Size detection limits of spICP-MS for analysis of nanoparticles in environmental media

Sammanfattning: Nanoparticles (NP) are widely integrated in our everyday lives, such as consumer products, health, agriculture, food, etc. This has resulted in a growing concern for their possible harmful effect on human health and the environment and the demand for characterization. To measure the emission of NPs and exposure levels in the environment, the methods have to be able of quantifying and sizing particles of interest at parts per billion (ppb) level concentrations or lower. However, their small size and expected low concentrations against high backgrounds of naturally occurring other particles make characterization impossible with non-specific measurement tech-niques. Single particle ICP-MS (spICP-MS) is a promising technique for quantifi-cation of both size and number concentration of NPs. The aim of this thesis is to evaluate the analytical capabilities of the Perkin Elmer Nexion 300 mass spectrometer for spICP-MS analysis. Gold nanoparticles were used as model NPs for all experiments and dissolved Au was added mimicking interferences or dissolved backgrounds often encountered in environmental samples. The particle size detection limit (DL) was determined and the factors determining the smallest measurable number concentration such as background count rate and particle contamination are discussed. Validation of outlier detection parameters has been done and it has been found that the more conservative n = 5 for  + n' gives a smaller false positives count rate than the more common n = 3, while still not counting too many false negatives. Size detection limits vary between 22 and 32 nm and concentration limits of detection (LOD) are in the range of several ng L-1 for smaller (40 nm) particles to ca. 40 ng L-1 for bigger (60 and 100 nm) particles, which is at least 10 times higher than the theoretical LOD. This made it impossible to detect particles of the size 5, 10 and 20 nm. In general, there was a lot of variation between size DLs obtained with a published equation and DLs were sometimes lower than found possible. Setting the DL to an intensity of 2 counts gave more consistent and realistic results. Adding dissolved gold (50 and 75 ppb) to artificially increase the background showed it was impossible to separate the particle events from this high background. Dwell times of 50 s were used, but it was found that these needed to be merged to bigger dwell times to get correct results in the outlier detection algorithm. Further research to detect NPs in the presence of high dissolved back-grounds is required, as well as research for DLs and LODs of other types of nanoparticles, such as NPs with different isotopes and / or spectral in-terferences. Despite these shortcomings, spICP-MS has shown to be very promising for future quantification of nanoparticles in environmental media.