Machine learning-based model construction and identification of dominant factor for simultaneous sulfide and nitrate removal process.

Accurate water quality prediction models are essential for the successful implementation of the simultaneous sulfide and nitrate removal process (SSNR). Traditional models, such as regression and analysis of variance, do not provide accurate predictions due to the complexity of microbial metabolic pathways. In contrast, Back Propagation Neural Networks (BPNN) has emerged as superior tool for simulating wastewater treatment processes. In this study, a generalized BPNN model was developed to simulate and predict sulfide removal, nitrate removal, element sulfur production, and nitrogen gas production in SSNR. Remarkable results were obtained, indicating the strong predictive performance of the model and its superiority over traditional mathematical models for accurately predicting the effluent quality. Furthermore, this study also identified the crucial influencing factors for the process optimization and control. By incorporating artificial intelligence into wastewater treatment modeling, the study highlights the potential to significantly enhance the efficiency and effectiveness of meeting water quality standards.

Response of simultaneous sulfide and nitrate removal process on acute toxicity of substrate concentration and salinity: Single toxicity and combined toxicity.

Simultaneous sulfide and nitrate removal process has performed excellent to treat nitrogen and sulfur pollutants in wastewater treatment. A high salinity stress poses a great challenge to the treatment of highly saline wastewater containing nitrate and sulfide. In addition, sulfide and nitrates are also toxic for the process, and their high concentration would inhibit the process. Therefore, the current work explores the single acute toxic effect and combined toxic effect of salinity and substrate concentration on the performance of the process from the perspective of toxicology. Considering sulfide and nitrate removal performance as an indicator, the IC50 values of sulfide were 293.20 mg S/L and 572.30 mg S/L, respectively; while those of salinity were 6.14% wt (91.78 mS/cm) and 6.63% wt (98.73 mS/cm), respectively. High substrate concentration or salinity resulted in elemental sulfur generation. The molar ratio of generated elemental sulfur to consumed sulfide(R-Sulfate) was close to 1. The response of nitrate reduction product to the elevating substrate concentration was not obvious, while its response to increasing salinity was on the contrary. With the increasing salinity (1.2% wt to 9.6% wt), molar ratio of generated nitrogen gas to consumed nitrate (R-Nitrogen gas) increased from 0.58 to 1, while molar ratio of generated nitrite to consumed nitrate (R-Nitrite) decreased from 0.43 to 0. Factorial analysis test revealed that the combined acute toxicity of substrate and salinity on sulfide oxidization and nitrate reduction were both antagonistic effects.

Metagenomic analysis of microbial community and metabolic pathway of simultaneous sulfide and nitrite removal process exposed to divergent hydraulic retention times.

The role of hydraulic retention time (HRT) on S0 production was assessed through metagenomics analyses. Considering comprehensive performance for the tested HRTs (0.25-13.33 h), the optimal HRT was 1 h, while respective sulfide and nitrite loading rate could reach 6.84 kg S/(m3·d) and 1.95 kg N/(m3·d), and total S0 yield was 0.36 kg S/(kg (VSS)·d). Bacterial community richness decreased along the shortening of HRT. Microbacterium, Sulfurimonas, Sulfurovum, Paracoccus and Thauera were highly abundant bacteria. During sulfur metabolism, high expression of sqr gene was the main reason of maintaining high desulfurization load, while lacking soxB caused the continuous increase of S0. Regarding nitrogen metabolism, the rapid decrease of nitrite transporter prevented nitrite to enter in cells, which caused a rapid decrease of nitrite removal under extreme HRT. Adjusting HRT is an effective way to enhance S0 production for the application of the simultaneous sulfide and nitrite removal process.

Recovering phosphate and energy from anaerobic sludge digested wastewater with iron-air fuel cells: Two-chamber cell versus one-chamber cell.

