Lincomycin solar photodegradation, algal toxicity and removal from wastewaters by means of ozonation

ARTICLE IN PRESS WAT E R R E S E A R C H

40 (2006) 630– 638

Available at www.sciencedirect.com

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

Lincomycin solar photodegradation, algal toxicity and removal from wastewaters by means of ozonation Roberto Andreozzia, Marisa Canterinoa, Roberto Lo Giudiceb, Raffaele Marottaa,, Gabriele Pintob, Antonino Polliob a

Dipartimento di Ingegneria Chimica, Facolta` di Ingegneria, Universita` di Napoli ‘‘Federico II’’, p.le V.Tecchio 80, 80125– Napoli, Italia Dipartimento delle Scienze Biologiche, Sezione Biologia Vegetale, Universita` di Napoli ‘‘Federico II’’, Via Foria, 80125 Napoli, Italia

b

ar t ic l e i n f o

A B S T R A C T

Article history:

Antibiotic molecules have been reported among the xenobiotics present at trace levels in

Received 14 July 2005

sewage treatment plant (STP) effluents and aquatic environment. Lincomycin, one of the

Received in revised form

most used in clinical practices whose presence in the STP effluents has been often

16 November 2005

documented, is submitted to an extensive investigation to assess its persistence in the

Accepted 17 November 2005

environment and toxicity towards different algal strains. The possibility to remove the

Available online 6 January 2006

lincomycin from water by means of ozonation is demonstrated and a reduction of toxicity

Keywords:

of ozonated solutions on S. leopoliensis, with respect to untreated solutions containing this

Lincomycin

compound, is obtained even just for 1 h of treatment. Kinetic constants for the attack to

Antibiotics

lincomycin of ozone (from 1.53
105 M1 s1 at pH ¼ 3.0 and 4.93
105 M1 s1 at pH ¼ 6.7)

Photodegradation

and OH radicals (4.37
109 M1 s1 at pH ¼ 5.5 and 4.59
109 M1 s1 at pH ¼ 7.5) are also

Algal toxicity

evaluated.

Ozonation

& 2005 Elsevier Ltd. All rights reserved.

Semicontinuous apparatus

1.

Introduction

Recently pharmaceuticals have been identified as a new class of environmental pollutants. Hundred of tons of drugs are annually sold only in Europe for human and veterinary medicine thus causing the release of many of these species (or their metabolites) to the environment (Heberer, 2002; Halling-Sorensen et al., 1998). Among these, many antibiotic molecules have been found in sewage treatment plant (STP) effluents, in surface waters and in soils (Stackelberg et al., 2004, Andreozzi et al., 2004; Hirsch et al., 1999; Zilles et al., 2005). This is not surprising if one considers that the use of antibiotics is very large for human beings and livestock. Between 30% and 90% of administered dose of most antibiotics is generally excreted with the urine and, often, they are not destroyed by conventional wastewater treatments (Rang et al., 1999). Once these xenobiotics are released Corresponding author. Tel.: +39 081 7682968; fax: +39 081 5936936.

E-mail address: [email protected] (R. Marotta). 0043-1354/$ - see front matter & 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2005.11.023

to the environment, they may undergo transformation processes in surface waters under the absorption of solar light (direct photolysis) or through the intervention of photosensitizers (indirect photolysis) such as nitrate and humic acids (Zepp et al., 1985). Natural degradation rates of these compounds, depending on their chemical-physical properties, may be very fast or completely ineffective, thus influencing in a relevant way the environmental persistence of the molecules. Although for many pharmaceuticals no adverse effects on living organisms and environment have been shown (Cleuvers, 2003), recent studies demonstrated the capability of some antibacterial agents (flumequine, oxytetracycline) and antidepressants and their metabolites (fluoxetine, sertraline and norfluoxetine) to bioaccumulate in living organisms such as fish and mussels (Delepee et al., 2004; Brooks et al., 2003) whereas anti-inflammatory diclofenac has been reported to have harmful effects on rainbow trout fish

ARTICLE IN PRESS WAT E R R E S E A R C H

(Laville et al., 2004; Schwaiger et al., 2004). Moreover it has been put forward that the presence of antibiotics in the environment could lead to the selection of resistant bacterial strains (Tendencia and de la Pena, 2001; Boon and Cattanach, 1999). Some evidence has been also collected that this resistance can pass to humans (Rhodes et al., 2000); this fact could render completely useless the related molecules in clinical practice, although this topic is still under debate (Ayscough et al., 2000). Due to their importance for the clinical practice, it is clear that, even when data on the adverse effects on living organisms are collected, the use of these drugs can not be banned and measures devoted to limit their discharge to the environment have to be found. Treatment of sewage waters by means of new removal processes could be a suitable solution. Among these advanced oxidation processes can be proposed as tertiary treatment for municipal wastewaters (Rosenfeldt and Linden, 2004; Vogna et al., 2002). Recently some studies appeared in the literature on the possibility to remove antibiotics from wastewaters by photocatalysis (Addamo et al., 2005), combined chemical and biological oxidation (Arslan Alaton et al., 2004) and by using reverse osmosis and ultrafiltration (Shi-zhong et al., 2004). The present work aims to study the solar photodegradation and the effects of lincomycin (Fig. 1), one of the most used antibiotics whose occurrence in surface waters and STP effluents (Calamari et al., 2003), on simple living organisms such as algae along with the possibility of removing it from aqueous solution by means of ozonation. Lincomycin is one of the antibiotics of the lincosamines class. It is generally used as Lincomycin hydrochloride, a well-established antibiotic drug used in human and veterinary medicine. It is effective primarily against gram-positive pathogens, and not effective against gram-negative bacteria. It is currently employed against susceptible strains of streptococci, pneumococci, and staphlylococci which usually can also be treated with penicillin or erythromycin. The antibacterial activity of Lincomycin hydrochloride is similar to the group of macrolide antibiotics to which erythromycin belongs. First of all, the toxicity of lincomycin towards microalgae has been assessed using a battery of microalgae; the most sensitive ones have been chosen to carry out tests on liquid medium. The effect of pH and starting concentration on the ozonation process have been investigated and the total organic carbon (TOC) abatement also evaluated. A mathematical model has been developed and the kinetic constant for the ozone attack to the substrate estimated by using the data

