npj Clean Water最新文献

Real-time modeling is vital for the intelligent management of urban water distribution systems (WDSs), enabling proactive decision-making, rapid anomaly detection, and efficient operational control. In comparison with traditional mechanistic simulators, data-driven models offer faster computation and reduced calibration demands, making them more suitable for real-time applications. However, existing models often accumulate long-term prediction errors and fail to capture the strong temporal dependencies in measured time series. To address these challenges, this study proposes the Kolmogorov–Arnold Attention Network for the real-time modeling of WDSs (KANSA), which combines Kolmogorov–Arnold Networks with attention mechanisms to extract temporal dependency features through bidirectional spatiotemporal processing. Additionally, a multi-equation soft-constraint formulation embeds mass and energy conservation laws into the loss function, mitigating cumulative errors and enhancing physical consistency. Evaluations on a benchmark network and a real-world system demonstrate that KANSA achieves high-accuracy real-time estimation and pattern fidelity while maintaining engineering-grade hydraulic balance.

Data-driven models often neglect the underlying physical principles, limiting generalization capabilities in water distribution systems (WDSs). This study presents a novel spatio-temporal graph physics-informed neural network (ST-GPINN) for water quality prediction in WDSs, integrating hydraulic simulations, physics-informed neural networks (PINNs), and graph neural networks (GNNs) to capture dynamics and graph-based network connectivity while approximating partial differential equations (PDEs). ST-GPINN discretizes WDSs using virtual nodes to enhance spatial granularity, employs an Encoder-Processor-Decoder architecture for predictions. Validated on Network A (a small-scale network with 9 junctions and 11 pipes) and Network B (a real large-scale WDS with 920 junctions and 1032 pipes), ST-GPINN outperforms others, achieving a MAE of 0.0073 mg/L, RMSE of 0.0121 mg/L, and R² of 88.91% in Network A, and a MAE of 0.008 mg/L, RMSE of 0.0098 mg/L, and R² of 98.91% in Network B. Its scalability and accuracy highlight ST-GPINN’s potential for water quality predictions.

The anaerobic digester liquor after thermal hydrolysis pretreatment (THP-AD liquor) is a highly hazardous wastewater containing high concentration of ammonium. Herein, a novel two-stage tandem-type simultaneous partial nitrification, anammox, and denitrification-integrated fixed biofilm activated sludge (SNAD-IFAS) hybrid process was successfully constructed for treating THP-AD liquor. The average removal efficiencies of ammonium, total nitrogen, and chemical oxygen demand in the stable phase were 94.0%, 89.9%, and 66.8%, respectively. The specific anammox activity of the two SNAD-IFAS reactors was 68.72 and 42.15 mg N/(g VSS·h), respectively. Candidatus Kuenenia and Candidatus Brocadia (13.49–20.94%) were the main genus of anammox bacteria. Although the relative abundance of Nitrosomonas (0.21–1.67%) was very lower than other bacteria, Nitrosomonas was a central genus in the co-occurrence network. Moreover, the genes involved in pyridine, dioxin, protein, and carbohydrate degradation were remarkably enriched in SNAD-IFAS, indicating the co-metabolism mechanism of refractory organic degradation. This study provides a low energy consumption, high efficiency, and low-carbon technology for treating THP-AD liquor.

Municipal sludge retains fecal pathogens that pose challenges in its treatment and utilization. Mesophilic anaerobic digestion (MAD) fails to achieve adequate pathogen reduction while demonstrating inconsistent methane. To address these, the feasibility of reducing fecal pollution indicators, including Escherichia coli (EC), human-specific Bacteroides HF183 (HF183), human adenovirus (HAdV), JC and BK polyomaviruses (JCPyV and BKPyV), and crAssphage by combining MAD with CaO₂ pre-treatment was assessed. Both ·OH and ·O₂⁻ were produced during CaO₂ pre-treatment. ·OH mainly accounted for the removal of these indicators, while ·O₂⁻ reduced their infectivity. In MAD process, CaO₂ pre-treatment accelerated the decay of most indicators with the elevated ammonia nitrogen in digested sludge. Overall, CaO₂ pre-treatment and the MAD process predominantly contributed to the reduction of HF183, JCPyV and HAdV and the reduction of crAssphage, EC, and BKPyV, respectively, and combining MAD with CaO₂ pre-treatment is an effective approach for removing fecal indicators from sludge.

Conventional flow-electrode capacitive deionization (FCDI) often exhibits performance constraints stemming from elevated ion migration resistance associated with diminished conductivity within the desalination chamber, particularly under complex aqueous matrices. This investigation introduces a symmetric anion-exchange membrane (AEM) configuration engineered to circumvent these conductivity limitations and enhance arsenic removal efficacy. Relative to conventional designs, the symmetric-AEM configuration demonstrated an approximate 19.4% enhancement in arsenic removal efficiency. For influent streams with initial arsenic concentrations of 1000 μg·L⁻¹, effluent concentrations were diminished below the analytical detection limit (0.02 μg·L⁻¹) employing a two-stage sequential process. This configuration sustains or potentially enhances desalination-chamber conductivity by optimizing ion migration pathways and facilitating anion compensation via highly mobile chloride ions. The contributions of chloride ions as supporting electrolytes and the transformations of arsenic valence states were interrogated, providing mechanistic insights into the observed performance improvements. Our findings signify a practical advancement in FCDI, presenting a potentially robust and efficacious strategy for arsenic remediation in contaminated groundwater. Thus, the symmetric-AEM configuration represents a significant advancement toward the broader implementation and practical application of FCDI systems for potable water production.

