Synergy of ozonation and photocatalysis to mineralize low concentration 2,4-dichlorophenoxyacetic acid in aqueous solution

Chemosphere 66 (2007) 1610–1617 www.elsevier.com/locate/chemosphere

Synergy of ozonation and photocatalysis to mineralize low concentration 2,4-dichlorophenoxyacetic acid in aqueous solution Rabindra Raj Giri a,*, Hiroaki Ozaki b, Takehiro Ishida b, Ryohei Takanami a, Shogo Taniguchi a b

a New Industrial R&D Center, Osaka Sangyo University, 3-1-1 Nakagaito, Daito City, 574-8530 Osaka, Japan Department of Civil Engineering, Osaka Sangyo University, 3-1-1 Nakagaito, Daito City, 574-8530 Osaka, Japan

Received 18 March 2006; received in revised form 5 August 2006; accepted 7 August 2006 Available online 18 September 2006

Abstract Concentration of 2,4-dichlorophenoxyacetic acid (2,4-D) may affect its degradation kinetics in advanced oxidation systems, and combinations of two or more systems can be more effective for its mineralization at low concentration levels. Degradations and mineralizations of 0.045 mM 2,4-D using O3, O3/UV, UV/TiO2 and O3/UV/TiO2 systems were compared, and influence of reaction temperature on the mineralization in O3/UV/TiO2 system was investigated. 2,4-D degradations by O3, O3/UV and UV/TiO2 systems were similar to the results of earlier investigations with higher 2,4-D concentrations. The degradations and total organic carbon (TOC) removals in the four systems were well described by the first-order reaction kinetics. The degradation and removal were greatly enhanced in O3/UV/TiO2 system, and further enhancements were observed with larger O3 supplies. The enhancements were attributed to hydroxyl radical (OH) generation from more than one reaction pathway. The degradation and removal in O3/UV/TiO2 system were very efficient with reaction temperature fixed at 20 C. It was suspected that reaction temperature might have influenced OH generation in the system, which needs further attention.  2006 Elsevier Ltd. All rights reserved. Keywords: Aromatic intermediates; Dechlorination; Degradation; Rate constant; TOC removal

1. Introduction Advanced oxidation processes (AOPs) are promising methods, and widely employed for degrading toxic recalcitrant organic pollutants. Degradation of organics in ozonation takes place by direct reaction with O3 molecules (Brillas et al., 2003; Chu and Ching, 2003; Li et al., 2005). However, this is an inefficient method for a complete mineralization due to its selective reaction with the organics (Brillas et al., 2003). OH is generated in ozonation at higher pH conditions as well as with UV radiation by O3 decomposition (Brillas et al., 2003). Unlike ozone molecules, OH is non-selective to organic molecules and pos*

Corresponding author. Tel.: +81 72 875 3001x7824; fax: +81 72 875 3076. E-mail address: [email protected] (R.R. Giri). 0045-6535/$ - see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2006.08.007

sesses the second highest oxidizing power. Catalyzed ozonation methods (e.g. O3/H2O2, O3/Fe2+, O3/Fe3+) produce more OH resulting in fast mineralization. Heterogeneous photocatalysis with TiO2 and UV is a rapidly emerging AOP. OH is considered as the sole oxidizing agent in photocatalysis, while other species also are involved in ozonation and catalyzed ozonation. However, photocatalysis alone has not been efficient enough for rapid mineralization of the organics (Farre et al., 2005; Li et al., 2005). Combinations of two or more AOPs have been the focus of current researches for efficient mineralization of organics due to synergistic effects of different oxidizing species in the combined systems. Photocatalytic ozonation is emerging as a promising AOP recently (Beltran et al., 2005; Farre et al., 2005; Li et al., 2005) among several combinations due to development of new TiO2 photocatalysts, although,

