Evaluation of Combined Anaerobic Membrane Bioreactor and Downflow Hanging Sponge Reactor for Treatment of Synthetic Textile Wastewater
Choerudin Choerudin1,2, Fauziyah Istiqomah Arrahmah1, Jonatan Kevin Daniel1, Takahiro Watari3, Takashi Yamaguchi3, Tjandra Setiadi1,4,*
1 Department of Chemical Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung, 40132, Indonesia.
2 Department of Chemical Engineering, Institut Teknologi Nasional Bandung, Jl. PHH. Mustafa 23, Bandung, 40124, Indonesia.
3Department of Civil and Environmental Engineering, Nagaoka University of Technology, Niigata 940-2188, Japan.
4Center for Environmental Studies (PSLH), Institut Teknologi Bandung, Jl. Sangkuriang 42A, Bandung, 40135, Indonesia
*Corresponding author: [email protected] Abstract
Various toxic chemicals in textile wastewater can cause serious problems for the ecosystem and human health if it is discharged without proper treatment. In this study, the performance of textile wastewater treatment using an anaerobic membrane bioreactor (AnMBR) combined with a downflow hanging sponge (DHS) reactor is evaluated. An AnMBR with a working volume of 5 L and a third-generation DHS with a working volume of 0.5 L is used to treat synthetic textile wastewater containing Reactive Black 5 azo dye. The experiment is performed under ambient conditions using AnMBR hydraulic retention times (HRTs) of 12 and 24 h, DHS HRTs of 1.4 and 2.8 h, and membrane fluxes of 2.65 and 5.21 LMH. The AnMBR-DHS combination significantly reduces the biochemical oxygen demand, chemical oxygen demand, and color of the synthetic textile wastewater by approximately 97.3 ± 1.8%, 94.4 ± 4.8%, and 95.0 ± 1.6%, respectively. Most of the reduction occurs in the AnMBR. The HRT and membrane flux do not significantly affect the performance of the AnMBR-DHS system. The microbial community in the AnMBR is dominated by the phyla Euryarchaeota, Caldiserica, and Proteobacteria, whereas that in the DHS is dominated by Proteobacteria. Some of the genera found in the AnMBR can reportedly reduce azo dyes, whereas some of those found in the DHS can reportedly degrade sulfonated aromatic amines.
Keywords: Anaerobic aerobic process; Biofilm; Microbial community; Reactive Black 5;
Textile wastewater
1. Introduction
The consumption and production of textiles grow exponentially with the growth of the global population [1]. The approximate annual textile consumption is 7 kg per capita with the global population is over 7 billion, the estimated annual textile production will be 49 billion kg. On the other hand, producing 1 kg of textiles requires 150–350 L of water for dyeing and finishing processes [2]. Therefore, population growth will significantly increase wastewater volume from this industry.
Various pollutants in textile wastewater can cause serious problems for aquatic ecosystems and human health if it is discharged without proper treatment [3]. Heavy metals, chlorine, phthalocyanine dyes, and other textile wastewater chemicals are known to be toxic, carcinogenic, and mutagenic [3]. The dangerous pollutants in textile wastewater are produced mainly during dyeing and printing and 85% of the wastewater is released from the dyeing process [4,5]. Only a maximum of 80% of dyes can be adsorbed in the dyeing process [5]; thus, approximately 20% of dyes are lost to wastewater during dyeing. Among various types of dyes, azo dyes are the most widely used dyes and account for 60–70% of total dye use [6]. Colored wastewater causes problems with photosynthesis for algae and aquatic plants. It also affects other aquatic organisms because it causes low light penetration and oxygen depletion [4].
To protect humans and the environment from the negative effects of wastewater, regulations regarding wastewater have been established. They continue to evolve to ensure acceptable management that meets health and environmental standards [7]. In Indonesia, the latest textile wastewater regulation was released in April 2019 and included a Pt-Co color standard of a maximum of 200 [8]. This shows the importance of dye removal in textile wastewater treatment.
Physicochemical, oxidation, and biological methods have been developed for treating textile wastewater to obtain the desired cost and efficiency. These methods have certain advantages and drawbacks with most of the methods have removal dye efficiencies of more than 80% and a few methods up to 90% [5]. Although they are easy to implement, physicochemical methods require high electricity consumption and have relatively low output because they generate a large number of by-products and non-reusable sludge. Besides, some can be used only for low- concentration wastewater, and some are not cost-effective [9,10]. On the other hands, the main drawbacks of oxidation methods, including chemical oxidation and advanced oxidation processes, are the release of toxins, high chemical usage, high energy usage, and high cost [4].
Biological methods have significant advantages over physical and oxidation methods. They
require fewer or even no chemicals and thus are more eco-friendly. One of alternative bioprocessing to textile wastewater treatment which stands out as a more cost-effective and environmentally friendly is the anaerobic–aerobic methods [11].
The anaerobic-aerobic methods have been employed in municipal and industrial wastewater treatment for many years [12]. This methods is the most logical concept that able to biodegrade the azo dyes [13]. In the anaerobic stage, azo dyes are broken down on the azo bond into a colorless aromatic amine compound. This compound is resistant to the anaerobic process, but it is easily degraded in the aerobic stage [14]. Most of azo dyes are following this degradation mechanism even though sometimes the fate of the degraded aromatic amines is difficult to predict because of the data limitation and the analytical problems [13]. However, the anaerobic- aerobic methods still be used because the toxicity after the treatment are less than the original azo dyes. Numerous types of azo dyes are available but Reactive Black 5 (RB5), a diazo dye and member of vinyl sulfone dyes family, is the most widely used dye in the textile industry [11]. The RB5 is also widely used as a representative compound to assess the efficiency of decolorization methods for azo dyes [15].
The anaerobic-aerobic methods can be conducted in four different types of unit operation [12].
The anaerobic-aerobic methods can be: (i) in separated units of anaerobic and aerobic; (ii) in separated zone of anaerobic and aerobic in a single unit; (iii) temporal anaerobic and aerobic phase in single unit (e.g. sequencing batch reactor); or (iv) combined anaerobic-aerobic culture system based on the limitation of oxygen diffusion in the medium (e.g. biofilm and granular sludge). As separated units, many types of the process have been used for anaerobic textile wastewater treatment, including the continuous stirred tank reactor, anaerobic sequencing batch reactor (SBR), upflow anaerobic sludge blanket (UASB), expanded granular sludge blanket, internal circulation reactor, anaerobic baffled reactor (ABR), anaerobic filter reactor, and anaerobic fluidized bed reactor [16]. Some of them have been used for industrial applications, whereas others are under development. However, the limited hydraulic retention time (HRT) and cell concentration during the separation process owing to washout are among the main drawbacks of anaerobic technologies [16]. To overcome the washout problem, the HRT should be independent of the cell concentration, which can be achieved by using a membrane as the separator. Despite promising results, few studies of the use of anaerobic membrane bioreactors (AnMBRs) for textile wastewater treatment have been conducted [17,18]. The use of a membrane always correlates with fouling as its drawbacks. However,
some methods to overcome the membrane fouling are provided based on the material of membrane.
