Sudden shock loads in wastewater influent can severely disrupt biological treatment processes and cause effluent quality exceedances in wastewater treatment plants, particularly in domestic-industrial integrated facilities. Timely and reliable early warning of such disturbances is critical for enabling proactive intervention and minimizing environmental and operational risks. This study develops a COD-centric closed-loop early warning framework that integrates a machine-learning-based soft-sensing module with a multi-step effluent prediction module, in which anomaly tagging is automatically triggered by effluent discharge limit thresholds. Two coupling architectures were evaluated: a serially coupled architecture (SCA) that sequentially connects influent sensing and effluent prediction, and a jointly coupled architecture (JCA) that enables end-to-end learning within a unified model. Feature importance was interpreted using SHAP analysis and validated through ablation studies to identify the key process variables that govern predictive performance and early-warning responsiveness. In a full-scale integrated wastewater treatment plant (IWTP) case study, the proposed framework achieved 95.0% accuracy and 87.2% anomaly detection precision for 12 h ahead warnings, with JCA outperforming SCA. These results demonstrate that backward inference from predicted effluent compliance risk enables timely identification of upstream disturbances before limit violations occur. This integrated and interpretable framework provides a novel, real-time, and cost-effective solution for linking effluent-risk forecasting with influent-anomaly diagnosis, substantially enhancing the operational resilience and proactive management of IWTPs.

The stringent effluent standards rise strong demand for efficient wastewater treatment, particularly for the superior removal of nitrogen and phosphorus. To address the issues of membrane fouling and insufficient phosphorus removal in the internal circulation aerated membrane bioreactor (ICAMBR) treating municipal wastewater, this study achieved significantly improved system performance and effective membrane fouling control through the addition of polyaluminum chloride (PAC). The findings showed that the addition of PAC resulted in average removal efficiencies of chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP) of 98.87%, 98.46%, and 94.83%, respectively, while the effluent TP level was reduced to 0.26 mg/L. Additionally, PAC effectively reduced the extracellular polymeric substances (EPS) content and enhanced sludge settleability, which led to a decrease in membrane fouling from 1.42 × 1012 m−1d−1 to 0.54 × 1012 m−1d−1. The irreversible membrane resistance decreased from 1.98 × 108 m−1 to 0.35 × 108 m−1, thereby effectively alleviating membrane fouling in the system. Microbial analysis revealed that PAC fostered the enrichment of denitrifying and ammonia-oxidizing bacteria, which in turn enhanced the abundance of key nitrogen metabolism genes, thereby reinforcing the nitrogen removal pathways in the system. This study provides both theoretical insights and technical support for the enhanced operation of ICAMBR through PAC addition in municipal wastewater treatment.

Background The removal of heavy metal ions from aqueous solutions has received increasing attention due to their non-biodegradability and potential for bioaccumulation in the food chain. Methods In this study, epichlorohydrin-crosslinked alginate fibers (EAFs) were synthesized via an organic covalent crosslinking strategy as an alternative to the traditional Ca(II) ionic crosslinking method, aiming to enhance the adsorption performance of alginate-based adsorbents toward heavy metals. Significant Findings Compared with Ca(II) crosslinked alginate fibers (Ca(II)-AFs), the Pb(II), Cd(II), Zn(II), Ni(II), Cu(II), and Co(II) uptakes on EAFs increased by 8.5%, 8.9%, 63.6%, 19.7%, 24.5%, and 46.1%, respectively, with adsorption capacities reaching 401.9 mg/g, 149.0 mg/g, 85.1 mg/g, 92.6 mg/g, 124.8 mg/g, and 82.6 mg/g. Adsorption equilibrium was achieved within 20 mins, indicating rapid kinetics. Furthermore, EAFs could be readily regenerated using a HNO3 solution, retaining >85% of the initial adsorption capacity after four adsorption-desorption cycles, demonstrating excellent reusability. Finally, the adsorption mechanism was elucidated through X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FT-IR), and X-ray photoelectron spectroscopy (XPS) analyses, which suggested that heavy metals were primarily captured through electrostatic attraction and coordination between the carboxyl groups of alginate and the heavy metal ions.

