European Studies for Nuclear Waste Transmutation

1
European Studies for Nuclear Waste Transmutation

• Paolo Pierini
• INFN Milano LASA

Work supported by the EURATOM 5 framework
program of the EC, under contract
FIKW-CT-2001-00179
2
Outline

• Introduction / Transmutation in an ADS
• ADS systems and accelerator requirements
• With emphasis on a rather unusual
  specification
• Accelerator reliability
• European ADS activities
• National programs on ADS
• The European Roadmap by ETWG (2001)
• PDS-XADS in the 5 Framework Program
• WP3 entirely dedicated to accelerator design
• Accelerator activities for the 6 FP (EUROTRANS)
• Synergies with HPPAs / Conclusion

3
Transmutation in an ADS
4
Transmutation in an ADS (made simple)

• Problem Disposal of Nuclear Waste
• Reduce the radiotoxicity of the waste
• Minimize volume and heat load of waste sent to
  deep underground repository
• EU 145 reactors, 125 GWe, 850 TWh yearly
  production (35)
• Yearly production of 2500 tons of spent fuel in
  EU (25 tons of Pu!)
• Strategy Partitioning and Transmutation
• Separate chemically the waste (separe Pu, MA,
  LLFF)
• Use the waste as fuel in dedicated transmuter
  systems
• Solution 2 ingredients are needed for a
  transmuter
• A subcritical reactor (klt1), with U-free fuel
• The chain reaction is not self-sustained
• An intense spallation source (high proton flux on
  liquid lead target)
• Provides the missing neutrons to keep the
  reaction going, with a broad energy spectrum
  (good for MA burning)

5
PT Goal
From ETWG Report, 2001

• Radiotoxicity of the spent fuel decreases to that
  of the starting raw material (uranium ore) only
  after period greater than 106 years
• In order not to contaminate the biosphere it is
  necessary to dispose of them in deep stable
  geological repositories
• Chemical separation (Partitioning) and
  irradiation in a fast and intense neutron flux
  with ADS systems (Transmutation) can reduce this
  time to 700-1000 years

6
ADS systems and accelerator requirements
7
Beam specifications for an ADS accelerator

• Very high duty cycle, possibly CW
• Efficiency
• Energy of the order of 1 GeV, determined by
• Neutron production rate per GeV and per
  proton(optimum value reached at 1 GeV)
• Energy dissipated in the input window(rapidly
  decreasing with energy, when E lt few GeV)
• But, target can be windowless!
• Beam power from several MW up to tens of MW
• few MW for a eXperimental ADS of up to 100
  MWth (XADS)
• 20 MW for an industrial burner of several 100s
  MWth (ETD)
• Design of a liquid metal spallation target for
  these power levels!

8
Reliability requirements

• A Nuclear Power Plant has a hard limit on the
  number of thermal cycles during its lifetime,
  mainly due to thermal fatigue in the vessel and
  main instrumentation
• After reaching that, it cannot be operated
  anymore
• Operation pattern dominated by the fuel cycle
• 3 months uninterrupted, 1 month maintenance
• Limit on beam trips
• Unexpected shutdowns (long beam interruption gtgt
  1 s) should not exceed a few per year
• in a conventional NPP is 1-2/year
• The allowable number of intermediate (0.1 s - 1
  s) beam trips depends on technological details of
  target and reactor (mainly fuel elements and
  cladding, etc.)
• Estimated of the order of hundreds/thousands per
  year
• Shorter beam trips (ltlt 1 s) can be tolerated due
  to the big thermal inertia
• Many trips experiences in current facilities
  could be moved to this time scale with modern
  controls based on fast electronics

9
Where do we stand with existing accelerators?

• Beam Specifications
• No particular concerns
• Average beam power of few (3.6) MW for XADS
• LAMPF/LANSCE currently above 1 MW
• SNS goal is 1.4 MW, with potential for upgrades
• Beam current limited to 20 mA also for industrial
  prototype
• Peak current at SNS and JPARC is higher
• Still, 4-20 MW represent a technological
  challenge for the target
• Reliability Specifications
• Existing accelerator facilities are many orders
  of magniture off
• Availability issues have been addressed recently
• By synchrotron light sources (user request)
• Medical accelerators
• Linear collider studies (in terms of expected
  luminosity)
• But no concern regarding number of beam trips
  until now

