Runaway electron beams in the gas discharge for UV nitrogen laser excitation

The review of the methods for obtaining the runaway electron beams in the gas discharge is performed. The new method is offered, using which the beam is first formed in a narrow gap (∼1 mm) between the cathode and the grid and then it is accelerated by the field of the plasma column of the anomalous self-sustained discharge in the main gap (10–20 mm long). The electron beams with an energy of about 10 keV and current density of 10³ A/cm² at a molecular nitrogen pressure of up to 100 Torr have been obtained experimentally. The results of research of the UV nitrogen laser with an excitation via runaway electron beam and radiation of energy of ∼1 mJ are given. The UV nitrogen laser generation with the energy of ∼1 mJ has been obtained by the runaway electron beams.     

Parameters of a supershort avalanche electron beam generated in atmospheric-pressure air

Conditions under which the number of runaway electrons in atmospheric-pressure air reaches ∼5 × 10¹⁰ are determined. Recommendations for creating runaway electron accelerators are given. Methods for measuring the parameters of a supershort avalanche electron beam and X-ray pulses from gas-filled diodes, as well as the discharge current and gap voltage, are described. A technique for determining the instant of runaway electron generation with respect to the voltage pulse is proposed. It is shown that the reduction in the gap voltage and the decrease in the beam current coincide in time. The mechanism of intense electron beam generation in gas-filled diodes is analyzed. It is confirmed experimentally that, in optimal regimes, the number of electrons generated in atmospheric-pressure air with energies T > eU _m , where U _m is the maximum gap voltage, is relatively small.     

Pulse-periodic generation of supershort avalanche electron beams and X-ray emission

Pulse-periodic generation of supershort avalanche electron beams (SAEBs) and X-ray emission in nitrogen, as well as the transition from a single-pulse mode to a pulse-periodic mode with a high repetition frequency, was studied experimentally. It is shown that, in the pulse-periodic mode, the full width at halfmaximum of the SAEB is larger and the decrease rate of the gap voltage is lower than those in the single-pulse mode. It is found that, when the front duration of the voltage pulse at a nitrogen pressure of 90 Torr decreases from 2.5 to 0.3 ns, the X-ray exposure dose in the pulse-periodic mode increases by more than one order of magnitude and the number of SAEB electrons also increases. It is shown that, in the pulse-periodic mode of a diffuse discharge, gas heating in the discharge gap results in a severalfold increase in the SAEB amplitude (the number of electrons in the beam). At a generator voltage of 25 kV, nitrogen pressure of 90 Torr, and pulse repetition frequency of 3.5 kHz, a runaway electron beam was detected behind the anode foil.     

Study of the Formation Time of a Self-Sustained Subnanosecond Discharge at High and Ultrahigh Gas Pressures

The formation times of self-sustained subnanosecond discharges in nitrogen at pressures of 1?40 atm and in hydrogen at pressures of 1–60 atm are analyzed in terms of the avalanche model. In experiments, a subnanosecond voltage pulse with an amplitude of 102 ± 2 kV was applied to a 0.5-mm-long discharge gap with a uniformly distributed electric field (the curvature radii of both the cathode and anode ends were 1 cm). The rise time of the voltage pulse from 0.1 to 0.9 of its amplitude value was about 250 ps. Breakdown occurred at the leading edge of the pulse. The discharge formation time was measured at different gas pressures with a step of 5–10 atm. Analysis of the experimental results shows that, in nitrogen at pressures of 10–40 atm and in hydrogen at pressures of 20–50 atm, breakdown occurs earlier than the electron avalanche reaches its critical length and that the critical avalanche length lies in the range of (2–8) × 10^–2 mm, which is one order of magnitude shorter than the discharge gap length. This means that the avalanche–streamer model is inapplicable in this case. The fast formation of a conducting channel under these conditions can be explained by ionization of gas by runaway electrons. In this case, the conducting column develops as a result of simultaneous development of a large number of electron avalanches in the gas volume. An increase in the hydrogen pressure from 50 to 60 atm leads to an abrupt increase in the discharge formation time by about 50%. As a result, the growth time of the electron avalanche to its critical length becomes shorter than the discharge formation time. In this case, the electrons cease to pass into the runaway regime and the discharge is initiated from the cathode due to field emission from microinhomogeneities on its surface. Under these conditions, the discharge formation time is well described by the avalanche–streamer model.     

