Ozonation of trace organic compounds in different municipal and industrial wastewaters: Kinetic-based prediction of removal efficiency and ozone dose requirements

Chemical Engineering Journal xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Ozonation of trace organic compounds in different municipal and industrial wastewaters: Kinetic-based prediction of removal efficiency and ozone dose requirements ⁎

Ze Liua, , Yongyuan Yanga, Chenjia Shaob, Zengwen Jib, Qiaoling Wangb, Shijie Wangb, ⁎ Yaping Guob, , Kristof Demeesterec, Stijn Van Hullea a b c

LIWET, Department of Green Chemistry and Technology, Ghent University, Campus Kortrijk, Graaf Karel De Goedelaan 5, B-8500 Kortrijk, Belgium Zhejiang University of Technology, College of Environment, Hangzhou 310014, PR China Research Group EnVOC, Department of Green Chemistry and Technology, Ghent University, Coupure Links 653, B-9000 Ghent, Belgium

H I GH L IG H T S

kinetic-based prediction was verified in both municipal and industrial wastewaters. • AOzone dose requirements were estimated for TrOCs removal in wastewaters. • OH scavenging from wastewater effluents were determined. • A comparison wasratesmade towards ozone and OH exposure and energy for TrOCs removal. • %

%

A R T I C LE I N FO

A B S T R A C T

Keywords: Trace organic compounds (TrOCs) Ozonation Advanced oxidation process (AOP) Kinetic-based prediction

For the wide application of ozonation in (industrial and municipal) wastewater treatment, prediction of trace organic compounds (TrOCs) removal and evaluation of energy requirements are essential for its design and operation. In this study, a kinetics approach, based on the correlation between the second order reaction rate constants of TrOCs with ozone and hydroxyl radicals (%OH) and the ozone and %OH exposure (i.e., ∫ [O3]dt and ∫ [%OH]dt, which are defined as the time integral concentration of O3 and %OH for a given reaction time), was validated to predict the elimination efficiency in not only municipal wastewaters but also industrial wastewaters. Two municipal wastewater treatment plant effluents from Belgium (HB-effluent) and China (QG-effluent) and two industrial wastewater treatment plant effluents respectively from a China printing and dyeing factory (PDeffluent) and a China lithium-ion battery factory (LZ-effluent) were used for this purpose. The %OH scavenging rate from the major scavengers (namely alkalinity, effluent organic matter (EfOM) and NO2−) and the total %OH scavenging rate of each effluent were calculated. The various water matrices and the %OH scavenging rates resulted in a difference in the requirement for ozone dose and energy for the same level of TrOCs elimination. For example, for more than 90% atrazine (ATZ) abatement in HB-effluent (with a total %OH scavenging rate of 1.9 × 105 s−1) the energy requirement was 12.3 × 10−2 kWh/m3, which was lower than 30.1 × 10−2 kWh/m3 for PD-effluent (with the highest total %OH scavenging rate of 4.7 × 105 s−1). Even though the water characteristics of selected wastewater effluents are quite different, the results of measured and predicted TrOCs abatement efficiency demonstrate that the kinetics approach is applicability for the prediction of target TrOCs elimination by ozonation in both municipal and industrial wastewater treatment plant effluents.

1. Introduction Trace organic compounds (TrOCs), such as pharmaceuticals (e.g. carbamazepine, ibuprofen and bezafibrate), personal care products (e.g. triethylcitrate and acetophenone), and endocrine disruptors (e.g. ⁎

bisphenol A and estrone) [1–3], present in various wastewater treatment plant effluents have received more attention in recent years: its potential risk to the aquatic environment and human health is a growing concern [4–6]. A variety of wastewater treatment technologies have been tested and reported to abate TrOCs. Ozonation, an advanced

Corresponding authors. E-mail addresses: [email protected] (Z. Liu), [email protected] (Y. Guo).

https://doi.org/10.1016/j.cej.2019.123405 Received 17 July 2019; Received in revised form 4 November 2019; Accepted 5 November 2019 1385-8947/ © 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Ze Liu, et al., Chemical Engineering Journal, https://doi.org/10.1016/j.cej.2019.123405

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

ozonation process. In this study, the kinetic-based prediction model was used to predict the removal efficiency of target TrOCs, which have a wide range of reactivity towards ozone and %OH and are typical contaminants presented in effluents, by ozonation in two effluents of municipal wastewater treatment plants located in Belgium and China, respectively, and two effluents of different types industrial wastewater treatment plants in China. The major objectives are (i) to investigate the abatement efficiency of target TrOCs in different types of effluents, (ii) to determine and compare the ozone exposures, %OH exposures and %OH scavenging rates of effluents during ozonation at specific ozone doses, (iii) to test the hypothesis on applying %OH scavenging rate as an operating parameter to assess %OH exposure, (iv) to evaluate the prediction applicability of the kinetics approach (by using the cited values of second order reaction rates of target TrOC with ozone and %OH from literature) in different wastewaters, and (v) to estimate the energy requirements by applying the prediction model.

