Polyamide Membranes with Tunable Surface Charge Induced by Dipole-Dipole Interaction for Selective Ion Separation.

Nanofiltration (NF) has the potential to achieve precise ion-ion separation at the subnanometer scale, which is necessary for resource recovery and a circular water economy. Fabricating NF membranes for selective ion separation is highly desirable but represents a substantial technical challenge. Dipole-dipole interaction is a mechanism of intermolecular attractions between polar molecules with a dipole moment due to uneven charge distribution, but such an interaction has not been leveraged to tune membrane structure and selectivity. Herein, we propose a novel strategy to achieve tunable surface charge of polyamide membrane by introducing polar solvent with a large dipole moment during interfacial polymerization, in which the dipole-dipole interaction with acyl chloride groups of trimesoyl chloride (TMC) can successfully intervene in the amidation reaction to alter the density of surface carboxyl groups in the polyamide selective layer. As a result, the prepared positively charged (PEI-TMC)-NH2 and negatively charged (PEI-TMC)-COOH composite membranes, which show similarly high water permeance, demonstrate highly selective separations of cations and anions in engineering applications, respectively. Our findings, for the first time, confirm that solvent-induced dipole-dipole interactions are able to alter the charge type and density of polyamide membranes and achieve tunable surface charge for selective and efficient ion separation.

A Tale of Two Foulants: The Coupling of Organic Fouling and Mineral Scaling in Membrane Desalination.

Membrane desalination that enables the harvesting of purified water from unconventional sources such as seawater, brackish groundwater, and wastewater has become indispensable to ensure sustainable freshwater supply in the context of a changing climate. However, the efficiency of membrane desalination is greatly constrained by organic fouling and mineral scaling. Although extensive studies have focused on understanding membrane fouling or scaling separately, organic foulants commonly coexist with inorganic scalants in the feedwaters of membrane desalination. Compared to individual fouling or scaling, combined fouling and scaling often exhibits different behaviors and is governed by foulant-scalant interactions, resembling more complex but practical scenarios than using feedwaters containing only organic foulants or inorganic scalants. In this critical review, we first summarize the performance of membrane desalination under combined fouling and scaling, involving mineral scales formed via both crystallization and polymerization. We then provide the state-of-the-art knowledge and characterization techniques pertaining to the molecular interactions between organic foulants and inorganic scalants, which alter the kinetics and thermodynamics of mineral nucleation as well as the deposition of mineral scales onto membrane surfaces. We further review the current efforts of mitigating combined fouling and scaling via membrane materials development and pretreatment. Finally, we provide prospects for future research needs that guide the design of more effective control strategies for combined fouling and scaling to improve the efficiency and resilience of membrane desalination for the treatment of feedwaters with complex compositions.

Exploring the Knowledge Attained by Machine Learning on Ion Transport across Polyamide Membranes Using Explainable Artificial Intelligence.

Recent studies have increasingly applied machine learning (ML) to aid in performance and material design associated with membrane separation. However, whether the knowledge attained by ML with a limited number of available data is enough to capture and validate the fundamental principles of membrane science remains elusive. Herein, we applied explainable artificial intelligence (XAI) to thoroughly investigate the knowledge learned by ML on the mechanisms of ion transport across polyamide reverse osmosis (RO) and nanofiltration (NF) membranes by leveraging 1,585 data from 26 membrane types. The Shapley additive explanation method based on cooperative game theory was used to unveil the influences of various ion and membrane properties on the model predictions. XAI shows that the ML can capture the important roles of size exclusion and electrostatic interaction in regulating membrane separation properly. XAI also identifies that the mechanisms governing ion transport possess different relative importance to cation and anion rejections during RO and NF filtration. Overall, we provide a framework to evaluate the knowledge underlying the ML model prediction and demonstrate that ML is able to learn fundamental mechanisms of ion transport across polyamide membranes, highlighting the importance of elucidating model interpretability for more reliable and explainable ML applications to membrane selection and design.

Capillary-Assisted Fabrication of Thin-Film Nanocomposite Membranes for Improved Solute-Solute Separation.

