Experimental Demonstration of Energy-Chirp Control in Relativistic Electron Bunches Using a Corrugated Pipe

The first experimental study is presented of a corrugated wall device that uses wakefields to remove a linear energy correlation in a relativistic electron beam (a “dechirper”). Time-resolved measurements of both longitudinal and transverse wakefields of the device are presented and compared with simulations. This study demonstrates the feasibility to employ a dechirper for precise control of the beam phase space in the next generation of free-electron-lasers.

Figure 1

Geometry of the vacuum chamber with corrugated walls. A rectangular coordinate system is centered on the symmetry axis of the chamber. The blue ellipse represents an electron beam propagating along the $z$ axis.

Figure 2

Wake of the experimental structure with $g=6\mathrm{mm}$, as calculated with the analytical model (dashed lines) and by using the mode-matching numerical method (solid lines): (a) the longitudinal, $w_z$, (b) the dipole, $w_d$, (blue) and quadrupole, $w_q$, (red) wake functions.

Figure 3

Layout of the PAL-ITF wakefield experiment.

Figure 4

Measured (black) and simulated (red) temporal profiles as seen on screen 6, compared to the fully resolved simulated profile at the dechirper (green).

Figure 5

Screen-6 images (top-row: measured, middle-row: measured and corrected, bottom-row: simulated and corrected) taken with rf deflector switched on and a net accelerating rf phase near crest. The left column of images has the gap open, and the right has the gap closed. The simulation uses 150 pC, rather than 200 pC, for better agreement with measurements.

Figure 6

Average relative energy loss over the bunch vs gap of the corrugated chamber. Error bars represent the rms variation (divided by $\sqrt{N-1}$) over $N=4$ separate measurements spanning two days. The solid curve is a simulation using 150 pC and the measured FWHM bunch length of 4.8 ps.

Figure 7

Screen 6 images (top-row: measured, bottom-row: simulated, $Q=150\mathrm{pC}$), with $g=6\mathrm{mm}$, rf deflector switched off and L0a phase adjusted 10 degrees off crest, showing no dipole kick when gap is centered on the beam (left), but a strong tail kick with gap off axis by 1 mm (right). The bunch head is at left in each image.

Figure 8

Measured (red) average vertical beam position on screen 5 vs vertical offset of the gap ($g=6\mathrm{mm}$) against Elegant simulations (dashed line) based on a linear approximation of the transverse wake and a calculation (solid line) including the wake nonlinearities for a parabolic bunch ($Q=150\mathrm{pC}$ in both simulation cases).

Figure 9

Screen 5 measured images with the rf deflector switched on, showing no quadrupole wake when the gap is open (left), but a strong time-dependent horizontal focusing with the gap closed to $g=6\mathrm{mm}$ (right). The bunch head is at the top.