technology for wastewater treatment, has been demonstrated as a promising method for the removal of a large number of TrOCs from secondary wastewater effluents and the improvement of water quality [7–12]. Despite many studies confirm that the application of ozonation in effluents degrades many types of TrOCs efficiently and significantly, challenges need to be solved that hamper wide application. For example, as thousands of different TrOCs with various chemical structures [13] are present in wastewater effluent and their reaction rates with ozone and hydroxyl radicals (%OH, a secondary oxidant formed during ozone decomposition) also vary greatly [14], the main oxidants and energy requirement during ozonation can differ significantly. Therefore, as ozonation is known as an energy consuming method, during application optimal doses needs to be determined. Furthermore, due to the variation in water matrices and physical-chemical effluents’ characteristics, the removal efficiencies of TrOCs are quite variable during ozone application. As such, a reliable method to predict the removal of TrOCs, to estimate the ozone demand and associated energy requirement, and to determine the optimal ozone doses is necessary. As such, indicators or surrogates such as UV absorbance at 254 nm (UVA254) or total fluorescence are studied to predict TrOCs elimination efficiency [15]. Moreover, quantitative structure-property relationships (QSPR) are developed as a tool to do prediction by means of modelling relationships between compounds’ properties and characteristics of their structures [16]. A general prediction approach on the basis of chemical kinetics (which is different from QSPR) is put forward by Lee et al. [17]. This approach allows the prediction of the abatement of TrOCs during wastewater treatment based on the correlation between the concentration of target compounds (C), the second order reaction rate constants of compounds with ozone and %OH (i.e. kO3 and k.OH ) and the ozone and %OH exposure (∫ [O3 ] dt and ∫ [·OH ] dt ) (as shown Eq. (1)),.

2. Materials and methods 2.1. Standard and reagents All chemicals and solvents (analytical grade, purity of 98% or above) were commercially available and used without any further purification. Six TrOCs with a large difference in ozone reactivity (Table 1) were selected, i.e. atrazine (ATZ), alachlor (ALA), bisphenol-A (BPA), carbamazepine (CBZ), 17α -thinylestradiol (EE2) and pentachlorophenol (PCP).

2.2. Wastewater effluents

[C ] ⎞ − ln ⎛ = k·OH [ ⎝ C ]0 ⎠ ⎜

⎟

∫ [·OH ] dt + kO ∫ [O] dt 3

Effluent samples were collected from a municipal wastewater treatment plants in Belgium (HB-effluent), a municipal wastewater treatment plants in China (QG-effluent), and two industrial wastewater treatment plants in China: one was from a printing and dyeing factory wastewater treatment plant (PD-effluent) and the other was from a lithium-ion battery factory wastewater treatment plant (LZ-effluent). All effluent water samples were used without any dilution, additional filtration or any other disinfection process, and refrigerated at 4 °C (in plastic barrels) before treatment.

(1)

Recently, a large number of constants for TrOCs towards the O3 and OH have been reported, and the ozone exposure and the %OH exposure can also be measured [18]. Additionally, previous studies have demonstrated that this kinetic approach is useful to predict target TrOCs elimination in several water types [19–21]. However, even though the ozone and %OH exposure can be measured, due to the consumption of ozone and %OH by water matrices, the ozone and %OH consumption kinetics of water matrices affect the ozone and %OH exposure. Thus, a challenge will be to predict and determine ozone and %OH exposure in various water matrices if the kinetic-based approach is to be used to make general prediction during the ozonation of wastewater. Additionally, if oxidants scavenging rates in water matrices are determined or predicted, oxidant exposure can be estimated and compared at the specific dosage in different wastewaters. As a result, oxidant scavenging rate could be an indicator of oxidant exposure and an operating parameter during applying this prediction method in ozonation. Furthermore, limited studies currently report the feasibility and reliability of this kinetic-based approach in different types of (domestic and/or industrial) wastewaters, so it is necessary to evaluate the kinetic-based prediction approach during ozonation in various wastewaters from different regions. For example, as the large number and diverse nature of TrOCs in industrial wastewater [22,23], it is interesting to investigate the TrOCs elimination efficiency and evaluate the kinetic-based prediction approach during ozonation in industrial wastewaters treatment plant effluents. However, to our knowledge, the studies on the removal of TrOCs and the development of prediction models in industrial wastewater effluents are never reported. Moreover, TrOCs are increasingly observed worldwide [24]. As such, it is also essential to investigate (municipal) wastewater treatment plants effluents from different regions in the world to estimate the applicability of the kinetic-based approach for the prediction of TrOCs abatement in %

2.3. Ozonation for TrOCs abatement in wastewater effluents Before ozonation, the TrOCs stock solution was prepared by mixing six selected TrOCs and dissolving them at a concentration of 10 mg/L into demineralized water. A fresh ozone stock solution (≈70 mg O3/L) was obtained by bubbling ozone gas into ice-cooled demineralized water in a 1 L gas-washing bottle. Ozone gas was produced by an ozone generator (Anseros COM-AD-02, 460 Watt) with pure oxygen at a flow of 300 mL/min [12]. Ozonation experiments were in the collected effluent and approximately 500 mL of the effluents was added in a reactor (10 cm inner diameter, 15 cm height). TrOCs stock solution was spiked into the reactor to obtain an initial concentration of 200 μg/L. The solutions were stirred by a magnetic stirrer at room temperature after ozone stock solution was dosed in the solutions. A specific ozone dose (expressed as g O3/g DOC), at a range of 0–1.5 g O3/g DOC, was applied in the experiments. In order to eliminate the impact of solution pH on ozone and %OH exposure, the pH of the solution was further adjusted to 8 using 1 M NaOH and buffered with 25 mM phosphate [30]. Experimental samples were collected to determine %OH exposure and the residual dissolved ozone. After quenching residual ozone by adding Na2SO3 (20 mM) [31], the residual TrOCs concentrations were also measured. 2

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

Table 1 An overview of selected TrOCs (their application, chemical structure and second order reaction rate constants towards ozone and %OH). Compounds

k%OH,

application

Alachlor (ALA)

MPs

M−1s−1

kO3,

9

7 × 10 [25]

Herbicide

MPs

M−1s−1

Structure

2.5 [26]

O Cl

N O

Atrazine (ATZ)

