Inquiry Regarding Two-Phase Flow and Partial Melting Models in ASPECT

1

Dear ASPECT Community,

I hope this message finds you well. I am currently working on a project that explores two-phase flow models, and I would like to inquire about the capabilities of ASPECT in this area.

Specifically, I am referring to the paper “https://doi.org/10.1093/gji/ggw329,”(Dannberg and Timo Heister,2016)which demonstrates a partial melting two-phase flow model. As I understand it, this work involves modeling the interaction between two phases, and I am interested to know if ASPECT is capable of implementing similar models, particularly those involving the interactions between metal and silicate melt components.

In addition, I am also inspired by the work of Korenaga & Marchi (2023) “https://doi.org/10.1073/pnas.2309181120”, which examines the percolation of metal in the mantle. I am interested in using ASPECT to replicate their findings related to metal percolation and to investigate the impact of metal percolation in the context of early Earth mantle differentiation.

Could you please clarify whether ASPECT can be extended or used to model such two-phase flow problems, particularly with a focus on metal and silicate melts?

I appreciate your time and look forward to your guidance.

Best regards,
Ting He

2

Hi Ting,

Thank you for posting the question to the forum.

Could you please clarify whether ASPECT can be extended or used to model such two-phase flow problems, particularly with a focus on metal and silicate melts?

Assuming the two-phase flow equations currently implemented in ASPECT are suitable for this class of problem, then what you would need to implement is a reaction model for the metal-silicate-silicate melt system. However, a quick look at the Korenaga and Marchi paper indicates this is a three phase problem (i.e., one needs to solve for the velocity of three distinct phases?)?

Aside, there is a new framework in ASPECT for combining material models for solid deformation with reactive fluid transport models (source code here, see reactive_fluid_transport_zero_solubility.prm in tests folder), which should simplify the overall process. The main new piece of code that would need to be added to this framework is a new reaction model.

I hope this helps.

Cheers,
John