气候变化领域集成服务门户

Detecting and quantifying engineered nanoparticles (ENPs) in complex environmental matrices requires the distinction between natural nanoparticles (NNPs) and ENPs. The distinction of NNPs and ENPs for regulatory purposes calls for cost-efficient methods, but is hampered by similarities in intrinsic properties, such as particle composition, size, density, surface chemistry, etc. Titanium dioxide (TiO2) ENPs, for instance, are produced in very large quantities but Ti also commonly occurs naturally in nano-scale minerals. In this work, we focus on utilizing particle size and composition to identify ENPs in a system with a significant background concentration of the target metal. We have followed independent approaches involving both conventional and state-of-the-art analytical techniques to detect and quantify TiO2 ENPs released into surface waters from sunscreen products and to distinguish them from Ti-bearing NNPs. To achieve this, we applied single particle inductively coupled plasma mass spectrometry with single-element (spICPMS) and multi-element detection (time-of-flight) spICP-TOFMS, together with transmission electron microscopy (TEM), automated scanning electron microscopy (autoSEM), and bulk elemental analyses. A background concentration of Ti-bearing NPs (approximately 5 x 10(3) particles per ml), possibly of natural origin, was consistently observed outside the bathing season. This concentration increased by up to 40% during the bathing season. Multi-element analysis of individual particles using spICP-TOFMS revealed that Al, Fe, Mn, and Pb are often present in natural Ti-bearing NPs, but no specific multi-element signatures were detected for ENPs. Our data suggests that TiO2 ENPs enter the lake water during bathing activities, eventually agglomerating and sedimenting. We found adhesion of the TiO2 ENPs to the air-water interface for short time periods, depending on wind conditions. This study demonstrates that the use of spICP-TOFMS and spICPMS in combination with other conventional analytical techniques offers significant advantages for discriminating between NNPs and ENPs. The quantitative data produced in this work can be used as input for modeling studies or as a benchmark for analysis protocols and model validations.