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617

enhanced organic mineralization by this method was pointed out earlier (Muller et al., 1998). The herbicide 2,4-D is extensively used as a plant growth regulator for agricultural and non-agricultural purposes. However, it has been proved to be toxic to human and animals. The World Health Organization has recommended 70 ppb as its maximum permissible concentration in drinking water (Yasman et al., 2004). Therefore, its presence in potential water sources for human and animal consumption is highly objectionable. Most of the investigations on 2,4-D degradation using AOPs done so far focused on its high concentrations ranging from ‘‘0.45 mM to 1.8 mM’’ (Muller et al., 1998; Piera et al., 2000; Brillas et al., 2003; Drzewicz et al., 2004; Singh and Muneer, 2004). The pollution levels in the environment have been significantly reduced due to advancement in treatment technologies and ever increasing stringent environmental regulations. Since degradation kinetic, mineralization efficiency and methods to be employed to treat waters polluted with organic chemicals, in general, depend on their initial concentrations, investigations on improved AOPs for efficient mineralization of 2,4-D at low concentration levels are essential. This paper aimed to compare mineralization of about 0.045 mM 2,4-D in aqueous solution in terms of its degradation rate, formation of aromatic intermediates, TOC removal and dechlorination efficiency using ozonation (O3), ozonation combined with UV (O3/ UV), photocatalysis (UV/TiO2) and ozonation combined with photocatalysis (O3/UV/TiO2), and evaluate synergistic effects of ozonation and photocatalysis on the mineralization. Influence of reaction temperature on 2,4-D mineralization in O3/UV/TiO2 system was also investigated. 2. Materials and methods 2.1. Materials A UV tube lamp (10 W, 254 nm, UVL10DL-12, SEN Lights Corporation, Japan) was enclosed vertically inside a high quality glass jacket fixed at center of the top of a Pyrex glass reactor (ID: 9.8 cm, h: 20 cm, 1.5 l). Ports at the top were used for sampling, ozone inflow, gas outflow and temperature monitoring. Tap water was circulated through the jacket to minimize solution temperature change. Ozone was produced by passing air (21% O2 + 79% N2) through ozone generator (Nippon Ozone Co. Ltd.). Carrier gas flow from the generator to reactor was regulated with a flow meter, while ozone concentration in the gas was continuously monitored using an ozone measuring device (EG-600, EBARA JITSUGYO Co. Ltd., Japan). Headspace gas in the reactor was continuously pumped and passed through a packed tower to absorb gas phase residual ozone. TiO2 silica gel powder (PSB-P200, ‘‘1.7–4.0 mm’’, ‘‘10– 11%’’ TiO2 by wt.) from Photocatalytic Materials Inc., Japan, was used in photocatalysis and photocatalytic ozonation experiments. Standards of all the chemicals including

1611

2,4-D were obtained from Wako Pure Chemicals Ltd. Na2SO3 solution (1.0 g l 1) was used to quench dissolved residual ozone in samples. KIO3, KI, H2SO4 and NaOH were used to prepare dissolved ozone standards in water (Merck, Germany). Spectroquant Picco Colorimeter test kit (Merck, Germany) was used to measure dissolved residual ozone. All other chemicals were of HPLC grade and obtained from Wako Pure Chemicals Ltd. Neocool Circulator (CF700, Yamato Scientific Co. Ltd., Japan) was used for maintaining solution temperature. 2.2. Experimental About 1.3 l of 0.045 mM 2,4-D solution was poured into the reactor. The ozone generator, carrier gas flow meter, ozone concentration monitoring device, reactor, pump and absorption tower were connected in series. The solution was mixed using magnetic stirrer with the same speed in all the experiments to avoid any possible effect of varying speeds. Samples were taken at 10 min interval, while residual ozone was measured at 20 min interval. Experiments for 2,4-D degradation in O3, O3/UV and O3/UV/TiO2 systems were carried out. The carrier gas flow rate was fixed at 1.0 l min 1 in all the cases, while 0.5, 1.5 and 3.0 mg l 1 min 1 O3 concentrations were used in each of the three cases. O3 concentration in carrier gas was regulated by varying electric voltage in ozone generator. A set of experiments with UV/TiO2 system was also conducted for comparison. Concentration of TiO2 catalyst in UV/ TiO2 and O3/UV/TiO2 systems was 5.0 g l 1. Degradation experiments at 10, 20 and 30 C solution temperatures in O3/UV/TiO2 system were also carried out. All the experiments were carried out for 80 min with sampling at every 10 min without prior adjustments in initial pH values of 0.045 mM 2,4-D solution. Wavelengths and intensities of the UV irradiation with and without UV jacket were also measured. The UV source emitted solely 254 nm wavelength, while more than 94% of the irradiation was transmitted into the reactor (the intensities with and without jacket were 20.8 and 22.2 lW cm 2 nm 1, respectively). 2.3. Analyses The samples were analyzed for remaining 2,4-D concentration and aromatic intermediates using HPLC with UV detector (Model: D-7000, Hitachi, Japan). Inertsil ODS-3 column (150 mm · 3 mm ID · 5 lm) with 35 C oven temperature, 20 ll sample volume and CH3CN:H2O:CH3COOH = 50:49:1 (v/v) mobile phase at 0.4 ml min 1 flow rate was used in the analysis. TOC was measured using TOC analyzer (TOC-VCSH, Shimadzu, Japan). Chloride ion concentrations were measured using Hitachi L-7000 ion chromatograph with conductivity detector (Model: L7000, Hitachi, Japan). GL-IC-A25 column (150 mm · 4.6 mm ID) with 40 C oven temperature and 4.0 mM Na2CO3 mobile phase at 1.0 ml min 1 flow rate was

