Negative-ion sources

Protons in hydrogen

plasmacoll.proton_hydrogen_processes(cross_sections, h2_ion=None) builds the six processes that dominate proton transport in the driver and expansion region of an RF-driven negative hydrogen ion source (the test-particle model of Wünderlich et al.):

Key	Reaction	Kind	Energy frame
mt_h	H⁺ + H → H⁺ + H	elastic	centre of mass
mt_h2	H⁺ + H₂ → H⁺ + H₂	elastic	lab
cx_h	H⁺ + H → H + H⁺	backscatter	lab
cx_h2	H⁺ + H₂ → H + H₂⁺	charge_transfer	lab
rot_h2	H⁺ + H₂(J=0) → H⁺ + H₂(J=2)	excitation	lab
vib_h2	H⁺ + H₂(v=0) → H⁺ + H₂(v>0)	excitation	lab

The function supplies the process definitions, not the cross sections. The model’s sources are Krstić and Schultz for H⁺ + H₂ and Janev et al. for H⁺ + H (see also AMJUEL/HYDHEL). The rotational (0.0453 eV) and vibrational (0.516 eV) thresholds replace those of the tables. With h2_ion="H2+" the charge exchange with H₂ creates H₂⁺ ions. Otherwise it only removes the proton.

Vibrationally excited H₂

H⁻ is produced in the volume by dissociative attachment, $e + \mathrm{H_2}(v) \to \mathrm{H^-} + \mathrm{H}$. The cross section rises by orders of magnitude from $v=0$ to the highest bound levels (Bardsley and Wadehra 1979; Wadehra 1984). A model that uses one ground-state cross section for all of H₂ gets the production rate wrong by orders of magnitude.

plasmacoll.vibrational supplies a prescribed population: the H₂(X) vibrational ladder of Fantz and Wünderlich (At. Data Nucl. Data Tables 92, 853 (2006)) with a Boltzmann distribution at one vibrational temperature.

from plasmacoll.vibrational import VibrationalDistribution

distribution = VibrationalDistribution.h2_ground_state_from_kelvin(5000.0)
distribution.populations  # n_v / n for v = 0..14
distribution.fraction_at_least(4)  # [H2(v >= 4)] / [H2]

attachment = plasmacoll.CrossSection.vibrational_mixture(
    level_tables, distribution.populations
)

A real population deviates from a single Boltzmann distribution at high $v$, so this underestimates the tail that drives attachment. It is a first approximation, not a solution of the vibrational kinetics. estimate_electronegativity turns the excited fraction into an order-of-magnitude $[\mathrm{H^-}]/[e]$ from the balance of attachment against detachment by H atoms (Kimura and Kasugai 2010, as given by Krištof et al. 2019).

plasmacoll does not ship level-resolved attachment cross sections. Read them from tables with CrossSection.from_table. Tutorial 5 runs the pipeline with synthetic levels.