9

Herbicide

3.0 × 10

Cl

6

N

N N

HN Bisphenol A (BPA)

Industrial chemical

1.6 × 109 [27]

1.7 × 104 [28]

Carbamazepine (CBZ)

Anti-epileptic

8.8 × 109 [3]

3.0 × 105 [3]

N H

HO

OH N

O 17α -thinylestradiol (EE2)

9.8 × 109 [3]

Contraceptive

NH2

1.8 × 105 [28]

OH H H

H

HO Pentachlorophenol (PCP)

4.0 × 109 [25]

Herbicide

3.78 × 107 [29]

OH Cl

Cl

Cl

Cl Cl

Table 2 Water quality characteristics of the sampled wastewater effluents. Wastewater effluent a,

b

pH

Municipal wastewater effluents

c

Industrial wastewater effluents

d

HB-effluent QG-effluent PD-effluent LZ-effluent

7.3 7.0 7.4 7.3

± ± ± ±

0.1 0.1 0.1 0.1

DOC mg C/L

Alkalinity mg/L as CaCO3

NO2− mg N/L

NO3− mg N/L

NH4+ mg N/L

9.6 ± 0.2 8.7 ± 0.3 11.4 ± 0.4 12.3 ± 0.4

2.5 3.1 4.9 3.3

0.24 0.17 0.45 0.28

6.61 1.81 3.94 0.32

1.54 3.43 0.67 9.21

± ± ± ±

0.3 0.2 0.1 0.3

± ± ± ±

0.01 0.05 0.03 0.02

± ± ± ±

0.03 0.01 0.02 0.01

± ± ± ±

0.03 0.02 0.01 0.04

a

No rain within 72 h before sampling. ‘… ± …’ indicates the average and standard deviation on triplicate measurements. c HB-effluent and QG-effluent were collected from municipal wastewater treatment plants in Belgium and China, respectively. d PD-effluent was picked up from a Chinese printing and dyeing factory wastewater treatment plant; LZ-effluent was gotten from a Chinese lithium-ion battery factory wastewater treatment plant. b

previous studies [21]. Ozone concentrations were determined by the indigo method [34]. %OH concentrations were determined by the measurement of p-chlorobenzoic acid (pCBA, a probe compound) by HPLC-DAD (Agilent 1100 series, USA) [35]. The determinations of ozone and %OH exposures have been shown in Text S1-2. A reaction time of 30 min was used to calculate the ozone and %OH exposures. The % OH scavenging rates of wastewater effluents were determined by competition kinetics, as detailed in Text S3 [36].

2.4. Analytical methods Standard methods were applied for measuring physical–chemical water quality parameters [32]. Nitrite (NO2−-N), nitrate (NO3−-N), ammonium (NH4+-N) and COD were measured by Hach-Lange cuvettes and a DR2800 spectrophotometer (Hach, Belgium). The pH was measured by a Metrohm 600 pH-meter (Metrohm, Belgium). UV − visible (UV − vis) absorption spectra analysis was obtained by a Shimadzu UV1601 spectrophotometer (with 1 cm quartz cuvettes) within a range from 200 to 800 nm (with 0.5 nm increments). The quantitative analysis of six selected TrOCs was performed on an Agilent 6890 GC Series gas chromatograph coupled to a Hewlett Packard 5973 mass selective (MS) detector after liquid–liquid extraction with dichloromethane (CH2Cl2) [33]. More details on the GC–MS method can be found in

3. Results and discussion 3.1. Effluent water quality The effluent water quality plays an important role in the TrOCs 3

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

Table 3 Summary of %OH reaction rate constants with EfOM in the selected wastewater treatment plant effluents, the %OH scavenging rate from EfOM, alkalinity and NO2−, and the total %OH scavenging rate of the tested wastewater treatment plant effluents.

k ·OH,EfOM (L mg C−1 s−1) a,

b

Scavenging rate from EfOM (s−1) Scavenging rate from alkalinity (s−1) Scavenging rate from NO2− (s−1) Total scavenging rate (s−1)

HB-effluent

QG-effluent

PD-effluent

LZ-effluent

1.2 × 104

2.1 × 104

2.9 × 104

2.3 × 104

1.2 × 105 1.3 × 104 5.2 × 104 1.9 × 105

1.8 × 105 2.1 × 104 3.7 × 104 2.4 × 105

3.3 × 105 4.6 × 104 9.7 × 104 4.7 × 105

2.8 × 105 2.7 × 104 6.1 × 104 3.7 × 105

a

More details on the calculation of k ·OH,EfOM can be found in [36].

b

The concentration of EfOM was determined as DOC (mg C L−1).

Fig. 1 for the four investigated effluents. With the increasing of ozone dose, the ozone exposure increased during ozonation in all tested effluents. The ozone exposures were (0.2–3.8) × 10−2 M s for HB-effluent, (0.2–3.0) × 10−2 M s for QG-effluent, (0.1–1.2) × 10−2 M s for PD-effluent, and (0.2–1.6) × 10−2 M s for LZ-effluent, respectively. The results showed that the ozone exposure in the two industrial wastewater treatment plant effluents was obviously lower than in the two municipal wastewater treatment plant effluents. It was attributed to the higher levels of EfOM in the two industrial wastewater treatment plant effluents (Table 3). These result which ozone exposure is influenced by water matrices is also in agreement with the research of Wert et al. [38]. It is worth to mention that almost no change of the ozone exposure in all selected effluents was observed when the ozone dose of less than 0.4 g O3/g DOC was applied. It seemed that ozone was consumed by water matrix or rapidly decomposed during the instantaneous ozone demand (IOD) when ozone dose was as low as less than 0.4 g O3/ g DOC during ozonation. It caused the no oxidants was available to abate contaminants under this circumstance. Thus, the results of ozone exposures suggest the lowest ozone requirements on TrOCs elimination during ozonation in this study.