Efficient separation of harmful contaminants (e.g., per- and polyfluoroalkyl substances, PFASs) from valuable components (water and nutrients) is essential to the resource recovery from domestic wastewater for agricultural purposes. Such selective recovery requires precise separation at the angstrom scale. Although nanofiltration (NF) has the potential to achieve solute-solute separation, the state-of-the-art polyamide (PA) membranes are typically constrained by limited precision of solute-solute selectivity and their well-documented permeability-selectivity trade-off. Herein, we present a novel capillary-assisted interfacial polymerization (CAIP) approach to generate metal-organic framework (MOF)-PA nanocomposite membranes with reduced surface charges and more uniform pore sizes that favor solute selectivity by enhanced size exclusion. By uniquely regulating the PA-MOF interactions using the capillary force, CAIP results in effective exposure of MOF nanochannels on the membrane surface and a PA matrix with a high cross-linking gradient in the vertical direction, both of which contribute to an exceptional water permeance of ∼18.7 LMH/bar and an unprecedentedly high selectivity between nutrient ions and PFASs. Our CAIP approach breaks new ground for utilizing nanoparticles with nanochannels in fabricating the next-generation, fit-for-purpose NF membranes for improved solute-solute separations.

Which Surface Is More Scaling Resistant? A Closer Look at Nucleation Theories for Heterogeneous Gypsum Nucleation in Aqueous Solutions.

Developing engineered surfaces with scaling resistance is an effective means to inhibit surface-mediated mineral scaling in various industries including desalination. However, contrasting results have been reported on the relationship between scaling potential and surface hydrophilicity. In this study, we combine a theoretical analysis with experimental investigation to clarify the effect of surface wetting property on heterogeneous gypsum (CaSO4·2H2O) formation on surfaces immersed in aqueous solutions. Theoretical prediction derived from classical nucleation theory (CNT) indicates that an increase of surface hydrophobicity reduces scaling potential, which contrasts our experimental results that more hydrophilic surfaces are less prone to gypsum scaling. We further consider the possibility of nonclassical pathway of gypsum nucleation, which proceeds by the aggregation of precursor clusters of CaSO4. Accordingly, we investigate the affinity of CaSO4 to substrate surfaces of varied wetting properties via calculating the total free energy of interaction, with the results perfectly predicting experimental observations of surface scaling propensity. This indicates that the interactions between precursor clusters of CaSO4 and substrate surfaces might play an important role in regulating heterogeneous gypsum formation. Our findings provide evidence that CNT might not be applicable to describing gypsum scaling in aqueous solutions. The fundamental insights we reveal on gypsum scaling mechanisms have the potential to guide rational design of scaling-resistant engineered surfaces.

Superhydrophobic Carbon Nanotube Network Membranes for Membrane Distillation: High-Throughput Performance and Transport Mechanism.

Despite increasing sustainable water purification, current desalination membranes still suffer from insufficient permeability and treatment efficiency, greatly hindering extensive practical applications. In this work, we provide a new membrane design protocol and molecule-level mechanistic understanding of vapor transport for the treatment of hypersaline waters via a membrane distillation process by rationally fabricating more robust metal-based carbon nanotube (CNT) network membranes, featuring a superhydrophobic superporous surface (80.0 ± 2.3% surface porosity). With highly permeable ductile metal hollow fibers as substrates, the construction of a superhydrophobic (water contact angle ∼170°) CNT network layer endows the membranes with not only almost perfect salt rejection (over 99.9%) but a promising water flux (43.6 L·m-2·h-1), which outperforms most existing inorganic distillation membranes. Both experimental and molecular dynamics simulation results indicate that such an enhanced water flux can be ascribed to an ultra-low liquid-solid contact interface (∼3.23%), allowing water vapor to rapidly transport across the membrane structure via a combined mechanism of Knudsen diffusion (more dominant) and viscous flow while efficiently repelling high-salinity feed via forming a Cassie-Baxter state. A more hydrophobic surface is more in favor of not only water desorption from the CNT outer surface but superfast and frictionless water vapor transport. By constructing a new superhydrophobic triple-phase interface, the conceptional design strategy proposed in this work can be expected to be extended to other membrane material systems as well as more water treatment applications.

Predicting Micropollutant Removal by Reverse Osmosis and Nanofiltration Membranes: Is Machine Learning Viable?

