Photodegradation of the antibiotics nitroimidazoles in aqueous solution by ultraviolet radiation

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Photodegradation of the antibiotics nitroimidazoles in aqueous solution by ultraviolet radiation G. Prados-Joya, M. Sa´nchez-Polo, J. Rivera-Utrilla*, M. Ferro-garcı´a Departamento de Quı´mica Inorga´nica, Facultad de Ciencias, Universidad de Granada, 18071, Granada, Spain

article info

abstract

Article history:

The objective of this study was to analyze the efficacy of ultraviolet (UV) radiation in the

Received 26 May 2010

direct photodegradation of nitroimidazoles. For this purpose, i) a kinetic study was per-

Received in revised form

formed, determining the quantum yield of the process; and ii) the influence of the different

4 August 2010

operational variables was analyzed (initial concentration of antibiotic, pH, presence of

Accepted 10 August 2010

natural organic matter compounds, and chemical composition of water), and the time

Available online 28 September 2010

course of total organic carbon (TOC) concentration and toxicity during nitroimidazole photodegradation was studied. The very low quantum yields obtained for the four nitro-

Keywords:

imidazoles are responsible for the low efficacy of the quantum process during direct

Nitroimidazoles

photon absorption in nitroimidazole phototransformation. The R254 values obtained show

Ultraviolet radiation

that the dose habitually used for water disinfection is not sufficient to remove this type of

Treatment

pharmaceutical; therefore, higher doses of UV irradiation or longer exposure times are

Oxidation

required for their removal. The time course of TOC and toxicity during direct photodegradation (in both ultrapure and real water) shows that oxidation by-products are not oxidized to CO2 to the desired extent, generating oxidation by-products that are more toxic than the initial product. The concentration of nitroimidazoles has a major effect on their photodegradation rate. The study of the influence of pH on the values of parameters 3 (molar absorption coefficient) and k0 E (photodegradation rate constant) showed no general trend in the behavior of nitroimidazoles as a function of the solution pH. The components of natural organic matter, gallic acid (GAL), tannic acid (TAN) and humic acid (HUM), may act as promoters and/or inhibitors of OH$ radicals via photoproduction of H2O2. The effect of GAL on the metronidazole (MNZ) degradation rate markedly differed from that of TAN or HUM, with a higher rate at low GAL concentrations. Differences in MNZ degradation rate among waters with different chemical composition are not very marked, although the rate is slightly lower in wastewaters, mainly due to the UV radiation filter effect of this type of water. ª 2010 Elsevier Ltd. All rights reserved.

1.

Introduction

Ultraviolet (UV) radiation is frequently applied to disinfect water intended for human consumption and wastewater. Due to the greater chemical contamination of water, UV radiation (Hijnen et al., 2006) is increasingly proposed as a technology to

remove organic micropollutants, underlining its high efficacy to eliminate certain pesticides and pharmaceuticals from water (Kang et al., 2004; Lazarova and Savoye, 2004). Significant advances have recently been made in our understanding of the photochemical processes undergone by organic contaminants and pharmaceuticals in aqueous

* Corresponding author. Tel.: þ34 958248523; fax: þ34 958248526. E-mail address: [email protected] (J. Rivera-Utrilla). 0043-1354/$ e see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2010.08.015

394

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

medium (Boreen et al., 2004; Latch et al., 2003; Packer et al., 2003). However, fewer data are available on their photochemical transformation (Boreen et al., 2003). The majority of pharmaceuticals are photo-active, i.e. able to absorb light. This is because their structures generally contain aromatic rings, heteroatoms, and other functional groups that make them prone to absorb UVevis radiation (direct photolysis) or to react with photosensitizing species capable of inducing pharmaceutical photodegradation in natural water (indirect photolysis). During direct photolysis, photon absorption gives rise to compounds in excited electronic states that are susceptible to chemical transformation. However, indirect or sensitized photolysis leads to the transformation of contaminants by energy transference or by chemical reactions with transitory species formed by the presence of light, such as hydroxyl radicals (HO), singlet oxygen (1O2), and triplet excited states of natural organic matter (3NOM*) (Schwarzenbach et al., 2003; Canonica et al., 1995; Canonica and Tratnyek, 2003; Gerecke et al., 2001; Zepp et al., 1985). Hence, the efficacy of direct photooxidation is governed by the contaminant absorption spectrum and the quantum yield of the process (V), whereas the dominant mechanism in indirect photolysis is the reaction between OH radicals and the micropollutant. Hence, addition of H2O2 during the photooxidation process accelerates the micropollutant removal rate, reducing the UV radiation required in comparison to direct photooxidation (Rosenfeldt and Linden, 2004); this is due to the generation of highly reactive radicals in H2O2 decomposition (Glaze et al., 1987). Nitroimidazole antibiotics were recently detected in waters at concentrations of 0.1e90.2 mg/L (Lindberg et al., 2004). They are widely used to treat infections caused by anaerobic and

protozoan bacteria (e.g., Trichomonas vaginalis and Giardia lamblia) in humans and animals and are added to chow for fish and fowl, leading to their accumulation in animals, fish-farm waters and, especially, meat industry effluents. Little is yet known about the capacity of current water treatment systems to remove nitroimidazoles (Wennmalm and Gunnarsson, 2005) but it is not expected to be very high given the complex chemical structure of these compounds. The objective of the present study was to analyze the efficacy of UV radiation in the direct photooxidation of nitroimidazoles. For this purpose, i) a kinetic study was conducted to determine the quantum yield of the process; and ii) the influence of the different operational variables (initial concentration of antibiotic, pH, presence of NOM components and chemical composition of water) was analyzed, and the time course of total organic carbon (TOC) concentration and toxicity during nitroimidazole photodegradation was studied.