10
Reliability Engineering

• Reliability engineering is a technical discipline
  for
• estimating,
• controlling and
• managing
• the probability of failures in complex systems,
  and has been applied in many industrial fields
  (aerospace, defense, automotive, electronics, )
• For most systems, due to the technical complexity
  of the design, it is not enough to specify and
  allocate the reliability of components in order
  to predict accurately the reliability of the
  system
• Formal mathematical and statistical methods can
  be applied to measure and assess reliability
  characteristics of components, but the associated
  uncertainties are high, leading to reliability
  estimates with limited credibility

11
Nature of connections is important

• Not only the component specifications (MTBF
  MTTR) are important for the reliability
  assessment of the system
• The logical or functional connection between
  components plays a major role in reliability
  mathematics
• Series connection
• Parallel connection
• Hot, warm and cold redundancy
• k out of n redundancy
• Different maintenance provisions capabilities,
  i.e. both repairable and non-repairable systems
  during the mission time
• E.g. 2-tunnel accelerator scheme (main linac
  service tunnel)
• Role of common cause failures
• Reliability is a property of the system
  configuration, it is not uniquely defined by the
  components specifications
• In other words, proper planning of redundancies
  allows building reliable systems out of
  moderately reliable components

12
Accelerator Reliability
13
Existing reliability data for accelerators

• Existing laboratories have huge data sets with
  many years of operating performances of
  components which can be a very useful source of
  information
• But, differently from the case of NPPs, there is
  no standardised way of identifying, collecting
  and analyzing performances, failures and
  maintenance data of accelerator component
• Every accelerator laboratory has its own format
  and procedures for the log keeping of its
  machines
• Very hard and time consuming to assemble a
  database of common accelerator components with
  these data sources
• And then, very few components are standard, and
  it is very difficult to validate, or benchmark
  the collected data
• Little experience with reliability practices in
  accelerator community
• Always aiming at the next and more performing
  machine rather than building a more robust
  version with the same performances...

14
MTBF/MTTR Considerations

• MTBF mean time between random statistical
  failures
• Highly predictable failures not in MTBF
• Components wear out
• Bad design issues or severe load/strength
  interference (very common)
• Aging (if we perform a constant failure rate
  analysis)
• MTTR mean time to repair i.e. provide back
  function to system
• Time to detect and identify the fault
• Time to prepare the system for access to the
  component
• Time to repair the component itself
• Time to prepare the system back for operation
• Time to restore nominal conditions for the beam
  at target
• bare component MTTR can be of no influence to the
  time to recover from a system fault
• In logbooks often part of the restoring time
  allocated to machine development runs, thus not
  correctly assigned as an effect of a fault

15
R/A oriented Design Criteria

• Fortunately a few reliability-oriented design
  criteria exists and have been followed early in
  the design stage of ADS accelerators
• Derating/Overdesign (and Load/Strength
  considerations)
• Stay well away from technological limitations in
  components operation (max applied load well below
  min component strength)
• Handles batch variation of components
• Ensures that marginal devices do not cause system
  failures
• Redundancy (parallelism)
• Provide the same function with several components
• Different strategies can be followed for standby
  redundancy
• Hot (failure rate standby failure rate
  operating)
• Warm (failure rate standby lt failure rate
  operating)
• Cold (failure rate standby 0)
• Fault Tolerance
• Failure of the component is not necessarily a
  system fault
• Implies a bottom-up approach for the assessment
  of each component fault on the system operation

16
Fault Tolerance

• Whereas the Derating/Parallelism are easy to
  implement, Fault Tolerance is the most difficult
  and time consuming feature to assess with
  precision for the accelerator operation
• Plenty of technological issues
• Complex hierarchy of dependent subsystems
• Requires bottom-up analysis of fault scenarios
• Interaction with beam physics and beam dynamics
  issues
• not all cavities or quadrupoles have the same
  effects, depending on their relative positions in
  the beamline, even when considering identical
  objects under identical operating conditions)
• Need extensive beam dynamics simulation scenario,
  transforming component faults into their effects
  (if any) on the particle beam (e.g. no field in
  cavity, bad field in magnet, etc.) and develop
  corrective actions, which later will have to be
  integrated (reliably) in the control system
• See JL Biarrotte talk this afternoon, parallel
  session B
• The control system plays a major role in
  guaranteeing fault tolerance to the accelerator

