Plasma sheath
A plasma sheath between two grounded (Dirichlet) walls.
Runs a two-species (electron/ion) particle simulation confined in the strip 0 < x < L between two grounded conducting plates, using GroundedStripField: its image-charge Green’s function enforces phi = 0 exactly at x = 0 and x = L (a Dirichlet boundary condition), not approximately. A quasi-neutral plasma starts uniform and Maxwellian; electrons, being lighter and faster, are lost to and repelled from the grounded walls more than the slower ions, leaving a thin layer of excess positive space charge against each wall — the classic plasma sheath — while the bulk stays quasi-neutral. Saves one figure per claim:
Source: examples/plasma_sheath.py. Run it yourself with python examples/plasma_sheath.py [--output examples/output] [--quick] (needs pip install "tree-d-3d[examples]").
Animations
Section titled “Animations”
Loading figure…an (x, v_x) phase-space animation of both species, from the initial uniform state to the developed sheath, with the particle counts in the top right corner.
Figures
Section titled “Figures”
Loading figure…the initial electron and ion positions, uniform and quasi-neutral, with the two grounded walls marked.

Loading figure…electron and ion density against x, time-averaged over a late window once the sheath has formed; quasi-neutral in the bulk, with an excess of ions over electrons in thin layers against both walls.

Loading figure…the electrostatic potential against x, same time average; it returns to (near) zero at both grounded plates — the Dirichlet condition the image charges enforce exactly — while the bulk floats higher, with most of the drop concentrated in the sheath layers.

Loading figure…the field E_x against x, same time average; close to zero in the quasi-neutral bulk and pointing from the bulk toward each wall inside the sheath layers.

Loading figure…electrons and ions remaining against time; the walls are a real particle sink, with the bulk only partly depleted over the plotted window.
--quick uses fewer particles and a shorter run (well under a minute instead of roughly ten). Needs Plotly (pip install 'tree-d-3d[examples]').