Ammonia-laden wastewater containing residual tetracycline (TC) and insufficient readily biodegradable carbon poses a challenge for stable biological nitrogen removal. In this study, a microaerophilic anoxic/oxic process based on the Biological Low Oxygen and High Activated Sludge concentration (Bio-LOHAS) strategy was applied to treat synthetic high-ammonia, low-C/N wastewater under stepwise TC loading (0-5.0 mg/L). Compared with conventional AO operation, Bio-LOHAS combines low dissolved oxygen and high sludge concentration to promote simultaneous nitrification and denitrification (SND) and improve tolerance to antibiotic stress. At TC <= 1.0 mg/L, the system maintained stable COD and total nitrogen (TN) removal above 90% and 85%, respectively. At 5.0 mg/L TC, nitrite accumulated because nitrite-oxidizing bacteria were more strongly inhibited than ammoniaoxidizing bacteria, shifting nitrogen conversion toward partial nitrification and denitrification. However, relatively high TN removal was still retained because denitrification was less affected than nitrification. Recovery was achieved after influent TC reduction or withdrawal, indicating improved tolerance of acclimated sludge during re-exposure rather than unrestricted stable operation under continuously high TC loading. Overall, BioLOHAS is a promising strategy for improving nitrogen-removal resilience in antibiotic-impacted, carbon-limited ammonia-rich wastewater.

Iron oxides can facilitate anaerobic digestion but are faced with electron competition between methanogens and iron-reducing bacteria. In this study, we developed a highly crystalline iron oxide material via the pyrolytic synthesis of metal-organic framework (MOF) precursors and systematically investigated its mechanistic role in improving anaerobic digestion. The results showed that 10 g/L of the MOF-derived iron oxide (MDF) enhanced the methane yield to 547.91 mL/g volatile solids. Characterisation of the MDF material revealed that it had a highly ordered crystalline structure (95.56 % crystallinity) and superior electrochemical activity, with a positively charged surface facilitating microbial adhesion. Microbial analysis revealed that MDF selectively enriched Methanosarcina (methanogens) and Clostridium_sensu_stricto_1 (syntrophic bacteria), while up-regulating genes associated with conductive pili (PilA) and quorum-sensing signalling molecules. This shift established a direct interspecies electron transfer-driven metabolic network. Furthermore, the high crystallinity of MDF suppressed the enrichment of iron-reducing bacteria (Trichococcus abundance, < 2.5 %), thereby mitigating iron reduction competition for acetate/hydrogen and overcoming the limitations of conventional Fe2O3. This superior stability was further confirmed in a long-term experiment, where MDF exhibited both higher methane yield and greater resistance to reduction than Fe2O3. This study provides fundamental insights into crystallinity-dependent electron transfer pathways for methanogenesis and a strategic framework for engineering stable conductive materials to enhance the efficiency and stability of anaerobic digestion.

Commercial adsorption resins are widely used for heavy metals removal from industrial wastewater, yet their selection still depends on labor-intensive and time-consuming experiments. Machine learning (ML) offers a promising alternative, but its predictive power is often constrained by insufficient mechanistic understanding. In this study, a hybrid grid search-extreme gradient boosting (GS-XGBoost) model was developed by integrating adsorption mechanisms into a data-driven framework. Key physicochemical descriptors, including chemical structure, atomic number, valence, electronegativity, atomic radius, as well as environmental conditions including temperature and pH, are incorporated, achieving high predictive accuracy (R2 = 0.915). SHAP analysis identified chemical structure as the most critical factor, where nitrogen-containing functional groups (excluding symmetrical N-containing motifs) enhanced heavy metal adsorption, whereas long-chain structures suppressed it. To further improve generalization, Density Functional Theory (DFT) were employed to identify actual adsorption sites, thereby refining the model's prediction of intramolecular adsorption configurations. This approach reduced the average relative error from 57.13 % to 24.27 %. This study establishes a robust framework for rapid adsorbent screening and provides a novel strategy for integrating mechanistic insights into data-driven models, advancing the intelligence design of adsorbents for environmental remediation.