Predictive models for micropollutant removal by membrane separation are highly desirable for the design and selection of appropriate membranes. While machine learning (ML) models have been applied for such purposes, their reliability might be compromised by data leakage due to inappropriate data splitting. More importantly, whether ML models can truly understand the mechanisms of membrane separation has not been revealed. In this study, we evaluate the capability of the XGBoost model to predict micropollutant removal efficiencies of reverse osmosis and nanofiltration membranes. Our results demonstrate that data leakage leads to falsely high prediction accuracy. By utilizing a model interpretation method based on the cooperative game theory, we test the knowledge of XGBoost on the mechanisms of membrane separation via quantifying the contributions of input variables to the model predictions. We reveal that XGBoost possesses an adequate understanding of size exclusion, but its knowledge of electrostatic interactions and adsorption is limited. Our findings suggest that future work should focus more on avoiding data leakage and evaluating the mechanistic knowledge of ML models. In addition, high-quality data from more diverse experimental conditions, as well as more informative variables, are needed to improve the accuracy of ML models for predicting membrane performance.

Differentiating Solutes with Precise Nanofiltration for Next Generation Environmental Separations: A Review.

Selective removal or enrichment of targeted solutes including micropollutants, valuable elements, and mineral scalants from complex aqueous matrices is both challenging and pivotal to the success of water purification and resource recovery from unconventional water resources. Membrane separation with precision at the subnanometer or even subangstrom scale is of paramount importance to address those challenges via enabling "fit-for-purpose" water and wastewater treatment. So far, researchers have attempted to develop novel membrane materials with precise and tailored selectivity by tuning membrane structure and chemistry. In this critical review, we first present the environmental challenges and opportunities that necessitate improved solute-solute selectivity in membrane separation. We then discuss the mechanisms and desired membrane properties required for better membrane selectivity. On the basis of the most recent progress reported in the literature, we examine the key principles of material design and fabrication, which create membranes with enhanced and more targeted selectivity. We highlight the important roles of surface engineering, nanotechnology, and molecular-level design in improving membrane selectivity. Finally, we discuss the challenges and prospects of highly selective NF membranes for practical environmental applications, identifying knowledge gaps that will guide future research to promote environmental sustainability through more precise and tunable membrane separation.

Contrasting Behaviors between Gypsum and Silica Scaling in the Presence of Antiscalants during Membrane Distillation.

Mineral scaling is a major constraint that limits the performance of membrane distillation (MD) for hypersaline wastewater treatment. Although the use of antiscalants is a common industrial practice to mitigate mineral scaling, the effectiveness and underlying mechanisms of antiscalants in inhibiting different mineral scaling types have not been systematically investigated. Herein, we perform a comparative investigation to elucidate the efficiencies of antiscalant candidates with varied functional groups for mitigating gypsum scaling and silica scaling in MD desalination. We show that antiscalants with Ca(II)-complexing moieties (e.g., carboxyl group) are the most effective to inhibit gypsum scaling formed via crystallization, whereas amino-enriched antiscalants possess the best performance to mitigate silica scaling created by polymerization. A set of microscopic and spectroscopic analyses reveal distinct mechanisms of antiscalants required for those two common types of scaling. The mitigating effect of antiscalants on gypsum scaling is attributed to the stabilization of scale precursors and nascent CaSO4 nuclei, which hinders phase transformation of amorphous CaSO4 toward crystalline gypsum. In contrast, antiscalants facilitate the polymerization of silicic acid, immobilizing active silica precursors and retarding the gelation of silica scale layer on the membrane surface. Our study, for the first time, demonstrates that antiscalants with different functionalities are required for the mitigation of gypsum scaling and silica scaling, providing mechanistic insights on the molecular design of antiscalants tailored to MD applications for the treatment of wastewaters containing different scaling types.

Distinct Behaviors between Gypsum and Silica Scaling in Membrane Distillation.

Mineral scaling constrains membrane distillation (MD) and limits its application in treating hypersaline wastewater. Addressing this challenge requires enhanced fundamental understanding of the scaling phenomenon. However, MD scaling with different types of scalants may have distinctive mechanisms and consequences which have not been systematically investigated in the literature. In this work, we compared gypsum and silica scaling in MD and demonstrated that gypsum scaling caused earlier water flux decline and induced membrane wetting that was not observed in silica scaling. Microscopic imaging and elemental mapping revealed contrasting scale morphology and distribution for gypsum and silica, respectively. Notably, while gypsum crystals grew both on the membrane surface and deep in the membrane matrix, silica only formed on the membrane surface in the form of a relatively thin film composed of connected submicrometer silica particles. We attribute the intrusion of gypsum into membrane pores to the crystallization pressure as a result of rapid, oriented crystal growth, which leads to pore deformation and the subsequent membrane wetting. In contrast, the silica scale layer was formed via polymerization of silicic acid and gelation of silica particles, which were less intrusive and had a milder effect on membrane pore structure. This hypothesis was supported by the result of tensile testing, which showed that the MD membrane was significantly weakened by gypsum scaling. The fact that different scaling mechanisms could yield different consequences on membrane performance provides valuable insights for the future development of cost-effective strategies for scaling control.

