Jean-Luc BIARROTTE

1
Proton driver for ADS applications

• Jean-Luc BIARROTTE CNRS / IPN Orsay

1
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

2
The PDS-XADS European project
2
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

3
Proton beam main specifications

• An ADS driver accelerator is a High Power Proton
  Accelerator (HPPA)
• XADS accelerator 3.6 MW beam power

• For the  small-scale  ADS (MYRRHA project)
  350 MeV, 5 mA
• For an industrial transmuter ? 1 GeV, 20 mA

3
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

4
Proton beam time structure

• A CW beam is favoured to avoid thermal stresses
  on the ADS target and sub-critical assembly (
  possible beam window)
• A pulsed beam mode (sharp short beam
  interruptions) is also needed to enable the
  in-line measurement of the sub-criticality level
• A CW RF operation is preferred for the
  accelerator (maximum reliability, lower RD
  effort)

A simple natural choice is CW accelerator
CW-based beam time structure
4
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

5
Superconducting LINAC Vs Cyclotron

• Superconducting linac
• No limitation in energy in intensity
• Highly modular and upgradeable
• Excellent potential for reliability
  (fault-tolerance)
• High efficiency (optimized operation cost)

• Cyclotron
• Attractive (construction) cost
• Required parameters at limits of feasibility
  ("dream machine")
• Compact, but therefore not modular

• In complete agreement with findings of the
  OCDE/NEA report
• Cyclotrons of the PSI type should be considered
  as the natural and cost-effective choice for
  preliminary low power experiments, where
  availability and reliability requirements are
  less stringent.
• CW linear accelerators must be chosen for
  demonstrators and full scale plants, because of
  their potentiality, once properly designed, in
  term of availability, reliability and power
  upgrading capability.

5
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

6
A classical High Power Superconducting Proton
LINAC
Example layout of the EURISOL proton linac
(Conceptual Design Report)
6
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

7
The low-energy section
High-intensity proton injectors are quite
straightforward ECR source RFQ
LEDA project at Los Alamos 100 mA CW, 6.7
MeV, 350 MHz in full operation (now stopped)
TRASCO project in Italy (INFN) 30
mA CW, 5 MeV, 352 MHz source in
operation, RFQ under fabrication
IPHI project in France (CEA / CNRS) 100 mA
CW, 3 MeV, 352 MHz source in operation,
RFQ under fabrication
7
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

8
The high-energy section (1)

• General agreement for using SC multi-cell
  elliptical cavities _at_700 MHz
• High performance (high gradients, high
  efficiency, security, reliability, modularity)
• Well-established solution (TTF, SNS)
• Comfortable margins can be chosen on critical
  values to ensure a design as robust as possible
  it consists in limiting, in a reasonable way, the
  minimum beam apertures, the fields in the
  cavities, the phase advances along the linac, the
  sensibility to beam mismatch, or the possibility
  of halo creation.

Layout of the XADS high-energy section
?0.47 (90-192 MeV), ?0.65 (192-498 MeV)
?0.85 (498 MeV 600 MeV or 1 GeV)
(14 modules / 59 m) (17 modules / 99
m) (3 or 12 modules / 26
or 101 m)
8
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

9
The high-energy section (2)
Active RD and prototyping are going on
successfully at CEA Saclay, CNRS / IPN Orsay
INFN Milano
Test results _at_2K of 5-cell ?0.5 prototypes (INFN
Milano)
A 5-cell ?0.65 prototype (CEA Saclay / IPN
Orsay)
9
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

10
The intermediate section (1)

• Two concepts have been retained
• Extend the injector philosophy towards high
  energies using DTL-type structures
• Extend the high-energy SC linac philosophy
  towards low energies using low-beta
  superconducting resonators

• The SC cavities solution as compared to a
  room-temperature DTL solution
• Roughly same length construction cost
• Excellent RF to beam efficiency
• gt Significant operation cost savings (7 MW AC,
  i.e. 2 M/year for a 6 mA beam)
• Very large beam aperture gt High safety (less
  structure activation)
• Independant RF structures gt High flexibility
  (power adjustments)
• gt High reliability (low power RF
  components, fault tolerance capability)
• - Poor real estate gradient at very low energies
  (lt 20 MeV)

