Quickstart

Define the cross sections, the neutral gas and the processes, build a MonteCarloCollisions operator, and collide markers with it:

import plasmacoll
from plasmacoll.constants import ATOMIC_MASS, ELECTRON_MASS, ev_to_kelvin

argon_mass = 39.95 * ATOMIC_MASS
gas = plasmacoll.NeutralBackground(
    "Ar", density=1e21, temperature=300.0, mass=argon_mass
)

elastic = plasmacoll.CrossSection.constant(1e-19)  # m^2
ionization = plasmacoll.CrossSection.from_table("ionization.dat", threshold=15.76)

mcc = plasmacoll.MonteCarloCollisions(
    species_masses={"e": ELECTRON_MASS, "Ar+": argon_mass},
    backgrounds=[gas],
    processes={
        "e": [
            plasmacoll.CollisionProcess("elastic", "Ar", elastic, energy_frame="lab"),
            plasmacoll.CollisionProcess(
                "ionization",
                "Ar",
                ionization,
                energy_frame="lab",
                products={"electron": "e", "ion": "Ar+"},
            ),
        ]
    },
    seed=1,
)

Collide arrays

collide takes the arrays of one species, positions (N, ndim), velocities (N, 3) in m/s and weights (N,), and a time step in seconds:

result = mcc.collide("e", positions, velocities, weights, dt=1e-11)
# the velocities after the step
result.velocities
# markers consumed by attachment, detachment and charge transfer
result.removed
# new markers: a list of NewMarkers(positions, velocities, weights)
result.created["Ar+"]
# real collisions per process
result.diagnostics.counts

Collide marker sets

collide_species updates every species in place. It removes consumed markers and adds the products once all species have collided:

electrons = plasmacoll.ParticleArrays.maxwellian(
    100_000, ELECTRON_MASS, ev_to_kelvin(5.0), seed=2
)
ions = plasmacoll.ParticleArrays.empty()
diagnostics = mcc.collide_species({"e": electrons, "Ar+": ions}, dt=1e-11)

A 0D reactor

ZeroDReactor repeats that step and records densities, mean energies, temperatures and collision counts as arrays over time:

reactor = plasmacoll.ZeroDReactor(mcc, {"e": electrons}, volume=1e-6)
history = reactor.run(dt=1e-11, num_steps=1000, every=10)
history.density["Ar+"]  # m^-3, one value per recorded step

With electric_field=(Ex, Ey, Ez) (V/m) and charges={"e": -1, "Ar+": 1} the reactor is a swarm experiment: every step accelerates the charged species before colliding them, and history.mean_velocity holds the drift velocities. max_markers=N merges a species back to $N/2$ markers whenever ionization grows it beyond $N$.

Next: the null-collision method, or the tutorials.