Rf Breakdown in Accelerator Structures: from Plasma Spots to Surface Melting*

Plasma spots are known to form at field emission sites in regions of high dc or rf electric field. Several mechanisms for the formation of plasma spots in an rf field have been proposed, and one such mechanism which fits experimental data is presented in this paper. However, a plasma spot by itself does not produce breakdown. A single plasma spot, with a lifetime on the order of 30 ns, extracts only a negligible amount of energy from the rf field. The evidence for its existence is a small crater, on the order of 10 μm in diameter, left behind on the surface. In this paper we present a model in which plasma spots act as a trigger to produce surface melting on a macroscopic scale (~ 0.1 mm2). Once surface melting occurs, a plasma that is capable of emitting several kiloamperes of electrons can form over the molten region. A key observation that must be explained by any theory of breakdown is that the probability of breakdown is independent of time within the rf pulse—breakdown is just as likely to occur at the beginning of the pulse as toward the end. In the model presented here, the conditions for breakdown develop over many pulses until a critical threshold for breakdown is reached. Presented at the 22 International Linear Accelerator Conference (LINAC 2004) Lübeck, Germany, August 16–20, 2004 *Work supported by Department of Energy Contract DE-AC02-76SF00515 RF BREAKDOWN IN ACCELERATOR STRUCTURES: FROM PLASMA SPOTS TO SURFACE MELTING* Perry B. Wilson, SLAC, Stanford CA 94309, USA