HO CH3 N

O C N H

CH3 H C

HO

O OH CH3CH2CH2

SCH3

Fig. 1 – Chemical structure of lincomycin.

631

collected during the experiments in a semicontinuous laboratory apparatus. The efficiency of ozonation in reducing toxicity has been measured by liquid-phase algal bioassays.

2.

Experimental

2.1.

Solar irradiations set-up

Sunlight irradiation runs were performed in Naples (401N–141E) in glass disk-reactors placed horizontally in thermostated bath at 25 1C. Actinometry was carried out by using a solution of p-nitroacetophenone (PNAP, 2.0
105 M) and different concentrations of pyridine (Dulin and Mill, 1982). The pyridine concentration was chosen to adjust the quantum yield of the PNAP (fAct ¼ 0.0169 [pyridine]) to modify the rate of loss of PNAP to match the rate of consumption of lincomycin (Organisation for Economic Cooperation and Development (OECD), 2000). The molar extinction coefficients of lincomycin at different wavelength were determined by means of a UV–VIS diode array spectrophotometer (HP 8452A).

2.2.

Toxicological assessments

2.2.1.

Disk diffusion assay

To detect antialgal activity of lincomycin, agar diffusion tests with different microalgal strains were performed. Eight bluegreen and 10 green algal strains (Table 1) were grown in 100 ml Erlenmayer flasks containing 50 ml of Bold Basal Medium (Nichols, 1973). The cultures were incubated on a shaking apparatus at 24 1C and continuously illuminated with fluorescent lamps (Philips TDL 30w/55) at approximately 85 mE m–2 s–1. Growth was daily determined microscopically with a Bu¨rker blood-counting chamber and by monitoring absorbance at 550 nm with a Secoman 250 I spectrophotometer. Plates were prepared with the media described above containing 15.0 g of agar/l. For preparation of the test plates, 1 ml of each culture in the mid-log phase of growth, were spread with a sterile glass rod on the surface of the plates containing the agarized Bold Basal Medium. The plates were incubated for 24 h, under the same conditions of light and temperature above described. Then, lincomycin (20 ml of solutions corresponding to 4.5 or 9 mg/l) was added to sterile paper disks (Becton Dickinson, diameter 6 mm). Once dry, each paper disks was placed onto the agar surface containing the test alga, and the plates were incubated for a week under the same experimental conditions. The antibiotic activity was recorded as the diameter of clear zones of inhibited algal growth around the paper disk.

2.2.2. OH

C H

40 (20 06) 63 0 – 638

Bioassay on liquid medium

The stock solutions of lincomycin were prepared by dissolving a known quantity of the compound in water. No added solvent were used for their preparation. The solutions were stirred for 24 h in the dark at ambient temperature. The test solutions were prepared by mixing the appropriate volumes of the stock solutions and of the culture media. The pH of all dilutions prepared in culture media and used for liquid bioassays was 6.871.

ARTICLE IN PRESS 632

WA T E R R E S E A R C H

40 (2006) 630– 638

Table 1 – Inhibition of algal growth by lincomycin.+diameter of inhibition 7–14 mm; ++ diameter of inhibition 15–23 mm; +++ diameter of inhibition423 mm;—no inhibition Lincomycin Species Cyanochloronta Anabaena flos-aquae Calotrix membranacea Fremyella diplosiphon Gloeotrichia sp. Nostoc commune Phormidium autumnale Scytonema hofmanni Synechococcus leopoliensis Chlorophyta Ankistrodesmus braunii Bracteococcus minor v. desertorum Chlamydomonas reinhardtii Chlorosarcinopsis gelatinosa Chlorella vulgaris Coccomyxa simplex Pseudokirchneriella subcapitata Scenedesmus quadricauda Scotiellopsis terrestris Stichococcus bacillaris

Provenience

Strain n.