Real-time and reliable monitoring of nitrogen in wastewater treatment plants (WWTPs) is essential for effluent quality control and mitigation of environmental risks such as eutrophication. However, conventional online analyzers for total nitrogen (TN) and related nitrogen indicators remain costly, maintenance-intensive and difficult to deploy at high temporal resolution, especially under data-scarce conditions. To address this limitation, a hybrid soft sensing framework was developed by integrating a simplified prior model with a calibration model through a bagging ensemble strategy, based on 1068 laboratory-simulated datasets and 576 real-world observations from a WWTP. The prior model, derived from controlled water-mixing experiments, provided a constrained reference representation, while the calibration model learned the system-level discrepancy between this reference system and real wastewater observations. Using pH, electrical conductivity (EC), oxidationreduction potential (ORP), and dissolved oxygen (DO) as inputs, the proposed framework improved both robustness and predictive accuracy under data-scarce conditions. Compared with purely data-driven models, it reduced MSE, MAE, and RMSE by 41.21%, 24.31%, and 23.58% for ammonia nitrogen (NH4+-N), and by 45.79%, 25.74%, and 26.36% for TN, respectively. Reliable prediction was maintained with as little as 10% and 40% of the original training data for NH4+-N and TN, respectively. In addition, lifecycle monitoring cost was reduced by 91.34% relative to conventional online analyzers. These results demonstrate that the proposed framework offers a cost-effective and practical solution for nitrogen monitoring in WWTPs under data-scarce conditions.

The efficient recovery of tin (Sn) from acidic steel wastewater remains a critical challenge for resource sustainability and environmental protection. Herein, epichlorohydrin (ECH) cross-linked alginate fibers (ECHalginate fibers) were developed via covalent bonding between the C3-OH groups of alginate and epoxide/C-Cl groups of ECH. Unlike conventional Ca(II)-cross-linked alginate prepared in CaCl2 solutions, this ethanol-based solidification strategy avoids ion-exchange consumption of carboxyl groups, thereby preserving active adsorption sites. The resultant fibers exhibited excellent structural integrity and a high Sn(II) uptake of 497.6 +/- 92.6 mg/g at pH 2.3 within 180 min, 7.4-48.5 times higher than that of commercial carbons and ion exchange resins. Comprehensive characterization (FE-SEM, BET, XRD, FT-IR, XPS) and DFT calculations confirmed that ECH crosslinking occurred predominantly at C3-OH sites, while Sn(II) adsorption proceeded through chelation with-COO-groups and Na(I)/Sn(II) exchange, followed by oxidation to Sn(IV). A speciation-weighted adsorption energy model Eads,adj was established, showing a strong linear correlation (R2 = 0.995) with experimental capacities, enabling quantitative prediction of adsorption performance under variable pH. Moreover, the ECH-alginate fibers demonstrated rapid regeneration in 0.5 mol/L HCl with complete desorption within 5 min and minimal capacity loss after five cycles. The practical application of ECH-alginate fibers yielded a remarkable recovery efficiency of 99.7 +/- 1.0% for Sn from real Sn-plating wastewater. This work provides a mechanistically guided and scalable approach for designing durable, high performance biopolymer adsorbents for sustainable metal resource recovery.

Monitoring ammonia nitrogen (NH4-N) in wastewater treatment plants (WWTPs) is essential for environmental protection and process optimization. Traditional detection methods are often time-consuming and costly, limiting their widespread application. To address this, a soft sensing approach based on the random forest (RF) algorithm was developed for NH4-N monitoring. To handle non-periodic fluctuations in influent water quality, a grid search-optimized random forest model (GS-RF) was constructed, enabling automatic hyperparameter tuning to adapt to varying conditions. The model was validated using real-world data from a WWTP, demonstrating high predictive accuracy with low error metrics. This study not only offers a novel and cost-effective strategy for NH4-N monitoring in WWTPs but also contributes to the advancement of intelligent process control in wastewater treatment operations.

