the diii–d advanced tokamak program · br feac-20 the challenge of self-consistent profiles mhd...

NATIONAL FUSION FACILITYS A N D I E G O

DIII–D QTYUIOP045-99/TST

THE DIII–D ADVANCED TOKAMAK PROGRAM

byT.S. Taylor

Presented atWorkshop on Physics Requirements

for Advanced TokamaksSan Diego, California

March 9–11, 1999

NATIONAL FUSION FACILITYDIII–D 031–99/TCS

OUTLINE

Introduction— What is an advanced tokamak

— DIII–D AT program goals

Steady state and high βN

Principal AT scenarios

DIII–D AT program elements— ITB physics

— Profile control [J(r)]

— Stability limits and pressure profile

— Edge stability

— Neoclassical tearing modes

— Wall stabilization

DIII–D — A NATIONAL FUSION RESEARCH PROGRAM

Collaborations with 50 institutions — 300 users1999 DIII–D National Physics Team (93 FTE)

General Atomics

LLNL

Visitors

ORNL

PPPLOther

Universities

045–99NATIONAL FUSION FACILITYDIII–D

AlaskaAlbertaCal TechChalmers U.Columbia U.Georgia TechHampton U.Helsinki U.Johns Hopkins U.LehighMITMoscow State U.RPIU. MarylandU.TexasU.TorontoU.WalesU.WashingtonU.WisconsinUC BerkeleyUC IrvineUCLAUCSD

ASIPP (China)Cadarache (France)CCFM (Canada)Culham (England)FOM (Netherlands)Frascati (Italy)Ioffe (Russia)IPP (Germany)JAERI (Japan)JET (EC)KAIST (Korea)Keldysh Inst. (Russia)KFA (Germany)Kurchatov (Russia)Lausanne (Switzerland)NIFS (Japan)Troitsk (Russia)SINICA (China)SWIP (China)Southwestern Inst. (China)Tsukuba U. (Japan)

ANLINELLANLLLNLORNLPNLPPPLSNLASNLL

CompXCPI (Varian)GAGycomOrincon

NATIONAL LABS UNIVERSITIES INTERNATIONAL LABS

INDUSTR


WHAT IS AN ADVANCED TOKAMAK

"Improvement of the tokamak concept towards higher performance andsteady-state operation through profile modification and control, plasmashape, and MHD stabilization"

Goal: concept improvement— High performance

3333 High β3333 High τE

— Steady state

How, techniques— Profile control

— Shape

— MHD stabilization

500

0

– 500

"SMART": R = 4.8 m,$rel = 0.59, I = 10/7 MA,

Qss = 27

500

0

– 500

– 1000

Z (c

m)

ITER CDA: R = 6.0 m,$rel = 1.0, I = 22/19 MA,

Qss = 4

IMPROVED CONFINEMENT AND STABILITY LIMITSLEAD TO MORE COMPACT POWER PLANT DESIGNS

142–96

β ↑ 50% β ~ 4.5 I/aB

τ ↑ 80% τ ~ 3.6 τITER–89P

(Perkins, LLNL, 1993)

QTYUIOP

ADVANCEDTOKAMAK

CONVENTIONALTOKAMAK

QTYUIOP

ADVANCED TOKAMAK APPROACH RELIESON MODIFICATION OF PROFILES

"Conventional Tokamak" Advanced Physics Approach

Global parameters (0–D) Internal profiles (1–D, 2–D)

Confinement: 1 < H < 2 2 < H < 4Increase Ip JJJJ Shape, δ, κ, ε, . . .JJJJ Larger size, or JJJJ Plasma rotation → sheared (E×B)JJJJ Large B JJJJ Current density

JJJJ Pressure po / , βp JJJJ Density profile control, edge control

Stability: 2 I/aB ≤ β ≤ 3 I/aB 3 I/aB ≤ β ≤ 6 I/aBJJJJ Limited power/A JJJJ ShapeJJJJ Increase Ip, lower q JJJJ Plasma rotation, wall stabilization

JJJJ Current density, lllli, qo, < J(a) > JJJJ Pressure po / , p′(a)

Current Inductive RF "smart" current drive bootstrapDrive: JJJJ Pulsed reactor JJJJ Steady-state

Disruptions: High Ip → low q Lower Ip — higher q → lower disruptivity→ increased risk Disruption avoidance – "maintain" stable profilesof disruptions Disruption control — local heating and

current drive, passive and active mode control

⇒ Advanced Tokamak achieves same performance at lower Ip

045–99

reduced force

NATIONAL FUSION FACILITYDIII–D QTYUIOP

045–99

THE MISSION OF THE DIII–D PROGRAM IS

To establish scientific basis for the optimization of the tokamak approach tofusion energy production