On the other hands, the use of a downflow hanging sponge (DHS) reactor is known to be a low-cost aerobic treatment, and the DHS has been applied for the post-treatment of high- strength industrial wastewater and municipal wastewater [19,20]. Compared to other post- treatment technologies like polishing ponds (PP), conventional activated sludge process (CAS), and PP with aeration, DHS technology has proven to give much better in BOD, COD, TSS, pathogens, orthophosphate, and nitrogen removal. It was also proven that DHS can work at an extremely low HRT with good organics removal and ammonium oxidation efficiency. DHS is recommended to be applied in developing countries because it is cost-effective and requires less area compared to other conventional processes [21,22]. Moreover, DHS is also a low energy technology because it requires only electrical energy for wastewater distribution, not for aeration devices. It was reported that the energy requirement for DHS systems is estimated to be approximately 75 % lower than CAS [23]. The drawback of DHS-G3 is the declining DO concentration followed by a decrease of nitrification activity after several months of operations.
It was reported that this is caused by the random arrangement of sponge media, as it is the reason for the lack of air transfer between sponge media inside the reactor [23].
A third-generation (G3) DHS reactor is easier to construct than other types of DHS reactors, as it involves the simple random packing of the sponge media [24]. The G3-DHS reactor was used recently to treat low-strength textile wastewater and exhibited high organic and color removal [25]. Thus, the treatment of textile wastewater using a combination of an AnMBR and a G3-DHS reactor shows promise. However, this combination has not been extensively studied.
In this study, the treatment of synthetic textile wastewater by an AnMBR-DHS system is evaluated. The chemical oxygen demand (COD), biochemical oxygen demand (BOD3), and Pt- Co color removal performance are evaluated using various values of two operational parameters (HRT and flux). Furthermore, the microbial communities are also investigated using the 16S rRNA gene sequence.
2. Materials and methods
This section describes the wastewater, reactor setup, and analytical methods. The analytical methods for the microbial community are covered in a separate subsection.
2.1. Synthetic dyeing wastewater
The synthetic textile wastewater used in this study was obtained using a previously published recipe [26]. The ratios of COD, N, and P for the anaerobic and aerobic synthetic wastewater were 250:5:1 and 100:5:1, respectively. The synthetic textile wastewater consisted of hydrolyzed starch (100 g of starch with 40 g of NaOH and water were hydrolyzed for 15 h at room temperature and then neutralized and diluted to 1 L) and hydrolyzed RB5 (5 g of RB5 with pH 12 were heated for 1 h at 80°C and then neutralized and diluted to 1 L). RB5 is a vinyl sulfone dye. Anaerobic synthetic wastewater for a continuous experiment was prepared once every two or three days using a dye concentration of 50 ppm and a COD concentration of 750 ppm. The BOD3/COD ratio was maintained at approximately 0.4.
2.2. System description and operational conditions
A schematic diagram of the AnMBR-DHS system is shown in Fig. 1. The anaerobic reactor was made of polyvinyl chloride and had a working volume of 5 L. There was an internal recirculation flow with a rate of 0.6 L/h. Hollow fiber membranes made of polyvinylidene difluoride with a pore size, an inner diameter, an outer diameter, and an effective length of 0.07–0.1 µm, 1.0 mm, 2.2 mm, and 15 cm, respectively, were used in this experiment.
Membrane areas of 0.06 and 0.10 m2 were used. The DHS reactor was a column made of acrylic and had an inner diameter of approximately 50 mm and a working volume of 0.5 L. A number of G3-DHS medium was used in this experiment and packed randomly in the column. The G3- DHS consisted of a sponge cube made of polyurethane in a net ring which made of polypropylene with a diameter and length of 32 mm. The influent of the DHS reactor was received directly from the AnMBR effluent without any pretreatment.
Fig. 1. Schematic diagram of AnMBR-DHS system.
Anaerobic sludge with an initial mixed liquor suspended solids (MLSS) concentration of 47.8 g/L was obtained from Bioprocess Technology Laboratory ITB, Bandung. The sludge originated from local municipal wastewater treatment in Bandung and was acclimatized to textile wastewater. The sludge and anaerobic synthetic wastewater were added to the reactor at a 1:1 ratio. The anaerobic synthetic wastewater was fed into the reactor at 1 L/day for three weeks. The DHS was acclimatized by soaking the sponges in the aerobic synthetic wastewater for three weeks. For the objectives of the study, the basis for selecting the specified HRT and the flux are as shown in Table 1. These operation parameter was the common operation used either in the AnMBR [17,18] or the DHS [25,27]. The entire system was operated continuously for 20 days at 30°C except for run III which only operated for 7 days due to the coronavirus diseases 2019 (COVID-19) pandemic. The shorten operation time of run III did not affect the biological processes, because a flux change was carried out by replacing the membrane with a different area. The fouled membrane was cleaned when its transmembrane pressure (TMP) reached 500 mbar. Physical cleaning was performed by air scouring with nitrogen gas for 15 min. Chemical cleaning was performed using NaOCl and a citric acid mixture after air scouring was conducted twice.
Table 1. Operational parameters used in this study.
Run Anaerobic Conditions Aerobic Conditions
HRT (h) Flux (LMH) HRT (h)
Evaluation of the AnMBR-DHS performances with varied HRT
I 24 5.21 2.8
II 12 5.21 1.4
Evaluation of the AnMBR-DHS performances with varied membrane fluxes
I 24 5.21 2.8
III 24 2.65 2.8
2.3.Analytical methods
Samples were collected daily from the influent system, AnMBR effluent, and DHS effluent.
They were centrifuged to separate the supernatant and total solids. The supernatant was then filtered and used to analyze the color removal, COD, and BOD3 parameters. The color removal and COD were monitored daily. To obtain the color removal parameter, the sample absorbance was measured by scanning spectrophotometry (UV-1800, Shimadzu) in the visible light region (380–700 nm). The color removal parameter was measured by comparing the peak absorbance
of the chromatograph at these wavelengths for each sample. The BOD3 parameter was obtained according to standard methods, with a temperature and time modification of 30°C and 3 days, respectively [28]. The COD and MLSS were analyzed as described in standard methods [29].
