Gradient limitation in accelerating structures imposed by surface melting

A rough picture is beginning to emerge of the physics behind the maximum gradient that can be sustained in an accelerating structure without producing surface damage at a level sufficient to cause a measurable change in the rf properties of the structure. Field emission sites are known to trigger the formation of so-called plasma spots in regions of high dc or rf surface electric fields. A single plasma spot has a finite lifetime (/spl sim/20-50 ns) and leaves behind a single crater. In the rf case, some fraction of the electrons emitted from the spot pick up energy from the rf field and back-bombard the area around the spot. Depending on the gradient, pulse length and available rf energy, multiple spots can form in close proximity. The combined back-bombardment power density from such a spot cluster can be sufficient to raise the surface temperature to the melting point in tens of nanoseconds over an area on the order of 100 microns in diameter. This molten area can now support a plasma capable of emitting several kiloamperes of electrons with an average energy of 50-100 kV. This is sufficient beam power to collapse the field in a traveling structure in 30 ns or so. The plasma also exerts a tremendous pressure on the molten surface, sufficient to cause a macroscopic amount of material to migrate toward a region of lower surface field. Over time, this process can modify the profile of the iris tip and produce an unacceptable change in the phase shift per cell.