Rf guns: bright injectors for FEL - Triangle Universities Nuclear ...

More documents

Recommendations

Info

292 C Travier / Rf guns: bright injectors for FEL introduces mixing between longitudinal and transverse phase-space planes . AFEL design is based on this theory . 3 3. CEA Bruyères-le-Châtel The Commissariat à l'Energie Atomique (CEA) in Bruy6res-le-Châtel (France) has built a laser-driven rf gun as an injector for a high-power infrared FEL [17] . In order to improve the beam characteristics, a frequency of 433 MHz was chosen for the linac and a subharmonic frequency of 144 MHz for the gun . At first, one cavity is used providing a 1 MeV beam (fig . 5) . Later, a second cavity will be added . Cs 3Sb and CsK,Sb photocathodes are used . Other types are also tested . The goal is to produce a 50-100 ps, 10-20 nC beam pulse . Design parameters [18] are summarized in table 2 . This gun is now under testing . An accelerating field of 25 MV/m on the cathode has been obtained . 20 nC, 100 ps pulses have been accelerated to 1 MeV . An emittance of 901T mmmrad was measured for a 10 nC pulse [191 . 3 .4 . Brookhaven National Laboratory . In order to study several ideas related to future particle acceleration methods, Brookhaven National Laboratory (BNL) is building an accelerator test facility (ATF) [20] which will also be used for FEL physics . The injector for the accelerator is a laser-driven rf gun operating at 2856 MHz. It consists of a one and a half cell cavity incorporating a photocathode in the end wall of the first cell . The cavity shape was designed so that rf fields cause minimal nonlinear distortion of the phase spaces (fig . 6) . This was obtained by using a Fourier decomposition of the rf fields and by setting the radial field to be linear [21] . The gun is designed to operate with a 100 MV/m electric field at cathode. The laser will provide a 4 to 6 ps pulse . 1 nC is expected from the yttrium photocathode The main design parameters are shown in table 2 . The gun has already been tested with a 20 ns pulse from an excimer laser . A 90 MV/m accelerating field was obtained at the cathode . A peak current of 0.6 A was accelerated to 3.6 MeV [20] . The picosecond Nd : YAG laser is now available and experiments will proceed soon . Meanwhile, extensive studies of metallic photocathodes are done [22] . 10 4 " ~ M LASER-DRIVEN GUNS EXPERIMEN-AL RESU_TS -ASER-DRIVEN GUNS DESIGN PARAhETEPS HERMIONIC GJNS EX~EPir'ENTAL PARAMEIERS the brightness as a function of the peak current and the bunch length respectively, for the different rf gun projects. These figures are only indicative and should be considered with some care . Fig . 7 also shows the brightness corresponding to different wavelength FEL . From these figures a few general comments can be made. Both thermionic and laser driven rf guns have demonstrated their capability to produce very bright beams (roughly one order of magnitude higher than conventional injectors) . Thermionic rf guns provide smaller peak currents than photoinlectors . The best laser-driven rf gun should be sufficient to reach wavelengths m the XUV domain provided that suitable wigglers can be built for such short wavelengths . X-ray domain is out of reach for the current state of the art . E Q cn cn w z 10' 4 10 12 10 100 1000 10000 CURRENT (A) Fig. 7 Rf gun brightness vs peak current . LASER-DRIVEN GUNS EXPERIMENTAL RESU-IS O LASER-D-2IVEN GUNS DESIGN PARAMETERS 'HER - 1ONIC GUNS EXPERIMENTAL PARA - E - ERS 4 . State of the art A figure of merit of the beam quality can be the peak brightness defined here as two times the peak current divided by the square of the normalized emittance . It is expressed in A/(m 2 rad2 ) . Figs . 7 and 8 show ro 10 1 10 100 1000 BUNCH LENGTH (PS) Fig . 8 . Rf gun brightness vs bunch length
C Tracter / Rf guns bright injectors for FEL 293 What are the present limitations in rf gun technology? As mentioned earlier, thermionic rf guns suffer from the backbombardment problem. The extra heat due to the backbombarded electrons increases the peak current during the rf macropulse which is not good for FEL operation. It also limits the duty cycle . Adding a transverse magnetic field improves the situation since the most energetic electrons come back on the cavity wall beside the cathode . Using this technique, Stanford could obtain a 10 ps macropulse repeated at 15 Hz without current increase during the macropulse [10) . If more current and higher duty cycles are needed, the backbombardment problem will still be a limitation . Conventional thertmonic cathodes have