Detection, toxicology, environmental fate and risk assessment of nanoparticles in the aquatic environment (DeTER)

Abstract

Nanotechnology is an emerging technology that has the potential to impact on all aspects of life and the economy and is expected to form the basis of several technological innovations and advances in the 21st century. The European Commission defines a nanomaterial as “a natural, incidental or manufactured material containing particles, in an unbound state or as an aggregate or as an agglomerate and where, for 50% or more of the particles in the number size distribution, one or more external dimensions is in the size range 1 nm–100 nm”. The production of and demand for products containing nanomaterials has increased significantly in recent years. The Nanodatabase developed by the Technical University of Denmark is a “living” inventory of commercially available products in the European consumer market that claim to contain engineered nanomaterials (ENMs) and currently lists 3037 products. Nanomaterials have a wide range of potential applications, from everyday uses (such as improvements in fabrics, paints, cosmetics and packaging) to medical applications, water and soil remediation and renewable energy production. Nanomaterials include both nanoparticles and nanosynthesised materials. Nanomaterials can be naturally occurring, inadvertently generated or engineered. ENMs are intentionally produced and exhibit unique electrical, magnetic, optical, antimicrobial and other properties. Silver nanoparticles (AgNPs) are among the ENMs most often incorporated in nanofunctionalised consumer products. AgNPs have been incorporated into a diverse range of consumer products including plastics, soaps, pastes, metals and fabrics and have applications in water and soil remediation. In the Nanodatabase, currently 539 of the 3037 products listed contain silver. Although advances in nanotechnology and the continued development of novel ENMs are expected to lead to significant societal benefits, there is increasing concern that the unique properties of ENMs may result in potential hazards for both humans and the environment. ENMs can be released into various environmental matrices during their production, use and disposal. ENMs pose a potential risk to human health through ingestion, inhalation and contact. To date, the concentrations of AgNPs in the aquatic environment have primarily been estimated through modelling, with predicted environmental concentrations in the ng/l range. This is largely because of a dearth of appropriate detection methods. The aims of this 3-year research project were to (1) develop and implement methods for the detection of AgNPs in water; (2) determine the toxicological properties and environmental fate of AgNPs in the aquatic environment and (3) develop risk assessment protocols that can be used to evaluate the environmental fate of and likely risk from AgNPs in aquatic pathways. The suitability of activated charcoal as a capture material for AgNPs from water was examined. Samples of 100 μg/l of AgNPs were initially generated and exposed to activated charcoal for 24 hours to examine the ability of charcoal to capture AgNPs. The decrease in silver concentration was measured using an inductively coupled plasma mass spectrometer. Following initial investigations, the surface area of the charcoal was increased, first, with a pestle and mortar and, second, by milling. The increased surface area of the milled charcoal increased the capture of the AgNPs from 11.9% to 63.6%. A hydrochloric acid leaching procedure was developed that successfully removed the captured silver, allowing the fraction captured by the charcoal to be quantified, with an average recovery rate of 94.8%. The results show that milled activated charcoal can successfully capture AgNPs from water samples. Activated charcoal therefore represents a cost-effective material for the remediation of waters impacted by AgNPs or other nano-wastes. A multi-trophic test battery that included three trophic levels was adopted to assess the ecotoxicity of AgNPs and silver nitrate (AgNO3) to the algae Pseudokirchneriella subcapitata, the crustacean Daphnia spp. and the cnidarian Hydra attenuata. The standard medium (Jaworski’s medium) and an ethylenediaminetetraacetic acid (EDTA)-free medium (Chu#10) were tested concurrently. An approximately 10-fold improvement in test sensitivity using EDTA-free medium was observed overall. No significant difference between the toxicity of AgNP and x