- Email: [email protected]
anaerobic landscape water system
Environmental Pollution 255 (2019) 112989
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Dose effects of calcium peroxide on harmful gases emissions in the anoxic/anaerobic landscape water system* Wen-Huai Wang a, b, Yi Wang a, b, *, Jia-Jun Li a, b, Heng Zhang a, b, Fei-Long Yan a, b, Lu-Qin Sun c a b c
School of Environmental and Municipal Engineering, Xi'an University of Architecture and Technology, China Key Laboratory of Northwest Water Resource, Environment and Ecology, MOE, Xi'an University of Architecture and Technology, Xi'an 710055, China Environmental Science Department, University of San Francisco, California, CA 94117, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 April 2019 Received in revised form 15 July 2019 Accepted 31 July 2019 Available online 1 August 2019
Large-area hypoxia of urban landscape water often causes the emissions of harmful gases in summer, which not only reduces its sensory effects, but also brings a potential threat to aquatic ecosystem and human health. This study explored the dose effects of calcium peroxide (CaO2) on inhibiting harmful gases emissions and restoring the scenic effect (including visual sense and olfactory sense) of anoxic/ anaerobic landscape water system. The results indicated that the emissions of H2S, CO2 and CH4 from the anoxic/anaerobic water system were obviously inhibited in the reactors with CaO2 additions and the effect was positively correlated with the CaO2 dose. Meanwhile, the concentrations of total chemical oxygen demand (TCOD) and soluble sulfide (S2), and turbidity in the overlying water (the water-layer above the sediment-water interface) were also decreased in the reactors dosed with CaO2. The reason was ascribed to the improvement of the anoxic/anaerobic condition in the water system and the increase of the species richness, bacteria count and aerobic microorganism abundance in sediment. Furthermore, 0.12 kg-CaO2 m2-sediment was selected as the optimal dose, which was based on considering the inhibiting effect of the harmful gases emissions, comprehensive influence and costs. Compared with control check (CK, the reactor without adding CaO2), the optimal dose of CaO2 could reduce 75.10% CH4, 81.02% CO2 and 100% H2S in gases, and decrease 81.52% S2, 42.85% TCOD and 84.01% turbidity in the overlying water. In conclusion, all the dosages of CaO2 could improve the anoxic condition of water system and 0.12 kg-CaO2 m2-sediment was the optimal dose in inhibiting harmful gases emissions, which could keep an excellent water quality in this simulation experiment. Therefore, this study may provide a feasible method and the optimal dose for inhibiting the emissions of harmful gases and restoring the scenic effect in the similar anoxic/anaerobic landscape water. © 2019 Published by Elsevier Ltd.
Keywords: Calcium peroxide dosages Landscape water system Anoxic/anaerobic condition Harmful gases Sulfur transformation
1. Introduction Urban landscape water is normally located in the populated area (Wang et al., 2017b). It is widely believed that exogenous pollution and endogenous release from the sediments are the main reasons for the water system to become anoxic or anaerobic (especially in summer), which would further lead to the generation and emissions of harmful gases, such as carbon dioxide (CO2), nitrous oxide
* This paper has been recommended for acceptance by Maria Cristina Fossi. * Corresponding author. School of Environmental and Municipal Engineering, Xi'an University of Architecture and Technology, China. E-mail addresses: [email protected], [email protected] (Y. Wang).
https://doi.org/10.1016/j.envpol.2019.112989 0269-7491/© 2019 Published by Elsevier Ltd.
(N2O), methane (CH4) and hydrogen sulfide (H2S) (Xiao et al., 2017). Among these harmful gases, CO2, N2O and CH4 belong to greenhouse gases, which will aggravate the global warming trend in a large area of water system (Gao et al., 2018). Meanwhile, as a typical odor gas, H2S would not only bring unpleasant odor and environmental impact, but also cause serious harm to the aquatic system, human health and ecological environment (Fang et al., 2016). Furthermore, the emissions of harmful gases under anoxic condition will also cause other adverse effects (such as sediment floating, turbidity and suspended matter increasing, and transparency decreasing), thereby reducing the sensory effects (visual sense and olfactory sense) of landscape water (Lee et al., 2017). Therefore, researches on the control of harmful gases emissions in the water system have been paid more attention in recent years.
2
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
At present, the methods of inhibiting harmful gases emissions mainly include increasing the dissolved oxygen (DO) and oxidation-reduction potential (ORP) of landscape water by aerating or adding oxidant (Ziegler et al., 2018). Although these methods can be used to control the emissions of harmful gases to some extent, they all have their nonnegligible defects. Aeration can blow off harmful gases that have been produced in the water system to a certain extent, but it would cause a sharp increase of harmful gases concentrations in the local atmospheric environment, thus its essence is just a transfer of harmful gases rather than inhibition (Zhangliang et al., 2018). Moreover, the aeration could cause the hydraulic disturbance of landscape water, and accelerate the pollutants released from sediment (Boog et al., 2017). As for oxidants addition, calcium nitrate would lead to a significant increase of nitrate concentration in the overlying water (Liu et al., 2017); ozone and hydrogen peroxide (H2O2) have short life cycle and are toxic to aquatic organisms at higher concentrations (Wert et al., 2014); chlorine and hypermanganate could introduce toxic substances (chloride and manganese ions) to further produce various byproducts and precursors (Shin et al., 2018). Consequently, it is urgent to find a more safe, efficient, stable and economical method to control harmful gases emissions in anoxic landscape water. As an environment-friendly oxidant, calcium peroxide (CaO2) can slowly release oxygen and strong oxidizing free radicals in humid air and water (Wang et al., 2016). Compared with traditional oxidants, it has the advantages of slow oxygen release, long lifetime, high stability and purity, low cost and environment safe (Lu et al., 2017). Our previous studies demonstrated that CaO2 could effectively increase the DO concentration and ORP level of water and sediment, obviously inhibit the pollutants release from sediment, and fundamentally avoid the occurrence of eutrophication in anoxic landscape water (Wang et al., 2018b; Wang et al., 2019). However, whether CaO2 has the ability to inhibit the emissions of harmful gases is unclear. If it can, how much is the optimal dose should be applied to the anoxic landscape water system to achieve a better effect. In order to answer the above questions, a new experiment was designed with different dosages of CaO2 to test how well it could lower the harmful gases emissions in a simulated anoxic/anaerobic landscape water system. The main objectives of this study were as followings: C To explore the impact of CaO2 dose on the physicochemical parameters and microbial community in the anoxic/anaerobic water system. C To investigate the dose effects and mechanism of CaO2 on inhibiting harmful gases emissions and to select the optimal dose by considering both the control effect and economic cost synthetically. C To provide a valuable reference and new technology for controlling harmful gases emissions in the anoxic/anaerobic landscape water system.
water was relatively clear and used directly in the experiment. The sediment was filtered with a 30 screen mesh to remove stones, leaves and other impurities. The filtered sediment had then been settled for 12 h and the supernatant was dumped before homogenization. The final sediment had a dark black color and a distinct odor after the pretreated procedure as aforementioned and would be used in the simulated experiment. CaO2 used in this experiment was purchased from Shanghai Macklin Biochemical Co., Ltd, with a purity of 72.08%. 2.2. Experimental device and design The experiment was carried out in four glass reactors, which had a diameter of 23 cm, a height of 40 cm and an effective volume of 10 L. On the top of the reactor, there was a rubber stopper with a diameter of 13 cm, where four holes were arranged on each rubber stopper for charging nitrogen gas, collecting the overlying water, sediment and gas samples respectively. The overlying water was added to the four reactors which had pre-loaded with 3 L sediments to the calibration line of 10 L by siphoning. The experiment included three treatment groups dosed with different dosages of CaO2 and one control check (CK) which was without CaO2 addition. The reactors dosed with CaO2 of 2.5 g, 5.0 g and 7.5 g were regarded as treatment groups, which were amount to 0.06, 0.12 and 0.18 kgCaO2 m2-sediment respectively. Furthermore, the reactors were charged with nitrogen gas and shaded with aluminum foil to simulate the actual anoxic and dark condition of the landscape water at a certain depth. The experiment was conducted in summer, with an average water temperature of 32.3 ± 1.9 C. 2.3. Sampling Samples were collected in three different locations at the same depth of each reactor respectively, and the mean values of parallel samples were regarded as the representative values of the experiment results. The water at 5 cm below the surface water and the sediment at 3 cm below the surface sediment were collected by peristaltic pump (LONGER, L100-1S-2) and regarded as representative samples of the overlying water and sediment. The interstitial water was obtained by centrifuging the sediment through a high speed freezing centrifuge (Eppendorf, 5804R) and the gas sample was collected at 3 cm above the water surface by a microsyringe with a volume of 500 mL. Water samples were collected at 9:00 a.m. approximately every 7 days (corresponding to the day 0, 4, 11, 18, 25, 32, 39, 46, 53 and 60 of the experiment period). Sediment and interstitial water were collected at 9:00 a.m. basically every 14 days (corresponding to the day 0, 4, 18, 32, 46 and 60 of the experiment period), while the gas samples were collected on the day 18, 32, 46, 53 and 60. At the end of the experiment, the sediment under the same treatment was mixed and used as samples to determine the microbial diversity and composition. 2.4. Analysis of samples
2. Materials and methods 2.1. Test materials The sediment and overlying water used in the experiment were collected with the water collector and grab sediment collector from the northeast corner of Xi'an moat, where the eutrophication and black-odor phenomenon occur annually, leading the water to unpleasant colors and odors. The collected sediment and overlying water were quickly transported to the Key Laboratory of Northwest Water Resource, Environment and Ecology, MOE. The overlying
In the water quality monitoring, the DO, temperature (T) and ORP were detected with a HACH portable multifunctional water quality meter (HQ-30d). The pH value was monitored with a pH meter (PHS-3C), turbidity was measured using a spectrophotometer (XINMAO752N), and total chemical oxygen demand (TCOD) was analyzed by titrimetry. The soluble sulfide (S2) was measured by dimethyl-aniline spectrophotometry after pretreatment by NaOHezinc acetate, whereas SO2 4 was analyzed by ion chromatography (Anion Chromatograph, ICS 1100). The measurements of ORP and pH in sediment were similar to
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
those of water quality. The analyses of microbial diversity and bacteria community structure in the sediments of different reactors were achieved with the Illumina MiSeq system, which mainly included DNA extraction, PCR amplification, high-throughput sequencing, bioinformatics and statistical analysis. In the analysis of gases, the gas composition was detected by the gas chromatograph (BEIFEN Corp.3420A). The parameters of the chromatograph were set as follows: TDX-01 packed column was selected as the chromatographic column and TCD thermal conductivity detector was used for detection. In the process of detection, the argon was used as carrier gas with a flow rate of 50 mL min1, the column temperature was set at 80 C and the detector temperature was set at 100 C. In addition, the instrument was calibrated by the calibration normalization method before sampling. The selected standard gas was consisted of 65.42% methane, 29.68% carbon dioxide, 4.41% nitrogen and 0.49% hydrogen. The H2S concentration was measured with the gas chromatograph (Clarus, PE 600), with a chromatographic column of PoraPakQ, a column temperature of 120 C, and the FPD detector temperature of 350 C. The carrier gas used in the measurement was nitrogen with a flow rate of 40 mL min1, the flow rates of air and hydrogen were set to 98 mL min1 and 76 mL min1, respectively. The instrument was calibrated by the H2S standard gas with the original concentration of 2000 ppm, and the H2S concentration gradient during the calibration process were set as 0, 400, 800, 1200, 1600, and 2000 ppm. 2.5. Statistical analysis The experimental data were calculated and plotted using the
3
Excel and Origin 9.0 software. In the analyses of sediments microorganisms, the base sequence obtained from high-throughput sequencing was first decontaminated using Trimmomatic, and high-quality double-end sequences were connected by the Flash software to obtain optimized sequences. Meanwhile, the base sequences of microorganisms in sediments were clustered into operational taxonomic unit (OTU) at a 97% identity threshold with the Uparse software, and the OTU representative sequences were compared with corresponding databases to obtain OTU species information with RDP classifier. Furthermore, the R language barplot and the R language Venn Diagram packages were used to analyze Alpha diversity of the strains and to plot the species abundance diagram and Venn map of different samples. The average values, standard deviation and variance homogeneity were analyzed using SPSS software (IBM SPSS Statistics 22.0). The significant difference of means was tested using t-test method with a significance level of p < 0.05. The data in the figures and tables were all the mean values and standard errors of parallel samples. 3. Results and discussion 3.1. Effects of CaO2 dosages on harmful gases emissions 3.1.1. Effects of CaO2 dosages on DO, ORP and pH in landscape water system The time course of DO concentration and ORP level in the water and sediment of different reactors were shown in Fig. 1, which indicated that all the dosages of CaO2 could elevate DO and ORP, especially in the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment.
