- Email: [email protected]
Biological treatment of selenate-containing saline wastewater by activated sludge under oxygen-limiting conditions
Water Research 154 (2019) 327e335
Contents lists available at ScienceDirect
Water Research journal homepage: www.elsevier.com/locate/watres
Biological treatment of selenate-containing saline wastewater by activated sludge under oxygen-limiting conditions Yuanyuan Zhang a, Masashi Kuroda a, Shunsuke Arai b, Fumitaka Kato b, Daisuke Inoue a, Michihiko Ike a, * a b
Division of Sustainable Energy and Environmental Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka, 565-0871, Japan Nippon Steel & Sumitomo Metal Corporation, 20-1 Shintomi, Futtu, Chiba, 293-8511, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 November 2018 Received in revised form 28 January 2019 Accepted 29 January 2019 Available online 15 February 2019
Selenium often coincides with high salinity in certain industrial wastewaters, which can be a limitation in the practical application of biological treatment. However, there are no studies on the biological treatment of selenate-containing saline wastewater. A sequencing batch reactor inoculated with activated sludge was applied to treat selenate in the presence of 3% (w/v) NaCl. Start-up of the sequencing batch reactor with a 7-day cycle duration and excessive acetate as the sole carbon source succeeded in removing above 98% and 72% soluble and solid selenium, respectively, under oxygen-limiting conditions. Further selenium removal experiments with a shorter cycle duration of 3 days and a stepwise decrease of acetate addition achieved soluble and total selenium removal efficiencies in most batches above 96% and 80%, respectively. Mass balance analysis revealed that selenate was converted into elemental selenium, most of which was accumulated in the sludge. Microscopic analyses also found that elemental selenium particles were primarily present as approximately 2 mm large rods, with some extremely large particles above 10 mm. Although the bacterial populations responsible for selenium removal, especially selenate reduction, could not be identified by microbial community analysis, this study reported for the first time that selenate could be biologically treated in the presence of considerable salinity, offering implications for the practical treatment of selenium in certain industrial wastewaters. © 2019 Elsevier Ltd. All rights reserved.
Keywords: Activated sludge Selenate bioreduction Saline wastewater Sequencing batch reactor Oxygen-limiting conditions
1. Introduction Although selenium (Se) is utilized in multiple applications as a valuable metalloid element, it is also an acute and reproductive toxicant for living organisms (Luoma and Presser, 2009). Secontaining wastewaters originate from various anthropogenic processes, including agricultural irrigation and industrial activities like mining, oil refining, coal-fired power generation, and photoelectric production (Lawson and Macy, 1995; Muscatello and Janz, 2009; Soda et al., 2011; Yan et al., 2001). Wastewaters should be well treated before discharge into natural environments to avoid bioaccumulation and toxicity of Se. Se mainly exists as oxyanions including selenate (SeO2 4 ) and selenite (SeO2 3 ), both of which are highly soluble in wastewater. Although physicochemical technologies such as ferric salt coagulation have been conventionally used for Se removal, they are costly
* Corresponding author. E-mail address: [email protected] (M. Ike). https://doi.org/10.1016/j.watres.2019.01.059 0043-1354/© 2019 Elsevier Ltd. All rights reserved.
and ineffective for selenate removal. Recently, biological treatment has been recognized as the preferred alternative for Se removal due to its low-cost and eco-friendly characteristics. In biological treatment, microorganisms reduce selenate to selenite, and further to insoluble elemental Se particles, which can be separated from the aqueous phase through post-treatment methods (Nancharaiah and Lens, 2015a). Also, selenate and selenite can be microbially volatilized into a gaseous phase as methyl selenide (Chasteen and Bentley, 2003; Kagami et al., 2013). Se can be recovered with a high purity by trapping the gaseous phase through the biovolatilization, though toxicity of methyl selenide is problematic. In such microbial Se transformations, selenate reduction proceeds mainly through dissimilatory metabolism under anoxic conditions, in which selenate is used as the terminal electron acceptor for anaerobic respiration (Fujita et al., 1997; Kuroda et al., 2011b; Macy and Lawson, 1993). Also, assimilatory metabolism by both aerobes and anaerobes partially contributes to selenate reduction (Van Hullebusch, 2017). The microbial activity of selenite reduction is considered to exist more ubiquitously in natural environments
328
Y. Zhang et al. / Water Research 154 (2019) 327e335
2. Materials and methods Abbreviations 2.1. Activated sludge DO Dissolved oxgen DOC Dissolved organic carbon ES Excess sludge MLSS Mixed liquor suspended solids OTUs Operational taxonomic units PCR Polymerase chain reaction SBR Sequencing batch reactor Se Selenium SeO2Selenite 3 SeO2Selenate 4 STEM-EDX Scanning transmission electron microscopyenergy dispersive X-ray spectroscopy TEM Transmission electron microscopy
than selenate reduction (Ike et al., 2000), and various bacteria capable of reducing selenite under aerobic or anaerobic conditions have been reported (Gonzalez-Gil et al., 2016; Mishra et al., 2011; Nguyen et al., 2016). Kuroda et al. (2011a) pointed out that anaerobic selenite-reducers usually show a weaker capability of selenite reduction than aerobic selenite-reducers. However, the mechanisms of microbial selenite reduction have not been fully elucidated. It should be noted that Se occurrence often coincides with high salinity in certain industrial wastewaters, which can be a limitation in practical application of biological treatment. Typical concentrations of Se and salinity in Se refinery wastewater were reported to range from 13.2 to 74.0 mg/L and 6e7%, respectively (Soda et al., 2011). Kiln powder leachate from a cement-manufacturing plant contained 2e42 mg/L Se and 4.4e13.2% salinity (Soda et al., 2015). High salinity in these wastewaters may pose an adverse impact on bacterial activity through osmotic stress and reaction pathway inhibition in substrate degradation processes (Pollice et al., 2000), resulting in a significant reduction in treatment performance. Actually, inhibition of Se reduction caused by salinity was observed for Pseudomonas stutzeri NT-I, of which specific reduction rates for selenate and selenite significantly dropped at 20 g-NaCl/L compared to those at 0.05 g-NaCl/L (Kuroda et al., 2011a). Therefore, developing biological Se treatment technologies for saline wastewater in order to expand application is highly sought. In our previous research, removal of selenite from artificial wastewater with high salinity (70 g-NaCl/L) was achieved through biovolatilization using aerobic treatment with activated sludge gradually acclimated to high salinity prior to treatment (Zhang et al., 2019). On the other hand, successful selenate removal from saline wastewater by biological treatment has not been reported, though Mal et al. (2017) recently reported that selenate could be successfully removed from synthetic wastewater without significant salinity by activated sludge. It is worth noting that the removal of selenate is more challenging than selenite because a two-step reduction, namely selenate and selenite reductions (which cannot always proceed simultaneously), are necessary to obtain readilyremovable elemental Se. The objective of this work was to establish biological treatment technology for selenate removal from saline wastewater (3% NaCl, w/v). A sequencing batch reactor (SBR) was fed with synthetic saline wastewater containing high concentrations of selenate (1e5 mM), and treatment experiments were carried out under oxygen-limiting conditions. SBR performance on Se removal was investigated, focusing on cycle duration (i.e. hydraulic retention time), and amount of carbon source addition (i.e. organic loading).
Activated sludge obtained from a coke-oven wastewater treatment process in a steel manufacturing plant, which contains sulfur compounds such as sulfate, thiosulfate, and thiocyanate, was used as the seed of model SBR because bacterial Se metabolism is known to be associated with sulfur metabolism (Zannoni et al., 2007). Since the wastewater was treated with marine water after dilution, this activated sludge was adapted to high salinity. Activated sludge was washed three times with 50 mM potassium phosphate buffer (pH 7.5) and re-suspended into the buffer before use. 2.2. Synthetic wastewater Thirty grams per liter NaCl (3%, w/v) was added to synthetic wastewater containing 1 or 5 mM Na2SeO4, 8.25e66 mM sodium acetate, 3.5e28 mM NH4Cl, 0.2e1.7 mM K2HPO4, 0.09e0.7 mM KH2PO4, 0.2 mM MgSO4$7H2O, and 2 mM CaCl2$2H2O to create high salinity conditions. Trace element stock solution prepared as previously reported (Nancharaiah et al., 2008) was supplemented at 0.1 mL/L-medium. During different operational phases, sodium acetate concentration in synthetic wastewater was changed to maintain dissolved organic carbon (DOC) concentrations of 200e1600 mg/L in the reactor, and nitrogen and phosphate concentrations were adjusted on the same scale. The pH of synthetic wastewater was adjusted to 7.0e7.1 by addition of 2 M NaOH. 2.3. SBR operation Forty milliliters of seed sludge with a mixed liquor suspended solids (MLSS) concentration of 12,000 mg/L in potassium phosphate buffer and 160 mL synthetic wastewater were put into a 200 mL Erlenmeyer flask as a model experimental SBR system. The SBR was sealed with a rubber cap and was enclosed with parafilm to establish oxygen-limiting conditions (dissolved oxygen (DO) after reaction < 0.2 mg/L) though head-space air was not removed, and operated with a 7 day start-up cycle duration and 3 days for further treatment experiments with decreasing substrate donation. An SBR cycle consisted of: (1) rotary shaking at 120 rpm and 28 C, (2) withdrawal of 1e12 mL excess sludge (ES), (3) sludge settling for 0.5e1.5 h, (4) effluent decantation to give an 80% volumetric exchange ratio, and (5) wastewater refill. Withdrawal of 1e12 mL ES was for Se measurement (1 mL), bacterial community analysis (1 mL), and MLSS measurement (10 mL). SBR operation was divided into 7 phases with variation in cycle duration (start-up or treatment), selenite concentrations, and DOC concentrations as summarized in Table 1. 2.4. Analytical procedures Soluble Se and DOC levels of influent and effluent were measured after centrifugation of the samples (15,000g, 10 min, 4 C) and filteration of the supernatants through a 0.45 mm poresize polycarbonate membrane filter (Toyo Roshi Kaisha Ltd., Tokyo, Japan). For determination of solid Se in the effluent, effluent precipitate after centrifugation was washed with 5 mg/L Na5P3O10 buffer, and the pellet digested in a mixture of 69% HNO3 and 95% H2SO4 (20:1) at 100 C for 10 min. Similar pretreatment was performed to determine total solid Se in ES. Total soluble and solid Se concentrations were measured using inductively coupled plasmaatomic emission spectroscopy (SPS7800; SII NanoTechnology Inc., Chiba, Japan). Selenite and selenate concentrations were separately determined by ion chromatography (HIC-SP system; Shimadzu,
Y. Zhang et al. / Water Research 154 (2019) 327e335
329
Table 1 Operational phases of sequencing batch reactor.