4.5 mg/l

9.0 mg/l

UTEX UTEX UTEX UTEX UTEX UTEX UTEX UTEX

1444 379 481 583 584 1580 2349 625

++ +++ ++ +++ +++

+++ +++ ++ ++ +++ +++

CCAP UTEX UTEX UTEX CCAP UTEX UTEX UTEX CCALA CCAP

202.7a 1386 89 1180 211/11b 274 1648 76 476 379/1c

+ -

++ -

Liquid growth inhibition tests against lincomycin were performed with the strains UTEX 1648 Pseudokirchneriella subcapitata, SAG 1020-1a Cyclotella meneghiniana and UTEX 625 Synechococcus leopoliensis. Inocula corresponding to 10 000 and 100 000 cells/ml from cultures of each strain in mid exponential phase were grown in 100 ml Erlenmeyer flasks containing either Bold Basal Medium (Nichols, 1973) for P. subcapitata and S. leopoliensis, or Bacillariophyta medium (Schlo¨sser, 1994) for C. meneghiniana. Lincomycin was tested at concentrations ranging from 625 to 40 mg/l. All tests were carried out in triplicate in axenic conditions at 2471 1C with lighting of 85 mE m2 s1, under continuous illumination. Control containing only distilled water and BBM was also tested. After 96 incubation, algal growth was determined by counting the cells with a Bu¨rker blood-counting chamber. During the course of the experiments (96 h) the algal cultures remained in the exponential phase of growth. For liquid growth inhibition tests the lowest observed effect concentration (LOEC), which differs significantly from that of the control, was determined by hypothesis tests. Dunnett’s tests, after verifying the Shapiro–Wilk’s test for normality and the Hartley’s test for homogeneity of variance, were used. Calculations were performed using TOXSTAT 3.0 software (Gulley et al., 1989). The no observed effect conservation (NOEC), corresponding to the next lower concentration to the LOEC, was deduced. EC50 for lincomycin (the concentration that cause 50% of the effect) was determined by regression using a log-logistic model.

2.3.

Ozonation runs

The ozonation experiments of the aqueous solutions of lincomycin were carried out in a semicontinuous stirred tank

thermostated at 25 1C. The experimental apparatus has been previously reported (Andreozzi et al., 1996). Before the run the solutions were buffered at desired pH with H3PO4, KH2PO4 and Na2HPO4 salts but keeping constant at 0.1 M the ionic strength value (with NaCl salt). The ozone concentration in the outlet gaseous stream was monitored by continuous UV measurements at 253 nm by means of an UV spectrophotometer equipped with a quartz cell (optical length ¼ 2.0
102 dm).

2.4.

UV/H2O2 system

The UV/H2O2 runs were carried out at 25 1C in an annular glass reactor equipped with a low-pressure lamp with a monochromatic wavelength emission at 254 nm (Andreozzi et al., 1999). The aqueous solutions were regulated at desired pH value with dilute HClO4 and NaOH mixtures. Samples were taken at fixed reaction times and analysed.

2.5.

Analytical measurements

2.5.1.

HPLC analysis

The substrates were analysed by HPLC (HP 1100 L, Hewlett Packard) equipped with a quaternary pump, a vacuum degasser, a diode array detector and a Synergi C12 4u POLAR column. The following mixtures with different ratio (90/10 for lincomycin and 70/30 for benzoic acid) of buffered aqueous solutions (water/methanol/phosphoric acid 500:25:2) and acetonitrile as mobile phase in isocratic mode (flow of 1.0 ml min1) were used. The detection wavelengths were 210 and 230 nm for lincomycin and benzoic acid, respectively.

ARTICLE IN PRESS WAT E R R E S E A R C H

2.5.2.

633

40 (20 06) 63 0 – 638

LC-MS analysis

2500

The LC system consisted of HPLC (Agilent, 1100). Chromatographic separation was performed on Synergi C12 4u MAX-RP column. Chromatographic conditions: a mixture of 90% formic acid (0.1% by volume) and 10% acetonitrile, flow rate 0.7 ml min1. The mass-spectrometric detection was performed on MSD Quad VL (Agilent Mass Spectrometer) equipped with electrospray ionization. MS data were acquired in ESI+ mode (capillary temperature 350 1C; source voltage 3.5 kV, drying N2 gas flow 11 l/min). Collision energy to produce the desired quantity of [M+H]+ molecular ion was individually optimized. The pH of aqueous solutions was measured by using a pHmeter with a glass pH electrode. The total organic carbon was monitored by means of TOC analyzer (Shimadzu 5000 A).

half-life time (day)

2000

1500

1000

500

0

2.6.

Chemicals

1

2

3

4

season Lincomycin, benzoic acid and hydrogen peroxide (not stabilized) were purchased from ICN, Carlo Erba and Fluka, respectively. Commercial humate sodium salt (from Aldrich) was employed as a substitute for aquatic humic acids (Zepp et al., 1981).

3.

Fig. 2 – Influence of the latitude (K: 201N, ’: 301N, m: 401N, E: 501N) and season (1: spring, 2: summer, 3: fall, 4: winter) on the photolysis of lincomycin at pH ¼ 5.5 and T ¼ 25 1C.

