO2 or •COOH). Based on the molecular weight (12 units lower than that of CLP, which corresponds to the loss of COO and the introduction of two oxygen atoms), the compound was identified as 3,6-dichloro pyridinediol (the exact positions of hydroxyls could not be determined).

Fig. 9. Tentative pathways for photocatalytic degradation of CLP.

#### **3.9 Cell growth activity**

The cell growth activity of CLP, as well as of the mixture of CLP and its photocatalytic degradation intermediates was evaluated *in vitro* in a panel of two cell lines: H-4-II-E (rat hepatoma) and MRC-5 (human fetal lung) at 160, 80, 40, 20 and 10-fold dilutions that correspond to 6.25, 12.5, 25.0, 50.0 and 100 µM concentrations of CLP at the beginning of experiment (before the irradiation process). The toxicity was evaluated using the SRB assay (Skehan et al., 1990), which determines both specific rate of protein synthesis and cell growth rate.

Cell growth inhibition of CLP reached 5 to 7% in the MRC-5 and H-4-II-E cell line, respectively (Figure 10). The reaction mixture obtained after different irradiation times showed a higher toxicity toward the MRC-5 cell line compared to the parent compound after 120 min of irradiation at 20-fold dilution and after 240 min in the whole concentration range (Figure 10b). A comparison of the evolution of toxicity and degradation kinetics indicates that the toxicity toward the MRC-5 cell line was mildly increased after 120 min of irradiation at higher concentrations, i.e. at 20-fold dilution, and after 240 min in the whole concentration range. This implies that irradiation longer than 120 min contributed to the

and in negative mode (no loss of CO2 or •COOH). Based on the molecular weight (12 units lower than that of CLP, which corresponds to the loss of COO and the introduction of two oxygen atoms), the compound was identified as 3,6-dichloro pyridinediol (the exact

HO Cl

OH

**(1) (2)**

Cl N COOH

Cl (OH)2

**(6)**

The cell growth activity of CLP, as well as of the mixture of CLP and its photocatalytic degradation intermediates was evaluated *in vitro* in a panel of two cell lines: H-4-II-E (rat hepatoma) and MRC-5 (human fetal lung) at 160, 80, 40, 20 and 10-fold dilutions that correspond to 6.25, 12.5, 25.0, 50.0 and 100 µM concentrations of CLP at the beginning of experiment (before the irradiation process). The toxicity was evaluated using the SRB assay (Skehan et al., 1990), which determines both specific rate of protein synthesis and cell

Cell growth inhibition of CLP reached 5 to 7% in the MRC-5 and H-4-II-E cell line, respectively (Figure 10). The reaction mixture obtained after different irradiation times showed a higher toxicity toward the MRC-5 cell line compared to the parent compound after 120 min of irradiation at 20-fold dilution and after 240 min in the whole concentration range (Figure 10b). A comparison of the evolution of toxicity and degradation kinetics indicates that the toxicity toward the MRC-5 cell line was mildly increased after 120 min of irradiation at higher concentrations, i.e. at 20-fold dilution, and after 240 min in the whole concentration range. This implies that irradiation longer than 120 min contributed to the

Cl N

**(4,7)**

HO Cl N COOH

HO H + –

Cl

Cl N COOH

OH

Cl

positions of hydroxyls could not be determined).

*or*

Cl N COOH

**(5)**

OH

Cl N COOH

Cl N OH

**3.9 Cell growth activity** 

growth rate.

**(3)**

Cl

+ HO – Cl

HO N COOH

HO H + –

HO H + –

Fig. 9. Tentative pathways for photocatalytic degradation of CLP.

Cl

Cl

– COOH + HO

concentration of toxic degradation intermediates and to the toxicity of the mixture that is no longer dominated by the parent compound.

Fig. 10. Cell growth activity of the serial dilutions of CLP and its photocatalytic degradation intermediates obtained after different irradiation times in a) H-4-II-E and b) MRC-5 cell line. (One-way analysis of variance, compared to the control; **a**: p 0.05, **b**: p 0.01). Results are expressed as mean SD of two independent experiments, each performed in quadruplicate (*n* = 8).

On the other hand, in the H-4-II-E cell line, the reaction mixture obtained after 60 minutes of irradiation produced significant (p 0.01) stimulation of cell growth compared to control (Figure 10a). The effects that are not concentration-dependent may be explained by the well known concept of hormesis (low dose stimulation and high dose inhibition). Since hormetic effects have been reported in a highly diverse array of biological models, for numerous organs and endpoints and chemical/physical stressors, it is evident that no single mechanism can account for these phenomena. In pharmacology, such dose responses have been studied with the aid of synthetic agonists and antagonists of receptors which mediate hormetic biphasic effects (Calabrese & Baldwin, 2001). A single agonist with differential

Comparative Assessment of the Photocatalytic Efficiency

control that was treated with DDW (data not shown).

of controls and HgCl2.

