First you have to know that the crystallite size is different than particle size. A crystal is the building block of a any particle so its a obvious thing that a crystallite size is always less than that of particle
On the subject, I am dealing with an issue about this. There is this SEM image a student sent me, in which particles appear to have about 50-60 nm. But when I try to calculate crystallite size by GSAS-II or Williamson-Hall method, the crystallite size is about 60-70 nm. But how is possible that the crystallite size is the same measured by SEM?
Anand Sagar I would not say “always”. Some materials are intentionally synthesized to achieve large crystallite sizes in order to minimize size broadening, as in some SRM where cry size of 0.5-1 microns (determined by the FPA) are seen.
Ricardo Tadeu Maia I would not say that this is a large difference. Both measurements involve errors; what is the deviation of the cry size measured by SEM?
For your XRD determination, I suggest using a physically-based method such as the fundamental parameters approach (FPA). Cry size measurement by XRD requires a good knowledge of your equipment, so you can properly differentiate peak broadening from the instrument (e.g. axial divergence) and from your sample (basically cry size and strain). For this purpose, you should measure a known SRM (e.g. LaB6) and model your equipment.
Professor Paulo Ricardo, thanks for your answer. What I wanted to say was that we know that crystallite sizes can range between 2000 nm ( a very large crystallite, like Silicon NIST) to a few nanometers ( like some nanoparticles). And then in this SEM , I can see structures with ~50nm diameter. My question is: this what I see in SEM, are the crystallites?
TEM is localized measurement while XRD is global measurement. Therefore, they should not be the same. Also, what method you have used to calculate the crystal size? have you used modified Williamson-Hall method or Scherrer Equation?. I recomend using Uniform Deformation Model (UDM) or Uniform Stress Deformation Model (USDM) to calculate the crystal size. Also, using synchrotron xrd will be much accurate since the crystal size is very small.