Modeling of Transport Barrier Based on Drift Alfv´en ...
Transcript of Modeling of Transport Barrier Based on Drift Alfv´en ...
9th IAEA TM on H-mode Physics and Transport BarriersCatamaran Resort Hotel, San Diego
2003-09-25
Modeling of Transport Barrier Based on
Drift Alfven Ballooning Mode Transport Model
A. Fukuyama, M. Uchida and M. HondaDepartment of Nuclear Engineering,
Kyoto University, Kyoto 606-8501, Japan
Motivation
• Development of Robust Transport Model
L-mode confinement time scaling Large transport near the plasma edge in L-mode H-mode confinement time scaling for given edge temperature Formation of internal transport barrier Profile database (ITPA) Behavior of fluctuation
• Purpose of the present model
To describe both
— Electrostatic ITG mode
· enhanced transport for large ion temperature gradient
— Electromagnetic Ballooning mode (CDBM)
· transport reduction for negative s− α: ITB formation
magnetic shear: s =r
q
dq
dr
pressure gradient (Shafranov shift): α = −q2Rdβ
dr
Turbulent Transport Model
CDBMReduced MHD equation
ElectromagneticIncompressible
Toroidal ITG[hIon Fluid Equation
ElectrostaticBoltzmann Distribution of Electron
Drift Alfven Ballooning ModeReduced Two-Fluid Equation
ElectromagneticCompressible
• Small pressure gradient: Electrostatic ITG
• Large pressure gradient: Electromagnetic BM
• Without drift motion, it reduces to CDBM
• Ion parallel viscosity and compressibility desta-bilizes the mode
• s− α dependence similar to the CDBM mode
Reduced Two-Fluid Equation (Slab Plasma)
• Equation of Vorticity[niΛ0
ΩiB0− ε0
e− neΛ0e
ΩeB
]∂∇2
⊥φ1
∂t+
ni
ΩiB
∂
∂t
(∇2⊥p1i
qini
)− ne
ΩeB
∂
∂t
(∇2⊥p1e
qene
)
−∇‖(niv1‖i − nev1‖e) +
(ieniΛ0iω∗i
Ti+
ieneΛ0eω∗eTe
)φ1
=1
eB
(b× κ + b× ∇B
B
)· (∇p1i +∇p1e)
• Parallel Equation of Motion (j = e, i)
mjn0j
∂v1j‖∂t
+∇‖p1j − qjn0jE1‖ = 0
• Equation of State (j = e, i)
∂pj1
∂t+ vE1 · ∇pj0 + Γjpj0∇‖v‖j1 = 0
• Ampere’s Law
∇2⊥A1‖ = −µ0
∑j
(n0qjv1j‖)
Linear Analysis: Slab Plasma
• Slab ITG mode (Comparison of three models)
Ion Fluid Model (3 eqs.)
Reduced Two-Fluid Model (6 eqs.)
Full Two-Fluid Model (14 eqs.)
0 0.2 0.4−4
−3
−2
−1
0
1
2
kyρi
ωr/ω∗e
3 equations(φ,v||,pi)
14 equations (φ,nj,vjx,vjy,vjz,pj,A)
γ/ω∗e
6 equations (φ,vj||,pj,A||)
0 2 4
−1
0
1
ηe
γ/ω*e
ωr/ω*e
14 equations (ηi=5)14 equations (ηi=3)14 equations (ηi=1) 6 equations (ηi=5)
• Reduced two-fluid model is very close to the full two-fluid model.
Reduced Two-Fluid Equation (Toroidal Plasma)
• Ballooning transformation: ξ
• Equation of Vorticity
−iω
ΩiB0
m2
r2f 2
(n0iΛ0iφ1 +
p1i
qi
)− −iω
ΩeB0
m2
r2f 2
(n0eΛ0eφ1 +
p1e
qe
)
−−iωem2f 2φ1
ε0r2+
Bθ
rB0
∂
∂ξ(n0iv1j‖ − n0ev1e‖)−
imBϕ
erR0B20
H(ξ)(p1i + p1e) = 0
• Parallel Equation of Motion (j = e, i)
−iωmjn0jv1j‖ +Bθ
rB0
∂p1
∂ξ+ qjn0j
(BθΛ0j
rB0
∂φ
∂ξ− iω∗ajA‖
)= 0
• Equation of State (j = e, i)
−iωp1j − iqjn0jΛ0jω∗j(1 + ηj)φ +Γjp0jBθ
rB0
∂v1j‖∂ξ
= 0
• Ampere’s law
−m2
r2f 2A‖ = −µ0e(n0iv1i‖ − n0ev1e‖)
• H(ξ) ≡ κ0 + cos ξ + (sξ − α sin ξ) sin ξ, f 2(ξ) = 1 + (sξ − α sin ξ)2
Linear Analysis: Toroidal Plasma
• Ballooning mode (Transition from electrostatic to electromagnetic)
Electrostatic Toroidal ITG Mode
Electromagnetic Ballooning Mode
0 0.2 0.4 0.6−2
−1
0
1
α=0.8α=0.3α=0.08
k⊥ρi
ωr/ω
*e
0 0.2 0.4 0.60
0.5
1
α=0.8α=0.3α=0.08
k⊥ρi
γ/ω
*e
0 5 10
−0.4
0.1
0.6
0 0.5 1
ηj
γ/ω*e
ωr/ω*e
Eelectro−Magnetic
Eelectro−Static
γ/ω*e
ωr/ω*e
α
Ballooning Mode
ITG Mode
• Electromagnetic effect becomes dominant for large ηi or α
Nonlinear Reduced Two-Fluid Equation (Toroidal Plasma)
• Turbulent transport coefficients are included.
