Wide Range Of Plasma Conditions Research Articles

Atomic chlorine concentration measurements in a plasma etching reactor. II. A simple predictive model

Atomic chlorine concentrations in Cl2 plasmas have been measured using the technique of infrared absorption spectroscopy. A wide range of plasma conditions were examined, including power input densities of 0.05–0.5 W/cm−3, pressures of 200–800 mTorr, and excitation frequencies of 0.1–13.5 MHz. A simple kinetic model accurately predicts atomic chlorine concentrations as a function of discharge parameters. The primary source of atomic chlorine was assumed to be an electron-impact dissociation of Cl2, and the reaction rate coefficient was assumed to be independent of plasma conditions. The electron density was estimated from a plasma impedance analysis. The depletion of atomic chlorine was successfully modeled assuming second-order recombination on the electrode surfaces; depletion by gas-gas recombination and by flow from the reactor were insignificant over the range of plasma conditions studied. Because the diffusion of atomic chlorine is fast relative to the formation and depletion rates, the atomic chlorine concentration is fairly independent of interelectrode position. Atomic chlorine concentrations were accurately represented by the equation [Cl]=3.5×108 [e−]0.5P0.5, where [Cl] and [e−] are in cm−3 and P is in mTorr.

Parametric investigation of the density profile in the scrape-off layer of ASDEX

Systematic investigations of the scrape-off layer (SOL) in the midplane of ASDEX have been carried out in He, D 2 and H 2 for diverted ohmic discharges over a wide range of plasma conditions: n e ∼ 0.5−4.7 × 10 13 cm −3 , I p = 200–450 kA, B T = 16–23 kG, q a ∼ 2.4–4.4 and P OH = 200–480 kW. For the first two cm outside the separatix, n e is found to decay exponentially with an e-folding length λ n given by λ n = kq α (He, k = 1.32 cm, α = 0.52; D 2, k =1.29 cm, α = 0.35; H 2, k = 1.18 cm, α = 0.4) when from which follows for q a = 3: λ n ( D 2) ∼ λ n ( H 2) ∼ 0.8 λ n ( He). The q α scaling is roughly predicted by the simple formula λ n = D ⊥ L υ ∥ under the assumption D ⊥ ∝ m i −0.5 (as has been observed on ASDEX for H 2 and D 2). There appears to be no explicit λ n dependence on heating power. λ n varies strongly with n e in the range n e ≤ 1 × 10 13 cm −3 , decreasing for example (D 2,H 2; q a = 3.0), from λ n ≥ 3 cm at n e ∼ 0.5 × 10 13 cm −3 to λ n ∼ 1.9 cm for n e ≥ 1.5 × 10 13 cm −3 , n e at the separatrix is primarily a function of n e .

Flux, energy, and particle lifetime measurements for well formed field reversed configurations

Experimental flux, energy and particle lifetimes are reported for a wide range of plasma conditions (Ti = 100–800 eV, rs = 3.5–7 cm) in the slow risetime TRX-2 field reversed theta pinch. The lifetimes exhibit a scaling with the parameter rs/ρi01/2 to an exponent of between 2.7 and 3.6. This favourable scaling, especially for the flux lifetime, implies an average plasma resistivity which scales as the drift parameter vE/vi to some power greater than unity. For present small devices, typical flux diffusion coefficients are about 4 m2·s−1 and, based on the above scaling, would decrease rapidly in larger devices with smaller pressure gradient scalelengths. The above scaling is obtained only for well formed FRCs with low levels of formation turbulence, and it remains to be seen if this low turbulence level can be maintained in larger, less kinetic, FRCs.

Fusion-neutron production in the TFTR with Deuterium neutral-beam injection

We report measurements of the fusion reaction rate in the Tokamak Fusion Test Reactor (TFTR) covering a wide range of plasma conditions and injected neutral-beam powers up to 6.3 MW. The fusion neutron production rate in beam-injected plasmas decreases slightly with increasing plasma density ne, even though the energy confinement parameter neτE generally increases with density. The measurements indicate and Fokker-Planck simulations show that with increasing density the source of fusion neutrons evolves from mainly beam-beam and beam-target reactions at very low ne to a combination of beam-target and thermonuclear reactions at high ne. At a given plasma current, the reduction in neutron source strength at higher ne is due to both a decrease in electron temperature and in beam-beam reaction rate. The Fokker-Planck simulations also show that at low ne, plasma rotation can appreciably reduce the beam-target reaction rate for experiments with coinjection only. The variation of neutron source strength with plasma and beam parameters is as expected for beam-dominated regimes. However, the Fokker-Planck simulations systematically overestimate the measured source strength by a factor of 2–3; the source of this discrepancy has not yet been identified.

Measurements of ion cyclotron instability characteristics in a mirror-confined plasma

Frequency spectrum, wavelength, and phase velocity measurements of ion cyclotron oscillations in a mirror-confined plasma are described. The measured potential amplitudes reach, but do not exceed, the electron temperature. Other experiments in this device, not discussed here, reveal a wide range of plasma conditions where these oscillations are not detected.

