Modeling Nanomaterial Environmental Fate in Aquatic Systems

Transcription

1 pubs.acs.org/est Modeling Nanomaterial Environmental Fate in Aquatic Systems Amy L. Dale,,, Elizabeth A. Casman,, Gregory V. Lowry,, Jamie R. Lead, Enrica Viparelli, and Mohammed Baalousha*,, Department of Civil and Environmental Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States Center for Environmental Implications of Nanotechnology, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States Center for Environmental Nanoscience and Risk, Department of Environmental Health Sciences, University South Carolina, Columbia, South Carolina 29208, United States Department of Civil and Environmental Engineering, University of South Carolina, Columbia, South Carolina 29208, United States Mathematical models improve our fundamental understanding of the environmental behavior, fate, and transport of engineered nanomaterials (NMs, chemical substances or materials roughly nm in size) and facilitate risk assessment and management activities. Although today s large-scale environmental fate models for NMs are a considerable improvement over early efforts, a gap still remains between the experimental research performed to date on the environmental fate of NMs and its incorporation into models. This article provides an introduction to the current state of the science in modeling the fate and behavior of NMs in aquatic environments. We address the strengths and weaknesses of existing fate models, identify the challenges facing researchers in developing and validating these models, and offer a perspective on how these challenges can be addressed through the combined efforts of modelers and experimentalists. USES OF NM ENVIRONMENTAL FATE MODELS Models are powerful tools for describing the behavior of contaminants in complex natural systems, and thus are essential for scientific and regulatory purposes. 4,5 For fundamental scientific understanding of environmental fate and behavior of NMs, models can provide a framework to categorize concepts, fill knowledge gaps, and extrapolate and predict missing data. They can also be used to generate scientific hypotheses, which would require further testing via experimental or field studies, test alternative theories about the mechanism underlying an observed experimental result, identify significant scientific uncertainties, and evaluate possible release and exposure scenarios. Fate and behavior models can add value to scientificefforts by exploring the relative influences of the processes contributing to NM fate and behavior in complex ecosystems, thereby identifying dominant processes. 1 3 For regulatory purposes, the aim of investigating the fate and behavior of pollutants is to determine if they pose a risk to environmental and human health and to guide management strategies. 4 Models are of particular value for NMs, as detection of NMs in environmental systems is very challenging and suitable analytical methods are still under development. 6 Most NM fate models for aquatic systems are used to predict environmental concentrations (PECs) and compare them to no adverse effect concentrations (PNECs) to determine risk. 7 They have also been used (with mixed success) for (1) comparison of the relative fate of NMs with different chemical identities and/or prioritizing NMs for additional scrutiny, 8,9 (2) identification of important fate processes or parameters via parametric analysis or sensitivity analysis, 10,11 (3) estimation of potential for long-range transport, 10,12 and (4) estimation of overall residence times. 13 Likely future applications of such models include (5) comparison of the impact of spatially and temporally heterogeneous environmental conditions on fate, (6) evaluation of likely recovery times of contaminated environments should loadings cease, and (7) as a tool for ranking the sources and nature of contamination and feasible remediation strategies. Additionally, advancement in NM modeling may also be beneficial and stimulating for other fields, for instance, to improve our understanding of the fate and transport of contaminants with strong affinity to natural NMs and colloids such as trace metals and organic compounds. The processes that influence NM behavior in the environment include NM surface coating changes (e.g., degradation or replacement by natural organic matter), oxidation, dissolution, sulfidation, advection, diffusion, hetero- and homoaggregation or disaggregation and sedimentation/resuspension 14 (Figure 1). The modeling approaches that have been adapted and applied over the past decade to describe NM fate vary widely in terms of which of these processes are included and how they are modeled. We briefly highlight seminal papers in the field, introduce the approaches they apply, and provide our perspective on the strengths and weaknesses of each approach. We refer the reader to Gottschalk et al. (2014) and Scheringer et al. (2014) for more extensive reviews of environmental fate models for NMs. 7,15 Published: January 22, American Chemical Society 2587