Start-up Se treatment with decreasing substrate donation
Phase (batch)
Cycle duration (d)
Inf. selenate (mM)
Inf. DOC (mg/L)
I (1e3) II (4e6) III (7e35) IV (36e41) V (42e47) VI (48e53) VII (54e)
7 7 3 3 3 3 3
5 1 1 1 1 1 1
1600 1600 1600 1200 800 400 200
Kyoto, Japan) with an IonPac AS4A-SC column (Dionex, CA, USA) and using 3 mM Na2CO3 as the mobile phase. DOC was measured using a total organic carbon analyzer (TOC-VCSH; Shimadzu, Kyoto, Japan). DO was measured using a DO analyzer (HQ30D; Hach, Loveland, USA). The method for MLSS concentration measurement was described previously (Takada et al., 2018). 2.5. Microscopic analysis Transmission electron microscopy (TEM) and scanning transmission electron microscopy-energy dispersive X-ray spectroscopy (STEM-EDX) were performed to determine the location and morphology of selenium in the sludge. For TEM analysis, the ES sample was diluted and gently washed with 3% NaCl solution three times, then dropped on a carbon-coated 150 mesh copper grid. After 2 min, excess liquid was absorbed by holding a piece of moist filter paper against the edge of the grid. The grid was then examined in an H-800 TEM (Hitachi, Tokyo, Japan) at 200 kV of electron accelerating voltage. STEM-EDX analysis was performed on selected particles at 20 kV of electron accelerating voltage using EMAX-7000 (Horiba, Kyoto, Japan) associated with TEM.
selenate-reducers. Excessive sodium acetate levels of 1600 mg/L DOC were supplemented to form anoxic conditions by SBR oxygen consumption. Influent concentration of selenate and effluent concentrations of total and soluble Se during start-up phases (phases I and II) are shown in Fig. 1. During phase I batch 1, negligible amounts of soluble Se were removed in 7 days, whereas soluble Se removal at 56e65% was observed during batches 2 and 3 (Fig. 1a). Along with soluble Se removal, sludge color varied to a red hue specific to elemental Se (Fig. 1b), which confirmed the reduction of selenate to elemental Se. By contrast, solid Se in the effluent increased during phase I, which resulted in a slight decline in total Se removal from batch 2 to batch 3. Therefore, during phase II batch 4, influent selenate concentration decreased from 5 mM to 1 mM, which drastically improved the removal of soluble and solid Se removal to above 98% and 72%, respectively (Fig. 1a). Monitoring temporal changes of soluble Se during batch 6 revealed a maximum soluble Se removal of 97% in 72 h (data not shown). In addition, during phase II sludge color changed from red to black (Fig. 1b), which might indicate the formation of immobilized elemental Se as reported previously (Nancharaiah et al., 2018). By contrast, during the start-up period the removal of acetate-derived DOC was always less
2.6. Bacterial community analysis Fast DNA Spin Kit for Soil (MP Biomedicals, Illkirch, France) was used for extraction of genomic DNA in ES samples according to manufacturer's protocol. Extracted DNA was subjected to 16S rRNA amplicon sequencing at Bioengineering Laboratory Co. Ltd. (Kanagawa, Japan). A 2-step tailed polymerase chain reaction (PCR) method was performed to construct amplicon libraries targeting the V4 region of bacterial 16S rRNA genes with the 515F and 806R primer set (Peiffer et al., 2013). At the second PCR, the resulting amplicon was barcoded using the Index PCR primers. Sequence reads obtained were quality filtered, processed and analyzed using QIIME ver. 1.9.1 (Caporaso et al., 2010) according to companyrecommended setting conditions. High-quality reads were clustered into operational taxonomic units (OTUs) with a 97% similarity threshold, and taxonomic assignments were conducted using the Greengenes database (http://greengenes.secondgenome.com). Details in data processing and methods for calculation of Shannon diversity index and Chao1 richness estimator were described previously (Takada et al., 2018). 3. Results and discussion 3.1. SBR start-up During phases I and II, the SBR was started with a 7 day cycle duration, with the aim of acclimatizing seed sludge to artificial wastewater and increasing levels of microorganisms associated with Se removal. As selenate-reducers are reported to grow by anaerobic selenate respiration (Nancharaiah and Lens, 2015b), 5 mM selenate was added in phase I to promote the growth of
Fig. 1. Selenium concentrations (a) and sludge color change (b) during start-up phases I and II. I and II over Figure (a) indicate phase No. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
330
Y. Zhang et al. / Water Research 154 (2019) 327e335
than 16% (data not shown). Achieving successful selenate removal during start-up within a short period indicated that halotolerant selenate-reducers existed in the seed sludge and could be enriched relatively readily. The presence of halotolerant selenite-reducers has been reported previously (Mishra et al., 2011), and our earlier study clarified that they can be enriched from activated sludge in a municipal wastewater treatment plant under aerobic conditions without difficulty (Zhang et al., 2019). By contrast, this study clearly indicated that enrichment of halotolerant selenate-reducers is possible under oxygenlimiting conditions. During the SBR start-up period (especially in batch 3), solid Se remained at considerable concentrations in the effluent despite efficient removal of soluble Se. However, this problem was solved by the decline of influent Se concentration, and consequently the start-up of an SBR capable of removing selenate at a stable high ratio without flowing solid Se was achieved. 3.2. Se treatment experiment Since both soluble and solid Se could be removed almost completely within 3 days in batch 6 (phase II), the SBR was operated with a shorter cycle duration of 3 days from phase III to phase VII. During phases III to VII, the influent substrate (acetate) concentration was decreased gradually after confirming stable Se removal. Since high substrate supplementation leads to increased cost not only for the substrate itself but also the DOC removal posttreatment, it is important to identify appropriate substrate donation levels. The performance of Se and DOC removal during phases III to VII are shown in Fig. 2. During phase III, substrate was added at a DOC concentration of 1600 mg/L and soluble Se was removed efficiently at > 90% (excepting temporal worsening in batch 31), which confirmed stable removal of soluble Se. By contrast, solid Se was detected in the effluent at 0.05e0.40 mM, and total Se removal efficiency was 76 ± 12%. Although DOC removal efficiency increased to 20e40% during the former part of phase III, it declined to around 10% after batch 30. The decline in DOC removal efficiency might be associated with the decrease of soluble Se removal observed in batch 31, and could have been caused by incomplete sealing of the flask. During phases IV to VII where the influent concentration of DOC was decreased to 1200 (phase IV), 800 (phase V), 400 (phase VI), and 200 mg/L (phase VII) at intervals of 6 batches, soluble Se removal performance was maintained above 96%. Solid Se concentration in the effluent was also maintained at 0.05e0.26 mM, and above 80% total Se removal was achieved in most batches. The abnormally high solid Se in the effluent in batch 60 might be caused by unsuitable placement of the SBR flask in the rotary shaker. DOC consumption during phases IV to VII was generally 100e200 mg/L per batch (Fig. S1), and DOC removal efficiency varied largely with average removal efficiencies of 6, 21, 55, and 58% in phases IV, V, VI, and VII, respectively. Consequently, high concentrations of DOC remained in the effluent. The results during phases III to VII verified that soluble Se can be removed from high-salinity selenate wastewater stably with a high efficiency by SBR with acclimatized activated sludge. It was also found that residual solid Se in the effluent can be avoided by appropriately adjusting substrate concentrations. Further residual solid Se can be easily removed by appropriate post-treatment, like chemical coagulation and filtration (Staicu et al., 2015). Thus, the results demonstrated biological removal of selenate in high-salinity wastewater for the first time. Organic substrate is an important factor that greatly affects the biological Se removal process. Acetate used as the substrate in this study has been practically applied as a denitrification substrate (Dong et al., 2017; Gong et al., 2013), and thus would be
economically acceptable in wastewater treatment. However, even in phase VII with the lowest acetate supplementation (200 mg/L as DOC), during which nearly 8 mM carbon on average was consumed along with nearly complete removal of 1 mM selenate in the influent, acetate remained at considerable concentration in the effluent. Residual DOC should be further removed by posttreatment, like subsequential aerobic biological treatment by activated sludge (Mal et al., 2017). Also, exploration of alternative substrates could prove useful. Time courses of soluble Se (selenite and selenate) in a batch cycle were monitored in batches 34 and 71 (Fig. 3). During batch 34, selenate was gradually depleted in 70 h with little (<0.1 mM) selenite accumulation (Fig. 3a). In contrast, selenate was reduced completely within 28 h during batch 71, with a partial accumulation of selenite which was further removed to an undetectable level within 48 h (Fig. 3b). More efficient Se removal in batch 71 (phase VII) than in batch 34 (phase III) might be due to further enrichment of microorganisms associated with Se removal, especially selenatereducers, after the start-up period, or by the negative influence of excessive organic substrate on selenate reduction or total Se removal. Although the reason for the difference in Se removal levels between batches could not be identified, the slight succession of bacterial communities as described below may support the former possibility. Soluble Se time course monitoring verified that selenate and selenite reductions proceed synchronously. Biological reduction of selenate to selenite has been known to occur by anaerobic respiration under anoxic conditions. Thus, it could be speculated that selenite removal also occurs under anoxic conditions, which differs from our earlier study performing aerobic selenite removal in activated sludge (Zhang et al., 2019). 3.3. Se mass balance To clarify the fate of selenate added to the SBR and removed from the water phase, the Se mass balance was investigated by monitoring the amount of Se accumulated in the sludge in addition to soluble and solid Se in the influent, effluent, and ES discharged from the SBR. Se accumulation in the sludge was observed from batch 2, when detectable soluble Se removal occurred, which gradually increased to around 2800 mg/L (Fig. 2e). Figure 4 gives the cumulative Se distribution during phases III and VII. Soluble Se that existed as selenate in the influent at the beginning of a batch cycle was transformed mostly to solid Se, namely elemental Se. For example, during phase VII, 82% of elemental Se generated was accumulated in the sludge, including 54% remaining in the SBR, 28% removed as ES, and 12% present in the effluent. These results indicated that the removal of soluble Se (selenate) in the SBR occurred mainly by bioprecipitation, namely in a series of reductive reactions from selenate to elemental selenium via selenite. Although the results of mass balance analysis during phase III also inferred the possibility of biovolatilization, its contribution to Se removal from the water phase was minor, if any. Biovolatilization of selenium has been reported to occur when protein substrates are supplemented (Thompson-Eagle and Frankenberger, 1990; Zhang et al., 2019; Zhang and Frankenberger, 1999), and thus minimally occurred in this study where acetate was used as the substrate. In contrast, the amount of solid Se discharged as ES increased from 12% (phase III) to 28% (phase VII) along with an increase of solid Se accumulated in the SBR sludge. Nevertheless, solid Se in effluent decreased slightly from 20% (phase III) to 12% (phase VII). This might be a result of generated Se particle properties that settle easily and are resistant to suspension in the water phase. A previous study has reported that elemental Se was generated as nanoparticles and readily discharged from the reactor in fresh water,
Y. Zhang et al. / Water Research 154 (2019) 327e335
331
Fig. 2. Se removal efficiencies (a), Se concentrations (b), dissolved organic carbon (DOC) removal efficiencies (c), DOC concentrations (d), and solid Se and mixed liquor suspended solids (MLSS) (e). III to VII over the figure indicate phase No.