2.

Experimental

2.1.

Reagents

All chemical reagents used (phosphoric acid, sodium hydroxide, hydrogen peroxide, atrazine, gallic acid, tannic acid, humic acid, acetonitrile, ammonium acetate, and nitroimidazoles) were high purity analytical grade reagents supplied by SigmaeAldrich. Ultrapure water was obtained using Milli-Q equipment (Millipore). Fig. 1 depicts the chemical structure and the acidity constant values of the nitroimidazoles selected for this study: Metronidazole (MNZ), Dimetridazole (DMZ), Tinidazole (TNZ) and Ronidazole (RNZ).

Fig. 1 e Chemical structure of the nitroimidazoles studied.

395

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

natural waters (ground and surface waters) and wastewaters from the Motril (Granada, Spain) drinking water treatment (DWTP) and wastewater treatment (WWTP) plants, respectively. After characterization and filtering, water samples were kept refrigerated (T ¼ 297 K) until use.

1.0

[M]/[M] o

0.8

0.6

0.4

0.2

0.0 0

50

100

150

200

t (min) Fig. 2 e Removal kinetics of the four nitroimidazoles by means of UV photodegradation. [nitroimidazole]o [ 200 mM, pH [ 5e7, T [ 298 K; (>), MNZ; (,), DMZ; (6), TNZ; (B), RNZ.

2.2.

UV irradiation experimental device

The photoreactor used for nitroimidazoles photodegradation was equipped with a low-pressure mercury lamp (Heraeus Noblelight model TNN 15/32, nominal power 15 W). The problem solution is set in 6 quartz tubes with 1 cm diameter and 30 mL placed around and equidistant from the Hg lamp. These tubes are immersed in bidistilled water in continuous recirculation for temperature control, using a Frigiterm ultrathermostat, with a magnetic agitation system in each tube. In each experiment, after stabilizing the lamp and controlling the temperature (298 K), the photoreactor was turned on, and aliquots were withdrawn from the reactor at different time intervals in order to assess: i) nitroimidazole concentration, ii) total organic carbon (TOC), and iii) toxicity of the photodegradation products. The substrate concentration is higher than it was found on environmental samples. This high concentration was used to follow better its analysis by HPLC method. The influence of the presence of organic matter was analyzed by adding gallic acid (GAL), tannic acid (TAN), or humic acid (HUM) to the solution in some experiments. The concentration of these acids (components of natural organic matter) ranged from 10e80 mg/L. Each photooxidation experiment was repeated three times.

2.3.

Water sampling and characterization

The influence of the chemical composition of water on the direct photodegradation of nitroimidazoles was studied in

2.4.

Analytical methods

2.4.1.

Determination of nitroimidazole concentration

Chromatographic follow-up of nitroimidazole concentrations was done using a WATERS ALLIANCE 2690a analytical HPLC with WATERS M-996 photodiode detector and automatic injector with capacity for 120 flasks. The chromatographic column was Nova-Pak C18 Cartridge, particle size 4 mm and 3.9  150 mm inner diameter. The mobile phase used was a buffer solution of pH 4.3 with 96% 5.0 mM ammonic acetate and 4% acetonitrile in isocratic mode at a 1 mL min1 flow.

2.4.2.

Determination of atrazine concentration

Atrazine was used as actinometer to determine the radiant energy of the lamp (Canonica et al., 1995). Atrazine concentration was determined by high performance liquid chromatography (HPLC) using a chromatographic column with the same characteristics as in the nitroimidazole determination. The mobile phase was a buffer solution of pH 4.5, with 50% 2.5 mM ammonium acetate and 50% acetonitrile in isocratic mode and a flow of 1 mL min1.

2.4.3.

Determination of total organic carbon concentration

TOC was determined by using a Shimadzu V-CSH equipment with ASI-V autosampler.

2.4.4.

Toxicity determination

A LUMISTOX 300 system was used to measure toxicity, based on the standardized biotest (DIN/EN/ISO 11348-2) of Vibrio fischeri bacteria inhibition (NRRL B-11177). The measurement is based on inhibition of the luminosity intensity of marine bacteria Vibrio fisheri. Toxicity is expressed as percentage inhibition after 15 min of exposure. The results presented in this manuscript were obtained considering tree measurements of the same sample and the error associated to these data is between 5 and 10%.

3.

Results and discussion

3.1. Direct photodegradation of nitroimidazoles: quantum yields; time course of total organic carbon and toxicity Fig. 2 depicts the kinetics of direct nitroimidazole photodegradation with low-pressure Hg lamp (254 nm). It shows

Table 1 e Parameters obtained from direct irradiation at 254 nm of the four nitroimidazoles. Nitroimidazole MNZ DMZ TNZ RNZ

3 (m2 mol1) 209.72 224.45 233.92 226.13

k 104 (s1) 1.72  1.67  1.09  1.18 

0.09 0.20 0.09 0.18

V 104 (mol Eins1) 34.7  31.5  19.6  22.1 

1.8 3.9 1.7 3.4

k0 E (m2 Eins1) 1.68 1.63 1.06 1.15

   

0.09 0.20 0.09 0.18

R254 (%) 0.14  0.14  0.09  0.10 

0.01 0.02 0.01 0.01

396

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

a

25

50

b

40

% Inhibition

TOC (mg/L)

