View
22
Download
0
Category
Preview:
Citation preview
Page 1 CERN Academic Training Programme, 12-13 April 2011
Fusion Plasma Physics and ITER: An Introduction
D J Campbell ITER Organization, Route de Vinon sur Verdon, 13115 St Paul lez Durance, France
Acknowledgements: Many colleagues in the ITER Organization and the international fusion programme
The views and opinions expressed herein do not necessarily reflect those of the ITER Organization.
2. Physics of Tokamak Plasmas
Page 2 CERN Academic Training Programme, 12-13 April 2011
• Controlling the plasma: − heating − current drive − measurement − control
• Key physics challenges: − confinement − magnetohydrodynamic stability − power and particle exhaust − fusion power and the burning plasma − Steady-state operation
Lecture 2 - Synopsis
Page 3 CERN Academic Training Programme, 12-13 April 2011
Controlling the plasma
Page 4 CERN Academic Training Programme, 12-13 April 2011
• A magnetically confined plasma is a complex state − it cannot be created and sustained without a sophisticated feedback control
system
• Free energy which is available within the plasma tends to generate turbulence and magnetohydrodynamic instabilities (mhd) which reduce plasma confinement quality − much effort is expended to control and sustain a high quality plasma
capable of producing significant fusion power
• In a thermonuclear plasma with significant fusion power, additional control requirements are imposed in ensuring the fusion power can be sustained for extended periods − failures of the control system lead to conditions which tend to reduce fusion
power or extinguish the plasma
Plasma Control
Page 5 CERN Academic Training Programme, 12-13 April 2011
• There are several “actuators” which can influence the plasma behaviour:
− magnetic geometry: changing the plasma equilibrium shape
− plasma heating
− injecting current – changing the current profile
− rotation – injecting torque
− fuelling – injecting particles (even impurities)
• Over the years we have learned how to use these tools to influence plasma behaviour − “optimizing plasma performance” means largely finding the right
combination of control tools
How can we control the plasma?
Page 6 CERN Academic Training Programme, 12-13 April 2011
The plasma shape can be very flexible
• Plasma shape can affect both confinement and stability properties
R Stambaugh, APS (2000)
Page 7 CERN Academic Training Programme, 12-13 April 2011
• Tokamaks have a built in heating scheme: “ohmic” heating by the current − but plasma resistivity varies as Te
-3/2, so heating power declines with increasing Te
− so plasma temperatures of several keV are possible, but additional heating is required to achieve 10-20 keV
• Two basic heating schemes: − injecting neutral particle beams − injecting radiofrequency waves – because the plasma refractive index
depends on density and magnetic fields, several RF options are possible
• Each heating technique also provides some current drive
• A further question: − how much of each type do we need to optimize plasma performance?
Plasma Heating
Page 8 CERN Academic Training Programme, 12-13 April 2011
• Neutral beam injection (NBI): − intense particle beams are accelerated,
neutralized and injected into plasma − Eb ~ 100 keV, Pb up to 40MW − very effective:
− heating − current drive − fuelling − rotation drive
• For ITER: − Eb ~ 1 MeV to penetrate plasma/ drive
current − negative ion source technology − higher energy ⇒ little fuelling, little rotation
drive
Injection of Neutral Particle Beams fuelling
heating
TFTR
Page 9 CERN Academic Training Programme, 12-13 April 2011
• Ion Cyclotron Radiofrequency Heating (ICRF): − launched at frequencies ~ ωci ⇒ f ~ 50 MHz
− technology conventional − wave coupling to plasma problematic – penetration through edge
• Electron cyclotron resonance heating (ECRH): − launched at frequencies ~ ωce ⇒ f > 100 GHz
− source technology non-conventional: “gyrotrons” − coupling, absorption, space localization very good
• Lower Hybrid Heating/ Current Drive (LHCD): − “lower hybrid” a complex wave resonance in plasma: f ~ 5 GHz − technology fairly conventional (source: klystrons) − wave coupling to plasma problematic – penetration through edge
Radiofrequency Heating
Page 10 CERN Academic Training Programme, 12-13 April 2011
• Current drive provides: − replacement of the transformer drive ⇒ towards steady-state plasma
− manipulation of the current profile to improve confinement/ stability − direct suppression of plasma instabilities
Current Drive
• Current drive efficiency (driven current/ input power - ηCD): − typically increases with Te
− for beams, also increases with Eb
⇒ favourable for ITER
C Gormezano et al, Nucl Fusion 47 S285 (2007)
Page 11 CERN Academic Training Programme, 12-13 April 2011
Paux for Q=10 nominal scenario: 40-50MW
NBI Layout
DNB
120GHz
Heating the ITER Plasma
126 or 170GHz
Page 12 CERN Academic Training Programme, 12-13 April 2011
• Technology: − ICRF and LHCD fairly conventional − NBI and ECRH source technology challenging
• Coupling to plasma: − NBI and ECRH straightforward
− ICRF and LHCD problematic: antenna design challenging due to difficulty in coupling wave through (evanescent) plasma edge
• Radial localization: − Resonance condition favours ECRH and ICRF radial localization − NBI and LHCD more global in effects
• Current drive: − NBI and LHCD most efficient − ECRH and ICRF used in more specialized applications where space
localization important
Why 4 Heating Systems?
