Aggregation and dispersion of silver nanoparticles in exposure media for aquatic toxicity tests

Abstract Silver nanoparticles (AgNPs) are currently being very widely used in industry, mainly because of their anti-bacterial properties, with applications in many areas. Once released into the environment, the mobility, bioavailability, and toxicity of AgNPs in any ecosystem are dominated by colloidal stability. There have been studies on the stability or the aggregation of various nanoparticles (NPs) under a range of environmental conditions, but there is little information on fully characterised AgNPs in media used in (eco)toxicity studies. In this study, monodisperse 7, 10 and 20 nm citrate-stabilised AgNPs were synthesised, characterised and then fractionated and sized by flow field-flow fractionation (FFF) and measured with dynamic light scattering (DLS) in different dilutions of the media recommended by OECD for Daphnia magna (water flea) toxicity testing. Stability of NPs was assessed over 24 h, and less so over 21 days, similar time periods to the OECD acute and chronic toxicity tests for D. magna . All particles aggregated quickly in the media with high ionic strength (media1), resulting in a loss of colour from the solution. The size of particles could be measured by DLS in most cases after 24 h, although a fractogram by FFF could not be obtained due to aggregation and polydispersity of the sample. After diluting the media by a factor of 2, 5 or 10, aggregation was reduced, although the smallest NPs were unstable under all media conditions. Media diluted up to 10-fold in the absence of AgNPs did not induce any loss of mobility or fecundity in D. magna . These results confirm that standard OECD media causes aggregation of AgNPs, which result in changes in organism exposure levels and the nature of the exposed particles compared to exposure to fully dispersed particles. Setting aside questions of dose metrics, significant and substantial reduction in concentration over exposure period suggests that literature data are in the main improperly interpreted and nanoparticles are likely to have far greater biological effects than suggested thus far by poorly controlled exposures. We recommend that the standard OECD media is diluted by a factor of ca. 10 for use with these NPs and this test media, which reduces AgNP aggregation without affecting the viability of the text organism.