— The DIII–D Program's primary focus is the Advanced Tokamak thatseeks to find the ultimate potential of the tokamak as a magneticconfinement system

Strategy:— Demonstrate improvements

separately, then simultaneously

— Develop solid scientific understandingand predictive capability3333 Diagnostics3333 Theory/modeling3333 Require strong coupling of theory

and experiment

— Develop control scenarios and tools basedon scientific understanding

— Increase performance and duration

Performance

GOALβN H

τDUR/τE

⇒ Steady State

QTYUIOP112–98/TST

THE DIII–D ADVANCED TOKAMAK PROGRAM SEEKS TOEXPLORE THE ULTIMATE POTENTIAL OF THE TOKAMAK

Performance High confinement enhancement factor, H 4 High β, βN 6

Duration, Steady State Non-inductively driven, Eφ(ρ) 0

— High bootstrap fraction

Heat Removal and Particle Control High volume recombination and

radiated power fraction Low core impurity content

SimultaneouslyIntegrated

Approach: Active Control of Profiles Plasma shape Current density profile

⇒ Localized current drive Pressure profile

⇒ Localized heating for ITB⇒ ΩE×B, poloidal rotation with RF

Rotation profile, E×B Density profile Radiation, impurity, profile Recombination, fueling source

016–98 QTYUIOP

ADVANCED TOKAMAK PROGRAMDEMANDS NEW RESEARCH PARADIGM

Details of profiles are important— q(ρ); P′(ρ); Ω(ρ) or ωE×B; P(a), J(a); turbulence, n(ρ), T(ρ)— Geometry is important, plasma shape, divertor shape; interaction with profiles— Inherently 1 and 2 dimensional— Requires new profile diagnostics— Requires new analysis approaches, new analysis tools

Self-consistency and simultaneity increase complexity— Strong, nonlinear interaction of many elements

3333 Current profile → transport → pressure profile → current profile3333 Heating deposition → pressure profile → ωE×B → transport → pressure profile →

current profile3333 Strong coupling between pressure profile, plasma shape, wall stabilization and stability limit3333 Divertor → SOL → pedestal → edge stability and core transport

— Synergism among different physics must be fully explored— Demands more integrated experimental research program— Requires new analysis code capabilities — more “integrated” codes

Fundamental physics theories are required to understand, lead the experiments, andprovide predictive capability— Strong coupling between theory, modeling, and experiment

⇒ Increased challenge ⇒ exciting

QTYUIOPBR FEAC-20

THE CHALLENGE OF SELF-CONSISTENT PROFILES

MHDStability

Shape

ECCD and FWCDCurrent Drive

q ProfilePressureProfile

Bootstrap

Magneticshear

E×B shea

r

BeamPower

TransportRotation, ΩωE×B, γmax

Magnetic

shear

Shafranov

shift

InductiveShaping

045–99

LocalizedHeating ECH


DIII–D

FIGURES OF MERIT

045–99NATIONAL FUSION FACILITY

S A N D I E G O

DIII–D

Confinement enhancement

Normalized beta

Bootstrap fraction

Fusion gain

Steady state Q

Bootstrap alignment (Politzer)

Duration

H = τE/τEITER–89P

βN = (I/aB)BT

fBS ∝ ξ √A q βN

PfusPloss

∝ nT τ ∝ βN HI1 B3 ∝ βN H q2

PfusPCD

∝ γcur εeff βN B

3

n q (1 – ξ √A qβN)

falign = 1 – ∫ dv

∫ dv

neT abs (J – JBS)

neT abs (J)

τDUR/τE ; τDUR/τCR

QTYUIOP112–98 TT/df

A COMPACT STEADY-STATE TOKAMAK REQUIRES OPERATION AT HIGH βN

βN = 3.5

Large Bootstrap Fraction (Steady State)

0.0 0.5 1.0 2.0

10

εβp

β

ε 1 + κ2

Equi

libriu

m L

imit

q* = 4

LargePower

Density

βN = 5

• High power density⇒ high βT

• Steady state ⇒ high IB/Ip

⇒ high βp

• High βT + high βp ⇒ high βN

2 5

01.5

Advanced Tokamak

Curr

ent L

imit

Ideal MHD

Pressure Limit

β β κ βT Np ∝ +

1

2

22

β βN T aB= ( )I

StableUnsta

ble (Exte

rnal Kink)

Unstable (Internal Kink)p

r

0.5 0.7 0.9 1.10.0

2.0

4.0

6.0

βN

li (INTERNAL INDUCTANCE)