2000

Results and discussion

3.1. Solar photodegradation: determination of quantum yield for direct photolysis half-life time (day)

1600

As reported in the literature (EPA, 1996), the primary quantum yield for direct photolysis of a substance under solar irradiation is readily calculated by the following equation: fLin ¼ fAct

kLin Sðel Ll ÞAct , kAct Sðel Ll ÞLin

(1)

where kLin and kAct are the pseudo-first order kinetic constants for lincomycin and PNAP actinometer, fAct is the quantum yield for actinometer photolysis calculated by the equation fAct ¼ 0.069 [pyridine] (Dulin and Mill, 1982), el (M1 cm1) and Ll (103 Einstein cm–2 day–1) (Zepp and Cline, 1977) are, respectively, molar absorption coefficients and average daily solar irradiance over wavelength interval centred at wavelength l. More precisely fLin evaluated through the Eq. (1) is an overall quantum yield being lincomycin capable of giving rise to different species according to the pH of the solution and its pKa value (7.79) (Qiang and Adams, 2004). Quantum yields of 1.1
104 and 1.3
104 have been calculated at pH 5.5 and 7.5, respectively. Half-life times (t1=2 ) for lincomycin at both investigated pH are thus calculated at varying the season and latitude based on estimated quantum yield: t1=2 ¼

ln 2 . fLin Sðel Ll ÞLin

(2)

In Figs. 2 and 3 the values found with Eq. (2) are shown. It is evident from these diagrams that lincomycin is characterized by a certain refractoriness to undergo photolytic reactions with a half-life time in the worst conditions (winter, 501N latitude) up to 2033 days (at pH 5.5) and 1760 days (at pH 7.5).

1200

800

400

0 1

2

3

4

season Fig. 3 – Influence of the latitude (K: 201N, ’: 301N, m: 401N, E: 501N) and season (1: spring, 2: summer, 3: fall, 4: winter) on the photolysis of lincomycin at pH ¼ 7.5 and T ¼ 25 1C.

3.2.

Solar photodegradation: indirect photolysis

In Fig. 4, the concentration decays of lincomycin recorded during photolytic solar experiments in the presence of photosensitizers (nitrate and humic acids) are compared with the data collected during direct photolysis. Both nitrates and humic acids markedly increase the rate of lincomycin photolysis with respect to the case in which no photosensitizers were added. This effect is independent from the pH and is approximately of the same amplitude when 15 mg/l of NO 3 and 5 mg/l of humic acid were used. For both the species the

ARTICLE IN PRESS 634

WA T E R R E S E A R C H

results obtained during these runs well agree with those reported by others researchers for different organic molecules (Mack and Bolton, 1999; Zepp et al., 1981).

3.3.

Hydrolytic degradation

After 22 days under dark conditions only 10.4% and 9.00% of lincomycin disappeared at pH 5.5 and 7.5, respectively. This data undoubtedly indicate that hydrolytic degradation is just a minor pathway for lincomycin disappearance in the aquatic environment.

3.4.

Toxicity of lincomycin on algae

Lincomycin toxicity toward algae was preliminarily tested with the paper disk bioassay on ten Chlorophyta and eight Cyanochloronta (Table 1). Lincomycin affected mainly the growth of Cyanobacteria, but some of them as C. membranacea and S. hofmanni were insensitive to the antibiotic, probably due to the large gelatinous layer enclosing the cells. The strain UTEX 625, belonging to Synechococcus leopoliensis, showed a marked sensitivity to lincomycin, together with P. autumnale and F. diplosiphon. Among the Chlorophycean strains, most were completely insensitive to lincomycin, but, P. subcapitata (formerly Selenastrum capricornutum) was

1.00

0.80

C/Co

0.60

0.40

0.20

0.00 0

40

80 120 Time (hours)

160

200

Fig. 4 – Effect of nitrate and humic acids on the photolysis of lincomycin at pH 5.5 (full symbols) and 7.5 (empty symbols): no photosensitizers (K, J); 15 mg/l of NO–3 (m, n); 5 mg/l of humic acids (’, &).

40 (2006) 630– 638

sensitive, confirming that this strain is particularly useful to assess the toxicity of a large array of xenobiotics. On the basis of these results the toxicity of lincomycin towards algae was tested in broth on the two most sensitive strains, namely the cyanobacterium S. leopoliensis and the chlorophycean P. subcapitata. In addition lincomycin was also assayed on the Bacillariophyta Cyclotella meneghiniana, strain SAG 1020-1a, since diatoms form a relevant part of freshwater phytoplankton. In Table 2 the results of these tests are reported. Chronic NOECs of lincomycin were determined for the three algae, and the relative 96-h EC50 found were considered as acute measured end-points, as suggested by the European Technical Guidance on risk assessment (1996), in the case of algal studies. The cyanobacterium S. leopoliensis confirmed to be the most sensitive organism, either on the basis of EC50 than on NOECs. The range of acute EC50 values varied from 195 mg/l for S. leopoliensis to 1510 mg/l for P. subcapitata and 1630 mg/l for C. meneghiniana, whereas chronic NOEC data were in the range from 78 (S. leopoliensis) to 781 mg/l (C. meneghiniana).

3.5.