17
Requisites for Fault Tolerance

• Fault tolerance requires at least five necessary
  functions, many of which need to be handled by
  the accelerator diagnostics and control system
• Fault detection
• It happened!
• Fault isolation
• why did it happen?
• Fault containment
• avoid fault propagation
• next weakest link effect
• common cause failures
• Fault masking
• no spurious value on system state due to a faulty
  component is passed out of the system boundary as
  representative of the system state
• Fault compensation
• Capabilities to compensate functions of the
  faulty component with the use of redundant
  components

18
Operating considerations Maintenance

• Reliability and availability specifications need
  to be fulfilled during the required mission time
• In ADS is at least 3 month of continuous
  operation
• To meet reliability and availability
  specifications ( keep them during time)
  maintenance and spare parts policies need to be
  set up
• In existing accelerator facilities short and
  frequent maintenance periods are scheduled in
  order not to affect adversely
• beam availability to scheduled experiments in
  user facilities (as synchrotron light sources) or
• integrated yearly luminosity for colliders
• ADS maintenance policy needs to be compatible
  with the fuel cycle
• Either adequate redundancy must be planned
• Or access to redundant devices failing frequently
  (e.g. power supplies in separate tunnel, with
  free access)
• Always plan to avoid the infant mortality and
  wear out decrease in reliability of components
  (bath tub curve)

19
European ADS activities
20
National programs on ADS accelerators (pre-2001)

• France (CEA/CNRS) IPHI-SILHI ASH, 1996
• Built the SILHI source, gt100 mA H _at_ 95 keV
• Reliability runs
• IPHI RFQ, 100 mA, H, 5 MeV
• Now 3 MeV for CERN/SPL collaboration
• Cfr. R. Ferdinand, Session B this afternoon
• DTL hot model built and tested
• Development of SC cavity prototypes at various
  beta
• Elliptical, single cells _at_ 0.5 0.65
• Elliptical, 5 cell _at_ 0.65
• Spoke, 2 gaps _at_ 0.35
• Italy (INFN) TRASCO (TRAsmutazione SCOrie), 1998
• Built the TRIPS source, gt 50 mA H _at_ 80 keV
• Reliability runs
• TRASCO RFQ, 30 mA, H, 5 MeV under construction
• Will be used for the SPES project for radioactive
  beams
• Development of SC elliptical cavity prototypes
• Single cells _at_ 0.5
• Two 5 cells _at_ 0.5

21
Outcome of National Programs/1

• The performances needed by the components for an
  ADS accelerator are well within the state of the
  art of the technology
• Example 1 SILHI/TRIPS sources
• Preliminary reliability assessments with
  dedicated runs

TRIPS _at_ LNS gt 50 mA
SILHI _at_ Saclay gt 100 mA
22
Outcome of National Programs/2

• Example 2 spoke cavities
• Can be designed with safe margins on peak fields

b0.35Eacc 16.2 MV/m Bp 134 mT Ep 49.5 MV/m
Bulk Nb, TESLA technology
IPN/Orsay
23
Outcome of National Programs/3

• Example 3 SC Elliptical cavities of reduced beta
  (0.47 0.65 built)
• Can be designed with safe margins on peak fields

Bulk Nb, TESLA technology
b0.47Eacc 17 MV/m Bp 100 mT Ep 60 MV/m
INFN
Z501 _at_ TJNAF 31/03/2004 Z502 _at_ Saclay
24/06/2004
24
Conclusions from prototyping