elimination during ozonation. Based on the results shown in Table 2, the water quality of four effluents are greatly different. Alkalinity, DOC and NO2− present in the two industrial wastewater treatment plant effluents were higher than in the two municipal wastewater treatments plant effluents. The concentration of DOC was the highest in the LZeffluent, and the concentrations of alkalinity and NO2− were higher in the PD-effluent than k·OH , HCO3− in the other three effluents. In the selected effluents, alkalinity (=8.5 × 106 M−1s−1 8 −1 −1 andk·OH , co32 − = 3.9 × 10 M s [37]), effluent organic matter (EfOM, the values of k·OH , EfoM for effluents were shown in Table 3) and NO2− (k·OH , NO2− =1.0 × 1010 M−1s−1 [37]) are the major %OH scavengers. These scavengers influenced TrOCs elimination by means of consuming % OH and reducing the availability of %OH that can react with target TrOCs. Meanwhile, decreased %OH exposure in the effluents affects the prediction of TrOCs elimination by the kinetics-based approach. To further evaluate the influences of these scavengers on TrOCs elimination and the application of the prediction approach, the %OH scavenging rates from different scavengers and selected effluents were calculated and provided in Table 3. As the results show in Table 3, the total scavenging rate of the PDeffluent (4.7 × 105 s−1) was the highest among all selected effluents, and the total scavenging rates of the other three effluents ranged from 1.9 × 105 s−1 to 3.7 × 105 s−1. The relative high alkalinity and NO2− resulted in a higher total %OH scavenging rate for the PD-effluent. It should be noted that, in all selected effluents, the %OH scavenging rates from EfOM contributed for more than 60% to the total %OH scavenging rates, indicating that EfOM was the most dominant scavenger in the selected effluents. The research of Lee et al. is consistent with this result [17]. Lee et al. applied a competition kinetic method to confirm that around 80% %OH was consumed by EfOM in their investigated effluents, although only EfOM and bicarbonate – and not NO2− – were considered as the major scavengers. The results of the %OH scavenging rates indicate that most of %OH is not consumed by reacting with TrOCs but reduced by water matrices. By comparing the results of %OH scavenging rates from different effluents, it can be estimated that the effluent with a higher %OH scavenging rate needs more ozone to achieve the same level of %OH exposure. %OH scavenging rate shows the consumption capacity of %OH by water matrix, it can be used not only to characterize the assignment of %OH consumption during ozonation but also to evaluate and determine the optimal ozone dosage range during wastewater ozonation. Furthermore, based on the contributed rates of each scavenger in wastewater towards the total %OH scavenging rate of wastewater, targeted water quality improvement measures (e.g. prefiltration of the waste water [21]) could be made to decrease %OH scavenging rate, increase ozone utilization rate on TrOCs abatement and reduce energy and cost of ozonation application.

3.3. %OH exposure %

OH exposure was calculated by the time-integrated concentration of OH for a given reaction period (∫ [%OH] dt). It can be seen in Fig. 2 that %OH exposure increases with increasing ozone dose. The increase of the %OH exposure with the ozone dose shows to be linear (R2 = 0.9232), which is in agreement with [39]. %OH is mainly generated by ozone decay, so the formation of %OH is primarily dependent on the ozone dose [40], but the results shown in Fig. 2 showed differences towards %OH exposures during ozonation in different effluents at the same ozone doses. Higher %OH exposures were observed in the effluents (QG-effluent and particularly HB-effluent) with the lowest %OH scavenging rate (Table 3 and Fig. 2). Hence, it seems that %OH exposure is affected by the water matrix in this study. To further investigate the possibly effect of water matrix on %OH exposure during wastewater effluents ozonation, the PD-effluent was diluted with demineralized water (1:1) to achieve a total %OH scavenging rate similar to that of the QG-effluent. Fig. 2 shows that the %OH exposure of the 50% diluted PD-effluent is still 9.7–33.5% lower than that of the QG-effluent despite %OH exposure at equal ozone dose obtains 0.1–0.5 × 10−10 M s increase by dilution. Fig. 1 showed that the ozone exposure in PD-effluent was lower than that in QG-effluent. It can be explained that ozone in PD-effluent may have a faster decay, which leaded to a short time that the formed %OH presented in the effluent. Consequently, %OH exposure in PD-effluent was lower. Besides, it is possible that the EfOM of PD-effluent is more reactive towards ozone via non-radical mechanisms, or the EfOM of QG-effluent contains more electron-rich compounds. If a considerable number of ozone-reactive compounds which react with ozone through non-radical mechanisms present in EfOM of PD-effluent, less %OH could form from ozone decomposition at the specific ozone dosages. Meanwhile, if more electronrich compounds contained in the EfOM of QG-effluent, more %OH could %

3.2. Ozone exposure According to the measured ozone concentrations in the experimental solutions at specific reaction times, ozone exposure was calculated by the time-integrated concentration of O3 for a given reaction period (∫ [O3] dt). Ozone exposures at specific ozone doses are shown in 4

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

Fig. 1. Ozone exposure at specific ozone doses for the four different types of wastewater treatment plant effluents treated by ozonation.

