Cation-π Interactions Contribute to Hydrophobic Humic Acid Removal for the Control of Hydraulically Irreversible Membrane Fouling.

Hydraulically irreversible membrane fouling is a major problem encountered during membrane-based water purification. Membrane foulants present large hydrophobic fractions, with humic acid (HA) being a prevalent example of hydrophobic natural organic matter. Furthermore, HA contains numerous aromatic rings (π electrons), and its hydrophobic interactions are a major cause of irreversible membrane fouling. To address this issue, in this study, we used the cation-π interaction, which is a strong noncovalent, competitive interaction present in water. Because the strength of cation-π interactions depends on the combination of cations and π molecules, utilizing the appropriate cations will effectively remove irreversible fouling caused by hydrophobic HA. We performed macroscale experiments to determine the cleaning potential of the test cations, nanomechanically analyzed the changes in HA cohesion caused by the test cations using a surface force apparatus and an atomic force microscope, and used molecular dynamics simulations to elucidate the HA removal mechanism of test studied cations. We found that the addition of 1-ethyl-3-methylimidazolium, an imidazolium cation with an aromatic moiety, effectively removed the HA layer by weakening its cohesion, and the size, hydrophobicity, and polarity of the HA layer synergistically affected the HA removal mechanism based on the cation-π interactions.

Selective removal of color substances by carbon-based adsorbents in livestock wastewater effluents.

Livestock wastewater effluent generated after the anaerobic treatment process contains the considerable amount of color-causing organic matter. In this study, a quantitative comparison of three carbon-based adsorbents included granular activated carbon (GAC), expanded graphite (EG), and multi-walled carbon nanotubes (MWNTs) was carried out for the potential application to the removal of color substances, and their mechanism was proposed. Although GAC showed the highest specific dissolved organic carbon (DOC) adsorption capacity, the color removal efficiency was the smallest among three adsorbents. The selective color removal ratios of EG and MWNTs reached 22.7 ± 0.1 PtCo/mg-DOC-removed and 21.2 ± 0.1 PtCo/mg-DOC-removed, respectively, while that of GAC was only 12.3 ± 0.1 PtCo/mg-DOC-removed. The selective adsorption of color substances by graphene-based carbon materials was due to the aromatic π-π interaction between organic matter and the hexagonal carbon lattice of graphene. The analysis of molecular weight distribution also confirmed that the exposed surface area and macro-pores were responsible for the adsorption of high molecular weight color substances. The chemical regeneration of three adsorbents was examined using 1% NaOCl solution and MWNTs showed almost complete recovery of the initial color removal capacity. In conclusion, MWNTs were the most suitable carbon nanomaterial for the selective color removal from livestock wastewater effluent.

Increased biodegradability of low-grade coal wastewater in anaerobic membrane bioreactor by adding yeast wastes.

Demineralization is required in upgrading low-grade coal to serve as an alternative energy resource for the production of fuel and valuable chemicals but generates a large amount of low-grade coal wastewater (LCWW). The objective of this study was to investigate the effects of a co-substrate on an anaerobic membrane bioreactor (AnMBR) treating LCWW. CH4 was not produced during the operation fed by LCWW alone. When yeast wastes (YW) were supplemented, there was a gradual increase in the biodegradability of LCWW, achieving 182 CH4 mL/g COD with 58% COD removal efficiency. The analysis of physicochemical characteristics in the effluent of AnMBR, done by excitation-emission matrix (EEM) and size exclusion chromatography (SEC), showed that the proportion of soluble microbial products (SMPs) and aromatic group with high-molecular weight (>1 kDa) increased. Microbial analysis revealed that the increased dominance of bacteria Comamonas, Methanococcus, and Methanosarcina facilitated biodegradation of LCWW in the presence of YW.

Dissolved organic matter characterization of biochars produced from different feedstock materials.