The aerobic granular sludge (AGS) process has emerged as a viable alternative to landfill leachate treatment. The mechanisms by which dissolved organic matter (DOM) in landfill leachate, a potential stimulant, is utilized during treatment with AGS systems remain unclear. In this study, we revealed DOM-mediated nitrogen removal in AGS receiving the effluent from up-flow anaerobic sludge blanket (UASB). The results showed that granules were successfully formed with real fresh leachate contents increasing from 10% to 30%. The established AGS bioreactor could achieve good nitrogen removal (76.16% on average) through partial nitrification-denitrification (PND), effectively reducing carbon demand and selectively utilizing DOM from leachate. Fourier transform ion cyclotron resonance mass spectrometry analysis showed the utilization of DOM of aliphatic, protein-like and amino sugar-like compounds with low saturation in situ of leachate. Microbial analysis identified that Thauera and Rhodobacter belong to Proteobacteria as the dominant nitrogen-removing bacteria, and OLB13, OLB12, and Devosia, accounting for 16.11% of the microbial community, are the primary DOM degraders, with significant correlation (p < 0.05). High-throughput analysis showed that protein-like and amino sugar-like compounds were the main DOM components used to facilitate PND via diverse metabolic pathways. This study suggests that the critical role of DOM interactions with AGS microbial consortia could shed light on the regulation of nitrogen removal in landfill leachate treatment.

To address the dewatering challenges associated with anionic polyacrylamide (APAM) in water treatment processes, APAM-filled alginate soft capsules (AFASCs) were prepared through encapsulating APAM within an alginate-Ca(II) shell, and used to efficiently remove heavy metals from aqueous solutions. The structural properties and adsorption mechanisms of AFASCs were systematically analyzed using a microscope, FT-IR, XPS, XRD, and DFT calculations. The results confirmed that APAM was successfully encapsulated by the alginate-Ca(II) shell with a thickness of about 177.6 mu m, and it acted as the primary component responsible for removing heavy metals. The adsorption kinetics followed an intra-particle diffusion model, and the maximum adsorption capacities for Cu(II), Cd(II), and Pb(II) were found to be 270.6 +/- 8.1 mg/g, 274.6 +/- 13.6 mg/g, and 563.8 +/- 10.5 mg/g, respectively. Mechanistic studies revealed that the main way heavy metals were removed was through coordination with the oxygenated functional groups (-COOH/C-OH on alginate, and O=C on APAM) and-NH2 groups on APAM. Additionally, anion exchange between heavy metals and calcium ions (Ca(II)) adsorbed by the-COO-groups on alginate also played a role. Furthermore, the adsorbed Cu(II) was reduced to Cu(I) due to electron transfer when C-C/C-H were oxidized to C-O and C=O. The AFASCs could be regenerated using an acidic CaCl2 solution. The adsorption capacity decreased only approximately 20 % after being reused five times, making this approach sustainable for both APAM utilization and heavy metal remediation. This research provides a promising strategy for addressing APAM utilization challenges and reducing heavy metal contamination.

