Average Composition Conservation

solid-solid phase transformations, influence of stresses and strains
Post Reply
Posts: 3
Joined: Sat Apr 21, 2018 12:02 am
anti_bot: 333

Average Composition Conservation

Post by jmueller » Thu Jan 10, 2019 1:56 am

I'm simulating the nucleation and growth of a solid phase at grain boundaries in a solid matrix.
The nucleating phase is expected to grow with an enriched composition that is partitioned from the matrix.

I've noticed that the 'average concentration' of the partitioning component increases quite substantially over the course of the simulation.
It appears the the nucleating phase is growing at an enriched composition without requiring the diffusion of the enriching component to the interface, leaving the matrix with (almost) the nominal composition. (i.e. concentration is not conserved in the simulation.)

I'm coupling to a T-C ges file.

Is there a way to restrict the composition for a simulation so that the average concentration is conserved?


Posts: 997
Joined: Mon Jun 23, 2008 9:29 pm

Re: Average Composition Conservation

Post by Bernd » Fri Jan 11, 2019 12:21 am

Dear Josh,

The average composition of the domain should always be conserved if proper boundary conditions are used for the concentration field (isolated, symmetric, periodic), while gradient or fixed conditions of course would change average composition. The same can happen if a 1d-extension of the concentration field is used ("1d_far_field").
How did you check the average composition? Is it changing in the screen output, where it is calculated at each output time step (i.e. the variation is not close to zero)? If this is the case, and it is not because of the boundary conditions, then it looks like huge numerical trouble. Do you see huge or infinite concentrations anywhere?


Post Reply