English

• Chinese medicine residues (CMRs) are underutilized resources, which are abundant in polysaccharides, saponins, flavonoids and other bioactive compounds [1-3]. Compared to traditional crude and low-value resource utilization, the model of employing chemical separation technology or biotechnology to transform inexpensive waste into high-value-added substances has attracted significant attention [4-6]. Despite the necessity of dehydration pre-treatment as the initial step during high-value resource utilization of CMRs, the management of generated wastewater is frequently neglected. This wastewater differs from conventional traditional Chinese medicine wastewater due to its complex composition, high concentration of organics, and poor biochemistry characteristics. Nevertheless, research on the treatment of this type of wastewater is currently lacking. Therefore, it is crucial to investigate an efficient and environmentally sustainable technology for treating CMR dehydration wastewater to ensure the sustainable utilization of high-value CMRs.

  Biological methods are widely regarded as the most promising approach for the treatment of organic wastewater [7]. Previous research has demonstrated that anaerobic treatment technologies, such as expanded granular sludge bed (EGSB) reactors, play a central role in the treatment of Chinese medicine wastewater [8]. However, anaerobic systems are susceptible to changes in substrate conditions and suffer from drawbacks, such as low hydrolysis efficiency and poor stability [9]. The intricate composition and elevated concentration of organics may lead to the accumulation of organic acids within anaerobic system, causing acidification, inhibition of anaerobe growth, and disruption of the stable operation. In recent years, microaeration has been recognized as an effective strategy for enhancing the stability of anaerobic systems by expediting hydrolysis and preventing excessive accumulation of volatile fatty acids (VFAs) [10,11]. It has been reported that microaeration creates a specific environment characterized by the overlap of anaerobic and aerobic conditions, which offers ecological niches for facultative anaerobes while preserving survival space for anaerobes [12]. The presence of oxygen enhances the growth of acid-producing bacteria and stimulates the release of extracellular enzymes, thereby expediting the process of hydrolysis [13,14]. Hence, for the CMR dehydration wastewater with complex composition and high concentration of organics, the addition of microaeration to anaerobic system is an effective measure to improve capacity and stability. Based on the aforementioned research, a microaerobic EGSB (Mi-EGSB) was employed for preliminary treatment of CMR dehydration wastewater to establish a foundation for subsequent treatment.

  To ensure compliance with discharge standards, additional aerobic biological treatment is still necessary. In recent years, moving bed sequencing batch reactor (MBSBR) has gained favor as a new type of aerobic reactor [15]. Limited dissolved oxygen (DO) mass transfer within the carrier creates a gradient, providing attachment sites for slow-growing microorganisms like denitrifying bacteria [16]. These microorganisms form biofilms on the surface of suspended carriers [17], allowing simultaneous nitrification and denitrification (SND) in a single reactor [18]. Furthermore, the sequential batch operation enables dynamic changes in DO levels over time. The temporal and spatial gradient of DO empowers the MBSBR to exhibit exceptional efficacy in the removal of nitrogen and phosphorus. However, current researches are limited to synthetic wastewater, and further research is needed to investigate its treatment effect on actual wastewater.

  In this study, a novel microaerobic-aerobic combined process is proposed for the dehydration wastewater from CMRs. The Mi-EGSB reactor effectively removed majority of chemical oxygen demand (COD). Microaeration facilitated hydrolysis and acidification process of macromolecular organics, thereby enhancing the biochemical characteristics of wastewater. Furthermore, the distribution of microbial communities in Mi-EGSB was analyzed to investigate the impact of microaeration on microbial diversity and community structure. Finally, operational efficiency of MBSBR was compared under different DO conditions, ultimately achieving advanced removal of organics as well as nitrogen and phosphorus pollutants. This study offers theoretical support and technical guidance for the treatment of CMR dehydration wastewater, ensuring the sustainable utilization of CMR resources.

  The EGSB reactor has a height of 1000 mm, an inner diameter of 80 mm, and an effective volume of 5 L. The MBSBR is constructed from a column with an inner diameter of 11 cm, a height of 40 cm, and an effective volume of 3.5 L. Cubic modified polyurethane foam with a side length of 1 cm was used as the carrier. To investigate the impact of DO on operation performance, two MBSBRs with different DO concentrations were established.

  The characteristics of dehydration wastewater from licorice residues are presented in Table 1. The inoculated sludge utilized in this study originated from the secondary sedimentation tank of Harbin Wenchang Wastewater Plant and was added into the reactor after static settling and impurity removal. The mixed liquor volatile suspended solid (MLVSS) of the EGSB inoculated sludge was 21.655 g/L, with an inoculation volume of 50%. Powdered activated carbon was simultaneously added as nucleation carrier for granular sludge at a ratio of 0.5 g PAC/g VSS to expedite the granularization of sludge. The mixed liquor suspended solid (MLSS) of sludge inoculated with MBSBR was 23.558 mg/L, with an inoculum volume of 30% (v/v).