20 15 10 5

30 20 10

0

0

0

50

100

150

200

0

t (min)

50

100

150

200

t (min)

Fig. 3 e Time course of the concentration of total organic carbon (a) and toxicity (b) during UV irradiation of the four nitroimidazoles in ultrapure water. [nitroimidazole]o [ 200 mM, pH [ 5e7, T [ 298 K; (>), MNZ; (,), DMZ; (6), TNZ; (B), RNZ.

slightly higher removal rates for MNZ and DMZ than for TNZ and RNZ. At 3 h of treatment, the concentration of all nitroimidazoles was reduced by 70e85%. A pseudo-first order kinetic model was applied to the above results to determine the nitroimidazole photodegradation rate constant. Equation (1) (Sharpless and Linden, 2003) was then used to calculate the quantum yield (V) for each nitroimidazole. Table 1 shows the results obtained. Fl ¼

kl 2:303$El $3l

(1)

where kl is the photodegradation rate constant (s1); El is the rate of energy emitted, corresponding to the photon flow emitted by the lamp (E s1 m2); 3l is the molar absorption coefficient at the wavelength in question (m2 mol1); and Vl is the quantum yield (mol E1).

0.05

-1

Ф (mol·Eins ) · 10

0.04

0.03

0.02

The rate of energy irradiated by the lamp was determined by actimometry, using a solution of 5 mM atrazine as actinometer (Canonica et al., 1995) and obtaining an energy of 1.027$104 E s1 m2 for the lamp used. For this purpose, the quantum yield of atrazine was considered to be 0.046 mol E1 and the molar absorption coefficient to be 386 m2 mol1 at 254 nm wavelength (Hessler et al., 1993). Table 1 shows the values of the study parameters, showing a very low quantum yield for all four nitroimidazoles (values from 34.7$104 for MNZ to 19.6$104 for TNZ), responsible for a low efficacy of the quantum process in nitroimidazole phototransformation. This low efficacy (Table 1), alongside the results in Fig. 2, confirm the need for long irradiation times to achieve their complete removal. Except for MNZ, no data on quantum yields of nitroimidazoles have been reported in the literature; nonetheless, the values obtained are similar to those reported by other authors for pharmaceutical compounds (Rosenfeldt and Linden, 2004; Canonica et al., 2008; Sa´nchez-Polo et al., 2007; Lo´pez et al., 2002; Andreozzi et al., 2003). The value of (34.7  1.8)$104 mol E1 obtained for MNZ is close to the value of 0.0033 mol E1 reported by Shemer et al. (2006). For comparative purposes, it is essential to consider the apparent photodegradation rate constant normalized by the energy of the lamp, k0 E (m2 E1), using Equation (2). This constant is independent of the fluctuations in energy irradiated by the lamp and permits direct comparisons among phototransformation rate constants obtained with different photoreactors (Canonica et al., 2008). 0

kE ¼

0.01

0.00 0

500

1000

1500

2000

[nitroimidazole]o (μM) Fig. 4 e Quantum yield of the photodegradation process as a function of initial nitroimidazole concentration. pH [ 5e7, T [ 298 K; (>), MNZ; (,), DMZ; (6), TNZ; (B), RNZ.

kl El

(2)

where kl (s1) is the photodegradation rate constant of contaminants used in Equation (1), and El (Einstein$s1 m2) is the radiant energy emitted by the lamp at a wavelength of 254 nm, calculated by actinometry (Canonica et al., 1995). The k0 E values obtained are shown in Table 1. Table 1 also shows the percentage nitroimidazole removal for an irradiation dose of 400 J m2 (R254). This parameter determines the applicability of UV radiation in nitroimidazole photodegradation under the real conditions of a water treatment plant. An irradiation dose

397

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

Table 2 e Specific constants of the photodegradation prediction model for the four nitroimidazoles. Nitroimidazole MNZ DMZ TNZ RNZ

m

n

0.3293 1.0143 1.0938 0.1135

0.5857 0.7328 0.8483 0.4742

of 400 J m2 was selected as reference because it is the minimum value recommended by different European bodies ¨ Norm, 2001; Cabaj et al., for water disinfection (DVGW, 1997; O 1996). With the present lamp, the time required to reach an irradiation dose of 400 J m2 is 8.3 s, and the radiant energy is equivalent to 8.49$104 E m2 for a wavelength of 254 nm. Hence, the percentage of nitroimidazole removed with this irradiation dose, R254, can be calculated by means of Equation (3). 0 4 R254 ¼ 1  eðkE 8:49$10 Þ

(3)

0

where k E is the apparent photodegradation rate constant normalized by the energy of the lamp Equation (2). As shown in Table 1, a very low percentage of nitroimidazole removal was achieved at the minimum irradiation dose (400 J m2) (Table 1). Hence, the usual dose for water disinfection is not adequate to remove this type of pharmaceutical, which requires higher doses of UV irradiation or longer exposure times to be eliminated by direct photolysis. Data in Table 1 demonstrates the low performance of the nitroimidazole photodegradation process, decreasing in the order: MNZ > DMZ > RNZ > TNZ.

5.0

-1

ε

3.2. Influence of the different operational variables on nitroimidazole photodegradation This section analyzes the effects on nitroimidazole photodegradation performance of the operational parameters studied: nitroimidazole concentration, medium pH and presence of natural organic matter.

3.2.1.

