Summary: An atmospheric‐pressure glow discharge has been generated using an inductive energy storage pulsed power generator. A pulsed high voltage with a short rise time of under 30 ns is employed to generate streamer discharges simultaneously at all tips of a needle‐array electrode in nitrogen. The large number of streamer discharges prevent glow‐to‐arc transitions caused by inhomogeneous thermalization. SOS diodes are used as an opening switch to shorten the rise time. Circuit parameters such as the capacitance of a primary energy storage capacitor and/or the inductance of a secondary energy storage inductor are determined by experimental results using dummy resistive loads. The results show that circuit parameters adequate for producing atmospheric glow discharges are different from those for producing corona discharges. The energy transfer efficiency of the pulsed power generator has a maximum value of 75% at 193 Ω dummy load resistance. Voltage‐current characteristics of the glow discharge show two regions: constant voltage (normal glow) and an increasing voltage with discharge current (abnormal glow). The glow current is drastically decreased by eliminating the SOS diodes, in which case the charging voltage is directly applied to the electrode. Spatial‐ and time‐averaged electron densities in a positive column are estimated from calculations based on nitrogen swarm data. The electron density is estimated to be 1.8 × 1011 cm−3, which is much larger than 9.7 × 109 cm−3 in the case without the SOS diodes.

An atmospheric pressure glow discharge was generated using a needle-array electrode in nitrogen. A pulsed high voltage with short rise time under 10 ns was employed to generate streamer discharges simultaneously at all needle tips. The large number of streamer discharges prevented glow-to-arc transition caused by inhomogeneous thermalization. Semiconductor opening switch (SOS) diodes were employed as an opening switch to shorten the rise time. The glow current was drastically decreased by eliminating the SOS diode, in which case the charging voltage was directly applied to the electrode. Spatial- and time-averaged electron densities in a positive column were estimated from calculations based on nitrogen swarm data. The density was estimated to be 1.8×1011 cm-3, which was much larger than 9.7×109 cm-3 in the case without the SOS diode.

No abstract