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617

3. Results and discussion 3.1. 2,4-D degradation and rate constant Concentration profiles of 2,4-D in O3, O3/UV, UV/TiO2 and O3/UV/TiO2 oxidation systems are presented in Fig. 1. The increased 2,4-D degradations with larger ozone supply rates were obvious. The degradations were enhanced in O3/UV system due to photolysis and oxidation with OH in addition to direct ozonation. Role of photolysis on 2,4-D degradation appeared larger at smaller O3 supply rates. Although, direct ozonation is dominant in O3/UV system with acidic pH conditions (Chu and Ching, 2003) and solution pH values in the system were <4.6 throughout the experiments, formation of OH could not be ruled out (Yao et al., 1998 cited in Masten et al., 1996; Beltran et al., 2005). Since UV absorption by O3 in O3/UV system increases with larger O3 concentrations possibly generating more OH, the reduced role of photolysis on 2,4-D degradation with larger O3 supplies was obvious. Degradation of 2,4-D was further enhanced in O3/UV/TiO2 system. For instance, half-life periods of 2,4-D in O3, O3/UV and O3/UV/TiO2 systems with 1.5 and 3.0 mg l 1 min 1 O3 supply rates were about 45, 28, 20 and 18, 17, 14 min, respectively. The enhanced 2,4-D degradation in O3/UV/ TiO2 system can be attributed to additional source of  OH due to action of O3 on TiO2 (Agustina et al., 2005; Beltran et al., 2005). Photocatalysis alone did not appear very effective for 2,4-D degradation, where the half-life period was more than 80 min. Photocatalytic degradation of 2,4-D with TiO2 is enhanced at low pH, and the degrada-

tion is very effective with pH between 2 and 3 (Singh and Muneer, 2004; Terashima et al., 2006). Initial pH of 2,4D solutions with TiO2 in suspension were relatively larger (‘‘5.1–5.2’’) than those of without TiO2 (<4.6) possibly resulting in reduced 2,4-D degradation in UV/TiO2 system in this investigation. Being a weak acid, dissociation of 2,4D in water is very small (<1.0%), which could be favorable for its oxidation. Apparent 2,4-D degradation rate constant (k) was calculated based on the first-order reaction kinetic (inset of Fig. 2) with R2 values more than 0.98 in all the cases. The k values for the cases considered were between ‘‘0.11 h 1 and 5.52 h 1’’, which were about 0.43 h 1 for UV/TiO2 system. Ozone supply rate did have a major influence on k value (Fig. 2) especially in O3 system, where direct ozonation of 2,4-D was presumably dominant. For instance, k value in O3 system with 1.5 and 3.0 mg l 1 min 1 O3 supplies were, respectively, about 8.0 and 27.2 times of the value with 0.5 mg l 1 min 1 O3 supply. Although, k values in O3/UV and O3/UV/TiO2

6.0 5.0 4.0

3.0 -ln(2,4-D/2,4-Do)

employed in the measurement. Residual aqueous phase ozone was measured using photometric DPD (diphenylendiamine)-based method at 528 nm UV wavelength. Wavelengths and intensities of UV irradiation from the tube light were checked using UV spectroradiometer (USR-40D, Kansai Scientific Instruments Co. Ltd., Japan).

k for 2,4-D (h-1)

1612

2.0

1.0

0.0 0

20

40

60

80

3.0 2.0 O3 O3/UV

1.0

O3/UV/TiO2

0.0 0.0

0.5

1.0 1.5 2.0 O3 flow rate (mg l-1min-1)

O3(0.5) O3(0.5)/UV

2,4-D (C/Co)

0.8

O3(0.5)/UV/TiO2 O3(1.5)

0.6

O3(1.5)/UV O3(1.5)/UV/TiO2 O3(3.0)

0.4

O3(3.0)/UV O3(3.0)/UV/TiO2

0.2

O3(0.0)/UV O3(0.0)/UV/TiO2

0.0 10

20

30

40

50

3.0

Fig. 2. Effect of O3 supply on apparent 2,4-D degradation rate constant (k).

1.0

0

2.5

60

70

80

Time (min) Fig. 1. Relative 2,4-D concentration profiles in different oxidation systems.

