Article pubs.acs.org/est
Kinetics and Modeling of Degradation of Ionophore Antibiotics by UV and UV/H2O2 Hong Yao,†,‡ Peizhe Sun,‡ Daisuke Minakata,‡,§ John C. Crittenden,‡,§ and Ching-Hua Huang*,‡ †
Department of Municipal and Environmental Engineering, Beijing Jiaotong University, Beijing 100044, People’s Republic of China School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States § Brook Byers Institute for Sustainable Systems, Georgia Institute of Technology, Atlanta, Georgia 30332, United States ‡
S Supporting Information *
ABSTRACT: Ionophore antibiotics (IPAs), one of the major groups of pharmaceuticals used in livestock industry, have been found to contaminate agricultural runoff and surface waters via land application of animal manures as fertilizers. However, limited research has investigated the means to remove IPAs from water sources. This study investigates the degradation of IPAs by using ultraviolet (UV) photolysis and UV combined with hydrogen peroxide (UV/H2O2) advanced oxidation process (AOP) under low-pressure (LP) UV lamps in various water matrices. Three widely used (monensin, salinomycin, and narasin) and one model (nigericin) IPAs exhibit low light absorption in the UV range and degrade slowly at the light intensity of 3.36 × 10−6 Einstein·L−1·s−1 under UV photolysis conditions. However, IPAs react with hydroxyl radicals produced by UV/H2O2 at fast reaction rates, with second-order reaction rate constants at (3.49−4.00) × 109 M−1·s−1. Water matrix constituents enhanced the removal of IPAs by UV photolysis but inhibited UV/H2O2 process. A steady-state kinetic model successfully predicts the impact of water constituents on IPA degradation by UV/H2O2 and determines the optimal H2O2 dose by considering both energy consumption and IPA removal. LC/MS analysis of reaction products reveals the initial transformation pathways of IPAs via hydrogen atom abstraction and peroxidation during UV/H2O2. This study is among the first to provide a comprehensive understanding of the degradation of IPAs via UV/H2O2 AOP.
■
SAL in manure and soil was reported by several studies.9−12 In contrast, few studies have investigated abiotic degradation of IPAs. An earlier study indicated that SAL is relatively resistant to abiotic degradation; it was reported that under conditions such as white fluorescent light exposure or in solution pH ranging from 5 to 9, SAL was stable over 40 days of experimental period.1 In sterilized soil, Sassman and Lee12 reported that half of MON decreased in the first 5 days, but remained at the same concentration afterward for 40 days. Sun et al. reported that MON, SAL, and NAR were stable at pH 7− 10, but prone to acid-catalyzed hydrolysis at acidic conditions.13 Based on IPAs’ significant usage, contamination of water sources, potential environmental persistence, and toxicity of concern,14−16 it is necessary to investigate treatment processes that can quickly and effectively remove IPA contaminants from water. Advanced oxidation processes (AOPs) that produce highly reactive and nonselective electrophiles such as hydroxyl radicals are attractive and promising methods for removal of trace and nonbiodegradable residual organic contaminants in a wide range of waters.17−21 Hydroxyl radicals can yield complete
INTRODUCTION Ionophore antibiotics (IPAs) are antiparasitic compounds commonly used as feed additives for coccidiosis prevention and growth promotion in livestock production throughout the world.1,2 In a survey conducted by the Animal Health Institute (AHI) in 1999, IPAs along with arsenicals were estimated to be the top antibiotics consumed in the U.S. In 2012, the U.S. Food and Drug Administration (FDA) reported the annual sales of 3 821 138 kg of IPAs in the U.S., which was the second top sold pharmaceutical group next to tetracyclines.3,4 The commonly used IPAs in agricultural applications include monensin (MON), salinomycin (SAL) and narasin (NAR) (Table 1). An increasing number of studies have reported the occurrence of IPAs in manure, sewage, soil, sediments, surface water, and groundwater in recent years.5−7 For example, Watkinson et al. detected MON and SAL in different waters including hospital discharge, wastewater treatment plant (WWTP) effluent, environmental surface waters, and drinking waters, among which MON and SAL were detected at 94% and 21% frequencies, respectively, in environmental surface waters, and occasionally detected in WWTP effluent.5 Kim and Carlson found that IPAs were likely accumulated in sediments near agricultural fields.8 To date, however, information regarding the fate and transformation of IPAs in different environments is still quite limited. Potential biodegradation of IPAs such as MON and © 2013 American Chemical Society
Received: Revised: Accepted: Published: 4581
December 26, 2012 March 30, 2013 April 9, 2013 April 9, 2013 dx.doi.org/10.1021/es3052685 | Environ. Sci. Technol. 2013, 47, 4581−4589
Environmental Science & Technology
Article
Table 1. Chemical Properties and Structures of IPAs Investigated
a
Sodium adducts of IPAs on ESI LC/MS were used for quantification. bFranco et al. (2009). cEuropean Food Safety Authority (2004).