Detection, Toxicology, Environmental Fate and Risk Assessment of Nanoparticles in the Aquatic Environment (DeTER) the toxicity of AgNO3 was observed. Both Daphnia pulex and Daphnia magna were compared using AgNO3 and AgNPs. Daphnia pulex, with a 24-hour half-maximal inhibitory concentration (IC50) of 9.3 μg/l, was less sensitive to AgNO3 than Daphnia magna, with an IC50 of 1.22 μg/l. When tested with AgNPs, both species yielded similar results, with an IC50 of 7.85 μg/l for Daphnia magna and 4.2 μg/l for Daphnia pulex. As these IC50 values were substantially higher than the predicted environmental concentrations, sub-lethal end points were investigated. Fecundity was assessed in Daphnia magna, with the number of offspring reduced by 50% after 14 days and 75% after 28 days when cultured in 100 ng/l of AgNPs. This demonstrates that the effects of AgNPs may be seen at close to environmentally relevant concentrations on population numbers rather than single individual organisms. Assessment of the ecotoxicity of AgNPs using Hydra attenuata gross morphology as the end point yielded a 96-hour half-maximal effective concentration (EC50) of 29 μg/l for AgNPs. The effect of silver on the regeneration of Hydra attenuata was the most environmentally relevant bioassay investigated as it is very sensitive and robust and Hydra attenuata represents benthic dwellers likely to be exposed to higher concentrations of AgNPs. The risk assessment involved a number of interlinking stages. Stage 1 included a review of the state of the art regarding natural attenuation processes that affect ENPs in the natural aquatic environment and a review of current risk assessment strategies. In stage 2 a suite of laboratory-scale studies were conducted to better characterise the aggregation potential of AgNPs. The behavioural indications derived from stages 1 and 2 were used to develop an aquatic risk model (stage 3), which was used to characterise the likely residual levels of AgNPs in surface waters. Estimated initial values indicated a mean AgNP concentration of 4.34 × 10–2 μg/l and this was assumed as a worst-case scenario for surface water concentrations in Ireland and used as an initial input value in the risk model. Seasonal factors were incorporated in the risk model to account for potential fluctuations in organic matter and ionic strength, which have been identified as key influencers of particle stability and eventual fate in natural water systems. The predicted results from the model developed indicate that citrate-coated particles underwent greater removal than polyvinylpyrrolidone (PVP)-coated AgNPs in both stream water and lake water, with predicted removal rates after 7 days of ≈70% (stream water) and ≈67% (lake water) for citrate-coated particles and ≈45% (stream water) and ≈50% (lake water) for PVP-coated AgNPs. Predicted aquatic concentrations of AgNPs were compared with toxicity data from project partners to establish if a risk is posed by current estimated concentrations of AgNPs in natural waters. The EC50 values from primary producer (algae) to primary consumer (Daphnia pulex) to secondary consumer (Hydra attenuata) exposed to PVP-coated AgNPs were compared with persistent concentrations of AgNPs in freshwater systems. The concentrations of AgNPs used in the model were at levels deemed unlikely to have toxicity concerns to aquatic organisms (mean levels in water of 4.34 × 10–2 μg/l). Therefore, at current predicted water concentrations, AgNPs are unlikely to present a toxic concern to the aquatic food chain. Stages 4 and 5 used information generated to assess potential human exposure through drinking water. The model incorporated estimated removal rates for the differing treatment processes. Risk to human health was calculated based on water consumption and potential exposure to residual AgNPs using the hazard quotient (HQ). The HQ is a ratio of the possible exposure to a particular substance and the level at which it is expected that no adverse effects will occur. If the calculated HQ is less than 1 then it is expected that no adverse health effects will result from exposure. The predicted HQ indicated that there was no existing risk through the consumption of drinking water (HQ of 3.24 × 10–7 for males and 3.84 × 10–7 for females). However, the increased industrial usage of nanomaterials in many sectors, in conjunction with the persistence of AgNPs during drinking water treatment, suggests the need to constantly monitor levels and re-assess exposure through drinking water into the future.