Anaerobic sludge digested (ASD) wastewater is widespread in wastewater treatment plants. Recovering phosphate from ASD wastewater not only removes pollutants but also solves the phosphorus deficiency problem. Iron-air fuel cells were chosen to recover phosphate and generate electricity from ASD wastewater. To optimize cell configuration, a two-chamber and a one-chamber iron-air fuel cell were set up. The phosphate removal efficiency, the vivianite yield and the electricity generation efficiency of the two fuel cells were evaluated. It turned out that the volumetric removal rate (VRR) of phosphate of the two-chamber cell was 11.60 mg P·L-1·h-1, which was about five times of that in the one-chamber cell. The phosphate recovery product vivianite was detected on the surface of the iron anodes and the calculated purities of the two-chamber fuel cell and one-chamber fuel cell were 90.6% and 58.7%, respectively. Considering the content and purity, the iron anode surface in the two-chamber fuel cell was the best point to recover phosphate. The proton exchange membrane (PEM) in the two-chamber fuel cell provided low pH conditions suitable for vivianite formation. Moreover, under the low pH condition, metal ions of Fe2+, Ca2+, Al3+ and so on were kept soluble, leading to a high conductivity. The high conductivity caused low internal resistance, which benefited the electricity generation. The total output electric power of the two-chamber fuel cell was 2.4 times that of the one-chamber fuel cell when treating 25 mL ASD wastewater (0.62 vs. 0.26 mW·h). Overall, the two-chamber fuel cell was the better choice for phosphate recovery and electricity generation from ASD wastewater. Further studies on the long-term operation of two-chamber fuel cells should be carried out.

Response difference of simultaneous sulfide and nitrite removal process to different cooling modes.

The effects of various cooling modes (sudden cooling (25℃→10℃) and step cooling (25℃→20℃→15℃→10℃)) on the performance of simultaneous sulfide and nitrite removal process were reported. Regardless of cooling mode adopted, the process maintained good sulfide removal performance, and removal percentage was 100.00%. Considering nitrite removal percentage, the process was more sensitive to step cooling mode (k = 0.06707) in comparison to sudden cooling mode (k = 0.02760). Lowering temperature promoted the transformation from sulfate to elemental sulfur, and it was easier to increase the proportion of elemental sulfur (79.90%) by means of step cooling. The sulfide oxidation rate and nitrite reduction rate were 0.01540 mg /(L∙min) and 0.00354 mg /(L∙min), respectively, in the sudden cooling mode, and 0.01168 mg /(L∙min) and 0.00138 mg /(L∙min), respectively, in the step cooling mode. Low temperature reduced the diversity of microbial community, and Sulfurovum was still a dominant bacterial member in both cooling modes.

Production and characteristics of elemental sulfur during simultaneous nitrate and sulfide removal.

The production and characteristics of elemental sulfur were examined during simultaneous sulfide and nitrate removal, with abiotic assays as control. The biotic assay showed good sulfide and nitrate removal, with the respective removal percentage of which were 90.67-96.88% and 100%. Nitrate reduction resulted in the production of nitrogen gas, while sulfate formed due to sulfide oxidation. The concentration of elemental sulfur in the effluent was greater than that in the sludge, which accounted for 73.70-86.28% of total elemental sulfur produced. Furthermore, the elemental sulfur of the effluent and sludge from the biotic assays was orthorhombic crystal S8. Elemental sulfur was normally distributed in the effluent, but its average diameter increased with the increasing influent sulfide concentration (60-300 mg S/L), where the average diameter increased from 10 (60 mg S/L) to 29 µm (300 mg S/L).

Insights into microbial community in microbial fuel cells simultaneously treating sulfide and nitrate under external resistance.

