Proceedings of EPAC 2002, Paris, France
DESIGN OF A 40 MEV LINEAR ACCELERATOR FOR PROTONS AND
DEUTERONS USING SUPERCONDUCTING HALF WAVE RESONATORS
M. Pekeler, K. Dunkel, C. Piel, H. Vogel, P. vom Stein
ACCEL Instruments, Bergisch Gladbach, Germany, [email protected]
Abstract
The design of a medium energy ion linear accelerator
(linac) for research applications is presented. The use of
superconducting resonators allows continuous wave (cw)
operation at high average currents. Independently phased
RF supplies for each resonator give the ability to
accelerate different particle species. The study gives an
overview of the accelerator set-up and its subsystems.
Parameter
Value
Unit
Particles
Protons/Deuterons
Energy
40
MeV
Current
4
mA
Operation
Continuously
Table 1: Design specifications
1 LAYOUT
The basic design specifications as given in table 1 will
be reached by a system consisting out of an ion source, a
RFQ and several modules equipped with super
conducting Half Wave Resonators (HWR). Starting at the
ion source the particles are pre accelerated in a RF
Quadrupole (RFQ) copper resonator up to an energy of
1.5 MeV/amu. The RFQ is capable of cw operation at an
RF power level of approx. 220 kW. This transition energy
of 1.5 MeV/amu between RFQ and the following
superconducting accelerating modules was optimised in
order to minimize the overall RF power consumption. The
main part of the linac (figure 1) consists of 12
superconducting accelerator modules (figure 2), which
accelerate the beam to its final energy of 40 MeV.
2 SUPERCONDUCTING MODULE
Each module contains four Half Wave Resonators
(HWR) with an operating frequency of 176 MHz. The RF
design is shown in figure 3. Two superconducting
solenoid lenses with an field strength of up to 6 T are
used for beam focusing. The resonators will be cooled by
saturated liquid helium at 4.5 K.
Figure 1: General layout
936
TTF
Proceedings of EPAC 2002, Paris, France
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Protons
Deuterons
8 x HWR 176 MHZ
L: 110 mm
40 x HWR 176 MHZ
L: 150 mm
E acc : 10.6 MV/m
E acc : 9.25 MV/m
B peak : 65.3 mT
B peak : 76.9 mT
P theo: 4.3 W
P theo: 5.9 W
0
5
10
15
20
25
30
35
40
45
Cavity No.
Figure 4: Transit Time Factors per cavity
The average energy gain per resonator is approx. 800
keV. The corresponding RF peak surface fields in the
resonator are 25 MV/m electric field and 60 to 80 mT
magnetic field. For an average beam current of 4 mA the
needed RF power per cavity is max. 4 kW.
Øout
Øin
The design is based on a peak electric field of 25
MV/m, the resulting energy gain per cavity is shown in
figure 5.
1.2
1
Delta E (MeV)
Figure 2: Layout of four HWR module
0.8
Protons
Deuterons
0.6
0.4
8 x HWR 176 MHZ
L: 110 mm
0.2
0
40 x HWR 176 MHZ
L: 150 mm
E acc : 10.6 MV/m
E acc : 9.25 MV/m
B peak : 65.3 mT
B peak : 76.9 mT
P theo: 4.3 W
P theo: 5.9 W
0
5
10
15
20
25
30
35
40
45
Cavity No.
Figure5: Energy gain per HWR
Lacc
Figure 6 shows that already two HWR families
optimised for different β are sufficient to reach 40 MeV.
The use of 352 MHz spoke type resonators can improve
the layout.
h
Lgap
E_kin (MeV)
50
45
Figure 3: HWR design for RF field simulation
3 DESIGN
The first phase of the system design started with a beam
dynamics analysis of different RF resonator types. A
comparison between Quarter Wave Resonators (QWR),
which are well established technology, and HWR showed
a clear advantage of the HWR type with respect to its
beam optical properties for acceleration of light ions [1].
Based on this resonator type a layout of the whole linac
was established.
First results determined by single particle tracking in
RF fields, as calculated by Microwave Studio, had been
used to optimise transit time factors (figure 4).
937
46.77 MeV
40.66 MeV
40
35
30
25
20
Protons
Deuterons
15
10
5
0
0
4
8
12 16 20 24 28 32 36 40 44 48
Cavity No.
Figure 6: Particle energy
First multi particle simulations based on the simulation
code TRACK [2] had been performed showing that an
energy of above 40 MeV can be reached for protons and
deuterons (figure 6 and 7).
Proceedings of EPAC 2002, Paris, France
BCS Losses [W]
700
600
500
400
300
200
100
0
5
10
15
20
25
30
35
40
45
E_peak [MV/m]
Figure 8: Cryogenic losses of linac with 48 cavities
according to BCS theory
Figure 6: Simulation results for Deuterons
5 CONCLUSIONS
Based on the results it was shown that a 40 MeV proton
and deuteron linac is feasible based on superconducting
technology with an injection energy from a normal
conducting RFQ as low as 1.5 MeV.
6 REFERENCES
[1] A. Facco et al, ”Study on Beam Steering in
Intermediate-β s.c. QWRs” , PAC’01, Chicago, 2001
[2] P. N. Ostroumov, K. W. Shepard, “Correction of
Beam-steering Effects in Low-Velocity Superconducting
Quarter-wave Cavities” Phys. Rev. ST Accel. Beams,
Vol. 4, 110101, 2001
[3] J.R. Delayen et al., “Application of RF superconductivity to high-brightness ion-beam accelerators”,
Nucl. Instr. & Meth. B56/57, p.1025-1028, 1991
Figure 7: Simulation results for protons
4 RESULTS
Within this work a general layout for the 40 MeV linac
based on sc HWRs was established. Based on RF
simulations the transit time factors and energy gain per
cavity were calculated, showing that particle energies
above 40 MeV can be achieved with a total number of 48
cavities. All calculations are based on a peak electric field
of 25 MV/m.
Cooling requirements at 4.5° K based on BCS theory
were calculated (figure 8). According to operation
experience of comparable systems [3] RF losses should
be at or below twice the BCS losses up to the operating
fields. With static losses of about 200 W the total
refrigerator power needed amounts to 800 W at 4.5 K.
938