Picosecond runaway electron beams in air

Experimental data on the generation of picosecond runaway electron beams in an air gap with an inhomogeneous electric field at a cathode voltage of up to 500 kV are presented. The methods and equipment developed for these experiments made it possible to measure the beam characteristics with a time resolution of better than 10⁻¹¹ s, determine the voltage range and the beam formation time in the breakdown delay stage, and demonstrate the influence of the state of the cathode surface on the stability of runaway electron generation. It is demonstrated that the critical electron runaway field in air agrees with the classical concepts and that the accelerated beam can be compressed to ∼20 ps. It is unlikely that, under these conditions, the beam duration is limited due to the transition of field emission from the cathode to a microexplosion of inhomogeneities. The maximum energy acquired by runaway electrons in the course of acceleration does not exceed the value corresponding to the electrode voltage.     

Amplitude−temporal characteristics of a supershort avalanche electron beam generated during subnanosecond breakdown in air and nitrogen

The amplitude?temporal characteristics of a supershort avalanche electron beam (SAEB) with an amplitude of up to 100 A, as well as of the breakdown voltage and discharge current, are studied experimentally with a picosecond time resolution. The waveforms of discharge and SAEB currents are synchronized with those of the voltage pulses. It is shown that the amplitude?temporal characteristics of the SAEB depend on the gap length and the designs of the gas diode and cathode. The mechanism for the generation of runaway electron beams in atmospheric-pressure gases is analyzed on the basis of the obtained experimental data.     

Efficient algorithm for simulating a multidimensional glow discharge

A normal glow discharge in nitrogen between two infinite plane-parallel electrodes is simulated numerically in three dimensions in the drift-diffusion model. An algorithm is proposed that is based on solving the model equations by a hybrid of the sweep and relaxation methods and that combines a satisfactorily fast rate of convergence with the efficiency of individual iterations. The algorithm developed was used to carry out a three-dimensional simulation of a discharge in nitrogen at the pressures p = 5 and 1 Torr, the source voltage and ballast resistance being 2 kV and 300 kΩ, respectively. It was found that, at the pressure p = 1 Torr, the current density and charged particle densities change to a torioidal distribution in the anode region.     

Hydrodynamic model of a dielectric-barrier discharge in pure chlorine

A one-dimensional hydrodynamic model of a dielectric-barrier discharge (DBD) in pure chlorine is developed, and the properties of the discharge are modeled. The discharge is excited in an 8-mm-long discharge gap between 2-mm-thick dielectric quartz layers covering metal electrodes. The DBD spatiotemporal characteristics at gas pressures of 15–100 Torr are modeled for the case in which a 100-kHz harmonic voltage with an amplitude of 8 kV is applied to the electrodes. The average power density deposited in the discharge over one voltage period is 2.5–5.8 W/cm³. It is shown that ions and electrons absorb about 95 and 5% of the discharge power, respectively. In this case, from 67 to 97% of the power absorbed by electrons is spent on the dissociation and ionization of Cl₂ molecules. Two phases can be distinguished in the discharge dynamics: the active (multispike) phase, which follows the breakdown of the discharge gap, and the passive phase. The active phase is characterized by the presence of multiple current spikes, a relatively high current, small surface charge density on the dielectrics, and large voltage drop across the discharge gap. The passive phase (with no current spikes) is characterized by a low current, large surface charge density on the dielectrics, and small voltage drop across the discharge gap. The peak current density in the spikes at all pressures is about 4 mA/cm². In the multispike phase, there are distinct space charge sheaths with thicknesses of 1.5–1.8 mm and a mean electron energy of 4.3–7 eV and the central region of quasineutral plasma with a weak electric field and a mean electron energy of 0.8–3 eV. The degree of ionization of chlorine molecules in the discharge is ~0.02% at a pressure of 15 Torr and ~0.01% at 100 Torr. The DBD plasma is electronegative due to the fast attachment of electrons to chlorine atoms: e + Cl₂ → Cl + Cl^–. The most abundant charged particles are Cl ₂ ⁺ and Cl^? ions, and the degree of ionization during current spikes in the active phase is (4.1–5.5) × 10^–7. The mechanism of discharge sustainment is analyzed. The appearance of a series of current spikes in the active phase of the discharge is explained.     