Fig. 1. Time course of DO and ORP in the various reactors under different CaO2 dosages.
4
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
As shown in Fig. 1(a) and (b), the initial concentrations of DO in the overlying water and at sediment-water interface were 0.02 and 0.01 mg L1, which belonged to the severely anoxic conditions. Meanwhile, similar DO concentration trends were observed both in the overlying water and at sediment-water interface, which demonstrated that the DO concentrations of CK remained steadily at a low level throughout the test, while others were increased obviously with CaO2 additions. Furthermore, the DO concentrations in the reactor dosed with 0.06 kg-CaO2 m2-sediment, were increased first and then declined gradually, while others were increased gradually and kept nearly at a constant (＞3.0 mg L1) with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. In the whole process, the mean concentrations of DO in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 0.03, 0.93, 2.62 and 2.80 mg L1 respectively, while the corresponding concentrations at the sediment-water interface were 0.01, 0.78, 2.52 and 2.65 mg L1. Therefore, adding CaO2 with different dosages could enhance the DO concentration of the anoxic water system compared with CK significantly (p < 0.05), while the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment had better performances for a long time (p < 0.01), which was probably attributed to the characteristic anen et al., 2012). of slow-release oxygen of CaO2 (Nyk€ As shown in Fig. 1(c) and (d), the ORP values in the overlying water of all the samples were decreased sharply at the beginning of the test due to the initial turbulence of sediments and the anaerobic condition created by nitrogen pumping in. Subsequently, the ORP values in the overlying water of CK kept decreasing continuously, while it was relatively steady during the whole test in the reactor with 0.06 kg m2 dose. However, the ORP values increased gradually in the reactors adding CaO2 with 0.12 and 0.18 kg m2 dosages. By contrast, the ORP values in the sediment of CK decreased continually, while others were all increased gradually with CaO2 additions. Furthermore, the growth rate of ORP value dosed with 0.06 kg-CaO2 m2-sediment was much lower than those with the larger dosages. The average values of ORP in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 291.7, 189.1, 69.6 and 51.5 mV respectively, while the corresponding values in the sediments were 350.9, 260.2, 160.3 and 151.2 mV. It was reported that CaO2 could not only release oxygen and hydrogen peroxide, but also produce strong oxidizing free radicals of hydroxyl and superoxide, thereby elevating the ORP level of anoxic landscape water when it dissolved in water (Qian et al., 2016). Consequently, CaO2 addition could elevate the ORP level of the overlying water and sediment significantly compared with CK (p < 0.05), especially in the dosages of 0.12 and 0.18 kg-CaO2 m2-
sediment (p < 0.01). Fig. 2 shows that the time course of pH values in the water systems under different CaO2 dosages. As shown in Fig. 2, the pH values in the overlying water of all the reactors were increased first and then decreased. However, the pH values in the sediments of CK and 0.06 kg m2 dose declined gradually while those were basically stable in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment. The initial values of pH in the overlying water and sediment were 7.18 and 7.12. However, the minimal pH values in CK and 0.06 kg m2 dose were decreased to 6.33 and 6.71 in the overlying water, and 6.23 and 6.72 in the sediment at the end of the test, while those were neutral or weak alkaline environment under the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment throughout the test. Therefore, CaO2 could avoid the acidification of water and sediment under anoxic condition. However, the pH value of the overlying water dosed with 0.18 kg-CaO2 m2-sediment could reach up to 9.32 during a short period, which exceeded the pH threshold value (6.0 < pH < 9.0) for Chinese Surface Water Quality Standards (GB3838-2002). That implied the buffering effect of the sediment on the pH value of the overlying water was connected with the dose adopted, especially in a limited environment. In summary, CaO2 could effectively improve the hypoxic condition of landscape water system and avoid the acidification of water and sediment. The improvement effect was positively correlated with the CaO2 dose, but the pH value of the overlying water rose sharply to exceed Chinese Surface Water Safety Threshold in a short term under the 0.18 kg m2 dose.
3.1.2. Effects of CaO2 dosages on microbial population in sediment 3.1.2.1. The OTU sequence of microorganisms. The sequencing coverages of sediment microorganism in four reactors were greater than 99%, which indicated that the sequencing depth of the highthroughput analysis was sufficient (Chen et al., 2016). Meanwhile, the values of Chao index in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 1062, 1176, 1315 and 1404, respectively, while the corresponding values of ACE index were 1038, 1162, 1311 and 1389. Furthermore, Fig. 3 shows that the total OTU numbers of sediment microorganism in CK, 0.06, 0.12 and 0.18 kg m2 dosages were 982, 1098, 1205 and 1233 respectively. The mutual OTU was 736 in the overlapping region of four reactors, which indicated the species compositions of CK and 0.06 kg m2 dose were similar, while those were alike in the reactors dosed with 0.12 and 0.18 kgCaO2 m2-sediment. Therefore, CaO2 could obviously increase the total number and diversity of microorganism in the sediments.
Fig. 2. Time course of pH values in the various reactors under different CaO2 dosages.
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
Fig. 3. Venn diagram of microorganism OTU sequence in sediments under different CaO2 dosages.
3.1.2.2. The relative abundance of bacteria in genera level. Fig. 4 shows the species abundance (in genera level) in the sediments of different reactors. The bacteria with a relative abundance of less than 0.3% and the undetermined class were uniformly defined by “Others.” In the experiment, the detected species (in genera level) in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 168, 173, 184 and 186, respectively, while the corresponding values of species were 22, 23, 24 and 24 without considering “Others.” Although the relative abundances of species in the sediments of different reactors were various, the dominant strains were similar. The dominant species in these four reactors were Petrimonas (7.16%e 11.37%), Proteus Hauser (2.52%e6.03%), Smithella (6.31%e12.63%)
5
and Anaerolineaceae (8.25%e13.22%). In addition to the dominant species, the relative abundances of other bacteria also were varied with CaO2 addition. The relative abundances in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 2.52%, 1.38%, 0.56% and 0.51%, respectively, for Caldisericum; 5.21%, 3.19%, 0.52% and 0.38% for Syntrophorhabdus; whereas those were 0.58%, 1.22%, 2.67% and 2.78% for Planctomyces. Furthermore, the unique bacteria included Gemella, Haemophilus, and Streptococcus appeared in the sediments of CK and 0.06 kg m2 dose, while those included Nitrobacter, Nitrosomonas, Syntrophomonas and Thiobacillus became visible in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment, besides these same species. So, the microbial community structures in sediments of CK and 0.06 kg m2 dosage were similar, while those were alike in 0.12 and 0.18 kg m2 dosages. In general, CaO2 could significantly increase the total number and diversity of bacteria in sediments, enhance the abundance of aerobic bacteria, reduce the anaerobic bacteria richness, and derive some functional bacteria adapted to new environment condition, especially in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2sediment (Gholami et al., 2018). 3.1.3. Effects of CaO2 dosages on harmful gases in landscape water system 3.1.3.1. Gas composition in different reactors. As shown in Table 1, N2, CH4 and CO2 accounted for 64.80%, 24.62% and 10.58% of the total gas composition in CK, respectively, which indicated that it had a great quantity of greenhouse gases emissions in anoxic landscape water without CaO2 addition. In the treatment groups with CaO2 additions, the N2 proportion in the gas composition of different reactors were all increased significantly (p < 0.05), while the proportions of CH4 and CO2 were decreased accordingly (p < 0.01). Meanwhile, the more CaO2 was added, the less greenhouse gases were detected. In addition, the gas compositions in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment were similar (p > 0.05), and the CO2 concentration in the gas composition
Fig. 4. Relative abundance of bacteria (in genera level) in sediments under different CaO2 dosages.
6
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
Table 1 Gas composition in various reactors under different CaO2 dosages. Gas Composition
N2
CH4
CO2
CK 0.06 kg m2 0.12 kg m2 0.18 kg m2
64.80 ± 4.92% 76.51 ± 2.73% 91.86 ± 2.58% 94.02 ± 1.32%
24.62 ± 3.54% 16.20 ± 1.90% * 6.13 ± 1.65% ** 5.98 ± 1.32% **
10.58 ± 2.43% 7.29 ± 0.91% ** 2.01 ± 1.29% ** BDL **
* ** **
Note: Values are the means ± standard deviations (n ¼ 5) of five measurements during the whole experiment (on the day 18, 32, 46, 53 and 60). BDL means that the measured value is below the detection limit. * Significant difference between the three treatment groups and CK at the level of p < 0.05. ** Significant difference between the three treatment groups and CK at the level of p < 0.01.
of 0.18 kg m2 dosage was below the detection limit. Compared with CK, the dose of 0.12 kg-CaO2 m2-sediment could reduce 75.10% CH4 and 81.03% CO2 in gas composition, while it could reduce 75.71% CH4 and 100% CO2 with 0.18 kg m2 dosage. Therefore, the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment could effectively control the anaerobic fermentation of the sediment, which would increase the proportion of N2 in the gas composition while decrease the proportions of greenhouse gases (CH4 and CO2). 3.1.3.2. The emissions of H2S in different reactors. Table 2 shows the time course of H2S concentration in various reactors. As shown in this table, the H2S concentration in CK was increased sharply and then decreased, while it was stable first and then increased quickly in the reactor dosed with 0.06 kg-CaO2 m2-sediment. The mean concentrations of H2S in the reactors with CK and 0.06 kg m2 dosage were 1350.95 and 782.31 ppm, which were far beyond its lethal concentration to aquatic organism (such as fish, shellfish and shrimp, etc.) (Bagarinao, 1992). It was noteworthy that the H2S concentrations in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment were below the detection limit, which indicated that CaO2 with different dosages all had significant effect on inhibiting H2S emission in anoxic water than other researches (p < 0) (Wu et al., 2018). Therefore, CaO2 could effectively control the production of H2S in the anoxic landscape water, and completely inhibit H2S emission with the dosages of 0.12 and 0.18 kg m2, thereby fundamentally avoiding its unpleasant odor and high toxicity to ecological environment. 3.1.4. Effects of CaO2 dosages on sulfur transformation in water system The effects of CaO2 on sulfur transformation in water system of different reactors are shown in Fig. 5. As shown in this figure, the S2 concentrations in different zones of CK and 0.06 kg m2 dosage were all increased with time progressed, while those were decreased with the dosages of 0.12 and 0.18 kg m2. On the contrary, the SO2 concentrations in the overlying water and 4 Table 2 Time course of H2S concentration in various reactors under different CaO2 dosages. H2S (ppm)
CK
0.06 kg m2
0.12 kg m2
0.18 kg m2
18th 32nd 46th 53rd 60th Mean value
1106.35 ± 36.22 1624.13 ± 35.22 1353.06 ± 36.22 1236.86 ± 51.28 1434.35 ± 40.37 1350.95 ± 39.86
627.52 ± 12.36 722.45 ± 15.33 767.69 ± 19.11 744.39 ± 17.20 1049.48 ± 16.15 782.31 ± 16.03 **
BDL BDL BDL BDL BDL BDL
BDL BDL BDL BDL BDL BDL
**
**
Note: Values are the means ± standard deviations (n ¼ 3) of parallel samples. BDL means that the measured value is below the detection limit. ** Significant difference between the three treatment groups and CK at the level of p < 0.01.
interstitial water of CK kept stable at a low level, but those were increased in the treatment groups with CaO2 additions and the growth trend was positively correlated with the CaO2 dose. The initial values of S2 concentration in the overlying water and interstitial water were 0.19 and 0.36 mg L1, which were considered as higher S2 concentrations. The S2 concentrations in CK and 0.06 kg m2 dose were increased gradually and reached the maximum values (0.73 and 0.35 mg L1 for the overlying water; 0.95 and 0.46 mg L1 for the interstitial water) at the end of the test, while those were decreased in the reactors with dosages of 0.12 and 0.18 kg m2 and reached the minimum values (0.02 and 0.01 mg L1 for the overlying water; 0.11 and 0.08 mg L1 for the interstitial water) at the same time. Contrarily, the initial values of SO2 4 concentration in the overlying water and interstitial water were 0.08 and 0.12 mg L1, which were normally categorized as low levels. 2 The variations of SO2 dose 4 concentration in CK and 0.06 kg m were small, while those were enhanced steadily in the reactors with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. In the experiment, the mean values of S2 concentration in the overlying water and interstitial water of CK were 0.54 and 0.65 mg L1. Therefore, the S2 concentrations in the reactors with the dosages of 0.06, 0.12 and 0.18 kg m2 were 40.82% (p ¼ 0), 18.48% (p ¼ 0) and 16.61% (p ¼ 0) of that in CK for the overlying water, and 56.82% (p ¼ 0.005), 30.77% (p ¼ 0) and 27.79% (p ¼ 0) of that in CK for the interstitial water. Furthermore, the mean SO2 4 concentrations in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 0.18, 4.22, 14.70 and 16.83 mg L1, respectively, throughout the test, and the corresponding concentrations in the interstitial water were 0.27, 2.11, 6.51 and 7.06 mg L1. However, there were no significant differences of S2 and SO2 concentrations between 0.12 and 4 0.18 kg m2 dosages in the overlying water (p ¼ 0.908 and p ¼ 0.501) and interstitial water (p ¼ 0.851 and p ¼ 0.838). Therefore, CaO2 could obviously promote sulfur oxidation and inhibit its reduction in the landscape water, thereby, significantly decreasing the S2 concentration (p < 0.01) and increasing the SO2 4 concentration (p < 0.05) with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. 3.1.5. Effects of CaO2 dosages on TCOD and turbidity in the overlying water Fig. 6 shows the time course of TCOD and turbidity in the overlying water. Compared with CK, the TCOD concentration and turbidity in the treatment groups with CaO2 additions were all decreased obviously, and the downward trend was positively correlated with the dose of CaO2. As shown in Fig. 6(a), the initial value of TCOD concentration in the overlying water was 65.38 mg L1, and it was increased constantly and reached to the maximum value of 106.36 mg L1 (at the end of the test) in CK. However, the TCOD concentration in the overlying water with 0.06 kg m2 dose remained stable relatively but it decreased steadily to the minimum values of 42.31 and 40.28 mg L1 respectively (at the end of the test) under 0.12 and 0.18 kg m2 dosages. Throughout the test, the mean TCOD concentrations in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 91.86, 63.64, 52.50 and 49.78 mg L1, respectively. So, CaO2 could inhibit the organic matter release from sediment and degrade a part of the organic matter due to its strong oxidizing property, which led to a significant reduction (p < 0.01) of TCOD concentration in the overlying water (Wang et al., 2017a). As shown in Fig. 6(b), the turbidity in all reactors were increased first and decreased later, but the values in the treatment groups were much lower than that in CK. The turbidity in CK and 0.06 kg m2 dose were increased sharply to the maximum values of 82.50 and 49.02 NTU on the 11th of the test, and then reduced gradually. But the turbidity remained higher than 15 NTU in the
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
7
Fig. 5. Time course of S2 and SO2 4 concentrations in the overlying water and interstitial water under different CaO2 dosages.
Fig. 6. Time course of TCOD and turbidity in the overlying water under different CaO2 dosages.
whole process. However, the turbidity in the reactors with the CaO2 dosages of 0.12 and 0.18 kg m2 were elevated first to the maximum values of 18.16 and 18.04 NTU on the 11th of the test, and then decreased to the minimum values of 2.02 and 1.37 NTU in the later stage. The mean values of turbidity in the overlying water with the dosages of 0.06, 0.12 and 0.18 kg-CaO2 m2-sediment were 21.69, 8.70, and 7.85 NTU, which were 39.86% (p ¼ 0), 15.99% (p ¼ 0) and
14.43% (p ¼ 0) of that in CK respectively. The reduced turbidity in the overlying water through CaO2 addition could be ascribed to the improvement of the anoxic condition in water system, which caused the suspended solids back to the sediment by inhibiting the harmful gases productions and emissions (Glunk et al., 2011). In summary, CaO2 with the dosages of 0.12 and 0.18 kg m2 could obviously reduce the values of TCOD concentration and
8
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
turbidity in the overlying water, thereby, effectively restoring the sensory effect of anoxic landscape water. 3.2. Feasibility of CaO2 on inhibiting harmful gases emissions in anoxic landscape water system The anoxic environment is the main reason of harmful gases emissions and sensory effect decrease. So improving the hypoxia condition is the key to treat the blackening and stinking of landscape water (Andreasen et al., 2013). Compared with other methods (such as sediment dredging, diversion dilution and submerged aeration), in-situ addition of oxygen-releasing material has the advantages of cost-effective, convenience and no secondary pollution. Additives commonly used in the anoxic water are H2O2, MgO2, Ca(NO3)2 and others. However, H2O2 has a short period of lifetime, while the purity of MgO2 is low, and Ca(NO3)2 will increase the nitrogen concentration in water (Shen et al., 2013). Being alkaline-earth metal peroxide, CaO2 has a good thermal stability and can release oxygen slowly in humid air and water without other adverse effects. Therefore, CaO2 can be theoretically used to improve anoxic condition of landscape water, thereby inhibiting harmful gases emissions and restoring its landscape effect. It was reported that the lower DO and ORP were the main reasons of harmful gases emissions in the anoxic water, and the sediment ORP below 300 mV was the precursor to this problem (Wang et al., 2014). CaO2 could chronically release O2, H2O2, and strong oxidizing free radicals of HO and O2 in its dissolution process, thus significantly increased the DO concentration and ORP level in water and sediment (Xue et al., 2018). In this experiment, it was demonstrated that CaO2 with the dosages of 0.12 and 0.18 kg m2 could effectively increase the DO concentrations in the overlying water and at sediment-water interface, and keep them up to above 3.0 mg L1 for a long time. Meanwhile CaO2 would increase the ORP level in sediment from 312.6 mV to at least 69.6 mV with the dosages of 0.12 and 0.18 kg m2, while the 0.06 kg m2 dose had a weaker effect on improving anoxic condition. Under anoxic condition, organic sulfur compounds and sulfates are converted to H2S, HS and S2 with the action of sulfate reducing bacteria, among which H2S is the main olfactory substance and has a strong toxicity (Niu et al., 2018). Meanwhile, it will not only produce odor compounds, but also emit a large number of greenhouse gases (such as CH4 and CO2) in the process of water blackening and smelling (Chen et al., 2018). In the experiment, CaO2 increased the relative abundances of Proteus Hauser, Planctomyces and Verrucomicrobium, reduced the corresponding values of Bacteroides, Caldisericum, Clostridium, Desulfobacterium, Parabacteroides, and Smithella, and derived some new bacteria of Nitrobacter, Nitrosomonas, Syntrophomonas and Thiobacillus. As common aerobic bacteria in sediment, Proteus Hauser, Planctomyces and Verrucomicrobium can effectively degrade organic matter for cell synthesis, thus achieving the purpose of water purification (Chen et al., 2017). As typical anaerobic bacteria in anaerobic fermentation, Bacteroides, Clostridium, Parabacteroides, and Smithella can degrade alkanes to produce acetic acid, butyric acid, CH4 and CO2 under anaerobic conditions; while Caldisericum and Desulfobacterium can use small molecular organic-acid as electron acceptors to accelerate the reduction process of sulfate to S2 (Krishnan et al., 2017). Among these newly derived functional bacteria, Nitrobacter and Nitrosomonas can convert ammonia nitrogen to nitrate nitrogen to further improve the hypoxia condition of water system (Mellbye et al., 2018); Syntrophomonas can degrade organic acid to CO2 and H2O, thereby avoiding the acidification of anoxic water (Kong et al., 2018); and Thiobacillus can oxidize sulfide to sulfate, thus reducing the production of H2S and odor
compounds (Li et al., 2017). Therefore, CaO2 could effectively control the emissions of harmful gases (such as H2S, CO2 and CH4) and avoid the acidification and smelling of landscape water by changing microbial community in sediment. The blackening of water system is a necessary but insufficient condition for its smelling, because FeS and MnS are the main black substances, while COD is the main indicator to reflect its degree (Liu et al., 2009). Under the anoxic and reductive conditions, the complex bonds in metal oxides break to release metal ions, which then are combined with S2 to form the blackening substance. Some studies indicated that the blackening of water was positively correlated with the hypoxia degree and the concentrations of COD and S2 (Novotortsev et al., 2012). Furthermore, the increase of sediment floatation and suspended matter concentration in water caused by harmful gases emissions could also obviously reduce the transparency and sensory effects of landscape water. The addition of CaO2 improved the anoxic condition of landscape water significantly and reduced the COD and S2 concentrations in the overlying water and interstitial water obviously, thus alleviating the water blackening effectively. Meanwhile, the strong oxidizing free radicals produced by the dissolution of CaO2 have the functions of bleaching and disinfection, which can effectively reduce the water chromaticity (Yu et al., 2016). In addition, as a by-product of CaO2, calcium hydroxide is an excellent coagulant and could accelerate the removal of suspended particulate matters by settlement (Wang et al., 2013). Therefore, CaO2 could effectively control the productions of blackening compounds and odor substances, obviously reduce the turbidity in landscape water, thereby realizing the purpose of restoring the sensory effects of anoxic landscape water. In summary, this study showed that CaO2 could effectively improve the environment condition and microbial population in anoxic/anaerobic landscape water system, obviously inhibit harmful gases emissions and sulfate reduction, and notably promote the COD degradation and turbidity decrease. Furthermore, the lower price and convenient production of CaO2 also provide a higher probability for the application of CaO2 in practice. So, it is a feasible approach to adopt CaO2 for inhibiting harmful gases emissions in anoxic/anaerobic landscape water system. 3.3. The optimal dose of CaO2 for inhibiting harmful gases emissions Although the CaO2 with different dosages could inhibit harmful gases emissions and restore landscape function in anoxic water system, the overall performances of different dosages were diverse. The experiment showed that the improvement effect of adding CaO2 with 0.06 kg m2 dose was much lower than those of 0.12 and 0.18 kg m2 dosages. However, there was no significant difference between 0.12 and 0.18 kg m2 dosages in DO, ORP, TCOD and sulfur concentrations (p > 0.05). Meanwhile, the gas compositions in those two different dosages were similar, and the H2S concentration was below the detection limit. In addition, the sediments observed at different depth of CK were dark black, while the surface sediments in the treatment groups with CaO2 additions were oxidized to yellowish brown, and the thickness of the oxide layer was positively correlated with the dose of CaO2. There were obvious phenomena, such as bubble-escape, sediment-uplift, and dark brown overlying water, observed in CK and 0.06 kg m2 dosage. However, the above phenomena did not happen in the reactors which adding CaO2 with the dosages of 0.12 and 0.18 kg m2 and the overlying water was relatively clear and transparent. The unit price of CaO2 in this study was $1.16 kg1, so the 0.12 kg m2 dosage was equivalent to $0.14 m-2, which was far below the cost of other common technologies (such as sediment dredging, adding aluminum salt and artificial aeration) (Wang et al.,
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
2018a). Although the dosages of 0.12 and 0.18 kg m2 had similar effects on improving anoxic landscape water, the former could save 1/3 of the cost than the latter. Furthermore, the pH value in the overlying water dosed with 0.18 kg-CaO2 m2-sediment would climb up to 9.32, which exceeded the safety threshold of surface water and would bring potential threat to aquatic ecosystem. Therefore, comprehensively considering the overall performance, economic cost and negative effect, it was advisable to use 0.12 kgCaO2 m2-sediment as the optimal dose to inhibit harmful gases emissions. 4. Conclusions This study explored and compared overall performances of different CaO2 dosages on inhibiting harmful gases emissions in anoxic/anaerobic landscape water system. The optimal dose was selected through a comprehensive analysis. The main conclusions were as follows: 1) CaO2 could effectively improve the anoxic condition and sediment microbial community in landscape water system, obviously increase SO2 concentration while decrease S2 4 concentration in the water, and the improvement effect was positively correlated with the CaO2 dose. 2) CaO2 with the dosages of 0.12 and 0.18 kg m2 could significantly reduce the productions of greenhouse gases and completely control H2S emission, thereby improving the water quality by inhibiting anaerobic fermentation, harmful gases emissions, sediment floating and nutrients release in the anoxic/ anaerobic landscape water system. 3) The 0.12 kg-CaO2 m2-sediment was selected as the optimal dose, which could reduce 75.10% CH4, 81.02% CO2 and 100% H2S in gases, and decrease 81.52% S2, 42.85% TCOD and 84.01% turbidity in the overlying water. In summary, CaO2 could obviously improve the environmental condition, water quality and sediment microorganisms, effectively inhibit harmful gases emissions and blackening substances productions, and ultimately solve the blackening and stinking problem in anoxic/anaerobic landscape water system. Conflicts of interest The manuscript is an original work and has not been submitted or is under consideration for publication in another journal. We also confirm that all the listed authors have participated actively in the study and approved the submitted manuscript. The authors do not have any possible conflicts of interest. Acknowledgments This work was funded by the National Natural Science Foundation of China (Project No. 21677115). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.envpol.2019.112989. References Andreasen, Rune R., Liu, D., Ravn, Sebastian, Feilberg, Anders, Poulsen, Tjalfe G., 2013. Airdwater mass transfer of sparingly soluble odorous compounds in granular biofilter media. Chem. Eng. J. 220, 431e440. Bagarinao, T., 1992. Sulfide as an environmental factor and toxicant: tolerance and adaptations in aquatic organisms. Aquat. Toxicol. 24, 21e62.
