PROJECT EPSILON – A WAY TO STEADY STATE HIGH FUSION REACTOR
V.M. Kulygin, V.V. Arsenin, V.A. Zhil’tsov, A.V. Zvonkov, A.A. Skovoroda A.V. Timofeev
Russian Research Center “Kurchatov Institute”
Nuclear Fusion Institute
IAEA XXI Fusion Energy Conference, 16 -21 October 2006, Chengdu, China
Abstract• Pseudosymmetry (PS) principle being used for hot plasma
confinement magnetic systems construction could provide realization of all the magnetic confinement potential. Namely: to realize hot plasma confinement at steady state operational mode with a “tokamak” confinement time level. The high regime lets a possibility to think about advanced fuels (D- 3He, D-D).
• It is proposed to check the main idea and preliminary theoretical examination experimentally in frame of EPSILON project – closed crimped magnetic trap without rotational transform and with poloidal PS. The EPSILON consists of two (or more) crimped mirror traps (OME), closed with curvilinear elements (CREL). MHD stability and plasma cleaning are provided with axially symmetrical divertors disposed along the mirror traps with zero magnetic field at separatrix.
Introduction • The fusion parameters of magnetically confined hot plasma will be
got in ITER – the first experimental thermonuclear reactor that is now prepared for construction.
• As the nearest principal goal of magnetic fusion development is coming near, next stage problems become actual. They are the problems of magnetic confinement system optimization from reactortechnology point of view.
• Power reactor should have high power, technology and economic efficiency, be serviceable and meet safety and ecology demands.
• Because of the reason, world fusion development efforts along with the largest tokamak experiment realization should take aim at search for possible improvements of magnetic confinement systems. The improvements can consist in inherent steady state high operation, technology simplification, efficiency increasing and possibility to use ecologically attractive fuels (D-3He).
EPSILON is Closed Mirrors
• Closed Mirrors = Linked Mirrors, closed magnetic systems without rotational transform
• All previous attempts to use Linked Mirror configuration were unsuccessful
• Now the reason of that is clear: all of the previously tested systems contravened the principle of the same type pseudosymmetry for each element
Kadomtsev’s Rippled Trap (1958)
KADOMTSEV B.B., Magnetic traps with rippled magnetic field, Plasma Physics and the Problem of Controlled Thermonuclear Reactions, Ed. by M.A.Leontovich (Pergamon Press, New York, 1960). Vol.3.
Summary:At H0/Hm <<1, particle losses in curvilinear elements can be comparable with classical losses in rippled solenoids
EPSILON magnetic configuration
The main Improvement: Toroidal drift with correspondent particle losses iseliminated. The ripped magnetic configuration in the solenoids is arrangedWith coils of opposite current (toroidal divertors)
OME sell
EPSILON fundamentals
• Pseudo symmetrical geometry of a magnetic configuration
• Nested drift surfaces• Practical absence of rotational transform• MHD stability is provided by a magnetic
“hump” instead of a magnetic “well”, divertor stabilization
• High β
Pseudosymmetry Principle
• The PS General Condition
Φ is toroidal magnetic flux, F is a bounded function
F=F(ρ,θ,B) – weak isometry or quasiisodynamic (QID)F=F(ρ,θ) – isometryF=F(ρ) - quasisymmetryF=0 - isodynamic or orthogonality condition
The PS physical meaning: all |B| =const contours on the equilibrium magnetic surface encircle either the magnetic axis (Poloidal PS) or the
major axis of the torus (Toroidal PS)
No superbananas in that case
F
B
B
B
B
SKOVORODA, A.A., 3D Geometry of Currentless Magnetic Configurations with Improved Plasma Confinement, Plasma Phys.Control.Fusion 47 (2005) 1911
Kinds of symmetry
Toroidal symmetry
Red circles – isomagnetic lines
Poloidal symmetry
Red circles – isomagnetic lines
Poloidal pseudosymmetry
Isomagnetic lines
Magnetic axis
PS system
Breaking of the PS
Divertors influence on particles confinement.
The collisionless transversal particle losses take place only from rather thin plasma layer (its thickness is of Larmor radius, ρ, order of value) close to edge magnetic surface (of confinement region). For fusion products ρ is relatively large, but they are born rather far from the separatrix. For some little fraction of passing particles the losses because of non adiabaticity can be important. That can occur if such a particle endures displacement from one magnetic surface to another during passing through a CREL. It takes place in a layer with a thickness of ρ order of value as well.
Plasma equilibrium in mirror trap with a
divertor Plasma equilibrium in mirror trap with a divertor was calculated foranisotropic plasma on base of generalized Grad – ShafranovEquation and taking into consideration longitudinal equilibrium equation(which binds and )
The result is: presence of a divertor don’t prohibit from equilibriumachievement at , where the value should be calculatedusing plasma pressure and vacuum magnetic field in a divertor centre. It should be noted that because of unlimited rising of
near the separatrix, the equilibrium isn’t destroyed by CRELs influence.
)||( pp
p ||p
dlBU 1
1~
ARSENIN, V.V., KUYANOV, A.YU., Plasma Equilibrium in Axisymmetric Open Divertor Configurations, Plasma Phys. Rep., 26 (2000)741.
