Successful user operation of a superconducting radio-frequency photoelectron gun with Mg cathodes

At the electron linac for beams with high brilliance and low emittance (ELBE) center for high-power radiation sources, the second version of a superconducting radio-frequency (SRF) photoinjector has been put into operation and has been routinely applied for user operation at the ELBE electron accelerator. SRF guns are suitable for generating a continuous wave electron beam with high average currents and high beam brightness. The SRF gun at ELBE has the goal to generate short electron pulses with bunch charges of 200–300 pC at typical repetition rates of 100 kHz for the production of superradiant, coherent terahertz radiation. The SRF gun includes a 3.5-cell, 1.3-GHz niobium cavity and a superconducting solenoid. A support system with liquid nitrogen (${\mathrm{LN}}_2$) cooling allows the operation of normal-conducting, high quantum efficiency photocathodes. We present the design and performance of the SRF gun as well as beam measurement results of the operation with Mg photocathodes at an acceleration gradient of $8\mathrm{MV}/\mathrm{m}$ (4 MeV kinetic energy). In the last section, we discuss the SRF gun application for production of coherent terahertz radiation at the ELBE facility.

Figure 1

Cross section of the ELBE SRF gun II cryomodule, holding a 3.5-cell gun cavity and a superconducting solenoid at about 70 cm from the cathode.

Figure 2

Comparison of the intrinsic quality factor $Q_0$ as a function of the accelerating gradient $E_{\mathrm{acc}}$ of SRF gun II. The diagram shows a significant improvement compared to SRF gun I, even though not all of the earlier vertical performance could be transferred into routine operation.

Figure 3

Cathode holder and ${\mathrm{LN}}_2$ cooling system.

Figure 4

Block diagram of the analog low-level rf controller.

Figure 5

Block diagram of rf system and laser synchronization to the ELBE master clock.

Figure 6

Geometry of photocathode. The insert shows a Mg plug photograph.

Figure 7

Photograph of Mg photocathode Mg#214 after second laser cleaning (left) and surface structure changes after cleaning (optical microscope image, right).

Figure 8

Scheme of the two-channel laser developed by MBI, Berlin.

Figure 9

Laser spot profile on the cathode before and after optimization of the UV conversion (the spot is 3.8 mm in diameter and is formed as an optical image of a physical aperture on the optical table).

Figure 10

The diagnostic beamline of the ELBE SRF gun II.

Figure 11

Measured Faraday cup current and corresponding bunch charge versus drive laser phase. Curves are shown for various UV laser powers.

Figure 12

Kinetic energy at gun exit versus laser phase for acceleration gradients between 6 and $9\mathrm{MV}/\mathrm{m}$.

Figure 13

Energy spread as a function of the laser phase for different bunch charges measured with the 180° bending magnet of the diagnostics beam line.

Figure 14

Measured normalized transverse emittance with slit scan versus bunch charge and comparison with astra simulation results.

Figure 15

Bunch length versus laser phase for 3 and 200 pC at the entrance of accelerator module 1 (cavity C1) measured with the cavity phase scan method.

Figure 16

Beamline layout for THz radiation production at ELBE.

Figure 17

Evolution of rms bunch length in the beam line from SRF gun to THz radiator for three different SRF gun drive laser phases. The simulation results are obtained from astra and elegant with typical experimental gun parameters: $20.5\mathrm{MV}/\mathrm{m}$ gun peak field on axis, 4 mm laser spot diameter, 2.3 ps rms laser pulse length, and 200 pC bunch charge.

Figure 18

Simulation results of longitudinal phase space distribution for 35° laser phase (a), 45° laser phase (b), and 50° laser phase (c) at the THz radiator position. The associated longitudinal charge distributions are shown in the plots (d) to (f).

Figure 19

Bunch length measurement after bunch compression with magnetic chicane versus cavity C4 phase for different bunch charges (a) and different drive laser phases (b).

Figure 20

Dependence of undulator radiation power at 0.7 THz on electron bunch charge.

Figure 21

Normalized THz intensity histogram for a typical 10 min measurement. The standard deviation of the Gaussian distribution is 0.0328.

Figure 22

Undulator radiation power produced in cw operation for different frequencies at the TELBE facility.