back to results for ""
Below is a cache of http://www.slac.stanford.edu/econf/C000821/MOD04.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
A Charge State Breeder Based on a Low Frequency RFQ for the ISAC Accelerator Complex A CHARGE STATE BREEDER BASED ON A LOW FREQUENCY RFQ
FOR THE ISAC ACCELERATOR COMPLEX
P. Bricault, TRIUMF, Canada
Abstract
The new ISAC-II facility has just been funded. The
main goals of this extension to the ISAC facility are:
expanding the mass range from A 30 to A 150 and
increasing the final energy from 1.5 MeV/nucleon to 6.5
MeV/nucleon. The accepted mass to charge ratio for the
actual 36 MHz RFQ is A/q 30/1. In order to increase
the mass range to 150 we have to increase the charge-state
to at least 5+. This can be achieved by three different
methods: i) using a charge state breeder based on an EBIS
as developed at REX-ISOLDE [1], ii) using an ECRIS or
ECRIT based on an ECR as developed at Grenoble [2, 3],
iii) using a gas charge-state stripper based on a low
frequency RFQ [4]. Since intense RIB are difficult to
produce it is of primary importance to obtain the highest
overall charge-state breeding efficiency and to preserve the
beam quality, small emittance and beam purity. In
comparing the three above methods, our analysis shows
that the combination of a low frequency RFQ and a gas
stripper offers the best overall performance.
1 INTRODUCTION
The TRIUMFs ISAC uses the isotope separation on
line (ISOL) technique to produce radioactive ion beams
(RIB). The ISOL system consists of a primary production
beam, a target/ion source, a mass separator, and a
separated beam transport system. These systems together
act as the source of radioactive ion beams to be provided
to the accelerator or the low-energy experimental areas.
We use the 500 MeV - 100µA primary proton beam
extracted from the H- cyclotron [1]. The accelerator
complex comprises an RFQ [4] to accelerate beams of q/A 1/30 from 2 keV/u to 150 keV/u and a LINAC (DTL)
to accelerate ions of q/A 1/6 to a final energy between
0.15 MeV/u to 1.5 MeV/u. In April 2000 the ISACII
proposal was funded. the funding agency. The main goals
of this extension to the ISAC facility are: expanding the
mass range from A 30 to A 150 and increasing the
final energy from 1.5 MeV/nucleon to 6.5 MeV/nucleon.
The accepted mass to charge ratio for the actual 36 MHz
RFQ is A/q 30/1. In order to increase the mass range to
150 we have to increase the charge-state to at least 5+.
This paper reviewed the different schemes that can
allow the production of higher charge-state.
With the PIAFE proposal early 1993-1995 [5,6,7] and,
followed with the REX-ISOLDE [8] proposal the idea of
the charge state breeder make it way in the accelerator
design scheme. Now, every new proposal or upgrade
looks seriously into the use of a charge-state breeder. The
reason being, the higher the charge-state of an ion, the
higher will be the acceleration efficiency. In consequence,
a beam accelerated to several MeV/nucleon will require
less voltage and reduce considerably the size and the cost
of the accelerator.
2
CHARGE-STATE BOOSTER FOR RIB
FACILITY
The principle of the 1+ n+ consists in injecting an
ion beam into an other ion source. The charge-state
breeding ion source can be either an Electron Beam Ion
Source (EBIS) or and Electron Cyclotron Resonance Ion
Source (ECRIS). Since it takes some time to obtain the
desired high charge-state, the challenge is to be able to
capture with high efficiency the mono-charged beam into
the breeding ion source.
2.1 TRAP-EBIS combination
In the REX-ISOLDE scheme, the charge breeding will
be done by an EBIS. Since the acceptance of this ion
source is very small, 3 mm mrad at 60 kV, they used
a Penning trap for accumulation, cooling, and bunching.
However, the number of particles that can be trapped in
the Penning trap is limited. The maximum ion density for
the REX-Penning trap is estimated for A = 140 at 10
6
-10
8
ions per accumulation cycle.
The Penning trap is located on a high voltage-platform.