These results demonstrate that this approach is applicable to predict TrOCs elimination during ozonation in both municipal and industrial wastewater treatment plant effluents. It should be noted that the predicted abatement of three ozone-reactive compounds (BPA, CBZ and EE2) overestimates the measured abatement by a factor of 1.1–3.4 in all selected effluents. This can be explained by the uncertainty on the ozone exposure determination caused by mass transfer limitation or inefficient mixing during ozonation experiments [44]. The values ofk·OH ∫ [OH ] dt (briefly denoted as k1) + kO3 ∫ [O3 ] dt (briefly denoted as k2) were calculated for all selected contaminants in the four types of effluents to further elucidate the effect of ozone and % OH exposure on the predicted TrOCs elimination. The results are shown in Fig. 4 at an ozone dose of 0.9 g O3/g DOC as an example. For ATZ and ALA, the values of k1 + k2 were lower than 4.5, and k1 was dominant (> 93% contribution to k1 + k2) in all selected effluents. This indicates that the ozonation abatement of ATZ and ALA is primarily attributed by %OH-induced oxidation, which can be explained by the relatively low reaction rate constants of both compounds towards ozone (Table 1). Conversely, k2 dominates the k1 + k2 for BPA, CBZ, EE2 and PCP(> 83%, > 74%, > 61% and > 89% contribution to k1 + k2 for BPA, CBZ, EE2 and PCP, respectively), which demonstrates that the

be produced from the reaction ozone towards electron-rich compounds. It could be a possible pathway to generate %OH even though it is not main way for %OH generation during ozonation [41–43]. The ozone dose is the dominant factor affecting %OH exposure, but the impact of water matrix on %OH exposure is indispensable. It should be mentioned that due to the differences of water matrices, the similar %OH scavenging rates does not mean that %OH exposure is similar at the same ozone dose. 3.4. Prediction of ozonation TrOCs elimination efficiency in wastewater effluents According to the kinetic approach (Text S4), the experimentally determined ozone and %OH exposure were used to predict the elimination efficiency of the six selected TrOCs in four different wastewater effluents. Fig. 3 shows that the model predicts well the measured removal efficiency: R2 = 0.9021 and the standard deviation of the linear regression (S) = 9.294 for HB-effluent, R2 = 0.9267 and S = 9.009 for QG-effluent, R2 = 0.9512 and S = 7.655 for PD-effluent, and R2 = 0.9007 and S = 5.599 for LZ-effluent, respectively. It is suggested by Fig. 3 that the predicted results agree well with the measured results.

Fig. 2. OH exposure at specific ozone doses for the four different types of wastewater treatment plant effluents and PD-effluent diluted 1:1 with demineralized water treated by ozonation. The line represents linear regression of the data of four effluents without dilution. 5

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

100

100

a. HB-effluent n=114 r²=0.9021 S=9.294

60

c. PD-effluent n=90 r²=0.9512 S=7.655

80

40

ATZ

ALA

BPA

CBZ

EE2

PCP

Elimination predicted, %

Elimination predicted, %

80

60

40

20

20

0

0 0

20

40

60

80

0

100

20

100

60

80

100

80

100

100

b. QG-effluent n=114 r²=0.9267 S=9.009

60

d. LZ-effluent n=90 r²=0.9007 S=5.599

80 Elimination predicted, %

80 Elimination predicted, %

40

Elimination measured, %

Elimination measured, %

40

60

40

20

20

0

0 0

20

40

60

80

0

100

20

40

60

Elimination measured, %

Elimination measured, %

Fig. 3. Measured and predicted elimination efficiency of TrOCs during ozonation in (a) HB-effluent, (b) QG-effluent, (c) PD-effluent and (d) LZ-effluent. “n” is the number of data points. “S” is standard deviation of each linear regression. The S is calculated by the correlation between the sum-of-squares of the distance of the linear regression from the data points (SS) and the degrees of freedom of the fit (df), i.e. (SS/df)2. The full line represents the equation y = x.

lower %OH scavenging rate, the HB-effluent clearly requires less energy to achieve the same level of TrOC removal than in the industrial effluents. For the ozone-reactive compounds, it is noticed that the energy costs are slightly lower in the QG-effluent than in the HB-effluent. Even though the QG-effluent has a higher %OH scavenging rate, most TrOCs were degraded by ozone in the effluent (as shown in Fig. 4). Hence, this result indicated that more ozone was consumed by the water matrix in the HB-effluent and the ozone utilization rate on reacting to TrOCs was higher in QG-effluent comparing to in HB-effluent.

largest fraction of these more ozone-reactive TrOCs is degraded by direct ozone oxidation. Due to the higher reaction rate with ozone, the values of k1 + k2 for BPA, EE2, CBZ and PCP are also a factor of 42 to 4017 higher than for ALA and ATZ. The values of k1 + k2 for ATZ and ALA (in Fig. 4) were less than 10, however, the values of k1 + k2 for BPA, EE2 CBZ and PCP were at least two orders of magnitude higher. According to Text S4, this indicated that less BPA, EE2 CBZ and PCP remained in the effluents than ATZ and ALA. Additionally, as shown in Fig. 4, high removals of ATZ (> 60%) and ALA (> 50%) and completely removals of BPA, EE2 CBZ and PCP were achieved with applying 0.9 g O3/g DOC during ozonation. By means of calculating with the kinetics-based method on basis of the values of k1 + k2, an ozone dose of 0.9 g O3/g DOC is excessive for the abatement of BPA, EE2 CBZ and PCP in the selected effluents. This calculation result is also consistent with the experimental observations.

4. Conclusions In this study, the kinetics approach on the basis of the correlation between the second order reaction rate constant, ozone exposure and % OH exposure was demonstrated as a feasible method to predict TrOCs elimination efficiency during ozonation. Although water matrix affects the ozone exposure and TrOCs removal efficiency, the kinetic-based prediction model is applicability in not only municipal wastewater treatment plant effluents but also industrial wastewater treatment plant effluents. Furthermore, the %OH scavenging rates from the major scavengers in effluents and the total %OH scavenging rate of each effluent were calculated. With a comprehensive comparison of ozone and %OH exposure during ozonation in effluents, %OH scavenging rate could be used as an operating parameter to assess the %OH exposure. This is very meaningful for the further application of the TrOCs elimination prediction method during ozonation.