Lincomycin oxidation by means of ozone

Above-reported results indicate that lincomycin exerts toxic effects on algae and is characterized by a certain persistence in the aquatic environment due to its low tendency to undergo photolytic degradation and hydrolytic reaction. Moreover, although no specific experiments have been performed during the present investigation, a reduced biodegradability of this species can be inferred from its presence in STP effluents. Therefore there is an interest to collect indications about the possibility of removing lincomycin from water through new non-biological treatments to prevent or reduce its discharge into the environment. Among non-conventional treatment processes, the ozonation has already reached a degree of technological development which makes possible its utilization at full industrial scale (Rice, 1997). At first sight, the use of ozone could seem not particularly recommendable for lincomycin removal. In fact, this antibiotic does not present in its molecular structure any unsaturated site such as double bonds or aromatic rings, which are well known to be capable to readily react with ozone (Bailey, 1978). However, a more careful analysis evidences the presence of two nucleophilic centres, –S–CH3 and 4N–, towards which the ozone attack can be directed, provided that, for adopted experimental conditions, they are unprotonated. A pKa of 7.79 has been recently measured for lincomycin (Qiang and Adams, 2004) even though in the quoted paper this value has not been attributed to any specific group in the molecule itself. The comparison of

Table 2 – Acute (EC50) and chronic (NOEC) toxicity test results of lincomycin in liquid medium Species Synechococcus leopoliensis Pseudokirchneriella subcapitata Cyclotella meneghiniana

Exposure duration (h)

Assesment endpoint

NOEC (mg/l)

EC50 (mg/l)

96 96 96

Growth Growth Growth

78 156 781

195 1510 1630

ARTICLE IN PRESS WAT E R R E S E A R C H

literature data, calculated using Advanced Chemistry Development Software Solaris V 4.67, for N-methylpyrrolidine [12094-5] (pKa ¼ 10.5) and methyl-N-methyl-L-prolinate [27957-911] (pKa ¼ 8.48) allows to ascribe the pKa reported for lincomycin to the tertiary aminogroup in its molecule. Moreover, as documented by others (Perez Rey et al., 1987) ozone molecule may attack C-H bonds of glucosidic ring:

C H + O3

C

OH

O

OH

ring cleavage

O OH

and—at highest pHs—the activation of the radical mechanism of ozone decomposition makes an OH-mediated oxidation of lincomycin possible. For adopted experimental conditions at pH ¼ 5.5 a complete disappearance of lincomycin was observed after 2 min of ozonation with a specific ozone consumption equal to 1. The extension of the treatment up to 180 min resulted into a low degree of mineralization (o10%). LC-MS analysis indicates the presence, in a sample collected after 30 s of ozonation, of a peak (Rt ¼ 5.30 min) for which the following signals (m/z) were recorded in the mass spectrum: 423.2 (424.2, 425.1), 359.1 (360.2), 126.1 (127). A careful examination of these data allowed identifying this peak as due to lincomycin S-oxide:

HO CH 3 N

O C N H

H C

CH 3 HO OH

C H O

OH CH 3 CH 2 C H 2

S O

CH 3

which forms from the direct ozone attack on sulfur atom of the substrate. Unfortunately the unavailability of the chemical standard for this species prevented any quantitative evaluation. As expected due to the low pH of the solution, no intermediates containing in their structure N-oxide fragment have been identified, being the tertiary amino group protonated for adopted conditions. Further LC-MS runs on samples withdrawn from the reactor at higher ozonation times evidenced the presence of a second peak (Rt ¼ 9.1 min) with the following m/z ratios: 439.2 (440.2), 375.1 (376.1), 126.1 (127.1) indicating a further reaction intermediate. The [M+H]+ signal at 439.2 m/z and that [[M+H]+HSCH3O] at 375.1 m/z are consistent with the presence in the structure of this oxidation intermediate of an additional oxygen atom with respect to lincomycin S-oxide not bonded to sulfur atom. These data just allow ruling out the formation of lincomycin sulfone the identification of the involved species being not possible and requiring further information. Low degrees of mineralization and the presence in the first stage of ozonation in the reacting solution of intermediates whose structures are very similar to the parent compound prompted an extension of the investigations at higher pH.

635

40 (20 06) 63 0 – 638

Measurements performed during a successive run at pH 7.5 indicated a TOC decrease of just 50% after 3 h of ozonation treatment. Due to the incomplete mineralization observed even for long treatment times, an assessment of the real detoxification capability of ozonation has been carried out. Toxicity removal by ozonation was tracked through the use of toxicity tests in liquid medium performed on S. leopoliensis, the most sensitive alga. Comparison of bioassay results on this alga (at both pHs 5.5 and 7.5) before and after treatment with O3 was utilized to demonstrate that this process efficiently removed lincomycin toxicity. All the attempted treatments were equally effective to completely detoxify the solution, even at the shortest time (1 h). The removal of toxicity was independent of the pH of lincomycin solution, and no stimulation of growth was observed at any treatment.

4.

Kinetic assessments

For the scale-up and proper design of ozonation reactors and the choice of the best operating conditions the knowledge of reaction kinetics is required. Therefore, both kinetic constants for direct ozone attack and that of OH radicals to lincomycin need to be evaluated.