9
~¼Ller, R.A., 2017. Resilience of Boog, J., Nivala, J., Aubron, T., Mothes, S., Van, A.M., Ma carbon and nitrogen removal due to aeration interruption in aerated treatment wetlands. Sci. Total Environ. 621, 960e969. Chen, C., Liu, Q., Liu, C., Yu, J., 2017. Effect of different enrichment strategies on microbial community structure in petroleum-contaminated marine sediment in Dalian, China. Mar. Pollut. Bull. 117, 274e282. Chen, L., Zhu, B., Guo, Y., Xu, T., Lee, J.S., Qian, P.Y., Zhou, B., Chen, L., Zhu, B., Guo, Y., 2016. High-throughput transcriptome sequencing reveals the combined effects of key e-waste contaminants, decabromodiphenyl ether (BDE-209) and lead, in zebrafish larvae. Environ. Pollut. 214, 324e333. Chen, Q., Guo, B., Zhao, C., Xing, B., 2018. Characteristics of CH4 and CO2 emissions and influence of water and salinity in the Yellow River delta wetland, China. Environ. Pollut. 239, 289e299. Fang, H., Liu, Z., Jin, Z., Zhang, L., Liu, D., Pei, Y., 2016. An emphasis of hydrogen sulfide-cysteine cycle on enhancing the tolerance to chromium stress in Arabidopsis *. Environ. Pollut. 213, 870e877. Gao, J., Hou, H., Zhai, Y., Woodward, A., Vardoulakis, S., Kovats, S., Wilkinson, P., Li, L., Song, X., Xu, L., 2018. Greenhouse gas emissions reduction in different economic sectors: mitigation measures, health co-benefits, knowledge gaps, and policy implications *. Environ. Pollut. 240, 683e698. Gholami, F., Shavandi, M., Dastgheib, S.M.M., Amoozegar, M.A., 2018. Naphthalene remediation form groundwater by calcium peroxide (CaO2) nanoparticles in permeable reactive barrier (PRB). Chemosphere 212, 105e113. Glunk, C., Dupraz, C., Braissant, O., Gallagher, K.L., Verrecchia, E.P., Visscher, P.T., 2011. Microbially mediated carbonate precipitation in a hypersaline lake, Big Pond (Eleuthera, Bahamas). Sedimentology 58, 720e736. Kong, X., Yu, S., Fang, W., Liu, J., Li, H., 2018. Enhancing syntrophic associations among Clostridium butyricum, Syntrophomonas and two types of methanogen by zero valent iron in an anaerobic assay with a high organic loading. Bioresour. Technol. 257, 181e191. Krishnan, Y., Bong, C.P.C., Azman, N.F., Zakaria, Z., Othman, N.A., Abdullah, N., Ho, C.S., Lee, C.T., Hansen, S.B., Hara, H., 2017. Co-composting of palm empty fruit bunch and palm oil mill effluent: microbial diversity and potential mitigation of greenhouse gas emission. J. Clean. Prod. 146, 94e100. Lee, J., Rai, P.K., Jeon, Y.J., Kim, K.H., Kwon, E.E., 2017. The role of algae and cyanobacteria in the production and release of odorants in water. Environ. Pollut. 227, 252e262. Li, L., Wang, T., Sun, Y., Wang, P., Yvette, B., Meng, J., Li, Q., Zhou, Y., 2017. Identify biosorption effects of Thiobacillus towards perfluorooctanoic acid (PFOA): pilot study from field to laboratory. Chemosphere 171, 31e39. Liu, G.F., Zhong, J.C., He, J., Zhang, L., Fan, C.X., 2009. Effects of black spots of deadcyanobacterial mats on Fe-S-P cycling in sediments of Zhushan Bay, Lake Taihu. Huanjing Kexue 30, 2520e2526. Liu, X., Tao, Y., Zhou, K., Zhang, Q., Chen, G., Zhang, X., 2017. Effect of water quality improvement on the remediation of river sediment due to the addition of calcium nitrate. Sci. Total Environ. 575, 887e894. Lu, S., Zhang, X., Xue, Y., 2017. Application of calcium peroxide in water and soil treatment: a review. J. Hazard. Mater. 337, 163e177. Mellbye, B.L., Giguere, A.T., Murthy, G.S., Bottomley, P.J., Sayavedra-Soto, L.A., Fwr, C., 2018. Genome-Scale, constraint-based modeling of nitrogen oxide fluxes during coculture of Nitrosomonas europaea and nitrobacter winogradskyi. mSystems 2018, 3. https://doi.org/10.1128/mSystems.00170-17. Niu, Z.S., Pan, H., Guo, X.P., Lu, D.P., Feng, J.N., Chen, Y.R., Tou, F.Y., Liu, M., Yang, Y., 2018. Sulphate-reducing bacteria (SRB) in the Yangtze Estuary sediments: abundance, distribution and implications for the bioavailability of metals. Sci. Total Environ. 634, 296e304. Novotortsev, V.M., Mel'Nikov, I.O., Busygina, N.S., Rodionova, S.A., Artemov, A.V., 2012. Examination of the disinfecting properties of calcium peroxide and its suitability for improving hydrodynamic characteristics of surface water bodies. Russ. J. Appl. Chem. 85, 726e730. Nyk€ anen, A., Kontio, H., Klutas, O., Penttinen, O.P., Kostia, S., Mikola, J., Romantschuk, M., 2012. Increasing lake water and sediment oxygen levels using slow release peroxide. Sci. Total Environ. 429, 317e324. Qian, Y., Jing, Z., Zhang, Y., Chen, J., Zhou, X., 2016. Degradation of 2,4dichlorophenol by nanoscale calcium peroxide: implication for groundwater remediation. Separ. Purif. Technol. 166, 222e229. Shen, Q., Liu, C., Zhou, Q., Shang, J., Zhang, L., Fan, C., 2013. Effects of physical and chemical characteristics of surface sediments in the formation of shallow lake algae-induced black bloom. J. Environ. Sci. 25, 2353e2360. Shin, J., Lee, D., Hwang, T.-M., Lee, Y., 2018. Oxidation kinetics of algal-derived taste and odor compounds during water treatment with ferrate(VI). Chem. Eng. J. 334, 1065e1073. Wang, B.Y., Chen, Z.L., Zhu, J., Shen, J.M., Han, Y., 2013. Pilot-scale fluoridecontaining wastewater treatment by the ballasted flocculation process. Water Sci. Technol. J. Int. Assoc. Water Pollut. Res. 68, 134e143. Wang, G., Li, X., Fang, Y., Huang, R., 2014. Analysis on the formation condition of the algae-induced odorous black water agglomerate. Saudi J. Biol. Sci. 21, 597e604. Wang, G.B., Wang, Y., Zhang, Y., 2018a. Combination effect of sponge iron and calcium nitrate on severely eutrophic urban landscape water: an integrated study from laboratory to fields. Environ. Sci. Pollut. Res. Int. 25, 1e14. Wang, H., Zhao, Y., Li, T., Zhen, C., Wang, Y., Qin, C., 2016. Properties of calcium peroxide for release of hydrogen peroxide and oxygen: a kinetics study. Chem. Eng. J. 303, 450e457. Wang, H., Zhao, Y., Yan, S., Li, T., Meng, Y., Qin, C., 2017a. Fenton-like degradation of 2,4-dichlorophenol using calcium peroxide particles: performance and
10
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
mechanisms. RSC Adv. 7, 4563e4571. Wang, W.H., Wang, Y., Fan, P., Chen, L.F., Chai, B.H., Zhao, J.C., Sun, L.Q., 2018b. Effect of calcium peroxide on the water quality and bacterium community of sediment in black-odor water. Environ. Pollut. 248, 18e27. Wang, W.H., Wang, Y., Li, Z., Wei, C.Z., Zhao, J.C., Sun, L.Q., 2017b. Effect of a strengthened ecological floating bed on the purification of urban landscape water supplied with reclaimed water. Sci. Total Environ. 622, 1630e1639. Wang, Y., Wang, W.H., Yan, F.L., Ding, Z., Feng, L.L., Zhao, J.C., 2019. Effects and mechanisms of calcium peroxide on purification of severely eutrophic water. Sci. Total Environ. 650, 2796e2806. Wert, E.C., Korak, J.A., Trenholm, R.A., Rosario-Ortiz, F.L., 2014. Effect of oxidant exposure on the release of intracellular microcystin, MIB, and geosmin from three cyanobacteria species. Water Res. 52, 251e259. Wu, H., Yu, Z., Bian, S., Ko, J.H., Li, S.F.Y., Xu, Q., 2018. H2S adsorption by municipal solid waste incineration (MSWI) fly ash with heavy metals immobilization. Chemosphere 195, 40e47.
Xiao, W., Liu, X., Irwin, A.J., Laws, E.A., Wang, L., Chen, B., Zeng, Y., Huang, B., 2017. Warming and eutrophication combine to restructure diatoms and dinoflagellates. Water Res. 128, 206e216. Xue, Y., Sui, Q., Brusseau, M.L., Zhang, X., Qiu, Z., Lyu, S., 2018. Insight on the generation of reactive oxygen species in the CaO2/Fe(II) Fenton system and the hydroxyl radical advancing strategy. Chem. Eng. J. 353, 657e665. Yu, D., Xie, P., Zeng, C., Xie, L., Chen, J., 2016. In situ enclosure experiments on the occurrence, development and decline of black bloom and the dynamics of its associated taste and odor compounds. Ecol. Eng. 87, 246e253. Zhangliang, H., Dezhi, S., Hui, W., Ruoyu, L., Zhiyuan, B., Fei, Q., 2018. Effects of ambient temperature and aeration frequency on emissions of ammonia and greenhouse gases from a sewage sludge aerobic composting plant. Bioresour. Technol. 270, 457e466. Ziegler, G., Tamburri, M.N., Fisher, D.J., 2018. Long-term algal toxicity of oxidant treated ballast water. Mar. Pollut. Bull. 133, 18e29.
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Dose effects of calcium peroxide on harmful gases emissions in the anoxic/anaerobic landscape water system* Wen-Huai Wang a, b, Yi Wang a, b, *, Jia-Jun Li a, b, Heng Zhang a, b, Fei-Long Yan a, b, Lu-Qin Sun c a b c
School of Environmental and Municipal Engineering, Xi'an University of Architecture and Technology, China Key Laboratory of Northwest Water Resource, Environment and Ecology, MOE, Xi'an University of Architecture and Technology, Xi'an 710055, China Environmental Science Department, University of San Francisco, California, CA 94117, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 4 April 2019 Received in revised form 15 July 2019 Accepted 31 July 2019 Available online 1 August 2019
Large-area hypoxia of urban landscape water often causes the emissions of harmful gases in summer, which not only reduces its sensory effects, but also brings a potential threat to aquatic ecosystem and human health. This study explored the dose effects of calcium peroxide (CaO2) on inhibiting harmful gases emissions and restoring the scenic effect (including visual sense and olfactory sense) of anoxic/ anaerobic landscape water system. The results indicated that the emissions of H2S, CO2 and CH4 from the anoxic/anaerobic water system were obviously inhibited in the reactors with CaO2 additions and the effect was positively correlated with the CaO2 dose. Meanwhile, the concentrations of total chemical oxygen demand (TCOD) and soluble sulfide (S2), and turbidity in the overlying water (the water-layer above the sediment-water interface) were also decreased in the reactors dosed with CaO2. The reason was ascribed to the improvement of the anoxic/anaerobic condition in the water system and the increase of the species richness, bacteria count and aerobic microorganism abundance in sediment. Furthermore, 0.12 kg-CaO2 m2-sediment was selected as the optimal dose, which was based on considering the inhibiting effect of the harmful gases emissions, comprehensive influence and costs. Compared with control check (CK, the reactor without adding CaO2), the optimal dose of CaO2 could reduce 75.10% CH4, 81.02% CO2 and 100% H2S in gases, and decrease 81.52% S2, 42.85% TCOD and 84.01% turbidity in the overlying water. In conclusion, all the dosages of CaO2 could improve the anoxic condition of water system and 0.12 kg-CaO2 m2-sediment was the optimal dose in inhibiting harmful gases emissions, which could keep an excellent water quality in this simulation experiment. Therefore, this study may provide a feasible method and the optimal dose for inhibiting the emissions of harmful gases and restoring the scenic effect in the similar anoxic/anaerobic landscape water. © 2019 Published by Elsevier Ltd.
Keywords: Calcium peroxide dosages Landscape water system Anoxic/anaerobic condition Harmful gases Sulfur transformation
1. Introduction Urban landscape water is normally located in the populated area (Wang et al., 2017b). It is widely believed that exogenous pollution and endogenous release from the sediments are the main reasons for the water system to become anoxic or anaerobic (especially in summer), which would further lead to the generation and emissions of harmful gases, such as carbon dioxide (CO2), nitrous oxide
* This paper has been recommended for acceptance by Maria Cristina Fossi. * Corresponding author. School of Environmental and Municipal Engineering, Xi'an University of Architecture and Technology, China. E-mail addresses: [email protected], [email protected] (Y. Wang).
https://doi.org/10.1016/j.envpol.2019.112989 0269-7491/© 2019 Published by Elsevier Ltd.
(N2O), methane (CH4) and hydrogen sulfide (H2S) (Xiao et al., 2017). Among these harmful gases, CO2, N2O and CH4 belong to greenhouse gases, which will aggravate the global warming trend in a large area of water system (Gao et al., 2018). Meanwhile, as a typical odor gas, H2S would not only bring unpleasant odor and environmental impact, but also cause serious harm to the aquatic system, human health and ecological environment (Fang et al., 2016). Furthermore, the emissions of harmful gases under anoxic condition will also cause other adverse effects (such as sediment floating, turbidity and suspended matter increasing, and transparency decreasing), thereby reducing the sensory effects (visual sense and olfactory sense) of landscape water (Lee et al., 2017). Therefore, researches on the control of harmful gases emissions in the water system have been paid more attention in recent years.
2
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
At present, the methods of inhibiting harmful gases emissions mainly include increasing the dissolved oxygen (DO) and oxidation-reduction potential (ORP) of landscape water by aerating or adding oxidant (Ziegler et al., 2018). Although these methods can be used to control the emissions of harmful gases to some extent, they all have their nonnegligible defects. Aeration can blow off harmful gases that have been produced in the water system to a certain extent, but it would cause a sharp increase of harmful gases concentrations in the local atmospheric environment, thus its essence is just a transfer of harmful gases rather than inhibition (Zhangliang et al., 2018). Moreover, the aeration could cause the hydraulic disturbance of landscape water, and accelerate the pollutants released from sediment (Boog et al., 2017). As for oxidants addition, calcium nitrate would lead to a significant increase of nitrate concentration in the overlying water (Liu et al., 2017); ozone and hydrogen peroxide (H2O2) have short life cycle and are toxic to aquatic organisms at higher concentrations (Wert et al., 2014); chlorine and hypermanganate could introduce toxic substances (chloride and manganese ions) to further produce various byproducts and precursors (Shin et al., 2018). Consequently, it is urgent to find a more safe, efficient, stable and economical method to control harmful gases emissions in anoxic landscape water. As an environment-friendly oxidant, calcium peroxide (CaO2) can slowly release oxygen and strong oxidizing free radicals in humid air and water (Wang et al., 2016). Compared with traditional oxidants, it has the advantages of slow oxygen release, long lifetime, high stability and purity, low cost and environment safe (Lu et al., 2017). Our previous studies demonstrated that CaO2 could effectively increase the DO concentration and ORP level of water and sediment, obviously inhibit the pollutants release from sediment, and fundamentally avoid the occurrence of eutrophication in anoxic landscape water (Wang et al., 2018b; Wang et al., 2019). However, whether CaO2 has the ability to inhibit the emissions of harmful gases is unclear. If it can, how much is the optimal dose should be applied to the anoxic landscape water system to achieve a better effect. In order to answer the above questions, a new experiment was designed with different dosages of CaO2 to test how well it could lower the harmful gases emissions in a simulated anoxic/anaerobic landscape water system. The main objectives of this study were as followings: C To explore the impact of CaO2 dose on the physicochemical parameters and microbial community in the anoxic/anaerobic water system. C To investigate the dose effects and mechanism of CaO2 on inhibiting harmful gases emissions and to select the optimal dose by considering both the control effect and economic cost synthetically. C To provide a valuable reference and new technology for controlling harmful gases emissions in the anoxic/anaerobic landscape water system.
water was relatively clear and used directly in the experiment. The sediment was filtered with a 30 screen mesh to remove stones, leaves and other impurities. The filtered sediment had then been settled for 12 h and the supernatant was dumped before homogenization. The final sediment had a dark black color and a distinct odor after the pretreated procedure as aforementioned and would be used in the simulated experiment. CaO2 used in this experiment was purchased from Shanghai Macklin Biochemical Co., Ltd, with a purity of 72.08%. 2.2. Experimental device and design The experiment was carried out in four glass reactors, which had a diameter of 23 cm, a height of 40 cm and an effective volume of 10 L. On the top of the reactor, there was a rubber stopper with a diameter of 13 cm, where four holes were arranged on each rubber stopper for charging nitrogen gas, collecting the overlying water, sediment and gas samples respectively. The overlying water was added to the four reactors which had pre-loaded with 3 L sediments to the calibration line of 10 L by siphoning. The experiment included three treatment groups dosed with different dosages of CaO2 and one control check (CK) which was without CaO2 addition. The reactors dosed with CaO2 of 2.5 g, 5.0 g and 7.5 g were regarded as treatment groups, which were amount to 0.06, 0.12 and 0.18 kgCaO2 m2-sediment respectively. Furthermore, the reactors were charged with nitrogen gas and shaded with aluminum foil to simulate the actual anoxic and dark condition of the landscape water at a certain depth. The experiment was conducted in summer, with an average water temperature of 32.3 ± 1.9 C. 2.3. Sampling Samples were collected in three different locations at the same depth of each reactor respectively, and the mean values of parallel samples were regarded as the representative values of the experiment results. The water at 5 cm below the surface water and the sediment at 3 cm below the surface sediment were collected by peristaltic pump (LONGER, L100-1S-2) and regarded as representative samples of the overlying water and sediment. The interstitial water was obtained by centrifuging the sediment through a high speed freezing centrifuge (Eppendorf, 5804R) and the gas sample was collected at 3 cm above the water surface by a microsyringe with a volume of 500 mL. Water samples were collected at 9:00 a.m. approximately every 7 days (corresponding to the day 0, 4, 11, 18, 25, 32, 39, 46, 53 and 60 of the experiment period). Sediment and interstitial water were collected at 9:00 a.m. basically every 14 days (corresponding to the day 0, 4, 18, 32, 46 and 60 of the experiment period), while the gas samples were collected on the day 18, 32, 46, 53 and 60. At the end of the experiment, the sediment under the same treatment was mixed and used as samples to determine the microbial diversity and composition. 2.4. Analysis of samples
2. Materials and methods 2.1. Test materials The sediment and overlying water used in the experiment were collected with the water collector and grab sediment collector from the northeast corner of Xi'an moat, where the eutrophication and black-odor phenomenon occur annually, leading the water to unpleasant colors and odors. The collected sediment and overlying water were quickly transported to the Key Laboratory of Northwest Water Resource, Environment and Ecology, MOE. The overlying
In the water quality monitoring, the DO, temperature (T) and ORP were detected with a HACH portable multifunctional water quality meter (HQ-30d). The pH value was monitored with a pH meter (PHS-3C), turbidity was measured using a spectrophotometer (XINMAO752N), and total chemical oxygen demand (TCOD) was analyzed by titrimetry. The soluble sulfide (S2) was measured by dimethyl-aniline spectrophotometry after pretreatment by NaOHezinc acetate, whereas SO2 4 was analyzed by ion chromatography (Anion Chromatograph, ICS 1100). The measurements of ORP and pH in sediment were similar to
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
those of water quality. The analyses of microbial diversity and bacteria community structure in the sediments of different reactors were achieved with the Illumina MiSeq system, which mainly included DNA extraction, PCR amplification, high-throughput sequencing, bioinformatics and statistical analysis. In the analysis of gases, the gas composition was detected by the gas chromatograph (BEIFEN Corp.3420A). The parameters of the chromatograph were set as follows: TDX-01 packed column was selected as the chromatographic column and TCD thermal conductivity detector was used for detection. In the process of detection, the argon was used as carrier gas with a flow rate of 50 mL min1, the column temperature was set at 80 C and the detector temperature was set at 100 C. In addition, the instrument was calibrated by the calibration normalization method before sampling. The selected standard gas was consisted of 65.42% methane, 29.68% carbon dioxide, 4.41% nitrogen and 0.49% hydrogen. The H2S concentration was measured with the gas chromatograph (Clarus, PE 600), with a chromatographic column of PoraPakQ, a column temperature of 120 C, and the FPD detector temperature of 350 C. The carrier gas used in the measurement was nitrogen with a flow rate of 40 mL min1, the flow rates of air and hydrogen were set to 98 mL min1 and 76 mL min1, respectively. The instrument was calibrated by the H2S standard gas with the original concentration of 2000 ppm, and the H2S concentration gradient during the calibration process were set as 0, 400, 800, 1200, 1600, and 2000 ppm. 2.5. Statistical analysis The experimental data were calculated and plotted using the
3
Excel and Origin 9.0 software. In the analyses of sediments microorganisms, the base sequence obtained from high-throughput sequencing was first decontaminated using Trimmomatic, and high-quality double-end sequences were connected by the Flash software to obtain optimized sequences. Meanwhile, the base sequences of microorganisms in sediments were clustered into operational taxonomic unit (OTU) at a 97% identity threshold with the Uparse software, and the OTU representative sequences were compared with corresponding databases to obtain OTU species information with RDP classifier. Furthermore, the R language barplot and the R language Venn Diagram packages were used to analyze Alpha diversity of the strains and to plot the species abundance diagram and Venn map of different samples. The average values, standard deviation and variance homogeneity were analyzed using SPSS software (IBM SPSS Statistics 22.0). The significant difference of means was tested using t-test method with a significance level of p < 0.05. The data in the figures and tables were all the mean values and standard errors of parallel samples. 3. Results and discussion 3.1. Effects of CaO2 dosages on harmful gases emissions 3.1.1. Effects of CaO2 dosages on DO, ORP and pH in landscape water system The time course of DO concentration and ORP level in the water and sediment of different reactors were shown in Fig. 1, which indicated that all the dosages of CaO2 could elevate DO and ORP, especially in the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment.
