A Numerical Simulation Code for Electronic and Ionic Transients From a Time-Resolved Pulsed Townsend Experiment

Abstract

A numerical code which calculating the spatiotemporal development of the discharge current in the Townsend regime has been developed. This code was written in Fortran 77. The individual contribution of electrons, positive, and negative ions to the total measurable current in the pulsed discharge is calculated by solving the corresponding set of coupled continuity equations for each species, for which boundary and initial conditions are set adequately. This simulation code is also capable of considering a more realistic scheme of photoelectron emission, such as that produced using a pulsed ultraviolet laser. In addition, the processes of ion and electron drift and longitudinal diffusion, electron detachment, negative-ion stabilization, and ion conversion have been calculated successfully. Some of these effects are extremely difficult to model analytically, and the simultaneous inclusion of two or more of them proves practically impossible. The power of this method in simulating complex discharge situations is discussed in this paper by means of examples considering the processes of electron and ion drift, diffusion, electron impact ionization and attachment, electron detachment, and negative-ion stabilization.Index Terms-Attachment, continuity equations, diffusion, drift velocity and diffusion, electron and ion swarms, electron detachment, finite difference methods, ionization, ionization and attachment, low temperature plasmas, negative ions, numerical simulation codes, pulsed Townsend experiment.

Measurement of ionization, attachment, detachment and charge transfer rate coefficients in dry air around the critical electric field

“…We further present a rate-fitting method for pulsed Townsend experiments, based on simulation on graphics cards. A comparable approach has been published by Urquijo et al [16], and Bekstein et al [17], on CPUs. The publication is structured as follows: The experimental setup is presented in section 2.…”

Foundations and interpretations of the pulsed-Townsend experiment

“…Using thatW to then determine any further parameters (e.g.α T orD L ) will further propagate uncertainties in the other derived coefficients/parameters. Non-linear curve fitting to the full equation (24) should in fact be performed (as in, for example, reference [81]) in all cases.…”

Electron swarm coefficients and critical field strength of the gaseous ternary mixtures CF3I-SF6-N2 and CF3I-SF6-CO2

“…On the other hand, for a given amount of N 2 in the mixture, the values of W are slightly higher for correspondingly higher concentrations of SF 6 than CF 3 I, which is indicative of the stronger interaction of the electrons with the latter gas. In other words, the elastic momentum transfer cross section of N 2 is smaller than either those of SF 6 or CF 3 I, both in the range of 10 -15 cm 2 at low electron energies, while that of N 2 is of the order of 10 -16 cm 2 at low collision energies [35,36]. The velocity values of W SF6 are in turn higher than those for CF 3 I for E/N < 500 Td, albeit for different reasons related to the strong dipole moment of CF 3 I whereas that of SF 6 is zero [24].…”

“…to measure accurately since its effects are mostly shown on the falling portion of the trailing edge of the avalanche pulse [36], which is convoluted with the approximately Gaussian photoelectron pulse shape with 3-5 ns duration FWHM. The ND L values presented in figures 4 and 5 are those measured at the lowest possible gas pressures, since the trailing edge of the pulse lasts much longer than that due to the finite width of the photoelectron pulse, thereby rendering more reliable ND L values, in addition to the effects of the amplifier bandwidth (8.75 ns risetime).…”