2 Figure 1. Schematic representation of sources and flow of nanomaterials in the environment, and the key processes determining the fate and behavior of nanomaterials in aquatic environments. MATERIAL FLOW ANALYSIS OF NMS The earliest approaches to NM fate modeling relied on material flow analysis (MFA), which is a specific assessment methodology developed in the field of industrial ecology to track the stocks and flows of substances into and between technological compartments (e.g., wastewater treatment plants, incinerators) and environmental compartments (e.g., soil, air, water). Such models have no spatial detail (i.e., they average predicted contaminant mass over the entire environmental compartment, such as all surface water in Switzerland), but they do help conceptualize a material s life cycle. Mueller and Nowack (2008) predicted best and worst case estimates of environmental concentrations (PECs) of Ag, TiO 2, and carbon nanotubes released from products in air, water, soil, and landfills in Switzerland using a deterministic MFA. 16 All model parameters were later treated as probability distributions rather than singlevalue estimates in order to better account for large uncertainty in NM production volumes and behavior. 17,18 The probabilistic approach was then used to model the environmental flow of several NMs for the U.S., Europe and Switzerland. 17 MFAs have also been applied recently at regional scales. 19 MFAs spurred early research into the life cycle of NMs in the environment by identifying sewage treatment plants, biosolids, landfills, and incinerators as key intermediaries between the usage phase of nanoenabled products and the environment. They have provoked the risk community to ask if risk assessment methods (both for hazard and exposure) developed for other pollutants can be directly applied to NMs. They have highlighted ongoing uncertainties in the estimation of production volumes and emissions and have been used to rank the environmental risks from NMs. They have also provided provisional quantitative estimates of environmental concentrations, albeit with large uncertainties and limited spatial resolution. Because they are simple, they accommodate probabilistic or statistical methods such as Monte Carlo simulation more easily than other model types. Until data on NM production volumes and environmental releases become available or analytical methods are developed and applied, estimates of environmental emissions from MFAs will continue to be used in process-based environmental fate models such as those described below (e.g., 8,10,21 ). However, MFAs as they are currently applied to NMs lack spatial resolution and thus are not the most appropriate tool for predicting environmental concentrations. Experimental research and heteroaggregation models show that current generation NMs typically associate strongly with solid phases, leading to sedimentation and the accumulation of NMs in sediments at hot spots near points of release. 11,22,23 Thus, the nationally or even regionally averaged PECs reported in MFAs have little practical relevance for regulatory purposes. MFAs also rely more extensively on simplifying assumptions than other fate model types. Size, shape, aggregation state, surface coatings, particle chemistry, and phase changes of NMs are generally not treated explicitly, but are implicit in the choice of transfer factors between environmental compartments. 17,19,20 Recent probabilistic MFAs have included simple estimates of NM mass loss during sewage treatment due to Ag NM sulfidation and ZnO dissolution, 9 but this modeling framework is not designed to handle the complexities of NM processes in wastewater treatment facilities and environmental media. 2588

3 Environmental Science & Technology PROCESS-BASED ENVIRONMENTAL FATE AND TRANSPORT MODELS OF NMS NMs are largely subject to the same fate and transport processes that have been modeled successfully for organic and ionic contaminants, aerosols, sediments, water, and so on in the atmosphere, soils, sediments, and surface waters (advection, diffusion, and transformation, schematically depicted in Figure 1). Existing fate and transport (F&T) modeling frameworks are capable, with some adaptation, of describing all major NM fate processes. However, different F&T models employ different assumptions with respect to model scale and spatial resolution, whether processes are modeled at steady state or treated as timevariable, and whether transformations and heteroaggregation are expressed as dependent on NM properties, environmental conditions, or both. Some models conserve mass, while others conserve particle number. Such assumptions can radically change predicted results. Several researchers have adapted existing mass balance-based fate models developed for organic contaminants in order to assess the fate of NMs. Boxall et al. (2007) were the first to estimate the concentration of NMs in air, soil and water. 24 Blaser et al. (2008) used a Rhine river model to estimate silver PECs originating from biocidal plastics and fabrics in wastewater treatment plant effluent, surface waters, and sediments. 25 Gottschalk et al. (2011) estimated environmental concentrations of TiO 2, ZnO, and Ag NMs in a spatially resolved probabilistic model of Swiss rivers. Nanomaterial loss from the water column due to heteroaggregation and settling or chemical transformations was handled simply by comparing an optimistic complete removal scenario to a conservative no removal scenario. 