Comparison between millimeter wave interferometry and Langmuir probe measurements in low pressure discharges

Summary form only given. used to measure the electron density in a low pressure, rare gas plasma maintained in a cylindrical discharge tube. A comparison of measurements shows that the interferometer-derived density is greater than the probe-derived density. Initial measurements indicate that the ratio (interferometer/probe) increases with increasing gas pressure. Measurements were made over ranges of gas pressure and discharge current where the corresponding discharge was free of striations. Both diagnostics were expected to be valid in the parameter regime under investigation. Descriptions of the diagnostics, comparisons between the measurements, and suggestions for the discrepancies will be presented. The interferometer, which has a Mach-Zehnder configuration, employs a heterodyne detection system and has two free-running oscillators, both operating nominally at 75 GHz and separated in frequency by about 60 MHz. The radiation from one oscillator is split along two paths. In one path the radiation is directed to, and mixed with, radiation from a second (local) oscillator; this constitutes a reference signal. In the other path the radiation is frequency doubled to 150 GHz, directed through the discharge, and subsequently mixed with the local oscillator; this constitutes a test signal. The line-averaged density through the discharge is determined from the phase difference between the paths, which is given by the phase difference between the test and reference signals. Both probe and interferometer were scanned across the transverse direction of the cylindrical discharge tube. The 150 GHz (/spl lambda/=2mm) radiation that passes through the plasma has a fundamental Gaussian transverse mode with 6 min diameter beam waist that is located at the axis of the discharge tube (midway between transmitting and receiving antennae, which are 75 min apart). The probe and interferometer measurements were compared directly by integrating the probe-derived density over position to obtain line-averaged density. This was compared to the line-averaged density obtained by the interferometer for the same chord. In addition, density versus position was obtained from the interferometer data by assuming a cylindrically symmetric model for the electron density and finding free parameter(s) in the model which allow for the best fit, in the least squares sense, to the interferometer data. One assumed model was a constant multiplied by the density obtained from the probe. Other models included cylindrically symmetric polynomials containing terms of various order. Interferometer-derived densities were found to be fairly independent of these model choices. The probe and interferometer were compared by then considering the average density over the cross section of the discharge tube. These comparisons were quantitatively the same as those from the direct comparison cited earlier.