332
Y. Zhang et al. / Water Research 154 (2019) 327e335
Fig. 3. Time course of soluble Se concentration and dissolved organic carbon consumption monitored on batch 34 (phase III) and batch 71 (phase VII).
whereas biogenic Se was colloidally destabilized and was expected to settle faster in marine environments (Buchs et al., 2013). 3.4. Characterization of sludge and Se deposition
Fig. 4. Cumulative Se mass balance during phases III and VII. Batch 60 (phase VII) was excluded from calculation of Se mass balance.
The Se mass balance study revealed that most soluble Se removed was accumulated in the sludge. As shown in Fig. 2e, solid Se in the sludge accounted for approximately 22% (902 mg/L) of MLSS concentration at 4050 mg/L in batch 23, and 47% (2577 mg/L) of MLSS concentration at 5460 mg/L in batch 69, respectively. This indicated that biomass concentration was maintained stably at around 3000 mg/L, and that accumulation of solid Se in the sludge increased throughout the SBR operation period. The color change of the sludge to red during the start-up period indicated the formation of amorphous Se. This resulted in poor sludge settleability and effluent containing relatively high concentrations of solid Se. Thereafter sludge color change to black indicated the formation of crystalline Se, and excellent sludge settleability was maintained independently from the MLSS concentration, excepting the temporal worsening of settleability at batch 60 due to suspected equipment trouble. In general, activated sludge with MLSS as high
Y. Zhang et al. / Water Research 154 (2019) 327e335
333
Fig. 5. TEM image of Se particles (a) and corresponding EDX spectrum (b) in SBR sludge after phase VII.
as 5000 mg/L encountered difficulty in solid-liquid separation. Thus, it was suggested that enhancement of sludge settleability was attributable to the accumulation of solid Se in the sludge as similar phenomena have been observed in previous studies (Jain et al., 2015; Zhang et al., 2019). TEM analysis was conducted to characterize the sludge Se accumulation. TEM analysis of the sludge after phase VII where Se was removed efficiently found that the particles, most of which were confirmed to consist of Se by STEM-EDX, were present mainly as large rods approximately 2 mm long (Fig. 5). Rod Se particles were further confirmed by EDX mapping of the same area (Fig. S2). Extremely large rod particles of above 10 mm were also observed, which has never been reported before. By contrast, smaller nanospheres, which have been reported as the major morphology of elemental selenium accumulated in aerobic granular sludge (Nancharaiah et al., 2018) and activated sludge (Jain et al., 2015), were rarely observed inside or at the surface of bacterial cells (data not shown). The formation of such large crystalline Se particles in the sludge would strongly support good sludge settleability throughout the SBR operation period as reported previously (Jain et al., 2015). Although the reason for the formation of large crystalline Se particles is unclear, it might be a specific phenomenon for solid Se formation from high Se-containing wastewater under highsalinity conditions. 3.5. SBR bacterial communities 16S rRNA gene amplicon sequencing of the nine sludge samples obtained 41,449 to 59,912 sequence reads per a sample (483,575 sequence reads in total), which were further classified into 3,756 OTUs (Table 2). Based on the OTU number, Shannon diversity index and Chao1 richness estimator, the diversity of bacterial communities was highest in the seed sludge and phase I, declined from phase I to phase VI, and slightly increased from phase VI to phase VII. The results clearly indicated the enrichment of bacterial
populations associated with reduction and removal of Se under high salinity conditions. Bacterial community composition at the phylum level is summarized in Fig. 6. Overall phylum composition did not change greatly throughout the SBR operation period. Proteobacteria was found to be the most abundant phylum across all samples, accounting for 49e69% of total populations. The abundance of Bacteroidetes increased during phase I, but thereafter decreased along with the cultivation. At the OTU level, 35 OTUs were abundantly detected (>1%) in at least one of the nine samples analyzed. The most abundant representative OTUs are summarized in Table 3. Denovo9544 (16%), denovo22361 (8%), and denovo4423 (5%) were present in relatively high abundance in the seed sludge, and were assigned to Dermatophilaceae, Arcobacter, and Chromatiales, respectively; however, their abundance declined following batch 3 (phase I). Alternatively, denovo2041 and denovo6468, which were assigned to Marinobacterium and Marinobacter respectively, grew from previously undetectable levels in the seed sludge to highly dominant colonies of 42% and 11% in phase V (batch 47). Along with decreasing DOC input, the abundance of denovo2041 declined to 24% (phase VII), whereas denovo6468 maintained its increasing trend till the end of SBR operation. The dominance of these halophilic bacteria would be attributable to the saline condition of the synthetic wastewater in this study. Also, sulfur-reducers (denovo18020 and denovo3537) and a nitrate-reducer (denovo7316) were detected, accounting for 7% and 6% of the total population in batch 92 (phase VII). Throughout the SBR operation period in this study, bacterial populations that have been previously reported as capable of reducing selenate/selenite under saline conditions, such as Bacillus (Blum et al., 1998; Mishra et al., 2011) and Selenihalanaerobacter (Blum et al., 2001), were never detected. In addition, our attempt by conventional cultivation methods could not isolate halotolerant bacteria with a strong selenate reducing ability that could explain the selenate removal performance observed in the SBR (data not
Table 2 Summary of 16S rRNA gene amplicon sequencing of sludge samples. Sample (phase)
Qualified sequence reads
OTUs
Shannon diversity
Chao1 richness
Seed Batch Batch Batch Batch Batch Batch Batch Batch
41,449 59,912 56,784 56,932 51,653 57,633 48,002 53,827 57,383
1,456 1,717 1,099 852 763 711 644 692 696
4.52 4.21 3.17 2.77 2.69 2.49 2.51 2.76 2.69
3,794 4,674 2,925 2,027 1,737 1,656 1,346 1,718 1,837
3 (I) 6 (II) 35 (III) 41 (IV) 47 (V) 53 (VI) 69 (VII) 92 (VII)
334
Y. Zhang et al. / Water Research 154 (2019) 327e335
Fig. 6. Phylum level bacterial community structure. Relative abundance is expressed as the ratio of targeted sequence reads to total high-quality reads. Rare phyla with less than 1% abundance were grouped as “Unassigned; Others”.
Table 3 Relative abundance of representative predominant operational taxonomic units in sludge samples. OTU ID
denovo2041 denovo3537 denovo4423 denovo6468 denovo7316 denovo9544 denovo18020 denovo22361
Taxonomic classificationa
Marinobacterium Desulfuromonadales Chromatiales Marinobacter Nitratireductor Dermatophilaceae Desulfuromonas Arcobacter
Batch (phase) seed
3 (I)
6 (II)
35 (III)
41 (IV)
47 (V)
53 (VI)
69 (VII)
92 (VII)
0% 0% 5% 0% 0% 16% 1% 8%
14% 0% 1% 1% 0% 9% 2% 0%
38% 5% 0% 3% 0% 3% 1% 0%
34% 14% 0% 4% 1% 4% 0% 0%
36% 14% 0% 7% 1% 3% 0% 0%
42% 9% 0% 11% 1% 3% 0% 0%
38% 9% 0% 16% 1% 3% 0% 0%
23% 6% 0% 25% 8% 3% 0% 0%
24% 7% 0% 27% 6% 2% 0% 0%
a Taxonomic classification of the operational taxonomic units indicates the genus. If operational taxonomic units could not be assigned at the genus level, the lowest taxonomy (family/order/class) was shown.
shown). Thus, the bacterial populations responsible for Se removal in the SBR were not identified in this study. Some of the predominant bacteria detected in the 16S rRNA gene amplicon sequencing, such as Marinobacterium and Marinobacter, might be associated with selenate reduction, but are difficult to isolate by conventional cultivation methods. Another possibility would be cooperative selenate reduction by weak selenate-reducers. Identification of responsible bacteria or clarification of the cooperative relationship for selenate reduction is an important factor for understanding Se removal mechanisms and developing more efficient treatments of selenate-containing saline wastewater under anoxic conditions.
Acknowledgements The TEM analysis conducted in this study was supported by “Advanced Characterization Nanotechnology Platform, Nanotechnology Platform Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan” at the Research Center for Ultra-High Voltage Electron Microscopy (Nanotechnology Open Facilities) in Osaka University. This work was partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI [grant numbers 15K16145]. Appendix A. Supplementary data
4. Conclusions Biological reduction of selenate in the presence of considerable salinity (3% w/v NaCl) was demonstrated for the first time by SBR inoculated with activated sludge under oxygen-limiting conditions. Soluble Se removal above 96% and total Se removal above 80% were achieved in most batches. Biologically reduced Se was accumulated mainly as large rods in the sludge, which contributed to good sludge settleability. In contrast, the considerable remaining DOC concentrations following treatment, even at the lowest acetate donation, suggested the requirement of residual DOC post-treatment. This study offers a possibility for biological reduction of Se from Se-containing saline wastewater.
Declaration of interests None.