Fig. 1. Time course of DO and ORP in the various reactors under different CaO2 dosages.
4
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
As shown in Fig. 1(a) and (b), the initial concentrations of DO in the overlying water and at sediment-water interface were 0.02 and 0.01 mg L1, which belonged to the severely anoxic conditions. Meanwhile, similar DO concentration trends were observed both in the overlying water and at sediment-water interface, which demonstrated that the DO concentrations of CK remained steadily at a low level throughout the test, while others were increased obviously with CaO2 additions. Furthermore, the DO concentrations in the reactor dosed with 0.06 kg-CaO2 m2-sediment, were increased first and then declined gradually, while others were increased gradually and kept nearly at a constant (＞3.0 mg L1) with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. In the whole process, the mean concentrations of DO in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 0.03, 0.93, 2.62 and 2.80 mg L1 respectively, while the corresponding concentrations at the sediment-water interface were 0.01, 0.78, 2.52 and 2.65 mg L1. Therefore, adding CaO2 with different dosages could enhance the DO concentration of the anoxic water system compared with CK significantly (p < 0.05), while the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment had better performances for a long time (p < 0.01), which was probably attributed to the characteristic anen et al., 2012). of slow-release oxygen of CaO2 (Nyk€ As shown in Fig. 1(c) and (d), the ORP values in the overlying water of all the samples were decreased sharply at the beginning of the test due to the initial turbulence of sediments and the anaerobic condition created by nitrogen pumping in. Subsequently, the ORP values in the overlying water of CK kept decreasing continuously, while it was relatively steady during the whole test in the reactor with 0.06 kg m2 dose. However, the ORP values increased gradually in the reactors adding CaO2 with 0.12 and 0.18 kg m2 dosages. By contrast, the ORP values in the sediment of CK decreased continually, while others were all increased gradually with CaO2 additions. Furthermore, the growth rate of ORP value dosed with 0.06 kg-CaO2 m2-sediment was much lower than those with the larger dosages. The average values of ORP in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 291.7, 189.1, 69.6 and 51.5 mV respectively, while the corresponding values in the sediments were 350.9, 260.2, 160.3 and 151.2 mV. It was reported that CaO2 could not only release oxygen and hydrogen peroxide, but also produce strong oxidizing free radicals of hydroxyl and superoxide, thereby elevating the ORP level of anoxic landscape water when it dissolved in water (Qian et al., 2016). Consequently, CaO2 addition could elevate the ORP level of the overlying water and sediment significantly compared with CK (p < 0.05), especially in the dosages of 0.12 and 0.18 kg-CaO2 m2-
sediment (p < 0.01). Fig. 2 shows that the time course of pH values in the water systems under different CaO2 dosages. As shown in Fig. 2, the pH values in the overlying water of all the reactors were increased first and then decreased. However, the pH values in the sediments of CK and 0.06 kg m2 dose declined gradually while those were basically stable in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment. The initial values of pH in the overlying water and sediment were 7.18 and 7.12. However, the minimal pH values in CK and 0.06 kg m2 dose were decreased to 6.33 and 6.71 in the overlying water, and 6.23 and 6.72 in the sediment at the end of the test, while those were neutral or weak alkaline environment under the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment throughout the test. Therefore, CaO2 could avoid the acidification of water and sediment under anoxic condition. However, the pH value of the overlying water dosed with 0.18 kg-CaO2 m2-sediment could reach up to 9.32 during a short period, which exceeded the pH threshold value (6.0 < pH < 9.0) for Chinese Surface Water Quality Standards (GB3838-2002). That implied the buffering effect of the sediment on the pH value of the overlying water was connected with the dose adopted, especially in a limited environment. In summary, CaO2 could effectively improve the hypoxic condition of landscape water system and avoid the acidification of water and sediment. The improvement effect was positively correlated with the CaO2 dose, but the pH value of the overlying water rose sharply to exceed Chinese Surface Water Safety Threshold in a short term under the 0.18 kg m2 dose.
3.1.2. Effects of CaO2 dosages on microbial population in sediment 3.1.2.1. The OTU sequence of microorganisms. The sequencing coverages of sediment microorganism in four reactors were greater than 99%, which indicated that the sequencing depth of the highthroughput analysis was sufficient (Chen et al., 2016). Meanwhile, the values of Chao index in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 1062, 1176, 1315 and 1404, respectively, while the corresponding values of ACE index were 1038, 1162, 1311 and 1389. Furthermore, Fig. 3 shows that the total OTU numbers of sediment microorganism in CK, 0.06, 0.12 and 0.18 kg m2 dosages were 982, 1098, 1205 and 1233 respectively. The mutual OTU was 736 in the overlapping region of four reactors, which indicated the species compositions of CK and 0.06 kg m2 dose were similar, while those were alike in the reactors dosed with 0.12 and 0.18 kgCaO2 m2-sediment. Therefore, CaO2 could obviously increase the total number and diversity of microorganism in the sediments.
Fig. 2. Time course of pH values in the various reactors under different CaO2 dosages.
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
Fig. 3. Venn diagram of microorganism OTU sequence in sediments under different CaO2 dosages.
3.1.2.2. The relative abundance of bacteria in genera level. Fig. 4 shows the species abundance (in genera level) in the sediments of different reactors. The bacteria with a relative abundance of less than 0.3% and the undetermined class were uniformly defined by “Others.” In the experiment, the detected species (in genera level) in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 168, 173, 184 and 186, respectively, while the corresponding values of species were 22, 23, 24 and 24 without considering “Others.” Although the relative abundances of species in the sediments of different reactors were various, the dominant strains were similar. The dominant species in these four reactors were Petrimonas (7.16%e 11.37%), Proteus Hauser (2.52%e6.03%), Smithella (6.31%e12.63%)
5
and Anaerolineaceae (8.25%e13.22%). In addition to the dominant species, the relative abundances of other bacteria also were varied with CaO2 addition. The relative abundances in the sediments of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 2.52%, 1.38%, 0.56% and 0.51%, respectively, for Caldisericum; 5.21%, 3.19%, 0.52% and 0.38% for Syntrophorhabdus; whereas those were 0.58%, 1.22%, 2.67% and 2.78% for Planctomyces. Furthermore, the unique bacteria included Gemella, Haemophilus, and Streptococcus appeared in the sediments of CK and 0.06 kg m2 dose, while those included Nitrobacter, Nitrosomonas, Syntrophomonas and Thiobacillus became visible in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment, besides these same species. So, the microbial community structures in sediments of CK and 0.06 kg m2 dosage were similar, while those were alike in 0.12 and 0.18 kg m2 dosages. In general, CaO2 could significantly increase the total number and diversity of bacteria in sediments, enhance the abundance of aerobic bacteria, reduce the anaerobic bacteria richness, and derive some functional bacteria adapted to new environment condition, especially in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2sediment (Gholami et al., 2018). 3.1.3. Effects of CaO2 dosages on harmful gases in landscape water system 3.1.3.1. Gas composition in different reactors. As shown in Table 1, N2, CH4 and CO2 accounted for 64.80%, 24.62% and 10.58% of the total gas composition in CK, respectively, which indicated that it had a great quantity of greenhouse gases emissions in anoxic landscape water without CaO2 addition. In the treatment groups with CaO2 additions, the N2 proportion in the gas composition of different reactors were all increased significantly (p < 0.05), while the proportions of CH4 and CO2 were decreased accordingly (p < 0.01). Meanwhile, the more CaO2 was added, the less greenhouse gases were detected. In addition, the gas compositions in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment were similar (p > 0.05), and the CO2 concentration in the gas composition
Fig. 4. Relative abundance of bacteria (in genera level) in sediments under different CaO2 dosages.
6
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
Table 1 Gas composition in various reactors under different CaO2 dosages. Gas Composition
N2
CH4
CO2
CK 0.06 kg m2 0.12 kg m2 0.18 kg m2
64.80 ± 4.92% 76.51 ± 2.73% 91.86 ± 2.58% 94.02 ± 1.32%
24.62 ± 3.54% 16.20 ± 1.90% * 6.13 ± 1.65% ** 5.98 ± 1.32% **
10.58 ± 2.43% 7.29 ± 0.91% ** 2.01 ± 1.29% ** BDL **
* ** **
Note: Values are the means ± standard deviations (n ¼ 5) of five measurements during the whole experiment (on the day 18, 32, 46, 53 and 60). BDL means that the measured value is below the detection limit. * Significant difference between the three treatment groups and CK at the level of p < 0.05. ** Significant difference between the three treatment groups and CK at the level of p < 0.01.
of 0.18 kg m2 dosage was below the detection limit. Compared with CK, the dose of 0.12 kg-CaO2 m2-sediment could reduce 75.10% CH4 and 81.03% CO2 in gas composition, while it could reduce 75.71% CH4 and 100% CO2 with 0.18 kg m2 dosage. Therefore, the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment could effectively control the anaerobic fermentation of the sediment, which would increase the proportion of N2 in the gas composition while decrease the proportions of greenhouse gases (CH4 and CO2). 3.1.3.2. The emissions of H2S in different reactors. Table 2 shows the time course of H2S concentration in various reactors. As shown in this table, the H2S concentration in CK was increased sharply and then decreased, while it was stable first and then increased quickly in the reactor dosed with 0.06 kg-CaO2 m2-sediment. The mean concentrations of H2S in the reactors with CK and 0.06 kg m2 dosage were 1350.95 and 782.31 ppm, which were far beyond its lethal concentration to aquatic organism (such as fish, shellfish and shrimp, etc.) (Bagarinao, 1992). It was noteworthy that the H2S concentrations in the reactors dosed with 0.12 and 0.18 kg-CaO2 m2-sediment were below the detection limit, which indicated that CaO2 with different dosages all had significant effect on inhibiting H2S emission in anoxic water than other researches (p < 0) (Wu et al., 2018). Therefore, CaO2 could effectively control the production of H2S in the anoxic landscape water, and completely inhibit H2S emission with the dosages of 0.12 and 0.18 kg m2, thereby fundamentally avoiding its unpleasant odor and high toxicity to ecological environment. 3.1.4. Effects of CaO2 dosages on sulfur transformation in water system The effects of CaO2 on sulfur transformation in water system of different reactors are shown in Fig. 5. As shown in this figure, the S2 concentrations in different zones of CK and 0.06 kg m2 dosage were all increased with time progressed, while those were decreased with the dosages of 0.12 and 0.18 kg m2. On the contrary, the SO2 concentrations in the overlying water and 4 Table 2 Time course of H2S concentration in various reactors under different CaO2 dosages. H2S (ppm)
CK
0.06 kg m2
0.12 kg m2
0.18 kg m2
18th 32nd 46th 53rd 60th Mean value
1106.35 ± 36.22 1624.13 ± 35.22 1353.06 ± 36.22 1236.86 ± 51.28 1434.35 ± 40.37 1350.95 ± 39.86
627.52 ± 12.36 722.45 ± 15.33 767.69 ± 19.11 744.39 ± 17.20 1049.48 ± 16.15 782.31 ± 16.03 **
BDL BDL BDL BDL BDL BDL
BDL BDL BDL BDL BDL BDL
**
**
Note: Values are the means ± standard deviations (n ¼ 3) of parallel samples. BDL means that the measured value is below the detection limit. ** Significant difference between the three treatment groups and CK at the level of p < 0.01.
