Everything you always wanted to know about fusion reactors,
but were afraid to ask
Minh Quang Tran
Swiss Plasma Center
Ecole Polytechnique Fédérale de Lausanne-‐
Switzerland
11.10.15 ICFRM 17-‐ Tutorial Session
Be ware of my plagiarism
• Everything You Always Wanted to Know About Sex * But Were Afraid to Ask (1972)
• Director: Woody Allen
Plan
• More will come with the talk on DEMO by Dr. R. Wenninger
• IntroducOon: The energy issue
• The Physics basis: reacOons and fuels
• Why fusion is considered as a “disrupOve energy”?
• Some (not all) issues
• ITER
• Technology besides Materials
• Road map towards a fusion reactor
• InerOal confinement
• 15 minutes for Q&A: Everything you always wanted to know about fusion reactors, but dare to ask
11.10.15 ICFRM 17-‐ Tutorial Session
Introduction: The energy problem
• The constraints:
1. Increase of world population and therefore energy needs: 7.3 billions in 2015 to 8.9 billions by 2050, remaining stable beyond (UN study), coupled with a today inequality in energy access (inverse champagne glass)
2. Change in energy mix requirement: stronger reliance on electricity for an increasing urban population
3. Necessity to have “sustainable” energy “Development that meets the needs of the present without compromising the ability of future generations to meet their own needs” ( Brundtland report)
o Environment aspects: global warming
o Safety: accidents should not impose population evacuation
o Legacy towards next generations: depletion of fossil fuels; waste repository on a “human” (not geological) time scale
Electricity
consumpOon/ capita
• World bank data (h\p://data.worldbank.org/indicator/
EG.USE.ELEC.KH.PC/countries/all?display=graph)
• World 3064 kWh/capita
• EU: 6144 kW/h/ capita; Germany: 7270 kWh/
capita ; Switzerland: 7343 kWh/capita
• China: 3810 kWh/capita; India: 744 kWh/capita
• Vietnam: 1273 kWh/capita, HaiO: 50 kWh/
capitaà = 0.007 of Germany
11.10.15 ICFRM 17-‐ Tutorial Session
Fusion reactions
• Fusion reactions: “Easiest” to achieve reaction:
D + Tà He ( 3.5 MeV) + neutron
(14.1 MeV)
Other reactions
D + D à He3 ( 0.82 MeV) + neutron
(2.45 MeV
D + D à T (1.0MeV) + H ( 3.0MeV)
D + He3à He4 (3.76 MeV) +p (14.7MeV)
Energy release E = Δm * c2
2-3
n
Fission products
U
Fusion cross section
1 keV à T = 10 millions degrees through the relaOon kBT = E
11.10.15 ICFRM 17-‐ Tutorial Session
Lawson criterion (1)
• Fusion power release > Power loss
• Ions have a Maxwellian distribuOon f(E) characterized by a temperature T
• CalculaOon of the fusion reacOvity <σv> , which is average over the distribuOon funcOon
• Triple product Density n* Temperature T*Energy confinement Ome τE > 5* 1021 m-‐3keV s
Lawson criterion (1a)
• Plot of <σv> versus T
11.10.15 ICFRM 17-‐ Tutorial Session
DerivaOon of Lawson
criterion (1)
DerivaOon of Lawson criterion (2)
11.10.15 ICFRM 17-‐ Tutorial Session
Lawson criterion (2)
• Q definiOon= Fusion power/ External HeaOng power
• But the fusion reacOons provide energeOc He ions (3.5 MeV) which can thermalize with the D and T ions (10-‐20 keV): He ions is a source of heaOng
• Qà infinity (igniOon) if External HeaOng power is null: all needed heaOng is provided by He ions
• For a reactor, igniOon is not required Q = 30-‐40
The challenge of fusion
Density n* Temperature T*Energy confinement Ome τE > 5*
1021 m-‐3keV s The challenge:
1. To create a plasma with n about 10 20 m-‐3 and T about 108K, i.e. 10 keV
2. To confine its energy during τE of a few seconds
There are many Ome scale: ParOcle confinement
Ome τP, Plasma duraOon, Energy confinement Ome
11.10.15 ICFRM 17-‐ Tutorial Session
Fuel issues
• D: abundant as natural isotope of H ( 1/7000)
• T: Short life ( 12.3 years) radioacOve ( β decay at about 5.7 keV)
• Needs to “ breed” triOum using the neutron from the fusion reacOon D+ T à n + He
6Li + nà T + 4He + 4.8 MeV (exothermic)
7Li + nà T + 3He + n – 2.5 MeV (endothermic)
• Importance for the “Breeding Blanket” in a reactor
Fuel issues (a)
• Cross secOon of the fusion reacOon of n and 6Li and 7Li
11.10.15 ICFRM 17-‐ Tutorial Session
Fusion : a
“disruptive” energy
• Why a “disruptive” energy?
