A Python toolkit for fast computation of near and far-field quantities for plasmonic nanoparticles in the quasistatic and weakly retarded regime.

Overview

fastplasmon is a Python library for ultrafast dipolar electrostatic modeling of plasmonic nanoparticles with arbitrary geometry.

This method assumes that the surface charge density can be expanded as a series of multipole moments:

$$ \sigma(\mathbf{s}) = \sigma_{\text{monopole}}(\mathbf{s}) + \sigma_{\text{dipole}}(\mathbf{s}) + \sigma_{\text{quadrupole}}(\mathbf{s}) + \cdots $$

and retains only the dipole component

For strongly retarded regimes or higher-order multipolar effects, full-wave solvers (e.g., BEM, FEM, DDA) remain necessary.

This package currently supports single plasmonic nanoparticles composed of a single material embedded in a homogeneous medium, allowing the calculation of:

• LSPR spectra
• Geometry-only extraction of dipolar Neumann–Poincaré (K) eigenvalues
• Near-field mapping
• Efficient refractive-index sensitivity and biosensing analysis

Citing fastplasmon

If you use fastplasmon in your research, please cite the accompanying paper:

P. S. S. dos Santos, J. P. Mendes, J. M. M. M. de Almeida, L. C. C. Coelho, "Ultrafast Dipolar Electrostatic Modeling of Plasmonic Nanoparticles with Arbitrary Geometry" (2026).

https://doi.org/10.48550/arXiv.2601.16797

Core Features

• Support for arbitrary triangulated nanoparticle meshes
• Calculation of surface charge density
• Geometry-only extraction of intrinsic dipolar eigenvalues $\kappa_n \in (-1/2, 1/2)$
• Polarizibility tensor extraction
• Spectral response: Extinction, scattering, and absorption cross sections
• Inclusion of Modified Long-Wavelength Approximation (MLWA) for weak retardation (ka<0.7)
• Near-field evaluation: Field decay profiles for conformal dielectric coatings and effective refractive-index modeling for biosensing layers
• Computation time scales with $O(N^2)$ for matrix assembly, however, runtime is nearly independent of the number of wavelengths

Algorithmic Highlights

The key algorithmic ideas implemented in fastplasmon are:

• Dipole-subspace operator projection
  The K operator is projected onto the basis ${x(s), y(s), z(s)}$, yielding a 3×3 generalized eigenproblem.

• Matrix-free K application
  The K operator is applied directly to the dipole basis using Numba-accelerated kernels, avoiding dense matrix storage.

• Symmetrized reduced operators
  Discrete self-adjointness is enforced at the reduced level, ensuring physically admissible eigenvalues.

• Separation of geometry and material response
  All geometry-dependent quantities are computed once; material dispersion enters only through scalar expressions.

Installation

From PyPI:

pip install fastplasmon

From source:

git clone https://github.com/INESCTEC/fastplasmon.git
cd fastplasmon
pip install -e .

Basic Usage

A minimal workflow:

from fastplasmon import ArbitraryShapeDipolarSolver

solver = ArbitraryShapeDipolarSolver(
    vertices, faces,
    epsilon=eps_metal,
    medium_epsilon=eps_host
)

# Project K operator onto dipole subspace
kappa, R_dip, a_dip, w_dip = solver.projectK_modes()

# Compute polarizability and extinction
Cext = solver.extinction(
    wavelength,
    eps_medium,
    eps_metal
)

Authors

Paulo S. S. dos Santos

paulo.s.santos@inesctec.pt

INESC TEC - Institute of Systems and Computer Engineering, Technology and Science. Porto, Portugal

License

This project is licensed under the AGPLv3 license.