4.1. Determination of the kinetic constant for ozone direct attack to lincomycin Ozone can react with organic molecules directly as an electrophile and indirectly through the generation of OH radicals which are kinetically more reactive than ozone, being able to attack organic species through radical addition, hydrogen abstraction, and electron transfer. In a first phase the experiments were carried out by adding to the reacting solutions t-butyl alcohol, a well known radical scavenger species (Ma and Graham, 2000). In this way the radical mechanism of oxidation was strongly inhibited, thus making possible to kinetically characterize just the direct (ionic) attack of ozone to the molecule. The results collected during these runs were used in a kinetic analysis performed by means of a model developed by coupling an overall reaction: k

lincomycin þ zO3 ! products

(3)

to a proper fluidynamic submodel previously reported elsewhere (Andreozzi et al., 1996). The stoichiometric coefficient z in the equation (z) can be considered constant (z ¼ a) or a simple linear function of time (z ¼ a+bt) in consideration of the fact that ozone can react at starting time only with lincomycin but during the process there is a formation of reaction products whose still react with ozone. The parameters a and b can be estimated along with the kinetic constant k by means of a suitable optimization procedure which minimizes the differences between experimental lincomycin concentrations and those calculated by the model. The values estimated for k plotted against the pH in Fig. 5. These values were obtained, for each pH value, with starting lincomycin and ozone concentrations equal to 0.5

ARTICLE IN PRESS 636

WA T E R R E S E A R C H

40 (2006) 630– 638

0.035

1E+6 9 8

0.030

7 6

0.025 Concentration (mM)

5

k (1/M*s)

4

3

2

0.020 0.015 0.010 0.005 0.000

1E+5 2

3

4

5

6

0

7

50

pH Fig. 5 – Kinetic constant values for direct ozone attack to lincomycin against pH. Starting lincomycin concentration: 0.5 mM; starting ozone concentration: 0.4 mM.

100 Time (s)

150

200

Fig. 6 – Concentration profiles of lincomycin (K) and benzoic acid (’) with UV/H2O2 at pH ¼ 5.5 in presence of terz-butanol (10 mM).

0.030

4.2. Determination of the kinetic constant for OH attack to lincomycin When unscavenged ozonation of an organic species is carried out at pH45.0, the disappearance of the studied molecule is due to both the direct ozone attack and that of OH radicals arising from the radical ozone decomposition. The kinetic constant which regulates the attack of these radicals to the organic molecule can be more easily determined by means of H2O2/UV experiments with respect to ozonation runs. In fact in this case OH radicals are generated through the UV cleavage of H2O2 and the constant of HO attack can be estimated by means of the comparison method (Xiong and

0.025 Concentration (mM)

and 0.4 mM, respectively. As it is evident from the figure, constant values are found for pH lower than 6.0 as expected on the basis of reported pKa of lincomycin (7.79). For pH higher than 6.0 the reactivity of the molecule with ozone increases. Higher standard deviations are associated with kinetic constants estimated in this pH range, probably due to the reduced efficiency of radical scavenger to prevent the occurrence of the radical mechanism. The simple assumption z ¼ constant ¼ 1.0 holds for pH less than 6.0 whereas a linear dependence upon the time, z ¼ a+bt, was used for the runs at pHX6.0. The values found during the present investigation fairly agree with those reported by Qiang et al. (2004) who studied the lincomycin reaction kinetics by means of a stopped-flow apparatus. The use of an instrument dedicated to kinetic assessments enabled these authors to extend the investigations in a more wide range of pH than in the present work. However, it is noteworthy to stress that the procedure adopted here does not require any special experimental apparatus but just the adoption of a proper mathematical model for the data analysis.

0.020

0.015

0.010

0.005

0.000 0

20

40 Time (s)

60

80

Fig. 7 – Concentration profiles of lincomycin (K) and benzoic acid (’) with UV/H2O2 at pH ¼ 7.5 in presence of terz-butanol (10 mM).

Graham, 1992). Benzoic acid was used as reference substance during the experiments. In Figs. 6 and 7 the concentration decays for lincomycin and benzoic acid are reported for the run at pH 5.5 and 7.5, respectively. The following equations are adopted to describe the concentration decay of both the reactants during the ozonation: d½Lin ¼ kLin ½HOd ½Lin, dt

(4)

d½BA ¼ kBA ½HOd ½BA. dt

(5)

ARTICLE IN PRESS WAT E R R E S E A R C H

If Eq. (4) is divided by (5) and integrated, the following relationship is obtained: Ln

½Lin k ½BA ¼ Lin Ln . ½Lin0 kBA ½BA0

(6)

The ratio between the two kinetic constants kLin/kBA is thus derived by plotting the natural logarithm of the ratio [Lin]/ [Lin]0 against that of [BA]/[BA]0. From the two values obtained at pH 5.5 (kLin/kBA ¼ 0.728) and 7.5 (kLin/kBA ¼ 0.765) and that taken from the literature for the kinetic constant of HO. attack to benzoic acid, 6.0
109 M1 s1 (Buxton et al., 1988), the following data were calculated: kLin ¼ 4.37
109 M1 s1 at pH 5.5 and kLin ¼ 4.59
109 M1 s1 at pH 7.5. These values can be used along with those found for direct ozone attack to simulate the substrate disappearance during an unscavenged ozonation. Mathematical models which allow to simulate the system behaviour during unscavenged ozonation can be found elsewhere (Glaze and Kang, 1989).

5.

