Journal of Environmental Engineering

Membrane and reactor performance for an anaerobic membrane bioreactor (AnMBR) treating a high-strength starch-based wastewater [average chemical oxygen demand (COD) of $84.8\pm 8.9\mathrm{g}/\mathrm{L}$] was examined under six varied cleaning protocols to find optimal operating conditions. The AnMBR was composed of a 1,000-L anaerobic continuous stirred tank reactor (CSTR) integrated with an external, tubular nanofiltration membrane module operated in cross-flow mode. The polyvinylidene fluoride–based membrane module had a nominal pore size of 30 nm, a total membrane area of $0.42{\mathrm{m}}^2$, and a cross-flow velocity of $2.5\mathrm{m}/\mathrm{s}$. The different cleaning protocols were provided by varying the membrane’s permeate backwash duration, frequency, and flow rate and by incorporating chemically enhanced backwash cycles with the use of a mixed 250-ppm sodium hydroxide and 250-ppm hypochlorite solution. Reactor performance did not appear to be cleaning protocol dependent; however, a decline in reactor-specific methanogenic activity (SMA) was noticed over the duration of the study. A single, longer-duration chemically enhanced backwash offered the highest improvement in membrane performance over all other cleaning protocols.

This study focused on the selective recovery of zinc (Zn) from a leaching solution emerging from a sulfuric acid leaching process applied to unsorted spent batteries. Precipitation and solvent extraction were investigated. According to the results, solvent extraction using Cyanex 272 allowed for the selective removal of Zn from the solution containing high amounts of metals ($\sim 19.4\mathrm{g}\mathrm{Zn}/\mathrm{L}$, $\sim 23.4\mathrm{g}\mathrm{Mn}/\mathrm{L}$, $\sim 3.27\mathrm{g}\mathrm{Cd}/\mathrm{L}$, $\sim 3.19\mathrm{g}\mathrm{Ni}/\mathrm{L}$, and $\sim 0.25\mathrm{g}\mathrm{Co}/\mathrm{L}$). According to the results, the solvent extraction process was capable of recovering 97.6% of Zn from this leaching solution under the following conditions: two stages of extraction in the presence of an organic solution made of Cyanex 272 (30%, v/v) and tributylphosphate (TBP—2%, v/v) in kerosene, $\mathrm{pH}=2.2$, organic/aqueous (O/A) ratio = 2/1, and $\mathrm{T}=50°\mathrm{C}$. The Zn present in the organic phase was then stripped using 0.4 M ${\mathrm{H}}_2{\mathrm{SO}}_4$ with an O/A ratio fixed at 2/1. This stripping step allowed for the recovery of 81.8% of the Zn initially present in the organic phase. Subsequently, 82.4% of the Zn stripped in the aqueous solution was then electrically deposited after 3 h at $\mathrm{pH}=2$ with a current density fixed at $360\mathrm{A}/{\mathrm{m}}^2$.

Bioretention systems have become an increasingly common method for treating stormwater in urban areas, which help reduce peak flows and remove contaminants from stormwater. However, nutrients often leach out of the bioretention soil mix, which can contribute to the degradation of receiving waters in bioretention systems with underdrains. Development of mycelium may improve retention of nutrients and increase the water-holding capacity. To evaluate the impact of mycelium on nutrient leaching from bioretention systems, ectomycorrhizal and endomycorrhizal fungi were added to the bioretention soil mix to promote mycelium growth. A proprietary mix with bacteria and mycorrhizal fungi also was tested. Mesocosms were planted with Carex stipata, a native sedge with endomycorrhizal associations. Four tests were conducted with collected stormwater. Lower rates of phosphorus export were observed in mescocosms with mycorrhizal fungi; the export of total phosphorus was reduced by 13–48%, and the export of phosphate was reduced by 14–60%. There also was evidence of additional copper and nitrate uptake in mesocosms with mycorrhizal fungi. Retention of total phosphorus and phosphate, rather than export, was observed in mesocosms with the proprietary mix, but export rates of nitrate were high. This study indicates that mycelium may help reduce phosphorus export from bioretention systems.