2.4. DNA extraction, PCR amplification, and 16S rRNA gene sequencing
Genomic DNA was extracted from the collected samples using a bead-beating system (ZymoBIOMICS™ DNA Miniprep Kit). The quality of the extracted DNA was monitored using a NanoDrop® and Qubit® kit/instrument (spectrophotometer and microvolume fluorometer). DNA amplification was conducted by the polymerase chain reaction (PCR) with bacterial region V3-V4 16S rRNA as the target using the primer sets 341F (5'-3':
CCTAYGGGRBGCASCAG) and 806R (5'-3': GGACTACNNGGGTATCTAAT). The PCR product was purified by gel electrophoresis, and PCR products with a length of 466 bp were chosen. 16S rRNA gene sequencing was performed using the Ion S5™XL system. Taxonomic classification was performed using the SILVA database (www.arb-silva.de) for bacteria, archaea, and eukaryotes. The analysis was performed by a representative of Novogen AIT Genomics Singapore in Indonesia at PT. Genetika Science Indonesia. Data mining of the identified microorganisms was conducted using Bergey’s Manual, a bacteria database (bacterio.net), and other available sources.
3. Results and discussion
This section provides the results of the experiments, followed by discussions. The first subsection describes the system removal performance, the second subsection covers the operation of the membrane bioreactor, and the third subsection describes the microbiological diversity.
3.1. Performance of the AnMBR-DHS system
Several samples of colored water with various RB5 concentrations were analyzed by scanning spectrophotometry at 380–700 nm. Fig. 2 shows that RB5 exhibits the maximum absorbance at 476 and 595 nm. The color removal efficiency was evaluated based on the obtained maximum absorbance.
Fig. 2. The absorbance of RB5 dye at different concentrations in the visible light region (380–700 nm).
Fig. S1 in the Supplementary Material shows the color removal efficiency obtained by comparing the absorbance reduction at 476 and 595 nm for each run. Overall, there is not much difference in color removal efficiency between the AnMBR and AnMBR-DHS processes. High decolorization was obtained in the AnMBR; the removal efficiency was approximately 92%
for both HRTs (12 and 24 h). The use of different membrane areas at the same operational HRT also had little effect, as indicated by the high decolorization [Fig. 3a and 3c] from the first day of operation and the significant decolorization efficiency throughout the operation time. The slight difference in the total removal by the AnMBR alone and by the AnMBR-DHS indicates that color removal occurs mainly in the anaerobic process (Fig. S2 in the Supplementary Material). The azo bond can reportedly be decolorized under aerobic and anaerobic conditions [30]. The anaerobic step causes reductive cleavage of the azo bond and thus leads to decolorization. However, in the presence of oxygen, this cleavage is not significant. These facts explain why most of the decolorization occurred in the AnMBR, and only slightly decolorization occurred in the DHS reactor. Both aerobic and anaerobic conditions reportedly exist in the DHS system [23]. The reason is the decrease in oxygen concentration toward the sponge center, which results in anaerobic conditions at the center of the sponge. The existence of anaerobic conditions in the DHS system might be the largest contributor to the system's color removal efficiency. Absorption of dyes on the sponge are not occurred because the sponge is made of polyurethane; as well as the net ring that supported the sponge which made of polyethylene. Absorption on these material are not occurred because the mechanism of azo
dye attached on the fiber is based on chemically bounding (covalent bond), but not physically bounding (hydrogen bond, Van der Waals bond, and ionic bond) [31]. Besides, the azo dyes have to meet some chemical condition to be attached on these materials which may not be provided in the DHS reactor.
The COD removal rate of the AnMBR was approximately 86%, and the DHS process increased this rate to approximately 95% total COD removal. This result indicates that COD removal occurred mainly in the AnMBR, which suggests that a significant amount of COD was converted to methane (data not shown). Other azo dye treatment studies [26,32] also produced similar results. Although the COD removal is quite high after the AnMBR process, aerobic treatment is required. The reason is the production of sulfonated aromatic amines, which cannot be degraded under anaerobic conditions but degrade easily under aerobic conditions.
Table 2 summarizes the COD, BOD3, and color removal efficiency of the combined AnMBR- DHS treatment from this study compared to other anaerobic-aerobic systems on textile wastewater treatment. The detailed removal efficiency of the AnMBR-DHS and other compared system are respectively available in Table S.1 and Table S.2 (Supplementary Material). The HRT and membrane flux do not significantly affect the COD, BOD3, and color removal. This result is consistent with another study [18], which reported nearly the same AnMBR decolorization at different operational HRTs and fluxes. However, another study [17]
showed that the HRT and flux in an AnMBR significantly affected the total decolorization. In an anaerobic study [33], azo decolorization occurred in a certain period at the beginning of the anaerobic process. Thus, an HRT longer than the decolorization period will not affect the color removal effectiveness. This decolorization period depends on the biodegradability of the wastewater. Thus, the use of different feeds might be the cause of these different results, as the results using synthetic textile wastewater non azo dyes [18] in compared to the real textile wastewater [17]. Real textile wastewater is much more complex because it contains many azo dye variations. It typically has a low BOD3/COD ratio or a biodegradability index (BI) of approximately 0.1–0.25 [34].
Table 2. Performance of other anaerobic-aerobic systems on textile wastewater treatment in compared to this study results.
Type of Textile Wastewater
Technology Removal Performance (%)
Reference
Anaerobic Aerobic COD BOD5 Color
Studies of anaerobic-aerobic textile wastewater treatment
Real ABR DHS 90 - 58 [11]
Synthetic,
azo dyes UASB activated sludge 67-88 - 85 [35]
Real Anaerobic-aerobic SBR 75 - 85 [32]
Synthetic,
non-azo Flocculent anaerobic-aerobic SBR 70 - 75-80 [26]
Synthetic,
non-azo Granular anaerobic-aerobic SBR 80 - 75-80 [26]
Studies of AnMBR for textile wastewater treatment
Real AnMBR - 88 - 94 [17]
Synthetic,
non-azo AnMBR - 94 - 99 [18]
Studies of DHS for textile wastewater treatment Synthetic,
azo dyes - DHS 83±8 93±12 72±34 [25]
Synthetic,
non-azo - DHS 66.5±7.1 - 90.2±3.1 [27]
This Study Synthetic,
azo dyes AnMBR DHS 97.3±0.9 95.3±3.4
(BOD3) 94.0 ±2.0 Run I Synthetic,
azo dyes AnMBR DHS 96.5±0.5 98.3±1.6
(BOD3) 97.5±3.2 Run II Synthetic,
azo dyes AnMBR DHS 96.2±0.6 98.4±0.5
(BOD3) 94.2±0.7 Run III
In this study, the BI of the influent system was maintained at approximately 0.4. As a reference, wastewater with a BI of less than 0.35 is difficult to biodegrade [36]. The results indicate that in this experiment, all the HRTs were longer than the azo decolorization period.
The quality of the treated wastewater on day 20 of run II (12 h, 5.21 LMH) is shown in Fig.3.
The effluent of the AnMBR [Fig. 3b] is far more colorless than the system influent [Fig. 3a], and the effluent from the DHS [Fig. 3c] is slightly less colorful than the effluent of the AnMBR.
Most aromatic amines from azo dye biodegradation are susceptible to autoxidation and form dark-colored polymers [37]. However, in this study, colored products of aromatic compounds were not observed, suggesting that these colored polymers were removed during treatment.
Table 3 shows the characteristics of the system influent, AnMBR effluent, and DHS effluent at days 10 and 20 of run II. It shows that all measured parameters in the influent are largely decreased in the AnMBR stage and slightly decreased in the DHS stage except for TSS.
Increasing TSS in the DHS stage was caused by the detached solids (biofilms /microorganisms) from the DHS as the microorganisms grow along the operation time. The detached solids were flowing out with the effluent and measured as the TSS. Therefore, a small settling tank as a post-treatment has been used in the full-scale operation [38]. However, the TSS value in the DHS stage still meets the regulation standard used in this study. These results indicate that the effluent of the AnMBR-DHS system meets Indonesia’s regulation standards.
This research firstly reported the application of the AnMBR-DHS system for treating textile wastewater. Our previous study [11] demonstrated that anaerobic treatment is beneficial for dye degradation, but frequently sludge washout occurred and remained industrial application problems. The AnMBR examined in this study can resolve these problems and generate a high- quality effluent. The post-treatment DHS reactor can remove the ammonia that cannot be removed in membrane reactors. Therefore, this obtained result will bring immense benefits to textile wastewater treatment.
Fig. 3. The appearance of water on day 20 of run II (12 h, 5.21 LMH): (A) system influent, (B) AnMBR effluent, and (C) DHS effluent.
Table 3. Characteristics of the influent, AnMBR effluent, and DHS effluent on days 10 and 20 of run II (12 h, 5.21 LMH).
Parameter Units Maximum Value* Influent AnMBR Effluent DHS Effluent
Color Pt-Co 200 611±39 102±5 66±5
BOD3 mg/L 60 414±145 25±19 20±2
COD mg/L 150 1021±363 64±51 50±5
TSS mg/L 50 231±29 18±8 32±3
Phenol mg/L 0.5 <0.001 <0.001 <0.001
Total chromium (Cr) mg/L 1 <0.001 <0.001 <0.001
Ammonia (NH3-N) mg/L 8 9.8±3.2 4.2±0.6 2.6±0.3
Sulfide (H2S) mg/L 0.3 0.09±0.04 0.07±0.03 0.07±0.02
Oil and fat mg/L 3 20±14 1.8±1.4 1.7±0.3
pH - 6–9 6.5±0.2 7.6±0.3 7.5±0.3
*Standard regulation based on Permen-LH No. 16 Year 2019 Attachment II [8]
3.2. Transmembrane pressure changes in AnMBR-DHS system
Fig. 4 shows the relationship between the TMP and the operation time of the AnMBR. The TMP increases more rapidly at a higher HRT and lower operational flux. The MLSS measured at an HRT of 12 h and flux of 5.21 LMH shows an increase of up to 23.7%. The TMP behavior in this experiment is consistent with the theoretical behavior of fouled membranes [39], that is, that the increase in TMP in the operating membrane at constant flux permeation is the result of membrane fouling. A shorter HRT also results in higher MLSS, soluble microbial product, and extracellular polymeric substance concentrations, and thus a higher membrane fouling rate [40].
Fig. 4. Relationship between TMP and AnMBR operation time.
3.3. Biodiversity and distribution of microorganisms in AnMBR-DHS system
Sludge samples for microbial analysis were collected from the AnMBR and DHS on day 10 of run II (12 h, 5.21 LMH) to investigate the microbial diversity. The numbers of organism sequences obtained for the AnMBR and DHS were 134,784 tags and 81,458 tags, respectively.
The AnMBR organisms were clustered into 1,222 operational taxonomic units (OTUs) based on 97% similarity. The DHS organisms were clustered into 1,318 OTUs. The alpha diversity index was evaluated to characterize the biodiversity of the AnMBR and DHS microbial communities in terms of community richness and community evenness. Higher community richness as indicated by higher values of the Observed species, Chao 1, and Shannon indicators,
whereas higher community evenness was indicated by higher values of the Gini-Simpson indicator [41,42].
The AnMBR community had 1,142 observed species, a Chao 1 value of 1,235, a Shannon value of 6.3, and a Gini-Simpson value of 0.958. By contrast, the DHS community had 1,318 observed species, a Chao 1 value of 1,390, a Shannon value of 7.1, and a Gini-Simpson value of 0.965. The observed species, Chao 1, and Shannon values indicate that the DHS has greater microbial richness than the AnMBR. The Gini-Simpson value indicates that the AnMBR has higher evenness than the DHS. The difference in oxygen concentration between the two bioreactors might explain this diversity analysis. The AnMBR provides only anaerobic conditions and thus is less diverse than the DHS, which provides anaerobic and aerobic conditions. However, the anaerobic conditions in the DHS occur only near and at the center of the sponge, and thus the microbial community of the DHS is less even than that of the AnMBR, which has an oxygen concentration of almost zero throughout the bioreactor [23]. This may become another advantage of DHS which can provide a combined culture of anaerobic-aerobic in the medium as well as the concept of granular sludge for the textile wastewater treatment [26,43].
Fig. 5 shows the major phyla observed in the AnMBR: Euryarchaeota (29%), Proteobacteria (24%), Bacteroidetes (11%), Caldiserica (9%), Aminicenantes (5%), Acidobacteria (5%), Tenericutes (4%), Firmicutes (4%), Verrucomicrobia (2%), and Spirochetes (2%). These phyla are commonly found in anaerobic digesters [11,44,45].
Fig. 5. The relative abundance of phyla in AnMBR.
Fig. 6a and 6b show the major observed genera and the phylogenetic tree from the AnMBR microbial community, respectively. The 10 major genera are Methanosaeta (17%), Methanolinea (10%), Caldisericum (9%), Smithella (4%), Spirillum (3%), Tolumonas (3%), Desulfomonile (2%), Aeromonas (1%), Methanoregula (1%), and Syntrophus (1%).
