Observation of nonthermal turbulent electric fields in a nanosecond plasma opening switch experiment

Abstract

Nonthermal turbulent electric fields due to plasma instabilities were studied in a 100-ns duration plasma opening switch using observations of hydrogen line spectral profiles. The H ␣ and H ␤ widths were seen to rise by 2-3 times during the current pulse, shown to result from the presence of nonthermal electric fields in the plasma. The spectral profiles are analyzed using two recently developed methods based on short and intermediate time behaviors of the line profile Fourier transforms. One method gives the mean amplitude of the nonthermal fields with no dependence on their frequencies. The second method uses calculations of the autocorrelation functions for various field amplitudes and frequencies to yield bounds on these two parameters. The field amplitude is determined to be 14.5Ϯ2.5 kV/cm, and the fluctuation frequency is found to be of the order of the electron plasma frequency. Based on their high frequency, the oscillations probably result from Langmuir waves, driven by the voltage drop on the plasma opening switch ͑POS͒. The waves have no significant effect on the POS operation, since they do not give rise to anomalous resistivity, and therefore have no effect on the magnetic-field evolution. We obtain an upper limit for the amplitude of possible low-frequency fields ͑ion-acoustic waves͒, that may give rise to anomalous resistivity, and estimate the resulting diffusion velocity and current channel width. Both quantities are found to be much lower than the values observed in the experiment, and the low-frequency field amplitude is much lower than the saturation limit predicted by previous theoretical treatments. This implies that in our experiment possible low-frequency waves have little influence on the magnetic-field distribution.

Observation of faster-than-diffusion magnetic field penetration into a plasma

“…The instability that can result in the highest collision frequency, the ion-acoustic instability, has been investigated 11,38,39 in different configurations. Let us estimate the anomalous plasma collision frequency using the measured nonthermal electric field amplitude ͑as described in Sec.…”

“…2,11,38,39 Here, in order to estimate the anomalous plasma collisionality we measured the amplitude and obtained bounds on the frequency of the turbulent electric fields using Stark broadening of hydrogen and helium lines. For these instabilities the amplitude of the turbulent electric fields is found to result in a collisionality that is too small to explain the magnetic field penetration into the carbon plasma by diffusion.…”

Plasma dynamics in pulsed strong magnetic fields

“…Spectroscopy combined with novel plasma doping techniques allowed for time-dependent, threedimensional ͑3D͒ spatially resolved measurements of the plasma composition, 17,18 magnetic field evolution, [19][20][21] ion dynamics, [22][23][24] electron energy distribution, and nonthermal electric fields. 21,22,25,26 In particular, the use of Zeeman splitting of emission lines from the plasma constituents is highly advantageous, since it yields unambiguously the magnetic field inside the plasma.…”

Plasma and Magnetic-Field Dynamics in Conduction Phase of Opening Switch

“…For the linearly rising magnetic field, i.e., B(r c , t) =Ḃ 0 t = 2İt/cr c , these conditions are reduced to t c (m i /Zm) 1/2 ((1/2)((mc 3 /e)/İ)(r c /c)) 1/2 and n (1/4πr 2 c )(m i c 2 /Ze 2 ), where m and e are the electron mass and charge, m i and Z are the ion mass and charge, respectively, c is the velocity of light, r c is the cathode radius, n is the plasma concentration, t c is the switch conduction time, andİ is the Manuscript [7]. In addition, the current-channel width according to EMG theory is two orders of magnitude smaller than that observed in experiments [8]. In recent papers, it is also pointed out that magnetic-field evolution cannot be explained in the EMG framework [9], and after finite time, plasma pushing is dominant [10].…”