current densities of a few tens of A/cm2 . Recently, however, more than 100 A/cm 2 has been reported [23] . These values are nevertheless much smaller than that reported for photocathodes. Since in a thermionic rf gun, every rf bucket is filled, the beam loading is very important . Reflected electrons and secondary electrons also participate to beam loading . At equivalent peak current, thermionic rf guns require more rf power than laser-driven rf guns . This fact could be a limitation for high currents . Photoinjectors rely on photocathodes and laser technology . The optimum photocathode (good quantum efficiency, long lifetime, good ability to withstand high fields) has not yet been found . The most successful photocathodes mainly studied at Los Alamos (Cs3Sb, CsK2Sb) have a very good quantum efficiency (2 to 8%) . Their lifetime depends on the operating conditions . For their typical operating conditions (200 times 5 nC micropulses repeated at 1 Hz, 2 x 10-9 Torr pressure), Los Alamos HIBAF photocathodes have a 15 h lifetime [24), the production time for one cathode being 3 h . In order to improve this lifetime, Sheffield suggested to use a glow discharge to clean the cavity prior to its operation [251 . The main alternative to these multi-alkaline photocathodes are the thermionic cathodes (dispenser, LaB6 ) and the metallic cathodes . They both have a lower quantum efficiency and are more robust . Only LaB6 has been seriously tested in an rf gun [6). Dispenser cathodes are now studied in a do gun at Orsay [261. Metallic cathodes have been extensively studied at BNL both in picosecond and femtosecond regimes [27,28] in a do configuration . Quantum efficiency as high as 7.25 x 10-4 has been measured for samarium photocathodes . A current of 21 kA/cm2 has been extracted from a 7 mm2 yttrium photocathode . This result corresponds to the "highest current density reported to date for a macroscopic photoemitter" . These photocathodes have now to be tested in an rf gun configuration . Several other photocathode studies are currently un- . . der way in many different laboratories . A good overview can be found in ref. [29) . An important issue for any type of photocathode is the dark current (current due to field emission) . This current depends on the photocathode material and the field level in the cavity At Los Alamos, for the operating conditions mentioned above, a current of 100 fC per rf cycle was measured [301 . In case of cw operation, this dark current could carry a significant power . The laser used to illuminate the photocathode should provide very short pulses, at the wavelength corresponding to the photocathode spectral response . Given the photocathode quantum yield, the laser macropulse should carry enough energy to extract the desired charge . The laser should be synchronized to the rf, i .e . the laser micropulse should correspond to a determined phase of the rf signal . This phase should be repeated from pulse to pulse with a timing jitter much less than the pulse length The laser micropulse should be as uniform as possible in both transverse and longitudinal directions . Uniform trains of micropulses are required at a repetition rate that varies from a few tens of MHz for FEL to a few GHz for linear colliders' driving beams . Over the last few years, lasers suitable for rf gun operation have been developed . A lot more work is required to improve some of the performances and bring them to a high level of reliability . The most critical issues are the uniform pulse shape, the phase and amplitude stability and the very high micropulse repetition rate especially when very short energetic pulses are required. The typical present performances are : micropulse length 10-20 ps, micropulse energy 10-100 pJ, repetition rate 20-100 MHz, amplitude jitter 1%, timing jitter 1-2 ps . Table 3 Comparison of rf gun vs conventional injectors Conventional injector Rf gun thermionic laser-driven High accelerating field - + + Current density Short pulse - - - - + + Pulse format flexibility - - + Cathode lifetime + + - Rf system simplicity - + + Compactness - - + Timmgjitter + + Energy fitter + + Development state + - Cost V. ACCELERATOR TECHNOLOGY
Page 1 and 2: Nuclear Instruments and Methods in
Page 3 and 4: C. Trainer / Rf guns - bright injec
Page 5 and 6: C Travier / Rf guns .- bright injec
Page 7: C. Travier / Rf guns : bright injec
Page 11 and 12: C. Traurer / Rf guns . bright injec

Rf guns: bright injectors for FEL - Triangle Universities Nuclear ...

Create successful ePaper yourself

Delete template?

Save as template?