Organic Contaminant Abatement in Reclaimed ... - ACS Publications

85 downloads 121 Views 3MB Size Report
Feb 24, 2016 - Rebecca A. Trenholm,. ⊥. Silvio Canonica,. ‡. Shane A. Snyder,. ○ ...... Fax: (702) 895-3936. E-mail: Daniel. [email protected] (D.G.). Notes.
Article pubs.acs.org/est

Organic Contaminant Abatement in Reclaimed Water by UV/H2O2 and a Combined Process Consisting of O3/H2O2 Followed by UV/H2O2: Prediction of Abatement Efficiency, Energy Consumption, and Byproduct Formation Yunho Lee,*,†,‡ Daniel Gerrity,*,§,∥,⊥ Minju Lee,‡,# Sujanie Gamage,⊥ Aleksey Pisarenko,∥,⊥ Rebecca A. Trenholm,⊥ Silvio Canonica,‡ Shane A. Snyder,○ and Urs von Gunten‡,# †

School of Environmental Science and Engineering, Gwangju Institute of Science and Technology, 123, Oryong-dong, Buk-gu, Gwangju 500-712, Korea ‡ Eawag, Swiss Federal Institute of Aquatic Science and Technology, Ueberlandstrasse 133, P.O. Box 611, 8600 Duebendorf, Switzerland § Department of Civil and Environmental Engineering, University of Nevada, Las Vegas, Box 454015, 4505 S. Maryland Parkway, Las Vegas, Nevada 89154-4015, United States ∥ Trussell Technologies, Inc., 6540 Lusk Boulevard, Suite C274, San Diego, California 92121, United States ⊥ Applied Research and Development Center, Southern Nevada Water Authority, P.O. Box 99954, Las Vegas, Nevada 89193-9954, United States # School of Architecture, Civil, and Environmental Engineering (ENAC), École Polytechnique Fédérale de Lausanne, CH-1015, Lausanne, Switzerland ○ Department of Chemical and Environmental Engineering, University of Arizona, 1133 E. James E. Rogers Way, Harshbarger 108, Tucson, Arizona 85721-0011, United States S Supporting Information *

ABSTRACT: UV/H2O2 processes can be applied to improve the quality of effluents from municipal wastewater treatment plants by attenuating trace organic contaminants (micropollutants). This study presents a kinetic model based on UV photolysis parameters, including UV absorption rate and quantum yield, and hydroxyl radical (·OH) oxidation parameters, including second-order rate constants for ·OH reactions and steady-state ·OH concentrations, that can be used to predict micropollutant abatement in wastewater. The UV/H2O2 kinetic model successfully predicted the abatement efficiencies of 16 target micropollutants in benchscale UV and UV/H2O2 experiments in 10 secondary wastewater effluents. The model was then used to calculate the electric energies required to achieve specific levels of micropollutant abatement in several advanced wastewater treatment scenarios using various combinations of ozone, UV, and H2O2. UV/H2O2 is more energy-intensive than ozonation for abatement of most micropollutants. Nevertheless, UV/H2O2 is not limited by the formation of N-nitrosodimethylamine (NDMA) and bromate whereas ozonation may produce significant concentrations of these oxidation byproducts, as observed in some of the tested wastewater effluents. The combined process of O3/H2O2 followed by UV/H2O2, which may be warranted in some potable reuse applications, can achieve superior micropollutant abatement with reduced energy consumption compared to UV/H2O2 and reduced oxidation byproduct formation (i.e., NDMA and/or bromate) compared to conventional ozonation.



proposed, tested, or will be implemented in several countries.4,5 Micropollutant mitigation has also been an important issue for potable reuse of municipal wastewater effluents.6

INTRODUCTION

Effluents from municipal wastewater treatment plants (WWTPs) contain organic contaminants in the range of nanograms per liter to micrograms per liter (i.e., micropollutants), and some of these compounds have been shown to be problematic in aquatic ecosystems or potentially for human health.1−3 To reduce the input of such problematic micropollutants to drinking water supplies and the environment, advanced municipal wastewater treatment has been © 2016 American Chemical Society