The effect of electricity, induced by external resistance, on microbial community performance is investigated in Microbial Fuel Cells (MFCs) involved in simultaneous biotransformation of sulfide and nitrate. In the experiment, three MFCs were operated under different external resistances (100 Ω, 1000 Ω and 10,000 Ω), while one MFC was operated with open circuit as control. All MFCs demonstrate good capacity for simultaneous sulfide and nitrate biotransformation regardless of external resistance. MFCs present similar voltage profile; however, the output voltage has positive relationship with external resistance, and the MFC1 with lowest external resistance (100 Ω) generated highest power density. High-throughput sequencing confirms that taxonomic distribution of suspended sludge in anode chamber encompass phylum level to genus level, while the results of principal component analysis (PCA) suggest that microbial communities are varied with external resistance, which external resistance caused the change of electricity generation and substrate removal at the same, and then leads to the change of microbial communities. However, based on Pearson correlation analyses, no strong correlation is evident between community diversity indices (ACE index, Chao index, Shannon index and Simpson index) and the electricity (final voltage and current density). It is inferred that the performance of electricity did not significantly affect the diversity of microbial communities in MFCs biotransforming sulfide and nitrate simultaneously.

Coupled substrate removal and electricity generation in microbial fuel cells simultaneously treating sulfide and nitrate at various influent sulfide to nitrate ratios.

The current work coupled simultaneous sulfide and nitrate removal in a Microbial Fuel Cell (MFC). The substrate removal and electricity generation were coupled at influent Sulfide to Nitrate molar ratios (S/N ratios) of 5:0, 5:1, 5:2 and 5:3. The sulfide concentrations used included: 60 mg S/L, 300 mg S/L, 540 mg S/L, 780 mg S/L and 1020 mg S/L. The effect of S/N ratio on the performance of substrate removal was greater at higher influent sulfide concentration. The electricity generation also varied at different influent sulfide concentrations and S/N ratios. The number of electrons generated at S/N ratio of 5:2 was the largest at any fixed influent sulfide concentration. The Pearson correlation showed that effluent sulfate concentration and nitrogen gas had significant positive correlations with steady state voltage (or electronic quantity). Moreover, the simulation models were developed to establish the relation between substrate removal and electricity generation at various S/N ratios.

Effect of external resistance on substrate removal and electricity generation in microbial fuel cell treating sulfide and nitrate simultaneously.

The effect of external resistance on substrate removal and electricity generation was explored in microbial fuel cells (MFCs) simultaneously treating sulfide and nitrate. The MFCs were operated under three different conditions keeping open-circuit MFC as control. In batch mode, all the MFCs showed good capacity of simultaneously removing sulfide and nitrate regardless of external resistance. The voltage profile could be divided into rapid descent zone, bulge zone, and stability zone, which represents typical polarization behavior. Taking open circuit as control, low external resistance promoted the production of sulfate and nitrogen gas, while a strong link between product production and external resistance was evident based on Pearson correlation analyses. In addition, low external resistance improved the amount of transferred electrons, while the peak electronic quantity was noticed when the external resistance was equivalent to internal resistance. Moreover, the mechanism of substrate removal and electricity generation was hypothesized for the MFCs simultaneously treating sulfide and nitrate which explained the results well.

Prediction and quantifying parameter importance in simultaneous anaerobic sulfide and nitrate removal process using artificial neural network.

The present investigation deals with the prediction of the performance of simultaneous anaerobic sulfide and nitrate removal in an upflow anaerobic sludge bed (UASB) reactor through an artificial neural network (ANN). Influent sulfide concentration, influent nitrate concentration, S/N mole ratio, pH, and hydraulic retention time (HRT) for 144 days' steady-state condition were the inputs of the model; whereas output parameters were sulfide removal percentage, nitrate removal percentage, sulfate production percentage, and nitrogen production percentage. The prediction performance was evaluated by calculating root mean square error (RMSE), mean absolute error (MAE), mean absolute relative error (MARE), and determination coefficient (R (2)) values. Generally, the ANN model exhibited good prediction of the simultaneous sulfide and nitrate removal process. The effect of five input parameters to the performance of the reactor was quantified and compared using the connection weights method, Garson's algorithm method, and partial derivatives (PaD) method. The results showed that HRT markedly affects the performance of the reactor.