  Table 1

  

  Table 1.  Characteristics of the wastewater.

  DownLoad: CSV

  COD (mg/L)	pH	NH4+-N (mg/L)	TN (mg/L)	TP (mg/L)	UV254
  28,000-34,000	5–6	530	560	80	96–98

  The design parameters for the operational phase of the EGSB reactor are presented in Table 2. NaHCO₃ was added for pH adjustment of the influent, while trace elements were supplemented to meet the nutritional requirements of microorganisms. The reactor was started under anaerobic conditions, and granular sludge gradually formed. After startup, the EGSB was adjusted to microaerobic condition through aeration in the reflux section, and the DO at the top outlet was maintained at 0.1 mg/L. The organic loading rate (OLR) was gradually increased to facilitate the domestication of the sludge, and the impact of hydraulic retention time (HRT) and DO on the removal performance were also examined.

  Table 2

  

  Table 2.  Parameters of Mi-EGSB operating process.

  DownLoad: CSV

  Phase	Time (d)	COD (mg/L)	HRT (h)	OLR (kg COD m−3 d−1)	DO (mg/L)
  Ⅰ	0-49	2000	24	2	/
  Ⅱ	50-76	2000	24	2	0.1
  Ⅲ	77-114	3000	24	3	0.1
  Ⅳ	115-138	4000	24	4	0.1
  Ⅴ	139-151	5000	24	5	0.1
  Ⅵ	152-190	5000	18	6.7	0.1
  Ⅶ	191-220	5000	18	6.7	0.2

  The HRT of the MBSBR was set to 10 h: instantaneous FILL, 0 h; anoxic REACT, 3 h; aeration REACT, 6 h; SETTLE, 1 h; and instantaneous DRAW, 0 h. During phase Ⅰ, EGSB effluent was diluted and used for domesticating the inoculated sludge. In phase Ⅱ, effluent from the EGSB without dilution was introduced into the MBSBR, along with approximately 25% polyurethane fillers added. During the initial two phases, DO concentrations in the aeration phase were maintained within the range of 2–4 mg/L. To investigate the impact of aeration intensity on removal performance, DO concentrations of the two MBSBRs were adjusted to 4 mg/L (MBSBR-1) and 2 mg/L (MBSBR-2) during phase Ⅲ. The daily sludge discharge was approximately 200 mL, indicating that SRT was maintained at 17.5 days.

  DO was determined using a portable dissolved oxygen meter (HQ30d, HACH, America) and pH was determined with a FE20-FiveEasy pH meter. Based on the potassium dichromate colourimetric method, COD was analyzed using a digester (DRB200, HACH, America) and spectrophotometer (DR1010, HACH, America). The concentrations of ammonia (NH₄⁺), nitrite (NO₂⁻), nitrate (NO₃⁻) and total phosphorus (TP) were determined by a UV–visible spectrophotometer (UV-2600, Shimadzu, Japan) according to standard methods. Specifically, ammonia nitrogen was specifically analyzed using Nessler's spectrophotometry, nitrite was determined via N-1-naphthylethylenediamine spectrophotometry, nitrate was measured using phenol disulfonic acid spectrophotometry, and total phosphorus was quantified using ammonium molybdate spectrophotometry. SND was determined as follows:

  $ \mathrm{SND}=\left(1-\frac{\Delta \mathrm{NO}_2^{-}+\Delta \mathrm{NO}_3^{-}}{\Delta \mathrm{NH}_4^{+}}\right) \times 100 \% $

  (1)

  where $ \Delta \mathrm{NO}_2^{-}, \Delta \mathrm{NO}_3^{-} \text {and } \Delta \mathrm{NH}_4^{+}$ represent the changes in concentration of NO₂⁻, NO₃⁻, and NH₄⁺, respectively, during the aeration phase. Total nitrogen (TN) was determined by AnalytikJena Multi N/C 3100 TOC Analyzer. The composition of VFAs were detected by gas chromatography (Agilent 7890A GC system). Typical pollutants liquiritin and glycyrrhizic acid were detected by ultra performance liquid chromatography (Waters ACQUITY). The gases generated by EGSB were collected through gas bags and their composition was analyzed with gas chromatography (Agilent 7890A GC system).