3.5. Energy requirement for TrOCs elimination The energy requirements for TrOCs elimination during ozonation were calculated and compared for the four selected effluents (Table 4). Based on the used ozone generator and capacity (0.05–0.15 kg O3/kg O2) [35], the average energy requirement for O3 is 12 kWh/kg. To compare the energy requirements, all calculations are based on a TrOC removal of 95%. Due to the high second order reaction rate constants of BPA, CBZ, EE2 and PCP with ozone, the ozone dose and thus also energy requirements for these compounds is very similar (0.3 g O3/g DOC) in the different wastewater effluents. According to the results in Table 4, a quite large (up to a factor of 2.5) difference in energy requirements for TrOCs elimination is noticed among the different wastewater effluents, particularly for the less ozone-reactive compounds. As a result of the

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to 6

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

a. ALA

4.5

d. EE2

100

k₁

3500

k₁

k₂

k₂

ALA elimination

EE2 elimination Elimination, %

50

k₁ + k₂

2.5

2500

Elimination, %

75

k₁ + k₂

3.0

1500 100 75 50 25 0

4 2

0

0

0 HB

QG PD Wastewater treatment plants effluents

1.5

HB

LZ

QG PD Wastewater treatment plants effluents

b. ATZ

LZ

e. CBZ

k₁

k₁

5000

k₂

k₂ 75

0.5

Elimination, %

3000

k₁ + k₂

k₁ + k₂

50

CBZ elimination

4000

1.0

Elimination, %

ATZ elimination

2000

25

0 QG PD Wastewater treatment plants effluents

100

2

50

0

0 HB

4

LZ

0 HB

QG PD Wastewater treatment plants effluents

f. PCP

c. BPA

280

600000

k₁

k₁

k₂

k₂

BPA elimination

PCP elimination

k₁ + k₂

k₁ + k₂ 100

200000

100

0.8

0 HB

QG PD Wastewater treatment plants effluents

100

4 2 0

50 0

Elimination, %

400000

Elimination, %

200

LZ

50 0 HB

LZ

QG PD Wastewater treatment plants effluents

LZ

Fig. 4. The values of k·OH ∫ [OH ] dt (denoted as k1) + kO3 ∫ [O3 ] dt (denoted as k2) (left y-axis) and the measured TrOCs elimination efficiency (right y-axis) at a specific ozone dose of 0.9 g O3/g DOC in four different effluents.

influence the work reported in this paper.

Table 4 Energy requirements (×10−2 kWh/m3) for 95% TrOCs elimination by ozonation in four selected effluents. Target contaminant

HB-effluent

QG-effluent

PD-effluent

LZ-effluent

ATZ ALA BPA, EE2, CBZ and PCP

17.28 8.06 3.46

17.75 8.78 3.13

30.10 20.52 4.10

29.52 17.71 4.43

Acknowledgement Ze Liu is financially supported by the China Scholarship Council (CSC) by a CSC PhD grant. This research fits within the LED H2O project. The LED H2O is financially supported by The Flanders Knowledge Centre Water. Finally, Liu Z. would like to thank the care, support and love from Liu H. over the years. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// 7

Chemical Engineering Journal xxx (xxxx) xxxx

Z. Liu, et al.

doi.org/10.1016/j.cej.2019.123405. References

[22]

