Quantification and classification of engineered, incidental, and natural cerium-containing particles by spICP-TOFMS

Abstract

Cerium containing nanoparticles (Ce-NPs) from geogenic and anthropogenic sources are frequently found in the environment, and the ability to determine the origins of Ce-NPs relies on the presence of other rare earth elements (REEs), such as La. In this study, we develop a scheme to classify individual natural, incidental, and engineered Ce-containing particles using spICP-TOFMS. Well-characterized CeO₂ engineered particles (Ce-ENPs), incidental particles (Ce-INPs) from ferrocerium mischmetal, and natural particles (Ce-NNPs) from ground minerals (bastnaesite and parisite) are used as a model particle system. Based on mixtures of these three Ce-NP types, we demonstrate that the measured signals of Ce, La, and Nd in Ce-NNPs follow Poisson statistics and have conserved element ratios. The Ce-INPs we measure have similar Ce : La mass ratios to those of the Ce-NNPs, and Ce : Nd mass ratios can be used to distinguish these two Ce-NP types. Based on this, we develop particle-type-specific detection limits (L_D,sp) for the measurement of La and Nd in Ce-NNPs. Our approach establishes L_D,sp values with defined confidence intervals to control false-positive particle-type assignments, and allows us to accurately classify engineered, incidental, and natural Ce-NPs down to effective spherical diameters of 32, 35, and 45 nm, respectively. In pure Ce-NNP suspensions, this approach accurately classifies 68% of all detected Ce-NPs with <4% false assignments. For ternary mixtures of Ce-ENPs, INPs, and NNPs, we classify 56% to 76% of all detected Ce-NPs. We demonstrate a linear response with increasing concentrations of Ce-INPs and Ce-ENPs across approximately three orders of magnitude and can classify particles down to ratios of ∼1 : 50 for anthropogenic : natural Ce-NPs.