  Sludge samples from each stage were collected, centrifuged at 10,000 rpm for 10 min, and the supernatant was decanted and stored in a −80 ℃ refrigerator. The microbial community structure was subsequently identified through high-throughput sequencing. PCR amplification of the Ⅴ3~4 region of bacterial 16SrRNA gene was performed using universal primers 515F (5′-GTGYCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACNVGGGTWTCTAAT-3′). The sequencing based on Illumina MiSeq PE250 platform was completed by Majorbio Bio-pharm Technology Co., Ltd. (Shanghai, China). For operational taxonomic unit (OUT) and diversity analysis, the optimal clustering similarity was selected to be 97%.

  As shown in Fig. 1a, in phase Ⅰ, the COD removal efficiency gradually increased during the start-up process under anaerobic conditions. The pH of the effluent is basically higher than that of the influent (Fig. 1b). Complex organic compounds were degraded to simple soluble monomers by extracellular hydrolases of hydrolytic bacteria. These monomers were then further degraded to small-molecule VFAs, primarily acetic acid and propionic acid (Fig. 1c). Additionally, while the overall concentration of VFAs decreased, the proportion of acetic acid gradually increased, providing suitable substrate for methanogenic bacteria. By day 49, methane production and gas production rate reached 1.628 L CH₄/d and 0.202 L CH₄/g COD_r, respectively (Fig. 1d).

  Figure 1

  

  Figure 1.  (a) COD removal efficiencies of Mi-EGSB during each phase. (b) pH of influent and effluent of Mi-EGSB. (c) Composition of VFAs in the effluent of Mi-EGSB. (d) Methane production rate and methane yield of Mi-EGSB. (e) NH₄⁺-N and TN removal efficiencies of Mi-EGSB. (f) TP removal efficiencies of Mi-EGSB during Ⅵ and Ⅶ phases.

  DownLoad: Full-Size Img PowerPoint

  From phase Ⅱ to phase Ⅴ, with OLR increasing from 2 to 5 kg COD m⁻³ d⁻¹, the average COD removal rate increased from 78.87% to 92.13% under microaeration conditions. This demonstrates that microorganisms had gradually adapted to the wastewater environment. It has been reported that microaeration creates ecological niches for facultative anaerobes [13]. Meanwhile, facultative microorganisms retain strict anaerobes by maintaining low redox potential and providing increased growth factors [12,19]. The presence of oxygen was found to increase the production rate and number of facultative acidogens, resulting in the secretion of a large number of excretory enzymes that accelerate the conversion of organic compounds to VFAs [20]. In phase Ⅴ, the average concentration of VFAs in the effluent was 57.73 mg/L, with an average acetic acid concentration of 46.49 mg/L. The sustained low concentration of VFAs in the effluent indicates that the system exhibits a degree of resistance to shock loading. Furthermore, in phase Ⅳ, the average removal of NH₄⁺-N and TN reached 85.09% and 83.63%, respectively. This may be attributed to the unique environment created by microaeration with overlapping anaerobic and aerobic conditions, which promotes the SND process [21].

  In phase Ⅵ, HRT was reduced to 18 h, resulting in an OLR increase to 6.7 kg COD m⁻³ d⁻¹. The average concentrations of acetic acid and propionic acid in the effluent were 163.50 mg/L and 24.79 mg/L, respectively. The elevated organic load exceeded the metabolic capacity of the microorganisms, leading to a significant accumulation of VFAs. Consequently, the average COD removal rate decreased to 83.54%. The average removal rate of NH₄⁺-N and TN also decreased to 67.21% and 62.12% (Figs. 1e and f), respectively. However, the pH of the effluent remained higher than that of the influent, indicating stable resistance to shock loading [22]. When DO was increased to 0.2 mg/L during phase Ⅶ, a decrease in VFAs concentration was observed, and COD removal rebounded to 87.04%. This may be due to the further enrichment of facultative microbes and the facilitation of extracellular enzyme release. Nitrogen removal was enhanced, and the average total phosphorus (TP) removal efficiency also increased from 39.8% to 67.14%. Additionally, methanogens gradually adapted to the microaerobic environment and recovered their activity. Increasing aeration intensity appropriately can effectively enhance methane production.

  In conclusion, micro-aeration solves the problem of acidification and subsequent operational failure in anaerobic systems under shock loading conditions. Furthermore, it enhances the removal efficiency of EGSB for organic, nitrogen and phosphorus pollutants while reducing the loading for subsequent aerobic unit.