Fluorescence excitation-emission matrix (EEM) spectroscopy coupled with parallel factor analysis (PARAFAC) enables better understanding of the nature of dissolved organic matter (DOM). In the current study, we characterized 10 biochar samples produced from different feedstocks using EEM/PARAFAC analysis. The composition and distribution of DOM substances present in biochar varied significantly according to feedstock, activation, and pyrolysis temperature. The integration of proximate and ultimate analyses of the solid phase together with water extractable organic matter (WEOM) phase of biochar provided new insights into the characterization of biochars, including nature and functionality. Characterization of both WEOM and solid phases is recommended for biochar research before large-scale production for various environmental and industrial applications.

Impact of polymeric membrane filtration of oil sands process water on organic compounds quantification.

The interaction between organic fractions in oil sands process-affected water (OSPW) and three polymeric membranes with varying hydrophilicity (nylon, polyvinylidene fluoride and polytetrafluoroethylene) at different pHs was studied to evaluate the impact of filtration on the quantification of acid-extractable fraction (AEF) and naphthenic acids (NAs). Four functional groups predominated in OSPW (amine, phosphoryl, carboxyl and hydroxyl) as indicated by the linear programming method. The nylon membranes were the most hydrophilic and exhibited the lowest AEF removal at pH of 8.7. However, the adsorption of AEF on the membranes increased as the pH of OSPW decreased due to hydrophobic interactions between the membrane surfaces and the protonated molecules. The use of ultra pressure liquid chromatography-high resolution mass spectrometry (UPLC/HRMS) showed insignificant adsorption of NAs on the tested membranes at pH 8.7. However, 26±2.4% adsorption of NAs was observed at pH 5.3 following the protonation of NAs species. For the nylon membrane, excessive carboxylic acids in the commercial NAs caused the formation of negatively charged assisted hydrogen bonds, resulting in increased adsorption at pH 8.2 (25%) as compared to OSPW (0%). The use of membranes for filtration of soluble compounds from complex oily wastewaters before quantification analysis of AEF and NAs should be examined prior to application.

Different susceptibilities of bacterial community to silver nanoparticles in wastewater treatment systems.

The release of silver (Ag) nanoparticles (NPs) into sewage streams has heightened concerns about potential adverse impacts on wastewater treatment processes. Here, we show that the rate constants of both biological nitrification and organic oxidation decreased exponentially with an increase in the Ag NP concentration, but nitrification was more severely inhibited than the organic oxidation even at low Ag NP concentrations (<1 mg Ag L(-1)) in batch experiments. The long-term exposure effects of Ag NPs on activated sludge bacteria were evaluated in sequencing batch reactors (SBRs) fed with two different substrates favoring heterotrophic and autotrophic bacteria. From a continuous operation for 50 days, it was found that heterotrophic bacteria in the organic removal process have higher tolerance to Ag NPs than do nitrifying bacteria. The effects of Ag NPs on the microbial community in both SBRs were analyzed using 16S ribosomal ribonucleic acid (rRNA) gene sequences obtained from pyrosequencing. The results showed that the level of microbial susceptibility is different for each type of microorganism and that the microbial diversity decreased dramatically after continuous exposure to Ag NPs for 50 days, resulting in a decrease of wastewater treatment efficiency.

Removal of urea in ultrapure water system by urease-coated reverse osmosis membrane.

Among the various substances found in the feed source for the production of ultrapure water (UPW), urea is challenging to remove because it is a small molecular weight molecule that is not easily oxidized and does not carry a charge under neutral pH conditions. Urease enzyme, found in various organisms such as plants and bacteria, catalyze the hydrolysis of urea into carbon dioxide and ammonia. In this study, urease was immobilized on the polyamide layer of a reverse osmosis (RO) membrane to remove urea in UPW systems. The removal efficiency of urea by urease-coated RO membrane showed up to 27.9 % higher urea removal efficiency compared to the pristine membrane. This increase in urea removal can be attributed to both physical and biological effects from the urease coating on the membrane. Firstly, urease on the membrane surface can act as an additional physical barrier for urea to pass through. Secondly, urea can be hydrolyzed by the enzyme when it passes through the urease-coated RO membrane. In a two-pass RO system typical for UPW production, the removal of urea by a urease-coated membrane would be enhanced by twofold. This overall method can significantly increase the removal efficiency of urea in UPW systems, especially when considering the compounded removal by the urease coating, rejection by RO, and additional reactions by other treatment processes. Moreover, urea in UPW systems can be removed without the installment of additional processes by simply coating urease on the existing RO membranes.

