In this paper, we report on results from a computational investigation of nanosecond pulsed discharges in air of negative polarity using a two-dimensional fluid and fluid-Monte Carlo simulations. The computational parameters and conditions are taken from an experiment. The discharge is initiated near the cathode having a small radius of curvature and propagates towards the flat anode. The simulations are done in atmospheric pressure air with a nanosecond pulse of 120 kV amplitude in a 12 mm gap with a sharp cathode. The essential difference between discharges initiated by low-and high-energy beams of electrons originating in the vicinity of the cathode was observed. The beam of electrons with the initial energy of 20 keV was shown to result in a diffuse discharge, while the 20 eV beam had a negligible influence on the discharge evolution. By analyzing the experimental and computational results, we conclude that a diffuse discharge is formed due to fast electrons without any assumptions on the critical role of the runaway electrons. More study is required to explore the intermediate range of energies and investigate the transition to the diffuse mode of the discharge.

In this work, the propagation of fast ionisation wave (FIW) discharges driven by negative nanosecond high-voltage (HV) pulses and the x-ray spectrum from the FIW discharges are investigated. The measurement using a capacitive probe shows that the peak axial electric field (E z peak ) at each position rises with the increase of the applied voltage amplitude (U max ), resulting in an increase of the FIW velocity (V FIW ). The influence of the pre-ionisation on the FIW propagation is estimated from the relationship between V FIW and E . z peak The pre-ionisation effect appears to be enhanced with the increase of U max and it is suspected to be due to the presence of more high energy electrons. This is supported by the measured x-ray spectrum, showing a higher count rate and a more elevated high energy tail with a larger U max . The spatially resolved measurement of the x-ray spectrum shows that, with the increase of the distance away from the HV electrode (as the cathode), the lower energy part (< ∼15 keV) of the x-ray spectrum keeps decaying, while its high energy tail (> ∼25 keV) rises at first and then decays. Based on the spatially resolved x-ray spectrum, it is inferred that, as high energy electrons move away from the HV electrode, the peak of their energy distribution is shifted toward a higher value, while their total number decreases. Further analysis implies that the electrons emitted from the cathode obtain a large percent of the applied potential in the vicinity of the cathode and become runaway electrons. The electrons with a high initial energy travel freely in the discharge region, due to a decreasing electron collisional cross section with its energy.

This paper presents an experimental investigation into the runaway electron spectrum with a gas diode composed of a rough spherical cathode and plane anode under the excitation of a nanosecond-pulse generator in atmospheric air. The runaway electron beams are measured by a collector covered with aluminum foil with a thickness from 0 μm (mesh grid) to 50 μm. The energy spectrum is calculated by an improved Tikhonov regularization called the maximum entropy method. The experimental results show that the transition state of the discharge consisted of multiple streamer channels stretched from the cathode with glow-like plasma uniformly distributed over the anode. The number of runaway electrons measured by the collector is in the order of 1010 in atmospheric pressure air with a gap spacing of 5 mm and applied voltages of 70–130 kV. The cathode with a rough surface creates a more inhomogeneous electric field and larger emission site for the runaway electrons around the cathode, providing conditions for the coexistence of filamentary streamer and diffuse discharge. The reconstructed spectra show that the energy distribution of the runaway electrons presents a single-peak profile with energies from eU m/2–2eU m/3 (U m is maximal voltage across the gap).