  After the addition of carriers in phase Ⅱ, two MBSBRs were operated for six days without sludge discharge. The removal efficiency of COD is maintained at around 80% (Fig. 2a). The surface of carries gradually became covered with a thin film, and the removal rates of NH₄⁺-N and TN rose to 80% and 70% (Figs. 2b and c). The results indicated that the biofilm provided living space for both facultative and anaerobic microorganisms including denitrifying bacteria, and SND was achieved in MBSBRs [15]. The effectiveness of TP removal fluctuates consistently (Fig. 2d).

  Figure 2

  

  Figure 2.  (a) COD, (b) NH₄⁺-N, (c) TN and (d) TP removal efficiencies at different phases of MBSBR operated under different DO conditions. (e, f) Variations in the concentrations of nitrogen, COD and PO₄³⁻ during a cycle under different DO conditions.

  DownLoad: Full-Size Img PowerPoint

  In phase Ⅲ, DO concentrations in MBSBR-1 and MBSBR-2 were adjusted to 4 and 2 mg/L, respectively, during the aeration phase. With the SND system, COD was consumed not only by aerobic heterotrophs bacteria and other aerobes, but also by nitrifying bacteria during denitrification [15,16]. The average COD removal rates were 77.41% and 78.21%, respectively, with DO concentration having minimal impact. Notably, DO significantly affected nitrogen removal. The high activity of nitrifying bacteria (NOB) accelerated the oxidation of NH₄⁺-N under higher DO conditions (Fig. 2e) [23], resulting in higher removal efficiencies in MBSBR-1 (83.70%) compared to MBSBR-2 (80.37%). However, the accumulation of NO₃⁻-N indicated that NH₄⁺-N was not completely converted to gaseous nitrogen, but accumulated as an intermediate product of nitrate. In the single-cycle experiments, the average NO₃⁻-N concentrations at the end of aeration were 4.59 and 2.45 mg/L, with SND rates of 25.57% and 63.75%, respectively (Figs. 2e and f). This phenomenon can be attributed to the influence of DO concentration on biofilm structure and profile. Excessive aeration intensity resulted in increased oxygen penetration depth, thereby inhibiting the activity of denitrifying bacteria and the expression of essential enzyme [24]. Consequently, the electron acceptor NO₃⁻-N involved in denitrification is unable to reduction and therefore accumulated, ultimately leading to diminished efficiency of total nitrogen removal.

  MBSBR-2 (93.48%) exhibited a higher TP removal efficiency compared to MBSBR-1 (90.68%). TP concentration of MBSBR-2 effluent consistently remained below 0.5 mg/L. Additionally, phosphorus absorption to release ratios were 1.11 (MBSBR-1) and 1.17 (MBSBR-2), respectively (Figs. 2g and h), with higher concentrations of absorbed and released phosphorus observed in MBSBR-2. Considering low concentration of NO₃⁻-N in influent and an equivalent level of organics, these factors did not significantly influence the release of inorganic phosphorus and the synthesis of poly-β-hydroxybutyrate (PHB). This phenomenon may be attributed to the increased abundance of polyphosphate-accumulating organisms (PAOs) in MBSBR-2, which facilitating greater uptake of organics in anaerobic phase for PHB synthesis. Subsequently, PHB can be degraded under aerobic conditions to provide more energy for phosphorus uptake. In conclusion, the removal efficiency of MBSBR-2 surpassed that of MBSBR-1 for both nitrogen and phosphorus pollutants, indicating that controlling DO at 2 mg/L during aeration phase contributed to achieving higher removal rates while conserving energy consumption.

  In summary, the microaerobic EGSB and MBSBR combined process can effectively treat dehydration wastewater from CMRs, achieving high removal rates of 98.25% for COD, 90.49% for TN, and 98.55% for TP. The effluent met the discharge standard of the "Discharge Standard of Water Pollutants for Pharmaceutical Industry Chinese Traditional Medicine Category" (GB21906–2008).

  The dehydration wastewater from licorice residues used in this study contains typical components such as liquorice flavonoids, saponins, and polysaccharides. Specifically, liquiritin (LQ) belongs to dihydroflavonoids, while glycyrrhizic acid (GA) is classified as a saponin. The complex structures of both compounds have been discovered to inhibit microbial activity [25]. Therefore, it is crucial to investigate the longitudinal degradation trends of these two typical pollutants. As shown in the Fig. 3, the average influent concentrations of LQ and GA were 94.46 mg/L and 98.87 mg/L, respectively. The Mi-EGSB achieved removal rates of 80.74% for LQ and 94.07% for GA, with GA being nearly undetectable in the Mi-EGSB effluent during the stabilization phase. Furthermore, sequential batch experiments revealed hydrolysis rates of 16.743 and 18.732 mg L⁻¹ h⁻¹ for LQ and GA by microaeration sludge, respectively, indicating rapid degradation of GA when relevant hydrolases were expressed by microorganisms (Fig. S1 in Supporting information). It is noteworthy that GA, as a plant-based natural surfactant, has the potential to generate bubbles during reflux aeration and reduce the efficiency of oxygen dissolution [26]. However, oxygen mass transfer was facilitated by efficient degradation in the Mi-EGSB. Both components were undetectable in MBSBR effluent, indicating the effective degradation of typical pollutants through the combined process.

