Field line breaking and Microtearing Turbulence · Les Houches, March 9 2011. A brief history of...
Transcript of Field line breaking and Microtearing Turbulence · Les Houches, March 9 2011. A brief history of...
Field line breaking andMicrotearing Turbulence
Hauke Doerk
Thanks to F. Jenko and the GENE team
Max-Planck-Institut für Plasmaphysik, Garching
Winter School on dynamicsand turbulent transport in plasmas and conducting fluids,
Les Houches, March 9 2011
A brief history of microtearing research
• 1968: Tearing instability (Furth, Killeen, Rosenbluth)
• 1975: Instability due to ∇Te: µ-tearing (Hazeltine et al.)
• 1980: Model for saturation (Drake et al.)
• 1990: µ-tearing should be stable for realistic tokamakscenarios (Connor et al.)
• 1999: Focus on µ-tearing in plasma edge (Kesner et al.)
• 2003: Linear gyrokinetic simulations (Redi et al., Applegateet al.); Large electron heat transport in spherical tokamakscaused by µ-tearing?
• 2008: µ-tearing modes also found in conventional tokamaks(linear GK, Vermare et al., Told et al.)
• 2010: Nonlinear Gyrokinetic simulatios (Guttenfelder, thiswork)
Scope of this work
Problems
• Existence of microtearing instability in Tokamak geometry
• Electromagnetic heat transport caused by microtearing
• Nonlinear saturation of microtearing turbulence
Strategy
• Linear and nonlinear simulations using GENE
• Examine impact of steeper gradients, collisional effects...
• Comparison to analytical models
The GENE codeGyrokinetic Electromagnetic Numerical Experiment
Solves gyrokinetic equations on fixed grid in 5D phase space(⇒continuum code)
• Comprehensive physics
• Massively parallel
• Open source
http://gene.rzg.mpg.de
Characteristics of µ-tearing modes
Ballooning representation
• Fluctuating electrostaticpotential φ extends alongfield line
• Vector potential A‖ isstrongly localized aroundθ = 0
µ-tearing modes found in
• Spherical tokamaks(NSTX, MAST)
• Conventional tokamaks(ASDEX Upgrade)
• Model geometry: Circular(Lapillonne et al. 2009)
-10 -5 0 5 10
|A‖|,
ℜ(φ)(a.u.)
ballooning angle θp/π
|A‖|ℜ(φ)
Turbulent field fluctuations
Vector potential A‖
• large structures
Electrostatic potential φ
• small scale eddies
Microtearing turbulence is hard to resolve innumerical sumulation
Standard theory for fluctuation amplitude
Drake et al. 1980
• γlin ∼ DMk2⊥
• ⇒ δB/B0 ∼ %e/LTe
• simplified modelunderlying
Challenge
• 384×64×24×32×16grid points inx , y , z, v‖, µ space
• ∼ 25000 CPU hoursper run
0
1
2
3
4
5
6
0 1 2 3 4 5 6(δB
x/B
0)/(ρ
e/R
)
R/LTe
Gene simulationsδB/B0 = ρe/LTe
lowered resolution as a convergencestudy: scatter is moderate
Can gyrokinetic codes be used to prove or refine Drake’smodel?
Influence of temperature gradients
Ions
• R/LTi notimportant
Electrons
• RLTe
= − RTe
∂Te∂x
crucial
•(
RLTe
)crit∼ 1.3
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 1 2 3 4 5 6 7
γ/(v t
i/R)
R/LTe
(R/LTe)0
(R/LTe)crit
γ(kyρs = 0.05)γ(kyρs = 0.1)
Existence of a critical electron temperaturegradient confirmed
Influence of other parameters, Example: βe
• Bx and γlin aresensitive to βe
• Critical valueβe,crit ∼ 10−3
• Also geometry(q0, s) andcollisionalityplay a role..