d
dtXj → −iωXj − χj∇2
⊥Xj, χj =
⟨φ2
1
⟩
γnj∼
√〈φ2
1〉
• Equation of Vorticity:[(−iω + µi
m2f 2
r2)
n0i
ΩiB0− −iωe
ε0− (−iω + µe
m2f 2
r2)
n0e
ΩeB0
]m2f 2
r2φ1
+
(−iω + χi
m2f 2
r2
)1
ΩiB0
m2
r2f 2p1i
qi−
(−iω + χe
m2f 2
r2
)1
ΩeB0
m2
r2f 2p1e
qe
+Bθ
rB0
∂
∂ξ(n0iv1j‖ − n0ev1e‖)−
imBϕ
erR0B20
H(ξ)(p1i + p1e) = 0
• Parallel Equation of Motion
mjn0j
(−iω + µj
m2f 2
r2
)v1j‖ +
Bθ
rB0
∂p1
∂ξ+ qjn0j
(BθΛ0j
rB0
∂φ
∂ξ− iω∗ajA‖
)= 0
• Equation of State(−iω + χj
m2f 2
r2
)p1j − iqjn0jΛ0jω∗j(1 + ηj)φ +
Γjp0jBθ
rB0
∂v1j‖∂ξ
= 0
• Ampere’s Law
−m2
r2f 2A‖ = −µ0e(n0iv1i‖ − n0ev1e‖)
• CDBM Eigenmode Equation
∂
∂ξ
γf
γ + ηrm2f + λm4f 2
∂φ1
∂ξ− (γ2f + γµm2f )φ1 +
αγ
γ + χm2fH(ξ)φ = 0
Marginal Stability Condition (γ = 0)
1
λ
∂2φ
∂ξ2− µm6f 3φ +
αm2f
χH(ξ)φ = 0
• Low β DABM Eigenmode Equation: Marginal Stability Condition
∂2φ
∂ξ2− µiχi
τ 2APc2m6f 3
r6ω2pi
φ +m2c2αi
2r2ω2pi
H(ξ)φ = 0
• Eigenmode equations for CDBM and DABM are similar.
Nonlinear Analysis (Toroidal Plasma)
• Amplitude Dependence of the Growth Rate
Linear growth rate (χj = 0) is sensitive to k⊥ρi.
For large α, electromagnetic effect becomes important.
Saturation level can be estimated from the marginal condition.
0 1 20
0.2
0.4
0.6
0.8
χj, µj [m2/s]
γ/ω
*e
(α=0.02, s=0.4, k⊥ρi=0.26)
(a)
0 1 20
0.1
µj, χj[m2/s]
γ/ω
*e
(α=0.02, s=0.4, k⊥ρi=0.36)
(b)
0 10 200
0.2
0.4
µj, χj[m2/s]
γ/ω
*e
DABM
ITGM
(α=0.2, s=0.4, k⊥ρi=0.11)
(c)
• Dependence on s and α
Transport of DABM is much larger than that of CDBM.
Negative magnetic shear reduces the transport.
χ is proportional to α3/2 for small α.
There exists critical α above which transport is strongly enhanced.