**3.10 Effect of water type** 

consequence of the presence of HCO

concentration of HCO

water contained more HUM.

(Neppolian et al., 2002). Namely, the addition of HCO

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 183

H-4-II-E cell line appeared to be more sensitive to the presence of catalyst compared to the MRC-5 cell line (Figure 11). In the H-4-II-E cell line, the inhibition of cell growth influenced by TiO2 Wackherr was significantly different compared to control (p < 0.01), even at the 160 dilution (Figure 11), but in all cases was below 10%. Solvent (DDW) was also tested after different irradiation times and it was shown to be nontoxic, i.e. all values were at the level of

The effects of examined samples on the growth of selected cell lines were dependent on the type of cell line, concentration and time of irradiation. It can be concluded that CLP and reaction mixture of CLP and its photocatalytic degradation intermediates effected mildly the cell growth of both cell lines. In the examined concentration range, none of the treatments produced cell growth inhibition higher than 50%, i.e. the IC50 values were not reached, either by the parent compound nor by samples obtained after different irradiation times. All examined samples exhibited lower toxicity in selected cell lines compared to the cytotoxicity

A low level of free oxygen species is necessary for the promotion of cell proliferation (Burdon & Gill, 1993; Wei & Lee, 2002). The redox alterations play a significant role in a signal transduction pathway important for cell growth regulation. It is reasonable to propose that the examined samples obtained using UV irradiation in the presence of O2 and TiO2 Wackherr catalyst might influence the cell redox state, altering the cell proliferation.

Multi-endpoint bioassays that are based on whole cell response in human cell lines are a powerful indicator of metabolic, biochemical and genetic alterations that arise under the influence of evaluated compounds. This study presents an example of a systematic and

Since natural aquatic systems contain dissolved organic matter (DOM) and different ionic species, it can be expected that they may complicate the photodegradation process. It has been reported that higher contents of inorganic and organic matter in tap and river water affect the efficiency of removal by UV/TiO2 process (Buxton et al., 1988). To functionalise a TiO2 water treatment process, the basic understanding of the effect of these inorganic ions on the photocatalytic performance is essential (Crittenden et al., 1996). Due to the zwitter ionic nature of the TiO2 particles, it is also possible that the pH might have a profound effect on the selective inhibition of inorganic ions on the surface of the TiO2 particles (Guillard et al., 2003). In this study, the effect of the matrix on the photocatalytic degradation of CLP was studied on the example of drinking and Danube water. As can be seen from Table 4, the rate of CLP removal from the samples of tap and river water was by about two/three times slower than that from DDW. The observed decrease in the degradation rate can be a

present in Danube water (Table 1), caused a decrease in the rate of photocatalytic degradation compared to that observed in DDW. As can been seen in Table 4, the

**<sup>3</sup>** and HUM in the examined water samples

**<sup>3</sup>** in the examined tap water was somewhat higher, whereas the river

**<sup>3</sup>** and HUM to DDW in the amounts

simple first tier method to assess the toxicity of degradation products.

binding (i.e. high and low receptor affinities) that affects two opposite acting receptors will induce hormetic-like biphasic dose responses in numerous biological systems, as has been shown for dozens of receptor systems. Pollutants, for example, may initiate significant changes in complex receptor systems, and affect biphasic dose responses by inducing changes in the concentrations of endogenous agonists. When such changes occur over a broad dose range, biphasic dose responses typically become manifested (Calabrese, 2005).

The IC50 values of Aspirin®, two well known cytotoxic drugs (Doxorubicin® and Gemcitabine®) and HgCl2 (Table 3), as well as cell growth inhibition of TiO2 Wackherr (Figure 11) were obtained in the same panel of cell lines.


aAspirin ; bDoxorubicin®; cGemcitabine®

Table 3. IC50 values (M) of Aspirin, Doxorubicin, Gemcitabine and HgCl2 in selected cell lines.

Fig. 11. Cell growth activity of different concentrations of TiO2 Wackherr catalyst in the MRC-5 and H-4-II-E cell lines. (One-way analysis of variance, compared to the control; **a**: p0.05, **b**: p 0.01). Results are expressed as mean SD of two independent experiments, each performed in quadruplicate (*n* = 8).