(1)
Li is abundant in the crust and in the sea water ( WEC 2013)
2) No chain reactions; no severe accidents: fuel inside the reactor sufficient only for a few
minutes (no “Tchernobyl” type accident);
Fusion : a
“disruptive” energy
• (3) Environmental friendly
(2)
!"#
$!"#
%!"#
&!"#
'!"#
(!"#
)!"#
*!"#
+!"#
,!"#
$!!"#
!"#$%&'()
-./%0123#
$!#
$!!#
$!!!#
''# %&# %$# $+# (,#
!)"#$%&'() ,,!#
11.10.15 ICFRM 17-‐ Tutorial Session
Fusion : a
“disruptive” energy
4. Development of low activation materials ( Cf.
(3)
Talk by A Moeslang and A. Kimura)à no need of geological repository
5. This contributes as well to the safety aspect by reducing the “after heat”
11.10.15 ICFRM 17- Tutorial Session
Market penetration
• Economics studies confirm the possibility for fusion to penetrate the market
0 5 10 15 20 25
base 750 650 550 450
CO2 target (ppm)
Electricity Production (EJe) solar
wind biomass fusion fission gas coal hydro
0 105 1520 2530 35 4045
Coal Gas Nuclear fission Biomass Photovoltaics Wind Fusion
External Costs [mEuro/kWh]
Global Warming Other
Sum
State of ma\er at T>
10
5K , about 10 eV
• Binding energy e-‐ion: about 10 eV
• At fusion temperature, the state of ma\er is
plasma, i.e. a “gas” formed by electrons and ions , globally neutral and dominated by “collecOve”
effect
• Debye shielding: a charge is surrounded by a cloud of opposite charges which “shield” its Coulombian potenOal Vc. Beyond a few Debye lengths λD, Vc is no longer felt. λD = (ε0kBT/n0e2)1/2
Confinement
• Due to the Debye screening, electrostaOc confinement is not possible: electric field is shielded aqer a few Debye length
• Two possibiliOes:
u No confinementà InerKal confinement and
realisaOon of the triple product through very high density n ( very short τE and τP )
u By magneOc field à MagneOc confinement
11.10.15 ICFRM 17-‐ Tutorial Session
MagneOc confinement (1)
• Through the Lorentz force Florentz = q(v x B )
• B is generated either or both current by external coils or by the plasma it self
∇ × B = µo j
f = Force density in fluid description =ρel
(
j × B)
MagneOc confinement (2)
• Simple toroidal magneOc field (closed field lines) created by a a wire is not sufficient : parOcles
“driq” across magneOc field due to curvature and spaOal variaOon (1/r) of B: the driq direcOon
(verOcal) depends on the charge, leading to charge separaOon and hence a verOcal electric field E.
This E combined with the B leads to a global driq of both charge species according to (E x B),
leading to loss of confinement
11.10.15 ICFRM 17-‐ Tutorial Session
Tokamak
• Confinement is provided by a toroidal magneOc field BT provided by external toroidal coils( TF
coils) and by a current carried by the plasma itself IP.
Tokamak
11.10.15 ICFRM 17-‐ Tutorial Session
Tokamak field lines and
aspect raOo
T3 (USSR)
11.10.15 ICFRM 17-‐ Tutorial Session
1968: 1 keV confirmed by a team of scienOsts from UKAEA ( cold war). It
opens the era of tokamak
Confinement modes
H mode: High energy
confinement mode
H à L mode transiOon
11.10.15 ICFRM 17-‐ Tutorial Session
Divertor
Divertor: Where parOcle and energy from the
plasma are removed.
Divertor plate: W
High heat flux material Cf. Talk by Dr. S. Lisgo
Divertor:
• 54 Divertor cassettes
• High heat flux components capable of 10MWm-2 in stationary operation and 20MWm-2 transiently
CFC
W – “reflector plates” DIVERTOR
Divertor (ITER)
Heat load up to 20 MW m-‐2
11.10.15 ICFRM 17-‐ Tutorial Session
Divertor:
• 54 Divertor cassettes
• High heat flux components capable of 10MWm-2 in stationary operation and 20MWm-2 transiently
CFC
W – “reflector plates”
Divertor (ITER)
Stellarator
• 3D magneOc confinement created exclusively by external magneOc coils. No plasma current: no disrupOon
11.10.15 ICFRM 17-‐ Tutorial Session
W7 X stellarator
HeaOng (1)
• In the case of a tokamak where we have a plasma current IP, what is the heaOng by this current?