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617

systems, respectively, were larger than the corresponding values in O3 system, the rates of increase with larger O3 supplies were smaller than those of O3 system. As an example, k values in O3/UV and O3/UV/TiO2 systems with 1.5 and 3.0 mg l 1 min 1 O3 supplies were about 3.9, 3.1 and 8.6, 5.6 times of the corresponding values with 0.5 mg l 1 min 1 O3 supply, respectively. Rate constants of O3/UV and O3/UV/TiO2 systems for a fixed O3 supply were, respectively, higher compared to those of O3 system, which demonstrated synergistic effects of direct ozonation, photolysis and OH oxidation of 2,4-D. The k values of O3/ UV/TiO2 system for 0.5, 1.5 and 3.0 mg l 1 min 1 O3 supplies were about 1.5, 1.0 and 0.7 times of the corresponding combined (summation) k values of O3 and O3/UV systems, respectively. Similarly, the values were about 1.7, 2.0 and 1.4 times of the corresponding combined k values of O3 and UV/TiO2 systems, respectively. Although, the apparent k values in O3/UV/TiO2 system increased with larger O3 supply, UV distributions among the absorbing species (e.g., O3, 2,4-D and TiO2) might have changed considerably, which were not known. Presumably OH generation due to O3 decomposition might have dominated in O3/ UV/TiO2 system with larger O3 supplies resulting to increased k values. Reaction temperature plays important role in AOPs (Lee et al., 2003). 2,4-D solution temperatures at the beginning of degradation experiments with the three systems using UV illumination were between 18 C and 21 C, which increased by about 4 C during the experiments. 2,4-D degradations with O3/UV and O3/UV/TiO2 systems at three fixed solution temperatures (10, 20 and 30 C) were also investigated to understand temperature effect on the degradation. The degradation appeared to increase slightly at higher reaction temperatures in O3/UV system (not shown), but the effect was more distinct in O3/UV/TiO2 system (Fig. 3). 2,4-D degraded faster at 20 C, while the differences between 10 C and 30 C were small among

1.0 -ln(2,4-D/2,4-Do)

2.8

2,4-D (C/Co)

0.8

0.6

1.9

0.9

0.0

0

20

40

60

80

O3 (1.5 mg l-1min-1),

0.4

TiO2 (5.0 g l-1)

10 ºC 0.2

20 ºC 30 ºC

0.0

0

10

20

30

40 50 Time (min)

60

70

80

Fig. 3. Effect of reaction temperature on 2,4-D degradation in O3/UV/ TiO2 system.

1613

the three cases with 1.5 mg l 1 min 1 O3 supply, 5.0 g l 1 TiO2. The half-life period of 2,4-D at 20 C appeared to be slightly smaller (16 min) than the corresponding value without temperature control (20 min). Apparent k values of 2,4-D degradation at 10, 20 and 30 C were 2.65, 3.35 and 2.25 h 1, respectively (inset, Fig. 3). The value at 20 C in O3/UV/TiO2 system was larger than the corresponding value (3.01 h 1) without temperature control (Fig. 1). This result was consistent with our earlier result of photocatalytic oxidation of 2,4-D with TiO2 fiber catalyst (Terashima et al., 2006), where 2,4-D degraded faster with solution temperature fixed at 20 C among the three cases considered (20, 30 and 40 C). Half-life of ozone both in gas and aqueous phases, and its solubility in water are greatly reduced at higher temperatures. However, the enhanced 2,4-D degradation at 20 C both in UV/TiO2 and O3/UV/TiO2 systems may not suggest a major influence of ozone solubility on the 2,4-D degradation in O3/UV/TiO2 system. Owing to similar solution pH values and their variations during oxidation, any influence of pH on 2,4-D degradation in O3/UV/TiO2 system with fixed reaction temperatures may be ruled out, and the reason behind enhanced 2,4-D degradation at 20 C is unknown. 3.2. Aromatic intermediates Number and characteristics of intermediates generated during 2,4-D oxidation have direct impact on its mineralization. Only benzyl alcohol (retention time (RT)  2.90 min) and phenol (RT  3.46 min) were identified, while all other detected peaks in HPLC analysis (Fig. 4) were not known. Therefore, their identification using LCMS/MS and GCMS/MS is under consideration as the next step. All the detected aromatic byproducts in O3 system (e.g., Fig. 4(i)) except the model pollutant (2,4-D) were unknown, while one with RT  3.91 min was a major peak. Five intermediates were observed in O3/UV system with the three O3 supplies (Fig. 4(ii)), and the higher number of intermediates may be attributed to additional 2,4-D oxidation by photolysis. Phenol and benzyl alcohol were two known intermediates in O3/UV system, while phenol and two other unknowns (RT  5.14 and 10.0 min) were major byproducts. Only two intermediates were observed in UV/TiO2 system and 2,4-dichlorophenol (2,4-DCP) was one of them, which was not observed in O3 and O3/ UV systems. The number of detected intermediates in UV/TiO2 system was fewer than those in our earlier investigation using TiO2 fiber catalyst (Terashima et al., 2006) possibly due to characteristics of the catalysts. The intermediates in O3/UV/TiO2 system (Fig. 4(iii)) varied significantly with the three O3 supplies. Phenol and benzyl alcohol were two known intermediates among the five numbers detected in this system. Benzyl alcohol and an unknown (RT  4.01 min) were two major intermediates (Fig. 4(iii)), while small amounts of 2,4-DCP were also detected with smaller O3 supplies. 2,4-DCP, which was always a major intermediate in photocatalytic oxidation

