Core Plasma Constraints on Divertor Design By A.W. Leonard - - PowerPoint PPT Presentation

▶
core plasma constraints on divertor design
SMART_READER_LITE
LIVE PREVIEW

Core Plasma Constraints on Divertor Design By A.W. Leonard - - PowerPoint PPT Presentation

Aug 30, 2022 141 likes •393 views

Core Plasma Constraints on Divertor Design By A.W. Leonard Presented to IAEA-TM on Divertor Concepts Vienna, Austria Sept. 29 Oct. 2, 2015 JETs ILW experience illustrates important roles of boundary plasma JET Operational space

slide-1
SLIDE 1

Core Plasma Constraints on Divertor Design

By

A.W. Leonard

Presented to

IAEA-TM on Divertor Concepts Vienna, Austria

  • Sept. 29 – Oct. 2, 2015
slide-2
SLIDE 2

2

IAEA-TM Div. Concepts, Vienna Sept. 2015

JET’s ILW experience illustrates important roles

  • f boundary plasma
  • Operational space to limit impurity

accumulation

– Divertor conditions to limit W source – Fueling for ELM frequency, W transport

  • High Confinement

– Role of Zeff on pedestal pressure – Pedestal dependence on neutral density

  • Learning curve to optimize

performance

– Core performance optimization must be designed into future tokamak divertors

JET

  • C. Challis NF 2015
slide-3
SLIDE 3

3

IAEA-TM Div. Concepts, Vienna Sept. 2015

Core Performance a Critical Constraint on Divertor Design

  • Divertor design must simultaneously accommodate core plasma

as well as divertor target constraints

– Divertor: q⊥≤10 MWm-3, Te≤ 5 eV, no transients (ELMs) – Core: High confinement, High β, etc.

  • What are the Core constraints on divertor operation?

– Maintain robust H-mode confinement – High β operation – Low central impurity density to limit radiation and fuel dilution – No/small ELMs

  • Core constraints must be translated into divertor/SOL design

metrics

– Separatrix values: Density, temperature, neutral flux, impurity density, turbulence, etc. – These values are not well defined for future devices

slide-4
SLIDE 4

4

IAEA-TM Div. Concepts, Vienna Sept. 2015

Core Confinement is a Convolution of Multiple Physics Processes

  • Confinement degradation can occur through several pathways

– Pedestal pressure degradation – Profile peaking/flattening – Rotation – MHD instabilities; NTMs, RWMs, locked modes, etc.

  • Global confinement is not a good metric for core compatibility

divertor solutions in existing devices

– The divertor and SOL most directly affects the pedestal through the separatrix

– Most other transport processes can be described in terms of pedestal top conditions; Density, Temperature, Rotation, impurities, collisionality, etc.

slide-5
SLIDE 5

5

IAEA-TM Div. Concepts, Vienna Sept. 2015

Fusion Performance Relies on Robust H-mode Pedestal

  • Fusion gain scales strongly with

pedestal pressure

  • Predictive EPED model accurate
  • ver range of conditions
  • EPED requires Div/SOL input

– Density: Dependent on separatrix density and fueling – Zeff: Dependent on SOL impurity transport

  • Operational space for EPED validity

uncertain for

– PsepàPLH – High collisionality – High/Low edge recycling

100 101 102 100 101 102

EPED Predicted Pedestal Height (kPa) Measured Pedestal Height (kPa)

Comparison of EPED Model to 288 Cases on 5 Tokamaks JET (137) DIII-D ELM (109) DIII-D QH (11) JT-60U (16) C-Mod (10) AUG (5) ITER

  • P. Snyder NF 2015

ITER Fusion Power

Pedestal Pressure

slide-6
SLIDE 6

6

IAEA-TM Div. Concepts, Vienna Sept. 2015

Uncertain relationship between Pedestal density and SOL conditions

  • Optimal pedestal density a

combination of several factors

– Pedestal pressure – Fusion Gain – Current drive efficiency

  • Div/SOL models predict nsep and Γn0

– nped results from ionization source and transport

– nsep/nped ~25% - 50% in existing experiments

  • Progress needed in two areas

– Pedestal density transport to predict nped from nsep and Γn0 – Divertor design techniques to optimize nsep and Γn0 for dissipative divertor

  • peration

5 10 15 20 25 30 25 50 75 100 125 150

Pedestal Density (1019 m-3) EPED1 Pedestal Height (kPa)

Density and N Dependence of EPED1 for ITER Ref and Hybrid Reference (Ip=15MA, N=2) Hybrid (Ip=12MA, N=2) Hybrid (Ip=12MA, N=2.6) Hybrid (Ip=12MA, N=3.2)