Effect of operating modes on simultaneous anaerobic sulfide and nitrate removal in microbial fuel cell.

The effect of operating modes on the simultaneous sulfide and nitrate removal were studied in two-chamber microbial fuel cells (MFCs). The batch and continuous operating modes were compared and evaluated in terms of substrate removal and electricity generation. Upon gradual increase in the influent sulfide concentration from 60 to 1,020 S mg L(-1), and the hydraulic retention time decrease from 17.2 to 6 h, the MFC accomplished a good substrate removal efficiency whereby nitrogen and sulfate were the main end products. The removal efficiency of the MFC in the continuous mode was much higher than that in the batch mode, and its current densities in the continuous mode were more stable and higher than in the batch mode, which could be explained by the linear relationship between electrons released by the substrates and accepted on the electrodes. The electricity output in the continuous mode of the MFC was higher than that in the batch mode. MFC's operation in the continuous mode was a better strategy for the simultaneous treatment of sulfide and nitrate.

Dynamic behavior and concentration distribution of granular sludge in a super-high-rate spiral anaerobic bioreactor.

Dynamic behavior and concentration distribution of granular sludge is highly dependent on the ecological environment of microbial communities and substrate degradation efficiency along bed height. Both were modeled and verified through experiments in a super-high-rate spiral anaerobic bioreactor (SSAB). The sludge transport efficiency of upmoving biogas (k(t),(n)(-1)) displaying dynamic behavior of granular sludge in SSAB were predicted and found to be much lesser than of upflow anaerobic sludge blanket (UASB). The bed concentration distribution (C(m),(n)(-1)/C(m),(n)) which represented concentration distribution of granular sludge were also quantitatively predicted in two feeding strategies. Parametric sensitivity suggested that k(t),(n)(-1) was significantly influenced by spiral angle, outer radii of spiral rectangular channel, settling velocity of granular sludge and superficial liquid velocity (v(l)), while C(m),(n)(-1)/C(m),(n) was affected by v(l) and superficial biogas velocity. In addition, some measures were also suggested to optimize designs and operations of such bioreactors.

Bed expansion behavior and sensitivity analysis for super-high-rate anaerobic bioreactor.

Bed expansion behavior and sensitivity analysis for super-high-rate anaerobic bioreactor (SAB) were performed based on bed expansion ratio (E), maximum bed sludge content (V(pmax)), and maximum bed contact time between sludge and liquid (tau(max)). Bed expansion behavior models were established under bed unfluidization, fluidization, and transportation states. Under unfluidization state, E was 0, V(pmax) was 4 867 ml, and tau(max) was 844-3 800 s. Under fluidization state, E, V(pmax), and tau(max) were 5.28%-255.69%, 1 368-4 559 ml, and 104-732 s, respectively. Under transportation state, washout of granular sludge occurred and destabilized the SAB. During stable running of SAB under fluidization state, E correlated positively with superficial gas and liquid velocities (u(g) and u(l)), while V(pmax) and tau(max) correlated negatively. For E and V(pmax), the sensitivities of u(g) and u(l) were close to each other, while for tau(max), the sensitivity of u(l) was greater than that of u(g). The prediction from these models was a close match to the experimental data.

Quantitative comparison of stability of ANAMMOX process in different reactor configurations.

A new stability evaluating system for ANAMMOX comprising three instability indices i.e. coefficient of variation ratio, coefficient of range ratio and coefficient of regression function derivative was established. Three lab-scale ANAMMOX reactors viz upflow anaerobic sludge blanket (UASB) reactor, upflow stationary fixed film (USFF) reactor and anaerobic sequencing batch reactor (ASBR) were compared for their stability based on the established criterion against the hydraulic and substrate concentration shocks. The results showed that all ANAMMOX reactors under investigation were more tolerant to the hydraulic shock than substrate concentration shock. The UASB reactor was the most stable reactor configuration towards substrate concentration shock, followed by the USFF reactor and ASBR. However, the ASBR proved the most tolerant to hydraulic shock, followed by the UASB reactor and USFF reactor. In terms of stability, UASB reactor was more suitable configuration compared with USFF reactor. The instability indices proved to be effective and explicit for the evaluation of ANAMMOX systems.