U
Plasma stability questions
0|| 2
U
UpUpwMHD
00 U
Upp
Kadomtsev – Bernstein MHD flute mode stability condition
3/5
Marginally stable profile
KADOMTSEV, B.B., On convective instability of plasma, Plasma Physics and the Problem of Controlled Thermonuclear Reactions, Ed. by M.A. Leontovich (Pergamon Press, New York, 1960). Vol.4.
Plasma stability questions II
Displacement in balloon disturbance
1 – long wave disturbance which defines in MHD model;
2 – disturbance with zeros in each divertor, which defines with account of trapped particles;
3 – antisymmetric mode localized in divertor;
22
||~
L
acr
1 22
||~
l
acr
2
ARSENIN, V.V., KUYANOV, A.YU., Nonparaxial plasma equilibria in axisymmetric mirror, Trans. Fusion Sci. Tech., 39 1T (2001) 175.
About Rotational Transform • Project EPSILON does abandon in fact the rotational transform. MHD
stabilization is provided by plasma compressibility effects
• Magnetic surface topological stability against magnetic field disturbances is provided in reality because any such disturbance generates unipolar longitudinal plasma current, which results in small rotational transform that provides topological stabilization of magnetic surfaces.
• The plasma compressibility stabilizing effect will be kept at presence of such rotational transform up to values
where - ion Larmor radius, a – minor radius, R – major radius,
2~a
Ri
i
Some problems of high β High β could raise some problems. For instance we should meet the QID conditions for passing particles. At low average β the QID conditions are realized with magnetic hump conditions. At high β those can be provided with magnetic well. But there is some intermediate range of the parameter where the J contours are fully destroyed that leads to high losses. So, high mode of operation should be achieved with some force crossing of that barrier.
Radial and azimuthal dependence of the longitudinal invariant J for trapped particles with different pitch angles in closed mirror trap
at <β> = 0.1%. at <β> = 8%.
EPSILON magnetic configuration
а)
b)
(b) Magnetic surface (U=const, above) and drift surface for strongly passing partcles (full length of magnetic force lines L=const). Two cross sections for CREL (different curvature positions) and for central section of solenoid. External circles – turns with a current.
(a) EPSILON magnetic configuration with plain magnetic axis and divertors in solenoidalparts. Color of “near separatrix” magnetic surface depends on magnetic field module value (minimal magnetic field at solenoid axis 0.1T).
EPSILON Reactor Features
• Steady state operational mode. Current drive is not necessary i.e. energy injection systems are necessary only for ignition stage.
• High β mode of operation provides more flexibility for in D-T reactor reactor realization and exploitation. Advanced fuel cycles (D-3He and D-D catalyzed) can be considered for EPSILON type reactor in connection with high β operational mode.
• Magnetic system structure additivity. Each OME length can be prolonged without principal limitations that let increase reactor installed nuclear capacity.
EPSILON Reactor Features II
Specific power, MW/m3, in D-T reactor at T=15 keV, B=5T,as a function of β. It's increasing (limited by wall thermal load) lets to get more compact reactor. EPSILON geometry causes more uniform power flow to the first wall
0,00 0,05 0,10 0,15 0,20 0,25 0,300,01
0,1
1
10
spec
ific
pow
er, M
W/m
3
beta
EPSILON Reactor Features III
Critical confinement time tE, [s],
as function of β at B=5T.
1) D-T,2) D-D cat.,3) D-3He
Critical triple product, ntET, [m-3s keV] as function of β at B=5T.
1) D-T,2) D-D cat.,3) D-3He
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,90,1
1
10
100
1000
3
2
1t E, s
beta
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,91E21
1E22
1E23
1E24
1E25
3
2
1
nT
t E, [m
-3 k
eV
s]
beta
Proof-of-Principle Experiment EPSILON-M (Proposal)
1 Plasma volume, m3 1.2
2 Average plasma diameter in solenoids, m 0.3
3 Main axis dimensions, m 1.6 x 5.2
4 Maximal plasma density, m-3 1019
5 Plasma temperature, keV 0.5
6 Confinement time, s 0.005
7 Average in solenoids 0.1
8 Magnetic field in solenoids, T 0.1
9 Magnetic field in CRELs, T 1
10 Heating power, MW 0,5
Mode of operation – steady state. Working gases – hydrogen, deuterium. Plasma heating methods: ICR, ECR, LH. Magnetic system – water cooled copper coils
Conclusion• Theoretical investigations show that the proposed magnetic
confinement system should provide mirror confinement of majority of plasma ions in practically axisymmetric mirror traps (solenoids with set of divertors) and confinement of significant part of passing particles. Plasma stability should be also provided by the divertor system.
• Previous mirror experimental activity can be used for EPSILON development. In particular it is possible to use the ambipolar methods of particle and energy confinement which are well developed and checked.
• The CRELs have a service function of energy loss diminishing. They should liquidate thermal plasma loss to a wall along magnetic lines and solve the problem of electron temperature.
• Due to high β operational mode and specific structure, the system seems to be rather attractive from its possible reactor perspective point of view.