  Figure 3

  

  Figure 3.  Degradation trends of typical pollutants liquiritin and glycyrrhizic acid.

  DownLoad: Full-Size Img PowerPoint

  Shannon and Simpson indices are indicators for characterizing the richness and evenness of microbial communities, while ACE and Chao are frequently utilized to evaluate the abundance of microbial species [27]. During the startup phase, aerobes and certain facultative anaerobes were not suited for anaerobic environment and were consequently eliminated, leading to lower microbial diversity and abundance, as shown in Fig. 4a. After the addition of microaeration, OLR remained constant. The Shannon and Simpson indices changed to 3.75 and 0.06, while the ACE and Chao indices increased to be 635.91 and 635.6, respectively. During phase Ⅲ, despite the increase in OLR to 3 kg COD m⁻³ d⁻¹, aforementioned indicators continued to exhibit a consistent trend. Under the microaerobic condition, a DO gradient was formed within granular sludge, providing simultaneous survival space for facultative and anaerobic bacteria from the outer to the inner layers. The activities of facultative anaerobes were promoted while preserving the ecological niches of anaerobes [12,28]. Therefore, anaerobic digestion system exhibited a higher abundance and diversity of microbial communities in microaerobic environment. However, with a further increase in OLR to 5 kg COD m⁻³ d⁻¹, there was a corresponding rise in the presence of toxic substances such as aromatic compounds in the influent. Shannon and ACE indices decreased to 3.58 and 638.04, respectively, indicating the gradual elimination of microorganisms unable to produce enzymes to degrade the corresponding pollutants. In contrast, microbes adapted to the wastewater were enriched and became the dominant bacteria.

  Figure 4

  

  Figure 4.  (a) Changes in microbial community diversity and richness in Mi-EGSB during different phase. (b) The relative abundance changes of microbial community at phylum level in different phases of operation in Mi-EGSB. (c) Heatmap of microbial community at genus level with average abundance ranking in the top 20 in Mi-EGSB.

  DownLoad: Full-Size Img PowerPoint

  The relative abundance of sludge samples at phylum level is shown in Fig. 4b. The dominant phylum within Mi-EGSB were Bacteroidota, Synergistota and Firmicutes, all of which are essential phyla bacterial involved in hydrolytic acidification [29,30]. Miroaeration changed the microbial community characteristics. From phase Ⅰ to phase Ⅴ, the abundance of Bacteroidota increased from 20.00% to 59.65%. Bacteroidota have been reported to secrete a variety of extracellular enzymes to accelerate the conversion of complex organics into short-chain volatile acids, which play an important role in the degradation of proteins and polysaccharides [31,32]. They also secrete EPS, facilitating microbial aggregation, particle formation and structural stability. This phylum has the capability to expedite the hydrolytic acidification process in a microaerobic environment. Firmicutes produce extracellular enzymes to degrade organic matter such as cellulose, proteins, and lipids [31,33], with abundance increasing from 9.03% to 22.92% after microaeration. Proteobacteria, the phylum containing nitrogen and phosphorus removing microorganisms [34-36], were inhibited during anaerobic start-up. The abundance increased to 7.72% when microaeration was introduced in phase Ⅱ, explaining the rising NH₄⁺-N and TN removal. Notably, the relative abundance of Cloacimonadota increased from nearly zero to 12.42% after microaeration, and the abundance of Spirochaetota increased abruptly to 11.87% in phase Ⅲ. It was shown that some genera in these two phyla were able to be able to degrade long-chain fatty acids [37], and Spirochaetota was also able to use carbohydrate fermentation to produce VFAs [38], which accelerated the hydrolysis process. Verrucomicrobiota is also a phylum capable of degrading organics [39], reaching an abundance of 11.16% in phase Ⅱ and 6.76% in phase Ⅳ. Members of Chloroflexi are facultative anaerobes capable of participating in the degradation of polysaccharides and monosaccharides. They were inhibited during anaerobic start-up reducing the abundance to 0.27%. By phase Ⅳ, it increased to 3.29%, indicating that the microaerobic environment provided survival space for facultative anaerobes and enhanced microbial diversity and abundance.