Page 13 CERN Academic Training Programme, 12-13 April 2011
1 – 10 eV 10 – 100 eV 1 – 20 keV
Measurement: “Seeing” a Burning Plasma
Page 14 CERN Academic Training Programme, 12-13 April 2011
• About 40 large scale diagnostic systems are foreseen: • Diagnostics required for protection, control and physics studies • Measurements from DC to γ-rays, neutrons, α-particles, plasma species • Diagnostic Neutral Beam for active spectroscopy (CXRS, MSE ….)
UPPER PORT (12 used)
EQUATORIAL PORT (6 used)
DIVERTOR PORT (6 used)
DIVERTOR CASSETTES (16 used)
VESSEL WALL (Distributed Systems)
Analyzing the Plasma - ITER Diagnostics
Page 15 CERN Academic Training Programme, 12-13 April 2011
• Puffing a small amount of tritium into the edge of a deuterium plasma allows hydrogenic particle transport to be investigated: • The 14MeV neutron signal provides a distinct marker for the location of the
tritium as it diffuses into the plasma
14MeV neutron signal Derived tritium profile
P C Efthimion et al, Phys Rev Lett 75 85 (1995)
Example: Particle Transport Studies
TFTR
Page 16 CERN Academic Training Programme, 12-13 April 2011
• Sophisticated imaging systems in various spectral regions allow detailed internal images of plasma to be constructed
• X-ray tomography allows internal magnetic structure of the plasma to be investigated: • figure shows an image of a
“magnetic island” formation inside the plasma
G Huijsmans, JET
Example: Magnetic Island Structures
Page 17 CERN Academic Training Programme, 12-13 April 2011
• Controlling a burning plasma in ITER will require an extensive set of control functions based on measurement systems and actuators described here:
− Plasma equilibrium control: (routine and robust) • plasma shape, position and current
− Basic plasma parameter control: (routine and robust) • plasma density
− Plasma kinetic control: (ongoing research) • fuel mixture, fusion power, radiated power …
− Control for advanced operation: (ongoing research) • current profile
− Control of magnetohydrodynamic instabilities: (ongoing research) • several key MHD instabilities
• All of these elements have been developed and demonstrated to varying extents in existing tokamak devices
Plasma Control in ITER
Page 18 CERN Academic Training Programme, 12-13 April 2011
• Plasma equilibrium control is routine: • based on magnetic sensors which
provide signals to reconstruct plasma boundary or equilibrium
• feedback signals control voltages to poloidal field coils to maintain required plasma equilibrium parameters
• very long pulses require particular care to avoid drifts in magnetic diagnostic signals
Plasma Equilibrium Control
-5
-4
-3
-2
-1
0
1
2
3
4
5
3 4 5 6 7 8 9
Z, m
R, m
0.4MA
1.5MA
2.5MA
3.5MA4.5MA
5.5MA
SOH,SOB
XPF
11.5MA
6.5MA
ITER simulation
Page 19 CERN Academic Training Programme, 12-13 April 2011
(M R Wade et al, Fusion Energy Conf 2006)
Control of MHD Instabilities
• An MHD instability is detected magnetically: − localized electron cyclotron
current drive is used to suppress the instability
DIII-D
Page 20 CERN Academic Training Programme, 12-13 April 2011
Key Physics Challenges
Page 21 CERN Academic Training Programme, 12-13 April 2011
• A well developed “neoclassical” theory of transport processes in toroidal plasmas has been developed: − unfortunately, it doesn’t do a good job
of describing plasma processes ⇒ turbulence normally dominates
− turbulence produces “radial streamers” which represent regions of rapid heat transport
− rotational shear in the plasma can break up turbulent structures, reducing transport
Plasma Confinement
(P Beyer et al, Uni Provence/ CEA, 1999)
Page 22 CERN Academic Training Programme, 12-13 April 2011
Plasma Confinement: H-mode
JET
• It is found that the plasma confinement state (τE) can bifurcate: − two distinct plasma regimes, a low confinement (L-mode) and a high
confinement (H-mode), result − this phenomenon has been shown to arise from changes in the plasma