Influence of nitroimidazole concentration

Fig. 4 depicts the quantum yield of nitroimidazoles as a function of their concentration, showing that the initial concentration

300

5.0

250

4.0

300

ε

250

200

200 3.0

3.0

pka1 150

2

150 2.0

2.0

k’E

1.0

50

0.0

0 0

2

4

6

k’E

100

pka1

8

250

4.0 -1

0 0

300

ε

50

0.0

10

5.0

100

1.0

2

4

6

8

10

5.0

300

ε

4.0

250 200

3.0

2

150

pka1

2.0

k’ E

100 50 0

0.0 2

4

6

pH

150

k’ E

2.0

1.0

0

pka1

8

10

100 1.0

ε (m2·mol-1)

200 3.0

ε (m2·mol-1)

k'E (m ·Eins )

4.0

k'E (m ·Eins )

Two key parameters of the efficacy of any treatment system are: i) the concentration of total organic carbon and ii) the toxicity of the oxidation by-products generated. Fig. 3 shows the time course of TOC and toxicity values during the photodegradation of the four nitroimidazoles under study. According to the results in Fig. 3a, although the nitroimidazole concentration considerably decreases with radiation time (Fig. 2), oxidation by-products do not transform into CO2 to the desired extent. Consequently, they generate fractions of smaller molecular weight than the original nitroimidazole and maintain the TOC concentration constant throughout the treatment time. Moreover, as shown in Fig. 3b, the mixture of by-products generated during nitroimidazole photodegradation sometimes shows higher toxicity than the original product. Hence, nitroimidazole photodegradation may give rise to compounds that are pharmacologically active and have higher toxicity than the original compound. We highlight the results obtained for TNZ, which show a substantial increase in toxicity at 3 h of treatment (Fig. 3b).

50

0.0

0 0

2

4

6

8

10

pH

Fig. 5 e Molar absorption coefficient (3) and apparent photodegradation constant normalized by the energy emitted by the lamp (k0 E) as a function of solution pH for the four nitroimidazoles studied. T [ 298 K (>), MNZ; (,), DMZ; (6), TNZ; (B), RNZ.

398

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

Table 3 e Kinetic parameters of direct phototransformation of ionic (j [ 1) and neutral (j [ 2) species of nitroimidazoles. Nitro imidazole pka1 31 (m2 mol1) k0E1 (m2 Eins1) V1 103 (mol Eins1) 32 (m2 mol1) k0E2 (m2 Eins1) V2 103 (mol Eins1) MNZ DMZ TNZ RNZ

2.58 2.81 2.30 1.32

271.18 288.12 178.88 171.31

3.78 1.33 0.28 0.31

6.05 2.01 0.69 0.79

affects the quantum yield and therefore the photodegradation rate Equation (1). The decrease in photodegradation rate in nitroimidazoles with the increase of their concentration is related to the energy absorbed by each nitroimidazole molecule. Hence, given that the radiation energy deposited in the medium per unit volume is constant, nitroimidazole molecules can accept more radiant energy at lower concentrations, explaining the behavior observed. The equations that predict the quantum yield of each nitroimidazoles are based on the data in Fig. 4. These are potential-type equations Equation (4) and allow the quantum yield of the photochemical process to be obtained at the temperature and pH shown in the footnote of Fig. 4. n

F254 ¼ m$½nitroimidazolo

(4)

where V254 is the quantum yield (mol E1) at a wavelength of 254 nm, m and n are the specific constants obtained from Fig. 4 for each nitroimidazole (Table 2) and [nitroimidazole]o is the initial concentration of the nitroimidazole. Once the quantum yield is known, the photodegradation rate constant of any of the four nitroimidazoles can be determined as a function of their initial concentration by applying Equation (1). These results are of great interest from

209.72 224.45 233.91 226.13

2.10 1.73 1.42 1.80

4.35 3.35 2.64 3.46

an industrial standpoint because they allow pre-treatment prediction of the effectiveness of UV radiation to remove nitroimidazoles.

3.2.2.

Influence of solution pH

The influence of the solution pH in nitroimidazole photodegradation was analyzed in experiments with pH values ranging from 2e9. Based on the pKa of each nitroimidazole (Fig. 1) and the distribution of species that each presents, the four nitroimidazoles are in their cationic and/or neutral form in the pH range studied. Fig. 5 shows the variation in global molar absorption coefficient (3) and global photodegradation rate constant (k0 E) as a function of the solution pH for each nitroimidazole. Interestingly, the changes observed are determined by the pKa of the antibiotic. According to Equation (1), an increase in the molar absorption coefficient (3) would produce an increase in the k0 E value, but the results show no general tendency that describes the influence of pH on the 3 and k0 E values of these nitroimidazoles. In fact, the variation in trends among these nitroimidazoles differentiates four distinct behaviors: a) MNZ: increase in 3 and k0 E at pH < 4

Fig. 6 e Chemical structure of gallic acid (a), humic acid (b), and tannic acid (c).