• What is the resisOvity of a plasma? In the keV regime, it is like Cu, but it decreases as T-‐3/2.
• Ohmic heaOng , taking into account
phenomenological loss rate, cannot bring a
tokamak plasma to the 10-‐20 keV regime. The temperature will be about 4 keV ( Freiberg)
Heating (2)
• Heating by absorption RF waves or by injection of fast neutral particles, which thermalize with the plasma particles
RF waves at ion cyclotron frequency( about 50 MHz) or electron cyclotron frequency ( about 150-200 GHz)
Injection of fas t neutral partciles (1 MeV)
11.10.15 ICFRM 17-‐ Tutorial Session
Current drive (1)
• What is current drive? The toroidal plasma IP is an essenOal component of a tokamak
IP is induced as the current in a transformer. So it cannot be
sustained in steady state
Current drive (2)
• Non inducOve current drive can be achieved by
injecOon of parOcle momentum ( neutral beam) or by preferenOally injecOng EM waves using
Doppler shiqed resonance e.g. For electron cyclotron wave: ω = Ωce +/-‐ k//v// or by
manipulaOng the pressure
• Example of full non inducOve current drive in a medium size tokamak
11.10.15 ICFRM 17-‐ Tutorial Session
EX/W-5
2
analysis is performed with eight energy bins available for each chord, with adjustable thresholds within the 10-200 keV range.
The second harmonic X-mode ECE radiometer observes the plasma along one of three possible horizontal viewlines, two on the high field side and one on the low field side, and operates in the 78-114 GHz range with 24 channels of 0.75 GHz bandwidth [6]. The EC radiation observed on the high field side is dominated by relativistically downshifted emission by the high energy end of the electron distribution function and can thus be employed to diagnose the suprathermal population.
The quasilinear Fokker-Planck code CQL3D [7] is employed to model the dynamics of the elec- tron distribution function. The code is coupled to the TORAY-GA ray-tracing module [8] and solves the Fokker-Planck equation in two velocity and one spatial dimensions. The equation in- cludes a quasilinear EC wave damping term, a relativistic collision operator and a model for ra- dial diffusion, with an optional linear dependence on the parallel velocity and a particle- conserving advection term.
2. ECCD and suprathermal electrons
The ability to control the deposition location and toroidal injection angle accurately is instru- mental in the application of ECCD to current profile tailoring. This high degree of control was clearly demonstrated in TCV by sustaining the non-inductive plasma current with two X2 gy- rotrons at the time, and firing two sets of two gyrotrons in succession for their maximum pulse durations (2 s). As shown in Fig. 1, the TCV discharge length was thus extended to a record 4.3 s, well beyond the maximum length achievable in Ohmic conditions. Matching the powers, dep- osition locations and parallel wave numbers of the two sets of beams is essential for a smooth switch-over. This external control was further demonstrated by an interlaced square-wave mod- ulation of the two clusters (180 degrees out of phase), with no visible resulting modulation of the plasma parameters.
500 100
Plasma current (kA)
0 0.5 1 ECRH power cluster A (MW)
0 0.5 1
ECRH power cluster B (MW)
−1 0 Loop voltage (V) 1
6
7 Ohmic transformer current (kA)
1 1.5 Edge elongation
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
1.5
2.5 Internal inductance
Time (s)
FIG. 1. TCV discharge 20881 of record length (4.3 s), sustained by 0.9 MW ECCD.
!
Physics issues (1)
• A magneOcally confined plasma contains free energy which can be released as instabiliOes
• Example: Consider a tokamak as a levitated ringin a magneOc field topology. Eanrshaw theorem
indicates that the equilibrium is not stable. One degree of freedom is unstable: in a tokamak it is called the VerOcal Displacement Event VDE,
leading , if uncontrolled to disrupOon
Physics issues (2)
• Another example: The confinement of the 3.5 MeV He ions produced by the fusion reacOons
• These energeOc ions may be lost by interacOon
with waves ( Alfven waves) excited in the plasma, before thermalizing with the D and T ions.