Methanosaeta and Methanoregula are acetoclastic methanogens [41]. By contrast, Methanolinea is a hydrogenotrophic methanogen [11]. Acetoclastic methanogens convert acetate to methane and carbon dioxide, whereas hydrogenotrophic methanogens convert hydrogen and carbon dioxide to methane. Hydrogenotrophic methanogens are believed to have a significant effect on azo reduction and have better resistance to high azo concentration (comparing to acetoclastic methanogens). By contrast, acetoclastic methanogens have no significant effect and quite susceptible to the high azo compound concentration; thus, acetoclastic methanogens grow more slowly than hydrogenotrophic methanogens during azo dye treatment [46]. In agreement with this theory, an earlier experiment [11] showed a higher population of hydrogenotrophic methanogens compared to acetoclastic methanogens.
Although most of the phyla identified in the AnMBR in this experiment are quite similar to a study [11], the relative abundance of the methanogens is not; that is, this experiment shows a higher population of acetoclastic methanogens compared to hydrogenotrophic methanogens.
This difference may be attributable to the higher COD removal than those aforementioned study [11] and the lower concentration of azo compound than those another study [46]. In a typical anaerobic digester, the relative abundance of acetoclastic methanogens is much higher than that of hydrogenotrophic methanogens because of their crucial role in converting acetate (main products of acetogenesis) to methane and CO2. High hydrogen production inhibits acetogenic metabolism, resulting in low methane production and the accumulation of volatile fatty acids. Thus, hydrogenotrophic methanogens play an important role in the stabilization of the low hydrogen partial pressure in anaerobic digesters [41]. In contrast to typical anaerobic digesters, the AnMBR used in this experiment contained both types of methanogens at a high relative abundance, indicating that although the hydrogen production is high, acetogenic metabolism is not inhibited and so the COD removal is high. The reason may be that the produced hydrogen does not accumulate as a partial pressure inside the bioreactor; instead, it is used as an electron donor for azo decolorization as it has been mentioned by a study that hydrogen has been proven to be effective on dye reduction [46]. Additionally, the low
concentration of azo concentration in this research may also be the cause of the high abundance of acetoclastic methanogens.
Fig. 6. (A, C) Relative abundance of genera and (B, D) phylogenetic tree of the microbial community in (A, B) AnMBR and (C, D) DHS.
Degradation of azo dyes begins with the reduction of the azo bond. This reaction is not specific, and some suggested pathways were reviewed [47]. Many bacteria can reportedly reduce azo dyes. However, the high color removal of RB5 in this experiment, especially after the AnMBR process, demonstrates that azo reduction by mixed cultures can be effective. Although most of the major genera in the AnMBR in this experiment have not yet been reported as azo-reducing bacteria, their roles in providing a simple substrate for azo reduction and in COD removal are very important. Tolumonas is reportedly an acidogenic bacterium [48], whereas Syntrophus, Aeromonas, and Smithella are reportedly acetogenic bacteria [11,49,50]. Spirillum reportedly uses a limited number of organic acids employing only oxygen as an electron acceptor, with optimal operation at a 2% oxygen level [51]. The conditions inside the AnMBR are not strictly anaerobic, which might be the reason for the high abundance of Spirillum. Caldisericum and
Desulfomonile might be involved in the sulfur cycle; they can reportedly act as sulfur-reducing bacteria [52,53]. Moreover, several studies have documented that azo dyes are reduced under sulfate-reducing conditions [47], suggesting that sulfur bacteria contribute to azo reduction.
However, there is little information on the mechanism.
The 10 major genera identified in the DHS and the phylogenetic tree of the DHS microorganism community are shown in Fig. 6c and 6d, respectively. The major genera are Janthinobacterium (11%), Amaricoccus (3%), Pseudomonas (3%), Aeromonas (2%), Sulfurospirillum (2%), Rhodobacter (2%), Raoultella (1%), Tolumonas (1%), Desulfoprunum (1%), and Stella (1%). These genera all belong to the phylum Proteobacteria. In textile wastewater treatment, aerobic processes are required for aromatic amine degradation. Simple aromatic amines can be biodegraded under anaerobic conditions; however, sulfonated aromatic amines (which are reportedly azo dye degradation products) are generally biodegraded under aerobic conditions [54]. Some of the sulfonated aromatic amine-degrading bacteria that were reported in other studies and also identified in the DHS used in this experiment were Pseudomonas, Alcaligenes, Hydrogenophaga, Comamonas, Acinetobacter, and Rhodococcus [37,54,55].
As the DHS also provided anaerobic conditions near the center of the sponge, an azo-reducing reaction might occur in the DHS system. The azo-reducing bacteria found in the AnMBR are also found in the DHS, although at a much lower relative abundance. Other major genera identified in the DHS might play an important role in removing organic pollutants from the AnMBR effluent. Janthinobacterium, Amaricoccus, Sulfurospirillum, Rhodobacter, Raoultella, Tolumonas, Desulfoprunum, and Stella can reportedly use a wide range of organic matter in wastewater as a substrate under aerobic and/or microaerophilic conditions [56–63].
Sulfospirillum and Desulfoprunum might also contribute to the sulfur cycle in the DHS system,
as they can reduce sulfur compounds [58,62].
4. Conclusion
The AnMBR-DHS combination was shown to significantly reduce BOD3, COD, and color in synthetic textile wastewater. The AnMBR-DHS process resulted in BOD3, COD, and color removal values of 97.3 ± 1.8%, 94.4 ± 4.8%, and 95.0 ± 1.6%, respectively. Most of the removal occurred in the AnMBR. This AnMBR-DHS system meets Indonesia’s regulatory standards. The performance of the AnMBR-DHS system at HRTs of 12 and 24 h and membrane fluxes of 2.65 and 5.21 LMH did not differ significantly. However, fouling of the
membrane surface occurred more rapidly at a higher membrane flux or shorter HRT. The microbial community in the AnMBR was dominated by the phyla Euryarchaeota, Caldiserica, and Proteobacteria, whereas that in the DHS was dominated by Proteobacteria. Some genera found in the AnMBR can reportedly reduce azo dyes, and some genera found in the DHS can reportedly degrade sulfonated aromatic amines. Further studies of in-situ fouling countermeasures and prevention are needed. In addition, real textile wastewater should be treated using an AnMBR to observe the effect of HRT and flux on the decolorization efficiency in future studies. More scale up studies (design and operation) on textile wastewater are needed for AnMBR-DHS reactors to run optimally. Moreover, further measurements and assessments of the microbial community contributing to textile wastewater treatment are needed.