[1] A. Kumar, I. Xagoraraki, Pharmaceuticals, personal care products and endocrinedisrupting chemicals in U.S. surface and finished drinking waters: a proposed ranking system, Sci. Total Environ. 408 (2010) 5972–5989, https://doi.org/10. 1016/j.scitotenv.2010.08.048. [2] D. Gerrity, S. Snyder, Review of ozone for water reuse applications: toxicity, regulations, and trace organic contaminant oxidation, Ozone Sci. Eng. 33 (2011) 253–266, https://doi.org/10.1080/01919512.2011.578038. [3] M.M. Huber, S. Canonica, G.Y. Park, U. Von Gunten, Oxidation of pharmaceuticals during ozonation and advanced oxidation processes, Environ. Sci. Technol. 37 (2003) 1016–1024, https://doi.org/10.1021/es025896h. [4] Y. Luo, W. Guo, H.H. Ngo, L.D. Nghiem, F.I. Hai, J. Zhang, S. Liang, X.C. Wang, A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment, Sci. Total Environ. 473 (2014) 619–641, https://doi.org/10.1016/j.scitotenv.2013.12.065. [5] M.O. Barbosa, N.F.F. Moreira, A.R. Ribeiro, M.F.R. Pereira, A.M.T. Silva, Occurrence and removal of organic micropollutants: An overview of the watch list of EU Decision 2015/495, Water Res. 94 (2016) 257–279, https://doi.org/10.1016/ j.watres.2016.02.047. [6] S.-W. Nam, B.-I. Jo, Y. Yoon, K.-D. Zoh, Occurrence and removal of selected micropollutants in a water treatment plant, Chemosphere 95 (2014) 156–165, https:// doi.org/10.1016/j.chemosphere.2013.08.055. [7] M. Chys, K. Demeestere, A.S. Ingabire, J. Dries, H. Van Langenhove, S.W.H. Van Hulle, Enhanced treatment of secondary municipal wastewater effluent: comparing (biological) filtration and ozonation in view of micropollutant removal, unselective effluent toxicity, and the potential for real-time control, Water Sci. Technol. 76 (2017) 236–246, https://doi.org/10.2166/wst.2017.207. [8] M. Bourgin, B. Beck, M. Boehler, E. Borowska, J. Fleiner, E. Salhi, R. Teichler, U. von Gunten, H. Siegrist, C.S. McArdell, Evaluation of a full-scale wastewater treatment plant upgraded with ozonation and biological post-treatments: Abatement of micropollutants, formation of transformation products and oxidation by-products, Water Res. 129 (2018) 486–498, https://doi.org/10.1016/j.watres. 2017.10.036. [9] G. Knopp, C. Prasse, T.A. Ternes, P. Cornel, Elimination of micropollutants and transformation products from a wastewater treatment plant effluent through pilot scale ozonation followed by various activated carbon and biological filters, Water Res. 100 (2016) 580–592, https://doi.org/10.1016/j.watres.2016.04.069. [10] K. Yang, J. Yu, Q. Guo, C. Wang, M. Yang, Y. Zhang, P. Xia, D. Zhang, Z. Yu, Comparison of micropollutants’ removal performance between pre-ozonation and post-ozonation using a pilot study, Water Res. 111 (2017) 147–153, https://doi. org/10.1016/j.watres.2016.12.043. [11] K.M.S. Hansen, A. Spiliotopoulou, R.K. Chhetri, M. Escolà Casas, K. Bester, H.R. Andersen, Ozonation for source treatment of pharmaceuticals in hospital wastewater – Ozone lifetime and required ozone dose, Chem. Eng. J. 290 (2016) 507–514, https://doi.org/10.1016/j.cej.2016.01.027. [12] Z. Liu, S. Hosseinzadeh, N. Wardenier, Y. Verheust, M. Chys, S.V. Hulle, Combining ozone with UV and H2O2 for the degradation of micropollutants from different origins: lab-scale analysis and optimization, Environ. Technol. (2018) 1–10, https://doi.org/10.1080/09593330.2018.1491630. [13] J.E. Drewes, P. Anderson, N. Denslow, A. Olivieri, D. Schlenk, S.A. Snyder, K.A. Maruya, Designing monitoring programs for chemicals of emerging concern in potable reuse – what to include and what not to include? Water Sci Technol. 67 (2013) 433–439, https://doi.org/10.2166/wst.2012.520. [14] R.L. Oulton, T. Kohn, D.M. Cwiertny, Pharmaceuticals and personal care products in effluent matrices: a survey of transformation and removal during wastewater treatment and implications for wastewater management, J. Environ. Monit. 12 (2010) 1956–1978, https://doi.org/10.1039/C0EM00068J. [15] M. Park, T. Anumol, K.D. Daniels, S. Wu, A.D. Ziska, S.A. Snyder, Predicting trace organic compound attenuation by ozone oxidation: Development of indicator and surrogate models, Water Res. 119 (2017) 21–32, https://doi.org/10.1016/j.watres. 2017.04.024. [16] X. Jin, S. Peldszus, P.M. Huck, Predicting the reaction rate constants of micropollutants with hydroxyl radicals in water using QSPR modeling, Chemosphere 138 (2015) 1–9, https://doi.org/10.1016/j.chemosphere.2015.05.034. [17] Y. Lee, D. Gerrity, M. Lee, A.E. Bogeat, E. Salhi, S. Gamage, R.A. Trenholm, E.C. Wert, S.A. Snyder, U. von Gunten, Prediction of micropollutant elimination during ozonation of municipal wastewater effluents: use of kinetic and water specific information, Environ. Sci. Technol. 47 (2013) 5872–5881, https://doi.org/10. 1021/es400781r. [18] U. von Gunten, Ozonation of drinking water: Part I. Oxidation kinetics and product formation, Water Res. 37 (2003) 1443–1467, https://doi.org/10.1016/S00431354(02)00457-8. [19] Y. Lee, L. Kovalova, C.S. McArdell, U. von Gunten, Prediction of micropollutant elimination during ozonation of a hospital wastewater effluent, Water Res. 64 (2014) 134–148, https://doi.org/10.1016/j.watres.2014.06.027. [20] Y. Guo, H. Wang, B. Wang, S. Deng, J. Huang, G. Yu, Y. Wang, Prediction of micropollutant abatement during homogeneous catalytic ozonation by a chemical kinetic model, Water Res. 142 (2018) 383–395, https://doi.org/10.1016/j.watres. 2018.06.019. [21] Z. Liu, K. Demeestere, S. Van Hulle, Enhanced ozonation of trace organic