In order to check whether the cell growth activity presented in Figure 10 is a consequence of the presence of CLP and its degradation intermediates only, it was necessary to run a blank test. To this end, aqueous suspension of TiO2 Wackherr (2 mg cm–1 without CPL) was sonicated in the dark for 15 min, as in the case of photodegradation of CLP, filtered through Millipore membrane filter, to apply then the same dilutions from 10 to 160. There were no significant effects (p < 0.01) on the growth of MRC-5 cell line. On the other hand, the H-4-II-E cell line appeared to be more sensitive to the presence of catalyst compared to the MRC-5 cell line (Figure 11). In the H-4-II-E cell line, the inhibition of cell growth influenced by TiO2 Wackherr was significantly different compared to control (p < 0.01), even at the 160 dilution (Figure 11), but in all cases was below 10%. Solvent (DDW) was also tested after different irradiation times and it was shown to be nontoxic, i.e. all values were at the level of control that was treated with DDW (data not shown).

The effects of examined samples on the growth of selected cell lines were dependent on the type of cell line, concentration and time of irradiation. It can be concluded that CLP and reaction mixture of CLP and its photocatalytic degradation intermediates effected mildly the cell growth of both cell lines. In the examined concentration range, none of the treatments produced cell growth inhibition higher than 50%, i.e. the IC50 values were not reached, either by the parent compound nor by samples obtained after different irradiation times. All examined samples exhibited lower toxicity in selected cell lines compared to the cytotoxicity of controls and HgCl2.

A low level of free oxygen species is necessary for the promotion of cell proliferation (Burdon & Gill, 1993; Wei & Lee, 2002). The redox alterations play a significant role in a signal transduction pathway important for cell growth regulation. It is reasonable to propose that the examined samples obtained using UV irradiation in the presence of O2 and TiO2 Wackherr catalyst might influence the cell redox state, altering the cell proliferation.

Multi-endpoint bioassays that are based on whole cell response in human cell lines are a powerful indicator of metabolic, biochemical and genetic alterations that arise under the influence of evaluated compounds. This study presents an example of a systematic and simple first tier method to assess the toxicity of degradation products.

### **3.10 Effect of water type**

182 Herbicides – Properties, Synthesis and Control of Weeds

binding (i.e. high and low receptor affinities) that affects two opposite acting receptors will induce hormetic-like biphasic dose responses in numerous biological systems, as has been shown for dozens of receptor systems. Pollutants, for example, may initiate significant changes in complex receptor systems, and affect biphasic dose responses by inducing changes in the concentrations of endogenous agonists. When such changes occur over a broad dose range, biphasic dose responses typically become manifested (Calabrese, 2005). The IC50 values of Aspirin®, two well known cytotoxic drugs (Doxorubicin® and Gemcitabine®) and HgCl2 (Table 3), as well as cell growth inhibition of TiO2 Wackherr

> H-4-II-E >5551 0.272 0.004 3.189 MRC-5 >5551 0.408 0.384 69.578

Table 3. IC50 values (M) of Aspirin, Doxorubicin, Gemcitabine and HgCl2 in selected cell

**MRC-5 H-4-II-E**

**Control 160 80 40 20 10 Dilution**

Fig. 11. Cell growth activity of different concentrations of TiO2 Wackherr catalyst in the MRC-5 and H-4-II-E cell lines. (One-way analysis of variance, compared to the control; **a**: p0.05, **b**: p 0.01). Results are expressed as mean SD of two independent experiments,

In order to check whether the cell growth activity presented in Figure 10 is a consequence of the presence of CLP and its degradation intermediates only, it was necessary to run a blank test. To this end, aqueous suspension of TiO2 Wackherr (2 mg cm–1 without CPL) was sonicated in the dark for 15 min, as in the case of photodegradation of CLP, filtered through Millipore membrane filter, to apply then the same dilutions from 10 to 160. There were no significant effects (p < 0.01) on the growth of MRC-5 cell line. On the other hand, the

ASPa DOXb GEMc HgCl2

**b a a b b b**

(Figure 11) were obtained in the same panel of cell lines.

aAspirin ; bDoxorubicin®; cGemcitabine®

each performed in quadruplicate (*n* = 8).

lines.