399

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

60

3.5

[NOM]= 0 mg/L

2.5 -1

k·10 (s )

30

2.0

4

Transmittance %

3.0 45

1.5 1.0

15

0.5 0

0.0 0

20

40

60

80

0

1

[NOM] (mg/L)

b) DMZ: increase in 3 and decrease in k0 E at pH < 5 c) TNZ: decrease in 3 and k0 E at pH < 5 d) RNZ: 3 and k0 E remain virtually unchanged in the studied pH range. The variations in these parameters as a function of the medium pH are similar to those observed for other pharmaceutical compounds (Canonica et al., 2008). This global behavior was studied in greater depth by quantitative analysis of the photodegradation of each nitroimidazole species in the medium as a function of the pH, using Equation (5), which is obtained by combining Equations (1) and (3). 0

kEj ¼ 2:303$3j $Fj

(5)

where k0 Ej is the apparent rate constant for each nitroimidazole species “j” present in the solution. The contribution of each of each species towards total k0 E can be represented by Equation (6): 0

X

0

aj $kEj

(6)

j

Table 4 e MNZ photodegradation rate constants as a function of the concentration of acid present. GAL (mg L1) 0 9.8 18.1 36.0 76.4 0 0 0 0 0 0 0 0

TAN (mg L1)

HUM (mg L1)

0 0 0 0 0 15.9 27.2 43.1 79.0 0 0 0 0

0 0 0 0 0 0 0 0 0 9.5 23.2 36.0 79.2

3

4

[NOM]/[MNZ]o

Fig. 7 e Transmittance of NOM components as a function of concentration at l [ 254 nm. (>), GAL; (,), TAN; (6), HUM.

kE ¼

2

k 104 (s1) 2.36 3.08 2.87 1.97 0.73 1.81 1.50 1.16 0.60 1.84 1.53 1.26 0.66

            

0.17 0.11 0.15 0.11 0.04 0.05 0.15 0.05 0.03 0.12 0.04 0.04 0.02

Fig. 8 e MNZ photodegradation rate constant as a function of the relationship between NOM and MNZ concentrations. [MNZ]o [ 20 mg/L, T [ 298 K; (>), GAL/MNZ; (,), TAN/ MNZ; (6), HUM/MNZ.

where aj represents the molar fraction of each of the species (Sjaj ¼ 1). Once the species distribution and pKa of each nitroimidazole is known, the rate constant and molar absorption coefficient can be calculated by means of linear regression:   0 0 0 0 0 0 kE ¼ ð1  a2 Þ$kE1 þ a2 $kE2 ¼ kE1 þ kE2  kE1 a2

(7)

3 ¼ ð1  a2 Þ$31 þ a2 $32 ¼ 31 þ ð32  31 Þa2

(8)

0 kEj

After calculating and 3j for the ionic ( j ¼ 1) and neutral ( j ¼ 2) species of each nitroimidazole, the quantum yield of the corresponding species can be obtained by means of Equation (9). 0

Fj ¼

kEj 2:303$3j

(9)

Table 3 shows the results obtained by applying these equations. Using these parameters, the values of k0 E and V can be determined for any pH value in the studied range. As shown in Table 3, the rate constant (k0 E) and quantum yields (V) values obtained for the species of each nitroimidazole vary by only one order of magnitude, with minimum values for the ionic form of TNZ and maximum values for the ionic form of MNZ.

3.2.3.

Influence of the presence of gallic, tannic or humic acid

The influence of natural organic matter (NOM) components during nitroimidazole photodegradation was analyzed with experiments of MNZ photodegradation in the presence of three components of NOM. MNZ was selected for these experiments because it is the most representative nitroimidazole and the one most frequently detected in waters. The concentration of NOM in waters ranges from approximately 0.1 mg to >100 mg of total organic carbon (TOC) per liter, depending on their origin (Artinger et al., 2000). Fig. 6 shows the chemical structures of GAL, TAN, and HUM, the three NOM components selected for the study. GAL is

400

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

Table 5 e Chemical characteristics of the waters used. Water

pH

Ultrapure Surface Ground Waste

6.8 8.1 7.5 7.8

[HCO 3] 1

[SO2 4 ] 1

[NO 3] 1

0.0 5.2 8.9 10.5

0 104 150 257

0.0 1.5 3.7 11.7

a

T (%) TOC (mg L1) (meq L ) (mg L ) (mg L ) 0.0 9.8 17.4 25.2

100.0 97.4 97.8 60.1

Table 6 e Parameters obtained from UV irradiation of MNZ in waters with different chemical compositions. Water

k 104 (s1)

k0 E (m2 Eins1)

Ultrapure Surface Ground Waste

1.72  0.09 1.73  0.09 2.08  0.15 1.66  0.08

1.68  0.09 1.68  0.09 2.03  0.15 1.61  0.08

R254 (%) 0.14 0.14 0.17 0.14

 0.01  0.01  0.01  0.01

k0 E: photodegradation rate constant normalized by the energy of the lamp. R254: percentage removal for an irradiation dose of 400 J m2.

a: Transmittance (%) at 254 nm.

considered the basic structural unit of NOM, and HUM is considered its main component. HUM has a high molecular weight and a complex structure with a large number of aromatic rings and oxygenated functional groups (Steelink, 2002; Choudhry, 1984; Cooper et al., 2008). Knowledge of the amount of light absorbed by the different components of NOM is necessary to study the influence of humic matter during UV irradiation of MNZ and its effects on the removal rate. For this purpose, the transmittance of the three acids was determined at the studied concentrations for a wavelength of 254 nm (Fig. 7). According to the transmittance data in Fig. 7, the amount of UV light absorbed by NOM can be known, hence reducing the number of photons that reach the MNZ molecules (Canonica et al., 2008). The little transmittance shown by the three acids at high concentrations will have a very negative effect on the direct photodegradation of MNZ. Table 4 shows the MNZ photodegradation rate constant values in the presence of different concentrations of the selected acids; the MNZ photodegradation rate increases at small concentrations of GAL but decreases at any of the TAN and HUM concentrations studied. Fig. 8 plots the photodegradation rate constant against the relationship between concentrations of NOM constituents and MNZ. The presence of GAL has a markedly different effect on MNZ degradation in comparison to the presence of TAN or HUM. MNZ removal is favored only at a GAL/MNZ