On-Site Treatment of Shale Gas Flowback and Produced Water in Sichuan Basin by Fertilizer Drawn Forward Osmosis for Irrigation.

Fertilizer drawn forward osmosis (FDFO) was proposed to extract fresh water from flowback and produced water (FPW) from shale gas extraction for irrigation, with fertilizer types and membrane orientations assessed. The draw solution (DS) with NH4H2PO4 displayed the best performance, while the DS with (NH4)2HPO4 resulted in the most severe membrane fouling. The DS with KCl and KNO3 led to substantial reverse solute fluxes. The FDFO operation where the active layer of the membrane was facing the feed solution outperformed that when the active layer was facing the DS. The diluted DS and diluted FPW samples were used for irrigation of Cherry radish and Chinese cabbage. Compared to deionized water, irrigation with the diluted DS (total dissolved solid (TDS) = 350 mg·L-1) promoted plant growth. In contrast, inhibited plant growth was observed when FPW with high salinity (TDS = 5000 mg·L-1) and low salinity (TDS = 1000 mg·L-1) was used for irrigation of long-term (8-week) plant cultures. Finally, upregulated genes were identified to illustrate the difference in plant growth. The results of this study provide a guide for efficient and safe use of FPW after FDFO treatment for agricultural application.

Elucidating the Trade-off between Membrane Wetting Resistance and Water Vapor Flux in Membrane Distillation.

Membrane distillation (MD) has been receiving considerable attention as a promising technology for desalinating industrial wastewaters. While hydrophobic membranes are essential for the process, increasing membrane surface hydrophobicity generally leads to the reduction of water vapor flux. In this study, we investigate the mechanisms responsible for this trade-off relation in MD. We prepared hydrophobic membranes with different degrees of wetting resistance through coating quartz fiber membranes with a series of alkylsilane molecules while preserving the fiber structures. A trade-off between wetting resistance and water vapor flux was observed in direct-contact MD experiments, with the least-wetting-resistant membrane exhibiting twice as high vapor flux as the most wetting-resistant membrane. Electrochemical impedance analysis, combined with fluorescence microscopy, elucidated that a lower wetting resistance (still water-repelling) allows deeper penetration of the liquid-air interfaces into the membrane, resulting in an increased interfacial area and therefore a larger evaporative vapor flux. Finally, we performed osmotic distillation experiments employing anodized alumina membranes that possess straight nanopores with different degrees of wetting resistance, observed no trade-off, and substantiated this proposed mechanism. Our study provides a guideline to tailor the membrane surface wettability to ensure stable MD operations while maximizing the water recovery rate.

Electrochemical Oxidation of Hexafluoropropylene Oxide Dimer Acid (GenX): Mechanistic Insights and Efficient Treatment Train with Nanofiltration.

Hexafluoropropylene oxide dimer acid (HFPO-DA, trade name GenX) is a perfluoroalkyl ether carboxylic acid (PFECA) that has been detected in watersheds around the world. Similar to other per- and polyfluoroalkyl substances (PFASs), few processes are able to break HFPO-DA's persistent carbon-fluorine bonds. This study provides both experimental and computational lines of evidence for HFPO-DA mineralization during electrochemical oxidation at a boron-doped diamond anode with a low potential for the generation of stable organofluorine intermediates. Our density functional theory calculations consider the major operative mechanism, direct electron transfer, throughout the entire pathway. Initial oxidative attack does not break the ether bond, but leads to stepwise mineralization of the acidic side chain. Our mechanistic investigations reveal that hydroxyl radicals are unreactive toward HFPO-DA, while electrochemically activated sulfate facilitates its oxidation. Furthermore, we demonstrate that an NF90 membrane is capable of removing 99.5% of HFPO-DA from contaminated water. Electrochemical treatment of the nanofiltration rejectate is shown to reduce both energy and electrode costs by more than 1 order of magnitude compared to direct electrochemical treatment of the raw water. Overall, a nanofiltration-electrochemical oxidation treatment train is a sustainable destructive approach for the cost-effective elimination of HFPO-DA and other PFASs from contaminated water.