Acknowledgment
The authors would like to thank the Ministry of Research, Technology and Higher Education of Indonesia (Kementerian Ristek-Dikti) for the scholarship and master-doctoral program (Program Magister-Doktor untuk Sarjana Unggul, PMDSU 2016 – 2018 with the most recent contract number of 127/SP2H/PTNBH/DRPM/2018). The first author thanks to Institut Teknologi Bandung for providing the scholarships (No. 186/SK/I1.B01/KM/2019).
References
[1] U. Nimkar, Sustainable chemistry: A solution to the textile industry in a developing world, Curr. Opin. Green Sustain. Chem. 9 (2018) 13–17.
https://doi.org/10.1016/j.cogsc.2017.11.002.
[2] L. Bilińska, M. Gmurek, S. Ledakowicz, Comparison between industrial and simulated textile wastewater treatment by AOPs – Biodegradability, toxicity and cost assessment, Chem. Eng. J. 306 (2016) 550–559. https://doi.org/10.1016/j.cej.2016.07.100.
[3] M. Berradi, R. Hsissou, M. Khudhair, M. Assouag, O. Cherkaoui, A. El Bachiri, A. El Harfi, Textile finishing dyes and their impact on aquatic environs, Heliyon. 5 (2019) e02711. https://doi.org/10.1016/j.heliyon.2019.e02711.
[4] M. Yusuf, Handbook of Textile Effluent Remediation, Jenny Stanford Publishing, 2018.
https://doi.org/10.1201/9780429505478.
[5] V. Katheresan, J. Kansedo, S.Y. Lau, Efficiency of various recent wastewater dye removal methods: A review, J. Environ. Chem. Eng. 6 (2018) 4676–4697.
https://doi.org/10.1016/j.jece.2018.06.060.
[6] A.K. Verma, P. Bhunia, R.R. Dash, Performance of UASB reactor treating synthetic textile wastewater: effect of physicochemical pretreatment, Desalination Water Treat.
57 (2016) 8050–8060. https://doi.org/10.1080/19443994.2015.1017739.
[7] M. Salgot, M. Folch, Wastewater treatment and water reuse, Curr. Opin. Environ. Sci.
Health. 2 (2018) 64–74. https://doi.org/10.1016/j.coesh.2018.03.005.
[8] KLHK-RI, P.16/MENLHK/SETJEN/KUM.1/4/2019P: Second Amendment to
Regulation of the Minister of Environment Number 5 Year 2014 regarding Waste Water Quality Standard, (2019). https://peraturan.go.id/common/dokumen/bn/2019/bn433- 2019.pdf (accessed December 3, 2020).
[9] C.R. Holkar, A.J. Jadhav, D.V. Pinjari, N.M. Mahamuni, A.B. Pandit, A critical review on textile wastewater treatments: Possible approaches, J. Environ. Manage. 182 (2016) 351–366. https://doi.org/10.1016/j.jenvman.2016.07.090.
[10] D. Bhatia, N.R. Sharma, J. Singh, R.S. Kanwar, Biological methods for textile dye removal from wastewater: A review, Crit. Rev. Environ. Sci. Technol. 47 (2017) 1836–
1876. https://doi.org/10.1080/10643389.2017.1393263.
[11] T.H. Nguyen, T. Watari, M. Hatamoto, D. Sutani, T. Setiadi, T. Yamaguchi, Evaluation of a combined anaerobic baffled reactor–downflow hanging sponge biosystem for treatment of synthetic dyeing wastewater, Environ. Technol. Innov. 19 (2020) 100913.
https://doi.org/10.1016/j.eti.2020.100913.
[12] Y.J. Chan, M.F. Chong, C.L. Law, D.G. Hassell, A review on anaerobic–aerobic treatment of industrial and municipal wastewater, Chem. Eng. J. 155 (2009) 1–18.
https://doi.org/10.1016/j.cej.2009.06.041.
[13] F.P. van der Zee, S. Villaverde, Combined anaerobic–aerobic treatment of azo dyes—A short review of bioreactor studies, Water Res. 39 (2005) 1425–1440.
https://doi.org/10.1016/j.watres.2005.03.007.
[14] W.-E. Thung, S.-A. Ong, L.-N. Ho, Y.-S. Wong, F. Ridwan, H.K. Lehl, Y.-L. Oon, Y.- S. Oon, Biodegradation of Acid Orange 7 in a combined anaerobic-aerobic up-flow membrane-less microbial fuel cell: Mechanism of biodegradation and electron transfer, Chem. Eng. J. 336 (2018) 397–405. https://doi.org/10.1016/j.cej.2017.12.028.
[15] W. Ben Mbarek, E. Pineda, L. Escoda, J.J. Suñol, M. Khitouni, High efficiency
decolorization of azo dye Reactive Black 5 by Ca-Al particles, J. Environ. Chem. Eng. 5 (2017) 6107–6113. https://doi.org/10.1016/j.jece.2017.11.032.
[16] Q. Liao, J. Chang, C. Herrmann, A. Xia, eds., Bioreactors for Microbial Biomass and Energy Conversion, Springer Singapore, 2018. https://doi.org/10.1007/978-981-10- 7677-0.
[17] B.E.L. Baêta, D.R.S. Lima, S.Q. Silva, S.F. Aquino, B.E.L. Baêta, D.R.S. Lima, S.Q.
Silva, S.F. Aquino, Influence of the applied organic load (OLR) on textile wastewater treatment using submerged anaerobic membrane bioreactors (SAMBR) in the presence of redox mediator and powdered activated carbon (PAC), Braz. J. Chem. Eng. 33 (2016) 817–825. https://doi.org/10.1590/0104-6632.20160334s20150031.
[18] A. Yurtsever, E. Sahinkaya, Ö. Aktaş, D. Uçar, Ö. Çınar, Z. Wang, Performances of anaerobic and aerobic membrane bioreactors for the treatment of synthetic textile wastewater, Bioresour. Technol. 192 (2015) 564–573.
https://doi.org/10.1016/j.biortech.2015.06.024.
[19] T. Watari, T. Cuong Mai, D. Tanikawa, Y. Hirakata, M. Hatamoto, K. Syutsubo, M.
Fukuda, N.B. Nguyen, T. Yamaguchi, Development of downflow hanging sponge (DHS) reactor as post treatment of existing combined anaerobic tank treating natural rubber processing wastewater, Water Sci. Technol. 75 (2017) 57–68.
https://doi.org/10.2166/wst.2016.487.
