Beam dynamics driven design of powerful energy recovery linac for experiments

Powerful ERL for experiments (PERLE) is a novel energy recovery linac (ERL) test facility [1], designed to validate choices for a 50 GeV ERL foreseen in the design of the Large Hadron Electron Collider and the Future Circular Collider and to host dedicated nuclear and particle physics experiments. Its main goal is to demonstrate the high current, continuous wave, multipass operation with superconducting cavities at 802 MHz. With very high beam power (10 MW), PERLE offers an opportunity for controllable study of every beam dynamic effect of interest in the next generation of ERLs and becomes a “stepping stone” between the present state-of-the-art 1 MW ERLs and the future 100 MW scale applications.

Figure 1

Top and side views of PERLE, featuring two parallel linacs each hosting an 82.2 MeV cryomodule, achieving 500 MeV in three passes.

Figure 2

Linac configured with the SPL cryomodule. The injection optics are tuneable by an initial quadrupole doublet (red).

Figure 3

Multipass linac optics. Red/green curves illustrate symmetrically optimized horizontal/vertical beta functions across different passes through the linac; red/blue arrows indicate the accelerating/decelerating passes.

Figure 4

The lowest and highest energy arcs (arc 1 and arc 6). Optics architecture based on the FMC cell. Horizontal (red curve) and vertical (green curve) beta-function amplitudes ($\beta$) are illustrated. Blue and black curves represent the horizontal and vertical dispersion ($\eta$), respectively. The arc as configured above is tuned to the isochronous condition.

Figure 5

Top and side views of the 250 MeV version of PERLE, featuring one cryomodule on one side and three straight sections on the opposite side, achieving about 250 MeV at the IPs.

Figure 6

The highest energy turn optics of the 250 MeV version of PERLE. Beta-function amplitude ($\beta$) and dispersion ($\eta$) are illustrated on the top and middle plots, respectively, in horizontal (red curves) and vertical (green curves) planes. Momentum compaction factor ($M_{56}$) is presented on the bottom plot. Dipole and quadrupole magnets are the blue and orange areas.

Figure 7

Layout of the PERLE injector from the electron gun to the injection into the main linac with the components labeled.

Figure 8

Evolution of the rms beam sizes along the injector. The location of elements is marked below the plot. The electron gun is marked in cyan, solenoids in yellow, normal conducting rf cavities in orange, SRF cavities in gray, quadrupoles in red, and dipoles in blue.

Figure 9

Evolution of the rms transverse emittance along the injector. The location of elements is marked below the plot. The electron gun is marked in cyan, solenoids in yellow, normal conducting rf cavities in orange, SRF cavities in gray, quadrupoles in red, and dipoles in blue.

Figure 10

Longitudinal phase space after the merger and before the main linac with projected particle position (horizontal) and energy (vertical) distributions.

Figure 11

rf beam load diagram of linacs 1 and 2 with phase choices during acceleration (black), deceleration (red), and resultant rf load (blue).

Figure 12

Longitudinal phase spaces at (a) the interaction region and (b) dump for (c) different combinations of linearizing $T_{566}$ in arcs 1 and 4, with a line indicating a continuous set of solutions available.

Figure 13

rf beam load plots for linacs 1 and 2 for a longitudinal match with a nonisochronous arc 1 showing phase choices during acceleration (black), deceleration (red), and 10 times the resultant (blue).

Figure 14

Longitudinal phase space toward the dump.

Figure 15

Longitudinal phase space at the dump including CSR for a different beam current and a longitudinal emittance of 25 keV. The histogram in red represents the longitudinal distribution. Left: Results with 10 mA. Center: Results with 20 mA Right: Results with 30 mA.

Figure 16

Longitudinal phase space at the dump including CSR for 20 mA beam current and different injection longitudinal emittance. The histogram in red represents the longitudinal distribution. Left: Results with 25 keV. Center: Results with 98 keV mm. Right: Results with 195 keV mm.

Figure 17

bmad simulations showing longitudinal phase space at the dump including CSR for 20 mA beam current and different bunch length. The histograms in red represents the longitudinal and energy distributions. Top left: Results with 0.5 mA. Top center: Results with 1.0 mA. Top right: Results with 1.4 mm. Bottom left: Results with 2.0 mm. Bottom right: Results with 3.0 mm.

Figure 18

Study of the HOM amplitudes as a function of the time. A new electron bunch is injected every 25 ns. Left: A beam current of 20 mA. Right: A beam current of 200 mA.

Figure 19

Bunch recombination pattern optimizing the distance between the newly injected bunch 1 and bunch 6 going to the dump (a) and the pattern obtained from the unoptimized arc turn-by-turn path length (b). The cavity voltage is presented as a function of the time. The bunch 1 represents the injection of a new bunch every 25 ns. The bunches 2–6 represent the position of the bunch after 1–5 turns, respectively. There is an alternation of accelerating and decelerating bunches.

Figure 20

Study of the HOM amplitudes as a function of the time. A new electron bunch is injected every 25 ns. Left: The arc lattice without the optimized bunch recombination pattern. Right: Optimized bunch recombination pattern.

Figure 21

PERLE BBU tracking model with eight cavities. (a) HOM voltages when ${\mathrm{Q}}_L$ is below (blue), at (red), and above (green) the critical ${\mathrm{Q}}_L$ value and (b) corresponding rms of beam centroid.

Figure 22

Critical ${\mathrm{Q}}_L$s for first 15 dominant HOM modes estimated in simulation and analytical model for $I_{\mathrm{op}}=120\mathrm{mA}$.