A facultative bacterium was isolated from the sludge of a local dyeing industry for bioremediation of wastewater containing azo dyes. The identified Klebsiella pneumoniae by 16S ribosomal DNA (rDNA) sequence analysis was used for complete degradation of the recalcitrant sulfonated azo dyes in a two-stage process. In Stage I, different azo dyes (monosulfonated, disulfonated, and trisulfonated, $100\mathrm{mg}{\mathrm{L}}^{-1}$), were decolorized by more than 95% in less than 24 h. The addition of nicotinamide adenine dinucleotide hydrogen as a redox mediator resulted in faster decolorization of the model azo dye, Methyl Orange. In Stage II, the accumulated 4-aminobenzenesulfonic acid was further degraded under aerobic conditions. Dye degradation products obtained during biodegradation were analyzed by ultraviolet–visible spectroscopy and high-performance liquid chromatography. The phytotoxicity test performed with the dicot seed Cicer arietinum showed that the final degraded products after two-stage sequential process were less toxic. Moreover, the isolated Klebsiella pneumoniae was observed to be efficient compared with the previously reported strains. Thus, the isolated strain has great potential for the degradation of azo dyes and aromatic amines at the source points.

To evaluate two practical coking wastewater treatment systems, the removal efficacy of major pollutants was compared and bacterial profiles were analyzed using high-throughput sequencing. The A/O (anoxic/oxic) process (named as 1 system) used the activated sludge method, whereas the ${\mathrm{O}}_1/\mathrm{A}/{\mathrm{O}}_2$ (oxic/anoxic/oxic) process (named as 2 system) adopted the biocontact oxidation method, and the latter added an oil-removal tank and a preaeration tank. The oil-removal tank removed 37.0% of the oil, and phenol and ${\mathrm{CN}}^{-}$ were decreased to 398 and $17\mathrm{mg}/\mathrm{L}$, respectively. For the final effluent of the ${\mathrm{O}}_1/\mathrm{A}/{\mathrm{O}}_2$ process, the concentrations of chemical oxygen demand (COD), volatile phenol, total nitrogen, and total ${\mathrm{CN}}^{-}$ were low at 121, 0.09, 45.1, and $0.18\mathrm{m}\mathrm{g}/\mathrm{L}$, respectively. The $2\text{-}{\mathrm{O}}_1$ and $2\text{-}{\mathrm{O}}_2$ tanks presented higher bacterial diversity and lower species richness than the 1-O tank; the most abundant phylum in five biological tanks were Proteobacteria; and Nitrospirae had high richness in the oxic tanks. The microbial diversity and principal component analysis (PCA) revealed that different biological treatment processes and operation modes have significant impacts on the microbial population and species richness. Nitrifying and denitrifying bacteria, such as Nitrosomonas, Nitrospira, Thauera, and Thiobacillus, were relatively abundant, contributing to nitrogen removal.

Previous research has demonstrated that biological phosphorus removal (BPR) could be induced by an anoxic-aerobic sequencing batch reactor (SBR). This study applies a continuous-flow anoxic-oxic (AO) reactor at laboratory scale. Synthetic wastewater is used as feed water, in which soluble starch is the sole organic substance added as a carbon source. Effects of dissolved oxygen (DO) concentration on the BPR in the continuous-flow AO process are investigated. Results show that P removal is achieved mainly by denitrifying phosphate-accumulating organisms (DPAOs) without a specific anaerobic period in the system. Glycogen is the main energy source for phosphorous removal in this study, which is different from that of traditional biological phosphorous removal. The DO concentration in the aerobic unit affects anoxic P removal efficiency. Phosphorus removal efficiency as high as 82% is achieved at a lower DO level ($0.5\mathrm{mg}/\mathrm{L}$) due to the presence of more nitrate as an electron acceptor for P uptake. Polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) analysis proves that the microbial community structure changes significantly due to the change of the DO level. Acidobacterium and Proteobacterium play dominant roles in BPR. This study confirms that bacteria using starch can achieve high N and P removal efficiency, and a controlled concentration of dissolved oxygen is of benefit to the BNR of the continuous-flow AO system.