[20] T. Okubo, T. Onodera, S. Uemura, T. Yamaguchi, A. Ohashi, H. Harada, On-site evaluation of the performance of a full-scale down-flow hanging sponge reactor as a post-treatment process of an up-flow anaerobic sludge blanket reactor for treating sewage in India, Bioresour. Technol. 194 (2015) 156–164.
https://doi.org/10.1016/j.biortech.2015.07.015.
[21] S. Uemura, S. Suzuki, K. Abe, K. Kubota, T. Yamaguchi, A. Ohashi, Y. Takemura, H.
Harada, Removal of organic substances and oxidation of ammonium nitrogen by a down-flow hanging sponge (DHS) reactor under high salinity conditions, Bioresour.
Technol. 101 (2010) 5180–5185. https://doi.org/10.1016/j.biortech.2010.02.040.
[22] R.Z. Gaur, M. Owaes, T.C. D’silva, M.N. Hasan, P.K. Mutiyar, A.A. Khan, V.K. Tyagi, Micro-aerobic technologies for UASB effluent polishing: An overview, (2019).
https://doi.org/10.2166/9781780409740_0057.
[23] A. Nurmiyanto, A. Ohashi, Downflow Hanging Sponge (DHS) Reactor for Wastewater Treatment - A Short Review, MATEC Web Conf. 280 (2019) 05004.
https://doi.org/10.1051/matecconf/201928005004.
[24] M. Hatamoto, T. Okubo, K. Kubota, T. Yamaguchi, Characterization of downflow hanging sponge reactors with regard to structure, process function, and microbial community compositions, Appl. Microbiol. Biotechnol. 102 (2018) 10345–10352.
https://doi.org/10.1007/s00253-018-9406-6.
[25] T. Watari, Y. Hata, Y. Hirakata, P.N. Nguyet, T.H. Nguyen, S. Maki, M. Hatamoto, D.
Sutani, T. Setia, T. Yamaguch, Performance evaluation of down-flow hanging sponge reactor for direct treatment of actual textile wastewater; Effect of effluent recirculation to performance and microbial community, J. Water Process Eng. (2020) 101724.
https://doi.org/10.1016/j.jwpe.2020.101724.
[26] N.D. Lourenço, R.D.G. Franca, M.A. Moreira, F.N. Gil, C.A. Viegas, H.M. Pinheiro, Comparing aerobic granular sludge and flocculent sequencing batch reactor
technologies for textile wastewater treatment, Biochem. Eng. J. 104 (2015) 57–63.
https://doi.org/10.1016/j.bej.2015.04.025.
[27] A. Tawfik, D.F. Zaki, M.K. Zahran, Degradation of reactive dyes wastewater supplemented with cationic polymer (Organo Pol.) in a down flow hanging sponge (DHS) system, J. Ind. Eng. Chem. 20 (2014) 2059–2065.
https://doi.org/10.1016/j.jiec.2013.09.031.
[28] M.P. de Matos, A.C. Borges, A.T. de Matos, E.F. da Silva, M.A. Martinez, Effect of time-temperature binomial in obtaining biochemical oxygen demand of different wastewaters, Eng. Agríc. 34 (2014) 332–340. https://doi.org/10.1590/S0100- 69162014000200014.
[29] American Public Health Association, American Water Works Association, Water Environment Federation, AWWA, 1995. Standard Methods for the Examination of Water and Wastewater, Amer Pub Health Assoc. Wash. DC. (1998).
[30] R.L. Singh, P.K. Singh, R.P. Singh, Enzymatic decolorization and degradation of azo dyes – A review, Int. Biodeterior. Biodegrad. 104 (2015) 21–31.
https://doi.org/10.1016/j.ibiod.2015.04.027.
[31] A.E. Tret’yakova, V.V. Safonov, A.Yu. Yusina, Study of processes of dyeing polyurethane fibres by various classes of dyes, Fibre Chem. 44 (2013) 284–287.
https://doi.org/10.1007/s10692-013-9447-7.
[32] M.C. Tomei, D. Mosca Angelucci, A.J. Daugulis, Sequential anaerobic-aerobic decolourization of a real textile wastewater in a two-phase partitioning bioreactor, Sci.
Total Environ. 573 (2016) 585–593. https://doi.org/10.1016/j.scitotenv.2016.08.140.
[33] A. Assadi, M. Naderi, M.R. Mehrasbi, Anaerobic–aerobic sequencing batch reactor treating azo dye containing wastewater: effect of high nitrate ions and salt, J. Water Reuse Desalination. 8 (2018) 251–261. https://doi.org/10.2166/wrd.2017.132.
[34] F. Ghanbari, M. Moradi, A comparative study of electrocoagulation, electrochemical Fenton, electro-Fenton and peroxi-coagulation for decolorization of real textile wastewater: Electrical energy consumption and biodegradability improvement, J.
Environ. Chem. Eng. 3 (2015) 499–506. https://doi.org/10.1016/j.jece.2014.12.018.
[35] M. Bahia, F. Passos, O.F.H. Adarme, S.F. Aquino, S.Q. Silva, Anaerobic-Aerobic Combined System for the Biological Treatment of Azo Dye Solution using Residual Yeast, Water Environ. Res. 90 (2018) 729–737.
https://doi.org/10.2175/106143017x15131012153167.
[36] S. Pradhan, P. Kumar, I. Mehrotra, River pollution: Assessment of hydro-philic and phobic nature of persistent organic contaminants, Environ. Nanotechnol. Monit. Manag.
3 (2015) 47–54. https://doi.org/10.1016/j.enmm.2014.12.002.
[37] K.M. Kodam, Y.M. Kolekar, Bacterial Degradation of Textile Dyes, in: S.N. Singh (Ed.), Microb. Degrad. Synth. Dyes Wastewaters, Springer International Publishing, Cham, 2015: pp. 243–266. https://doi.org/10.1007/978-3-319-10942-8_11.
[38] T. Onodera, T. Okubo, S. Uemura, T. Yamaguchi, A. Ohashi, H. Harada, Long-term performance evaluation of down-flow hanging sponge reactor regarding nitrification in a full-scale experiment in India, Bioresour. Technol. 204 (2016) 177–184.
https://doi.org/10.1016/j.biortech.2016.01.005.
[39] O.T. Iorhemen, R.A. Hamza, J.H. Tay, Membrane Bioreactor (MBR) Technology for Wastewater Treatment and Reclamation: Membrane Fouling, Membranes. 6 (2016) 33.
https://doi.org/10.3390/membranes6020033.