Supplementary data to this article can be found online at https://doi.org/10.1016/j.watres.2019.01.059. References Blum, J.S., Bindi, A.B., Buzzelli, J., Stolz, J.F., Oremland, R.S., 1998. Bacillus arsenicoselenatis, sp. nov., and Bacillus selenitireducens, sp. nov.: two haloalkaliphiles from Mono Lake, California that respire oxyanions of selenium and arsenic. Arch. Microbiol. 171, 19e30. https://doi.org/10.1007/s002030050673. Blum, J.S., Stolz, J.F., Oren, A., Oremland, R.S., 2001. Selenihalanaerobacter shriftii gen. nov., sp. nov., a halophilic anaerobe from Dead Sea sediments that respires selenate. Arch. Microbiol. 175, 208e219. https://doi.org/10.1007/ s002030100257. Buchs, B., Evangelou, M.W.H., Winkel, L.H.E., Lenz, M., 2013. Colloidal properties of nanoparticular biogenic selenium govern environmental fate and bioremediation effectiveness. Environ. Sci. Technol. 47 (5), 2401e2407. https://doi.org/10. 1021/es304940s. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., ~ a, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Fierer, N., Pen Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of highthroughput community sequencing data. Nat. Methods 7 (5), 335e336. https://doi.org/10.1038/nmeth.f.303. Chasteen, T.G., Bentley, R., 2003. Biomethylation of selenium and tellurium:
Y. Zhang et al. / Water Research 154 (2019) 327e335 microorganisms and plants. Chem. Rev. 103 (1), 1e26. https://doi.org/10.1021/ cr010210þ. Dong, H., Wang, W., Song, Z., Dong, H., Wang, J., Sun, S., Zhang, Z., Ke, M., Zhang, Z., Wu, W.-M., Zhang, G., Ma, J., 2017. A high-efficiency denitrification bioreactor for the treatment of acrylonitrile wastewater using waterborne polyurethane immobilized activated sludge. Bioresour. Technol. 239, 472e481. https://doi. org/10.1016/j.biortech.2017.05.015. Fujita, M., Ike, M., Nishimoto, S., Takahashi, K., Kashiwa, M., 1997. Isolation and characterization of a novel selenate-reducing bacterium, Bacillus sp. SF-1. J. Ferment. Bioeng. 83 (6), 517e522. https://doi.org/10.1016/S0922-338X(97) 81130-0. Gong, L., Huo, M., Yang, Q., Li, J., Ma, B., Zhu, R., Wang, S., Peng, Y., 2013. Performance of heterotrophic partial denitrification under feast-famine condition of electron donor: A case study using acetate as external carbon source. Bioresour. Technol. 133, 263e269. https://doi.org/10.1016/j.biortech.2012.12.108. Gonzalez-Gil, G., Lens, P.N.L., Saikaly, P.E., 2016. Selenite reduction by anaerobic microbial aggregates: microbial community structure, and proteins associated to the produced selenium spheres. Front. Microbiol. 7, 1e14. https://doi.org/10. 3389/fmicb.2016.00571. Ike, M., Takahashi, K., Fujita, T., Kashiwa, M., Fujita, M., 2000. Selenate reduction by bacteria isolated from aquatic environment free from selenium contamination. Water Res. 34 (11), 3019e3025. https://doi.org/10.1016/S0043-1354(00)000415. Jain, R., Seder-Colomina, M., Jordan, N., Dessi, P., Cosmidis, J., van Hullebusch, E.D., Weiss, S., Farges, F., Lens, P.N.L., 2015. Entrapped elemental selenium nanoparticles affect physicochemical properties of selenium fed activated sludge. J. Hazard Mater. 295, 193e200. https://doi.org/10.1016/j.jhazmat.2015.03.043. Kagami, T., Narita, T., Kuroda, M., Notaguchi, E., Yamashita, M., Sei, K., Soda, S., Ike, M., 2013. Effective selenium volatilization under aerobic conditions and recovery from the aqueous phase by Pseudomonas stutzeri NT-I. Water Res. 47 (3), 1361e1368. https://doi.org/10.1016/j.watres.2012.12.001. Kuroda, M., Notaguchi, E., Sato, A., Yoshioka, M., Hasegawa, A., Kagami, T., Narita, T., Yamashita, M., Sei, K., Soda, S., Ike, M., 2011a. Characterization of Pseudomonas stutzeri NT-I capable of removing soluble selenium from the aqueous phase under aerobic conditions. J. Biosci. Bioeng. 112 (3), 259e264. https://doi.org/10. 1016/j.jbiosc.2011.05.012. Kuroda, M., Yamashita, M., Miwa, E., Imao, K., Fujimoto, N., Ono, H., Nagano, K., Sei, K., Ike, M., 2011b. Molecular cloning and characterization of the srdBCA operon, encoding the respiratory selenate reductase complex, from the selenate-reducing bacterium Bacillus selenatarsenatis SF-1. J. Bacteriol. 193 (9), 2141e2148. https://doi.org/10.1128/JB.01197-10. Lawson, S., Macy, J.M., 1995. Bioremediation of selenite in oil refinery wastewater. Appl. Microbiol. Biotechnol. 43 (4), 762e765. https://doi.org/10.1007/ BF00164785. Luoma, S.N., Presser, T.S., 2009. Emerging opportunities in management of selenium contamination. Environ. Sci. Technol. 43 (22), 8483e8487. https://doi.org/10. 1021/es900828h. Macy, J.M., Lawson, S., 1993. Cell yield (YM) of Thauera selenatis grown anaerobically with acetate plus selenate or nitrate. Arch. Microbiol. 160 (4), 295e298. https:// doi.org/10.1007/BF00292080. Mal, J., Nancharaiah, Y.V., van Hullebusch, E.D., Lens, P.N.L., 2017. Biological removal of selenate and ammonium by activated sludge in a sequencing batch reactor. Bioresour. Technol. 229, 11e19. https://doi.org/10.1016/j.biortech.2016.12.112. Mishra, R.R., Prajapati, S., Das, J., Dangar, T.K., Das, N., Thatoi, H., 2011. Reduction of selenite to red elemental selenium by moderately halotolerant Bacillus megaterium strains isolated from Bhitarkanika mangrove soil and characterization of reduced product. Chemosphere 84 (9), 1231e1237. https://doi.org/10.1016/j. chemosphere.2011.05.025. Muscatello, J.R., Janz, D.M., 2009. Selenium accumulation in aquatic biota downstream of a uranium mining and milling operation. Sci. Total Environ. 407 (4),
335
1318e1325. https://doi.org/10.1016/j.scitotenv.2008.10.046. Nancharaiah, Y.V., Joshi, H.M., Hausner, M., Venugopalan, V.P., 2008. Bioaugmentation of aerobic microbial granules with Pseudomonas putida carrying TOL plasmid. Chemosphere 71 (1), 30e35. https://doi.org/10.1016/j. chemosphere.2007.10.062. Nancharaiah, Y.V., Lens, P.N.L., 2015a. Selenium biomineralization for biotechnological applications. Trends Biotechnol. 33 (6), 323e330. https://doi.org/10. 1016/j.tibtech.2015.03.004. Nancharaiah, Y.V., Lens, P.N.L., 2015b. Ecology and biotechnology of seleniumrespiring bacteria. Microbiol. Mol. Biol. Rev. 79 (1), 61e80. https://doi.org/10. 1128/MMBR.00037-14. Nancharaiah, Y.V., Sarvajith, M., Lens, P.N.L., 2018. Selenite reduction and ammoniacal nitrogen removal in an aerobic granular sludge sequencing batch reactor. Water Res. 131, 131e141. https://doi.org/10.1016/j.watres.2017.12.028. Nguyen, V.K., Park, Y., Yu, J., Lee, T., 2016. Microbial selenite reduction with organic carbon and electrode as sole electron donor by a bacterium isolated from domestic wastewater. Bioresour. Technol. 212, 182e189. https://doi.org/10.1016/j. biortech.2016.04.033. Peiffer, J.A., Spor, A., Koren, O., Jin, Z., Tringe, S.G., Dangl, J.L., Buckler, E.S., Ley, R.E., 2013. Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc. Natl. Acad. Sci. Unit. States Am. 110 (16), 6548e6553. https://doi.org/10.1073/pnas.1302837110. Pollice, A., Rozzi, A., Tomei, M.C., Di Pinto, A.C., Limoni, N., 2000. Monitoring the inhibitory effect of NaCl on anaerobic wastewater treatment processes by the rantox biosensor. Environ. Technol. 21 (5), 535e544. https://doi.org/10.1080/ 09593332408618095. Soda, S., Hasegawa, A., Kuroda, M., Hanada, A., Yamashita, M., Ike, M., 2015. Selenium recovery from kiln powder of cement manufacturing by chemical leaching and bioreduction. Water Sci. Technol. 72 (8), 1294e1300. https://doi.org/10. 2166/wst.2015.339. Soda, S., Kashiwa, M., Kagami, T., Kuroda, M., Yamashita, M., Ike, M., 2011. Laboratory-scale bioreactors for soluble selenium removal from selenium refinery wastewater using anaerobic sludge. Desalination 279 (1e3), 433e438. https:// doi.org/10.1016/j.desal.2011.06.031. Staicu, L.C., van Hullebusch, E.D., Oturan, M.A., Ackerson, C.J., Lens, P.N.L., 2015. Removal of colloidal biogenic selenium from wastewater. Chemosphere 125, 130e138. https://doi.org/10.1016/j.chemosphere.2014.12.018. Takada, K., Shiba, T., Yamaguchi, T., Akane, Y., Nakayama, Y., Soda, S., Inoue, D., Ike, M., 2018. Cake layer bacterial communities during different biofouling stages in full-scale membrane bioreactors. Bioresour. Technol. 259, 259e267. https://doi.org/10.1016/j.biortech.2018.03.051. Thompson-Eagle, E.T., Frankenberger Jr., W.T., 1990. Protein-mediated selenium biomethylation in evaporation pond water. Environ. Toxicol. Chem. 9 (12), 1453e1462. https://doi.org/10.1002/etc.5620091204. Van Hullebusch, E.D. (Ed.), 2017. Bioremediation of selenium contaminated wastewater. Springer, Cham. https://doi.org/10.1007/978-3-319-57831-6. Yan, R., Gauthier, D., Flamant, G., Peraudeau, G., Lu, J., Zheng, C., 2001. Fate of selenium in coal combustion: volatilization and speciation in the flue gas. Environ. Sci. Technol. 35 (7), 1406e1410. https://doi.org/10.1021/es0001005. Zannoni, D., Borsetti, F., Harrison, J.J., Turner, R.J., 2007. The bacterial response to the chalcogen metalloids Se and Te. Adv. Microb. Physiol. 53, 1e312. https://doi.org/ 10.1016/S0065-2911(07)53001-8. Zhang, Y., Kuroda, M., Nakatani, Y., Soda, S., Ike, M., 2019. Removal of selenite from artificial wastewater with high salinity by activated sludge in aerobic sequencing batch reactors. J. Biosci. Bioeng. (in press) https://doi.org/10.1016/j. jbiosc.2018.11.002. Zhang, Y., Frankenberger Jr., W.T., 1999. Effects of soil moisture, depth, and organic amendments on selenium volatilization. J. Environ. Qual. 28 (4), 1321e1326. https://doi.org/10.2134/jeq1999.00472425002800040037x.