DIII–D

MODELING INDICATES PROFILE OPTIMIZATIONAS A WAY TO INCREASE THE BETA LIMIT

Beta limit sensitive to— Current profile— Pressure profile

Inductively driven currentconstrains allowable currentprofiles (and possiblypressure profiles)

Strong heating, fueling,current drive alters theconstraints

Implies two time scales— Pressure profile

relaxation → τE— Current profile

relaxation

045–99

(J.R. Ferron, Phys. Fluids B, 1990)


HIGH PERFORMANCE DURATION IS A RELEVANTFIGURE OF MERIT

Profiles impact the stability limit

— Pressure profile

— Current density profile

⇒ Two obvious time scales of interest to develop physics basis

— Pressure profile relaxation, τE3333 Required to demonstraate self-consistency between transport,

pressure profile, and stability limit

— Current profile relaxation, τCR

τCR ~ 1. 4a2 K

Te3 2

Zeff3333 Required to demonstraate self-consistency between pressure profile,

current density profile, and stability limit


DIII–D 045–99

THE DIII–D AT PROGRAM BALANCE — FOCUS

The DIII–D program aims to develop the best possible operational scenario forfusion energy production— Many opportunities to make improvements— Many complex interdependencies ⇒ many possible advanced tokamak solutions

A key to our approach to research.Maintain research attitude that is open to— Evolving and improving knowledge base— Innovations— New discoveries⇒ Device flexibility⇒ Diagnostic capabilities⇒ Control capabilities

BUT,

Focus sufficiently to test specific scenarios— Indicated by the theory— Identified potential for high performance + steady state— Consistent with limited resources

031-99 tcs/rsNATIONAL FUSION FACILITYDIII–D

Plasma Parameters DIII–D Plasma ShapeIp = 1.6 mega-amperesB = 2 Teslaq95 = 4.4Ti(0) = 10 keVTe(0) = 6 keVne = 6.5×1019 m–3

τE = 0.21 seconds

0

4

8 βNH98hy

0

1

2 H98hy

1.0 1.5Conventional Tokamak (ITER–EDA)

2.0 2.5 3.0 3.5Time (s)

0

2

4

Plasma Pressure

Plasma Energy Confinement Improvement

Advanced Performance Product

β (%)96686

96686

AT duration limited by neoclassical tearing mode

βN = βT/(I/AB)

H98y = τE/τEITER98y

Recent Progress

0

2

4

6

8

10

0.1 1 10 100

Perf

orm

ance

, βN

H9 8

hy

Duration, ∆t/τE

1998

1997

Conventional Tokamak(ITER–EDA)

(Max QDD)

≈ H89P /1.7

Advanced Tokamak(ARIES–RS)

DIII–DJT–60UAUG

DIII–D 1998 RESEARCH HAS EXTENDEDTOKAMAK PERFORMANCE AND DURATION


WE HAVE IDENTIFIED FOUR ADVANCED TOKAMAK SCENARIOSTHAT HAVE POTENTIAL FOR IMPROVED PERFORMANCE

Negative central magnetic shear

High internal inductance

Radiative improved modes

— Consistent with high lllli

GyroBohm scaling of ELMing H–mode to improved confinement

— ρ* scaling to ignition gives H > 3

— Requires improvement of the β-limit (NTM)

045–99

STEADY STATE CONSIDERATIONS ALSOLEAD TO TWO “NATURAL” CURRENT PROFILES

QTYUIOP

High current drive

J ⇒ high li

⇒ NCS

ρ

High bootstrap

Qss = PfusPCD

∝γcur Pfus

nRICD∝

γcur PfusnRI (1 – JBS)

QTYUIOP045–99

IMPROVED PERFORMANCE CONSIDERATIONSLEAD TO THE IDENTIFICATION OF TWO AT SCENARIOS

BASED ON THE CURRENT PROFILE

γLS HS

LS

HS

unstable

High n ballooning stability diagram

Magnetic Shear

Both low magnetic shear (LS) and high shear (HS) are favorable for:

Growth rate of trapped particle modes

Reduced turbulence andReduced transport Higher Beta

Motivated by theoretical considerations

— High lllli — high magnetic shear (HS) in the outer region of the plasma— Negative central magnetic shear → low or negative shear (LS) in the plasma core

α

Magnetic Shear1

0

HSLS

0

LS

HS

QTYUIOP124–97

lllli

10

2

00 1 3

β N (%

–T

–m/M

A)

Ballooning limit (Calc.)Experiment

4lllli

8

6

4

12

5

1.5 2.0 2.51.0

2

1

3

4

3.0li (target)

