1. Wakefield and impedance studies with ultra-relativistic beams on VELA/CLARA and UK-FEL
- Author
-
Sullivan, Michael, Jones, Roger, and Bertsche, William
- Subjects
Electron Beam ,X-band ,XARA ,Free Electron Laser ,Emittance ,Relativistic ,RF ,FEL ,Linac ,Accelerator ,Daresbury ,CLARA ,Impedance ,Wakefieds ,Beam Dynamics - Abstract
The effects of wakefields are a universally challenging topic in accelerator physics and are critically important due to their impact on beam quality. Wakefields can dilute the beam emittance and in the worst case can cause a beam break up (BBU) instability. Several types of wakefields exist, each with their own separate sources and influences on the beam. The wakefields considered in this Thesis are geometric, where the electric fields excited by the charged particles diffract off the accelerator walls at a transition in the beam pipe and travel back to the beam. These can be further split into longitudinal wakefields which alter the energy spread and length of the bunch, and transverse wakefields which transversely kick the beam and increase the transverse emittance, reducing the beam quality. Long-range wakefields span from one particle bunch to a trailing bunch, whereas short-range wakes effect the bunch that excites the wakefield. This Thesis presents a framework which includes the cell and structure design of a linac and calculation of the wakefields excited in disc-loaded linac structures. In this framework, I have produced analytical expressions of the long-range wakefields, and a method to fit numerical short-range wakefields to analytical formula. These procedures would be advantageous to the accelerator community for use in general linac design. This framework has been used to calculate the wakefields of the Compact Linear Accelerator for Research and Applications (CLARA) S-band linac structure as well as the X-band Accelerator for Research and Applications (XARA) structure, a linac designed to fulfil the aims of the CLARA project to be a Free Electron Laser (FEL) test facility to study novel acceleration techniques. The XARA project serves as an alternative or an upgrade to the CLARA phase 2 system, upgrading the beam energy from 250 MeV to ~ 1 GeV. This will reduce the achievable wavelength of the FEL setup by a factor of ~ 16, allowing many biological and ultra-fast dynamics processes to be studied, which is particularly attractive to the accelerator community. This thesis introduces a design of an X-band linac for the XARA project capable of an average gradient of = 80 MeV/m, much higher than the CLARA S-band linac gradient of = 25 MeV/m (operated at 17.5 MeV/m). This gradient was chosen to ensure a beam momentum of greater than 70 MeV/m could be reached with only three RF modules due to spatial constraints. I have created a numerical tool which optimises the structural parameters and iris tapering to increase its efficiency while avoiding breakdown. This is combined with a study of the long-range and short-range wakefields to decide the parameters of the initial design. The calculations of the wakefields agree well with results from established wakefield codes ABCI and ECHO2D, while reducing the required CPU time by a factor of ~ 20-30. This allows the short-range wakefields to be calculated rapidly if design changes need to be made. The short-range wakefields, which I have calculated, are used in the beam dynamics code ELEGANT to study the electron beam in the XARA structure. The variation of the initial offset and charge of the bunch gives a preliminary indication of how sensitive the beam quality is to misalignments and how much charge can be placed in a bunch. The normalised emittance growth for a bunch charge of Q_b = 250 pC for a transverse offset of C_x = 0.253 mm was found to be Δ∈Nx/∈Nx = 270.8%. These simulations were carried out without a focusing magnetic lattice which would recenter the beam and alleviate the wakefields, however this could be done in a future study which will include a full design of a FODO lattice. These calculations are then validated by an analytical formulation where the emittance growth at the end of the linac agrees within ~ 12% between the two methods. There are a suite of misalignment studies needed to fully quantify systematic and random errors likely to be present in the XARA/UK-XFEL machine. My work serves as an important initial study on the linac design towards a fully realised XFEL facility at Daresbury Labs. This facility would be available to the accelerator community to perform a wide range of studies as well as test novel FEL schemes.
- Published
- 2021