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

8

contaminants in municipal wastewater plant effluent by adding a preceding filtration step: comparison and prediction of removal efficiency, ACS Sustainable Chem. Eng. (2019), https://doi.org/10.1021/acssuschemeng.9b02571. T. Reemtsma, B. Zywicki, M. Stueber, A. Kloepfer, M. Jekel, Removal of sulfur−organic polar micropollutants in a membrane bioreactor treating industrial wastewater, Environ. Sci. Technol. 36 (2002) 1102–1106, https://doi.org/10.1021/ es010185p. A. Joss, H. Siegrist, T.A. Ternes, Are we about to upgrade wastewater treatment for removing organic micropollutants? Water Sci. Technol. 57 (2008) 251–255, https://doi.org/10.2166/wst.2008.825. European Chemical Agency (2017), (n.d.). W.R. Haag, C.C.D. Yao, Rate constants for reaction of hydroxyl radicals with several drinking water contaminants, Environ. Sci. Technol. 26 (1992) 1005–1013, https:// doi.org/10.1021/es00029a021. Z. Qiang, C. Liu, B. Dong, Y. Zhang, Degradation mechanism of alachlor during direct ozonation and O-3/H2O2 advanced oxidation process, Chemosphere 78 (2010) 517–526, https://doi.org/10.1016/j.chemosphere.2009.11.037. M. Umar, F. Roddick, L. Fan, H.A. Aziz, Application of ozone for the removal of bisphenol A from water and wastewater – a review, Chemosphere 90 (2013) 2197–2207, https://doi.org/10.1016/j.chemosphere.2012.09.090. M. Deborde, S. Rabouan, J.-P. Duguet, B. Legube, Kinetics of aqueous ozone-induced oxidation of some endocrine disruptors, Environ. Sci. Technol. 39 (2005) 6086–6092, https://doi.org/10.1021/es0501619. F. Javier Benitez, J.L. Acero, F.J. Real, J. Garcı́a, Kinetics of photodegradation and ozonation of pentachlorophenol, Chemosphere 51 (2003) 651–662, https://doi. org/10.1016/S0045-6535(03)00153-X. J.-K. Im, I.-H. Cho, S.-K. Kim, K.-D. Zoh, Optimization of carbamazepine removal in O-3/UV/H2O2 system using a response surface methodology with central composite design, Desalination 285 (2012) 306–314, https://doi.org/10.1016/j.desal. 2011.10.018. R.R. Giri, H. Ozaki, S. Ota, R. Takanami, S. Taniguchi, Degradation of common pharmaceuticals and personal care products in mixed solutions by advanced oxidation techniques, Int. J. Environ. Sci. Technol. 7 (2010) 251–260. A.P.H. Association, editor.) Baird Rodge, editor.) Eaton Andrew, editor.) Rice Eugene W., editor.) Bridgewater Laur, A.W.W. Association, W.E. Federation, Standard methods for the examination of water and wastewater, 23rd edition, Washington, DC American Public Health Association, 2017. https://trove.nla.gov. au/version/250853389. N. Wardenier, Z. Liu, A. Nikiforov, S.W.H. Van Hulle, C. Leys, Micropollutant elimination by O3, UV and plasma-based AOPs: An evaluation of treatment and energy costs, Chemosphere 234 (2019) 715–724, https://doi.org/10.1016/j. chemosphere.2019.06.033. H. Bader, J. Hoigné, Determination of ozone in water by the indigo method, Water Res. 15 (1981) 449–456, https://doi.org/10.1016/0043-1354(81)90054-3. M. Chys, W.T.M. Audenaert, H. Stapel, A. Ried, A. Wieland, M. Weemaes, H. Van Langenhove, I. Nopens, K. Demeestere, S.W.H. Van Hulle, Techno-economic assessment of surrogate-based real-time control and monitoring of secondary effluent ozonation at pilot scale, Chem. Eng. J. 352 (2018) 431–440, https://doi.org/10. 1016/j.cej.2018.07.041. Z. Liu, M. Chys, Y. Yang, K. Demeestere, S. Van Hulle, Oxidation of trace organic contaminants (TrOCs) in wastewater effluent with different ozone-based AOPs: comparison of ozone exposure and •oh formation, Ind. Eng. Chem. Res. 58 (2019) 8896–8902, https://doi.org/10.1021/acs.iecr.9b00293. G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical Review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (⋅OH/⋅O− in Aqueous Solution, J. Phys. Chem. Ref. Data 17 (1988) 513–886, https://doi.org/10.1063/1.555805. E.C. Wert, F.L. Rosario-Ortiz, S.A. Snyder, Effect of ozone exposure on the oxidation of trace organic contaminants in wastewater, Water Res. 43 (2009) 1005–1014, https://doi.org/10.1016/j.watres.2008.11.050. W.T.M. Audenaert, D. Vandierendonck, S.W.H. Van Hulle, I. Nopens, Comparison of ozone and HO induced conversion of effluent organic matter (EfOM) using ozonation and UV/H2O2 treatment, Water Res. 47 (2013) 2387–2398, https://doi.org/ 10.1016/j.watres.2013.02.003. E.J. Rosenfeldt, K.G. Linden, S. Canonica, U. von Gunten, Comparison of the efficiency of (OH)-O-center dot radical formation during ozonation and the advanced oxidation Processes O-3/H2O2 and UV/H2O2 (vol 40, pg 20, 2006), Water Res. 42 (2008) 2836–2838, https://doi.org/10.1016/j.watres.2008.01.027. S. Naumov, G. Mark, A. Jarocki, C. von Sonntag, The reactions of nitrite ion with ozone in aqueous solution – new experimental data and quantum-chemical considerations, Ozone Sci. Eng. 32 (6) (2010) 430–434, https://doi.org/10.1080/ 01919512.2010.522960. E. Mvula, S. Naumov, C. von Sonntag, Ozonolysis of Lignin Models in Aqueous Solution: Anisole, 1,2-Dimethoxybenzene, 1,4-Dimethoxybenzene, and 1,3,5Trimethoxybenzene, Environ. Sci. Technol. 43 (2009) 6275–6282, https://doi.org/ 10.1021/es900803p. P. Dowideit, C. von Sonntag, Reaction of ozone with ethene and its methyl- and chlorine-substituted derivatives in aqueous solution, Environ. Sci. Technol. 32 (1998) 1112–1119, https://doi.org/10.1021/es971044j. U. Hübner, S. Keller, M. Jekel, Evaluation of the prediction of trace organic compound removal during ozonation of secondary effluents using tracer substances and second order rate kinetics, Water Res. 47 (2013) 6467–6474, https://doi.org/10. 1016/j.watres.2013.08.025.