1614

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617 -1

(i) O3(1.5 mg l-1 min-1)

Intensity (mV)

0.5

2,4-D (8.07 min)

3

-70 min

0.81

/UV-40 min unknown (8.44 min)

unknown (3.91 min)

0.4

-1

(ii) O3 (1.5 mg l min )

2,4-D (7.91 min)

unknown (5.14 min)

Benzyl alcohol (2.90 min) Phenol (3.46 min)

unknown (10.00 min)

unknown (9.96 min)

unknown (5.32 min) 0.2

0.49

unknown (4.26 min)

0.32

0.0 0.5 -1

-1

3

0.28

Benzyl alcohol (2.77 min)

(iii) O3(3.0 mg l min )/UV/TiO2

2,4-D (7.44 min)

2

-1

(5.0 g l )-20 min Intensity (mV)

0.65

Phenol (3.15 min)

2,4-D(8.30 min)

Benzyl alcohol (3.48 min)

0.3

unknown (4.04 min)

Phenol (3.48 min)

O3 (1.5 mg l-1 min-1)/ UV/TiO 2 (5.0 g l )

0.07

unknown (5.07 min) 0.0

3

4

5

6

7

0.14

-1

unknown (4.01 min)

0.2

0.21

(iv) T = 20 o C-40 min,

0.00

8

9

10

11

3

4

5

Time (min)

6

7

8

9

10

11

-0.07

Time (min)

Fig. 4. Typical HPLC chromatograms of 2,4-D samples.

of 2,4-D in our earlier investigation (Terashima et al., 2006), can be degraded significantly faster by ozonation than by photocatalysis (Muller et al., 1998). The absence of 2,4-DCP as an intermediate in O3, O3/UV and O3/ UV/TiO2 systems may be attributed to its fast decomposition with ozone. Totally seven intermediates including phenol and benzyl alcohol were detected in O3/UV/TiO2 system at three fixed temperatures (e.g. Fig. 4(iv)). The numbers of intermediates at 10, 20 and 30 C temperatures were 5, 7 and 3, respectively. Benzyl alcohol was a major intermediate and detected throughout the experiments at 10 C and 20 C, while an unknown with RT  4.33 min was a major intermediate at 30 C case. UV distribution among O3, 2,4-D

and TiO2, and decreased aqueous solubility and half-life of O3 with increasing solution temperature, which were not quantified in this investigation, might have influenced generation of oxidizing species and hence 2,4-D degradation pathways in O3/UV/TiO2 system. 3.3. TOC removal Relative TOC profiles in 2,4-D degradations in O3, O3/ UV, UV/TiO2 and O3/UV/TiO2 systems are shown in Fig. 5. TOC removals were significantly slower than the corresponding 2,4-D degradations (Fig. 1) although, the trends in both the cases were similar. Although, 2,4-D degradations in O3/UV/TiO2 system were comparable to

1.0

O3(0.5) O3(0.5)/UV

0.8

TOC (C/Co)

O3(0.5)/UV/TiO2 O3(1.5)

0.6

O3(1.5)/UV O3(1.5)/UV/TiO2 0.4

O3(3.0) O3(3.0)/UV

0.2

O3(3.0)/UV/TiO2 O3(0.0)/UV

0.0 0

10

20

30

40

50

60

70

80

O3(0.0)/UV/TiO2

Time (min) Fig. 5. Relative TOC concentration profiles in different oxidation systems.