βN

BT = 1.5T, Ip = 0.4 – 0.5 MA, PNBabs = 12– 16MW,p (0)/ = 1.7 – 2.2

βN

(a)

(c)

JT-60U

DIII–D

βN INCREASES WITH INTERNAL INDUCTANCE,lllli

20.0

0.5

1.0

1.5

2.0

2.5

3.0

0

1

2

3

li (magnetics)

β N (

diam

agne

tic)

1.1 - 1.9 MA2.0 - 2.7 MA

D DT Ip

[Ferron, Phys. Fluids B5, 2534, (1993)]

(b)

[Sabbagh, IAEA, Montreal, F1-CN-64/AP2-17]

[Kamada, Nucl. Fusion 34, 1603, (1994)]

QTYUIOP

ENERGY CONFINEMENT INCREASES WITHPEAKING OF CURRENT DENSITY PROFILE

τ TH/

τ JET

/DIII

–D

2.0

1.5

1.0

0.5

0.00.5 1.0 1.5 2.0 2.5

Internal Inductance (li)

H–modeκ ∼ 1.7

L–mode κ ∼ 1.7

Current RampL–mode, κ ∼ 1.2

κ Ramp

H–mode, κ ∼ 2.1

124–97

τJET/DIII–D = 0.106 PL–0.46 I1.03 R1.48

Confinement improvement with li observed in many experiments

— TFTR [Zarnstorff, Phys. FluidB 3, 2338, (1991)]— JET [Christiansen, Proc. 19th EPS Conf., Innsbruck, Austria, Vol. I, p. 13. (1992)]— Tore Supra [Hoang, ibid, p. 27]— JT–60U [Kamada,Nucl. Fusion 34 1605, (1994)]

TFTR has obtained high fusion power in high li discharges (Sabbagh, IAEA, Montreal, F1-CN-64/AP2-17):

— PFus = 8.7 MW

— lllli = 1.4

— Ip /BT = 2.0 MA, 4.8 T

Ferron, Phys. Fluids B5, 2534 (1993)

145–97 TT QTYUIOP

INCREASING H AND βN WITH lllli SUGGESTAN ATTRACTIVE ADVANCED TOKAMAK SCENARIO

Advantages

Ease of central current drive

High βN, high H observed on manyexperiments

No power threshold Compatibility with ELMing H–mode,

radiative I–mode

Challenge: Self-Consistent High β, High lllli Scenario

0.0 0.4 0.80.0

1.0

2.0

ψ 1/2

<J>

(MA/

m2 )

Total current

Bootstrapcurrent

Limitation

Alignment of bootstrap current:

High lllli

⇒ Low edge J, high Sn⇒ Reduced edge transport⇒ High edge p′⇒ High edge bootstrap

High edge J

qo = 1.05

lllli = 1.2

q95 ~ 8 qo = 0.55

IBS/Ip ~ 50%–60% ⇒ ⇒ ⇒ lllli = 1.4–1.6

βN ~ 4 with sawtooth

H ~ 2–3 stabilization

DIII–D QTYUIOP298–98

NEGATIVE CENTRAL MAGNETIC SHEAR (NCS) SCENARIO IS BEINGPURSUED BY DIII–D AS STEADY-STATE HIGH PERFORMANCE

ADVANCED TOKAMAK (AT) SCENARIO Reduced current drive requirements

with aligned bootstrap is predicted— Ozeki et al., Nucl. Fusion 33, 1025 (1993)— Kessel, Phys. Rev. Lett. 72, 1212 (1994)— Manickam, Phys. Plasma 1, 1601 (1994)— Turnbull, Phys. Rev. Lett. 74, 718 (1995)

Improved performance observed experimentally— Reduced core transport observed

in a number of experiments— Highest performance in DIII–D is in NCS

with H–mode edge

Great progress in understanding the iontransport in internal transport barriers— ωE×B > γLIN— χ ~ χNeo 0 < ρ < 1

Challenges remain in understandingelectron thermal transport

Reverse Magnetic Shear

Current profile control and transport barriercontrol is needed to increase duration

Broad pressure profiles and wall stabilizationare needed for improved stability

016–98 QTYUIOP

SELF-CONSISTENT SIMULATIONS INDICATE DESIRED CURRENTDENSITY PROFILE CAN BE OBTAINED WITH OFF-AXIS ECCD

— Extended duration —

βN = 4, H = 2.8

Ip = 1.0 MA, ne = 3.2 × 1019

PEC = 2.3 MW, IEC = 0.15 MA

150

0.0ρ

100

50

00.5 1.0

J(A/

cm2 )