12 Liu and Cohen (2014) developed a multimedia (atmosphere, soil, surface water, biota, and sediment) model for NMs based on the fugacity approach developed for organic contaminants. They adapted the model for NMs by binning the NMs by size and using time-independent partitioning ratios for aggregation and attachment of NMs in the different environmental compartments. 8 NM dissolution was accounted for, but other NM chemical transformations (e.g., sulfidation or interaction with organic matter) and their effect on the pristine NM properties and behavior were not considered. The most recent fate models for NMs have incorporated approaches from colloid science, most particularly kinetic descriptions of heteroaggregation. NMs have been shown to exhibit system and time-dependent heteroaggregation behavior, which challenges the mechanistic foundation 26 (but not the immediate practical utility) 27 of equilibrium partition coefficients commonly used in fate models of organic compounds. Colloid models, referred to here as particle transport models (PTMs), balance particle number rather than mass and are generally mechanistic, meaning that model equations attempt to describe all of the fundamental processes that might govern NM behavior Empirical and semiempirical models, in contrast, simplify process descriptions by relying on observed relationships. Although PTMs have been applied extensively to soil systems, little agreement is currently found between these models and experimental data when NMs exhibit nonhyperexponential retention profiles. 31 A recent analysis of PTMs by Goldberg et al. (2014) finds that simple models can outperform complex alternatives for NMs in soil systems and that multiple PTM alternatives should be considered. 31 Cornelis (2015) reviewed the three major fate descriptors for NMs in soils (attachment efficiencies, partition coefficients, and retention 2589 coefficients) and found all of them lacking, especially partition coefficients. 32 Heteroaggregation of NMs in soils has not yet been modeled mechanistically in watershed models. Several recent works attempt to scale PTM equations up for use in semiempirical models of aquatic systems. Arvidsson et al. developed a particle transport model for NMs which accounted for homoaggregation and sedimentation of nano-tio 33 2 in an aqueous suspension. Praetorius et al. adapted the Rhine river model used by Blaser et al. (2008) 25 to include the aggregation framework postulated by Arvidsson et al. (2011) 33 and Petosa et al. (2010). 29 Meesters et al. (2014) developed a mass-based fate model for NMs (SimpleBox4nano) in rivers that used first-order kinetic rates (aggregation, attachment) to track the colloidal behavior of NMs. Heteroaggregation was calculated based on particle collisions and assumed attachment efficiency. Their model framework can also describe dissolution, but the case study presented only considered TiO 2, which is largely insoluble. 21 In surface water models based on PTMs, the major challenge is scaling up mechanistic models, since they are computationally demanding and currently impossible to parametrize adequately for complex environments. Praetorius et al. (2014) favor semiempirical PTMs and attachment efficiencies, 27 whereas Dale et al. favor mass balance alternatives that use simple heuristics to describe heteroaggregation until such time as data become available to improve model parametrization and validation. 26 Quik et al. showed for simple systems that PTMs can be reliably replaced with first-order mass balance alternatives. 34 In summary, no consensus on model formulation has yet been reached. A great deal of experimental research on heteroaggregation in complex environments, as well as comparative analysis of alternative NM frameworks, is needed to resolve this debate. 15,26 Several of the current challenges for the modeling community are common to all NM F&T models. Spatially resolved processbased (F&T) models can provide better estimates of NM environmental concentrations than MFAs because they explicitly model relevant environmental processes at a higher spatial resolution than MFAs (river segments as opposed to all surface water ). However, many F&T models (especially multimedia box models) similarly lack adequate spatial resolution for identifying areas of enhanced accumulation, and still require estimates of sources of NMs often predicted using MFAs. Also, the preponderance of recent NM fate models have assumed steady state conditions and therefore cannot capture time varying effects such as seasonal trends or flow-dependent patterns of sediment accumulation and scour. The development of dynamic models to study the impact of temporal variation on NM fate is likely to be valuable in understanding the behavior of NMs in large scale (e.g., watershed scale) simulations. Modelers have been slow to adopt the most recent research on NM chemical reaction kinetics, even though dissolution rates and other transformations rates have been published recently for several reactive metal and metal oxide NMs including Ag and ZnO NMs. These allow parametrization of even second-order (e.g., sulfide or oxygen-dependent) reactions and nonlinear kinetics such as surface passivation during sulfidation (e.g., refs 35 43). Indeed, no large-scale NM fate models to date have explored transformations beyond simple first-order descriptions of NM dissolution. Similarly, many models do not track dissolution byproducts (metal ions) and their speciation, even though the generation of toxic dissolved ions is a primary source of reactive metal and metal oxide NM ecotoxicity. 44 Dale et al.