Memory effect in the ignition of a low-pressure glow discharge in nitrogen in a long discharge tube

Results are presented from experimental studies of breakdown and the initial stage of a discharge in a long tube (with an interelectrode distance of 40 cm and a diameter of 2.8 cm) at a pressure of ～1 Torr and pulse discharge current of ～10 mA. Breakdown was produced by positive voltage pulses with a linearly growing leading edge with a steepness of dU/dt ～ 10⁶–10⁸ V/s. The time interval between pulses was varied from τ = 0.5 ms to 1 s, the pulse duration being 10 ms. The work was aimed at studying the memory effect of the discharge gap, namely, the influence of the previous pulse on the breakdown characteristics of the next one. In the experiments, the breakdown voltage was measured at different values of dU/dt and τ. It was found that the memory effect was absent at τ ～ 1 s. At the same time, an increase in the breakdown voltage with increasing dU/dt was observed. In the range of τ ≈ 50–200 ms, the breakdown voltage also did not depend on τ, but the memory effect took place. The memory effect in this case consisted in that the breakdown voltage decreased with increasing dU/dt, so that, at dU/dt ～ 10⁷ V/s, the breakdown voltage was two times lower than in the case of τ ～ 1 s. For τ ～ 1–10 ms, the memory effect manifested itself in that the breakdown voltage depended on τ: it could either decrease (the “normal” effect) or increase (the “anomalous” effect) with increasing τ. Breakdown of the discharge gap was preceded by the propagation of an ionization wave, except for the case of small τ values in the domain of existence of the anomalous effect. Estimates allowing one to qualitatively explain the experimental results are made.     

Conditions for electron runaway under leader breakdown of long gaps

An original hydrodynamic model in which inelastic collisions in the equations of motion and energy balance play a decisive role is developed and applied to simulate electron avalanches in strong electric fields. The mean energy and drift velocity of electrons, as well as the ionization coefficient and electric field in a wide range of mean electron energies, are determined for helium and xenon. A criterion is derived for the runaway of the average electron in discharges with ionization multiplication. It is shown that runaway can take place at any value of E/p, provided that the momentum mean free path exceeds the gap length. The voltage corresponding to electron runaway is found for helium, xenon, and air as a function of the electric field, the electron mean energy, and the parameter pd. Conditions for the formation of a precursor in electronegative gases are analyzed. It is shown that the presence of a precursor with a high electric conductance is necessary for the formation of a new leader step. The voltage and time ranges corresponding to efficient electron runaway and X-ray generation during leader breakdown in air are determined.     

Ultrahigh charging of dust grains by the beam−plasma method for creating a compact neutron source

Generation of high-voltage high-current electron beams in a low-pressure (P = 0.1–1 Torr) gas discharge is studied experimentally as a function of the discharge voltage and the sort and pressure of the plasma-forming gas. The density of the plasma formed by a high-current electron beam is measured. Experiments on ultrahigh charging of targets exposed to a pulsed electron beam with an energy of up to 25 keV, an electron current density of higher than 1 A/cm², a pulse duration of up to 1 μs, and a repetition rate of up to 1 kHz are described. A numerical model of ultrahigh charging of dust grains exposed to a high-energy electron beam is developed. The formation of high-energy positive ions in the field of negatively charged plane and spherical targets is calculated. The calculations performed for a pulse-periodic mode demonstrate the possibility of achieving neutron yields of higher than 10⁶ s^–1 cm^–2 in the case of a plane target and about 10⁹ s^–1 in the case of 10³ spherical targets, each with a radius of 250 μm.     