Matrix spectroscopy of Ag, Ta, W, and Mo trapped in Ar and Xe using a triode sputtering source

Atomic species of tungsten, molybdenum, and tantalum were extracted from a triode sputtering source and subsequently trapped in xenon and argon matrices at 10 °K. It is shown that this approach is particularly convenient for matrix isolation spectroscopy of materials which are difficult to vaporize. Good correlation with the gas spectrum is obtained for the strong spectral features. There is evidence that some spectral features are greatly enhanced in the matrix spectra over those in the gas spectra. Using silver/xenon as a prototype system it is shown that a wide range of plasma conditions can be used without aggregation. However, certain plasma conditions are identified which lead to weak spectra of silver dimers. Spectral shifts due to matrix effects are found to be W/Ar = 850 cm−1, W/Xe = 700 cm−1 for ν̄⊥ < 37 000 cm−1 and 0 cm−1 above, Mo/Ar = 2650 cm−1, Mo/Xe = 1150 cm−1, Ta/Xe = 980 cm−1 for ν⊥ < 33 000 cm−1 and - 3230 cm−1 above. In the spectrum of Ta/Xe a remarkable red shift of 3230 cm−1 for energies above 33 000 cm−1 is observed. This is in sharp contrast to the Ta spectrum in argon, where a surprisingly consistent perturbation of approximately 1000 cm−1 to the blue is observed.

A neutral-potassium-beam measurement of plasma density

A 6-keV beam of potassium atoms has been used to measure the integrated line density of a deuterium plasma in which the electron temperature was in the range of a few hundred eV. and the mean ion energy was in the range of 5–10 keV. Line densities from 5 × 1011 to 1014 particles per cm2 can be effectively measured. The various contributions to the attenuation of the neutral beam have been examined in sufficient detail to permit an evaluation of this measurement to a wide range of plasma conditions.

The effect of ionization on heat transfer to wires immersed in a highly thermallyionized plasma

Results obtained with a previously developed technique for measuring local heat transfer to bare and ceramic coated metal wires from a dense, highly thermallyionized plasma are presented and compared with theoretical predictions. A free-burning arc in argon at atmospheric pressure provides a plasma in local thermodynamic equilibrium. Wire probes are swept through this plasma in planes parallel to the anode. The wires do not draw a net current from the plasma. Experimental evidence indicates that the layer between the undisturbed plasma and the probe surface is not in chemical equilibrium. The data show that heat transfer to the wire in this situation is dominated by the energy transport associated with electron and ion partial currents to the probe surface as determined by the potential of the probe with respect to the plasma. The data obtained using this technique provide an experimentally verified model for the heat transfer to wires immersed in a plasma over a wide range of plasma conditions including a 100 percent variation in the degree of ionization.

Thermal conduction energy loss in an argon afterglow

The decay of neutral gas temperature in low pressure argon afterglows has been measured by monitoring the acoustic standing waves excited by the ionizing discharge, with a microwave interferometer. Calculations, based on heat loss by atom thermal conduction only, give very good agreement with experimental results over a wide range of plasma conditions.

Thermal conduction energy loss in a helium afterglow

A free-space microwave interferometer (35 GHz) is used to monitor the acoustic standing waves excited by the ionizing discharge in low-pressure helium afterglows. From the frequency of these waves, the decaying gas temperature is evaluated. Theoretical calculations, based on heat loss by atom thermal conduction only, give very good agreement with the experimental results over a wide range of plasma conditions.

Retuning Effects and Dynamic Instability of a Radio-Frequency Capacitive Discharge

A discharge plasma is established in air at pressures of 0.2–20 Torr in the rf electric field between parallel-plate electrodes in a resonant circuit at 14 MHz. By adjusting a calibrated tuning capacitor in parallel with the electrodes, one can observe a wide range of plasma conditions extending from normal extinction at low rf potential and power to an apparent dynamic instability at high potential and power. The loading effect of the plasma is found to be that of a lossy capacitance. Net input rf power to the plasma-loaded circuit at the tuning condition just below the threshold for dynamic instability is found to be ∼10 times that for a plasma maintained at the tuning condition for the resonant circuit without plasma. The retuning effects are discussed on the basis of the nonlinearity of the plasma medium, and a possible mechanism for the dynamic instability is proposed.

An investigation of the electrical breakdown of a plasma - electrode system

Electrical breakdown between a pair of electrodes immersed in a flowing plasma corresponds to the transition between low- and high-current conduction régimes and has been attributed to many causes. In the work reported here, which involved cold electrodes in contact with the plasma produced by an electromagnetic shock tube, it was found that the minimum breakdown voltage corresponded to an average electric field in the cathode sheath of about 2 MV cm−1 over a wide range of plasma conditions. This was probably adequate to initiate the formation of non-thermionic cathode spots, and the voltage-current characteristics in the high-current régime were consistent with a field-emission theory of this mechanism, originally suggested by Smy. Moreover, electric fields in the plasma, sheath thicknesses and mean free paths were such that alternative explanations of the high-current régime were reasonably eliminated.