When dealing with nanoparticles, size usually indicates a number average. This is obtained directly by measuring the diameter of a sufficiently large number of unique particles, using electron or probe microscopy techniques, or indirectly by analyzing data collected by X-ray diffraction or light scattering.

In both cases, the result is a hystogram representing the nanoparticle distribution, or dispersity, across various intervals of size. A perfectly monodispersed distribution means that all the nanoparticles have the very same size; for most applications, this is the ideal condition, as a collection of identical particles is expected to be absolutely homogeneous in terms of typically size-dependent physico-chemical properties.

Due to the complexity in the synthesis of nanoparticles, achieving narrow distributions of size is extremely challenging and must not be given for granted: for this reason, the indication of size should be always integrated with some indication of the degree of monodispersity, often provided as a ± variation.

A broad size distribution implies that for a given average size, there exist a significant population of nanoparticles with much larger and much smaller sizes. Each of these distribution tails will affect the final properties: larger nanoparticles will settle faster, block larger pores and channels, scatter more light, show a smaller specific surface; smaller nanoparticles will be more difficult to precipitate, show enhanced quantum behaviors and a higher specific surface, determining for instance larger corrosion effects. This dishomogeneity transfers directly to processing and performance issues, thereby affecting negatively the final product performance.