  To further explore the impact of microaeration on microbial community, the top 20 genera in relative abundance at genus level were categorized based on their functional characteristics as shown in Fig. 4c. Sixteen of these genera are associated with the process of hydrolytic acidification. The highest abundance was observed in genus Bacteroides of the phylum Bacteroidota, which is known for its capacity to hydrolyze and ferment complex macromolecules of organics [40]. In phase Ⅳ, their relative abundance reached 24.69%. Furthermore, norank_f__Eubacteriaceae, Petrimonas, Paludibacter, norank_f__norank_o__WCHB1–41, Christensenellaceae_R-7_group, Ralstonia and S50_wastewater-sludge_group are all representative genera of hydrolytic fermenters [41-43]. Their relative abundance increased within microaerobic environment, indicating that microaeration effectively accelerated the degradation of complex substrates into small-molecule VFAs and alcohols, thereby facilitating hydrolysis process. It is worth noting that Ralstonia, a member of phylum Proteobacteria, is a facultative anaerobic bacterium capable of heterotrophic metabolism using glucose, fructose, and other organics, as well as growing autotrophically with CO₂ [44]. Its enrichment in phase Ⅱ further demonstrated the ecological niche provided by microaeration for facultative anaerobes and contributed to microbial diversity. Methanosarcina is able to produce methane utilizing acetic acid and CO₂/H₂ [45], and has been reported as being tolerant to oxygen [28], with higher abundance under microaeration conditions. However, the growth of Methanobacterium and Methanobrevibacter was partially inhibited. In conclusion, microaeration can stimulate the growth of facultative anaerobes and enhance the presence of functional microorganisms, leading to improved hydrolysis efficiency and greater stability within the anaerobic digestion system.

  In this study, a microaerobic EGSB and MBSBR combined process was established to treat CMR dehydration wastewater. Under microaeration conditions, the enrichment of Bacteroidota, Synergistota and Firmicutes phyla associated with hydrolytic acidification accelerates degradation of organics. The increased microbial richness and diversity not only facilitated the removal of nitrogen and phosphorus pollutants, but also enhanced the stability of the system. Controlling DO concentration to 2 mg/L during aeration phase within MBSBR fostered SND, enhanced pollutant removal efficiency, and led to energy conservation. This study successfully addressed the challenge of acidification failure in anaerobic systems under shock loading, providing theoretical support and technical guidance for the treatment of CMR dehydration wastewater.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Yongqi Liang: Writing – original draft, Investigation, Formal analysis, Data curation, Conceptualization. Chuchu Chen: Writing – review & editing, Data curation. Yihong Chen: Writing – review & editing, Investigation. Huazhe Wang: Writing – review & editing, Supervision, Resources, Funding acquisition. Qi Zhao: Writing – review & editing, Formal analysis. Qinglian Wu: Writing – review & editing. Wan-Qian Guo: Writing – review & editing, Supervision, Resources, Funding acquisition.

  Acknowledgments

  This work received funding from the National Key R&D Program of China (No. 2019YFC1906600), the National Natural Science Foundation of China (No. 52200049), the China Postdoctoral Science Foundation (No. 2022TQ0089), the Heilongjiang Province Postdoctoral Science Foundation (No. LBH-Z22181), the State Key Laboratory of Urban Water Resource and Environment (Harbin Institute of Technology) (No. 2023DX06), and the Fundamental Research Funds for the Central Universities.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110780.

  
  
• 1. [1]

     Q. Lu, C. Li, J. Clean. Prod. 279 (2021) 123856. doi: 10.1016/j.jclepro.2020.123856

  2. [2]

     T. Wang, Q. Liao, Y. Wu, et al., Int. J. Biol. Macromol. 164 (2020) 3846–3857. doi: 10.1016/j.ijbiomac.2020.08.156

  3. [3]

     J. Luo, R. Yang, F. Ma, et al., Environ. Sci. Pollut. Res. 30 (2023) 32153–32167. doi: 10.1007/s11356-023-25614-4

  4. [4]

     C. Huang, Z.X. Li, Y. Wu, et al., J. Environ. Manage. 299 (2021) 113607. doi: 10.1016/j.jenvman.2021.113607

  5. [5]

     A. Saha, B.B. Basak, A. Banerjee, J. Appl. Res. Med. Aromat. Plants 31 (2022) 100428.

  6. [6]

     X. Su, Q. Xue, M. Sun, et al., Bioresour. Technol. 331 (2021) 125073. doi: 10.1016/j.biortech.2021.125073

  7. [7]

