Fields & Visualization¶

Where numbers become pictures. Real-space field reconstruction (ikarus.core.fields) and the matplotlib plotting helpers (ikarus.visualization).

Optional dependency

Plotting needs matplotlib (pip install "ikarus-rcwa[viz]"); field reconstruction itself runs on NumPy/SciPy alone.

Field reconstruction¶

`FieldMap`¶

A dataclass holding real-space vector fields on a sampling grid.

Attribute	Description
`E`, `H`	Complex arrays, shape `(..., 3)` → `(Ex, Ey, Ez)` / `(Hx, Hy, Hz)`.
`coords`	Dict mapping axis name (`"x"`, `"y"`, `"z"`) → 1-D coordinate array (m).
`z`	The depth (m) for an `xy` slice, else `None`.
`eps`	Real permittivity on the same grid (attached by `RCWA.get_fields`; used for structure overlays).
`intensity` (property)	(

`RCWA.get_fields(...)`¶

The high-level entry point — full signature at RCWA.get_fields. Returns a dict of FieldMaps keyed by depth label (xy) or plane name (xz/yz).

# xy slices at three depths:
maps = rcwa.get_fields(z_positions=[0, 100e-9, 200e-9], plane="xy", nx=128, ny=128)
for label, fm in maps.items():
    print(label, fm.intensity.max())

# an xz cross-section through the cell center:
xz = rcwa.get_fields(plane="xz", nx=200, y_position=rcwa.period_y / 2)["xz"]

The z convention

z = 0 is the cover / first-interior-layer interface; z grows into the stack. Negative z = inside the cover; beyond the total interior thickness = inside the substrate.

`reconstruct(...)` (low-level)¶

from ikarus.core.fields import reconstruct
reconstruct(sol, z_positions, nx=64, ny=64, plane="xy",
            x_position=0.0, y_position=0.0) -> dict

Operates directly on a FieldSolution (result.solution). RCWA.get_fields wraps this and attaches the structure permittivity for overlays.

Plotting¶

from ikarus.visualization import plot_field, plot_stack, plot_topology, plot_field_xy

`plot_field(field_map, component="intensity", ax=None, savefig=None, cmap=None, overlay=True, overlay_color="white", overlay_alpha=0.45)`¶

Plot a 2-D FieldMap cross-section. component is "intensity", a component ("Ex"…"Hz") for magnitude, or "<comp>phase" (e.g. "Eyphase") for phase. With overlay=True and eps present, material boundaries are drawn as semi-transparent contours — your built-in sanity check that the field lives where the structure is.

ax = plot_field(xz, component="intensity")
ax.figure.savefig("field.png", dpi=150)

plot_field(xz, component="Eyphase", cmap="twilight")  # a phase map

`plot_field_xy(field_dict, component="intensity", savefig=None)`¶

Several xy maps (the dict from get_fields) side by side.

`plot_stack(rcwa, ax=None, savefig=None, finite_frac=0.25)`¶

The layer stack as an xz cross-section, color-coded by material. Semi-infinite cover/substrate get a finite visual height (finite_frac of the interior).

`plot_topology(rcwa, layer_index, wavelength=None, ax=None, savefig=None)`¶

A patterned layer's permittivity (real part) over the unit cell.

Via the façade¶

rcwa.visualize_structure(plane="xz")                       # the stack
rcwa.visualize_structure(plane="xy", layer_index=1,        # a topology
                         savefig="topology.png")
rcwa.visualize_fields()                                    # auto xz field map

Full example¶

import numpy as np
from ikarus import RCWA, shapes
from ikarus.visualization import plot_field

period = 700e-9
cross = shapes.cross(arm_length=0.75, arm_width=0.25, grid_shape=(96, 96))

rcwa = RCWA(period_x=period, period_y=period, resolution=(96, 96), n_orders=(8, 8))
rcwa.add_uniform_layer(np.inf, "Air")
rcwa.add_layer(160e-9, cross, ["Air", "TiO2"])
rcwa.add_uniform_layer(np.inf, "SiO2")
rcwa.set_source(wavelength=600e-9, theta=0, polarization="linear")
rcwa.simulate()

xz = rcwa.get_fields(plane="xz", nx=120, y_position=period / 2)["xz"]
ax = plot_field(xz, component="intensity")     # |E|² with structure outline
ax.figure.savefig("cross_field_xz.png", dpi=150, bbox_inches="tight")

Best practices¶

Run a solve before get_fields (it will solve implicitly otherwise, with whatever settings are current).
Reconstruct on a finer nx/ny than the solver resolution for publication figures — the reconstruction grid is independent and free.
Leave overlay=True on: a field that ignores its structure outline is a bug telling you about itself.

Fields & Visualization¶

Field reconstruction¶

FieldMap¶

RCWA.get_fields(...)¶

reconstruct(...) (low-level)¶