Jtot JEC + JNBJboot

— Fully penetrated profiles —

βN = 5.7, H = 3.6

Ip = 1.6 MA, ne = 5.7 × 1019

PEC = 7 MW, IEC = 0.32 MA


KEY CHALLENGE FOR NCS SCENARIO

⇒ Consistency of resulting pressure profile in discharges with transportbarriers with stability at high beta

Steep pressure gradients at ITB or at the boundary can lead toinstability at low beta

⇒ Develop scenarios that have "naturally" favorable profiles

— Requires fundamental understanding of ITB, and stabilityboundaries

— Magnetic shear has an input

or

⇒ Develop transport barrier control techniques

NCSNCS

OPTIMIZEDSCENARIO

(SUSTAINED)

WALL STABILIZATIONFEEDBACK

PHYSICS UNDERSTANDING DRIVES DIII–D AT RESEARCH PLAN1999

PHYSICSPRINCIPLES

PLASMACONTROL

INTEGRATEDPHYSICS

2000 2001 2002

WALL STABILIZATIONPRINCIPLES

NTM PHYSICS NTM CONTROL

NCS DEVELOP

EDGE STABILITYPHYSICS

COUNTER NBI RF CDCOUNTER

ADVANCED TOKAMAK

ITB PHYSICS

OPTIMAL EDGE ANDDIVERTORδ, SN/DN

TOOLAPPLICATION

OPTIMALMODE SPECTRUM

AT DIVERTORNEUTRAL, FLOW

CONTROL

HIGH liSCENARIO

DEVELOPMENT

INTERMEDIATESCENARIO

(EXISTENCE)

HIGH li(EXISTENCE)


NA

TIO

NA

L F

USIO

N F

AC

ILIT

YS

AN

D

IE

GO

DIII–D

031-99 KHB/jy

STEEPEST CORE G

RADIENTS FORM

IN SHOTS

WITH NEG

ATIVE MAG

NETIC SHEAR

20

Shot 959891500 m

s10

Ti (keV)Vϕ/R (103s–1)ne (1019m–3)

5066

44

22

00

0.00.2

0.40.6

ρ0.8

1.0

q

200

100 15

INTERNAL TRANSPORT BARRIER (ITB) HAS BEEN SUSTAINEDFOR ~5 s IN DIII–D L–MODE DISCHARGES


DIII–D011–99/RDS

Fully penetrated current, profile with weak central magnetic shear and q > 1

Tem

pera

ture

(keV

)

0

5

10

0.0 0.2 0.4 0.6 0.8 1.00

2

4

6

8

Ti

Te

q

ρ

I p (M

A)T i

(keV

)

0

5

10

0

1

0

24

Pb (M

W)

0 1000 2000 4000 5000Time (ms)

3000

R (m) 1.834 1.930 1.993 2.043 2.123 2.200

ITB

BES shows suppression of turbulence ρ < 0.4∼ Key to long-pulse ITB is sustainment of the current profile

IAEA-F-CN-69 EX5/6 Synakowski

MICROWAVE ELECTRON CYCLOTRON HEATINGPROVIDES LOCALIZED CURRENT DRIVE


J ECCD

(A c

m–2

)

0.0 0.2 0.4 0.6 0.8 1.0

IECCD = 92 kA

–200

20

40

60

80

100

ρ

ρ

0.0 0.2 0.4 0.6 0.8 1.0

0–2–4

2468

10

J ECC

D (A

cm

–2)

IECCD = 35 kA Second Harmonic Resonance

SteerableAntenna

ρ = 0.15

ρ = 0.5


NCS DISCHARGES CAN BE FURTHER CATEGORIZEDBY THE EDGE CONDITION

L–Mode Edge ELMing H–Mode Edge ELM-Free H–Mode Edge

Advantages Low p′ edge Large SOL Neoclassical

transport in core Good beam

penetration

Steady-state edge ELMs purge impurities Cryopump can control

density

Highest βNH product Neoclassical χi over

entire radius Better bootstrap

alingment High performance even

with Ti ~ Te

PresentLimitations/Status

H ~ 2.3, βN ~ 2.3 Low beta limit Misaligned

bootstrap current Small core volume Control of qmin

H ~ 2.4, βN ~ 2.8 Difficult to establish ITB Type I ELM perturbations

are too large Neoclassical tearing

modes

Little control over edgedensity, p′ and Jbs

⇒ peeling modes Carbon impurity

accumulation at edge Pump ineffective for

density control

ResearchEffort/Directions

Expand radius ofqmin

Control core p′ usingRF

Use ECH to slowcurrent diffusion

Reduce edge 2nd stabilityaccess and ELM size byvarying shape(squareness)

Sustain ELMing H–modeduring Ip ramp foradditional profile control

Change edgecollisionality to reduceJbs

Trigger ELMs prior to X-event (Global MHD event)using pellets or other?