interstitial water of CK kept stable at a low level, but those were increased in the treatment groups with CaO2 additions and the growth trend was positively correlated with the CaO2 dose. The initial values of S2 concentration in the overlying water and interstitial water were 0.19 and 0.36 mg L1, which were considered as higher S2 concentrations. The S2 concentrations in CK and 0.06 kg m2 dose were increased gradually and reached the maximum values (0.73 and 0.35 mg L1 for the overlying water; 0.95 and 0.46 mg L1 for the interstitial water) at the end of the test, while those were decreased in the reactors with dosages of 0.12 and 0.18 kg m2 and reached the minimum values (0.02 and 0.01 mg L1 for the overlying water; 0.11 and 0.08 mg L1 for the interstitial water) at the same time. Contrarily, the initial values of SO2 4 concentration in the overlying water and interstitial water were 0.08 and 0.12 mg L1, which were normally categorized as low levels. 2 The variations of SO2 dose 4 concentration in CK and 0.06 kg m were small, while those were enhanced steadily in the reactors with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. In the experiment, the mean values of S2 concentration in the overlying water and interstitial water of CK were 0.54 and 0.65 mg L1. Therefore, the S2 concentrations in the reactors with the dosages of 0.06, 0.12 and 0.18 kg m2 were 40.82% (p ¼ 0), 18.48% (p ¼ 0) and 16.61% (p ¼ 0) of that in CK for the overlying water, and 56.82% (p ¼ 0.005), 30.77% (p ¼ 0) and 27.79% (p ¼ 0) of that in CK for the interstitial water. Furthermore, the mean SO2 4 concentrations in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 0.18, 4.22, 14.70 and 16.83 mg L1, respectively, throughout the test, and the corresponding concentrations in the interstitial water were 0.27, 2.11, 6.51 and 7.06 mg L1. However, there were no significant differences of S2 and SO2 concentrations between 0.12 and 4 0.18 kg m2 dosages in the overlying water (p ¼ 0.908 and p ¼ 0.501) and interstitial water (p ¼ 0.851 and p ¼ 0.838). Therefore, CaO2 could obviously promote sulfur oxidation and inhibit its reduction in the landscape water, thereby, significantly decreasing the S2 concentration (p < 0.01) and increasing the SO2 4 concentration (p < 0.05) with the dosages of 0.12 and 0.18 kg-CaO2 m2-sediment. 3.1.5. Effects of CaO2 dosages on TCOD and turbidity in the overlying water Fig. 6 shows the time course of TCOD and turbidity in the overlying water. Compared with CK, the TCOD concentration and turbidity in the treatment groups with CaO2 additions were all decreased obviously, and the downward trend was positively correlated with the dose of CaO2. As shown in Fig. 6(a), the initial value of TCOD concentration in the overlying water was 65.38 mg L1, and it was increased constantly and reached to the maximum value of 106.36 mg L1 (at the end of the test) in CK. However, the TCOD concentration in the overlying water with 0.06 kg m2 dose remained stable relatively but it decreased steadily to the minimum values of 42.31 and 40.28 mg L1 respectively (at the end of the test) under 0.12 and 0.18 kg m2 dosages. Throughout the test, the mean TCOD concentrations in the overlying water of CK, 0.06, 0.12 and 0.18 kg m2 dosages were 91.86, 63.64, 52.50 and 49.78 mg L1, respectively. So, CaO2 could inhibit the organic matter release from sediment and degrade a part of the organic matter due to its strong oxidizing property, which led to a significant reduction (p < 0.01) of TCOD concentration in the overlying water (Wang et al., 2017a). As shown in Fig. 6(b), the turbidity in all reactors were increased first and decreased later, but the values in the treatment groups were much lower than that in CK. The turbidity in CK and 0.06 kg m2 dose were increased sharply to the maximum values of 82.50 and 49.02 NTU on the 11th of the test, and then reduced gradually. But the turbidity remained higher than 15 NTU in the
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
7
Fig. 5. Time course of S2 and SO2 4 concentrations in the overlying water and interstitial water under different CaO2 dosages.
Fig. 6. Time course of TCOD and turbidity in the overlying water under different CaO2 dosages.
whole process. However, the turbidity in the reactors with the CaO2 dosages of 0.12 and 0.18 kg m2 were elevated first to the maximum values of 18.16 and 18.04 NTU on the 11th of the test, and then decreased to the minimum values of 2.02 and 1.37 NTU in the later stage. The mean values of turbidity in the overlying water with the dosages of 0.06, 0.12 and 0.18 kg-CaO2 m2-sediment were 21.69, 8.70, and 7.85 NTU, which were 39.86% (p ¼ 0), 15.99% (p ¼ 0) and
14.43% (p ¼ 0) of that in CK respectively. The reduced turbidity in the overlying water through CaO2 addition could be ascribed to the improvement of the anoxic condition in water system, which caused the suspended solids back to the sediment by inhibiting the harmful gases productions and emissions (Glunk et al., 2011). In summary, CaO2 with the dosages of 0.12 and 0.18 kg m2 could obviously reduce the values of TCOD concentration and
8
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
turbidity in the overlying water, thereby, effectively restoring the sensory effect of anoxic landscape water. 3.2. Feasibility of CaO2 on inhibiting harmful gases emissions in anoxic landscape water system The anoxic environment is the main reason of harmful gases emissions and sensory effect decrease. So improving the hypoxia condition is the key to treat the blackening and stinking of landscape water (Andreasen et al., 2013). Compared with other methods (such as sediment dredging, diversion dilution and submerged aeration), in-situ addition of oxygen-releasing material has the advantages of cost-effective, convenience and no secondary pollution. Additives commonly used in the anoxic water are H2O2, MgO2, Ca(NO3)2 and others. However, H2O2 has a short period of lifetime, while the purity of MgO2 is low, and Ca(NO3)2 will increase the nitrogen concentration in water (Shen et al., 2013). Being alkaline-earth metal peroxide, CaO2 has a good thermal stability and can release oxygen slowly in humid air and water without other adverse effects. Therefore, CaO2 can be theoretically used to improve anoxic condition of landscape water, thereby inhibiting harmful gases emissions and restoring its landscape effect. It was reported that the lower DO and ORP were the main reasons of harmful gases emissions in the anoxic water, and the sediment ORP below 300 mV was the precursor to this problem (Wang et al., 2014). CaO2 could chronically release O2, H2O2, and strong oxidizing free radicals of HO and O2 in its dissolution process, thus significantly increased the DO concentration and ORP level in water and sediment (Xue et al., 2018). In this experiment, it was demonstrated that CaO2 with the dosages of 0.12 and 0.18 kg m2 could effectively increase the DO concentrations in the overlying water and at sediment-water interface, and keep them up to above 3.0 mg L1 for a long time. Meanwhile CaO2 would increase the ORP level in sediment from 312.6 mV to at least 69.6 mV with the dosages of 0.12 and 0.18 kg m2, while the 0.06 kg m2 dose had a weaker effect on improving anoxic condition. Under anoxic condition, organic sulfur compounds and sulfates are converted to H2S, HS and S2 with the action of sulfate reducing bacteria, among which H2S is the main olfactory substance and has a strong toxicity (Niu et al., 2018). Meanwhile, it will not only produce odor compounds, but also emit a large number of greenhouse gases (such as CH4 and CO2) in the process of water blackening and smelling (Chen et al., 2018). In the experiment, CaO2 increased the relative abundances of Proteus Hauser, Planctomyces and Verrucomicrobium, reduced the corresponding values of Bacteroides, Caldisericum, Clostridium, Desulfobacterium, Parabacteroides, and Smithella, and derived some new bacteria of Nitrobacter, Nitrosomonas, Syntrophomonas and Thiobacillus. As common aerobic bacteria in sediment, Proteus Hauser, Planctomyces and Verrucomicrobium can effectively degrade organic matter for cell synthesis, thus achieving the purpose of water purification (Chen et al., 2017). As typical anaerobic bacteria in anaerobic fermentation, Bacteroides, Clostridium, Parabacteroides, and Smithella can degrade alkanes to produce acetic acid, butyric acid, CH4 and CO2 under anaerobic conditions; while Caldisericum and Desulfobacterium can use small molecular organic-acid as electron acceptors to accelerate the reduction process of sulfate to S2 (Krishnan et al., 2017). Among these newly derived functional bacteria, Nitrobacter and Nitrosomonas can convert ammonia nitrogen to nitrate nitrogen to further improve the hypoxia condition of water system (Mellbye et al., 2018); Syntrophomonas can degrade organic acid to CO2 and H2O, thereby avoiding the acidification of anoxic water (Kong et al., 2018); and Thiobacillus can oxidize sulfide to sulfate, thus reducing the production of H2S and odor
compounds (Li et al., 2017). Therefore, CaO2 could effectively control the emissions of harmful gases (such as H2S, CO2 and CH4) and avoid the acidification and smelling of landscape water by changing microbial community in sediment. The blackening of water system is a necessary but insufficient condition for its smelling, because FeS and MnS are the main black substances, while COD is the main indicator to reflect its degree (Liu et al., 2009). Under the anoxic and reductive conditions, the complex bonds in metal oxides break to release metal ions, which then are combined with S2 to form the blackening substance. Some studies indicated that the blackening of water was positively correlated with the hypoxia degree and the concentrations of COD and S2 (Novotortsev et al., 2012). Furthermore, the increase of sediment floatation and suspended matter concentration in water caused by harmful gases emissions could also obviously reduce the transparency and sensory effects of landscape water. The addition of CaO2 improved the anoxic condition of landscape water significantly and reduced the COD and S2 concentrations in the overlying water and interstitial water obviously, thus alleviating the water blackening effectively. Meanwhile, the strong oxidizing free radicals produced by the dissolution of CaO2 have the functions of bleaching and disinfection, which can effectively reduce the water chromaticity (Yu et al., 2016). In addition, as a by-product of CaO2, calcium hydroxide is an excellent coagulant and could accelerate the removal of suspended particulate matters by settlement (Wang et al., 2013). Therefore, CaO2 could effectively control the productions of blackening compounds and odor substances, obviously reduce the turbidity in landscape water, thereby realizing the purpose of restoring the sensory effects of anoxic landscape water. In summary, this study showed that CaO2 could effectively improve the environment condition and microbial population in anoxic/anaerobic landscape water system, obviously inhibit harmful gases emissions and sulfate reduction, and notably promote the COD degradation and turbidity decrease. Furthermore, the lower price and convenient production of CaO2 also provide a higher probability for the application of CaO2 in practice. So, it is a feasible approach to adopt CaO2 for inhibiting harmful gases emissions in anoxic/anaerobic landscape water system. 3.3. The optimal dose of CaO2 for inhibiting harmful gases emissions Although the CaO2 with different dosages could inhibit harmful gases emissions and restore landscape function in anoxic water system, the overall performances of different dosages were diverse. The experiment showed that the improvement effect of adding CaO2 with 0.06 kg m2 dose was much lower than those of 0.12 and 0.18 kg m2 dosages. However, there was no significant difference between 0.12 and 0.18 kg m2 dosages in DO, ORP, TCOD and sulfur concentrations (p > 0.05). Meanwhile, the gas compositions in those two different dosages were similar, and the H2S concentration was below the detection limit. In addition, the sediments observed at different depth of CK were dark black, while the surface sediments in the treatment groups with CaO2 additions were oxidized to yellowish brown, and the thickness of the oxide layer was positively correlated with the dose of CaO2. There were obvious phenomena, such as bubble-escape, sediment-uplift, and dark brown overlying water, observed in CK and 0.06 kg m2 dosage. However, the above phenomena did not happen in the reactors which adding CaO2 with the dosages of 0.12 and 0.18 kg m2 and the overlying water was relatively clear and transparent. The unit price of CaO2 in this study was $1.16 kg1, so the 0.12 kg m2 dosage was equivalent to $0.14 m-2, which was far below the cost of other common technologies (such as sediment dredging, adding aluminum salt and artificial aeration) (Wang et al.,
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
2018a). Although the dosages of 0.12 and 0.18 kg m2 had similar effects on improving anoxic landscape water, the former could save 1/3 of the cost than the latter. Furthermore, the pH value in the overlying water dosed with 0.18 kg-CaO2 m2-sediment would climb up to 9.32, which exceeded the safety threshold of surface water and would bring potential threat to aquatic ecosystem. Therefore, comprehensively considering the overall performance, economic cost and negative effect, it was advisable to use 0.12 kgCaO2 m2-sediment as the optimal dose to inhibit harmful gases emissions. 4. Conclusions This study explored and compared overall performances of different CaO2 dosages on inhibiting harmful gases emissions in anoxic/anaerobic landscape water system. The optimal dose was selected through a comprehensive analysis. The main conclusions were as follows: 1) CaO2 could effectively improve the anoxic condition and sediment microbial community in landscape water system, obviously increase SO2 concentration while decrease S2 4 concentration in the water, and the improvement effect was positively correlated with the CaO2 dose. 2) CaO2 with the dosages of 0.12 and 0.18 kg m2 could significantly reduce the productions of greenhouse gases and completely control H2S emission, thereby improving the water quality by inhibiting anaerobic fermentation, harmful gases emissions, sediment floating and nutrients release in the anoxic/ anaerobic landscape water system. 3) The 0.12 kg-CaO2 m2-sediment was selected as the optimal dose, which could reduce 75.10% CH4, 81.02% CO2 and 100% H2S in gases, and decrease 81.52% S2, 42.85% TCOD and 84.01% turbidity in the overlying water. In summary, CaO2 could obviously improve the environmental condition, water quality and sediment microorganisms, effectively inhibit harmful gases emissions and blackening substances productions, and ultimately solve the blackening and stinking problem in anoxic/anaerobic landscape water system. Conflicts of interest The manuscript is an original work and has not been submitted or is under consideration for publication in another journal. We also confirm that all the listed authors have participated actively in the study and approved the submitted manuscript. The authors do not have any possible conflicts of interest. Acknowledgments This work was funded by the National Natural Science Foundation of China (Project No. 21677115). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.envpol.2019.112989. References Andreasen, Rune R., Liu, D., Ravn, Sebastian, Feilberg, Anders, Poulsen, Tjalfe G., 2013. Airdwater mass transfer of sparingly soluble odorous compounds in granular biofilter media. Chem. Eng. J. 220, 431e440. Bagarinao, T., 1992. Sulfide as an environmental factor and toxicant: tolerance and adaptations in aquatic organisms. Aquat. Toxicol. 24, 21e62.
