Exchangeable SRF Gun Reaches 30 MV/m Continuously

High-gradient electron guns need photocathodes that can be prepared, transported, replaced, and operated without sacrificing the vacuum and RF performance of the superconducting cavity. That combination has been difficult for continuous-wave SRF guns: an exchange mechanism introduces alignment, cleanliness, multipacting, and field-emission risks at the cathode port. The authors report a prototype 185.7 MHz SRF gun cryomodule for the proposed LCLS-II-HE Low Emittance Injector that operated continuously at a 30 MV/m cathode field using a copper plug.

Core Contribution

The paper's central result is not simply a high field measurement. It is the demonstration that the gun can retain that field after the cathode plug is fully removed and reinserted while the cavity remains at 4.4 K. The authors report no measurable degradation in high-field cavity performance and no increase in field emission attributable to the exchange. This directly addresses a practical requirement for semiconductor photocathodes, whose preparation and lifetime make replacement capability necessary.

The result is consequently well supported as a cryomodule-and-transfer-system demonstration, rather than merely a cavity test without a removable cathode interface.

Technical Approach

The cryomodule integrates an SRF cavity, cathode plug, cathode stalk, fundamental power coupler, frequency tuner, rinse ports, a superconducting solenoid package, and an ultra-high-vacuum suitcase. The cathode plug travels on a transfer rod from the suitcase to its operating position in the cathode stalk. The cavity design concentrates the relevant cathode field at the plug edge while controlling the maximum field on the niobium cavity surface.

Figure 1 lays out this integration and the RF-field distribution near the cathode region. The engineering point is the interface: the plug must be exchanged without converting the cathode port into a source of particulate contamination, vacuum degradation, or excess RF emission.

The insertion mechanism manages a 1.37 m path. Two sets of contact electrodes determine the stalk-axis position and angle, enabling trajectory corrections before the plug reaches its operating position. That arrangement is intended to prevent mechanical interference during insertion and to make repeated installation compatible with the small tolerances of the cathode region.

Figure 2 shows the three-stage path from the UHV suitcase, through transfer to the insertion axis, and into the operating position.

Results and Analysis

The reported operating point is a 30 MV/m cathode field in continuous-wave RF operation. During the full-field test, vacuum pressure remained below 1×10⁻¹⁰ Torr, the lower measurement limit of the gauge. The measured FE X-ray level was about 1 mR/hr at a distance of 1 m; the authors note that this level appeared after a pressure excursion associated with an inductive-filter failure. A subsequent test with the plug completely removed produced a comparable X-ray level, which supports their interpretation that the plug was not the dominant X-ray source.

The exchange test is the more consequential result. The gun operated in CW mode for 1.5 hours with the plug in its nominal flush position at 30 MV/m, and complete removal and reinsertion occurred with the cavity held at 4.4 K. In a separate thermal-control test, the plug was heated to approximately 300 K while the cavity remained at 4.4 K. Together, these observations show that the exchange hardware, UHV conditions, and high-field SRF operation can coexist in this particular copper-plug configuration.

The evidence is convincing for the stated integration milestone, especially because it includes reinsertion at operating temperature rather than a one-time assembly result. It does not yet establish photocathode performance: the reported high-field campaign used a copper plug, not a high-quantum-efficiency semiconductor cathode, and the paper does not report beam, emittance, quantum-efficiency, lifetime, or repeated-exchange statistics. The result therefore reduces a central hardware risk for the LCLS-II-HE injector, while leaving the photocathode and beam-quality program to subsequent tests.