Annealing a thin film can have different consequences depending on the nature of the thin film. For example, if the film was deposited eptiaxially on a substrate, annealing may relief strain due to lattice mismatch and also modify the surface morphology. As a rule of thumb, annealing at 1/3 of the melting temperature of the bulk material, should in principle improve the surface roughness. In some cases, annealing can favor formation of specific structural phases (e,g, annealing FePd or FePt thin films may lead to the formation of the L!0 ordered phase). It is also possible that diffusion processes can be triggered by annealing. So the bottom line is what is the desired effect.

Dear Phuong Anh Ngoc Vo,

actually the main purpose of post annealing to any inorganic thin film is to improve its surface quality. In broad vision, due to annealing (better in the atmosphere of some ambient gas), the nucleation occurs which causes to enhance the grain size. Therefore, a vital change in the surface morphology with reduced defects can be observed.

Postannealing and heat treatment are used for diffusion of amorphous

precursors and phase formation.

PULSED LASER DEPOSITION OF THIN FILMS, Edited by Robert Eason

Annealing a thin film can have different consequences depending on the nature of the thin film. For example, if the film was deposited eptiaxially on a substrate, annealing may relief strain due to lattice mismatch and also modify the surface morphology. As a rule of thumb, annealing at 1/3 of the melting temperature of the bulk material, should in principle improve the surface roughness. In some cases, annealing can favor formation of specific structural phases (e,g, annealing FePd or FePt thin films may lead to the formation of the L!0 ordered phase). It is also possible that diffusion processes can be triggered by annealing. So the bottom line is what is the desired effect.

In addition to above, annealing of thin films could also have an impact upon the composition of films. For example, for sulfide based layers (In2S3, MoS, etc.), it is easy to form different phases or achieve improper stoichiometry (excess sulfur), annealing treatment may evaporate excess sulfur and help achieve proper stoichiometry.

there is no straight forward answer for post annealing, it depends on material, process parameters and what you really want from it.

a) it can reduce stress/strain generated during fabrication of thin film, provided film is not constrained during annealing,

b) can increase the crystallinity of films, depending upon the working temperature. for instance, for polymer films, working above tg but less than tm may improve overall crystallinity.

c) it can also affect the bonding between the film and the substrate,etc

In most cases, the effect of post annealing on thin film‎

‎1-‎ Crystalline property and grain size of the films were found to increase after ‎annealing. ‎

‎2-‎ The optical band gap of thin film increase due to the reduction of crystal ‎defect

‎3-‎ Annealing of thin films improves (increases) the transmission spectra in ‎transparent and medium absorption region thus the refractive index decreases.‎

‎4-‎ Annealing effect improves the film quality with reduced roughness.‎

The most important effect of the controlled annealing of a  hetero epitaxially grown thin film is to relax  the  stored elastic strain energy associated with the lattice mismatch strain through the formation and rearrangements  of line defects at the interface into their stationary equilibrium positions,  thereby to reduce the unwanted  surface roughness and undulations caused  by the  athermal relaxation of  elastic strain energy through the spontaneous surface morphological evolution.  

I agree with most of the comments here. Basically the answer of that question depends on the material that we are talking about.

I would like to add a couple of points that I think no one has mention yet. 

First, annealing parameters (especially annealing environment) can affect the vacancies of the crystal. For example, the enhance in conductivity when SnO2 crystal is anneal in O2 environment can be attribute to the creation of oxygen vacancies (we can say that this is still a live controversy going on). On the other hand, annealing in H2 environment can reduce the conductivity. So in order to say what does annealing do, you need to specify many things starting with the material that we are annealing.

Second, it is true that in most cases annealing releases the stress in the crystal structure, but there are cases were the opposite happens. As example you can read papers about LCMO on different substrates, where in some cases the electronic resistance is increase with post-annealing process.

Anyway, a short answer is hard for your question, but hope you have a better idea after all the comments.

All our colleagues especially Dr. Lukaszew have successfully answered the questions of WHAT ? but not for  the HOW?  especially one comes to answering the effects of post heat treatment on the surface roughness through the surface drift diffusion and its main driving force, which acts spontaneously, and it  is solely associated with the stored elastic strain energy in isochoric body,  regardless of its origin.

post annealing has lot of advantages but its make the film to get cracks, shrinkage, and make it brittle when the process condition is not optimised. so we must need to take care of process condition.

It depends on the annealing temperature. Typically, annealing at 1/3 of the melting temperature (in K) promotes diffusion in 2D so the effect is smoothening of the surface. Higher temperature may have detrimental effects since it may disturb the interface with the substrate. But again, it depends on the thin film material as well as the substrate material.

Dear all, annealing strongly depends on the interaction (potentials) of the film and the substrate materials. If the interaction os films and subtrate atoms is weak, heating will promote the formation of clusters or islands of the film material. If the interaction is strong, a smoothing of the film is possible. But, as said before, temperature regimes are difficult to predict! Kind regards, Dirk