Received: Revised: Accepted: Published: 3809

October 6, 2015 January 26, 2016 February 24, 2016 February 24, 2016 DOI: 10.1021/acs.est.5b04904 Environ. Sci. Technol. 2016, 50, 3809−3819

Article

Environmental Science & Technology

municipal wastewater effluents,1 and they cover a wide range of reactivity toward UV and ·OH. The validated kinetic model was then used to estimate the energy consumption associated with various treatment train configurations consisting of ozone and UV with/without H2O2 for the mitigation of micropollutants, including NDMA. Formation of NDMA and bromate was determined during ozonation and O3/H2O2 treatment of the selected wastewater effluents, which served as a further parameter for process efficiency comparisons.

Ozonation has been extensively tested as an option for micropollutant mitigation in wastewater treatment. Studies conducted at laboratory-, pilot-, and full-scale, either with ozone alone or in combination with a biological filtration step, have demonstrated significant abatement of many micropollutants and reduction of in vitro and in vivo toxicities.7−15 In the past, the formation of oxidation byproducts, such as aldehydes, bromate, and N-nitrosodimethylamine (NDMA), was not a critical issue for most wastewater treatment applications. However, with the increasing use of advanced treated wastewater as a drinking water source, it is now critical to consider the potential formation of these16−19 and other potentially toxic oxidation byproducts and/or transformation products.20 The combination of UV with hydrogen peroxide (UV/H2O2) is an advanced oxidation process (AOP) which has been shown to be a potential option to attenuate micropollutants in wastewater effluents.21−23 Formation of toxic byproducts is typically less of an issue for UV/H2O2,24,25 although formation of mutagenic/genotoxic byproducts was recently reported after medium pressure UV/H2O2 treatment of nitrate-containing waters.26,27 Studies in clean water matrices (e.g., drinking water) have shown that UV/H2O2 is energy-intensive due to the low UV absorption of H2O2, which results in low OH radical (·OH) formation efficacy. This is compounded by the high ·OH scavenging rate by wastewater matrix components such as dissolved organic matter (DOM)28,29 and by the limited UV transmittance of conventionally treated wastewaters. The use of UV or UV/H2O2 following ozonation can be an attractive alternative because of the increase in UV transmittance achieved by preozonation15,30 and the potential synergistic benefits for disinfection31 and micropollutant mitigation, reductions in NDMA formation,32 and minimization of bromate formation.24 In one study, micropollutant abatement, bromate formation, and the energy demand of O3/H2O2 followed by UV/H2O2 were studied and compared to UV/H2O2 alone for drinking water treatment.33 A substantial decrease in energy consumption was achieved for the combined process while bromate formation was minimal. However, no detailed information is currently available for the process combination of ozonation with UV or UV/H2O2 to eliminate micropollutants in wastewater effluent matrices. The presence of a large number of structurally diverse micropollutants in varying wastewater matrices poses a challenge for a generalizable design and operation of advanced wastewater treatment plants (WWTPs). For ozonation, previous studies have demonstrated that a chemical kinetics approach can be successfully applied to predict micropollutant abatement efficiency.34,35 For UV or UV/H2O2, some chemical kinetics models have been proposed and tested.36−40 However, a systematic evaluation of these models for a range of wastewater effluent matrices to predict the abatement of various micropollutants has not been reported previously. This study presents a kinetic model of low pressure (LP)-UV and LP-UV/H2O2 for the prediction of micropollutant abatement in various wastewater effluent matrices. The model includes kinetic parameters for direct UV photolysis, such as UV absorption spectra and quantum yields, and for UV/H2O2, including second order rate constants for ·OH reactions and the steady-state ·OH concentrations (calculated as the ratio of the ·OH formation rates and scavenging rate constants). The kinetic model was used to predict micropollutant abatement efficiency in bench-scale experiments with 16 target compounds in 10 secondary treated wastewater effluents. The 16 target compounds were selected based on their frequent occurrence in



