The Multipole Resonance Probe -- Concept, Theory, Experiments

Plasma diagnostics is a highly developed science. Of the many available techniques, however, only few are suitable for an industrial setting. To be useful for the supervision or control of technical plasmas, a diagnostic must be i) robust and stable, ii) insensitive against perturbation by the process, iii) itself not perturbing to the process, iv) clearly and easily interpretable without the need of calibration, v) compliant with the requirements of process integration, and, last not least, vi) economical in terms of investment, footprint, and maintenance. Plasma resonance spectroscopy -- exploiting the natural ability of a plasma to resonate on or near the electron plasma frequency -- is believed to provide a good basis for such an ``industry compatible'' plasma diagnostics. The idea is time-honored but has found renewed interest in recent years: An electric probe is used to feed a high frequency signal into the plasma and the response is evaluated with the help of a mathematical model. Once $\omega_{\rm pe}$ is known, one can calculate the electron density as $n_{\rm pe}=\omega_{\rm pe}^2\epsilon_0 m_{\rm e}/e^2=1.24 f_{\rm GHz}^2 \times 10^{10} {\rm cm}^{-3}$. The contribution will describe the general idea of active plasma resonance spectroscopy and introduce a mathematical formalism for its analysis. It will then focus on a novel realization, the so-called multipole resonance probe MRP, where the excited resonances can be mathematically classified and the connection between the probe response and the desired electron density can be evaluated analytically. The current state of the MRP project will be described, including the outcome results of first experiments.