concentration ratio <1.4. The data in Table 4 and Fig. 8 show that low concentrations of GAL in the medium favor MNZ photodegradation, despite the lesser transmittance of GAL versus TAN or HUM (Fig. 7). Therefore, GAL may act as a promoter of OH$ radicals, which oxidizes MNZ molecules. In contrast, the behavior in the presence of TAN and HUM suggests a predominant OH$ radical inhibition effect, due to their complex structure (Fig. 6) and the high reactivity of NOM against OH$ radicals (kOH$ ¼ 108 M1s1) (Basfar et al., 2005; Buxton et al., 1988). An increase in the concentration of all three acids decreases the MNZ removal rate constant to very similar levels, k < 0.73$104 s1, (Table 4 and Fig. 8), due to the little or null transmittance of GAL, TAN and HUM at high concentrations (Fig. 7). The different effects of NOM components on contaminant photodegradation have been studied by various authors (Zepp et al., 1985, 1981a, 1981b; Choudhry, 1984; Simmons and Zepp, 1986; Van Noort et al., 1988; Zheng and Ye, 2001; Zhan et al., 2006; Xu et al., 2007; Garbin et al., 2007). Cooper and Zika (1983) were the first to report that exposure of natural waters to high energy solar radiation gives rise to photoproduction of H2O2 (Cooper et al., 1988), which is an efficient OH radical producer under the action of UVevis light (Glaze et al., 1987).

3.3. Applicability of UV radiation for nitroimidazole degradation in water with different chemical compositions

1.0

[MNZ]/[MNZ] o

0.8

0.6

0.4

0.2

0.0 0

50

100

150

200

t (min) Fig. 9 e Influence of the chemical composition of water on the MNZ photodegradation rate. [MNZ]o [ 200 mM, pH [ 6e8, T [ 298 K; (>), ultrapure water; (B), surface water; (6), groundwater; (,), wastewater.

The applicability of UV radiation to remove nitroimidazole from water was studied by analyzing the influence of water chemical composition during MNZ photodegradation. Experiments were conducted with water of different origin and chemical composition: ultrapure, groundwater and wastewater. Table 5 depicts the results of their chemical characterization. Fig. 9 shows the results obtained from UV irradiation of MNZ in the different waters studied. The differences in MNZ degradation rate among the different waters are not very marked. There is a slight decrease in the wastewater samples, largely to the lower transmittance (T ¼ 60%) in this type of water, causing absorption of UV radiation and considerably reducing the number of photons reaching the nitroimidazole. In this case, the organic matter present in wastewater acts as a filter of UV radiation, reducing the efficacy of the treatment to remove MNZ from the medium. Table 6 shows the kinetic parameters and the percentage removal (R254) obtained by UV irradiation of MNZ in the studied waters. MNZ photodegradation rate constant values

401

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

a

b

50

20

% Inhibition

TOC (mg/L)

40

25

30

20

15

10

5

10

0

0 0

20

40

60

80

100

% MNZ removed

0

20

40

60

80

100

% MNZ removed

Fig. 10 e Time course of total organic carbon (a) and toxicity by Vibrio Fischeri inhibition (b) as a function of %MNZ removed during UV irradiation in waters with different chemical compositions: (,), surface water; (>), groundwater; (6), wastewater; (B), ultrapure. [MNZ]o [ 200 mM, pH [ 6e8, T [ 298 K.

(Table 6) are similar among ultrapure, surface and wastewaters. The rate is slightly higher in groundwater, suggesting a small indirect photodegradation similar to that observed in MNZ photodegradation in the presence of low concentrations of GAL. These results show that the presence of NOM with low humification rate, e.g., GAL and hydrobenzoic acids, favors the indirect photodegradation of contaminants (Fukushima and Tatsumi, 1999). Consequently, there is a need for a wider study to characterize the humic acids present in natural waters and to determine the role of each in contaminant photodegradation. In addition, the presence in this type of water of a high concentration of ions susceptible of transformation into oxidant radicals (sulfates, nitrates) that can react with MNZ, would also explain this increase in MNZ degradation rate in groundwaters. Fig. 10 depicts the time course of TOC values and toxicity during MNZ removal in ultrapure, surface, groundwater and wastewater. According to these results, UV radiation does not significantly reduce the TOC concentration during the 3 h irradiation in any of the studied waters; slight TOC reduction is only detected in wastewater and groundwater. Hence, although there is a considerable reduction in nitroimidazole concentration (Fig. 9), the degradation compounds are not mineralized to the desired extent and remain in the medium. Moreover, Fig. 10b shows that the mixture of by-products generated during MNZ photodegradation has a highly variable toxicity depending on the type of water studied. Interestingly, regardless of the type of water studied, minimum toxicity values are reached when around 40% of the MNZ has been removed.

4.