Adsorptive chito-beads for control of membrane fouling.

Chitosan has excellent antimicrobial, adsorption, heavy metal removal, and adhesion properties, making it a good substitute for microplastic-based cleaners. Here, chitosan microbeads (chito-beads) of various sizes ranging from 32 µm to 283 µm were prepared via emulsion using a liquid on oil method and the feasibility of using them as an essential constituent in a chemical cleaning solution for a reverse-osmosis (RO) membrane-fouling-control process was assessed. Prior to the assessment the cleaning efficiency of a solution containing chito-beads, the interaction energy between chitosan and a representative organic foulant (humic acid (HA)) in a RO membrane fouling was analyzed using colloidal atomic force microscopy, and the strongest attraction between chitosan and HA was observed in an aqueous solution. When comparing the membrane cleaning efficiency of cleaning solutions with and without chito-beads, smaller chito-beads (32 µm and 70 µm) were found to have higher cleaning efficiency. Applications of chito-beads to the membrane cleaning process can enhance the cleaning efficiency through the physicochemical interaction with organic foulants. This study can widen the use of chito-beads as an additive to membrane chemical cleaning solutions to control membrane fouling in other membrane processes as well.

Boosted micropollutant removal over urchin-like structured hydroxyapatite-incorporated nickel magnetite catalyst via peroxydisulfate activation.

In this work, urchin-like structured hydroxyapatite-incorporated nickel magnetite (NiFe3O4/UHdA) microspheres were developed for the efficient removal of micropollutants (MPs) via peroxydisulfate (PDS) activation. The prepared NiFe3O4/UHdA degraded 99.0 % of sulfamethoxazole (SMX) after 15 min in 2 mM PDS, having a first-order kinetic rate constant of 0.210 min-1. In addition, NiFe3O4/UHdA outperformed its counterparts, i.e., Fe3O4/UHdA and Ni/UHdA, by giving rise to corresponding 3.6-fold and 8.6-fold enhancements in the SMX removal rate. The outstanding catalytic performance can be ascribed to (1) the urchin-like mesoporous structure with a large specific surface area and (2) the remarkable synergistic effect caused by the redox cycle of Ni3+/Ni2+ and Fe2+/Fe3+ that enhances multipath electron transfers on the surface of NiFe3O4/UHdA to produce more reactive oxygen species. Moreover, the effects of several reaction parameters, in this case the initial solution pH, PDS dosage, SMX concentration, catalyst loading, co-existing MPs and humic acid level on the catalytic performance of the NiFe3O4/UHdA + PDS system were systematically investigated and discussed in detail. The plausible catalytic mechanisms in the NiFe3O4/UHdA + PDS system were revealed via scavenging experiments and electron paramagnetic resonance analysis, which indicated a radical (•OH and SO4•-) as the major pathway and a nonradical (1O2) as the minor pathway for SMX degradation. Furthermore, NiFe3O4/UHdA exhibited fantastic magnetically separation and retained good catalytic activity with a low leached ion concentration during the performance of four cycles. Overall, the prepared NiFe3O4/UHdA with outstanding PDS activation could be a promising choice for the degradation of persistent organic pollutants from wastewater.

Optimizing coagulant dosage using deep learning models with large-scale data.

