Recrystallization during Thermo-mechanical process

ripening phenomena, dislocations, grainboundary topology
Post Reply
anoopnn11
Posts: 1
Joined: Sun May 23, 2021 8:34 am
anti_bot: 333

Recrystallization during Thermo-mechanical process

Post by anoopnn11 » Thu Sep 23, 2021 9:26 pm

Dear Bernd,

I am pretty new to MICRESS tool. I want to predict microstructure evolution during the solid-state transition of friction stir welding (FSW) in Aluminum alloys. FSW is the thermo-mechanical process where material undergoes severe plastic deformation, and solid-state transition take place at a high temperature below the melting point.

The weld region/nugget region undergoes dynamic recrystallization, where both thermal cycle and plastic strain contribute. At the same time, the heat-affected zone (HAZ) undergoes static recrystallizations adjacent to the stir zone. FEM simulations provide the required thermal cycle and plastic strain, given as input for phase-field modeling representative volume element in 2D.

Some of my concerns are as follows.

How could I consider the stored energy of plastic strain in the phase-field model as a boundary condition?

I had referred to some of the examples i.e., ReX_deterministic/ReX_local_Humphreys/ReX_local_recovery/ReX_mean_dislocation/ReX_random. Which approach would be a more appropriate choice for my case.

What assumptions need to be considered for good correlative results?

Any additional suggestions from you on similar problems solved would be helpful to solve my case. Looking forward to your kind response and valuable feedback/suggestions.

Thank you.

Best Regards,
Anoop

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Recrystallization during Thermo-mechanical process

Post by Bernd » Thu Sep 23, 2021 11:43 pm

Dear Anoop,

Welcome to the MICRESS forum.

The recrystallisation model in MICRESS is a phenomenological approach which has been developed for the case of static recrystallisation. Thus, we typically start from an already deformed material and define dislocation density or stored energy per grain as initial state, or a field of dislocation densities (in single phase case only). Recrystallisation is initiated by nucleation of new grains typically without dislocations which grow into deformed grains due to the difference of stored energy. Temperature information can be reflected by T-dependent nucleation and interface mobility/energy. In this simple setting, stored energy would always decrease.

Meanwhile, we have developed some tools for manipulation of grains during runtime which can also be used to include or alter dislocation densities dynamically. Unfortunately, I am not really expert in the field of dynamic recrystallisation in general (and FSW specifically), but I could imagine that strain rates as boundary condition could be translated into a rate of dislocation density production, which could be obtained by this mechanism (it would be "add_to_grain" in the nucleation options which indirectly allows changing properties of existing grains) an which could drive an ongoing recrystallisation process. However, we did not do something similar yet, and it is thus also not reflected in the MICRESS examples. (I only once simulated a process with several deformation steps and intermediate heating, where the deformation steps were simplified to the periodic introducing dislocation densities with random distribution). Enhanced solute diffusion due to dislocations has also been introduced lately in MICRESS, so that effects of plastic strains on the element distribution could also be addressed.

In principle we also could imagine to add missing functionalities according to your specification (given the effort would be not too high and that they would provide a useful extension to the MICRESS capabilities for other users also...)

Maybe I could give you more specific hints if you would describe with a bit more detail which phenomena of FSW are typically observed at the microstructure level and what exactly would be the focus of your studies.

Best wishes
Bernd

Post Reply