A Highly Flexible Bunch Compressor for the Aps Leutl Fel

The Low-Energy Undulator Test Line (LEUTL) freeelectron laser (FEL) [1] at the Advanced Photon Source (APS) has achieved gain at 530 nm with an electron beam current of about 100 A [2, 3]. In order to push to 120 nm and beyond, we have designed and are commissioning a bunch compressor using a four-dipole chicane at 100-210 MeV to increase the current to 600 A or more. To provide options for control of emittance growth due to coherent synchrotron radiation (CSR), the chicane has variable R56. The symmetry of the chicane is also variable via longitudinal motion of the final dipole, which is predicted to have an effect on emittance growth [4]. Following the chicane, a three-screen emittance measurement system should permit resolution of the difference in emittance growth between various chicane configurations. A vertical bending magnet analysis line will permit imaging of correlations between transverse and energy coordinates [5]. A companion paper discusses the physics design in detail [4]. 1 APS LINAC OVERVIEW The APS injector consists of a linac, an accumulator ring, and a 7-GeV booster synchrotron. In addition to delivering beam to the accumulator, the linac can be configured [6] to deliver beam to the LEUTL experiment hall [1]. The linac consists of 13 Stanford Linear Accelerator Center (SLAC) type accelerating sections powered by four klystrons, two thermionic rf guns (TRFG) [7, 8, 9] powered (one at a time) by a single klystron, and one photocathode gun (PCG) [10] powered by a single klystron. Figure 1 shows a schematic of the system and the location of the newly-installed bunch compressor. The original purpose of the linac was to create positron beams and deliver them to the accumulator ring for injection into the APS. The positron target was subsequently removed when the APS switched to electron operation. In both situations, the requirements on the linac were modest in terms of emittance, energy spread, bunch length, and stability. However, the requirements for reliability were and are very high, which was one reason for elimination of positron operation. The FEL project requires much higher beam quality and beam stability. The required beam quality is typically only achieved using a photocathode gun; however, the reliability of such guns (particularly the drive laser) is insufficient to act as an injector for the APS. The dual thermionic guns have a distinct advantage here, having proven themselves as components of the injector at SSRL Work supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. W-31-109-ENG-38. [11]. The use of alpha magnets [7] for magnetic bunch compression in these guns allows the guns to be placed off-axis, leaving the on-axis position for the PCG. This is an important consideration in preserving the PCG beam brightness. 2 MAGNETIC BUNCH COMPRESSION The principle of magnetic bunch compression is wellknown, so we only review the basic idea here. In a magnetic chicane (see Figure 1) the path length traveled by a particle is s = so+R56 , where so is the central path length and = (p po)=po is the fractional momentum deviation. For simple chicanes, R56 < 0 so that high-energy particles take a shorter path. Phasing the beam ahead of the crest in the precompressor linac introduces an “energy chirp” into the beam, so that the tail of the beam has higher energy than the head. As a result, the tail will catch up to the head in the chicane, giving a shorter bunch. If the beam is undercompressed, then the energy spread imparted in the precompressor linac can be removed by phasing behind the crest in the postcompressor linac. It is possible to derive formulae for the phasing required to obtain a desired bunch length and minimized energy spread. However, accurate calculation requires including wakefield effects and depends on the detailed initial bunch shape. Hence, we used simulation to find the optimal values [4]. 3 LEUTL BEAM REQUIREMENTS The primary goal of the bunch compressor is to provide higher current beam to the LEUTL FEL. A secondary goal is characterization of CSR effects. The bunch compressor was designed with two LEUTL operating points in mind. These operating points, distinguished primarily by the beam current of 300 or 600 A, are summarized in Table 1. The requirements for charge and emittance are not difficult compared to the state-of-the-art for photoinjector systems. We hope that these parameters can be achieved repeatably and easily to provide for routine and stable operation. Because of the very non-Gaussian longitudinal phasespace distributions one typically sees in the compressor, we use the following definition for the beam current: I80 = 0:8 Qtotal t80 where Qtotal is the total charge in the beam and t80 is the length in time of the central 80% of the beam. The value of 80% was used because this includes most of the particles but typically excludes high-current spikes that tend to occur at the head and tail. Also, when we refer to XX International Linac Conference, Monterey, California