Table 2. Characteristics of the Water Samplesa water DI SW WW
pH 7.0 7.2 7.4
DOC 0.93) between ln([IPA]/[IPA]0) and UV fluence or irradiation time, indicating the reactions followed pseudofirst-order kinetics. The first-order rate constants of IPA degradation at 254 nm with receiving light intensity of 3.36 × 10−6 Einstein·L−1·s−1 at pH 7.0−7.4 ranged from 1.90 × 10−3 to 6.97 × 10−3 min−1, 8.25 × 10−3 to 1.09 × 10−2 min−1, 9.3 × 10−3 to 1.23 × 10−2 min−1 in DI, SW and WW, respectively (Table 3). Degradation of IPAs was faster in SW and WW than in DI. Compared with the fluence-based rate constants of other commonly used pharmaceuticals,30,31 IPAs degraded 1−2 orders of magnitude slower under UV 245 nm irradiation. In the DI water buffered by phosphate, IPAs were degraded by direct UV photolysis. In direct photolysis, compounds absorb photons and electrons are excited to a higher energy state, which enables structural transformation and leads to degradation of parent compounds. Hence, the overall transformation of IPAs by direct photolysis under UV 254 nm depends on (i) the ability of IPA to absorb light at 254 nm (i.e., molar absorption coefficient), and (ii) the possibility of structural transformation after light absorption (i.e., quantum yield). The ability to absorb light depends on electron distribution on the molecule. In general, sigma bonds have low ability to absorb light above 200 nm, whereas pi bonds and especially conjugated pi systems increase a compound’s ability to absorb UV light at 254 nm by lowering energy barrier between ground and excited states.32 The IPAs (Table 1) examined in this study consist of mostly sigma bonds (C−H, C−C, C−O, and O−H) except for a carboxylic carbonyl group at one end. SAL and NAR contain one CC bond and one CO bond additionally than MON and NIG. The aforementioned structural features result in IPAs’ very low light absorption at 254 nm. However, removal of parent IPAs by direct photolysis in DI was still observed, suggesting high quantum yields for IPAs. The obtained direct photolysis rate constants for the four IPAs were within a small range (Table 3), owning to their very similar structures. The faster reaction rate of SAL and NAR than MON is likely due to their extra double bonds, which facilitate direct photolysis. This hypothesis is consistent with the stronger UV absorbance of SAL than MON quantified in a methanol medium (SI Figure S2). Under the same UV irradiation intensity, degradation of IPAs in SW and WW was faster than in DI. Indirect photolysis in SW and WW likely contributed to the faster degradation. Indeed, in
mM sodium phosphate and SW and WW were not adjusted for pH. Reaction was initiated by exposing the solution to UV light. A sample aliquot was taken at each time interval and injected into a 2 mL amber glass vial. The vial contained a methanol/0.1 M Na2HPO4 mixture (v:v = 1:2) for improving the consistency and accuracy of IPA detection. The LP-UV photolysis of IPAs was investigated at UV fluences from 0 to 6.05 × 10−4 Einstein·L−1 to determine the fluence- and time-based rate constants. UV/H2O2 AOP. The UV/H2O2 experiments were conducted in DI, SW and WW using similar procedures described above with addition of H2O2 to the sample prior to UV exposure. IPAs at 0.8−3.0 μM and 30 mg·L−1 H2O2 were spiked into different water matrices. Control experiments with H2O2 were conducted without UV exposure. Sample aliquots were taken at different exposure times and placed in amber vials as described above for analysis. A competition kinetic approach was used to determine the rate constants of IPAs with hydroxyl radicals using para-chlorobenzoic acid (pCBA) as the reference compound.28,29 Experiments were conducted by spiking H2O2 into the reactor containing pCBA (1 μM) and individual IPA (1 μM) in DI and then the samples were exposed to UV light. A high H2O2 concentration (3000 mg·L−1) was used to ensure excess H2O2 so that a stable concentration of hydroxyl radicals was achieved during the competition kinetic reaction period. Organic Compound Analysis. IPAs in the reaction solutions were analyzed by an Agilent 1100 Series high performance liquid chromatography mass spectrometry (HPLC/MS) system (Agilent, Palo Alto, CA) with a reversephase Ascentis RP-amide column (2.1 × 150 mm, 3 μm) equipped with a guard column (Supelco, Bellefonte, PA). pCBA was analyzed by an Agilent 1100 Series HPLC system consisted of a diode array UV detector and an Agilent Eclipse XDB-C18 (150 × 4.6 mm, 5 μM) column. The details of instrumental conditions are provided in SI Text S3. To evaluate the UV-visible absorbance of IPAs, 50 mg/L of MON or SAL were dissolved in methanol and measured by an Agilent 8453 UV-vis spectrophotometer.
■
RESULTS AND DISCUSSION IPA Degradation under Low Pressure UV Lamp Irradiation. Within their solubility limit in water, the UV absorbance of MON, SAL, NAR, and NIG at 254 nm was below the detection limit of typical spectrophotometer. However, degradation of all four IPAs was observed under 4583
dx.doi.org/10.1021/es3052685 | Environ. Sci. Technol. 2013, 47, 4581−4589
Environmental Science & Technology
Article
natural or wastewater, cosolutes such as NO3−, Cl−, and dissolved organic matter (DOM) may produce certain radicals and reactive transient species under UV irradiation33−35 that may quickly react with IPAs. The rate constant of MON photolysis was increased by 4−5 folds in SW and WW compared to that in DI, indicating indirect photolysis could be the predominant pathway for MON removal under UV 254 nm in raw water. Comparatively, a smaller (
