Enabling Research Project - EUROFUSION: Enabling research runaway

Mar 6, 2015 - −16. −12. −8. −4. 0. A. Stahl, E. Hirvijoki, J. Decker, O. Embréus and T. Fülöp,. Accepted for publication in PRL, http://arxiv.org/abs/1412.4608. 7 ...
686KB taille 3 téléchargements 355 vues
Enabling Research Project Tünde Fülöp and Ola Embréus, Eero Hirvijoki and Adam Stahl

Chalmers University of Technology

March 6, 2015

1

Ion runaway: Motivation

Low-mode number TAEs observed?

f ≈ 60 -260 kHz Resonant velocity:

vA /3 < vT e Runaway of ions responsible?

[Fülöp & Newton, PoP 21, 080702 (2014)]

2

Ion runaway: Intro

Largely analogous to electron

1

Non-monotonic friction

2

Non-monotonic distribution!

CODION:

F//

runaway, but:

eE

an extension of CODE to ions vc1

vc2

3

Ion runaway: Results

10

Initial dist. E=200 V/m 220 V/m 240 V/m 260 V/m

-1

fD / n D

10-2

10

-3

10 -4

10

-5

0

2

4

6

8

10

v|| / vTD

Typical runaway ion distribution

Bump energy too low to drive

exhibiting a large high-energy

Alfvénic instabilities:

bump.

vA /3 ∼ 30 − 50 vD 4

Guiding-center radiation-reaction force Lorentz-Abraham-Dirac force

The Lorentz-Abraham-Dirac force    3γ 2 γ2 3γ 2 2 ¨ + 2 (v · v) ˙ v˙ + 2 v · v ¨ + 2 (v · v) ˙ v v , c c c

Force on a particle:

K=

e2 γ 2 6πε0 c3



insert the Lorentz force into K, (no E − f ield)   p2⊥ e4 B 2 −1 −1 ˙ p −  ν Ω B B × p, with ν = K = −νr p⊥ + r r B (mc)2 6πε0 γ(mc)3

Guiding-center transformation: Z α = (X, pk , µ), · is the gyroaverage n o   −1 α −1 i α Tgc Ki , Kgc = Tgc x , Z Landau Approximation:

gc

5

Guiding-center radiation-reaction force Gyro-averages

˙ is the guiding-center velocity X  νr 2µB  ˆ ˙ + 3vk %k κ , = −B ? b × X Ωk mc2

Force on spatial position:

KX

Force on the parallel momentum:

Kpk = −νr pk

 p⊥ γ 2 µB 2 +  % τ −  ν %⊥ τB , B k B B r mc2 2

Force on the magnetic moment:



2µB K = −νr µ 1 + mc2 µ



 2 + B %k τB ,

Denitions:

%k =

pk p⊥ ˆ · ∇ × b, ˆ κ=b ˆ · ∇b ˆ , %⊥ = , Ω?k = Ω (1 + B %k τB ), τB = b eB eB

6

Critical electric eld for runaway electron generation: • • • •

Experiments seem to show E/Ec > 3 − 5 needed for RE generation We study the RE dynamics using CODE RE growth rate strongly temperature dependent at xed E/Ec Raises the eective critical eld 0

3000 1000

E/E

D =1

−8

300 100 30 10 1.2 2 3 5 10

−4

0.1

−12 0.0 2 0.0

−16

1

30

100

300

−1 log10 (n−1 e dnr /dt) [s ]

At Te = 24 keV, a eld of at least E/Ec = 35 is needed for signicant growth

Te [eV]

10000

−20

E/Ec

A. Stahl, E. Hirvijoki, J. Decker, O. Embréus and T. Fülöp, Accepted for publication in PRL, http://arxiv.org/abs/1412.4608

7

Critical electric eld for runaway electron generation: Synchrotron back-reaction Plots show change in growth rate

• Classical critical eld derivation • Synchrotron radiation

back-reaction eectively introduces additional drag

• Leads to reduction in growth

(Γw. synch /Γno synch ) 10000



3/2 B 2 Te /ne

• Important at high temperature

and low density

0.5 300

0.3

100

0.1 30

2

3

5

10

30

E/Ec 30

ne [1019 m−3]

• Strength of synchrotron eects

0.7

1000

rate which can be very strong for weak electric elds

• Raises the eective critical eld

0.9

a)

3000

Te [eV]

only considers Coulomb collisions

0.9

b)

10

0.7

3

0.5

1

0.3

0.3 0.1

0.1 2

3

4

5

6

7

8

9

10

E/Ec

8

Critical electric eld for runaway electron generation: Growth-to-decay transition

• Experiments also considered transition from RE growth to decay • Build up RE tail, then ramp down E/Ec (ramp up density) • Visual synchrotron and HXR signals shows transition at E/Ec = 35 •

Simulations show transition at only slightly above

Ec

( 1.1)

E/Ec

BUT Synchrotron emission agrees with experiments!



Emitted synchrotron power sensitive to



Runaways are still gaining energy when the emission declines

4.2

1.3

B = 2.5

T

a)

0 1 (a.u.)

not RE decay but

7.4

E RE

Observed reduction is

redistribution of REs in momentum space

B

10.6

.5 T =3

B = 1.5 T

particle energies and pitches



12

1

Psynch.



12

0

b) 0.5

1

1.5

2

Time (s)

9

Critical electric eld for runaway electron generation: Summary

Experiments showing increased threshold electric eld for runaway growth can be explained by:

1

Temperature dependence of RE growth rate at xed E/Ec

2

Dampening eect of synchrotron emission (to some extent)

Apparent elevated RE growth-to-decay threshold is likely an artifact due to:

• Redistribution of runaways in momentum space as the eld

strength decreases

=⇒

• Change in emitted synchrotron and bremsstrahlung power