@M. Sahar: what you estimate by Scherrer equation should be better called "domain size". Be careful when you translate this number into something that you want to visualize: it is the length of the edge of a cube-shaped domain under the hypothesis that your powder is made of monodispersed cubes (scattering incoherently each other) under the further hypothesis that your peak is well approximated by a Gaussian. In all other conditions corrections should be applied and the meaning of the number becomes more and more faint...

If you check the reference I pointed to, you see that the cloud was already cleared:

"grains can be crystalline or amorphous. Coherently scattering domains (I don't like to call them crystallites as there is no universally accepted definition for this term) are the crystalline regions of material that scatter coherently (and in general they show no coherence with the neighboring ones). They are single crystal in nature (but they are in a powder as defined in the diffraction sense). A grain can contain multiple domains (consider e.g. two domains highly misoriented each other). The grain and the domain size are seldom the same in a polycrystalline aggregate but can be the same in a proper nanosized powder"

From XRD I are measuring the crystallite size: the smaller monocrystal in the system. It may, or may not be similar to the particle size of powder. To affect the XRD peak width the crystal size should be rather small, in order of 100nm or less. Then powders with such particle size tend to agglomerate, so the TEM investigation should be held to clarify this on your object.

Just a few general comments about all these "sizes". It is clear that particle size referes to the size of individual particles of a powder or particulate material. On the other hand, grain size is related to a microstructural element. The grains of a microstructure are the cristals formed during solidification or sintering and that are separeted one from another by grain bounderies. In principle, a grain would be a monocrystall, but it is known that a grain can be constituted of sub-grains separated by low-angle boundaries, like a mosaic. Then, the term crystallite is to distinguish the "macro" crystall or grain to the real small monocrystal that form the grain. I hope I have helped.

XRD measurement gives you the crystallite size, which is the smaller monocrystal you have in your samples. Yo can calculate the average crystallite size in a good approach by using the Scherrer's equation.

@Venkat: my suggestion is that you have a look at the various threads in here on the subject (you follow already some).

It seems that many people are a bit reluctant to use the search button when it comes to find the answer to a question. It took a few seconds to pull out a couple of threads on this (well I am biased...):

https://www.researchgate.net/post/Can_anyone_clarify_the_difference_between_Grain_size_particle_size_and_crystallite_size

https://www.researchgate.net/post/Application_of_Scherrers_equation2

Call them crystallites if you like, or call them little piggies : this does not change the fact that with diffraction you get information on the coherently scattering domains. You do what people are now doing with a free electron laser, but on a huuuuge amount of crystalline regions simultaneously

Answers given by J.Rodrgues are quite clear. What we are estimated using Scherrer method on the XRD data is the crystallite size while particle size refers to the smallest size of the powder form. Upon heat treatment of the sample around crystallization temperature, the multiphase of crystal start to grow, each is separated by the grain boundary. The grain size is the microcrystal of each phase. Hopefully, this would clear the cloud.

@M. Sahar: what you estimate by Scherrer equation should be better called "domain size". Be careful when you translate this number into something that you want to visualize: it is the length of the edge of a cube-shaped domain under the hypothesis that your powder is made of monodispersed cubes (scattering incoherently each other) under the further hypothesis that your peak is well approximated by a Gaussian. In all other conditions corrections should be applied and the meaning of the number becomes more and more faint...

If you check the reference I pointed to, you see that the cloud was already cleared:

"grains can be crystalline or amorphous. Coherently scattering domains (I don't like to call them crystallites as there is no universally accepted definition for this term) are the crystalline regions of material that scatter coherently (and in general they show no coherence with the neighboring ones). They are single crystal in nature (but they are in a powder as defined in the diffraction sense). A grain can contain multiple domains (consider e.g. two domains highly misoriented each other). The grain and the domain size are seldom the same in a polycrystalline aggregate but can be the same in a proper nanosized powder"

Thanks. Matteo Leoni for your nice explanation.

I agree with your statement that 'They are single crystal in nature (but they are in a powder as defined in the diffraction sense)'. However, we also apply X-ray Diffraction on planar surface of flat samples like thin films deposited on substrates or even bulk, sintered samples cut and polished into proper size and shape. Do you think we introduce a lot of error in peak position and peak broadening measurement in these samples which will eventually be reflected in estimating the structural parameters like lattice parameter, domain size, inhomogeneous lattice strain. defect concentration etc?