LAUSR

Abstract The identification of nanomaterials for regulatory purposes in Europe requires a statistically significant comparison of the number metrics median diameter to the 100 nm criterion. Methods should be robust against minor imperfections of the implementation, transferable and reproducible between laboratories, and applicable to a wide range of diameters and material compositions. Sample preparation and metrics conversion are additional challenges that must be included in validation. Here we report on the validation of a protocol for analytical ultracentrifugation with fixed and ramped speed, specifically intended to serve the European Commission recommended definition, as relevant for REACH regulation, cosmetics and food labelling and national inventories. All the determined measurement uncertainties remain below 12% for volume and number metrics median diameters, in both laboratories. Traces of sub-100-nm particles with around 2% mass contribution (but > 50% number contribution) were reproducibly quantified. For powders, sample preparation is a critical step, and sonication intensities above 0.4 W/mL are recommended. The ramp operation can reduce user bias by eliminating the choice of options in data acquisition. The identification of nanomaterials and non-nano-materials by AUC in either fixed or ramp speed is consistent with TEM, excluding the platelet Kaolin material, but including the monodisperse silica, multimodal silica, a nanoform and a non-nano-form of BaSO 4 , irregularly shaped CaCO 3 and coated non-nano TiO 2 .