In the European Union, nanomaterials are regulated through different pieces of sectoral legislation.This legislation often requires risk assessments and thus reliable characterization data, for whichregulatory guidance generally recommend electron microscopy. The guidance provides best practicesfor measurements but lacks requirements on how many particles to measure. Using transmissionelectron microscopy data of nanomaterials, a strategy based on repeated subsampling is proposedto establish, for different particle size and shape measurands, mathematical relationships betweenparticle count and precision, and subsequently to determine the minimum particle count. Our resultsconfirm that the minimum particle count generally depends on the width of the size and shapedistributions and that the median of the distribution can be determined with the highest precisioncompared to other percentiles. Upon combining the precision uncertainty related to particle numberwith uncertainties from other sources, such as sample preparation, calibration and trueness, wereach an optimal particle count above which additional particle measurements only yield negligibleimprovements to the combined measurement uncertainty. Our findings offer an experimentalapproach for determining the minimum particle count to measure particle size and shape by electronmicroscopy. It enables efficient analyses and facilitates compliance with legislation addressingnanomaterials across various application domains.