  • P. Snyder NF 2011
slide-7
SLIDE 7

7

IAEA-TM Div. Concepts, Vienna Sept. 2015

Pedestal Density Transport Progress Will Require More Attention

  • Uncertain role of edge recycling

in pedestal density profile

– Conflicting evidence of pedestal density pinch – Sets possible range of nsep/nped – Lack of pedestal ionization profile measurements hampers testing of emerging pedestal transport models

  • Pedestal ionization source

– High poloidal asymmetry – Difficult to measure

  • Pedestal ionization profile

measurements a key capability for progress

0.0 0.1 0.2 0.3 0.90 0.95 1.00 1.05

ψ

0.0 0.2 0.4 0.6 0.8 1.0

Density (1020 m-3) Deff (m2s-1) Deff

  • A. Leonard JNM 2013

DIII-D ASDEX-Upgrade

  • M. Willensdorfer NF 2013
slide-8
SLIDE 8

8

IAEA-TM Div. Concepts, Vienna Sept. 2015

Pedestal Pressure Often Degrades with Divertor Heat Flux Control and Detachment

  • Confinement and pedestal

degradation with high density

  • peration widely reported
  • Mechanisms for pedestal

degradation must be understood to scale results to future tokamaks

– Excessive core radiation – Lower MHD stability and increased transport at high collisionality – Neutrals, charge-exchange directly degrading transport barrier – Induced turbulence

0.4 0.6 0.8 1 1.2 0.2 0.3 0.4 0.5 0.6 0.7 0.8

ne/nGW HH-factor Ar puff (δ=0.36) D2 puff (δ=0.16) D2 puff (δ=0.36)

0.5 1 1.5 2 2.5 2 2.5 3 3.5 4 4.5 0.4 0.5 0.6 0.7 0.8 ne/nGW

ne [1019m-3] Wth [MJ] Ar puff D2 puff δ=0.36

(a) (b)

JT-60U

  • H. Urano NF 2015
slide-9
SLIDE 9

9

IAEA-TM Div. Concepts, Vienna Sept. 2015

Adequate power across separatrix required for robust pedestal

  • Pedestal typically degrades for

Psep≤PLH

  • Pedestal degradation for:

– Lower Psep due to instrinsic or seeded impurity radiation – Density increase raises PLH

  • Implications for divertor design

– Limited radiated power fraction in main chamber – SOL impurity density limit – Upper limit to core (and SOL) density

P

LH = 4.9x104ne,20 0.72Bt 0.8S0.94

Alcator C-mod

  • A. Loarte PoP 2011
slide-10
SLIDE 10

10

IAEA-TM Div. Concepts, Vienna Sept. 2015

Excessive X-point Radiation Can Degrade Pedestal and Confinement

  • ASDEX-Upgrade: Achieving

complete detachment across divertor target with N2 injection results in X-point radiation

– Pedestal pressure degrades ~60% – Profile peaking limits confinement loss to ~10% – Unknown correlation of Pped with PsepàPLH

  • Is X-point radiation as detrimental

as radiating mantle?

– On one hand, L|| may allow large Te gradient to midplane – On the other, X-point radiation may rob q⊥ across separatrix

ASDEX-Upgrade

D2 injection N2 injection Inner Detachment Complete Detachment

  • F. Reimold Nucl. Fusion 2015
slide-11
SLIDE 11

11

IAEA-TM Div. Concepts, Vienna Sept. 2015

Impurities can also Improve Pedestal Through Zeff and ν*

  • Optimal impurity density can be > 0

– Low collisionality can lower pedestal pressure stability limit through excessive bootstrap current – Similar effect to pedestal density – Zeff and ν* may also affect local pedestal transport

  • Ideal impurity level for DEMO will

depend on operational scenario

– Best plasma and impurity densities will depend on overall performance

  • ptimization

– Ideal pedestal collisionality may require some ‘leak’ of seeded divertor impurities

  • G. Maddison Nucl. Fusion 2014

JET

slide-12
SLIDE 12

12

IAEA-TM Div. Concepts, Vienna Sept. 2015

Low Pedestal Ionization May Lead to Higher Pedestal Pressure and Confinement

  • Reduced pedestal fueling

– Lower density and pressure gradients – Wider pedestal with higher pressure limit – Uncertain transport relationship between ne and Te

  • Effect of pedestal ionization in future devices

will require development and validation of pedestal transport models

– Low appears better. Is zero the best? – Measurement of 2D pedestal ionization profile a key capability needed for this development

NSTX Lithium Injection JET

  • J. Canik PoP 2011
  • C. Challis NF 2015
slide-13
SLIDE 13

13

IAEA-TM Div. Concepts, Vienna Sept. 2015

SOL Turbulent Transport May Limit Compatibility of Divertor Detachment with H-mode

  • As density increases:

– Far SOL turbulence with rapid radial transport moves inward towards separatrix – SOL radial transport correlated with collisionality

  • If increased SOL turbulence linked with

collisionality at field-line/material interface:

– Divertor detachment may inherently induce excessive turbulence at midplane separatrix – Potentially linked to density limit

223 225 227 229 231 233 235 237 R (cm) 0.1 1 10 100 ne (×1018 m-3) ~ 3 cm LCFS (a) BL ~ 3 cm ~ 8 cm ~ 5 cm ~ 2 cm LSOL OWS fGW

GW ~ 0.27

0.27 fGW

GW ~ 0.35

0.35 fGW

GW ~ 0.4

~ 0.4 fGW

GW ~ 0.5

~ 0.5 DSOL

DIII-D

  • D. Carrarelo JNM 2015
  • D. Rudakov NF 2005
slide-14
SLIDE 14

14

IAEA-TM Div. Concepts, Vienna Sept. 2015

Pedestal Degradation Not Inherently Linked with Divertor Detachment in DIII-D

  • Full detachment across target before

degradation of pedestal pressure

  • Reduction of pedestal pressure gradient

consistent with MHD stability

  • Further work needed on pedestal

response to high collisionality

Normalized Pressure gradient (p’*w1/4) 1.0 1.5 2.0 2.5 0.0 0.5 Normalized Pedestal Current (j*w1/4) 0.10 0.20 0.30 0.00

nped=3.8x10

19

Stable Unstable

γ=0.05ωA γ=0.5ωDia

nped=4.7x10

19

nped=5.3x10

19

nped=6.0x10

19

nped=6.7x10

19

nped=7.0x10

19

nped=7.1x10

19

Z (m)

  • 1.2
  • 1.1
  • 1.0
  • 0.9

1.35 1.40 1.45 1.50 1.55 1.60 1.65 Major Radius (m) 0.1 1.0 10.0 100.0 1000.0 Te (eV)

ne,ped=7.0×1019m-3

0.0 2.0 4.0 6.0 8.0 10.0 12.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 Pedestal ne,ped (1019 m-3) Pedestal Pressure (kPa) Pedestal Pressure

4.9 MW 6.2 MW

slide-15
SLIDE 15

15

IAEA-TM Div. Concepts, Vienna Sept. 2015

Examine Compatibility Scaling of Detachment with H-mode to Highlight Critical Issues

  • Premise: Compatibility with H-mode determined by maintaining

Psep>PLH

  • Figure of merit: ne,sep(detached) / ne,ped(Là

àH) ≤ 1.0

– As this ratio approaches 1.0, H-mode confinement is degraded – Use all existing empirical and model scalings

  • H-mode power threshold
  • Density at Psep=PLH

P

LH = 4.9x104ne,20 0.72Bt 0.8S0.94

ne(1020) = 2.5×10−4P

sep 1.39Bt −1.11R−2.61ε1.31κ −1.31

Alcator C-mod

  • A. Loarte PoP 2011
slide-16
SLIDE 16

16

IAEA-TM Div. Concepts, Vienna Sept. 2015

Use 2PM to estimate scaling of separatrix density at detachment onset

  • The 2-point model for midplane separatrix density
  • Simplifying assumptions for q||

– Simple ITPA λq scaling, λq∝Bp

  • 1

– q||,in=q||,out, qelectron=qion

  • Assume Te,div=15 eV

– Transition to convective transport – Minimum miplane density at detachment onset – Minimal pressure loss, lesser radiative losses

  • Resulting ratio of nsep,det/nped,Hà

àL

nsep = 1.1×1015q||

5 7

L||

2 7Te,div 1 2

1− frad

( )

fmom # $ % & ' ( nsep nH→L =1.36×104 P

sep

R # $ % & ' (

−0.68

Bt

1.82R0.93q−0.29ε−1.31κ1.31 1− frad

( )

slide-17
SLIDE 17

17

IAEA-TM Div. Concepts, Vienna Sept. 2015

Caveats and Lessons from Scaling Exercise

  • Scaling indicates

– Compatibility easier at higher power – H-mode compatibility much tougher in large high field devices – Range of accessible nsep/nped important – Radiative losses upstream of pressure loss beneficial

  • Critical assumptions to study

– nsep scaling with q|| – ITPA λq scaling to high power – Role of Greenwald density limit – Coupling of divertor entrance to midplane density

  • Similar exercise for scaling of ν*ped at detachment onset

nsep nH→L =1.36×104 P

sep

R # $ % & ' (

−0.68

Bt

1.82R0.93q−0.29ε−1.31κ1.31 1− frad

( )

υ ped

*

∝ P

sep

R " # $ % & '

2 nsep

nped " # $ $ % & ' '