THE PATH TO THE NCS–AT GOAL LEADS THROUGHMANY STABILITY ISSUES

031-99 rds/jyNATIONAL FUSION FACILITYDIII–D

PressureProfile

BroadHigher H

Larger ITBBetter Bootstrap

Alignment

H–mode EdgeBroader P (r)

NCS–ATGoal

Edge Stability(T1, T5)

Neoclassical Tearing(T3)

Wall Stabilization(T4)

L–mode EdgeMore Peaked P (r)

Kink Modeβ–limit

PeakedLower H

Narrow ITB

Neoclassical Tearing

(T3)Stable? Low ∇Pin s > 0 Region

031-99 RDS/jyNATIONAL FUSION FACILITYS A N D I E G O

DIII–D

EXPERIMENTAL β LIMITS CONSISTENT WITHCALCULATED DEPENDENCE ON po/⟨p⟩

TFTR high po/⟨p⟩ ~ 6.0 (ERS–mode):βN < 2

— Limited by fast n = 1 disruption~

DIII–D high po/⟨p⟩ ~ 6.0 (L–mode):βN < 2.5

— Limited by fast n = 1 disruption~

DIII–D low po/⟨p⟩ ~ 1.5 (H–mode):βN < 4

— No disruptionlimited by ELM-like activity fromfinite edge pressure gradients

~

0 2 4 6 8 10

Resistive

Ideal

Unstable

H–modeL–mode

P(O) / ⟨P⟩

β N (%

-m-T

/MA)

HIGH BETA REQUIRES BOTH BROAD PRESSURE AND STRONG SHAPING

QTYUIOP124–97

Calculated ideal n = 1 stability limit, wall at r/a = 1.5

Fixed q profile: q0 = 3.9, qmin = 2.1, q95 = 5.1

2 3 4 5 61

3

5

βNδ = 0.7κ = 1.8

βN = β (I/aB)–1 β∗ = <p2>1/2 2 µo/B2

δ = 0κ = 1.0

p0 /⟨p⟩

10

8

6

4

2

02 4 63 5

β*

κ = 1.8 δ = 0.7

κ = 1.0 δ = 0

p0 /⟨p⟩[A. Turnbull, IAEA 1996]



PressureProfile

BroadHigher H


Alignment


NCS–ATGoal





Kink Modeβ–limit

PeakedLower H

Narrow ITB



ENERGY CONFINEMENT TIME INCREASES WITHTHE PEDESTAL PRESSURE IN DIII–D ELMing H–MODES


DIII–D248–98

QTYUIOP

"Stiff" transport models predictτE increasing with PPED

For fixed shape H ∝ βPED1/2

ITER SHAPE(fixed)δ = 0.24κ = 1.75ε = 0.34

H-I

TE

R93

H

0 2 4 60

0.4

0.8

1.2

1.6

Type I ELMsType III ELMsL–mode

ITER Shape, q = 3.2, I = 1.5 MA95 P

PPED (kPa)elec

m = 4m = 5

m = 6m = 7

q = 1 q = 4/3 5/3 2

PressurePedestal

m = 8m = 9

Discharge #75121

n = 3VH–Mode

0.0 0.2 0.4 0.6 0.8 1.0ρ

0.0 0.2 0.4 0.6 0.8 1.0ρ

1.0

0.0

0.5

m = 4m = 5

m = 6

m = 7

x m

1.0

0.0

0.5

x m

q = 4/3 q = 2

m = 8

n = 3

Discharge #92001

GATO CALCULATIONS

ELM

WIDTH OF THE EDGE MODE IS LARGER WITHA LARGER PRESSURE PEDESTAL

031-99 RDS/jyNATIONAL FUSION FACILITYDIII–D

VH–mode (βN = 3): termination H–mode (βN = 2): ELM PressurePedestal

confinement.

