First Experimental Characterization of Plasma Parameters and Carbon Decontamination Rates in a Microwave Resonator Used in Particle Accelerators

Abstract

In-situ plasma processing of superconducting radio frequency (SRF) cavities is a performance recovery technique used to mitigate the field emission limiting phenomenon. It has been proved very effective at major particle accelerator facilities such as SNS, CEBAF, FRIB, FNAL and C-ADS. This technique is based on the ignition of a noble-gas/oxygen plasma inside the cavity over several hours to remove hydrocarbon-based contamination, responsible for the parasitic field emission degradation observed after several years of operation. Despite a large experimental R\&D effort from the community, plasma parameters and cleaning rates under various experimental conditions have never been directly evaluated. In this study, plasma parameters were measured using a Langmuir probe and cleaning rates thanks to a quartz crystal microbalance (QCM) coated with an amorphous carbon film to simulate a carbon-based contamination. In this article, the main results from a large parameter space are discussed along with guidelines for improving the plasma processing effectiveness in SRF cavities. The encountered technical challenges are also discussed, as the SRF cavity is by design not intended to be a plasma reactor.

Figures (20)

• Figure 1: Electric field distribution for various resonant modes called higher-order modes (HOMs). Every mode has a unique E-field distribution, allowing the plasma to ignite in an area of interest by choosing the right HOM. For example, the 5th HOM has a high electric field area in the middle-height of the cavity, leading the plasma to ignite there in a toroidal shape, as the E-field. RF simulations were performed with COMSOL Multiphysics comsol
• Figure 2: Schematic of the experimental setup. The Langmuir probe can be moved on the $x$ axis, while the quartz crystal microbalance is locked in a fixed position for all experiments. Cameras are installed at the top and bottom parts of the cavity to allow a qualitative diagnostic of plasma distribution. In this case, the plasma is ignited in both accelerating gaps, thus benefiting from the numerous vacuum ports providing access to the plasma for diagnostics.
• Figure 3: Electric field distribution and plasma light associated to the 11th HOM at 683 MHz
• Figure 4: Coupling factor $\beta_{\text{ext}}$ as a function of several HOM frequencies. High-frequency HOMs are significantly better coupled than the fundamental mode at 88 MHz. The 683 MHz HOM is highlighted with the green dotted line, showing a 21 time increase in coupling compared to the fundamental mode.
• Figure 5: Experimental Paschen curves for the fundamental mode at 88 MHz, the second mode at 251 MHz, the fifth mode at 439 MHz, and the eleventh mode at 683 MHz, measured in an Ar/O$_2$(10%) gas mixture. For comparison, data for a He/O$_2$(10%) mixture are also shown for the 683 MHz mode. The solid line indicates the cavity plasma ignition threshold, while the dashed line represents the coupler breakdown threshold. The coupler breakdown threshold for the 683 MHz mode is not represented, as it could not be reached below $\approx 10^{-1}$ mbar.