0
0.01
0.02
0.03
0.04
0.05
0.001 0.01
γ/(v t
i/R)
βe
βe0
βe(ASDEXupgrade)
kyρs = 0.06
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 0.002 0.004 0.006 0.008
(Bx/B
)/(ρ
e/L
Te)
βe
Gene simulationDrake’s formula
Drake’s Formula gives rough estimate for the fluctuationamplitude
Magnetic Field Stochastization
weak drive strong drive
Magnetic field fluctuation amplitude ofmicrotearing turbulence determines degree of field
stochastization
Heat Transport in Stochastic Magnetic Fields
Diffusivity model
• χeme = vteDM
• DM = LC(δB/B0)2
(quasilinear result)
• B− correlation lengthLC = qR
Heuristic: random walk
• D = (∆r)2/∆t
• ∆r = (δB/B0)LC
• ∆t = LC/vte
0
0.5
1
1.5
2
2.5
3
3.5
4
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
χe em
/(ρ2 iv t
i/R)
(δBx/B0)/(ρi/R)
modelR/LTe variation
β variation
χeme = 1.36vteqR(δB/B0)2
GENE resultsconfirm simple model (e.g.Liewer
1985) collisionless case(breaks down at weak drive)
Effects of collision rate νe
Collisional regime
• Transition expectedaround νc = 0.01 (linearphysics)
• Lc → λmfp
• λmfp = vte/νe
• χeme = vteλmfp(δB/B0)2
∝ 1/νe expected
Gyrokinetic results
• χeme decreases weaker
than 1/νc
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.1 1 10
γ/(v t
i/R)
collision frequency ν∗
λmfp
2πq0R=1
ν∗0
λmfp
2πq0R=10
γ ∝−νc
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.1 1 10
γ/(v t
i/R)
collision frequency ν∗
λmfp
2πq0R=1
ν∗0
λmfp
2πq0R=10
γ ∝−νc
kyρs = 0.01kyρs = 0.01kyρs = 0.05
0.1
1
1
L‖/qR
collision rate ν∗
model: χeem = L‖vte(δB/B0)
2
Gene simulationγlin, ky = 0.12 (a.u.)∼ 1/νc for νc > 0.01
Nonlinear Saturation of Microtearing Turbulence
Model by Drake ’80
• γlin ∼ vteλmfp (%e/LTe)2 k2⊥
• γNL ∼ −DMk2⊥
• Balance: ⇒ δB/B0 ∼ %e/LTe
Gyrokinetic simulations
• Accessible parameter regime:γlin,max ∼ (R/LTe)2vti/R
• Dissipation scale:kdiss ≈ 0.2/%i
• γNL = −χeme k2
diss
• Balance: ⇒ δB/B0 ∼ %e/LTe
0
1
2
3
4
5
6
0 1 2 3 4 5 6
(δB
x/B
0)/(ρ
e/R
)
R/LTe
Gene simulationsδB/B0 = ρe/LTe
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.1 1
a.u.
kyρi
linear growth rate γlfree energy sources/sinks
magnetic electron heat flux Qeme
Effects of E × B shear flow
Simulation results
• Heat flux is reduced
• It does not seem to vanishup to vE×B ≈ 10γlin
• Low ky remain (unusualcompared to ITG/TEM)
Possible explanation:
• Special microtearingmode structure
0
0.2
0.4
0.6
0.8
1
1.2
0 2 4 6 8 10 12
χem e
/χem e,0
E ×B shearing rate/γmax
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
0 0.1 0.2 0.3 0.4 0.5 0.6
heatflux
kyρi
lin. growth rate γL (A.U.)Qe
em ExB rate=0.84×Qe
em ExB rate=0.8Qe
em R/LTe = 4.5
Summary and Conclusions
• Microtearing modes can be unstable in conventionalTokamaks
• Heat transport can be substantial (∼ 1m2/s)
• Drake’s Formula for saturation amplitude is roughlyconfirmed, but parameters other than LTe (βe) are crucial
• Transition to collisional regime is seen
• Further simulations:System size: globalmicrotearing + ITG. . .
• Very recent: compare to ASDEX Upgrade measurements inthe outer core (Wolfrum et al.)
Thank you for your attention!