EMERGING EDGE STABILITY PICTURE SUGGESTS NEED TO REGULATETHE EDGE BOOTSTRAP CURRENT AND/OR THE EDGE PRESSURE

GRADIENT TO EXTEND THE DURATION OF AT MODES (T1)

Need ELMs to provide density and impurity control

Pressure gradient and bootstrapcurrent work in a positive feedbackloop until second stability limitis reached

But, most of our high performance discharges are terminated by an ELM that couplescouples tolow–n modes, causing a drop in plasma confinement

Need to regulate the edge ∇P and or JBS either through changes in transport or stability

Techniques: shaping, impurity radiation, fueling, edge ECH

Reducing the edge current should, in general, be the favorable path since this willreduce the kink-like (low–n) character of the edge instability

Reference Equilibrium:κ = 1.8, δ = 0.3, A = 170/65, ψped = 0.98,ψwid = 0.0125, q95 = 3.5, and qaxis = 1.1

031–99 RDS/jyNATIONAL FUSION FACILITY

S A N D I E G O

DIII–D

Unstable

Un

Stable

Stable Equil. p'

0.9 0.95 1.0

p'

ψ~

SHAPING: HIGH AND LOW SQUARENESS ELIMINATES SECONDSTABLE ACCESS AND HAS A LARGE IMPACT ON ELMs

031–99/RDS/jyNATIONAL FUSION FACILITYS A N D I E G O

DIII–D

1

0

0

4

8

1.9

Edge Te (keV)

βN = 1.8

UnstableUnstable

βN = 1.9

Edge Te (keV)

Dα (a.u.) Dα (a.u.)

Time (s) Time (s)2.0

0.8ψ

Nψ

N

1.00.9 0.8 1.00.9

2.5 2.6

IAEA-F1-CN-69/EX8/1 Lao

No second regime accesssmall ELMs

Second regime accesslarge ELMs



PressureProfile

BroadHigher H


Alignment


NCS–ATGoal





Kink Modeβ–limit

PeakedLower H

Narrow ITB



NA

TIO

NA

L F

USIO

N F

AC

ILIT

YS

AN

D

IE

GO

DIII–D

FUTURE WO

RK: RADIALLY LOCALIZED O

FF–AXIS ECCDFO

R SUPPRESSION O

F NEOCLASSICAL TEARING

MO

DES

MO

TIVATION – Expected beta lim

it for ITER-like discharge

—Island sustained by “m

issing” bootstrap current in O–PT

G

OAL – Replace “m

issing” bootstrap current (Hegna, Callen, Zohm)

—Suppress island size or m

ake vanish

3

1.0

2960 ms

before 3/2 island

3120 ms

with 3/2 island

(and 2/2 kink?)

# 86176

0.80.6

0.40.2

q ≈ 2/2q ≈ 3/2

ρ0.0

0.1

ε1/2 ∇P/Bθ (MA/m2)

0.0

0.2

0.3

ECHf = 110.0 G

Hzfacet ang = 19.0 degtilt ang = 52.0 deg

2τRr

0 1

12

–1

34

56

SaturatedIsland

Modified Rutherford Eqn

(unmodulated,

δFW

HM = 3 cm,

localized at q = 3/2)

W (cm

) Iaux = 0

Iaux = 30 kA

dwdt

011-99 EJS/jy



PressureProfile

BroadHigher H


Alignment


NCS–ATGoal





Kink Modeβ–limit

PeakedLower H

Narrow ITB



0.0 0.2 0.4 0.6 0.8 1.0

RPlasma /Rwall

0

5

10

15 = 1.5

βN

p0/ = 4.8

p0/ = 2.4

Wall atInfinity

βN< 3.5

R PlasmaRwall

0.2 0.4 0.6 0.8 1.00.0

J

JBS

JFTJ

(MA/m2)

2.0

0.0

1.0

ρ

f ~70%BS βN =5.5

p0/ = 4.8

p0/ = 2.4

031-99 RDS/jyNATIONAL FUSION FACILITY

S A N D I E G O

DIII–D

WALL STABILIZATION IS CRUCIALFOR ADVANCED TOKAMAK OPERATION

AT operation with

— High normalized beta, βN = β/I/aB— Large bootstrap current fraction— Good bootstrap current alignment

Requires plasmas with

— Broad pressure profile— Broad current profile

Such plasmas have

— Low βN stability limit to n = 1external kink without a wall

— Significantly higher limit with aconducting wall


DIII–D038–99/EJS

IDEAL KINK MODE STABILIZED BY ROTATION AND RESISTIVE WALLABOVE NO-WALL βN LIMIT FOR > 30 τwall

Wall stabilization sustained with βN up to 1.4 × βN Plasma rotation slows as βN exceeds the no-wall limit

Resistive wall mode grows when rotation drops below a critical value

0

80

60

40

20

3

2

1

PNB = 10 MW

βN No Wall Limit

Plasma ToroidalRotation

δBr (n = 1)