Conventional biological treatment of low carbon-to-nitrogen (C/N) municipal wastewater is challenged by the need for supplemental carbon sources and high aeration energy. Here, we first introduce a Biological Low Oxygen and High Activated Sludge concentration (Bio-LOHAS) system and evaluate its performance under two low dissolved oxygen (DO) gradients strategies: a monotonically increasing DO profile (M-O) and a reverse profile (O-M). At an influent C/N ratio of 4.41 ± 0.92, the O-M strategy outperformed the M-O strategy, achieving total nitrogen (TN) and total phosphorus (TP) removal efficiency of 77.8 % and 95.8 %, respectively, compared with 64.3 % and 68.9 % under M-O. Integration of in-situ pathway profiling and batch experiments revealed that the O-M strategy facilitated carbon allocation and enhanced microbial synergy. Concurrently, 16S rRNA-based community analysis indicated that the O-M strategy favored the enrichment of denitrifying glycogen-accumulating organisms (DGAOs) and denitrifying phosphorus-accumulating organisms (DPAOs), thereby driving polyhydroxyalkanoates (PHAs) -mediated denitrifying phosphorus removal. Notably, it promoted a dynamic balance between DGAOs and DPAOs, optimized internal carbon source conversion efficiency, and increased microbial network complexity. Metagenomic analysis further confirmed the activation of endogenous denitrification and polyphosphate metabolic pathways, with increased abundance of key functional genes involved in PHAs-glycogen cycling and polyphosphate synthesis. More importantly, the Bio-LOHAS process reduced external carbon demand by 40 % and aeration energy input by 27-33 %. This study demonstrates the potential of the Bio-LOHAS process as a promising and energy-efficient strategy for low C/N municipal wastewater treatment and provides a rational basis for optimizing DO gradients in full-scale applications.

The purification of mercury-contaminated water has garnered significant attention due to the severe biotoxicity of mercury. To efficiently remove mercury ions (Hg(II)) from aqueous solutions, a nitrogen-rich biomass adsorbent material, polyethyleneimine (PEI)-functionalized pectin fibers (PEI-pectin fibers), were synthesized using glutaraldehyde as a cross-linking agent between PEI and pectin. PEI-pectin fibers demonstrated excellent stability and Hg(II) adsorption performance across a broad pH range. The maximum adsorption capacity of PEIpectin fibers for Hg(II) reached 466.2 +/- 23.8 mg/g at pH 5, significantly surpassing the performance of previously reported adsorbents. The synthesis mechanism and adsorption behavior were elucidated using density functional theory, Fourier-transform infrared spectroscopy, and X-ray photoelectron spectroscopy. Adsorption was primarily driven by coordination between Hg(II) and nitrogen-containing functional groups on PEI, along with cation exchange between Hg(II) and Ca(II) ions associated with the carboxyl groups in pectin. Furthermore, Hg(II)-loaded PEI-pectin fibers were easily regenerated using a 0.05 M Na2EDTA solution. PEI-pectin fibers maintained excellent Hg(II) removal efficiency under various conditions, including ultra-low mercury concentrations, varying adsorbent dosages, and the presence of competing ions. The fibers demonstrated strong selectivity for Hg(II), high resistance to interference from common cations and anions, and retained over 94 % removal efficiency even in high-salinity environments. These findings highlight PEI-pectin fibers as a highly promising and cost-effective biomass adsorbent material for the efficient removal of Hg(II) from wastewater.

Unidentified illegal industrial wastewater discharges pose significant threats to the effective operation of domestic-industrial integrated wastewater treatment plants (IWTPs). Identifying the pollutants sources and implementing targeted source control is considered the most effective strategy to mitigate recurrence. In this study, a Genetic Algorithm-optimized Extreme Learning Machine (GA-ELM) model was developed using an ion characteristic database constructed from 1975 valid ion data points collected from 14 industrial dischargers and the domestic influent of IWTPs. By analyzing the weights of different input features in the activation function, less relevance features were excluded, enhancing the model's performance. The GA-ELM achieved high accuracy, with 90.63 % for sector identification and 86.15 % for identifying discharging enterprises. Moreover, the GA-ELM showcased strong computational efficiency, with source identification times of 25.48 s for sectors and 24.13 s for enterprise, reflecting improvements of 9.80 % and 13.48 %, respectively, compared to the model with all features. Validation experiments attested to the GA-ELM model's robust generalization capabilities, particularly in recognizing metal manufacturing wastewater, with an accuracy of approximately 99 %. This study presents a novel approach for identifying industrial wastewater, particularly inorganic pollutants, and contributes to advancing the application of artificial intelligence in the management of IWTPs.

For recovering gold from gold cyanide (Au(I)) leachate, quaternary ammonium (R4N+) functionalized chitosan fibers (QECFs) were developed. The synthesis process was finely tuned via the Box-Behnken design to maximize Au(I) adsorption efficacy. Additionally, the environmental impacts and estimated manufacturing costs of preparing QECFs were assessed through life cycle assessment and economic analysis. The QECFs demonstrated exceptional maximum adsorption capacities of 669.75 +/- 47.20 mg/g at pH = 9.5 within 15 min, surpassing reported adsorbents in both capacity and rate. Comprehensive characterization of adsorbents and mechanisms was conducted using DFT conclusion, FE-SEM, FT-IR, and XPS. The findings elucidated those electrostatic interactions between the R4N+ on QECFs and Au(I), along with anion exchange involving Cl- and Au(I), are integral to the adsorption mechanism. Furthermore, the QECFs could be swiftly regenerated using NaCl solution. The practical application of QECFs yielded a remarkable recovery efficiency of 98.2 % for Au(I) from WPCBs leachates, underscoring their potential utility in the gold recovery.

Traditional methods for developing adsorbents often rely on trial-and-error methods in the laboratory, which are not only time-consuming and labor-intensive but also increase reagent waste. In this study, to efficiently design a high-performance adsorbent for silver (Ag(I)) recovery, a series of functionalized chitosan structures were constructed through density functional theory (DFT). Among the designed structures, thiourea-modified chitosan emerged as the most promising candidate due to its lowest adsorption energy for Ag(I). Based on this designed structure, thiourea-modified chitosan fibers (CFs-TU) was prepared. The resulting CFs-TU exhibited excellent stability and selective capture performance for Ag(I) across a broad pH range. The characteristics of CFs-TU and the underlying adsorption mechanism for Ag(I) were investigated using characterization analysis and DFT calculations. The analysis revealed that at pH 6, the adsorption was primarily governed by the coordination between Ag(I) and -SH/C=S/C-N moieties. However, at pH 2, a synergistic effect was observed, involving electrostatic attraction between NO3- adsorbed on the protonated amino group in the CFs-TU and Ag(I) cations, alongside coordination between Ag(I) and -SH/C=S, which facilitated Ag(I) adsorption. During the adsorption process, Ag(I) was generally reduced to Ag(0) owing to electron transfer as the -NH2/-NH- and -SH functional groups of CFs-TU were oxidized to =N- and C=S. Furthermore, the exhausted CFs-TU could be efficiently regenerated using an acidic thiourea solution. The CFs-TU prepared through this DFT-based approach demonstrated significant potential for the recovery of Ag(I) from aqueous solutions, highlighting a novel and effective method for adsorbent development.

An emergency water pollution incident poses a significant risk to the proper functioning of wastewater treatment plants, particularly in domestic-industrial integrated facilities. Source tracing is recognized as an effective method to mitigate ongoing impacts. Machine learning-assisted traceability is emerging as a more efficient and faster method compared to traditional methods. In this study, a total of 712 sets of characterization wastewater information from effluent samples from14 discharge enterprises across 6 different sectors, as well as domestic wastewater was collected using 3-dimensional fluorescence spectroscopy. After data cleaning and augmentation, a feature fingerprint database of wastewater was constructed to train a traceability model. Several machine learning algorithms, including Back Propagation neural network (BP), Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB) and K-Nearest Neighbors (KNN), were selected for constructing the traceability framework. Subsequently, an advanced Particle Swarm Optimization Random Forest model (PSO-RF), capable of automatically optimizing model parameters, was proposed and applied to trace the sources of wastewater in integrated wastewater treatment plant. The PSO-RF achieved and accuracy of 96.55 % in sector identification and 94.25 % in manufacturer identification. As part of the validation process, laboratory simulations were conducted using blended wastewater with different volume ratios of domestic and industrial wastewater to evaluated the potential application of PSO-RF. The results consistently demonstrated PSO-RF's effectiveness, particularly in tracing pharmaceutical wastewater sources, maintaining an accuracy of over 85 %. This work presents a novel strategy for tracing abnormal sources during emergency pollutant incidents, providing essential support for integrating artificial intelligence (AI) into meticulous wastewater management.