Contents lists available at ScienceDirect
Water Research journal homepage: www.elsevier.com/locate/watres
Biological treatment of selenate-containing saline wastewater by activated sludge under oxygen-limiting conditions Yuanyuan Zhang a, Masashi Kuroda a, Shunsuke Arai b, Fumitaka Kato b, Daisuke Inoue a, Michihiko Ike a, * a b
Division of Sustainable Energy and Environmental Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka, 565-0871, Japan Nippon Steel & Sumitomo Metal Corporation, 20-1 Shintomi, Futtu, Chiba, 293-8511, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 November 2018 Received in revised form 28 January 2019 Accepted 29 January 2019 Available online 15 February 2019
Selenium often coincides with high salinity in certain industrial wastewaters, which can be a limitation in the practical application of biological treatment. However, there are no studies on the biological treatment of selenate-containing saline wastewater. A sequencing batch reactor inoculated with activated sludge was applied to treat selenate in the presence of 3% (w/v) NaCl. Start-up of the sequencing batch reactor with a 7-day cycle duration and excessive acetate as the sole carbon source succeeded in removing above 98% and 72% soluble and solid selenium, respectively, under oxygen-limiting conditions. Further selenium removal experiments with a shorter cycle duration of 3 days and a stepwise decrease of acetate addition achieved soluble and total selenium removal efficiencies in most batches above 96% and 80%, respectively. Mass balance analysis revealed that selenate was converted into elemental selenium, most of which was accumulated in the sludge. Microscopic analyses also found that elemental selenium particles were primarily present as approximately 2 mm large rods, with some extremely large particles above 10 mm. Although the bacterial populations responsible for selenium removal, especially selenate reduction, could not be identified by microbial community analysis, this study reported for the first time that selenate could be biologically treated in the presence of considerable salinity, offering implications for the practical treatment of selenium in certain industrial wastewaters. © 2019 Elsevier Ltd. All rights reserved.
Keywords: Activated sludge Selenate bioreduction Saline wastewater Sequencing batch reactor Oxygen-limiting conditions
1. Introduction Although selenium (Se) is utilized in multiple applications as a valuable metalloid element, it is also an acute and reproductive toxicant for living organisms (Luoma and Presser, 2009). Secontaining wastewaters originate from various anthropogenic processes, including agricultural irrigation and industrial activities like mining, oil refining, coal-fired power generation, and photoelectric production (Lawson and Macy, 1995; Muscatello and Janz, 2009; Soda et al., 2011; Yan et al., 2001). Wastewaters should be well treated before discharge into natural environments to avoid bioaccumulation and toxicity of Se. Se mainly exists as oxyanions including selenate (SeO2 4 ) and selenite (SeO2 3 ), both of which are highly soluble in wastewater. Although physicochemical technologies such as ferric salt coagulation have been conventionally used for Se removal, they are costly
* Corresponding author. E-mail address: [email protected] (M. Ike). https://doi.org/10.1016/j.watres.2019.01.059 0043-1354/© 2019 Elsevier Ltd. All rights reserved.
and ineffective for selenate removal. Recently, biological treatment has been recognized as the preferred alternative for Se removal due to its low-cost and eco-friendly characteristics. In biological treatment, microorganisms reduce selenate to selenite, and further to insoluble elemental Se particles, which can be separated from the aqueous phase through post-treatment methods (Nancharaiah and Lens, 2015a). Also, selenate and selenite can be microbially volatilized into a gaseous phase as methyl selenide (Chasteen and Bentley, 2003; Kagami et al., 2013). Se can be recovered with a high purity by trapping the gaseous phase through the biovolatilization, though toxicity of methyl selenide is problematic. In such microbial Se transformations, selenate reduction proceeds mainly through dissimilatory metabolism under anoxic conditions, in which selenate is used as the terminal electron acceptor for anaerobic respiration (Fujita et al., 1997; Kuroda et al., 2011b; Macy and Lawson, 1993). Also, assimilatory metabolism by both aerobes and anaerobes partially contributes to selenate reduction (Van Hullebusch, 2017). The microbial activity of selenite reduction is considered to exist more ubiquitously in natural environments
328
Y. Zhang et al. / Water Research 154 (2019) 327e335
2. Materials and methods Abbreviations 2.1. Activated sludge DO Dissolved oxgen DOC Dissolved organic carbon ES Excess sludge MLSS Mixed liquor suspended solids OTUs Operational taxonomic units PCR Polymerase chain reaction SBR Sequencing batch reactor Se Selenium SeO2Selenite 3 SeO2Selenate 4 STEM-EDX Scanning transmission electron microscopyenergy dispersive X-ray spectroscopy TEM Transmission electron microscopy
than selenate reduction (Ike et al., 2000), and various bacteria capable of reducing selenite under aerobic or anaerobic conditions have been reported (Gonzalez-Gil et al., 2016; Mishra et al., 2011; Nguyen et al., 2016). Kuroda et al. (2011a) pointed out that anaerobic selenite-reducers usually show a weaker capability of selenite reduction than aerobic selenite-reducers. However, the mechanisms of microbial selenite reduction have not been fully elucidated. It should be noted that Se occurrence often coincides with high salinity in certain industrial wastewaters, which can be a limitation in practical application of biological treatment. Typical concentrations of Se and salinity in Se refinery wastewater were reported to range from 13.2 to 74.0 mg/L and 6e7%, respectively (Soda et al., 2011). Kiln powder leachate from a cement-manufacturing plant contained 2e42 mg/L Se and 4.4e13.2% salinity (Soda et al., 2015). High salinity in these wastewaters may pose an adverse impact on bacterial activity through osmotic stress and reaction pathway inhibition in substrate degradation processes (Pollice et al., 2000), resulting in a significant reduction in treatment performance. Actually, inhibition of Se reduction caused by salinity was observed for Pseudomonas stutzeri NT-I, of which specific reduction rates for selenate and selenite significantly dropped at 20 g-NaCl/L compared to those at 0.05 g-NaCl/L (Kuroda et al., 2011a). Therefore, developing biological Se treatment technologies for saline wastewater in order to expand application is highly sought. In our previous research, removal of selenite from artificial wastewater with high salinity (70 g-NaCl/L) was achieved through biovolatilization using aerobic treatment with activated sludge gradually acclimated to high salinity prior to treatment (Zhang et al., 2019). On the other hand, successful selenate removal from saline wastewater by biological treatment has not been reported, though Mal et al. (2017) recently reported that selenate could be successfully removed from synthetic wastewater without significant salinity by activated sludge. It is worth noting that the removal of selenate is more challenging than selenite because a two-step reduction, namely selenate and selenite reductions (which cannot always proceed simultaneously), are necessary to obtain readilyremovable elemental Se. The objective of this work was to establish biological treatment technology for selenate removal from saline wastewater (3% NaCl, w/v). A sequencing batch reactor (SBR) was fed with synthetic saline wastewater containing high concentrations of selenate (1e5 mM), and treatment experiments were carried out under oxygen-limiting conditions. SBR performance on Se removal was investigated, focusing on cycle duration (i.e. hydraulic retention time), and amount of carbon source addition (i.e. organic loading).
Activated sludge obtained from a coke-oven wastewater treatment process in a steel manufacturing plant, which contains sulfur compounds such as sulfate, thiosulfate, and thiocyanate, was used as the seed of model SBR because bacterial Se metabolism is known to be associated with sulfur metabolism (Zannoni et al., 2007). Since the wastewater was treated with marine water after dilution, this activated sludge was adapted to high salinity. Activated sludge was washed three times with 50 mM potassium phosphate buffer (pH 7.5) and re-suspended into the buffer before use. 2.2. Synthetic wastewater Thirty grams per liter NaCl (3%, w/v) was added to synthetic wastewater containing 1 or 5 mM Na2SeO4, 8.25e66 mM sodium acetate, 3.5e28 mM NH4Cl, 0.2e1.7 mM K2HPO4, 0.09e0.7 mM KH2PO4, 0.2 mM MgSO4$7H2O, and 2 mM CaCl2$2H2O to create high salinity conditions. Trace element stock solution prepared as previously reported (Nancharaiah et al., 2008) was supplemented at 0.1 mL/L-medium. During different operational phases, sodium acetate concentration in synthetic wastewater was changed to maintain dissolved organic carbon (DOC) concentrations of 200e1600 mg/L in the reactor, and nitrogen and phosphate concentrations were adjusted on the same scale. The pH of synthetic wastewater was adjusted to 7.0e7.1 by addition of 2 M NaOH. 2.3. SBR operation Forty milliliters of seed sludge with a mixed liquor suspended solids (MLSS) concentration of 12,000 mg/L in potassium phosphate buffer and 160 mL synthetic wastewater were put into a 200 mL Erlenmeyer flask as a model experimental SBR system. The SBR was sealed with a rubber cap and was enclosed with parafilm to establish oxygen-limiting conditions (dissolved oxygen (DO) after reaction < 0.2 mg/L) though head-space air was not removed, and operated with a 7 day start-up cycle duration and 3 days for further treatment experiments with decreasing substrate donation. An SBR cycle consisted of: (1) rotary shaking at 120 rpm and 28 C, (2) withdrawal of 1e12 mL excess sludge (ES), (3) sludge settling for 0.5e1.5 h, (4) effluent decantation to give an 80% volumetric exchange ratio, and (5) wastewater refill. Withdrawal of 1e12 mL ES was for Se measurement (1 mL), bacterial community analysis (1 mL), and MLSS measurement (10 mL). SBR operation was divided into 7 phases with variation in cycle duration (start-up or treatment), selenite concentrations, and DOC concentrations as summarized in Table 1. 2.4. Analytical procedures Soluble Se and DOC levels of influent and effluent were measured after centrifugation of the samples (15,000g, 10 min, 4 C) and filteration of the supernatants through a 0.45 mm poresize polycarbonate membrane filter (Toyo Roshi Kaisha Ltd., Tokyo, Japan). For determination of solid Se in the effluent, effluent precipitate after centrifugation was washed with 5 mg/L Na5P3O10 buffer, and the pellet digested in a mixture of 69% HNO3 and 95% H2SO4 (20:1) at 100 C for 10 min. Similar pretreatment was performed to determine total solid Se in ES. Total soluble and solid Se concentrations were measured using inductively coupled plasmaatomic emission spectroscopy (SPS7800; SII NanoTechnology Inc., Chiba, Japan). Selenite and selenate concentrations were separately determined by ion chromatography (HIC-SP system; Shimadzu,
Y. Zhang et al. / Water Research 154 (2019) 327e335
329
Table 1 Operational phases of sequencing batch reactor.