     L. Lv, W. Li, C. Wu, et al., Bioresour. Technol. 240 (2017) 84–93. doi: 10.1016/j.biortech.2017.01.053

  8. [8]

     W. Li, C. Su, X. Liu, et al., Environ. Sci. Pollut. Res. 21 (2014) 8167–8175. doi: 10.1007/s11356-014-2709-y

  9. [9]

     D. Nguyen, V. Gadhamshetty, S. Nitayavardhana, et al., Bioresour. Technol. 193 (2015) 513–522. doi: 10.1016/j.biortech.2015.06.080

  10. [10]

      F. Girotto, W. Peng, R. Rafieenia, et al., Waste Biomass Valori. 9 (2018) 161–174. doi: 10.1007/s12649-016-9785-9

  11. [11]

      N. Duc, S.K. Khanal, Biotechnol. Adv. 36 (2018) 1971–1983. doi: 10.1016/j.biotechadv.2018.08.007

  12. [12]

      Q. Chen, W. Wu, D. Qi, et al., Sci. Total Environ. 710 (2020) 136388. doi: 10.1016/j.scitotenv.2019.136388

  13. [13]

      M. Peces, S. Astals, W.P. Clarke, et al., Bioresour. Technol. 200 (2016) 631–638. doi: 10.1016/j.biortech.2015.10.085

  14. [14]

      S. Xu, A. Selvam, J.W.C. Wong, Waste Manage. 34 (2014) 363–369. doi: 10.1016/j.wasman.2013.10.038

  15. [15]

      J. Wang, H. Rong, Y. Cao, et al., Bioprocess. Biosyst. Eng. 43 (2020) 1833–1846. doi: 10.1007/s00449-020-02374-w

  16. [16]

      Y. Cao, C. Zhang, H. Rong, et al., Water Res. 108 (2017) 86–94. doi: 10.1016/j.watres.2016.10.063

  17. [17]

      V. Hoang, R. Delatolla, T. Abujamel, et al., Water Res. 49 (2014) 215–224. doi: 10.1016/j.watres.2013.11.018

  18. [18]

      J.W. Lim, C.E. Seng, P.E. Lim, et al., Bioresour. Technol. 102 (2011) 9876–9883. doi: 10.1016/j.biortech.2011.08.014

  19. [19]

      L. Li, X. Peng, X. Wang, et al., Bioresour. Technol. 248 (2018) 20–28. doi: 10.1016/j.biortech.2017.07.012

  20. [20]

      J.W. Lim, J.Y. Wang, Waste Manage. 33 (2013) 813–819. doi: 10.1016/j.wasman.2012.11.013

  21. [21]

      S. Sun, M. Zhang, X. Gu, et al., Bioresour. Technol. 369 (2023) 128495. doi: 10.1016/j.biortech.2022.128495

  22. [22]

      Z. Li, Y. Hu, C. Liu, et al., Bioresour. Technol. 275 (2019) 94–100. doi: 10.1016/j.biortech.2018.12.003

  23. [23]

      Y. Fan, S. Sun, S. He, Water Res. 235 (2023) 119837. doi: 10.1016/j.watres.2023.119837

  24. [24]

      W. Ma, Y. Han, W. Ma, et al., Bioresour. Technol. 244 (2017) 84–91. doi: 10.1016/j.biortech.2017.07.083

  25. [25]

      Y. Mei, Y. Yang, R. Gao, et al., Curr. Res. Food Sci. 7 (2023) 100584. doi: 10.1016/j.crfs.2023.100584

  26. [26]

      R.A. Isbrucker, G.A. Burdock, Regul. Toxicol. Pharm. 46 (2006) 167–192. doi: 10.1016/j.yrtph.2006.06.002

  27. [27]

      Y. Ye, P. Yang, Y. Deng, et al., Chin. Chem. Lett. 33 (2022) 3127–3132. doi: 10.1016/j.cclet.2021.10.009

  28. [28]

      S.F. Fu, F. Wang, X.S. Shi, et al., Chem. Eng. J. 287 (2016) 523–528. doi: 10.1016/j.cej.2015.11.070

  29. [29]

      S.F. Fu, S. He, X.S. Shi, et al., Bioresour. Technol. 198 (2015) 497–502. doi: 10.1016/j.biortech.2015.09.029

  30. [30]

      L. Feng, W. Zhao, Y. Liu, et al., Chin. Chem. Lett. 34 (2023) 107439. doi: 10.1016/j.cclet.2022.04.037

  31. [31]

      J.W. Lim, J.A. Chiam, J.Y. Wang, Bioresour. Technol. 171 (2014) 132–138. doi: 10.3339/jkspn.2014.18.2.132

