Design and cold model experiment of a continuous-wave deuteron radio-frequency quadrupole

A deuteron radio-frequency quadrupole (RFQ) is being built by the RFQ group at Peking University. It is a very compact high-current RFQ, operating at 162.5 MHz in continuous-wave mode. By optimizing the beam dynamics design, our simulations reached 98% transmission efficiency for acceleration of the 50-mA deuteron beam from 50 keV to 1 MeV, with an intervane voltage of 60 kV and a length of 1.809 m. This RFQ adopts a window-type structure, with low power consumption and sufficient mode separation, with no stabilizing rods required. Its magnetic coupling windows have been optimized by both electromagnetic simulation and the construction of an equivalent circuit model. The empirical equation based on the circuit model provides a new way to evaluate the effect of the window size on the frequency. In addition, an aluminum model of the full-length RFQ has been built and tested, and the results show good agreement with the simulations. During the tuning process, the magnetic coupling effect between quadrants was found to be unique to the window-type RFQ. We also propose a method to estimate the effects of different degrees of electric field unflatness on the beam transmission. For the cooling system design, the results of thermostructural analysis, verified by comparing results from ansys and cst, show that the special cooling channels provide a high cooling efficiency around the magnetic coupling windows. The maximal deformation of the structure was approximately $75\mu \mathrm{m}$. The beam-loading effect caused by a high current, and the coupler design, are also discussed.

Figure 1

Detailed 3D engineering model of the copper RFQ cavity made in two segments. The adjustable aluminum tuners will be replaced with fixed copper tuners after low-power tuning.

Figure 2

RFQ design parameters after optimization.

Figure 3

Beam envelope and emittance of the original and optimized designs.

Figure 4

Comparison of the transverse and longitudinal limiting current of the original and optimized designs.

Figure 5

Beam transmission along the RFQ. Plots from top to bottom are the beam profiles in the $x$ and $y$ planes and phase and energy spectrums, respectively.

Figure 6

The beam losses and energy change with a cell number in the original and optimized designs.

Figure 7

Mode frequencies and quality factor for the ${\mathrm{TE}}_{210}$ mode as a function of the four-vane RFQ length.

Figure 8

Parameters of the coupling window.

Figure 9

Evaluation of ${\mathrm{TE}}_{210}$, mode separation between ${\mathrm{TE}}_{210}$ and ${{}^0\mathrm{TE}}_{110}$, and power loss as a function of the depth for three width values. The stored energy of the RFQ cavity remains 1 J during the simulation.

Figure 10

Rectangular window and elliptical window.

Figure 11

Electric field comparison of two different window shapes.

Figure 12

The electric field along the RFQ before and after optimization.

Figure 13

On-axis longitudinal fields along the four-vane and window-type RFQs.

Figure 14

The surface current of a window-type RFQ in one quadrant.

Figure 15

Equivalent circuit of a window-type RFQ. (a) The whole circuit without equivalent inductance $L_1$. (b) The simplified circuit per section per quadrant.

Figure 16

The vanes (before processing the modulations), cavity walls, and assembly (without end plates) of the cold model.

Figure 17

The measurement of the cold model.

Figure 18

The left-hand side shows the magnetic field along the RFQ and the tuner positions. The right-hand side shows the frequency shift as a function of the inserted depth of no. 1 and no. 2 tuners.

Figure 19

Tuned electric field along the RFQ cold model for four quadrants. The simulated field is also included for comparison.

Figure 20

The electric field of the four quadrants along the RFQ with all the tuners of the third quadrant inserted 50 mm into the cavity. The black line in each graph is the electric field without tuners for comparison.

Figure 21

The left-hand side shows the coupling effect between the quadrants, and the right-hand side shows the magnetic field of the normal four-vane RFQ at the cavity end.

Figure 22

Output transverse and longitudinal emittance at different multiples of the field deviation.

Figure 23

Beam envelopes at (a) multiple 0 and (b) multiple 3.

Figure 24

Multiphysics scheme based on cst and ansys.

Figure 25

(a) Power density and (b) power loss for RFQ components, with the intervane voltage normalized to 60 kV.

Figure 26

The cooling channel design.

Figure 27

The simulated heat transfer coefficient distribution.

Figure 28

(a) Temperature map and (b) deformation map of the RFQ vanes.

Figure 29

Voltage phasors for a beam-loaded cavity.

Figure 30

The reflected power as a function of the input beam current with the optimum coupling factor of 2. The resonant frequency has been detuned to the optimal value for different values of the beam current.

Figure 31

The two coupler positions relative to the RFQ cavity.

Figure 32

The relationship between the coupling factor and the loop area.

Figure 33

The rf structure of the coupler with a matching port.

Figure 34

The change of the simulated $S_{11}$ parameter with length $B$.

Figure 35

Temperature distribution of the coupler without including the loop or any cooling.