flow in a narrow edge region, or pedestal
Page 23 CERN Academic Training Programme, 12-13 April 2011
τth ∝ IpR2P-2/3
• Turbulent transport is difficult to predict quantitatively: − so, we use scaling experiments to predict the level of energy
confinement in ITER − H-mode turns out to be robust enough to provide the basis for the ITER
design ⇒ significant reduction in size of device
Plasma Confinement: τE Scaling
Plasma Current Major Radius
Input Power
Page 24 CERN Academic Training Programme, 12-13 April 2011
• Predictions of fusion performance in ITER rely essentially on a small number of physics rules: • H-mode energy confinement scaling (IPB98(y,2)):
• H-mode threshold power:
(ie a certain level of power needs to flow across the plasma boundary to trigger an H-mode)
!
"E,th98(y,2) = 0.144 I0.93B0.15P#0.69n0.41M0.19R1.97$0.58%0.78 (s)
!
"E # IR2P$2 / 3
!
PLH = 0.098M"1B0.80n 200.72S0.94 (MW)
!
H98(y,2) = "E,thexp / "E,th
98(y,2)NB:
ITER Physics Basis I
Page 25 CERN Academic Training Programme, 12-13 April 2011
• MHD stability:
• Divertor physics:
!
q95 = 3
!
n/ nGW " 1
!
nGW (1020 ) =I(MA)"a2
!
"N = "(%) aBI(MA)
!
"N # 2.5
!
", # det ermined by control considerations
!
q95 = 2.5 a2BRI
f(",#,$)
!
Peak t arget power ~ 10MWm"2
!
Helium transport : "He* / "E ~ 5
!
Impurity content : nBe / ne = 0.02 (+ ~ 0.1% Ar for radiation)
ITER Physics Basis II
Page 26 CERN Academic Training Programme, 12-13 April 2011
• The interaction of the plasma fluid and the magnetic field is described by magnetohydrodynamic (MHD) stability theory
− provides a good qualitative, and to a significant extent quantitative, description of stability limits and the associated instabilities
• There are two basic types of instability: − “ideal” instabilities produce field line bending – can grow very rapidly − “resistive” instabilities cause tearing and reconnection of the magnetic
field lines ⇒ formation of “magnetic islands”
MHD Stability - Plasma Operational Limits
• Plasma control techniques are being applied to suppress the most significant instabilities − allows access to higher fusion performance
Page 27 CERN Academic Training Programme, 12-13 April 2011
MHD Stability: Disruptions
H-mode L-
mode CQ TQ
Plasma current Plasma energy
RE current
t Typical chain of events during
plasma disruption
• The ultimate stability limit in tokamak plasmas is set by major disruptions: large scale MHD instabilities − loss of plasma energy in milliseconds (thermal quench – TC) − Plasma current decays in 10s of milliseconds (current quench – QC)
• Produces: − very large heat loads on plasma facing surfaces − significant electromagnetic forces in vacuum vessel − large runaway electron beam
Mitigation techniques
essential
Page 28 CERN Academic Training Programme, 12-13 April 2011
• li-qa diagram describes stable plasma operating region, limited by disruptions: − low li typically has to be negotiated during the plasma current ramp-up − high-li limit typically occurs due to excessive radiation at plasma edge,
resulting in cold edge plasma and narrow current channel (eg at density limit)
qa=2 limit
JET Limiter plasmas
MHD Stability - Plasma Equilibrium Limits
Page 29 CERN Academic Training Programme, 12-13 April 2011
• Experiments have shown that tokamak plasmas can sustain a maximum density: − limit depends on operating regime
(ohmic, L-mode, H-mode …)
− limit may be determined by edge radiation imbalance or edge transport processes
− limit can be disruptive or non-disruptive
• Comprehensive theoretical understanding still limited − “Greenwald” density:
nGW = I(MA)/ πa2
− operational figure of merit
JET
MHD Stability - Density Limits
Page 30 CERN Academic Training Programme, 12-13 April 2011
• Maximum value of normalized plasma pressure, β, is limited by mhd instabilities:
• Typically, “Troyon” limit describes tokamak plasmas:
βN ≤ 2.8-3.5
• More generally, “no-wall” limit: βN ≤ 4×li
!
"(%) = 100 < p >
B2 / 2µo
!
"N ="(%)
Ip(MA) / aB
Plasma MHD Stability – Pressure Limit: β
Page 31 CERN Academic Training Programme, 12-13 April 2011
• Essential problem is: − handle power produced by plasma with
(steady-state) engineering limit for plasma facing surfaces of 10 MWm-2
− extract helium from the core plasma to limit concentrated below ~6%
− prevent impurities from walls penetrating into plasma core
− ensure plasma facing surfaces last sufficiently long
Power and Particle Exhaust
Core plasma
Scrape-off layer (SOL) plasma: region of open field lines
Divertor targets
Private plasma
X-point
Page 32 CERN Academic Training Programme, 12-13 April 2011
• The divertor is a significant element of the solution − surfaces for high heat fluxes (10 MWm-2) − cryopumping to extract particles leaving the plasma, including helium
Power and Particle Exhaust
ITER divertor cassette – 54 cassettes make up the complete toroidal ring
Page 33 CERN Academic Training Programme, 12-13 April 2011
• The concept of divertor “detachment” is fundamental to power exhaust in the burning plasma environment: − additional impurities are added to
edge plasma to increase radiation − a large pressure gradient develops
along the field lines into the divertor − the divertor plasma temperature
falls to a few eV − a large fraction of the plasma
exhaust power is redistributed by radiation and ion-neutral collisions
0 1 2 3 4 5MW/m3
Deut.
Carbon
Radiationfrom
DIII-DRadiationLevelsDetachedRegion
0
1
2
3
4
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
Radius (m)
IRTV Divertor Heat Flux
Before Gas Puffing
After Detachment
Hea
t Flu
x (M
W/m
2 )
Power and Particle Exhaust
Page 34 CERN Academic Training Programme, 12-13 April 2011
• CFC divertor targets (~50m2): − high thermal conductivity and good
thermal shock resistance (doesn’t melt)
− but combines chemically with hydrogen (ie tritium)
• Be first wall (~700m2): − good thermal conductivity − low-Zi – low core radiation
• W-clad divertor elements (~100m2): − high melting point and sputtering
resistance − but might still melt during thermal
transients − will eventually replace CFC in entire
divertor
Plasma Facing Components - Challenges
Page 35 CERN Academic Training Programme, 12-13 April 2011
• Access to plasmas which are dominated by a-particle heating will open up new areas of fusion physics research, in particular:
− confinement of α’s in plasma − response of plasma to α-heating − influence of α-particles on MHD stability
• Experiments in existing tokamaks have already provided some positive evidence
− ”energetic” particles (including α-particles) are well confined in the plasma
− such particle populations interact with the background plasma and transfer their energy as predicted by theory
− but energetic particles can induce MHD instabilities (Alfvén eignemodes) - for ITER parameters at Q=10, the impact is expected to be tolerable
Burning Plasma Physics in ITER
Page 36 CERN Academic Training Programme, 12-13 April 2011
• In existing experiments single particle theory of energetic ion confinement confirmed: − simple estimate, based on banana
orbit width shows that Ip ≥ 3MA required for α-particle confinement
• Classical slowing down of fast ions well validated: − data range 30keV NBI (ISX-B) to
3.5MeV α-particles (TFTR)
• Energetic ion heating processes routinely observed in additional heating experiments
.
2.0
1.5
1.0
0.5
0.010-2 10-1 1 10
SLOWING-DOWN TIME (s)
EXPE
RIM
ENT
/ THE
ORY
****
ITER reference value
W W Heidbrink, G J Sadler, Nucl Fusion 34 535 (1994)
Energetic Ion Confinement
Page 37 CERN Academic Training Programme, 12-13 April 2011
• Electron heating by α-particles validated in JET and TFTR DT experiments: − in JET, maximum electron
temperature correlated with optimum fuel mixture for fusion power production
P R Thomas et al, Phys Rev Lett 80 5548 (1998)
Electron Heating by α-Particles
Page 38 CERN Academic Training Programme, 12-13 April 2011
• In a tokamak plasma, the Alfvén wave continuum splits into a series of bands, with the gaps associated with various features of the equilibrium: • a series of discrete frequency Alfvén eigenmodes can exist in these gaps:
• toroidicity-induced (TAE) gap created by toroidicity • ellipticity-induced (EAE) gap created by elongation • triangularity-induced (NAE) gap created by additional non-
circular effects
• beta-induced (BAE) gap created by field compressibility • kinetic toroidal (KTAE) gap created by non-ideal effects
such as finite Larmor radius … and others!
