Scattering

kind: "scattering", model lightsmith.Simulation. One or more particles in a homogeneous, non-absorbing host medium, illuminated by a plane wave.

Fields

Field	Meaning
particles	One or more Particle(shape, material, position_nm)
medium_index	Real refractive index of the host (default 1)
illumination	PlaneWave(direction, polarization), default +z with x polarisation. polarization must be orthogonal to direction.
wavelengths	Vacuum wavelengths in nm (a list or a WavelengthRange)
orientation_average	Average the cross sections over all particle orientations
outputs	cross_sections (default on); optional near_field and far_field requests
substrate	A half-space below the particles. Part of the schema, but no built-in backend supports it in 0.1.
options	Per-method solver options

Shapes

kind	Parameters	Notes
sphere	radius_nm
layered_sphere	core_radius_nm, shells: [{thickness_nm, material}]	The particle material is the core
spheroid	semi_axis_xy_nm, semi_axis_z_nm	Symmetry axis along z
cylinder	radius_nm, length_nm	Flat ends, axis along z

Field outputs

• NearFieldRequest: the total |E|² enhancement, and optionally the complex E, sampled on a square plane. Settings:
• plane: xy, xz or yz;
• offset_nm, half_width_nm, points;
• wavelengths_nm.
• FarFieldRequest: dC_sca/dΩ in nm²/sr on a θ × φ grid. θ is measured from the incident direction. φ = 0 is the E-plane and φ = 90° the H-plane.

Result

Result contains:

• wavelengths_nm, c_ext_nm2, c_sca_nm2, c_abs_nm2;
• equivalent_radius_nm;
• optional near_field and far_field lists;
• provenance and warnings.

Result.to_xarray() converts the cross sections to an xarray Dataset (xarray extra).

Solvers

method="auto" tries Mie, then T-matrix, then DDA, then FDTD. The quasistatic solver is never picked automatically.

	Mie	T-matrix	DDA	FDTD	Quasistatic
Spheres	✅	✅	✅	✅	✅
Layered spheres	✅ (native)	✅	✅	✅	✅
Spheroids, cylinders	—	—	✅	✅	spheroids
Several particles	—	✅ (spheres)	✅	✅	—
Near field	native	✅	✅	✅	—
Far field	native	✅	✅	—	✅
Orientation average	✅	✅ (analytic)	✅ (sampled)	—	✅
GPU	native	—	✅	✅	—

Example

import lightsmith as ls

sim = ls.Simulation(
    particles=[ls.Particle(shape=ls.Spheroid(semi_axis_xy_nm=10, semi_axis_z_nm=30),
                           material=ls.LibraryMaterial(name="Au"))],
    medium_index=1.33,
    illumination=ls.PlaneWave(direction=(1, 0, 0), polarization=(0, 0, 1)),  # E along the long axis
    wavelengths=ls.WavelengthRange(start_nm=450, stop_nm=900, num=91),
)
ls.estimate(sim, method="dda")
res = ls.solve(sim, method="dda")