Polymer networks have gained great attention due to their unique structural arrangement and characteristic properties. The interconnection of polymer chains can be achieved through covalent or noncovalent interactions. However, it remains a challenge to explore a strategy for the precise integration of covalent and noncovalent cross-linking within polymer network architectures. In this work, we designed and synthesized hydrogen-bonding preorganized arylhydrazone dual-arm monomer 1 (M1) and triarm monomer 2 (M2), which were used to fabricate covalent polymer networks (CPNs), CPN1 and CPN2, via acid-catalyzed macrocyclization. The resulting CPN1 and CPN2 exhibited well-defined electron-rich macrocyclic cavities, enabling subsequent supramolecular cross-linking with bipyridinium guest (BP) to generate supramolecular/covalent polymer networks (SCPNs), SCPN1 and SCPN2. CPN1 formed denser, more flexible films compared to the brittle CPN2, highlighting the importance of appropriate covalent cross-linking density. Importantly, incorporation of BP significantly improved the mechanical properties of the networks, including Young's modulus, tensile strength, and toughness, demonstrating the effectiveness of host-guest interactions in reinforcing polymer network structures and the potential of macrocycle-driven supramolecular engineering for advanced material design.

Harmless treatment of waste activated sludge (WAS) is the primary goal for sludge management while bioresource recovery is raising wide scholarly and industrial interest due to the abundant resource potential in WAS. This study employed sulfite and freezing/thawing to pretreat WAS for its versatility in simultaneously enhancing methane production, improving dewaterability, and inactivating pathogenic microorganisms. The addition of 100 mg S/L of sulfite and subsequent freezing/thawing significantly promoted the release of organics from a low organic-contained WAS (volatile solid/total solid = 0.4), leading to a substantial increase in methane production by 10.99 %. The combined pretreatment significantly reduced the capillary suction time of sludge by 77.24 % and effectively inactivated pathogenic microorganisms to below detectable levels. From a micro level, sulfur aggregation was observed around sludge particles, which can be attributed to the process of ice formation. During this process, sulfite was continuously transferred to the aqueous phase surrounding the particles where were froze even slower. Moreover, the formation of capillary water ice crystals further compromised the protective function of extracellular polymeric substances (EPS) on sludge cells. Economic, and environmental analyses suggest this combined sulfite and freezing/thawing pretreatment has a potential for efficient sludge treatment towards energy recovery and safe disposal with high environmental and economic value.

Sensitive receiving waters place greater demands on the treatment of wastewater treatment plant (WWTP) tailwaters. In this study, a three-stage constructed rapid infiltration system (CRIs) was developed and modelled for the deep treatment of WWTP tailwater. The average removal of COD, NH4+-N and TN in the effluent in the long-term experiment were 84.5 %, 95.6 % and 77.4 %, respectively. In the system, the first and second stages carry out nitrification and denitrification processes with the mediation of nitrifying bacteria (Nitrospira and Ellin6067) and denitrifying bacteria (Azospira), while dynchronised nitrification denitrification exists in the third stage with nitrifying bacteria (Nitrospira and Nitrosomonas) and denitrifying bacteria (Hyphomicrobium and Thauera). A mathematical model of CRIs was developed and calibrated by coupling biokinetic Constructed Wetland Model No. 1 and Activated Sludge Model No. 3, which was successfully applied to predict the changes in the quality of water entering and leaving the system under different operating conditions, including COD, NH4+-N and TN with the overall error less than 30 %. The simulation results show that the system can withstand a hydraulic load of 1.0 m/d and treat wastewater with influent COD and NH4+-N within 68.9 mg/L and 10 mg/L, respectively. However, the removal rate of the system is obviously affected by the composition of the influent and dissolved oxygen. The results of the study suggest that the application of three-stage CRIs in tailwater treatment is feasible, and the model is a useful tool for predicting the operational effectiveness of CRIs, which can provide technical guidance for wider application.