4 Table 1. Proposed Next-Generation Enhancements of Nanoparticle F&T Models Possible through Collaboration of Modelers and Experimentalists improvements to... descriptions of NM heteroaggregation and transport descriptions of reactive NM chemistry examples 1. Apply kinetic descriptors of heteroaggregation rather than equilibrium descriptors 2. Model heteroaggregate breakup/disaggregation 3. Express heteroaggregation as a function of environmental drivers (e.g., natural organic matter, ph, ionic strength) and NM properties (e.g., particle size, engineered coating, ph PZC ) 4. Include bedload shift and other relevant sediment transport processes in stream models 5. Express reaction rates as a function of environmental drivers (e.g., oxygen, temperature, ph) and particle properties (e.g., surface area, size) 6. Express rates as a function of particle transformations (e.g., NM dissolution rate as a function of NM sulfidation) 7. Determine rates for both heteroaggregated and unaggregated nanoparticles 8. Determine rates in complex environmental media (e.g., microbially mediated oxidation rates) 9. Track formation and speciation of reaction byproducts (e.g., metal ions) in models model spatial resolution model temporal resolution sensitivity analyses to allow simplification of model structure selection of model boundaries model calibration and validation 10. Shift from national/regional scale models to watershed scale and smaller to predict local accumulation and effects, facilitate policy analysis 11. Parameterize models with site-specific data (e.g., environmental emissions, streamflow and sediment transport parameters, dissolved oxygen, sulfide, ph, organic carbon) 12. Consider the influence of system hydrology/streamflow dynamics on NM fate 13. Consider seasonal trends 14. Identify NM properties and environmental conditions that are irrelevant in complex systems for exclusion from models 15. Compare the influence of NM properties to that of environmental drivers to determine whether one or the other can be ignored in complex systems 16. Determine the level of detail needed to model heteroaggregation with natural colloids (NCs) (How many NC size classes to model? Should NC geochemical identity be accounted for?) 17. Perform comparative analysis of mass-based versus particle number-based (PTM) models 18. Devise first-order alternatives to more complex rate equations, where appropriate 19. Determine whether sources and sinks generally ignored in NM fate models are potentially significant (e.g., incinerators, landfill leachate, crop soil runoff, biouptake and trophic transfer) 20. Develop analytical tools to determine NM concentrations and speciation in complex media at environmentally relevant concentrations 21. Collect relevant data from field and mesocosm-scale studies risk assessment (2013) adapted a one-dimensional sediment diagenetic model 45 to account for sulfidation and oxidative dissolution of silver NMs in sediments and speciation of resulting Ag ions as a function of dissolved oxygen, sulfide, and temperature. 13 The model was used to examine the persistence of NMs in the sediments and the system properties controlling Ag ion efflux from the sediment to the water column. There are no inherent scale-up issues prohibiting the development of large-scale models implementing such system-dependent chemical transformations. In spite of existing uncertainties surrounding reaction rates, it is better to postulate flawed kinetic models than to ignore the transformations of highly reactive particles (e.g., Ag NMs, ZnO NMs, CuO NMs) entirely. 22. Make site-specific estimates of NM concentrations and speciation in wastewater effluent and biosolids to facilitate the estimation of realistic PECs 23. Provide PECs at appropriately fine spatial scale 24. Report uncertainty/variability (e.g., spatiotemporal variation in model results) 25. Perform toxicology on environmentally transformed species at environmentally relevant concentrations to update hazard assessments EXPERIMENTAL RESEARCH UNDERPINNING NM FATE MODELING Fundamental Research Is Still Needed to Determine Model Structure and Parameter Values. As a result of over a decade of research into NM fate and effects, it is now relatively straightforward to identify processes affecting NM fate and anticipate the impact of various particle properties and environmental conditions on NM fate. However, experimental research is still needed to prioritize and quantif y observed relationships in order to determine model structure and input values. For model development, fundamental research is still badly needed in the following areas: (1) Identif ication of the key physicochemical properties of NMs and environmental conditions that impact NM fate and transport in the environment: Molecular and ionic 2590