• Experience during the prototyping sponsored by
  national programs
• in terms of performances all components of an ADS
  system are well within the capabilities of
  technology
• All components realized through European
  industries important point for the industrial
  ADS perspective
• No needed RD for performance enhancement of the
  main components (sources, RF structures, )
• The RD prototypical activities were possible
  since they were greatly synergic with our
  accelerator community
• Proton Drivers neutrino factories (SPL),
• Radioactive ion beams (EURISOL),
• Spallation neutron sources (SNS/ESS),
• Irradiation tools (IFMIF)
• But, can we integrate these components in a
  system with the necessary reliability
  characteristics?
• It is a system design issue regarding only
  accelerators for ADS!
• Reliability requirements (in terms of number of
  trips and operation cycle!) have no parallel in
  HEP community

25
ETWG Roadmap (April 2001)

• European Technical Working Group on the use of
  ADS for nuclear waste transmutation,
• Set in 1998, chair C. Rubbia
• Representatives from 9 countries A, B, D, ES,
  FIN, F, I, P, S
• Issued a Report to Ministers Advisors of
  participating countries
• A European Roadmap for Developing ADS for
  Nuclear Waste Incineration
• 12 year roadmap on the path to operation of an
  eXperimental ADS, using standard MOX fuel and
  demonstrating the coupling with a high intensity
  accelerator and a subcritical system
• Switch to new fuel based on MA by 2025
• Industrial scale by 2040
• Time scale needed for the technological
  development of a new fuel in a NPP is of the
  order of 25 years
• Lead to integrated project in 5 Framework
  Program

26
PDS-XADS (2001-2004)

• Preliminary Design Study for an eXperimental
  Accelerator Driven System
• 3 year, 12 M, 6 M from EC
• 25 participants 11 Industries, 14 research
  institutions
• Ending October 2004
• Assess and compare different options for an 80
  MWth ADS
• Pb/Bi cooled reactor (in 2 sizes) vs. He gas
  cooled reactor
• Spallation target, with beam window or windowless
• Window severe thermal-mechanical loads, beam
  damage and highly corrosive environment (Pb)
• Windowless vacuum technology challenge, how to
  safely maintain a differential vacuum between the
  beam line and the liquid Pb/Bi spallation target
• Accelerator choice linac vs cyclotron
• Preliminary studies on plant Safety
• Analysis of safety approach and study of
  accidental conditions
• WP3 dedicated to the Accelerator 600 MeV 6 mA
• CNRS, CEA, INFN, U.Frankfurt, ITN, ENEA, IBA,
  Framatome, Ansaldo

27
PDS-XADS Deliverables

• Requirements for the XADS Accelerator and
  Technical Answers
• Set baseline design choices on the basis of
  requirements
• 600 MeV and 6 mA
• Accelerator Feedback Systems, Safety Grade
  Shutdown and Power Control
• Provisions for routine operation, commissioning
  and ramping
• Discussion on controls and diagnostics
• Accelerator Radiation Safety and Maintenance
• Radioprotection aspects
• Maintenance provisions for high reliability
• Potential for Reliability Improvement and Cost
  Optimisation of Linac and Cyclotron Accelerators
• Preliminary reliability analysis
• Costing
• Ruled out cyclotron for reliability potential
• Definition of the XADS-Class Reference
  Accelerator Concept and Needed RD
• Reference linac design, outline of needed RD
• Extrapolation from XADS Accelerator to the
  Accelerator of an Industrial Transmuter

28
The PDS-XADS linac
from J.L.Biarrotte
All prototypical activities have been performed
within National programs by thePDS-XADS
participants
29
How reliability is implemented

• Reliability guidelines extensively used in the
  linac design
• Derating
• Redundancies
• Fault tolerance
• Provide redundancy in the most critical items
• Source, RFQ, low energy nc stage
• Achieved by injector duplication!
• Handle the natural redundancy in the
  superconducting linac
• A SC linac has a high degree of modularity
• The whole beamline is an array of nearly
  identical periods (array of focusing and
  accelerating components)
• All components are highly derated with respect to
  technological limitations
• A high degree of fault tolerance with respect to
  cavity/magnets can be expected in the SC linac
• Implies a reliable and sophisticated digital RF
  control system with preset set points for
  implementation

30
Nominal beam dynamics

• Beam dynamics design criteria aimed at
• Planning smoothness
• Smooth lattice phase advance per meter across all
  beamline
• Smooth lattice variations
• Smooth beam matching between section
• Avoiding resonances
• Clear from structure/tune resonances and big tune
  depression
• Providing beam clearance
• Ratio between beam aperture to rms beam size 10-25