10
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

11
The intermediate section (2)
? The independently-phased SC option is agreed
from X MeV (X still to be defined between 5 and
50 MeV), especially because of it allows to
implement the fault-tolerance concept (see
later) ? 352 MHz Spoke cavities, developed by
CNRS / IPN Orsay, are used in this region (high
shunt impedance, good mechanical stability
tunability, no steering effect, possibility to
design multi-gap structures if needed, excellent
test results)
11
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

12
The intermediate section (3)
? Between 5 MeV and 20 MeV, the ?0.15 spoke
solution is not so efficient in terms of real
estate gradient DTL-type solutions are also
explored - Superconducting CH-DTL structure,
developed by IAP Frankfurt - Room-temperature
IH-DTL structure, developed by IBA
Prototyping on SC CH-DTL IH-DTL
12
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

13
The final beam transport line

• Doubly achromatic beam line concept
• (non-dispersive optical system energy
  monitoring)
• Beam scanning method to paint the target
• (very reliable used for protontherapy)
• Meets the specifications of the 3 different XADS
• (Gas-cooled XADS, LBE-cooled XADS, MYRRHA)

13
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

14
The XADS reliability analysis (1)

• Main conclusions of the reliability analysis
  performed within the XADS project
• ? The cyclotron option for PDS-XADS does not seem
  to offer a sufficient perspective of reaching the
  requested reliability level
• No showstopper to reach high availability high
  reliability with the XADS reference linac if
  OVER DESIGN REDUNDANCY are used
• ? FAULT TOLERANCE has been identified as a key
  element in order to guarantee reliability by
  design and operation (as an example, several tens
  of RF failure are foreseen per year )

14
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

15
The XADS reliability analysis (2)
15
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

16
The XADS accelerator, optimized for reliability
16
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

17
Fault tolerance calculations for XADS (1)
Fault tolerance ability to loose a cavity or a
Q-pole on line without loosing the beam
? A systematic study has been performed to
analyse the fault tolerance capability of the
XADS SC linac (5 600 MeV)
? The local compensation method has been used
17
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

18
Fault tolerance calculations for XADS (2)
? The conclusion of the study is that in every
case, with an appropriate retuning, the beam can
be transported up to the high energy end without
any beam loss (100 transmission, small
emittance growth), and within the nominal target
parameters. ? An adequate fast LLRF control
system is required margins of 30 are needed on
the Eacc operating point of the cavities ? The
transient states of the recovery procedure have
still to be carefully studied and optimized.
Simulation in the case where spoke cavity 4 is
lost (5.5 MeV), and proper retuning is performed
Sample of the retuning values for each cavity
failure
18
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

19
Feedback systems Maintenance
19
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

20
Accelerator shielding design
? Conservative beam losses assumption - RFQ 3 -
Intermediate energy 10 nA/m - High energy 1
nA/m ? ALARA shielding design criteria - lt 1
mSv/year (ICRP-60 recommendations) - I.e. dose
rate lt 0.5 mSv/h (2000h/year) - Occupancy factor
1 ? Tunnel design for the XADS linac -
Shielding calculations _at_ 600 MeV - Chicane access
design - Shielding for normal losses also copes
with accidental losses
Required concrete/earth combined thickness at 600
MeV
20
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

21
The PDS-XADS roadmap
21
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

22
A few PRELIMINARY figures
22
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

23
RD for the FP6 (2004-2008) IP EUROTRANS
23
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.

24
Conclusion
IP EUROTRANS begins on April 1st, 2005 DM1
(Design) / WP1.3 (Accelerator) is coordinated by
CNRS/IN2P3, and aims to the demonstration of the
ADS linac technology This RD program is
highly reliability-oriented (the practical
implementation of the fault-tolerance concept,
which is the major key point to achieve the high
reliability requirement, has especially to be
demonstrated) has a lot of synergies with other
applications based on a HP proton linac
(EURISOL, neutrino factories) should finally
lead, in 2008, to a frozen detailed design of
the XADS / MYRRHA linac, with assessed
reliability and costing
24
J-L. BIARROTTE, 16-19 March 2005 BENE week, CERN.