0 1
10−3
10−2
10−1
100
(α=0.01)
s
χ j, µ
j
[m2/s]
DABMCDBM
10−2 10−1 10010−1
100
101
102
s=0.8
α
χ js=0.4s=0.2s=−0.4(EM)s=−0.4(ES)
10−2 10−1 100
−1
0
1
s=0.8
α
ωr/ω
*e
s=0.4s=0.2s=−0.4s=−0.4
• Contour of χ on s-α plane
3 1
0.7 0.3 0.1
-0.4
-0.2
0
0.2
0.4
0.6
0 0.1 0.2 0.3 0.4 0.5 0.6
µj, χj [m2/s]
α
s
30 10 7 3 1
-0.4
-0.2
0
0.2
0.4
0.6
0 0.1 0.2 0.3 0.4 0.5 0.6α
s
Uncertain (nq<10)
µj, χj [m2/s]
CDBM DABM
DABM Turbulence Model
• Low β DABM Eigenmode Equation: (γ = 0)
∂2φ
∂ξ2− µiχi
τ 2APc2m6f 3
r6ω2pi
φ +m2c2αi
2r2ω2pi
H(ξ)φ = 0
• Marginal Stability Condition
χDABM = F (s, α, κ) αc2
ω2pe
vTe
qR
Magnetic shear s ≡ r
q
dq
dr
Pressure gradient α ≡ − q2Rdβ
dr
Magnetic curvature κ ≡ − r
R
(1− 1
q2
)
Different gradient dependence with CDBM
• Weak and negative magnetic shear andShafranov shift reduce χ.
s− α dependence ofF (s, α, κ, ωE1)
0.0
0.5
1.0
1.5
2.0
2.5
-0.5 0.0 0.5 1.0
F ωE1
=0.2
ωE1
=0.3
ωE1
=0.1
ωE1
=0
s - α
Fitting Formula
F =
1
1 + G1ω2E1
1√2(1− 2s′)(1− 2s′ + 3s′2)
for s′ = s− α < 0
1
1 + G1ω2E1
1 + 9√
2s′5/2
√2(1− 2s′ + 3s′2 + 2s′3)
for s′ = s− α > 0
High βp mode
• Empirical factor of 3 was chosen to reproduce the L-mode scaling.
• R = 3 m, a = 1.2 m, κ = 1.5, B0 = 3 T, Ip = 1.5 MA
• Time evolution during the first one second after 10 MW heating switched on.
Temperature evolution Temperature Safety factor
TE,TD,<TE>,<TD> [keV] vs t
1.0 1.2 1.4 1.6 1.8 2.00
1
2
3
4TD0
Te0
<Te><TD>
TD [keV] vs r
0.0 0.2 0.4 0.6 0.8 1.0 1.20
1
2
3
4TD
QP vs r
0.0 0.2 0.4 0.6 0.8 1.0 1.20
2
4
6
8
q
Current evolution Current density Thermal diffusivity
IP,IOH,INB,IRF,IBS [MA] vs t
1.0 1.2 1.4 1.6 1.8 2.00.0
0.4
0.8
1.2
1.6Itot
IOHIBS
JTOT,JOH,JNB,JRF,JBS [MA/m^2] vs r
0.0 0.2 0.4 0.6 0.8 1.0 1.20.0
0.2
0.4
0.6
0.8
1.0
Jtot
JBS
JOH
AKD [m^2/s] vs r
0.0 0.2 0.4 0.6 0.8 1.0 1.20
1
2
3
4
χi
Summary
• In order to describe both the electrostatic ion temperature gradi-ent (ITG) mode and the electromagnetic current diffusive ballooningmode (CDBM), we have derived a set of reduced two-fluid equationsin both slab and toroidal configurations and numerically solved themas an eigenvalue problem.
• Linear analysis in a toroidal configuration describes a ballooningmode, which we call DABM (Drift Alfven Ballooning Mode).
Small pressure gradient: Toroidal ITG mode
Large pressure gradient: Ballooning mode with stabilizing ω∗
• Based on the theory of self-sustained turbulence, we have numeri-cally calculated the transport coefficients from the marginal stabilitycondition.
When α is small, χ is approximately proportional to α3/2.
χ is an increasing function of s−α; similar to CDBM model whichsuccessfully reproduces the ITB formation.
When α exceeds a critical value, χ starts to increase strongly withα, which may suggest the stiffness of the profile
• Using an electrostatic approximation, we have derived a formula ofthermal diffusivity slightly different from the CDBM model.
• Preliminary transport simulation reproduces the formation of ITB.The barrier locates in outer region and the gradient is steeper thanthe CDBM model.
• More general expression is required for high-β electromagnetic re-gion.