EXPERIMENTAL SECTION Standards and Reagents. All chemicals and solvents (95% purity or higher) were used as received from various commercial suppliers. Further descriptions of chemical sources and stock solutions are provided in the Supporting Information (SI), SI-Text-1. Wastewater Effluents. Secondary wastewater effluents were collected from 10 WWTPs in the United States (U.S.), Switzerland (CH), and Australia (AUS). Nine WWTPs used conventional activated sludge treatment, and one WWTP used a trickling filter as the main secondary treatment process. Table S1 summarizes the general water quality data for the 10 secondary effluents from this study, which encompass a wide range of water quality in relation to DOM (DOC = 4.7−26.4 mgC L−1 and UV254 absorbance = 0.10−0.41 cm−1), alkalinity (65−332 mg L−1 as CaCO3), nitrite ( 1 × 10−4 cm2 mJ−1), the relative standard deviation (i.e., one standard deviation/ mean ×100) of kUV‑meas was 3−75% (after aggregating the data from the various matrices). The 16 micropollutants were classified into four groups according to the magnitude of their kUV‑meas and kUV/H2O2‑meas values (Table 1), and the criteria for grouping micropollutants will be further discussed below. Group I (Figure 1 and Figures S1a−d). The group I micropollutants (diclofenac, NDMA, triclosan, and sulfamethoxazole) were characterized by kUV‑meas values of ≥1.4 × 10−3 cm2 mJ−1 (Table 1). The average percent abatements with a UV dose of 500 mJ cm−2 (no H2O2) were 93%, 89%, 86%, and 61% for diclofenac, NDMA, triclosan, and sulfamethoxazole, respectively. The kUV‑meas/kUV/H2O2‑meas values for the group I compounds were higher than 0.8, indicating that UV photolysis is mainly responsible for their abatement under the chosen experimental conditions (UV dose = 0−2700 mJ cm−1, [H2O2]0 = 10 mg L−1). For NDMA, direct UV photolysis contributes almost 100% to its

′ } = {2.303εH2O2 Φ·OH,H2O2[H 2O2 ]Ep,ave /{k ·OH,DOC[DOC] + k ·OH,HCO3−[HCO3−] + k ·OH,CO32−[CO32 −] + k ·OH,NO2−[NO2−] + k ·OH,Br−[Br −]}

(3)

where εH2O2 (= 1.96 m2 mol−1) is the molar absorption coefficient of H2O2 at 254 nm,45 Φ·OH,H2O2 (= 1 mol einstein−1) is the ·OH quantum yield from H2O2 photolysis,45 and k·OH,Si is the secondorder rate constant for the reaction of ·OH with the matrix components, with Si being components such as DOC, HCO3−, CO32−, NO2−, and Br−. k·OH,DOC values for the 10 wastewater effluents were determined previously34 using ozone as a source of ·OH, and the values range from 1.0 × 104 to 3.4 × 104 (mgC L−1)−1 s−1 (Table S1). The values of k·OH,HCO3− (= 8.5 × 106 M−1 s−1), k·OH,CO32− (= 3.9 × 108 M−1 s−1), k·OH,NO2− (= 1.0 × 1010 M−1 s−1), and k·OH,Br− (= 1.1 × 109 M−1 s−1) were taken from the literature.46 The initial conditions for the relevant parameters (i.e., the concentration of H2O2, DOC, carbonate species, NO2−, and Br−, and Ep,ave ′ ) were used for the [·OH]ss calculation for the 3811

DOI: 10.1021/acs.est.5b04904 Environ. Sci. Technol. 2016, 50, 3809−3819

Figure 1. Abatement of diclofenac (group I), naproxen (group II), carbamazepine (group III), and meprobamate (group IV) as a function of the UV dose during treatment of the indicated wastewater effluents (Table S1) with UV and UV/H2O2 ([H2O2]0 = 10 mg L−1). Symbols and lines represent the measured and predicted data, respectively. The entire micropollutant abatement data are shown in Figure S1.