Conclusions

Lincomycin was submitted to the photodegradative and hydrolytic experiments and its toxicity towards algae measured. The results of these studies indicate that this species is characterized by a certain persistence in the environment and exerts toxic effects on different algal strains. Half-life times of 2033 and 1760 days were calculated for lincomycin solar photodegradation at pH 5.5 and 7.5, respectively, for winter and 501N latitude whereas EC50’s varying from 195 mg/l for S. leopoliensis to 1510 mg/l for P. subcapitata and 1630 mg/l for C. meneghiniana were estimated. The possibility of removing this antibiotic from water by means an ozonation process was also investigated. Ozone was demonstrated to be capable of promptly attacking lincomycin molecules, the resulting samples being less toxic on S. leopoliensis with respect to the parent compound even just for 1 h of treatment. Kinetic constants for the attack to lincomycin of ozone (from 1.53
105 M1 s1 at pH 3.0 to 4.93
105 M1 s1 at pH 6.7) and OH radicals (4.37
109 M1 s1 at pH 5.5 and 4.59
109 M1 s1 at pH 7.5) were also evaluated. R E F E R E N C E S

Addamo, M., Augugliaro, V., Di Paola, A., Garcia-Lopez, E., Loddo, V., Marci, G., Palmisano, L., 2005. Removal of drugs in aqueous systems by photoassisted degradation. J. Appl. Electrochem. 35 (7-8), 765–774. Andreozzi, R., Caprio, V., Insola, A., Tufano, V., 1996. Measuring ozonation rate constants in gas-liquid reaction under the kinetic-diffusional transition regime. Chem. Eng. Commun. 143, 195–197. Andreozzi, R., Caprio, V., Insola, A., Marotta, R., 1999. The oxidation of metol (N-methyl-p-aminophenol) in aqueous solution by UV/H2O2 photolysis. Water Res. 34 (2), 463–472. Andreozzi, R., Caprio, V., Ciniglia, C., de Champdore, M., Lo Giudice, R., Marotta, R., Zuccato, E., 2004. Antibiotics in the environment: occurrence in Italian STPs, fate, and preliminary assessment on algal toxicity of amoxicillin. Environ. Sci. Technol. 38 (24), 6832–6838.

40 (20 06) 63 0 – 638

637

Arslan Alaton, I., Dogruel, S., Baykal, E., Gerone, G., 2004. Combined chemical and biological oxidation of penicillin formulation effluent. J. Environ. Manage. 73 (2), 155–163. Ayscough, N.J., Fawell, J., Franklin, G., Young, W., 2000. Review of Human Pharmaceuticals in the Environment, Report P390, UK Environment Agency, Bristol, UK Bailey, P.S., 1978. Ozonation in Organic Chemistry. Academic Press, New York. Boon, P.I., Cattanach, M., 1999. Antibiotic resistance of native and faecal bacterial isolated from rivers, reservoirs and sewage treatment facilities in Victoria, South-Eastern Australia. Lett. Appl. Microbiol. 28, 164–168. Brooks, B.W., Foran, C.M., Richards, S.M., Weston, J.T., Philip, K., Stanley, J.K., Solomon, K.R., Slattery, M., La Point, T.W., 2003. Aquatic ecotoxicology of fluoxetine. Toxicol. Lett. 142 (3), 169–183. 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 (d OH=d O ) in aqueous solution. J. Phys. Chem. Ref. Data 17 (2), 513–886. Calamari, D., Zuccato, E., Castiglioni, S., Bagnati, R., Fanelli, R., 2003. Strategic survey of therapeutic drugs in the Rivers Po and Lambro in Northern Italy. Environ. Sci. Technol. 37, 1241–1248. Cleuvers, M., 2003. Aquatic ecotoxicity of pharmaceuticals including the assessment of combination effects. Toxicol. Lett. 142 (3), 185–194. Delepee, R., Pouliquen, H., Le Bris, H., 2004. The bryophyte Fontinalis antipyretica Hedw. bioaccumulates oxytetracycline, flumequine and oxolinic acid in the freshwater environment. Sci. Total Environ. 322 (1-3), 243–253. Dulin, D., Mill, T., 1982. Development and evaluation of sunlight actinometers. Environ. Sci. Technol. 16, 815–820. EPA, 1996. OPPTS 835.2210—Fate, transport and transformation test guidelines. Direct photolysis rate in water by sunlight. United States Environmental Protection Agency, Washington, DC. Glaze, W.H., Kang, J.W., 1989. Advanced oxidation processes. Description of a kinetic model for the oxidation of hazardous materials in aqueous media with ozone and hydrogen peroxide in a semibatch reactor. Ind. Eng. Chem. Res. 28 (11), 1573–1580. Gulley, D.D., Boelter, A.N., Bergman, H.L., 1989. Toxstat/Datasys Statistical Software, Relase 3.0. Department of Zoology and Physiology, University of Wyoming, Laramie, WY. Halling-Sorensen, B., Nielsen, S.N., Lanzky, P.F., Ingerslev, F., Lutzhorft, H.C., Jorgensen, S.E., 1998. Occurrence, fate and effects of pharmaceutical substances in the environment- a review. Chemosphere 36, 357–393. Heberer, T., 2002. Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment: a review of recent research data. Toxicol. Lett. 131 (1/2), 5–17. Hirsch, R., Ternes, T., Hebere, K., Kratz, K.L., 1999. Occurrence of antibiotics in the aquatic environment. Sci. Total Environ. 225 (1/2), 109–118. Laville, N., Ait-Aissa, S., Gomez, E., Casellas, C., Porcher, J.M., 2004. Effects of human pharmaceuticals on cytotoxicity, EROD activity and ROS production in fish hepatocytes. Toxicology 196 (1/2), 41–55. Ma, J., Graham, N.J.D., 2000. Degradation of atrazine by manganese-catalyzed ozonation-influence of radical scavengers. Water Res. 34 (15), 3822–3828. Mack, J., Bolton, J.R., 1999. Photochemistry of nitrite and nitrate in aqueous solution: a review. J. Photochem. Photobiol. A: Chem. 128, 1–13. Nichols, H.W., 1973. Growth media—freshwater. In: Stein, J.R. (Ed.), Handbook of Phycological Methods. Cambridge University Press, Cambridge, pp. 7–24.