The platform is at the same potential as the singly
charged ion source at the ISOLDE target. A transfer beam
line consists of two electrostatic benders and two
electrostatic quadrupole doublets. The confinement time to
reach the charge-to-mass ratio larger than 1/4.5 is less
than 20 ms. The EBIS magnet has a magnetic field of 2 T
with an homogeneity of about 2.5 along the
confinement length of 0.8 m. The ions ejected from the
EBIS are mass analyzed with a magnetic achromat
composed of two 90º dipoles.
This scheme is very complex. It requires a lot of beam
manipulation and tuning can be very long and difficult. It
is not ideal for a facility, which has to serve many users.
Furthermore, it is not suitable for high intensity because
the Penning trap is space charge limited.
2.2 Charge-state breeder based on ECRIS
Several tests have been done over the past five years to
developed the charge breeding using an ECR[9,10,11].
Recently at Grenoble [12], they tested many different
elements, noble gases, alkali, and metallic species. The
charge-state breeding efficiency is define as
XX International Linac Conference, Monterey, California
211
MOD04 charge-state breeding efficiency is define as and the global
efficiency is defined as,
Adjustment
voltage at
injection
Ion source
Voltage
(V
IS
= V
ext
)
Mass Separator
Charge State Booster
(V
CSB
= V
ext
)
RFQ, 2keV/nucleon
E
K
= 2 keV*A=q * V
ext
E
K
= 300 keV,
If we want to avoid the
stripper, q/A
II
>1/7
Implies q=22
V
ext
= 13.6 kV !!!
Potential diagram for the combined ion source
and charge state booster, A=150
Figure 1: Potential diagram for an ECRIS as charge-state
breeder.
For the noble gases, the global ionization efficiency is
between 40% to 50%. This efficiency drops significantly
for other elements. These results can be explained by the
fact that noble gases do not stick to the surface walls of
the ECRIS very long. The ions, which are not directly
capture by the plasma, are neutralized and the atoms are
release after a short period and then be ionized to higher
charge-state. The charge breeding efficiency is of the order
of 10%.
For species with a large enthalpy potential, the sticking
time on the walls can be much longer. In some cases the
atoms can react with the surface material and make stable
compound. The global efficiency is then much lower and
the charge breeding efficiency is between 2 and 3%.
We can see several problems arising from those results.
The global efficiency being less than 100% means that
radioactive species will stay into the ion source. It will
become a major source a radioactivity. We have to be
concerned for the services and maintenance of the ECRIS.
Furthermore, the decay of the radioactive species will
create contamination of the desired beam. This can be a
big issue since we are developing a resonant LASER ion
source to produce pure radioactive ion beams.
Furthermore, in order to inject at 2 keV/nucleon into
our RFQ the extraction voltage of the 1+ RIB has to be
adapted accordingly. The for ISACII has to be greater to
q/A 1/7 in order to avoid any further stripping. This
means that we will have to extract the 1+ RIB at very low
voltage. It will result a reduction of the transmission by a
factor 2 in the mass separator section. Finally, the charge
breeding efficiency being very low will somehow reduce
the physics opportunities. Experiments will take longer
and in some other cases, simply impossible.
3 RFQ-STRIPPER COMBINATION
The other solution we envisage to increase the mass
range to A 150 is to use a gas stripper at low energy.
Two different scenarios can be used, a 12 MHz RFQ, gas
stripper and a new 36 MHz RFQ or a 12 MHz RFQ, gas
stripper and deceleration to 2 keV/nucleon and injection
into the actual RFQ. In both cases, the residual intensity
will be the same. Figure 2 shows schematic diagram of
the two options.
Figure3 shows the relative equilibrium charge-state as a
function of the reduced velocity for ions from
32
S to
181
Ta.
In our case, the necessary energy will be between 16 and
20 keV/nucleon.
To preserve beam intensity the RFQ has to operate in
CW mode. This design uses an 12 MHz RFQ similar to
the one developed at ANL [4]. To maintain an input
energy of 2 keV/nucleon at the entrance of the existing
ISAC 36 MHz split ring RFQ [13] the new RFQ and the
gas stripper would have to be installed on a 300 kV high
voltage platform.
The ANL design will not provide the necessary energy.
It was shown by the INS-Tokyo group that it