Conclusions

Quantum yields obtained for the four nitroimidazoles are very low R254 values obtained show that the dose habitually used for the disinfection of waters is not adequate to remove this type of pharmaceutical compounds. The time course of TOC and of toxicity during direct photodegradation of nitroimidazoles, in both ultrapure and real

waters, shows that oxidation by-products do not transform into CO2 to the desired extent. Therefore, they generate fractions of lower molecular weight than the original molecule, maintaining a constant TOC concentration throughout the treatment time and possibly giving rise to pharmacologically active compounds with higher toxicity than the original nitroimidazole. The concentration of nitroimidazole has a major effect on its photodegradation rate. The study of the influence of pH on the values of parameters 3 and k0 E shows no general tendency for the behavior of nitroimidazoles as a function of the pH. The rate constants (k0 E) and quantum yields (V) obtained for the different species of each nitroimidazoles vary by only one order of magnitude, with minimum values for the ionic form of TNZ and maximum values for the ionic form of MNZ. The NOM components GAL, TAN, and HUM may act as promoters and/or inhibitors of OH$ radicals generated by H2O2 photoproduction. Results show that the presence of GAL has a markedly different effect on the MNZ degradation rate from that of TAN or HUM, with an increase in this rate at low GAL concentrations. These results appear to show that, under these conditions, GAL mainly acts as a promoter of OH$ radicals, which oxidize MNZ molecules. In contrast, the presence of TAN or HUM decreases MNZ degradation rate, suggests a predominant effect of their OHradical inhibiting capacity, due to their complex structure and high reactivity against OHradicals. Differences in MNZ degradation rate among the studied waters, which have different chemical compositions, are not very marked, although there is a slight decrease in wastewaters, mainly because of the UV radiation filter effect of this type of water.

Acknowledgments The authors are grateful for the financial support provided by MEC-DGI, FEDER (Project: CTQ2007-67792-C02-01/PPQ), Junta de Andalucı´a (Project: RNM3823).

402

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

references

Andreozzi, R., Raffaele, M., Nicklas, P., 2003. Pharmaceuticals in STP effluents and their solar photodegradation in aquatic environment. Chemosphere 50 (10), 1319e1330. Artinger, R., Buckau, G., Geyer, S., Fritz, P., Wolf, M., Kim, J.I., 2000. Characterization of groundwater humic substances: influence of sedimentary organic carbon. Applied Geochemistry 15 (1), 97e116. Basfar, A.A., Khan, H.M., Al-Shahrani, A.A., Cooper, W.J., 2005. Radiation induced decomposition of methyl tert-butyl ether in water in presence of chloroform: kinetic modelling. Water Research 39 (10), 2085e2095. Boreen, A.L., Arnold, W.A., McNeill, K., 2004. Photochemical fate of sulfa drugs in then aquatic environment: sulfa drugs containing five-membered heterocyclic groups. Environmental Science and Technology 38 (14), 3933e3940. Boreen, A.L., Arnold, W.A., McNeill, K., 2003. Photodegradation of pharmaceuticals in the aquatic environment: a review. Aquatic Sciences 65 (4), 320e341. Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. Journal of Physical Chemistry Reference Data 17 (2), 513e886. Cabaj, A., Sommer, R., Schoenen, D., 1996. Biodosimetry: model calculations for u.v. water disinfection devices with regard to dose distributions. Water Research 30 (4), 1003e1009. Canonica, S., Meunier, L., von Gunten, U., 2008. Phototransformation of selected pharmaceuticals during UV treatment of drinking water. Water Research 42 (12), 121e128. Canonica, S., Tratnyek, P.G., 2003. Quantitative structure-activity relationships for oxidation reactions of organic chemicals in water. Environmental Toxicology and Chemistry 22 (8), 1743e1754. Canonica, S., Jans, U., Stemmler, K., Hoigne´, J., 1995. Transformation kinetics of phenols in water: photosensitization by dissolved natural organic material and aromatic ketones. Environmental Science and Technology 29 (7), 1822e1831. Choudhry, G.G., 1984. Humic substances, structural aspects, and photophysical, photochemical and free radical characteristics. In: The Handbook of Environmental Chemistry: The Natural Environment and the Biogeochemical Cycles, vol. 1C, pp. 1e24. Cooper, W.J., Song, W., Gonsior, M., Kalnina, D., Peake, B.M., Mezyk, S.P., 2008. Recent advances in structure and reactivity of dissolved organic matter in natural waters. Water Science and Technology: Water Supply 8, 615e623. Cooper, W.J., Zika, R.G., Petasne, R.G., Plane, J.M.C., 1988. Photochemical formation of H2O2 in natural waters exposed to sunlight. Environmental Science and Technology 22 (10), 1156e1160. Cooper, W.J., Zika, R.G., 1983. Photochemical formation of hydrogen peroxide in surface and ground waters exposed to sunlight. Science 220 (4598), 711e712. DVGW, 1997. W 294. UVeDesinfektionsanlagen fu¨r die Trinkwasserversorgung e Anforderungen und Pru¨fung. Fukushima, M., Tatsumi, K., 1999. Photosensitized production of hydrogen peroxide in the aqueous system containing dissolved organic carbon. Toxicology Environmental Chemistry 69 (3e4), 435e448. Garbin, J.R., Milori, D.M.B.P., Simo˜es, M.L., da Silva, W.T.L., Neto, L.M., 2007. Influence of humic substances on the photolysis of aqueous pesticide residues. Chemosphere 66 (9), 1692e1698.