Start-up Se treatment with decreasing substrate donation
Phase (batch)
Cycle duration (d)
Inf. selenate (mM)
Inf. DOC (mg/L)
I (1e3) II (4e6) III (7e35) IV (36e41) V (42e47) VI (48e53) VII (54e)
7 7 3 3 3 3 3
5 1 1 1 1 1 1
1600 1600 1600 1200 800 400 200
Kyoto, Japan) with an IonPac AS4A-SC column (Dionex, CA, USA) and using 3 mM Na2CO3 as the mobile phase. DOC was measured using a total organic carbon analyzer (TOC-VCSH; Shimadzu, Kyoto, Japan). DO was measured using a DO analyzer (HQ30D; Hach, Loveland, USA). The method for MLSS concentration measurement was described previously (Takada et al., 2018). 2.5. Microscopic analysis Transmission electron microscopy (TEM) and scanning transmission electron microscopy-energy dispersive X-ray spectroscopy (STEM-EDX) were performed to determine the location and morphology of selenium in the sludge. For TEM analysis, the ES sample was diluted and gently washed with 3% NaCl solution three times, then dropped on a carbon-coated 150 mesh copper grid. After 2 min, excess liquid was absorbed by holding a piece of moist filter paper against the edge of the grid. The grid was then examined in an H-800 TEM (Hitachi, Tokyo, Japan) at 200 kV of electron accelerating voltage. STEM-EDX analysis was performed on selected particles at 20 kV of electron accelerating voltage using EMAX-7000 (Horiba, Kyoto, Japan) associated with TEM.
selenate-reducers. Excessive sodium acetate levels of 1600 mg/L DOC were supplemented to form anoxic conditions by SBR oxygen consumption. Influent concentration of selenate and effluent concentrations of total and soluble Se during start-up phases (phases I and II) are shown in Fig. 1. During phase I batch 1, negligible amounts of soluble Se were removed in 7 days, whereas soluble Se removal at 56e65% was observed during batches 2 and 3 (Fig. 1a). Along with soluble Se removal, sludge color varied to a red hue specific to elemental Se (Fig. 1b), which confirmed the reduction of selenate to elemental Se. By contrast, solid Se in the effluent increased during phase I, which resulted in a slight decline in total Se removal from batch 2 to batch 3. Therefore, during phase II batch 4, influent selenate concentration decreased from 5 mM to 1 mM, which drastically improved the removal of soluble and solid Se removal to above 98% and 72%, respectively (Fig. 1a). Monitoring temporal changes of soluble Se during batch 6 revealed a maximum soluble Se removal of 97% in 72 h (data not shown). In addition, during phase II sludge color changed from red to black (Fig. 1b), which might indicate the formation of immobilized elemental Se as reported previously (Nancharaiah et al., 2018). By contrast, during the start-up period the removal of acetate-derived DOC was always less
2.6. Bacterial community analysis Fast DNA Spin Kit for Soil (MP Biomedicals, Illkirch, France) was used for extraction of genomic DNA in ES samples according to manufacturer's protocol. Extracted DNA was subjected to 16S rRNA amplicon sequencing at Bioengineering Laboratory Co. Ltd. (Kanagawa, Japan). A 2-step tailed polymerase chain reaction (PCR) method was performed to construct amplicon libraries targeting the V4 region of bacterial 16S rRNA genes with the 515F and 806R primer set (Peiffer et al., 2013). At the second PCR, the resulting amplicon was barcoded using the Index PCR primers. Sequence reads obtained were quality filtered, processed and analyzed using QIIME ver. 1.9.1 (Caporaso et al., 2010) according to companyrecommended setting conditions. High-quality reads were clustered into operational taxonomic units (OTUs) with a 97% similarity threshold, and taxonomic assignments were conducted using the Greengenes database (http://greengenes.secondgenome.com). Details in data processing and methods for calculation of Shannon diversity index and Chao1 richness estimator were described previously (Takada et al., 2018). 3. Results and discussion 3.1. SBR start-up During phases I and II, the SBR was started with a 7 day cycle duration, with the aim of acclimatizing seed sludge to artificial wastewater and increasing levels of microorganisms associated with Se removal. As selenate-reducers are reported to grow by anaerobic selenate respiration (Nancharaiah and Lens, 2015b), 5 mM selenate was added in phase I to promote the growth of
Fig. 1. Selenium concentrations (a) and sludge color change (b) during start-up phases I and II. I and II over Figure (a) indicate phase No. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
330
Y. Zhang et al. / Water Research 154 (2019) 327e335
than 16% (data not shown). Achieving successful selenate removal during start-up within a short period indicated that halotolerant selenate-reducers existed in the seed sludge and could be enriched relatively readily. The presence of halotolerant selenite-reducers has been reported previously (Mishra et al., 2011), and our earlier study clarified that they can be enriched from activated sludge in a municipal wastewater treatment plant under aerobic conditions without difficulty (Zhang et al., 2019). By contrast, this study clearly indicated that enrichment of halotolerant selenate-reducers is possible under oxygenlimiting conditions. During the SBR start-up period (especially in batch 3), solid Se remained at considerable concentrations in the effluent despite efficient removal of soluble Se. However, this problem was solved by the decline of influent Se concentration, and consequently the start-up of an SBR capable of removing selenate at a stable high ratio without flowing solid Se was achieved. 3.2. Se treatment experiment Since both soluble and solid Se could be removed almost completely within 3 days in batch 6 (phase II), the SBR was operated with a shorter cycle duration of 3 days from phase III to phase VII. During phases III to VII, the influent substrate (acetate) concentration was decreased gradually after confirming stable Se removal. Since high substrate supplementation leads to increased cost not only for the substrate itself but also the DOC removal posttreatment, it is important to identify appropriate substrate donation levels. The performance of Se and DOC removal during phases III to VII are shown in Fig. 2. During phase III, substrate was added at a DOC concentration of 1600 mg/L and soluble Se was removed efficiently at > 90% (excepting temporal worsening in batch 31), which confirmed stable removal of soluble Se. By contrast, solid Se was detected in the effluent at 0.05e0.40 mM, and total Se removal efficiency was 76 ± 12%. Although DOC removal efficiency increased to 20e40% during the former part of phase III, it declined to around 10% after batch 30. The decline in DOC removal efficiency might be associated with the decrease of soluble Se removal observed in batch 31, and could have been caused by incomplete sealing of the flask. During phases IV to VII where the influent concentration of DOC was decreased to 1200 (phase IV), 800 (phase V), 400 (phase VI), and 200 mg/L (phase VII) at intervals of 6 batches, soluble Se removal performance was maintained above 96%. Solid Se concentration in the effluent was also maintained at 0.05e0.26 mM, and above 80% total Se removal was achieved in most batches. The abnormally high solid Se in the effluent in batch 60 might be caused by unsuitable placement of the SBR flask in the rotary shaker. DOC consumption during phases IV to VII was generally 100e200 mg/L per batch (Fig. S1), and DOC removal efficiency varied largely with average removal efficiencies of 6, 21, 55, and 58% in phases IV, V, VI, and VII, respectively. Consequently, high concentrations of DOC remained in the effluent. The results during phases III to VII verified that soluble Se can be removed from high-salinity selenate wastewater stably with a high efficiency by SBR with acclimatized activated sludge. It was also found that residual solid Se in the effluent can be avoided by appropriately adjusting substrate concentrations. Further residual solid Se can be easily removed by appropriate post-treatment, like chemical coagulation and filtration (Staicu et al., 2015). Thus, the results demonstrated biological removal of selenate in high-salinity wastewater for the first time. Organic substrate is an important factor that greatly affects the biological Se removal process. Acetate used as the substrate in this study has been practically applied as a denitrification substrate (Dong et al., 2017; Gong et al., 2013), and thus would be
economically acceptable in wastewater treatment. However, even in phase VII with the lowest acetate supplementation (200 mg/L as DOC), during which nearly 8 mM carbon on average was consumed along with nearly complete removal of 1 mM selenate in the influent, acetate remained at considerable concentration in the effluent. Residual DOC should be further removed by posttreatment, like subsequential aerobic biological treatment by activated sludge (Mal et al., 2017). Also, exploration of alternative substrates could prove useful. Time courses of soluble Se (selenite and selenate) in a batch cycle were monitored in batches 34 and 71 (Fig. 3). During batch 34, selenate was gradually depleted in 70 h with little (<0.1 mM) selenite accumulation (Fig. 3a). In contrast, selenate was reduced completely within 28 h during batch 71, with a partial accumulation of selenite which was further removed to an undetectable level within 48 h (Fig. 3b). More efficient Se removal in batch 71 (phase VII) than in batch 34 (phase III) might be due to further enrichment of microorganisms associated with Se removal, especially selenatereducers, after the start-up period, or by the negative influence of excessive organic substrate on selenate reduction or total Se removal. Although the reason for the difference in Se removal levels between batches could not be identified, the slight succession of bacterial communities as described below may support the former possibility. Soluble Se time course monitoring verified that selenate and selenite reductions proceed synchronously. Biological reduction of selenate to selenite has been known to occur by anaerobic respiration under anoxic conditions. Thus, it could be speculated that selenite removal also occurs under anoxic conditions, which differs from our earlier study performing aerobic selenite removal in activated sludge (Zhang et al., 2019). 3.3. Se mass balance To clarify the fate of selenate added to the SBR and removed from the water phase, the Se mass balance was investigated by monitoring the amount of Se accumulated in the sludge in addition to soluble and solid Se in the influent, effluent, and ES discharged from the SBR. Se accumulation in the sludge was observed from batch 2, when detectable soluble Se removal occurred, which gradually increased to around 2800 mg/L (Fig. 2e). Figure 4 gives the cumulative Se distribution during phases III and VII. Soluble Se that existed as selenate in the influent at the beginning of a batch cycle was transformed mostly to solid Se, namely elemental Se. For example, during phase VII, 82% of elemental Se generated was accumulated in the sludge, including 54% remaining in the SBR, 28% removed as ES, and 12% present in the effluent. These results indicated that the removal of soluble Se (selenate) in the SBR occurred mainly by bioprecipitation, namely in a series of reductive reactions from selenate to elemental selenium via selenite. Although the results of mass balance analysis during phase III also inferred the possibility of biovolatilization, its contribution to Se removal from the water phase was minor, if any. Biovolatilization of selenium has been reported to occur when protein substrates are supplemented (Thompson-Eagle and Frankenberger, 1990; Zhang et al., 2019; Zhang and Frankenberger, 1999), and thus minimally occurred in this study where acetate was used as the substrate. In contrast, the amount of solid Se discharged as ES increased from 12% (phase III) to 28% (phase VII) along with an increase of solid Se accumulated in the SBR sludge. Nevertheless, solid Se in effluent decreased slightly from 20% (phase III) to 12% (phase VII). This might be a result of generated Se particle properties that settle easily and are resistant to suspension in the water phase. A previous study has reported that elemental Se was generated as nanoparticles and readily discharged from the reactor in fresh water,
Y. Zhang et al. / Water Research 154 (2019) 327e335
331
Fig. 2. Se removal efficiencies (a), Se concentrations (b), dissolved organic carbon (DOC) removal efficiencies (c), DOC concentrations (d), and solid Se and mixed liquor suspended solids (MLSS) (e). III to VII over the figure indicate phase No.
