Extracting Metallic Nanoparticles from Soils for Quantitative Analysis: Method Development Using Engineered Silver Nanoparticles and SP-ICP-MS

The lack of an efficient and standardized method to disperse soil particles and quantitatively subsample the nanoparticulate fraction for characterization analyses is hindering progress in assessing the fate and toxicity of metallic engineered nanomaterials in the soil environment. This study investigates various soil extraction and extract preparation techniques for their ability to remove nanoparticulate Ag from a field soil amended with biosolids contaminated with engineered silver nanoparticles (AgNPs), while presenting a suitable suspension for quantitative single-particle inductively coupled plasma mass spectroscopy (SP-ICP-MS) analysis. Extraction parameters investigated included reagent type (water, NaNO₃, KNO₃, tetrasodium pyrophosphate (TSPP), tetramethylammonium hydroxide (TMAH)), soil-to-reagent ratio, homogenization techniques as well as procedures commonly used to separate nanoparticles from larger colloids prior to analysis (filtration, centrifugation, and sedimentation). We assessed the efficacy of the extraction procedure by testing for the occurrence of potential procedural artifacts (dissolution, agglomeration) using a dissolved/particulate Ag mass ratio and by monitoring the amount of Ag mass in discrete particles. The optimal method employed 2.5 mM TSPP used in a 1:100 (m/v) soil-to-reagent ratio, with ultrasonication to enhance particle dispersion and sedimentation to settle out the micrometer-sized particles. A spiked-sample recovery analysis shows that 96% ± 2% of the total Ag mass added as engineered AgNP is recovered, which includes the recovery of 84.1% of the particles added, while particle recovery in a spiked method blank is ∼100%, indicating that both the extraction and settling procedure have a minimal effect on driving transformation processes. A soil dilution experiment showed that the method extracted a consistent proportion of nanoparticulate Ag (9.2% ± 1.4% of the total Ag) in samples containing 100%, 50%, 25%, and 10% portions of the AgNP-contaminated test soil. The nanoparticulate Ag extracted by this method represents the upper limit of the potentially dispersible nanoparticulate fraction, thus providing a benchmark with which to make quantitative comparisons, while presenting a suspension suitable for a myriad of other characterization analyses.