• Heat removal in divertor
• …. And many more
11.10.15 ICFRM 17-‐ Tutorial Session
Material science issues
• A very exciOng field to deliver materials which ü Have the necessary thermo-‐mechanical
properOes under irradiaOon
ü Are compaOble with the operaOon of magneOcally confined plasma
ü Fulfil the promises of waste disposal
• In itself it is an exciOng field
11.10.15 ICFRM 17-‐ Tutorial Session
Space
Time
1-30 nm, 1-10 ps
1-5 nm, 2 ps - s 1 nm - 1 µm, 10 ps - s
10 nm - µm, ns - s
10 nm - 10 µm, µs - hours
DDD
kRT
0.1 nm - 1 m, 1 ps - years
MD
Formation energies
of point defects 0.1 nm
Ab Initio
kMC FEM
0 50 100 150
-250 -200 -150 -100 -50 0 50 100
K (MPa m1/2)
T(˚C)
6 2 1
3
4 5
1. RT, unirradiated 2. RT, 0.37 dpa 3. RT, 0.93 dpa 4. 523 K, unirradiated 5. 523 K, 0.30 dpa 6. 523 K, 0.75 dpa 800
700 600 500 400 300 200 100 0
σ [MPa]
10 9 8 7 6 5 4 3 2 1 0
ε[%]
Ttest = Tirrad.
1 µm - 1 cm Kinetic rate
theory
Discrete dislocation dynamics
Finite element modelling
Kinetic Monte Carlo Molecular dynamics
Atomic displacement cascade
Stacking fault
tetrahedron Interaction edge dislocation- void
Nanocrystal
TEM simulated image TEM image
TEM image
TEM image
kRT equations Interaction
dislocations- defects
Tensile tests
Creep tests Fracture toughness versus temperature Bend bar with
FEM mesh
ITER objectives (1)
• ITER objectives:
1. Produce P fusion = 500 (360) MWth of during 400
(3000) s with an external additional heating Pheating 50 MW (Power gain Q =P fusion /Pheating = 10)
2. Study physics of a “burning” plasma, i.e. when the energetic 3.5 MeV He nuclei from fusion reactions are confined and provide a dominant heating power (100 MW compared to the 50 MW of external
heating)
ITER objectives (1)
3. Integrate in a single device the different
technologies (e.g. superconductivity SC, heating methods and all associated power electronics) and the physics constraints
4. Prove the safety aspects of a fusion reactor: ITER is the first fusion reactor to be licenced as a nuclear
reactor
11.10.15 ICFRM 17-‐ Tutorial Session
Extrapolation for
ITER
The ITER tokamak
Plasma
Current IP :15 MA Major radius R= 6.2 m
Plasma radius a = 2m
R/a= 3.1
Fusion power: 500 MW
Pulse : 400s SC Poloidal
field (PF) coils
SC Toroidal field (TF) coils BT= 5.3 T
SC Central solenoid (CS)
Generates IP
Ports for Heating &
Current Drive and Diagnostics
R
11.10.15 ICFRM 17-‐ Tutorial Session
ITER safety
• Aqer the Fukishima accident, ITER also went through the so called “stress test” as any other nuclear reactor
ITER was licenced as “réacteur nucléaire de base”
according to the French nuclear legislaOon in2012, becoming the first fusion reactor to be licenced
Superconductors
• Coils to create the magneOc configuraOon of ITER are superconducOng (either Nb3Sn or NbTi)
11.10.15 ICFRM 17-‐ Tutorial Session
NB3Sn condcutors for CS ( leq) and TF (right) coils
Toroidal field coils (
ITER)Some features of ITER
• ITER will include Test Blanket Modules TBMs, which are mock-‐ups of Breeding Blanket for a reactor
11.10.15 ICFRM 17-‐ Tutorial Session
EU HCLL and HCPB TBM
TBM
He-Cooled Ceramic Breeder concepts F proposal to install a specific-design TBM (China, EU, India)
F proposal to contribute with a specific- design sub-module in other Parties TBM (Korea, Japan, RF, USA)
Lithium-Lead concepts
F Helium-Cooled design (EU)
F Dual-Coolant (He+LiPb) design (US, India) F Dual-Functional design, which is initially a HCLL evolving later to DCLL (China)
Water-Cooled Ceramic Breeder concept F specific-design TBM (Japan)
Molten Lithium concepts
F Self-Cooled design (SCLi) (RF)
F He-Cooled design type (HCLi) (Korea)
Solid Breeders Designs Liquid Breeders Designs
TBM
TBMs tests need a whole TBM system
T B M P O R T
PFC (ITER)
• 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
− melting during VDEs
• W-clad divertor elements (~100m2):
− high melting point and sputtering resistance
− but might still melt during thermal transients
− will eventually replace CFC
11.10.15 ICFRM 17-‐ Tutorial Session
HeaOng and Current Drive in ITER
• ITER will have 3 methods to heat and perform non inducOve current:
• Electron cyclotron wave at 170 GHz and 20 MW power deposited to plasma
• Ion cyclotron wave in the frequency range of 55 MW and 20MW at plasma
• Neutral beam injecOon at 1 MeV and about 30 A
Where are we?