332
Y. Zhang et al. / Water Research 154 (2019) 327e335
Fig. 3. Time course of soluble Se concentration and dissolved organic carbon consumption monitored on batch 34 (phase III) and batch 71 (phase VII).
whereas biogenic Se was colloidally destabilized and was expected to settle faster in marine environments (Buchs et al., 2013). 3.4. Characterization of sludge and Se deposition
Fig. 4. Cumulative Se mass balance during phases III and VII. Batch 60 (phase VII) was excluded from calculation of Se mass balance.
The Se mass balance study revealed that most soluble Se removed was accumulated in the sludge. As shown in Fig. 2e, solid Se in the sludge accounted for approximately 22% (902 mg/L) of MLSS concentration at 4050 mg/L in batch 23, and 47% (2577 mg/L) of MLSS concentration at 5460 mg/L in batch 69, respectively. This indicated that biomass concentration was maintained stably at around 3000 mg/L, and that accumulation of solid Se in the sludge increased throughout the SBR operation period. The color change of the sludge to red during the start-up period indicated the formation of amorphous Se. This resulted in poor sludge settleability and effluent containing relatively high concentrations of solid Se. Thereafter sludge color change to black indicated the formation of crystalline Se, and excellent sludge settleability was maintained independently from the MLSS concentration, excepting the temporal worsening of settleability at batch 60 due to suspected equipment trouble. In general, activated sludge with MLSS as high
Y. Zhang et al. / Water Research 154 (2019) 327e335
333
Fig. 5. TEM image of Se particles (a) and corresponding EDX spectrum (b) in SBR sludge after phase VII.
as 5000 mg/L encountered difficulty in solid-liquid separation. Thus, it was suggested that enhancement of sludge settleability was attributable to the accumulation of solid Se in the sludge as similar phenomena have been observed in previous studies (Jain et al., 2015; Zhang et al., 2019). TEM analysis was conducted to characterize the sludge Se accumulation. TEM analysis of the sludge after phase VII where Se was removed efficiently found that the particles, most of which were confirmed to consist of Se by STEM-EDX, were present mainly as large rods approximately 2 mm long (Fig. 5). Rod Se particles were further confirmed by EDX mapping of the same area (Fig. S2). Extremely large rod particles of above 10 mm were also observed, which has never been reported before. By contrast, smaller nanospheres, which have been reported as the major morphology of elemental selenium accumulated in aerobic granular sludge (Nancharaiah et al., 2018) and activated sludge (Jain et al., 2015), were rarely observed inside or at the surface of bacterial cells (data not shown). The formation of such large crystalline Se particles in the sludge would strongly support good sludge settleability throughout the SBR operation period as reported previously (Jain et al., 2015). Although the reason for the formation of large crystalline Se particles is unclear, it might be a specific phenomenon for solid Se formation from high Se-containing wastewater under highsalinity conditions. 3.5. SBR bacterial communities 16S rRNA gene amplicon sequencing of the nine sludge samples obtained 41,449 to 59,912 sequence reads per a sample (483,575 sequence reads in total), which were further classified into 3,756 OTUs (Table 2). Based on the OTU number, Shannon diversity index and Chao1 richness estimator, the diversity of bacterial communities was highest in the seed sludge and phase I, declined from phase I to phase VI, and slightly increased from phase VI to phase VII. The results clearly indicated the enrichment of bacterial
populations associated with reduction and removal of Se under high salinity conditions. Bacterial community composition at the phylum level is summarized in Fig. 6. Overall phylum composition did not change greatly throughout the SBR operation period. Proteobacteria was found to be the most abundant phylum across all samples, accounting for 49e69% of total populations. The abundance of Bacteroidetes increased during phase I, but thereafter decreased along with the cultivation. At the OTU level, 35 OTUs were abundantly detected (>1%) in at least one of the nine samples analyzed. The most abundant representative OTUs are summarized in Table 3. Denovo9544 (16%), denovo22361 (8%), and denovo4423 (5%) were present in relatively high abundance in the seed sludge, and were assigned to Dermatophilaceae, Arcobacter, and Chromatiales, respectively; however, their abundance declined following batch 3 (phase I). Alternatively, denovo2041 and denovo6468, which were assigned to Marinobacterium and Marinobacter respectively, grew from previously undetectable levels in the seed sludge to highly dominant colonies of 42% and 11% in phase V (batch 47). Along with decreasing DOC input, the abundance of denovo2041 declined to 24% (phase VII), whereas denovo6468 maintained its increasing trend till the end of SBR operation. The dominance of these halophilic bacteria would be attributable to the saline condition of the synthetic wastewater in this study. Also, sulfur-reducers (denovo18020 and denovo3537) and a nitrate-reducer (denovo7316) were detected, accounting for 7% and 6% of the total population in batch 92 (phase VII). Throughout the SBR operation period in this study, bacterial populations that have been previously reported as capable of reducing selenate/selenite under saline conditions, such as Bacillus (Blum et al., 1998; Mishra et al., 2011) and Selenihalanaerobacter (Blum et al., 2001), were never detected. In addition, our attempt by conventional cultivation methods could not isolate halotolerant bacteria with a strong selenate reducing ability that could explain the selenate removal performance observed in the SBR (data not
Table 2 Summary of 16S rRNA gene amplicon sequencing of sludge samples. Sample (phase)
Qualified sequence reads
OTUs
Shannon diversity
Chao1 richness
Seed Batch Batch Batch Batch Batch Batch Batch Batch
41,449 59,912 56,784 56,932 51,653 57,633 48,002 53,827 57,383
1,456 1,717 1,099 852 763 711 644 692 696
4.52 4.21 3.17 2.77 2.69 2.49 2.51 2.76 2.69
3,794 4,674 2,925 2,027 1,737 1,656 1,346 1,718 1,837
3 (I) 6 (II) 35 (III) 41 (IV) 47 (V) 53 (VI) 69 (VII) 92 (VII)
334
Y. Zhang et al. / Water Research 154 (2019) 327e335
Fig. 6. Phylum level bacterial community structure. Relative abundance is expressed as the ratio of targeted sequence reads to total high-quality reads. Rare phyla with less than 1% abundance were grouped as “Unassigned; Others”.
Table 3 Relative abundance of representative predominant operational taxonomic units in sludge samples. OTU ID
denovo2041 denovo3537 denovo4423 denovo6468 denovo7316 denovo9544 denovo18020 denovo22361
Taxonomic classificationa
Marinobacterium Desulfuromonadales Chromatiales Marinobacter Nitratireductor Dermatophilaceae Desulfuromonas Arcobacter
Batch (phase) seed
3 (I)
6 (II)
35 (III)
41 (IV)
47 (V)
53 (VI)
69 (VII)
92 (VII)
0% 0% 5% 0% 0% 16% 1% 8%
14% 0% 1% 1% 0% 9% 2% 0%
38% 5% 0% 3% 0% 3% 1% 0%
34% 14% 0% 4% 1% 4% 0% 0%
36% 14% 0% 7% 1% 3% 0% 0%
42% 9% 0% 11% 1% 3% 0% 0%
38% 9% 0% 16% 1% 3% 0% 0%
23% 6% 0% 25% 8% 3% 0% 0%
24% 7% 0% 27% 6% 2% 0% 0%
a Taxonomic classification of the operational taxonomic units indicates the genus. If operational taxonomic units could not be assigned at the genus level, the lowest taxonomy (family/order/class) was shown.
shown). Thus, the bacterial populations responsible for Se removal in the SBR were not identified in this study. Some of the predominant bacteria detected in the 16S rRNA gene amplicon sequencing, such as Marinobacterium and Marinobacter, might be associated with selenate reduction, but are difficult to isolate by conventional cultivation methods. Another possibility would be cooperative selenate reduction by weak selenate-reducers. Identification of responsible bacteria or clarification of the cooperative relationship for selenate reduction is an important factor for understanding Se removal mechanisms and developing more efficient treatments of selenate-containing saline wastewater under anoxic conditions.
Acknowledgements The TEM analysis conducted in this study was supported by “Advanced Characterization Nanotechnology Platform, Nanotechnology Platform Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan” at the Research Center for Ultra-High Voltage Electron Microscopy (Nanotechnology Open Facilities) in Osaka University. This work was partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI [grant numbers 15K16145]. Appendix A. Supplementary data
4. Conclusions Biological reduction of selenate in the presence of considerable salinity (3% w/v NaCl) was demonstrated for the first time by SBR inoculated with activated sludge under oxygen-limiting conditions. Soluble Se removal above 96% and total Se removal above 80% were achieved in most batches. Biologically reduced Se was accumulated mainly as large rods in the sludge, which contributed to good sludge settleability. In contrast, the considerable remaining DOC concentrations following treatment, even at the lowest acetate donation, suggested the requirement of residual DOC post-treatment. This study offers a possibility for biological reduction of Se from Se-containing saline wastewater.
Declaration of interests None.