Water treatment plants are facing challenges that necessitate transition to automated processes using advanced technologies. This study introduces a novel approach to optimize coagulant dosage in water treatment processes by employing a deep learning model. The study utilized minute-by-minute data monitored in real time over a span of five years, marking the first attempt in drinking water process modeling to leverage such a comprehensive dataset. The deep learning model integrates a one-dimensional convolutional neural network (Conv1D) and gated recurrent unit (GRU) to effectively extract features and model complex time-series data. Initially, the model predicted coagulant dosage and sedimentation basin turbidity, validated against a physicochemical model. Subsequently, the model optimized coagulant dosage in two ways: 1) maintaining sedimentation basin turbidity below the 1.0 NTU guideline, and 2) analyzing changes in sedimentation basin turbidity resulting from reduced coagulant dosage (5-20%). The findings of the study highlight the effectiveness of the deep learning model in optimizing coagulant dosage with substantial reductions in coagulant dosage (approximately 22% reduction and 21 million KRW/year). The results demonstrate the potential of deep learning models in enhancing the efficiency and cost-effectiveness of water treatment processes, ultimately facilitating process automation.

Bacteria-polymeric membrane interactions: atomic force microscopy and XDLVO predictions.

Atomic force microscopy (AFM) in conjunction with a bioprobe developed using a polydopamine wet adhesive was used to directly measure the adhesive force between bacteria and different polymeric membrane surfaces. Bacterial cells of Pseudomonas putida and Bacillus subtilis were immobilized onto the tip of a standard AFM cantilever, and force measurements made using the modified cantilever on various membranes. Interaction forces measured with the bacterial probe were compared, qualitatively, to predictions by the extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) theory with steric interactions included. The XDLVO theory predicted attractive interactions between low energy hydrophobic membranes with high energy hydrophilic bacterium (P. putida). It also predicted a shallow primary maximum with the most hydrophilic bacterium, B. subtilis . Discrepancies between predictions using the XDLVO theory and theory require involvement of factors such as bridging effects. Differences in interaction between P. putida and B. subtilis are attributed to acid-base interactions and steric interactions. P. putida is Gram negative with lipopolysaccharides present in the outer cell membrane. A variation in forces of adhesion for bacteria on polymeric membranes studied was interpreted in terms of hydrophilicity and interfacial surface potential calculated from physicochemical properties.

Coagulant dosage determination using deep learning-based graph attention multivariate time series forecasting model.

Determination of coagulant dosage in water treatment is a time-consuming process involving nonlinear data relationships and numerous factors. This study provides a deep learning approach to determine coagulant dosage and/or the settled water turbidity using long-term data between 2011 and 2021 to include the effect of various weather conditions. A graph attention multivariate time series forecasting (GAMTF) model was developed to determine coagulant dosage and was compared with conventional machine learning and deep learning models. The GAMTF model (R2 = 0.94, RMSE = 3.55) outperformed the other models (R2 = 0.63 - 0.89, RMSE = 4.80 - 38.98), and successfully predicted both coagulant dosage and settled water turbidity simultaneously. The GAMTF model improved the prediction accuracy by considering the hidden interrelationships between features and the past states of features. The results demonstrate the first successful application of multivariate time series deep learning model, especially, a state-of-the-art graph attention-based model, using long-term data for decision-support systems in water treatment processes.

Enhanced anaerobic treatment of sulfate-rich wastewater by electrical voltage application.

Treatment of sulfate-rich wastewater with high methane recovery is a major concern due to sulfide inhibition. Here, an electrical voltage (EV) aims to enhance methanogenesis and sulfidogenesis to treat sulfate-rich wastewater. Two (control and EV-applied) reactors were operated with a gradual decrease in chemical oxygen demand (COD)/SO42- ratios (CSR). EV-applied reactor (EVR) demonstrated an increase of ∼30 % in methane production and ∼40 % in sulfate removal, compared to the control till CSR of 2.0. At CSR 1.0, the control failed, while EVR still exhibited a stable performance of 50 % COD-methane recovery. Microbial community results showed that the relative abundance of sulfate-reducing bacteria in EVR was 1.5 times higher than the control. Furthermore, higher relative abundance of dissimilatory sulfate reductase (>50 %) and Ni/Fe hydrogenase (x15) genes demonstrated an improved tolerance against H2S toxicity. This study highlights the importance of EV application by minimizing the byproduct inhibition in sulfate-rich wastewater.

