Skip to content

Add scalar frequency-dependent permittivity models - #877

Draft
hughcars wants to merge 4 commits into
mainfrom
hughcars/dispersive-material
Draft

Add scalar frequency-dependent permittivity models#877
hughcars wants to merge 4 commits into
mainfrom
hughcars/dispersive-material

Conversation

@hughcars

@hughcars hughcars commented Aug 10, 2026

Copy link
Copy Markdown
Collaborator

Summary

Adds a scalar frequency-dependent Permittivity model for driven and 3D eigenmode simulations:

"Permittivity":
{
  "HighFrequency": 2.08,
  "Terms":
  [
    {
      "Type": "Drude",
      "PlasmaFrequency": 1.0,
      "CollisionFrequency": 0.1
    }
  ]
}

Supported additive terms are:

  • Drude
  • Debye
  • Lorentz
  • explicit pole-residue
  • Djordjevic–Sarkar

Legacy scalar and principal-axis Permittivity inputs remain unchanged.

Frequency-dependent scalar coefficients reuse one unit vector-mass operator per material support. The high-frequency permittivity remains in the ordinary mass matrix, exact zero poles are folded into conductivity, and nonlinear terms are evaluated directly in the frequency-dependent operator.

The change also adds documentation and a Drude cylindrical-cavity example with serial and parallel regression coverage.

Validation

  • Spack build/install with test dependencies and -j4 passed
  • Non-regression, non-long CTest matrix passed: 185 tests passed and one expected source-schema test skipped
  • Eight-rank cylinder_cavity_drude and existing floquet_lumped regressions passed locally
  • Source schema validation, full documentation build, formatting, and diff checks passed
  • Full standard CI matrix passed across Linux, macOS, Spack, GPU, containers, documentation, style, and CodeQL
  • Long tests were not requested or run
  • Final Palace-specific review: no findings

@hughcars hughcars added the no-long-tests This PR does not require the long tests to be merged label Aug 11, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

no-long-tests This PR does not require the long tests to be merged

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant