Integrated power exhaust strategies for ITER:

1
Integrated power exhaust strategies for ITER a
few remarks on the problem W. Fundamenski
(UKAEA, EFDA-JET) with contributions from
V.Philipps, G.Matthews, S.Brezinsek, A.Loarte,
F.Sartori, C.Lowry, P.Lang, Y.Liang, T.Eich
2
Ignition vs. exhaust criteria for a reactor

• In order for any exothermic reactor to operate in
  steady-state,
• fresh fuel must be added at the rate at which it
  is consumed,
• this fuel must be heated, ideally by the
  reactions themselves,
• fuel must be confined, by whatever means are
  available, for sufficiently long to allow the
  exothermic processes to continue,
• the energy and ash must be removed from the
  system at the rate at which they are created,
• the impurities released from the reactor walls
  during this exhaust process must not inhibit the
  ignition (burn) of fuel, and
• the reactor itself, primarily its walls, must not
  be damaged by all the exhaust processes.
• Conditions (i)-(iii) represent ignition criteria,
  conditions (iv)-(vi) as exhaust criteria.
• Taken together these constitute the criteria of
  mutual compatibility between the reaction
  processes and materials/components in an
  exothermic reactor
• Since the latter provide the boundary conditions
  for the thermodynamic quantities, they
  effectively determine the maximum achievable
  energy gain for a given reactor design, i.e. a
  give combination of fuel and hardware.

3
Ignition vs. exhaust criteria for ITER

• In order for ITER to operate in steady-state,
• D T must be added (NBI pellet fuelling) at
  the rate at which they are consumed
• D T must be heated by RF, NBI and dominant a
  heating
• D T must be confined for sufficiently long to
  ensure Q Pfus / Pheat 10
• The (neutron, photon and plasma) power and He ash
  must be removed from the system at the rate at
  which they are created
• the Be, C W impurities released from the
  reactor walls must not inhibit the continued burn
  (Q10) of D T by either dilution and/or
  radiation,
• The plasma facing components must not be damaged
  by the above processes.
• e.g. joint requirement of partial detachment of
  both divertor legs (to reduce the inter-ELM heat
  loads below 10 MW/m2), and ELM-mitigation (to
  reduce the ELM transient energy loads below 1
  MJ/m2).
• (iv)-(vi) effectively impose the boundary
  conditions for the plasma density and
  temperature, and hence determine the maximum
  achievable Q in ITER, for a given set of internal
  and external hardware, i.e. TF, PF and control
  coils PFCs heating, fuelling, pumping and
  cooling systems, etc.

4
Ignition and MHD stability limits
5
Power exhaust limits
6
MHD stability, ignition exhaust b limits
7
Transient heat load limits for ITER
CFC
W
ITER adopted a value of 0.5 MJ/m2 for the
maximum allowed ELM energy load
8
Maximum permitted ELM size for ITER
Using best estimates for divertor wetted area and
in-out asymmetry, one finds DWELM QELM x Sin x
(1 Pout/Pin) 0.5 MJ/m2 x 1.3 m2 x 1.5 1 MJ
Assuming Wdia 400 MJ and Wped / Wdia 1/3,
then DWELM/ Wped lt 1 This requires a decrease in
the natural ELM size by a factor of 20 Some
caveats, e.g. the above assumes that the
simulated ELM pulse shape in the plasma gun is
the same as the real ELM pulse in a tokamak
2
1
9
Divertor heat loads due to ELMs
Determination of divertor ELM power flux time
dependence
W. Fundamenski
TRINITI plasma gun (from A.Loarte)
JET-T. Eich
more than 60 of DWELM,div arrives after
qELM,divmax ? smaller DTsurfELM This inherent
skewness could allow for a moderately larger ELM
load ( 30 higher)
10
Steady and transient wall loads on ITER
Type of Interaction Units 2001 PID Latest Proposal
Outer Midplane Outer Midplane Outer Midplane Outer Midplane Outer Midplane Outer Midplane
Radiation - MWm-2 0.5 0.5 0.5
Power Conducted between ELMs ? MWm-2 None None 3
Power Conducted by ELMs ? MWm-2 None 0.93 2.5 (3.4)
Energy Conducted by ELMs ? MJm-2 None 0.19 (0.93) 0.06 (1.7)
Near second X-point Near second X-point Near second X-point Near second X-point Near second X-point Near second X-point
Radiation - MWm-2 0.5 0.5 0.5
Power Conducted between ELMs ? MWm-2 None None 5
Power Conducted by ELMs ? MWm-2 None 3.8 33
Energy Conducted by ELMs ? MJm-2 None 0.77 (3.8) 0.8 (17)
Baffle Region Baffle Region Baffle Region Baffle Region Baffle Region Baffle Region
MARFE - MWm-2 1.3 1.4 0.6
Parallel (?)
1
Perpendicular (-)
Start-up Ramp-down ? MW 15 15 8
11
ELM control and effect on inter-ELM loads

• On ITER, ELM size can be reduced by a combination
  of
• pellet pacing
• Necessary tool for deep (pedestal) plasma
  fuelling
• external magnetic field perturbation (EFCC coils,
  RMP coils?, TF ripple)
• 6x3 EFCC coils envision for error field
  correction
• TF ripple in the range of 0.2-0.5
• magnetic pacing (vertical kicks)
• Not clear if suitable to ITER at present
• impurity seeding (Type-III ELMs)
• In the absence of C as radiator with Be/W mix,
  impurity seedind necessary for partial detachment
• For all the above techniques it is imperative to
  determine
• The maximum reduction in ELM size (increase of
  ELM frequency)
• Associated reduction in confinement (H98) and
  fusion gain (Q)
• Associated increase in inter-ELM heat loads
  (q_div, Te_div, Ti_div, q_lim) and any
  detrimental effects on divertor plasma detachment
• Associated increase in core plasma pollution
  (Z_eff)
• Synergistic effects
• all the above processes (pellets/EFCCs/TF
  ripple/seeding) concurrent on ITER

12
ELM size reduction by pellet injection

• Type-I ELM frequency can be increased by
  injection of small deuterium pellets, provided
  that pellet freq. gt 1.5 natural ELM freq.
  (results from AUG)
• Can the effects of plasma fuelling and ELM pacing
  be decoupled?
• Can ELM pacing be demonstrated at N_GW 0.75?

fPel gt 1.5 f0ELM
13
ELM size reduction by EFCC coils with n2
Type-I ELM frequency can also be increased by
introducing steady state n1 or n2
fields What is the change in
confinement when the magnetic pump out is
compensated by external fuelling? What is the
effect in impurity seeded, highly radiative
plasmas?

• fELM ? 15 Hz ? 40 Hz
• DTe ?
• 650 eV ? 250 eV
• n_e ?(pump-out)
• T_e ?(not fully understood)
• H98 ? ( 0-20 )

14
ELM size reduction with vertical kicks
Type-I ELM frequency can likewise be increased by
fast changes (vertical kicks) to magnetic
equilibrium
70426, 2MA 2.35T

• Natural ELM frequency ( 5 Hz) increased to
  10-25 Hz
• At freq. gt 35 Hz kicks do not always trigger an
  ELM
• Small reduction of Wdia and pedestal quantities
  ne , Te
• Promising technique for ILW, in which case the
  ELM size need only be reduced by 2-3 times

15
Impurity seeded, highly radiative discharges
Finally, ELM frequency can increased
substantially (factor of 10) by affecting a
I-III transition, that is, by replacing Type-I by
Type-III ELMs.

• Best pulse at 2.75MA/2.2T, high d
• frad0.75, Zeff 1.5 2, N-seeded
• H98(y,2) 0.83 ( 17 degradation)
• bN 1.9, n reduced by 2.5
• Both divertors detached !
• At present the only scenario compatible with ITER
  requirements of 1 ELM energy loss partial
  detachment, although at the penalty of
  confinement degradation of 15-20, which yields Q
  4-5 _at_ 15 MA.

Q10 domain
Zeff1.7
Old results
16
Effect of TF ripple on ELMs

• ELM size, as well as H98, decrease with
  increasing TF ripple
• For the same ?ped, ?WELM/Wped decreases by
  factor of two when TF ripple increases from 0.1
  to 1
• Change related to smaller conductive loss
  (?Te/Teped), rather than convective loss
  (?n/nped)
• The reduction must less pronounced at higher
  density, i.e. close to N_GW 1.

17
THE END
18
Some lessons from Be/W wall on JET
The presence of Be on main wall (limiters, dump
plate) and W in the divertor (especially W-coated
CFC tiles) imposes new limits on plasma scenarios

• Energy and power limits
• Main chamber PSI, mainly during transients
• Divertor steady state transient loads
• NBI shine-through
• Special effects associated with RF power (ICRH
  and LH)

Plasma compatibility issues 1. The risk of W
contamination of (fuel dilution in) core
plasma 2. Reduced edge and divertor radiation (in
the absence of Carbon) 3. Hence, the need for
impurity seeding to replace Carbon as the main
radiating species
19
Transient heat load limits for W-coatings on JET

• Maximum coating test temperature in cyclic
  loading (200 pulses) 1600C
• W-C carbide formation starts at about 1000C
  (exponential increase), carbidisation of the W
  layer should be avoided
• W-C carbides have lower melting point, less
  ductility, and release C

To have some margin for ELMs, Tmax should be
below 1600C (? 1200C) ? Surface temperature is
limiting in most cases (presently set by energy
limits given by metallic base structures)
Preliminary heat load tests in Judith simulator
on 200?m VPS (2000 Elms, 1 ms) found power limit
of 0.3 GW/m2 (T. Hirai) ongoing analysis
about failure mode at higher loads Recommend
maximum transient heat load 0.2 GW/m2
20
Maximum ELM size for W-coatings on JET
Using best estimates for divertor wetted area and
in-out asymmetry, one finds DWELM QELM x Sin x
(1 Pout/Pin) 0.2 GW/m2 x 500 ms x 1 m2 x 1.5
150 kJ For typical JET stored energies of Wdia
5 MJ and Wped / Wdia 1/3, which translates
into DWELM/ Wped lt 9 (less for larger
Wdia). This requires a decrease in the natural
ELM size by a factor of 2
9
21
Maximum ELM size for W-coatings on JET
Another possible methods of estimating maximum
ELM size is based on maximum tile temperature
rise, and is thus dependent on tile
temperature. This gives somewhat higher limit for
cold tiles ( 250 kJ with safety margin).
22
Maximum T_e_div for W-coatings on JET

• W erosion proceeds by physical sputtering with an
  ion energy threshold (for deuterium ions) of
  150 eV, or T_e_div 150 / 5 eV 30 eV.
• Erosion of bulk W plate is not a critical issue
  for its life time
• Erosion of W coatings, might be an issue,
  especially for the thinner coatings, which pose a
  high risk of gradually revealing the C substrate
• Need to find an optimum between cooling the
  divertor plasma (lt 30 eV) to reduce the erosion
  of W and introduction of seeding impurities which
  can themselves increase erosion (lower energy
  threshold for higher Z)
• Need to aim for partial detachment (T_e_div lt 5
  eV) at both divertor legs !

23
ITER-like wall (ILW) Be wall
Be limiter Surfaces moved forward 3cm (W- coated
CFC recessed)
Restraint ring protection
Shinethrough Area
10µm W - coated CFC recessed
Be coating of inconel
24
ITER-like wall (ILW) W/W-CFC divertor

• W coatings
• 200 ?m VPS (Plansee) selected at first but more
  RD show that thick VPS on CFC may not be
  reliable enough
• change to 14 ?m thin Re-W multilayer coating is
  very likely (to be decided January 2008)

25
Power and energy load limits for ILW
7.5 MW/m2 on JET CFC

• Surface Temperature limits
• Steady state
• Transient (ELMs)
• Energy limit for metallic for substructures

Elms
Steady state
26
Steady state heat load limits for ILW
The limit on the steady state heat load
determined by the maximum allowed coating
temperature, heating power, radiation, SOL width
and shot duration
Maximum coating temperature 1600C ? 1200C for
ELM window
27
ELM induced material loss on JET
JET experiments at high Ip with ITER-like values
of ELM size up to 1 MJ
ELM affected area on JET 1 m2. Hence, above
increase occurs at 0.7 MJ/m2, although the
increase of radiation associated with ablation of
surface layers, rather than bulk CFC
28
Main chamber heat loads due to ELMs
ELM energy deposition at main chamber given by
competition of parallel and perpendicular
transport and filament size detachment dynamics
T. Eich/W. Fundamenski/R. Pitts
Do larger ELM filaments travel faster? What is
their spatial structure?