Dear Issac,

 please have a look at the answers of a similar question here on RG.

https://www.researchgate.net/post/What_is_the_reason_mechanism_behind_the_increase_in_grain_size_with_annealing_temperature_of_nanoparticles_or_semiconductors?_tpcectx=search

Dear Gerhard martens thanks for your kind reply. My concern is decrease in grain size upon annealing not increasing grain growth.

Recrystallization is a process where new (smaller) grains nucleate from an existing microstructure.  This usually occurs after 'cold-working' a metal to induce a large number of defects ("dislocations").  The annealing of a cold-worked metal can induce the nucleation of small grains (defect free).  If the annealing is long enough/high enough temperature continued growth of these new grains (defect free - therefore lower energy) may eventually become larger than the original grain size.  So one can expect annealing to induce a grain size reduction (initially, incomplete recrystallization) and then an ultimately (possibly) larger grain size.

What method do you use to determine the size of the grains? Whether the sample is single phase?

If in Your system have a place of decay, for example, a solid solution with the release of the secondary phases?

If You had originally amorphous-crystalline structure, it can be observed the formation (crystallization) of new phases with small grain size during  annealing. It is also possible that you have increased the degree of perfection of the grain: defects are removed, the pores are healed. The process of formation of the microstructure during heat treatment is strongly dependent on the mode of heating and cooling and the previous state of your alloy (for example, subjected to any deformation). The appearance of the finegrained structure can promote recrystallization with grinding of coarse grains in allotropical transition of a substance from one modification to another.

I have used scherrer formula to determine the grain size.. My material is polycrystalline nature.. Is there any references to your sayings or link to know about it further more..

So, You used the XRD data, and therefore You can check whether the changed phase composition. Depending on the grain size, also try independent methods of electron or optical microscopy of thin sections of Your samples.

As for publications, it is perhaps you can find a lot of books in this topic about the crystallization of steels and sintering processes, e.g., in ceramics, ferrites.

Thanks for the suggestions. 

Dear Issac Nelson,

I make some remarks.

1. I assume that the kinetics of nucleation & growth determine the grain size of the deposited film.

2. I assume that nucleation in your case is heterogeneous.

3. The temperature of first deposition seems essential, i.e. large undercoolings will yield smaller grains and smaller undercoolings can be expected to produce larger grains.

4. Ostwald ripening can explain grain growth due to annealing. 

 Indeed, in the crystallization process during fast cooling, it is observed the creation of small crystallites and even the formation of the amorphous phase. In this case, we're not talking about cooling, but about the annealing of the object with the already formed structure of the grains (with Tann below the melting point). And recrystallization (Ostwald ripeninig) is the consolidation and growth of larger grains at the expense of small ones (the rich get richer and the poor get poorer). And ultimately according Ostwald, the coarse grains should remain in the system. However, Isaac's question is that he is watching the reducing the size of the grains, but not their growth.

I also got a similar result with my experiment. The Indium droplets on Si (100) surface seems to break up to form smaller dots when annealed in short time but the grain size increase when the annealed time is increase. What the actual explanation?

amir che you can refer my paper

Article Impact of sequential annealing on the characteristics of the...

Thanks Issac Nelson on the sharing