Search

A method is presented for direct determination of the frequency of ionization and of the coefficient of free diffusion for electrons in a microwave discharge. The method is based on determining the rate of growth of the electron density in a microwave cavity in which an electric field larger than that necessary for breakdown is applied. The ionization frequency and the coefficient of free diffusion for electrons in hydrogen were measured in a pressure range of 15 to 45 mm Hg. Comparison is made with existing theory.

In a high frequency discharge plasma resonance maximizes the electric field, thus producing a high ionization rate in the regions near resonance. The effect of this on the distribution of electrons and of ionization in a parallel plane discharge is calculated and compared with the observation that the light from such a discharge often is a minimum at the center.

Experimental methods are given for determining the complex admittance and electron density of gas discharges by the use of microwave techniques. Applications are discussed where the discharge is contained in either high or low Q resonant cavities.

The effect of magnetic field on the high frequency breakdown of gases has been studied. The presence of energy resonance and the modification of diffusion are shown experimentally and explained theoretically. An application is made of both the average electron theory and the Boltzmann theory, and the correspondence between these two theories is discussed.

An analysis is presented of an experimental study of microwave scattering from nonequilibrium plasmas in which the electrons had a large steady drift velocity ${v}_{d}$ with respect to the ions. According to theory, such plasmas must support "relatively undamped" acoustic-ion-plasma waves due to a two-stream instability type of mechanism. Measurements of the frequency-power spectrum of the scattered signal show agreement with theory. In particular, as predicted, the measured scattering cross sections were large in comparison with the Thomson cross section. The effects of collisions predicted by theory did not show up in the experimental results. As expected from theoretical considerations of nonlinear effects (such as the nonlinear coupling between the plasma modes of different wave numbers), the relative drift velocity and the electron-ion temperature ratio tend toward a saturation value as the plasma current is increased. The experimental results indicate that the nonlinear effects are relatively weak.

Microwave techniques are applied to the study of the diffusion of electrons and positive ions in helium. For the electron and ion densities realized experimentally, the diffusion takes place ambipolarly; that is, the flow of charged particles of both signs is equalized by the space charge field which they set up. At 1 mm Hg pressure and 0.039-ev average energy, the measured value of the ambipolar diffusion coefficient, ${D}_{a}$, is 540 ${\mathrm{cm}}^{2}$/sec. The measured variation of ${D}_{a}$ with gas density, energy of the particles, and diffusion container size and shape agrees closely with theory.

Values for the collision cross section for momentum transfer and the fractional energy lost by an electron on collision with a hydrogen molecule were obtained from measurements of the microwave conductivity of a gaseous plasma. The experiments were made in the afterglow of a pulsed discharge in a cavity resonating at a wavelength of 10 cm. The mean electron energy was varied from 0.04 ev (room temperature) to 0.08 ev by heating the gas, and from 0.04 ev to 1.6 ev by microwave agitation of the electrons alone. The collision probability in molecular hydrogen at electron energies in the neighborhood of 0.04 ev is found to be $28.5{(\frac{v}{{v}_{0}})}^{0.55} {(\mathrm{c}\mathrm{m}\ensuremath{-}\mathrm{m}\mathrm{m}\phantom{\rule{0ex}{0ex}}\mathrm{H}\mathrm{g})}^{\ensuremath{-}1}\ifmmode\pm\else\textpm\fi{}3%$, where $v$ is the electron velocity, and ${v}_{0}$ the most probable velocity at 300\ifmmode^\circ\else\textdegree\fi{}K. Measurements in deuterium gas gave the same result. The collision probability in hydrogen increases to a peak value of 64 ${(\mathrm{c}\mathrm{m}\ensuremath{-}\mathrm{m}\mathrm{m}\phantom{\rule{0ex}{0ex}}\mathrm{H}\mathrm{g})}^{\ensuremath{-}1}$\ifmmode\pm\else\textpm\fi{}7% at an electron energy of approximately 1.4 ev. The energy loss of electrons near room temperature was found by comparing the conductivity measurements obtained by thermal agitation with those obtained by microwave agitation of the electrons. The fraction of the excess electron energy lost on collision with the hydrogen molecule is (3.5\ifmmode\pm\else\textpm\fi{}0.5)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}3}$. In the higher energy range, the loss and the collision probability were computed from a comparison between microwave conductivity measurements and electron drift velocity and diffusion experiments.