Filamentous granular sludge bulking in a laboratory scale UASB reactor.

Filamentous bulking was observed in a lab scale upflow anaerobic sludge blanket (UASB) reactor. Granules failed to settle normally and disintegrated. The characteristics of the granules in structure and microbial composition during the granulation process were investigated by means of scanning electron microscopy (SEM) and denaturing gradient gel electrophoresis (DGGE) technique. Granules with high porosity instead of compact ones were developed in the reactor and Methanosaeta concilii and Methanobacterium formicicum were identified as the predominant methanogens present in granules. The excess growth of the filamentous bacteria could be the contributing factor causing floatation and disintegration.

Anoxic biodegradation of dimethyl phthalate (DMP) by activated sludge cultures under nitrate-reducing conditions.

Worldwide extensive use of plasticized plastics has resulted in phthalates pollution in different environment. Nitrates from industry and agriculture are also widely disseminated in the soils, natural waters and wastewaters. Dimethyl phthalate (DMP) biodegradation by activated sludge cultures under nitrate-reducing conditions was investigated. Under one optimized condition, DMP was biodegraded from 102.20 mg/L to undetectable level in 56 h under anoxic conditions and its reaction fitted well with the first-order kinetics. Using the high-performance liquid chromatography (HPLC) and liquid chromatography mass spectrometry (LC-MS) analysis, mono-methyl phthalate (MMP) and phthalic acid (PA) were detected as the major intermediates of DMP biodegradation. When combined with the determination of chemical oxygen demand (COD(Cr)) removal capacity and pH, DMP was found to be mineralized completely under anoxic conditions. The biodegradation pathway was proposed as DMP -->MMP-->PA-->...-->CO2 + H2O. The molar ratio ofDMP to nitrate consumed was found to be 9.0:1, which agrees well with the theoretical stoichiometric values of DMP biodegradation by nitrate-reducing bacteria. The results of the non-linear simulation showed that the optimum pH and temperature for the degradation were 7.56 and 31.4 degrees C, respectively.

Production and application of anaerobic granular sludge produced by landfill.

Sludge granulation is considered to be the most critical parameter governing successful operation of an upflow anaerobic sludge blanket and expanded granular sludge bed (EGSB) reactors. Pre-granulated seeding sludge could greatly reduce the required startup time. Two lab-scale and a pilot-scale EGSB reactors were operated to treat Shaoxing Wastewater Treatment Plant (SWWTP) containing wastewater from real engineering printing and dyeing with high pH and sulfate concentration. The microbiological structure and the particle size distribution in aerobic excess sludge, sanitary landfill sludge digested for one year, and the granular sludge of EGSB reactor after 400 d of operation were analyzed through scanning electron microscopy (SEM) and sieves. The lab-scale EGSB reactor seeded with anaerobic sludge after digestion for one year in landfill showed obviously better total chemical oxygen demand (TCOD) removal efficiency than one seeded with aerobic excess sludge after cation polyacrylamide flocculation-concentration and dehydration. The TCOD removed was 470.8 mg/L in pilot scale EGSB reactor at short hydraulic retention time of 15 h. SEM of sludge granules showed that the microbiological structure of the sludge from different sources showed some differences. SEM demonstrated that Methanobacterium sp. was present in the granules of pilot-scale EGSB and the granular sludge produced by landfill contained a mixture of anaerobic/anoxic organisms in abundance. The particle size distribution in EGSB demonstrated that using anaerobic granular sludge produced by sanitary landfill as the seeding granular sludge was feasible.