Presented here is a mechanistic model of the biological dynamics of the photic zone of a single-cell arctic waste stabilization pond (WSP) for the prediction of oxygen concentration and the removal of oxygen-demanding substances. The model is an exploratory model to assess the limiting environmental factors affecting treatment performance in arctic WSPs. A sensitivity analysis was used to provide a quantification of the relative uncertainties of parameters that exist within the described modeling framework. The model was able to qualitatively reproduce mesocosm experiment trends in phytoplankton growth, dissolved oxygen concentration, and the reduction of carbonaceous biochemical oxygen demand on Day 5 (${\mathrm{CBOD}}_5$). These results demonstrated that ${\mathrm{CBOD}}_5$ reduction and oxygen state are very sensitive to organic loading regimes at low temperatures (5–15°C). The sensitivity analysis identified that it was the difference in phytoplankton growth rates, and the associated change in photosynthetic oxygen production, that mainly contribute to creating differences in ${\mathrm{CBOD}}_5$ removal rates and the development of aerobic conditions. The model was also sensitive to atmospheric aeration rates at low temperature, providing further evidence that low oxygen availability limits the treatment of ${\mathrm{CBOD}}_5$ in cold-climate WSPs. During the development process, it was discovered that common formulations of depth-integrated phytoplankton growth performed poorly for the arctic system to be modeled, which was a quiescent eutrophic environment. This paper presents a new phytoplankton growth formula within the paradigm of a poorly mixed eutrophic system that may find utilization in other eutrophic, colored, or turbid systems. The novel aspect of the approach is that the depth-integrated phytoplankton growth function was formulated upon the premise that the phytoplankton would be capable of orienting themselves to optimize their growth under poorly mixed conditions, and the average growth rate of the phytoplankton population must decrease as crowding puts pressure on shared resources. The general agreement of the model with the experiments, combined with the simplicity of the depth-integrated box model, suggests there is potential for further development of the model as a tool for assessing proposed arctic WSP designs. The sensitivity analysis highlighted the uncertainty and importance of the parameterization of bacterial and phytoplankton physiology and metabolism in WSP models.

Regenerative stormwater conveyance (RSC) is a stormwater control measure (SCM) built in a channel that uses a series of riffles, grade control structures, and pools with an underlying sand media layer to detain and treat stormwater runoff. A purpose-built experimental field RSC was subjected to simulated hydrographs by mixing dissolved nitrogen and phosphorus into inflow from a managed retention pond. The system storage capacity of $30.5{\mathrm{m}}^2$ within the pools ($14.2{\mathrm{m}}^2$) and sand layer ($16.3{\mathrm{m}}^2$) was sized to store the runoff volume from the 25-mm event. Twelve storm simulations were conducted under varying conditions of storm depth (25 and 38 mm) and antecedent dry periods (1 day and 3 days). Simulations confirmed that RSC can convert large portions of flow volumes from surface to subsurface flow; mean reductions in surface flow volumes were 80% for the 25-mm event and 40% for the 38-mm event. Reductions in dissolved pollutant concentrations in surface flow were insignificant or very minor for all analytes. However, mean reductions in subsurface flow concentrations were high for both total Kjeldahl nitrogen (TKN, 79%) and orthophosphate (81%). Because of well-drained subsurface conditions, nitrification of TKN to nitrate was observed, leading to a 250% increase in nitrate concentration. Results of the simulation study indicate that microbial transformation and sorption mechanisms are present in RSC subsurface outflow. Subsurface flow contributed 66% of the total nitrogen load and 42% of the total phosphorus load for the 25-mm event. When adequately sized, RSCs can provide substantial hydrologic mitigation for urban drainage areas. As implementation of RSC continues, regulators and designers are encouraged to consider RSC subsurface flow as an important contribution to downstream pollutant load. Proper filter media selection and differentiated aerobic/anaerobic conditions are essential to optimize subsurface flow treatment.

