Effect of Dielectric Photoemission on Surface Breakdown: An LDRD Report

Abstract

The research discussed in this report was conceived during our earlier attempts to simulate breakdown across a dielectric surface using a Monte Carlo approach. While cataloguing the various ways that a dielectric surface could affect the breakdown process, we found that one obvious effect -photoemission from the surface -had been ignored. Initially, we felt that inclusion of this effect could have a major impact on how an ionization front propagates across a surface because of the following argument chain: (1) The photon energy required to release electrons from a surface via photoemission is less than the photon energy required to ionize gas molecules directly. (2) The mean free path of a photon in gas is longer for low-energy photons than for high-energy photons. (3) Photoionization is a major effect in advancing the ionization front for breakdown in gas without a surface, therefore, we know that even high-energy photons can be released from the head of a streamer and propagate some distance through the gas. Our hypothesis, therefore, was that photons with energies near the threshold of photoemission could travel further in front of the streamer before being absorbed than higher-energy photons needed for photoionization, yet the lower-energy photons, with the help of the surface, could still create seed electrons for new avalanches. Thus, the streamer would advance more rapidly next to a surface than in gas alone. Additionally, the photoemission from the surface would add to the electrons in the avalanche and cause the avalanche to grow faster. After some study, however, we are forced to conclude that although photoemission does contribute to avalanche growth at fields near breakdown threshold, secondary electron emission causes electrons to stick to the surface and cancels out the growth due to photoemission. This conclusion assumes a discharge that occurs over a short period of time so that charging of the surface, which could alter its secondary electron emission characteristics, does not occur. This report documents the numerical work we did on investigating this effect and the experimental work we did on pre-breakdown phenomena in gas.

Ionization coefficient approach to modeling breakdown in nonuniform geometries.

“…At the field levels studied here, the secondary electron emission phenomenon acts as a sink to the electrons in the avalancheretarding the electron growth. More details on the code BREAKDOWN can be found in [22], [23].…”

“…Table 10 gives parameters to be used in the canonical curve for three dielectric materials. Figure 30 shows the normalized curves for Teflon Material Table 9.Excitation energy levels and spectral bands [23]. and polyethylene.…”

The influence of tube diameter and gas gap on breakdown characteristic of capillary discharge

“…At atmospheric-pressure, non-equilibrium plasmas are susceptible to instabilities and, in particular, to arcing (glowto-arc transition) [13][14][15][16]. Once the transition occurs, the gap voltage drops sharply and the discharge current rises rapidly to dozens of ampere.…”

“…According to the relative researches, the diameter of capillary (D), the gas gap (d), the length-to-diameter ratios (d/D), and the feed gas were critical parameters for the microdischarge characteristic; secondary electron emission was the key factor in discharge process [12][13][14][15]17]. As a simple reference, in [13][14], the flashover across a dielectric surface tended to follow the surface in air but follow the electric field in nitrogen. In [15][16] the pin-pin electrode geometry was placed over a dielectric flat in argon.…”

Physics of dielectric surface flashover at atmospheric pressure

“…Obviously, this energy is still below the fvst cross-over point for secondary electron emission, SEE, due to impacting electrons (typically several IOeV for dielectrics) so that the electrons that impact the surface will be trapped in the bulk of the dielectric [3]. Hence, the surface will be negatively charged and the head of any electron avalanche that-is developing across the gap will be repelled from the surface, which we believe is true for N2 or AI.…”