Discharge 92544

1.51.1 1.3 1.41.2Time (s) RWM Onset

qmin ~ 2

q = 3

2015

5

0

10kHz

gaus

s

100

–100

0

–200

60 Hz

1.465 1.470 1.475 1.480 1.485Time (s)

Outboard dBθ /dt (T/s)

10–1

Toro

idal

Ang

le

no-wall

RESISTIVE WALL MODE (RWM) IS SUPPRESSED (~30 ms)BY OPEN-LOOP ACTIVE CONTROL


DIII–D248–98

QTYUIOP

Near stationary RWM is reproducibly obtained

IAEA F1-CN-69/EXP3/10 Strait

Feed-forward static n = 1 field is preprogrammed at RWM onset, with phase opposing the mode

kHz

This result is encouraging for active feedback

o

x

C-coil

o

x

o

x

PlannedC-coil

extension

8

4

01.20 1.25 1.30

Plasma rotationat q = 3

Time (s)

2

1

βN

96625

96633

Onset of RWM

kA

5.0

2.5

C–coil current

2.0

keV

3.0 Te at R = 2.1 m

2.5

Gaus

s

4.0

2.0

δBr

VALEN 3D FEEDBACK CONTROL MODEL PREDICTS IMPROVED β LIMIT IN DIII–D

• Existing 6 coil set can increase RWM stability limit to β ~ 3.4N

βN

Gro

wth

Rat

e (s

)

–1

1 2 3 4 51

100

104

106

ideal wallβ limit

no wallβ limit

ideal kink

resistivewall mode

No feedback6 coil feedback

031-99 RDS/jy


DIII–D

N

• Extended 18 coil set can increase RWM stability limit to β ~ 3.8(not optimized design)

18 coil feedback

• Gain in fusion powerat fixed bootstrapfraction = = 2.6! (3.8

3.0 )4

NCSNCS

OPTIMIZEDSCENARIO

(SUSTAINED)

WALL STABILIZATIONFEEDBACK

PHYSICS UNDERSTANDING DRIVES DIII–D AT RESEARCH PLAN1999

PHYSICSPRINCIPLES

PLASMACONTROL

INTEGRATEDPHYSICS

2000 2001 2002

WALL STABILIZATIONPRINCIPLES

NTM PHYSICS NTM CONTROL

NCS DEVELOP

EDGE STABILITYPHYSICS

COUNTER NBI RF CDCOUNTER

ADVANCED TOKAMAK

ITB PHYSICS

OPTIMAL EDGE ANDDIVERTORδ, SN/DN

TOOLAPPLICATION

OPTIMALMODE SPECTRUM

AT DIVERTORNEUTRAL, FLOW

CONTROL

HIGH liSCENARIO

DEVELOPMENT

INTERMEDIATESCENARIO

(EXISTENCE)

HIGH li(EXISTENCE)


1999 2000 2001 2002 2003

Inside launch pellet Inside Divertor Pump and Baffle

18 Coil Feedback6 Coil Feedback

3 MW ECH

Fueling and Edge Control

CurrentDrive

Resistive WallMode Control

Operation

10 MW ECH

10-s pulseLower divertor

Liquid jet

Counter NBI

• Verify off-axis ECCD

• βN = 1.4βN(no wall)

• Increase NTM onset threshold (βN)

• χi ~ neoclassical

• βN*H98 ≥ 6 (1 s); ≥ 3.5 (25 τE)

6 MW ECH 10 MW ECH

Progress Checkpoint

Simultaneously Demonstrate High: βN, H98, fBS, radiative divertorAT Research Program

Extend duration, performance, reactor similarity

Upgrade Options

DIII–D ADVANCED TOKAMAK 5–YEAR RESEARCH PLAN


Hardware upgrades with new diagnostics in 1999–2000 supports a two-phase AT physics development and integration plan

An initial test of AT integration with a progress checkpoint in 2001will evaluate upgrade options to extend AT integration

Physics IntegrationPhysics Development


DIII–D 045–99

DIII–D AT PROGRAM:REMAINING CHALLENGES AND OPPORTUNITIES

Understand transport barrier dynamics; broaden pressure profiles

— Develop ITB control as needed

Implement methods to sustain hollow current profiles and high bootstrap fractions

Deepen the physics understanding of neoclassical tearing modes; avoid or stabilize

Confirm our edge stability physics picture; find a compromise

Understand the physics of wall stabilization; implement feedback

WHAT DO WE HAVE TO GAIN?

An understanding of the ultimate potential of the tokamak as a magneticconfinement system

Greatly increased fusion power output

Much improved prospects for steady-state

the diii–d advanced tokamak program · br feac-20 the challenge of self-consistent profiles mhd...

Documents