5 contaminants can be characterized solely by their chemical identities. In contrast, even NMs with the same chemical identity may exhibit different sizes, shapes, crystallinities, surface coatings, etc., and these differences will affect NM behavior. Similarly, environmental fate can be affected by ph, natural organic matter, dissolved oxygen, ionic strength, and so on. 46,47 Models may be extremely sensitive to some of these factors but insensitive to others. Modeling efforts at large scales require an objective consensus on which NM properties, environmental properties, and fate and transport processes are needed to assess the environmental fate of NMs and which can be safely disregarded. This requires close coordination between modelers and experimentalists. For example, there is increasing consensus that homoaggregation can be safely ignored in environmental media models. 10,21 Coupled with experimental research, side-by-side comparisons of alternate modeling frameworks and sensitivity and uncertainty analysis will enable modelers to identify the major drivers of NM fate and to determine which simplifying assumptions can be made safely at various scales and which cannot. (2) Quantif ication of reaction rates and heteroaggregation rates in a manner that ref lects their dependence on NM properties and environmental conditions. Determining rates of heteroaggregation or chemical transformations is challenging for many reasons: (i) Rates depend not only on the NM intrinsic properties, but also the chemistry of the environment in which NMs occur, both of which are highly variable. (ii) Several NM transformation processes are interdependent. 48 For instance, aggregation of NMs may slow down their dissolution due to the decrease in the total surface area exposed to the surrounding environment, 49 with greater effects on compact aggregates. (iii) Many NMs transform to new entities or partially new entities that can retain some of the properties of the pristine NMs, 14,42 with the transformation rates of the partially transformed NMs being intermediate between pristine and fully transformed NMs. (iv) Simple laboratory systems have not captured behaviors such as disaggregation of heteroaggregates under changing environmental conditions or reaction rate modification by microbes (such as has been postulated for the oxidation of sulfide in metal sulfides formed following NM oxidation 13,47 ). Most of the research on NM aggregation performed to date has focused on homoaggregation in simple electrolyte solutions. 50 The impact of natural organic matter (NOM) has been investigated widely in generally simple solutions, but the impact of more polydispersed and heterogeneous NOM or natural sediments has been rarely investigated. Similarly, the impacts of ionic strength, heterogeneities in natural sediment properties (e.g., size, geochemical identity), NOM types, NM surface coatings, and NM type on heteroaggregation behavior are not yet understood in sufficient detail to be modeled. For inclusion in models, relationships between NM properties, environmental properties, and behavior must be quantif ied rather than simply observed. This will require assays that are well-controlled but that contain sufficient complexity to capture environmental behaviors. In particular, they can be used to describe rates as a function of environmental parameters and NM properties, both to 2591 determine the choice of model inputs and their values, and to determine the functional form of rate equations. (3) Finally, we note an ongoing need for f urther research into the form and (if possible) concentration of NMs within wastewater eff luent and biosolids (e.g., refs 51,52) for use in realistic model scenarios. Field and Mesocosm-Scale Data Are Still Needed to Calibrate and Validate Models. Models used to estimate PECs of conventional pollutants are calibrated and validated by field and/or laboratory observations. 4 This is not generally possible for NMs, since current NM loadings are poorly quantified and detection of NMs in complex environmental media is difficult to impossible. Absence of data for calibrating and validating NM fate models is the biggest obstacle to obtaining accurate predictions of environmental NM concentrations and to confidence that NM fate models are predictive as well as descriptive. 7,53 That said, models have many uses (noted previously) and can often provide insights even without complete model calibration and validation. Progress in NM risk assessment modeling will be facilitated by in situ measurement of NM environmental concentrations and forms. Data collection, itself, would be facilitated with the development of analytical tools that can quantify NM concentration and speciation at environmentally relevant concentrations, and the development of NM-specific sample extraction protocols from environmentally complex samples. 