31
Minimal e growth in nominal conditions

• Beam from RFQ output to 600 MeV in all-SC linac
• 2 spoke (0.15, 0.35) and 3 elliptical sections
  (0.47, 0.65, 0.85)
• 6D gaussian beam

32
Reliability Assessments

• Even in the absence of a validated reliability
  DB, reliability analyses were addressed in
  PDS-XADS
• Failure Modes and Effect Analysis (FMEA)
  performed on whole linac to assess critical areas
  in the design with a bottom-up approach
• Reliability Block Diagram (RBD) analysis to
  derive from top-down reliability estimations of
  different configurations, varying in the degree
  of redundancy according to the foreseen level of
  fault tolerance in the final system
• Software tools commercially available

If the same number of components, with the same
reliability characteristics, were to be put in a
series connections, the system would have an MTBF
of 30 hours
33
IP EUROTRANS in the 6 FP (2005-2008)

• Develop a reference design for a European
  Facility for Industrial Transmutation (ETD/EFIT)
  with a power of up to several 100s MWth, together
  with a more detailed design of a short-term
  eXperimental demonstration of the technical
  feasibility of Transmutation in an Accelerator
  Driven sub-critical System (XT-ADS)
• Experimentally demonstrate the stable operation
  and dynamic behavior of an ADS at power (TRADE
  experiment)
• Develop and demonstrate the necessary associated
  technologies, especially fuels, heavy liquid
  metal technologies, and nuclear data
• Overall technical feasibility, and perform an
  economic assessment
• Integrated Project with 5 subprojects (23 M from
  EC )
• Design, TRADE experiment, Fuel, Materials and
  Nuclear Data
• Accelerator workpackage in Design domain, with
  2.2 M
• Program aimed at experimental characterization of
  reliability characteristics of selected
  components (in source, sc cryomodule, RF control)

34
Synergies with HPPA
35
ADS is a HPPA

• Many components for an ADS SC linac are
  frequently found in other programs
• High current front ends based on RFQs
• LEDA, SNS, CERN Linac4,
• Reduced beta elliptical cavities
• SNS, JPARC, EURISOL, RIA, FNAL 8GeV PD, SPL,
• Spoke cavities
• RIA, SPL, FNAL 8 GeV PD
• Machine design is driven by similar
  considerations
• Sectioning of the SC linac
• Optimal choices of structure types ( cells,
  type, ) and b
• Limit number of different sections
• high current beam dynamics
• Smooth design, Resonances, tune depression,
• Accelerating gradients often limited at low
  energy by strong longitudinal phase advances per
  cell

36
A few HPPA based on SC linacs
Same building blocks, especially in SC linacs!
37
Conclusions

• Accelerators for ADS are HPPAs where
• No record performances on components are asked
• Moderate beam currents are needed (lt20 mA, but
  CW)
• Hard system reliability goals need to be met
• Few beam stops per year gt 1 s
• Uninterrupted operation for 3-6 months
• Encouraging results from preliminary assessments
• Very stimulating environment (interaction with
  reactor designers)
• Future ADS studies focused on reliability
  assessments of an HPPA
• Design for Reliability
• Use of RAMS (Reliability Availability
  Maintainability and Safety) tools and
  methodologies from early in the conception stage
• Experimental demonstration of reliability
  characteristics of components
• Other accelerator applications are investigating
  availability design
• Example the Linear Collider study ( 30 km
  linac) to develop a luminosity budget based on
  hardware availability estimations

38
Acknowledgements

• INFN and CEA/Saclay, IN2P3/Orsay
• TRASCO and IPHI teams
• Whole PDS-XADS Collaboration
• Particularly Working Package 3 Accelerator
• CNRS, CEA, INFN, U.Frankfurt, ITN, ENEA, IBA,
  Framatome, Ansaldo
• A wish for a great time with EUROTRANS in the
  next years
• Work supported by the EURATOM 5 framework
  program of the EC, under contract
  FIKW-CT-2001-00179