ARTICLE IN PRESS 638

WA T E R R E S E A R C H

Organisation for Economic Cooperation and Development (OECD), 2000. Guidance document for testing of chemicals. Phototransformation of Chemicals in water—Direct and indirect photolysis. OECD Environmental Health and Safety Publication. Office for Official Publications of the European Communities. Technical guidance document in support of council directive 93/67/EEC on risk assessment for new notified substances and commission regulation (EC) 1488/94 on risk assessment for existing substances. 1996, Luxembourg, Luxembourg. Perez Rey, P., Gomez Moraleda, M., Ramos Lazcano, R., 1987. Ozone reactions with carbohydrates in aqueous medium. In: Proceedings of the Eighth Ozone World Congress of IOA, 15–18 September 1987, Vol. 2, pp. E107–E127. Qiang, Z., Adams, C., Surampalli, R., 2004. Determination of ozonation rate constants for lincomycin and spectinomycin. Ozone: Sci. Eng. 26, 1–13. Qiang, Z., Adams, C., 2004. Potentiometric determination of acid dissociation constants (pKa) for human and veterinary antibiotics. Water Res. 38, 2874–2890. Rang, H.P., Dale, M., Ritter, J.M., 1999. Pharmacology. Churcill Livingstone, Edinburgh. Rhodes, G., Huys, G., Swings, J., McGann, P., Hiney, M., Smith, P., Pickup, R.W., 2000. Distribution of oxytetracycline resistance plasmids between aeromonads in hospital and aquaculture environments: implication of Tn1721 in dissemination of the tetracycline resistance determinant tet A. Appl. Environ. Microbiol. 66 (9), 3883–3890. Rice, R.G., 1997. Applications of ozone for industrial wastewater treatment—a review. Ozone: Sci. Eng. 18 (6), 477–515. 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. Environ. Sci. Technol. 38 (20), 5476–5483. Schlo¨sser, U.G., 1994. SAG—Sammlung von Algenkulturen at the University of Go¨ttingen. Catalogue of strains 1994. Botan. Acta 107, 111–186.

40 (2006) 630– 638

Schwaiger, J., Ferling, H., Mallow, U., Wintermayr, H., Negele, R.D., 2004. Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part I: histopathological alterations and bioaccumulation in rainbow trout. Aquatic Toxicol. 68, 141–150. Shi-zhong, L., Xiao-yan, L., Dian-zuo, W., 2004. Membrane (RO-UF) filtration for antibiotic wastewater treatment and recovery of antibiotics. Separ. Purific. Technol. 34 (1/3), 109–114. Stackelberg, P.E., Furlong, E.T., Meyer, M.T., Zaugg, S.D., Henderson, A.K., Reissman, D.B., 2004. Persistence of pharmaceutical compounds and other organic wastewater contaminants in a conventional drinking-water-treatment plant. Sci. Total Environ. 329 (1/3), 99–113. Tendencia, T.A., de la Pena, L.D., 2001. Antibiotic resistance of bacteria from shrimp ponds. Aquaculture 195, 193–204. Vogna, D., Marotta, R., Napoletano, A., Andreozzi, R., d’Ischia, M., 2002. Advanced oxidation of the pharmaceutical drug diclofenac with UV/H2O2 and ozone. Water Res. 38 (2), 414–422. Xiong, F., Graham, N.J., 1992. Rate constants for herbicide degradation by ozone. Ozone: Sci. Eng. 14, 283–301. Zepp, R.G., Cline, D.M., 1977. Rates of direct photolysis in aquatic environment. Environ. Sci. Technol. 11 (4), 359–366. Zepp, R.G., Baughman, G.L., Schlotzhauer, P.F., 1981. Comparison of photochemical behaviour of various humic substances in water: I sunlight induced reactions of aquatic pollutants photosensitized by humic substances. Chemosphere 10, 109–117. Zepp, R.G., Schlotzhauer, P.F., Sink, R.M., 1985. Photosensitized transformations involving electronic energy transfer in natural waters: role of humic substances. Environ. Sci. Technol. 19 (1), 74–81. Zilles, J., Shimada, T., Jindal, A., Robert, M., Raskin, L., 2005. Presence of macrolide-lincosamide-streptogramin B and tetracycline antimicrobials in swine waste treatment processes and amended soil. Water Environ. Res. 77 (1), 57–62.