**0**

**20**

**40**

**60**

**Cell growth (%)**

**80**

**100**

**120**

Cell line IC50 (M)

Since natural aquatic systems contain dissolved organic matter (DOM) and different ionic species, it can be expected that they may complicate the photodegradation process. It has been reported that higher contents of inorganic and organic matter in tap and river water affect the efficiency of removal by UV/TiO2 process (Buxton et al., 1988). To functionalise a TiO2 water treatment process, the basic understanding of the effect of these inorganic ions on the photocatalytic performance is essential (Crittenden et al., 1996). Due to the zwitter ionic nature of the TiO2 particles, it is also possible that the pH might have a profound effect on the selective inhibition of inorganic ions on the surface of the TiO2 particles (Guillard et al., 2003). In this study, the effect of the matrix on the photocatalytic degradation of CLP was studied on the example of drinking and Danube water. As can be seen from Table 4, the rate of CLP removal from the samples of tap and river water was by about two/three times slower than that from DDW. The observed decrease in the degradation rate can be a consequence of the presence of HCO **<sup>3</sup>** and HUM in the examined water samples (Neppolian et al., 2002). Namely, the addition of HCO **<sup>3</sup>** and HUM to DDW in the amounts present in Danube water (Table 1), caused a decrease in the rate of photocatalytic degradation compared to that observed in DDW. As can been seen in Table 4, the concentration of HCO **<sup>3</sup>** in the examined tap water was somewhat higher, whereas the river water contained more HUM.

Comparative Assessment of the Photocatalytic Efficiency

suggests a predominant effect of the

2011).

**4. Conclusion** 

involved free

HUM and their high reactivity with

of TiO2 Wackherr in the Removal of Clopyralid from Various Types of Water 185

irradiation (Chu et al., 2009b). Expectedly, the degradation rate decreased after the addition of HUM up to 20 mg L–1 (Figure 13). The behaviour observed in the presence if HUM

Fig. 13. Effect of the concentration of HUM on photodegradation of CLP in DDW. Operation

The results of this study clearly indicate that under the UV irradiation TiO2 Wackherr was more efficient than Degussa P25 in both the process of removal of CLP from water and its mineralisation. The reaction followed the pseudo-first order kinetics. The optimum loading of TiO2 Wackherr was 1.0 mg mL–1 at pH 3.5. The photodegradation rate was dependent on the temperature, and the apparent activation energy was 37.9 kJ mol–1. Along with molecular oxygen, KBrO3 was the most efficient electron acceptor when concerning the degradation of the parent compound, whereas its mineralisation was most efficient in the presence of O2 only. It was found that the presence of ethanol as a scavenger of

inhibited the CLP photodecomposition, suggesting that the reaction mechanism mainly

that six intermediates were formed. The analysis of the intermediate product formed during the photocatalytic degradation could be a useful source of information about the degradation pathways. The rate of photodegradation of CLP in DDW was about two/three times higher than in tap and river waters. The photodegradation rate was dominantly

validates the presented screening methodology of ecotoxicological risk assessment for transformation products, and can be used as a first step in toxicity assessment of degradation products and for prioritisation and planning of more detailed investigations.

influenced by the pH of the medium and the presence of HCO

OH. The LC–DAD, and LC–ESI–MS/MS monitoring of the process showed

conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH~7.0.

OH radical inhibition, due to the complex structure of

OH radicals (Basfar et al., 2005; Prados-Joya et al.,

OH

**<sup>3</sup>** and DOM. Our work


Table 4. The influence of water type on the degradation rate (*R*) of CLP determined after 120 min of irradiation. Operation conditions: *c*(CLP) 0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH ~7.0.

In the literature, the inhibition of photocatalytic properties in the presence of ions is often explained by the scavenging of OH radical by ions. Of ionic species, HCO <sup>3</sup> can especially, inhibit the degradation rate due to the high rate constant *k′* of its reaction with OH (8.5 × 106 M–1 s–1) (Buxton et al., 1988). Because of that we focused our attention on the influence of different concentrations of this ion on the photocatalytic degradation (Figure 12). Expectedly, an inhibition of CLP degradation was observed after adding HCO3 to DDW up to about 285 mg L–1.

Fig. 12. Effect of the concentration of HCO3 on photodegradation of CLP in DDW. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH~7.0.

The effect of HUM can be explained by the reaction with OH, which lowers the availability of the latter for the reaction with CLP. Moreover, the actually available UV radiation reduces because some organic matters (especially aromatic compounds) absorb strongly UV irradiation (Chu et al., 2009b). Expectedly, the degradation rate decreased after the addition of HUM up to 20 mg L–1 (Figure 13). The behaviour observed in the presence if HUM suggests a predominant effect of the OH radical inhibition, due to the complex structure of HUM and their high reactivity with OH radicals (Basfar et al., 2005; Prados-Joya et al., 2011).

Fig. 13. Effect of the concentration of HUM on photodegradation of CLP in DDW. Operation conditions: *c*(CLP)0 = 1.0 mM, TiO2 Wackherr = 2.0 mg mL–1, *t* = 25 oC, pH~7.0.