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617

those in O3 and O3/UV systems with higher O3 supplies, TOC removals in the latter cases were significantly smaller. It is evident from Fig. 5 that O3/UV/TiO2 system was very effective for 2,4-D mineralization. TOC removal in O3/UV/ TiO2 system with the lowest O3 supply was higher than those of all the cases in O3 and O3/UV systems. The mineralization was greatly enhanced with higher O3 supplies in the system, which was consistent with earlier investigation results (Agustina et al., 2005). A complete TOC removal was achieved with 3.0 mg l 1 min 1 O3 supply in less than 80 min, while the removal efficiencies with 1.5 and 0.5 mg l 1 min 1 supplies at 80 min were about 65% and 40% only, respectively. However, the efficiencies for all other cases considered were less than 35%. The first-order apparent TOC removal rate constants (k) with R2 values more than 0.98 in all the cases varied from ‘‘0.03 h 1 to 2.52 h 1’’ (inset, Fig. 6), which were very small than the corresponding values for 2,4-D degradation. The k values in O3/UV/TiO2 system were about ‘‘25%–46%’’ of the corresponding k values for 2,4-D degradation, while they were less than 15% in O3 and O3/UV systems. O3 supply highly influenced k values in O3/UV/TiO2 system, while the effect in O3 and O3/UV systems was insignificant (Fig. 6). The rate constants for O3, O3/UV and O3/UV/ TiO2 systems with 1.5 and 3.0 mg l 1 min 1 O3 supplies were about 2.0, 1.9, 2.2 and 5.6, 4.3, 7.5 times, respectively, of the corresponding values with 0.5 mg l 1 min 1 O3 supply. The k values for O3/UV and O3/UV/TiO2 systems with 0.5, 1.5 and 3.0 mg l 1 min 1 O3 supplies were, respectively, about 1.5, 1.4, 1.2 and 9.6, 10.3, 12.7 times of the corresponding values in O3 system. 2,4-D mineralizations in those systems were presumably guided mainly by characteristics of intermediates and relative concentrations of oxidizing species. Vary small TOC reductions in O3 and O3/ UV systems may be attributed to selective reactions of O3 with organic molecules, negligibly small contribution of direct photolysis to oxidation, and presence of relatively

2.5

k for TOC (h-1)

2.0

-ln(TOC/TOCo)

3.0 2.5

1.7

0.8

0.0

1.5

0

20

40

60

80

O3

1.0

O3/UV O3/UV/TiO2

0.5 0.0 0.0

0.5

1.0

1.5

TOC (20 ºC) TOC (30 ºC) Chloride (10 ºC) 60

Chloride (20 ºC) Chloride (30 ºC) -ln(TOC/TOCo)

1.6

O3 = 1.5 mg l-1 min -1

1.2 0.8

TiO 2 = 5.0 g l-1

0.4 0.0

0

20

40

60

80

0 0

10

20

30

40

3.0

resistive intermediates like phenol. However, formation of 2,4-DCP (a more resistive intermediate), and possibly limited OH generation could have resulted to small TOC removals in UV/TiO2 system. As discussed elsewhere (Beltran et al., 2005), OH is generated from three parallel reactions (actions of UV and O3 on TiO2, and O3 decomposition) in O3/UV/TiO2 system. Therefore, the highest TOC removals in the system can be attributed to increased  OH generation and synergistic effect of several oxidizing species to decompose varieties of intermediates more effectively. Influence of reaction temperature on TOC removal was more distinct (Fig. 7) than on 2,4-D degradation in O3/UV/ TiO2 system. TOC removal was the highest at 20 C, which was consistent with 2,4-D degradation. Then, it gradually decreased at 30 C and 10 C cases. A distinct difference in TOC removal was observed between 30 C and 10 C, while 2,4-D degradations were almost similar until 40 min, and it appeared to be faster at 10 C later. Apparent TOC removal rate constants (k) with the three temperatures in increasing order were 0.66 h 1, 1.11 h 1 and

80

20

2.5

Fig. 6. Effect of O3 supply on apparent TOC removal rate constant (k).

TOC (10 ºC)

40

2.0

O3 flow rate (mg l-1min-1)

100

Remaining TOC & dechlorination (%)

1615

50

60

70

80

Time (min) Fig. 7. Effect of reaction temperature on TOC removal and dechlorination in O3/UV/TiO2 system.

