CSR Shielding Experiment

It is well known that the emission of coherent synchrotron radiation (CSR) in a dipole magnets leads to increase in beam energy spread and emittance. At the Brookhaven National Laboratory Accelerator Test Facility (ATF) we study the suppression of CSR emission affect on electron beam in a dipole magnet by two vertically spaced conducting plates. The gap between the plates is controlled by four actuators and could be varied from 0 to 14 mm. Our experimental results show that closing the plates significantly reduces both the beam energy loss and CSRinduced beam energy spread. In this paper we present selected results of the experiment and compare then with rigorous analytical theory. INTRODUCTION There are a number of publications investigating the self-effect of CSR on a bunch energy spectrum. A comprehensive collection can be found in [1]. The possibility of using shielding plates to compensate this effect was suggested in [2]. Mean energy loss compensation was experimentally observed [3] and agrees well with predictions [4]. Authors are not aware of experimental results that report direct experimental observation of the shielding plates effect on the beam energy spread. Experimental observations and supporting theoretical estimates presented in this paper lead to different conclusion from previous work [5]: shielding plates can significantly reduce the energy spread increase induced on a bunch by CSR in a bending magnet. DESCRIPTION OF EXPERIMENT The experiment was carried out at Brookhaven Accelerator Test Facility. For optimized beam parameters, an energy spectrum change at the level of 10 keV was expected in a single pass through a dipole magnet. A beam energy stability at the level of δE/E~10 was needed to reliably characterize the effect. The high brightness electron beam was produced by a 1.6 cell photoinjector RF gun and then accelerated by a 6-m long S-band linac to 57.6MeV. Running the linac off crest induced a energy-time correlation on the beam. In the dispersive region of the transport line an energy selection collimator was used to select a ≈2ps long bunch by allowing a 200keV slice of the original beam to pass. Very stable beam conditions were achieved as a result: a slightly different part of the original beam in time was selected due to natural shot to shot energy jitter, yet the selected beam ended up with the same energy spectrum. The slow feedback system that stabilizes the beam energy at the 10 level and the RF phase at the 0.25 degree level was utilized. The method is very similar to the mask technique that was developed and actively used for different experiments at ATF and described in more details in [6]. Large ratio of horizontal dispersion to the beta functions at energy collimator location (~12.5 m in figure 1), low emittance and local energy spread allow for the generation of a flat top pulse with sharp rise and fall in times. The linear energy time correlation of the selected beam was utilized to measure the beam time by analyzing beam images on the energy spectrometer. Aluminum shielding plates were installed in the second dipole (s=20m on Fig. 1). The material polished to the submicron level of flatness was used to reduce influence of surface roughness wake-fields. A 0.4-meter long dipole magnet with a 20-degree bending angle was used for the experiment. The plates expand 0.15 cm on each side of the dipole to cover the edge field regions for a total length of 0.7 m. The beam was drastically shortened in the dispersive region of the beam line, after the first dipole. As a result, the CSR effect is much stronger in the second dipole and shielding plates were not installed in the first dipole. Figure 1: Optical function of the transport line from the linac exit to the spectrometer beam profile monitor. The energy collimator is located at position of 12.5 m and the magnet with shielding plates at the 20 m position. Beam Twiss parameters starting from the linac exit and to the energy spectrometer are shown in Figure 1. The focusing was design to have a zero dispersion function in the center of the magnet with the shielding plates in order to minimize the influence of transverse effects. Small vertical beta function in combination with low emittance were essential for loss-free transmission of the beam through the 0.7 m long shielding plates closed down to a 0.0 5.0 10.0 15.0 20.0 25.0