6,54,55 We note finally that F&T models only predict exposures. To determine risk, a function of both hazard and exposure, toxicology studies are needed that consider all relevant NM transformation states (as opposed to the current focus on toxicity of pristine NMs). Table 1 summarizes our recommendations for possible enhancements of next-generation NM fate and transport models. Experimental research and data collection are critically needed alongside model development to facilitate these improvements. CONCLUSION Although there are ongoing challenges, the NM fate modeling field has evolved in the direction of better environmental and NM chemistry and better descriptions of particle physics. Existing models tend to focus on one or the other, but the next generation will probably incorporate both. The field has also moved from steady state box models to spatially resolved dynamic models, the latter being better suited for identifying risks and evaluating management strategies. Given the evolving state of the science, predicted environmental concentrations from NM fate models of any type must currently be viewed as merely suggestive. However, as frameworks improve and data limitations decrease, the predictive performance of NM fate models could become a valid basis for regulatory decisionmaking. To date, attention in the NM modeling community has mainly focused on three processes: heteroaggregation, dissolution, and sedimentation. Model formulations describing these three processes have been relatively simple thus far, since data were not previously available to allow model parametrization (e.g., reaction rates and attachment efficiencies). However, NM fate models are developing rapidly. Many other processes, such as disaggregation, resuspension, and reactions with ligands such as NOM, sulfide, phosphate, and chloride are likely to also be considered in the near future, as will be the fate of NM transformation products, which can also have significant

6 environmental impacts. The impact of environmental conditions such as ph, temperature, oxygen or sulfide availability, ionic strength, and natural colloid properties on rates of transformation and heteroaggregation is important, but difficult to quantify at present and rarely modeled as temporally variable, or spatially variable, when, in fact, it is always both. This is a major impediment to environmental realism in modeling NM fate and transport. Lastly, the role of surface coatings and NOM on NM fate is not currently understood well enough to be modeled. NM models are evolving to reflect the dynamic and increasingly quantitative state of the science surrounding NM behaviors. These models will better address key NM-specific behaviors, including heteroaggregation and surface area-dependent chemical transformations. They will have improved spatial and temporal resolution. They will be able to handle a greater range of NM types including nonmetallic and hybrid NMs, 56,57 and also other environmental systems including estuarine and coastal environments, and relevant engineered systems, such as landfills and wastewater treatment plants. It may take another generation or more before these models can credibly interface with NM bioaccumulation and trophic transfer models, which are being developed in parallel, but this too is expected. Such efforts, and discussions within the scientific community, should contribute to the development of an acceptable minimum set of principles for modeling NM fate in the aquatic environment in the not-distant future. Though lack of calibration and validation data is a persistent issue, and though the level of detail required to capture all the important features of the fate of NMs in the environment is still debated, significant strides have been made, with each increment adding to the diversity of questions that the models can address and improving confidence in the results. AUTHOR INFORMATION Corresponding Author * mbaalous@mailbox.sc.edu. Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS This work was supported by nanobee (UK Natural Environmental Research Council, NE/H013148/1), Transatlantic Initiative for Nanotechnology and the Environment (TINE), ModNanoTox (European Union Framework Program VII, NMP-2010-EU-USA), MARINA (European Union Framework Program VII, ), the US EPA Science to Achieve Results program (R834574), the National Science Foundation (NSF) and the Environmental Protection Agency (EPA) under NSF Cooperative Agreement EF /EF , the Smart- State Center for Environmental Nanoscience and Risk (CENR) at the University of South Carolina, and the Center for the Environmental Implications of Nanotechnology (CEINT). A.L.D. acknowledges the Philip and Marsha Dowd-Institute for Complex Engineered Systems (ICES) Engineering Fellowship and the EPA STAR Fellowship program for additional financial support. Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF or the EPA. This work has not been subjected to EPA review and no official endorsement should be inferred. REFERENCES (1) Johnston, J. M.; Lowry, M.; Beaulieu, S.; Bowles, E. State-of-the- Science Report on Predictive Models and Modeling Approaches for Characterizing and Evaluating Exposure to Nanomaterials, EPA/600/R- 10/129; U.S. Environmental Protection Agency, (2) Kaiser, D. L.; Standridge, S.; Friedersdorf, L.; Geraci, C. L.; Kronz, F.; Meador, M. A.; PateB. D.RudnitskyR. G.SloterL. E.SteppD. M. National Nanotechnology Initiative Strategic Plan, (3) OECD Co-Operation on Risk Assessment: Prioritisation of Important Issues on Risk Assessment of Manufactured Nanomaterials- Final Report, ENV/JM/MONO(2013)18; Environment Directorate; Organisation for Economic Co-Operation and Development, (4) Ramaswami, A.; Milford, J. B.; Small, M. J. Integrated Environmental Modeling: Pollutant Transport, Fate, And Risk in the Environment. 