Environmental context. Soils are the environmental compartment likely to be exposed most to manufactured nanoparticles, but there is no method available at present to assess their retention, which determines potential mobility and bioavailability. Optimisation and application of a method to determine retention values for silver (Ag) and cerium oxide (CeO₂) manufactured nanoparticles in soils found in many cases that they differed from the partitioning of their bulk and soluble counterparts. Wider application of this method can assist in comparing the risk of many different manufactured nanoparticles with other contaminants in soil systems and model their relationship to soil properties.

Abstract. Methods to study the retention of manufactured nanoparticles (MNP) are lacking for soils that are likely to be increasingly exposed to MNP. In this study we present, for the first time, a method to determine retention values (K_r) of Ag and CeO₂ MNP, that can be ranked among solid–liquid partitioning (K_d) values of bulk (micrometre-sized) forms, soluble salts and other possible contaminants of soils. After method optimisation, suspensions containing 1.24 mg kg^–1 Ag as Ag MNP and 1.30 mg kg^–1 Ce as CeO₂ MNP were added to five soils. More than 7% of Ag MNP occurred as soluble Ag^I after 24 h and the range of K_r values of Ag MNP (77–2165 L kg^–1) and CeO₂ MNP (1.1–2828 L kg^–1) contrasted with K_d values of soluble Ag^I, Ce^III and Ce^IV salts and bulk Ag and CeO₂ powders in different soils.