Combined Organic Fouling and Inorganic Scaling in Reverse Osmosis: Role of Protein-Silica Interactions.

We investigated the relationship between silica scaling and protein fouling in reverse osmosis (RO). Flux decline caused by combined scaling and fouling was compared with those by individual scaling or fouling. Bovine serum albumin (BSA) and lysozyme (LYZ), two proteins with opposite charges at typical feedwater pH, were used as model protein foulants. Our results demonstrate that water flux decline was synergistically enhanced when silica and protein were both present in the feedwater. For example, flux decline after 500 min was far greater in combined silica scaling and BSA fouling experiments (55 ± 6% decline) than those caused by silica (11 ± 2% decline) or BSA (9 ± 1% decline) alone. Similar behavior was observed with silica and LYZ, suggesting that this synergistic effect was independent of protein charge. Membrane characterization by scanning electron microscopy and Fourier transform infrared spectroscopy revealed distinct foulant layers formed by BSA and LYZ in the presence of silica. A combination of dynamic light scattering, transmission electron microscopy , and energy dispersive X-ray spectroscopy analyses further suggested that BSA and LYZ facilitated the formation of aggregates with varied chemical compositions. As a result, BSA and LYZ were likely to play different roles in enhancing flux decline in combined scaling and fouling. Our study suggests that the coexistence of organic foulants, such as proteins, largely alters scaling behavior of silica, and that accurate prediction of RO performance requires careful consideration of foulant-scalant interactions.

Relating Silica Scaling in Reverse Osmosis to Membrane Surface Properties.

We investigated the relationship between membrane surface properties and silica scaling in reverse osmosis (RO). The effects of membrane hydrophilicity, free energy for heterogeneous nucleation, and surface charge on silica scaling were examined by comparing thin-film composite polyamide membranes grafted with a variety of polymers. Results show that the rate of silica scaling was independent of both membrane hydrophilicity and free energy for heterogeneous nucleation. In contrast, membrane surface charge demonstrated a strong correlation with the extent of silica scaling (R2 > 0.95, p < 0.001). Positively charged membranes significantly facilitated silica scaling, whereas a more negative membrane surface charge led to reduced scaling. This observation suggests that deposition of negatively charged silica species on the membrane surface plays a critical role in silica scale formation. Our findings provide fundamental insights into the mechanisms governing silica scaling in reverse osmosis and highlight the potential of membrane surface modification as a strategy to reduce silica scaling.

Performance and Mechanisms of Ultrafiltration Membrane Fouling Mitigation by Coupling Coagulation and Applied Electric Field in a Novel Electrocoagulation Membrane Reactor.

A novel electrocoagulation membrane reactor (ECMR) was developed, in which ultrafiltration (UF) membrane modules are placed between electrodes to improve effluent water quality and reduce membrane fouling. Experiments with feedwater containing clays (kaolinite) and natural organic matter (humic acid) revealed that the combined effect of coagulation and electric field mitigated membrane fouling in the ECMR, resulting in higher water flux than the conventional combination of electrocoagulation and UF in separate units (EC-UF). Higher current densities and weakly acidic pH in the EMCR favored faster generation of large flocs and effectively reduced membrane pore blocking. The hydraulic resistance of the formed cake layers on the membrane surface in ECMR was reduced due to an increase in cake layer porosity and polarity, induced by both coagulation and the applied electric field. The formation of a polarized cake layer was controlled by the applied current density and voltage, with cake layers formed under higher electric field strengths showing higher porosity and hydrophilicity. Compared to EC-UF, ECMR has a smaller footprint and could achieve significant energy savings due to improved fouling resistance and a more compact reactor design.

One-Time Addition of Nano-TiO2 Triggers Short-Term Responses in Benthic Bacterial Communities in Artificial Streams.