[40] J. Lee, J.W. Lee, Y.M. Kim, C. Park, K.Y. Park, Performance and Fouling in Pre- Denitrification Membrane Bioreactors Treating High-Strength Wastewater from Food Waste Disposers, Water. 9 (2017) 512. https://doi.org/10.3390/w9070512.
[41] L.N. Nguyen, A.Q. Nguyen, L.D. Nghiem, Microbial Community in Anaerobic
Digestion System: Progression in Microbial Ecology, in: X.-T. Bui, C. Chiemchaisri, T.
Fujioka, S. Varjani (Eds.), Water Wastewater Treat. Technol., Springer, Singapore, 2019: pp. 331–355. https://doi.org/10.1007/978-981-13-3259-3_15.
[42] B.-R. Kim, J. Shin, R.B. Guevarra, J.H. Lee, D.W. Kim, K.-H. Seol, J.-H. Lee, H.B.K.
and R.E. Isaacson, Deciphering Diversity Indices for a Better Understanding of
Microbial Communities, 27 (2017) 2089–2093. https://doi.org/10.4014/jmb.1709.09027.
[43] C. Choerudin, A. Harimawan, T. Setiadi, The Comparison of Different Nutrient Sources in Development of Aerobic Granular Sludge, MATEC Web Conf. 156 (2018) 03034.
https://doi.org/10.1051/matecconf/201815603034.
[44] S. Yang, H.V. Phan, H. Bustamante, W. Guo, H.H. Ngo, L.D. Nghiem, Effects of shearing on biogas production and microbial community structure during anaerobic digestion with recuperative thickening, Bioresour. Technol. 234 (2017) 439–447.
https://doi.org/10.1016/j.biortech.2017.03.051.
[45] L. Li, Q. He, Y. Ma, X. Wang, X. Peng, A mesophilic anaerobic digester for treating food waste: process stability and microbial community analysis using pyrosequencing, Microb. Cell Factories. 15 (2016) 65. https://doi.org/10.1186/s12934-016-0466-y.
[46] R. Dai, X. Chen, Y. Luo, P. Ma, S. Ni, X. Xiang, G. Li, Inhibitory effect and
mechanism of azo dyes on anaerobic methanogenic wastewater treatment: Can redox mediator remediate the inhibition?, Water Res. 104 (2016) 408–417.
https://doi.org/10.1016/j.watres.2016.08.046.
[47] F.J. Cervantes, A.B.D. Santos, Reduction of azo dyes by anaerobic bacteria:
Microbiological and biochemical aspects, Rev. Environ. Sci. Biotechnol. 10 (2011) 125–137. https://doi.org/10.1007/s11157-011-9228-9.
[48] X. Cao, H. Wang, S. Zhang, O. Nishimura, X. Li, Azo dye degradation pathway and bacterial community structure in biofilm electrode reactors, Chemosphere. 208 (2018) 219–225. https://doi.org/10.1016/j.chemosphere.2018.05.190.
[49] A. Martin‐Carnahan, S.W. Joseph, Aeromonas, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2015: pp. 1–44.
https://doi.org/10.1002/9781118960608.gbm01081.
[50] A. Galushko, J. Kuever, Smithella, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2019: pp. 1–4.
https://doi.org/10.1002/9781118960608.gbm01063.pub2.
[51] S.I. Abou-Elela, G. Golinielli, E.M. Abou-Taleb, M.S. Hellal, Municipal wastewater treatment in horizontal and vertical flows constructed wetlands, Ecol. Eng. 61 (2013) 460–468. https://doi.org/10.1016/j.ecoleng.2013.10.010.
[52] K. Mori, Caldisericum, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2018: pp. 1–4. https://doi.org/10.1002/9781118960608.gbm01477.
[53] A. Galushko, J. Kuever, Desulfomonile, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2019: pp. 1–7.
https://doi.org/10.1002/9781118960608.gbm01062.pub2.
[54] L. Pereira, P.K. Mondal, M. Alves, Aromatic Amines Sources, Environmental Impact and Remediation, in: E. Lichtfouse, J. Schwarzbauer, D. Robert (Eds.), Pollut. Build.
Water Living Org., Springer International Publishing, Cham, 2015: pp. 297–346.
https://doi.org/10.1007/978-3-319-19276-5_7.
[55] A. Pandey, P. Singh, L. Iyengar, Bacterial decolorization and degradation of azo dyes, Int. Biodeterior. Biodegrad. 59 (2007) 73–84.
https://doi.org/10.1016/j.ibiod.2006.08.006.
[56] M. Gillis, N.A. Logan, Janthinobacterium, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2015: pp. 1–12.
https://doi.org/10.1002/9781118960608.gbm00964.
[57] D. Tanikawa, S. Yamashita, T. Kataoka, H. Sonaka, Y. Hirakata, M. Hatamoto, T.
Yamaguchi, Non-aerated single-stage nitrogen removal using a down-flow hanging sponge reactor as post-treatment for nitrogen-rich wastewater treatment, Chemosphere.
233 (2019) 645–651. https://doi.org/10.1016/j.chemosphere.2019.06.012.
[58] J.F. Stolz, R.S. Oremland, B.J. Paster’, F.E. Dewhirst, P. Vandamme, Sulfurospirillum, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2015: pp. 1–7.
https://doi.org/10.1002/9781118960608.gbm01072.
[59] J.F. Imhoff, Rhodobacter, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2015: pp. 1–12. https://doi.org/10.1002/9781118960608.gbm00862.
[60] M. Drancourt, C. Bollet, A. Carta, P. Rousselier, Phylogenetic analyses of Klebsiella species delineate Klebsiella and Raoultella gen. nov., with description of Raoultella ornithinolytica comb. nov., Raoultella terrigena comb. nov. and Raoultella planticola comb. nov., Int. J. Syst. Evol. Microbiol. 51 (2001) 925–932.
https://doi.org/10.1099/00207713-51-3-925.
[61] X. Wang, Z. Chen, J. Kang, X. Zhao, J. Shen, L. Yang, The key role of inoculated sludge in fast start-up of sequencing batch reactor for the domestication of aerobic granular sludge, J. Environ. Sci. 78 (2019) 127–136.
https://doi.org/10.1016/j.jes.2018.08.008.
[62] A. Galushko, J. Kuever, Desulfoprunum, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2019: pp. 1–3.
https://doi.org/10.1002/9781118960608.gbm01593.
[63] L.V. Vasilyeva, Stella, in: Bergeys Man. Syst. Archaea Bact., American Cancer Society, 2015: pp. 1–5. https://doi.org/10.1002/9781118960608.gbm00889.