• These modes can be driven unstable by the free energy arising from energetic particle populations with velocities above the Alfvén velocity, eg α-particles
Alfvén Eigenmodes
Page 39 CERN Academic Training Programme, 12-13 April 2011
A Q=10 scenario with: Ip=15MA, Paux=40MW, H98(y,2)=1 Te
Ti
ne
10nHe
Zeff
fHe
qΨ
χe
keV
%
MA
/m2
1019m
-3 m
2s-1
Current Ramp-up Phase
-5
-4
-3
-2
-1
0
1
2
3
4
5
3 4 5 6 7 8 9
Z, m
R, m
Plasma start-up in PF Scenario 2
4.5MA (XPF)
0.5MA - XPF
0.5MA
1.5MA
2.5MA
3.5MA
-5
-4
-3
-2
-1
0
1
2
3
4
5
3 4 5 6 7 8 9
Z, m
R, m
Plasma start-up in PF Scenario 2
4.5MA (XPF)
XPF - SOF15MA (SOF)
Dashed line: reference plasma
ITER Plasma Scenario - H-mode
Page 40 CERN Academic Training Programme, 12-13 April 2011
• Discovery of internal transport barriers ⇒ “advanced scenarios”
• But development of an integrated plasma scenario satisfying all reactor-relevant requirements remains challenging
plasma with reversed central shear + sufficient rotational shear
internal transport barrier ⇒ enhanced confinement
reduced current operation + large bootstrap current fraction
reduced external current drive + current well aligned for mhd stability and confinement enhancement
active mhd control
Steady-state operation + High fusion power density
Steady-State Operation
Page 41 CERN Academic Training Programme, 12-13 April 2011
• Calculations of plasma profiles for a possible ITER steady-state scenario with Q=5
0
5
10
15
20
25
30
0.0 0.2 0.4 0.6 0.8 1.0
T, keV
X
1
0
ne, 1020m-3
Te
Ti ne
0.0
0.2
0.4
0.6
0.8
1.0
0.0 0.2 0.4 0.6 0.8 1.0
j, MA/m2
X
10
0
q
jbs
jtot
jnb+lh
q
(A Polevoi et al, IAEA-CN-94/CT/P-08, IAEA FEC2002)
ITER Steady-State Operation
Page 42 CERN Academic Training Programme, 12-13 April 2011
• To fulfill its missions ITER must carry out an ambitious and exciting physics programme
• Its essential design features give it the capability to do this: − pulse length and duty cycle − flexible heating and current drive systems
• total power • variety of systems
− diagnostic access and facilities − additional plasma engineering systems
• flexible fuelling systems • capability for control of important MHD instabilities
− equilibrium shape flexibility − divertor and first wall exchange capability
ITER has many assets as a burning plasma experiment and the key step towards the realization of fusion energy
Conclusions
Page 43 CERN Academic Training Programme, 12-13 April 2011
ITER Physics Basis, ITER Physics Expert Groups et al, Nucl Fusion 39 2137-2638 (1999)
Progress in the ITER Physics Basis, ITPA Topical Physics Groups et al, Nucl Fusion 47 S1-S413 (2007)
Hawryluk, R. J. Results from deuterium-tritium tokamak confinement experiments, Reviews of Modern Physics 70, 537 (1998).
Gibson, A. et al. Deuterium-tritium plasmas in the Joint European Torus (JET): Behavior and implications, Physics of Plasmas 5, 1839 (1998).
Keilhacker, M. et al. High fusion performance from deuterium-tritium experiments in JET, Nuclear Fusion 39, 209 (1999).
Jacquinot, J. et al. Overview of ITER physics deuterium-tritium experiments in JET, Nuclear Fusion 39, 235 (1999).
http://www.iter.org - and associated links
References: Tokamak Fusion Physics
Recommended