Nano-TiO2 is an engineered nanomaterial whose production and use are increasing rapidly. Hence, aquatic habitats are at risk for nano-TiO2 contamination due to potential inputs from urban and suburban runoff and domestic wastewater. Nano-TiO2 has been shown to be toxic to a wide range of aquatic organisms, but little is known about the effects of nano-TiO2 on benthic microbial communities. This study used artificial stream mesocosms to assess the effects of a single addition of nano-TiO2 (P25 at a final concentration of 1 mg l(-1)) on the abundance, activity, and community composition of sediment-associated bacterial communities. The addition of nano-TiO2 resulted in a rapid (within 1 day) decrease in bacterial abundance in artificial stream sediments, but bacterial abundance returned to control levels within 3 weeks. Pyrosequencing of partial 16S rRNA genes did not indicate any significant changes in the relative abundance of any bacterial taxa with nano-TiO2 treatment, indicating that nano-TiO2 was toxic to a broad range of bacterial taxa and that recovery of the bacterial communities was not driven by changes in community composition. Addition of nano-TiO2 also resulted in short-term increases in respiration rates and denitrification enzyme activity, with both returning to control levels within 3 weeks. The results of this study demonstrate that single-pulse additions of nano-TiO2 to aquatic habitats have the potential to significantly affect the abundance and activity of benthic microbial communities and suggest that interactions of TiO2 nanoparticles with environmental matrices may limit the duration of their toxicity.

Lifetime and dissolution kinetics of zinc oxide nanoparticles in aqueous media.

We have assessed the persistence and lifetime of ZnO nanoparticles (ZnO-NPs) by performing dissolution experiments in three different aqueous media. These experiments were performed at ZnO-NP concentration levels close to the solubility of zincite (∼8 µM or 650 µg l(-1) of ZnO)-a concentration that is orders of magnitude higher than current estimated relevant environmental concentrations. The kinetics were followed by voltammetry, while maintaining the pH at about 7.5 using a CO2/N2 gas mixture to remove di-oxygen interference. Our results show that, under these conditions, ZnO-NPs readily dissolve with a lifetime expectancy that does not exceed 90 min. Water chemistry, especially the presence of dissolved organic matter (DOM), plays an important role in ZnO-NP dissolution. Dissolution rates significantly increase in the presence of strong chelating agents, EDTA and L-cysteine, while the addition of polymeric DOM, such as sodium alginate, has the opposite effect. Our results suggest that ZnO-NPs are unlikely to persist in natural aqueous media and that the toxicity should be primarily related to the released Zn(2+) ions rather than effects commonly associated to the presence of nanoparticles.

The Global Rise of Zero Liquid Discharge for Wastewater Management: Drivers, Technologies, and Future Directions.

Zero liquid discharge (ZLD)-a wastewater management strategy that eliminates liquid waste and maximizes water usage efficiency - has attracted renewed interest worldwide in recent years. Although implementation of ZLD reduces water pollution and augments water supply, the technology is constrained by high cost and intensive energy consumption. In this critical review, we discuss the drivers, incentives, technologies, and environmental impacts of ZLD. Within this framework, the global applications of ZLD in the United States and emerging economies such as China and India are examined. We highlight the evolution of ZLD from thermal- to membrane-based processes, and analyze the advantages and limitations of existing and emerging ZLD technologies. The potential environmental impacts of ZLD, notably greenhouse gas emission and generation of solid waste, are discussed and the prospects of ZLD technologies and research needs are highlighted.

Attenuation of Microbial Stress Due to Nano-Ag and Nano-TiO2 Interactions under Dark Conditions.

Engineered nanomaterials (ENMs) are incorporated into thousands of commercial products, and their release into environmental systems creates complex mixtures with unknown toxicological outcomes. To explore this scenario, we probe the chemical and toxicological interactions of nanosilver (n-Ag) and nanotitania (n-TiO2) in Lake Michigan water, a natural aqueous medium, under dark conditions. We find that the presence of n-Ag induces a stress response in Escherichia coli, as indicated by a decrease in ATP production observed at low concentrations (in the µg L-1 range), with levels that are environmentally relevant. However, when n-Ag and n-TiO2 are present together in a mixture, n-TiO2 attenuates the toxicity of n-Ag at and below 20 µg L-1 by adsorbing Ag+(aq). We observe, however, that toxic stress cannot be explained by dissolved silver concentrations alone and, therefore, must also depend on silver associated with the nanoscale fraction. Although the attenuating effect of n-TiO2 on n-Ag's toxicity is limited, this study emphasizes the importance of probing the toxicity of ENM mixtures under environmental conditions to assess how chemical interactions between nanoparticles change the toxicological effects of single ENMs in unexpected ways.