Increasingly stringent water quality standards are forcing more water treatment facilities to implement adsorption steps. Activated carbon is efficient but has a high environmental impact due to CO₂ emissions and energy demand. Adsorbents derived from water treatment residuals offer a potential solution. In this study, a novel laboratory rotary furnace was designed to produce clay-carbon composite adsorbents from drinking water treatment residues. The process was optimized using a statistical design of experiments, representing the first comprehensive statistical analysis of the thermal activation of such residuals. Thermal activation increased the specific surface area almost tenfold (112–201 m²/g). The adsorbents were tested for removal of ibuprofen, caffeine, diclofenac (1 µg/L), and brilliant blue FCF (5 mg/L). Response surface models showed that heating rate (p < 0.003) and ramp duration (p < 0.00002) significantly influenced adsorption capacity. Mass balance calculations suggest on-site production could fully substitute activated carbon and generate surplus material.

Biotic methylation of inorganic mercury (iHg) in aquatic systems is largely driven by microorganisms such as sulfate-reducing bacteria (SRB). Using the SRB model strain Pseudodesulfovibrio hydrargyri BerOc1 we investigated biotic iHg methylation aiming to assess the rates of mono-methylmercury (CH₃Hg) production and to characterize the carbon (C) isotopic signatures (δ¹³C) of the CH₃Hg product. BiogenicCH₃Hg exhibited δ¹³C values averaging −23.1 ± 2.0‰, representing a ¹³C-depletion of 14.4‰ compared to the pyruvate carbon source used for the growing of the strain and a 9‰ depletion relative to the microbial biomass. The maximum methylation yield observed in our samples was around 15% of the available iHg and a constant C isotope fractionation was detected over time. We propose that the methyl group is metabolically transferred from the carbon sources to cobalamin in the HgcA protein and subsequently to inorganic mercury (iHg), leading to consistent light C isotope enriched CH₃Hg signatures.

The surge in Africa’s economy has led to increased use of per- and polyfluoroalkyl substances (PFASs), resulting in environmental pollution. This has spurred regulatory and research efforts to mitigate these impacts. We used data analytics to study PFAS pollution in aquatic ecosystems in Africa and model PFAS concentrations using socio-economic variables. In agreement with literature, urbanization, gross domestic product, and population growth drive PFAS pollution.

In this contribution, we report the synthesis of a poly(4-vinylpyridine)-reduced graphene oxide-magnetite (P4VP-rGO-Fe₃O₄) organo-magnetogel (OMG), designed for high-performance pollutant adsorption. In the OMG, rGO and Fe₃O₄ nanoparticles are in situ encapsulated during the chemical cross-linking of the 4-vinylpyridine polymer. The adsorption performance of OMG was evaluated using three model water pollutants, viz., organic dyes, heavy metal ions, and waterborne pathogens. The equilibrium adsorption capacity exceeded 400 mg/g for the organic dyes. Beyond dye removal, the OMG also adsorbed heavy metal ions, such as AsO₂⁻, Pb²⁺, Cr₂O₇^2−, and Cd²⁺ ions, with removal efficiencies exceeding 60% and adsorption capacities exceeding 200 mg/g. The OMG also exhibited remarkable antibacterial activity against E. coli and S. Typhi, with almost zero viability for S. Typhi. The OMG promises a broad-spectrum applicability in wastewater treatment, offering a sustainable and efficient solution for water decontamination.

In this study, the composite photocatalyst UiO-66-NH₂@TiO₂ and a polysulfone membrane modified with it were evaluated for removing humic acid (HA) under visible light in a photocatalytic membrane reactor (PMR). The photocatalyst and membrane were analyzed using FTIR, XRD, FESEM, PL, DRS, AFM, BET, contact angle, and porosity tests to assess pollutant removal, water flux, and membrane resistance. Synthesis was confirmed by FTIR and XRD, while DRS showed a 2.87 eV bandgap, indicating visible light activity. PL results revealed reduced electron-hole recombination. FTIR and SEM confirmed photocatalyst presence and uniform dispersion in the membrane, improving hydrophilicity by decreasing the contact angle by 8°. In suspension, the photocatalyst removed 93% of HA under visible light. Among modified membranes, PS-UNT6% performed best, showing a 26% increase in pure water flux and 12% higher HA removal (total 95.2%) compared to the unmodified membrane. Fouling was significantly reduced, with only a 14% flux drop after 12 h under light, versus 19% without modification. Pore-blocking resistance decreased by over 98%, along with reductions in intrinsic, fouling, and cake resistances, demonstrating enhanced membrane performance through photocatalytic modification.