OF RECENT PROGRESS ON HIGH REPETITION RATE , HIGH BRIGHTNESS ELECTRON GUNS

In the last few years, the formidable results of x-ray light sources based on FELs opened the door to classes of experiments not accessible before. Operating facilities have relatively low repetition rates (~ 10-100 Hz), and the natural step forward consists in the development of FEL light sources capable of extending their rates by orders of magnitude in the MHz regime. Additionally, ERL based x-ray facilities with their promise of outstanding performance also require extremely high, GHz-class repetition rates. The development of such facilities would represent the next revolutionary step in terms of science capability. To operate such light sources, an electron injector capable of MHz/GHz repetition rates and with the brightness required by X-ray FELs or ERLs is required. Such injector presently does not exist. In response to that, many groups around the world are intensively working on different schemes and technologies that show the potential for achieving the desired results. This paper includes a description of the requirements for such injectors, an overview of the pursued technologies and of the results obtained so far by the groups active in the field. INTRODUCTION With the advent of 4 generation light sources, X-ray source facilities based on FELs or ERLs, the availability of high brightness electron guns became of paramount importance. Indeed, the x-ray performance of such facilities directly depends on the electron beam quality and in particular on its brightness. Up to many hundreds of pC bunches are usually required with sub-micron normalized emittances in both transverse planes. Storage rings cannot satisfy such a requirement and linac based accelerators are required. The highest brightness obtainable in a linac is ultimately defined by the injector and in particular by the electron gun performance. The spectacular success of the first x-ray FELs [1-4] is in large part due to the high brightness performance of their electron guns. Such facilities operate at relatively low repetition rates (<~100 Hz), but proposed high average brightness x-ray FELs [5] and FEL oscillators [6] requires MHz-class repetition rates, and proposed ERL-based 4 generation light sources ask for even higher rates beyond 1 GHz [7-10]. The normalconducting (NC) high-frequency RF technology (1.3-3 GHz) used in the excellently performing electron guns used in existing 4 generation light sources cannot be scaled to repetition rates higher than ~10 kHz (while maintaining the required accelerating fields) because of the excessive power density dissipated on the gun cavity walls [11]. Electron guns capable of 4 generation light source quality beams at MHz/GHz repetition rates presently do not exist, and a number of groups around the world are working on schemes to fill that gap up. This paper includes a review of the requirements for such guns and of the technologies adopted by the different groups, and an overview of the active projects and of their recent results. HIGH-REPETITION RATE HIGHBRIGHTNESS GUN REQUIREMENTS Table 1 contains the list of the requirements that an electron gun has to satisfy in order to operate in a highrepetition rate 4 generation light source. Table 1: Fourth Generation Light Source Electron Gun Requirements Parameter Value or Comment Repetition rate from ~1 MHz to ~1.3 GHz Charge per bunch from ~1 pC to ~1 nC Norm. transverse emittance from ~0.1 to ~ 1 μm with values increasing with increasing charge/bunch Beam energy at gun exit > ~500 keV (to control space charge induced emittance increase) Electric field at the cathode at emission > ~10 MV/m (for controlling the space charge limit at emission) Bunch length from ~100 fs to ~50 ps for controlling charge density Compatibility with magnetic fields in the cathode/gun area mainly for emittance compensation but also for advanced beam manipulation Acceptable dark current to avoid quenching in the superconductive linac and for controlling radiation doses Capability of operating high quantum efficiency photo-cathodes to deliver the required charge and current with available laser technology Operating vacuum pressure ~10 – ~10 Torr for cathode lifetime Capability of an “easy” cathode replacement requires a vacuum loadlock system Reliability compatible with a user facility operation The requirements in Table 1 have been discussed in detail elsewhere [12], here we want to remark the particular importance that some of those specifications ____________________________________________ *Work supported by the Director of the Office of Science of the US Department of Energy under Contract no. DEAC02-05CH11231 FSannibale@lbl.gov FRXBA01 Proceedings of IPAC2012, New Orleans, Louisiana, USA ISBN 978-3-95450-115-1 4160 C op yr ig ht c ○ 20 12 by IE E E – cc C re at iv e C om m on sA tt ri bu tio n 3. 0 (C C B Y 3. 0) — cc C re at iv e C om m on sA tt ri bu tio n 3. 0 (C C B Y 3. 0) 03 Particle Sources and Alternative Acceleration Techniques A15 New Acceleration Techniques have. In order to obtain the required brightness, space charge must be carefully controlled along the gun and the injector. Because of the need of controlling field emission to minimize dark current, and because limitations of some of the gun technologies, the field at the cathode during photoemission can be significantly smaller than in the case of low repetition rate guns. This situation forces the beam into a different beam dynamics regime where relatively long bunches must be used to reduce the charge density and control space charge effects. Simulations [13] and initial experimental results [14] indicate that the desired performance can be obtained if the beam energy and accelerating gradients in Table 1 are achieved. Nonetheless a complete experimental demonstration is still required. In addition to the control of the bunch length, the optimization of the beam brightness also requires control of the spatial distribution of the bunch. A 3D uniform ellipsoidal distribution generates linear space charge forces that do not increase the rms emittance and thus represents the ideal distribution to pursue. Such a distribution is difficult to generate and the so-called “beer can” distribution is often adopted as a reasonable and easier to generate tradeoff. Alternatively a special technique, the so-called “beam blow-out” [15, 16] generates the 3D uniform ellipsoid by starting with a very short (< ~100 fs) and relatively wide beam. Under the action of its own space charge forces this “pancake” beam evolves in the desired ellipsoidal distribution. The technique that has been demonstrated for bunches with charges/bunch <~100 pC [17], generates an ellipsoidal distribution with a very regular longitudinal phase space and a reasonable transverse emittance. The bunch distribution control requirement makes of photocathodes the most chosen cathode system. A notable exception is represented by the SACLA FEL injector [18] where a thermionic cathode is used in combination of a complex and articulated compression system to successfully generate the high brightness beam. Delivering the required charge and current with the power available by the present laser technology requires using high quantum efficiency photocathodes with QE > ~10. Semiconductor cathodes, such as Cs2Te, GaAs:Cs or CsK2Sb for example, offer the required QE but are quite reactive and sensitive to ion back bombardment. Because of this, in order to achieve acceptable lifetimes, they require the low operational pressures shown in Table 1. A complete review of photocathode systems can be found in reference [19]. AVAILABLE GUN TECHNOLOGIES The list of technologies presently pursued for the development of high-repetition rate high-brightness electron guns includes direct current (DC), superconducting RF (SRF) and low frequency (< ~700 MHz) normal conducting (NC) continuous wave (CW) RF guns plus some hybrid configuration. DC schemes allow for arbitrarily high repetition rates, are compatible with the application of magnetic fields in the cathode/gun area, have demonstrated extremely low vacuum pressures (~10 Torr) and are compatible with all cathodes presently under consideration. The areas where improvements are still required include higher beam energies at the gun exit and higher gradients at the cathode. Field emission from the gun metallic parts creates charge build up in the gun ceramic insulator generating voltage breakdown and ceramic punctuation when energies higher of the presently achieved ~350 kV are attempted. Relatively recent results from the JAEA group [20] are showing promising progress towards higher energies, more details in the next section. Electron guns based on SRF schemes allow for very high repetition rates (> ~1 GHz), are potentially capable of high gradients at the cathode, and because of the effective cryopumping performed by the superconducting walls are capable of a very good vacuum performance. Meissner field exclusion does not allow for externally applied magnetic fields forcing the use of cryogenic solenoids for the emittance compensation. Questions on compatibility of semiconductor cathodes with the SRF environment needs to be carefully investigated, even though promising results with Cs2Te have been demonstrated at the Rossendorf gun [21]. Reduction of multipacting (especially in the RF coupler area) and of field emission at higher gradients is actively pursued by the groups developing SRF schemes. Performance reproducibility of the SRF structures also requires further improvements. In NC CW RF guns the frequency is lowered below ~700 MHz to make the power dissipated on the gun cavity walls small enough to be removed by water cooling while operating in CW mode with the gradients required for the generation of high brightness electron beams. NC CW RF schemes can achieve repetition rates of hundreds of MHz and are compatible with magnetic fields in the cathode/gun area. If the frequency is lowered down into the VHF range (30-3