A study of plateau slopes of self‐quenching Geiger‐Muller counters shows that no quenching vapor tested produced flat characteristics. On the assumption that the slopes were given by spurious counts arising from the gas discharge phenomena, a study was made of the controlling factors. Neither the quenching gas nor the vehicular gas positive ions were found to be significant. A study of the negative ions produced by molecular dissociation of the quenching gas at the cathode, and electron attachment in the body of the gas shows that these factors are directly correlated with the slopes of the plateau characteristics.

The criterion for breakdown of a low pressure gas at microwave frequencies is that the ionization rate equal the diffusion rate for electrons. This principle is applied to compute the breakdown of air in coaxial cylinders using previously reported high frequency ionization coefficient data obtained from breakdown experiments between parallel plates. Agreement between computed and measured results at 9.6-cm wave-length supports the validity of the above criterion for breakdown. The method of computation is illustrative of the general problem of computing breakdown fields for tubes of arbitrary shape and field configuration using ionization coefficient data obtained from uniform field breakdown.

The diffusion equation for electrons in a nonuniform field is solved and the breakdown condition derived. The breakdown condition is expressed in such a manner that an effective characteristic diffusion length Λe is determined; the meaning of Λe expresses the equivalent characteristic diffusion length for uniform electric fields. From the experimental breakdown fields, in which the electric field is uniform, Λe is determined and used to predict theoretical breakdown curves for the nonuniform cases. Theory and experiment are compared, the agreement verifying the correctness of the approach.

Microwave techniques are applied to the measurement of the recombination between electrons and positive ions. The measured recombination coefficients for monatomic gases range from 1.7\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}8}$ cc/ion-sec. for helium to 3\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}7}$ cc/ion-sec. for argon. The recombination coefficients for diatomic gases range from 2.5\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}6}$ cc/ion-sec. for hydrogen to 2.8\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}7}$ cc/ion-sec. for oxygen. At electron temperatures of 300\ifmmode^\circ\else\textdegree\fi{}K, the recombination is found to be independent of pressure for the monatomic gases. A pressure dependence is observed in diatomic gases, the recombination increasing with increasing pressure in nitrogen and oxygen.

Cyclotron emission from high-energy plasmas is calculated for two classes of electron distribution functions: (a) those that decrease monotonically with increasing electron energy, and (b) those that have one or more maxima displaced from zero energy. In (a) the emission does not differ greatly compared with the emission from a Maxwellian plasma of the same energy. In (b) the emission can grow exponentially with distance traversed in the plasma, resulting in a greatly enhanced loss of radiant power.

The high-frequency electric field required to maintain a hydrogen plasma has been measured as a function of pressure and plasma electron density. A theory of the plasma based on a solution of the Boltzmann transport equation has been developed to predict this field; it agrees satisfactorily with experiment. The theory has no adjustable parameters, and uses only the probabilities of collision, excitation, and ionization of the gas by electrons, and the ionic mobility.

Characteristics of a gamma‐ray counter making use of a 3000‐megacycle gas discharge in a coaxial mode cavity resonator are described. The breakdown of a discharge in gases at these frequencies does not depend upon secondary effects. Since the breakdown is not controlled by positive ions, shorter breakdown and resolving times than those of a Geiger‐Muller counter are realized. In the high frequency discharge, ambipolar diffusion is the controlling factor in loss of electrons. A d.c. clearing field is superimposed on the a.c. field to eliminate electrons rapidly. A quenching agent is present to suppress electron emission from positive ion bombardment of the walls. The discharge is extinguished by reducing the magnetron power electronically. The plateau and sensitivity characteristics are similar to those of a Geiger‐Muller counter.

A plasma in a dc magnetic field has a transverse conductivity component whose reactive part depends, as to both magnitude and sign, on the strength of the magnetic field. By measuring the value of magnetic field that is necessary for bringing this reactive part to zero—and hence removing any resonant‐frequency shifts of a TE011 mode of a cylindrical cavity containing the plasma—determination of the probability of elastic collision of electrons in helium is obtained as a check on the method. The value of Pm=20±1 cm−1 mm Hg−1 for the electrons at room temperature thus obtained agrees substantially with the value obtained by previous workers. Sources of error in the measurement are discussed.

An electric field of sufficiently high frequency applied to electrons in a gas may deliver energy to the electrons without imparting to them any continuous drift motion resulting from the field. The criterion for breakdown of a low pressure gas at microwave frequencies is therefore that ionization by collision of electrons with neutral gas molecules replace loss by diffusion to the walls of the discharge tube. The condition is mathematically expressed as a simple boundary value problem. This breakdown principle is applied to converting microwave breakdown measurements into measurements of ionization rates as a function of the electric field strength, pressure, and frequency. A new ionization coefficient is introduced appropriate to the high frequency discharge conditions, and its relation to the d.c. Townsend coefficient is explained. The energy transfer from the electric field to the electrons at a given $\frac{E}{p}$ is shown to be most efficient when the pressure is high enough or the frequency low enough to result in many collisions of electrons with gas molecules per cycle. This maximum efficiency is equal to the d.c. energy transfer efficiency. When the pressure is lower or the frequency is higher, the electrons have an out-of-phase component of motion and do not receive energy so efficiently, resulting in lower ionization rates observed experimentally.