1st ed.; John Wiley & Sons: New York, NY, (5) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic Chemistry; Wiley-Interscience, (6) von der Kammer, F.; Ferguson, P. L.; Holden, P. A.; Masion, A.; Rogers, K. R.; Klaine, S. J.; Koelmans, A. A.; Horne, N.; Unrine, J. M. Analysis of engineered nanomaterials in complex matrices (environment and biota): General considerations and conceptual case studies. Environ. Toxicol. Chem. 2012, 31 (1), (7) Gottschalk, F.; Sun, T.; Nowack, B. Environmental concentrations of engineered nanomaterials: Review of modeling and analytical studies. Environ. Pollut. 2013, 181 (0), (8) Liu, H. H.; Cohen, Y. Multimedia environmental distribution of engineered nanomaterials. Environ. Sci. Technol. 2014, 48 (6), (9) Sun, T. Y.; Gottschalk, F.; Hungerbuḧler, K.; Nowack, B. Comprehensive probabilistic modelling of environmental emissions of engineered nanomaterials. Environ. Pollut. 2014, 185, (10) Praetorius, A.; Scheringer, M.; Hungerbuḧler, K. Development of environmental fate models for engineered nanoparticles: A case study of TiO 2 nanoparticles in the Rhine River. Environ. Sci. Technol. 2012, 46 (12), (11) Therezien, M.; Thill, A.; Wiesner, M. R. Importance of heterogeneous aggregation for NP fate in natural and engineered systems. Sci. Total Environ. 2014, (0), (12) Gottschalk, F.; Ort, C.; Scholz, R. W.; Nowack, B. Engineered nanomaterials in rivers Exposure scenarios for Switzerland at high spatial and temporal resolution. Environ. Pollut. 2011, 159 (12), (13) Dale, A. L.; Lowry, G. V.; Casman, E. A. Modeling nanosilver transformations in freshwater sediments. Environ. Sci. Technol. 2013, 47 (22), (14) Lowry, G. V.; Gregory, K. B.; Apte, S. C.; Lead, J. R. Transformations of nanomaterials in the environment. Environ. Sci. Technol. 2012, 46 (13), (15) Scheringer, M.; Praetorius, A.; Goldberg, E. S. Environmental fate and exposure modeling of nanomaterials. Frontiers of Nanoscience, Nanoscience and the Environment 2014, No. 7, (16) Mueller, N. C.; Nowack, B. Exposure modeling of engineered nanoparticles in the environment. Environ. Sci. Technol. 2008, 42 (12), (17) Gottschalk, F.; Sonderer, T.; Scholz, R. W.; Nowack, B. Modeled environmental concentrations of engineered nanomaterials (TiO 2, ZnO, Ag, CNT, fullerenes) for different regions. Environ. Sci. Technol. 2009, 43 (24), (18) Gottschalk, F.; Sonderer, T.; Scholz, R. W.; Nowack, B. Possibilities and limitations of modeling environmental exposure to engineered nanomaterials by probabilistic material flow analysis. Environ. Toxicol. Chem. 2010, 29 (5), (19) Keller, A. A.; Lazareva, A. Predicted releases of engineered nanomaterials: From global to regional to local. Environ. Sci. Technol. Lett. 2013, 1 (1), (20) Keller, A. A.; McFerran, S.; Lazareva, A.; Suh, S. Global life cycle releases of engineered nanomaterials. J. Nanopart. Res. 2013, 15 (6),

7 (21) Meesters, J. A. J.; Koelmans, A. A.; Quik, J. T. K.; Hendriks, A. J.; van de Meent, D. Multimedia modeling of engineered nanoparticles with SimpleBox4nano: Model definition and evaluation. Environ. Sci. Technol. 2014, 48 (10), (22) Velzeboer, I.; Quik, J. T. K.; van de Meent, D.; Koelmans, A. A. Rapid settling of nanoparticles due to heteroaggregation with suspended sediment. Environ. Toxicol. Chem. 2014, 33 (8), (23) Lowry, G. V.; Espinasse, B. P.; Badireddy, A. R.; Richardson, C. J.; Reinsch, B. C.; Bryant, L. D.; Bone, A. J.; Deonarine, A.; Chae, S.; Therezien, M.; Colman, B. P.; Hsu-Kim, H.; Bernhardt, E. S.; Matson, C. W.; Wiesner, M. R. Long-term transformation and fate of manufactured Ag nanoparticles in a simulated large scale freshwater emergent wetland. Environ. Sci. Technol. 2012, 46 (13), (24) Boxall, A. B. A.; Chaudhry, Q.; Sinclair, C.; Jones, A.; Aitken, R.; Jefferson, B.; Watts, C. Current and Future Predicted Environmental Exposure to Engineered Nanoparticles; Central Science Laboratory, Department of the Environment and Rural Affairs: London, UK., (25) Blaser, S. A.; Scheringer, M.; MacLeod, M.; Hungerbuḧler, K. Estimation of cumulative aquatic exposure and risk due to silver: Contribution of nano-functionalized plastics and textiles. Sci. Total Environ. 2008, 390 (2 3), (26) Praetorius, A.; Tufenkji, N.; Goss, K. U.; Scheringer, M.; von der Kammer, F.; Elimelech, M. The road to nowhere: equilibrium partition coefficients for nanoparticles. Environ. Sci. Nano. 2014, 1 (4), (27) Dale, A. L.; Lowry, G. V.; Casman, E. A. Much ado about α: reframing the debate over appropriate fate descriptors in nanoparticle environmental risk modeling. Environ. Sci.: Nano. 2015, DOI: / C4EN00170B. (28) Elimelech, M.; Gregory, J.; Jia, X.; Williams, R. A,. Particle Deposition and Aggregation: Measurement, Modeling and Simulation; Butterworth-Heinemann: Oxford, (29) Petosa, A. R.; Jaisi, D. P.; Quevedo, I. R.; Elimelech, M.; Tufenkji, N. Aggregation and deposition of engineered nanomaterials in aquatic environments: Role of physicochemical interactions. Environ. Sci. Technol. 2010, 44 (17), (30) Phenrat, T.; Song, J. E.; Cisneros, C. M.; Schoenfelder, D. P.; Tilton, R. D.; Lowry, G. V. Estimating attachment of nano-and submicrometer-particles coated with organic macromolecules in porous media: development of an empirical model. Environ. Sci. Technol. 