Environmental Science & Technology Article

3812

DOI: 10.1021/acs.est.5b04904 Environ. Sci. Technol. 2016, 50, 3809−3819

Article

Environmental Science & Technology

Table 1. Summary of Kinetic Parameters for the Abatement of the Selected Micropollutants (MPs) in UV and UV/H2O2 Processesa

compound diclofenac (pKa = 4.2) NDMA triclosan (pKa = 8.1) sulfamethoxazole (pKa = 5.7) phenytoin atrazine naproxen (pKa = 4.2)

ibuprofen (pKa = 4.9) bisphenol A (pKa = 9.6/10.2) carbamazepine primidone atenolol (pKa = 9.6) trimethoprim (pKa = 7.1) gemfibrozil (pKa = 4.4) DEET meprobamate

−1

2 b

kUV‑meas, (mJ cm )

kUV‑pred, kUV‑meas/ (mJ−1 cm2)c kUV‑pred

εMP, (m mol−1)d 2

kUV‑meas/ k·OH‑pred ′ for kUV/H2O2‑meas × representative FMP (FMP‑ww) k·OH,MP, f −1 e −1 −1 g 100 (mol einstein ) (M s ) wastewater effluent, s−1h

Group I: kUV‑meas ≥ 1.4 × 10−3 cm2 mJ−1 1.0 5.20 × 102 0.29 (0.29) 91 ± 12 2.0 1.53 × 102 0.31 (0.61) ∼100 1.7 1.95 × 102 0.28 (0.48) 90 ± 14 0.7 1.51 × 103 3.8 × 10−2 88 ± 11 (2.7 × 10−2) Group II: 0.2 × 10−3 ≤ kUV‑meas < 1.4 × 10−3 cm2 mJ−1 −3 (1.1 ± 0.4) × 10 1.7 × 10−3 0.7 1.26 × 102 0.28 (0.18) 66 ± 20 (7.1 ± 2.2) × 10−4 8.1 × 10−4 0.9 3.51 × 102 4.7 × 10−2 83 ± 18 (4.2 × 10−2) (3.4 ± 1.7) × 10−4 5.4 × 10−4 0.6 4.02 × 102 2.8 × 10−2 31 ± 19 (1.7 × 10−2) Group III: kUV‑meas < 0.2 × 10−3 cm2 mJ−1 and k·OH,MP ≥ 5 × 109 M−1 s−1 −4 (1.3 ± 0.4) × 10 2.4 × 10−4 0.5 25.6 0.19 (0.10) 17 ± 7 1.7 × 10−5 9.4 75 4.6 × 10−3 (1.6 ± 1.2) × 10−4 17 ± 9 (1.9 × 10−2) (7.0 ± 4.8) × 10−5 1.8 × 10−5 3.9 6.07 × 102 6.0 × 10−4 7±5 (2.4 × 10−3) (7.8 ± 0.2) × 10−5 8.8 × 10−5 0.3 22 8.2 × 10−2 11 ± 3 (2.2 × 10−2) (7.1 ± 2.2) × 10−5 nai nai 43 nai (2.4 × 10−2) 8±4 1.7 × 10−5 1.8 2.94 × 102 1.2 × 10−3 (7.8 ± 4.6) × 10−5 7±4 (2.2 × 10−3) (6.0 ± 2.9) × 10−5 nai nai 27 nai (6.9 × 10−2) 8±4 (5.8 ± 2.7) × 10−5 nai nai 1.29 × 102 nai (5.2 × 10−3) 7±1 Group IV: kUV‑meas < 0.2 × 10−3 cm2 mJ−1 and k·OH,MP is < 5 × 109 M−1 s−1 ∼0 ∼0 nai