[Isolation, identification and characterization of nitrate-reducing and sulfide-oxidizing bacteria].

Bacterial strains CB and CS were isolated from the Anoxic Sulfide Oxidizing (ASO) reactor working under steady-state with sulfide and nitrate as electron donor and acceptor, respectively. Based on electron microscopy, physiological test and 16S rDNA phylogenetic sequence analysis, the isolate CB was very close to Bacillus pseudofirmus and the isolate CS was very close to Bacillus hemicellulosilytus and Bacillus halodurans. The 16S rDNA sequences of both isolates were submitted to GenBank, NCBI and accession numbers EF542806 and EF542807 were allotted for isolates CB and CS, respectively. According to Biolog carbon source utilization test, the isolate CB is weak to use the carbon sources, while the isolate CB can use many carbon sources. Both the isolate CB and the isolate CS are able to use nitrate for sulfide oxidation. The isolate CS has greater capability to oxidize sulfide with nitrate as electron acceptor.

Assessing potential dietary toxicity of heavy metals in selected vegetables and food crops / 浙江大学学报（英文版）（B辑：生物医学和生物技术）

Heavy metals, such as cadmium, copper, lead, chromium and mercury, are important environmental pollutants, particularly in areas with high anthropogenic pressure. Their presence in the atmosphere, soil and water, even in traces can cause serious problems to all organisms, and heavy metal bioaccumulation in the food chain especially can be highly dangerous to human health. Heavy metals enter the human body mainly through two routes namely: inhalation and ingestion, ingestion being the main route of exposure to these elements in human population. Heavy metals intake by human populations through food chain has been reported in many countries. Soil threshold for heavy metal toxicity is an important factor affecting soil environmental capacity of heavy metal and determines heavy metal cumulative loading limits. For soil-plant system, heavy metal toxicity threshold is the highest permissible content in the soil (total or bioavailable concentration) that does not pose any phytotoxic effects or heavy metals in the edible parts of the crops does not exceed food hygiene standards. Factors affecting the thresholds of dietary toxicity of heavy metal in soil-crop system include: soil type which includes soil pH, organic matter content, clay mineral and other soil chemical and biochemical properties; and crop species or cultivars regulated by genetic basis for heavy metal transport and accumulation in plants. In addition, the interactions of soil-plant root-microbes play important roles in regulating heavy metal movement from soil to the edible parts of crops. Agronomic practices such as fertilizer and water managements as well as crop rotation system can affect bioavailability and crop accumulation of heavy metals, thus influencing the thresholds for assessing dietary toxicity of heavy metals in the food chain. This paper reviews the phytotoxic effects and bioaccumulation of heavy metals in vegetables and food crops and assesses soil heavy metal thresholds for potential dietary toxicity.

Prediction of anoxic sulfide biooxidation under various HRTs using artificial neural networks / 生物医学与环境科学（英文）

<p><b>OBJECTIVE</b>During present investigation the data of a laboratory-scale anoxic sulfide oxidizing (ASO) reactor were used in a neural network system to predict its performance.</p><p><b>METHODS</b>Five uncorrelated components of the influent wastewater were used as the artificial neural network model input to predict the output of the effluent using back-propagation and general regression algorithms. The best prediction performance is achieved when the data are preprocessed using principal components analysis (PCA) before they are fed to a back propagated neural network.</p><p><b>RESULTS</b>Within the range of experimental conditions tested, it was concluded that the ANN model gave predictable results for nitrite removal from wastewater through ASO process. The model did not predict the formation of sulfate to an acceptable manner.</p><p><b>CONCLUSION</b>Apart from experimentation, ANN model can help to simulate the results of such experiments in finding the best optimal choice for ASObased denitrification. Together with wastewater collection and the use of improved treatment systems and new technologies, better control of wastewater treatment plant (WTP) can lead to more effective maneuvers by its operators and, as a consequence, better effluent quality.</p>