Supplementary data to this article can be found online at https://doi.org/10.1016/j.watres.2019.01.059. References Blum, J.S., Bindi, A.B., Buzzelli, J., Stolz, J.F., Oremland, R.S., 1998. Bacillus arsenicoselenatis, sp. nov., and Bacillus selenitireducens, sp. nov.: two haloalkaliphiles from Mono Lake, California that respire oxyanions of selenium and arsenic. Arch. Microbiol. 171, 19e30. https://doi.org/10.1007/s002030050673. Blum, J.S., Stolz, J.F., Oren, A., Oremland, R.S., 2001. Selenihalanaerobacter shriftii gen. nov., sp. nov., a halophilic anaerobe from Dead Sea sediments that respires selenate. Arch. Microbiol. 175, 208e219. https://doi.org/10.1007/ s002030100257. Buchs, B., Evangelou, M.W.H., Winkel, L.H.E., Lenz, M., 2013. Colloidal properties of nanoparticular biogenic selenium govern environmental fate and bioremediation effectiveness. Environ. Sci. Technol. 47 (5), 2401e2407. https://doi.org/10. 1021/es304940s. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., ~ a, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Fierer, N., Pen Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of highthroughput community sequencing data. Nat. Methods 7 (5), 335e336. https://doi.org/10.1038/nmeth.f.303. Chasteen, T.G., Bentley, R., 2003. Biomethylation of selenium and tellurium:
Y. Zhang et al. / Water Research 154 (2019) 327e335 microorganisms and plants. Chem. Rev. 103 (1), 1e26. https://doi.org/10.1021/ cr010210þ. Dong, H., Wang, W., Song, Z., Dong, H., Wang, J., Sun, S., Zhang, Z., Ke, M., Zhang, Z., Wu, W.-M., Zhang, G., Ma, J., 2017. A high-efficiency denitrification bioreactor for the treatment of acrylonitrile wastewater using waterborne polyurethane immobilized activated sludge. Bioresour. Technol. 239, 472e481. https://doi. org/10.1016/j.biortech.2017.05.015. Fujita, M., Ike, M., Nishimoto, S., Takahashi, K., Kashiwa, M., 1997. Isolation and characterization of a novel selenate-reducing bacterium, Bacillus sp. SF-1. J. Ferment. Bioeng. 83 (6), 517e522. https://doi.org/10.1016/S0922-338X(97) 81130-0. Gong, L., Huo, M., Yang, Q., Li, J., Ma, B., Zhu, R., Wang, S., Peng, Y., 2013. Performance of heterotrophic partial denitrification under feast-famine condition of electron donor: A case study using acetate as external carbon source. Bioresour. Technol. 133, 263e269. https://doi.org/10.1016/j.biortech.2012.12.108. Gonzalez-Gil, G., Lens, P.N.L., Saikaly, P.E., 2016. Selenite reduction by anaerobic microbial aggregates: microbial community structure, and proteins associated to the produced selenium spheres. Front. Microbiol. 7, 1e14. https://doi.org/10. 3389/fmicb.2016.00571. Ike, M., Takahashi, K., Fujita, T., Kashiwa, M., Fujita, M., 2000. Selenate reduction by bacteria isolated from aquatic environment free from selenium contamination. Water Res. 34 (11), 3019e3025. https://doi.org/10.1016/S0043-1354(00)000415. Jain, R., Seder-Colomina, M., Jordan, N., Dessi, P., Cosmidis, J., van Hullebusch, E.D., Weiss, S., Farges, F., Lens, P.N.L., 2015. Entrapped elemental selenium nanoparticles affect physicochemical properties of selenium fed activated sludge. J. Hazard Mater. 295, 193e200. https://doi.org/10.1016/j.jhazmat.2015.03.043. Kagami, T., Narita, T., Kuroda, M., Notaguchi, E., Yamashita, M., Sei, K., Soda, S., Ike, M., 2013. Effective selenium volatilization under aerobic conditions and recovery from the aqueous phase by Pseudomonas stutzeri NT-I. Water Res. 47 (3), 1361e1368. https://doi.org/10.1016/j.watres.2012.12.001. Kuroda, M., Notaguchi, E., Sato, A., Yoshioka, M., Hasegawa, A., Kagami, T., Narita, T., Yamashita, M., Sei, K., Soda, S., Ike, M., 2011a. Characterization of Pseudomonas stutzeri NT-I capable of removing soluble selenium from the aqueous phase under aerobic conditions. J. Biosci. Bioeng. 112 (3), 259e264. https://doi.org/10. 1016/j.jbiosc.2011.05.012. Kuroda, M., Yamashita, M., Miwa, E., Imao, K., Fujimoto, N., Ono, H., Nagano, K., Sei, K., Ike, M., 2011b. Molecular cloning and characterization of the srdBCA operon, encoding the respiratory selenate reductase complex, from the selenate-reducing bacterium Bacillus selenatarsenatis SF-1. J. Bacteriol. 193 (9), 2141e2148. https://doi.org/10.1128/JB.01197-10. Lawson, S., Macy, J.M., 1995. Bioremediation of selenite in oil refinery wastewater. Appl. Microbiol. Biotechnol. 43 (4), 762e765. https://doi.org/10.1007/ BF00164785. Luoma, S.N., Presser, T.S., 2009. Emerging opportunities in management of selenium contamination. Environ. Sci. Technol. 43 (22), 8483e8487. https://doi.org/10. 1021/es900828h. Macy, J.M., Lawson, S., 1993. Cell yield (YM) of Thauera selenatis grown anaerobically with acetate plus selenate or nitrate. Arch. Microbiol. 160 (4), 295e298. https:// doi.org/10.1007/BF00292080. Mal, J., Nancharaiah, Y.V., van Hullebusch, E.D., Lens, P.N.L., 2017. Biological removal of selenate and ammonium by activated sludge in a sequencing batch reactor. Bioresour. Technol. 229, 11e19. https://doi.org/10.1016/j.biortech.2016.12.112. Mishra, R.R., Prajapati, S., Das, J., Dangar, T.K., Das, N., Thatoi, H., 2011. Reduction of selenite to red elemental selenium by moderately halotolerant Bacillus megaterium strains isolated from Bhitarkanika mangrove soil and characterization of reduced product. Chemosphere 84 (9), 1231e1237. https://doi.org/10.1016/j. chemosphere.2011.05.025. Muscatello, J.R., Janz, D.M., 2009. Selenium accumulation in aquatic biota downstream of a uranium mining and milling operation. Sci. Total Environ. 407 (4),
335
1318e1325. https://doi.org/10.1016/j.scitotenv.2008.10.046. Nancharaiah, Y.V., Joshi, H.M., Hausner, M., Venugopalan, V.P., 2008. Bioaugmentation of aerobic microbial granules with Pseudomonas putida carrying TOL plasmid. Chemosphere 71 (1), 30e35. https://doi.org/10.1016/j. chemosphere.2007.10.062. Nancharaiah, Y.V., Lens, P.N.L., 2015a. Selenium biomineralization for biotechnological applications. Trends Biotechnol. 33 (6), 323e330. https://doi.org/10. 1016/j.tibtech.2015.03.004. Nancharaiah, Y.V., Lens, P.N.L., 2015b. Ecology and biotechnology of seleniumrespiring bacteria. Microbiol. Mol. Biol. Rev. 79 (1), 61e80. https://doi.org/10. 1128/MMBR.00037-14. Nancharaiah, Y.V., Sarvajith, M., Lens, P.N.L., 2018. Selenite reduction and ammoniacal nitrogen removal in an aerobic granular sludge sequencing batch reactor. Water Res. 131, 131e141. https://doi.org/10.1016/j.watres.2017.12.028. Nguyen, V.K., Park, Y., Yu, J., Lee, T., 2016. Microbial selenite reduction with organic carbon and electrode as sole electron donor by a bacterium isolated from domestic wastewater. Bioresour. Technol. 212, 182e189. https://doi.org/10.1016/j. biortech.2016.04.033. Peiffer, J.A., Spor, A., Koren, O., Jin, Z., Tringe, S.G., Dangl, J.L., Buckler, E.S., Ley, R.E., 2013. Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc. Natl. Acad. Sci. Unit. States Am. 110 (16), 6548e6553. https://doi.org/10.1073/pnas.1302837110. Pollice, A., Rozzi, A., Tomei, M.C., Di Pinto, A.C., Limoni, N., 2000. Monitoring the inhibitory effect of NaCl on anaerobic wastewater treatment processes by the rantox biosensor. Environ. Technol. 21 (5), 535e544. https://doi.org/10.1080/ 09593332408618095. Soda, S., Hasegawa, A., Kuroda, M., Hanada, A., Yamashita, M., Ike, M., 2015. Selenium recovery from kiln powder of cement manufacturing by chemical leaching and bioreduction. Water Sci. Technol. 72 (8), 1294e1300. https://doi.org/10. 2166/wst.2015.339. Soda, S., Kashiwa, M., Kagami, T., Kuroda, M., Yamashita, M., Ike, M., 2011. Laboratory-scale bioreactors for soluble selenium removal from selenium refinery wastewater using anaerobic sludge. Desalination 279 (1e3), 433e438. https:// doi.org/10.1016/j.desal.2011.06.031. Staicu, L.C., van Hullebusch, E.D., Oturan, M.A., Ackerson, C.J., Lens, P.N.L., 2015. Removal of colloidal biogenic selenium from wastewater. Chemosphere 125, 130e138. https://doi.org/10.1016/j.chemosphere.2014.12.018. Takada, K., Shiba, T., Yamaguchi, T., Akane, Y., Nakayama, Y., Soda, S., Inoue, D., Ike, M., 2018. Cake layer bacterial communities during different biofouling stages in full-scale membrane bioreactors. Bioresour. Technol. 259, 259e267. https://doi.org/10.1016/j.biortech.2018.03.051. Thompson-Eagle, E.T., Frankenberger Jr., W.T., 1990. Protein-mediated selenium biomethylation in evaporation pond water. Environ. Toxicol. Chem. 9 (12), 1453e1462. https://doi.org/10.1002/etc.5620091204. Van Hullebusch, E.D. (Ed.), 2017. Bioremediation of selenium contaminated wastewater. Springer, Cham. https://doi.org/10.1007/978-3-319-57831-6. Yan, R., Gauthier, D., Flamant, G., Peraudeau, G., Lu, J., Zheng, C., 2001. Fate of selenium in coal combustion: volatilization and speciation in the flue gas. Environ. Sci. Technol. 35 (7), 1406e1410. https://doi.org/10.1021/es0001005. Zannoni, D., Borsetti, F., Harrison, J.J., Turner, R.J., 2007. The bacterial response to the chalcogen metalloids Se and Te. Adv. Microb. Physiol. 53, 1e312. https://doi.org/ 10.1016/S0065-2911(07)53001-8. Zhang, Y., Kuroda, M., Nakatani, Y., Soda, S., Ike, M., 2019. Removal of selenite from artificial wastewater with high salinity by activated sludge in aerobic sequencing batch reactors. J. Biosci. Bioeng. (in press) https://doi.org/10.1016/j. jbiosc.2018.11.002. Zhang, Y., Frankenberger Jr., W.T., 1999. Effects of soil moisture, depth, and organic amendments on selenium volatilization. J. Environ. Qual. 28 (4), 1321e1326. https://doi.org/10.2134/jeq1999.00472425002800040037x.