Peculiarities of detecting pulses of runaway electrons and X-rays generated by high-voltage nanosecond discharges in open atmosphere

Peculiarities of detecting pulses of high-energy runaway electrons and the accompanying X-rays generated by high-voltage nanosecond discharges at high overvoltages in dense gases are discussed. Methods for overcoming difficulties encountered in such measurements are demonstrated. Different techniques for detecting runaway electrons and X-rays generated by discharges in open atmosphere are described, and typical errors that may be done when interpreting measurement results are considered. Experiments with the use of a small-size generator of nanosecond pulses with the idle-running voltage of 250–270 kV and stored energy of ～0.5 J have been carried out. It is shown that, in measuring runaway electron pulses by using an experimental configuration with a grid anode, a major portion of the recorded signal is attributed to electromagnetic disturbances. It is found that X-rays are mainly generated due to the deceleration of runaway electrons in the anode, rather than in gas. The number of runaway electrons with energies ～300 keV does not exceed 10⁹ per shot, and the X-ray dose is 60–200 μR/shot, depending on the anode material.     

Experimental investigation of X-ray lasing in a capillary argon discharge

Results are presented from experiments on the laser generation of X-ray radiation at the wavelength λ=469 ? (ε=26.4 eV) on the 3p(J=0)−3s(J=1) transition of Ne-like Ar ions. Experiments were carried out on the SIGNAL electrophysical facility with a 3.1-mm-diameter 157-mm-long Al₂O₃ ceramic capillary filled with argon at a pressure of 0.2–1.0 Torr. The discharge current amplitude was I ∼ 25–40 kA, the current rise rate being dI/dt ∼ 10¹² A/s. By a vacuum X-ray diode tuned to detect X-ray photons with energies in the range 10–40 eV, laser pulses with a duration of t ₁ ∼ 1 ns and maximum energy of E _1,max ∼ 1 μJ were recorded. The pulses were generated 35 ns after the discharge current was switched on. The line spectra in the wavelength range of 150–500 ? showed the bright λ=469 ? line. The angular divergence of the generated X-ray laser beam was estimated to be Δϑ ∼ 2 mrad. Original Russian Text ? O.N. Gilev, V.I. Afonin, V.I. Ostashev, V.Yu. Politov, A.M. Gafarov, A.L. Zapysov, A.V. Andriyash, é.P. Magda, L.N. Shamraev, A.A. Safronov, A.V. Komissarov, N.A. Khavronin, N.A. Pkhaĭko, L.V. Antonova, L.N. Shushlebin, 2006, published in Fizika Plazmy, 2006, Vol. 32, No. 2, pp. 160–165.     

Energy distribution of runaway electrons generated by a nanosecond discharge in atmospheric-pressure air

The spectra of an ultrashort avalanche electron beam generated by a nanosecond discharge in atmospheric-pressure air were investigated. The temporal characteristics of the beam current pulses, gap voltage, and discharge current in a gas diode were measured with a time resolution of ～0.1 ns. A simple technique was developed for recovering electron spectra from the curves of beam attenuation by aluminum foils. The effect of the cathode design, electrode gap length, and generator parameters on the electron spectra were studied using seven setups. It is shown that generation of electrons with anomalously high energies requires the use of cathodes with increased curvature radius.     

Ecton processes in the generation of pulsed runaway electron beams in a gas discharge

As was shown earlier for pulsed discharges that occur in electric fields rising with extremely high rates (10¹⁸ V/(cm s)) during the pulse rise time, the electron current in a vacuum discharge is lower than the current of runaway electrons in an atmospheric air discharge in a 1-cm-long gap. In this paper, this is explained by that the field emission current from cathode microprotrusions in a gas discharge is enhanced due to gas ionization. This hastens the initiation of explosive electron emission, which occurs within 10^–11 s at a current density of up to 10¹⁰ A/cm². Thereafter, a first-type cathode spot starts forming. The temperature of the cathode spot decreases due to heat conduction, and the explosive emission current ceases. Thus, the runaway electron current pulse is similar in nature to the ecton phenomenon in a vacuum discharge.     

Influence of a Nitrogen Admixture on the Anomalous Memory Effect in the Breakdown of Low-Pressure Argon in a Long Discharge Tube