1616

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617

0.88 h 1, respectively. The values at 10 C and 30 C were about 59% and 78%, respectively, of the value at 20 C. The corresponding k value in the system without temperature control was 0.78 h 1, which is about 118%, 70% and 113% of the values at 10, 20 and 30 C, respectively. This result is consistent with our earlier investigation result on photocatalysis of 2,4-D with TiO2 fiber catalyst (Terashima et al., 2006), in which the highest TOC removal was observed at 20 C, and it gradually decreased at 30 C and 40 C. OH may be considered as major oxidizing specie in O3/UV/TiO2 system, because it is generated from more than one parallel path (Beltran et al., 2005). The fast 2,4-D mineralization at 20 C among the cases considered may indicate that reaction temperature might have affected the use of OH for oxidizing 2,4-D and its degradation byproducts, which needs further attention. 3.4. Dechlorination Chlorine content in hazardous organics is considered as a measure of their toxicity. Dechlorination in 2,4-D oxidation was evaluated as the ratio of measured free chloride ion concentration in a sample to its calculated concentration in 2,4-D solution before oxidation. Dechlorination profiles in O3, O3/UV, UV/TiO2 and O3/UV/TiO2 were more or less similar to 2,4-D degradation profiles (Fig. 1). O3 supply highly influenced dechlorination, which increased with increasing the supply. Final dechlorinations in O3, O3/UV and O3/UV/TiO2 with 0.5, 1.5 and 3.0 mg l 1 min 1 O3 supplies were about 15%, 50%, 86%; 31%, 69%, 90% and 66%, 89%, 99%, respectively. Additional effects of UV irradiation and photocatalysis were distinctly reflected on dechlorinations with a fixed O3 supply in O3/UV and O3/UV/TiO2 systems. Dechlorinations in O3, O3/UV and O3/UV/TiO2 systems with 0.5 mg l 1 min 1 O3 supply were about 15%, 31% and 66%, respectively. The value in O3/UV/TiO2 system was about 4.4 and 2.1 times of the values in O3 and O3/UV systems, respectively. Almost a complete dechlorination was achieved in O3/UV/TiO2 system with the highest O3 supply. The values for photocatalysis, and UV irradiation only were about 23% and 9%, respectively. The results did indicate that O3/UV/TiO2 system was very effective for dechlorination than other systems considered. The enhanced dechlorinations in O3/UV/TiO2 system may be attributed to generation of OH from more than one pathway (Beltran et al., 2005). The increased dechlorinations in the system with larger O3 supplies may also be attributed to increased O3 decomposition and its reaction with TiO2 resulting in higher OH generations (Beltran et al., 2005). Since OH possesses the second highest oxidizing power after positively charged hole on TiO2 (Munter, 2001),  OH generation and its effective utilization for oxidation might have affected dechlorination too. Reaction temperature did influence dechlorination in O3/UV/TiO2 system (Fig. 7). The highest dechlorination (99%) was achieved at 20 C, while the values at 10 C

and 30 C were about 81% and 86%, respectively. Dechlorination at 20 C increased steadily until the end of the experiment. The values at 10 C and 30 C were almost similar until 40 min. But, it slightly increased at 10 C case after 40 min and then started to decrease after 60 min until the end. The steady and biggest chloride ion concentrations at 20 C, smaller dechlorinations at two other temperatures and sudden decrease in dechlorination after 60 min at 10 C (Fig. 7) may have a significance in 2,4-D mineralization. The enhanced 2,4-D degradation, TOC removal and highest dechlorination at 20 C may indicate significance of this particular reaction temperature in O3/UV/TiO2 system. Reaction temperature might have influenced the oxidation process in two possible ways: generation of OH and its effective utilization in the oxidation. Although, free chloride ions are good OH scavengers (Buston et al., 1988; Farre et al., 2005), the effect is significant at high concentrations only (Farre et al., 2005 cited Kiwi et al., 2000). Therefore, influence of reaction temperature on OH scavenging action of chloride ions in this case appeared to be less significant. Thus, it becomes apparent that generations of positive holes and OH might have been significantly enhanced at 20 C case, which needs more attention. 4. Conclusion Oxidative degradations of 0.045 mM 2,4-D in aqueous solution with O3, O3/UV and UV/TiO2 systems were similar to the results of earlier investigations carried out with high 2,4-D concentrations (‘‘0.45–1.80 mM’’). Decays of 2,4-D and TOC in O3, O3/UV, UV/TiO2 and O3/UV/ TiO2 systems were well described by the first-order reaction kinetic. Removals of 2,4-D and TOC were greatly enhanced in O3/UV/TiO2 system, and further enhancement was observed with larger O3 supplies. Apparent 2,4-D removal rate constant (k) values in O3/UV/TiO2 system were ‘‘1.4–1.7’’ times larger than summation of the corresponding k values in O3 and UV/TiO2 systems, while they were ‘‘1.2–5.8’’ times larger for TOC. The enhancements were attributed to OH generation from more than one reaction pathway in O3/UV/TiO2 system. Removals of 2,4-D and TOC in O3/UV/TiO2 system with reaction temperature fixed at 20 C were very efficient. Reaction temperature was suspected to play a major role on OH generation in the system, which needs further attention. References Agustina, T.E., Ang, H.M., Vareek, V.K., 2005. A review of synergistic effect of photocatalysis and ozonation on wastewater treatment. J. Photochem. Photobiol. C 6, 264–273. Beltran, F.J., Rivas, F.J., Gimeno, O., 2005. Comparison between photocatalytic ozonation and other oxidation processes for the removal of phenols from water. J. Chem. Technol. Biotechnol. 80, 973–984. Brillas, E., Calpe, J.C., Cabot, P., 2003. Degradation of the herbicide 2,4dichlorophenoxyacetic acid by ozonation catalyzed with Fe2+ and UVA light. Appl. Catal. B Environ. 46, 381–391.