A peptide of PilZ domain-containing protein controls wastewater-treatment-membrane biofouling by inducing bacterial attachment.

Membrane-based wastewater reclamation is used to mitigate water scarcity; however, irreversible biofouling is an elusive problem that hinders the efficiency of a forward-osmosis (FO) membrane-based process, and the protein responsible for fouling is unknown. Herein, we identified fouling proteins by analyzing the microbiome and proteome of wastewater extracellular polymeric substances responsible for strong irreversible FO-membrane fouling. The IGLSSLPR peptide of a PilZ domain-containing protein was found to recruit bacterial attachment when immobilized on the membrane surface while suppressing it when dissolved, in a similar manner to the Arg-Gly-Asp (RGD) peptide in mammalian cell cultures. Bacteria adhere to IGLSSLPR and poly-l-lysine-coated membranes with similar energies and exhibit water fluxes that decline similarly, which is ascribable to interaction as strong as electrostatic interactions in the peptide-coated membranes. We conclude that IGLSSLPR is the key domain responsible for membrane fouling and can be used to develop antifouling technology against bacteria, which is similar to the current usage of RGD peptide in mammalian cell cultures.

Upflow anaerobic sludge blanket reactor operation under high pressure for energy-rich biogas production.

Autogenerative high-pressure digestion has an advantage of producing CH4-rich biogas directly from the reactor. However, its continuous operation has rarely been reported, and has never been attempted in an upflow anaerobic sludge blanket reactor (UASB). Here, UASB was continuously operated at 10 g COD/L/d with increasing pressure from 1 to 8 bar. As the pressure increased, the CH4 content in the biogas increased gradually, reaching 96.7 ± 0.8% at 8 bar (309 MJ/m3 biogas). The pH was dropped from 8.2 to 7.2 with pressure increase, but COD removal efficiency was maintained > 90%. The high pressure up to 8 bar did not adversely impact the physicochemical properties of granules, which was due to the increased production of extracellular polymeric substances (EPS), particularly, tightly bound EPS (34% increase). With pressure increase, there was no changes in the microbial community and ATPase gene expression, but 41% increase in carbonic anhydrase gene expression was observed.

Impact of an extracellular polymeric substance (EPS) precoating on the initial adhesion of Burkholderia cepacia and Pseudomonas aeruginosa.

Extracellular polymeric substances (EPS) significantly influence bacterial adhesion to solid surfaces, but it is difficult to elucidate the role of EPS on bacterial adhesion due to their complexity and variability. In the present study, the effect of EPS on the initial adhesion of B. cepaciaepacia PC184 and P. aeruginosa PAO1 on glass slides with and without an EPS precoating was investigated under three ionic strength conditions. The surface roughness of EPS coated slides was evaluated by atomic force microscopy (AFM), and its effect on initial bacterial adhesion was found to be trivial. X-ray photoelectron spectroscopy (XPS) studies were performed to determine the elemental surface compositions of bacterial cells and substrata. The results showed that an EPS precoating hindered bacterial adhesion on solid surfaces, which was largely attributed to the presence of proteins in the EPS. This observation can be attributed to the increased steric repulsion at high ionic strength conditions. A steric model for polymer brushes that considers the combined influence of steric effects and DLVO interaction forces is shown to adequately describe bacterial adhesion behaviors.

Novel method for the facile control of molecular weight cut-off (MWCO) of ceramic membranes.

This study demonstrates a simple and novel preparation method to prepare ceramic nanofiltration membranes with a precise and tunable molecular weight cut-off (MWCO) by packing variously sized nanoparticles into existing membrane pores. As a result, ceramic membranes with a MWCO from 1000 Da to 10,000 Da were successfully prepared with the narrow distribution of the pore size after the filtration-coating process. In addition, the effective porosity of the ceramic membranes was calculated from the results of the membrane properties by the Hagen-Poiseuille equation which fit within the range of the sphere packing theory from 17.3% to 41.8%. Furthermore, the results of nonlinear curve fitting between the MWCO and the nanoparticle size show a high accuracy, which implies that the MWCO of the ceramic membranes can be predicted using the curve fitting model with variously sized nanoparticles in the filtration-coating process. In conclusion, the novel filtration-coating method enables precise pore control and provides a tunable MWCO to ceramic membranes by preparing various sizes of nanoparticles.