−3

q95

2

Bt

2βn,ped 2

P

sepBtq95

( )

1 7

ε 7 2

slide-18
SLIDE 18

18

IAEA-TM Div. Concepts, Vienna Sept. 2015

H-mode Compatibility Scaling is Evident in Existing Models Used for Design

  • ITER modeling

– Operational space bounded by H-mode and detachment – Factor of 3 lower λq à Factor ~2 increase in nsep

  • Fig. 5. Operational window showing

l contours for

qpk < 10 MW/m 2 and (a) fperp = 1 (kqk = 3.6 mm), (b) fperp = 1/2 ( kqk = 1.6 mm) and (c) fperp = 1/4 ( kqk = 1.2 mm) vs. alpha particle power. Vertical axis is linear in alpha power fraction, Q/(5 + Q), and is labelled by the corresponding Q. A.S. Kukushkin et al. / Journal of Nuclear Materials 438 (2013) S203–S207

Narrowing λq à à

  • A. Kukushkin JNM 2013
slide-19
SLIDE 19

19

IAEA-TM Div. Concepts, Vienna Sept. 2015

Divertor and SOL Design Must Limit Core Impurity Influx

  • Low-Z control to limit fuel dilution

– Helium ash must be pumped even for detached divertor – Whether helium ash or seeded impurities are of greater concern will be core scenario dependent

  • High-Z control to limit core radiation

– Tolerable high-Z SOL contamination from divertor and main chamber PFCs dependent on core operational scenario – ELM control will affect core confinement of high-Z – Lowest high-Z separatrix density is desirable, but upper limit requires improved pedestal transport models

slide-20
SLIDE 20

20

IAEA-TM Div. Concepts, Vienna Sept. 2015

Small Main Chamber Gap to Separatrix Allows High β scenarios

  • High β scenarios require a nearby

conducting wall

  • RF current drive technologies also

benefit from proximity to separatrix

  • Optimal spacing a trade off of

competing effects

– SOL interaction with wall for dissipative divertor operation – High β stability – RF current drive

Ideal-wall stability limit

DIII-D

Courtesy of A. Garofalo

slide-21
SLIDE 21

21

IAEA-TM Div. Concepts, Vienna Sept. 2015

Double-null a Consideration for Overall Performance Optimization

  • Double-null advantages for core

performance

– Higher triangularity for confinement and global β – Potential greater use of Bt volume – Easier access for divertor components – Design flexibility for heat and particle flux control

  • Double-null concerns

– Magnetic balance for best heat flux control may not match best particle and impurity pumping – Higher H-mode power threshold for magnetic balance optimized exhaust pumping efficiency

slide-22
SLIDE 22

22

IAEA-TM Div. Concepts, Vienna Sept. 2015

Each ELM Control Scheme Imposes Its Unique Constraints on Divertor Solution

  • ELM control solutions will affect the

boundary solution

– Heat and particle flux control – SOL and PFC interaction

  • ELM control may also constrain divertor
  • peration with unique requirements
  • RMP 3D fields

– Operational constraints on density/ collisionality and neutral flux

  • Pellet ELM pacing

– Pumping to handle increased particle flux

  • QH-mode, or other ELM-free reqimes

– Maximum density/collisionality – Plasma rotation

slide-23
SLIDE 23

23

IAEA-TM Div. Concepts, Vienna Sept. 2015

High Performance Plasmas will Make Requirements of Divertor and SOL Plasma

  • Core plasma response to divertor plasmas is a convolution of

many processes that must be individually scaled to DEMO

– Pedestal stability and transport – Core radiation and fuel dilution – Profile peaking – Rotation effects on stability and transport – Global MHD instabilities; NTMs, RWMs, Locked modes, etc.

  • Most core effects are mitigated through the pedestal, but

requirements for the Pedestal/SOL interface are not yet defined

– Tsep, nsep, nimp, Γn0, q⊥sep,nsep, ñ, etc.

slide-24
SLIDE 24

24

IAEA-TM Div. Concepts, Vienna Sept. 2015

Core Performance Issues Needing Resolution for Divertor Design

  • Pedestal will likely require lower nsep than standard divertor

solution, but quantitative limit requires further development

– Pedestal density transport, response to neutral flux – Operational limits of robust pedestal operation – Measurement of pedestal 2D ionization profile a key measurement for progress

  • Impurity control in divertor and SOL needed

– Limits to nimp,sep depedend strongly on operational scenario, ELM control

  • Small wall gap beneficial for overall performance

– Tradeoffs of mainchamber PMI, divertor operation and global b

  • Consider implications of double-null configurations

– Lots of physics and PFC considerations to tradeoff

  • Not too early for scaling exercises to guide solutions and

identify tests of key processes