• How far are we from the goal, regarding nTτE?
11.10.15 ICFRM 17-‐ Tutorial Session
Another way to ask the quesOon
(courtesy of IO)• Electrical power consumpOon to answer the quesOon: Steady state:120 MW conOnuous power consumpOon,180 MVA connected loads (mainly motors), During plasma pulse: 500 MW peak pulse consumpOon, 2.2 GVA connected power converters
The roadmap towards fusion
• The roadmap is NOT a single machine but rather a programme:
1. Build and exploit ITER
2. A programme on material based on an Early Neutron Source /IFMIF
3. PreparaOon of DEMO to be operaOonal by 2050 (Cf Talk by Dr. R. Wenninger)
11.10.15 ICFRM 17-‐ Tutorial Session
The Roadmap
Present
experiments +
JET 15 MWth
ITER
In construction 500 MWth
DEMO
Preconceptual design
500 MWe
Fusion
Power Plant
1.5 GWe
2025-20 30
2050
2015
Man
From Chinese Road map:
DEMO: 2030 FPP: 2050
JET
ENS/IFMIF
Why do we need a dedicated n source?
• The fusion reacOons produces neutron with a well defined energy of 14 MeV and hence to test
material one needs a high flux and fluence source close to this energy ( Cf. Dr. J. Knaster talk P9)
§ TransmutaOon; Frenkel pair formaOon; He and H embri\lement (56Fe(n,α)53Cr (incident n threshold at 2.9 MeV) and 56Fe(n, p)56Mn (incident n threshold at 0.9 MeV)
§ InteracOon of the 14 MeV with material (Cf. Dr. A.
Kimura presentaOon)
11.10.15 ICFRM 17-‐ Tutorial Session
InerOal fusion
• The plasma is NOT confined. Its expansion rate is given by the ion acousOc speed cs = (kBT/ Ion
mass)0.5
EM wave interacOon with plasma
11.10.15 ICFRM 17-‐ Tutorial Session
Interaction zone
InteracOon zone
The prospect of inertial fusion energy derives from scientific advancements in different arenas
Fast Ignition/Shock Ignition Indirect Drive Laser Fusion:
Central Hot Spot Ignition
Direct Drive Laser Fusion:
Central Hot Spot Ignition
Heavy Ion Fusion
How ICF could be achieved
11.10.15 ICFRM 17-‐ Tutorial Session
The National Ignition Facility (NIF) provides the
opportunity for ignition physics research at full scale
° ° ° °
Laser Beams (enter through
laser entrance
hole (LEH)
Hohlraum (laser target) Coupling: laser energy couples to
hohlraum and converts to x-rays Drive: x-rays bathe capsule,
heating it up -- it expands
• conservation of momentum: ablated shell expands outward, rest of shell (frozen DT) is
forced inward
Fusion initiates in a central hot spot and a burn front propagates outward
Fuel DT
Tr (eV)
Symmetry: radiation compresses capsule and it implodes
Time (ns)
11.10.15 ICFRM 17-‐ Tutorial Session
NIF laser amplifiers
11.10.15 ICFRM 17-‐ Tutorial Session
The hohlraum
11.10.15 ICFRM 17-‐ Tutorial Session
11.10.15 ICFRM 17-‐ Tutorial Session
Conclusion (1)
• Fusion is a mulO generaOon endeavour.
• To the quesOon : is this worth devoOng your career as a material scienOst?
• From a personal perspecOve, at the end of my
career , without any regret, I would join again this quest
• The only thing I would do differently is to study material sciences, to be able to solve the most challenging field of fusion research: a no –no go.
Conclusion (2)
22nd World Energy Congress, Daegu 2013
Capturing the Moments
20 November 2013
“The fusion challenge is much bigger than Apollo … It’s like a mission to Mars or jumping from the Wright brothers airplane to the jet engine.” It is generally agreed that the middle of this century is a realisOc Omeline for commercial scale fusion energy, though some it can happen faster.
-‐ Nebojsa Nakicenovic, Deputy Director
& Deputy CEO of IIASA
11.10.15 ICFRM 17-‐ Tutorial Session