A phenomenological description of high-frequency gas-discharge breakdown is given, describing the similarities and differences between these discharges and the more familiar dc type. High-frequency discharge breakdown is controlled by the process of electron diffusion and, besides the theory of its behavior, the physical limitations of tube size, gas pressure, and frequency for this type of breakdown are given. The particular case of hydrogen is cited. The effects of superimposing a small dc field and a magnetic field on the ac field are also discussed.

Microwave measurements of the radiation from a weakly ionized low-energy plasma immersed in a magnetic field show significant departures from the Kirchhoff-Planck law. The departures can be explained by taking account of the non-Maxwellian distribution of the radiating electrons. By comparing the measured and calculated radiation temperatures we estimate the distribution of electron velocities and their mean energy. (auth)

A previously reported microwave method for determining the collision probability for momentum transfer of slow electrons has been modified so that a variation in average electron energy from 0.012 ev to 3 ev may be obtained. Measurements of the ratio of the real part to the imaginary part of the electron conductivity are performed in the afterglow of a pulsed helium discharge in a microwave resonant cavity. The average electron energy is varied by applying a microwave electric field in the afterglow and, under appropriate assumptions, the average electron energy is determined theoretically from this field. Measurements are also obtained by varying the gas temperature from 77\ifmmode^\circ\else\textdegree\fi{}K to 700\ifmmode^\circ\else\textdegree\fi{}K. The value of the collision probability for momentum transfer in helium is 18.3\ifmmode\pm\else\textpm\fi{}2 percent ${\mathrm{cm}}^{2}$/${\mathrm{cm}}^{3}$ per mm Hg from 0 to 0.75 electron volts and increases slowly to a peak value of 19.2\ifmmode\pm\else\textpm\fi{}2 percent at 2.2 ev.

The theory of the positive ion saturation region of probes at low pressure is modified to take into account the directed current at the sheath edge. This drift current explains the discrepancy between the theoretical and experimental ratio of the electron saturation probe current to the positive ion saturation probe current. The theory is checked by using microwave methods to obtain an independent measurement of the ion density. Probe and microwave measurements are compared in the pressure region of 0.05-6 mm Hg in hydrogen, argon, and helium.

A microwave method is given for determining the collision probability for momentum transfer for electrons in thermal equilibrium with a gas. An integral equation is developed which gives the ratio of the resistive and reactive components of the microwave conductivity in terms of the collision frequency for electrons with the gas atoms. Two approximate solutions are given for the collision frequency as a function of electron energy. An experimental procedure is described in which the conductivity ratio is determined from the shift in the resonant frequency and the change in conductance of a resonant cavity. Measurements are made during the decay of a pulsed discharge after the electrons have cooled to the gas temperature. Preliminary results are given for several gases.

Calculations are given for the radiation temperature in terms of the average electron energy, to be used for interpreting microwave radiation measurements from plasmas with non‐Maxwellian distributions of electron velocities.

The average electron concentration of a gas-discharge plasma can be inferred from the phase change suffered by an electromagnetic wave in its passage through the medium. With available microwave generators, the maximum electron density that can be measured is approximately 3×1014 cm−3. On the other hand, interferometers that operate at optical wavelengths become insensitive at electron densities less than approximately 1016 cm−3. An infrared interferometer is described here which operates at 0.03 cm wavelength and is able to detect densities above 5×1013 cm−3 in a plasma column 1 cm thick. In the design, incoherent radiation from a mercury arc lamp is split by a diffraction grating and the plasma placed in one of the diffracted beams. The beams are then recombined and fringe shifts measured by a Golay detector.

Measurements of externally excited moving striations in a low‐pressure (0.005‐0.03 Torr) argon discharge are presented. The wave propagates with little radial structure, and corresponds to the lowest radial mode. A theory which uses the Boltzmann equation for the electron motion, and fluid equations for the ions gives results in agreement with the experiment. Included in this model are the effects of ion inertia, and of depletion of the high‐energy tail of the electron distribution function due to loss of electrons to the tube walls. Results indicate that both effects are important at low pressure. A simple model, which uses fluid equations for electrons and ions, shows the relationship between moving striations and ion acoustic waves.