In this work, a model was built to simulate the multiphase extraction (MPE) applied to soil polluted by toluene. The model is composed of three differential partial equations simulating the multiphase flow of the three present phases using the capillarity equations and four equations to simulate the transport and the transfer between the three phases. The model was applied to a hypothetical pollution at field scale. A first step was applied to simulate the toluene propagation in the soil in saturated and unsaturated zones during a period of 2 months and its distribution in different phases. Results show a decrease of nonaqueous-phase liquid (NAPL) residual saturation under the effect of mass transfers between phases. It was observed that the propagation by gas-phase diffusion was greater than liquid-phase diffusion. The MPE application shows the important effect of the strainer location and the applied vacuum on the velocity fields. The proposed model shows the limitation of the MPE method in the capillary zone, which could not be remediated after 1 year of treatment. This observation confirms the necessity to ameliorate the process by lowering or deepening the groundwater or by using several extraction wells. The mass transfer coefficients were estimated from the literature. This work also highlighted the importance of a good identification of parameters.

In this study, a process consisting of cooling crystallization, Fenton reactions, and a sequencing batch reactor was investigated for the treatment of organic wastewater with high chemical oxygen demand (COD) and sulfate from a chemical company. Cooling crystallization was first employed to remove sulfate from the wastewater. At 4.0°C, sulfate concentration was reduced from 212 to $96.0\mathrm{g}/\mathrm{L}$, and total wastewater volume decreased by 49.1% through two-stage cooling crystallization. COD concentration was increased from 18.0 to $18.5\mathrm{g}/\mathrm{L}$ due to the precipitation of ${\mathrm{Na}}_2{\mathrm{SO}}_4$ crystals mainly containing crystal water. The effluent was then treated by Fenton reactions. At pH 2.0, $8.0\mathrm{mmol}/\mathrm{L}$ of ferric/ferrous, $40\mathrm{mL}/\mathrm{L}$ of 30% ${\mathrm{H}}_2{\mathrm{O}}_2$, and a duration of reaction of 60 min, COD removal efficiency maximized at 77.9%, and COD concentration in the effluent leaving the Fenton reactions was $\mathrm{4,100}\mathrm{mg}/\mathrm{L}$. At last, the mixture of the effluent from the Fenton reactions and sewage at a volume ratio of $1:4$ was treated by a sequencing batch reactor (SBR) in which the microorganisms were acclimated to be sulfate tolerant. The efficiency of COD removal by the SBR steadily exceeded 85%, and COD concentration leaving the SBR was below $150\mathrm{mg}/\mathrm{L}$. These data showed that the three-step process was cost-effective for the treatment of high-salt organic wastewater.

Fecal indicator bacteria (FIB) are used as proxies to measure the microbial water quality of aquatic ecosystems. Methods of modeling FIB have evolved in order to provide accurate and timely prediction to inform decisions by governing authorities to prevent risks to public health. A predictive model to forecast the FIB concentrations of an urban waterway, the Chicago River, utilizing the artificial neural network (ANN) method was developed. To address tuning of hyperparameters of the ANN model, an exhaustive testing was performed to select optimal hyperparameters. The root-mean-square propagation (RMSprop) optimizer performed better than the stochastic gradient descent (SGD) and adaptive moment estimation (Adam) optimizers in this study. Eight input variables were eventually selected from 10 initially proposed variables: water temperature; turbidity; daily, 2-day, and 7-day cumulative rainfall; river flow discharge; distance from the upstream water reclamation plant; and number of upstream combined sewer outfalls. Water reclamation plants and combined sewer overflows were found to be critical contributors of microbial pollution in this urban waterway and should be considered in the ANN model. The developed model has an accuracy of 86.5% to predict whether fecal coliform concentration is above or below a regulatory threshold.