The commissioning of the ALICE DC (direct current) photogun was a major project in it's own right. This type of electron source for particle accelerators was the first of its type to be commissioned in the UK.
The commissioning over the history of ALICE has involved several stages. The initial commissioning to operate at 350 kV was performed in 2006-2007 and is detailed below. Although the goal of 350 kV operation was achieved (using the original large diameter ceramic insulator) serious difficulties with gun performance and integrity persisted, and eventually it was decided to install a smaller diameter ceramic insulator and operate at a lower voltage of 230 kV, in order to progress to full machine commissioning including demonstration of energy recovery and operation of the infra red free electron laser.
ALICE operated reliably with the DC photogun at 230 kV from 2008 to 2012. In 2012, as part of the ceramic development research at Daresbury, a new improved large diameter ceramic was installed. This allowed operation of the gun at 325 kV.
The initial commissioning of the ALICE DC photogun took place over four separate periods. Each period ended in the gun being either disassembled because of problems developed or opened up to replace the photocathode. The aims of commissioning were twofold. Firstly, to condition the gun for the operational voltage of 350kV and to ensure a reliable cathode activation for the quantum efficiencies of at least 2-3% that are sufficient to generate electron bunches of nominal 80pC charge.
Secondly, to fully characterise the electron beam and compare the measurements with the ASTRA simulations. The beam investigation was conducted with a dedicated diagnostic beamline which, after completion of the gun commissioning, was disassembled to allow positioning of the superconductive RF linac. (We however plan to re-install this beamline on ALICE, see gun beamline upgrade)
This period was significant in that we have gained first experience working with high voltage DC photoguns and that ALICE generated its first beam on 16th August 2006.
The gun HV conditioning was not however completed fully partially because of the conditioning resistor failures. The quantum efficiency (QE) of activated photocathodes never exceeded 0.5% and the cathode lifetime was very low.
Eventually, the gun has developed a serious current leakage that prevented the gun from further operation. We have conducted a number of experiments with the aim to pinpoint the problem but the results were not entirely conclusive. After the gun disassembly, a distinctive black spot has been found on the surface of the anode electrode that confirmed the current leakage took place on the vacuum side of the gun.
The gun was reassembled after general cleaning, electrode polishing and installation of a new GaAs wafer. This time, a strong field emission was observed from the cathode that prevented us from operating the gun at its nominal voltage and optimal solenoid settings. The QE was higher (normally above 1% on the freshly activated cathode) but the cathode lifetime was again very low. Moreover, the QE was highly non-uniform - only one half of the cathode area could be activated. This was traced down to a problem with a non-uniform heat cleaning of the wafer surface.
We nevertheless managed to test and characterise all gun beamline and diagnostic components thus identifying a number of issues to be resolved.
At the end of this period, a mechanical failure inside the cathode ball took place and that necessitated another disassembly of the gun for repairs.
This period of the gun commissioning was plagued by two problems. One was a huge beam halo surrounding the main beam that was probably caused by the overheating of the cathode during the heat clean process. The other was a significant contamination of the gun surfaces after each cathode activation that required a prolonged HV gun conditioning to restore the high voltage handling capability of the gun.
The gun was eventually vented for the cathode replacement after several unsuccessful attempts to activate the cathode without compromising the HV integrity of the gun. At the same time, important changes were introduced to the activation technique including the light modulation for controlling the activation process and switching from O2 to NF3 gas for the cathode activation.
This last period of the gun commissioning proved to be the most fruitful in terms of the gun operation and the electron beam characterization. We made, perhaps for the first time, a full set of measurements of the electron bunches including the transverse emittance, longitudinal profiles, energy spectra at various bunch charges from 1 to 80pC. The QE was routinely above 3% after cathode activations that allowed us to achieve the bunch charges well above 100pC however the cathode lifetime remained to be an issue. Unfortunately, the cathode still exhibited a few field emission spots and this prevented us from operating the gun at optimal solenoid settings. This and the very fact of the FE presence accounted for a higher transverse emittance compared to that predicted by the computer simulations.
In terms of energy spectra and the bunch longitudinal profiles the ASTRA model matched the experimental results remarkably well. Some other experiments have been conducted including a comparison of the bunch characteristics at two different lengths of laser pulses. The commissioning results were reported at the EPAC and presented in a number of reports that can be found in documentation and publications.
During the gun commissioning, the most frustrating were several failures of the HV ceramic braze while baking out that resulted in losses of time and huge amount of work for technical staff. A collaborative effort between the JLab, Cornell and Daresbury to design and manufacture improved ceramics would eventually rectify this problem in 2012, see gun ceramic development.