Electron Beam Dynamics in the 50 Mev Thomx Compact Storage Ring

ThomX is a high flux compact X ray source based on Compton back scattering between a relativistic electron beam and an intense laser pulse. To increase the repetition rate, the electron beam is stored in a 50 MeV ring. The main drawback of such a scheme is the low energy of the electrons regarding collective effects and intrabeam scattering. These effects tend to spread or even disrupt the stored bunch and they limit its charge, especially in low energy rings machine where damping plays a negligible role. Thus such collective effects reduce the maximum Xray flux and it is important to investigate them to predict the performance of this type of X-ray source. In addition, Compton back scattering acts on the electron beam by increasing its energy spread. This presentation will show firstly the impact of collective effects on the electron beam, essentially during the first few thousand turns when they are the most harmful. Then, the reduction of the Xray flux due to Compton back scattering and intrabeam scattering will be investigated on a longer time scale. INTRODUCTION ThomX is based on a 50 MeV storage ring operating in a pulsed mode at a target current of 20 mA [1]. The electron bunch charge considered is then 1 nC. The synchrotron equilibrium is not reached. Because of the low beam energy, the degradation of the electron bunch is not slowed down by damping. Accordingly, we have to take care of all sources of electron bunch degradation. The Compton interaction rate is determined by the stored electron bunch characteristics which follow themselves from the linac performances. The bunch is ejected when the scattered radiation characteristics are no more suitable for users. This paper will present first the collective effects involved mainly during the first turns and the IntraBeam Scattering (IBS) and Compton Back Scattering (CBS) effects that take place on a longer time scale. In the last part of this paper, the flux reduction evaluation and the spectrum distribution evolution will be dealt with. COLLECTIVE EFFECTS The ThomX electron bunch is very sensitive to wakefield effects mainly due to its low energy, its short length and the fact that there is no damping during the storage time. There are various sources of wakefields: Beam pipe geometry (bellows, pump ports, RF taper, RF HOM etc ...), resistive wall effect, Coherent Synchrotron Radiation (CSR enhanced by the short bend radius). In addition the injected bunch from the linac suffers from a strong longitudinal phase space mismatch. 2D and 6D tracking simulation codes including these effects have been developed. The injected bunch is short, only 4 ps rms long, and exhibits the standard curved banana shape in the longitudinal phase space resulting from the 3 GHz linac acceleration. During the first turns in the ring, the bunch phase space profile will undergo a complex dynamics under the effect of the collective effects as shown in figure 1. This turbulent process will progressively lead to the matching of the ring longitudinal phase space over the first thousand turns. The bunch length will reach about 25 ps rms while the energy spread will remain unchanged. In this process, the CSR is the dominant collective effect. The important point is the need for longitudinal and transverse feedbacks. Their design is under way to cope with any source of bunch oscillation and to provide a center of mass damping. To summarize, these sources are for each plane: ! Injection miss steering: Position and time/energy jitter from the linac and from the transfer line, ! Transition time: First ~1000 turns when strong collective effects occur, enhanced by injection mismatch between the linac and the ring, ! Storage time : Impedance, HOM and ions. Figure 1: Longitudinal phase space evolution during the first few thousand turns after injection from the linac. Charge 1 nC. Simulation realised with a 2D tracking code including the effects of the longitudinal feedback, the CSR, the space charge and the resisting wall. Proceedings of IPAC2011, San Sebastián, Spain MOPS050 05 Beam Dynamics and Electromagnetic Fields D05 Instabilities Processes, Impedances, Countermeasures 715 C op yr ig ht c © 20 11 by IP AC ’1 1/ EP SAG — cc C re at iv e C om m on sA ttr ib ut io n 3. 0 (C C BY 3. 0) In the transverse plane, the most critical instabilities are supposed to be the head-tail and resistive-wall arising from the vacuum chamber impedance, as well as those due to ions. The resistive-wall instability threshold estimated in frequency domain with 28 mm aperture stainless steel chambers turns out to be as low as ~5 mA due to the long radiation damping time. The growth rate at 20 mA is nearly 1 ms with little dependence on the number of bunches. MOSES calculations were carried out with a broadband resonator impedance scaled from that modelled for SOLEIL. While the TMCI (transverse mode coupling instability) threshold turned out to be as high as 90 mA, that of the head-tail was low for the same reason as the resistive-wall, corresponding to ~0.16 ms growth time at 20 mA. Regarding the ion instabilities, the critical mass for the 2-bunch case is less than 1 implying the possibility of trapping all species. The asymptotic growth rate of the fast beam-ion instability for 40 mA consisting of 2 bunches is ~0.1 ms, as deduced from the linear model of T. Raubenheimer and F. Zimmermann [2]. Multibunch tracking is being performed to pursue these instabilities in more detail. As regards transverse feedback, a simple analogue system damping the few most dangerous coupled-bunch modes may be better suited for ThomX in view of the 1 or 2 bunch operation envisaged. The optimisation is underway.