Electrical voltage application as a novel approach for facilitating methanogenic granulation.

Despite having high-rate methanogenic performance, up-flow anaerobic sludge blanket reactor still has challenges regarding long-start up period (3-8 months) for granulation. In this study, "electrical voltage (EV, 0.3 V) application" was attempted for facilitating granulation in the continuous operation with increased organic loading rates (0.5-11.0 kg COD/m3/d). Up to 11.0 kg COD/m3/d, EV-reactor exhibited the stable performance, while the control failed. After 49 days of operation (at 7 kg COD/m3/d), the granules collected from EV-reactor had larger diameter (2.3 vs 1.6 mm), higher settling velocity (2.6 vs 1.9 cm/s), and higher hydrophobicity (52.1 % vs 34.5 %), compared to the control. EV application also increased the specific methanogenic activity for propionate and hydrogen almost by two times. The relative abundance of Pseudomonas sp. (quorum sensing (QS)-related microbe) in EV-reactor was 17 % higher than that in the control. In addition, EV application increased the expression of QS genes significantly by 27 times.

Urchin-like structured magnetic hydroxyapatite for the selective separation of cerium ions from aqueous solutions.

In this study, bio-inspired urchin-like structured hydroxyapatite (UHdA) and its magnetic composite (UHdA@Fe3O4) were developed for efficient and easy separation of cerium ions (Ce3+) from aquatic waste streams. UHdA and UHdA@Fe3O4 exhibited superior Ce3+ adsorption capacities of 248.39 and 230.01 mg/g-UHdA respectively, compared to a commercial HdA (141.71 mg/g-HdA) due to their hierarchical mesoporous structure and large specific surface area. The adsorption of Ce3+ to UHdA and UHdA@Fe3O4 were heterogeneous, pseudo-second-order-kinetic, and the rate-limiting step was external mass transfer and intra-particle diffusion. Moreover, thermodynamic studies revealed that the adsorption process was spontaneous and endothermic nature. The high selectivity towards Ce3+ in multi-ionic systems is attributed to the strong affinity between strong Lewis acid (Ce3+) and base (PO43- and OH-) interactions. XRD, FTIR, and XPS analysis demonstrated that the adsorption was mainly attributable to the ion exchange of Ce3+ with Ca2+ and to surface complexation. The desorption of Ce3+ was efficiently accomplished using 0.1 M HNO3. The results suggest that UHdA and UHdA@Fe3O4 could be promising choices for the adsorption and recovery of rare earth elements.

Parametric studies during the removal of ammonia by membrane contactor with various stripping solutions.

Although membrane contactors (MCs) have been recognized to be an efficient approach for the removal of ammonia from water streams, factors affecting the MCs performance were not clearly investigated. In this study, the effects of stripping solution chemistry (acid types and concentration), feed solution chemistry (pH, temperature, and ammonia concentration), and stages of MCs system have been comprehensively evaluated. Interestingly, the type of stripping solutions significantly affected the removal of ammonia, and the comparative effectiveness were in the order of H3PO4 > H2SO4 > HCOOH. However, the concentration of stripping solutions and ammonia in the feed has little impact to the performance of MCs. Among the feed solution chemistry, pH and temperature were the most crucial factors for ammonia removal in MCs, because the increase of pH and temperature enhanced the free ammonia fraction in the solution and facilitated the mass transfer through pores. At the absorbent concentration of 0.5 M H3PO4, pH of 10, and temperature of 40 °C, single-stage MCs could achieve 51% of ammonia removal within 40 s, and the ammonia removal rate in two-stage MCs reached 90% at the 1.5 min of hydraulic retention time (HRT). The results suggested the superior feasibility of multi-stage MCs system compare to the conventional stripping processes for the removal of ammonia in various waste or wastewater.