  32. [32]

      L. Zhou, X. Yan, X. Pei, et al., Chin. Chem. Lett. 33 (2022) 428–433. doi: 10.1016/j.cclet.2021.06.055

  33. [33]

      S.G. Shin, G. Han, J. Lim, et al., Water Res. 44 (2010) 4838–4849. doi: 10.1016/j.watres.2010.07.019

  34. [34]

      D. Pan, S. Shao, J. Zhong, et al., Bioresour. Technol. 348 (2022) 126726. doi: 10.1016/j.biortech.2022.126726

  35. [35]

      J. Li, L. Zheng, C. Ye, et al., Bioresour. Technol. 329 (2021) 124897. doi: 10.1016/j.biortech.2021.124897

  36. [36]

      Y. Peng, X. Gu, M. Zhang, et al., Water Res. 248 (2024) 120868. doi: 10.1016/j.watres.2023.120868

  37. [37]

      M. Sun, Z. Shi, C. Zhang, et al., Appl. Environ. Microbiol. 88 (2022) e01042-22.

  38. [38]

      B. Yang, H. Xu, Y. Liu, et al., Chem. Eng. J. 383 (2020) 123211. doi: 10.1016/j.cej.2019.123211

  39. [39]

      B. Hu, C. Liu, W. Jiang, et al., Int. J. Biol. Macromol. 192 (2021) 398–406. doi: 10.1016/j.ijbiomac.2021.09.127

  40. [40]

      J. Du, C. Tian, J. Xiao, et al., Water Res. 251 (2024) 121110. doi: 10.1016/j.watres.2024.121110

  41. [41]

      I. Zeb, J. Ma, F. Mehboob, et al., J. Cleaner Prod. 219 (2019) 797–808. doi: 10.1016/j.jclepro.2019.01.295

  42. [42]

      T. Inaba, M. Yamaguchi, A. Taniguchi, et al., J. Gen. Appl. Microbiol. 68 (2022) 242–247. doi: 10.2323/jgam.2022.04.003

  43. [43]

      Y. Zuo, Y. Shao, L. Wang, et al., J. Environ. Manage. 328 (2023) 116977. doi: 10.1016/j.jenvman.2022.116977

  44. [44]

      B. Lin, C. Hyacinthe, S. Bonneville, et al., Environ. Microbiol. 9 (2007) 1956–1968. doi: 10.1111/j.1462-2920.2007.01312.x

  45. [45]

      Q. Cao, W. Zhang, T. Lian, et al., Bioresour. Technol. 365 (2022) 128140. doi: 10.1016/j.biortech.2022.128140

• 

  Figure 1  (a) COD removal efficiencies of Mi-EGSB during each phase. (b) pH of influent and effluent of Mi-EGSB. (c) Composition of VFAs in the effluent of Mi-EGSB. (d) Methane production rate and methane yield of Mi-EGSB. (e) NH₄⁺-N and TN removal efficiencies of Mi-EGSB. (f) TP removal efficiencies of Mi-EGSB during Ⅵ and Ⅶ phases.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) COD, (b) NH₄⁺-N, (c) TN and (d) TP removal efficiencies at different phases of MBSBR operated under different DO conditions. (e, f) Variations in the concentrations of nitrogen, COD and PO₄³⁻ during a cycle under different DO conditions.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Degradation trends of typical pollutants liquiritin and glycyrrhizic acid.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Changes in microbial community diversity and richness in Mi-EGSB during different phase. (b) The relative abundance changes of microbial community at phylum level in different phases of operation in Mi-EGSB. (c) Heatmap of microbial community at genus level with average abundance ranking in the top 20 in Mi-EGSB.

  下载: 全尺寸图片 幻灯片

  Table 1.  Characteristics of the wastewater.

  COD (mg/L)	pH	NH4+-N (mg/L)	TN (mg/L)	TP (mg/L)	UV254
  28,000-34,000	5–6	530	560	80	96–98

  下载: 导出CSV

  Table 2.  Parameters of Mi-EGSB operating process.

  Phase	Time (d)	COD (mg/L)	HRT (h)	OLR (kg COD m−3 d−1)	DO (mg/L)
  Ⅰ	0-49	2000	24	2	/
  Ⅱ	50-76	2000	24	2	0.1
  Ⅲ	77-114	3000	24	3	0.1
  Ⅳ	115-138	4000	24	4	0.1
  Ⅴ	139-151	5000	24	5	0.1
  Ⅵ	152-190	5000	18	6.7	0.1
  Ⅶ	191-220	5000	18	6.7	0.2

  下载: 导出CSV
•