The memory effect (the dependence of the dynamic breakdown voltage U _b on the time interval τ between voltage pulses) in pulse-periodic discharges in pure argon and the Ar + 1%N₂ mixture was studied experimentally. The discharge was ignited in a 2.8-cm-diameter tube with an interelectrode distance of 75 cm. The measurements were performed at gas pressures of P = 1, 2, and 5 Torr and discharge currents in a steady stage of the discharge of I = 20 and 56 mA. Breakdown was produced by applying positive-polarity voltage pulses, the time interval between pulses being in the range of τ = 0.5–40 ms. In this range of τ values, a local maximum (the anomalous memory effect) was observed in the dependence U _b (τ). It is shown that addition of nitrogen to argon substantially narrows the range of τ values at which this effect takes place. To analyze the measurement results, the plasma parameters in a steady-state discharge (in both pure argon and the Ar + 1%N₂ mixture) and its afterglow were calculated for the given experimental conditions. Analysis of the experimental data shows that the influence of the nitrogen admixture on the shape of the dependence U _b (τ) is, to a large extent, caused by the change in the decay rate of the argon afterglow plasma in the presence of a nitrogen admixture.     

Breakdown in helium in high-voltage open discharge with subnanosecond current front rise

Investigations of high-voltage open discharge in helium have shown a possibility of generation of current pulses with subnanosecond front rise, due to ultra-fast breakdown development. The open discharge is ignited between two planar cathodes with mesh anode in the middle between them. For gas pressure 6 Torr and 20 kV applied voltage, the rate of current rise reaches 500 A/(cm² ns) for current density 200 A/cm² and more. The time of breakdown development was measured for different helium pressures and a kinetic model of breakdown in open discharge is presented, based on elementary reactions for electrons, ions and fast atoms. The model also includes various cathode emission processes due to cathode bombardment by ions, fast atoms, electrons and photons of resonant radiation with Doppler shift of frequency. It is shown, that the dominating emission processes depend on the evolution of the discharge voltage during the breakdown. In the simulations, two cases of voltage behavior were considered: (i) the voltage is kept constant during the breakdown; (ii) the voltage is reduced with the growth of current. For the first case, the exponentially growing current is maintained due to photoemission by the resonant photons with Doppler-shifted frequency. For the second case, the dominating factor of current growth is the secondary electron emission. In both cases, the subnanosecond rise of discharge current was obtained. Also the effect of gas pressure on breakdown development was considered. It was found that for 20 Torr gas pressure the time of current rise decreases to 0.1 ns, which is in agreement with experimental data.     

Analysis of cylindrical Langmuir probe using experiment and different theories

Cylindrical probe data have been analyzed using different theories in order to determine some plasma parameters (electron temperature and electron and ion densities). Langmuir probe data are obtained in a cylindrical DC glow discharge in the positive column plasma at argon gas pressures varied from 0.5 to 6 Torr and at constant discharge current equal to 10 mA. The electron density has calculated from the electron current at the space potential and from Orbital Motion Limited (OML) collisionless theory. Ion density has obtained from the OML analysis of the ion saturation currents. In addition, the electron temperature has measured by three different methods using probe and electrons currents. The electron temperature T _e , plasma density n _e , and space potential V _s , have been obtained from the measured single cylindrical probe I–V characteristic curves. The radial distribution of the electron temperature and plasma density along the glow discharge are measured and discussed. Using the collisionless theories by Langmuir cylindrical probe and up to several Torr argon gas pressures the differences between the values of electron temperature and electron and ion densities stay within reasonable error limits.     

Specificity of probe measurements in diffuse plasmas of dense gases in strong electric fields

The article is devoted to extending the applicability of the probe diagnostics to the range of higher pressures of the plasma-forming gas by taking into account the effect of the probe shadow on the anode. The probe current–voltage characteristic in the diffuse plasma of a dense gas in a strong electric field was measured, and the influence of the probe potential and probe current on the dimensions of the probe shadow on the anode was studied experimentally. The experiments were carried at different currents of a steady-state glow discharge and different velocities of the gas flow through the discharge. The plasma-forming gas was nitrogen at a pressure of P = 100 Torr.     

Effect of the Air Flow Velocity on the Characteristics of a Pulsating Discharge Produced by a DC Power Source

Results are presented from experimental studies of a pulsating discharge produced by a dc power source in subsonic and supersonic cold (T = 150–300 K) air flows at static air pressures in the flow of 40–760 Torr. Two modes of pulsating discharge were implemented experimentally: without and with (from one to five) intermediate breakdowns. The discharge pulsation frequency, the maximum attainable voltage across the discharge gap, the length of the plasma channel, and the electric field in the discharge plasma were studied as functions of the air flow velocity and discharge current.