R.R. Giri et al. / Chemosphere 66 (2007) 1610–1617 Buston, G.V., Greenstock, C.L., Helman, W.P., Rose, A.B., 1988. Critical review of data constants for reaction of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solutions. J. Phys. Chem. Ref. Data 17, 513–886. Chu, W., Ching, M.H., 2003. Modeling the ozonation of 2,4-dichlorophenoxyacetic acid through a kinetic approach. Water Res. 37, 39–46. Drzewicz, P., Trojanowicz, M., Zona, R., Solar, S., Gehringer, P., 2004. Decomposition of 2,4-dichlorophenoxyacetic acid by ozonation, ionizing radiation as well as ozonation combined with ionizing radiation. Radiat. Phys. Chem. 69, 281–287. Farre, M.J., Franch, M.I., Malato, S., Ayllon, J.A., Pearl, J., 2005. Degradation of some biorecalcitrant pesticides by homogeneous and heterogeneous photocatalytic ozonation. Chemosphere 58, 1127–1133. Kiwi, J., Lopez, A., Nadtochenko, V., 2000. Mechanism and kinetics of the OH-radical intervention during Fenton oxidation in the presence of a significant amount of radical scavenger (Cl ). Environ. Sci. Technol. 34, 2162–2168. Lee, Y., Lee, C., Yoon, J., 2003. High temperature dependence of 2,4-dichlorophenoxyacetic acid degradation by Fe3+/H2O2 system. Chemosphere 51, 963–971. Li, L., Zhu, W., Chen, L., Zhang, P., Chen, Z., 2005. Photocatalytic ozonation of dibutyl phthalate over TiO2 film. J. Photochem. Photobiol. A 175, 171–177. Munter, R., 2001. Advanced oxidation processes: current status and prospects. P. Estonian Acad. Sci.: Chem. 50, 59–80.

1617

Masten, S.J., Shu, M., Galbraith, M.J., Davies, S.H., 1996. Oxidation of chlorinated benzenes using advanced oxidation processes. Hazard. Waste Hazard. Mater. 13, 265–282. Muller, T.S., Sun, Z., Kumar M.P., G., Itoh, K., Murabayashi, M., 1998. The combination of photocatalysis and ozonolysis as a new approach for cleaning 2,4-dichlorophenoxyacetic acid polluted water. Chemosphere 36, 2043–2055. Piera, E., Calpe, J.C., Brillas, E., Domenech, X., Pearl, J., 2000. 2,4Dichlorophenoxyacetic acid degradation by catalyzed ozonation: TiO2/UVA/O3 and Fe(II)/UVA/O3 systems. Appl. Catal. B Environ. 27, 169–177. Singh, H.K., Muneer, M., 2004. Photodegradation of a herbicide derivative, 2,4-dichlorophenoxyacetic acid in aqueous suspensions of titanium dioxide. Res. Chem. Intermed. 30, 317–329. Terashima, Y., Ozaki, H., Giri, R.R., Tano, T., Nakatsuji, S., Takanami, R., Taniguchi, S., 2006. Photocatalytic oxidation of low concentration 2,4-D solution with new TiO2 fiber catalyst in a continuous flow reactor. IWA Conference, Beijing, in press. Yao, J.J., Huang, Z.H., Masten, J., 1998. The ozonation of pyrine: pathway and product identification. Water Res. 32, 3001– 3012. Yasman, Y., Bulatov, V., Girdin, V.V., Agur, S., Galil, N., Armon, R., Schechter, I., 2004. A new sono-electrochemical method for enhanced detoxification of hydrophilic chloroorganic pollutants in water. Ultrason. Sonochem. 11, 365–372.