Add a kernel (|\nabla \|^2) in phase field module

[kkq561]

Rationale

I would like to include a kernel (|\nabla |^2) in phase field module for phase field model of oxidation (oxide kinetic growth problem via cation/anion vacancies and electrons).

Description

In oxide kinetics problem, we need to add an expression (|\nabla |^2) in A-C equation. Specifically, it is for including electrostatic contributions in phase field simulation (|\nabla \electric field|^2). The detail information is described in the recent paper (Sherman and Voorhees, Physical Review E 2017, https://journals.aps.org/pre/abstract/10.1103/PhysRevE.95.032801).

Impact

This will be useful for users to directly perform phase field model of oxidation using phase field module in moose rather than connecting similar kernels in other modules (e.g., JouleHeatingKernel in heat conduction module).

[permcody]

Don’t do that, just turn on both the phase field AND hear conduction modules in your app (do this in the Makefile). On Fri, Dec 1, 2017 at 10:35 PM kkq561 notifications@github.com wrote:

Rationale

I want to add the JouleHeatingSource kernel in heat_conduction module to phase_field module for including |\nabla \psi|^2 in A-C equation for oxidation modeling. Description Impact

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/idaholab/moose/issues/10347, or mute the thread https://github.com/notifications/unsubscribe-auth/AC5XIII-dK39lboqiiovpeRTmBPvimi-ks5s8OG2gaJpZM4QzLvH .

[dschwen]

@permcody, I discussed this with @laagesen. The issue is a bit misleading. We'd like to add a kernel for a very similar PDE to phase field because we think using the "Joule Heating" kernel would be very confusing. All constants and coupled variables have names that are specific to a Joule heating problem. We'd have to supply some PF order parameter as the "Temperature" in the way we use that PDE - that would look odd.

[dschwen]

I'm open to other suggestions. Do we have a general policy for structurally identical PDEs that have coefficients that can mean multiple different physical things?

[laagesen]

Kyoungdoc, I think it would help if you changed the wording of this issue, since you are not literally adding the JouleHeatingSource to the phase-field module, but adding a kernel with similar physics with different materials as input. Can you please change the issue title and the wording of the text so that it focuses more on the problem you are working on, rather than referencing the JouleHeatingKernel?