2010, 44 (12), (31) Goldberg, E.; Scheringer, M.; Bucheli, T. D.; Hungerbuḧler, K. Critical assessment of models for transport of engineered nanoparticles in saturated porous media. Environ. Sci. Technol. 2014, 48 (21), (32) Cornelis, G. Fate descriptors for engineered nanoparticles: the good, the bad, and the ugly. Environ. Sci.: Nano 2015, DOI: / C4EN00122B. (33) Arvidsson, R.; Molander, S.; Sandeń, A.; Hassello v, M. Challenges in exposure modeling of nanoparticles in aquatic environments. Hum. Ecol. Risk Assess. 2011, 17 (1), (34) Quik, J. T.; van de Meent, D.; Koelmans, A. A. Simplifying modeling of nanoparticle aggregation-sedimentation behavior in environmental systems: A theoretical analysis. Water Res. 2014, 62, (35) Liu, J.; Hurt, R. H. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ. Sci. Technol. 2010, 44 (6), (36) Liu, J.; Sonshine, D. A.; Shervani, S.; Hurt, R. H. Controlled release of biologically active silver from nanosilver surfaces. ACS Nano 2010, 4 (11), (37) Liu, J.; Pennell, K. G.; Hurt, R. H. Kinetics and mechanisms of nanosilver oxysulfidation. Environ. Sci. Technol. 2011, 45 (17), (38) Levard, C.; Reinsch, B. C.; Michel, F. M.; Oumahi, C.; Lowry, G. V.; Brown, G. E. Sulfidation processes of PVP-coated silver nanoparticles in aqueous solution: Impact on dissolution rate. Environ. Sci. Technol. 2011, 45 (12), (39) Ma, R.; Stegemeier, J.; Levard, C.; Dale, J.; Noack, C. W.; Yang, T.; Brown, G.; Lowry, G. Sulfidation of copper oxide nanoparticles and 2593 properties of resulting copper sulfide. Environ. Sci. Nano. 2014, 1 (4), (40) Ma, R.; Levard, C. m.; Michel, F. M.; Brown, G. E.; Lowry, G. V. Sulfidation mechanism for zinc oxide nanoparticles and the effect of sulfidation on their solubility. Environ. Sci. Technol. 2013, 47 (6), (41) Mudunkotuwa, I. A.; Rupasinghe, T.; Wu, C. M.; Grassian, V. H. Dissolution of ZnO nanoparticles at circumneutral ph: A study of size effects in the presence and absence of citric acid. Langmuir 2011, 28 (1), (42) Bian, S. W.; Mudunkotuwa, I. A.; Rupasinghe, T.; Grassian, V. H. Aggregation and dissolution of 4 nm ZnO nanoparticles in aqueous environments: Influence of ph, ionic strength, size, and adsorption of humic acid. Langmuir 2011, 27 (10), (43) Baalousha, M.; Arkill, K. P.; Romer, I.; Palmer, R. E.; Lead, J. R. Transformations of citrate and Tween coated silver nanoparticles reacted with Na2S. Sci. Total Environ. 2015, 502 (0), (44) Maurer-Jones, M. A.; Gunsolus, I. L.; Murphy, C. J.; Haynes, C. L. Toxicity of engineered nanoparticles in the environment. Anal. Chem. 2013, 85 (6), (45) Di Toro, D. M.; Mahony, J. D.; Hansen, D. J.; Berry, W. J. A model of the oxidation of iron and cadmium sulfide in sediments. Environ. Toxicol. Chem. 1996, 15 (12), (46) Baalousha, M.; Manciulea, A.; Cumberland, S.; Kendall, K.; Lead, J. R. Aggregation and surface properties of iron oxide nanoparticles; Influence of ph and natural organic matter. Environ. Toxicol. Chem. 2008, 27 (9), (47) Baalousha, M. Aggregation and disaggregation of iron oxide nanoparticles: Influence of particle concentration, ph and natural organic matter. Sci. Total Environ. 2009, 407 (6), (48) Misra, S. K.; Dybowska, A.; Berhanu, D.; Luoma, S. N.; Valsami- Jones, E. The complexity of nanoparticle dissolution and its importance in nanotoxicological studies. Sci. Total Environ. 2012, 438 (0), (49) Lin, D.; Tian, X.; Wu, F.; Xing, B. Fate and transport of engineered nanomaterials in the environment. J. Environ. Qual. 2010, 39 (6), (50) Baalousha, M.; Nur, Y.; Ro mer, I.; Tejamaya, M.; Lead, J. R. Effect of monovalent and divalent cations, anions and fulvic acid on aggregation of citrate-coated silver nanoparticles. Sci. Total Environ. 2013, (0), (51) Kaegi, R.; Voegelin, A.; Ort, C.; Sinnet, B.; Thalmann, B.; Krismer, J.; Hagendorfer, H.; Elumelu, M.; Mueller, E. Fate and transformation of silver nanoparticles in urban wastewater systems. Water Res. 2013, 47 (12), (52) Ma, R.; Levard, C.; Judy, J. D.; Unrine, J. M.; Durenkamp, M.; Martin, B.; Jefferson, B.; Lowry, G. V. Fate of zinc oxide and silver nanoparticles in a pilot wastewater treatment plant and in processed biosolids. Environ. Sci. Technol. 2014, 48 (1), (53) Gottschalk, F.; Kost, E.; Nowack, B. Engineered nanomaterials in water and soils: A risk quantification based on probabilistic exposure and effect modeling. Environ. Toxicol. Chem. 2013, 32 (6), (54) Kookana, R. S.; Boxall, A. B.; Reeves, P.; Ashauer, R.; Beulke, S.; Chaudhry, Q.; Cornelis, G.; Fernandes, P. T.; Gan, J. J.; Kah, M. Nanopesticides: Guiding principles for regulatory evaluation of environmental risks. J. Agric. Food Chem. 2014, 62 (19), (55) Baalousha, M.; Prasad, A.; Lead, J. R. Quantitative measurement of the nanoparticle size and number concentration from liquid suspensions by atomic force microscopy. Environ. Sci. Process. Impact. 2014, 16 (6), (56) Saleh, N. B.; Afrooz, A. R. M.; Bisesi, J. H., Jr.; Aich, N.; Plazas- Tuttle, J.; Sabo-Attwood, T. Emergent properties and toxicological considerations for nanohybrid materials in aquatic systems. Nanomaterials 2014, 4 (2), (57) Aich, N.; Plazas-Tuttle, J.; Lead, J.; Saleh, N. A critical review of nanohybrids: Synthesis, applications, and environmental implications. Environ. Chem. 2014, 11 (6),