Gerecke, A.C., Canonica, S., Muller, S.R., Scharer, M., Schwarzenbach, R.P., 2001. Quantification of dissolved natural organic matter (DOM) mediated phototransformation of phenylurea herbicides in lakes. Environmental Science and Technology 35 (19), 3915e3923. Glaze, W.H., Kang, J., Chapin, D.H., 1987. Chemistry of water treatment processes involving ozone, hydrogen peroxide and ultraviolet radiation. Ozone: Science and Engineering 9, 335e352. Hessler, D.P., Gorenflo, V., Frimmel, F.H., 1993. Degradation of aqueous atrazine and metazachlor solutions by UV and UV/H2O2einfluence of pH and herbicide concentration. Acta Hydrochimica Hydrobiology 21 (4), 209e214. Hijnen, W.A.M., Beerendonk, E.F., Medema, G.J., 2006. Inactivation credit of UV radiation for viruses, bacteria and protozoan (oo) cysts in water: a review. Water Research 40 (1), 3e22. Kang, S.J., Allbaugh, T.A., Reynhout, J.W., Erickson, T.L., Olmstead, K.P., Thomas, L., Thomas, P., 2004. Selection of an ultaviolet disinfection system for a municipal wastewater treatment plant. Water Science and Technology 50, 163e169. Latch, D.E., Stender, B.L., Packer, J.L., Arnold, W.A., McNeill, K., 2003. Photochemical fate of pharmaceuticals in the environment: cimetidine and ranitidine. Environmental Science and Technology 37 (15), 3342e3350. Lazarova, V., Savoye, P., 2004. Technical and sanitary aspects of wastewater disinfection by UV irradiation for landscape irrigation. Water Science and Technology 50, 203e209. Lindberg, R., Jarnheimer, P., Olsen, B., Johansson, M., Tysklind, M., 2004. Determination of antibiotic substances in hospital sewage water using solid phase extraction and liquid chromatography/mass spectrometry and group analogue internal standards. Chemosphere 57, 1479e1488. Lo´pez, A., Bozzi, A., Mascolo, G., Ciannarella, R., Passino, R., 2002. UV and H2O2/UV degradation of a pharmaceutical intermediate in aqueous solution. Annali di Chimica 92 (1e2), 41e51. ¨ Norm, 2001. M 5873e1. Anlagen zur Desinfektion von Wasser O mittels Ultraviolett-Strahlen. Packer, J.L., Werner, J.J., Latch, D.E., McNeill, K., Arnold, W.A., 2003. Photochemical fate of pharmaceuticals in the environment: naproxen, diclofenac, clofibric acid, and ibuprofen. Aquatic Sciences 65 (4), 342e351. Rosenfeldt, E.J., Linden, K.G., 2004. Degradation of endocrine disrupting chemicals bisphenol A, ethinyl estradiol, and estradiol during UV photolysis and advanced oxidation processes. Environmental Science and Technology 38 (20), 5476e5483. Sa´nchez-Polo, M., Rivera-Utrilla, J., Me´ndez-Dı´az, J.D., Canonica, S., von Gunten, U., 2007. Photooxidation of naphthalenesulfonic acids: comparison between processes based on O3, O3/activated carbon and UV/H2O2. Chemosphere 68 (10), 1814e1820. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003. Environmental Organic Chemistry. John Wiley & Sons, New York, p. 1311. Sharpless, C.M., Linden, K.G., 2003. Experimental and model comparisons of low- and medium-pressure Hg lamps for the direct and H2O2 assisted UV photodegradation of N-nitrosodimethylamine in simulated drinking water. Environmental Science and Technology 37 (9), 1933e1940. Shemer, H., Kunukcu, Y.K., Linden, K.G., 2006. Degradation of the pharmaceutical metronidazole via UV, fenton and photofenton processes. Chemosphere 63 (2), 269e276. Simmons, M.S., Zepp, R.G., 1986. Influence of humic substances on photolysis of nitroaromatic compounds in aqueous systems. Water Research 20 (7), 899e904. Steelink, C., 2002. Investigating humic acids in soils. Anal. Chem. 74 (11), 327Ae333A.

w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 3 9 3 e4 0 3

Van Noort, P., Lammers, R., Verboom, H., Wondergem, E., 1988. Rates of triplet humic acid sensitized photolysis of hydrophobic compounds. Chemosphere 17 (1), 35e38. Wennmalm, A., Gunnarsson, B., 2005. Public health care management of water pollution with pharmaceuticals: environmental classification and analysis of pharmaceutical residues in sewage water. Journal of Drug Information 39, 291e297. Xu, G., Li, F., Wang, Q., 2007. Effect of humic acids on photodegradation of chloroacetanilide herbicides under UV irradiation. Journal of Environmental Science and Health. Part. B, Pesticides, Food Contaminants, and Agricultural Wastes 42 (2), 165e171. Zepp, R.G., Schlotzhauer, P.F., Sink, R.M., 1985. Photosensitized transformations involving electronic energy transfer in natural waters: role of humic substances. Environmental Science and Technology 19 (1), 74e81.

403

Zepp, R.G., Baughman, G.L., Schlotzhauer, P.F., 1981a. Comparison of photochemical behavior of various humic substances in water: I. Sunlight induced reactions of aquatic pollutants photosensitized by humic substances. Chemosphere 10 (1), 109e117. Zepp, R.G., Baughman, G.L., Schlotzhauer, P.F., 1981b. Comparison of photochemical behavior of various humic substances in water: II. Photosensitized oxygenations. Chemosphere 10 (1), 119e126. Zhan, M., Yang, X., Xian, Q., Kong, L., 2006. Photosensitized degradation of bisphenol A involving reactive oxygen species in the presence of humic substances. Chemosphere 63 (3), 378e386. Zheng, H., Ye, C., 2001. Photodegradation of acetochlor and butachlor in waters containing humic acid and inorganic ion. Bulletin Environmental Contamination Toxicology 67 (4), 601e608.