Should I Learn Fenics for this Problem? Phase-Field Simulations

Hello Fenics users,

I am trying to study Polymer Crystallization using the phase field technique. I don’t consider myself to be an expert programmer so the notion of coding a finite element model from scratch seems to be both silly and not conducive to me ever graduating with a PhD in any reasonable amount of time. I am also looking at a code called PRISMS-PF which is implemented in C++ that I am comparatively less proficient in (I would consider myself a fairly proficient Python programmer). The system that I am looking to simulate would be a coupled Allen-Cahn and and Heat equation system. But I have learned that the code I’m trying to use may not be able to simultaneously solve the heat-transfer and Allen-Cahn equation that turns out to be very important when modeling polymer crystallization. From what I know about Fenics, it seems to be VERY extensible and could be well suited to solve problems like this. I’ve seen a few posts implementing phase field models for other applications, but not crystallization. Is that something that could be done? Ideally I’d like a model that could capture the fact that the amorphous phase of the polymer and the crystalline phase of the polymer would have different heat transfer coefficients. I’ve already spent a ton of time researching this other code, but if it can’t handle this aspect, then it might be a complete dealbreaker, so any advice would be most useful.

Thanks!

Erik

Hi,
This should be totally possible. Just write your PDEs in weak form and you are read to go with FEniCS.

3 Likes

I second Jeremy’s comments, I’m currently working on a few applications for this exact scenario. You can implement the coupled phase-temperature models in FEniCS in a number of ways including mixed functionspaces and in a staggered approach. I highly encourage you to peruse literature and experiment to get a sense of the affect of the mobility coefficients and surface energy on the interface between phases.