CDBM Turbulence Model
• Marginal Stability Condition (γ = 0)
χCDBM = F (s, α, κ, ωE1) α3/2 c2
ω2pe
vA
qR
Magnetic shear s ≡ r
q
dq
dr
Pressure gradient α ≡ − q2Rdβ
dr
Magnetic curvature κ ≡ − r
R
(1− 1
q2
)
E ×B rotation shear ωE1 ≡ r2
svA
d
dr
E
rB
• Weak and negative magnetic shear,
Shafranov shift and
E ×B rotation shear
reduce thermal diffusivity.
s− α dependence ofF (s, α, κ, ωE1)
0.0
0.5
1.0
1.5
2.0
2.5
-0.5 0.0 0.5 1.0
F ωE1
=0.2
ωE1
=0.3
ωE1
=0.1
ωE1
=0
s - α
Fitting Formula
F =
1
1 + G1ω2E1
1√2(1− 2s′)(1− 2s′ + 3s′2)
for s′ = s− α < 0
1
1 + G1ω2E1
1 + 9√
2s′5/2√
2(1− 2s′ + 3s′2 + 2s′3)for s′ = s− α > 0
Heat Transport Simulation
• Simple One-Dimensional Analysis
No impurity, No neutral, No sawtooth Fixed density profile: ne(r) ∝ (1− r2/a2)1/2
Thermal diffusivity (adjustable parameter C = 12)
χe = CχTB + χNC,e
χi = CχTB + χNC,i
• Transport Equation∂
∂t
3
2neTe = −1
r
∂
∂rr neχe
∂Te
∂r+ POH + Pie + PHe
∂
∂t
3
2niTi = −1
r
∂
∂rr niχi
∂Ti
∂r− Pie + PHi
∂
∂tBθ =
∂
∂rηNC
[1
µ0
1
r
∂
∂rrBθ − JBS − JLH
]
• Standard Plasma Parameter
R = 3 m Bt = 3 T Elongation = 1.5
a = 1.2 m Ip = 3 MA ne0 = 5× 1019 m−3
High βp mode (1)
• R = 3 m, a = 1.2 m, κ = 1.5, B0 = 3 T, Ip = 1 MA
• Time evolution during the first one second after heating switched on
Temperater Current Safety factor
Te
T [k
eV]
r [m]0.0 0.2 0.4 0.6 0.8 1.0 1.2
0
2
4
6JTOT
J [M
A/m
2 ]r [m]
0.0 0.2 0.4 0.6 0.8 1.0 1.20.0
0.2
0.4
0.6
r [m]
q q
0.0 0.2 0.4 0.6 0.8 1.0 1.20
4
8
12
Shear Normalized Pressure Thermal diffusivity
s
s
r [m]0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.0
0.5
1.0
1.5 α
α
r [m]0.0 0.2 0.4 0.6 0.8 1.0 1.2
0
1
2
3
4
χD
r [m]χ
[m2 /
s]
0.0 0.2 0.4 0.6 0.8 1.0 1.20
5
10
15
20
High βp mode (2)
• One second after heating power of PH = 20 MW was switched on
Temperater profile Current profile Safety factor
Te
TD
T [k
eV]
r [m]0.0 0.2 0.4 0.6 0.8 1.0 1.2
0
2
4
6JTOT
JOH
J [M
A/m
2 ]r [m]
JBS
0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.0
0.2
0.4
0.6
r [m]
q q
0.0 0.2 0.4 0.6 0.8 1.0 1.20
4
8
12
Shear and pressure Thermal diffusivity s− α diagram
0.0 0.2 0.4 0.6 0.8 1.0 1.2
0
1
2
3
4
r [m]
s
α F(s,α)
s, α
, F(s
,α) χ
D
χTB
χNC
r [m]
χ [m
2 /s]
0.0 0.2 0.4 0.6 0.8 1.0 1.20
5
10
15
20
0 1 2 3 4 50.0
0.4
0.8
1.2
1.6
α
s before heating
1s after heating on
r=0m
r=0.6m
r=0.6m
r=1.2m
Simulation of Current Hole Formation
• Current ramp up: Ip = 0.5 −→ 1.0 MA
• Moderate heating: PH = 5 MW
• Current hole is formed.
• The formation is sensitive to the edge temperature.
.1*2+
./-*2+
.,0*2+
.*/+
-0
,0
.*/+
-0
,0
/.*0+
,1
-110.0*2+
1.-*2+
1,/*2+0./*1+
-2
,2
Simulation of Reversed Shear Configuration
Ip : 3 MA constantHeating : 20 MWH factor ' 0.95
I [M
A]
T [k
eV]
τ E [
s]
IBS
IOH
Te0 TD0
<Te><TD>
τE τEITER89
t [s]
Ip : 1 MA −→ 3 MA/1 sHeating : 20 MW
H factor ' 1.6
I [M
A]
T [k
eV]
τ E [
s]
Ip IBS IOH
Te0TD0
<Te><TD>
τE
τEITER89
t [s]