9
~¼Ller, R.A., 2017. Resilience of Boog, J., Nivala, J., Aubron, T., Mothes, S., Van, A.M., Ma carbon and nitrogen removal due to aeration interruption in aerated treatment wetlands. Sci. Total Environ. 621, 960e969. Chen, C., Liu, Q., Liu, C., Yu, J., 2017. Effect of different enrichment strategies on microbial community structure in petroleum-contaminated marine sediment in Dalian, China. Mar. Pollut. Bull. 117, 274e282. Chen, L., Zhu, B., Guo, Y., Xu, T., Lee, J.S., Qian, P.Y., Zhou, B., Chen, L., Zhu, B., Guo, Y., 2016. High-throughput transcriptome sequencing reveals the combined effects of key e-waste contaminants, decabromodiphenyl ether (BDE-209) and lead, in zebrafish larvae. Environ. Pollut. 214, 324e333. Chen, Q., Guo, B., Zhao, C., Xing, B., 2018. Characteristics of CH4 and CO2 emissions and influence of water and salinity in the Yellow River delta wetland, China. Environ. Pollut. 239, 289e299. Fang, H., Liu, Z., Jin, Z., Zhang, L., Liu, D., Pei, Y., 2016. An emphasis of hydrogen sulfide-cysteine cycle on enhancing the tolerance to chromium stress in Arabidopsis *. Environ. Pollut. 213, 870e877. Gao, J., Hou, H., Zhai, Y., Woodward, A., Vardoulakis, S., Kovats, S., Wilkinson, P., Li, L., Song, X., Xu, L., 2018. Greenhouse gas emissions reduction in different economic sectors: mitigation measures, health co-benefits, knowledge gaps, and policy implications *. Environ. Pollut. 240, 683e698. Gholami, F., Shavandi, M., Dastgheib, S.M.M., Amoozegar, M.A., 2018. Naphthalene remediation form groundwater by calcium peroxide (CaO2) nanoparticles in permeable reactive barrier (PRB). Chemosphere 212, 105e113. Glunk, C., Dupraz, C., Braissant, O., Gallagher, K.L., Verrecchia, E.P., Visscher, P.T., 2011. Microbially mediated carbonate precipitation in a hypersaline lake, Big Pond (Eleuthera, Bahamas). Sedimentology 58, 720e736. Kong, X., Yu, S., Fang, W., Liu, J., Li, H., 2018. Enhancing syntrophic associations among Clostridium butyricum, Syntrophomonas and two types of methanogen by zero valent iron in an anaerobic assay with a high organic loading. Bioresour. Technol. 257, 181e191. Krishnan, Y., Bong, C.P.C., Azman, N.F., Zakaria, Z., Othman, N.A., Abdullah, N., Ho, C.S., Lee, C.T., Hansen, S.B., Hara, H., 2017. Co-composting of palm empty fruit bunch and palm oil mill effluent: microbial diversity and potential mitigation of greenhouse gas emission. J. Clean. Prod. 146, 94e100. Lee, J., Rai, P.K., Jeon, Y.J., Kim, K.H., Kwon, E.E., 2017. The role of algae and cyanobacteria in the production and release of odorants in water. Environ. Pollut. 227, 252e262. Li, L., Wang, T., Sun, Y., Wang, P., Yvette, B., Meng, J., Li, Q., Zhou, Y., 2017. Identify biosorption effects of Thiobacillus towards perfluorooctanoic acid (PFOA): pilot study from field to laboratory. Chemosphere 171, 31e39. Liu, G.F., Zhong, J.C., He, J., Zhang, L., Fan, C.X., 2009. Effects of black spots of deadcyanobacterial mats on Fe-S-P cycling in sediments of Zhushan Bay, Lake Taihu. Huanjing Kexue 30, 2520e2526. Liu, X., Tao, Y., Zhou, K., Zhang, Q., Chen, G., Zhang, X., 2017. Effect of water quality improvement on the remediation of river sediment due to the addition of calcium nitrate. Sci. Total Environ. 575, 887e894. Lu, S., Zhang, X., Xue, Y., 2017. Application of calcium peroxide in water and soil treatment: a review. J. Hazard. Mater. 337, 163e177. Mellbye, B.L., Giguere, A.T., Murthy, G.S., Bottomley, P.J., Sayavedra-Soto, L.A., Fwr, C., 2018. Genome-Scale, constraint-based modeling of nitrogen oxide fluxes during coculture of Nitrosomonas europaea and nitrobacter winogradskyi. mSystems 2018, 3. https://doi.org/10.1128/mSystems.00170-17. Niu, Z.S., Pan, H., Guo, X.P., Lu, D.P., Feng, J.N., Chen, Y.R., Tou, F.Y., Liu, M., Yang, Y., 2018. Sulphate-reducing bacteria (SRB) in the Yangtze Estuary sediments: abundance, distribution and implications for the bioavailability of metals. Sci. Total Environ. 634, 296e304. Novotortsev, V.M., Mel'Nikov, I.O., Busygina, N.S., Rodionova, S.A., Artemov, A.V., 2012. Examination of the disinfecting properties of calcium peroxide and its suitability for improving hydrodynamic characteristics of surface water bodies. Russ. J. Appl. Chem. 85, 726e730. Nyk€ anen, A., Kontio, H., Klutas, O., Penttinen, O.P., Kostia, S., Mikola, J., Romantschuk, M., 2012. Increasing lake water and sediment oxygen levels using slow release peroxide. Sci. Total Environ. 429, 317e324. Qian, Y., Jing, Z., Zhang, Y., Chen, J., Zhou, X., 2016. Degradation of 2,4dichlorophenol by nanoscale calcium peroxide: implication for groundwater remediation. Separ. Purif. Technol. 166, 222e229. Shen, Q., Liu, C., Zhou, Q., Shang, J., Zhang, L., Fan, C., 2013. Effects of physical and chemical characteristics of surface sediments in the formation of shallow lake algae-induced black bloom. J. Environ. Sci. 25, 2353e2360. Shin, J., Lee, D., Hwang, T.-M., Lee, Y., 2018. Oxidation kinetics of algal-derived taste and odor compounds during water treatment with ferrate(VI). Chem. Eng. J. 334, 1065e1073. Wang, B.Y., Chen, Z.L., Zhu, J., Shen, J.M., Han, Y., 2013. Pilot-scale fluoridecontaining wastewater treatment by the ballasted flocculation process. Water Sci. Technol. J. Int. Assoc. Water Pollut. Res. 68, 134e143. Wang, G., Li, X., Fang, Y., Huang, R., 2014. Analysis on the formation condition of the algae-induced odorous black water agglomerate. Saudi J. Biol. Sci. 21, 597e604. Wang, G.B., Wang, Y., Zhang, Y., 2018a. Combination effect of sponge iron and calcium nitrate on severely eutrophic urban landscape water: an integrated study from laboratory to fields. Environ. Sci. Pollut. Res. Int. 25, 1e14. Wang, H., Zhao, Y., Li, T., Zhen, C., Wang, Y., Qin, C., 2016. Properties of calcium peroxide for release of hydrogen peroxide and oxygen: a kinetics study. Chem. Eng. J. 303, 450e457. Wang, H., Zhao, Y., Yan, S., Li, T., Meng, Y., Qin, C., 2017a. Fenton-like degradation of 2,4-dichlorophenol using calcium peroxide particles: performance and
10
W.-H. Wang et al. / Environmental Pollution 255 (2019) 112989
mechanisms. RSC Adv. 7, 4563e4571. Wang, W.H., Wang, Y., Fan, P., Chen, L.F., Chai, B.H., Zhao, J.C., Sun, L.Q., 2018b. Effect of calcium peroxide on the water quality and bacterium community of sediment in black-odor water. Environ. Pollut. 248, 18e27. Wang, W.H., Wang, Y., Li, Z., Wei, C.Z., Zhao, J.C., Sun, L.Q., 2017b. Effect of a strengthened ecological floating bed on the purification of urban landscape water supplied with reclaimed water. Sci. Total Environ. 622, 1630e1639. Wang, Y., Wang, W.H., Yan, F.L., Ding, Z., Feng, L.L., Zhao, J.C., 2019. Effects and mechanisms of calcium peroxide on purification of severely eutrophic water. Sci. Total Environ. 650, 2796e2806. Wert, E.C., Korak, J.A., Trenholm, R.A., Rosario-Ortiz, F.L., 2014. Effect of oxidant exposure on the release of intracellular microcystin, MIB, and geosmin from three cyanobacteria species. Water Res. 52, 251e259. Wu, H., Yu, Z., Bian, S., Ko, J.H., Li, S.F.Y., Xu, Q., 2018. H2S adsorption by municipal solid waste incineration (MSWI) fly ash with heavy metals immobilization. Chemosphere 195, 40e47.
Xiao, W., Liu, X., Irwin, A.J., Laws, E.A., Wang, L., Chen, B., Zeng, Y., Huang, B., 2017. Warming and eutrophication combine to restructure diatoms and dinoflagellates. Water Res. 128, 206e216. Xue, Y., Sui, Q., Brusseau, M.L., Zhang, X., Qiu, Z., Lyu, S., 2018. Insight on the generation of reactive oxygen species in the CaO2/Fe(II) Fenton system and the hydroxyl radical advancing strategy. Chem. Eng. J. 353, 657e665. Yu, D., Xie, P., Zeng, C., Xie, L., Chen, J., 2016. In situ enclosure experiments on the occurrence, development and decline of black bloom and the dynamics of its associated taste and odor compounds. Ecol. Eng. 87, 246e253. Zhangliang, H., Dezhi, S., Hui, W., Ruoyu, L., Zhiyuan, B., Fei, Q., 2018. Effects of ambient temperature and aeration frequency on emissions of ammonia and greenhouse gases from a sewage sludge aerobic composting plant. Bioresour. Technol. 270, 457e466. Ziegler, G., Tamburri, M.N., Fisher, D.J., 2018. Long-term algal toxicity of oxidant treated ballast water. Mar. Pollut. Bull. 133, 18e29.